Annual Review of Medicine Volume 55 2004

Annual Review of Medicine Volume 55, 2004 CONTENTS Effect of Completed Human Genome Sequence on Development of Novel Th...

Author: C. Thomas Caskey

39 downloads 1207 Views 5MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!

Report copyright / DMCA form

DOWNLOAD PDF

Annual Review of Medicine Volume 55, 2004

CONTENTS Effect of Completed Human Genome Sequence on Development of Novel Therapeutics for Human Disease, Christopher P. Austin Toward Alzheimer Therapies Based on Genetic Knowledge, John Hardy

Annu. Rev. Med. 2004.55:1-13. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Inherited Diseases Involving G Proteins and G Protein--Coupled Receptors, Allen M. Spiegel, Lee S. Weinstein The Scientific Basis for the Current Treatment of Parkinson's Disease, C. Warren Olanow Progress in Antisense Technology, Stanley T. Crooke Serum Proteomics in the Early Diagnosis of Cancer, Kevin P. Rosenblatt, Peter Bryant-Greenwood, J. Keith Killian, Arpita Mehta, David Geho, Virginia Espina, Emanuel F. Petricoin, Lance A. Liotta Molecular Neurobiology of Drug Addiction, Jennifer Chao, Eric J. Nestler Beta Cell Replacement for Type 1 Diabetes, Peter G. Stock, Jeffrey A. Bluestone Cochlear Implantation for the Treatment of Deafness, Benjamin J. Copeland, Harold C. Pillsbury Drug-Eluting Stents, T. Cooper Woods, Andrew R. Marks New Approaches to Hemodialysis, Andreas Pierratos Emerging Infectious Threats to the Blood Supply, Roger Y. Dodd, David A. Leiby Lead Poisoning, Herbert Needleman The Impact of Minimally Invasive Surgical Techniques, Sir Ara Darzi, Yaron Munz Implementing a Research Agenda for Complementary and Alternative Medicine, Jonathan D. Berman, Stephen E. Straus Basic Advances and New Avenues in Therapy of Spinal Cord Injury, Bruce H. Dobkin, Leif A. Havton Clinical Management of Tuberculosis in the Context of HIV, Bouke C. de Jong, Dennis M. Israelski, Elizabeth L. Corbett, Peter M. Small HIV-Associated Lipodystrophy: Pathogenesis, Prognosis, Treatment, and Controversies, Polyxeni Koutkia, Steven Grinspoon Human Papillomavirus Vaccines and Prevention of Cervical Cancer, Kathrin U. Jansen, Alan R. Shaw Opportunities for Control of Meningococcal Disease in the United States, Pratima L. Raghunathan, Scott A. Bernhardt, Nancy E. Rosenstein Recent Advances in the Development of HIV-1 Vaccines Using Replication-Incompetent Adenovirus Vectors, John W. Shiver, Emilio A. Emini Left Ventricular Diastolic Dysfunction and Diastolic Heart Failure, William H. Gaasch, Michael R. Zile Mechanisms of Pulmonary Fibrosis, Victor J. Thannickal, Galen B. Toews, Eric S. White, Joseph P. Lynch III, Fernando J. Martinez Systemic Mastocytosis, Cem Akin, Dean D. Metcalfe

1 15 27 41 61

97 113 133 157 169 179 191 209 223 239 255

283

303 319

333

355 373 395 419

The erbB Family: Targets for Therapeutic Development Against Cancer and Therapeutic Strategies Using Monoclonal Antibodies and Tyrosine Kinase Inhibitors, Eric K. Rowinsky Nonmyeoablative Immunotherapy for Solid Tumors, Richard W. Childs, John Barrett Rituximab: Expanding Role in Therapy for Lymphomas and Autoimmune Diseases, William Rastetter, Arturo Molina, Christine A. White

Annu. Rev. Med. 2004.55:1-13. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Botulinum Toxin and Other New Approaches to Migraine Therapy, Avi Ashkenazi, Stephen D. Silberstein Management of Infections in the Neutropenic Patient, Kenneth V.I. Rolston

433 459

477 505 519

12 Dec 2003

13:30

AR

AR206-ME55-01.tex

AR206-ME55-01.sgm

LaTeX2e(2002/01/18) P1: GBC 10.1146/annurev.med.55.091902.104426

Annu. Rev. Med. 2004. 55:1–13 doi: 10.1146/annurev.med.55.091902.104426 First published online as a Review in Advance on Nov. 3, 2003

THE IMPACT OF THE COMPLETED HUMAN GENOME SEQUENCE ON THE DEVELOPMENT OF NOVEL THERAPEUTICS FOR HUMAN DISEASE∗ Annu. Rev. Med. 2004.55:1-13. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Christopher P. Austin National Human Genome Research Institute, National Institutes of Health, Bethesda, Maryland 20892; email: [email protected]

Key Words translational research, chemical genomics, drug development, Human Genome Project ■ Abstract With the official completion of the Human Genome Project in April 2003, we have both the opportunity and the imperative to translate this unprecedented scientific accomplishment into tangible improvements in human health. Medical benefits from the genome will come in stages and can be conceptualized as occurring in three areas: improved understanding of disease causation at the molecular level, improved diagnosis and disease classification based on genetic profiles, and new therapeutics based on targets identified in the genome. These improvements will require increased physician understanding of genetic principles applied to common diseases.

WHAT WAS THE HUMAN GENOME PROJECT? The Human Genome Project (HGP) was an international collaboration begun in 1990 and officially completed April 14, 2003, 50 years to the month after the discovery of the double helical structure of DNA by Watson & Crick (1). The concept of sequencing the genome was being informally discussed within the scientific community as far back as the early 1980s but only became a reality after a National Research Council report in 1988 recommended that a “special project” to map and sequence the genome of humans and several model organisms be undertaken (2). The report predicted that the project would take 15 years; the HGP was actually completed in 12.5 years at a cost of ∼$2.7 billion, significantly less than originally projected. From the beginning, the HGP was an international cooperative effort, with 20 sequencing centers in six countries participating, although five centers in the United States and United Kingdom contributed the bulk of the sequence. A particularly noteworthy aspect of the HGP was the earmarking of ∼5% ∗ The U.S. Government has the right to retain a nonexclusive, royalty-free license in and to any copyright covering this paper.

1

12 Dec 2003

Annu. Rev. Med. 2004.55:1-13. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

2

13:30

AR

AR206-ME55-01.tex

AR206-ME55-01.sgm

LaTeX2e(2002/01/18)

P1: GBC

AUSTIN

of its budget for study of the ethical, legal, and social implications (ELSI) of the science being done. Unprecedented at the time, concurrent study of ELSI issues raised by scientific research has now become more common. Excellent histories and perspectives on the HGP have recently been published (3, 4). Thanks to principles of full and immediate data release established by the members of the HGP consortium, the entire human genome sequence is now available to anyone with an Internet connection at multiple sites on the World Wide Web, including http://www.ncbi.nih.gov/genome/guide/human/. The genomes of many other eukaryotic organisms have also been sequenced, including those of yeast, the nematode worm Caenorhabditis elegans, the fruit fly Drosophila melanogaster, and the mouse Mus musculus. Many other genomes are currently being sequenced, including those of the chimpanzee, dog, cat, cow, chicken, and honeybee; interspecies genome comparisons have begun to alter classical conceptions of evolutionary relationships among species (5). Equally important from a medical perspective has been the sequencing of the genomes of well over 100 prokaryotes, parasites, and viruses, including many human pathogens (Table 1). A recent example of the power of genome sequencing to influence human health was the complete sequencing of the causative organism for severe acute respiratory syndrome (SARS) within two weeks of its detection in SARS patients in April 2003 (http:// www.ncbi.nlm.nih.gov/genomes/SARS/sars more.html#Goto sars; see also 6).

NEXT STEPS FOR THE HUMAN GENOME Sequencing the genomes of humans and other organisms is an enormous basic scientific achievement, but it has no immediate clinical benefit. Early, hyperbolic prophesies of the medical benefits of the human genome sequence, often coming from those seeking professional or financial gain, have been discredited, often replaced by equally erroneously pessimistic prophesies that genomics will have little if any effect on clinical medicine for the foreseeable future (7). The truth is probably somewhere in between (8), and the remainder of this review discusses steps being taken to purposefully translate the genome(s) into tangible improvements in human health. A series of meetings of over 600 scientists convened by the National Human Genome Research Institute during 2001 and 2002 resulted in a recently published vision for the future of genomics research (9). This plan foresees impacts of genomics in three overlapping arenas: basic biology, health, and society, pictured as three floors of a house built on the foundation of the HGP (Figure 1). Crosscutting elements that apply to all three areas are pictured as supporting beams; these include education and training, technology development, and consideration of ELSI implications of developments on any of the three floors. Floor 2, Genomics to Health, includes the following goals: (a) to identify genes and pathways involved in disease; (b) to develop and apply genome-based diagnostic methods for disease diagnosis, disease classification, and prospective prediction of disease susceptibility and drug response; and (c) to catalyze development of new therapeutics based on genomic information. These goals are discussed below.

12 Dec 2003

13:30

AR

AR206-ME55-01.tex

AR206-ME55-01.sgm

LaTeX2e(2002/01/18)

HUMAN GENOME SEQUENCE AND DRUG DEVELOPMENT

P1: GBC

3

TABLE 1 Selected human pathogens with sequenced genomes Bacillus anthracis Borrelia burgdorferi Campylobacter jejuni Chlamydia pneumoniae Chlamydia trachomatis Clostridium tetani Annu. Rev. Med. 2004.55:1-13. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Enterococcus faecalis Escherichia coli Haemophilus influenzae Helicobacter pylori Human immunodeficiency virus Listeria monocytogenes Mycobacterium tuberculosis Mycoplasma pneumoniae Neisseria meningitidis Plasmodium falciparum Pseudomonas aeruginosa Rickettsia prowazekii Salmonella typhi SARS coronavirus Shigella flexneri Staphylococcus aureus Streptococcus pneumoniae Treponema pallidum Tropheryma whipplei Vibrio cholerae Yersinia pestis

Identifying Genes and Pathways Responsible for Human Disease The most immediate application of genome sequences is to accelerate identification of genes whose dysfunction causes or modifies human diseases. The human genome will be the source of such genes for most diseases, although in the special case of infectious diseases, the genomes of pathogenic organisms are scrutinized to identify genes responsible for viability, pathogenesis, virulence, and host interaction (e.g., 10).

12 Dec 2003

Annu. Rev. Med. 2004.55:1-13. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

4

13:30

AR

AR206-ME55-01.tex

AR206-ME55-01.sgm

LaTeX2e(2002/01/18)

P1: GBC

AUSTIN

Figure 1 The impacts of genomics in three overlapping arenas—basic biology, health, and society—can be pictured as three floors of a house built on the foundation of the Human Genome Project (see text for details). From Reference 9 with permission.

There are at least three basic strategies to identify human disease genes, and all will be accelerated by the human genome sequence. The most conventional method is via genetic linkage or association studies; this strategy can identify genes that are unequivocally related to the disease, but it is arduous and may identify disease genes of unknown function or mechanism. The advent of the genome sequence has allowed researchers to move rapidly from large chromosomal regions to individual candidate genes, since all genes identified within any linkage region are immediately available; this process formerly took months to years but now can be reduced to days. This change is illustrated by the recent recognition that the lamin A gene is responsible for Hutchinson-Gilford progeria syndrome (HGPS) very soon after the linked region was identified (11). The search for the causes of single-gene disorders such as HGPS has been greatly accelerated by the genome sequence. However, more powerful tools will be needed to find the genes responsible for common diseases such as diabetes, major depression, and hypertension, which are caused by the combined effects of multiple genes along with environmental factors. Among such tools under development is a haplotype map of the human genome. Haplotypes are sets of genetic markers that together represent blocks of contiguous human DNA; these blocks are roughly consistent among all humans, but different individuals have different versions of the blocks, allowing identification of relationships between these

12 Dec 2003

13:30

AR

AR206-ME55-01.tex

AR206-ME55-01.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:1-13. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

HUMAN GENOME SEQUENCE AND DRUG DEVELOPMENT

P1: GBC

5

haplotype blocks and diseases in groups of individuals. The International Haplotype Map Project aims to define the haplotype structure of human populations, and markers to identify them unambiguously (see http://genome.gov/10001688). This “HapMap” will greatly simplify genetic association studies by decreasing the number of genetic markers that must be queried by a factor of six or more (12; see also http://genome.gov/10005336). Construction of the HapMap began in October 2002 and is expected to be complete by the end of 2004. A second approach that will improve the identification of disease genes is a project called ENCODE, or the Encyclopedia of DNA Elements (see http://genome. gov/10005107), which began this year and aims to comprehensively identify all structural and functional features in a selected 1% of the human genome. Successful techniques will then be applied to annotate the remaining 99%. The current best estimate of the number of genes (defined as protein-coding regions) in the human genome is ∼30,000, based on comparison of known genes (identified by other methods) with the genome sequence and comparison of the human genome with the genomes of related organisms such as the mouse (13). However, it is likely that some genes in the human genome have not been found because they are unlike previously identified genes and thus are not identified by homology methods. Conversely, some genes currently counted will be found to be bioinformatic artifacts or pseudogenes that are no longer functional. In addition, the regions of DNA responsible for correct timing and location of gene expression are almost entirely unknown, but they are critical to human diseases, many of which are due in part to incorrect gene expression. A well-known example of this principle is the overexpression of the growth factor receptor Her2/Neu in a subset of breast cancers, making them susceptible to the specific monoclonal antibody trastuzumab (Herceptin®). The new genes and regulatory elements identified by ENCODE will be immediately useful in identifying new human disease genes, but equally importantly, the methods developed will enable researchers to develop a complete catalogue of functional elements in the entire genome. A third approach for identifying human disease genes relies not on the analysis of gene sequences but rather on their expression in pathological conditions. “Expression” in this context refers to the production of messenger RNA (mRNA) or protein from the gene encoding them; the presence of the mRNA or protein in pathological but not normal tissues is interpreted as circumstantial evidence that the gene is involved in disease. The caveat for this type of study is that it shows correlation, not causation, and as such, it is most useful in providing candidates for further investigations of causation. For diagnostic purposes, however, such correlation is often sufficient, so these methods are used increasingly to identify markers to improve diagnosis (see below). DNA microarrays (sometimes referred to as DNA chips) allow the simultaneous identification of many tens of thousands of genes expressed in a particular pathological tissue and allow this gene pattern to be compared to the normal condition. Using this approach, novel genes involved in cancer, diabetes, schizophrenia, and many other diseases have been identified (e.g., 14, 15). Knowing the complete human genome sequence has enabled the

12 Dec 2003

Annu. Rev. Med. 2004.55:1-13. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

6

13:30

AR

AR206-ME55-01.tex

AR206-ME55-01.sgm

LaTeX2e(2002/01/18)

P1: GBC

AUSTIN

production of increasingly comprehensive microarrays; several companies now market microarrays representing all known human genes. These complete arrays will further accelerate the identification of candidate disease genes by improving the identification of genes whose expression is altered in human disease tissue and in animal models of human disease. A final approach to disease gene identification that the human genome sequence has dramatically accelerated is the identification of new members of known gene families whose function is then associated with disease. A striking example of this approach was the identification of a new member of the apolipoprotein family, well known as regulators of cholesterol transport and metabolism, by comparison of conserved regions of the human and mouse genomes. This gene, ApoA5, has been shown to influence triglyceride levels in populations and, when mutated, causes hypertriglyceridemia (16, 17). Related gene family approaches, in which genome databases were searched for novel G-protein–coupled receptors, identified the urotensin II receptor (18), which is the most potent mammalian vasoconstrictor identified and may play a role in essential hypertension (19). The acceleration in identification of disease genes promises to continue and will shed increasing light on the pathogenesis of many human diseases. Although this improved understanding will be helpful to patients and caregivers, its most immediate effect on health care is in the arena of diagnosis and prognosis of disease.

Genome-Based Diagnostics To the extent that a particular disease is caused by genetic factors, its diagnosis will be improved by the identification of gene sequence or expression variants that are causally related or correlated with the disease. The percentage of disease risk that can be assigned to genetic factors varies greatly, from perhaps a low of <5% for traumatic injuries to 100% for single-gene (Mendelian) disorders such as cystic fibrosis or Huntington’s disease. Over 1500 Mendelian disease genes have been identified, out of a total of ∼3000 human disease syndromes attributed to single gene mutations, and new gene-disease associations are being made at a rate of approximately five per week (http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db= OMIM). As these associations are made, many are rapidly converted into diagnostic tests, although in some cases gene patents have restricted the use of the tests, as is the case with BRCA1 testing in breast cancer (20). The increasing availability of these tests will allow accurate diagnosis for an expanding number of rare disorders. Although single-gene disorders are of great importance to families affected by them, they have relatively little impact on most clinical practice, since even when added together these disorders are relatively rare. However, the common disorders of Western populations—e.g., cardiovascular disease, cancer, and Alzheimer’s disease—have a genetic component, and elucidation of genetic risk factors for these disorders is having an increasing impact on clinical medicine. For example, testing for Factor V Leiden mutations as a risk factor for thromboembolism, HFE

12 Dec 2003

13:30

AR

AR206-ME55-01.tex

AR206-ME55-01.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:1-13. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

HUMAN GENOME SEQUENCE AND DRUG DEVELOPMENT

P1: GBC

7

mutation testing for hereditary hemochromatosis as a risk factor for cardiovascular disease, BRCA1 testing in breast cancer patients and their family members, and population screening for cystic fibrosis mutations are all in routine clinical practice, although the appropriate use of these tests remains the subject of debate (21). For most common disorders, the disease risk attributable to genetic factors is around 50%, with the other 50% attributable to environmental factors such as smoking, diet, and exercise. The genetic risk is due to the additive effects of many genes, however, each of which individually accounts for only a small degree of disease susceptibility, usually ∼5% or less. A critical distinction exists between genetic alterations that cause rare, singlegene (Mendelian) disorders and more common, multigenic (complex) disorders. Genetic alterations in single-gene disorders tend to be sequence changes, insertions, or deletions that cause major disruption of a gene and its protein product, and these changes generally cause the diseases with which they are associated, either directly or indirectly. In contrast, the genetic alterations that are associated with common multigenic disorders tend to be much more subtle, for example, changing the expression or activity of the gene product, sometimes only in certain environmental conditions. Thus, the genetic alterations that are associated with common diseases only increase or decrease susceptibility to these diseases, rather than cause them. This distinction has two important consequences for practicing physicians. First, because virtually all diseases commonly seen by physicians have a genetic component, all physicians will need to become familiar with the principles of genetics and genetic testing, so that they can properly diagnose, treat, and counsel their patients. Medical genetics, traditionally the realm of specialists, is rapidly becoming the realm of primary care physicians. For this reason, the National Human Genome Research Institute and the Health Resources and Services Administration jointly sponsored a “Genetics in Primary Care” program that brought together genetics faculty and primary care physicians in a number of academic institutions (http://genes-r-us.uthscsa.edu/resources/genetics/primary care.htm). Recently, the American Academy of Family Physicians designated genetics and genomics as their Annual Clinical Focus for 2005. There is a clear need for further programs of this kind (22), and the National Coalition for Health Professional Education in Genetics (http://www.nchpeg.org/) is an excellent resource. Second, because common diseases are associated with multiple genetic alterations that each confer limited risk but cause disease when multiply inherited and/or combined with environmental risk factors, results of genetic tests for common diseases must be interpreted in the context of other risk factors in assessing aggregate disease risk. Physicians are familiar with this challenge in, for example, explaining the additive effects of environmental factors such as smoking and sedentary lifestyle on the risk of developing cardiovascular disease. Genetic tests will provide an additional type of laboratory data that will complement the traditional blood and body fluid analyses, and will allow patients and their physicians to more precisely determine individual risk for common diseases.

12 Dec 2003

Annu. Rev. Med. 2004.55:1-13. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

8

13:30

AR

AR206-ME55-01.tex

AR206-ME55-01.sgm

LaTeX2e(2002/01/18)

P1: GBC

AUSTIN

This discussion has focused on DNA sequence alterations that are diagnostic of disease, but gene expression alterations in disease are also moving into clinical practice. The most widespread tool to detect them is the DNA microarray, which can simultaneously determine the expression levels of thousands of genes in a tissue or blood sample. This method was initially used to identify gene expression patterns that define subsets of lymphomas (23, 24) and breast cancers (25, 26) that, though histologically indistinguishable, respond differently to chemotherapy and confer distinct prognoses. The microarray approach has now been applied to a broad variety of other cancers, including brain, ovarian, lung, colon, kidney, prostatic, gastric, and leukemia (see 27 for review), and the first large clinical trial assigning breast cancer patients to treatment on the basis of their gene expression profiles has recently begun (28). One of the first studies to discriminate among patients with a non-neoplastic disorder—renal allograft rejection—using gene expression profiling was recently published (29), and it is expected that as microarray profiling increasingly allows more accurate diagnoses and appropriate treatment, its use in the clinic will become more common.

New Therapeutics Based on New Targets Identified in the Genome The path from the identification of a gene to a new drug is well known to be long, expensive, and failure-prone. A recent study from the Tufts Center for the Study of Drug Development estimates that development of one drug from target validation to approval by the US Food and Drug Administration (FDA) takes 12–15 years and $403 million in out-of-pocket costs (30). In the mid and late 1990s, in the midst of the HGP, some optimists predicted that the abundance of data it generated would make drug development easier and cheaper. In fact, the opposite has occurred: Drug development times and costs have increased, and the rate of new drug approvals has declined (31). Among the many factors probably driving these trends is that the very characteristic that makes new “genomic” targets attractive—their novelty—also makes drugs directed against them more likely to fail, since less is known about the targets’ functions in complex biological systems (32). In addition, having genomic sequence information does not directly affect the primary reasons for drug development failure, namely pharmacokinetics and toxicity problems and lack of adequate efficacy in human trials (33). A dramatic illustration of these realities is the fate of the genomics-centered biotechnology companies that achieved stratospheric levels of publicity and market capitalization in 2000. Many of these companies, which came to symbolize the hoped-for power of genomics to transform drug development, have now been acquired or have reinvented themselves as drug development organizations (34). The supposition that the human genome sequence can have a positive effect on the development of new drugs has therefore been questioned (7, 35). However, even though new therapeutics will continue to depend on the old rules and timelines of drug development, there are signs that the genome will indeed contribute

12 Dec 2003

13:30

AR

AR206-ME55-01.tex

AR206-ME55-01.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:1-13. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

HUMAN GENOME SEQUENCE AND DRUG DEVELOPMENT

P1: GBC

9

meaningfully and increasingly to making better drugs faster. First, the concept that therapies based on molecular understanding of disease may be more effective than those based on phenotypes has been spectacularly validated by the success of several targeted cancer therapies. Trastuzumab (Herceptin®), a monoclonal antibody directed against the Her2/Neu growth factor receptor amplified in a subset of breast cancers (36); imatinib (Gleevec®), a Bcr-Abl tyrosine kinase inhibitor specific for the molecular defect in chronic myeloid leukemia (37); and most recently bortezomib (Velcade®), a proteasome inhibitor that prevents activation of the NFκB pathway in multiple myeloma (38), have all produced remarkable clinical improvements in patients with these diseases. Several new drugs in late stages of clinical development are novel proteins derived directly from genome sequence information, including Keratinocyte Growth Factor-2 (Human Genome Sciences, Rockville, MD) and Famoxin [Fatty Acid Metabolic Oxidation Inducer (Genset, Evry, France)]. Second, the same array technologies that are used to diagnose diseases more precisely (see previous section) are also used to determine specificity and mechanism of action of novel drugs, allowing more precise molecular evaluations of drug effects than have been possible using whole-animal phenotypic testing (39). Importantly, the FDA has initiated a program to work with pharmaceutical and biotechnology companies to develop data standards for submission of array data, which should accelerate the use of this and other complex genomic data in regulatory submissions (40). More precise molecular diagnosis of disease, whether based on gene expression or gene sequence patterns, promises to improve both the development and effectiveness of drugs by allowing their use to be targeted to appropriate patients. The first prominent example of this was Herceptin®. More recently, in a trial of the reverse transcriptase inhibitor abacavir to treat HIV/AIDS patients, genetic sequence profiles of potential subjects were used to exclude those who may be predisposed to hypersensitivity reactions (41). Finally, Roche Diagnostics has released the first DNA sequence microarray intended for routine clinical use in the pharmacogenetic profiling of two polymorphic drug-metabolizing enzymes, CYP2D6 and CYP2C19. Variances in their sequence account for much of the variance in patient response to several commonly used drugs (42; see also http://www.rochediagnostics.com/media/pdf/press release/2003/background amplichip 450.pdf). A final trend that is facilitating the translation of genome information into medically important advances is the utilization of medicinal small-molecule compounds for basic biological studies and, increasingly, the early stages of drug development in the public sector. Although there are estimated to be ∼30,000 genes in the human genome and ∼100,000 protein products (many genes make more than one protein), fewer than 500 proteins have been targeted by small-molecule drugs (43, 44). Of those, only 120 are targeted by currently marketed drugs (45). Despite vigorous debate concerning the probable number of unappreciated drug targets in the human genome, ranging from a few hundred to several thousand (45, 46), it is clear that well over 90% of gene products are recognized by no small-molecule

12 Dec 2003

Annu. Rev. Med. 2004.55:1-13. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

10

13:30

AR

AR206-ME55-01.tex

AR206-ME55-01.sgm

LaTeX2e(2002/01/18)

P1: GBC

AUSTIN

Figure 2 Technological developments that enable large-scale compound screening in the public sector.

compound. Because small-molecule modulators can be invaluable in determining “druggability” (i.e., manipulability of a target with therapeutic effect and no obvious adverse effects), the availability of more small-molecule compounds for this purpose should expand and speed discovery of new targets from the genome. Furthermore, because small-molecule probes may be starting points for drug development, such tools may be expected to accelerate the discovery and development of new drugs, particularly for “orphan diseases” with too low a prevalence to be attractive to the private sector. Traditionally, most of the work to evaluate the therapeutic potential of a drug target, and all of the drug screening and medicinal chemistry required to identify new drug leads, has been done in the private sector. Several events in the past decade have begun to shift some of these activities into the public sector (Figure 2). First, the HGP has provided a plethora of new potential targets in need of functional characterization and therapeutic assessment. Second, combinatorial chemistry and the advent of specialized companies that sell high-quality chemical libraries have made small-molecule compounds available to the public sector on a large scale for the first time. Third, the robotics, instrumentation, and informatics required to track and screen hundreds of thousands of compounds in multiwell plates have become well developed, partly in response to the robotics demands of the HGP. Together, these developments have made sophisticated public-sector chemistry and screening initiatives possible on a scale that has traditionally existed only in pharmaceutical companies (47, 48; see also http://iccb.med.harvard.edu). At the National Institutes of Health (NIH), a major new effort has been initiated to bring the tools of small-molecule chemistry to biology, as part of the NIH Director’s “Roadmap” process. This “Molecular Libraries” initiative will establish public-sector resources for chemical synthesis and library construction, assay development, screening, medicinal chemistry, and cheminformatics. It will include a substantial technology development component focused on expanding chemical diversity, improving assay development and robotics, and developing programs for predictive pharmacokinetics and toxicology. All tools developed will be made available to the entire biomedical research community, as was the case with DNA

12 Dec 2003

13:30

AR

AR206-ME55-01.tex

AR206-ME55-01.sgm

LaTeX2e(2002/01/18)

P1: GBC

HUMAN GENOME SEQUENCE AND DRUG DEVELOPMENT

11

sequence information during the HGP, in order to enable the broadest possible range of research and therapeutic applications (48a).

Annu. Rev. Med. 2004.55:1-13. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

CONCLUSIONS The year 2003 has been an epochal one for science, with the complete DNA sequence of our species becoming available to us all. This accomplishment has come, however, in the context of a decreasing rate of new drug development and evidence that many physicians lack the necessary genetics knowledge and skills to take full advantage of new genome-based diagnostic and therapeutic tools. Improved understanding of disease biology, and in some cases reclassification of phenotypically defined diseases in molecular terms (49, 50), has already begun. Within two to three years, clinicians can expect the introduction of gene expression diagnostic tests for common diseases such as cancer and the increasing use of gene sequence–based tests to customize drug dosages to individual patients and assess the risk of common diseases, and physicians will need to learn to interpret results of these tests correctly. Finally, although it can be expected to require at least five years, new, more effective, and more customized therapeutics can be expected from public- and private-sector efforts in drug discovery and development. Together, these advances will bring the medical promise of the genome to fruition (51). The Annual Review of Medicine is online at http://med.annualreviews.org

LITERATURE CITED 1. Watson JD, Crick FHC. 1953. A structure for deoxyribose nucleic acid. Nature 171:737–38 2. National Research Council. 1988. Mapping and Sequencing the Human Genome. Washington, DC: Natl. Acad. 3. Collins FS, Morgan M, Patrinos A. 2003. The Human Genome Project: lessons from large-scale biology. Science 300:286–90 4. Judson HF. 2003. “The greatest surprise for everyone”—notes on the 50th anniversary of the double helix. N. Engl. J. Med. 348:1712–14 5. Thomas JW, Touchman JW, Blakesley RW, et al. 2003. Comparative analyses of multi-species sequences from targeted genomic regions. Nature 424:788–93 6. Kuiken T, Fouchier RA, Schutten M, et al.

7.

8. 9.

10.

11.

2003. Newly discovered coronavirus as the primary cause of severe acute respiratory syndrome. Lancet 362:263–70 Holtzman NA, Marteau TM. 2000. Will genetics revolutionize medicine? N. Engl. J. Med. 343:141–44 Bell JI. 2003. The double helix in clinical practice. Nature 421:414–16 Collins FS, Green ED, Guttmacher AE, Guyer MS. 2003. A vision for the future of genomics research. Nature 422:835–47 Read TD, Peterson SN, Tourasse N, et al. 2003. The genome sequence of Bacillus anthracis Ames and comparison to closely related bacteria. Nature 423:81–86 Eriksson M, Brown WT, Gordon LB, et al. 2003. Recurrent de novo point mutations in lamin A cause Hutchinson-Gilford progeria syndrome. Nature 423:293–98

12 Dec 2003

13:30

Annu. Rev. Med. 2004.55:1-13. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

12

AR

AR206-ME55-01.tex

AR206-ME55-01.sgm

LaTeX2e(2002/01/18)

P1: GBC

AUSTIN

12. Gabriel SB, Schaffner SF, Nguyen H, et al. 2002. The structure of haplotype blocks in the human genome. Science 296:2225–29 13. Waterston RH, Lindblad-Toh K, Birney E, et al. 2002. Initial sequencing and comparative analysis of the mouse genome. Nature 420:520–62 14. Aitman TJ, Glazier AM, Wallace CA, et al. 1999. Identification of Cd36 (Fat) as an insulin-resistance gene causing defective fatty acid and glucose metabolism in hypertensive rats. Nat. Genet. 21:76–83 15. Mirnics K, Middleton FA, Marquez A, et al. 2000. Molecular characterization of schizophrenia viewed by microarray analysis of gene expression in prefrontal cortex. Neuron 28:53–67 16. Pennacchio LA, Olivier M, Hubacek JA, et al. 2001. An apolipoprotein influencing triglycerides in humans and mice revealed by comparative sequencing. Science 294:169–73 17. Kao JT, Wen HC, Chien KL, et al. 2003. A novel genetic variation of apolipoprotein A5 gene is associated with hypertriglyceridemia. Hum. Mol. Genet. 12:2533–39 18. Ames RS, Sarau HM, Chambers JK, et al. 1999. Human urotensin-II is a potent vasoconstrictor and agonist for the orphan receptor GPR14. Nature 401:282–86 19. Behm DJ, Harrison SM, Ao Z, et al. 2003. Deletion of the UT receptor gene results in the selective loss of urotensin-II contractile activity in aortae isolated from UT receptor knockout mice. Br. J. Pharmacol. 139:464–72 20. Sevilla C, Julian-Reynier C, Eisinger F, et al. 2003. Impact of gene patents on the cost-effective delivery of care: the case of BRCA1 genetic testing. Int. J. Technol. Assess. Health Care 19:287–300 21. Khoury MJ, McCabe LL, McCabe ER. 2003. Population screening in the age of genomic medicine. N. Engl. J. Med. 348: 50–58 22. Burke W, Emery J. 2002. Genetics education for primary-care providers. Nat. Rev. Genet. 3:561–66

23. Rosenwald A, Wright G, Chan WC, et al. 2002. The use of molecular profiling to predict survival after chemotherapy for diffuse large-B-cell lymphoma. N. Engl. J. Med. 346:1937–47 24. Alizadeh AA, Eisen MB, Davis RE, et al. 2000. Distinct types of diffuse large B-cell lymphoma identified by gene expression profiling. Nature 403:503–11 25. Chang JC, Wooten EC, Tsimelzon A, et al. 2003. Gene expression profiling for the prediction of therapeutic response to docetaxel in patients with breast cancer. Lancet 362:362–69 26. Huang E, Cheng SH, Dressman H, et al. 2003. Gene expression predictors of breast cancer outcomes. Lancet 361: 1590–96 27. Chung CH, Bernard PS, Perou CM. 2002. Molecular portraits and the family tree of cancer. Nat. Genet. 32(Suppl.):533– 40 28. Branca M. 2003. Genetics and medicine. Putting gene arrays to the test. Science 300:238 29. Sarwal M, Chua MS, Kambham N, et al. 2003. Molecular heterogeneity in acute renal allograft rejection identified by DNA microarray profiling. N. Engl. J. Med. 349:125–38 30. DiMasi JA, Hansen RW, Grabowski HG. 2003. The price of innovation: new estimates of drug development costs. J. Health Econ. 22:151–85 31. Peakman T, Franks S, White C, Beggs M. 2003. Delivering the power of discovery in large pharmaceutical organizations. Drug Discov. Today 8:203–11 32. Lehman Brothers, McKinsey and Co. 2001. The fruits of genomics: drug pipelines face indigestion until the new biology ripens 33. Kubinyi H. 2003. Drug research: myths, hype and reality. Nat. Rev. Drug Discov. 2: 665–68 34. Gura T. 2002. Genomics. After the gold rush: gene firms reinvent themselves. Science 297:1982–84

12 Dec 2003

13:30

AR

AR206-ME55-01.tex

AR206-ME55-01.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:1-13. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

HUMAN GENOME SEQUENCE AND DRUG DEVELOPMENT 35. Noble D. 2003. Will genomics revolutionise pharmaceutical R&D? Trends Biotechnol. 21:333–37 36. Untch M, Ditsch N, Hermelink K. 2003. Immunotherapy: new options in breast cancer treatment. Expert Rev. Anticancer Ther. 3:403–8 37. O’Dwyer ME, Mauro MJ, Druker BJ. 2002. Recent advancements in the treatment of chronic myelogenous leukemia. Annu. Rev. Med. 53:369–81 38. Richardson PG, Barlogie B, Berenson J, et al. 2003. A phase 2 study of bortezomib in relapsed, refractory myeloma. N. Engl. J. Med. 348:2609–17 39. Gerhold DL, Jensen RV, Gullans SR. 2002. Better therapeutics through microarrays. Nat. Genet. 32(Suppl.):547–51 40. Hackett JL, Lesko LJ. 2003. Microarray data—the US FDA, industry and academia. Nat. Biotechnol. 21:742–43 41. Telenti A, Aubert V, Spertini F. 2002. Individualising HIV treatment—pharmacogenetics and immunogenetics. Lancet 359:722–23 42. Oscarson M. 2003. Pharmacogenetics of drug metabolising enzymes: importance for personalised medicine. Clin. Chem. Lab. Med. 41:573–80

P1: GBC

13

43. Drews J, Ryser S. 1997. The role of innovation in drug development. Nat. Biotechnol. 15:1318–19 44. Drews J. 1996. Genomic sciences and the medicine of tomorrow. Nat. Biotechnol. 14:1516–18 45. Hopkins AL, Groom CR. 2002. The druggable genome. Nat. Rev. Drug Discov. 1:727–30 46. Drews J. 2000. Drug discovery: a historical perspective. Science 287:1960–64 47. Alper J. 2003. Drug development. Biotech thinking comes to academic medical centers. Science 299:1303–5 48. Strausberg RL, Schreiber SL. 2003. From knowing to controlling: a path from genomics to drugs using small molecule probes. Science 300:294–95 48a. Zerhouni E. 2003. The NIH roadmap. Science 302:63–72 49. Alizadeh AA, Ross DT, Perou CM, van de Rijn M. 2001. Towards a novel classification of human malignancies based on gene expression patterns. J. Pathol. 195:41–52 50. Wagner KR. 2002. Genetic diseases of muscle. Neurol. Clin. 20:645–78 51. Guttmacher AE, Collins FS. 2003. Welcome to the genomic era. N. Engl. J. Med. 349:994–96

5 Jan 2004

11:14

AR

AR206-ME55-02.tex

AR206-ME55-02.sgm

LaTeX2e(2002/01/18) P1: GBC 10.1146/annurev.med.55.091902.103607

Annu. Rev. Med. 2004. 55:15–25 doi: 10.1146/annurev.med.55.091902.103607 First published online as a Review in Advance on Sept. 15, 2003

TOWARD ALZHEIMER THERAPIES BASED ∗ ON GENETIC KNOWLEDGE John Hardy Annu. Rev. Med. 2004.55:15-25. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Laboratory of Neurogenetics, National Institute on Aging, National Institutes of Health, Bethesda, Maryland 20892; email: [email protected]

Key Words animal models, cholesterol, anti-inflammatory, immunotherapy, amyloid ■ Abstract Genetic analysis has allowed the dissection of the pathogenic pathway that leads to Alzheimer’s disease. It has also been integral to the development of earlier and more accurate diagnostic practices. This analysis has identified many potential therapeutic targets, and clinical trials aimed at these targets are now under way. If these approaches are successful, it will be a spectacular validation of genetic-knowledge– based treatment strategies; if they are not, researchers will need to re-evaluate this approach toward understanding and developing strategies for treating complex diseases.

INTRODUCTION We live in the third golden age of the study of neurodegenerative diseases. The first was around the turn of the twentieth century, with the classification of these diseases as clinicopathological entities largely by physician scientists from the “German School.” The second occurred from about 1955 to 1980, when the elucidation of the mechanism of neurotransmission, along with the recognition of the imperfect cellular selective vulnerability of neurologic and psychiatric disease, spawned the explosion in transmitter pharmacology (most elegantly summarized in Reference 1) that underpins the current pharmaceutical industry. This, the third golden era, is based on the use of genetic discoveries, including but not limited to those made possible by the Human Genome Project, in order to determine the mechanisms of disease pathogenesis. Current therapies for Alzheimer’s disease (AD) and essentially all other neurodegenerative diseases are based on transmitter-replacement therapies that are conceptually based on the wish to replicate the miracle of L-dopa therapy in Parkinson’s disease. These therapies, such as cholinergic therapy for AD, are not aimed at altering the underlying disease pathogeneses, and like L-dopa, they probably ∗ The U.S. Government has the right to retain a nonexclusive, royalty-free license in and to any copyright covering this paper.

15

5 Jan 2004

11:14

Annu. Rev. Med. 2004.55:15-25. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

16

AR

AR206-ME55-02.tex

AR206-ME55-02.sgm

LaTeX2e(2002/01/18)

P1: GBC

HARDY

have little or no effect on this process. These therapies in general, therefore, suffer from three deficiencies: first, they allow the underlying disease process to continue unabated; second, they work only to the extent to which the disease has a selective vulnerability; and third, it is unreasonably optimistic to expect that coarsely turning up synaptic efficacy by (for example) inhibiting transmitter breakdown will ever satisfactorily mimic the exquisite control of synaptic transmission we believe operates in the undamaged brain. It is worth mentioning that these deficiencies are largely not addressed by “gene therapy” and stem cell approaches, whose philosophical bases are similar to those of conventional transmitter-based therapies. In AD and related diseases, genetic analysis has firmly implicated an outline of the pathways to cell death, the “amyloid cascade hypothesis” (1a). It has allowed the construction of useful animal models of the disease based on this hypothesis in which pathogenic mechanisms can be investigated and in which therapies can be tested, and it has allowed, in mutation carriers, the earliest symptoms and signs of the disease process to be tracked with the eventual aim of starting diseasemodifying treatments early. It has not, as yet, led to effective therapies. However, the hope is that the steps in the pathway will lead to points at which pharmacological intervention will become possible. Such strategies are currently under way. The purpose of this review is to summarize this progress.

THE AMYLOID CASCADE HYPOTHESIS The amyloid hypothesis has been recently reviewed (1a) and is only briefly summarized here. Mutations in three genes, the APP gene and the presenilin 1 and 2 genes, all lead to early-onset autosomal dominant AD (2). All of these mutations are believed to cause disease by altering APP processing such that Aβ deposition is more likely (3). Although the precise mechanism of the presenilin mutations’ action is unclear, presenilins appear to be the central component of the enzyme, γ -secretase, responsible for the C-terminal cleavage of Aβ, and the mutations alter the position of this cleavage (4, 5). These genetic findings imply that, in these rare Mendelian cases at least (6), Aβ is the initiating molecule. It is much less evident that the same cascade of events operates in typical, late-onset AD. Parsimony would suggest that it does, as well as three pieces of circumstantial evidence: First, the product of APOE, the only gene known to be involved in late-onset disease, is involved somehow in Aβ deposition (7); second, genetic variability in APP expression appears to contribute to the risk of late-onset AD (8) [and trisomy of the APP gene in Down’s syndrome is associatied with AD (9)]; and third, there appears to be genetic variability on chromosome 10 that both contributes to the risk of late-onset AD (10) and influences peripheral Aβ metabolism (11). Together, these findings suggest, but do not prove, that typical AD has a similar pathogenesis to the early-onset Mendelian forms, and we might expect other genes involved in the risk of the common forms of the disorder to be involved in Aβ metabolism (12). The strategy of using gene discovery in Mendelian forms of disease to develop new therapies is also being applied to other neurodegenerative diseases. Alpha-synuclein mutations are known to cause early-onset Parkinson’s

5 Jan 2004

11:14

AR

AR206-ME55-02.tex

AR206-ME55-02.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:15-25. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

ALZHEIMER THERAPIES

P1: GBC

17

Figure 2 The schematic relationship between APP/Aβ and tau in both Alzheimer’s disease and frontal temporal dementia linked to chromosome 17 (FTDP-17) suggests that tau dysfunction/tangle formation is downstream of Aβ in Alzheimer pathogenesis (12).

disease (13) and are clearly related to the pathogenesis of apparently sporadic cases (14, 15). Superoxide dismutase mutations in amyotrophic lateral sclerosis (16) are less clearly linked to the pathological entity generally. However, in both of these disorders, drug development programs are attempting to influence these processes (for examples of initiatives in this area, see http://grants.nih.gov/grants/guide/noticefiles/NOT-NS-03–004.html). Tangles, made of the tau protein (17), are an invariant feature of AD; Lewy bodies, which consist largely of α-synuclein, are a common feature of AD (18). Mutations in these genes lead to dementing disorders in which the pathology largely consists of the mutant protein (i.e., tangles and Lewy bodies, respectively) without Aβ (plaque) deposition, largely proving that these are downstream events in AD, but also suggesting that these pathologies are final common pathways for neurodegenerative cell loss [Figure 1 and Figure 2; see also References 19 and 20]. Experimental evidence has come from transgenic animals and is discussed below.

TRANSGENIC MOUSE MODELING OF ALZHEIMER PATHOGENESIS The production of transgenic mice with, first, single APP mutations (21), then both APP and PS1 mutations (22) and two APP mutations (23), has gradually led to the production of mice that develop florid and convincing neuritic plaques and have been extremely useful in the development of anti-Aβ therapies. However, these mice have rather subtle behavioral phenotypes (24), essentially no tau or synuclein pathology (25), and very little cell loss (26). Thus, they allow the mechanisms of plaque development to be studied and therapies aimed at this process to be tested, but they are not useful for testing strategies aimed at other events.

5 Jan 2004

11:14

Annu. Rev. Med. 2004.55:15-25. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

18

AR

AR206-ME55-02.tex

AR206-ME55-02.sgm

LaTeX2e(2002/01/18)

P1: GBC

HARDY

Mice with tau mutations develop tangles (27) and cell loss. Such mice are potentially useful for developing an understanding of the pathogenesis of this lesion and for testing therapies aimed at tangles. The available mice have considerable practical limitations, (short life expectancy, poor breeding and so on) but these are likely to be overcome in the near future. Remarkably, crossing mice having APP mutations with mice having tau mutations considerably potentiates the tau pathology (tangles) but has little or no effect on the Aβ (plaque) pathology. This observation provides experimental support for the suggestion that tau is downstream of Aβ in AD pathology (Figure 2) and also provides an experimental model in which that relationship can be experimentally investigated (28). Mice with α-synuclein transgenes have less impressive pathology than the corresponding tau mice (29), but, like the tau mice, when they are crossed with APP transgenic mice the pathology is potentiated (30). Thus, one can construct a hierarchical sequence of events between Aβ and α-synuclein similar to the sequence connecting Aβ and tau (Figure 3) It is interesting to note that these models may be more, rather than less, like the human situation, in which overlapping pathology is the rule rather than the exception (31). How much Lewy bodies contribute to dementia in “mixed” cases is unknown; however, therapies that target a common pathogenic pathway may be more effective than those tailored to a particular “disease” (32). Additional strategies to impact both vascular and amyloid pathologies may be very relevant to the development of clinical treatment. In fact, it is the latter overlap that makes current statin trials additionally hopeful (33; http://www.clinicaltrials.gov/ct/gui/show/NCT00046358?order = 20; see below).

Figure 3 A scheme similar to Figure 2, derived from the analysis of Lewybody disease, including the Lewy-body variant of Alzheimer’s disease (derived from Reference 13).

5 Jan 2004

11:14

AR

AR206-ME55-02.tex

AR206-ME55-02.sgm

LaTeX2e(2002/01/18)

ALZHEIMER THERAPIES

P1: GBC

19

Thus, combining mouse models allows us to study and modulate either Aβ deposition and plaque formation, or tangle formation, and/or allows investigation of the interaction between either Aβ and tau or Aβ and α-synuclein. Thus, although there is no one model of AD in mice, the separate stages of pathogenesis can now be targeted for detailed investigation and therapeutic modulation.

Annu. Rev. Med. 2004.55:15-25. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

IMPROVING DIAGNOSIS AND DEFINING DISEASE PROGRESSION Since the realization that AD was a common disorder of the elderly in addition to being a rare disease in the middle-aged (34), there has been a steady improvement in the accuracy of clinical diagnosis at both ends of the temporal spectrum of the disease. Clinical diagnosis has been refined using pathology as the gold standard (35); this has been correlated with find as well. In addition there has been an increasing ability to accurately identify those at a very early stage of disease, “mild cognitive impairment” (36). This progress is a necessary corollary to therapeutic strategies aimed at slowing disease pathogenesis, since obviously the intention is to maintain people at high levels of function rather than maintaining them in an impaired state. Although the progress in this area has been impressive, it would be preferable to be able to identify AD patients before the disease process has had any clinical effects. In order to identify presymptomatic changes in the very earliest stages of disease pathogenesis, the most efficient approach has been to follow prospectively those with APP and presenilin mutations in the years before they reach the characteristic age of onset for their families. As the gene discovery in patients with Mendelian causes of disease helped reveal the general principles of pathogenesis, similarly, studying the clinical course in patients with these rare Mendelian disorders will allow us to extend presymptomatic marker discovery into general clinical diagnostic use. To date, this approach has led to the identification of hippocampal atrophy as the earliest abnormality that distinguishes those who will develop AD (37); this atrophy begins 2–3 years before the earliest, very subtle, clinical signs of disease. Although this study needs to be validated in more typical AD, it demonstrates that preclinical identification of people who will develop disease is technically possible. It is likely that early diagnosis will be the key to successful treatment, especially for treatments aimed at preventing and minimizing, rather than reversing, the disease process.

EXPERIMENTAL THERAPIES BASED ON THE AMYLOID CASCADE HYPOTHESIS If AD represents the effects of a chronic imbalance between Aβ production and Aβ clearance, and this imbalance can be caused by numerous distinct initiating factors, how should we treat and prevent the disorder? Five broad strategies have been put forward.

5 Jan 2004

11:14

Annu. Rev. Med. 2004.55:15-25. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

20

AR

AR206-ME55-02.tex

AR206-ME55-02.sgm

LaTeX2e(2002/01/18)

P1: GBC

HARDY

First, one could attempt to partially inhibit either of the two proteases, β- and γ secretase, that generate Aβ from APP (Figure 1). In the case of the former enzyme, compound screening and medicinal chemistry are being vigorously pursued to identify potent small-molecule inhibitors that can fit the large active site of this aspartyl protease and still penetrate the blood-brain barrier. In the case of the latter, potent membrane-permeable inhibitors are already in hand. However, their testing in humans has barely been attempted because of the theoretical concern that most such compounds might interfere importantly with signaling by Notch proteins and other cell-surface receptors, which would probably lead to serious disorders (38). Second, one could attempt to prevent the oligomerization of Aβ or enhance its clearance from the cortex. This approach is exemplified by the use of active or passive Aβ immunization, in which antibodies to Aβ decrease cerebral levels of the peptide by promoting microglial clearance (39, 40) and/or by redistributing the peptide from the brain to the systemic circulation (41). Active immunization with synthetic Aβ1-42 peptide produces robust benefits in APP transgenic mice without detectable toxicity. However, when this approach was extended to AD patients, a small but unacceptable fraction of the study subjects developed a usually transient nervous-system inflammatory reaction that was fatal in a few cases (42), precluding further testing with this preparation. Those who were enrolled in the clinical trial surely had a significant plaque and tangle burden; it is possible that if this treatment were possible very early, even presymptomatically, this unacceptable adverse effect could be avoidable. An interesting example of worsening when toxic load is leached from the nervous system can be seen in the treatment of Wilson’s disease, in which copper accumulates in the nervous system as well as in the liver and other organs due to abnormalities of copper transport. When first treated, those with nervous-system involvement often worsen before they improve; the copper that was inactive in the nervous system is freed by the treatment and transiently increases before the new baseline of low copper is manifest as a result of chronic treatment (43). Analysis of a subset of Aβ-immunized patients suggested that the treatment may have slowed disease progression (44), although definitive conclusions on this await the report of the analysis of the whole trial (45). Several alternative preparations intended to provide Aβ antibodies by either active or passive routes have been formulated (see, e.g., Reference 46), and one or more of these is likely to reach clinical testing before long. The third broad approach to the treatment of AD is anti-inflammatory. This strategy is based on evidence that a cellular inflammatory response in the cortex is elicited by the progressive accumulation of Aβ (47). Additionally, it was recently shown that some anti-inflammatory drugs may have direct effects on the cleavage of APP by γ -secretase, independent of their inhibition of cyclooxygenase 2 and other inflammatory mediators (48). Some such drugs have been shown to reduce cytopathology in APP transgenic mice (49, 50). Clinical trials of

5 Jan 2004

11:14

AR

AR206-ME55-02.tex

AR206-ME55-02.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:15-25. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

ALZHEIMER THERAPIES

P1: GBC

21

compounds based on these findings are currently under way (see, e.g., http://www. clinicaltrials.gov/ct/gui/show/NCT00046358?order = 20). It is worth noting that the recently reported trial of naproxen that failed to show benefit in AD (50a) is not directly interpretable because although naproxen is a potent cyclo-oxygenase inhibitor, it does not affect APP processing (48). Thus, this trial will have to be repeated with a different compound (48–50). The fourth approach is based on modulating cholesterol homeostasis (51). Chronic use of cholesterol-lowering drugs has recently been associated with a lower incidence of AD (52, 53). Concurrently, high-cholesterol diets have been shown to increase Aβ pathology in animals (54, 55), and cholesterol-lowering drugs have been shown to reduce pathology in APP transgenic mice (56). These effects seem to be caused by a direct (though poorly understood) effect of cholesterol on APP processing (57, 58). A particular advantage of this approach is that statin drugs are generally well-tolerated and have already been widely prescribed. Again, clinical trials are under way. The fifth approach is based on the surprising observation that Aβ aggregation partially dependent on Cu2+ and Zn2+ ions (59). This has led to the suggestion that chelation of these ions may have therapeutic potential. Clinical trials of such a chelating agent, clioquinol, which has been used successfully in the transgenic model (60), are currently under way.

CONCLUSIONS Clinical AD research has reached a pivotal point. There is a general, though not universal, belief that we now have a skeletal understanding of disease pathogenesis. As a field, we have greatly improved our diagnosis of the disease and have learned how to assess its severity and rate of progress. Thus, we appear to have all the technologies and tools necessary to test whether a particular treatment is showing efficacy. As outlined above, we have at least five different approaches that target different aspects of this crucial central pathogenic pathway. Clinical trials are by nature slow and complex, and side effects may be difficult to predict. However, it is hoped that one or more of these five approaches, either alone or in combination, will yield clinical benefit. Such an outcome would validate the enormous amount of basic work done on the molecular genetics and molecular biology of the disease. This in turn would further focus drug-discovery efforts so that further rapid clinical advances would be expected. Such translational research is necessarily reiterative, as improvements in diagnosis and understanding of pathogenesis help to refine each other. In a general sense, successful treatments for AD would also validate the more general concept of pathogenic-knowledge–based therapeutics. The current “third golden age” would have clinical relevance analogous to that conferred on basic discoveries of neurotransmitters by the success of L-dopa therapy (61).

5 Jan 2004

11:14

22

AR

AR206-ME55-02.tex

AR206-ME55-02.sgm

LaTeX2e(2002/01/18)

P1: GBC

HARDY

The Annual Review of Medicine is online at http://med.annualreviews.org

Annu. Rev. Med. 2004.55:15-25. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

LITERATURE CITED 1. Cooper JR, Bloom FE, Roth RH. 2003. The Biochemical Basis of Neuropharmacology. Oxford, UK: Oxford Univ. Press 1a. Hardy J, Selkoe DJ. 2002. The amyloid hypothesis of Alzheimer’s disease: progress and problems on the road to therapeutics. Science 297:353–56 2. Cruts M, Van Broeckhoven C. 1998. Molecular genetics of Alzheimer’s disease. Ann. Med. 30:560–65 3. Scheuner D, Eckman C, Jensen M, et al. 1996. Secreted amyloid β-protein similar to that in the senile plaques of Alzheimer’s disease is increased in vivo by the presenilin 1 and 2 and APP mutations linked to familial Alzheimer’s disease. Nat. Med. 2: 864–70 4. De Strooper B, Saftig P, Craessaerts K, et al. 1998. Deficiency of presenilin-1 inhibits the normal cleavage of amyloid precursor protein. Nature 391:387–90 5. Wolfe MS, Xia W, Ostaszewski BL, et al. 1999. Two transmembrane aspartates in presenilin-1 required for presenilin endoproteolysis and gamma-secretase activity. Nature 398(6727):513–17 6. Rogaeva E. 2002. The solved unsolved mysteries of the genetics of early-onset Alzheimer’s disease. Neuromol. Med. 2: 1–10 7. Bales KR, Verina T, Cummins DJ, et al. 1999. Apolipoprotein E is essential for amyloid deposition in the APP(V717F) transgenic mouse model of Alzheimer’s disease. Proc. Natl. Acad. Sci. 96:15233– 38 8. Wavrant-De Vrieze F, Crook R, Holmans P, et al. 1999. Genetic variability at the amyloid-beta precursor protein locus may contribute to the risk of lateonset Alzheimer’s disease. Neurosci. Lett. 269:67–70 9. Prasher VP, Farrer MJ, Kessling AM, et al. 1998. Molecular mapping of Alzheimer-

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

type dementia in Down’s syndrome. Ann. Neurol. 43:380–83 Myers A, Holmans P, Marshall H, et al. 2000. Susceptibility locus for Alzheimer’s disease on chromosome 10. Science 290:2304–5 Ertekin-Taner N, Graff-Radford N, Younkin LH, et al. 2000. Linkage of plasma Aβ42] to a quantitative locus on chromosome 10 in late-onset Alzheimer’s disease pedigrees. Science 290:2303–4 Tanzi RE, Bertram L. 2001. New frontiers in Alzheimer’s disease genetics. Neuron 32:181–84 Polymeropoulos MH, Lavedan C, Leroy E, et al. 1997. Mutation in the alphasynuclein gene identified in families with Parkinson’s disease. Science 276:2045– 47 Spillantini MG, Schmidt ML, Lee VM, et al. 1997. Alpha-synuclein in Lewy bodies. Nature 388:839–40 Farrer M, Maraganore DM, Lockhart P, et al. 2001. Alpha-synuclein gene haplotypes are associated with Parkinson’s disease. Hum. Mol. Genet. 10:1847–51 Rosen DR, Siddique T, Patterson D, et al. 1993. Mutations in Cu/Zn superoxide dismutase gene are associated with familial amyotrophic lateral sclerosis. Nature 362:59–62 Goedert M, Wischik CM, Crowther RA, et al. 1988. Cloning sequencing of the cDNA encoding a core protein of the paired helical filament of Alzheimer disease: identification as the microtubuleassociated protein tau. Proc. Natl. Acad. Sci. 85:4051–55 Hansen LA, Masliah E, Galasko D, et al. 1993. Plaque-only Alzheimer disease is usually the Lewy body variant, and vice versa. J. Neuropathol. Exp. Neurol. 52:648–54 Hardy J, Duff K, Gwinn-Hardy K, et al.

5 Jan 2004

11:14

AR

AR206-ME55-02.tex

AR206-ME55-02.sgm

LaTeX2e(2002/01/18)

ALZHEIMER THERAPIES

20.

Annu. Rev. Med. 2004.55:15-25. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

21.

22.

23.

24.

25.

26.

27.

28.

29.

1998. Genetic dissection of Alzheimer’s disease and related dementias: amyloid and its relationship to tau. Nat. Neurosci. 1:95–99 Hardy J. 2003. The relationship between amyloid and tau. J. Mol. Neurosci. 20:203–6 Games D, Adams D, Alessandrini R, et al. 1995. Alzheimer-type neuropathology in transgenic mice overexpressing V717F beta-amyloid precursor protein. Nature 373:523–27 Duff K, Eckman C, Zehr C, et al. 1996. Increased amyloid β42(43) in brains of mice expressing mutant presenilin 1. Nature 383:710–13 Chishti MA, Yang DS, Janus C, et al. 2001. Early-onset amyloid deposition and cognitive deficits in transgenic mice expressing a double mutant form of amyloid precursor protein 695. J. Biol. Chem. 276:21562–70 Morgan D, Diamond DM, Gottschall P, et al. 2000. Vaccination with Aβ peptide prevents memory deficits in an animal model of Alzheimer’s disease. Nature 408:982–85 Masliah E, Sisk A, Mallory M, et al. 2001. Neurofibrillary pathology in transgenic mice overexpressing V717F beta-amyloid precursor protein. J. Neuropathol. Exp. Neurol. 60:357–68 Irizarry MC, Soriano F, McNamara M, et al. 1997. Aβ deposition is associated with neuropil changes, but not with overt neuronal loss in the human amyloid precursor protein V717F (PDAPP) transgenic mouse. J. Neurosci. 17:7053–59 Lewis J, McGowan E, Rockwood J, et al. 2000. Neurofibrillary tangles, amyotrophy and progressive motor disturbance in mice expressing mutant (P301L) tau protein. Nat. Genet. 25:402–5 Lewis J, Dickson DW, Lin WL, et al. 2001. Enhanced neurofibrillary degeneration in transgenic mice expressing mutant tau APP. Science 293:1487–91 Masliah E, Rockenstein E, Veinbergs I,

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

P1: GBC

23

et al. 2000. Dopaminergic loss and inclusion body formation in alpha-synuclein mice: implications for neurodegenerative disorders. Science 287:1265–69 Masliah E, Rockenstein E, Veinbergs I, et al. 2001. Beta-amyloid peptides enhance alpha-synuclein accumulation and neuronal deficits in a transgenic mouse model linking Alzheimer’s disease and Parkinson’s disease. Proc. Natl. Acad. Sci. 98:12245–50 Barker WW, Luis CA, Kashuba A, et al. 2002. Relative frequencies of Alzheimer disease, Lewy body, vascular and frontotemporal dementia, and hippocampal sclerosis in the State of Florida Brain Bank. Alzheimer Dis. Assoc Disord. 16:203–12 Hardy J, Gwinn-Hardy K. 1999. Neurodegenerative disease: a different view of diagnosis. Mol. Med. Today 5:514–17 Kalaria RN, Ballard C. 1999. Overlap between pathology of Alzheimer disease and vascular dementia. Alzheimer Dis. Assoc. Disord. 13(Suppl. 3):S115–23 Tomlinson BE, Blessed G, Roth M. 1970. Observations on the brains of demented old people. J. Neurol. Sci. 11(3):205–42 Mirra SS. 1997. Neuropathological assessment of Alzheimer’s disease: the experience of the Consortium to Establish a Registry for Alzheimer’s Disease. Int. Psychogeriatr. 9(Suppl. 1):263–68 Petersen RC, Stevens JC, Ganguli M, et al. 2001. Practice parameter: early detection of dementia: mild cognitive impairment (an evidence-based review). Report of the Quality Standards Subcommittee of the American Academy of Neurology. Neurology 56:1133–42 Schott JM, Fox NC, Frost C, et al. 2003. Assessing the onset of structural change in familial Alzheimer’s disease. Ann. Neurol. 53:181–88 Haass C, De Strooper B. 1999. The presenilins in Alzheimer’s disease—proteolysis holds the key. Science 286:916–19 Schenk D, Barbour R, Dunn W, et al.

5 Jan 2004

11:14

24

40.

Annu. Rev. Med. 2004.55:15-25. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

41.

42.

43.

44.

45.

46.

47.

48.

49.

AR

AR206-ME55-02.tex

AR206-ME55-02.sgm

LaTeX2e(2002/01/18)

P1: GBC

HARDY 1999. Immunization with amyloid-beta attenuates Alzheimer-disease-like pathology in the PDAPP mouse. Nature 400: 173–77 Bard F, Cannon C, Barbour R, et al. 2000. Peripherally administered antibodies against amyloid beta-peptide enter the central nervous system and reduce pathology in a mouse model of Alzheimer disease. Nat. Med. 6:916–19 DeMattos RB, Bales KR, Cummins DJ, et al. 2002. Brain to plasma amyloid-beta efflux: a measure of brain amyloid burden in a mouse model of Alzheimer’s disease. Science 295:2264–67 Nicoll JA, Wilkinson D, Holmes C, et al. 2003. Neuropathology of human Alzheimer disease after immunization with amyloid-beta peptide: a case report. Nat. Med. 9:448–52 Stuerenburg HJ. 2000. CSF copper concentrations, blood-brain barrier function, and coeruloplasmin synthesis during the treatment of Wilson’s disease. J. Neural. Transm. 107(3):321–29 Hock C, Konietzko U, Streffer JR, et al. 2003. Antibodies against beta-amyloid slow cognitive decline in Alzheimer’s disease. Neuron 38:547–54 Winblad B, Blum KI. 2003. Hints of a therapeutic vaccine for Alzheimer’s? Neuron 38:517–18 Weiner HL, Lemere CA, Maron R, et al. 2000. Nasal administration of amyloidbeta peptide decreases cerebral amyloid burden in a mouse model of Alzheimer’s disease. Ann. Neurol .48:567–79 Rogers J, Strohmeyer R, Kovelowski CJ, et al. 2002. Microglia inflammatory mechanisms in the clearance of amyloid beta peptide. GLIA 40(2):260–69 Weggen S, Eriksen JL, Sagi SA, et al. 2003. Evidence that nonsteroidal anti-inflammatory drugs decrease Aβ42 production by direct modulation of gamma-secretase activity. J. Biol. Chem. 278:31831–37 Lim GP, Yang F, Chu T, et al. 2001.

50.

50a.

51.

52.

53.

54.

55.

56.

57.

Ibuprofen effects on Alzheimer pathology and open field activity in APPsw transgenic mice. Neurobiol. Aging 22:983–91 Jantzen PT, Connor KE, DiCarlo G, et al. 2002. Microglial activation and betaamyloid deposit reduction caused by a nitric oxide-releasing nonsteroidal antiinflammatory drug in amyloid precursor protein plus presenilin-1 transgenic mice. J. Neurosci. 15:2246–54 Aisen PS, Schafer KA, Grundman M, et al. 2003. Effects of rofecoxib or naproxen vs placebo on Alzheimer disease progression: a randomized controlled trial. Alzheimer’s Disease Cooperative Study. JAMA 289(21):2819–26 Burns M, Duff K. 2003. Use of in vivo models to study the role of cholesterol in the etiology of Alzheimer’s disease. Neurochem. Res. 28:979–86 Wolozin B, Kellman W, Ruosseau P, et al. 2000. Decreased prevalence of Alzheimer disease associated with 3-hydroxy-3methyglutaryl coenzyme A reductase inhibitors. Arch. Neurol. 57:1439–43 Jick H, Zornberg GL, Jick SS, et al. 2000. Statins and the risk of dementia. Lancet 356:1627–31 Sparks DL, Martin TA, Gross DR, et al. 2000. Link between heart disease, cholesterol, and Alzheimer’s disease: a review. Microsc. Res. Tech. 50:287–90 Refolo LM, Pappolla MA, LaFrancois J, et al. 2001. A cholesterol-lowering drug reduces beta-amyloid pathology in a transgenic mouse model of Alzheimer’s disease. Neurobiol. Dis. 8(5):890–99 Refolo LM, Malester B, LaFrancois J, et al. 2000. Hypercholesterolemia accelerates the Alzheimer’s amyloid pathology in a transgenic mouse model. Neurobiol. Dis. 7:321–31 Fassbender K, Simons M, Bergmann C, et al. 2001. Simvastatin strongly reduces levels of Alzheimer’s disease beta-amyloid peptides Aβ42 and Aβ40 in vitro and in vivo. Proc. Natl. Acad. Sci. 98:5856–61

5 Jan 2004

11:14

AR

AR206-ME55-02.tex

AR206-ME55-02.sgm

LaTeX2e(2002/01/18)

ALZHEIMER THERAPIES

Annu. Rev. Med. 2004.55:15-25. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

58. Wahrle S, Das P, Nyborg AC, et al. 2002. Cholesterol-dependent gamma-secretase activity in buoyant cholesterol-rich membrane microdomains. Neurobiol. Dis. 9:11–23 59. Bush AI, Pettingell WH, Multhaup G, et al. 1994. Rapid induction of Alzheimer A beta amyloid formation by zinc. Science 265:1464–67

P1: GBC

25

60. Cherny RA, Atwood CS, Xilinas ME, et al. 2001. Treatment with a copperzinc chelator markedly and rapidly inhibits beta-amyloid accumulation in Alzheimer’s disease transgenic mice. Neuron 30:665–76 61. Hornykiewicz O. 2002. Dopamine miracle: from brain homogenate to dopamine replacement. Mov. Disord. 17:501–8

HARDY.qxd

1/5/2004

3:25 PM

Page 1

Annu. Rev. Med. 2004.55:15-25. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

ALZHEIMER THERAPIES

C-1

Figure 1 The overall structure of the APP protein showing the Aβ segment of the molecule in red (left) and showing the position of the mutations relative to the specific processing sites of the APP molecule (right). On the right, red circles denote pathogenic mutations; orange circles represent variants that are probably not pathogenic.

2 Dec 2003

22:30

AR

AR206-ME55-03.tex

AR206-ME55-03.sgm

LaTeX2e(2002/01/18) P1: GBC 10.1146/annurev.med.55.091902.103843

Annu. Rev. Med. 2004. 55:27–39 doi: 10.1146/annurev.med.55.091902.103843 First published online as a Review in Advance on Aug. 14, 2003

INHERITED DISEASES INVOLVING G PROTEINS ∗ AND G PROTEIN–COUPLED RECEPTORS Allen M. Spiegel and Lee S. Weinstein Annu. Rev. Med. 2004.55:27-39. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, Bethesda, Maryland 20892; email: [email protected]

Key Words signal transduction, gain- and loss-of-function mutations, hormone resistance ■ Abstract Heterotrimeric G proteins couple seven-transmembrane receptors for diverse extracellular signals to effectors that generate intracellular signals altering cell function. Mutations in the gene encoding the α subunit of the G protein–coupling receptors to stimulation of adenylyl cyclase cause developmental abnormalities of bone, as well as hormone resistance (pseudohypoparathyroidism caused by loss-offunction mutations) and hormone hypersecretion (McCune-Albright syndrome caused by gain-of-function mutations). Loss- and gain-of-function mutations in genes encoding G protein–coupled receptors (GPCRs) have been identified as the cause of an increasing number of retinal, endocrine, metabolic, and developmental disorders. GPCRs comprise an evolutionarily conserved gene superfamily (1). By coupling to heterotrimeric G proteins, GPCRs transduce a wide variety of extracellular signals including monoamine, amino acid, and nucleoside neurotransmitters, as well as photons, chemical odorants, divalent cations, hormones, lipids, peptides and proteins. Following a brief overview of G protein–coupled signal transduction, we review the growing body of evidence that mutations in genes encoding GPCRs and G proteins are an important cause of human disease.

OVERVIEW OF G PROTEIN–COUPLED SIGNAL TRANSDUCTION G Protein–Coupled Receptors All members of the GPCR superfamily share a common structural feature: seven membrane-spanning helices connected by three intracellular loops and three extracellular loops with an extracellular amino terminus and an intracellular carboxy terminus (2). This basic structure has now been verified by X-ray crystallography for rhodopsin (3). Superimposed on the basic structure of GPCRs are a number of ∗ The U.S. Government has the right to retain a nonexclusive, royalty-free license in and to any copyright covering this paper.

27

2 Dec 2003

22:30

Annu. Rev. Med. 2004.55:27-39. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

28

AR

AR206-ME55-03.tex

SPIEGEL

¥

AR206-ME55-03.sgm

LaTeX2e(2002/01/18)

P1: GBC

WEINSTEIN

variations relevant to differences in ligand binding, G protein coupling, and interaction with other proteins. Sequence alignment, especially of the transmembrane helices, allows one to divide the superfamily into subfamilies (1, 4). Of these, family 1 is the largest and includes opsins, odorant receptors, and receptors for monoamines, purines, opiates, chemokines, some small peptide hormones, and the large glycoprotein hormones, thyroid stimulating hormone (TSH), luteinizing hormone (LH), and follicle-stimulating hormone (FSH). Family 2 shows essentially no sequence homology to family 1, even within the transmembrane helices. Members include receptors for peptide hormones, such as parathyroid hormone (PTH), parathyroid hormone–related protein (PTHrP), and calcitonin. Family 3 members include the metabotropic glutamate receptors, an extracellular Ca2+-sensing receptor, and putative taste and pheromone receptors. Family 3 GPCRs form dimers, and this may be true of other GPCRs as well. A general model of GPCR activation postulates that GPCRs are in equilibrium between an activated state and an inactive state (5). These states presumptively differ in the disposition of the transmembrane helices, and in turn, the cytoplasmic domains that determine G protein coupling. Agonists, according to this model, stabilize the activated state. For some family 1 GPCRs, agonists activate the receptor by binding directly within the seven-transmembrane domain and altering the disposition of the helices. For other GPCRs in families 1, 2, and 3, agonist binding involves portions of the receptor’s extracellular domain, and the mechanism whereby this signal is transmitted to the seven-transmembrane domain remains to be clarified. All GPCRs act as guanine nucleotide exchange factors (6). In their activated (agonist-bound) conformation, they catalyze exchange of guanosine diphosphate (GDP) tightly bound to the α subunit of heterotrimeric G proteins for guanosine triphosphate (GTP) (Figure 1). This in turn leads to activation of the α subunit and its dissociation from the G protein βγ dimer. Both G protein subunits are capable of regulating effector activity. Identified G protein–regulated effectors include enzymes of second-messenger metabolism such as adenylyl cyclase and phospholipase C-β, and a variety of ion channels. Agonist binding to GPCRs thus alters intracellular second-messenger and ion concentrations with resultant rapid effects on hormone secretion, muscle contraction, and a variety of other physiologic functions. Longer-term changes in gene expression are also seen as a result of second messenger–mediated phosphorylation of transcription factors.

G Proteins The G protein subunits are encoded by distinct genes (7). The α subunit binds guanine nucleotides with high affinity and specificity and possesses intrinsic GTPase activity. The β and γ polypeptides are tightly but noncovalently associated in a functional dimer subunit. There is considerable diversity in G protein subunits, with multiple genes encoding all three subunits. There are at least 16 distinct α subunit genes in mammals. These vary widely in range of expression. Some, such as Gsα, which couples many GPCRs to stimulation of adenylyl cyclase and cAMP formation (Figure 1), are ubiquitous; others, such as Gt1α (transducin),

2 Dec 2003

22:30

AR

AR206-ME55-03.tex

AR206-ME55-03.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:27-39. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

G PROTEIN AND RECEPTOR DISORDERS

P1: GBC

29

Figure 1 The G protein GTPase cycle (described in text). Potential sites for disease-causing abnormalities are numbered. In each panel, the stippled region denotes the plasma membrane with extracellular above and intracellular below. Under physiologic conditions, effector regulation by G protein subunits is transient and is terminated by the GTPase activity of the α subunit. The latter converts bound GTP to GDP, thus returning the α subunit to its inactivated state with high affinity for the βγ dimer, which reassociates to form the heterotrimer. The figure shows the G protein Gs with its effector, adenylyl cyclase. Activation of adenylyl cyclase generates the intracellular second messenger, cAMP, which activates protein kinase A (PKA). The latter enzyme phosphorylates a variety of proteins that mediate the physiologic effects of agonists for Gs-coupled receptors. Cholera toxin covalently modifies the Gsα subunit blocking its GTPase activity. Somatic mutations of the Gsα subunit likewise block GTPase activity. In both cases, constitutive activation and agonist-independent cAMP formation result.

which couples the GPCR rhodopsin to cGMP phosphodiesterase in retinal rod photoreceptor cells, are highly localized. There are 5 distinct mammalian β subunit genes and at least 11 γ subunit genes. Within any given cell, multiple distinct GPCRs, G proteins, and effectors are expressed, but there is relative specificity in G protein coupling to GPCRs and to effectors based on unique sequence determinants of the respective components (8).

2 Dec 2003

22:30

30

AR

AR206-ME55-03.tex

SPIEGEL

¥

AR206-ME55-03.sgm

LaTeX2e(2002/01/18)

P1: GBC

WEINSTEIN

Annu. Rev. Med. 2004.55:27-39. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

GENERAL FEATURES OF DISEASES CAUSED BY MUTATIONS IN G PROTEIN AND GPCR GENES Mutations in genes encoding GPCRs and G proteins can cause loss of function by impairing any of several steps in the normal GPCR/GTPase cycle (Figure 1) (9). G protein and GPCR loss-of-function mutations block signaling in response to the corresponding agonist(s). Gain-of-function mutations in either GPCRs or G proteins lead to constitutive, agonist-independent activation of signaling. In endocrine signaling, loss-of-function mutations cause hormone resistance, mimicking hormone deficiency, whereas gain-of-function mutations mimic states of hormone excess. Defective G protein–mediated signaling can also lead to neoplasia and developmental and sensory abnormalities. The phenotype caused by GPCR and G protein mutations depends on the range of expression of the involved gene and on whether the mutation is somatic or germline. For germline mutations, the phenotype will be pleiotropic for widely expressed genes and more focal for genes expressed more narrowly. In contrast, somatic mutation of even a ubiquitously expressed gene can cause focal manifestations.

G PROTEIN GENE DISORDERS To date, mutation of two Gα subunits (transducin and Gsα) have been associated with human disease. The Nougaret form of autosomal dominant stationary night blindness is associated with a transducin mutation that uncouples it from its effector (10). No Gβ or Gγ mutations have been associated with monogenic human disorders, but a polymorphism of the β 3 subunit has been implicated in several common multigenic disorders.

Activating Gsα Mutations Activating Gsα mutations encode substitutions of either Arg201 or Gln227—two residues that are critical for the GTPase reaction—and lead to constitutive activation by disrupting the intrinsic GTPase activity, which prolongs the active state (Figure 1). These dominant, somatic mutations are present in ∼40% of growth hormone–secreting pituitary adenomas (11) and infrequently in nonsecreting and adrenocorticotropic hormone (ACTH)–secreting pituitary tumors, as well as in thyroid, parathyroid, and adrenal tumors (11, 12). In growth hormone–secreting tumors, the mutation is almost always in the maternal allele, presumably because Gsα is expressed almost exclusively from the maternal allele in pituitary cells (13). Somatic Gsα Arg201 mutations are also present in fibrous dysplasia (FD) of bone (14) and in a more widespread tissue distribution in the McCune-Albright syndrome (MAS) (15, 16). MAS is classically defined by the triad of polyostotic FD, caf´e-au-lait skin lesions, and gonadotropin-independent sexual precocity (15), although these patients may also develop tumors (or nodular hyperplasia)

2 Dec 2003

22:30

AR

AR206-ME55-03.tex

AR206-ME55-03.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:27-39. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

G PROTEIN AND RECEPTOR DISORDERS

P1: GBC

31

of pituitary somatotrophs, thyroid, or adrenal cortex with associated hormonal oversecretion and other nonendocrine abnormalities (e.g., cardiomyopathy, sudden death, liver abnormalities) (16). It is believed that the somatic mutation in MAS patients occurs early in development, and therefore the clinical spectrum in each individual is determined by the tissue distribution of mutant-bearing cells (15). Although MAS is generally not inherited, presumably because Gsα-activating mutations are lethal in the germline, one patient with severe skeletal, endocrine, and developmental abnormalities was reported to possibly have a germline Arg201 Leu mutation (17). Gsα-activating mutations have also been identified in intramuscular myxomas, both those that occur alone and those that present with accompanying FD (Mazabraud syndrome) (18). Most, if not all, of the clinical manifestations associated with Gsα-activating mutations result from increased intracellular cAMP level. In many endocrine organs, growth and hormone secretion are stimulated by trophic hormones that activate Gs/cAMP pathways, and constitutive Gs activation leads to endocrine tumors and hormonal hypersecretion in the absence of circulating trophic hormones. FD is a focal bone lesion consisting of immature mesenchymal cells with interspersed woven bone spicules and cartilaginous islands. This lesion results from excess cAMP within bone marrow stromal cells, which stimulates their proliferation and alters their differentiation (19). The hypophosphatemia often associated with MAS appears to result from hypophosphatemic factors secreted from FD lesions (20). Hyperpigmentation in MAS results from excess cAMP in melanocytes, which stimulates the expression of tyrosinase, the rate-limiting enzyme for melanin production (21). The cardiac manifestations occasionally associated with MAS likely reflect the effects of Gs/cAMP overstimulation in cardiomyocytes.

Inactivating Gsα Mutations ALBRIGHT HEREDITARY OSTEODYSTROPHY Patients who inherit a heterozygous Gsα null mutation develop Albright hereditary osteodystrophy (AHO), a syndrome characterized by one or more of the following clinical features: short stature, brachydactyly, subcutaneous ossifications, centripetal obesity, depressed nasal bridge, hypertelorism, and mental or developmental deficits (22, 23). The severity of the AHO phenotype is variable, and some patients with Gsα mutations have few or no symptoms. Although the ectopic ossifications in AHO are generally subcutaneous and limited, some patients develop progressive osseous heteroplasia, a severe form of ossification that invades the deep tissues (24). PSEUDOHYPOPARATHYROIDISM TYPE 1a In addition to the AHO phenotype, patients who inherit Gsα mutations from their mother also develop resistance to various hormones (PTH, TSH, LH, and FSH) that stimulate Gs/cAMP in their target tissues, a condition referred to as pseudohypoparathyroidism type 1a (PHP1a) (22, 23). In contrast, patients who inherit the same mutations from their father develop only the AHO phenotype, a condition also referred to as

2 Dec 2003

22:30

Annu. Rev. Med. 2004.55:27-39. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

32

AR

AR206-ME55-03.tex

SPIEGEL

¥

AR206-ME55-03.sgm

LaTeX2e(2002/01/18)

P1: GBC

WEINSTEIN

pseudopseudohypoparathyroidism (PPHP). This is because Gsα is imprinted in a tissue-specific manner. Although Gsα is biallelically expressed in most tissues, it is expressed primarily from the maternal allele in various hormonal target tissues, including renal proximal tubule, thyroid, and pituitary (13, 25). Null mutations in the active maternal allele lead to Gsα deficiency and hormone resistance, whereas mutation of the inactive paternal allele has little effect on Gsα expression or hormonal signaling. In fact, the Gsα gene GNAS at 20q13 has multiple gene products owing to the use of alternative promoters that are also imprinted (22). In most other tissues, Gsα is not imprinted, and therefore its expression is similarly reduced by ∼50% in both PHP1a and PPHP (23). Gsα haploinsufficiency probably leads to the AHO phenotype (22). One mutation (Ala366Ser) produces both PHP1a and gonadotropin-independent precocious puberty in males. This mutation leads to increased GDP dissociation, which at core body temperature denatures the protein, resulting in PHP1a. At the lower temperature of the testis, the mutant protein is stable but is activated because it can bind GTP in the absence of receptor stimulation (26).

Gsα Imprinting Defects: Pseudohypoparathyroidism Type 1b Patients with pseudohypoparathyroidism type 1b (PHP1b) have renal PTH resistance in the absence of AHO or resistance to other hormones (except mild TSH resistance in some cases). Most cases of PHP1b are sporadic, but it can be familial. In these families, PTH resistance only occurs when the disease is inherited maternally, similar to the parental inheritance pattern of PTH resistance within AHO kindreds (27). Familial PHP1b has been mapped to 20q13 in the vicinity of GNAS (27). However, erythrocyte Gs function is normal in PHP1b patients (23), ruling out Gsα null mutations. Within GNAS is an imprinted region (the exon 1A region) located just upstream of the Gsα promoter in which the DNA is methylated on the maternal allele (28). In virtually all PHP1b patients, the maternal-specific methylation of exon 1A is absent, consistent with failure to establish a maternal imprint in the oocyte (28). In one case, PHP1b was associated with paternal uniparental disomy of chromosome 20, which results in the same imprinting pattern (29). Familial PHP1b presumably results from mutations that disrupt the establishment of the maternal imprint, although no specific mutations have been identified to date (27, 29). In PHP1b, both GNAS alleles have a paternal-specific imprinting pattern. In renal proximal tubules, this probably results in Gsα deficiency and PTH resistance because Gsα is normally expressed primarily from the maternal allele (22). In most other tissues, Gsα expression would be unaffected by the imprinting defect because in these tissues it is normally expressed equally from both parental alleles. This would explain the lack of Gsα haploinsufficiency in erythrocytes or the AHO phenotype in PHP1b patients (23). The exon 1A region appears to be important for the tissue-specific imprinting of Gsα, although the specific mechanisms have not been delineated (22, 28). In one kindred, three affected siblings had a Gsα

2 Dec 2003

22:30

AR

AR206-ME55-03.tex

AR206-ME55-03.sgm

LaTeX2e(2002/01/18)

G PROTEIN AND RECEPTOR DISORDERS

P1: GBC

33

mutation that deletes residue Ile382 within the carboxyl terminus, which results in selective uncoupling of Gsα from the PTH receptor (30).

Annu. Rev. Med. 2004.55:27-39. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Role of Gβ 3-C825T Polymorphism in Multigenic Disorders A single base substitution (C825T) polymorphism within the Gβ 3 subunit was identified and initially shown to be associated with hypertension (31). More recent studies link the C825T allele to other features of the metabolic syndrome, including obesity and insulin resistance (32, 33), although these associations have not been confirmed in all studies (34). The C825T polymorphism generates a shortened Gβ 3 through alternative splicing, which enhances signaling through Gi pathways (31). The mechanisms by which this variant affects Gi signaling or human phenotypes are presently unknown.

DISEASES CAUSED BY MUTATIONS OF GPCR GENES GPCR Gene Loss-of-Function Mutations Clinically significant impairment of signal transduction generally requires loss of function of both alleles of a GPCR gene; thus, most such diseases are autosomal recessive, but there are several exceptions (Table 1). Loss-of-function

TABLE 1 Diseases caused by GPCR loss-of-function mutations Receptor

Disease

Inheritance

Cone opsins

Color blindness

X-linked; autosomal recessive

Rhodopsin

Retinitis pigmentosa

Autosomal dominant; recessive

V2 vasopressin Nephrogenic diabetes insipidus

X-linked

ACTH

Familial ACTH resistance

Autosomal recessive

LH

Male pseudohermaphroditism

Autosomal recessive

Ca2+ sensing

Familial hypocalciuric hypercalcemia Autosomal dominant

Ca2+ sensing

Neonatal hyperparathyroidism

Autosomal recessive

Endothelin-B

Hirschsprung disease

Complex

FSH

Hypergonadotropic ovarian failure

Autosomal recessive

TSH

Congenital hypothyroidism

Autosomal recessive

TRH

Central hypothyroidism

Autosomal recessive

GHRH

Growth hormone deficiency

Autosomal recessive

GnRH

Central hypogonadism

Autosomal recessive

Melanocortin 4 Extreme obesity

Codominant

PTH/PTHrP

Autosomal recessive

Blomstrand chondrodysplasia

2 Dec 2003

22:30

Annu. Rev. Med. 2004.55:27-39. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

34

AR

AR206-ME55-03.tex

SPIEGEL

¥

AR206-ME55-03.sgm

LaTeX2e(2002/01/18)

P1: GBC

WEINSTEIN

mutations may be missense as well as nonsense or frameshift mutations that truncate the normal receptor protein. They may involve any portion of the receptor, although the membrane-spanning helices are a particularly frequent site (35). Lossof-function mutations of receptors for ACTH, TSH, FSH, and the hypothalamic hormones—gonadotropin-releasing hormone (GnRH) (36), thyrotropin-releasing hormone (TRH), and growth hormone–releasing hormone (GHRH)—mimic deficiency of the respective hormones. Subjects with heterozygous loss-of-function mutations of the TSH receptor gene are generally euthyroid with compensatory elevated serum TSH, but homozygous mutations result in congenital hypothyroidism associated with a hypoplastic or even absent thyroid gland (37, 38). Loss-of-function mutations in LH, endothelin B, and PTH/PTHrP receptors cause developmental anomalies, reflecting the critical role of the respective hormones in normal development. Hirschsprung disease, congenital intestinal obstruction, is caused by lack of enteric ganglia secondary to defects in development of neural crest–derived cells. Loss-of-function mutations of the endothelin type B receptor can lead to Hirschsprung disease, but inheritance is complex (39). As many as 20% of subjects with heterozygous endothelin B receptor mutations may show the disease, but even when homozygous mutations are present, penetrance for Hirschsprung disease is not complete. Homozygotes may also show pigmentary defects (Shah-Waardenburg syndrome), presumably because melanocytes are also neural crest–derived. Loss-of-function mutations of the RET gene encoding a tyrosine kinase receptor for several neurotrophic factors can also cause Hirschsprung disease, and evidence in humans and mice supports genetic interactions between mutations in the RET gene and endothelin B receptor gene in the complex inheritance of this disease. Loss-of-function mutations of both copies of the LH receptor gene cause a rare form of 46,XY male pseudohermaphroditism known as Leydig cell hypoplasia (40). Absence of functional PTH/PTHrP receptors causes a rare, lethal form of dwarfism known as Blomstrand chondrodysplasia (41). Defects in breast and tooth formation in affected subjects show that this receptor is important for normal development of bone, teeth, and breast. X-linked nephrogenic diabetes insipidus (renal vasopressin resistance) is caused by loss-of-function mutations in the V2 vasopressin receptor gene located on the X chromosome (42). Males inheriting a mutant gene develop the disease, whereas most females do not show overt disease because random X inactivation results, on average, in 50% normal receptor genes. Identification of the mutation in carrier females facilitates early treatment of affected male neonates to avoid hypernatremia and brain damage. In familial hypocalciuric hypercalcemia, there is relative resistance to extracellular Ca2+ action caused by loss-of-function mutation of one copy of the gene encoding the Ca2+-sensing receptor that controls PTH secretion from the parathyroid and reabsorption of Ca2+ by the kidney (43). If two defective copies are inherited, extreme Ca2+ resistance causing neonatal severe primary hyperparathyroidism results. Loss-of-function mutations in the gene encoding the melanocortin 4 receptor, which regulates hypothalamic pathways controlling

2 Dec 2003

22:30

AR

AR206-ME55-03.tex

AR206-ME55-03.sgm

LaTeX2e(2002/01/18)

G PROTEIN AND RECEPTOR DISORDERS

P1: GBC

35

appetite and energy metabolism, result in a distinct obesity syndrome characterized by hyperphagia and increased linear growth (44). Inheritance is codominant, with homozygotes showing a more severe phenotype than heterozygotes. Autosomal dominant retinitis pigmentosa is an exceptional case in which certain mutations in one copy of the rhodopsin gene impair normal folding and synthesis of rhodopsin (45). This ultimately leads to degeneration of retinal rod photoreceptor cells in which rhodopsin synthesis accounts for a high proportion of total protein synthesis.

Annu. Rev. Med. 2004.55:27-39. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

GPCR Gene Gain-of-Function Mutations Given the dominant nature of activating mutations, most diseases caused by GPCR gain-of-function mutations are inherited in an autosomal dominant manner (Table 2). Unlike loss-of-function mutations, GPCR gain-of-function mutations are almost always missense mutations. Activating missense mutations are thought to disrupt normal inhibitory constraints that maintain the receptor in its inactive conformation (5). Mutations disrupting these constraints mimic the effects of agonist binding and shift the equilibrium toward the activated state of the receptor. Germline gain-of-function mutations in the LH and TSH receptor genes may mimic states of hormone excess, familial male precocious puberty (40), and familial nonautoimmune hyperthyroidism (38), respectively. Women inheriting gain-of-function mutations in the LH receptor gene do not show precocious puberty because, unlike in males, the combined action of LH and FSH is required for female pubertal development. As discussed above for activating Gsα mutations, increased cAMP in many endocrine cells leads to increased proliferation and hormone hypersecretion. Thus, somatic gain-of-function mutations of the LH and TSH receptor genes cause sporadic tumors of Leydig cells and the thyroid cells, respectively. Most germline activating mutations of the Ca2+-sensing receptor gene are not truly constitutively activating. Instead, they increase the receptor’s sensitivity to agonist stimulation (43). Because the agonist, extracellular Ca2+, is always present, such mutations lead to suppression of PTH secretion and increased urinary Ca2+ excretion at inappropriately low concentrations of serum Ca2+. This results in

TABLE 2 Diseases caused by GPCR gain-of-function mutation Receptor

Disease

Inheritance

Rhodopsin

Congenital night blindness

Autosomal dominant

LH

Familial male precocious puberty

Autosomal dominant

LH

Sporadic Leydig cells tumors

Somatic

TSH

Familial nonautoimmune hyperthyroidism

Autosomal dominant

TSH

Sporadic hyperfunctional thyroid adenomas

Somatic

Ca2+ sensing

Familial hypocalcemia

Autosomal dominant

PTH/PTHrP

Jansen metaphyseal chondrodysplasia

Autosomal dominant

2 Dec 2003

22:30

Annu. Rev. Med. 2004.55:27-39. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

36

AR

AR206-ME55-03.tex

SPIEGEL

¥

AR206-ME55-03.sgm

LaTeX2e(2002/01/18)

P1: GBC

WEINSTEIN

hypocalcemia and relative hypercalcuria. The latter may become clinically significant if calcium supplements and vitamin D are used aggressively to raise the serum Ca2+. Activating mutations of the PTH/PTHrP receptor gene cause Jansen’s metaphyseal chondrodysplasia (46). The phenotype includes hypercalcemia and hypophosphatemia mimicking the effects of PTH hypersecretion, but also abnormal bone development (short-limb dwarfism), reflecting the critical role of PTHrP in endochondral bone formation. A mutant PTH/PTHrP receptor has also been identified in human endochondromatosis, a condition characterized by abnormal proliferation and differentiation of growth-plate cartilage cells (47). Normal sensitivity to light depends on constraining rhodopsin tightly in its inactive state under dark conditions. Activating rhodopsin mutations cause congenital night blindness by disrupting normal constraints (45).

GPCR Gene Polymorphisms Variations in GPCR gene sequence can have important consequences beyond causing Mendelian diseases. Homozygous loss-of-function mutations of the type 5 chemokine receptor confer resistance to HIV infection because this receptor serves as a coreceptor for HIV entry into cells (48). As more polymorphisms are discovered in the human genome, many examples of variations in GPCR gene sequence will be found (35). The challenge will be to elucidate their possible functional significance, for example, whether such differences are important in individual variation in drug response (pharmacogenomics) or whether they could confer susceptibility to disease (complex disease genes). Specific polymorphisms in adrenergic GPCR genes have already been shown to confer susceptibility to congestive heart failure (49).

CONCLUSIONS By careful study of the phenotypic consequences of naturally occurring mutations in genes encoding G proteins and GPCRs, we can learn a great deal about the normal function of these genes. These mutations also help define critical structure-function relationships. Artificial knockouts of G protein and GPCR genes in mice reveal many nonredundant, functionally important G proteins and GPCRs for which no human disease-causing mutations have yet been identified. There are, moreover, multiple GPCRs (so-called orphans) for which the endogenous agonist is unknown (1). This suggests that further search for disease-causing mutations, informed by careful analysis of relevant phenotypes, is likely to be fruitful. Even more likely is the identification of additional polymorphisms in genes encoding G proteins and GPCRs that play a role in complex inheritance of common diseases. Identification and study of diseases caused by mutations in G proteins and GPCRs should also lead to advances in novel forms of treatment for such diseases, including development of inverse agonists (5) and “pepducins” (50) to inhibit

2 Dec 2003

22:30

AR

AR206-ME55-03.tex

AR206-ME55-03.sgm

LaTeX2e(2002/01/18)

P1: GBC

G PROTEIN AND RECEPTOR DISORDERS

37

constitutively activated GPCRs, and methods to rescue function of misfolded or truncated GPCRs (42). The Annual Review of Medicine is online at http://med.annualreviews.org

Annu. Rev. Med. 2004.55:27-39. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

LITERATURE CITED 1. Vassilatis DK, Hohmann JG, Zeng H, et al. 2003. The G protein-coupled receptor repertoires of human and mouse. Proc. Natl. Acad. Sci. USA 100(8):4903–8 2. Pierce KL, Premont RT, Lefkowitz RJ. 2002. Seven-transmembrane receptors. Nat. Rev. Mol. Cell Biol. 3:639–50 3. Palczewski K, Kumasaka T, Hori T, et al. 2000. Crystal structure of rhodopsin: a G protein-coupled receptor. Science 289: 739–45 4. Bockaert J, Pin JP. 1999. Molecular tinkering of G protein-coupled receptors: an evolutionary success. EMBO J. 18:1723–29 5. Seifert R, Wenzel-Seifert K. 2002. Constitutive activity of G-protein coupled receptors: cause of disease and common property of wild-type receptors. Naunyn. Schmiedebergs. Arch. Pharmacol. 366:381–416 6. Hamm H. 2001. How activated receptors couple to G proteins. Proc. Natl. Acad. Sci. USA 98:4819–21 7. Dessauer CW, Posner BA, Gilman AG. 1996. Visualizing signal transduction: receptors, G proteins and adenylate cyclases. Clin. Sci. 91:527–37 8. Neer EJ. 1995. Heterotrimeric G proteins: organizers of transmembrane signals. Cell 80:249–57 9. Spiegel AM. 1998. G Proteins, Receptors, and Disease. Totawa, NJ: Humana 10. Muradov KG, Artemyev NO. 2000. Loss of the effector function in a transducin-α mutant associated with Nougaret night blindness. J. Biol. Chem. 275:6969–74 11. Lyons J, Landis CA, Harsh G, et al. 1990. Two G protein oncogenes in human endocrine tumors. Science 249:655–59 12. Williamson EA, Johnson SJ, Foster S, et al. 1995. G protein gene mutations in patients

13.

14.

15.

16.

17.

18.

19.

20.

with multiple endocrinopathies. J. Clin. Endocrinol. 80:1702–5 Hayward BE, Barlier A, Korbonits M, et al. 2001. Imprinting of the Gsα gene GNAS1 in the pathogenesis of acromegaly. J. Clin. Invest. 107:R31–36 Shenker A, Weinstein LS, Sweet DE, et al. 1994. An activating Gsα mutation is present in fibrous dysplasia of bone in the McCuneAlbright syndrome. J. Clin. Endocrinol. Metab. 79:750–55 Weinstein LS, Shenker A, Gejman PV, et al. 1991. Activating mutations of the stimulatory G protein in the McCune-Albright syndrome. N. Engl. J. Med. 325:1688–95 Shenker A, Weinstein LS, Moran A, et al. 1993. Severe endocrine and non-endocrine manifestations of the McCune-Albright syndrome associated with activating mutations of stimulatory G protein Gs. J. Pediatr. 123:509–18 Mockridge KA, Levine MA, Reed LA, et al. 1999. Polyendocrinopathy, skeletal dysplasia, organomegaly, and distinctive facies associated with a novel, widely expressed Gsα mutation. Am. J. Hum. Genet. 65(Suppl.): A426 (Abstr.) Okamoto S, Hisaoka M, Ushijima M, et al. 2000. Activating Gsα mutation in intramuscular myxomas with and without fibrous dysplasia of bone. Virchows Arch. 437:133–37 Candeliere GA, Glorieux FH, Prud’homme J, et al. 1995. Increased expression of the c-fos proto-oncogene in bone from patients with fibrous dysplasia. N. Engl. J. Med. 332:1546–51 Collins MT, Chebli C, Jones J, et al. 2001. Renal phosphate wasting in fibrous dysplasia of bone is part of a generalized

2 Dec 2003

22:30

38

21.

Annu. Rev. Med. 2004.55:27-39. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

22.

23.

24.

25.

26.

27.

28.

29.

AR

AR206-ME55-03.tex

SPIEGEL

¥

AR206-ME55-03.sgm

LaTeX2e(2002/01/18)

P1: GBC

WEINSTEIN

renal tubular dysfunction similar to that seen in tumor-induced osteomalacia. J. Bone Miner. Res. 16:806–13 Kim I, Kim ER, Nam HJ, et al. 1999. Activating mutation of Gsα in McCuneAlbright syndrome causes skin pigmentation by tyrosinase gene activation on affected melanocytes. Horm. Res. 52:235–40 Weinstein LS, Yu S, Warner DR, et al. 2001. Endocrine manifestations of stimulatory G protein α-subunit mutations and the role of genomic imprinting. Endocr. Rev. 22:675– 705 Spiegel AM, Weinstein LS. 2001. Pseudohypoparathyroidism. In The Metabolic and Molecular Bases of Inherited Disease, ed. CR Scriver, AL Beaudet, WS Sly, D Valle, pp. 4205–21. New York: McGraw-Hill Shore EM, Ahn J, Jan de Beur S, et al. 2002. Paternally inherited inactivating mutations of the GNAS1 gene in progressive osseous heteroplasia. N. Engl. J. Med. 346:99–106 Germain-Lee EL, Ding C-L, Deng Z, et al. 2002. Paternal imprinting of Gαs in the human thyroid as the basis of TSH resistance in pseudohypoparathyroidism type 1a. Biochem. Biophys. Res. Commun. 296:67–72 Iiri T, Herzmark P, Nakamoto JM, et al. 1994. Rapid GDP release from Gs-alpha in patients with gain and loss of endocrine function. Nature 371:164–67 Juppner H, Schipani E, Bastepe M, et al. 1998. The gene responsible for pseudohypoparathyroidism type Ib is paternally imprinted and maps in four unrelated kindreds to chromosome 20q13.3. Proc. Natl. Acad. Sci. USA 95:11798–803 Liu J, Litman D, Rosenberg MJ, et al. 2000. A GNAS1 imprinting defect in pseudohypoparathyroidism type IB. J. Clin. Invest. 106:1167–74 Bastepe M, Lane AH, J¨uppner H. 2001. Paternal uniparental disomy of chromosome 20q- and the resulting changes in GNAS1 methylation as a plausible cause of pseudohypoparathyroidism. Am. J. Hum. Genet. 68:1283–89

30. Wu WI, Schwindinger WF, Aparicio LF, et al. 2001. Selective resistance to parathyroid hormone caused by a novel uncoupling mutation in the carboxyl terminus of Gαs. A cause of pseudohypoparathyroidism type Ib. J. Biol. Chem. 276:165–71 31. Siffert W, Rosskopf D, Siffert G, et al. 1998. Association of a human G-protein β3 subunit variant with hypertension. Nat. Genet. 18:45–48 32. Brand E, Wang JG, Herrmann SM, et al. 2003. An epidemiological study of blood pressure and metabolic phenotypes in relation to the Gβ3 C825T polymorphism. J. Hypertens. 21:729–37 33. Von Beckerath N, Schusterschitz Y, Koch Y, et al. 2003. G protein β3 subunit 825T allele carriage and risk of coronary artery disease. Atherosclerosis 167:135–39 34. Fernandez-Real JM, Penarroja G, Richart C, et al. 2003. G protein β3 gene variant, vascular function, and insulin sensitivity in type 2 diabetes. Hypertension 41:124–29 35. Lee A, Rana BK, Schiffer HH, et al. 2003. Distribution analysis of nonsynonymous polymorphisms within the G-proteincoupled receptor gene family. Genomics 81:245–48 36. Karges B, Karges W, Mine M, et al. 2003. Mutation A1a171 Thr stabilizes the gonadotropin-releasing hormone receptor in its inactive conformation, causing familial hypogonadotropic hypogonadism. J. Clin. Endocrinol. Metab. 88(4):1873–79 37. Jordan N, Williams N, Gregory JW, et al. 2003. The W546X mutation of the thyrotropin receptor gene: potential major contributor to thyroid dysfunction in a caucasian population. J. Clin. Endocrinol. Metab. 88(3):1002–5 38. Corvilain B, Van Sande J, Dumont JE, et al. 2001. Somatic and germline mutations of the TSH receptor and thyroid diseases. Clin. Endocrinol. 55:143–58 39. Carasquillo MM, McCallion AS, Puffenberger EG, et al. 2002. Genome-wide association study and mouse model identify interaction between RET and EDNRB

2 Dec 2003

22:30

AR

AR206-ME55-03.tex

AR206-ME55-03.sgm

LaTeX2e(2002/01/18)

G PROTEIN AND RECEPTOR DISORDERS

40.

Annu. Rev. Med. 2004.55:27-39. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

41.

42.

43.

44.

pathways in Hirschsprung disease. Nat. Genet. 32:237–44 Themmen AP, Verhoef-Post M. 2002. LH receptor defects. Semin. Reprod. Med. 20(3):199–204 Wysolmerski JJ, Cormier S, Philbrick WM, et al. 2001. Absence of functional type 1 parathyroid hormone (PTH)/PTH-related protein receptors in humans is associated with abnormal breast development and tooth impaction. J. Clin. Endocrinol. Metab. 86(4):1487–88 Schulz A, Sangkuh K, Lennert T, et al. 2002. Aminoglycoside pretreatment partially restores the function of truncated V2 vasopressin receptors found in patients with nephrogenic diabetes insipidus. J. Clin. Endocrinol. Metab. 87(11):5247–57 Hu J, Spiegel AM. 2003. Naturally occurring mutations of the extracellular Ca++sensing receptor: implications for understanding its structure and function. Trends Endocrinol. Metab. 14:282–88 Farooqi S, Keogh JM, Yeo GSH, et al. 2003. Clinical spectrum of obesity and mutations in the melanocortin 4 receptor gene. N. Engl. J. Med. 348(12):1085–95

P1: GBC

39

45. Berson EL. 2002. Retinitis pigmentosa and allied diseases. In Principles and Practice of Ophthalmology, ed. DM Albert, FA Jakobiec, DT Azar, ES Gragoudas, pp. 2262–90. Philadelphia: Saunders 46. Calvi LM, Schipani E. 2000. The PTH/PTHrP receptor in Jansen’s metaphyseal chondrodysplasia. J. Endocrinol. Invest. 23(8):545–54 47. Hopyan S, Gokgoz N, Poon R, et al. 2002. A mutant PTH/PTHrP type I receptor in enchondromatosis. Nat. Genet. 30(3):306– 10 48. Liu R, Paxton WA, Choe S, et al. 1996. Homozygous defect in HIV-1 coreceptor accounts for resistance of some multiply exposed individuals to HIV-1 infection. Cell 86:367–377 49. Small KM, Wagoner LE, Levin AM, et al. 2002. Synergistic polymorphisms of β 1and α 2c-adrenergic receptors and the risk of congestive heart failure. N. Engl. J. Med. 347(15):1135–42 50. Covic L, Misra M, Badar J, et al. 2002. Pepducin-based intervention of thrombinreceptor signaling and systemic platelet activation. Nat. Med. 8(10):1161–65

5 Jan 2004

11:20

AR

AR206-ME55-04.tex

AR206-ME55-04.sgm

LaTeX2e(2002/01/18) P1: GBC 10.1146/annurev.med.55.091902.104422

Annu. Rev. Med. 2004. 55:41–60 doi: 10.1146/annurev.med.55.091902.104422 c 2004 by Annual Reviews. All rights reserved Copyright °

THE SCIENTIFIC BASIS FOR THE CURRENT TREATMENT OF PARKINSON’S DISEASE

Annu. Rev. Med. 2004.55:41-60. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

C. Warren Olanow Department of Neurology, Mount Sinai School of Medicine, Annenberg 14–94, One Gustave L. Levy Place, Box 1137, New York, New York 10029; email: [email protected]

Key Words dopamine, levodopa, dopamine agonist, neuroprotection ■ Abstract Parkinson’s disease (PD) is an age-related neurodegenerative disease that affects approximately one million people in the United States. The introduction of levodopa revolutionized the treatment for this disorder, but the long-term utility of the drug is limited by motor complications, the development of features such as postural instability and dementia that do not respond to treatment, and continued disease progression. Insights into the organization of the basal ganglia in the normal and PD conditions has permitted the design of new treatment strategies that reduce the risk of developing motor complications. Additionally, increased knowledge of the mechanisms responsible for cell death in PD has permitted the development of putative neuroprotective drugs that might slow or stop disease progression. No drug has yet been established to alter the rate of disease progression, but the rapid pace of research offers reason for optimism.

INTRODUCTION Parkinson’s disease (PD) is an age-related neurodegenerative disorder with a mean age of onset of 60 years. It affects ∼1 million individuals in the United States at an estimated annual cost of $28 billion. As the population ages in the coming decades, these numbers are likely to increase substantially, creating an even greater public health problem. Clinically, PD is characterized by four cardinal features: resting tremor, rigidity, bradykinesia (slowing of movement), and gait disturbance with postural instability (1, 2). Nonmotor features include autonomic dysfunction, sleep disturbances, depression, and dementia. Pathologically, there is preferential degeneration of melanized dopaminergic neurons in the substantia nigra pars compacta (SNc) coupled with proteinaceous intracellular inclusions (Lewy bodies) (Figure 1) and a reduction in striatal dopamine (3, 4). Although the nigrostriatal dopamine system bears the brunt of damage, neurodegeneration can also involve norepinephrine neurons in the locus ceroeleus, cholinergic neurons in the nucleus basalis of 0066-4219/04/0218-0041$14.00

41

5 Jan 2004

11:20

Annu. Rev. Med. 2004.55:41-60. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

42

AR

AR206-ME55-04.tex

AR206-ME55-04.sgm

LaTeX2e(2002/01/18)

P1: GBC

OLANOW

Meynert and dorsal motor nucleus of the vagus, serotonin neurons in the dorsal raphe nucleus, and neurons of the cerebral cortex, brain stem, spinal cord, and peripheral autonomic nervous system (3). Neuropathological studies from the London Brain Bank suggest that the clinical diagnosis of PD is frequently incorrect (5) and that resting tremor, asymmetry, and a good response to levodopa are the clinical features that best predict PD pathology (6). Since its introduction in the late 1960s, treatment for PD has been primarily based on dopamine replacement by means of the prodrug levodopa, which is converted to dopamine in the brain by aromatic amino acid decarboxylase (AADC) (7). Levodopa is typically administered in combination with a peripheral AADC inhibitor such as carbidopa (Sinemet®) or benserazide (Madopar®). These combinations reduce the peripheral metabolism of levodopa and prevent side effects such as nausea and vomiting due to activation of dopamine receptors in the area postrema of the medulla that are not protected by the blood-brain barrier. Levodopa remains the most widely employed and most effective antiparkinsonian agent and provides dramatic clinical benefits for almost all PD patients. Treatment is associated with reduced disability, extended employability, prolonged independence, and increased survival in comparison to the pre-levodopa era. However, three problems limit the long-term utility of levodopa (1, 2): 1. Motor complications: These consist of dyskinesias and motor fluctuations (8). Dyskinesias are involuntary movements that are usually choreiform in nature and typically occur at the time of maximal clinical benefit and peak plasma concentration (“peak-dose dyskinesias”). Less commonly, patients experience dyskinesias at the beginning and end of the levodopa dose response (diphasic dyskinesia). These typically consist of stereotyped rhythmic movements that primarily involve the lower limbs. Motor fluctuations consist of alternations between “on” periods, in which levodopa provides a good antiparkinsonian benefit, and “off” periods, when the levodopa benefit “wears off” prior to the next dose and patients experience worsening parkinsonism. Indeed, patients may cycle between “on” periods that are complicated by peak-dose dyskinesia and “off” periods associated with disabling parkinsonism. Motor complications are common, affecting as many as 80% of PD patients who have been treated with levodopa for >5 years. Indeed, motor complications, and not worsening of parkinsonism, are presently the primary reason for performing most neurosurgical procedures for PD (see below). 2. Features that are not adequately controlled by levodopa: Many PD patients develop features that do not respond to, or are inadequately controlled by, levodopa therapy. These include freezing periods, postural instability, falls, autonomic dysfunction (orthostatic hypotension, bowel and bladder disturbances, and sexual dysfunction), depression, and dementia. These can be important sources of disability, and indeed dementia is the commonest reason for nursing-home placement of PD patients.

5 Jan 2004

11:20

AR

AR206-ME55-04.tex

AR206-ME55-04.sgm

LaTeX2e(2002/01/18)

PARKINSON’S DISEASE

P1: GBC

43

Annu. Rev. Med. 2004.55:41-60. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

3. Disease progression: Despite the symptomatic value of levodopa, patients continue to deteriorate, and most eventually experience unacceptable disability. Indeed, there has been a theoretical concern that levodopa might actually accelerate neuronal degeneration and disease progression as a result of cytotoxic metabolites generated by the drug’s oxidative metabolism, although there is no direct evidence that this actually occurs in PD (9). Collectively, these problems limit the long-term value of levodopa, despite the drug’s profound symptomatic benefits. Accordingly, there has been an intensive search for more effective treatment strategies. Over the past decade, insight into the organization of the basal ganglia in the normal and disease states has permitted the development of new medical and surgical treatment approaches that greatly improve motor function and quality of life for PD patients. In addition, increased understanding of the mechanisms responsible for cell death in PD has permitted the identification of putative neuroprotective and restorative therapies. This review considers the scientific basis for these new treatment approaches and how they influence current management. Because of space limitations, it does not cover the routine treatment of PD or the management of autonomic dysfunction, sleep disturbances, depression, psychosis, and dementia that can also be important problems and a source of disability for PD patients (see References 1 and 10 for detailed discussion of these issues).

MANAGEMENT OF EARLY PARKINSON’S DISEASE Levodopa remains the most effective antiparkinsonian agent and provides excellent symptomatic control, particularly in the early stages of the disease. However, with chronic treatment and advancing disease, patients experience disability related to motor complications, and it becomes increasingly difficult to deliver a dose of levodopa that both controls parkinsonian motor features and avoids dyskinesia. The management of early PD must therefore consider the long-term consequences of initial treatment, and strategies should be selected to reduce the risk of eventual motor complications.

Experimental Observations The classic model of the basal ganglia proposes that basal ganglia input neurons in the striatum project to basal ganglia output neurons in the globus pallidus pars interna (GPi) and substantia nigra pars reticularis (SNr) by way of direct and indirect pathways (11, 12; Figure 2a). Neurons in the direct and indirect pathways exert inhibitory and excitatory influences, respectively, on GPi/SNr neurons and regulate their inhibitory effect on thalamocortical and brain-stem neurons involved in motor circuitry. In this model, SNc dopamine neurons differentially influence the basal ganglia system by activating D1 and inhibiting D2 receptors on striatal neurons that give rise to the direct and indirect pathways, respectively (13).

5 Jan 2004

11:20

Annu. Rev. Med. 2004.55:41-60. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

44

AR

AR206-ME55-04.tex

AR206-ME55-04.sgm

LaTeX2e(2002/01/18)

P1: GBC

OLANOW

The model predicts that dopamine depletion results in a net increase in the firing rate of GPi/SNr neurons leading to excessive inhibition of thalamo-cortical projection neurons and the emergence of parkinsonian motor features (Figure 2b). This concept is supported by neurophysiologic, metabolic, and imaging studies that confirm increased activity in the subthalamic nucleus (STN) and GPi in the parkinsonian state (14–16). It can thus be predicted that lesions of the STN or GPi that suppress neuronal overactivity in these structures might provide antiparkinsonian benefits, and indeed such benefits have been demonstrated in both MPTP monkeys and PD patients (17–20; see below). The model also predicts that levodopa-induced dyskinesias result from a reduced firing frequency in basal ganglia output neurons with consequent disinhibition of thalamocortical neurons and uncontrolled activation of cortical motor regions (Figure 2c). In support of this hypothesis, neuronal firing rates in GPi are reduced in experimental models of dyskinesia (21); furthermore, intraoperative recordings from the GPi of PD patients undergoing pallidotomy demonstrate that administration of a dopamine agonist is associated with both the onset of dyskinesia and a reduction in neuronal firing rate (22). However, the classic model does not account for all of the metabolic findings in levodopa-induced dyskinesia (23). More importantly, it does not explain why pallidotomy, which profoundly reduces the output of the GPi, is consistently associated with amelioration and not induction of dyskinesia (23). It is now apparent that the basal ganglia are much more complex than the classic model depicted. There is evidence of extensive axonal collateralization within the basal ganglia, and most striatal neurons coexpress both D1 and D2 receptors. Thus, the direct and indirect pathways are less clearly delineated than was initially proposed (24). It is also now appreciated that SNc dopamine neurons provide dopaminergic innervation to virtually all regions of the basal ganglia, and not just the striatum. Further, dopamine neurons fire continuously at a low frequency (tonic firing) so as to provide continuous stimulation of dopamine receptors (25). Dopamine neurons may also exhibit short bursts of high-frequency (phasic) firing, typically in association with reward (26), but the terminal reuptake system is sufficiently robust that dopamine delivered by phasic firing is rapidly removed from the synaptic cleft, thereby maintaining synaptic dopamine concentrations and receptor activation at a relatively constant level. Collectively, these findings suggest that the basal ganglia functions more as a dynamic, self-correcting network that is modulated by dopamine than as the linear, firing-rate–dependent system proposed by the classic model (Figure 2d). Based on these new findings, it has been proposed that levodopa-induced motor complications are related to disruption of the basal ganglia network as a result of abnormal intermittent or pulsatile stimulation of striatal dopamine receptors, which does not mirror the normal continuous activation of these receptors that occurs physiologically (27). Pulsatile stimulation is prone to occur in PD where there is a loss of striatal dopamine terminals and their capacity to buffer fluctuations in striatal dopamine concentration. With the loss of nigral neurons that occurs

5 Jan 2004

11:20

AR

AR206-ME55-04.tex

AR206-ME55-04.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:41-60. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

PARKINSON’S DISEASE

P1: GBC

45

in PD, striatal dopamine levels become increasingly dependent on the peripheral availability of levodopa. In this situation, fluctuations in the plasma concentration of short-acting levodopa may be directly translated to the striatum, causing striatal dopamine receptors to be exposed to alternating high and low levels of striatal dopamine. This pulsatile stimulation leads to dysregulation of receptor-mediated intracellular gene and protein signals and changes in the firing patterns of basal ganglia output neurons with resultant motor complications. Gene changes reported in striatal neurons in association with pulsatile stimulation of dopamine receptors and the development of dyskinesia in rodents and primates include upregulation of preproenkephalin, prodynorphan, c-fos, and c-jun. Additionally, in levodopainduced dyskinesia, firing patterns in GPi neurons are altered with respect to the number and duration of pauses and bursts, and degree of synchrony, as well as firing frequency. These experimental observations have led to the concept that motor complications might be prevented or delayed by delivering dopaminergic therapy in a more continuous manner so as to avoid pulsatile stimulation of striatal dopamine receptors. Indeed, it has now been shown that short-acting dopaminergic agents that are prone to induce pulsatile stimulation (such as levodopa or the dopamine agonists quinpirole, PHNO, and SKF 82958) rapidly induce severe dyskinesias in parkinsonian monkeys, whereas long-acting dopamine agonists (such as bromocriptine, ropinirole, and cabergoline) do not, even though they provide comparable motor benefit (28–30). The importance of pulsatile stimulation in the induction of dyskinesia is further illustrated by the finding that a short-acting dopamine agonist induces dyskinesia when administered intermittently but not when the same agent is administered continuously (31). Interestingly, in these experiments, gene changes in striatal neurons that are associated with dyskinesia, such as upregulation of preproenkephalin, are found with intermittent, but not continuous, delivery of the same agent (32).

Clinical Observations Based on the concept of continuous dopaminergic stimulation, several clinical trials have compared long-acting dopamine agonists to short-acting levodopa as initial therapy for PD. Dopamine agonists are antiparkinson drugs that have historically been used as adjuncts to levodopa in patients who already experience motor complications. However, currently available dopamine agonists each have a longer plasma half-life than levodopa (Table 1) and might therefore be expected to activate striatal dopamine receptors in a less pulsatile and more physiologic manner. In prospective double-blind clinical trials, early PD patients were randomized to initiate therapy with either standard levodopa or a dopamine agonist (33–35). Patients in both groups were permitted to receive open-label supplemental levodopa in addition to their blinded study medication if deemed necessary. In each of these studies, patients randomized to start therapy with the dopamine agonist had a reduced risk of developing motor complications than did patients randomized to levodopa, whether or not they took open-label levodopa (Figure 3a). If

5 Jan 2004

11:20

46

AR

AR206-ME55-04.tex

AR206-ME55-04.sgm

LaTeX2e(2002/01/18)

P1: GBC

OLANOW TABLE 1 Half-life of dopaminergic agents Drug

Half-life (Hours)

Sinemet® (Carbidopa/Levodopa)

2.5–3

Parlodel® (bromocriptine)

12–15

Permax® (pergolide) ®

7–16

Requip (ropinirole)

6–8

Mirapex® (pramipexole)

8–12

®

Annu. Rev. Med. 2004.55:41-60. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

1–1.5

Sinemet® + Comtan® (Entacapone)

Casbar (cabergoline)

>24

they could remain on monotherapy, patients in the levodopa group had a 15-fold greater risk of developing dyskinesia than did patients in the agonist group (33). When levodopa was added to the dopamine agonist, the risk of experiencing motor complications increased but was still significantly lower than in patients treated with levodopa alone (odds ratio of 4:1; Figure 3b). It has been hypothesized that initiating levodopa in combination with a catechol-O-methyl transferase (COMT) inhibitor to block its peripheral metabolism and extend its elimination half-life might further reduce the likelihood that the drug will induce pulsatile stimulation and motor complications (36). Preliminary studies in MPTP monkeys support this concept (37) and clinical trials are under way to test this hypothesis. The results of clinical trials in PD patients are consistent with observations in the MPTP monkey and support the notion that levodopa-induced motor complications are related to the short half-life of the drug. They further indicate that the motor complications can be prevented or reduced by therapies that provide more continuous dopaminergic stimulation. Interestingly, in each of these studies, patients who were started on a dopamine agonist had slightly less improvement on the Unified Parkinson Disease Rating Scale (UPDRS) than those started on levodopa, even though patients in either treatment group could receive supplemental open-label levodopa therapy in addition to their blinded study medication. The basis of this small difference in motor score is not known and may not be clinically relevant, since patients in both groups had comparable quality-of-life scores. Based on these results, many specialists now institute symptomatic therapy for PD with a dopamine agonist except in the elderly and cognitively impaired, supplementing with levodopa when agonist monotherapy no longer achieves adequate symptomatic control (1). This view is not universal, however (38). Some experts continue to recommend initiating treatment with levodopa, arguing that it is superior symptomatically and that introducing agonists at a later stage might provide comparable long-term benefits with respect to motor complications, although such studies have not yet been performed. There remains a theoretical concern that levodopa might be toxic to dopamine neurons based on its oxidative metabolism and its capacity to damage cultured

5 Jan 2004

11:20

AR

AR206-ME55-04.tex

AR206-ME55-04.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:41-60. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

PARKINSON’S DISEASE

P1: GBC

47

Figure 3 (a) Comparison of frequency of developing dyskinesia in 3 different doubleblind prospective trials in which patients have been randomized to initiate treatment with a dopamine agonist or levodopa (derived from References 33 and 34). In each study, the frequency of dyskinesias was significantly lower in patients started on the dopamine agonist than in those started on levodopa. The names and durations of the trials are noted on the x-axis. (b) Percentage of patients who remained free of dyskinesia while on monotherapy (left) and of those who received open-label levodopa supplementation (right). Note that the frequency of dyskinesias increases when levodopa is added to agonist-treated patients but is still lower than in patients receiving levodopa alone. (Adapted from Reference 33 with permission.)

5 Jan 2004

11:20

Annu. Rev. Med. 2004.55:41-60. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

48

AR

AR206-ME55-04.tex

AR206-ME55-04.sgm

LaTeX2e(2002/01/18)

P1: GBC

OLANOW

dopamine neurons. However, levodopa toxicity in vitro depends on the study conditions, and under some circumstances the drug has even been shown to be protective. In vivo, levodopa has not been shown to promote the loss of dopamine neurons in normal rodents, primates, or humans. Further, levodopa treatment does not increase dopamine neuronal death in animals that have been dopamine-lesioned or suffer oxidative stress, models that are more relevant to PD. The recently completed Elldopa study tried to compare the rate of disease progression in early PD patients randomized to treatment with placebo or one of three doses of levodopa (39). The primary endpoint was the change in UPDRS motor score between an untreated baseline and a follow-up evaluation performed after nine months of treatment and two weeks of withdrawal from the study intervention. Patients randomized to levodopa had less deterioration from baseline than did placebo patients, consistent with slower, not faster, disease progression. However, the motor response to levodopa can persist for weeks after withdrawal. It is therefore not possible to say with certainty whether the results are due to a protective or prolonged symptomatic effect. The problem is compounded by the results of concomitantly performed neuroimaging studies in these patients, which used striatal β-CIT uptake on single-photon-emission computerized tomagraphy (SPECT) as a surrogate index of nigrostriatal function. Patients randomized to levodopa had a significantly greater rate of decline from baseline than placebo patients, consistent with a toxic effect. However, the potential of levodopa to induce regulatory changes in dopamine terminal function and other components of the nigrostriatal system confounds interpretation of these results as well. The Elldopa study thus does not clarify whether levodopa is toxic in PD, and it is generally recommended that the drug be prescribed based solely on its efficacy and side-effect profile (40).

MANAGEMENT OF ADVANCED PARKINSON’S DISEASE Although current treatment strategies are designed to prevent or limit the development of motor complications, hundreds of thousands of PD patients currently suffer from these problems. Manipulations of levodopa, dopamine agonists, COMT inhibitors, amantadine, and diet may help some of these patients but fail to provide satisfactory control in the majority. Modern surgical therapies offer an effective alternative. In addition, medical therapies based on the principles of continuous dopaminergic stimulation may be helpful in the management of established motor complications as well as in their prevention.

Surgical Approaches Surgery has been used in the treatment of PD for more than 50 years, but it fell into disfavor with the introduction of levodopa. The recent resurgence of interest in surgery relates to the limitations of levodopa, improvements in stereotactic

5 Jan 2004

11:20

AR

AR206-ME55-04.tex

AR206-ME55-04.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:41-60. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

PARKINSON’S DISEASE

P1: GBC

49

operative techniques, the use of microelectrode recording to more accurately define the target site, and insights into the anatomical and physiological organization of the basal ganglia suggesting that the STN or GPi might be appropriate surgical targets (see above). Several groups have noted dramatic benefit following pallidotomy, particularly when the lesion was placed in the postero-lateral or motor-sensory region of the GPi (19, 41, 42). Interestingly, the most striking benefit observed in these studies was a consistent and marked reduction in contralateral dyskinesia. This seemingly contradicts the classic model of the basal ganglia, which predicts that pallidotomy should induce, rather than ameliorate, dyskinesia. As discussed above, it is now believed that dyskinesia is related to the induction of altered neuronal firing patterns in basal ganglia output neurons that result in miscommunication of abnormal signals to cortical motor regions. It is the abolition of these abnormal neuronal firing patterns that likely accounts for the antidyskinetic benefits observed with pallidotomy. This implies that it is better for cortical motor regions to receive no information from the basal ganglia than incorrect or miscoded information. Side effects associated with pallidotomy are primarily related to hemorrhage and damage to neighboring structures, which include the internal capsule and optic radiation. Bilateral procedures, which are frequently required in PD, can also induce dysphagia, dysarthria, and cognitive impairment. Consequently, physicians have been reluctant to recommend bilateral pallidotomy. Ablation procedures have now been largely replaced by high-frequency deep brain stimulation (DBS), which involves implanting an electrode into a brain target and stimulating at a frequency of 100–180 Hz. DBS simulates the effect of a lesion, although the precise mechanism of action is not known. The desired brain target is identified by a combination of neuroimaging techniques, microelectrode recording, and microstimulation. A permanent electrode is then implanted into the selected target site and connected to a stimulator that is placed subcutaneously over the chest wall. DBS does not require making a brain lesion, and the stimulation settings (contacts, voltage, frequency, and pulse width) can be adjusted at any time so as to maximize benefits and minimize adverse effects. DBS thus permits bilateral procedures to be performed with relative safety and allows the targeting of brain structures one might be reluctant to lesion, such as the STN. DBS for PD was initially performed in the ventral intermediate (VIM) nucleus of the thalamus and provided striking and long-term amelioration of tremor (43). However, DBS-VIM does not affect the other, more disabling features of PD, so it is only occasionally performed today in PD patients. DBS of the STN or the GPi has been shown to provide significant improvement for all of the cardinal features of PD, including tremor (20, 44). These benefits have been confirmed in a large multicenter double-blind cross-over trial (45). In this study, the percentage of “on” time without dyskinesia during the waking day improved from 27% to 74% following bilateral stimulation of the STN and from 28% to 64%

5 Jan 2004

11:20

Annu. Rev. Med. 2004.55:41-60. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

50

AR

AR206-ME55-04.tex

AR206-ME55-04.sgm

LaTeX2e(2002/01/18)

P1: GBC

OLANOW

with bilateral stimulation of the GPi. Stimulation of either target also significantly reduced the percentage of daily “off” time and the severity of parkinsonism during “off” periods. Direct comparisons of stimulation of the STN versus GPi have not been performed, but open-label comparisons suggest that DBS of the STN may be superior. Adverse effects associated with DBS include those related to the surgical intervention (hemorrhage, stroke, infection), the stimulator system (lead breaks, mechanical malfunction, infection), and stimulation of the target (muscle twitch, paresthesia, dysarthria, eye-movement disorder). Stimulation-related side effects are typically transient and of little or no clinical consequence. Clinically significant adverse events occur in ∼2%–4% of patients. In the multicenter study, bilateral DBS procedures performed in 143 patients were associated with seven brain hemorrhages, four of which resulted in persistent disability. DBS thus offers promise for patients with advanced PD who suffer severe motor complications that cannot be controlled satisfactorally with currently available medical therapies. It is very important to appreciate that antiparkinsonian benefits obtained with DBS and other surgical procedures are no greater than what can be achieved with levodopa and primarily consist of reduced motor complications. This observation illustrates the importance of initiating PD treatment with strategies that reduce the likelihood of motor complications and avoid the later need for surgical interventions.

Medical Approaches Laboratory and clinical evidence suggests that therapies based on the principles of continuous dopamine stimulation can reverse, as well as prevent, motor complications. Chronic treatment with the long-acting dopamine agonist cabergoline reduces dyskinesia in levodopa-treated MPTP monkeys without compromising behavioral benefits (46). Similar benefits have been observed in PD patients with continuous infusion of levodopa, or the dopamine agonists apomorphine and lisuride (47). In a four-year prospective trial performed in advanced PD patients, “off” time was reduced by 59.3% in patients randomized to receive a continuous subcutaneous infusion of lisuride, whereas “off time” increased by 21.4% in patients randomized to continue to receive oral formulations of standard levodopa (48). Dyskinesia scores were similarly significantly improved in the lisuride infusion group, compared to both their original baseline and to patients treated with levodopa. Patients did not develop tolerance. However, the administration of infusion therapies does present practical problems for patients, care givers, and physicians. In addition, agents are employed that are not routinely available in many countries (e.g., lisuride, apomorphine, domperidone, and levodopa methyl ester). If, however, levodopa could be administered in a modified oral formulation that mirrors the pharmacokinetic profile of levodopa infusion, it might provide similar benefits that can be applied to a larger group of patients. This hypothesis is currently being tested.

5 Jan 2004

11:20

AR

AR206-ME55-04.tex

AR206-ME55-04.sgm

LaTeX2e(2002/01/18)

PARKINSON’S DISEASE

P1: GBC

51

FUTURE DIRECTIONS Despite the recent medical and surgical advances in the treatment of PD, patients continue to experience disability due to disease progression and the development of motor (e.g., freezing, falling, postural instability) and nonmotor (e.g., autonomic dysfunction, sleep disorders, depression, dementia) features that are not adequately controlled by levodopa therapy. This has led to an intensive search for a neuroprotective or restorative therapy.

Annu. Rev. Med. 2004.55:41-60. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Neuroprotective Therapies Neuroprotection can be defined as an intervention that stops or slows neuronal degeneration and disease progression. Numerous factors have been implicated in the etiopathogenesis of PD (Figure 4), providing a daunting list of candidate targets for a neuroprotective therapy. A full discussion of each is beyond the scope of this review and is extensively reviewed elsewhere (49). Mutations in several different genes have been linked to the development of PD in small numbers of familial cases (50). In contrast, twin studies suggest that genetic factors do not play a dominant role in the majority of patients who have a sporadic form of the disorder (51). Postmortem studies provide evidence that oxidative stress, excitotoxicity, mitochondrial dysfunction, and inflammation contribute to the pathogenetic cascade of events leading to neurodegeneration and that cell death in PD ultimately occurs by way of a signal-mediated apoptotic process (52). The extent to which these various factors contribute to cell death, however, may vary in individual cases. It

Figure 4 Mechanism of cell death in Parkinson’s disease.

5 Jan 2004

11:20

Annu. Rev. Med. 2004.55:41-60. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

52

AR

AR206-ME55-04.tex

AR206-ME55-04.sgm

LaTeX2e(2002/01/18)

P1: GBC

OLANOW

is thus becoming increasingly clear that PD is not a single disorder and probably results from a complex interplay between different genetic and environmental factors. Accordingly, it is uncertain whether a single strategy directed at a specific etiopathogenetic factor will be neuroprotective in all of the different forms of the disease. Recent studies suggest that a defect in the clearance of misfolded proteins by the ubiquitin proteasome system may be common to both the familial and sporadic forms of PD and could represent a common target for neuroprotective therapy in the different forms of the disorder (53). The first clinical study that tried to detect a neuroprotective effect in PD was the DATATOP study (54). Untreated PD patients were randomly assigned to receive treatment with the antioxidant vitamin E, the monoamine oxidase-B (MAO-B) inhibitor selegiline (Eldepryl®), or their placebos using a 2 × 2 factorial design. The primary outcome measure was time to deterioration necessitating the introduction of levodopa therapy. In this study, vitamin E in doses of 2000 IU per day was not superior to placebo whether administered alone or in combination with selegiline. In contrast, selegiline significantly delayed the emergence of disability. However, post hoc analyses demonstrated that selegiline had a symptomatic effect that confounded interpretation of the study. Thus, it was not possible to determine if the apparent delay in disease progression was due to a neuroprotective effect with slowing of neuronal degeneration or a symptomatic effect that simply masked ongoing neurodegeneration. To control for a possible symptomatic effect, a prospective double-blind controlled study was performed using as a primary endpoint the change in UPDRS motor score between an untreated baseline visit and an untreated final visit performed two months after withdrawal of the study intervention (55). In this study, deterioration from baseline was significantly greater in patients randomized to placebo than in those randomized to Selegiline. However, it is possible that the washout was too short, so that a protracted symptomatic effect might have confounded delineation of any neuroprotective benefit. Interestingly, laboratory studies demonstrate that selegiline-related neuroprotection does not depend on MAO-B inhibition and is related to its propargyl metabolite desmethylselegiline, which has an antiapoptotic effect (56). Two other propargylamine compounds, Rasagiline and TCH346, also show neuroprotective effects in the laboratory and are currently being tested in PD. The glutamate release inhibitor Riluzole has been tested using multiple primary endpoints, but the clinical trial was negative. More recently, a pilot study showed promising results with high doses of the bioenergetic agent coenzyme Q10, using change in UPDRS score between untreated baseline and initial treatment visits as the primary endpoint (57). However, because of the small sample size and mild symptomatic effects detected following introduction of the drug, it is impossible to state with certainty that benefits are due to a protective effect of the drug. To avoid confounding symptomatic effects in neuroprotective trials, studies have used surrogate neuroimaging markers of nigrostriatal function as the primary endpoint. One prospective double-blind trial used the rate of decline in striatal β-CIT uptake on SPECT, a measure of dopamine transporter density, to compare

5 Jan 2004

11:20

AR

AR206-ME55-04.tex

AR206-ME55-04.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:41-60. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

PARKINSON’S DISEASE

P1: GBC

53

Figure 5 Note reduced rate of decline in neuroimaging markers of the nigrostriatal system in patients treated with dopamine agonists versus levodopa (derived from References 58 and 59). It remains to be determined whether this difference is due to a protective effect provided by agonists, levodopa-induced toxicity, or pharmacologic differences in the capacities of these drugs to regulate measures of nigrostriatal function.

the rate of disease progression in patients randomized to initiate therapy with levodopa or the dopamine agonist pramipexole (58). A similar study compared the rate of decline in striatal fluorodopa uptake on positron emission tomography (PET) in untreated patients randomized to levodopa or ropinirole (59). Both studies demonstrated that patients randomized to initiate therapy with a dopamine agonist had a slower rate of decline in these measures of nigrostriatal function than those receiving levodopa (Figure 5). Because there was no placebo group in either of these studies, it is not possible to say if dopamine agonists are protective or if levodopa accelerates the loss of nigrostriatal function. A variety of in vitro and in vivo studies demonstrating the capacity of dopamine agonists to protect dopamine neurons suggest that dopamine agonists might be protective in PD (60). The results of the Elldopa study described above, however, raise the possibility that these differences may relate to levodopa toxicity (39). It is also possible that the study drugs tested differ in their capacity to regulate the components of the nigrostriatal system that were imaged and that the findings represent a pharmacologic rather than a neuroprotective effect (61). Thus, despite the many promising candidate drugs that might be neuroprotective in PD, none has been proven to alter disease progression, and physicians are forced to rely on their best judgment rather than clinical data in deciding whether to introduce a putative neuroprotective drug for PD. Further, none of the currently employed clinical or imaging outcome measures has been established to measure disease progression and accordingly are not presently accepted by regulators for

5 Jan 2004

11:20

54

AR

AR206-ME55-04.tex

AR206-ME55-04.sgm

LaTeX2e(2002/01/18)

P1: GBC

OLANOW

purposes of registration or labeling. The determination of an endpoint that reflects the rate of PD progression and will allow testing of the many putative neuroprotective drugs is an urgent priority.

Annu. Rev. Med. 2004.55:41-60. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Restorative Therapies An alternative approach to the treatment of PD involves transplantation strategies designed to replace dopamine neurons or to restore function in damaged cells. Experimentally, implanted dopamine neurons can survive, reinnervate the striatum, manufacture dopamine, and provide motor benefits to dopamine-lesioned rodents and monkeys (62). In open-label trials, fetal nigral transplantation has been reported to provide clinical benefit and to increase striatal fluorodopa uptake on PET (63, 64). Further, autopsy studies in PD patients demonstrated robust survival of implanted dopamine neurons with reinnervation of the striatum in an organotypic manner (65). However, these benefits were not reproduced in two double-blind, placebo-controlled trials. One study conducted in 40 advanced patients failed to meet its primary endpoint (a quality-of-life measure), although modest improvement was observed in UPDRS motor scores, particularly in patients younger than 60 years (66). Some of the transplanted patients developed a previously unreported and disabling form of dyskinesia that persisted for hours and even days after stopping levodopa (off-medication dyskinesia). The second study similarly did not did not detect significant transplant-related benefits with respect to the primary endpoint (UPDRS motor scores during the off period) in 34 patients who were randomized to receive tranplantation with one or four donors per side or a placebo procedure (66a). Post hoc analysis demonstrated significant improvements with transplantation in patients with milder disease, but no age-related benefits were detected. In this study, more than half of the transplanted patients developed offmedication dyskinesia, which was severe enough to require surgical intervention in three. It is not yet clear why implanted mesencephalic dopamine neurons in these studies failed to provide more dramatic clinical benefits despite evidence of graft survival on PET and at postmortem examination. Clearly, it will be necessary to find a way to expand on these clinical benefits and to prevent the development of off-medication dyskinesia if transplantation strategies are to become a viable treatment option. There is also considerable interest in the use of stem cells as a source of virtually unlimited numbers of standardized dopamine neurons for transplantation (67). Embryonic stem (ES) cells have the capacity to differentiate into dopamine neurons (68) and have been shown to provide motor benefits following transplantation into 6-hydroxydopamine lesioned rodents (69, 70). However, the number of surviving dopamine cells after grafting is small and benefits are less dramatic than those obtained with transplantation of differentiated fetal nigral cells. In addition, several animals developed intracerebral tumors following grafting (69) and the potential for dyskinesia formation remains unknown. It therefore remains to be established that ES cells will be useful as a treatment for PD. In addition, results in animal

5 Jan 2004

11:20

AR

AR206-ME55-04.tex

AR206-ME55-04.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:41-60. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

PARKINSON’S DISEASE

P1: GBC

55

models must be optimized and safety issues addressed before clinical trials can begin. An alternative approach involves the use of trophic factors and specifically glial cell line–derived neurotrophic factor (GDNF). GDNF has been shown to promote the survival and phenotypic differentiation of cultured fetal rat midbrain dopaminergic neurons (71). Further, direct striatal implantation of GDNF improves behavioral function and restores tyrosine hydroxylase immunostaining in MPTPtreated monkeys (72). An initial trial of GDNF administered by intraventricular infusion did not provide benefit for PD patients and was poorly tolerated (73), but it is likely that the GDNF did not cross the blood-brain barrier. More striking clinical benefits were reported in a small number of PD patients following direct infusion of GDNF into the striatum (74). Double-blind controlled trials are now testing striatal infusion of GDNF in PD patients. Interest has also begun to focus on the use of gene therapy to deliver trophic factors and other potentially therapeutic proteins to the PD brain. Lentivirus delivery of GDNF into the striatum of MPTP-treated monkeys was well tolerated and provided dramatic behavioral benefits and restoration of striatal and nigral dopaminergic innervation (75). Other gene therapy approaches that are currently being investigated in PD involve the use of adeno-associated virus (AAV) to deliver aromatic amino acid decarboxylase (AADC) into the striatum to promote a more continuous conversion of levodopa to dopamine and glutamate acid decarboxylase (GAD) into the STN to reduce neuronal hyperactivity. In addition to choosing the protein, vector, and target site, gene therapy teams will also have to consider whether to employ a regulatable system to limit excessive activity or distribution of the protein or virus.

SUMMARY Figure 6 presents an algorithm for the management of PD based on the principles discussed in this article. The determination that dopamine is reduced in the striatum of PD patients and the introduction of levodopa as a therapy represent landmark achievements in the clinical neuroscience of the twentieth century. However, chronic levodopa therapy is associated with motor complications that limit its effectiveness in the majority of patients. Recent studies indicate that motor complications in PD are related to abnormal pulsatile stimulation of striatal dopamine receptors with consequent intracellular gene and protein changes and altered firing patterns in basal ganglia output neurons. These findings have led to the notion that treating PD patients with therapies that provide more continuous dopaminergic stimulation might reduce the risk of causing motor complications. Indeed, prospective double-blind controlled trials in early PD patients demonstrate that initiating therapy with a long-acting dopamine agonist is associated with a significant reduction in the frequency and severity of motor complications in comparison to patients randomized to the shortacting regular formulation of levodopa.

5 Jan 2004

11:20

Annu. Rev. Med. 2004.55:41-60. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

56

AR

AR206-ME55-04.tex

AR206-ME55-04.sgm

LaTeX2e(2002/01/18)

P1: GBC

OLANOW

Figure 6 Treatment algorithm for the management of Parkinson’s disease. “Combination therapy” refers to dopamine agonist plus L-dopa. “Drug modification” refers to adjustments of dosage/frequency or drug type. (Adapted from Reference 1 with permission.)

Although these treatment strategies are beneficial in early PD, many patients already suffer motor complications that cannot be satisfactorily controlled with available treatment strategies. In this population of patients, DBS of the STN or GPi can reduce both “off” periods and dyskinesia and dramatically improve quality of life. Restorative therapies such as transplantation have provided promising results in the laboratory, but these have not been confirmed in double-blind trials in PD patients. They also illustrate the need to monitor for unanticipated side effects with these novel therapies. Despite major improvements in the treatment of PD, patients continue to progress and experience disability from features that do not respond to levodopa, such as dementia and postural instability. An intensive search is under way for a neuroprotective therapy that can slow or even stop disease progression. Laboratory studies have identified many promising candidates, but no agent has been proven to alter the natural history of PD, and clinicians have little guidance on whether to prescribe a putative neuroprotective agent. Indeed, no satisfactory marker of disease progression that can be used as a primary endpoint in clinical trials has yet

5 Jan 2004

11:20

AR

AR206-ME55-04.tex

AR206-ME55-04.sgm

LaTeX2e(2002/01/18)

P1: GBC

PARKINSON’S DISEASE

57

been defined. Nonetheless, it is encouraging that recent advances in the treatment of PD have been based on insights gained in the laboratory, and it is anticipated that the wealth of research presently being conducted in PD will lead to better treatments in the future. The Annual Review of Medicine is online at http://med.annualreviews.org

Annu. Rev. Med. 2004.55:41-60. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

LITERATURE CITED 1. Olanow CW, Watts RL, Koller WC. 2001. An algorithm (decision tree) for the management of Parkinson’s disease (2001): treatment guidelines. Neurology. 56(Suppl. 5):1–88 2. Lang AP, Lozano AE. 1998. Parkinson’s disease. N. Engl. J. Med. 339:1044–53 3. Forno LS. 1996. Neuropathology of Parkinson’s disease. J. Neuropathol. Exp. Neurol. 55:259–72 4. Ehringer H, Hornykiewicz O. 1960. Verteilung von Noradrenalin und Dopamin (3-hydroxytyramin) in Gehrin des Menschen und ihr Verhalten bei Erkrankungen des extrapyramidalen Systems. Klinische Wochenschrift 38:1238– 39 5. Hughes AJ, Daniel SE, Kilford L, et al. 1992. Accuracy of clinical diagnosis of idiopathic Parkinson’s disease: a clinicopathological study of 100 cases. J. Neurol. Neurosurg. Psychiatry 55:181–84 6. Hughes AJ, Ben-Shlomo Y, Daniel SE, et al. 1992. What features improve the accuracy of clinical diagnosis in Parkinson’s disease: a clinicopathologic study. Neurology 42:1142–46 7. Cotzias GC, Van Woert MH, Schiffer LM. 1967. Aromatic amino acids and modification of parkinsonism. N. Engl. J. Med. 276:374–79 8. Obeso JA, Olanow CW, Nutt JG. 2000. Levodopa motor complications in Parkinson’s disease. Trends Neurosci. 23(Suppl. 2):2–7 9. Olanow CW. 1990. Oxidation reactions in Parkinson’s disease. Neurology 40:32–37 10. Goetz C, Koller W, Poewe W, et al. 2002. Management of Parkinson’s dis-

11.

12.

13.

14.

15.

16.

17.

18.

19.

ease: an evidence-based review. Mov. Disord. 17(Suppl. 4):1–166 Albin RL, Young AB, Penney JB. 1989. The functional anatomy of basal ganglia disorders. TINS 12:366–75 DeLong MR. 1990. Primate models of movement disorders of basal ganglia origin. TINS 13:281–89 Gerfen CR. 2000. Molecular effects of dopamine on striatal-projection pathways. TINS 23:64–70 (Suppl.) Bergman H, Wichmann T, Karmon B, et al. 1994 The primate subthalamic nucleus. II. Neuronal activity in the MPTP model of parkinsonism. J. Neurophysiol. 72:507–20 Crossman AR, Mitchell IJ, Sambrook MA. 1985. Regional brain uptake of 2deoxy glucose in MPTP-induced parkinsonism in the Macaque monkey. Neuropharmacology 24:587–591 Eidelberg D, Moeller JR, Dhawan V, et al. 1994. Metabolic topography of parkinsonism. J. Cereb. Blood Flow Metab. 14:783–801 Bergman H, Wichmann T, DeLong MR. 1990. Reversal of experimental parkinsonism by lesions of the subthalamic nucleus. Science 249:1436–38 Brotchie JM, Mitchell IJ, Sambrook MA, et al. 1991. Alleviation of parkinsonism by antagonist of excitatory amino acid transmission in the medial segment of the globus pallidus in rat and primate. Mov. Disord. 6:133–38 Laitinen LV, Bergenheim AT, Hariz MI. 1992. Leksell’s posteroventral pallidotomy in the treatment of Parkinson’s disease. J. Neurosurg. 76:53

5 Jan 2004

11:20

Annu. Rev. Med. 2004.55:41-60. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

58

AR

AR206-ME55-04.tex

AR206-ME55-04.sgm

LaTeX2e(2002/01/18)

P1: GBC

OLANOW

20. Limousin P, Krack P, Pollack P, et al. 1998. Electrical stimulation of the subthalamic nucleus in advanced Parkinson’s disease. N. Engl. J. Med. 339:1105–11 21. Papa SM, Desimone R, Fioranni M, et al. 1999. Internal globus pallidus discharge is nearly suppressed during levodopainduced dyskinesia. Ann. Neurol. 46:732– 38 22. Hutchison WD, et al. 1997. Effects of apomorphine on globus pallidus neurons in Parkinsonian patients. Ann. Neurol. 42:767–75 23. Obeso JA, Rodriguez-Oroz MC, Rodriguez M, et al. 2000. Pathophysiology of levodopa-induced dyskinesias in Parkinson’s disease: problems with current models of the basal ganglia. Ann. Neurol. 47:22–34 24. Olanow CW, Obeso JA, Nutt J, eds. 2000. Basal ganglia, Parkinson’s disease, and levodopa therapy. Trends Neurosci. 23(10) (Suppl.) 25. Grace AA. 1991. Phasic versus tonic dopamine release and the modulation of dopamine system responsivity: a hypothesis for the etiology of schizophrenia. Neuroscience 41:1–24 26. Schultz W. 1998. Predictive reward signal of dopamine neurons. J. Neurosci. 80:1– 27.25 27. Olanow CW, Obeso JA. 2000. Preventing levodopa-induced dyskinesia. Ann. Neurol. 47:167–78 28. B´edard PJ, Di Paolo T, Falardeau P, et al. 1986. Chronic treatment with levodopa, but not bromocriptine induces dyskinesia in MPTP-treated parkinsonian monkeys. Correlation with [3H] spiperone binding. Brain Res. 379:294–99 29. Pearce RK, Banerji T, Jenner P, et al. 1998. De novo administration of ropinirole and bromocriptine induces less dyskinesia than levodopa in MPTPtreated marmoset. Mov. Disord. 13:234– 41 30. Jenner P. 2000. Factors influencing the onset and persistence of dyskinesia in

31.

32.

33.

34.

35.

36.

37.

38.

MPTP treated primates. Ann. Neurol. 47(4, Suppl. 1):90–99 Blanchet PJ, Calon F, Martel JC, et al. 1995. Continuous administration decreases and pulsatile administration increases behavioral sensitivity to a novel dopamine D2 agonist (U-91356A) in MPTP-exposed monkeys. J. Pharmacol. Exp. Ther. 272:854–59 Morisette M, Goulet M, Soghomonian JJ, et al. 1997. Preproenkephalin mRNA expression in the caudate-putamen of MPTP monkeys after chronic treatment with the D2 agonist U91356A in continuous or intermittent mode of administration: comparison with L-dopa therapy. Brain Res. Mol. Brain Res. 49:55–62 Rascol O, Brooks DJ, Korczyn AD, et al. 2000. A five year study of the incidence of dyskinesia in patients with early Parkinson’s disease who were treated with ropinirole or levodopa. N. Engl. J. Med. 342:1484–91 Parkinson Study Group. 2000. Pramipexole vs levodopa as initial treatment for Parkinson disease. JAMA 284:231–38 Rinne UK, Bracco F, Chouza C, et al. 1998. Early treatment of Parkinson’s disease with cabergoline delays the onset of motor complications. Results of a doubleblind levodopa controlled trial. Drugs 55(Suppl. 1):23–30 Olanow CW, Obeso JA. 2000. Pulsatile stimulation of dopamine receptors and levodopa-induced motor complications in Parkinson’s disease: implications for the early use of COMT inhibitors. Neurology 55(Suppl. 4):72–81 Jenner P, Al-Barghouthy G, Smith L, et al. 2002. Initiation of entacapone with L-DOPA further improves antiparkinsonian activity and avoids dyskinesia in the MPTP primate model of Parkinson’s disease. Neurology 58(Suppl. 3):374– 75 Albin RL, Frey KA. 2003. Initial agonist treatment of Parkinson disease. Neurology 60:390–94

5 Jan 2004

11:20

AR

AR206-ME55-04.tex

AR206-ME55-04.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:41-60. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

PARKINSON’S DISEASE 39. Parkinson Study Group. 2003. Does levodopa slow or hasten the rate of progression of Parkinson’s disease? The results of the Elldopa trial. Neurology 60(Suppl. 1):80–81 40. Agid Y, Olanow CW, Mizuno Y. 2002. Levodopa—why the controversy? Lancet 360:575 41. Lang AE, Lozano AM, Montgomery E, et al. 1997. Posteroventral medial pallidotomy in advanced Parkinson’s disease. N. Engl. J. Med. 337:1036–42 42. Vitek JL, Bakay RA, Freeman A, et al. 2003. Randomized trial of pallidotomy versus medical therapy for Parkinson’s disease. Ann. Neurol. 53:558–69 43. Benabid AL, Pollak P, Gervason C, et al. 1991. Long-term suppression of tremor by chronic stimulation of the ventral intermediate thalamic nucleus. Lancet. 337:403 44. Volkmann J, Sturm V, Weiss P, et al. 1998. Bilateral high frequency stimulation of the internal globus pallidus in advanced Parkinson’s disease. Ann. Neurol. 44:953– 61 45. The Deep Brain Stimulation for PD Study Group. 2001. Deep brain stimulation of the subthalamic nucleus or globus pallidus pars interna in Parkinson’s disease. N. Engl. J. Med. 345:956–63 46. Hadj Tahar A, Gregoire L, Bangassoro E, et al. 2000. Sustained cabergoline treatment reverses levodopa-induced dyskinesias in parkinsonian monkeys. Clin. Neuropharmacol. 23:195–202 47. Nutt JG, Obeso JA, Stocchi F. 2000. Continuous dopamine receptor stimulation in advanced Parkinson’s disease. Trends Neurosci. 23:109–15 48. Stocchi F, Ruggieri S, Vacca L, et al. 2002. Prospective randomized trial of lisuride infusion versus oral levodopa in PD patients. Brain 25:2058–66 49. Olanow CW, Schapira A, Agid Y. 2003. Causes of cell death and prospects for neuroprotection in Parkinson’s disease. Ann. Neurol. 53:1–170 (Suppl.) 50. Warner TT, Schapira AH. 2003. Ge-

51.

52.

53.

54.

55.

56.

57.

58.

59.

60.

P1: GBC

59

netic and environmental factors in the cause of Parkinson’s disease. Ann. Neurol. 53(Suppl. 3):16–23 Tanner CM, Ottman R, Goldman SM, et al. 1999. Parkinson disease in twins: an etiologic study. JAMA 281:341–46 Olanow CW, Tatton WG. 1999. Etiology and pathogenesis of Parkinson’s disease. Ann. Rev. Neurosci. 22:123–44 McNaught K St. P, Olanow CW, Halliwell B, et al. 2001. Failure of the ubiquitinproteasome system in Parkinson’s disease. Nat. Rev. Neurosci. 2:589–94 Parkinson’s Study Group. 1993. Effects of tocopherol and deprenyl on the progression of disability in early Parkinson’s disease. N. Engl. J. Med. 328:176–83 Olanow CW, Hauser RA, Gauger L, et al. 1995. A longtitudinal double blind controlled study of the affect of deprenyl and levodopa on the progression of the signs and symptoms of Parkinson’s disease. Ann. Neurol. 38:771–77 Tatton WG, Chalmers-Redman RME, Ju WJH, 2002. Propargylamines induce antiapoptotic new protein synthesis in serum- and nerve growth factor (NGF)withdrawn, NGF-differentiated PC-12 cells. J. Pharmacol. Exp. Ther. 301:753– 64 Shults CW, Oakes D, Kieburtz K, et al. 2002. Effect of coenzyme Q10 in early Parkinson’s disease: evidence of slowing of the functional decline. Arch. Neurol. 59:1541–50 Parkinson Study Group. 2002. Dopamine transporter brain imaging to assess the effects of pramipexole vs levodopa on Parkinson disease progression. JAMA 287:1653–61 Whone AL, Watts R, Stoessl J, et al. 2003. Slower progression in early Parkinson’s disease treated with ropinirole compared with L-DOPA: the REAL-PET study. Ann. Neurol. 54:93–101 Schapira AHV, Olanow CW. 2003. Rationale for the use of dopamine agonists as neuroprotective agents in Parkinson’s

5 Jan 2004

11:20

60

61.

62.

Annu. Rev. Med. 2004.55:41-60. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

63.

64.

65.

66.

66a.

67.

AR

AR206-ME55-04.tex

AR206-ME55-04.sgm

LaTeX2e(2002/01/18)

P1: GBC

OLANOW disease. Ann. Neurol. 53(Suppl. 3):149– 59 Ahlskog JE. 2003. Slowing Parkinson’s disease progression: recent dopamine agonist trials. Neurology. 60:381–89 Lindvall O, Hagell P. 2001. Cell therapy and transplantation in Parkinson’s disease. Clin. Chem. Lab. Med. 39:356–61 Lindvall O, Widner H, Rehncrona S, et al. 1992. Transplantation of fetal dopamine neurons in Parkinson’s disease: one-year clinical and neurophysiological observations in two patients with putaminal implants. Ann. Neurol. 31:155–65 Freeman TB, Olanow CW, Hauser RA, et al. 1995. Bilateral fetal nigral transplantation into the postcommissural putamen in Parkinson’s disease. Ann. Neurol. 38:379–88 Kordower JH, Freeman TB, Snow BJ, et al. 1995. Post mortem evidence of dopamine graft survival and striatal reinnervation in a Parkinson’s disease patient displaying improved motor function following fetal nigral transplantation. N. Eng. J. Med. 332:1118–24 Freed CR, Greene PE, Breeze RE, et al. 2001. Transplantation of embryonic dopamine neurons for severe Parkinson’s disease. N. Engl. J. Med. 344:710–19 Olanow CW, Goetz CG, Kordower JH, et al. 2003. A double blind controlled trial of fetal nigral transplantation in Parkinson’s disease. Ann. Neurol. 54:403–14 Lindvall O. 2003. Stem cells for cell therapy in Parkinson’s disease. Pharm. Res. 47:279–87

68. Lee SA, Lumelsky N, Studer L, et al. 2000. Efficient generation of midbrain and hindbrain neurons from mouse embryonic stem cells. Nat. Biotechnol. 18: 675–79 69. Bjorklund LM, S´anchez-Pernaute R, Chung S, et al. 2002. Embryonic stem cells develop into functional dopaminergic neurons after transplantation in a Parkinson rat model. Proc. Natl. Acad. Sci. USA. 99:2344–49 70. Kim J-H, Auerbach JM, RodriguezGomez JA, et al. 2002. Dopamine neurons derived from embryonic stem cells function in an animal model of Parkinson’s disease. Nature 418:50–56 71. Lin L-F, Doherty DH, Lile JD, et al. 1993. A glial cellline-derived neurotrophic factor for midbrain dopaminergic neurons. Science 260:1130–32 72. Gash DM, Zhang Z, Ovadia A, et al. 1996. Functional recovery in parkinsonian monkeys treated with GDNF. Nature 380:252– 55 73. Nutt JG, Burchiel KG, Comella CL, et al. 2003. Randomized double blind trial of glial cell line derived neurotrophic factor (GDNF) in PD. Neurology 14:69–73 74. Gill SS, Patel NK, Hotton GR, et al. 2003. Direct brain infusion of glial cell line derived neurotrophic factor in Parkinson’s disease. Nat. Med. 9:589–95 75. Kordower JH, Emborg ME, Bloch J, et al. 2000. Neurodegeneration prevented by lentiviral vector delivery of GDNF in primate models of Parkinson’s disease Science 290:767–73

Olanow.qxd

1/5/2004

3:28 PM

Page 1

Annu. Rev. Med. 2004.55:41-60. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

PARKINSON’S DISEASE

C-1

Figure 1 Low- and high-power photomicrographs of an intracellular Lewy body (arrow) within a melanized dopaminergic neuron in the substantia nigra pars compacta in a patient with Parkinson’s disease (H & E stain). Note that it comprises a dense core and a pale halo composed of protein aggregates and neurofilament, respectively. (Picture courtesy of Dr. Dan Perl.)

Annu. Rev. Med. 2004.55:41-60. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Olanow.qxd

C-2

1/5/2004 3:28 PM

OLANOW

See legend on next page

Page 2

Olanow.qxd

1/5/2004

3:28 PM

Page 3

Annu. Rev. Med. 2004.55:41-60. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

PARKINSON’S DISEASE

Figure 2 Classic model of the basal ganglia in the (a) normal, (b) parkinsonian, and (c) levodopa-induced dyskinesia states. The model proposes that the major input region of the basal ganglia (the striatum, comprising the putamen and caudate nucleus) is connected to the major output region (GPi and SNr) by a direct pathway and by an indirect pathway that has synaptic connections in the GPe and STN. (a) Dopamine neurons in the SNc act to excite inhibitory neurons in the direct pathway and inhibit the excitatory influence of the indirect pathway. (b) In Parkinson’s disease, the model proposes that dopamine depletion leads to overactivity in the GPi and SNr with excess inhibition of the thalamus, reduced activation of cortical motor regions, and the development of parkinsonian features. (c) In contrast, the model proposes that dyskinesia results from excess levels of dopaminergic activation causing suppression of firing in GPi and SNr with disinhibition of the thalamus and overexcitation of cortical motor regions. (d) It is now appreciated that this model is not completely correct (see text) and that the basal ganglia more likely represent a complex interactive network in a normal state. Plus and minus symbols indicate excitatory and inhibitory, respectively. SNc, substantia nigra pars compacta; SNr, substantia nigra pars reticularis; GPi, globus pallidus pars interna; GPe, globus pallidus pars externa; VL, ventrolateral thalamus; PPN, pedunculopontine nucleus; STN, subthalamic nucleus. (Adapted from Reference 24 with permission.)

C-3

12 Dec 2003

13:36

AR

AR206-ME55-05.tex

AR206-ME55-05.sgm

LaTeX2e(2002/01/18) P1: GBC 10.1146/annurev.med.55.091902.104408

Annu. Rev. Med. 2004. 55:61–95 doi: 10.1146/annurev.med.55.091902.104408 c 2004 by Annual Reviews. All rights reserved Copyright ° First published online as a Review in Advance on Oct. 6, 2003

PROGRESS IN ANTISENSE TECHNOLOGY Stanley T. Crooke

Annu. Rev. Med. 2004.55:61-95. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Isis Pharmaceuticals, 2292 Faraday Avenue, Carlsbad, California 92008; email: [email protected]

Key Words RNA, therapeutics, RNase H, RNAi, double-strand RNase ■ Abstract Antisense technology exploits oligonucleotide analogs to bind to target RNAs via Watson-Crick hybridization. Once bound, the antisense agent either disables or induces the degradation of the target RNA. Antisense agents can also alter splicing. During the past decade, much has been learned about the basic mechanisms of antisense, the medicinal chemistry, and the pharmacologic, pharmacokinetic, and toxicologic properties of antisense molecules. Antisense technology has proven valuable in gene functionalization and target validation. With one drug marketed, Vitravene®, and approximately 20 antisense drugs in clinical development, it appears that antisense drugs may prove important in the treatment of a wide range of diseases.

INTRODUCTION Antisense technology exploits oligonucleotide analogs (typically 15–20 nucleotides) to bind to cognate RNA sequences through Watson-Crick hybridization, resulting in the destruction or disablement of the target RNA. Thus, antisense technology represents a new pharmacology. The “receptor,” mRNA, has never before been considered in the context of drug receptor interactions. Prior to the advent of antisense technology, no medicinal chemistry had been practiced on oligonucleotides, the putative drugs. The basis of the drug receptor interaction, Watson-Crick hybridization, had never been considered as a potential binding event for drugs and put into a pharmacologic context. Finally, postbinding events such as recruitment of nucleases to degrade the receptor RNA had never been considered from a pharmacologic perspective. A key to understanding antisense technology is to consider it in pharmacologic context. It is essential to understand the structure, function, and metabolism of the receptors for antisense oligonucleotides, and to consider their effects in the context of dose-response curves. In the future, advances in antisense biology and medicinal chemistry will improve pharmacologic behaviors.

0066-4219/04/0218-0061$14.00

61

12 Dec 2003

62

13:36

AR

AR206-ME55-05.tex

AR206-ME55-05.sgm

LaTeX2e(2002/01/18)

P1: GBC

CROOKE

EVALUATION OF ANTISENSE DRUGS

Annu. Rev. Med. 2004.55:61-95. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Elucidating the mechanism(s) of action of any drug is challenging, but drugs of novel structure and action such as antisense oligonucleotides present a particularly difficult challenge. For RNase H and double-strand RNase-activating oligonucleotides, demonstration of reduction of target mRNA abundance by Northern blot, RT-PCR, or RNase protection assays, or transcriptional assay analyses are required. Ideally, demonstration that the levels of closely related mRNAs are unaffected should be included. In brief, the following data-gathering steps should be taken with any new antisense oligonucleotide: 1. Generate dose-response curves in vitro using several cell lines and methods of in vitro delivery. 2. Generate dose-response curves in vivo and correlate the rank-order potencies in vitro and in vivo. 3. Perform “gene walks” for all RNA species and oligonucleotide chemical classes, i.e, evaluate the effects of ASOs designed to bind to 40–80 sites in a target RNA, then select the best. 4. Perform time courses. 5. Directly demonstrate the proposed mechanism of action by measuring the target RNA and/or protein. 6. Evaluate specificity and therapeutic indices via studies on closely related mRNA isotypes and appropriate toxicologic studies. 7. Use RNase H protection assays and transcriptional arrays to provide broader analyses of specificity. 8. Perform sufficient pharmacokinetics to define rational dosing schedules for pharmacologic studies. 9. When control oligonucleotides display surprising activities, determine the mechanisms involved.

MOLECULAR MECHANISMS OF ANTISENSE DRUGS Occupancy-Only Mediated Mechanisms Classic competitive antagonists are thought to alter biological activities because they bind to receptors, thereby preventing agonists from binding to them. Binding of oligonucleotides to specific sequences may inhibit the interaction of the RNA with proteins, other nucleic acids, or other factors required for essential steps in the intermediary metabolism of the RNA or its utilization by the cell. This can inhibit processes such as translation or splicing, or alter the metabolism of the RNA (for review see 1).

12 Dec 2003

13:36

AR

AR206-ME55-05.tex

AR206-ME55-05.sgm

LaTeX2e(2002/01/18)

PROGRESS IN ANTISENSE TECHNOLOGY

P1: GBC

63

Occupancy-Inducted Destabilization

Annu. Rev. Med. 2004.55:61-95. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Any oligonucleotide analog that can form a duplex with the target RNA that is a substrate for a nuclease may destabilize a target RNA. Two groups of enzymes have been exploited widely to date: the RNases H and the double-strand RNases. ACTIVATION OF RNase H RNase H is a ubiquitous enzyme that degrades the RNA strand of an RNA-DNA duplex. It has been identified in organisms as diverse as viruses and human cells (2). At least two classes of RNase H have been identified in eukaryotic cells. Multiple enzymes with RNase H activity have been observed in prokaryotes (2). Although RNase H is involved in DNA replication, it may play other roles in the cell and is found in the cytoplasm as well as the nucleus (3). However, its concentration in the nucleus is thought to be greater, and some of the enzyme found in cytoplasmic preparations may be due to nuclear leakage. Recently, a knockout of the RNase H1 gene was shown to be embryonically lethal and to inhibit mitochondrial DNA synthesis (4). The precise recognition elements for RNase H are not known. However, oligonucleotides with DNA-like properties as short as tetramers can activate RNase H (5). Changes in the sugar influence RNase H activation, since sugar modifications that result in RNA-like oligonucleotides, e.g., 20 -fluoro or 20 -methoxy, do not appear to serve as substrates for RNase H (6, 7). Alterations in the orientation of the sugar to the base can also affect RNase H activation, as α-oligonucleotides are unable to induce RNase H or may require parallel annealing (8, 9). Additionally, backbone modifications influence the ability of oligonucleotides to activate RNase H. Methylphosphonates do not activate it (10, 11), whereas phosphorothioates are excellent substrates (12–14). In addition, chimeric molecules have been studied as oligonucleotides that bind to RNA and activate RNase H (15, 16). For example, oligonucleotides comprising wings of 20 -methoxy phosphonates and a five-base gap of deoxyoligonucleotides bind to their target RNA and activate RNase H. Furthermore, a single ribonucleotide in a sequence of deoxyribonucleotides was sufficient to serve as a substrate for RNase H when bound to its complementary deoxyoligonucleotide (17). It is possible to take advantage of chimeric oligonucleotides designed to activate RNase H, with greater affinity for their RNA receptors, and to enhance specificity (18, 19). In one study, RNase H–mediated cleavage of target transcript was much more selective when deoxyoligonucleotides consisting of methylphosphonate deoxyoligonucleotide wings and phosphodiester gaps were compared to full phosphodiester oligonucleotides (19). Given the emerging role of chimeric oligonucleotides with modifications in the 30 and 50 wings designed to enhance affinity for the target RNA and nuclease stability, and with a DNA-type gap to serve as a substrate for RNase H, studies focused on understanding the effects of various modifications on the efficiency of the enzyme(s) are also important. In one such study on E. coli RNase H the enzyme displayed minimal sequence specificity and was processive. When a chimeric

12 Dec 2003

Annu. Rev. Med. 2004.55:61-95. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

64

13:36

AR

AR206-ME55-05.tex

AR206-ME55-05.sgm

LaTeX2e(2002/01/18)

P1: GBC

CROOKE

oligonucleotide with 20 -modified sugars in the wings was hybridized to the RNA, the initial site of cleavage was the nucleotide adjacent to the methoxy-deoxy junction closest to the 30 end of the RNA substrate. The initial rate of cleavage increased as the DNA gap increased, and the enzyme was considerably less efficient against an RNA target duplexed with a chimeric antisense oligonucleotide than a full DNA-type oligonucleotide (20). Subsequent studies evaluated the interactions of antisense oligonucleotides with structured and unstructured targets, and detailed the impacts of these interactions on RNase H (21). Using a series of noncleavable substrates and Michaelis-Menten analyses, we evaluated both binding and cleavage and showed that, in fact, E. coli RNase H1 is a double-strand RNA binding protein. The Kd for the RNA duplex was 1.6 µM; the Kd for a DNA duplex was 176 µM; and the Kd for single-strand DNA was 942 µM. In contrast, the enzyme could only cleave RNA in an RNA-DNA duplex. Any 20 modification in the antisense drug at the cleavage site inhibited cleavage, but significant charge reduction and 20 modifications were tolerated at the binding site. Finally, placing a positive charge (e.g., 20 -propoxyamine) in the antisense drug reduced affinity and cleavage. We also examined the effects of antisense oligonucleotide-induced RNA structures on the activity of E. coli RNase H1 (21). Any structure in the duplex substrate had a significant negative effect on the cleavage rate. Further, cleavage of selected sites was inhibited entirely, and this was explained by steric hindrance imposed by the RNA loop traversing either the minor or major grooves of the heteroduplex. Recently, both human RNase H genes have been cloned and expressed (22, 23). The type 1 enzyme is a 286–amino acid protein with a calculated mass of 32 kDa. The enzyme is encoded by a single gene that is at least 10 kb long and expressed ubiquitously in human cells and tissues. The amino acid sequence of human RNase H1 displays strong homology with RNase H1 from yeast (21.8% amino acid identity), chicken (59%), Escherichia coli (33.6%), and mouse (84.3%). The type 1 enzymes are all small proteins (<40 kDa) and their estimated pI values are all 8, 0.7 and greater. The amino acid resides in E. coli RNase H1, which is thought to be involved in the Mg2+ binding site, catalytic center, and substrate binding region (24–26). The human RNase H2 enzyme is a 299–amino acid protein with a calculated mass of 33.4 kDa and is also ubiquitously expressed in human cells and tissues (22; Wu, unpublished data). Human RNase H2 shares strong amino acid sequence homology with RNase H2 from Caenorhabditis elegans (45.5% amino acid identity), yeast (25.7%), and E. coli (14.4%). Unlike the RNase H1 isotype, the type 2 enzyme is an acidic protein exhibiting a pI of 4.94. The properties of the cloned and expressed human RNase H1 have been characterized (27). The activity of the type 1 enzyme is Mg2+-dependent and inhibited by Mn2+ and the sulfhydryl blocking agent N-ethylmaleimide. Human RNase H1 was also inhibited by increasing ionic strength, with optimal activity for both KCl and NaCl observed at 10–20 mM. The enzyme exhibited a bell-shaped response to divalent cations and pH; the optimum conditions for catalysis were 1 mM Mg2+ and pH 7 to 8. The protein was shown to be reversibly denatured under the

12 Dec 2003

13:36

AR

AR206-ME55-05.tex

AR206-ME55-05.sgm

LaTeX2e(2002/01/18)

PROGRESS IN ANTISENSE TECHNOLOGY

P1: GBC

65

Annu. Rev. Med. 2004.55:61-95. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

influence of temperature and destabilizing agents such as urea. Renaturation of human RNase H1 was observed to be highly cooperative and did not require divalent cations. Human RNase H1 was shown to bind selectively to “A-form” duplexes with approximately 10- to 20-fold greater affinity than that observed for E. coli RNase H1 (21, 28). Finally, human RNase H1 displays a strong positional preference for cleavage, i.e., the enzyme cleaves between 8 and 12 nucleotides from the 50 -RNA-30 -DNA terminus of the duplex. ACTIVATION OF DOUBLE-STRAND RNases By using phosphorothioate oligonucleotides with 20 -modified wings and a ribonucleotide center, we have shown that mammalian cells contain enzymes that can cleave double-strand RNAs (29). This was important because it added to the repertoire of intracellular enzymes that may be used to cleave target RNAs, and because chimeric oligonucleotides with 20 modified wings and oligoribonucleotide gaps have higher affinity for RNA targets than chimeras with oligodeoxynucleotide gaps. More recently, the double-strand RNase mechanism, RNAi or siRNA, has generated excitement (30–40). Figure 1 compares RNase and RNAi mechanisms. At Isis Pharmaceuticals, we exploited double-strand RNases by chemically modifying single-strand RNA to stabilize it instead of using double-strand RNA. Recently, several comparisons of RNase H to RNAi as approaches for gene functionalization have been reported (41). Double-strand RNases, especially the processes involved in RNAi, have provided new opportunities to identify new mechanisms for antisense that may differ from RNase H.

CHARACTERISTICS OF PHOSPHOROTHIOATE OLIGONUCLEOTIDES Of the first-generation oligonucleotide analogs, the phosphorothioate class is the best known and has yielded the broadest range of activities. Phosphorothioate oligonucleotides were first synthesized in 1969, when a poly (rlrC) phosphorothioate was synthesized. This modification clearly achieves the objective of increased nuclease stability. In this class of oligonucleotides, one of the oxygen atoms in the phosphate group is replaced with a sulfur. The resulting compound is negatively charged, is chiral at each phosphorothioate phosphodiester, and is much more resistant to nucleases than the parent phosphorothioate (42).

Interactions with Proteins Phosphorothioate oligonucleotides bind to proteins. Their interactions with proteins can be divided into nonspecific, sequence-specific, and structure-specific binding events, each of which may have different characteristics and effects. Nonspecific binding to a wide variety of proteins has been demonstrated. Exemplary

12 Dec 2003

Annu. Rev. Med. 2004.55:61-95. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

66

13:36

AR

AR206-ME55-05.tex

AR206-ME55-05.sgm

LaTeX2e(2002/01/18)

P1: GBC

CROOKE

Figure 1a (1) The double-strand oligonucleotide passes through the cell membrane and enters the cytoplasm. (2) The helicase separates the sense and antisense strands of the oligonucleotide. (3) The RISC complex, an endogenous conglomerate of functional components, associates with the antisense oligonucleotide. (4) The antisense strand of the oligonucleotide binds (hybridization) to the target mRNA, forming a sense-antisense duplex. (5) The nuclease component of RISC is an endogenous nuclease that degrades the target mRNA. This inhibits target mRNA expression.

of this type of binding is the interaction of phosphorothioate oligonucleotides with serum albumin. The affinity of such interactions is low. The Kd for albumin is ∼200 µM, in a similar range with aspirin or penicillin (43, 44). Binding to serum protein, including albumin, prevents rapid clearance by glomerular filtration, thus providing the opportunity for these drugs to distribute to peripheral tissues. Phosphorothioates interact with many different proteins and these interactions play important roles in the distribution, clearance, and toxicological profiles of the

12 Dec 2003

13:36

AR

AR206-ME55-05.tex

AR206-ME55-05.sgm

LaTeX2e(2002/01/18)

67

Annu. Rev. Med. 2004.55:61-95. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

PROGRESS IN ANTISENSE TECHNOLOGY

P1: GBC

Figure 1b (1) The single-strand DNA oligonucleotide passes through the cell membrane and enters the cytoplasm. (2) The oligonucleotide enters the nucleus. (3) The oligonucleotide binds (hybridizes) to the target mRNA, forming a sense-antisense duplex. (4) The formation of the duplex initiates the recruitment of the RNase H enzyme, an endogenous nuclease. (5) RNase H degrades the target mRNA, inhibiting target mRNA expression. The oligonucleotide moves on and binds (hybridizes) to another target RNA.

12 Dec 2003

68

13:36

AR

AR206-ME55-05.tex

AR206-ME55-05.sgm

LaTeX2e(2002/01/18)

P1: GBC

CROOKE

drugs (44–57). In this regard, antisense drugs do not significantly differ from most drugs and the summation of these interactions is measured in the therapeutic index that the first-generation ASOs display.

Annu. Rev. Med. 2004.55:61-95. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Nuclease Stability The principal metabolic pathway for oligonucleotides is cleavage via endo- and exonucleases. Phosphorothioate oligonucleotides, though quite stable to various nucleases, are competitive inhibitors of nucleases (50, 58–61). Consequently, the stability of phosphorothioate oligonucleotides to nucleases is probably a bit less than initially thought, since early studies used concentrations of oligonucleotides that were high enough to inhibit nucleases. Similarly, phosphorothioate oligonucleotides are degraded slowly by cells in tissue culture with a half-life of 12–24 h and are slowly metabolized in animals (60, 62, 63). The pattern of metabolites suggests primarily exonuclease activity with perhaps modest contributions by endonucleases. However, several lines of evidence suggest that, in many cells and tissues, endonucleases are important in the metabolism of oligonucleotides. For example, 30 - and 50 -modified oligonucleotides with phosphodiester backbones are degraded relatively rapidly in cells and after administration to animals (64, 65). Thus, stability-enhancing strategies in which oligonucleotides are modified after only the 30 and 50 terminus have failed.

In Vitro Cellular Uptake Phosphorothioate oligonucleotides are taken up by a wide range of cells in vitro (50, 59, 66–68). In fact, uptake of phosphorothioate oligonucleotides into a prokaryote, Vibrio parahaemoyticus, has been reported, as has uptake into Schistosoma mansoni (69, 70). Uptake is time- and temperature-dependent. It is also influenced by cell type, cell-culture conditions, media and sequence, and length of the oligonucleotide (59). No correlation has been found between uptake and the lineage of cells, whether the cells are transformed or virally infected. Nor are the factors causing differences in uptake of different sequences of oligonucleotide understood. Although several studies have suggested that receptor-mediated endocytosis may be a significant mechanism of cellular uptake, the data are not yet conclusive (71). Numerous studies have shown that phosphorothioate oligonucleotides distribute broadly in most cells once taken up (59, 72). Again, however, significant differences in subcellular distribution between various types of cells are noted. Cationic lipids have been used to enhance uptake of phosphorothioate oligonucleotides in cells that take up little oligonucleotide in vitro (73–75). Here too there are substantial variations from cell type to cell type. Other approaches to enhanced intracellular uptake in vitro have included streptolysin D treatment of cells and the use of dextran sulfate and other liposome formulations, as well as physical means such as microinjections (59, 76, 77).

12 Dec 2003

13:36

AR

AR206-ME55-05.tex

AR206-ME55-05.sgm

LaTeX2e(2002/01/18)

PROGRESS IN ANTISENSE TECHNOLOGY

P1: GBC

69

Annu. Rev. Med. 2004.55:61-95. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

In Vivo Pharmacokinetics Phosphorothioate oligonucleotides bind to serum albumin and α-2 macroglobulin. The apparent affinity for albumin is quite low (200–400 µM), comparable to the low-affinity binding observed for a number of drugs, including aspirin and penicillin (62–64). Serum protein binding, therefore, provides a repository for these drugs and prevents rapid renal excretion. Because serum protein binding is saturable, at higher doses, intact oligomer may be found in urine (52, 78). Our studies suggest that in rats, oligonucleotides administered intravenously at doses of 15–20 mg/kg saturate the serum protein binding capacity (79). Phosphorothioate oligonucleotides are rapidly and extensively absorbed after parenteral administration. In rats, after an intradermal dose 3.6 mg/kg of 14CISIS 2105, a 20-mer phosphorothioate, ∼70% of the dose was absorbed within 4 h and total systemic bioavailability was in excess of 90% (80). Absorption of ISIS 2105 was similar after intradermal injection in humans (81). Subcutaneous administration to rats and monkeys results in somewhat lower bioavailability and greater distribution to lymph nodes, as would be expected (72). Distribution of phosphorothioate oligonucleotides from blood after absorption or intravenous (IV) administration is extremely rapid. Distribution half-lives of less than 1 h are reported (52, 62, 78, 80). Blood and plasma clearance is multiexponential, with a terminal elimination half-life of 40–60 h in all species except humans. In humans the terminal elimination half-life may be somewhat longer (82). Phosphorothioates distribute broadly to all peripheral tissues. Liver, kidney, bone marrow, skeletal muscle, and skin accumulate the highest percentage of a dose, but other tissues contain small quantities of drug (80, 83). No evidence of significant penetration of the blood-brain barrier has been reported. The rates of incorporation and clearance from tissues vary among organs, with the liver accumulating drug most rapidly (20% of a dose within 1–2 h). Similarly, elimination of drug from liver is rapid compared to many other tissues; the terminal half-life from liver is 62 h, whereas from renal medulla it is 156 h. The distribution into the kidney has been studied more extensively. Phosphorothioates were found in Bowman’s capsule, in the proximal convoluted tubule, in the bush-border membrane, and within renal tubular epithelial cells (84). The data suggested that the oligonucleotides are filtered by the glomerulus and then reabsorbed by the proximal convoluted tubule epithelial cells. Reabsorption might be mediated by interactions with specific proteins in the bush-border membranes (84). In addition, oligonucleotide is accumulated in a nonfiltering kidney, suggesting that there is uptake from the basal side also. Clearance of phosphorothioate oligonucleotides is due primarily to metabolism (62, 78, 80). Metabolism is mediated by exo- and endonucleases that result in shorter oligonucleotides and, ultimately, nucleosides that are degraded by normal metabolic pathways. Base excision or modification are theoretically possible, although no direct evidence of them has been reported. One study found, but did

12 Dec 2003

Annu. Rev. Med. 2004.55:61-95. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

70

13:36

AR

AR206-ME55-05.tex

AR206-ME55-05.sgm

LaTeX2e(2002/01/18)

P1: GBC

CROOKE

not fully characterize, a larger-molecular-weight radioactive material in urine (52). Clearly, the potential for conjugation reactions and extension of oligonucleotides via these drugs serving as primers for polymerases must be explored in more detail. In a very thorough study, 20 nucleotide phosphodiester and phosphorothioate oligonucleotides were administered intravenously to mice at a dose of 6 mg/kg. The oligonucleotides were internally labeled with 3H-CH3 by methylation of an internal deoxycytidine residue using Hha1 methylase and S-(3H) adenosyl methionine (85). The observed pharmacokinetic properties were consistent with our results as described above. Additionally, autoradiographic analyses showed drug in renal cortical cells (84, 85). A study that carefully measured intact drug by capillary gel electrophoresis determined the pharmacokinetics of Alicaforsen (a 20-mer phosphorothioate oligodeoxynucleotide) after a 2-h infusion. Doses from 0.06 mg/kg to 2.0 mg/kg were studied. Peak plasma concentrations increased linearly with dose; the 2-mg/kg dose resulted in peak plasma concentrations of intact drug of ∼9.5 µg/ml. Clearance from plasma, however, was dose-dependent. The clearance of the 2-mg/kg dose was 1.28 ml min–1 kg–1, whereas that of 0.5 mg/kg was 2.07 ml min–1 kg–1. Essentially no intact drug was found in urine. In addition to the pharmacologic effects of phosphorothioate oligonucleotides on animals and humans, several other lines of evidence show that these drugs enter cells in organs. Autoradiographic, fluorescent, and immunohistochemical approaches have shown that these drugs are localized in endopromal convoluted tubular cells, various bone marrow cells, skin cells, and liver cells (83, 84, 86). Perhaps more compelling and of more long-term value is a study showing the distribution of phosphorothioate oligonucleotides in the liver of rats treated intravenously at various doses (87). The kinetics and extent of the accumulation into Kuppfer, endothelial, and hepatocyte cell populations varied, and as doses were increased, the distribution changed. Moreover, subcellular distribution was significant and somewhat dependent on the cell type. In summary, pharmacokinetic studies of several phosphorothioates demonstrate that they are well absorbed from parenteral sites, distribute broadly to many peripheral tissues, do not cross the blood-brain barrier, and are eliminated primarily by nuclease metabolism. In short, once-daily or every-other-day systemic dosing is feasible. Although the similarities between oligonucleotides of different sequences far outweigh the differences, additional studies are required to see whether sequence exerts subtle effects on the pharmacokinetic profile of this class of drugs.

Modes of Administration Phosphorothioate oligodeoxynucleotides are attractive for inhalation delivery to the lung and upper airway (71, 88, 89). Target reduction in the lung has been demonstrated, and these drugs have been shown to distribute broadly to all cell types in the lung after aerosol administration. Further, these drugs are well tolerated at doses up to 12 mg/kg (71).

12 Dec 2003

13:36

AR

AR206-ME55-05.tex

AR206-ME55-05.sgm

LaTeX2e(2002/01/18)

PROGRESS IN ANTISENSE TECHNOLOGY

P1: GBC

71

Annu. Rev. Med. 2004.55:61-95. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Phosphorothioate oligonucleotides in a simple cream formulation penetrate normal mouse, pig, and human skin, and they penetrate and accumulate in human psoriatic skin grown on nude mice 6.5.2a, 6.5.2b. Further, a phase IIa study in patients with plaque psoriasis showed that Alicaforsen, an inhibitor of intercellular adhesion molecule 1 (ICAM-1), accumulated throughout the dermis and epidermis after topical administration and resulted in a positive trend in the primary endpoint, induration (Kruger, unpublished observations). Studies have also demonstrated reduction of targets such as ICAM-1, B71, and B72 (for review see 90). Alicaforsen has also been administered by enema and shown to be effective in treating ulcerative colitis (91).

PHARMACOLOGIC PROPERTIES Molecular Pharmacology Although there are multiple mechanisms by which an oligonucleotide may terminate the activity of an RNA species to which it binds, evidence has been reported for only three of these mechanisms. Antisense oligonucleotides have been reported to inhibit RNA splicing, affect translation of mRNA, and induce degradation of RNA by RNase H (92–94). Without question, the mechanism that has resulted in the most potent compounds and is best understood is RNase H activation. To serve as a substrate for RNase H, a duplex between RNA and “DNA-like” oligonucleotide is necessary. Specifically required are a charged phosphate and a sugar moiety in the oligonucleotide that induces a duplex conformation equivalent to that of a DNA-RNA duplex (95). Thus, phosphorothioate oligodeoxynucleotides are expected to induce RNase H-mediated cleavage of the RNA when bound. Many chemical approaches that enhance the affinity of an oligonucleotide for RNA result in duplexes that are no longer substrates for RNase H. Selection of sites for induction of optimal antisense activity in an RNA molecule is dependent on the terminating mechanism and influenced by the chemical class of the oligonucleotide. Each RNA appears to display unique patterns of sites of sensitivity. Within the phosphorothioate oligodeoxynucleotide chemical class, antisense activity can vary from undetectable to 100% by shifting an oligonucleotide by just a few bases in the RNA target (92, 96, 97). Despite significant progress in developing general rules that help define potentially optimal sites in RNA species, to a large extent, this remains an empirical process that must be performed for each RNA target and every new chemical class of oligonucleotides. Phosphorothioates have also shown effects inconsistent with the antisense mechanism for which they were designed. Some of these effects are due to sequence or are structure-specific. Others are due to nonspecific interactions with proteins. These effects are particularly prominent in in vitro tests for antiviral activity, where high concentrations of cells, viruses, and oligonucleotides are often coincubated (98, 99). Human immune deficiency virus (HIV) is particularly problematic because many oligonucleotides bind to the gp120 protein (56). However,

12 Dec 2003

Annu. Rev. Med. 2004.55:61-95. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

72

13:36

AR

AR206-ME55-05.tex

AR206-ME55-05.sgm

LaTeX2e(2002/01/18)

P1: GBC

CROOKE

the potential for confusion arising from the misattribution of an activity to an antisense mechanism when, in fact, it is due to nonantisense effects is certainly not limited to antiviral or in vitro tests (100–102). These data urge caution and argue for careful dose-response curves, direct analyses of target protein or RNA, and inclusion of appropriate controls before drawing conclusions concerning the mechanisms of action of oligonucleotide-based drugs. In addition to protein interactions, other factors, such as overrepresented sequences of RNA and unusual structures that may be adopted by oligonucleotides, can contribute to unexpected results (56). Given the variability in cellular uptake of oligonucleotides, the variability in potency as a function of binding site in an RNA target, and the potential nonantisense activities of oligonucleotides, careful evaluation of dose-response curves and clear demonstration of the antisense mechanism are required before drawing conclusions from in vitro experiments. Nevertheless, numerous well-controlled studies have conclusively demonstrated antisense activity against a variety of targets (12, 51, 96, 103, 104).

In Vivo Pharmacologic Activities In vivo activities of phosphorothioate oligonucleotides after both local and systemic administration have been reported (for review see 105). Here I review only the few reports that provide sufficient data to support a relatively firm conclusion about mechanism of action. Local effects have been reported for phosphorothioate and methylphosphonate oligonucleotides. In one study (106), a phosphorothioate oligonucleotide designed to inhibit c-myb production and applied locally inhibited intimal accumulation in the rat carotid artery. A Northern blot analysis showed a significant reduction in c-myb RNA in animals treated with the antisense compound, a control oligonucleotide had no effect. A study attributed the effects of the oligonucleotide to a nonantisense mechanism (101). However, only one dose level was studied, so definitive conclusions are not yet possible. Similar effects were reported for phosphorothioate oligodeoxynucleotides designed to inhibit cyclin-dependent kinases (CDC-2 and CDK-2). Again, the antisense oligonucleotide inhibited intimal thickening and cyclin-dependent kinase activity, whereas a control oligonucleotide had no effect (107). Additionally, local administration of a phosphorothioate oligonucleotide designed to inhibit N-myc reduced N-myc expression and slowed the growth of a subcutaneously transplanted human tumor in nude mice (108, 109). Antisense oligonucleotides administered intraventricularly have induced various effects in the central nervous system. Intraventricular injection of antisense oligonucleotides to neuropeptide-y-y1 receptors reduced the density of the receptors and caused anxious behavior (109). Similarly, an antisense oligonucleotide designed to bind to NMDA-R1 receptor channel RNA inhibited the synthesis of these channels and reduced the volume of focal ischemia produced by occlusion of the middle cerebral artery in rats (110).

12 Dec 2003

13:36

AR

AR206-ME55-05.tex

AR206-ME55-05.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:61-95. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

PROGRESS IN ANTISENSE TECHNOLOGY

P1: GBC

73

In a series of well-controlled studies, intraventricular administration of antisense oligonucleotides selectively inhibited dopamine type-2 receptor expression, dopamine type-2 receptor RNA levels, and behavioral effects in animals with chemical lesions. Controls included randomized oligonucleotides and the observation that no effects were observed on dopamine type-1 receptor or RNA levels (16–18). This laboratory also reported the selective reduction of dopamine type-1 receptor and RNA levels with the appropriate oligonucleotide (111). Numerous other targets have been inhibited after central nervous system administration as well (for review see 112). In rabbits, aerosol administration of an antisense phosphorothioate oligodeoxynucleotide designed to inhibit the production of adenosine A1 receptor reduced receptor numbers in the airway smooth muscle and inhibited adenosine, house dust-mite allergen, and histamine-induced bronchoconstriction (113). Neither a control nor an oligonucleotide complementary to bradykinin B2 receptors reduced the density of adenosine A1 receptors, but oligonucleotides complementary to bradykin in B2 receptor mRNA did. In addition to local and regional effects of antisense oligonucleotides, wellcontrolled studies have demonstrated systemic effects of phosphorothioate oligodeoxynucleotides. Expression of interleukin-1 in mice was inhibited by systemic administration of antisense oligonucleotides (113). Oligonucleotides to the NF-κB p65 subunit administered intraperitoneally at 40 mg/kg every 3 days slowed tumor growth in mice transgenic for the human T-cell leukemia viruses (21). Another in vivo tumor model yielded similar results after either prolonged subcutaneous infusion or intermittent subcutaneous injection of other antisense oligonucleotides (114). Several recent reports further explore phosphorothioate oligonucleotides as antitumor agents in mice. In one study, a phosphorothioate oligonucleotide directed to inhibit the bcr-abl oncogene was administered intravenously at a dose of 1 mg/day for 9 days to immunodeficient mice injected with human leukemic cells. The drug inhibited the development of leukemic colonies in the mice and selectively reduced bcr-abl RNA levels in peripheral blood lymphocytes, spleen, bone marrow, liver, lungs, and brain (115). However, the effects on the RNA levels could have been secondary to effects on the growth of various cell types. In the second study, a phosphorothioate oligonucleotide antisense to the protooncogene myb inhibited the growth of human melanoma in mice. Again, myb mRNA levels appeared to be selectively reduced (116). A number of studies have directly examined target RNA levels, target protein levels, and pharmacologic effects using a wide range of control oligonucleotides, and have examined the effects on closely related isotypes. Single and chronic daily administration of a phosphorothioate oligonucleotide designed to inhibit mouse protein kinase C-α (PKC-α) selectively inhibited expression of PKC-α RNA in mouse liver without effects on any other isotype. The effects lasted at least 24 h after a dose, and a clear dose-response curve was observed, with an intraperitoneal dose of 10–15 mg/kg reducing PKC-α RNA levels in liver by 50% after 24 h (117).

12 Dec 2003

Annu. Rev. Med. 2004.55:61-95. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

74

13:36

AR

AR206-ME55-05.tex

AR206-ME55-05.sgm

LaTeX2e(2002/01/18)

P1: GBC

CROOKE

IV administration of a phosphorothioate oligonucleotide designed to inhibit human PKC-α expression selectively inhibited expression of PKC-α RNA and PKC-α protein in human tumor cell lines implanted subcutaneously in nude mice (118). In these studies, effects on RNA and protein levels were highly specific and observed at doses lower than 6 mg/kg. A large number of control oligonucleotides failed to show activity. In a similar series of studies, Monia et al. demonstrated highly specific loss of human c-raf kinase RNA in human tumor xenografts and antitumor activity that correlated with the loss of RNA (119, 120). Finally, a single injection of a phosphorothioate oligonucleotide designed to inhibit c-AMP-dependent protein kinase type 1 was reported to selectively reduce RNA and protein levels in human tumor xenografts and to reduce tumor growth (121). This growing body of evidence indicates that phosphorothioate oligonucleotides can induce potent systemic and local effects in vivo. More importantly, there have been several studies with sufficient controls and direct observation of target RNA and protein levels to suggest highly specific effects that are difficult to explain via any mechanism other than antisense. As would be expected, the potency of these effects varies with the target, the organ, and the endpoint measured, as well as the route of administration and the time after a dose when the effect is measured. In conclusion, although in vivo activity data must be interpreted cautiously, and it is clearly necessary to include a range of controls and to evaluate effects on target RNA and protein levels and control RNA and protein levels directly, some effects have been observed in animals that are most likely primarily due to an antisense mechanism.

CLINICAL ACTIVITIES A supplemental table (see Supplemental Material at http://www.annualreviews.org/ supmat/supmat.asp) provides a comprehensive summary of all the clinical reports on antisense drugs. VitraveneTM, a phosphorothioate oligonucleotide designed to inhibit cytomegalovirus-induced retinitis, has been shown to be safe and effective in the treatment of this disease after intravitreal injection and has been approved by regulatory agencies in the United States, Europe, and South America (for review see 122).

Alicaforsen Alicaforsen (Isis Pharmaceuticals, Carlsbad, CA) is a phosphorothioate oligonucleotide inhibitor of human ICAM-1. It reduces ICAM-1 levels in several organs in various animals and has potent antiinflammatory effects (for review see 123). A number of clinical trials have evaluated Alicaforsen in patients with inflammatory diseases (for review see 124). The drug was evaluated in a randomized

12 Dec 2003

13:36

AR

AR206-ME55-05.tex

AR206-ME55-05.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:61-95. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

PROGRESS IN ANTISENSE TECHNOLOGY

P1: GBC

75

placebo-controlled phase IIa trial in 43 patients with moderate to severe rheumatoid arthritis (125). The effects of 0.5, 1.0, and 2.0 mg/kg IV three times weekly for one month were compared to placebo. The 0.5 mg/kg and 2.0 mg/kg doses resulted in prolonged improvement in rheumatoid arthritis activity. All doses were extremely well tolerated. Alicaforsen was evaluated as a potential therapeutic for patients with psoriasis (126). This study compared IV and topical routes of administration. The dosing for the IV study was equivalent to that used for the rheumatoid arthritis study, but only 17 patients were evaluated. Again, all doses were well tolerated. At the end of the treatment period, ∼30% reduction in psoriasis activity index score was reported (126). Additionally, several concentrations of Alicaforsen, applied topically in a simple cream formulation, were evaluated in patients with moderate plaque psoriasis. The drug was applied once a day for 3 months, and patients were followed for a total of 6 months. In approximately half the patients, skin biopsies were taken. Alicaforsen achieved very high dermal and epidermal concentrations that increased as the concentration in the cream was increased. At 2% and 4% concentrations, the creams produced dermal concentrations similar to those at which pharmacologic effects were observed in animals. In regard to the primary endpoint, induration, there was trend (p = 0.053) in favor of increasing Alicaforsen concentrations. Again the drug was very well tolerated. Alicaforsen has been evaluated most extensively in Crohn’s disease. In the initial placebo-controlled study in steroid-dependent patients, Alicaforsen reduced symptoms, steroid use, and ICAM-1 levels in the small intestines of patients treated and evaluated with serial colonoscopies. The duration of effect after a month of dosing was in excess of 6 months (127). In a subsequent 300-patient trial, in which patients were treated with placebo or 2 mg/kg every other day for 2 or 4 weeks, again the drug was extremely well tolerated. A population pharmacokinetic study demonstrated that heavier patients and women achieved greater exposures to the drug, and in the patients in the upper two quartiles of drug exposure, the drug produced a statistically significant increase in complete remissions compared to placebo (128). More recently, a higher-dose phase II study in Crohn’s showed that the higher doses were well tolerated and active (129). Additional phase III trials at higher doses are in progress (91). In addition, Alicaforsen has been administered as a retention enema to patients with ulcerative colitis. A randomized double-blind trial demonstrated a clear doseresponse relationship, and the top dose was extremely active. All treated patients responded. The mean reduction in the ulcerative colitis disease index was 73% versus 23% for placebo (p = 0.004). All patients experienced prolonged benefit with no relapses for 6 months after 1 month of dosing (91). In a related syndrome, pouchitis, Alicaforsen has shown considerable promise. Pouchitis occurs in patients with ulcerative colitis who have undergone resection of their colon, reanastomosis of the small bowel to the rectum, and construction of a pseudocolon or pouch. A significant fraction of these patients developed inflammation in the pouch. Alicaforsen provided a highly statistically significant

12 Dec 2003

76

13:36

AR

AR206-ME55-05.tex

AR206-ME55-05.sgm

LaTeX2e(2002/01/18)

P1: GBC

CROOKE

benefit in these patients (91). Thus, evidence indicates that Alicaforsen is active in inflammatory disease when administered intravenously, topically, or by enema.

Annu. Rev. Med. 2004.55:61-95. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Affinitak Affinitak (Isis Pharmaceuticals, Carlsbad, CA) is a phosphorothioate oligodeoxynucleotide inhibitor of protein kinase Cα (for review see 130). This drug has been evaluated as a single agent in a variety of solid tumors, using a range of dose schedules at doses as high as 30 mg/kg/day, and has been well tolerated. The supplemental table shows results in patients with advanced chemoresistant malignancies. Affinitak has also been evaluated in combination with various chemotherapeutic agents in a number of solid tumors. In a phase I/II study in 53 patients with non–small-cell carcinoma of the lung, Affinitak was associated with a substantial improvement in survival (16 months) compared to historical controls (131). Based on these data, a large phase III program was designed. The first study in the phase III program was a 616-patient randomized trial in patients with stage IIIb or IV non–small-cell carcinoma of the lung. Patients were randomized to receive either standard doses of carboplatinum-taxol or carboplatinumtaxol plus Affinitak. Affinitak was administered via an indwelling central venous catheter as a continuous IV infusion of 2 mg/kg/day for 2 weeks out of every 3-week cycle. Six cycles were considered a complete course of therapy. Although the Affinitak arm did not achieve the primary statistical endpoint, substantial evidence suggested activity (132). Table 1 analyzes survival as a function of whether patients were able to complete all six cycles of therapy. Note that survival improved in both the “completers” and “noncompleters.” Figure 2 shows the survival of completers in the two groups. These analyses suggest that the survival advantage afforded by Affinitak was total-dose–related and potentially significant. Affinitak, when added to carboplatinum and taxol, was well tolerated. Aside from the expected central-line complications, the only clinically meaningful observation was an increase in mild to moderate thrombocytopenia. A second study, in which Affinitak is added to gemcitabine and cisplatinum in patients with non–small-cell carcinoma of the lung, is in progress and results are expected in 2004.

TABLE 1 Summary of major survival parameters Log Rank Test Exp.

ITT overall Completers Noncompleters

Control

Median

n

Median

N

p Value

10.0 17.3 6.5

309 114 195

9.7 14.4 5.2

307 142 165

0.8054 0.0548 0.6406

12 Dec 2003

13:36

AR

AR206-ME55-05.tex

AR206-ME55-05.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:61-95. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

PROGRESS IN ANTISENSE TECHNOLOGY

P1: GBC

77

Figure 2 Patients who completed therapy appeared to experience Affinitak-associated survival benefit relative to controls.

GeneSense GeneSense is a phosphorothioate oligodeoxynucleotide designed to inhibit BCL2. In a non-Hodgkin’s lymphoma trial, this single agent reduced BCL2 levels in peripheral blood cells (133). GeneSense is also being studied in combination with chemotherapeutic agents. A phase I study with decarbazine in patients with malignant melanoma reported a 21% response rate in 14 patients (134). This drug is being evaluated in other combinations for other malignant diseases.

ISIS 2503 ISIS 2503 is a phosphorothioate oligodeoxynucleotide that inhibits Ha-ras. It is currently completing phase II single-agent and combination studies in patients with various solid malignancies (for review see 130). It is well tolerated and shows preliminary suggestions of activity (see supplemental table at http://www.annualre views.org/supmat/supmat.asp). More recently, ISIS 2503 was combined with gemcitabine in a 40-patient phase II study in patients with pancreatic cancer. The combination was well tolerated and displayed potential benefit.

ISIS 5132 ISIS 5231 is a phosphorothioate oligodeoxynucleotide inhibitor of c-raf kinase. In phase I studies, this drug reduced c-raf kinase levels in peripheral blood cells and displayed activity in patients with ovarian cancer (135). The phase II evaluation, however, failed to show meaningful activity for the drug (for review see 130).

12 Dec 2003

78

13:36

AR

AR206-ME55-05.tex

AR206-ME55-05.sgm

LaTeX2e(2002/01/18)

P1: GBC

CROOKE

TOXICOLOGY

Annu. Rev. Med. 2004.55:61-95. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Genotoxicity Concerns about the genotoxicity of phosphorothioate oligodeoxynucleotides cannot be dismissed; our understanding of the basic mechanisms of action is limited, little in vitro testing has been performed, and no data from long-term studies of oligonucleotides are available. We have performed mutagenicity studies on five phosphorothioate oligonucleotides [Alicaforsen, ISIS 2105, ISIS 5132, ISIS 14803, and VitraveneTM (Isis Pharmaceuticals, Carlsbad, CA)] and found them to be nonmutagenic at all concentrations studied (82, 136). Two mechanisms of genotoxicity that may be unique to oligonucleotides have been considered. One possibility is that an oligonucleotide analog could be integrated into the genome and produce mutagenic events. Although such integration is conceivable, it is likely to be extremely rare. For most viruses, viral DNA integration is itself a rare event and, of course, viruses have evolved specialized enzyme-mediated mechanisms to achieve integration. Moreover, our preliminary studies have shown that phosphorothioate oligodeoxynucleotides are generally poor substrates for DNA polymerases, and enzymes such as integrases, gyrases, and topoisomerases (which have obligate DNA cleavage as intermediate steps in their enzymatic processes) are unlikely to accept these compounds as substrates. Consequently, it would seem that the risk of genotoxicity due to genomic integration is no greater than and probably less than that of other potential mechanisms (e.g., alteration of the activity of growth factors, cytokine release, and nonspecific effects on membranes that might trigger arachidonic acid release or inappropriate intracellular signaling). Presumably, new analogs that deviate significantly more from natural DNA would be even less likely to be integrated. A second genotoxicity concern is that oligonucleotides might be degraded to toxic or carcinogenic metabolites. However, metabolism of phosphorothioate oligodeoxynucleotides by base excision would release normal bases, which presumably would be nongenotoxic. Similarly, oxidation of the phosphorothioate backbone to the natural phosphodiester structure would also yield nonmutagenic (and probably nontoxic) metabolites. Finally, it is possible that phosphorothioate bonds could be hydrolyzed slowly, releasing nucleoside phosphorothioates that presumably would be rapidly oxidized to natural (nontoxic) nucleoside phosphates. However, oligonucleotides with modified bases and/or backbones may pose different risks.

Acute and Transient Toxicities In Vivo COMPLEMENT ACTIVATION Rapid infusion of phosphorothioate oligodeoxynucleotides in nonhuman primates can result in cardiovascular collapse (137–139). Complement activation is necessary but not sufficient to produce the observed cardiovascular effects (for review see 46). The other factors contributing to cardiovascular collapse are not yet fully elucidated. However, it has been suggested

12 Dec 2003

13:36

AR

AR206-ME55-05.tex

AR206-ME55-05.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:61-95. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

PROGRESS IN ANTISENSE TECHNOLOGY

P1: GBC

79

that dosing monkeys that are restrained may exacerbate cardiovascular events. Activation of the complement cascade due to activation of the alternate pathway is insensitive to sequence, but is absolutely related to peak plasma concentration (140). The threshold concentration for activation of complement in the monkey is 40–50 µg/ml phosphorothioate oligodeoxynucleotide, and once the threshold is reached, variable but potentially dangerous levels of complement activation are observed. Studies in humans have avoided peak plasma concentrations that would induce complement. However, it appears that monkeys are substantially more sensitive to complement activation than humans. A comparison of the effects on complement activation in human versus monkey serum demonstrates a dramatic difference. In monkey serum, phosphorothioate oligodeoxynucleotides activate complement in a concentration-dependent fashion. It human serum, complement activation is actually inhibited at higher concentrations of oligonucleotide (46). This effect is thought to be due to the sensitivity of human serum to inhibition of complement activators. The mechanism of complement activation is currently believed to be an interaction with factor H (46). These effects can be reduced by chemical modifications and formulations that reduce plasma protein binding (46). INHIBITION OF CLOTTING In all species studies, phosphorothioate oligodeoxynucleotides induce a transient, apparently self-limited, peak plasma concentration– related inhibition of clotting, manifested as an increase in activated partial thromboplastin time (aPTT) (141–144). Increases in aPTT are minimally affected by sequence, but the effects are directly proportional to the length of the phosphorothioate oligodeoxynucleotides (46). The mechanism of aPTT increase seems to be an interaction with the intrinsic tenase complex (145, 146). This complex interaction involves effects on multiple clotting factors, including factors VIIIa, IXa, and X. The effects on clotting are transient, appear to be self-limited, and have not caused bleeding diatheses in animals or humans. Clotting inhibition can be ameliorated by chemical modifications such as 20 -O-(methoxyethyl) (20 -O-MOE) substitution (46).

Subchronic Toxicities IMMUNE STIMULATION In rodents, the most prominent toxicity is immune stimulation. This is also frequently a confounding variable that must be evaluated with regard to the mechanism of action of pharmacologic effects (for review see 136). Subchronic administration of doses as low as 10 mg/kg/day in rodents results in splanomegaly, lymphoid hyperplasia, and diffuse multiorgan mixed mononuclear cell infiltrates (140, 147–149). These effects are reminiscent of stimulator effects observed on isolated splemocytes (150–152). In fact, these in vitro studies are reasonably predictive of the relative potencies of phosphorothioates in moving in vivo immune stimulation. Immune stimulation in rodents is a property common to all phosphorothioate oligodeoxynucleotides, but potency varies substantially as

12 Dec 2003

Annu. Rev. Med. 2004.55:61-95. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

80

13:36

AR

AR206-ME55-05.tex

AR206-ME55-05.sgm

LaTeX2e(2002/01/18)

P1: GBC

CROOKE

a function of sequence. Rodent immune stimulatory motifs include palindromic sequences and CG motifs (153, 154). In primates, immune stimulation is far less prominent. Doses that produce immune stimulation in monkeys have not been identified despite evaluation of numerous oligonucleotides at a variety of doses and schedules (46). At least one factor contributing to this species difference is that the optimal rodent immune stimulator sequence is simple and immune stimulation is induced by all sequences. In contrast, in primates, the optimal sequences are different and more complex (155) and the response is much less promiscuous with regard to sequence. Modifications of the base, sugar, and backbone can reduce immune stimulation in rodents. For example, 50 -methycytosine and 20 -O-MOE containing oligonucleotides display substantially reduced potency for immune stimulation in rodents (46).

Other Toxicities In rodents but not in monkeys, single-cell hepatolyte neurons are occasionally observed and have been related to immunostimulation. In monkeys, transient thromobcytopenia is occasionally observed, perhaps associated with complement activation (for review see 156). Other toxicities in animals are mild and infrequent at therapeutic doses. For example, occasional increases in liver function enzymes are noted, but these occur at high doses and are not associated with histopathological changes (46).

Human Safety At Isis, we have studied ∼10 phosphorothioate oligonucleotides in humans. We have studied antisense drugs administered intravitreally, intradermally, subcutaneously, intravenously, and topically. VitraveneTM, an intravitreally administered drug, is commercially available around the world. After systemic administration, in clinical trials, we have studied more than 2000 patients treated with more than 70,000 doses. IV doses have ranged from 0.5 mg/kg every other day to as much as 30 mg/kg/day as a continuous infusion. Most patients have been treated with multiple doses and many have been treated for multiple months. COMPLEMENT ACTIVATION With 2-hour infusions at 2 mg/kg, no increases in complement split products were observed in more than 300 patients with inflammatory diseases treated with Alicaforsen, an inhibitor of ICAM-1. Similarly, ISIS 5132, an inhibitor of c-raf kinase, was dosed from 0.5–6.0 mg/kg with no meaningful increases in complement split products. Affinitak, a PKCα inhibitor, gave equivalent data and both of these drugs were studied in patients with various malignant diseases. Similar results were observed with all the phosporothioate oligodeoxynucleotides administered on this schedule in a variety of patients (for review see 157, 158). Three phosphorothioate oligodeoxynucleotides—Affinitak, ISIS 5132, and ISIS 2503 (an H-ras inhibitor)—have been thoroughly characterized in patients with malignant diseases who were dosed with long-term infusions.

12 Dec 2003

13:36

AR

AR206-ME55-05.tex

AR206-ME55-05.sgm

LaTeX2e(2002/01/18)

PROGRESS IN ANTISENSE TECHNOLOGY

P1: GBC

81

Annu. Rev. Med. 2004.55:61-95. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

When given as 21-day antitumor infusions at doses as high as 10 mg/kg/day, no increases in complement split products were observed (159–161). In contrast, when these drugs were given as 24-h infusions, significant increases in complement split products occurred at doses of 18 mg/kg/day and greater (157). Despite these increases, only a few patients experienced mild fevers and myalgias at these very high doses. CYTOKINES At very high doses, i.e., 24 mg/kg/day as a continuous IV infusion, significant increases in IL-6, IL-IRα, and TNFα were observed and often correlated with flu-like syndromes (157). At present, the precise roles of each of the cytokines and complement activation in the clinical signs and symptoms (myalgia and fever) at high doses are not defined. COAGULATION All phosphorothioate oligodeoxynucleotides studied in normal volunteers and in patients have resulted in transient, self-limited increases in activated partial thromboplastin time (aPTT). The effects are dose and peak plasma concentrations dependent (157). These effects appear to be more prominent after the first dose. In no patients treated have we observed any evidence of bleeding, so the effects on aPTT have not proven to be a problem in the clinic and humans appear to behave similarly to other animals with regard to this side effect. PLATELET EFFECTS When phosphorothioate oligodeoxynucleotides have been administered by continuous IV infusion in patients with malignant diseases, transient thrombocytoplastin has been observed in a few patients (157). This effect was more frequent during the first course of therapy, was not obviously dose-related, and was not associated with bone marrow effects (158, 162). Most of the time, platelet counts returned to normal while dosing was continued and no bleeding was observed. The mechanism for this effect is not clear but probably involves margination.

THE MEDICINAL CHEMISTRY OF OLIGONUCLEOTIDES The core of any rational drug discovery program is medicinal chemistry. Although the synthesis of modified nucleic acids has been studied for some time, the intense focus on the medicinal chemistry of oligonucleotides dates to perhaps five years ago. The scope of medicinal chemistry has expanded enormously, but the biological data to support conclusions about synthetic strategies are only beginning to emerge. The base, sugar, and phosphate moieties of oligonucleotides and oligonucleotide conjugates have been modified. The subjects of medicinal chemical programs include approaches to create stronger and more selective affinity for RNA or duplex structures; to provide the ability to cleave nucleic acid targets; and to enhance nuclease stability, cellular uptake and distribution, and in vivo tissue distribution, metabolism, and clearance.

12 Dec 2003

13:36

Annu. Rev. Med. 2004.55:61-95. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

82

AR

AR206-ME55-05.tex

AR206-ME55-05.sgm

LaTeX2e(2002/01/18)

P1: GBC

CROOKE

Figure 3 Isis oligonucleotide modifications.

Figure 3 shows a dinucleotide and the sites and type of modifications studied. Perhaps the most fruitful area for modifications has been the 20 position. Several hundred 20 modifications have been studied, of which the most interesting is the 20 methoxyethyl modification (Figure 4). 20 methoxyethyl chimeric oligonucleotides have significantly higher affinity than phosphorothioates for RNA. They are at least ten times more potent and have an elimination half-life of 25–30 days in all organs and species studied. Further, this modification reduces the proinflammatory effects of oligonucleotides and has supported oral administration in humans (for review see 1). Several antisense drugs based on this chemistry are in clinical trials today. Phosphate replacement has also been a fruitful area of research. The morpholino replacement has been studied and several drugs in development are based on morpholino chemistry (for review see 163). Additionally, PNAs have been studied extensively (for review see 1).

20 -O-(2-METHOXYETHYL) CHIMERAS: SECOND-GENERATION ANTISENSE DRUGS The 20 -O-(2-methoxyethyl) substitution represents a significant advance in antisense therapeutics and the culmination of more than a decade of progress in antisense technology. Oligonucleotides in which every nucleotide contains a

12 Dec 2003

13:36

AR

AR206-ME55-05.tex

AR206-ME55-05.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:61-95. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

PROGRESS IN ANTISENSE TECHNOLOGY

P1: GBC

83

Figure 4 Example of 20 modification.

20 -O-(2-methoxyethyl) modification are of great value when used for mechanisms mediated by occupancy only, e.g., induction of alternative splicing. They have been used most broadly in chimeric structures designed to serve as substrates for RNase H. Numerous chimeric 20 -O-(2-methoxyethyl) oligonucleotides have not been studied extensively. Although such oligonucleotides typically display affinity for target RNA that is several orders of magnitude greater than for phosphorothioate oligodeoxynucleotides, they are less attractive substrates for RNase H (for review see 164). Consequently, they typically display only 5–15-fold increases in potency both in vitro and in vivo. For example, a direct comparison of two antisense inhibitors of c-raf kinase with the same sequence—one being a phosphorothioate oligodeoxynucleotide, the other a 20 -O-(2-methoxyethyl) chimera—showed that in vitro the 20 -O-(2-methoxyethyl) chimera was approximately fivefold more potent. Similarly, after IV administration in a rat heart allograph model, the modified

12 Dec 2003

Annu. Rev. Med. 2004.55:61-95. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

84

13:36

AR

AR206-ME55-05.tex

AR206-ME55-05.sgm

LaTeX2e(2002/01/18)

P1: GBC

CROOKE

oligonucleotide was at least fivefold more potent (165). Similar data were reported from a direct comparison of antisense inhibitors of survivin in vitro and in vivo in a human tumor xenograft model (166). Pharmacologic studies of antisense oligomers modified with 20 -O-(2-methoxyethyl) chemistry have recently been reviewed (167). Perhaps more importantly, 20 -O-(2-methoxyethyl) chimeras are substantially more stable than phosphorothioate oligonucleotides. In mice, rats, and monkeys, the elimination half-life is nearly 30 days in plasma and several tissues (168– 171). Furthermore, in the liver, elegant correlations between pharmacokinetic and pharmacodynamic effects have been reported showing a duration of pharmacologic action of nearly 20 days in the mouse (172). Studies in animals have recently been extended to humans. ISIS 104838, a 20 -O-(2-methoxyethyl) chimera that acts as a TNFα antisense inhibitor, reduced TNFα secretion at doses at least tenfold lower than would be expected in humans by first-generation antisense drugs (Dorr, unpublished observations) and to have an elimination half-life in plasma of ∼30 days (Geary, unpublished observations). These studies demonstrate the feasibility of monthly dosing. Despite significant progress in achieving acceptable oral bioavailability of antisense inhibitors, much remains to be accomplished, and important clinical studies now under way will help determine the feasibility of oral delivery (for review see 173). Two key barriers to oral bioavailability have been identified and partially overcome: presystemic metabolism and penetration across the gastrointestinal (GI) mucosa. Because the gut has a very high level of nucleases contributed by both the host and bacteria resident in the GI tract, metabolism of phosphorothioate oligodeoxynucleotides in the gut occurs much too rapidly to support adequate oral bioavailability (174). Chimeric modifications of the oligonucleotides have been shown to enhance stability and oral bioavailability (175, 176). With the development of 20 -O-MOE phosphorothioates, oral bioavailability is potentially feasible (173). In rodents, dogs, and monkeys, the permeability of the GI tract to antisense inhibitors has been significantly enhanced by formulations containing penetration enhancers such as bile acid salts and fatty acids. After intrajejunal administration of several 20 -O-MOE modified oligonucleotides in the presence of penetration enhancer, systemic bioavailability in excess of 20% was observed in all three species (173). Initial studies with solid dose forms containing penetration enhancers resulted in significantly less systemic bioavailability (estimated tissue bioavailability: 10%–15%) (168, 170, 173). Additional clinical trials of various solid dose forms containing ISIS 104838, a 20 -O-MOE chimeric antisense inhibitor of TNFα, and various penetration enhancers are in progress, so we should soon have some sense of the near-term potential for oral delivery of antisense inhibitors in humans. Finally, 20 -O-(2-methoxyethyl) chimeras have been shown to be less proinflammatory (46, 177). After repeated subcutaneous dosing in humans, ISIS 104838

12 Dec 2003

13:36

AR

AR206-ME55-05.tex

AR206-ME55-05.sgm

LaTeX2e(2002/01/18)

PROGRESS IN ANTISENSE TECHNOLOGY

P1: GBC

85

produced dramatically lower local inflammation than first-generation antisense drugs. Thus, more convenient and better locally tolerated subcutaneous administration may be feasible.

Annu. Rev. Med. 2004.55:61-95. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

ISIS 104838 ISIS 104838 is a second-generation inhibitor of TNF-α. Clinical trials have demonstrated that it reduces in normal volunteers at doses significantly lower than would be required for a first-generation antisense drug. Further, the drug was extremely well tolerated after repeated subcutaneous dosing. Additionally, the study confirmed that second-generation antisense drugs have a long elimination half-life, i.e., 30 days (178). ISIS 104838 was also reported to be orally bioavailable in humans after administration of solid dose forms. The estimated tissue bioavailability was ∼15%. Current studies are optimizing the oral formulations. These formulations would be expected to provide equivalent results for all second-generation 20 methoxyethyl antisense drugs. A large phase II trial of the subcutaneous form of the drug in patients with rheumatoid arthritis is in progress.

LIMITATIONS Antisense technology is a versatile and effective tool to evaluate gene function in vitro. In our laboratory, we have made antisense inhibitors to perhaps as many as 4000 genes without failure. Antisense agents are also active in vivo and so can be used to validate drug discovery targets. The limitations of antisense agents are defined by their pharmacokinetic and toxicologic properties. For systemic therapy, at lower doses, these drugs distribute primarily to liver, kidney, spleen, and fat cells. They do not cross an intact bloodbrain barrier and do not seem to enter skeletal muscle. Second-generation antisense drugs display similar distribution properties. The administration of antisense drugs by alternative routes, such as aerosol, topical, and enema, enhance their potential utility for organs that are not the primary sites of distribution after systemic delivery. The principal dose-limiting toxicity is proinflammatory effects. In humans, for first-generation antisense drugs, these effects are encountered at doses in excess of 10 mg/kg/day; second-generation antisense drugs are much less prone to cause these side effects. Antisense drugs have a slow onset of action (24–48 h). This limits their utility in acutely life-threatening diseases such as sepsis and cardiovascular events. In the future, new approaches that further enhance potency should reduce the impediments to treating some diseases and ease the challenges of oral delivery.

12 Dec 2003

86

13:36

AR

AR206-ME55-05.tex

AR206-ME55-05.sgm

LaTeX2e(2002/01/18)

P1: GBC

CROOKE

The Annual Review of Medicine is online at http://med.annualreviews.org

Annu. Rev. Med. 2004.55:61-95. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

LITERATURE CITED 1. Crooke ST. 2003. Oligonucleotide therapeutics. In Burger’s Medicinal Chemistry, ed. DJ Abraham, pp. 115–66. New York: Wiley. 6th ed. 2. Crouch RJ, Dirksen M-L. 1985. Ribonucleases H. In Nucleases, ed. SM Linn, RJ Roberts, pp. 211–41. Cold Spring Harbor, NY: Cold Spring Harbor Lab. Press 3. Crum C, Johnson JD, Nelson A, Roth D. 1988. Complementary oligodeoxynucleotide mediated inhibition of tobacco mosaic virus RNA translation in vitro. Nucleic Acids Res. 16:4569–81 4. Cerritelli SM, Frolova EG, Feng C, et al. 2003. Failure to produce mitochondrial DNA results in embryonic lethality in RNase H1 null mice. Mol. Cell 11:807– 15 5. Donis-Keller H. 1979. Site specific enzymatic cleavage of RNA. Nucleic Acids Res. 7:179–92 6. Kawasaki AM, Casper MD, Freier SM, et al. 1993. Uniformly modified 20 -deoxy-20 -fluoro-phosphorothioate oligonucleotides as nuclease-resistant antisense compounds with high affinity and specificity for RNA targets. J. Med. Chem. 36:831–41 7. Sproat BS, Lamond AI, Beijer B, et al. 1989. Highly efficient chemical synthesis of 20 -O-methyloligoribonucleotides and tetrabiotinylated derivatives; novel probes that are resistant to degradation by RNA or DNA specific nucleases. Nucleic Acids Res. 17:3373–86 8. Morvan F, Rayner B, Imbach JL. 1991. Alpha-oligonucleotides: a unique class of modified chimeric nucleic acids. Anticancer Drug Des. 6:521–29 9. Gagnor C, Rayner B, Leonetti JP, et al. 1989. Alpha-DNA.IX: parallel annealing of alpha-anomeric oligodeoxyribonucleotides to natural mRNA is required for interference in RNase H me-

10.

11.

12.

13.

14.

15.

16.

17.

18.

diated hydrolysis and reverse transcription. Nucleic Acids Res. 17:5107–14 Miller PS. 1989. Non-ionic antisense oligonucleotides. In Oligodeoxynucleotides: Antisense Inhibitors of Gene Expression, ed. JS Cohen, p. 79. Boca Raton, FL: CRC Press Maher LJ, III, Wold B, Dervan PB. 1989. Inhibition of DNA binding proteins by oligonucleotide-directed triple helix formation. Science 245:725–30 Stein CA, Cheng Y-C. 1993. Antisense oligonucleotides as therapeutic agents— Is the bullet really magical? Science 261:1004–12 Mirabelli CK, Bennett CF, Anderson KP, Crooke ST. 1991. In vitro and in vivo pharmacologic activities of antisense oligonucleotides. Anti-Cancer Drug Design 6:647–61 Cazenave C, Stein CA, Loreau N, et al. 1989. Comparative inhibition of rabbit globin mRNA translation by modified antisense oligodeoxynucleotides. Nucleic Acids Res. 17:4255–73 Quartin RS, Brakel CL, Wetmur JG. 1989. Number and distribution of methylphosphonate linkages in oligodeoxynucleotides affect exo- and endonuclease sensitivity and ability to form RNase H substrates. Nucleic Acids Res. 17:7253–62 Furdon PJ, Dominski Z, Kole R. 1989. RNase H cleavage of RNA hybridized to oligonucleotides containing methylphosphonate, phosphorothioate and phosphodiester bonds. Nucleic Acids Res. 17:9193–204 Eder PS, Walder JA. 1991. Ribonuclease H from K562 human erythroleukemia cells. J. Biol. Chem. 266:6472–79 Monia BP, Lesnik EA, Gonzalez C, et al. 1993. Evaluation of 20 -modified oligonucleotides containing 20 -deoxy gaps as

12 Dec 2003

13:36

AR

AR206-ME55-05.tex

AR206-ME55-05.sgm

LaTeX2e(2002/01/18)

PROGRESS IN ANTISENSE TECHNOLOGY

19.

Annu. Rev. Med. 2004.55:61-95. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

20.

21.

22.

23.

24.

25.

26.

27.

28.

antisense inhibitors of gene expression. J. Biol. Chem. 268:14514–22 Giles RV, Tidd DM. 1992. Increased specificity for antisense oligodeoxynucleotide targeting of RNA cleavage by RNase H using chimeric methylphosphonodiester/phosphodiester structures. Nucleic Acids Res. 20:763–70 Crooke ST, Lemonidis KM, Neilson L, et al. 1995. Kinetic characteristics of Escherichia coli RNase H1: cleavage of various antisense oligonucleotide-RNA duplexes. Biochem. J. 312:599–608 Lima WF, Crooke ST. 1997. Binding affinity and specificity of Escherichia coli RNase H1: impact on the kinetics of catalysis of antisense oligonucleotideRNA hybrids. Biochemistry 36:390–98 Wu H, Lima WF, Crooke ST. 1998. Molecular cloning and expression of cDNA for human RNase H. Antisense Nucleic Acid Drug Dev. 8:53–61 Frank P, Braunshofer-Reiter C, Wintersberger U, et al. 1998. Cloning of the cDNA encoding the large subunit of human RNase HI, a homologue of the prokaryotic RNase HII. PNAS 95:12872–77 Kanaya S, Katsuda-Nakai C, Ikebara M. 1991. Importance of the positive charge cluster in Escherichia coli ribonuclease HI for the effective binding of the substrate. J. Biol. Chem. 266:11621–27 Katayanagi K, Okumura M, Morikawa K. 1993. Crystal structure of Escherichia coli RNase HI in complex with Mg2+ ˚ resolution: proof for a sinat 2.8 A gle Mg2+-binding site. Proteins Struct. Funct. Genet. 17:337–46 Nicholson A. 1997. Ribonucleases: Structures and Functions. New York: Academic Lima WF, Mohan V, Crooke ST. 1997. The influence of antisense oligonucleotide-induced RNA structure on Escherichia coli RNase H1 activity. J. Biol. Chem. 272:18191–99 Wu H, Lima WF, Crooke ST. 1999. Prop-

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

P1: GBC

87

erties of cloned and expressed human RNase H1. J. Biol. Chem. 274:28270–78 Lima WF, Crooke ST. 1997. Cleavage of single strand RNA adjacent to RNADNA duplex regions by Escherichia coli RNase H1. J. Biol. Chem. 272:27513–16 Fraser AG, Kamath RS, Zipperlen P, et al. 2000. Functional genomic analysis of C. elegans chromosome I by systematic RNA interference. Nature 408:325– 30 Gonczy P, Echeverri C, Oegema K, et al. 2000. Functional genomic analysis of cell division in C. elegans using RNAi of genes on chromosome III. Nature 408:331–36 Grishok A, Tabara H, Mello CC. 2000. Genetic requirements for inheritance of RNAi in C. elegans. Science 287:2494– 97 Zamore PD, Tuschl T, Sharp PA, Bartel DP. 2000. RNAi: double-stranded RNA directs the ATP-dependent cleavage of mRNA at 21 to 23 nucleotide intervals. Cell 101:25–33 Zamore PD. 2002. Ancient pathways programmed by small RNAs. Science 296:1265–69 Tuschl T, Zamore PD, Lehmann R, et al. 1999. Targeted mRNA degradation by double-stranded RNA in vitro. Genes Dev. 13:3191–97 Caplen NJ, Fleenor J, Morgan AFA. 2000. dsRNA-mediated gene silencing in cultured Drosophila cells: a tissue culture model for the analysis of RNA interference. Gene 252:95–105 Der SD, Yang Y-L, Weissmann C, Williams BRG. 1997. A doublestranded RNA-activated protein kinase-dependent pathway mediating stress-induced apoptosis. Proc. Natl Acad. Sci. USA 94:3279–83 Elbashir SM, Harborth J, Lendeckel W, et al. 2001. Duplexes of 21-nucleotide RNAs mediate RNA interference in cultured mammalian cells. Nature 411:494– 98

12 Dec 2003

13:36

Annu. Rev. Med. 2004.55:61-95. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

88

AR

AR206-ME55-05.tex

AR206-ME55-05.sgm

LaTeX2e(2002/01/18)

P1: GBC

CROOKE

39. Caplen NJ, Parrish S, Imani F, et al. 2001. Specific inhibition of gene expression by small double-stranded RNAs in invertebrate and vertebrate systems. Proc. Natl. Acad. Sci. USA 98:9742–47 40. Novina CD, Murray MF, Dykxhoorn DM, et al. 2002. siRNA-directed inhibition of HIV-1 infection. Nat. Med. 8:681–86 41. Vickers TA, Koo S, Bennett CF, et al. 2003. Efficient reduction of target RNAs by small interfering RNA and RNase H-dependent antisense agents. J. Biol. Chem. 278:7108–18 42. Cohen JS. 1993. Phosphorothioate oligodeoxynucleotides. See Ref. 42a, pp. 205–22 42a. Crooke ST, Lebleu B, eds. 1993. Antisense Research and Applications. Boca Raton, FL: CRC Press 43. Crooke ST, Graham MJ, Zuckerman JE, et al. 1996. Pharmacokinetic properties of several novel oligonucleotide analogs in mice. J. Pharmacol. Exp. Ther. 277:923–37 44. Joos RW, Hall WH. 1969. Determination of binding constants of serum albumin for penicillin. J. Pharmacol. Exp. Ther. 166:113 45. Srinivasan SK, Tewary HK, Iversen PL. 1995. Characterization of binding sites, extent of binding, and drug interactions of oligonucleotides with albumin. Antisense Res. Dev. 5:131–39 46. Levin AA, Henry S, Monteith D, Templin M. 2001. Toxicity of antisense oligonucleotides. See Ref. 46a, pp. 201– 67 46a. Crooke ST, ed. 2001. Antisense Drug Technology: Principles, Strategies and Applications. New York: Marcel Dekker 47. Loke SL, Stein CA, Zhang XH, et al. 1989. Characterization of oligonucleotide transport into living cells. Proc. Natl. Acad. Sci. USA 86:3474– 78 48. Hawley P, Gibson I. 1996. Interaction of oligodeoxynucleotides with mammalian

49.

50.

51.

52.

53.

54.

55.

56.

57.

cells. Antisense and Nucleic Drug Dev. 6:185–95 Crooke RM, Graham MJ, Cooke ME, Crooke ST. 1995. in vitro pharmacokinetics of phosphorothioate antisense oligonucleotides. J. Pharmacol. Exp. Ther. 275:462–73 Gao W-Y, Han F-S, Storm C, et al. 1992. Phosphorothioate oligonucleotides are inhibitors of human DNA polymerases and RNase H: implications for antisense technology. Mol. Pharmacol. 41:223–29 Crooke ST. 1995. Oligonucleotide therapeutics. In Burger’s Medicinal Chemistry and Drug Discovery, Vol. 1, ed. ME Wolff, pp. 863–900. New York: Wiley Agrawal S, Temsamani J, Tang JY. 1991. Pharmacokinetics, biodistribution, and stability of oligodeoxynucleotide phosphorothioates in mice. Proc. Natl. Acad. Sci. USA 88:7595–99 Majumdar C, Stein CA, Cohen JS, et al. 1989. Stepwise mechanism of HIV reverse transcriptase: primer function of phosphorothioate oligodeoxynucleotide. Biochemistry 28:1340–46 Cheng Y, Gao W, Han F. 1991. Phosphorothioate oligonucleotides as potential antiviral compounds against human immunodeficiency virus and herpes viruses. Nucleosides Nucleotides 10:155–66 Stein CA, Neckers M, Nair BC, et al. 1991. Phosphorothioate oligodeoxycytidine interferes with binding of HIV-1 gp120 to CD4. J. Acquired Immune Deficiency Syndromes 4:686–93 Wyatt JR, Vickers TA, Roberson JL, et al. 1994. Combinatorially selected guanosine-quartet structure is a potent inhibitor of human immunodeficiency virus envelope–mediated cell fusion. Proc. Natl. Acad. Sci. USA 91:1356–60 Hartmann G, Krug A, Waller-Fontaine K, Endres S. 1996. Oligodeoxynucleotides enhance lipopolysaccharidestimulated synthesis of tumor necrosis factor: dependence on phosphorothioate

12 Dec 2003

13:36

AR

AR206-ME55-05.tex

AR206-ME55-05.sgm

LaTeX2e(2002/01/18)

PROGRESS IN ANTISENSE TECHNOLOGY

58.

59.

Annu. Rev. Med. 2004.55:61-95. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

60.

61.

62.

63.

64.

65.

66.

67.

68.

modification and reversal by Heparin. Mol. Med. 2:429–38 Wickstrom E. 1986. Oligodeoxynucleotide stability in subcellular extracts and culture media. J. Biochem. Biophys. Methods 13:97–102 Crooke ST. 1995. Therapeutic Applications of Oligonucleotides. Austin, TX: R.G. Landes. 138 pp. Hoke GD, Draper K, Freier SM, et al. 1991. Effects of phosphorothioate capping on antisense oligonucleotide stability, hybridization and antiviral efficacy versus herpes simplex virus infection. Nucleic Acids Res. 19:5743–48 Campbell JM, Bacon TA, Wickstrom E. 1990. Oligodeoxynucleoside phosphorothioate stability in subcellular extracts, culture media, sera and cerebrospinal fluid. J. Biochem. Biophys. Methods 20:259–67 Cossum PA, Sasmor H, Dellinger D, et al. 1993. Disposition of the 14C-labeled phosphorothioate oligonucleotide ISIS 2105 after intravenous administration to rats. J. Pharmacol. Exp. Ther. 267:1181– 90 Crooke ST. 1993. Therapeutic potential of oligonucleotides. Curr. Opin. Invest. Drugs 2:1045–48 Sands H, Gorey-Feret LJ, Ho SP, et al. 1995. Biodistribution and metabolism of internally 3H-labeled oligonucleotides. II. 30 ,5-blocked oligonucleotides. Mol. Pharmacol. 47:636–46 Miyao T, Takakura Y, Akiyama T, et al. 1995. Stability and pharmacokinetic characteristics of oligonucleotides modified at terminal linkages in mice. Antisense Res. Dev. 5:115–21 Crooke RM. 1991. in vitro toxicology and pharmacokinetics of antisense oligonucleotides. Anti-Cancer Drug Design 6:609–46 Crooke RM. 1993. in vitro and in vivo toxicology of first generation analogs. See Ref. 42a, pp. 471–92 Neckers LM. 1993. Cellular internaliza-

69.

70.

71.

72.

73.

74.

75.

76.

77.

P1: GBC

89

tion of oligodeoxynucleotides. See Ref. 42a, pp. 451–60 Chrisey LA, Walz SE, Pazirandeh M, Campbell JR. 1993. Internalization of oligodeoxyribonucleotides by Vibrio parahaemolyticus. Antisense Res. Dev. 3:367–81 Tao LF, Marx KA, Wongwit W, et al. 1995. Uptake, intracellular distribution, and stability of oligodeoxynucleotide phosphorothioate by Schistosoma mansoni. Antisense Res. Dev. 5:123–29 Templin MV, Levin AA, Graham MJ, et al. 2000. Pharmacokinetic and toxicity profile of a phosphorothioate oligonucleotide following inhalation delivery to lung in mice. Antisense Nucleic Acid Drug Dev. 10:359–68 Leeds JM, Henry SP, Geary R, et al. 2000. Comparison of the pharmacokinetics of subcutaneous and intravenous administration of a phosphorothioate oligodeoxynucleotide in cynomolgus monkeys. Antisense Nucleic Acid Drug Dev. 10:435–41 Bennett CF, Chiang MY, Chan H, Grimm S. 1993. Use of cationic lipids to enhance the biological activity of antisense oligonucleotides. J. Liposome Res. 3:85– 102 Bennett CF, Chiang MY, Chan H, et al. 1992. Cationic lipids enhance cellular uptake and activity of phosphorothioate antisense oligonucleotides. Mol. Pharmacol. 41:1023–33 Quattrone A, Papucci L, Schiavone N, et al. 1994. Intracellular enhancement of intact antisense oligonucleotide steadystate levels by cationic lipids. AntiCancer Drug Design 9:549–53 Giles RV, Spiller DG, Tidd DM. 1995. Detection of ribonuclease H–generated mRNA fragments in human leukemia cells following reversible membrane permeabilization in the presence of antisense oligodeoxynucleotides. Antisense Res. Dev. 5:23–31 Wang S, Lee RJ, Cauchon G, et al. 1995.

12 Dec 2003

13:36

90

78.

Annu. Rev. Med. 2004.55:61-95. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

79.

80.

81.

82.

83.

84.

85.

86.

AR

AR206-ME55-05.tex

AR206-ME55-05.sgm

LaTeX2e(2002/01/18)

P1: GBC

CROOKE Delivery of antisense oligodeoxyribonucleotides against the human epidermal growth factor receptor into cultured KB cells with liposomes conjugated to folate via polyethylene glycol. Proc. Natl. Acad. Sci. USA 92:3318–22 Iversen P. 1992. In vivo studies with phosphorothioate oligonucleotides: pharmacokinetics prologue. Anticancer Drug Des. 6:531–38 Yu RZ, Geary RS, Leeds JM, et al. 2000. Comparison of pharmacokinetics and tissue disposition of an antisense phosphorothioate oligonucleotide targeting human Ha-ras mRNA in mouse and monkey. J. Pharm. Sci. 90:182–93 Cossum PA, Troung L, Owens SR, et al. 1994. Pharmacokinetics of a 14C-labeled phosphorothioate oligonucleotide, ISIS 2105, after intradermal administration to rats. J. Pharmacol. Exp. Ther. 269:89–94 Crooke S, Grillone I, Tendolkar A, et al. 1994. A pharmacokinetic evaluation of 14 C-labeled afovirsen sodium in patients with genital warts. Clin. Pharmacol. Ther. 56:641–46 Crooke ST, Grillone LR, Tendolkar A, et al. 1994. A pharmacokinetic evaluation of 14C-labeled afovirsen sodium in patients with genital warts. Clin. Pharmacol. Ther. 56:641–46 Butler M, Stecker K, Bennett CF. 1997. Cellular distribution of phosphorothioate oligodeoxynucleotides in normal rodent tissues. Lab. Invest. 77:379–88 Rappaport J, Hanss B, Kopp JB, et al. 1995. Transport of phosphorothioate oligonucleotides in kidney: implications for molecular therapy. Kidney Int. 47:1462–69 Sands H, Gorey-Feret LJ, Cocuzza AJ, et al. 1994. Biodistribution and metabolism of internally 3H-labeled oligonucleotides. I. Comparison of a phosphodiester and phosphorothioate. Mol. Pharmacol. 45:932–43 Takakura Y, Mahato RI, Yoshida M, et al. 1996. Uptake characteristics of oligonu-

87.

88.

89.

90.

91.

92.

93.

94.

95.

cleotides in the isolated rat liver perfusion system. Antisense Nucleic Acid Drug Dev. 6:177–83 Graham MJ, Crooke ST, Monteith DK, et al. 1998. In vivo distribution and metabolism of a phosphorothioate oligonucleotide within rat liver after intravenous administration. J. Pharmacol. Exp. Ther. 286:447–58 Nyce JW. 1997. Respirable antisense oligonucleotides as novel therapeutic agents for asthma and other pulmonary diseases. Exp. Opin. Invest. Drugs 6:1149–56 Nyce JW. 2001. Respirable antisense oligonucleotides (RASONs). See Ref. 46a, pp. 679–85 Saxena SK, Ackerman EJ. 1990. Microinjected oligonucleotides complementary to the α-sarcin loop of 28 S RNA abolish protein synthesis in Xenopus oocytes. J. Biol. Chem. 265:3263–69 Miner PB, Bane BL, Sadiq SA, et al. 2003. Antisense inhibitor of ICAM-1 given as enema improves symptomatic pouchitis. Presented at Digestive Disease Week, Orlando, FL Chiang M-Y, Chan H, Zounes MA, et al. 1991. Antisense oligonucleotides inhibit intercellular adhesion molecule 1 expression by two distinct mechanisms. J. Biol. Chem. 266:18162–71 Agrawal S, Goodchild J, Civeira MP, et al. 1988. Oligodeoxynucleoside phosphoramidites and phosphorothioates as inhibitors of human immunodeficiency virus. Proc. Natl. Acad. Sci. USA 85:7079–83 Kulka M, Smith CC, Aurelian L, et al. 1989. Site specificity of the inhibitory effects of oligo(nucleoside methylphosphonate)s complementary to the acceptor splice junction of herpes simplex virus type 1 immediate early mRNA 4. Proc. Natl. Acad. Sci. USA 86:6868– 72 Mirabelli CK, Crooke ST. 1993. Antisense oligonucleotides in the context of

12 Dec 2003

13:36

AR

AR206-ME55-05.tex

AR206-ME55-05.sgm

LaTeX2e(2002/01/18)

PROGRESS IN ANTISENSE TECHNOLOGY

96.

Annu. Rev. Med. 2004.55:61-95. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

97.

98.

99.

100.

101.

102.

103.

104.

modern molecular drug discovery and development. See Ref. 42a, pp. 7–35 Crooke ST. 1992. Therapeutic applications of oligonucleotides. Bio/Technology 10:882–86 Bennett CF, Crooke ST. 1996. Oligonucleotide-based inhibitors of cytokine expression and function. In Therapeutic Modulation of Cytokines, ed. BB Henderson, W Mark, pp. 171–93. Boca Raton, FL: CRC Press Azad RF, Driver VB, Tanaka K, et al. 1993. Antiviral activity of a phosphorothioate oligonucleotide complementary to RNA of the human cytomegalovirus major immediate-early region. Antimicrob. Agents Chemother. 37:1945–54 Wagner RW, Matteucci MD, Lewis JG, et al. 1993. Antisense gene inhibition by oligonucleotides containing C-5 propyne pyrimidines. Science 260:1510–13 Barton CM, Lemoine NR. 1995. Antisense oligonucleotides directed against p53 have antiproliferative effects unrelated to effects on p53 expression. Br. J. Cancer 71:429–37 Burgess TL, Fisher EF, Ross SL, et al. 1995. The antiproliferative activity of c-myb and c-myc antisense oligonucleotides in smooth muscle cells is caused by a nonantisense mechanism. Proc. Natl. Acad. Sci. USA 92:4051–55 Hertl M, Neckers LM, Katz SI. 1995. Inhibition of interferon-gamma–induced intercellular adhesion molecule-1 expression on human keratinocytes by phosphorothioate antisense oligodeoxynucleotides is the consequence of antisense-specific and antisense–nonspecific effects. J. Invest. Dermatol. 104: 813–18 Crooke ST, Mirabelli CK. 1993. Antisense oligonucleotides in the context of modern molecular drug discovery and development. See Ref. 42a, pp. 7–35 Nagel KM, Holstad SG, Isenberg KE. 1993. Oligonucleotide pharmacother-

105.

105a.

106.

107.

108.

109.

110.

111.

112.

113.

P1: GBC

91

apy: an antigene strategy. Pharmacotherapy 13:177–88 Crooke ST. 1998. Basic principles of antisense therapeutics. See Ref. 105a, pp. 1–50 Crooke ST, ed. 1998. Antisense Research and Application. Berlin/Heidelberg: Springer-Verlag Simons M, Edelman ER, DeKeyser J-L, et al. 1992. Antisense c-myb oligonucleotides inhibit arterial smooth muscle cell accumulation in vivo. Nature 359:67–70 Abe J, Zhou W, Taguchi J, et al. 1994. Suppression of neointimal smooth muscle cell accumulation in vivo by antisense CDC2 and CDK2 oligonucleotides in rat carotid artery. Biochem. Biophys. Res. Commun. 198:16–24 Whitesell L, Rosolen A, Neckers LM. 1991. In vivo modulation of N-myc expression by continous perfusion with an antisense oligonucleotide. Antisense Res. Dev. 1:343–50 Wahlestedt C, Pich EM, Koob GF, et al. 1993. Modulation of anxiety and neuropeptide Y-Y1 receptors by antisense oligodeoxynucleotides. Science 259:528–31 Wahlestedt C. 1994. Antisense oligonucleotide strategies in neuropharmacology. Trends Pharmacol. Sci. 15:42–46 Zhang S-P, Zhou L-W, Weiss B. 1994. Oligodeoxynucleotide antisense to the D1 dopamine receptor mRNA inhibits D1 dopamine receptor–mediated behaviors in normal mice and in mice lesioned with 6-hydroxydopamine. J. Pharmacol. Exp. Ther. 271:1462–70 Dean NM, McKay R, Condon TP, Bennett CF. 1994. Inhibition of protein kinase C-alpha. expression in human A549 cells by antisense oligonucleotides inhibits induction of intercellular adhesion molecule 1 (ICAM-1) mRNA by phorbol esters. J. Biol. Chem. 269:16416– 24 Burch RM, Weitzberg M, Blok N,

12 Dec 2003

13:36

92

114.

Annu. Rev. Med. 2004.55:61-95. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

115.

116.

117.

118.

119.

120.

121.

122.

AR

AR206-ME55-05.tex

AR206-ME55-05.sgm

LaTeX2e(2002/01/18)

P1: GBC

CROOKE et al. 1991. N-(Fluorenyl-9-methoxycarbonyl) amino acids, a class of antiinflammatory agents with a different mechanism of action. Proc. Natl. Acad. Sci. USA 88:355–59 Higgins KA, Perez JR, Coleman TA, et al. 1993. Antisense inhibition of the p65 subunit of NF-κB blocks tumorigenicity and causes tumor regression. Proc. Natl. Acad. Sci. USA 90:9901–5 Skorski T, Nieborowska-Skorska M, Nicolaides NC, et al. 1994. Suppression of Philadelphia leukemia cell growth in mice by BCR-ABL antisense oligodeoxynucleotide. Proc. Natl. Acad. Sci. USA 91:4504–8 Hijiya N, Zhang J, Ratajczak MZ, et al. 1994. Biologic and therapeutic significance of MYB expression in human melanoma. Proc. Natl. Acad. Sci. USA 91:4499–503 Dean NM, McKay R. 1994. Inhibition of protein kinase C-alpha expression in mice after systemic administration of phosphorothioate antisense oligodeoxynucleotides. Proc. Natl. Acad. Sci. USA 91:11762–66 Dean N, McKay R, Miraglia L, et al. 1996. Inhibition of growth of human tumor cell lines in nude mice by an antisense oligonucleotide inhibitor of protein kinase C-alpha expression. Cancer Res. 56:3499–507 Monia BP, Johnston JF, Geiger T, et al. 1995. Antitumor activity of a phosphorothioate oligodeoxynucleotide targeted against C-raf kinase. Nat. Med. 2:668– 75 Monia BP, Johnston JF, Sasmor H, Cummins LL. 1996. Nuclease resistance and antisense activity of modified oligonucleotides targeted to Ha-ras. J. Biol. Chem. 271:14533–40 Nesterova M, Cho-Chung YS. 1995. A single-injection protein kinase A– directed antisense treatment to inhibit tumor growth. Nat. Med. 1:528–33 Grillone L. 2001. The development of

123.

124.

125.

126.

127.

128.

129.

130.

131.

antisense oligonucleotides as antivirals. See Ref. 46a, pp. 725–38 Bennett F. 2001. General pharmacology of phosphorothioate oligodeoxynucleotides. See Ref. 46a, pp. 291–318 Shanahan W. 2001. ISIS 2302, an antisense inhibitor of intercellular adhesion molecule 1 (ICAM–1). See Ref. 46a, pp. 773–94 Maksymowych W, Blackburn W, Hutchings E, et al. 1999. A pilot study of ISIS 2302, an antisense oligodeoxynucleotide, in patients with active rheumatoid arthritis. Arthritis Rheumatol. 42:S170 Fredriksson T, Pettersson U. 1978. Severe psoriasis, oral therapy with a new retinoid. Dermatologica 157:238–44 Yacyshyn BR, Bowen-Yacyshyn MB, Jewell L, et al. 1998. A placebocontrolled trial of ICAM-1 antisense oligonucleotide in the treatment of Crohn’s disease. Gastroenterology 114:1133–42 Yu RZ, Su JG, Grundy JS, et al. 2003. Prediction of clinical responses in a simulated phase III trial of Crohn’s patients administered the antisense phosphorothioate oligonucleotide ISIS 2302: comparison of proposed dosing regimens. Antisense Nucleic Acid Drug Dev. 13:57–66 Yacyshyn B, Woloschuk B, Yacyshyn M, et al. 1997. Efficacy and safety of ISIS 2302 (ICAM-1 antisense oligonucleotide) treatment of steroid-dependent Crohn’s disease. Presented at Annu. Meet. Am. Gastroenterol. Assoc. and Am. Assoc. Study Liver Dis., Washington, DC Holmlund J. 2001. Clinical activities in patients with solid tumors or lymphoma. See Ref. 46a, pp. 739–59 Yuen A, Advani R, Fisher G, et al. 2001. A phase I/II trial of ISIS 3521, an antisense inhibitor of protein kinase C alpha, combined with carboplatin and paclitaxel in patients with non–small cell

12 Dec 2003

13:36

AR

AR206-ME55-05.tex

AR206-ME55-05.sgm

LaTeX2e(2002/01/18)

PROGRESS IN ANTISENSE TECHNOLOGY

132.

Annu. Rev. Med. 2004.55:61-95. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

133.

134.

135.

136.

137.

138.

139.

lung cancer. Presented at Annu. Meet. Am. Soc. Clin. Oncol., San Francisco, CA Lynch T, Raju R, Lind M, et al. 2002. Randomized phase III trial of chemotherapy and antisense oligonucleotide LY900003 (ISIS 3521) in patients with advanced NSCLC: initial report. Presented at Annu. Meet. Am. Soc. Clin. Oncol., Orlando, FL Waters J, Webb A, Cunningham D, et al. 2000. Phase I clinical and pharmacokinetic study of Bcl-2 antisense oligonucleotide therapy in patients with nonHodgkin’s lymphoma. J. Clin. Oncol. 18:1812–23 Jansen B, Wacheck V, Heere-Ress E. 2000. Systemic treatment with Bcl-2 antisense down-regulates tumor content of Bcl-2 protein and induces major antitumor responses in patients with melanoma when combined with dacarbazine. Presented at Am. Assoc. Cancer Res. 91st Annu. Meet., San Francisco, CA O’Dwyer PJ, Stevenson JP, Gallagher M, et al. 1999. c-raf-1 depletion and tumor responses in patients treated with the c-raf-1 antisense oligodeoxynucleotide ISIS 5132 (CGP 69846A). Clin. Cancer Res. 5:3977–82 Levin AA, Monteith DK, Leeds JM, et al. 1998. Toxicity of oligodeoxynucleotide therapeutic agents. See Ref. 105a, pp. 169–214 Galbraith WM, Hobson WC, Giclas PC, et al. 1994. Complement activation and hemodynamic changes following intravenous administration of phosphorothioate oligonucleotides in the monkey. Antisense Res. Dev. 4:201–6 Henry SP, Giclas PC, Leeds J, et al. 1997. Activation of the alternative pathway of complement by a phosphorothioate oligonucleotide: potential mechansim of action. J. Pharmacol. Exp. Ther. 281:810–16 Cornish KG, Iversen P, Smith L, et al.

140.

141.

142.

143.

144.

145.

146.

147.

148.

P1: GBC

93

1993. Cardiovascular effects of a phosphorothioate oligonucleotide to p53 in the conscious rhesus monkey. Pharmacol.Comm. 3:239–47 Henry SP, Monteith D, Levin AA. 1997. Antisense oligonucleotide inhibitors for the treatment of cancer. 2. Toxicological properties of phosphorothioate oligodeoxynucleotides. AntiCancer Drug Des. 12:395–408 Henry SP, Novotny W, Leeds J, et al. 1997. Inhibition of coagulation by a phosphorothioate oligonucleotide. Antisense Nucleic Acid Drug Dev. 7:503–10 Wallace TL, Bazemore SA, Kornbrust DJ, Cossum PA. 1996. Single-dose hemodynamic toxicity and pharmacokinetics of a partial phosphorothioate anti-HIV oligonucleotide (AR177) after intravenous infusion to cynomolgus monkeys. J. Pharmacol. Exp. Ther. 278:1306–12 Griffin LC, Tidmarsh GF, Bock LC, et al. 1993. In vivo anticoagulant properties of a novel nucleotide-based thrombin inhibitor and demonstration of regional anticoagulation in extracorporeal circuits. Blood 81:3271–76 Nicklin PL, Ambler J, Mitchelson A, et al. 1997. Preclinical profiling of modified oligonucleotides: anticoagulation and pharmacokinetic properties. Nucleosides Nucleotides 16:1145–53 Sheehan JP, Lan H-C. 1998. Phosphorothioate oligonucleotides inhibit the intrinsic tenase complex. Blood 92:1617– 25 Henry S, Larkin R, Novotny W, Kornbrust D. 1994. Effects of ISIS 2302, a phosphorothioate oligonucleotide, on in vitro and in vivo coagulation parameters. Pharm. Res. 11:S-353/S1-S490 Monteith DK, Geary RS, Leeds JM, et al. 1998. Preclinical evaluation of the effects of a novel antisense conpound targeting C-raf kinase in mice and monkeys. Toxicol. Sci. 46:365–75 Henry SP, Bolte H, Auletta C, Kornbrust

12 Dec 2003

13:36

94

149.

Annu. Rev. Med. 2004.55:61-95. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

150.

151.

152.

153.

154.

155.

156.

157.

158.

AR

AR206-ME55-05.tex

AR206-ME55-05.sgm

LaTeX2e(2002/01/18)

P1: GBC

CROOKE DJ. 1997. Evaluation of the toxicity of ISIS 2302, a phosphorothioate oligonucleotide, in a 4-week study in cynomolgus monkeys. Toxicology 120:145–55 Sarmiento UM, Perez JR, Becker JM, Ramaswamy N. 1994. In vivo toxicological effects of rel A antisense phosphorothioates in CD-1 mice. Antisense Res. Dev. 4:99–107 Pisetsky DS, Reich CF. 1994. Stimulation of murine lymphocyte proliferation by a phosphorothioate oligonucleotide with antisense activity for herpes simplex virus. Life Sci. 54:101–7 Liang H, Nishioka Y, Reich CF, et al. 1996. Activation of human B cells by phosphorothioate oligodeoxynucleotides. J. Clin. Invest. 98:1119–29 Lemonidis-Farrar K, Templin M, et al. 2000. Cytokine levels and cell population changes as predictive measures of immunostimulatory effects of phosphorothioate oligonucleotides (PS ODN) in mice. Toxicologist 42:120 Krieg AM. 1998. Immune stimulation by oligonucleotides. See Ref. 105a, pp. 243–62 Davis HL, Weeranta R, Waldschmidt TJ, et al. 1998. CpG DNA is a potent enhancer of specific immunity in mice immunized with recombinant hepatitis B surface antigen. J. Immunol. 160:870– 76 Harmann G, Krieg A. 2000. Mechanism and function of a newly identified CpG DNA motif in human primary B cells. J. Immunol. 164:944–53 Krieg AM, Yi A-K, Matson S, et al. 1995. CpG motifs in bacterial DNA trigger direct B-cell activation. Nature 374:546– 49 Dorr FA, Glover JM, Kwoh J. 2001. Clinical safety of phosphorothioate oligodeoxynucleotides. See Ref. 46a, pp. 269–90 Nemunaitis J, Holmlund JT, Kraynak M, et al. 1999. Phase I evaluation of ISIS 3521, an antisense oligodeoxynucleotide

159.

160.

161.

162.

163.

164. 165.

166.

to protein kinase C-alpha, in patients with advanced cancer. J. Clin. Oncol. 17:3586–95 Dorr A, Nemunaitis J, Bruce J, et al. 1999. Phase I and pharmacokinetic trial of ISIS 2503, a 20-mer antisense oligonucleotide against h-ras, by 14-day continuous infusion (CIV) in patients with advanced cancer. Presented at Annu. Meet. Am. Soc. Clin. Oncol., Atlanta, GA Holmlund J, Nemunaitis J, Schiller J, et al. 1998. Phase I trial of c-raf antisense oligonucleotide—ISIS 5132 (CGP 69846A) by 21-day continuous intravenous infusion (CIV) in patients with advanced cancer. Presented at Annu. Meet. Am. Soc. Clin. Oncol., Los Angeles, CA Sikic GI, Yuen AR, Halsey J, et al. 1998. A Phase I trial of an antisense oligonucleotide targeted to protein kinase Calpha (ISIS 3521) delivered by twentyone day continuous intravenous infusion. Presented at Annu. Meet. Am. Soc. Clin. Oncol., Denver, CO Yuen AR, Halsey J, Fisher GA, et al. 1999. Phase I study of an antisense oligonucleotide to protein kinase Calpha (ISIS 3521/CGP 64128A) in patients with cancer. Clin. Cancer Res. 5:3357–63 Summerton J, Weller D. 1997. Morpholino antisense oligomers: design, preparation, and properties. Antisense Nucleic Acid Drug Dev. 7:187–95 Crooke ST. 2001. Basic principles. See Ref. 46a, pp. 1–28 Stepkowski SM, Qu X, Wang M-E, et al. 2000. Inhibition of c-raf expression by antisense oligonucleotides extends heart allograft survival in rats. Transplantation 70:656–61 Zellweger T, Miyake H, Cooper S, et al. 2001. Antitumor activity of antisense clusterin oligonucleotides is improved in vitro and in vivo by incorporation of 20 O-(2-methoxy)ethyl chemistry. J. Pharmacol. Exp. Ther. 298:1–7

12 Dec 2003

13:36

AR

AR206-ME55-05.tex

AR206-ME55-05.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:61-95. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

PROGRESS IN ANTISENSE TECHNOLOGY 167. Dean NM, Butler M, Monia BP, Manoharan M. 2001. Pharmacology of 20 O-(2-methoxyethyl)-modified antisense oligonucleotides. See Ref. 46a, pp. 319– 38 168. Geary RS, Khatsenko O, Bunker K, et al. 2001. Absolute bioavailability of 20 O-(2-methoxyethyl)-modified antisense oligonucleotides following intraduodenal instillation in rats. J. Pharmacol. Exp. Ther. 296:898–904 169. Manoharan M. 2001. Oligonucleotide conjugated in antisense technology. See Ref. 46a, pp. 391–470 170. Geary RS, Watanabe TA, Truong L, et al. 2001. Pharmacokinetic properties of 20 -O-(2-methoxyethyl)-modified oligonucleotide analogs in rats. J. Pharmacol. Exp. Ther. 296:890–97 171. Crooke R, Graham M. 2001. Suborgan pharmacokinetics. See Ref. 46a, pp. 155–82 172. Zhang Z, Cook J, Nickel J, et al. 2000. Reduction of liver Fas expression by an antisense oligonucleotide protects mice from fulminant hepatitis. Nat. Biotechnol. 18:862–67 173. Hardee G, Weinbach S, Tillman L. 2001. New routes and novel formulations for delivery of antisense oligonu-

174.

175.

176.

177.

178.

P1: GBC

95

cleotides. See Ref. 46a, pp. 795– 832 Nicklin PL, Craig SJ, Phillips JA. 1998. Pharmacokinetic properties of phosphorothioates in animals—absorption, distribution, metabolism and elimination. See Ref. 105a, pp. 141–68 Zhang R, Lu Z, Zhao H, et al. 1995. In vivo stability, disposition and metabolism of a “hybrid” oligonucleotide phosphorothioate in rats. Biochem. Pharmacol. 50:545–56 Zhang R, Diasio RB, Lu Z, et al. 1995. Pharmacokinetics and tissue distribution in rats of an oligodeoxynucleotide phosphorothioate (GEM 91) developed as a therapeutic agent for human immunodeficiency virus type-1. Biochem. Pharmacol. 49:929–39 Henry SP, Stecker K, Brooks D, et al. 2000. Chemically modified oligonucleotides exhibit decreased immune stimulation in mice. J. Pharmacol. Exp. Ther. 292:468–79 Sewell L, Tami J, Geary R, et al. 2001. Phase I study of S.C and I.V. ISIS 104838 (anti-TNF-alpha), a second generation oligonucleotide. Presented at British Pharmaceutical Society, Dublin, Ireland

5 Jan 2004

11:28

AR

AR206-ME55-06.tex

AR206-ME55-06.sgm

LaTeX2e(2002/01/18) P1: GBC 10.1146/annurev.med.55.091902.105237

Annu. Rev. Med. 2004. 55:97–112 doi: 10.1146/annurev.med.55.091902.105237 First published online as a Review in Advance on Nov. 3, 2003

Annu. Rev. Med. 2004.55:97-112. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

SERUM PROTEOMICS IN CANCER DIAGNOSIS ∗ AND MANAGEMENT Kevin P. Rosenblatt,1 Peter Bryant-Greenwood,3 J. Keith Killian,1 Arpita Mehta,1 David Geho,1 Virginia Espina,1 Emanuel F. Petricoin III,2 and Lance A. Liotta1 1

Laboratory of Pathology, National Cancer Institute, Center for Cancer Research, Bethesda, Maryland 20892 2 Food and Drug Administration—National Cancer Institute Clinical Proteomics Program, Center for Biologics Evaluation and Research, Food and Drug Administration, Bethesda, Maryland 20892 3 Department of Pathology, John A. Burns School of Medicine, University of Hawaii, Honolulu, Hawaii 96822; email: [email protected]

Key Words proteomics, mass spectrometry, SELDI, protein profiling ■ Abstract Mass spectrometry–based diagnostics has the potential to revolutionize molecular medicine. Using modern mass-spectrometer technologies, clinical tests can be developed that are practical, robust, accurate, and inexpensive. Serum proteomic pattern profiling couples mass spectrometry with adaptive artificial-intelligence–based bioinformatics, which can now be employed to detect pathological states reflected in the serum proteome. With this approach, rapid and cost-effective tests with exquisite clinical sensitivity and specificity are emerging. These tools may dramatically change how disease is detected, monitored, and managed.

INTRODUCTION Cancer has traditionally been thought of as a genetic disease, but functionally, it is a proteomic affliction. Genetic defects are selected for during carcinogenesis only when they present a survival advantage for the developing cancer cell (1, 2); these survival advantages are manifest within the altered protein networks and signaling pathways that drive cancer growth, cell survival, tumor invasion, and distant metastasis (3). High-throughput genomics techniques have facilitated a better understanding of diseases such as cancer by deciphering the unique molecular signature of each tumor for subclassification into groups that predict clinical outcomes ∗ The U.S. Government has the right to retain a nonexclusive, royalty-free license in and to any copyright covering this paper.

97

5 Jan 2004

11:28

Annu. Rev. Med. 2004.55:97-112. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

98

AR

AR206-ME55-06.tex

AR206-ME55-06.sgm

LaTeX2e(2002/01/18)

P1: GBC

ROSENBLATT ET AL.

and therapeutics (4, 5). However, these techniques suffer a severe limitation in their ability to monitor changes in protein levels, post-translational modifications, or protein-protein interactions. Thus, they cannot predict the activation state of key signaling molecules in important protein networks. Proteins are the molecules of action of the cell; this is why the vast majority of drug targets are proteins. In order to understand the molecular basis of cancer, we will have to study proteins themselves. Moreover, among the critical elements of the central biological paradigm for expressing genetic information—DNA, RNA, and protein—proteins (and peptides) provide unparalleled clinical access to a treasure trove of sensitive biological information that can be used to detect and characterize disease, including cancer. The nascent field of proteomics—the complete description of all the proteins encoded by the genome—promises to rapidly expand our understanding of normal human physiology and diseases such as cancer. An important goal of clinical proteomics is to develop robust, sensitive, and specific methodologies for the simultaneous analysis of all the proteins expressed by the human genome, referred to as the human proteome, and to establish “biosignature” profiles that discriminate between disease states (6–8). Serum proteomic pattern analysis is an emerging technology that is increasingly employed for the early detection of disease, the measurement of therapeutic toxicity and disease responses, and the discovery of new drug targets for therapy. Prior to the advent of serum proteomics, two-dimensional gel electrophoresis (2D-PAGE) coupled with mass spectrometric (MS)-based protein identification (9, 10) was the classic tool of proteomic analysis. Indeed, it is still utilized to investigate the differential expression of protein patterns between disease or physiological states. This technique has recently been complemented by protein microarrays, a developing proteomic technology that examines the precise protein-protein signaling events of discrete disease states in a high-throughput and parallel fashion. Protein microarrays have been used successfully to profile the activation state of cellular signaling pathways, which is not possible by existing gene microarrays (11, 12). However, the application of 2D-PAGE MS and protein microarrays, as well as DNA microarray technology, to clinical samples requires the direct acquisition of patient tissue for the extraction of proteins or nucleic acids. Proteomic pattern analysis requires only relatively small amounts of easily accessible serum and body fluids for protein detection. Whereas both transcript and protein profiling can be applied directly to diseased tissues, the profiling of serum and body fluids is restricted to proteomic analysis. Because the procedures are simple, inexpensive, and minimally invasive, serum proteomic methods readily lend themselves to screening-test development; their robustness and ease promise to translate into routine clinical practice. A recent clinical application for proteomics to cancer research has been the discovery of new biomarkers that can serve as early detection surrogates for disease (6, 13–15). This review focuses on three areas. First, we review mass spectrometry and applied technologies that make analysis of the serum proteome possible. This is followed by an overview of the bioinformatics applications used to mine the serum proteome—a hitherto untapped information archive. We conclude with our

5 Jan 2004

11:28

AR

AR206-ME55-06.tex

AR206-ME55-06.sgm

LaTeX2e(2002/01/18)

SERUM PROTEOMICS FOR CANCER

P1: GBC

99

model for serum proteomic applications to the early diagnosis of cancer and its clinical management.

Annu. Rev. Med. 2004.55:97-112. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

MASS SPECTROMETRY ANALYSIS OF PROTEINS Mass spectrometry dates back to the early years of the twentieth century (16). The first instrument was developed by J.J. Thompson in 1912, and numerous advances by chemists and engineers over the past century resulted in an assortment of different technologies and analytical methods. The instruments are made up of three primary components: the source, which produces ions for analysis; the mass analyzer, which identifies the ions based on their mass-to-charge ratios (m/z); and the detector, which quantifies the ions resolved by the analyzer (Figure 1). Multiple subtypes of ion sources, analyzers, and detectors have been developed since the inception of mass spectrometers, and different components can be combined to create different instruments, but the principle is the same in each case: The spectrometers create ion mixtures from a sample and then resolve them into their component ions based on their m/z values. The past two decades have seen a proliferation of significantly improved spectrometric devices, allowing precision analysis of biomolecules too fragile to survive earlier instrumentation. The three ionization methods most commonly used for analysis of peptides and proteins are electrospray ionization (ESI), matrix-assisted laser desorption/ionization (MALDI), and surface-enhanced laser desorption/ionization (SELDI); these ionization sources are often coupled to time-of-flight (TOF) or quadrupole analyzers or some combination of these analyzers (for review see 17). Historically, mass spectrometry was limited to the analysis of small molecules. Larger biomolecules such as peptides or proteins simply did not survive the harsh ionization methods available to create the ions. The ESI, MALDI, and SELDI techniques permit a tender ionization of large biomolecules, called soft ionization, without too much fragmentation of the principal ions. ESI (18, 19) and MALDI (20) were both developed during the late 1980s and were the foundation for the emergence of mass spectrometry as a tool of investigation of biological samples. SELDI, developed in the early 1990s, is essentially a modification of the MALDI approach to ionization (21). All three of these ionization techniques are sensitive to the picomole-to-femtomole range that is required for application to biological samples, including carbohydrates; oligonucleotides; small polar molecules; and peptides, proteins, and post-translationally modified proteins, such as glycoproteins and phosphoproteins. MALDI and SELDI are similar in that both involve the spotting of biological samples onto a solid surface, sometimes referred to as a probe, or for SELDI a ProteinChip® (Ciphergen Biosystems, Fremont, CA). In MALDI, before the samples are applied to the probe, they are mixed with energy-absorbing compounds typically referred to as a chemical matrix. The matrix contains small chromophores that absorb light at a particular wavelength. Commonly used matrix chemicals include α-cyano-4-hydroxycinnamic acid, 3,5-dimethoxy-4-hydroxycinnamic acid, and 2,5-dihydroxybenzoic acid. After spotting, the evaporation of water and

5 Jan 2004

11:28

Annu. Rev. Med. 2004.55:97-112. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

100

AR

AR206-ME55-06.tex

AR206-ME55-06.sgm

LaTeX2e(2002/01/18)

P1: GBC

ROSENBLATT ET AL.

solvents from the mixture results in sample proteins embedded in a crystalline lattice made up of matrix molecules. The probes are inserted into the source, which includes a laser that can fire at the excitation wavelength of the matrix. The exact mechanism is unclear, but the matrix molecules absorb the photon energy from the laser, become excited, and transmit some of this energy to the peptides and proteins in the sample. The energy transfer causes the matrix and sample to vaporize into a gas cloud. The matrix then plays a role in the ionization of the specimen; some protons are ejected from matrix molecules and transferred to nearby proteins and peptides, resulting in positive ions. The matrix molecules will absorb most of the energy from the photons, minimizing the amount of energy absorbed and damage incurred by the biomolecules, but some fragmentation of peptides and proteins occurs anyway. The ion gas cloud is subsequently accelerated into the mass analyzer electrostatically. The difference between SELDI and MALDI is that the SELDI probes employ selective surfaces to capture only a fraction of proteins from a complex mixture in biological samples (for review see 22). After the sample is applied to the ProteinChip®, the surface is washed to remove unbound proteins and impurities, and the photoactivatable matrix can then be applied to the chip surface. This technique has been modified so that laser desorption can be carried out without the addition of a chemical matrix (22). Surfaces with diverse affinities for different proteins of interest can be generated to carry out on-probe chromatography, including cation/anion exchange, reverse phase (for hydrophobic interactions), and metal affinity chromatography, among others. This “quick and dirty” chromatography has distinct advantages over more involved separation technologies such as column chromatography and electrophoresis: It is swift, relatively inexpensive, simple to perform, and versatile. The technique is sensitive to the femtomole range and reproducible (23). These attributes lend themselves readily to high-throughput diagnostics. ESI creates an ion gas cloud in the source directly from the sample solution. Directing a stream of the aqueous sample through a needle held at a high voltage produces a fine mist of highly charged droplets. These droplets, which contain the proteins and peptides of interest, are electrostatically driven through nitrogen gas, air, heat, solvents, or other drying agents in order to evaporate water and solvents from the surface. As the droplets decrease in size, the surface charges are deposited onto the peptides and proteins. This ionization method is more gentle than SELDI or MALDI, resulting in much less fragmentation and producing larger molecules with multiple charge states. This multicharging of proteins generates ions within the m/z range of ESI-coupled analyzers, which often have a smaller mass range than TOF analyzers, allowing the observation of rather large molecules. Although modern mass spectrometers can measure the mass of intact proteins with a high degree of accuracy, many methods call for the measurement of peptide mass (10). As the mass of a molecule increases, the error in measurement also increases. Meanwhile, for larger proteins, the efficiency of ionization decreases and the sensitivity of their detection diminishes. Analysis of peptide fragments of proteins permits sensitive detection and accurate mass measurement. The enzymatic fragmentation of proteins using proteases such as trypsin results in peptide

5 Jan 2004

11:28

AR

AR206-ME55-06.tex

AR206-ME55-06.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:97-112. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

SERUM PROTEOMICS FOR CANCER

P1: GBC

101

lengths within the range of current mass analyzers. Because these enzymes cleave proteins at well-defined positions, i.e., they are sequence-specific, a peptide-mass fingerprint pattern will be generated that can be predicted from protein-sequence information provided by the Human Genome Project. Protein spots picked from 2D gels, or proteins and peptides purified by chromatography, can be collected, digested, and analyzed, and the mass fingerprint patterns can be compared to the database to identify proteins of interest. One advantage of serum- and fluid-based proteomics analysis is that digestion is unnecessary; the mass range of many of the biomolecules of interest in these biological samples is already well within the mass range of current mass analyzers. The MALDI and SELDI ionization approaches are often connected to TOF analyzers and have a practical mass limit between 150,000 and 300,000 Da. By comparison, ESI instruments have a mass limit of 70 kDa and are most commonly paired with quadrupole and other analyzers (16, 17) (Figure 1). TOF analyzers are more simply designed than other instruments and rely on calculations of mass from the time required for ions to travel to the detector from the ionization source. After ionization in the source, groups of ionized proteins and peptides are accelerated into the analyzer with the same kinetic energy (KE) using static electric fields. The KE imparted to each ion depends on its charge. The relationship between KE, mass (m), and velocity (v) is KE = 1/2mv2. By this formula, if two ions of different mass enter the analyzer with the same KE, the ion with smaller mass will reach the detector first because it will have greater velocity. Because the path length, d, to the detector is the same for all ions, the time that each ion takes to reach the detector is used to back-calculate its velocity (v = d/t) and mass. Thus, knowing the time it takes for each ion to hit the detector allows us to know its mass. The resolution of TOF instruments is lower than that of other analyzers because of the KE spread of ions reaching the detectors; this results in broadened peaks for each m/z value in the chromatograms. Resolution has been improved for TOF instruments by the addition of reflectron components within the analyzers that reduce the variation in KE for each ion species hitting the detector (24). Quadrupole analyzers are constructed from four parallel metal rods to which are applied direct current (DC) and radiofrequency (RF) voltages (Figure 1). These voltages create selective magnetic fields that control which ions pass through to the detector. The fields cause the ions to travel through the rods in a corkscrew trajectory, and for a particular combination of DC and RF voltages, only ions of a specific m/z value can pass through. At set voltages, the rods act as a band pass filter, but sweeping the RF voltages allows ions of ever increasing or decreasing m/z values to reach the detector. Tandem mass analyzers are instruments used for detailed structural analysis of selected peptides. These instruments are often made from combinations of analyzers such as ABI’s QSTAR® (Applied Biosystems, Foster City, CA), a hybrid system that joins two quadrupoles in tandem with a TOF analyzer (Figure 1). Particular tryptic peptide fragments can be selected and subfragmented sequentially in the two quadrupoles and then the subfragments can be measured in the TOF

5 Jan 2004

11:28

Annu. Rev. Med. 2004.55:97-112. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

102

AR

AR206-ME55-06.tex

AR206-ME55-06.sgm

LaTeX2e(2002/01/18)

P1: GBC

ROSENBLATT ET AL.

analyzer. Because the peptides will subfragment in a reproducible way, the resulting pattern is somewhat like the sequence-ladder pattern obtained in DNA sequencing. Although the analysis of the pattern is more involved than DNA sequencing, available software will allow the direct determination of the amino acid sequence of the peptide. This sequence can then be used to query the human genome and protein databases to directly identify the parent protein. An advantage of MALDI and SELDI over ESI is a higher tolerance for salts, up to the millimolar range. This benefit, in addition to the utility of MALDI/SELDI for the investigation of complex samples, makes these techniques better suited to the examination of biological samples such as serum. ESI, a kinder, gentler ionizing method, permits the formation of multicharged species; because these spectrometers are measuring m/z ratios, many relatively large ion species can be adequately detected in the apparently more limited m/z range of ESI instruments. The fact that ESI produces gas clouds from a liquid solution makes it highly compatible with liquid chromatography–based separation technologies; the entire analysis from fractionation to MS characterization can be done online up to the sequence-analysis stage. ESI is also easily adapted to quadrupole analyzers, which are particularly suited to structural analysis and protein sequencing. SELDI and MALDI sources are well suited to high-throughput applications. The recent coupling of MALDI and SELDI sources to quadrupole analyzers for serum proteomic profiling (13, 25) will enable the future rapid identification of m/z peaks from diagnostic and predictive patterns discovered in molecular profiling studies.

BIOINFORMATICS PLATFORMS FOR SERUM- AND FLUID-BASED PROTEOMICS Recently, several groups have demonstrated the viability of using serum proteomic pattern analysis for the diagnosis of ovarian (6), breast (26, 27), prostate (15, 28), and liver cancer (29). Various bioinformatics algorithms have been used for protein pattern discovery, but all of these studies have used the SELDI ionization technique along with low-resolution TOF MS analysis (see above). Earlier studies demonstrated proof of principle of biomarker development for prostate cancer using SELDI-TOF (30, 31), but some of the studies relied on the isolation of actual malignant cells from pathology specimens. Body-fluid–based diagnostics, using lavage, effluent, or effusion material, offers a less invasive approach to biomarker discovery than biopsy or surgical-specimen–dependent approaches. Additionally, serum-based approaches may offer a superior repository of biomarkers because serum is easy and inexpensive to obtain. The data within the serum proteome is both complex and poorly understood. Our utilization of powerful bioinformatics applications to analyze these data is based on four propositions: 1. The proteins or peptides present in the serum reflect the status of the various tissues that they bathe. The diseased tissue may not secrete these proteins

5 Jan 2004

11:28

AR

AR206-ME55-06.tex

AR206-ME55-06.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:97-112. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

SERUM PROTEOMICS FOR CANCER

P1: GBC

103

or peptide fragments. However, they may arise from downstream cleavage products of upregulated enzymes, immune-mediated molecules, basement membrane or extracellular matrix material, growth factors, or secretory products from another organ system generated in response to disease at a distant site—such as the liver’s attempt to clear a protein secreted by a lymphocyte in response to a tumor. 2. The proteins or peptides in the serum are specific not only to the tissue affected by disease but also to the disease process itself. For example, the serum proteomic spectrum generated by a serous carcinoma of the ovary might differ from that produced by a serous carcinoma of the endometrium—two morphologically similar tumors that share biological behavior but occur at different sites. Similarly, a serous carcinoma of the ovary might reveal key differences in its serum proteomic profile when compared to a mucinous tumor of the ovary. More subtle levels of analysis may be possible. For example, one type of ovarian serous carcinoma could be 99.9% sensitive to cisplatin, whereas another serous carcinoma may only be 80% chemosensitive; this difference in sensitivity might be reflected in the serum proteome, before, during, and after treatment. 3. The successful diagnosis of a disease does not require identification of each of the proteins or peptides that constitute the serum-proteome differences between one disease and another, or between a disease and a normal state. What is diagnostic is the fingerprint or summation of these differences. Bioinformatic analysis tools are required to detect these differences. These tools allow clinicians to find those needles in the haystack that we call serum markers. 4. Detecting differences in the serum proteome requires a large number of data points to deal with the biological spectrum of the host and of the disease state. As mentioned above, most serum proteomic methods have utilized the TOF mass analyzer coupled with a SELDI source to generate a fingerprint spectrum of serum proteins made up of peptide and protein m/z values. Our group has focused singularly on the low-mass end of the proteomic spectrum (6, 13, 15). We define this generally as all of the proteins and peptides present in the serum between 800 Da and 20,000 Da. Although this mass range has previously been disparaged as cellular flotsam, it contains more analyzable data points than the rest of the serum proteome (20,000 Da to 300,000 Da) combined (Figure 2). In order to take advantage of the data-point density in the low-mass range, our instruments need exquisite resolution. High resolution is required for accurate peak discrimination, and for MS measurements, it reflects the ability to distinguish two species of similar m/z value. Figure 3 compares the serum proteomic spectra derived from two different mass spectrometers. The first is derived from a chip read by the first instrument utilized by the National Cancer Institute/US Food and Drug Administration (NCI-FDA) Clinical Proteomics Program for the early diagnosis

5 Jan 2004

11:28

Annu. Rev. Med. 2004.55:97-112. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

104

AR

AR206-ME55-06.tex

AR206-ME55-06.sgm

LaTeX2e(2002/01/18)

P1: GBC

ROSENBLATT ET AL.

Figure 2 TOF mass-spectrometry chromatogram with ion peaks in the range of 2000–200,000 Da. Notice that there are more m/z peaks between 2000 and 20,000 Da than between 20,000 and 200,000 Da. For the NCI-FDA Clinical Proteomics Program studies, the low-mass range (generally below 20,000 or 10,000 Da) is concentrated on exclusively.

of ovarian cancer (6). The second is the same portion of the serum proteome as revealed by a high-performance instrument (13, 25). Given more data points, we can ask more sophisticated questions about our samples and generate more precise answers. Determining the mere presence or absence of disease may not require a large number of data points, but interrogating the status of a cellular pathway or detecting subclinical toxicity may. Sophisticated bioinformatics algorithms are required to investigate these serum fingerprints for useful clinical stratification as defined by pathologists (for diagnosis) or clinicians (for therapy response or toxicity). One such algorithm is the analytical platform used by the NCI-FDA ovarian cancer diagnostic application. The Proteome Quest (Correlogic Systems, Inc., Bethesda, MD) software application uses a combination of two analytical tools. The first is a genetic algorithm first described by Holland (32). Genetic algorithms parallel the natural selection process. The 15,500 m/z values generated by the SELDI-TOF process are collectively analyzed to determine which values best segregate disease from nondisease. To achieve this, the computer creates hundreds to tens of thousands of “virtual” chromosomes, which are simply small sets of m/z values randomly selected from the x-axis of the data input. Each candidate subset of m/z values contains from 5 to 20 of the 15,500 potential x-axis values from the spectra, which can be likened to combinations of genes or alleles on a chromosome (Figure 4a). A cluster-analysis method is then used as a “fitness test” of the genetic algorithm

5 Jan 2004

11:28

AR

AR206-ME55-06.tex

AR206-ME55-06.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:97-112. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

SERUM PROTEOMICS FOR CANCER

P1: GBC

105

(33, 34); it plots the pattern formed by the combined y-axis amplitudes of the candidate m/z subsets in N-dimensional space, where N is the number of m/z values in the test set (Figure 4b). The pattern formed by the relative amplitude of a particular m/z value subset is then rated for its ability to distinguish between disease and nondisease; only the best discriminator sets are saved. The m/z values within the highest-rated sets are recombined to form new subsets (Figure 4c). These new subsets are then rerated accordingly until a set of m/z values emerges that fully discriminates diseased from the nondiseased serum samples. Literally trillions of iterations can be run until the best combination of m/z values surfaces. The final output of the bioinformatics tool is a diagnostic pattern defined by the relative combination of m/z amplitudes; this pattern of amplitudes acts as a discriminator for the training-set cohorts (Figure 4b). This completes the initial step in developing the diagnostic pattern—deriving a set of m/z values and relative amplitudes that distinguish one study set (disease) from another (nondisease). The second phase of the bioinformatic analysis involves testing the derived m/z values on unknown patient serum samples. In this validation process, a pattern of relative amplitudes in N-dimensional space is generated for each unknown serum sample using the key m/z values identified by the genetic algorithm. The unknown patterns are sample-matched against the optimum values of the training set. Each unknown is then classified into one of three possible categories: cancer, unaffected, or a new, nonmatching cluster (Figure 4d). This system enabled the segregation of all ovarian cancer cases from noncancer cases using only a few microliters of serum in the assay (6). Most important, all 18 patients with stage I disease were correctly identified as well as the patients with later-stage disease. Of the 66 unknowns who did not have cancer, 63 were correctly identified. The sensitivity and specificity of this platform are, respectively, 100% and 95%. Using the high-resolution QSTAR® instrument described above, the specificity of the ovarian serum diagnostic test has recently been improved to 100%, while maintaining a sensitivity of 100% (13, 25). In an expanded validation set, which entailed testing two rounds of blinded validation serum samples not used in the original training set, every stage I ovarian cancer was correctly identified, and no benign diseases were misinterpreted as cancer. These results demonstrate the utility of the diagnostic platform as an early-detection screening test as well as a reliable diagnostic test. The algorithm above represents only one of the possible bioinformatic-analysis packages used in serum proteomics. Additional software recently used to classify serum samples include hierarchical-clustering algorithms, neural networks, and other statistical algorithms used previously in the analysis of DNA microarrays (27–29). However, to date, none of these other algorithms have yielded the exquisite level of sensitivity and specificity for serum proteomic platforms as the Proteome Quest software. An interesting recent development is the use of visualization tools to analyze the massive complex data sets encountered in proteomics (35). Using these new tools, for example, it was discovered that most of the m/z values needed to define

5 Jan 2004

11:28

106

AR

AR206-ME55-06.tex

AR206-ME55-06.sgm

LaTeX2e(2002/01/18)

P1: GBC

ROSENBLATT ET AL.

discriminatory molecular profiles were of fairly low amplitude; most of these data had been either obscured in the noise floor on low-resolution/low-performance instruments or simply ignored in favor of more prevalent peaks.

Annu. Rev. Med. 2004.55:97-112. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

SERUM CANCER DIAGNOSIS AND MANAGEMENT For many patients, cancer is diagnosed too late, after the tumor has already metastasized throughout the body. Perhaps >60% of patients with breast, colon, lung, or ovarian carcinoma have microscopic or obvious metastatic disease at the time of diagnosis; the success of therapy in such cases is often very limited. Early detection of disease often translates into higher cure rates. Ovarian cancer is an excellent example. More than 80% of patients present at an advanced clinical stage, when tumor cells have left the ovary and disseminated throughout the pelvic and peritoneal cavities or beyond (36). The five-year survival rate for these latestage patients is ∼35%, even for patients receiving the most advanced surgical and pharmacological treatment. By contrast, early-stage ovarian cancer is associated with five-year survival rates in excess of 90% with conventional treatment (36–41). It is clear that an early-detection method that identifies disease well in advance of the symptoms, when the tumor is confined to the ovary, will have a profound effect on patient survival. The present scheme for the diagnosis of cancer using an initial screening test is as follows: A screening test detects a suspicious result, which is confirmed using a “gold standard” diagnostic test; then definitive therapy can begin. The inherent inaccuracy of our present screening modalities for cancer, from the Pap smear to a serum marker such as carcinoembryonic antigen (CEA), demands a confirmatory “gold standard” test that is invasive and often expensive. These standard tests, though often viewed as definitive, present several problems in our present cancer-diagnosis strategy. The most common format for confirmatory cancer tests is a combination of two completely different tests, for example, imaging followed by biopsy. Correct diagnosis in this context requires multiple types of diagnostic instruments and multiple experts. The inaccuracy inherent in this process is the first problem with our “gold standard” modalities. The inevitable false positives and false negatives result in unnecessary treatment or the morbidity and mortality associated with undiagnosed or misdiagnosed disease. An additional problem with many confirmatory tests is that they are invasive. Invasive tests in themselves are associated with morbidity; for example, a lung biopsy can cause pulmonary edema or pneumothorax, and radiologic contrast material can trigger allergic reactions. The error rate and the invasive nature of our confirmatory tests result in increased costs and patient avoidance of the initial screening test (42). Patient avoidance, combined with the inherent error rates, represents a glass ceiling for cancer survival rates. The diagnostic bioinformatics platform used for early diagnosis can also be applied to other cancer-management issues. These include the diagnosis of minimal residual disease, the detection and localization of microscopic metastatic

5 Jan 2004

11:28

AR

AR206-ME55-06.tex

AR206-ME55-06.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:97-112. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

SERUM PROTEOMICS FOR CANCER

P1: GBC

107

disease to various sites before radiologic detection, and the monitoring of drug toxicity during treatment. Applications to non-neoplastic disease processes such as infection, autoimmune disease, prenatal disease, and dementia are being explored. An especially useful application of serum proteomic profiling has been the detection of drug-toxicity profiles for drugs such as doxorubicin. Currently, a limited number of assays are available to detect cardiotoxicity; most of these involve screening for troponin isoforms. There is no clinical diagnostic marker for early detection of cardiotoxicity, before significant cardiac damage occurs; however, a recent study utilizing SELDI-TOF and the same bioinformatics software package used in the NCI-FDA ovarian study discovered proteomic patterns that distinguish doxorubicin-induced cardiotoxicity with sensitivities and specificities of >90% (43). It is clear that using only a minute amount of patient serum, drug toxicity could be monitored in real time and therapeutics appropriately discontinued or altered before end organ damage.

LOW-MOLECULAR-WEIGHT BIOMARKERS The ovarian cancer data are important because they provide insight into the kinetics of the serum proteome. Small molecules that would otherwise be cleared by the body must be shed into the serum to the point where they achieve a steadystate concentration. Rapid changes in disease may not be detected or may be more difficult to diagnose because the small molecules fail to achieve the required steady-state concentration in the patient’s serum compartment. Thus, potentially important proteins or peptides may be missed using platforms currently under development. A related problem is protein or peptide half-life. A cohort of proteins or peptides may elude detection because of their very short half-lives. They will simply degrade into spectral noise in the time it takes to get the sample from the patient to the mass spectrophotometer. These drawbacks perhaps explain why the same platform that is 100% sensitive and 95% specific for stage I ovarian cancer can only distinguish women in labor from those not in labor with 85% sensitivity and 75% specificity (P. Bryant-Greenwood, K.P. Rosenblatt, personal communication). Mass spectrometry works best for small proteins and metabolic products. Traditional protein-based discovery tools such as 2D PAGE cannot discriminate in the low-molecular-weight range. Recent work in the NCI-FDA proteomics lab has determined that virtually all of the low-molecular-weight proteins that serve as biomarkers are complexed to high-molecular-weight serum proteins such as albumin (44). There are several reagents and protocols used presently that prefractionate and discard plentiful large serum proteins such as albumin with the intent of relieving the anticipated obscuring effect of these species. However, many, perhaps most, of the biomarkers are also being thrown away. The association of the small peptides and proteins with the carrier proteins is actually expected. Without an association with large biomolecules, the smaller species would easily be cleared by the kidneys, which effectively filter small peptides/proteins and metabolic byproducts

5 Jan 2004

11:28

108

AR

AR206-ME55-06.tex

AR206-ME55-06.sgm

LaTeX2e(2002/01/18)

P1: GBC

ROSENBLATT ET AL.

Annu. Rev. Med. 2004.55:97-112. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

from the blood. Modeling studies have shown that although the affinities between the carriers and small markers are weak, the enormous excess of carriers such as albumin or immunoglobulins guarantees a constant interaction between the two (44). Even with significant off rates for small proteins and peptides, every time a putative biomarker dissociates from its larger partner, numerous other binding partners are found nearby to immediately bind it up. Thus, carrier proteins potentially store valuable information about the physiological and pathological state and may be mined preferentially for markers in the future.

PERSPECTIVES The individual protein and peptide constituents of the patterns generated by SELDITOF, though diagnostic, are mostly unidentified. These species need not be specified at the amino acid–sequence level to be useful as a diagnostic test. The sequence information does not reveal the functional properties of the molecules; for example, what do the molecules PSA, CEA, CA-125, and CA-19-9 actually do in the human body? Certainly, they do not exist just to serve as possible markers of early cancer or recurrence. Although it is the pattern of m/z values and their relative amplitudes that discriminates diseased from nondiseased patients, this does not mean that the constituent molecules cannot be identified. Mass spectrometry, coupled with algorithmic interrogation of the data set, allows us to home in on the most significant data points without using electrophoresis. In general it has been assumed that to adequately resolve the proteins of interest, it is necessary to prefractionate the sample to some degree (45). The complex mixture of peptides in a cell lysate or serum proteome was previously believed to be too intricate, or too degraded, to yield useful information. This may indeed be the case for peptide fingerprint or MS-MS identification of serum samples, where the complexity of sequence information would be overwhelming for database searching; however, fluid proteomics does not rely on trypsin digestion, peptide fragmentation, or database searching to derive useful information from the proteome. Identification of the peptides and proteins corresponding to particular m/z values generated by SELDI-TOF analysis may be achieved using the ABI QSTAR® system, which enables MS-MS peptide sequencing. It is important to keep in mind, however, that these m/z values, though significant diagnostically, may not correlate with molecules that emanate from the tumor itself. Why a particular m/z peptide is important is another question and requires an understanding of cellular signaling and tissue interactions that, for the most part, are completely unknown. The serum diagnostic patterns may be a byproduct of the complex tumor-host microenvironment and may reflect aberrant modifications to host proteins rather than tumor-cell–specific markers (13, 25). The salient point is that the serum proteomic platform provides a sensitive tool for segregating a few of the important proteins and peptides from the hundreds of thousands to millions that exist in a serum sample or tissue lysate.

5 Jan 2004

11:28

AR

AR206-ME55-06.tex

AR206-ME55-06.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:97-112. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

SERUM PROTEOMICS FOR CANCER

P1: GBC

109

Ideally, algorithms employed for serum proteomic profiling should filter out temporal fluctuations in the serum proteome unrelated to the biological condition being considered. For example, in the case of cancer diagnosis, the atypical neoplastic cells themselves may directly spill evidence of their existence into the bloodstream for detection by the QSTAR®, such as a mutant protein or byproducts of a deranged cell signaling pathway. In this scenario, there would be no need for a “fasting” serum proteome or a “glucose challenge” serum profile, for the evidence of disease would be manifest continually. However, as biological entities, malignancies—and therefore their serum signatures—may vary with host environment and cellular behavior. At the present time, since we do not know precisely what contributes to the uniqueness of the profile, we must consider the possibility of serum proteome constituents that fluctuate in response to outside influence. Ultimately, serum proteomic profiling may be enhanced through physiological or pharmacological interventions that bring out the signal. It is important to surmise the origin of significant contributory factors in the design of disease/control serum training sets, to ensure the generation of marker profiles with maximum diagnostic sensitivity. Such considerations also may highlight potential shortcomings of some training sets. If suitable mimics of the disease in question are not included in the training sets, the bioinformatics tools may not be able to discriminate between benign and malignant disease. For example, the original ovarian cancer serum profiling study included serum samples from female patients with inflammatory diseases such as rheumatoid arthritis, colitis, and sinusitis (6). Because malignancies often elicit an inflammatory response from the host, testing the algorithm against inflammatory diseases, or including inflammatory diseases in the original training set, is crucial to weeding out proteomic signatures that represent inflammation or generalized systemic reactions to disease rather than cancer profiles. Until now, the disease populations have been carefully chosen so that the samples fall into simple groups—e.g., serum samples were taken from ovarian cancer patients before surgery or other treatment (6, 13). It is possible that use of common overthe-counter NSAIDs for a few days prior to testing may suffice to perturb serum proteomic profiles and effectively knock out evidence of an early-stage lymphoproliferative neoplasm, not to mention the use of steroid and other immunomodulator treatments. Recent antibiotic therapy for an upper respiratory tract infection may likewise knock out the microbial contribution to the serum profile. As described above, one may suspect many factors to challenge the sensitivity of cancer-detecting algorithms. But what about the specificity of the assay? Is it possible that more than one cancer will share a serum proteomic signature, especially if neoplastic signaling pathways are being detected? We would not want to tell a patient, “Your proteomic signature is consistent with both colorectal carcinoma and breast cancer.” There is a mounting need for broadly trained anatomic pathologist/physician scientists to be consulted on the design of algorithm training sets, drawing on knowledge of embryogenesis, histogenesis, and carcinogenesis, as well as molecular cancer pathways shared by neoplasms. These same future pathologists will certainly be called on when algorithms fall short, and future

5 Jan 2004

11:28

Annu. Rev. Med. 2004.55:97-112. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

110

AR

AR206-ME55-06.tex

AR206-ME55-06.sgm

LaTeX2e(2002/01/18)

P1: GBC

ROSENBLATT ET AL.

autopsy reports may well include discussions of algorithm-pathological correlations. Although host, inflammatory, and other factors may modify the biology of cancer, the extent to which they antagonize proteomic algorithms remains to be tested; an “overly healthy” training set may lower the specificity of algorithms when applied to human populations. The pathologist is an ideal and requisite consultant in the programming of cancer algorithms. Defining the signatures of cancer and comparing them to host-response signatures and to innumerable environmental and pharmacological conditions will move early-stage cancer diagnosis in the direction of cancer prevention. The application of serum proteomics to early diagnosis and the monitoring of toxicity and disease state is a daunting challenge. It requires a host of disparate disciplines to work together to input meaningful data into the system. These data may be in the form of clinical information, human tissue samples, pathological diagnoses, clinical pharmacology, or MS-derived data. The challenge is no longer just the development of new technologies but rather the best use and integration of these technologies for the diagnosis and treatment of disease; the process of this integration is a new, evolving field of translational medicine. The Annual Review of Medicine is online at http://med.annualreviews.org

LITERATURE CITED 1. Hanahan D, Weinberg RA. 2000. The hallmarks of cancer. Cell 100:57–70 2. Hunter T. 2000. Signaling—2000 and beyond. Cell 100:113–27 3. Liotta LA, Kohn EC. 2001. The microenvironment of the tumour-host interface. Nature 411:375–79 4. Chung CH, Bernard PS, Perou CM. 2002. Molecular portraits and the family tree of cancer. Nat. Genet. 32:533–40 (Suppl.) 5. Staudt LM. 2002. Gene expression profiling of lymphoid malignancies. Annu. Rev. Med. 53:303–18 6. Petricoin EF, Ardekani AM, Hitt BA, et al. 2002. Use of proteomic patterns in serum to identify ovarian cancer. Lancet 359:572–77 7. Bichsel VE, Liotta LA, Petricoin EF 3rd. 2001. Cancer proteomics: from biomarker discovery to signal pathway profiling. Cancer J. 7:69–78 8. Liotta L, Petricoin E. 2000. Molecular profiling of human cancer. Nat. Rev. Genet. 1:48–56

9. Link AJ. 1999. 2-D Proteome Analysis Protocols. Totowa, NJ: Humana. xvii, 601 pp. 10. Westermeier R, Naven T. 2002. Proteomics in Practice: a Laboratory Manual of Proteome Analysis. Weinheim: WileyVCH. xxiv, 316 pp. 11. Liotta LA, Espina V, Mehta AI, et al. 2003. Protein microarrays: meeting analytical challenges for clinical applications. Cancer Cell 3:317–25 12. Espina V, Mehta AI, Liotta LA, et al. 2003. Pathology of the future: molecular profiling for targeted therapy. Cancer Invest. Submitted 13. Conrads TP, Fusaro VA, Ross S, et al. 2003. Multiple high-resolution serum proteomic features for ovarian detection. J. Natl. Cancer Inst. Submitted 14. Petricoin EF, Zoon KC, Kohn EC, et al. 2002. Clinical proteomics: translating benchside promise into bedside reality. Nat. Rev. Drug Discov. 1:683–95 15. Petricoin EF 3rd, Ornstein DK, Paweletz

5 Jan 2004

11:28

AR

AR206-ME55-06.tex

AR206-ME55-06.sgm

LaTeX2e(2002/01/18)

SERUM PROTEOMICS FOR CANCER

16.

17.

Annu. Rev. Med. 2004.55:97-112. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

18.

19.

20.

21.

22.

23.

24.

25.

26.

CP, et al. 2002. Serum proteomic patterns for detection of prostate cancer. J. Natl. Cancer Inst. 94:1576–78 Siuzdak G. 1996. Mass Spectrometry for Biotechnology. San Diego: Academic. xvi, 161 pp. Mann M, Hendrickson RC, Pandey A. 2001. Analysis of proteins and proteomes by mass spectrometry. Annu. Rev. Biochem. 70:437–73 Fenn JB, Mann M, Meng CK, et al. 1989. Electrospray ionization for mass spectrometry of large biomolecules. Science 246:64–71 Whitehouse CM, Dreyer RN, Yamashita M, Fenn JB. 1985. Electrospray interface for liquid chromatographs and mass spectrometers. Anal. Chem. 57:675–79 Karas M, Hillenkamp F. 1988. Laser desorption ionization of proteins with molecular masses exceeding 10,000 daltons. Anal. Chem. 60:2299–301 Hutchens TW, Yip T-T. 1993. New desorption strategies for the mass spectrometric analysis of macromolecules. Rapid Commun. Mass Spectrom. 7:576–80 Merchant M, Weinberger SR. 2000. Recent advancements in surface-enhanced laser desorption/ionization-time of flightmass spectrometry. Electrophoresis 21: 1164–77 Frears ER, Stephens DJ, Walters CE, et al. 1999. The role of cholesterol in the biosynthesis of beta-amyloid. Neuroreport 10:1699–705 Cornish TJ, Cotter RJ. 1993. A curvedfield reflectron for improved energy focusing of product ions in time-of-flight mass spectrometry. Rapid Commun. Mass Spectrom. 7:1037–40 Conrads TP, Zhou M, Petricoin III EF, et al. 2003. Cancer diagnosis using protomic patterns. Expert Rev. Mol. Diagn. 3:411–20 Rai AJ, Zhang Z, Rosenzweig J, et al. 2002. Proteomic approaches to tumor marker discovery. Arch. Pathol. Lab. Med. 126:1518–26

P1: GBC

111

27. Li J, Zhang Z, Rosenzweig J, et al. 2002. Proteomics and bioinformatics approaches for identification of serum biomarkers to detect breast cancer. Clin. Chem. 48:1296–304 28. Adam BL, Qu Y, Davis JW, et al. 2002. Serum protein fingerprinting coupled with a pattern-matching algorithm distinguishes prostate cancer from benign prostate hyperplasia and healthy men. Cancer Res. 62:3609–14 29. Poon TC, Yip TT, Chan AT, et al. 2003. Comprehensive proteomic profiling identifies serum proteomic signatures for detection of hepatocellular carcinoma and its subtypes. Clin. Chem. 49:752–60 30. Verma M, Wright GL Jr, Hanash SM, et al. 2001. Proteomic approaches within the NCI early detection research network for the discovery and identification of cancer biomarkers. Ann. NY Acad. Sci. 945:103– 15 31. Paweletz CP, Liotta LA, Petricoin EF 3rd. 2001. New technologies for biomarker analysis of prostate cancer progression: laser capture microdissection and tissue proteomics. Urology 57:160–63 32. Holland JH. 1994. Adaptation in Natural and Artificial Systems: an Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence. Cambridge, MA: MIT Press 33. Kohonen T. 1982. Self-organizing formation of topologically correct feature maps. Biol. Cybern. 43:59–69 34. Kohonen T. 1990. The self-organizing map. Proc. Inst. Electr. Electron. Eng. 78:1464–80 35. Johann DJ, McGuigan MD, Fusaro VA, et al. 2003. Visualization and datamining of serum proteomic data for early cancer detection. See Ref. 35a, Abstr. #1020 35a. 2003. Proc. 94th Annu. Meet. Am. Assoc. Cancer Res., Vol. 44. Washington, DC: Am. Assoc. Cancer Res. 36. Ozols RF, Rubin SC, Thomas GM, Robboy SJ. 2000. Epithelial ovarian cancer. See Ref. 36a, pp. 981–1058

5 Jan 2004

11:28

Annu. Rev. Med. 2004.55:97-112. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

112

AR

AR206-ME55-06.tex

AR206-ME55-06.sgm

LaTeX2e(2002/01/18)

P1: GBC

ROSENBLATT ET AL.

36a. Hoskins WJ, Perez CA, Young RC. 2000. Principles and Practice of Gynecologic Oncology. Philadelphia: Lippincott Williams & Wilkins. 3rd ed. 37. Bast RC Jr, Klug TL, St John E, et al. 1983. A radioimmunoassay using a monoclonal antibody to monitor the course of epithelial ovarian cancer. N. Engl. J. Med. 309:883–87 38. Menon U, Jacobs I. 2000. Tumor markers. See Ref. 36a, pp. 165–82 39. Menon U, Jacobs IJ. 2000. Recent developments in ovarian cancer screening. Curr. Opin. Obstet. Gynecol. 12:39–42 40. Jacobs IJ, Skates SJ, MacDonald N, et al. 1999. Screening for ovarian cancer: a pilot randomised controlled trial. Lancet 353:1207–10 41. Cohen LS, Escobar PF, Scharm C, et al.

42.

43.

44.

45.

2001. Three-dimensional power Doppler ultrasound improves the diagnostic accuracy for ovarian cancer prediction. Gynecol. Oncol. 82:40–48 Sung NS, Crowley WF Jr, Genel M, et al. 2003. Central challenges facing the national clinical research enterprise. JAMA 289:1278–87 Ardekani AM, Fusaro V, Ross SJ, et al. 2003. Early doxorubicin-induced toxicity is identified by proteomic patterns in serum. See Ref. 35a Mehta AI, Ross SJ, Lowenthal MS, et al. 2003. Biomarker amplification by serum carrier protein binding. Submitted to Dis. Markers Liebler DC. 2002. Introduction to Proteomics: Tools for the New Biology. Totowa, NJ: Humana. ix, 198 pp.

Rosenblatt.qxd

1/5/2004

3:29 PM

Page 1

Annu. Rev. Med. 2004.55:97-112. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

SERUM PROTEOMICS FOR CANCER

C-1

Figure 1 Schematic of electrospray ionization (ESI), matrix-assisted laser desorption/ionization (MALDI), and surface-enhanced laser desorption/ionization (SELDI) mass spectrometry. (a) Mass spectrometers generally have three components: the source, the analyzer, and the dectector. In MALDI and SELDI, the protein solution is dried down onto a probe or chip. A matrix is then applied, which results in the ionization and desorption of the proteins after a laser pulse. The time-of-flight (TOF) analyzer separates the ions by charge and they are measured in the detector. The ions with smaller mass reach the detector first. The difference between MALDI and SELDI is in the probes used to bind the protein solution. The SELDI chips have a selective surface that binds a subset of the serum peptides and proteins based on their physiochemical properties; a range of selective surfaces can be used to trap different subsets from serum. (b) ESI ionizes proteins by passing the solution through a highvoltage nozzle. Nitrogen gas can be used to desolvate the solution, depositing charge directly onto the peptides and proteins. The quadrupole analyzers use radiofrequency and direct-current voltages to select ions based on their mass and charge. Only a single quadrupole analyzer is depicted, but many modern instruments use these analyzers in tandem. Dissociation of peptides into subfragments using inert gases occurs in the second quadrupole (not shown) for peptides sequencing; in the case of the ABI QSTAR® instrument, these subfragments are resolved in a TOF analyzer. The ions are detected by the detector. For direct serum profiling of patient samples in the NCIFDA studies, SELDI chips are used and these ions are initially scanned in the QSTAR® without dissociation of peptide/protein fragments. In this case, the instrument is used for its high resolution and mass accuracy. In principle, the mass-spectrometry sequencing mode can subsequently be used to sequence selected peptides.

Rosenblatt.qxd

Annu. Rev. Med. 2004.55:97-112. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

C-2

1/5/2004

3:29 PM

Page 2

ROSENBLATT ET AL.

Figure 3 Comparison between chromatograms generated on a low-performance mass spectrometer (Ciphergen SELDI-TOF MS) and a high-performance instrument (ABI QSTAR® Pulsar QqTOF MS). The same sample was run on both instruments using the identical protein chip. Note the significant increase in data points possible with the high-performance machine.

Rosenblatt.qxd

1/5/2004

3:29 PM

Page 3

Annu. Rev. Med. 2004.55:97-112. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

SERUM PROTEOMICS FOR CANCER

C-3

Figure 4 The Proteome Quest software application. (a) Graphic representation of the genetic algorithm. In the case of data generated on the Ciphergen mass spectrometer, the ~15,500 data points are randomly distributed among up to 50,000 possible “virtual” chromosomes. [For searches testing five discriminators per chromosome, (15,500)5 different chromosomes are possible, but a maximum of 50,000 is allowed by the program.] Each chromosome then undergoes a “fitness test,” which assesses the ability of the particular subset of m/z values to discriminate between diseased and nondiseased patients. The algorithm “knows” which patients are diseased because this information is explicit in the testing phase. (b) Proteomic pattern clustering in N-space. The “fitness test” is an N-dimensional cluster analysis that determines whether diseased and nondiseased patients cluster together based on the particular combination of m/z values selected by the genetic algorithm. Clusters are illustrated for only three discriminators (three-dimensional analysis) for clarity. The patterns are derived from Euclidian plots in N-space (three-dimensional space in the figure) of the amplitude values for each m/z peak selected. In our example, the same three m/z values are used to discriminate cancer from noncancer; however, these selected peaks differ in relative intensity between the two samples. Thus, the cancer and noncancer patients cluster separately in three-dimensional space. The Proteome Quest application typically can consider from 5 to 20 discriminators during modeling (5-space to 20-space clustering). (c) The algorithm saves virtual chromosomes (combinations of m/z values) that pass the fitness test. These chromosomes can be recombined in a manner similar to crossing over during meiosis. Only relatively rare permutations are desired so as not to completely reshuffle the combinations of m/z peaks. Thus, new combinations of peaks can be created on the virtual chromosomes in order to create better discriminators. Literally trillions of iterations of the algorithm can be run until the best chromosomes emerge. (d) Validation phase of the algorithm. The best chromosomes and their corresponding patterns in N-space that emerge from the modeling are validated against a blinded set of patients. At this point, only the particular m/z values on the best chromosomes are used to plot the relative amplitudes of the unknowns in N-dimensional space (3space). Three possible outcomes are illustrated: pattern matches to the cancer group, matches to the benign or nondiseased group, and a “no match” category. This last category is important to avoid forcing samples into erroneous designations.

Annu. Rev. Med. 2004.55:97-112. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Rosenblatt.qxd

C-4

1/5/2004 3:29 PM

ROSENBLATT ET AL.

Figure 4 (Continued)

Page 4

5 Jan 2004

11:31

AR

AR206-ME55-07.tex

AR206-ME55-07.sgm

LaTeX2e(2002/01/18) P1: GBC 10.1146/annurev.med.55.091902.103730

Annu. Rev. Med. 2004. 55:113–32 doi: 10.1146/annurev.med.55.091902.103730 c 2004 by Annual Reviews. All rights reserved Copyright ° First published online as a Review in Advance on Sept. 15, 2003

MOLECULAR NEUROBIOLOGY OF DRUG ADDICTION

Annu. Rev. Med. 2004.55:113-132. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Jennifer Chao and Eric J. Nestler Department of Psychiatry and Center for Basic Neuroscience, The University of Texas, Southwestern Medical Center, 5323 Harry Hines Boulevard, Dallas, Texas 75390-9070; email: [email protected]

Key Words reward, nucleus accumbens, mesolimbic dopamine, CREB, 1FosB ■ Abstract Addiction can be viewed as a form of drug-induced neural plasticity. One of the best-established molecular mechanisms of addiction is upregulation of the cAMP second messenger pathway, which occurs in many neuronal cell types in response to chronic administration of opiates or other drugs of abuse. This upregulation and the resulting activation of the transcription factor CREB appear to mediate aspects of tolerance and dependence. In contrast, induction of another transcription factor, termed 1FosB, exerts the opposite effect and may contribute to sensitized responses to drug exposure. Knowledge of these mechanisms could lead to more effective treatments for addictive disorders.

INTRODUCTION The Diagnostic and Statistical Manual of Mental Disorders refers to drug addiction as “substance dependence,” the essential characteristic of which is a compulsive pattern of drug-seeking and drug-taking behavior that continues despite adverse consequences. “Addiction,” however, is by far the preferable term, since “dependence”—a pharmacological term—describes only one of the many types of adaptations to drug exposure that comprise addiction. “Dependence” refers to drug-induced adaptations that compensate for drug exposure and lead to an array of withdrawal symptoms when drug use ceases. Withdrawal symptoms vary with the substance but usually involve a significant negative affective state (dysphoria) and in some cases profound somatic abnormalities. “Tolerance” refers to drug-induced adaptations that lead to diminishing effects of a constant drug dose. “Sensitization,” or reverse tolerance, refers to drug-induced adaptations that enhance drug responsiveness with repeated drug exposure. Many drugs cause both tolerance and sensitization, with some drug effects decreasing over time while others increase. The term drug abuse is often used, though confusing, since it can refer to widely varying levels of drug intake. 0066-4219/04/0218-0113$14.00

113

5 Jan 2004

11:31

Annu. Rev. Med. 2004.55:113-132. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

114

AR

CHAO

AR206-ME55-07.tex

¥

AR206-ME55-07.sgm

LaTeX2e(2002/01/18)

P1: GBC

NESTLER

The key questions in the study of addiction are why some individuals make the transition from casual drug use to compulsive use (addiction) whereas others do not, and why relapse is so common, independent of the time elapsed since last drug use. Theories proposed to explain the compulsive element of addiction include hedonic or opponent-process, incentive-sensitization, and learning-based theories (for review, see 1). As the name implies, the hedonic theory characterizes the transition to addiction in terms of affective states, either positive or negative, experienced by the individual. The basic tenet of this theory draws from the traditional view of addiction, in which initial drug-taking results in a positive affective state (e.g., euphoria or pleasure) but, upon cessation of the drug, a withdrawal reaction of anhedonia or dysphoria occurs. The need to alleviate this negative affective state by continued drug use would underlie the compulsive element of addiction. The dueling processes of euphoria and dysphoria, measured on a hedonic scale, make up the components of the opponent-process theory (2, 3). As the hedonic set point is raised, the same amount of drug results in weaker hedonic effects and stronger negative after-effects when the drug is withdrawn. The transition from the initial positive hedonic state to an increasing negative hedonic state draws the individual into a spiral of homeostatic dysregulation of brain reward pathways, resulting in the development of addiction and vulnerability to relapse. The theory of incentive sensitization draws a distinction between drug “liking” (an affective response, as described above) and drug “wanting.” This theory proposes that the excessive wanting of drug and the excessive incentive salience attached to drug-associated stimuli drives compulsive drug seeking, drug taking, and relapse (4, 5). Learning-based theories of addiction propose that repeated drug exposure is associated with particularly strong memories, mediated by drug-induced changes in brain reward regions. Accordingly, drug taking is a learned response to conditioned stimuli, such as drug-associated cues (6–8). It is likely that a combination of factors proposed in each of these theories contributes to the neural and behavioral pathology that underlies addiction. A major challenge in drug-abuse research is to identify the molecular and cellular changes that drugs cause in the brain to produce the complex behavioral syndrome called addiction. Before discussing these molecular and cellular mechanisms of addiction, we briefly summarize the brain’s reward circuitry. It is generally believed that drugs of abuse usurp neural circuitry in the brain that normally controls responses to natural rewards, such as food, sex, and social interactions (Figure 1). Perhaps the most important mediator of drug reward per se is the mesolimbic dopamine system, comprised of dopamine neurons with cell bodies in the ventral tegmental area (VTA) of the midbrain and the projection areas of these neurons in the limbic forebrain, in particular, the nucleus accumbens (NAc). This VTA-NAc circuit is a key detector of a rewarding stimulus; drug-induced changes in these regions could increase or decrease an individual’s sensitivity to the rewarding effects of drug exposure. The amygdala is particularly

5 Jan 2004

11:31

AR

AR206-ME55-07.tex

AR206-ME55-07.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:113-132. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

MOLECULAR NEUROBIOLOGY OF DRUG ADDICTION

P1: GBC

115

important for conditioned aspects of drug exposure, for example, establishing associations between environmental cues and both the rewarding actions of acute drug exposure and the aversive symptoms during drug withdrawal. The hippocampus, a traditional memory circuit, is no doubt crucial for memories of the context of drug exposure and withdrawal. The hypothalamus is important in mediating many effects of drugs on the body’s physiological state. Probably most important, but least understood, is the role of the frontal regions of the cerebral cortex, such as the medial prefrontal cortex, anterior cingulate cortex, and orbitofrontal cortex. These regions provide executive control over drug use, which is severely impaired in many addicts. Of course, these brain regions, and many more, do not function separately but are parts of a complex and highly integrated circuit that is profoundly altered by drug exposure.

MOLECULAR MECHANISMS OF ADDICTION Not surprisingly, drugs of abuse have been reported to change literally hundreds of proteins in the various reward-related brain regions mentioned above. Rather than attempting a comprehensive review of these drug-induced changes, we focus on a small number of well-characterized changes that have been shown to contribute to certain features of the behavioral syndrome of addiction. Moreover, we focus on drug-induced changes in transcription factors, which are nuclear proteins that bind to the regulatory regions of certain genes and thereby regulate their transcription into mRNA. This focus on transcription factors is based on the notion that druginduced changes at the level of gene expression could explain the longevity of the behavioral abnormalities associated with addiction.

cAMP and CREB CREB (cAMP response element binding protein) is a member of the bZIP superfamily of transcription factors. It is composed of a C-terminal basic domain that is responsible for binding to DNA and a leucine zipper domain that mediates dimerzation with itself or other members of the CREB family of transcription factors, including CREM (cAMP response element modulator) and ATF-1 (activating transcription factor 1). The CRE (consensus cAMP response element) to which CREB dimers bind consists of the palindromic sequence TGACGTCA. Many genes have CRE sites in their promoters, including neuropeptides, neurotransmitter synthesizing enzymes, neurotransmitter receptors, signaling proteins, and other transcription factors (9, 10). CRE-mediated transcription requires CREB activation via phosphorylation at Ser-133. Phosphorylation and subsequent activation of CREB is a site of convergence for several signal transduction cascades, including the cAMP pathway via protein kinase A (PKA), intracellular Ca2+ via Ca2+calmodulin-dependent kinases (CaMK), the Ras/extracellular signal regulated kinase (ERK) protein kinase pathway, the phosphotidylinositol-3-kinase (PI3K)/Akt

5 Jan 2004

11:31

Annu. Rev. Med. 2004.55:113-132. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

116

AR

CHAO

AR206-ME55-07.tex

¥

AR206-ME55-07.sgm

LaTeX2e(2002/01/18)

P1: GBC

NESTLER

kinase pathway, and stress-induced signaling cascades (10). CREB binding protein (CBP) subsequently binds to the phosphorylated CREB dimer and serves as an adaptor to the transcription initiation complex. The histone acetyltransferase (HAT) activity endogenous to CBP unravels chromatin and facilitates transcription (9, 11). CREB is of particular interest in drug addiction because its activation is downstream of the cAMP signaling pathway, whose upregulation has been extensively characterized as an adaptation to chronic exposure to drugs of abuse (12). Beginning with cultured neuronal cell lines and extending to several brain regions involved in addiction, chronic opiate exposure has been shown to upregulate the cAMP signaling cascade (13, 14). This upregulation is viewed as a homeostatic compensatory response to the acute inhibitory actions of opiates, which bind to Gi-coupled receptors and inhibit adenylyl cyclase production of cAMP. Upregulation of the cAMP pathway mediates several aspects of addiction, depending on the specific region of the brain involved (15). Effects on the locus coeruleus (LC) and NAc are discussed here. The response to repeated exposure to morphine has been studied extensively in the LC, which has served as a useful model system (Figure 2). The LC, located at the base of the fourth ventricle, is the major noradrenergic nucleus in the brain, mediating the control of attention, vigilance, and the sympathetic nervous system (16, 17). Upregulation of the cAMP pathway and CREB in the LC is implicated in mediating some of the symptoms underlying physical opiate dependence and withdrawal (14, 18). Acute exposure to opiates inhibits the cAMP signaling cascade; however, upon chronic opiate administration, CREB expression is increased in the LC (12, 19), implying a homeostatic or compensatory regulatory mechanism. This increased CREB activity appears to play an important role in physical opiate dependence and withdrawal (18). Among the genes involved are adenylyl cyclase type VIII and tyrosine hydroxylase, whose expression is upregulated by chronic morphine administration via a CREB-dependent mechanism (18, 20). Consistent with these data, mice containing targeted mutations of the α and 1 isoforms of the CREB gene show attenuated physical symptoms of morphine withdrawal (21). Interestingly, these mice also exhibit a strong aversion to opiate withdrawal in a conditioned-aversion paradigm despite their attenuated physical withdrawal symptoms, indicating that the mechanisms or circuitry of physical dependence may be distinct from those mediating the negative motivational aspects of morphine withdrawal (22). Chronic exposure to opiates, cocaine, and alcohol also upregulates the cAMP pathway in the NAc (14, 23, 24). As would be expected from this upregulation, activation of CREB and CRE-mediated transcription have also been observed in response to chronic morphine and amphetamine treatments in this brain region (25–27). Understanding of the functional role of CREB activity in addiction was facilitated by a series of studies examining the behavioral manifestations of a localized increase in CREB activity in the NAc. First, bilateral intra-NAc infusions of a PKA activator, which would be expected to activate CREB, decreased cocaine reward in rats as demonstrated by reduced baseline cocaine self-administration,

5 Jan 2004

11:31

AR

AR206-ME55-07.tex

AR206-ME55-07.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:113-132. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

MOLECULAR NEUROBIOLOGY OF DRUG ADDICTION

P1: GBC

117

whereas infusion of PKA inhibitors increased cocaine reward (28). Studies overexpressing CREB in the rat NAc via viral-mediated gene transfer provided more direct evidence of the effects of CREB activity on reward. Increased CREB expression decreases the rewarding effects of cocaine, opiates, and sucrose, a natural reward (25, 29), whereas expression of the dominant-negative mutant form of CREB resulted in the opposite effects. Finally, inducible transgenic mice overexpressing CREB in the NAc and dorsal striatum (30) demonstrate decreased preference for cocaine in the conditioned-place-preference paradigm (CA McClung, EJ Nestler, unpublished observations). Together, these data indicate that upregulation of the cAMP pathway and CREB in the NAc as a result of chronic drug administration decreases the rewarding effects of cocaine and morphine. CREB α/1 mutant mice showed partially consistent results, demonstrating increased rewarding responses to cocaine as assayed by conditioned-place-preference assays while showing no increase in response to morphine (22). The anatomically unrestricted nature of the CREB mutation in these mice makes it unclear whether these responses are due to lack of CREB in the NAc or perhaps elsewhere in the brain; however, these interesting results highlight the importance of generating inducible, region-specific knockout mice to further elucidate these mechanisms. Beyond attenuating the rewarding effects of drugs of abuse, upregulation of the cAMP pathway and CREB in the NAc may also contribute to states of dysphoria seen early in withdrawal (15, 32). Thus, CREB overexpression in the NAc, achieved with viral vectors or in inducible transgenic mice, produces depressionlike responses in the forced-swim and learned-helplessness tests, whereas mutant CREB expression causes antidepressant-like responses (33, 33a). Recent data more precisely define the depression-like state mediated by CREB, namely, a general state of emotional numbness and anhedonia (25). Efforts are under way to identify target genes for CREB in the NAc. One apparent target is dynorphin, an opioid peptide expressed in a subset of medium spiny neurons in the NAc, which is induced in this region after chronic drug exposure (29, 34–36) (Figure 3). Dynorphin release from the NAc contributes to dysphoria during withdrawal through what amounts to a negative-feedback loop to VTA dopamine neurons (32, 37). Dynorphin binds to κ opioid receptors on VTA dopamine neuron cell bodies and terminals to inhibit their activity and decrease dopamine release in the NAc (38). The cocaine aversion caused by CREB overexpression in the NAc can be attenuated with a κ opioid antagonist (29), as can depression-like responses seen under these conditions (33). The search for additional CREB target genes in the NAc has been extended to the use of DNA microarrays on bitransgenic mice (33a; CA McClung, EJ Nestler, unpublished observations) expressing CREB or a dominant negative CREB (mCREB) in an inducible, region-specific manner. These studies have shown that the vast majority of the genes upregulated by CREB in the NAc are downregulated by mCREB, attesting to both the functional implications of increased expression of these genes and the reliability of the microarray detection technique. In addition, some of these CREB-regulated genes comprise a small subset of the genes regulated in this brain

5 Jan 2004

11:31

Annu. Rev. Med. 2004.55:113-132. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

118

AR

CHAO

AR206-ME55-07.tex

¥

AR206-ME55-07.sgm

LaTeX2e(2002/01/18)

P1: GBC

NESTLER

Figure 3 Regulation of CREB by drugs of abuse. The figure shows a ventral tegmental area (VTA) dopamine (DA) neuron innervating a class of nucleus accumbens (Nac) GABAergic projection neurons that expressed dynorphin (dyn). Dynorphin serves as a negative-feedback mechanism in this circuit: Dynorphin, released from terminals of the NAc neurons, acts on κ opioid receptors located on nerve terminals and cell bodies of the DA neurons to inhibit their functioning. Chronic exposure to cocaine or opiates upregulates the activity of this negative-feedback loop by upregulating the cAMP pathway, activation of CREB, and induction of dynorphin. (From Reference 113 with permission.)

region by cocaine administration. These potential target genes provide possible mechanisms of drug-induced plasticity and deserve further study.

1FosB Immediate early genes are a class of genes whose expression is induced within minutes of exposure to a stimulus. Of particular interest in the study of addiction are the Fos and Jun families of immediate early genes, which encode transcription factors. The Fos family of transcription factors includes c-Fos, FosB, Fos-related antigens 1 and 2 (Fra-1 and -2), and 1FosB. 1FosB is a truncated splice variant of full-length FosB, and lacks a portion of the C-terminal transactivation domain present in other Fos proteins (39). Fos family members heterodimerize with Jun family transcription factors (c-Jun, JunB, JunD) to form the activator protein– 1 (AP-1) complex. The AP-1 complex binds to specific DNA sequences in the promoters of various target genes, with the consensus sequence TGAC/GTCA. AP-1 complexes can act as either a transcriptional inducer or repressor, depending on the specific AP-1 binding site and promoter in question.

5 Jan 2004

11:31

AR

AR206-ME55-07.tex

AR206-ME55-07.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:113-132. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

MOLECULAR NEUROBIOLOGY OF DRUG ADDICTION

P1: GBC

119

Acute exposure to drugs of abuse rapidly (in 1–4 h) induces all Fos family members in the NAc and dorsal striatum (40, 41) (Figure 4). Even with continued drug exposure, levels of these proteins decline rapidly toward basal levels within 8– 12 h. However, biochemically modified isoforms of 1FosB exhibit a very different expression pattern. Acutely, 1FosB expression is only modestly induced, but it persists long after the other Fos family members have returned to basal levels. In fact, several lines of evidence point toward 1FosB being a unique target of chronic exposure to drugs of abuse. First, as noted above, whereas other Fos family proteins respond to drugs of abuse with a characteristic sharp upregulation followed by a quick decline to basal levels within hours (42–44), 1FosB isoforms are very stable and demonstrate in vivo half-lives of weeks (45). They therefore persist for weeks after the drug is withdrawn (46). As a result, 1FosB levels gradually accumulate with repeated drug exposure, suggesting that its dynamics allow it to play a longer-term role in subsequent regulation of gene expression. Second, 1FosB expression is significantly induced in response to chronic exposure to several drugs of abuse, including cocaine, amphetamine, opiates, nicotine, ethanol, and phencyclidine (40, 47–51). Importantly, these substances induce 1FosB most prominently in the NAc and dorsal striatum, but to a lesser extent in other brain regions known to be important in addiction, including the NAc amygdala and prefrontal cortex (51). Because of its unique temporal properties and its induction by virtually all drugs of abuse, the functional significance of 1FosB in drug-related behaviors has been studied extensively. In response to chronic exposure to drugs of abuse, 1FosB is selectively upregulated within a subpopulation of medium spiny neurons containing the neuropeptides substance P and dynorphin in the NAc and dorsal striatum (41, 51). Exposure to antipsychotic drugs also induces 1FosB expression in these same regions, but this induction occurs in the other major subpopulation of medium spiny neurons in the NAc, namely those neurons containing the neuropeptide enkephalin (52, 53). Further, 1FosB accumulates in dynorphin-containing neurons of the NAc after excessive running behavior, suggesting that the induction of 1FosB in this specific subset of neurons in the NAc may be triggered by many types of compulsive behaviors (54). Transgenic mice were generated that exhibit inducible expression of 1FosB primarily in the NAc and dorsal striatum (55). When 1FosB is expressed specifically within the dynorphin-positive neurons in these regions of adult mice, the mice exhibit sensitized behavioral responses to drugs of abuse (56). Inducible expression of 1FosB increases sensitivity to the locomotor activating properties of cocaine (56). The mice also demonstrate enhanced sensitivity to the rewarding effects of cocaine and morphine in place-conditioning assays (51, 56). In addition, they self-administer cocaine at lower doses than their littermate controls that do not overexpress 1FosB, and they maintain self-administration at even lower doses (57). These mice also work harder to self-administer cocaine in progressive ratio self-administration assays, indicating that 1FosB may be involved in sensitizing mice to the motivational effects of cocaine as well, leading to a greater likelihood

5 Jan 2004

11:31

Annu. Rev. Med. 2004.55:113-132. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

120

AR

CHAO

AR206-ME55-07.tex

¥

AR206-ME55-07.sgm

LaTeX2e(2002/01/18)

P1: GBC

NESTLER

Figure 4 Gradual accumulation of 1FosB versus the rapid and transient induction of acute Fos family proteins in brain. Top: Several waves of Fos-like proteins are induced in neurons by acute stimuli (e.g., single drug administration). c-Fos is induced rapidly and degraded within several hours of the acute stimulus, whereas other “acute Fos proteins” [e.g., FosB, 1FosB, and Fos-related antigen (Fra)-1 and -2] are induced somewhat later and persist somewhat longer than c-Fos. Stable isoforms of 1FosB are also induced at low levels following a single acute simulus but persist in brain for long periods. In a complex with Jun-like proteins, these waves of Fos proteins form AP-1 binding complexes with shifting composition over time. Bottom: With repeated (e.g., twice daily) stimulation, for example by repeated drug administration, each acute stimulus induces low levels of stable 1FosB isoforms. This is indicated by the lower set of overlapping lines, which indicate 1FosB induced by each acute stimulus. The result is a gradual increase, indicated by the stepped line in the graph, in the total levels of 1FosB with repeated stimuli during a course of chronic treatment. The increasing levels of 1FosB with repeated stimulation would result in the gradual induction of significant levels of a long-lasting AP-1 complex, which is hypothesized to underlie persisting forms of neural plasticity in the brain. (From Reference 113 with permission.)

5 Jan 2004

11:31

AR

AR206-ME55-07.tex

AR206-ME55-07.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:113-132. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

MOLECULAR NEUROBIOLOGY OF DRUG ADDICTION

P1: GBC

121

of relapse when the drug is withheld (57, 58). 1FosB expression also increases running activity, demonstrating a similar effect on natural rewards (54). If an increase in 1FosB activity heightens sensitivity to the behavioral effects of cocaine and morphine, then manipulations that decrease 1FosB activity should block these effects. Indeed, mice that inducibly express a dominant negative antagonist of 1FosB, 1c-Jun, in the NAc and dorsal striatum show a decrease in cocaine place conditioning, suggesting reduced sensitivity to the rewarding effects of cocaine as expected (59). Together these data indicate that accumulation of 1FosB both enhances drug sensitivity and increases the incentive properties of cocaine. Thus, 1FosB accumulation could amount to a “molecular switch,” whose uniquely stable expression bridges the gap between acute responses to drug exposure and long-term adaptations in the neural and behavioral plasticity of addiction (15). Some of the findings of earlier studies using fosB knockout mice were similar to those of the 1FosB inducible transgenic studies described above (60). For example, the knockout mice do not sensitize to repeated cocaine administration. However, their initial exposure to cocaine results in enhanced behavioral responses, discrepant with findings from the transgenic studies. Interpretation of data from the fosB knockout is complicated by the inability to ascribe these effects specifically to 1FosB, given that the mice lack both gene products, 1FosB and FosB. It is interesting that the results of chronic cocaine administration in the knockout mice, where 1FosB would be expected to have a greater effect, are consistent with those of the 1FosB transgenic studies. In contrast, the discrepant behavioral responses after acute drug administration support the possibility that the immediate and transient induction of FosB expression may play the more dominant role in short-term behavioral responses. Also, in the fosB knockout mice, the fosB gene is ubiquitously absent from the earliest stages of development, so the results from these mice are more complicated to interpret than those from the inducible, regionspecific expression of the 1FosB and 1cJun transgenic mice. A major goal of current research is to identify 1FosB target genes. Using the candidate-gene approach, two target genes have been identified. Transgenic mice that overexpress 1FosB show increased expression of the AMPA (α-amino-3hydroxy-5-methyl-4-isoxazoleproprionic acid) glutamate receptor subunit GluR2 (56), whereas 1cJun expression blocks the ability of chronic cocaine exposure to induce this protein (59). In addition, the promoter region of this gene contains an AP-1 site that binds 1FosB, and overexpression of GluR2 in the NAc via viralmediated gene transfer increases rewarding responses to cocaine, comparable to the result of 1FosB overexpression (56). Another potential target gene of 1FosB is the neuropeptide dynorphin. In contrast to the actions of CREB on dynorphin expression, 1FosB decreases expression of the neuropeptide, which could further contribute to the enhancement of reward sensitivity seen with 1FosB induction (58, 61). Another approach to identifying potential 1FosB target genes has been through the use of DNA microarrays (62, 63; CA McClung, EJ Nestler, unpublished observations). Inducible overexpression of 1FosB regulates the expression of

5 Jan 2004

11:31

Annu. Rev. Med. 2004.55:113-132. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

122

AR

CHAO

AR206-ME55-07.tex

¥

AR206-ME55-07.sgm

LaTeX2e(2002/01/18)

P1: GBC

NESTLER

several genes in the NAc and other regions (62, 63). The transcriptional regulation of these genes by 1FosB requires additional confirmation, and their significance to drug-related plasticity has yet to be elucidated. However, one putative 1FosB target gene identified by DNA microarray analysis is cyclin-dependent kinase 5 (Cdk5) (62, 64). Subsequently found to be induced in the NAc and dorsal striatum by chronic cocaine administration (64), Cdk5 has an AP-1 site in its promoter region, and 1FosB increases promoter activity via this site (62). A possible function of Cdk5 in addiction plasticity is discussed below. Another recent study using DNA microarrays indicates that the expression profile of genes induced by 1FosB in the NAc can account for close to 30% of all the genes regulated by chronic cocaine (CA McClung, EJ Nestler, unpublished observations). Interestingly, short-term 1FosB expression results in gene-expression effects opposite to those seen with long-term 1FosB expression. These differences are reflected in opposing effects on cocaine reward, which is reduced by short-term 1FosB expression and increased by long-term 1FosB expression. This dynamic regulation of gene expression profiles by 1FosB in either a time-dependent or, because of its stability, concentration-dependent manner is unusual, and the relationship of its putative target genes to drug-induced gene expression and addiction behavior warrants further investigation.

Other Transcription Factors Although this review emphasizes CREB and 1FosB, other transcription factors related to addiction have also been studied. For example, NAC-1 is a transcription factor whose mRNA expression is increased in the NAc of rats after chronic cocaine self-administration (65), and it is believed to play a role in some of the behavioral responses to cocaine, including sensitization (66, 67). Levels of the transcription factor NURR1, shown to activate transcription of the dopamine transporter in vitro, are markedly low in midbrain dopamine neurons of human cocaine abusers, which implies that the cocaine-induced decrease in NURR1 mRNA levels may mediate decreased dopamine transporter gene transcription in these neurons after repeated drug exposure (68). Immunoreactivity of another transcription factor, nuclear factor-κB, is induced in the NAc of mice with repeated cocaine exposure, and nuclear factor-κB has been identified as a target gene of 1FosB by use of DNA microarrays (63). As more microarray studies elucidate transcriptional changes in animal models of addiction, it is expected that additional transcription factors, their target genes, and their roles in addiction plasticity will be identified and studied (62, 69; CA McClung, EJ Nestler, unpublished observations).

NEUROTROPHIC MECHANISMS OF ADDICTION Neuronal Morphology As the relationship between a cell’s chemistry and structure becomes increasingly apparent, it is not surprising that chronic exposure to drugs of abuse has been

5 Jan 2004

11:31

AR

AR206-ME55-07.tex

AR206-ME55-07.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:113-132. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

MOLECULAR NEUROBIOLOGY OF DRUG ADDICTION

P1: GBC

123

shown to alter the morphology of neurons in reward circuits of the brain. Chronic morphine administration decreases the size and caliber of VTA dopamine neurons (70), and changes in cytoskeletal proteins and impairment in axoplasmic transport have been observed specific to these neurons (71, 72). It is possible that these morphological alterations reflect a decrease in dopaminergic transmission to the NAc, which may in turn contribute to dysphoria during withdrawal. In addition, chronic morphine has been shown to decrease the complexity of dendritic branching and the number of spines on medium spiny neurons in the NAc and prefrontal cortex in rats (73). In contrast, chronic cocaine or amphetamine increases dendritic branching and spine density in the NAc and prefrontal cortex in rats (73, 74). It is hypothesized that these alterations in limbocortical circuitry may contribute to incentive-motivational effects as well as impaired decision making and judgment (75). Results of studies in which rats self-administered heroin or cocaine were similar to results from rats receiving experimenter-administered drugs (75, 76), indicating that alterations of neuronal morphology occur in both settings. Significantly, these dendritic changes persist for up to one month, leading some to surmise that the structural modifications are important in mediating long-term behavioral changes after chronic drug exposure.

Neurotrophic Factors Although the relationship between drug-induced alterations in dendritic morphology and behavior remains conjectural, studies have implicated neurotrophic factors, which can control neuronal morphology, in aspects of drug addiction. Evidence accumulated during the past decade suggests that neurotrophic factors, whose role in the development of the nervous system is well-characterized, also mediate plasticity in the adult nervous system via their ability to regulate synaptic transmission as well as maintain growth, survival, and differentiation of neurons (77–80). Dopaminergic neurons of the VTA express brain-derived neurotrophic factor (BDNF) and neurotrophin 3 (NT3) mRNA, as well as their receptors, the tropomyosin-related kinase (Trk) receptors TrkB and TrkC, respectively. Medium spiny neurons of the NAc express TrkB and TrkC receptors as well as low levels of BDNF (81). An early study showed that BDNF infusion could reverse certain morphological and biochemical changes seen in VTA dopaminergic neurons following repeated morphine exposure (70, 82). More recent experiments have shown that BDNF and NT3 potently influence behavioral sensitization to cocaine and the regulation of dopaminergic transmission to the NAc (83–85). Further, BDNF has been implicated in mediating cue-induced cocaine craving even after 90 days of withdrawal, possibly through sustained increases in BDNF but not nerve growth factor (NGF) levels within the VTA, NAc, and amygdala (86). Knockout mouse studies show that BDNF is responsible for inducing normal expression of D3 dopamine receptors in the NAc shell and plays an important role in behavioral sensitization (87). This may be relevant to addictive behaviors; infusion of D3 receptor partial agonists has been shown to achieve selective inhibition of cocaine-seeking

5 Jan 2004

11:31

Annu. Rev. Med. 2004.55:113-132. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

124

AR

CHAO

AR206-ME55-07.tex

¥

AR206-ME55-07.sgm

LaTeX2e(2002/01/18)

P1: GBC

NESTLER

behavior (88), and D3 receptor blockade attenuates both the rewarding effects of cocaine (as assayed by conditioned place preference) and cocaine-induced drugseeking behavior (89). Thus, BDNF-dependent induction of D3 receptor gene expression in the NAc may be one mechanism by which the neurotrophic factor can modulate drug-associated behaviors. Finally, conditional BDNF knockout mice also have an attenuated opiate withdrawal syndrome, which implies their involvement in opiate-induced behaviors as well (90). Other neurotrophic factors, such as NT3, ciliary neurotrophic factor, basic fibroblast growth factor, and glial-cell–derived neurotrophic factor (GDNF) also influence psychostimulant-induced behavioral sensitization (see 81 for review). GDNF infusion into the VTA decreases the usual cocaine-induced upregulation of intracellular proteins, and GDNF knockout mice show increased behavioral sensitization to cocaine (91). BDNF and NT3 have each been shown to modulate opiate withdrawal and noradrenergic signaling (90, 92). These and other findings suggest that neurotrophic factors play an important role in regulating psychostimulantor opiate-induced behaviors. Given the function of neurotrophic factors in influencing neuronal morphology, it is tempting to speculate that they may mediate the cocaine- and morphine-induced structural changes described above, but there is as yet no direct evidence that they are responsible for such long-term changes. Studies show that repeated exposure to drugs of abuse alters neurotrophicfactor signaling cascades in neurons of the mesolimbic dopamine system (85, 93–97). Neurotrophic factor binding to the Trk family of receptor tyrosine kinases activates several signal-transduction cascades, including the Ras/ERK protein kinase pathway, the PI3K/Akt kinase pathway, and an isoform of phospholipase C, phospholipase Cγ 1 (PLCγ 1) (98). For example, repeated exposure to morphine increases the expression of PLCγ 1, an activator of the phosphatidylinositol pathway, in the VTA (97). When PLCγ 1 is expressed by viral-mediated gene transfer at levels similar to the upregulation in response to chronic morphine in the VTA, the behavioral response to the rewarding effects of morphine is altered (96). These and other studies in the MAPK cascade and the JAK-STAT pathway indicate that neurotrophic-factor signaling pathways are altered by chronic drug exposure and provide possible mechanisms whereby changes in neurotrophic-factor signaling affect behavioral plasticity. Recent studies identify a molecular substrate that may link chronic drug abuse with dendritic changes in neurons in response to cocaine. Cyclin-dependent kinase 5 (Cdk5), part of a family of serine/threonine cyclin-dependent kinases, is best characterized for regulation of neuronal cytoarchitecture (99). It is known to play a role in neuronal migration, actin dynamics, microtubule stability, synaptic structure, and plasticity. Unlike other Cdk family members, which are regulators of eukaryotic cell-cycle transitions, Cdk5 functions in neurons, which are postmitotic cells. Cdk5 activity is regulated by its interaction with the noncyclin coactivators p35 and p39, and by transcriptional regulation and post-translational events such as phosphorylation. Interestingly, it has been shown that Cdk5 is transcriptionally

5 Jan 2004

11:31

AR

AR206-ME55-07.tex

AR206-ME55-07.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:113-132. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

MOLECULAR NEUROBIOLOGY OF DRUG ADDICTION

P1: GBC

125

regulated by 1FosB (62). cDNA-array analysis of brain regions overexpressing 1FosB in inducible transgenic mice reveals that Cdk5 is consistently upregulated, and promoter analysis identified an AP-1 binding site whereby 1FosB activates Cdk5 transcription in vitro. In addition, either chronic cocaine administration or overexpression of 1FosB increases Cdk5 and p35 expression in the striatum (64). Further, inhibition of Cdk5 activity in the striatum both potentiates the behavioral effects of chronic cocaine and attenuates the cocaine-induced dendritic spine outgrowth in the NAc core and shell (64, 100). Together, these data point toward a scheme whereby chronic cocaine upregulates the uniquely long-lasting transcription factor 1FosB, which increases Cdk5 transcription and activity, which then contributes to the structural modification of dendritic spines in the NAc, which may in turn play a role in the development of behavioral sensitization to cocaine (Figure 5). Neurotrophic factors may also be involved in this pathway, since NGF has been shown to induce p35 expression in PC12 cells via the ERK cascade (101), and BDNF induces Cdk5 kinase activity in primary neuronal cultures (102). This pathway represents at least one potential mechanism whereby drug-induced upregulation of a transcription factor can lead to structural and behavioral changes that are strongly implicated in addiction plasticity. The significance of these schemes

Figure 5 Regulation of dendritic structure by drugs of abuse. The figure shows the expansion of a neuron’s dendritic tree after chronic exposure to a drug of abuse, as has been observed in the NAc and prefrontal cortex for cocaine and related psychostimulants. The areas of magnification show an increase in number of dendritic spines, which is postulated to coincide with activated nerve terminals. Such alterations in dendritic structure, which are similar to those observed in some learning models (e.g., longterm potentiation), could mediate long-lived sensitized responses to drugs of abuse or environmental cues. (From Reference 15 with permission.)

5 Jan 2004

11:31

126

AR

CHAO

AR206-ME55-07.tex

¥

AR206-ME55-07.sgm

LaTeX2e(2002/01/18)

P1: GBC

NESTLER

requires further study, but it is tantalizing to view synaptic, and therefore neuralcircuit, reorganization as a potential mechanism of the long-term effects of drugs on reward, learning, and relapse seen in addiction.

Annu. Rev. Med. 2004.55:113-132. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Neurogenesis Traditionally, the hippocampus is viewed as a critical mediator of declarative memory, but increasing evidence points to its role in the acquisition and maintenance of drug-taking behavior (103–105). Still, the details of its involvement in addiction remain poorly understood. Chronic exposure to drugs of abuse has been shown to decrease the birth of new neurons in the subgranular zone of the adult hippocampus. Both chronic morphine treatment and self-administration of opiates decrease neurogenesis in this region (106). In addition, studies have shown that ethanol inhibits neural progenitor-cell differentiation and survival in the adult rat subgranular zone (107), and self-administration of nicotine decreases neurogenesis in the dentate gyrus in a dose-dependent manner (108). Although acute exposure to psychostimulants has yielded partially contradictory results, studies involving chronic cocaine or amphetamine administration have yet to be done (see 109 for review). A recent study shows that a cannabinoid receptor antagonist given in vivo increases adult neurogenesis in the hippocampus (110). The functional significance of these findings remains unclear, given the current controversy in the field regarding the physiologic importance of adult neurogenesis. Growing evidence supports a correlation between increased neurogenesis and learning and memory (111, 112); further studies are needed to clarify the role of adult neurogenesis in addiction plasticity. ACKNOWLEDGMENT Preparation of this review was supported by grants from the National Institute on Drug Abuse. The Annual Review of Medicine is online at http://med.annualreviews.org

LITERATURE CITED 1. Robinson TE, Berridge KC. 2003. Addiction. Annu. Rev. Psychol. 54:25–53 2. Solomon RL. 1980. The opponentprocess theory of acquired motivation: the costs of pleasure and the benefits of pain. Am. Psychol. 35:691–712 3. Koob GF, Le Moal M. 2001. Drug addiction, dysregulation of reward, and allostasis. Neuropsychopharmacology 24: 97–129 4. Robinson TE, Berridge KC. 1993. The

neural basis of drug craving: an incentivesensitization theory of addiction. Brain Res. Rev. 18:247–91 5. Robinson TE, Berridge KC. 2000. The psychology and neurobiology of addiction: an incentive-sensitization view. Addiction 95:S91–117 6. Berke JD, Hyman SE. 2000. Addiction, dopamine, and the molecular mechanisms of memory. Neuron 25:515–32 7. Di Chiara G. 1999. Drug addiction as

5 Jan 2004

11:31

AR

AR206-ME55-07.tex

AR206-ME55-07.sgm

LaTeX2e(2002/01/18)

MOLECULAR NEUROBIOLOGY OF DRUG ADDICTION

8.

Annu. Rev. Med. 2004.55:113-132. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

dopamine-dependent associative learning disorder. Eur. J. Pharmacol. 375:13– 30 Everitt BJ, Dickinson A, Robbins TW. 2001. The neuropsychological basis of addictive behaviour. Brain Res. Brain Res. Rev. 36:129–38 Lonze BE, Ginty DD. 2002. Function and regulation of CREB family transcription factors in the nervous system. Neuron 35:605–23 Mayr B, Montminy M. 2001. Transcriptional regulation by the phosphorylationdependent factor CREB. Nat. Rev. Mol. Cell Biol. 2:599–609 Chan HM, La Thangue NB. 2001. p300/CBP proteins: HATs for transcriptional bridges and scaffolds. J. Cell Sci. 114:2363–73 Guitart X, Thompson MA, Mirante CK, et al. 1992. Regulation of cyclic AMP response element-binding protein (CREB) phosphorylation by acute and chronic morphine in the rat locus coeruleus. J. Neurochem. 58:1168–71 Sharma SK, Klee WA, Nirenberg M. 1975. Dual regulation of adenylate cyclase accounts for narcotic dependence and tolerance. Proc. Natl. Acad. Sci. USA 72:3092–96 Nestler EJ, Aghajanian GK. 1997. Molecular and cellular basis of addiction. Science 278:58–63 Nestler EJ. 2001. Molecular basis of longterm plasticity underlying addiction. Nat. Rev. Neurosci. 2:119–28 Foote SL, Bloom FE, Aston-Jones G. 1983. Nucleus locus ceruleus: new evidence of anatomical and physiological specificity. Physiol. Rev. 63:844–914 Aston-Jones G, Rajkowski J, Cohen J. 1999. Role of locus coeruleus in attention and behavioral flexibility. Biol. Psychiatry 46:1309–20 Lane-Ladd SB, Pineda J, Boundy VA, et al. 1997. CREB (cAMP response element-binding protein) in the locus coeruleus: biochemical, physiological,

19.

20.

21.

22.

23.

24.

25.

26.

27.

P1: GBC

127

and behavioral evidence for a role in opiate dependence. J. Neurosci. 17:7890–901 Widnell KL, Russell DS, Nestler EJ. 1994. Regulation of expression of cAMP response element-binding protein in the locus coeruleus in vivo and in a locus coeruleus–like cell line in vitro. Proc. Natl. Acad. Sci. USA 91:10947–51 Chao JR, Ni YG, Bolanos CA, et al. 2002. Characterization of the mouse adenylyl cyclase type VIII gene promoter: regulation by cAMP and CREB. Eur. J. Neurosci. 16:1284–94 Maldonado R, Blendy JA, Tzavara E, et al. 1996. Reduction of morphine abstinence in mice with a mutation in the gene encoding CREB. Science 273:657–59 Walters CL, Blendy JA. 2001. Different requirements for cAMP response element binding protein in positive and negative reinforcing properties of drugs of abuse. J. Neurosci. 21:9438–44 Terwilliger RZ, Beitner-Johnson D, Sevarino KA, et al. 1991. A general role for adaptations in G-proteins and the cyclic AMP system in mediating the chronic actions of morphine and cocaine on neuronal function. Brain Res. 548:100–10 Unterwald EM, Cox BM, Kreek MJ, et al. 1993. Chronic repeated cocaine administration alters basal and opioid-regulated adenylyl cyclase activity. Synapse 15:33– 38 Barrot M, Olivier JD, Perrotti LI, et al. 2002. CREB activity in the nucleus accumbens shell controls gating of behavioral responses to emotional stimuli. Proc. Natl. Acad. Sci. USA 99:11435–40 Shaw-Lutchman TZ, Barrot M, Wallace T, et al. 2002. Regional and cellular mapping of cAMP response element– mediated transcription during naltrexoneprecipitated morphine withdrawal. J. Neurosci. 22:3663–72 Shaw-Lutchman TZ, Impey S, Storm D, Nestler EJ. 2003. Regulation of CREmediated transcription in mouse brain by amphetamine. Synapse 48:10–17

5 Jan 2004

11:31

Annu. Rev. Med. 2004.55:113-132. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

128

AR

CHAO

AR206-ME55-07.tex

¥

AR206-ME55-07.sgm

LaTeX2e(2002/01/18)

P1: GBC

NESTLER

28. Self DW, Genova LM, Hope BT, et al. 1998. Involvement of cAMP-dependent protein kinase in the nucleus accumbens in cocaine self-administration and relapse of cocaine-seeking behavior. J. Neurosci. 18:1848–59 29. Carlezon WA Jr, Thome J, Olson VG, et al. 1998. Regulation of cocaine reward by CREB. Science 282:2272–75 30. Sakai N, Thome J, Newton SS, et al. 2002. Inducible and brain region–specific CREB transgenic mice. Mol. Pharmacol. 61:1453–64 31. Deleted in proof 32. Hyman SE, Malenka RC, Spangler R, et al. 2001. Addiction and the brain: the neurobiology of compulsion and its persistence. Nat. Rev. Neurosci. 2:695–703 33. Pliakas AM, Carlson RR, Neve RL, et al. 2001. Altered responsiveness to cocaine and increased immobility in the forced swim test associated with elevated cAMP response element–binding protein expression in nucleus accumbens. J. Neurosci. 21:7397–403 33a. Newton SS, Thome J, Wallace T, et al. 2002. Inhibition of CREB or dynorphin in the nucleus accumbens produces an antidepressant-like effect. J. Neurosci. 22:10883–90 34. Daunais JB, McGinty JF. 1995. Cocaine binges differentially alter striatal preprodynorphin and zif/268 mRNAs. Brain Res. Mol. Brain Res. 29:201–10 35. Spangler R, Ho A, Zhou Y, et al. 1996. Regulation of kappa opioid receptor mRNA in the rat brain by “binge” pattern cocaine administration and correlation with preprodynorphin mRNA. Brain Res. Mol. Brain Res. 38:71–76 36. Cole RL, Konradi C, Douglass J, Hyman SE. 1995. Neuronal adaptation to amphetamine and dopamine: molecular mechanisms of prodynorphin gene regulation in rat striatum. Neuron 14:813–23 37. Shippenberg TS, Rea W. 1997. Sensitization to the behavioral effects of cocaine: modulation by dynorphin and

38.

39.

40.

41.

42.

43.

44.

45.

46.

kappa-opioid receptor agonists. Pharmacol. Biochem. Behav. 57:449–55 Spanagel R, Herz A, Shippenberg TS, et al. 1992. Opposing tonically active endogenous opioid systems modulate the mesolimbic dopaminergic pathway. Proc. Natl. Acad. Sci. USA 89:2046–50 Mumberg D, Lucibello FC, Schuermann M, Muller R. 1991. Alternative splicing of fosB transcripts results in differentially expressed mRNAs encoding functionally antagonistic proteins. Genes Dev. 5:1212– 23 Hope BT, Nye HE, Kelz MB, et al. 1994. Induction of a long-lasting AP-1 complex composed of altered Fos-like proteins in brain by chronic cocaine and other chronic treatments. Neuron 13:1235–44 Moratalla R, Elibol B, Vallejo M, et al. 1996. Network-level changes in expression of inducible Fos-Jun proteins in the striatum during chronic cocaine treatment and withdrawal. Neuron 17:147–56 Young ST, Porrino LJ, Iadarola MJ. 1991. Cocaine induces striatal c-fos– immunoreactive proteins via dopaminergic D1 receptors. Proc. Natl. Acad. Sci. USA 88:1291–95 Graybiel AM, Moratalla R, Robertson HA. 1990. Amphetamine and cocaine induce drug-specific activation of the cfos gene in striosome-matrix compartments and limbic subdivisions of the striatum. Proc. Natl. Acad. Sci. USA 87:6912– 16 Hope B, Kosofsky B, Hyman SE, Nestler EJ. 1992. Regulation of immediate early gene expression and AP-1 binding in the rat nucleus accumbens by chronic cocaine. Proc. Natl. Acad. Sci. USA 89:5764–68 Chen J, Kelz MB, Hope BT, et al. 1997. Chronic Fos-related antigens: stable variants of 1FosB induced in brain by chronic treatments. J. Neurosci. 17:4933–41 Andersson M, Westin JE, Cenci MA, et al. 2003. Time course of striatal 1FosBlike immunoreactivity and prodynorphin

5 Jan 2004

11:31

AR

AR206-ME55-07.tex

AR206-ME55-07.sgm

LaTeX2e(2002/01/18)

MOLECULAR NEUROBIOLOGY OF DRUG ADDICTION

47.

Annu. Rev. Med. 2004.55:113-132. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

48.

49.

50.

51.

52.

53.

54.

55.

56.

57.

mRNA levels after discontinuation of chronic dopaminomimetic treatment. Eur. J. Neurosci. 17:661–66 Nye HE, Hope BT, Kelz MB, et al. 1995. Pharmacological studies of the regulation of chronic FOS-related antigen induction by cocaine in the striatum and nucleus accumbens. J. Pharmacol. Exp. Ther. 275:1671–80 Nye HE, Nestler EJ. 1996. Induction of chronic Fos-related antigens in rat brain by chronic morphine administration. Mol. Pharmacol. 49:636–45 Pich EM, Pagliusi SR, Tessari M, et al. 1997. Common neural substrates for the addictive properties of nicotine and cocaine. Science 275:83–86 Ryabinin AE, Wang YM. 1998. Repeated alcohol administration differentially affects c-Fos and FosB protein immunoreactivity in DBA/2J mice. Alcohol Clin. Exp. Res. 22:1646–54 Kelz MB, Nestler EJ. 2000. 1FosB: a molecular switch underlying long-term neural plasticity. Curr. Opin. Neurol. 13: 715–20 Atkins JB, Chlan-Fourney J, Nye HE, et al. 1999. Region-specific induction of deltaFosB by repeated administration of typical versus atypical antipsychotic drugs. Synapse 33:118–28 Hiroi N, Graybiel AM. 1996. Atypical and typical neuroleptic treatments induce distinct programs of transcription factor expression in the striatum. J. Comp. Neurol. 374:70–83 Werme M, Messer C, Olson L, et al. 2002. 1FosB regulates wheel running. J. Neurosci. 22:8133–38 Chen J, Kelz MB, Zeng G, et al. 1998. Transgenic animals with inducible, targeted gene expression in brain. Mol. Pharmacol. 54:495–503 Kelz MB, Chen J, Carlezon WA Jr, et al. 1999. Expression of the transcription factor 1FosB in the brain controls sensitivity to cocaine. Nature 401:272–76 Colby CR, Whisler K, Steffen C, et al.

58.

59.

60.

61.

62.

63.

64.

65.

66.

67.

P1: GBC

129

2003. Striatal cell type–specific overexpression of 1FosB enhances incentive for cocaine. J. Neurosci. 23:2488–93 Nestler EJ, Barrot M, Self DW. 2001. 1FosB: a sustained molecular switch for addiction. Proc. Natl. Acad. Sci. USA 98:11042–46 Peakman MC, Colby C, Perrotti LI, et al. 2003. Inducible, brain region–specific expression of a dominant negative mutant of c-Jun in transgenic mice decreases sensitivity to cocaine. Brain Res. 970:73–86 Hiroi N, Brown JR, Haile CN, et al. 1997. FosB mutant mice: loss of chronic cocaine induction of Fos-related proteins and heightened sensitivity to cocaine’s psychomotor and rewarding effects. Proc. Natl. Acad. Sci. USA 94:10397–402 Shaw TZ, Gilden L, Kelz MB, et al. 2000. Regulation of neuropeptide expression by 1FosB in striatum. Society for Neuroscience Abstracts 26:525 Chen J, Zhang Y, Kelz MB, et al. 2000. Induction of cyclin-dependent kinase 5 in the hippocampus by chronic electroconvulsive seizures: role of 1FosB. J. Neurosci. 20:8965–71 Ang E, Chen J, Zagouras P, et al. 2001. Induction of nuclear factor–κB in nucleus accumbens by chronic cocaine administration. J. Neurochem. 79:221–24 Bibb JA, Chen J, Taylor JR, et al. 2001. Effects of chronic exposure to cocaine are regulated by the neuronal protein Cdk5. Nature 410:376–80 Cha XY, Pierce RC, Kalivas PW, Mackler SA. 1997. NAC-1, a rat brain mRNA, is increased in the nucleus accumbens three weeks after chronic cocaine selfadministration. J. Neurosci. 17:6864–71 Kalivas PW, Duffy P, Mackler SA. 1999. Interrupted expression of NAC-1 augments the behavioral responses to cocaine. Synapse 33:153–59 Mackler SA, Korutla L, Cha XY, et al. 2000. NAC-1 is a brain POZ/BTB protein that can prevent cocaine-induced sensitization in the rat. J. Neurosci. 20:6210–17

5 Jan 2004

11:31

Annu. Rev. Med. 2004.55:113-132. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

130

AR

CHAO

AR206-ME55-07.tex

¥

AR206-ME55-07.sgm

LaTeX2e(2002/01/18)

P1: GBC

NESTLER

68. Bannon MJ, Pruetz B, Manning-Bog AB, et al. 2002. Decreased expression of the transcription factor NURR1 in dopamine neurons of cocaine abusers. Proc. Natl. Acad. Sci. USA 99:6382–85 69. Toda S, McGinty JF, Kalivas PW. 2002. Repeated cocaine administration alters the expression of genes in corticolimbic circuitry after a 3-week withdrawal: a DNA microarray study. J. Neurochem. 82:1290–99 70. Sklair-Tavron L, Shi WX, Lane SB, et al. 1996. Chronic morphine induces visible changes in the morphology of mesolimbic dopamine neurons. Proc. Natl. Acad. Sci. USA 93:11202–7 71. Beitner-Johnson D, Nestler EJ. 1993. Chronic morphine impairs axoplasmic transport in the rat mesolimbic dopamine system. Neuroreport 5:57–60 72. Beitner-Johnson D, Guitart X, Nestler EJ. 1992. Neurofilament proteins and the mesolimbic dopamine system: common regulation by chronic morphine and chronic cocaine in the rat ventral tegmental area. J. Neurosci. 12:2165–76 73. Robinson TE, Kolb B. 1999. Alterations in the morphology of dendrites and dendritic spines in the nucleus accumbens and prefrontal cortex following repeated treatment with amphetamine or cocaine. Eur. J. Neurosci. 11:1598–604 74. Robinson TE, Kolb B. 1997. Persistent structural modifications in nucleus accumbens and prefrontal cortex neurons produced by previous experience with amphetamine. J. Neurosci. 17:8491– 97 75. Robinson TE, Gorny G, Mitton E, Kolb B. 2001. Cocaine self-administration alters the morphology of dendrites and dendritic spines in the nucleus accumbens and neocortex. Synapse 39:257–66 76. Self DW, McClenahan AW, BeitnerJohnson D, et al. 1995. Biochemical adaptations in the mesolimbic dopamine system in response to heroin selfadministration. Synapse 21:312–18

77. Barde YA. 1989. Trophic factors and neuronal survival. Neuron 2:1525–34 78. Lu B, Figurov A. 1997. Role of neurotrophins in synapse development and plasticity. Rev. Neurosci. 8:1–12 79. Korsching S, Turgeon SM, Pollack AE, Fink JS. 1993. The neurotrophic factor concept: a reexamination. J. Neurosci. 13:2739–48 80. Thoenen H. 1995. Neurotrophins and neuronal plasticity. Science 270:593–98 81. Pierce RC, Bari AA. 2001. The role of neurotrophic factors in psychostimulantinduced behavioral and neuronal plasticity. Rev. Neurosci. 12:95–110 82. Nestler EJ, Berhow MT, Brodkin ES. 1996. Molecular mechanisms of drug addiction: adaptations in signal transduction pathways. Mol. Psychiatry 1:190–99 83. Horger BA, Iyasere CA, Berhow MT, et al. 1999. Enhancement of locomotor activity and conditioned reward to cocaine by brain-derived neurotrophic factor. J. Neurosci. 19:4110–22 84. Martin-Iverson MT, Todd KG, Altar CA. 1994. Brain-derived neurotrophic factor and neurotrophin-3 activate striatal dopamine and serotonin metabolism and related behaviors: interactions with amphetamine. J. Neurosci. 14:1262–70 85. Pierce RC, Pierce-Bancroft AF, Prasad BM. 1999. Neurotrophin-3 contributes to the initiation of behavioral sensitization to cocaine by activating the Ras/mitogenactivated protein kinase signal transduction cascade. J. Neurosci. 19:8685–95 86. Grimm JW, Lu L, Hayashi T, et al. 2003. Time-dependent increases in brainderived neurotrophic factor protein levels within the mesolimbic dopamine system after withdrawal from cocaine: implications for incubation of cocaine craving. J. Neurosci. 23:742–47 87. Guillin O, Diaz J, Carroll P, et al. 2001. BDNF controls dopamine D3 receptor expression and triggers behavioural sensitization. Nature 411:86–89 88. Pilla M, Perachon S, Sautel F, et al.

5 Jan 2004

11:31

AR

AR206-ME55-07.tex

AR206-ME55-07.sgm

LaTeX2e(2002/01/18)

MOLECULAR NEUROBIOLOGY OF DRUG ADDICTION

89.

Annu. Rev. Med. 2004.55:113-132. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

90.

91.

92.

93.

94.

95.

96.

97.

1999. Selective inhibition of cocaineseeking behaviour by a partial dopamine D3 receptor agonist. Nature 400:371– 75 Vorel SR, Ashby CR Jr, Paul M, et al. 2002. Dopamine D3 receptor antagonism inhibits cocaine-seeking and cocaineenhanced brain reward in rats. J. Neurosci. 22:9595–603 Akbarian S, Rios M, Liu RJ, et al. 2002. Brain-derived neurotrophic factor is essential for opiate-induced plasticity of noradrenergic neurons. J. Neurosci. 22:4153–62 Messer CJ, Eisch AJ, Carlezon WA Jr, et al. 2000. Role for GDNF in biochemical and behavioral adaptations to drugs of abuse. Neuron 26:247–57 Akbarian S, Bates B, Liu RJ, et al. 2001. Neurotrophin-3 modulates noradrenergic neuron function and opiate withdrawal. Mol. Psychiatry 6:593–604 Berhow MT, Hiroi N, Kobierski LA, et al. 1996. Influence of cocaine on the JAK-STAT pathway in the mesolimbic dopamine system. J. Neurosci. 16:8019– 26 Berhow MT, Hiroi N, Nestler EJ. 1996. Regulation of ERK (extracellular signal regulated kinase), part of the neurotrophin signal transduction cascade, in the rat mesolimbic dopamine system by chronic exposure to morphine or cocaine. J. Neurosci. 16:4707–15 Berhow MT, Russell DS, Terwilliger RZ, et al. 1995. Influence of neurotrophic factors on morphine- and cocaine-induced biochemical changes in the mesolimbic dopamine system. Neuroscience 68:969– 79 Bolanos CA, Perrotti LI, Edwards S, et al. 2003. Phospholipase Cγ in distinct regions of the ventral tegmental area differentially modulates mood-related behaviors. J. Neurosci. 23:7569–76 Wolf DH, Numan S, Nestler EJ, Russell DS. 1999. Regulation of phospholipase Cγ in the mesolimbic dopamine system

98.

99.

100.

101.

102.

103.

104.

105.

106.

107.

108.

P1: GBC

131

by chronic morphine administration. J. Neurochem. 73:1520–28 Patapoutian A, Reichardt LF, Berhow MT, Brodkin ES. 2001. Trk receptors: mediators of neurotrophin action. Curr. Opin. Neurobiol. 11:272–80 Dhavan R, Tsai LH, Berhow MT, Brodkin ES. 2001. A decade of CDK5. Nat. Rev. Mol. Cell Biol. 2:749–59 Norrholm SD, Bibb JA, Nestler EJ, et al. 2003. Cocaine-induced proliferation of dendritic spines in nucleus accumbens is dependent on the activity of cyclin-dependent kinase-5. Neuroscience 116:19–22 Harada T, Morooka T, Ogawa S, Nishida E. 2001. ERK induces p35, a neuronspecific activator of Cdk5, through induction of Egr1. Nat. Cell Biol. 3:453–59 Tokuoka H, Saito T, Yorifuji H, et al. 2000. Brain-derived neurotrophic factor– induced phosphorylation of neurofilament-H subunit in primary cultures of embryo rat cortical neurons. J. Cell Sci. 113:1059–68 White NM. 1996. Addictive drugs as reinforcers: multiple partial actions on memory systems. Addiction 91:921–49; discussion 51–65 Koob GF, Le Moal M. 2001. Drug addiction, dysregulation of reward, and allostasis. Neuropsychopharmacology 24:97– 129 Vorel SR, Liu X, Hayes RJ, et al. 2001. Relapse to cocaine-seeking after hippocampal theta burst stimulation. Science 292:1175–78 Eisch AJ, Barrot M, Schad CA, et al. 2000. Opiates inhibit neurogenesis in the adult rat hippocampus. Proc. Natl. Acad. Sci. USA 97:7579–84 Crews FT, Miller MW, Ma W, et al. 2003. Neural stem cells and alcohol. Alcohol Clin. Exp. Res. 27:324–35 Abrous DN, Adriani W, Montaron MF, et al. 2002. Nicotine self-administration impairs hippocampal plasticity. J. Neurosci. 22:3656–62

5 Jan 2004

11:31

132

AR

CHAO

AR206-ME55-07.tex

¥

AR206-ME55-07.sgm

P1: GBC

NESTLER

109. Eisch AJ. 2002. Adult neurogenesis: implications for psychiatry. Prog. Brain Res. 138:315–42 110. Rueda D, Navarro B, Martinez-Serrano A, et al. 2002. The endocannabinoid anandamide inhibits neuronal progenitor cell differentiation through attenuation of the Rap1/B–Raf/ERK pathway. J. Biol. Chem. 277:46645–50 111. Gould E, Gross CG. 2002. NeurogeneAnnu. Rev. Med. 2004.55:113-132. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

LaTeX2e(2002/01/18)

sis in adult mammals: some progress and problems. J. Neurosci. 22:619–23 112. Shors TJ, Townsend DA, Zhao M, et al. 2002. Neurogenesis may relate to some but not all types of hippocampal-dependent learning. Hippocampus 12: 578–84 113. Nestler EJ. 2001. Molecular neurobiology of addiction. Am. J. Addiction 201:201– 217

CHAO.qxd

1/5/2004

3:31 PM

Page 1

Annu. Rev. Med. 2004.55:113-132. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

MOLECULAR NEUROBIOLOGY OF DRUG ADDICTION

C-1

Figure 1 Key neural circuits of addiction as visualized in a mid-sagittal diagrammatic representation of the rat brain. Dotted lines indicate limbic afferents to the nucleus accumbens (NAc). Arrows represent efferents from the NAc thought to be involved in drug reward. Dopamine pathways indicate projections of the mesolimbic dopamine system thought to be a critical substrate for drug reward. This system originates in the ventral tegmental area (VTA) and projects to the NAc and other limbic structures, including olfactory tubercle (OT), ventral domains of the caudate-putamen (C-P), amygdala (AMG), and prefrontal cortex (PFC). Opioid peptide pathways indicate opioid peptidecontaining neurons, which are involved in opiate, ethanol, and possibly nicotine reward. These opioid peptide systems include the local enkephalinergic circuits (short segments) and the hypothalamic midbrain beta-endorphin circuit (long segment). Blue areas indicate the hypothesized distribution of GABAA receptor complexes, which may contribute to ethanol reward. Nicotinic acetylcholine receptors are hypothesized to be located on dopaminergic and opioid peptidergic systems. ARC, arcuate nucleus; Cer, cerebellum; DMT, dorsomedial thalamus; IC, inferior colliculus; LC, locus coeruleus; LH, lateral hypothalamus; PAG, periaqueductal gray; SC, superior colliculus; SNr, substantia nigra pars reticulata; VP, ventral pallidum. (From Reference 15 with permission.)

Annu. Rev. Med. 2004.55:113-132. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

CHAO.qxd 1/5/2004

C-2

3:31 PM

CHAO ■

See legend on next page

Page 2

NESTLER

CHAO.qxd

1/5/2004

3:31 PM

Page 3

Annu. Rev. Med. 2004.55:113-132. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

MOLECULAR NEUROBIOLOGY OF DRUG ADDICTION

C-3

Figure 2 Opiates acutely inhibit neurons in the locus coeruleus (LC) by increasing the conductance of an inwardly rectifying K+ channel via coupling with subtypes of Gi/o and by decreasing a Na+-dependent inward current via coupling with Gi/o and the consequent inhibition of adenylyl cyclase. Reduced levels of cAMP decrease protein kinase A (PKA) activity and the phosphorylation of the responsible channel or pump. Inhibition of the cAMP pathway also decreases phosphorylation of numerous other proteins and thereby affects many additional processes in the neuron. For example, it reduces the phosphorylation state of CREB, which may initiate some of the longer-term changes in LC function. Red arrows summarize effects of chronic morphine in the LC. Chronic morphine increases levels of types I (ACI) and VIII (ACVIII) adenylyl cyclase, PKA catalytic (C) and regulatory type II (RII) subunits, and several phosphoproteins, including CREB and tyrosine hydroxylase (TH), the rate-limiting enzyme in norepinephrine biosynthesis. These changes contribute to the altered phenotype of the drug-addicted state. For example, the intrinsic excitability of LC neurons is increased via enhanced activity of the cAMP pathway and Na+-dependent inward current, which contributes to the tolerance, dependence, and withdrawal exhibited by these neurons. Upregulation of ACVIII and TH is mediated via CREB, whereas upregulation of ACI and the PKA subunits appears to occur via a CREB-independent mechanism not yet identified. (From Reference 113 with permission.)

2 Dec 2003

22:9

AR

AR206-ME55-08.tex

AR206-ME55-08.sgm

LaTeX2e(2002/01/18) P1: GCE 10.1146/annurev.med.55.091902.103539

Annu. Rev. Med. 2004. 55:133–56 doi: 10.1146/annurev.med.55.091902.103539 c 2004 by Annual Reviews. All rights reserved Copyright °

BETA-CELL REPLACEMENT FOR TYPE I DIABETES Peter G. Stock1 and Jeffrey A. Bluestone2

Annu. Rev. Med. 2004.55:133-156. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

1

Department of Surgery, Division of Transplantation, University of California, San Francisco, San Francisco, California 94143; 2Diabetes Center, Department of Medicine, Pathology, Microbiology and Immunology, University of California, San Francisco, San Francisco, California 94143; email: [email protected], [email protected]

Key Words pancreas, islet, transplantation, tolerance, stem cell ■ Abstract The ability to achieve insulin independence with either solid-organ pancreas or islet transplantation has increased the number of patients seeking beta-cell replacement as an alternative to insulin therapy. Despite dramatic improvements in the ability to achieve insulin independence following solid-organ pancreas transplantation, the secondary complications of long-standing diabetes are frequently irreversible by the time surgical intervention is justified based on the risk of this procedure. Pancreatic islet transplantation provides a safer and less invasive alternative for beta-cell replacement that could be justified earlier in the course of diabetes to prevent the development of secondary complications. Recent advances in the technology of islet isolation, as well as the ability to prevent the alloimmune and recurrent autoimmune response following islet transplantation with immunosuppressive regimens that are not toxic to beta cells, have rekindled an interest in this field. Widespread application of islet transplantation will depend on further improvements in selective immunosuppression, development of immunologic tolerance, and finding new sources of beta cells.

INTRODUCTION Dramatic improvements in the success of the transplantation of pancreatic tissue, either the whole organ or pancreatic islets, have sparked a renewed interest in transplantation as a treatment for diabetes mellitus. Pancreas transplantation has become a widely accepted treatment for type I diabetic patients who have undergone a previous or simultaneous kidney transplant. The success rate (success defined as normoglycemia and insulin independence) is currently >80% at 3 years (1, 2). However, solid-organ pancreas transplantation in the preuremic recipient is not widely accepted mainly because of the associated surgical complications and the need for vigorous immunosuppression, both of which contribute considerably to the overall morbidity and costs of this procedure (3). The secondary complications of long-standing diabetes are frequently irreversible by the time surgical intervention is justified based on the risks of the procedure, underscoring the 0066-4219/04/0218-0133$14.00

133

2 Dec 2003

22:9

Annu. Rev. Med. 2004.55:133-156. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

134

AR

STOCK

AR206-ME55-08.tex

¥

AR206-ME55-08.sgm

LaTeX2e(2002/01/18)

P1: GCE

BLUESTONE

necessity of a safer and less invasive procedure for beta-cell replacement that could be justified prior to the development of the secondary complications. The efficacy of pancreatic islet transplantation, a significantly less invasive procedure, was only recently demonstrated by the achievement of insulin independence in seven consecutive type I diabetic recipients after percutaneous portal-vein transplantation of pancreatic islets (4). As a result of the significant improvements attained in the “Edmonton protocol,” multiple centers across the United States and worldwide are developing programs to transplant pancreatic islets as an alternative method of beta-cell replacement. This review addresses the development of the current protocols for islet transplantation, as well as future strategies for extending the application of this new technology.

BACKGROUND: THE EVOLUTION OF BETA-CELL REPLACEMENT The Problem The incidence of diabetes mellitus is predicted to increase significantly in the next decade, and it already affects an estimated 130 million people worldwide. It affects 16 million Americans and consumes one out of every eight health care dollars. Despite the efficacy of insulin therapy, the devastating secondary complications, including nephropathy, neuropathy, retinopathy, and cardiovascular disease can shorten life expectancy by as much as one third. The Diabetes Control and Complications Trial demonstrated that tight regulation of blood sugars with intensive insulin therapy significantly lowered the level of the glycosylated hemoglobin (HbA1C) and minimized the progression of the secondary complications. Nonetheless, even intensive therapy did not abrogate the development of secondary complications, and tight control resulted in a significantly higher risk of severe hypoglycemic reactions leading to seizure or coma (5, 6).

Solid-Organ Pancreas Transplantation Solid-organ pancreas transplantation has undergone significant progress in the past decade. It has been the most consistent method of beta-cell replacement, resulting in sustained euglycemia, insulin independence, and normalization of HbA1C. The most important advances have been in preventing rejection of this highly immunogenic transplant. The addition of mycophenolate mofetil and tacrolimus to immunosuppressive regimens decreased the incidence of rejection following pancreas transplantation from 80% to <20% at most centers performing this procedure. Antibody induction therapy with antilymphocyte preparations, along with the addition of sirolimus, an immunosuppressive agent that lacks nephrotoxicity or beta-cell toxicity, has permitted steroid-free maintenance therapy following solid-organ pancreas transplantation with continued low rejection rates. Despite the elimination of steroids from maintenance regimens, the incidence of rejection

2 Dec 2003

22:9

AR

AR206-ME55-08.tex

AR206-ME55-08.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:133-156. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

BETA-CELL REPLACEMENT

P1: GCE

135

of this highly immunogenic organ is currently reported to be <10% (7, 8). The dramatic improvement in the ability to prevent rejection of pancreas transplants is reflected in the results reported by the International Pancreas Transplant Registry (9). The three-year graft survival, as defined by insulin independence, is ∼80% for pancreas transplants performed simultaneously with a kidney transplant as well as for pancreas transplants performed after a successful kidney transplant. Progression of the secondary complications of diabetes is curtailed by the presence of a functioning pancreas transplant, and the development of diabetic nephropathy is prevented in the simultaneously transplanted kidney (10). Similarly, there are gradual improvements in neuropathy (11) and overall improvements in the quality of life (12). These highly significant improvements in success have been the impetus to proceed with solitary pancreas transplantation in the preuremic patient with lifethreatening diabetes. Although pancreas allografts in the preuremic diabetic recipient are more vulnerable to rejection (rejection loss at one year equals 8% for these patients versus 2% for those who had simultaneous pancreas and kidney transplants), the improved results have increased the frequency of this procedure (2). Nonetheless, because of the invasive nature of the surgery and long-term complications of the rigorous immunosuppressive regimens, this procedure remains limited to patients with hypoglycemic unawareness or metabolic lability despite intensive insulin therapy. The transplantation of pancreatic islets provides an attractive and less invasive alternative to solid-organ transplantation. In addition to avoiding the technical complications of solid-organ transplantation, related to thrombosis of the blood supply to the whole-organ allograft and the danger of activation of the digestive enzymes associated with the exocrine function, pancreatic islet transplantation provides the opportunity to manipulate the islets prior to transplantation in order to decrease immunogenicity of the allograft. Of equal significance, this technique has the potential to provide an unlimited source of beta cells that would ultimately be independent of the limited donor pool. The proliferation of beta cells from either embryonic or adult stem cells is most promising, although xenogeneic islets may prove to be an attractive alternative source pending further refinement in immunosuppressive protocols and resolution of concerns related to endogenous animal viruses.

The History of Islet Transplantation The first successful transplantation of pancreatic islets in experimental animal models was performed in 1972 (13, 14) and the first clinical islet allograft in a diabetic recipient of a previous kidney transplant in 1974 (15–17), but pancreatic islet transplants did not immediately emerge as a successful alternative to solidorgan transplantation. Despite enthusiasm for this procedure and promises that islet transplantation was just around the corner, the success of islet transplantation as defined by insulin independence was minimal. Over the 25 years following the first clinical attempt at islet transplantation, over 300 alloislet transplants in type I

2 Dec 2003

22:9

Annu. Rev. Med. 2004.55:133-156. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

136

AR

STOCK

AR206-ME55-08.tex

¥

AR206-ME55-08.sgm

LaTeX2e(2002/01/18)

P1: GCE

BLUESTONE

diabetic patients were performed and described in a comprehensive report by Hering & Ricordi (18). Although many of the islet allografts had some function as evidenced by C-peptide, few of the transplants provided even transient states of insulin independence despite the use of potent immunosuppressive drugs. This comprehensive report also reviewed over 3000 fetal grafts, neonatal grafts, and xenografts performed worldwide during the same time period and reported to the International Islet Registry; 96% of these reports came from China and Eastern Europe. Although there were occasional reports of insulin independence (18, 19), the vast majority of these cases failed to demonstrate insulin independence or long-term engraftment as defined by presence of C-peptide. Disappointment in islet transplantation based on the low frequency of insulin independence as compared to solid-organ transplantation was tempered by the success of islet autotransplantation to prevent diabetes following total pancreatectomy for chronic pancreatitis (20). The long-term efficacy of islet autotransplants has been reported by the transplant group at the University of Minnesota, who had successfully performed this procedure since the 1970s. The intraportal infusion of pancreatic islets isolated from nondiabetic patients suffering from pancreatitis has produced states of insulin independence for as long as 13 years following total pancreatectomy. This finding demonstrates that islet transplantation can be a viable alternative to solid-organ transplantation (21). Successful states of insulin independence required the successful isolation of a significant number of islets following pancreatectomy, which underscores the necessity of a sufficient mass of islets to achieve insulin independence. Achieving insulin independence following allogeneic islet transplantation into a type I diabetic recipient is further complicated by the necessity for preventing both alloimmune and autoimmune destruction. The historic success of alloislet transplantation for nondiabetic patients with diabetes precipitated by surgical removal of the pancreas may imply the existence of an additional barrier to autoimmune disease directed against beta cells, which presumably recurs following islet transplantation. In the largest series, reported from the University of Pittsburgh (22, 23), simultaneous islet-liver allotransplantation after upper abdominal exenteration including total pancreatectomy resulted in insulin independence in seven patients. Several patients were able to sustain insulin independence until death secondary to malignancy, including one patient who was insulin-independent for five years until his death from malignancy. The recurrence of malignant disease following this procedure has prevented the continued application of this strategy. Nonetheless, the success of islet transplantation in this setting relative to islet allotransplants performed into type I diabetic recipients during the same era provided indirect evidence of the additional immunologic barrier in patients with autoimmunity against beta cells. Further reports of success following islet allotransplantation into patients without autoimmune disease, including diabetes secondary to cystic fibrosis (24) and hemochromatosis (25), underscored the importance of developing immunosuppressive strategies effective against both autoimmune and alloimmune responses.

2 Dec 2003

22:9

AR

AR206-ME55-08.tex

AR206-ME55-08.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:133-156. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

BETA-CELL REPLACEMENT

P1: GCE

137

The strength of the autoimmune response contributing to beta-cell destruction was conclusively demonstrated in the landmark paper by Steffes et al. (26), in which a living-donor pancreas transplant between identical twins was rejected by the recipient. Pathologic confirmation that the recurrent autoimmune process specifically destroyed beta cells, but left the rest of the pancreas intact, attested to the potency of the autoimmune response leading to the destruction of beta cells. While the field of solid-organ transplantation for the treatment of type I diabetes enjoyed increasing success throughout the 1990s related to improvement in surgical techniques and immunosuppressive regimens, islet transplantation stagnated as a result of its relative failure to achieve insulin independence. The international registry, maintained in Giesen, Germany, demonstrated that as of 1990, <10% of the 447 islet allografts performed worldwide were able to sustain insulin independence (9). These disappointing results were attributed to the difficulty of obtaining satisfactory yields of functional islets following isolation, the toxic effects of the available immunosuppressive agents known to have beta-cell toxicity, and the inability to prevent alloimmunity as well as recurrent autoimmunity. In the majority of these procedures, the regimen of immunosuppression consisted of antibody induction with an antilymphocyte globulin combined with cyclosporine, azathioprine, and glucocorticoids. However, two centers, the University of Giesen (27) and the University of Alberta in Edmonton, Canada (28), persisted in perfecting the technology and immunosuppression essential for successful islet transplantation for patients with type I diabetes. The Giesen team’s meticulous attention to optimal conditions for pancreas procurement, preservation, islet isolation, islet engraftment, and immunosuppressive strategies demonstrated in the late 1990s that clinical islet transplantation could produce states of insulin independence (29). This rekindled the interest in islet transplantation and prompted several centers worldwide to reinitiate clinical islet transplantation programs (Figures 1 and 2).

The Breakthrough of the Edmonton Protocol In 1999, the University of Alberta group established that islet transplantation could consistently produce states of insulin independence (4). Two unique changes were responsible for the dramatic and consistent success observed in the Edmonton protocol. The first consisted of the intraportal infusion of freshly isolated islets, followed by a second and sometimes third infusion of additional islets from different donors (at a later date) in order to achieve insulin independence. Based on a review of their data, it appears that ∼8000–9000 islet equivalents per kilogram (Ieq/kg) were necessary to achieve a state of insulin independence, and this yield was rarely obtained with one donor. Thus, the majority of recipients required the infusion of islets isolated from at least two cadaveric donors. The other significant change from previous unsuccessful protocols was the use of effective immunosuppressive agents that do not cause toxicity to pancreatic islets (Figure 3). It has long been recognized that steroids and calcineurin inhibitors are toxic to beta cells. Although these agents have provided adequate immunosuppression for solid-organ

2 Dec 2003

22:9

Annu. Rev. Med. 2004.55:133-156. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

138

AR

STOCK

AR206-ME55-08.tex

¥

AR206-ME55-08.sgm

LaTeX2e(2002/01/18)

P1: GCE

BLUESTONE

Figure 1 Number of islet allografts over time (1974–2002). There has been a dramatic increase in islet allotransplants in the past decade. However, the number decreased significantly after 1995 until 1999 when the success of the Edmonton Protocol led to a significant increase of transplants each year. These data were compiled by the Islet Transplant Registry (ITR) and kindly provided by Dr. James Shapiro, Edmonton, Canada.

Figure 2 Number of individual sites throughout the world actively engaged in islet transplantation (1999–2002) with a total of 282 (as of early 2003). The number of transplants per site is shown in the parentheses. (Figure kindly provided by Dr. James Shapiro, Edmonton, Canada.)

2 Dec 2003

22:9

AR

AR206-ME55-08.tex

AR206-ME55-08.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:133-156. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

BETA-CELL REPLACEMENT

P1: GCE

139

Figure 3 There has been a significant change in the range and number of immunosuppressive drugs used in the islet transplant setting, as depicted in this timeline chart. Importantly, in the late 1990s there was a shift from the use of the highly diabetogenic corticosteroids to cocktails of immunosuppressive drugs that rely on direct anti–T-cell activity, sirolimus and low-dose calcineurin inhibitors. ATG, anti-thymocyteglobulin; ALG, anti–lymphocyte globulin; OKT3/OKT3γ 1 (ala-ala), anti-CD3; other names defined in text.

pancreas transplants, the quantity of beta cells transplanted with a whole pancreas was not sufficient to tolerate the toxic insult from the diabetogenic agents. By eliminating steroids and decreasing the dose of calcineurin inhibitors, Shapiro and colleagues provided effective immunosuppression for the decreased islet mass infused with islet transplants (4).

CURRENT STATE OF THE ART Pancreas Procurement, Preservation, and Islet Isolation Several important conclusions can be inferred from reviews of the largest series of islet isolations performed at centers that persisted in improving the technology of islet isolation (28, 30–33). Perhaps the most important observation is that successful isolation of pancreatic islets depends on the careful procurement of the donor pancreas. Many of the pancreases obtained for islet isolation were removed after procurement and preservation of other organs to be used in solid-organ transplant, with minimal attention dedicated to the procurement of the pancreas. Consequently, the initial preservation was less than optimal, and the initial insult ultimately translated into the poor quality and yield of the isolated islets (Figure 4). A second conclusion drawn from these large series of human islet isolation is the importance of minimizing cold ischemia time following the procurement of the donor pancreas. Pancreases procured from deceased donors can tolerate cold ischemic times of up to 24 h and consistently produce states of insulin independence following transplantation in patients with insulin-dependent diabetes (2).

2 Dec 2003

22:9

Annu. Rev. Med. 2004.55:133-156. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

140

AR

STOCK

AR206-ME55-08.tex

¥

AR206-ME55-08.sgm

LaTeX2e(2002/01/18)

P1: GCE

BLUESTONE

Figure 4 Essential elements of successful islet transplantation. Recent abstracts presented by James Shapiro and Jonathan Lakey at the American Society of Transplantation Conference (May 31 and June 2, 2003) were used to develep this list of criteria that influenced the outcome of islet transplantation during a controlled multicenter trial sponsored by the Immune Tolerance Network/JDRF/NIH (adapted figure kindly provided by Barbara DiMercurio, NIAID).

However, islets cannot tolerate the same amount of cold ischemia as their solidorgan counterpart. In the most complete review of worldwide experience with islet transplantation, Hering & Ricordi documented 16 cases of insulin independence following islet transplantation from a single donor prior to 1996 and observed that the ischemic time in 15/16 of these cases was <8 h (18). Subsequent trials of successful islet transplantation resulting in insulin independence since 1990 (4, 34, 35) correlate successful islet isolation and transplants to cold ischemic times of <8 h. The fact that islets have a lower tolerance for cold ischemia than the solid-organ pancreas is undoubtedly related to the multiple insults an islet must overcome during the isolation process. Important progress in minimizing the detrimental effect of cold ischemia by using a two-layer preservation technique for storing the solid-organ pancreas prior to the digestion process has been reported (36–38). The two-layer technique utilizes a perfluorochemical preservation fluid that improves oxygen delivery to the stored pancreas and increases islet yields after prolonged storage (36, 37). The isolation of human pancreatic islets from the solid-organ pancreas was significantly improved by the use of an automated digestion protocol (39, 40) and large-scale purification by continuous ficoll gradients on the Cobe 2991 bloodcell processor (41–43). Despite significant improvements, there continue to be unpredictable variables resulting in inconsistent yields of viable islets. The current isolation technology remains dependent on the distension of the pancreatic duct of the procured solid-organ pancreas with collagenase in order to dissociate the pancreas to free the imbedded islets. A significant Achilles heel of this procedure, however, relates to the constitution, consistency, and consequent reliability of the digestive enzymes required

2 Dec 2003

22:9

AR

AR206-ME55-08.tex

AR206-ME55-08.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:133-156. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

BETA-CELL REPLACEMENT

P1: GCE

141

to free the islets from the surrounding islet tissue. If the collagenase (a mixture of bacterially derived enzymes) is too potent and the exposure time too long, the digestion process will destroy the islets. If the enzymatic activity is weak or the exposure time too short, the islets will not be freed from the surrounding acinar tissue, resulting in poor yields. An advance in the ability to isolate islets was the development of Liberase (Boehringer-Mannheim, now Roche, Indianapolis, IN). Liberase is a blend of three distinct collagenase enzymes purified from bacteria. It is constituted to be low in endotoxin content, consistent in enzyme activity, and low in contaminants. The use of this consistent cocktail minimizes the variability between different preparations of the enzyme (43, 44). However, different lots of Liberase continue to demonstrate significant variability. This variability is reflected in continued inconsistency in yields of functional islets necessary for transplantation. In preliminary reports for several institutions performing islet transplantation, centers are reporting that among isolations performed with the intent to transplant, ∼50% are achieving islet preparations suitable for transplantation. The inconsistency in achieving adequate yields of islets is further complicated by the variability in the quality of the donor pancreas. In the most extensive analysis of the effect of donor variables on islet yield, Lakey et al. found that high donor body mass index (BMI), increased donor age, and procurement of the pancreas by an experienced local team improved yields of functional islets (28). Poor yields of functional islets correlated with hyperglycemia or hemodynamic instability of the donor prior to death, as well as increased duration of cold ischemia of the donor pancreas prior to the isolation procedure. Other analyses of the effect of donor variables have consistently correlated high donor BMI and donor hemodynamic stability with successful islet isolation (30, 32, 33, 46). Unfortunately, the current state of the art in terms of islet isolation suggests that only one third to one half of the preparations of islets isolated with the intent to transplant will yield enough functional islets (>250,000 Ieq) for transplantation. If up to three successful isolations are necessary to achieve an islet yield adequate for transplantation, as many as four to six donors may be necessary to achieve insulin independence for a given recipient. Clearly, the widespread application of islet transplantation as a treatment for diabetes mellitus will depend on improvements in the current technology of islet isolation and purification in order to produce more consistent yields of functional islets. It is important to note that some pancreases which are not suitable for solidorgan transplantation may provide excellent quality and quantity of pancreatic islets. For example, donors with a high BMI (>30) are rarely used for solid-organ transplantation because of the high risk of postoperative complications, yet are ideal for islet isolation (28, 33, 46, 47). Transplant groups from the University of Minnesota (34, 48) and University of Pennsylvania (49) have demonstrated consistent ability to achieve insulin independence from a single donor. Hering et al. (34, 48) have achieved insulin independence from a single donor by culturing the islets prior to transplantation, permitting pretransplant immunosuppression with immunodepleting agents to

2 Dec 2003

22:9

142

AR

STOCK

AR206-ME55-08.tex

¥

AR206-ME55-08.sgm

LaTeX2e(2002/01/18)

P1: GCE

BLUESTONE

facilitate engraftment. However, even this practical method for treatment of an increasing number of diabetic patients will rapidly be challenged by a severely limited donor pool.

Annu. Rev. Med. 2004.55:133-156. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Patient Selection for Trials Successful islet transplantation requires the prevention of the alloimmune response as well as the recurrent autoimmune response, so it remains dependent on immunosuppressive agents. Despite the ultimate possibility of transplantation tolerance (see below), current recipients of islet transplants will be subjected to the long-term sequelae of immunosuppressive agents, including increased risks of infections and malignancies. Therefore, as in whole-organ pancreas transplantation, the exchange of insulin therapy for immunosuppressive agents must be justified by an assessment of the risk/benefit ratio. For this reason, subjects included in the majority of the current trials are limited to type I diabetic recipients with hypoglycemic unawareness or metabolic instability despite intensive insulin therapy (34, 48–52). In addition to the requirements for “life-threatening” diabetes, most current trials limit the weight of the recipients, as a result of the reports from the large Edmonton trial demonstrating the requirement for 8000–9000 Ieq/kg to achieve states of insulin independence. The ITN [sponsored by the National Institutes of Health (NIH) and Juvenile Diabetes Research Foundation (JDRF)] is the first large multicenter trial designed in an attempt to confirm the Edmonton protocol at nine sites (45). Patients weighing more than 70 kg are excluded from the trial. Another technique for predicting the potential for success as defined by insulin independence could be pretransplant daily insulin requirements, although the inclusion/exclusion criteria for the ITN and other trials continue to rely on absolute weight. An important exclusion criterion for islet transplantation in the preuremic diabetic recipient is the presence of significant renal insufficiency. The current immunosuppressive strategies depend on the use of calcineurin inhibitors, which can further impair renal function and exacerbate existing diabetic nephropathy. For this reason, patients with significant impairment of renal function (creatinine clearance less than 60%–70%) are excluded from islet transplantation prior to receiving a kidney transplant. Given limited yields of islets per pancreas as well as the requirements for chronic immunosuppression, most type I diabetic patients selected for current trials of islet transplantation have life-threatening diabetes, with hypoglycemic unawareness or metabolic instability despite intensive insulin therapy. Although most of these patients have the secondary complications of neuropathy and retinopathy, the presence of significant nephropathy excludes patients from trials of islet transplantation in the absence of kidney transplantation. As islet trials are expanded, kidney transplant recipients will be simultaneously transplanted with islets, or islet transplantation will follow successful kidney transplant. A successful sequential kidney/islet transplant, dependent on steroid-free immunosuppression, was recently reported (53).

2 Dec 2003

22:9

AR

AR206-ME55-08.tex

AR206-ME55-08.sgm

LaTeX2e(2002/01/18)

BETA-CELL REPLACEMENT

P1: GCE

143

Many potential recipients for solid-organ pancreas transplantation cannot undergo the stress of the operative procedure because of advanced cardiovascular disease. Although the primary focus of islet transplantation is to provide a safer technique to justify earlier intervention prior to the development of secondary complications, it also allows beta-cell replacement in patients with life-threatening diabetes who cannot tolerate the cardiovascular stresses of solid-organ pancreas transplantation.

Annu. Rev. Med. 2004.55:133-156. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Techniques for Islet Transplantation The intraportal infusion of islets is the only technique that has successfully led to insulin independence following islet transplantation in humans. The intrahepatic site is the only one currently being pursued in human trials, although other sites have been studied in animals, including the omental pouch and the submucosal space of the upper gastrointestinal tract (54–56). Access to the portal system is accomplished with either percutaneous transhepatic cannulation (4, 49, 51, 52) or via branches of the mesenteric venous system cannulated by direct exposure using a mini-laparotomy (34). The advantage of the percutaneous approach is the avoidance of an abdominal wall incision, although the risk of postinfusion bleeding is greater. The risk of bleeding requiring transfusion is minimized by the administration of a hemostatic agent through the catheter tract at the conclusion of the islet infusion, as well as the use of a smaller 4Fr-gauge islet infusion catheter (50). The current infusion process requires anticoagulation with heparin to prevent thrombosis of the portal vein, although portal-vein thrombosis has been rare. In the most recent update from the Edmonton group, partial portal-vein thrombosis was detected following two of the 54 islet infusions. Five of the patients in this series had evidence of bleeding following the percutaneous infusion of the islets, three required transfusion, and one required transfusion and open surgery because of an expanding intrahepatic and subcapsular hemorrhage (52). Transient elevations in portal pressures have been noted during and immediately after the intraportal infusion of islets, but no long-term sequelae have been reported when purified islets have been infused (57). Although some of the earlier trials in islet transplantation used nonpurified preparations of islets, most of these preparations were obtained from pancreatectomy specimens that were processed for autoislet transplantation (20). In these cases, the amount of exocrine and endocrine tissue obtained from diseased pancreases was smaller than that obtained from a normal pancreas, and purification was not attempted because of concerns about further islet loss. Of note, one of the earliest reports of insulin independence following islet transplants from a single donor was accomplished with nonpurified islets (58). Because a significant increase in portal pressures has been associated with infusion of tissue pellets greater than 10 cc, the current islet protocols used purified preparations of islets with significantly smaller tissue pellets. Most current strategies for portal-vein cannulation have adopted the percutaneous route now that

2 Dec 2003

22:9

144

AR

STOCK

AR206-ME55-08.tex

¥

AR206-ME55-08.sgm

LaTeX2e(2002/01/18)

P1: GCE

BLUESTONE

technologic advances have minimized the risks of postinfusion bleeding. The mini-laparotomy and transjugular cannulation of the portal system remain viable alternatives.

Annu. Rev. Med. 2004.55:133-156. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Immunosuppressive Strategies A major advance toward achieving insulin independence following islet transplantation was the elimination or minimization of immunosuppressive agents known to have beta-cell toxicity. The success of the Edmonton protocol has in part been attributed to the elimination of steroids as well as the minimization of the calcineurin inhibitor tacrolimus. The ability to provide regimens that are not toxic to beta cells following islet transplantation has been facilitated by access to newer non–beta-cell–toxic agents, most notably sirolimus, a TOR inhibitor that blocks IL-2–dependent proliferation of T lymphocytes. Current immunosuppressive regimens depend on an induction agent used only around the time of the islet infusion, followed by maintenance immunosuppression. For the Edmonton trial as well as the NIH/JDRF ITN trial the induction agent was dacluzimab, a humanized monoclonal antibody directed against the IL-2 receptor (CD25) (4, 45). Maintenance immunosuppressive therapy has consisted of sirolimus and low-dose tacrolimus, thus minimizing the nephrotoxic and diabetogenic effects of tacrolimus but providing enough immunosuppression to protect the islets from the immune response. For the Edmonton protocol, islets were transplanted immediately following preparation without any culture period (4, 45). Although protocols utilizing fresh islets for immediate transplantation after islet preparation offer the benefit of minimizing further islet loss in culture, they do not allow functional testing of the isolated islets prior to transplantation, and there is some risk of transplanting impaired islets. Moreover, the immediate transplantation of the islets precludes any pretreatment of the recipient or the islets themselves. A strategy that is gaining popularity includes a 48-hour period of islet culture prior to transplantation. This permits the pretransplant evaluation of the purified islets for purity, sterility, endotoxin content, and in vitro response to a glucose challenge. Of equal significance, the pretransplant culture period has permitted induction therapy with immunodepleting (i.e., anti-CD3) and anti-inflammatory (i.e., anti-TNF) agents prior to the infusion of islets, thus facilitating the engraftment of islets by minimizing the initial immune-mediated “hit” to the islet allograft. Another successful strategy, which has resulted in insulin independence following the infusion of islets isolated from a single donor, has utilized induction therapy with a nonmitogenic humanized antibody directed against CD3 (34) followed by maintenance therapy with sirolimus and low-dose tacrolimus (see below). As more protocols are developed, other promising immunodepleting agents being tested include Thymoglobulin (polyclonal anti–T-cell agent) and Campath-1H (monoclonal anti-CD52). In addition, non-nephrotoxic and non–beta-cell–toxic agents are being developed, including biologic agents that block costimulation (CTLA4-Ig,

2 Dec 2003

22:9

AR

AR206-ME55-08.tex

AR206-ME55-08.sgm

LaTeX2e(2002/01/18)

BETA-CELL REPLACEMENT

P1: GCE

145

Annu. Rev. Med. 2004.55:133-156. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

anti-CD11a, anti-CD40L), anti-CD45RB, and FTY720. Despite the development of more selective immunosuppression and less toxic immunosuppression, any use of these agents will inhibit widespread application of islet transplantation regardless of islet availability, based on the risk/benefit ratio of immunosuppression versus insulin therapy.

RESULTS IN THE CURRENT ERA: RATES OF INSULIN INDEPENDENCE, ISLET FUNCTION AND LONGEVITY, COMPLICATIONS Since the initial report (4), an increasing number of centers have reported achieving insulin independence following islet transplantation. Recent reports of success have come from Miami (59), the NIH (60), Philadelphia (49), Minnesota (34, 48), Milan (61), and the Swiss-French Group (35). The Minnesota and Philadelphia teams achieved insulin independence with islets isolated from a single donor most of the time. The long-term function and viability of islet transplantation remains to be determined; the longest published follow-up to date is that reported by the Edmonton group (52). Of the 17 consecutive patients who completed the protocol and obtained insulin independence, 15 were available for one-year follow-up. Of these patients, 12 (80%) remained insulin-independent and had normalization of HbA1C levels. All patients required islets isolated from at least two donors, and a total islet mass of 8000–9000 Ieq/kg was necessary to achieve insulin independence. At the time of the most recent publication, the mean follow-up of the 17 patients who had obtained initial insulin independence was 20.4 months. Eleven of the 17 remained insulin-independent, although 2 of the 11 were started on an oral agent because of a rise in HbA1C levels. C-peptide was lost in 3 of the 6 patients who returned to insulin therapy. Half of the patients who returned to insulin therapy had an increase in the titer of islet cell antigen (ICA) and glutamic acid decarboxylase (GAD) antibodies, which suggests recurrence of the autoimmune response. The etiology of the islet loss in the other patients remains speculative, but it could be related to “burnout” of the islet tissue versus the alloimmune response. In terms of function, only two patients with insulin independence had normal oral glucose tolerance testing (OGTT), despite stable glucose control and normalization of HbA1C. The abnormal OGTT seen in the majority of the recipients presumably represents an islet mass that is marginal but sufficient to produce states of insulin independence. All patients with detectable C-peptide, regardless of insulin requirements, had resolution of glycemic instability and hypoglycemic unawareness. The most significant complications associated with the islet-transplant procedure have been related to the side effects of immunosuppression. In general, the procedure is well-tolerated and requires a single day of hospitalization if performed percutaneously. To date, none of the centers that perform islet transplantation

2 Dec 2003

22:9

Annu. Rev. Med. 2004.55:133-156. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

146

AR

STOCK

AR206-ME55-08.tex

¥

AR206-ME55-08.sgm

LaTeX2e(2002/01/18)

P1: GCE

BLUESTONE

have reported any long-term sequelae to the liver from the intraportal injection of the islets, based on resolution of liver function tests. However, the development of portal hypertension and long-term effect of islets on the liver have yet to be formally addressed. As for toxicity from the immunosuppressive drugs, even lowdose tacrolimus (trough <5 ng/ml) has resulted in progression of renal insufficiency in 2 patients in the Edmonton trial who had evidence of pre-existing disease, as well as exacerbation of proteinuria in 4 other patients (52). Antihypertensive therapy was started in 53% of the patients in the Edmonton trial, presumably related to tacrolimus. A known complication of sirolimus therapy is hypercholesterolemia, and 53% of patients in the Edmonton trial required statin therapy (52). Anemia and leukopenia have also been seen in several protocols using sirolimus as maintenance therapy; patients have required granulocyte colony-stimulating factor to correct the leukopenia. A frequent complaint in all regimens utilizing therapeutic levels of sirolimus has been mouth ulcers, which resolve after decreases in the dosage of sirolimus. Diarrhea has also been a significant complaint of patients on sirolimus/tacrolimus-based therapy but has not been a long-term problem. No serious infectious complications have been reported, including infectious complications related to the infusion of contaminated islets. Of equal significance, there has been no evidence of cytomegalovirus (CMV) disease despite the transplantation of islets isolated from CMV-positive donors into CMV-negative recipients. None of the islet trials have reported lymphoproliferative disease. There has been no sensitization of islet recipients to HLA antigen, despite the infusion of islets isolated from multiple donors (4, 34, 52). The lack of sensitization reflects effective immunosuppression from the alloimmune response, which is reassuring, since these patients are at significant risk for needing a kidney transplant. On the other hand, the necessity of reinstituting insulin therapy in 6/17 patients at a mean follow-up of 20 months in the Edmonton trial suggests that further improvement in immunosuppressive strategies may be necessary (52). This concern is confirmed by evidence of recurrent autoimmunity in half of the patients who returned to insulin therapy. The cause for graft failure in the other three patients remains unclear. The lack of the development of anti-HLA antibodies as well as autoantibodies suggests that the loss of islet function may be nonimmunologic; rather, the loss of insulin production may reflect “burnout” of the islets, as many of the patients have only a marginal islet mass (see above). Recent evidence suggests that neo-islet formation from islet progenitor cells in the ducts of the pancreas is necessary to maintain beta-cell mass (62). Further follow-up of the successful islet transplant will provide insight into the requirement for islet precursors to maintain long-term function. Although current reports suggest the safety and efficacy of islet transplants, the ultimate utility of this procedure will depend on finding a better source of islets, as well as even better immunosuppression to prevent alloimmunity and recurrent autoimmunity. For this reason, the next sections focus on two areas highly relevant to further development of islet-transplantation strategies: the development of transplantation tolerance and the potential for stem cell–derived islets for transplantation.

2 Dec 2003

22:9

AR

AR206-ME55-08.tex

AR206-ME55-08.sgm

LaTeX2e(2002/01/18)

BETA-CELL REPLACEMENT

P1: GCE

147

Annu. Rev. Med. 2004.55:133-156. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

ISLETS AS A MODEL FOR TOLERANCE STRATEGIES Islet transplantation has become an accepted treatment option for selected patients with inadequate glucose control even under stringent insulin therapy. However, the application of this procedure for the treatment of type 1 diabetes is limited by the need for potent nonselective immunosuppression, including nephrotoxic and diabetogenic calcineurin-inhibitor therapy. Islet transplantation has unique pluses and minuses as a venue for testing tolerogenic therapies (therapies that will lead to a rejection-free state without ongoing immunotherapies). On the plus side, diabetes is not an acutely life-threatening disease. Additionally, even though the graft might be lost should the tolerogenic therapies fail, even short-lived blood-sugar control has significant long-term benefit to diabetic patients, reducing complications and morbidity as reported by the Diabetes Control and Complications Trial Group (63). Moreover, in contrast to solid-organ transplantation, the risk of the transplant procedure itself is quite modest. There is no surgery, and the procedural risks are limited to portal hypertension and, in rare cases, bleeding. On the other hand, one weakness of choosing islet transplantation as a target for tolerance therapies in patients with type 1 diabetes is that, unlike other organ-transplantation settings, in type 1 diabetes the induction of potent and persistent tolerance must be considered in the context of an ongoing pathogenic autoimmune response. Thus, tolerogenic strategies for the treatment of type 1 diabetes are confounded by the memory T-cell response, the presence of autoantibodies, and the genetic immunologic abnormalities found in the autoimmune individual. As such, a series of novel therapeutic approaches must be developed to target clonal deletion of T and B cells, regulatory T-cell expansion, and altered T-cell receptor-mediated signaling. Over the past few decades, a more detailed understanding of the molecular events associated with T-cell recognition and activation has led to various tolerance approaches in numerous models of both autoimmunity and transplantation (64). The results of these studies have suggested a number of sites for intervention in the immune response. For instance, it is clear that T lymphocytes require the engagement of both the T-cell receptor and a series of coreceptors, notably costimulatory signals, for complete activation. Blockade of these cell-surface molecules results in incomplete activation and T-cell anergy. Thus, costimulation antagonists present an attractive means of promoting tolerance (65). Clinical trials of costimulation blockade have provided important data pertaining to the safety, efficacy, and mechanisms of tolerance induction. For instance, the kidney transplant program at the University of California, San Francisco has been involved in the initial trials of LEA29Y, a high-affinity mutant form of CTLA-4Ig that antagonizes the CD28 costimulatory pathway. Preliminary studies have shown that this therapy provides effective immunosuppression following renal transplantation in humans and islet allografts in nonhuman primates (66). Similarly, in the preclinical nonhuman-primate allotransplant setting, anti-CD154 monoclonal antibody (IDEC-131) and other CD28 antagonists—such as antibodies directed against the CD28 ligands, B7.1 and B7.2 (67)—have been effectively combined

2 Dec 2003

22:9

Annu. Rev. Med. 2004.55:133-156. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

148

AR

STOCK

AR206-ME55-08.tex

¥

AR206-ME55-08.sgm

LaTeX2e(2002/01/18)

P1: GCE

BLUESTONE

with rapamycin for maintenance immunosuppression (68, 69). It should be emphasized that all of these preclinical and clinical studies successfully prevented rejection without corticosteroid and calcineurin inhibitors. Thus, the use of single or combination costimulation antagonists is likely to be a productive avenue of clinical research in the islet-transplantation arena. The use of T-cell–depleting induction therapy has become widespread, and new trials have been designed with a focus on immunosuppressive drug withdrawal. The hypothesis behind the use of these drugs is that a transient but profound T-cell depletion can reset the immune system to a tolerized state in the presence of autoantigen and alloantigen expression of the foreign islets. Preclinical studies have shown that a depleting anti-CD3 immunotoxin, combined with more general macrophage inhibitor 15-deoxyspergualin, is an effective tolerogenic therapy (70). There has been increasing use of Campath 1H (Alemtuzumab), a humanized anti-CD52 monoclonal antibody (mAb), currently approved for the treatment of B-cell chronic lymphocytic leukemia. The mAb has been shown to rapidly deplete peripheral blood B cells and T cells. In some cases, it can take over a year to reconstitute the immune system. A number of studies are testing the efficacy of Campath 1H in transplantation, with particular regard to calcineurin-free regimens. In one study, over a dozen patients received transplants under the cover of Campath 1H with encouraging results (71). Based on these results and others, the ITN (http://www.immunetolerance.org) has approved an islet-transplant protocol by the Edmonton group that will treat islet-transplant recipients with a combination of Campath 1H and sirolimus (rapamycin) with the intent to withdraw all drugs at ∼1 year, assuming that the patients meet certain requirements of operational tolerance. Thymoglobulin, a polyclonal rabbit antihuman thymocyte globulin (SangStat), is approved for the treatment of acute renal-transplant rejection and is a powerful lymphocyte-depleting agent. At the 2002 American Transplantation Congress, the transplant group from the University of Pittsburgh reported a small clinical trial that used thymoglobulin as a pretreatment drug (72). The investigators reported weaning success in 64 kidney-transplant patients, 7 of whom took the calcineurin inhibitor (tacrolimus) just once weekly. In addition, of 18 lung-transplant patients put through the same protocol, 5 are on tacrolimus four times weekly, 5 on a once-daily dose, and 8 on a twice-daily dose. Instead of the usual three-drug combination, the lung-transplant recipients are receiving only the tacrolimus plus very-low-dose prednisone. It is anticipated that many islet-transplant trials in the next few years will focus on these T-depleting agents. As an example, an ongoing thymoglobulin-based islet-transplant trial at the University of Minnesota uses the thymoglobulin therapy in combination with low-dose conventional immunosuppressive drugs (48). Thus, it is likely that these types of therapies will be evolving and moving into the islet-transplantation area. Preclinical studies have suggested that one approach to attaining tolerance is the creation of a chimeric state in which large numbers of donor cells are maintained in the recipient. The most clinically relevant approaches have been those that use nonmyeloablative host-conditioning regimens, since the whole-body irradiation

2 Dec 2003

22:9

AR

AR206-ME55-08.tex

AR206-ME55-08.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:133-156. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

BETA-CELL REPLACEMENT

P1: GCE

149

used in other regimens to allow donor bone marrow to become established carries excessive risks of toxicity. Nonmyeloablative mixed chimeric approaches have allowed complete immunosuppressive withdrawal in some limited cases (73). Thus, these therapies may provide a robust approach to tolerance in the islet-transplant setting. In this regard, Ricordi and colleagues were recently approved to perform allogeneic islet transplants using a combination of Campath 1H, sirolomus, and CD34+ stem cell therapy in preconditioned recipients (http://www.immuneto lerance.org). The goal of this study is to determine whether this protocol will lead to chimerism that can be the basis for total drug withdrawal (74). An approach that has been successful in moderating both the autoimmune and alloimmune responses is the use of T-cell receptor antagonists such as anti-CD4 or anti-CD3 mAbs. Over the past decade, the potential for anti-CD3 mAbs to induce tolerance in a safe and effective manner has been studied by several groups. In mice, a five-day course of anti-CD3 antibodies at the time of disease onset was sufficient to reverse the disease, induce long-term remission, and prevent recurrent immune responses even against transplanted syngeneic islets (75). Mechanistic studies demonstrated that the mAb has short-lived effects on naive T cells but delivers a partial signal in activated T cells, inactivating Th1 cells while permitting proliferation/cytokine production by Th2 cells and regulatory T cells (76). Thus, the antibody therapy “tips” the balance of immune homeostasis so that the Th2 and regulatory T cells block the residual pathogenic response. Based on in vitro and small animal studies, a phase I/II trial of patients with new-onset type 1 diabetes was initiated. In this trial, a humanized FcR nonbinding anti-CD3 mAb was administered for two weeks at the time of disease onset. The FcR nonbinding anti-CD3 mAb, given without other immune-suppressive agents, halted the progression of disease for >1 year (74). Moreover, significant increases in IL-10 in the serum of approximately two thirds of treated patients and an IL10+CD4+ T-cell population was observed in vivo after drug treatment. These studies led to a pilot trial in which “brittle” type 1 diabetic patients were transplanted with allogeneic islets under the cover of FcR nonbinding anti-CD3, low-dose tacrolimus, and sirolimus. Sixty-six percent of these patients maintained long-term insulin independence after a single-donor islet transplant (34). However, to date, the therapies have not been shown to be toleragenic in the allotransplant setting. In this regard, the ITN recently agreed to sponsor a drug-withdrawal trial in islettransplant recipients at the University of Minnesota. Investigators will administer a combination of the FcR nonbinding anti-CD3 mAb in combination with sirolimus. Patients will be withdrawn from all drugs at ∼1 year, assuming that they meet certain requirements of operational tolerance (http://www.immunetolerance.org).

STEM CELL–DERIVED ISLETS FOR TRANSPLANTATION The shortage of functional beta cells from available donors is becoming one of the major limiting factors for the treatment of diabetes by islet transplantation. Even if all the available cadaveric pancreases could be used effectively to prepare islets

2 Dec 2003

22:9

Annu. Rev. Med. 2004.55:133-156. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

150

AR

STOCK

AR206-ME55-08.tex

¥

AR206-ME55-08.sgm

LaTeX2e(2002/01/18)

P1: GCE

BLUESTONE

and if each recipient needed only one pancreas equivalent, the supply of pancreases is believed sufficient to treat only a small fraction of all individuals with type 1 diabetes. A potential solution is the use of xenogeneic tissue (see above). However, barriers to successful xenotransplantation include the risk of transmitting infections agents from one species to another and the inherent increased immunological reactivity to tissues from other species. Given the significant hurdles facing these approaches, other possible sources for islet tissue are being sought. In this section, we describe two parallel approaches to developing islet cells: the use of adult and embryonic stem (ES) cells. ES cells are derived from embryos that develop from eggs that have been fertilized in vitro. They are not derived from eggs fertilized in a woman’s body. Human ES cells are derived from the blastocyst, which is a group of 32 cells. Because of ES cells’ special properties of renewable growth and selective differentiation, deriving insulin-secreting cells from this source is an exciting prospect for generating an unlimited supply of specialized beta cells for transplantation. Additionally, ES cells are amenable to stable genetic modification, through which they could be manipulated so as to escape or inhibit the immune responses of the patient and prevent rejection. In fact, recent studies have suggested that stem cells themselves are protolerogenic, suggesting that these cells may be used as the tolerogen for subsequent islet-cell therapy (77). Finally, differentiated cell types derived from ES cells possess many of the physiologic and functional capacities of normal cells. Under certain conditions, with selected differentiation factors, the ES cells can be differentiated into insulin-producing beta cells (78, 79). More important, the longterm function or expansion of such cells when they are transplanted into diabetic animals has been demonstrated (79). Unfortunately, this field remains extremely controversial, with some scientists challenging the validity of the experimental results (80). Despite several publications suggesting that mouse and human ES cells can be used to develop islet cells, convincing evidence that this is feasible is still lacking. An alternative possibility is that multipotent progenitor cells reside within the adult pancreas and that regulated differentiation of these cells could fill the role of ES-cell–derived beta cells. In the adult pancreas, beta cells have a limited life span, and cell replacement is critical to maintain glucose homeostasis (81). Regeneration of beta cells after tissue injury has been observed in several model systems, indicating that certain cells within the mature pancreas retain the ability to partially restore beta-cell mass after injury (82). Recent studies indicate that differentiated exocrine acinar and/or ductal cells may trans-differentiate into beta cells (81). In fact, recent studies have demonstrated progressive beta-cell regeneration in beta-cell–deficient mice (83). Whether this in vivo observation can be adapted to an in vitro system that will result in sufficient expansion for human transplantation remains to be seen. Nonetheless, it should be pointed out that even as few as five doublings of already mature beta cells could lead to a 30-fold increase in islet transplants. Several studies over the past 4–5 years have suggested that stem cells for endocrine pancreas are present in distant organs such as bone marrow (84, 85).

2 Dec 2003

22:9

AR

AR206-ME55-08.tex

AR206-ME55-08.sgm

LaTeX2e(2002/01/18)

BETA-CELL REPLACEMENT

P1: GCE

151

Annu. Rev. Med. 2004.55:133-156. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

A recent study has suggested that cells transferred during bone marrow transplantation can be detected in the pancreas and differentiate locally into functioning beta-like cells (84). However, there is evidence that the bone marrow cells fuse to endogenous tissues and may not differentiate into the target tissue (86). Still, it is encouraging that a unique population of bone marrow progenitors developed by Verfaillie and colleagues can be induced to differentiate to most somatic cells may provide a novel approach to developing a new islet source (85).

CONCLUSIONS: INSULIN THERAPY VERSUS BETA-CELL REPLACEMENT The ability to achieve insulin independence with either solid-organ pancreas or islet transplantation has increased the number of diabetic patients seeking beta-cell replacement as an alternative to insulin therapy. Given the known complications of chronic immunosuppression, either transplantation procedure should be limited to patients requiring immunosuppression for a kidney transplant or patients with lifethreatening diabetes mellitus. The latter category would include diabetic patients with hypoglycemic unawareness or metabolic instability/lability despite intensive insulin regimens. However, significant achievements have been made in the field of clinical islet transplantation in the past three years. The demonstration of diabetes reversal on a consistent basis in islet-transplant recipients marked a turning point in the history of islet transplantation and cell-based diabetes therapies (4). These findings have now been confirmed at ∼10 additional institutions. Yet, for islet transplantation to become the treatment of choice for type 1 diabetics, additional advances are necessary, including maximization of the islet preparation and the transplant protocol as well as the introduction of new, potentially tolerogenic drugs to control both alloimmune and autoimmune responses. Corticosteroids and calcineurin inhibitors must be avoided to circumvent inhibitory effects on insulin secretion and insulin action, which are both particularly deleterious if a marginal mass is present. Avoiding calcineurin inhibitors will further eliminate nephrotoxic side effects and increase the likelihood of successful tolerance induction. Most important, widespread use of the procedure will depend on new sources of islets either using adult or embryonic stem cells. Meanwhile, many insulin-dependent diabetics meet the indications for betacell replacement today, so an algorithm based on the current state of the art must be developed. Because the success of islet transplantation as defined by insulin independence requires the infusion of ∼9000 Ieq/kg, this procedure would be most effective in patients with low BMI or low insulin requirements. Islets are a viable option for such patients if they are hesitant to undergo an abdominal operation. For patients receiving transplants at experienced centers, the chance of insulin independence one year after the procedure is 80% for islet transplants versus

2 Dec 2003

22:9

Annu. Rev. Med. 2004.55:133-156. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

152

AR

STOCK

AR206-ME55-08.tex

¥

AR206-ME55-08.sgm

LaTeX2e(2002/01/18)

P1: GCE

BLUESTONE

90% for solid-organ transplants. For larger patients with life-threatening diabetes, solid-organ pancreas transplant remains a better alternative, offering a high chance of insulin independence. The operative procedure for solid-organ transplantation is rigorous, and patients with significant cardiovascular risks would be better served by an islet transplant. Over 20 years ago, high hopes were pinned on islet transplantation as a cure for diabetes mellitus. Early results from the Edmonton trial suggest that we are closer to that goal but still have room for significant progress. The widespread application of islet transplantation will depend on further improvements in immunosuppressive strategies, advances in the area of transplantation tolerance, increases in the longevity of islet transplants, and development of an unlimited source of beta cells. The Annual Review of Medicine is online at http://med.annualreviews.org

LITERATURE CITED 1. Gruessner AC, Sutherland DE, Dunn DL, et al. 2001. Pancreas after kidney transplants in posturemic patients with type I diabetes mellitus. J. Am. Soc. Nephrol. 12:2490–99 2. Gruessner A, Sutherland D. 2002. Pancreas transplant outcomes for United States (US) and non-US cases as reported to the United Network for Organ Sharing (UNOS) and the International Pancreas Transplant Registry (IPTR) as of October, 2001. In Clinical Transplants 2001, ed. M Cecka, P Terasaki, pp. 41–72. Los Angeles: Regents of Univ. Calif. 3. Sutherland DE, Gruessner RW, Najarian JS, et al. 1998. Solitary pancreas transplants: a new era. Transplant. Proc. 30: 280–81 4. Shapiro AM, Lakey JR, Ryan EA, et al. 2000. Islet transplantation in seven patients with type 1 diabetes mellitus using a glucocorticoid-free immunosuppressive regimen. N. Engl. J. Med. 343:230–38 5. Nathan DM. 1993. Long-term complications of diabetes mellitus. N. Engl. J. Med. 328:1676–85 6. The DCCT Research Group. 1993. The effect of intensive treatment of diabetes on the development of long-term complica-

7.

8.

9.

10.

11.

12.

tions in insulin-dependent diabetes mellitus. N. Engl. J. Med. 329:977–86 Kaufman DB, Leventhal JR, Koffron AJ, et al. 2002. A prospective study of rapid corticosteroid elimination in simultaneous pancreas-kidney transplantation: comparison of two maintenance immunosuppression protocols: tacrolimus/mycophenolate mofetil versus tacrolimus/sirolimus. Transplantation 73:169–77 Freise C, Hirose R, Feng S, et al. 2002. Minimal rejection with a steroid sparing protocol in simultaneous pancreas kidney transplantation. Am. J. Transplant. 2(Suppl. 3): 203 (Abstr.) Brendel M, Hering B, Schultz A, et al. 2001. International Islet Transplant Registry Report. Int. Transplant Reg. Newsl. 8:1–20 Fioretto P, Steffes MW, Sutherland DE, et al. 1998. Reversal of lesions of diabetic nephropathy after pancreas transplantation. N. Engl. J. Med. 339:69–75 Navarro X, Sutherland DE, Kennedy WR. 1997. Long-term effects of pancreatic transplantation on diabetic neuropathy. Ann. Neurol. 42:727–36 Gross CR, Limwattananon C, Matthees BJ. 1998. Quality of life after pancreas

2 Dec 2003

22:9

AR

AR206-ME55-08.tex

AR206-ME55-08.sgm

LaTeX2e(2002/01/18)

BETA-CELL REPLACEMENT

13.

14.

Annu. Rev. Med. 2004.55:133-156. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

transplantation: a review. Clin. Transplant. 12:351–61 Ballinger WF, Lacy PE. 1972. Transplantation of intact pancreatic islets in rats. Surgery 72:175–86 Reckard CR, Barker CF. 1973. Transplantation of isolated pancreatic islets across strong and weak histocompatibility barriers. Transplant. Proc. 5:761–63 Najarian JS, Sutherland DE, Matas AJ, et al. 1977. Human islet transplantation: a preliminary report. Transplant. Proc. 9:233–36 Matas AJ, Sutherland DE, Payne WD, et al. 1977. Islet transplantation. The critical period of donor ischemia in neonatal rats. Transplantation 23:295–98 Sutherland DE, Matas AJ, Najarian JS. 1978. Pancreatic islet cell transplantation. Surg. Clin. North Am. 58:365–82 Hering B, Ricordi C. 1999. Islet transplantation for patients with type I diabetes. Graft 2:12–27 Hu YF, Gu ZF, Zhang HD, Ye RS. 1992. Fetal islet transplantation in China. Transplant. Proc. 24:1998–99 Wahoff DC, Papalois BE, Najarian JS, et al. 1995. Autologous islet transplantation to prevent diabetes after pancreatic resection. Ann. Surg. 222:562–75; discussion 75–79 Robertson RP, Lanz KJ, Sutherland DE, et al. 2001. Prevention of diabetes for up to 13 years by autoislet transplantation after pancreatectomy for chronic pancreatitis. Diabetes 50:47–50 Ricordi C, Tzakis AG, Carroll PB, et al. 1992. Human islet isolation and allotransplantation in 22 consecutive cases. Transplantation 53:407–14 Tzakis AG, Ricordi C, Alejandro R, et al. 1990. Pancreatic islet transplantation after upper abdominal exenteration and liver replacement. Lancet 336:402–5 Cretin N, Buhler L, Fournier B, et al. 1998. Results of human islet allotransplantation in cystic fibrosis and type I diabetic patients. Transplant. Proc. 30:315–16 Brunicardi FC, Atiya A, Stock P, et al. 1995.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

P1: GCE

153

Clinical islet transplantation experience of the University of California Islet Transplant Consortium. Surgery 118:967–71; discussion 71–72 Steffes MW, Sutherland DE, Goetz FC, et al. 1985. Studies of kidney and muscle biopsy specimens from identical twins discordant for type I diabetes mellitus. N. Engl. J. Med. 312:1282–87 Hering BJ, Bretzel RG, Hopt UT, et al. 1994. New protocol toward prevention of early human islet allograft failure. Transplant. Proc. 26:570–71 Lakey JR, Warnock GL, Rajotte RV, et al. 1996. Variables in organ donors that affect the recovery of human islets of Langerhans. Transplantation 61:1047–53 Bretzel RG, Brandhorst D, Brandhorst H, et al. 1999. Improved survival of intraportal pancreatic islet cell allografts in patients with type-1 diabetes mellitus by refined peritransplant management. J. Mol. Med. 77:140–43 Zeng Y, Torre MA, Karrison T, et al. 1994. The correlation between donor characteristics and the success of human islet isolation. Transplantation 57:954–58 Benhamou PY, Watt PC, Mullen Y, et al. 1994. Human islet isolation in 104 consecutive cases. Factors affecting isolation success. Transplantation 57:1804–10 Brandhorst D, Hering BJ, Brandhorst H, et al. 1994. Influence of donor data and organ procurement on human islet isolation. Transplant. Proc. 26:592–93 Brandhorst H, Brandhorst D, Hering BJ, et al. 1995. Body mass index of pancreatic donors: a decisive factor for human islet isolation. Exp. Clin. Endocrinol. Diabetes 103(Suppl. 2):23–26 Hering B, Kandaswamy R, Harmon J, et al. 2001. Insulin independence after singledonor islet transplantation in type I diabetes with hOKT3 (ala-ala), sirolimus, and tacrolimus. Am. J. Transplant. 1(Suppl. 1):180A Benhamou PY, Oberholzer J, Toso C, et al. 2001. Human islet transplantation network

2 Dec 2003

22:9

154

Annu. Rev. Med. 2004.55:133-156. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

AR

STOCK

AR206-ME55-08.tex

¥

AR206-ME55-08.sgm

LaTeX2e(2002/01/18)

P1: GCE

BLUESTONE

for the treatment of type I diabetes: first data from the Swiss-French GRAGIL consortium (1999–2000). Groupe de Recherche Rhin Rhjne Alpes Geneve pour la transplantation d’Ilots de Langerhans. Diabetologia 44:859–64 Matsumoto S, Kandaswamy R, Sutherland DE, et al. 2000. Clinical application of the two-layer (University of Wisconsin solution/perfluorochemical plus O2) method of pancreas preservation before transplantation. Transplantation 70:771–74 Hering BJ, Matsumoto I, Sawada T, et al. 2002. Impact of two-layer pancreas preservation on islet isolation and transplantation. Transplantation 74:1813–16 Kuroda Y, Kawamura T, Suzuki Y, et al. 1988. A new, simple method for cold storage of the pancreas using perfluorochemical. Transplantation 46:457–60 Warnock GL, Ellis D, Rajotte RV, et al. 1988. Studies of the isolation and viability of human islets of Langerhans. Transplantation 45:957–63 Ricordi C, Lacy PE, Finke EH, et al. 1988. Automated method for isolation of human pancreatic islets. Diabetes 37:413–20 Olack B, Swanson C, McLear M, et al. 1991. Islet purification using EuroFicoll gradients. Transplant. Proc. 23:774– 76 Finke E, Marchetti P, Falqui L, et al. 1991. Large scale isolation, function, and transplantation of islets of Langerhans from the adult pig pancreas. Transplant. Proc. 23:772–73 Lake SP, Bassett PD, Larkins A, et al. 1989. Large-scale purification of human islets utilizing discontinuous albumin gradient on IBM 2991 cell separator. Diabetes 38(Suppl. 1):143–45 Linetsky E, Bottino R, Lehmann R, et al. 1997. Improved human islet isolation using a new enzyme blend, liberase. Diabetes 46:1120–23 Shapiro J, Hering B, Ricordi C, et al. 2003. International multicenter trial of islet transplantation using the Edmonton protocol in

46.

47.

48.

49.

50.

51.

52.

53.

54.

55.

56.

patients with type I diabetes. Am. J. Transplant. 3 (Suppl. 5):A3 (Abstr.) Rastellini C, Rilo HH, Fontes P, et al. 1995. The effect of donor age on the outcome of islets isolated from human pancreata. Transplant. Proc. 27:3257–58 Une S, Atiya A, Ohtsuka S, et al. 1997. Reevaluation of donor factors affecting islet isolation from human pancreas using a twostep digestion method. Transplant. Proc. 29:1969 (Abstr.) Hering B, Kandaswamy R, Ansite J, et al. 2003. Successful single donor islet transplantation in type I diabetes. Am. J. Transplant. 3(Suppl. 5):A567 (Abstr.) Markmann JF, Deng S, Huang X, et al. 2003. Insulin independence following isolated islet transplantation and single islet infusions. Ann. Surg. 237:741–49; discussion 49–50 Shapiro AM, Ryan EA, Lakey JR. 2001. Pancreatic islet transplantation in the treatment of diabetes mellitus. Best Pract. Res. Clin. Endocrinol. Metab. 15:241–64 Ryan EA, Lakey JR, Rajotte RV, et al. 2001. Clinical outcomes and insulin secretion after islet transplantation with the Edmonton protocol. Diabetes 50:710–19 Ryan EA, Lakey JR, Paty BW, et al. 2002. Successful islet transplantation: continued insulin reserve provides long-term glycemic control. Diabetes 51:2148–57 Kaufman DB, Baker MS, Chen X, et al. 2002. Sequential kidney/islet transplantation using prednisone-free immunosuppression. Am. J. Transplant. 2:674–77 Yasunami Y, Lacy PE, Finke EH. 1983. A new site for islet transplantation—a peritoneal-omental pouch. Transplantation 36:181–82 Ao Z, Matayoshi K, Lakey JR, et al. 1993. Survival and function of purified islets in the omental pouch site of outbred dogs. Transplantation 56:524–29 Tchervenivanov N, Metrakos P, Kokugawa Y, et al. 1994. Submucosal transplantation of pancreatic islets. Transplant. Proc. 26:680–81

2 Dec 2003

22:9

AR

AR206-ME55-08.tex

AR206-ME55-08.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:133-156. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

BETA-CELL REPLACEMENT 57. Casey JJ, Lakey JR, Ryan EA, et al. 2002. Portal venous pressure changes after sequential clinical islet transplantation. Transplantation 74:913–15 58. Gores PF, Najarian JS, Stephanian E, et al. 1993. Insulin independence in type I diabetes after transplantation of unpurified islets from single donor with 15deoxyspergualin. Lancet 341:19–21 59. Alejandro R, Ferreira J, Froud T, et al. 2003. Insulin independence in 13 patients following transplantation of cultured islets. Am. J. Transplant. 3(Suppl. 5):A565 (Abstr.) 60. Rother K, Hirshberg B, Gaglia J, et al. 2002. Islet transplantation in patients with type I diabetes. NIH experience in 6 patients. Am. J. Transplant. 2(Suppl. 3):A359 (Abstr.) 61. Luzi L, Perseghin G, Brendel MD, et al. 2001. Metabolic effects of restoring partial beta-cell function after islet allotransplantation in type 1 diabetic patients. Diabetes 50:277–82 62. Butler AE, Janson J, Bonner-Weir S, et al. 2003. Beta-cell deficit and increased betacell apoptosis in humans with type 2 diabetes. Diabetes 52:102–10 63. Nathan DM, Lachin J, Cleary P, et al. 2003. Intensive diabetes therapy and carotid intima-media thickness in type 1 diabetes mellitus. N. Engl. J. Med. 348:2294–303 64. Matthews JB, Ramos E, Bluestone JA. 2003. Clinical trials of transplant tolerance: slow but steady progress. Am. J. Transplant. 3:794–803 65. Masteller EL, Bluestone JA. 2002. Immunotherapy of insulin-dependent diabetes mellitus. Curr. Opin. Immunol. 14:652–59 66. Adams AB, Shirasugi N, Durham MM, et al. 2002. Calcineurin inhibitor-free CD28 blockade-based protocol protects allogeneic islets in nonhuman primates. Diabetes 51:265–70 67. Lenschow DJ, Walunas TL, Bluestone JA. 1996. CD28/B7 system of T cell costimulation. Annu. Rev. Immunol. 14:233–58 68. Kirk AD, Tadaki DK, Celniker A, et al. 2001. Induction therapy with monoclonal antibodies specific for CD80 and CD86 de-

69.

70.

71.

72.

73.

74.

75.

76.

77.

P1: GCE

155

lays the onset of acute renal allograft rejection in non-human primates. Transplantation 72:377–84 Hausen B, Klupp J, Christians U, et al. 2001. Coadministration of either cyclosporine or steroids with humanized monoclonal antibodies against CD80 and CD86 successfully prolong allograft survival after life supporting renal transplantation in cynomolgus monkeys. Transplantation 72:1128–37 Contreras JL, Jenkins S, Eckhoff DE, et al. 2003. Stable alpha- and beta-islet cell function after tolerance induction to pancreatic islet allografts in diabetic primates. Am. J. Transplant. 3:128–38 Knechtle SJ, Pirsch JD, Fechner JH, et al. 2003. Campath-1H induction plus rapamycin monotherapy for renal transplantation: results of a pilot study. Am. J. Transplant. 3:722–30 Murase N, Metes D, Zeevi A, et al. 2003. Immunomonitoring in abdominal whole organ recipients treated with a toleranceenhancing regimen of immunosuppression. Am. J. Transplant. 3(Suppl. 5):A1156 (Abstr.) Spitzer TR, McAfee SL, Dey BR, et al. 2003. Nonmyeloablative haploidentical stem-cell transplantation using anti-CD2 monoclonal antibody (MEDI-507)-based conditioning for refractory hematologic malignancies. Transplantation 75:1748– 51 Herold KC, Hagopian W, Auger JA, et al. 2002. Anti-CD3 monoclonal antibody in new-onset type 1 diabetes mellitus. N. Engl. J. Med. 346:1692–98 Chatenoud L. 2003. CD3-specific antibody-induced active tolerance: from bench to bedside. Nat. Rev. Immunol. 3:123–32 Smith JA, Bluestone JA. 1997. T cell inactivation and cytokine deviation promoted by anti-CD3 mAbs. Curr. Opin. Immunol. 9:648–54 Fandrich F, Lin X, Chai GX, et al. 2002. Preimplantation-stage stem cells induce long-term allogeneic graft acceptance

2 Dec 2003

22:9

156

78.

Annu. Rev. Med. 2004.55:133-156. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

79.

80.

81.

82.

AR

STOCK

AR206-ME55-08.tex

¥

AR206-ME55-08.sgm

LaTeX2e(2002/01/18)

P1: GCE

BLUESTONE

without supplementary host conditioning. Nat. Med. 8:171–78 Hori Y, Rulifson IC, Tsai BC, et al. 2002. Growth inhibitors promote differentiation of insulin-producing tissue from embryonic stem cells. Proc. Natl. Acad. Sci. USA 99:16105–10 Lumelsky N, Blondel O, Laeng P, et al. 2001. Differentiation of embryonic stem cells to insulin-secreting structures similar to pancreatic islets. Science 292:1389–94 Rajagopal J, Anderson WJ, Kume S, et al. 2003. Insulin staining of ES cell progeny from insulin uptake. Science 299:363 Bonner-Weir S. 2000. Islet growth and development in the adult. J. Mol. Endocrinol. 24:297–302 Hayashi KY, Tamaki H, Handa K, et al. 2003. Differentiation and proliferation of

83.

84.

85.

86.

endocrine cells in the regenerating rat pancreas after 90% pancreatectomy. Arch. Histol. Cytol. 66:163–74 Pelengaris S, Khan M, Evan GI. 2002. Suppression of Myc-induced apoptosis in beta cells exposes multiple oncogenic properties of Myc and triggers carcinogenic progression. Cell 109:321–34 Hess D, Li L, Martin M, et al. 2003. Bone marrow-derived stem cells initiate pancreatic regeneration. Nat. Biotechnol. 21:763– 70 Verfaillie CM, Schwartz R, Reyes M, et al. 2003. Unexpected potential of adult stem cells. Ann. NY Acad. Sci. 996:231– 34 Wurmser AE, Gage FH. 2002. Stem cells: cell fusion causes confusion. Nature 416:485–87

2 Dec 2003

22:6

AR

AR206-ME55-09.tex

AR206-ME55-09.sgm

LaTeX2e(2002/01/18) P1: GBC 10.1146/annurev.med.55.091902.105251

Annu. Rev. Med. 2004. 55:157–67 doi: 10.1146/annurev.med.55.091902.105251 c 2004 by Annual Reviews. All rights reserved Copyright °

COCHLEAR IMPLANTATION FOR THE TREATMENT OF DEAFNESS Benjamin J. Copeland1 and Harold C. Pillsbury III2 1

Otolaryngology Associates, 9002 North Meridian, Suite 213, Indianapolis, Indiana 46260 Department of Otolaryngology Head and Neck Surgery, University of North Carolina at Chapel Hill, G0412 Neurosciences Hospital, Chapel Hill, North Carolina 27599–7600; email: [email protected], [email protected]

Annu. Rev. Med. 2004.55:157-167. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

2

Key Words sensorineural hearing loss, cochlear prosthesis, adult, pediatric ■ Abstract Cochlear implants have dramatically changed the treatment and prognosis for patients with profound sensorineural hearing loss. Deaf adults and children can be successfully (re)integrated into the hearing world through a multidisciplinary approach involving otolaryngologists, audiologists, and speech/language pathologists. As the technology of the cochlear prosthesis advances, the candidacy for these devices continues to broaden. This review addresses the basic technology, candidacy criteria, and important issues in the fields of adult and pediatric cochlear implantation. Cost utility and future directions in the treatment of the profoundly hearing impaired are discussed.

INTRODUCTION Significant hearing loss affects one in three adults over the age of 60 and half of adults over the age of 75. In addition, ∼1 of every 1000 children is born deaf (1, 2). The vast majority of these hearing impairments result from the loss of haircell receptors in the inner ear, limiting the cochlea’s ability to transduce sound information from the environment to neural transmissions that can be interpreted by the central nervous system as auditory sensations. Because the nervous elements that transmit information from the cochlea to the brain (spiral ganglion cells) primarily remain functional, it is possible to stimulate the brain with a cochlear prosthesis. This is the basis for the development of the cochlear implant system. The treatment of acquired or congenital deafness has rapidly evolved in the past several decades. The development of an implantable system to restore stimulation to the inner ear has revolutionized the therapeutic approach and prognosis for those afflicted with profound sensorineural hearing loss. Cochlear implant systems have evolved from early experimental wires and posts to sophisticated multichannel devices, approved by the U.S. Food and Drug Administration (FDA), which enable deaf patients to participate fully in the hearing world. 0066-4219/04/0218-0157$14.00

157

2 Dec 2003

22:6

158

AR

AR206-ME55-09.tex

COPELAND

¥

AR206-ME55-09.sgm

LaTeX2e(2002/01/18)

P1: GBC

PILLSBURY

TECHNOLOGICAL AND SURGICAL ASPECTS OF COCHLEAR IMPLANTS All cochlear implants consist of four basic components necessary to deliver acoustic information from the environment to the cochlea (3). The system has three external components and one surgically placed internal device:

Annu. Rev. Med. 2004.55:157-167. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

1. The external receiver is a directional microphone that captures sounds in the environment. This analog information is received by the external speech processor through a direct wire connection. 2. The speech processor employs various coding strategies to convert the analog signal to digital output. This component operates on an external power source (batteries). The newly digitized acoustic information is sent via another direct wire connection to the externally worn transmitting coil. 3. The transmitting coil contains a magnet and a transmitter. The magnet holds the transmitting coil in close proximity to the implanted portion of the device, and the transmitter transfers the digital information across the patient’s skin to the surgically implanted receiver/stimulator. 4. The receiver/stimulator is the only component of the system that is not directly accessible. It is anchored to the postauricular skull and is composed of a magnet (of opposite polarity to the transmitting coil) and a multichannel electrode array that is placed into the cochlea at the time of surgery. The receiver/stimulator receives the electromagnetic signal from the transmitting coil and distributes it to the intracochlear electrodes, thereby delivering it to the surviving nervous tissue, the spiral ganglion cells of the cochlear nerve located in the bony modiolus of the cochlea. The surgery to implant a cochlear implant system involves ∼2–3 h of general anesthesia and is usually performed on an outpatient basis. The typical approach is through a postauricular incision with a gentle S-curve into the scalp behind the superior portion of the ear. The bone of the mastoid cortex and the air-cell system of the mastoid are removed with an otologic drill. Having identified the facial nerve and inner ear, the surgeon approaches the middle ear through the facial recess, taking care not to injure the facial nerve. A separate well for the receiver/ stimulator is drilled in the postauricular skull and the receiver/stimulator is secured in the well. The round window of the cochlea is identified through the facial recess and a small (∼1 mm) hole is drilled into the cochlea to the scala tympani. The electrode array is placed into the cochleostomy. The cochleostomy is then sealed with soft tissue, and the remaining wound is closed in layers and dressed with a compressive headwrap. The surgical sites are allowed to heal for ∼1 month before the internal component is linked to the external ones. The speech processor and receiver/stimulator must then be programmed for the individual to achieve maximal auditory benefit and minimize any discomfort. The extensive process of aural (re)habilitation

2 Dec 2003

22:6

AR

AR206-ME55-09.tex

AR206-ME55-09.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:157-167. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

COCHLEAR IMPLANTATION

P1: GBC

159

requires the input of experienced audiologists and speech and language experts to achieve maximal performance. Several technical concerns arise in patients with cochlear implants. Because the implanted receiver/stimulator is permanently attached to the patient, large electrical currents moving through the patient can damage the delicate circuitry of the device. To this end, patients with cochlear implants are instructed to undergo no procedure involving monopolar cautery. Bipolar cautery is permitted, since the electrical current flows between the two tips of the instrument and not through the patient’s body to the grounding pad. Additionally, magnetic resonance imaging (MRI) modalities may not be performed unless the implanted magnet of the receiver/stimulator is removed first. This can be of special concern when deciding to implant a patient who will probably need future MRI evaluation. The implanted component of the device has evolved from wires to singlechannel devices to the present-day multichannel devices. These advances have improved patient performance with the devices. Perhaps the most important advances have been in the external components of the implants, specifically in the arena of speech processing. The development of improved coding strategies has facilitated delivery of auditory information to the implanted devices and improved patient performance. However, better coding strategies require greater power consumption. Several strategies have been devised to try to extend battery life. The future will undoubtedly hold advances in both the internal and external components of the cochlear implant system, with the possible result of a completely implantable device that could be recharged through the skin if needed.

CANDIDACY ISSUES The FDA has approved the use of cochlear implants for adults and children with bilateral profound sensorineural hearing loss. For postlingually deafened adults (those who had acquired spoken language before losing their hearing), the current criteria for implantation require speech reception to be <40% of words in sentences in the best-aided condition. This often translates into ∼10% correct responses in open-set (no contextual or visual cues) single-word testing. Prelingually deafened adults (those whose hearing loss predated the acquisition of spoken language) are seldom appropriate candidates for implantation because of general poor performance with the device. The FDA-approved minimum age of cochlear implantation in children is 1 year. Pediatric patients should be demonstrated to be profoundly hearing impaired and gain no advantage from an adequate trial (usually 4–6 months) of hearing-aid amplification. Although they have no spoken language abilities, experience with cochlear implantation in these patients shows remarkable performance results. Children are frequently able to obtain high degrees of open-set speech recognition and excellent speech production within several years. The mechanism of this early neural plasticity is poorly understood to date.

2 Dec 2003

22:6

160

AR

AR206-ME55-09.tex

COPELAND

¥

AR206-ME55-09.sgm

LaTeX2e(2002/01/18)

P1: GBC

PILLSBURY

Annu. Rev. Med. 2004.55:157-167. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

CURRENT ISSUES IN ADULT COCHLEAR IMPLANTATION Once controversial, multichannel devices have been demonstrated to provide better speech perception than the single-channel devices (4, 5). The selection criteria for adult implant candidates show a trend of becoming less stringent; individuals with greater levels of residual hearing are being considered for cochlear implantation. This trend is likely due to the improved performance observed in postlingually deafened adults with newer devices and coding strategies. The average performance for implanted adults on open-set word testing is now 40%–50% (5). It is likely that as the average performance level rises, so will the acceptable levels of residual hearing in potential candidates. Traditionally, the surgery to implant an electrode array into the cochlea was widely believed to destroy any native acoustic ability by disrupting the delicate infrastructure and fluids of the cochlea (6–8). The decision to proceed with implantation meant sacrificing the ear’s residual hearing. However, several lines of evidence now suggest that some implant patients retain residual hearing in the implanted ear (9–12) despite the surgical trauma to the cochlea. The mechanisms underlying this hearing preservation are uncertain and as yet there are no predictors of which patients will retain hearing. Numerous studies are investigating the effects of electrode design, surgical technique, and potential mediators of the hearing loss to more predictably preserve residual hearing (13, 14). Adults often present with substantially different levels of hearing loss for each ear and with different durations of deafness for each ear. In numerous studies, the duration of deafness has shown significant correlation with poorer speech perception performance (5, 15). A common, though controversial, practice has been to implant the ear that has the most residual hearing, potentially taking advantage of any “auditory memory” stored in the neural connections. Recent retrospective analysis has demonstrated patients implanted in the poorer-hearing ear perform similarly to those implanted in the better-hearing ear (16). Because the numbers of neural elements in the better-hearing ear are higher on histopathologic studies (17–19), the capacity of the better-hearing ear to perform better has been attributed to undetermined central auditory mechanisms that receive bilateral inputs after the initial synapse of the cochlear nerve in the cochlear nucleus (16). There has been additional investigation into the possibility that wearing a conventional hearing aid in addition to a cochlear implant might contribute to speech perception and sound localization. The hearing aid would be worn in the nonimplanted ear and might provide some bilateral advantage. A very limited study found that this arrangement conferred binaural advantages (20). Although cochlear implants are currently approved for unilateral implantation, trials are investigating the effects of bilateral implantation in adults. The theoretical advantages of bilateral implantation would be improved sound localization and improved hearing in noise due to the “head-shadow effect.” (The ear on the opposite side from a noise source is partially shielded from the noise by the head and therefore has a better signal-to-noise ratio, and presumably better speech reception,

2 Dec 2003

22:6

AR

AR206-ME55-09.tex

AR206-ME55-09.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:157-167. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

COCHLEAR IMPLANTATION

P1: GBC

161

than the ear exposed directly to the noise source.) Early results from bilateral cochlear implant studies have shown improved sound localization and speech perception in noise (21–23). Further results from these trials, with longer followup times, are pending and should reveal advantages and disadvantages to bilateral implantation. Numerous groups have investigated the effects of cochlear implantation in the elderly (24–26). There has been no significant decrement in performance associated with advancing age at implantation. Surgery in this population may carry some additional risks; however, advanced age is not an exclusionary finding when considering a patient for cochlear implantation. The surgical procedure is not particularly painful and is not associated with fluid shifts or a prolonged period of convalescence, so it is well tolerated in the elderly population. The cost of cochlear implantation is significant. The device itself costs $20,000– 25,000. A cost-utility analysis found cochlear implantation comparable to many commonly performed medical procedures, such as coronary artery angioplasty and coronary artery bypass grafting. The cost per quality-adjusted life-year is substantially less than that of hormone replacement therapy, implantable defibrillators, tuberculin screening, knee replacements, and peritoneal or hemodialysis for end-stage renal disease (27, 28). With new investigations into the possibilities of bilateral cochlear implantation, the cost-utility analysis will need to be recalculated for bilateral implantees. Early analysis suggests the cost-effectiveness of the second implant will be significantly lower than the high degree of efficacy demonstrated for the first device (29).

CURRENT ISSUES IN PEDIATRIC COCHLEAR IMPLANTATION One in 1000 children are born with a severe hearing impairment, the cause for which is often unknown. Over 90% of hearing-impaired children are born to parents who both hear, and 97% have at least one hearing parent (30). Despite a lack of speech and language development in congenitally deaf children, cochlear implantation can provide dramatic results in this population. The FDA has approved implantation in children as young as 12 months, although there are anecdotal reports of implantation in younger children in unusual circumstances (31, 32). Given the difficulty in properly testing hearing in such young patients and the necessity for an adequate trial of amplification, it will be difficult to lower the age for pediatric implantation much further. Current protocols for evaluation include imaging studies of the inner ear (typically an MRI scan to demonstrate the presence of a cochlear nerve and possibly a computed tomography scan to define the bony architecture of the inner ear) and extensive audiologic testing during the amplification trial (30). Many syndromes can present as congenital sensorineural hearing loss, and an appropriate work-up for these is directed by individual physical findings and family history (33).

2 Dec 2003

22:6

Annu. Rev. Med. 2004.55:157-167. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

162

AR

AR206-ME55-09.tex

COPELAND

¥

AR206-ME55-09.sgm

LaTeX2e(2002/01/18)

P1: GBC

PILLSBURY

Numerous inner ear malformations can lead to hearing loss in the pediatric population. Deaf children with inner ear malformations, once considered a contraindication to cochlear implantation, have been successfully implanted and can perform well with the device (34–40). The performance of this subset of pediatric patients appears to be at least partially related to the specific malformation. Certain patterns of inner ear malformations are correlated with better performance levels than others (34). Performance in pediatric cochlear implantation has been the subject of much research. A myriad of factors contribute to an individual child’s success. In general, younger age at implantation is associated with higher levels of performance (41– 44), with younger patients achieving open-set speech recognition more rapidly than older children (42). In addition, research has demonstrated that children with cochlear implants who are educated with a strong auditory-oral–based approach perform at a higher level in both speech perception and language production than similar children educated with combinations of oral and manual communication (45, 46). As in adults, previous language experience has been correlated with improved performance with a cochlear implant (47, 48). Another issue in pediatric cochlear implantation concerns implantation in multiply handicapped children. These children often face numerous other challenges to communication aside from their profound sensorineural hearing loss (33); however, they show notable responses to improved environmental awareness of sound provided by a cochlear implant. Implants in these patients are often difficult to program reliably because of their multiple cognitive and physical conditions (43). Recent advances in cochlear implant technology have begun to simplify the process of programming in these complex situations. This process takes advantage of objective data in the form of electrically evoked whole-nerve action potentials from device stimulation to aid in setting program levels (49–52). This novel application shows promise in treating these challenging patients and may confer a greater objectivity to the programming of all implants in pediatric patients. Recently, concerns over the development of meningitis in children with cochlear implants have come to the forefront of the field. The organism mainly responsible for these infections has been Streptococcus pneumoniae. Children with severe hearing loss are at greater risk for developing meningitis even without cochlear implantation (53, 54); however, there have been several recent reports of meningitis in implanted children. The cases appeared to cluster primarily around one particular type of implant device, and its manufacturer has voluntarily ceased production until the concerns can be addressed at the design level (55). Current recommendations are that all children with hearing loss and cochlear implants receive vaccination against Str. pneumoniae every five years for life to help lessen the chances of developing meningitis. A review of issues in pediatric cochlear implantation would not be complete without addressing concerns regarding the impact of cochlear implants on Deaf Culture. Much has been written of the losses to Deaf Culture by the placement of cochlear implants into children (56–58). The arguments of those advocates for

2 Dec 2003

22:6

AR

AR206-ME55-09.tex

AR206-ME55-09.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:157-167. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

COCHLEAR IMPLANTATION

P1: GBC

163

Deaf Culture who would prohibit cochlear implantation in children have been internally contradictory: Deafness is not a disability, yet disability benefits must be preserved for deaf individuals; cochlear implants do not work, yet they work so well that they threaten “genocide” by eliminating deafness. The central issue in this debate appears to be who has the right to determine what is in the best interests of a child: the child’s parents (who most often are hearing) or deaf society as a whole. With over 90% of deaf children born to hearing parents, the availability of cochlear implantation raises the real possibility of dwindling numbers of individuals remaining in the deaf community. This necessarily threatens the existence of the deaf educational system and may require a rethinking of the manner in which this subset of the population is educated.

FUTURE DIRECTIONS The reality of cochlear implants is exciting for those who treat patients with hearing loss, but the future may hold even greater possibilities. Many of the significant advances in implant technology have been made at the programming level. As coding strategies improve, the speech perception abilities of cochlear implant users will advance. These advances are likely to improve hearing in normal and challenging listening environments. The advancement of materials science may ultimately lead to the completely implantable device with no permanent external components. Power will have to be provided by an implantable, reliable, rechargeable battery. Alternatively, the application of advanced piezoelectric technology to cochlear implants may obviate the need for any internal power source at all (59). The development of an implantable microphone for collecting the environmental sound is also a prerequisite for the completely implantable device. As progress is made in predictably preserving residual hearing in ears receiving implants, the level of hearing loss effectively treated by cochlear implantation will broaden. This will offer effective rehabilitation to a larger number of patients with significant levels of hearing loss and will allow selective augmentation of the hearing frequencies that are deficient. Progress in understanding the pathophysiology of hair-cell and nerve-cell loss within the cochlea is critical to the future of cochlear implantation. There appear to be many growth factors that promote the maintenance, rescue, and repair of auditory neurons (60). These factors may also promote the transmission of electrical signals by the implant device through modulation of the native ion channels in the surviving neurons (60). Integrating growth factors and the cochlear implant system would potentially enhance the performance of patients with the device. Finally, as experience with cochlear implant users has grown, so has the educational system required to train these patients to gain maximal benefit from their devices. Strong auditory-oral–based educational systems in which therapists and teachers are familiar with cochlear implants are being developed in response to

2 Dec 2003

22:6

164

AR

AR206-ME55-09.tex

COPELAND

¥

AR206-ME55-09.sgm

LaTeX2e(2002/01/18)

P1: GBC

PILLSBURY

Annu. Rev. Med. 2004.55:157-167. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

this vital need. One such program to educate both children with cochlear implants and the educators who will work with them, the CASTLE program, is in place at the University of North Carolina at Chapel Hill. These programs enhance the gain provided by the implant to both the individual and to society as a whole by maximizing performance at the level of the individual. If this approach is integrated into the mainstream educational system, not only will individual patients benefit, but the cost to society of educating patients with profound hearing loss will fall dramatically.

CONCLUSION Cochlear implants have revolutionized the treatment and prognosis for patients afflicted with profound sensorineural hearing loss. Further work will shed light not only on the pathophysiology of the sense of hearing but also on specific mechanisms of neural plasticity and central nervous system integration that will be applicable to a wide range of medical fields. ACKNOWLEDGMENTS While working on this article, B.J.C. was a fellow in Otology/Neurolotogy at the Department of Otolaryngology Head and Neck Surgery, University of North Carolina at Chapel Hill. The Annual Review of Medicine is online at http://med.annualreviews.org

LITERATURE CITED 1. NIH Consensus Statement. 1993. Early identification of hearing impairment in infants and young children. Int. rep. 11(1):1– 24 2. Marcincuk MC, Roland PS. 2002. Geriatric hearing loss. Geriatrics 57(4):44–59 3. Linstrom CJ. 1998. Cochlear implantation. Prim. Care 25(3):583–617 4. Rubenstein JT, Parkinson WS, Lowder MW, et al. 1998. Single channel to multi channel conversions in adult cochlear implant subjects. Am. J. Otol. 19:461–66 5. Rubenstein JT, Miller CA. 1999. How do cochlear prostheses work? Curr. Opin. Neurobiol. 9:399–404 6. Brimacombe JA, Arndt PE, Staller SJ, et al. 1994. Multichannel cochlear implanta-

7.

8.

9.

10.

tion in adults with severe to profound sensorineural hearing loss. In Cochlear Implants, ed. I Hochmair-Desoyer, E Hochmair, pp. 387–92. Wein, Germ.: Manz Rizer FM. 1998. Postoperative audiometric evaluation of cochlear implant patients. Otolaryngol. Head Neck Surg. 98:203–6 Boggess WJ, Baker JE, Balkany TJ. 1989. Loss of residual hearing after cochlear implantation. Laryngoscope 99:1002–5 Hodges AV, Schloffman J, Balkany T. 1997. Conservation of residual hearing with cochlear implantation. Am. J. Otol. 18:179–83 Dye L, House WF, O’Connor C. 1987. Measurable residual hearing following cochlear implantation. Presented at

2 Dec 2003

22:6

AR

AR206-ME55-09.tex

AR206-ME55-09.sgm

LaTeX2e(2002/01/18)

COCHLEAR IMPLANTATION

11.

Annu. Rev. Med. 2004.55:157-167. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

12.

13.

14.

15.

16.

17.

18.

19.

20.

Am. Acad. Otolaryngol.-Head and Neck Surgery, Chicago, IL, September Lorens A, Geremek A, Walkowiak A, et al. 2000. Residual acoustic hearing in the ear before and after cochlear implantation. In Fourth European Congress of Oto-rhinolaryngology Head and Neck Surgery, ed. K Jahnke, M Fischer, pp. 135–38. Bologna: Monduzzi Skarzynski H, Lorens A, D’Haese P, et al. 2002. Preservation of residual hearing in children and postlingually deafened adults after cochlear implantation. ORL 64:247– 53 Lehnhardt E. 1993. Intracochlear placement of cochlear implant electrodes in soft surgery technique. HNO 41:356–59 (German) Peeters S, Van Immerseel L, Zarowski A, et al. 1998. New developments in cochlear implants. Acta Otorhinolaryngol. Belg. 52:115–27 Blamey P, Arndt P, Bergeron F, et al. 1996. Factors affecting auditory performance of postlinguistically deaf adults using cochlear implants. Audiol. Neurotol. 1:293–306 Francis HW, Yeagle JD, Renna T, et al. 2003. Choice of ear for cochlear implantation in asymmetric hearing loss. Presented at Annu. Meet. Am. Laryngol. Rhinol. Otol. Soc., 106th, Nashville, TN, May 2–6 Fayad J, Linthicum FH, Otto SR, et al. 1991. Cochlear implants: histopathologic findings related to performance in 16 human temporal bones. Ann. Otol. Rhinol. Laryngol. 100:807–11 Nadol JB, Young YS, Glynn RJ. 1989. Survival of spiral ganglion cells in profound sensorineural hearing loss. Ann. Otol. Rhinol. Laryngol. 98:411–16 Incesulu A, Nadol JB. 1998. Correlation of acoustic threshold measures and spiral ganglion cell survival in severe to profound sensorineural hearing loss. Ann. Otol. Rhinol. Laryngol. 107:906–11 Tyler RS, Parkinson AJ, Wilson BS, et al.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

P1: GBC

165

2002. Patients utilizing a hearing aid and a cochlear implant. Ear Hear. 23:98–105 Tyler RS, Gantz BJ, Rubinstein JT, et al. 2002. Three month results with bilateral cochlear implants. Ear Hear. 23:80S– 89S Van Hoesel R, Ramsden R, O’Driscoll M. 2002. Sound direction identification, interaural time delay discrimination, and speech intelligibility advantages in noise for a bilateral cochlear implant user. Ear Hear. 23:137–49 Balkany T, Boggess W, Dinner B. 1988. Binaural cochlear implantation. Laryngoscope 98:1040–43 Chatelin V, Larky J, Polite C, et al. 2003. Cochlear implantation outcome in the elderly. Presented at Annu. Meet. Am. Laryngol. Rhinol. Otol. Soc., 106th, Nashville, TN, May 2–6 Labadie RF, Carrasco VN, Gilmer CH, et al. 2000. Cochlear implant performance in senior citizens. Otolaryngol. Head Neck Surg. 123(4):419–24 Waltzman SB, Cohen NL, Shapiro WH. 1993. The benefits of cochlear implantation in the geriatric population. Otolaryngol. Head Neck Surg. 108(4):329–33 Wyatt JR, Niparko JK, Rothman ML, et al. 1995. Cost effectiveness of the multichannel cochlear implant. Am. J. Otol. 16:52– 62 Palmer CS, Niparko JK, Wyatt JR, et al. 1999. A prospective study of the costutility of the multichannel cochlear implant. Arch. Otolaryngol. Head Neck Surg. 125(11):1221–28 Summerfield AQ, Marshall DH, Barton GR, et al. 2002. A cost-utility scenario analysis of bilateral cochlear implantation. Arch. Otolaryngol. Head Neck Surg. 128:1255–62 Balkany TJ, Hodges A, Miyamoto RT, et al. 2001. Cochlear implants in children. Otolaryngol. Clin. N. Am. 34(2):455–67 Lenarz T, Lesinski-Schiedat RD, Battmer RD, et al. 2000. Cochlear implants in children under the age of one. Presented at

2 Dec 2003

22:6

166

32.

Annu. Rev. Med. 2004.55:157-167. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

AR

AR206-ME55-09.tex

COPELAND

¥

AR206-ME55-09.sgm

LaTeX2e(2002/01/18)

P1: GBC

PILLSBURY

Int. Conf. Cochlear Implants, 6th, Miami Beach, FL, Feb. 3–5 Miyamoto RT. 2003. Language and speech development in deaf children and infants following cochlear implantation. Presented at Annu. Meet. Am. Laryngol. Rhinol. Otol. Soc., 106th, Nashville, TN, May 2–6 Gurtler N, Lalwani AK. 2002. Etiology of syndromic and nonsyndromic sensorineural hearing loss. Otolaryngol. Clin. N. Am. 35(4):891–908 Buchman CA, Copeland BJ, Yu KK, et al. 2003. Cochlear implantation in children with congenital inner ear malformations. Laryngoscope. In press Weber BP, Dillo W, Dietrich B, et al. 1998. Pediatric cochlear implantation in cochlear malformations. Am. J. Otol. 19:747–53 Luntz M, Balkany T, Hodges AV, et al. 1997. Cochlear implants in children with congenital inner ear malformations. Arch. Otolaryngol. Head Neck Surg. 123:974– 77 Slattery WH, Luxford WM. 1995. Cochlear implantation in the congenital malformed cochlea. Laryngoscope 105:1184–87 Tucci DL, Telian SA, Zimmerman-Phillips S, et al. Cochlear implantation in patients with cochlear malformations. Arch. Otolaryngol. Head Neck Surg. 121:833–38 Woolley AL, Jenison V, Stroer BS, et al. 1998. Cochlear implantation in children with inner ear malformations. Ann. Otol. Rhinol. Laryngol. 107:492–500 Harker LA, Vanderheiden S, Veazey D, et al. 1999. Multichannel cochlear implantation in children with large vestibular aqueduct syndrome. Ann. Otol. Rhinol. Laryngol. 108:39–43 Nikolopoulos TP, O’Donoghue GM, Archbold S. 1999. Age at implantation, its importance in pediatric cochlear implantation. Laryngoscope 109:595–99 Waltzman SB, Cohen NL. 1998. Cochlear implantation in children younger than 2 years old. Am. J. Otol. 19:158–62 Hassanzadeh S, Farhadi M, Daneshi A, et al. 2002. The effects of age on au-

44.

45.

46.

47.

48.

49.

50.

51.

52.

53.

ditory speech perception in cochlear implanted prelingually deaf children. Otolaryngol. Head Neck Surg. 126:524– 27 Richter B, Eissele S, Laszig R, et al. 2002. Receptive and expressive language skills of 106 children with a minimum of 2 years’ experience in hearing with a cochlear implant. Int. J. Ped. Otorhinolaryngol. 64:111–25 Hodges AV, Dolan-Ash M, Balkany TJ, et al. 1999. Speech perception results in children with cochlear implants. Otolaryngol. Head Neck Surg. 121:31–34 Robbins AM, Bollard PM, Green J. 1999. Language development in children implanted with the CLARION cochlear implant. Ann. Otol. Rhinol. Laryngol. 108:113–18 Rubenstein JT, Parkinson WS, Tyler RS, et al. 1999. Residual speech recognition and cochlear implant performance. Am. J. Otol. 20:445–52 Zwolen TA, Zimmerman-Phillips S, Ashbaugh CJ, et al. 1997. Cochlear implantation of children with minimal open set speech recognition skills. Ear Hear. 18: 240–51 Brown CJ, Abbas PJ. 1990. Electrically evoked whole nerve action potentials. J. Acoust. Soc. Am. 88:1385–91 Brown CJ, Hughes ML, Lopez SM, et al. 1999. Relationship between EABR and thresholds and levels used to program the CLARION speech processor. Ann. Otol. Rhinol. Laryngol. 108:50–57 Brown CJ. 2000. Neural response telemetry: pediatric applications. Presented at Int. Conf. Cochlear Implants, 6th, Miami Beach, FL, Feb. 3–5 Shallop JK, Facer GW, Peterson A. 1999. Neural response telemetry with the Nucleus CI24M cochlear implant. Laryngoscope 109:1755–59 Ohlms LA, Edwards MS, Mason EO. 1990. Recurrent meningitis and Mondini dysplasia. Arch. Otolaryngol. Head Neck Surg. 116(5):608–12

2 Dec 2003

22:6

AR

AR206-ME55-09.tex

AR206-ME55-09.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:157-167. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

COCHLEAR IMPLANTATION 54. Parisier SC, Birken EA. 1976. Recurrent meningitis secondary to idiopathic oval window CSF leak. Laryngoscope 86(10): 1503–15 55. Cohen NL. 2003. Meningitis in cochlear implant recipients: the North American experience. Presented at Annu. Meet. Am. Laryngol. Rhinol. Otol. Soc., 106th, Nashville, TN, May 2–6 56. Balkany TJ, Hodges AV, Goodman KW. 1996. Ethics of cochlear implantation in young children. Otolaryngol. Head Neck Surg. 114:748–55

P1: GBC

167

57. Lane H, Grodin M. 1997. Ethical issues in cochlear implant surgery. Kennedy Inst. Eth. J. 7(3):231–51 58. Lane H. 1995. Letter to the Editor. Am. J. Otol. 16:393–99 59. Mukherjee N, Roseman RD, Willging JP. 2000. The piezoelectric cochlear implant. J. Biomed. Mater. Res. 53:181– 87 60. Marzella PL, Clark GM. 1999. Growth factors, auditory neurons and cochlear implants. Acta Otolaryngol. 119:407– 12

5 Jan 2004

11:52

AR

AR206-ME55-10.tex

AR206-ME55-10.sgm

LaTeX2e(2002/01/18) P1: GBC 10.1146/annurev.med.55.091902.105243

Annu. Rev. Med. 2004. 55:169–78 doi: 10.1146/annurev.med.55.091902.105243 c 2004 by Annual Reviews. All rights reserved Copyright ° First published online as a Review in Advance on Sept. 15, 2003

DRUG-ELUTING STENTS T. Cooper Woods and Andrew R. Marks

Annu. Rev. Med. 2004.55:169-178. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Department of Physiology and Cellular Biophysics, Department of Medicine, Center for Molecular Cardiology, Columbia University College of Physicians and Surgeons, New York, New York 10032; email: [email protected]

Key Words sirolimus, taxol, restenosis, rapamycin, paclitaxel ■ Abstract Advances in catheter and stent design have made stent implantation the standard coronary angioplasty procedure. Unfortunately, in-stent restenosis continues to plague this procedure, with the optimum binary restenosis rates reaching ∼10% to 20%. In the past few years, it has become clear that in-stent restenosis is largely due to the migration and proliferation of vascular smooth muscle cells to form a neointima. To address this issue, stents coated with drug-delivery vehicles have been developed to deliver antiproliferative therapeutics. Two drugs, rapamycin and taxol, have been the lead compounds for testing the idea of a drug-eluting stent. These drugs have been successful largely because of the solid mechanistic understanding of their effects and extensive preclinical examination. The result of these years of work is that the rapamycin-coated stent entered the US market in April of 2003, and the taxol-coated stent appears poised to follow soon.

INTRODUCTION The treatment of coronary artery disease, which remains a leading cause of mortality in the developed world, has undergone revolutionary changes in the past decade. Multivessel coronary artery disease, once considered a surgical problem, is now routinely treated with intravascular therapy including balloon angioplasty and coronary artery stent implantation (1–3). A major limitation of this approach has been in-stent restenosis, which occurs in ∼30% of stented coronary arteries and often results in recurrent ischemic episodes that require repeat intravascular procedures and/or surgery (4). A major advance in the treatment of coronary artery disease is the development of drug-eluting stents that dramatically reduce the incidence of in-stent restenosis to <5%. The first drug-eluting stent was approved in April 2002 for use in Europe and in April 2003 for use in the United States by the Food and Drug Administration (FDA). The development of this rapamycin (sirolimus)–coated stent represents the application of fundamental studies that focused on understanding the role of vascular smooth muscle cell (VSMC) proliferation and migration in stent restenosis (5). New knowledge about how VSMC proliferation and migration are regulated based on the use of rapamycin as a molecular 0066-4219/04/0218-0169$14.00

169

5 Jan 2004

11:52

170

AR

AR206-ME55-10.tex

WOODS

¥

AR206-ME55-10.sgm

LaTeX2e(2002/01/18)

P1: GBC

MARKS

probe (6–8) led directly to its development as the first successful therapeutic for stent restenosis.

Annu. Rev. Med. 2004.55:169-178. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

CAUSES OF IN-STENT RESTENOSIS The development of percutaneous transluminal coronary (balloon) angioplasty (PTCA) changed the treatment of coronary artery disease due to atherosclerosis (2, 3). Unfortunately, revascularization by PTCA is limited by elastic recoil and thrombosis at the site of angioplasty, as well as VSMC growth in response to the balloon-induced injury that results in neointimal formation. Within minutes following balloon deflation, the coronary artery undergoes elastic recoil due to contraction of the elastin fibers of the inner and external elastic laminae, causing up to a 40% lumen loss. A thrombotic response triggered by endothelial denudation, and medial dissection due to the mechanical injury of PTCA, lead to platelet adherence and aggregation on the stent struts and the exposed subendothelial surface. Antiplatelet therapy has essentially eliminated the problem of thrombosis following stent implantation (9, 10). However, the response to injury following PTCA with or without stent implantation, combined with the permanent presence of a foreign body (the metal stent) in the vasculature, stimulates the growth of neointima that causes restenosis. Neointimal formation is a complicated process involving the recruitment of inflammatory cells to the site of injury, the migration of VSMC from the media to the intima, and the proliferation of these cells (11– 13). The final mechanism of this response is the late remodeling of the vessel. These processes combine to result in a biological response to PTCA plus stent implantation that impairs the utility of this therapeutic approach. The use of coronary artery stenting following balloon angioplasty essentially solved the problem of elastic recoil and vessel remodeling (14). Moreover, thanks to advances in stent design over the past decade, the rate of restenosis when stents are placed in large vessels with short lesions has been reduced to ∼10%–20% (Figure 1). In contrast, balloon angioplasty without stenting has a restenosis rate of ∼50% (15–19). Thus, under less than optimal conditions with respect to the type of lesion and the anatomy of the target vessel, the restenosis rate is still substantial and severely limits the use of stents in patients with complex coronary artery disease or left main coronary artery disease, as well as in diabetics, in whom the incidence of restenosis is essentially twice that seen in nondiabetics (20). When stents are used to treat lesions in smaller-diameter vessels, the in-stent restenosis rate increases to >20% in nondiabetics and >40% in diabetics (20). The substantial rate of stent restenosis led to the redesigning of stents in hopes of improving their antithrombotic properties and thereby reducing the rate of restenosis. Inorganic elements (e.g., gold), polymers (e.g., polylactic acid and fibrin), and immobilized drugs (e.g., heparin and abciximab) have been successfully applied to stents. However, these agents have generally caused slight decreases in thrombosis without substantially decreasing in-stent restenosis.

5 Jan 2004

11:52

AR

AR206-ME55-10.tex

AR206-ME55-10.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:169-178. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

DRUG-ELUTING STENTS

P1: GBC

171

When stent redesign alone failed to prevent restenosis, attention turned toward antithrombotics. Research into antithrombotic therapies led to great excitement over the possibility that blocking platelet aggregation would eliminate in-stent restenosis. One such agent, the antibody abciximab, which targets glycoprotein IIb/IIIa (GP IIb/IIIa) on platelets, seemed to have the potential to reduce in-stent restenosis to insignificant levels (9). Instead, it became clear that although coronary stenting combined with antithrombotic therapies had essentially eliminated the problems of elastic recoil, thrombus formation, and vessel remodeling, in-stent restenosis remained a major problem. Although in retrospect it is obvious that VSMC growth and migration are the culprits with respect to in-stent restenosis, when the magnitude of the problem began to emerge in the late 1990s it was firmly believed that elastic recoil and thrombosis were the most important therapeutic targets.

INHIBITORS OF VSMC PROLIFERATION Examination of occluded stents showed evidence of VSMC growth inside the devices. Meanwhile, studies using rapamycin as a molecular probe had revealed that it was a potent inhibitor of VSMC proliferation and migration (6–8). Nevertheless, there was great resistance to the use of rapamycin for cardiovascular purposes because it was also a potent immunosuppressant. That drawback, combined with the lingering hope that antiplatelet therapy would solve the problem, delayed the development of rapamycin for prevention of stent restenosis. Extensive data have shown that numerous agents, including the antineoplastic drug taxol, inhibit VSMC proliferation. The final piece of the puzzle was the demonstration in a pig model that systemic rapamycin could potently inhibit neointimal formation following PTCA (21). This first demonstration of dramatic inhibition of neointimal formation in a large animal model, combined with detailed understanding of the mechanism by which rapamycin achieved this exciting result, provided the impetus for the development of stents coated with a biopolymer that contains rapamycin.

MECHANISMS OF ACTION Rapamycin is a 31-membered macrocycle lactone similar in structure to the immunosuppressant FK506 (tacrolimus). The major intracellular receptor for both of these compounds is the immunophilin FKBP12 (FK506 binding protein, 12 kDa). The rapamycin-FKBP12 complex binds to the mammalian target of rapamycin [mTOR, also known as the FKBP12 and rapamycin-associated protein (FRAP) or rapamycin and FKBP12 target (RAFT)] (22). Rapamycin blocks cell-cycle progression at the G1-to-S transition in VSMC (6, 7). Rapamycin increases levels of the cyclin-dependent kinase inhibitor p27Kip1. According to the “threshold model” of cell-cycle regulation, commitment to the cell cycle is controlled by a careful balance between cyclin-dependent kinases and their inhibitors (23, 24).

5 Jan 2004

11:52

Annu. Rev. Med. 2004.55:169-178. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

172

AR

AR206-ME55-10.tex

WOODS

¥

AR206-ME55-10.sgm

LaTeX2e(2002/01/18)

P1: GBC

MARKS

During cell-cycle progression, cyclin E/cdk2 complexes phosphorylate the retinoblastoma protein, leading to its dissociation from the transcription factor E2F. E2F regulates expression of genes that encode proteins required for the G1-to-S transition. Thus, increases in p27Kip1, which binds and inactivates the cyclin E/cdk2 complex, inhibit cell-cycle progression. The mechanism by which rapamycin treatment blocks growth-factor–mediated repression of p27Kip1 levels is unclear. Further evidence of a key role for p27Kip1 in the ability of rapamycin to inhibit VSMC growth comes from studies showing that rapamycin-resistant cell lines are unable to upregulate p27Kip1 in response to rapamycin treatment, and cells from p27 knockout mice are partially resistant to growth inhibition by rapamycin (25). Rapamycin has other cellular effects that probably contribute to its ability to prevent stent restenosis. Inhibition of mTOR by rapamycin-FKBP12 inhibits phosphorylation of p70S6K and the eukaryotic initiation factor 4E (eIF-4E) binding protein, 4E-BP1 (26, 27). Phosphorylation activates p70S6K, which in turn activates the S6 ribosome involved in protein translation. Phosphorylation of 4E-BP1 dissociates it from the eIF-4E. When 4E-BP1 binds to eIF-4E, it blocks assembly of the eIF-4F complex and the recruitment of the 40S ribosome to the 7-methyl-GTP cap of mRNA. Phosphorylation of the ribosomal protein S6 by p70S6K leads to translation of the 50 TOP mRNAs that are required for cell growth. Regulation of protein translation through these two pathways may contribute to mTOR’s role in modulating cellular proliferation and/or migration, but the precise contribution of inhibition of protein translation to the ability of rapamycin to prevent stent restenosis remains to be determined (28). Taxol induces its antirestenotic effects by stabilizing the microtubule assembly (29, 30). Microtubules are polymers assembled from tubulin dimers. The assembly and disassembly of microtubules is maintained in a dynamic equilibrium that is regulated according to cell-cycle phase. Microtubules are best known as the major component of the mitotic spindle apparatus, but they also play a role in cell shape, migration, and growth-factor signaling (31–34). Taxol binds to the Nterminal 31 residues of the beta-tubulin subunit in the microtubule and promotes polymerization of microtubules. At subnanomolar concentrations this leads to stabilization of microtubules, whereas higher concentrations lead to greater mass and number of microtubules. Thus, treatment with taxol shifts the equilibrium between microtubules and the tubulin dimers toward assembly, interrupting cell division and inhibiting other microtubule-dependent processes (30). At all phases of the cell cycle bundles of disorganized microtubules are formed, and in mitosis an abnormal number of asters form, preventing progression through M-phase (35, 36).

LOCAL DRUG DELIVERY One of the common explanations for the failure of systemically administered therapeutics to prevent restenosis is the inability to achieve the required dose at the site of injury. The goal of drug-eluting stents is to place the right amount of drug

5 Jan 2004

11:52

AR

AR206-ME55-10.tex

AR206-ME55-10.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:169-178. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

DRUG-ELUTING STENTS

P1: GBC

173

at the site of injury at the time of injury. Equally important is that for drugs with a narrow therapeutic window it may be difficult to achieve desired levels in the artery without unwanted side effects. For these reasons, drug-eluting stents have been aggressively pursued as a therapeutic modality. The development of drug-eluting stents required solving the important problem of how to apply sufficient concentrations of drug to the stent with appropriate release kinetics. To date, only rapamycin and taxol have achieved significant efficacy in preventing stent restenosis in clinical trials, and only rapamycin-coated stents have been approved for clinical use. There is a long list of agents that have failed to prevent in-stent restenosis or shown only limited efficacy compared to rapamycin and taxol.

CLINICAL RESULTS Radioactive Stents An alternative to coating stents with drugs that prevent in-stent restenosis is the delivery of radiation at the site of the stent. However, this approach has achieved only limited success because of high rates of restenosis at the edges of the stent. Radioactive stents were the first form of local delivery used to prevent in-stent restenosis (37). The first clinical trial of 32P-radioactive stents was the IRIS (IsoStent for Restenosis Intervention Study) 1A trial, in which 57 patients received stents with low 32P activity (0.5–1.0 µC) of radiation (38). Since then, several more trials with doses up to 21 µC have been conducted. These studies showed that although radioactive stents prevented neointimal growth within the stent, restenosis due to neointimal formation in the areas immediately proximal and distal to the stent was highly prevalent (39). No large multicenter trial has assessed the efficacy of radioactive stents and their future is uncertain.

Rapamycin The Cordis CypherTM stent is a BX VELOCITYTM stent coated with a 50:50 mixture of polyethylenevinylacetate and polybutylmethacrylate containing 140 µg/mm2 of rapamycin. The first-in-man study with this stent yielded dramatic results (40). The stents were implanted in 30 patients using angiography and intravascular ultrasound. At four-month follow-up, the restenosis rate was 0%, with no evidence of in-stent or edge restenosis. Furthermore, no major adverse cardiovascular events (death, myocardial infarction, stent thrombosis, or target lesion revascularization) were observed at eight months. A two-year follow-up study indicated that in-stent lumen dimension remained unchanged and restenosis “catchup” did not appear to be occurring (41). The next study, RAVEL (Randomized Study with the Sirolimus-Eluting BX VELOCITYTM balloon-expandable stent), compared 120 patients receiving the drug-eluting stents to 118 receiving bare metal

5 Jan 2004

11:52

Annu. Rev. Med. 2004.55:169-178. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

174

AR

AR206-ME55-10.tex

WOODS

¥

AR206-ME55-10.sgm

LaTeX2e(2002/01/18)

P1: GBC

MARKS

stents (42). Patients were divided into three groups based on the reference diameter of the vessel prior to intervention. At one year, binary restenosis rates were 26.6% for patients who received the bare stent and 0% for those with the rapamycineluting stents in all groups. The rate of major cardiac events at one year was 5.8% for the sirolimus-eluting stent group versus 28.8% for the bare stent group (42). The SIRIUS trial (a Multicenter, Randomized, Double-Blind Study of Sirolimus-Coated BX VELOCITYTM Balloon-Expandable Stent in the Treatment of Patients with De Novo Coronary Artery Lesions), was the pivotal trial for achieving FDA approval of rapamycin-coated stents for use in the United States. This study, completed in 2002, demonstrated a reduction of in-stent restenosis from 35.4% to 3.2% in 1101 patients (43).

Taxol Numerous clinical trials have evaluated taxol for treatment of in-stent restenosis. A NIRTM stent is coated with a proprietary polymer system loaded with either a “low” dose of 1.0 µg/mm2 or a “moderate” dose of 2.0 µg/mm2 of taxol. In the initial study, TAXUS I, 61 patients received either a slow-release low-dose stent or a bare stent. The restenosis rate for this trial was 0% for the drug-eluting stent and 10.3% for the bare stent (44). This study was followed by the TAXUS II trial, which tested two different polymer configurations, both loaded with the low dose (45). The first cohort tested a polymer exhibiting a slow-release profile. At six months, the restenosis rates were found to be reduced to 2.3% from 17.9% when the taxol-eluting stent was used (45). The second cohort, which examined a polymer with a slightly higher release rate, showed a similar reduction in restenosis, from 20.2% to 4.6% (46). Recently, the TAXUS III trial reported efficacy in the treatment of in-stent restenosis (47). Multiple clinical studies are now examining the efficacy of taxol-eluting stents using various drug-release formulations on different types of stents. Most clinical studies have shown efficacy in the prevention of in-stent restenosis; however, the DELIVER trial, which enrolled 1043 patients and compared treatment with a MULTI-LINK PENTATM stent coated with 3 µg/mm2 of taxol to treatment with the bare stent, reported an insignificant difference in restenosis rates (16% versus 21%, respectively) (48). Nevertheless, a taxol-eluting stent was approved for use in Europe in February 2003.

Other Drugs The SCORE trial (Study to Compare Restenosis Rate Between QueST and QuaDSQP2) trial compared the use of a QuanamTM stent coated with the taxol derivative 7-hexanoyltaxol to a bare stent (49). The drug was actually released from sleeves wrapped around the stent. The study showed a significant reduction in restenosis, but thrombotic complications occurred in 8% of patients as compared to 0% for the bare stent. The ACTION trial (Actinomycin-Eluting Stent Improves Outcomes by

5 Jan 2004

11:52

AR

AR206-ME55-10.tex

AR206-ME55-10.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:169-178. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

DRUG-ELUTING STENTS

P1: GBC

175

Reducing Neointimal Hyperplasia) was testing actinomycin D, an antineoplastic agent, but the trial was halted when preliminary analysis showed an unacceptably high rate of target lesion revascularization in patients treated with the drugeluting stent (50). Two trials are examining tacrolimus, EVIDENT (EndoVascular Investigation Determining the Safety of a New Tacrolimus-Eluting Stent Graft) and PRESENT (Preliminary Safety Evaluation of Nanoporous Tacrolimus-Eluting Stents), and have reported encouraging initial safety data. Recent results from the FUTURE trial (First Used to Underscore Reduction in Restenosis with Everolimus) indicated that a stent coated with Everolimus (a macrolide from the same family as rapamycin) achieved a statistically significant reduction in neointimal volume compared to the bare stent at six months (51).

FUTURE CONSIDERATIONS Rapamycin- and taxol-eluting stents are the culmination of the decades of research into stent and balloon design, polymer and materials chemistry, vascular biology, and pharmacokinetic modeling. A major challenge for the future will be to understand why some patients develop restenosis despite rapamycin or taxol (or other agents) on coated stents. In particular, as noted above, diabetics appear to be relatively resistant to rapamycin. Because diabetics comprise a substantial portion of patients with coronary artery disease, it is important to understand the molecular basis for their resistance to rapamycin in terms of development of in-stent restenosis. Reducing the persistent high rates of stent restenosis in diabetics may require the development of new agents for drug-eluting stents. Another important issue is whether all stents should be drug-eluting stents, which cost substantially more than bare metal stents, or whether some lesions (e.g., short lesions in large vessels in nondiabetic patients?) may be successfully treated with noncoated stents without subjecting patients to unacceptable risks of in-stent restenosis. ACKNOWLEDGMENT A.R.M. is a Distinguished Clinical Scientist Awardee of the Doris Duke Charitable Foundation. The Annual Review of Medicine is online at http://med.annualreviews.org

LITERATURE CITED 1. Williams DO, Holubkov R, Yeh W, et al. 2000. Percutaneous coronary intervention in the current era compared with 1985– 1986: the National Heart, Lung, and Blood

Institute Registries. Circulation 102:2945– 51 2. Berger E, Williams DO, Reinert S, et al. 1986. Sustained efficacy of percutaneous

5 Jan 2004

11:52

176

3.

4.

Annu. Rev. Med. 2004.55:169-178. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

AR

AR206-ME55-10.tex

WOODS

¥

AR206-ME55-10.sgm

LaTeX2e(2002/01/18)

P1: GBC

MARKS

transluminal coronary angioplasty. Am. Heart J. 111:233–36 Block PC. 1980. Percutaneous transluminal coronary angioplasty. Am. J. Roentgenol. 135:955–59 Holmes DR Jr. 2003. State of the art in coronary intervention. Am. J. Cardiol. 91:50A– 53A Marx SO, Marks AR. 2001. Bench to bedside: the development of rapamycin and its application to stent restenosis. Circulation 104:852–55 Jayaraman T, Marks AR. 1993. Rapamycin-FKBP12 blocks proliferation, induces differentiation, and inhibits cdc2 kinase activity in a myogenic cell line. J. Biol. Chem. 268:25385–88 Marx SO, Jayaraman T, Go LO, et al. 1995. Rapamycin-FKBP inhibits cell cycle regulators of proliferation in vascular smooth muscle cells. Circ. Res. 76:412–17 Poon M, Marx SO, Gallo R, et al. 1996. Rapamycin inhibits vascular smooth muscle cell migration. J. Clin. Invest. 98:2277– 83 Topol EJ, Lincoff AM, Kereiakes DJ, et al. 2002. Multi-year follow-up of abciximab therapy in three randomized, placebocontrolled trials of percutaneous coronary revascularization. Am. J. Med. 113:1–6 Harrington RA, Kleiman NS, KottkeMarchant K, et al. 1995. Immediate and reversible platelet inhibition after intravenous administration of a peptide glycoprotein IIb/IIIa inhibitor during percutaneous coronary intervention. Am. J. Cardiol. 76:1222–27 Welt FG, Rogers C. 2002. Inflammation and restenosis in the stent era. Arterioscler. Thromb. Vasc. Biol. 22:1769–76 Ross R. 1999. Atherosclerosis—an inflammatory disease. N. Engl. J. Med. 340:115– 26 Ross R. 1999. Atherosclerosis is an inflammatory disease. Am. Heart J. 138:S419– 20 Mudra H, Regar E, Klauss V, et al. 1997. Serial follow-up after optimized

15.

16.

17.

18.

19.

20.

21.

22.

23.

ultrasound-guided deployment of PalmazSchatz stents. In-stent neointimal proliferation without significant reference segment response. Circulation 95:363–70 Heuser R, Houser F, Culler SD, et al. 2000. A retrospective study of 6,671 patients comparing coronary stenting and balloon angioplasty. J. Invasive. Cardiol. 12:354– 62 Serruys PW, de Jaegere P, Kiemeneij F, et al. 1994. A comparison of balloonexpandable-stent implantation with balloon angioplasty in patients with coronary artery disease. Benestent Study Group. N. Engl. J. Med. 331:489–95 Fischman DL, Leon MB, Baim DS, et al. 1994. A randomized comparison of coronary-stent placement and balloon angioplasty in the treatment of coronary artery disease. Stent Restenosis Study Investigators. N. Engl. J. Med. 331:496–501 Kimmel SE, Localio AR, Brensinger C, et al. 2000. Effects of coronary stents on cardiovascular outcomes in broadbased clinical practice. Arch. Intern. Med. 160:2593–99 Rogers CDK. 2002. Drug-eluting stents: role of stent design, delivery vehicle, and drug selection. Rev. Cardiovasc. Med. 3:S10–15 Suselbeck T, Latsch A, Siri H, et al. 2001. Role of vessel size as a predictor for the occurrence of in-stent restenosis in patients with diabetes mellitus. Am. J. Cardiol. 88:243–47 Gallo R, Padurean A, Chesebro JH, et al. 1998. Inhibition of intimal thickening after balloon angioplasty in porcine coronary arteries by rapamycin. Circulation 99:2164– 70 Sabers CJ, Martin MM, Brunn GJ, et al. 1995. Isolation of a protein target of the FKBP12-rapamycin complex in mammalian cells. J. Biol. Chem. 270:815– 22 Bresnahan WA, Boldogh I, Ma T, et al. 1996. Cyclin E/Cdk2 activity is controlled by different mechanisms in the G0 and G1

5 Jan 2004

11:52

AR

AR206-ME55-10.tex

AR206-ME55-10.sgm

LaTeX2e(2002/01/18)

DRUG-ELUTING STENTS

24.

25.

Annu. Rev. Med. 2004.55:169-178. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

phases of the cell cycle. Cell Growth Differ. 7:1283–90 Koff A, Polyak K. 1995. p27KIP1, an inhibitor of cyclin-dependent kinases. Prog. Cell Cycle Res. 1:141–47 Luo Y, Marx SO, Kiyokawa H, et al. 1996. Rapamycin resistance tied to defective regulation of p27Kip1. Mol. Cell. Biol. 16:6744–51 Brunn GJ, Hudson CC, Sekulic A, et al. 1997. Phosphorylation of the translational repressor PHAS-I by the mammalian target of rapamycin. Science 277:99–101 Burnett PE, Barrow RK, Cohen NA, et al. 1998. RAFT1 phosphorylation of the translational regulators p70 S6 kinase and 4EBP1. Proc. Natl. Acad. Sci. USA 95:1432– 37 Marx SO, Marks AR. 1999. Cell cycle progression and proliferation despite 4BP1 dephosphorylation. Mol. Cell Biol. 19: 6041–47 Schiff PB, Fant J, Horwitz SB. 1979. Promotion of microtubule assembly in vitro by taxol. Nature 277:665–67 Schiff PB, Horwitz SB. 1980. Taxol stabilizes microtubules in mouse fibroblast cells. Proc. Natl. Acad. Sci. USA 77:1561– 65 Dugina VB, Alexandrova AY, Lane K, et al. 1995. The role of the microtubular system in the cell response to HGF/SF. J. Cell Sci. 108(Pt. 4):1659–67 Rosette C, Karin M. 1995. Cytoskeletal control of gene expression: depolymerization of microtubules activates NF-kappa B. J. Cell Biol. 128:1111–19 Pletjushkina OJ, Ivanova OJ, Kaverina IN, et al. 1994. Taxol-treated fibroblasts acquire an epithelioid shape and a circular pattern of actin bundles. Exp. Cell Res. 212:201–8 Rosania GR, Swanson JA. 1996. Microtubules can modulate pseudopod activity from a distance inside macrophages. Cell Motil. Cytoskelet. 34:230–45 Rowinsky EK, Donehower RC, Jones RJ, et al. 1988. Microtubule changes and cyto-

36.

37.

38.

39.

40.

41.

42.

43.

44.

P1: GBC

177

toxicity in leukemic cell lines treated with taxol. Cancer Res. 48:4093–100 De Brabander M, Geuens G, Nuydens R, et al. 1981. Taxol induces the assembly of free microtubules in living cells and blocks the organizing capacity of the centrosomes and kinetochores. Proc. Natl. Acad. Sci. USA 78:5608–612 Hafeez N. 2002. Prevention of coronary restenosis with radiation therapy: a review. Clin. Cardiol. 25:313–22 Carter AJ, Fischell TA. 1998. Current status of radioactive stents for the prevention of in-stent restenosis. Int. J. Radiat. Oncol. Biol. Phys. 41:127–33 Albiero R, Adamian M, Kobayashi N, et al. 2000. Short- and intermediate-term results of 32P radioactive beta-emitting stent implantation in patients with coronary artery disease: the Milan Dose-Response Study. Circulation 101:18–26 Sousa JE, Costa MA, Abizaid A, et al. 2001. Lack of neointimal proliferation after implantation of sirolimus-coated stents in human coronary arteries: a quantitative coronary angiography and three-dimensional intravascular ultrasound study. Circulation 103:192–95 Sousa JE, Costa MA, Sousa AG, et al. 2003. Two-year angiographic and intravascular ultrasound follow-up after implantation of sirolimus-eluting stents in human coronary arteries. Circulation 107:381–83 Morice MC, Serruys PW, Sousa JE, et al. 2002. A randomized comparison of a sirolimus-eluting stent with a standard stent for coronary revascularization. N. Engl. J. Med. 346:1773–80 Moses JW, Leon MB, Popma JJ, et al. 2002. SIRIUS: a U.S. multicenter, randomized, double-blind study of the SIRolImUSeluting stent in de novo native coronary lesions. Presented at Transcatheter Cardiovascular Therapeutics Conf., Washington, DC, Sep. 24–28 Grube E, Silber S, Hauptmann KE, et al. 2003. TAXUS I: six- and twelve-month results from a randomized, double-blind trial

5 Jan 2004

11:52

Annu. Rev. Med. 2004.55:169-178. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

178

AR

AR206-ME55-10.tex

WOODS

¥

AR206-ME55-10.sgm

LaTeX2e(2002/01/18)

P1: GBC

MARKS

on a slow-release paclitaxel-eluting stent for de novo coronary lesions. Circulation 107:38–42 45. Columbo A. 2002. TAXUS-II international study cohort I: slow-release formulation— six month results intent to treat analysis. Presented at Transcatheter Cardiovascular Therapeutics Conf., Washington, DC, Sep. 24–28 46. Columbo A. 2002. TAXUS-II international study cohort II: moderate-release formulation—six month results intent to treat analysis. Presented at Transcatheter Cardiovascular Therapeutics Conf., Washington, DC, Sep. 24–28 47. Tanabe K, Serruys PW, Grube E, et al. 2003. TAXUS III Trial: in-stent restenosis treated with stent-based delivery of pacli-

48.

49.

50.

51.

taxel incorporated in a slow-release polymer formulation. Circulation 107:559– 64 2003. Guidant Reports Preliminary Results of DELIVER Clinical Trial. Guidant Corp., Indianapolis, IN de la Fuente LM, Miano J, Mrad J, et al. 2001. Initial results of the Quanam drug eluting stent (QuaDS-QP-2) registry (BARDDS) in human subjects. Catheter Cardiovasc. Interv. 53:480–88 2002. Guidant Halts Further Development on Actinomycin-D Drug Eluting Stent Program. Guidant Corp., Indianapolis, IN 2003. Guidant Reports Positive Six-Month Clinical Results of Everolimus Eluting Coronary Stent. Guidant Corp., Indianapolis, IN

sample.qxd

1/5/2004

3:36 PM

Page 1

C-1

Annu. Rev. Med. 2004.55:169-178. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

DRUG-ELUTING STENTS

Figure 1 The response to percutaneous transluminal coronary angioplasty (PTCA). Beginning with a coronary artery that has already stenosed owing to plaque formation, PTCA is performed either with or without stent implantation. With PTCA alone, the artery is subject to elastic recoil, thrombus formation, neointimal growth, and late vessel remodeling. If a stent is implanted, the effects of elastic recoil and late vessel remodeling are limited, but the artery is still at risk of neointimal hyperplasia. If a drug-eluting stent is implanted, neointimal hyperplasia can be nearly abolished by blocking cell-cycle progression.

2 Dec 2003

22:4

AR

AR206-ME55-11.tex

AR206-ME55-11.sgm

LaTeX2e(2002/01/18) P1: GBC 10.1146/annurev.med.55.091902.103801

Annu. Rev. Med. 2004. 55:179–89 doi: 10.1146/annurev.med.55.091902.103801 c 2004 by Annual Reviews. All rights reserved Copyright ° First published online as a Review in Advance on Sept. 22, 2003

NEW APPROACHES TO HEMODIALYSIS Andreas Pierratos

Annu. Rev. Med. 2004.55:179-189. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Department of Medicine, Humber River Regional Hospital, University of Toronto, Toronto, Ontario, Canada; email: [email protected]

Key Words daily hemodialysis, nocturnal hemodialysis, home hemodialysis ■ Abstract Treatment of end-stage renal disease with dialysis is characterized by high mortality rate, low quality of life, and high cost. Recent randomized controlled studies showed that increasing the dialysis dose above the currently recommended levels in thrice-weekly hemodialysis does not decrease the patient mortality rate. Short daily hemodialysis or daily home nocturnal hemodialysis are promising alternatives. Both improve quality of life and control blood pressure and anemia; nocturnal hemodialysis additionally controls serum phosphates without phosphate binders, allows a free diet, and corrects sleep apnea. Although the direct cost of daily hemodialysis is higher than that of conventional hemodialysis, the cost of total care, especially when delivered at home, seems to be lower. Further confirmation of these results is important. Restructuring of the dialysis reimbursement system is necessary to make the use of daily hemodialysis possible. Hemofiltration techniques, sorbents, and the renal tubular assist device may also help change the current grim statistics.

BACKGROUND Renal replacement options for patients with end-stage renal disease include hemodialysis, peritoneal dialysis, or rarely a preemptive transplantation from a living donor. Quality of life of patients on dialysis is poor, about 50% of the level of the nondialysis population, as is their vocational rehabilitation potential (1). This poor quality of life results from the combined effects of the uremic syndrome, the dialysis procedure itself, and the presence or development of comorbidities. Despite some recent improvement, patient mortality is very high (15%–25% per year), with cardiovascular disease being the primary cause of death (2). The uremic syndrome is usually ascribed to the retention of uremic toxins, which have molecular weights of <100 to >12,000 Da (3). The toxicity of the larger-molecular-weight solutes probably contributes significantly to the uremic syndrome, but this has not been fully proven.

What Affects Solute Removal? Solute removal by dialysis depends on the concentration gradient of the unbound solute across the dialysis membrane (dialyzer or peritoneum), the permeability of 0066-4219/04/0218-0179$14.00

179

2 Dec 2003

22:4

Annu. Rev. Med. 2004.55:179-189. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

180

AR

AR206-ME55-11.tex

AR206-ME55-11.sgm

LaTeX2e(2002/01/18)

P1: GBC

PIERRATOS

the membrane to the specific solute, and the length of dialysis. Parameters that decrease the efficacy of solute removal by dialysis include large molecular size, high protein binding, and slow diffusion of the solutes to the blood compartment. The blood concentration of the dialyzable molecules decreases during dialysis and therefore the rate of solute removal by dialysis decreases exponentially with time. The early part of the dialysis session provides most of the solute removal. Shortening the duration and increasing the frequency of dialysis increases the overall efficiency (4). However, prolonged dialysis sessions, despite the decrease in the solute removal rate with time, do increase removal of solutes. This is particularly relevant to larger molecules, which are characterized by slow removal rate (5, 6). The dialysis filter (dialyzer) membranes are divided into two broad groups, low and high flux, depending on their permeability characteristics (7). High-flux dialyzers are more effective in removing larger molecules as well as water and electrolytes (8). There is evidence but no proof that high-flux membranes are associated with improved patient outcomes (9). Compared to the diffusive dialysis process, ultrafiltration (fluid removal during dialysis), a convective process, is associated with more effective removal of larger molecules because of the “solvent drug” effect. This quality is exploited by the technique of hemofiltration, which involves forced removal of large amounts of fluid from the intravascular space and replacement by a similar volume of an appropriate electrolytic solution. The efficacy of removal of small molecules by hemofiltration is lower than by conventional hemodialysis. A combination of hemodialysis and hemofiltration in the form of hemodiafiltration offers the highest solute removal of both small and large molecules (10).

Dialysis Dose and Outcomes What constitutes adequate or optimal dialysis is still an unanswered question. A dimensionless parameter, urea Kt/V (where K is the urea dialysis clearance, t is dialysis time, and V is the volume of urea distribution in the body), has been popularized as a measure of dialysis dose. Unfortunately, Kt/V does not correlate with outcomes across different dialysis methods. A modification, the standard Kt/V (stdKt/V), tries to rectify this problem. StdKt/V is presumed to be the same in all dialysis methods with the same outcomes. The National Cooperative Dialysis Study (NCDS), a randomized prospective study done in the 1980s, concluded that low dialysis dose with a urea Kt/V below 1 per dialysis session in a thrice-weekly treatment schedule correlated with worse outcomes (11). As a result of the NCDS study, highly efficient dialysis (high K) allowed the time on dialysis (low t) to be shortened without changing urea Kt/V. This was perceived to have a negative impact on dialysis outcomes. In the mid 1990s, the Dialysis Outcome Quality Initiative (DOQI) guidelines sponsored by the National Kidney Foundation recommended a urea Kt/V of 1.2 per dialysis session (12). Most recently, the Hemodialysis (HEMO) study, a large prospective randomized controlled study supported by the National Institutes of Health,

2 Dec 2003

22:4

AR

AR206-ME55-11.tex

AR206-ME55-11.sgm

LaTeX2e(2002/01/18)

NEW APPROACHES TO HEMODIALYSIS

P1: GBC

181

examined the effect of a urea Kt/V of 1.65 versus the DOQI standard of 1.25. It also evaluated the possible benefit of using high-flux dialysis membranes to enable more efficient removal of larger molecules. The results, published in 2003 (13), indicated no further beneficial effect either from increased dialysis dose or from use of high-flux dialyzers in the thrice-weekly hemodialysis.

Annu. Rev. Med. 2004.55:179-189. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Decline in Home Hemodialysis Although the utilization of home hemodialysis was as high as 90% in the 1970s, it has declined to about 2% over time. The main reason for the decline has been the creation of many accessible satellite dialysis units and the increase in patient comorbidities that prevent hemodialysis at home (14).

Decline in Peritoneal Dialysis Peritoneal dialysis in the form of CAPD (continuous ambulatory peritoneal dialysis) was introduced in the late 1970s. Despite its simplicity and low cost, its popularity has declined because of a perception of decreased efficacy over time. The Canada-USA (CANUSA) Peritoneal Dialysis Study Group showed increasing patient survival on CAPD with a higher dialysis dose (15). Reanalysis of the CANUSA data revealed that the benefit was due mainly to residual kidney function rather to dialysis-derived clearance (16). The recently published ADEMEX (ADEquacy of Peritoneal Dialysis in MEXico) study, a prospective randomized controlled study, failed to show any increase in patient survival with increasing dialysis dose (17). Therefore, one would not expect significant improvement in patient outcomes by increasing the dialysis dose on CAPD.

At the Crossroads In summary, patients in end-stage renal disease have high mortality and low quality of life. Increasing the dose in either thrice-weekly hemodialysis or CAPD does not improve outcomes. Therefore, new options are needed.

ALTERNATIVES TO CONVENTIONAL DIALYSIS Long Intermittent Hemodialysis Long intermittent hemodialysis, a regimen of 8 h dialysis three times per week, has been known since the early days of dialysis (18–20). It offers excellent bloodpressure control and improved patient survival. Blood pressure is controlled by decreasing the extracellular fluid volume and dietetic sodium restriction. More than 90% of patients come off antihypertensives. However, despite excellent bloodpressure control, left ventricular hypertrophy still persists (21). Laurent and colleagues reported a 10-year survival of 75%, and this high figure was only partially related to the early inclusion of healthier patient cohorts (22). Despite the long

2 Dec 2003

22:4

182

AR

AR206-ME55-11.tex

AR206-ME55-11.sgm

LaTeX2e(2002/01/18)

P1: GBC

PIERRATOS

Annu. Rev. Med. 2004.55:179-189. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

dialysis and the increased β 2-microglobulin removal, they found no prevention of dialysis-associated amyloidosis (23). Long intermittent hemodialysis improves phosphate removal, but many patients still require the use of phosphate binders. A modified regimen of dialysis every other day rather than three times per week provides better hemodynamic stability (24). Intermittent nocturnal hemodialysis in the hospital or at home has attracted more attention recently in North America (25). A significant advantage of long intermittent hemodialysis is its relatively low incremental cost compared to conventional hemodialysis.

Daily Hemodialysis SHORT DAILY HEMODIALYSIS The first clinical report on short daily hemodialysis was published in 1969 (26). All studies have reported significant improvement in uremic symptomatology and better control of anemia and hypertension. Systematic use of daily hemodialysis has been reported by Buoncristiani’s group in Italy over the past 25 years (27, 28). It includes hemodialysis for 2–2.5 h six or seven days per week. It can be performed either in in-center facilities or at home. Interest in daily hemodialysis has increased over the past few years (29). Increasing the frequency of dialysis while maintaining the same number of hours per week enhances solute removal (30). Both small- and larger-molecule removal is affected (31, 32). StdKt/V increases from ∼2 on conventional hemodialysis to almost 3, representing almost a 50% increase in the dose. Short daily hemodialysis is better tolerated than conventional hemodialysis. Standardized instruments show significant improvement in quality of life after patients are converted to short daily hemodialysis (33–35). The improvement is quite striking, especially in patients with significant comorbidities. Blood-pressure control improves on short daily hemodialysis, leading to regression of cardiac hypertrophy (28, 36). This is achieved by a gradual decrease in the post-dialysis weight (“dry weight”) and therefore of the extracellular fluid volume by dialysis which is well tolerated. There is evidence of improved nutrition characterized by increased appetite, weight gain, increase in serum albumin, and other nutritional parameters (37). Anemia control, enabling up to a 40% decrease in the dose of erythropoietin, was reported in most but not all studies (28, 38). Other reported advantages include lower levels of advanced glycated end products (AGE) (39) and protein bound toxins (32). Despite increased phosphate removal, serum phosphate and the need for phosphate binders are not significantly affected because the improved appetite leads to greater phosphate intake (40).

DAILY HOME NOCTURNAL HEMODIALYSIS Daily home nocturnal hemodialysis is performed at night during sleep, usually for 8 h, six or seven nights a week using a standard hemodialysis machine. The method was introduced in 1994 (41, 42). Most centers practice “live” remote monitoring of the machines during hemodialysis through a telephone or internet connection (43, 44). Use of the internet allows centralized monitoring of large geographical areas and cost containment. Lower

2 Dec 2003

22:4

AR

AR206-ME55-11.tex

AR206-ME55-11.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:179-189. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

NEW APPROACHES TO HEMODIALYSIS

P1: GBC

183

dialysis intensity and blood flow allow the use of only one dialysis needle (intermittent bidirectional flow). Because of the high level of hemodynamic stability, a partner is not considered necessary. Safety features include moisture sensors on the floor and on the fistula needles to detect possible blood or dialysate leaks or dislodgement of the fistula needle. The dose of dialysis is higher than in the other regimens. StdKt/V increases to >5 as compared to 2 on conventional hemodialysis and 3 on short daily hemodialysis (45). Therefore, this modality is particularly suitable for large patients. The fourfold increase in the total dialysis duration leads to a fourfold increase in removal of middle-sized molecules, including β 2-microglobulin (46, 47). Phosphate removal is twice as effective as in conventional hemodialysis (48). Phosphate binders are discontinued, diet is unrestricted, and many patients require addition of phosphate into the dialysate. The calcium-phosphorus product, which represents a risk factor for arterial calcification and cardiovascular morbidity and mortality, normalizes (49). Massive extraosseous calcifications can dissolve (50). High dialysate calcium or vitamin D analogues can be used without the fear of elevated calcium/phosphorus product, resulting in improvement in hyperparathyroid bone disease (51). The quality of life improves and patients are enthusiastic about the improvement. Brissenden et al. (52) report improved scores on the Sickness Impact Profile, Beck Depression Index, and three categories of the SF-36 instrument. Diet is unrestricted, including phosphate, potassium, sodium, and fluid intake. Except for vitamin supplements, patients need no dialysis-related medication. Blood pressure normalizes with the use of almost no antihypertensives (53). This is achieved by decreasing the patients’ “dry weight” and sequential discontinuation of their medications. Echocardiographic data also indicate a decrease in systemic peripheral resistance. Regression of left ventricular hypertrophy follows the conversion to daily home nocturnal hemodialysis (53), and there is evidence of improved endothelial function (54). Furthermore, left ventricular function improves significantly (55). Patients previously ineligible for kidney transplantation have been added to the transplantation list. The method is associated with lower serum homocysteine levels than conventional hemodialysis (56). Most groups report a decrease in erythropoietin dose by about 30% and an increase in hemoglobin (33, 57). Improved appetite and weight gain are seen on daily home nocturnal hemodialysis (42, 58). Polysomnography in sleep studies showed normalization of the high apnea hypopnea index seen in sleep apnea, highly prevalent among end-stage renal disease patients (59). Periodic limb movements persisted (60). This is the only dialysis modality associated with improvement in sleep apnea. COST OF DAILY HEMODIALYSIS The direct cost of daily hemodialysis is higher than that of conventional hemodialysis. However, in a retrospective study including medication and hospitalization costs, Mohr et al. (33) showed savings of $6400

2 Dec 2003

22:4

184

AR

AR206-ME55-11.tex

AR206-ME55-11.sgm

LaTeX2e(2002/01/18)

P1: GBC

PIERRATOS

Annu. Rev. Med. 2004.55:179-189. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

(U.S.) per patient-year for in-center short daily hemodialysis and about $9500 for home daily hemodialysis. In a prospective controlled study comparing daily home nocturnal hemodialysis to in-center conventional hemodialysis, McFarlane et al. (61) calculated a mean annual savings of about $7000 (U.S.) per patient with the daily home regimen (p = 0.003). The main reasons for the benefit were decreased labor cost, hospitalization rate, and medication use. The cost utility was $50,041/QALY (quality-adjusted life year) for daily home nocturnal hemodialysis and $83,556/QALY for conventional hemodialysis (62). This study demonstrated that daily home nocturnal hemodialysis offers improved quality at lower cost. OBSTACLES The higher direct cost of daily hemodialysis and the existing method of dialysis reimbursement represent the most important obstacles for its utilization. Currently, only three dialysis sessions per week are reimbursed in the United States, Canada, and other countries. Savings in the form of lower hospitalization rates and decreased cost of medications, including erythropoietin, are not realized by the providers, who have to absorb the increased direct cost of the method. In the United States, a significant portion of the profit margin of dialysis units comes from the use of the very medications that daily dialysis makes less necessary. The need for prospective randomized controlled studies prior to allocation of funding has been debated. The National Institutes of Health, in collaboration with the Center for Medicare and Medicaid Services, has recently announced a request for applications for pilot prospective randomized controlled studies to compare either short daily hemodialysis or daily nocturnal hemodialysis to in-center conventional hemodialysis (63). The results of this pilot study will be available in three years and may lead to larger studies. If the results are very encouraging, they may lead to adequate funding for daily hemodialysis. Another significant obstacle for daily hemodialysis is that most dialysis units lack experience with home hemodialysis.

Hemofiltration/Hemodiafiltration Although hemofiltration and hemodiafiltration have been utilized since the 1970s, there is renewed interest in these methods (64). They are better at larger-molecule removal than hemodialysis is (65). Their main clinical advantages are superior hemodynamic stability, fewer symptoms during treatment, and improved bloodpressure control (66). The main disadvantage of hemodiafiltration is the significant expense of producing large volumes of sterile replacement solution. Newer techniques lower costs by implementing “online” production of endotoxin-free replacement solution from dialysate. This will probably increase utilization of hemofiltration and hemodiafiltration (67). Short or nocturnal daily hemofiltration and hemodiafiltration would combine the advantages of hemofiltration and the daily schedule (64, 68). Clinical data on these methods are awaited with interest.

2 Dec 2003

22:4

AR

AR206-ME55-11.tex

AR206-ME55-11.sgm

LaTeX2e(2002/01/18)

NEW APPROACHES TO HEMODIALYSIS

P1: GBC

185

Sorbents

Annu. Rev. Med. 2004.55:179-189. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Sorbents have been used for renal replacement therapy since the 1970s (69). New sorbent materials have been shown to remove larger-molecular-weight toxins such as β 2-microglobulin, angiogenin, leptin, and cytokines (70). The sorbent columns are added to the regular hemodialysis circuit. The future use of sorbents as a replacement or as an adjunct to renal replacement therapy depends on clinical results from ongoing studies and on the added cost of this treatment.

Renal Tubule Assist Device Renal replacement therapy in its current forms does not replace the metabolic and endocrinologic functions of the renal tubule cells. Recently, renal tubule cells have been successfully cultured and a renal tubule-cell assist device containing human cells in an extracorporeal circuit has been created. This has been shown to replace filtration, metabolic, and endocrinologic functions in acutely uremic dogs (71). Its preliminary use in humans in the setting of acute renal failure in the intensive care unit has been successful, increasing hemodynamic stability and clinical improvement (72).

THE FUTURE The need for alternative solutions to the problem of poor patient outcomes is imperative. Daily hemodialysis in both forms is the most promising new alternative. If the use of daily dialysis is reimbursed adequately, most symptomatic patients dialyzed in in-center facilities will be converted to daily hemodialysis. Progressively, daily hemodialysis is likely to become the standard of care for in-center facilities. Adequate reimbursement for daily hemodialysis at home will create incentives for industry to create more patient-friendly hemodialysis machines and for patients to adopt home hemodialysis. Even using the current hemodialysis machines, the penetration of home hemodialysis will probably increase initially to 10% and then to 30%–50% of the dialysis population. Both short daily hemodialysis and daily nocturnal hemodialysis will be utilized at home; more patients will prefer nocturnal hemodialysis because of the higher dialysis dose, better hemodynamic profile, phosphate control, and time management. If the appropriate funding for daily hemodialysis is delayed, the utilization of long hemodialysis three times a week or every other day will gain momentum. Long overnight dialysis will be practiced in the in-center facilities but mainly at home. This will also increase the utilization of home hemodialysis. The use of hemofiltration and hemodiafiltration, either in the conventional thrice-weekly treatment on in a daily form, will depend on the cost of the method.

2 Dec 2003

22:4

186

AR

AR206-ME55-11.tex

AR206-ME55-11.sgm

LaTeX2e(2002/01/18)

P1: GBC

PIERRATOS

The Annual Review of Medicine is online at http://med.annualreviews.org

Annu. Rev. Med. 2004.55:179-189. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

LITERATURE CITED 1. Evans RW, Manninen DL, Garrison LPJ, et al. 1985. The quality of life of patients with end-stage renal disease. N. Engl. J. Med. 312:553–59 2. United States Renal Data System— USRDS. 2003. Mortality and causes of death. http://www.usrds.org/2002/pdf/ h.pdf 3. Vanholder R, De Smet R, Glorieux G, et al. 2003. Review on uremic toxins: classification, concentration, and interindividual variability. Kidney Int. 63:1934–43 4. Clark WR, Leypoldt JK, Henderson LW, et al. 1999. Quantifying the effect of changes in the hemodialysis prescription on effective solute removal with a mathematical model. J. Am. Soc. Nephrol. 10:601–9 5. Henderson LW, Clark WR, Cheung AK. 2001. Quantification of middle molecular weight solute removal in dialysis. Semin. Dial. 14:294–99 6. Pierratos A. 2001. Effect of therapy time and frequency on effective solute removal. Semin. Dial. 14:284–88 7. Clark WR. 2000. Hemodialyzer membranes and configurations: a historical perspective. Semin. Dial. 13:309–11 8. Leypoldt JK, Cheung AK. 1996. Removal of high-molecular-weight solutes during high-efficiency and high-flux haemodialysis. Nephrol. Dial. Transplant. 11:329–35 9. Hakim RM, Held PJ, Stannard DC, et al. 1996. Effect of the dialysis membrane on mortality of chronic hemodialysis patients. Kidney Int. 50:566–70 10. Canaud B, Bosc JY, Leray H, et al. 1998. On-line haemodiafiltration: state of the art. Nephrol Dial. Transplant. 13(Suppl. 5):3– 11 11. Gotch FA, Sargent JA. 1985. A mechanistic analysis of the National Cooperative Dialysis Study (NCDS). Kidney Int. 28:526– 34 12. National Kidney Foundation. 1997. NKF-

13.

14.

15.

16.

17.

18. 19.

20.

21.

22.

DOQI clinical practice guidelines for hemodialysis adequacy. Am. J. Kidney Dis. 30:S15–S66; Erratum. 1997. Am. J. Kidney Dis. 30(4 Suppl. 3) Eknoyan G, Beck GJ, Cheung AK, et al. 2002. Effect of dialysis dose and membrane flux in maintenance hemodialysis. N. Engl. J. Med. 347:2010–19 Blagg CR. 1996. A brief history of home hemodialysis. Adv. Renal Replace. Ther. 3:99–105 Canada-USA (CANUSA) Peritoneal Dialysis Study Group. 1996. Adequacy of dialysis and nutrition in continuous peritoneal dialysis: association with clinical outcomes. J. Am. Soc. Nephrol. 7:198–207 Bargman JM, Thorpe KE, Churchill DN. 2001. Relative contribution of residual renal function and peritoneal clearance to adequacy of dialysis: a reanalysis of the CANUSA study. J. Am. Soc. Nephrol. 12:2158–62 Paniagua R, Amato D, Vonesh E, et al. 2002. Effects of increased peritoneal clearances on mortality rates in peritoneal dialysis: ADEMEX, a prospective, randomized, controlled trial. J. Am. Soc. Nephrol. 13:1307–20 Barber S, Appleton DR, Kerr DN. 1975. Adequate dialysis. Nephron 14:209–27 Shaldon S, Oakley JJ. 1966. Experience with regular haemodialysis in the home. Br. J. Urol. 38:616–20 Charra B, Calemard E, Ruffet M, et al. 1992. Survival as an index of adequacy of dialysis. Kidney Int. 41:1286–91 Covic A, Goldsmith DJ, Georgescu G, et al. 1996. Echocardiographic findings in long-term, long-hour hemodialysis patients. Clin. Nephrol. 45:104–10 Laurent G, Charra B. 1998. The results of an 8 h thrice weekly haemodialysis schedule. Nephrol. Dial. Transplant. 13(Suppl.):125–31

2 Dec 2003

22:4

AR

AR206-ME55-11.tex

AR206-ME55-11.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:179-189. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

NEW APPROACHES TO HEMODIALYSIS 23. Laurent G, Calemard E, Charra B. 1988. Dialysis related amyloidosis. Kidney Int. Suppl. 24:S32–S34 24. Scribner BH, Twardowski ZJ. 2000. The case for every-other-day dialysis. Hemodialysis Int. 4:5–7 25. Kurella M, Hung A, Tichy M, et al. 2002. Intermittent nocturnal in-center hemodialysis: UCSF–Mt Zion Experience. J. Am. Soc. Nephrol. 13:410A (Abstr.) 26. DePalma JR, Pecker EA, Maxwell MH. 1999. A new automatic coil dialyser system for ‘daily’ dialysis. Semin. Dial. 12:410– 18 27. Buoncristiani U, Quintaliani G, Cozzari M, et al. 1988. Daily dialysis: long-term clinical metabolic results. Kidney Int. 24:S137– 40 28. Woods JD, Port FK, Orzol S, et al. 1999. Clinical and biochemical correlates of starting “daily” hemodialysis. Kidney Int. 55:2467–76 29. Pierratos A. 1998. Daily hemodialysis: Why the renewed interest? Am. J. Kidney Dis. 32:S76–S82 30. Galland R, Traeger J, Delawari E, et al. 1999. Daily hemodialysis versus standard hemodialysis: urea TAC, TAD, weekly eKt/v, Std(Kt/v) and nPCR. Home Hemodial. Int. 3:33–36 31. Goldfarb-Rumyantzev AS, Cheung AK, Leypoldt JK. 2002. Computer simulation of small-solute and middle-molecule removal during short daily and long thrice-weekly hemodialysis. Am. J. Kidney Dis. 40:1211– 18 32. Fagugli RM, De Smet R, Buoncristiani U, et al. 2002. Behavior of non-proteinbound and protein-bound uremic solutes during daily hemodialysis. Am. J. Kidney Dis. 40:339–47 33. Mohr PE, Neumann PJ, Franco SJ, et al. 2001. The case for daily dialysis: its impact on costs and quality of life. Am. J. Kidney Dis. 37:777–89 34. Kooistra MP, Vos J, Koomans HA, et al. 1998. Daily home haemodialysis in The Netherlands: effects on metabolic con-

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

P1: GBC

187

trol, haemodynamics, and quality of life. Nephrol. Dial. Transplant. 13:2853–60 Buoncristiani U, Cairo G, Giombini L, et al. 1989. Dramatic improvement of clinicalmetabolic parameters and quality of life with daily dialysis. Int. J. Artif. Organs 12:133–36 Fagugli RM, Reboldi G, Quintaliani G, et al. 2001. Short daily hemodialysis: blood pressure control and left ventricular mass reduction in hypertensive hemodialysis patients. Am. J. Kidney Dis. 38:371–76 Galland R, Traeger J, Arkouche W, et al. 2001. Short daily hemodialysis rapidly improves nutritional status in hemodialysis patients. Kidney Int. 60:1555–60 Fagugli RM, Buoncristiani U, Ciao G. 1998. Anemia and blood pressure correction obtained by daily hemodialysis induce a reduction of left ventricular hypertrophy in dialysed patients. Int. J. Artif. Organs 21:429–31 Floridi A, Antolini F, Galli F, et al. 2002. Daily haemodialysis improves indices of protein glycation. Nephrol. Dial. Transplant. 17:871–78 Traeger J, Galland R, Ferrier ML, et al. 2002. Optimal control of phosphatemia by short daily hemodialysis. J. Am. Soc. Nephrol. 13:410A–411A (Abstr.) Uldall R, Ouwendyk M, Francoeur R, et al. 1996. Slow nocturnal home hemodialysis at the Wellesley Hospital. Adv. Ren. Replace Ther. 3:133–36 Pierratos A, Ouwendyk M, Francoeur R, et al. 1998. Nocturnal hemodialysis: threeyear experience. J. Am. Soc. Nephrol. 9:859–68 Pierratos A. 1999. Nocturnal home haemodialysis: an update on a 5-year experience. Nephrol. Dial. Transplant. 14:2835–40 Hoy CD. 2001. Remote monitoring of daily nocturnal hemodialysis. Hemodial. Int. 4:8–12 Gotch FA. 2001. Is Kt/V urea a satisfactory measure for dosing the newer dialysis regimens? Semin. Dial. 14:15–17 Clark WR, Leypoldt JK, Henderson LW,

2 Dec 2003

22:4

188

47.

Annu. Rev. Med. 2004.55:179-189. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

48.

49.

50.

51.

52.

53.

54.

55.

56.

AR

AR206-ME55-11.tex

AR206-ME55-11.sgm

LaTeX2e(2002/01/18)

P1: GBC

PIERRATOS et al. 1999. Quantifying the effect of changes in the hemodialysis prescription on effective solute removal with a mathematical model. J. Am. Soc. Nephrol. 10:601–9 Raj DS, Ouwendyk M, Francoeur R, et al. 2000. β 2-microglobulin kinetics in nocturnal haemodialysis. Nephrol. Dial. Transplant. 15:58–64 Mucsi I, Hercz G, Uldall R, et al. 1998. Control of serum phosphate without any phosphate binders in patients treated with nocturnal hemodialysis. Kidney Int. 53:1399–404 Block GA, Hulbert-Shearon TE, Levin NW, et al. 1998. Association of serum phosphorus and calcium × phosphate product with mortality risk in chronic hemodialysis patients: a national study. Am. J. Kidney Dis. 31:607–17 Kim SJ, Goldstein M, Szabo T, et al. 2003. Resolution of massive uremic tumoral calcinosis with daily nocturnal home hemodialysis. Am. J. Kidney Dis. 41:E12 Pierratos A, Hercz G, Sherrard DJ, et al. 2001. Calcium, phosphorus metabolism and bone pathology on long term nocturnal hemodialysis. J. Am. Soc. Nephrol. 12:274A (Abstr.) Brissenden JE, Pierratos A, Ouwendyk M, et al. 1998. Improvements in quality of life with nocturnal hemodialysis. J. Am. Soc. Nephrol. 9:168A (Abstr.) Chan CT, Floras JS, Miller JA, et al. 2002. Regression of left ventricular hypertrophy after conversion to nocturnal hemodialysis. Kidney Int. 61:2235–39 Chan CT, Harvey PJ, Pierratos A, et al. 2002. Impact of nocturnal hemodialysis on blood pressure and endothelially mediated vasodilation. J. Am. Soc. Nephrol. 13:60A (Abstr.) Chan C, Floras JS, Miller JA, et al. 2002. Improvement in ejection fraction by nocturnal haemodialysis in end-stage renal failure patients with coexisting heart failure. Nephrol. Dial. Transplant. 17:1518–21 Friedman AN, Bostom AG, Levey AS, et al. 2002. Plasma total homocysteine lev-

57.

58.

59.

60.

61.

62.

63.

64.

65.

66.

67.

els among patients undergoing nocturnal versus standard hemodialysis. J. Am. Soc. Nephrol. 13:265–68 McFarlane PA, Pierratos A, Redelmeier DA. 2002. Cost savings of home nocturnal versus conventional in-center hemodialysis. Kidney Int. 62:2216–22 Pierratos A, Ouwendyk M, Rassi M. 1999. Total body nitrogen increases on nocturnal hemodialysis. J. Am. Soc. Nephrol. 10:299A (Abstr.) Hanly PJ, Pierratos A. 2001. Improvement of sleep apnea in patients with chronic renal failure who undergo nocturnal hemodialysis. N. Engl. J. Med. 344:102–7 Hanly PJ, Gabor JY, Chan C, et al. 2003. Daytime sleepiness in patients with CRF: impact of nocturnal hemodialysis. Am. J. Kidney Dis. 41:403–10 McFarlane PA, Pierratos A, Redelmeier DA. 2002. Cost savings of home nocturnal versus conventional in-center hemodialysis. Kidney Int. 62:2216–22 McFarlane PA, Bayuni A, Pierratos A, et al. 2003. The quality of life and cost utility of home nocturnal and conventional in-center hemodialysis. Kidney Int. 64:1004–11 National Institutes of Health. 2003. Frequent hemodialysis clinical trials. http:// grants1.nih.gov/grants/guide/rfa-files/RF A-DK-03-005.html Ledebo I. 2002. On-line hemofiltration. Old concept—new approach. Contrib. Nephrol. 221–26 Henderson LW, Colton CK, Ford CA. 1975. Kinetics of hemodiafiltration. II. Clinical characterization of a new blood cleansing modality. J. Lab. Clin. Med. 85:372– 91 Altieri P, Sorba G, Bolasco P, et al. 2001. Predilution haemofiltration—the Second Sardinian Multicentre Study: comparisons between haemofiltration and haemodialysis during identical Kt/V and session times in a long-term cross-over study. Nephrol. Dial. Transplant. 16:1207– 13 Ledebo I. 2002. On-line preparation of

2 Dec 2003

22:4

AR

AR206-ME55-11.tex

AR206-ME55-11.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:179-189. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

NEW APPROACHES TO HEMODIALYSIS solutions for dialysis: practical aspects versus safety and regulations. J. Am. Soc. Nephrol. 13(Suppl. 1):S78–S83 68. Zimmerman DL, Swedko P, Posen G, et al. 2002. Daily hemofiltration with a simplified method of delivery. J. Am. Soc. Nephrol. 13:61A (Abstr.) 69. Lewin AJ, Greenbaum MA, Gordon A, et al. 1972. Current status of the clinical application of the Redy R dialysate delivery system. Proc. Clin. Dial. Transplant. Forum 2:52–56 70. Winchester JF. 2002. Sorbent hemoperfu-

P1: GBC

189

sion in end-stage renal disease: an in-depth review. Adv. Renal Replace. Ther. 9:19– 25 71. Fissell WH, Lou L, Abrishami S, et al. 2003. Bioartificial kidney ameliorates gram-negative bacteria-induced septic shock in uremic animals. J. Am. Soc. Nephrol. 14:454–61 72. Humes HD, Weitzel WF, Bartlett RH, et al. 2003. Renal cell therapy is associated with dynamic and individualized responses in patients with acute renal failure. Blood Purif. 21:64–71

2 Dec 2003

22:3

AR

AR206-ME55-12.tex

AR206-ME55-12.sgm

LaTeX2e(2002/01/18) P1: GBC 10.1146/annurev.med.55.091902.103831

Annu. Rev. Med. 2004. 55:191–207 doi: 10.1146/annurev.med.55.091902.103831 c 2004 by Annual Reviews. All rights reserved Copyright ° First published online as a Review in Advance on Aug. 18, 2003

EMERGING INFECTIOUS THREATS TO THE BLOOD SUPPLY

Annu. Rev. Med. 2004.55:191-207. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Roger Y. Dodd and David A. Leiby Transmissible Diseases Department, Jerome H. Holland Laboratory for the Biomedical Sciences, American Red Cross, Rockville, Maryland 20855; email: [email protected], [email protected]

Key Words blood safety, disease, interventions, transfusion, transmission ■ Abstract During the past 15 years, it has become clear that new agents and new strains of existing agents continue to emerge worldwide as protagonists of infectious disease. These emerging agents pose threats not only to the general human population but also to recipients of blood transfusions. Indeed, the modern era of blood safety perhaps began with the recognition of HIV as an emerging agent transmissible by blood transfusion. Today, emerging infectious agents that pose a threat to the blood supply are not limited to viruses, but include bacterial, protozoan, and prion agents. Preventing the transmission of these new agents by blood transfusion is often problematic, as the available tools may be inadequate. It is certain, however, that new agents will continue to emerge as threats to blood safety and these agents are likely to require novel approaches to prevent their transmission.

INTRODUCTION Emerging Infections Over the past 10 or 15 years, it has become abundantly clear that the war against infectious diseases has not been won. Rather, new infections are being recognized, new and virulent strains of existing agents are appearing, and the geographic ranges and prevalence rates of well-known agents are expanding. These phenomena, collectively termed emerging infections, are attracting a great deal of attention and concern and there are numerous publications and even a specialized journal on the topic. This review discusses the impact of emerging infections on the safety of the blood supply, with particular reference to the United States (Table 1).

Blood Safety A number of infectious agents are known to be transmissible by blood transfusion and vigorous effort goes into preventing or minimizing such transmission. All donated blood products are tested for evidence of infection with the hepatitis B and 0066-4219/04/0218-0191$14.00

191

2 Dec 2003

22:3

AR

192

DODD

AR206-ME55-12.tex

¥

AR206-ME55-12.sgm

LaTeX2e(2002/01/18)

P1: GBC

LEIBY

TABLE 1 Blood safety: emerging and other agents of current concern in the U. S.

Annu. Rev. Med. 2004.55:191-207. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Interventions Agent

Transmissibilitya

Questionsb

Tests

HBV

Known, high

Hx, risk

Yes

HCV

Known, high

Hx, risk

Yes

HIV 1-2

Known, high

Hx, risk

Yes

HTLV I/II

Known

WNV

Known

CMV

Known

T. pallidum

Known, rare

Hx

Plasmodium spp.

Known, high

Travel

T. cruzi

Known

Hx

B. microti

Known

Hx

Dengue

Suspected

Ehrlichia/Anaplasma

Suspected

HHV-8

Theoretical

SARS virus

Theoretical

Travel, risk

vCJD

Theoretical

Hx, travel

Yes Symptoms

Other

LRc

Yes Yesd

LRd

Yes In development

Transmissibility: Known = multiple historical reports. High = transmission at unacceptably high frequencies absent any intervention. Suspected = limited case reports. Theoretical = properties of the agent imply transmissibility but no cases reliably reported. a

b Questions: Hx = donor is asked about a history of the disease. Risk = donor is asked about risk factors or behaviors associated with likelihood of exposure to the agent. Symptoms = donor is asked about the presence or recent presence of symptoms associated with the agent. Travel = Donor is asked about residence in or travel to a location judged to offer increased risk of exposure to the agent. c

LR, leukoreduction. HTLV and CMV are largely cell-associated in the circulation.

d

Procedures are used to reduce risk of CMV infection in products specifically destined for transfusion to highly susceptible patients.

C viruses, human immunodeficiency virus (HIV) (one of the best examples of an emerging pathogen), human T cell lymphotropic virus (HTLV), and syphilis. In addition, all donors are interviewed in order to elicit medical or risk histories associated with infection with these and other pathogens. As a result of these precautions, the residual risk of infection for HIV and hepatitis C virus is about 1 in every 2,000,000 units transfused; that for hepatitis B is somewhat higher at 1 in 250,000– 500,000; and the risk of HTLV infection appears to be vanishingly small (1). It has been more than 40 years since a case of syphilis transmission by transfusion. Although these figures are comforting, they also illustrate that other risks may now be higher when considered relative to these major infections. Historically, almost all infections that are transmissible by transfusion have been characterized by a

2 Dec 2003

22:3

AR

AR206-ME55-12.tex

AR206-ME55-12.sgm

LaTeX2e(2002/01/18)

EMERGING THREATS TO THE BLOOD SUPPLY

P1: GBC

193

prolonged, silent carrier state with the infectious agent circulating in the blood but not causing symptoms. However, this is not a prerequisite; some infections with a very short period of infectivity in the blood (i.e., a few days) have been transmitted via transfusion.

Annu. Rev. Med. 2004.55:191-207. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Interventions The risk of transfusion from emerging infections should be managed systematically. In general, a systematic approach should include a broad mechanism of surveillance to identify emerging infections, followed by a process to assess whether the agent could be transmitted by transfusion. Prerequisites would include (a) an asymptomatic phase with the infectious agent present in the blood, (b) the ability of the agent to survive in stored blood, and (c) patients susceptible to intravenous infection. Likely candidate agents should be prioritized for further study both on the basis of public health importance and public concern (2). Should time permit, it would be highly desirable to assess the prevalence (and ideally, incidence) of infection in the donor population. Finally, appropriate interventions should be implemented where necessary. Ideally, such interventions should be continuously evaluated for efficacy. Available interventions include measures based on selection of epidemiologically safe populations from which donors are drawn; measures based on a history elicited from the donor, leading to permanent or temporary deferral; and test methods designed to detect evidence of infection or infectivity with the agent in question. A subset of tests, known as surrogate tests, use indirect indicators to test out a population of donors thought to be at increased risk of transmitting the agent in question. An example of this approach was the introduction of tests for elevated serum levels of alanine aminotransferase and/or the presence of antibodies to the hepatitis B core antigen in order to reduce the risk of transfusion-transmitted hepatitis C (which was then known as non-A, non-B hepatitis) (3). In this review, we focus on a number of emerging infections relevant to blood safety, although we also discuss a few counter examples of infections that have not yet been shown to offer any threat.

VIRAL INFECTIONS West Nile Virus West Nile virus (WNV) is a flavivirus of the JE serogroup. It was originally recognized in the West Nile region of Uganda in 1937. Until recently, it was endemic in a large part of Africa, Southern and Southeastern Europe, the Middle East, and the western part of the Indian subcontinent. In those areas, it would probably not have qualified as an emerging infection. It does, however, result in periodic outbreaks, which may be quite large but are generally geographically restricted. However, in 1999, an outbreak of WNV infection occurred in New York City, involving

2 Dec 2003

22:3

Annu. Rev. Med. 2004.55:191-207. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

194

AR

DODD

AR206-ME55-12.tex

¥

AR206-ME55-12.sgm

LaTeX2e(2002/01/18)

P1: GBC

LEIBY

66 human cases with 22 deaths (4). During 2000 and 2001, a similar number of human cases were noted, but evidence of viral infections in birds, farm animals, and mosquito pools showed that the virus had spread up and down the entire East coast of the United States and into some of the midwestern and southern central states. In the year 2002, 4156 cases with 284 deaths were reported from 39 states and the District of Columbia, plus a smaller number from three Canadian provinces. Clearly, WNV is a new and emerging pathogen in North America and it seems to be displaying many of the properties typical of pathogens newly introduced into a previously unaffected geographic area. The natural infection cycle of the virus is among birds; transmission occurs predominantly via culicine mosquitoes, although there is also a tick-borne cycle in the Old World. Humans and domestic animals are an accidental, end-stage host and the relatively low level of viremia suggests that they cannot serve as a source of secondary transmission by vectors. In contrast, viremic levels in birds can be extremely high, suggesting that birds are the amplifying hosts (5). Over the first four years of the epidemic in the United States, avian mortality has been significant. The incubation period in humans is on the order of 2–7 days. Current estimates suggest that only ∼20% of infections lead to symptoms and the vast majority of these are relatively mild and flu-like (West Nile fever). However, about 1 in every 150 infections will result in severe, life-threatening or fatal neurologic disease, termed West Nile encephalitis or West Nile meningoencephalitis. Early studies based on experimental inoculations indicated that there was a brief viremic period prior to and during symptomatic disease (5). Consequently, it was thought that a small potential existed for transmission of WNV by transfusion. In August 2002, Biggerstaff & Petersen (6) published an estimate of the risk of transfusion transmission of WNV during the 1999 epidemic in New York. They concluded that the risk was about two viremic donations in every 10,000 in the borough of Queens during the epidemic period that year. That risk is high, relative to more familiar transfusion-transmitted agents, but it was confined in both time and location. Only a few weeks after the publication of that paper, four cases of WNV were transmitted from a single organ donor who had, in turn, been infected by transfusion. By the end of 2002, more than 60 suspected transfusion-transmitted cases of WNV had been reported and 23 of these cases were confirmed. In a number of cases, samples from the index donations were available and were shown to be viremic by nucleic acid amplification and/or viral isolation. Additionally, follow-up samples from implicated donors showed that the donor had seroconverted and become WNV IgM–antibody-positive. Interestingly, in 9 of 14 cases, the implicated donor did report some typical symptoms of WNV fever in the three weeks before or after the donation (6a; L.R. Petersen, personal communication). Apparently, symptomatic individuals may have higher viral titers in the blood than those who are aymptomatic. Evaluations by the Centers for Disease Control and Prevention (CDC) suggested that the risk of transmission of WNV via transfusion varied greatly by time and location during 2002, with peak risk rates as high as 1.5 in 1000 donations.

2 Dec 2003

22:3

AR

AR206-ME55-12.tex

AR206-ME55-12.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:191-207. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

EMERGING THREATS TO THE BLOOD SUPPLY

P1: GBC

195

However, the nationwide mean risk was not thought to exceed 3 per 100,000 during the 2002 epidemic. As might be expected, these observations prompted considerable concern about transfusion safety. In 2002, blood collection agencies began routinely alerting donors to report any symptoms of WNV infection that occurred after donation, and measures were put in place to defer individuals with diagnosed disease or with evidence of implication in a transfusion transmission. Subsequently, the US Food and Drug Administration (FDA) recommended that the donor interview be supplemented to elicit a history of fever and headache up to one week prior to donations during the months in which peak human infection is anticipated (June– November). A positive response would result in deferral of the donor for four weeks (7). Manufacturers designed nucleic acid tests for routine screening of blood donations in small pools, as is currently in place for HIV and hepatitis C virus. Such tests were uniformly in place (under Investigational New Drug procedures) by July of 2003. Preliminary data based on samples taken during the 2002 epidemic suggest that the frequency of donations positive for WNV RNA is compatible with projections based on infectivity models, derived from estimates of incidence and of the length of asymptomatic viremia.

Severe Acute Respiratory Syndrome (SARS) In November 2002, an apparently new human infection emerged, probably in Guangdong, China. The infection caused fever and severe respiratory symptoms with a 7%–15% fatality rate. Of greatest concern was the highly contagious nature of the disease, which was quite frequently transmitted to close contacts (including caregivers) of affected patients (8). As of May 8, 2003, a total of 7053 cases had been observed, with 506 fatalities, and 33 countries or discrete localities had reported at least one case. There had been 63 cases in the United States but no fatalities. More than 50% of all cases had occurred in mainland China. The World Health Organization and CDC issued travel warnings suggesting cancellation of all nonessential travel to areas where transmission was judged to be a particular risk. These included Beijing, Singapore, Hong Kong, Hanoi (Vietnam), Taiwan, and Toronto (Canada). At the time of writing, the outbreaks have apparently been controlled in Hanoi and Toronto and are thought to be limited in Singapore, Hong Kong, and the United States. By August 2003, all travel warnings had been lifted. The leading etiologic hypothesis is a novel coronavirus, not previously observed among humans or other mammals, which belongs to a genetically distinct group (9, 10). The virus has proven quite hardy, with significant survival on surfaces, perhaps accounting for some elements of its transmissibility. In addition, it appears to be shed in the feces of patients for a fairly prolonged period after disease. Some patients appear to be more infectious than others. The SARS outbreak has caused global concern, and massive public health efforts have been initiated to control and contain it. Such measures include interventions designed to reduce the risk of transmission by transfusion. This is

2 Dec 2003

22:3

Annu. Rev. Med. 2004.55:191-207. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

196

AR

DODD

AR206-ME55-12.tex

¥

AR206-ME55-12.sgm

LaTeX2e(2002/01/18)

P1: GBC

LEIBY

highly unusual, as respiratory diseases are not usually believed to offer any risk of transmission by transfusion. (For example, the US outbreak of hantavirus disease during the 1990s was not considered a transfusion risk.) However, there was evidence that the putative infectious agent of SARS did have a viremic phase, at least during symptomatic disease. Such evidence included isolation of the virus from the kidney of one patient and finding of viral RNA in patient blood samples by PCR. In the absence of any knowledge about the presence of viremia during the presymptomatic phase, or of the extent to which inapparent infection occurred, it seemed prudent to attempt to reduce the risk of transmission by transfusion. In many countries, potential blood donors were deferred on the basis of recent travel to a SARS-affected area (as defined by CDC or WHO), contact with a known case, or a history of the disease itself. In the United States, deferral for travel or contact was set at two weeks, and donors with a history of SARS were deferred for four weeks after recovery. These were interim measures pending more information about the etiology and pathogenesis of SARS. As of July 2003, there have been no cases of SARS attributable to transmission by transfusion. However, the disease has caused widespread disruption of the blood supply in affected areas, largely because potential donors are unwilling to risk exposure outside their homes. Although diagnostic tests for SARS are under development, at the time of writing there are no tests to identify potentially viremic blood donors.

Other Viral Infections There are many other emerging viral infections, and each must be evaluated for transfusion risk. High-profile hemorrhagic fever agents, such as Ebola and Marburg viruses, are not considered to be a major risk at this time, because of their rarity and the rapidity with which symptoms and death occur. However, serologic evidence of asymptomatic infection with Ebola virus has been reported (11). Thus, a significant outbreak that is not geographically restricted would generate concern about blood safety. Hantavirus, as mentioned above, is not considered a transfusion risk (12). Selected herpes viruses, particularly cytomegalovirus (CMV) and Epstein-Barr virus (EBV), have long been known to be transmitted by transfusion. Transfusiontransmitted CMV is known to cause serious disease in some susceptible patient populations, such as low-birthweight infants and immunocompromised individuals. A significant degree of control of this outcome has been achieved by the use of CMV-seronegative and/or leukoreduced blood components for at-risk patients. HHV-8, the most recently characterized human herpes virus, is also of some concern. This virus is considered to be the etiologic agent of Kaposi’s sarcoma (KS), and is also known as Kaposi’s sarcoma–associated herpesvirus (KSHV). It is apparent that HHV-8 has been endemic in some human populations in Africa and parts of southern Europe for many years and is associated with “classical” KS. However, HHV-8 is also “emerging” in the sense that it is highly associated with HIV infection and is probably responsible for the occurrence of KS among AIDS patients.

2 Dec 2003

22:3

AR

AR206-ME55-12.tex

AR206-ME55-12.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:191-207. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

EMERGING THREATS TO THE BLOOD SUPPLY

P1: GBC

197

HHV-8 was reported in a blood donation, raising the question of its tranmissibility by transfusion (12a). This concern is heightened by the demonstration that HHV-8 is transmissible by organ transplantation, with resultant disease in some patients (12b–12d). Additionally, an epidemiologic study of a population of women at risk for HIV infection showed an independent association of HHV-8 infection with injecting drug use, clearly implying blood-borne infection (12e). However, there has been no clear evidence of transmission of this virus by blood transfusion (12f). Indeed, Operskalski and colleagues showed that recipients of blood from individuals coinfected with HIV and HHV-8 were infected only with HIV (13). Continued monitoring of the potential risk of HHV-8 transmission is indicated because of the propensity of herpes viruses to be transmitted by body fluids and the potentially serious consequences of infection. Modern molecular technology has resulted in the recent identification of a number of viruses, some of which are actually or potentially transmissible by transfusion. Examples include hepatitis C virus, hepatitis G virus, and TTV and related circoviruses. However, epidemiologic evidence shows that all of these viruses have been present in human populations for millennia and thus should not be considered “emerging.”

BACTERIAL INFECTIONS Ehrlichia chaffeensis Ehrlichia chaffeensis is a rickettsial agent that causes human monocytic ehrlichiosis (HME). First recognized in 1986, this agent infects monocytes and thus is a likely candidate for transmission by blood transfusion (14). E. chaffeensis occurs primarily in the southeastern and southwestern United States, with as yet unconfirmed infections in patients from Asia, Europe, and Africa (15–17). The US vector for this agent is the Lone Star tick, Amblyomma americanum, and the white-tailed deer serves as the main reservoir host. As with many emergent agents discussed in this chapter, symptoms are often described as subclinical or mild and flu-like. About one third of patients with E. chaffeensis infections may develop a rash. In 1999, HME became a notifiable disease in the United States, but only in 36 states. Two hundred cases of HME were reported during 2000 (18), but this figure is probably an underestimate of the true number of cases because of incomplete reporting. Despite a growing number of reported cases, an intracellular location that would appear ideal for transfusion transmission, and a demonstrated ability to remain viable in erythrocytes treated with ADSOL (adenine, dextrose, sorbitol, sodium chloride, and mannitol) at 4–6oC for at least 11 days (19), no cases of transfusiontransmitted E. chaffeensis have been reported. Partly because of its recent description and nonspecific symptoms, transfusion cases of E. chaffeensis infection may be misdiagnosed or overlooked. Alternatively, the widespread use of leukoreduction filters may have the inadvertent effect of lowering the likelihood of transmission.

2 Dec 2003

22:3

198

AR

DODD

AR206-ME55-12.tex

¥

AR206-ME55-12.sgm

LaTeX2e(2002/01/18)

P1: GBC

LEIBY

Annu. Rev. Med. 2004.55:191-207. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Anaplasma phagocytophilum Perhaps of greater concern for blood safety than E. chaffeensis is Anaplasma phagocytophilum, the etiologic agent of human granulocytic ehrlichiosis (HGE), first described in 1994. The known geographic distribution of A. phagocytophilum includes the United States, Europe, and Korea (16, 20). In the United States, A. phagocytophilum is transmitted by black-legged ticks, also called deer ticks— Ixodes scapularis in the Northeast and Upper Midwest, and probably I. pacificus in the Pacific Northwest. In Europe I. ricinus, the sheep tick, has been implicated as the vector. Symptoms of HGE are similar to those of HME, but an accompanying rash is rare. Like HME, HGE has only recently become a notifiable disease in the United States, reportable in a limited number of states. The most recent data indicate that 351 cases of HGE were reported in 2000 (18). These findings are not insignificant because severe cases of HGE may involve renal failure, gastrointestinal bleeding, acute respiratory distress, and death in 5% of patients (15). Largely because of its recent description, seroprevalence and transmission estimates for A. phagocytophilum are limited. Studies in the general US population suggest seroprevalence rates of 0.4%–14%, whereas studies in blood donors from Connecticut and New York revealed rates as high as 3.5% and 11.3%, respectively (21–25). The reported survival of A. phagocytophilum in refrigerated blood for up to 18 days, coupled with a purported transfusion case in Minnesota, indicates that this agent may pose a threat to blood safety (26, 27). At present, continued surveillance is recommended; however, suggestions that A. phagocytophilum may be removed by leukoreduction filters should be considered.

Borrelia burgdorferi An epidemic of arthritis of unknown etiology in the late 1970s near Old Lyme, Connecticut led to the description of a new tick-borne spirochete, Borrelia burgdorferi, identified as the agent of Lyme disease (28, 29). Since its initial description, innumerable cases of Lyme disease have been reported in North America, Europe, and other parts of the world. Despite the widespread nature of this agent, there have been no documented cases of transfusion transmission. In several instances, donors have been diagnosed with Lyme disease a few days after donation, but follow-up testing of recipients did not indicate spirochete transmission (30). Several explanations have been put forth for the lack of transmission. It has been suggested that the spirochetes do not survive blood storage conditions, but studies in red cells, platelets, and fresh frozen plasma indicate that the agent survives for the duration of the storage period (31). Alternatively, cases of post-transfusion Lyme disease are unlikely to be recognized owing to the nonspecific nature of symptoms and the suspected absence of a characteristic erythema migrans (32). It has been difficult to demonstrate spirochetes in Lyme disease patients (33); thus, the spirochetemic phase may be very short and not conducive to transmission by transfusion. Even though no transmission cases have been demonstrated to date, continued vigilance with regard to B. burgdorferi is required.

2 Dec 2003

22:3

AR

AR206-ME55-12.tex

AR206-ME55-12.sgm

LaTeX2e(2002/01/18)

EMERGING THREATS TO THE BLOOD SUPPLY

P1: GBC

199

PROTOZOAN INFECTIONS

Annu. Rev. Med. 2004.55:191-207. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Plasmodium spp. The genus Plasmodium contains the protozoan parasites that cause human malaria: P. falciparum, P. vivax, P. malariae, and P. ovale. These agents are found primarily in tropical and subtropical regions, where they annually cause 300–500 million cases of malaria and 1.5–2.7 million deaths (34). Humans generally acquire infection through the bite of an infected female anophelene mosquito. Once introduced into the human host, sporozoites infect hepatic cells and undergo one round of replication before their release into the peripheral blood as merozoites. These merozoites in turn invade erythrocytes, where they undergo synchronous replication, the timing of which is species-specific and directly related to the episodic febrile paroxysms and anemia characteristic of malaria. More severe complications, attributable in part to high parasitemias, include coma and death. Available drug treatments for malaria consist of chloroquine, primaquine, quinine, doxycycline, and, in the case of drug-resistant falciparum malaria, mefloquine. Patients with extremely high and life-threatening levels of parasitemia can also be treated by exchange transfusions designed to rapidly reduce the number of parasites (35). In countries where malaria is not endemic, relatively few cases are acquired through blood transfusion. In the United States, ∼1–2 cases of transfusion-acquired malaria are reported annually, and Canada has seen only 3 cases in the past 10 years (36, 37). Most of these cases can be attributed to asymptomatic or semiimmune donors who have recently returned from an endemic country, or, in rare cases, emigrated to North America many years earlier. Prevention of Plasmodium transmission depends on completion of an accurate travel history. However, misinterpretation of travel questions has led to transfusion cases on more than one occasion. Transfusion-transmitted Plasmodium is perhaps a larger problem in Europe because of greater proximity to, and large numbers of immigrants from, endemic countries. Indeed, in parts of Europe, blood is screened for antibodies to Plasmodium. Although transfusion-transmitted malaria in nonendemic countries can be attributed largely to immigration and international travel, one must also consider the possibility of re-emerging malaria in non-endemic countries. Many locations harbor competent anophelene vectors in the absence of the malaria parasite. The phenomenon of “airport malaria,” by which an infected mosquito arrives from an endemic area in association with an international flight, suggests one mechanism for the reintroduction of the parasite (38). An alternative scenario would be the reintroduction of the parasite into the local mosquito population by an infected person. In the summer of 2002, three people in suburban Washington, DC were apparently infected with P. vivax that was probably introduced into the local mosquito population by an infected immigrant from an endemic country (39). Endemic foci of malaria could be reestablished in nonendemic countries by either of these mechanisms.

