Oxaliplatin (eloxatin): Fortschritte In Der Tumortherapie 2004 (Onkologie)

Dokumentation und Evaluation der Weiterbildung Originalarbeit · Original Article

Supplement 1 zu Band 27, August 2004

Oxaliplatin (Eloxatin®) Fortschritte in der Tumortherapie 2004

Herausgeber H.-J. Schmoll, Halle/Saale

Basel · Freiburg · Paris · London · New York · Bangalore · Bangkok · Singapore · Tokyo · Sydney

Contents · InhaltDokarbeit · Original Article Vol. 27, Supplement 1, August 2004

1

Vorwort Schmoll, H.-J. (Halle/Saale)

2

Einleitung Oxaliplatin beim Kolorektalkarzinom: Von der Palliation zur Kuration

12

ASCO-Abstracts

12 13

16 16 17 33 37 40 44 44 46 50 51 52 55 56 57 58

Breast Cancer Developmental Therapeutics – Cytotoxic Chemotherapy Developmental Therapeutics – Molecular Therapeutics Gastrointestinal Cancer – Colorectal/Liver A. Adjuvant Colorectal Cancer B. Previously Untreated Advanced Colorectal Cancer C. Previously Treated Advanced Colorectal Cancer D. Hepatic Metastases of Colorectal Cancer E. Rectal Cancer Gastrointestinal Cancer – Non-Colorectal/Liver A. Pancreas Cancer B. Gastric and Esophageal Cancer C. Hepatocellular and Biliary Cancer Genitourinary Cancer Gynecologic Cancer Adult Leukemia, Lymphoma and Transplantation Lung Cancer Patient Care Tumor Biology / Immunobiology / Human Genetics

59

Autorenverzeichnis

II

Impressum

14

© 2004 S. Karger GmbH, Freiburg Fax +49 761 4 52 07 14 E-mail [email protected] www.karger.com

Accessible online at: www.karger.com/onk

VorwortDokumentation und Evaluation der Weit Onkologie 2004;27(suppl 1):1

Vorwort Auch der kritische und eher skeptische Beobachter wird feststellen, dass sich die klinische Tumorforschung – basierend auf den mit großer Dynamik voranschreitenden Erkenntnissen der molekularen Tumorbiologie – in einem seit vielen Jahren nicht mehr gekannten Aufschwung befindet: Davon zeugt die Vielfalt der neuen, auf spezifische molekulare Mechanismen der Tumorzellen zielenden Substanzen, die «small molecules», die Antikörper oder Kombinationsprodukte sowie die damit verbundene große Zahl klinischer Explorations- und randomisierter Studien, die in kurzer Zeit neue Erkenntnisse zur Wirksamkeit und zum klinischen Nutzen bringen. Wir erleben jetzt den Beginn der Ernte von intensiver Forschung in der Vergangenheit; zugleich ist auf Grund der technologischen Entwicklungen in Zukunft mit einer noch schnelleren Entwicklung zu rechnen. Wer jedoch geglaubt hatte, neue molekulare Therapeutika würden die klassischen Zytostatika überflüssig machen, hat sich getäuscht. Vielmehr verhelfen die Substanzen, die auf neue Targets zielen, nicht nur dem «targeted treatment» zu neuer Blüte (schon Methotrexat und 5-FU sind klassische «targeted drugs» mit Dihydrofolatreduktase bzw. Thymidilatsynthase als Target), da sie vielfach mit den Zytostatika synergistisch wirken und Resistenzen vermindern. Glücklicherweise ist das Nebenwirkungsspektrum dieser neuen Substanzen ein anderes als das der klassischen Zytostatika, so dass sich der Kombination dieser beiden «Substanzgruppen» große Möglichkeiten eröffnen. Für die Entwicklung dieser neuen Therapieoptionen und -kombinationen bei den verschiedenen Tumoren ist der Einschluss vieler Patienten in klinische Studien nötig und damit ganz besonders die Aktivität von kooperativen Studiengruppen gefragt. Nur eine Gruppe, die in kurzer Zeit viele Patienten einbinden kann, wird an den neuen Entwicklungen in vollem Maß teilhaben und sie voranbringen. Dies zeigt sich auch in den in diesem Heft zusammengestellten Abstracts: Während ein großer Teil von Studien solche Kombinationen, die vor 1–2 Jahren noch neu waren, erneut oder in abgewandelter Form präsentiert, ohne dass hieraus eine wichtige, