Pineal Region Tumors. Diagnosis and Treatment Options
Progress in Neurological Surgery Vol. 23
Series Editor
L. Dade Lunsford
Pittsburgh, Pa.
Pineal Region Tumors Diagnosis and Treatment Options
Volume Editors
Tatsuya Kobayashi Nagoya L. Dade Lunsford Pittsburgh, Pa. 58 figures, 19 in color, and 28 tables, 2009
Basel · Freiburg · Paris · London · New York · Bangalore · Bangkok · Shanghai · Singapore · Tokyo · Sydney
Tatsuya Kobayashi, MD, PhD
L. Dade Lunsford, MD, FACS
Nagoya Radiosurgery Center Nagoya Kyoritsu Hospital Nagoya, Japan
Lars Leksell and Distinguished Professor of Neurological Surgery The University of Pittsburgh Pittsburgh, Pa., USA
Library of Congress Cataloging-in-Publication Data Pineal region tumors: diagnosis and treatment options / volume editors, Tatsuya Kobayashi, L. Dade Lunsford. p. ; cm. – (Progress in neurological surgery, ISSN 0079-6492 ; v. 23) Includes bibliographical references and indexes. ISBN 978-3-8055-9077-8 (hardcover: alk. paper) 1. Pineal gland–Tumors. 2. Pineal gland–Tumors–Surgery. I. Kobayashi, Tatsuya, 1938- II. Lunsford, L. Dade. [DNLM: 1. Pinealoma–therapy. 2. Brain Neoplasms–therapy. 3. Pineal Gland–physiopathology. W1 PR673 v. 23 2009 / WK 350 P649245 2009] RC280. P5P56 2009 616. 99’447 – dc22 2008055657
Bibliographic Indices. This publication is listed in bibliographic services, including Current Contents® and PubMed/MEDLINE. Disclaimer. The statements, opinions and data contained in this publication are solely those of the individual authors and contributors and not of the publisher and the editor(s). The appearance of advertisements in the book is not a warranty, endorsement, or approval of the products or services advertised or of their effectiveness, quality or safety. The publisher and the editor(s) disclaim responsibility for any injury to persons or property resulting from any ideas, methods, instructions or products referred to in the content or advertisements. Drug Dosage. The authors and the publisher have exerted every effort to ensure that drug selection and dosage set forth in this text are in accord with current recommendations and practice at the time of publication. However, in view of ongoing research, changes in government regulations, and the constant flow of information relating to drug therapy and drug reactions, the reader is urged to check the package insert for each drug for any change in indications and dosage and for added warnings and precautions. This is particularly important when the recommended agent is a new and/or infrequently employed drug. All rights reserved. No part of this publication may be translated into other languages, reproduced or utilized in any form or by any means electronic or mechanical, including photocopying, recording, microcopying, or by any information storage and retrieval system, without permission in writing from the publisher. © Copyright 2009 by S. Karger AG, P.O. Box, CH–4009 Basel (Switzerland) www.karger.com Printed in Switzerland on acid-free and non-aging paper (ISO 9706) by Reinhardt Druck, Basel ISSN 0079–6492 ISBN 978–3–8055–9077–8 e-ISBN 978–3–8055–9078–5
Contents
VII Series Editor’s Note Lunsford, L.D. (Pittsburgh, Pa.) IX Preface Kobayashi, T. (Nagoya) Introduction 1 Statistical Analysis of Pineal Tumors Based on the Data of Brain Tumor Registry of Japan Shibui, S.; Nomura, K. (Tokyo) Tumors of Pineal Cell Origin 12 Pathology of Pineal Parenchymal Tumors Sato, K.; Kubota, T. (Fukui) 26 Occipital Transtentorial Approach and Combined Treatments for Pineal Parenchymal Tumors Tsumanuma, I. (Niigata); Tanaka, R. (Tsubame); Fujii, Y. (Niigata) 44 Role of Stereotactic Radiosurgery in the Management of Pineal Parenchymal Tumors Kano, H.; Niranjan, A.; Kondziolka, D.; Flickinger, J.C.; Lunsford, L.D. (Pittsburgh, Pa.) Tumors of Germ Cell Origin 59 Pathology of Intracranial Germ Cell Tumors Sato, K.; Takeuchi, H.; Kubota, T. (Fukui) 76 Pineal Germ Cell Tumors Matsutani, M. (Saitama) 86 Strategy of Combined Treatment of Germ Cell Tumors Sawamura, Y. (Sapporo/Tokyo) 96 Radiation Therapy for Intracranial Germ Cell Tumors Aoyama, H. (Sapporo) 106 Stereotactic Radiosurgery for Pineal and Related Tumors Mori, Y.; Kobayashi, T. (Nagoya); Hasegawa, T.; Yoshida, K.; Kida, Y. (Komaki)
V
119 Management of Central Nervous System Germinoma: Proposal for a Modern Strategy Shibamoto, Y. (Nagoya) 130 Quality of Life of Extremely Long-Time Germinoma Survivors Mainly Treated with Radiotherapy Sugiyama, K.; Yamasaki, F.; Kurisu, K.; Kenjo, M. (Hiroshima) 140 Author Index 141 Subject Index
VI
Contents
Series Editor’s Note
Despite their relative rarity, tumors of the pineal region have remained of great interest to neurosurgeons, radiation, and medical oncologists. At many international meetings, controversy reigns as the merits and risks of various treatment strategies are debated. In Asia, the incidence of germ cell tumors seems much greater than the incidence seen in North American or European populations. Classification systems are occasionally confusing as tumors of the pineal region may arise from cells within the pineal parenchyma, from germ cell origin, from adjacent structures including the dorsal midbrain, from other midline structures, or from the tentorium. In this volume, the authors report various diagnostic and treatment strategies for tumors of the pineal region, emphasizing multimodality management in many patients. For some patients of the pineal region, the evaluation often needs to include pertinent serum or cerebrospinal fluid markers, which may be elevated in nongerminomatous germ cell tumors (a rather oxymoronic term developed by pathologists to confuse clinicians, I suppose). High-resolution magnetic resonance imaging is critical to assess the tumor and to follow the response to treatment. A variety of treatment modalities as well as various surgical approaches are discussed in this volume. We have asked many Asian colleagues to present their experience in this volume, in part because of the more frequent management of these tumors in the Asian population. I am sure that their findings are applicable to all patients and all centers that diagnose pineal region tumors. L. Dade Lunsford, Pittsburgh, Pa., USA
VII
Preface
The pineal region is an anatomic location where a wide variety of intracranial tumors occur. Germ cell tumors (GCTs) and pineal parenchymal tumors are the most frequently encountered. The frequency of pineal GCTs is higher in Asian countries, including Japan, while pineal parenchymal tumors are less frequently detected in Asia than in the United States and European countries. Emeritus Prof. Naoki Kageyama, Nagoya University School of Medicine, my teacher and one of the pioneers of GCT studies in Japan, described ‘ectopic pinealoma’ as another name for suprasellar germinomas in 1961. Many original studies of intracranial GCTs by Japanese investigators have contributed to the evolution of the treatment for pineal region tumors. The current volume of Progress in Neurological Surgery takes advantage of the knowledge of Japanese experts on pineal tumors, with special emphasis on epidemiology, pathological diagnosis, and surgical, radiotherapeutic, radiosurgical and chemotherapeutic management options. We hope that this volume will enhance the knowledge of our colleagues about the various manifestations and treatment options available in the modern era of neurosurgery. Tatsuya Kobayashi, Nagoya
IX
Introduction Kobayashi T, Lunsford LD (eds): Pineal Region Tumors. Diagnosis and Treatment Options. Prog Neurol Surg. Basel, Karger, 2009, vol 23, pp 1–11
Statistical Analysis of Pineal Tumors Based on the Data of Brain Tumor Registry of Japan Soichiro Shibuia ⭈ Kazuhiro Nomurab a
Neurosurgery Division, National Cancer Center Hospital, Tokyo, bTokyo Labour Welfare Hospital, Tokyo, Japan
Abstract In this study, we present statistical analyses of pineal tumors based on the data from Brain Tumor Registry of Japan. The most frequent tumor in the pineal region was germinoma, and it accounted for 49.2% of all pineal tumors; it was followed by pineocytoma (8.5%), glioma (6.5%), pineoblastoma (5.1%), malignant teratoma (5.2%) and teratoma (5.1%). Germinoma is most frequent among patients between 10 and 19 years of age, and there are some patients aged >30 years; however, there are few patients with choriocarcinoma, embryonal carcinoma, and yolk sac tumor who are aged >30 years. Pineoblastoma is most frequent among patients under 5 years of age, while pineocytoma is evenly distributed in patients between 10 and 60 years of age. The 5-year survival rate of germinoma was 89.4%, while those of embryonal carcinoma, yolk sac tumor and choriocarcinoma were 35.3, 37.3 and 58.1%, respectively. Copyright © 2009 S. Karger AG, Basel
The Committee of Brain Tumor Registry of Japan (BTRJ) was founded in 1975 in order to investigate the incidence and characteristics of brain tumors in Japan; it is organized by 80 professors of neurosurgical departments of universities and medical colleges throughout Japan. The first report was published in 1978, and in September 2003 the 11th edition of the report was issued [1]. The number of collaborating neurosurgical institutions is now 473. This report was based on the data of the 11th edition of BTRJ (1969–1996) and 12th edition of BTRJ (1984– 2000), which we are now preparing to publish. Statistical analyses of the pineal region tumors were performed mainly for the patients registered during 1984 and 2000. Permission to use the data was obtained from the Committee of BTRJ in June 2008.
Registration
Patients with brain tumors who were treated during 1969 and 2000 have been registered in the brain tumor registration office of National Cancer Center Hospital according to the registration forms which had been sent from the collaborating neurosurgical institutions throughout Japan. The numbers of the patients by diagnosis, age, location and treatment were summarized, and survival rates were calculated using Cutler’s method [1]. Every year about 5,000 cases are registered, and more than 117,000 cases have been registered so far. According to the Central Brain Tumor Registry of the United States (CBTRUS), the incidence rate of all primary and brain and central nervous system tumors was 14.8 cases per 100,000 person-years [2]. If the incidence rate in Japan was similar to that of the United States, the registration rate of the primary brain tumors would be estimated at 40% of the brain tumors occurring in Japan in a year.
Frequencies of Primary Brain Tumors
The frequencies of various primary brain tumors registered between 1984 and 2000 are listed in table 1. Meningioma was the most frequent and accounted for 26.5% of the primary brain tumors; it was followed by glioma (25.6%), pituitary adenoma (17.9%) and schwannoma (10.5%); however, the frequency of each tumor was different by age of patients. Among patients under 15 years of age, glioma was the most frequent. It accounted for 54.1% of all primary brain tumors; it was followed by germ cell tumors (14.6%), craniopharyngiomas (8.5%), meningioma (2.0%) and pituitary adenomas (2.0%). Frequency was quite different among patients over 70 years of age. Among elderly patients, meningioma was the most frequent. It accounted for 41.3%, followed by glioma 25.6%, pituitary adenoma 9.6% and schwannoma 7.0%. Malignant lymphoma was as high as 6.7%. Compared with the data of CBTRUS 1998–2002, glioma and meningioma were less frequent, and pituitary adenoma and germ cell tumor were much more frequent in Japan.
Neuroepithelial Tumor
Frequencies of neuroepithelial tumors are listed in table 2. Glioblastoma was the most frequent neuroepithelial tumor. It accounted for 34.5% of all neuroepithelial tumors and was followed by astrocytoma (26.7%) and anaplastic astrocytoma (17.6%). Among patients under 15 years of age, astrocytoma was the most frequent. It accounted for 32.9% of all neuroepithelial tumors and was followed by
2
Shibui · Nomura
Table 1. Frequencies of primary brain tumors (Brain Tumor Registry of Japan 1984–2000) Total
Age ≤14 years
Meningioma
26.5
Glioma
25.6
Schwannoma Pituitary adenoma
15–69 years ≥70 years 26.5
41.3
31.4
54.1
23.4
25.6
43.6
10.5
1.4
11.9
7.0
8.0
17.9
2.0
20.6
9.6
6.3
Germ cell tumor
2.7
14.6
1.9
0.0
0.6
Craniopharyngioma
3.9
8.5
3.3
1.6
0.7
Dermoid, epidermoid
1.5
1.4
1.6
0.4
N/A
Hemangioblastoma
1.7
4.9
1.9
1.0
0.9
Sarcoma
2.1
5.2
1.9
0.2
N/A
Malignant lymphoma 3.0
0.3
2.7
6.7
3.1
Others
5.6
4.3
6.6
5.4
Total
4.6
2.0
CBTRUS 1998–2002
100.0 100.0 100.0 100.0 (n = 67,293) (n = 5,148) (n = 53,674) (n = 8,471)
100.0 (n = 63,698)
Figures indicate percentages.
medulloblastoma (22.2%), anaplastic astrocytoma (9.1%), ependymoma (8.7%) and glioblastoma (6.9%). Among patients over 70 years of age, glioblastoma was the most frequent and it accounted for 59.4% of all neuroepithelial tumors; it was followed by anaplastic astrocytoma (19.6%). It was observed that 80% of neuroepithelial tumors in aged patients were malignant gliomas.
Pineal Region Tumor
Apart from 105 histologically unknown cases, 1,188 cases of pineal region tumor were registered during 1984 and 2000. The most frequent tumor was germinoma. It accounted for 49.2% of all pineal region tumors and was followed by pineocytoma (8.5%), glioma (6.5%), pineoblastoma (5.5%), malignant teratoma (5.2%), and teratoma (5.1%; table 3). There was a 13:1 male predominance in germinoma, 14:1 in teratoma and malignant teratoma, 2:1 in pineoblastoma and totally 5:1, while there was no predominance in pineocytoma.
Statistical Analysis of Pineal Tumors from BTRJ
3
Table 2. Frequencies of neuroepithelial tumors (Brain Tumor Registry Japan 1984–2000) All
Age ≤14 years
15–69 years ≥70 years
Glioblastoma
34.5
6.6
36.1
60.9
Astrocytoma
26.7
32.9
27.6
13.2
Anaplastic astrocytoma
17.6
9.5
19.0
19.8
Oligodendroglioma
3.5
1.4
4.3
1.2
Anaplastic oligodendroglioma
0.9
0.1
1.1
0.7
Ependymoma
3.1
8.1
2.4
0.4
Anaplastic ependymoma
0.9
3.4
0.5
0.2
Plexus papilloma
1.0
2.8
0.7
0.4
Medulloblastoma
4.0
21.3
0.8
0.1
Other glioma
7.8
14.0
7.5
3.1
Total
100.0 100.0 100.0 100.0 (n = 17,492) (n = 2,787) (n = 12,536) (n = 2,169)
Figures indicate percentages.
Age distribution of patients with various pineal tumors is shown in figure 1. Germinoma is most frequent among patients between 10 and 19 years of age and there are some patients over 30; however, there are a few patients over 30 with choriocarcinoma, embryonal carcinoma, and yolk sac tumor. Pineoblastoma is most frequent among patients under 5 years of age, while pineocytoma is evenly distributed in patients between 10 and 60 years of age. Benign teratoma is common in patients under 20 years of age, and so is malignant teratoma. Treatment of pineal region tumors depends on the histological diagnosis, and it will be discussed in the other chapters. Generally, the benign tumors, such as teratoma, dermoid and epidermoid tumors, are resected surgically, and the malignant tumors are treated with chemotherapy and radiotherapy after histological diagnosis by biopsy or surgical removal. Germinoma is sensitive to chemotherapeutic agents such as cisplatin and carboplatin. They are usually administered with etoposide and followed by low-dose radiation [3–6]. Most of malignant tumors such as choriocarcinoma, embryonal carcinoma and yolk sac tumors are usually treated with ifosfomide, cisplatin and etoposide and radiation in Japan.
4
Shibui · Nomura
180
Male Female
160 140
Patients
120 100 80 60 40 20 0 0
5
10
15
20
25
30
35
a
40 45 Age (years)
50
55
60
65
70
75
20
80
Male Female
18 16 14
Patients
12 10 8 6 4 2 0 0
5
10
15
20
b
25
30
35
40 45 Age (years)
50
55
60
65
70
Fig. 1. Age distribution and sex ratio of patients with primary brain tumors in the pineal region. a Germinoma (n = 585; male:female ratio = 12.6:1). b Choriocarcinoma (n = 27; male:female ratio = 5.8:1).
Statistical Analysis of Pineal Tumors from BTRJ
5
75
80
20
Male Female
18 16 14
Patients
12 10 8 6 4 2 0 0
5
10
15
20
25
30
35
c
40 45 Age (years)
50
55
60
65
70
75
20
80
Male Female
18 16 14
Patients
12 10 8 6 4 2 0 0
5
d
10
15
20
25
30
35
40 45 Age (years)
50
55
60
65
70
75
80
Fig. 1. c Embryonal carcinoma (n = 34; male:female ratio = 10.3:1). d Yolk sac tumor (n = 34; male:female ratio = 10.3:1).
6
Shibui · Nomura
25 Male Female 20
Patients
15
10
5
0 0
5
10
15
20
25
30
35
e
40 45 Age (years)
50
55
60
65
70
75
20
80
Male Female
18 16 14
Patients
12 10 8 6 4 2 0 0
5
10
15
20
f
25
30
35
40 45 Age (years)
50
55
60
65
70
Fig. 1. e Other germ cell tumor (n = 68; male:female ratio = 7.5:1). f Pineoblastoma (n = 65; male:female ratio = 1.7:1).
Statistical Analysis of Pineal Tumors from BTRJ
7
75
80
20
Male Female
18 16 14
Patients
12 10 8 6 4 2 0 0
5
10
15
20
25
30
35
g
40 45 Age (years)
50
55
60
65
70
75
20
80
Male Female
18 16 14
Patients
12 10 8 6 4 2 0 0
5
h
10
15
20
25
30
35
40 45 Age (years)
50
55
60
65
70
75
80
Fig. 1. g Pineocytoma (n = 101; male:female ratio = 1:1). h Teratoma (n = 61; male:female ratio = 14.3:1).
8
Shibui · Nomura
20
Male Female
18 16 14
Patients
12 10 8 6 4 2 0 0
5
10
15
20
i
25
30
35 40 45 Age (years)
50
55
60
65
70
Fig. 1. i Malignant teratoma (n = 62; male:female ratio = 14.5:1).
Cumulative survival rates of each pineal tumor were calculated by Cutler’s method. Five-year survival rate of germinoma was 89.4%, while those of embryonal carcinoma, yolk sac tumor and choriocarcinoma were 35.3, 37.5 and 58.1%, respectively (table 4).
Discussion
The incidence of the pineal tumors according to the centralized brain tumor registries varies from 0.4 to 1% among adult patients and from 3 to 5% among children [7]. Especially germ cell tumors are very frequent in Asian countries such as Japan and Korea [8]. The frequency in Japan is 5 times as high as in the western countries. In the registries of the United States, germ cell tumors were classified only into germinoma and mixed germ cell tumors, and the details of choriocarcinoma, embryonal carcinoma and yolk sac tumors were not reported. Even in Japan these tumors are rare, but 95 cases apart from 68 cases with mixed germ cell tumors were registered in BTRJ during 1984 and 2000 and survival rates were calculated. Compared with pure germinomas, survival rates of these tumors were very low, and especially those of embryonal carcinoma and yolk sac
Statistical Analysis of Pineal Tumors from BTRJ
9
75
80
Table 3. Frequencies of pineal tumors by histology (BTRJ 1984–2000) Male Germinoma
Female
Total
542
43
585 (49.2%)
Pineoblastoma
51
50
101 (8.5%)
Pineoblastoma
41
24
65 (5.5%)
Teratoma
57
4
61 (5.1%)
Malignant teratoma
58
4
62 (5.2%)
Embryonal carcinoma
31
3
34 (2.9%)
Yolk sac tumor
31
3
34 (2.9%)
Choriocarcinoma
23
4
27 (2.3%)
Other germ cell tumor
60
8
68 (5.7%)
Glioma
42
35
77 (6.5%)
4
1
5 (0.4%)
Epidermoid
13
2
15 (1.3%)
Others
32
22
54 (4.5%)
985
203
1,188 (100.0%)
84
72
Dermoid
Total Unknown
105
Table 4. Cumulative survival rates of pineal tumors (BTRJ 1984–2000) Patients
1 year
2 years
3 years
4 years
5 years
Germinoma
486
96.2
92.4
91.3
90.3
89.4
Pineocytoma
77
95.4
89
87.5
87.5
84.1
Pineoblastoma
30
74.8
57.1
50.7
48.5
46.1
Teratoma
50
96.4
96.4
92.2
92.2
89.6
Malignat teratoma
41
84.2
74.9
72.8
70.6
70.6
Embryonal carcinoma
12
64.7
38.2
35.3
35.3
35.3
Yolk sac tumor
13
55.6
44.8
37.3
37.3
37.3
Choriocarcinoma
15
62.3
62.3
62.3
62.3
58.1
Rates are expressed as percentages.
10
Shibui · Nomura
tumors were less than 40%. Although chemotherapies such as ifosfomide, cisplatin and etoposide and carboplatin + etoposide improved the survival, the latter is still unsatisfactory. Nationwide registry is important. It is a retrospective study and the evidence level is not as high as it would be in a prospective study. But thanks to it we can understand the global aspects of the tumors, especially rare tumors such as pineal tumors. In 2008, BTRJ started to register tumors online. The registration rate is expected to be much higher, and most of the brain tumors in Japan will be registered.
References 1
2
3
4
The Committee of Brain Tumor Registry of Japan: Brain Tumor Registry of Japan (1969– 1996). Neurol Med Chir 2003;43(suppl):i–vii, 1–111. Central Brain Tumor Registry of the United States: Statistical Report: Primary Brain Tumors in the United States, 1998–2003. Chicago, Central Brain Tumor Registry of the United States, 2005. Matsutani M, Sano K, Takakura K, Fujimaki T, Nakamura O, Funata N, Seto T: Primary intracranial germ cell tumors: a clinical analysis of 153 histologically verified cases. J Neurosurg 1997;86: 446–455. Matsutani M, The Japanese Pediatric Brain Tumor Study Group: Combined chemotherapy and radiation therapy for CNS germ cell tumors – the Japanese experience. J Neurooncol 2001;54: 311–316.
5
6
7
8
Shibamoto Y, Abe Y, Yamashita J, Takahashi M, Hiraoka M, Ono K, Tsutsui K: Treatment result of intracranial germinoma as a function of the irradiated volume. Int J Radiat Oncol Biol Phys 1998; 15:285–290. Sawamura Y, Shirato H, Ikeda J, Tada M, Ishii N, Kato T, Abe H, Fujieda K: Induction chemotherapy followed by reduced-volume radiation therapy for newly diagnosed central nervous system germinoma. J Neurosurg 1998;88:66–72. Villano JL, Propp JM, Porter KR, Stewart AK, Valyi-Nagi T, Li X, Engelhard HH, McCarthy BJ: Malignant pineal germ-cell tumors: an analysis if cases from three tumor registries. Neuro Oncol 2008;10:121–130. Nomura K: Epidemiology of germ cell tumors in Asia of pineal region tumor. J Neurooncol 2001; 54:211–217.
Soichiro Shibui, MD, DMSc Neurosurgery Division, National Cancer Center Hospital 5-1-1 Tsukiji, Chuo-ku Tokyo 104-0045 (Japan) Tel. +81 3 3542 2511, Fax +81 3 3542 3815, E-Mail
[email protected]
Statistical Analysis of Pineal Tumors from BTRJ
11
Tumors of Pineal Cell Origin Kobayashi T, Lunsford LD (eds): Pineal Region Tumors. Diagnosis and Treatment Options. Prog Neurol Surg. Basel, Karger, 2009, vol 23, pp 12–25
Pathology of Pineal Parenchymal Tumors Kazufumi Sato ⭈ Toshihiko Kubota Division of Neurosurgery, Department of Sensory and Locomotor Medicine, Faculty of Medical Science, University of Fukui, Fukui, Japan
Abstract Pineal parenchymal tumors (PPTs) are neuroepithelial tumors that arise from pineocytes or their precursors. According to the currently revised WHO classification of tumors of the central nervous system, PPTs are subdivided into well-differentiated pineocytoma, poorly differentiated pineoblastoma, and PPT with intermediate differentiation (PPTID). Pineocytomas are slow-growing neoplasms composed of small mature cells resembling pineocytes. Large pineocytomatous rosettes are the most characteristic appearance. Pineoblastomas are the most primitive form and have a highly malignant biological behavior. PPTIDs show an intermediate histological grade of malignancy between pineocytomas and pineoblastomas. Immunohistochemically, PPTs are positive for several neuronal markers, including synaptophysin, neurofilaments, class III β-tubulin, and chromogranin A. Photosensory differentiation is associated with immunoreactivity for retinal S-antigen and rhodopsin. Ultrastructurally, dense core vesicles and clear vesicles are present in both cytoplasm and cellular processes, the latter showing occasional synapse-like junctions. In some cases, ultrastructural evidence of photoreceptor differentiation, such as synaptic ribbons, microtubular sheaves, and cilia, is observed. Little is known about the genetics responsible for the development of PPTs. Several chromosomal abnormalities have been identified frequently in pineoblastomas and PPTIDs but less commonly in pineocytomas. Pineoblastomas are known to occur in patients with RB1 gene abnormalities, and these tumors also develop in patients with familial bilateral retinoblastomas (trilateral retinoblastoma syndrome). However, specific gene abnormalities involved in the tumorigenesis of PPTs have not been identified. Copyright © 2009 S. Karger AG, Basel
Histology of the Normal Human Pineal Gland
The normal adult pineal gland has a lobulated architecture in that parenchymal cells (pineocytes) are intersected by vascular-rich connective tissue stroma (fig. 1a). These pineal parenchymal cells are a specialized form of neuronal cells, and are immunopositive for synaptophysin (fig. 1b) and neurofilaments, as well as
a
b
c
Fig. 1. Normal adult human pineal gland. a The pineal gland shows a lobulated structure, and pineal parenchymal cells are intersected by vascular-rich connective tissue stroma. HE. b Immunohistochemistry of synaptophysin showing positivity for parenchymal cells. c Immunohistochemistry of GFAP showing positivity for most of interstitial cells and a small number of parenchymal cells.
chromogranin A, retinal S-antigen, serotonin, and melatonin [1, 2]. A pineocyte has a club-shaped cytoplasmic process terminating on the neighboring blood vessel wall. The pineal gland also contains a moderate number of astrocytes, which are concentrated at the stroma and less frequently distributed in the lobules. These cells are strongly positive for glial fibrillary acidic protein (GFAP; fig. 1c) and S-100 protein [1, 3].
Classification and Grading of Pineal Parenchymal Tumors
In older literature, pineal parenchymal tumors (PPTs) have occasionally been referred to as ‘true pinealomas’ [4]. Herrick and Rubinstein [4] reported 28 cases of PPTs, and confirmed that these tumors have a potential towards divergent glial and neuronal differentiation. In one example, differentiation towards neurosensory photoreceptors was demonstrated. In 1993, Schild et al. [5] proposed the classification of PPTs into pineocytomas, pineoblastomas, PPTs with intermediate differentiation (PPTIDs), and mixed PPTs. In the 1993 WHO classification of tumors of the central nervous system (CNS), mixed pineocytoma/pineoblastoma was referred to as ‘a tumor containing area composed of immature (pineoblastic) and differentiated (pineocytic) neoplastic cells’. In the recently revised WHO classification of tumors of the CNS [6], PPTs are subdivided into well-differentiated pineocytoma, poorly differentiated pineoblastoma, and PPTID. Pineocytomas are slow-growing neoplasms composed of small mature cells resembling pineocytes. The WHO classification designates pineocytomas as grade I lesions. Pineoblastomas are the most primitive, and have a highly malignant biological behavior. These tumors correspond to WHO grade IV. PPTIDs show an intermediate-grade of malignancy, which may correspond to
Pathology of Pineal Parenchymal Tumors
13
WHO grade II or III. Although definite grading criteria have not been established, this malignancy scale showed a correlation (p < 0.001) with overall survival, event free survival, and local control [7]. Histologically, a strong correlation was shown between overall survival and necrosis or expression of neurofilament protein. In terms of event-free survival, a strong correlation was found with mitosis and necrosis [7]. Grade I pineocytomas do not metastasize, and the 5-year event-free survival is 100%. Local recurrence (21%) and rare spinal metastasis (5%) may occur with grade II PPTIDs. Grade III PPTIDs and grade IV pineoblastomas can recur locally and metastasize [7]. The MIB-1 (Ki-67) labeling index (LI) is significantly higher in pineoblastomas than in other types of PPTs. The LI for pineocytomas and pineocytomas with anaplasia was less than 7%, but that for pineoblastomas was more than 8% [8, 9]. The MIB-1 LI correlated well with histological malignancy and neuronal differentiation evaluated immunohistochemically by both neurofilaments and synaptophysin [9, 10]. In PPTIDs, the mean MIB-1 LI was reported to range from 3 to 10% [9, 10].
Incidence, Age, and Sex Distribution
Pineal region tumors are uncommon and account for less than 1% of primary tumors of the CNS. PPTs account for 14–30% of pineal tumors [5, 6]. Based on previous criteria, pineocytomas represent 14–60% of PPTs [5, 6, 11]. Pineocytomas occur throughout life but most frequently affect adults (mean age: 38–45 years) [4–6, 11–14]. They are distributed evenly between the sexes. Pineoblastomas comprise approximately 40% of all PPTs, and they usually occur in the first two decades of life (mean age: 18.5 years) [4, 6, 7, 11, 13, 15]. There is no sex predilection. PPTIDs comprise at least 20% of all PPTs [6]. The reported incidence of 0–60% may reflect the frequent misdiagnosis of this tumor and/or the inclusion of mixed pineocytoma/pineoblastoma and other unusual PPTs [6]. PPTIDs affect all ages, with a peak incidence in early adulthood (mean age: 38 years) [5, 6, 11, 13, 15]. There is a slight female preponderance [6, 7].
Neuroimaging
On computed tomography, pineocytomas typically show round, well-demarcated, isodense or hypodense masses, measuring less than 3 cm in diameter. Most tumors show homogenous contrast enhancement. Some cases show calcification (tumoral and/or pineal gland origin) or occasional cystic changes [16, 17]. Mild hydrocephalus is a common feature. On MRI, pineocytomas tend to be hypointense or
14
Sato · Kubota
Fig. 2. Pineocytoma. Enhanced MRI showing well-demarcated pineal mass with homogeneous enhancement.
isointense on T1-weighted images and hyperintense on T2-weighted images with usually homogenous contrast enhancement [16, 17] (fig. 2). On computed tomography, pineoblastomas appear as a large, lobulated mass, which usually shows heterogeneous contrast enhancement [16]. Hydrocephalus is present in most cases. The tumors are variably cystic. On MRI, large portions of the tumors are hypointense to isointense on T1-weighted images and hyperintense on T2-weighted images with heterogeneous contrast enhancement [16–19] (fig. 3). The margins between tumors and adjacent structures are occasionally less well defined, suggesting an invasive nature [16, 17]. Extensive cystic change is rare. In PPTIDs, there are no specific neuroradiological manifestations that permit their reliable distinction from other PPTs.
Pathology
Pineocytoma Macroscopy
Pineocytomas are usually well-defined lesions with a pale and grayish appearance, and the cut surface shows homogeneous or granular appearance [4, 20]. Degenerative changes including cyst formation and focal hemorrhage can be present [11]. There is no evidence of local infiltration into adjacent brain parenchyma and leptomeninges.
Pathology of Pineal Parenchymal Tumors
15
Fig. 3. Pineoblastoma. Enhanced MRI showing marked enhancement of a pineal mass with cystic change.
Microscopy
Pineocytomas are well-differentiated neoplasm composed of relatively small, uniform, mature cells resembling mature pineocytes (fig. 4a) [1, 4–7, 13, 20]. The majority of nuclei are round-to-oval with inconspicuous nucleoli. The cells grow in sheets or in a lobulated structure with large anucleate areas called pineocytomatous rosettes (fig. 4b). These rosettes are a characteristic feature of pineocytoma, and they have been interpreted as evidence of neuronal differentiation. Pineocytomas are also characterized by cytoplasmic processes with club-like terminal expansion oriented towards blood vessels. These processes are argyrophlic and can be visualized with specific silver carbonate or Bodian stains (fig. 4c). Mitotic figures or necrotic areas are not conspicuous. The pleomorphic variant of pineocytomas contains ganglioid giant cells with hyperchromatic and bizarre nuclei, but usually mitotic figures are not seen (fig. 5) [5, 8, 13, 20, 21]. This pleomorphism is characteristic of neuronal differentiation [4]. Immunohistochemistry
Pinocytomas show immunopositivity for several neuronal markers, including synaptophysin, neurofilaments (fig. 6a), class III β-tubulin, and chromogranin A [6, 7, 14, 21, 22]. Photosensory differentiation is associated with immunoreactivity for
16
Sato · Kubota
a
b
c
Fig. 4. Pineocytoma. a The tumor is composed of relatively small, uniform, mature cells resembling mature pineocytes. HE. b Pineocytomatous rosettes. HE. c Club-like processes oriented towards blood vessels. Bodian stain.
Fig. 5. Pleomorphic pineocytoma. Photomicrograph showing numerous ganglioid giant cells with hyperchromatic and bizarre nucleus. HE.
retinal S-antigen and rhodopsin [10, 11, 22]. Pineocytomatous rosettes are strongly positive for synaptophysin and neurofilaments, and the latter demonstrates the fibrillar meshwork of cytoplasmic processes [1]. In typical pineocytoma, interstitial cells are usually positive for GFAP (fig. 6b) and S-100 protein [1, 8]. In some pineocytomas, tumors show evidence of an extensive astrocytic component [4, 20]. In pleomorphic pineocytoma, tumor cells are strongly positive for neuronal markers especially for synaptophysin (fig. 7) and neurofilaments [7, 8, 13, 21]. Electron Microscopy
Pineocytoma cells are characterized by moderate size, round to oval nuclei with moderate amounts of heterochromatin. Nucleoli are not prominent. Their
Pathology of Pineal Parenchymal Tumors
17
a
b
Fig. 6. Pineocytoma. a Immunohistochemistry for neurofilament (68 kDa) showing numerous positive cell processes. b Immunohistochemistry of GFAP showing strong positivity in interstitial cells.
cytoplasm is abundant and contains well-developed organelles, including smooth and rough endoplasmic reticulum, Golgi apparatus, mitochondria, microtubules, and lysosomes. Membrane-bound dense core vesicles (fig. 8a) and clear vesicles (fig. 8b) are present in both cytoplasm and cellular processes, the latter showing occasional synapse-like junctions (fig. 8b) [1, 4, 14, 23–25]. Pineocytoma cells share ultrastructural findings with normal pineocytes, such as synaptic ribbons, 9 + 0 cilia, microtubular sheaves, annulate lamellae (fig. 8c), paired twisted filaments, membranous whorls (fig. 8d), and fibrous bodies (fig. 8e) [1, 4, 14, 23–26]. In cases with retinoblastic differentiation, microrosettes bearing microvilli and bulb-ended 9 + 0 cilia are observed (fig. 8f) [4, 11, 14, 25]. Cell junctions are rarely seen.
Pineoblastoma Macroscopy
Pineoblastomas are pinkish-gray, soft, and gelatinous tumors and occasionally show diffuse infiltration of local structures [20]. Hemorrhage and/or necrosis are common, and craniospinal dissemination through the CSF occurs frequently [4, 5, 11, 20, 27]. Extracranial metastatic spread of the tumor may occur to the peritoneal cavity via the ventriculoperitoneal shunt. Extremely rare extracranial metastases to the bone [28] or lung [29] have been described in a few cases with pineoblastomas. Microscopy
Pineoblastomas are composed of hypercellular and closely packed patternless sheets of small cells, mimicking primitive neuroectodermal tumor or
18
Sato · Kubota
Fig. 7. Pleomorphic pineocytoma. Immunohistochemistry of synaptophysin showing strong positivity in most of the tumor cells.
medulloblastoma. The tumor cells have high nuclear-to-cytoplasmic ratios and indistinct cellular borders (fig. 9a). Nuclei are round to irregular and rich in chromatin. Necrosis is common, but mitotic activity varies considerably. No lobular arrangement could be discerned. Numerous Homer Wright rosettes (fig. 9b) and occasional Flexner-Wintersteiner rosettes are observed, but pineocytomatous rosettes are absent. In rare cases, fleurettes (a bundle of cytoplasmic processes showing bulbous expansion at their distal portion) indicative of photoreceptor differentiation may be found [4, 30]. Silver carbonate stains show scant cytoplasm and few cellular processes. Melanin production as well as cartilaginous and rhabdomyoblastic differentiation are encountered in rare pineoblastomas referred to as ‘pineal anlage tumors’ [31, 32]. Immunohistochemistry
The immunophenotype of pineoblastoma is similar to that of pineocytoma and includes reactivity for neuronal and photoreceptor markers, such as synaptophysin, neurofilaments, class III β-tubulin, chromogranin A, and retinal S-antigen [7, 10, 11, 13, 22, 33]. However, reactivities are variable and usually lower in intensity than in other PPTs. Flexner-Wintersteiner rosettes and fleurettes are strongly positive for retinal S-antigen. Thus, pineoblastomas share morphological and immunohistochemical features with photoreceptor cells of the developing pineal gland and retina. In rare cases immunopositive for GFAP and αB-crystallin, the presence of entrapped reactive astrocytes should be excluded [6].
Pathology of Pineal Parenchymal Tumors
19
a
b
d
c
e
f
Fig. 8. Pineocytoma. Electron micrographs. a Dense core vesicles (arrows) in the cell processes. b Numerous clear vesicles and a few synapse-like junctions (arrows). c Annulate lamellae. d Membranous whorls. e Fibrous body. f Microvillous projections and cilia.
a
b
Fig. 9. Pineoblastoma. a The tumor is composed of hypercellular and closely packed patternless sheets of small cells with necrosis. HE. b Homer Wright rosettes (arrows). HE.
Electron Microscopy
The fine structure of pineoblastomas is similar to that of any poorly differentiated neuroectodermal tumors. Tumor cells are composed of closely packed small cells with round to oval nuclei, which are often irregularly indented (fig.
20
Sato · Kubota
Fig. 10. Pineoblastoma. Electron micrograph showing poorly differentiated tumor cells with no recognizable neuronal structures.
10). Cytoplasm is scant and contains few profiles of rough and smooth endoplasmic reticulum, annulate lamellae, membranous whorls as well as occasional microtubules, intermediate filaments and lysosomes [1, 10, 14, 24, 34]. Scattered dense core vesicles are observed in the cytoplasm, but they are very infrequent compared with those seen in pineocytomas [14, 24]. Poorly formed, short cell processes may contain microtubules as well as scant dense core vesicles [24]. In some cases, tumor cells show ultrastructural evidence of photoreceptor differentiation, such as synaptic ribbons, microtubular sheaves, and club-shaped giant cilia with a 9 + 0 configuration [1, 14, 24, 34]. Junctional complexes are usually inconspicuous.
Pineal Parenchymal Tumor of Intermediate Differentiation Macroscopy
The gross appearance of PPTIDs is similar to that of pineocytomas. The tumors are circumscribed, soft in texture, and usually lacking in gross evidence of necrosis. Microscopy
PPTIDs are either diffuse (neurocytoma-like) or somewhat lobulated tumors characterized by moderately high cellularity, mild to moderate nuclear atypia, and low
Pathology of Pineal Parenchymal Tumors
21
Fig. 11. PPTID. Tumor cells showing moderately high cellularity and mild nuclear atypia. HE.
to moderate mitotic activity (fig. 11) [6]. Preliminary studies suggest that tumors corresponding to grade II or III can be distinguished on the basis of mitotic activity and neurofilament protein immunoreactivity [7, 13]. PPTIDs include transitional form cases in which typical pineocytomatous areas are associated with a diffuse pattern [6]. In the current WHO classification [6], mixed pineocytoma/pineoblastoma is considered an extremely rare neoplasm interpreted as harboring both primitive and mature components and being subsumed under the category of pineoblastomas. Immunohistochemistry
Immunohistochemically, the tumor cells show synaptophysin positivity. Variable labeling is seen with antibodies to neurofilaments, chromogranin A, and retinal S-antigen [1, 9, 10, 13].
Genetics
Little is known regarding molecular genetic alterations underlying the formation of PPTs. Cytogenetic studies have suggested that monosomy in the distal 12q region and partial deletion or loss of chromosome 11 are related to tumor progression [35–37]. A microarray analysis of pineocytoma has shown high-level expression of genes coding for enzymes related to melatonin synthesis (HIOMT) and phototransduction
22
Sato · Kubota
(OPN4, RGS16), and such reactivities indicate bi-directional neurosecretory and photosensory differentiation [38]. A relationship between the RB1 gene and pineocytoma has not been established. In pineoblastomas, monosomy for chromosomes 20 and 22, and trisomy for chromosome 14 have been described [7]. In another analysis of pineoblastomas, both monosomy 22 and INI1 gene (22q11) mutation were reported [39]. Pineoblastomas are known to occur in patients with RB1 gene abnormalities, and their prognosis is significantly worse than that of sporadic cases [40]. Pineoblastomas develop in patients with familial bilateral retinoblastomas, an occurrence termed ‘trilateral retinoblastoma syndrome’ [19, 41], and have also been reported in patients with familial adenomatous polyposis [42]. Familial pineoblastoma has been described, and a genetic basis and mutual exposure to an environmental factor responsible for the gene abnormalities were suggested [43]. By comparative genomic hybridization, frequent chromosomal gains and losses have been identified in PPTIDs but not in pineocytomas [44]. The most common chromosomal imbalances in PPTID are +4q, +12q, and –22. In a real-time RT-PCR analysis, the expression level of four genes (PRAME, CD24, POU4F2, and HOXD13) in PPTID is distinctly higher, almost the same level as in pineoblastomas, and in contrast to the low expression level of these genes in pineocytomas [38].
References 1 Jouvet A, Fèvre-Montange M, Besançon R, Derrington E, Saint-Pierre G, Belin MF, Pialat J, Lapras C: Structural and ultrastructural characteristics of human pineal gland, and pineal parenchymal tumors. Acta Neuropathol (Berl) 1994;88:334–348. 2 Huang SK, Klein DC, Korf HW: Immunocytochemical demonstration of rodopsin, S-antigen, and neuron-specific proteins in the human pineal gland. Cell Tissue Res 1992;267:493–498. 3 Zang X, Nilaver G, Stein BM, Fetell MR, Duffy PE: Immunocytochemistry of pineal astrocytes: species differences and functional implications. J Neuropathol Exp Neurol 1985;44:486–495. 4 Herrick MK, Rubinstein LJ: The cytological differentiating potential of pineal parenchymal neoplasms (true pinealomas). A clinicopathological study of 28 tumours. Brain 1979;102:289–320. 5 Schild SE, Scheithauer BW, Schomberg PJ, Hook CC, Kelly PJ, Frick L, Robinow JS, Buskirk SJ: Pineal parenchymal tumors. Clinical, pathologic, and therapeutic aspects. Cancer 1993;72:870–880.
Pathology of Pineal Parenchymal Tumors
6 Nakazato Y, Jouvet A, Scheithauer BW: Tumours of the pineal region; in Louis DN, Ohgaki H, Wiestler OD, Cavenee WK (eds): WHO Classification of Tumours of the Central Nervous System, ed 4. Lyon, IARC, WHO press, 2007, pp 123–127. 7 Jouvet A, Fauchon F, Bouffet E, Saint-Pierre G, Champier J, Fevre-Montange M: Tumors of pineal parenchymal and glial cells; in McLendon RE, Rosenblum MK, Bigner DD (eds): Russell & Rubinstein’s Pathology of Tumors of the Nervous System, ed 7. London, Hodder Arnold, 2006, pp 413–425. 8 Kuchelmeister K, von Borcke IM, Klein H, Bergmann M, Gullotta F: Pleomorphic pineocytoma with extensive neuronal differentiation: report of two cases. Acta Neuropathol (Berl) 1994;88: 448–453. 9 Tsumanuma I, Tanaka R, Washiyama K: Clinicopathological study of pineal parenchymal tumors: correlation between histopathological features, proliferative potential, and prognosis. Brain Tumor Pathol 1999;16:61–68.
23
10 Numoto RT: Pineal parenchymal tumors: cell differentiation and prognosis. J Cancer Res Clin Oncol 1994;120:683–690. 11 Mena H, Rushing EJ, Ribas JL, Delahunt B, McCarthy WF: Tumors of pineal parenchymal cells: a correlation of histological features, including nucleolar organizer regions, with survival in 35 cases. Hum Pathol 1995;26:20–30. 12 Cho BK, Wang KC, Nam DH, Kim DG, Jung HW, Kim HJ, Han DH, Choi KS: Pineal tumors: experience with 48 cases over 10 years. Childs Nerv Syst 1998;14:58. 13 Jouvet A, Saint-Pierre G, Fauchon F, Privat K, Bouffet E, Ruchoux MM, Chauveinc L, Fèvre-Montange M: Pineal parenchymal tumors: a correlation of histological features with prognosis in 66 cases. Brain Pathol 2000;10:49–60. 14 Min KW, Scheithauer BW, Bauserman SC: Pineal parenchymal tumors: an ultrastructural study with prognostic implications. Ultrastruct Pathol 1994;18:69–85. 15 Hoffman HJ, Yoshida M, Becker LE, Hendrick EB, Humphreys RP: Pineal region tumors in childhood. Experience at the Hospital for Sick Children. 1983. Pediatr Neurosurg 1994;21:91–104. 16 Chiechi MV, Smirniotopoulos JG, Mena H: Pineal parenchymal tumors: CT and MR features. J Comput Assist Tomogr 1995;19:509–517. 17 Nakamura M, Saeki N, Iwadate Y, Sunami K, Osato K, Yamaura A: Neuroradiological characteristics of pineocytoma and pineoblastoma. Neuroradiology 2000;42:509–514. 18 Korogi Y, Takahashi M, Ushio Y: MRI of pineal region tumors. J Neurooncol 2001;54:251–261. 19 Provenzale JM, Weber AL, Klintworth GK, McLendon RE: Radiologic-pathologic correlation. Bilateral retinoblastoma with coexistant pinealoblastoma (trilateral retinoblastoma). AJNR 1995; 16:157–165. 20 Borit A, Blackwood W, Mair WG: The separation of pineocytoma from pineoblastoma. Cancer 1980;45:1408–1418. 21 Sato K, Kubota T, Kawano H: Immunohistochemical and ultrastructural study of pineocytoma with pure neuronal differentiation. J Clin Electron Microscopy (Jpn) 1992;25:191–198. 22 Perentes E, Rubinstein LJ, Herman MM, Donoso LA: S-antigen immunoreactivity in human pineal glands and pineal parenchymal tumors. A monoclonal antibody study. Acta Neuropathol (Berl) 1986;71:224–227.
24
23 Hassoun J, Gambarelli D, Peragut JC, Toga M: Specific ultrastructural markers of human pinealomas. A study of four cases. Acta Neuropathol (Berl) 1983;62:31–40. 24 Markesbery WR, Haugh RM, Young AB: Ultrastructure of pineal parenchymal neoplasms. Acta Neuropathol (Berl) 1981;55:143–149. 25 Nielsen SL, Wilson CB: Ultrastructure of a ‘pineocytoma’. J Neuropathol Exp Neurol 1975;34: 148–158. 26 Hassoun J, Devictor B, Gambarelli D, Peragut JC, Toga M: Paired twisted filaments: a new ultrastructural marker of human pinealomas? Acta Neuropathol (Berl) 1984;65:163–165. 27 DeGirolami U, Schmidek H: Clinicopathological study of 53 tumors of the pineal region. J Neurosurg 1973;39:455–462. 28 Constantine C, Miller DC, Gardner S, Balmaceda C, Finlay J: Osseous metastasis of pineoblastoma: a case report and review of the literature. J Neurooncol 2005;74:53–57. 29 Banerjee AK, Kak VK: Pineoblastoma with spontaneous intra and extracranial metastasis. J Pathol 1974;114:9–12. 30 Stefanko SZ, Manschot WA: Pinealoblastoma with retinoblastomatous differentiation. Brain 1979;102:321–332. 31 Berns S, Pearl G: Review of pineal anlage tumor with divergent histology. Arch Pathol Lab Med 2006;130:1233–1235. 32 Schmidbauer M, Budka H, Pilz P: Neuroepithelial and ectomesenchymal differentiation in a primitive pineal tumor (‘pineal anlage tumor’). Clin Neuropathol 1989;8:7–10. 33 Yamane Y, Mena H, Nakazato Y: Immunohistochemical characterization of pineal parenchymal tumors using novel monoclonal antibodies to the pineal body. Neuropathology 2002;22: 66–76. 34 Kline KT, Damjanov I, Katz SM, Schmidek H: Pineoblastoma: an electron microscopic study. Cancer 1979;44:1692–1699. 35 Bello MJ, Rey JA, de Campos JM, Kusak ME: Chromosomal abnormalities in a pineocytoma. Cancer Genet Cytogenet 1993;71:185–186. 36 Rainho CA, Rogatto SR, de Moraes LC, BarbieriNeto J: Cytogenetic study of a pineocytoma. Cancer Genet Cytogenet 1992;64:127–132. 37 Sreekantaiah C, Jockin H, Brecher ML, Sandberg AA: Interstitial deletion of chromosome 11q in a pineoblastoma. Cancer Genet Cytogenet 1989;39: 125–131.
Sato · Kubota
38 Fèvre-Montange M, Champier J, Szathmari A, Wierinckx A, Mottolese C, Guyotat J, FigarellaBranger D, Jouvet A, Lachuer J: Microarray analysis reveals differential gene expression patterns in tumors of the pineal region. J Neuropathol Exp Neurol 2006;65:675–684. 39 Biegel JA, Fogelgren B, Zhou JY, James CD, Janss AJ, Allen JC, Zagzag D, Raffel C, Rorke LB: Mutations of the INI1 rhabdoid tumor suppressor gene in medulloblastomas and primitive neuroectodermal tumors of the central nervous system. Clin Cancer Res 2000;6:2759–2763. 40 Plowman PN, Pizer B, Kingston JE: Pineal parenchymal tumours. II. On the aggressive behaviour of pineoblastoma in patients with an inherited mutation of the RB1 gene. Clin Oncol (R Coll Radiol) 2004;16:244–247. 41 Bader JL, Miller RW, Meadows AT, Zimmerman LE, Champion LA, Voûte PA: Trilateral retinoblastoma. Lancet 1980;2:582–583.
42 Gadish T, Tulchinsky H, Deutsch AA, Rabau M: Pinealoblastoma in a patient with familial adenomatous polyposis: variant of Turcot syndrome type 2? Report of a case and review of the literature. Dis Colon Rectum 2005;48: 2343–2346. 43 Lesnick JE, Chayt KJ, Bruce DA, Rorke LB, Trojanowski J, Savino PJ, Schatz NJ: Familial pineoblastoma. Report of two cases. J Neurosurg 1985; 62:930–932. 44 Rickert CH, Simon R, Bergmann M, DockhornDworniczak B, Paulus W: Comparative genomic hybridization in pineal parenchymal tumors. Genes Chromosomes Cancer 2001;30:99–104.
Kazufumi Sato, MD Division of Neurosurgery, Department of Sensory and Locomotor Medicine Faculty of Medical Science, University of Fukui 23 Shimoaizuki, Matsuoka, Eiheiji-cho Yoshida-gun, Fukui 910-1193 (Japan) Tel. +81 776 61 8474, Fax +81 776 61 8121, E-Mail
[email protected]
Pathology of Pineal Parenchymal Tumors
25
Tumors of Pineal Cell Origin Kobayashi T, Lunsford LD (eds): Pineal Region Tumors. Diagnosis and Treatment Options. Prog Neurol Surg. Basel, Karger, 2009, vol 23, pp 26–43
Occipital Transtentorial Approach and Combined Treatments for Pineal Parenchymal Tumors Itaru Tsumanumaa ⭈ Ryuichi Tanakab ⭈ Yukihiko Fujiia Departments of Neurosurgery, aBrain Research Institute, Niigata University, Niigata, and bTsubame Rosai Hospital, Tsubame, Japan
Abstract The deep-seated location of pineal parenchymal tumors (PPTs) and their associations with critical structures make their surgical resection technically challenging; further, the rarity of PPTs and repeated changes in their histopathological diagnostic criteria makes the study of their biological behavior and clinical outcomes difficult. Here, we describe the surgical techniques and results of an occipital transtentorial approach for PPTs together with the results in the clinicopathological study of PPTs. Since 1982, we have treated 93 patients with pineal region tumors, including 17 PPTs, with the occipital transtentorial approach using the lateral semiprone position. The infrasplenial approach is helpful in separating the internal cerebral veins from the tumor, particularly when the tumor is tightly adherent to the veins. Permanent homonymous hemianopsia occurred in 1 of the 17 patients with PPTs. Permanent ocular movement disorders were not encountered. Extensive removal of the tumor significantly prolongs survival at least in patients with pineocytomas and PPT of intermediate differentiation (PPTIMD). Despite extensive resection and adjuvant radiochemotherapy, the prognosis of the patients with pineoblastomas is extremely poor. Although the proliferative potentials of pineocytomas and PPTIMD were significantly lower than those of pineoblastomas, there was no such difference between pineocytomas and PPTIMD.
Copyright © 2009 S. Karger AG, Basel
Pineal parenchymal tumors (PPTs) are very rare neoplasms arising from the pineal gland located on the diencephalic roof at the posterior extremity of the third ventricle. The term ‘pinealoma’, coined by Krabbe [1], was originally used for tumors arising from pineal parenchymal cells. It had become erroneously used to designate neoplasms that are now widely recognized to be germinomas. On the basis of histopathological findings, del Rio-Hortega [2] classified the tumors arising from
pineal parenchymal cells into ‘pineoblastomas’ and ‘pineocytomas’. In 1993, the World Health Organization (WHO) international classification added a category named ‘mixed/transitional pineal tumors’ and organized these three entities as subtypes of ‘PPTs’ [3]. The name ‘mixed/transitional pineal tumors’ was changed to ‘pineal parenchymal tumor of intermediate differentiation (PPTIMD)’ in the 2000 WHO classification of nervous system tumors. Recently, the ‘papillary tumor of the pineal region’ subtype was added to PPTs in the 2007 WHO classification [4]. The deep-seated location of PPTs and their association with critical structures make their surgical resection technically challenging. Pineal region tumors have most commonly been treated using the infratentorial supracerebellar approach [5] or the occipital transtentorial approach (OTA) [6]. One of the advantages of the infratentorial supracerebellar approach is that the deep venous system does not interfere with access to the pineal region. However, this approach requires patients to be placed in the sitting position, which is associated with a risk of air embolism. In contrast, the OTA can be performed with patients positioned either prone or semiprone, thereby avoiding the risk of air embolism. The disadvantage of this approach is that the deep venous system obstructs access to the tumor. Moreover, the rarity of PPTs and repeated changes in the histopathological diagnostic criteria makes the study of their biological behavior and clinical outcomes difficult. The heterogeneity of their management also renders the interpretation of published data more difficult. This chapter describes the role, surgical techniques, and results obtained with the OTA. The clinicopathological features of the PPTs treated have also been mentioned.
Indication and Role of Occipital Transtentorial Approach in the Management of Pineal Parenchymal Tumors
Tumors of varied histological and biological types arising in the pineal region can be removed using the OTA. Since 1982, we have treated 93 patients with pineal region tumors (table 1) with this approach using a lateral semiprone position and an infrasplenial approach [7]. Surgery plays the most important role in the management of patients with PPTs. The establishment of an accurate histological diagnosis has implications for the choice of postoperative adjuvant therapy, necessity for metastatic workup, planning of optimal long-term follow-up, and prediction of long-term prognosis. In patients with pineocytomas, completeness of excision seems to be the most important factor for avoiding disease progression. Tumors are well controlled after complete resection in spite of receiving no radiation therapy in pineocytomas [8] as also in pineocytomas or PPTIMD with cytologic pleomorphism [9]. On the other hand, there is no evidence that surgery improves the prognosis in pineoblastomas.
Occipital Transtentorial Approach and Combined Treatments
27
Table 1. Histological types and extent of resection with OTA in 93 pineal region tumors Histological type
Patients
Extent of tumor removal total
subtotal
partial
9
Germ cell tumors1 Germinomas Teratomas Embryonal carcinomas Yolk sac tumors Choriocarcinomas
44 17 13 6 6 2
4 11 6 5 1
4 2
PPTs Pineocytomas PPT of intermediate differentiation Pineoblastomas
17 7 6 4
5 4 2
2 1 2
Glial tumors Astrocytomas Anaplastic astrocytomas Glioblastomas Ependymomas
16 6 4 1 5
2 1
1 1 1 1
3 2
Miscellaneous Cavernous angioma Cancer metastasis Lipoma Meningioma Epidermoids
11 4 2 2 1 2
1
1 2
Pineal cyst
5
4
1 1
1
4
1 2 4
1
1
Mixed germ cell tumors are classified into each subgroup according to their dominant histopathological features.
Endoscopic biopsy with third ventriculostomy is becoming an important alternative for the initial surgical management of pineal region tumors including PPTs. These procedures can be performed in a single step that further includes tumor biopsy for histological diagnosis, cerebrospinal fluid sampling for tumor markers and cytological diagnosis, and resolution of hydrocephalus [10]. Seeding of the tumor by the endoscopic procedures is a matter of concern. However, for their series of 12 patients with intracranial germinomas, Shono et al. [11] reported that the risk of tumor dissemination due to neuroendoscopic procedures appears to be minimal when the appropriate chemotherapy and radiotherapy are provided
28
Tsumanuma · Tanaka · Fujii
a
b
Fig. 1. a T1-weighted contrast-enhanced magnetic resonance images of a 3-year-old boy with pineoblastoma. The heterogeneously enhanced tumor obstructs the aqueduct, displaces the splenium of the corpus callosum and cerebellar vermis, and extends into the third ventricle. b T1-weighted contrast-enhanced magnetic resonance images of a 40-year-old male patient with pineocytoma. The well-demarcated and homogeneously enhanced tumor causes obstructive hydrocephalus.
postoperatively. The risk of tumor seeding by endoscopic procedures for patients with PPTs remains largely unknown.
Preoperative Management
Diagnosis It is often difficult to differentially diagnose PPTs from pineal region germ cell tumors with no teratomatous component or glial tumors including ependymomas (fig. 1). Therefore, the initial treatment for PPTs is generally the same as that for pineal region tumors including tumors which can be distinguished from PPTs. PPTs have a propensity for leptomeningeal dissemination similar to germ cell tumors. Preoperative evaluation of leptomeningeal dissemination is essential for selecting appropriate postoperative adjuvant therapies and for predicting the prognosis of the patients. Whole neuraxis magnetic resonance imaging and cytological examination of the cerebrospinal fluid should be undertaken depending on whether the patient’s condition permits it. Tumor markers including human chorionic gonadotropin β-subunit, α- fetoprotein, and placental alkaline phosphatase in the serum or cerebrospinal fluid are assayed for the purpose of excluding germ cell tumors in the pineal region.
Occipital Transtentorial Approach and Combined Treatments
29
a
b
Fig. 2. Lateral semiprone position for the OTA with the approach side downward. Lateral (a) and anterior (b) views.
Neuroophthalmological evaluation is useful for detecting Parinaud’s syndrome (paralysis of upgaze or convergence, light-near pupillary dissociation, and retraction nystagmus) – which is caused by midbrain tectum compression by the tumor – and deterioration of visual acuity and choked disc caused by obstructive hydrocephalus.
Preoperative Care At the time of admission, most patients with a pineal region tumor have obstructive hydrocephalus associated with symptoms or signs of increased intracranial pressure. Administration of corticosteroids generally improves these symptoms, and patients can then tolerate the preoperative examinations described above. If the increased intracranial pressure is progressive, ventricular drainage is instituted before or during the craniotomy. Even in such cases, shunt operations should be avoided not only because hydrocephalus generally improves after tumor resection but also because it can result in peritoneal dissemination of the tumor cells via the shunting system.
Surgical Technique for Occipital Transtentorial Approach
Positioning (Lateral Semiprone Position) We preferentially use a lateral semiprone position with the approach side downward (fig. 2) [7]. The patient is placed in the head-up position by tilting the
30
Tsumanuma · Tanaka · Fujii
a
b
Fig. 3. a Lateral semiprone position for the OTA. The sagittal plane of the head is bent toward the surgeon standing at the back of the patient. b Skin incision and craniotomy for the OTA.
operating table, leaving the confluence of the venous sinuses approximately 10 cm above the right atrium to avoid air embolism. The head is rotated toward the floor at approximately a 60° angle. The neck is slightly extended and the sagittal plane of the head is bent toward the surgeon standing at the back of the patient (fig. 3). The occipital lobe on the approach side spontaneously falls away from the midline under the influence of gravity, which minimizes the retraction of the brain. The surgeon can look down not only at the pineal region but also at the dorsorostral or ipsilateral portions of the tumor with great comfort (fig. 3). The cerebellar vermis or aqueduct can also be viewed by standing at the parietal side of the patient.
Craniotomy and Tentorial Incision An occipitoparietal scalp flap and a craniotomy beyond the midline are created to expose the superior sagittal sinus and the superior sagittal sinus-straight sinus corner. The dura is opened in two flaps: one based on the superior sagittal sinus and the other on the transverse sinus. Ventricular drainage via the posterior horn of the ipsilateral lateral ventricle minimizes the retraction of the occipital lobe. Bridging veins that can disturb the retraction of the occipital lobe rarely present within this craniotomy. The tentorium is divided slightly obliquely to the straight sinus. The medial cut edge of the tentorium is pulled up and retracted to expose the posterior fossa widely.
Occipital Transtentorial Approach and Combined Treatments
31
Tumor Exposure The arachnoid spanning the quadrigeminal cistern is generally thick and cloudy in patients with pineal region tumors. It is sharply dissected to expose the deep venous system. The vein of Galen and its tributaries tightly guard the dorsorostral aspect of the tumor (fig. 4). The vein of Galen and both internal cerebral veins are located above the tumor. The internal cerebral veins are generally elevated by the tumor and hidden under the splenium of the corpus callosum. Both basal veins are displaced rostrally and laterally. The precentral cerebellar vein is displaced rostrally and dorsally. The ipsilateral basal vein and precentral cerebellar vein can hamper the access to the tumor in the OTA. The tumor must be resected through the space between these deep veins. It has been reported that the precentral cerebellar vein can be sacrificed to gain additional exposure without increasing the morbidity significantly [5].
Infrasplenial Approach to Expose the Internal Cerebral Veins The infrasplenial approach is useful for exposing the internal cerebral veins and the tumor through the velum interpositum cistern in an earlier stage of tumor resection. This optional approach is indicated when it is difficult to separate the tumor from the ventral aspect of the vein of Galen or the internal cerebral veins, or when internal decompression of the tumor is tough on account of its hardness, largeness, or hypervascularity. The arachnoid between the splenium and the vein of Galen is dissected, and the splenium is retracted 5–10 mm ventrally using a thin spatula to expose the internal cerebral veins displaced laterally by the tumor (fig. 4). Then the internal cerebral veins are separated from the tumor. The feeding arteries from both the medial posterior choroidal arteries can be coagulated in these procedures. Wide arachnoid dissection of the posterior interhemispheric fissure above the corpus callosum facilitates the infrasplenial approach.
Tumor Removal Pineal region tumors are removed through the narrow space between the important deep veins. Great attention should be paid in order to preserve these veins. Repeated internal decompression, piecemeal resection, and shrinking the tumor with coagulation of the tumor surface are mandatory to avoid deep venous system injury. The tumor is removed with a variety of techniques, such as suction, section with surgical scalpels and scissors, or ultrasonic aspiration.
32
Tsumanuma · Tanaka · Fujii
a
b
Fig. 4. A schematic drawing of the OTA (a) combined with the infrasplenial approach (b). The vein of Galen and its tributaries tightly guard the dorsorostral aspect of the tumor. The tumor must be resected through the space between these veins (1). The infrasplenial approach can expose the internal cerebral veins and the tumor through the velum interpositum cistern (2). F = Falx; GV = vein of Galen; PCV = precentral cerebellar vein; BV = basal vein; ICV = internal cerebral vein; SS = straight sinus; Sp = splenium of corpus callosum; Te = tentorium; Tu = tumor.
PPTs, except for invasive pineoblastomas, are generally well demarcated and can be separated from the surrounding structures including the thalamus, velum interpositum, cerebellar vermis, deep venous system, or the medial surface of the occipital lobe. The only attachment of the tumor to the brain is around the midbrain tectum and the posterior commissure. To avoid brainstem injury by the retraction of the tumor, the attachment should be detached at an earlier stage of the resection after adequate internal decompression of the tumor. Even while using the infrasplenial approach, it is occasionally difficult to confirm the completeness of resection of the tumor at the posterior roof of the third ventricle. Endoscopy is very useful in visualizing the velum interpositum, the inferior aspect of which is difficult to perceive under the microscopic view in the OTA (fig. 5).
Closure The tentorium divided in the approach procedures does not have to be sutured. The dura covering brain surface is closed, and the bone flap is replaced and fixed using titanium plates. The ventricular drainage system should remain in place for a few days even if the tumor is completely resected, because it takes 3–4 days for
Occipital Transtentorial Approach and Combined Treatments
33
Fig. 5. Endoscopic image showing residual tumor at the inferior aspect of the left internal cerebral vein. ICV = Internal cerebral vein; Tu = tumor; V = velum interpositum.
the obstruction of the aqueduct to be released after tumor resection. The aqueduct obstruction is generally relieved even after partial tumor resection and a shunt operation is unnecessary.
Complications There were no surgical deaths among our 17 patients with PPTs. However, permanent homonymous hemianopsia occurred in one patient who had huge and hard tumors, which likely reflected lengthy retraction of the occipital lobe. Permanent ocular movement disorders were not encountered. Bruce and Ogden [12] reported that the surgical major morbidity rate associated with pineal region tumors was 3–6.8% and the permanent minor morbidity rate was 3–28%.
Adjuvant Therapy
Little clinical data is available in the literature regarding the role of radiation therapy in the management of PPTs. Schild et al. [13] recommended craniospinal irradiation in patients with tumors with seeding potential (PPTs other than
34
Tsumanuma · Tanaka · Fujii
pineocytomas) and localized fields for pineocytomas. In the patients receiving radiation therapy with 50 Gy or more, 0 of 12 had local failure compared with 6 of 7 (86%) patients receiving lesser doses even in patients with PPTIMD or pineoblastomas. Furthermore, in the patients with PPTs other than pineocytomas, 1 of 7 (14%) developed leptomeningeal failure when treated with craniospinal irradiation, compared with 4 of 8 (50%) treated with lesser volumes. Radiation therapy should be considered for patients with pineocytomas that are resected incompletely. It is still controversial whether irradiation is required when the tumor is excised completely. Recently, in a clinicopathological study of 14 patients with pineocytomas or PPTIMD with cytologic pleomorphism, Fèvre-Montange et al. [9] reported that no tumor progression was encountered in any of the 10 patients who underwent complete resection of the tumor, even in the 7 patients who received no radiation therapy. Prophylactic spinal irradiation for patients with PPTs is still a matter of debate. Spinal irradiation is generally taken into account when magnetic resonance imaging shows evidence of leptomeningeal dissemination. Ghim et al. [14] treated 3 children with pineoblastomas with neoadjuvant chemotherapy consisting of etoposide, cisplatin, and vincristine followed by craniospinal irradiation. Although 2 patients remained in complete remission (CR) or near CR, one died of the progressing tumor. Combined chemotherapy and radiotherapy may be effective in older children with pineoblastoma. Hinkes et al. [15] studied a series of 11 children with pineoblastoma treated within prospective multicenter trials. All of 5 children younger than 3 years of age who received chemotherapy (carboplatin, vincristine, etoposide, procarbazine, vincristine, or methotrexate) after surgery died of tumor progression with a median progression-free survival of 0.6 years. In contrast, 5 of 6 children older than 3 years who received chemotherapy (ifosfamide, etoposide, high-dose methotrexate, cisplatin, or cytarabin) and craniospinal irradiation immediately after the surgery remained in CR with a median progression survival of 7.9 years. A recent trial has shown that the outcome of patients with relapses remained poor even with a multimodal high-dose chemotherapy regimen [16].
Prognosis
Schild et al. [13] reported good prognosis of pineocytomas with 1-year, 3-year, and 5-year survival rates of 100, 100, and 67%, respectively; the median survival rate was 88%. However, other PPTs including pineoblastomas, PPTIMD, and mixed PPTs had the propensity for leptomeningeal failure with 1-, 3-, and 5-year survival rates of 88, 78, and 58%, respectively. Min et al. [17] classified PPTs into pineoblastomas, PPT of intermediate or mixed differentiation, and pineocytomas based on
Occipital Transtentorial Approach and Combined Treatments
35
a combination of their light microscopic and ultrastructural features. Although pineoblastomas behaved as highly malignant tumors, a correlation between the morphology and prognosis was less evident between intermediate tumors and pineocytomas. Pineoblastomas generally occur with a predilection for children, whereas pineocytomas most frequently affect adults [18]. Chang et al. [19] reported that among 11 adult patients with pineoblastomas, 4 patients died of the disease with a median survival of 10 months and 5 patients remained alive without disease progression after a median follow-up of 26 months. Lutterbach et al. [20] documented that the 3-, 5-, and 10-year survival rates of adult patients with pineoblastomas were 88, 78, and 58%, respectively. D’Andrea et al. [21], on the other hand, reported that 4 of 6 children with pineocytomas had tumor recurrence at a median of 2 years after diagnosis; further, these tumors were considered to be aggressive with a high propensity for leptomeningeal dissemination in the pediatric population. As mentioned above, pineoblastomas are highly malignant tumors with a propensity for leptomeningeal dissemination, and pineocytomas have a good prognosis if they are properly treated with surgery and additional radiation therapy. However, the incidence and prognosis of each histological subtype including PPTIMD vary considerably in the literature. The characterization of PPTs has been far from adequate and no firm diagnostic criteria – light microscopic or ultrastructural – have been established. The heterogeneity of management also renders the interpretation of published data more difficult.
Clinicopathological Study
We have previously reported a clinicopathological study of 13 patients with PPTs for whom the tumor specimens and precise clinical data were available [22]. The data described below include these 13 patients and 1 additional patient (patient 9). In the previous report [22], we classified the tumors showing intermediate differentiation as ‘pineocytoma with anaplasia’ based on the histological findings including Bodian’s staining. This category is very similar to ‘PPTIMD’ described later in the WHO 2000 Classification [18].
Patient Characteristics The patients evaluated consisted of 9 male and 5 female patients with an age range of 0–73 years (mean, 30.4 years; median, 28 years), and the follow-up periods from the time of diagnosis ranged from 7 to 294 years (mean, 103.9 months; median, 111 months). Clinical data are briefly summarized in table 2.
36
Tsumanuma · Tanaka · Fujii
Histopathological Diagnosis In cases 1–4, the tumors were moderately cellular with round to oval-shaped nuclei. The tumor cells had abundant cytoplasmic processes and were arranged in a sheet-like or ill-defined lobular pattern, which consisted of pineocytomatous rosettes. Occasional tumor cells with markedly large nuclei were seen. Mitotic figures were rare or absent. These tumors were diagnosed as pineocytomas (fig. 6). In cases 5–8, the tumors were highly cellular with intermediate-sized nuclei showing moderate atypia. The tumor cells had scant cytoplasmic processes. Mitotic figures were occasional or absent. In cases 5–7, a few Homer-Wright rosette-like structures were seen. Bodian’s staining revealed argyrophilic processes with terminal expansions in the center of the rosette (data not shown). These tumors were classified into PPTIMD. In case 9, the tumor tissue was characterized for the major part by features suggestive of pineocytoma, but it also included a localized area showing nuclear atypia and necrosis; this case was also diagnosed as PPTIMD (fig. 6). In cases 10–14, the tumors were composed of densely packed cells with scanty, ill-defined, and wispy cytoplasm. The tumors had polygonal or carrot-shaped, small and hyperchromatic nuclei with marked atypia. The cells were diffusely arranged and interrupted only by Homer-Wright rosettes or Flexner-Wintersteiner rosettes. Hemorrhage, necrosis, and prominent mitotic figures were common. No argyrophilic processes were demonstrated by Bodian’s stain (data not shown). These findings are common in primitive neuroectodermal tumors. These tumors were diagnosed as pineoblastomas (fig. 6).
Extent of Tumor Removal and Outcomes Pineocytomas were resected totally in 2 out of 4 patients and subtotally in 2 other patients. Adjuvant external-beam radiation therapies, including prophylactic whole neuraxis irradiation, were carried out for all the patients. In the 5 PPTIMD patients, the tumors were resected totally in 3 patients and subtotally in 1 patient. Radiation therapies, including prophylactic whole neuraxis irradiation, were performed for all of these 4 patients. Another case (case 8) received local radiation therapy with no biopsy. Although the tumor regressed after radiation, it recurred at the left cerebellopontine angle. With the use of open biopsy the recurrent tumor was diagnosed as PPTIMD, and local brain radiation therapy resulted in CR. Eventually, the patient died of leptomeningeal dissemination of the PPTIMD, which was confirmed on autopsy. Apart from this one case and another pineocytoma patient who died of other causes, the patients with pineocytomas or PPTIMD are still alive with no progression of the tumor. Completeness of excision
Occipital Transtentorial Approach and Combined Treatments
37
a
d
b
e
c
Fig. 6. Histopathological findings in the PPTs. a Pineocytoma (case 2). Moderately cellular tumor cells with abundant cytoplasmic processes are arranged in a sheet-like pattern. b PPT of intermediate differentiation (case 8). The tumors are highly cellular and the nuclei have medium-toabundant chromatin and show moderate atypia. Tumor cells have scant cytoplasmic processes. c PPT of intermediate differentiation (case 5) showing Homer-Wright rosette-like structures. d Pineoblastoma (case 14). Densely packed tumor cells with small and hyperchromatic nuclei with marked atypia are diffusely arranged and interrupted only by Homer-Wright rosettes. e Pineoblastoma (case 11) showing Flexner-Wintersteiner rosettes. HE staining.
38
Tsumanuma · Tanaka · Fujii
100 Pineocytoma and PPTIMD
Percent surviving
80
60 Pineoblastoma 40
20
0 0
100 200 Survival (months)
300
Fig. 7. Kaplan-Meier estimate of overall survival for all patients with PPTs. The survival of the patients with pineocytomas or PPTIMD is significantly longer than that of the patients with pineoblastomas (p = 0.00022, log-rank test).
appears to be the most important factor for avoiding disease progression; this has also been mentioned in previous reports [9, 13]. Pineoblastomas were removed totally in 1 patient, subtotally in 1 patient and partially in 2 patients. Another patient underwent local brain radiation with no surgical intervention. In spite of adjuvant radiation therapy and/or chemotherapy, all the 5 patients with pineoblastoma died of leptomeningeal dissemination of the tumor within 14 months after the time of diagnosis. As previously reported [13], surgery does not improve the prognosis in pineoblastomas. Hinkes et al. [15] reported a better prognosis in children older than 3 years who received chemotherapy and craniospinal irradiation after the surgery. Unfortunately, we have no experience of older children with pineoblastoma except for an 8-year-old boy who was treated only with radiation therapy and died due to leptomeningeal dissemination of the tumor. This may be the reason for the prognosis of our patients with pineoblastoma being extremely poor (fig. 7).
Proliferative Potential and Neuronal or Neurosecretory Characteristics The MIB-1 staining indices in the 4 pineocytomas, 5 PPTIMD, and 4 pineoblastomas examined are shown in table 2. The indices for the pineocytomas and PPTIMD specimens were less than 7% but those for all the pineoblastomas examined were more than 8%. The mean MIB-1 staining index of pineoblastomas was
Occipital Transtentorial Approach and Combined Treatments
39
Table 2. Case representations Patient Age no. years
Sex
Extent of resection
Adjuvant therapy
Outcome
Suvival months
MIB-1 index %
Pineocytoma 1 25
M
subtotal
progression free
2511
6.5
2
40
M
total
progression free
1841
1.0
3
44
F
subtotal
72
6.3
4
57
F
total
whole neuraxis and local brain Rx whole neuraxis and local brain Rx whole neuraxis and local brain Rx local brain Rx
progression free
2941
1.6
local brain Rx local brain Rx whole neuraxis and local brain Rx local brain Rx
progression free progression free progression free
1241 1351 361
3.4 1.6 2.0
CR; recurrence at left 168 cerebellopontine angle 62 months later, whole neuraxis and local brain Rx after biopsy resulting in CR; leptomeningeal dissemination whole neuraxis and progression free 1361 local brain Rx
6.2
PPT of intermediate differentiation 5 61 M total 6 73 F total 7 23 F subtotal 8
31
F
ND
9
57
M
total
Pineoblastoma 10 8 M
ND
11
0
M
partial
12
1
M
subtotal
13
2
M
partial
14
3
M
total
local brain Rx
died of other disease
leptomeningeal dissemination local brain Rx, Cx leptomeningeal (BLM, MTX) dissemination Cx (CBDCA, VP16) PD in spite of whole neuraxis Rx and Cx (VBL, CTX, DACT, BLM, CDDP); leptomeningeal dissemination whole neuraxis and PR; leptomeningeal local brain Rx, Cx dissemination, whole brain (ACNU) and local Rx, Cx (ACNU, MTX) whole neuraxis and CR; leptomeningeal local brain Rx, Cx dissemination, Cx (CBDCA, (CBDCA, VP16) VP16, L-PAM) and local Rx
0.5
10
ND
9
8.2
14
15.8
10
9.1
14
29.5
ND = Not done; Rx = irradiation; Cx = chemotherapy; BLM = bleomycin; MTX = methotrexate; CBDCA = carboplatin; VP16 = etoposide; ACNU = nimustine; VBL = vinblastine; CTX = cyclophosphamide; DACT = actinomycin D; CDDP = cisplatin; L-PAM = melphalan; CR = complete response; PR = partial response; PD = progressive disease. 1 Alive. 2 Died of other disease.
40
Tsumanuma · Tanaka · Fujii
Table 3. Correlations between histopathological features and mean MIB-1 indices Histopathological features
MIB-1 index, %
Diagnosis Pineocytoma PPTIMD Pineoblastoma
3.9 ± 3.0a 2.7 ± 2.2b 15.7 ± 9.8a, b
Neurofilament protein + –
3.7 ± 3.4c 22.7 ± 9.7c
Chromogranin A + –
6.1 ± 7.0 10.8 ± 13.0
Common superscripts indicate statistically significant difference (p < 0.05). Values are expressed as mean ± SD.
statistically significantly higher than those of pineocytomas and PPTIMD (p < 0.05, Mann-Whitney test). However, there was no significant difference between pineocytomas and PPTIMD (table 3). It is generally believed that PPTs exhibit a continuous spectrum of differentiation, the extent of which is paralleled by differences in biological behavior and clinical outcomes. The proliferative potential of PPTs has been reported by a few authors. Mena et al. [23] investigated 35 patients with PPTs using silver-stained nucleolar organizer region (AgNOR) counts and showed that pineoblastomas were more actively proliferative than pineocytomas, with mixed pineocytoma/pineoblastomas showing intermediate activity. However, no correlation was observed between the mean AgNOR score and the prognosis within tumor groups. Our data suggest that pineoblastomas are more actively proliferative than pineocytomas or PPTIMD and that the proliferative potential of PPTIMD is equivalent to that of pineocytomas. Is the spectrum of biological behavior in PPTs continuous? On the basis of our observations, with regard to the malignant potential, we emphasize that a clear distinction should be made between pineoblastomas in children and other types of PPTs in adults. In our immunohistochemical characterization studies, synaptophysin, a 38-kDa glycoprotein found in the synaptic vesicle membranes of neuroendocrine cells, was immunopositive in all of the 2 pineocytoma, 4 PPTIMD, and 3 pineoblastoma specimens examined. On the other hand, both pineocytomas, all of the 3 PPTIMD, and 1 of 3 pineoblastomas examined were immunopositive for the 160-kDa neurofilament protein. The mean MIB-1 index in neurofilament
Occipital Transtentorial Approach and Combined Treatments
41
protein-immunopositive specimens was significantly lower than that in immunonegative specimens (p < 0.05, Mann-Whitney test; table 3). Further, 1 of 2 pineocytomas, 2 of 3 PPTIMD, and 1 of 3 pineoblastomas examined were immunopositive for chromogranin A, a member of the secretogranin/chromogranin class of proteins expressed in the secretory granules of endocrine cells and neurons. The level of expression of cytoskeletal elements in PPT is likely to be inversely associated with the proliferative potential of the tumor cells. We reported that the in situ expression of hydroxyindole-O-methyl transferase, the enzyme catalyzing the final step of the melatonin biosynthesis, is observed not only in pineocytomas but also in pineoblastomas [24]. This observation, as well as the synaptophysin immunopositivity in pineoblastomas, suggests that the neurosecretory characteristics of the pinealocytes are retained not only in pineocytomas but also in pineoblastomas, which are the most undifferentiated and proliferative tumors among PPTs.
Conclusion
The OTA using lateral semiprone position can remove PPTs extensively and safely. Combining the infrasplenial approach with the OTA is useful in visualizing the internal cerebral veins directly, particularly when the tumor is tightly adherent to the ventral aspect of the vein of Galen and the internal cerebral veins. Endoscopy can increase the visibility of the posterior part of the third ventricular roof. Extensive removal of the tumor significantly prolongs survival at least in the patients with pineocytomas and PPTIMD, the proliferative potentials of which are likely to be the same. In spite of extensive resection and adjuvant radiochemotherapy, the prognosis of the patients with pineoblastomas is extremely poor. Further elucidation of the biological features of pineoblastomas is required to resolve this issue.
References 1 Krabbe KH: The pineal gland, especially in relation to the problem on its supposed significance in sexual development. Endocrinology 1923;7:379–414. 2 del Rio-Hortega: Pineal gland; in Penfield W (ed): Cytology and Cellular Pathology of the Nervous System. New York, Hoeber, 1932, vol 2, pp 636–703. 3 Kleihues P, Burger PC, Scheithauer BW: Histological Typing of Tumours of the Central Nervous System. World Health Organization International Histological Classification of Tumours, ed 2. Berlin, Springer, 1993.
42
4 Louis DN, Ohgaki H, Wiestler OD, Cavenee WK, Burger PC, Jouvet A, Scheithauer BW, Kleihues P: The 2007 WHO classification of tumours of the central nervous system. Acta Neuropathol 2007;114: 97–109. 5 Stein BM: The infratentorial supracerebellar approach to pineal lesions. J Neurosurg 1971;35: 197–202. 6 Lazar ML, Clark K: Direct surgical management of masses in the region of the vein of Galen. Surg Neurol 1974;2:17–21.
Tsumanuma · Tanaka · Fujii
7 Tanaka R, Washiyama K: Occipital transtentorial approach to pineal region tumors. Oper Tech Neurosurg 2003;6:215–221. 8 Vaquero J, Ramiro J, Martínez R, Coca S, Bravo G: Clinicopathological experience with pineocytomas: report of five surgically treated cases. Neurosurgery 1990;27:612–619. 9 Fèvre-Montange M, Szathmari A, Champier J, Mokhtari K, Chrétien F, Coulon A, FigarellaBranger D, Polivka M, Varlet P, Uro-Coste E, Fauchon F, Jouvet A: Pineocytoma and pineal parenchymal tumors of intermediate differentiation presenting cytologic pleomorphism: a multicenter study. Brain Pathol 2008;18:354–359. 10 Chernov MF, Kamikawa S, Yamane F, Ishihara S, Kubo O, Hori T: Neurofiberscopic biopsy of tumors of the pineal region and posterior third ventricle: indications, technique, complications, and results. Neurosurgery 2006;59:267–277. 11 Shono T, Natori Y, Morioka T, Torisu R, Mizoguchi M, Nagata S, Suzuki SO, Iwaki T, Inamura T, Fukui M, Oka K, Sasaki T: Results of a long-term followup after neuroendoscopic biopsy procedure and third ventriculostomy in patients with intracranial germinomas. J Neurosurg 2007;107:193–198. 12 Bruce JN, Ogden AT: Surgical strategies for treating patients with pineal region tumors. J Neurooncol 2004;69:221–236. 13 Schild SE, Scheithauer BW, Schomberg PJ, Hook CC, Kelly PJ, Frick L, Robinow JS, Buskirk SJ: Pineal parenchymal tumors. Clinical, pathologic, and therapeutic aspects. Cancer 1993;72:870–880. 14 Ghim TT, Davis P, Seo JJ, Crocker I, O’Brien M, Krawiecki N: Response to neoadjuvant chemotherapy in children with pineoblastoma. Cancer 1993;72:1795–1800. 15 Hinkes BG, von Hoff K, Deinlein F, WarmuthMetz M, Soerensen N, Timmermann B, Mittler U, Urban C, Bode U, Pietsch T, Schlegel PG, Kortmann RD, Kuehl J, Rutkowski S: Childhood pineoblastoma: experiences from the prospective multicenter trials HIT-SKK87, HIT-SKK92 and HIT91. J Neurooncol 2007;81:217–223.
16 Broniscer A, Nicolaides TP, Dunkel IJ, Gardner SL, Johnson J, Allen JC, Sposto R, Finlay JL: High-dose chemotherapy with autologous stem-cell rescue in the treatment of patients with recurrent non-cerebellar primitive neuroectodermal tumors. Pediatr Blood Cancer 2004;42:261–267. 17 Min KW, Scheithauer BW, Bauserman SC: Pineal parenchymal tumors: an ultrastructural study with prognostic implications. Ultrastruct Pathol 1994;18:69–85. 18 Mena H, Nakazato Y, Jouvet A, Scheithauer BW: Pineal parenchymal tumours; in Kleihues P, Cavenee WK (eds): World Health Organization Classification of Tumours. Pathology and Genetics of Tumours of the Nervous System. Lyon, IARC Press, 2000, pp 115–122. 19 Chang SM, Lillis-Hearne PK, Larson DA, Wara WM, Bollen AW, Prados MD: Pineoblastoma in adults. Neurosurgery 1995;37:383–391. 20 Lutterbach J, Fauchon F, Schild SE, Chang SM, Pagenstecher A, Volk B, Ostertag C, Momm F, Jouvet A: Malignant pineal parenchymal tumors in adult patients: patterns of care and prognostic factors. Neurosurgery 2002;51:44–56. 21 D’Andrea AD, Packer RJ, Rorke LB, Bilaniuk LT, Sutton LN, Bruce DA, Schut L: Pineocytomas of childhood. A reappraisal of natural history and response to therapy. Cancer 1987;59:1353–1357. 22 Tsumanuma I, Tanaka R, Washiyama K: Clinicopathological study of pineal parenchymal tumors: correlation between histopathological features, proliferative potential, and prognosis. Brain Tumor Pathol 1999;16:61–68. 23 Mena H, Rushing EJ, Ribas JL, Delahunt B, McCarthy WF: Tumors of pineal parenchymal cells: a correlation of histological features, including nucleolar organizer regions, with survival in 35 cases. Hum Pathol 1995;26:20–30. 24 Tsumanuma I, Tanaka R, Ichikawa T, Washiyama K, Kumanishi T: Demonstration of hydroxyindoleO-methyltransferase (HIOMT) mRNA expression in pineal parenchymal tumors: histochemical in situ hybridization. J Pineal Res 2000;28:203–209.
Itaru Tsumanuma, MD, PhD Department of Neurosurgery Yamagata Prefectural Central Hospital 1800 Aoyagi Yamagata 990-2292 (Japan) Tel. +81 23 685 2626, Fax +81 23 685 2608, E-Mail
[email protected]
Occipital Transtentorial Approach and Combined Treatments
43
Tumors of Pineal Cell Origin Kobayashi T, Lunsford LD (eds): Pineal Region Tumors. Diagnosis and Treatment Options. Prog Neurol Surg. Basel, Karger, 2009, vol 23, pp 44–58
Role of Stereotactic Radiosurgery in the Management of Pineal Parenchymal Tumors Hideyuki Kano ⭈ Ajay Niranjan ⭈ Douglas Kondziolka ⭈ John C. Flickinger ⭈ L. Dade Lunsford Departments of Neurological Surgery and Radiation Oncology, The University of Pittsburgh, and Center for Image-Guided Neurosurgery, University of Pittsburgh Medical Center, Pittsburgh, Pa., USA
Abstract We evaluated 20 pineal parenchymal tumor patients who underwent radiosurgery at our institution over a 20-year period. Thirteen patients had pineocytoma, 5 patients had pineoblastoma and 2 patients had mixed pineal parenchymal tumors. The median radiosurgery prescription dose to the tumor margin was 15.0 (12–20) Gy. At an average of 54.1 (range, 7.7–149.2) months, 6 patients had died and 14 patients were living. The overall survival after radiosurgery was 95.0, 68.6, and 51.4% at 1, 5 and 10 years, respectively. Patients with pineocytomas had 1-, 3- and 5-year overall survivals of 100, 92.3 and 92.3%, respectively. In 19 patients who were evaluated with imaging, 5 (26%) demonstrated complete regression, 9 (47%) had partial regression, 2 (11%) had stable tumors and 2 (11%) showed local in-field progression. The progression-free survival after stereotactic radiosurgery for all type of pineal parenchymal tumors was 100, 89.2 and 89.2% at 1, 3, 5 years after radiosurgery, respectively. Stereotactic radiosurgery is an effective and safe alternative to the surgical resection of pineocytomas as well as part of multimodal therapy for more aggressive pineal parenchymal tumors. Copyright © 2009 S. Karger AG, Basel
Pineal region tumors (PRTs) are rare neoplasms. PRTs account for 0.4–1% of intracranial tumors in Western countries, but their incidence is higher (2.2–8% of intracranial tumors) in northeastern Asian countries [1–3]. Germ cell tumors are the most frequent, of which germinomas and teratomas account for 25–53.5% of all tumors in the pineal region [4–6]. Pineal parenchymal tumors (PPTs) are the second most common pineal tumors in adults and account for 15–30% of PRTs [7]. The designation of PPT includes pineocytomas, PPTs of intermediate differentiation or mixed pineocytomas/pineoblastomas, and pineoblastomas. Pineocytoma is a slowly growing tumor with a relatively favorable prognosis in
most cases. In contrast, pineal parenchymal tumors of intermediate differentiation have less predictable growth rates and clinical behavior. Pineoblastoma is the malignant variant which has many common features with medulloblastoma and other primitive neuroectodermal tumors. Metastatic spread of pineoblastoma via cerebrospinal fluid pathways is a frequent complication and is often associated with a fatal outcome. Despite improved microsurgical techniques, resection of PPT remains a challenge because of their deep location and associated critical structures [8]. The role of adjuvant radiotherapy has not been clearly defined. The role of radiosurgery in the treatment of PPT is not well documented as only a few studies have been published. The purpose of this report is to review our experience with radiosurgery for PPT. We retrospectively evaluated survival, imaging response, and treatmentrelated morbidity in patients who underwent radiosurgery for PPTs.
Methods and Materials Patient Population
Twenty histologically confirmed PPT patients had stereotactic radiosurgery (SRS) at the University of Pittsburgh between April 1989 and January 2006. The series included 12 males and 8 females with a median age of 34 years (range, 3.5–68.4 years). Six patients had undergone prior surgical resection of their PPTs and 15 patients had either stereotactic or open biopsies (table 1). Prior adjuvant management included fractionated radiation therapy (RT; n = 3), chemotherapy (n = 3), and both RT and chemotherapy (n = 2). All pathological data were reviewed by neuropathologists. Thirteen patients had pineocytomas which corresponded to WHO grade 2 tumors. Five patients had pineoblastomas which corresponded to WHO grade 4 tumors. Two patients had mixed PPTs. All tumors were located in the pineal region.
Radiosurgery Technique Our radiosurgical technique has been described in detail in previous reports [9]. In brief, patients underwent application of an imaging-compatible stereotactic head frame. The procedure was performed under local anesthesia for patients over 13 year of age and under general anesthesia for children younger that 13. After frame application, high resolution magnetic resonance imaging (MRI) was performed. Patients underwent either a sagittal scout MRI or a 3-D localizer sequence which included axial, coronal and sagittal images. The tumor was then imaged using contrast-enhanced volume acquisition images. T2-weighted MR images using Fast Spin Echo technique also were acquired to assess the infiltrative tumor volume. The target volume included enhanced and nonenhanced tumor regions. In all patients, the radiosurgery dose was prescribed to the whole tumor volume. The median tumor volume was 3.1 (0.9–14.2) cm3. A median of five isocenters (1–10) were used for dose planning. The median prescription dose delivered to the tumor margin was 15.0 (12–20) Gy. The maximum dose varied from 24 to 40 Gy (median, 30.0 Gy). SRS was performed
Radiosurgery for Pineal Parenchymal Tumors
45
Table 1. Summary of 20 PPT patients treated by SRS Characteristics
Patients
Sex Male Female
12 8
Prior surgical removal Prior biopsy Prior FRT Prior chemotherapy Prior FRT + chemotherapy
6 15 3 3 2
Histopathology Pineocytoma Pineoblastoma Mixed pineal parenchymal tumor
13 5 2
Target volume, cm3 Margin dose, Gy Maximum dose, Gy Age, years
Median
Mean
Range
3.1 15.0 30.0 34.4
4.4 15.2 30.4 33.6
0.9 ∼14.2 12∼20 24∼40 3.5∼68.4
FRT = Fractionated radiation therapy
with either a Model U, B, C, or 4-C Leksell Gamma Knife (Elekta Inc., Atlanta Ga., USA). All patients received an intravenous dose of 20–40 mg methylprednisolone after radiosurgery and all were discharged from the hospital within 24 h. Patients were evaluated clinically and radiologically using MR imaging at intervals of 3–6 months. Nineteen patients had follow-up that varied from 7.7 to 149.2 months. One patient had no radiological images after SRS. Fourteen patients (70.0%) had follow-up of 24 months or more. The mean follow-up time was 54.1 (range, 7.7–149.2) months. The follow-up MR images were compared with the intraoperative images and tumor dimensions were measured in axial, sagittal, and coronal planes. A complete response was defined as the complete disappearance of enhancing or nonenhancing tumor; partial response was defined as a shrinkage >50% of the tumor area; stable disease was defined as a reduction of tumor area from 0 to 50%, and progressive disease was defined as more than 25% increase in size of enhancing or nonenhancing tumor. Tumor growth adjacent to the irradiated tumor and outside the isodose volume was defined as marginal recurrence; tumors that progressed outside the SRS-treated volume were considered distant recurrences. For statistical analysis, we constructed Kaplan-Meier plots for survival and progression-free survival using the dates of diagnosis, first surgery, first SRS, follow-up MRIs, and death or last follow-up. Progression-free survival and overall survival time were calculated from the day of the first SRS using the Kaplan-Meier method. Univariate analysis was performed on the
46
Kano · Niranjan · Kondziolka · Flickinger · Lunsford
Table 2. Tumor type and response after radiosurgery, based on imaging results Lesions Tumor type
CR
PR
SD
PD
No images
Pineocytoma Pineoblastoma/mixed PPT
3 2
8 1
2 1
0 2
0 1
CR = Complete response; PR = partial response; SD = stable disease; PD = progressive disease.
Kaplan-Meier curves using a log rank statistic with p < 0.05 set as significant. Standard statistical processing software (SPSS, version 15.0) was used. This retrospective study was approved by the University of Pittsburgh institutional review board. All chart reviews and data analyses were conducted before October 31, 2007.
Results
Fourteen patients (70.0%) were alive and 6 (30.0%) had died at an average follow-up of 42.5 (range 7.7–149.2) months from radiosurgery, and 45.2 (range 7.8–149.8) months from their initial diagnosis. One pineocytoma and one pineoblastoma patient died secondary to dissemination throughout the neuraxis at 14.7 and 18.6 months after SRS, respectively. One patient died from systemic metastasis from pineoblastoma at 7.7 months after SRS. One pineoblastoma patient had tumor progression and died at 12.8 months after SRS. One pineocytoma and one pineoblastoma patient died from an unknown cause at 149.2 and 48.2 months after SRS, respectively. The mean follow-up time after SRS in the entire series was 54.1 (range, 7.7– 149.2) months. Follow-up imaging studies demonstrated tumor control in 17 (89.5%) of 19 lesions at a median and mean of 43.7 and 54.7 months after SRS, respectively. In 19 patients who were evaluated with follow-up imaging, 5 (3 pineocytomas and 2 pineoblastomas) had complete tumor resolution, 9 (8 pineocytomas and 1 pineoblastoma) had tumor regression, and 3 (2 pineocytomas and 1 pineoblastoma) had stable tumors. Local in-field tumor progression was seen in 2 patients with pineoblastomas at 12.8 and 31.2 months (table 2). One patient underwent repeat radiosurgery for local tumor progression. At the time of this analysis, 14 patients were alive at a median and mean of 43.7 and 59.0 months after radiosurgery, respectively (range 19.2–148.7 months). For surviving patients, the median and mean survival was 47.1 and 66.5 months after the initial diagnosis (range 25.8–150.2). The overall survival after radiosurgery
Radiosurgery for Pineal Parenchymal Tumors
47
was 95.0, 68.6, and 51.4% at 1, 5 and 10 years, respectively. Of the 20 patients, 4 (20%) had intracranial (n = 2) or systemic metastasis (n = 2) after SRS. The progression-free survival after SRS was 100, 89.2 and 89.2% at 1, 3, 5 years after radiosurgery, respectively.
Statistical Analysis We performed a univariate analysis using the log rank test to assess factors that might influence the length of overall and progression-free survival. The following variables were assessed: sex (male or female), age (older or younger than 21 years), histopathology (pineocytoma or pineoblastoma/mixed PPT), radiosurgical target volume (≥ or <3.0 cm3), marginal dose (≥ or <15 Gy). In univariate analysis, the histopathology of pineocytoma was associated with an improved overall survival (fig. 1). Patients with intracranial or systemic metastasis after SRS had significantly shorter survival time after radiosurgery compared to patients who did not have metastasis (p = 0.0006; fig. 2). In univariate analysis only the histopathology of pineocytoma was associated with an improved progressionfree survival (fig. 3).
Tumor Grade Two of 13 (15.4%) patients with pineocytoma died and 4 of 7 (57.1%) patients with pineoblastoma or mixed pineocytoma/pineoblastoma died. Patients with pineocytomas had 1-, 3- and 5-year overall survivals of 100, 92.3 and 92.3%, respectively. Patients with pineoblastoma or mixed pineocytoma/pineoblastoma had 1-, 3- and 5-year overall survivals of 85.7, 57.1 and 28.6%, respectively (fig. 1). Patients with low-grade tumors had significantly better overall survivals (p = 0.020). None of the pineocytoma patients showed progression in the SRS volume (fig. 4). In contrast, 2 of 6 (33.3%) patients with pineoblastoma or mixed pineocytoma/ pineoblastoma exhibited local progression within the SRS volume (table 2). Patients with pineocytomas had 1- and 5-year progression-free survivals of 100% and 100%, respectively. Patients with pineoblastoma or mixed pineocytoma/pineoblastoma had 1- and 3-year progression free survivals of 100 and 66.7%, respectively (table 3; fig. 3; 5-year survival could not be assessed). Patients with low-grade tumors had significantly better progression-free survivals (p = 0.033). Other variables (age, sex, target volume, margin dose and RT and/or chemotherapy) were not significantly associated with better progression-free survival (table 4).
48
Kano · Niranjan · Kondziolka · Flickinger · Lunsford
1.0
Pineocytoma
Overall survival
0.8
0.6
0.4
Pineoblastoma and mixed PPT
0.2
0 0
24
48
72 96 Time (months)
120
144
Fig. 1. Kaplan-Meier estimate of overall survival curve in all patients after SRS with pineocytoma or pineoblastoma and mixed PPT. Pineocytoma was a statistically significant factor for longer overall survival (p = 0.0197).
1.0
Overall survival
0.8 Metastases (–) 0.6
0.4 Metastases (+) 0.2
0 0
24
48
72 96 Time (months)
120
144
Fig. 2. Kaplan-Meier estimate of overall survival curve in all patients after SRS with metastases and without metastases. The patients with intracranial or systemic metastasis after SRS had significantly shorter survival time after radiosurgery than patients who had no metastasis (p = 0.0006).
Radiosurgery for Pineal Parenchymal Tumors
49
1.0 Pineocytoma Progression free survival
0.8
0.6
Pineoblastoma and mixed PPT
0.4
0.2
0 0
24
48
72 96 Time (months)
120
144
Fig. 3. Kaplan-Meier progression-free survival curves in all patients with pineocytoma or pineoblastoma and mixed PPT after the SRS. Pineocytoma was a statistically significant factor for longer progression-free survival (p = 0.0327).
a
b
Fig. 4. a T1-weighted contrast-enhanced axial MR image of a 33-year-old male show a pineocytoma at the time of SRS. b Axial T1-weighted contrast-enhanced MR image obtained 2 years after SRS showing complete regression of enhanced area. His diplopia improved.
Metastasis to Central Nervous System and Other Organs Four (2 pineocytomas and 2 pineoblastomas) patients developed central nervous system (CNS) and systemic metastases. Three (2 pineocytomas and 1 pineoblastoma) of 4 patients had multiple neuraxis metastases despite having undergone craniospinal radiotherapy (fig. 5). One pineoblastoma patient had multiple systemic metastases to adrenal gland and pelvis. The median time to develop CNS
50
Kano · Niranjan · Kondziolka · Flickinger · Lunsford
Table 3. Survival based on tumor grade Pineocytoma
Pineoblastoma mixed PPT
Total
Survival rate Actuarial 1-year Actuarial 3-year Actuarial 5-year
84.6 100 92.3 92.3
42.9 85.7 57.1 28.6
70.0 95.0 80.0 68.6
Recurrence-free rate 1-Year PFS 3-Year PFS 5-Year PFS
100 100 100 100
66.7 100 66.7 66.7
89.5 100 89.2 89.2
PFS = Progression-free Survival. Rates are expressed as percentages.
Table 4. Results of univariate and multivariate analysis for 20 patients treated for tumors, with OAS and PFS measured from date of radiosurgery p value OAS Age (21 Years old) Sex Tumor grade Metastases Prior FRT and/or chemotherapy
0.8701 0.4400 0.0197* 0.0006* 0.2692
PFS Age (21 Years old) Sex Tumor grade Targer volume (3.0 cm3) Margin dose (15 Gy) Prior FRT and/or chemotherapy
0.8105 0.2576 0.0327* 0.0119 0.6111 0.3462
OAS = Overall survival. *p<0.05.
or systemic metastasis was 6.0 (range, 0.5–56.1) months after diagnosis and 2.9 (range, 0.4–47.1) months after SRS. Three of 4 patients with the CNS or systemic metastases died at a median of 14.7 (range, 7.7–48.2) months after SRS. Patients with CNS or systemic metastases had 1-, 3- and 5-year overall survivals of 100, 93.8 and 80.4%, respectively. Patients without metastases had 1-, 3- and 5-year
Radiosurgery for Pineal Parenchymal Tumors
51
a
b
c
d
Fig. 5. A 61-year-old male underwent stereotactic biopsy for a pineocytoma. a Axial T1-weighted contrast-enhanced MR images show an enhanced tumor at the time of SRS. b Coronal T1-weighted contrast-enhanced MR images show an enhanced tumor at the time of SRS. c Axial T1-weighted contrast-enhanced MR images obtained 4.4 months after SRS show regression of enhancing tumor, but reveal a left temporal metastasis. d Coronal T1-weighted contrast-enhanced MR images obtained 4.4 months after SRS show regression of enhancing tumor with central necrosis of the tumor. The patient died 14.7 months after SRS because of dissemination of the tumor.
overall survivals of 75.0, 25.0 and 25.0%, respectively (fig. 2). Patients with intracranial or systemic metastasis after SRS had significantly shorter survival after radiosurgery compared to patients who did not develop metastasis (p = 0.0006).
Symptom Improvement Nine (45%) of 20 patients had upward gaze paresis at the time of SRS. Six of these patients (66.7%) reported improvement in their symptoms after radiosurgery. The median time to improvement after SRS was 2.9 (range, 0.4–11.5) months. One of 9 patients (11.1%) had no change in his upward gaze paresis in the last follow-up. Two of 9 patients (22.2%) experienced worsening of their symptoms. On
52
Kano · Niranjan · Kondziolka · Flickinger · Lunsford
a
b
c
d
e
f
Fig. 6. a, d T1- and T2-weighted contrast-enhanced axial MR images of a 35-year-old male show a pineocytoma at the time of SRS. b, e Axial T1- and T2-weighted contrast-enhanced MR images obtained 7 months after SRS showing progression of enhanced area with peritumoral edema and central necrosis. His diplopia got worse. After administration of corticosteroids, his diplopia improved 13 months after SRS. c, f Axial T1-weighted contrast-enhanced MR images obtained 5 years after SRS showing almost total disappearance of tumor but cystic change.
follow-up imaging, one showed tumor progression and the other tumor regression with development of multiple brain metastases.
Adverse Radiation Effects Four patients (21.1%) with pineocytoma developed adverse radiation effects. One patient who received a tumor margin dose of 16 Gy developed peritumoral edema and ptosis 6 months after SRS which resolved at 11 months after SRS. The two patients who received 13 and 15 Gy developed peritumoral edema and upward gaze palsy 7 and 26 months after SRS. These symptoms resolved 13 and 29 months after SRS, respectively (fig. 6). One patient who received 15 Gy developed asymptomatic MRI-defined
Radiosurgery for Pineal Parenchymal Tumors
53
peritumoral edema 3 months after SRS. These imaging changes resolved 6 months after SRS and were successfully managed with corticosteroids. The marginal dose was not significantly associated with the probability of adverse radiation effect (ARE).
Discussion
Tumors arising from the pineal gland parenchyma have been classified into three subgroups according to their histopathological findings and clinical results: pineocytoma, pineoblastoma and mixed tumors that contain both pineal elements and germ cell elements [10]. Histological diagnosis and the presence of germ cell markers in the serum and CSF (e.g. placental alkaline phosphatase, α-fetoprotein, human chorionic gonadotropin) are very important for decision making prior to SRS. Pineocytoma is a slowly growing tumor that caries a relatively favorable prognosis in most cases. Pineocytomas correspond histologically to WHO grade 2 tumors and are analogous to grade 2 astrocytomas in other areas of the brain. Pineocytoma is typically localized to the pineal area and compresses adjacent structures, including the cerebral aqueduct, brain stem and cerebellum. Their growth may extend into the third ventricle. The majority of patients exhibit neuro-ophthalmologic findings, particularly Parinaud syndromes [11]. In the present series 9 (45%) of 20 patients had upward gaze paresis at presentation for SRS and 2 patients had balance disorders. Nine patients became asymptomatic after a CSF shunt. In this series, we did not find any association between pre-SRS symptoms and overall survival. Pineoblastomas are malignant pineal tumors that resemble primitive neuroectodermal tumors. Pineoblastomas correspond histologically to WHO grade 4 tumors. The interval between initial symptoms and diagnosis may be brief for these patients [12, 13]. Metastasis via cerebrospinal fluid pathways is common and is often associated with fatal outcome [12]. Schild et al. [14], reported 1-, 3-, and 5-year survival rates of pineoblastoma patients treated by various modalities are 88, 78 and 58%, respectively. In our small series, the overall survival after SRS was 95.0, 80.0, and 68.6% at 1, 3 and 5 years, respectively (fig. 1). This suggests that adjuvant SRS (as part of an aggressive oncologic management that includes neuraxis fractionated RT and chemotherapy) may enhance outcome for this difficult tumor.
Treatment Modalities Surgical Resection and Stereotactic Biopsy
Radical tumor removal is not always feasible because of the critical location of PPTs. In the last 20 years, the surgical mortality rates associated with PRT removal
54
Kano · Niranjan · Kondziolka · Flickinger · Lunsford
Table 5. Series reporting SRS of PRTs First author
Year
Patients Tumor type
Mean CR F/U months
PR
NC
PG
Meta- Death stases
Kobayashi Deshmuck Reyns Present study Kobayashi Reyns Present study
2001 2004 2006 2007 2001 2006 2007
3 5 8 12 2 5 7
23.3 14.6 34.0 51.6 23.3 34.0 54.1
33 N/A 57 67 50 20 14
0 100 29 17 0 0 29
0 0 0 0 50 60 29
0 0 0 17 0 20 29
Pineocytoma Pineocytoma Pineocytoma Pineocytoma Pineoblastoma Pineoblastoma Pineoblastoma
68 N/A 14 25 0 20 29
0 0 0 15 50 40 57
NC = No change; PG = progression; F/U = follow-up Values for CR, PR, NC, PG, metastases and death indicate percentages.
have been reduced to less than 2% [15, 16]. Bruce and Ogden [16] reported that the surgical major morbidity rate associated with PRTs was 3–6.8% and permanent minor morbidity rate was 3–28%. Stereotactic biopsy has the benefit of less risk and minimal morbidity but may be associated with a greater likelihood of diagnostic inaccuracy (diagnostic yield varies from 87 to 97%). After stereotactic diagnosis, SRS as the therapeutic option was posed as a less invasive treatment. In our series, the SRS procedural mortality rate was zero.
Fractionated Radiation Therapy
Pineocytomas have been considered to be tumors that are relatively radioresistant to conventional fractionated RT [17]. In our series radiosurgery was used as an alternative to surgical resection or RT. Because pineoblastomas frequently metastasized to CNS and/or systemic organs, these patients require craniospinal RT and chemotherapy in addition.
Stereotactic Radiosurgery
SRS is now considered an option for the treatment of primary or recurrent PPTs. Few studies of SRS for pineal parenchymal tumors have been reported in the literature [9, 18–24]. Deshmukh et al. [20] reported 100% local tumor control after SRS for five pineocytoma at a mean follow-up of 14.6 months (table 5). Reyns et al. [24] treated 13 patients with PPT using radiosurgery. At a mean follow-up of
Radiosurgery for Pineal Parenchymal Tumors
55
34 months, these authors reported 100% local tumor control for pineocytoma (n = 8) and 40% for pineoblastoma (n = 5). In an SRS study, Kobayashi et al. [22] documented local tumor control in all 3 (100%) patients with pineocytoma at a mean follow-up of 21 months. Pineoblastoma patients (n = 2) showed an unfavorable response to SRS with progression rate of 50% [22]. In our series, the local tumor control after SRS for pineocytoma (n = 13) and pineoblastoma or mixed pineal parenchymal tumor (n = 7) at a mean follow-up of 54.1 months was 100 and 50%, respectively. Patients with pineocytomas had 1- and 5-year overall survivals of 100 and 92.3%, respectively. Patients with pineoblastoma or mixed pineocytoma/pineoblastoma had 1-, and 5-year overall survivals of 85.7 and 28.6%, respectively.
Metastases to Central Nervous System and Systemic Organs Metastasis of aggressive PPTs to other CNS sites is the most common cause of death. Aggressive and prone to metastases, pineoblastomas are histologically indistinguishable from other primitive neuroectodermal tumors. Historically, they have been treated with radical surgery followed by adjuvant fractionated radiation and/or chemotherapy [23]. Reyns et al. [24] reported that one of 11 patients with pineoblastoma presented with brain metastases. In the present series, patients with intracranial or extracranial metastasis after SRS had significantly shorter survival times compared to patients who did not have metastasis (p = 0.0006). The median duration for developing metastasis was 6.0 and 2.9 months after initial diagnosis and SRS, respectively (range, 0.5–56.1 months after diagnosis, 0.4–47.1 months after SRS). Because of their aggressive nature and high rate of recurrence, adjunctive treatment modalities are necessary. These options include fractionated external-beam radiotherapy, chemotherapy and repeat SRS when indicated.
Symptom Improvement and Adverse Radiation Effects In our series, 45% of all patients had diplopia and upward gaze paresis at the time of SRS. The symptom improvement rate was 66.7%. The median time to improvement after SRS was 2.9 months (range, 0.4–11.5 months). SRS seemed to improve the quality of life for patients with pineocytomas. Some investigators reported major morbidity after surgical resection was 3–6.8% and permanent minor morbidity was 3–28% [25]. In our series, one patient had asymptomatic T2 signal changes on imaging and 3 patients (15.8%) developed transient ARE after SRS (one was asymptomatic). However, the permanent
56
Kano · Niranjan · Kondziolka · Flickinger · Lunsford
morbidity rate was 0%. The tumor marginal dose ranging from 13 to 16 Gy was not significantly associated with the probability of ARE.
Conclusions
SRS is an effective and safe alternative to the surgical resection of pineocytomas. SRS is an important adjunct in the management of more aggressive PPTs, which require multimodal therapy.
References 1 Kondziolka D, Hadjipanayis CG, Flickinger JC, Lunsford LD: The role of radiosurgery for the treatment of pineal parenchymal tumors. Neurosurgery 2002;51:880–889. 2 Lutterbach J, Fauchon F, Schild SE, Chang SM, Pagenstecher A, Volk B, Ostertag C, Momm F, Jouvet A: Malignant pineal parenchymal tumors in adult patients: patterns of care and prognostic factors. Neurosurgery 2002;51:44–56. 3 Sano K: Pineal region tumors: problems in pathology and treatment. Clin Neurosurg 1983;30: 59–91. 4 Cho BK, Wang KC, Nam DH, Kim DG, Jung HW, Kim HJ, Han DH, Choi KS: Pineal tumors: experience with 48 cases over 10 years. Childs Nerv Syst 1998;14:53–58. 5 Kang JK, Jeun SS, Hong YK, Park CK, Son BC, Lee IW, Kim MC: Experience with pineal region tumors. Childs Nerv Syst 1998;14:63–68. 6 Schild SE, Scheithauer BW, Haddock MG, Wong WW, Lyons MK, Marks LB, Norman MG, Burger PC: Histologically confirmed pineal tumors and other germ cell tumors of the brain. Cancer 1996;78:2564–2571. 7 Kun LE: The brain and spinal cord; in Moss WT, Cox JD (eds): Radiation Oncology: Rationale, Technique, Results. St Louis, CV Mosby, 1989, vol 6, pp 597–616. 8 Bruce JN, Stein BM: Surgical management of pineal region tumors. Acta Neurochir 1995; 134:130–135. 9 Hasegawa T, Kondziolka D, Hadjipanayis CG, Flickinger JC, Lunsford LD: The role of radiosurgery for the treatment of pineal parenchymal tumors. Neurosurgery 2002;51:880–889.
Radiosurgery for Pineal Parenchymal Tumors
10 Fauchon F, Jouvet A, Paquis P, Saint-Pierre G, Mottolese C, Ben Hassel M, Chauveinc L, Sichez JP, Philippon J, Schlienger M, Bouffet E: Parenchymal pineal tumors: a clinicopathological study of 76 cases. Int J Rad Oncol Biol Phys 2000;46: 959–968. 11 Grimoldi N, Tomei G, Stankov B, Lucini V, Masini B, Caputo V, Repetti ML, Lazzarini G, Gaini SM, Lucarini C, Fraschini F, Villani R: Neuroendocrine, immunohistochemical, and ultrastructural study of pineal region tumors. J Pineal Res 1998;25:147–158. 12 Chang SM, Lillis H, Larson DA, Wara WM, Bollen AW, Prados MD: Pineoblastoma in adults. Neurosurgery 1995;37:383–390. 13 Jakacki RI, Zeltzer PM, Boyett JM, Albright AL, Allen JC, Geyer JR, Rorke LB, Stanley P, Stevens KR, Wisoff J: Survival and prognostic factors following radiation and/or chemotherapy for primitive neuroectodermal tumors of the pineal region in infants and children: a report of the Childrens Cancer Group. J Clin Oncol 1995; 13:1377–1383. 14 Schild SE, Scheithauer BW, Schomberg PJ, Hook CC, Kelly PJ, Frick L, Robinow JS, Buskirk SJ: Pineal parencymal tumors. Clinical, pathologic, and therapeutic aspects. Cancer 1993;72:870– 880. 15 Friedman JA, Lynch JJ, Buckner JC, Scheithauer BW, Raffel C: Management of malignant pineal germ cell tumors with residual mature teratoma. Neurosurgery 2001;48:518–523. 16 Bruce JN, Ogden AT: Surgical strategies for treating patients with pineal region tumors. J NeuroOncol 2004;69:221–236. 17 Borit A, Blackwood W, Mair WG: The separation of pineocytoma from pineoblastoma. Cancer 1980;45:1408–1418.
57
18 Dempsey PK, Kondziolka D, Lunsford LD: Stereotactic diagnosis and treatment of pineal region tumours and vascular malformations. Acta Neurochir (Wien) 1992;116:14–22. 19 Dempsey PK, Lunsford LD: Stereotactic radiosurgery for pineal region tumors. Neurosurg Clin N Am 1992;3:245–253. 20 Deshmukh, VR, Smith KA, Rekate HL, Coon S, Spetzler RF: Diagnosis and management of pineocytomas. Neurosugery 2004;55:349–357. 21 Regis J, Bouillot P, Rouby-Volot F, FigarellaBranger D, Dufour H, Peragut JC: Pineal region tumors and the role of stereotactic biopsy: review of the mortality, morbidity, and diagnostic rates in 370 cases. Neurosurgery 1996;39: 907–912.
22 Kobayashi T, Kida Y, Mori Y: stereotactic radiosurgery for pineal and related tumors. J Neurooncol 2001;54:301–309. 23 Borit A, Blackwood W, Mair WG: The separation of pineocytoma from pineoblastoma. Cancer 1980;45:1408–1418. 24 Reyns N, Hayashi M, Chinot O, Manera L, Peragut JC, Blond S, Regis J: The role of Gamma Knife radiosurgery in the treatment of pineal parenchymal tumours. Acta Neurochir (Wien) 2006;148: 5–11. 25 Brude JN, Ogden AT: Surgical strategies for treating patients with pineal region tumors. J Neuro Oncol 2004;69:221–236.
L. Dade Lunsford, MD, FACS Lars Leksell and Distinguished Professor of Neurological Surgery, The University of Pittsburgh B400, UPMC Pittsburgh, PA 15213 (USA) Tel. +1 412 647 6781, Fax +1 412 647 6783, E-Mail
[email protected]
58
Kano · Niranjan · Kondziolka · Flickinger · Lunsford
Tumors of Germ Cell Origin Kobayashi T, Lunsford LD (eds): Pineal Region Tumors. Diagnosis and Treatment Options. Prog Neurol Surg. Basel, Karger, 2009, vol 23, pp 59–75
Pathology of Intracranial Germ Cell Tumors Kazufumi Sato ⭈ Hiroaki Takeuchi ⭈ Toshihiko Kubota Division of Neurosurgery, Department of Sensory and Locomotor Medicine, Faculty of Medical Science, University of Fukui, Fukui, Japan
Abstract Intracranial germ cell tumors (GCTs) usually arise in midline structures, including the pineal or suprasellar regions of children and young adults. The classification of GCTs includes germinoma, teratoma, yolk sac tumor, embryonal carcinoma, and choriocarcinoma. However, intracranial GCTs are often of mixed histologic composition (mixed GCTs), and only germinoma and teratoma are likely to be encountered as pure tumor types. Although GCTs are usually identified using conventional histological techniques, immunohistochemical studies are very useful for delineating these entities, using special markers such as human chorionic gonadotropin, α-fetoprotein, human placental alkaline phosphatase, cytokeratin, as well as c-kit and OCT4. Ultrastructural examination is also useful in confirming the identity of these tumors. Genetic alterations specifically encountered in central nervous system GCTs are largely unknown. Patients with Klinefelter syndrome or Down syndrome appear to be predisposed to the development of gonadal as well as intracranial germinomas. Frequent imbalances of chromosomes have been described in intracranial GCTs, including chromosomes 1, 8, 12, 13, 18 and X. Recently, p14 and c-kit gene alterations have been reported, particularly in some intracranial germinomas; Copyright © 2009 S. Karger AG, Basel however, their importance remains unclear.
Histogenesis and Classification
Embryologically, primordial germ cells appear in the wall of the yolk sac at the end of the 3rd gestational week, and migrate toward the developing gonads (primitive sex glands) by the beginning of the 5th week [1]. If some primitive germ cells abberate into other sites in this migration stage, extragonadal germ cell tumors (GCTs) may develop even in the central nervous system (CNS). Classifications of extragonadal GCTs follow those adopted for neoplasms of germ cell origin involving the testis and ovary, and are essentially modifications of Teilum’s concepts (fig. 1) [2]. In 1965, Teilum [2] regarded embryonal carcinomas as tumors of totipotential cells that may give rise to either embryonal neoplasms,
Table 1. CNS GCTs 1
Germinoma Pure (grade II) With STGCs (grade II–III)
2
Teratoma Mature (grade I) Immature (grade III–IV) With malignant transformation (grade IV)
3
Yolk sac tumor (endodermal sinus tumor) (grade IV)
4
Embryonal carcinoma (grade IV)
5
Choriocarcinoma (grade IV)
6
Mixed GCT
with the potential to differentiate into the derivatives of all three germ layers (teratoma), or extraembryonal neoplasms (yolk sac tumor and choriocarcinoma). Takei and Pearl [3] suggested that all subtypes of GCTs are derived from neoplastic totipotential cells. According to this concept, embryonal carcinomas arise from anaplastic endodermal cells. The histology of intracranial GCTs is similar to that of GCTs elsewhere in the body. The current WHO classification of CNS tumors recognizes five major forms of GCTs: (1) germinoma, (2) teratoma, (3) yolk sac tumor (endodermal sinus tumor), (4) embryonal carcinoma, and (5) choriocarcinoma [4]. Tumors with more than one of the above histological types are termed mixed GCTs (table 1). Only germinoma and teratoma are likely to be encountered as pure tumor types [4–6]. Although WHO grades of GCTs are variable and have not yet been clearly established to date, the tentative grades are shown in table 1.
Incidence, Age and Sex Distribution
The incidence of primary intracranial GCTs is higher in Asian countries than in Western countries. CNS GCTs account for 2–3% of primary intracranial neoplasms and 8–15% of pediatric brain tumors in series from Far East Asian countries including Japan [4, 7]. In Western countries, the incidence of these tumors accounts for only 3–4% of all brain tumors in children [4]. The pineal gland is the most common site of intracranial GCTs, which occur most frequently in the first two decades of life and usually more frequently in males.
60
Sato · Takeuchi · Kubota
Germ cell
Germinoma
Tumors of totipotential cell
Embryonal carcinoma
Extraembryonal tissue Trophoblast
Choriocarcinoma
Yolk sac
Embryonal tissue Ectoderm
Yolk sac tumor (endodermal sinus tumor)
Mesoderm
Endoderm
Teratoma
Fig. 1. Histogenesis and relationship of intracranial GCTs. Modified from Teilum [2].
Fig. 2. Germinoma. Enhanced MRI showing well-defined and homogeneously enhanced tumor in the pineal region.
Pathology of Intracranial Germ Cell Tumors
61
Fig. 3. Yolk sac tumor. Enhanced MRI showing multilobular and heterogeneously enhanced tumor in the pineal region.
Neuroimaging
The neuroradiological profiles of intracranial GCTs are largely nonspecific, but a few useful generalizations can be provided [8]. On MRI, GCTs other than teratomas usually appear as hypo- to isointense solid masses on T1-weighted images and iso- to hyperintense solid masses on T2-weighted images, and show prominent contrast enhancement (fig. 2). Nongerminomatous GCTs tend to show heterogeneous enhancement (fig. 3). A diagnosis of teratoma should be considered for a lesion containing intratumoral cysts admixed with calcified regions and foci having a low signal density on computed tomography, which is characteristic of fats. Intratumoral hemorrhage is particularly characteristic of choriocarcinoma and mixed tumors with choriocarcinomatous elements.
Pathology
Germinoma Germinomas are the most common GCTs in the neuraxis. The preferential sites of these tumors are the pineal and suprasellar (neurohypophyseal) regions.
62
Sato · Takeuchi · Kubota
a
b
Fig. 4. Germinoma. a Tumor showing diffuse proliferation of large and oval cells. HE. b Numerous lymphocytes infiltrating the fibrovascular stroma. HE.
Macroscopy
Germinomas are usually composed of a well-circumscribed solid and soft tissue with a grey-pink cut surface, and occasionally accompanying small cysts. Although conspicuous necrosis and hemorrhage are usually absent, these findings usually suggest the presence of more malignant tumors such as embryonal carcinoma and choriocarcinoma when present. Germinomas have the capacity to infiltrate the adjacent brain and commonly disseminate along cerebrospinal fluid pathways. Extracranial metastasis of tumors usually occurs in the peritoneal cavity via ventriculoperitoneal shunt [9]. Extremely rare extracranial metastases to the lungs or bone have also been described [10]. Spontaneous regression of the tumors has rarely been reported after ventriculoperitoneal shunt or very low doses of irradiation, as demonstrated by computed tomography [11]. Microscopy
Typical germinomas contain two cell populations (two-cell pattern). One cell population contains large neoplastic tumor cells, and the other consists of lymphocytes. The tumors are composed of polygonal cells with central nuclei and prominent nucleoli (fig. 4a). The cytoplasm is pale to slightly clear due to the presence of abundant glycogen, which is demonstrated by the periodic acid Shiff reaction. Mitotic figures are frequent, but necrosis is uncommon. These cells are arranged in sheets or large lobules separated by fibrovascular stroma. In the
Pathology of Intracranial Germ Cell Tumors
63
Fig. 5. Germinoma with STGCs. STGCs are scattered in the tumor. HE.
stroma, variable numbers of inflammatory cells are present, including T lymphocytes of both helper/inducer and cytotoxic/suppressor types, as well as considerable number of B lymphocytes [12] (fig. 4b). Plasma cells are also occasionally observed. In some cases, granulomatous inflammatory reaction is conspicuous and constitutes the predominant element of the tumor [4, 6]. Syncytiotrophoblastic giant cells (STGCs) are sometimes present (germinomas with STGCs; fig. 5). However, cytotrophoblastic cells are not usually observed; when present, mixed GCTs with choriocarcinoma should be considered. Immunohistochemistry
Germinomas are characterized by a surface membrane with diffuse cytoplasmic labeling for placental alkaline phosphatase (PLAP; fig. 6a), normally expressed by primordial germ cells and syncytiotrophoblasts. Cell membrane labeling for the proto-oncogene c-kit (fig. 6b) is also characteristic of germinomas and not shared by other GCTs [13]. Furthermore, nuclear staining for OCT4 is a specific and sensitive immunohistochemical method for detecting germinomas [14]. Cytokeratin immunoreactivity in germinomas is typically focal, often weak, and restricted to minor tumor cell populations [6, 15]. Although the positive immunoreactive rate of cytokeratin in germinomas is up to 30%, other GCTs are usually positive for this antibody [4]. STGCs show strong immunopositive reactions for β-human chorionic gonadotropin (β-HCG; fig. 7), human placental lactogen (HPL), and cytokeratin. The immunohistochemical profiles of CNS GCTs are shown in table 2.
64
Sato · Takeuchi · Kubota
a
b
Fig. 6. Germinoma. a Immunohistochemistry for PLAP showing strong cytoplasmic positivity in most of the large tumor cells. b Immunohistochemistry for c-kit showing positive membranous staining in the neoplastic cells.
Fig. 7. Germinoma with STGCs. Immunohistochemistry for β-HCG showing positive staining in multinucleated giant cells.
Electron Microscopy
Ultrastructurally, two distinct cell populations constitute germinomas, the neoplastic large cells and reactive small cells. The tumor cells are polygonal and contain large eccentric nuclei and prominent nucleoli. A large number of glycogen granules, Golgi apparatus, rough endoplasmic reticulum, and annulate lamellae are present in the cytoplasm [15 -17] (fig. 8a). On the other hand, reactive small cells have round and invaginated or lobulated nuclei. The chromatin is usually
Pathology of Intracranial Germ Cell Tumors
65
Table 2. Immunohistochemical findings of CNS GCTs AFP
β-HCG
HPL
PLAP
c-kit
OCT4
Cytokeratin
–
–
–
⫹⫹
⫹
⫹
–1
–
⫹
⫹
⫹⫹
⫹
⫹
⫹
Teratoma
⫹2~–
–
–
–
–
–
⫹
Yolk sac tumor
⫹⫹
–
–
–~⫹
–
–
⫹
Embryonal carcinoma
⫹3~–
–~⫹
–~⫹
⫹~–
–
⫹
⫹
Choriocarcinoma
–
⫹⫹
⫹⫹
⫹~–
–
–
⫹
Germinoma Pure With STGC
1
Positive for 20–30% of cases. AFP is restricted to enteric type. 3 Positive when yolk sac tumor component is present. 2
located at the margin of the nucleus, and nucleoli are rarely seen. The cytoplasm consists of a small layer around the nucleus and contains a few organelles, including free ribosomes, mitochondria, and Golgi apparatus (fig. 8b). These morphologic features are compatible with those of nonneoplastic lymphocytes. Thus, ultrastructural findings of germinomas are quite similar to those of the primordial germ cells observed during migration within the normal human embryo [18].
Teratoma Teratomas differentiate along ectodermal, endodermal and mesodermal lines. Immature teratomas are the most common type of teratomas arising from the CNS. Macroscopy
Teratomas are often large, well-demarcated tumors tending to adhere firmly to neighboring structures. The tumors are usually multicystic, and the cut surface demonstrates a mixed composition, including aggregates of keratin debris, fat, cartilage and bone. CNS teratomas rarely contain teeth or hairs. Advanced organogenesis and somatic organization may result in intracranial fetus-in-fetu [19]. In contrast, the immature elements appear as ill-defined areas of pinkish grey tissue marked by focal necrosis and hemorrhage.
66
Sato · Takeuchi · Kubota
Microscopy
The histological appearance of teratomas varies with the presence of differentiating cell types and diverse tissue. Mature teratomas consist of fully differentiated tissues representative of the three germ layers. The elements are often organized in a stereotypic manner, including skin, cartilage and bone, adipose tissue, smooth muscle, and glioneuronal tissue with choroid plexus (fig. 9a). Mitosis and necrosis are absent. Immature teratomas contain incompletely differentiated components resembling fetal tissues. The most immature elements are primitive embryonal mesenchymal tissue or neuroectodermal tissue with histological patterns of canalicular structures resembling a developing neural tube or ependymal rosettes (fig. 9b). Although the prognosis of intracranial immature teratomas is usually poor, the ‘maturation of intracranial immature teratoma’ has rarely been described [20]. ‘The growing teratoma syndrome’ represents tumor recurrence and the presence of mature teratoma in resected specimens despite decreasing levels of tumor serum markers [21]. This type of tumor maturation reflects the selective ablation of their more malignant components by chemotherapy and/ or radiotherapy [4]. ‘Teratomas with malignant transformation’ are rare variants of teratoma having additional somatic type malignant elements. Sarcomas with undifferentiated or rhabdomyosarcomatous features [5, 7, 22], squamous carcinomas [7], and adenocarcinomas [5] have been described as malignant components.
Immunohistochemistry
Immunohistochemically, the constituent elements of teratomas express antigens similar to the native somatic counterparts. Both pure mature and immature teratomas show positivity for α-fetoprotein (AFP) in enteric type glands. This positive reaction could explain the occasional association of teratomas with elevated CSF levels of AFP [6, 7], which should not be interpreted as an indication of malignant germ cell elements or yolk sac tumor. In some cases, teratomas show a positive reaction for carcinoembryonic antigen in glandular and squamous elements [5].
Yolk Sac Tumor (Endodermal Sinus Tumor) Yolk sac tumors are highly malignant GCTs mimicking normal yolk sac structures. About half of primary intracranial yolk sac tumors are mixed GCTs containing various proportions of other germ cell elements. An elevated level of AFP alone in blood and/or CSF is strongly suggestive of yolk sac tumor.
Pathology of Intracranial Germ Cell Tumors
67
a
b
Fig. 8. Germinoma. a Electron micrograph of neoplastic large cells showing abundant glycogen granules, mitochondria, and rough endoplasmic reticulum in the cytoplasm. b Electron micrograph of small cells with dense nucleus and scant cytoplasm resembling lymphocytes.
a
b
Fig. 9. Mature teratoma. a Photomicrograph showing differentiated gland structures and cartilage tissue. HE. b Photomicrograph showing canals mimicking the primitive neural tube. HE.
Macroscopy
The tumors are firm in consistency and contain gelatinous or myxomatous areas. Hemorrhagic and necrotic areas are frequently seen. Microscopy
Yolk sac tumors are often composed of clear, cuboidal to columnar epithelial cells organized in various patterns from sheets, cords and tubules to papillary structures
68
Sato · Takeuchi · Kubota
embedded in a loose myxoid matrix (fig. 10a). The tumor cells are small and contain rounded nuclei with inconspicuous nucleoli surrounded by scant cytoplasm. The papillary projections are sustained by fibrovascular stroma with thin-walled blood vessels. The presence of a particular glomerular-like structure consisting of perivascular epithelial-lined space, termed ‘Schiller-Duval bodies’ (fig. 10b), is a hallmark of yolk sac tumors. Another characteristic feature is the presence of periodic acid Shiff-positive intra- and extracytoplasmic ‘hyaline globules’ (fig. 10c). Mitotic figures are also commonly observed, but necrosis is less frequent in yolk sac tumors than in embryonal carcinoma or choriocarcinoma.
Immunohistochemistry
Cytoplasmic immunoreactivity for AFP of the epithelial component of yolk sac tumors is characteristic and may be of considerable value in distinguishing these tumors from germinoma and embryonal carcinoma (fig. 10d) [5, 16, 22]. The hyaline globules also exhibit strong immunoreactivity for AFP, facilitating differential diagnosis from other GCTs. The tumor cells usually show positivity for cytokeratin and epithelial membrane antigen [23]. Focal immunopositivity for PLAP and carcinoembryonic antigen is detected in some tumors [6]. Yolk sac tumors are negative for c-kit, OCT4, HPL, and β-HCG.
Electron Microscopy
Ultrastructurally, yolk sac tumors are recognized for pseudoglandular formations of the tumor cells. The cytoplasm contains conspicuous endoplasmic reticulum and well-developed Golgi apparatus, glycogen, and moderate numbers of mitochondria, microtubules, and lysosomes [3, 15, 16]. Nuclei are oval and elongated with prominent nucleoli. The apical surfaces of the tumor cells of the pseudoglandular structures have numerous microvilli, and amorphous intracytoplasmic vesicles are occasionally seen. The latter vesicles presumably contain AFP [3]. These apical cells are connected to each other by junctional complexes such as tight junctions and desmosomes.
Embryonal Carcinoma Intracranial embryonal carcinomas are rare variants of GCTs [24]. Tumor cells may exceptionally replicate the structure of an early embryo, forming ‘embryoid bodies’, which include germ discs and miniature amniotic cavities. Embryonal
Pathology of Intracranial Germ Cell Tumors
69
a
b
c
d Fig. 10. Yolk sac tumor. a Tumor showing a reticular growth pattern. Strands and cords of tumor cells are embedded in a myxoid matrix. HE. b Tumor showing a papillary structure covering a centrally located blood vessel (Schiller-Duval bodies). HE. c Various sizes of intra- and extracytoplasmic ‘hyaline globules’. HE. d Immunohistochemistry for AFP showing positive staining in most of the tumor cells.
carcinomas often differentiate into yolk sac tumor and choriocarcinoma, accompanied by elevation of serum AFP and β-HCG levels. Microscopy
Embryonal carcinomas are characterized by immature tumor cells arranged in sheets or cords, or appear as gland-like structures with abortive papillae [4, 5, 7, 24]. The tumor cells are large, polygonal or round, and have moderate amounts of eosinophilic or clear cytoplasm (fig. 11a). Nuclei are large and round with markedly enlarged nucleoli. Mitotic figures are usually numerous, and these tumors have more prominent multiple necrotic foci than yolk sac tumors.
70
Sato · Takeuchi · Kubota
a
b
Fig. 11. Embryonal carcinoma. a Tumors are composed of pleomorphic immature cells arranged in sheets with abortive papillae. HE. b Immunohistochemistry of cytokeratin showing positive membranous staining in the cytoplasm.
Immunohistochemistry
The tumor cells uniformly show dense and diffuse cytoplasmic labeling for cytokeratin (fig. 11b) and epithelial membrane antigen [23], attesting to their differentiation along epithelial lines and distinguishing these neoplasms from germinomas (with which they share PLAP and OCT 4 immunoreactivity) [6, 15]. Embryonal carcinomas show negative staining for c-kit.
Electron Microscopy
The pure type of embryonal carcinomas is characterized by microvilli-bearing tumor cells bound together by junctional complexes such as tight junctions and desmosomes. These findings are similar to the morphologic appearance of yolk sac tumors [3, 16]. Takei and Pearl [3] stressed that ultrastructurally, embryonal carcinomas are composed predominantly of anaplastic endodermal cells, which are only multiple elements in yolk sac tumors. As opposed to Teilum’s concepts [2], these ultrastructural findings suggest the possibility that yolk sac tumors can arise directly from neoplastic germ cells [3].
Choriocarcinoma Choriocarcinomas are extremely rare forms of intracranial GCTs. Similar to yolk sac tumors, these tumors are presumably derived from totipotential cell lineages
Pathology of Intracranial Germ Cell Tumors
71
with the capacity for trophoblastic differentiation. Pure CNS choriocarcinomas are rare. These tumors usually include one element of other GCTs, particularly embryonal carcinoma and yolk sac tumor as well as teratoma.
Macroscopy
Choriocarcinomas are usually well demarcated and prone to extensive hemorrhagic necrosis. These tumors are highly malignant and usually invade adjacent structures. The accumulation of a myxoid material lends a gelatinous appearance and consistency to some yolk sac tumors. In addition, the development of extraneural metastasis, especially in the lungs, has been reported [25].
Microscopy
The definitive diagnosis of choriocarcinomas requires the identification of two characteristic cell types: cytotrophoblastic cells and STGCs. Cytotrophoblastic cells are round to polygonal with clear cytoplasm, round nuclei and prominent nucleoli. Conversely, STGCs typically contain pleomorphic densely hyperchromatic multiple nuclei, and frequently show mitotic figures. The cytoplasm is large and vacuolization is common (fig. 12a). Sinusoidal stromal vascular channels forming blood lakes accompanied by extensive hemorrhagic necrosis are present.
Immunohistochemistry
STGCs show immunopositivity for β-HCG (fig. 12b), HPL, and cytokeratin [5, 6, 15, 22]. HPL immunoreactivity allows a clear differentiation from most other GCTs [5]. The majority of the tumor cells show immunopositivity for cytokeratin and PLAP. Choriocarcinomas show negative staining for c-kit, OCT4 and AFP.
Genetics
Little is known about the genetic alterations of CNS GCTs. Intracranial germinomas have been reportedly associated with genetic diseases, including Klinefelter syndrome [26], Down syndrome [27], and neurofibromatosis type 1 [28]. Thus, the relationship between some chromosome or gene abnormalities and intracranial germinomas is suggested. Intracranial teratomas [29] and germinomas [30] have been reported in otherwise normal siblings; however, genetic abnormalities of these tumors have remained unclarified to date.
72
Sato · Takeuchi · Kubota
a
b
Fig. 12. Choriocarcinoma. a Photomicrograph showing admixture of cytotrophoblastic cells and STGCs. HE. b Immunohistochemistry for β-HCG showing strong positive staining in STGCs.
Recent studies using comparative genomic hybridization in CNS GCTs found several chromosomal imbalances. The most common chromosomal imbalance is gains of 1p, 8p, and 12q, and losses of 13q and 18q [31, 32]. Another report showed a nearly universal X chromosome gain in CNS GCTs [33]. The most frequent genotype abnormality is XXY, similar to that in Klinefelter syndrome. Interestingly, all intracranial tumors reported in Klinefelter patients are GCTs [33]. Frequent alterations of the p14 gene were detected especially in intracranial pure germinomas, suggesting that this gene plays an important role in the development of these tumors [34]. More recently, Sakuma et al. [35] reported the presence of mutations of the c-kit gene in 23–25% of intracranial germinomas. They speculated that such mutations promote, but are not essential to, the development of some germinomas.
References 1 Sadler TW: Langman’s Medical Embryology, ed 9. Baltimore, Wiliams & Wilkins, 2004. 2 Teilum G: Classification of endodermal sinus tumour (mesoblastoma vitellinum) and so-called ‘embryonal carcinoma’ of the ovary. Acta Pathol Microbiol Scand 1965;64:407–429. 3 Takei Y, Pearl GS: Ultrastructural study of intracranial yolk sac tumor: with special reference to the oncologic phylogeny of germ cell tumors. Cancer 1981;48:2038–2046.
Pathology of Intracranial Germ Cell Tumors
4 Rosenblum MK, Nakazato Y, Matsutani M: CNS germ cell tumours; in Louis DN, Ohgaki H, Wiestler OD, Cavenee WK (eds): WHO Classification of Tumours of the Central Nervous System, ed 4. Lyon, IARC, WHO press, 2007, pp 198–204. 5 Bjornsson J, Scheithauer BW, Okazaki H, Leech RW: Intracranial germ cell tumors: pathobiological and immunohistochemical aspects of 70 cases. J Neuropathol Exp Neurol 1985;44:32–46. 6 Ho DM, Liu H: Primary intracranial germ cell tumor. Pathologic study of 51 patients. Cancer 1992;70:1577–1584.
73
7 Matsutani M, Sano K, Takakura K, Fujimaki T, Nakamura O, Funata N, Seto T: Primary intracranial germ cell tumors: a clinical analysis of 153 histologically verified cases. J Neurosurg 1997;86:446–455. 8 Liang L, Korogi Y, Sugahara T, Ikushima I, Shigematsu Y, Okuda T, Takahashi M, Kochi M, Ushio Y: MRI of intracranial germ-cell tumours. Neuroradiology 2002;44:382–388. 9 Altundağ OO, Celik I, Kars A: Pineal germ cell tumor metastasis via ventriculoperitoneal shunt. Am J Clin Oncol 2002;25:104–105. 10 Kim K, Koo BC, Delaflor RR, Shaikh BS: Pineal germinoma with widespread extracranial metastases. Diagn Cytopathol 1985;1:118–122. 11 Murai Y, Kobayashi S, Mizunari T, Ohaki Y, Adachi K, Teramoto A: Spontaneous regression of a germinoma in the pineal body after placement of a ventriculoperitoneal shunt. J Neurosurg 2000;93:884–886. 12 Saito T, Tanaka R, Kouno M, Washiyama K, Abe S, Kumanishi T: Tumor-infiltration lymphocytes and histocompatibility antigens in primary intracranial germinomas. J Neurosurg 1989;70:81–85. 13 Nakamura H, Takeshima H, Makino K, Kuratsu J: C-kit expression in germinoma: an immunohistochemistry-based study. J Neurooncol 2005;75: 163–167. 14 Hattab EM, Tu PH, Wilson JD, Cheng L: OCT4 immunohistochemistry is superior to placental alkaline phosphatase (PLAP) in the diagnosis of central nervous system germinoma. Am J Surg Pathol 2005;29:368–371. 15 Felix I, Becker LE: Intracranial germ cell tumors in children: an immunohistochemical and electron microscopic study. Pediatr Neurosurg 1990;16:156–162. 16 Masuzawa T, Shimabukuro H, Nakahara N, Iwasa H, Sato F: Germ cell tumors (germinoma and yolk sac tumor) in unusual sites in the brain. Clin Neuropathol 1986;5:190–202. 17 Min KW, Scheithauer BW: Pineal germinomas and testicular seminoma: a comparative ultrastructural study with special references to early carcinomatous transformation. Ultrastruct Pathol 1990;14:483–496. 18 Funkuda T: Ultrastructure of primordial germ cells in human embryo. Virchows Arch B Cell Pathol 1976;20:85–89. 19 Afshar F, King TT, Berry CL: Intraventricular fetusin-fetu. Case report. J Neurosurg 1982;56: 845–849.
74
20 Shaffrey ME, Lanzino G, Lopes MB, Hessler RB, Kassell NF, VandenBerg SR: Maturation of intracranial immature teratomas. Report of two cases. J Neurosurg 1996;85:672–676. 21 Bi WL, Bannykh SI, Baehring J: The growing teratoma syndrome after subtotal resection of an intracranial nongerminomatous germ cell tumor in an adult: case report. Neurosurgery 2005;56:188. 22 Rueda-Pedraza ME, Heifetz SA, Sesterhenn IA, Clark GB: Primary intracranial germ cell tumors in the first two decades of life. A clinical, light-microscopic, and immunohistochemical analysis of 54 cases. Perspect Pediatr Pathol 1987;10:160–207. 23 Nakagawa Y, Perentes E, Ross GW, Ross AN, Rubinstein LJ: Immunohistochemical differences between intracranial germinomas and their gonadal equivalents. An immunoperoxidase study of germ cell tumors with epithelial membrane antigen, cytokeratin, and vimentin. J Pathol 1988;156:67–72. 24 Borit A: Embryonal carcinoma of the pineal region. J Pathol 1969;97:165–168. 25 Page R, Doshi B, Sharr MM: Primary intracranial choriocarcinoma. J Neurol Neurosurg Psychiatry 1986;49:93–95. 26 Arens R, Marcus D, Engelberg S, Findler G, Goodman RM, Passwell JH: Cerebral germinomas and Klinefelter syndrome. A review. Cancer 1988;61:1228–1231. 27 Hashimoto T, Sasagawa I, Ishigooka M, Kubota Y, Nakada T, Fujita T, Nakai O: Down’s syndrome associated with intracranial germinoma and testicular embryonal carcinoma. Urol Int 1995;55:120–122. 28 Wong TT, Ho DM, Chang TK, Yang DD, Lee LS: Familial neurofibromatosis 1 with germinoma involving the basal ganglion and thalamus. Childs Nerv Syst 1995;11:456–458. 29 Wakai S, Segawa H, Kitahara S, Asano T, Sano K, Ogihara R, Tomita S: Teratoma in the pineal region in two brothers. Case reports. J Neurosurg 1980;53:239–243. 30 Aoyama I, Kondo A, Ogawa H, Ikai Y: Germinoma in siblings: case reports. Surg Neurol 1994;41:313–317. 31 Rickert CH, Simon R, Bergmann M, DockhornDworniczak B, Paulus W: Comparative genomic hybridization in pineal germ cell tumors. J Neuropathol Exp Neurol 2000;59:815–821. 32 Schneider DT, Zahn S, Sievers S, Alemazkour K, Reifenberger G, Wiestler OD, Calaminus G, Göbel U, Perlman EJ: Molecular genetic analysis of central nervous system germ cell tumors with comparative genomic hybridization. Mod Pathol 2006;19:864–873.
Sato · Takeuchi · Kubota
33 Okada Y, Nishikawa R, Matsutani M, Louis DN: Hypomethylated X chromosome gain and rare isochromosome 12p in diverse intracranial germ cell tumors.JNeuropatholExpNeurol2002;61:531–538. 34 Iwato M, Tachibana O, Tohma Y, Arakawa Y, Nitta H, Hasegawa M, Yamashita J, Hayashi Y: Alterations of the INK4a/ARF locus in human intracranial germ cell tumors. Cancer Res 2000;60:2113–2115.
35 Sakuma Y, Matsukuma S, Yoshihara M, Sakurai S, Nishii M, Kishida T, Kubota Y, Nagashima Y, Inayama Y, Sasaki T, Nakamura Y, Miura T, Kameda Y, Tsuchiya E, Miyagi Y: c-kit gene mutations in intracranial germinomas. Cancer Sci 2004;95:716–720.
Kazufumi Sato, MD Division of Neurosurgery, Department of Sensory and Locomotor Medicine Faculty of Medical Science, University of Fukui 23 Shimoaizuki, Matsuoka, Eiheiji-cho Yoshida-gun, Fukui 910-1193 (Japan) Tel. +81 776 61 8474, Fax +81 776 61 8121, E-Mail
[email protected]
Pathology of Intracranial Germ Cell Tumors
75
Tumors of Germ Cell Origin Kobayashi T, Lunsford LD (eds): Pineal Region Tumors. Diagnosis and Treatment Options. Prog Neurol Surg. Basel, Karger, 2009, vol 23, pp 76–85
Pineal Germ Cell Tumors Masao Matsutani Department of Neuro-Oncology, Saitama International Medical Center, Saitama Medical University, Saitama, Japan
Abstract Intracranial germ cell tumors (GCTs), especially pineal tumors have attracted the special attention of neuropathologists and neurosurgeons because of their unique growth sites, characteristic subtypes with different histology, and high incidence in Japan and other Asian countries. This chapter describes the general clinical features of pineal GCTs and current treatment of intracranial GCTs. Despite excellent long-term results for patients with germinoma treated with radiation therapy, the potential for late effects makes the treatment controversial. Most patients with nongerminomatous tumors treated by conventional treatment with surgery and radiation therapy failed to survive longer than 3 years. After combination chemotherapy with cisplatin was confirmed to be effective in gonadal GCTs, GCTs of the brain became candidates for chemotherapy. For germinoma, a trial with chemotherapy alone failed with a high rate of recurrence, but Japanese and European trials with chemotherapy and reduced dose and volume of radiation therapy demonstrate good event-free survival rates. Ongoing phase II studies with combined chemotherapy and radiation therapy for nongerminomatous tumors will result in a 5-year survival rate of >50%, which is better than that by radiation therapy alone. Copyright © 2009 S. Karger AG, Basel
According to the Brain Tumor Registry in Japan [1], 53,654 cases of primary brain tumors were registered in the period between 1984 and 1996. There were 966 histologically verified pineal tumors; there, germ cell tumor (GCT) had the highest frequency (65.7%, 635 cases) followed by pineal parenchymal tumors (12.0%, 76 cases). Intracranial GCTs have attracted the special attention of neuropathologists and neurosurgeons because of their unique growth sites, characteristic subtypes with different histology, and high incidence in Japan. They grow at various brain sites as a solitary mass or as multiple foci; most of them occur in the pineal and neurohypophyseal (suprasellar) region, and they are rarely found in the basal ganglia or
at other sites in the brain. Epidemiologically, the incidence of these tumors differs. Japan has the highest incidence; GCTs comprise 2.8% of all primary brain tumors and 15.4% of them occur in children younger than 15 years old [1]. In the USA and Europe, their incidence is lower than in Japan; the reported incidence in the USA is 0.5% [2]. This chapter describes general aspects of intracranial GCTs with special reference to those in the pineal region.
Pathology
According to the World Health Organization classification of tumors of the nervous system [3], GCTs are categorized into 5 basic types (germinomas, teratomas, choriocarcinomas, yolk sac or endodermal sinus tumors, and embryonal carcinomas) and their mixture (mixed GCT). Germinomas are composed of large polygonal cells with a pale eosinophilic or clear cytoplasms, and small lymphocytes. Their cytoplasm stains positive for placental alkaline phosphatase (PLAP) and there is infiltration by small lymphocytes along the stroma of vascular connective tissue. Germinomas containing syncytiotrophoblastic giant cells (germinomas with STGC) stain positive for human chorionic gonadotropin (HCG) or β-HCG. Teratomas are divided into three subtypes according to the degree of tumor cell differentiation: mature and immature teratomas, and teratoma with malignant transformation. Mature teratomas contain 3 well-differentiated germ cell layers: ectoderm, endoderm and mesoderm layer. Immature teratomas are composed of incompletely differentiated tissues resembling fatal tissue. Teratomas that, like carcinomas and sarcomas, contain elements exhibiting unequivocal malignant transformation are referred to as teratomas with malignant transformation. Yolk sac or endodermal sinus tumors are composed of primitive epithelial cells that proliferate in a loose-kit reticular network or compact sheets. Diagnostic features are Schiller-Duval bodies and PAS-positive, cytoplasmic and extracellular eosinophilic droplets immunopositive for α-fetoprotein (AFP). Choriocarcinomas are composed of two characteristic cell types, syncytiotrophoblasts and cytotrophoblasts, arranged in a two-layer pattern. These cells are strongly immunopositive for HCG or β-HCG. Embryonal carcinomas contain primitive epithelial cells growing in solid sheets or poorly formed glands. They are sometimes positive for AFP or HCG (β-HCG). In mixed GCTs, the most frequent component is germinomas, followed by mature or immature teratomatous components.
Pineal Germ Cell Tumors
77
General Aspects of Pineal Germ Cell Tumors
GCTs are most commonly located in the pineal region; their next most frequent site of occurrence is the hypothalamic-neurohypophyseal region, followed by the basal ganglia. Pineal and basal ganglia GCTs predominantly affect males, while suprasellar GCTs have no sex preponderance. According to a recent statistical analysis of the Committee of Brain Tumor Registry in Japan [1], 1,463 patients with intracranial GCTs were registered between 1984 and 1996; of these, 1,068 were males (73.0%) and 395 females (27.0%). Only 43 patients (2.9%) were younger than 5 and 90 (6.2%) were older than 35 years; 1,026 patients (70.1%) were between 10 and 24 years old. Patients with GCTs manifest symptoms and signs attributable to the affected site [4]. Tumors in the pineal region tend to compress and obstruct the cerebral aqueduct, resulting in progressive hydrocephalus with intracranial hypertension. They also compress the tectal plate and produce the characteristic upward and downward gaze palsy (Parinaud sign) and Argyll-Robertson pupils. Patients with suprasellar or neurohypophyseal tumors present with bitemporal hemianopsia and decreased visual acuity due to compression of the optic chiasm. Tumor invasion into the neurohypophysis results in pan-hypopituitarism and diabetes insipidus. A survey of pituitary function revealed decreased levels of anterior pituitary hormones (especially GH, FSH, and LH) and vasopressin and elevated prolactin titers [5]. Patients with tumors in both the pineal and neurohypophyseal region exhibited the signs characteristic of tumors at these sites [4]. Tumors in the basal ganglia or thalamus invade the pyramidal tract and result in contralateral hemiparesis. With the exception of HCG-secreting tumors, the clinical signs and symptoms of the different histological subtypes are not tumor specific. Some HCG-secreting tumors manifest intratumoral hemorrhage that results in acute intracranial hypertension.
Neuroimaging
On magnetic resonance imaging (MRI) T1-weighted scans, germinomas are visualized as iso- or slightly low-signal round, square-round, or oval masses; they appear as iso- or high-signal areas on T2-weighted images. They are mostly homogeneously and partly heterogeneously enhanced by gadolinium. The MRI pattern of HCG- or β-HCG-secreting germinomas is almost identical to that of pure germinomas except for the occasional presence of intratumoral hemorrhage. Existence of teratomatous component should be considered for a lesion that can be shown to contain intratumoral cysts admixed with calcified regions and foci having the low signal attenuation characteristics of fat. The MRI pattern
78
Matsutani
of highly malignant GCTs including yolk sac tumors, embryonal carcinomas and mixed tumors with those elements manifest as an irregularly shaped, iso-, low-, or high-signal, homogeneously enhanced mass, and cystic components and perifocal edema are also observed. Intratumoral hemorrhage is particularly characteristic of choriocarcinoma and mixed neoplasms with choriocarcinomatous or HCGsecreting elements. These findings are largely nonspecific, and it is difficult to differentiate pineal GCTs from pineal parenchymal tumors as well as to differentiate between pathological subtypes of tumors, including GCTs [6].
Endoscopic Findings of Pineal Germ Cell Tumors
Pineal GCTs tend to protrude anteroinferiorly to the tectal plate and posterior portion of the third ventricle, leading to hydrocephalus despite the small mass. Endoscopic third ventriculostomy is increasingly used as a safe and effective procedure to relieve hydrocephalus [7, 8]. During this procedure, neurosurgeons frequently observe small tumor fragments deposited on the ventricular wall of the anterior horn, infundibular recess or optic recess which were not found on the preoperative MR images; these plaques are histopathologically identical to the primary germinomas [9]. This finding suggests that in the treatment of germinomas radiation volume should involve the whole ventricular system including obex of the fourth ventricle and pituitary fossa.
Serum Titer of α-Fetoprotein, Human Chorionic Gonadotropin (β-Gonadotropin) and Placental Alkaline Phosphatase
Tumor marker studies of AFP, HCG and β-HCG and PLAP in the serum and cerebrospinal fluid of patients have recently been attributed the clinical importance for diagnosis of intracranial GCTs. HCG (or HCG-β) is usually secreted by choriocarcinoma cells, STGCs or immature tumor cells; AFP is secreted by yolk sac tumor cells or immature tumor cells, and PLAP by germinoma cells. In some cases, these markers reflect the number of cells that secrete these proteins; they are useful for differentiating tumors with predominant choriocarcinoma or yolk sac tumor from other kinds of marker-secreting tumors if HCG or AFP titer is more than 2,000 mIU/ml or 2,000 ng/ml [4]. However, these two titers are sometimes secreted by immature teratomas or embryonal carcinomas, and they cannot give us precise histological subtypes in most patients. Elevation of serum and cerebrospinal fluid PLAP is characteristic of tumors composed wholly,
Pineal Germ Cell Tumors
79
Table 1. Ten-year survival rate in pure germinoma patients treated primarily with craniospinal radiation therapy alone First author
Cases
10-year survival rate, %
Haddock [11] Sawamura [12] Aoyama [13] Shibamoto [14] Ogawa [15]
32 60 110 38 126
91 90 87 91 90
or in part of germinoma; the elevated PLAP can only suggest the tumor is a tumor containing germinoma. Recently, the ultrasensitive EIA was developed which can detect 1/1,000 levels, a level of pictogram, of β-HCG detected by the standard method [10]. According to this analysis, all patients with germinomas presented with abnormally high titer of β-HCG as compared to those with nongerminomas.
Treatment
The traditional approach to germinomas consisted of biopsy followed by prophylactic craniospinal irradiation, and a boost to the local area resulted in an excellent outcome. According to 5 reports with more than 30 patients [11–16], 87–91% of 366 patients with pure germinomas treated primarily with craniospinal radiation therapy alone survived for at least 10 years (table 1). However, with modern imaging procedures, the proportion of patients presenting with metastatic disease at the time of diagnosis is low, and risk of secondary seeding outside the irradiated volume in germinoma does not exceed 12% in histologically verified series [4, 13, 16]. Furthermore, in children, neuroaxis irradiation resulted in mental retardation, pituitary gland dysfunction, and short stature in later life [17, 18]. The benefit of craniospinal radiotherapy thus seems questionable, and the optimal radiation treatment has been a matter of debate. Single institutional experiences have shown that, with complete diagnostic craniospinal evaluation, spinal irradiation is not necessary and the cure rates for localized germinomas are excellent with extended focal irradiation including the third and lateral ventricles as well as the sella and the pineal region (whole ventricular field) [4, 16, 19]. Without craniospinal irradiation, the Tokyo University series (n = 50) reported the 10-year survival rate of 92.7%; 39 patients (78%) received postoperative radiation therapy (40–62 Gy, median 50 Gy) to a generous local field encompassing the tumor site and third and the lateral ventricles as well
80
Matsutani
as the sellar and pineal regions, or to the tumor area (limited local field) alone [4]. Rogers et al. [20] reviewed publications since 1988 to compare patterns of disease relapse and cure rates after craniospinal radiotherapy, reduced-volume irradiation alone (i.e. whole brain or whole ventricular irradiation followed by a boost), and focal or localized irradiation alone. They found the recurrence rate after whole brain or whole ventricular radiotherapy plus boost was 7.6 and 3.8% respectively, with no predilection for isolated spinal relapses (2.9 vs. 1.2%); they concluded that reduced-volume radiotherapy plus boost should replace craniospinal radiotherapy when a radiotherapy-alone approach is used. After combination chemotherapy with cisplatin was confirmed to be effective in gonadal GCTs, GCTs in the brain became candidates for chemotherapy; platinum-based chemotherapy yielded a complete response (CR) rate of 85–100% in patients with germinomas [21, 22]. This has led to the speculation that chemotherapy may eventually replace radiotherapy as the primary treatment modality for central nervous system (CNS) germinoma. Another strategy for germinoma was to reduce the radiotherapy volume and dose and combine systemic chemotherapy. The international CNS GCT study group tried to treat germinoma patients with chemotherapy alone if they showed complete remission; they failed to obtain a good event-free survival (EFS) in the first and the second trials [23, 24]. The Japanese pediatric brain tumor study group conducted a multi-institutional phase II study including surgical debulking of the tumor and verification of its histological composition, followed by preirradiation chemotherapy and subsequent radiation therapy [25, 26]. Patients with pure germinoma received 3 courses of the carboplatin-etoposide (CARE) combination with carboplatin (450 mg/m2) on day 1 and etoposide (150 mg/m2) on days 1–3, followed by local irradiation (24 Gy) to the generous local field encompassing the tumor site, the 3rd and lateral ventricles, and the sellar and pineal regions. The 5-year overall survival (OS) and disease-free survival rate in 123 patients was 98 and 88%, respectively [27]. They found the recurrent rate was 6% in patients treated by extended local field irradiation, whereas that was 28% in patients treated by limited local field irradiation with a margin of less than 2 cm; they insist again that the radiation volume should involve the whole ventricular system in the treatment of germinoma. The French Society for Pediatric Oncology (SFOP) conducted a study that combined chemotherapy (alternating courses of etoposide/carboplatin and etoposide/ifosphamide for a recommended total of 4 courses) with 40-Gy local irradiation to treat 60 patients with localized germinomas [28]. Their 8-year OS and EFS rate in 60 patients was 98 and 83%, respectively. The International Society for Pediatric Hematology and Oncology treated 41 patients with germinoma using the SFOP CNS GCT-96 protocol. The 5-year EFS rate was 82%, with a median follow-up of 27 months. In an extended protocol, they used identical chemotherapy regimens; under option A, the patients
Pineal Germ Cell Tumors
81
received craniospinal irradiation with 24 Gy followed by a tumor boost of 16 Gy. Under option B, they received focal irradiation with 40 Gy. The EFS rates were 93 (median follow-up: 35 months) and 90% (median follow-up: 26 months) in patients treated by option A and B, respectively [29]. The Children’s Oncology Group treated children with germinomas and normal markers with cisplatin + etoposide, alternating with vincristine + cyclophosphamide, for four cycles followed by focal irradiation (CR: 30.6 Gy;
Nongerminomatous Germ Cell Tumors
Nongerminomatous tumors proved refractory to conventional treatments with surgery and irradiation. Jennings et al. [31], who analyzed the survival of 216 patients with intracranial GCTs who had received conventional treatment, noted that most patients with nongerminomatous tumors did not survive beyond 3 years. According to the results of the University of Tokyo series, the 3-year survival rate of 11 patients with highly malignant GCTs (choriocarcinoma, yolk sac tumor, and embryonal carcinoma) was 27.3%; it was 9.3% for patients with mixed tumors mainly consisting of pure malignant elements [4]. Of the 61 nongerminomatous GCTs treated, 34 (55.7%) recurred or metastasized. The rate of tumor recurrence increased with histological malignancy. The site of recurrence was the primary site alone (20 cases), the primary site and a remote site (5 cases), the brain outside the primary site (3 cases), and the spinal cord alone (4 cases); 2 patients had systemic metastasis alone. Overall, 25 patients (73.5%) had tumor recurrence at the primary site with or without remote recurrence. Currently, several prospective phase II studies are being investigated to assess the effect of combination chemotherapy and radiation therapy for GCTs. The Japanese pediatric brain tumor study group divided nongerminomas into two groups, intermediate prognosis (moderate malignancy), and poor prognosis (high malignancy) groups (table 2) [25]. Patients with HCG- or β-HCG-secreting germinomas were placed in the intermediate prognosis group because these tumors have a higher recurrence rate than pure germinomas [4]. Patients in the intermediate prognosis group received 3 courses of CARE combination followed by 30-Gy irradiation to a generous local field and 20-Gy irradiation to the tumor site. They then received additional CARE chemotherapy every 3–4 months for a total of 5 times. Patients in the poor prognosis group received 3 courses of ifosphamide-cisplatin-etoposide (ICE) combination followed concurrently by whole brain and spinal irradiation with a dose of 30 Gy, and a 30-Gy boost delivered to a generous local field; they received additional ICE chemotherapy every 3–4 months for a total of 5 times. The 5-year OS and EFS rates in 38 patients with HCG (β-HCG)-secreting germinoma were 97.3 and 86.8%, respectively. The 5-year OS and EFS rates of 40 patients in the
82
Matsutani
Table 2. Therapeutic classification Good prognostic group Germinoma, pure Intermediate prognostic group Germinoma with STGC Immature teratoma Teratoma with malignant transformation Mixed tumors mainly composed of germinoma or teratoma Poor prognostic group Choriocarcinoma Yolk sac tumor Embryonal carcinoma Mixed tumors mainly composed of choriocarcinoma, yolk sac tumor, or embryonal carcinoma
intermediate prognosis group excluding HCG-secreting germinoma were 92.2 and 87.2%, respectively; they were 61.0 and 65.6% in the poor prognosis group [32]. The international CNS GCT study group tested the effectiveness and safety of 2 chemotherapy regimens that consisted of the administration of cisplatin, etoposide, cyclophosphamide, and bleomycin (regimen A) and carboplatin, etoposide, and bleomycin (regimen B). Patients without CR underwent salvage surgery with or without irradiation. Of 20 patients with nongerminomatous tumors, 8 remained in complete remission and 6 remained in enduring second or third complete remission. The 5-year EFS and OS rates were 45 and 75%, respectively. The results obtained in patients with nongerminomatous tumors were considered encouraging [33]. The International Society for Pediatric Hematology and Oncology treated 122 nongerminomatous patients with focal radiation therapy of 54 Gy for localized tumor or craniospinal irradiation of 54 Gy for disseminated tumors after 4 courses of cisplatin/ etoposide/ifosphamide. The EFS rate was 68% in the absence and 72% in the presence of dissemination, and the median follow-up was 25 and 33 months, respectively [34]. The Children’s Oncology Group treated children with germinomas and normal markers with cisplatin + etoposide, alternating with vincristine + cyclophosphamide, for four cycles followed by focal irradiation (CR: 30.6 Gy;
Pineal Germ Cell Tumors
83
An interim evaluation of these trials reviewed here is encouraging; patients with germinoma are successfully treated by combination treatment with chemotherapy and a reduced dose of irradiation to increase the cure rate while decreasing radiation-induced side effects including anterior pituitary dysfunction, and patients with nongerminomatous, moderately malignant tumors present with good 5-year or median survival without tumor progression.
References 1 The Committee of Brain Tumor Registry of Japan: Report of Brain Tumor Registry of Japan (1969–1996), 11th Edition. Neurol Med Chir (Tokyo) 2003;43(suppl i–vii);1–111. 2 Central Brain Tumor Registry of the United States: CBTRUS (2008). Statistical Report: Primary Brain Tumors in the United States, 2000– 2004. Chicago, CBTRUS, 2008. 3 Louis DN, Ohgaki H, Wiestler OD, Cavenee WK (eds): WHO Classification of Tumours of the Central Nervous System. Lyon, IARC, 2006, p 309. 4 Matsutani M, Sano K, Takakura K, Fujimaki T, Nakamura O, Funata N, Seto T: Primary intracranial germ cell tumors: a clinical analysis of 153 histologically verified cases. J Neurosurg 1997;86: 446–455. 5 Saeki N, Takami K, Murai H, Kubota M, Yamaura A, Uchida D, Koguchi Y, Nakamura S, Tatsuno I, Wada K, Minagawa M, Yasuda T: Long-term outcome of endocrine function in patients with neurohypophyseal germinomas. Endocr J 2000;47: 83–89. 6 Korogi Y, Takahashi M, Ushio Y: MRI of pineal region tumors. J Neurooncol 2001;54:251–261. 7 O’Brien DF, Hayhurst C, Pizer B, Mallucci CL: Outcomes in patients undergoing single-trajectory endoscopic third ventriculostomy and endoscopic biopsy for midline tumors presenting with obstructive hydrocephalus. J Neurosurg 2006;105 (suppl 3):219–226. 8 Shono T, Natori Y, Morioka T, Torisu R, Mizouguchi M, Nagata S, Inamura T, Fukui M, Oka K, Sasaki T: Results of a long-term follow-up after neuroendoscopic biopsy procedure and third ventriculostomy in patients with intracranial germinomas. J Neurosurg 2007;107(suppl 3):193–198. 9 Wellons III JC, Reddy AT, Tubbs RS, Abdullatif H, Oakes WJ, Blount JP, Grabb PA: Neuroendoscopic findings in patients with intracranial germinomas correlating with diabetes insipidus. J Neurosurg 2004;100:430–436.
84
10 Katakami H, Hashida S, Matsutani M, The Japanese Pediatric Brain Tumor Study Group: All CNS germinomas produce human chorionic gonadotropin as assessed by real time PCR and an ultrasensitive EIA (abstract). Neuro-oncology 2007;9: 212. 11 Haddock M, Schild SE, Scheithauer BW, Schomberg PJ: Radiation therapy for histologically confirmed primary central nervous system germinoma. Int J Radiat Oncol Biol Phys 199;738: 915–923. 12 Sawamura Y, Ikeda J, Shirato H, Tada M, Abe H: Germ cell tumours of the central nervous system: treatment consideration based on 111 cases and their long-term clinical outcomes. Eur J Cancer 1998;34:104–110. 13 Aoyama H, Shirato H, Kakuto Y, Inakoshi H, Nishino M, Yoshida H, Hareyama M, Yanagisawa T, Watarai J, Miyasaka K: Pathologically-proven intracranial germinoma treated with radiation therapy. Radiother Oncol 1998;47:201–205. 14 Shibamoto Y, Sasai K, Oya N, Hiraoka M: Intracranial germinoma: radiation therapy with tumor volume-based dose selection. Radiology 2001; 218:452–456. 15 Ogawa K, Shikama N, Toita T, Nakamura K, Uno T, Onishi H, Itami J, Kakinohara Y, Kinjo T, Yoshii Y, Ito H, Murayama S: Long-term results of radiotherapy for intracranial germinoma: a multi-institutional retrospective review of 126 patients. Int J Radiat Oncol Biol Phys 2004; 58: 705–713. 16 Shirato H, Nishio M, Sawamura Y, Myohjin M, Kitahara T, Nishioka T, Mizutani Y, Abe H, Miyasaka K: Analysis of long-term treatment of intracranial germinoma. Int J Radiat Oncol Biol Phys 1997;37:511–515. 17 Jenkin D, Berry M, Chan H, Greenberg M, Hendrick B, Hoffman H, Humphereys R, Sonley M, Weitzman S: Pineal region germinomas in childhood. Treatment considerations. Int J Radiat Oncol Biol Phys 1990;18:541–545.
Matsutani
18 Matsutani M, Sano K, Takakura K: Long-term follow-up of patients with primary intracranial germinomas; in Packer R, Bleyer WL, Pochedly C (eds): Pediatric Neuro-Oncology. Chur, Harwood Academic Publishers, 1992, pp 254–260. 19 Shibamoto Y, Abe Y, Yamashita J, Takahashi M, Hiraoka M, Ono K, Tsutsui K: Treatment result of intracranial germinoma as a function of the irradiated volume. Int J Radiat Oncol Biol Phys 1988; 15:285–290. 20 Rogers SJ, Mosleh-Shirazi MA, Saran FH: Radiotherapy of localised intracranial germinoma: time to sever historical ties? Lancet Oncol 2005;6:509– 519. 21 Allen JC, Kim JH, Packer RJ: Neoadjuvant chemotherapy for newly diagnosed germ cell tumors of the central nervous system. J Neurosurg 1987; 67:65–70. 22 Yoshida J, Sugita K, Kobayashi T, Shitara N, Matsutani M, Tanaka R, Nagai H, Yamada H, Yamashita J, Oda Y, Hayakawa T, Ushio Y: Prognosis of intracranial germ cell tumours: effectiveness of chemotherapy with cisplatin and etoposide (CDDP and VP-16). Acta Neurochir (Wien) 1993; 120:111– 117. 23 Balmaceda C, Heller G, Rosenblum M, Diez B, Villablanca JG, Kellie S, Maher P, Vlamis V, Walker RW, Leibel S: Chemotherapy without irradiation-a novel approach for newly diagnosed CNS germ cell tumors: results of an international cooperative trial. J Clin Oncol 1996;14:2908–2915. 24 Kellie SJ, Boyce H, Dunkel IJ, Rosenblum M, Brualdi L Finlay JL: Primary Chemotherapy for intracranial nongerminomatous germ cell tumors: results of the second international CNS germ cell study group protocol. J Clin Oncol 2004;22: 846–853. 25 Matsutani M, Ushio Y, Abe H, Yamashita J, Shibui S, Fujimaki T, Takakura K, Nomura K, Tanaka R, Fukui M, Yoshimoto T, Hayakawa T, Nagahsima T, Kurisu K, Kayama T: Combined chemotherapy and radiation therapy for central nervous system germ cell tumors: preliminary results of a phase II study of the Japanese Pediatric Brain Tumor Study Group. Neurosurg Focus 1998;5:e7.
26 Matsutani M, The Japanese Pediatric Brain Tumor Study Group: Combined chemotherapy and radiation therapy for CNS germ cell tumors – the Japanese experience. J Neurooncol 2001;54: 311–316. 27 Matsutani M, The Japanese Pediatric Brain Tumor Study Group: Treatment for intracranial germinoma; final results of Japanese Study Group (abstract). Neuro-oncology 2005;7:519. 28 Alapetite C, Patte C, Frappaz D, Sainte-Rose C, Kieffer R, Raquin MA, Baranzelli MC: Long-term follow-up of intracranial germinoma treated with primary chemotherapy followed by focal radiation treatment: The SFOP-90 experience (abstract). Neuro-oncology 2005;7:517. 29 Calaminus G, Alapetite C, Frappaz D, Garré ML, Koch S, Kortmann RD, Nicholson J, Ricardi U, Saran F: Update of protocol patients with CNS germinoma treated according to slop CNS GCT 96 (abstract). Neuro-oncology 2005;7:518. 30 Kretschmar C, Kleinberg L, Greenberg M, Burger P, Holmes E, Wharam M: Pre-radiation chemotherapy with response-based radiation therapy in children with central nervous system germ cell tumors: a report from the Children’s Oncology Group. Pediatr Blood Cancer 2007;48: 285–291. 31 Jennings MT, Gelman R, Hochberg F: Intracranial germ-cell tumors: natural history and pathogenesis. J Neurosurg 1985;63:155–167. 32 Matsutani M, The Japanese Pediatric Brain Tumor Study Group: Treatment for intracranial nongerminoma; final results of Japanese Study Group (abstract). Neuro-oncology 2005;7:527. 33 Kellie SJ, Boyce H, Dunkel IJ, Rosenblum M, Brualdi L Finlay JL: Primary chemotherapy for intracranial nongerminomatous germ cell tumors: results of the second international CNS germ cell study group protocol. J Clin Oncol 2004;22:846– 853. 34 Calaminus G, Alapetite C, Frappaz D, Garré ML, Koch S, Kortmann RD, Nicholson J, Ricardi U, Saran F: Update of protocol patients with CNS nongerminoma treated according to slop CNS GCT 96 (abstract). Neuro-oncology 2005;7:526.
Prof. Masao Matsutani, MD, DMSci Department of Neuro-Oncology, Saitama International Medical Center, Saitama Medical University Yamane 1397-1, Hidaka-City Saitama 350–1298 (Japan) Tel. +81 42 984 4400, Fax +81 42 984 0438, E-Mail
[email protected]
Pineal Germ Cell Tumors
85
Tumors of Germ Cell Origin Kobayashi T, Lunsford LD (eds): Pineal Region Tumors. Diagnosis and Treatment Options. Prog Neurol Surg. Basel, Karger, 2009, vol 23, pp 86–95
Strategy of Combined Treatment of Germ Cell Tumors Yutaka Sawamura Departments of Neurosurgery, Hokkaido University Hospital, Sapporo, and Teikyo University Hospital, Tokyo, Japan
Abstract The histopathological entity ‘germ cell tumor’ (GCT) encompasses a number of histological subtypes. Pineal GCTs can be grossly divided into three categories: those with a good, intermediate, and poor prognostic. Germinoma and mature teratoma are curable and classified into the good prognostic group, whereas embryonal carcinoma, yolk sac tumor, and other highly malignant neoplasms leave patients with a dismal prognosis. There are other types of GCT that have an intermediate prognosis, such as immature teratoma. Only mature teratomas are curable by surgical resection alone; the other types require adjuvant therapy. To plan a surgical strategy, the neurosurgeon has to acquire enough knowledge of the effect of adjuvant therapies and biological behavior of the GCTs. Germinoma can be cured by low-dose radiotherapy in combination with chemotherapy, and nowadays needs only to be biopsied. Other tumors, such as highly malignant tumors need a sophisticated combination therapy that includes surgery, craniospinal radiation therapy, and intensive chemotherapy. An appropriate neoadjuvant therapy prior to radical surgical removal will remarkably reduce the surgical risk. The goal of treatment should be tightly focused on the reduction of posttreatment sequelae, including surgical morbidity, and Copyright © 2009 S. Karger AG, Basel not on a complete microsurgical resection.
The histopathological entity ‘germ cell tumor’ (GCT) encompasses a number of histological subtypes whose responses to adjuvant therapy and prognoses are diverse. Furthermore, the similarity of clinical presentation and radiological findings among pineal GCTs with different histological malignancies makes their management complex [1, 2]. No prospective study to elucidate a proper staging has been reported in the literature. For selecting a therapeutic regimen, CNS GCTs have been traditionally divided into two major groups, so called germinomatous GCTs and nongerminomatous GCTs, as a simple extrapolation from gonadal GCTs. To consider the prognoses and to select an appropriate therapeutic plan, CNS GCTs
Table 1. Classification of pineal GCTs to consult when selecting appropriate management 1. Good prognostic group Germinoma Mature teratoma 2. Intermediate prognostic group Immature teratoma Mixed GCTs consisting of germinoma with either mature or immature teratoma 3. Poor prognostic group Teratoma with malignant transformation including a component of squamous cell carcinoma, adenocarcinoma, or sarcoma Embryonal carcinoma Yolk sac tumor Choriocarcinoma Mixed GCTs including a component of embryonal carcinoma, yolk sac tumor, choriocarcinoma, or teratoma with malignant transformation
can be grossly divided into three categories, namely good, intermediate, and poor prognostic groups (table 1) [2]. Germinoma and mature teratoma are highly curable and classified into the good prognostic group, whereas embryonal carcinoma, yolk sac tumor, choriocarcinoma, teratoma with malignant transformation, and mixed GCT including a component of cancer or sarcoma leave patients with a dismal prognosis. Between these good and poor prognostic groups, there are other types of GCT that have an intermediate prognosis, such as immature teratoma and mixed GCT composed of teratoma and germinoma [3, 4]. Among the diverse histological subtypes, only mature teratomas are curable by a surgical resection alone, and the other types of pineal GCTs require adjuvant therapy. To plan a surgical strategy, the neurosurgeon treating patients with pineal GCTs has to acquire enough knowledge of the effect of adjuvant therapies and biological behavior on the GCTs. The planning of the neurosurgical management greatly depends on the biological nature of the individual neoplasm and should be determined by evaluating preoperative radiological findings, levels of serum/cerebrospinal fluid (CSF) tumor markers, and an intraoperative histological diagnosis using frozen sections, as well as the surgeon’s experience. Germinoma, which is the most common tumor originating from the pineal body, can be cured by low-dose radiotherapy in combination with chemotherapy, and nowadays needs only to be biopsied (fig. 1). Conversely, mature teratoma can
Combined Treatment of Germ Cell Tumors
87
a
b
Fig. 1. a Contrast axial MRI showing a pineal mass in a 17-year-old man. Endoscopic biopsy revealed a germinoma and simultaneous third ventriculostomy resolved obstructive hydrocephalus. b After one cycle of chemotherapy using carboplatin and etoposide, the tumor disappeared on MRI.
be cured only by a radical surgical resection. Other tumors, such as malignant teratomas, embryonal carcinomas, choriocarcinomas, and yolk sac tumors need a sophisticated combination therapy that includes surgical intervention, craniospinal radiation therapy, and intensive chemotherapy. For such tumors, neurosurgeons have to recognize that surgical intervention is just a part of the combination therapy. For a large and highly malignant pineal neoplasm, an appropriate neoadjuvant therapy prior to radical surgical removal will remarkably reduce the surgical risk. In general, a greater resection of the malignant neoplasm is associated with a better prognosis, and a radical surgical resection of invasive tumor in the pineal region carries a significant risk of operative morbidity. Although the occipital transtentorial approach has become a safe surgical procedure in the experienced neurosurgeon’s hands [5], the goal of treatment should be tightly focused on the reduction of posttreatment sequelae including surgical morbidity, and not on a complete microsurgical resection itself. When a primary pineal GCT is considered, endoscopic tumor biopsy is first recommended [6].
Germinoma
Germinomas account for approximately 70% of intracranial GCTs. The level of serum/CSF β-human chorionic gonadotropin (β-HCG) in patients with
88
Sawamura
pineal germinoma may be slightly elevated, approximately <100 mIU/ml. The primary goal of surgery for pineal germinomas is to obtain a sufficient volume of tumor tissue for an accurate histological examination. If preoperative radiological studies indicate a strong suspicion of germinoma, biopsy samples should be obtained by a craniotomy, stereotactic, or endoscopic procedure. Because the initial manifestation of a pineal germinoma is usually caused by obstructive hydrocephalus due to the aqueduct obstruction, currently an endoscopic biopsy with simultaneous third ventriculostomy is most commonly applied. Endoscopic biopsy of marker-negative GCTs is a safe, reliable method of establishing a diagnosis of germinoma [6]. If an intraoperative histological diagnosis of germinoma was made during craniotomy, the resection should be stopped as germinoma is an invasive tumor. Near the end of tumor resection we often encounter a residual mass invading the posterior commissure, the periaqueductal white matter, and the superior colliculus. Stopping the procedure at this point reduces the complication rate significantly without compromising the high cure rate of this unique neoplasm. In addition, a complete surgical removal alone inevitably causes early relapse of germinoma. It is, therefore, clear that a radical resection of pineal germinoma offers no benefit over biopsy [7]. It is known that germinomas are highly chemosensitive tumors, and the agents that have been examined in previous clinical studies are bleomycin, carboplatin, cisplatin, cyclophosphamide, etoposide, ifosfamide, and vinblastine [2]. The most common combination in chemotherapeutic regimens includes carboplatin/ cisplatin plus etoposide. Germinomas tend to be treated with a lower dose of irradiation than those used with conventional radiotherapy of 40–55 Gy [8, 9]. Preirradiation chemotherapy has been advocated as an adjuvant therapy to further decrease the total volume of radiation therapy [10–13]. A European study has suggested that preirradiation chemotherapy followed by 30–40 Gy of irradiation may be adequate for treating germinomas [14]. Aoyama et al. [10] reported the results of an induction chemotherapy followed by 24 Gy in 12 fractions to the involved field. With a mean follow-up duration of 58 months, however, 6 of 27 patients with germinoma had a relapse. This high relapse rate is attributed to the small radiation field that they employed. The whole ventricle field, therefore, is recommended as the smallest target volume for germinoma [13]. The combined chemoradiotherapy approach is associated with minimal endocrinopathy and minimal neurocognitive dysfunction. Patients with relapses after low-dose radiation therapy can respond well to salvage therapy without significant sequelae [11]. Biopsy failure of a mixed GCT is not rare. If a mixed GCT is mistakenly interpreted as ‘pure germinoma’ after an initial biopsy, at least the germinoma
Combined Treatment of Germ Cell Tumors
89
a
b
d
e
c
f
Fig. 2. A pineal mixed GCT in an 11-year-old man. Contrast-enhanced computed tomography image (a), T2-weighted (b) and contrast-enhanced (c) MR images showing a pineal mass which was heterogeneously enhanced and isointense on T2. Endoscopic biopsy revealed a germinoma. The patient received 4 cycles of ICE chemotherapy using ifosfamide, cisplatin, and etoposide, and then whole-ventricle field irradiation with 25.2 Gy in 14 fractions. During the first cycle of the chemotherapy, the enhanced portion of the mass rapidly decreased in size, but a well-demarcated mass remained. The mass showed similar intensity to the CSF on T2-weighted image (d) and contrast-enhanced T1-weighted image (e), and slightly low signal intensity on cisternography (f). It was totally resected through occipital transtentorial approach and was determined to be an epidermoid cyst.
component can be eradicated by adjuvant therapy, and other components may remain. After that, a second-look surgery is feasible to resect the residual tumor that was resistant to the adjuvant therapy (fig. 2). When the level of HCG is normalized, the residual mass is almost always a teratoma.
90
Sawamura
Mature Teratoma
The presurgical examination showing heterogeneous appearance on magnetic resonance imaging (MRI), dense calcification, multiple cysts, and negative/low levels of HCG/α-fetoprotein (AFP) indicates a mature teratoma. Pineal mature teratoma is a target for radical removal by craniotomy. If a complete resection of a mature teratoma has been performed, it is not necessary to give adjuvant therapy [15]. A pure pineal mature teratoma is, however, rare and often part of a mixed GCT, including germinoma and/or immature teratoma. Even after a complete resection, histological misdiagnosis may occur in determining the subtypes of teratoma, which leads to an inadequate therapy. Emphasis must be placed on performing an extremely careful histological examination in which the entire part of the resected specimen is searched. A large pineal teratoma may be partially removed or biopsied to avoid surgical complication. Even if histological diagnosis of the resected specimen is mature teratoma, the residual part may contain a small portion of immature or other malignant tissue. It has been a matter of controversy whether the partially removed mature teratoma should be treated with adjuvant therapy or simply observed until relapse. A true mature teratoma is resistant to chemoradiation therapy and eventually grows during the adjuvant therapy. Children with mature teratoma should not be exposed to the potential toxicities of unnecessary chemotherapy or radiation therapy. When a histological diagnosis of mature teratoma is surely obtained after a partial resection, a second radical resection is recommended. Mixed GCTs occasionally show growing teratoma syndrome that is a paradoxical response to chemoradiation therapy [16]. During an adjuvant therapy, a secreting or malignant portion responds, but simultaneously a portion of mature teratoma continues to grow (fig. 3). The enlarging tumor consists of elements of a mature teratoma that are refractory to chemotherapy or radiation. In such a case, radical surgical removal is also indispensable.
Immature Teratoma
Immature teratomas require adjuvant therapy because they often recur even after a gross total resection. There has been minimal information available in the literature concerning the response of immature teratomas to chemotherapy [15, 17]. A moderate dose of local irradiation, approximately 40–50 Gy to the primary site, in combination with cisplatin-based chemotherapy may be an appropriate adjuvant therapy for immature teratomas.
Combined Treatment of Germ Cell Tumors
91
a
b
c
d
e
Fig. 3. Plain (a) and contrast-enhanced (b) MR images showing a suprasellar teratoma; most of the enhancement is due to contrast medium. Biopsy revealed a mature teratoma mixed with germinoma. During chemotherapy, the enhanced portion of the mass decreased in size, but simultaneously the adipose tissue component, indicated by the high-signal area on plain MR imaging (c) or the low-density area on computed tomography scanning (d), remarkably increased. e The growing mass was surgically resected and was determined to be a pure dermoid cyst. From Sawamura, et al. [15].
A rigid histological examination of immature teratomas often finds a tiny component of other GCT. It should be noted that an immature teratoma contains substantially heterogeneous phenotypes which show various responses to adjuvant therapy. In addition, an immature teratoma may recur as a different histological subtype of GCTs such as pure germinoma, yolk sac tumor, choriocarcinoma, embryonal carcinoma, dermoid cyst, or epidermoid cyst. Because an adjuvant therapy for pineal immature teratoma usually results in partial response, if possible, the residual mass should be completely resected to circumvent a relapse. In contrast to
92
Sawamura
the teratoma occurring in the hypothalamic pituitary axis, a pineal immature teratoma can be resected without causing intolerable neurological sequelae.
Malignant Germ Cell Tumors
GCTs including a component with highly malignant histology, such as embryonal carcinoma, yolk sac tumor, choriocarcinoma, teratoma with malignant transformation with either squamous cell carcinoma, adenocarcinoma, mesenchymal sarcoma, or other sarcomas are distressing tumors. Since these tumors are invariably fatal, the selection of therapy is directed toward improving the survival rate. For these highly aggressive GCTs, more extensive resection may be associated with improved survival, and adjuvant therapy includes multidrug intensive chemotherapy as well as craniospinal axis radiotherapy. For the highly malignant GCTs, standard combined chemotherapy is platinum-based regimen. The tumor markers β-HCG and/or AFP in serum and CSF must be examined prior to any treatment, including surgical biopsy. Preoperative examination of these serum markers has a significant value in predicting the prognosis of a patient, though it is never definitive for histology. Choriocarcinomas secrete extremely high levels of HCG. The value of AFP found in patients with yolk sac tumors may be over several thousands ng/ml. The extremely high levels of AFP in the body fluids have been thought to be the hallmark of yolk sac tumors, though embryonal carcinomas also produce high levels of AFP. Furthermore, teratomas including malignant components produce various levels of these tumor markers. Chemoradiation sensitivity also varies among these malignant neoplasms. Prior to a radical resection, neoadjuvant therapy including chemotherapy and radiation therapy has recently been advocated in the treatment of large and malignant pineal GCTs secreting tumor markers [18]. The purpose of the neoadjuvant therapy is to obtain a visible tumor bulk reduction on neuroimaging, to reduce the vascularity of a hemorrhagic tumor, and to avoid surgery-induced dissemination during radical resection of a biologically active tumor. A safer and complete surgical resection can be performed after an effective neoadjuvant therapy (fig. 4). The patients with residual mass and normal tumor markers should undergo a second look resection because most patients have residual teratoma or necrotic tissue and can be spared additional chemotherapy or radiation [11]. The remaining mass may contain a component of mature teratoma, immature teratoma, yolk sac tumor, or no viable tumor cells in fibrous tissue with necrosis [19]. Malignant pineal GCTs show a high incidence of subarachnoid dissemination or spinal metastases. Craniospinal radiotherapy with high-dose local boost
Combined Treatment of Germ Cell Tumors
93
a
b
c
Fig. 4. a Contrast-enhanced MR image showing a malignant pineal GCT in a 5-year-old male. Because the level of serum AFP was over 2,000 ng/ml, surgical biopsy was not performed. This patient was treated with ICE chemotherapy using ifosfamide, cisplatin, and etoposide, and then craniospinal radiation therapy with 25.2 Gy in 14 fractions and 19.8 Gy local boost. b The tumor reduced in size forming cysts, and the level of AFP was normalized. c A complete removal of the residual tumor was achieved and histological examination determined a mature teratoma with fibrous necrotic tissue.
remains indispensable, though the effects of radiotherapy alone have been discouraging. Additional intensive chemotherapy should be utilized to improve the poor prognosis. For a small residual tumor after initial irradiation, stereotactic irradiation may be indicated to increase the dose of local boost, giving an additional high dose solely to the tumor tissue avoiding excessive dose to the normal brain tissue. Combination chemotherapy regimens using cisplatin, carboplatin, etoposide, ifosfamide, or other agents appear to be effective for these highly malignant GCTs. The author has employed a regimen including ifosfamide, etoposide, and cisplatin, and obtained some promising results [10, 15]. The relatively low success rate of chemotherapy in patients with systemic GCT who have failed to respond to etoposide-cisplatin combinations has led to preliminary evaluations of high-dose chemotherapy with autologous bone marrow transplant or, more recently, peripheral blood stem cell support. The high-dose chemotherapy followed by autologous stem cell rescue is to be studied, but only in the salvage setting of patients with relapsed or disseminated GCT.