2 Dec 2003

22:3

200

AR

DODD

AR206-ME55-12.tex

¥

AR206-ME55-12.sgm

LaTeX2e(2002/01/18)

P1: GBC

LEIBY

Annu. Rev. Med. 2004.55:191-207. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Babesia spp. Intraerythrocytic protozoan parasites of the genus Babesia cause the zoonotic disease babesiosis. Human cases of babesiosis have been reported in North America, Europe, Latin America, Africa, and Southeast Asia (40). In general, Babesia spp. are transmitted to humans by ticks of the genus Ixodes, the same group of ticks responsible for transmission of B. burgdorferi and A. phagocytophilum. Most human cases of babesiosis have been limited to the United States and Europe. US cases are primarily attributed to infection with B. microti, whereas in Europe most cases are caused by B. divergens (41, 42). As for A. phagocytophilum, vectors are I. scapularis and I. pacificus in the United States and I. ricinus in Europe. Both nymphal and adult stages of these ticks can transmit infection; however, for successful transmission of Babesia spp., infected ticks must feed for at least 48 h (43). In most cases, babesiosis presents as an asymptomatic or mild self-limiting disease. It can be characterized by fever, headache, and myalgia. More severe symptoms are observed in infant, elderly, immunocompromised, or asplenic hosts. Infections with B. divergens or the newly emergent US agent WA-1 are more virulent than B. microti infections, producing fulminant disease with rapid onset (44, 45). In addition to WA-1, other emergent isolates of Babesia have been identified in the United States, including CA-1 and MO-1 (46, 47). With the recent report of B. divergens in the United States and transfusion cases of a B. microti-like parasite in Japan, the Babesia spp. are clearly emerging as new threats to blood safety (48, 49). There have been at least 40 reported cases of transfusion-transmitted B. microti, (50) a number that now appears conservative. Except for two cases reported in Japan and Canada (49, 51), the transfusion-transmission cases have all occurred in the United States. Additionally, although it was initially recognized 10 years ago, there have already been two cases of transfusion-transmitted WA-1 (52). Not surprisingly, the implicated donor in practically all these cases was asymptomatic. The blood component primarily associated with transmission was erythrocytes, but platelet units probably containing B. microti–infected erythrocytes have been linked with transmission in several cases. Based on a transfusion case, the parasite is capable of surviving and remaining viable for at least 35 days at 4oC (53). Indeed, a recent study from Connecticut reported that the risk of transfusion-transmitted B. microti is 1 in 1800 transfused red cell units (54). As for many of these agents, options for preventing transmission of Babesia spp. are limited. Because the risk-factor questions lack sufficient sensitivity, interview questions cannot adequately screen donors for Babesia exposure. The diagnostic method of choice remains the immunofluorescence assay, which is impractical for blood screening (55). However, the recent development of an enzyme immunoassay for B. microti suggests that blood screening tests may become available in the near future (56). Similarly, the existence of sensitive PCR assays for Babesia DNA suggests that nucleic acid testing is also a viable option, particularly for detecting infections in their window period (i.e., before humoral immune responses are detectable) (57).

2 Dec 2003

22:3

AR

AR206-ME55-12.tex

AR206-ME55-12.sgm

LaTeX2e(2002/01/18)

EMERGING THREATS TO THE BLOOD SUPPLY

P1: GBC

201

Annu. Rev. Med. 2004.55:191-207. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Trypanosoma cruzi Trypanosoma cruzi is the etiologic agent of Chagas’ disease, also called American trypanosomiasis. This protozoan parasite is only endemic to the Western Hemisphere, primarily South America, Central America, and Mexico, where an estimated 16–18 million people are infected (58). Human infections are acquired naturally following exposure to a triatomine insect vector; during a blood meal the bug defecates, and infective trypomastigotes within the feces enter the bite wound or the conjunctiva or other mucosal surface. Infections with T. cruzi are also acquired congenitally, by organ transplant, and by blood transfusion. Once introduced into the host, the parasite disseminates hematogenously to muscle tissue, with a predilection for the heart. The subsequent acute phase is generally mild and short-lived, lasting only 1–2 months. Infected persons then enter a lifelong indeterminate phase of disease that is generally asymptomatic but characterized by detectable antibody and intermittent parasitemia. Later in life, 20%–30% of infected persons will develop clinical Chagas’ disease, manifested by potentially fatal cardiac and gastrointestinal involvement (59). The efficacy of drug therapy for Chagas’ disease remains controversial, and treatment options are limited to nifurtimox and benznidazole. Although reported cases of Chagas’ disease are generally in decline throughout much of Latin America, T. cruzi can be viewed as an emerging infectious agent in the United States and Canada. During the past several decades, millions of people have emigrated to the United States and Canada from countries in which T. cruzi is endemic. Estimates suggest that in the United States alone, 50,000 to 100,000 of these immigrants may harbor chronic, asymptomatic T. cruzi infections (60). The children of these immigrants may also be infected, having acquired the infection congenitally (61). Thus, these immigrants and their children represent a growing reservoir population for the potential transmission of T. cruzi by blood transfusion. During the past 15 years, there have been 6 reported cases of transfusiontransmitted T. cruzi in Canada (n = 2) and the United States (n = 4) (62–66). However, these were fulminant cases recognized in immunocompromised recipients, which suggests that many other cases probably occur in immunocompetent persons but are not recognized or diagnosed as Chagas’ disease. Nationwide estimates suggest that 1 in 25,000 US blood donors are infected with T. cruzi. Local infection rates can be much higher, depending on the proportion of at-risk donors in the population. In Los Angeles and Miami, seropositivity rates for T. cruzi were reported to be 1 in 7500 and 1 in 9000 blood donors, respectively (67). This same study indicated that the prevalence of T. cruzi in Los Angeles blood donors increased significantly from 1996 to 1998, with a local rate of 1 in 5400 donors during 1998. This rapid increase in prevalence could be traced to enhanced minority recruitment efforts necessitated by changes in donor demographics. Another study has demonstrated that a majority (63%) of blood donors with antibodies to T. cruzi show evidence of active parasitemia in their peripheral blood (D.A. Leiby et al., unpublished data). Taken together, US

2 Dec 2003

22:3

Annu. Rev. Med. 2004.55:191-207. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

202

AR

DODD

AR206-ME55-12.tex

¥

AR206-ME55-12.sgm

LaTeX2e(2002/01/18)

P1: GBC

LEIBY

blood collection organizations are faced with increasing numbers of at-risk and seropositive donors, who in many cases are parasitemic and potentially capable of transmitting T. cruzi to blood recipients. In general, preventing transmission of T. cruzi by blood transfusion is relatively straightforward. In much of Latin America, blood donations are screened for T. cruzi antibodies. US and Canadian blood is not screened for this parasite, mainly because screening assays licensed by the respective governments are unavailable. However, at a September 2002 meeting of the FDA’s Blood Products Advisory Committee, the FDA requested manufacturers to develop and submit for approval antibody screening tests for T. cruzi (68). It is anticipated that once a test has been licensed, universal blood screening for T. cruzi will be implemented in the United States. Alternative control measures such as risk-factor questions lack sufficient sensitivity (69), and some risk-factor questions pertaining to country of origin may be culturally offensive and counterproductive.

TRANSMISSIBLE SPONGIFORM ENCEPHALOPATHIES Transmissible spongiform encephalopathies (TSEs) are characterized by irreversible, terminal brain disease, pathologically recognized by spongiform change and amyloid plaque deposition. This group of diseases, which (in differing forms) affect animals and man, are thought to be caused by an unconventional infectious agent termed a prion. Prions are cellular proteins that normally are cycled by physiologic mechanisms. However, they can also exist in one or more conformational forms that are no longer subject to enzymatic degradation. This pathologic form accumulates and is thought to cause the disease. The pathologic prion can act as a nucleus for the conversion of more normal protein to the insoluble form, resulting in multiplication of the pathogenic agent (70). As their name suggests, TSEs are transmissible. For the classical human TSEs, the predominant (about 85%) form of disease is spontaneous and the majority of residual cases are genetic. In animals, the disease is readily transmitted by intracranial inoculation, a situation mirrored in the human TSE, Creutzfeldt Jakob Disease (CJD), which has been transmitted by reuse of neurologic surgery instruments and transplant of dura mater. Additionally, iatrogenic infections resulted from the use of pituitary-derived human growth hormones. The majority of TSEs do not qualify as emerging infections. Indeed, the incidence rates of CJD and some similar naturally occurring TSEs appear to be remarkably stable over time and geography. However, this is not the case for variant CJD (vCJD). This human TSE emerged recently as an outcome of the huge epidemic of bovine spongiform encephalopathy (BSE) that occurred primarily in the United Kingdom between 1980 and 1996. The BSE epidemic resulted from the use of meat and bone meal as cattle feed. Subsequently vCJD arose, probably as a result of the pathogenic prion entering the human food chain from affected cattle (71). As of early March 2003, there had been 132 confirmed or suspected cases of vCJD in

2 Dec 2003

22:3

AR

AR206-ME55-12.tex

AR206-ME55-12.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:191-207. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

EMERGING THREATS TO THE BLOOD SUPPLY

P1: GBC

203

the United Kingdom, 6 in France, 1 in Italy, and 1 in the Republic of Ireland. In addition, Canada, Hong Kong, and the United States had each reported a single case. Each of these cases, however, were attributable to exposure in the United Kingdom. To date, there has been no evidence of the transmission of either CJD or vCJD by blood transfusion, and there is reason to believe that CJD is unlikely to be transmissible by this route (72). However, at the time of writing, it is not possible to be as confident about vCJD, largely because the pathogenesis of disease differs in some important ways from CJD and because there is less experimental information. Experiments by Houston and colleagues showed transmission of BSE via transfusion in sheep (73, 74). As a consequence, measures have been taken to attempt to reduce the theoretical risk of transmission of vCJD by transfusion. In the United States, the primary approach has been to ask presenting blood donors about their travel history and to reject those who have spent time in BSE-affected areas during times when BSE was occurring, and prior to the implementation of effective controls on the human food chain. The absence of any premortem diagnostic or screening tests for TSEs complicates efforts to understand and control their transmission in human populations.

SUMMARY AND CONCLUSIONS Prevention of transfusion-transmissible infections attracts a great deal of public and political concern. The continual effort to maintain and improve blood safety goes well beyond accepted norms for cost-effectiveness in medical interventions. This approach has been singularly effective in managing agents traditionally considered to offer the most risk to the patient. Transfusionists and regulatory agencies are now focusing a similar degree of attention on emerging infections. However, the tools are not always available or adequate (as in the cases of vCJD and babesiosis). Some imperfect tools, such as history-based exclusions, may have been implemented too hastily, or without adequate attention to their impact on the availability of blood. Finally, the astute reader will have recognized that the enormous effort to prevent rare events in the United States is not matched by the resources to manage blood safety in the developing world, where the blood recipient population continues to be at serious risk of infection with HIV, Plasmodium spp., T. cruzi, and other emerging infections. The Annual Review of Medicine is online at http://med.annualreviews.org

LITERATURE CITED 1. Dodd RY, Notari EP, Stramer SL. 2002. Current prevalence and incidence of infectious disease markers and estimated window-period risk in the American Red

Cross blood donor population. Transfusion 42:975–79 2. Kleinman S. 2001. Transfusion safety when you don’t know what you are

2 Dec 2003

22:3

204

3.

Annu. Rev. Med. 2004.55:191-207. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

4.

5.

6.

6a.

7.

8.

9.

10.

11.

AR

DODD

AR206-ME55-12.tex

¥

AR206-ME55-12.sgm

LaTeX2e(2002/01/18)

P1: GBC

LEIBY

looking for: issues related to emerging pathogens. In Blood Safety in the New Millennium, ed. SL Stramer, pp. 79–96. Bethesda, MD: Am. Assoc. Blood Banks Blajchman MA, Bull SB, Feinman SV. 1995. Post-transfusion hepatitis: impact of non-A, non-B hepatitis surrogate tests. Lancet 345:21–25 Nash D, Mostashari F, Fine A, et al. 2001. The outbreak of West Nile virus infection in the New York City area in 1999. N. Engl. J. Med. 344:1807–14 Petersen LR, Marfin AA. 2002. West Nile virus: a primer for the clinician. Ann. Intern. Med. 137:173–79 Biggerstaff BJ, Petersen LR. 2002. Estimated risk of West Nile virus transmission through blood transfusion during an epidemic in Queens, New York City. Transfusion 42:1019–26 Pealer LN, Marfin AA, Petersen LR, et al. 2003. Transmission of West Nile virus through blood transfusion—United States, 2002. N. Engl. J. Med. In press US Food and Drug Administration. 2003. Guidance for industry, revised recommendations for the assessment of donor suitability and blood and blood product safety in cases of known or suspected West Nile virus infection. http://www.fda.gov/cber/ gdlns/wnvguid.pdf Donnelly CA, Ghani AC, Leung GM, et al. 2003. Epidemiological determinants of spread of causal agent of severe acute respiratory syndrome in Hong Kong. Lancet 361:1761–6 Marra MM, Jones SJM, Astell CR. 2003. The genome sequence of the SARS-associated coronavirus. Science 300:1399– 404 Rota PA, Oberste MA, Monroe SS, et al. 2003. Characterization of a novel coronavirus associated with severe acute respiratory syndrome. Science 300:1394–99 Leroy EM, Baize S, Volchkov VE, et al. 2000. Human asymptomatic Ebola infection and strong inflammatory response. Lancet 355:2210–15

12. Duchin JS, Koster FT, Peters CJ, et al. 1994. Hantavirus pulmonary syndrome: a clinical description of 17 patients with a newly recognized disease. N. Engl. J. Med. 330:949–55 12a. Blackbourn DJ, Ambroziak J, Lennette E, et al. 1997. Infectious human herpesvirus 8 in a healthy North American blood donor. Lancet 349:609–11 12b. Farge D, Lebb´e C, Marjanovic Z, et al. 1999. Human herpes virus-8 and other risk factors for Kaposi’s sarcoma in kidney transplant recipients. Transplantation 67: 1236–42 12c. Milliancourt C, Barete S, Marcelin AG, et al. 2001. Human herpesvirus-8 seroconversions after renal transplantation. Transplantation 72:1319–20 12d. Regamey N, Tamm M, Wernli M, et al. 1998. Transmission of human herpesvirus 8 infection from renal-transplant donors to recipients. N. Engl. J. Med. 339:1358– 63 12e. Cannon MJ, Dollard SC, Smith DK, et al. 2001. Blood-borne and sexual transmission of human herpesvirus 8 in women with or at risk for human immunodeficiency virus infection. N. Engl. J. Med 344:637–43 12f. Challine D, Roudot-Thoraval F, Sarah T, et al. 2001. Seroprevalence of human herpes virus 8 antibody in populations at high or low risk of transfusion, graft, or sexual transmission of viruses. Transfusion 41:1120–5 13. Operskalski EA, Busch MP, Mosley JW, et al. 1997. Blood donations and viruses. Lancet 349:1327 14. Anderson BE, Dawson JE, Jones DC, Wilson KH. 1991. Ehrlichia chaffeensis, a new species associated with human ehrlichiosis. J. Clin. Microbiol. 29:2838–42 15. Dumler JS, Bakken JS. 1995. Ehrlichial diseases of humans: emerging tick-borne infections. Clin. Infect. Dis. 20:1102–10 16. Heo EJ, Park JH, Koo JR, et al. 2002. Serologic and molecular detection of Ehrlichia chaffeensis and Anaplasma

2 Dec 2003

22:3

AR

AR206-ME55-12.tex

AR206-ME55-12.sgm

LaTeX2e(2002/01/18)

EMERGING THREATS TO THE BLOOD SUPPLY

17.

Annu. Rev. Med. 2004.55:191-207. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

18.

19.

20.

21.

22.

23.

24.

25.

26.

phagocytophila (human granulocytic ehrlichiosis agent) in Korean patients. J. Clin. Microbiol. 40:3082–85 Keysary A, Amram L, Keren G, et al. 1999. Serologic evidence of human monocytic and granulocytic ehrlichiosis in Israel. Emerg. Infect. Dis. 5:775–78 Centers for Disease Control and Prevention. 2002. Summary of notifiable diseases—United States, 2000. Morbid. Mortal. Wkly. Rep. 49:x–xi McKechnie DB, Slater KS, Childs JE, et al. 2000. Survival of Ehrlichia chaffeensis in refrigerated, ADSOL-treated RBCs. Transfusion 40:1041–47 Bakken JS, Dumler JS. 2000. Human granulocytic ehrlichiosis. Clin. Infect. Dis. 31:554–60 Fritz CL, Kjemtrup AM, Conrad PA, et al. 1997. Seroepidemiology of emerging tickborne infectious diseases in a Northern California community. J. Infect. Dis. 175:1432–39 Hilton E, DeVoti J, Benach JL, et al. 1999. Seroprevalence and seroconversion for tick-borne diseases in a high-risk population in the northeast United States. Am. J. Med. 106:404–9 Bakken JS, Goellner P, Van Etten M, et al. 1998. Seroprevalence of human granulocytic ehrlichiosis among permanent residents of northwestern Wisconsin. Clin. Infect. Dis. 27:1491–96 Leiby DA, Chung APS, Cable RG, et al. 2002. Relationship between tick bites and the seroprevalence of Babesia microti and Anaplasma phagocytophila (previously Ehrlichia sp.) in blood donors. Transfusion 42:1585–91 Aguero-Rosenfeld ME, Donnarumma L, Zentmaier L, et al. 2002. Seroprevalence of antibodies that react with Anaplasma phagocytophila, the agent of human granulocytic ehrlichiosis, in different populations in Westchester County, New York. J. Clin. Microbiol. 40:2612–15 Kalantarpour F, Chowdhury I, Wormser GP, et al. 2000. Survival of the human

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

P1: GBC

205

granulocytic ehrlichiosis agent under refrigeration conditions. J. Clin. Microbiol. 38:2398–99 Eastlund T, Persing D, Mathiesen D, et al. 1999. Human granulocytic ehrlichiosis after red cell transfusion. Transfusion 39(Suppl.):117S (Abstr.) Burgdorfer W, Barbour AG, Hayes SF, et al. 1982. Lyme disease—a tick-borne spirochetosis? Science 216:1317–19 Steere AC, Grodzicki RL, Kornblatt AN, et al. 1983. The spirochetal etiology of Lyme disease. N. Engl. J. Med. 308:733– 40 Cable R, Krause P, Badon S, et al. 1993. Acute blood donor co-infection with Babesia microti (Bm) and Borrelia burgdorferi (Bb). Transfusion 33(Suppl.):50S (Abstr.) Badon SJ, Fister RD, Cable RG. 1989. Survival of Borrelia burgdorferi in blood products. Transfusion 29:581–83 Cable RG, Trouern-Trend J. 2001. Tickborne infections. In Blood Safety and Surveillance, ed. JV Linden, C Bianco, pp. 399–422. New York: Marcel Dekker Shrestha M, Grodzicki RL, Steere AC. 1985. Diagnosing early Lyme disease. Am. J. Med. 78:235–40 World Health Organization. 1997. World malaria situation in 1994. Wkly. Epidemiol. Rec. 72:269–76 Phillips P, Nantel S, Benny WB. 1990. Exchange transfusion as an adjunct to the treatment of severe falciparum malaria: case report and review. Rev. Infect. Dis. 12:1100–8 Mungai M, Tegtmeier G, Chamberland M, et al. 2001. Transfusion-transmitted malaria in the United States from 1963 through 1999. N. Engl. J. Med. 344:1973– 78 Slinger R, Giulivi A, Bodie-Collins M, et al. 2001. Transfusion-transmitted malaria in Canada. Can. Med. Assoc. J. 164:377–79 Thang HD, Elsas RM, Veenstra J. 2002. Airport malaria: report of a case and a

2 Dec 2003

22:3

206

39.

40.

Annu. Rev. Med. 2004.55:191-207. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

41.

42.

43.

44.

45.

46.

47.

48.

49.

AR

DODD

AR206-ME55-12.tex

¥

AR206-ME55-12.sgm

LaTeX2e(2002/01/18)

P1: GBC

LEIBY

brief review of the literature. Neth. J. Med. 60:441–43 Centers for Disease Control and Prevention. 2002. Local transmission of Plasmodium vivax malaria—Virginia, 2002. Morbid. Mortal. Wkly. Rep. 51:921– 23 Kjemtrup AM, Conrad PA. 2000. Human babesiosis: an emerging tick-borne disease. Int. J. Parasitol. 30:1323–37 McQuiston JH, Childs JE, Chamberland ME, et al. 2000. Transmission of tickborne agents by blood transfusion: a review of known and potential risks in the United States. Transfusion 40:274– 84 Homer MJ, Aguilar-Delfin I, Telfored SR, et al. 2000. Babesiosis. Clin. Microbiol. Rev. 13:451–69 Piesman J, Spielman A. 1982. Babesia microti: infectivity of parasites from ticks for hamsters and white-footed mice. Exp. Parasitol. 53:242–48 Quick RE, Herwaldt BL, Thromford JW, et al. 1993. Babesiosis in Washington State: a new species of Babesia? Ann. Intern. Med. 119:284–90 Gorenflot A, Moubri K, Precigout E, et al. 1998. Human babesiosis. Ann. Trop. Med. Parasitol. 92:489–501 Persing DH, Herwaldt BL, Glaser C, et al. 1995. Infection with a Babesia-like organism in northern California. N. Engl. J. Med. 332:298–303 Herwaldt BL, Persing DH, Pr´ecigout EA, et al. 1996. A fatal case of babesiosis in Missouri: identification of another piroplasm that infects humans. Ann. Intern. Med. 124:643–50 Beattie JF, Michelson ML, Holman PJ. 2002. Acute babesiosis caused by Babesia divergens in a resident of Kentucky. N. Engl. J. Med. 347:697–98 Matsui T, Inoue R, Kajimoto K, et al. 2000. First documentation of transfusionassociated babesiosis in Japan. Rinsho Ketsueki 41:628–34 (Japanese with English summary)

50. Lux JZ, Weiss D, Linden JV, et al. 2003. Transfusion-associated babesiosis after heart transplant. Emerg. Infect. Dis. 9:116–19 51. Jassoum BS, Fong IW, Hannach B, et al. 2000. Transfusion-transmitted babesiosis in Ontario: first reported case in Canada. Can. Commun. Dis. Rep. 26:9– 13 52. Kjemtrup AM, Lee B, Fritz CL, et al. 2002. Investigation of transfusion transmission of a WA1-type babesial parasite to a premature infant in California. Transfusion 42:1482–87 53. Mintz ED, Anderson JF, Cable RG, et al. 1991. Transfusion-transmitted babesiosis: a case report from a new endemic area. Transfusion 31:365–68 54. Cable RG, Badon S, Trouern-Trend, et al. 2001. Evidence for transmission of Babesia microti from Connecticut blood donors to recipients. Transfusion 41(Suppl.):12S–13S (Abstr.) 55. Leiby DA, Chung APS, Cable RG, et al. 2002. Relationship between tick bites and the seroprevalence of Babesia microti and Anaplasma phagocytophila (previously Ehrlichia sp.) in blood donors. Transfusion 42:1585–91 56. Houghton RL, Homer MJ, Reynolds LD, et al. 2002. Identification of Babesia microti-specific immunodominant epitopes and development of a peptide EIA for detection of antibodies in serum. Transfusion 42:1488–96 57. Persing DH, Mathiesen D, Marshall WF, et al. 1992. Detection of Babesia microti by polymerase chain reaction. J. Clin. Microbiol. 30:2097–103 58. 2000. Chagas disease, Chile. Certification of interruption of transmission. Wkly. Epidemiol. Rec. 75:10–12 59. Kirchhoff LV. 1999. American trypanosomiasis (Chagas’ disease). In Tropical Infectious Diseases: Principles, Pathogens, and Practice, ed. RL Guerrant, DH Walker, PF Weller, pp. 785–96. New York: Churchill Livingstone

2 Dec 2003

22:3

AR

AR206-ME55-12.tex

AR206-ME55-12.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:191-207. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

EMERGING THREATS TO THE BLOOD SUPPLY 60. Kirchhoff LV, Gam AA, Gilliam FC. 1987. American trypanosomiasis (Chagas’ disease) in Central American immigrants. Am. J. Med. 82:915–20 61. Leiby DA, Fucci MH, Stumpf RJ. 1999. Trypanosoma cruzi in a low- to moderaterisk blood donor population: seroprevalence and possible congenital transmission. Transfusion 39:310–15 62. Leiby DA, Lenes BA, Tibbals MA, et al. 1999. Prospective evaluation of a patient with Trypanosoma cruzi infection transmitted by transfusion. N. Engl. J. Med. 341:1237–39 63. Cimo PL, Luper WE, Scouros MA. 1993. Transfusion-associated Chagas’ disease in Texas: report of a case. Texas Med. J. 89:48–50 64. Grant IH, Gold JW, Wittner M, et al. 1989. Transfusion-associated acute Chagas disease acquired in the United States. Ann. Intern. Med. 111:849–51 65. Nickerson P, Orr P, Schroeder M-L, et al. 1989. Transfusion-associated Trypanosoma cruzi infection in a nonendemic area. Ann. Intern. Med. 111:851– 53 66. Lane DJ, Sher G, Ward B, et al. 2000. Investigation of the second case of transfusion transmitted Chagas disease in Canada. Blood 96:60a (pt. 1) 67. Leiby DA, Herron RM, Read EJ, et al. 2002. Trypanosoma cruzi in Los Angeles and Miami blood donors: impact of

68.

69.

70.

71. 72.

73.

74.

P1: GBC

207

evolving donor demographics on seroprevalence and implications for transfusion transmission. Transfusion 42:549–55 Blood Products Advisory Committee. 2002. Transcripts of the September 12, 2002 Blood Products Advisory Committee Meeting, CBER, FDA. http://www. fda.gov/OHRMS/DOCKETS/ac/02/trans cripts/3892t1-03.pdf Leiby DA, Read EJ, Lenes BA, et al. 1997. Seroepidemiology of Trypanosoma cruzi, etiologic agent of Chagas’ disease, in U.S. blood donors. J. Infect. Dis. 176:1047– 52 Prusiner SB. 2001. Shattuck Lecture— neurodegenerative diseases and prions. N. Engl. J. Med. 344:1516–26 Collinge J. 1999. Variant CreutzfeldtJakob disease. Lancet 354:317–23 Brown P, Cervenakova L, McShane LM, et al. 1999. Further studies of blood infectivity in an experimental model of transmissible spongiform encephalopathy, with an explanation of why blood components do not transmit CreutzfeldJakob disease in humans. Transfusion 39: 1169–78 Houston F, Foster JD, Chong A, et al. 2000. Transmission of BSE by blood transfusion in sheep. Lancet 356:999– 1000 Hunter N, Foster J, Chong A, et al. 2002. Transmission of prion diseases by blood transfusion. J. Gen. Virol. 83:2897–905

3 Dec 2003

21:51

AR

AR206-ME55-13.tex

AR206-ME55-13.sgm

LaTeX2e(2002/01/18) P1: GBC 10.1146/annurev.med.55.091902.103653

Annu. Rev. Med. 2004. 55:209–22 doi: 10.1146/annurev.med.55.091902.103653 c 2004 by Annual Reviews. All rights reserved Copyright ° First published online as a Review in Advance on Aug. 18, 2003

LEAD POISONING Herbert Needleman

Annu. Rev. Med. 2004.55:209-222. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Professor of Psychiatry and Pediatrics, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania 15213; email: [email protected]

■ Abstract Understanding of lead toxicity has advanced substantially over the past three decades, and focus has shifted from high-dose effects in clinically symptomatic individuals to the consequences of exposure at lower doses that cause no symptoms, particularly in children and fetuses. The availability of more sensitive analytic methods has made it possible to measure lead at much lower concentrations. This advance, along with more refined epidemiological techniques and better outcome measures, has lowered the least observable effect level until it approaches zero. As a consequence, the segment of the population who are diagnosed with exposure to toxic levels has expanded. At the same time, environmental efforts, most importantly the removal of lead from gasoline, have dramatically reduced the amount of lead in the biosphere. The remaining major source of lead is older housing stock. Although the cost of lead paint abatement is measured in billions of dollars, the monetized benefits of such a Herculean task have been shown to far outweigh the costs.

INTRODUCTION In recent years, the focus in lead poisoning has shifted away from adults exposed to high doses in industrial settings to the larger population of asymptomatic children with lesser exposures. This chapter surveys the past three decades of lead research and reviews the evolving knowledge of the distribution, toxicology, and remediation of lead toxicity.

EARLY HISTORY Warnings of lead’s poisonous properties extend at least as far back as the second century B.C., when Nikander, a Greek physician, described the colic and paralysis that followed lead ingestion. The early victims of lead toxicity were mainly lead workers and wine drinkers. Lead’s sweet flavor made it useful in winemaking, to counteract the astringent flavor of tannic acid in grapes. Lead-sweetened wine, containing as much as 20 mg of lead per liter, was an important part of the diet of upper-class Romans. The synchronous decrease in fertility and increase in psychosis among the Roman aristocracy has raised speculation implicating lead poisoning in the fall of Rome (1). 0066-4219/04/0218-0209$14.00

209

3 Dec 2003

21:51

210

AR

AR206-ME55-13.tex

AR206-ME55-13.sgm

LaTeX2e(2002/01/18)

P1: GBC

NEEDLEMAN

Annu. Rev. Med. 2004.55:209-222. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Widespread outbreaks of lead colic continued in Europe until as late as the sixteenth century, when Eberhard Gockel, a German physician, traced a colic epidemic to lead-adulterated wine. Duke Ludwig of W¨urrtemberg, upon learning of an epidemic of lead colic in his duchy, banned its use in winemaking, imposing the death penalty for violators. Workers in the metals trades remain an important risk group; lead exposure remains one of the leading causes of workplace illness. In the United States, more than 320,000 American workers were occupationally exposed to lead in 1998.

DISCOVERY OF CHILDHOOD LEAD POISONING It was only a century ago that childhood lead poisoning was recognized. The rapid growth of scientific understanding can be divided into four stages. The first reports of lead-poisoned children in Brisbane, Australia, in 1892 were greeted with widespread disbelief that lead toxicity could afflict children (2). Although the disease had reached epidemic proportions, there was considerable doubt that lead was the cause. Many of the homes in Brisbane were raised on piles, with large wooden-enclosed verandas that served as play areas for children. The rails were painted with white lead, which chalked and powdered in the hot Brisbane sun. The cause of the epidemic, lead-containing paint, was established in 1904, and lead paint was banned for household use in Brisbane in 1920. Childhood lead poisoning was first described in the United States in 1914 (3). The prevailing belief in the second stage of knowledge was that acute poisoning had only two outcomes: death or complete recovery without any residua. This misconception was discarded in 1943 with the first follow-up of children who had recovered from acute toxicity. Nineteen of 20 surviviors had significant deficits: behavioral disorders, learning difficulties, and school failure (4). In this third stage, it was generally accepted that lead toxicity caused long-term deficits, but these deficits were thought to occur only in those children who had displayed clinical signs of encephalopathy during the acute episode. The fourth stage began in the 1970s, when studies of children with no clinical signs of toxicity showed deficits in IQ scores, attention, and language (see below).

TOXICOLOGY OF LEAD Lead is a divalent cation, and it binds strongly to sulfhydryl groups on proteins. Of the many organs affected by lead, the most important is the central nervous system (CNS). Much of lead’s toxicity can be attributed to distortion of enzymes and structural proteins, but this versatile toxicant has many other targets. Lead interferes with the development of the endogenous opiate system (5). It efficiently cleaves the ribophosphate backbone of tRNA catalytically at specific sites, with no evidence of a threshold (6). Many of lead’s toxic properties are due to its ability to mimic or compete with calcium. At picomolar concentrations, lead competes

3 Dec 2003

21:51

AR

AR206-ME55-13.tex

AR206-ME55-13.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:209-222. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

LEAD POISONING

P1: GBC

211

successfully with calcium for binding sites on cerebellar phosphokinase C and thereby affects neuronal signaling (7). It inhibits calcium entry into cells (8). Lead is picked up by mitochondria and produces swelling and distortion of mitochondrial cristae. Uncoupled energy metabolism, inhibited cellular respiration, and altered calcium kinetics follow (9). Lead has a binary impact on neurotransmitter release: Spontaneous neurotransmitter release is enhanced, whereas stimulated release is inhibited (10). Attention has also focused on the heme synthetic pathway, where many sites for lead activity are found. Delta aminolevulinic acid dehydratase is extremely sensitive to lead. Inhibition of this enzyme results in increased circulating aminolevulinic acid (ALA). ALA is a weak gamma-aminobutyric acid (GABA) agonist that decreases GABA release by presynaptic inhibition. Increased circulating ALA may account for some of the behavioral disorders seen in patients with porphyria and perhaps in lead toxicity. Lead has diverse impacts on the CNS. Immature astrocytes are sensitive to lead, and lead interferes with myelin formation and the integrity of the blood-brain barrier (11). Lead interferes with the synthesis of collagen and affects vascular permeability. At high enough doses, this results in brain edema and hemorrhage (12). At lower doses, lead given to lactating rats interferes with synaptogenesis in their pups (13). Lead’s interference with brain development has been demonstrated using the rodent barrel field cortex as a model (14). Behavioral alterations secondary to lead exposure in rodents and primates are analogous to changes in humans. In one study, monkeys received lead acetate in their food from birth to 200 days of age and achieved blood lead levels ranging from 3 µg/dl to 25 µg/dl. At 7 to 8 years, they were given a delayed alternation test, in which the critical positive stimulus was alternated. Treated monkeys showed impaired ability to learn, particularly at longer intervals of delay (15). Lead-exposed primates also demonstrate impaired social function (16). Rodents given lead show deficits in learning mediated by dopaminergic and glutamatergic systems (17). In one interesting report, untreated rats found 15% solutions of alcohol aversive in a free-choice situation, but when their blood lead levels were raised to 61 µg/dl, they increased their alcohol intake in both free- and forced-choice paradigms (18). The author speculated that lead increased the irritability of the rats and that they sought alcohol as a tranquilizer.

CLINICAL ASPECTS OF TOXICITY Although adult lead poisoning is mainly of occupational origin, cases of acute lead poisoning from leaded dishware, bootlegged moonshine liquor, certain cosmetics, and folk remedies continue to be reported. Lead is still mined and smelted, although this has declined with the removal of lead from gasoline. Lead poisoning in adults can affect the peripheral and central nervous systems, the kidneys, and blood pressure. Classical descriptions of occupational toxicity

3 Dec 2003

21:51

Annu. Rev. Med. 2004.55:209-222. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

212

AR

AR206-ME55-13.tex

AR206-ME55-13.sgm

LaTeX2e(2002/01/18)

P1: GBC

NEEDLEMAN

depict peripheral neuropathy with wrist or foot drop. At lesser exposures, slowed peripheral nerve conduction has been reported (19). Patients with high blood lead levels may present with severe, intractable colic, motor clumsiness, clouded consciousness, weakness, and paralysis. Lead has adverse effects on both male and female reproduction. The fetotoxic properties of lead were known to British factory inspectors at the end of the nineteenth century; they found a high incidence of stillbirths (60%), neonatal deaths, and a decreased fertility rate in women employed in the ceramic industry (20). Males may manifest decreased sperm counts and teratospermia (21). Hypertension has been associated with acute lead poisoning, along with renal failure. At lesser exposures, both experimental and epidemiological evidence of interference with renal function and elevations in blood pressure have been reported. Using data from the third National Health and Nutrition Examination Survey (NHANES III), a recent reevaluation of blood pressure in relation to contemporary blood lead levels found that black men and women had higher blood lead levels (5.4 µg/dl, 3.4 µg/dl) than their white counterparts (4.4 µg/dl, 3.0 µg/dl). Black subjects, both men and women, had a statistically significant association of blood lead with blood pressure after covariate adjustment. The association was not seen in whites (22). In its alkyl form, lead is a powerful neurotoxin. When tetraethyl lead (TEL) was first produced for use as a motor fuel additive in 1925, workers at all three operating plants began to die. After a brief moratorium imposed by the Surgeon General, production resumed and continued until the 1980s. TEL is fat-soluble; absorption through the skin and uptake by the brain is rapid. Because of growing evidence of neurotoxic effects at low doses, TEL was removed from gasoline in stepwise fashion beginning in 1978. Lead has been classified as an animal carcinogen, but the data on human carcinogenesis are considered inadequate. Some recent studies of cancer rates in lead trade workers (e.g., smelters, painters, body and fender repairmen) have shown an increase in standard mortality rates, but others have not (23).

CHILDHOOD LEAD TOXICITY Children are more sensitive to lead than adults for several reasons: Their exposure is increased by their universal hand-to-mouth activity; their gut absorbs lead more readily than an adult’s; and the developing CNS is more vulnerable to toxicants than the mature CNS. At high doses, generally blood lead levels >60 µg/dl, clinical symptoms become visible in children. Abdominal pain and arthralgia are common early complaints. Clumsiness and staggering may be seen, followed by headache and behavioral changes, which are signs of early encephalopathy. This may progress to alterations of consciousness, stupor, and convulsions. Encephalopathy, fortunately, has become rare in the United States. A high percentage of those children who recover

3 Dec 2003

21:51

AR

AR206-ME55-13.tex

AR206-ME55-13.sgm

LaTeX2e(2002/01/18)

LEAD POISONING

P1: GBC

213

from clinical encephalopathy have severe cognitive, attentional, and behavioral impairments.

Annu. Rev. Med. 2004.55:209-222. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

ASYMPTOMATIC LEAD TOXICITY In the 1960s, the accepted toxic threshold for lead in children was 60 µg/dl. Screening studies in eastern U.S. cities found that 10%–20% of inner-city children had blood lead levels over 40 µg/dl. This finding raised the conjecture, first made by Byers & Lord in 1943, that a proportion of school failure and behavioral disorder resulted from unrecognized lead toxicity. Five studies of lead levels and behavior in children without signs or symptoms of classical lead toxicity were undertaken in the early 1970s. Three reported an association between lead and IQ (24–26); two did not (27, 28). These early studies were constrained by design flaws: The number of subjects in each study was small, and each relied on blood lead levels to rank exposure. Blood lead, a short-term marker, may misclassify earlier exposure. Lead in blood has a half-life of 35 days. Exposure peaks at two years, then drops. Most studies enrolled children at about age 6 or later. There are no data on the validity of blood lead measurements four years after exposure has ended. Some studies used relatively insensitive outcome measures, such as group or screening tests. Control of confounders and statistical procedures was limited. Some investigators studied clinic samples and their data may thus have suffered selection bias. In 1979, we conducted a study that attempted to address these design issues. Lead concentration in deciduous teeth was selected as the marker of exposure. The sample comprised asymptomatic primary school students from the public schools of Chelsea and Somerville, Massachusetts. Subjects were classified on a large number of covariates, and these were controlled in the analysis by Analysis of Covariance (ANCOVA). Excluded were children with a history of lead exposure or toxicity. Children with elevated lead levels (in the ninetieth percentile for lead concentration) in their teeth were found, after covariate control, to be significantly impaired on the Wechsler Intelligence Scale for Children—Revised (WISC-R) IQ test, on language processing, and on reaction time under varying conditions of delay, a measure of attention. When teachers’ ratings of 2146 children on an 11-item forced-choice scale were classified by dentine lead level, we found a dosedependent increase in bad classroom behavior in direct relation to tooth lead level (Figure 1) (29). In the 1980s, following the removal of lead from gasoline, the blood lead levels in the referent group dropped. This enabled well-designed studies, employing larger samples, better measures of outcome and lead burden, and more sophisticated statistical analyses, to discover effects of lower blood lead levels (30–33). Three meta-analyses confirmed that low-level lead exposure was associated with IQ deficits (34–36). In response to the new data, in 1991, the Centers for Disease Control revised its limit of acceptable blood lead level downward in steps, from 60 µg/dl in the 1970s to its current status of 10 µg/dl.

3 Dec 2003

21:51

Annu. Rev. Med. 2004.55:209-222. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

214

AR

AR206-ME55-13.tex

AR206-ME55-13.sgm

LaTeX2e(2002/01/18)

P1: GBC

NEEDLEMAN

Figure 1 Distribution of negative teachers’ ratings in relation to dentine lead concentrations.

In 2002 and 2003, two new studies found effects at levels below 10 µg/dl, further lowering the observed threshold for effect. Lanphear et al. examined psychological performance and blood lead levels in 4853 children who were NHANES III subjects. The mean blood lead in this sample was 1.9 µg/dl; 2.1% of the sample had blood lead levels in excess of 10 µg/dl. After adjustment for covariates, significant inverse relationships between blood lead and math and reading subtests and the Block Design and Digit Span of the WISC-III were reported down to 2.5 µg/dl (37). Canfield et al. (38) studied 166 36-month-old children whose mean blood lead level was 7.9 µg/dl. Seventy-three percent of the subjects had blood lead levels under 10 µg/dl. Significant inverse relationships were found between IQ scores and lead after covariate adjustment. The slope of the effect was greater at the lower blood levels of lead (38). The import of the recent studies is that a threshold for lead and neurobehavioral function has not yet been demonstrated.

FOLLOW-UP STUDIES OF CHILDREN The late effects of early-childhood lead exposure have been examined by several investigators, who found persistence of deficits over time. Our subjects, first examined in 1979, were seen again 12 years later at mean age of 18.7 years. Subjects whose dentine lead levels were in the high-lead group (ninetieth percentile) had more school failure, reading disabilities, lower class standing in their final year of high school, and disturbances in fine motor function (39) (Figure 2). Fergusson et al. followed a sample of New Zealand children into their eighteenth year and reported that elevated lead levels were associated with poorer reading

3 Dec 2003

21:51

AR

AR206-ME55-13.tex

AR206-ME55-13.sgm

LaTeX2e(2002/01/18)

P1: GBC

Annu. Rev. Med. 2004.55:209-222. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

LEAD POISONING

215

Figure 2 Rate of failure to graduate from high school in relation to dentine lead levels.

scores, more failure to graduate from high school, and poorer examination scores (40).

FORWARD STUDIES OF PRENATAL AND INFANT LEAD EXPOSURE Bellinger et al. found significant associations between umbilical blood lead levels and neurodevelopmental scores at 2 years of age (41). At later ages, the association between umbilical cord blood lead and outcome was attenuated. The 2-year blood lead concentrations, however, were significantly related to scores at 10 years of age. Exposure at 2 years had no observable threshold, demonstrating neurotoxic effects at blood lead levels below 10 µg/dl. Dietrich et al. followed a group of 253 children from birth until 6 years of age. Blood lead levels at age 6 were associated with deficits in performance scores of the WISC, after covariate adjustment, and in motor function (42).

LEAD AND BEHAVIOR Cognitive function, measured by psychometric IQ tests, has been the major focus of most studies of lead exposure in childhood. There are persuasive reasons to believe that cognitive dysfunction may not be the most important effect of lead, and that we may be entering a fifth stage of understanding of lead’s effects, in which lead is recognized to adversely affect social behavior.

3 Dec 2003

21:51

Annu. Rev. Med. 2004.55:209-222. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

216

AR

AR206-ME55-13.tex

AR206-ME55-13.sgm

LaTeX2e(2002/01/18)

P1: GBC

NEEDLEMAN

This is not an entirely new notion. Parents have frequently reported that after recovery from an episode of acute lead poisoning, their child’s behavior changed dramatically, and they became restless, inattentive, and aggressive. In 1943, Byers & Lord reported attentional dysfunction and aggression in a sample of lead-poisoned children on follow-up (4). We studied 301 primary-school students and found that children with elevated bone lead levels scored higher on the attention deficit, aggression, and delinquency clusters of the Child Behavior Checklist after adjustment of covariates (43). Dietrich et al. found that prenatal lead exposure was associated with parents’ reports of delinquency and aggression, and postnatal lead exposure was associated with self reports of delinquent acts (44). A recent case-control study of 195 arrested and convicted delinquent youths found an increased risk of delinquency associated with bone lead concentrations measured by X-ray fluorescence. The covariateadjusted odds ratio was 4 (95% CL 1.4–11.1). The population-attributable risk for delinquency due to lead exposure ranged from 11% to 38% in this sample (45). A number of recent ecological investigations correlating leaded gasoline sales or ambient lead levels with crime rates support an association between lead exposure and crime. Stretesky & Lynch compared homicide rates in 3311 counties in the United States (46) After adjustment for 15 covariates, they reported a fourfold increase in homicide rates in those counties with the highest air lead levels compared to controls. Nevin correlated sales of leaded gasoline with violent crime rates and, adjusting for unemployment and percent of population in the high-crime age group, found a statistically significant association (47). It has been speculated that one of the reasons for the recent decline in crime rates is decreased exposure to lead.

LEAD EFFECTS IN OLDER SUBJECTS The greatest storage pool for lead is in bone, and the question of lead’s fate in older subjects when bone demineralizes has attracted considerable speculation. It is estimated that 50% of trabecular bone in women is lost over a lifetime. Lead is mobilized when bone resorption begins; significantly higher blood lead levels have been measured in postmenopausal women than in premenopausal women (48). Elevated lead appears to adversely affect cognitive function in elderly subjects as well. Older women (mean age 70.5 years) with blood lead levels >8 µg/dl had poorer performance on cognitive measures and slower reaction times than women with blood lead levels <3 µg/dl after covariate adjustment (49). Results of a large-cohort study of former TEL workers support a causal association between lead and dementia (50). The subjects (n = 535) were studied a mean of 16 years after their workplace exposure. The investigators report elevated bone lead levels and dose-related deficits in verbal and visual memory, executive ability, and manual dexterity. TEL workers exhibited a greater decline in function measured at yearly intervals than controls. The same investigators examined the interaction between APOE genotypes with bone lead levels. An interaction effect for bone lead × APOE genotype was found for 19 of 20 regression models, indicating that the toxic effect of lead is greater in subjects with at least one APOE allele (51).

3 Dec 2003

21:51

AR

AR206-ME55-13.tex

AR206-ME55-13.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:209-222. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

LEAD POISONING

P1: GBC

217

Lead has induced apoptosis in a number of experimental systems, including rat midbrain (52), rat testis (53), rat fibroblasts (54), rodent lung (55), and rodent retinal rod cells (56, 57). These findings, and similarities in the distribution of lead exposure and the rates of Alzheimer’s, make the topic of lead-related dementia worthy of further study. Lead exposure may also decrease lifespan. This is borne out by a recent study of subjects from the second NHANES study (1976–1980), who were followed up in 1992. The mortality of 4292 subjects with blood lead levels of 20–29 µg/dl was compared to those with levels <10 µg/dl. Subjects with higher lead levels had a 46% increased all-cause mortality, 39% increased cardiovascular mortality, and 68% increased cancer mortality (58).

DIAGNOSIS AND MANAGEMENT OF CLINICAL LEAD POISONING In adults, lead toxicity should be considered in the differential diagnosis of abdominal pain, arthralgia, hypertension, severe headache, increased intracranial pressure, CNS dysfunction, anemia, and renal dysfunction. An occupational history and an inventory of possible sources of exposure are useful. A blood lead level >10 µg/dl should be considered elevated, even though clinical symptoms are rarely seen below 60 µg/dl. Any child with growth failure, abdominal pain, behavior change, hyperactivity, language delay, or anemia should have a blood lead test to rule out lead toxicity. The cornerstone of lead toxicity management is the termination of exposure. For children, this means inspection of the home, and if this does not reveal lead, a survey of other possible sources. For lead levels >40 µg/dl, chelation therapy is effective in lowering the blood lead level. Calcium disodium edathamil (EDTA) was the preferred method until recently, when dimercaptosuccinic acid (succimer), an oral agent, was found to have equal efficacy. Both agents will reduce an elevated blood lead level to 40%–50% of its baseline. After treatment is concluded (5 days for EDTA, 19 days for succimer), body pools tend to equilibrate, and blood lead levels begin to rise, often requiring repeated courses. EDTA has drastically reduced the mortality rate from encephalopathy, but its efficacy at lower exposures has never been systematically studied. As a result, whether it conveys any benefit to children without encephalopathy remains unknown. After succimer had been in use for a few years, a multicenter study evaluated its efficacy in children with moderate elevations of lead (25–44 µg/dl). Blood lead levels in the treatment group were reduced to significantly lower levels than controls at the completion of treatment, but two years later, there were no differences between the two groups. At the conclusion of the study, no significant differences were found between treatment subjects and controls in cognitive, behavioral, or neuropsychological function (59). The only remedy at this time for low-level lead exposure is therefore primary prevention.

3 Dec 2003

21:51

218

AR

AR206-ME55-13.tex

AR206-ME55-13.sgm

LaTeX2e(2002/01/18)

P1: GBC

NEEDLEMAN

Annu. Rev. Med. 2004.55:209-222. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

PRIMARY PREVENTION In the early 1970s, the question of silent lead toxicity became the focus of intense controversy because of its regulatory implications. In 1973, when the Environmental Protection Agency began examining the health effects of TEL, industrial representatives claimed that the associations between lead and IQ were spurious, and that removing lead from gasoline would have no impact on body lead burdens. In 1977, after review of the health effects, the Environmental Protection Agency established an air lead standard of 1.5 µg/M3. The stepwise removal of lead from gasoline, based on the new air standard, began in the late 1970s. Figure 3 shows the effect of removing lead from gasoline on blood lead levels in the United States between 1975 and 1980. With the removal of lead from gasoline, a single major source remains for American children: leaded paint. Although it has been banned in household paint since 1971, 80% of the houses built before 1950, or 23,000,000 units, contain leaded paint. A cost-benefit analysis by the Public Health Service estimated the cost of abatement in these houses over a 30-year period at $33.7 billion in 1991. The estimated benefit from avoided health care costs and increased income due to raised IQ was $61.7 billion. This cost analysis may be conservative; it does not include avoided delinquency and cardiovascular disease, both demonstrated effects of lead exposure, among the health effects (60).

Figure 3 Parallel decreases in blood lead values observed in the NHANES II study and amounts of lead used in gasoline during 1976–1980.

3 Dec 2003

21:51

AR

AR206-ME55-13.tex

AR206-ME55-13.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:209-222. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

LEAD POISONING

P1: GBC

219

Current analyses also demonstrate that primary prevention yields large economic benefits. Grosse et al. calculated that each present-day preschool child’s IQ was increased by 2.2–4.7 points over what it would have been had the reduction in leaded gasoline and blood lead not taken place (61). From this, they calculated the IQ-related increase in income and estimated the economic benefit for each year’s birth cohort of 3.8 million children. The benefit range for the 1998 birth cohort was between $110 billion and $319 billion (61). Landrigan et al., assuming no threshold for the lead-IQ association, estimated the loss of future earnings for the one-year cohort of children aged 5 in 1997 at $43.4 billion (62). The evidence that lead toxicity extends down to the lowest measurable levels, that pharmacological therapies are ineffective at preventing sequelae in those with low levels, and that reduction of exposure yields huge economic as well as health benefits are strong warrants for a systematic program of abatement of lead from the single remaining major source: lead in older homes. The Annual Review of Medicine is online at http://med.annualreviews.org

LITERATURE CITED 1. Gilfillan SC. 1965. Lead poisoning and the fall of Rome. J. Occup. Med. 7:53–60 2. Gibson JL. 1892. Notes on lead-poisoning as observed among children in Brisbane. Proc. Intercolonial Med. Congr. Austr. 3:76–83 3. Blackfan KD. 1917. Lead poisoning in children with especial reference to lead as a cause of convulsions. Am. J. Med. Sci. 53:877–87 4. Byers RK, Lord EE. 1943. Late effects of lead poisoning on mental development. Am. J. Dis. Child. 66:471–83 5. Bailey C, Kitchen I. 1985. Ontogenesis of proenkephalin products in rat striatum and the inhibitory effects of low-level lead exposure. Dev. Brain Res. 22:75–79 6. Brown RS, Hingerty BE, Dewan JC, et al. 1983. Pb(II)-catalysed cleavage of the sugar-phosphate backbone of yeast tRNA(Phe)—implications for lead toxicity and self-splicing RNA. Nature 303:543– 46 7. Markovac J, Goldstein GW. 1988. Picomolar concentrations of lead stumulate brain protein kinase C. Nature 334:71–73

8. Simons T. 1993. Lead-calcium interactions in cellular lead toxicity. Neurotoxicology 14:77–85 9. Holtzman D, DeVries C, Nguyen H, Bensch K. 1984. Maturation of resistance to lead encephalopathy: cellular and subcellular mechanisms. Neurotoxicology 5:97–124 10. Bressler J, Goldstein G. 1991. Mechanisms of lead neurotoxicity. Biochem. Pharnacol. 41:479–84 11. Krigman MR. 1978. Neuropathology of heavy metal intoxication. Environ. Health Persp. 26:117–20 12. Pentschew A. 1965. Morphology and morphogenesis of lead encephalopathy. Acta Neuropathol. 5:133–60 13. Averill D, Needleman HL. 1980. Neonatal lead exposure retards cortical synaptogenesis in the rat. In Low Level Lead Exposure: The Clinical Implications of Current Research, ed. HL Needleman, pp. 201–10. New York: Raven 14. Wilson MA, Johnston MV, Goldstein GW, et al. 2000. Neonatal lead exposure impairs development of rodent barrel field

3 Dec 2003

21:51

220

15.

Annu. Rev. Med. 2004.55:209-222. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

16.

17.

18.

19.

20. 21.

22.

23.

24.

25.

26.

AR

AR206-ME55-13.tex

AR206-ME55-13.sgm

LaTeX2e(2002/01/18)

P1: GBC

NEEDLEMAN cortex. Proc. Natl. Acad. Sci. USA 97: 5540–45 Rice DC. 1992. Behavioral impairment produced by developmental lead exposure: evidence from primate reseach. In Human Lead Exposure, ed. HL Needleman, pp. 137–54. Boca Raton, FL: CRC Bushnell PJ, Bowman RE. 1979. Effects of chronic lead ingestion on social development in infant rhesus monkeys. Neurobehav. Tox. 1:207–19 Cory-Slechta DA. 1997. Relationships between Pb-induced changes in neurotransmitter system function and behavioral toxicity. Neurotox. 18:673–88 Nation JR, Baker DM, Taylor B, et al. 1986. Dietary lead increases ethanol consumption in the rat. Behav. Neurosci. 100:525– 30 Seppalainen A, Hernberg S. 1980. Subclinical lead neuropathy. Am. J. Ind. Med. 1:413–20 Oliver ST. 1911. A lecture on lead poisoning and the race. BMJ. 1:1096–98 Lancrajan, Popescu JL, Gavenescu O, et al. 1975. Reproductive ability of workmen occupationally exposed to lead. Arch. Environ. Health. 30:396–401 Vuppurturi S, Muntner P, Bazzano LA, et al. 2003. Blood lead level is associated with elevated blood pessure in blacks. Hypertension 41:463–68 Landrigan PJ, Boffetta P, Apostoli P. 2000. The reproductive toxicity and carcinogenicity of lead: a critical review. Am. J. Ind. Med. 38:231–43 De la Burde B, NcLin S, Choate S, 1972. Does asymptomatic lead exposure in children have latent sequelae? J. Pediatr. 81:1088–91 Landrigan PJ, Baloh RW, Barthel WF, et al. 1975. Neuropsychological dysfunction in children with low level lead absorption. Lancet, pp. 708–12, March 29 Perino J, Ernhart CB. 1974. The relation of subclinical lead level to cognitive and sensorimotor impairment in black preschoolers. J. Learn. Disabil. 7:26–30

27. Kotok D. 1972. Development of children with elevated blood lead levels: a controlled study. J. Pediatr. 80:57–61 28. Lansdown RG, Shepherd J, Clayton BE, et al. 1974. Blood-lead levels, behaviour, and intelligence. A population study. Lancet 1:538–41 29. Needleman HL, Gunnoe C, Leviton A, et al. 1979. Deficits in psychological and classroom performance of children with elevated dentine lead levels. N. Engl. J. Med. 300:689–95 30. Fulton M, Raab G, Thomson G, et al. 1987. Influence of blood lead on the ability and attainment of children in Edinburgh. Lancet 1:1221–26 31. Lyngbye T, Hansen ON, Trillingsgaard A, et al. 1990. Learning disabilities in children: significance of low level lead exposure and confounding factors. Acta Pedriatr. Scand. 79:352–60 32. Fergusson DM, Horwood J, Lynskey MT. 1997. Early dentine lead levels and educational outcomes at 18 years. J. Child Psychol. Psychiatry 38:471–78 33. McMichael AJ, Baghurst PA, Wigg NR, et al. 1988. Port Pirie Cohort Study: environmental exposure to lead and children’s abilities at the age of four years. N. Engl. J. Med. 319:468–75 34. Needleman HL, Gatsonis C. 1990. Low level lead exposure and the IQ of children. J. Am. Med. Assoc. 263(5):673–78 35. Schwartz J. 1993. Beyond LOEL’s, p values and vote counting: methods for looking at the shapes and strengths of associations. Neurotoxicology 14:237–46 36. Pocock SJ, Smith M, Baghurst P. 1994. Environmental lead and children’s intelligence: a systematic review of the epidemiological evidence. BMJ 309:11889– 97 37. Lanphear BP, Dietrich K, Auinger P, Cox C. 2000. Cognitive deficits associated with blood lead concentrations <10 microg/dL in US children and adolescents. Public Health Rep. 115(6):521–29 38. Canfield RL, Henderson CR, Cory-Slechta

3 Dec 2003

21:51

AR

AR206-ME55-13.tex

AR206-ME55-13.sgm

LaTeX2e(2002/01/18)

LEAD POISONING

39.

Annu. Rev. Med. 2004.55:209-222. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

DA, et al. 2003. Intellectual impairments in children with blood lead concentrations below 10 µg per deciliter. N. Engl. J. Med. 348:1517–26 Needleman HL, Schell A, Bellinger D, et al. 1990. The long-term effects of exposure to low doses of lead in childhood. An 11-year follow-up report. N. Engl. J. Med. 11:83– 88 Fergusson DM, Horwood J, Lynskey MT. 1997. Early dentine lead levels and educational outcomes at 18 years. J. Child Psychol. Psychiatry 38:471–78 Bellinger D, Leviton A, Waternaux C, et al. 1987. Longitudinal analyses of prenatal and postnatal lead exposure and early cognitive development. N. Engl. J. Med. 316:1037–43 Dietrich KN, Krafft KM, Bornschein RL, et al. 1987. Low-level fetal lead exposure effect on neurobehavioral development in early infancy. Pediatrics 80:721–30 Needleman HL, Riess JA, Tobin MJ, et al. 1996. Bone lead levels and delinquent behavior. J. Am. Med. Assoc. 275:363– 69 Dietrich KN, Ris MD, Succop PA, et al. 2001. Early exposure to lead and juvenile delinquency. Neurotoxicol. Teratol. 23:511–18 Needleman H, McFarland CE, Ness R, et al. 2002. Bone lead levels in adjudicated delinquency: a case-control study. Neurotoxicol. Teratol. 24:711–17 Stretesky PB, Lynch MJ. 2001. The relationship between lead exposure and homicide. Pediatr. Adolesc. Med. 155:579– 82 Nevin R. 2000. How lead exposure relates to temporal changes in IQ, violent crime, and unwed pregnancy. Environ. Res. 83:1– 22 Silbergeld EK, Schwartz J, Mahaffey K. 1988. Lead and osteoporosis: mobilization of lead from bone in postmenopausal women. Environ. Res. 47:79–94 Muldoon SB, Cauley JA, Kuller LH, et al. 1996. Effects of blood lead levels on cog-

50.

51.

52.

53.

54.

55.

56.

57.

58.

59.

60.

P1: GBC

221

nitive function of older women. Neuroepidemiology 15:62–72 Schwartz BS, Stewart WF, Bolla KI, et al. 2000. Past adult lead exposure is associated with longitudinal decline in cognitive function. Neurology 8:1144–50 Stewart WF, Schwartz BS, Simon D, et al. 2002. APOE genotype, past adult lead exposure, and neurobehavioral function. Environ. Health Persp. 110:501–5 Tavakili-Nezhad M, Barron AJ, Pitts DK. 2001. Postnatal inorganic lead exposure decreases the number of spontaniously acive midbrain dopamine neurons in the rat. Neurotoxicology 22:259–69 Adhikari N, Sinha N, Narayan R, et al. 2001. Lead induces cell death in testes of young rats. J. Appl. Toxicol. 21:275–77 Iavocoli I, Carelli G, Sgambato A, et al. 2001. Lead inhibits growth and induces apoptosis in normal rat fibroblasts. Altern. Lab. Anim. 29:461–69 Shabani A, Rabbani A. 2000. Lead nitrate induced apoptosis in alveolar macrophages from rat lung. Toxicology 149:109– 14 He L, Poblenz AT, Medrano CJ, et al. 2000. Lead and calcium produce photoreceptor cell apoptosis by opening the mitochondria permeability pore. J. Biol. Chem. 275:12175–84 Fox DA, Campbell ML, Blocker YS. 1997. Functional alterations and apoptotic cell death in the retina following developmental or adult lead exposure . Neurotoxicology 18:645–64 Lustberg M, Silbergeld E. 2002. Blood lead levels and mortality. Arch. Intern. Med. 25:2443–49 Rogan WJ, Dietrich KN, Ware JH, et al. 2001. Treatment of Lead-Exposed Children Trial Group. The effect of chelation therapy with succimer on neuropsychological development in children exposed to lead. N. Engl. J. Med. 344:1421–26 US Dep. Health Hum. Serv., Public Health Serv./Cent. Dis. Control. 1991. Strategic plan for the elimination of childhood lead

3 Dec 2003

21:51

222

AR

AR206-ME55-13.tex

AR206-ME55-13.sgm

P1: GBC

NEEDLEMAN

poisoning. Prepared for Risk Manag. Subcomm. of Dep. Health Hum. Serv. 61. Grosse SD, Matte TD, Schwartz J, et al. 2002. Economic gains resulting from the reduction in children’s exposure to lead in the United States. Environ. Health Persp. 110:563–70

Annu. Rev. Med. 2004.55:209-222. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

LaTeX2e(2002/01/18)

62. Landrigan PJ, Schechter CB, Lipton JM, et al. 2002. Environmental pollutants and disease in American children: estimates of morbidity, mortality, and costs for lead poisoning, asthma, cancer, and developmental disabilities. Environ. Health Persp. 110:721–28

12 Dec 2003

14:0

AR

AR206-ME55-14.tex

AR206-ME55-14.sgm

LaTeX2e(2002/01/18) P1: GBC 10.1146/annurev.med.55.091902.105248

Annu. Rev. Med. 2004. 55:223–37 doi: 10.1146/annurev.med.55.091902.105248 c 2004 by Annual Reviews. All rights reserved Copyright ° First published online as a Review in Advance on Aug. 14, 2003

THE IMPACT OF MINIMALLY INVASIVE SURGICAL TECHNIQUES

Annu. Rev. Med. 2004.55:223-237. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Sir Ara Darzi and Yaron Munz The Department of Surgical Oncology and Technology, Imperial College London, Praed Street, W2 1NY London, United Kingdom; email: [email protected], [email protected]

Key Words innovation, technology, robotic, research, surgical care ■ Abstract The use of minimally invasive techniques (MIT) in patient care is well documented in ancient history; however, it was not until the 1990s that advancements in technology enabled surgeons to realize the true potential of this approach. The minimally invasive approach has revolutionized surgical care, significantly reducing postoperative pain, recovery time, and hospital stays with marked improvements in cosmetic outcome and overall cost-effectiveness. It is now used around the world and in all major fields of surgery, compelling changes in training programs in order to assure quality control and patient safety. The bond between surgeons practicing minimally invasive surgery (MIS) and the high-tech industry is of utmost importance to future developments. Surgical robotic systems represent the most technologically advanced product of this collaboration, and their potential application in MIS shows much promise. As technology advances, additional developments in MIT are likely.

INTRODUCTION Modern surgery, unlike modern medicine, retains its link with ancient traditions and thousands of years of collective experience. Consequently, surgeons have attempted to promote natural healing processes rather than suppress them. Hippocrates (460–375 BC), Celsius (AD 357–412), Galen (AD 129–201), and other physicians believed that a healthy body was maintained through careful attention to diet, environment, and exercise (1, 2). Modern experimental medicine diverged from this ancient tradition in the early twentieth century to explore new and unproved concepts of therapy assuming that the body’s response to illness or trauma is the normal physiologic reaction rather than the pathologic consequences of any given condition or pathology. The most important advancement in this direction has been the introduction of laparoscopic surgery or minimally invasive surgery (MIS). This new approach has completely revolutionized modern surgical practices, significantly changing the surgical way of thinking, surgical techniques, and 0066-4219/04/0218-0223$14.00

223

12 Dec 2003

14:0

Annu. Rev. Med. 2004.55:223-237. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

224

AR

DARZI

AR206-ME55-14.tex

¥

AR206-ME55-14.sgm

LaTeX2e(2002/01/18)

P1: GBC

MUNZ

all other aspects of modern surgical patient care (3). MIS significantly reduces postoperative pain, recovery time, and hospital stays and markedly improves cosmetic outcome and overall cost-effectiveness. Minimally invasive techniques (MIT) are currently implemented in a wide range of medical fields as well as in other sciences such as molecular biology and genetics. It is almost impossible to imagine that, until recently, procedures such as the removal of a gallbladder were considered major surgery with an average hospital stay of a week and a relatively high postoperative complication rate. During the earlier years of MIS, this approach raised ambivalence because of the lack of appropriate teaching and training, combined with increased complication rates and long learning curves. MIS was advisable only in limited conditions, when performed by those who had already completed their learning curves. Experience brought improved patient outcomes and perhaps a deeper understanding of how this art should be taught to the younger generation of surgeons. Nowadays, laparoscopic cholecystectomy is the “gold standard” procedure (4), performed routinely all over the world (5–7). Furthermore, this procedure, among others, is now performed as day surgery in a growing number of medical centers in the United States and Europe (8–11). Hospital stays have become significantly shorter because recovery from laparoscopic procedures is faster compared to the open technique. The economic impact on healthcare organizations attributed to shorter hospital stays and faster return to normal activities is undisputed (12). As MIT are continuously evolving, new technologies are introduced, pushing further the limits of this concept. The combination of technological advancements in surgical techniques and cost-effectiveness issues, as well as ethical and political issues, have raised the need for change in training programs, ensuring that lessons learned from the early days of MIS are fully implemented. New technologies, targeted at teaching and training outside the operating theater, are being developed to provide the skills needed to perform MIS without subjecting patients to unnecessary risks (13–16). The implementation of MIS has had a great impact on a wide range of fields, and in this review we highlight some of the most important procedures within each subspecialty and point out the most important areas influenced by this approach. Other, more focused surgical specialties utilizing MIT, such as neurosurgery and ENT, are beyond the scope of this review.

MINIMALLY INVASIVE SURGICAL TECHNIQUES General Surgery Although gynecologists were probably the first to use laparoscopy in the early 1960s, it was not until general surgeons started to use MIT that these techniques became an established approach and not just an alternative to conventional surgical techniques. The modern era of MIS really started when the French general surgeon Mouret realized that laparoscopy could be used for more than just exploration, and in 1987 he performed the first laparoscopic cholecystectomy (17). Only five years

12 Dec 2003

14:0

AR

AR206-ME55-14.tex

AR206-ME55-14.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:223-237. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

MINIMALLY INVASIVE SURGERY

P1: GBC

225

later, this procedure was acknowledged as the “gold standard” technique for the removal of the gall bladder (18–20). Since the early 1990s, surgeons have been pushing the limits of MIS by attempting the most complicated procedures known in surgery. MIT are now being utilized in all surgical subspecialties and in some nonsurgical specialties such as interventional radiology (21), cardiology (22), and gastroenterology (23). To date, most of the procedures known in surgery have been successfully managed by MIT; it is no longer a question of how to use these techniques but rather why, where, or when to use them. Following the success of the laparoscopic cholecystectomy, other MIS procedures gained acceptance as methods of choice, such as gastric fundoplication (Nissen’s and other types) (24–26), laparoscopic procedures for the treatment of morbid obesity, e.g., gastric banding (27–30) and Roux-en-Y (31–35), and solid organ resections for various conditions, e.g., splenectomy (for chronic idiopathic thrombocytopenic purpura, Gaucher’s disease, and others) (36–39) and adrenalectomy (for a variety of benign conditions) (40, 41). Colorectal surgery was never considered complicated using the open technique; however, with the application of MIT, it has become one of the most challenging types of surgery. It requires the most advanced psychomotor skills to perform and takes years to master (42–45). Any type of colorectal resection can be now safely performed using MIT; however, there is an ongoing debate within the surgical community regarding their use in patients with malignant disease (46–49). Some concerns were raised regarding the dissemination of cancer to the peritoneal cavity or port sites, either due to the pressurized carbon dioxide intraperitoneal environment (commonly known as the chimney effect) or by the use of devices such as the harmonic shirrs, which produce vapor while executing their task (50). Reports of port-site metastasis do exist in the literature, but these should be carefully examined; the majority of such complications result from technical inadequacy and insufficient laparoscopic skills rather than the pressurized gas environment (51, 52). MIS is increasingly accepted even for total mesorectal excision as long as surgeons follow the classic rules of cancer surgery; among other advantages, MIS provides better vision with powerful magnification, which enables the surgeon to avoid unnecessary damage to adjacent important structures (nerves, blood vessels, and other viscera) (53–55). The consensus among the surgical community remains solid regarding the use of laparoscopic surgery for malignant disease: It should be carried out only within randomized controlled trials. Space-age technology combined with advancements in MIT gave rise to the new field of surgical robotics (56). Systems such as the Zeus (Computer Motion, California) and the da Vinci surgical system (Intuitive Surgical, California), referred to as telemanipulator systems, have the potential to overcome most of the disadvantages of conventional MIS (57–59). Surgical robots fully restore stereoscopic vision (three-dimensional vision) and hand-eye coordination, both of which are significantly lost in conventional MIS (60, 61). The robotic working tips restore all seven degrees of freedom of motion exactly like the human wrist. In addition to tremor abolition and motion scaling, they enhance technical dexterity, allowing

12 Dec 2003

14:0

Annu. Rev. Med. 2004.55:223-237. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

226

AR

DARZI

AR206-ME55-14.tex

¥

AR206-ME55-14.sgm

LaTeX2e(2002/01/18)

P1: GBC

MUNZ

Figure 1 Port positioning requires experience with all features of the system. This figure demonstrates positioning of the da Vinci arms for a robotic-assisted laparoscopic Nissen’s fundoplication in a patient suffering from gastroesophageal reflux disease (St. Mary’s Hospital, London, United Kingdom, 2002).

surgeons to perform complicated procedures with greater precision and perhaps less damage to the surroundings (62). Currently, these systems are challenged by surgeons in the search for specific procedures where patients, surgeons, and health care systems may benefit most (Figure 1). Because this technology is still in its infancy, it is believed that further advancements in the fields of robotics, computing, and imaging will have even greater impact on the future of surgery (63–65).

Gynecology Soon after general surgery had embraced MIS, gynecologists realized the full potential of the technique they had practically invented and began using it in a variety of procedures and conditions (66). MIS has become the preferred method of treatment of a variety of adnexal conditions such as benign ovarian cysts (67) and ectopic pregnancy (68). Surgery during pregnancy has become much more feasible with the rapid advancements in MIT and the introduction of highly specialized instrumentation; procedures such as cholecystectomy and appendectomy during pregnancy are widely accepted by surgeons and patients alike (69). More advanced procedures such as myomectomy (70–75) and hysterectomy (76) for

12 Dec 2003

14:0

AR

AR206-ME55-14.tex

AR206-ME55-14.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:223-237. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

MINIMALLY INVASIVE SURGERY

P1: GBC

227

benign disease, or even reconstructive genitourinary surgery such as retropubic urethropexy, are also well accepted (77–79). In all surgical fields, however, there is still some uncertainty regarding the use of MIT in the management of malignant disease (80–83). Currently, a preponderance of data supports the oncological use of MIT; however, only results from randomized controlled trials could validate its use for malignant disease. As results from prospective trials are beginning to emerge, it is becoming clear that MIS can be safely used for cancer operations, provided that the same basic rules that apply to the open technique are carefully followed (84–87). Complicated procedures, such as retroperitoneal lymphadenectomy for either therapeutic purposes or sampling for cancers of the cervix, uterus, ovaries, vagina, and fallopian tubes, are currently performed in a number of centers (88); however, the outcome improvements of these procedures cannot be assessed yet. With the introduction of telemanipulator surgical robotic systems, more procedures can be performed, thus further expanding the use of MIT. The field of female infertility is perceived as one of the major beneficiaries of robotic surgery because delicate procedures such as fallopian tube anastomosis (89, 90) or lysis of adnexal adhesions can be now performed with higher rates of success than before (91). At present, the range of gynecological procedures that justify robotic use seems quite limited, as cost-effectiveness issues determine the disbursement of resources; however, this may change in the future, as the variety of systems offered will grow while costs will surely drop.

Urology The use of MIT in urology has deep roots; procedures such as transurethral prostatectomy have been performed for almost 40 years. However, MIT in urological surgery gained wide acceptance only during the 1990s (92). Urologists and surgeons are now performing adrenalectomies for a wide range of conditions, including Conn’s syndrome and incidentalomas (93). Nephrectomies and especially donor nephrectomies (94) have been adopted as the procedure of choice by a growing number of transplant surgeons. These procedures are supported by the public because they offer the advantages of MIS to otherwise perfectly healthy patients (donors) (95). More challenging procedures, such as pyloplasty and uretero-uretrostomies, as well as bladder reconstructions are carried out with considerable success (96–98). It is not surprising that surgical robots are now deployed in urology. Adrenalectomies as well as nephrectomies are currently performed using robotic systems, utilizing their unsurpassed vision, precision, and dexterity enhancement where these are most likely to contribute to better operative and overall outcomes (99).

Cardiothoracic Surgery Cardiothoracic surgery is perhaps the only field where conventional MIS never gained much popularity, although the potential benefits for the average cardiac patient were supposed to be the greatest (100, 101). At present, a relatively

12 Dec 2003

14:0

Annu. Rev. Med. 2004.55:223-237. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

228

AR

DARZI

AR206-ME55-14.tex

¥

AR206-ME55-14.sgm

LaTeX2e(2002/01/18)

P1: GBC

MUNZ

small number of cardiothoracic surgeons use MIT and have achieved considerable success. Procedures such as mitral valve replacement and reconstructive surgery for a number of heart conditions can be performed by means of MIS with high rates of success and minimal discomfort to patients. However, more complicated cardiac surgeries such as bypass procedures demand advanced laparoscopic skills and require a long learning curve (102–104). The uses of MIT in cardiothoracic surgery expanded with the introduction of robotic surgery, which was approved by the US Food and Drug Administration in 2001. At present, cardiothoracic surgery is the most appropriate setting for the use of surgical robotic systems (105–108). Robotics has broadened the range of procedures that can be carried out using MIT, overcoming all the disadvantages and restrictions that kept MIS from becoming accepted in this field. The use of MIT for other, probably simpler, thoracoscopic procedures such as lung biopsies, lung lobectomies, and others is more accepted and holds a number of advantages that are unique to this field as a result of the anatomy of the thorax. Because it is a rigid cavity, there is no need for insufflation of gas into the thorax in order to maintain adequate working space. Furthermore, the use of ports for insertion of other instruments is not always necessary, so the surgeon can use instruments usually used in the open technique with the additional advantages of better visualization and flexibility (109–116).

PATIENT CARE AND HEALTH SERVICES The use of MIT in medicine and especially in surgery has brought about fundamental changes in patient care and health services worldwide. Patients no longer perceive surgery as a threat to their well-being or their ability to regain their normal life style; thus, patients with surgically amenable conditions are now more likely to pursue this option, regardless of age (27, 117, 118). Less pain and faster recovery during the immediate postoperative period, a significantly shorter hospital stay, a faster return to normal lifestyle, and better cosmesis are all considered to be highly advantageous for the patient (119–125). The short hospital stay, one of the most recognized and highly appreciated advantages of MIT, is crucial in assessing the cost-effectiveness of this approach (126). Patient turnover has dramatically grown without the need to expand the number of hospital beds; thus, the savings for health care services grew significantly. The availability of a wide range of elective procedures amenable to MIS is the result of surgical advancement as well as public demand. Never has surgery been more patient-driven than it is today, probably because the short- and long-term benefits for patients make it so popular. The hospital stay for most of these operations, when carried out by the open technique, was at least a week, but with the use of MIT it dropped to 24 h or less when procedures were performed in day surgery units (124). Fast recovery and return to normal lifestyle have had an equal effect on national productivity. The availability of surgical robotic systems may further

12 Dec 2003

14:0

AR

AR206-ME55-14.tex

AR206-ME55-14.sgm

LaTeX2e(2002/01/18)

MINIMALLY INVASIVE SURGERY

P1: GBC

229

improve overall cost-effectiveness because they perform more efficiently, carrying out more procedures within the same period of time and with a reduced staff.

Annu. Rev. Med. 2004.55:223-237. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

SURGICAL TRAINING AND TECHNOLOGIES DEVELOPMENT During the earlier years of MIS, a steep rise in intra- and postoperative complication rates resulted from the lack of proper training of both surgeons and operatingtheater personnel (128, 129). These issues recently came under public scrutiny, raising complex issues such as clinical safety, quality control in surgery, and costeffectiveness. For centuries, the surgical profession has been taught according to the Halstedian method, or master-and-apprentice model (130); however, MIS requires more time to master than conventional surgery and is far more complicated to teach (131). New methods were needed in order to improve teaching, training, and consequently patients’ outcome (132–137). In the aviation industry, simulator training is a crucial part of any flying training program (138–141). It has been shown that one hour of simulator training equals half an hour of real flying (139, 142, 144, 145). This observation has driven the

Figure 2 The da Vinci system can be used in a number of settings. Here it is used to train general surgeons in a virtual operating room. Use of the da Vinci system provides a stress-free environment where different types of skills can be taught, trained, and assessed (St. Mary’s Hospital, London, United Kingdom, 2002).

12 Dec 2003

14:0

Annu. Rev. Med. 2004.55:223-237. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

230

AR

DARZI

AR206-ME55-14.tex

¥

AR206-ME55-14.sgm

LaTeX2e(2002/01/18)

P1: GBC

MUNZ

development of surgical simulators to teach MIT. Numerous studies have confirmed the ability of these simulators to teach the basic psychomotor skills required for performing in the laparoscopic environment (146–150). The greatest advantage of simulation is that it can provide objective assessment of performance, which can be used for feedback during training as well as skills assessment and revalidation (151–158). Utilizing these new technologies in surgery provides the logical solution for all issues discussed above, i.e., quality control, safety, and cost-effectiveness, thus improving the overall efficiency of surgical patient care (159–161). Other technologies are currently under investigation, and some are already being used as prototypes in research facilities around the world. Eye and hand motion tracking devices can determine a variety of parameters, which can be used for feedback in teaching and training. Augmented-reality technology can superimpose three-dimensional computerized tomography data over real-time imaging, creating anatomical transparency of the patient, which allows much more precision and efficiency in certain delicate procedures. Such technologies can be used both for training and for real procedures, ensuring efficient use of resources while providing the best service possible (Figure 2).

CONCLUSION MIT were used in a wide range of procedures long before they were applied to surgery, but only then did they receive proper attention and development. This approach was rapidly advanced by innovative surgeons in close collaboration with technologists (scientists, engineers, software and hardware experts, etc). Within the medical profession, this type of collaboration is unique to the surgical community (162, 163), especially to those who are directly involved in MIS. The bond between surgery and technology is now stronger than ever and is likely to get stronger in the future, but we should not forget that at the sharp end of these highly complicated technologies lie patients whose well-being is our first priority. The Annual Review of Medicine is online at http://med.annualreviews.org

LITERATURE CITED 1. Osler W. 1921. The Evolution of Modern Medicine. New Haven, CT: Yale Univ. Press 2. Nutton V. 1996. The rise of medicine. In The Cambridge Illustrated History of Medicine, ed. Porter R, pp. 52–81. Cambridge, UK: Cambridge Univ. Press 3. Darzi A, Mackay S. 2002. Recent ad-

vances in minimal access surgery. BMJ 324:31–34 4. Begos DG, Modlin IM. 1994. Laparoscopic cholecystectomy: from gimmick to gold standard. J. Clin. Gastroenterol. 19:325–30 5. Gaskin TA, Isobe JH, Mathews JL, et al. 1991. Laparoscopy and the general

12 Dec 2003

14:0

AR

AR206-ME55-14.tex

AR206-ME55-14.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:223-237. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

MINIMALLY INVASIVE SURGERY surgeon. Surg. Clin. N. Am. 71:1085–97 6. Peters JH, Ellison EC, Innes JT, et al. 1991. Safety and efficacy of laparoscopic cholecystectomy. A prospective analysis of 100 initial patients. Ann. Surg. 213:3– 12 7. Voyles CR, Petro AB, Meena AL, et al. 1991. A practical approach to laparoscopic cholecystectomy. Am. J. Surg. 161: 365–70 8. Hollington P, Toogood GJ, Padbury RT. 1999. A prospective randomized trial of day-stay only versus overnight-stay laparoscopic cholecystectomy. Aust. N. Z. J. Surg. 69:841–43 9. Huang A, Stinchcombe C, Phillips D, McWhinnie DL. 2000. Prospective 5-year audit for day-case laparoscopic cholecystectomy. Br. J. Surg. 87:362–73 10. Jensen BN, Larsen JF, Nielsen FB. 1999. Laparoscopic cholecystectomy as a sameday surgical procedure. Ugeskr. Laeger 161:5911–14 11. Kelstrup M, Poulsen KA, Frohmuller S, et al. 1999. Laparoscopic cholecystectomy as a same-day surgical procedure. Results and patient satisfaction. Ugeskr. Laeger 161:5928–30 12. Brazier JE, Johnson AG. 2001. Economics of surgery. Lancet 358:1077– 81 13. Bann SD, Khan MS, Darzi AW. 2003. Measurement of surgical dexterity using motion analysis of simple bench tasks. World J. Surg. 27:390–94 14. Darzi A, Mackay S. 2002. Skills assessment of surgeons. Surgery 131:121–24 15. Datta V, Mandalia M, Mackay S, et al. 2002. Relationship between skill and outcome in the laboratory-based model. Surgery 131:318–23 16. Gallagher AG, Satava RM. 2002. Virtual reality as a metric for the assessment of laparoscopic psychomotor skills. Learning curves and reliability measures. Surg. Endosc. 16:1746–52 17. Dubois F, Berthelot G, Levard H. 1991. Laparoscopic cholecystectomy: historic

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

P1: GBC

231

perspective and personal experience. Surg. Laparosc. Endosc. 1:52–57 Perissat J, Collet D, Belliard R, et al. 1992. Laparoscopic cholecystectomy: the state of the art. A report on 700 consecutive cases. World J. Surg. 16:1074–82 Soper NJ, Stockmann PT, Dunnegan DL, Ashley SW. 1992. Laparoscopic cholecystectomy. The new “gold standard”? Arch. Surg. 127:917–21 Suc B, Fontes DI, Fourtanier G, Escat J. 1992. 3606 cholecystectomies under celioscopy. The Register of the French Society of Digestive Surgery. Ann. Chir. 46:219–26 Raj GV, Reddan DJ, Hoey MB, Polascik TJ. 2003. Management of small renal tumors with radiofrequency ablation. Urology 61:23–29 Katzen BT. 2002. The future of catheterbased angiography: implications for the vascular interventionalist. Radiol. Clin. N. Am. 40:689–92 Neff R. 2001. Angiography and interventional radiology: percutaneous approaches to benign pancreatic disorders. Surg. Clin. N. Am. 81:359–61, xi Balaji NS, Peters JH. 2002. Minimally invasive surgery for esophageal motility disorders. Surg. Clin. N. Am. 82:763–82 Todd S, Corsnitz D, Ray S, Nassar J. 2002. Outpatient laparoscopic Nissen fundoplication. AORN J. 75:956, 959–64 Terry M, Smith CD, Branum GD, et al. 2001. Outcomes of laparoscopic fundoplication for gastroesophageal reflux disease and paraesophageal hernia. Surg. Endosc. 15:691–99 Zinzindohoue F, Chevallier JM, Douard R, et al. 2003. Laparoscopic gastric banding: a minimally invasive surgical treatment for morbid obesity: prospective study of 500 consecutive patients. Ann. Surg. 237:1–9 Cadiere GB, Himpens J, Hainaux B, et al. 2002. Laparoscopic adjustable gastric banding. Semin. Laparosc. Surg. 9:105– 14

12 Dec 2003

14:0

Annu. Rev. Med. 2004.55:223-237. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

232

AR

DARZI

AR206-ME55-14.tex

¥

AR206-ME55-14.sgm

LaTeX2e(2002/01/18)

P1: GBC

MUNZ

29. Pontiroli AE, Pizzocri P, Librenti MC, et al. 2002. Laparoscopic adjustable gastric banding for the treatment of morbid (grade 3) obesity and its metabolic complications: a three-year study. J Clin. Endocrinol. Metab. 87:3555–61 30. Vertruyen M. 2002. Experience with lap-band system up to 7 years. Obes. Surg. 12:569–72 31. Hamilton EC, Sims TL, Hamilton TT, et al. 2003. Clinical predictors of leak after laparoscopic Roux-en-Y gastric bypass for morbid obesity. Surg. Endosc. 32. McCormick JT, Papasavas PK, Caushaj PF, Gagne DJ. 2003. Laparoscopic revision of failed open bariatric procedures. Surg. Endosc. 17:413–15 33. Patterson EJ, Urbach DR, Swanstrom LL. 2003. A comparison of diet and exercise therapy versus laparoscopic Roux-en-Y gastric bypass surgery for morbid obesity: a decision analysis model. J. Am. Coll. Surg. 196:379–84 34. Schauer P, Ikramuddin S, Hamad G, Gourash W. 2003. The learning curve for laparoscopic Roux-en-Y gastric bypass is 100 cases. Surg. Endosc. 17:212–15 35. DeMaria EJ, Sugerman HJ, Kellum JM, et al. 2002. Results of 281 consecutive total laparoscopic Roux-en-Y gastric bypasses to treat morbid obesity. Ann. Surg. 235:640–45 36. Rosen M, Brody F, Walsh RM, Ponsky J. 2002. Hand-assisted laparoscopic splenectomy vs conventional laparoscopic splenectomy in cases of splenomegaly. Arch. Surg. 137:1348–52 37. Brodsky JA, Brody FJ, Walsh RM, et al. 2002. Laparoscopic splenectomy. Surg. Endosc. 16:851–54 38. Anthony ML, Hardee EM. 1999. Laparoscopic splenectomy in children with sickle cell disease. AORN J. 69:567–82 39. Park A, Targarona EM, Trias M. 2001. Laparoscopic surgery of the spleen: state of the art. Langenbecks Arch. Surg. 386:230– 39 40. Matsuda T, Murota T, Oguchi N, et al.

41.

42.

43.

44.

45.

46.

47.

48.

49.

50.

51.

2002. Laparoscopic adrenalectomy for pheochromocytoma: a literature review. Biomed. Pharmacother. 56(Suppl. 1): 132s–138s Sung GT, Gill IS. 2000. Laparoscopic adrenalectomy. Semin. Laparosc. Surg. 7: 211–22 Senagore AJ, Madbouly KM, Fazio VW, et al. 2003. Advantages of laparoscopic colectomy in older patients. Arch. Surg. 138:252–56 Braga M, Vignali A, Gianotti L, et al. 2002. Laparoscopic versus open colorectal surgery: a randomized trial on shortterm outcome. Ann. Surg. 236:759–66 Duepree HJ, Senagore AJ, Delaney CP, et al. 2002. Advantages of laparoscopic resection for ileocecal Crohn’s disease. Dis. Colon Rectum 45:605–10 Rose J, Schneider C, Scheidbach H, et al. 2002. Laparoscopic treatment of rectal prolapse: experience gained in a prospective multicenter study. Langenbecks Arch. Surg. 387:130–37 Canter RJ, Williams NN. 2002. Surgical treatment of colon and rectal cancer. Hematol. Oncol. Clin. N. Am. 16:907–26 Hartley JE, Monson JR. 2002. The role of laparoscopy in the multimodality treatment of colorectal cancer. Surg. Clin. N. Am. 82:1019–33 Wu FP, Sietses C, von Blomberg BM, et al. 2003. Systemic and peritoneal inflammatory response after laparoscopic or conventional colon resection in cancer patients: a prospective, randomized trial. Dis. Colon Rectum 46:147–55 Pikarsky AJ. 2001. Update on prospective randomized trials of laparoscopic surgery for colorectal cancer. Surg. Oncol. Clin. N. Am. 10:639–53 Hubens G. 2002. Port site metastases: Where are we at the beginning of the 21st century? Acta Chir. Belg. 102:230–37 Zmora O, Gervaz P, Wexner SD. 2001. Trocar site recurrence in laparoscopic surgery for colorectal cancer. Surg. Endosc. 15:788–93

12 Dec 2003

14:0

AR

AR206-ME55-14.tex

AR206-ME55-14.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:223-237. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

MINIMALLY INVASIVE SURGERY 52. Zmora O, Weiss EG. 2001. Trocar site recurrence in laparoscopic surgery for colorectal cancer. Myth or real concern? Surg. Oncol. Clin. N. Am. 10:625–38 53. Pikarsky AJ, Rosenthal R, Weiss EG, Wexner SD. 2002. Laparoscopic total mesorectal excision. Surg. Endosc. 16: 558–62 54. Poulin EC, Schlachta CM, Gregoire R, et al. 2002. Local recurrence and survival after laparoscopic mesorectal resection for rectal adenocarcinoma. Surg. Endosc. 16:989–95 55. Morino M, Parini U, Giraudo G, et al. 2003. Laparoscopic total mesorectal excision: a consecutive series of 100 patients. Ann. Surg. 237:335–42 56. Satava RM. 2001. Minimally invasive surgery and its role in space exploration. Surg. Endosc. 15:1530 57. Boyd WD, Kodera K, Stahl KD, Rayman R. 2002. Current status and future directions in computer-enhanced video- and robotic-assisted coronary bypass surgery. Semin. Thorac. Cardiovasc. Surg. 14:101–9 58. Hashizume M, Konishi K, Tsutsumi N, et al. 2002. A new era of robotic surgery assisted by a computer-enhanced surgical system. Surgery 131:S330–S333 59. Sung GT, Gill IS. 2001. Robotic laparoscopic surgery: a comparison of the DA Vinci and Zeus systems. Urology 58:893– 98 60. Munz Y, Hernandez JD, Bann SD, et al. 2002. The advantages of 3D visualization in surgical performance with the da Vinci tele-manipulation robotic system. JSLS 6:265 61. Gulbins H, Boehm DH, Reichenspurner H, et al. 1999. 3D-visualization improves the dry-lab coronary anastomoses using the Zeus robotic system. Heart Surg. Forum 2:318–24 62. Ballantyne GH. 2002. Robotic surgery, telerobotic surgery, telepresence, and telementoring. Review of early clinical results. Surg. Endosc. 16:1389–402

P1: GBC

233

63. Satava RM. 2003. Disruptive visions. Surg. Endosc. 17:104–7 64. Satava RM. 2002. Surgical robotics: the early chronicles: a personal historical perspective. Surg. Laparosc. Endosc. Percutan. Tech. 12:6–16 65. Satava RM, Jones SB. 2000. Preparing surgeons for the 21st century. Implications of advanced technologies. Surg Clin. N. Am. 80:1353–65 66. Loh FH, Hameed N, Ng SC. 2002. The impact of minimal access surgery on gynaecological surgery in a university gynaecological unit over a 10-year period from 1991 to 2000. Singapore Med. J. 43:177– 81 67. Mendilcioglu I, Zorlu CG, Trak B, et al. 2002. Laparoscopic management of adnexal masses. Safety and effectiveness. J. Reprod. Med. 47:36–40 68. Luciano AA, Roy G, Solima E. 2001. Ectopic pregnancy: from surgical emergency to medical management. Ann. NY Acad. Sci. 943:235–54 69. Fatum M, Rojansky N. 2001. Laparoscopic surgery during pregnancy. Obstet. Gynecol. Surv. 56:50–59 70. Dubuisson JB, Chapron C, Fauconnier A, Babaki-Fard K. 2001. Laparoscopic myomectomy fertility results. Ann. NY Acad. Sci. 943:269–75 71. Rossetti A, Sizzi O, Soranna L, et al. 2001. Long-term results of laparoscopic myomectomy: recurrence rate in comparison with abdominal myomectomy. Hum. Reprod. 16:770–74 72. Seidman DS, Nezhat CH, Nezhat F, Nezhat C. 2001. The role of laparoscopicassisted myomectomy (LAM). JSLS 5: 299–303 73. Di Gregorio A, Maccario S, Raspollini M. 2002. The role of laparoscopic myomectomy in women of reproductive age. Reprod. Biomed. Online. 4(Suppl. 3):55–58 74. Falcone T, Bedaiwy MA. 2002. Minimally invasive management of uterine fibroids. Curr. Opin. Obstet. Gynecol. 14: 401–7

12 Dec 2003

14:0

Annu. Rev. Med. 2004.55:223-237. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

234

AR

DARZI

AR206-ME55-14.tex

¥

AR206-ME55-14.sgm

LaTeX2e(2002/01/18)

P1: GBC

MUNZ

75. Seracchioli R, Colombo FM, Bagnoli A, et al. 2003. Laparoscopic myomectomy for fibroids penetrating the uterine cavity: Is it a safe procedure? Br. J. Obstet. Gynecol. 110:236–40 76. Kohler C, Hasenbein K, Klemm P, et al. 2003. Laparoscopic-assisted vaginal hysterectomy with lateral transsection of the uterine vessels. Surg. Endosc. 17:485– 90 77. Hannah SL, Roland B, Gengenbacher PM. 2001. Extraperitoneal retropubic laparoscopic urethropexy. J. Am. Assoc. Gynecol. Laparosc. 8:107–10 78. Walter AJ, Morse AN, Hammer RH, et al. 2002. Laparoscopic versus open Burch retropubic urethropexy: comparison of morbidity and costs when performed with concurrent vaginal prolapse repairs. Am. J. Obstet. Gynecol. 186:723–28 79. Zullo F, Morelli M, Russo T, et al. 2002. Two techniques of laparoscopic retropubic urethropexy. J. Am. Assoc. Gynecol. Laparosc. 9:178–81 80. Pejovic T, Nezhat F. 2001. Laparoscopic management of adnexal masses: the opportunities and the risks. Ann. NY Acad. Sci. 943:255–68 81. Kopjar M, Bukovic D, Zadro M, et al. 1999. Minimally invasive surgery in the field of gynecologic oncology. Coll. Antropol. 23:629–32 82. Canis M, Botchorishvili R, Wattiez A, et al. 2000. Cancer and laparoscopy, experimental studies: a review. Eur. J. Obstet. Gynecol. Reprod. Biol. 91:1–9 83. Canis M, Rabischong B, Botchorishvili R, et al. 2001. Risk of spread of ovarian cancer after laparoscopic surgery. Curr. Opin. Obstet. Gynecol. 13:9–14 84. Dargent DF. 2001. Laparoscopic surgery in gynecologic oncology. Surg. Clin. N. Am. 81:949–64 85. Newman L. 2001. Surgical oncology focusing on minimally invasive surgery, more randomized clinical trials. J. Natl. Cancer Inst. 93:897–900 86. Guicciardi MA, Andreoni B, Croce E.

87.

88.

89.

90.

91.

92. 93. 94.

95.

96.

97.

98.

99.

2000. Mini-invasive surgery may be advised in oncology. Crit. Rev. Oncol. Hematol. 34:169–73 Chi DS, Curtin JP. 1999. Gynecologic cancer and laparoscopy. Obstet. Gynecol. Clin. N. Am. 26:201–15 Guy L, Giraud B, Boiteux JP. 2001. Minimally invasive extraperitoneal lymphadenectomy: clinical validation of an original technique. J. Endourol. 15:903–8 Degueldre M, Vandromme J, Huong PT, Cadiere GB. 2000. Robotically assisted laparoscopic microsurgical tubal reanastomosis: a feasibility study. Fertil. Steril. 74:1020–23 Falcone T, Goldberg JM, Margossian H, Stevens L. 2000. Robotic-assisted laparoscopic microsurgical tubal anastomosis: a human pilot study. Fertil. Steril. 73:1040– 42 Margossian H, Falcone T. 2001. Robotically assisted laparoscopic hysterectomy and adnexal surgery. J. Laparoendosc. Adv. Surg. Tech. A 11:161–65 Jackson CL. 2001. Urologic laparoscopy. Surg. Oncol. Clin. N. Am. 10:571–78 Gill IS. 2001. The case for laparoscopic adrenalectomy. J. Urol. 166:429–36 Pace KT, Dyer SJ, Phan V, et al. 2003. Laparoscopic versus open donor nephrectomy. Surg. Endosc. 17:134–42 Lind MY, Liem YS, Bemelman WA, et al. 2003. Live donor nephrectomy and return to work. Surg Endosc. 17:591–95 Jarrett TW, Chan DY, Charambura TC, et al. 2002. Laparoscopic pyeloplasty: the first 100 cases. J. Urol. 167:1253–56 Soulie M, Salomon L, Seguin P, et al. 2001. Multi-institutional study of complications in 1085 laparoscopic urologic procedures. Urology 58:899–903 Pattaras JG, Moore RG. 2000. Laparoscopic pyeloplasty. J. Endourol. 14:895– 904 Gettman MT, Blute ML, Peschel R, Bartsch G. 2003. Current status of robotics in urologic laparoscopy. Eur. Urol. 43:106–12

12 Dec 2003

14:0

AR

AR206-ME55-14.tex

AR206-ME55-14.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:223-237. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

MINIMALLY INVASIVE SURGERY 100. Chitwood WR. 1998. State of the art review: videoscopic minimally invasive mitral valve surgery. Trekking to a totally endoscopic operation. Heart Surg. Forum 1:13–16 101. Nataf P, Al Attar N, Ramadan R, et al. 2000. Thoracoscopic IMA takedown. J. Card. Surg. 15:278–82 102. Falk V, Autschbach R, Krakor R, et al. 1999. Computer-enhanced mitral valve surgery: toward a total endoscopic procedure. Semin. Thorac. Cardiovasc. Surg. 11:244–49 103. Mohr FW, Onnasch JF, Falk V, et al. 1999. The evolution of minimally invasive valve surgery—2 year experience. Eur. J. Cardiothorac. Surg. 15:233–38 104. Grossi EA, Galloway AC, LaPietra A, et al. 2002. Minimally invasive mitral valve surgery: a 6-year experience with 714 patients. Ann. Thorac. Surg. 74:660– 63 105. Argenziano M. 2003. Totally endoscopic, robotic cardiac surgery. Heart Surg. Forum 6:104 106. Chitwood WR Jr. 2003. Robotic mitral valve surgery. Heart Surg. Forum 6:106– 7 107. Donias HW, Karamanoukian HL, D’Ancona G, Hoover EL. 2003. Minimally invasive mitral valve surgery: from port access to fully robotic-assisted surgery. Angiology 54:93–101 108. Nifong LW, Chu VF, Bailey BM, et al. 2003. Robotic mitral valve repair: experience with the da Vinci system. Ann. Thorac. Surg. 75:438–42 109. Manasse E, Infante M, Ghiselli S, et al. 2002. A video-assisted thoracoscopic technique to encircle the four pulmonary veins: a new surgical intervention for atrial fibrillation ablation. Heart Surg. Forum 5:337–39 110. Fukui S, Paraskevas N, Soury P, et al. 2003. Totally videoendoscopic descending thoracic aorta to femoral artery bypass. J. Vasc. Surg. 37:191–93 111. Perry Y, Fernando HC, Buenaventura

112.

113.

114.

115.

116.

117.

118.

119.

120.

121.

122.

P1: GBC

235

PO, et al. 2002. Minimally invasive esophagectomy in the elderly. JSLS 6: 299–304 Savcenko M, Wendt GK, Prince SL, Mack MJ. 2002. Video-assisted thymectomy for myasthenia gravis: an update of a single institution experience. Eur. J. Cardiothorac. Surg. 22:978–83 Perez P Jr. 2002. Thoracoscopy and videoassisted thoracic surgery. Rev. Invest. Clin. 54:364–65 Burdine J, Joyce LD, Plunkett MB, et al. 2002. Feasibility and value of videoassisted thoracoscopic surgery wedge excision of small pulmonary nodules in patients with malignancy. Chest 122: 1467–70 Swanson SJ, Batirel HF. 2002. Videoassisted thoracic surgery (VATS) resection for lung cancer. Surg. Clin. N. Am. 82:541–59 Moffatt SD, Mitchell JD, Whyte RI. 2002. Role of video-assisted thoracoscopic surgery and classic thoracotomy in lung cancer management. Curr. Opin. Pulm. Med. 8:281–86 Barthel SW, Strome M. 1999. Snoring, obstructive sleep apnea, and surgery. Med. Clin. N. Am. 83:85–96 Todd S, Corsnitz D, Ray S, Nassar J. 2002. Outpatient laparoscopic Nissen fundoplication. AORN J. 75:956, 959–54 Malina M, Nilsson M, Brunkwall J, et al. 2000. Quality of life before and after endovascular and open repair of asymptomatic AAAs: a prospective study. J. Endovasc. Ther. 7:372–79 Magovern JA, Benckart DH, Landreneau RJ, et al. 1998. Morbidity, cost, and sixmonth outcome of minimally invasive direct coronary artery bypass grafting. Ann. Thorac. Surg. 66:1224–29 Beals DA. 2002. Does minimal access mean minimal relationship? Defining the physician-patient relationship in postmodern culture. Semin. Laparosc. Surg. 9:218–21 Pace KT, Dyer SJ, Stewart RJ, et al. 2003.

12 Dec 2003

14:0

236

123.

124.

Annu. Rev. Med. 2004.55:223-237. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

125.

126.

127. 128.

129.

130.

131.

132.

133.

AR

DARZI

AR206-ME55-14.tex

¥

AR206-ME55-14.sgm

LaTeX2e(2002/01/18)

P1: GBC

MUNZ

Health-related quality of life after laparoscopic and open nephrectomy. Surg. Endosc. 17:143–52 Bruce J, Russell EM, Mollison J, Krukowski ZH. 2001. The measurement and monitoring of surgical adverse events. Health Technol. Assess. 5:1–194 Watts G. 2001. Day surgery. All in a day’s work. Health Serv. J. 111:1 (Suppl.) Edison MI, Horgan S, Helton WS. 2001. Using small-group workshops to improve surgical residents’ technical skills. Acad. Med. 76:557–58 Calland JF, Tanaka K, Foley E, et al. 2001. Outpatient laparoscopic cholecystectomy: patient outcomes after implementation of a clinical pathway. Ann. Surg. 233:704–15 Deleted in proof Krahenbuhl L, Sclabas G, Wente MN, et al. 2001. Incidence, risk factors, and prevention of biliary tract injuries during laparoscopic cholecystectomy in Switzerland. World J. Surg. 25:1325–30 Berg AA. 2002. “Licence” requirements for physicians who are to perform laparoscopic surgery! Frequency of complications still too high. Lakartidningen 99:16– 17 Toledo-Pereyra LH. 2002. William Stewart Halsted: father of American modern surgery. J. Invest Surg. 15:59–60 Nussbaum MS. 2002. Surgical endoscopy training is integral to general surgery residency and should be integrated into residency and fellowships abandoned. Semin. Laparosc. Surg. 9:212–15 Hawn MT, Cook D, Deveney C, Sheppard BC. 2002. Quality of life after laparoscopic bilateral adrenalectomy for Cushing’s disease. Surgery 132:1064–68 Nilsson G, Larsson S, Johnsson F. 2002. Randomized clinical trial of laparoscopic versus open fundoplication: evaluation of psychological well-being and changes in everyday life from a patient perspective. Scand. J. Gastroenterol. 37:385– 91

134. Granderath FA, Kamolz T, Schweiger UM, et al. 2002. Quality of life and symptomatic outcome three to five years after laparoscopic Toupet fundoplication in gastroesophageal reflux disease patients with impaired esophageal motility. Am. J. Surg. 183:110–16 135. Luketich JD, Fernando HC, Christie NA, et al. 2001. Outcomes after minimally invasive esophagomyotomy. Ann. Thorac. Surg. 72:1909–12 136. Velanovich V, Shurafa MS. 2001. Clinical and quality of life outcomes of laparoscopic and open splenectomy for haematological diseases. Eur. J. Surg. 167:23–28 137. Hasson HM. 1995. Presidential address. Endoscopic surgery, cost effectiveness, and the quality of life. J. Am. Assoc. Gynecol. Laparosc. 2:115–21 138. Preston N. 1992. Pilot training simulators. J. Air Med. Transp. 11:15–21 139. Bodrov VA, Kharin VV. 1989. Possibilities for the use of aviation simulators during dynamic medical observation of flight personnel. Voen. Med. Zh. 51–55 140. Braune RJ, Trollip SR. 1982. Towards an internal model in pilot training. Aviat. Space Environ. Med. 53:996–99 141. Ginsburg AP, Evans DW, Sekule R, Harp SA. 1982. Contrast sensitivity predicts pilots’ performance in aircraft simulators. Am. J. Optom. Physiol. Opt. 59:105– 9 142. Barker JM, Clothier CC, Woody JR, et al. 1996. Crew resource management: a simulator study comparing fixed versus formed aircrews. Aviat. Space Environ. Med. 67:3–7 143. Deleted in proof 144. Torkington J, Smith SG, Rees BI, Darzi A. 2001. Skill transfer from virtual reality to a real laparoscopic task. Surg. Endosc. 15:1076–79 145. Figert PL, Park AE, Witzke DB, Schwartz RW. 2001. Transfer of training in acquiring laparoscopic skills. J. Am. Coll. Surg. 193:533–37 146. Gallagher AG, Richie K, McClure N,

12 Dec 2003

14:0

AR

AR206-ME55-14.tex

AR206-ME55-14.sgm

LaTeX2e(2002/01/18)

MINIMALLY INVASIVE SURGERY

Annu. Rev. Med. 2004.55:223-237. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

147.

148.

149.

150.

151.

152.

153.

154.

McGuigan J. 2001. Objective psychomotor skills assessment of experienced, junior, and novice laparoscopists with virtual reality. World J. Surg. 25:1478– 83 Grantcharov TP, Bardram L, FunchJensen P, Rosenberg J. 2003. Learning curves and impact of previous operative experience on performance on a virtual reality simulator to test laparoscopic surgical skills. Am. J. Surg. 185:146–49 Haluck RS, Webster RW, Snyder AJ, et al. 2001. A virtual reality surgical trainer for navigation in laparoscopic surgery. Stud. Health Technol. Inform. 81:171–76 Hoffman H, Vu D. 1997. Virtual reality: teaching tool of the twenty-first century? Acad. Med. 72:1076–81 Kothari SN, Kaplan BJ, DeMaria EJ, et al. 2002. Training in laparoscopic suturing skills using a new computer-based virtual reality simulator (MIST-VR) provides results comparable to those with an established pelvic trainer system. J. Laparoendosc. Adv. Surg. Tech. A 12:167– 73 Cerilli GJ, Merrick HW, Staren ED. 2001. Objective Structured Clinical Examination technical skill stations correlate more closely with postgraduate year level than do clinical skill stations. Am. Surg. 67:323–26 Darzi A, Smith S, Taffinder N. 1999. Assessing operative skill. Needs to become more objective. BMJ 318:887–88 Darzi A, Mackay S. 2001. Assessment of surgical competence. Qual. Health Care 10(Suppl. 2):ii64–ii69 Darzi A, Datta V, Mackay S. 2001. The

155.

156.

157.

158.

159.

160.

161.

162.

163.

P1: GBC

237

challenge of objective assessment of surgical skill. Am. J. Surg. 181:484–86 Datta V, Mackay S, Mandalia M, Darzi A. 2001. The use of electromagnetic motion tracking analysis to objectively measure open surgical skill in the laboratory-based model. J. Am. Coll. Surg. 193:479–85 Datta V, Chang A, Mackay S, Darzi A. 2002. The relationship between motion analysis and surgical technical assessments. Am. J. Surg. 184:70–73 Downes M, Cavusoglu MC, Gantert W, et al. 1998. Virtual environments for training critical skills in laparoscopic surgery. Stud. Health Technol. Inform. 50:316– 22 Satava RM, Cuschieri A, Hamdorf J. 2003. Metrics for objective assessment. Surg. Endosc. 17:220–26 Fielding GA. 2002. Technical developments and a team approach leads to an improved outcome: lessons learnt implementing laparoscopic splenectomy. ANZ J. Surg. 72:459 Torkington J, Smith SG, Rees BI, Darzi A. 2000. The role of simulation in surgical training. Ann. R. Coll. Surg. Engl. 82:88–94 Wysocki WM, Moesta KT, Schlag PM. 2003. Surgery, surgical education and surgical diagnostic procedures in the digital era. Med. Sci. Monit. 9:RA69–RA75 Bell RL, Seymour NE. 2002. Laparoscopic treatment of fulminant ulcerative colitis. Surg. Endosc. 16:1778–82 Carolin K, Turfah FH. 2001. Multiple small bowel metastasis after laparoscopically assisted abdominoperineal resection. Surg. Endosc. 15:518

2 Dec 2003

21:57

AR

AR206-ME55-15.tex

AR206-ME55-15.sgm

LaTeX2e(2002/01/18) P1: GBC 10.1146/annurev.med.55.091902.103657

Annu. Rev. Med. 2004. 55:239–54 doi: 10.1146/annurev.med.55.091902.103657 First published online as a Review in Advance on Aug. 14, 2003

IMPLEMENTING A RESEARCH AGENDA FOR COMPLEMENTARY AND ALTERNATIVE MEDICINE∗

Annu. Rev. Med. 2004.55:239-254. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Jonathan D. Berman and Stephen E. Straus National Center for Complementary and Alternative Medicine, National Institutes of Health, Bethesda, Maryland 20892; email: [email protected]

Key Words complementary medicine, alternative medicine, research ■ Abstract Complementary and alternative medicine (CAM) consists of diverse clinical interventions that are practiced because of their popularity rather than the prior demonstration of safety and efficacy required for conventional agents. CAM therapies can be grouped into five categories: biologically based therapies, manipulative and body-based interventions, mind-body interventions, “energy” therapies, and alternative medical systems. The present evidence that individual CAM interventions are efficacious is largely anecdotal, but hundreds of small trials have yielded positive results. For a few modalities, existing data are either very encouraging or else sufficient to conclude that they are ineffective. CAM interventions are presumed to be safe, yet they may not be, particularly in the case of botanical agents with inherent toxicities, significant drug interactions, or potent adulterants. The public health questions regarding CAM can only be addressed through a research agenda that defines which interventions have favorable therapeutic indices. Implementation of this agenda involves adequate characterization and standardization of the product or practice, with rigorous investigation to demonstrate its safety, mechanism of action, and efficacy.

PRESENT STATUS OF COMPLEMENTARY AND ALTERNATIVE MEDICINES Complementary and alternative medicine (CAM) consists of clinical interventions that are practiced because of their popularity. In contrast, conventional treatments must be demonstrated to be safe and effective before registration authorities approve them for use. Alternative interventions are used in place of conventional treatments; complementary interventions are used with conventional treatments. The 20 most common CAM therapies used in the United States in the 1990s are listed by Kessler et al. (1). The many diverse CAM therapies are frequently ∗ The U.S. Government has the right to retain a nonexclusive, royalty-free license in and to any copyright covering this paper.

239

2 Dec 2003

21:57

240

AR

AR206-ME55-15.tex

BERMAN

¥

AR206-ME55-15.sgm

LaTeX2e(2002/01/18)

P1: GBC

STRAUS

grouped into five overlapping categories: biologically based therapies, manipulative and body-based interventions, mind-body interventions, “energy” therapies, and alternative medical systems.

Annu. Rev. Med. 2004.55:239-254. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Biologically Based Therapies Biologically based therapies entail administration of botanical or animal products, vitamins, minerals, and amino acids. These agents reflect some of mankind’s first attempts to improve the human condition. The personal effects of the “ice man” included medicinal herbs (2). By the Middle Ages, thousands of botanical products had been inventoried for their medicinal effects. Many of these, including digitalis and quinine, form the basis of contemporary medicines. In the United States, botanicals/herbs, vitamins, minerals, and amino acids have been formally classified and regulated as dietary supplements since the Dietary Supplement and Health Education Act (DSHEA) of 1994 (3). The difference between a dietary supplement and a drug depends on the use of the agent, not on the nature of the agent itself. If an herb, vitamin, mineral, or amino acid is used to resolve nutritional deficiency or to improve or sustain the structure or function of the body, the agent is considered a dietary supplement. If the agent is used to diagnose, prevent, treat, or cure a disease, the agent is considered a drug. The distinction between the use of a product for a “structure-function” indication and its use for a therapeutic aim is something of an exercise in semantics. For example, administering calcium for the declared purpose of counteracting diminished plasma calcium levels or to “improve bone health” would be considered as giving a dietary supplement. On the other hand, administering calcium to treat or prevent the disease osteoporosis would be considered as giving a drug. Similarly, menopause is viewed as a normal, age-related physiological process; thus, black cohosh would be a botanical dietary supplement when administered to improve bodily function in menopausal women. However, treating osteoporosis with this same botanical would be considered administering a drug. These semantic distinctions are important because under law, dietary supplements can be marketed without prior approval of their safety and efficacy by the US Food and Drug Administration (FDA), whereas prior approval is required for drugs. Thus, the passage of DSHEA in 1994 permitted countless botanicals, amino acids, vitamins, and minerals to be sold in grocery stores and health food stores without regulatory review. Although all dietary supplements are grouped together, botanicals and animal products differ from vitamins, minerals, and amino acids in being complex mixtures. Plant extracts, for example, consist of literally thousands of components. CAM practitioners use the whole extract for treatment, as opposed to identifying an active compound within the extract to be purified and used as a single agent. The use of whole extracts, rather than the active component alone, to treat

2 Dec 2003

21:57

AR

AR206-ME55-15.tex

AR206-ME55-15.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:239-254. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

COMPLEMENTARY AND ALTERNATIVE MEDICINE

P1: GBC

241

human diseases is based on the assumption that components in an extract combine additively or synergistically to produce the total therapeutic effect. Biologically based therapies have been used to treat diseases of all human organs. For a few of them, data are already sufficient to warrant large-scale trials. For example, multicenter trials are in progress for the following botanicals: Ginkgo biloba for prevention of dementia, St. John’s wort for depression, saw palmetto and Pygeum africanum for benign hypertrophy of the prostate, glucosamine and chondroitin sulfate from animal tissues for a musculoskeletal disorder (osteoarthritis), shark cartilage for lung cancer, and vitamin E and selenium to prevent prostate cancer. These examples illustrate that in general, the diseases for which biological products are directed tend to be chronic ones. The vast number of biologically based approaches being applied to disorders of various bodily systems renders it impossible to draw sweeping conclusions about their efficacy. A survey of dietary supplements used to prevent or treat infectious diseases illustrates the range of existing data regarding the clinical efficacy of CAM biologicals. For infectious diseases, biologically based treatments range from being effective or likely effective in specific situations, to being of undetermined efficacy, to being proven ineffective. Among the effective botanical modalities for infectious diseases, artemisinins are a striking success. For severe malaria, artemisinins kill parasites more rapidly (48 h and 72 h in two studies) than does standard therapy with quinine (60 h and 90 h) (4, 5). As an example of an approach that is likely to be effective in selected settings, cranberry prevented urinary tract infections in women in one prominent study. The number of infections was reduced from 18 per 50 control women to 8 per 50 treated women (6). However, the treatment was not an effective prophylaxis for children with neurogenic bladder, for whom 75% of both controls and treated subjects experienced infection (7). Similarly, vitamin A supplementation statistically reduced the duration of measles pneumonia in vitamin A–deficient children from 8.5 days to 6.3 days (8), but it did not reduce the duration of respiratory syncytial virus pneumonia in US children, who are far less likely to be vitamin-deficient (9). Studies of Echinacea purpurea conflicted regarding its ability to treat or prevent viral colds. One study found that prophylactic administration of 4 ml of fresh expressed juice of whole flowering echinacea plants, twice a day for eight weeks, did not reduce statistically the incidence of natural colds (74% in placebo, 65% in the echinacea group) (10). Another study found that 5 ml of pressed juice, twice a day for 10 days, was effective treatment in that the duration of cold symptoms (6 days) was statistically diminished compared to placebo (9 days) (11). Finally, careful study proved some biological interventions ineffective. Contrary to popular belief, vitamin C prophylaxis does not diminish the incidence or duration of the common cold (12, 13), nor does treatment with “megadoses” of vitamin C (daily administration of ∼300 times the recommended daily allowance) reduce the duration or severity of cold symptoms (14).

2 Dec 2003

21:57

242

AR

AR206-ME55-15.tex

BERMAN

¥

AR206-ME55-15.sgm

LaTeX2e(2002/01/18)

P1: GBC

STRAUS

Annu. Rev. Med. 2004.55:239-254. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Manipulative and Body-Based Therapies Manipulative and body-based approaches include chiropractic or osteopathic manipulation and many forms of massage. Spinal manipulation has been used since the time of the ancient Greeks (15), but it was incorporated into two specific healing disciplines in the American heartland of the late nineteenth century: chiropractic and osteopathic medicine. The essence of chiropractic is “spinal adjustment,” in which force is applied to the spine with therapeutic intent. “Dozens of adjusting techniques exist, and discussions about their relative merits make up much chiropractic academic discourse” (15). The force is generally delivered by hand. The medical indication is low-back pain in 60% of cases, and head, neck, and extremity complaints for most other cases. The rationale for spinal manipulation has been postulated to be alleviation of an entrapped facet, repositioning of a disk fragment, alleviation of stiffness induced by fibrotic tissue, or modification of neuronal activity. Chiropractic is literally a hands-on rather than a high-technology approach and is frequently performed along with other CAM approaches such as massage, acupuncture, administration of botanicals, and physical therapy, largely for acute or chronic musculoskeletal disorders. The efficacy of chiropractic remains uncertain. Most controlled prospective studies of chiropractic involve manipulation for acute or subacute low-back pain. When chiropractic manipulation was compared to receiving an educational booklet on low-back pain, the group receiving manipulation had significantly less severe symptoms at 4 weeks (p = 0.02) but not at 12 weeks (p = 0.06) with respect to “bothersome” complaints, and only an insignificant improvement (p = 0.32–0.45) on the more definitive Roland Disability Scale at both time points (16). Previous clinical opinion was more encouraging as to the short-term benefit of chiropractic manipulation, with high rates of patient satisfaction. Yet, in the aggregate, the repeated manipulations are at least as expensive as any other interventions for low-back pain except surgical ones. Osteopathy is also rooted in physical manipulation, but over the past half century, osteopathic practice has become increasingly mainstream and is now largely indistinguishable from allopathic (medical) practice.

Mind-Body Techniques The mind and the body engage in a constant interplay that is mediated physically through neural pathways and chemically through the endocrine system and other systems of circulating mediators. Bodily functions are influenced by emotions, and emotions contribute to symptoms. Given the universal prevalence of superstition, ritual, and religious practice throughout human experience, one can assume safely that psychic interventions such as meditation and prayer have essentially always been used in efforts to forestall or ameliorate physical and mental suffering. Over the past century, a range of psychotherapeutic and psychopharmacological tools were developed to soothe the mind and ameliorate somatic complaints.

2 Dec 2003

21:57

AR

AR206-ME55-15.tex

AR206-ME55-15.sgm

LaTeX2e(2002/01/18)

COMPLEMENTARY AND ALTERNATIVE MEDICINE

P1: GBC

243

Annu. Rev. Med. 2004.55:239-254. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Numerous techniques rooted in far more ancient healing methods coexisted with these newer techniques and have emerged as elements of CAM. The boundary between these CAM and more conventional mind-body techniques is not well-defined but is largely decided according to each technique’s acceptance by mainstream medical communities. Meditation, dance, music therapy, art therapy, and intercessory prayer are considered complementary or alternative approaches; most uses of hypnosis, cognitive therapies, and biofeedback, however, are well integrated with conventional medicine.

Energy Therapies A variety of approaches invoke energy systems as vehicles for healing. Some are understood through contemporary physics, such as magnets, whereas others engage putative energies that remain to be demonstrated objectively. For example, certain Asian traditions attribute health to the proper flow of vital energies through the body. Efforts to restore or strengthen this flow are goals of modalities such as qi gong (pronounced “chee gun”), a traditional Chinese form of health maintenance based on the concepts of qi, meaning the vital energy of the body, and gong, cultivation of the qi (17). Qi gong exercises involve meditation, rhythmical movements, and breathing. The exercises are intended to regulate the mind and to increase circulation of the blood (18), and also to increase respiratory capacity. Qi gong has been specifically used for hypertension and respiratory diseases, among other diseases (17). As concluded by one review of the extant data, “The reported studies do not necessarily measure up to the strict protocols required for randomized controlled trials” (17). Distant healing refers to “the intentional influence of one or more persons upon another living system without utilizing known physical means of intervention” (19). This can take the form of intercessory prayer, in which a person extends mental energies on behalf of another, even at great distances, and “Therapeutic Touch,” in which the healing energies are believed to extend from the healer’s hands as they pass close to a patient’s body (20). The lack of plausible biophysical mechanisms for many of these energy healing approaches makes them exceedingly controversial. Yet, one energy approach is gaining traction both in the American marketplace and in academic medical centers. Acupuncture was originally conceived to restore the flow of qi. Acupuncture was already in use in East Asia by the first century BC, with the intention to “harmonize” a patient’s state of being (21). Two important states of being identified by the ancient Chinese were termed yin and yang. Yin is associated with cold, darkness, passivity, receptivity, tranquility, and quiescence. Yang is associated with heat, stimulation, excess, assertiveness, and dynamism. Another recognized state of being is “dampness,” which has the yin qualities of coldness and softness, and the yang quality of excess. To harmonize imbalances in yin-yang and other states of being, fine needles were inserted into specific points on the body along “meridians” connected to vital centers.

2 Dec 2003

21:57

Annu. Rev. Med. 2004.55:239-254. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

244

AR

AR206-ME55-15.tex

BERMAN

¥

AR206-ME55-15.sgm

LaTeX2e(2002/01/18)

P1: GBC

STRAUS

There have been numerous studies of acupuncture’s clinical effects, and with evolving explanations of how they may be mediated, there is growing acceptance of their potential benefits. One recent review listed 235 randomized clinical trials of acupuncture for pain, as well as 200 randomized clinical trials for other conditions such as asthma and pulmonary disease, emesis, and various addictions (21). In 1997, a National Institutes of Health Consensus Development Panel concluded that needle acupuncture is probably efficacious for adult postoperative and chemotherapy-associated nausea and vomiting, and for postoperative dental pain (22). Shen et al. subsequently confirmed the effectiveness of acupuncture for chemotherapy-induced emesis in breast cancer patients. The group receiving electroacupuncture (in which mild electrical stimulation is delivered through the acupuncture needles) experienced a mean of 6.3 episodes of vomiting, compared to 10.7 episodes for a minimal needling negative-control group and 13.4 episodes for a positive-control group receiving conventional antiemetic drugs (23). Recent research has not established the basis for the traditional acupuncture points or meridians, but the stimulation by needling of small-diameter nerves within the musculature appears to have physiological effects. Such stimulation activates the spinal cord, brain stem, and hypothalamus, which in turn can trigger release of endogenous opioids. Functional magnetic resonance imaging and positron-emission tomography scanning have shown that needling affects specific areas of the brain. Some of the neuronal pathways that must be engaged in these activities would not have been predicted by conventional medicine. For example, part of the occipital lobe normally activated by a visual cue was also activated by acupuncture to the medial aspect of the big toe, the site traditionally needled by acupuncturists aiming to address certain visual disorders (24).

Alternative Medical Systems Alternative medical systems comprise interventions that do not rely on any elements of conventional medicine. Examples of alternative medical systems include traditional Chinese medicine, Ayurveda (meaning “science of life”) of India, and various Native American healing approaches. Each is a complex system that incorporates natural products, diet, spiritual elements, and other modalities. Newer alternative systems include homeopathic medicine and naturopathic medicine, both founded in Germany in the nineteenth century. Homeopathy began as a reaction to the unsafe interventions of that time, such as purging and leeching. Homeopathic medicine is based on the concepts of “similars” and “potencies.” These concepts signify that symptoms of an illness will respond to administration of infinitesimal amounts of an agent that in higher concentrations would produce similar symptoms in healthy subjects (25). The notion that small amounts of a substance can be healthful was reinforced by the success of vaccines and allergy desensitization. Frequently, however, homeopathic medications are so dilute as to invite serious skepticism. For example, Jacobs et al. treated infantile diarrhea in developing countries with a homeopathic agent “diluted 1:100 in a water/alcohol

2 Dec 2003

21:57

AR

AR206-ME55-15.tex

AR206-ME55-15.sgm

LaTeX2e(2002/01/18)

COMPLEMENTARY AND ALTERNATIVE MEDICINE

P1: GBC

245

solution 30 times for a final concentration of 1 × 10−60,” well beyond Avogadro’s number of molecules in a mole (26). Naturopathic medicine derives from nineteenth-century German practices that aimed to promote self-healing with such nontoxic measures as a proper diet. Today the practice involves lifestyle recommendations, conventional drugs in tolerated doses, herbs, acupuncture, and more.

Annu. Rev. Med. 2004.55:239-254. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Prevalence and Appeal of CAM Interventions Despite remarkable gains in lifespan and life quality over the past century, and the spotty evidence supporting CAM approaches, CAM interventions are widely practiced. In 1997, Americans made ∼600 million visits to CAM practitioners with ∼$25 billion out-of-pocket costs for alternative therapies. Among US adults in 1999, 33% of women and 24% of men used CAM, as did 31%–33% of 35– 54-year-olds. The most commonly used modalities were prayer (used by 14% of adults), herbal medicine (10%), and chiropractic therapies (8%) (27). For some chronic, debilitating diseases, use of CAM interventions is even more prevalent; as many as 84% of AIDS patients use CAM (28). These large expenditures of time and money signify that CAM has fundamental appeal for patients. At the least, CAM interventions offer hope as a defense against diseases for which there are few conventional medicines, and warm personal relationships as a defense against medical procedures that can be cold and isolating. (29).

Disadvantages of CAM Interventions The resources expended on CAM interventions are substantial and could be unwise. Reliance on an ineffective CAM agent may dissuade a patient from pursuing or adhering to more conventional approaches that are proven effective. Some who use complementary modalities may be merely “hedging their bets,” doing something extra, hoping it will make up for any inherent deficiencies in the conventional approaches. This strategy rests on the assumption that the complementary approach is natural and, even if proven ineffective, is at least safe—an assumption that may not be valid. As De Smet observed, “Since the drug receptors in our body cannot distinguish whether a molecule comes from the plant kingdom or from the chemical laboratory, naturalness does not, by definition, guarantee harmlessness” (30). Herbal medicines are plagued by contamination with heavy metals and filth, by adulteration with prescription drugs, wide divergence from labeled content, interference with the pharmacokinetics of life-saving drugs, and even some inherent toxicities. Between 1990 and 1999, the FDA recalled more than 100 dietary supplements because they caused eosinophilia-myalgia syndrome, digitalis intoxication, hypervitaminosis D, and Salmonella infection (31). There are several prominent recent examples of unexpected clinical problems related to herbal medicines. The product PC-SPES was used by thousands of men with advanced prostate cancer. It was removed from the market in mid-2002 when

2 Dec 2003

21:57

Annu. Rev. Med. 2004.55:239-254. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

246

AR

AR206-ME55-15.tex

BERMAN

¥

AR206-ME55-15.sgm

LaTeX2e(2002/01/18)

P1: GBC

STRAUS

found to be contaminated with the synthetic drugs warfarin and diethylstilbesterol (32, 33). “Chinese herb nephropathy” refers to nephropathy and/or cancer caused by weight-reducing pills supposed to contain the Chinese herb Sephania tetrandara for which the nephrotoxic herb, Aristolochia fangchi, was inadvertently substituted (34). Dietary supplements containing ephedra may not provide much benefit with respect to weight loss or athletic performance (35) but may result in cardiovascular and central nervous system toxicity (36). Kava (Piper methysticum), used for its central nervous system effects, may result in liver toxicity (reviewed in 37). Furthermore, botanicals can modify host metabolic enzymes to such an extent that the concentrations of conventional agents are materially altered. St John’s wort decreased the area-under-the-curve concentration of the anti-HIV drug indinavir by a mean of 57% in normal volunteers (38), and garlic decreased the area-underthe-curve concentration of another anti-HIV drug, saquinavir, by 51% (39).

RESEARCH AGENDA FOR COMPLEMENTARY AND ALTERNATIVE MEDICINES The above summary of the domains and present clinical status of CAM interventions indicates that although a vast number are in use, few have been shown to be safe or beneficial. The costs of CAM approaches and their potential risks and benefits provide the public-health rationale for subjecting them to critical appraisal. In pursuit of this vision, the US Congress authorized in 1998 the establishment of a new component of the National Institutes of Health—the National Center for Complementary and Alternative Medicine (NCCAM)—with a mandate to conduct CAM research, train CAM investigators, and disseminate authoritative information to practitioners and the public. At NCCAM, our own evolving experience with studies of CAM modalities is that too many are weakened by bad assumptions: that the product is well defined; that the optimal dose is known; that the right target population for study has been selected; that practitioners administer the product similarly; that any honest investigator could verify the received wisdom that the treatment is safe and effective. In challenging these assumptions, we are forced to reflect on lessons learned from the more orderly sequence by which conventional treatments are developed and to exploit those strategies that would best suit the particular characteristics of CAM agents. In the end, our goal is to give CAM modalities the best possible chance of proving beneficial. The fundamental requirements for conventional clinical agents are that they be reproducibly manufactured and administered, be safe, and be effective. In regulatory parlance, the term “manufacturing” refers to the totality of chemistrymanufacturing-control: synthesis of a drug, manufacturing of the product to be administered to the patient, and control of these processes so that the drug and product are made reproducibly. This is certainly relevant to dietary supplements, given their distressing variability. First, nonclinical (in vitro or animal) studies attempt to predict the efficacy and safety likely to be seen in clinical trials, and

2 Dec 2003

21:57

AR

AR206-ME55-15.tex

AR206-ME55-15.sgm

LaTeX2e(2002/01/18)

COMPLEMENTARY AND ALTERNATIVE MEDICINE

P1: GBC

247

Annu. Rev. Med. 2004.55:239-254. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

they attempt to determine the pharmacokinetic and biochemical mechanisms for efficacy and safety. Next, clinical trials provide the actual clinical data on safety, efficacy, and mechanisms. The tools and techniques required to conduct CAM research are essentially the same as those for conventional research. Yet, because CAM approaches are already in practice, some confidence in their safety pre-exists formal investigation, and some of the regulatory mandates designed for new chemical entities can be modified.

Chemistry-Manufacturing-Control Considerations for Botanical Dietary Supplements Unlike synthetic drugs, botanical drugs are complex mixtures. The active ingredient can be hypothesized but is rarely known for certain. For example, St. John’s wort extracts contain napthodianthrones such as hypericin and pseudohypericin; flavones such as hyperoside, quercitrin, and rutin; aglycones such as quercetin, kampferol, luteolin, and myricetin; and phloroglucinaols such as hyperforin, in addition to probably hundreds of unnamed components. The presumed value of St. John’s wort for affective disorders has been tied to a number of its constituents, but ultimately, its psychoactive compound(s) are unproven. For this and other botanicals, reproducible manufacturing requires consistent generation of unknown and therefore unanalyzable constituents. To do this, the process by which the final product is generated—botanical growth, harvest, and extraction; the manufacturing processes leading to the final clinical product—must be rigorously controlled. It has long been known that differences in plant species, growth conditions, time of harvest, and extraction conditions materially affect the composition of the final product. A good historical example is that of digitalis. The publication of a new Pharmacopoeia circa 1907 stimulated the following comments in JAMA: “One of the first essentials. . ., satisfactory to both patient and physician, is that the remedy used shall not only possess the qualities usually ascribed to it, but also that different preparations of the same drug shall possess these qualities to a uniform degree, or in other words, shall be of uniform strength. . .,” and further, “it is impossible to fix any chemical standard to which preparations of digitalis and its allies shall conform, on account of the difficulty connected with the isolation of the glucosides. . .the crude drug itself varies notoriously in its content of active principle. This variation depends on the time when the leaves are gathered, and the country from which the plant is obtained. . .. The US Pharmacopoeia directs that the leaves of the second year’s growth shall be gathered at the commencement of flowering. Experience has shown that these are best dried in a current of warm air and then should be preserved in tin containers to protect the drug from light, air, and moisture” (40). These admonitions are broadly relevant to the problem of standardizing botanical remedies today. A more modern, and ultimately highly successful, example is that of the new antimalarial agent artemisinin. Pure artemisinin is purified from A. annua leaves. A. annua grows on the banks of the Potomac river in Maryland, USA, but the yield

2 Dec 2003

21:57

Annu. Rev. Med. 2004.55:239-254. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

248

AR

AR206-ME55-15.tex

BERMAN

¥

AR206-ME55-15.sgm

LaTeX2e(2002/01/18)

P1: GBC

STRAUS

is ∼0.05%. Artemisinin yields from China range from 0.01% to 0.5%, with the highest from varieties grown in Sichuan Province (41). Because the yield varies with local varieties and growth conditions, all artemisinin is currently derived from high-yielding regions of Southeast Asia. As a result of its recalls of contaminated and adulterated dietary supplements, in 2003 the FDA proposed new chemistry-manufacturing-control (CMC) standards for marketed dietary supplements. These standards will make manufacturing controls more like those for conventional drugs. A key feature of the new standards is the introduction of process controls to supplement analysis of the final product. With respect to botanicals used as drugs in clinical trials, the FDA has issued draft requirements for phase 1/2 studies and phase 3 studies (42). NCCAM promulgated its adaptation of these requirements for all studies it supports, as summarized in Table 1, columns 4 and 5 (43). Proposals for phase 1/2 studies are to include a description of the plant and the extraction procedure; analysis of the clinical formulation for putative active ingredient(s), for a chemical fingerprint representing the unidentified ingredients, and for lack of contamination by pesticides, heavy metals, and synthetic drug TABLE 1 Chemistry-manufacturing-control considerations for clinical trials

Study subject

Study parameter

Study details

PLANT SUBSTANCE

Starting material

Botanical description Extraction procedure Quantity of active moeity Identity: chemical/ biologic assay Stability

PLANT PRODUCT

Manufacturing Finished product Product assay

Storage Stability Excipients Impurities Reference standard In process controls Bioavailability Microbiology Environmental

Reagents/process Quantity of active moeity Methods/specifications Identity: chemical/ biologic assay Purity Describe conditions Light/heat/time List List/analyze Standard batch SOPs Disintegration/ dissolution rate Contamination Assessment

Data to support Phase 1/2

Data to support Phase 3

X X

Expanded Expanded X X X

X X

X X X

X

X X X X X X X X X X X X X X

2 Dec 2003

21:57

AR

AR206-ME55-15.tex

AR206-ME55-15.sgm

LaTeX2e(2002/01/18)

COMPLEMENTARY AND ALTERNATIVE MEDICINE

P1: GBC

249

Annu. Rev. Med. 2004.55:239-254. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

adulterants; demonstration that the analyzed components dissolve in physiological solutions; and demonstration that the ingredients are stable over the length of the trial. Phase 3 studies involve a larger number of patients and are penultimate to registration, marketing, and the less-monitored use of the agent by the general public. Thus, for phase 3 studies, extra confidence in the information is needed. This is provided by formal process controls demonstrating that the plants were cultivated/harvested according to good agricultural/harvesting practices, and that extraction, analysis, and manufacturing were performed according to good manufacturing practices.

Chemistry-Manufacturing-Control Considerations for Nonbotanical Dietary Supplements Some nonbotanical dietary supplements, such as glucosamine or melatonin, differ from botanical drugs in being single chemical entities. CMC requirements for these substances can therefore approximate those of a conventional agent.

Nonclinical Considerations for All CAM Interventions Nonclinical information generally needed to support investigation of a conventional drug consists of in vitro and animal data on efficacy, toxicity, and pharmacokinetics/mechanisms. In the case of CAM agents, because of their substantial human use prior to formal investigation, clinical tolerance may be reasonably predicted and animal studies are not required to protect human subjects. Therefore, the FDA draft botanical guidelines and NCCAM policy do not require in vitro or animal experimentation prior to clinical experimentation. The ultimate criterion for clinical use is a favorable clinical therapeutic index, irrespective of a mechanistic understanding of the effects or their correlation with nonclinical data. Nevertheless, nonclinical experimentation has considerable value. Attractive nonclinical and mechanistic data frequently provide the confidence to pursue a clinical program and/or result in a better design of that program. Moreover, some understanding of mechanism helps lead to general acceptance of an approach.

Clinical Considerations for All CAM Interventions Demonstration of the clinical value of both biological agents and nonbiological approaches can follow similar developmental schemata (Table 2). Whether the interventions have arisen from laboratory investigations or have evolved over the centuries or in modern clinics through empirical trial and observation, clinical value must be demonstrated in the manner that is used for new conventional interventions. A first step in the formal prospective evaluation of CAM interventions is to determine how much to give and how often. Each intervention can be underdosed, and so have little effect, or overdosed so that it is toxic (biological agents), physically painful (manipulative and body-based interventions), or unnecessarily costly (all interventions). Once the apparent optimum dose is determined in

2 Dec 2003

21:57

250

AR

AR206-ME55-15.tex

BERMAN

¥

AR206-ME55-15.sgm

LaTeX2e(2002/01/18)

P1: GBC

STRAUS

TABLE 2 Clinical considerations for CAM clinical trials Study

Study parameter

Study details

PHASE I

Toxicity

Single dose Multiple dose Fed vs fasted drug-drug interactions

Annu. Rev. Med. 2004.55:239-254. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Pharmacokinetics (biologics) PHASE II

Efficacy

Multiple dose

PHASE III

Efficacy and Toxicity

Large comparative trials in adults Additional trial in children

Data to support Phase 1/2

Data to support Phase 3

X

X (if possible) (if possible) X

these phase 1/2 trials, large phase 3 trials can compare the optimum dose to a conventional intervention.

Design of Clinical Studies Many of the disorders for which CAM modalities are prescribed have variable natural histories and are highly subjective. These features make their study relatively difficult. Consequently, particular attention should be paid to the essential features of study design: control groups, entry criteria, endpoints, sample size, reproducible administration of the interventions, and compliance. PLACEBO CONTROL GROUPS Responses to CAM and conventional approaches can be influenced by patients’ expectation that they will benefit from a powerful intervention by a skilled practitioner. Thus, the best efficacy studies are placebocontrolled, unless this exposes patients to unacceptable suffering or morbidity. Studies of certain CAM modalities, particularly the physical interventions such as chiropractic manipulation, can be as difficult to blind as conventional surgical interventions. Nevertheless, the placebo effect can be significant for physical interventions, and randomized placebo controls are useful if they can be instituted. A randomized study of arthroscopic surgery suggested that surgical techniques elicit such strong placebo effects that the long-standing practice had no specific value (44). The need for placebo controls is particularly important in studies of CAM agents because of their widespread clinical use and social acceptance prior to formal evaluation. Countless patients and their physicians may be correct that a particular CAM therapy produced a major improvement in the patient’s condition, without the CAM therapy having any specific value, if there is a large placebo effect for that condition. Perhaps one example of this dilemma is St. John’s wort. In one large study of its value for major depression of moderate severity, a partial

2 Dec 2003

21:57

AR

AR206-ME55-15.tex

AR206-ME55-15.sgm

LaTeX2e(2002/01/18)

COMPLEMENTARY AND ALTERNATIVE MEDICINE

P1: GBC

251

Annu. Rev. Med. 2004.55:239-254. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

response (50% improvement in the standard Hamilton D index of depression) or full response was seen in 38% of patients but in 43% of placebo recipients (45). This study’s finding that St. John’s wort had no specific efficacy could not have been revealed without a placebo study arm. DISEASE SEVERITY CRITERIA Because the diseases treated with CAM agents are often chronic, have widely variable placebo response rates, and have small effect sizes in response to conventional treatment, the characteristics of the patients enrolled in a study can influence its likelihood of success. Davidson’s study of St. John’s wort for major depression showed no effect greater than placebo, whereas Lecrubier’s study (46) did show a statistically significant effect. The studies involved different formulations of St. John’s wort but the same concentration of the putative active ingredient hypericin, and both used the same efficacy criterion. Why did the outcomes differ? One hypothesis is that the patients were differentially seeking care and thus may have differed in severity of depression (46). On the other hand, increased severity of disease may modify drug response rates and placebo response rates equally, so that enrollment of sicker patients may not generally demonstrate the specific efficacy of CAM interventions (47). DISEASE ENDPOINTS It is important to apply objective measures of disease severity where possible, or at least to utilize well-validated instruments for subjective endpoints. For example, the best studies of glucosamine for osteoarthritis use not only the well-accepted WOMAC index of pain, physical function, and stiffness, but also coded flexion X-rays to measure the joint space (48).

Many CAM interventions are used in conditions for which conventional drugs are only modestly effective, yet these small effect sizes are of clinical significance. In the above-mentioned St. John’s wort study, only 25% of patients treated with the standard serotonin-specific reuptake inhibitor sertraline demonstrated a full response. Small effect sizes mean that large numbers of patients must be enrolled to statistically differentiate standard therapy from placebo. Controlled trials of agents with small effect sizes are therefore very expensive. The problem of expending large resources on a study that may fail is an unfortunate challenge of CAM research and reflects the fact that these approaches enter formal clinical experimentation on the basis of anecdotes.

SAMPLE SIZE

REPRODUCIBLE ADMINISTRATION OF INTERVENTION Botanical drugs are difficult to manufacture reproducibly but (like conventional drugs) are simple to administer. In contrast, other CAM modalities, such as manipulative and body-based interventions, mind-body interventions, and energy therapies, have no manufacturing issues but (like conventional interventions such as surgery and psychotherapy) can be difficult to administer reproducibly. In each of these settings, the best study designs

2 Dec 2003

21:57

252

AR

AR206-ME55-15.tex

BERMAN

¥

AR206-ME55-15.sgm

LaTeX2e(2002/01/18)

P1: GBC

STRAUS

incorporate a “manual of administration” so that the interventions are performed in a standardized manner to which each practitioner agrees. Strong preferences of patients towards CAM or conventional interventions can threaten study blinding and randomization. Because dietary supplements are widely marketed, patients enrolled in a blinded placebo-controlled study who are not improving as much as they had anticipated might “cheat” by purchasing and taking the study supplement (49). In a study of an intensive nutritional intervention for pancreatic cancer, it was difficult to enroll subjects. Subjects who strongly desired conventional chemotherapy refused to risk being randomized to the nutritional approach, whereas patients who feared the toxicity of chemotherapy and its modest benefits in pancreatic cancer only wanted to receive the nutritional approach. In the end, the study had to be redesigned to allow self-selection, with the hope that the disease severity of subjects who do enroll into each study arm will, in the end, be comparable.

Annu. Rev. Med. 2004.55:239-254. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

COMPLIANCE

CONCLUSIONS Complementary and alternative medicines are interventions that have entered clinical use without the prior demonstration of safety and efficacy required for conventional interventions. Given the high costs of developing new drugs and the popular appeal of CAM interventions, investigation of their therapeutic index is an attractive means of augmenting our clinical armamentarium. The agenda for CAM research focuses on providing the confidence in manufacturing, administration, safety, and efficacy afforded by conventional agents. The Annual Review of Medicine is online at http://med.annualreviews.org

LITERATURE CITED 1. Kessler RC, Davis RB, Foster DF, et al. 2001. Long-term trends in the use of complementary and alternative medical therapies in the United States. Ann. Intern. Med. 135:262–68 2. Goldman P. 2001. Herbal medicines today and the roots of modern pharmacology. Ann. Intern. Med. 135:594–600 3. US Food and Drug Administration. Dietary supplements: overview. http://www. CFSAN.FDA.gov/∼dms/supplmnt.html 4. Hien TT, Day NPJ, Phu NH, et al. 1996. A controlled trial of artemether or quinine in Vietnamese adults with severe fal-

ciparum malaria. N. Engl. J. Med. 335:76– 83 5. Van Hensbroek MB, Onyioray E, Jarrar S, et al. 1996. A trial of artemether or quinine in children with cerebral malaria. N. Engl. J. Med. 335:69–75 6. Kontiokari T, Sundqvist K, Nuutinen M, et al. 2001. Randomized trial of cranberrylingonberry juice and lactobacillus GG drink for the prevention of urinary tract infections in women. BMJ 322:1–5 7. Schlager TA, Anderson S, Trudell J, et al. 1999. Effect of cranberry juice on bacteriuria in children with neurogenic bladder

2 Dec 2003

21:57

AR

AR206-ME55-15.tex

AR206-ME55-15.sgm

LaTeX2e(2002/01/18)

COMPLEMENTARY AND ALTERNATIVE MEDICINE

8.

Annu. Rev. Med. 2004.55:239-254. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

receiving intermittent catheterization. J. Pediatr. 135:698–702 Hussey GD, Klein M. 1990. A randomized, controlled trial of vitamin A in children with severe measles. N. Engl. J. Med. 323:160–64 Bresee JS, Fischer M, Dowell SF. 1996. Vitamin A therapy for children with respiratory syncytial virus infection: a multicenter trial in the United States. Pediatr. Infect. Dis. J. 15:777–82 Grimm W, Muller H-H. 1999. A randomized controlled trial of the effect of fluid extract of Echinacea purpurea on the incidence and severity of colds and respiratory infections. Am. J. Med. 106:138–43 Schulten B, Bulitta M, Ballering-Bruhl B, et al. 2001. Efficacy of Echinacea purpurea in patients with a common cold. Arzneimit. Forsch. 51:563–68 Coulehan JL, Everhard S, Kapner L, et al. 1976. Vitamin C and acute illness in Navajo schoolchildren. N. Engl. J. Med. 295:973– 77 Pitt HA, Costrini AM. 1979. Vitamin C prophylaxis in marine recruits. JAMA 241:908–11 Audera C, Patulny RV, Sander BH, et al. 2001. Mega-dose vitamin C in treatment of the common cold: a randomized controlled trial. Med. J. Aust. 175:359–62 Meeker WC, Haldeman S. 2002. Chiropractic: a profession at the crossroads of mainstream and alternative medicine. Ann. Intern. Med. 136:216–27 Cherkin D, Deyo RA, Battie M, et al. 1998. A comparison of physical therapy, chiropractic manipulation, and provision of an educational booklet for the treatment of patients with low back pain. N. Engl. J. Med. 339:1021–29 Sancier KM. 1999. Therapeutic benefits of qigong exercises in combination with drugs. J. Altern. Complement. Med. 5:383– 89 Ng B-Y. 1999. Qigong-induced mental disorders: a review. Aust. NZ J. Psychiatr. 33:197–206

P1: GBC

253

19. Astin JA, Harkness E, Ernst E. 2000. The efficacy of “distant healing”: a systematic review of randomized trials. Ann. Intern. Med. 132:903–10 20. Rosa L, Rosa E, Sarner L, et al. 1998. A close look at Therapeutic Touch. JAMA 279:1005–10 21. Kaptchuk TJ. 2002. Acupuncture: theory, efficacy, and practice. Ann. Intern. Med. 136:374–83 22. NIH Consensus Development Panel on Acupuncture. 1998. Acupuncture. JAMA 280:1518–24 23. Shen J, Wenger N, Glaspy J, et al. 2000. Electroacupuncture for myeloablative chemotherapy-induced vomiting: a randomized controlled trial. JAMA 284:2755–61 24. Cho ZH, Chung SC, Jones JP, et al. 1998. New findings of the correlation between acupoints and corresponding brain cortices using functional MRI. Proc. Natl. Acad. Sci.USA 95:2670–73 25. Jonas WB, Kaptchuk TJ, Linde K. 2003. A critical overview of homeopathy. Ann. Intern.Med. 138:393–99 26. Jacobs J, Jimenez M, Malthouse S, et al. 2000. Homeopathic treatment of acute childhood diarrhea: results from a clinical trial in Nepal. J. Altern. Complement. Med. 6:131–39 27. Ni H, Simile C, Hardy AM. 2002. Utilization of complementary and alternative medicine by United States adults. Results from the 1999 National Health Interview Survey. Med. Care 40:353–58 28. Sparber A, Wootton JC, Bauer L, et al. 2000. Use of complementary medicine by adult patients participating in HIV/AIDS clinical trials. J. Altern.Complement. Med. 6:415–22 29. Richardson MA, Straus S. 2002. Complementary and alternative medicine: opportunities and challenges for cancer management and research. Semin. Oncol. 29:531– 45 30. De Smet PAGM. 1995. Health risks of herbal remedies. Drug Saf. 13:81–93

2 Dec 2003

21:57

Annu. Rev. Med. 2004.55:239-254. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

254

AR

AR206-ME55-15.tex

BERMAN

¥

AR206-ME55-15.sgm

LaTeX2e(2002/01/18)

P1: GBC

STRAUS

31. US Food and Drug Administration. 2003. Current good manufacturing practice in manufacturing, packing, or holding dietary ingredients and dietary supplements; proposed rule. Fed. Regist. 68:12157–263 32. Straus SE. 2002. Herbal medicines— What’s in the bottle? N. Engl. J. Med. 347: 1997–98 33. Ko R. 1998. Adulterants in Asian patent medicines. N. Engl. J. Med. 339:847 34. Nortier JL, Martinez M-C M, Schmeiser HH, et al. 2000. Urothelial carcinoma associated with the use of a Chinese herb (Aristolochia fangchi). N. Engl. J. Med. 342:1686–92 35. Shekelle PG, Hardy ML, Morton SC, et al. 2003. Efficacy and safety of ephedra and ephedrine for weight loss and athletic performance: a meta-analysis. JAMA 289: 1537–45 36. Haller CA, Benowitz NL. 2000. Adverse cardiovascular and central nervous system events associated with dietary supplements containing ephedra alkaloids. N. Engl. J. Med. 343:1833–38 37. Ernst E. 2002. The risk-benefit profile of commonly used herbal therapies: ginkgo, St. John’s wort, ginseng, echinacea, saw palmetto, and kava. Ann. Intern. Med. 136: 42–53 38. Piscitelli SC, Burstein AH, Chaitt D, et al. 2000. Indinavir concentrations and St. John’s wort. Lancet 355:547–48 39. Piscitelli SC, Burnstein AH, Welden N, et al. 2002. The effect of garlic supplements on the pharmacokinetics of saquinavir. Clin. Infect. Dis. 34:234–38

40. Edmunds CW. 1907. The standardization of cardiac remedies. JAMA XLVIII:1744– 47 41. Klayman DL. 1985. Qinghaosu (artemisinin): an antimalarial drug from China. Science 228:1049–55 42. Food and Drug Administration Center for Drug Evaluation and Research. 2000. Botanical drug products draft guidance. 43. 2003. Considerations for NCCAM clinical trial applications. http://nccam.nih.gov/res earch/policies/clinical-considerations.htm. 44. Moseley JB, O’Malley K, Petersen NJ, et al. 2002. A controlled trial of arthroscopic surgery for osteoarthritis of the knee. N. Engl. J. Med. 347:81–88 45. Hypericum Depression Trial Study Group. 2002. Effect of Hypericum perforatum (St. John’s wort) in major depressive disorder. A randomized controlled trial. JAMA 287:1807–14 46. Lecrubier Y, Clerc G, Didi R, et al. 2002. Efficacy of St. John’s wort extract WS 5570 in major depression: a double-blind placebo controlled trial. Am. J. Psychiatr. 159:1361–66 47. Montgomery SA. 1999. The failure of placebo-controlled studies. Eur. Neuropsychopharmacol. 9:271–76 48. Reginster JY, Deroisy R, Rovati LC, et al. 2001. Long-term effects of glucosamine sulphate on osteoarthritis progression: a randomized placebo-controlled clinical trial. Lancet 357:251–56 49. Nahin RL, Straus SE. 2001. Research into complementary and alternative medicine: problems and potential. BMJ 322:161–64

12 Dec 2003

14:2

AR

AR206-ME55-16.tex

AR206-ME55-16.sgm

LaTeX2e(2002/01/18) P1: GBC 10.1146/annurev.med.55.091902.104338

Annu. Rev. Med. 2004. 55:255–82 doi: 10.1146/annurev.med.55.091902.104338 c 2004 by Annual Reviews. All rights reserved Copyright ° First published online as a Review in Advance on Sept. 8, 2003

BASIC ADVANCES AND NEW AVENUES IN THERAPY OF SPINAL CORD INJURY

Annu. Rev. Med. 2004.55:255-282. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Bruce H. Dobkin and Leif A. Havton Neurologic Rehabilitation and Neural Repair Research Programs, Department of Neurology, Geffen School of Medicine, University of California Los Angeles, 710 Westwood Plaza, Los Angeles, California 90095-1769; email: [email protected]

Key Words myelopathy, neuroprotection, rehabilitation, neural regeneration, activity-dependent plasticity, functional neuroimaging ■ Abstract The prospects for successful clinical trials of neuroprotective and neurorestorative interventions for patients with acute and chronic myelopathies depend on preclinical animal models of injury and repair that reflect the human condition. Remarkable progress continues in the attempt to promote connections between the brain and the sensory and motor neurons below a spinal cord lesion. Recent experiments demonstrate the potential for biological therapies to regenerate or remyelinate axons and to incorporate new neural cells into the milieu of a traumatic spinal cord injury. The computational flexibility and plasticity of the sensorimotor systems of the brain, spinal cord, and motor unit make functional use of new circuitry feasible in patients. To incorporate residual and new pathways, neural repair strategies must be coupled to rehabilitation therapies that drive activity-dependent plasticity for walking, for reaching and grasping, and for bowel and bladder control. Prevention of pain and dysautonomia are also clinical targets. Research aims to define the temporal windows of opportunity for interventions, test the safety and efficacy of delivery systems of agents and cells, and provide a better understanding of the cascades of gene expression and cell interactions both acutely and chronically after injury. These bench-to-bedside studies are defining the neurobiology of spinal cord injury rehabilitation.

BACKGROUND Investigators have learned valuable lessons from developmental and experimental neurobiology about the cues that enable axons to grow and stem cells to differentiate and migrate. As a result, the potential to reduce the neurological impairments and functional disabilities of people with traumatic spinal cord injury (SCI) has grown. Greater knowledge about how to manipulate activity-dependent plasticity to retrain sensorimotor skills complements strategies for neural repair (1). At the same time, patients and their families have become forceful advocates for a SCI cure. The possibilities for partial neurorestoration must be balanced against risky, premature clinical trials that employ poorly understood interventions drawn from 0066-4219/04/0218-0255$14.00

255

12 Dec 2003

14:2

Annu. Rev. Med. 2004.55:255-282. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

256

AR

AR206-ME55-16.tex

DOBKIN

¥

AR206-ME55-16.sgm

LaTeX2e(2002/01/18)

P1: GBC

HAVTON

limited pilot data in rodents. We review the promising biological, engineering, and rehabilitative strategies that may lessen disability and enhance quality of life for patients with acute and chronic myelopathies, primarily due to traumatic SCI. Traumatic SCI causes disability in 8000–10,000 Americans yearly. Causes include penetrating bullet wounds and other forms of violence (26%) and nonpenetrating lesions from vehicular accidents (38%), and sports accidents (7%), as well as falls (22%), especially in elderly persons (2). Males 15 to 35 years old are most often injured, but another peak occurs in people over age 60 who have an underlying cervical spinal stenosis from degenerative disk disease. Approximately 50% of victims present with complete sensorimotor impairment below the lesion. Fewer than 5% of these patients will regain the ability to walk. When some movement is present, motor gains may continue for many months. Patients who present with even modest voluntary movement below the lesion by the time they begin inpatient rehabilitation usually recover some useful movement in the arms or legs and up to 60% will walk, though often with assistive devices. Table 1 shows the incidence and yearly costs of care for patients with tetraplegia and paraplegia. The American Spinal Cord Injury Association (ASIA) Impairment Scale defines the degree of neurological loss as A (complete sensorimotor loss below the lesion including absent sacral sensation), B (sensory but no motor function below the lesion level), C (some motor preservation, but the majority of muscles are less than 3/5), and D (muscle grade is 3 or greater in the majority of groups below the lesion). By the time of admission for inpatient rehabilitation, ∼50% of cervical lesions and 75% of thoracic lesions are graded A. About 8% of lesions at all levels are ASIA B. About 32% of cervical lesions are ASIA D and 65% of L-1–L-5 lesions are ASIA D. Causes of nontraumatic SCI include cervical myelopathy associated with cervical spondylosis or a spinal stenosis; epidural metastatic and primary cord tumors; cord ischemia from atherothrombosis, emboli, and venous anomalies; infection; and viral or immune-mediated transverse myelitis. Multiple sclerosis may present with spastic paraparesis. Thoracoabdominal aneurysm repair causes paraparesis or TABLE 1 Severity and costs beyond the first year after traumatic spinal cord injury Severity of injury Tetraplegia Complete C-1–C-4 C-5–C-8 Incomplete Paraplegia Complete Incomplete

Incidence

Yearly health expenses (in 1998 dollars)

19%

30%

$95,000 $39,000 $12,000

28% 21%

$20,000 $11,000

Source: National Spinal Cord Injury Statistical Center (http://www.spinalcord.uab.edu).

12 Dec 2003

14:2

AR

AR206-ME55-16.tex

AR206-ME55-16.sgm

LaTeX2e(2002/01/18)

SPINAL CORD INJURY

P1: GBC

257

paralysis in up to 10% of surgeries, often after a postoperative delay (3). This review emphasizes traumatic SCI, but most of the interventions discussed for rehabilitation and neural repair can be applied to any cause of an acute or chronic myelopathy.

EXPERIMENTAL ACUTE SPINAL CORD INJURY

Annu. Rev. Med. 2004.55:255-282. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Pathogenesis Traumatic SCI may include torsional and tearing mechanical forces on the cord, axonal tears and demyelination, ionic fluxes, ischemia, loss of vascular autoregulation, neurotoxicity from glutamate and products of cell membrane breakdown such as free radicals, central gray neuronal death, delayed cell death by apoptosis, and an early and a delayed inflammatory response. Debris cleared by macrophages leaves a cystic matrix with trabeculae of new vascular structures, basal lamina, fibroblasts, sprouting axons tacked to the wall, and a rim of preserved white matter with some spared axons surrounding the central damage. Thus, gray matter of the dorsal and ventral horns is destroyed over one or more levels of the cord, ascending and descending white matter tracts are partially to completely disconnected, and the dorsal and ventral roots at the level of the lesion may be avulsed or torn. Models of traumatic SCI in rodents (4) and in transgenic mice (5) reproduce some aspects of human SCI, but because of neurobiological differences among strains and between rodents and humans, such models are most valuable for studies of particular biological processes. Each species and injury model has unique characteristics. Models include complete transection of the cord by surgical incision, incomplete transection that may include 1–3 quadrants of the cord, circumferential compression using a surgical clip, and an impact contusion using a weight or electromechanical device dropped over the cord after removing the dorsal vertebral bone. The impact contusion is reproducible in a graded fashion to create a focal injury similar to human lesions. Of course, a single blow at one angle does not reenact all of the torsional forces and prolonged compression that occur with nonpenetrating traumatic SCI from accidents.

Therapies Experimental studies of acute SCI target partially characterized components of the pathophysiological cascades that follow injury within hours to days. The cellular and molecular mechanisms of injury resemble those associated with acute stroke and traumatic brain injury, including ischemia, hypoxia, mechanical injury, changes in gene expression, inflammation, and necrotic and apoptotic cell death. Microarray technology is revealing the upregulation and downregulation of gene expression associated with injury over the minutes, hours, and days after SCI (6), which may better focus future neuroprotective and repair therapies (7). Recent murine models of SCI permit genetic manipulations to better determine the effects of a single molecular component of injury.

12 Dec 2003

14:2

Annu. Rev. Med. 2004.55:255-282. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

258

AR

AR206-ME55-16.tex

DOBKIN

¥

AR206-ME55-16.sgm

LaTeX2e(2002/01/18)

P1: GBC

HAVTON

Results from experimental models of neuroprotection in rodents have suggested interventions for human clinical trials. Highly inbred rodent species, usually of one sex, are typically injured under controlled experimental conditions and studied with outcome measures that do not fully reflect the behavioral and pathophysiological complexity of humans who suffer a SCI (2). Indeed, success in an animal model has led to only one proven, though controversial, human intervention, namely methylprednisolone. Potential targets for clinical trials of neuroprotection within hours to several days after acute SCI are similar to those derived from models of stroke and traumatic brain injury. Interventions that have been or are being studied include the following: ion channel and alpha-amino-3-hydroxy5-methyl-4-isoxazoleproprionic (AMPA) receptor antagonists to protect axons; glutamate receptor antagonists or other forms of inhibition of the N-methyl-Daspartate (NMDA) receptor to protect neurons; antioxidants and cell membrane stabilizers (8); caspase inhibitors to prevent apoptosis; neurotrophins, such as brain-derived neurotrophic factor (BDNF), or nitric oxide synthetase inhibition to prevent cell death; inhibition of cytokines and proteinases that contribute to inflammation and disruption of the blood-brain barrier, such as matrix metalloproteinase9 and minocycline (8a); and modulation of macrophage activity to limit demyelination and secondary injury from inflammatory responses (9). Interventions for neuroprotection overlap with potential biological and pharmacological approaches that inhibit or accelerate mechanisms of spontaneous neural repair. For example, several families of metalloproteases activate or inhibit axon elongation and guidance signaling under differing conditions (10). Activated microglia and macrophages at the wound site appear to augment secondary tissue damage within the first several days after injury, then provide both neurotrophins and other substrates to increase peri-wound axonal sprouting toward the microglia along the edge of the injury. Thus, the timing of an anti-inflammatory or proinflammatory cell or immunemediated approach may be important for strategies of protection and repair.

ACUTE SPINAL CORD INJURY IN HUMANS Pathogenesis Human traumatic SCI often produces a central hematoma, which is visible by magnetic resonance imaging in the most severe cases. Sparing of ascending and descending axons at the periphery of the central lesion may account for later sensorimotor gains. An autopsy study revealed that 28% of 130 patients who survived a complete SCI had some intact axons (11). Patients had been able to move against gravity, but not resistance, if 4%–10% of the axons at the fourth thoracic level were intact. Voluntary foot movement was associated with the presence of only 3000 of the usual 41,000 fibers in the corticospinal tract. Intact touch and vibration sensation required approximately 117,000 of the 452,000 sensory fibers. Sparing of >25% of the lateral or ventral white matter in studies of nonhuman primates permitted walking. Thus, rather modest incremental sparing of sensorimotor

12 Dec 2003

14:2

AR

AR206-ME55-16.tex

AR206-ME55-16.sgm

LaTeX2e(2002/01/18)

SPINAL CORD INJURY

P1: GBC

259

pathways from an acute therapeutic intervention, or modest regeneration elicited by subsequent biological intervention, may dramatically improve motor control. Many patients who present with ASIA A or B impairments improve 1–2 levels in motor function and 2–3 levels in sensory function below the lesion. Some of these gains may be related to physiological and structural improvements in the dorsal and ventral roots at the level of SCI and locally in gray and white matter.

Annu. Rev. Med. 2004.55:255-282. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Therapies Traumatic SCI is managed with surgery within one week of onset in 40%–70% of patients. Importantly, however, no clinical trials have been carried out to demonstrate the efficacy of spinal decompression, realignment, or stabilization for lessening sensorimotor or bowel and bladder impairments (12). Various procedures probably help prevent late complications such as cord tethering, syringomyelia, recurrent cord compression, and perhaps some forms of neuropathic pain. Surgical approaches for cervical myelopathy due to spondylosis also lack support from formal trials but may prevent progression of spastic paraparesis (13). Clinical trials of interventions for ameliorating the pathologic and behavioral effects of acute SCI suffer from problems in translating a single intervention in a rodent model to the complex, multifactorial disorder that exists outside the laboratory. Differences between humans and rodents in size, therapeutic window, dose-response relationships, drug penetration, and gene expression must be taken into account; differences in age and sex may also be important. In addition, interventions may aim at behavioral compensation or at innate restitutive or substitutive capacities. (Restitutive capacity allows the same neural pathways to be used after injury; substitutive capacity entails adapting a defective or partially spared network, usually via external stimulation such as rehabilitation of motor skills.) Most important, the designs of randomized human trials for neuroprotection have been less than optimal for showing clinically meaningful improvements, as opposed to statistically significant differences, due to an intervention. Acute randomized clinical trials for neuroprotection have been carried out using methylprednisolone, tirilizad, naloxone, and GM-1 ganglioside (Table 2). The results of these trials have been rather disappointing and controversial (14). The culmination of three large trials from the National Acute Spinal Cord Injury Study (NASCIS I, II, and III) led the investigators to recommend that patients with acute SCI receive methylprednisolone (30 mg/kg bolus, then 5.4 mg/kg/h infusion) for 23 h if started within 3 h of injury. If the drug is started 3–8 h after injury, patients should stay on the regimen for 48 h. This recommendation is based on modest gains in motor scores, but not functional gains. The outcome measures of the NASCIS studies aimed rather low in looking for a benefit; the GM-1 studies may have statistically powered their trial based on an effect size that was too high to reveal efficacy. An Israeli research group has taken rodent studies into a Phase 1 human trial of activated macrophages (15), which is leading to a Phase 2 trial that includes American sites. A subject’s peripheral blood monocytes are activated with

12 Dec 2003

14:2

260

AR

AR206-ME55-16.tex

DOBKIN

¥

AR206-ME55-16.sgm

LaTeX2e(2002/01/18)

P1: GBC

HAVTON

Annu. Rev. Med. 2004.55:255-282. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

TABLE 2 Clinical trials to enhance outcomes after spinal cord injury (SCI) Intervention RCTs∗

Outcome

Confounds

Methylprednisolone (95)

Motor gains small; side effects greater with 48-h infusion

Design and statistical issues; limited impact of preventing lipid peroxidation

GM-1 ganglioside (96)

No better than steroid

Design, control, and statistical issues

4-aminopyridine (fampridine) (92)

Improves axonal conduction to modestly increase strength or lessen fatigability with repeated movements

Plugs potassium channels in partially demyelinated fibers; in Phase 3 trial with 150 subjects for spasticity, bladder, and motor outcomes (Acorda Therapeutics, Hawthorne, NY)

BWSTT∗ for walking

Nonrandomized trials show value for poor walkers

High manpower and experience needed by therapists

Robotic- and FES∗ -assisted BWSTT

Robot-assistive stepper may be more practical than manual training for patients with no motor function; design ought to include proprioceptive/torque feedback

SAFETY TRIALS FOR CHRONIC SCI (http://carecure.atinfopop.com; no peer-reviewed reports available) Fetal tissue in cord syrinx

Cell plug persists without adverse reaction; no clear functional gains

Not sociopolitically viable in United States

Human fetal stem cells injected into lesion

Oral reports from Beijing and Moscow suggest no efficacy

No formal report of Phase 1/2 trials

Fetal OEGs∗ injected into cord

Report from Beijing in first 150 subjects and from Lisbon pending

Clinical assessments and outcome measures unclear

Fetal pig embryo cells cultured to become oligodendrocytes

Phase 1 trial not yet reported (Diacrin Inc., Charlestown, MA)

No cells found in one autopsy

Autologous monocytes activated against skin and injected into cord below lesion

Some sensorimotor gains in Phase 1 trial by oral report

Phase 2 trial of immune strategy pending (Proneuron Biotechnologies, Rehovat, Israel)

Peripheral nerve bridge with neurotrophins

Oral reports from Taiwan (H Cheng, http://www.nature.com/nsu/nsu pf/ 030512-12.html)

Technical difficulty and could reinjure cord

∗ RCT, randomized clinical trial; BWSTT, body weight–supported treadmill training; FES, functional electrical stimulation; OEG, olfactory ensheathing glial cells.

12 Dec 2003

14:2

AR

AR206-ME55-16.tex

AR206-ME55-16.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:255-282. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

SPINAL CORD INJURY

P1: GBC

261

epidermal cells and placed into the injury site about two weeks after a traumatic SCI, where they presumably counteract the negative impact of inflammatory responses and tie up myelin products that may inhibit axonal regeneration. The mechanism of efficacy is still unclear. Risks include an exacerbation of injury by chronic macrophage activation and autoimmune responses (16). Medical complications (17) in the first weeks after a SCI may decrease the benefits of rehabilitation and of future biological interventions for neural repair. Unsuspected fractures or local trauma and inflammation outside the spinal column may produce heterotopic ossification and limit joint range of motion and weight bearing on a limb (2). Low-level, focal irradiation may reduce this risk, along with disodium etidronate. Cerebral trauma may accompany up to 50% of traumatic SCIs, which may alter behavior, cognitive abilities, and ability to cooperate with rehabilitation if not managed. Even minor pain, or pain signals not appreciated from below the lesion, such as from a rectal fissure, bladder infection, thrombophlebitis, or traumatized joint, may induce spasms and dysautonomia with associated hypertension, sweating, and the potential for a cardiac arrythmia. With increasing time after the injury, the initial flaccid paralysis may be superseded by spacticity, manifested as abnormal postures, spasms, hyperreflexia, and, in some instances, contractures. Weakness, impaired motor control for skilled movements, and fatigability are even more disabling residua of an upper motoneuron lesion. Dystonic postures that interfere with hygiene or self-image and spasms that cause pain or make wheelchair activities dangerous require first-order oral antispasticity agents such as baclofen or tizanidine, local injection of botulinum toxin, or intrathecal injection of baclofen. Motor control for walking, reaching, and grasping is rarely improved by these drugs (2).

NEUROLOGIC REHABILITATION AND NEUROPLASTICITY Much of rehabilitation for highly impaired patients involves behavioral compensation. Patients learn to use whatever sensorimotor function they possess, along with assistive devices, to become more independent in their daily mobility, self care, and community roles (2). Task-oriented practice, development of problem-solving skills, management of skin, bowel, and bladder care, and use of a wheelchair for those who cannot walk are priorities during inpatient therapy. Additional strategies take advantage of the growing understanding of the neurobiology of rehabilitation. Successful rehabilitative techniques must induce activity in residual spinal pathways that interact with cortical, subcortical, and brainstem nodes for motor control to drive the plasticity required at all CNS levels for relearning motor skills (2). The spinal cord itself serves far more physiological functions than its appearance as a simple conduit for ascending and descending axons may suggest. The dorsal and ventral neurons of the gray matter integrate and project sensory information related to proprioception and cutaneous inputs related to posture, skilled actions,

12 Dec 2003

14:2

262

AR

AR206-ME55-16.tex

DOBKIN

¥

AR206-ME55-16.sgm

LaTeX2e(2002/01/18)

P1: GBC

HAVTON

and automatic flexor and extensor movements such as stepping and reaching. This information drives learning and representational plasticity for movements along the neuroaxis. Interactions among spinal cord columns of rostrocaudal motor pools and oscillating circuits increase the flexibility of supraspinal projections for motor control.

Annu. Rev. Med. 2004.55:255-282. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Locomotor Training Walking over flat ground depends, at minimum, on segmental sensory feedback to spinal and supraspinal networks and on output from cortical, brainstem, and lumbosacral locomotor regions, all of which interact with biomechanical and postural mechanisms for balance and motor control. Segmental proprioceptive and cutaneous sensory inputs associated with walking had powerful modulatory effects on spinal motor pools after complete spinal cord transection at a low thoracic level in the cat and rat (18, 19). These animals could be trained to walk on a moving treadmill when given truncal support. During training on the treadmill, they incorporated signals for stance that emphasized limb loading and for swing that emphasized hip extension at the end of the stance phase (20). Spinal locomotion requires oscillating flexor and extensor motor pools in the lumbar cord called central pattern generators. Studies in spinalized rodents and cats reveal that training to step reduces spinal levels of inhibitory neurotransmitter-associated substances such as glutamic acid decarboxylase (GAD-67) and glycine, whereas injury-induced paralysis raises these levels (21). The exercise also increases neurotrophin levels in muscle and cord. These biochemical changes, then, are activity-dependent and can perhaps be manipulated by physical therapies in patients. Lessons learned from training spinal-transected animals to step have led to a technique called body weight–supported treadmill training (BWSTT) in human subjects with acute and chronic SCI, to improve walking over ground. In this training, subjects are placed on a treadmill with their weight partially supported by a climbing harness attached to a lift that provides 0%–50% unloading of the paretic legs (22). Therapists manually assist the legs to step with the joint angles and timing of stance and swing typical of normal gait. Hip kinematics and load bearing are critical elements in eliciting rhythmical electromyographic activity in the flexor and extensor muscles of the legs in patients with complete and incomplete spinal cord lesions (23, 24). This form of task-oriented massed practice enhances motorskills learning in the presence of spared pathways for locomotor control (22, 25). Quasiexperimental studies suggested efficacy of BWSTT for walking in highly impaired subjects, especially with chronic SCI (25–27). However, a large multicenter randomized clinical trial of patients with recent SCI graded ASIA B, C, or D did not find significant differences for its primary outcome measures. In this study, 140 patients received either conventional mobility training or BWSTT for 12 weeks. Interventions started during initial inpatient rehabilitation, within 8 weeks of an incomplete SCI that spared sensation or some sensorimotor function below the lesion (28). At a 6-month follow-up, no significant differences

12 Dec 2003

14:2

AR

AR206-ME55-16.tex

AR206-ME55-16.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:255-282. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

SPINAL CORD INJURY

P1: GBC

263

were found in the percentage of subjects who recovered the ability to walk with minimal assistance or less help, and no differences were found in walking speed for those who could walk (29). It remains possible that BWSTT may be valuable for patients beyond the first 3–6 months after an incomplete SCI or other causes of a myelopathy, especially in subjects who have some ability to flex the hip and extend the knee. Treadmill training using robotic-assistive stepping devices is also coming into use to entrain sensorimotor integration for walking without physically taxing therapists (30). Cortical representational plasticity within the sensorimotor network has been demonstrated after the deafferentation and deefferentation caused by SCI. Functional neuroimaging studies, for example, have shown that the primary sensorimotor cortex representation for the hand may expand into the representation of the trunk that no longer receives sensory input from below a thoracic SCI (31). Functional magnetic resonance imaging performed during passive dorsiflexion of an affected ankle reveals intact inputs to the cortex in some patients who were thought to have a complete lesion, and step training using BWSTT in patients who walk poorly over ground can induce cortical reorganization in the leg representation associated with improved walking ability (32). Such reorganization may occur primarily within the cortical representation or associated with collateral sprouting within the cord, which was demonstrated after a complete experimental lesion (33).

Interventions to Reverse Muscle Atrophy Chronic nonuse, nonloading, and inactive biomechanical stress on skeletal muscle causes a drop in protein synthesis, an increase in protein degradation, and a preferential reduction in the myosin heavy-chain contractile proteins that are found in slow-twitch, fatigue-resistant muscles (34). Limb immobilization alone will cause atrophy at a rate of 1% to 5% a day for several weeks and a 40% loss of strength by 6 weeks. An upper motoneuron lesion that reduces selective motor control and strength may be magnified when patients do not perform resistance exercises. For muscles that can contract, selective isometric exercises against 60% of the maximal force of a muscle group may be the safest form of exertion for paretic groups. Neurotrophic factors such as insulin growth factor-1, neurotrophin-4, and BDNF are among the substances that have been used exogeneously in rodent experiments to enhance muscle function (35). Muscle normally produces these factors in response to stretch. Indeed, the neurotrophins help link mechanical stimuli to protein synthesis for muscle mass, motoneuron health, and synaptic efficacy. Beta-2 agonists, anabolic steroids, growth hormone, and angiotensin-converting enzyme inhibitors are among the other substances shown to exert an anabolic effect on muscle fibers in animals and humans (36). Clinical studies will be needed to reveal how to best employ such drugs to reverse atrophy and augment the strengthening effect of resistance exercises. Functional electrical stimulation (FES) can increase muscle mass in the legs and improve cardiovascular conditioning in patients with paraplegia (37). Surface

12 Dec 2003

14:2

264

AR

AR206-ME55-16.tex

DOBKIN

¥

AR206-ME55-16.sgm

LaTeX2e(2002/01/18)

P1: GBC

HAVTON

electrodes over the gluteal, hamstring, and quadriceps muscles are stimulated to sequentially contract against the resistance of bicycle pedals connected to an ergometer. However, more than half of patients who start a pedaling program drop out within 6 months and few continue beyond one year. The expense of the equipment, the assistance needed for set-up, the time taken from other daily activities, and the meager visible effects of FES exercise deter patients.

Annu. Rev. Med. 2004.55:255-282. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Neuroaugmentation Devices FES has also been used to stimulate or potentiate functional movements. For example, it has been used to stimulate standing and simple stepping, usually with some bracing and upper-extremity aides (ParaStep from Sigmedics, Northfield, IL). Embedded muscle electrode stimulators have also been commercialized for finger pinching and grasping (the FreeHand from Neurocontrol, Cleveland, OH) and for emptying the bowel and bladder (Vocare from Neurocontrol, Cleveland, OH) (38, 39). The necessity for surgical laminectomy, implantation of electrodes into the S-2 to S-4 ventral roots, and destruction of several dorsal roots has dissuaded some patients with upper motoneuron neurogenic bladder from employing the bladder stimulator, even though its positive economic and quality-of-life consequences have been demonstrated (39). FES products have not been financially successful, which may constrain future investments in sophisticated consumer aids. Newer nerve and neuromuscular stimulation systems using injectable bionic neurons (BIONsTM) are approaching large-scale clinical trials. BIONs are wireless electronic stimulators (2 mm × 15 mm) with a circuit chip and antenna coil that have been safely implanted by hypodermic needle injection in patients (39a). Control paradigms are being developed for a variety of applications. Microstimulation of the spinal cord is being studied to externally activate an alternating pattern of flexor and extensor muscle groups for stepping. The stimuli may drive small modules for synergistic movements called motor primitives (40) or central pattern generators for locomotion (41). Extradural stimulation at about L-1 near the dorsal roots elicits oscillating electromyographic activity in paraplegic patients (42). A patient who walked poorly after an incomplete SCI improved modestly with the combination of epidural electrical stimulation and treadmill training until walking ability was the same with or without stimulation (43). Brain-computer interfaces are in development to use biofeedback-trained cortical signals to drive an exoskeleton or FES device that translates the thought of a movement into a stimulated action (44). A variety of robotic exoskeletons and upper- and lower-extremity assistive devices are being tested (45). Passive step training or upper-extremity movement induced by FES or robotic assists may temporarily alter some of the membrane properties of motoneurons and motorcircuit physiology or augment proliferation of neural progenitors and produce neurotrophins in the spinal cord to enhance a strategy for neural repair. Some investigators speculate that the modest sensorimotor gains by Christopher Reeve five years after his cervical SCI can be attributed, in part, to automatically evoked

12 Dec 2003

14:2

AR

AR206-ME55-16.tex

AR206-ME55-16.sgm

LaTeX2e(2002/01/18)

SPINAL CORD INJURY

P1: GBC

265

movements during FES bicycling. The notion is that FES signals are a form of patterned neural activity that may alter central activity-dependent plasticity, leading to regeneration and recovery of function (46). However, sensory feedback induced by FES is not likely to lead to persistent central sensorimotor activation or motor skills learning (47). At present, passive and FES-driven repetitive movement cannot be justified as medically necessary.

Annu. Rev. Med. 2004.55:255-282. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

NEURAL REPAIR IN ANIMAL MODELS Studies of pathological specimens and animal models of injury demonstrate that injured neurons and axons are capable of turning on genes that lead to attempted axonal regeneration, but the milieu for extension of axons inhibits such growth (48, 49). Regeneration-associated genes include those that make (a) cytoskeletal proteins, such as the tubulins, that extend the axon growth cone’s protrusions or filipodia; (b) growth cone proteins, including GAP-43, that mediate signal transduction; and (c) cell adhesion molecules (NCAMS) that guide the growth cone, such as L1. Many of these genes require constant expression and interaction of their proteins for an axon to reach a target. The axon-elongation machinery interacts with the growth cone’s assessment of signals in its immediate environment. An environment of obstacles, such as scar tissue that contains both physical and chemical barriers and other inhibitory molecules, ordinarily shuts these genes down soon after SCI. Table 3 lists some specific approaches to neural repair that have led to axonal regrowth or sprouting of axons for 5–20 mm in a modest percentage of descending or ascending fibers beyond the site of injury. The efficacy of any intervention or combination of interventions depends on the extent of injury, the particular behavioral goal sought from regeneration, and the time from onset of injury. For example, most biological interventions have been attempted within one day to two weeks of the experimental SCI. The same approach that encouraged axonal regeneration within that period tends to fail when undertaken more than four weeks after injury (49a). Many behavioral studies have related the growth of axons to modest improvements in a few rodent behaviors, such as forelimb grasping and hindlimb movements (50, 51). Studies in some animal models of repair are compelling. Often, however, the tissue evidence for regenerating axons does not prove that the cut axons are regrowing. The apparent axons may actually be spared fibers or collateral branches from an uninjured axon (52). Sometimes, the rapidity of behavioral gains does not bear any plausible temporal relationship to the time needed for axons to regenerate past the lesion. Thus, evidence is still wanting for morphologic, physiologic, and behavioral proof of the influence of new inputs on their targets. Figure 1 shows the potential effects of one or more neural repair strategies when combined with training. An initial goal might be to regenerate 10%–20% of one or more descending tracts for 3–4 cm, which would be about two levels in the human cervical cord. Figure 2 makes it clear that 10 mm of axonal growth in

Nogo receptor antibody

Nogo peptide antagonist

Bind to Nogo receptor, blocking inhibition of axon growth by Nogo, MAG,∗ and OMGP∗ Bind receptor

Injected locally by osmotic CSF pump; immunization with CNS myelin component (e.g., Nogo-66) or injection of activated macrophages Potential for oral administration CSF pump; possible intravenous route

Inject Cell type needed may not differentiate or integrate; ethical issues for human studies with fetal tissues Technically complex, risky surgery

AR206-ME55-16.sgm

Binds Nogo to block inhibition of axon growth

Differentiate into matrix cells, neurons, and oligodendroglia; serve as neuronal relays within the bridge, repopulate gray and white matter, provide trophins White-to-gray and gray-to-white matter connections

Bone marrow stromal/stem cells Stem/progenitor cells Fetal spinal cord Injury-induced neurogenesis Peripheral nerve micrografts

AR206-ME55-16.tex

Nogo myelin inhibitor Nogo-A antibody

Align axons, migrate, produce trophins

¥

Schwann cell graft

Injection of alginate or smart biodegradable fibers may release cells and factors, but timing of release and integration with cord uncertain Inject just above and below lesion or within a matrix into a cystic area Quantity and quality of cells, associated matrix, ability of axon to travel beyond a bridge are uncertain; could induce scar, tumor

DOBKIN

OEGs∗

Fill cavity; may contain growth substances and cells

AR

Bridges Polymers, conduits

Delivery/Confounds

266

Action

14:2

Intervention

TABLE 3 Rodent studies of promising human interventions for spinal cord injury (SCI) repair

Annu. Rev. Med. 2004.55:255-282. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

12 Dec 2003 LaTeX2e(2002/01/18) P1: GBC

HAVTON

∗

Aid axonal growth and transport Neurite outgrowth Activates axonal growth cone Decrease number of cells making inhibitory molecules; lessen negative inflammatory cells

Other trophins∗ GDNF, FGF, VEGF, IGF-1 Immunophilin Inosine (93)

Electrical stimulation

X-irradiation

Limit neuronal apoptosis, aid axonal regeneration and guidance to targets; aid dendritic sprouting and learning mechanisms, such as LTP∗

Higher ratio of cAMP/cGMP for axon extension Block inhibitory effect of Mag, OMGP, Nogo A

(Continued)

Safety and potential negative impact in humans

AR206-ME55-16.sgm

Methods and efficacy in humans uncertain

Provide orally or infuse near injury site Oral or intravenous route

Inject engineered fibroblasts that secrete a trophin; inject or pump factor into CSF

Provide soon after injury for brief time by local infusion near injury site

Must be taken up by neuron (potential for oral administration)

AR206-ME55-16.tex

Neurotrophic factors BDNF,∗ NT-3

Rho GTPases

cGMP

Overcomes growth cone inhibitors

Infuse locally above and below edge of injury site

AR

Growth cone signaling cAMP

Digests inhibitors to foster axon growth in white matter

14:2

Proteoglycan inhibitor Chondroitinase

Annu. Rev. Med. 2004.55:255-282. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

12 Dec 2003 LaTeX2e(2002/01/18)

SPINAL CORD INJURY

P1: GBC

267

OEGs

∗ OEG, olfactory ensheathing glial cells; MAG, myelin-associated glycoprotein; OMGP, oligodendrocyte myelin glycoprotein; AMP, adenosine monophosphate; GMP, guanosine monophosphate; BDNF, brain-derived neurotrophic factor; GDNF, glial cell–derived neurotrophic factors; FGF, fibroblast growth factor; VEGF, vascular endothelial growth factor; IGF-1, insulin growth factor-1; LTP, long-term potentiation

Inject at dorsal horn entry zone Graded timing of interventions; more manipulations increase risk of tissue damage and adverse interactions to substances given

Local infusion or implant cells that secrete agent

AR206-ME55-16.sgm

OEGs Combinations of above (Figure 1)

Prevent small-fiber sprouting that leads to pain; increase large to small fiber ratio Regenerate sensory axons Bridge lesion, aid axons to grow beyond the bridge, and target spinal neurons

Inject near injury site Can reinnervate muscle within 1 month of cervical plexus avulsion in humans

CSF infusion

AR206-ME55-16.tex

Dorsal horn Neurotrophins

Neurotrophins, antiapoptotic proteins, caspase inhibitors Replace lower motoneurons Regenerate axons into peripheral nerve to muscle, sphincters, bladder

Migrate to surround axons

¥

Ventral horn Prevent apoptosis of injured or axotomized cells Implant neuronal precursors Reimplant ventral roots from below the lesion into cord above lesion

DOBKIN

Abundant in adult brain

Must proliferate, differentiate, and migrate to where needed; only travel short distances Activated in situ precursors may inhibit growth cone Inject into cord

AR

Oligodendrocyte precursors

Stimulate to become oligodendrocytes

Remyelination Neural stem cells

Delivery/Confounds

268

Action

14:2

Intervention

TABLE 3 (Continued)

Annu. Rev. Med. 2004.55:255-282. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

12 Dec 2003 LaTeX2e(2002/01/18) P1: GBC

HAVTON

12 Dec 2003

14:2

AR

AR206-ME55-16.tex

AR206-ME55-16.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:255-282. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

SPINAL CORD INJURY

P1: GBC

269

Figure 1 Experimental strategies for SCI repair. At the site of this hypothetical cystic lesion, a cellular matrix or biopolymer bridge is injected to fill the cavity. The matrix contains neuronal or oligodendroglial precursors and neurotrophins. Chondroitinase is injected into the white matter above and below the cavity to lessen inhibitory proteoglycans. An inhibitor to the Nogo receptor is injected around white matter tracts near the cavity to prevent growthcone collapse. Fibroblasts modified to secrete brain-derived neurotrophic factor (BDNF) or neurotrophin-3 (NT-3) are placed intermittently along the white matter tracts below the injury site to help signal the growth cones of regenerating axons. Neurotrophins or other diffusible attractive cueing molecules are injected to lead regenerating axons into the ventral gray matter of the cord at the levels of targeted motoneuron pools. Task-oriented sensorimotor training promotes regeneration and specific neuronal targeting, incorporates the sensorimotor pools into functional units, and enhances cortical representational plasticity for the motor control needed to relearn functional skills.

a rodent cord covers many more segments than in the human cord; the distance from a high thoracic injury to the lumbar motor pools in the human can exceed 20 cm.