94
Sawamura
References 1 Reis F, Faria AV, Zanardi VA, Menezes JR, Cendes F, Queiroz LS: Neuroimaging in pineal tumors. J Neuroimaging 2006;16:52–58. 2 Sawamura Y, de Tribolet N: Current diagnosis and treatment of central nervous system germ cell tumours. Current Opin Neurol 1996;9:419–423. 3 Ogawa K, Toita T, Nakamura K, Uno T, Onishi H, Itami J, Shinkwa N, Saeki N, Yoshii Y, Murayama S: Treatment and prognosis of patients with intracranial nongerminomatous malignant germ cell tumors: a multiinstitutional retrospective analysis of 41 patients. Cancer 2003;98:369–376. 4 Sawamura Y, Ikeda J, Shirato H, Tada M, Abe H: Germ cell tumors of the central nervous system: treatment consideration based on 111 cases and their long-term clinical outcomes. Eur J Cancer 1998;34:104–110. 5 Sawamura Y, de Tribolet N: Neurosurgical management of pineal tumours. Adv Tech Stand Neurosurg 2001;27:1–22. 6 Luther N, Edger MA, Dunkel U, Souweidane MM: Correlation of endoscopic biopsy with tumor marker status in primary intracranial germ cell tumors. J Neurooncol 2006;79:45–50. 7 Sawamura Y, de Tribolet N, Ishii N, Abe H: Surgical management of primary intracranial germinomas. Diagnostic surgery or radical resection? J Neurosurg 1997;87:262–266. 8 Bamberg M, Kortmann RD, Calaminus G, Becker G, Meisner C, Harms D, Gobel U: Radiation therapy for intracranial germinoma: results of the German cooperative prospective trials MAKEI 83/86/89. J Clin Oncol 1999;17:2585–2592. 9 Haas-Kogan DA, Missett BT, Wara WM, Donaldson SS, Lamborn KR, Prados MD, Fisher PG, Huhn SL, Fisch BM, Berger MS, Le QT: Radiation therapy for intracranial germ cell tumors. Int J Radiat Oncol Biol Phys 2003;56:511–518. 10 Aoyama H, Shirato H, Ikeda J, Fujieda K, Miyasaka K, Sawamura Y: Induction chemotherapy followed by low-dose involved-field radiotherapy for intracranial germ cell tumors. J Clin Oncol 2002;20: 857–865.
11 Kaur H, Singh D, Peereboom DM: Primary central nervous system germ cell tumors. Curr Treat Options Oncol 2003;4:491–498. 12 Sawamura Y, Shirato H, Ikeda J, Tada M, Ishii M, Kato T, Abe H, Fujieda K: Induction chemotherapy followed by reduced-volume irradiation for newly diagnosed CNS germinoma. J Neurosurg 1998;88:66–72. 13 Shirato H, Aoyama H, Ikeda J, Fujieda K, kato N, Ishi N, Miyasaka K, Iwasaki Y, Sawamura Y: Impact of margin for target volume in low-dose involved field radiotherapy after induction chemotherapy for intracranial germinoma. Int J Radiat Oncol Biol Phys 2004;60:214–217. 14 Calaminus G, Bamberg M, Baranzelli MC, Benoit Y, di Montezemolo LC, Fossati-Bellani F, Jurgens H, Kuhl HJ, Lenard HG, Curto ML: Intracranial germ cell tumors: a comprehensive update of the European data. Neuropediatrics 1994;25:26–32. 15 Sawamura Y, Kato T, Ikeda J, Murata J-I, Tada M, Shirato H: Teratomas of the central nervous system: treatment consideration based on 34 cases. J Neurosurg 1998;89:728–737. 16 Bi WL, Bannykh SI, Baehring J: The growing teratoma syndrome after subtotal resection of an intracranial nongerminomatous germ cell tumor in an adult: case report. Neurosurgery 2005;56: 188. 17 Garre ML, El-Hossainy MO, Fondelli P, Gobel U, Brisigotti M, Donati PT, Nantron M, Ravegnani M, Garaventa A, De Bernardi B: Is chemotherapy effective therapy for intracranial immature teratoma? A case report. Cancer 1996;77:977–982. 18 Kochi M, Itoyama Y, Shiraishi S, Kitamura I, Marubayashi T, Ushio Y: Successful treatment of intracranial nongerminomatous malignant germ cell tumors by administering neoadjuvant chemotherapy and radiotherapy before excision of residual tumors. J Neurosurg 2003;99:106–114. 19 Nakamura H, Takeshima H, Makino K, Kuratsu J: Evaluation of residual tissues after adjuvant therapy in germ cell tumors. Pediatr Neurosurg 2007; 43: 82–91.
Yutaka Sawamura, MD, PhD Department of Neurosurgery, Hokkaido University Hospital North 15, West 7, Kita-ku Sapporo 060-8638 (Japan) Tel. +81 11 706 5987, Fax +81 11 708 7737, E-Mail
[email protected]
Combined Treatment of Germ Cell Tumors
95
Tumors of Germ Cell Origin Kobayashi T, Lunsford LD (eds): Pineal Region Tumors. Diagnosis and Treatment Options. Prog Neurol Surg. Basel, Karger, 2009, vol 23, pp 96–105
Radiation Therapy for Intracranial Germ Cell Tumors Hidefumi Aoyama Department of Radiology, Hokkaido University Graduate School of Medicine, Sapporo, Japan
Abstract Although radiation therapy (RT) is essential to the management of intracranial germ cell tumors, the ideal radiation dose and field remain controversial. For the treatment of germinoma, whole central nervous system radiation, which was once the standard RT field, is being replaced by whole ventricle (WV) field radiation for localized disease. The use of induction chemotherapy has been expected to further reduce the RT field and dose; however, use of a localized field smaller than the WV field has resulted in a higher recurrence rate. Therefore, the WV field should be considered appropriate even after induction chemotherapy. With regard to the radiation dose to the primary tumor site, it can be reduced to 40–45 Gy in RT alone. The further reduction of the radiation dose when using a combination of chemotherapy and RT is yet to be determined. Unlike germinomas, nongerminomatous germ cell tumors, with the exception of mature teratomas, are refractory to conventional RT. The whole central nervous system field should thus be used for all but immature teratomas. Given that local progression is the primary pattern of recurrence even after effective induction chemotherapy, RT dose increase through the use of modern techniques, including stereotactic irradiation and intensity-modulated RT, should be investigated. Copyright © 2009 S. Karger AG, Basel
Primary central nervous system (CNS) germ cell tumors (GCTs) are rare, accounting for 1–2% of all brain tumors and for 3–10% of brain tumors in children. Approximately half occur in the pineal region. The second most predominant site is the neurohypophyseal region. Multifocal presentation in these regions is occasionally seen. Among intracranial GCTs, approximately 60% are germinomas [1–4]. Radiation therapy (RT) plays a primary role in the treatment of germinomas and a 90–100% curative rate can be expected after RT alone [1–4]. On the other hand, nongerminomatous GCTs (NGGCTs) are generally aggressive and most patients with nongerminomatous tumors treated conventionally with surgery and RT fail to survive longer than 3 years [1].
Table 1. Germinoma – WCNS RT First author
Patients
Radiation dose, Gy
All
WONS WB
WV
Primary
Dose per fraction
36 30
– –
– –
50 50
1.8 1.5
30.6 25.6 21
36 – –
– – 30
50.4 50.4 49.5
1.8 1.5–1.8 1.5
with Spinal dissemination
Bamberg [5] Bamberg [5]
11 49
0 0
Maity [6] Merchant [7] Schoenfeld [8]
39 12 31
13 1 0
Followup years
Outcomes, %
Relapses
OS
RFS
All
9.8 7.4
100 (5) 92 (5)
100 (5) 87 (5)
0 (0%) 0 (0%) 5 (10.2%) 1 (2%)
7.1 5.8 7.0
100 (10) 100 88 (10)
100 (10) 100 94 (10)
0 (0%) 0 (0%) 2 (9.5%)
Figures in parentheses in the outcomes columns indicate years.
Treatment of Intracranial Germinoma
Reduction of Whole Central Nervous System Dose Historically, whole CNS (WCNS) radiation followed by a local boost has been the standard treatment for intracranial germinomas. Radiation doses of 30–36 Gy to the WCNS and 50 Gy to the primary tumor site have frequently been used. Although this treatment schedule has achieved a satisfactory outcome, several investigators have tried to reduce the dose of RT to the WCNS. A German multiinstitutional prospective trial (MAKEI 83/86/89) assessed whether the dose to the WCNS could be reduced in an RT alone approach [5]. They initially used 36 Gy to the WCNS and then reduced this dose to 30 Gy. They observed only 1 recurrence within the CNS. Several others investigators assessed the possibility of reducing the WCNS dose to 30.6 Gy, 25.6 Gy, or 21 Gy [6–8]. The results of these attempts are summarized in table 1. The long-term relapse-free survival (RFS) and overall survival (OS) rates were 87–100% and 88–100%, respectively. It is notable that spinal failure was observed in only 1 case, which was treated with 30 Gy to the WCNS. Thus, it could be said that the prophylactic WCNS radiation dose could be reduced to around 21–25 Gy without increasing the risk of spinal relapses.
Attempts to Reduce the Radiation Field and Dose to the Primary Site Despite the excellent outcome of tumor control with a WCNS RT approach, radiation-induced late sequelae have been a matter of concern. Reductions of
Radiation Therapy for Intracranial Germ Cell Tumors
97
Spinal W/Wo others
0 (0%) 0 (0%) 0 (0%)
Table 2. Germinoma – Neoadjuvant chemotherapy followed by focal radiation First author
Aoyama [14] Lafay-Cousin [12] Bouffet [20] Matsutani [24] Nguyen [16]
Year
2002 2006 1999 2001 2006
Patients
28 6 57 75 9
RT dose, Gy
24 (Focal) 24-40 WV 40 (Focal) 24 (Focal) 30.6 (Focal)
Followup years
4.8 4 3.5 2.9 7.8
Outcomes, %
Relapses
OS
RFS
All
Spinal w/ wo others
93.8 (5) 100 98 (3) ND 89 (10)
66 (5) 100 96.4 (3) ND 62/42 (5/10)
6 (21%) 0 (0%) 4 (7%) 9 (12%) 4 (44%)
2 (7%) 0 (0%) 2 (4%) ND 2 (2%)
ND = Not documented Figures in parentheses in the outcomes columns indicate years.
the radiation field and dose have been investigated in the context of RT alone or a combination treatment using chemotherapy and RT (table 2). Ogawa et al. [9] assessed the relationship between radiation field and treatment outcome in a review of 126 patients with intracranial germinoma who were treated by RT alone. More than half of the patients were treated with a whole brain (WB) radiation field, while a WCNS field was used only for 56 patients (44%). It is noteworthy that the incidence of spinal relapses was 4% (2 of 56) for patients treated with spinal irradiation and 3% (2 of 70) for those without spinal irradiation. Shirato et al. [10] reviewed 51 patients with germinoma treated by RT alone. The radiation field was the WCNS in 16, WB in 9, whole ventricle (WV) in 21, and local field without ventricle coverage in 5 patients. The 10-year cause-specific survival rates for pathologically verified and unverified germinomas were 100 and 96%, respectively. No relapse was noted in patients in whom the WV field radiation was used. They concluded that 40 Gy WV irradiation is appropriate for most intracranial germinomas. Haas-Kogan et al. [11] analyzed 41 cases of localized germinoma patients and reached a similar conclusion. Of 41 localized germinoma patients, 18 patients were treated with WV field radiation with a median dose of 32.4 Gy. None of them experienced tumor recurrence. Lafay-Cousin et al. [12] examined whether bifocal germinomas could be treated with chemotherapy followed by WV field radiation. The radiation field was WV with or without a boost to the primary tumor locations. All patients remain in complete remission at a median follow-up of 48.1 months. The authors suggested that bifocal germinoma can be considered a loco-regional rather than a metastatic disease and could be treated with WV RT. The use of an RT field smaller than the WV has been mainly investigated for the
98
Aoyama
a
b
c
Fig. 1. Pineal germinoma treated with cisplatin-etoposide followed by focal radiation of 24 Gy. a Enhanced MRI after surgery. b Dose distribution of focal RT. c Tumor recurrence developed 34 months after treatment. The patient was successfully treated with WCNS radiation, and remains free of recurrence.
combination approach of induction chemotherapy and RT. A Group at Hokkaido University conducted a prospective study aimed at reducing the RT volume and dose after 3–4 cycles of cisplatin-based induction chemotherapy [13, 14]. Twentyfive patients with intracranial germinoma were enrolled in the protocol treatment. The actuarial survival rate and progression-free survival rate were 93.8 and 66%, respectively. Recurrence of disease was observed in 6 patients. The reason for these recurrences was examined in detail [15], and it was determined that all the relapses occurred in patients who received focal irradiation without WV coverage (fig. 1), while there were no relapses among those treated with a wider safety margin. Nguyen et al. [16] retrospectively compared 9 patients treated with neoadjuvant chemotherapy and focal radiation (median, 30.6 Gy) and 12 patients who received WCNS (median, 24 Gy) and a local boost up to 50 Gy. All 4 tumor recurrences occurred among those who received neoadjuvant chemotherapy and focal radiation. The 10-year progression-free survival was 41.5% in the focal radiation group versus 100% in the WCNS group. The rate of distant control in the spine at 5 years was 62% for patients who received focal irradiation and 100% for patients who received WCNS radiation (p = 0.04). The relation between the radiation field and pattern of recurrence in publications is summarized in table 3 [5–20]. The
Radiation Therapy for Intracranial Germ Cell Tumors
99
Table 3. Germinoma – Radiation field and pattern of recurrences First author
Year
RT field Protocoled Patients chemotherapy
Total relapses
Failure at spine w/wo other sites
Bamberg [5] Maity [6] Merchant [7] Schoenfeld [8] Ogawa [9] Haddock [17] Aoyama [18] Shirato [10] Shibamoto [19] Shirato [15] Nguyen [16]
1999 2004 2000 2006 2004 1997 1998 1997 2001 2004 2006
WCNS WCNS WCNS WCNS WCNS WCNS WCNS WCNS WCNS WCNS WCNS
5 (8%) 0 (0%) 0 (0%) 2 (6%) 3 (5%) 0 (0%) 2 (9%) 1 (6%) 1 (4%) 0 (0%) 0 (0%)
1 (1.6%) 0 (0%) 0 (0%) 0 (0%) 2 (3.5%) 0 (0%) 0 (0%) 0 (0%) 0 (0%) 0 (0%) 0 (0%)
14 (4.8%)
3 (1%)
62 10 10 9
3 (5%) 1 (10%) 2 (20%) 0 (0%)
2 (3.2%) 0 (0%) 1 (10%) 0 (0%)
91
6 (6.5%)
3 (3.2%)
2 21 18 6 6
1 (50%) 2 (10%) 0 (0%) 0 (0%) 0 (0%)
0 (0%) 1 (5%) 0 (0%) 0 (0%) 0 (0%)
53
3 (5.6%)
1 (1.8%)
6 11 8 5 3 13 18 57 9
3 (50%) 5 (45.4%) 5 (62.5%) 2 (40%) 1 (33.3%) 1 (7.6%) 6 (33.3%) 4 (7.0%) 4 (44.4%)
0 (0%) 4 (36.3%) 1 (12.5%) 1 (20%) 0 (0%) 0 (0%) 2 (11%) 2 (3.5%) 3 (33.3%)
31 (23.8%)
13 (10%)
No No No No No No No No No Yes No
60 39 12 31 56 10 23 16 25 3 12 287
Ogawa [9] Haddock [17] Aoyama [18] Shirato [10]
Ogawa [9] Shirato [10] Haas-Kogan [11] Shirato [15] Lafay-Oousin [12]
Ogawa [9] Haddock [17] Aoyama [18] Shirato [10] Haas-Kogan [11] Shibamoto [19] Shirato [15] Bouffet [20] Nguyen [16]
2004 1997 1998 1997
2004 1997 2003 2004 2006
2004 1997 1998 1997 2003 2001 2004 1999 2006
WB WB WB WB
WV WV WV WV WV
Focal Focal Focal Focal Focal Focal Focal Focal Focal
No No No No
No No No Yes Yes
No No No No No No Yes Yes Yes
130
100
Aoyama
overall recurrence rates are around 5% when the WCNS, WB, or WV fields are used, whereas they range from 7 to 62% (mean, 23.8%, 31/130) and exceed 30% in most series when a focal field is used. The frequency of spinal failure with or without failure at other sites is about 1–3% when WCNS, WB, or WV fields are used, whereas it is significantly increased to 10% or more (range 0–36) when a focal field is used. Therefore, the radiation field could be reduced from WCNS to WV without deterioration of the frequency of tumor recurrence. However, an RT field smaller than the WV might result in a significant increase in tumor recurrence even after effective induction chemotherapy. With regard to radiation dose, Shibamoto et al. [19] prospectively investigated the possibility of radiation dose reduction to 40–45 Gy for germinomas smaller than 4 cm. Although 2 patients developed meningeal dissemination, none had local failure. In the series of induction-chemotherapy followed by RT, The French Society for Pediatric Oncology (SFOP) used 40 Gy to the initial tumor volume and treated 59 germinoma patients [20]. After a median follow-up of 42 months, 4 recurrences of disease were observed. The 3-year actuarial progressionfree survival was 96.4%. Lafay-Cousin et al. [12] used 24–40 Gy WV field RT and no recurrence was observed among 18 patients for 4 years. The Hokkaido University Group used 24 Gy to the primary tumor site. Although 6 recurrences were observed, these recurrences had more to do with the radiation field. That is, all the recurrences were seen among the 16 patients treated with a localized field, and no relapses were seen in the 9 patients treated with WV (24 Gy) or WCNS (24 Gy). Based on these findings, it is still unclear whether a radiation dose lower than 40 Gy can be safely used even after effective induction chemotherapy [13–15].
Nongerminomatous Germ Cell Tumors: Prognosis after a Conventional Approach
Unlike germinomas, the treatment outcomes of NGGCTs are less than satisfactory (table 4). With the exception of mature teratomas, NGGCTs are generally refractory to treatment: most patients with NGGCTs treated conventionally with surgery and RT fail to survive longer than 3 years [1]. Spinal dissemination is more common than in germinomas and systemic metastases, especially to the lung and bone, and occurs in 3% of patients [1]. In a review by Sawamura et al. [2] of 111 cases treated at Hokkaido University, the probability of 10-year survival was 67% for immature teratomas, but only 25% for GCTs that included a highly malignant component. In a review of 153 histologically verified GCTs by Matsutani et al. [3], the 10-year survival rates of patients with mature teratomas and malignant teratomas were 92.9 and 70.7%, respectively, whereas those with pure malignant GCTs (embryonal carcinoma, yolk sac tumor, or choriocarcinoma) had a 3-year
Radiation Therapy for Intracranial Germ Cell Tumors
101
Table 4 . NGGCTs, prognosis First author
Sawamura [2]
Patients Pathology
9 7 15
Matsutani [3] 11 11
10 12
Schild [21] 57
Immature teratoma IMT with Germinoma Highly malignant GCTs Malignant teratomas Pure malignant GCTs (embryonal carcinoma, yolk sac tumor, choriocarcinoma) Mixed (germinoma or teratoma) Mixed (embryonal carcinoma, yolk sac tumor, choriocarcinoma) Immature teratoma Mixed GCTs Others
Follow- OS, % up years
7.2
8.1
3
Recurrences Total
Local w/wo other site
67 (10) 69 (10) 25 (10)
ND ND ND
ND ND ND
70.7 (10) 27.3 (5)
6 (54.5%) 6 (54.5%) 10 (90.9%) 8 (72.7%)
35 (10)
4 (40%)
9.3 (5)
10 (83.3%) 8 (66.7%)
67 (3) 44 (3) 13 (3)
ND ND ND
2 (20%)
ND ND ND
Figures in parentheses in the OS column indicate years.
survival rate of as low as 27.3%. In the review of 57 cases reported by Schild et al. [21], the 3-year survival rate was 86% for patients with mature teratomas, 67% for patients with immature teratomas, 44% for patients with mixed GCTs, and 13% for patients with the other histologic types. A retrospective review by Aoyama et al. [22] considered the cases of 24 NGGCT patients: 5 patients with mature teratoma with/without germinoma (group 1), 6 patients with immature teratoma with/without germinoma (group 2), and 13 patients with other highly malignant tumors (group 3). The 5-year actuarial RFS rate was 100% for group 1, 63% for group 2, and 44% for group 3. RT constitutes an essential part of treatment strategy. Schild et al. [21] reported that patients who received radiotherapy had a 3-year survival rate of 46% compared to 11% for those patients who did not receive radiotherapy (p = 0.0015). The radiation field should be selected according to the frequency of spinal dissemination. Immature teratomas have a generally lower rate of spinal relapse compared
102
Aoyama
with other highly malignant histologies. Aoyama et al. [22] examined 12 patients with immature teratomas with/without germinomas. There was no spinal failure in patients with immature teratoma, irrespective of the fact that 5 of the 6 did not receive WCNS radiation. There were 2 patients who experienced tumor recurrence at the primary tumor site. In the above-mentioned review of Matsutani et al. [3], all 6 recurrences among the 11 cases of malignant teratomas arose in the primary tumor site. On the other hand, teratomas with malignant transformation are thought to have a higher risk of spinal dissemination [21, 22]. For other highly malignant NGGCTs, the risk of spinal dissemination is also high. Aoyama et al. [22] reported that spinal failure was observed in 3 of 8 patients (37.5%) who did not receive WCNS radiation, but in none of the 5 who did receive it. Matsutani et al. [3] observed a 21.7% (5/23) rate of spinal failure in patients with highly malignant NGGCTs treated with WB RT. The radiation dose must be determined according to the probability of local tumor recurrence. When a conventional radiation dose (around 50 Gy) is used, the local tumor control rate is generally low. Matsutani et al. [3] reported that 34 (55.7%) of 61 NGGCTs recurred or metastasized. Twentyfive of those 34 patients (73.5%) suffered from tumor recurrence at the primary site with or without recurrence at a remote site. Hass-Kogan et al. [11] reported that in 5 of the 7 NGGCT patients with disease progression or relapse, the primary site was a component of failure. This high incidence of failure at the primary site after a conventional radiation dose highlights the need for more intensive chemotherapy or higher radiation doses to the primary site for patients with poor responses to chemotherapy or radiation.
Prospective Clinical Studies: Chemotherapy with Radiation
Calaminus et al. [23] reported the long-term outcome of 41 patients with intracranial malignant NGGCTs enrolled in the German prospective protocol MAKEI 89. The protocol recommended, after a clinically or histologically proven diagnosis and cisplatin-based chemotherapy, a resection of tumor and WCNS RT (30 Gy) with a tumor boost (20 Gy). The 5-year RFS rate was 74% in those treated according to the protocol. The use of WCNS RT had a significant influence on survival (p = 0.035), as did a cumulative cisplatin dose ≥400 mg/m2 (p = 0.002). The Japanese Pediatric Brain Tumor Study Group is now conducting a phase II study of platinum-based induction chemotherapy followed by RT [24]. The 5-year OS and event-free survival for malignant teratoma with/without germinoma were 93.6 and 83%, respectively. The 3-year OS of 23 patients with highly malignant GCTs was 65.5%. The 3-year survival rates of 74% obtained from the German trial and 65.5% obtained from the Japanese trial seem to be significantly better than those obtained by a conventional approach.
Radiation Therapy for Intracranial Germ Cell Tumors
103
Late Radiation Toxicities
Cranial RT has a potential risk of causing late radiation sequelae. This is especially true for medulloblastoma patients. Perhaps the most devastating sequela in cranial radiation in medulloblastoma patients, IQ decline, progresses more rapidly in association with several risk factors, including younger age, hydrocephalus, use of radiation, radiation dose and field. However, it is not fully understood if the findings obtained in patients with medulloblastoma could be extrapolated to patients with GCTs. Recently, two authors reported a favorable outcome in cognitive function after WCNS radiation for germinomas. In one study, Merchant et al. [7] observed no IQ decline in 25 patients with germinoma who were treated with WCNS RT. In another study, Sutton et al. [25] assessed the QOL of 22 patients treated for intracranial germinoma with WCNS irradiation. All 22 patients are in or have completed high school, 9 are in or have completed college, and 5 have higher degrees. Intracranial GCTs are usually diagnosed after puberty, and thus the influence of RT on the brain parenchyma or pituitary gland must be smaller than that in patients with medulloblastoma. Therefore, in reporting late radiation toxicities, great care must be taken not to include changes caused by the tumor itself or other confounding factors.
References 1 Jennings MT, Gelman R, Hochberg F: Intracranial germ-cell tumors: natural history and pathogenesis. J Neurosurg 1985;63:155–67. 2 Sawamura Y, Ikeda J, Shirato H, Tada M, Abe H: Germ cell tumours of the central nervous system: treatment consideration based on 111 cases and their long-term clinical outcomes. Eur J Cancer 1998;34:104–110. 3 Matsutani M, Sano K, Takakura K, Fujimaki T, Nakamura O, Funata N, Seto T: Primary intracranial germ cell tumors: a clinical analysis of 153 histologically verified cases. J Neurosurg 1997;86:446–455. 4 Matsutani M: Clinical management of primary central nervous system germ cell tumors. Semin Oncol 2004;31:676–683. 5 Bamberg M, Kortmann RD, Calaminus G, Becker G, Meisner C, Harms D, Göbel U: Radiation therapy for intracranial germinoma: results of the German cooperative prospective trials MAKEI 83/86/89. J Clin Oncol 1999;17:2585–2592.
104
6 Maity A, Shu HK, Janss A, Belasco JB, Rorke L, Phillips PC, Sutton LN, Goldwein JW: Craniospinal radiation in the treatment of biopsy-proven intracranial germinomas: twenty-five years’ experience in a single center. Int J Radiat Oncol Biol Phys 2004;58:1165–1170. 7 Merchant TE, Sherwood SH, Mulhern RK, Rose SR, Thompson SJ, Sanford RA, Kun LE: CNS germinoma: disease control and long-term functional outcome for 12 children treated with craniospinal irradiation. Int J Radiat Oncol Biol Phys 2000;46:1171–1176. 8 Schoenfeld GO, Amdur RJ, Schmalfuss IM, Morris CG, Keole SR, Mendenhall WM, Marcus RB Jr: Low-dose prophylactic craniospinal radiotherapy for intracranial germinoma. Int J Radiat Oncol Biol Phys 2006;65:481–485. 9 Ogawa K, Shikama N, Toita T, Nakamura K, Uno T, Onishi H, Itami J, Kakinohana Y, Kinjo T, Yoshii Y, Ito H, Murayama S: Long-term results of radiotherapy for intracranial germinoma: a multi-institutional retrospective review of 126 patients. Int J Radiat Oncol Biol Phys 2004;58:705–713.
Aoyama
10 Shirato H, Nishio M, Sawamura Y, Myohjin M, Kitahara T, Nishioka T, Mizutani Y, Abe H, Miyasaka K: Analysis of long-term treatment of intracranial germinoma. Int J Radiat Oncol Biol Phys 1997;37:511–515. 11 Haas-Kogan DA, Missett BT, Wara WM, Donaldson SS, Lamborn KR, Prados MD, Fisher PG, Huhn SL, Fisch BM, Berger MS, Le QT: Radiation therapy for intracranial germ cell tumors. Int J Radiat Oncol Biol Phys 2003;56:511–518. 12 Lafay-Cousin L, Millar BA, Mabbott D, Spiegler B, Drake J, Bartels U, Huang A, Bouffet E: Limited-field radiation for bifocal germinoma. Int J Radiat Oncol Biol Phys 2006;65:486–492. 13 Sawamura Y, Shirato H, Ikeda J, Tada M, Ishii N, Kato T, Abe H, Fujieda K: Induction chemotherapy followed by reduced-volume radiation therapy for newly diagnosed central nervous system germinoma. J Neurosurg 1998; 88:66–72. 14 Aoyama H, Shirato H, Ikeda J, Fujieda K, Miyasaka K, Sawamura Y: Induction chemotherapy followed by low-dose involved-field radiotherapy for intracranial germ cell tumors. J Clin Oncol 2002;20:857–865. 15 Shirato H, Aoyama H, Ikeda J, Fujieda K, Kato N, Ishi N, Miyasaka K, Iwasaki Y, Sawamura Y: Impact of margin for target volume in lowdose involved field radiotherapy after induction chemotherapy for intracranial germinoma. Int J Radiat Oncol Biol Phys 2004;60: 214–217. 16 Nguyen QN, Chang EL, Allen PK, Maor MH, Ater JL, Mahajan A, Wolff JE, Weinberg JS, Woo SY: Focal and craniospinal irradiation for patients with intracranial germinoma and patterns of failure. Cancer 2006;107:2228–2236. 17 Haddock MG, Schild SE, Scheithauer BW, Schomberg PJ: Radiation therapy for histologically confirmed primary central nervous system germinoma. Int J Radiat Oncol Biol Phys 1997; 38:915–923.
18 Aoyama H, Shirato H, Kakuto Y, Inakoshi H, Nishio M, Yoshida H, Hareyama M, Yanagisawa T, Watarai J, Miyasaka K: Pathologically-proven intracranial germinoma treated with radiation therapy. Radiother Oncol 1998;47:201–205. 19 Shibamoto Y, Sasai K, Oya N, Hiraoka M: Intracranial germinoma: radiation therapy with tumor volume-based dose selection. Radiology 2001; 218:452–456. 20 Bouffet E, Baranzelli MC, Patte C, Portas M, Edan C, Chastagner P, Mechinaud-Lacroix F, Kalifa C: Combined treatment modality for intracranial germinomas: results of a multicentre SFOP experience. Société Française d’Oncologie Pédiatrique. Br J Cancer 1999;79:1199–1204. 21 Schild SE, Haddock MG, Scheithauer BW, Marks LB, Norman MG, Burger PC, Wong WW, Lyons MK, Schomberg PJ: Nongerminomatous germ cell tumors of the brain. Int J Radiat Oncol Biol Phys 1996;36:557–563. 22 Aoyama H, Shirato H, Yoshida H, Hareyama M, Nishio M, Yanagisawa T, Kakuto Y, Watarai J, Inakoshi H, Miyasaka K: Retrospective multi-institutional study of radiotherapy for intracranial non-germinomatous germ cell tumors. Radiother Oncol 1998;49:55–59 23 Calaminus G, Bamberg M, Harms D, Jürgens H, Kortmann RD, Sörensen N, Wiestler OD, Göbel U: AFP/beta-HCG secreting CNS germ cell tumors: long-term outcome with respect to initial symptoms and primary tumor resection. Results of the cooperative trial MAKEI 89. Neuropediatrics 2005;36:71–77. 24 Matsutani M; Japanese Pediatric Brain Tumor Study Group. Combined chemotherapy and radiation therapy for CNS germ cell tumors – the Japanese experience. J Neurooncol 2001;54:311–316. 25 Sutton LN, Radcliffe J, Goldwein JW, Phillips P, Janss AJ, Packer RJ, Zhao H: Quality of life of adult survivors of germinomas treated with craniospinal irradiation. Neurosurgery 1999;45:1292–1297.
Hidefumi Aoyama, MD, PhD Department of Radiology, Hokkaido University Graduate School of Medicine North 15, West 7, Kia-ku Sapporo 060-8638 (Japan) Tel. +81 11 706 5977, Fax +81 11 706 7876, E-Mail
[email protected]
Radiation Therapy for Intracranial Germ Cell Tumors
105
Tumors of Germ Cell Origin Kobayashi T, Lunsford LD (eds): Pineal Region Tumors. Diagnosis and Treatment Options. Prog Neurol Surg. Basel, Karger, 2009, vol 23, pp 106–118
Stereotactic Radiosurgery for Pineal and Related Tumors Yoshimasa Moria ⭈ Tatsuya Kobayashia ⭈ Toshinori Hasegawab ⭈ Kouta Yoshidab ⭈ Yoshihisa Kidab a Nagoya Radiosurgery Center, Nagoya Kyoritsu Hospital, Nagoya, and bGamma Knife Center, Komaki City Hospital, Komaki, Japan
Abstract Radiosurgery is increasingly being used to treat pineal region tumors, either as an additional therapy after conventional treatments or as a primary treatment. We report our experience with Gamma Knife radiosurgery (GKRS) for the treatment of pineal and related tumors. Forty-nine patients underwent GKRS for pineal and related tumors (n = 74) between February 1992 and September 2007. The diagnosis was germ cell tumors (GCTs) in 38 patients (53 tumors), pineal parenchymal tumors (PPTs) in 9 (19 tumors), and unknown in 2 (2 tumors). The mean treatment volume was 3.3 ml (range 0.1–22 ml) in GCT cases and 3.7 ml (range 0.3–23 ml) in PPT cases. Prescribed doses around 50% isodose line ranged from 9.9 to 25.7 Gy. One patient (one tumor) with pineocytoma was lost to follow-up. Median clinical and imaging follow-up in the remaining 48 cases was 33.5 months (range, 3–192 months). Survival rates at 5 years and 10 years after GKRS in GCT cases (n = 38) were both 68%. They were 100 and 67%, respectively, in PPT cases (n = 8). We evaluated the treatment results with categorization of GCT cases into 2 groups, i.e. germinoma (group 1), and germinoma with syncytiotrophoblastic giant cell and malignant GCT (group 2). PPT cases were also divided into 2 groups, i.e. pineocytoma (group 3) and pineoblastoma and mixed pineocytoma/pineoblastoma (group 4). Local tumor control (LTC) rates at 3 and 5 years were 82% in group 1 (n = 18), 72 and 62% in group 2 (n = 35), and 85% in group 3 (n = 13). LTC rate at 2 years was 30% in group 4 (n = 5). In group 1 (n = 16), progression-free survival (PFS) rates at 3 and 5 years were 79 and 63%. They were 43 and 37% in group 2 (n = 22), and 80% in group 3 (n = 5). PFS rate at 2 years was 33% in group 4 (n = 3). Germinoma and pineocytoma showed higher LTC and PFS rates after GKRS, though pineoblastoma was liable to relapse. Intermediate prognosis was obtained in germinoma with syncytiotrophoblastic giant cell and malignant GCT. GKRS is expected to be an effective and safe adjuvant treatment approach to Copyright © 2009 S. Karger AG, Basel pineal and related tumors.
Several types of tumors are found in the pineal region. Germ cell tumors (GCTs) are the most frequent, of which germinomas and teratomas account for more than 80% [1]. GCTs are also found in the suprasellar, basal ganglia, thalamus and other regions in addition to the pineal region. Pineal parenchymal tumors (PPTs) such as pineocytoma and pineoblastoma are found infrequently. Other infrequent tumors are glial tumors, meningiomas, epidermoid tumors and so on. Treatment strategies also vary according to the pathological type. Germinomas can be cured by chemotherapy with relatively small doses of fractionated irradiation, and benign tumors, such as mature teratomas and meningiomas, can be cured by microsurgery alone. However, optimal treatments for other tumors have not been established yet. Radiosurgery is being increasingly used to treat pineal region tumors, either as the principal treatment modality or in conjunction with conventional external beam radiation therapy. The reason behind the increased use of radiosurgery is that the highly conformal treatment planning possible with this technique allows delivery of cytotoxic doses of radiation to the target field while minimizing the dose delivered to neighboring vital structures. Hence, radiosurgery may be a less morbid alternative for treating deep-seated lesions surrounded by eloquent structures. We report our series of 49 patients with pineal region tumors treated by Gamma Knife radiosurgery (GKRS). The role of GKRS for treatment of pineal and related tumors is discussed.
Materials and Methods We treated 49 patients (74 tumors) during the period from February 1992 through September 2007. Diagnosis of pineal and related tumors was GCT in 38 patients, PPT in 9, and unknown in 2 (table 1). In 13 of 38 patients with GCT, GCT was histologically diagnosed, but in others GCT was diagnosed by elevation of serum tumor markers or clinical course. In all 9 cases of PPT, diagnosis was histologically confirmed. In 2 cases, pathological diagnosis had not been obtained and no elevation of serum tumor markers was found.
Characteristics of Cases Out of 38 GCT patients, 27 were male and 11 female. The mean age was 21 years (range 8–66 years). Initial symptoms were, as shown in table 2, headache and gait disturbance, which was due to hydrocephalus, and extraocular muscle movement disorder, such as Parinaud phenomenon, diabetes insipidus, pituitary gland dysfunction, and so on. Thirteen patients had undergone surgical resection or biopsy prior to GKRS. Twenty patients had received conventional external beam radiation therapy previously, the mean dose of which was 45 Gy and 7 of whom had received only focal irradiation. Twenty-seven had received chemotherapy previously. Only 2 (one germinoma and one malignant GCT) had undergone GKRS as an initial treatment. In 7 patients, ventriculoperitoneal shunt had been done for hydrocephalus prior to GKRS. All 9 cases of PPT had surgical resection before GKRS. Two patients previously received radiation therapy
Radiosurgery for Pineal Tumors
107
Table 1. Histology of pineal and related tumors (n = 49) Histology
Patients n
%
38 16 6 16
78 33 12 33
PPT Pineocytoma Pineoblastoma Mixed tumor
9 6 2 1
18 12 4 2
Unknown (no pathology)
2
4
GCT Germinoma Germinoma with STGC Malignant GCT
Table 2. Initial symptoms in cases of pineal and related tumors Symptom
Headache Diabetes insipidus Diplopia Hydrocephalus Nausea Gait disturbance Hemiparesis Growing retardation Amenorrhea Incidental Unknown
Histology GCT
PPT
unknown
14 9 6 4 3 2 2 1 1 3 1
2
1
3 1 3
1 1
of the mean dose of 65 Gy and 3 had chemotherapy. In 7 of 9 cases, ventriculoperitoneal shunt (6) or third ventriculostomy (1) was done before. One of 2 patients with histologically unknown pineal tumor had undergone radiation therapy and chemotherapy previously but they were not effective. The other patient underwent GKRS as an initial treatment. The median period between diagnosis and GKRS was 262 days in GCT cases and 95 days in PPT cases.
Dose Planning Seventy-four lesions were treated by GKRS. Fifty-three tumors were GCTs, 19 PPTs, and 2 histologically unknown pineal tumors. Among GCT cases, treated tumors were in the primary
108
Mori · Kobayashi · Hasegawa · Yoshida · Kida
Table 3. Dose planning of GKRS Variable
Tumor diameter, mm Tumor volume, ml Maximum dose, Gy Marginal dose, Gy Percent isodose Isocenters
GCT (n = 53)
PPT (n = 19)
range
mean
range
mean
6.0–34.8 0.1–22.0 14.5–40.0 9.9–25.7 45–95 1–14
15.5 3.3 27.4 15.5 58.3 3.4
8.0–35.3 0.3–23.0 17.0–40.0 12.5–20.3 45–95 1–8
16.4 3.7 27.9 16.7 61.9 2.6
pineal region in 36 cases and in other location in 17. Among PPT cases, they were in the primary pineal region in 8 and in other location in 11 cases. Both of the histologically unknown pineal tumors had no other region tumor. Three-dimensional dose planning was conducted using KULA system (Elekta, Tokyo) or Leksell GammaPlan (Elekta) with magnetic resonance imaging (MRI) T1-weighted images or spoiled gradient recalled acquisition images with gadolinium enhancement. Tumor size and treatment dose of GKRS are shown in table 3.
Methods of Evaluation The effects of GKRS were evaluated based mainly on changes in tumor size on MRI taken every 3–6 months after treatment. A five-grade system was devised using criteria proposed by the Japan Brain Tumor Registry [2]. The five grades were complete response (CR, tumor disappearance), partial response (PR, ≥50% tumor volume reduction), minor response (MR, 25–50% reduction), no change (NC, <25% reduction), and progression (tumor enlargement). From these values, the control rate can be calculated as CR + PR + MR + NC/total. Further evaluation based on changes in neurological signs and/or tumor markers were made as well.
Results
Imaging follow-up was obtained in 48 out of 49 cases. One case of pineocytoma (one pineal tumor) was lost to follow-up. The median and mean follow-up period was 33.5 and 49 (range, 3–192) months from GKRS. The mean follow-up period from diagnosis was 70 (range, 8–264) months. The 5-year and 10-year survival rates after GKRS were both 68% in GCT cases (n = 38); they were analyzed using Kaplan-Meyer curve. The survival rates from diagnosis were 84 and 63%, respectively. In PPT cases (n = 8), the survival rates from GKRS were 100 and 67%, and those from diagnosis were 100 and 50%, respectively. Local tumor control (LTC) rates at 5 years after GKRS were 70% in GCT cases and 71% in PPT cases.
Radiosurgery for Pineal Tumors
109
1.0
Local tumor control
N.S. (p = 0.42) 0.8 0.6 0.4 0.2
Group 1 (n = 18) Group 2 (n = 35)
0 0
25
50
75
100 125 Months
150
175
200
Fig. 1. LTC of group 1 (germinoma) and group 2 (germinoma with STGC and malignant GCT).
The LTC rates for primary pineal lesions and disseminated lesions at 5 years after GKRS were 67 and 84%, respectively. We evaluated the treatment results with categorization of GCT cases into 2 groups, i.e. cases of germinoma (group 1), and those of germinoma with syncytiotrophoblastic giant cell (STGC) and malignant GCT (group 2). PPT cases were also divided into 2 groups, i.e. cases of pineocytoma (group 3) and those of pineoblastoma and mixed pineocytoma/pineoblastoma (group 4). LTC rates at 3 and 5 years were both 82% in group 1 (n = 18; fig. 1). They were 72 and 62% in group 2 (n = 35; fig. 1), and both 85% in group 3 (n = 13; fig. 2). The LTC rate at 2 years was 30% in group 4 (n = 5; fig. 2). The progression-free survival (PFS) rate of at 3 and 5 years was 79 and 63% in group 1 (n = 16; fig. 3). They were 43 and 37% in group 2 (n = 22; fig. 3). They were both 80% in group 3 (n = 5; fig. 4). The PFS rate at 2 years was 33% in group 4 (n = 3; fig. 4). LTC in group 3 had a statistically significantly better rate than that in group 4 (p < 0.01, log-rank test). The PFS rate tended be better in group 1 than in group 2 (p = 0.10). It also tended to be better in group 3 than in group 4 (p = 0.12). Actually, in the cases of pure germinoma (group 1) only 4 of 16 (25%) had tumor progression after GKRS. Among 5 cases of pineocytoma with follow-up (group 3), the tumors were controlled in 4 cases. Only one case developed cerebrospinal fluid dissemination after GKRS. In the 2 cases of histologically unknown pineal tumors, the tumors were controlled for 22 and 30 months, respectively. Concerning side effects, only one patient who underwent reirradiation by GKRS after whole brain radiation therapy of 40 Gy in 20 fractions developed optic neuropathy in both eyes 7 months after GKRS.
110
Mori · Kobayashi · Hasegawa · Yoshida · Kida
Local tumor control
1.0 0.8
p < 0.01
0.6 0.4 0.2
Group 3 (n = 13) Group 4 (n = 5)
0 0
20
40
60
80
100
120
140
Months
Fig. 2. LTC of group 3 (pineocytoma) and group 4 (pineoblastoma and mixed pineocytoma/ pineoblastoma). There was a significant difference (p < 0.01) between the two groups in logrank, Mantel-Cox test.
1.0 N.S. (p = 0.10)
0.8
PFS
0.6 0.4 0.2
Group 1 (n = 16) Group 2 (n = 22)
0 0
25
50
75
100
125
150
175
200
Months
Fig. 3. PFS of group 1 and group 2.
Illustrative Cases Case 1: 17-Year-Old Boy
A 17-year-old boy (fig. 5) had shown dwarfism since his childhood. Hypertrophy of the pituitary stalk and a small pineal mass were found incidentally on routine head CT. Serum human chorionic gonadotropin (HCG) was slightly elevated, and β-HCG was also elevated in the cerebrospinal fluid (CSF). The clinical diagnosis was germinoma with STGC. Three sessions of chemotherapy with cisplatin and etoposide were administered from January to October 1998.
Radiosurgery for Pineal Tumors
111
1.0 N.S. (p = 0.12) 0.8
PFS
0.6 0.4 0.2
Group 3 (n = 5) Group 4 (n = 3)
0 0
20
40
80 60 Months
100
120
140
Fig. 4. PFS of group 3 and group 4.
a
b
c
d
Fig. 5. MRI with gadolinium enhancement of case 1. a Suprasellar and pineal masses were found with increased HCG in CSF. Tumors showed PR after chemotherapy (b) but pineal tumor showed regrowth (c) and GKRS was performed. d The pineal tumor showed PR again within 19 months after GKRS.
112
Mori · Kobayashi · Hasegawa · Yoshida · Kida
a
b
c
d
e
f
Fig. 6. MRI with gadolinium enhancement of case 2. a Large pineal tumor was treated by first GKRS after chemotherapy. The tumor showed slight decrease in size (b), but CSF dissemination occurred frequently and subtotal resection of the tumor was made (c). d As the residual tumor gradually increased, second GKRS was given. The tumor showed CR within 18 months (e), and tumor-free state lasted without increase in tumor marker for more than 81 months (f).
Both tumors showed PR, but the pineal tumor regrew with CSF dissemination in February 1999. GKRS was conducted with a margin dose of 14 Gy in March 1999. CSF tumor markers became negative and the tumor showed PR at 19 months after treatment.
Case 2: 14-Year-Old Girl A 14-year-old girl (fig. 6) complained of headache in July 1991. This progressed to double vision, nausea and vomiting in November when CT scans revealed a pineal tumor and hydrocephalus with disseminated lesions in the frontal ventricle. Serum HCG and α-fetoprotein were elevated. Under a clinical diagnosis of malignant GCT, 2 sessions of chemotherapy with cisplatin and etoposide were conducted. The tumor diminished slightly and GKRS was performed with a margin dose of 15 Gy in February 1992. However, CSF dissemination was widespread
Radiosurgery for Pineal Tumors
113
a
b
c
d
Fig. 7. MRI with gadolinium enhancement of case 3. GKRS with a margin dose of 15.3 Gy was performed (axial section, a; coronal, b). The tumor was decreased in size and complete remission was obtained within 45 months after GKRS (axial, c; coronal, d).
after the treatments. Intrathecal injections of methotrexate through an Ommaya reservoir were followed by whole skull irradiation of 40 Gy. Dissemination stopped and subtotal removal of the tumor was carried out. The diagnosis was teratoma with embryonal carcinoma. The residual tumor gradually increased in size and a second GKRS was carried out in February 1994. The tumor showed CR with undetectable tumor markers for more than 81 months since the initial treatment.
Case 3: 68-Year-Old Female This patient (fig. 7) had developed gait disturbance and Parinaud’s sign. She was diagnosed as having hydrocephalus. Ventriculoperitoneal shunt relieved her symptoms. Three years later, a pineal tumor was detected on MRI. Histological diagnosis based on the specimen from biopsy was pineocytoma. GKRS with a margin dose of 15.3 Gy was performed. The tumor decreased in size and complete remission was obtained within 45 months after GKRS.
114
Mori · Kobayashi · Hasegawa · Yoshida · Kida
Discussion
The treatment strategy for pineal region tumors remains controversial [3]. Optimal management must be adapted on a case-by-case basis and include microsurgery, radiotherapy, and chemotherapy, alone or in combination. Specific management decisions depend mainly on the histology, age, and spread of the tumor. Pure teratoma can be cured only by microsurgical tumor removal [4]. In Japan, germinoma treatment has essentially been standardized to use combined chemotherapy and/ or conventional irradiation [5, 6]; with this treatment, a cure rate of more than 90% has been obtained. Therefore, treatment of recurrent or resistant tumor after this combination therapy is a matter of interest. GKRS is a potential choice as an additional treatment [3]. The advantage of stereotactic radiosurgery is the increase in the therapeutic ratio achieved by reducing the volume of surrounding normal tissue receiving radiation. This has particular implications for pineal tumors near the brainstem. In our study, only 3 of 16 cases of germinoma had relapse at 3, 6, and 28 months after GKRS. Only a few studies on the use of stereotactic radiosurgery for germinoma have been reported. Regine et al. [7] reported a single case with a thalamic germinoma treated only by GKRS. The patient was tumor free for 2 years. Zissiadis et al. [8] reported the results of stereotactic radiotherapy (or radiosurgery) and whole brain radiotherapy or craniospinal radiation therapy in 13 cases of newly-diagnosed germinoma. All 10 cases were alive and free from disease at follow-up from 12.4 to 72.6 months. Casentini et al. [9] reported 6 cases of germinoma treated by combined Linac-based radiosurgery and external-beam radiotherapy. Tumor disappearance was achieved in 4 cases for a mean period of 36 months. Manera et al. [10] reported 2 cases of germinoma treated by GKRS. Complete tumor disappearance was obtained in both patients at 7 and 15 days. No recurrence was observed during a follow-up period of 13 and 19 months. Subach et al. [11] reported 2 cases of germinoma treated by GKRS. Both tumors were decreased in size after GKRS. Germinoma with STGC has been classified as belonging to the intermediate prognostic group [5] known to show local recurrence or extraneural metastasis. There was only one report on the use of radiosurgery for this tumor. Kobayashi et al. [3] were the first to report on the use of GKRS in 4 patients with germinoma with STGC. Two patients showed PR, but 2 showed progression, one of whom died 12 months after GKRS. Among the 6 patients in our study, 3 showed tumor progression at 7 months (dissemination), 25 months (distant failure then local tumor progression), and 60 months (local tumor progression and dissemination) after GKRS. Regarding malignant GCTs, there have been reports suggesting the effectiveness of combined chemotherapy with conventional irradiation for these tumors. However, the results have not been encouraging; the 3-year survival rate was
Radiosurgery for Pineal Tumors
115
27.3% in 10 patients given chemotherapy using cisplatin and etoposide and focal irradiation of 50–55 Gy [5]. Lekovic et al. [12] reported the result of GKRS in a case with nongerminomatous GCT. The patient received 15-Gy GKRS boost after reduced-dose craniospinal radiation therapy of 36 Gy and was free from relapse for 73 months. Hasegawa et al. [13] reported the results of GKRS combined with external-beam radiation therapy and chemotherapy in 4 patients with nongerminomatous GCT. In 3 of 4 patients, the tumor was controlled for 4–58 months after GKRS, and one patient died from tumor progression at 5 months after radiosurgery. Zissiadis et al. [8] reported the results of stereotactic radiotherapy with or without craniospinal radiation therapy combined with chemotherapy in 5 patients with newly-diagnosed mixed germinoma. Three of 5 patients were alive and free from disease during a follow-up period from 26.7 to 46.9 months. One patient had relapse (elevation of tumor markers) at 37 months after treatment, and the other one had persistent disease for 32.1 months following treatment. Our present results using GKRS in 16 cases of malignant GTC showed mortality rate was high (31%) during a median follow-up of 20 months (range, 3–120 months). PFS was obtained only in 8 of 16 cases for 3–120 months (median, 65 months) after GKRS. Tumors arising from the pineal gland parenchyma have been classified into three subgroups according to the malignancy; pineocytoma, pineoblastoma, and mixed tumor [14]. These tumors are extremely rare, but recent reports suggest that these tumors may be treated either by stereotactic radiosurgery or in combination with chemotherapy or radiation therapy. Backlund et al. [15] reported in 1974 that 2 pineocytoma cases showed tumor disappearance after GKRS. Hasegawa et al. [16] reported 16 PPT cases, including 10 pineocytomas, 4 pineoblastomas, and 2 mixed tumors, treated by GKRS. Though 5 patients died (4 from leptomeningeal or extracranial tumor spread and one from unknown reason) during the mean follow-up period of 52 months, the local tumor was controlled in all 14 cases with imaging follow-up. Lekovic et al. [12] showed similar results in 9 cases of PPT including 8 pineocytomas and one pineoblastoma. One patient died, but in all 9 patients the LTC rate was 100% during a follow-up period of 1–96 months. Reyns et al. [17] also reported similar results of GKRS in 8 patients with pineocytoma and 5 with pineoblastoma. With a mean follow-up of 34 (range 6–88) months, all tumors responded to treatment and disappeared or stopped growing, though 2 patients with pineoblastoma had out of treatment field tumor progression requiring several radiosurgery procedures, and died because of carcinomatous meningitis or tumor size progression. Deshmukh et al. [18] reported that 5 pineocytoma patients undergoing radiosurgery showed attenuation of local disease (mean follow-up, 19.3 months; range, 6–36 months). Our present results are apparently comparable to theirs in that in 4 out of 5 pineocytoma patients the tumor was controlled. Both pineoblastoma patients had tumor progression at 3 and 13 months after GKRS.
116
Mori · Kobayashi · Hasegawa · Yoshida · Kida
GKRS was the initial treatment for 2 pineal tumors in which a pathological diagnosis had not been obtained. CR was obtained in one case and PR in the other over a follow-up period of 22 and 30 months. Clinical diagnosis may be suggestive of pineocytoma or pineoblastoma based on sensitivity to GKRS, clinical presentation, and patient age. We treated 2 patients with germinoma and malignant GCT by GKRS as an initial treatment. One germinoma patient developed CSF dissemination 15 months after the GKRS and the other patient with malignant GCT developed distant failure 3 months after GKRS. GKRS may not be recommended as an initial treatment for GCT as GCT has a nature to disseminate. If radiosurgery is used, close followup observation is thought to be necessary. Despite the small sample size, we conclude that GKRS may be useful for the treatment of tumors of benign nature, pure germinoma and pineocytoma. GKRS must, however, be used with caution if it is to be used for malignant disease.
Conclusions
Based on our experience, we believe that GKRS as an adjuvant treatment after conventional therapy for optimizing patient survival and quality of life is expected to be an effective and novel approach to pineal and related tumors. Usefulness of GKRS as an initial therapy is to be determined.
References 1 Committee of Brain Tumor Registry of Japan: Incidence of tumors in pineal region. Brain Tumor Registry of Japan 1990;7:56. 2 Committee of Brain Tumor Registry of Japan: The Criteria of the Effects of Treatments for Brain Tumors. Tokyo, Kanehara Publisher, 1995. 3 Kobayashi T, Kida Y, Mori Y: Stereotactic gamma radiosurgery for pineal and related tumors. J Neurooncol 2001;54:301–309. 4 Blakeley JO, Grossman SA: Management of pineal region tumors. Curr Treat Options Oncol 2006;7: 505–516. 5 Matsutani M, Sano K, Takakura K, Fujimaki T, Nakamura O: Combined treatment with chemotherapy and radiation therapy for intracranial germ cell tumors. Childs Nerv Syst 1998;14:59– 62.
Radiosurgery for Pineal Tumors
6 Ogawa K, Toita T, Kakinohara Y, Yamaguchi K, Miyagi K, Kinjyo T, Yamashiro K, Sawada S: Radiation therapy for intracranial germ cell tumors; predictive value of tumor response as evaluated by computed tomography. Int J Clin Oncol 1997;2:67–71. 7 Regine WF, Hodes JE, Patchel RA: Incracranial germinoma: treatment with radiosurgery alone – a case report. J Neurooncol 1998;37:75–77. 8 Zissiadis Y, Dutton S, Kieran M, Goumnerova L, Scott RM, Kooy HM, Tarbell NJ: Stereotactic radiotherapy for pediatric intracranial germ cell tumors. Int J Radiat Oncol Biol Phys 2001;51:108– 112. 9 Casentini L, Colombo F, Pozza F, Benedetti A: Combined radiosurgery and external radiotherapy of intracranial germinomas. Surg Neurol 1990; 34:79–86.
117
10 Manera L, Regis J, Chinot O, Porcheron D, Levrier O, Farnarier P, Peragut JC: Pineal region tumors: the role of stereotactic radiosurgery. Stereotact Funct Neurosurg 1996;66(suppl 1):164– 173. 11 Subach BR, Lunsford LD, Kondziolka D: Stereotactic radiosurgery in the treatment of pineal region tumors. Prog Neurol Surg 1998;14:175– 194. 12 Lekovic GP, Gonzalez LF, Shetter AG, Porter RW, Smith KA, Brachman D, Spetzler RF: Role of Gamma Knife surgery in the management of pineal region tumors. Neurosurg Focus 2007; 23:1–5. 13 Hasegawa T, Kondziolka D, Hadjipanayis G, Flickinger JC, Lunsford LD: Stereotactic radiosurgery for CNS nongerminomatous germ cell tumors. Report of four cases. Pediatr Neurosurg 2003;38:329–333.
14 Fauchon F, Jouvet A, Paquis P, Sant-Pierre G, Mottolese C, Ben Hassel M, Chauvenic L, Sichez JP, Philippon J, Schlienger M, Bouffet E: Parenchymal pineal tumors: a clinicopathological study of 76 cases. Int J Rad Oncol Biol Phys 2000;46: 959–968. 15 Backlund EO, Rahn T, Sorby B: Treatment of pinealomas by stereotactic radiation surgery. Acta Radiol Ther Phys Biol 1974;13:368–376. 16 Hasegawa T, Kondziolka D, Hadjipanayis G, Flickinger JC, Lunsford LD: The role of radiosurgery for the treatment of pineal parenchymal tumors. Neurosurgery 2002;51:880–889. 17 Reyns N, Hayashi M, Chinot L, Manera L, Peragut J-C, Blond S, Regis J: The role of Gamma Knife radiosurgery in the treatment of pineal parenchymal tumors. Acta Neurochir (Wien) 2006;148:5–11. 18 Deshmukh VR, Smith KA, Rekate HL, Coons S, Spetzler RF: Diagnosis and management of pineocytomas. Neurosurgery 2004;55:349–357.
Yoshimasa Mori, MD Nagoya Radiosurgery Center, Nagoya Kyoritsu Hospital 1-172 Hokke, Nakagawa Nagoya, Aichi 454-0933 (Japan) Tel. +81 52 362 5151, Fax +81 52 353 9126, E-Mail
[email protected]
118
Mori · Kobayashi · Hasegawa · Yoshida · Kida
Tumors of Germ Cell Origin Kobayashi T, Lunsford LD (eds): Pineal Region Tumors. Diagnosis and Treatment Options. Prog Neurol Surg. Basel, Karger, 2009, vol 23, pp 119–129
Management of Central Nervous System Germinoma: Proposal for a Modern Strategy Yuta Shibamoto Department of Radiology, Nagoya City University Graduate School of Medical Sciences, Nagoya, Japan
Abstract With the development of diagnostic, radiologic, and therapeutic modalities, strategies for management of central nervous system (CNS) germinoma are changing gradually. The author advocates that typical germinomas can be diagnosed based on their typical clinical and radiological findings, together with slight elevation of β-human chorionic gonadotropin levels in the serum and/or cerebrospinal fluid (CSF) and quick response to radiation or chemotherapy. Radiation therapy has been the standard treatment for CNS germinoma until recently. Germinomas 4 cm or less in diameter can be cured with radiation doses of 40–45 Gy. Regarding the treatment volume, an individualized approach is recommended and a focal radiation field covering at least major parts of the ventricular system is recommended if no CSF dissemination is present and CSF cytology is negative. Such irradiation is best given by intensity-modulated radiation therapy. Systemic chemotherapy with reduced doses (24–30 Gy) of radiation has to some extent been successful, but longer follow-up periods are necessary to draw conclusions regarding the superiority of this treatment over standard-dose radiation therapy. CNS germinoma patients should be completely cured with minimum morbidity, probably by employing appropriate doses of chemotherapy and intensity-modulated radiation therapy in the future. Copyright © 2009 S. Karger AG, Basel
Central nervous system (CNS) germinoma is a rare neoplasm. However, in East Asia it is not quite as uncommon as it is in Western countries. Its incidence among all childhood brain tumors is 0.1–2% in Western countries, whereas it is 1.4–6% in East Asia [1]. According to the Brain Tumor Registry of Japan [2], 1,037 patients have been reported during the years 1984–1996, and this represented 2.0% of all primary brain tumors including those of adult patients. This tumor registry is considered to cover about 30% of all primary brain tumors occurring in the Japanese population.
Pure germinoma is a highly curable tumor. Since this tumor most often affects teenagers and is highly sensitive to both radiotherapy and chemotherapy, curing germinoma with minimum morbidity is an important issue. To establish the best management for this tumor, various strategies for choosing diagnostic procedures and treatment options have been proposed. However, any strategy should change in accord with the progress in medical sciences. In this article, the author proposes a refined strategy for the best management of CNS germinoma.
Diagnosis of Central Nervous System Germinoma
In any type of tumors, histopathology is the gold standard for establishing the diagnosis. With the improvements in modern diagnostic imaging modalities and tumor marker measurements, however, do readers think histological diagnosis is still mandatory in every patient? Even with the development of neurosurgery, the probability of complications associated with taking a tissue specimen is never zero. Any complication associated with brain surgery always becomes a serious problem for the patient. CNS germinoma arises at the pineal and/or neurohypophyseal (suprasellar) regions or the basal ganglia; the most common site is the pineal region (50–60%) followed by the neurohypophyseal region (30–40%). The incidence in the basal ganglia to thalamus is usually 3–10% [3]. Occasionally, germinoma arises in both the pineal and neurohypophyseal regions, and this is a characteristic finding although nongerminomatous germ cell tumors occasionally arise at both sites too. In the study by Ogino et al. [4] reporting 103 patients, the tumors were in the pineal region in 44%, the neurohypophyseal region in 26%, both the pineal and neurohypophyseal region in 12%, the basal ganglia in 17%, and the frontal lobe in 2%. The vast majority of CNS germinoma patients are between 10 and 25 years of age, and the median age is 15–17 years. For pineal region and basal ganglia germinomas, the male to female ratio is higher than 5, whereas for neurohypophyseal germinoma, the ratio is near unity. Overall, the male to female ratio is 3–5:1. Table 1 shows proportions of male and female patients according to the tumor site in the author’s series of 156 patients with CNS germinoma. In this series, solitary pineal and neurohypophyseal lesions were present in 46 and 31% of cases, respectively. Occurrence at both sites was seen in nearly 10% and as high as 13% of the tumors arose at the basal ganglia to thalamus. The male to female ratio was over 10 in patients with a tumor in the pineal region or the basal ganglia to thalamus, while it was near unity for patients with a solitary neurohypophyseal legion. Radiologically, germinoma shows homogenous density on CT and homogenous signal intensity on T1-weighted MR images [5]. Typically, it shows homogenous and striking contrast enhancement. It may sometimes possess cysts. From these
120
Shibamoto
Table 1. Distribution of male and female patients with CNS germinoma according to tumor siite Site Pineal Neurohypophyseal Pineal + neurohypophyseal Basal ganglia, thalamus Total
Patients
Male
Female
Male:female
72 (46%) 49 (31%) 15 (9.6%) 20 (13%)
67 26 14 19
5 23 1 1
13 1.1 14 19
156 (100%)
126
30
4.2
clinical and radiological findings, it is not difficult to suspect germinoma even before any treatment. Recent development in immunohistochemistry enabled detection of very low levels of human chorionic gonadotropin (HCG), especially its β-subunit, using an ultrasensitive kit, and a study by Katakami et al. [6] showed that elevation of β-HCG levels in the serum or CSF is seen in most CNS germinoma patients. Elevation of β-HCG over 1,000 mIU/ml or 200 ng/ml would indicate presence of pure or mixed nongerminomatous germ cell tumors, especially choriocarcinoma, but low levels of elevation below 10 mIU/ml or 2 ng/ml would suggest that the tumor is pure germinoma. It should be noted that elevation of β-HCG levels can be observed in craniopharyngioma patients [7] and this tumor should be ruled out from clinical and radiological findings. Conversely, elevation of α-fetoprotein or carcinoembryonic antigen levels indicates the presence of nongerminomatous germ cell tumors or their components. Placental alkaline phosphatase is also known as a marker for germinoma [8]. Although the usefulness of placental alkaline phosphatase does not seem to be well established, the true and false positivity rates are reported to be about 50 and 1.6%, respectively, in testicular seminoma [9]. CNS germinoma is known to often show positive cerebrospinal fluid (CSF) cytology [10, 11]. It frequently shows the ‘two-cell pattern’ that is a characteristic pathological finding of germinoma. Previously, this cytological pattern was considered to be almost pathognomonic [10], but later, it was realized that other diseases can also show similar cytological patterns. Thus, CNS germinoma cannot be diagnosed by the CSF cytological findings alone, although it may be helpful in establishing the clinical diagnosis. Pure germinoma is known to quickly respond to radiation and chemotherapy. The lesions less than 3 cm in diameter usually disappear after 20 Gy of radiation therapy [12]. Marked shrinkage is also observed following systemic chemotherapy such as carboplatin and etoposide. Figure 1 shows a case of CNS germinoma with a huge disseminated lesion within the anterior horns of the lateral ventricle. The
Management of CNS Germinoma
121
primary tumor is seen at the pineal region. Both primary and disseminated lesions responded to radiation quickly, although the disseminated lesion did not disappear before a dose of 20 Gy due to its large size. At the end of treatment, no tumors were visible and the patient was free from recurrence for more than 8 years. With the typical clinical and radiological findings, mild elevation of β-HCG levels and quick response to treatment, the diagnosis of germinoma appears firm. In the author’s opinion, such patients should never be biopsied, since it is a meaningless procedure. If one insists that biopsy is necessary to exclude the possibility of mixed germ cell tumors, then a small biopsy specimen would be insufficient and total removal becomes necessary. Even when β-HCG levels are within normal limits, germinoma can be diagnosed with all the other findings stated above. Of course, there exist germinomas that do not possess above-mentioned characteristics; biopsy should be considered for such patients. Some pure germinomas do not respond so quickly to treatment, and one of the reasons for this is considered to be the presence of granulomatous components [13]. In summary, tumors with typical characteristics of germinoma need not be biopsied, especially when mild elevation of β-HCG is present. With no elevation of the tumor markers, indication for biopsy should be determined based on the typicality of the clinical and radiological findings.
Treatment of Central Nervous System Germinoma
Radiation Therapy Until recently, the standard treatment for CNS germinoma has been conventional radiation therapy. CNS germinoma can be cured by radiation with a probability as high as over 90% [14–17]. Nearly all germinomas respond quickly to radiation therapy, but a small proportion (5–10%) of the tumors recur; in the author’s experience, recurrence is more often seen at distant sites (i.e. as meningeal dissemination) rather than at the primary tumor site. Meningeal dissemination is observed at a certain probability (around 10%) after focal radiation therapy, but in some patients it may develop even after craniospinal irradiation. Metastasis to the abdomen can also develop, although rarely, in patients harboring a ventricularperitoneal shunt tube with no filter. Radiation treatment volume for CNS germinoma has been a controversial issue. Since germinoma can develop CSF dissemination at a certain percentage, it has been argued whether patients with this disease should be irradiated to the whole cerebrospinal axis. The author’s group reported treatment results according to the irradiated volume at Kyoto University [18, 19]. Depending on the era of treatment, germinoma patients were irradiated to the primary tumor site plus a margin
122
Shibamoto
a
b
c
Fig. 1. Response of disseminated CNS germinoma to radiation therapy. Scans show CNS germinoma before radiation (a), at 20 Gy (b) and 50 Gy (c). The high-density area in the pineal region after 50 Gy of radiation is a calcification.
or to the cerebrospinal axis regardless of the clinical and radiological findings. Afterwards, we have used an individualized approach. There was no significant difference in overall survival between patients receiving craniospinal irradiation and those receiving focal radiation. Figure 2 shows an updated result. The 10-year survival rate was 95% for 38 patients receiving craniospinal irradiation and 88% for 33 patients receiving focal radiation therapy; again, there was no significant difference between the two groups. In a review, Rogers et al. [20] found that the recurrence rate was 7.6% after whole brain or whole ventricular radiotherapy plus boost compared with 3.8% after craniospinal radiotherapy. Although craniospinal radiation plus boost had been perceived to be the gold standard treatment in Western countries, they concluded that reduced volume radiotherapy plus boost should replace craniospinal radiotherapy when a radiotherapy only approach is used. Accordingly, a recent worldwide trend seems to be to use an individualized approach. Since the overall incidence of CSF dissemination to the cranial subarachnoid space and within the spinal canal is 10% or less, a routine use of craniospinal irradiation appears to be an overtreatment. It is now recommended to treat the whole cerebrospinal axis when CSF cytology is positive or CSF dissemination is present. Even when dissemination is present only in the ventricular system, the author would recommend craniospinal irradiation, because it is not so harmful for patients with CNS germinoma who are usually older than 10 years. Previously, the standard dose for pure germinoma was 50 Gy in 1.8- to 2.0Gy daily fractions. However, the author’s group conducted a prospective study employing radiation therapy alone to investigate dose reduction for germinoma
Management of CNS Germinoma
123
Survival rate
1.0
0.5
0 0
24
48
72
96
120
144
168
Months
Fig. 2. Survival curves for CNS germinoma patients according to the radiation treatment volume. 䊊 = Focal radiation; 䊉 = craniospinal radiation.
[19, 21]. Using a daily dose of 1.8 Gy (focal radiation) or 1.6 Gy (craniospinal radiation), the total dose was 36 Gy after macroscopic total removal of the tumor, 40 Gy for tumors less than 2.5 cm, 45 Gy for tumors between 2.5 and 4 cm, and 50 Gy for larger tumors. An individualized approach was employed to determine the treatment volume. After a median follow-up period of 120 months, the 10-year relapse-free survival rate for 38 patients treated with this policy was 95%; this cure rate was considered to be equivalent to that obtained using a policy of uniformly giving 50 Gy. Therefore, the dose of 40–45 Gy appeared to be sufficient to cure CNS germinoma with a diameter 4 cm or less. The possibility of further dose reduction was also suggested. The German Society for Pediatric Oncology and Hematology also conducted a prospective study to reduce total radiation doses, and they found no local recurrence after a dose of 45 Gy [22]. The optimal dose for cerebrospinal axis prophylaxis has not yet been clarified. Considering the high radiosensitivity of germinoma, our group used 20–24 Gy in 13–15 fractions, and obtained a favorable outcome [18, 19, 21]. In our dose reduction study stated above, we used craniospinal doses around 20 Gy for 10 patients who had negative or equivocal CSF cytology findings but wished to be treated to the cerebrospinal axis, and found no recurrences in these patients [21]. We administered 24 Gy for patients with positive CSF cytology or CSF dissemination, but one patient developed CSF dissemination later; the reason for the recurrence remained unclear. Therefore, doses of 20 and 24 Gy were recommended for CSF dissemination-negative and positive cases, respectively. The German Society for Pediatric Oncology and Hematology also attempted at reducing the craniospinal radiation dose from 36 Gy to 30 Gy and found no difference in the outcome of patients in both groups [22].
124
Shibamoto
The use of 40-Gy dose to the primary tumor site should be meaningful when a possibility of recurrence, especially CSF dissemination, after focal radiation therapy is considered. If a patient had been treated to the primary tumor site with a dose of 40 Gy and developed CSF dissemination later (outside of the treated volume), then a second course of radiotherapy might be given to the whole cerebrospinal axis up to the dose of 24 Gy, because a total dose of 64 Gy with an interval of more than 6 months may not produce complications at a high probability. Thereafter, disseminated sites could be boosted up to 40–45 Gy. With such a treatment, patients should have a good chance of cure with second-line radiotherapy after first-line focal radiation to the primary tumor site.
Chemotherapy and Reduced-Dose Radiation The radiation dose of 40–50 Gy may not cause brain necrosis at a high probability, but it may have other unfavorable effects in adolescent as well as adult patients. Such adverse effects include hormonal insufficiency and decline in neurocognitive functions [23, 24]. To avoid possible adverse events of radiation therapy, systemic chemotherapy without employing radiation was investigated by some groups. However, the recurrence rate was unacceptably high (50–60%) [25–27]. Farng et al. [25] reported 6 recurrences among 11 patients treated with a combination of vinblastine, bleomycin, cisplatin, and etoposide. In a series of Kumabe et al. [26], 5 out of 8 patients treated with cisplatin-based chemotherapy relapsed at a mean period of 19 months during a mean follow-up period of 53 months. In a multi-institutional study, 45 patients were treated with 4–6 courses of carboplatin, bleomycin and etoposide, and the recurrence rate was 54% with a median follow-up of 35 months [27]. Although radiation therapy can salvage patients developing recurrence after primary chemotherapy [28, 29], the high recurrence rate was considered unacceptable, and the strategy of using chemotherapy alone was denied in the 1990s. Thereafter, systemic chemotherapy in combination with a reduced dose of radiation has been tested by a few groups [3, 30–32]. The dose was reduced to 24–30 Gy, and some success has been reported as compared with the results obtained by chemotherapy alone. Matsutani et al. [32] carried out a multi-institutional study in which 3 courses of carboplatin (450 mg/m3) and etoposide (150 mg/m3 × 3 days) were given before 24 Gy of focal radiation. Following the chemotherapy, 85% of 108 germinoma patients achieved complete response. With a follow-up period of 3.8 years, however, 12 out of 108 (11%) patients developed recurrence. This recurrence rate was considered to be higher than that seen after conventional radiation treatment with 50 Gy. They analyzed patterns of recurrence, and found that several recurrences developed at the margin of focal radiation fields, so it
Management of CNS Germinoma
125
was considered that the radiation field employed was too small. Based on these data, the subsequent study by the same group was carried out with the same chemotherapy regimens and extended fields of radiation therapy; the radiation field covered the entire ventricular systems and the dose remained identical (24 Gy in 12 fractions). Although the final results of this study have not been published yet, treatment results appeared to have improved as compared to the previously conducted one using a localized radiation field. To draw definitive conclusions regarding systemic chemotherapy plus generous-margin focal radiation with reduced doses, long-term results of these studies are necessary. The influence of HCG level elevation in the serum or CSF on prognosis has been a subject of controversy. Some studies reported a worse prognosis in patients with an elevated HCG level [33, 34], while others indicated similarly good prognosis in patients with normal and elevated HCG levels [4, 17, 35]. A large retrospective study by the Chubu Radiation Oncology Group in Japan suggested the latter [4]. If the prognosis of patients with an elevated HCG level is worse, treatment intensity should be increased for such patients, and this was indeed done in a study by Matsutani et al. [32]. However, the group later realized that the HCG level does not influence the prognosis, and intensified treatment is no longer used.