Preventing Growth-Cone Inhibition Gliotic scarring and the traumatic cavity represent a physical barrier to regeneration of axons. The glial scar includes many cell types, including glial progenitors, injured oligodendrocytes, reactive astrocytes, microglia, fibroblasts, macrophages, and meningeal cells. Most of these cells produce molecules that can inhibit axonal growth, including proteoglycans, collagen, growth cone–collapsing substances, and repulsive guidance molecules. Their identification has led to therapeutic strategies. For example, enzymatic digestion of a glycosaminoglycan chain from

12 Dec 2003

14:2

Annu. Rev. Med. 2004.55:255-282. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

270

AR

AR206-ME55-16.tex

DOBKIN

¥

AR206-ME55-16.sgm

LaTeX2e(2002/01/18)

P1: GBC

HAVTON

Figure 2 Physical comparison of human and rodent spinal cord anatomy. The overall length of a rat brain and spinal cord, including the cauda equina, measure about 15 cm. The human cord alone, up to the cauda equina averages about 65 cm in a person who is 68 inches tall. The splay of roots of the cauda equina of the human cord reveals that 2–3 lumbar roots equal the diameter of the rat spinal cord. The differences in diameter of the cord sectioned at the lumbar enlargement are shown in the axial sections. Approximately 20 axial sections from the rat would fit within the human cord at this level. Relative distances necessary for targeting the cervical or lumbar motor pools for axonal regeneration from any level is at least four times greater in humans.

chondroitin sulfate proteoglycans has made the environs more permissive of axonal regeneration (53). Recent studies point to the therapeutic potential for targeting myelin-associated inhibitory substances produced by oligodendrocytes—myelin-associated glycoprotein (MAG), oligodendrocyte-myelin glycoprotein (OMGP), and Nogo-A. Growth inhibition in neurites is caused by their Nogo-66 and amino-Nogo domains when oligodendrocytes, periaxonal CNS myelin, and myelin debris are exposed by an injury. The normal role of Nogo-A may be to prevent collateral sprouting of axons after development. MAG, OMGP, and Nogo bind to the same

12 Dec 2003

14:2

AR

AR206-ME55-16.tex

AR206-ME55-16.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:255-282. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

SPINAL CORD INJURY

P1: GBC

271

region of the Nogo receptor at the growth cone. When the receptor complex is signaled by one of these inhibitory substances, a small guanosine triphosphatase (GTPase) called Rho and other cascades of intracellular activity stop support for the growth cone. Partial reversal of this inhibition has been accomplished using an intrathecal infusion of the small peptide NEP1-40, which inhibits binding of Nogo-66 to the Nogo receptor (54). The inhibition of Rho using small molecules that cross the blood-brain barrier may eventually accomplish the same effect in patients (55). Infusion of a substance that irreversibly inhibits Rho or the Nogo receptor may best be carried out in the first 3–10 days after SCI and continued for several weeks. Monoclonal antibodies raised against Nogo-A permit spinal axons to regenerate modest distances below a lesion (56). Immunization with myelin also promotes axonal regeneration in rodents after transection of the corticospinal tracts by blocking myelin-associated inhibitors (56a). Another approach is to increase the amount of the second messenger cyclic nucleotide adenosine monophosphate (cAMP), which induces genes to activate protein kinase A and to synthesize polyamines (57). The cAMP also inactivates Rho to make the growth cone less sensitive to MAG, OMGP, and Nogo. Indeed, a high ratio of intracellular cAMP to cGMP may stimulate growth-cone motility, whereas a low ratio or just a low level of cAMP may lead to collapse of the growth cone. More information about the Nogo receptor’s binding sites and the mechanisms that signal genes to make actin and other components of an extending axon growth cone may lead to practical ways to regenerate axons in an otherwise hostile milieu (57a).

Augmenting Axonal Regeneration The neurotrophins are polypeptide growth factors that include nerve growth factor (NGF), neurotrophins 3 and 4 (NT-3 and NT-4), and brain-derived neurotrophic factor (BDNF). They act specifically on one of three Trk receptor tyrosine kinases and the p75 neurotrophin receptor. In many ways, the receptors act as omnipresent neural sensors for extracellular and intracellular signals. The p75 receptor increases its expression after SCI, axotomy, or ischemia in motoneurons and oligodendrocytes. With the Nogo receptor, it forms a complex that inhibits neurite outgrowth by MAG and OMGP (48). The neurotrophins induce axonal extension and dendritic arborization by acting on cytoskeletal proteins. In addition, they guide axons over distances, aid neuronal cell survival, and participate in the regulation of synapse formation, synaptic plasticity, long-term potentiation for learning, and the release of neurotransmitters (58). BDNF seems better at protecting corticospinal neurons than eliciting axonal growth, whereas NT-3 elicits axonal regeneration in the spinal cord (59). BDNF has reversed the atrophy of rubrospinal neurons and promoted the regeneration of spinal axons into a peripheral nerve graft, even when given one year after SCI (60). Thus, trophic support may rescue chronically injured neurons, probably by triggering regeneration-associated genes. Other growth factors, including insulin-related growth factor, cytokine family members such as leukemia inhibitory factor, and ciliary neurotrophic factor also have trophic effects on

12 Dec 2003

14:2

272

AR

AR206-ME55-16.tex

DOBKIN

¥

AR206-ME55-16.sgm

LaTeX2e(2002/01/18)

P1: GBC

HAVTON

particular types of central axons, as well as on peripheral nerve axons. Trophic factors are diffusible and could be manipulated to make a gradient in their concentration for better axonal signaling over long distances. They can be delivered by a CSF pump or by perilesional implantation of fibroblasts, Schwann cells, or neural progenitor cells that have been genetically modified to express neurotrophins. The temporal pattern of providing neurotrophins may be important. For example, axotomized rubrospinal neurons showed greater survival and axonal regeneration after a delay, rather than after acute treatment with BDNF or NT-3 (50, 61).

Annu. Rev. Med. 2004.55:255-282. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Guiding Regenerating Axons The demonstration that CNS axons may be enticed to regenerate is an important breakthrough. It is unclear, however, which axons among the descending tracts are most critical for restoring voluntary movement and which targets in the spinal gray matter below the lesion will enable restitution of function. In addition, strategies that induce axonal regeneration may result in aberrant sprouting of axons into both sides of the cord and into dorsal and ventral horns. Thus, strategies to guide the regenerating axons in SCI lesions are also needed. The growth cone can be steered by local forces of attraction and repulsion. These include variations in the cone’s adhesion to different surfaces in the milieu, the mechanical forces generated by the cone, and transduction of signals in the milieu. Growth cones may be guided by direct-current electrical fields, as well as by chemotropic gradients of attractive (e.g., netrin-1) and repulsive (e.g., MAG) cues (62). Physical, chemical, and electrical cues, then, may aid axonal guidance through the spinal cord white matter and into gray matter targets. Extracellular axon guidance molecules present during development include the netrins, semaphorins, slits, and ephrins. Transmembrane proteins on the surface of the growth cone, such as plexin, initiate a signal transduction cascade for collapse, repulsion, or turning of the growth cone when they bind to one of these families of developmental guidance molecules (63). In general, guidance molecules affect actin-based motility by directing and stabilizing the assembly of microtubules. These steering molecules signal different receptors at a growth cone to attract or repel fingers of actin webs and bundles, as well as microtubule filaments within the stretching and shrinking protrusions of the growth cone. Interactions are complex. A guidance cue may act on a navigating growth cone differently over the course of the axon’s journey, as occurs during development, by attracting, repelling, or modulating internal and external factors (64). Although investigators can only speculate on how to best control the distribution of cues over space and time for rewiring the spinal cord, manipulations have shown some success. For example, the Roundabout (Robo) family of slit receptors can be inhibited to promote axon sprouting. Cyclic AMP and cGMP can reverse growth-cone repulsion. Neurotrophins also have neurotropic actions that steer the growth cone toward them. Mechanical guidance of regenerating axons in damaged peripheral nerves depends heavily on the integrity of the nerve’s endoneurial tubes. An experimental SCI by surgical transection leaves much of the structure above and below the

12 Dec 2003

14:2

AR

AR206-ME55-16.tex

AR206-ME55-16.sgm

LaTeX2e(2002/01/18)

SPINAL CORD INJURY

P1: GBC

273

cut intact for axonal regeneration, compared, for example, to the focal contusion model, which disrupts this structure. Mechanical guidance by the intact central myelin or by implanted guidance channels such as peripheral nerve grafts may be a useful repair strategy (64a).

Annu. Rev. Med. 2004.55:255-282. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Implanting Cells or Stimulating Neurogenesis Exogenous and endogenous stem cells, or more differentiated precursors of neurons and oligodendrocytes, offer an exciting but challenging approach for neural repair (59, 65). The cellular and extracellular milieu of the injury and the genetic makeup of the implants may alter the survival, differentiation, migration, and integration of these cells. The injury itself leads to signaling of progenitors of neurons and especially of glia to stream out from the central canal of the cord or proliferate from within the white matter of the brain and spinal cord (66). The fate and function of these endogenous multipotential cells is uncertain. However, exogenous multipotent stem cells and neural progenitor cells may be programmed in cell culture and by environmental signals to serve as potential tools for repair. A few examples of experimental manipulations follow (Table 2). After a complete cord transection in adult rodents, fetal spinal cord tissue placed into the lesion produced greater descending axonal regeneration when combined with BDNF or NT-3 (50). Because fetal tissue poses ethical constraints, other cell types have been tried in rodent SCI. Progenitor cells used for spinal cord implantation include neural progenitors derived from subventricular and olfactory regions of embryonic and adult rodent and human brain tissue, and nonneural progenitors from marrow stromal cells. Embryonic stem cells in rodents have been differentiated into specific neuronal cell types, such as motoneurons of the ventral horn of the cord (67), by exposure to signals drawn from normal development, such as neurotrophins. The motoneurons integrated within ventral horns of the cord and sent out axons that made cholinergic synapses on muscle. In one study, embryonic stem cells from mice were transplanted into a spinal contusion in rats after being manipulated into a neural lineage. Although most cells died, the majority of the survivors became oligodendrocytes that remyelinated adjacent axons, and 10% became neurons (68). Other studies have had less success, finding only a small percentage of surviving cells that did not become astrocytes. The milieu of a SCI may inhibit neuronal precursor differentiation and require interventions beyond transplantation (69), such as adding a neurotrophin. Some lines of neural stem cells produce their own neurotropic factors, which leads to host axonal regeneration (69a). Newer techniques may generate pure populations of neurons from human fetal stem cells that are capable of producing, for example, acetylcholine, regardless of the milieu into which they are grafted (70). Schwann cells from peripheral nerve and cells from olfactory mucosa have special characteristics that make them candidates for use after SCI in humans. Cultured Schwann cells injected into a rat spinal cord contusion one week after injury were somewhat superior to cultured olfactory ensheathing glial (OEG) cells,

12 Dec 2003

14:2

Annu. Rev. Med. 2004.55:255-282. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

274

AR

AR206-ME55-16.tex

DOBKIN

¥

AR206-ME55-16.sgm

LaTeX2e(2002/01/18)

P1: GBC

HAVTON

grown from cells of the olfactory bulb, in enabling axonal growth, myelination, and partial hindlimb stepping (71). Cultured OEGs have promoted axonal regeneration over distances of 10 mm, repressed axonal branching, aligned growing axons, and improved forepaw and hindlimb movement when injected into a focal corticospinal tract lesion (72) or into a transected cord (73). The combination of olfactory fibroblasts and ensheathing cells embedded in a matrix improved phrenic nerve and climbing skills after a high cervical cord hemisection associated with possible dorsal and ventral tract repair (74). However, these models of injury may produce a different level of inflammatory and local signaling molecule response compared to a traumatic contusion. In addition, most regenerating axons in these models seem to end just below the lesion, suggesting that if they are making effective synapses 1–2 weeks after the transplantation, any restitution of function must be related to spared fibers or to propriospinal or other local polysynaptic connections that were engaged, and not to restitution of a pathway. Other types of cells have been implanted to try to restore a particular descending pathway. For example, brainstem raphe cells that produce serotonin have restored treadmill locomotion after implantation into a T-11 injury in the rat. Engineered fibroblasts that produce BDNF or NT-3 injected below the lesion can signal axons from above a lesion to extend toward them (59), protect the cell bodies, and regenerate specific pathways such as dorsal root, serotonergic, and rubrospinal fibers (75). These trophins may also induce oligodendrocytes to myelinate fibers in a contusion (76). Certain neural precursor cells derived from the human brain or from marrow have been induced to myelinate fibers to improve central conduction (77, 78). In a mouse model, cultured adult neural precursors drawn from the subventicular zone reversed much of the demyelination caused by experimental multiple sclerosis (79). Interventions for focal demyelination may be especially useful for patients with a transverse myelitis, focal infarct, or spinal multiple sclerosis (80). The success of implants will depend, in part, on the local cellular milieu produced by each disease and the timing of an implant in relation to onset of the injury. Transplantation studies for SCI in humans have started in North America, Europe, Australia, South America, China, and Russia (Table 3). In the United States and Europe, patients with multiple sclerosis have received autologous Schwann cells and OEGs placed into plaques in safety trials. Oligodendrocyte precursors from pigs were implanted into the lesion bed of six subjects after SCI in another safety trial (Diacrin Corp). Results are pending. Reports from American physicians who visited Beijing describe implantation of OEGs in patients six or more months after a traumatic SCI (http://carecure.atinfopop.com). The cells were expanded from olfactory tissue of one fetus and injected just above and below the lesion during open visualization of the cord. Sensation supposedly improved within days of the injection over a few levels and motor function increased for up to two levels below the lesion. A study using embryonic stem cells in China was halted, apparently because no recovery was seen in 30 cases. One hopes that meaningful data will be published by these and other investigators.

12 Dec 2003

14:2

AR

AR206-ME55-16.tex

AR206-ME55-16.sgm

LaTeX2e(2002/01/18)

SPINAL CORD INJURY

P1: GBC

275

Annu. Rev. Med. 2004.55:255-282. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Reimplanting Nerve Roots Nerve roots may be torn or avulsed by traumatic SCI. This injury may occur at any segmental level but is perhaps a more critical cause of disability in patients who suffer a burst fracture, vertebral subluxation, or penetrating injury below the T-10 vertebral body, which accounts for 20% of traumatic SCIs. Here, the tip of the cord or conus medullaris and the nearby floating nerve roots of the cauda equina may be sheared, torn, crushed, or avulsed. Clinically, this lower motoneuron lesion causes flaccid paralysis, muscle atrophy, and loss of stretch reflexes from ventral root and anterior horn injury; flaccid bladder and sphincters from preganglionic parasympathetic and autonomic nerve injury; and at-level and below-level sensory loss and, often, central neuropathic pain from dorsal horn and ganglion injury. The motor and autonomic neurons of these roots start to die by apoptosis within one week after being disconnected at the interface between the central and peripheral nervous system (80a). Neurotrophic factors may be able to rescue these cell bodies. Remarkably, reimplantation of avulsed ventral roots, both experimentally and in humans after brachial plexus avulsions (81), can lead to at least partial recovery of motor function (82). The motoneurons and preganglionic parasympathetic neurons, at least in animal models, regenerate toward the reimplanted ventral root, probably lured by substances such as neurotrophins within the nerve’s Schwann cells. This axonal regeneration is associated with remyelination of nerve root fibers (80a). Thus, at the time of surgical management of acute conus/cauda SCI, identification and reimplantation of ventral roots into the cord at the level of a tear or avulsion or into the ventrolateral cord at a level above the gray matter injury could restore motor and bladder function when combined with rehabilitation training.

Maintaining Dorsal Horn Function Dorsal root and dorsal horn injury may cause sensory loss and pain at the segmental levels of the SCI. Dorsal root ganglion cells do not ordinarily regenerate into the dorsal horn after an avulsion. Strategies to promote spinal ingrowth of subpopulations of these neurons are similar to strategies for ventral root and central axonal regeneration. Interventions include placing fibroblast-secreting cells or a local intrathecal infusion of neurotrophins such as BDNF, NGF, or NT-3 into the dorsal horn, upregulation of cAMP, transplantation of embryonic spinal tissue or OEGs into the transition zone (83), and implantation of a peripheral nerve conduit from the entry zone to the dorsal root. These manipulations have restored sensation and bladder function after an experimental dorsal rhizotomy in rodents (84). Attempts to restore sensory inputs within and near the injured dorsal horn may, however, activate pathways for pain, autonomic dysreflexia, and flexor or extensor spasms. This potential negative aspect of exogenous neurotrophins derives from their role in pain-associated neuroplasticity, both neurite outgrowth and physiologic reorganization (85, 86). Several therapeutic strategies have prevented

12 Dec 2003

14:2

276

AR

AR206-ME55-16.tex

DOBKIN

¥

AR206-ME55-16.sgm

LaTeX2e(2002/01/18)

P1: GBC

HAVTON

these complications in rodents. Antibodies to NGF prevented small fiber afferents from sprouting in the dorsal horn, which reduced autonomic responses to pain. Other neurotrophins, such as GDNF, enhance large-fiber proliferation and offset the balance toward pain induced by small-fiber proliferation. Neurotrophins may also increase the regeneration of ascending axons from the dorsal horn.

Annu. Rev. Med. 2004.55:255-282. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Bridges to Span SCI-Induced Scar and Cavities The cavity and scar tissue that form after SCI may have to be filled or bridged to allow regenerating axons to descend and ascend across a cystic cord to reach their targets (Figure 1). Schwann cells, OEGs, embryonic spinal cord tissue, and tiny peripheral nerve segments have allowed at least some types of axons to cross from white matter into white or gray matter below the lesion. The reproducibility of successful experiments in other laboratories has been poor, however. Engineered substrates may include synthetic channels or biopolymers laced with guidance channels, Schwann cells or OEGs, a permissive substrate such as laminin, and neurotrophic factors. A bridge may release substances on a schedule that attracts axons into and then out of the bridge. Many forms of “smart” (biodegradable) bridges for central and peripheral axons have shown some efficacy in animal models (87, 88).

Pharmacologic Interventions After a complete SCI in cats and rodents, inhibitory neurotransmitters are expressed in the motor pools below the lesion (89). Other neurotransmitters, such as serotonin and norepinephrine, tend to excite the pools related to locomotion (90). Thus, a strategy to augment rehabilitation and neural repair strategies may include medications that act as antagonists or agonists for the control of excitation and inhibition, as well as routine avoidance of drugs that may have a negative impact on excitation and inhibition of motor and bowel and bladder autonomic activity. The spinal cord is also capable of learning through training (91). Drugs that affect membrane properties and synaptic efficacy are being studied to determine if they may augment motor skills learning by affecting spinal motor pools or mechanisms of cortical representational plasticity for the practiced movements.

FUTURE DIRECTIONS Animal models of injury and repair allow the researcher to study or manipulate specific biological changes and responses to interventions. Robust changes in preclinical studies should include anatomical, physiological, and behavioral correlative evidence of the effects of a biological and training strategy (94). Positive outcomes for limiting injury or inducing greater repair in a transgenic mouse or inbred rodent species cannot predict a similar outcome in human subjects. In this decade, interventions for neural repair that produce robust gains in rodents may

12 Dec 2003

14:2

AR

AR206-ME55-16.tex

AR206-ME55-16.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:255-282. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

SPINAL CORD INJURY

P1: GBC

277

have to be tested in small studies with larger mammals or nonhuman primates to evaluate safety and dose-response relationships. Clinicians, including neurologists, physiatrists, neurosurgeons, and orthopedic surgeons, as well as rehabilitation therapists, must develop ethical designs to test these approaches and measure important and relevant outcomes in patients with recent motor-complete SCI. Clinical trials with chronically injured subjects may run in parallel, especially for safety trials of invasive interventions. Primary outcomes for trials could include, for example, a robust increase in the likelihood that patients with high quadriplegia breathe without a ventilator, gain better use of an upper extremity by improving two or more motor levels, or have fewer complications of dysautonomia. For subjects with paraplegia from a mid- to low-thoracic lesion, reciprocal stepping at a low energy cost, control of sphincters and voluntary bowel and bladder emptying, and less pain at and below the level of injury will be clinically meaningful outcome measures. The latter endpoints will also be appropriate for studies in patients with conus or cauda equina injuries. The potential pool of participants eligible for clinical trials of neural repair interventions who have incurred a recent, profound SCI is rather small compared to other common neurological diseases. If only subjects classified as ASIA A are included, and 10%–20% of patients meet entry criteria (as is typical of such studies), as few as 600 subjects a year would be available across the United States and 800 in Europe. ASIA B subjects may also be reasonable to enter into early interventional studies if the repair strategy does not surgically disrupt the cord, as well as studies of chronically paraplegic subjects. However, experimental interventions in patients with chronic SCI, which perhaps means >3 months after trauma for certain approaches, probably should not include patients with debilitating pain, dysautonomia or spasms, or a symptomatic tethered cord or syrinx. These complications could worsen with biological manipulation and would interfere with interpretation of any adverse responses to the intervention. Thus, multicenter clinical trials will require great cooperation between institutions and adequate funding to identify and randomize just 50–100 subjects with a cervical or a thoracic injury of interest and to monitor outcomes for at least two years. Present care and optimal deployment of biological interventions demand welldefined physical rehabilitation and pharmacological therapies. Clinicians must continue to develop interventions that drive mechanisms of activity-dependent plasticity to best incorporate spared and new circuitry into a neural matrix for important behavioral gains.

ACKNOWLEDGMENTS We thank the National Institutes of Health for funding through HD37439, HD39629, T32HD07416, NS42719, NS16333, and T32NS07449; the Nathan Shapell Foundation; the State of California Roman Reed Bill; and the Larry L. Hillblom Foundation.

12 Dec 2003

14:2

278

AR

AR206-ME55-16.tex

DOBKIN

¥

AR206-ME55-16.sgm

LaTeX2e(2002/01/18)

P1: GBC

HAVTON

The Annual Review of Medicine is online at http://med.annualreviews.org

Annu. Rev. Med. 2004.55:255-282. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

LITERATURE CITED 1. Raineteau O, Schwab M. 2001. Plasticity of motor systems after incomplete spinal cord injury. Nat. Rev. Neurosci. 2:263– 73 2. Dobkin BH. 2003. The Clinical Science of Neurologic Rehabilitation. New York: Oxford Univ. Press 3. Maniar H, Sundt T, Prasad S, et al. 2003. Delayed paraplegia after thoracic and thoracoabdominal aneurysm repair: a continuing risk. Ann. Thorac. Surg. 75:113–20 4. Kwon B, Oxland T, Tetzlaff W. 2002. Animal models used in spinal cord regeneration research. Spine 27:1504–10 5. Inman D, Guth L, Steward O. 2002. Genetic influences on secondary degeneration and wound healing following spinal cord injury in various strains of mice. J. Comp. Neurol. 451:225–35 6. Di Giavanni S, Knoblach S, Brandoli C, et al. 2003. Gene profiling in spinal cord injury shows role of cell cycle neuronal death. Ann. Neurol. 53:454–68 7. Nesic O, Svrakic N, Xu G, et al. 2002. DNA microarray analysis of the contused spinal cord: effect of NMDA inhibition. J. Neurosci. Res. 68:406–23 8. Gilgun-Sherki Y, Rosenbaum Z, Melamed E, et al. 2002. Antioxidant therapy in acute central nervous system injury. Pharmacol. Rev. 54:271–84 8a. Wells J, Hurlbert RJ, Fehlings M, Yong V. 2003. Neuroprotection by minocycline facilitates significant recovery from spinal cord injury in mice. Brain 126:1628–37 9. Jones T, Basso D, Sodhi A, et al. 2002. Pathological CNS autoimmune disease triggered by traumatic spinal cord injury: implications for autoimmune vaccine therapy. J. Neurosci. 22:2690–2700 10. McFarlane S. 2003. Metalloproteases: carving out a role in axon guidance. Neuron 37:559–62 11. Kakulas B. 1999. A review of the neuro-

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

pathology of human spinal cord injury with emphasis on special features. J. Spinal Cord Med. 22:119–24 Fehlings M, Tator C. 1999. An evidencebased review of decompressive surgery in acute spinal cord injury: rationale, indications, and timing based on experimental and clinical studies. J. Neurosurg. 91(Spine 1):1–11 Edwards C, Riew K, Anderson P, et al. 2003. Cervical myelopathy: current diagnostic and treatment strategies. Spine J. 3:68–81 Hadley M. 2002. Pharmacological therapy after acute cervical spinal cord injury. Neurosurgery 50:S63–S72 Hauben E, Schwartz M. 2002. Therapeutic vaccination for spinal cord injury: helping the body cure itself. Trends Pharmacol. Sci. 24:7–12 Popovich P, Jones T. 2002. Manipulating neuroinflammatory reactions in the injured spinal cord: back to basics. Trends Pharmacol. Sci. 24:13–17 Consortium for Spinal Cord Medicine. 1999. Outcomes following traumatic spinal cord injury: clinical practices guidelines for health-care professionals. Washington, DC: Paralyzed Veterans of America de Leon R, Tamaki H, Hodgson J, et al. 1999. Hindlimb locomotor and postural training modulates glycinergic inhibition in the spinal cord of the adult spinal cat. J. Neurophysiol. 82:359–69 Barbeau H, McCrea D, O’Donovan M, et al. 1999. Tapping into spinal circuits to restore motor function. Brain Res. Rev. 30:27–51 Duysens J, Van de Crommert H. 1998. Neural control of locomotion: the central pattern generator from cats to humans. Gait Posture 7:131–41 Edgerton V, de Leon R, Harkema S, et al.

12 Dec 2003

14:2

AR

AR206-ME55-16.tex

AR206-ME55-16.sgm

LaTeX2e(2002/01/18)

SPINAL CORD INJURY

22.

Annu. Rev. Med. 2004.55:255-282. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

23.

24.

25.

26.

27.

28.

29.

30.

31.

2001. Retraining the injured spinal cord. J. Physiol. 533:15–22 Dobkin B. 1999. Overview of treadmill locomotor training with partial body weight support: a neurophysiologically sound approach whose time has come for randomized clinical trials. Neurorehabil. Neural Repair 13:157–65 Harkema S, Hurley S, Patel U, et al. 1997. Human lumbosacral spinal cord interprets loading during stepping. J. Neurophysiol. 77:797–811 Dietz V, Muller R, Colombo G. 2002. Locomotor activity in spinal man: significance of afferent input from joint and load receptors. Brain 125:2626–34 Barbeau H. 2003. Locomotor training in neurorehabilitation: emerging rehabilitation concepts. Neurorehabil. Neural Repair 17:3–11 Dietz V. 2003. Spinal cord pattern generators for locomotion. Clin. Neurophysiol. 114:1379–89 Wernig A, Nanassy A, Miller S. 1998. Maintenance of locomotor abilities following Laufband (treadmill) therapy in para- and tetraplegic persons: follow-up studies. Spinal Cord 36:744–49 Dobkin B, Apple D, Barbeau H, et al. 2003. Methods for a randomized trial of weight-supported treadmill training versus conventional training for walking during inpatient rehabilitation after incomplete traumatic spinal cord injury. Neurorehabil. Neural Repair 17:153– 67 Dobkin B, Apple D, Barbeau H, et al. 2003. Results of the Spinal Cord Injury Locomotor Trial (SCILT). Presented at annual joint scientific meeting of Am. Soc. Neurorehabil. and Am. Congr. Rehabil. Med., Tucson, AZ, Oct. 23 Colombo G, Joerg M, Schreier R, et al. 2000. Treadmill training of paraplegic patients using a robotic orthosis. J. Rehabil. Res. Dev. 37:693–700 Bruehlmeier M, Dietz V, Leenders K, et al. 1998. How does the human brain deal

32.

33.

34.

35.

36.

37.

38.

39.

39a.

40.

P1: GBC

279

with a spinal cord injury? Eur. J. Neurosci. 10:3918–22 Dobkin B. 2000. Spinal and supraspinal plasticity after incomplete spinal cord injury: correlations between functional magnetic resonance imaging and engaged locomotor networks. In Progress in Brain Research, ed. F Seil, 128:99–111. Amsterdam: Elsevier Fouad K, Pedersen V, Schwab M, et al. 2001. Cervical sprouting of corticospinal fibers after thoracic spinal cord injury accompanies shifts in evoked motor responses. Curr. Biol. 11:1766–70 Caiozzo VJ. 2002. Plasticity of skeletal muscle phenotype: mechanical consequences. Muscle Nerve 26:740–68 Singleton J, Feldman E. 2001. Insulin-like growth factor-1 in muscle metabolism and myotherapies. Neurobiol. Dis. 8:541–54 Onder G, Penninx B, Balkrishnan R, et al. 2002. Relation between use of angiotensin-converting enzyme inhibitors and muscle strength and physical function in older women: an observational study. Lancet 359:926–30 Scremin A, Kurta L, Gentili A, et al. 2000. Increasing muscle mass in spinal cord injured persons with a functional electrical stimulation exercise program. Arch. Phys. Med. Rehabil. 80:1531–36 Peckham P, Keith M, Kilgore K, et al. 2001. Efficacy of an implanted neuroprosthesis for restoring grasp in tetraplegia: a multicenter study. Arch. Phys. Med. Rehabil. 82:1380–88 Creasey G, Grill J, Korsten J, et al. 2001. An implantable neuroprosthesis for restoring bladder and bowel control to patients with spinal cord injuries: a multicenter trial. Arch. Phys. Med. Rehabil. 82:1512–19 Loeb GE, Peck RA, Moore W, Hood K. 2001. BIONTM system for distributed neural prosthetic interfaces. Med. Eng. Phys. 23:9–18 Thoroughman K, Shadmehr R. 2000. Learning of action through adaptive

12 Dec 2003

14:2

280

41.

42.

Annu. Rev. Med. 2004.55:255-282. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

43.

44.

45.

46.

47.

48.

49.

49a.

50.

AR

AR206-ME55-16.tex

DOBKIN

¥

AR206-ME55-16.sgm

LaTeX2e(2002/01/18)

P1: GBC

HAVTON

combination of motor primitives. Nature 407:742–47 Pearson K. 2000. Neural adaptation in the generation of rhythmic behavior. Annu. Rev. Physiol. 62:723–53 Dimitrijevic M, Gerasimenko Y, Pinter M. 1998. Evidence for a spinal central pattern generator in humans. Ann. NY Acad. Sci. 860:360–76 Herman R, He J, D’Luzansky S, et al. 2002. Spinal cord stimulation facilitates functional walking in a chronic incomplete spinal cord injury. Spinal Cord 40: 65–68 Wolpaw J, Birbaumer N, McFarland D, et al. 2002. Brain-computer interfaces for communication and control. Clin. Neurophysiol. 113:767–91 Reinkensmeyer D, Lum P, Winters J. 2002. Emerging technologies for improving access to movement therapy following neurologic injury. In Emerging and Accessible Telecommunications, Information and Healthcare Technologies, ed. J Winters, C Robinson, R Simpson, G Vanderheiden, pp. 123–38. New York: IEEE Press McDonald J, Becker D, Sadowsky C, et al. 2002. Late recovery following spinal cord injury. Case report and review of the literature. J. Neurosurg. 97:252–65 Dobkin B. 2003. Do electrically stimulated sensory inputs and movements lead to long-term plasticity and rehabilitation gains? Curr. Opin. Neurol. 16:In press David S, Lacroix S. 2003. Molecular approaches to spinal cord repair. Annu. Rev. Neurosci. 26:411–40 Selzer M. 2003. Promotion of axonal regeneration in the injured CNS. Lancet Neurol. 2:157–66 Houle JD, Tessler A. 2003. Repair of chronic spinal cord injury. Exp. Neurol. 182:247–60 Coumans J, Lin T, Dai H, et al. 2001. Axonal regeneration and functional recovery after complete spinal cord transection in rats by delayed treatment with transplants

51.

52.

53.

54.

55.

56.

56a.

57.

57a. 58.

59.

and neurotrophins. J. Neurosci. 21:9334– 44 Murray M, Kim D, Liu Y, et al. 2002. Transplantation of genetically modified cells contributes to repair from spinal injury. Brain Res. Rev. 40:292–300 Steward O, Zheng B, Tessier-Levigne M. 2003. False resurrections: distinguishing regenerated from spared axons in the injured central nervous system. J. Comp. Neurol. 459:1–8 Bradbury E, Moon L, Popat R, et al. 2002. Chrondroitinase ABC promotes functional recovery after spinal cord injury. Nature 416:636–40 GrandPre T, Li S, Strittmatter S. 2002. Nogo-66 receptor antagonist peptide promotes axonal regeneration. Nature 417: 547–51 Dergham P, Ellezam B, Essagian C, et al. 2002. Rho signaling pathway targeted to promote spinal cord repair. J. Neurosci. 22:6570–77 Bareyre F, Haudenschild B, Schwab M. 2002. Long-lasting sprouting and gene expression changes induced by the monoclonal antibody IN-1 in the adult spinal cord. J. Neurosci. 22:7097–110 Sicotte M, Tsatas O, Jeong S, et al. 2003. Immunization with myelin or recombinant Nogo-66/MAG in alum promotes axon regeneration and sprouting after corticospinal tract lesions in the spinal cord. Mol. Cell. Neurosci. 23:251–63 Cai D, Qiu J, Cao Z, et al. 2001. Neuronal cyclic AMP controls the developmental loss in ability of axons to regenerate. J. Neurosci. 21:4731–39 Woolf CJ. 2003. No Nogo: Now where to go? Neuron 38:153–56 Chao M. 2003. Neurotrophins and their receptors: a convergence point for many signalling pathways. Nat. Neurosci. 4: 299–309 Blesch A, Lu P, Tuszynski M. 2002. Neurotrophic factors, gene therapy, and neural stem cells for spinal cord repair. Brain Res. Bull. 57:833–38

12 Dec 2003

14:2

AR

AR206-ME55-16.tex

AR206-ME55-16.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:255-282. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

SPINAL CORD INJURY 60. Plunet W, Kwon B, Tetzlaff W. 2002. Promoting axonal regeneration in the central nervous system by enhancing the cell body response to axotomy. J. Neurosci. Res. 68:1–6 61. Novikova L, Novikov L, Kellerth J-O. 2000. Survival effects of BDNF and NT3 on axotomized rubrospinal neurons depend on the temporal pattern of neurotrophin administration. Eur. J. Neurosci. 12:776–80 62. Ming G-L, Henley J, Tessier-Lavigne M, et al. 2001. Electrical activity modulates growth cone guidance by diffusible factors. Neuron 29:441–52 63. Fournier A, Strittmatter S. 2001. Repulsive factors and axon regeneration in the CNS. Curr. Opin. Neurobiol. 11:89–94 64. Dickson B. 2002. Molecular mechanisms of axon guidance. Science 298:1959–64 64a. Cheng H, Cao Y, Olson L. 1996. Spinal cord repair in adult paraplegic rats: partial restoration of hind limb function. Science 273:510–14 65. Horner P, Power A, Kempermann G, et al. 2000. Proliferation and differentiation of progenitor cells throughout the intact adult spinal cord. J. Neurosci. 20:2218– 28 66. Nunes M, Roy N, Keyoung H, et al. 2003. Identification and isolation of multipotential neural progenitor cells from the subcortical white matter of the adult human brain. Nat. Med. 9:439–47 67. Wichterle H, Lieberam I, Jessell T. 2002. Directed differentiation of embryonic stem cells into motor neurons. Cell 110:385–97 68. McDonald J, Liu X-Z, Qu Y, et al. 1999. Transplanted embryonic stem cells survive, differentiate, and promote recovery in injured rat spinal cord. Nat. Med. 5:1410–12 69. Cao Q, Howard R, Dennison J, et al. 2002. Differentiation of engrafted neuronalrestricted precursor cells is inhibited in the traumatically injured spinal cord. Exp. Neurol. 177:349–59

P1: GBC

281

69a. Lu P, Jones LL, Snyder EY, Tuszynski M. 2003. Neural stem cells constitutively secrete neurotrophic factors and promote extensive host axonal growth after spinal cord injury. Exp. Neurol. 181:115–29 70. Wu P, Tarsenko Y, Gu Y, et al. 2002. Region-specific generation of cholinergic neurons from fetal human neural stem cells grafted in adult rat. Nat. Neurosci. 5:1271–78 71. Takami T, Oudega M, Bates M, et al. 2002. Schwann cell but not olfactory ensheathing glia transplants improve hindlimb locomotor performance in the moderately contused adult rat thoracic spinal cord. J. Neurosci. 22:6670–81 72. Li Y, Field P, Raisman G. 1998. Regeneration of adult rat corticospinal axons induced by transplanted olfactory ensheathing cells. Neuroscience 18:10514–24 73. Ramon-Cueto A, Cordero M, Santos-Benito F, et al. 2000. Functional recovery of paraplegic rats and motor axon regeneration in their spinal cords by olfactory ensheathing glia. Neuron 25:425–36 74. Li Y, Decherchi P, Raisman G. 2003. Transplantation of olfactory ensheathing cells into spinal cord lesions restores breathing and climbing. J. Neurosci. 23:727–31 75. Liu Y, Kim D, Himes T. 1999. Transplants of fibroblasts genetically modified to express BDNF promote regeneration of adult rat rubrospinal axons and recovery of forelimb function. J. Neurosci. 19:4370–87 76. McTigue D, Horner P, Stokes B, et al. 1998. Neurotrophin-3 and brain-derived neurotrophic factor induce oligodendrocyte proliferation and myelination of regenerating axons in the contused adult rat spinal cord. J. Neurosci. 18:5354– 65 77. Akiyama Y, Radtke C, Kocsis J. 2002. Remyelination of the rat spinal cord by transplantation of identified bone marrow stromal cells. J. Neurosci. 22:6623–30 78. Akiyama Y, Honmou O, Kato T, et al. 2001. Transplantation of clonal neural

12 Dec 2003

14:2

282

79.

Annu. Rev. Med. 2004.55:255-282. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

80.

80a.

81.

82.

83.

84.

85.

AR

AR206-ME55-16.tex

DOBKIN

¥

AR206-ME55-16.sgm

LaTeX2e(2002/01/18)

P1: GBC

HAVTON

precursor cells derived from adult human brain establishes functional peripheral myelin in the rat spinal cord. Exp. Neurol. 167:27–39 Pluchino S, Quattrini A, Brambilla E, et al. 2003. Injection of adult neurospheres induces recovery in a chronic model of multiple sclerosis. Nature 422:686–94 Keirstead H, Hur T, Rogister B. 1999. Polysialylated neural cell adhesion molecule-positive CNS precursors generate both oligodendrocytes and Schwann cells to remyelinate the CNS after transplantation. J. Neurosci. 19:7529–36 Nieto J, Hoang TX, Tillakaratne NJK, Havton LA. 2002. Lumbosacral spinal cord neurons reinnervate implanted avulsed ventral roots in a cauda equina injury injury and repair model. Soc. Neurosci. Abstr. 31:852 Carlstedt T, Anand P, Hallin R, et al. 2000. Spinal nerve root repair and reimplantation of avulsed ventral roots into the spinal cord after brachial plexus injury. J. Neurosurg. 93:237–47 Cullheim S, Wallquist W, Mammarberg H, et al. 2002. Properties of motoneurons underlying their regenerative capacity after axon lesions in the ventral funiculus or at the surface of the spinal cord. Brain Res. Rev. 40:309–16 Navarro X, Valero A, Gudino G. 1999. Ensheathing glia transplants promote dorsal root regeneration and spinal reflex restitution after multiple lumbar rhizotomy. Ann. Neurol. 45:207–15 Pascual J, Gudino-Cabrera G, Insausti R, et al. 2002. Spinal transplants of olfactory ensheathing cells promote axon regeneration and bladder activity after bilateral lumbosacral dorsal rhizotomy in the adult rat. J. Urol. 167:1522–26 Ji R, Woolf C. 2000. Neuronal plasticity and signal transduction in nociceptive neurons: implications for the initiation

86.

87.

88.

89.

90.

91.

92.

93.

94.

95.

96.

and maintenance of pathological pain. Neurobiol. Dis. 1–10 Ikeda H, Heinke B, Ruscheweyh R, et al. 2003. Synaptic plasticity in spinal lamina I projection neurons that mediate hyperalgesia. Science 299:1237–40 Novikov L, Novikova L, Mosahebi A, et al. 2002. A novel biodegradable implant for neuronal rescue and regeneration after spinal cord injury. Biomaterials 23:3369– 76 Woerly S. 2000. Restorative surgery of the central nervous system by means of tissue engineering using NeuroGel implants. Neurosurg. Rev. 23:59–77 Tillakaratne N, Hoang T, de Leon R, et al. 2002. Use-dependent modulation of inhibitory capacity in the feline lumbar spinal cord. J. Neurosci. 22:3130– 43 Capaday C. 2002. The special nature of human walking and its neural control. Trends Neurosci. 25:370–76 Chen X, Carp J, Chen L, Wolpaw J. 2002. Corticospinal tract transection prevents operantly conditioned H-reflex increase in rats. Exp. Brain Res. 144:88–94 Blight A. 2000. Animal models of spinal cord injury. Top. Spinal Cord Inj. Rehabil. 6:1–13 Benowitz L, Goldberg D, Madsen J, et al. 1999. Inosine stimulates extensive axon collateral growth in the rat corticospinal tract after injury. Proc. Natl. Acad. Sci. USA 96:13486–90 Ramer M, Harper G, Bradbury E. 2000. Progress in spinal cord research. Spinal Cord 38:449–72 Bracken M, Shepard M, Holford T, et al. 1998. Results of the third acute spinal cord injury randomized controlled trial. J. Neurosurg. 89:699–706 Geisler F, Coleman W, Grieco G, et al. 2001. The Sygen multicenter acute spinal cord injury study. Spine 26:S87–S98

12 Dec 2003

14:3

AR

AR206-ME55-17.tex

AR206-ME55-17.sgm

LaTeX2e(2002/01/18) P1: GBC 10.1146/annurev.med.55.091902.103753

Annu. Rev. Med. 2004. 55:283–301 doi: 10.1146/annurev.med.55.091902.103753 c 2004 by Annual Reviews. All rights reserved Copyright ° First published online as a Review in Advance on Oct. 6, 2003

CLINICAL MANAGEMENT OF TUBERCULOSIS IN THE CONTEXT OF HIV INFECTION

Annu. Rev. Med. 2004.55:283-301. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Bouke C. de Jong,1 Dennis M. Israelski,1,2 Elizabeth L. Corbett,3 and Peter M. Small1,4 1

Division of Infectious Diseases and Geographic Medicine, Stanford University, Stanford, California 94305 2 Division of Infectious Diseases, San Mateo Medical Center, San Mateo, California 94403 3 Department of Infectious and Tropical Diseases, London School of Hygiene and Tropical Medicine, London, United Kingdom 4 Bill and Melinda Gates Foundation, Seattle, Washington 98102; email: [email protected]; [email protected]; [email protected]; [email protected]

Key Words treatment, drug interactions, immune reconstitution syndrome ■ Abstract Globally, the HIV and tuberculosis epidemics are stoking each other, creating a public health crisis of enormous proportions. At the level of individuals, contemporaneous infection with M. tuberculosis and HIV poses great challenges to clinical management. This chapter provides an overview of active and latent tuberculosis treatment in HIV-infected and -uninfected individuals. The discussion focuses on medication issues, including interactions between antitubercular drugs, antiretroviral drugs, and medicines used for opportunistic infections and treatment in the face of comorbidities. Clinical questions specific to coinfection are discussed, including duration and timing initiation of therapy and immune reconstitution. Most of the data presented were generated in industrialized settings and are presented to assist patient management in such settings. However, given the disproportionate amount of TB/HIV in lessdeveloped nations and the increasing availability of antiretroviral therapy in resourcelimited settings, the issues presented will become increasingly relevant globally.

INTRODUCTION The convergence of the global epidemics of tuberculosis and human immunodeficiency virus (HIV) is one of the most significant challenges to global public health. It is estimated that one third of the world population is infected with Mycobacterium tuberculosis, more than 8 million of whom develop active disease each year (1). An estimated 5 million people were infected with HIV in 2002, amounting to a global prevalence of 42 million people (2). In combination, the toll is devastating. In the year 2000, an estimated 11 million people were coinfected with HIV and M. tuberculosis; in that same year, there were half a million new cases of 0066-4219/04/0218-0283$14.00

283

12 Dec 2003

14:3

Annu. Rev. Med. 2004.55:283-301. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

284

AR

AR206-ME55-17.tex

AR206-ME55-17.sgm

LaTeX2e(2002/01/18)

P1: GBC

DE JONG ET AL.

tuberculosis disease (“active” tuberculosis) associated with HIV (1). Three quarters of the world’s HIV-infected individuals live in sub-Saharan Africa, where annual tuberculosis case notification rates have risen up to fourfold since the mid1980s, reaching peaks of more than 400 cases per 100,000 population. In some areas of sub-Saharan Africa, over two thirds of patients with active tuberculosis are coinfected with HIV (3). Similar though less catastrophic effects were observed in the developed world, where the HIV epidemic contributed to a temporary reversal of the downward trend in tuberculosis rates in the United States in the 1980s and early 1990s. Relative to the expected continued decline, there were 15,000–25,000 excess cases of tuberculosis attributed to HIV (4), disproportionately afflicting minority populations and intravenous drug users (5, 6). Through increased rates of case detection, contact investigations, and increased use of directly observed therapy, a decrease in incidence of tuberculosis was re-established in 1993. Today, despite an absolute increase of the number of people living with HIV, the widespread use of antiretroviral therapy by acquired immunodeficiency syndrome (AIDS) patients has further decreased the incidence of tuberculosis (7). Conversely, the estimated rate of HIV coinfection among reported tuberculosis cases in the United States decreased from ∼15% to 9% between 1993 and 2000 (8). Overall, these favorable trends in tuberculosis incidence and the prevalence of coinfection continue, although rates of decline are lower in high-risk populations, including foreign-born persons, intravenous drug users, homeless persons, African Americans, and Hispanics (9). Where available, antiretroviral therapy has greatly improved the prognosis of HIV and tuberculosis coinfected individuals. However, drug-drug interactions and paradoxical reactions caused by immune reconstitution add considerable complexity to the management of coinfected patients. The aim of this article is to provide a comprehensive and up-to-date review of the treatment of patients coinfected with HIV and tuberculosis. This article focuses on experiences in the developed world, which can serve as a scientific foundation for devising management strategies for tuberculosis/HIV in resource-poor settings.

HIV AND TUBERCULOSIS: A DEADLY INTERACTION The interaction between HIV and tuberculosis is synergistic, each increasing the pathogenicity of the other. M. tuberculosis causes disease at any stage of HIV infection, exacerbates the course of HIV infection, and is the leading cause of HIV-related morbidity and mortality in less-developed countries. Immune activation by M. tuberculosis increases HIV plasma viremia and quasispecies diversity in coinfected people, potentially increasing both the rate at which HIV disease progresses and infectivity to others (10). In turn, HIV affects the course of tuberculosis in several ways. HIV promotes progression to active tuberculosis in people with either recently acquired or latent tuberculosis infection. HIV is the most common identifiable risk factor for progression from latent tuberculosis infection to active disease. In contrast to

12 Dec 2003

14:3

AR

AR206-ME55-17.tex

AR206-ME55-17.sgm

LaTeX2e(2002/01/18)

TUBERCULOSIS/HIV COINFECTION

P1: GBC

285

Annu. Rev. Med. 2004.55:283-301. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

HIV-uninfected people, for whom the lifetime risk of progressing to tuberculosis is 10%–20%, up to 10% of coinfected individuals develop active tuberculosis each year (1). HIV also increases the rate of recurrence of tuberculosis following treatment completion or cure, mainly because of an increased risk of disease following reinfection (11, 12). Dramatic point source outbreaks of tuberculosis have been documented in settings where HIV-infected persons congregate in the United States (13, 14). Given the ubiquity of such circumstances in developing countries, it is likely that this phenomenon contributes significantly to the global tuberculosis epidemic.

TREATMENT OF LATENT TUBERCULOSIS IN COINFECTED PATIENTS Prior to initiating treatment of latent tuberculosis infection, it is essential to exclude the diagnosis of active disease with a thorough history, physical examination, and chest radiography. Failure to do so may result in administering a single agent (isoniazid) to a person with active disease, which carries a great risk of creating drug resistance. Because of their high risk of progression to active disease, HIV-infected patients with tuberculin skin-test–induced induration of ≥5 mm should receive treatment for latent tuberculosis infection (15). The use of isoniazid in tuberculin skin-test– positive, HIV-uninfected individuals decreases their likelihood of developing tuberculosis by ∼90% (16). The impact that has been demonstrated in HIV-infected persons is somewhat smaller (35%–76%) (17), presumably due to exogenous reinfection after eradication of latent infection. This presumption is supported by studies demonstrating that after treatment of latent tuberculosis infection with isoniazid, only the persistence of risk factors for exposure to tuberculosis is associated with development of disease (18). In addition, HIV-infected individuals who are close contacts of someone with untreated pulmonary tuberculosis should themselves be treated for presumptive latent tuberculosis infection, regardless of their tuberculin skin-test results or history of treatment for latent tuberculosis infection (16). The therapeutic options for latent tuberculosis infection are similar to those for HIV-uninfected persons (19): (a) isoniazid for 9 months (preferred) or (b) rifampin or rifabutin for 4 months. Isoniazid can be given once daily or twice weekly using directly observed therapy, whereas the other regimen is recommended only for daily therapy. The advantages of isoniazid are the greater abundance of efficacy data, the lack of drugdrug interactions with antiretroviral medications, and the lower cost. In addition, the consequences of inducing resistance by using rifampin or rifabutin alone as therapy for latent infection are an order more serious both to the patient and the community at large than is the case with isoniazid, as loss of a rifamycin from the regimen prolongs treatment of active tuberculosis to 18–24 months. Therefore,

12 Dec 2003

14:3

Annu. Rev. Med. 2004.55:283-301. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

286

AR

AR206-ME55-17.tex

AR206-ME55-17.sgm

LaTeX2e(2002/01/18)

P1: GBC

DE JONG ET AL.

rifampin or rifabutin alone is unlikely to be widely used as single agent for latent tuberculosis infection and recommended only in cases of isoniazid resistance or intolerance. The CDC recently recommended that physicians stop offering the regimen consisting of rifampin or rifabutin with pyrazinamide for 2 months (19a). Although effective, this regimen was found to be associated with unacceptably high rates of liver toxicity (8% versus 1% with isoniazid) (20), including several deaths (19). When treating HIV-infected individuals exposed to multi-drugresistant tuberculosis, two drugs with activity to the isolate (e.g., pyrazinamide and a fluoroquinolone or pyrazinamide/ethambutol) should be used for a minimum of 12 months (16). Antiretroviral therapy alone also appears to be highly effective at reducing the risk of progression to active tuberculosis in latently infected individuals, perhaps by as much as 80%–92% (17, 21, 22), although this has not been the subject of any clinical trial. However, patients on or about to start antiretroviral therapy should still be offered treatment for latent tuberculosis if skin-test positive because there is a risk of antiretroviral therapy failure in the long term.

TREATMENT OF ACTIVE TUBERCULOSIS IN COINFECTED PATIENTS Whereas HIV treatment can be delayed in most patients, tuberculosis treatment does not offer the same luxury of time. The rapidly progressive nature of tuberculosis in HIV-infected patients requires that empirical antituberculous therapy be started for patients at high risk for tuberculosis or whose conditions are deteriorating, after cultures of sputum, blood, and urine have been collected. The joint tuberculosis treatment guidelines from the American Thoracic Society, the Centers for Disease Control, and the Infectious Diseases Society of America (ATS/CDC/IDSA) were recently updated (23). Treatment recommendations for tuberculosis are the same for HIV-infected and HIV-uninfected patients, with two exceptions regarding intermittent therapy, as addressed below. The two main measures of effectiveness for tuberculosis regimens are the treatment failure rate and the relapse rate. Both are low for the standard 6-month regimen consisting of isoniazid, rifampin, and pyrazinamide for 2 months followed by isoniazid and rifampin for another 4 months. Ethambutol is also given during the first 2 months to increase the efficacy of the regimen against primary-drug-resistant strains of M. tuberculosis. Any deviation from this regimen is likely to be associated with higher relapse or treatment failure rates unless drugs are given for a prolonged period. Studies comparing treatment outcomes have shown that clinical response rates, time to culture conversion from positive to negative, and treatment failure rates differ little between HIV-infected and -uninfected patients treated with standard regimens (23–25), although the risk of death during treatment is increased by HIV coinfection. Relapse rates in most studies (26) are 5% or less for disease caused by pansensitive M. tuberculosis, unless patients have positive cultures beyond

12 Dec 2003

14:3

AR

AR206-ME55-17.tex

AR206-ME55-17.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:283-301. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

TUBERCULOSIS/HIV COINFECTION

P1: GBC

287

2 months or are noncompliant with their antimycobacterial regimen (27). Similarly, studies in HIV-uninfected patients reported high relapse rates in patients with cavitation and positive cultures at the completion of 2 months of therapy (28). In response to these findings, the new ATS/CDC/IDSA guidelines recommend, regardless of HIV status, prolonging therapy in patients with cavitation noted on initial chest radiograph and a delayed response, defined as positive cultures at completion of 2 months of therapy. These patients should receive rifampin and isoniazid for an additional 3 months for a total of 9 months of treatment. The World Health Organization (WHO) guidelines, however, do not distinguish between cavitary and noncavitary disease and recommend 6 months of therapy in all cases. There is a paucity of specific data about therapy of extrapulmonary tuberculosis in HIV-infected individuals. However, in general, a standard 6-month course of antituberculous therapy is adequate. Exceptions to this include central nervous system disease (including military tuberculosis with cerebrospinal-fluid evidence of meningitis), which should be treated for 9–12 months, and bone and joint disease, for which treatment can be prolonged to 9 months. Regardless of HIV status, adjuvant corticosteroids are indicated for central nervous system disease and pericardial disease (23). For pleural disease, there is evidence of a more rapid clinical and radiographic response with steroids in HIV-uninfected patients (29), although a Cochrane review found insufficient evidence to recommend steroids in that setting (30). HIV-infected tuberculosis patients are more likely to relapse with rifamycinresistant organisms than HIV-uninfected patients, especially when they are given highly intermittent regimens (31, 32). Other risk factors for rifamycin-monoresistant tuberculosis are nonadherence, severe immune suppression, and a positive acid fast smear (33). The loss of a rifamycin from the treatment regimen is likely to delay sputum conversion, prolong the duration of therapy, and, in endemic areas, result in higher relapse and mortality rates (34, 35). Thus, the ATS/CDC/IDSA guidelines recommend two differences in the treatment of HIV-infected tuberculosis patients as compared with HIV-uninfected tuberculosis patients: Once-weekly isoniazid-rifapentine in the continuation phase should not be used in HIV-infected patients, and twice-weekly isoniazid-rifampin or isoniazid-rifabutin should not be used for patients with CD4 counts less than 100 cells/µl (23). Therapy with isoniazid should be supplemented with 25 mg of pyridoxine (B6) per day, or 50 mg twice weekly with intermittent antituberculous therapy (36), a regimen similar to that for HIV-uninfected tuberculosis patients at risk for peripheral neuropathy. Regardless of HIV infection, patients receiving ethambutol should be questioned at monthly intervals regarding visual disturbances; those receiving doses higher than the recommended 15–20 mg/kg, or taking ethambutol for more than 2 months, should have monthly testing of visual acuity and color vision (23). Some studies found the incidence of serious side effects from antituberculous medications among HIV-infected individuals to be similar to rates in HIVuninfected subjects (26, 32), whereas others found significantly increased rates of

12 Dec 2003

14:3

Annu. Rev. Med. 2004.55:283-301. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

288

AR

AR206-ME55-17.tex

AR206-ME55-17.sgm

LaTeX2e(2002/01/18)

P1: GBC

DE JONG ET AL.

side effects, namely rifampin-associated rash (37) and hepatotoxicity from antituberculous combination therapy (38, 39). Regardless of whether they have tuberculosis, HIV-infected persons should receive trimethoprim-sulfamethoxazole for prophylaxis of Pneumocystis carinii pneumonia if their CD4 count is less than 200 cells/µl. Neither HIV infection nor diarrhea alter pharmacokinetic characteristics of standard antituberculous drugs (40). Directly observed therapy should be used in all patients with HIV-related tuberculosis. If the HIV care provider and tuberculosis care provider are not the same person, communication between these providers should occur regularly throughout the course of treatment. Recommendations for the management of treatment failure, relapse, and multidrug-resistant tuberculosis are similar regardless of whether patients are coinfected with HIV, and these topics are covered in the ATS/CDC/IDSA guidelines (23).

TIMING OF ANTIRETROVIRAL TREATMENT IN PATIENTS WITH ACTIVE TUBERCULOSIS Most coinfected patients are not receiving antiretroviral therapy when they develop tuberculosis, and some are first tested for HIV at the time they are diagnosed with tuberculosis (41). Delaying antiretroviral therapy until after completion of tuberculosis treatment has several advantages. Standard regimens can be used for both diseases, and potential drug-drug interaction is avoided. In patients who are contemporaneously diagnosed, the period of antituberculous therapy can be used to lay a foundation for better adherence to antiretroviral therapy. Furthermore, paradoxical reactions (described below, e.g., new fevers in a patient who is otherwise responding to antituberculous therapy) may be less common as the time interval between treatment of active tuberculosis and initiation of antiretroviral therapy is increased. A final advantage of delaying antiretroviral therapy is that 90% of antituberculoustherapy–induced hepatitis among HIV-infected individuals occurs in the first 2– 4 weeks, as do common side effects such as rash and gastric disturbances (42, 43). The management of these common side effects before the introduction of antiretroviral therapy is crucial to the success of treatment. The current recommendation for starting antiretroviral treatment in patients without tuberculosis is to wait until the CD4 cell count drops to 200–350 cells/µl (44). However, the high mortality rates in the first few months of tuberculosis treatment observed in less-developed countries and in the pre–antiretroviral-therapy era in the United States caution against routinely withholding antiretroviral therapy until after tuberculosis treatment has been completed. As a compromise between ease of management and the risks of leaving HIV disease untreated, the WHO recommends starting antiretroviral therapy as soon as tuberculosis therapy is tolerated in patients with a CD4 count below 50 cells/µl, waiting 2 months into tuberculosis therapy in those with CD4 counts in the 50–200 cells/µl range, and

12 Dec 2003

14:3

AR

AR206-ME55-17.tex

AR206-ME55-17.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:283-301. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

TUBERCULOSIS/HIV COINFECTION

P1: GBC

289

waiting until the end of tuberculosis therapy for those with CD4 counts over 200 cells/µl (42). The ATS/CDC/IDSA guidelines suggest an individualized decision on timing of antiretroviral therapy in patients with CD4 counts below 350 cells/µl and advise delaying antiretroviral therapy for 4–8 weeks if possible (23). Specific treatment of M. tuberculosis, however, can confer immunological benefit without antiretroviral therapy; a likely mechanism to explain the observed improvements in CD4 counts and plasma viral load (43) is that, as tuberculosis is effectively treated, a decrease in immune activation leads to downregulation of viral replication (43). For the minority of patients who are on antiretroviral therapy when they are diagnosed with tuberculosis, the antiretroviral regimen will need to be adjusted for compatibility with use of a rifamycin (see Table 1).

PHARMACOKINETIC INTERACTIONS IN COINFECTED PATIENTS Several important pharmacokinetic interactions must be considered when starting antiretroviral therapy in patients also on treatment for tuberculosis with a regimen that includes rifamycins. This review provides an update on guidelines published by the CDC (15, 45).

Cytochrome P450-3A Because rifamycins induce the cytochrome P450-3A (CYP3A) system in the liver and intestinal wall, they increase the metabolism of both protease inhibitors and non-nucleoside reverse transcriptase inhibitors. This effect is weaker with rifabutin than with rifampin. Conversely, some antiretrovirals diminish CYP3A activity, which can significantly increase the serum concentrations of rifabutin. Rifampin, on the other hand, is metabolized through deacetylation and is not itself affected by the P450 system (41). DOSE ADJUSTMENTS As a consequence of these drug-induced alterations in cytochrome P450, dose adjustments are needed when coadministering rifamycins and some antiretroviral drugs. Rifabutin at 150–300 mg/day can safely and effectively be exchanged for rifampin in patients who receive antiretroviral therapy concomitantly (46, 47). When a tuberculosis patient receiving rifampin needs to be started on a protease inhibitor or a non-nucleoside reverse transcriptase inhibitor that is compatible with the use of rifabutin, rifampin should be changed to rifabutin 2 weeks prior to the start of antiretroviral therapy to allow reduction of the enzymeinducing activity of rifampin (23). At the time of writing, no studies have been published on interactions between the rifamycins and the newer antiretrovirals, such as tenofovir, atazanavir, and enfuvirtide. Enfuvirtide is the first drug of the new class of HIV-fusion inhibitors to be approved by the U.S. Food and Drug Administration (FDA). It does not inhibit or induce the metabolism of CYP3A and is therefore unlikely to show significant interaction with the rifamycins.

12 Dec 2003

14:3

290

AR

AR206-ME55-17.tex

AR206-ME55-17.sgm

LaTeX2e(2002/01/18)

P1: GBC

DE JONG ET AL.

TABLE 1 Dose adjustments between antiretroviral therapy and rifamycins (modified from Reference 41) Rifampin (normal daily dose 600 mg) NRTIs

Rifabutin∗ (daily, unless mentioned; normal daily dose 300 mg)

Caution when using with triple nucleoside therapy

Normal rifabutin dose

Efavirenz

Efavirenz dose at 600–800 mg

Increase rifabutin dose to 450–600 mg

Nevirapine

Probably can be used at normal nevirapine dose

Normal dose of each

NR

NR

Amprenavir

NR

Lower rifabutin dose to 150 mg

Atazanavir

No data

Lower rifabutin dose to 150 mg three times weekly

Indinavir

NR

Lower rifabutin dose to 150 mg Increase indinavir dose to 1000 mg q8

Nelfinavir

NR

Lower rifabutin dose to 150 mg

Saquinavir

Can use with low-dose ritonavir “boost”

Normal rifabutin dose

Ritonavir

Use only with full-dose ritonavir

Lower rifabutin dose to 150 mg two times weekly

Lopinavir/ritonavir

Probably can be used with lopinavir/ritonavir dose increase to 400/400 mg or 800/200 mg

Lower rifabutin dose to 150 mg three times weekly

Annu. Rev. Med. 2004.55:283-301. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

NNRTIs

Normal efavirenz dose

Delavirdine Protease inhibitors

∗

∗

Consider checking rifabutin levels when using concomitantly with protease inhibitors.

Abbreviations: NRTIs, nucleoside reverse transcriptase inhibitors; NNRTIs, non-nucleoside reverse transcriptase inhibitors; NR, not recommended.

RIFAMPIN WITH NUCLEOSIDE REVERSE TRANSCRIPTASE INHIBITORS Rifampin can be used with nucleoside reverse transcriptase inhibitors without dose adjustments, but it somewhat decreases levels of zidovudine and possibly of abacavir. Experts therefore do not agree on its use in combination with the triple nucleoside therapy zidovudine-lamivudine-abacavir (41, 45). RIFAMPIN WITH NON-NUCLEOSIDE REVERSE TRANSCRIPTASE INHIBITORS Rifampin can be combined with efavirenz, and some experts recommend increasing the

12 Dec 2003

14:3

AR

AR206-ME55-17.tex

AR206-ME55-17.sgm

LaTeX2e(2002/01/18)

TUBERCULOSIS/HIV COINFECTION

P1: GBC

291

Annu. Rev. Med. 2004.55:283-301. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

dose from 600 to 800 mg to compensate for a slight decrease in serum efavirenz concentration (45). Rifampin reduces nevirapine levels by 31%, although trough levels still appear adequate without dose adjustments owing to nevirapine’s high therapeutic index (48). There is, however, theoretical concern about combined hepatotoxicity of nevirapine and tuberculosis medications (49). RIFAMPIN WITH PROTEASE INHIBITORS Rifampin decreases levels of saquinavir, indinavir, nelfinavir, amprenavir, and lopinavir by >75% (reviewed in Reference 41), which might result in selection of resistant HIV quasispecies. Rifampin can be administered with full-dose ritonavir (400–600 mg twice daily), and limited data support using low-dose ritonavir “boosting” in combination with saquinavir in patients taking rifampin (50). Similarly, a study in healthy volunteers indicated that increased doses of lopinavir/ritonavir (i.e., from the usual 400/100 mg to 400/400 mg and possibly 800/200 mg, twice daily), in conjunction with therapeutic drug monitoring, may allow for concomitant use of rifampin 600 mg once daily (50a). Ritonavir in this role counteracts the CYP3A effects of rifampin, which makes it a theoretically appealing combination with other protease inhibitors as well. Unfortunately, the effect is not consistent with different protease inhibitors, and the level of, for instance, indinavir is still lowered by almost 90% in the presence of ritonavir and rifampin (51). Rifabutin, which is not generally available in resource-limited settings, is much easier to administer in combination with antiretrovirals, as it is a much less potent inducer of the cytochrome P450 system. RIFABUTIN WITH NUCLEOSIDE REVERSE TRANSCRIPTASE INHIBITORS Rifabutin does not interact significantly with nucleoside reverse transcriptase inhibitors. RIFABUTIN WITH NON-NUCLEOSIDE REVERSE TRANSCRIPTASE INHIBITORS Efavirenz reduces levels of rifabutin by almost 40%, and the dose of rifabutin should be increased to 450–600 mg/day (52). Nevirapine and rifabutin can be coadministered at normal doses. RIFABUTIN WITH PROTEASE INHIBITORS Protease inhibitors increase rifabutin levels, resulting in higher rates of adverse effects including arthralgias, uveitis, and leukopenia. With most protease inhibitors, therefore, the rifabutin dose should be lowered to 150 mg/day, or even to 150 mg three times weekly when using atazanavir or lopinavir/ritonavir and 150 mg twice weekly when combined with full-dose ritonavir (50a, 52a). Therapeutic drug monitoring of rifabutin levels can be a valuable tool in patients taking protease inhibitors and in those taking three or more interactive drugs, in the absence of pharmakokinetic data on these complex interactions (53). Therapeutic drug monitoring of non-nucleoside reverse transcriptase inhibitors and protease inhibitors has theoretical appeal but remains unvalidated in clinical practice.

12 Dec 2003

14:3

292

AR

AR206-ME55-17.tex

AR206-ME55-17.sgm

LaTeX2e(2002/01/18)

P1: GBC

DE JONG ET AL. SPECIAL CONSIDERATIONS OF PHARMACOKINETIC INTERACTIONS These same effects on the cytochrome system have consequences for other drugs that are frequently administered to HIV-infected patients. For example, rifampin decreases methadone levels by ∼75%, whereas rifabutin has a less dramatic effect (54).

Annu. Rev. Med. 2004.55:283-301. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Patients on Prophylaxis for Other Opportunistic Infections Rifampin decreases levels of trimethoprim-sulfamethoxazole (cotrimoxazole), which is the most widely used prophylaxis in HIV-infected patients for the prevention of both P. carinii pneumonia and toxoplasmosis (55). At lower dosing levels of cotrimoxazole, this interaction was found to reduce its efficacy in preventing toxoplasmic encephalitis (56). In the absence of prospective data, it is probably reasonable to give double-strength cotrimoxazole once daily as opposed to three times a week, when it is combined with rifampin. Rifampin has been shown to decrease dapsone levels in HIV-infected patients using dapsone at 100 mg twice weekly as prophylaxis for P. carinii (57), but it is unknown whether this would significantly affect either drug levels or efficacy when taking the usual dose of 100 mg once daily. The combination of rifabutin and daily azithromycin did not result in significant drug interactions, but it was associated with high rates of neutropenia [66% in one series (58) and 100% in a smaller series that also included clarithromycin (59)] and gastrointestinal side effects. It is unknown whether weekly azithromycin in combination with rifabutin has the same effect. Clarithromycin doubles the area under the curve (AUC) of rifabutin, which conversely almost halves the AUC of clarithromycin (60). In the absence of clear guidelines, it is prudent to monitor the neutrophil count regularly. Rifampin can dramatically decrease levels of azoles such as itraconazole and ketoconazole.

IMMUNE RECONSTITUTION SYNDROME Occasionally, patients being treated for tuberculosis may experience a temporary worsening of symptoms or development of new signs or symptoms of tuberculosis disease. This phenomenon is known as immune reconstitution syndrome or a paradoxical reaction. Immune reconstitution syndrome is thought to represent an enhanced immunologic response to mycobacterial antigens during the course of treatment, resulting in a stronger inflammatory response at sites of tuberculosis infection. It is not associated with changes in M. tuberculosis bacteriology (i.e., no change from negative to positive cultures). Immune reconstitution syndrome can occur in HIV-uninfected tuberculosis patients but is probably more common in coinfected patients. It typically appears ∼6 weeks after the start of antiretroviral therapy in a patient receiving concurrent treatment for active tuberculosis (61). Antiretroviral therapy has been shown to improve antigen-specific CD4+ responses to M. tuberculosis (62), although not to the levels seen in healthy control subjects (63).

12 Dec 2003

14:3

AR

AR206-ME55-17.tex

AR206-ME55-17.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:283-301. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

TUBERCULOSIS/HIV COINFECTION

P1: GBC

293

The incidence of immune reconstitution syndrome in coinfected patients ranges from 7%–36%. In a prospective study, immune reconstitution syndrome occurred in 36% of patients receiving antiretroviral therapy, compared with 7% in HIVinfected patients not receiving antiretroviral therapy and 2% in HIV-uninfected patients (64). In this study, 6 of 7 patients with immune reconstitution syndrome converted their previously negative tuberculin skin test at a median of 6.5 weeks after the start of antiretroviral therapy. Common symptoms and signs of immune reconstitution syndrome include high fevers, lymphadenopathy, worsening of chest radiographic findings, and worsening of original tuberculosis lesions. Less frequent manifestations include pleural effusions, psoas abscesses, central nervous system tuberculomas, and epididymitis and orchitis. We found only one case of meningitis (65). The initiation of antiretroviral therapy can also unmask previously undiagnosed infections by augmenting the inflammatory response. Thus, the diagnosis of immune reconstitution syndrome should be made only after a thorough evaluation has excluded other etiologies. Immune reconstitution syndrome can be brief or prolonged, with multiple recurrences, but no deaths have been clearly associated with it (66). In general, antiretroviral therapy should not be interrupted if immune reconstitution syndrome occurs. Nonsteroidal inflammatory drugs may provide some relief, but some patients have required the use of corticosteroids (in addition to tuberculosis treatment) to treat these reactions. Use of steroids in this setting has not been associated with tuberculosis treatment failure (66), but data are limited and the decision to use corticosteroids must be made on a case-by-case basis. Indications for use of concomitant corticosteroids for immune reconstitution syndrome include severe hypoxemia, airway obstruction, neurologic impairment, or possibly enlarged painful lymph nodes. The ATS/CDC/IDSA guidelines recommend, based on expert opinion, use of prednisone at 1 mg/kg per day with a gradual reduction after 1–2 weeks (23). Manifestations of immune reconstitution syndrome in patients being treated for coinfection will undoubtedly receive wider attention as antiretroviral therapy becomes increasingly available to resource-poor areas where tuberculosis is highly endemic.

TREATMENT CONSIDERATIONS IN SPECIAL POPULATIONS Patients with Underlying Liver Disease and Latent Tuberculosis Infection In a prospective study, isoniazid was given for treatment of latent infection to 138 chronically hepatitis C–infected individuals with normal baseline hepatic transaminase levels, 25% of whom were HIV positive as well (67). Thirty-two patients (22%) developed transaminase value elevations to >3 times the upper limit of normal, and isoniazid was withdrawn in 11 patients (8%). Transaminase

12 Dec 2003

14:3

Annu. Rev. Med. 2004.55:283-301. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

294

AR

AR206-ME55-17.tex

AR206-ME55-17.sgm

LaTeX2e(2002/01/18)

P1: GBC

DE JONG ET AL.

elevation was only significantly associated with concurrent alcohol use, not with HIV or hepatitis B infection. The risks of transaminase elevation and discontinuation were within the range reported for populations with a low prevalence of hepatitis C virus. Similarly, a series of patients with liver disease awaiting transplantation who had positive tuberculin skin-test results tolerated isoniazid well (68). No data are available on treating these patients with rifampin/pyrazinamide, although data in individuals without liver disease suggest that this regimen has more hepatotoxic effects than isoniazid. In summary, the majority of individuals with latent tuberculosis infection and liver disease will benefit from treatment for latent tuberculosis infection, and it is impossible to predict which patients will develop hepatotoxicity. Thus, it is prudent to initiate therapy with isoniazid for coinfected individuals regardless of liver abnormalities. However, patients with liver disease merit close clinical and laboratory monitoring, and medications should be promptly discontinued if adverse effects occur.

Patients with Underlying Liver Disease and Active Tuberculosis Infection A study of tuberculosis patients with hepatitis C and/or HIV infection demonstrated a relative risk of hepatotoxicity of five- and fourfold, respectively, for patients with either hepatitis C or HIV. In contrast, those infected with both hepatitis C and HIV faced a 14-fold increase in risk of hepatoxicity. Possible “liver sparing” regimens have been described for use in patients with overt liver failure. These exclude isoniazid, rifampin, and pyrazinamide and typically include a fluoroquinolone, ethambutol, streptomycin, and/or cycloserine, and they last for a minimum of 18 months (23). Such a regimen can also be used in the patient who develops severe hepatotoxicity on standard tuberculosis drugs (69). However, there are limited data on these regimens. Patients with underlying liver disease can often be started on the usual four drugs with close laboratory and clinical follow-up; however, these patients do have an increased rate of drug-induced hepatitis, and therapy should be held or changed to another regimen if serum aminotransferase levels rise above five times the upper limit of normal, or three times the upper limit of normal in the presence of symptoms.

Patients with Underlying Neuropathy Neuropathy is very common in people living with HIV. Distal symmetrical polyneuropathy is more common as HIV disease progresses (70). Data on the effect of isoniazid on HIV-infected patients with pre-existing neuropathy are scarce. A retrospective review of patients who had used antiretrovirals with and without isoniazid found higher rates of incident neuropathy in those on d4T and isoniazid than in those taking d4T alone (55% versus 11%). Pyridoxine intake was not mentioned in this report (71). Patients at risk for neuropathy should preferably still receive isoniazid, with 25 mg of pyridoxine. The guidelines of the Bureau of Tuberculosis

12 Dec 2003

14:3

AR

AR206-ME55-17.tex

AR206-ME55-17.sgm

LaTeX2e(2002/01/18)

TUBERCULOSIS/HIV COINFECTION

P1: GBC

295

Control of the New York City Department of Health recommend discontinuing isoniazid if a patient develops peripheral neuropathy, while continuing pyridoxine (25 mg daily), until symptoms abate (36). Patients on hemodialysis typically receive pyridoxine as part of a multivitamin, and are at risk of accumulation of pyridoxine, which has itself been associated with neuropathy.

Annu. Rev. Med. 2004.55:283-301. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Safety of Antituberculous Medications in Pregnancy Pyrazinamide is not FDA approved for use in pregnancy because of a lack of data, although it is recommended for use in pregnancy by the WHO and has an excellent safety record outside of the United States. In HIV-infected pregnant patients, the benefits of pyrazinamide outweigh the risks (15), although it should probably not be given during the first trimester unless drug resistance to isoniazid or rifampin is strongly suspected (36). The incidence of drug-induced hepatitis, especially with isoniazid, is increased during and immediately after pregnancy. Thus, close monitoring of liver function is recommended. Streptomycin should not be used in pregnancy because of the potential for ototoxicity in the fetus (72).

Children In the absence of data, the American Academy of Pediatrics recommends that HIVinfected children should receive tuberculosis therapy for a minimum of 9 months (73). Ethambutol can safely be used, even in those too young to be evaluated by eye examinations, when susceptibility to isoniazid and rifampin is not (yet) known (15). For latent infection, isoniazid for 9 months or rifampin for 4 months can be used.

RESPONSE TO TREATMENT OF COINFECTED PATIENTS Published information on the clinical outcome of concurrent treatment for HIV and tuberculosis is limited, and results of randomized clinical trials have yet to be published. One study showed a good clinical response, including a decline in viral load, in 25 hospitalized patients (74). Another study of 188 patients, 45% of whom commenced antiretroviral therapy during tuberculosis treatment, found significant reductions in viral load, AIDS-defining illnesses, and mortality. Furthermore, CD4 counts increased and viral load decreased in patients not taking antiretroviral therapy. Adverse events occurred in 54% of patients on treatment for both HIV and tuberculosis, one third of whom changed or interrupted their treatment regimens. The major side effects were peripheral neuropathy (21%), rash (17%), and gastrointestinal upset (10%), the majority of which occurred in the first 2 months. Only 5% experienced a paradoxical reaction (43). A recent US Public Health Service study, presented at a 2003 conference on retroviruses and opportunistic infection, compared outcome of HIV-infected patients on twice-weekly rifabutin plus isoniazid with historical controls. At 12

12 Dec 2003

14:3

Annu. Rev. Med. 2004.55:283-301. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

296

AR

AR206-ME55-17.tex

AR206-ME55-17.sgm

LaTeX2e(2002/01/18)

P1: GBC

DE JONG ET AL.

months of follow-up, mortality among these patients was 5% compared to 15% in the preantiretroviral therapy era (75). At the same conference, two more studies were presented on the interaction between rifampin and efavirenz at a dose of 600 mg in combination with two nucleoside reverse transcriptase inhibitors. A prospective study from India compared HIV-infected patients with and without tuberculosis, who were all started on efavirenz. The CD4 response to antiretroviral therapy was at least as good in the tuberculosis patients, but 8% of them developed paradoxical worsening and 10% hepatotoxicity (76). In a study from Brazil, the success rate of tuberculosis treatment in patients receiving antiretroviral therapy was 84% at 24 months, and 2 of 49 patients relapsed. Paradoxical worsening was observed in 12% (77). These studies suggest that efavirenz at the normal dose of 600 mg is efficient when given concomitantly with rifampin, although the study subjects were on average smaller than US patients.

SUMMARY Treatment of latent tuberculosis infection is especially important in HIV-infected individuals, given the high rate of progression to active tuberculosis. The treatment of latent tuberculosis infection and active tuberculosis is similar in HIV-infected and -uninfected persons, except for some of the intermittent regimens. Concurrent antiretroviral therapy complicates the tuberculosis treatment regimen, but limited data suggest that combination therapy can be safe and improves outcome, and it should be strongly considered in persons with CD4 counts under 200 cells/µl. Paradoxical reactions occur in 7%–36% of coinfected patients on concurrent treatment and, though associated with worsened signs and/or symptoms, generally resolve without complications. The past decade has witnessed remarkable progress in the management of HIV and HIV-associated tuberculosis in the First World. Potent antiviral drugs have dramatically changed the outcome for HIV-infected individuals with and without tuberculosis coinfection. In combination with public health measures, these drugs have been associated with a gratifying reduction in morbidity and mortality from both conditions. To date, however, these advances have had little impact on those regions where they are most needed. The twin epidemics of AIDS and tuberculosis continue unabated in the majority of the world. It is currently estimated that only one third of the world has access to high-quality treatment for tuberculosis, and antiretroviral therapy is available to only a tiny minority of those in need (78). Fortunately, we are at an unprecedented time in global health; mechanisms such as the Global Fund Against Tuberculosis, AIDS and Malaria provide hope that resources for addressing this inequity may be forthcoming. Establishing the scientific and clinical base for the practical and cost-effective administration of the therapies described in this paper to those in need is one of the greatest public health challenges of our lifetime.

12 Dec 2003

14:3

AR

AR206-ME55-17.tex

AR206-ME55-17.sgm

LaTeX2e(2002/01/18)

TUBERCULOSIS/HIV COINFECTION

P1: GBC

297

The Annual Review of Medicine is online at http://med.annualreviews.org

Annu. Rev. Med. 2004.55:283-301. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

LITERATURE CITED 1. Corbett EL, Watt CJ, Walker N, et al. 2003. The growing burden of tuberculosis: global trends and interactions with the HIV epidemic. Arch. Intern. Med. 163:1009–21 2. Joint United Nations Programme on HIV/AIDS (UNAIDS) WHO. 2002. AIDS epidemic update. Rep. 02.46E, UNAIDS/ WHO, Geneva 3. Dye C, Scheele S, Dolin P, et al. 1999. Consensus statement. Global burden of tuberculosis: estimated incidence, prevalence, and mortality by country. WHO Global Surveillance and Monitoring Project. JAMA 282:677–86 4. Cantwell MF, Snider DE Jr, Cauthen GM, et al. 1994. Epidemiology of tuberculosis in the United States, 1985 through 1992. JAMA 272:535–39 5. Chaisson RE, Schecter GF, Theuer CP, et al. 1987. Tuberculosis in patients with the acquired immunodeficiency syndrome. Clinical features, response to therapy, and survival. Am. Rev. Respir. Dis. 136:570–74 6. Shafer RW, Edlin BR. 1996. Tuberculosis in patients infected with human immunodeficiency virus: perspective on the past decade. Clin. Infect. Dis. 22:683–704 7. Kirk O, Gatell JM, Mocroft A, et al. 2000. Infections with Mycobacterium tuberculosis and Mycobacterium avium among HIV-infected patients after the introduction of highly active antiretroviral therapy. EuroSIDA Study Group JD. Am. J. Respir. Crit. Care Med. 162:865–72 8. CDC. 2002. Reported tuberculosis in the United States, 2001. Atlanta, GA: US Dep. Health Hum. Serv., CDC 9. Badri M, Wilson D, Wood R. 2002. Effect of highly active antiretroviral therapy on incidence of tuberculosis in South Africa: a cohort study. Lancet 359:2059–64 10. Collins KR, Quinones-Mateu ME, Toossi

11.

12.

13.

14.

15.

16.

17.

18.

19.

Z, et al. 2002. Impact of tuberculosis on HIV-1 replication, diversity, and disease progression. AIDS Rev. 4:165–76 Sonnenberg P, Murray J, Glynn JR, et al. 2001. HIV-1 and recurrence, relapse, and reinfection of tuberculosis after cure: a cohort study in South African mineworkers. Lancet 358:1687–93 van Rie A, Warren R, Richardson M, et al. 1999. Exogenous reinfection as a cause of recurrent tuberculosis after curative treatment. N. Engl. J. Med. 341:1174–79 CDC. 1991. Nosocomial transmission of multidrug-resistant tuberculosis among HIV-infected persons—Florida and New York, 1988–1991. Morb. Mortal. Wkly. Rep. 40:585–91 Frieden TR, Fujiwara PI, Washko RM, et al. 1995. Tuberculosis in New York City—turning the tide. N. Engl. J. Med. 333:229–33 Centers for Disease Control and Prevention. 1998. Prevention and treatment of tuberculosis among patients infected with human immunodeficiency virus: principles of therapy and revised recommendations. MMWR Recomm. Rep. 47:1–58 American Thoracic Society. 2000. Targeted tuberculin testing and treatment of latent tuberculosis infection. MMWR Recomm. Rep. 49:1–51 Bucher HC, Griffith LE, Guyatt GH, et al. 1999. Isoniazid prophylaxis for tuberculosis in HIV infection: a meta-analysis of randomized controlled trials. AIDS 13:501–7 Casado JL, Moreno S, Fortun J, et al. 2002. Risk factors for development of tuberculosis after isoniazid chemoprophylaxis in human immunodeficiency virus-infected patients. Clin. Infect. Dis. 34:386–89 Jasmer RM, Nahid P, Hopewell PC. 2002. Clinical practice. Latent tuberculosis infection. N. Engl. J. Med. 347:1860–66

12 Dec 2003

14:3

Annu. Rev. Med. 2004.55:283-301. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

298

AR

AR206-ME55-17.tex

AR206-ME55-17.sgm

LaTeX2e(2002/01/18)

P1: GBC

DE JONG ET AL.

19a. Centers for Disease Control and Prevention (CDC), American Thoracic Society. 2003. Update: adverse event data and revised American Thoracic Society/CDC recommendations against the use of rifampin and pyrazinamide for treatment of latent tuberculosis infection—United States, 2003. Morb. Mortal. Wkly. Rep. 52(31):735–39 20. Jasmer RM, Saukkonen JJ, Blumberg HM, et al. 2002. Short-course rifampin and pyrazinamide compared with isoniazid for latent tuberculosis infection: a multicenter clinical trial. Ann. Intern. Med. 137:640–47 21. Girardi E, Antonucci G, Vanacore P, et al. 2000. Impact of combination antiretroviral therapy on the risk of tuberculosis among persons with HIV infection. AIDS 14:1985–91 22. Jones JL, Hanson DL, Dworkin MS, et al. 2000. HIV-associated tuberculosis in the era of highly active antiretroviral therapy. The Adult/Adolescent Spectrum of HIV Disease Group. Int. J. Tuberc. Lung Dis. 4:1026–31 23. Blumberg HM, Burman WJ, Chaisson RE, et al. 2003. American Thoracic Society/Centers for Disease Control and Prevention/Infectious Diseases Society of America: treatment of tuberculosis. Am. J. Respir. Crit. Care Med. 167:603–62 24. Murray J, Sonnenberg P, Shearer SC, et al. 1999. Human immunodeficiency virus and the outcome of treatment for new and recurrent pulmonary tuberculosis in African patients. Am. J. Respir. Crit. Care Med. 159:733–40 25. Ackah AN, Coulibaly D, Digbeu H, et al. 1995. Response to treatment, mortality, and CD4 lymphocyte counts in HIV-infected persons with tuberculosis in Abidjan, Cote d’Ivoire. Lancet 345:607–10 26. Chaisson RE, Clermont HC, Holt EA, et al. 1996. Six-month supervised intermittent tuberculosis therapy in Haitian patients with and without HIV infection. Am. J. Respir. Crit. Care Med. 154:1034–38

27. Driver CR, Munsiff SS, Li J, et al. 2001. Relapse in persons treated for drugsusceptible tuberculosis in a population with high coinfection with human immunodeficiency virus in New York City. Clin. Infect. Dis. 33:1762–69 28. Mitchison DA. 1993. Assessment of new sterilizing drugs for treating pulmonary tuberculosis by culture at 2 months. Am. Rev. Respir. Dis. 147:1062–63 29. Lee CH, Wang WJ, Lan RS, et al. 1988. Corticosteroids in the treatment of tuberculous pleurisy. A double-blind, placebo-controlled, randomized study. Chest 94:1256–59 30. Matchaba PT, Volmink J. 2000. Steroids for treating tuberculous pleurisy. Cochrane Database Syst. Rev.: CD001876 31. CDC. 2002. Acquired rifamycin resistance in persons with advanced HIV disease being treated for active tuberculosis with intermittent rifamycin-based regimens. Morb. Mortal Wkly. Rep. 51:214– 15 32. El-Sadr WM, Perlman DC, Matts JP, et al. 1998. Evaluation of an intensive intermittent-induction regimen and duration of short-course treatment for human immunodeficiency virus-related pulmonary tuberculosis. Terry Beirn Community Programs for Clinical Research on AIDS (CPCRA) and the AIDS Clinical Trials Group (ACTG). Clin. Infect. Dis. 26:1148–58 33. Munsiff SS, Joseph S, Ebrahimzadeh A, et al. 1997. Rifampin-monoresistant tuberculosis in New York City, 1993–1994. Clin. Infect. Dis. 25:1465–67 34. Hawken M, Nunn P, Gathua S, et al. 1993. Increased recurrence of tuberculosis in HIV-1-infected patients in Kenya. Lancet 342:332–37 35. Wallis RS, Helfand MS, Whalen CC, et al. 1996. Immune activation, allergic drug toxicity and mortality in HIV-positive tuberculosis. Tuberc. Lung Dis. 77:516– 23 36. Fujiwara PI. 1999. Clinical Policies and

12 Dec 2003

14:3

AR

AR206-ME55-17.tex

AR206-ME55-17.sgm

LaTeX2e(2002/01/18)

TUBERCULOSIS/HIV COINFECTION

37.

Annu. Rev. Med. 2004.55:283-301. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

38.

39.

40.

41.

42.

43.

44.

45.

Protocols. New York: Bur. Tuberc. Control, New York City Dep. Health Small PM, Schecter GF, Goodman PC, et al. 1991. Treatment of tuberculosis in patients with advanced human immunodeficiency virus infection. N. Engl. J. Med. 324:289–94 Ungo JR, Jones D, Ashkin D, et al. 1998. Antituberculosis drug-induced hepatotoxicity. The role of hepatitis C virus and the human immunodeficiency virus. Am. J. Respir. Crit. Care Med. 157:1871– 76 Pedral-Sampaio DB, Martins Netto E, Alcantara AP, et al. 1997. Use of standard therapy for tuberculosis is associated with increased adverse reactions in patients with HIV. Braz. J. Infect. Dis. 1:123– 30 Choudhri SH, Hawken M, Gathua S, et al. 1997. Pharmacokinetics of antimycobacterial drugs in patients with tuberculosis, AIDS, diarrhea. Clin. Infect. Dis. 25:104– 11 Barnes PF, Lakey DL, Burman WJ. 2002. Tuberculosis in patients with HIV infection. Infect. Dis. Clin. North Am. 16:107– 26 Pedral-Sampaio DB, Netto EM, Brites C, et al. 2002. Treating tuberculosis in AIDS patients: when to start and how long to keep giving drugs? AIDS 16:1845 Dean GL, Edwards SG, Ives NJ, et al. 2002. Treatment of tuberculosis in HIVinfected persons in the era of highly active antiretroviral therapy. AIDS 16:75– 83 Yeni PG, Hammer SM, Carpenter CC, et al. 2002. Antiretroviral treatment for adult HIV infection in 2002: updated recommendations of the International AIDS Society—USA Panel. JAMA 288:222– 35 Centers for Disease Control and Prevention. 2000. Updated guidelines for the use of rifabutin or rifampin for the treatment and prevention of tuberculosis among HIV-infected patients taking protease in-

46.

47.

48.

49.

50.

50a.

51.

52.

52a. 53.

54.

P1: GBC

299

hibitors or nonnucleoside reverse transcriptase inhibitors. Morb. Mortal Wkly. Rep. 49:185–89 Gonzalez-Montaner LJ, Natal S, Yongchaiyud P, et al. 1994. Rifabutin for the treatment of newly-diagnosed pulmonary tuberculosis: a multinational, randomized, comparative study versus rifampicin. Rifabutin Study Group. Tuberc. Lung Dis. 75:341–47 McGregor MM, Olliaro P, Wolmarans L, et al. 1996. Efficacy and safety of rifabutin in the treatment of patients with newly diagnosed pulmonary tuberculosis. Am. J. Respir. Crit. Care Med. 154:1462–67 Ribera E, Pou L, Lopez RM, et al. 2001. Pharmacokinetic interaction between nevirapine and rifampicin in HIVinfected patients with tuberculosis. J. Acquir. Immune. Defic. Syndr. 28:450–53 2002. Scaling up ART in resource-limited settings. Guidelines for a public health approach, World Health Organization, Geneva, Switz. Veldkamp AI, Hoetelmans RM, Beijnen JH, et al. 1999. Ritonavir enables combined therapy with rifampin and saquinavir. Clin. Infect. Dis. 29:1586 Cvetkovic RS, Goa KL. 2003. Lopinavir/ritonavir. A review of its use in the management of HIV infection. Drugs 63:769–802 Andrade A, Flexner C. 2003. Progress in pharmacology and drug interactions from the 10th CROI. Hopkins HIV Rep. 15:7,11 Lopez-Cortes LF, Ruiz-Valderas R, Viciana P, et al. 2002. Pharmacokinetic interactions between efavirenz and rifampicin in HIV-infected patients with tuberculosis. Clin. Pharmacokinet. 41:681– 90 Pham PA. 2003. Drug profile: atazanavir (reyataz, ATV). Hopkins HIV Rep. 15:4,5 Peloquin CA. 2002. Therapeutic drug monitoring in the treatment of tuberculosis. Drugs 62:2169–83 Brown LS, Sawyer RC, Li R, et al. 1996. Lack of a pharmacologic interaction

12 Dec 2003

14:3

300

55.

Annu. Rev. Med. 2004.55:283-301. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

56.

57.

58.

59.

60.

61.

62.

AR

AR206-ME55-17.tex

AR206-ME55-17.sgm

LaTeX2e(2002/01/18)

P1: GBC

DE JONG ET AL. between rifabutin and methadone in HIVinfected former injecting drug users. Drug Alcohol. Depend. 43:71–77 Ribera E, Pou L, Fernandez-Sola A, et al. 2001. Rifampin reduces concentrations of trimethoprim and sulfamethoxazole in serum in human immunodeficiency virus–infected patients. Antimicrob. Agents Chemother. 45:3238–41 Ribera E, Fernandez-Sola A, Juste C, et al. 1999. Comparison of high and low doses of trimethoprim-sulfamethoxazole for primary prevention of toxoplasmic encephalitis in human immunodeficiency virus–infected patients. Clin. Infect. Dis. 29:1461–66 Gatti G, Merighi M, Hossein J, et al. 1996. Population pharmacokinetics of dapsone administered biweekly to human immunodeficiency virus–infected patients. Antimicrob. Agents Chemother. 40:2743– 48 Hafner R, Bethel J, Standiford HC, et al. 2001. Tolerance and pharmacokinetic interactions of rifabutin and azithromycin. Antimicrob. Agents Chemother. 45:1572– 77 Apseloff G, Foulds G, LaBoy-Goral L, et al. 1998. Comparison of azithromycin and clarithromycin in their interactions with rifabutin in healthy volunteers. J. Clin. Pharmacol. 38:830–35 Hafner R, Bethel J, Power M, et al. 1998. Tolerance and pharmacokinetic interactions of rifabutin and clarithromycin in human immunodeficiency virus–infected volunteers. Antimicrob. Agents Chemother. 42:631–39 Orlovic D, Smego RA Jr. 2001. Paradoxical tuberculous reactions in HIV-infected patients. Int. J. Tuberc. Lung Dis. 5:370– 75 Wendland T, Furrer H, Vernazza PL, et al. 1999. HAART in HIV-infected patients: restoration of antigen-specific CD4 T-cell responses in vitro is correlated with CD4 memory T-cell reconstitution, whereas improvement in delayed type hypersensi-

63.

64.

65.

66.

67.

68.

69.

70.

71.

72.

tivity is related to a decrease in viraemia. AIDS 13:1857–62 Schluger NW, Perez D, Liu YM. 2002. Reconstitution of immune responses to tuberculosis in patients with HIV infection who receive antiretroviral therapy. Chest 122:597–602 Narita M, Ashkin D, Hollender ES, et al. 1998. Paradoxical worsening of tuberculosis following antiretroviral therapy in patients with AIDS. Am. J. Respir. Crit. Care Med. 158:157–61 Eyer-Silva WA, Pinto JF, Arabe J, et al. 2002. Paradoxical reaction to the treatment of tuberculosis uncovering previously silent meningeal disease. Rev. Soc. Bras. Med. Trop. 35:59–61 Wendel KR, Sterling TR. 2001. HIV and paradoxical worsening of tuberculosis. Hopkins HIV Rep. 13:8,12 Sadaphal P, Astemborski J, Graham NM, et al. 2001. Isoniazid preventive therapy, hepatitis C virus infection, and hepatotoxicity among injection drug users infected with Mycobacterium tuberculosis. Clin. Infect. Dis. 33:1687–91 Singh N, Wagener MM, Gayowski T. 2002. Safety and efficacy of isoniazid chemoprophylaxis administered during liver transplant candidacy for the prevention of posttransplant tuberculosis. Transplantation 74:892–95 Salomon N, Perlman DC, Goldstein S. 1994. Special problems in treating tuberculosis. Ann. Intern. Med. 120:440– 41 Wulff EA, Wang AK, Simpson DM. 2000. HIV-associated peripheral neuropathy: epidemiology, pathophysiology and treatment. Drugs 59:1251–60 Breen RA, Lipman MC, Johnson MA. 2000. Increased incidence of peripheral neuropathy with co-administration of stavudine and isoniazid in HIV-infected individuals. AIDS 14:615 Bothamley G. 2001. Drug treatment for tuberculosis during pregnancy: safety considerations. Drug Safety 24:553–65

12 Dec 2003

14:3

AR

AR206-ME55-17.tex

AR206-ME55-17.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:283-301. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

TUBERCULOSIS/HIV COINFECTION 73. Pickering LJ, ed. 2000. Tuberculosis. In Red Book Report of the Committee on Infectious Diseases, pp. 593–613. Elk Grove Village, IL: Am. Acad. Pediatr. 74. Narita M, Stambaugh JJ, Hollender ES, et al. 2000. Use of rifabutin with protease inhibitors for human immunodeficiency virus–infected patients with tuberculosis. Clin. Infect. Dis. 30:779–83 75. Burman WJ, Benator D, Vernon A, et al. 2003. Use of antiretroviral therapy during treatment of active tuberculosis with a rifabutin-based regimen. Presented at Conf. Retroviruses Opportunistic Infect., 10th, Feb. 10–14, Boston, MA 76. Patel A, Patel K, Patel J, et al. 2003. To study the safety and antiretroviral effi-

P1: GBC

301

cacy of concomitant use of rifampicin and efavirenz in antiretroviral-na¨ıve tuberculosis co-infected HIV-1 patients in India. Presented at Conf. Retroviruses Opportunistic Infect., 10th, Feb. 10–14, Boston, MA 77. Pedral-Sampaio D, Alves C, Netto E, et al. 2003. Efficacy of efavirenz 600 mg dose in the ARV therapy regimen for HIV patients receiving rifampicin in the treatment of tuberculosis. Conf. Retroviruses Opportunistic Infect., 10th, Feb. 10–14, Boston, MA 78. Reynolds SJ, Bartlett JG, Quinn TC, et al. 2003. Antiretroviral therapy where resources are limited. N. Engl. J. Med. 348:1806–9

2 Dec 2003

21:50

AR

AR206-ME55-18.tex

AR206-ME55-18.sgm

LaTeX2e(2002/01/18) P1: GBC 10.1146/annurev.med.55.091902.104412

Annu. Rev. Med. 2004. 55:303–17 doi: 10.1146/annurev.med.55.091902.104412 c 2004 by Annual Reviews. All rights reserved Copyright ° First published online as a Review in Advance on July 30, 2003

HIV–ASSOCIATED LIPODYSTROPHY:

Annu. Rev. Med. 2004.55:303-317. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Pathogenesis, Prognosis, Treatment, and Controversies Polyxeni Koutkia and Steven Grinspoon Massachusetts General Hospital Program in Nutritional Metabolism and Neuroendocrine Unit, Harvard Medical School, 55 Fruit Street, Boston, Massachusetts 02114; email: [email protected]

Key Words lipodystrophy, insulin resistance, HIV ■ Abstract Potent antiretroviral agents markedly suppress HIV and have dramatically improved the clinical course, prognosis, and survival of HIV-infected patients. Unfortunately, highly active antiretroviral therapy is often compromised by metabolic complications, including insulin resistance, dyslipidemia, and fat redistribution. Together these changes have been termed the HIV-lipodystrophy syndrome, which is estimated to affect a majority of patients treated with potent combination antiretroviral therapy. Routine testing of fasting glucose is recommended for all HIV-infected patients, particularly those who are obese, have a family history of diabetes mellitus, or are receiving protease inhibitor therapy. Preliminary investigations have demonstrated the potential utility of insulin-sensitizing agents and lipid-lowering therapies to ameliorate these metabolic disturbances. Patients with HIV infection who demonstrate fat redistribution and develop hyperinsulinemia and dyslipidemia may be at increased risk of cardiovascular disease. However, the long-term effects on cardiovascular disease have not yet been determined.

INTRODUCTION Human immunodeficiency virus (HIV) is a major global health problem. However, significant progress has been made in the treatment of HIV infection with highly active antiretroviral therapies (HAART). Potent antiretroviral agents markedly suppress HIV and have dramatically improved the clinical course, prognosis, and survival of HIV-infected patients. Unfortunately, HAART is often compromised by metabolic complications, including insulin resistance, dyslipidemia, and fat redistribution. Together these changes have been termed the HIV-lipodystrophy syndrome. Fat redistribution is characterized by selective loss of subcutaneous fat from the face and extremities, and in some patients, accumulation of fat around the neck, dorsocervical region, abdomen, and trunk (1, 2). Discrete subcutaneous fat deposits, particularly in the dorsocervical area, have also been described (3). A 0066-4219/04/0218-0303$14.00

303

2 Dec 2003

21:50

Annu. Rev. Med. 2004.55:303-317. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

304

AR

AR206-ME55-18.tex

KOUTKIA

¥

AR206-ME55-18.sgm

LaTeX2e(2002/01/18)

P1: GBC

GRINSPOON

minority of patients exhibit pure fat atrophy, a second group demonstrates increased abdominal visceral adiposity alone, and a third, larger group demonstrates both abnormalities (4). Breast enlargement has been observed in both women and men but whether it is due to excess subcutaneous fat, glandular hypertrophy, or both is not clear. Lipodystrophy in HIV-infected patients is associated with a cluster of metabolic abnormalities, including insulin resistance, impaired glucose tolerance, hypertriglyceridemia, and low serum levels of high-density lipoprotein (HDL) cholesterol (5, 6). Diabetes is a less common finding (4). Patients affected by severe lipodystrophy are often troubled by the disfiguring facial fat atrophy and changes in body habitus, including discomfort from dorsocervical fat accumulation in some cases. The disturbing changes in body composition may contribute to noncompliance or discontinuation of HAART despite adequate HIV suppression (7). In the United States, the prevalence of lipodystrophy is estimated at 25%–50% of HIV-infected patients receiving combined antiretroviral therapy. Although increased truncal fat has been described in patients investigated prior to the current era of HAART (5), most studies suggest that the changes in body composition are associated with use of the two major classes of antiretroviral drugs, nucleoside reverse transcriptase inhibitors (NRTIs) and protease inhibitors (PIs) (8). Use of both agents simultaneously may result in the most severe changes in body composition. Much of the initial search for the cause of fat redistribution among HIV-infected patients focused on PI therapy, and studies compared patients who were PI-exposed to PI-na¨ıve patients. Exposure to NRTI therapy has also been implicated in the development of fat redistribution, particularly fat atrophy (9). Mallal and colleagues (9) evaluated 277 patients participating in the Western Australian HIV Cohort Study and found that fat atrophy was associated with increased age and longer duration of NRTI treatment in addition to PI treatment. Stavudine exposure, in this study and others (10), significantly increased the risk of developing fat atrophy. Abnormalities of glucose homeostasis, including insulin resistance and related metabolic abnormalities (hypertriglyceridemia, low HDL, atherogenic lipid profile) frequently accompany changes in body composition among HIV-infected patients receiving HAART (4, 11). Fasting hyperinsulinemia, inappropriate insulin responses to standard glucose challenge, and decreased glucose disposal rates have now been shown both in association with changes in body composition, e.g., loss of subcutaneous abdominal and extremity fat and increased abdominal visceral fat (12), and in response to specific antiviral therapies (13). An effect of PIs on GLUT-4–mediated glucose transport in vitro (14) and on insulin sensitivity in vivo has been demonstrated (13). Insulin resistance is associated with an atherogenic lipid profile and impaired fibrinolysis in HIV-infected patients with fat redistribution (15) and may contribute independently to an increased risk of cardiovascular disease in this population.

2 Dec 2003

21:50

AR

AR206-ME55-18.tex

AR206-ME55-18.sgm

LaTeX2e(2002/01/18)

HIV-ASSOCIATED LIPODYSTROPHY

P1: GBC

305

PATHOGENESIS

Annu. Rev. Med. 2004.55:303-317. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Fat Redistribution The mechanisms responsible for fat atrophy and visceral adiposity in HIV-infected patients are not known. However, specific medications and drug classes may effect adipocyte differentiation and promote apoptosis (16). Domingo et al. (17) performed subcutaneous fat cell biopsies on HIV-infected patients treated with a PI who were experiencing subcutaneous fat atrophy and truncal adiposity. Ten of eleven samples demonstrated apoptosis, suggesting increased adipocyte cell death in the subcutaneous fat compartment (17). The disturbances in fat redistribution and metabolism among HIV-infected patients cannot currently be explained by a single agent or class of agents, but there is mounting evidence that antiretroviral therapies play a role in the development of lipodystrophy (18). Similarities in body fat distribution between patients with HIV lipodystrophy and Cushing’s syndrome have prompted investigation of the hypothalamicpituitary-adrenal axis as a possible mechanism by which PIs or NRTIs cause fat redistribution (19, 20). However, serum and urine cortisol concentrations are usually normal and dexamethasone suppression appropriate, ruling out true Cushing’s syndrome. Futhermore, subjects with HIV lipodystrophy demonstrate facial fat atrophy rather than plethora and lack many of the specific signs and symptoms of Cushing’s syndrome. Cortisol may be locally produced in adipose tissue from conversion of biologically inactive cortisone by the enzyme 11β-hydroxysteroid dehydrogenase type 1. The expression of this enzyme and of glucocorticoid receptors is significantly higher in omental fat than in subcutaneous fat (21, 22). Therefore, locally increased glucocorticoid concentrations or action without systemic hypercortisolism may induce regional adiposity, but whether this mechanism contributes to the development of HIV lipodystrophy is unknown. Carr et al. (23) identified a homology between a 12–amino acid sequence of the catalytic domain of HIV-1 protease and the retinoic acid–binding domain of cytoplasmic retinoic acid–binding protein-1 (CRABP-1) and the lipid-binding domain of low-density lipoprotein receptor–like protein (LRP). CRABP-1 carries retinoic acid (24), which, when isomerized to cis-9-retinoic acid, activates a nuclear retinoid X receptor-α peroxisome proliferator-activated receptor-γ (PPARγ ) complex known to regulate adipocyte proliferation and differentiation (25). Thus, PIs, by inhibiting CRABP-1, may inhibit adipocyte differentiation. However, preliminary results of in vitro studies do not support this hypothesis. Three-dimensional crystal analyses of CRABP-1 and HIV-1 protease showed no structural similarity (26). Furthermore, none of the PIs directly binds to retinoid X receptor-α or PPARγ (16, 27). Inhibition of LRP by PIs may not account for hyperlipidemia, since inactivation of LRP in the liver of wild-type mice does not result in hyperlipidemia (28). Other investigators propose that nonspecific inhibition of human proteins, such as insulin-degrading enzymes or cathepsins (aspartyl proteases), by PIs can cause primary hyperinsulinemia (29, 30). However, this mechanism cannot explain the loss of body fat.

2 Dec 2003

21:50

Annu. Rev. Med. 2004.55:303-317. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

306

AR

AR206-ME55-18.tex

KOUTKIA

¥

AR206-ME55-18.sgm

LaTeX2e(2002/01/18)

P1: GBC

GRINSPOON

Limited information on the histopathology of fine-needle biopsy or surgical specimens of the adipose deposition reveals nonencapsulated mature adipose tissue (31), some with fibrotic changes, ruling out dysplastic or neoplastic pathology. Subcutaneous adipocyte apoptosis has also been reported in patients with HIV lipodystrophy (32). Furthermore, apoptosis has not been reversed by switching from indinavir to nevirapine despite improvement in the metabolic parameters (32). In vitro studies with C3H10T1/2 murine mesenchymal stem cells, 3T3L1 preadipocytes, and human preadipocytes demonstrated that several PIs inhibit adipocyte differentiation. Recently, Bastard et al. (33) showed reduced mRNA expression of many transcription factors, including sterol regulatory element–binding protein-1 (SREBP-1) in adipocytes from patients with HIV lipodystrophy receiving PIs compared with healthy subjects; however, SREBP-1 protein levels were increased. As SREBP-1c overexpression in adipose tissue causes lipodystrophy in mice, PI-induced alterations of SREBP-1 expression may contribute to changes in body composition and fat distribution in HIV-infected patients. Autosomal dominant familial partial lipodystrophies have been attributed to defects in lamin A/C (34), and autosomal recessive congenital generalized lipodystrophy has been traced to mutations in 1-acylglycerol-3-phosphate Oacyltransferase-2. There are phenotypic similarities between HIV lipodystrophy and familial partial lipodystrophies. Whether lipodystrophy in HIV-infected patients is due to PI-induced changes in the expression of these genes or other homologous genes, particularly those involved in the triglyceride or phospholipid biosynthetic pathways, or with adipocyte differentiation remains to be determined. The mechanism for fat loss and gain in HIV-infected patients remains unknown. In vitro studies to date suggest that PIs may decrease adipogenesis, but no unifying hypothesis has thus far emerged. Furthermore, sequence homology of the PIs to lipoprotein receptor-like protein (LRP) suggests a potential effect of the PI’s on lipid clearance, but the clinical importance of this observation remains unknown (23). Alternatively, NRTIs may inhibit DNA polymerase-γ , decreasing oxidative phosphorylation. However, no specific link between any effects of NRTI therapy on DNA polymerase-γ and clinical fat loss has thus far been demonstrated. Some studies suggest that drug exposure alone is not sufficient to cause changes in fat distribution (35, 36). Potential interactions with immune function, cytokines, and other mediators have been postulated. For example, one study of HIV-infected men related an increase in soluble type 2 TNF-α receptor levels to the severity of extremity fat loss (37).

Lipid Abnormalities Lipid disorders are seen frequently among HIV-infected patients. Prior to the era of HAART, hypertriglyceridemia was common among HIV-infected patients (38). Increased hepatic very-low-density lipoprotein (VLDL) synthesis and decreased triglyceride clearance were demonstrated among HIV-infected patients (39). More recently, severe hypertriglyceridemia has been found, particularly among patients

2 Dec 2003

21:50

AR

AR206-ME55-18.tex

AR206-ME55-18.sgm

LaTeX2e(2002/01/18)

HIV-ASSOCIATED LIPODYSTROPHY

P1: GBC

307

Annu. Rev. Med. 2004.55:303-317. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

receiving PIs. Recent studies have investigated the effects of various PIs on lipids in HIV-negative patients and people with occupational exposure to HIV infection. For example, Purnell et al. demonstrated hypertriglyceridemia in response to ritonavir in non-HIV-infected patients (40). Several mechanisms have been proposed for PIinduced hypertriglyceridemia, including reduction of lipoprotein lipase activity (41) and protection of apolipoprotein B from degradation by proteasomes (42). These mechanisms may contribute to the development of dyslipidemia in patients receiving antiretroviral therapy, and individual PIs may have varying effects on triglyceride synthesis.

Abnormalities of Insulin and Glucose Insulin resistance and impaired glucose tolerance are commonly seen among HIVinfected patients with fat redistribution who are on HAART. HIV-infected patients with fat redistribution demonstrate fasting hyperinsulinemia and decreased insulin sensitivity. Among patients with fat redistribution in the era of HAART, fasting glucose levels do not differ from those of control subjects of comparable age and body mass index (BMI) (4). In contrast, levels of fasting insulin, twohour insulin, and two-hour glucose on standard glucose challenge are markedly increased. In a recent study, 35% of the HIV-infected patients with fat redistribution demonstrated glucose intolerance compared with only 5% of age- and BMImatched healthy control subjects. In contrast, 7% of HIV-infected patients with fat redistribution compared with 0.5% of control subjects demonstrated markedly increased two-hour glucose responses, diagnostic of diabetes (two-hour glucose >200 mg/dl) (4). Increased diastolic blood pressure, hypertriglyceridemia, low HDL, increased LDL, and markedly increased levels of tissue type plasminogen activator (tPA) and plasminogen activator inhibitor-I (PAI-1) are also associated with hyperinsulinemia in this population (4, 43, 44). The percentage of HIV-infected patients with fasting hyperinsulinemia that will develop type II diabetes is not known. Changes in body composition may also contribute to insulin resistance. Using an insulin clamp technique, Mynarcik et al. demonstrated decreased glucose disposal (insulin resistance) among HIV-infected patients, in association with reduced peripheral fat on dual energy–X-ray absorptiometry (DEXA) determination (37). Fasting insulin levels are most elevated among patients with significant peripheral fat loss and increased abdominal visceral adiposity, whereas patients with either fat loss or increased visceral adiposity demonstrate lesser degrees of hyperinsulinemia (4). Severity of body composition abnormalities, use of specific antiretroviral therapies, and family history may all contribute to the development of diabetes in this population. In multivariate modeling, Meininger et al. demonstrated a 1% increase in fasting insulin for each 1% increase in visceral fat and an independent 1% increase in fasting insulin for each 1% reduction in abdominal subcutaneous fat (12). Eighty-nine percent of the variability in fasting insulin was explained in

2 Dec 2003

21:50

Annu. Rev. Med. 2004.55:303-317. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

308

AR

AR206-ME55-18.tex

KOUTKIA

¥

AR206-ME55-18.sgm

LaTeX2e(2002/01/18)

P1: GBC

GRINSPOON

multivariate modeling, which included age, BMI, PI use, and waist-to-hip ratio (12). Direct effects of protease inhibitors on glucose regulation have recently been shown. Murata et al. demonstrated a specific effect of PI therapy on insulinstimulated glucose transport and GLUT-4 in 3T3-L1 adipocytes (14). The protease inhibitor indinavir did not affect early insulin signaling events (insulin receptor autophosphorylation and subsequent tyrosine phosphorylation or phosphatidylinositol-3 kinase activation) or the translocation of intracellular GLUT-4. At physiological concentrations, indinavir was associated with a 26% reduction in glucose uptake. Similar inhibition of insulin-mediated glucose uptake was seen with ritonavir and amprenavir. Noor et al. investigated the use of indinavir over four weeks in HIV-negative patients. A significant decrease in glucose disposal occurred with no clinically significant change in body composition (13). These data suggest more direct effects of antiretroviral agents, particularly PIs, on glucose uptake. However, changes in body composition resulting from PIs and other antiretrovirals may also simultaneously contribute to changes in glucose metabolism among HIV-infected patients receiving chronic antiretroviral therapy. Increased lipolysis resulting from direct effects of antiretroviral drugs and changes in body composition may also contribute to insulin resistance in this population. Increased free fatty acid levels and lipolytic rates have been shown among HIV-infected patients receiving HAART and predict insulin responses to standard glucose-tolerance testing (45, 46). Furthermore, acute dosing studies have shown that reduction in free fatty acids by inhibition of lipolysis increases insulin sensitivity (47).

Clinical Consequences of Metabolic Abnormalities in HIV-Infected Patients Whether the anthropometric (increased waist-to-hip ratio, increased visceral fat, reduced subcutaneous fat) and metabolic (hypertriglyceridemia, low HDL, modest increases in LDL and diastolic blood pressure) abnormalities seen among HIVinfected patients receiving HAART contribute to an actual increase in cardiovascular disease remains unknown. Recent studies using Framingham risk equations suggest increased risk for myocardial infarction among HIV-infected patients with fat redistribution (48). In addition, recent case reports suggest premature coronary artery disease in this population (49). Two recent cross-sectional studies suggest increased risk of myocardial infarction and cardiovascular events in HIV-infected patients. Mary-Krause et al. demonstrated a substantially increased risk of myocardial infarction among French HIV-infected patients receiving a PI for >30 months compared to <18 months (50). In contrast, Klein et al. investigated coronary heart disease (CHD) event rates among 4541 persons followed from 1996 to 2000 in the Kaiser Permanente system (51). CHD event rates were not significantly different among PI users versus non–PI users (5.8 versus 5.2 events/1000 PY), but overall CHD event rates were higher in the HIV-infected patients than in control subjects

2 Dec 2003

21:50

AR

AR206-ME55-18.tex

AR206-ME55-18.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:303-317. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

HIV-ASSOCIATED LIPODYSTROPHY

P1: GBC

309

(2.8 events/1000 PY of follow-up). In contrast, use of antiretroviral medication was not associated with increased coronary artery disease events in a recently published Veteran’s Affairs cohort (52). Further longitudinal studies with larger numbers of patients are needed to determine whether HIV-infected patients in the era of HAART are at increased risk for myocardial infarction or CHD. One mechanism by which hyperinsulinemia and insulin resistance may contribute to increased cardiovascular disease is through an effect on fibrinolysis. Recent data indicate that patients with HIV infection and fat redistribution have markedly elevated PAI-1 and tPA antigen levels in association with significant hyperinsulinemia (15). Data from the Framingham Offspring Study suggest a direct, independent effect of hyperinsulinemia to impair fibrinolysis in glucose-intolerant and diabetic subjects (53). Furthermore, increased tPA antigen, a marker of impaired fibrinolysis, predicts increased risk of (a) coronary artery disease mortality among patients with a history of angina pectoris and CHD and (b) cerebral vascular events among individuals without a prior history of CHD (54). PIs may also directly affect plaque formation by upregulation of CD36 and subsequent sterol accumulation in macrophages (55). In addition, studies using surrogate markers suggest an effect of PIs on carotid intimal and endothelial function. Maggi and colleagues (56) evaluated carotid arteries for premature lesions in PI-treated and PI-naive HIV-infected patients and compared them to HIV-negative control subjects. Premature carotid lesions were identified in 52.7% of PI-treated patients, 14.9% of PI-naive patients, and only 6.4% of healthy control subjects. In this study, PI exposure, smoking, and CD4 cell count predicted the occurrence of carotid lesions. At concentrations near clinical plasma levels, the PI ritonavir significantly decreased cell viability and increased cytotoxicity in human endothelial cell cultures (57). A clinical study demonstrated reduced flow-mediated vascular dilation in PI-treated versus non-PI treated patients (58).

TREATMENT Dietary and Lifestyle Modification Although data from large-scale treatment studies are not yet available, diet and increased physical activity, particularly aerobic exercise, should be encouraged to improve dyslipidemia and insulin resistance among HIV-infected patients. A 16week pilot study found resistance exercise beneficial in reducing truncal fat as well as total body fat (59). Exercise training can reduce serum triglycerides by as much as 25% and, along with dietary restriction, can reduce total abdominal and visceral fat as well (60, 61). Severely hypertriglycemic patients should be advised to consume low-fat diets and avoid ethanol consumption to prevent chylomicronemia and acute pancreatitis. Low-fat, high-carbohydrate diets, however, may exacerbate hyperinsulinemia in nondiabetic subjects and in patients with type 2 diabetes, and therefore dietary changes should always be individualized. Increased doses of

2 Dec 2003

21:50

310

AR

AR206-ME55-18.tex

KOUTKIA

¥

AR206-ME55-18.sgm

LaTeX2e(2002/01/18)

P1: GBC

GRINSPOON

Annu. Rev. Med. 2004.55:303-317. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

ζ -3 polyunsaturated fatty acids (5–10 g/d) from concentrated fish-oil preparations may effectively lower plasma triglyceride concentrations. A cross-sectional study demonstrated that a diet high in polyunsaturated to saturated fats and low in fiber was associated with hyperinsulinemia in HIV-infected patients with fat redistribution (62). There are currently no published interventional studies of the use of diet and exercise to treat or prevent the metabolic disturbances associated with fat redistribution in HIV-infected patients, but dietary and lifestyle counseling should be considered for all patients who present with such disturbances, as potential therapies to reduce the consequences of insulin resistance.

Treatment for Dyslipidemia Given the evidence that PI treatment may directly affect lipid metabolism, interruption or replacement of the PI component of antiretroviral therapy has been investigated as a potential strategy to reverse hyperlipidemia. Martinez and colleagues (63) prospectively evaluated 23 patients who discontinued PI therapy, despite sustained virologic suppression (<200 HIV-1 RNA copies/ml), because of lipodystrophy symptoms. Nevirapine was substituted for the PI, and after approximately 8 months, patients experienced a 22% reduction in total cholesterol and a 57% reduction in triglycerides. These changes were associated with improved indices of insulin resistance and decreased waist-to-hip ratio. However, subsequent reports, including a randomized trial comparing a switch to nevirapine versus continuation of PI (64), failed to demonstrate a significant improvement in hyperlipidemia with the discontinuation of PI treatment. Similarly, switching to efavirenz after discontinuing a PI has not been as effective for the reversal of hypertriglyceridemia and hypercholesterolemia (65). As previously described, Hatano et al. (66) showed significant reductions in cholesterol, LDL cholesterol, and triglycerides following interruption of HAART for a median of 7 weeks. Discontinuing PIs may indeed improve triglyceride and cholesterol levels in HIVinfected patients, but such a course of action is controversial in the well-controlled patient with improved immunologic function, and it remains experimental. Drug therapy for hypertriglyceridemia in HIV-infected patients may be initiated with a fibric acid inhibitor, e.g., gemfibrozil and/or fenofibrate, which can lower serum triglycerides by ∼40%. Hydroxymethylglutaryl enzyme coenzyme A (HMG CoA) reductase inhibitors (statins) have also been used effectively but are most effective in lowering LDL and total cholesterol (67). Simultaneous use of PIs and statins can therefore elevate plasma levels of statins and increase the risk of myopathy. Fichtenbaum et al. (68) demonstrated a significant 32-fold increase in simvastatin area-under-the-curve (AUC) concentrations in the presence of ritonavir and saquinavir. In contrast, the AUC concentration of atorvostatin increased 4.5-fold in ritonavir/saquinavir-treated patients, whereas little effect was seen on pravastatin AUC concentrations. Potential interactions between PIs and HMG CoA reductase inhibitors are important to consider in HIV-infected patients requiring lipid therapy. Pravastatin is not metabolized by cytochrome P4503A4

2 Dec 2003

21:50

AR

AR206-ME55-18.tex

AR206-ME55-18.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:303-317. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

HIV-ASSOCIATED LIPODYSTROPHY

P1: GBC

311

and can be used to treat dyslipidemia in HIV-infected patients. Similarly, atorvastatin is only moderately affected by PI use and can also be used at low dose with careful monitoring. The interaction of specific PIs other than ritonavir/saquinavir with HMG-CoA reductase inhibitors remains unknown. Fibrates are also metabolized through cytochrome P450, but the predominant pathway is CYP4A and thus fibrate metabolism is less likely to be affected by PI use. Hyperlipidemia, especially hypertriglyceridemia and low HDL, are common in HIV-infected patients. Dyslipidemia may be further exacerbated by direct effects of PIs and indirect effects mediated through fat redistribution in HAARTtreated patients. Patients with severe dyslipidemia should be treated first with diet changes and then medications if necessary. Until further data become available, it is reasonable to use the National Cholesterol Education Program (NCEP) treatment guidelines. However, special care should be taken to avoid interactions with PIs when using HMG CoA reductase inhibitors.

Treatment of Glucose Abnormalities Recent studies have emphasized the potential utility of insulin-sensitizing agents in treating abnormalities of glucose metabolism in the HIV-lipodystrophy syndrome. A randomized, placebo-controlled, 12-week pilot study investigated the effects of low-dose metformin (500 mg PO BID) (69). Metformin treatment was associated with reductions in insulin, weight, waist circumference, diastolic blood pressure, and concentrations of tPA and PAI-I and was not associated with any significant adverse events (4, 45). Development of lactic acidosis is a rare but potentially serious side effect of metformin. It should be used with caution in HIV-infected patients, especially those recently on NRTI therapy. Modest 10%–20% reduction in plasma triglyceride levels may also be seen with metformin and related to decreased hepatic VLDL production. Taken together, the preliminary studies on metformin suggest a beneficial effect on cardiovascular risk parameters, but additional studies are needed to determine the long-term safety and efficacy of metformin and the optimal target population for this intervention. Furthermore, metformin should not be used in patients with azotemia and significant liver dysfunction. Insulin resistance may also occur in HIV-infected patients from the loss of subcutaneous fat. Hadigan et al. demonstrated that reduced thigh circumference, a marker of peripheral fat atrophy, was an independent predictor of hyperinsulinemia in HIV-positive patients with fat redistribution (4). Mynarcik et al. demonstrated decreased insulin sensitivity in association with reduced subcutaneous fat (37). A novel class of therapeutic agents, the thiazolidinediones, has been shown to promote adipogenesis, primarily through an action on PPARγ . Although the thiazolidinediones have effects on both hepatic and peripheral insulin resistance, the dominant effect is to improve peripheral glucose uptake. In a recent study among non-HIV-infected patients with lipodystrophy, troglitazone significantly increased subcutaneous fat and decreased visceral fat in association with improved levels of triglyceride, free fatty acids, and insulin resistance

2 Dec 2003

21:50

Annu. Rev. Med. 2004.55:303-317. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

312

AR

AR206-ME55-18.tex

KOUTKIA

¥

AR206-ME55-18.sgm

LaTeX2e(2002/01/18)

P1: GBC

GRINSPOON

(70). A small pilot study of HIV-infected patients with subcutaneous fat atrophy demonstrated increased peripheral subcutaneous fat in response to rosiglitazone (71). In a randomized, placebo-controlled study, rosiglitazone affected insulin but not subcutaneous fat (72). Further clinical trials are needed to determine the safety and efficacy of thiazolidinediones in HIV-infected patients. Preliminary studies suggest improvement in insulin levels with substitution of a non-nucleoside reverse transcriptase inhibitor (NNRTI) for a PI, but further studies of the efficacy of switching strategies are needed (63). As with insulin resistance and lipid abnormalities, treatment strategies to minimize or reverse fat redistribution are under investigation. In a pilot study of resistance training and aerobic exercise for HIV-infected patients with increased abdominal fat (n = 10), Roubenoff and colleagues (59) showed significant decreases in trunk fat after 16 weeks of exercise. Exercise may have additional benefits in reducing insulin resistance and improving lipid profiles and should be considered in all patients, particularly those with increased central adiposity. Routine performance of fasting glucose is recommended for all HIV-infected patients, particularly those who are obese, have a family history of diabetes, or are receiving PI therapy. Because recent data demonstrate a high prevalence of impaired glucose tolerance among HIV-infected patients with fat redistribution, performance of a standard 75-g glucose tolerance test may also be useful, particularly in patients receiving HAART. Measurement of fasting insulin may be helpful but is not generally recommended outside the research setting.

Treatment of Fat Redistribution Cosmetic surgery has been used to improve lipodystrophic changes and reduce excess dorsocervical fat. To date, there have been only small case series published with few or no follow-up data on the use of plastic surgery to correct fat redistribution. Wolfort and colleagues (73) report on two patients who underwent liposuction of the dorsocervical fat pad and one patient who had liposuction of the abdomen and flanks. Patients reported satisfaction with the outcome, but no longterm follow-up information was presented. Carefully designed studies that control for antiretroviral exposure and evaluate the safety and efficacy of plastic surgery for lipodystrophy, including liposuction and fat transplantation, are imperative. Rietschel and colleagues (74) found reduced mean concentrations of growth hormone (GH) as well as reduced basal and pulse-amplitude GH concentrations among HIV-infected men with lipodystrophy compared to HIV-infected controls without lipodystrophy and compared to healthy men. Recombinant human GH (rhGH) has been used successfully in HIV-negative GH-deficient adults to reduce abdominal adiposity (75). Wanke and colleagues (76) conducted a 12-week open-label trial of rhGH (6 mg/d s.c.) in 10 patients with HIV infection and fat redistribution. Waist-to-hip ratio, a measure of central adiposity, was significantly reduced with rhGH therapy. However, the use of rhGH may induce or exacerbate insulin resistance and glucose intolerance, and one patient in this trial developed

2 Dec 2003

21:50

AR

AR206-ME55-18.tex

AR206-ME55-18.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:303-317. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

HIV-ASSOCIATED LIPODYSTROPHY

P1: GBC

313

hyperglycemia. Therefore, the utility of rhGH for the treatment of lipodystrophy may be limited in this population with increased risk of insulin resistance and diabetes. Treatment with lower doses of rhGH may be useful, particularly for patients with central adiposity, but additional information from controlled trials is necessary to determine the long-term efficacy and safety of rhGH to treat HIV-infected patients with fat redistribution. Antiretroviral switching and structured therapy interruptions have also been used to attempt reversal of fat redistribution. Martinez and colleagues (63) found significant reductions in waist-to-hip ratio after 8 months in patients who switched from a PI-containing regimen to nevirapine. In addition, after 6 months of nevirapine, 91% of patients reported subjective improvement of body fat changes. In another study, 50% of patients who switched from a PI to nevirapine reported improvements in body shape abnormalities after 6 months, whereas no patient who remained on a PI reported such improvement (64). Short-term interruption of HAART, however, has not been found to improve anthropometric measurements in patients with HIV infection and fat redistribution (66).

CONCLUSIONS HIV-associated lipodystrophy is estimated to affect the majority of patients treated with potent combination antiretroviral therapy. Significant metabolic disturbances have been identified in association with fat redistribution in these patients, including insulin resistance and hyperlipidemia. Preliminary investigations have demonstrated potential utility in the use of insulin-sensitizing agents and lipid-lowering therapies to ameliorate these metabolic disturbances. Patients with HIV infection who demonstrate fat redistribution and who develop hyperinsulinemia and dyslipidemia may be at increased risk of cardiovascular disease. However, the long-term effects on cardiovascular disease have not yet been determined. The Annual Review of Medicine is online at http://med.annualreviews.org

LITERATURE CITED 1. Hengel RL, Watts NB, Lennox JL. 1997. Benign symmetric lipomatosis associated with protease inhibitors. Lancet 350: 1596 2. Carr A, Cooper DA. 1998. Lipodystrophy associated with an HIV-protease inhibitor. N. Engl. J. Med. 339:1296 3. Lo JC, Mulligan K, Tai VW, et al. 1998. “Buffalo hump” in men with HIV-1 infection. Lancet 351:871–75 4. Hadigan C, Meigs JB, Corcoran C, et al.

2001. Metabolic abnormalities and cardiovascular disease risk factors in adults with human immunodeficiency virus infection and lipodystrophy. Clin. Infect. Dis. 32:130–39 5. Kotler DP, Rosenbaum K, Wang J, et al. 1999. Studies of body composition and fat distribution in HIV-infected and control subjects. J. Acquir. Immune Defic. Syndr. Hum. Retrovirol. 20:228–37 6. Panse I, Vasseur E, Raffin-Sanson ML,

2 Dec 2003

21:50

314

7.

Annu. Rev. Med. 2004.55:303-317. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

8.

9.

10.

11.

12.

13.

14.

15.

AR

AR206-ME55-18.tex

KOUTKIA

¥

AR206-ME55-18.sgm

LaTeX2e(2002/01/18)

P1: GBC

GRINSPOON

et al. 2000. Lipodystrophy associated with protease inhibitors. Br. J. Dermatol. 142: 496–500 Kasper T, Arboleda C, Halpern M. 2000. The impact of patient perceptions of body shape changes and metabolic abnormalities on antiretroviral therapy. Presented at XIII Int. AIDS Conf., Durban, South Africa, Jul. 9–14 Hadigan C, Corcoran C, Stanley T, et al. 2000. Fasting hyperinsulinemia in human immunodeficiency virus-infected men: relationship to body composition, gonadal function, and protease inhibitor use. J. Clin. Endocrinol. Metab. 85:35–41 Mallal SA, John M, Moore CB, et al. 2000. Contribution of nucleoside analogue reverse transcriptase inhibitors to subcutaneous fat wasting in patients with HIV infection. AIDS 14:1309–16 Saint-Marc T, Partisani M, Poizot-Martin I, et al. 1999. A syndrome of peripheral fat wasting (lipodystrophy) in patients receiving long-term nucleoside analogue therapy. AIDS 13:1659–67 Carr A, Samaras K, Burton S, et al. 1998. A syndrome of peripheral lipodystrophy, hyperlipidaemia and insulin resistance in patients receiving HIV protease inhibitors. AIDS 12:F51–58 Meininger G, Hadigan C, Rietschel P, et al. 2002. Body-composition measurements as predictors of glucose and insulin abnormalities in HIV-positive men. Am. J. Clin. Nutr. 76:460–65 Noor MA, Seneviratne T, Aweeka FT, et al. 2002. Indinavir acutely inhibits insulinstimulated glucose disposal in humans: a randomized, placebo-controlled study. AIDS 16:F1–8 Murata H, Hruz PW, Mueckler M. 2000. The mechanism of insulin resistance caused by HIV protease inhibitor therapy. J. Biol. Chem. 275:20251–54 Hadigan C, Meigs JB, Rabe J, et al. 2001. Increased PAI-1 and tPA antigen levels are reduced with metformin therapy in HIV-infected patients with fat redistribu-

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

tion and insulin resistance. J. Clin. Endocrinol. Metab. 86:939–43 Zhang B, MacNaul K, Szalkowski D, et al. 1999. Inhibition of adipocyte differentiation by HIV protease inhibitors. J. Clin. Endocrinol. Metab. 84:4274–77 Domingo P, Matias-Guiu X, Pujol RM, et al. 1999. Subcutaneous adipocyte apoptosis in HIV-1 protease inhibitor-associated lipodystrophy. AIDS 13:2261–67 Mulligan K, Grunfeld C, Tai VW, et al. 2000. Hyperlipidemia and insulin resistance are induced by protease inhibitors independent of changes in body composition in patients with HIV infection. J. Acquir. Immune Defic. Syndr. 23:35–43 Yanovski JA, Miller KD, Kino T, et al. 1999. Endocrine and metabolic evaluation of human immunodeficiency virusinfected patients with evidence of protease inhibitor-associated lipodystrophy. J. Clin. Endocrinol. Metab. 84:1925–31 Miller KK, Daly PA, Sentochnik D, et al. 1998. Pseudo-Cushing’s syndrome in human immunodeficiency virus-infected patients. Clin. Infect. Dis. 27:68–72 Bujalska IJ, Kumar S, Stewart PM. 1997. Does central obesity reflect “Cushing’s disease of the omentum”? Lancet 349:1210– 13 Rebuffe-Scrive M, Bronnegard M, Nilsson A, et al. 1990. Steroid hormone receptors in human adipose tissues. J. Clin. Endocrinol. Metab. 71:1215–19 Carr A, Samaras K, Chisholm DJ, Cooper DA. 1998. Pathogenesis of HIV1-protease-inhibitor-associated peripheral lipodystrophy, hyperlipidemia and insulin resistance. Lancet 352:1881–83 Li E, Norris AW. 1996. Structure/function of cytoplasmic vitamin A-binding proteins. Annu. Rev. Nutr. 16:205–34 Tontonoz P, Hu E, Spiegelman BM. 1995. Regulation of adipocyte gene expression and differentiation by peroxisome proliferator activated receptor gamma. Curr. Opin. Genet. Dev. 5:571–76 Stevens GJ, Chen M, Grecko R, Lankford

2 Dec 2003

21:50

AR

AR206-ME55-18.tex

AR206-ME55-18.sgm

LaTeX2e(2002/01/18)

HIV-ASSOCIATED LIPODYSTROPHY

27.

Annu. Rev. Med. 2004.55:303-317. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

28.

29.

30.

31.

32.

33.

34.

A, et al. 1999. Investigations into proposed mechanisms of HIV-associated peripheral lipodystrophy, hyperlipidemia and insulin resistance. Presented at Int. Workshop Adverse Drug Reactions and Lipodystrophy in HIV, 1st, San Diego, CA, Jun. 26–29 Lenhard JM, Furfine ES, Jain RG, et al. 2000. HIV protease inhibitors block adipogenesis and increase lipolysis in vitro. Antiviral. Res. 47:121–29 Rohlmann A, Gotthardt M, Hammer RE, et al. 1998. Inducible inactivation of hepatic LRP gene by cre-mediated recombination confirms role of LRP in clearance of chylomicron remnants. J. Clin. Invest. 101:689–95 Martinez E, Gatell J. 1998. Metabolic abnormalities and use of HIV-1 protease inhibitors. Lancet 352:821–22 Stricker RB, Goldberg B. 1998. Fat accumulation and HIV-1 protease inhibitors. Lancet 352:1392 Roth VR, Kravcik S, Angel JB. 1998. Development of cervical fat pads following therapy with human immunodeficiency virus type 1 protease inhibitors. Clin. Infect. Dis. 27:65–67 Domingo P, Matias-Guiu X, Pujol RM, et al. 2001. Switching to nevirapine decreases insulin levels but does not improve subcutaneous adipocyte apoptosis in patients with highly active antiretroviral therapy-associated lipodystrophy. J. Infect. Dis. 184:1197–201 Bastard JP, Caron M, Vidal H, et al. 2002. Association between altered expression of adipogenic factor SREBP1 in lipoatrophic adipose tissue from HIV-1-infected patients and abnormal adipocyte differentiation and insulin resistance. Lancet 359:1026–31 Speckman RA, Garg A, Du F, et al. 2000. Mutational and haplotype analyses of families with familial partial lipodystrophy (Dunnigan variety) reveal recurrent missense mutations in the globular C-terminal domain of lamin A/C. Am. J. Hum. Genet. 66:1192–98

P1: GBC

315

35. Suzawa M, Takada I, Yanagisawa J, et al. 2003. Cytokines suppress adipogenesis and PPAR-gamma function through the TAK1/TAB1/NIK cascade. Nat. Cell. Biol. 5:224–30 36. Lichtenstein KA, Ward DJ, Moorman AC, et al. 2001. Clinical assessment of HIVassociated lipodystrophy in an ambulatory population. AIDS 15:1389–98 37. Mynarcik DC, McNurlan MA, Steigbigel RT, et al. 2000. Association of severe insulin resistance with both loss of limb fat and elevated serum tumor necrosis factor receptor levels in HIV lipodystrophy. J. Acquir. Immune Defic. Syndr. 25:312–21 38. Grunfeld C, Pang M, Doerrler W. 1991. Circulating interferon alpha levels and hypertriglyceridemia in the acquired immunodeficiency syndrome. Am. J. Med. 90:154–62 39. Grunfeld C, Feingold KR. 1992. Metabolic disturbances and wasting in the acquired immunodeficiency syndrome. N. Engl. J. Med. 327:329–37 40. Purnell JQ, Zambon A, Knopp RH, et al. 2000. Effect of ritonavir on lipids and postheparin lipase activities in normal subjects. AIDS 14:51–57 41. Ranganathan S, Kern PA. 2002. The HIV protease inhibitor saquinavir impairs lipid metabolism and glucose transport in cultured adipocytes. J. Endocrinol. 172:155– 62 42. Liang JS, Distler O, Cooper DA, et al. 2001. HIV protease inhibitors protect apolipoprotein B from degradation by the proteasome: a potential mechanism for protease inhibitor-induced hyperlipidemia. Nat. Med. 7:1327–31 43. Hadigan C, Rabe J, Grinspoon S. 2002. Sustained benefits of metformin therapy on markers of cardiovascular risk in human immunodeficiency virus-infected patients with fat redistribution and insulin resistance. J. Clin. Endocrinol. Metab. 87:4611–15 44. Sattler FR, Qian D, Louie S, et al. 2001. Elevated blood pressure in subjects with lipodystrophy. AIDS 15:2001–10

2 Dec 2003

21:50

Annu. Rev. Med. 2004.55:303-317. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

316

AR

AR206-ME55-18.tex

KOUTKIA

¥

AR206-ME55-18.sgm

LaTeX2e(2002/01/18)

P1: GBC

GRINSPOON

45. Hadigan C, Borgonha S, Rabe J, et al. 2002. Increased rates of lipolysis among HIVinfected men receiving highly active antiretroviral therapy. Metabolism 51:1143– 47 46. Meininger G, Hadigan C, Laposata M, et al. 2002. Elevated concentrations of free fatty acids are associated with increased insulin response to standard glucose challenge in human immunodeficiency virusinfected subjects with fat redistribution. Metabolism 51:260–66 47. Hadigan C, Rabe J, Meininger G, et al. 2003. Inhibition of lipolysis improves insulin sensitivity in protease inhibitortreated HIV-infected men with fat redistribution. Am. J. Clin. Nutr. 77:490–94 48. Hadigan C, Meigs JB, Wilson PWF, et al. 2003. Prediction of coronary heart disease in HIV-infected patients with fat redistribution. Clin. Infect. Dis. 36:909–16 49. Henry K, Melroe H, Huebsch J, et al. 1998. Severe premature coronary artery disease with protease inhibitors. Lancet 351:1328 50. Mary-Krause M, Cotte L, Partisani M, et al. 2001. Impact of treatment with protease inhibitor (PI) on myocardial infarction (MI) occurrence in HIV-infected men. Presented at Conf. Retroviruses and Opportunistic Infections, 8th, Chicago, IL, Feb. 4–8 51. Klein D, Hurley LB, Quesenberry CP Jr, et al. 2002. Do protease inhibitors increase the risk for coronary heart disease in patients with HIV-1 infection? J. Acquir. Immune Defic. Syndr. 30:471–77 52. Bozzette SA, Ake CF, Tam HK, et al. 2003. Cardiovascular and cerebrovascular events in patients treated for human immunodeficiency virus infection. N. Engl. J. Med. 348:702–10 53. Meigs JB, Mittleman MA, Nathan DM, et al. 2000. Hyperinsulinemia, hyperglycemia, and impaired hemostasis: the Framingham Offspring Study. JAMA 283: 221–28 54. Johansson L, Jansson JH, Boman K, et al. 2000. Tissue plasminogen activator, plasminogen activator inhibitor-1, and tissue

55.

56.

57.

58.

59.

60.

61.

62.

63.

plasminogen activator/plasminogen activator inhibitor-1 complex as risk factors for the development of a first stroke. Stroke 31:26–32 Dressman J, Kincer J, Matveev SV, et al. 2003. HIV protease inhibitors promote atherosclerotic lesion formation independent of dyslipidemia by increasing CD36dependent cholesteryl ester accumulation in macrophages. J. Clin. Invest. 111:389– 97 Maggi P, Serio G, Epifani G, et al. 2000. Premature lesions of the carotid vessels in HIV-1-infected patients treated with protease inhibitors. AIDS 14:F123–28 Zhong DS, Lu XH, Conklin BS, et al. 2002. HIV protease inhibitor ritonavir induces cytotoxicity of human endothelial cells. Arterioscler. Thromb. Vasc. Biol. 22:1560–66 Stein JH, Klein MA, Bellehumeur JL, et al. 2001. Use of human immunodeficiency virus-1 protease inhibitors is associated with atherogenic lipoprotein changes and endothelial dysfunction. Circulation 104:257–62 Roubenoff R, Weiss L, McDermott A, et al. 1999. A pilot study of exercise training to reduce trunk fat in adults with HIV-associated fat redistribution. AIDS 13:1373–75 Jones SP, Doran DA, Leatt PB, et al. 2001. Short-term exercise training improves body composition and hyperlipidaemia in HIVpositive individuals with lipodystrophy. AIDS 15:2049–51 Yarasheski KE, Tebas P, Stanerson B, et al. 2001. Resistance exercise training reduces hypertriglyceridemia in HIV-infected men treated with antiviral therapy. J. Appl. Physiol. 90:133–38 Hadigan C, Jeste S, Anderson EJ, et al. 2001. Modifiable dietary habits and their relation to metabolic abnormalities in men and women with human immunodeficiency virus infection and fat redistribution. Clin. Infect. Dis. 33:710–17 Martinez E, Conget I, Lozano L, et al. 1999. Reversion of metabolic abnormalities after

2 Dec 2003

21:50

AR

AR206-ME55-18.tex

AR206-ME55-18.sgm

LaTeX2e(2002/01/18)

HIV-ASSOCIATED LIPODYSTROPHY

64.

Annu. Rev. Med. 2004.55:303-317. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

65.

66.

67.

68.

69.

70.

switching from HIV-1 protease inhibitors to nevirapine. AIDS 13:805–10 Barreiro P, Soriano V, Blanco F, et al. 2000. Risks and benefits of replacing protease inhibitors by nevirapine in HIV-infected subjects under long-term successful triple combination therapy. AIDS 14:807–12 Lafon E, Landman R, Quertainmont M, et al. 2000. LIPSTOP study: evolution of clinical lipodystrophy (LD) blood lipids, visceral (VAT) and subcutaneous (SAT) adipose tissue after switching from protease inhibitor (PI) to efavirenx (EFV) in HIV-1 infected patients. Presented at Int. AIDS Conf. Durban, South Africa, XIII, Jul. 9–14 Hatano H, Miller KD, Yoder CP, et al. 2000. Metabolic and anthropometric consequences of interruption of highly active antiretroviral therapy. AIDS 14:1935–42 Henry K, Melroe H, Huebesch J, et al. 1998. Atorvastatin and gemfibrozil for protease-inhibitor-related lipid abnormalities. Lancet 352:1031–32 Fichtenbaum CJ, Gerber JG, Rosenkranz SL, et al. 2002. Pharmacokinetic interactions between protease inhibitors and statins in HIV seronegative volunteers: ACTG Study A5047. AIDS 16:569–77 Hadigan C, Corcoran C, Basgoz N, et al. 2000. Metformin in the treatment of HIV lipodystrophy syndrome: a randomized controlled trial. JAMA 284:472–77 Arioglu E, Duncan-Morin J, Sebring N,

71.

72.

73.

74.

75.

76.

P1: GBC

317

et al. 2000. Efficacy and safety of troglitazone in the treatment of lipodystrophy syndromes. Ann. Intern. Med. 133:263–74 Gelato MC, Mynarcik DC, Quick JL, et al. 2002. Improved insulin sensitivity and body fat distribution in HIV-infected patients treated with rosiglitazone: a pilot study. J. Acquir. Immune Defic. Syndr. 31: 163–70 Yki-Jarvinen H, Sutinen J, Silveira A, et al. 2003. Regulation of plasma PAI-1 concentrations in HAART-associated lipodystrophy during rosiglitazone therapy. Arterioscler. Thromb. Vasc. Biol. 23:688–94 Wolfort FG, Cetrulo CL, Nevarre DR. 1999. Suction-assisted lipectomy for lipodystrophy syndromes attributed to HIVprotease inhibitor use. Plast. Reconstr. Surg. 104:1814–20; discussion 1821–22 Rietschel P, Hadigan C, Corcoran C, et al. 2001. Assessment of growth hormone dynamics in human immunodeficiency virusrelated lipodystrophy. J. Clin. Endocrinol. Metab. 86:504–10 Cuneo RC, Judd S, Wallace JD, et al. 1998. The Australian Multicenter Trial of Growth Hormone (GH) Treatment in GHDeficient Adults. J. Clin. Endocrinol. Metab. 83:107–16 Wanke C, Gerrior J, Kantaros J, et al. 1999. Recombinant human growth hormone improves the fat redistribution syndrome (lipodystrophy) in patients with HIV. AIDS 13:2099–103

2 Dec 2003

21:49

AR

AR206-ME55-19.tex

AR206-ME55-19.sgm

LaTeX2e(2002/01/18) P1: GBC 10.1146/annurev.med.55.091902.103520

Annu. Rev. Med. 2004. 55:319–31 doi: 10.1146/annurev.med.55.091902.103520 c 2004 by Annual Reviews. All rights reserved Copyright °

HUMAN PAPILLOMAVIRUS VACCINES AND PREVENTION OF CERVICAL CANCER

Annu. Rev. Med. 2004.55:319-331. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Kathrin U. Jansen and Alan R. Shaw Merck & Co., Inc., 770 Sumneytown Pike, West Point, Pennsylvania 19486; email: kathrin [email protected]; alan [email protected]

Key Words virus-like particle vaccines, vaccine delivery, proof-of-concept, clinical endpoints ■ Abstract Cervical cancer and precancerous cervical lesions constitute a major problem in women’s health. Every year 470,000 cases of cervical cancer are diagnosed worldwide, and about half the women afflicted will die. In the United States alone, ∼14,000 cases of cervical cancer are diagnosed each year despite the availability of screening and access to high-quality gynecological care. With the confirmation that cervical cancer is caused by an infectious agent, human papillomavirus, the possibility of fighting this disease with either prophylactic or therapeutic vaccination arose. This review describes advances in vaccine development and very promising first results for prophylactic vaccination against cervical cancer.

INTRODUCTION Cervical cancer and precancerous cervical lesions constitute a major problem in women’s health. Every year 470,000 cases of cervical cancer are diagnosed worldwide, and about half the women afflicted will die. The prevalence of cervical cancer is estimated to be 1.4 million cases worldwide (1). A substantial proportion of modern gynecological effort and expense is devoted to the detection and treatment of cervical dysplasias and cervical cancer. In the United States alone, ∼50 million Pap tests are performed each year, and they discover ∼1.2 million cases of lowgrade dysplasia (CIN1), ∼300,000 cases of high-grade dysplasia (CIN2/3), and ∼14.000 cases of cervical cancer (2). The total health care cost associated with the screening and treatment of cervical cancer in the United States is estimated to be $6 billion per year. Despite this screening and treatment, ∼5000 women die from the disease each year in the United States. In areas of the world where most women do not have access to regular, high-quality gynecological care and screening, cervical cancer is second only to breast cancer as a cancer-related cause of death (1). Four times as many cases of cervical cancer occur in the developing world as in more developed countries. Although screening has dramatically reduced the incidence of the disease in the developed world, it is still overall only ∼75% effective. 0066-4219/04/0218-0319$14.00

319

2 Dec 2003

21:49

320

AR

JANSEN

AR206-ME55-19.tex

¥

AR206-ME55-19.sgm

LaTeX2e(2002/01/18)

P1: GBC

SHAW

A VIRUS THAT CAUSES CANCER

Annu. Rev. Med. 2004.55:319-331. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

The etiology of cervical dysplasia and cancer was unclear for a long time. Environmental factors as well as smoking and alcohol consumption have received attention. Further study focused on candidate sexually transmitted infectious agents such as Chlamydia, Neisseria gonorrhea, and particularly Herpes simplex virus. Starting in the early 1980s, several lines of investigation converged on the human papillomaviruses (HPVs) as the primary suspects to cause cervical cancer. The important steps in this convergence were as follows: 1. Finding HPV genomes in cervical carcinomas and cloning the viruses (3). The advent of more sophisticated detection technologies, such as polymerase chain reaction (PCR), greatly aided the detection of HPV in cancer specimens. 99.7% of all cervical cancers harbor HPV types (4). 2. Defining HPVs as a large family of closely related viruses. Over 80 HPV types have now been identified, of which 40 infect the genital tract (5). 3. Elucidating the mechanisms of transformation. During the 1990s, several teams of investigators demonstrated that HPV is a true “tumor virus” in that it carries genes encoding multiple proteins that interfere with cell-cycle control, leading to transformation and uncontrolled cell growth (6). 4. Demonstrating that infection with rabbit and bovine papillomavirus can cause tumors in rabbits and cows (7, 8). 5. Completing epidemiological case-control studies (9–12) in humans that supplied sufficient evidence to prompt the International Agency for Research on Cancer (IARC) to declare HPV types 16 and 18 human carcinogens in 1995. Recently, IARC published data from their pooled 11 case-control studies suggesting that an additional 13 HPV types should be considered carcinogens (13). This body of information was slow to develop, and its evolution depended on advances in cellular, molecular, and immunological diagnostic technologies that have occurred during the past 20 years.

VIRUS GENETICS, MOLECULAR BIOLOGY, AND LIFE CYCLE HPVs are small, nonenveloped viruses with relatively small (∼8 kb pairs), doublestranded circular DNA genomes (6). As members of the papovavirus family, they share numerous characteristics with the better-known examples, SV40 and polyoma virus. The HPV genome codes for only eight proteins. The late L1 and L2 genes code for the virus capsid proteins, early proteins E1 and E2 are responsible for viral replication and transcription, and E4 seems to aid virus release from infected cells (14). HPVs are DNA tumor viruses that also encode proteins that

2 Dec 2003

21:49

AR

AR206-ME55-19.tex

AR206-ME55-19.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:319-331. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

CERVICAL CANCER VACCINES

P1: GBC

321

interfere with the activity of host cell-cycle–regulating proteins such as RB and p53, among others. HPV early proteins E6 and E7 inhibit or target for destruction cellular proteins RB and p53, and E5 has also been implicated in cellular transformation (15). There is now ample evidence that HPV has the tools to cause cancer based on molecular mechanisms alone. To date, over 80 genotypes of HPV have been described (5). Genotyping of HPV is based on DNA sequences of the L1, E6, and E7 genes. A 10% difference in sequence with respect to previously established strains is sufficient to define a new type of virus. HPV has also been associated with several cutaneous hyperplastic conditions, including common warts of the hands and feet (types 1, 2, and 4), genital warts and recurrent respiratory papillomatosis (types 6 and 11), epidermodysplasia verruciformis (types 5 and 8), and anal as well as cervical carcinomas and adenocarcinomas (types 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 68, and 73). Although many HPV types are known, only a few of them cause the vast majority of disease. HPV types 16 and 18 are responsible for >70% of all cervical cancer cases, followed by HPV types 31 and 45 (4), which cause an additional 10% of cases. Two additional HPV types, HPV6 and 11, are not considered high-risk or oncogenic types but are responsible for >90% of all anogenital warts in men and women (16). In general, HPV infects the basal cells of a human epithelial surface (17). Infected basal cells divide; some progeny remain as infected basal cells while others, also infected, move away from the basement membrane, differentiate, and become epithelial cells. Virus replication and assembly is tightly linked to the differentiation program of the epithelial cell. Infectious virions are produced only in the terminally differentiated cell and are shed as virus-laden squames. This explains why HPV cannot grow in tissue culture. Over the years, various methods— including the nude mouse (18, 19) and SCID mouse (20) xenograft systems as well as raft-culture systems (21)—have achieved limited propagation of infectious virus for some HPV types, but the lack of a convenient culture system has hampered vaccine development.

EPIDEMIOLOGY OF HUMAN PAPILLOMAVIRUS By infecting only the cells of the basal layer and executing virus replication and assembly only in a fully differentiated cell that is destined to die, the virus avoids the immune system of the host. The success of this strategy is documented by the very poor immune responses (humoral as well as cell-mediated) to HPV infection. However, most HPV infections seem ultimately controlled and eliminated by an immune response. The nature of this response is still under investigation. Infections that are not controlled and persist for a long time can cause more severe pathologies and, ultimately, cancer (22). Infection in both women and men is clearly related to sexual activity (23, 24). For women, the most striking risk factors for HPV infection and development

2 Dec 2003

21:49

322

AR

JANSEN

AR206-ME55-19.tex

¥

AR206-ME55-19.sgm

LaTeX2e(2002/01/18)

P1: GBC

SHAW

Annu. Rev. Med. 2004.55:319-331. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

of detectable pathology are numerous lifetime sexual partners and early onset of sexual activity. Although women bear the brunt of dysplasias and cancer, HPVrelated carcinomas of the anal mucosa are relatively frequent in men having sex with men (25), and the rate seems to be increasing. A third group at risk for severe HPV infections are neonates. At birth, neonates sometimes acquire HPV types 11 and 6 from the infected birth canal of their mother. HPV can infect the mucosa of the pharynx and cause large wart-like growths that can obstruct the airway. Recurrent respiratory papillomatosis is rare but potentially life-threatening. Patients with this condition undergo multiple surgical procedures each year in order to breathe and speak (26–29).

APPROACHES TO A VACCINE AND CURRENT CLINICAL STATUS Multiple approaches are being tried in the quest for an effective vaccine. An excellent review of the status of the field, citing company websites and other internet sources, was published in July 2000 (30). However, much progress has occurred since then. There are two major alternatives to consider when one sets out to make a vaccine against HPV. Should one strive to prevent infection by creating a prophylactic vaccine, or should one focus on a therapeutic vaccine for individuals already infected? Although vaccines in the classical sense are prophylactic, as are all currently licensed vaccines, multiple groups are pursuing a therapeutic vaccine. Depending on the approach, the choice of viral antigens to use as immunogens as well as the choice of an appropriate delivery system could be quite different. Prophylactic vaccines are simpler in that they need only to raise an immune response sufficient to limit infection and prevent clinical disease. A therapeutic vaccine must elicit an immune response that can clear an already established infection. This requires a vaccine to make the immune system do something it has failed to do during the primary infection. In other words, the therapeutic vaccine has to do better than nature. In the case of HPV, antigens to be presented fall into two general classes: the early proteins, which do not become part of the virion but are expressed at some level in infected cells, and the late proteins, which make up the virus coat (6). For a successful therapeutic vaccine, the chosen antigens should be expressed in every infected cell and the vaccine should induce an immune response that mobilizes the cell-mediated arm of the immune system to rid the body of virally infected cells. Primary target antigens for a therapeutic vaccine are the oncoproteins E6 and E7 because they are expressed throughout the life cycle of HPV as well as in cancer cells. Additional targets are the E1 and E2 proteins associated with viral replication/transcription. For a successful prophylactic vaccine, the obvious target antigens would be the capsid proteins because they are the only antigens accessible for a classical neutralizing antibody response to prevent infection. Delivery of these antigens results in the mobilization of the humoral arm of the immune system and induces

2 Dec 2003

21:49

AR

AR206-ME55-19.tex

AR206-ME55-19.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:319-331. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

CERVICAL CANCER VACCINES

P1: GBC

323

a strong, long-lasting, virus-neutralizing antibody response. This is not to say that early viral antigens might not also be useful. Early work by Christensen & Kreider demonstrated that antibodies against the L1 capsid protein neutralized HPV11 in a nude-mouse xenograft system (31). However, the prophylactic vaccine field really took off in 1991, after Zhou et al. demonstrated that the HPV16 L1 capsid protein, when expressed in a recombinant system, forms virus-like particles (VLPs) resembling native virions (32). Numerous groups have now shown the successful expression of L1 VLPs from a number of HPV types. Unfortunately, because papillomavirus infection is species-specific, there is no HPV disease model. Preclinical animal papillomavirus models have helped to assess the feasibility of the various vaccine approaches. The cottontail rabbit has a naturally occurring papillomavirus that causes large, pigmented, exophytic skin warts. This model is widely favored for its convenience, simplicity, and reproducibility. Both therapeutic and prophylactic cottontail rabbit papillomavirus vaccines have been tested successfully (33, 34). Several bovine papillomaviruses (BPVs) have been useful. The well-known BPV1 can be used to transform mouse C127 cells grown in tissue culture. It causes skin growths analogous to those of the rabbit. BPV4 causes papillomas of the digestive tract in cows. These mucosal lesions are somewhat closer to the disease targeted for a vaccine in humans. Prophylactic vaccination has been observed to prevent BPV disease, and therapeutic vaccination has ameliorated established warts (35, 36). The dog, particularly the beagle, develops wart-like growths on the lips and gums caused by canine oral papillomavirus. This is a transient condition that blooms and resolves in a matter of weeks. In these animals, too, vaccination has been shown to prevent infection and disease (37, 38). For therapeutic vaccines, mouse models that use tumor cells expressing viral antigens have also shown success (39, 40). However, mouse and other animal models do not necessarily predict whether a particular approach will work in nonhuman primates or humans.

ANTIGEN DELIVERY Naked DNA In the early 1990s, it was shown that injecting the gene for an antigen, surrounded by the appropriate control signals, in the context of a plasmid, could raise a cellular and humoral response against the antigen (41). Vaccination of rabbits with a cottontail rabbit papillomavirus L1 DNA construct was shown to protect rabbits from papillomavirus infection and wart formation (42). The naked DNA approach has several advantages. It is simple, it is relatively quick to develop, and the same methods can be applied to almost any antigen, carbohydrates being the major exception. Naked DNA was examined as a therapeutic as well as prophylactic vaccine by several groups and showed promise in preclinical models. However, results in

2 Dec 2003

21:49

324

AR

JANSEN

AR206-ME55-19.tex

¥

AR206-ME55-19.sgm

LaTeX2e(2002/01/18)

P1: GBC

SHAW

humans and nonhuman primates were disappointing; multiple very high doses of naked DNA vaccines are required to elicit immune responses.

Annu. Rev. Med. 2004.55:319-331. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Vectored Gene Delivery The relative inefficiency of naked DNA led to the next step of delivering the antigen coding sequence in a viral vector. This allows the antigen gene to enter cells more efficiently, and it permits targeting to particular cell types; different virus vectors have different cell tropisms. One of the first vaccines into the clinic was vaccinia virus expressing the transforming proteins E6 and E7 of HPV types 16 and 18 (43, 44). This vaccine was intended as a therapeutic adjunct to traditional cervical cancer therapy. Though discontinued for cancer therapy, it is still under evaluation for its ability to ameliorate cervical dysplasia. Vaccinia is also being used in two Chinese development programs delivering genes for either HPV16 or 58 L1 fused to their respective E7 genes. HPV 58 is highly prevalent in China. A highly attenuated form of vaccinia, modified vaccinia Ankara, is being developed by Transgene S.A. as a vehicle for the codelivery of HPV genes and the gene for interleukin-2. Addition of IL-2 to the vaccine is intended to boost the T-cell response of the host. Our group looked preclinically at a replication-deficient adenovirus expressing HPV16 L1 and compared it to DNA and VLP immunization by evaluating humoral as well as cell-mediated immune responses in rhesus macaques. Although the recombinant adenovirus induced strong cell-mediated immune responses, it induced a weaker neutralizing antibody response than VLPs did (45).

Fusion Proteins Preclinical experiments in cattle, using fusion proteins between the minor capsid protein L2 and the oncogene E7, suggested that vaccination with such a fusion protein and an appropriate adjuvant could ameliorate disease. A human version of the vaccine targeting HPV6 was licensed and tested by Glaxo SmithKline for the treatment of genital warts. This approach demonstrated no efficacy and the vaccine was abandoned. A second fusion-protein vaccine had a similar fate. MediGene AG tested an HPV16 chimeric VLP in which the HPV16 L1 capsid protein was fused to the E7 protein in 36 patients with cervical cancer. MediGene recently announced the discontinuation of this approach because the phase I/II trial results did not fulfill its predetermined efficacy criteria. HPV16 immune responses, however, were detected in immunized patients. A third approach using a fusion protein of a Bacillus Calmette-Guerin (BCG) heat-shock protein with E7 (Stressgen Biotechnologies Corp.) is still being evaluated clinically against high-grade cervical dysplasia.

Virus-Like Particles Expression of the capsid or virus-coat protein genes of certain viruses in a heterologous system, such as yeast or insect cells, produces a VLP through

2 Dec 2003

21:49

AR

AR206-ME55-19.tex

AR206-ME55-19.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:319-331. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

CERVICAL CANCER VACCINES

P1: GBC

325

self-directed assembly of the recombinant protein. Hepatitis B surface antigen was the first example of a VLP that became a commercially viable vaccine. As mentioned above, Zhou et al. (32) were the first to report that the HPV major capsid protein L1, expressed in a recombinant system, self-assembles into a VLP. Preclinical efficacy was demonstrated using VLPs formulated on aluminum adjuvants. They induced a strong virus-neutralizing antibody response in nonhuman primates (46, 47). Preclinical results were confirmed by several phase I studies that tested the immunogenicity and safety of monovalent VLP-based vaccines. These vaccines were generally well tolerated and generated high levels of neutralizing antibodies (48–50). These initial results were encouraging, but would such a vaccine actually prevent HPV infection? After all, the antibody response induced by the VLP-based vaccine is generated systemically, whereas the target cells for HPV infection are epithelial cells of the genital tract. Therefore, the proposition that systemic antibody may prevent local infection of the genital mucosa was untested. We designed a double-blind, placebo-controlled proof-of-concept study to answer this question (51). Young women (n = 2392) were assigned to receive placebo or yeast-derived HPV16 L1 VLPs (40-µg dose) formulated on Merck aluminum adjuvant at day 0, month 2, and month 6 by intramuscular injection. Samples from the genital tract were obtained at enrollment, one month after the booster immunization, and every six months thereafter. In addition, the women underwent gynecological examinations and were referred for colposcopy according to protocol. Biopsy tissue was evaluated for intraepithelial neoplasia and analyzed by PCR for the presence of HPV16 DNA. DNA was prepared from the specimens using routine methods. HPV16 DNA was amplified by PCR using typeand gene-specific primers for the HPV16 L1, E6, and E7 genes. PCR products were visualized by dot-blot hybridization using type- and gene-specific oligonucleotides. The assays were validated to have a 95% probability to detect 13 copies of HPV16 DNA per sample. The primary endpoint of the trial was persistent HPV16 infection, defined by (a) HPV16 DNA detection in samples obtained at two or more visits at least four months apart; (b) a cervical biopsy showing cervical intraepithelial neoplasia or cancer and HPV16 DNA in the biopsy and in a genital sample collected at the antecedent or subsequent visit; or (c) HPV16 DNA detected in a sample collected during the last visit before being lost to follow-up. Women were followed for a median of 17.4 months after completion of the vaccination regimen, at which time 41 cases of persistent HPV16 infection were accrued. All 41 cases occurred in the placebo group, none in the vaccine group. Of these 41 cases, 31 were persistent HPV16 infection, 5 were HPV16-related CIN 1, 4 were HPV16-related CIN2, and 1 occurred in a woman who first tested positive for HPV16 on the last visit before she was lost to follow-up. These results translate to 100% efficacy (95% confidence interval, 90–100; p < 0.001). Since all 9 cases of HPV16-related CIN were in the placebo group, there is great hope that an HPV VLP-based vaccine may reduce the incidence of cervical cancer.

2 Dec 2003

21:49

326

AR

JANSEN

AR206-ME55-19.tex

¥

AR206-ME55-19.sgm

LaTeX2e(2002/01/18)

P1: GBC

SHAW

Annu. Rev. Med. 2004.55:319-331. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Merck is currently conducting phase III clinical trials to assess the efficacy of a quadrivalent HPV vaccine covering HPV types 16 and 18 as well as types 6 and 11. The National Cancer Institute and Glaxo SmithKline are slated to start phase III clinical trials at the end of 2003 with a bivalent VLP-based vaccine covering HPV types 16 and 18 (51a). From a technical perspective, vaccination with VLPs appears promising. Nevertheless, several practical issues must be addressed before these vaccines can be licensed and deployed in clinical practice and public health programs.

PRACTICAL ISSUES Is Prevention of Persistent Infection an Acceptable Clinical Endpoint for a Phase III Study and Licensure? Epidemiological studies have established that persistent infection with HPV is a prerequisite for the development of the vast majority of high-grade dysplasia and cervical cancer. Therefore, one would logically assume that a vaccine preventing HPV infection would prevent cervical cancer. In the current regulatory environment, if one wishes to claim in the vaccine’s labeling that it prevents cancer, one must actually demonstrate prevention of cancer or at least a defined precursor, not just infection. This was the view of the U.S. Food and Drug Administration’s External Advisory Committee at a recent review of several HPV-vaccine projects (52). Given the frequency of self-clearance of infection and self-resolution of earlygrade lesions, an efficacy study measuring prevention of clinical lesions rather than infection is not an unreasonable requirement. Clinical trials generally take place in an environment where there is an established “standard of care” for the disease to be addressed. Ethically, any deviation from this standard of care would put trial participants at unacceptable risk. Because high-grade dysplasia (CIN2/3) is regarded as a direct precursor to cervical cancer and must be treated, a cervical cancer endpoint would be unethical. An appropriate compromise is to use the appearance of CIN2/3 lesions related to vaccine types as an endpoint in a placebo-controlled study. Because true CIN2/3 lesions are relatively rare in current practice (53), the phase III trial now under way will need to include tens of thousands of volunteers.

Public Health Aspects of HPV Vaccination Assuming that the great promise seen in Merck’s proof-of-concept study is borne out in the phase III trials, a vaccine against HPV infection and disease should become available in several years. However, there are several questions that will generate substantial debate. Current clinical studies are focused on young women in their late teens and early twenties, often in university settings. Entry into university is generally associated with increased sexual activity, and students tend to stay in

WHEN TO VACCINATE?

2 Dec 2003

21:49

AR

AR206-ME55-19.tex

AR206-ME55-19.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:319-331. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

CERVICAL CANCER VACCINES

P1: GBC

327

one place for at least a few years, making follow-up more practical. Realistically, however, entry into university is frequently later than onset of sexual activity. A recent study published in Communicable Disease Reports (54) shows the rates of diagnosis of genital warts (first attack) in English and Welsh clinics by age and gender. The data suggest that vaccination of adolescents aged 10–12 years would be most desirable to control HPV infection and disease, since it is most likely that this population is still uninfected. Experience with vaccines in general has shown that the most efficient way to get as many people vaccinated as possible is to vaccinate them when they are infants. Public health programs have shown that coverage is best for pediatric programs and less successful in adults. A further advantage is that immune responses to vaccines are generally better in children. Because the duration of a protective immune response after immunization with VLP-based vaccines is not yet known, it remains to be seen whether effective infant immunization could be achieved. It also would further crowd the already intense infant-vaccination schedule. Clinical studies to date have focused on females because women suffer most from the pathology of HPV infection. Males, however, are the vectors. With the notable exception of penile warts and some cases of penile and anal cancer, there is little obvious pathology associated with HPV in heterosexual males. HPV is very difficult to detect in this population. This is partly because of the lack (until recently) of an acceptable method of sampling. Men having sex with men do suffer from anal intraepithelial neoplasia. The anal epithelium has a transition zone similar to that of the cervix, and this is the most frequent site of HPV disease in this group. Since vaccines work best when given to large proportions of the population, vaccination trials to show some efficacy in men are also being considered. We can learn from our experience with rubella vaccination. When effective rubella vaccines became available in the 1970s, some public health authorities chose to vaccinate only girls, in some cases not until their early teens. This seemed reasonable because the main deleterious consequence of rubella infection was fetal rubella syndrome, a major cause of devastating birth defects. Vaccinating women before childbearing age should have been sufficient. However, experience, best documented in Sweden, showed that sex-specific vaccination was not an effective policy. Only when both boys and girls were vaccinated in the first years of life did rubella and fetal rubella syndrome essentially vanish (55).

WHAT ABOUT MALES?

MARKETING A VACCINE FOR A SEXUALLY TRANSMITTED DISEASE Even if HPV vaccines are shown to be safe and effective, marketing a vaccine against a sexually transmitted disease to the general public may be problematic. The public health and health economic benefits are likely to be considerable. Public health authorities would most likely recommend a vaccine that would prevent cervical cancer, at least for women. Parental resistance, however, can easily be imagined: “Why does my child need to be vaccinated against this pathogen? She/he won’t be sexually active for a long time. Let’s hold off on this until later.” The most effective strategy

2 Dec 2003

21:49

328

AR

JANSEN

AR206-ME55-19.tex

¥

AR206-ME55-19.sgm

LaTeX2e(2002/01/18)

P1: GBC

SHAW

Annu. Rev. Med. 2004.55:319-331. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

will be to maintain philosophical distance from the sexual aspects of the question and focus on the prevention of a common cause of cancer. IMPLEMENTING HPV VACCINATION IN THE DEVELOPING WORLD Cervical cancer screening and access to high-quality gynecological care are limited for women in developing countries, yet ∼80% of cases occur in these parts of the world. An effective prophylactic vaccine would have an enormous impact on women’s health, if it could be delivered. Three major issues must be resolved in order to take full advantage of the promise of HPV vaccines. First, the global infrastructure must be reinforced to accommodate the logistics of delivery of a new vaccine to a, perhaps, nonpediatric population. This is a rather tall order, and in practice, this may become a pediatric vaccine in developing countries even if the developed world makes a different choice. There is no adolescent vaccination visit in most parts of the world. The World Health Organization’s Expanded Program for Immunization delivers the “basic six” vaccines (diphtheria, tetanus, pertussis, polio, measles and BCG) to a large fraction of the world’s birth cohort. If effective immunity could be shown to last into adulthood, then pediatric administration may be the easier solution for developing countries. Second, the capacity for producing these vaccines on a global scale must be created. The “chicken-and-egg” aspect of this problem might not be obvious to those outside the vaccine industry. In order to justify the capital and other ancillary investments necessary to create manufacturing capacity approximately ten times greater than one might normally contemplate, there must be some reasonable assurance of a market for the product. This is tightly linked to the third issue, funding. Today, in 2003, most of the developing world’s vaccines are paid for by governmental or international donor agencies. Until recently, the vaccines provided through this funding mechanism have been “traditional” vaccines such as the basic six above. In the past few years, there has been a growing uptake of hepatitis B vaccine and conjugated polysaccharide vaccine against Haemophilus influenza. This expansion has been greatly aided by funding from the Gates Foundation and an overall reinvigorated interest in vaccines. To deliver an HPV vaccine for cervical cancer to the women in greatest need, many of whom live in the very poorest countries, one can only hope that industry, governments, and donor organizations will make similar efforts and alliances.

The Annual Review of Medicine is online at http://med.annualreviews.org

LITERATURE CITED 1. Ferlay J, Bray F, Pisani P, et al. 2001. Globocan 2000: Cancer Incidence, Mortality and Prevalence Worldwide. Lyons, France: IARC Press

2. Koutsky LA, Galloway DA, Holmes KK. 1988. Epidemiology of genital human papillomavirus infection. Epidemiol. Rev. 10:122–63

2 Dec 2003

21:49

AR

AR206-ME55-19.tex

AR206-ME55-19.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:319-331. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

CERVICAL CANCER VACCINES 3. International Agency for Research on Cancer. 1995. IARC Working Group on the Evaluation of Carcinogenic Risks to Humans. Lyons, France: Int. Agency Res. Cancer. 87 pp. 4. Bosch FX, Lorincz A, Munoz N, et al. 2002. The causal relation between human papillomavirus and cervical cancer. J. Clin. Pathol. 55:244–65 5. de Villiers EM. 2001. Taxonomic classification of papillomaviruses. Papillomavirus Rep. 57–63 6. Howley PM. 1996. Papillomaviridae: the viruses and their replication. In Fields Virology, ed. BN Fields, DM Knipe, PM Howley, pp. 2045–76. Philadelphia: Lippincott-Raven 7. Jarrett WF, Murphy J, O’Neil BW, et al. 1978. Virus-induced papillomas of the alimentary tract of cattle. Int. J. Cancer 22:323–28 8. Kreider JW. 1980. Neoplastic progression of the Shope rabbit papilloma. In Viruses in Naturally Occurring Cancers, ed. M Essex, G Todaro, H zur Hausen, pp. 283– 99. Cold Spring Harbor, NY: CSH Press 9. Kjaer SK, vandenBrule AJC, Bock JE, et al. 1996. Human papillomavirus—the most significant risk determinant of cervical intraepithelial neoplasia. Int. J. Cancer 65:601–6 10. Walboomers JMM, Jacobs MV, Manos MM, et al. 1999. Human papillomavirus is a necessary cause of invasive cervical cancer worldwide. J. Pathol. 189:12–19 11. Schiffman MH, Bauer HM, Hoover RN, et al. 1993. Epidemiologic evidence showing that human papillomavirus infection causes most cervical intraepithelial neoplasia. J. Natl. Cancer Inst. 85:958– 64 12. Munoz N, Bosch FX, de Sanjose S, et al. 1992. The causal link between human papillomavirus and invasive cervical cancer: a population-based case-control study in Colombia and Spain. Int. J. Cancer 52:743–49 13. Munoz N, Bosch FX, de Sanjose S, et al.

14.

15.

16.

17.

18.

19.

20.

21.

22.

P1: GBC

329

2003. Epidemiologic classification of human papillomavirus types associated with cervical cancer. N. Engl. J. Med. 348:518– 27 Bryan JT, Brown DR. 2001. Transmission of human papillomavirus type 11 infection by desquamated cornified cells. Virology 281:35–42 Auvinen E, Crusius K, Steuer B, et al. 1997. Human papillomavirus type 16 E5 protein (review). Int. J. Oncol. 11:1297– 304 Gissmann L, Wolnik L, Ikenberg H, et al. 1983. Human papillomavirus types 6 and 11 DNA sequences in genital and laryngeal papillomas and in some cervical cancers. Proc. Natl. Acad. Sci. USA 80:560– 63 Taichman LB, LaPorta R. 1986. The Expression of Papillomaviruses in Epithelial Cells, pp. 109–39. New York.: Plenum Kreider JW, Howett MK, Wolfe SA, et al. 1985. Morphological transformation in vivo of human uterine cervix with papillomavirus from condylomata acuminata. Nature 317:639–41 Kreider JW, Howett MK, Leure Dupree AE, et al. 1987. Laboratory production in vivo of infectious human papillomavirus type 11. J. Virol. 61:590–93 Bonnez W, Rose RC, Da Rin C, et al. 1993. Propagation of human papillomavirus type 11 in human xenografts using the severe combined immunodeficiency (SCID) mouse and comparison to the nude mouse model. Virology 197:455– 58 Meyers C, Frattini MG, Hudson JB, et al. 1992. Biosynthesis of human papillomavirus from a continuous cell line upon epithelial differentiation. Science 257:971–73 Nobbenhuis MAE, Walboomers JMM, Helmerhorst TJM, et al. 1999. Relation of human papillomavirus status to cervical lesions and consequences for cervicalcancer screening: a prospective study. Lancet 354:20–25

2 Dec 2003

21:49

Annu. Rev. Med. 2004.55:319-331. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

330

AR

JANSEN

AR206-ME55-19.tex

¥

AR206-ME55-19.sgm

LaTeX2e(2002/01/18)

P1: GBC

SHAW

23. Bauer HM, Ting Y, Greer CE, et al. 1991. Genital human papillomavirus infection in female university students as determined by a PCR-based method. JAMA 265:472–77 24. Ho GYF, Bierman R, Beardsley L, et al. 1998. Natural history of cervicovaginal papillomavirus infection in young women. N. Engl. J. Med. 338:423–28 25. Goldstone SE, Winkler B, Ufford LJ, et al. 2001. High prevalence of anal squamous intraepithelial lesions and squamous cell carcinoma in men who have sex with men as seen in a surgical practice. Dis. Colon Rectum 44:690–98 26. Mahnke CG, Werner JA, Frohlich O, et al. 1998. Clinical course of recurrent laryngeal papillomatosis and detection of human papillomavirus DNA sequences in laryngeal papillomas. Laryngorhinootologie 77:157–64 27. Abramson AL, Steinberg BM, Winkler B. 1987. Laryngeal papillomatosis: clinical, histopathologic and molecular studies. Laryngoscope 97:678–85 28. Elo J, Hidvegi J, Bajtai A. 1995. Papova viruses and recurrent laryngeal papillomatosis. Acta Otolaryngol. 115:322–25 29. Terry RM, Lewis FA, Griffiths S, et al. 1987. Demonstration of human papillomavirus types 6 and 11 in juvenile laryngeal papillomatosis by in-situ DNA hybridization. J. Pathol. 153:245–48 30. Kols A, Sherris J. 2000. HPV Vaccines: Promise and Challenges. Seattle: Program for Appropriate Technology in Health 31. Christensen ND, Kreider JW. 1990. Antibody-mediated neutralization in vivo of infectious papillomaviruses. J. Virol. 64:3151–56 32. Zhou J, Sun XY, Stenzel DJ, et al. 1991. Expression of vaccinia recombinant HPV 16 L1 and L2 ORF proteins in epithelial cells is sufficient for assembly of HPV virion-like particles. Virology 185:251–57 33. Jansen KU, Rosolowsky M, Schultz LD, et al. 1995. Vaccination with yeast-

34.

35.

36.

37.

38.

39.

40.

41.

42.

expressed cottontail rabbit papillomavirus (CRPV) virus-like particles protects rabbits from CRPV-induced papilloma formation. Vaccine 13:1509–14 Leachman SA, Shylankevich M, Slade MD, et al. 2002. Ubiquitin-fused and/or multiple papillomavirus early genes from cottontail rabbit as DNA vaccines. J. Virol. 76:7616–24 Campo MS, Grindlay GJ, O’Neil BW, et al. 1994. Prophylactic and therapeutic vaccination against a mucosal papillomavirus. J. Gen. Virol. 74:945–53 Campo MS. 1997. Vaccination against papillomavirus in cattle. Clin. Dermatol. 15:275–83 Suzich JA, Ghim SJ, Palmer-Hill FJ, et al. 1995. Systemic immunization with papillomavirus L1 protein completely prevents the development of viral mucosal papillomas. Proc. Natl. Acad. Sci. USA 92:11553–57 Moore RA, Santos EB, Nicholls PK, et al. 2002. Intraepithelial DNA immunization with a plasmid encoding a codon optimized COPV E1 gene sequence, but not the wild-type gene sequence completely protects against mucosal challenge with infectious COPV in beagles. Virology 304:451–59 Jochmus I, Schafer K, Faath S, et al. 1999. Chimeric virus-like particles of the human papillomavirus type 16 (HPV 16) as a prophylactic and therapeutic vaccine. Arch. Med. Res. 30:269–74 Greenstone HL, Nieland JD, deVisser KE, et al. 1998. Chimeric papillomavirus virus-like particles elicit antitumor immunity against the E7 oncoprotein in an HPV16 tumor model. Proc. Natl. Acad. Sci. USA 95:1800–5 Donnelly JJ, Ulmer JB, Shiver JW, et al. 1997. DNA vaccines. Annu. Rev. Immunol. 15:617–48 Donnelly JJ, Martinez D, Jansen KU, et al. 1996. Protection against papillomavirus with a polynucleotide vaccine. J. Infect. Dis. 173:314–20

2 Dec 2003

21:49

AR

AR206-ME55-19.tex

AR206-ME55-19.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:319-331. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

CERVICAL CANCER VACCINES 43. Borysiewicz LK, Fiander A, Nimako M, et al. 1996. A recombinant vaccinia virus encoding human papillomavirus types 16 and 18, E6 and E7 proteins as immunotherapy for cervical cancer. Lancet 347:1523–27 44. Boursnell MEG, Rutherford E, Hickling JK, et al. 1996. Construction and characterization of a recombinant vaccinia expressing human papillomavirus proteins for immunotherapy of cervical cancer. Vaccine 14:1485–94 45. Tobery TW, Smith JF, Kuklin N, et al. 2003. Effect of vaccine delivery system on the induction of HPV16 L1-specific humoral and cell-mediated immune responses in immunized rhesus macaques. Vaccine 21:1539–47 46. Lowe RS, Brown DR, Bryan JT, et al. 1997. Human papillomavirus type 11 (HPV-11) neutralizing antibodies in the serum and genital mucosal secretions of African green monkeys immunized with HPV-11 virus-like particles expressed in yeast. J. Infect. Dis. 176:1141–45 47. Palker TJ, Monteiro JM, Martin MM, et al. 2001. Antibody, cytokine and cytotoxic T lymphocyte responses in chimpanzees immunized with human papillomavirus virus-like particles. Vaccine 19:3733–43 48. Brown DR, Bryan JT, Schroeder JM, et al. 2001. Neutralization of human papillomavirus type 11 (HPV-11) by serum from women vaccinated with yeast-derived HPV-11 L1 virus-like particles: correlation with competitive radioimmunoassay titer. J. Infect. Dis. 184:1183–86 49. Harro CD, Pang YYS, Roden RBS, et al.

50.

51.

51a.

52.

53.

54.

55.

P1: GBC

331

2001. Safety and immunogenicity trial in adult volunteers of a human papillomavirus 16 L1 virus-like particle vaccine. J. Natl. Cancer Inst. 93:284–92 Evans TG, Bonnez W, Rose RC, et al. 2001. A phase 1 study of a recombinant viruslike particle vaccine against human papillomavirus type 11 in healthy adult volunteers. J. Infect. Dis. 183:1485–93 Koutsky LA, Ault KA, Wheeler CM, et al. 2002. A controlled trial of a human papillomavirus type 16 vaccine. N. Engl. J. Med. 347:1645–51 Billich A. 2003. HPV vaccine MedImmune/GlaxoSmithKline. Curr. Opin. Investig. Drugs 4(2):210–13 Food Drug Admin. Center Biol. Eval. Res. 2001. Summary minutes, vaccines and related biological products advisory committee. Meeting #88. http://www.fda.gov/ ohrms/dockets/ac/01/minutes/3805m1.pdf Woodman CBJ, Collins S, Winter H, et al. 2001. Natural history of cervical human papillomavirus infection in young women: a longitudinal cohort study. Lancet 357:1831–36 Communicable Disease Reports. 2001. Sexually transmitted infections quarterly report: anogenital warts and HSV infection in England and Wales. Commun. Dis. Rep. 11:11–15 Boettinger M, Forsgren M. 1997. Twenty years’ experience of rubella vaccination in Sweden: 10 years of selective vaccination (of 12-year-old girls and of women postpartum) and 13 years of a general two-dose vaccination. Vaccine 15:1538– 44

2 Dec 2003

21:47

AR

AR206-ME55-20.tex

AR206-ME55-20.sgm

LaTeX2e(2002/01/18) P1: GBC 10.1146/annurev.med.55.091902.103612

Annu. Rev. Med. 2004. 55:333–53 doi: 10.1146/annurev.med.55.091902.103612 First published online as a Review in Advance on Oct. 20, 2003

Annu. Rev. Med. 2004.55:333-353. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

OPPORTUNITIES FOR CONTROL OF MENINGOCOCCAL DISEASE IN THE UNITED STATES∗ Pratima L. Raghunathan, Scott A. Bernhardt, and Nancy E. Rosenstein Meningitis and Special Pathogens Branch, Centers for Disease Control and Prevention, 1600 Clifton Road, Atlanta, Georgia 30333; email: [email protected], [email protected], [email protected]

Key Words Neisseria meningitidis, epidemiology, chemoprophylaxis, meningococcal vaccines, meningococcal diagnostics ■ Abstract The United States currently has relatively low rates of meningococcal disease caused by Neisseria meningitidis. Serogroups Y, C, and B are most common. Although most cases are sporadic, a minority are associated with outbreaks. Pediatric populations have disproportionately higher rates of disease, but nearly two thirds of all cases occur in persons aged 15 years and older. The major challenge to control of domestic meningococcal disease is the absence of a vaccine to prevent sporadic cases spanning many age groups. The quadrivalent A/C/Y/W-135 meningococcal polysaccharide vaccine is licensed in the United States, but because of its limited efficacy in children under two years of age, it is recommended for high-risk groups and outbreak response rather than routine childhood immunization. New conjugate meningococcal vaccines have successfully reduced endemic disease in the United Kingdom, and similar vaccines promise to have a dramatic impact on the burden of meningococcal disease in the United States.

INTRODUCTION The epidemiology of meningococcal disease in the United States has undergone a tremendous shift over the past hundred years. In the first half of the twentieth century, large, explosive “cerebrospinal meningitis epidemics” raged periodically, with primary attack rates as high as 310 per 100,000 population and case fatality ratios approaching 70% (1–3). Mortality rates dropped with the advent of sulfonamide antibiotics, but major epidemics in both civilian and military populations ∗ The U.S. Government has the right to retain a nonexclusive, royalty-free license in and to any copyright covering this paper.

333

2 Dec 2003

21:47

Annu. Rev. Med. 2004.55:333-353. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

334

AR

AR206-ME55-20.tex

RAGHUNATHAN

¥

AR206-ME55-20.sgm

BERNHARDT

¥

LaTeX2e(2002/01/18)

P1: GBC

ROSENSTEIN

recurred during World War II troop mobilizations (4). These regular meningococcal disease epidemics disappeared from the United States in the postwar period (5). Since the 1950s, the United States has experienced low and relatively stable rates of endemic meningococcal disease at 1–2 per 100,000 population (5). Superimposed on this background rate, the meningococcus causes occasional outbreaks within organizations or communities. This pattern of predominantly endemic disease overlaid with infrequent outbreaks is also observed in other industrialized nations (6). In the United States, the major challenge to control of meningococcal disease is the absence of a vaccine to prevent sporadic cases. Because of their limited efficacy in young children, meningococcal polysaccharide vaccines are recommended for high-risk groups and outbreak response rather than routine childhood immunization (7). However, new conjugate meningococcal vaccines have successfully reduced endemic disease in the United Kingdom, and similar vaccines promise to have a dramatic impact on the burden of meningococcal disease in the United States.

Microbiology Meningococcal disease is caused by the encapsulated Gram-negative diplococcus Neisseria meningitidis. The meningoccal capsule consists of chemically distinct polysaccharides that can be classified antigenically into at least 13 serogroups (A, B, C, H, I, K, L, W-135, X, Y, Z, Z0 , 29E), five of which cause the vast majority of disease (A, B, C, Y, W-135). Meningococci are further distinguished by serotype and serosubtype based on the outer membrane proteins (OMPs) PorB and PorA, which lie within the meningococcal outer membrane beneath the polysaccharide capsule. Other OMPs include Opa (class 5), Opc (class 5c), and transferrin binding proteins (Tbps). The serogroup A, C, Y, and W-135 polysaccharide capsules elicit serogroup-specific bactericidal antibody responses (8, 9), which correlate with protection against serogroup A and serogroup C disease (10). These polysaccharide moieties form the basis of the quadrivalent serogroup A/C/Y/W-135 meningococcal polysaccharide vaccine. In contrast, the serogroup B polysaccharide capsule is poorly immunogenic, probably because of its similarity to polysialosyl glycopeptides expressed on the surface of developing neural cells, which induce selftolerance (11). Therefore, vaccine strategies against serogroup B meningococci have focused on OMPs (12).

Carriage and Immunity Meningococci colonize the human nasopharynx, which is the organism’s only natural reservoir. Asymptomatic carriage of both pathogenic and nonpathogenic strains is relatively common, yet few carriers develop invasive disease. In the United States, baseline meningococcal carriage rates are 5%–10% (13). The duration of carriage ranges from weeks to months (14). Transmission occurs through direct contact with respiratory droplets from colonized individuals. Increased carriage rates can be observed in crowded settings, such as military barracks (15). Meningococcal carriage is an immunizing event, resulting in the development of

2 Dec 2003

21:47

AR

AR206-ME55-20.tex

AR206-ME55-20.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:333-353. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

MENINGOCOCCAL DISEASE CONTROL

P1: GBC

335

serogroup-specific protective antibody (16). Adolescents and young adults have the highest meningococcal carriage rates; children and infants more frequently carry the nonpathogenic species Neisseria lactamica, which may be an important means of acquiring cross-protective immunity (14, 17). The classic studies of Goldschneider et al. found that the age-dependent risk of meningococcal disease correlated with carriage and naturally acquired immunity to meningococcus (10, 16). Infants in the first month of life have a moderate rate of disease because they are protected by transplacentally derived maternal antibodies (16, 18). As this protective immunity wanes, meningococcal disease risk increases, with rates peaking at 3–4 months of age (10, 18). As children gradually become exposed to meningococci and N. lactamica through nasopharyngeal carriage, and to antigenically similar enteric flora such as E. coli K1 and K92 (19, 20), they develop bactericidal antibody and have lower disease rates. By adulthood, 65%–85% of individuals possess bactericidal antibody against meningococci and consequently remain at low disease risk (10). Age-related waning of natural immunity may contribute to increased meningococcal disease rates observed in persons aged 65 years and older (21).

Clinical Features In a small proportion of carriers, meningococci invade the mucosa and proliferate in the bloodstream, causing invasive disease. Invasive meningococcal disease encompasses three common clinical forms: meningitis, meningococcal bacteremia, and pneumonia. Meningitis (meningeal infection), observed in ∼50% of invasive meningococcal infections, is characterized by abrupt onset of fever, headache, and neck stiffness, sometimes with nausea, vomiting, photophobia, and altered mental status (21). Meningococcal bacteremia (bloodstream infection) occurs in 40% of invasive disease cases, and a subset exhibit clinical signs of meningococcemia, or fulminant meningococcal sepsis (21, 22). Key signs of meningococcemia are sudden onset of fever and a petechial or purpuric rash. The clinical course is characterized by hemodynamic instability leading to shock, diffuse intravascular coagulation, and death; case fatality ratios have been reported to range from 18% to 53% (23). Meningococcal pneumonia occurs in ∼6% of invasive disease cases (21). In contrast to the other clinical forms of meningococcal disease, pneumonia primarily affects older patients and results in case fatality ratios below 10% (24, 25). Despite presumed improvements in clinical care since the 1970s, case fatality ratios for all meningococcal infections have remained relatively stable between 9% and 12% (5). Between 8 and 19% of survivors suffer from serious sequelae such as deafness, neurologic deficits, or limb loss (22, 23, 26).

Risk Factors Risk factors for meningococcal disease can be categorized into organism characteristics that promote virulence; environmental conditions that facilitate exposure to meningococci; and host factors that increase bacterial colonization, invasion, and survival in the bloodstream (22). Meningococcal virulence determinants include

2 Dec 2003

21:47

Annu. Rev. Med. 2004.55:333-353. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

336

AR

AR206-ME55-20.tex

RAGHUNATHAN

¥

AR206-ME55-20.sgm

BERNHARDT

¥

LaTeX2e(2002/01/18)

P1: GBC

ROSENSTEIN

capsular polysaccharide, adhesins, nutrient-acquisition factors, and the ability to release outer membrane vesicles containing endotoxin (27). In the environment, crowded living conditions are likely to facilitate respiratory droplet transmission of meningococci (28–32). Black race (21, 33) and low socioeconomic status (2, 3, 34), both linked to higher rates of meningococcal disease, may also be considered environmental risk factors, in that they are presumably markers for increased exposure to high-transmission settings. Risk factors that likely influence meningococcal colonization or invasion include include active or passive smoking (30, 32, 35, 36) and recent Mycoplasma pneumoniae or viral upper respiratory tract infections (30, 37, 38). Meningococci may be better able to attach to and penetrate nasopharyngeal mucosa that have been damaged by other pathogens or by tobacco smoke (30, 32, 35, 36). Risk factors related to host immune defense include age (10, 16), chronic illness (30), and rare immune deficiencies (39–41). Natural immunity is acquired with age, and this inverse relationship between age and susceptibility is thought to explain high rates of meningococcal disease in children aged less than two years (10, 16). Chronic underlying illness may reduce humoral immune defense (30). Rare host immune deficiencies, such as late component complement deficiency (39), properdin deficiency (40), and asplenia (41), also favor the proliferation of meningococci in the bloodstream, the former two by interfering with classical and alternative pathways for complement-mediated lysis. However, because these conditions are rare, persons with these known risk factors account for only a small fraction of meningococcal disease cases (42).

EPIDEMIOLOGY OF MENINGOCOCCAL DISEASE IN THE UNITED STATES Each year, 2400–3000 cases of meningococcal disease occur in the United States (21, 43). Approximately 97% of cases are sporadic and represent background endemic disease; the remaining 3% are associated with outbreaks (21, 43). Meningococcal disease is seasonal, with incident cases peaking in December and January (21). Both passive and active surveillance systems are used to monitor meningococcal disease, a reportable disease in the United States. In the passive National Notifiable Diseases Surveillance System (NNDSS), state health departments collect and transmit weekly reports of cases to the Centers for Disease Control and Prevention (CDC) through the National Electronic Telecommunications System for Surveillance (44). From 1996 through 2001, the average annual incidence of meningococcal disease reported to NNDSS varied greatly by state, ranging from 0.6 per 100,000 population in Delaware to 2.8 per 100,000 population in Oregon (Figure 1). Regional variation in meningococcal disease was also apparent, with elevated rates detected in the Pacific Northwest, midwestern Mississippi Valley, and South. The higher disease rates in the Pacific Northwest were probably due to the well-documented

2 Dec 2003

21:47

AR

AR206-ME55-20.tex

AR206-ME55-20.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:333-353. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

MENINGOCOCCAL DISEASE CONTROL

P1: GBC

337

Figure 1 Distribution of mean annual meningococcal disease incidence rates in the United States, 1996–2001. Incidence rates per 100,000 population per year are averaged for the years 1996–2001 by state. Shading represents rate quartiles. Confirmed (clinically compatible illness with isolation of N. meningitidis from a normally sterile site) and probable (clinically compatible cases with positive antigen test in cerebrospinal fluid or clinical purpura fulminans in the absence of a positive blood culture) cases are included. Sources: National Notifiable Diseases Surveillance System and US Census Bureau.

epidemic of serogroup B meningococcal disease in Oregon and neighboring areas of Washington, which was first detected in 1993 (45, 46). Low-incidence states were concentrated in the Northeast, along the Canadian border, and in the Southwest. The factors governing clustering of high- and low-incidence states deserve further investigation, although these crude rates are not adjusted for differing age and race structures of the underlying state populations. As a complement to the passive NNDSS system, CDC coordinates active laboratory-based surveillance for invasive meningococcal disease as part of the Emerging Infections Program through Active Bacterial Core surveillance (ABCs) (47). Participating surveillance sites collect data from all patients with sterile site N. meningitidis isolates, allowing detection of trends in causative meningococcal serogroup and accurate estimation of age-specific incidence rates. From 1996 through 2001, the largest proportion of meningococcal disease cases was due to serogroup Y (39%), followed by serogroup C (31%) and serogroup B (23%) (Figure 2). The increasing proportion of serogroup Y has been previously noted in the United States (21), whereas serogroups B and C predominate in Canada and Europe (6, 48). Persons with serogroup Y meningococcal disease were more likely to be older, to be black, to have chronic underlying illnesses, and to present

2 Dec 2003

21:47

Annu. Rev. Med. 2004.55:333-353. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

338

AR

AR206-ME55-20.tex

RAGHUNATHAN

¥

AR206-ME55-20.sgm

BERNHARDT

¥

LaTeX2e(2002/01/18)

P1: GBC

ROSENSTEIN

Figure 2 Serogroup distribution among meningococcal isolates received from participating Active Bacterial Core surveillance sites (California, Colorado, Connecticut, Georgia, Maryland, Minnesota, New York, Tennessee), 1996–2001. NG = nongroupable. Analysis excludes Oregon because of its unusual serogroup B meningococcal disease epidemic.

with meningococcal pneumonia (21, 49). Serogroup A was notably absent and serogroup W-135 was rare in this US population, yet both have recently caused major meningococcal epidemics in Africa (50, 51). Following the 2000 serogroup W-135 outbreak associated with the Hajj in Saudi Arabia, serogroup W-135 cases were detected among a few pilgrims returning to the United States and their close contacts (52, 53). Nevertheless, serogroup W-135 meningococcal disease rates have not increased in the United States (CDC, unpublished data). Importantly, approximately one fourth of US meningococcal cases were caused by the non– vaccine-preventable B serogroup. Because the population under active ABCs surveillance is defined, these data can also be used to generate national age-specific meningococcal disease incidence rates and disease burden (Figure 3). As has been historically observed, in 1996– 2001, children under two years of age had the highest age-specific incidence of meningococcal disease (5.5 per 100,000 population), followed by children aged 2–4 years. However, children under five years accounted for only 25% of the total disease burden. Although pediatric populations had disproportionately higher rates of disease, nearly two thirds of all meningococcal disease cases occurred in adolescents and adults aged 15 years and older. Consistent with previous data,

2 Dec 2003

21:47

AR

AR206-ME55-20.tex

AR206-ME55-20.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:333-353. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

MENINGOCOCCAL DISEASE CONTROL

P1: GBC

339

Figure 3 Age-specific annual incidence rates and burden of meningococcal disease, race-adjusted and projected to US population from Active Bacterial Core surveillance (ABCs) data, 1996–2001. ABCs sites included California, Colorado, Connecticut, Georgia, Maryland, Minnesota, New York, Oregon, and Tennessee; aggregate population under surveillance ranged from 24.1 million in 1996 to 35.4 million in 2001.

slightly elevated rates of disease were observed in adolescents and young adults aged 15–24, and in adults over 65 years (21). Therefore, primary prevention strategies for the United States must consider the dispersed disease burden that spans many age groups.

DIAGNOSTIC TECHNIQUES USEFUL IN CHARACTERIZING MENINGOCOCCAL DISEASE The current US confirmed case definition for meningococcal disease requires isolation of N. meningitidis from a sterile site, typically blood or cerebrospinal fluid (CSF) but occasionally joint, pleural, or pericardial fluid. In cases of meningococcemia, aspirates or skin biopsies of purpura or petechiae can also yield meningococci (54). As an adjunct to culture, latex agglutination testing can rapidly detect meningococcal polysaccharide antigens in CSF and provide serogroup identification. Although commercial latex agglutination kits detect N. meningitidis capsular antigens with high sensitivity and specificity among culture-confirmed cases (55), these tests appear to have low sensitivity when Gram stain and culture of CSF are

2 Dec 2003

21:47

Annu. Rev. Med. 2004.55:333-353. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

340

AR

AR206-ME55-20.tex

RAGHUNATHAN

¥

AR206-ME55-20.sgm

BERNHARDT

¥

LaTeX2e(2002/01/18)

P1: GBC

ROSENSTEIN

negative (56, 57). Ultrasound has been reported to enhance the sensitivity of latex agglutination testing for N. meningitidis (58). Determination of the meningococcal serogroup becomes critically important in the context of investigating suspected meningococcal disease outbreaks, because public health actions differ for vaccine-preventable and non–vaccine-preventable serogroups. Patients with suspected meningitis often receive parenteral antibiotics prior to lumbar puncture, which interfere with culture confirmation. This has prompted the development of nonculture meningococcal diagnostics (59). Polymerase chain reaction (PCR) assays can detect meningococcal-specific nucleic acid sequences in CSF and blood. Most involve an initial screening reaction to confirm meningococcal infection and a subsequent reaction to determine serogroup. The first PCR test amplifies the meningococcal-specific capsular transport gene ctrA; specimens that test positive are subjected to the second test, which distinguishes serogroup B, C, Y, and W-135 alleles of the siaD sialyltransferase gene (60–63). These techniques have been adapted to a fully automated TaqMan system (64) that allows the rapid, sensitive, and specific confirmation of meningococcal etiology as well as identification of the main disease-causing serogroups. A LightCycler PCR system has also recently been developed that detects and genogroups A, B, C, Y, and W-135 meningococci within a few hours (65). Because of its different polysaccharide biosynthesis pathway, serogroup A capsule is detected by PCR amplification of the sacC gene in this system (65). In England and Wales, 36% of meningococcal disease cases are confirmed by PCR alone (64). Similar technology is being evaluated in the United States. Both phenotypic and genotypic methods have been used to investigate meningococcal diversity and global epidemiology. Serogrouping, serotyping, and serosubtyping are phenotypic methods that require specialized reagents for serologic discrimination of variant meningococcal surface structures—namely, capsular polysaccharide (serogroup) and porin proteins PorB (serotype) and PorA (serosubtype). Multilocus enzyme electrophoresis (MEE) is the established phenotyping technique for analyzing the temporal and geographic distribution of meningococcal strains across the world. MEE detects allelic variants of conserved metabolic enzymes revealed through electrophoretic mobility differences on starch gels (66). Although labor-intensive and time-consuming, this phenotypic subtyping method has been used to classify meningococci into electrophoretic types (ETs) and to identify hypervirulent lineages (67). For example, serogroup B meningococci of the ET-5 complex were shown to have caused an epidemic in Norway that was first detected in 1974 and lasted through 1991 (67, 68). This clonal complex spread across Europe and South America in the 1980s. In the United States, ET-5 strains were subsequently associated with the serogroup B meningococcal disease epidemic in Oregon from 1992 through 1996 (46). MEE has also demonstrated that meningococci from a different clonal lineage, the ET-37 complex, have caused serogroup C outbreaks in the United States (69). MEE has been used for two decades for meningococcal subtyping, but the technique is restricted to a few reference laboratories, and its results are difficult to

2 Dec 2003

21:47

AR

AR206-ME55-20.tex

AR206-ME55-20.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:333-353. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

MENINGOCOCCAL DISEASE CONTROL

P1: GBC

341

standardize between groups. Molecular genotyping techniques are increasingly being explored to classify disease isolates both in the localized outbreak setting and within the global context of meningococcal disease (70). Pulsed-field gel electrophoresis (PFGE) can be a valuable short-term molecular subtyping tool to determine whether isolates from different individuals in a suspected outbreak represent the same strain. This technique exploits the rapid evolution of variability in restriction enzyme sites within the meningococcal genome, and thus can distinguish unrelated sporadic disease isolates with multiple PFGE patterns from an outbreak clone. A retrospective analysis of PFGE profiles within serogroup C outbreaks in the United States demonstrated that isolates were not identical but had a very high degree of similarity (>95% pattern relatedness), and this knowledge would have provided additional evidence for public health action (69). PFGE can also discriminate among highly diverse serogroup B meningococci (71, 72). In contrast to its utility for serogroup C outbreaks, however, PFGE is not as frequently employed for investigating serogroup B meningococcal disease because of the organism’s great diversity (69). Multi-locus sequence typing (MLST) employs a similar rationale to MEE’s, but entails sequencing seven conserved “housekeeping” genes and classifying allelic differences into sequence types. The main advantage of MLST is its reliance on standard molecular biology techniques, which enables different laboratories to document and compare their results quite readily; typing results can be deposited in a public database accessible by the Internet (http://neisseria.org/nm/typing/mlst). The congruence between MLST sequence types and MEE electrophoretic types has been established for some hypervirulent lineages of meningococci (70). However, a recent comparison of meningococcal subtyping methods revealed that MLST may not discriminate between sporadic and outbreak isolates as well as a newer technique, 16S ribosomal RNA gene sequencing (73). Different combinations of classical and molecular subtyping techniques may be appropriate for public health investigations and population genetic studies of meningococci.

CHEMOPROPHYLAXIS TO PREVENT MENINGOCOCCAL DISEASE Persons who have close contact with meningococcal disease patients are at substantially increased risk for acquiring carriage and disease (74–76). Among close contacts, household members of index cases have a dramatically elevated risk of acquiring disease compared to the general population in industrialized countries, with relative risk estimates ranging from 500 to 1200 (77–80). The secondary attack rate among this exposed group has been estimated at 2–4 per 1000 exposed persons (77, 78). Rates of secondary disease also appear somewhat elevated among daycare attendees (80) and schoolchildren (81). One study in the United Kingdom estimated secondary disease among health care workers to be 0.8 per 100,000 persons, a small absolute risk but 25 times greater than in the general population (82).

2 Dec 2003

21:47

Annu. Rev. Med. 2004.55:333-353. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

342

AR

AR206-ME55-20.tex

RAGHUNATHAN

¥

AR206-ME55-20.sgm

BERNHARDT

¥

LaTeX2e(2002/01/18)

P1: GBC

ROSENSTEIN

Systemic antibiotics can eradicate nasopharyngeal carriage of meningococci among contacts of sporadic cases and thus prevent secondary disease. Consequently, the US Advisory Committee on Immunization Practices (ACIP) (7) and the Red Book (83) recommend antimicrobial chemoprophylaxis for close contacts of meningococcal disease cases. Approximately 70% of secondary cases occur within seven days of disease onset in the index case, necessitating prompt antibiotic administration, ideally within 24 h of identifying the case (7, 79, 80). Antibiotic chemoprophylaxis is unlikely to be helpful after 14 days. Anecdotal evidence suggests widespread implementation of these recommendations. Since secondary cases are rare, chemoprophylaxis represents the most significant means of prevention of meningococcal disease in the United States. In serogroup C meningococcal outbreaks, mass chemoprophylaxis is not often considered because of the existence of effective polysaccharide vaccines with longer duration of protection. However, because of the lack of a serogroup B vaccine, mass chemoprophylaxis has been employed to control organization-based serogroup B meningococcal outbreaks. In an evaluation of rifampicin administered prophylactically to 900 students in a school outbreak of serogroup B disease, meningococcal carriage was reduced by 85%, and no further cases were detected (84). However, rifampicin-resistant meningococcal isolates rapidly emerged, although they did not cause disease (84). Mass chemoprophylaxis appears most effective in focal serogroup B outbreaks in small, well-defined populations such as schools (84), rather than in community-wide serogroup B outbreaks of longer duration (85). An analysis of school-based meningococcal disease clusters lent further support to the potential utility of chemoprophylaxis in school settings (81). Within these school clusters, one third of subsequent cases appeared within two days of disease onset in the index case. Thus, even when an organization-based outbreak is caused by a vaccine-preventable serogroup, antibiotic distribution may be a more timely intervention than vaccination, because protective antibodies take 7–10 days to develop after vaccination. The potential benefit of mass chemoprophylaxis in these settings needs to be weighed against the possible emergence of antibiotic resistance, rare adverse events associated with chemoprophylaxis, and the logistic difficulties of prophylaxis campaigns (84).

Antimicrobial Agents for Chemoprophylaxis Current ACIP guidelines recommend rifampicin, ciprofloxacin, or ceftriaxone as chemoprophylactic agents because of their demonstrated efficacy in eradicating meningococcal carriage (Table 1) (7). A two-day regimen of rifampicin is effective in clearing carriage but is unsuitable for pregnant women because of its teratogenicity (84, 86). A single dose of ciprofloxacin can eradicate carriage (87, 88), but it is not generally recommended for pregnant and lactating women and children under 18 years owing to findings of cartilage damage in animal models (89). However, ciprofloxacin has been used to eradicate carriage in Malawian children without adverse events (90). Ceftriaxone is also effective as a single dose, but it must

2 Dec 2003

21:47

AR

AR206-ME55-20.tex

AR206-ME55-20.sgm

LaTeX2e(2002/01/18)

P1: GBC

MENINGOCOCCAL DISEASE CONTROL

343

TABLE 1 Antibiotics recommended by the US Advisory Committee on Immunization Practices for chemoprophylaxis against meningococcal disease (7)

Adult dose

Pediatric dose

Route

Duration

Antimicrobial resistance documented?

Rifampicin (84, 86)

600 mg/12 h

10 mg/kg/12 h

oral

2 days

Yes

Ciprofloxacin (87, 88)

500 mg

—

oral

Single dose

No

Ceftriaxone (91)

250 mg

125 mg

IM

Single dose

No

Annu. Rev. Med. 2004.55:333-353. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Generic name (References)

be administered parenterally (91). More recently, a single dose of azithromycin was shown to eradicate carriage of meningococci in a cohort of Egyptian nursing students (92); further validation of these results in a pediatric population (e.g., mass chemoprophylaxis in a school) could expand the battery of meningococcal chemoprophylactic agents for specific outbreak settings.

VACCINES TO PREVENT MENINGOCOCCAL DISEASE Meningococcal Polysaccharide Vaccines The quadrivalent serogroup A/C/Y/W-135 polysaccharide vaccine (Menomune®) is the only meningococcal vaccine licensed in the United States. Although the vaccine is recommended for controlling serogroup A, C, Y, and W-135 meningococcal epidemics, it is not routinely used against endemic disease because of its immunologic shortcomings. The protective efficacy of serogroup C polysaccharide has been estimated at ∼85% in both clinical trials and epidemic settings (93–95). However, the serogroup C polysaccharide does not induce strong or lasting immune responses in children under two years of age (96–98). Even in vaccinated adults, serogroup C serum bactericidal antibody levels decline markedly within two years of vaccination (99). The serogroup A polysaccharide has a similarly high protective efficacy, between 89% and 100% in clinical trials (100, 101), and the vaccine has proven effective in controlling epidemics (102–104). Infants as young as three months develop antibodies to serogroup A polysaccharide (97, 105) and can develop short-term protection (101). However, the antibody response declines within 12 months to background levels (98), and the duration of protection against serogroup A disease appears short-lived in children and adults (99, 106). In children vaccinated before the age of four years, vaccine efficacy declines from 100% to 8% within three years; in children vaccinated after four years of age, the vaccine efficacy decreases from 85% to 67% over the same time period (106). The protective efficacy of the

2 Dec 2003

21:47

Annu. Rev. Med. 2004.55:333-353. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

344

AR

AR206-ME55-20.tex

RAGHUNATHAN

¥

AR206-ME55-20.sgm

BERNHARDT

¥

LaTeX2e(2002/01/18)

P1: GBC

ROSENSTEIN

serogroup Y and serogroup W-135 meningococcal polysaccharides has not been established, although immunogenicity has been demonstrated (9). The utility of meningococcal polysaccharide vaccines is further restricted because they do not sustainably reduce meningococcal carriage (102, 107, 108) and therefore do not lead to herd immunity. Furthermore, repeated immunization with the serogroup A (109, 110) and serogroup C (111–113) polysaccharide has induced immunologic hyporesponsiveness in children and adults, although the clinical relevance of these findings is unknown. In summary, plain meningococcal polysaccharide vaccine is not considered for routine use in the general population because of its poor immunogenicity in children, short duration of protection, and inability to induce herd immunity. Despite these limitations, in the United States the quadrivalent meningococcal polysaccharide vaccine is useful for certain high-risk groups, such as military recruits, laboratory workers exposed to N. meningitidis, persons with asplenia or complement deficiencies, and travelers to highly endemic or epidemic areas (7, 114). Freshmen living in dormitories have a modestly increased risk of invasive meningococcal disease (115, 116). Because studies demonstrated that 68% of cases in college students were vaccine-preventable, ACIP recommended that college freshmen, especially those who live in dormitories, receive education about meningococcal disease and the quadrivalent meningococcal vaccine (117).

Conjugate Meningococcal Polysaccharide Vaccines Conjugate vaccine technology can overcome the immunologic limitations of meningococcal polysaccharide vaccines, which provoke T-cell–independent responses. When the capsular polysaccharide antigen is conjugated to a protein carrier, a T-cell–dependent host immune response develops, resulting in long-lasting protection and immunologic memory even in infants. This technology was first successfully exploited for the H. influenzae serotype b (Hib) conjugate vaccine, which has reduced the US burden of Hib disease by 99% in children less than five years of age (118). This remarkable decline can partly be attributed to herd immunity: Hib vaccine also reduces nasopharyngeal carriage in vaccinated individuals, thereby lowering disease transmission and indirectly benefiting unvaccinated individuals (118). A pneumococcal conjugate vaccine was licensed in February 2000 in the United States; it has already substantially reduced the rate of invasive disease caused by Streptococcus pneumoniae among toddlers and may also be reducing the rate in adults (119). Using the same technology, serogroup A, C, Y, and W-135 polysaccharides have been conjugated to tetanus toxoid and CRM197 proteins. The safety and immunogenicity of bivalent A+C and monovalent C conjugate vaccines have been demonstrated among infants and adults in the United States, England, and Africa (120–123). Because of the relatively low burden of endemic meningococcal disease, clinical efficacy trials are difficult to implement in industrialized countries. In the United Kingdom, meningococcal serogroup C conjugate vaccines were

2 Dec 2003

21:47

AR

AR206-ME55-20.tex

AR206-ME55-20.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:333-353. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

MENINGOCOCCAL DISEASE CONTROL

P1: GBC

345

licensed based on immunologic data in 1999 and introduced in the routine infant immunization schedule (124). A mass “catchup” vaccination campaign also targeted all persons under the age of 18 years (124). The serogroup C vaccine efficacy was ∼90% among all age groups, and two years after the introduction of the vaccine, serogroup C disease incidence declined 87% among vaccinees (125, 126). Moreover, carriage of serogroup C meningococci among teenagers decreased 66% within one year of vaccination (127), and disease decreased 34%–61% among unvaccinated individuals (125). Carriage of other meningococcal serogroups was unaffected (127). These exciting results indicate that serogroup C conjugate vaccines provide serogroup-specific protection against meningococcal carriage and have at least a short-term impact on herd immunity, although the duration of this effect remains to be seen. In addition, the length of protection and need for a booster dose will need to be evaluated in all age groups, particularly in infants. Potential complications of the vaccine implementation strategy include the emergence of replacement disease due to other serogroups and the development of capsule switching, as has been documented for serogroups B and C (128, 129). Thus far, the United Kingdom has not reported either of these problems, although surveillance is ongoing (125). Several other countries in Europe, as well as Canada and Australia, are in the process of implementing serogroup C conjugate vaccine programs. A quadrivalent conjugate polysaccharide A/C/Y/W-135 vaccine has recently been shown to be safe and immunogenic in healthy adults and may eventually become available in the United States (130).

Serogroup B Vaccines The serogroup B capsular polysaccharide is poorly immunogenic in humans because it resembles a self-antigen (11). However, because serogroup B N. meningitidis causes about one third of meningococcal disease in the United States (21) and can cause outbreaks (45, 46), a serogroup B vaccine is critical for long-term control. Serogroup B vaccine development has focused on subcapsular antigens, using preparations of outer membrane proteins (OMPs) from epidemic strains (12). OMP vaccines have been moderately useful in the control of native epidemics caused by the homologous vaccine strain, but they have had limited to no efficacy in young children and infants (131, 132). Moreover, OMP vaccines have failed to induce protective responses against heterologous serogroup B strains (133). Because of the diversity of OMPs associated with endemic disease, this approach may be best suited for the development of designer vaccines for outbreaks (134, 135). Because OMP vaccines produce poor cross-protective immune responses and low efficacy in young children, novel serogroup B vaccine strategies are being explored. In 2000, the genome of a virulent serogroup B meningococcal strain was sequenced (136), and a functional screen of open reading frames yielded seven novel surface-exposed proteins with the potential to elicit bactericidal immune responses in mice (137). Further studies will determine whether any of these

2 Dec 2003

21:47

346

AR

AR206-ME55-20.tex

RAGHUNATHAN

¥

AR206-ME55-20.sgm

BERNHARDT

¥

LaTeX2e(2002/01/18)

P1: GBC

ROSENSTEIN

proteins will be immunogenic and efficacious in humans, but this genome-based strategy is one of multiple approaches to serogroup B vaccine development (12).

Annu. Rev. Med. 2004.55:333-353. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

PROSPECTS FOR THE CONTROL OF MENINGOCOCCAL DISEASE IN THE UNITED STATES Although most meningococcal disease in the United States is endemic, meningococcal outbreaks often create public fear and panic and consequently command disproportionate attention and resources. Currently, two strategies exist for controlling meningococcal disease outbreaks: antimicrobial chemoprophylaxis and polysaccharide vaccines. Unfortunately, these approaches do not significantly reduce the overall burden of meningococcal disease. To accomplish this objective, new tools are needed. Meningococcal conjugate vaccines will soon be available in the United States, but complicated questions remain about formulations, target age groups, and combinations with other vaccines. Serogroups A and W-135 are rare in the United States, but the occurrence of international outbreaks and the potential for imported disease suggest that the broadest possible vaccine formulation would be preferable. Use of conjugate vaccines in infants, toddlers, or adolescents could have a substantial impact on disease (138). If conjugate meningococcal vaccines reduce carriage and thus create herd immunity, immunizing adolescents, who have the highest carriage rates, might rapidly reduce transmission. Finally, because of the already crowded infant immunization schedule, multiple combination vaccines are being explored. The significant presence of serogroup B disease also requires the development and implementation of serogroup B vaccines, which are likely to have different immunologic and epidemiologic properties from the conjugate proteinpolysaccharide antigens. In the long run, serogroup-specific vaccines may not be the final solution, and the pendulum may shift toward common protein vaccines that protect against all pathogenic meningococcal serogroups (12). Improved surveillance and diagnostic techniques will become increasingly important to monitor trends in meningococcal disease epidemiology after the introduction of these much-anticipated vaccines in the United States. The Annual Review of Medicine is online at http://med.annualreviews.org

LITERATURE CITED 1. Brundage JF, Ryan MA, Feighner BH, et al. 2002. Meningococcal disease among United States military service members in relation to routine uses of vaccines with different serogroup-specific components,

1964–1998. Clin. Infect. Dis. 35:1376– 81 2. Pickett WH. 1931. An epidemic of cerebrospinal meningitis in Saginaw, Michigan. Am. J. Public Health 21:139–46

2 Dec 2003

21:47

AR

AR206-ME55-20.tex

AR206-ME55-20.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:333-353. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

MENINGOCOCCAL DISEASE CONTROL 3. Lee WW. 1931. Epidemic meningitis in Indianapolis, 1929–1930. J. Prevent. Med. 5:203–9 4. Brundage JF, Zollinger WD. 1987. Evolution of meningococcal disease epidemiology in the U.S Army. In Evolution of Meningococcal Disease, ed. NA Vedros, 1:5–23. Boca Raton, FL: CRC Press 5. Rosenstein NE, Perkins BA. 2000. Update on Haemophilus influenzae serotype b and meningococcal vaccines. Pediatr. Clin. North Am. 47:337–52, vi 6. Connolly M, Noah N. 1999. Is group C meningococcal disease increasing in Europe? A report of surveillance of meningococcal infection in Europe 1993– 6. European Meningitis Surveillance Group. Epidemiol. Infect. 122:41–49 7. CDC. 1997. Control and prevention of meningococcal disease: recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR Recomm. Rep. 46:1–10 8. Gotschlich EC, Goldschneider I, Artenstein MS. 1969. Human immunity to the meningococcus. IV. Immunogenicity of group A and group C meningococcal polysaccharides. J. Exp. Med. 129:1367– 84 9. Hankins WA, Gwaltney JM Jr, Hendley JO, et al. 1982. Clinical and serological evaluation of a meningococcal polysaccharide vaccine groups A, C, Y, W135. Proc. Soc. Exp. Biol. Med. 169:54–57 10. Goldschneider I, Gotschlich EC, Artenstein MS. 1969. Human immunity to the meningococcus. I. The role of humoral antibodies. J. Exp. Med. 129:1307–26 11. Finne J, Bitter-Suermann D, Goridis C, et al. 1987. An IgG monoclonal antibody to group B meningococci crossreacts with developmentally regulated polysialic acid units of glycoproteins in neural and extraneural tissues. J. Immunol. 138:4402–7 12. Morley SL, Pollard AJ. 2001. Vaccine prevention of meningococcal disease, coming soon? Vaccine 20:666–87

P1: GBC

347

13. Greenfield S, Sheehe PR, Feldman HA. 1971. Meningococcal carriage in a population of “normal” families. J. Infect. Dis. 123:67–73 14. Blakebrough IS, Greenwood BM, Whittle HC, et al. 1982. The epidemiology of infections due to Neisseria meningitidis and Neisseria lactamica in a northern Nigerian community. J. Infect. Dis. 146:626– 37 15. Fraser PK, Bailey GK, Abbott JD, et al. 1973. The meningococcal carrier-rate. Lancet 1:1235–37 16. Goldschneider I, Gotschlich EC, Artenstein MS. 1969. Human immunity to the meningococcus. II. Development of natural immunity. J. Exp. Med. 129:1327–48 17. Cartwright KA, Stuart JM, Jones DM, et al. 1987. The Stonehouse survey: nasopharyngeal carriage of meningococci and Neisseria lactamica. Epidemiol. Infect. 99:591–601 18. Shepard CW, Rosenstein NE, Fischer M, et al. 2003. Neonatal meningococcal disease in the United States, 1990 to 1999. Pediatr. Infect. Dis. J. 22:418–22 19. Glode MP, Robbins JB, Liu TY, et al. 1977. Cross-antigenicity and immunogenicity between capsular polysaccharides of group C Neisseria meningitidisof Escherichia coli K92. J. Infect. Dis. 135:94–104 20. Kasper DL, Winkelhake JL, Zollinger WD, et al. 1973. Immunochemical similarity between polysaccharide antigens of Escherichia coli 07: K1(L):NM and group B Neisseria meningitidis. J. Immunol. 110:262–68 21. Rosenstein NE, Perkins BA, Stephens DS, et al. 1999. The changing epidemiology of meningococcal disease in the United States, 1992–1996. J. Infect. Dis. 180:1894–901 22. van Deuren M, Brandtzaeg P, van der Meer JW. 2000. Update on meningococcal disease with emphasis on pathogenesis and clinical management. Clin. Microbiol. Rev. 13:144–66

2 Dec 2003

21:47

Annu. Rev. Med. 2004.55:333-353. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

348

AR

AR206-ME55-20.tex

RAGHUNATHAN

¥

AR206-ME55-20.sgm

BERNHARDT

23. Kirsch EA, Barton RP, Kitchen L, et al. 1996. Pathophysiology, treatment and outcome of meningococcemia: a review and recent experience. Pediatr. Infect. Dis. J. 15:967–79 24. Koppes GM, Ellenbogen C, Gebhart RJ. 1977. Group Y meningococcal disease in United States Air Force recruits. Am. J. Med. 62:661–66 25. Winstead JM, McKinsey DS, Tasker S, et al. 2000. Meningococcal pneumonia: characterization and review of cases seen over the past 25 years. Clin. Infect. Dis. 30:87–94 26. Edwards MS, Baker CJ. 1981. Complications and sequelae of meningococcal infections in children. J. Pediatr. 99:540–45 27. Tzeng YL, Stephens DS. 2000. Epidemiology and pathogenesis of Neisseria meningitidis. Microbes Infect. 2:687–700 28. Kaiser AB, Hennekens CH, Saslaw MS, et al. 1974. Seroepidemiology and chemoprophylaxis of disease due to sulfonamide-resistant Neisseria meningitidis in a civilian population. J. Infect. Dis. 130: 217–24 29. Stanwell-Smith RE, Stuart JM, Hughes AO, et al. 1994. Smoking, the environment and meningococcal disease: a case control study. Epidemiol. Infect. 112:315– 28 30. Fischer M, Hedberg K, Cardosi P, et al. 1997. Tobacco smoke as a risk factor for meningococcal disease. Pediatr. Infect. Dis. J. 16:979–83 31. Moodley JR, Coetzee N, Hussey. G 1999. Risk factors for meningococcal disease in Cape Town. S. Afr. Med. J. 89:56–59 32. Baker M, McNicholas A, Garrett N, et al. 2000. Household crowding a major risk factor for epidemic meningococcal disease in Auckland children. Pediatr. Infect. Dis. J. 19:983–90 33. Jackson LA, Wenger. JD 1993. Laboratory-based surveillance for meningococcal disease in selected areas, United States, 1989–1991. MMWR CDC Surveill. Summ. 42:21–30

¥

LaTeX2e(2002/01/18)

P1: GBC

ROSENSTEIN 34. Pearce MC, Sheridan JW, Jones DM, et al. 1995. Control of group C meningococcal disease in Australian aboriginal children by mass rifampicin chemoprophylaxis and vaccination. Lancet 346:20– 23 35. Haneberg B, Tonjum T, Rodahl K, et al. 1983. Factors preceding the onset of meningococcal disease, with special emphasis on passive smoking, symptoms of ill health. NIPH Ann. 6:169–73 36. Yusuf HR, Rochat RW, Baughman WS, et al. 1999. Maternal cigarette smoking and invasive meningococcal disease: a cohort study among young children in metropolitan Atlanta, 1989–1996. Am. J. Public Health 89:712–17 37. Young LS, LaForce FM, Head JJ, et al. 1972. A simultaneous outbreak of meningococcal and influenza infections. N. Engl. J. Med. 287:5–9 38. Moore PS, Hierholzer J, DeWitt W, et al. 1990. Respiratory viruses and mycoplasma as cofactors for epidemic group A meningococcal meningitis. JAMA 264:1271–75 39. Figueroa J, Andreoni J, Densen P. 1993. Complement deficiency states and meningococcal disease. Immunol. Res. 12:295–311 40. Ross SC, Densen P. 1984. Complement deficiency states and infection: epidemiology, pathogenesis and consequences of neisserial and other infections in an immune deficiency. Medicine (Baltimore) 63:243–73 41. Francke EL, Neu HC. 1981. Postsplenectomy infection. Surg. Clin. North Am. 61:135–55 42. Pollard AJ, Frasch C. 2001. Development of natural immunity to Neisseria meningitidis. Vaccine 19:1327–46 43. Rosenstein NE, Perkins BA, Stephens DS, et al. 2001. Meningococcal disease. N. Engl. J. Med. 344:1378–88 44. CDC. 2003. Summary of notifiable diseases, United States, 2001. Morbid. Mortal. Wkly. Rep. 50:1–108

2 Dec 2003

21:47

AR

AR206-ME55-20.tex

AR206-ME55-20.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:333-353. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

MENINGOCOCCAL DISEASE CONTROL 45. Centers for Disease Control and Prevention. 1995. Serogroup B meningococcal disease—Oregon, 1994. Morbid. Mortal. Wkly. Rep. 44:121–24 46. Diermayer M, Hedberg K, Hoesly F, et al. 1999. Epidemic serogroup B meningococcal disease in Oregon: the evolving epidemiology of the ET-5 strain. JAMA 281:1493–97 47. Schuchat A, Hilger T, Zell E, et al. 2001. Active Bacterial Core surveillance of the Emerging Infections Program network. Emerg. Infect. Dis. 7:92–99 48. Pollard AJ, Scheifele D. 2001. Meningococcal disease and vaccination in North America. J. Paediatr. Child Health 37: S20–S27 49. Racoosin JA, Whitney CG, Conover CS, et al. 1998. Serogroup Y meningococcal disease in Chicago, 1991–1997. JAMA 280:2094–98 50. 2001. Epidemics of meningococcal disease. African meningitis belt, 2001. Wkly. Epidemiol. Rec. 76:282–88 51. Bertherat E, Yada A, Djingarey MH, et al. 2002. First major epidemic caused by Neisseria meningitidis serogroup W135 in Africa? Med. Trop. 62:301–4 52. Popovic T, Sacchi CT, Reeves MW, et al. 2000. Neisseria meningitidis serogroup W135 isolates associated with the ET-37 complex. Emerg. Infect. Dis. 6:428–29 53. Lingappa JR, Al-Rabeah AM, Hajjeh R, et al. 2003. Serogroup W-135 meningococcal disease during the Hajj, 2000. Emerg. Infect. Dis. 9:665–71 54. van Deuren M, van Dijke BJ, Koopman RJ, et al. 1993. Rapid diagnosis of acute meningococcal infections by needle aspiration or biopsy of skin lesions. BMJ 306:1229–32 55. Camargos PA, Almeida MS, Cardoso I, et al. 1995. Latex particle agglutination test in the diagnosis of Haemophilus influenzae type B, Streptococcus pneumoniae and Neisseria meningitidis A and C meningitis in infants and children. J. Clin. Epidemiol. 48:1245–50

P1: GBC

349

56. Tarafdar KRS, Recco RA, Zaman MM. 2001. Lack of sensitivity of the latex agglutination test to detect bacterial antigen in the cerebrospinal fluid of patients with culture-negative meningitis. Clin. Infect. Dis. 33:406–8 57. Perkins MD, Mirrett S, Reller LB. 1995. Rapid bacterial antigen detection is not clinically useful. J. Clin. Microbiol. 33:1486–91 58. Gray SJ, Sobanski MA, Kaczmarski EB, et al. 1999. Ultrasound-enhanced latex immunoagglutination and PCR as complementary methods for non-culturebased confirmation of meningococcal disease. J. Clin. Microbiol. 37:1797–801 59. Fox AJ. 2001. Nucleic acid technologies and meningococcal infection. J. Infect. 42:100–3 60. Maiden MC, Frosch M. 2001. Molecular techniques for the investigation of meningococcal disease epidemiology. Mol. Biotechnol. 18:119–34 61. Porritt RJ, Mercer JL, Munro R. 2000. Detection and serogroup determination of Neisseria meningitidis in CSF by polymerase chain reaction (PCR). Pathology 32:42–45 62. Borrow R, Claus H, Guiver M, et al. 1997. Non-culture diagnosis and serogroup determination of meningococcal B and C infection by a sialyltransferase (siaD) PCR ELISA. Epidemiol. Infect. 118:111– 17 63. Borrow R, Claus H, Chaudhry U, et al. 1998. siaD PCR ELISA for confirmation and identification of serogroup Y and W135 meningococcal infections. FEMS Microbiol. Lett. 159:209–14 64. Guiver M, Borrow R, Marsh J, et al. 2000. Evaluation of the Applied Biosystems automated Taqman polymerase chain reaction system for the detection of meningococcal DNA. FEMS Immunol. Med. Microbiol. 28:173–79 65. Molling P, Jacobsson S, Backman A, et al. 2002. Direct and rapid identification and genogrouping of meningococci and

2 Dec 2003

21:47

350

66.

Annu. Rev. Med. 2004.55:333-353. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

67.

68.

69.

70.

71.

72.

73.

74.

AR

AR206-ME55-20.tex

RAGHUNATHAN

¥

AR206-ME55-20.sgm

BERNHARDT

porA amplification by LightCycler PCR. J. Clin. Microbiol. 40:4531–35 Selander RK, Caugant DA, Ochman H, et al. 1986. Methods of multilocus enzyme electrophoresis for bacterial population genetics and systematics. Appl. Environ. Microbiol. 51:873–84 Caugant DA, Froholm LO, Bovre K, et al. 1986. Intercontinental spread of a genetically distinctive complex of clones of Neisseria meningitidis causing epidemic disease. Proc. Natl. Acad. Sci. USA 83:4927–31 Lystad A, Aasen S. 1991. The epidemiology of meningococcal disease in Norway 1975–91. NIPH Ann. 14:57–65; discussion 65–66 Popovic T, Schmink S, Rosenstein NA, et al. 2001. Evaluation of pulsed-field gel electrophoresis in epidemiological investigations of meningococcal disease outbreaks caused by Neisseria meningitidis serogroup C. J. Clin. Microbiol. 39:75– 85 Maiden MC, Bygraves JA, Feil E, et al. 1998. Multilocus sequence typing: a portable approach to the identification of clones within populations of pathogenic microorganisms. Proc. Natl. Acad. Sci. USA 95:3140–45 Yakubu DE, Pennington TH. 1995. Epidemiological evaluation of Neisseria meningitidis serogroup B by pulsed-field gel electrophoresis. FEMS Immunol. Med. Microbiol. 10:185–89 Swaminathan B, Matar GM, Reeves MW, et al. 1996. Molecular subtyping of Neisseria meningitidis serogroup B: comparison of five methods. J. Clin. Microbiol. 34:1468–73 Sacchi CT, Whitney AM, Reeves MW, et al. 2002. Sequence diversity of Neisseria meningitidis 16S rRNA genes and use of 16S rRNA gene sequencing as a molecular subtyping tool. J. Clin. Microbiol. 40:4520–27 Munford RS, Taunay AdE, de Morais JS, et al. 1974. Spread of meningococ-

¥

LaTeX2e(2002/01/18)

P1: GBC

ROSENSTEIN

75.

76.

77.

78.

79.

80.

81.

82.

83.

cal infection within households. Lancet 1:1275–78 Olcen P, Kjellander J, Danielsson D, et al. 1981. Epidemiology of Neisseria meningitidis; prevalence and symptoms from the upper respiratory tract in family members to patients with meningococcal disease. Scand. J. Infect. Dis. 13:105–9 Broome CV. 1986. The carrier state: Neisseria meningitidis. J. Antimicrob. Chemother. 18:25–34 Meningococcal Disease Surveillance Group. 1974. Meningococcal disease: secondary attack rate and chemoprophylaxis in the United States. JAMA 235:261– 65 Meningococcal Disease Surveillance Group. 1976. Analysis of endemic meningococcal disease by serogroup and evaluation of chemoprophylaxis. J. Infect. Dis. 134:201–4 Hastings L, Stuart J, Andrews N, et al. 1997. A retrospective survey of clusters of meningococcal disease in England and Wales, 1993 to 1995: estimated risks of further cases in household and educational settings. Commun. Dis. Rep. CDR Rev. 7:R195–R200 De Wals P, Hertoghe L, Borlee-Grimee I, et al. 1981. Meningococcal disease in Belgium. Secondary attack rate among household, day-care nursery and preelementary school contacts. J. Infect. 3: 53–61 Zangwill KM, Schuchat A, Riedo FX, et al. 1997. School-based clusters of meningococcal disease in the United States. Descriptive epidemiology and a case-control analysis. JAMA 277:389– 95 Gilmore A, Stuart J, Andrews N. 2000. Risk of secondary meningococcal disease in health-care workers. Lancet 356:1654– 55 American Academy of Pediatrics. 2003. Meningococcal infections. In Red Book: 2003 Report of the Committee on Infectious Diseases, ed. LK Pickering,

2 Dec 2003

21:47

AR

AR206-ME55-20.tex

AR206-ME55-20.sgm

LaTeX2e(2002/01/18)

MENINGOCOCCAL DISEASE CONTROL

84.

Annu. Rev. Med. 2004.55:333-353. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

85.

86.

87.

88.

89.

90.

91.

92.

pp. 430–36. Elk Grove Village, IL: Am. Acad. Pediatr. 26th ed. Jackson LA, Alexander ER, DeBolt CA, et al. 1996. Evaluation of the use of mass chemoprophylaxis during a school outbreak of enzyme type 5 serogroup B meningococcal disease. Pediatr. Infect. Dis. J. 15:992–98 Shehab S, Keller N, Barkay A, et al. 1998. Failure of mass antibiotic prophylaxis to control a prolonged outbreak of meningococcal disease in an Israeli village. Eur. J. Clin. Microbiol. Infect. Dis. 17:749–53 Devine LF, Johnson DP, Hagerman CR, et al. 1970. Rifampin: levels in serum and saliva and effect on the meningococcal carrier state. JAMA 214:1055–59 Gaunt PN, Lambert BE. 1988. Singledose ciprofloxacin for the eradication of pharyngeal carriage of Neisseria meningitidis. J. Antimicrob. Chemother. 21:489– 96 Dworzack DL, Sanders CC, Horowitz EA, et al. 1988. Evaluation of single-dose ciprofloxacin in the eradication of Neisseria meningitidis from nasopharyngeal carriers. Antimicrob. Agents Chemother. 32:1740–41 1997. Control and prevention of serogroup C meningococcal disease: evaluation and management of suspected outbreaks: recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR Recomm. Rep. 46: 13–21 Cuevas LE, Hart CA. 1993. Chemoprophylaxis of bacterial meningitis. J Antimicrob. Chemother. 31(Suppl. B):79–91 Schwartz B, Al-Tobaiqi A, Al-Ruwais A, et al. 1988. Comparative efficacy of ceftriaxone and rifampicin in eradicating pharyngeal carriage of group A Neisseria meningitidis. Lancet 1:1239–42 Girgis N, Sultan Y, Frenck RW, et al. 1998. Azithromycin compared with rifamin for eradication of nasopharyngeal colonization by Neisseria meningitidis. Pediatr. Infect. Dis. J. 17:816–19

P1: GBC

351

93. Artenstein MS, Gold R, Zimmerly JG, et al. 1970. Prevention of meningococcal disease by group C polysaccharide vaccine. N. Engl. J. Med. 282:417–20 94. Gold R, Artenstein MS. 1971. Meningococcal infections. II. Field trial of group C meningococcal polysaccharide vaccine in 1969–70. Bull. World Health Org. 45:279–82 95. Rosenstein N, Levine O, Taylor JP, et al. 1998. Efficacy of meningococcal vaccine and barriers to vaccination. JAMA 279:435–39 96. Goldschneider I, Lepow ML, Gotschlich EC. 1972. Immunogenicity of the group A and group C meningococcal polysaccharides in children. J. Infect. Dis. 125:509– 19 97. Goldschneider I, Lepow ML, Gotschlich EC, et al. 1973. Immunogenicity of group A and group C meningococcal polysaccharides in human infants. J. Infect. Dis. 128:769–76 98. Gold R, Lepow ML, Goldschneider I, et al. 1979. Kinetics of antibody production to group A and group C meningococcal polysaccharide vaccines administered during the first six years of life: prospects for routine immunization of infants and children. J. Infect. Dis. 140:690–97 99. Zangwill KM, Stout RW, Carlone GM, et al. 1994. Duration of antibody response after meningococcal polysaccharide vaccination in US Air Force personnel. J. Infect. Dis. 169:847–52 100. Makela PH, Kayhty H, Weckstrom P, et al. 1975. Effect of group-A meningococcal vaccine in army recruits in Finland. Lancet 2:883–86 101. Peltola H, Makela PH, Kayhty H, et al. 1977. Clinical efficacy of meningococcus group A capsular polysaccharide vaccine in children three months to five years of age. N. Engl. J. Med. 297:686–91 102. Bosmans E, Vimont-Vicary FE, Andre PJ, et al. 1980. Protective efficacy of a bivalent (A + C) meningococcal vaccine during a cerebrospinal menigitis epidemic

2 Dec 2003

21:47

352

103.

Annu. Rev. Med. 2004.55:333-353. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

104.

105.

106.

107.

108.

109.

110.

111.

112.

AR

AR206-ME55-20.tex

RAGHUNATHAN

¥

AR206-ME55-20.sgm

BERNHARDT

in Rwanda. Ann. Soc. Belge Med. Trop. 60:297–306 Binkin N, Band J. 1982. Epidemic of meningococcal meningitis in Bamako, Mali: epidemiological features and analysis of vaccine efficacy. Lancet 2:315– 18 Lennon D, Gellin B, Hood D, et al. 1992. Successful intervention in a group A meningococcal outbreak in Auckland, New Zealand. Pediatr. Infect Dis. J. 11:617–23 Gold R, Lepow ML, Goldschneider I, et al. 1977. Immune response of human infants of polysaccharide vaccines of group A and C Neisseria meningitidis. J. Infect. Dis. 136:S31–S35 (Suppl.) Reingold AL, Broome CV, Hightower AW, et al. 1985. Age-specific differences in duration of clinical protection after vaccination with meningococcal polysaccharide A vaccine. Lancet 2:114–18 Hassan-King MK, Wall RA, Greenwood BM. 1988. Meningococcal carriage, meningococcal disease and vaccination. J. Infect. 16:55–59 Moore PS, Harrison LH, Telzak EE, et al. 1988. Group A meningococcal carriage in travelers returning from Saudi Arabia. JAMA 260:2686–89 MacLennan J, Obaro S, Deeks J, et al. 1999. Immune response to revaccination with meningococcal A and C polysaccharides in Gambian children following repeated immunisation during early childhood. Vaccine 17:3086–93 Borrow R, Joseph H, Andrews N, et al. 2001. Reduced antibody response to revaccination with meningococcal serogroup A polysaccharide vaccine in adults. Vaccine 19:1129–32 MacDonald NE, Halperin SA Law BJ, et al. 1998. Induction of immunologic memory by conjugated vs plain meningococcal C polysaccharide vaccine in toddlers: a randomized controlled trial. JAMA 280:1685–89 Granoff DM, Gupta RK, Belshe RB, et al.

¥

LaTeX2e(2002/01/18)

P1: GBC

ROSENSTEIN

113.

114.

115.

116.

117.

118.

119.

120.

121.

122.

1998. Induction of immunologic refractoriness in adults by meningococcal C polysaccharide vaccination. J. Infect. Dis. 178:870–74 Richmond P, Kaczmarski E, Borrow R, et al. 2000. Meningococcal C polysaccharide vaccine induces immunologic hyporesponsiveness in adults that is overcome by meningococcal C conjugate vaccine. J. Infect. Dis. 181:761–64 CDC. 2002. Laboratory-acquired meningococcal disease—United States, 2000. Morbid. Mortal. Wkly. Rep. 51:141–44 Harrison LH, Dwyer DM, Maples CT, et al. 1999. Risk of meningococcal infection in college students. JAMA 281:1906– 10 Bruce MG, Rosenstein NE, Capparella JM, et al. 2001. Risk factors for meningococcal disease in college students. JAMA 286:688–93 CDC. 2000. Meningococcal disease and college students. Recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR Recomm. Rep. 49:13–20 CDC. 1998. Progress toward eliminating Haemophilus influenzae type b disease among infants and children—United States, 1987–1997. Morbid. Mortal. Wkly. Rep. 47:993–98 Whitney CW. 2002. The potential of pneumococcal conjugate vaccines for children. Pediatr, Infect. Dis. J. 21:961–70 Lieberman JM, Chiu SS, Wong VK, et al. 1996. Safety and immunogenicity of a serogroups A/C Neisseria meningitidis oligosaccharide-protein conjugate vaccine in young children. JAMA 275:1499– 1503 MacLennan JM, Shackley F, Heath PT, et al. 2000. Safety, immunogenicity, and induction of immunologic memory by a serogroup C meningococcal conjugate vaccine in infants: a randomized controlled trial. JAMA 283:2795–2801 Campagne G, Garba A, Fabre P, et al. 2000. Safety and immunogenicity of three

2 Dec 2003

21:47

AR

AR206-ME55-20.tex

AR206-ME55-20.sgm

LaTeX2e(2002/01/18)

MENINGOCOCCAL DISEASE CONTROL

123.

Annu. Rev. Med. 2004.55:333-353. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

124.

125.

126.

127.

128.

129.

130.

doses of a Neisseria meningitidis A + C diphtheria conjugate vaccine in infants from Niger. Pediatr. Infect. Dis. J. 19:144– 50 Twumasi PA, Kumah S, Leach A, et al. 1995. A trial of group A plus group C meningococcal polysaccharide-protein conjugate vaccine in African infants. J. Infect. Dis. 171:632–38 Miller E, Salisbury D, Ramsay M. 2001. Planning, registration, and implementation of an immunisation campaign against meningococcal serogroup C disease in the UK: a success story. Vaccine 20(Suppl. 1):S58–S67 Balmer P, Borrow R, Miller E. 2002. Impact of meningococcal C conjugate vaccine in the UK. J. Med. Microbiol. 51:717–22 Bose A, Coen P, Tully J, et al. 2003. Effectiveness of meningococcal C conjugate vaccine in teenagers in England. Lancet 361:675–76 Maiden MC, Stuart JM. 2002. Carriage of serogroup C meningococci 1 year after meningococcal C conjugate polysaccharide vaccination. Lancet 359:1829–31 Swartley JS, Marfin AA, Edupuganti S, et al. 1997. Capsule switching of Neisseria meningitidis. Proc. Natl. Acad. Sci. USA 94:271–76 Kertesz DA, Coulthart MB, Ryan JA, et al. 1998. Serogroup B, electrophoretic type 15 Neisseria meningitidis in Canada. J. Infect. Dis. 177:1754–57 Campbell JD, Edelman R, King JC Jr, et al. 2002. Safety, reactogenicity, and immunogenicity of a tetravalent meningococcal polysaccharide-diphtheria toxoid conjugate vaccine given to healthy adults. J. Infect. Dis. 186:1848–51

P1: GBC

353

131. Sierra GV, Campa HC, Varcacel NM, et al. 1991. Vaccine against group B Neisseria meningitidis: protection trial and mass vaccination results in Cuba. NIPH Ann. 14:195–207; discussion 208–10 132. Bjune G, Hoiby EA, Gronnesby JK, et al. 1991. Effect of outer membrane vesicle vaccine against group B meningococcal disease in Norway. Lancet 338:1093–96 133. Tappero JW, Lagos R, Ballesteros AM, et al. 1999. Immunogenicity of 2 serogroup B outer-membrane protein meningococcal vaccines: a randomized controlled trial in Chile. JAMA 281:1520– 27 134. Sacchi CT, Whitney AM, Popovic T, et al. 2000. Diversity and prevalence of PorA types in Neisseria meningitidis serogroup B in the United States, 1992–1998. J. Infect. Dis. 182:1169–76 135. Sacchi CT, Lemos AP, Popovic T, et al. 2001. Serosubtypes and PorA types of Neisseria meningitidis serogroup B isolated in Brazil during 1997–1998: overview and implications for vaccine development. J. Clin. Microbiol. 39:2897– 903 136. Tettelin H, Saunders NJ, Heidelberg J, et al. 2000. Complete genome sequence of Neisseria meningitidis serogroup B strain MC58. Science 287:1809–15 137. Pizza M, Scarlato V, Masignani V, et al. 2000. Identification of vaccine candidates against serogroup B meningococcus by whole-genome sequencing. Science 287:1816–20 138. Lingappa JR, Rosenstein N, Zell ER, et al. 2001. Surveillance for meningococcal disease and strategies for use of conjugate meningococcal vaccines in the United States. Vaccine 19:4566–75

12 Dec 2003

14:11

AR

AR206-ME55-21.tex

AR206-ME55-21.sgm

LaTeX2e(2002/01/18) P1: GBC 10.1146/annurev.med.55.091902.104344

Annu. Rev. Med. 2004. 55:355–72 doi: 10.1146/annurev.med.55.091902.104344 c 2004 by Annual Reviews. All rights reserved Copyright ° First published online as a Review in Advance on Dec. 10, 2003

Annu. Rev. Med. 2004.55:355-372. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

RECENT ADVANCES IN THE DEVELOPMENT OF HIV-1 VACCINES USING REPLICATION-INCOMPETENT ADENOVIRUS VECTORS John W. Shiver and Emilio A. Emini Vaccine Research, Merck Research Laboratories, West Point, Pennsylvania 19486; email: john [email protected]; emilio [email protected]

Key Words CTL, AIDS, heterologous prime/boost immunization ■ Abstract An increasing body of evidence suggests that a vaccine that elicits anti-HIV-1 cellular immunity could provide the basis for an effective AIDS vaccine. Comparative immunization experiments testing a variety of vaccine approaches have demonstrated that replication-incompetent adenovirus vectors are an effective means for eliciting cytotoxic T-lymphocyte (CTL) immune responses against HIV-1 antigens. These immune responses effectively control viremia in nonhuman primates following challenge with simian AIDS viruses. Such data, coupled with epidemiology studies that identify HIV-1 gag, pol, and nef as the best antigens for broadly directed cellular immune responses, provide guidance for the development of a potential AIDS vaccine.

INTRODUCTION An effective HIV-1 vaccine is urgently needed as the only means for arresting the worldwide AIDS epidemic. There are currently ∼15,000 new HIV-1 infections each day despite identification of high-risk activities and improved efforts to inform people of how to minimize the possibility of becoming infected (1). Potent, novel antiretroviral drugs have been developed and introduced into clinical usage over the past few years (2). Combinations of these drugs have produced dramatic results as treatments of HIV-1 infection. However, multiple-drug therapy has significant limitations. Anti-HIV-1 retroviral therapy does not cure infection, drug adherence is challenging, adverse events are common, and the costs of therapy are considerable, especially in developing countries. Early efforts to develop a preventive vaccine focused on eliciting virusneutralizing antibodies. Unfortunately, a large body of research demonstrated that recombinant HIV-1 env protein (e.g., gp120) and V3 peptide immunogens did not elicit antibodies capable of neutralizing primary viral isolates in either 0066-4219/04/0218-0355$14.00

355

12 Dec 2003

14:11

Annu. Rev. Med. 2004.55:355-372. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

356

AR

SHIVER

AR206-ME55-21.tex

¥

AR206-ME55-21.sgm

LaTeX2e(2002/01/18)

P1: GBC

EMINI

preclinical or human clinical studies (3, 4). Even when improved immunogens based on primary isolates were tested, the neutralizing antibodies produced were extremely virus-isolate–specific because of the high variability of gp120 and therefore could not provide broad vaccine coverage (5). The recent failure in a phase III human-vaccine clinical-efficacy trial of a recombinant gp120 protein vaccine based on both laboratory and primary viruses confirmed predictions of the limitations of these immunogens (6). In recent years it has become increasingly apparent that cell-mediated immune responses, particularly those involving virus-specific CD8+ cytotoxic T lymphocytes (CTL), play a key role in control of both acute and chronic HIV-1 infections. The appearance of HIV-1-specific CTL correlates with control of the acute HIV-1 viremia observed early in infection, in contrast to virus-neutralizing antibodies, which appear much later (7, 8). Also, experiments in rhesus monkeys involving ablation of CD8+ T cells have directly demonstrated that these cells are absolutely required for control of the persistent viremia established by the related simian immunodeficiency virus (SIV) (9). Finally, several studies suggest that the elicitation in monkeys of virus-specific CTL can influence the outcome of virus challenge with either SIV or hybrid SIV/HIV-1 viruses known as SHIVs [see discussion below (10–12)]. In these studies, the vaccine-induced cellular immune response does not prevent infection but typically mediates the establishment of a lower virus load compared to unimmunized control animals.

CONSTRUCTION AND CHARACTERIZATION OF ADENOVIRUS TYPE 5 AS A VACCINE VECTOR Basis for Selection of Adenovirus as the Primary Vaccine Vector The generation of CTL responses requires that the immunogen be expressed within antigen-presenting cells (13). The immunogen is then processed into small peptides (typically about nine amino acids in length) by the proteasome complex and translocated into the endoplasmic reticulum/Golgi complex secretory pathway for eventual association with major histocompatibility complex (MHC) class I proteins. CD8+ T lymphocytes recognize antigen in association with class I MHC via the T-cell receptor (TCR). Activation of naive CD8+ T cells into activated effector or memory cells generally requires both TCR engagement of antigen, as described above, and engagement of costimulatory cell-surface molecules. Optimal induction of CTL responses and the establishment of CD8+ cell memory populations also require immunological “help” in the form of cytokines from antigen-specific CD4+ T lymphocytes (14). At present, there are no licensed vaccines specifically designed to elicit cellular immune responses, although several do, particularly live, attenuated viral vaccines (15). Considerable research will be required to develop a vaccine whose efficacy is entirely predicated on producing antiviral cellular immunity. There are numerous

12 Dec 2003

14:11

AR

AR206-ME55-21.tex

AR206-ME55-21.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:355-372. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

ADENOVIRUS VECTORS AS HIV-1 VACCINES

P1: GBC

357

potential approaches for constructing vaccines that elicit CTL responses. We conducted extensive immunization experiments on nonhuman primates in order to compare many of these vaccine approaches and to determine their relative immunogenicity. In order to make these experiments as stringent and definitive as possible, a substantial effort was made to keep variables to a minimum. All vaccines tested were based on the same HIV-1 gag antigen, and quantitative cellular immune response assays were developed and standardized to allow clear discrimination among the tested vaccines. These assays are described in more detail below. Over the course of these preclinical studies, we tested synthetic peptides, plasmid DNA, and recombinant protein antigens prepared in a variety of adjuvants or delivery vehicles, as well as various bacterial and viral vectors including BCG (bacille Calmette-Guerin), adenoviruses, pox viruses, and alphaviruses. The results from these experiments showed that recombinant viral vectors, particularly those based on replication-defective adenoviruses, were most effective in eliciting specific CTL responses. Viral vectors can be constructed to encode genes that express selected antigens from pathogens. When the recombinant vaccine vector is injected into a host, it binds cell types consistent with its known cellular tropism and mediates the expression of the vaccine antigens. The optimal vaccine vector would produce the vaccine antigen in excess of its own proteins (so that the immune response focuses on the target antigen) and would primarily be produced in cells that can serve as antigen-presenting cells for induction of CTL responses. Adenoviruses have a broad cell tropism including professional antigen-presenting cells such as macrophages and immature dendritic cells, and they can enter (if not necessarily replicate in) cells from most animal species (16). Our initial adenovirus vaccine vectors were based on a replication-defective adenovirus type 5 (Ad5). The viral vector was rendered replication-defective by removing the E1 gene, which is required both for adenovirus replication and for downstream viral gene expression. Genes encoding vaccine antigens may be inserted in place of the E1 gene with expression driven by heterologous regulatory elements. Our adenovirus vaccine vectors use the cytomegalovirus (CMV) immediate/early promoter (17) and bovine growth hormone transcription termination/polyadenylation (18) regulatory elements to drive expression of the vaccine antigen gene. These expression-regulatory elements were selected for their ability to mediate high levels of gene expression. The HIV-1 gag, pol, and nef proteins were selected as vaccine antigens on the basis of comprehensive experiments described in greater detail below. Synthetic forms of each gene were constructed using optimal codons for human gene expression (19). We have shown previously, and were the first to demonstrate in the HIV-1 vaccine field, that codon-optimized genes provide much higher gene expression (>100-fold) than wild-type HIV-1 genes in heterologous expression vectors (20). This was a key observation in our early development of vaccines based on viral and DNA vectors. Each synthetic gene was inserted into a different Ad5 vector to create a trivalent vaccine mixture. Figure 1 shows schematic diagrams of the three Ad5 vectors.

12 Dec 2003

14:11

Annu. Rev. Med. 2004.55:355-372. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

358

AR

SHIVER

AR206-ME55-21.tex

¥

AR206-ME55-21.sgm

LaTeX2e(2002/01/18)

P1: GBC

EMINI

Figure 1 Schematic diagrams of replication-incompetent Ad5 HIV-1 gag, pol, and nef vaccine vector genomes. Abbreviations: BGH pA, bovine growth hormone polyadenylation signal; CMV, cytomegalovirus; ORF, open reading frame; 9, packaging signal; ITR, inverted terminal repeat.

Although they are not replication-competent, adenovirus vectors retain the ability to efficiently bind cells, become internalized, and deliver their genome to the nucleus for subsequent expression of encoded antigen genes. E1-deleted adenoviruses can be produced in large quantities in appropriate human cell lines designed to provide the E1 gene product in trans. The most typical cell lines used for E1-deleted adenovirus propagation are 293 and PER.C6TM cells. For our vaccine studies, we have used PER.C6TM, which is a human retinoblastoma cell line transduced with an E1 gene segment that complements production of adenovirus vectors but prevents generation of replication-competent adenovirus through homologous recombination.

Measuring Cellular Immune Responses Against HIV-1 Antigens The immunological objective of our HIV-1 vaccine is to elicit anti-HIV-1 cytotoxic and helper T-cell immune responses. These immune responses are mediated by CD8+ and CD4+ T lymphocytes, respectively. As noted above, CTL with the appropriate antigen specificity can directly kill virus-infected cells, thus playing a primary role in clearing or limiting viral infections. CD4+ T lymphocytes provide immunological “help” for developing antibody responses and for promoting lymphocyte memory. Classically, CTL responses were measured by the in vitro capacity of CD8+ T cells to kill autologous cells presenting appropriate peptide epitopes (cytotoxicity assays). In their usual format, these assays are not quantitative and are extremely tedious to perform. However, additional assays have emerged in the past five years that enable more accurate measurement of the overall antigen-specific population for both CD8+ and CD4+ T cells.

12 Dec 2003

14:11

AR

AR206-ME55-21.tex

AR206-ME55-21.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:355-372. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

ADENOVIRUS VECTORS AS HIV-1 VACCINES

P1: GBC

359

The deficiency of classical limiting dilution cytotoxicity assays became clear with the advent of MHC class I tetramer assays that quantitate the number of CD8+ T cells that express a specific T-cell receptor (21). The tetramer method showed that precursor frequency assays detect only a small fraction of the actual antigenspecific cell population. Although tetramer reagents permit accurate assessment of antigen-specific CD8+ T cells, these reagents are limited to only a few epitopes and HLA alleles, so it is not possible to assess more than a minor fraction of the overall CTL response against a protein antigen. Other assays measure the frequencies of T cells that produce cytokines following short-term in vitro treatment of lymphocytes with peptide antigens. Memory T cells assume a quiescent, nondividing state in the absence of antigen and express no or little cytokine. When these cells are exposed to antigen bound to MHC class I molecules, they become activated and initiate entry into the cell-division cycle. The most prominent of the cytokines produced by activated T cells are γ -interferon, tumor necrosis factor alpha (TNFα), and interleukin 2 (IL-2). The frequencies of activated T cells may be obtained by using assays that allow the cytokine-producing T cells to be directly visualized. The two most common and convenient assays for this are the ELIspot and intracellular cytokine staining (ICS) assays. ELIspot assays are performed in 96-well plates coated with anticytokine antibodies into which peripheral-blood mononuclear cells (PBMCs) are added along with peptides representing a processed form of the target antigen. Once activated by the processed antigen, cells produce and secrete cytokines that are captured by the anticytokine antibodies. The cells are washed away after overnight stimulation with antigen (a time period too short for significant cell proliferation to occur) and a second anticytokine antibody, conjugated to an enzyme that cleaves a colorimetric substrate that becomes insoluble, is added. After the substrate is added and plates further developed, spots representing individual cytokine-secreting cells may be visualized, counted, and normalized with respect to the number of cells added to each well. The ELIspot assay is an efficient and sensitive means of assessing the total T-cell (sum of CD8+ and CD4+) response to a given antigen and detects a much larger fraction of the actual responding T-cell population (up to 50% as gauged by tetramer assays) than precursor frequency-based cytotoxicity assays. In practice, detection of γ -interferon is the best readout because there is a low background of this cytokine in the absence of antigen, and it is produced in greater quantity and by a greater proportion of T cells than TNFα and IL-2. The ICS assay is similar in principle to the ELIspot assay except that the induced cytokines are trapped within the cells by brefeldin A treatment. These cytokines accumulate within cells and are identified using anticytokine antibodies, as with the ELIspot assay, but are visualized by flow cytometry. As for the ELIspot assay, γ -interferon is the best readout. The ICS assay is somewhat less sensitive than the ELIspot assay but provides additional, complementary information concerning the responding T-cell population. Using additional fluorophore-tagged monoclonal antibodies against T-cell surface proteins, antigen-specific T cells can be further

12 Dec 2003

14:11

Annu. Rev. Med. 2004.55:355-372. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

360

AR

SHIVER

AR206-ME55-21.tex

¥

AR206-ME55-21.sgm

LaTeX2e(2002/01/18)

P1: GBC

EMINI

phenotyped as either CD8+ or CD4+ so that the total and relative contributions of each T-cell subset representing cytotoxic and helper functions can be quantified. CD8+ T-cell responses identified by ELIspot and ICS assays correlate well with presence or absence of cytotoxic activities, confirming that these assays detect cells with biologically meaningful functions. Both the ELIspot and ICS assays rely on the addition of peptide antigens to stimulate cytokine production by memory T cells. Known T-cell epitopes, typically nine amino acids in length for CD8+ T cells, and more generic peptide pools comprising large antigens may be used to stimulate cell cultures. We have relied extensively on the use of peptide pools that encompass entire protein antigens. We have optimized the use of the peptides so that only one or two pools are required for each antigen. Pools composed of peptides that are 15 amino acid residues in length and overlap by 11 residues can detect both CD8+ and CD4+ T cells. Pools composed of nine-amino-acid-residue-length peptides, overlapping by eight residues, specifically stimulate CD8+ T cells. For the ELIspot assay, we utilize four pools of 15-residue-length peptides (one for gag, two for pol, one for nef) and four pools of nine-residue-length peptides (also one for gag, two for pol, one for nef). This permits us to simultaneously measure immune responses to all three vaccine antigens and capture the total T-cell, as well as CD8+ T-cell, responses to each antigen. These assays were validated operationally with defined criteria for positive responses (at least 55 spot-forming cells per million lymphocytes and at least a fourfold increase in cytokine-producing cells in the presence of peptide antigens compared to no antigen). The ICS assay uses similar peptide pools as described for the ELIspot assay except that only the 15-residue-length peptides are used. This is because the contributions of the CD4+ versus the CD8+ T-cell responses can be obtained by staining against these cell-surface markers. This assay was also operationally validated with defined criteria for positive responses (at least 300 cytokine-producing cells per million lymphocytes and at least a threefold increase in cytokine-producing cells in the presence of peptide antigens compared to no antigen). It is also important to determine which and how many peptide epitopes are recognized by CTL from an immunized subject. Each CD8+ T-cell response defined against each T-cell epitope may represent distinct and complementary immune pressure against the virus. Accordingly, the ELIspot assay may also be used to deconvolute the total CD8+ T-cell response to assess individual responses against the 9-amino-acid-residue peptides. The summary estimations of cellular immune responses, in a given HIV-1infected subject or vaccinee, against HIV-1 proteins are derived from each of the assays described above. Overall, the following information is obtained: (a) total T-cell response against each viral antigen (e.g., gag, pol, or nef); (b) relative and total responses for the CD8+ and CD4+ T-cell subsets for each viral antigen; and (c) the number and identity of individual determinants targeted by the responding CD8+ T cells. These measurements should allow us to associate vaccine-mediated control of infection with an accompanying vaccine-elicited immune response.

12 Dec 2003

14:11

AR

AR206-ME55-21.tex

AR206-ME55-21.sgm

LaTeX2e(2002/01/18)

ADENOVIRUS VECTORS AS HIV-1 VACCINES

P1: GBC

361

Annu. Rev. Med. 2004.55:355-372. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Selection of Vaccine Antigens and the Potential for Cross-Clade Immune Responses The high genetic variability of HIV-1 is uniformly observed between and even within single infected individual subjects. This genetic diversity groups into at least 10 defined families (clades) of viruses that are assigned alphabetic designations (22). Clades C, A, and B, in that order, are the most prevalent worldwide, representing ∼85% of infections overall. Clade B infections predominate in the Western Hemisphere and Europe, Clade A in central Africa and eastern Asia, and Clade C in southern Africa and southern Asia. An optimally effective HIV-1 vaccine must provide protection that would cover this breadth of diversity. The best T-cell-based vaccine for accomplishing this type of broad coverage would probably need to elicit T-cell immune responses against multiple epitopes. This would help ensure sufficient match between vaccine antigen and infecting virus to trigger an effective immune response. The need for a vaccine to encode a large number of potential T-cell epitopes is even more important, given the high degree of MHC class I polymorphism within the human population. The leading vaccine candidate for initial efficacy studies is a trivalent mixture of Ad5 vectors, each encoding either HIV-1 gag, pol, or nef as vaccine antigens. These antigens are among the most conserved and largest gene products of HIV-1 that serve as targets for CTL responses in HIV-1-infected humans. HIV-1 pol is the most conserved protein (>90%) across all HIV-1 clades and is ∼850 amino acids in length. HIV-1 gag and nef are also relatively well-conserved (∼85% and 80%, respectively) across clades A, B, and C and are 500 and ∼200 amino acids long, respectively. In contrast, env is typically <80% conserved even within a single clade. This level of sequence conservation for gag, pol, and nef coupled with the relatively large sizes of these proteins suggest the existence of multiple T-cell epitopes that are common to diverse viruses. To evaluate this hypothesis, and to provide relevant human immunological data supporting the choice of vaccine antigens, T-cell immune responses were tested using ELIspot and intracellular cytokine assays with PBMC samples obtained from HIV-1-infected humans from different locations worldwide (230 HIV-1-positive subjects from the United States, Brazil, Thailand, Malawi, and South Africa) and infected with genetically diverse viruses (clades A/E, B, C). Antigen was provided in the assays as peptide pools for gag, nef, pol, rev, and tat sequences corresponding to near-consensus clade B viruses. High frequencies of these subjects (∼75–100%) showed positive T-cell responses against clade B gag, pol, and nef antigens regardless of either the donor’s origin or the clade of the infecting founder virus, whereas rev and tat responses were detected only infrequently (<25%). The highest levels of T-cell responses were also observed against gag, pol, and nef. These data demonstrate that humans are capable of producing T-cell immune responses against gag, pol, and nef antigens and suggest that an appropriate vaccine vector expressing gag, pol, and nef could elicit T-cell responses in most human subjects.

12 Dec 2003

14:11

Annu. Rev. Med. 2004.55:355-372. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

362

AR

SHIVER

AR206-ME55-21.tex

¥

AR206-ME55-21.sgm

LaTeX2e(2002/01/18)

P1: GBC

EMINI

These same HIV-1-positive subjects were also tested for T-cell immune responses against gag and nef antigens that corresponded to near-consensus viral sequences representing clades A and C. As expected from the high levels and frequencies of T-cell responses seen against clade B antigens, antigens based on different clades demonstrated significant levels of cross-reactivity. Cross-reactivities were determined by calculating the ratio of T-cell responses against nonfounder virus antigen to founder virus antigen using two different mathematical approaches: (a) linear regression coefficients (no-intercept model) of log tranformed responses and (b) the geometric means of ratios of T-cell responses. The two calculation methodologies provide a similar assessment of cross-strain T-cell responses. The data show that most of the T-cell response (50–100%) against nef or gag antigen is retained across clades regardless of the founder virus infection that initiated the immune responses. These data represent the most complete and only populationdirected analysis of cross-clade T-cell responses conducted to date and support the view that the diversity of HIV-1 genomic variability may not preclude constructing a T-cell vaccine with broad coverage. Based on these investigations, we selected three HIV-1 clade B antigens, gag, pol, and nef. These genes were constructed using sequences of near-consensus viral strains designated CAM-1, IIIB, and JR-FL for gag, pol, and nef respectively. Each gene encodes a complete, native protein, with pol commencing directly after the terminus of the protease open reading frame. Both pol and nef were functionally inactivated with site mutations. For pol these changes converted all nine catalytic acidic residues to alanine; nef was inactivated by conversion of glycine-2 to alanine to eliminate its myristoylation signal, as nef function is strictly tied to specific intracellular location (23). The selection of these vaccine antigens should help ensure immune responses in most vaccine recipients with potential coverage against viral diversity.

Immunogenicity of Ad5 Vaccines in Nonhuman Primates Immunization studies in nonhuman primates compared potencies of different vaccines, dose-response relationships, effect of prior adenovirus immunity, phenotypes of responding T cells, cross-strain T-cell responses, heterologous viral vector combinations, and protective efficacy against simian immunodeficiency viruses. Figure 2 summarizes antigag T-cell responses in rhesus monkeys elicited by the Ad5–HIV-1 gag vaccine as measured by ELIspot assay. In this study, different vaccine doses ranging from 107 to 1011 viral particles (vp; based on quantitative PCR determination of viral genome equivalents) were given as intramuscular injections at weeks 0 and 24 (24). Unlike humans, rhesus monkeys are not natural hosts to Ad5 infection and do not exhibit neutralizing antibodies to Ad5 vectors. (The lefthand three panels of the graph depict T-cell responses obtained in the absence of pre-existing immunity to Ad5.) These monkeys show a clear dose response to the vaccine immediately after the first injection, with the highest responses observed at the highest dose levels and responses approaching background, but still positive, at the lowest dose. As stated above, these were the strongest T-cell immune responses

12 Dec 2003

14:11

AR

AR206-ME55-21.tex

AR206-ME55-21.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:355-372. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

ADENOVIRUS VECTORS AS HIV-1 VACCINES

P1: GBC

363

Figure 2 ELIspot T-cell responses for monkeys immunized with Ad5-gag. Immunizations were given at weeks 0 and 24 using the indicated doses to animals that were either naive for Ad5 exposure or had been pretreated with one or three 1010-vp doses of non-gag-encoding Ad5 virus (producing neutralization titers of ∼30–90 and ∼270–810, respectively, at the time of the first Ad5-gag injection). Peripheral-blood mononuclear cell (PBMC) samples were stimulated using an HIV-1 gag peptide pool. Mock-corrected responses are shown for each monkey. SFC = IFN-γ spot-forming cells.

for any individual vaccine we have tested. The right-hand panels show immune responses obtained at the highest vaccine dose in the presence of pre-established anti-Ad5 immunity. Serum neutralizing antibodies against Ad5 were generated by prior injection of adenovirus vector that did not encode an immunogen. The resulting neutralizing antibody titers were similar to low (30–90) and relatively high (270–810) titers observed in humans. In this way, pre-existing Ad5 immunity attenuated but did not abrogate the vaccine response. The net effect of pre-existing adenovirus immunity was an effective dose reduction of vaccine; that is, a 1011 vp dose exhibited immunogenicity similar to a 109 vp dose. As noted above (“Measuring Cellular Immune Responses Against HIV-1 Antigens”), γ -interferon ELIspot assays measure the total (CD8+ and CD4+) T-cell response to antigen. Over the course of our research program, we have established a strong concordance between this marker and functional assays. For CD8+ T cells, function has been measured classically by cytotoxicity assays that are semiquantitative in nature. Cytotoxicity assays provide a relevant but indirect, semiquantitative glimpse of functional response. Figure 3 depicts cytotoxicity profiles for each animal tested for ELIspot responses above. Each profile consists of plots

12 Dec 2003

14:11

Annu. Rev. Med. 2004.55:355-372. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

364

AR

SHIVER

AR206-ME55-21.tex

¥

AR206-ME55-21.sgm

LaTeX2e(2002/01/18)

P1: GBC

EMINI

Figure 3 CTL responses for monkeys immunized with Ad5-gag. Peripheral-blood mononuclear cells from Ad5-gag vaccinees that had been restimulated for two weeks with vaccinia virus-gag treatment were mixed with gag peptide pool-pulsed autologous B cell lines as target cells at the indicated effector-to-target (E:T) ratios. The percentages of lysed targets are shown for samples without (squares) or with (diamonds) HIV-1 gag peptide pool antigens.

showing the percentage of autologous target cells, either treated (diamond symbols) or untreated (square symbols) with gag peptide antigen, killed in the presence of cultured lymphocytes at fixed ratios (shown as effector-to-target ratios). Depletion of CD8+ T cells prior to setting up the cytotoxicity assays eliminated all cytotoxic activities (data not shown). These data demonstrate that T lymphocytes from every immunized monkey exhibited cytotoxic activities against gag antigen-sensitized autologous cells, confirming that these antigen-specific T cells have relevant functional responses. Finally, epitope mapping using the ELIspot assay showed that the immunized monkeys recognized from one to seven distinct peptides, with three epitopes being the most common (data not shown). These data confirm that the Ad5-gag vaccine elicits responses against multiple T-cell determinants. Intracellular γ -interferon cytokine staining visualized by flow cytometry was also used to determine the relative CD8+/CD4+ T-cell contributions to the antigag T-cell response. The scatter plots in Figure 4 show the fraction of gag-specific T cells that are either CD8+ (upper right quadrant) or CD4+ (lower right quadrant) cells. The top row of plots shows the background responses without antigen and the lower row shows the gag-specific responses. These data demonstrate that CD8+

12 Dec 2003

14:11

AR

AR206-ME55-21.tex

AR206-ME55-21.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:355-372. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

ADENOVIRUS VECTORS AS HIV-1 VACCINES

P1: GBC

365

Figure 4 Phenotyping antigag T-cell responses by flow cytometry for monkeys immunized with Ad5-gag. CD3+ lymphocytes are characterized for CD8+ staining and IFN-γ production. Numbers reflect the percentage of CD3+ lymphocytes that are CD8+IFN-γ + (upper right quadrant) or CD4+IFN-γ + (lower right quadrant).

T cells comprise most of the T-cell response induced by Ad5-gag vaccines, in corroboration of the cytotoxic-assay data summarized above, with CD4+ cells as a smaller but consistent proportion of the total response. The final potential HIV-1 vaccine will be a trivalent vaccine composed of an equal mixture of Ad5 vectors encoding HIV-1 gag, pol, and nef, as discussed above. The trivalent vaccine was also tested in rhesus monkeys for immunogenicity against each vaccine antigen. Figure 5 summarizes ELIspot T-cell responses against the three antigens four weeks after immunizations at weeks 0 and 24 using 1010 vp per vector. T-cell responses against all three antigens were observed to be of similar magnitude as those obtained previously with immunization against gag alone.

Efficacy of Ad5-SIV gag Vaccine in Nonhuman Primates Against Immunodeficiency Virus Challenge The data summarized above show that an Ad5-gag vaccine stimulates gag-specific CD8+ T cells with cytotoxic activity. We next tested an Ad5 vaccine vector encoding SIV gag to determine whether this vaccine could protect nonhuman primates against infection with a simian AIDS virus (11). We specifically excluded an envcontaining vaccine component from this study to clearly segregate the protection conferred by cellular immune responses from that of challenge-virus-specific antibody responses. A chimeric SIV/HIV called SHIV89.6P was used to challenge monkeys by intravenous inoculation. SHIV89.6P comprises an SIV gag/pol/nef backbone and an HIV-1 tat/rev/env. It is extremely pathogenic in monkeys, causing a profound acute CD4 lymphopenia, set-point viral loads of ∼106 viral RNA (vRNA) molecules/ml of plasma, and progression to AIDS within 6–12 months.

12 Dec 2003

14:11

Annu. Rev. Med. 2004.55:355-372. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

366

AR

SHIVER

AR206-ME55-21.tex

¥

AR206-ME55-21.sgm

LaTeX2e(2002/01/18)

P1: GBC

EMINI

Figure 5 ELIspot T-cell responses for monkeys immunized with trivalent Ad5–HIV1 gag, pol, nef vaccine. Immunizations were administered at weeks 0 and 24 using 1010 vp per vector. Peripheral-blood mononuclear cell (PBMC) samples were either untreated (mock) or antigen-sensitized using HIV-1 gag, nef, and pol peptide pools. SFC = IFN-γ spot-forming cells.

Three monkeys were immunized at 0, 6, and 24 weeks with 1011 vp of Ad5SIV gag vaccine. All three developed strong antigag CD8+ T-cell responses (see Table 1) similar to those directed against HIV-1 gag in the immunogenicity studies described above. Three months after the last injection, these and six unimmunized control monkeys were challenged intravenously with SHIV89.6P using ∼50 monkey-infectious doses. All monkeys became infected; five of the six controls quickly developed CD4 lymphopenia and high viral loads, as expected (see Figure 6, color insert). These animals also subsequently progressed to AIDS. The

TABLE 1 Anti-SIV gag T-cell responses in monkeys immunized with Ad5-SIV gaga ELIspotb CM9 peptide

Total gag

Tetramerc CM9/tet

54G

6

508

180

1798

56G

16

435

114

4571

82F

8

314

118

4376

Monkey #

Mock

a

Injections at 0, 6, 24 wks; challenge at wk 36; immune responses measured prior to challenge.

b

SFC/million peripheral-blood mononuclear cells.

Tetramer-positive CD3+ CD8+ T cells per million lymphocytes; background <100.

c

12 Dec 2003

14:11

AR

AR206-ME55-21.tex

AR206-ME55-21.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:355-372. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

ADENOVIRUS VECTORS AS HIV-1 VACCINES

P1: GBC

367

Ad5-SIV-gag–immunized animals, however, did not exhibit CD4 lymphopenia, had a significantly dampened acute viremia, and resolved chronic viremia close to undetectable levels (∼50 vRNA/ml of plasma). The immunized monkeys have not progressed to AIDS or shown a viral rebound in at least 933 days following infection. These data demonstrate that CTL responses elicited by an adenovirus vector vaccine can provide significant control of a simian AIDS virus. Nonhuman primate immunization/immunodeficiency-virus-challenge studies have shown that, in general, the animals best able to control viremia after challenge exhibited relatively higher levels of ELIspot- and ICS-based T-cell responses after immunization, against relatively more epitopes. The magnitude of T-cell responses at the time of challenge demonstrated significant negative correlations with both peak and chronic viremias. It is difficult from these small studies to ascribe strict criteria to the levels of these immune responses and the attenuation of infection post challenge. Nevertheless, such experiments provide important guidance for potential human efficacy studies and suggest that the immune-response assays we use capture relevant immunological information.

ALTERNATIVE IMMUNIZATION STRATEGIES Vectors Derived from Rare Human Adenovirus Serotypes The data summarized above provide strong support for an adenovirus-based vaccine approach against HIV-1 infection. However, pre-existing antiadenoviral immunity can dampen or possibly abrogate vaccine responses. The scientific literature and data from our own epidemiological studies suggest that most North Americans have anti-Ad5 neutralizing antibody titers, and about one third have relatively high titers (>200). Other parts of the world typically exhibit higher frequencies and levels of anti-Ad5 antibodies. The use of vectors derived from rare human adenovirus serotypes is an option for overcoming pre-existing Ad5 immunity. There are ∼50 known human adenovirus serotypes, many of which show much less frequent seroprevalence in diverse human populations (25). In particular, adenoviruses from serogroups B and D are much less seroprevalent than group C viruses. However, the use of vectors based on these rarer serotypes was complicated by the fact that the currently available cell lines, such as PER.C6TM cells, that support replication of E1-deleted adenovirus vectors are specific for type 5 adenoviruses and for those belonging to the closely related group C. Specific E1 gene products expressed by the cell line must interact with a gene product expressed by the adenoviral E4 gene region to support propagation of the adenoviral vector. This interaction is serogroup-specific. Accordingly, this constraint was overcome by replacing the E4 coding domain of non–group C adenovirus vectors with the Ad5 E4. The resulting chimeric vectors are capable of propagating in PER.C6TM cells with productivity similar to that of the group C adenovirus vectors. This observation has enabled us to produce essentially any human adenovirus vector using PER.C6TM cells.

12 Dec 2003

14:11

Annu. Rev. Med. 2004.55:355-372. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

368

AR

SHIVER

AR206-ME55-21.tex

¥

AR206-ME55-21.sgm

LaTeX2e(2002/01/18)

P1: GBC

EMINI

Figure 7 ELIspot T-cell responses for monkeys immunized with rare serotype adenovirus vectors expressing HIV-1 gag. Immunizations were given at weeks 0 and 4 using 1010 vp for each vector. Peripheral-blood mononuclear cell samples were stimulated using an HIV-1–gag peptide pool. Mock-corrected responses are shown for each monkey.

Using this approach, we derived a variety of vectors based on uncommon serotypes selected from groups B (adenoviruses 34 and 35) and D (Ad24). Each of these viruses has low seroprevalence in humans. Figure 7 summarizes immunization results in monkeys using several of these vectors. In this experiment, monkeys were immunized with 1011 vp doses of Ad5, Ad24, Ad34, or Ad35 HIV1–gag vaccines at 0 and 4 weeks. The Ad5 vaccinees, as noted above, quickly developed T-cell responses after the first immunization reaching magnitudes of ∼500–1000 γ -interferon–secreting cells per million PBMCs. All three of the rare serotype adenoviruses also were immunogenic but generally less so than Ad5. Interestingly, responses were observed following a single immunization with Ad24 but not until after two immunizations with Ad34 and Ad35. In addition, each of the new adenoviral vectors elicits primarily CD8+ T-cell responses, as we observed with Ad5 (data not shown). Collectively, these experiments provide evidence that vaccine approaches using rare human adenoviruses, chimeric with Ad5 for E4, are immunogenic in primates.

Heterologous Viral Vector Prime/Boost Immunizations The relationship between pre-existing Ad5 neutralization titer, Ad5 vaccine dose, and immune response in humans is unclear at this time. As noted above, it will

12 Dec 2003

14:11

AR

AR206-ME55-21.tex

AR206-ME55-21.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:355-372. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

ADENOVIRUS VECTORS AS HIV-1 VACCINES

P1: GBC

369

be important to have alternative vaccine components/approaches ready for clinical testing if pre-existing adenovirus immunity is found to be an important limitation of the initial Ad5 vaccine candidates. Possible alternative approaches include the use of rare adenovirus serotypes in place of Ad5 as well as heterologous viral vector “prime/boost” immunizations to increase overall immune responses compared to the use of a single vector. Heterologous “prime/boost” immunizations could use two different adenovirus serotypes (one type for the prime and another for the boost) or use an adenoviral vector prime followed by boosting with a nonadenoviral vector. These immunization strategies should help ensure that most vaccinees respond to the vaccine. We tested several vector combinations in nonhuman primate immunization studies and identified two combinations that showed substantial immune-response enhancements compared to single-vector immunizations. The first of these combinations involved initial priming with Ad5-gag and subsequently boosting with canarypox or modified vaccinia Ankara virus vectors encoding gag. Monkeys immunized with this combination exhibited consistently higher T-cell responses than animals immunized using a homologous Ad5 prime/boost regimen. The T-cell responses following the poxvirus boost comprised similar proportions of CD4+ and CD8+ T cells rather than predominantly CD8+ T cells as observed for adenovirus vector vaccines (data not shown). The order of immunization was critical: Priming first with poxvirus vector and then Ad5 produced responses no greater than immunization with Ad5 alone. Figure 8 shows results from a monkey immunization study in which heterologous prime/boost immunizations were conducted using different adenovirus serotypes. Low doses of Ad5 primed for strong responses following boosting with Ad24. In addition, immunizations with either Ad24 or Ad35 primed for significant boosting following Ad5 immunization, particularly for the Ad24/Ad5 combination. Collectively, these data exemplify a number of potential vaccine approaches that could be used if a vaccine based on Ad5 alone does not provide sufficient immunogenicity in diverse human populations.

CONCLUSIONS Developing an effective AIDS vaccine has proven to be much more difficult than we recognized at the beginning of the epidemic. These difficulties have served to highlight the deficiencies in our overall understanding of the underlying virologic and immunologic processes. For HIV-1, the most salient of these difficulties are the virus’s high degree of genetic diversity, resistance to neutralization by antibodies, and relentless destruction of the host’s immune system. Once the role of CTL responses in containment of the virus was appreciated, researchers began to explore the potential of CTL-eliciting vaccines to either prevent or significantly

12 Dec 2003

14:11

Annu. Rev. Med. 2004.55:355-372. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

370

AR

SHIVER

AR206-ME55-21.tex

¥

AR206-ME55-21.sgm

LaTeX2e(2002/01/18)

P1: GBC

EMINI

Figure 8 ELIspot T-cell responses for monkeys immunized with heterologous adenovirus vector prime/boost combinations. Immunizations were given at weeks 0 and 4 at the indicated doses for the prime followed by boosting with the different viral vectors at week 24. Peripheral-blood mononuclear cell samples were stimulated using an HIV-1 gag peptide pool. Mock-corrected responses are shown for each monkey.

ameliorate the infection. Our studies in this regard have led us to conclude that replication-defective adenovirus vector vaccines are very effective for eliciting CTL responses against HIV-1 antigens. In simian models of immunodeficiency virus infection, such responses substantially mitigated experimental infection and significantly delayed the onset of clinical AIDS. It is hoped that these vaccineelicited cellular immune responses will have a similar beneficial effect in humans who become infected with HIV-1. The resulting reduced viral load should also lessen the probability of virus transmission by the infected host. The vaccine would, therefore, have a substantial effect on the spreading epidemic. ACKNOWLEDGMENTS We would like to thank our colleagues whose talents and dedication have proven essential in the conduct of this research. It is not possible to identify all of them individually but we note, in particular, the contributions of D. Casimiro, A. Bett, X. Liang, T. Fu, L. Chen, F. Wang, S. Dubey, J. Condra, W. Schleif, Z. Zhang, P. Coplan, M. Davies, D. Freed, L. Hand, M. Caulfield, C.T. Caskey, and E.M. Scolnick.

12 Dec 2003

14:11

AR

AR206-ME55-21.tex

AR206-ME55-21.sgm

LaTeX2e(2002/01/18)

ADENOVIRUS VECTORS AS HIV-1 VACCINES

P1: GBC

371

The Annual Review of Medicine is online at http://med.annualreviews.org

Annu. Rev. Med. 2004.55:355-372. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

LITERATURE CITED 1. UNAIDS/WHO Working Group on Global HIV/AIDS STD Surveillance. 1998. Report on the Global HIV/AIDS Epidemic, December 1998 2. Pomerantz RJ, Horn DL. 2003. Twenty years of therapy for HIV-1 infection. Nat. Med. 9:867–72 3. Mathews TJ. 1994. Tenth anniversary perspectives on AIDS. AIDS Res. Hum. Retrovir. 10:631–32 4. Moore JP, Cao Y, Qing L, et al. 1995. Primary isolates of human immunodeficiency virus type 1 are relatively resistant to neutralization by monoclonal antibodies to gp120, and their neutralization is not predicted by studies with monomeric gp120. J. Virol. 69:101–9 5. Fouts T, Godfrey K, Bobb K, et al. 2002. Crosslinked HIV-1 envelope-CD4 receptor complexes elicit broadly cross-reactive neutralizing antibodies in rhesus macaques. Proc. Natl. Acad. Sci. USA 99:1842– 47 6. Cohen J. 2003. AIDS vaccine trial produces disappointment and confusion. Science 299:1290–91 7. Koup RA, Safrit JT, Cao Y, et al. 1994. Temporal association of cellular immune responses with the initial control of viremia in primary human immunodeficiency virus type 1 syndrome. J. Virol. 68:4650–55 8. Borrow P, Lewicki PH, Hahn BH, et al. 1994. Virus-specific CD8+ cytotoxic Tlymphocyte activity associated with control of viremia in primary human immunodeficiency virus type 1 infection. J. Virol. 68:6103–10 9. Schmitz JE, Kuroda MJ, Santra S, et al. 1999. Control of viremia in simian immunodeficiency virus infection by CD8+ lymphocytes. Science 283:857–60 10. Barouch DH, Santra S, Schmitz JE, et al. 2000. Control of viremia and prevention of clinical AIDS in rhesus monkeys

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

by cytokine-augmented DNA vaccination. Science 290:486 Shiver JW, Fu T, Chen L, et al. 2002. Replication-incompetent adenoviral vaccine vector elicits effective antiimmunodeficiency-virus immunity. Nature 415:331–35 Amara RR, Villinger F, Altman JD, et al. 2001. Control of a mucosal challenge and prevention of AIDS by a multiprotein DNA/MVA vaccine. Science 292:69– 74 Pamer E, Cresswell P. 1998. Mechanisms of MHC class I–restricted antigen processing. Annu. Rev. Immunol. 16:323–58 Matloubian M, Concepcion RJ, Ahmed R. 1994. CD4+ T-cells are required to sustain CD8+ cytotoxic T-cell responses during chronic viral infection. J. Virol. 68:8056– 63 Jaye A, Magnusen AF, Sadiq AD, et al. 1998. Ex vivo analysis of cytotoxic T lymphocytes to measles antigens during infection after vaccination in Gambian children. J. Clin. Invest. 102:1969–77 Koostra NA, Verma IM. 2003. Gene therapy with viral vectors. Annu. Rev. Pharmacol. Toxicol. 43:413–39 Boshart M, Weber F, Jahn G, et al. 1985. A very strong enhancer is located upstream of an immediate early gene of human cytomegalovirus. Cell 41:521–30 Pfarr DS, Rieser LA, Woychik RP, et al. 1986. Differential effects of polyadenylation regions on gene expression in mammalian cells. Nucleic Acids Res. 19:3979– 86 Lathe R. 1985. Synthetic oligonucleotide probes deduced from amino acid sequence data. Theoretical and practical considerations. J. Mol. Biol. 183:1–12 Shiver JW, Davies ME, Freed DC, et al. 2003. Vaccines comprising synthetic genes. U.S. Patent No. 6,534,312 B1

12 Dec 2003

14:11

372

AR

SHIVER

AR206-ME55-21.tex

¥

AR206-ME55-21.sgm

P1: GBC

EMINI

21. Doherty PC, Christensen JP. 2000. Accessing complexity: the dynamics of virusspecific T cell responses. Annu. Rev. Immunol. 18:561–92 22. Korber B, Kuiken C, Foley B, et al. 1998. In Human Retroviruses and AIDS. Los Alamos, NM: Los Alamos Natl. Lab. 23. Pandori MW, Fitch NJ, Craig HM, et al. 1996. Producer cell modification of human immunodeficiency virus type 1: nef is a virion protein. J. Virol. 70:4283–90 Annu. Rev. Med. 2004.55:355-372. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

LaTeX2e(2002/01/18)

24. Casimiro D, Chen L, Fu T, et al. 2003. Comparative immmunogenicity in rhesus monkeys of DNA plasmid, recombinant vaccinia virus, and replication-defective adenovirus vectors expressing a human immunodeficiency virus type 1 gag gene. J. Virol. 77:6305–13 25. D’Ambrosio E, Del Grosso N, Chicca A, et al. 1982. Neutralizing antibodies against 33 human adenoviruses in normal children in Rome. J. Hyg. (Lond). 89:155–61

Shiver.qxd

1/5/2004

3:37 PM

Page 1

Annu. Rev. Med. 2004.55:355-372. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

ADENOVIRUS VECTORS AS HIV-1 VACCINES

C-1

Figure 6 SHIV89.6P challenge of rhesus monkeys immunized with Ad5-SIV-gag vaccine. Immunizations were given at weeks 0, 6, and 24 and viral challenge was administered three months later. Post-challenge viremia and CD4+ cell counts are shown for control naive monkeys (upper panels) and vaccinees (lower panels). Plasma viral levels were determined using a branched DNA amplification assay with detection limits of either 500 (before day 637) or 50 (from day 637) viral RNA copies/ml.

12 Dec 2003

19:40

AR

AR206-ME55-22.tex

AR206-ME55-22.sgm

LaTeX2e(2002/01/18) P1: GBC 10.1146/annurev.med.55.091902.104417

Annu. Rev. Med. 2004. 55:373–94 doi: 10.1146/annurev.med.55.091902.104417 c 2004 by Annual Reviews. All rights reserved Copyright ° First published online as a Review in Advance on Aug. 18, 2003

LEFT VENTRICULAR DIASTOLIC DYSFUNCTION AND DIASTOLIC HEART FAILURE William H. Gaasch1 and Michael R. Zile2 Annu. Rev. Med. 2004.55:373-394. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

1

Department of Cardiovascular Medicine, Lahey Clinic, Burlington, Massachusetts 01805 and University of Massachusetts Medical School, Worcester, Massachusetts 01655; email: [email protected]; 2Cardiology Division, Department of Medicine, Gazes Cardiac Research Institute, Medical University of South Carolina, Charleston, South Carolina 29425 and the Ralph H. Johnson Department of Veterans Affairs Medical Center, Charleston, South Carolina 29401

Key Words cardiac function, congestive heart failure, diastole, ejection fraction ■ Abstract Thirty to fifty percent of patients presenting with signs and symptoms of heart failure have a normal left ventricular (LV) systolic ejection fraction. The clinical examination cannot distinguish these patients (diastolic heart failure) from those with a depressed ejection fraction (systolic heart failure), but echocardiography can. The management of diastolic heart failure has two major objectives. The first is to reverse the consequences of diastolic dysfunction (e.g., venous congestion), and the second is to eliminate or reduce the factors responsible for diastolic dysfunction (e.g., myocardial hypertrophy, fibrosis, and ischemia).

INTRODUCTION Clinicians and physiologists have, in the past three decades, reexamined their concepts about the pathophysiology of heart failure. A major focus of this deliberation has been to clarify the important distinctions between left ventricular (LV) systolic dysfunction and diastolic dysfunction (Figure 1). Simply stated, systolic dysfunction can be considered a defect in the ability of the myofibrils to shorten against a load; the ventricle loses its ability to eject blood into a high-pressure aorta and the ejection fraction falls. The term diastolic dysfunction implies that the myofibrils do not rapidly or completely return to their resting length; the ventricle cannot accept blood at low pressures, and ventricular filling is slow or incomplete unless atrial pressure rises. Cardiac structure and function differ substantially in systolic and diastolic dysfunction, but the clinical consequences (i.e., the signs and symptoms of heart failure) are similar. A balanced view of the differences and similarities of these two conditions is essential to the process of diagnosing and treating patients with heart failure. In this review, we describe the clinical and pathophysiologic features of diastolic dysfunction and provide an approach to the diagnosis and treatment of patients with diastolic heart failure. 0066-4219/04/0218-0373$14.00

373

12 Dec 2003

19:40

Annu. Rev. Med. 2004.55:373-394. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

374

AR

GAASCH

AR206-ME55-22.tex

¥

AR206-ME55-22.sgm

LaTeX2e(2002/01/18)

P1: GBC

ZILE

Figure 1 Diagram of left ventricular (LV) pressure-volume loops in systolic dysfunction and diastolic dysfunction. In systolic dysfunction, contractility is depressed, and the end-systolic pressure-volume line is displaced downward and to the right; there is diminished capacity to eject blood into a high-pressure aorta. In diastolic dysfunction, chamber stiffness is increased and the diastolic pressure-volume relation is displaced up and to the left; there is diminished capacity to fill at low diastolic pressures. The LV ejection fraction is low in systolic dysfunction and normal in diastolic dysfunction. (Adapted from Reference 40.)

Interest in the diastolic properties of the left ventricle, the nature of myocardial relaxation, and diastolic filling of the ventricles was stimulated by the early work of Henderson (1), Wiggers & Katz (2), and Meek (3)—and by the recognition that normal diastolic “function” is an important ingredient in the coordinated pump function of the heart. Over the past three decades, clinical investigators and basic scientists have studied the pathophysiology of diastole and have sought to understand the relationship between abnormalities in diastolic function and the clinical syndrome of heart failure. During the 1970s, the study of diastole in humans was largely confined to the cardiac catheterization laboratory. Clinical investigators measured LV diastolic pressure and volume and described the mechanisms that underlie LV diastolic dysfunction. LV compliance or distensibility was decreased in patients with pressure overload hypertrophy, and it was suggested that such diastolic dysfunction limits the potential to utilize the Frank-Starling mechanism and thereby has a negative impact on LV systolic performance (4). Other investigators demonstrated the dramatic impact of ischemia on the diastolic properties of the ventricle (5, 6). The substantial increase in LV diastolic pressure seen during angina pectoris was largely a consequence of diminished diastolic distensibility or compliance of the ventricle. These and other studies, performed in the 1970s, examined the physical

12 Dec 2003

19:40

AR

AR206-ME55-22.tex

AR206-ME55-22.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:373-394. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

DIASTOLIC HEART FAILURE

P1: GBC

375

properties of the fully relaxed ventricle and defined the determinants of passive chamber stiffness. The major determinants are the chamber volume, wall mass, and composition of the wall (7–9). They also examined dynamic factors that are intrinsic to the myocardium (i.e., the process of relaxation) as well as those extrinsic to the ventricle [i.e., the pericardium and the right ventricle (10, 11)]. Alterations in these factors, alone or in concert, could contribute to acute or chronic alterations in the diastolic properties of the ventricle, and that diastolic dysfunction could lead to heart failure (12, 13). In the 1980s, clinicians reported what seemed to be a surprisingly high prevalence (approximately 30%) of a normal LV ejection fraction in patients with congestive heart failure (14–16). Recurrent pulmonary edema was reported in elderly patients with ischemic heart disease—despite a normal LV ejection fraction (17). A syndrome of hypertensive hypertrophic cardiomyopathy was described in elderly patients with dyspnea or chest pain (18). These patients had “abnormal diastolic function” in the presence of “excessive systolic emptying” (high normal ejection fraction), and it was suggested that treatment of such patients differed from that used in patients with a depressed LV ejection fraction. Others emphasized diastolic dysfunction in patients with hypertension (19) and the elderly (20). Thus, it became apparent that a significant number of patients with heart failure had a normal LV ejection fraction, and the terms diastolic dysfunction and diastolic heart failure emerged in the medical literature (21–24). In the 1990s, the use of echocardiography and Doppler techniques to assess LV diastolic relaxation and filling grew at an exponential rate. The noninvasive nature and the widespread availability of these techniques contributed to this growth, and as a result, a prodigious volume of reports and clinical research on diastolic dysfunction was published. These studies provide considerable insight into the dynamics of LV relaxation and filling in health and disease (25–27). A most important contribution was the use of echocardiography to estimate LV filling pressure (28–31). These techniques also provide significant prognostic information. For example, the finding of echocardiographic evidence of diastolic dysfunction in an asymptomatic patient is a risk factor for the development of heart failure (32); the early identification of such patients provides a window of opportunity to prevent progression of what appears to be “preclinical heart disease” (33). In recent years, our understanding of the epidemiology and clinical pathophysiology of diastolic dysfunction and heart failure has expanded (34–39). The prevalence of diastolic heart failure in the community is now known to be at least as high as that reported in previous studies of hospitalized patients; almost half of all patients with heart failure have diastolic heart failure. It is especially prevalent in elderly women. Clinicians now recognize that the manifestations of heart failure are similar in patients with systolic and diastolic heart failure—despite major differences in LV volume, mass, geometry, and function (37). Considerable progress in our understanding of the basic mechanisms underlying diastolic dysfunction is being made (38, 39), but to date, there have been no published double-blind

12 Dec 2003

19:40

376

AR

GAASCH

AR206-ME55-22.tex

¥

AR206-ME55-22.sgm

LaTeX2e(2002/01/18)

P1: GBC

ZILE

placebo-controlled randomized multicenter therapeutic trials in patients with diastolic heart failure.

Annu. Rev. Med. 2004.55:373-394. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

TERMINOLOGY Diastolic dysfunction refers to abnormal mechanical properties of the myocardium and includes abnormal LV diastolic distensibility, impaired filling, and slow or delayed relaxation—regardless of whether the ejection fraction is normal or depressed and whether the patient is asymptomatic or symptomatic. In this review, the term asymptomatic diastolic dysfunction is used to refer to an asymptomatic patient with a normal ejection fraction and an abnormal echo-Doppler pattern of LV filling; this is often seen, for example, in patients with hypertensive heart disease. If such a patient were to exhibit symptoms of effort intolerance and dyspnea, especially if there were evidence of venous congestion and edema, the term diastolic heart failure is used. This terminology parallels that used in asymptomatic and symptomatic patients with LV systolic dysfunction, and it facilitates the use of a pathophysiologic, diagnostic, and therapeutic framework that includes all patients with LV dysfunction—whether or not they have symptoms (40).

CAUSES OF DIASTOLIC DYSFUNCTION The factors that influence and determine the LV diastolic pressure-volume relations are shown in Figure 2. The basic mechanisms that underly diastolic dysfunction may be intrinsic to the cardiomyocyte (e.g., abnormal calcium homeostasis), or they may be a consequence of abnormalities in the extracellular matrix (e.g., alterations in collagen). Neurohormonal and cardiac endothelial activity also modulate ventricular stiffness and relaxation (39). Clinically, the common causes of LV diastolic dysfunction are hypertrophy and ischemia, but several other conditions may cause heart failure in the presence of a normal LV ejection fraction (Table 1).

PATHOPHYSIOLOGY The diastolic properties of the left ventricle are determined largely by the size or volume of the LV chamber, the thickness and physical properties of the ventricular wall, and the process of myocardial relaxation. Thus, a combination of increased myocardial mass and alterations in the extracellular collagen network may cause or contribute to an increase in passive elastic stiffness of the ventricle and a steep diastolic pressure-volume relation (Figure 1). Disorders of the active process of myocardial relaxation, acting alone or in concert with abnormal passive properties of the ventricle, can also stiffen the ventricle. As a result, LV compliance or distensibility is reduced, the dynamics of filling are altered, and the end-diastolic pressure is increased (9, 12, 13, 27, 38). Under these circumstances, a relatively

12 Dec 2003

19:40

AR

AR206-ME55-22.tex

AR206-ME55-22.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:373-394. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

DIASTOLIC HEART FAILURE

P1: GBC

377

Figure 2 Diastolic pressure-volume relations (upper panel) and the factors that influence these relations. The effects of each of these factors depends on the time at which they occur; some factors exert their influence in early diastole (i.e., relaxation rate and diastolic suction), others in late diastole (i.e., pericardium), and still others throughout diastole (i.e., elastic properties). VES = end-systolic volume; Vo = equilibrium volume; VED = end-diastolic volume; RV = right ventricle. (Adapted from Reference 41).

small increase in central blood volume can produce a substantial increase in LV diastolic pressure and consequently pulmonary venous hypertension and pulmonary edema. Several factors can promote fluid retention and precipitate overt heart failure in patients with heart disease. Common precipitants include uncontrolled hypertension, atrial fibrillation, noncompliance with or inappropriate discontinuation of heart failure medications, myocardial ischemia, anemia, renal insufficiency, administration of nonsteroidal anti-inflammatory drugs, and overindulgence in high-salt foods (42). The glitazones may have a similar effect (43). Patients with LV diastolic dysfunction are especially sensitive to these precipitants because of the steep LV diastolic pressure-volume relation, which results in large changes in LV diastolic pressure with only small changes in volume. Thus, elevated LV diastolic and pulmonary venous pressures in patients with a normal ejection fraction (in the absence of valvular disease) are directly related to abnormalities in the diastolic properties of the ventricle (i.e., diastolic dysfunction). This is not to say that contractile function is entirely normal in patients with diastolic heart failure. Indeed, subtle abnormalities of contractile function are present in many if not most such patients; diastolic dysfunction, however, is the

12 Dec 2003

19:40

378

AR

GAASCH

AR206-ME55-22.tex

¥

AR206-ME55-22.sgm

LaTeX2e(2002/01/18)

P1: GBC

ZILE TABLE 1 Causes of heart failure with a normal ejection fraction

Annu. Rev. Med. 2004.55:373-394. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Left ventricular diastolic dysfunction Pressure-overload hypertrophy hypertensive heart disease aortic stenosis Ischemic heart disease acute myocardial ischemia chronic coronary heart disease Cardiomyopathy hypertrophic infiltrative restrictive Valvular heart disease Acute aortic or mitral regurgitation Mitral stenosis Aortic stenosis Pericardial disease Constriction Tamponade Circulatory congestive states Rapid fluid administration Arterio-venous fistula Severe anemia Thyrotoxicosis

dominant feature. It should also be recognized that diastolic function is rarely normal in patients with heart failure and a low ejection fraction; however, in this case, systolic dysfunction is the dominant feature. The differences and similarities between systolic and diastolic heart failure are summarized in Table 2. LV volume, geometry, and function differ substantially in these two conditions; at the microscopic level, significant differences in the cardiomyocyte and extracellular matrix are seen. Even brain natriuretic peptide (BNP) levels and survival differ. Despite these major differences, the clinical signs and symptoms of heart failure are present to a similar degree in patients with systolic and diastolic heart failure. Thus, the clinical history and physical examination do not provide information that allows a differentiation of systolic from diastolic heart failure. Patients with diastolic heart failure, as well as those with diastolic dysfunction and little or no congestion, exhibit exercise intolerance for two principal reasons. First, elevated LV diastolic and pulmonary venous pressure causes a reduction in lung compliance, which increases the work of breathing and evokes the symptom of dyspnea. Second, a substantial number of patients who have LV hypertrophy, a

12 Dec 2003

19:40

AR

AR206-ME55-22.tex

AR206-ME55-22.sgm

LaTeX2e(2002/01/18)

P1: GBC

DIASTOLIC HEART FAILURE

379

Annu. Rev. Med. 2004.55:373-394. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

TABLE 2 Differences and similarities between systolic and diastolic heart failure Systolic heart failure

Diastolic heart failure

Signs and symptoms

present

present

BNP

↑↑

↑

Exercise testing Duration Systolic BP Pulse pressure VO2

↓ ↑ ↑ ↓↓

↓ ↑↑ ↑↑ ↓

LV remodeling End-diastolic volume End-systolic volume Myocardial mass Relative wall thickness Cardiomyocyte EC matrix (collagen)

↑↑ ↑↑ ↑ (eccentric LVH) ↓ ↑ length ↓

N ↓ ↑ (concentric LVH) ↑↑ ↑ diameter ↑↑

LV systolic function Ejection fraction Stroke volume Myocardial contractility

↓↓ N-↓ ↓↓

N-↑ N-↓ ↓

LV diastolic function Chamber stiffness Myocardial stiffness Relaxation time-constant Filling dynamics End-diastolic pressure Preload reserve

N-↓ N-↑ ↑ abnormal ↑↑ exhausted

↑↑ ↑ ↑ abnormal ↑↑ limited

Morbidity

↑↑

↑↑

Survival

↓↓

↓

Abbreviations: BNP, brain natriuretic peptide; BP, blood pressure; VO2, oxygen consumption; LV, left ventricular; LVH, LV hypertrophy; EC, extracellular.

high relative wall thickness, and a small end diastolic volume exhibit a low stroke volume and a depressed cardiac output (44); a ventricle with a normal ejection fraction cannot produce a normal stroke volume if the chamber size is small. Third, these hearts exhibit a limited ability to utilize the Frank-Starling mechanism during exercise (45, 46). Such limited preload reserve, especially if coupled with the chronotropic incompetence that is seen with advancing age, limits the cardiac output during exercise. This leads to lactate accumulation and structural as well as functional abnormalities of skeletal muscles; the result can be skeletal muscle fatigue (the legs and the accessory muscles of respiration). This latter mechanism helps to explain the poor relationship between exercise tolerance and changes

12 Dec 2003

19:40

380

AR

GAASCH

AR206-ME55-22.tex

¥

AR206-ME55-22.sgm

LaTeX2e(2002/01/18)

P1: GBC

ZILE

in LV diastolic pressure. Other mechanisms, including physical deconditioning, contribute to exercise intolerance.

Annu. Rev. Med. 2004.55:373-394. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

DIAGNOSIS The clinical diagnosis of diastolic heart failure requires reliable evidence of heart failure in the presence of a normal or near-normal LV ejection fraction (greater than 40%–50%). It remains controversial whether measurements of diastolic function are necessary for the diagnosis of diastolic heart failure. Some insist that there must be hemodynamic (echocardiographic or catheterization) evidence of abnormal LV relaxation, filling, diastolic distensibility, or stiffness (47). Others do not accept echocardiographic data; rather, they require cardiac catheterization to document the presence of diastolic dysfunction before making a diagnosis of definite diastolic heart failure (48). Still others point out that virtually all patients with heart failure and a normal ejection fraction exhibit evidence of diastolic dysfunction, and they argue that echocardiographic or other documentation of diastolic dysfunction is merely confirmatory (49). Likewise, blood levels of BNP have been used to confirm the diagnosis of congestive heart failure (50), but further study will be required before BNP levels can be recommended as a method to differentiate systolic from diastolic heart failure (51). Echocardiography is often used to assess the LV ejection fraction, but the method of choice depends on the availability of and institutional experience with other techniques such as radionuclide angiography and cineangiograpy. In addition to the ejection fraction, the echocardiogram provides other information on LV function, LV geometry and wall thickness, regional wall motion abnormalities, valvular disease, pericardial disease, and left atrial size. Echocardiography has proven to be most useful in the assessment of LV size and ejection fraction, but the use of Doppler-derived indices of diastolic function has had less impact on the management of individual patients with acute heart failure (52). Recognizing these uncertainties, many if not most clinicians make the diagnosis of diastolic heart failure in patients with the signs and symptoms of heart failure in the presence of a normal LV ejection fraction, especially if there is evidence of LV hypertrophic remodeling (49). In the absence of hypertrophy, the diagnosis of probable diastolic heart failure may be more appropriate (48). After successful treatment, symptomatic patients who no longer exhibit evidence of congestion should carry the diagnosis of chronic diastolic heart failure.

MANAGEMENT There are no large randomized placebo-controlled trials that provide evidencebased therapeutic strategies in patients with diastolic dysfunction or diastolic heart failure. The recommendations presented in this review are based on the results of small clinical studies, anecdotal experience, and an understanding of the

12 Dec 2003

19:40

AR

AR206-ME55-22.tex

AR206-ME55-22.sgm

LaTeX2e(2002/01/18)

DIASTOLIC HEART FAILURE

P1: GBC

381

Annu. Rev. Med. 2004.55:373-394. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

TABLE 3 Management of diastolic heart failure Initial management Treat the presenting syndrome pulmonary edema/congestive state systemic atrial hypertension myocardial ischemia atrial fibrillation/tachycardia Clarify the diagnosis history and physical examination echocardiography cardiac catheterization/angiography biopsy Long-term management Consider mechanisms promote regression of left ventricular hypertrophy prevent/promote regression of fibrosis modify cellular/extracellular mechanisms Correct the pathophysiology salt restriction and diuretics block the renin-angiotensin-aldosterone system maintain atrial contraction prevent excessive tachycardia treat hypertension prevent myocardial ischemia

pathophysiology of diastole. In general, the management of diastolic heart failure has two objectives. The first is to reverse the consequences of diastolic dysfunction (e.g., venous congestion and exercise intolerance). The second is to eliminate or reduce the factors that are responsible for diastolic dysfunction (e.g., hypertrophy, fibrosis, ischemia). See Table 3.

ASYMPTOMATIC DIASTOLIC DYSFUNCTION The prevalence of asymptomatic diastolic dysfunction is not known, but there is reason to believe that the condition is common—especially in people with hypertension or advanced age. The finding of diastolic dysfunction in an asymptomatic patient is a risk factor for the future development of congestive heart failure and the early identification of such patients provides a window of opportunity to prevent progression of what appears to be preclinical heart disease (32, 33, 40). There are no data that might support the use of treatment directed primarily at the diastolic dysfunction. Rather, the goal should be an aggressive management of hypertension and other potential causes of diastolic dysfunction.

12 Dec 2003

19:40

382

AR

GAASCH

AR206-ME55-22.tex

¥

AR206-ME55-22.sgm

LaTeX2e(2002/01/18)

P1: GBC

ZILE

ACUTE DIASTOLIC HEART FAILURE The initial management of acute heart failure and pulmonary edema consists of measures that relieve pulmonary congestion while maintaining oxygenation, arterial pressure, and perfusion of vital organs. With few exceptions, the initial treatment of patients with diastolic heart failure is similar to that used in those with systolic heart failure. Treatment is frequently initiated prior to hospitalization, often in an ambulance, and later in a hospital emergency department.

Annu. Rev. Med. 2004.55:373-394. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Prehospital Management Airway management and the administration of oxygen should be the first consideration. A face mask is generally used, but there is evidence that continuous positive airway pressure can improve lung mechanics, lessen the work of breathing, and reduce the need for intubation (53). Morphine reduces anxiety and has vasodilator properties, but its use can increase the need for mechanical ventilation. A second goal should be to reduce the pulmonary venous pressure. This can be achieved safely and effectively with sublingual or intravenous nitroglycerin (10–50 µg/min depending on clinical response). The use of nitroglycerine avoids electrolyte disturbances that can be seen when intravenous diuretics are administered as the initial treatment. Indeed, when treatments with nitrates, furosemide, and morphine are compared, optimal clinical outcomes are seen with nitrates (54). Moreover, the effects of nitrates are rapidly reversible in the event of hypotension. Rotating tourniquets also can effect a rapidly reversible reduction in central venous pressures. If the prehospitalization period is expected to be prolonged, the intravenous administration of nitroprusside (0.1–10 µg/kg/min) may be necessary in patients with severe hypertension, but there are no clear indications for angiotensin-converting enzyme (ACE) inhibitors or beta-adrenergic agonists during the prehospitalization period.

Initial Hospital Treatment On admission to the emergency department, the diagnosis of heart failure should be confirmed and associated, or complicating problems should be considered (e.g., pneumonia, myocardial infarction, pulmonary embolism, dissection of the aorta). At the same time, treatment should be initiated. In most patients, arterial hypoxemia can be reversed by oxygen administration with a Venturi mask. If this is not effective, continuous positive airway pressure can be used (vide supra). Endotracheal intubation may be required if arterial oxygenation cannot be maintained or if there is progressive hypercapnia.

OXYGEN

This agent (administered intravenously in a dose of 3–5 mg over several minutes) diminishes the patient’s distress and reduces the work of breathing. Morphine achieves its beneficial hemodynamic effects by acting as a vasodilator

MORPHINE

12 Dec 2003

19:40

AR

AR206-ME55-22.tex

AR206-ME55-22.sgm

LaTeX2e(2002/01/18)

DIASTOLIC HEART FAILURE

P1: GBC

383

Annu. Rev. Med. 2004.55:373-394. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

and thereby pooling blood in the splanenic circulation. Special caution is necessary if the pulmonary edema is associated with hypotension, stroke, or independent pulmonary disease—especially in patients with hypercapnia. PRELOAD REDUCTION A decrease in left atrial and pulmonary venous pressure is obviously desirable in patients with acute pulmonary edema. In patients with LV systolic dysfunction and acute exacerbation of chronic heart failure (most of whom have an expanded central blood volume), a substantial reduction in pulmonary venous pressure can be achieved without a significant drop in arterial pressure. However, patients with diastolic heart failure often develop a dramatic decrease in arterial pressure when attempts are made to reduce preload. This occurs as a consequence of a steep LV diastolic pressure-volume relation; even a small reduction in diastolic volume can result in relatively large reduction in LV diastolic pressure and systemic arterial pressure. Therefore, if there is reason to believe that diastolic heart failure is present, the initial attempts at preload reduction should be conservative. Diuretics are effective and commonly used in the initial treatment. Furosamide is administered intravenously at an initial dose of 40–80 mg; subsequent doses depend on the response to the initial dose. Intravenous nitroglycerine (10–50 µg/min) is also used to reduce preload. It has the advantage of being anti-ischemic and does not result in electrolyte abnormalities. Nesiritide (0.015–0.06 µg/kg/min) produces a dose-dependent decrease in pulmonary capillary wedge pressure and systemic vascular resistance, and an increase in cardiac output in patients with systolic heart failure (55). There is little published experience in patients with diastolic heart failure. AFTERLOAD REDUCTION Many, if not most, patients with acute pulmonary edema and diastolic heart failure are hypertensive (56). Although nitroglycerine or nesiritide are both effective in reducing blood pressure and relieving pulmonary edema, nitroprusside is the vasodilator of choice when a substantial reduction in pressure is required. It is administered by intravenous infusion a dose of 0.1–10 µg/kg/min; the dose is adjusted to obtain the desired hemodynamic effects. Nitroprusside is used only in situations requiring short-term reductions in blood pressure; early arrangements should be made to substitute other antihypertensive agents. Beta-adrenergic receptor blockers may be used alone or in combination with nitroprusside.

CHRONIC DIASTOLIC HEART FAILURE Any attempt to develop a long-term therapeutic plan must be based on a careful consideration of the cause of the diastolic dysfunction and its potential response to treatment. For example, verapamil can be effective in symptomatic patients who have hypertrophic cardiomyopathy (57), but it is contraindicated in patients who have cardiac amyloidosis (58). Coronary artery disease, hypertensive heart disease,

12 Dec 2003

19:40

384

AR

GAASCH

AR206-ME55-22.tex

¥

AR206-ME55-22.sgm

LaTeX2e(2002/01/18)

P1: GBC

ZILE

Annu. Rev. Med. 2004.55:373-394. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

chronic constrictive pericarditis, and aortic stenosis provide relatively specific therapeutic targets, but the problem is often less specific. For example, the complex mechanisms that underlie heart failure in elderly patients with hypertensive hypertrophic cardiomyopathy tend to preclude simple therapeutic interventions (59). In general, emphasis is placed on control of arterial hypertension, management of the congestive state, maintenance of normal sinus rhythm, and prevention of myocardial ischemia. It is particularly important to avoid the known “precipitants” of heart failure (see “Pathophysiology,” above).

Specific Therapeutic Targets VENOUS CONGESTION The renin-angiotensin-aldosterone system is activated in patients who have chronic heart failure, but the mechanisms that evoke its activation remain unclear in patients who have LV diastolic dysfunction. In some, myocardial ischemia, uncontrolled hypertension, and excessive dietary sodium may promote the development of congestion, whereas in others, low systemic vascular resistance or low arterial pressure may contribute to salt and water retention (60). Elevated venous pressure can directly cause renal sodium retention (61). Despite a limited understanding of the pathogenesis of salt and water retention in patients with diastolic dysfunction, diuretics remain the mainstay of therapy for venous congestion. After the initial treatment (vide supra), salt restriction is necessary, and long-term administration of a diuretic is usually required. It should be recognized that diuretics have the potential to reduce cardiac output, especially in patients with small LV chambers. With the exception of their antihypertensive effects, diuretics do not alter the primary disease processes that lead to diastolic dysfunction. Thiazide diuretics (HCTZ 25–50 mg p.o. QD) can suffice for management of mild heart failure, but the side effects of carbohydrate intolerance and hyperuricemia can be undesirable. Loop diuretics such as furosemide (40–240 mg p.o. QD) are more potent than the thiazide diuretics, especially when the glomerular filtration rate is reduced. The combination of furosemide and a thiazide diuretic can be especially useful when edema is refractory to either agent alone. For patients who develop hypokalemia, the potassium-sparing diuretic spironolactone may be added. Spironolactone also has the potential to retard the fibrosis that contributes to abnormal chamber stiffness. The reduction in blood volume produced by diuretics may trigger an increase in sympathetic tone and renin-angiotensin activation, which can lead to vasoconstriction and worsening of the pathophysiology. Some vasodilators, particularly nitrates and pure arteriolar vasodilators, evoke a similar response. ACE inhibitors (and beta-blockers) blunt the neurohormonal activation and decrease the salt and water retention that complicates the treatment of heart failure. In addition, the ACE inhibitors may also have salutary effects on the active and passive properties of the left ventricle. If hypotension does not limit their use, ACE inhibitors (e.g., Lisinopril 10–40 mg p.o. QD) can at the very least provide useful adjunctive therapy in

12 Dec 2003

19:40

AR

AR206-ME55-22.tex

AR206-ME55-22.sgm

LaTeX2e(2002/01/18)

DIASTOLIC HEART FAILURE

P1: GBC

385

Annu. Rev. Med. 2004.55:373-394. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

patients with diastolic dysfunction, particularly those with evidence of a chronic congestive state. ATRIAL ARRHYTHMIAS Rapid atrial fibrillation in patients with LV diastolic dysfunction is usually accompanied by a substantial increase in ventricular diastolic and atrial pressures, leading to pulmonary edema and hypotension. Overt decompensation occurs because of inadequate time for complete ventricular relaxation and also because of the loss of atrial mechanical function and its contribution to ventricular filling. An attempt to restore and maintain sinus rhythm is mandatory. Direct current cardioversion may be necessary on an emergency basis. In less urgent situations, electrical or chemical cardioversion can be performed after rate control with beta-blockers, calcium channel blockers, or digitalis.

Tachycardia is poorly tolerated in most cardiac disorders. Atrial tachyarrhythmias, and even sinus tachycardia, have a negative impact on diastolic function for several reasons. A rapid heart rate causes an increase in myocardial oxygen demand and a decrease in coronary perfusion time, which can promote ischemic diastolic dysfunction even in the absence of coronary artery disease. In addition, tachycardia does not allow sufficient time for relaxation, and as a result there is incomplete relaxation between beats, which causes an increase in diastolic pressure relative to volume. Tachycardia also reduces the LV diastolic filling time and the coronary perfusion time. Accordingly, most clinicians use betablockers or calcium channel blockers to prevent excessive tachycardia and produce a relative bradycardia in patients who have diastolic dysfunction. It should be recognized, however, that bradycardia can result in a fall in cardiac output despite some potential for improved filling pressures. Such considerations underscore the need for individualizing therapeutic interventions that affect heart rate; an initial goal might be a resting rate of approximately 65–70 bpm with a blunted exerciseinduced increase in heart rate (and a blunted increase in blood pressure during exercise). Although the optimal heart rate for hypertrophic or failing hearts is not known, it is likely that such hearts would function most efficiently at relatively slow rates. This has several potentially beneficial effects that are largely related to the salutary effects on myocardial energetics and the prolonged diastolic interval that allows complete relaxation between beats. Furthermore, hypertrophied and failing hearts exhibit a flat or even negative force-frequency relationship, and in contrast to normal hearts, function may improve as the rate is slowed (62, 63).

RATE CONTROL

MYOCARDIAL ISCHEMIA An extensive clinical and experimental literature documents the deleterious effect of ischemia on diastolic function of the left ventricle. A transient increase in LV stiffness and diastolic pressure develops during myocardial ischemia caused by coronary spasm, exercise, rapid atrial pacing, angioplasty balloon inflation, and spontaneous angina (64). Ischemia can be treated with nitrates, beta-blockers, and calcium channel blockers, percutaneous coronary intervention,

12 Dec 2003

19:40

386

AR

GAASCH

AR206-ME55-22.tex

¥

AR206-ME55-22.sgm

LaTeX2e(2002/01/18)

P1: GBC

ZILE

Annu. Rev. Med. 2004.55:373-394. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

or coronary artery bypass surgery. When the signs of ischemic diastolic dysfunction are prominent, bypass surgery is appropriate (65); yet even after successful surgery there may be recurrent episodes of heart failure (66). HYPERTENSION AND HYPERTROPHY Several factors contribute to the diastolic dysfunction seen in hypertensive heart disease (67). First, the abnormal loading conditions imposed by arterial hypertension reduce LV relaxation and filling rates. Second, concentrically hypertrophied hearts exhibit increased passive stiffness (caused by a low LV volume-mass ratio and fibrosis of the myocardium) and impaired relaxation that is independent of hemodynamic loads. Third, limited coronary vascular reserve can be responsible for myocardial ischemia, even in the absence of epicardial coronary disease. Each of these factors should be considered in the treatment of patients with hypertensive heart disease and diastolic dysfunction. Abnormalities of diastolic function can be detected in asymptomatic hypertensive patients with or without measurable hypertrophy (68). Adequate control of the arterial pressure in these patients with preclinical heart disease should favorably alter loading conditions in the short term, while, in the long-term, promoting regression of hypertrophy. Although the short-term treatment of elevated systemic arterial pressure tends to augment diastolic function, the effect of load reduction is difficult to demonstrate during long-term therapy; indeed there is considerable variation in the effects of different antihypertensive agents on myocardial relaxation. For example, despite an equivalent reduction in arterial pressure, nifedipine augments LV filling rate and other relaxation indices, but propranolol does not (69). Some studies of patients who have hypertensive heart disease indicate that diastolic dysfunction improves as LV hypertrophy regresses (70, 71). Other studies have confirmed improved diastolic function, prolonged exercise duration, and better heart failure scores in verapamil-treated patients who have hypertensive heart disease and clinically significant LV diastolic dysfunction, but these clinical benefits were not closely related to changes in blood pressure or heart rate (72). Differences in the effects of treatment on diastolic function probably depend on the amount of hypertrophy regression, the alterations in LV loading conditions, the direct myocardial effect of the antihypertensive agent, and possibly changes in coronary reserve. Progressive interstitial fibrosis accompanies the hypertrophic response to systemic arterial hypertension; fibrosis can also be prominent in the hypertrophy seen with aortic stenosis and hypertrophic cardiomyopathy. This abnormal accumulation of fibrillar collagen is a result of enhanced collagen synthesis (and/or decreased degredation) by cardiac fibroblasts that is related in part to the activity of the reninangiotensin-aldosterone system. The important functional consequences of progressive interstitial and perivascular fibrosis include increased myocardial stiffness and impaired coronary flow reserve. In experimental studies, ACE inhibitors or spironolactone appears to protect against this exaggerated fibrous tissue response (73). Thus, the imperative to treat arterial hypertension may include prevention of the deleterious effects of angiotensin II and aldosterone. Angiotensin-converting

12 Dec 2003

19:40

AR

AR206-ME55-22.tex

AR206-ME55-22.sgm

LaTeX2e(2002/01/18)

DIASTOLIC HEART FAILURE

P1: GBC

387

Annu. Rev. Med. 2004.55:373-394. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

enzyme inhibitors are widely used, effective antihypertensive agents that can produce regression of LV hypertrophy, and a salutary effect on cardiac fibrosis may constitute an unexpected bonus. As a preventive or treatment strategy this has not yet been tested in humans with diastolic dysfunction. HYPERTROPHIC CARDIOMYOPATHY AND RELAXATION Although hypertrophic cardiomyopathy is a unique entity, diastolic dysfunction is an important component of its pathophysiology. The diastolic dysfunction is caused by increased passive stiffness of the ventricle, (due to a low LV volume-mass ratio, fibrosis, and fiber disarray) and abnormal myocardial relaxation (due to abnormal calcium metabolism, altered loading conditions, and nonuniformity) (74). As a result of depressed calcium sequestration by the sarcoplasmic reticulum and perhaps an increase in membrane calcium channels, myocardial calcium overload contributes to a slow or delayed myocardial relaxation (a slow and prolonged dissociation of actin-myosin), which leads to increased diastolic tension (75). Such prolonged relaxation can persist throughout the entire diastolic interval, especially in the presence of tachycardia. Assuming that the alterations in passive stiffness are relatively fixed and irreversible (which may not be true), medical therapy has generally been directed toward the relaxation abnormalities. The calcium channel blockers verapamil, diltiazem, and nifedipine can improve many of the abnormal indices of relaxation and provide symptomatic relief (75), but these agents do not directly benefit myocardial calcium homeostatis. Verapamil, the most widely used calcium channel blocker in hypertrophic cardiomyopathy, has a beneficial effect on angina and dyspnea; it also improves exercise capacity (57). Therapy is initiated at 120–240 mg/day and gradually increased to 360– 480 mg/day; the optimal dose is determined by the symptomatic response of the patient. Unfortunately, the vasodilating effects of calcium channel blockers can lead unpredictably to intensification of an outflow obstruction or hypotension even in the absence of obstruction. Although beta-blockers can impair ventricular relaxation, they are commonly used in patients who have hypertrophic cardiomyopathy. Angina, dyspnea, and presyncopal symptoms tend to improve during treatment. Angina seems to respond more favorably than dyspnea. Thus, treatment with metoprolol may be initiated at 25–50 mg. twice a day; the dose is then gradually increased to 100–200 mg depending on the clinical response. Some patients require higher doses to achieve a beneficial effect on exercise capacity and symptoms (76). When symptoms prove refractory to medical therapy, alcohol ablation of the interventricular septum, surgical procedures such as myotomy-myectomy, mitral valve replacement, or atrioventricular sequential pacemakers may favorably influence the diastolic properties of the left ventricle and produce symptomatic relief in selected patients with hypertrophic obstructive cardiomyopathy. EXERCISE INTOLERANCE Patients who have a history of diastolic heart failure (even those who have diastolic dysfunction and little or no congestion) often exhibit

12 Dec 2003

19:40

Annu. Rev. Med. 2004.55:373-394. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

388

AR

GAASCH

AR206-ME55-22.tex

¥

AR206-ME55-22.sgm

LaTeX2e(2002/01/18)

P1: GBC

ZILE

substantial exercise intolerance (vide supra). Given the limited understanding of the precise factors responsible for dyspnea and fatigue (77, 78), it has been difficult to develop a standard treatment plan for patients with LV diastolic dysfunction. Certainly, hypertension, myocardial ischemia, and clinically apparent congestion must be treated, but caution must be exercised to avoid even mild volume depletion that can contribute to a reduced cardiac output. A most important (and largely ignored) treatment is directed against the deconditioning that is prominent in many patients with diastolic heart failure. Cardiac rehabilitation programs can be very helpful in this regard (79). Although calcium channel blockers and beta blockers improve symptoms in some patients with LV diastolic dysfunction, the benefit on exercise capacity is not always paralleled by improved measures of LV diastolic function. For example, in symptomatic patients who have hypertrophic cardiomyopathy, a placebocontrolled double-blind comparison of the effects of verapamil and propranolol on exercise tolerance indicated that both agents produced an increase in exercise duration; however, relaxation rate increased with verapamil and decreased with propranolol (80). The observation that such verapamil effects persist in the longterm (57) and that it is effective in patients who have other causes of diastolic dysfunction (72) makes this agent a treatment of choice for exercise intolerance. Beta-blockers are an acceptable alternative, despite a direct depressant effect on myocardial relaxation. ACE inhibitors or angiotensin II–receptor-blocking agents also have the potential to improve exercise tolerance in patients who have diastolic dysfunction. For example, treatment with losartan is associated with an increase in exercise capacity and improved quality of life in patients who have hypertensive cardiovascular disease and documented diastolic dysfunction (81). These responses are similar to those observed in patients treated with verapamil (72). The salutory effects of losartin and verapamil are at least in part related to their antihypertensive effect.

Positive Inotropic Drugs Most patients with heart failure caused by LV systolic dysfunction benefit from treatment with positive inotropic agents. Such therapy is generally not used in the long-term treatment of patients with diastolic heart failure because the LV ejection fraction is preserved and there appears to be little potential for a beneficial effect. Moreover, positive inotropic agents have the potential to worsen the pathophysiologic processes that cause diastolic dysfunction. Digitalis, by inhibiting the sodium-potassium adenosine triphosphatase pump, augments intracellular calcium through a sodium-calcium exchange mechanism and enhances the contractile state. By doing so, digitalis produces an increase in systolic energy demands while adding to a diastolic calcium overload. These effects may not be clinically apparent in many circumstances, but during hemodynamic stress or ischemia, digitalis may promote diastolic dysfunction (82). Data from the Digitalis Investigators

Candesartin

24

Treatment Study drug(s)

Duration (months)

2004

2004

X

X X

X X X

18

Nebivolol

2000 1000

X X X

12

Perindapril

1000 1000

Yes II–IV Yes >35

2004

X

X X X

Diuretic alone Diuretic+Ramiprel Diuretic+Irbesartan 12

1000 1000

Yes II–IV No >45

2006

X

X X X

24

Irbesartan

3600 3600

Yes II–IV Yes >45

I-Preserve

LaTeX2e(2002/01/18)

DIASTOLIC HEART FAILURE

Abbreviations: HF, heart failure; NYHA, New York Heart Association; MRI, magnetic resonance imaging; VO2, oxygen consumption.

X 2003

Completion of study (year)

2003

Brain natriuretic peptide

X X X X X X

6

MCC-135

500 230

Yes II–IV Yes >40

Hong Kong

AR206-ME55-22.sgm

X X X

6500 2500

Enrollment Total Diastolic heart failure

Yes II–III Yes >40

Seniors

AR206-ME55-22.tex

Endpoints & measurements Death Hospitalization Echocardiography Cardiac MRI Exercise tolerance Exercise VO2

Yes II–IV Yes >40

PEP-CHF

AR

Entrance criteria Signs and symptoms of HF NYHA functional class Prior hospitalization Ejection fraction (%)

MCC-135

19:40

CHARM

TABLE 4 Ongoing clinical trials in patients with diastolic heart failure. The CHARM-Preserved Trial data indicate that treatment with candesartin reduces hospitalization rates in patients with diastolic heart failure (87)

Annu. Rev. Med. 2004.55:373-394. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

12 Dec 2003 P1: GBC

389

12 Dec 2003

19:40

Annu. Rev. Med. 2004.55:373-394. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

390

AR

GAASCH

AR206-ME55-22.tex

¥

AR206-ME55-22.sgm

LaTeX2e(2002/01/18)

P1: GBC

ZILE

Group study, however, suggest that digitalis might have a beneficial effect, despite a normal LV ejection fraction, on some clinical outcome measures, such as heart failure hospitalizations (83). However, there appears to be a corresponding increase in endpoints related to myocardial ischemia and arrhythmias. Recognizing conflicting opinions on this issue, most clinicians do not use digitalis in patients with diastolic heart failure. Beta-adrenergic agonists, by increasing intracellular cyclic adenosine monophosphate, enhance calcium sequestration by the sarcoplasmic reticulum and thereby promote a more rapid and complete myocardial relaxation between beats (84). Beta-agonists can also increase venous capacitance, which leads to a reduction in ventricular filling pressures. Phosphodiesterase inhibitors can produce similar salutary effects on myocardial relaxation and venous capacitance (85). Unfortunately, all cyclic adenosine monophosphate-dependent agents promote calcium influx into the cell and augment myocardial energy demands. Thus, dopamine, amrinone, and similar agents are used only in the short-term management of acute diastolic heart failure.

ONGOING CLINICAL TRIALS There are at least six ongoing large randomized placebo-controlled therapeutic trials in patients with diastolic heart failure (see Table 4). All six require that the patients exhibit signs and symptoms of heart failure, a recent hospitalization for heart failure, and no more than a mild reduction in the ejection fraction. Three of them include subgroups with low ejection fractions. Four are directed at inhibition or blocking the renin-angiotensin system; one uses a beta-adrenergic blocker; and one is directed at calcium homeostatis. These studies will make available a variety of information (especially echocardiographic and exercise data) that should better characterize the syndrome of diastolic heart failure and hopefully provide direction for management and treatment (86). The Annual Review of Medicine is online at http://med.annualreviews.org

LITERATURE CITED 1. Henderson Y. 1906. The volume curve of the ventricles of the mammalian heart, and the significance of this curve in respect to the mechanics of the heart beat and the filling of the ventricles. Am. J. Physiol. 16:32–36 2. Wiggers CJ, Katz LN. 1922. The contour of the ventricular volume curves under different conditions. Am. J. Physiol. 58:439– 75

3. Meek WJ. 1927. Cardiac tonus. Physiol. Rev. 7:259–87 4. Gaasch WH, Battle WE, Oboler AO, et al. 1972. Left ventricular stress and compliance in man: with special reference to normalized ventricular function curves. Circulation 45:746–62 5. Dwyer EM Jr. 1970. Left ventricular pressure-volume alterations and regional disorders of contraction during

12 Dec 2003

19:40

AR

AR206-ME55-22.tex

AR206-ME55-22.sgm

LaTeX2e(2002/01/18)

DIASTOLIC HEART FAILURE

6.

7.

Annu. Rev. Med. 2004.55:373-394. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

myocardial ischemia induced by atrial pacing. Circulation 42:1111–22 McLaurin LP, Rolett EL, Grossman W. 1973. Impaired left ventricular relaxation during pacing-induced ischemia. Am. J. Cardiol. 32:751–57 Covell JW, Ross J Jr. 1973. Nature and significance of alterations in myocardial compliance. Am. J. Cardiol. 32:449–55 Mirsky I. 1976. Assessment of passive elastic stiffness of cardiac muscle. Mathematical concepts, physiologic and clinical considerations, directions of future research. Prog. Cardiovasc. Dis. 18:277–308 Glantz SA, Parmley WW. 1978. Factors which affect the diastolic pressure-volume curve. Circ. Res. 42:171–80 Ross J Jr. 1979. Acute displacement of the diastolic pressure-volume curve of the left ventricle. Role of the pericardium and the right ventricle. Circulation 59:32–37 Brutsaert DL, Housmans PR, Goethals MA. 1980. Dual control of relaxation: its role in the ventricular function in the mammalian heart. Circ. Res. 47:637–52 Gaasch WH, Levine HJ, Quinones MA, et al. 1976. Left ventricular compliance: mechanisms and clinical implications. Am. J. Cardiol. 38:645–53 Grossman W, McLaurin LP. 1976. Diastolic properties of the left ventricle. Ann. Intern. Med. 84:316–26 Echeverria HH, Bilsker MS, Myerburg RJ, et al. 1983. Congestive heart failure: echocardiographic insights. Am. J. Med. 75:750–55 Dougherty AH, Naccarelli GV, Gray EL, et al. 1984. Congestive heart failure with normal systolic function. Am. J. Cardiol. 54:778–82 Soufer R, Wohlgelernter D, Vita NA, et al. 1985. Intact systolic left ventricular function in clinical congestive heart failure. Am. J. Cardiol. 55:1032–36 Kunis R, Greenberg H, Yeoh CB, et al. 1985. Coronary revascularization for recurrent pulmonary edema in elderly patients with ischemic heart disease and pre-

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

P1: GBC

391

served ventricular function. N. Engl. J. Med. 313:1207–10 Topol EJ, Traill TA, Fortuin NJ. 1985. Hypertensive hypertrophic cardiomyopathy of the elderly. N. Engl. J. Med. 312:277–83 Given BD, Lee TH, Stone PH, et al. 1985. Nifedipine in severely hypertensive patients with congestive heart failure and preserved ventricular function. Arch. Intern. Med. 145:281–85 Caidahl K, Eriksson H, Hartford M, et al. 1988. Dyspnea of cardiac origin in 67 year old men: II. Relation to diastolic left ventricular function and mass. The study of men born in 1913. Br. Heart J. 59:329–38 Kessler KM. 1988. Heart failure with normal systolic function: update on prevalence, differential diagnosis, prognosis, and therapy. Arch. Intern. Med. 148:2109–11 Kessler KM. 1989. Diastolic heart failure: diagnosis and management. Hosp. Prac. July:111–38 Grossman W. 1990. Diastolic dysfunction and congestive heart failure. Circulation 81(Suppl. III):1–7 Stauffer JC, Gaasch WH. 1990. Recognition and treatment of left ventricular diastolic dysfunction. Prog. Cardiovasc. Dis. 32:319–32 Cohen GJ, Pietrolungo DO, Thomas JD, et al. 1996. A practice guide to assessment of ventricular diastolic function using Doppler echocardiography. J. Am. Coll. Card. 27:1753–60 Rakowski H, Appleton C, Chan KL, et al. 1996. Canadian consensus recommendations for measurement and reporting of diastolic dysfunction by echocardiography. J. Am. Soc. Echocardiogr. 9:736–60 Nishimura RA, Tajik AJ. 1997. Evaluation of diastolic filling of left ventricle in health and disease: Doppler echocardiography is the clinician’s rosetta stone. J. Am. Coll. Cardiol. 30:8–18 Mulvagh S, Quinones MA, Kleiman NS, et al. 1992. Estimation of left ventricular end diastolic pressure from Doppler transmitral flow velocity in cardiac patients

12 Dec 2003

19:40

392

29.

Annu. Rev. Med. 2004.55:373-394. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

30.

31.

32.

33.

34.

35.

36.

37.

AR

GAASCH

AR206-ME55-22.tex

¥

AR206-ME55-22.sgm

LaTeX2e(2002/01/18)

P1: GBC

ZILE

independent of systolic performance. J. Am. Coll. Cardiol. 20:112–19 Garcia MJ, Ares MA, Asher C, et al. 1997. Color M-mode flow velocity propagation: an index of early left ventricular filling that combined with pulsed Doppler peak E velocity may predict capillary wedge pressure. J. Am. Coll. Cardiol. 29:448– 54 Nagueh SF, Middleton KJ, Kopelen HA, et al. 1997. Doppler tissue imaging: a non-invasive technique for evaluation of left ventricular relaxation and estimation of filling pressures. J. Am. Coll. Cardiol. 30:1527–33 Ommen SR, Nishimura RA, Appleton CP, et al. 2000. The clinical utility of Doppler echocardiography and tissue Doppler imaging in estimation of left ventricular filling pressure: a comparative simultaneous Doppler-catheterization study. Circulation 102:1788–94 Aurigemma GP, Gottdiener JS, Shemanski L, et al. 2001. Predictive valve of systolic and diastolic function for incident congestive heart failure in the elderly: the cardiovascular health study. Am. J. Coll. Cardiol. 37:1042–48 Redfield MM, Jacobsen SJ, Burnett JC, et al. 2003. Burden of systolic and diastolic ventricular dysfunction in the community. J. Am. Med. Assoc. 289:194–202 Senni M, Tribouilloy CM, Rodeheffer RJ, et al. 1998. Congestive heart failure in the community: a study of all incident cases in Olmsted county, Minnesota in 1991. Circulation 98:2282–89 Vasan RS, Larson MG, Benjamin EJ, et al. 1999. Congestive heart failure in subjects with normal versus reduced left ventricular ejection fraction: prevalence and mortality in a population-based cohort. J. Am. Coll. Cardiol. 33:1948–55 Senni M, Redfied MM. 2001. Heart failure with preserved systolic function: a different natural history? J. Am. Coll. Cardiol. 38:1277–82 Kitzman DW, Little WC, Brubaker PH,

38.

39.

40.

41.

42.

43.

44.

45.

46.

et al. 2002. Pathophysiological characterization of isolated diastolic heart failure in comparison to systolic heart failure. J. Am. Med. Assoc. 288:2144–50 Zile MR, Brutsaert DL. 2002. New concepts in diastolic dysfunction and diastolic heart failure: part 1. Diagnosis, prognosis and measurements of diastolic function. Circulation 105:1387–93 Zile MR, Brutsaert DL. 2002. New concepts in diastolic dysfunction and diastolic heart failure: part II. Causal mechanisms and treatment. Circulation 105:1505–8 Gaasch WH. 1994. Diagnosis and treatment of heart failure based on left venricular systolic or diastolic dysfunction. J. Am. Med. Assoc. 271:1276–80 Gilbert JC, Glantz SA. 1989. Determinants of left ventricular filling and of the diastolic pressure-volume relationship. Circ. Res. 64:827–52 Tsuyuki RT, McKelvic RS, Arnold JM, et al. 2001. Acute precipitants of congestive heart failure exacerbations. Arch. Intern. Med. 161:2337–42 Wang CH, Weisel RD, Liu PP, et al. 2003. Glitazones and heart failure: critical appraisal for the clinician. Circulation 107:1350–54 Aurigemma GP, Gaasch WH, McLaughlin M, et al. 1995. Reduced left ventricular systolic pump performance and depressed myocardial contractile function in patients >65 years of age with normal ejection fraction and high relative wall thickness. Am. J. Cardiol. 76:702–5 Cuocolo A, Sax FL, Brush JE, et al. 1990. Left ventricular hypertrophy and impaired diastolic filling in essential hypertension: diastolic mechanisms for systolic dysfunction during exercise. Circulation 81:978– 86 Kitzman DW, Higginbotham BM, Cobb FR et al. 1991. Exercise intolerance in patients with heart failure and preserved left ventricular systolic function: failure of the Frank-Starling mechanism. J. Am. Coll. Cardiol. 17:1065–72

12 Dec 2003

19:40

AR

AR206-ME55-22.tex

AR206-ME55-22.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:373-394. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

DIASTOLIC HEART FAILURE 47. Paulus WJ, Eur. Study Group Diastolic Heart Fail. 1998. How to diagnose diastolic heart failure. Eur. Heart J. 19:990–1003 48. Vasan RS, Levy D. 2000. Defining diastolic heart failure: a call for standardized diagnostic criteria. Circulation 101:2118–21 49. Zile MR, Gaasch WH, Carroll JD, et al. 2001. Heart failure with a normal ejection fraction: Is measurement of diastolic function necessary to make the diagnosis of diastolic heart failure? Circulation 104:779– 82 50. Dao Q, Krishnaswamy P, Kazanegra R, et al. 2001. Utility of B-type natriuretic peptide in the diagnosis of congestive heart failure in an urgent care setting. J. Am. Coll. Cardiol. 37:379–85 51. Lubien E, DeMaria A. Krishnaswamy P, Clopton P, et al. 2002. Utility of Bnatriuretic peptide in detecting diastolic dysfunction: comparison with Doppler velocity recordings. Circulation 105(5):595– 601 52. Vinch CS, Aurigemma GP, Hill JC, et al. 2003. Usefulness of clinical variables, echocardiography, and levels of brain natriuretic peptide and norepinepherine to distinguish systolic and diastolic causes of heart failure. Am. J. Cardiol. 91:1140–43 53. Pang D, Keenan SP, Cook DJ, et al. 1998. The effect of positive pressure airway support on mortality and the need for intubation in cardiogenic pulmonary edema: a systematic review. Chest 114(4):1185–92 54. Hoffmann JR, Reynolds S. 1987. Comparison of nitroglycerine, morphine, and furosemide in treatment of presumed prehospital pulmonary edema. Chest 92(4): 586–93 55. Colucci WS. 2001. Nesiritide for the treatment of decompensated heart failure. J. Card. Fail. 7:92–100 56. Gandhi SK, Powers JC, Nomeier AM, et al. 2001. The pathogenesis of acute pulmonary edema associated with hypertension. N. Engl. J. Med. 344:17–22 57. Bonow RO, Dilsizian V, Rosing DR, et al. 1985. Verapamil-induced improvement in

58.

59.

60.

61.

62.

63.

64.

65.

66.

P1: GBC

393

left ventricular filling and increased exercise tolerance in patients with hypertrophic cardiomyopathy: short and long term results. Circulation 72:853–64 Pollak A, Falk RH. 1993. Left ventricular systolic dysfunction precipitated by verapamil in cardiac amyloidosis. Chest 104:618–22 Topol EJ, Traill TA, Fortuin NJ. 1985. Hypertensive hypertrophic cardiomyopathy of the elderly. N. Engl. J. Med. 312:277–83 Anand IS, Chandrashekhar Y, Ferrari R, et al. 1992. Pathogenesis of congestive state in chronic obstructive pulmonary disease: studies of body water and sodium, renal function, hemodynamics and plasma hormones during edema and after recovery. Circulation 86:12–21 Firth JD, Raine AEG, Ledingham JGG. 1998. Raised venous pressure: a direct cause of renal sodium retention in edema. Lancet 1:1033–35 Liu CP, Ting CT, Lawrence W, et al. 1993. Diminished contractile response to increased heart rate in intact human left ventricular hypertrophy: systolic versus diastolic determinants. Circulation 88:1893– 906 Mulieri LA, Hasenfuss G, Leavitt B, et al. 1992. Altered myocardial force-frequency relation in human heart failure. Circulation 85:1743–50 Paulus WJ, Bronzwaer JGF, de Bruyne B, et al. 1994. Different effects of “supply” and “demand” ischemia on left ventricular diastolic function in humans. See Ref. 86, pp. 286–305 Kunis R, Greenberg H, Yeoh CG, et al. 1985. Coronary revascularization for recurrent pulmonary edema in elderly patients with ischemic heart disease and preserved ventricular function. N. Engl. J. Med. 313:1207–10 Kramer K, Kirkman P, Kitzman D. 2000. Flash pulmonary edema: association with hypertension and reoccurrence despite coronary revascularization. Am. Heart. J. 140(3):451–55

12 Dec 2003

19:40

Annu. Rev. Med. 2004.55:373-394. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

394

AR

GAASCH

AR206-ME55-22.tex

¥

AR206-ME55-22.sgm

LaTeX2e(2002/01/18)

P1: GBC

ZILE

67. Hoit BD, Walsh RA. 1994. Diastolic function in hypertensive heart disease. See Ref. 86, pp. 354–72 68. Inouye I, Massie B, Loge D, et al. 1984. Abnormal left ventricular filling: an early finding in mild to moderate systemic hypertension. Am. J. Cardiol. 53:120–26 69. Zusman RM. 1989. Nifedipine but not propranolol improves left ventricular systolic and diastolic function in patients with hypertension. Am. J. Cardiol. 64:F51–61 70. Smith VE, White WB, Meeran MK, et al. 1986. Improved left ventricular filling accompanies reduced left ventricular mass during therapy of essential hypertension. J. Am. Coll. Cardiol. 8:1449–54 71. Schulman SP, Weiss JL, Becker LC, et al. 1990. The effects of antihypertensive therapy on left ventricular mass in elderly patients. N. Engl. J. Med. 322:1350–56 72. Setaro JF, Zaret BL, Schulman DS, et al. 1990. Usefulness of verapamil for congestive heart failure associated with abnormal left ventricular diastolic filling and normal left ventricular systolic performance. Am. J. Cardiol. 66:981–86 73. Weber KT, Brilla CG. 1991. Pathological hypertrophy and cardiac interstitium: fibrosis and renin-angiotensin-aldosterone system. Circulation 83:1849–65 74. Wigle ED, Sasson Z, Henderson MA, et al. 1985. Hypertrophic cardiomyopathy: the importance of the site and the extent of hypertrophy. Prog. Cardiovasc. Dis. 28:1– 83 75. Udelson JE, Bonow RO. 1994. Left ventricular diastolic function and calcium channel blockers in hypertrophic cardiomyopathy. See Ref. 86, pp. 465–89 76. Maron BJ, Bonow RO, Cannon RO, et al. 1987. Hypertrophic cardiomyopathy: interrelationships of clinical manifestations, pathophysiology and therapy. N. Engl. J. Med. 316:844–52 77. Packer M. 1990. Abnormalities of diastolic function as a potential cause of exercise intolerance in chronic heart failure. Circulation 81(Suppl. III):78–86

78. Chikamori T, Counihan PJ, Doi YL, et al. 1992. Mechanisms of exercise limitation in hypertrophic cardiomyopathy. J. Am. Coll. Cardiol. 19:507–12 79. Kitzman DW, Brubaker PH, Anderson RA, et al. 1999. Exercise training improves aerobic capacity in elderly patients with diastolic heart failure: a randomized, controlled trial. Circulation 100(1):296 (Abstr.) 80. Rosing DR, Kent KM, Maron BJ, et al. 1979. Verapamil therapy: a new approach to the pharmacologic treatment of hypertrophic cardiomyopathy. II. Effects on exercise capacity and symptomatic status. Circulation 60:1208–13 81. Warner JG, Metzger DC, Kitzman DW, et al. 1999. Losartan improves exercise tolerance in patients with diastolic dysfunction and a hypertensive response to exercise. J. Am. Coll. Cardiol. 33:1567–72 82. Lorell BH, Isoyama S, Grice WN, et al. 1988. Effects of ouabain and isoproterenol on left ventricular diastolic function during low-flow ischemia in isolated, bloodperfused rabbit hearts. Circ. Res. 63:457– 67 83. Massie BM, Abdalla I. 1998. Heart failure in patients with preserved left ventricular systolic function: Do digitalis glycosides have a role? Prog. Cardiovasc. Dis. 40:357–69 84. Lang RM, Carroll JD, Nakamura S, et al. 1988. Role of adrenoceptors and dopamine receptors in modulating left ventricular diastolic function. Circ. Res. 63:126–34 85. Monrad ES, McKay R, Baim DS, et al. 1984. Improvement in indexes of diastolic performance in patients with congestive heart failure treated with milrinone. Circulation 70:1030–37 86. Gaasch WH, LeWinter MM, eds. 1994. Left Ventricular Diastolic Dysfunction and Heart Failure. Philadelphia: Lea & Febiger 87. Yusuf S, Pfeffer MA, Swedberg K, et al. 2003. Effects of candesartin in patients with chronic heart failure and preserved left ventricular ejection fraction; the CHARMPreserved Trial. Lancet 362:777–81

5 Jan 2004

12:34

AR

AR206-ME55-23.tex

AR206-ME55-23.sgm

LaTeX2e(2002/01/18) P1: GCE 10.1146/annurev.med.55.091902.103810

Annu. Rev. Med. 2004. 55:395–417 doi: 10.1146/annurev.med.55.091902.103810 c 2004 by Annual Reviews. All rights reserved Copyright °

MECHANISMS OF PULMONARY FIBROSIS Victor J. Thannickal, Galen B. Toews, Eric S. White, Joseph P. Lynch III, and Fernando J. Martinez Annu. Rev. Med. 2004.55:395-417. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Division of Pulmonary and Critical Care Medicine, University of Michigan Medical Center, Ann Arbor, Michigan 48109; email: [email protected]

Key Words inflammation, regeneration, fibroblasts, cell differentiation, apoptosis ■ Abstract Tissue injury evokes highly conserved, tightly regulated inflammatory responses and less well-understood host repair responses. Both inflammation and repair involve the recruitment, activation, apoptosis, and eventual clearance of key effector cells. In this review, we propose the concept of pulmonary fibrosis as a dysregulated repair process that is perpetually “turned on” even though classical inflammatory pathways may be dampened or “switched off.” Significant regional heterogeneity, with varied histopathological patterns of inflammation and fibrosis, has been observed in individual patients with idiopathic pulmonary fibrosis. We discuss environmental factors and host response factors, such as genetic susceptibility and age, that may influence these varied manifestations. Better understanding of the mechanisms of lung repair, which include alveolar reepithelialization, myofibroblast differentiation/activation, and apoptosis, should offer more effective therapeutic options for progressive pulmonary fibrosis.

INTRODUCTION Pulmonary fibrosis results from a variety of insults to the lung that include toxic, autoimmune, drug-induced, infectious, or traumatic injuries. It represents one end of a spectrum of types of tissue responses to injury. Specific types of tissue responses are likely to depend on multiple host response factors including age, genetic susceptibility, and environmental factors. The resulting histopathological changes in the lung can be diverse with overlapping features, characterized by varying degrees of inflammation and fibrosis (1) (Figure 1). If the etiological agent is known, simple avoidance of the agent may result in spontaneous resolution; in other cases, a short course of steroids may be warranted but resolution can be expected. Complete resolution is more likely when the underlying histopathology is characterized by inflammation rather than by fibrosis. Druginduced lung diseases that produce desquamative interstitial pneumonia (DIP)like reactions or cellular nonspecific interstitial pneumonia (NSIP) are more likely to respond to simple avoidance of the drug with or without a short course of steroid therapy than is usual interstitial pneumonia (UIP), characterized by greater 0066-4219/04/0218-0395$14.00

395

5 Jan 2004

12:34

Annu. Rev. Med. 2004.55:395-417. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

396

AR

AR206-ME55-23.tex

AR206-ME55-23.sgm

LaTeX2e(2002/01/18)

P1: GCE

THANNICKAL ET AL.

Figure 1 Histopathological patterns of idiopathic interstitial pneumonias (IIPs) represent a spectrum of tissue reactions with varying degrees of inflammation and fibrosis. This reaction pattern probably depends on multiple factors, including age, genetic susceptibility, environmental factors, and perhaps the nature of the injurious agent. Abbreviations: DIP, desquamative interstitial pneumonia; RB-ILD, respiratory bronchiolitis-associated interstitial lung disease; LIP, lymphocytic interstitial pneumonia; COP, cryptogenic organizing pneumonia; NSIP, nonspecific interstitial pneumonia; AIP, acute interstitial pneumonia; UIP, usual interstitial pneumonia.

fibrosis (Figure 2). In some cases, these histopathological patterns occur without an identifiable etiological agent and are therefore termed idiopathic. Idiopathic UIP is synonymous with the clinical syndrome of idiopathic pulmonary fibrosis (IPF). These inflammatory/fibrotic disease processes are grouped into “clinicalradiological-pathological” entities known as the idiopathic interstitial pneumonias (IIPs) (2).

Pathogenetic Relationships Between the IIPs There is controversy on relationships between the IIPs. It has been suggested that UIP/IPF represents a distinct process/disease entity that is separate from other IIPs (3, 4). However, recent studies have convincingly demonstrated that different histopathologic subtypes of IIP may coexist in the same patient and even in the

5 Jan 2004

12:34

AR

AR206-ME55-23.tex

AR206-ME55-23.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:395-417. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

PULMONARY FIBROSIS

P1: GCE

397

same lobe of the lung (5, 6). This suggests that there may be significant overlap in the pathophysiology of these diseases, similar to the varied reaction patterns in the lung produced in response to known etiological agents and in collagen vascular diseases (CVD). However, an important difference between nonidiopathic (drug-, asbestos-, or CVD-associated) and idiopathic UIP is the apparent stability in the former, provided the etiological agents are avoided or the underlying CVD is treated, and the more progressive nature of the latter (7). Whether this difference is due to the continued presence of an unidentified etiological agent/injury or a fundamental difference in disease pathogenesis is not known. Nevertheless, accumulating evidence indicates that histopathological subtypes of IIP may represent a spectrum of tissue reactions to injury that depend on a number of genetic, environmental, and age-related modifiers (Figure 1). This concept is further supported by a recent study of familial pulmonary fibrosis related to a mutation in surfactant protein C with associated epithelial cell injury; a histopathological pattern of NSIP was noted in an affected family member of younger age, whereas UIP was noted in an older member (8). Better understanding and definition of the natural histories of “sporadic” cases of IIP may shed more light on the pathogenetic relationships within this intriguing group of lung disorders.

HOST RESPONSES TO TISSUE INJURY: INFLAMMATION AND REPAIR Host responses to injury, regardless of the type of injury, evoke certain stereotypical responses to protect the host from real or perceived danger. Inflammatory responses involve a complex set of interactions among soluble factors and cells that can arise in any tissue in response to traumatic, infectious, postischemic, toxic, or autoimmune injury. Nathan (9) has argued that “evolution did not anticipate surgery with aseptic technique . . . the body reacts to trauma as if the emergency is infection, until proven otherwise.” Thus, even in the absence of microbes, injured tissues signal classical inflammatory responses (10). Tissue injury is also likely to simultaneously trigger “repair responses” that involve fibroblasts and myofibroblasts, although mechanisms for the initiation, regulation, and resolution of repair are not well understood. The complex regulation of and interactions between inflammation and repair determine the eventual outcome of the responses to tissue injury. The orchestration of inflammation and repair, under normal conditions, is tightly controlled in a timeand context-dependent manner. Thus, the role of any cell, cytokine/mediator, matrix molecule, or intracellular signaling molecule has to be considered contextually in complex-systems biology. The proinflammatory transcription factor, NF-κB, that mediates proinflammatory gene expression during the onset of inflammation, induces expression of antiinflammatory genes and apoptosis of inflammatory cells during the resolution of inflammation (11). Transforming growth factor-β1 (TGF-β1), released soon after tissue injury, serves primarily as a proinflammatory

5 Jan 2004

12:34

398

AR

AR206-ME55-23.tex

AR206-ME55-23.sgm

LaTeX2e(2002/01/18)

P1: GCE

THANNICKAL ET AL.

molecule because of its potent neutrophil chemotactic effects (12), but then its function switches to resolution of inflammation and repair during the healing phase (13). Dysregulation of inflammation and/or repair, particularly the ability to temporally “switch off” specific signals, is likely to have pathological consequences.

Annu. Rev. Med. 2004.55:395-417. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

The Concept of Fibrosis as Dysregulated Repair Patients respond poorly to antiinflammatory therapies because there is little or no inflammation in advanced UIP/IPF, although it is possible that inflammation accompanied the initial injury in early phases of the disease. The most plausible explanation for the absence of significant inflammation in the chronic disease process we recognize as UIP/IPF is that tissue responses have successfully “switched off” classic inflammatory responses, but persist in a state of chronic repair. The nature of the etiological agent, as well as the resultant tissue injury and host response, may dictate the relative degrees of inflammation and repair. Furthermore, the persistence and/or recurrence of injury may alter the tissue microenvironment to influence subsequent host responses. Thus, persistent or recurrent injury in the context of a chronic repair process may be sufficient to dampen the classical inflammatory response. Alterations in the alveolar microenvironment of the lung probably account for the dysregulated repair and aberrant tissue remodeling that characterizes the progressive fibrosis in UIP/IPF. These alterations include imbalances in the production of soluble mediators that include chemokines, cytokines, growth factors, and eicosanoids. Aberrant angiogenesis, impaired fibrinolysis, reduced extracellular matrix turnover, and increased oxidative stress responses contribute to dysregulated tissue homeostasis (Figure 3). Such alterations in the alveolar microenvironment favor the loss of epithelial cells and accumulation of fibroblasts and myofibroblasts, leading to unrelenting, progressive fibrosis. For the remainder of this discussion, we focus on the critical processes involved in this dysregulated repair process and, in particular, on the ineffectiveness of alveolar reepithelialization and on fibroblast/myofibroblast activation, key elements in the pathogenesis of UIP/IPF.

ALVEOLAR EPITHELIAL CELLS: REEPITHELIALIZATION FAILURE Alveolar epithelial cell (AEC) injury is an early and consistent finding in UIP/IPF (14–16). Early indications that AEC injury may be important in the pathogenesis of progressive pulmonary fibrosis (17, 18) have been substantiated by more recent evidence in familial pulmonary fibrosis (8). Normal AEC regeneration and reepithelialization of the alveolar wall are critical for normal healing without fibrosis. The ability to rapidly accomplish this explains, in large part, the “scarless” wound healing in fetal wounds (19). Impaired or delayed reepithelialization may result

5 Jan 2004

12:34

AR

AR206-ME55-23.tex

AR206-ME55-23.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:395-417. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

PULMONARY FIBROSIS

P1: GCE

399

Figure 3 Progressive pulmonary fibrosis results from dynamic alterations in the alveolar microenvironment that eventually promote loss of alveolar epithelial cells and accumulation of activated fibroblasts/myofibroblasts. Such alterations include the presence or activation of profibrotic cytokines, growth factors, and chemokines; eicosanoid imbalance with increased production of profibrotic leukotrienes and deficiency in prostaglandin E2; impaired fibrinolysis; overproduction of TIMPs relative to MMPs; and a state of elevated oxidative stress. Abbreviations: TIMPs, tissue inhibitors of matrix metalloproteinases; MMPs, matrix metalloproteinases.

from loss of proliferative capacity, increased apoptosis, or ineffective migration of AECs (Figure 4). There appears to be significant heterogeneity in the proliferative phenotype of AECs in UIP/IPF. Higher rates of AEC proliferation have been observed at the bronchoalveolar junction in UIP/IPF lungs (20). Reduced proliferative capacity of type II AECs and/or the inability to differentiate into type I AECs have also been observed in human pulmonary fibrosis (16) and in explant lung models of fibrosis (21). Few studies have examined the relative concentrations and expression of epithelial cell mitogens in UIP/IPF, although levels of hepatocyte growth factor appear to be elevated in bronchoalveolar lavage (BAL) (22, 23). In animal models of fibrosis, a deficiency of granulocyte macrophage-colony stimulating factor (GMCSF) worsens the fibrotic response to lung injury (24, 25). This may be attributed to loss of epithelial regenerative capacity, since GM-CSF is an AEC mitogen and, in addition, induces production of the antifibrotic eicosanoid, prostaglandin E2 (PGE2) (26). In other animal models, GM-CSF has been shown to exert profibrotic

5 Jan 2004

12:34

Annu. Rev. Med. 2004.55:395-417. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

400

AR

AR206-ME55-23.tex

AR206-ME55-23.sgm

LaTeX2e(2002/01/18)

P1: GCE

THANNICKAL ET AL.

effects (27, 28). Such differences may be related to variations in model systems and experimental design, once again highlighting the importance of spatial, temporal, and context-dependent effects of multifunctional cytokines. Ineffective repair may also be explained by increased rates of AEC “dropout” resulting from apoptotic or necrotic cell death. AECs in UIP/IPF have been reported to undergo increased apoptosis by a Fas ligand–mediated pathway (29). The generation of angiotensin-II, either by autocrine or paracrine mechanisms, appears to induce AEC apoptosis by this pathway (30, 31). More recently, the profibrotic cytokine TGF-β1 has been shown to induce Fas-mediated apoptosis of AECs (32). Mitochondrial-mediated apoptotic pathways may also contribute to increased apoptotic rates of AECs in UIP/IPF (33). Extrinsic oxidative stress promotes AEC apoptosis in association with reductions in glutathione (GSH) levels (34–37). TGF-β1 contributes to increased oxidative stress in UIP/IPF by inhibiting the synthesis of GSH (38) and by increasing the production of oxidants by myofibroblasts (39). The concept that AEC apoptosis is involved in the initiation and/or progression of fibrotic lung disease is also supported by various experimental animal models of fibrosis (40–42). Another alteration of AEC phenotype that may contribute to a failure of alveolar reepithelialization is ineffective or delayed migration, but the presence of such a phenotype has not been convincingly demonstrated in human UIP/IPF. In the bleomycin animal model of fibrosis, type II AECs express membrane type-I matrix metalloproteinase (MT1-MMP) and activate MMP-2 on their cell surfaces, an effect that appears to be critical for AEC migration during the repair process following lung injury (43). The roles of MMPs and their tissue inhibitors (TIMPs) in lung injury and repair are complex, involving multiple functions in addition to cell migration (44–46). Although MMP-2 may be important for AEC migration, it has also been implicated in basement-membrane damage in IPF (45). Similarly, MMP-7 (matrilysin) appears to be important for reepithelialization (47) but is upregulated in the bleomycin model of lung injury, and mice deficient in MMP-7 are protected from fibrotic sequelae (48). Other classes of proteases also regulate cell migration, including plasmin, which is derived from plasminogen within the provisional fibrin clot itself and can be activated by either tissue-type plasminogen activator or urokinase-type plasminogen activator (uPA). These plasminogen activators and the receptor for uPA are all upregulated in migrating keratinocytes in cutaneous wound healing, suggesting an important role for this system in reepithelialization (19). Mice deficient in plasminogen activator inhibitor-1 are protected from bleomycin-induced pulmonary fibrosis (49); a similar protective effect is observed when uPA is transgenetically expressed in an inducible, lung-specific manner (50). Whether these protective effects relate to a more favorable AEC phenotype or to increased turnover of the provisional matrix or another yet unidentified effect, perhaps on fibroblast phenotype, is currently unclear. Evidence is now accumulating that reepithelialization may, in part, be mediated by adult stem cells. Tissue- and bone marrow–derived stem cells appear to have the potential to replace and regenerate adult tissues (51, 52). Type I AEC may

5 Jan 2004

12:34

AR

AR206-ME55-23.tex

AR206-ME55-23.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:395-417. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

PULMONARY FIBROSIS

P1: GCE

401

derive from bone marrow–derived cells, independent of type II differentiation in the bleomycin injury model (53). In a bone marrow transplant model, marrowderived cells gave rise to type II cells that subsequently differentiated into type I AECs (54). Twenty percent of AECs were derived from the marrow following bone marrow stem cell transplantation (55). However, bone marrow progenitor cells do not differentiate into the respiratory epithelium of the healthy upper airway (56). Delayed or ineffective reepithelialization, regardless of the mechanism(s) involved, is a key trigger for recruiting, activating, and sustaining mesenchymal cells. A perpetual “on-signal” results in continual bidirectional signaling between the epithelial and mesenchymal tissue compartments, leading to a dysregulated, persistent repair process that culminates in fibrosis. AECs in UIP/IPF have been shown to produce a number of soluble recruitment and trophic factors, including TGF-β1 (57–59), monocyte chemoattractant protein-1 (60), platelet-derived growth factor (61), tumor necrosis factor-α (58, 62), connective tissue growth factor (63), and endothelin-1, that are known to promote activation of fibroblasts and mesenchymal cells (64). Additionally, epithelium-derived factors that suppress mesenchymal cell functions may be reduced in UIP/IPF; PGE2 is decreased in the lungs of IPF patients (65, 66). Nitric oxide is another epithelial-derived factor that may suppress mesenchymal cell activation (67–70).

FIBROBLASTS AND MYOFIBROBLASTS: EFFECTOR CELLS IN FIBROSIS Fibroblastic foci (FF), often localized at the “leading edge” of normal and fibrotic lung, represent a hallmark of the histopathology of UIP (1, 71). The presence and extent of FF in patients with UIP/IPF is one of the more reliable markers of a poor prognosis and decreased survival (72, 73). The profusion of FF in idiopathic UIP (IPF) is greater than in CVD-associated UIP and is associated with increased mortality (7). Early studies demonstrated the presence of phenotypically distinct fibroblasts, termed myofibroblasts, in FF of UIP/IPF (71). The transient appearance and disappearance of myofibroblasts is well documented in the granulation tissue of healing cutaneous wounds (74). Myofibroblasts possess ultrastructural features intermediate between fibroblasts and smooth muscle cells; they are defined primarily by their expression of contractile proteins (75). Myofibroblasts represent a synthetically active and contractile phenotype that is responsible for the connective tissue synthesis and remodeling characteristic of fibrotic disorders (76–78).