Future Treatment Strategy
Combination of systemic chemotherapy and reduced-dose radiation seems to have gained some success, but there still remain some questions to be answered before it becomes a universally accepted standard treatment. First, it should be clarified whether the cure rate obtained by this treatment is identical to that obtained by standard-dose radiation therapy. This requires longer follow-up periods. In the author’s opinion, radiation therapy, even when it is given around the primary tumor, is effective in eradicating tumor cells floating in CSF, because CSF is circulating. During the course of fractionated radiotherapy, tumor cells floating in CSF may come into the treatment volume so that they are easily killed because of their high radiosensitivity. Thus, the author fears that reducing the total radiation dose may increase the chance of CSF dissemination, since chemotherapeutic agents now in use do not cross the blood-brain barrier efficiently and their concentration in CSF does not become high enough to kill tumor cells in CSF. This needs to be clarified in the near future. Second, adverse events of chemotherapy, including the incidence of secondary cancers should be clarified. Regardless of the radiation dose used (reduced or full), radiation therapy to the ventricular system should be given by intensity-modulated radiation therapy (IMRT) [36]. Figure 3 shows dose distribution of an IMRT plan for whole ventricular irradiation, as compared with that of parallel-opposing and 4-field irradiation.
126
Shibamoto
a
b
c
Fig. 3. Dose distribution for whole ventricular irradiation by parallel-opposing beams (a), 4-field box method (b) and IMRT (c).
Fig. 4. Dose distribution for craniospinal irradiation by helical tomotherapy.
By using the IMRT technique, radiation doses to the normal brain including the hippocampus can be reduced. Boost to the primary tumor site can also be done by stereotactic irradiation. In using this technique, hypofractionated radiation schedules rather than single fraction are recommended from the standpoint of radiobiology to expect reoxygenation of hypoxic tumor cells during fractionation. When employing cerebrospinal axis irradiation, tomotherapy seems to be the best choice [37]. Figure 4 shows dose distribution for craniospinal irradiation by tomotherapy. Uniform doses are delivered with no gaps that are unavoidable with conventional craniospinal irradiation using a linear accelerator. Unnecessary doses to thoracic and abdominal organs are markedly reduced. Booster doses can also be given by tomotherapy. In Japan, many patients are sent to tomotherapy sites to receive this treatment. In conclusion, the standard treatment for CNS germinoma should be determined by evaluating the results of systemic chemotherapy and reduced-dose
Management of CNS Germinoma
127
radiation after a long enough follow-up. If the results are comparable to that obtained by conventional full-dose radiation therapy, then systemic chemotherapy with IMRT to the ventricular system and simultaneous boost to the primary tumor site may be regarded as the best treatment. If the results are not good, IMRT to the ventricular system and boost to the primary site (40–45 Gy) may then be further investigated without the use of chemotherapy.
References 1 Keene D, Johnston D, Strother D, Fryer C, Carret AS, Crooks B, Eisenstat D, Moghrabi A, Wilson B, Brossard J, Mpofu C, Odame I, Zelcer S, Silva M, Samson Y, Hand J, Bouffet E: Epidemiological survey of central nervous system germ cell tumors in Canadian children. J Neurooncol 2007; 82:289–295. 2 The Committee of Brain Tumor Registry of Japan: Report of Brain Tumor Registry of Japan (1969– 1996), 11th Edition. Neurol Med Chir 2003; 43(suppl):5. 3 Matsutani M: Clinical management of primary central nervous system germ cell tumors. Semin Oncol 2004;31:676–683. 4 Ogino H, Shibamoto Y, Takanaka T, Suzuki K, Ishihara S, Yamada T, Sugie C, Nomoto Y, Mimura M: CNS germinoma with elevated serum human chorionic godnadotropin level: clinical characteristics and treatment outcome. Int J Radiat Oncol Biol Phys 2005;62:803–808. 5 Fujimaki T, Matsutani M, Funada N, Kirino T, Takakura K, Nakamura O, Tamura A, Sano K: CT and MRI features of intracranial germ cell tumors. J Neurooncol 1994;19:217–226. 6 Katakami H, Hashida S, Yamaguchi H, Yazawa S, Nakano S, Wakisaka S, Matsutani M: Diagnosis and follow-up of CNS germ cell tumors using an ultrasensitive assay of HCG-beta (in Japanese). Horm Clin 2003;51:196–206. 7 Suzuki N, Konishi E, Kondo K, Fujita I, Ishikawa E: A highly sensitive colorimetric sandwichenzyme immunoassay for human chorionic gonadotropin using specific antibodies against the carboxy-terminal portion of the beta-subunit. Clin Chim Acta 1985;150:247–253. 8 Ono N, Naganuma H, Inoue HK, Tamura M: Cerebrospinal fluid placental alkaline phosphatase in the intracranial germinomas: results of enzyme antigen immunoassay. Neurol Med Chir (Tokyo) 1991;31:563–567.
128
9 Koshida K, Uchibayashi T, Yamamoto H, Hirano K: Significance of placental alkaline phosphatase (PLAP) in the monitoring of patients with seminoma. Br J Urol 1996;77:138–142. 10 Sano K: Pinealoma in children. Childs Brain 1976;2:67–72. 11 Shibamoto Y, Oda Y, Yamashita J, Takahashi M, Kikuchi H, Abe M: The role of cerebrospinal fluid cytology in radiotherapy planning for intracranial germinoma. Int J Radiat Oncol Biol Phys 1994; 29:1089–1094. 12 Inoue Y, Takeuchi T, Tamaki M, Nin K, Hakuba A, Nishimura S: Sequential CT observations of irradiated intracranial germinomas. Am J Roentogenol 1979;132:361–365. 13 Utsuki S, Oka H, Tanizaki Y, Kondo K, Kawano N, Fujii K: Histological features of intracranial germinomas not disappearing immediately after radiotherapy. Neurol Med Chir (Tokyo) 2006;46: 429–433. 14 Wolden SL, Wara WM, Larson DA, Prados MD, Edwards MS, Sneed PK: Radiation therapy for primary intracranial germ-cell tumors. Int J Radiat Oncol Biol Phys 1995;32:943–949. 15 Hardenbergh PH, Golden J, Billet A, Scott RM, Shrieve DC, Silver B, Loeffler JS, Tarbell NJ: Intracranial germinoma: the case for lower dose radiation therapy. Int J Radiat Oncol Biol Phys 1997;39:419–426. 16 Aoyama H, Shirato H, Kakuto Y, Inakoshi H, Nishio M, Yoshida H, Hareyama M, Yanagisawa T, Watarai J, Miyasaka K: Pathologically-proven intracranial germinoma treated with radiation therapy. Radiother Oncol 1998;47:201–205. 17 Ogawa K, Shikama N, Toita T, Nakamura K, Uno T, Onishi H, Itami J, Kakinohana Y, Kinjo T, Yoshii Y, Ito H, Murayama S: Long-term results of radiotherapy for intracranial germinoma: a multi-institutional retrospective review of 126 patients. Int J Radiat Oncol Biol Phys 2004; 58:705–713.
Shibamoto
18 Shibamoto Y, Abe M, Yamashita J, Takahashi M, Hiraoka M, Ono K, Tsutsui K: Treatment results of intracranial germinoma as a function of the irradiated volume. Int J Radiat Oncol Biol Phys 1988;15:285–290. 19 Shibamoto Y, Takahashi M, Abe M: Reduction of the radiation dose for intracranial germinoma: a prospective study. Br J Cancer 1994;70:984–989. 20 Rogers SJ, Mosleh-Shirazi MA, Saran FH: Radiotherapy of localized intracranial germinoma: time to sever historical ties? Lancet Oncol 2005;6:509– 519. 21 Shibamoto Y, Sasai K, Oya N, Hiraoka M: Intracranial germinoma: radiation therapy with tumor volume-based dose selection. Radiology 2001;218:452–456. 22 Bamberg M, Kortmann RD, Calaminus G, Becker G, Meisner C, Harms D, Göbel U: Radiation therapy for intracranial germinoma: results of the German cooperative prospective trials MAKEI 83/86/89. J Clin Oncol 1999;17:2585–2592. 23 Duffner PK, Cohen ME, Thomas PRM, Lansky SB: The long-term effects of cranial irradiation on the central nervous system. Cancer 1985;56:1841– 1846. 24 Roman DD, Sperduto PW: Neuropsychological effects of cranial radiation: current knowledge and future directions. Int J Radiat Oncol Biol Phys 1995;31:983–998. 25 Farng KT, Chang KP, Wong TT, Guo WY, Ho DM, Hu WL: Pediatric intracranial germinoma treated with chemotherapy alone. Zhonghua Yi Xue Za Zhi 1999;62:859–866. 26 Kumabe T, Kusaka Y, Joukura H, Ikeda H, Shirane R, Yoshimoto T: Recurrence of intracranial germinoma treated with chemotherapy only (in Japanese). No Shinkei Geka 2002;30:935–942. 27 Balmaceda C, Heller G, Rosenblum M, Diez B, Villablanca JG, Kellie S, Maher P, Vlamis V, Walker RW, Leibel S, Finlay JL: Chemotherapy without irradiation – a novel approach for newly diagnosed CNS germ cell tumors: results of an international cooperative trial. J Clin Oncol 1996;14:2908–2915.
28 Merchant TE, Davis BJ, Sheldon JM, Leibel SA: Radiation therapy for relapsed CNS germinoma after primary chemotherapy. J Clin Oncol 1998;16:204–209. 29 Shibamoto Y, Sasai K, Kokubo M, Hiraoka M: Salvage radiation therapy for intracranial germinoma recurring after primary chemotherapy. J Neurooncol 1999;44:181–185. 30 Allen JC, DaRosso RC, Donahue B, Nirenberg A: A phase II trial of preirradiation carboplatin in newly diagnosed germinoma of the central nervous system. Cancer 1994;74:940–944. 31 Bouffet E, Baranzelli MC, Patte C, Portas M, Edan C, Chastagner P, Mechinaud-Lacroix F, Kalifa C: Combined treatment modality for intracranial germinomas: results of a multicentre SFOP experience. Br J Cancer 1999;79:1199–1204. 32 Matsutani M, Japanese Pediatric Brain Tumor Study Group: Combined chemotherapy and radiation therapy for CNS germ cell tumors – the Japanese experience. J Neurooncol 2001;54:311– 316. 33 Uematsu Y, Tsuura Y, Miyamoto K, Itakura T, Hayashi S, Komai N: The recurrence of primary intracranial germinomas. Special reference to germinoma with STGC (syncytiotrophoblastic giant cell). J Neurooncol 1992;13:247–256. 34 Utsuki S, Kawano N, Oka H, Tanaka T, Suwa T, Fujii K: Cerebral germinoma with syncytiotrophoblastic giant cells: feasibility of predicting prognosis using the serum hCG level. Acta Neurochir (Wien) 1999;141:975–977. 35 Shibamoto Y, Takahashi M, Sasai K: Prognosis of intracranial germinoma with syncytiotrophoblastic giant cells treated by radiation therapy. Int J Radiat Oncol Biol Phys 1997;37:505–510. 36 Raggi E, Mosleh-Shirazi MA, Saran FH: An evaluation of conformal and intensity-modulated radiotherapy in whole ventricular radiotherapy for localised primary intracranial germinomas. Clin Oncol (R Coll Radiol) 2008;20:253–260. 37 Bauman G, Yartsev S, Coad T, Fisher B, Kron T: Helical tomotherapy for craniospinal radiation. Br J Radiol 2005;78:548–552.
Prof. Yuta Shibamoto, MD Department of Radiology, Nagoya City University Graduate School of Medical Sciences Nagoya 467-8601 (Japan) Tel. +81 52 853 8274, Fax +81 52 852 5244, E-Mail
[email protected]
Management of CNS Germinoma
129
Tumors of Germ Cell Origin Kobayashi T, Lunsford LD (eds): Pineal Region Tumors. Diagnosis and Treatment Options. Prog Neurol Surg. Basel, Karger, 2009, vol 23, pp 130–139
Quality of Life of Extremely Long-Time Germinoma Survivors Mainly Treated with Radiotherapy Kazuhiko Sugiyamaa ⭈ Fumiyuki Yamasakia ⭈ Kaoru Kurisua ⭈ Masahiro Kenjob Departments of aNeurosurgery and bRadiation Oncology, Hiroshima University Hospital, Hiroshima, Japan
Abstract Purpose: To assess the quality of life (QOL) of extremely long-time survivors with germinoma mainly treated with radiotherapy. Patients and Methods: We enrolled 52 of 68 patients who received radiotherapy between 1968 and 1995 at our hospital. They were 41 males and 11 females; the tumor location was pineal in 20, neurohypophyseal in 15, pineal and neurohypophyseal in 11 patients; in 6 it was located in another region. All underwent radiotherapy; the median dose was 48.2 (range 40.0–60.2) Gy. The median follow-up period was 226 (range 0–448) months. The clinical outcome and QOL were evaluated retrospectively. Results: In 6 patients, the tumor recurred; 6 other patients developed second tumors while in complete remission from the first tumor. The main cause of 12 deaths was complications due to primary tumor invasion, the initial treatment, or tumor recurrence rather than tumor progression. The 10-, 20-, and 30-year actuarial survival rate was 83.6, 77.5, and 64.2%, respectively. Of 44 patients, 6 were married and 3 males with solitary pineal tumors were fathers. Among 32 patients, 14 had, or had not, graduated from high school; the other 18 went on to higher education. Twenty-one patients had no occupation; 7 of 11 formerly employed patients had left their jobs. Conclusion: Radiotherapy delivered between 1968 and 1995 to patients with germinoma yielded satisfactory outcomes Copyright © 2009 S. Karger AG, Basel but a decline in the QOL.
The introduction of more effective treatments has improved the rate and length of survival of children with cancer [1, 2]. Since the 1960s, patients with germinoma, one of the primary brain tumors that frequently affect children, adolescents and young adults, have been given a good prognosis because radiotherapy led to the disappearance of the tumor [3–5]. Although various modalities, including chemotherapy, were developed in the mid-1990s to avoid the late adverse effects of radiotherapy [6, 7], little is known regarding the quality of life (QOL) of extremely
long-time germinoma survivors primarily treated with radiotherapy. Therefore, we retrospectively analyzed their clinicopathological status and identified social problems encountered by these individuals.
Patients and Methods Between 1968 and 1995, 68 germinoma patients underwent radiotherapy at Hiroshima University Hospital. By March 2006, we had lost contact with 16 patients primarily due to their leaving the Hiroshima area. Of the remaining 52 patients, 40 visited our hospital at least once a year after treatment; 12 died at our or affiliated hospitals between 1972 and 2006. After acquiring prior informed consent from the patient or legal representative, we retrospectively investigated the posttreatment course of the 52 patients by evaluating clinical information contained in their medical records and reviewing imaging studies. Event-free and overall survival rates were calculated with the Kaplan-Meier method and measured from the date of pathological diagnosis or the start of radiotherapy [8]. We also surveyed these patients or their caregivers regarding their academic and occupational careers, and their marital and social status. Patient characteristics are listed in table 1. Until the introduction in 1976 of computed tomography (CT) at our hospital, pneumoencephalography followed by test irradiation with 20 Gy was the primary tool to diagnose germinoma. The introduction of magnetic resonance imaging (MRI) in 1988 facilitated the clear localization and assessment of the characteristics of the tumors. The most frequent site of the tumors in the 52 patients included in this study (41 males and 11 females) was the pineal or neurohypophyseal region. In 11 patients, these tumors developed synchronously in the pineal- and neurohypophyseal region. Irradiation therapy was delivered 5 times/week using 60Co teletherapy units (8 cases) or 10 MV X-rays generated by a linear accelerator (44 cases); the typical daily dose to the primary brain tumor was 1.8 or 2.0 Gy. Both the treatment volume and radiation dose changed over time. Whole brain irradiation using the cone down method was performed almost routinely until 1991; the median local dose to the tumor lesion was 50.2 Gy. Whole spine irradiation with a median dose of 25.0 (20.0–30.0) Gy was additionally delivered to 7 patients for prophylaxis. Starting in 1992, germinoma patients received 40-Gy irradiation to the whole ventricle field with the administration, 4 times/week, of CBDCA (100 mg/m2/body surface) [9]. By March 2006, the follow-up period ranged from 0 to 448 months (median 226 months; table 1).
Results
In the course of 12–108 months after the initial therapy, 6 patients suffered tumor recurrence in the primary tumor region (n = 2), the spinal cord (n = 2), and the optic nerve (n = 1); in one patient, the entire wall of the ventricle was involved diffusely in tumor recurrence. Three patients underwent a second course of radiotherapy, one patient received chemotherapy, and 2 underwent concurrent chemoand radiotherapy. All recurrent lesions disappeared after these salvage treatments (table 2).
QOL of Extremely Long-Time Germinoma Survivors
131
Table 1. Patient characteristics Sex Male Female
41 11
Age at diagnosis, years Range Median
7–32 14.2
Diagnostic imaging PEG CT MR
7 22 23
Diagnosis Pathological Response to radiation
41 11
Location Pineal Neurohypophyseal Pineal + neurohypophyseal Basal ganglia Other
20 15 11 3 3
Treatment Radiation only (46.2–60.2 Gy/median 50.2 Gy) Radiation (40.0 Gy) + 4 times weekly CBDCA (100 mg/m2) Radiation apparatus 60 Co Linac
8 44
Follow-up periods, months Range Median
0–448 226
42 10
PEG = Pneumoencephalography.
Second tumors developed more than 84 months after the initial treatment in 6 patients without recurrence of the primary tumor. Four of these patients had received alternating doses of irradiation from both sides; the four-field technique was not used. The pathological diagnosis of the 6 second tumors was high-grade glioma (n = 4), atypical meningioma (n = 1), and cavernous angioma (n = 1); all developed within the irradiated field (table 2; fig. 1 and 2).
132
Sugiyama · Yamasaki · Kurisu · Kenjo
Table 2. Neuro-oncological follow-up Recurrence
61
Clinical course after recurrence Death with recurring tumor Suicide* Death with deterioration of QOL** Alive with schizophrenia
2 1 2 1
Death
12
Causes of death Uncontrolled recurring tumor Adrenal failure without tumor Second tumor Suicide* Syringobulbia Gradual deterioration after caregiver’s death**
2 3 3 1 1 2
Patients that withdrew from their family and social circle
2
Second tumor after treatment
62
Single and double asterisks indicate same cases. 1 Recurrence 108, 85, 96, 12, 22, 101 months after treatment. 2 Recurrence 182, 84, 456, 206, 181, 359 months after treatment (2 glioblastomas multiforme, 1 anaplastic astrocytoma, 1 anaplastic oligoastrocytoma, 1 atypical meningioma, 1 cavernous angioma).
Twelve patients died during the follow-up period; 2 of uncontrolled recurrent tumor, and the remaining 10 died despite complete remission of their primary or recurrent germinoma. The 10 deaths were attributable to adrenal failure (n = 3), glioma as a second tumor (n = 3), gradual deteriorations after the caregiver’s death (n = 2), suicide (n = 1), and syringobulbia (n = 1) [10]. Adrenal failure was characterized by slight fever lasting for a few days followed by sudden-onset cardiac arrest. Two patients completely lost touch with their family more than 20 years after treatment (table 2). Figure 3 shows the cause-specific event-free and actuarial event-free survival rates. These rates appear discordant because 4 patients died of adrenal failure or syringobulbia, and 9 experienced serious events including the development of second tumors, or were lost from follow-up without recurrence. Figure 4 shows cause-specific and actuarial survival rates. The actuarial survival rate at 10-, 20-, and 30 years was 83.6, 77.5, and 64.2%, respectively. The 2
QOL of Extremely Long-Time Germinoma Survivors
133
a
b
c
d
Fig. 1. A 20-year-old man. a T1-weighted gadolinium-enhanced MRI showing 2 lesions in the pineal and neurohypophyseal region. The histological diagnosis was germinoma. b Radiation field at the two-field technique. The tumor disappeared completely after 54.0-Gy irradiation. c T1-weighted gadolinium-enhanced MRI revealed a cerebellar tumor that developed 84 months after treatment. d Pathological study of surgical materials indicated anaplastic astrocytoma (HE stain).
graphs are dissimilar due to death by suicide, development of glioma as a second tumor, and death of the caregiver. The marital status was examined in 44 of the 52 patients; of the other patients, one died just after radiotherapy, 5 also died at the age of younger than 18 years, and 2 were already married at the time of undergoing the initial treatment. Six patients got married during the follow-up period; however, 3 divorced and the other 3 patients with solitary pineal tumors were fathers (table 3). We evaluated the academic and occupational careers of 32 patients; they had undergone radiotherapy at an age younger than 16 years and survived until the age
134
Sugiyama · Yamasaki · Kurisu · Kenjo
a
b
c
Fig. 2. A 14-year-old boy. a, b At 206 months after 51.6-Gy irradiation of a pineal tumor,T1weighted gadolinium-enhanced MRI showed a mass attached to the posterotemporal dura. c The histological diagnosis was atypical meningioma (HE stain).
of 18 or older. Of these, 3 graduated from junior and 3 from senior high school, 8 graduated from a special needs senior high school, 10 from community college, and 8 were university graduates. Of these patients, 21 had no occupation and 7 of the remaining 11 patients left their jobs at an age older than 30 years because of recent memory disturbance or dyscalculia (table 3).
QOL of Extremely Long-Time Germinoma Survivors
135
1.0
Cause specific
Rate
0.8 0.6 Actuarial
0.4 0.2 0 0
100
200
300
400
500
Time (months)
Fig. 3. Cause-specific event-free and actuarial event-free survival rates.
Cause specific
1.0
Rate
0.8 0.6 0.4
Actuarial
0.2 0 0
100
200
300
400
500
Time (months)
Fig. 4. Cause-specific and actuarial survival rates.
Representative Case This 21-year-old man (fig. 5) presented with headache and upgaze palsy in 1985. CT showed a calcified tumor in the pineal region with obstructive hydrocephalus. A biopsy specimen of the tumor yielded a histological diagnosis of germinoma. The tumor disappeared completely after 52-Gy radiotherapy. After graduating from the university in 1987, he was licensed as an architect by the Japanese National Board. In 1995, at the age of 31 years, he experienced frequent episodes of recent memory loss and dyscalculia and was forced to resign at the age of 36. He currently lives with his parents who help him with his daily activities.
136
Sugiyama · Yamasaki · Kurisu · Kenjo
Table 3. Marital and career status Marital status (n = 44) Married
6 (3 with pineal, 3 neurohypophyseal tumors) 3 3 (all with solitary pineal tumors)
Divorced Fathers Academic career and occupation (n = 32) Graduated from junior high school Graduated from senior high school Graduated from a special-needs senior high school Graduated from community college Graduated from university, or more Left their occupation at the age of over 30 years
a
3 3 8 10 8 7
b
c
d
Fig. 5. Representative case. a CT performed on admission showed a pineal tumor. b After irradiation, CT revealed disappearance of the tumor. c Pathological examination of surgical materials revealed germinoma (HE stain). d MRI obtained in 2006 when the patient was 42 years old. There was brain atrophy with no tumor recurrence.
QOL of Extremely Long-Time Germinoma Survivors
137
Discussion
Germinomas are radiosensitive radiocurable tumors; they account for 0.5–2.5% of all primary brain tumors [3, 5, 7]. They frequently affect children, adolescents, and young adults, and they develop primarily in the pineal or neurohypophyseal region, or synchronously in both regions [5, 7, 10–12, 17]. Some earlier reports addressed the QOL of patients treated with irradiation; the median follow-up duration in most of these reports was less than 15 years [7, 11, 12, 13]. Ours is the first analysis of the QOL of patients surviving longer than 15 years. Our follow-up study (median period 224 months) of extremely long-time survivors showed that the primary cause of death or severe posttreatment events rendering the QOL unsatisfactory was not germinoma progression but various kinds of complication. Our findings alert to the need for monitoring the development of second tumors by MRI, for certain replacement of oral adrenocorticosteroid therapy, and for maintaining a lifestyle that protects against the occurrence of adrenal failure or suicide in patients who are alive more than 20 years after undergoing irradiation therapy. Our finding that 6 patients developed a second tumor after radiotherapy to treat their germinoma is important. Of the 4 patients whose second tumors were high-grade gliomas, 3 with glioblastoma died and the other patient with an anaplastic oligodendroglioma required additional treatment to address the recurrent tumor [14]. All 4 neoplasms arose in an area of the brain parenchyma that had been exposed to alternating doses of irradiation from both sides [14]. Current radiotherapeutic techniques, e.g. three-dimensional irradiation and intensitymodulated irradiation therapy minimize the extent of brain injury induced by irradiation. Most of the survivors in our study population remained unmarried and did not have a regular job. Others resigned at the age of approximately 30 years due to sequelae attributable to their initial treatment. These individuals remain financially unstable and experience difficulties finding caregivers after the death of their parents; their gradual deterioration after the related caregiver’s death was a main cause of death. Our findings indicate that long-term cancer survivors, including germinoma survivors, need a supportive social system that includes follow-up and long-term care. In the past 10 years, regimens have been introduced that involve volume- and field-reduced radiotherapy with chemotherapy to address germinoma [6, 7]. Controversy surrounds the issue of whether these treatments are superior to irradiation from the point of view of tumor control. Studies are underway to evaluate how these therapies lessen the incidence and degree of pituitary impairment [15] or intelligence decline [16], and how they protect against the occurrence of second tumors [14], suicide [11], or withdrawal of these patients from their family and social circle.
138
Sugiyama · Yamasaki · Kurisu · Kenjo
Acknowledgements This study was supported by the Foundation of Children’s Association of Japan.
References 1 Bhatia S, Landier W: Evaluating survivors of pediatric cancer. Cancer J 2005;11:340–354. 2 Curry HL, Parkes SE, Powell JE, Mann JR: Caring for survivors of childhood cancers: the size of the problem. Eur J Cancer 2006;42:501–508. 3 Jennings MT, Gelman R, Hochberg F: Intracranial germ-cell tumors: natural history and pathogenesis. J Neurosurg 1985;63:155–167. 4 Kageyama N: Ectopic pinealoma in the region of the optic chiasm. Report of five cases. J Neurosurg 1971;35:755–759. 5 Matsutani M, Sano K, Takakura K, Fujimaki T, Nakamura O, Funata N, Seto T: Primary intracranial germ cell tumors: a clinical analysis of 153 histologically verified cases. J Neurosurg 1997;86: 446–455. 6 Buckner JC, Peethambaram PP, Smithson WA, Groover RV, Schomberg PJ, Kimmel DW, Raffel C, O’Fallon JR, Neglia J, Shaw EG: Phase II trial of primary chemotherapy followed by reduced-dose radiation for CNS germ cell tumors. J Clin Oncol 1999;17:933–940. 7 Sawamura Y, Shirato H, Ikeda J, Tada M, Ishii N, Kato T, Abe H, Fujieda K: Induction chemotherapy followed by reduced-volume radiation therapy for newly diagnosed central nervous system germinoma. J Neurosurg 1998;88:66–72. 8 Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958;53:457–481. 9 Allen JC, Walker R, Luks E, Jennings M, Barfoot S, Tan C: Carboplatin and recurrent childhood brain tumors. J Clin Oncol 1987;5:459–463. 10 Sugiyama K, Uozumi T, Arita K, Kiya K, Kurisu K, Sumida M, Harada K: Clinical evaluation of 33 patients with histologically verified germinoma. Surg Neurol 1994;42:200–210.
11 Bamberg M, Kortmann RD, Calaminus G, Becker G, Meisner C, Harms D, Göbel U: Radiation therapy for intracranial germinoma: results of the German cooperative prospective trials MAKEI 83/86/89. J Clin Oncol 1999;17:2585–2592. 12 Ogawa K, Shikama N, Toita T, Nakamura K, Uno T, Onishi H, Itami J, Kakinohana Y, Kinjo T, Yoshii Y, Ito H, Murayama S: Long-term results of radiotherapy for intracranial germinoma: a multi-institutional retrospective review of 126 patients. Int J Radiat Oncol Biol Phys 2004;58: 705–713. 13 Shibamoto Y, Takahashi M, Sasai K: Prognosis of intracranial germinoma with syncytiotrophoblastic giant cells treated by radiation therapy. Int J Radiat Oncol Biol Phys 1997;37:505–510. 14 Doskaliyev A, Yamasaki F, Kenjo M, Shrestha P, Saito T, Hanaya R, Sugiyama K, Kurisu K: Secondary anaplastic oligodendroglioma after cranial irradiation: a case report. J Neurooncol 2008; 88:299–303. 15 Aida T, Abe H, Fujieda K, Matsuura N: Endocrine functions in children with suprasellar germinoma. Neurol Med Chir (Tokyo) 1993;33:152– 157. 16 Merchant TE, Sherwood SH, Mulhern RK, Rose SR, Thompson SJ, Sanford RA, Kun LE: CNS germinoma: disease control and long-term functional outcome for 12 children treated with craniospinal irradiation. Int J Radiat Oncol Biol Phys 2000;46:1171–1176. 17 Sawamura Y, Ikeda J, Shirato H, Tada M, Abe H: Germ cell tumours of the central nervous system: treatment consideration based on 111 cases and their long-term clinical outcomes. Eur J Cancer 1998;34:104–110.
Kazuhiko Sugiyama, MD, DMSci Department of Neurosurgery, Hiroshima University Hospital 1-2-3 Kasumi, Minami-ku Hiroshima 734-8551 (Japan) Tel. +81 82 257 5227, Fax +81 82 257 5229, E-Mail
[email protected]
QOL of Extremely Long-Time Germinoma Survivors
139
Author Index
Aoyama, H. 96 Flickinger, J.C. 44 Fujii, Y. 26 Hasegawa, T. 106 Kano, H. 44 Kenjo, M. 130 Kida, Y. 106 Kobayashi, T. IX, 106 Kondziolka, D. 44 Kubota, T. 12, 59 Kurisu, K. 130 Lunsford, L.D. VII, 44 Matsutani, M. 76 Mori, Y. 106
140
Niranjan , A. 44 Nomura, K. 1 Sato, K. 12, 59 Sawamura, Y. 86 Shibamoto, Y. 119 Shibui, S. 1 Sugiyama, K. 130 Takeuchi, H. 59 Tanaka, R. 26 Tsumanuma, I. 26 Yamasaki, K. 130 Yoshida, K. 106
Subject Index
Age distribution pineal parenchymal tumors 14 pineal tumors in Brain Tumor Registry of Japan 4–9 Brain Tumor Registry of Japan (BTRJ) foundation 1 registration 2 statistical analysis neuroepithelial tumor frequencies 2–4 pineal tumors age distribution 4–9 frequency by histology 3, 10 sex distribution 5–9 survival rates 9–11 primary brain tumor frequencies 2 Chemotherapy combination radiation therapy for germ cell tumors 81, 82, 103, 125, 126 germ cell tumors 81, 82 Choriocarcinoma, see Germ cell tumors c-kit, mutation in germ cell tumors 73 Computed tomography (CT) germinoma diagnosis 120 pineal parenchymal tumors 14 Dose reduction, chemotherapy and reduceddose radiation for germinomas 125, 126 Electron microscopy embryonal carcinoma 71 germinoma 65, 66, 68 pineoblastoma 20, 21 pineocytoma 17, 18
teratoma 69 Embryonal carcinoma, see Germ cell tumors α-Fetoprotein (AFP), germ cell tumor marker 79, 80, 93 Gamma Knife, see Stereotactic radiosurgery Germ cell tumors (GCTs) brain distribution 78 choriocarcinoma pathology immunohistochemistry 72 macroscopy 72 microscopy 72 classification 59, 60, 77, 86, 87 clinical features of pineal tumors 78 embryonal carcinoma pathology electron microscopy 71 immunohistochemistry 71 microscopy 70, 71 endoscopic findings 79 genetics 72, 73 germinoma diagnosis 120, 121 pathology electron microscopy 65, 66, 68 immunohistochemistry 64 macroscopy 63 microscopy 63, 64 histogenesis 59, 61 imaging 61, 62, 78, 79 incidence 60 markers 79, 80, 88–90, 93 quality of life in germinoma survivors after radiation therapy case example 136
141
Germ cell tumors (GCTs) (continued) marital status 134, 137 neuro-oncological follow-up 131–133 occupation 134, 135, 137 patient characteristics 132 study design 131 survival rates 133 treatment-related tumors 138 teratoma pathology immunohistochemistry 67 macroscopy 66 microscopy 67 treatment chemotherapy 81, 82 combination chemotherapy 81, 82, 103, 125, 126 Gamma Knife, see Stereotactic radiosurgery germinoma 88–90, 122–128 malignant tumors 93, 94, 115–117 nongerminomatous tumors 82–84 radiation therapy germinoma 97–101, 122–125 intensity-modulated radiation therapy 126–128 late sequelae 104 nongerminomatous tumors 101–103 overview 80, 81 teratoma immature teratoma 91–93 mature teratoma 91 surgical resection 87, 88 yolk sac tumor pathology electron microscopy 69 immunohistochemistry 69, 70 macroscopy 68 microscopy 68, 69 Germinoma, see Germ cell tumors Human chorionic gonadotropin (HCG), germ cell tumor marker 79, 80, 88–90, 93, 121 Immunohistochemistry choriocarcinoma pathology 72 embryonal carcinoma pathology 71
142
germinoma pathology 64 pineal parenchymal tumor of intermediate differentiation 22 pineoblastoma 19 pineocytoma 16, 17 teratoma pathology 67 yolk sac tumor pathology 69, 70 Infratentororial supracerebellar approach, pineal parenchymal tumors 27 INI1, mutation in pineal parenchymal tumors 23 Intensity-modulated radiation therapy, see Radiation therapy Klinefelter’s syndrome, germ cell tumors 73 Magnetic resonance imaging (MRI) germinoma diagnosis 120 mixed germ cell tumor 90 pineal germ cell tumors germinoma 61 overview 78, 79 yolk sac tumor 62 pineal parenchymal tumors overview 14, 15 stereotactic radiosurgery planning 45, 46 suprasellar teratoma 92 Metastasis, pineal parenchymal tumor following stereotactic radiosurgery 50–52, 56 Nongerminomatous tumors, see Germ cell tumors Occipital transtentorial approach (OTA), pineal parenchymal tumors adjuvant therapy 34, 35 clinicopathological study histopathology 37, 38 neuronal characteristics 41, 42 patient characteristics 36 proliferative potential 39–41 tumor removal extent and outcomes 37, 39 closure 33, 34 complications 34
Subject Index
craniotomy and tentorial incision 31 indications and role 27–29 internal cerebral vein exposure 32 overview 27 patient positioning 30, 31 preoperative management care 30 diagnosis 29, 30 prognosis 35, 36 tumor exposure 32 tumor removal 32, 33 p14, mutation in germ cell tumors 73 Parenchymal tumors classification and grading 14, 26, 27 epidemiology 14 genetics 22, 23 imaging 14, 15 infratentororial supracerebellar approach 27 occipital transtentorial approach adjuvant therapy 34, 35 clinicopathological study histopathology 37, 38 neuronal characteristics 41, 42 patient characteristics 36 proliferative potential 39–41 tumor removal extent and outcomes 37, 39 closure 33, 34 complications 34 craniotomy and tentorial incision 31 indications and role 27–29 internal cerebral vein exposure 32 overview 27 patient positioning 30, 31 preoperative management care 30 diagnosis 29, 30 prognosis 35, 36 tumor exposure 32 tumor removal 32, 33 pineal parenchymal tumor of intermediate differentiation pathology immunohistochemistry 22 macroscopy 21 microscopy 21, 22 pineoblastoma pathology
Subject Index
electron microscopy 20, 21 immunohistochemistry 19 macroscopy 18 microscopy 18, 19 pineocytoma pathology electron microscopy 17, 18 immunohistochemistry 16, 17 macroscopy 15 microscopy 16 stereotactic radiosurgery study adverse radiation effects 53, 54, 56, 57 metastasis 50–52, 56 patients 45 statistical analysis 48 survival rates 47–49, 54 symptom improvement 52, 53, 56 technique 45–47 treatment modalities fractionated radiation therapy 55 stereotactic radiosurgery 55, 56 surgical resection and stereotactic biopsy 54, 55 tumor grade and survival 48, 54 Pineal gland, histology 12, 13 Pineal tumors, see Germ cell tumors; Parenchymal tumors Pineoblastoma, see Parenchymal tumors Pineocytoma, see Parenchymal tumors Placental alkaline phosphatase (PLAP), germ cell tumor marker 79, 80, 93 Quality of life (QOL), germinoma survivors after radiation therapy case example 136 marital status 134, 137 neuro-oncological follow-up 131–133 occupation 134, 135, 137 patient characteristics 132 study design 131 survival rates 133 treatment-related tumors 138 Radiation therapy, see also Stereotactic radiosurgery fractionated radiation therapy for pineal parenchymal tumor 55
143
Radiation therapy (continued) germ cell tumors combination chemotherapy 81, 82, 103, 125, 126 germinoma 97–101, 122–125 intensity-modulated radiation therapy 126–128 late sequelae 104 nongerminomatous tumors 101–103 overview 80, 81 quality of life in germinoma survivors after radiation therapy case example 136 marital status 134, 137 neuro-oncological follow-up 131–133 occupation 134, 135, 137 patient characteristics 132 study design 131 survival rates 133 treatment-related tumors 138 RB1, mutation in pineal parenchymal tumors 23 Sex distribution pineal parenchymal tumors 14 pineal tumors in Brain Tumor Registry of Japan 5–9 Statistical analysis, see Brain Tumor Registry of Japan Stereotactic radiosurgery (SRS) Gamma Knife radiosurgery for pineal tumors case characterization 107, 108, 110
144
case studies 111–114 dose planning 108, 109 evaluation methodology 109 local tumor control rates 109, 110 malignant germ cell tumors 115–117 parenchymal tumors 116 side effects 110 survival rates 115, 116 pineal parenchymal tumor study adverse radiation effects 53, 54, 56, 57 metastasis 50–52, 56 patients 45 statistical analysis 48 survival rates 47–49, 54 symptom improvement 52, 53, 56 technique 45–47 treatment modalities fractionated radiation therapy 55 stereotactic radiosurgery 55, 56 surgical resection and stereotactic biopsy 54, 55 tumor grade and survival 48, 54 Survival rates germinoma patients Gamma Knife radiosurgery 115, 116 radiation therapy 133 pineal germinomas 80 pineal parenchymal tumor following stereotactic radiosurgery 47–49, 54 pineal tumor overview 9–11 Teratoma, see Germ cell tumors
Subject Index