Source(s) and Regulation of the Myofibroblast The recruitment and activation of fibroblasts/myofibroblasts probably represent a normal response to epithelial or endothelial injury, in general. Fibroblasts are phenotypically the most versatile of the connective-tissue cell family, with exceptional plasticity to adapt to local, sometimes harsh, environments (79). In the early stages

5 Jan 2004

12:34

Annu. Rev. Med. 2004.55:395-417. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

402

AR

AR206-ME55-23.tex

AR206-ME55-23.sgm

LaTeX2e(2002/01/18)

P1: GCE

THANNICKAL ET AL.

following injury, their primary function is to secrete extracellular matrix (ECM) proteins that provide a tissue scaffold for normal repair events such as epithelial cell migration. In later stages of tissue repair, myofibroblasts, by virtue of their contractile function, facilitate wound closure and reepithelialization, a process that is better defined in cutaneous wound healing (19, 80). Eventual dissolution of the granulation tissue scaffold and apoptosis of fibroblasts/myofibroblasts is critical for restoration of normal tissue architecture (81, 82). The physiological mechanisms for apoptosis of these cells are not well understood, but it has been postulated that loss of mechanical tension may play a role (83). Persistence of myofibroblasts in areas of active fibrosis is a consistent finding in the pathology of human fibrotic diseases involving diverse organ systems including the lung (78, 84). The source or sources of myofibroblasts in UIP/IPF lung are unclear, although most current evidence points to the differentiation of fibroblasts under the influence of local growth factors/cytokines (85). Relative contributions from circulating mesenchymal stem cells or from the local transdifferentiation of epithelial cells to fibroblasts, events that have been demonstrated in other organ systems (86, 87), have not been well defined in the lung. The regulation of myofibroblast differentiation is complex and involves the actions of soluble growth factors and ECM-derived signals emanating from alterations in the biophysical and biochemical properties of the ECM. TGF-β1 is a central regulator of this phenotype in vitro and in vivo (78); other soluble factors such as endothelin (88) and thrombin (89) may mediate similar effects, although their relative roles in fibrotic diseases of the lung are unclear. The synthesis of alternatively spliced forms of fibronectin involving the ED-A domain and focal adhesion remodeling by TGF-β1 is required for myofibroblast differentiation (90, 91). Biophysical factors such as tension and compliance of the cell’s substratum are also crucial in the regulation of this phenotype (92, 93). Myofibroblast differentiation of lung fibroblasts by TGF-β1 depends on activation of integrinfocal adhesion kinase signaling, suggesting that this pathway may serve to integrate diverse extracellular signals to regulate this phenotypic transition (94). The c-JunNH2-terminal kinase pathway has also been reported to mediate myofibroblast differentiation (95).

Phenotypic Heterogeneity of Fibroblasts in UIP Several studies have attempted to characterize the phenotype of fibroblasts/ myofibroblasts in UIP/IPF, sometimes with conflicting results. Such differences may relate to inherent tissue fibroblast heterogeneity and changes in cellular microenvironment, including in vitro culture conditions. Fibroblasts derived from fibrotic tissue have been reported to demonstrate both high and low proliferative capacities (96–98); lower rates of proliferation appear to be associated with more advanced fibrosis (96), perhaps reflecting a more differentiated state. Fibrotic-lung fibroblasts demonstrate anchorage-independent growth in soft agar, whereas normal fibroblasts do not (99). In vivo apoptotic rates of fibroblasts/myofibroblasts from UIP/IPF appear to be lower than the apoptotic rate in the fibromyxoid

5 Jan 2004

12:34

AR

AR206-ME55-23.tex

AR206-ME55-23.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:395-417. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

PULMONARY FIBROSIS

P1: GCE

403

connective tissue of bronchiolitis obliterans organizing pneumonia (100). The paradoxically higher rates of apoptosis observed in in vitro culture of UIP/IPF fibroblasts (98) may relate to differences in cellular microenvironment and, in particular, to the presence of activated TGF-β1 (85). UIP/IPF fibroblasts/myofibroblasts in vivo are highly synthetic and produce a number of ECM proteins and integrin molecules (71, 76, 101, 102). This is accompanied by reduced capacity for ECM degradation from imbalances in the production of MMPs and TIMPs (46, 98). TIMP-2 expression by fibroblasts/myofibroblasts correlates with the structural remodeling seen in UIP/IPF (45, 103, 104). Other phenotypic characteristics described in UIP/IPF fibroblasts include enhanced migratory capacity (105), increased fibroblast contractility (106), and diminished COX-2 expression and PGE2 synthesis (65). Cumulatively, these studies suggest that fibroblasts/myofibroblasts in UIP/IPF represent a population of cells with diverse phenotypes and functions; this heterogeneity is likely to be dictated by dynamic changes in regionally distinct tissue microenvironments within the injured lung.

Expanded Roles for Fibroblasts/Myofibroblasts Myofibroblasts, in addition to their well-recognized synthetic and contractile functions, represent an “activated” phenotype with the capacity to produce a variety of growth factors (107), cytokines (108), and chemokines (109). Their secretory repertoire includes the generation and release of both reactive oxygen and nitrogen species (39, 110, 111). Such properties have led some to refer to myofibroblasts as inflammatory cells (108). The hypothesis that fibroblasts, in general, contribute to immune responses has been further bolstered by the finding of surface receptors, such as CD40, typically associated with immune cells (112, 113). It has been proposed that, under certain pathological conditions, fibroblasts as sentinel cells of the immune system regulate the switch from acute, resolving inflammation to chronic, persistent inflammation (114, 115). There is growing recognition that fibroblasts/myofibroblasts sustain their growth and activity even in the absence of classic inflammatory cells and their secreted products (3, 85, 116). This suggests that autocrine mechanisms and epithelialderived factors may be sufficient to drive the fibrotic process, similar to the “epithelial-mesenchymal trophic unit” described in airway remodeling of asthmatic patients (117). Consistent with this concept, myofibroblasts express a number of growth factor receptors (118) and cell adhesion receptors (119) that allow them to respond to mitogenic growth factors and ECM signals in their cellular microenvironment.

INFLAMMATORY CELLS: MODULATORS OF FIBROSIS The precise roles of the various inflammatory cells in disease pathogenesis, particularly of UIP/IPF, remain poorly understood (3, 120). It is likely that neutrophils and macrophages play significant modulatory roles in the fibrogenic process at

5 Jan 2004

12:34

Annu. Rev. Med. 2004.55:395-417. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

404

AR

AR206-ME55-23.tex

AR206-ME55-23.sgm

LaTeX2e(2002/01/18)

P1: GCE

THANNICKAL ET AL.

various stages of the disease. Additionally, activated mesenchymal cells themselves are potent sources of monocyte/neutrophil chemotactic activity because they elaborate proinflammatory chemokines/cytokines (121, 122). Thus, these cells may induce the recruitment/activation of inflammatory cells to sites of injury. Interestingly, although the numbers of neutrophils in the lower respiratory tract of IPF patients may not be elevated, they appear to be of a more activated phenotype (123–125). The release of primary and secondary granule enzymes by neutrophils isolated from UIP/IPF is increased in comparison to fibrosing alveolitis associated with systemic sclerosis (125). In addition to these enzymes, neutrophils release higher concentrations of oxidants, a proposed mechanism for epithelial cell injury in UIP/IPF (123). The presence and activation of alveolar macrophages are also likely to influence tissue remodeling and fibrogenesis. Although macrophages have long been considered an initiator of the fibrotic process in UIP/IPF (126, 127), recent evidence suggests that under certain conditions they may also mediate antifibrotic effects (128, 129).

EXTRACELLULAR MATRIX REMODELING The accumulation of ECM proteins, including collagens, fibronectin, proteoglycans, and elastin, has long been considered the hallmark of fibrosis. The ECM plays a vital and critical role early in the repair process, orchestrating the modulation and transitions of cell phenotype and function. This conceptual paradigm is better appreciated in tissue morphogenesis in developmental biological processes (130), but it is likely to be as important in adult tissue remodeling responses to injury. The phenotype of fibroblasts and epithelial cells is critically dependent on both soluble factors and ECM-generated signals. Aberrant signaling by the ECM may perpetuate the cycle of ECM deposition and impaired degradation that typifies the progressive, seemingly irreversible remodeling seen in end-stage fibrosis. It has been suggested that a nondegradative environment exists in UIP/IPF, in part due to an imbalance of MMPs and TIMPs (46). However, although the production of TIMPs is consistently elevated in UIP/IPF, MMPs often expressed on AECs are also upregulated, suggesting additional functions for these multifunctional enzymes (103). MMPs are known to exert a myriad of regulatory actions critical in tissue repair and remodeling, including epithelial cell migration, proliferation, differentiation, and apoptosis, as well as release of latent or bound growth factors from the ECM (131). MMPs may degrade and damage the basement membrane (132, 133), a consistent finding in UIP/IPF (71, 76, 103). Injury to the basement membrane may also occur through oxidant-mediated mechanisms (134, 135). Oxidants can alter the structure and function of ECM proteins by inducing dityrosine-dependent crosslinking reactions (136); moreover, TGF-β1-differentiated myofibroblasts may serve as a source of oxidant production and thereby alter ECM remodeling (94, 110). The role of TGF-β1 in inhibiting the expression of TIMPs, contributing to a nondegradative environment, is well recognized (137).

5 Jan 2004

12:34

AR

AR206-ME55-23.tex

AR206-ME55-23.sgm

LaTeX2e(2002/01/18)

PULMONARY FIBROSIS

P1: GCE

405

Annu. Rev. Med. 2004.55:395-417. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

TYPE I AND TYPE II CYTOKINE PROFILES A shift from a type I (IFN-γ , IL-2, IL-12, IL-18) to a type II (IL-4, IL-5, IL-10, IL13) cytokine profile is likely to be a key event in the progression of inflammation to fibrosis (138, 139). Type II cytokines are generally activators of fibroblasts and may promote ECM deposition and remodeling (140, 141). Type I cytokines, particularly IFN-γ , appear to exert antifibrotic effects by inhibiting the activation of myofibroblasts (142), an effect that is probably related to blockade of TGF-β1 signaling (143). A relative shift to a type II–predominant cytokine profile has been demonstrated in patients with IPF (144–146), suggesting that this imbalance may contribute to fibrogenesis. A recent phase III trial examined the efficacy of IFN-γ therapy for patients with IPF; although results have yet to be published, early indications are that it may not be as promising as initially hoped (146, 147).

ANGIOGENESIS The formation of new blood vessels (neovascularization) is an early response to tissue injury and is triggered by production of soluble angiogenic growth factors or chemokines, alteration in ECM or integrin molecules, and localized tissue hypoxia (148). Neovascularization is necessary to sustain the formation of granulation tissue at the sites of tissue injury, a process that is well characterized in cutaneous wound healing (80). Increased angiogenic activity has been demonstrated in lung homogenates of patients with IPF. This increased angiogenic activity was attributed to elevated levels of proangiogenic factors, namely IL-8 and epithelial-derived neutrophil-activating peptide 78 (ENA-78). A concomitant reduction in an antiangiogenic chemokine, IFN-induced protein of 10 kDa (IP-10) (149, 150), was noted. In contrast, other proangiogenic factors such as vascular endothelial growth factor (VEGF) appear to be decreased in BAL fluid of patients with IPF versus normal nonsmoking volunteers (151). Moreover, there appears to be greater vascularization of the fibromyxoid lesions of bronchiolitis obliterans organizing pneumonia than in the FF of UIP/IPF (103, 152), an effect that correlates with increased expression of VEGF and fibroblast growth factor–2 (153). Blood vessel density is reduced in the FF of UIP/IPF patients with net vascular ablation and redistribution of blood vessels in areas of interstitial thickening (154). Such differences in the role of angiogenesis in UIP/IPF relate, in part, to the specific methods utilized for analysis. Angiogenesis is probably critical for the early formation of granulation tissue and provisional matrix; it may regress in the later, more mature FF of UIP/IPF.

EICOSANOID IMBALANCE Eicosanoids, lipid metabolites of the 20-carbon arachadonic acid, are biologically active during many inflammatory and repair processes. Their roles in inflammation and remodeling of the asthmatic airway are well appreciated (155). Leukotrienes,

5 Jan 2004

12:34

Annu. Rev. Med. 2004.55:395-417. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

406

AR

AR206-ME55-23.tex

AR206-ME55-23.sgm

LaTeX2e(2002/01/18)

P1: GCE

THANNICKAL ET AL.

derived from the 5-lipoxygenase (5-LO) pathway, have been shown to be elevated in lung homogenates of IPF patients; alveolar macrophages from these patients demonstrate constitutive activation of 5-LO (156). Recent work in animal models supports a profibrotic role for leukotrienes in vivo; mice deficient in 5-LO are protected from bleomycin-induced fibrosis (157). In contrast, certain prostaglandins (in particular, PGE2) derived from the cyclooxygenase (COX) pathway may be decreased in IPF lung. Fibroblasts isolated from patients with UIP/IPF had diminished capacity for PGE2 synthesis due to a defect in the regulation of COX-2 (65). Moreover, PGE2 concentrations are decreased in the BAL of patients with IPF (66). Potential antifibrotic effects of PGE2 include inhibition of fibroblast proliferation (158), migration (159), contractility (160), and myofibroblast differentiation (161). In animal models of fibrosis, monocyte chemoattractant protein-1 signaling via the cysteine-cysteine chemokine receptor 2 (CCR2) decreases PGE2 production by AECs (162), a potential mechanism for the observed protection in CCR2-deficient mice from experimental pulmonary fibrosis (163). Moreover, diminished synthesis of PGE2 in GM-CSF–deficient mice correlates with increased fibrotic responses to bleomycin in these animals (26). Cumulatively, these observations support an altered eicosanoid imbalance in the alveolar microenvironment characterized by relative increases in profibrotic leukotrienes and a decrease in PGE2.

OXIDATIVE STRESS It has been proposed that IPF is a disorder of elevated oxidative stress, with the existence of an oxidant-antioxidant imbalance in distal alveolar air spaces (164, 165). Neutrophils from patients with IPF generate higher levels of oxidants, and the epithelial lining fluid in these patients demonstrates high levels of myeloperoxidase (123). In addition, the epithelial lining fluid of patients with IPF contains lower concentrations of the sulfhydryl-containing tripeptide GSH, a critical antioxidant that protects AECs from oxidant injury (37). Both oral and aerosolized administration of N-acetyl cysteine, a GSH precursor, safely and effectively augment lung GSH levels in association with a reduction in the spontaneous generation of oxidants by alveolar macrophages (166, 167). Nonphagocytic cells, including fibroblasts, can also generate oxidants in the setting of lung injury and repair (39). Oxidants may alter the nature of surrounding ECM (136). The potential for fibroblast/myofibroblast-derived oxidants to induce or perpetuate AEC injury and/or apoptosis in chronic fibrotic disease deserves further investigation.

GENETIC AND ENVIRONMENTAL FACTORS The complexity of the inflammatory and repair processes predicts that polygenic factors may influence biological outcomes of the host response to lung injury. Functional gene polymorphisms of a number of cytokines have been associated

5 Jan 2004

12:34

AR

AR206-ME55-23.tex

AR206-ME55-23.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:395-417. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

PULMONARY FIBROSIS

P1: GCE

407

with either the incidence or disease progression of IPF (168, 169). A mutation in the prosurfactant protein-C gene encoding a leucine-to-glutamine substitution in the C terminus was noted in a kindred of patients with familial IPF (8). This results in improper folding and processing of prosurfactant protein-C in type II AECs, as evidenced by electron microscopy showing aberrant subcellular localization of the protein and abnormal lamellar bodies (8). Up to 3% of IPF cases cluster in families, suggesting that in such cases genetic susceptibility may play a more dominant role in disease pathogenesis. Environmental factors have also been thought to contribute to the development of pulmonary fibrosis. Associations between latent viral infections and IPF have been demonstrated (170–172), but cause–effect relationships have not been established. Occupational and dust exposures have shown some loose associations (173, 174), but links to specific agents or exposures have not been identified. The most intriguing environmental risk factor studied is cigarette smoking. Although studies have demonstrated that a history of smoking may be an independent risk factor for developing IPF (175), recent studies strongly indicate that current smokers have improved mortality in established cases of IPF (72, 176). Current smokers also had lower granulation/connective tissue (FF) scores, another independent predictor of disease survival (7, 72). These observations raise the possibility that a factor or factors in cigarette smoke may negatively regulate fibroblast/myofibroblast activation (177). Thus, it is likely that the interplay between environmental factors and genetic susceptibility strongly influences the tissue injury and repair processes that culminate in fibrosis.

AGE-RELATED INFLUENCES IPF is a disease of the elderly, with a median age of onset of 50–60 years (178). The diagnosis is rarely made in a patient under 40 years old, and the incidence of IPF markedly increases with advancing age. Patients with idiopathic UIP (IPF) present at older ages than patients with CVD-associated UIP (7). The aging process and the biology of cellular senescence may influence tissue repair and remodeling responses. Genomic analyses of senescent fibroblasts demonstrate a pattern of gene expression that mimics inflammatory wound repair, suggesting that such cells may contribute to chronic wound pathologies (179). Aging is associated with oxidative stress (180) and impairment in the capacity to repair damaged cells (181); the latter may be further compounded by age-related limitations in adult stem cell responses.

CONCLUSION Accumulating evidence indicates that the IIPs represent a spectrum of tissue responses to an unidentified injurious agent that results in varying degrees of inflammation and fibrosis. The degree and types of inflammation and fibrosis

5 Jan 2004

12:34

Annu. Rev. Med. 2004.55:395-417. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

408

AR

AR206-ME55-23.tex

AR206-ME55-23.sgm

LaTeX2e(2002/01/18)

P1: GCE

THANNICKAL ET AL.

(histopathologic patterns) that result are determined by multiple factors, including, age, genetic susceptibility, environmental factors, and perhaps the nature of the etiological agent itself. The interplay between these multiple host and environmental factors influences the activity of cells and molecules within alveolar microenvironments of the lung; interactions between cells and molecules within these complex tissue microenvironments are likely to be dynamic and highly heterogeneous, ultimately determining pathological outcomes manifested by varied histopathological patterns. Stereotypical and highly conserved inflammatory and repair responses designed to protect the host from danger are activated in response to tissue injury. These cellular and molecular responses are regulated in a time- and context-dependent manner; that is, molecules and cells may switch their function and phenotype based on cues received from their microenvironment. The regulation of inflammation and repair, and particularly the mechanisms involved in the termination of repair responses, require further study. Dysregulation of these events leads to pathological outcomes. The challenge for clinicians is to accurately “phenotype” the disease process (i.e., define relative degrees of inflammation and fibrosis), a task that is complicated by the heterogeneity of the disease process. Future treatment approaches must take into account the disease heterogeneity observed in these patients. ACKNOWLEDGMENTS This work was supported by NIH grant HL-67967 (V.J.T.) and a Specialized Center of Research (SCOR) grant in the Pathobiology of Fibrotic Lung Disease, P50 HL056402 (G.B.T. and F.J.M.). The Annual Review of Medicine is online at http://med.annualreviews.org

LITERATURE CITED 1. Katzenstein AL, Myers JL. 1998. Idiopathic pulmonary fibrosis: clinical relevance of pathologic classification. Am. J. Respir. Crit. Care Med. 157:1301–15 2. American Thoracic Society/European Respiratory Society International Multidisciplinary Consensus Classification of the Idiopathic Interstitial Pneumonias. 2002. Am. J. Respir. Crit. Care Med. 165:277–304 3. Selman M, King TE, Pardo A. 2001. Idiopathic pulmonary fibrosis: prevailing and evolving hypotheses about its pathogenesis and implications for therapy. Ann. Intern. Med. 134:136–51

4. Collard HR, King TE Jr. 2003. Demystifying idiopathic interstitial pneumonia. Arch. Intern. Med. 163:17–29 5. Flaherty KR, Travis WD, Colby TV, et al. 2001. Histopathologic variability in usual and nonspecific interstitial pneumonias. Am. J. Respir. Crit. Care Med. 164:1722– 27 6. Katzenstein AL, Zisman DA, Litzky LA, et al. 2002. Usual interstitial pneumonia: histologic study of biopsy and explant specimens. Am. J. Surg. Pathol. 26:1567– 77 7. Flaherty KR, Colby TV, Travis WD, et al. 2003. Fibroblastic foci in usual

5 Jan 2004

12:34

AR

AR206-ME55-23.tex

AR206-ME55-23.sgm

LaTeX2e(2002/01/18)

PULMONARY FIBROSIS

8.

Annu. Rev. Med. 2004.55:395-417. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

9. 10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

interstitial pneumonia: idiopathic versus collagen vascular disease. Am. J. Respir. Crit. Care Med. 167:1410–15 Thomas AQ, Lane K, Phillips J 3rd, et al. 2002. Heterozygosity for a surfactant protein C gene mutation associated with usual interstitial pneumonitis and cellular nonspecific interstitial pneumonitis in one kindred. Am. J. Respir. Crit. Care Med. 165:1322–28 Nathan C. 2002. Points of control in inflammation. Nature 420:846–52 Matzinger P. 2002. The danger model: a renewed sense of self. Science 296:301– 5 Lawrence T, Gilroy DW, Colville-Nash PR, et al. 2001. Possible new role for NFkappaB in the resolution of inflammation. Nat. Med. 7:1291–97 Wahl SM, Hunt DA, Wakefield LM, et al. 1987. Transforming growth factor type beta induces monocyte chemotaxis and growth factor production. Proc. Natl. Acad. Sci. USA 84:5788–92 Grande JP. 1997. Role of transforming growth factor-β in tissue injury and repair. Proc. Soc. Exp. Biol. Med. 214:27–40 Coalson JJ. 1982. The ultrastructure of human fibrosing alveolitis. Virchows Arch. A Pathol. Anat. Histol. 395:181–99 Corrin B, Jagusch M, Dewar A, et al. 1987. Fine structural changes in idiopathic pulmonary haemosiderosis. J. Pathol. 153:249–56 Kasper M, Haroske G. 1996. Alterations in the alveolar epithelium after injury leading to pulmonary fibrosis. Histol. Histopathol. 11:463–83 Rennard SI, Bitterman PB, Crystal RG. 1983. Response of the lower respiratory tract to injury. Mechanisms of repair of the parenchymal cells of the alveolar wall. Chest 84:735–39 Katzenstein AL. 1985. Pathogenesis of “fibrosis” in interstitial pneumonia: an electron microscopic study. Hum. Pathol. 16:1015–24 Martin P. 1997. Wound healing—aiming

20.

21.

22.

23.

24.

25.

26.

27.

P1: GCE

409

for perfect skin regeneration. Science 276:75–81 Chilosi M, Poletti V, Murer B, et al. 2002. Abnormal re-epithelialization and lung remodeling in idiopathic pulmonary fibrosis: the role of deltaN-p63. Lab. Invest. 82:1335–45 Adamson IY, Young L, Bowden DH. 1988. Relationship of alveolar epithelial injury and repair to the induction of pulmonary fibrosis. Am. J. Pathol. 130:377– 83 Sakai T, Satoh K, Matsushima K, et al. 1997. Hepatocyte growth factor in bronchoalveolar lavage fluids and cells in patients with inflammatory chest diseases of the lower respiratory tract: detection by RIA and in situ hybridization. Am. J. Respir. Cell Mol. Biol. 16:388– 97 Hojo S, Fujita J, Yoshinouchi T, et al. 1997. Hepatocyte growth factor and neutrophil elastase in idiopathic pulmonary fibrosis. Respir. Med. 91:511–16 Christensen PJ, Bailie MB, Goodman RE, et al. 2000. Role of diminished epithelial GM-CSF in the pathogenesis of bleomycin-induced pulmonary fibrosis. Am. J. Physiol. Lung Cell Mol. Physiol. 279:L487–95 Moore BB, Coffey MJ, Christensen P, et al. 2000. GM-CSF regulates bleomycin-induced pulmonary fibrosis via a prostaglandin-dependent mechanism. J. Immunol. 165:4032–39 Charbeneau RP, Christensen PJ, Chrisman CJ, et al. 2003. Impaired synthesis of prostaglandin E2 by lung fibroblasts and alveolar epithelial cells from GM-CSF–/– mice: implications for fibroproliferation. Am. J. Physiol. Lung Cell Mol. Physiol. 284:L1103–11 Xing Z, Tremblay GM, Sime PJ, et al. 1997. Overexpression of granulocytemacrophage colony-stimulating factor induces pulmonary granulation tissue formation and fibrosis by induction of transforming growth factor-β 1 and

5 Jan 2004

12:34

410

28.

Annu. Rev. Med. 2004.55:395-417. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

29.

30.

31.

32.

33.

34.

35.

36.

AR

AR206-ME55-23.tex

AR206-ME55-23.sgm

LaTeX2e(2002/01/18)

P1: GCE

THANNICKAL ET AL. myofibroblast accumulation. Am. J. Pathol. 150:59–66 Chen G, Grotendorst G, Eichholtz T, et al. 2003. GM-CSF increases airway smooth muscle cell connective tissue expression by inducing TGF-β receptors. Am. J. Physiol. Lung Cell Mol. Physiol. 284:L548–56 Maeyama T, Kuwano K, Kawasaki M, et al. 2001. Upregulation of Fas-signalling molecules in lung epithelial cells from patients with idiopathic pulmonary fibrosis. Eur. Respir. J. 17:180–89 Wang R, Zagariya A, Ang E, et al. 1999. Fas-induced apoptosis of alveolar epithelial cells requires ANG II generation and receptor interaction. Am. J. Physiol. Lung Cell Mol. Physiol. 277:L1245–50 Wang R, Ramos C, Joshi I, et al. 1999. Human lung myofibroblast-derived inducers of alveolar epithelial apoptosis identified as angiotensin peptides. Am. J. Physiol. Lung Cell Mol. Physiol. 277:L1158–64 Hagimoto N, Kuwano K, Inoshima I, et al. 2002. TGF-β 1 as an enhancer of Fas-mediated apoptosis of lung epithelial cells. J. Immunol. 168:6470–78 Kuwano K, Hagimoto N, Maeyama T, et al. 2002. Mitochondria-mediated apoptosis of lung epithelial cells in idiopathic interstitial pneumonias. Lab. Invest. 82:1695–706 Lavrentiadou SN, Chan C, Kawcak T, et al. 2001. Ceramide-mediated apoptosis in lung epithelial cells is regulated by glutathione. Am. J. Respir. Cell Mol. Biol. 25:676–84 Goldkorn T, Balaban N, Shannon M, et al. 1998. H2O2 acts on cellular membranes to generate ceramide signaling and initiate apoptosis in tracheobronchial epithelial cells. J. Cell Sci. 111(Pt. 21):3209–20 Beeh KM, Beier J, Haas IC, et al. 2002. Glutathione deficiency of the lower respiratory tract in patients with idiopathic pulmonary fibrosis. Eur. Respir. J. 19:1119– 23

37. Cantin AM, Hubbard RC, Crystal RG. 1989. Glutathione deficiency in the epithelial lining fluid of the lower respiratory tract in idiopathic pulmonary fibrosis. Am. Rev. Respir. Dis. 139:370–72 38. Arsalane K, Dubois CM, Muanza T, et al. 1997. Transforming growth factor-β1 is a potent inhibitor of glutathione synthesis in the lung epithelial cell line A549: transcriptional effect on the GSH ratelimiting enzyme γ -glutamylcysteine synthetase. Am. J. Respir. Cell Mol. Biol. 17:599–607 39. Thannickal VJ, Fanburg BL. 1995. Activation of an H2O2-generating NADH oxidase in human lung fibroblasts by transforming growth factor β1. J. Biol. Chem. 270:30334–38 40. Crouch E. 1990. Pathobiology of pulmonary fibrosis. Am. J. Physiol. Lung Cell Mol. Physiol. 259:L159–84 41. Kuwano K, Hagimoto N, Kawasaki M, et al. 1999. Essential roles of the Fas-Fas ligand pathway in the development of pulmonary fibrosis. J. Clin. Invest. 104:13–19 42. Hagimoto N, Kuwano K, Miyazaki H, et al. 1997. Induction of apoptosis and pulmonary fibrosis in mice in response to ligation of Fas antigen. Am. J. Respir. Cell Mol. Biol. 17:272–78 43. Kunugi S, Fukuda Y, Ishizaki M, et al. 2001. Role of MMP-2 in alveolar epithelial cell repair after bleomycin administration in rabbits. Lab. Invest. 81:1309–18 44. Parks WC, Shapiro SD. 2001. Matrix metalloproteinases in lung biology. Respir. Res. 2:10–19 45. Hayashi T, Stetler-Stevenson WG, Fleming MV, et al. 1996. Immunohistochemical study of metalloproteinases and their tissue inhibitors in the lungs of patients with diffuse alveolar damage and idiopathic pulmonary fibrosis. Am. J. Pathol. 149:1241–56 46. Selman M, Ruiz V, Cabrera S, et al. 2000. TIMP-1, -2, -3, and -4 in idiopathic pulmonary fibrosis. A prevailing nondegradative lung microenvironment? Am.

5 Jan 2004

12:34

AR

AR206-ME55-23.tex

AR206-ME55-23.sgm

LaTeX2e(2002/01/18)

PULMONARY FIBROSIS

47.

Annu. Rev. Med. 2004.55:395-417. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

48.

49.

50.

51. 52.

53.

54.

55.

56.

57.

J. Physiol. Lung Cell Mol. Physiol. 279:L562–74 Dunsmore SE, Saarialho-Kere UK, Roby JD, et al. 1998. Matrilysin expression and function in airway epithelium. J. Clin. Invest. 102:1321–31 Zuo F, Kaminski N, Eugui E, et al. 2002. Gene expression analysis reveals matrilysin as a key regulator of pulmonary fibrosis in mice and humans. Proc. Natl. Acad. Sci. USA 99:6292–97 Eitzman DT, McCoy RD, Zheng X, et al. 1996. Bleomycin-induced pulmonary fibrosis in transgenic mice that either lack or overexpress the murine plasminogen activator inhibitor–1 gene. J. Clin. Invest. 97:232–37 Sisson TH, Hanson KE, Subbotina N, et al. 2002. Inducible lung-specific urokinase expression reduces fibrosis and mortality after lung injury in mice. Am. J. Physiol. Lung Cell Mol. Physiol. 283: L1023–32 Otto WR. 2002. Lung epithelial stem cells. J. Pathol. 197:527–35 Poulsom R, Alison MR, Forbes SJ, et al. 2002. Adult stem cell plasticity. J. Pathol. 197:441–56 Kotton DN, Ma BY, Cardoso WV, et al. 2001. Bone marrow-derived cells as progenitors of lung alveolar epithelium. Development 128:5181–88 Theise ND, Henegariu O, Grove J, et al. 2002. Radiation pneumonitis in mice: a severe injury model for pneumocyte engraftment from bone marrow. Exp. Hematol. 30:1333–38 Krause DS, Theise ND, Collector MI, et al. 2001. Multi-organ, multi-lineage engraftment by a single bone marrowderived stem cell. Cell 105:369–77 Davies JC, Potter M, Bush A, et al. 2002. Bone marrow stem cells do not repopulate the healthy upper respiratory tract. Pediatr. Pulmonol. 34:251–56 Khalil N, O’Connor RN, Unruh HW, et al. 1991. Increased production and immunohistochemical localization of transform-

58.

59.

60.

61.

62.

63.

64.

65.

66.

P1: GCE

411

ing growth factor-β in idiopathic pulmonary fibrosis. Am. J. Respir. Cell Mol. Biol. 5:155–62 Kapanci Y, Desmouliere A, Pache JC, et al. 1995. Cytoskeletal protein modulation in pulmonary alveolar myofibroblasts during idiopathic pulmonary fibrosis. Possible role of transforming growth factor β and tumor necrosis factor α. Am. J. Respir. Crit. Care Med. 152:2163–69 Khalil N, O’Connor RN, Flanders KC, et al. 1996. TGF-β 1, but not TGF-β 2 or TGF-β 3, is differentially present in epithelial cells of advanced pulmonary fibrosis: an immunohistochemical study. Am. J. Respir. Cell Mol. Biol. 14:131–38 Antoniades HN, Neville-Golden J, Galanopoulos T, et al. 1992. Expression of monocyte chemoattractant protein 1 mRNA in human idiopathic pulmonary fibrosis. Proc. Natl. Acad. Sci. USA 89: 5371–75 Antoniades HN, Bravo MA, Avila RE, et al. 1990. Platelet-derived growth factor in idiopathic pulmonary fibrosis. J. Clin. Invest. 86:1055–64 Nash JR, McLaughlin PJ, Butcher D, et al. 1993. Expression of tumour necrosis factor-alpha in cryptogenic fibrosing alveolitis. Histopathology 22:343–47 Pan LH, Yamauchi K, Uzuki M, et al. 2001. Type II alveolar epithelial cells and interstitial fibroblasts express connective tissue growth factor in IPF. Eur. Respir. J. 17:1220–27 Giaid A, Michel RP, Stewart DJ, et al. 1993. Expression of endothelin-1 in lungs of patients with cryptogenic fibrosing alveolitis. Lancet 341:1550–54 Wilborn J, Crofford LJ, Burdick MD, et al. 1995. Cultured lung fibroblasts isolated from patients with idiopathic pulmonary fibrosis have a diminished capacity to synthesize prostaglandin E2 and to express cyclooxygenase-2. J. Clin. Invest 95:1861–68 Borok Z, Gillissen A, Buhl R, et al. 1991. Augmentation of functional prostaglandin

5 Jan 2004

12:34

412

67.

Annu. Rev. Med. 2004.55:395-417. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

68.

69.

70.

71.

72.

73.

74.

75.

76.

AR

AR206-ME55-23.tex

AR206-ME55-23.sgm

LaTeX2e(2002/01/18)

P1: GCE

THANNICKAL ET AL. E levels on the respiratory epithelial surface by aerosol administration of prostaglandin E. Am. Rev. Respir. Dis. 144:1080–84 Folkerts G, Nijkamp FP. 1998. Airway epithelium: more than just a barrier! Trends Pharmacol. Sci. 19:334–41 Wang R, Ghahary A, Shen YJ, et al. 1997. Nitric oxide synthase expression and nitric oxide production are reduced in hypertrophic scar tissue and fibroblasts. J. Invest. Dermatol. 108:438–44 Vernet D, Ferrini MG, Valente EG, et al. 2002. Effect of nitric oxide on the differentiation of fibroblasts into myofibroblasts in the Peyronie’s fibrotic plaque and in its rat model. Nitric Oxide 7:262–76 Zhang HY, Phan SH. 1999. Inhibition of myofibroblast apoptosis by transforming growth factor β 1. Am. J. Respir. Cell Mol. Biol. 21:658–65 Kuhn C, McDonald JA. 1991. The roles of the myofibroblast in idiopathic pulmonary fibrosis. Ultrastructural and immunohistochemical features of sites of active extracellular matrix synthesis. Am. J. Pathol. 138:1257–65 King TE Jr., Schwarz MI, Brown K, et al. 2001. Idiopathic pulmonary fibrosis: relationship between histopathologic features and mortality. Am. J. Respir. Crit. Care Med. 164:1025–32 Nicholson AG, Fulford LG, Colby TV, et al. 2002. The relationship between individual histologic features and disease progression in idiopathic pulmonary fibrosis. Am. J. Respir. Crit. Care Med. 166:173–77 Gabbiani G, Ryan GB, Majne G. 1971. Presence of modified fibroblasts in granulation tissue and their possible role in wound contraction. Experientia 27:549– 50 Walker GA, Guerrero IA, Leinwand LA. 2001. Myofibroblasts: molecular crossdressers. Curr. Top. Dev. Biol. 51:91–107 Kuhn C 3rd, Boldt J, King TE Jr., et al. 1989. An immunohistochemical study of

77.

78.

79.

80.

81.

82.

83.

84.

85. 86.

87.

88.

architectural remodeling and connective tissue synthesis in pulmonary fibrosis. Am. Rev. Respir. Dis. 140:1693–703 Zhang K, Rekhter MD, Gordon D, et al. 1994. Myofibroblasts and their role in lung collagen gene expression during pulmonary fibrosis. A combined immunohistochemical and in situ hybridization study. Am. J. Pathol. 145:114–25 Tomasek JJ, Gabbiani G, Hinz B, et al. 2002. Myofibroblasts and mechanoregulation of connective tissue remodelling. Nat. Rev. Mol. Cell Biol. 3:349–63 Harris J. 1994. Differentiated cells and the maintenance of tissues. In Molecular Biology of the Cell, ed. B Alberts, D Bray, J Lewis, et al., pp. 1139–93. New York: Garland Singer AJ, Clark RA. 1999. Cutaneous wound healing. N. Engl. J. Med. 341:738– 46 Desmouliere A, Redard M, Darby I, et al. 1995. Apoptosis mediates the decrease in cellularity during the transition between granulation tissue and scar. Am. J. Pathol. 146:56–66 Lorena D, Uchio K, Costa AM, et al. 2002. Normal scarring: importance of myofibroblasts. Wound Repair Regen. 10:86–92 Grinnell F, Zhu M, Carlson MA, et al. 1999. Release of mechanical tension triggers apoptosis of human fibroblasts in a model of regressing granulation tissue. Exp. Cell Res. 248:608–19 Border WA, Noble NA. 1994. Transforming growth factor β in tissue fibrosis. N. Engl. J. Med. 331:1286–92 Phan SH. 2002. The myofibroblast in pulmonary fibrosis. Chest 122:286S–289S Pittenger MF, Mackay AM, Beck SC, et al. 1999. Multilineage potential of adult human mesenchymal stem cells. Science 284:143–47 Iwano M, Plieth D, Danoff TM, et al. 2002. Evidence that fibroblasts derive from epithelium during tissue fibrosis. J. Clin. Invest. 110:341–50 Shahar I, Fireman E, Topilsky M, et al.

5 Jan 2004

12:34

AR

AR206-ME55-23.tex

AR206-ME55-23.sgm

LaTeX2e(2002/01/18)

PULMONARY FIBROSIS

Annu. Rev. Med. 2004.55:395-417. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

89.

90.

91.

92.

93.

94.

95.

96.

1999. Effect of endothelin-1 on alphasmooth muscle actin expression and on alveolar fibroblasts proliferation in interstitial lung diseases. Int. J. Immunopharmacol. 21:759–75 Bogatkevich GS, Tourkina E, Silver RM, et al. 2001. Thrombin differentiates normal lung fibroblasts to a myofibroblast phenotype via the proteolytically activated receptor-1 and a protein kinase C-dependent pathway. J. Biol. Chem. 276:45184–92 Serini G, Bochaton-Piallat ML, Ropraz P, et al. 1998. The fibronectin domain ED-A is crucial for myofibroblastic phenotype induction by transforming growth factorβ1. J. Cell Biol. 142:873–81 Dugina V, Fontao L, Chaponnier C, et al. 2001. Focal adhesion features during myofibroblastic differentiation are controlled by intracellular and extracellular factors. J. Cell Sci. 114:3285–96 Arora PD, Narani N, McCulloch CA. 1999. The compliance of collagen gels regulates transforming growth factor-β induction of α-smooth muscle actin in fibroblasts. Am. J. Pathol. 154:871–82 Hinz B, Mastrangelo D, Iselin CE, et al. 2001. Mechanical tension controls granulation tissue contractile activity and myofibroblast differentiation. Am. J. Pathol. 159:1009–20 Thannickal VJ, Lee DY, White ES, et al. 2003. Myofibroblast differentiation by transforming growth factor-β1 is dependent on cell adhesion and integrin signaling via focal adhesion kinase. J. Biol. Chem. 278:12384–89 Hashimoto S, Gon Y, Takeshita I, et al. 2001. Transforming growth factor-β1 induces phenotypic modulation of human lung fibroblasts to myofibroblast through a c-Jun-NH2-terminal kinase-dependent pathway. Am. J. Respir. Crit. Care Med. 163:152–57 Raghu G, Chen YY, Rusch V, et al. 1988. Differential proliferation of fibroblasts cultured from normal and fibrotic hu-

97.

98.

99.

100.

101.

102.

103.

104.

105.

P1: GCE

413

man lungs. Am. Rev. Respir. Dis. 138:703– 8 Jordana M, Schulman J, McSharry C, et al. 1988. Heterogeneous proliferative characteristics of human adult lung fibroblast lines and clonally derived fibroblasts from control and fibrotic tissue. Am. Rev. Respir. Dis. 137:579–84 Ramos C, Montano M, Garcia-Alvarez J, et al. 2001. Fibroblasts from idiopathic pulmonary fibrosis and normal lungs differ in growth rate, apoptosis, and tissue inhibitor of metalloproteinases expression. Am. J. Respir. Cell Mol. Biol. 24:591– 98 Torry DJ, Richards CD, Podor TJ, et al. 1994. Anchorage-independent colony growth of pulmonary fibroblasts derived from fibrotic human lung tissue. J. Clin. Invest. 93:1525–32 Lappi-Blanco E, Soini Y, Paakko P. 1999. Apoptotic activity is increased in the newly formed fibromyxoid connective tissue in bronchiolitis obliterans organizing pneumonia. Lung 177:367–76 Kuhn C, Mason RJ. 1995. Immunolocalization of SPARC, tenascin, and thrombospondin in pulmonary fibrosis. Am. J. Pathol. 147:1759–69 Fukuda Y, Basset F, Ferrans VJ, et al. 1995. Significance of early intra-alveolar fibrotic lesions and integrin expression in lung biopsy specimens from patients with idiopathic pulmonary fibrosis. Hum. Pathol. 26:53–61 Fukuda Y, Ishizaki M, Kudoh S, et al. 1998. Localization of matrix metalloproteinases-1, -2, and -9 and tissue inhibitor of metalloproteinase-2 in interstitial lung diseases. Lab. Invest. 78:687–98 Fukuda Y, Mochimaru H, Terasaki Y, et al. 2001. Mechanism of structural remodeling in pulmonary fibrosis. Chest 120:41S– 43S Suganuma H, Sato A, Tamura R, et al. 1995. Enhanced migration of fibroblasts derived from lungs with fibrotic lesions. Thorax 50:984–89

5 Jan 2004

12:34

Annu. Rev. Med. 2004.55:395-417. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

414

AR

AR206-ME55-23.tex

AR206-ME55-23.sgm

LaTeX2e(2002/01/18)

P1: GCE

THANNICKAL ET AL.

106. Miki H, Mio T, Nagai S, et al. 2000. Fibroblast contractility: usual interstitial pneumonia and nonspecific interstitial pneumonia. Am. J. Respir. Crit. Care Med. 162:2259–64 107. Finlay GA, Thannickal VJ, Fanburg BL, et al. 2000. Transforming growth factor-β 1-induced activation of the ERK pathway/ activator protein-1 in human lung fibroblasts requires the autocrine induction of basic fibroblast growth factor. J. Biol. Chem. 275:27650–56 108. Phan SH, Zhang K, Zhang HY, et al. 1999. The myofibroblast as an inflammatory cell in pulmonary fibrosis. Curr. Top. Pathol. 93:173–82 109. Mezzano SA, Droguett MA, Burgos ME, et al. 2000. Overexpression of chemokines, fibrogenic cytokines, and myofibroblasts in human membranous nephropathy. Kidney Int. 57:147–58 110. Thannickal VJ, Aldweib KD, Fanburg BL. 1998. Tyrosine phosphorylation regulates H2O2 production in lung fibroblasts stimulated by transforming growth factor β1. J. Biol. Chem. 273:23611–15 111. Holstein AF, Maekawa M, Nagano T, et al. 1996. Myofibroblasts in the lamina propria of human seminiferous tubules are dynamic structures of heterogeneous phenotype. Arch. Histol. Cytol. 59:109–25 112. Brouty-Boye D, Pottin-Clemenceau C, Doucet C, et al. 2000. Chemokines and CD40 expression in human fibroblasts. Eur. J. Immunol. 30:914–19 113. Kaufman J, Graf BA, Leung EC, et al. 2001. Fibroblasts as sentinel cells: role of the CDcd40-CDcd40 ligand system in fibroblast activation and lung inflammation and fibrosis. Chest 120:53S–55S 114. Buckley CD, Pilling D, Lord JM, et al. 2001. Fibroblasts regulate the switch from acute resolving to chronic persistent inflammation. Trends Immunol. 22:199– 204 115. Lo D, Feng L, Li L, et al. 1999. Integrating innate and adaptive immunity in the whole animal. Immunol. Rev. 169:225–39

116. Selman M, Pardo A. 2002. Idiopathic pulmonary fibrosis: an epithelial/fibroblastic cross-talk disorder. Respir. Res. 3:3 117. Davies DE, Holgate ST. 2002. Asthma: the importance of epithelial mesenchymal communication in pathogenesis. Inflammation and the airway epithelium in asthma. Int. J. Biochem. Cell Biol. 34:1520–26 118. Thannickal VJ, Aldweib KD, Rajan T, et al. 1998. Upregulated expression of fibroblast growth factor (FGF) receptors by transforming growth factor-β1 (TGF-β1) mediates enhanced mitogenic responses to FGFs in cultured human lung fibroblasts. Biochem. Biophys. Res. Commun. 251:437–41 119. Heino J, Ignotz RA, Hemler ME, et al. 1989. Regulation of cell adhesion receptors by transforming growth factor-β. Concomitant regulation of integrins that share a common β 1 subunit. J. Biol. Chem. 264:380–88 120. Gross TJ, Hunninghake GW. 2001. Idiopathic pulmonary fibrosis. N. Engl. J. Med. 345:517–25 121. Hogaboam CM, Bone-Larson CL, Lipinski S, et al. 1999. Differential monocyte chemoattractant protein-1 and chemokine receptor 2 expression by murine lung fibroblasts derived from Th1- and Th2type pulmonary granuloma models. J. Immunol. 163:2193–201 122. Hogaboam CM, Steinhauser ML, Chensue SW, et al. 1998. Novel roles for chemokines and fibroblasts in interstitial fibrosis. Kidney Int. 54:2152–59 123. Cantin AM, North SL, Fells GA, et al. 1987. Oxidant-mediated epithelial cell injury in idiopathic pulmonary fibrosis. J. Clin. Invest. 79:1665–73 124. Obayashi Y, Yamadori I, Fujita J, et al. 1997. The role of neutrophils in the pathogenesis of idiopathic pulmonary fibrosis. Chest 112:1338–43 125. Cailes JB, O’Connor C, Pantelidis P, et al. 1996. Neutrophil activation in fibrosing alveolitis: a comparison of lone

5 Jan 2004

12:34

AR

AR206-ME55-23.tex

AR206-ME55-23.sgm

LaTeX2e(2002/01/18)

PULMONARY FIBROSIS

126.

Annu. Rev. Med. 2004.55:395-417. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

127.

128.

129. 130.

131.

132.

133.

134.

135.

cryptogenic fibrosing alveolitis and systemic sclerosis. Eur. Respir. J. 9:992–99 Martinet Y, Rom WN, Grotendorst GR, et al. 1987. Exaggerated spontaneous release of platelet-derived growth factor by alveolar macrophages from patients with idiopathic pulmonary fibrosis. N. Engl. J. Med. 317:202–9 Carre PC, Mortenson RL, King TE Jr., et al. 1991. Increased expression of the interleukin-8 gene by alveolar macrophages in idiopathic pulmonary fibrosis. A potential mechanism for the recruitment and activation of neutrophils in lung fibrosis. J. Clin. Invest. 88:1802–10 Nishida M, Fujinaka H, Matsusaka T, et al. 2002. Absence of angiotensin II type 1 receptor in bone marrow-derived cells is detrimental in the evolution of renal fibrosis. J. Clin. Invest. 110:1859–68 Oliver JA. 2002. Unexpected news in renal fibrosis. J. Clin. Invest. 110:1763–64 Bokel C, Brown NH. 2002. Integrins in development: moving on, responding to, and sticking to the extracellular matrix. Dev. Cell 3:311–21 Sternlicht MD, Werb Z. 2001. How matrix metalloproteinases regulate cell behavior. Annu. Rev. Cell Dev. Biol. 17:463–516 Collier IE, Wilhelm SM, Eisen AZ, et al. 1988. H-ras oncogene-transformed human bronchial epithelial cells (TBE-1) secrete a single metalloprotease capable of degrading basement membrane collagen. J. Biol. Chem. 263:6579–87 Pardo A, Ridge K, Uhal B, et al. 1997. Lung alveolar epithelial cells synthesize interstitial collagenase and gelatinases A and B in vitro. Int. J. Biochem. Cell Biol. 29:901–10 Johnson RJ, Couser WG, Chi EY, et al. 1987. New mechanism for glomerular injury. Myeloperoxidase–hydrogen peroxide–halide system. J. Clin. Invest. 79: 1379–87 Raats CJ, Bakker MA, van den Born J, et al. 1997. Hydroxyl radicals depolymerize glomerular heparan sulfate in vitro and

136.

137.

138.

139.

140.

141.

142.

143.

144.

145.

P1: GCE

415

in experimental nephrotic syndrome. J. Biol. Chem. 272:26734–41 Larios JM, Budhiraja R, Fanburg BL, et al. 2001. Oxidative protein cross-linking reactions involving L-tyrosine in transforming growth factor-β1-stimulated fibroblasts. J. Biol. Chem. 276:17437–41 Wright JK, Cawston TE, Hazleman BL. 1991. Transforming growth factor β stimulates the production of the tissue inhibitor of metalloproteinases (TIMP) by human synovial and skin fibroblasts. Biochim. Biophys. Acta 1094:207– 10 Kunkel SL, Lukacs NW, Strieter RM, et al. 1996. Th1 and Th2 responses regulate experimental lung granuloma development. Sarcoidosis Vasc. Diffuse Lung Dis. 13:120–28 Gharaee-Kermani M, McGarry B, Lukacs N, et al. 1998. The role of IL-5 in bleomycin-induced pulmonary fibrosis. J. Leukoc. Biol. 64:657–66 Lee CG, Homer RJ, Zhu Z, et al. 2001. Interleukin-13 induces tissue fibrosis by selectively stimulating and activating transforming growth factor β 1. J. Exp. Med. 194:809–21 Liu X, Kohyama T, Wang H, et al. 2002. Th2 cytokine regulation of type I collagen gel contraction mediated by human lung mesenchymal cells. Am. J. Physiol. Lung Cell Mol. Physiol. 282:L1049–56 Oldroyd SD, Thomas GL, Gabbiani G, et al. 1999. Interferon-γ inhibits experimental renal fibrosis. Kidney Int. 56:2116–27 Ulloa L, Doody J, Massague J. 1999. Inhibition of transforming growth factorbeta/SMAD signalling by the interferonγ /STAT pathway. Nature 397:710–13 Wallace WA, Ramage EA, Lamb D, et al. 1995. A type 2 (Th2-like) pattern of immune response predominates in the pulmonary interstitium of patients with cryptogenic fibrosing alveolitis (CFA). Clin. Exp. Immunol. 101:436–41 Ando M, Miyazaki E, Fukami T, et al.

5 Jan 2004

12:34

416

146.

Annu. Rev. Med. 2004.55:395-417. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

147.

148.

149.

150.

151.

152.

153.

154.

AR

AR206-ME55-23.tex

AR206-ME55-23.sgm

LaTeX2e(2002/01/18)

P1: GCE

THANNICKAL ET AL. 1999. Interleukin-4-producing cells in idiopathic pulmonary fibrosis: an immunohistochemical study. Respirology 4:383– 91 Ziesche R, Hofbauer E, Wittmann K, et al. 1999. A preliminary study of long-term treatment with interferon γ -1b and lowdose prednisolone in patients with idiopathic pulmonary fibrosis. N. Engl. J. Med. 341:1264–69 Honore I, Nunes H, Groussard O, et al. 2003. Acute respiratory failure after interferon-γ therapy of end-stage pulmonary fibrosis. Am. J. Respir. Crit. Care Med. 167:953–57 Folkman J, D’Amore PA. 1996. Blood vessel formation: What is its molecular basis? Cell 87:1153–55 Keane MP, Arenberg DA, Lynch JP 3rd, et al. 1997. The CXC chemokines, IL8 and IP-10, regulate angiogenic activity in idiopathic pulmonary fibrosis. J. Immunol. 159:1437–43 Keane MP, Belperio JA, Burdick MD, et al. 2001. ENA-78 is an important angiogenic factor in idiopathic pulmonary fibrosis. Am. J. Respir. Crit. Care Med. 164:2239–42 Koyama S, Sato E, Haniuda M, et al. 2002. Decreased level of vascular endothelial growth factor in bronchoalveolar lavage fluid of normal smokers and patients with pulmonary fibrosis. Am. J. Respir. Crit. Care Med. 166:382–85 Lappi-Blanco E, Kaarteenaho-Wiik R, Soini Y, et al. 1999. Intraluminal fibromyxoid lesions in bronchiolitis obliterans organizing pneumonia are highly capillarized. Hum. Pathol. 30:1192–96 Lappi-Blanco E, Soini Y, Kinnula V, et al. 2002. VEGF and bFGF are highly expressed in intraluminal fibromyxoid lesions in bronchiolitis obliterans organizing pneumonia. J. Pathol. 196:220–27 Renzoni EA, Walsh DA, Salmon M, et al. 2003. Interstitial vascularity in fibrosing alveolitis. Am. J. Respir. Crit. Care Med. 167:438–43

155. Holgate ST, Peters-Golden M, Panettieri RA, et al. 2003. Roles of cysteinyl leukotrienes in airway inflammation, smooth muscle function, and remodeling. J. Allergy Clin. Immunol. 111:S18–34 156. Wilborn J, Bailie M, Coffey M, et al. 1996. Constitutive activation of 5-lipoxygenase in the lungs of patients with idiopathic pulmonary fibrosis. J. Clin. Invest. 97:1827–36 157. Peters-Golden M, Bailie M, Marshall T, et al. 2002. Protection from pulmonary fibrosis in leukotriene-deficient mice. Am. J. Respir. Crit. Care Med. 165:229–35 158. Lama V, Moore BB, Christensen P, et al. 2002. Prostaglandin E2 synthesis and suppression of fibroblast proliferation by alveolar epithelial cells is cyclooxygenase-2-dependent. Am. J. Respir. Cell Mol. Biol. 27:752–58 159. Kohyama T, Ertl RF, Valenti V, et al. 2001. Prostaglandin E2 inhibits fibroblast chemotaxis. Am. J. Physiol. Lung Cell Mol. Physiol. 281:L1257–63 160. Mio T, Liu XD, Adachi Y, et al. 1998. Human bronchial epithelial cells modulate collagen gel contraction by fibroblasts. Am. J. Physiol. Lung Cell Mol. Physiol. 274:L119–26 161. Kolodsick JE, Peters-Golden M, Larios J, et al. 2003. Prostaglandin E2 inhibits fibroblast to myofibroblast transition via E. prostanoid receptor 2 signaling and cyclic adenosine monophosphate elevation. Am. J. Respir. Cell Mol. Biol. 29:537–44 162. Moore BB, Peters-Golden M, Christensen PJ, et al. 2003. Alveolar epithelial cell inhibition of fibroblast proliferation is regulated by MCP-1/CCR2 and mediated by PGE2. Am. J. Physiol. Lung Cell Mol. Physiol. 284:L342–49 163. Moore BB, Paine R 3rd, Christensen PJ, et al. 2001. Protection from pulmonary fibrosis in the absence of CCR2 signaling. J. Immunol. 167:4368–77 164. MacNee W, Rahman I. 1995. Oxidants/ antioxidants in idiopathic pulmonary fibrosis. Thorax 50(Suppl. 1):S53–58

5 Jan 2004

12:34

AR

AR206-ME55-23.tex

AR206-ME55-23.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:395-417. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

PULMONARY FIBROSIS 165. Rahman I, Skwarska E, Henry M, et al. 1999. Systemic and pulmonary oxidative stress in idiopathic pulmonary fibrosis. Free Radical Biol. Med. 27:60–68 166. Meyer A, Buhl R, Magnussen H. 1994. The effect of oral N-acetylcysteine on lung glutathione levels in idiopathic pulmonary fibrosis. Eur. Respir. J. 7:431–36 167. Borok Z, Buhl R, Grimes GJ, et al. 1991. Effect of glutathione aerosol on oxidantantioxidant imbalance in idiopathic pulmonary fibrosis. Lancet 338:215–16 168. Whyte M, Hubbard R, Meliconi R, et al. 2000. Increased risk of fibrosing alveolitis associated with interleukin-1 receptor antagonist and tumor necrosis factor-α gene polymorphisms. Am. J. Respir. Crit. Care Med. 162:755–58 169. Pantelidis P, Fanning GC, Wells AU, et al. 2001. Analysis of tumor necrosis factorα, lymphotoxin-α, tumor necrosis factor receptor II, and interleukin-6 polymorphisms in patients with idiopathic pulmonary fibrosis. Am. J. Respir. Crit. Care Med. 163:1432–36 170. Yonemaru M, Kasuga I, Kusumoto H, et al. 1997. Elevation of antibodies to cytomegalovirus and other herpes viruses in pulmonary fibrosis. Eur. Respir. J. 10:2040–45 171. Stewart JP, Egan JJ, Ross AJ, et al. 1999. The detection of Epstein-Barr virus DNA in lung tissue from patients with idiopathic pulmonary fibrosis. Am. J. Respir. Crit. Care Med. 159:1336–41 172. Kelly BG, Lok SS, Hasleton PS, et al. 2002. A rearranged form of Epstein-Barr virus DNA is associated with idiopathic pulmonary fibrosis. Am. J. Respir. Crit. Care Med. 166:510–13 173. Baumgartner KB, Samet JM, Coultas

174.

175.

176.

177.

178.

179.

180.

181. 182.

P1: GCE

417

DB, et al. 2000. Occupational and environmental risk factors for idiopathic pulmonary fibrosis: a multicenter casecontrol study. Collaborating Centers. Am. J. Epidemiol. 152:307–15 Mullen J, Hodgson MJ, DeGraff CA, Godar T. 1998. Case-control study of idiopathic pulmonary fibrosis and environmental exposures. J. Occup. Environ. Med. 40:363–67 Baumgartner KB, Samet JM, Stidley CA, et al. 1997. Cigarette smoking: a risk factor for idiopathic pulmonary fibrosis. Am. J. Respir. Crit. Care Med. 155:242– 48 Flaherty KR, Toews GB, Travis WD, et al. 2002. Clinical significance of histological classification of idiopathic interstitial pneumonia. Eur. Respir. J. 19:275– 83 Nobukuni S, Watanabe K, Inoue J, et al. 2002. Cigarette smoke inhibits the growth of lung fibroblasts from patients with pulmonary emphysema. Respirology 7:217– 23 Coultas DB, Zumwalt RE, Black WC, et al. 1994. The epidemiology of interstitial lung diseases. Am. J. Respir. Crit. Care Med. 150:967–72 Shelton DN, Chang E, Whittier PS, et al. 1999. Microarray analysis of replicative senescence. Curr. Biol. 9:939–45 Finkel T, Holbrook NJ. 2000. Oxidants, oxidative stress and the biology of ageing. Nature 408:239–47 Kudravi SA, Reed MJ. 2000. Aging, cancer, and wound healing. In Vivo 14:83–92 White ES, Lazar MH, Thannickal VJ. 2003. Pathogenetic mechanisms in usual interstitial pneumonia/idiopathic pulmonary fibrosis. J. Pathol. 201:343–54

Victor.qxd

1/5/2004

3:39 PM

Page 1

Annu. Rev. Med. 2004.55:395-417. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

PULMONARY FIBROSIS

C-1

Figure 2 Low-magnification photomicrographs (hematoxylin and eosin) of desquamative interstitial pneumonia (DIP), nonspecific interstitial pneumonia (NSIP) and usual interstitial pneumonia (UIP) demonstrating varying degrees of inflammation and fibrosis. DIP (left) is primarily an inflammatory reaction characterized by intraalveolar accumulation of macrophages. NSIP (middle) features both inflammation (primarily of interstitial lymphocytes) and fibrosis that is of diffuse and fairly uniform distribution. UIP (right) represents a more heterogeneous pattern with areas of dense fibrosis on the left and relatively normal-appearing alveoli on the right. A zone of microscopic honeycomb change is seen at top left (arrowheads), represented by enlarged air spaces filled with mucin and separated by dense bands of interstitial fibrosis. A fibroblastic focus (arrow) is seen in the center of the field in association with a few inflammatory cells. Photomicrographs were reproduced with permission (1).

Victor.qxd

Annu. Rev. Med. 2004.55:395-417. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

C-2

1/5/2004

3:39 PM

Page 2

THANNICKAL ET AL.

Figure 4 Epithelial-mesenchymal communication is dysregulated in progressive pulmonary fibrosis with aberrant bidirectional signaling between tissue compartments. Inflammatory cells that modulate this process may be recruited and activated by secreted products from these “resident” cells. Resultant effects on alveolar epithelial cells include increased apoptosis, dysregulated proliferation, and ineffective migration. Effects on mesenchymal cells include myofibroblast differentiation, resistance to apoptosis, and enhanced extracellular matrix (ECM) secretion. Migration of fibroblasts to sites of injury probably occurs early in the process. The proliferative phenotype of the mesenchymal cells in fibroblastic foci has not been well defined. It is important to recognize that the cellular composition and phenotypes as well as the intraalveolar milieu of soluble mediators, matrix molecules, and reactive oxygen/nitrogen species (ROS/RNS) may be in a state of flux and regulated in a time- and context-dependent manner. Dysregulation of the repair process leads to persistent mesenchymal cell activation that culminates in pulmonary fibrosis. Abbreviations: TGF, transforming growth factor; TNF, tumor necrosis factor; PGE2, prostaglandin E2; PDGF, platelet-derived growth factor; reactive oxygen/nitrogen species, ROS/RNS; HGF, hepatocyte growth factor; TIMP, tissue inhibitors of matrix metalloproteinases; MMP, matrix metalloproteinase; PAI-1, plasminogen activator inhibitor-1; uPA, urokinase-type plasminogen activator. (Adapted from Reference 182 with minor modifications.)

12 Dec 2003

14:27

AR

AR206-ME55-24.tex

AR206-ME55-24.sgm

LaTeX2e(2002/01/18) P1: GCE 10.1146/annurev.med.55.091902.103822

Annu. Rev. Med. 2004. 55:419–32 doi: 10.1146/annurev.med.55.091902.103822

SYSTEMIC MASTOCYTOSIS∗ Cem Akin and Dean D. Metcalfe

Annu. Rev. Med. 2004.55:419-432. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Laboratory of Allergic Diseases, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, Maryland 20892; email: [email protected], [email protected]

Key Words mast cells, urticaria pigmentosa, c-kit, bone marrow ■ Abstract Systemic mastocytosis is a clonal disorder of the mast cell and its progenitor. The symptoms of systemic mastocytosis are due to the pathologic accumulation and activation of mast cells in various tissues such as bone marrow, skin, gastrointestinal tract, liver, and spleen. Recent studies revealed striking differences between the molecular and cellular biology of mast cells in patients with mastocytosis and those of healthy individuals. These findings are being used in formulating diagnostic criteria as well as designing novel treatment approaches to the disease.

CLINICAL SPECTRUM Although clinical manifestations of mast-cell disease have been recognized since the middle of the nineteenth century (1), most of our current knowledge about the molecular and cellular basis of the disease is derived from studies performed within the past 10 years. This knowledge has allowed new insights into the etiology, diagnosis, classification, and therapy of systemic mastocytosis. The hallmark of mastocytosis is a pathologic accumulation of mast cells in various tissues. Symptoms related to mast-cell degranulation are frequently observed in all categories of mastocytosis. These may include episodic flushing, dyspepsia, diarrhea, abdominal pain, musculoskeletal pain, or hypotension. In addition, some patients may have a non–mast-cell clonal hematologic disorder, whose clinical presentation will reflect that of the associated hematologic disorder. Depending on the number and extent of the tissues involved, the disease may present with a spectrum of clinical manifestations. On the benign end of the spectrum is pediatric-onset mastocytosis limited to skin (2–4). Patients with this variant of mastocytosis are typically diagnosed with urticaria pigmentosa (UP) skin lesions within the first six months of life and experience regression or improvement of the skin lesions by the time they reach puberty. On the other hand, patients ∗ The U.S. Government has the right to retain a nonexclusive, royalty-free license in and to any copyright covering this paper.

419

12 Dec 2003

14:27

Annu. Rev. Med. 2004.55:419-432. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

420

AR

AKIN

AR206-ME55-24.tex

¥

AR206-ME55-24.sgm

LaTeX2e(2002/01/18)

P1: GCE

METCALFE

with adult-onset mastocytosis generally have evidence of systemic mastocytosis, defined by involvement of an organ system other than skin. Systemic mastocytosis is a persistent or progressive clonal disorder of the mast-cell progenitor associated with activating mutations in the c-kit gene. Systemic mastocytosis can follow a benign or indolent course, or it may be associated with life-threatening hematologic disorders. Therefore, correct diagnosis and classification of disease are important for an accurate prognosis. The diagnosis of systemic mastocytosis can be challenging. A physician may incorrectly diagnose mastocytosis in a patient who does not have the disease or miss it in a patient who does. Systemic mastocytosis is a clonal disorder of the hematopoietic system, and as with other clonal disorders of the bone marrow, its diagnosis cannot be based on symptoms alone. Cutaneous mastocytosis, when present, greatly facilitates the diagnosis. Most of the diagnostic challenges, however, arise in patients who do not have typical UP skin lesions.

DIAGNOSIS OF CUTANEOUS MASTOCYTOSIS Mastocytosis involves the skin in ∼80% of patients. The most common skin manifestation of mastocytosis is UP (5). Lesions of UP are fixed, hyperpigmented red-brown macules or papules. Their diameter usually does not exceed 0.5 cm, but individual lesions may be larger in children. UP lesions may urticate or flare when the skin is exposed to temperature changes or physical irritation. Darier’s sign, the local whealing of the UP lesion when rubbed or scratched, must be distinguished from generalized dermatographism, which does not discriminate between lesional and normal skin. It is recommended that the suspicion of cutaneous mastocytosis be confirmed by a skin biopsy of a lesion. Characteristic skin biopsy findings include multifocal aggregates of mast cells in the upper dermis (6). Perivascular areas are frequently involved. Typically, mast cells are increased tenfold or more in lesions of UP when compared to normal skin. Milder increases in mast-cell numbers are observed in patients with other conditions such as unexplained flushing, chronic urticaria, and atopic dermatitis (7, 8). It is therefore not advisable to establish the diagnosis based solely on mild histopathologic increases in mast-cell numbers without accompanying UP lesions. Lesions of UP are more variable in childhood-onset cutaneous mastocytosis, which usually presents within the first six months of life (9). The lesions may be larger and some are plaque-like. Skin lesions do not initially appear as hyperpigmented as the lesions of adult-onset cutaneous mastocytosis, although they usually take on pigmentation as the child matures. Formation of bullae over the lesions may be encountered up to age three and should be differentiated from other bullous diseases of childhood (10). Other rare forms of cutaneous mastocytosis include mastocytoma (11), diffuse cutaneous mastocytosis (12), and telangiectasia macularis eruptiva perstans (TMEP) (13). Cutaneous mastocytomas are solid, benign mast-cell tumors and are

12 Dec 2003

14:27

AR

AR206-ME55-24.tex

AR206-ME55-24.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:419-432. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

SYSTEMIC MASTOCYTOSIS

P1: GCE

421

diagnosed by biopsy. They may precede or coexist with UP. Diffuse cutaneous mastocytosis is a diffuse involvement of the entire skin and may result from confluence of UP lesions. Histopathology shows extensive infiltration of the dermis by mast cells. TMEP is a rare form of cutaneous mastocytosis characterized by generalized telangiectatic macules that exhibit Darier’s sign. As with other forms of cutaneous mastocytosis, this diagnosis cannot be established by histopathology alone. TMEP usually accompanies some other form of cutaneous or systemic mastocytosis. UP, TMEP, and plaque-like lesions are sometimes collectively referred to as maculopapular cutaneous mastocytosis (6). The recommended method of evaluation of mast cells in tissues is immunohistochemical staining with tryptase (14–16). This method is more sensitive and reliable than metachromatic stains such as Giemsa or toluidine blue. Skin biopsy usually demonstrates multifocal or diffuse aggregates of mast cells in the papillary dermis extending into the reticular dermis. It is important to determine whether the patient presenting with cutaneous mastocytosis also has evidence of systemic mastocytosis. Figure 1 outlines one approach to this diagnostic evaluation. Cutaneous mastocytosis in the absence of systemic disease is common in patients with pediatric-onset mastocytosis. In contrast, cutaneous mastocytosis is accompanied by systemic disease in most, if not all, patients who experience onset of lesions after age two.

DIAGNOSIS OF SYSTEMIC MASTOCYTOSIS Systemic mastocytosis is diagnosed when there is evidence of involvement of a tissue other than skin. The most common extracutaneous tissue sites are bone marrow, spleen, liver, lymph nodes, and gastrointestinal tract. Comparative information is not available about diagnostic aspects of normal and pathologic mast cells in tissues other than bone marrow in patients with systemic mastocytosis. This is because (a) these tissues are not generally biopsied in the routine evaluation of patients with mastocytosis, (b) mast cells may be elevated in other inflammatory and infectious diseases of these tissues, and particularly (c) bone marrow is almost always involved in systemic mastocytosis. For these reasons, a bone marrow biopsy and aspirate is the procedure of choice in cases of suspected systemic mastocytosis. Because bone marrow involvement is not common in the pediatric population, in contrast to patients with adult-onset disease, a bone marrow biopsy is generally reserved for children with a high likelihood of systemic involvement. These include patients with late onset of skin lesions (i.e., after two years of age) and those with hepatomegaly, splenomegaly, unexplained pathologic lymphadenopathy, abnormalities in complete blood count, or baseline serum total tryptase levels of >20 ng/ml. Examination of the bone marrow in patients with systemic mastocytosis provides valuable information about the extent of the disease and the presence or absence of a non–mast-cell hematologic disorder. It also helps the physician to counsel patients about prognosis.

12 Dec 2003

14:27

AKIN

AR206-ME55-24.tex

¥

AR206-ME55-24.sgm

LaTeX2e(2002/01/18)

METCALFE

Annu. Rev. Med. 2004.55:419-432. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

422

AR

Figure 1 Evaluation of patients with suspected mastocytosis.

P1: GCE

12 Dec 2003

14:27

AR

AR206-ME55-24.tex

AR206-ME55-24.sgm

LaTeX2e(2002/01/18)

SYSTEMIC MASTOCYTOSIS

P1: GCE

423

Annu. Rev. Med. 2004.55:419-432. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Diagnostic Criteria for Systemic Mastocytosis Recent years have witnessed important advancements in our understanding of the molecular and cellular pathology of mast-cell disease. Among these are the discovery of c-kit mutations (17, 18), identification of surrogate disease markers (19, 20), and characterization of surface immunophenotype of normal and neoplastic mast cells (21). Such research findings have been translated into clinical use in the recently established criteria for the diagnosis of systemic mastocytosis (22, 23). These consist of one major and four minor criteria. In order to satisfy the diagnosis of systemic mastocytosis, a patient should meet either the major criterion and one minor criterion, or three minor criteria. The major criterion for diagnosis of systemic mastocytosis is a finding of dense infiltrates of mast cells in bone marrow or other extracutaneous tissues. This is considered the most specific pathologic feature of mastocytosis. In order to satisfy this criterion, mast cells should be observed in aggregates of 15 or more, as confirmed by tryptase immunohistochemistry or metachromatic stains such as Giemsa or toluidine blue. In bone marrow, mast-cell aggregates can be observed in paratrabecular and perivascular locations and may be associated with a benign lymphoid aggregate consisting of B and T cells (24–26). Eosinophils are often present in variable numbers in these lesions. The decalcification process necessary for sectioning of the paraffin-embedded bone marrow tissue often interferes with the metachromatic staining of the mast cells. Immunohistochemical staining for mast-cell tryptase is thus more reliable and has largely replaced metachromatic staining for the diagnosis of mast-cell disease. Most patients with the major criteria also have one of the minor criteria and thus can be diagnosed with systemic mastocytosis. If the disease is at an early stage, or if the bone marrow biopsy sample is inadequate, the major criterion may be absent. In this case, three of the four minor criteria must be fulfilled for diagnosis. The minor criteria are (a) atypical mast-cell morphology, (b) aberrant mast-cell surface immunophenotype, (c) serum/plasma tryptase level of greater than 20 ng/ml, and (d) a codon 816 c-kit mutation in peripheral blood, bone marrow, or lesional tissue. ATYPICAL MAST-CELL MORPHOLOGY A normal mast cell has a round or oval shape, a round and centrally located nonlobated nucleus, and a fully granulated cytoplasm. The nucleus of a normal mast cell may be obscured by granular staining. Mast cells in bone marrow from patients with systemic mastocytosis may exhibit a number of phenotypic aberrations, including a spindle shape, cytoplasmic projections, and hypogranulation (27). Mast cells with a multilobular nucleus and an eccentric nucleus are also observed. In order to fulfill this criterion, at least 25% of all mast cells in the biopsy sections or aspirate smears must have aberrant morphology. Mast cells are rare in normal marrow and may be difficult to detect in aspirates. They are found in close proximity to spicules in aspirate smears in patients with

12 Dec 2003

14:27

424

AR

AKIN

AR206-ME55-24.tex

¥

AR206-ME55-24.sgm

LaTeX2e(2002/01/18)

P1: GCE

METCALFE

Annu. Rev. Med. 2004.55:419-432. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

mastocytosis. An aggressive variant of the disease (such as mast-cell leukemia) must be considered if the percentage of mast cells exceeds 20% of all nucleated cells in the bone marrow aspirate. ABERRENT MAST-CELL SURFACE IMMUNOPHENOTYPE Unlike normal bone marrow mast cells, those from patients with indolent systemic mastocytosis express CD2, CD25, and CD35 (21). The aberrant immunophenotype is readily detected by multicolor flow cytometric analysis of the bone marrow aspirate. Mast cells are distinguished from the other cells by bright staining with CD117 (Kit) and surface IgE and by high side-scatter properties. Because mast cells constitute a minor fraction of the bone marrow cells (usually from 0.002% to 0.05% in normal marrow), it is important to analyze a sufficient number of cells—at least 300,000—to acquire sufficient events to analyze mast cells by flow cytometry. Surface expression of CD25, but not CD2, has recently been reported in bone marrow mast cells in the myeloproliferative variant of hypereosinophilic syndrome (28). SERUM/PLASMA TRYPTASE LEVEL >20 NG/ML Currently available commercial assay systems for tryptase measure total tryptase levels. Total tryptase levels elevate in serum or plasma in association with anaphylaxis. In patients with mastocytosis, the total tryptase may show persistent elevation. Thus, in order to be of value in diagnosis of mastocytosis (29), the serum tryptase level must be obtained when the patient is at his or her baseline state of health and not after or during an anaphylactic episode. The mean total level of tryptase (believed to be mostly non-beta or immature tryptase as opposed to beta or mature tryptase released in anaphylaxis) in healthy individuals is ∼5 ng/ml. Levels of >20 ng/ml are associated with systemic mastocytosis (29). Patients with cutaneous disease alone or limited systemic disease may have tryptase levels of <20 ng/ml. High tryptase levels are also reported in some patients with myelodysplastic syndromes, hypereosinophilic syndrome, and other myeloid neoplasms (28, 30, 31). Thus, a tryptase level of more than 20 ng/ml alone is not diagnostic of a mast-cell disorder.

CODON 816 C-KIT MUTATION IN PERIPHERAL BLOOD, BONE MARROW, OR LESIONAL TISSUE Somatic point mutations in codon 816 (most commonly Asp816Val) of

the proto-oncogene c-kit are associated with adult-onset systemic mastocytosis (18, 32). The protein product of c-kit is a receptor tyrosine kinase (Kit), which is activated by binding of stem cell factor (SCF or Kit ligand), the major mast-cell growth and differentiation factor. Codon 816 point mutations result in ligandindependent activation and constitutive autophosphorylation of Kit. Such mutations have been detected in the lesional tissues (skin or bone marrow) of most patients with systemic mastocytosis examined to date (33, 34). In patients with extensive disease, the mutation may be detectable in peripheral blood cells, most frequently in monocytes and B cells (34). This observation suggests involvement of a hematopoietic progenitor cell in the etiology of mastocytosis. Thus, detectability

12 Dec 2003

14:27

AR

AR206-ME55-24.tex

AR206-ME55-24.sgm

LaTeX2e(2002/01/18)

SYSTEMIC MASTOCYTOSIS

P1: GCE

425

of the codon 816 mutation in unfractionated peripheral blood white cells correlates with the amount of clonal expansion derived from the mutated hematopoietic progenitor. Codon 816 mutations are not detectable in most patients with typical pediatric-onset cutaneous mastocytosis, although there are exceptions.

Annu. Rev. Med. 2004.55:419-432. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

CLASSIFICATION Classification of mastocytosis is based on World Health Organization criteria (Table 1) (22, 23). Cutaneous mastocytosis is the most frequently observed category in patients with pediatric-onset disease, whereas indolent systemic mastocytosis is the most common in patients with adult-onset disease. Indolent systemic mastocytosis is differentiated from more advanced categories of systemic mastocytosis by (a) lack of evidence of an associated clonal hematological non–mast-cell disorder, (b) lack of end organ dysfunction such as cytopenias, ascites, malabsorption, and pathological fractures (for aggressive systemic mastocytosis), and (c) lack of evidence of a mast-cell leukemia. A new subcategory of indolent systemic mastocytosis, termed indolent smoldering mastocytosis, was introduced in the revised classification to account for patients who have extensive mast-cell disease (as evidenced by 30% or more infiltration of the bone marrow cavity by mast cells, serum tryptase levels of >200 ng/ml, and hepatosplenomegaly) but no distinct evidence of an associated clonal non–mast-cell disease or aggressive mastocytosis (35–37). It is not known whether cases of indolent smoldering mastocytosis are more likely to advance to a more aggressive form of the disease. Figure 2 summarizes this approach to the classification of systemic mastocytosis.

TREATMENT Modalities used in the treatment of mastocytosis can be divided into two broad categories: (a) those intended to control symptoms due to mediators released from mast cells and (b) those intended to reduce mast-cell burden. Because there is no therapy proven to cure mast-cell disease, symptom control is an important part of the treatment strategy. Commonly used medications include H1 antihistamines for pruritus, H2 antihistamines for peptic symptoms, and epinephrine (as needed) for episodes of hypotension (38, 39). Oral cromolyn sodium may be beneficial in controlling gastrointestinal symptoms (40, 41). Glucocorticoids have been used successfully in patients with frequent hypotensive episodes, as well as those with ascites and diarrhea associated with malabsorption. Psoralen ultraviolet A therapy may provide transient relief of pruritus and fading of skin lesions in some patients (42). Chemotherapy has no recognized role in treatment of systemic indolent mastocytosis. Patients with more advanced categories of systemic mastocytosis may be candidates for several approaches that aim to control the mast-cell burden, although

Diagnostic features Lack of systemic involvement Age of onset generally <2 years Lack of advanced categories of mastocytosis Age of onset generally >2 years Most common category in adult-onset disease Commonly associated with myelodysplastic or myeloproliferative disorders Occasionally seen with acute leukemias and lymphomas Findings of end organ dysfunction due to mast-cell infiltration, such as: bone marrow failure liver dysfunction with ascites splenomegaly with hypersplenism skeletal osteolytes with pathologic fractures gastrointestinal involvement with malabsorption and weight loss Mast cells with high grade morphology (multilobular or multiple nuclei) >10% mast cells in peripheral blood or >20% mast cells in bone marrow aspirate smears Malignant and destructive soft tissue tumor Mast cells with high grade morphology Rare benign tumor consisting of mature mast cells

Cutaneous mastocytosis

Indolent systemic mastocytosis

Systemic mastocytosis with associated clonal hematological non–mast-cell lineage disease

Aggressive systemic mastocytosis

Mast-cell leukemia

Mast-cell sarcoma

Extracutaneous mastocytoma

LaTeX2e(2002/01/18)

Good

Poor

AR206-ME55-24.sgm

Poor

AR206-ME55-24.tex

Poor

¥

Same as the associated non–mast-cell disorder

AKIN

Good

AR

Good

Prognosis

14:27

Category

426

TABLE 1 Classification of mastocytosis (adapted from World Health Organization criteria)

Annu. Rev. Med. 2004.55:419-432. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

12 Dec 2003 P1: GCE

METCALFE

12 Dec 2003

14:27

AR

AR206-ME55-24.tex

AR206-ME55-24.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:419-432. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

SYSTEMIC MASTOCYTOSIS

P1: GCE

427

Figure 2 An algorithm for classification of systemic mastocytosis to substantiate this diagnosis in patients with one major and one minor criterion, or three minor criteria. MPD, myeloproliferative disorder; MDS, myelodysplastic syndrome; SM-AHNMD, systemic mastocytosis with associated clonal hematological non–mast-cell lineage disease; ISM, indolent systemic mastocytosis; ASM, aggressive systemic mastocytosis.

none of these approaches has consistently resulted in cure of the disease. Interferon alpha is the drug for which most experience has been reported (43). Although its mechanism of action in mastocytosis is not known, it is presumed to act by restricting the proliferative potential of hematopoietic progenitor cells. Using the recently proposed response criteria, one study determined that the patients who appeared to have derived the most benefit from interferon alpha were those with aggressive systemic mastocytosis (44). The overall response rate in 14 patients with aggres-

12 Dec 2003

14:27

Annu. Rev. Med. 2004.55:419-432. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

428

AR

AKIN

AR206-ME55-24.tex

¥

AR206-ME55-24.sgm

LaTeX2e(2002/01/18)

P1: GCE

METCALFE

sive systemic mastocytosis in this retrospective meta-analysis was 57%, although only 21% of the patients had a major response. However, interferon alpha failed to produce a beneficial effect in other studies (45), and its use is not routinely recommended for patients with indolent systemic disease, who generally have a favorable prognosis. 2-Chlorodeoxyadenosine (cladribine, 2-CDA), a nucleoside analogue, was reported to induce a major clinical and histopathological response in one patient with systemic mastocytosis (46). This drug does not appear to require cells in active cell cycle to exert its cytotoxic activity, and thus it may be beneficial in slowly progressing neoplastic processes. Because the drug has myelosuppressive and immunosuppressive properties, it cannot be recommended for patients with indolent disease with normal life expectancy. Anecdotal experience suggests that there exists a subpopulation of patients with advanced/aggressive mastocytosis who respond to 2-CDA, although the responses generally appear to be transient. Allogeneic bone marrow transplantation (BMT) is currently being investigated as a treatment option for patients with advanced categories of mastocytosis associated with poor survival. In only a handful of reported instances has BMT been performed to treat a hematologic disorder associated with mastocytosis (47–49). Although these studies reported favorable responses of the associated hematologic disorders, a complete remission of the mast-cell disease was seen in only one study, which used non–T cell–depleted peripheral blood stem cell transplantation in a patient with an associated myeloproliferative disorder (49). This suggests that the value of allogeneic BMT in mastocytosis may be due to the immunotherapeutic effects of the donor marrow rather than the initial myeloablative conditioning regimen. A protocol utilizing a nonmyeloablative peripheral blood stem cell transplantation in treatment of advanced systemic mastocytosis is under investigation at the National Institutes of Health Clinical Center. Of three patients transplanted under this protocol thus far, two have achieved a complete response of their associated hematologic disorder (myeloproliferative and myelodysplastic) and a partial response of their mast-cell disease, with reduced tryptase levels but persistent mast-cell collections in the bone marrow (50). The availability of small-molecular-weight inhibitors of tyrosine kinase suggested the mutated Kit tyrosine kinase in mastocytosis as a therapeutic target. Imatinib mesylate (Gleevec; Novartis, Basel, Switzerland) is currently the only such drug approved by the US Food and Drug Administration. It has a fairly specific inhibition profile that includes bcr-abl, Kit, and platelet-derived growth factor-receptor tyrosine kinases (51–53). In vitro studies investigating the ability of imatinib to inhibit various mutants of c-kit revealed that although the drug effectively inhibited wild-type Kit and Kit bearing juxtamembrane activating mutations (similar to those found in gastrointestinal stromal tumors), it failed to inhibit Kit bearing codon 816 mutations associated with most common forms of systemic mastocytosis (54, 55). This has been attributed to a conformational change in Kit bearing the codon 816 mutation, which interferes with the association of the drug with the ATP-binding domains of the receptor. Consistent with these observations,

12 Dec 2003

14:27

AR

AR206-ME55-24.tex

AR206-ME55-24.sgm

LaTeX2e(2002/01/18)

SYSTEMIC MASTOCYTOSIS

P1: GCE

429

imatinib showed a strong in vitro cytotoxic effect on mast cells bearing wild-type Kit, whereas mast cells bearing a codon 816 mutation isolated from bone marrow aspirates of patients with mastocytosis were fairly resistant to the drug (56). These studies suggest that imatinib is unlikely to be an effective therapy for patients who carry codon 816 mutations. However, imatinib might be of value in unusual clinical presentations of mastocytosis, which are not associated with codon 816 mutations. A careful mutational analysis of a sample enriched for lesional mast cells appears essential before contemplating therapy with imatinib.

Annu. Rev. Med. 2004.55:419-432. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

The Annual Review of Medicine is online at http://med.annualreviews.org

LITERATURE CITED 1. Nettleship E, Tay W. 1869. Rare forms of urticaria. BMJ 2:323 2. Carter MC, Metcalfe DD. 2002. Paediatric mastocytosis. Arch. Dis. Child 86:315– 19 3. Kettelhut BV, Metcalfe DD. 1991. Pediatric mastocytosis. J. Invest. Dermatol. 96:15S– 18S 4. Sondergaard J, Asbo-Hansen E. 1987. Mastocytosis in childhood. In Pediatric Dermatology, ed. R Happle, pp. 148–54. Berlin: Springer-Verlag 5. Soter NA. 1991. The skin in mastocytosis. J. Invest. Dermatol. 96:32S–38S; discussion 8S–9S 6. Wolff K, Komar M, Petzelbauer P. 2001. Clinical and histopathological aspects of cutaneous mastocytosis. Leuk. Res. 25: 519–28 7. Garriga MM, Friedman MM, Metcalfe DD. 1988. A survey of the number and distribution of mast cells in the skin of patients with mast cell disorders. J. Allergy Clin. Immunol. 82:425–32 8. Brockow K, Akin C, Huber M, et al. 2002. Levels of mast-cell growth factors in plasma and in suction skin blister fluid in adults with mastocytosis: correlation with dermal mast-cell numbers and mast-cell tryptase. J. Allergy Clin. Immunol. 109:82– 88 9. Brockow K, Akin C, Huber M, et al. 2003. Assessment of the extent of cutaneous involvement in children and adults with

10.

11.

12.

13.

14.

15.

16.

mastocytosis: relationship to symptomatology, tryptase levels, and bone marrow pathology. J. Am. Acad. Dermatol. 48:508– 16 Golitz LE, Weston WL, Lane AT. 1984. Bullous mastocytosis: diffuse cutaneous mastocytosis with extensive blisters mimicking scalded skin syndrome or erythema multiforme. Pediatr. Dermatol. 1:288– 94 Chagrin L, Sachs P. 1954. Urticaria pigmentosa appearing as a solitary nodular lesion. Arch. Dermatol. Syphilol. 69:345–55 Findlay GH, Schulz EJ, Pepler WJ. 1960. Diffuse cutaneous mastocytosis. S. Afr. Med. J. 34:353 Parkes Weber F, Rast H. 1930. Telangiectasia macularis eruptiva perstans. Br. J. Dermatol. Syph. 42:374–82 Walls AF, Jones DB, Williams JH, et al. 1990. Immunohistochemical identification of mast cells in formaldehyde-fixed tissue using monoclonal antibodies specific for tryptase. J. Pathol. 162:119–26 Li WV, Kapadia SB, Sonmez-Alpan E, et al. 1996. Immunohistochemical characterization of mast cell disease in paraffin sections using tryptase, CD68, myeloperoxidase, lysozyme, and CD20 antibodies. Mod. Pathol. 9:982–88 Horny HP, Sillaber C, Menke D, et al. 1998. Diagnostic value of immunostaining for tryptase in patients with mastocytosis. Am. J. Surg. Pathol. 22:1132–40

12 Dec 2003

14:27

Annu. Rev. Med. 2004.55:419-432. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

430

AR

AKIN

AR206-ME55-24.tex

¥

AR206-ME55-24.sgm

LaTeX2e(2002/01/18)

P1: GCE

METCALFE

17. Furitsu T, Tsujimura T, Tono T, et al. 1993. Identification of mutations in the coding sequence of the proto-oncogene c-kit in a human mast cell leukemia cell line causing ligand-independent activation of ckit product. J. Clin. Invest. 92:1736–44 18. Nagata H, Worobec AS, Oh CK, et al. 1995. Identification of a point mutation in the catalytic domain of the protooncogene c-kit in peripheral blood mononuclear cells of patients who have mastocytosis with an associated hematologic disorder. Proc. Natl. Acad. Sci. USA 92:10560–64 19. Schwartz LB, Sakai K, Bradford TR, et al. 1995. The alpha form of human tryptase is the predominant type present in blood at baseline in normal subjects and is elevated in those with systemic mastocytosis. J. Clin. Invest. 96:2702–10 20. Akin C, Schwartz LB, Kitoh T, et al. 2000. Soluble stem cell factor receptor (CD117) and IL-2 receptor alpha chain (CD25) levels in the plasma of patients with mastocytosis: relationships to disease severity and bone marrow pathology. Blood 96:1267– 73 21. Escribano L, Orfao A, Diaz-Agustin B, et al. 1998. Indolent systemic mast cell disease in adults: immunophenotypic characterization of bone marrow mast cells and its diagnostic implications. Blood 91:2731– 36 22. Valent P, Horny H, Escribano L, et al. 2001. Diagnostic criteria and classification of mastocytosis: a consensus proposal. Leuk. Res. 25:603–25 23. Valent P, Horny HP, Li CY, et al. 2001. Mastocytosis. In World Health Organization Classification of Tumours. Pathology and Genetics of Tumours of the Haematopoietic and Lymphoid Tissues, ed. ES Jaffe, NL Harris, H Stein, JW Vardiman, pp. 291– 302. Lyon: IARC Press 24. Parker RI. 1991. Hematologic aspects of mastocytosis. I. Bone marrow pathology in adult and pediatric systemic mast cell disease. J. Invest. Dermatol. 96:47S–51S 25. Horny HP, Kaiserling E. 1988. Lymphoid

26.

27.

28.

29.

30.

31.

32.

33.

34.

cells and tissue mast cells of bone marrow lesions in systemic mastocytosis: a histological and immunohistological study. Br. J. Haematol. 69:449–55 Akin C, Jaffe ES, Raffeld M, et al. 2002. An immunohistochemical study of the bone marrow lesions of systemic mastocytosis: expression of stem cell factor by lesional mast cells. Am. J. Clin. Pathol. 118:242– 47 Sperr WR, Escribano L, Jordan JH, et al. 2001. Morphologic properties of neoplastic mast cells: delineation of stages of maturation and implication for cytological grading of mastocytosis. Leuk. Res. 25:529– 36 Klion AD, Noel P, Akin C, et al. 2003. Elevated serum tryptase levels identify a subset of patients with a myeloproliferative variant of idiopathic hypereosinophilic syndrome associated with tissue fibrosis, poor prognosis, and imatinib responsiveness. Blood 101:4660–66 Schwartz LB. 2001. Clinical utility of tryptase levels in systemic mastocytosis and associated hematologic disorders. Leuk. Res. 25:553–62 Sperr WR, Jordan JH, Stehberger B, et al. 1999. Detection of elevated serum tryptase levels in MDS. Blood 94:287b Sperr WR, Jordan JH, Baghestanian M, et al. 2001. Expression of mast cell tryptase by myeloblasts in a group of patients with acute myeloid leukemia. Blood 98:2200–9 Longley BJ, Tyrrell L, Lu SZ, et al. 1996. Somatic c-KIT activating mutation in urticaria pigmentosa and aggressive mastocytosis: establishment of clonality in a human mast cell neoplasm. Nat. Genet. 12:312– 14 Longley BJ Jr, Metcalfe DD, Tharp M, et al. 1999. Activating and dominant inactivating c-KIT catalytic domain mutations in distinct clinical forms of human mastocytosis. Proc. Natl. Acad. Sci. USA 96:1609– 14 Akin C, Kirshenbaum AS, Semere T, et al. 2000. Analysis of the surface expression of

12 Dec 2003

14:27

AR

AR206-ME55-24.tex

AR206-ME55-24.sgm

LaTeX2e(2002/01/18)

SYSTEMIC MASTOCYTOSIS

35.

Annu. Rev. Med. 2004.55:419-432. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

36.

37.

38.

39.

40.

41.

42.

43.

44.

c-kit and occurrence of the c-kit Asp816Val activating mutation in T cells, B cells, and myelomonocytic cells in patients with mastocytosis. Exp. Hematol. 28:140–47 Jordan J, Fritsche-Polanz R, Sperr WR, et al. 2001. A case of ‘smouldering’ mastocytosis with high mast cell burden, monoclonal myeloid cells, and C-KIT mutation Asp-816-Val. Leuk. Res. 25:627–34 Akin C, Scott LM, Metcalfe DD. 2001. Slowly progressive systemic mastocytosis with high mast-cell burden and no evidence of a non-mast-cell hematologic disorder: an example of a smoldering case? Leuk. Res. 25:635–38 Valent P, Akin C, Sperr WR, et al. 2002. Smouldering mastocytosis: a novel subtype of systemic mastocytosis with slow progression. Int. Arch. Allergy Immunol. 127:137–39 Worobec AS. 2000. Treatment of systemic mast cell disorders. Hematol. Oncol. Clin. North Am. 14:659–87,vii Escribano L, Akin C, Castells M, et al. 2002. Mastocytosis: current concepts in diagnosis and treatment. Ann. Hematol. 81:677–90 Soter NA, Austen KF, Wasserman SI. 1979. Oral disodium cromoglycate in the treatment of systemic mastocytosis. N. Engl. J. Med. 301:465–69 Horan RF, Sheffer AL, Austen KF. 1990. Cromolyn sodium in the management of systemic mastocytosis. J. Allergy Clin. Immunol. 85:852–55 Czarnetzki BM, Rosenbach T, Kolde G, et al. 1985. Phototherapy of urticaria pigmentosa: clinical response and changes of cutaneous reactivity, histamine and chemotactic leukotrienes. Arch. Dermatol. Res. 277:105–13 Kluin-Nelemans HC, Jansen JH, Breukelman H, et al. 1992. Response to interferon alfa-2b in a patient with systemic mastocytosis. N. Engl. J. Med. 326:619–23 Valent P, Akin C, Sperr WR, et al. 2003. Aggressive systemic mastocytosis and related mast cell disorders: current treatment op-

45.

46.

47.

48.

49.

50.

51.

52.

53.

54.

P1: GCE

431

tions and proposed response criteria. Leuk. Res. 27:635–41 Worobec AS, Kirshenbaum AS, Schwartz LB, et al. 1996. Treatment of three patients with systemic mastocytosis with interferon alpha-2b. Leuk. Lymphoma 22:501–8 Tefferi A, Li CY, Butterfield JH, et al. 2001. Treatment of systemic mast-cell disease with cladribine. N. Engl. J. Med. 344:307–9 Ronnov-Jessen D, Lovgreen Nielsen P, Horn T. 1991. Persistence of systemic mastocytosis after allogeneic bone marrow transplantation in spite of complete remission of the associated myelodysplastic syndrome. Bone Marrow Transplant 8:413–15 Fodinger M, Fritsch G, Winkler K, et al. 1994. Origin of human mast cells: development from transplanted hematopoietic stem cells after allogeneic bone marrow transplantation. Blood 84:2954–59 Przepiorka D, Giralt S, Khouri I, et al. 1998. Allogeneic marrow transplantation for myeloproliferative disorders other than chronic myelogenous leukemia: review of forty cases. Am. J. Hematol. 57:24–28 Nakamura R, Akin C, Bahceci E, et al. 2002. Allogeneic non-myeloablative stem cell transplantation for advanced systemic mastocytosis: possible induction of a graftversus mastocytosis effect. Biol. Blood Marrow Transplant. 8:81 (Abstr.) Buchdunger E, Zimmermann J, Mett H, et al. 1996. Inhibition of the Abl proteintyrosine kinase in vitro and in vivo by a 2-phenylaminopyrimidine derivative. Cancer Res. 56:100–4 Druker BJ, Tamura S, Buchdunger E, et al. 1996. Effects of a selective inhibitor of the Abl tyrosine kinase on the growth of Bcr-Abl positive cells. Nature Med. 2:561– 66 Buchdunger E, Cioffi CL, Law N, et al. 2000. Abl protein-tyrosine kinase inhibitor STI571 inhibits in vitro signal transduction mediated by c-kit and platelet-derived growth factor receptors. J. Pharmacol. Exp. Ther. 295:139–45 Ma Y, Zeng S, Metcalfe DD, et al.

12 Dec 2003

14:27

432

AR

AKIN

AR206-ME55-24.tex

¥

AR206-ME55-24.sgm

P1: GCE

METCALFE

2002. The c-KIT mutation causing human mastocytosis is resistant to STI571 and other KIT kinase inhibitors; kinases with enzymatic site mutations show different inhibitor sensitivity profiles than wild-type kinases and those with regulatory-type mutations. Blood 99:1741–44 55. Zermati Y, De Sepulveda P, Feger F, et al. 2003. Effect of tyrosine kinase inhibitor

Annu. Rev. Med. 2004.55:419-432. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

LaTeX2e(2002/01/18)

STI571 on the kinase activity of wild-type and various mutated c-kit receptors found in mast cell neoplasms. Oncogene 22:660– 64 56. Akin C, Brockow K, D’Ambrosio C, et al. 2003. Effects of tyrosine kinase inhibitor STI571 on human mast cells bearing wild-type or mutated c-kit. Exp. Hematol. 31:686–92

2 Dec 2003

21:36

AR

AR206-ME55-25.tex

AR206-ME55-25.sgm

LaTeX2e(2002/01/18) P1: GBC 10.1146/annurev.med.55.091902.104433

Annu. Rev. Med. 2004. 55:433–57 doi: 10.1146/annurev.med.55.091902.104433 c 2004 by Annual Reviews. All rights reserved Copyright ° First published online as a Review in Advance on Sept. 8, 2003

THE ERBB FAMILY: Targets for Therapeutic

Annu. Rev. Med. 2004.55:433-457. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Development Against Cancer and Therapeutic Strategies Using Monoclonal Antibodies and Tyrosine Kinase Inhibitors Eric K. Rowinsky Director, Institute for Drug Development of the Cancer Therapy and Research Foundation, San Antonio, Texas, and Clinical Professor of Medicine, Division of Medical Oncology at The University of Texas, Health Science Center at San Antonio, San Antonio, Texas; email: [email protected]

Key Words epidermal growth factor receptor, EGFR, gefitinib, erlotinib CI-1033, EKB-569, GW572016 ■ Abstract The overexpression and aberrant function of members of the erbB family of receptors, particularly erbB1 (also known as epidermal growth factor receptor), and its ligands in many human cancers have provided a rationale for targeting this signaling network with novel approaches. erbB1 is a selective target for inhibiting cancers because its activation often confers a proliferative advantage. Activation of the erbB1 tyrosine kinase provides signals that drive dysregulated proliferation, invasion, metastasis, angiogenesis, and cell survival, and its inhibition has potential in both the treatment and prevention of these malignancies. Based on the structure and function of erbB1, two therapeutic strategies have been developed. The first uses human monoclonal antibodies (MAbs) generated against the receptor’s ligand-binding extracellular domain. These MAbs block binding of receptor-activating ligands, and, in some cases, can induce receptor endocytosis and downregulation. The second uses small molecules that compete with adenosine triphosphate (ATP) for binding to the receptor’s kinase pocket, thereby blocking receptor activation and the transduction of postreceptor signals. Early clinical studies suggest that both approaches are well tolerated and can induce clinical activity in many common malignancies.

INTRODUCTION Cells are continuously exposed to diverse external stimuli, ranging from soluble endocrine and paracrine factors to signaling molecules on neighboring cells. The cell must interpret these extracellular signals to produce an appropriate developmental or proliferative response. Receptors of the tyrosine kinase (TK) family play principal roles in these processes, as they integrate a multitude of external 0066-4219/04/0218-0433$14.00

433

2 Dec 2003

21:36

Annu. Rev. Med. 2004.55:433-457. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

434

AR

AR206-ME55-25.tex

AR206-ME55-25.sgm

LaTeX2e(2002/01/18)

P1: GBC

ROWINSKY

stimuli with specific internal signals and responses, ultimately allowing the cell to respond correctly to its environment. This review focuses on one family of structurally related TK receptors, known as the erbB type 1 receptors [also known as the epidermal growth factor receptor (EGFR) family], which are critical for mediating the proliferation and differentiation of normal cells. Experimental data support the suggestion that aberrant activation of the kinase activity of these receptors is important in the development and/or progression of human cancer. This article reviews current knowledge of the erbB family and its ligands, particularly their roles in signal transduction and the malignant phenotype, providing support for erbB, particularly erbB1, as a critical target for therapeutic development against malignant diseases.

Evolution and Signaling Diversity of the erbB Family The erbBs were first implicated in cancer in the early 1980s, when the avian erythroblastosis tumor virus was found to encode an aberrant form of the human EGFR, or erbB1. Over the past several decades, four members of the erbB receptor family have been identified, and the physiologic function of these receptors and their ligands, the mediation of cell-to-cell interactions, and the consequences of erbB dysregulation are being appreciated, including their associations with the malignant process.

erbB Family Ligands Most ligands of erbB family receptors are synthesized as transmembrane precursors that can be proteolytically cleaved to release the soluble form of the peptide or can function as membrane-anchored ligands in juxtacrine signaling (1–5). The peptides share a domain of homology that encompasses ∼50 amino acids. The salient feature of this domain is the EGF-like or EGF-homologous region, which is required for erbB binding and activation. Expression and processing of ligand precursors are highly regulated, in part by extensive horizontal connections with other modulators and signaling systems. erbB ligands have been classified into three major groups based on their direct binding to a particular erbB family member (Figure 1). The first group consists of epidermal growth factor (EGF), transforming growth factor–alpha (TGF-α), and amphiregulin (also known as schwannoma-derived growth factor), which bind exclusively to erbB1. The membrane-bound forms of EGF and TGF-α may interact with receptors on the surface of adjacent cells, thereby potentially contributing to cell-to-cell adhesions and cell-to-cell stimulatory interactions. The second group of erbB ligands is represented by heparin-binding EGF, betacellulin, and epiregulin. Heparin-binding EGF and betacellulin bind and activate both erbB1 and erbB4, whereas epiregulin appears to be a more broad-spectrum erbB receptor ligand, binding all receptors except homodimers of erbB2 (6, 7). The third group of erbB ligands consists of a large and complex family of polypeptides called heregulins or neuregulins (NRGs) (also known as neu differentiation factors). Different NRG isoforms have varying

2 Dec 2003

21:36

AR

AR206-ME55-25.tex

AR206-ME55-25.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:433-457. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

erbB TARGETS FOR CANCER THERAPY

P1: GBC

435

Figure 1 Structure of erbB family receptors and their cognate ligands. The receptor consists of three domains: a ligand-binding extracellular domain containing two cysteine-rich regions (CR1 and CR2), a transmembrane domain, and an intracellular domain containing a tyrosine kinase region. (EGF, epidermal growth factor; EGFR, EGF receptor; HER, human epidermal receptor; HB-EGF, heparin-binding EGF; NRG, neuregulin; TGF-α, transforming growth factor-α)

affinities for different receptor heterodimers, resulting in distinct but overlapping patterns of biologic responses (5).

Ligand-Receptor Signaling Specificity Because many erbB ligands bind and activate the same receptor, functional redundancy, as well as specificity, is evident. TGF-α and EGF are almost indistinguishable in their ability to bind, activate, and downmodulate erbB1, but their biologic activities differ substantially. TGF-α is more potent than EGF as an angiogenic factor in vivo and in stimulating epidermal cell-colony formation in tissue culture (4, 5). Ligands such as EGF and NRG-4, which bind to erbB1 and erbB4, respectively, have narrow specificities, whereas others, such as epiregulin NRG-1β and betacellulin, bind to two distinct primary receptors. Overexpression of erbB2, which favors receptor heterodimer formation, can broaden ligand specificity, and ligands that are more efficient at recruiting this coreceptor can reduce the binding of less effective ligands. Splice variants of NRGs and various ligand-receptor complexes also differ in their ability to recruit any particular partner receptor, which affects their potency and the kinetics of signaling.

erbB RECEPTOR: STRUCTURE AND FUNCTION All erbB receptor proteins belong to subclass I of the superfamily of receptor TKs (RTKs), classified according to their sequence homology and domain organization. erbB receptors are expressed in a variety of tissues of epithelial,

2 Dec 2003

21:36

Annu. Rev. Med. 2004.55:433-457. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

436

AR

AR206-ME55-25.tex

AR206-ME55-25.sgm

LaTeX2e(2002/01/18)

P1: GBC

ROWINSKY

mesenchymal, and neuronal origin, where they play fundamental roles in critical developmental, proliferative, and differentiation processes (5–11). The erbB family consists of four closely related transmembrane receptors: erbB1 (also termed EGFR or HER1), erbB2 (also termed HER2 or Neu), erbB3 (also termed HER3), and erbB4 (also termed HER4). With few exceptions (e.g., hematopoietic cells), erbB receptors are expressed in cells of mesodermal and ectodermal origins. In epithelial tissues, the basolateral distribution of erbB family members enables them to mediate signals required for growth between mesenchymal and epithelial tissue components. All four erbB receptors share a common molecular architecture composed of three distinct regions: (a) an extracellular region consisting of four glycosylated domains, two of which are cysteine-rich; (b) a transmembrane domain containing a single hydrophobic anchor sequence; and (c) an intracellular region containing the catalytic TK domain, which is responsible for the generation and regulation of intracellular signaling (Figure 1) (5–11). The formation of erbB homodimers and heterodimers, following ligand binding and receptor aggregation, activates the intrinsic RTK activity via intramolecular phosphorylation and generates a cascade of downstream chemical reactions that transmit a wide variety of cellular effects (Figure 2).

erbB1 erbB1 is essential to the regulation of normal cell growth and differentiation, and its dysregulation confers a proliferative advantage and malignant potential. The receptor transmits growth regulatory signals, particularly upon binding of EGF or TGF-α. In fact, except for the NRGs, all EGF ligands are capable of binding to the erbB1 receptor and producing mitogenic effects on EGF-responsive cells. erbB1 expression, overexpression, or dysregulation may alter intracellular

Figure 2 Cellular responses associated with signaling through erbB family receptors.

2 Dec 2003

21:36

AR

AR206-ME55-25.tex

AR206-ME55-25.sgm

LaTeX2e(2002/01/18)

erbB TARGETS FOR CANCER THERAPY

P1: GBC

437

signaling along pathways such as the mitogen-activated protein kinase (MAPK) and phosphatidylinositide-3-kinase (PI3K) signaling pathways. When activated, these pathways translate proteins required for G1 to S phase traverse or phosphorylation of antiapoptotic proteins leading to cell survival, respectively (Figure 2) (5–11). erbB1 currently serves as a target for therapeutic development against malignant diseases due to its ubiquitous nature and principal function as a regulator of proliferative signals; however, it is important to consider that most malignant tumors also have altered expression of erbB family members.

Annu. Rev. Med. 2004.55:433-457. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

erbB2 The second member of the erbB receptor family to be discovered was erbB2 (HER2) which shows considerable homology to erbB1. Since erbB2 is a more potent oncoprotein than other members of the erbB family and has no known high-affinity ligands, its function is somewhat uncertain (7, 12). The discovery that erbB1 and erbB2 could form heterodimeric complexes raised the possibility that the interaction of ligands with erbB2 involved heterodimerization of erbB2 with other erbB receptors. Subsequent studies showed that erbB3 and erbB4 can also form heterodimers with erbB2. In the absence of a high-affinity ligand that directly binds to erbB2, it is likely that heterodimerization and transmodulation of other erbB receptors is the preferred initiating event for signaling (7, 9–11). There is increasing evidence that the principal function of erbB2 is as a coreceptor or dimerization partner for all other erbB family members and that it is important in the potentiation of erbB signaling.

erbB3 and erbB4 erbB3 (HER3) and erbB4 (HER4) are structurally related family members, although relatively little is known about their function (5–11). Interestingly, erbB3 lacks TK activity and is activated by TKs on other receptors. Heterodimers formed with erbB3 and erbB4 preferentially signal through the PI3K survival pathway relative to other types of heterodimers.

SIGNALING THROUGH erbB On ligand binding, cellular responses are elicited through multiple divergent pathways (5–11). Briefly, ligand binding leads to receptor aggregation, facilitating the formation of both erbB homodimers and heterodimers, which are capable, to varying degrees, of activating the intrinsic receptor TK activity via intermolecular phosphorylation within its cytoplasmic domain. The resultant phosphorylated tyrosine residues serve as, or modulate the readiness of, docking sites for downstream signaling molecules and cytoplasmic messenger proteins, which, in turn, initiate a cascade of signals that emanate from the cytoplasm to the nucleus. Key tyrosine phosphorylation sites responsible for recruitment of downstream receptor targets are located in the juxtamembrane region and C-terminal tail of the receptor,

2 Dec 2003

21:36

438

AR

AR206-ME55-25.tex

AR206-ME55-25.sgm

LaTeX2e(2002/01/18)

P1: GBC

ROWINSKY

which flank the TK domain. Signaling through erbB1 and other family members triggers a rich network of downstream cellular pathways, culminating in responses that range from cell division to cell death, motility to adhesion, invasiveness, and angiogenesis. Ultimately, downstream effects on gene expression determine the biologic response to receptor activation. Because the network is often dysregulated in cancers, a molecular understanding of these processes may lead to the development of therapeutics.

Annu. Rev. Med. 2004.55:433-457. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Signaling Diversity The ability of the erbB family to undergo homodimerization and heterodimerization, both constitutively and in response to their different ligands, provides enormous diversity to erbB-related signaling. Receptors that do not bind a particular ligand when expressed alone can be cross-activated if a binding-competent receptor is also present. For example, although EGF does not bind to or activate erbB2 by itself, EGF induces phosphorylation of tyrosine residues on both erbB1 and erbB2 in cells expressing both receptors (7, 9). The diversity and hierarchy of heterodimeric receptor interactions may be due to differences in the affinities of the various ligand-receptor-receptor complexes. Although 10 possible dimeric complexes can potentially form, each having variable potential to induce downstream signaling, there appears to be a graded ligand-dependent hierarchy for the formation of heterodimers, and those containing erbB2 are the most stable and preferred (7, 13, 14). erbB2, which does not have a direct high-affinity ligand, acts instead as a common receptor for other erbB family members. When erbB2 is overexpressed, heterodimers form preferentially. erbB2-containing heterodimers have features (e.g., slow ligand dissociation, relaxed ligand specificity, slow endocytosis, rapid recycling, prolonged firing) that prolong and enhance downstream signaling and its effects such as proliferation, migration, and resistance to apoptosis (7). The interactions of erbB family members are also characterized by directionality. For example, NRG-1 induces formation of erbB1-erbB3 heterodimers and erbB1-erbB4 heterodimers more readily than EGF does (7). The specificity, potency, and diversity of intracellular signals are determined, in part, by positive and negative effectors of erbB proteins, the identity of the ligand, dimer composition, and specific structural determinants of the receptors. However, the principal determinant is the vast array of phosphotyrosine-binding proteins that associate with structurally diverse C-terminal “downstream docking” tails of each erbB receptor after engagement into dimeric complexes. These critical sequences, which contain tyrosine residues that undergo phosphorylation on ligand binding and receptor dimerization, represent docking sites for various proteins involved in signal transduction (7, 10, 13–15). Docking sites are provided for proteins containing Src homology 2 or phosphotyrosine-binding domains, which recognize specific phosphotyrosine residues in the context of their surrounding amino acids. Each erbB receptor displays a distinct pattern of C-terminal autophosphorylation sites. At least for erbB2, which does not have a direct activating ligand, these

2 Dec 2003

21:36

AR

AR206-ME55-25.tex

AR206-ME55-25.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:433-457. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

erbB TARGETS FOR CANCER THERAPY

P1: GBC

439

phosphotyrosine-binding sites are essential for its transforming properties. There is a great deal of overlap among the signaling pathways activated by the four erbB receptors. For example, the MAPK pathway is an invariable target of all erbB family members. On the other hand, there are also specific examples of preferential modulation of specific pathways. This is illustrated by the presence of multiple binding sites for the regulatory subunit of PI3K on erbB3 and erbB4, which render these receptors the most efficient activators of the PI3K pathway (16). Simultaneous activation of cascades, such as the MAPK pathway, the stressactivated protein kinase cascade, protein kinase C, and the PI3K pathway, translates in the nucleus into distinct transcriptional programs, the culmination of which is the net cellular response.

Turning Off the Signal The principal process by which erbB signaling is turned off is ligand-mediated receptor endocytosis, and the kinetics of this process is often understated with regard to the overall magnitude of signaling (7, 17, 18). The kinetics of signal degradation are determined in part by the composition of the receptors. For erbB, ligand stimulation results in rapid endocytosis and degradation of both the receptor and ligand. Ligand binding induces receptor clustering in clathrin-coated pits on the cell surface, followed by endocytosis, migration to multivesicular bodies, and eventual lysosomal degradation. Degradation of erbB depends on TK activity, and kinase-negative receptor mutants generally recycle to the cell surface for reutilization. erbB1 is more prone to degradation via endosome formation and hydrolysis and is the only erbB receptor that can interact directly with c-Cbl, a ubiquitin ligase that targets the erbB to lysosomal degradation following ligandinduced receptor internalization, whereas the other erbB receptors are relatively endocytosis impaired and tend to be recycled back to the cell surface (7, 17–19). The rapid endocytosis and degradation of the activated erbB receptor attenuate the signal generated at the cell surface in response to growth-factor stimulation. The particular mode and site of degradation are also determined in part by the composition of the dimer. For example, erbB1 homodimers are processed primarily to the lysosome, erbB3 molecules are constitutively recycled, and heterodimerization with erbB2 decreases the rate of endocytosis and increases recycling of its partners (7, 17–19). erbB2 homodimers, which are stable in the endosomal vacuole, are rapidly tagged with ubiquitin and processed for digestion, resulting in weak signals, whereas erbB2 heterodimers are relatively unstable in the endosome, resulting in a lower rate of degradation and a higher rate of receptor recirculation (7, 17–21). To make matters even more complex, networks integrate heterologous signals from other networks. In the case of erbB, heterologous signals induced by hormones, neurotransmitters, lymphokines, and stress inducers are integrated into downstream messengers (7). These interactions are mediated by protein kinases that directly phosphorylate the erbB receptors, thereby affecting their kinase

2 Dec 2003

21:36

Annu. Rev. Med. 2004.55:433-457. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

440

AR

AR206-ME55-25.tex

AR206-ME55-25.sgm

LaTeX2e(2002/01/18)

P1: GBC

ROWINSKY

activity or endocytoic transport. One type of trans-regulatory mechanism involves the activation of G-protein–coupled receptors (GPCRs), such as those for lysophosphatidic acid, thrombin, and endothelin. Agonists of GPCRs may result in a net increase of phosphorylation of erbB1 and erbB2. By a poorly defined mechanism, these agonists can also activate matrix metalloproteinases, which then cleave membrane-tethered erbB ligands, such as heparin-binding EGF, thereby freeing them to bind to erbBs. Activation of GPCRs may also activate Src family kinases, leading to phosphorylation of tyrosine residues on the intracellular domains of ErbB. These activities can subsequently trigger events downstream of ErbB1, possibly contributing to the mitogenic potential of heterologous agonists. Furthermore, interconnections between other signaling pathways help to integrate and coordinate cellular responses to extracellular stimuli. The erbB family and related signaling network provide enormous signaling diversity at many levels, including ligand specificity, receptor partnering, scaffolding sites for effector signaling proteins, and substrate specificity for their kinase activities, receptor degradation, and integration of heterologous signals. Diversity among types of cells and tissues also exists, depending on the expression and preferred stoichiometry for interactions of the receptors and ligands. In sum, erbB receptors couple to specific downstream pathways with differing efficiencies, thereby affording an astonishing range of signaling possibilities. The particular cellular response to erbB stimulation is a function of the cellular context, as well as the specific ligand and erbB dimer. This has been shown best for mitogenic and transforming responses; homodimeric receptor combinations are less mitogenic and transforming than the corresponding heterodimeric combinations, and erbB2-containing heterodimers are the most potent complexes.

erbB in Cancer In many different cancer cell types, the erbB pathway becomes hyperactivated or dysregulated by several mechanisms, including overproduction of ligands, overproduction of receptors, or constitutive activation of receptors. The particular mechanism by which erbB becomes hyperactivated may be very important in determining overall prognosis and guiding specific treatment. erbB1 is expressed on normal cells at levels ranging from 20,000 to 200,000 receptors per cell (20, 21). However, receptor levels can be much higher in malignant cells. Amplification of growth-factor receptor genes is one mechanism by which tumor cells can increase the number of cell-surface receptors. erbB1 gene amplication has been observed in several cancers, such as brain malignancies and non–small-cell lung cancer (NSCLC) (7, 21). In some renal-cell carcinomas and other malignancies, erbB1 may also be overexpressed in the absence of gene amplification owing to a variety of mechanisms, including mutations that increase erbB1 transcription, mRNA translation, or stability of the protein. Many malignancies and cancer cell lines, especially carcinomas (Table 1), overexpress erbB1 (21). Increased erbB1 expression has also been reported in

erbB1

2 Dec 2003

21:36

AR

AR206-ME55-25.tex

AR206-ME55-25.sgm

LaTeX2e(2002/01/18)

P1: GBC

erbB TARGETS FOR CANCER THERAPY

441

TABLE 1 Malignancies overexpressing wild-type and mutated forms of erbB receptors erbB1

erbB2

Breast (14%–91%)

∗

Breast (10%–37%)

∗

∗

Ovary (30%–75%)

Annu. Rev. Med. 2004.55:433-457. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

∗

Ovary (20%–32%)

erbB3

erbB4

Breast

Breast

Ovary

Ovary

Renal (50%–90%)

Renal (24%–40%)

Renal

Lung (NSCLC) ∗ (40%–80%)

Lung (NSCLC) ∗ (3%–56%)

Lung (NSCLC)

Head and neck (squamous) Head and neck ∗ (30%–75%) (squamous) ∗ (32%–62%) Colorectal (25%–77%)

Colorectal (7%)

Head and neck ∗ (squamous)

∗

∗

Lung (NSCLC) Head and neck ∗ (squamous)

∗

∗

Pancreas (30%–50%)

Pancreas

∗

Glioma (40%–50%)

Bladder (31%–48%)

∗

∗

Esophagus

Bladder (7%–36%) Esophagus (13%–73%)

∗

Stomach

Stomach (5%–55%) ∗

Prostate

Prostate

Melanoma

Melanoma

Thyroid ∗

Endometrial

Endometrial

∗

Prostate

Prostate

Melanoma

Melanoma

Thyroid

Thyroid

∗

Skin (squamous cell)

Skin (squamous)

Lung (small cell)

Lung (small cell)

Cervical

∗

Sarcomas Chronic myelogenous leukemia ∗

Clinical studies have linked overexpression and/or mutation of this erbB receptor to a worse prognosis.

melanoma and meningioma. The level of erbB1 expression and activity has been shown to vary widely within a tumor type, but this variability may be due, in part, to differences in detection methods. Table 2 illustrates the range of erbB1 overexpression observed in different studies for a given tumor type. erbB1 overexpression is associated with a higher grade, higher proliferation, or reduced survival in a variety of cancers including NSCLC, squamous-cell carcinoma of the head and neck (SCCHN), and carcinomas of the ovary and breast, among others. However, the data regarding the predictive value of erbB1 expression within a given tumor type are conflicting. Several variants of the erbB1 receptor have been identified in many malignancies. The most common variant is the mutated erbB1 receptor EGFRvIII, which

2 Dec 2003

21:36

442

AR

AR206-ME55-25.tex

AR206-ME55-25.sgm

LaTeX2e(2002/01/18)

P1: GBC

ROWINSKY

TABLE 2 Selected antibodies to erbB1 in clinical development

Annu. Rev. Med. 2004.55:433-457. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Agent (developer)

Class

Phase of development

Target

Pertinent clinical results

Cetuximab (ImClone Systems)

Human recombinant antibody (chimera)

III

Binds to external domains of erbB1

Nonlinear pharmacokinetics with erbB1 saturation at doses >200 mg/m2; T1/2 in plasma ≈7 days Major toxicities: folliculitis and/or acneiform rash (80%); severe allergic reactions (4%) Human antichimeric antibodies, which are occasionally neutralizing, reported in 3% of patients Early development in combination with chemotherapy or radiation Colorectal cancer (irinotecan-refractory): combined with irinotecan, 22.5% major response rate; single agent, 10% major response rate Pancreatic cancer (untreated): combined with gemcitabine, 12% major response rate Phase III studies in head and neck and colorectal cancers in progress

EMD72000 (Merck KGa)

Humanized recombinant IgG1κ antibody

I/II

Binds to the external domain of erbB1

Major toxicity: rash Major activity in colorectal cancer in phase I

ABX-EGF (Abgenix)

Human IgG2κ antibody (fully human)

II

Binds to external domains of erbB1

Major toxicity: rash At the recommended phase II dose (2.5 mg/kg/week), receptor-mediated clearance is saturated and rash occurs in 100% of patients Major activity in colorectal (10% major response rate) and renal cancers Phase II studies in prostate, renal, colorectal, and lung cancers in progress

MDX-447 (Medarex)

Bispecific monoclonal antibody

II

Binds to both erbB1 and CD64 on neutrophils and monocytes

Evidence of immunological activity, skin toxicity, and biological responsiveness in phase I Phase II studies in progress

is caused by deletion of exons 2 to 7 and subsequent loss of amino acid residues 6 through 276 in the extracellular domain. EGFRvIII is not found in normal tissues but is expressed on the cell membrane in certain tumors, including ∼50% of gliomas. EGFRvIII has also been detected in medulloblastoma, in carcinomas of the prostate, breast, ovary, and stomach, and in NSCLC in varying frequencies, suggesting broad clinical relevance (21). This variant possesses a constitutively activated TK that may result in ligand-independent transformation of cell lines, although the mutation results in the deletion of a part of the extracellular domain that renders the receptor incapable of ligand binding and dimerization (20). Specific monoclonal antibodies (MAbs) have been isolated that can help to detect this

2 Dec 2003

21:36

AR

AR206-ME55-25.tex

AR206-ME55-25.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:433-457. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

erbB TARGETS FOR CANCER THERAPY

P1: GBC

443

variant on tumor cells using immunohistochemistry. The use of such MAbs permits identification of EGFRvIII in certain tumors that may not bind MAbs against wildtype erbB (22, 23). The tumor-specific expression of this variant, coupled with the rapid internalization of EGFRvIII-MAb complexes, suggests that anti-EGFRvIII– targeted therapy using MAbs or immunoconjugates may be useful against tumors expressing this variant (23). Although both overexpression and structural alterations of erbB1 occur commonly in human malignancies, particularly carcinomas, in vitro studies suggest that overexpression of the normal receptor leads to transformation only in the presence of a ligand, and expression of EGF-like ligands often accompanies erbB1 overexpression in primary tumors. Autocrine stimulation of the erbB1 receptor by production of EGF and TGF-α from the tumor cells can complete the feedback loop by binding to its own receptors. Therefore, tumor cells may require a functional autocrine loop for continued survival rather than increased expression of erbB1 alone. In order to metastasize, tumor cells must complete several essential processes. These include stromal and vascular invasion, embolization, survival in the circulation, arrest in a distant capillary bed, and extravasation into and multiplication in organ paranchyma. erbB1 has been implicated in several pathways that affect tumor-cell survival and apoptosis, angiogenesis, motility, and invasion. Therefore, inhibition of EGFR activity could affect multiple aspects of tumor growth, progression, and metastasis. Both in vitro and animal studies have clearly indicated that erbB2 overexpression plays a pivotal role in oncogenic transformation and tumorigenesis (24, 25). Transfection of the erbB2 gene into human breast and ovarian tumor cell lines increases DNA synthesis, tumorigenicity, and metastatic potential. Furthermore, the growth of tumors and human breast cancer cell lines overexpressing the erbB2 receptor is inhibited by MAbs directed at the receptor. erbB2 amplification/ overexpression has been detected in subsets of a wide range of human cancers. When the erbB2 gene is amplified and the erbB2 receptor overexpressed, it is very likely that this overexpression contributes significantly to tumor development or progression. When only elevated levels of erbB2 protein are found with no erbB2 gene amplification, it is not clear whether these levels of protein are involved in the development or progression of the tumor. It is important to recognize the importance of detection methods and their interpretation in determining the percentage of tumors scored as positive for erbB2 protein expression. The association of erbB2 expression with cancer has been demonstrated best in breast cancer, in which overexpression correlates with tumor size, spread of the tumor to lymph nodes, high grade, high percentage of S-phase cells, aneuploidy, and lack of steroid hormone receptors. Evidentally erbB2 confers a strong proliferative advantage to tumor cells (24, 25). Other malignancies in which erbB2 gene amplification and/or protein overexpression are found include ovarian, endometrial, cervical, gastric, colorectal, bladder, kidney, pancreatic, and thyroid

erbB2

2 Dec 2003

21:36

Annu. Rev. Med. 2004.55:433-457. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

444

AR

AR206-ME55-25.tex

AR206-ME55-25.sgm

LaTeX2e(2002/01/18)

P1: GBC

ROWINSKY

carcinomas, as well as SCCHN and mesenchymal malignancies. Paradoxically, a higher degree of erbB2 overexpression is reported in early forms of breast cancer (such as ductal carcinoma in situ) than in more advanced invasive carcinomas, suggesting that alterations in erbB2 alone are insufficient for breast tumor progression from a relatively benign to a more malignant phenotype (24, 25). The structure of overexpressed erbB2 protein is the same as that found in normal cells; no mutational change in the erbB2 gene product has been identified in human cancers (26). Overexpression of erbB2 on the cell surface appears to lead to constitutive activation of erbB2 homodimers without the need for ligand binding, resulting in unregulated cell growth and oncogenic transformation. Amplification of the erbB2 gene is a prerequisite of erbB2 protein expression in most tumors (24, 27). However, the erbB2 gene is not amplified in a small percentage of tumors that overexpress erbB2 mRNA or protein. In these rarer cases, erbB2 protein overexpression may result from transcriptional or post-transcriptional dysregulation. The identification of erb2 amplification by fluorescence in situ hybridization (FISH) has now received regulatory approved in the United States and elsewhere to predict which breast cancer patients are at high risk for recurrence and disease-related death following definitive local treatment. Many investigations are attempting to relate erbB2 amplification status (as determined by FISH) to benefit from therapies targeting erbB2. Accumulating data indeed indicate that erbB2 gene amplification by FISH can identify patients who might benefit from more aggressive therapy (7). erbB3 AND erbB4 The catalytically inactive member of the erbB family, erbB3, is expressed in several epithelial cancers, but there is no evidence for gene amplification, and overexpression is limited. Coexpression of erbB3 and erbB4 with other erbB family members appears to improve the predictive power for prognosticating the overall course for certain tumor types; however, the roles of these receptors in cancer are not well understood.

RATIONALE AND TREATMENT STRATEGIES FOR TARGETING THE erbB1 The development of therapeutics targeting erbB2, particularly trastuzimab (Herceptin®, Genentech Inc., South San Francisco, CA), has been reviewed elsewhere (7). This section is limited to a discussion of therapeutics targeting erbB1. The coexpression of erbB and ligands at tumor sites allows erbB activation via autocrine/paracrine mechanisms. In support of the operational nature of these signaling pathways in erbB-expressing tumor cells, interruption of signaling with various erbB inhibitors has been shown to inhibit tumor proliferation in vitro and in vivo (28–33). These observations, coupled with (a) the ability to identify erbBexpressing human tumors from patients, (b) the association of erbB expression, particularly erbB1, with poor prognosis, and (c) the lack of a critical physiologic role of erbB (specifically erbB1) in healthy adults, have all suggested that this

2 Dec 2003

21:36

AR

AR206-ME55-25.tex

AR206-ME55-25.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:433-457. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

erbB TARGETS FOR CANCER THERAPY

P1: GBC

445

network is an ideal target for novel therapeutic strategies. Over the past decade, it has been recognized that antibodies and small molecules (drugs) can be used to perturb erbB1 signaling at the cellular level, thereby inhibiting cell growth and promoting cell death. Because erbB1 is an integral component of critical signaling pathways involved in regulating tumor growth, a rational approach to cancer therapy is to block the function of the receptor and thus inhibit cell proliferation and tumor progression. Strategies to target erbB1 for anticancer therapy include the following: (a) MAbs directed against the erbB1 receptor to block binding of EGF and TGF-α activated cell growth; (b) synthetic erbB1 TK inhibitors that act directly on the cytoplasmic TK domain of the receptor, preventing signal transduction and cell proliferation; and (c) ligand conjugates that bind specifically to erbB1 and deliver a lethal payload following ligand-toxin internalization (21). This review discusses the first two approaches, which are more developed at present.

Monoclonal Antibodies HUMANIZED ANTIBODIES A number of MAbs recognize the extracellular domain of erbBs, particularly erbB1 and erbB2. Table 2 summarizes relevant features of selected MAbs targeting erbB1. Those directed against erbB1 compete for ligand binding, induce erbB dimerization and internalization, and inhibit ligandstimulated TK activity, downstream signaling, and tumor growth (28–35). erbB is then degraded and downregulated. Many anti-erbB1 MAbs are more effective in vivo than in vitro, possibly because of the induction of antiangiogenic activity and/or enhancement of immune effector activity. The ability to induce receptor dimerization and downregulation from the cell surface and to block the receptor’s catalytic function has been best characterized for the 528 mouse IgG2a and the 225 mouse IgG1, as well as the 225 humanized MAb, which led to the development of cetuximab (36). In erbB1-dependent tumor cells, this approach inhibits erbB1 signaling, leading to cell-cycle arrest and/or cell death (37, 38). In addition to blocking autocrine erbB1 signaling, it has been proposed that erB1-targeting MAbs may recruit Fc receptor-expressing immune effector cells, leading to antibodydependent cellular cytotoxicity and tumor eradication. In one study, less complete inhibition of A431 tumor growth was observed with 225 F(ab0 )2 than with the bivalent 225 MAb, which suggests that, in addition to kinase blockade, immune mechanisms contribute to the antitumor activity of intact 225 MAb. The results of a phase I trial with the erbB1 mouse MAb 225 MAb showed selective antibody localization in NSCLC that had not been prescreened for erbB1, suggesting that the differential expression of erbB1 in tumor versus normal tissues can provide a therapeutic window for cancer syndromes with a high prevalence of detectable receptor expression (21, 28–33, 39–42). Cetuximab (IMC-C225), a chimeric humanized version of 225 MAb (Erbitux®, ImClone Systems Inc., New York, NY) was generated to avoid the host’s immune response against mouse antibodies. Interestingly, cetuximab’s binding affinity (Kd = 1–2 × 10–10 M) is approximately tenfold greater than those of the natural ligands (EGF and TGF-α) and the parental murine MAb (21). Cetuximab binding blocks EGF-induced activation, autophosphorylation, and

2 Dec 2003

21:36

Annu. Rev. Med. 2004.55:433-457. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

446

AR

AR206-ME55-25.tex

AR206-ME55-25.sgm

LaTeX2e(2002/01/18)

P1: GBC

ROWINSKY

internalization of erbB1. The MAb arrests cells in the G1 phase of the cell cycle, which is associated with increased levels of p27Kip1 (21). Cetuximab and similar MAbs, such as EMD72000 (Merck KGA, Darmstadt, Germany), also inhibit the growth of erbB1-expressing cancers in vitro, reduce tumor volume, and/or increase survival in mice with a wide range of erbB1-expressing human cancers. These effects have also been associated with apoptosis. Furthermore, these MAbs enhance the effects of radiation and various chemotherapy agents in vitro and in vivo. These agents include platinating (e.g., cisplatin), DNA-intercalating (e.g., doxorubicin), and antimicrotubule (e.g., paclitaxel, docetaxel) agents (21, 28–42). It has been proposed that in order to produce a significant antitumor response, erbB1 must be saturated with the MAb. Receptor saturation has generally been assumed to be achieved in vivo when the systemic elimination pathways for the MAb are saturated, at which point internalization of the MAb-erbB1 complexes becomes the principal mechanism of MAb clearance. The results of phase I studies indicate that cetuximab pharmacokinetics are nonlinear and saturation of elimination pathways occurs at intravenous doses of 200 and 400 mg/m2. The higher dose was associated with zero-order clearance during the first 96 h of the infusion and the half-life of the MAb was estimated to be 7 days (21). Cetuximab has principally been developed for weekly administration as a component of multi-agent regimens, since preclinical studies indicated that the predominant antitumor effect of therapeutics targeting erbB1 is delayed tumor growth, which may not impact the disease course of patients with advanced malignancies as profoundly as tumor regression does. The most common toxicity has been an acneiform rash or folliculitis involving the face, upper chest, and back, which occurred in 80% of patients and probably relates to the prominent roles of erbB1 and EGF-like ligands in epidermal tissues (27–42). Severe allergic reactions occurred in ∼4% of patients, with most occurring within minutes of the first infusion. Approximately 3% of patients had detectable human antichimeric antibodies, which were occasionally neutralizing (21). Overall, there was little evidence of significant production of human antichimeric antibodies in response to cetuximab treatment, and such antibodies have not precluded repetitive treatment. Prominent antitumor activity has been noted in patients with several types of advanced malignancies in phase I and II trials of cetuximab combined with other therapeutic modalities, as well as in limited studies of cetuximab as a single agent in colorectal cancer, as outlined in Table 1 (37–42). HUMAN ANTIBODIES A fully humanized IgG2κ Mab specific to erbB1 (ABXEGF, Abgenix, Inc., Freemont, CA) has also been generated using a XenoMouse® technology, in which human immunoglobulin genes were introduced into mice engineered to lack functional mouse immunoglobulin genes. Unlike humanized Mabs—which are constructed by implanting the complementary-determining regions of the mouse antibody into the human immunoglobulin framework, still contain ∼5%–10% mouse protein sequences, and may still be immunogenic— this human MAb does not contain murine protein sequences (43, 44). The full

2 Dec 2003

21:36

AR

AR206-ME55-25.tex

AR206-ME55-25.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:433-457. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

erbB TARGETS FOR CANCER THERAPY

P1: GBC

447

humanization would be expected to result in no immunogenicity and a slower clearance rate of the MAb as compared with mouse or mouse-derived MAbs, thus allowing repeated antibody administration in immunocompetent patients. The ABX-EGF MAb binds erbB1 with high affinity (Kd = 5 × 10−11 M), thereby blocking binding of both EGF and TGF-α, and inhibits EGF-dependent tumor-cell activation and proliferation (44, 45). ABX-EGF completely prevents the formation of human epidermoid carcinoma A431 xenografts in athymic mice (43, 44). More importantly, ABX-EGF without concomitant chemotherapy completely eradicates well-established A431 erbB1-overexpressing tumors, which do not recur. Human pancreatic, renal, breast, and prostate tumor xenografts, which express different levels of erbB1, were also inhibited by ABX-EGF treatment. The number of erbB1 molecules per cell correlated with the degree of tumor growth inhibition; MAb had no effect on tumors that did not express erbB1. Preliminary results of early clinical evaluations of ABX-EGF administered weekly in patients with various malignancies likely to express erbB1 indicate that the ABX-EGF is well tolerated at doses predicted to induce antitumor activity based on modeling of preclinical data (45, 46). The predominant toxicity is a transient acneiform rash and human antihuman antibodies have not been detected. Diseasedirected evaluations have recently begun in patients with advanced NSCLC and renal, prostate, and colorectal cancers, and intriguing activity has been observed to date in the most mature studies in renal and colorectal cancer patients (45–47). BISPECIFIC ANTIBODIES Bispecific antibodies, largely directed to erbB1, have dual specificity because they have two different antibody-binding regions. One is specific for the erbB1 receptor and the other is engineered to bind to an immunologic effector cell (48). The result is an antibody that binds to erbB (preventing signaling and related effects) and also enhances the host’s cytotoxic effector mechanisms. Antibodies specific for erbB1 and the immunoglobulin receptor CD64, which is found on monocytes and neutrophils, include MDX-447 (Medarex, Inc., Princeton, NJ). Both bispecific antibodies have been shown to reduce tumor growth and enhance immunologically mediated cytotoxicity. The results of phase I evaluations show immunologic activity, good tolerability, and biologic responses, and phase II disease-directed studies are ongoing (48).

SMALL MOLECULES TARGETING erbB1 TYROSINE KINASE Another approach to inhibiting erbB1 involves small molecules designed to inhibit RTK activity, erbB phosphorylation, and critical signaling downstream (21, 28, 33, 49, 50). Hypothetically, this approach could inhibit signaling mediated by ligands, as well as signaling that is independent of growth factors. In contrast to MAbs, such agents may also inhibit ligand-independent signaling due to constitutively active mutant receptors (e.g., EGFRvIII), but the relative merits of these modalities

2 Dec 2003

21:36

Annu. Rev. Med. 2004.55:433-457. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

448

AR

AR206-ME55-25.tex

AR206-ME55-25.sgm

LaTeX2e(2002/01/18)

P1: GBC

ROWINSKY

for inhibiting such mutant receptors are not clear. This strategy has involved the random screening from natural or synthetic compound libraries of small molecules with molecular weights of 300–400 Da that compete with the Mg-ATP binding site of the catalytic domain of the erbB1-TK (21, 28–33, 49, 50). The anilinoquinazolines have clearly set the boundaries for defining potency and specificity of inhibitors of erbB1 receptor TK (49, 50). Members of this class have demonstrated potencies as low as 6 ρM against the erbB1 TK, with almost total specificity relative to other receptor and intracellular TKs, as well as potent, specific inhibition of all EGF-mediated processes in viable cells. More recently, the quinazoline series of inhibitors has been expanded to include fused tricyclic derivatives, such as imidazoloquinazolines, pyrroloquinazolines, and pyrazoloquinazolines. These compounds have IC50 values against the isolated enzyme that are in the picomolar and low nanomolar ranges for inhibition of EGF-mediated receptor autophosphorylation. Tables 3 and 4 list several erbB1 TK inhibitors in clinical development, and this section discusses the salient preclinical and clinical characteristics of several of these compounds. The anilinoquinazolines, gefitinib (ZD1839; Iressa®; AstraZeneca, Wilmington, DE) and erlotinib (OSI-774; Tarceva®; OSI Pharmaceuticals, Uniondale, NY) are in the most advanced stages of clinical development. Both compounds inhibit the purified erbB1-TK enzyme in vitro with IC50 values in the low nanomolar range (51–54). Because of the high intracellular concentrations of ATP, much higher concentrations of the reversible erbB1-TK inhibitors are required to continuously block erbB1 phosphorylation in intact cells than in acellular experimental systems, which served, in part, as the rationale for developing of irreversible TK inhibitors such as CI-1033 (Pfizer, Inc., Groton, CT), EKB-569 (Wyeth-Ayerst, Philadelphia, PA) as these compounds bind irreversibly to the receptor, competing effectively with high concentrations of ATP for receptor binding. The ∼80% homology between the erbB1 and erbB2 TKs has allowed the generation of the RTK inhibitors, CI-1033, EKB-569, and GW572016 (Glaxo SmithKline, Philadelphia, PA), which inhibit multiple erbB receptor families (21, 55–57). This review does not discuss specific small-molecule inhibitors of erbB2 TK, which are also under development.

Gefitinib Gefitinib (ZD-1839; Iressa®) was initially shown to selectively inhibit erbB1 TK activity and EGF-dependent proliferation of KB SCCHN cells with IC50 values of 23 and 80 nM, respectively (Table 3) (22, 29–37, 52–54). However, it was at least 100-fold less active at inhibiting other kinases. The agent exerts various effects on tumor cells that express erbB1, such as blocking receptor autophosphorylation, inducing cell-cycle arrest, and reducing cell proliferation. Tumor growth inhibition is associated with a dose- and time-dependent upregulation of the cyclindependent kinase inhibitor p27Kip1, which may account for cell-cycle arrest in the G1 phase. Gefitinib also inhibits angiogenesis. Tumor growth inhibition has been the predominant therapeutic effect of gefitinib in vitro and in vivo, but DNA fragmentation indicative of apoptosis and tumor regression has been noted in several

2 Dec 2003

21:36

AR

AR206-ME55-25.tex

AR206-ME55-25.sgm

LaTeX2e(2002/01/18)

P1: GBC

erbB TARGETS FOR CANCER THERAPY

449

TABLE 3 Selected small-molecule reversible inhibitors of erbB receptor tyrosine kinase activity in clinical development

Annu. Rev. Med. 2004.55:433-457. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Agent (developer)

Chemical class

Target and mechanism of action

Pertinent clinical results

Gefitinib (ZD1839; AstraZeneca)

Quinazoline

Competitively inhibits ATP binding in the internal TK domain of erbB1 IC50 against erbB1 TK activity in vitro is 0.020 µM

Phase I studies evaluated both intermittent (once-daily for 14 days every 28 days) and continuous (once daily, uninterrupted treatment) schedules Major toxicities: rash, diarrhea Phase II studies in previously treated patients with NSCLC: • Iressa Dose Evaluation in Advanced Lung Cancer (IDEAL)-1 Study (Japan/Europe; no screening for erbB1): 250 vs. 500 mg/day, major response rate, 18.4% vs. 19% • IDEAL 2 (USA; no screening for erbB1): 250 vs. 500 mg/day, major response rate, 12% vs. 9%. Stable disease in 31% vs. 27%. Lung cancer symptoms respond in 95% and 71% of patients who had major responses and stable disease, respectively. Accelerated regulatory approval in May 2003 for patients with advanced NSCLC following failure of platinum- and docetaxel-based therapies Phase III study (no screening for erbB1): • Iressa NSCLC Trial Assessing Combination Treatment (INTACT)–1 and -2: Chemotherapy, consisting of paclitaxel/carboplatin or gemcitabine/cisplatin, plus placebo or gefitinib 250 or 500 mg/day; combined therapy for 6 courses followed by maintenance gefitinib or placebo: no differences in response rates, progression-free survival, or overall survival

Erlotinib (OSI-774; OSIP/ Genentech/ Roche)

Quinazoline

Competitively inhibits ATP binding to the ATP binding site in the internal TK domain of erbB1 IC50 against erbB1 TK activity in vitro is 0.002 µM

Phase I studies evaluated continuous (once-daily) dosing schedule Major toxicities: rash and diarrhea Phase II studies: activity noted in previously treated NSCLC, ovarian, and head and neck cancers, with major responses in 11%, 10%, and 13%, respectively Phase III studies in NSCLC and pancreatic cancers, as well as broad phase II evaluations, are ongoing Major responses in brain tumors in phase I studies

GW572016 (GlaxoSmithKline)

Quinazoline

Competitively inhibits ATP binding to the ATP binding site of the internal TK domains of erbB1 and erbB2 More potent erbB1 inhibition

Phase I studies evaluated continuous (once daily uninterrupted) schedule Major responses in breast cancer in phase I studies Major toxicities: rash, nausea and vomiting, diarrhea Phase II evaluations ongoing

2 Dec 2003

21:36

450

AR

AR206-ME55-25.tex

AR206-ME55-25.sgm

LaTeX2e(2002/01/18)

P1: GBC

ROWINSKY

TABLE 4 Selected small-molecule irreversible inhibitors of erbB receptor tyrosine kinase activity in clinical development

Annu. Rev. Med. 2004.55:433-457. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Agent Chemical (developer) class

Target and mechanism of action

Pertinent clinical results

CI-1033 (Pfizer)

Quinazoline

Irreversibly binds to cysteine residues in the ATP binding site of the internal TK domains of erbB1, erbB2, and erbB4 and prevents transmodulation of erbB3 Particularly potent against erbB1 and erbB2 TK

Phase I studies evaluated various once-daily intermittent and continuous treatment schedules Major toxicities: rash, diarrhea, nausea, vomiting. Also, dose-related myelosuppression and hypersensitivity reactions in studies evaluating intermittent treatment (e.g., once weekly) T1/2 = 4–6 h Phase II studies ongoing in ovarian, breast, and non–small-cell lung cancers

EKB-569 (Wyeth Ayrest)

Quinazoline

Irreversibly binds to cysteine residues in the ATP binding site of the internal TK domains of erbB1 and erbB2 IC50 = 1.33 nM against erbB1 TK in vitro and 15 nM against cells that overexpress erbB1 Striking protection against polyp formation in APCMin/+ mice when administered with sulindac

Phase I studies evaluated once-daily continuous (for 21 days) schedule Major toxicities: skin rash, diarrhea, mucositis, nausea T1/2 = 20–24 h Development in colorectal and pancreatic cancers as a single agent and in combinations

erbB1-overexpressing tumors. Although some tumor xenograft studies in animals have reported tumor regression (especially for erbB1-overexpressing tumors), the more usual finding has been dose-dependent inhibition of tumor growth, and tumors have resumed rates of growth comparable to those of controls following treatment. Combinations of gefitinib and various anticancer chemotherapeutics or radiation have generally shown enhanced activity compared with single-agent treatment against human tumor xenografts, even in tumors without high levels of erbB1 (21, 51–53). There is also ample experimental evidence that the autocrine activation of erbB1 signaling is important in breast cancer cells with acquired resistance to tamoxifen. Breast cancer cell lines with acquired tamoxifen resistance and impressive sensitivity to gefitinib have also been described, and these cells are more sensitive to gefitinib than tamoxifen-sensitive parental cells (41, 42). Because erbB1 and other erbB family members form heterodimers, particularly with erbB2, resulting in receptor transmodulation, it is likely that gefitinib also interferes with erbB1-driven transactivation of other erbB receptor types. Furthermore, low gefitinib concentrations are effective at inhibiting growth of erbB2-overexpressing breast cancer cells that also express erbB1 (41, 42). In fact, gefitinib exhibited greater growth

2 Dec 2003

21:36

AR

AR206-ME55-25.tex

AR206-ME55-25.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:433-457. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

erbB TARGETS FOR CANCER THERAPY

P1: GBC

451

inhibition than the erbB2-targeted MAb trastuzumab, and the agent, at higher concentrations, was effective against trastuzumab-resistant erbB2 breast cancer cells. Such results provide a rationale for the combined use of therapeutics targeting both erbB1 and erbB2. Mounting evidence suggests that gefitinib and other erbB1-targeted therapeutics may be useful in cancer prevention (58). Gefitinib, administered on an oral, once-daily, uninterrupted schedule, is undergoing clinical evaluations (Table 3). Most studies to date have been conducted in patients whose tumors have not been screened for erbB1 or potential determinants of response. Antitumor activity has been noted in patients with NSCLC, as well as brain, breast, head and neck, and prostate cancers. In May 2003, gefitinib received accelerated regulatory approval in the United States as monotherapy treatment (250 mg/day) for patients with locally advanced or metastatic NSCLC after failure of both platinum-based and docetaxel chemotherapy.

Erlotinib Erlotinib (OSI-774; Tarceva®) is a reversible, ATP-competitive inhibitor of the erbB1 TK, with an IC50 value of ∼2 nM and >1000 times the selectivity of pp60c-src, pp145c-abl, insulin receptor, and insulin-like growth factor–1 receptor TKs (21, 28–32, 54). The IC50 for inhibition of EGF-mediated receptor autophosphorylation of HN5 human head and neck cancer xenografts was 20 nM, and EGF-dependent mitogenesis and proliferation was reduced by 50% at comparable concentrations. Similar results were obtained in the DiFi colorectal carcinoma and the MDA-MB-468 breast carcinoma cell line. Erlotinib can induce cell-cycle arrest with an accumulation of cells in G0/G1, loss of the hyperphosphorylated form of the retinoblastoma protein, and accumulation of p27kip1. The agent also induces apoptosis, but higher drug concentrations are generally required. The combination of apoptosis and cell-cycle arrest lends further support to the concept that programmed cell death and differentiation may be the principal cellular mechanisms by which erbB-targeted TK therapeutics function. Erlotinib showed impressive antitumor activity both in vitro and in vivo. In preclinical studies, OSI-774 was well tolerated at doses of 10–200 mg/kg daily, had good oral bioavailability, and had a pharmacokinetic profile that was conducive to once-daily administration (54). Erlotinib showed significant in vivo activity: An oral dose of 10 mg/kg reduced HN5 xenografts by 50%, and a 50-mg/kg dose nearly achieved tumor stasis. Similar results were obtained against erbB1-overexpressing A431 xenografts. In clinical trials to date, in which the agent is administered once daily on an uninterrupted schedule, antitumor activity is observed in patients with NSCLC and cancers of the breast, head and neck, ovary, and brain (21).

GW572016, a Combined erbB1 and erbB2 TK Inhibitor GW572016, a 6-thiazolyquinazoline that reversibly inhibits the phosphorylation of erbB1 and erbB2 and downstream MAPK in a dose-dependent manner, has

2 Dec 2003

21:36

452

AR

AR206-ME55-25.tex

AR206-ME55-25.sgm

LaTeX2e(2002/01/18)

P1: GBC

ROWINSKY

Annu. Rev. Med. 2004.55:433-457. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

demonstrated potent tumor growth inhibitory activity in vitro (IC50 values below 0.15 µM) and appears selective for tumor cells relative to normal cells (56). In vivo studies showed antitumor activity against erbB1-overexpressing head and neck carcinoma and erbB2-overexpressing breast carcinoma xenografts. Clinical evaluations of GW572016 administered on a once-daily uninterrupted schedule are ongoing in tumor types such as breast and colorectal cancers, which are likely to express both erbB1 and erbB2. Preliminary activity has been reported in patients with advanced breast cancer (60).

Irreversible Inhibitors of Multiple erbB Receptor Tyrosine Kinases EKB-569 and CI-1033 are composed of chemical moieties that form covalent bonds within the RTK domain, resulting in irreversible receptor binding and sustained TK inhibition in vitro. This feature may also circumvent drug binding competition due to high intracellular ATP concentrations. However, the rates of receptor turnover and drug clearance in vivo are probably important factors to consider when comparing the merits of reversible and irreversible inhibitors of RTK. The optimal use of reversible erbB TK inhibitors mandates relevant plasma concentrations and/or agents with relatively long half-lives to keep the target suppressed, whereas the use of irreversible compounds would require that plasma concentrations be attained only long enough to briefly expose the receptors to the drug, which would then permanently suppress kinase activity. Like cetuximab, the erbB1-selective, reversible inhibitors appear to induce regression of tumors with either high or low erbB1 expression, suggesting that high erbB1 levels are not necessarily a predictor of tumor response to erbB1-targeting therapeutics. CI-1033 (Pfizer, Inc., Groton, CT) is a 4-anilinoquinazoline that irreversibly binds within the ATP-binding pocket of erbB TK and inhibits both activation and downstream signaling emanating from erbB1, erbB2, erbB3, and erbB4 (21, 49, 50, 55). CI-1033 binds irreversibly with high affinity to all erbB family members, particularly erbB1 and erbB2. It inhibits isolated erbB1 TK activity with an IC50 value of 1.5 nM and inhibits heregulin-mediated tyrosine phosphorylation in MDA-MB-453 human breast carcinoma, which expresses erbB2, erbB3, and erbB4, with an IC50 value of 9 nM (50, 55). CI-1033 has been shown to inhibit erbB1 phosphorylation in A431 carcinoma and MDA-MB-468 human breast cancer cells, and it inhibits the growth of several human xenografts (50, 55). It also induces regression of well-established A431 tumors. The results of studies of long-term drug administration indicate that CI-1033 maintains tumor growth suppression for extended time periods without the emergence of drug resistance (49, 50, 55). The activity of the compound appears to be independent of dose fractionation; significant activity is obtained on dosing regimens ranging from once daily to once weekly (50, 55). In addition, CI-1033 enhances the cytotoxic effects of other therapeutic modalities (21). For example, the agent enhances the cytotoxic CI-1033

2 Dec 2003

21:36

AR

AR206-ME55-25.tex

AR206-ME55-25.sgm

LaTeX2e(2002/01/18)

erbB TARGETS FOR CANCER THERAPY

P1: GBC

453

Annu. Rev. Med. 2004.55:433-457. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

effects of the topoisomerase inhibitors SN-38 and topotecan in vitro, possibly by interfering with a relevant drug-resistance mechanism. Phase I evaluations of CI1033 administered once daily on various intermittent and noninterrupted schedules have been completed, and phase II disease-directed evaluations are ongoing (Table 4). EKB-569 EKB-569 (Wyeth-Ayerst, Philadelphia, PA) is a 3-cyanoquinoline that binds covalently and irreversibly to erbB1 (Table 4). It is a potent inhibitor of a recombinant form of erbB1 TK (IC50 = 1.33 nM) and phosphorylation of erbB1 in cells (IC50 = 15 nM in cells that overexpress erbB1) (58, 59). Although >10 times more drug is required to inhibit other erbB TKs, including erbB2, EKB-569 is equipotent at inhibiting the growth of cells that overexpress erbB1 or erbB2. The effect is specific because 10- to 50-fold more drug is needed to inhibit the growth of cell lines that do not overexpress erbB1 or erbB2. Notable growth inhibition has also been observed in tumors derived from cells that overexpress erbB1 or erbB2, but EKB-569 does not inhibit the growth of tumors with low erbB expression. Consistent with its ability to irreversibly bind to erbB1 and erbB2, inhibition of receptor phosphorylation is sustained far longer than plasma levels of EKB-569. Furthermore, a combination of EKB-569 and the nonsteroidal anti-inflammatory agent sulindac provided a striking protection against colon tumor formation in a mouse model of familial adenomatous polyposis (>95% reduction in polyps), a surrogate for the development of colon cancer (59). Phase I evaluations of EKB administered once daily continuously and for 3 weeks every 4 weeks have been completed, and both phase II single-agent and combination evaluations are ongoing.

CLINICAL DEVELOPMENT AND OBSTACLES AHEAD In clinical evaluations, inhibitors of erbB1 TK have demonstrated excellent tolerability, both as single agents and combined with other anticancer therapeutic modalities (Tables 2–4). Interestingly, the principal toxicity of both MAbs and small molecules directed against erbB1 is an acneiform and/or follicular skin rash, which may be a pharmacodynamic marker of clinical benefit. Consistent antitumor activity with reversible small-molecule inhibitors of RTK and MAbs alike has been noted in patients with NSCLC. Interestingly, however, notable antitumor activity has occurred in patients with colorectal cancer following treatment with MAbs but not the small-molecule TK inhibitors. The relative benefits conferred by irreversible erbB1 TK binding and inhibition of multiple erbB subfamilies will be known following clinical evaluations of CI-1033, EKB-569, and GW572016 in relevant tumor types. Nevertheless, rates of major tumor regression in nonrandomized clinical evaluations have been fairly low, which probably reflects the indiscriminate treatment of unscreened patients with tumors that may or may not possess the appropriate target or determinants for response. The failure to determine which patients are likely to benefit from such therapeutics may also explain why

2 Dec 2003

21:36

Annu. Rev. Med. 2004.55:433-457. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

454

AR

AR206-ME55-25.tex

AR206-ME55-25.sgm

LaTeX2e(2002/01/18)

P1: GBC

ROWINSKY

randomized phase III studies have found no clear advantage of erbB1-targeted therapeutics combined with cytotoxic therapy in unscreened patients with NSCLC. It is important to note that the success of trastuzimab, specifically in increasing the survival of patients with breast cancer, would not have been appreciated if patients had not been screened before treatment for the principal target of trastuzimab, erbB2. Furthermore, since tumor growth delay (in contrast to tumor regression) appears to be the predominant benefit of erbB-targeted therapeutics in preclinical studies, particularly in tumors that are not exclusively driven by erbB dysregulation or overexpression, it will be increasingly necessary to select appropriate endpoints for phase II screening studies to appreciate and quantify drug-induced tumor growth delay. In practicality, however, some indication that the erbB inhibitors possess relevant clinical activity and can modify the natural history of disease progression will be needed before resource-intensive large randomized phase III studies are commenced. Adequately designed clinical trials will ensure that the usefulness of erbB-targeted therapeutics is correctly assessed, so that potentially useful agents are not rejected on the basis of poor performance with regard to an inappropriate clinical or biologic endpoint. The Annual Review of Medicine is online at http://med.annualreviews.org

LITERATURE CITED 1. Tzahar E, Pinkas-Kramarski R, Moyer JD, et al. 1997. Bivalence of EGF-like ligands drives the ErbB signaling network. EMBO J. 16:4938–50 2. Cohen S, Carpenter G. 1975. Human epidermal growth factor: isolation and chemical and biological properties. Proc. Natl. Acad. Sci. USA 72:1317–21 3. Shelly M, Pinkas-Kramarski R, Guarino BC, et al. 1998. Epiregulin is a potent pan-ErbB ligand that preferentially activates heterodimeric receptor complexes. J. Biol. Chem. 273:10496–505 4. Derynck R. 1998. Transforming growth factor alpha. Cell 54:593–95 5. Simon MA. 2000. Receptor tyrosine kinases: specific outcomes from general signals. Cell 103:13–15 6. Walker RA. 1998. The erbB/HER type 1 tyrosine kinase receptor family. J. Pathol. 185:234–35 7. Yarden Y, Sliwkowski MX. 2001. Untangling the ErbB signaling network. Nat. Rev. Mol. Cell. Biol. 2:127–37

8. Schlessinger J. 2000. Cell signaling by receptor tyrosine kinases. Cell 103:211– 25 9. Daly RJ. 1999. Take your partners, please: signal diversification by the erbB family of receptor tyrosine kinases. Growth Factors 16:255–63 10. Riese DJ II, Stern DF. 1998. Specificity within the EGF family/ErbB receptor family signaling network. Bioessays 20:41–48 11. Olayioye MA, Neve RM, Lane HA, et al. 2000. The ErbB signaling network: receptor heterodimerization in development and cancer. EMBO J. 19:3159–67 12. Stern DF, Kamps MP. 1998. EGFstimulated tyrosine phosphorylation of p185neu: a potential model for receptor interactions. EMBO J. 7:995–1001 13. Graus-Porta D, Beerli RR, Daly JM, et al. 1997. ErbB-2, the preferred heterodimerization partner of all ErbB receptors, is a mediator of lateral signaling. EMBO J. 16:1647–55

2 Dec 2003

21:36

AR

AR206-ME55-25.tex

AR206-ME55-25.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:433-457. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

erbB TARGETS FOR CANCER THERAPY 14. Sorkin A, Di Fiore PP, Carpenter G. 1993. The carboxyl terminus of epidermal growth factor receptor/erbB-2 chimerae is internalization impaired. Oncogene 8: 3021–28 15. Riese DJ II, van Raaij TM, Plowman GD, et al. 1995. The cellular response to neuregulins is governed by complex interactions of the erbB receptor family. Mol. Cell. Biol. 15:5770–76 16. Elenius K, Choi CJ, Paul S, et al. 1999. Characterization of a naturally-occurring ErbB4 isoform that does not bind or activate phosphatidyl inositol 3-kinase. Oncogene 8:2607–15 17. Yarden Y. 2001. Biology of HER2 and its importance in breast cancer. Oncology 61(Suppl. 2):1–13 18. Shtiegman K, Yarden Y. 2003. The role of ubiquitylation in signaling by growth factors: implications to cancer. Semin. Cancer Biol. 13:29–40 19. Levkowitz G, Waterman H, Zamir E, et al. 1996. c-Cbl/Sli-1 regulates endocytic sorting and ubiquitination of the epidermal growth factor receptor. Genes Dev. 12:3663–74 20. Chow NH, Liu HS, Lee EI, et al. 1997. Significance of urinary epidermal growth factor and its receptor expression in human bladder cancer. Anticancer Res. 17: 1293–96 21. Rowinsky EK. 2001. Targeting signal transduction. The erbB receptor family as a target for therapeutic development against cancer. Horizons in Cancer Therapeutics: From Bench to Bedside, 2:3–36. West Conshohocken, PA: Meniscus Educ. Inst. 22. Moscatello DK, Holgado-Madruga M, Emlet DR, et al. 1998. Constitutive activation of phosphatidylinositol 3-kinase by a naturally occurring mutant epidermal growth factor receptor. J. Biol. Chem. 273:200–6 23. Wikstrand CJ, Hale LP, Batra SK, et al. 1995. Monoclonal antibodies against EGFRvIII are tumor specific and react

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

P1: GBC

455

with breast and lung carcinomas and malignant gliomas. Cancer Res. 55:3140– 48 Slamon DJ, Godolphin W, Jones LA, et al. 1989. Studies of the HER-2/neu protooncogene in human breast and ovarian cancer. Science 244:707–12 Colomer R, Shamon LA, Tsai MS, et al. 2001. Herceptin: from the bench to the clinic. Cancer Invest. 19:49–56 Hudziak RM, Schlessinger J, Ullrich A. 1987. Increased expression of the putative growth factor receptor p185HER2 causes transformation and tumorigenesis of NIH 3T3 cells. Proc. Natl. Acad. Sci. USA 84: 7159–63 Pauletti G, Godolphin W, Press MF, et al. 1996. Detection and quantitation of HER2/neu gene amplification in human breast cancer archival material using fluorescence in situ hybridization. Oncogene 13: 63–72 Baselga J, Albanell J, Molina MA, et al. 2002. The ErbB receptor family: a therapeutic target for cancer. Trends Mol. Med. 8(Suppl. 4):S19–S26 Arteaga C. 2002. Overview of epidermal growth factor receptor biology and its role as a therapeutic target in human neoplasia. Semin. Oncol. 5(Suppl. 14):3–9 Mendelsohn J. 2002. Targeting the epidermal growth factor receptor for cancer therapy. J. Clin. Oncol. 20(Suppl. 18):1S– 13S Wakeling AE. 2002. Epidermal growth factor receptor tyrosine kinase inhibitors. Curr. Opin. Pharmacol. 2:382–87 Arteaga CL, Moulder SL, Yakes FM. 2002. HER (erbB) tyrosine kinase inhibitors in the treatment of breast cancer. Semin. Oncol. 29(Suppl. 11):4–10 De Bono JS, Rowinsky EK. 2002. The ErbB receptor family: a therapeutic target for cancer. Trends Mol. Med. 8(Suppl. 4):S19–26 Fan Z, Lu Y, Wu X, et al. 1994. Antibodyinduced epidermal growth factor receptor dimerization mediates inhibition of

2 Dec 2003

21:36

456

35.

Annu. Rev. Med. 2004.55:433-457. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

36.

37.

38.

39.

40.

41.

42.

43.

AR

AR206-ME55-25.tex

AR206-ME55-25.sgm

LaTeX2e(2002/01/18)

P1: GBC

ROWINSKY autocrine proliferation of A431 squamous carcinoma cells. J. Biol. Chem. 269: 27595–602 Mendelsohn J. 2000. Blockade of receptors for growth factors: an anticancer therapy: the fourth annual Joseph H. Burchenal American Association for Cancer Research Clinical Research Clinical Award Lecture. Clin. Cancer Res. 6:747– 52 Baselga J, Mendelsohn J. 1997. Type I receptor tyrosine kinases as targets for therapy in breast cancer. J. Mammary Gland Biol. Neoplasia 2:165–74 Wu X, Fan Z, Masui H, et al. 1995. Apoptosis induced by an anti-epidermal growth factor receptor monoclonal antibody to a human colorectal carcinoma cell line and its delay by insulin. J. Clin. Invest. 95:1897–905 Goldstein NI, Prewett M, Zuklys K, et al. 1993. Biological efficacy of a chimeric antibody to the epidermal growth factor receptor in a human tumor xenograft model. Clin. Cancer Res. 53:4322–28 Fontanini G, De Laurentiis M, Vignati S, et al. 1988. Evaluation of epidermal growth factor-related growth factors and receptors and neoangiogenesis in completely resected stage I-IIIA non-small cell lung cancer: amphiregulin and microvessel counts are independent prognostic indicators of survival. Clin. Cancer Res. 4:241–49 Cohen RB. 2003. Epidermal growth factor receptor as a therapeutic target in colorectal cancer. Clin. Colorectal Cancer 2:246– 51 Ritter CA, Arteaga CL. 2003. The epidermal growth factor receptor tyrosine kinase: a promising therapeutic target in solid tumors. Semin. Oncol. 30:1(Suppl. 1):3–11 Arteaga CL. 2002. Epidermal growth factor receptor dependence in human tumors: more than just expression? Oncologist 7(Suppl. 4):31–39 Yang XD, Jia XC, Corvalan JR, et al.

44.

45.

46.

47.

48.

49.

50.

51.

52.

1999. Eradication of established tumors by a fully human monoclonal antibody to the epidermal growth factor receptor without concomitant chemotherapy. Cancer Res. 59:1236–43 Yang X, Jia X, Corvalan JR, et al. 2001. Development of ABX-EGF, a fully human anti-EGF receptor monoclonal antibody, for cancer therapy. Crit. Rev. Oncol. Hematol. 38:17–23 Rowinsky E, Schwartz G, Dutcher J, et al. 2002. ABX-EGF, a fully human antiepidermal growth factor receptor monoclonal antibody: phase II clinical trial in renal cell cancer. Eur. J. Cancer. 38(Suppl. 7):S57–S65 Figlin RA, Belldegrun AS, Crawford J, et al. 2002. ABX-EGF, a fully human antiepidermal growth factor receptor (EGFR) monoclonal antibody (mAb) in patients with advanced cancer: phase 1 clinical results. Proc. Am. Soc. Clin. Oncol. 21:10a (Abstr.) Meropol NJ, Berlin J, Hecht JR, et al. 2003. Multicenter study of ABX-EGF monotherapy in patients with metastatic colorectal cancer. Proc. Am. Soc. Clin. Oncol. 22:256 (Abstr.) Pfister DG, Lipton A, Belt R, et al. 1999. A phase I trial of the epidermal growth factor receptor (EGFR)-directed bispecific antibody (BsAB) MDX-447 in patients with solid tumors. Proc. Am. Soc. Clin. Oncol. 18:256 (Abstr.) Levitt ML, Koty PP. 1999. Tyrosine kinase inhibitors in preclinical development. Invest. New Drugs 17:213–26 Fry DW. 2000. Site-directed irreversible inhibitors of the erbB family of receptor tyrosine kinases as novel chemotherapeutic agents for cancer. Anticancer Drug Des. 15:3–16 Baselga J, Averbuch SD. 2000. ZD1839 (‘Iressa’) as an anticancer agent. Drugs 60(Suppl. 1):33–40 Albanell J, Rojo F, Averbuch S, et al. 2002. Pharmacodynamic studies of the epidermal growth factor receptor inhibitor

2 Dec 2003

21:36

AR

AR206-ME55-25.tex

AR206-ME55-25.sgm

LaTeX2e(2002/01/18)

erbB TARGETS FOR CANCER THERAPY

53.

Annu. Rev. Med. 2004.55:433-457. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

54.

55.

56.

ZD1839 in skin from cancer patients. Histopathologic and molecular consequences of receptor inhibition. J. Clin. Oncol. 20:110–24 Moyer JD, Barbacci EO, Iwata KK, et al. 1997. Induction of apoptosis and cell cycle arrest by CP-358,774, an inhibitor of epidermal growth factor receptor tyrosine kinase. Cancer Res. 57:4838–48 Hidalgo M, Siu LL, Nemunaitis J, et al. 2001. A phase I and pharmacologic study of OSI-774, an epidermal growth factor receptor tyrosine kinase inhibitor, in patients with advanced solid malignancies. J. Clin. Oncol. 19:3267–79 Fry DW. 2000. Site-directed irreversible inhibitors of the erbB family of receptor tyrosine kinases as novel chemotherapeutic agents for cancer. Anticancer Drug Des. 15:3–16 Xia W, Mullin MJ, Keith BR, et al. 2002. Anti-tumor activity of GW572016: a dual tyrosine kinase inhibitor blocks EGF acti-

57.

58.

59.

60.

P1: GBC

457

vation of EGFR/erbB2 and downstream Erk 1/2 and AKT pathways. Oncogene 21:6255–63 Greenberger LM, Discafani C, Wang Y-F, et al. 2000. EKB-569: a new irreversible inhibitor of EGFR tyrosine kinase for the treatment of cancer. Clin. Cancer Res. 6(Suppl.):4544s Averbuch SD. 2002. Lung cancer prevention: retinoids and the epidermal growth factor receptor—a phoenix rising? Clin. Cancer Res. 1:1–3 Torrance CJ, Jackson PE, Montgomery E, et al. 2000. Combinatorial chemoprevention of intestinal neoplasia. Nat. Med. 6:1024–28 Spector N, Raefsky E, Hurwiz H, et al. 2003. Safety, clinical efficacy, and biologic assessments from EGF10004, a randomized phase IB study of GW572016 for patients with metastatic carcinomas overexpressing EGFR or erbB2. Proc. Am. Soc. Clin. Oncol. 18:193a (Abstr.)

15 Dec 2003

15:12

AR

AR206-ME55-26.tex

AR206-ME55-26.sgm

LaTeX2e(2002/01/18) P1: GBC 10.1146/annurev.med.55.091902.104511

Annu. Rev. Med. 2004. 55:459–75 doi: 10.1146/annurev.med.55.091902.104511 First published online as a Review in Advance on Oct. 27, 2003

NONMYELOABLATIVE ALLOGENEIC IMMUNOTHERAPY FOR SOLID TUMORS∗

Annu. Rev. Med. 2004.55:459-475. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Richard W. Childs and John Barrett Allogeneic Hematopoietic Cell Transplant Unit, Hematology Branch, National Heart, Lung, and Blood Institutes, National Institutes of Health, Bethesda, Maryland 20892; email: [email protected], [email protected]

Key Words allogeneic stem cell transplantation, nonmyeloablative transplant, graft-versus-tumor, graft-versus-leukemia, renal cell carcinoma ■ Abstract Over the past decade, considerable advances have been made in the field of allogeneic hematopoietic stem cell transplantation. Recognition that transplanted donor immune cells can cure patients with leukemia has led to the development of nonmyeloablative or “low-intensity” conditioning regimens, which have expanded the application of allogeneic transplantation to a growing number of hematological malignancies. The improved safety and preliminary success of this transplant approach have justified applying allogeneic immunotherapy to patients with treatment-refractory solid tumors.

INTRODUCTION Allogeneic bone marrow or peripheral blood cell transplantation (referred to as hematopoietic stem cell transplantation or SCT) was originally developed as a method to rescue bone marrow function following high-dose (myeloablative) therapy in the treatment of hematological malignancies (1). Despite recent advances in systemic therapies, for many patients allogeneic SCT remains the only treatment that offers a chance of cure. Over the past decade, our understanding of the mechanisms by which malignant cells are eradicated following transplantation has evolved considerably. Originally, high-dose conditioning was thought to be the main factor responsible for long-term disease-free survival. More recently, it has become clear that transplanted immune cells are capable of killing malignant cells. This so-called graft-versus-leukemia (GVL) or graft-versus-tumor (GVT) effect is a powerful form of immunotherapy that can eradicate advanced or even chemotherapy-resistant leukemias. ∗ The U.S. Government has the right to retain a nonexclusive, royalty-free license in and to any copyright covering this paper.

459

15 Dec 2003

15:12

Annu. Rev. Med. 2004.55:459-475. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

460

AR

CHILDS

AR206-ME55-26.tex

¥

AR206-ME55-26.sgm

LaTeX2e(2002/01/18)

P1: GBC

BARRETT

The list of hematological malignancies in which GVL effects have been described now includes not only acute and chronic leukemias but also myelodysplastic syndromes, myelofibrosis, Hodgkin’s and non-Hodgkin’s lymphomas, multiple myeloma, and Epstein-Barr virus–related lymphoproliferative disorder. The ability of donor lymphocyte infusions to induce remission in patients with relapsed chronic myelogenous leukemia (CML) has provided indisputable evidence of the curative potential of GVL and impelled the development of nonmyeloablative transplant approaches (2). Furthermore, it has stimulated oncologists to seek similar beneficial allogeneic immune effects in metastatic solid tumors (3). This article reviews the development and early clinical results of allogeneic stem cell transplantation as immunotherapy for solid tumors.

THE IMMUNE SYSTEM AND CANCER The first documented attempt to use the immune system to treat patients with advanced cancer was reported in the late nineteenth century by W. Coley, who injected bacterial toxins into the tumors of cancer patients (4). However, more than a half a century passed before investigators began systematic efforts to explore immunotherapy as an adjunct to other systemic therapies in the treatment of advanced cancer. Recent interest in the development of immune-based treatments for solid tumors has been motivated by the failure of conventional chemotherapy to cure most patients. The late twentieth century saw the birth of cytokine therapies designed to stimulate the host’s immune system against cancer. Pioneering studies by Rosenberg and colleagues in the 1980s provided some of the first evidence that natural killer (NK) cells and T lymphocytes could induce clinically relevant regression of advanced cancer (4–8). In particular, reports of treatment-refractory disease regressing following interleukin-2 (IL-2) and/or interferon-alpha treatment suggested that immunotherapy-based strategies could complement chemotherapy in some tumors. Remarkably, some patients with metastatic melanoma and renal cell carcinoma (RCC) achieved durable remissions following treatment with IL-2, an immune-enhancing agent with no direct antineoplastic effects. Unfortunately, most immunotherapy regimens using cytokines have generally had low response rates and are sometimes associated with considerable toxicity (9–11). It is generally agreed that the main contribution of cytokine-based treatment has been to establish proof of concept, laying the foundation for future immune-based therapies. Recently, the characterization of antigens overexpressed or restricted to cancer cells has led to the development of vaccines aimed at enhancing host immunity specifically at the tumor (12, 13). This remains a rapidly developing area of investigation with the potential for improving the safety and specificity of immunotherapy compared to nontargeted cytokine-based approaches. Nevertheless, these strategies are very much in their infancy; only a handful of patients treated with cancer vaccines have yet shown clinical benefit. Defects in the immune system of the tumor-bearing host may be partially responsible for the low response rates from treatments designed to boost self

15 Dec 2003

15:12

AR

AR206-ME55-26.tex

AR206-ME55-26.sgm

LaTeX2e(2002/01/18)

P1: GBC

ALLOGENEIC IMMUNOTHERAPY IN SOLID TUMORS

461

(autologous) immunity to cancer (14–18). Prior exposure to chemotherapy or global T-cell anergy to cancer as a consequence of tumors lacking immunostimulatory ligands (e.g., B7.1) may contribute to these abnormalities. Allogeneic SCT, which replaces the recipient’s defective immune system with that of the healthy donor, could potentially overcome some of these barriers.

Annu. Rev. Med. 2004.55:459-475. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

ALLOGENEIC IMMUNOTHERAPY: THE GRAFT-VERSUS-LEUKEMIA EFFECT Allogeneic SCT offers many patients with hematological malignancies the only chance for a cure. During its early development, it was believed that “mega-dose” conditioning (chemotherapy alone or in combination with total body irradiation) was an absolute requirement for the eradication of all malignant cells (1). However, the advent of highly sensitive molecular techniques to measure minimal residual disease showed that many patients with detectable residual leukemia in the first few months following SCT ultimately became “molecular cures.” The realization that high-dose, or myeloablative, conditioning frequently fails to eradicate all leukemic cells and the observation that the risk of leukemic relapse is lower in patients who develop graft-versus-host disease (GVHD) provided the first evidence of a GVL effect following SCT (19–23). The demonstration that patients with relapsed CML following allogeneic SCT could be cured by donor lymphocyte infusions, established beyond doubt both the existence and curative potential of the GVL effect (24, 25). Antimalignancy effects after lymphocyte infusions occur in a wide variety of hematological malignancies (26–31). However, response rates to donor lymphocyte infusion for relapsed malignancy after SCT vary according to stage and type of disease (Table 1). For example, ∼80% of patients with CML relapsing in chronic phase can be expected to be cured by donor lymphocyte infusion therapy. In contrast, durable responses to donor lymphocyte infusion are relatively rare in relapsed acute myelogenous leukemia (AML) or CML relapsing into blast crisis. TABLE 1 Targets for a graft-versus-leukemia effect Malignancy

Susceptibility to GVL

Chronic myelogenous leukemia (chronic phase)

High

Chronic lymphocytic leukemia

High

Low-grade non-Hodgkin’s lymphoma

High

Acute myelogenous leukemia

Intermediate

Myelodysplastic syndrome

Intermediate

Multiple myeloma

Intermediate

Intermediate/high grade non-Hodgkin’s lymphoma

Intermediate

Hodgkin’s disease

Intermediate

Acute lymphocytic leukemia

Low

Chronic myelogenous leukemia (blast crisis)

Low

15 Dec 2003

15:12

Annu. Rev. Med. 2004.55:459-475. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

462

AR

CHILDS

AR206-ME55-26.tex

¥

AR206-ME55-26.sgm

LaTeX2e(2002/01/18)

P1: GBC

BARRETT

The mechanisms that account for variability in susceptibility to GVL have not been defined, although in general, rapidly proliferating leukemias appear to be less responsive to the immune effect. Although allogeneic NK cells may play a role in GVL in HLA-mismatched transplants, most in vitro and in vivo evidence suggests donor T cells are the dominant immune population mediating disease regression (32–37). Indeed, the risk of relapse of some hematological malignancies (e.g., CML) increases substantially when donor T cells are depleted from the allograft in an effort to prevent GVHD (38). The antigens that serve as the primary target for GVL are not fully understood. The pattern of tissue distribution dictates whether GVL occurs in the presence or absence of GVHD. The scenario of relapsed disease remitting after donor lymphocyte infusion in the absence of GVHD would imply a response to antigens that are restricted to hematopoietic tissues or specifically to the leukemia (39–42). T cells with leukemia-restricted patterns of cytotoxicity have been expanded in vitro from the blood of responding patients. However, in many patients the GVL effect is accompanied by acute or chronic GVHD, suggesting a broader tissue distribution of target antigens. Such responses are directed against broadly expressed minor histocompatibility antigens (mHa) that are polymorphic between the patient and donor.

GRAFT-VERSUS-TUMOR EFFECTS IN SOLID TUMORS Rationale Expression of major histocompatibility complex (MHC) class I, in vitro susceptibility to peptide-specific T-cell killing, and regression of disease after treatment with IL-2 provide the basis for speculation that select solid tumors could be susceptible to a GVT effect following allogeneic SCT (5–13). As discussed, mHa differences between the patient and donor may be the dominant antigens targeted by the donor immune system during GVL effect. It can be hypothesized that cancers originating from tissues that are a target of GVHD (skin, liver, gastrointestinal tract, etc.) would express the same mHa that induce GVHD, thus making them a target of the donor alloresponse. Additionally, antigens restricted to the tumor could stimulate tumor-specific alloresponses from donor T cells in contrast to defective tolerized T cells in the tumor-bearing host. However, despite the theoretical advantages of allogeneic immunotherapy, it should be remembered that solid tumors are apt to evade immune-mediated killing. Tumor-cell downregulation of MHC molecules, secretion of soluble inhibitors of T-cell function (e.g., TGF-β), and expression of membrane-bound Fas ligand are just a few mechanisms that could potentially inhibit both autologous and allogeneic immune responses (3, 17, 18, 43).

Animal Models Before clinical trials were pursued, animal models were tested for the existence of a graft-versus-solid-tumor effect (44). Among mice inoculated with mammary

15 Dec 2003

15:12

AR

AR206-ME55-26.tex

AR206-ME55-26.sgm

LaTeX2e(2002/01/18)

ALLOGENEIC IMMUNOTHERAPY IN SOLID TUMORS

P1: GBC

463

adenocarcinoma cells, survival in those that received an allogeneic SCT was superior to survival in recipients of a syngeneic SCT (45). Further studies provided evidence that murine mammary adenocarcinoma cells expressed mHa that could be targeted by alloreactive donor T cells in the setting of allogeneic but not autologous bone marrow transplantation (46). These studies provided valuable insight into the plasticity of the GVL effect, suggesting for the first time that allogeneic immune responses might also be inducible against nonhematological malignancies.

Annu. Rev. Med. 2004.55:459-475. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Clinical Data One of the first studies investigating whether an alloresponse might occur outside of the setting of hematological malignancies was a pilot trial comparing autologous versus allogeneic SCT in children with metastatic neuroblastoma (47). Although GVT effects were not observed in the allogeneic cohort, this isolated observation was not sufficient to support generalizations about the susceptibility of other solid tumors to GVT. The earliest evidence supporting the existence of an allogeneic GVT effect in a solid tumor came from patients with metastatic breast carcinoma undergoing fully myeloablative allogeneic SCT. A letter describing the incidental regression of a metastatic breast carcinoma lesion in a patient receiving an allogeneic SCT for relapsed AML raised the possibility that a GVT effect was responsible (48). At the same time, regression of liver metastasis in association with severe acute GVHD was reported in a woman transplanted for metastatic breast carcinoma (49). The ability to kill breast cancer cell lines with alloreactive T cells expanded from the patient during GVHD suggested that disease regression resulted from donor T cells targeting broadly expressed (including on the tumor) mHa. In 1998, a series of 10 patients with metastatic breast cancer treated at a single institution with an allogeneic SCT was reported (50). Although disease regression was mainly attributable to myeloablative conditioning, two patients responded during acute GVHD following the withdrawal of immunosuppression. These reports provided the first evidence that a donor immune-mediated antitumor effect could occur after an allogeneic SCT. Unfortunately, enthusiasm for this approach was tempered by significant and sometimes fatal toxicities associated with the transplant.

NONMYELOABLATIVE CONDITIONING AS A PLATFORM TO EVALUATE ALLOGENEIC SCT IN SOLID TUMORS The observation of GVL effects following SCT and the demonstration that the immune system could be used to treat some metastatic cancers following cytokine therapy prompted exploration of allogeneic transplantation in patients with nonhematological malignancies. Despite progressive improvements in transplant safety, procedure-related mortality remains near 25%. Without evidence of efficacy, most considered this risk too high to justify studies of allogeneic SCT in patients with nonhematological malignancies.

15 Dec 2003

15:12

Annu. Rev. Med. 2004.55:459-475. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

464

AR

CHILDS

AR206-ME55-26.tex

¥

AR206-ME55-26.sgm

LaTeX2e(2002/01/18)

P1: GBC

BARRETT

Morbidity related to intensive conditioning contributes significantly to the risk of myeloablative transplants. Veno-occlusive disease of the liver, pneumonitis, and severe mucositis leading to opportunistic bacterial and fungal sepsis are the main toxicities directly related to high-dose conditioning that can be lethal. In an effort to improve the safety profile of allogeneic SCT, investigators recently developed reduced-intensity or nonmyeloablative conditioning regimens. The key factor motivating nonmyeloablative transplant trials was an increased confidence in the potential of the GVL effect to cure malignant diseases. Nonmyeloablative conditioning regimens use powerful immunosuppressants to allow engraftment of the donor immune system while reducing overall toxicity. Pilot trials of this approach were first evaluated in hematologic malignancies known to be sensitive to GVL (51–57). Although no direct comparisons of myeloablative versus nonmyeloablative regimens have yet been made, preliminary data on the safety of this new approach have been encouraging. Several centers reported transplant-related mortality rates of <20% in patient cohorts usually precluded from conventional SCT because the risk of procedure-related mortality was considered unacceptable. Nonmyeloablative transplants have already been shown to induce GVL effects sufficient to cure patients with a variety of advanced hematological malignancies, including acute and chronic leukemias. The reduced toxicity of this approach offered investigators a safer transplantation modality through which to test whether GVT effects could be induced in solid tumors.

NONMYELOABLATIVE ALLOGENEIC SCT FOR METASTATIC RENAL CELL CARCINOMA Trial Design The possibility that the immune system could control metastatic renal cell carcinoma (RCC) was first entertained in the late 1920s after a case report of a patient who had spontaneous regression of metastatic disease (58). The search for donormediated GVT effects in this disease was further encouraged by the observation that many patients with metastatic RCC have dysfunctional immunity, and by other in vitro and in vivo evidence of RCC’s susceptibility to immune attack (10, 59): ■ ■

■ ■

Isolation of tumor-infiltrating lymphocytes from metastatic lesions In vitro data showing susceptibility to lymphokine-activated killer (LAK) cells In vitro data showing susceptibility to killing by antigen-specific T cells Response of systemic disease to immunomodulator therapy (e.g., IL-2)

We began pursuing nonmyeloablative SCT in patients with metastatic RCC in late 1997, soon after the favorable results of nonmyeloablative SCT in hematological malignancies (60). Because of the experimental nature of the transplant approach, we restricted our pilot trial to patients with metastatic disease

15 Dec 2003

15:12

AR

AR206-ME55-26.tex

AR206-ME55-26.sgm

LaTeX2e(2002/01/18)

ALLOGENEIC IMMUNOTHERAPY IN SOLID TUMORS

P1: GBC

465

Annu. Rev. Med. 2004.55:459-475. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

who had failed cytokine-based immunotherapy. Patients with CNS metastasis were excluded in anticipation of an increased risk of intracranial bleeding from conditioning-related thrombocytopenia. Furthermore, concerns related to GVHD morbidity limited the procedure to those who had either an HLA-identical or single-HLA-antigen–mismatched sibling donor. Our strategy to minimize toxicity and optimize the induction of a GVT effect included reduced-intensity conditioning, early withdrawal of immunosuppression, and the administration of donor lymphocyte infusions or cytokines (interferon-alpha or IL-2) for those with progressing disease.

Toxicity and Engraftment Preliminary results of pilot trials evaluating nonmyeloablative SCT in patients with advanced RCC have recently been reported (60–65). Although most regimens have been well tolerated, complications associated with nonmyeloablative SCT vary with the type and intensity of conditioning agents. Common toxicities associated with conventional myeloablative SCT, such as severe mucositis and veno-occlusive disease of the liver, are rare in RCC patients undergoing nonmyeloablative SCT (Table 2). Because nonmyeloablative conditioning does eradicate recipient hematopoiesis, both donor and patient myeloid and lymphoid cells are usually detectable at the time of neutrophil recovery (54). This state, called mixed chimerism, is in contrast to the full donor myeloid and lymphoid chimerism that follows myeloablative SCT. Mixed T-lymphocyte chimerism appears to induce donor tolerance to recipient tissue, decreasing risk of acute GVHD. Donor immune effects do not usually occur until donor lymphocytes predominate in the blood. Patients with a prior history of chemotherapy exposure and those exposed to more immunosuppressive conditioning agents develop full donor chimerism faster than those who are chemotherapy-naive or who receive less intense conditioning. Because most

TABLE 2 Transplant-related toxicities after nonmyeloablative SCT Toxicity

Incidence (%)

Mucositis

0%–5%

Veno-occlusive disease of the liver

0%–5%

Pneumonitis

0%–10%

Febrile neutropenia

25%–90%

Thrombocytopenia requiring platelet transfusions

0%–30%

Graft rejection

0%–10%

Acute graft-versus-host disease

15%–55%

Cytomegalovirus reactivation

20%–40%

Chronic graft-versus-host disease

30%–70%

Transplant-related mortality (overall)

10%–20%

15 Dec 2003

15:12

466

AR

AR206-ME55-26.tex

CHILDS

¥

AR206-ME55-26.sgm

LaTeX2e(2002/01/18)

P1: GBC

BARRETT

patients with RCC have not received chemotherapy prior to the transplant, donor Tcell engraftment is delayed compared to patients with hematologic malignancies. As a consequence, the majority of nonmyeloablative SCT regimens incorporate strategies to accelerate the conversion from mixed to full donor chimerism by early discontinuation of GVHD prophylaxis (usually cyclosporine or tacrolimus) or by the infusion of donor lymphocytes.

Evidence That RCC is Susceptible to a GVT Effect Annu. Rev. Med. 2004.55:459-475. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Although the ultimate role of allogeneic immunotherapy in the treatment of advanced kidney cancer is still evolving, preliminary trials of nonmyeloablative SCT in RCC quickly established that this tumor is susceptible to a GVT effect (Table 3). The first patient with cytokine-refractory metastatic RCC treated at the NIH using cyclophosphamide (120 mg/m2) and fludarabine (125 mg/m2) conditioning had complete regression of pulmonary and bony metastatic disease four months after the procedure and remains in remission >5 years post-transplant (Figure 1). Subsequently, 10 of the first 19 patients transplanted (all cytokine-therapy failures) had a GVT effect with 7 partial and 3 complete responses (61). Disease responses were observed most commonly in patients with the clear-cell variant of RCC (∼80% of all RCC cases) in the setting of isolated pulmonary metastatic disease. However, dramatic responses occasionally occurred in patients with extensive metastatic disease in multiple metastatic sites including the bones, lymph nodes, and liver (Figure 2). Failure to observe disease regression following immunosuppression withdrawal did not always preclude the induction of a GVT effect, as some patients responded to a donor lymphocyte infusion. Remarkably, some patients who had been resistant to interferon-alpha before SCT responded to low doses of this agent given subcutaneously after transplantation. Investigators from The University of Chicago reported four partial responses in 15 patients who received an allograft from an HLA-identical sibling (62). Notably, one partial responder had regression in the primary kidney tumor, a rare event among responders to cytokine-based therapy. Their initial regimen using low doses of fludarabine (90 mg/m2) and cyclophosphamide (2 g/m2) resulted in a 75% graft rejection rate. When the doses of fludarabine and cyclophosphamide were increased to 150 mg/m2 and 4 g/m2, respectively, all subsequent patients achieved

TABLE 3 Published results of nonmyeloablative SCT for renal cell carcinoma Reference

Patients (#)

Conditioning agents

Response rate (PR + CR)

61

19

Fludarabine, cyclophosphamide

53%

62

15

Fludarabine, cyclophosphamide

33%

7

7

Fludarabine, cyclophosphamide

0%

63

7

Thiotepa, fludarabine, cyclophosphamide

71%

15 Dec 2003

15:12

AR

AR206-ME55-26.tex

AR206-ME55-26.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:459-475. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

ALLOGENEIC IMMUNOTHERAPY IN SOLID TUMORS

P1: GBC

467

Figure 1 The nonmyeloablative transplant approach used in the pilot trial at the National Institutes of Health for patients with treatment-refractory renal cell carcinoma. Post-transplant immune enhancement through donor lymphocyte infusion or cytokine administration (usually subcutaneous interferon-alpha or IL-2) is reserved for patients with disease progression in the absence of acute or chronic graft-versus-host disease.

sustained donor engraftment. These results highlight how small changes in the doses of conditioning drugs can dramatically influence engraftment. In contrast to the high incidence of GVHD (∼55%) observed following the NIH regimen, only 2 of 12 (17%) patients treated at The University of Chicago experienced grade 2 or greater acute GVHD, perhaps because of a more gradual withdrawal of GVHD prophylaxis. It is of some concern, however, that the lower incidence of GVHD may also be associated with a weaker GVT effect.

Figure 2 Regression of multiple pulmonary metastases in a patient with IL-2– refractory renal cell carcinoma (clear-cell type) 7.5 months after a nonmyeloablative allogeneic transplant.

15 Dec 2003

15:12

Annu. Rev. Med. 2004.55:459-475. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

468

AR

CHILDS

AR206-ME55-26.tex

¥

AR206-ME55-26.sgm

LaTeX2e(2002/01/18)

P1: GBC

BARRETT

Figure 3 Post-transplant events that provide evidence supporting the involvement of donor T cells in mediating disease regression in renal cell carcinoma events after nonmyeloablative stem cell transplantation.

Another group recently reported partial responses in 4 of 7 (57%) patients with metastatic RCC following treatment with a fludarabine- and thiotepa-based SCT (63). The regimen was associated with minimal toxicity and is now being used in a cooperative European trial investigating GVT effects in patients with a variety of metastatic solid tumors. Regression of metastatic RCC has also been described in nonmyeloablative transplants using 2-chlorodeoxyadenosine (2-CDA), lowdose busulphan, or low-dose total body irradiation (64–66). Trials investigating which nonmyeloablative strategy might be optimal in this malignancy are yet to be developed.

Mechanisms of GVT in RCC The development of more effective transplant approaches requires a better understanding of the immune cells responsible for the GVT effect. The nonmyeloablative SCT trials described above have all provided indirect evidence that donor immune cells are responsible for the regression of metastatic RCC. The observation that disease regression is delayed four months or more after conditioning and is associated with acute GVHD, immunosuppression withdrawal, donor lymphocyte infusions, and a state of predominantly donor T-cell chimerism strongly suggests that donor T lymphocytes play a central role in mediating disease responses (Figure 3). Furthermore, the observation that tumor regression occurs with or without acute GVHD suggests that both broadly expressed mHa and antigens restricted to the tumor may be target antigens for these allogeneic immune effectors. Preliminary in vitro data show that RCC cells express a broad range of mHa that could render them susceptible to a GVT effect in the setting of GVHD (67,

15 Dec 2003

15:12

AR

AR206-ME55-26.tex

AR206-ME55-26.sgm

LaTeX2e(2002/01/18)

ALLOGENEIC IMMUNOTHERAPY IN SOLID TUMORS

P1: GBC

469

68). T-cell clones with tumor-restricted cytotoxicity have been expanded from a few that responded without GVHD (69). These preliminary findings support the hypothesis that distinct T-cell populations recognizing tumor-restricted antigens and/or antigens shared by both the tumor and normal tissues (e.g., mHa) are targets for the GVT effect.

Annu. Rev. Med. 2004.55:459-475. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

LIMITATION OF ALLOGENEIC SCT IN RENAL CELL CARCINOMA The susceptibility of RCC to a GVT effect following allogeneic SCT has already provided clinical benefit to patients with advanced treatment-refractory disease. Unfortunately, several factors currently limit the broader application of this approach: ■ ■ ■ ■ ■

Allogeneic SCT requires an HLA-matched sibling donor There is a 4–6-month delay before the GVT effect occurs Patients with rapidly progressive disease are unlikely to benefit Risk of regimen related mortality is 10%–20% Complete response is rare

Currently, nonmyeloablative SCT is reserved for patients with an HLAcompatible sibling donor, i.e., approximately one third of all patients. Trials evaluating nonmyeloablative SCT using HLA-matched unrelated donors have started, and if effective, could potentially expand the application of allogeneic immunotherapy to the majority of patients with RCC. The regimen-related toxicity of nonmyeloablative SCT is clearly lower than would be expected with a conventional transplant, but life-threatening complications such as GVHD and opportunistic infection still occur. At present, ∼10%–15% of patients die as a complication of the procedure. Consequently, most referring oncologists prefer to reserve transplantation for patients failing cytokine therapy. However, because the GVT effect typically takes four months or longer to become established, delaying the transplant increases the risk that the patient will succumb to the tumor before an antitumor response occurs (70). Metastatic RCC is often a rapidly proliferating tumor associated with short survival. It is therefore important to be selective when choosing transplant candidates, as patients with “explosive” metastatic disease will not survive the time required for the generation of a GVT effect.

ALLOGENEIC SCT FOR MELANOMA Metastatic melanoma has shared a reputation with RCC as an immunoresponsive tumor. Given the positive results of nonmyeloablative SCT in patients with metastatic RCC, it is reasonable to presume that similar GVT effects could be

15 Dec 2003

15:12

Annu. Rev. Med. 2004.55:459-475. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

470

AR

CHILDS

AR206-ME55-26.tex

¥

AR206-ME55-26.sgm

LaTeX2e(2002/01/18)

P1: GBC

BARRETT

induced against this tumor. Several groups have attempted to use nonmyeloablative transplantation in patients with advanced treatment-refractory melanoma (54, 71). Surprisingly, preliminary results show allogeneic SCT to have little to no efficacy in this disease. Anecdotal reports of patients with “explosive” metastatic progression in the setting of acute and chronic GVHD are particularly discouraging. A retrospective analysis of 25 patients with metastatic melanoma treated at four different institutions with three different nonmyeloablative regimens showed disappointing results (71). Although 24 of 25 patients achieved sustained donor engraftment and about half developed acute GVHD, only one patient had evidence of a GVT effect. None of the six patients who received a donor lymphocyte infusion responded. Most depressing was the median survival of only 100 days for this patient group. These results may have been due to the inclusion of patients with treatment-refractory, rapidly proliferating tumors, who could not benefit from a delayed GVT effect. Further studies of allogeneic SCT in metastatic melanoma should therefore be reserved for the small subset of patients with slowly growing disease.

ALLOGENEIC PERIPHERAL BLOOD SCT FOR OTHER SOLID TUMORS Although the number of investigational transplants being conducted for solid tumors has increased substantially over the past few years, insufficient data preclude comment on the efficacy of allogeneic SCT in most solid tumors other than RCC. There are anecdotal reports of GVT effects following nonmyeloablative SCT in patients with metastatic breast carcinoma, colon carcinoma, pancreatic carcinoma, and osteosarcoma (63, 64, 66). A case report and a small case series of tumor responses in patients with metastatic ovarian cancer have also been described recently (72, 73). However, the susceptibility of this tumor to chemotherapy and the proximity of the responses to the transplant conditioning make it difficult to conclude with certainty that disease regression resulted from an immune effect. Trials designed to systematically investigate nonmyeloablative SCT in a variety of metastatic solid tumors are currently under way in the United States, Japan, and Europe. It will take several years for them to accrue sufficient patients to determine the sensitivity of individual solid tumors to GVT effects.

FUTURE DIRECTIONS These initial transplant trials have provided proof of principle that an allogeneic GVT effect can be used to treat advanced solid tumors. However, strategies to separate GVT from GVHD are needed to improve the safety and efficacy of this transplant technique. Based on observations described here, it appears that donor T cells recognizing the tumor can be distinct from those causing GVHD. Methods to selectively deplete alloreactive cells that respond to GVHD antigens while

15 Dec 2003

15:12

AR

AR206-ME55-26.tex

AR206-ME55-26.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:459-475. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

ALLOGENEIC IMMUNOTHERAPY IN SOLID TUMORS

P1: GBC

471

preserving T cells with antiviral and antileukemia effects are currently being investigated in clinical trials (74, 75). If effective, this strategy could provide a useful basis for establishing the donor immune system without a need for post-transplant immunosuppression. Furthermore, it could provide a platform to target allogeneic lymphocytes to the tumor through vaccination strategies or the adoptive transfusion of tumor-specific T cells. Another promising approach is to exploit the ability of donor NK cells to exert powerful allo-immune cytotoxicity in the setting of mismatched donor-recipient combinations. This effect has recently been illuminated in studies demonstrating heightened NK cell cytotoxicity to HLA-mismatched tumor targets as the consequence of killer IgG-like receptor incompatibility. In this situation, NK cells that are normally inhibited from exerting cytotoxicity by suppressing signals from autologous HLA class I molecules can be cytotoxic to HLA-mismatched targets. In HLA-mismatched transplantation, such incompatibility can result in powerful NK cell–mediated effects in which leukemic relapse is almost completely abrogated (33, 76). Whether solid tumors might be similarly susceptible to alloreactive NK cells is a current area of investigation (77). Although the use of allogeneic SCT for the treatment of solid tumors is still in its infancy, developments such as these provide a realistic expectation that SCT will be more widely and effectively used in the future. The Annual Review of Medicine is online at http://med.annualreviews.org

LITERATURE CITED 1. Thomas ED, Lochte HL Jr, Cannon JH, et al. 1959. Supra-ethal whole body irradiation and isologous marrow transplantation in man. J. Clin. Invest 38:1709–16 2. Barrett J, Childs R. 2000. Nonmyeloablative stem cell transplants. Br. J. Haematol. 111:6–17 3. Childs R, Barrett J. 2002. Nonmyeloablative stem cell transplantation for solid tumors: expanding the application of allogeneic immunotherapy. Semin. Hematol. 39(1):63–71 4. Coley W. 1896. Further observations upon the treatment of malignant tumors with the toxins of erysipelas and bacillus proddigious with a report of 160 cases. Johns Hopkins Hosp. Bull. 7:157–62 5. Rosenberg SA, Lotze MT, Muul LM, et al. 1987. A progress report on the treatment of 157 patients with advanced cancer using lymphokine-activated killer cells and

6.

7.

8.

9.

interleukin-2 or high-dose interleukin-2 alone. N. Engl. J. Med. 316:889–97 Topalian SL, Solomon D, Avis FP, et al. 1988. Immunotherapy of patients with advanced cancer using tumor-infiltrating lymphocytes and recombinant interleukin-2: a pilot study. J. Clin. Oncol. 6:839–53 Rosenberg SA, Yang JC, Topalian SL, et al. 1994. Treatment of 283 consecutive patients with metastatic melanoma or renal cell carcinoma using high-dose bolus interleukin-2. JAMA 271:907–13 Quesada JR, Swanson DA, Trindade A. 1983. Renal cell carcinoma: anti-tumor effects of leukocyte interferon. Cancer Res. 43:940 Negrier S, Escudier B, Lasset C, et al. 1998. Recombinant human interleukin-2, recombinant human interferon alpha 2a, or both in metastatic renal-cell carcinoma. N. Engl. J. Med. 338:1272–78

15 Dec 2003

15:12

Annu. Rev. Med. 2004.55:459-475. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

472

AR

CHILDS

AR206-ME55-26.tex

¥

AR206-ME55-26.sgm

LaTeX2e(2002/01/18)

P1: GBC

BARRETT

10. Motzer RJ, Bander NH, Nanus DM. 1996. Renal cell carcinoma. N. Engl. J. Med. 335:856 11. Gleave ME, Elhilali M, Fredet Y, et al. 1998. Interferon gamma-1b compared with placebo in metastatic renal cell carcinoma. N. Engl. J. Med. 338:1265–71 12. Rosenberg SA. 1999. A new era for cancer immunotherapy based on the genes that encode cancer antigens. Immunity 10:281–87 13. Boon T, Coulie PG, Van den Eynde B. 1997. Tumor antigens recognized by T cells. Immunol. Today 18:267–68 14. Uzzo RG, Rayman P, Bloom T. 1999. Mechanisms of apoptosis in T-cells from patients with renal cell carcinoma. Clin. Cancer Res. 5:1219–29 15. Li X, Liu J, Park JK, et al. 1994. T cells from renal cell carcinoma patients exhibit an abnormal pattern of kappa B-specific DNA binding activity: a preliminary report. Cancer Res. 54:5424–29 16. Finke JH, Zea AH, Stanley J, et al. 1993. Loss of T-cell receptor zeta chain and p56lck in T-cells infiltrating human renal cell carcinoma. Cancer Res. 53:5613– 16 17. Hahne M, Rimoldi D, Schroter M, et al. 1996. Melanoma cell expression of Fas(apo-1/CD95) ligand: implications for tumor immune escape. Science 274:1363– 66 18. Gattoni-Celli S, Kirsh K, Timpane R, et al. 1992. Beta 2 microglobulin is mutated in human colon cancer cell line (SCT) deficient in the expression of HLA class I antigens on the cell surface. Cancer Res. 52:1201–4 19. Mathe G, Amiel JL, Schwarzenberg L, et al. 1965. Adoptive immunotherapy of acute leukemia: experimental and clinical results. Cancer Res. 25:1525–31 20. Weiden P, Flournoy N, Thomas E, et al. 1979. Antileukemic effects of graft-versushost disease in human recipients of allogeneic marrow grafts. N. Engl. J. Med. 300:1068–73 21. Weiden PL, Sullivan KM, Flournoy N,

22.

23.

24.

25.

26.

27.

28.

29.

30.

et al. 1981. Antileukemic effect of chronic graft-versus-host disease: contribution to improved survival after allogeneic marrow transplantation. N. Engl. J. Med. 304:1529– 33 Horowitz M, Gale RP, Sondel P, et al. 1990. Graft-versus-leukemia reactions after bone marrow transplantation. Blood 75:552–62 Lee M, Champlin R, Kantarjian H, et al. 1992. Detection of minimal residual disease by polymerase chain reaction of bcr/abl transcripts in chronic myelogenous leukemia following allogeneic bone marrow transplantation. Br. J. Haematol. 82:708–14 Kolb HJ, Mittermueller J, Clemm C, et al. 1990. Donor leukocyte transfusions for treatment of recurrent chronic myelogenous leukemia in marrow transplant patients. Blood 76:2462–65 Kolb HJ, Schattenberg A, Goldman JM, et al. 1995. Graft-versus-leukemia effect of donor lymphocyte transfusions in marrow grafted patients. European group for Blood and Marrow Transplantation Working Party Chronic Leukemia. Blood 86:2041–50 van Besien KW, de Lima M, Giralt SA, et al. 1997. Management of lymphoma recurrence after allogeneic transplantation: the relevance of graft-versus-lymphoma effect. Bone Marrow Transplant. 19:977–82 Verdonck L, Lokhorst H, Dekker A, et al. 1996. Graft-versus-myeloma effect in 2 cases. Lancet 347:800–1 Jones RJ, Ambinder RF, Piantadosi S, et al. 1991. Evidence of a graft-versuslymphoma effect associated with allogeneic bone marrow transplantation. Blood 77:649–53 Heslop HE, Brenner MK, Rooney CM, et al. 1994. Donor T cells to treat EBVassociated lymphoma. N. Engl. J. Med. 331:679–80 Papadopoulos EB, Ladanyi M, Emanuel D, et al. 1994. Infusions of donor leukocytes to treat EBV associated lymphoproliferative disorders after allogeneic bone marrow

15 Dec 2003

15:12

AR

AR206-ME55-26.tex

AR206-ME55-26.sgm

LaTeX2e(2002/01/18)

ALLOGENEIC IMMUNOTHERAPY IN SOLID TUMORS

31.

Annu. Rev. Med. 2004.55:459-475. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

32.

33.

34.

35.

36.

37.

38.

39.

transplantation. N. Engl. J. Med. 330:1185– 91 Mehta J, Powles R, Singhal S, et al. 1996. Clinical and hematologic response of chronic lymphocytic and prolymphocytic leukemia persisting after allogeneic bone marrow transplantation with the onset of acute graft-versus-host disease: possible role of graft-versus-leukemia. Bone Marrow Transplant. 17:371–75 van der Harst D, Goulmy E, Falkenburg F, et al. 1994. Recognition of minor histocompatibility antigens on lymphocytic and myeloid leukemic cells by cytotoxic T-cell clones. Blood 83:1060–66 Ruggeri L, Capanni M, Casucci M, et al. 1999. Role of natural killer cell alloreactivity in HLA-mismatched hematopoietic stem cell transplantation. Blood 94:333–39 Appelbaum FR. 2001. Haematopoietic cell transplantation as immunotherapy. Nature 411:385–89 Truitt RL, Atasoylu AA. 1991. Contribution of CD4+ and CD8+ T cells to graft-versus-host disease and graft-versusleukemia reactivity after transplantation of MHC compatible bone marrow. Bone Marrow Transplant. 8:51–58 Lewalle P, Hensel N, Guimaraes A, et al. 1996. Helper and cytotoxic lymphocyte responses to chronic myeloid leukaemia: implications for adoptive immunotherapy with T cells. Br. J. Haematol. 92:587–94 Faber LM, Van Luxemburg-Heijs SAP, Willemze R, et al. 1992. Generation of leukemia-reactive cytotoxic T lymphocyte clones from the HLA-identical bone marrow donor of a patient with leukemia. J. Exp. Med. 176:1283–89 Goldman JM, Gale RP, Horowitz MM, et al. 1988. Bone marrow transplantation for chronic myelogenous leukemia in chronic phase: increased risk of relapse associated with T-cell depletion. Ann. Intern. Med. 108:806–11 Goulmy E. 1997. Human minor histocompatibility antigens: new concepts for marrow transplantation and adoptive im-

40.

41.

42.

43.

44.

45.

46.

47.

48.

P1: GBC

473

munotherapy. Immunol. Rev. 157:125– 40 Warren EH, Greenberg PD, Riddell SR. 1998. Cytotoxic T-lymphocyte-defined human minor histocompatibility antigens with a restricted tissue distribution. Blood 91:2197–207 Dickinson AM, Wang XN, Sviland L, et al. 2002. In situ dissection of the graft-versushost activities of cytotoxic T cells specific for minor histocompatibility antigens. Nat. Med. 8:410–14 Mutis T, Gillespie G, Schrama E, et al. 1999. Tetrameric HLA class I-minor histocompatibility antigen peptide complexes demonstrate minor histocompatibility antigen-specific cytotoxic T lymphocytes in patients with graft-versus-host disease. Nat. Med. 5:839–42 Huber D, Phillipp J, Fontana A, et al. 1992. Protease inhibitors interfere with transforming growth factor beta-dependent but not transforming growth factorindependent pathway of tumor cell mediated immunosuppression. J. Immunol. 148:277–84 Moscovitch M, Slavin S. 1984. Antitumor effects of allogeneic bone marrow transplantation in (NZB X NZW) F1 hybrids with spontaneous lymphosarcoma. J. Immunol. 132:997–1003 Morecki S, Moshel Y, Gelfand Y, et al. 1997. Induction of graft vs. tumor effect in a murine model of mammary carcinoma. Int. J. Cancer 71:59–63 Morecki S, Yacovlev E, Diab A, et al. 1998. Allogeneic cell therapy for murine mammary carcinoma. Cancer Res. 58:3891– 95 Ladenstein R, Lasset C, Hartmann O, et al. 1994. Comparison of auto versus allografting as consolidation of primary treatments in advanced neuroblastoma over one year of age at diagnosis: report from the European Group for Bone Marrow Transplantation. Bone Marrow Transplant. 14(1):37– 46 Ben-Yosef R, Or R, Nagler A, et al.

15 Dec 2003

15:12

474

49.

Annu. Rev. Med. 2004.55:459-475. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

50.

51.

52.

53.

54.

55.

AR

CHILDS

AR206-ME55-26.tex

¥

AR206-ME55-26.sgm

LaTeX2e(2002/01/18)

P1: GBC

BARRETT

1996. Graft-versus-tumour and graft-versus-leukaemia effect in patient with concurrent breast cancer and acute myelocytic leukaemia. Lancet 348(9036):1242–43 Eibl B, Schwaighofer H, Nachbaur D, et al. 1996. Evidence of a graft-versus-tumor effect in a patient treated with marrow ablative chemotherapy and allogeneic bone marrow transplantation for breast cancer. Blood 88:1501–8 Ueno NT, Rondon G, Mirza NQ, et al. 1998. Allogeneic peripheral-blood progenitorcell transplantation for poor-risk patients with metastatic breast cancer. J. Clin. Oncol. 16:986–93 Giralt S, Estey E, Albitar M, et al. 1997. Engraftment of allogeneic hematopoietic progenitor cells with purine analog-containing chemotherapy: harnessing graft versus leukemia without myeloablative therapy. Blood 89:4531–36 Slavin S, Nagler A, Naparastak E, et al. 1998. Nonmyeloablative stem cell transplantation and cell therapy as an alternative to conventional bone marrow transplantation with lethal cytoreduction for the treatment of malignant and nonmalignant hematologic diseases. Blood 91:756– 63 Khouri IF, Keating M, Korbling M, et al. 1998. Transplant-lite: induction of graft versus malignancy using fludarabine-based nonablative chemotherapy and allogeneic blood progenitor-cell transplantation as treatment for lymphoid malignancies. J. Clin. Oncol. 16:2817–24 Childs R, Clave E, Contentin N, et al. 1999. Engraftment kinetics after nonmyeloablative allogeneic peripheral blood stem cell transplantation: full donor Tcell chimerism precedes alloimmune responses. Blood 94:3234–41 Sykes M, Preffer F, McAfee S, et al. 1999. Mixed lymphohaemopoietic chimerism and graft-versus-lymphoma effects after non-myeloablative therapy and HLA mismatched bone-marrow transplantation. Lancet 353:1755–59

56. McSweeney PA, Niederwieser D, Shiruzu JA. 2001. Hematopoietic cell transplantation in older patients with hematologic malignancies: replacing high-dose cytotoxic therapy with graft-versus-tumor effects. Blood 97(11):3390–400 57. Champlin R, Khouri I, Shimoni A, et al. 2000. Harnessing graft-versusmalignancy: non-myeloablative preparative regimens for allogeneic haematopoietic transplantation, an evolving strategy for adoptive immunotherapy. Br. J. Haematol. 111:18–29 58. Bumpus HC. 1928. The apparent disappearance of pulmonary metastasis in a case of hypernephroma following nephrectomy. J. Urol. 20:185 59. Finke JH, Rayman P, Hart L, et al. 1994. Characterization of TIL subsets from human renal cell carcinoma: specific reactivity defined by cytotoxicity, INF-γ secretion and proliferation. J. Immunother. 15:91 60. Childs R, Clave E, Tisdale J, et al. 1999. Successful treatment of metastatic renal cell carcinoma with a nonmyeloablative allogeneic peripheral-blood progenitor-cell transplant: evidence for a graft-versustumor effect. J. Clin. Oncol. 17(7):2044–49 61. Childs R, Chernoff A, Contentin N, et al. 2000. Regression of metastatic renal-cell carcinoma after nonmyeloablative allogeneic peripheral-blood stem-cell transplantation. N. Engl. J. Med. 343:750–58 62. Rini B, Zimmerman TM, Stadler W, et al. 2002. Allogeneic stem-cell transplantation of renal cell cancer after nonmyeloablative chemotherapy: feasibility, engraftment, and clinical results. J. Clin. Oncol. 20(8):2017–24 63. Bregni M, Dodero A, Peccatori J, et al. 2002. Nonmyeloablative conditioning followed by hematopoietic cell allografting and donor lymphocyte infusions for patients with metastatic renal and breast cancer. Blood 99(11):4234–36 64. Hentschke P, Barkholt L, Uzunel M, et al. 2003. Low-intensity conditioning and hematopoietic stem cell transplantation in

15 Dec 2003

15:12

AR

AR206-ME55-26.tex

AR206-ME55-26.sgm

LaTeX2e(2002/01/18)

ALLOGENEIC IMMUNOTHERAPY IN SOLID TUMORS

65.

Annu. Rev. Med. 2004.55:459-475. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

66.

67.

68.

69.

70.

patients with renal and colon carcinoma. Bone Marrow Transplant. 31(4):253–61 Sandmaier B, Masood N, Thompson J, et al. 2002. Fludarabine and total body irradiation (TBI), followed by allogeneic hematopoietic stem cell transplant (HSCT) as immunotherapy for metastatic renal cell carcinoma. Proc. Am. Soc. Clin. Oncol. A1677 Makimoto A, Mineishi S, Tanosaki R, et al. 2001. Nonmyeloablative stem cell transplantation (NST) for refractory solid tumors. Proc. A. Soc. Clin. Oncol. 20:44 Childs RW, Mena OJ, Tykodi S, et al. 2002. Minor histocompatibility antigens (mHa) are expressed on renal cell carcinoma (RCC) cells and are potential targets for a graft-vs-tumor effect (GVT) following allogeneic blood stem cell transplantation (SCT). Proc. A. Soc. Clin. Oncol. 21:433a (Abstr.) Warren EH, Tykodi SS, Murata M, et al. 2002. T-cell therapy targeting minor histocompatibility Ags for the treatment of leukemia and renal-cell carcinoma. Cytotherapy 4(5):441 Mena O, Igarashi T, Srinivasan R, et al. 2001. Immunologic mechanisms involved in the graft versus tumor effect in renal cell carcinoma (RCC) following nonmyeloablative allogeneic peripheral blood stem cell transplantation. Blood 98:356 (Abstr.) Pedrazzoli P, Da Prada G, Giorgiani G, et al. 2002. Allogeneic blood stem cell transplantation after reduced intensity, preparative regimen—a pilot study in patients with refractory malignancies. Cancer 94(9):2409–16

P1: GBC

475

71. Childs R, Bradstock K, Gottlieb DJ, et al. 2002. Non-myeloablative allogeneic stem cell transplantation (NST) for metastatic melanoma: nondurable chemotherapy responses without clinically meaningful graft-vs-tumor (GVT) effects. Blood 100:429a (Abstr.) 72. Bay JO, Choufi B, Pomel C, et al. 2000. Potential allogeneic graft-versus-tumor effect in a patient with ovarian cancer. Bone Marrow Transplant. 25:681–82 73. Bay JO, Fleury J, Choufi B, et al. 2002. Allogeneic hematopoietic stem cell transplantation in 2002; ovarian carcinoma: results of five patients. Bone Marrow Transplant. 30:95–102 74. Amrolia PJ, Muccioli-Casadei G, Yvon E, et al. 2003. Selective depletion of donor allo-reactive T-cells without loss of antiviral or anti-leukemic responses. Blood 102:2292–99 75. Solomon SR, Tran T, Carter CS, et al. 2002. Optimized clinical-scale culture conditions for ex vivo selective depletion of hostreactive donor lymphocytes: a strategy for GvHD prophylaxis in allogeneic PBSC transplantation. Cytotherapy 4(5):395–406 76. Ruggeri L, Capanni M, Urbani E, et al. 2002. Effectiveness of donor natural killer cell alloreactivity in mismatched hematopoietic transplants. Science 295:2097–100 77. Igarashi T, Srinivasan R, Wynberg J, et al. 2002. Generation of allogeneic NK cells with selective cytotoxicity to melanoma and renal cell carcinoma based on KIRligand incompatibility. Blood 100:73a (Abstr.)

12 Dec 2003

19:43

AR

AR206-ME55-27.tex

AR206-ME55-27.sgm

LaTeX2e(2002/01/18) P1: GBC 10.1146/annurev.med.55.091902.104249

Annu. Rev. Med. 2004. 55:477–503 doi: 10.1146/annurev.med.55.091902.104249 c 2004 by Annual Reviews. All rights reserved Copyright ° First published online as a Review in Advance on Sept. 15, 2003

RITUXIMAB: Expanding Role in Therapy for Lymphomas and Autoimmune Diseases

Annu. Rev. Med. 2004.55:477-503. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

William Rastetter, Arturo Molina, and Christine A. White IDEC Pharmaceuticals Corporation, 3030 Callan Road, San Diego, California 92121; email: [email protected]

Key Words monoclonal antibody, immunotherapy, CD20 antigen, B-cell depletion ■ Abstract Rituximab (Rituxan® ) is a human-mouse chimeric monoclonal antibody that targets the B-cell CD20 antigen and causes rapid and specific B-cell depletion. Rituximab was approved in the United States in 1997 to treat low-grade or follicular, relapsed or refractory, CD20-positive B-cell non-Hodgkin’s lymphoma (NHL). Since then, further clinical experience with rituximab has been incorporated into the prescribing information, which now stipulates an extended eight-week schedule, treatment of patients with refractory or relapsed bulky disease measuring >10 cm, and retreatment of patients who responded to rituximab previously. In 1998, the European Union approved rituximab (MabThera® ) to treat stage III/IV, follicular, chemotherapy-resistant, or relapsed NHL. Recently, the European Union also approved the use of rituximab in combination with standard chemotherapy for aggressive NHL. Many clinical trials have evaluated rituximab, alone or with other therapies, in indolent and aggressive NHL as well as other B-cell lymphoproliferative disorders. New studies are evaluating rituximab’s role in first-line therapy, maintenance therapy, and stem-cell transplantation procedures. The use of rituximab against autoimmune disorders, such as rheumatoid arthritis, immune thrombocytopenic purpura, autoimmune hemolytic anemia, systemic lupus erythematosus, and multiple sclerosis, is also under investigation.

INTRODUCTION Rituximab (Rituxan® ) was the first monoclonal antibody approved by the US Food and Drug Administration for the treatment of malignancy. Since its approval in 1997 for the treatment of relapsed or refractory, low-grade or follicular, CD20 antigen-positive, B-cell non-Hodgkin’s lymphoma (NHL), more than 300,000 patients worldwide have been treated with rituximab either as a single agent or in combination with other therapies. As of April 2003, 78 countries have approved rituximab for the treatment of lymphoma. In the European Union, rituximab was approved under the trade name MabThera® to treat stage III/IV, follicular, chemotherapy-resistant, or relapsed (≥2 relapses) NHL and, in combination with conventional combination chemotherapy, to treat aggressive B-cell NHL. In 2001, 0066-4219/04/0218-0477$14.00

477

12 Dec 2003

19:43

478

AR

AR206-ME55-27.tex

RASTETTER

¥

MOLINA

¥

AR206-ME55-27.sgm

LaTeX2e(2002/01/18)

P1: GBC

WHITE

the US rituximab label was expanded to include an extended eight-week schedule in the original indication, treatment of patients with refractory or relapsed, bulky (largest lesion >10 cm), low-grade or follicular NHL, and rituximab retreatment (1–4). Rituximab is currently under investigation as a single agent and in combination with various other therapies for treatment of other B-cell lymphoproliferative and nonmalignant disorders.

Annu. Rev. Med. 2004.55:477-503. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Mechanisms of Action of Rituximab Rituximab is a genetically engineered, chimeric murine/human monoclonal antibody containing human IgG1 heavy-chain and kappa light-chain constant region sequences and murine variable region sequences (5). Rituximab binds specifically to the CD20 antigen, a 35-kDa transmembrane protein, which is involved in cellcycle progression and differentiation (6, 7). The CD20 antigen is expressed in the vast majority of B-cell NHLs (8). The CD20 antigen is also expressed on normal B lymphocytes, from pre-B cells to activated B cells, but not on differentiated plasma cells, T cells, hematopoietic stem cells, or nonhematopoietic normal tissues (8). Rituximab treatment causes rapid depletion of CD20-positive B cells in the peripheral blood. Despite B-cell depletion, antibody production is maintained by plasma cells, and normal peripheral B cells are subsequently replenished by hematopoietic stem cells in most patients 3–12 months after therapy (9). The mechanisms of action of rituximab are thought to include antibodydependent cellular cytotoxicity, complement-mediated cell lysis, induction of apoptosis, inhibition of cell growth, and sensitization to chemotherapy (5, 10–15). A genetic dimorphism in the gene that encodes the Fc gamma receptor IIIa (Fcγ RIIIa) results in either a phenylalanine or a valine residue at amino acid position 158, located in the region of the receptor that binds to IgG1. These polymorphisms correspond with differences in activation of antibody-dependent cellular cytotoxicity by rituximab. The homozygous valine Fcγ RIIIa (Fcγ RIIIa-158V) has a higher affinity for human IgG1 and mediates increased antibody-dependent cellular cytotoxicity in vitro relative to homozygous phenylalanine Fcγ RIIIa (Fcγ RIIIa-158F) or heterozygous (Fcγ RIIIa-158F-carrier) receptors. Cartron et al. identified the Fcγ RIIIa phenotype in 49 follicular NHL patients treated with rituximab as first-line therapy (16). The objective response rate in 10 patients with the homozygous Fcγ RIIIa-158V genotype was significantly higher at 2 months (100%) and at 12 months (90%) than response rates in 17 patients with homozygous Fcγ RIIIa-158F receptors (67%) and 22 patients with heterozygous receptors (51%) (p = 0.03). Similar results were reported in patients with relapsed follicular NHL (17) and in patients with Waldenstrom’s macroglobulinemia (18). In another study, the Fcγ RIIIa genotype correlated with the degree of B-cell depletion induced by rituximab in patients with systemic lupus erythematosus (19). In vitro studies have demonstrated rituximab sensitization of lymphoma cell lines to the cytotoxic and apoptotic effects of various therapeutic agents, such as cisplatin, fludarabine, vinblastine, and doxorubicin (20–22). A potential mechanism of sensitization is downregulation of interleukin (IL)-10, resulting in decreased

12 Dec 2003

19:43

AR

AR206-ME55-27.tex

AR206-ME55-27.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:477-503. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

RITUXIMAB

P1: GBC

479

activity of Stat3 protein, a transcriptional activator, which decreases the expression of bcl-2, an inhibitor of apoptosis (22). A recent study suggests that rituximab downregulation of IL-10 expression may be mediated by inhibition of p38 mitogen-activated protein kinase activity (23). Additional in vitro studies have demonstrated synergistic apoptotic effects between rituximab and the glucocorticoids dexamethasone or hydrocortisone in NHL cell lines (24, 25). Preliminary results from a recent study demonstrated the possibility of predicting rituximab response in lymphoid malignancies by using cDNA microarray analysis (26). A complete understanding of rituximab’s mechanisms of action awaits further investigation. However, it is clear that the actions of this therapeutic monoclonal antibody are distinct from conventional chemotherapeutic agents used in the treatment of lymphoma. Thus, rituximab may work synergistically with chemotherapeutic agents without added or overlapping toxicity.

RITUXIMAB MONOTHERAPY FOR LYMPHOID MALIGNANCIES Rituximab was initially approved for treatment of relapsed or refractory, low-grade or follicular NHL. Clinical studies using rituximab as a single agent for other histologic subtypes of lymphoma have been conducted in untreated patients and patients who have relapsed despite prior therapy. Table 1 summarizes the efficacy results from selected trials evaluating rituximab monotherapy. Clinical efficacy was assessed by evaluating the overall response rate (ORR), complete response (CR) rate, partial response (PR) rate, duration of response (DR), time to progression, time to treatment failure (TTF), and in one study, minor response rate (MR) (27). Molecular response, when evaluated, was assessed by clearance of the chromosomal translocation t(14;18) involving a rearrangement of the bcl-2 gene, which can be detected using the polymerase chain reaction (PCR) technique. This translocation results in overexpression of the bcl-2 protein, an inhibitor of apoptosis.

First-Line Therapy In a study of 49 previously untreated patients with low-grade, follicular lymphoma and low tumor burden (nodal or extranodal masses ≤7 cm), a standard four-week course of rituximab produced an ORR of 73% and a CR rate of 26% (28). On day 50 after treatment, molecular remissions were achieved in the peripheral blood of 57% of patients and in the bone marrow of 31% of patients. Progression was observed within one year in only 1 of 17 patients (6%) who achieved a molecular response in peripheral blood, compared with 8 of 13 (62%) patients who did not achieve a molecular remission. A statistically significant correlation between molecular remission and clinical response was demonstrated (28, 29). In this trial, patients who achieved a molecular remission had a significantly higher median progression-free survival (744 days) than those who did not (364 days, p < 0.005). In another trial, 37 newly diagnosed patients responded to a standard four-dose

12 Dec 2003

19:43

480

AR

AR206-ME55-27.tex

RASTETTER

¥

MOLINA

¥

AR206-ME55-27.sgm

LaTeX2e(2002/01/18)

P1: GBC

WHITE

Annu. Rev. Med. 2004.55:477-503. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

TABLE 1 Rituximab monotherapy for lymphoid malignancies (single regimen: 375 mg/m2, weekly × 4) Reference

Indication

N(Pts)b

Response

28

Newly diagnosed follicular NHL with low tumor burden

49c

ORR = 73% (20% CR, 6% CRu, 47% PR) Median DR = not available

30

Newly diagnosed follicular NHL

37

ORR = 61% (25% CR) Median TTP: 20 months

32

Newly diagnosed and relapsed or refractory mantle zone lymphoma (MALT type)

35

ORR = 74% (49% CR, 25% PR) Median DR = not reported

37

Newly diagnosed and relapsed mantle cell, immunocytoma, and small B-cell lymphocytic lymphoma

120c

ORR = 30% (8% CR, 22% PR) Median DR in mantle cell: 1.2 years

38

Newly diagnosed and relapsed or refractory Waldenstrom’s macroglobulinemia

30b

ORR = 60% (27% PR, 33% MR) Median TTF in responders: 8 months

43

Relapsed or refractory follicular NHL

70

ORR = 46% (3% CR, 43% PR) Median DR = 11 months

39

Relapsed or refractory low-grade or follicular B-cell NHL

166

ORR = 48% (6% CR, 42% PR) Median DR = 11.2 months

44

Recurrent indolent lymphoma

34c

ORR = 59% (24% CR, 35% PR) Median DR = not available Median TTP = 16 months

2

Relapsed or refractory, bulky disease, low-grade NHL

31

ORR = 39% (3% CR, 36% PR) Median DR = 5.9 months

53

Relapsed or refractory CLL

30

ORR = 23% (23% PR) Median DR = 20 week.

a

Abbreviations: CLL, chronic lymphocytic leukemia; CR, complete response rate; CRu, complete response (unconfirmed); DR, duration of response; MR, minor response rate; NHL, non-Hodgkin’s lymphoma; ORR, overall response rate; pts, patients; TTF, time to treatment failure; TTP, time to progression.

b c

Patients in the intent-to-treat population are listed, unless otherwise noted.

Evaluable patients only.

course of rituximab; the ORR was 61% with a 25% CR rate (30). Responses to first-line therapy with a four-week course of rituximab were also observed in other histologic subtypes of NHL (27, 31–38).

Relapsed or Refractory Disease The standard course of four weekly infusions of 375 mg/m2 rituximab in the pivotal study of 166 patients resulted in a 48% ORR with 6% CRs and a median

12 Dec 2003

19:43

AR

AR206-ME55-27.tex

AR206-ME55-27.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:477-503. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

RITUXIMAB

P1: GBC

481

DR of 11.2 months (39, 40). Twenty-nine percent of patients with chemotherapyresistant disease and 51% of patients who failed anthracycline therapy responded to a standard course of rituximab (39, 40). When the International Workshop Response Criteria were applied to this study, the ORR was 56% and the combined CR and unconfirmed CR(CRu) rate was 32% (41, 42). Other studies have confirmed the efficacy of rituximab monotherapy in relapsed or refractory indolent NHL (43, 44), including bulky disease with lesions measuring >10 cm (2). Rituximab treatment of relapsed or refractory lymphoma subsets, such as extranodal marginal zone lymphoma, mantle cell lymphoma (MCL), and diffuse large-cell lymphoma and other aggressive lymphomas has resulted in ORRs between 22% and 37% (32, 36, 45–47). Significant responses to rituximab, including CRs, have been observed in patients with diffuse large-cell lymphoma and MCL relapsing after, or refractory to, autologous stem-cell transplantation (48, 49). The impressive single-agent activity and favorable safety profile of rituximab in these settings have prompted investigations of extended and maintenance dosing schedules and combination studies with chemotherapeutic agents designed to improve treatment outcomes for NHL.

Alternative Dosing Schedules Investigations aimed at increasing the response rate to rituximab have included evaluations of higher doses and frequency of dosing (Table 2). Dosing of eight weekly infusions in 37 patients with relapsed or refractory, follicular NHL yielded a 57% ORR with 14% CRs and a median DR of 13.4+ months (3). A study by the Swiss Group for Clinical Cancer Research evaluated 202 patients with newly diagnosed or relapsed follicular NHL who responded or had stable disease 12 weeks after a standard course of rituximab. Patients were randomized to observation or to extended therapy with one additional infusion of rituximab every two months for a total of four additional infusions. A statistically significant improvement in DR and event-free survival (EFS) was demonstrated in patients who received extended therapy. The median DR and EFS were 36 and 23 months in the extended-therapy arm, respectively, compared with 16 and 12 months in the observation arm, respectively (50). The difference was more pronounced in chemotherapy-naive patients, with an EFS of 36 months for the extended-therapy arm and 19 months for the observation arm. Hainsworth et al. performed a study in which previously untreated patients with follicular or small lymphocytic NHL (SLL) were restaged 6 weeks after induction therapy with a standard 4-infusion course of rituximab. Those who responded or had stable disease were treated with 4-week maintenance courses of rituximab every 6 months for a maximum of 4 maintenance courses or until disease progression (35, 51). After the rituximab induction course, the ORR in 60 patients was 47%, with a 7% CR rate. However, after the first maintenance course, the ORR in 46 patients was 73%, with a 37% CR rate. Progression-free survival in patients who received maintenance therapy was 34 months. A lower incidence of toxicities was observed during rituximab maintenance treatments than during the induction phase.

12 Dec 2003

19:43

482

AR

AR206-ME55-27.tex

RASTETTER

¥

MOLINA

¥

AR206-ME55-27.sgm

LaTeX2e(2002/01/18)

P1: GBC

WHITE

TABLE 2 Rituximab monotherapy for lymphoid malignancies: alternative regimensa

Annu. Rev. Med. 2004.55:477-503. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Reference Indication

Dose and schedule

3

Relapsed or refractory 375 mg/m2, weekly × 8 follicular, low-grade NHL

45

Relapsed or refractory aggressive NHL

33

N(Pts)b Response 37

ORR = 57% (14% CR, 43% PR) Median DR = 13.4+ months

375 mg/m2, weekly × 8

57c

ORR = 37% (12% CR, 25% PR) Median DR = not available

Newly diagnosed and relapsed or refractory CLL or other mature B-cell lymphoid leukemias

4 weekly doses with dose escalation Dose 1: 375 mg/m2, Doses 2–4: 500 to 2250 mg/m2

45c

All pts: ORR = 40% Median TTP in responders: 8 months CLL pts at highest dose (N = 10c): 75% ORR (all PR)

34

Newly diagnosed and relapsed SLL/CLL

Randomized: 2 cohorts, 33 Dose 1: 100 mg/m2, followed by 250 or 375 mg/m2, 3×/wk × 4 weeks

4

Rituximab-relapsed low-grade or follicular NHL

Second or third course of rituximab in responders, 375 mg/m2, weekly × 4

57c

ORR = 40% (11% CR, 30% PR) Median DR = 17.8 months (Kaplan-Meier estimate)

51

First-line in low-grade follicular NHL and SLL

375 mg/m2, weekly × 4 At 6 weeks, if not PD, repeat courses every 6 months (not to exceed 4 repeats)

60b

After first course: ORR = 47% (7% CR, 40% PR) ≥1 course (N = 46): ORR = 73 % (37% CR, 36% PR) Median PFS = 34 months

50d

Newly diagnosed and relapsed or refractory follicular NHL

Randomized, 2 cohorts 375 mg/m2, weekly × 4 followed by: 1) Observation or 2) Single infusion, months 3, 5, 7, and 9

185c

128 relapsed/refractory pts: ORR = 46% 57 newly diagnosed pts: ORR = 67% EFS in responders: 36 months if maintenance 16 months if observation

31

Newly diagnosed and relapsed or refractory Waldenstrom’s macroglobulinemia

375 mg/m2 weekly × 4 After 3 months, if not PD, repeat course

27

ORR = 44% (44% PR) Median TTP: 16 months

27

Newly diagnosed and relapsed or refractory Waldenstrom’s macroglobulinemia

375 mg/m2 weekly × 4 After 3 months, if not PD, repeat course

22c

ORR = 73% (50% PR, 23% MR) Median TTF in responders: not reached

46

Relapsed or refractory aggressive lymphoma, including diffuse largecell lymphoma (DLCL) and mantle cell

Randomized: 2 cohorts 1) 375 mg/m2, weekly × 8 or 2) 375 mg/m2 on day 1, followed by 500 mg/m2, weekly × 7

52c

All patients: ORR = 33% (10% CR, 23% PR) No difference between treatment arms

ORR = 45% (3% CR, 42% PR) Median DR: 10 months

a

Abbreviations: CLL, chronic lymphocytic leukemia; EFS, event-free survival; SLL, small-cell lymphocytic lymphoma; PD, progressive disease; PFS, progression-free survival; other abbreviations as defined in Table 1.

b

Patients in the intent-to-treat population are listed, unless otherwise noted.

c

Evaluable patients only.

d

Data updated from presentation at the American Society of Hematology 44th Annual Meeting, 2002.

12 Dec 2003

19:43

AR

AR206-ME55-27.tex

AR206-ME55-27.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:477-503. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

RITUXIMAB

P1: GBC

483

Because serum levels of rituximab were statistically higher in responders than in nonresponders, and serum levels and response rates were lower in SLL patients than in follicular NHL patients (52), some investigators evaluated higher rituximab doses for treatment of SLL and chronic lymphocytic leukemia (CLL). Additionally, CD20 density is lower on SLL/CLL cells than on follicular NHLs (53, 54), suggesting a possible role for an alternative treatment schedule. Two separate trials evaluated dose intensity to determine whether increasing serum rituximab concentrations could overcome low CD20 density and increase response rates (33, 34). A 3×-weekly dose of rituximab for 4 weeks yielded an ORR of 45% in CLL patients with previously untreated and relapsed disease (34). In a separate study of 40 relapsed CLL patients, an ORR of 75% (all PR) was achieved with a total rituximab dose of 2250 mg/m2 compared with an ORR rate of 43% (all PR) achieved with a total dose of 1000–1500 mg/m2 (33). Subsequent efforts to treat CLL are focusing on rituximab-chemotherapy combinations. Patients who have responded to rituximab can respond to retreatment at relapse. In a study of 60 patients with relapsed or refractory, low-grade or follicular NHL who had responded to a previous course of rituximab in the pivotal trial, a 38% ORR (10% CR, 28% PR) was achieved after a second course (55 patients) or third course (5 patients) of the standard rituximab regimen (4). Interestingly, the median estimated DR in these patients was 15 months (range 3–25 months), longer than the 11.2 months reported in the pivotal trial (4). Dosing with an eight-week course of rituximab was evaluated in patients with diffuse large-cell lymphoma and MCL (46). Of 52 evaluable patients, 33% responded (10% CR and 23% PR) to treatment with 8 weekly standard doses of rituximab or 1 standard dose followed by 7 higher doses (500 mg/m2). In a recent phase II study, 57 patients with relapsed, aggressive B-cell NHL were treated with 8 weekly infusions of rituximab and achieved an ORR of 37% (45).

Safety of Rituximab Rituximab is generally well tolerated by most patients. The majority of adverse events are grade 1 or 2 and are infusion-related. The first infusion may cause a syndrome of fever, chills, and occasional hypotension and dyspnea. Severe infusion reactions may require interruption of the infusion, which in most cases can be resumed at a reduced rate when the symptoms have completely resolved. Premedications, such as antihistamines and corticosteroids, may be required in some instances. Patients with preexisting cardiac or pulmonary conditions, those with prior clinically significant cardiopulmonary adverse events, and those with high numbers of circulating malignant cells (≥25,000/mm3) or high tumor burden should be closely monitored during all infusions. In rare cases, infusion reactions have resulted in death or tumor lysis syndrome associated with acute renal failure, hyperkalemia, hypocalcemia, hyperuricemia, or hyperphosphatemia. The risk of tumor lysis syndrome appears to be higher in patients with high numbers of circulating malignant cells (≥25,000/mm3) or high tumor burden. Readministration

12 Dec 2003

19:43

484

AR

AR206-ME55-27.tex

RASTETTER

¥

MOLINA

¥

AR206-ME55-27.sgm

LaTeX2e(2002/01/18)

P1: GBC

WHITE

Annu. Rev. Med. 2004.55:477-503. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

of rituximab in conjunction with prophylactic therapy has been tolerated in some patients who experienced tumor lysis syndrome (55). In very rare instances, severe mucocutaneous reactions, some with fatal outcome, have been reported. These include paraneoplastic pemphigus, StevensJohnson syndrome, and toxic epidermal necrolysis (55). Rituximab can be safely readministered in patients who develop recurrent disease after an initial response (56).

RITUXIMAB IN COMBINATION WITH ALKYLATOR- OR ANTHRACYCLINE-BASED CHEMOTHERAPY FOR LYMPHOID MALIGNANCIES Based on in vitro studies that demonstrate synergistic effects between rituximab and certain chemotherapeutic agents, various clinical trials have combined rituximab with chemotherapy (Table 3). The Groupe d’Etude des Lymphomes de l’Adulte (GELA) reported an improvement in CR rate, EFS, and overall survival by adding rituximab to the CHOP regimen (57). This randomized clinical trial of CHOP with or without rituximab included 400 elderly patients (60–80 years old) with previously untreated diffuse large-B-cell lymphoma. The patients were randomized to receive 8 cycles of CHOP every 3 weeks or 8 cycles of CHOP plus rituximab, with rituximab given on day 1 of each cycle. The CR rate achieved in patients who received the combination treatment (76%) was higher than the CR rate in those treated with CHOP alone (63%) (p = 0.005). With a median follow-up of 2 years, median EFS (Figure 1) was significantly longer in patients treated with rituximab plus CHOP compared with CHOP alone (p < 0.001). The combination resulted in a significant increase in survival (Figure 2) compared with CHOP alone (p = 0.007). At 2 years posttreatment, 70% of patients treated with the combination were alive, compared with 57% of those treated with CHOP alone (57). In contrast to previous randomized trials, this is the first time an improvement in EFS and overall survival has been demonstrated with a combination regimen other than CHOP. Consistent with these results, a phase II study of CHOP and rituximab in 33 patients with untreated, aggressive lymphoma demonstrated an ORR of 97% (58, 59). At a median followup of 62 months, disease-free survival in this study was 82% and overall survival was 88%. The combination of rituximab and CHOP is also effective as therapy for previously untreated and relapsed indolent NHL. A phase II trial of 6 cycles of CHOP in combination with 6 standard infusions of rituximab was conducted in 40 patients with newly diagnosed (N = 31) or relapsed or refractory (N = 9) low-grade or follicular NHL. All treated patients responded, with a CR rate of 63% in patients completing CHOP/rituximab treatment. The median progression-free survival has not been reached as of 65 months of observation (60). Other studies of rituximab added to the CHOP regimen also suggest or confirm an improved response rate in patients with newly diagnosed follicular NHL,

12 Dec 2003

19:43

AR

AR206-ME55-27.tex

AR206-ME55-27.sgm

LaTeX2e(2002/01/18)

P1: GBC

RITUXIMAB

485

TABLE 3 Rituximab in combination with alkylator- or anthracycline-based chemotherapy for lymphoid malignanciesa

Annu. Rev. Med. 2004.55:477-503. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

References Indication

Dose and scheduleb

N(Pts)c

Response

142, 143

Newly diagnosed and (R + CHOP) × 6 relapsed or refractory low-grade B-cell lymphoma

38

All pts: ORR = 95% Completely treated pts, (N = 35): ORR = 100% (63% CR, 37% PR) Median DR: 64 months

61

Newly diagnosed CHOP vs R + CHOP follicular, mantle-cell lymphomas or immunocytoma

272

CHOP alone: ORR = 85% R + CHOP: ORR = 95% CHOP for mantle cell: ORR = 69% R + CHOP for mantle cell: ORR = 97%

62

Newly diagnosed (R + CHOP) × 6 mantle-cell lymphoma

40

ORR = 96% (48% CR, 48% PR) Median PFS: 16.6 months

58, 59

Newly diagnosed, advanced aggressive NHL

(R + CHOP) × 6

33

ORR = 97% (61% CR, 36% PR) PFS at 5 yrs = 80%

57

Newly diagnosed, Randomized: diffuse large-B-cell 1) (R + CHOP) × 8 or lymphoma: elderly pts 2) CHOP × 8

56

Relapsed low-grade and aggressive lymphoma

Rituximab retreatment of 59 previous responders (375 mg/m2, weekly × 4) with or without chemotherapy

Median TTP after first R = 12 months Median TTP after second R = 20 months

63

Newly diagnosed aggressive NHL

(R + CHOP, with liposomal vincristine) × 6–8

66

ORR = 100% (94% CR, 5% CRu, 1% PR)

64

HIV-related high-grade (R + CHOP) × 6 lymphoma

50d

CR + CRu: 80% PR: 8%

65

Newly diagnosed and (R + EPOCHe) × relapsed or refractory minimum of 6 aggressive NHL

34d

Newly diagnosed pts (N = 20): ORR = 85% (85% CR) Relapsed/refractory pts (N = 14): ORR = 85% (64% CR, 21% PR)

68

Newly diagnosed follicular NHL

CHOP × 6 followed by R, 375 mg/m2, weekly × 4 in pts with CR or PR

84

ORR = 72% (54% CR/CRu, 18% PR) 2-year PFS = 76% 2-year overall survival = 95%

67

Newly diagnosed follicular NHL

Randomized: 2 cohorts R 375 mg/m2, weekly × 4, then 1) (R + CHOP) × 3 or 2) (R + CVP) × 3

82d

ORR = 97% (57% CR, 40% PR) PFS at 15 months: 87%

69

Newly diagnosed follicular NHL

CHOP, then R for 128 responders with Bcl-2 positive cells in peripheral blood and/or bone marrow

1) N = 202 R + CHOP: ORR = 82% 2) N = 197 (76% CR/CRu, 7% PR) At 2 yrs, 70% survival CHOP: ORR = 69% (63% CR, 6% PR) At 2 yrs, 57% survival

After CHOP, ORR = 94% (57% CR, 37% PR) 77 Bcl-2 positive responders received R, 63% were Bcl-2 negative at 44 wk.

a

Abbreviations: CHOP: cyclophosphamide/vincristine/doxorubicin/prednisone; CVP: cyclophosphamide/vincristine/prednisone; EPOCH: etoposide/vincristine/doxorubicin/cyclophosphamide/prednisone; PFS, progression-free survival; R, rituximab; other abbreviations as defined in Table 1.

b

Unless otherwise noted, one infusion of rituximab (375 mg/m2) is included in each combination cycle.

c

Patients in the intent-to-treat population are listed, unless otherwise noted.

d

Evaluable patients only.

12 Dec 2003

19:43

Annu. Rev. Med. 2004.55:477-503. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

486

AR

AR206-ME55-27.tex

RASTETTER

¥

MOLINA

¥

AR206-ME55-27.sgm

LaTeX2e(2002/01/18)

P1: GBC

WHITE

Figure 1 Event-free survival (EFS) among 399 patients assigned to chemotherc 2002 Massachusetts apy with CHOP or with CHOP plus rituximab. Copyright ° Medical Society. All rights reserved.

mantle-cell lymphoma, immunocytomas (61, 62), aggressive NHL (63), and HIVrelated high-grade lymphoma (64). The combination of rituximab and EPOCH (etoposide, vincristine, doxorubicin, cyclophosphamide, and prednisone) resulted in an ORR >80% in newly diagnosed and relapsed aggressive lymphoma, including lymphomas associated with HIV infection (65). A recent trial evaluated the combination of rituximab and chlorambucil in 29 patients with previously untreated or relapsed or refractory low-grade NHL. The ORR was 93%, with a 59% CR rate (66). Studies are under way to determine the utility of rituximab given prior to chemotherapy or as maintenance therapy following chemotherapeutic regimens (67–69). An ORR of 97% was achieved in previously untreated follicular NHL treated with rituximab followed by a short, three-cycle course of either CHOP or CVP (cyclophosphamide, vincristine, and prednisone) (67). The Southwest Oncology Group studied six cycles of CHOP followed by a four-infusion course of rituximab. This study of previously untreated patients (N = 84) resulted in an ORR of 72% with 54% CRs; clinical responses improved in 19% of patients after the four-week course of rituximab (68). The Eastern Cooperative Oncology Group (ECOG) is completing a 400-patient study in indolent NHL in which previously untreated patients are treated with

12 Dec 2003

19:43

AR

AR206-ME55-27.tex

AR206-ME55-27.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:477-503. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

RITUXIMAB

P1: GBC

487

Figure 2 Overall survival (OS) among 399 patients assigned to chemotherapy c 2002 Massachusetts with CHOP or with CHOP plus rituximab. Copyright ° Medical Society. All rights reserved.

CVP and then randomized to observation or rituximab maintenance (375 mg/m2 weekly × 4 every 6 months for 2 years) (1). Another ECOG study is being conducted in elderly patients (≥60 years) with diffuse aggressive B-cell lymphoma (Figure 3). Previously untreated patients are randomized to CHOP or CHOP/ rituximab; responders are subsequently randomized to observation versus maintenance with 375 mg/m2 rituximab weekly × 4 every 6 months for 2 years (1). Preliminary results of the two ECOG trials should be available within the next year.

RITUXIMAB IN COMBINATION WITH FLUDARABINE-BASED CHEMOTHERAPY FOR LYMPHOID MALIGNANCIES Table 4 summarizes results from trials investigating the utility of combining rituximab with fludarabine-based regimens (61, 70–81). Concurrent administration of rituximab and fludarabine presented an acceptable toxicity profile in patients with previously untreated or anthracycline- and fludarabine-naive CLL (74).

12 Dec 2003

19:43

Annu. Rev. Med. 2004.55:477-503. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

488

AR

AR206-ME55-27.tex

RASTETTER

¥

MOLINA

¥

AR206-ME55-27.sgm

LaTeX2e(2002/01/18)

P1: GBC

WHITE

Figure 3 Treatment schema for ECOG Study 4494 of CHOP versus CHOP plus rituximab in the front-line treatment of patients (≥60 years) with intermediate-grade NHL. Responders are randomized to maintenance with rituximab or observation. The seventh and eighth cycles of CHOP and the fifth dose of rituximab are optional at the investigator’s discretion.

Fludarabine and rituximab in combination with cyclophosphamide resulted in a higher CR rate in previously untreated CLL patients than any other treatments evaluated (76). Furthermore, molecular remissions in marrow aspirates were achieved for the first time in CLL patients, raising the possibility that CLL may be potentially curable without allogeneic transplant techniques. A recent randomized study investigated the efficacy of fludarabine, cyclophosphamide, and mitoxantrone (FCM) therapy with or without rituximab in lowgrade and mantle-cell lymphomas (61). Rituximab appeared to improve the efficacy of the FCM combination, particularly in patients with mantle-cell lymphoma (Table 4). Additional studies of rituximab-fludarabine combinations for the treatment of follicular NHL and Waldenstrom’s macroglobulinemia have been reported (77, 78). In a study of chemotherapy-naive follicular NHL patients randomized to sequential treatment with fludarabine and mitoxantrone followed by rituximab (FMR) versus CHOP followed by rituximab, subsequent rituximab treatment improved PR to CR in 10 of the 12 patients who had only partially responded to FM (73).

TABLE 4 Rituximab in combination with fludarabine-based chemotherapy for lymphoid malignanciesa References

Indication

Dose and scheduleb

N(Pts)c

Response

70

Newly diagnosed, relapsed, or refractory low-grade lymphoma

1 infusion of R,b then (R + F) × 6

30

ORR = 93% (80% CR/CRu, 13% PR) Median DR: 14+ months (Continued)

12 Dec 2003

19:43

AR

AR206-ME55-27.tex

AR206-ME55-27.sgm

LaTeX2e(2002/01/18)

P1: GBC

RITUXIMAB

489

Annu. Rev. Med. 2004.55:477-503. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

TABLE 4 (Continued) References Indication

Dose and scheduleb

N(Pts)c

Response

78

Newly diagnosed and relapsed Waldenstrom’s macroglobulinemia

R: Weeks 1–4 F: Weeks 5, 9, 13 R: Weeks 17, 18 F: Weeks 19, 23, 27 R: Weeks 30, 31

14

ORR = 86% (1 CR, 8 PR, 3 MR) Median DR = not available

71

Newly diagnosed advanced low-grade NHL

F + M CR → R × 4 wk PR → F + M → R × 4 wk

32

ORR = 97% Median DR = 12 months

61

Relapsed or refractory follicular, mantle-cell lymphomas, or immunocytoma

Randomized: 2 cohorts 1) FCMc × 4 or 2) FCMR × 4

94d

FCM alone: ORR = 58% FCMR: ORR = 83% FCM for mantle cell: ORR = 33% FCMR for mantle cell: ORR = 65%

74

Front-line, relapsed, or (R + F) × 4 refractory, but anthracycline and fludarabine-naive CLL

31d

ORR = 87%, CR = 23%, CRu = 10%, PR = 55% Median follow-up of 54 wk., median DR = 75 wk.

75

Newly diagnosed CLL

FCR × 6 (R at 375 mg/m2 for cycle 1 and 500 mg/m2 for cycles 2–6)

135

(67% CR, 19% nodular PR, 18% PR) Median DR = not available

76

Newly diagnosed CLL

FCR × 6 (R at 375 mg/m2 for cycle 1 and 500 mg/m2 for cycles 2–6)

79d

ORR = 95% (66% CR, 14% nodular PR, 15% PR) Median DR = not available

77

Relapsed, follicular NHL

F + C → R → FCR × 3

52

ORR = 82%, CR = 65% Median DR = 20 months

72, 73

Newly diagnosed follicular NHL

Randomized: 2 cohorts 1) (FM) × 6 or 2) CHOP × 6 If CR or PR, then → R, 375 mg/m2, weekly × 4

93d

FM arm (N = 47): 68% CR, 26% PR CHOP arm (N = 46): 37% CR, 56% PR After addition of R, CR rate for FM was 87% and for CHOP was 76%; molecular response was 59% for FM and 40% for CHOP Median DR = not available

144, 80

Newly diagnosed stage IV indolent lymphoma

Randomized: 2 cohorts, IFN maintenance for both 1) Concurrent: (Rb + FND) × 6, FND × 2 2) Sequential: FND × 8 → R: 375 mg/m2, weekly × 6

Concurrent Molecular responses (12 months) N = 44 similar for both arms: ∼85% Sequential N = 34

79

Newly diagnosed CLL

Randomized: 2 cohorts, 104 1) Concurrent: (Rb + F) × 2 6 → R, 375 mg/m , weekly × 4 2) Sequential: F × 6 → R, 375 mg/m2, weekly × 4

Concurrent (N = 51): ORR = 90% (47% CR, 43% PR); Sequential (N = 53): ORR = 77% (28% CR, 49% PR) Median DR = not available

a

Abbreviations: CHOP: cyclophosphamide/vincristine/doxorubicin/prednisone; FCM, fludarabine/cyclophosphamide/mitoxantrone; FCR, fludarabine/cyclophosphamide/rituximab; FM, fludarabine/mitoxantrone; FND, fludarabine/mitoxantrone/dexamethasone; M, mitoxantrone; R, rituximab; other abbreviations as defined in Table 1.

b c

One standard infusion of rituximab (375 mg/m2) is included in each combination, unless otherwise specified.

Patients in the intent-to-treat population are listed, unless otherwise noted.

d

Evaluable patients only.

12 Dec 2003

19:43

490

AR

AR206-ME55-27.tex

RASTETTER

¥

MOLINA

¥

AR206-ME55-27.sgm

LaTeX2e(2002/01/18)

P1: GBC

WHITE

Supplemental Table 1 (follow the Supplemental Material link from the Annual Reviews home page at http://www.annualreviews.org) summarizes trials investigating rituximab for treatment of Hodgkin’s disease (82, 83) and posttransplant proliferative disorders (84–88).

Annu. Rev. Med. 2004.55:477-503. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

RITUXIMAB USE IN MYELOABLATIVE REGIMENS In an attempt to reduce the risk of relapse after stem-cell transplantation (SCT), rituximab has been evaluated for in vivo purging prior to stem-cell collection and posttransplantation for eradication of minimal residual disease (89–99) (Table 5). A dramatic difference in three-year progression-free survival was observed in a retrospective analysis comparing autologous SCT (ASCT) with rituximab consolidation versus conventional combination chemotherapy in patients with previously untreated mantle-cell lymphoma. In contrast, of 20 who had intensive chemotherapy followed by ASCT and rituximab, 89% were progression-free at three years. Despite some reports of delayed neutropenia, the incidence of infection does not appear to be increased when rituximab is added to SCT regimens (92, 94, 100). Randomized studies are addressing the impact of rituximab as part of SCT.

RITUXIMAB IN COMBINATION WITH CYTOKINES OR OTHER ANTIBODIES Advances in immunotherapy for malignancies include rituximab for treatment of lymphoma, trastuzumab for metastatic breast cancer (approved in the United States), alemtuzumab for treatment of chronic lymphocytic leukemia, edrecolomab for Dukes C colon cancer (approved in Europe), and other antibodies and cytokines currently being evaluated as single agents to treat a variety of carcinomas (101). Combination treatments with rituximab and some of these antibodies and biologic response modifiers are being investigated [see Supplemental Table 2; follow the Supplemental Material link from the Annual Reviews home page at http://www.annualreviews.org (102–111)]. Rituximab is administered as part of the ibritumomab tiuxetan radioimmunotherapy regimen. Ibritumomab tiuxetan (Zevalin® ) is indicated for patients with relapsed or refractory, low-grade, follicular, or transformed B-cell NHL, including patients with rituximab-refractory follicular NHL (102–104, 112–114). Ibritumomab tiuxetan is a radioimmunoconjugate formed by linking the murine anti-CD20 monoclonal antibody, ibritumomab, with the linker-chelator, tiuxetan. Ibritumomab is the murine parent of the rituximab chimeric monoclonal antibody. The ibritumomab tiuxetan radioimmunotherapy regimen includes an initial infusion of rituximab (250 mg/m2) followed by a 5-mCi imaging dose of 111Indium ibritumomab tiuxetan. After 7–9 days, another rituximab infusion is followed by a therapeutic dose (standard dose of 0.4 mCi/kg) of 90Yttrium ibritumomab tiuxetan. The rituximab infusions are administered to improve the biodistribution of the

12 Dec 2003

19:43

AR

AR206-ME55-27.tex

AR206-ME55-27.sgm

LaTeX2e(2002/01/18)

P1: GBC

RITUXIMAB

491

Annu. Rev. Med. 2004.55:477-503. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

TABLE 5 Rituximab use in myeloablative regimensa References Indication

Dose and schedule

N(Pts)b

94c

Relapsed follicular lymphoma

In vivo purge with rituximab and/or alpha-IFN followed by HDT/ASCT

IFN (N = 14); IFN: 71% relapse with R (N = 23); median follow-up (MFU) R + IFN (N = 12) 59 months. R: 30% with MFU 37 months IFN + R: 8% with MFU 10 months

92, 93

Previously untreated HDT/autologous PBSCT, then follicular lymphoma rituximab 375 mg/m2 and newly diagnosed weekly × 4) or relapsed mantle-cell lymphoma

30c

97% clinical CR At 30 months, 97% of pts alive

90, 91c

Previously untreated Retrospective matched-pair mantle-cell lymphoma analysis. Intensive chemotherapy/ASCT/R vs conventional CC

20 pts (ASCT/R) and 40 pts (CC)

At 3 yrs, PFS = 89% for ASCT/R and 29% for ASCT

95

Previously untreated First protocol: Induction with mantle-cell lymphoma CHOP (3 cycles) Second protocol: Induction with Ara-C (×2) + rituximab (×4) + CHOP (3 cycles)

41 in first protocol 25 in second protocol

First protocol: CR 24% before ASCT Second protocol: CR 69% before ASCT

96, 97c

Relapsed B-cell lymphoma

High-dose rituximab (1000 mg/m2) with G/GM-CSF and ASCT

42

Overall survival at 1 yr = 93%

98

Relapsed or refractory NHL

Rituximab/Zevalin (median 14 0.24 mCi/kg) followed by high-dose BEAM chemotherapy and autologous peripheral blood progenitor cell transplant

At 2 years, overall survival = 77%; PFS = 50%

99

Relapsed or refractory NHL

Rituximab/Zevalin (median 75 mCi) followed by VP-16/ cyclophosphamide, then autologous stem-cell transplantation (ASCT)

Overall survival at 8 months = 94%

18

Response

a

Abbreviations: ASCT, autologous stem-cell transplantation; ASCTR, ASCT plus rituximab; BEAM, bcnu/etoposide/ arabinoside/melphalan; CC, combination chemotherapy; CHOP, cyclophosphamide/vincristine/doxorubicin/prednisone; G/GM-CSF, granulocyte/granulocyte macrophage colony stimulating factor; HDT, high-dose therapy; IFN, interferon; MFU, median follow-up; PBSCT, peripheral-blood stem cell transplantation; PFS, progression-free survival.

b c

Patients in the intent-to-treat population are listed, unless otherwise noted.

Evaluable patients only.

radiolabeled antibody. Gamma camera scans are performed at 2–24 h and 48–72 h following the imaging dose to rule out altered biodistribution, which occurs in rare instances (112). Several clinical trials have demonstrated ∼80% response rates to Zevalin in indolent and transformed NHL. Additional studies are under way to evaluate the efficacy of ibritumomab tiuxetan followed by rituximab consolidation or maintenance. Treatment of 38 relapsed or refractory low-grade or follicular NHL patients with a standard course of rituximab, combined with a 12-week course of interferon

12 Dec 2003

19:43

Annu. Rev. Med. 2004.55:477-503. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

492

AR

AR206-ME55-27.tex

RASTETTER

¥

MOLINA

¥

AR206-ME55-27.sgm

LaTeX2e(2002/01/18)

P1: GBC

WHITE

alpha-2a, did not improve ORR. The combination treatment resulted in a longer median DR (22.3 months) and TTP (25.2 months) than rituximab alone (105). Two lymphokines, IL-2 and IL-12, have been combined with rituximab to treat NHL in an effort to augment the cytotoxicity of each agent alone (107, 108). Alemtuzumab is an anti-CD52 antibody, also known as CAMPATH-1H (115), which is approved for treatment of CLL. The combination of alemtuzumab and rituximab was evaluated in patients with CLL and other chronic lymphoid malignancies (109). The reported efficacy of the combination was higher than that seen in other studies with alemtuzumab alone. Apolizumab is an antibody directed against the HLA-DR β antigen. Preliminary results showed that the combination of apolizumab and rituximab had some efficacy in relapsed CLL and NHL (110). The cell-surface antigen CD80 is expressed on a subset of normal B cells and the majority of B-cell lymphoma cells. A recent study evaluated the anti-CD80 antibody, IDEC-114, for treatment of follicular NHL (116). In this dose-escalating study, 3 of 16 patients treated with IDEC-114 achieved a CR; however, time to response was longer in these patients than in those treated with rituximab. Encouraging preclinical studies and safety and efficacy results in patients have led to a dose-finding study combining IDEC-114 with rituximab. A single infusion of rituximab alone is followed two days later by an infusion of IDEC-114 alone. After a safety evaluation, three weekly infusions of the combination are administered. Thalidomide is an immunomodulator (117) and an inhibitor of angiogenesis (118). Following the initial report that thalidomide is active in the treatment of multiple myeloma (119), studies have been initiated to evaluate thalidomide in other lymphoproliferative diseases (120). In a recent study, the combination of rituximab and thalidomide was administered to patients with relapsed or CHOPresistant MCL (121). An ORR of 91% (27% CRs) was achieved after a standard course of rituximab with daily oral thalidomide. Larger studies are needed to evaluate this combination in MCL and other lymphoma subtypes. In another novel treatment for MCL, 5 cycles of an idiotype vaccine were administered to 13 patients 12 weeks after B-cell depletion and cytoreduction with rituximab and EPOCH (111). Follow-up is in progress to assess the efficacy of the vaccine in these patients versus historical controls.

RITUXIMAB FOR NONMALIGNANT INDICATIONS Because of its ability to deplete B cells, rituximab is being evaluated for treatment of various autoimmune disorders. Reduction of B cells decreases de novo antibody production, which may be useful for therapy of antibody-associated autoimmune diseases. Another effect of rituximab is a decline in the number of antigen-presenting B cells, which may decrease activation of T cells involved in mediating autoimmune disorders. Efficacy of rituximab treatment has been reported in autoimmune hemolytic anemia (122, 123), immune thrombocytopenic purpura (124–126), thrombotic

12 Dec 2003

19:43

AR

AR206-ME55-27.tex

AR206-ME55-27.sgm

LaTeX2e(2002/01/18)

RITUXIMAB

P1: GBC

493

TABLE 6 Rituximab for nonmalignant indicationsa

Annu. Rev. Med. 2004.55:477-503. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

References Indication

Rituximab dose and schedule

N(Pts)b Response rates ORR = 87%

122, 123

Pediatric autoimmune hemolytic anemia

375 mg/m2 weekly (median 3 doses)

124

Refractory immune thrombocytopenia

375 mg/m2 weekly × 4 57

ORR = 72%; CR = 32%

126

Chronic immune thrombocytopenia

375 mg/m2 weekly × 4 25

ORR = 52% (20% CR, 20% PR, 12% MR)

127

Refractory thrombotic thrombocytopenic purpura

375 mg/m2 weekly × 4–8

CR = 83%

128

Relapsing/refractory 375 mg/m2 weekly × 4 4 thrombotic thrombocytopenic purpura

140

Rheumatoid arthritis

375 mg/m2 weekly × 4 5

Rituximab was clinically beneficial in 80% of pts

129

Systemic lupus erythematosus

500 mg infusion × 2 6 + cyclophosphamide and high-dose corticosteroids

83% had improved BILAG global score at 6-month evaluation

a

15

6

ORR = 75%

BILAG, British Isles Lupus Assessment Group; other abbreviations as defined in Table 1.

b

Patients in the intent-to-treat population are listed.

thrombocytopenic purpura (127, 128), sytemic lupus erythematosus (129, 130), refractory dermatomyositis (131), Wegener’s granulomatosis (132), cold agglutinin disease associated with indolent lymphoma (133, 134), fludarabine-associated immune thrombocytopenic purpura (135), acquired factor VIII inhibitors (136), and graft-versus-host disease after allogeneic transplantation (137) (Table 6). One report described single cases of patients with idiopathic type II mixed cryoglobulinemia, Goodpasture’s syndrome, and primary biliary cirrhosis who had clinical responses to rituximab (138). Because of its demonstrated activity in a variety of autoimmune disorders, clinical trials of rituximab for the treatment of multiple sclerosis are also under development. A standard course of rituximab treatment in five patients with rheumatoid arthritis resulted in improvement in all patients evaluated using the American College of Rheumatology response criteria (139); reductions or normalization of rheumatoid factor and C-reactive protein were also noted in these patients (140). A large, randomized, double-blind, placebo-controlled study in patients with rheumatoid arthritis evaluated the efficacy and safety of rituximab alone or in combination with methotrexate or cyclophosphamide. This study compared the rituximab-based therapies with a single-agent weekly methotrexate regimen and a placebo control arm. An interim analysis of this trial showed that a short induction regimen with rituximab alone or in combination with either methotrexate or cyclophosphamide produced substantial clinical benefit in rheumatoid arthritis (141). Longer

12 Dec 2003

19:43

494

AR

AR206-ME55-27.tex

RASTETTER

¥

MOLINA

¥

AR206-ME55-27.sgm

LaTeX2e(2002/01/18)

P1: GBC

WHITE

follow-up of this trial and additional studies are warranted to determine the optimal dose and schedule of rituximab in these indications.

Annu. Rev. Med. 2004.55:477-503. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

SUMMARY Rituximab has been used in more than 300,000 patients worldwide either as a single agent or in combination with other therapies. Responses to rituximab therapy have been observed in a number of histologic subtypes of NHL (31, 32, 34–38). Alternative dosing strategies may improve the response rate in several of these lymphoma histologic subtypes. Because rituximab has a unique mechanism of action and may sensitize lymphoma cells to the cytotoxic effects of chemotherapeutic agents without added or overlapping toxicity, this antibody has been readily incorporated into numerous chemotherapy regimens with promising results. For the first time in decades, an improvement in event-free and overall survival in diffuse aggressive lymphoma was demonstrated, using the CHOP-rituximab combination as reported by the GELA randomized trial (Figures 1 and 2). Rituximab has been evaluated for in vivo purging before SCT and as consolidation therapy afterwards in an attempt to reduce the risk of relapse (89–94). Studies are being conducted to determine the utility of rituximab as maintenance therapy in patients with chemotherapyresponsive and rituximab-responsive lymphoma. Rituximab is also administered as part of the ibritumomab tiuxetan radioimmunotherapy regimen. Efficacy of rituximab treatment has been reported in nonmalignant autoimmune diseases and will be investigated in multiple sclerosis.

ACKNOWLEDGMENTS The authors thank Anne Larocca, Tim Wright, and Kim Smith for their editorial and technical assistance with the preparation of this manuscript. The Annual Review of Medicine is online at http://med.annualreviews.org

LITERATURE CITED 1. Grillo-Lopez AJ, Hedrick E, Rashford M, et al. 2002. Rituximab: ongoing and future clinical development. Semin. Oncol. 29:105–12 2. Davis TA, White CA, Grillo-Lopez AJ, et al. 1999. Single-agent monoclonal antibody efficacy in bulky non-Hodgkin’s lymphoma: results of a phase II trial of rituximab. J. Clin. Oncol. 17:1851–57

3. Piro LD, White CA, Grillo-Lopez AJ, et al. 1999. Extended rituximab (antiCD20 monoclonal antibody) therapy for relapsed or refractory low-grade or follicular non-Hodgkin’s lymphoma. Ann. Oncol. 10:655–61 4. Davis TA, Grillo-Lopez AJ, White CA, et al. 2000. Rituximab anti-CD20 monoclonal antibody therapy in non-Hodgkin’s

12 Dec 2003

19:43

AR

AR206-ME55-27.tex

AR206-ME55-27.sgm

LaTeX2e(2002/01/18)

RITUXIMAB

5.

Annu. Rev. Med. 2004.55:477-503. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

6.

7.

8.

9.

10.

11.

12.

13.

lymphoma: safety and efficacy of retreatment. J. Clin. Oncol. 18:3135–43 Reff ME, Carner K, Chambers KS, et al. 1994. Depletion of B cells in vivo by a chimeric mouse human monoclonal antibody to CD20. Blood 83:435–45 Valentine MA, Meier KE, Rossie S, et al. 1989. Phosphorylation of the CD20 phosphoprotein in resting B lymphocytes. Regulation by protein kinase C. J. Biol. Chem. 264:11282–87 Einfeld DA, Brown JP, Valentine MA, et al. 1988. Molecular cloning of the human B cell CD20 receptor predicts a hydrophobic protein with multiple transmembrane domains. EMBO J. 7:711–17 Anderson KC, Bates MP, Slaughenhoupt BL, et al. 1984. Expression of human B cell-associated antigens on leukemias and lymphomas: a model of human B cell differentiation. Blood 63:1424–33 Maloney DG, Grillo-Lopez AJ, White CA, et al. 1997. IDEC-C2B8 (rituximab) anti-CD20 monoclonal antibody therapy in patients with relapsed low-grade nonHodgkin’s lymphoma. Blood 90:2188– 95 Shan D, Ledbetter JA, Press OW. 2000. Signaling events involved in anti-CD20induced apoptosis of malignant human B cells. Cancer Immunol. Immunother. 48:673–83 Taji H, Kagami Y, Okada K, et al. 1998. Growth inhibition of CD20-positive B lymphoma cell lines by IDEC-C2B8 antiCD20 monoclonal antibody. Jpn. J. Cancer Res. 89:748–56 Flieger D, Renoth S, Beier I, et al. 2000. Mechanism of cytotoxicity induced by chimeric mouse human monoclonal antibody IDEC-C2B8 in CD20-expressing lymphoma cell lines. Cell. Immunol. 204:55–63 Pedersen IM, Buhl AM, Klausen P, et al. 2002. The chimeric anti-CD20 antibody rituximab induces apoptosis in B-cell chronic lymphocytic leukemia cells through a p38 mitogen acti-

14.

15.

16.

17.

18.

19.

20.

21.

22.

P1: GBC

495

vated protein-kinase-dependent mechanism. Blood 99:1314–19 Byrd JC, Kitada S, Flinn IW, et al. 2002. The mechanism of tumor cell clearance by rituximab in vivo in patients with B-cell chronic lymphocytic leukemia: evidence of caspase activation and apoptosis induction. Blood 99:1038–43 Clynes RA, Towers TL, Presta LG, et al. 2000. Inhibitory Fc receptors modulate in vivo cytotoxicity against tumor targets. Nat. Med. 6:443–46 Cartron G, Dacheux L, Salles G, et al. 2002. Therapeutic activity of humanized anti-CD20 monoclonal antibody and polymorphism in IgG Fc receptor FcgammaRIIIa gene. Blood 99:754–58 Weng W-K, Levy R. 2002. Analysis of IgG Fc receptor FcgRIIIa polymorphism in relapsed follicular non-Hodgkin’s lymphoma patients treated with rituximab. Blood 100(11):353a (Abstr.) Treon SP, Fox EA, Hansen M, et al. 2002. Polymorphisms in Fc gamma RIIIa (CD16) receptor expression are associated with clinical response to rituximab in Waldenstrom’s macroglobulinemia. Blood 100(11):573a (Abstr.) Anolik JH, Campbell D, Felgar RE, et al. 2003. The relationship of FcgammaRIIIa genotype to degree of B cell depletion by rituximab in the treatment of systemic lupus erythematosus. Arthritis Rheum. 48:455–59 Demidem A, Lam T, Alas S, et al. 1997. Chimeric anti-CD20 (IDEC-C2B8) monoclonal antibody sensitizes a B cell lymphoma cell line to cell killing by cytotoxic drugs. Cancer Biother. Radiopharm. 12:177–85 Alas S, Emmanouilides C, Hanna N, et al. 2000. Rituximab-mediated growth regulation and drug sensitization of B lymphoma: pivotal role of IL-10. Proc. Am. Assoc. Cancer Res. 41:288 (Abstr.) Alas S, Emmanouilides C, Bonavida B. 2001. Inhibition of interleukin 10 by rituximab results in down-regulation of bcl-2

12 Dec 2003

19:43

496

23.

Annu. Rev. Med. 2004.55:477-503. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

24.

25.

26.

27.

28.

29.

30.

31.

AR

AR206-ME55-27.tex

RASTETTER

¥

MOLINA

¥

AR206-ME55-27.sgm

LaTeX2e(2002/01/18)

P1: GBC

WHITE

and sensitization of B-cell non-Hodgkin’s lymphoma to apoptosis. Clin. Cancer. Res. 7:709–23 Vega MI, Huerta S, Garban H, et al. 2002. Involvement of the p38MAPK pathway in the inhibition of IL-10 synthesis and secretion by rituximab in 2F7 NHL cells: role in inhibition of bcl-2 expression and sensitization to chemotherapeutic drugs. Blood 100(11):572a–73a (Abstr.) Rose AL, Smith BE, Maloney DG. 2002. Glucocorticoids and rituximab in vitro: synergistic direct antiproliferative and apoptotic effects. Blood 100:1765– 73 Maloney DG, Smith B, Rose A. 2002. Rituximab: mechanism of action and resistance. Semin. Oncol. 29:2–9 Bohen SP, Troyanskaya OG, Alter O, et al. 2003. Variation in gene expression patterns in follicular lymphoma and the response to rituximab. Proc. Natl. Acad. Sci. USA 100:1926–30 Treon SP, Emmanouilides CE, Kimby E, et al. 2002. Extended rituximab therapy is highly effective and facilitates hematological recovery in patients with lymphoplasmacytic lymphoma (Waldenstrom’s macroglobulinemia, WM). Proc. Am. Soc. Clin. Oncol. 21:268a (Abstr.) Colombat P, Salles G, Brousse N, et al. 2001. Rituximab (anti-CD20 monoclonal antibody) as single first-line therapy for patients with follicular lymphoma with a low tumor burden: clinical and molecular evaluation. Blood 97:101–6 Solal-Celigny P. 2002. Increasing treatment options in indolent non-Hodgkin’s lymphoma. Semin. Oncol. 29:2–6 Witzig TE, Vukov AM, Habermann TM, et al. 2002. Rituximab therapy for patients with newly diagnosed, asymptomatic advanced-stage follicular grade I non-Hodgkin’s lymphoma (NHL): a phase II trial in the North Central Cancer Treatment Group (NCCTG). Blood 100(11):361a (Abstr.) Dimopoulos MA, Zervas C, Zomas A,

32.

33.

34.

35.

36.

37.

38.

39.

et al. 2002. Treatment of Waldenstrom’s macroglobulinemia with rituximab. J. Clin. Oncol. 20:2327–33 Conconi A, Thi´eblemont C, Martinelli G, et al. 2001. Activity of rituximab in extranodal marginal zone lymphomas (MALTtype). Blood 98(11):807a (Abstr.) O’Brien SM, Kantarjian H, Thomas DA, et al. 2001. Rituximab dose-escalation trial in chronic lymphocytic leukemia. J. Clin. Oncol. 19:2165–70 Byrd JC, Murphy T, Howard RS, et al. 2001. Rituximab using a thrice weekly dosing schedule in B-cell chronic lymphocytic leukemia and small lymphocytic lymphoma demonstrates clinical activity and acceptable toxicity. J. Clin. Oncol. 19:2153–64 Hainsworth JD, Litchy S, Burris HA, 3rd, et al. 2002. Rituximab as first-line and maintenance therapy for patients with indolent non-Hodgkin’s lymphoma. J. Clin. Oncol. 20:4261–67 Foran JM, Cunningham D, Coiffier B, et al. 2000. Treatment of mantle-cell lymphoma with rituximab (chimeric monoclonal anti-CD20 antibody): analysis of factors associated with response. Ann. Oncol. 11:S117–S21 Foran JM, Rohatiner AZ, Cunningham D, et al. 2000. European phase II study of rituximab (chimeric anti-CD20 monoclonal antibody) for patients with newly diagnosed mantle-cell lymphoma and previously treated mantle-cell lymphoma, immunocytoma, and small B-cell lymphocytic lymphoma. J. Clin. Oncol. 18: 317–24 Treon SP, Agus TB, Link B, et al. 2001. CD20-directed antibody-mediated immunotherapy induces responses and facilitates hematologic recovery in patients with Waldenstrom’s macroglobulinemia. J. Immunother. 24:272–79 McLaughlin P, Grillo-Lopez AJ, Link BK, et al. 1998. Rituximab chimeric anti-CD20 monoclonal antibody therapy for relapsed indolent lymphoma: half of

12 Dec 2003

19:43

AR

AR206-ME55-27.tex

AR206-ME55-27.sgm

LaTeX2e(2002/01/18)

RITUXIMAB

40.

Annu. Rev. Med. 2004.55:477-503. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

41.

42.

43.

44.

45.

46.

47.

48.

patients respond to a four-dose treatment program. J. Clin. Oncol. 16:2825–33 McLaughlin P, Hagemeister FB, GrilloLopez AJ. 1999. Rituximab in indolent lymphoma: the single-agent pivotal trial. Semin. Oncol. 26:79–87 Cheson BD, Horning SJ, Coiffier B, et al. 1999. Report of an international workshop to standardize response criteria for non-Hodgkin’s lymphomas. J. Clin. Oncol. 17:1244–53 Grillo-Lopez AJ, Cheson BD, Horning SJ, et al. 2000. Response criteria for NHL: importance of ‘normal’ lymph node size and correlations with response rates. Ann. Oncol. 11:399–408 Foran JM, Gupta RK, Cunningham D, et al. 2000. A UK multicentre phase II study of rituximab (chimaeric anti-CD20 monoclonal antibody) in patients with follicular lymphoma, with PCR monitoring of molecular response. Br. J. Haematol. 109:81–88 Walewski J, Kraszewska E, Mioduszewska O, et al. 2001. Rituximab (Mabthera, Rituxan) in patients with recurrent indolent lymphoma: evaluation of safety and efficacy in a multicenter study. Med.Oncol. 18:141–48 Igarashi T, Itoh K, Kobayashi Y, et al. 2002. Phase II and pharmacokinetic study of rituximab with eight weekly infusions in relapsed aggressive B-cell nonHodgkin’s lymphoma (B-NHL). Proc. Am. Soc. Clin. Oncol. 21(1):286a (Abstr.) Coiffier B, Haioun C, Ketterer N, et al. 1998. Rituximab (anti-CD20 monoclonal antibody) for the treatment of patients with relapsing or refractory aggressive lymphoma: a multicenter Phase II study. Blood 92:1927–32 Coiffier B. 2002. Rituximab in the treatment of diffuse large B-cell lymphomas. Semin. Oncol. 29:30–35 Tsai DE, Moore HCF, Hardy CL, et al. 1999. Rituximab (anti-CD20 monoclonal antibody) therapy for progressive intermediate-grade non-Hodgkin’s lym-

49.

50.

51.

52.

53.

54.

55.

56.

57.

P1: GBC

497

phoma after high-dose therapy and autologous peripheral stem cell transplantation. Bone Marrow Transplant. 24:521–26 Pan D, Moskowitz CH, Zelenetz AD, et al. 2002. Rituximab for aggressive nonHodgkin’s lymphomas relapsing after or refractory to autologous stem cell transplantation. Cancer J. 8:371–76 Ghielmini M, Hsu Schmitz S-F, Cogliatti S, et al. 2002. Prolonged treatment with rituximab significantly improves event free survival and duration of response in patients with follicular lymphoma: a randomised SAKK trial. Blood 100(11):161a (Abstr.) Hainsworth JD. 2002. Rituximab as firstline and maintenance therapy for patients with indolent non-Hodgkin’s lymphoma: interim follow-up of a multicenter phase II trial. Semin. Oncol. 29:25–29 Berinstein NL, Grillo-Lopez AJ, White CA, et al. 1998. Association of serum rituximab (IDEC-C2B8) concentration and anti-tumor response in the treatment of recurrent low-grade or follicular non-Hodgkin’s lymphoma. Ann. Oncol. 9:995–1001 Huhn D, von Schilling C, Wilhelm M, et al. 2001. Rituximab therapy of patients with B-cell chronic lymphocytic leukemia. Blood 98:1326–31 Almasri NM, Duque RE, Iturraspe J, et al. 1992. Reduced expression of CD20 antigen as a characteristic marker for chronic lymphocytic leukemia. Am. J. Hematol. 40:259–63 2001. Package insert. Rituxan® (rituximab), IDEC Pharmaceuticals Corporation, San Diego, CA Coiffier B, Bouaffia F, Thieblemont C, et al. 2002. Rituximab re-treatment in Bcell lymphoma patients: efficacy and toxicity in 59 patients treated in one center. Blood 100(11):359a (Abstr.) Coiffier B, Lepage E, Briere J, et al. 2002. CHOP chemotherapy plus rituximab compared with CHOP alone in elderly patients with diffuse large-B-cell

12 Dec 2003

19:43

498

58.

Annu. Rev. Med. 2004.55:477-503. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

59.

60.

61.

62.

63.

64.

AR

AR206-ME55-27.tex

RASTETTER

¥

MOLINA

¥

AR206-ME55-27.sgm

LaTeX2e(2002/01/18)

P1: GBC

WHITE

lymphoma. N. Engl. J. Med. 346:235– 42 Vose JM, Link BK, Grossbard ML, et al. 2001. Phase II study of rituximab in combination with CHOP chemotherapy in patients with previously untreated, aggressive non-Hodgkin’s lymphoma. J. Clin. Oncol. 19:389–97 Vose JM, Link BK, Grossbard ML, et al. 2002. Long-term follow-up of a phase II study of rituximab in combination with CHOP chemotherapy in patients with previously untreated aggressive non-Hodgkin’s lymphoma (NHL). Blood 100(11):361a (Abstr.) Czuczman MS, Grillo-Lopez AJ, McLaughlin P, et al. 2001. Clearing of cells bearing the bcl-2 [t(14;18)] translocation from blood and marrow of patients treated with rituximab alone or in combination with CHOP chemotherapy. Ann. Oncol. 12:109–14 Hiddemann W, Unterhalt M, Dreyling M, et al. 2002. The addition of rituximab (R) to combination chemotherapy (CT) significantly improves the treatment of mantle cell lymphomas (MCL): results of two prospective randomized studies by the German Low Grade Lymphoma Study Group (GLSG). Blood 100(11):92a (Abstr.) Howard OM, Gribben JG, Neuberg DS, et al. 2002. Rituximab and CHOP induction therapy for newly diagnosed mantle-cell lymphoma: molecular complete responses are not predictive of progression-free survival. J. Clin. Oncol. 20:1288–94 Rodriguez MA, Sarris A, East K, et al. 2002. A phase II study of liposomal vincristine in CHOP with rituximab for patients with untreated aggressive Bcell non-Hodgkin’s lymphoma (NHL): a safe and effective combination. Blood 100(11):92a (Abstr.) Boue F, Gabarre J, Gisselbrecht C, et al. 2002. CHOP chemotherapy plus rituximab in HIV patients with high grade

65.

66.

67.

68.

69.

70.

71.

72.

lymphoma—results of an ANRS trial. Blood 100(11):470a (Abstr.) Wilson WH, Gutierrez M, O’Connor P, et al. 2002. The role of rituximab and chemotherapy in aggressive Bcell lymphoma: a preliminary report of dose-adjusted EPOCH-R. Semin. Oncol. 29:41–47 Martinelli G, Laszlo D, Mancuso P, et al. 2002. Rituximab plus chlorambucil in low-grade non Hodgkin’s lymphomas (NHL): clinical results of a phase II study. Blood 100(11):777a (Abstr.) Hainsworth JD, Burris HA, Yardley DA, et al. 2002. Rituximab plus short duration chemotherapy as first-line treatment for follicular non-Hodgkin’s lymphoma (NHL): a Minnie Pearl Cancer Research Network phase II trial. Proc. Am. Soc. Clin. Oncol. 21(2):268a (Abstr.) Maloney D, Press OW, Braziel RM, et al. 2001. A phase II trial of CHOP followed by rituximab chimeric monoclonal antiCD20 antibody for treatment of newly diagnosed follicular non-Hodgkin’s lymphoma: SWOG 9800. Blood 98(11):843a (Abstr.) Rambaldi A, Lazzari M, Manzoni C, et al. 2002. Monitoring of minimal residual disease after CHOP and rituximab in previously untreated patients with follicular lymphoma. Blood 99:856–62 Czuczman MS, Fallon A, Mohr A, et al. 2002. Rituximab in combination with CHOP or fludarabine in low-grade lymphoma. Semin. Oncol. 29:36–40 Gregory SA, Venugopal P, Adler S, et al. 2002. Combined fludarabine, mitoxantrone, and rituximab achieves a high response as initial treatment for advances low grade non-Hodgkin’s lymphoma (LGNHL). Blood 100(11):362a (Abstr.) Zinzani PL. 2001. A randomized trial of fludarabine and mitoxantrone plus rituximab versus CHOP plus rituximab as firstline treatment in patients with follicular lymphoma. Blood 98(11):842a (Abstr.)

12 Dec 2003

19:43

AR

AR206-ME55-27.tex

AR206-ME55-27.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:477-503. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

RITUXIMAB 73. Zinzani PL. 2002. A multicenter randomized trial of fludarabine and mitoxantrone (FM) plus rituximab versus CHOP plus rituximab as first-line treatment in patients with follicular lymphoma (FL). Blood 100(11):93a (Abstr.) 74. Schulz H, Klein SK, Rehwald U, et al. 2002. Phase 2 study of a combined immunochemotherapy using rituximab and fludarabine in patients with chronic lymphocytic leukemia. Blood 100:3115–20 75. Keating M, Manshouri T, O’Brien S, et al. 2002. A high proportion of molecular remission can be obtained with a fludarabine, cyclophosphamide, rituximab combination (FCR) in chronic lymphocytic leukemia (CLL). Blood 100(11):205a (Abstr.) 76. Wierda W, O’Brien S, Albitar M, et al. 2001. Combined fludarabine, cyclophosphamide, and rituximab achieves a high complete remission as initial treatment of chronic lymphocytic leukemia. Blood 98(11):771a (Abstr.) 77. Sacchi S, Tucci A, Merli F, et al. 2002. Phase II study with fludarabine and cyclophosphamide plus rituximab (FC + R) in relapsed follicular lymphoma patients. Blood 100(11):572a (Abstr.) 78. Treon SP, Wasi P, Emmanouilides CA, et al. 2002. Combination therapy with rituximab and fludarabine is highly active in Waldenstrom’s macroglobulinemia. Blood 100(11):211a (Abstr.) 79. Byrd JC, Peterson BL, Park K, et al. 2001. Concurrent rituximab and fludarabine has a higher complete response rate than sequential treatment in untreated chronic lymphocytic lymphoma (CLL) patients: results from CALGB 9712. Blood 98(11):772a (Abstr.) 80. Sarris AH, Jiang Y, Tsimberidou AM, et al. 2002. Quantitative real-time polymerase chain reaction for monitoring minimal residual disease in patients with advanced indolent lymphomas treated with rituximab, fludarabine, mitoxantrone, and dexamethasone. Semin. Oncol. 29:48–55

P1: GBC

499

81. McLaughlin P, Hagemeister FB, Rodriguez MA, et al. 2000. Safety of fludarabine, mitoxantrone, and dexamethasone combined with rituximab in the treatment of stage IV indolent lymphoma. Semin. Oncol. 27:37–41 82. Schulz H, Rehwald U, Reiser M, et al. 2002. Phase-II trial of rituximab in patients with relapsed CD20-positive Hodgkin’s lymphoma: an update from the German Hodgkin’s Lymphoma Study Group (GHSG). Blood 100(11):775a (Abstr.) 83. Younes A, Fayad L, McLaughlin P, et al. 2002. Rituximab plus ABVD therapy for newly diagnosed patients with classical Hodgkin’s disease. Blood 100(11):308b (Abstr.) 84. Ferry CK, Maillard A, Agbalika F, et al. 2002. Treatment of B-lymphoproliferative disorder after hematopoietic stem cell transplantation. Role and immunological follow-up of rituximab® . Blood 100(11):113a–14a (Abstr.) 85. Oertel SHK, Zeidler K, Grefer J, et al. 2002. Monotherapy with the anti-CD20 antibody rituximab (Ritux) in patients with post-transplant lymphoproliferative disease (PTLD): results of a multicentre phase II study. Blood 100(11):573a (Abstr.) 86. van Esser JW, Niesters HG, van der Holt B, et al. 2002. Prevention of EpsteinBarr virus-lymphoproliferative disease by molecular monitoring and preemptive rituximab in high-risk patients after allogeneic stem cell transplantation. Blood 99:4364–69 87. Milpied N, Vasseur B, Parquet N, et al. 2000. Humanized anti-CD20 monoclonal antibody (rituximab) in post transplant Blymphoproliferative disorder: a retrospective analysis on 32 patients. Ann. Oncol. 11:s113–16 88. Choquet S, Herbrecht R, Soci´e G, et al. 2002. Efficacy and safety of rituximab in B-cell post transplantation lymphoproliferative disorders (B-PTLD):

12 Dec 2003

19:43

500

89.

Annu. Rev. Med. 2004.55:477-503. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

90.

91.

92. 93.

94.

95.

96.

AR

AR206-ME55-27.tex

RASTETTER

¥

MOLINA

¥

AR206-ME55-27.sgm

LaTeX2e(2002/01/18)

P1: GBC

WHITE

preliminary results of a multicenter, open label, phase II trial (M39037 TRIAL). Blood 100(11):467a (Abstr.) Mangel J, Buckstein R, Imrie K, et al. 2002. Pharmacokinetic study of patients with follicular or mantle cell lymphoma treated with rituximab as “in vivo purge” and consolidative immunotherapy following autologous stem cell transplantation. Blood 100(11):479b (Abstr.) Mangel J, Leitch HA, Connors JM, et al. 2002. Intensive chemotherapy and autologous stem cell transplantation plus rituximab is superior to conventional chemotherapy for newly diagnosed advanced stage mantle cell lymphoma: a matched pair analysis. Blood 100(11): 780a (Abstr.) Mangel J, Buckstein R, Imrie K, et al. 2002. Immunotherapy with rituximab following high-dose therapy and autologous stem-cell transplantation for mantle cell lymphoma. Semin. Oncol. 29:56–69 Brugger W. 2002. Improving outcomes in transplantation. Semin. Oncol. 29:23–26 Brugger W, Hirsch J, Repp R, et al. 2002. Treatment of follicular and mantle cell lymphoma with rituximab after high-dose chemotherapy and autologous blood stem cell transplantation: a multicenter phase II study. Blood 100(11):644a (Abstr.) Buckstein RJ, Mangel J, Imrie K, et al. 2002. High dose therapy/ASCT consolidated with rituximab and or alpha interferon immunotherapy for relapsed follicular lymphoma prolongs progression free survival and achieves durable molecular remissions. Blood 100(11):647a–48a (Abstr.) Geisler CH, Pedersen LB, Andersen NS, et al. 2002. Mantle cell lymphoma (MCL): increased clinical and molecular response rates adding Ara-C and rituximab to CHOP + BEAM and autologous stem cell transplantation. Results of the 1st and 2nd Nordic MCL Protocols. Blood 100(11):644a (Abstr.) Khouri IF, Saliba RM, Hosing C, et al.

97.

98.

99.

100.

101.

102.

2002. Concurrent administration of highdose rituximab (HD-R) with G/GM-CSF and autologous stem cell transplantation (ASCT) for relapsed B-cell lymphomas: increased efficacy without added toxicity. Blood 100(11):645a (Abstr.) Khouri IF, Saliba RM, Giralt SA, et al. 2001. Nonablative allogeneic hematopoietic transplantation as adoptive immunotherapy for indolent lymphoma: low incidence of toxicity, acute graft-versushost disease, and treatment-related mortality. Blood 98:3595–99 Winter JN, Inwards D, Erwin W, et al. 2002. Zevalin dose-escalation followed by high-dose BEAM and autologous peripheral blood progenitor cell (PBPC) transplant in non-Hodgkin’s lymphoma: early outcome results. Blood 100(11): 411a–12a (Abstr.) Nademanee A, Molina A, Forman SJ, et al. 2002. A phase I/II trial of high-dose radioimmunotherapy (RIT) with Zevalin in combination with high-dose etoposide (VP-16) and cyclophosphamide (CY) followed by autologous stem cell transplant (ASCT) in patients with poor-risk or relapsed B-cell non-Hodgkin’s lymphoma (NHL). Blood 100(11):182a (Abstr.) Horwitz SM, Breslin S, Negrin RS, et al. 2000. Adjuvant rituximab after autologous peripheral blood stem cell transplant (APBSCT) results in delayed immune reconstitution without increase in infectious complications. Blood 96(11):384a (Abstr.) White CA, Weaver RL, Grillo-Lopez AJ. 2001. Antibody targeted immunotherapy for treatment of malignancy. Annu. Rev. Med. 52:125–45 Witzig TE, Gordon LI, Cabanillas F, et al. 2002. Randomized controlled trial of yttrium-90-labeled ibritumomab tiuxetan radioimmunotherapy versus rituximab immunotherapy for patients with relapsed or refractory low-grade, follicular, or transformed B-cell non-Hodgkin’s lymphoma. J. Clin. Oncol. 20:2453–63

12 Dec 2003

19:43

AR

AR206-ME55-27.tex

AR206-ME55-27.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:477-503. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

RITUXIMAB 103. Wiseman GA, Gordon L, Multani PS, et al. 2002. Ibritumomab tiuxetan radioimmunotherapy for patients with relapsed or refractory non-Hodgkin’s lymphoma and mild thrombocytopenia: a phase II multicenter trial. Blood 99:4336– 42 104. Witzig TE, Flinn IW, Gordon LI, et al. 2002. Treatment with ibritumomab tiuxetan radioimmunotherapy in patients with rituximab-refractory follicular non-Hodgkin’s lymphoma. J. Clin. Oncol. 20: 3262–69 105. Davis TA, Maloney DG, Grillo-Lopez AJ, et al. 2000. Combination immunotherapy of relapsed or refractory low-grade or follicular non-Hodgkin’s lymphoma with rituximab and interferon-alpha-2a. Clin. Cancer Res. 6:2644–52 106. Sacchi S, Federico M, Vitolo U, et al. 2001. Clinical activity and safety of combination immunotherapy with IFN-alpha 2a and rituximab in patients with relapsed low grade non-Hodgkin’s lymphoma. Haematologica 86:951–58 107. Friedberg JW, Neuberg D, Gribben JG, et al. 2002. Combination immunotherapy with rituximab and interleukin 2 in patients with relapsed or refractory follicular non-Hodgkin’s lymphoma. Br. J. Haematol. 117:828–34 108. Ansell SM, Witzig TE, Kurtin PJ, et al. 2002. Phase 1 study of interleukin-12 in combination with rituximab in patients with B-cell non-Hodgkin lymphoma. Blood 99:67–74 109. Faderl S, Thomas DA, O’Brien S, et al. 2002. Combined use of alemtuzumab and rituximab in patients with relapsed and refractory chronic lymphoid malignancies—an update of the M.D. Anderson experience. Blood 100(11):206a (Abstr.) 110. Hegde U, White T, Stetler-Stevenson M, et al. 2002. Phase I study of combination rituximab (CD20) and apolizumab (Hu1D10) monoclonal antibody therapy in previously treated B-cell lymphoma

111.

112.

113.

114.

115.

116.

117.

118.

119.

P1: GBC

501

and chronic lymphocytic leukemia. Blood 100(11):358a (Abstr.) Wilson WH, Neelapu S, White T, et al. 2002. Idiotype vaccine following EPOCH-rituximab treatment in untreated mantle cell lymphoma. Blood 100(11):162a (Abstr.) 2002. Package insert. Zevalin® (ibritumomab tiuxetan), IDEC Pharmaceuticals Corporation, San Diego, CA Witzig TE, White CA, Gordon LI, et al. 2002. Updated results of a randomized controlled study of Zevalin radioimmunotherapy (RIT) vs rituximab in B-cell non-Hodgkin’s lymphoma (NHL). Ann. Oncol. 13(Suppl. 2):38 (Abstr.) Witzig TE, White CA, Wiseman GA, et al. 1999. Phase I/II trial of IDECY2B8 radioimmunotherapy for treatment of relapsed or refractory CD20+ B-cell non-Hodgkin’s lymphoma. J. Clin. Oncol. 17:3793–803 Osterborg A, Dyer MJ, Bunjes D, et al. 1997. Phase II multicenter study of human CD52 antibody in previously treated chronic lymphocytic leukemia. European Study Group of CAMPATH-1H treatment in chronic lymphocytic leukemia. J. Clin. Oncol. 15:1567–74 Czuczman M, Witzig TE, Younes A, et al. 2002. IDEC-114, an anti-CD80 monoclonal antibody for relapsed or refractory, follicular NHL: phase I/II study of safety, efficacy, and pharmacokinetics. Blood 100(11):163a (Abstr.) Haslett PA, Corral LG, Albert M, et al. 1998. Thalidomide costimulates primary human T lymphocytes, preferentially inducing proliferation, cytokine production, and cytotoxic responses in the CD8+ subset. J. Exp. Med. 187:1885– 92 D’Amato RJ, Loughnan MS, Flynn E, et al. 1994. Thalidomide is an inhibitor of angiogenesis. Proc. Natl. Acad. Sci. USA 91:4082–85 Singhal S, Mehta J, Desikan R, et al. 1999. Antitumor activity of thalidomide in

12 Dec 2003

19:43

502

120.

Annu. Rev. Med. 2004.55:477-503. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

121.

122.

123.

124.

125.

126.

127.

128.

AR

AR206-ME55-27.tex

RASTETTER

¥

MOLINA

¥

AR206-ME55-27.sgm

LaTeX2e(2002/01/18)

P1: GBC

WHITE

refractory multiple myeloma. N. Engl. J. Med. 341:1565–71 Dimopoulos MA, Zomas A, Viniou NA, et al. 2001. Treatment of Waldenstrom’s macroglobulinemia with thalidomide. J. Clin. Oncol. 19:3596–601 Drach J, Kaufman H, Puespoek A, et al. 2002. Marked anti-tumor activity of rituximab plus thalidomide in patients with relapsed/resistant mantle cell lymphoma. Blood 100(11):162a (Abstr.) Zecca M, Nobili B, Ramenghi U, et al. 2002. Rituximab for the treatment of refractory autoimmune hemolytic anemia in children. Blood 100(11):444a–45a (Abstr.) Zecca M, De Stefano P, Nobili B, et al. 2001. Anti-CD20 monoclonal antibody for the treatment of severe, immune-mediated, pure red cell aplasia and hemolytic anemia. Blood 97:3995– 97 Cooper N, Stasi R, Feuerstein M, et al. 2002. Transient B cell depletion with rituximab, an anti CD20 monoclonal antibody, resulted in lasting complete responses in 16 of 57 adults with refractory immune thrombocytopenic purpura. Blood 100(11):52a (Abstr.) Cooper D, Feuerstein M, McFarland J, et al. 2002. Investigating the mechanism of action of rituximab, an anti-CD20 monoclonal antibody in adults with immune thrombocytopenic purpura. Blood 100(11):479a (Abstr.) Stasi R, Pagano A, Stipa E, et al. 2001. Rituximab chimeric anti-CD20 monoclonal antibody treatment for adults with chronic idiopathic thrombocytopenic purpura. Blood 98:952–57 Tsai H-M, Gutterman LA, Kloster B, et al. 2002. Long-term remission of refractory thrombotic thrombocytopenic purpura after rituximab therapy. Blood 100(11):684a (Abstr.) Ahmad A, Aggarwal A, Sharma D, et al. 2002. Rituximab for the treatment of relapsing/refractory thrombotic

129.

130.

131.

132.

133.

134.

135.

136.

137.

thrombocytopenic purpura (TTP). Blood 100(11):686a (Abstr.) Leandro MJ, Edwards JC, Cambridge G, et al. 2002. An open study of B lymphocyte depletion in systemic lupus erythematosus. Arthritis Rheum. 46:2673–77 Anolik JH, Campbell D, Felgar R, et al. 2002. B lymphocyte depletion in the treatment of systemic lupus (SLE): phase I/II trial of rituximab (Rituxan®) in SLE. Arthritis Rheum. 46(9 Suppl.):S289 (Abstr.) Levine TD. 2002. A pilot study of rituximab therapy for refractory dermatomyositis. Arthritis Rheum. 46(9 Suppl.):S488 (Abstr.) Specks U, Fervenza FC, McDonald TJ, et al. 2001. Response of Wegener’s granulomatosis to anti-CD20 chimeric monoclonal antibody therapy. Arthritis Rheum. 44:2836–40 Cohen Y, Polliack A, Zelig O, et al. 2001. Monotherapy with rituximab induces rapid remission of recurrent cold agglutinin-mediated hemolytic anemia in a patient with indolent lymphoplasmacytic lymphoma. Leuk. Lymphoma 42:1405–8 Berentsen S, Tjonnfjord GE, Brudevold R, et al. 2001. Favourable response to therapy with the anti-CD20 monoclonal antibody rituximab in primary chronic cold agglutinin disease. Br. J. Haematol. 115:79–83 Hegde UP, Wilson WH, White T, et al. 2002. Rituximab treatment of refractory fludarabine-associated immune thrombocytopenia in chronic lymphocytic leukemia. Blood 100:2260–62 Wiestner A, Cho HJ, Asch AS, et al. 2002. Rituximab in the treatment of acquired factor VIII inhibitors. Blood 100:3426–28 Ratanatharathorn V, Carson E, Reynolds C, et al. 2000. Anti-CD20 chimeric monoclonal antibody treatment of refractory immune-mediated thrombocytopenia in a patient with chronic graft-versus-host disease. Ann. Intern. Med. 133:275–79

12 Dec 2003

19:43

AR

AR206-ME55-27.tex

AR206-ME55-27.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:477-503. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

RITUXIMAB 138. Arzoo K, Sadeghi S, Liebman HA. 2002. Treatment of refractory antibody mediated autoimmune disorders with an antiCD20 monoclonal antibody (rituximab). Ann. Rheum. Dis. 61:922–24 139. Arnett FC, Edworthy SM, Bloch DA, et al. 1988. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum. 31:315–24 140. De Vita S, Zaja F, Sacco S, et al. 2002. Efficacy of selective B cell blockade in the treatment of rheumatoid arthritis: evidence for a pathogenetic role of B cells. Arthritis Rheum. 46:2029–33 141. Edwards JCW, Szczepanski L, Filipowicz-Sosnowka A, et al. 2002. Efficacy and safety of rituximab, a B-cell targeted chimeric monoclonal antibody: a randomized, placebo-controlled trial in

P1: GBC

503

patients with rheumatoid arthritis. Arthritis Rheum. 46(9 Suppl.):S197 (Abstr.) 142. Czuczman MS, Grillo-Lopez AJ, White CA, et al. 1999. Treatment of patients with low-grade B-cell lymphoma with the combination of chimeric antiCD20 monoclonal antibody and CHOP chemotherapy. J. Clin. Oncol. 17:268– 76 143. Czuczman M. 2001. Progression free survival (PFS) after six years (median) follow-up of the first clinical trial of rituximab/CHOP chemoimmunotherapy. Blood 98(11):601a (Abstr.) 144. McLaughlin P, Hagemeister FB, Rodriguez MA, et al. 2000. Safety of fludarabine, mitoxantrone, and dexamethasone combined with rituximab in the treatment of stage IV indolent lymphoma. Semin. Oncol. 27:37–41

2 Dec 2003

21:31

AR

AR206-ME55-28.tex

AR206-ME55-28.sgm

LaTeX2e(2002/01/18) P1: GCE 10.1146/annurev.med.55.091902.104440

Annu. Rev. Med. 2004. 55:505–18 doi: 10.1146/annurev.med.55.091902.104440 c 2004 by Annual Reviews. All rights reserved Copyright °

BOTULINUM TOXIN AND OTHER NEW APPROACHES TO MIGRAINE THERAPY

Annu. Rev. Med. 2004.55:505-518. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Avi Ashkenazi and Stephen D. Silberstein Jefferson Headache Center, Thomas Jefferson University Hospital, Philadelphia, Pennsylvania 19107; email: [email protected]

Key Words headache, prevention, antinociception, topiramate, angiotensin blockers ■ Abstract The number of migraine treatments and our understanding of migraine pathophysiology are both increasing. Newer treatments focus on migraine prevention. Botulinum toxin (BTX) is a potent neurotoxin used primarily to treat diseases associated with increased muscle activity. Recently, BTX was found to have antinociceptive effects that are probably independent of its muscle-relaxant action. Clinical trials support the efficacy of BTX type A (and possibly also type B) in the treatment of migraine. The anticonvulsant topiramate was recently shown to be effective for migraine prevention. At the low doses used for this indication, cognitive side effects are not a major concern. Another new approach to migraine prevention is angiotensin type 1 (AT1) receptor blockade. The high tolerability of the AT1 receptor blocker candesartan warrants further studies to assess its role in migraine prevention.

INTRODUCTION Migraine is an episodic neurovascular disorder characterized by repeated attacks of headache, autonomic dysfunction, and gastrointestinal symptoms (1). Some patients experience an aura—transient neurological symptoms that precede attacks and usually last less than 60 minutes. Migraine is a common disorder, with a prevalence of 12% in the adult population (6% in men and 18% in women) and 4% in children. It has a strong impact on quality of life. The World Health Organization has ranked migraine as one of the most disabling of the chronic diseases (2). Significant progress has been made in our understanding of migraine pathophysiology. Migraine is now considered a primary CNS disorder with secondary effects on meningeal blood vessels (1). The treatment of acute migraine attacks has improved dramatically with the introduction of the 5-HT1B/1D agonists, known as the triptans. The prevention of migraine has been neglected and, owing to the relatively low efficacy and high adverse-event rates of previously available preventive medicines, unsatisfactory for many patients. 0066-4219/04/0218-0505$14.00

505

2 Dec 2003

21:31

506

AR

AR206-ME55-28.tex

ASHKENAZI

¥

AR206-ME55-28.sgm

LaTeX2e(2002/01/18)

P1: GCE

SILBERSTEIN

This review describes recent advances in the prevention and treatment of migraine, including neurotoxins as a novel approach to migraine prevention.

Annu. Rev. Med. 2004.55:505-518. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

BOTULINUM TOXIN Botulinum toxin (BTX) is a potent neurotoxin produced by the anaerobic bacterium Clostridium botulinum (3, 4). It causes a dose-dependent muscle relaxation by blocking acetylcholine release at the neuromuscular junction. Systemic human intoxication by BTX, usually caused by oral ingestion of organisms and/or toxin in spoiled food, results in botulism—an acute, potentially fatal muscle paralysis (5). Over the past two decades, locally injected BTX has been used to treat various disorders associated with increased muscle tone (6, 7, 39). Its first clinical use was for strabismus (40). It has also been used extensively for various forms of dystonia, achalasia, spastic bladder, and hyperhidrosis (18, 41, 42). Recent data have shown BTX to be effective in treating various pain syndromes, including neuropathic pain, low back pain, whiplash-associated disorders (43–45), and migraine and other types of chronic headaches (29, 46–49).

Structure and Preparations BTX belongs to the clostridial neurotoxin family and exists as seven antigenically distinct serotypes (A–G) (3). The toxin is noncovalently associated with nontoxic proteins. In its purified form, it is a ∼150-kDa polypeptide that consists of two subunits, a light chain and a heavy chain, linked by a disulfide bond. The light chain acts as a zinc-dependent endopeptidase. The heavy chain contains two domains. One, in the C terminus (HC), is the ganglioside-binding domain, which has a key role in binding the toxin to the target cell membrane and facilitating its internalization. The other, in the N terminus (HN), is the translocation domain, which promotes penetration of the light chain through the endosomal membrane into the cytosol. BTX type A (BTX-A) is the most widely used serotype in clinical practice. It is available in the United States as Botox® (Allergan, Irving, CA) and in Europe as Dysport® (Ipsen, Berkshires, UK). Although these two preparations contain the same serotype, they differ in potency and in antigenicity (4, 8). BTX type B has recently become available for clinical use in the United States (Myobloc® , Elan Pharmaceuticals, San Diego, CA), and in Europe (NeuroBloc® , Elan Pharmaceuticals, CA). Clinical experience with this serotype, however, is still limited. BTX is administered intramuscularly in the affected muscles in dystonia and spasticity. In migraine, it is usually administered in the frontal, temporal, and cervical musculature in the distribution of pain and tenderness.

Immunology Since BTX is a protein of nonhuman origin, it may evoke antibody formation (4, 9). Once neutralizing antibodies are present, the efficacy of the toxin is lost (10).

2 Dec 2003

21:31

AR

AR206-ME55-28.tex

AR206-ME55-28.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:505-518. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

NEW APPROACHES TO MIGRAINE THERAPY

P1: GCE

507

The reported prevalence of treatment resistance due to antibody formation depends on the assay used for antibody detection, the patient population, and the treatment protocol. Treatment resistance to an older formulation was estimated to occur in at least 5% to 10% of patients with cervical dystonia (11). Factors that increase the risk for antibody formation include higher doses and short intervals between doses (12–14). With the new formulation and the relatively low doses used in migraine treatment (see below), this complication is unlikely. Recommendations for minimizing immunoresistance include using the lowest effective dose at the longest possible intervals and avoiding booster injections (4). When resistance to one BTX serotype develops, switching to a different serotype may restore the therapeutic response (10). This response, however, may be only temporary. In a recent study, 10 patients with antibody-mediated therapeutic failure to BTX-A were given BTX-B (NeuroBloc® ). After an initial response, six patients developed secondary therapeutic failure with documented antibodies to BTX-B (15). Three other patients had a stable, continuous response. A new low-molecular-weight BTX-A has demonstrated efficacy in an animal model of spasticity (16). This novel toxin may be less immunogenic than the conventional BTX-A.

Mechanism of Action BTX affects the nervous system through a multistage process that ultimately blocks neurotransmitter release (3, 4, 17). The toxin binds to the target nerve terminal through its HC domain and is subsequently internalized into an endosome. The disulfide bond is then cleaved, and the light chain undergoes translocation to the cytosol. This stage is facilitated by the HN domain of the heavy chain. In the final stage, the enzymatically active light chain cleaves one or more proteins involved in neurotransmitter release. The type of protein cleaved depends on the toxin serotype. BTX-A cleaves a synaptosomal-associated protein of 25 kDa (SNAP25), whereas BTX-B attacks a vesicle-associated membrane protein (VAMP), also called synaptobrevin. In both cases, the protein cleavage prevents synaptic vesicle fusion with the plasma membrane and thus prevents neurotransmitter release. EFFECT AT THE NEUROMUSCULAR JUNCTION The main site of action of BTX is the neuromuscular junction. By interfering with acetylcholine release from the presynaptic axon terminal at this site, BTX causes a dose-dependent and reversible muscle relaxation. Axonal sprouting, which occurs following BTX entrance into the cell, causes termination of the toxin-induced effect in 2–6 months. The eventual wearing off of the toxin-induced effect allows the return of neuromuscular transmission and regression of sprouts. EFFECT ON THE AUTONOMIC NERVOUS SYSTEM Botulism is associated with autonomic nervous system dysfunction (5). BTX affects cholinergic synapses in both sympathetic and parasympathetic pathways. By acting at these sites, the toxin

2 Dec 2003

21:31

508

AR

AR206-ME55-28.tex

ASHKENAZI

¥

AR206-ME55-28.sgm

LaTeX2e(2002/01/18)

P1: GCE

SILBERSTEIN

reduced sweat secretion in patients with hyperhidrosis and improved bladder function in patients with detrusor sphincter dyssynergia (18, 19).

Annu. Rev. Med. 2004.55:505-518. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

EFFECT ON AFFERENT PATHWAYS BTX may also act via muscle-spindle afferent mechanisms (20). In animals, BTX reduced spindle afferent discharges, which suggests a direct effect on gamma motor nerve endings (21). It also caused atrophy of both intrafusal and extrafusal muscle fibers when injected into rats (22). EFFECTS ON THE CENTRAL NERVOUS SYSTEM There is increasing evidence that BTX affects the CNS (23). Earlier animal studies suggested retrograde transport of the toxin into the CNS (24). More recently, however, it was shown that only breakdown products of the toxin were transported in a retrograde manner (25). Janicki & Habermann found that BTX inhibits the release of methionine-enkephalin-like material in rat striatum in vitro (26). However, little evidence currently indicates that functional toxin penetrates the CNS in humans. The central effects of BTX are more likely to result from CNS neuroplasticity, induced by alterations in afferent input. Several studies support this hypothesis. Byrnes et al. found that BTX can reverse changes in cortical motor representation of the upper limb in patients with writer’s cramp (27). More recently, Gilio et al. showed that BTX-A normalizes intracortical circuits in patients with dystonia (28). ANTINOCICEPTIVE EFFECTS Although BTX has primarily been used for the treatment of dystonia and other disorders associated with increased muscle tone, its analgesic effect has long been recognized (29–31). In many patients, the toxin’s analgesic effect occurs earlier and lasts longer than its effect on muscular hyperactivity (4). Pain reduction may also be observed in the absence of weakness. Cui & Aoki examined the effect of BTX-A on the pain behavior of rats after formalin injection (32). Five and twelve days before formalin injection, rats were treated with BTX-A injected subcutaneously to the hind paw. BTX-A inhibited the delayed nociceptive response, as assessed by reduced pain behavior, at both time points. This effect was achieved at doses that did not cause muscle weakness. Lew et al. studied the analgesic effect of BTX-B, given in three different doses, in 122 patients with idiopathic cervical dystonia (33). Using the Toronto Western Spasmodic Torticollis Rating Scale (TWSTRS) pain score, 61%–83% of patients treated with BTX-B responded and the response was dose-dependent. In contrast, a study of healthy volunteers showed that BTX-A, injected intradermally, reduced neurogenic flare induced by electrical stimulation but had little effect on acute pain and allodynia (34). This model is of acute, rather than chronic, pain and may have little relevance to disorders associated with sensitization. However, several studies have shown little or no direct effect of BTX on acute cutaneous nociception (34, 35). The analgesic effect of BTX therefore cannot be explained solely by the reduction of muscle tone by the toxin. Other mechanisms must be involved, but their nature remains speculative. Several theories have been proposed (31):

2 Dec 2003

21:31

AR

AR206-ME55-28.tex

AR206-ME55-28.sgm

LaTeX2e(2002/01/18)

NEW APPROACHES TO MIGRAINE THERAPY

P1: GCE

509

Annu. Rev. Med. 2004.55:505-518. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

1. By decreasing prolonged muscle contraction, BTX may reduce the release of various substances that sensitize muscle nociceptors. 2. Through its effect on muscle spindle activity, BTX can indirectly disrupt muscle pain associated with abnormal muscle contraction. Because the spindle afferents have important supraspinal projections, the change in their firing pattern caused by BTX may also cause changes in sensory processing at higher levels of the nervous system. 3. BTX may suppress neurogenic inflammation, which has been implicated in the pathogenesis of migraine and other pain syndromes. 4. BTX may affect the release of neurotransmitters other than acetylcholine. There is evidence that it inhibits the release of substance P in vitro (36, 37). More evidence for this theory comes from an in vivo study showing that the effect of BTX on pain behavior in the rat formalin model is associated with reduced glutamate release (38).

BOTULINUM TOXIN FOR THE TREATMENT OF MIGRAINE The beneficial effect of BTX-A in migraine treatment was first noted in patients who were given the toxin for the treatment of facial wrinkles (50). Since then, an increasing number of studies have examined the efficacy of BTX-A, and more recently BTX-B, as migraine-preventive drugs (Table 1).

Botulinum Toxin Type A EFFICACY AND TOLERABILITY Binder et al. conducted an open-label study to examine the efficacy of BTX-A for acute and preventive treatment of migraine (50). Treatment protocols were individualized. Of 77 migraine patients who were treated prophylactically, 51% reported complete relief from migraine symptoms for a mean duration of 4.1 months. Partial response (>50% reduction in headache frequency or severity), with a mean response duration of 2.7 months, was achieved in 38% of patients. Of 10 migraine patients who were treated acutely, 70% reported complete response, with improvement occurring 1–2 h after treatment. A double-blind, vehicle-controlled study evaluated the efficacy of BTX-A for migraine prevention (51). The study included 123 patients with International Headache Society (IHS)-defined migraine with or without aura. Patients were randomized to receive a single administration of either BTX-A (at a dose of 25 units or 75 units) or vehicle. BTX-A was injected into the frontalis, temporalis, and glabellar muscles. At the 25-unit dose, BTX-A reduced migraine frequency by 1.88 attacks per month (compared with 0.98 attacks per month for placebo) for three months following treatment. BTX-A at this dose also reduced migraine severity and migraine-associated vomiting. The 75-unit dose was not significantly more effective than placebo. This was attributed to the lower baseline headache

2 Dec 2003

21:31

510

AR

AR206-ME55-28.tex

ASHKENAZI

¥

AR206-ME55-28.sgm

LaTeX2e(2002/01/18)

P1: GCE

SILBERSTEIN

Annu. Rev. Med. 2004.55:505-518. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

TABLE 1 Studies of botulinum toxin (BTX) for migraine prevention (n = number of patients) BTX serotype and total dose

Study design

n

Results

Reference

Prospective, double-blind, placebo-controlled

123 Type A, 25 U or 75 U

25 U: significant reduction in migraine frequency 75 U: no significant efficacy

51

Prospective, double-blind, placebo-controlled

56 Type A, dose variable

Significant reduction in migraine frequency, severity, and duration

52

Prospective, placebo-controlled

30 Type A, 50 U

Significant reduction in migraine frequency and duration

53

Prospective, open-label

29 Type A, 50 U

Complete elimination of headaches in 55% of patients, significant improvement in additional 28%

54

Prospective, double-blind, placebo-controlled

32 Type A, 200 U

Significant reduction in migraine-related disability and in consumption of acute-pain medications; no change in number of days with headache

55

Prospective, open-label

77 Type A, dose variable

Complete elimination of headaches in 51% of patients, significant improvement in additional 38%

50

Prospective, open-label

21 Type B, 5000 U Significant reduction in headache frequency and severity

63

Prospective, open-label

36 Type B, 5000 U Improvement in 66% of patients

64

frequency of patients who received this dose. BTX was well tolerated; transient side effects, lasting days to weeks, included blepharoptosis, diplopia, and injection-site weakness. Another double-blind, placebo-controlled study examined the effect of BTX-A on 56 migraine patients (52). BTX was injected into the frontalis and temporalis muscles; the outcome measures were attack frequency, attack duration, and pain intensity. The maximal effect was found at week 12 following treatment. BTX-A reduced migraine frequency by 1.8 attacks per month (compared with a reduction of 0.2 attack per month for placebo). It also reduced mean headache severity by 4.0 points on a scale of 0–10, compared with a decrease of 0.2 points for placebo. Headache duration was reduced in the BTX group by an average of 15.2 h, compared with a mean reduction of 5.6 h for placebo. Barrientos & Chana examined prospectively the efficacy and safety of BTX-A in the prophylactic treatment of migraine (53). Thirty patients with IHS-defined migraine were randomized to receive either placebo or BTX-A at a dose of 50 units, injected into six sites in the head and neck. BTX-A treatment significantly decreased the number of monthly days with headache, from 5.7 at baseline to 2.5 at 90 days post-treatment. It also reduced migraine frequency, duration of

2 Dec 2003

21:31

AR

AR206-ME55-28.tex

AR206-ME55-28.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:505-518. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

NEW APPROACHES TO MIGRAINE THERAPY

P1: GCE

511

migraine attacks, and consumption of acute-pain medications. BTX-A therapy was not associated with any significant adverse effects and was well tolerated. Behmand et al. evaluated prospectively the efficacy of BTX-A, injected into the corrugator muscle, on migraine (54). Twenty-nine patients received 25 units into each corrugator muscle to a total of 50 units. Two months later, 16 patients (55%) reported complete elimination of headaches and 8 (28%) experienced significant improvement (decreased migraine frequency from 6.4 to 2.1 per month and decreased pain intensity from 8.6 to 6.1 points on a scale of 0–10). A double-blind placebo-controlled study evaluated the effect of BTX-A treatment on 32 migraine patients (55). Two treatments, at a dose of 100 units each, were given three months apart. Although the total number of days with headache was not reduced, the character of the pain changed to a moderate, nonthrobbing headache responsive to standard analgesics, and triptan use was reduced. Several other retrospective studies have shown efficacy of BTX-A in migraine prevention (56–58). Other studies of patients with chronic daily headache, many of whom had chronic migraine, showed only a mild or no response to BTX (59, 60). SAFETY More than two decades of clinical use have established BTX-A as a remarkably safe drug. Based on animal studies, the lethal dose in humans is estimated at ∼3000 units (4). The doses used for migraine treatment (25–100 units) are unlikely to be toxic. An antitoxin is available in the event of accidental overdose (61). Permanent muscle paralysis and persistent weakness have not been reported in migraine studies. BTX should be used with caution in patients with neuromuscular junction diseases (e.g., myasthenia gravis). It is contraindicated in patients who take aminoglycosides, which interfere with neuromuscular transmission (62). Because there are limited data on the safety of BTX in pregnant and lactating women, it is not recommended for use in these circumstances. Evidence for the efficacy of BTX-A in migraine treatment is growing. BTX-A offers unique advantages over conventional migraine-preventing drugs: It has a long duration of action of up to four months, and it is safe and well tolerated, with almost no systemic side effects. These properties make BTX-A particularly appealing to patients who do not comply with daily drug treatments or cannot tolerate them. Several questions regarding BTX-A use for migraine remain to be answered: (a) What subgroup of migraine patients will benefit the most from treatment? (b) What is the optimal therapeutic dose? (c) Should the treatment program (dose and injection sites) be standardized or tailored individually for each patient? Large controlled studies, which are under way, may resolve these questions.

Botulinum Toxin Type B Clinical experience with BTX-B is far less extensive than that with BTX-A. Few preliminary studies have assessed its efficacy in headache treatment. Lake & Saper conducted an open-label study on the efficacy of BTX-B in the treatment of

2 Dec 2003

21:31

Annu. Rev. Med. 2004.55:505-518. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

512

AR

AR206-ME55-28.tex

ASHKENAZI

¥

AR206-ME55-28.sgm

LaTeX2e(2002/01/18)

P1: GCE

SILBERSTEIN

21 patients with IHS-defined migraine (63). Patients received a total of 5000 units of BTX-B injected into 11 sites (dosages of BTX-B are 25–50 times those of BTXA). Evaluation was done at baseline and four months following treatment. Mean monthly headache frequency declined from 7.7 preinjection to 4.6 at four months following treatment. Significant improvement also occurred on visual analog scales of headache, sleep, Migraine Disability Assessment Scale (MIDAS) scores, and overall treatment satisfaction. Adverse events were transient and rated as mild by five of six patients who experienced them. An open-label study examined the efficacy of BTX-B in treating transformed migraine (64). The study included 36 patients with at least four migraine episodes in a four-week period. Patients received a dose of 5000 units of BTX-B injected into three or more muscles. The sites of injection were chosen according to pain distribution, trigger points, and frown lines. Twenty-four patients (66%) reported improvement in headache severity as assessed by the numeric rating scale. Headache frequency was also reduced. Adverse events were mild and included dry mouth and transient pain at the injection site. A randomized double-blind placebo-controlled study evaluating the efficacy of BTX-B in patients with chronic headaches, including migraine, is under way (65). BTX-B may be effective for migraine treatment, but clinical experience is still limited. Currently, it may be considered for patients who develop antibodymediated resistance to BTX-A. Larger clinical trials may better define the role of BTX-B in migraine prevention.

OTHER NEW APPROACHES TO MIGRAINE TREATMENT Recent advances in migraine pathophysiology research have led to new concepts in migraine prevention (1, 66). Migraine is currently viewed as a neurovascular disorder with a CNS generator (1). Therefore, the current focus in migraine prevention is on attempting to modulate central neurotransmitter systems.

Topiramate Anticonvulsants are increasingly used for migraine prevention. Topiramate is a structurally unique anticonvulsant derived from D-fructose (67). In addition to seizure prevention, it is used to treat mood disorders and essential tremor (68, 69). Its role in migraine treatment was recently evaluated (70). Topiramate acts via several mechanisms that may be relevant to migraine treatment (71): (a) It has a state-dependent blocking effect on voltage-sensitive sodium and L-type calcium channels; (b) it acts on the γ -amino butyric acid (GABA) type-A receptor to enhance GABA transmission; (c) it reduces excitatory glutamatergic neurotransmission by binding to the α-amino-3-hydroxy-5-methylisoxazole-4-propionic acid (AMPA)/kainate receptor; and (d) it is a weak inhibitor of carbonic anhydrase. These effects of the drug are probably mediated by modulation of receptor and

2 Dec 2003

21:31

AR

AR206-ME55-28.tex

AR206-ME55-28.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:505-518. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

NEW APPROACHES TO MIGRAINE THERAPY

P1: GCE

513

channel phosphorylation. Since activation of neurons within the trigeminocervical complex is probably the biological substrate for pain in migraine, Storer & Goadsby studied the effect of topiramate on trigeminocervical activation in the cat (72). The superior sagittal sinus (SSS) of anesthetized cats was isolated and electrically stimulated to produce a model of trigeminocervical nociceptive activation. Topiramate reduced SSS-evoked firing of neurons in the trigeminocervical complex in a dose-dependent fashion, with the maximum effect seen at 30 min. At this time point, topiramate at 5 mg/kg reduced neuronal firing by 48 ± 5%. Topiramate may directly inhibit the trigeminocervical complex or influence the neural network that controls sensory input. Several recent clinical studies have found topiramate effective in migraine treatment (73–75). Silberstein et al. evaluated the effect of topiramate on migraine in a placebo-controlled trial of 487 patients (the MIGR-001 study) (75). At a daily dose of 100 mg, topiramate reduced the average migraine frequency by 2.1 episodes per month (from 5.4 to 3.3) compared with a corresponding reduction of 0.8 episodes per month by placebo (p = 0.001). Significantly more patients treated with topiramate at 100 mg/day were responders (i.e., they experienced a ≥50% reduction in migraine frequency) compared with placebo-treated patients (54% versus 23%). Topiramate treatment was also associated with a reduction in the mean number of monthly migraine days and a decrease in the consumption of acute-pain medications. The 200-mg dose was not significantly more effective than the 100-mg dose. The onset of drug effect was observed by the end of the first month of treatment. Topiramate was well tolerated. The most common adverse events were paresthesias, taste change, anorexia, fatigue, and nausea. Cognitive adverse events occurred in 19% of patients taking the 100-mg dose but led to withdrawal in only 4%. In contrast to many other migraine-preventive drugs, which cause weight gain, topiramate treatment was associated with an average of 3.8% weight loss. A similarly designed study that included 483 migraine patients (the MIGR-002 study) was recently completed (76). Patients were randomized to receive either topiramate (50, 100, or 200 mg/day) or placebo. Patients treated with topiramate at 100 mg/day had a mean decrease of 2.3 episodes per month in migraine frequency, compared with a decrease of 1.1 episodes per month for patients receiving placebo (p = 0.002). Responder rates (as defined above) were significantly higher in the topiramate 100-mg group (49%) compared with placebo (23%). As in the MIGR-001 study, the 100-mg dose had the most favorable efficacy/tolerability profile. Topiramate was safe and generally well tolerated. Side effects were similar to those found in the previous study. There was an average loss of 3.3% of body weight in the topiramate 100-mg group. The third trial (MIGR-003 study), which compared topiramate to propranolol and placebo, has recently been completed but results are not yet available. Young et al. evaluated the outcome of 74 migraine patients treated with topiramate in a case-series study (73). Twenty-four patients had episodic migraine and 50 had chronic migraine. The mean dose was 208 mg/day, and treatment continued for at least six weeks. Topiramate treatment resulted in a mean decrease of seven days

2 Dec 2003

21:31

Annu. Rev. Med. 2004.55:505-518. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

514

AR

AR206-ME55-28.tex

ASHKENAZI

¥

AR206-ME55-28.sgm

LaTeX2e(2002/01/18)

P1: GCE

SILBERSTEIN

per month with headache, from 21.6 to 13.6 (p < 0.0001). The responder rate to topiramate (as defined above) was 44.6% (58.3% for episodic migraine and 38.0% for chronic migraine). Headache severity was also significantly reduced. Adverse events were usually mild to moderate and included paresthesias, cognitive difficulties, dizziness, and nausea. Patients with comorbid depression had a similar outcome to those who were not depressed. In another retrospective study to evaluate the efficacy of topiramate in migraine treatment, the drug was given as add-on therapy for chronic migraine patients and as monotherapy for episodic migraine patients (74). The mean daily dose of topiramate was 87.5 mg and the mean duration of treatment was 8.4 months. In patients with chronic migraine, mean days with migraine per 28 days decreased from 6.3 to 3.7. Headache severity, consumption of acute-pain medications, and MIDAS-scale values also decreased significantly in this group. Patients with episodic migraine also had a significant decrease in headache severity and in monthly migraine days (from 5.8 to 1.9 per 28 days). In this study, which also included cluster-headache patients, topiramate was well tolerated, with only 8/178 patients discontinuing treatment. Several other studies have demonstrated topiramate’s efficacy in migraine prevention (77, 78). The necessary doses (100–200 mg/day) are significantly lower than those used for epilepsy, mitigating the concern of cognitive side effects. Absence of the weight gain seen with many other preventive migraine treatments is also a major benefit.

Angiotensin II Receptor Blockade Several reports on the efficacy of angiotensin-converting enzyme (ACE) inhibitors for migraine prevention have been published (79, 80). Angiotensin receptor antagonists block the renin-angiotensin system without the common side effects caused by ACE inhibitors (e.g., coughing and angioneurotic edema). A recent placebocontrolled study including 60 patients evaluated the efficacy of candesartan, an angiotensin II type 1 (AT1) receptor blocker, in migraine prevention (81). Patients received candesartan at 16 mg/day or placebo for two 12-week periods in a cross-over design. The mean number of days with headache during a 12-week period was significantly lower with candesartan than with placebo (13.6 versus 18.5). Candesartan also lowered headache severity, level of disability, and days of sick leave. It was very well tolerated, with a tolerability profile similar to that of placebo. This drug’s mechanism of action in migraine prevention is currently unknown. Angiotensin II affects cerebral blood flow through AT1 receptors (82). It also modulates the activity of various neurotransmitters, including serotonin, dopamine, and melatonin (83, 84). Finally, angiotensin may be an indirect activator of nitric oxide (NO) synthase, thereby increasing levels of NO, a molecule that affects nociceptive pathways (85). Blocking angiotensin II activity at any of these sites may prevent migraine. The high tolerability of candesartan compared with many other migraine-preventive drugs warrants further studies of this drug for this indication.

2 Dec 2003

21:31

AR

AR206-ME55-28.tex

AR206-ME55-28.sgm

LaTeX2e(2002/01/18)

NEW APPROACHES TO MIGRAINE THERAPY

P1: GCE

515

Annu. Rev. Med. 2004.55:505-518. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

CONCLUSION The number of drugs available for migraine-preventive treatment is increasing. More data are available regarding the efficacy of various drugs, enabling a more rational, rather than the previously used empiric, therapeutic approach. There is increasing evidence for the efficacy of BTX-A in the treatment of migraine. Its long duration of action and high tolerability make it especially appealing for patients whose compliance on orally administered drugs is poor. The anticonvulsant topiramate is effective for migraine prevention at relatively low doses, minimizing cognitive side effects. Angiotensin II receptor blockade, a new approach to migraine prevention, needs to be further tested. The Annual Review of Medicine is online at http://med.annualreviews.org

LITERATURE CITED 1. Goadsby PJ, Lipton RB, Ferrari MD. 2002. Migraine—current understanding and treatment. N. Engl. J. Med. 346:257– 70 2. World Health Organization. 2001. World Health Report. www.who.int/whr/index. htm. 3. Turton K, Chaddock JA, Acharya KR. 2002. Botulinum and tetanus neurotoxins: structure, function and therapeutic utility. Trends Biochem. Sci. 27:552–58 4. Brin MF. 1997. Botulinum toxin: chemistry, pharmacology, toxicity, and immunology. Muscle Nerve 20:S146–S168 5. Shapiro RL, Hatheway C, Swerdlow DL. 1998. Botulism in the United States: a clinical and epidemiologic review. Ann. Intern. Med. 129:221–28 6. Jankovic J, Brin MF. 1991. Therapeutic uses of botulinum toxin. N. Engl. J. Med. 324:1186–94 7. Jankovic J, Brin ME. 1997. Botulinum toxin: historical perspective and potential new indications. Muscle Nerve 20:S129– S145 8. Silberstein SD. 2001. Review of botulinum toxin type A and its clinical applications in migraine headache. Expert Opin. Pharmacother. 2:1649–54 9. Critchfield J. 2002. Considering the im-

10.

11.

12.

13.

14.

15.

16.

mune response to botulinum toxin. Clin. J. Pain 18:S133–S141 Borodic G, Johnson E, Goodnough M, et al. 1996. Botulinum toxin therapy, immunologic resistance, and problems with available materials. Neurology 46:26–29 Dauer WT, Burke RE, Greene P, et al. 1998. Current concepts on the clinical features, aetioogy and management of idiopathic cervical dystonia. Brain 121:547– 60 Zuber M, Sebald M, Bathien N, et al. 1995. Botulinum antibodies in dystonic patients treated with type A botulinum toxin: frequency and significance. Neurology 43:1715–18 Greene P, Fahn S, Diamond B. 1994. Development of resistance to botulinum toxin type A in patients with torticollis. Mov. Disord. 9:213–17 Jankovic J, Schwartz K. 1995. Response and immunoresistance to botulinum toxin injections. Neurology 45:1743–46 Dressler DW, Bigalke H, Benecke R. 2003. Botulinum toxin type B in antibodyinduced botulinum toxin type A therapy failure. Neurology 60:A465–A466 (Abstr.) Sakamoto T, Asanuma K, Kaji R, et al. 2003. Development of a new low

2 Dec 2003

21:31

516

17.

18.

Annu. Rev. Med. 2004.55:505-518. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

19.

20.

21.

22.

23. 24.

25.

26.

27.

AR

AR206-ME55-28.tex

ASHKENAZI

¥

AR206-ME55-28.sgm

LaTeX2e(2002/01/18)

P1: GCE

SILBERSTEIN

molecular weight botulinum type A neurotoxin preparation for treating muscle hyperactivity. Neurology 60:A466 (Abstr.) Aoki KR. 2001. Pharmacology and immunology of botulinum toxin serotypes. J. Neurol. 248:3–10 Naumann M, Jost WH, Toyka KV. 1999. Botulinum toxin in the treatment of neurological disorders of the autonomic nervous system. Arch. Neurol. 56:914–16 Heckmann M, Ceballos-Baumann AO, Plewig G, for the Hyperhidrosis Study Group. 2001. Botulinum toxin A for axillary hyperhidrosis (excessive sweating). N. Engl. J. Med. 344:488–93 Giladi N. 1997. The mechanism of action of botulinum toxin type A in focal dystonia is most probably through its dual effect on efferent (motor) and afferent pathways at the injection site. J. Neurol. Sci. 152:123– 35 Filippi GM, Errico P, Santarelli R, et al. 1993. Botulinum A toxin effects on rat jaw muscle spindles. Acta Otolaryngol. 113:400–4 Rosales RL, Arimura K, Takenaga S, et al. 1996. Extrafusal and intrafusal muscle effects in experimental botulinum toxin-A injection. Muscle Nerve 19:488–96 Hallett M. 2000. How does botulinum toxin work? Ann. Neurol. 48:7–8 Wiegand H, Wellhoner HH. 1977. The action of botulinum A neurotoxin on the inhibition by antidromic stimulation of the lumbar monosynaptic reflex. Naunyn Schmiedebergs Arch. Pharmacol. 298:235–38 Aoki R. 1998. The development of BOTOX—its history and pharmacology. Pain Digest 8:337–41 Janicki PK, Habermann E. 1983. Tetanus and botulinum toxins inhibit, and black widow spider venom stimulates the release of methionine-enkephalin-like material in vitro. J. Neurochem. 41:395–402 Byrnes ML, Thickbroom GW, Wilson SA, et al. 1998. The corticomotor representation of upper limb muscles in writer’s

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

cramp and changes following botulinum toxin injection. Brain 121:977–88 Gilio F, Curr`a A, Lorenzano C, et al. 2000. Effects of botulinum toxin type A on intracortical inhibition in patients with dystonia. Ann. Neurol. 48:20–26 Gobel H, Heinze A, Kuhn KH, et al. 2001. Botulinum toxin A in the treatment of headache syndromes and pericranial pain syndromes. Pain 91:195–99 Borodic GE, Acquadro M, Johnson EA. 2001. Botulinum toxin therapy for pain and inflammatory disorders: mechanisms and therapeutic effects. Expert Opin. Investig. Drugs 10:1531–44 Arezzo JC. 2002. Possible mechanisms for the effects of botulinum toxin on pain. Clin. J. Pain 18:S125–S132 Cui M, Aoki KR. 2000. Botulinum toxin type A (BTX-A) reduces inflammatory pain in the rat formalin model. Cephalalgia 20:414 Lew MF, Adornato BT, Duane D, et al. 1997. Botulinum toxin type B: a doubleblind, placebo-controlled, safety and efficacy study in cervical dystonia. Neurology 49:701–7 Kramer HH, Angerer C, Erbguth F, et al. 2003. Botulinum toxin A reduces neurogenic flare but has almost no effect on pain and hyperalgesia in human skin. J. Neurol. 250:188–93 Blersch W, Schulte-Mattler WJ, Przywara S, et al. 2002. Botulinum toxin A and the cutaneous nociception in humans: a prospective, double-blind, placebocontrolled, randomized study. J. Neurol. Sci. 205:59–63 Ishikawa H, Mitsui Y, Yoshitomi T, et al. 2000. Presynaptic effects of botulinum toxin type A on the neuronally evoked response of albino and pigmented rabbit iris sphincter and dilator muscles. Jpn. J. Ophthalmol. 44:106–9 Welch MJ, Purkiss JR, Foster KA. 2000. Sensitivity of embryonic rat dorsal root ganglia neurons to Clostridium botulinum neurotoxins. Toxicon 38:245–58

2 Dec 2003

21:31

AR

AR206-ME55-28.tex

AR206-ME55-28.sgm

LaTeX2e(2002/01/18)

Annu. Rev. Med. 2004.55:505-518. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

NEW APPROACHES TO MIGRAINE THERAPY 38. Cui M, Li Z, You S, et al. 2003. Mechanisms of the antinociceptive effect of subcutaneous Botox® : inhibition of peripheral and central nociceptive processing. Arch. Pharmacol. 365:33 (Abstr.) 39. Munchau A, Bhatia KP. 2000. Uses of botulinum toxin injection in medicine today. BMJ 320:165 40. Scott AB, Magoon EH, McNeer KW, et al. 1990. Botulinum treatment of childhood strabismus. Ophthalmology 97:1434–38 41. Hughes AJ. 1994. Botulinum toxin in clinical practice. Drugs 48:888–93 42. Pasricha PJ, Ravich WJ, Kaloo AN. 1993. Botulinum toxin for achalasia. Lancet 341:244–45 43. Argoff CE. 2002. A focused review on the use of botulinum toxins for neuropathic pain. Clin. J. Pain 6:S181 44. Difazio M, Jabbari B. 2002. A focused review of the use of botulinum toxins for low back pain. Clin. J. Pain 18:S155–S162 45. Freund BJ, Schwartz M. 2002. Use of botulinum toxin in chronic whiplashassociated disorder. Clin. J. Pain 18: S163–S168 46. Evers S, Rahmann A, Vollmer-Haase J, et al. 2002. Treatment of headache with botulinum toxin A—a review according to evidence-based medicine criteria. Cephalalgia 22:699–710 47. Loder E, Biondi D. 2002. Use of botulinum toxins for chronic headaches: a focused review. Clin. J. Pain 18:S169–S176 48. Mathew NT, Kaup AO. 2002. The use of botulinum toxin type A in headache treatment. Curr. Treatment Options Neurol. 4:365–73 49. Ashkenazi A, Silberstein SD. 2003. The evolving management of migraine. Curr. Opin. Neurol. 16(3):341–45 50. Binder WJ, Brin MF, Blitzer A, et al. 2000. Botulinum toxin type A (Botox) for treatment of migraine headaches: an openlabel study. Otolaryngol. Head Neck Surg. 123:669–76 51. Silberstein SD, Mathew N, Saper J, et al. 2000. Botulinum toxin type A as a migraine

52.

53.

54.

55.

56.

57.

58.

59.

60.

61. 62.

63.

64.

P1: GCE

517

preventive treatment. Headache 40:445– 50 Brin MF, Swope DM, O’Brien C, et al. 2000. Botox for migraine: double-blind, placebo-controlled, region-specific evaluation. Cephalalgia 20:421 (Abstr.) Barrientos N, Chana P. 2002. Efficacy and safety of botulinum toxin type A (Botox® ) in the prophylactic treatment of migraine. Headache 42:452 (Abstr.) Behmand RA, Tucker T, Guyuron B, et al. 2002. Single-site Botox injection for the elimination of migraine trigger points. Headache 42:403 (Abstr.) Relja MA, Klepac N. 2003. Botulinum toxin type-A reduces acute medication (triptans) use in migraine patients. Neurology 60:A321 (Abstr.) Mauskop A, Basdeo R. 2000. Botulinum toxin A is an effective prophylactic therapy for migraines. Cephalalgia 20:422 (Abstr.) Smuts JA, Barnard PW. 2000. Botulinum toxin type A in the treatment of headache syndromes: a clinical report on 79 patients. Cephalalgia 20:332 (Abstr.) Mathew NT, Kailasam J, Meadors L. 2002. “Disease modification” in chronic migraine with botulinum toxin type A: long-term experience. Headache 42:454–(Abstr.) Robbins LD. 2001. Botulinum toxin A for refractory chronic daily headache. Cephalalgia 21:465 (Abstr.) Ondo WG, Derman HS. 2002. Botulinum toxin A for chronic daily headache: a 60patient, randomized, placebo-controlled, parallel design study. Headache 42:431 (Abstr.) Scott AB. 1988. Antitoxin reduces botulinum side effects. Eye 2:29–32 Argov Z, Mastaglia FL. 1979. Disorders of neuromuscular transmission caused by drugs. N. Engl. J. Med. 301:409–13 Lake AE, Saper JR. 2003. Botulinum toxin type B for migraine prophylaxis: a 4-month open-label prospective outcome study. Neurology 60:A322 (Abstr.) Opida C. 2002. Open-label study of Myobloc (botulinum toxin type B) in the

2 Dec 2003

21:31

518

65.

66.

Annu. Rev. Med. 2004.55:505-518. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

67.

68.

69.

70.

71.

72.

73.

74.

75.

AR

AR206-ME55-28.tex

ASHKENAZI

¥

AR206-ME55-28.sgm

LaTeX2e(2002/01/18)

P1: GCE

SILBERSTEIN

treatment of patients with transformed migraine headaches. J. Pain 3:10 (Abstr.) Gwynn MW, Baker T, English J. 2003. Double-blind, placebo-controlled study of botulinum toxin B for the treatment of chronic headache. Neurology 60:A322 (Abstr.) Silberstein SD, Goadsby PJ. 2002. Migraine: preventive treatment. Cephalalgia 22:491–512 Langtry HD, Gillis JC, Davis R. 1997. Topiramate. A review of its pharmacodynamic and pharmacokinetic properties and clinical efficacy in the management of epilepsy. Drugs 54:752–73 Galvez-Jimenez N, Hargreave M. 2000. Topiramate and essential tremor. Ann. Neurol. 47:837–38 Marcotte D. 1998. Use of topiramate, a new antiepileptic as a mood stabilizer. J. Affect. Disord. 50:245–51 Silberstein SD, Neto W, Jacobs D, et al. 2003. Efficacy and safety of topiramate in migraine prevention: a dose ranging placebo-controlled, double blind, multicentered trial. Cephalalgia In press Cutrer FM. 2001. Antiepileptic drugs: how they work in headache. Headache 41(Suppl. 1):S3–S10 Storer RJ, Goadsby PJ. 2003. Topiramate inhibits trigeminovascular traffic in the cat: a possible locus of action in the prevention of migraine. Neurology 60:A238 (Abstr.) Young WB, Hopkins MM, Shechter AL, et al. 2002. Topiramate: a case series study in migraine prophylaxis. Cephalalgia 22:659–63 Mathew NT, Kailasam J, Meadors L. 2002. Prophylaxis of migraine, transformed migraine, and cluster headache with topiramate. Headache 42:796–803 Silberstein SD, Bhattacharaya S, Neto W, et al., for the MIGR-001 Study Group. 2003. Topiramate in the prevention of migraine headache: a randomized, double-

76.

77.

78.

79.

80.

81.

82.

83.

84.

85.

blind, placebo-controlled, multiple-dose study. Cephalagia In press Brandes JL, Jacobs D, Neto W, et al. 2003. Topiramate in the prevention of migraine headache: a randomized, double-blind, placebo-controlled parallel study (MIGR002). Neurology 60(Suppl. 1):A238 (Abstr.) Storey JR, Calder CS, Hart DE, et al. 2001. Topiramate in migraine prevention: a double-blind, placebo-controlled study. Headache 41:968–75 Von Seggern RL, Mannix LK, Adelman JU. 2002. Efficacy of topiramate in migraine prophylaxis: a retrospective chart analysis. Headache 42:804–9 Bender WI. 1995. ACE inhibitors for prophylaxis of migraine headaches. Headache 35:470–71 Schrader H, Stovner LJ, Helde G, et al. 2001. Prophylactic treatment of migraine with angiotensin converting enzyme inhibitor (lisinopril): randomised, placebocontrolled, crossover study. Br. J. Med. Psychol. 3222:19–22 Tronvik E, Stovner LJ, Helde G, et al. 2003. Prophylactic treatment of migraine with an angiotensin II receptor blocker: a randomized controlled trial. JAMA 289:65–69 Nishimura Y, Ito T, Saavedra JM. 2000. Angiotensin II AT1 blockade normalizes cerebrovascular autoregulation and reduces cerebral ischemia in spontaneously hypertensive rats. Stroke 31:2478–86 Mendelsohn FA, Jenkins TA, Berkovic SF. 1993. Effects of angiotensin II on dopamine and serotonin turnover in the striatum of conscious rats. Brain Res. 613:221–29 Baltatu O, Afeche SC, Jose-dos Santos SH, et al. 2002. Locally synthesized angiotensin modulates pineal melatonin generation. J. Neurochem. 80:328–34 Reuter U, Chiarugi A, Bolay H, et al. 2002. Nuclear factor-κB as a molecular target for migraine therapy. Ann. Neurol. 51:507–16

2 Dec 2003

21:30

AR

AR206-ME55-29.tex

AR206-ME55-29.sgm

LaTeX2e(2002/01/18) P1: GCE 10.1146/annurev.med.55.091902.103826

Annu. Rev. Med. 2004. 55:519–26 doi: 10.1146/annurev.med.55.091902.103826 c 2004 by Annual Reviews. All rights reserved Copyright ° First published online as a Review in Advance on Oct. 15, 2003

MANAGEMENT OF INFECTIONS IN THE NEUTROPENIC PATIENT

Annu. Rev. Med. 2004.55:519-526. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Kenneth V.I. Rolston Department of Infectious Diseases, Infection Control and Employee Health, The University of Texas, M.D. Anderson Cancer Center, Houston, Texas 77030; email: [email protected]

Key Words risk assessment, bacterial/fungal infections, empiric therapy, infection prevention ■ Abstract Neutropenic patients continue to be at increased risk for developing serious infections despite substantial advances in supportive care. Epidemiologic shifts occur periodically and need to be detected early because they influence prophylactic, empiric, and specific therapy strategies. Although effective in preventing bacterial and some fungal infections, prophylaxis must be used with caution because it is associated with the emergence of resistance. The choices for empiric therapy include combination regimens and monotherapy. Specific choices depend on local factors (epidemiology, susceptibility/resistance patterns, availability). Various treatment settings (hospitalbased, early discharge, outpatient) are also available, and the choice depends on the patient’s risk category. Early diagnosis and treatment of many fungal and viral infections remains suboptimal. Infection control and prevention are important strategies, especially with the emergence of multidrug-resistant organisms.

INTRODUCTION It has been four decades since Bodey et al. first described the relationship between neutropenia and infection (1). Although the risk of infection increases when the absolute neutrophil count (ANC) falls below 1000/mm3, the currently accepted definition of neutropenia is an ANC of ≤500/mm3 (2). The severity and duration of neutropenia are both important and influence not only the frequency and severity of infection but also the response to therapy and overall outcome. It has been estimated that all patients who have severe neutropenia (<100/mm3) for 3 weeks or more will develop a serious infection (3). In addition to quantitative neutropenia, many patients with hematologic disorders (e.g., acute leukemias) have defects in neutrophil function as well, and are at increased risk of infection despite adequate or even increased numbers of neutrophils. 0066-4219/04/0218-0519$14.00

519

2 Dec 2003

21:30

Annu. Rev. Med. 2004.55:519-526. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

520

AR

AR206-ME55-29.tex

AR206-ME55-29.sgm

LaTeX2e(2002/01/18)

P1: GCE

ROLSTON

Chemotherapy-induced neutropenia is often superimposed on other immunological deficits (impaired humoral or cell-mediated immunity) that might be present because of the underlying malignancy. In such patients, infections associated with those immunologic deficits must be considered in addition to infections generally associated with neutropenia. The spectrum of infection in neutropenic patients undergoes periodic changes, and geographic/institutional differences are also common. Knowledge of local epidemiology and of susceptibility/resistance patterns is vital to the appropriate management of neutropenic patients. Standard management includes the prompt administration of broad-spectrum, empiric antibiotic therapy after hospitalization (2). Recent understanding of the syndrome of “febrile neutropenia” has made it possible to recognize low-risk subsets among neutropenic patients, and to consider options such as oral, outpatient therapy (4).

SPECIAL CONSIDERATIONS Patients with neutropenia often fail to develop symptoms and signs of infection because of a blunted inflammatory response. Only 8% of patients with severe neutropenia who develop pneumonia produce purulent sputum compared to 84% who are not neutropenic (5). Chest radiographs show that neutropenic patients with pneumonia often do not develop pulmonary infiltrates; those with urinary tract infections may not have localized symptoms such as dysuria; and some may have meningitis without overt meningeal symptoms or signs. Fever may be the initial and often the only sign of infection. Occasionally, an infection may develop in the absence of fever, as with organisms such as Clostridium septicum, or if the patient is receiving corticosteroids. Approximately 50%–60% of febrile episodes in neutropenic patients never have clinical or microbiological evidence of infection, hence the designation “unexplained fever.” Most of these patients respond to antibiotic therapy, which suggests that these fevers are probably caused by low-grade, undetected infections. Patients with neutropenia can develop infections at unusual sites, with uncommon manifestations, and by opportunistic pathogens that seldom cause infection in immunocompetent hosts. Typhilitis, an inflammatory process most commonly involving the caecum, occurs almost exclusively in patients with acute leukemia (6). Perirectal infections with extensive tissue necrosis extending into the rectum also occur predominantly in patients with acute leukemia (7). Endocarditis is a rare infection in neutropenic patients, presumably because most of these patients are also thrombocytopenic and fail to make the fibrin/platelet mesh that is critical for the formation of bacterial vegetations. Infections can disseminate rapidly in many patients with severe neutropenia, underscoring the importance of early, broad-spectrum antibiotic therapy. However, a low-risk subset also exists, particularly among patients with solid tumors and short-lived neutropenia. Clinical and

2 Dec 2003

21:30

AR

AR206-ME55-29.tex

AR206-ME55-29.sgm

LaTeX2e(2002/01/18)

INFECTIONS IN NEUTROPENIC PATIENTS

P1: GCE

521

statistically derived risk-prediction rules have been developed in order to identify such patients (8–10). Finally, it is important to consider noninfectious causes of fever such as transfusion reactions, chemotherapy or other drug-related fever, allergic reactions, and occasionally tumor fever.

Annu. Rev. Med. 2004.55:519-526. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

SPECTRUM OF INFECTION Infections that occur during the early phases of a neutropenic episode are predominantly bacterial. Fungal infections are generally seen in patients with severe and prolonged neutropenia and in those receiving multiple courses of broad-spectrum antibiotics. Most large cancer treatment centers have reported a predominance of Gram-positive pathogens as a cause of bacteremia in neutropenic patients (11). However, institutional differences do exist; many centers often encounter Gramnegative bacilli instead. Additionally, most tissue-based infections (pneumonia, typhilis/enterocolitis, perirectal infections) and polymicrobial infections are frequently Gram-negative (12). Yeasts such as Candida spp. and Trichosporon spp. and molds such as Aspergillus spp., the Zygomycetes, and Fusarium spp. are the usual causes of fungal infections. Viral infections are uncommon, but herpesviruses (HSV, VZV, CMV, HHV6) and community respiratory viruses are the most frequent pathogens. Table 1 lists the common pathogens encountered in neutropenic patients.

EMPIRIC THERAPY The administration of empiric, broad-spectrum antibiotic therapy is considered the standard of care for febrile episodes in neutropenic patients (2). Several therapeutic choices are available (Table 2). Individual institutions must tailor the use of specific agents based on local epidemiology and local susceptibility/resistance patterns. Some experts favor combination regimens that are potentially bactericidal over single-agent regimens (monotherapy), especially in high-risk patients with documented Gram-negative infections. However, most head-to-head clinical trials have not demonstrated the superiority of combination therapy over monotherapy (2). Combination regimens may reduce the overall emergence of resistance, but may be associated with increased toxicity and cost. Combinations that do not include a glycopeptide (e.g., vancomycin) consist of an aminoglycoside (e.g., amikacin, tobramycin, gentamicin) along with a cephalosporin (e.g., cefepime, ceftazidime), carbapenem (e.g., imipenem, meropenem), quinolone (e.g., ciprofloxacin), or an antipseudomonal penicillin/beta-lactamase inhibitor (e.g., piperacillin/tazobactam). When more potent Gram-positive coverage is indicated, combining vancomycin with ceftazidime, imipenem, meropenem, or a quinolone might be indicated. Other specific Gram-positive agents (e.g., linezolid, quinupristin/dalfopristin) are not yet indicated for use in empiric regimens.

2 Dec 2003

21:30

522

AR

AR206-ME55-29.tex

AR206-ME55-29.sgm

LaTeX2e(2002/01/18)

P1: GCE

ROLSTON

Annu. Rev. Med. 2004.55:519-526. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

TABLE 1 Common causes of documented infections in neutropenic patients Bacteria Gram-positive Staphylococcus spp. Viridans streptococci Enterococcus spp. Corynebacterium spp. Bacillus spp. Gram-negative Escherichia coli Pseudomonas aeruginosa Klebsiella spp. Enterobacter spp. Proteus spp. Stenotrophomonas maltophilia Fungi Candida albicans and other Candida spp. Trichosporon beigelli Aspergillus fumigatus and other Aspergillus spp. Zygomycetes Fusarium spp. Viruses Herpesviruses Respiratory syncytial virus Influenza virus Parainfluenza virus Adenovirus

Based on current susceptibility patterns, cefepime, meropenem, and imipenem are the most appropriate agents for monotherapy (13, 14). The response rates of most empiric regimens range from 55% to 85% (2). It is customary to treat for 72–96 h before making alterations, in order to allow the initial regimen to produce a response. These alterations depend on the clinical setting (e.g., suspected catheter-related infection, abdominal or pelvic focus, central nervous system infection) and microbiologic data. The most common modifications are the addition of a glycopeptide, if not used initially, or the addition of an antifungal agent. Strengthening Gram-negative or anaerobic coverage when indicated is also commonplace. The choice of antifungal agents depends on the use of antifungal prophylaxis and the nature of the fungal infections (yeast versus mold). In general, amphotericin B or one of its lipid preparations is used in this setting. Newer agents such as voriconazole and caspofungin are also being evaluated for empiric antifungal therapy in persistently febrile neutropenic patients (15).

2 Dec 2003

21:30

AR

AR206-ME55-29.tex

AR206-ME55-29.sgm

LaTeX2e(2002/01/18)

INFECTIONS IN NEUTROPENIC PATIENTS

P1: GCE

523

TABLE 2 Traditional choices for empiric therapy in febrile neutropenic patients Combination Regimens (without glycopeptide) Aminoglycoside

Annu. Rev. Med. 2004.55:519-526. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Combination Regimens (with a glycopeptide)a Vancomycin

+antipseudomonal penicillin ± beta/lactamase inhibitor +extended spectrum cephalosporin +carbapenem or quinolone +antipseudomonal penicillin ± beta/lactamase inhibitor +extended spectrum cephalosporin +carbapenem +quinolone or monobactam

Single-Agent Regimens (monotherapy) Carbapenemb Extended spectrum cephalosporin Antipseudomonal penicillin + beta/lactamase inhibitorc a

Linezolid and quinoprinstin/dalfopristin not recommended for empiric therapy.

b

Imipenem and meropenem (but not ertapenem).

c

May need additional clinical data.

SPECIFIC THERAPY When a specific pathogen is isolated, therapy can be adjusted to the organism isolated, based on local susceptibility patterns. Table 3 lists the agents used most often, including some newer options. The isolation of a specific pathogen, especially if it is a Gram-positive organism, does not necessarily permit the use of narrow-spectrum agents in patients with severe neutropenia (particularly those with significant mucositis), since these patients may have occult Gram-negative or polymicrobial infections. Our antifungal armamentarium is expanding; several newer triazoles (voriconazole, posaconazole, ravuconazole) and echinocandins (caspofungin, anidulafungin, micafungin) have either recently become available or are nearing approval by the US Food and Drug Administration. Our ability to treat most viral infections is still quite limited; new antiviral agents are needed.

LENGTH OF THERAPY The length of therapy depends on the type of infection (bacteremia, urinary tract infection, pneumonia, etc.), the organism isolated (S. aureus, coagulase-negative staphylococci, P. aeruginosa, Candida spp., etc.), and the persistence of or recovery from neutropenia. Most experts continue therapy until (a) all signs and symptoms of infection have resolved, (b) the patient has been afebrile for 3–4 days, (c) cultures, if initially positive, have been rendered negative, and (d) radiographic evidence of infection, if initially present, shows signs of resolution. Some experts

2 Dec 2003

21:30

524

AR

AR206-ME55-29.tex

AR206-ME55-29.sgm

LaTeX2e(2002/01/18)

P1: GCE

ROLSTON

Annu. Rev. Med. 2004.55:519-526. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

TABLE 3 Therapeutic agents for the treatment of documented infections in febrile neutropenic patients Antibacterial Agents Narrow-spectrum (Gram-positive) Nafcillin, oxacillin Vancomycin (teicoplanin, where available) Linezolid Quinupristin/dalfopristin Narrow-spectrum (Gram-negative) Aminoglycosides Monobactams (aztreonam) Quinolones (ciprofloxacin) Narrow-spectrum (anaerobic) Metronidazole Clindamycin Broad-spectrum Meropenem, imipenem Cefepime, ceftazidime Piperacillin/tazobactam Moxifloxacin, gatifloxacin (additional clinical data needed) Trimethoprim/sulfamethoxazole Antifungal Agents Amphotericin B (including lipid preparations) Fluconazole, itraconazole, voriconazole Caspofungin Antiviral Agents Acyclovir Valacyclovir cGanciclovir Foscarnet Cidofovir Ribavirin

recommend continuing therapy until resolution of neutropenia (ANC ≥ 500/mm3 for 2 consecutive days) (2).

RISK ASSESSMENT AND RISK-BASED THERAPY It has long been recognized that not all febrile neutropenic patients have the same risk of developing serious infection-related or other complications. However, it has only recently become possible to identify low-risk patients accurately and in a timely manner, using clinical criteria and/or statistically derived prediction rules (8–10, 16, 17). Low-risk patients can be discharged early after initial stabilization in the hospital or can be treated for the febrile episode without any hospitalization (8, 18, 19). Most outpatient regimens are quinolone-based (e.g., ciprofloxacin +

2 Dec 2003

21:30

AR

AR206-ME55-29.tex

AR206-ME55-29.sgm

LaTeX2e(2002/01/18)

INFECTIONS IN NEUTROPENIC PATIENTS

P1: GCE

525

amoxicillin/clavulanate or clindamycin) and can be administered orally, although parenteral outpatient therapy is also feasible in low-risk patients who are unable to tolerate oral therapy. An experienced team and the appropriate infrastructure are essential for a risk-based program to be successful.

Annu. Rev. Med. 2004.55:519-526. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

INFECTION PREVENTION Antibacterial prophylaxis (generally with a quinolone such as ciprofloxacin) has been found to reduce the frequency of documented Gram-negative infections, but it may have no impact, or occasionally may lead to an increase in Grampositive infections (20). Increased survival as a result of antibacterial prophylaxis has not been demonstrated. The emergence of resistant organisms is a significant drawback of chemoprophylaxis (21). Consequently, it is recommended only for patients anticipated to have severe and prolonged neutropenia (ANC ≤ 100/mm3 for >10–14 days). Antifungal prophylaxis (e.g., fluconazole/itraconazole) has been shown to reduce the frequency of infections caused by Candida spp. (22, 23). Mold infections are much more difficult to prevent, and effective strategies are yet to be developed. Some of the newer antifungal agents with activity against filamentous fungi (e.g., voriconazole, posaconazole) are being evaluated for the prevention of fungal infections.

SUMMARY The management of febrile neutropenic patients has evolved considerably. Although hospital-based, empiric therapy remains the standard for high-risk patients, newer strategies such as early discharge or oral, out-patient therapy are becoming the norm for low-risk patients. The most important factors for the selection of antimicrobial agents for empiric use are local microbiology and susceptibility/ resistance patterns. Better strategies for infection prevention are needed, particularly for fungal and viral infections. These issues will continue to challenge clinicians caring for febrile neutropenic patients for the foreseeable future. The Annual Review of Medicine is online at http://med.annualreviews.org

LITERATURE CITED 1. Bodey GP, Buckley M, Sathe YS, et al. 1966. Quantitative relationships between circulating leukocytes and infection in patients with acute leukemia. Ann. Intern. Med. 64:328–40 2. Hughes WT, Armstrong D, Bodey GP, et al. 2002. 2002 Guidelines for the use of an-

timicrobial agents in neutropenic patients with cancer. Clin. Infect. Dis. 34:730–51 3. Rolston KVI, Bodey GP. 2003. Infections in patients with cancer. In Cancer Medicine, ed. JF Holland, E Frei, pp. 2633–58. Ontario: BC Decker. 6th ed. 4. Rolston KVI. 1998. Expanding the options

2 Dec 2003

21:30

526

5.

Annu. Rev. Med. 2004.55:519-526. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

6.

7.

8.

9.

10.

11.

12.

13.

14.

AR

AR206-ME55-29.tex

AR206-ME55-29.sgm

LaTeX2e(2002/01/18)

P1: GCE

ROLSTON for risk-based therapy in febrile neutropenia. Diagn. Microbiol. Infect. Dis. 31:411– 16 Sickles EA, Greene WH, Wiernik PH. 1975. Clinical presentation of infection in granulocytopenic patients. Arch. Intern. Med. 135:715–19 Gomez L, Martino R, Rolston KV. 1998. Neutropenic enterocolitis: spectrum of the disease and comparison of definite and possible cases. Clin. Infect. Dis. 27:695– 99 Rolston KVI, Bodey GP. 1993. Diagnosis and management of perianal and perirectal infection in the granulocytopenic patient. In Current Clinical Topics in Infectious Diseases, ed. J Remington, MN Swartz, pp. 164–71. Boston: Blackwell Sci. Rolston K. 1999. New trends in patient management: risk-based therapy for febrile patients with neutropenia. Clin. Infect. Dis. 29:515–21 Talcott JA, Siegel RD, Finberg R, et al. 1992. Risk assessment in cancer patients with fever and neutropenia: a prospective, two-center validation of a prediction rule. J. Clin. Oncol. 10:316–22 Klastersky J, Paesmans M, Rubenstein E, et al. 2000. The MASCC Risk Index: a multinational scoring system to predict low-risk febrile neutropenic cancer patients. J. Clin. Oncol. 18:3038–51 Zinner SH. 1999. Changing epidemiology of infections in patients with neutropenia and cancer: emphasis on gram-positive and resistant bacteria. Clin. Infect. Dis. 29:490– 94 Elting LS, Bodey GP, Fainstein V. 1986. Polymicrobial septicemia in the cancer patient. Medicine 65:2l8–25 Jones RN. 1999. Contemporary antimicrobial susceptibility patterns of bacterial pathogens commonly associated with febrile patients with neutropenia. Clin. Infect. Dis. 29:495–502 Rolston KVI, Kontoyiannis DP, Raad II, et al. 2003. Susceptibility surveillance among gram-negative bacilli at a compre-

15.

16.

17.

18.

19.

20.

21.

22.

23.

hensive cancer center. Presented at Gen. Meet. Am. Soc. Microbiol., 103rd, Washington, DC, May 18–22 (Abstr. #2362) Walsh TJ, Pappas P, Winston DJ, et al. 2002. Voriconazole compared with liposomal amphotericin B for empirical antifungal therapy in patients with neutropenia and persistent fever. N. Engl. J. Med. 346:225– 34 Freifeld A, Marchigiani D, Walsh T, et al. 1999. A double-blind comparison of empirical oral and intravenous antibiotic therapy for low-risk febrile patients with neutropenia during cancer chemotherapy. N. Engl. J. Med. 341:305–11 Kern WV, Cometta A, DeBock R, et al. 1999. Oral versus intravenous empirical antimicrobial therapy for fever in patients with granulocytopenia who are receiving cancer chemotherapy. N. Engl. J. Med. 341:312–18 Rubenstein EB, Rolston K, Benjamin RS, et al. 1993. Outpatient treatment of febrile episodes in low risk neutropenic cancer patients. Cancer 71:3640–46 Mullen CA, Petropoulos D, Roberts WM, et al. 1999. Outpatient treatment of febrile neutropenia in low risk pediatric cancer patients. Cancer 86:126–34 Cruciani M, Rampazzo R, Malena M, et al. 1996. Prophylaxis with fluoroquinolones for bacterial infections in neutropenic patients: a meta-analysis. Clin. Infect. Dis. 23:795–805 Rolston KVI. 1998. Commentary: chemoprophylaxis and bacterial resistance in neutropenic patients. Infect. Dis. Clin. Pract. 7:202–4 Goodman JL, Winston DJ, Greenfield A, et al. 1992. A controlled trial of fluconazole to prevent fungal infections in patients undergoing bone marrow transplantation. N. Engl. J. Med. 326:845–51 Marr KA, Seidel K, White TC, et al. 2000. Candidemia in allogeneic blood and marrow transplant recipients: evolution of risk factors after the adoption of prophylactic fluconazole. J. Infect. Dis. 181:309–16

Annual Review of Medicine Volume 55, 2004

CONTENTS Effect of Completed Human Genome Sequence on Development of Novel Therapeutics for Human Disease, Christopher P. Austin Toward Alzheimer Therapies Based on Genetic Knowledge, John Hardy

Annu. Rev. Med. 2004.55:519-526. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Inherited Diseases Involving G Proteins and G Protein--Coupled Receptors, Allen M. Spiegel, Lee S. Weinstein The Scientific Basis for the Current Treatment of Parkinson's Disease, C. Warren Olanow Progress in Antisense Technology, Stanley T. Crooke Serum Proteomics in the Early Diagnosis of Cancer, Kevin P. Rosenblatt, Peter Bryant-Greenwood, J. Keith Killian, Arpita Mehta, David Geho, Virginia Espina, Emanuel F. Petricoin, Lance A. Liotta Molecular Neurobiology of Drug Addiction, Jennifer Chao, Eric J. Nestler Beta Cell Replacement for Type 1 Diabetes, Peter G. Stock, Jeffrey A. Bluestone Cochlear Implantation for the Treatment of Deafness, Benjamin J. Copeland, Harold C. Pillsbury Drug-Eluting Stents, T. Cooper Woods, Andrew R. Marks New Approaches to Hemodialysis, Andreas Pierratos Emerging Infectious Threats to the Blood Supply, Roger Y. Dodd, David A. Leiby Lead Poisoning, Herbert Needleman The Impact of Minimally Invasive Surgical Techniques, Sir Ara Darzi, Yaron Munz Implementing a Research Agenda for Complementary and Alternative Medicine, Jonathan D. Berman, Stephen E. Straus Basic Advances and New Avenues in Therapy of Spinal Cord Injury, Bruce H. Dobkin, Leif A. Havton Clinical Management of Tuberculosis in the Context of HIV, Bouke C. de Jong, Dennis M. Israelski, Elizabeth L. Corbett, Peter M. Small HIV-Associated Lipodystrophy: Pathogenesis, Prognosis, Treatment, and Controversies, Polyxeni Koutkia, Steven Grinspoon Human Papillomavirus Vaccines and Prevention of Cervical Cancer, Kathrin U. Jansen, Alan R. Shaw Opportunities for Control of Meningococcal Disease in the United States, Pratima L. Raghunathan, Scott A. Bernhardt, Nancy E. Rosenstein Recent Advances in the Development of HIV-1 Vaccines Using Replication-Incompetent Adenovirus Vectors, John W. Shiver, Emilio A. Emini Left Ventricular Diastolic Dysfunction and Diastolic Heart Failure, William H. Gaasch, Michael R. Zile Mechanisms of Pulmonary Fibrosis, Victor J. Thannickal, Galen B. Toews, Eric S. White, Joseph P. Lynch III, Fernando J. Martinez Systemic Mastocytosis, Cem Akin, Dean D. Metcalfe

1 15 27 41 61

97 113 133 157 169 179 191 209 223 239 255

283

303 319

333

355 373 395 419

The erbB Family: Targets for Therapeutic Development Against Cancer and Therapeutic Strategies Using Monoclonal Antibodies and Tyrosine Kinase Inhibitors, Eric K. Rowinsky Nonmyeoablative Immunotherapy for Solid Tumors, Richard W. Childs, John Barrett Rituximab: Expanding Role in Therapy for Lymphomas and Autoimmune Diseases, William Rastetter, Arturo Molina, Christine A. White

Annu. Rev. Med. 2004.55:519-526. Downloaded from arjournals.annualreviews.org by Universitaet Heidelberg on 10/04/05. For personal use only.

Botulinum Toxin and Other New Approaches to Migraine Therapy, Avi Ashkenazi, Stephen D. Silberstein Management of Infections in the Neutropenic Patient, Kenneth V.I. Rolston

433 459

477 505 519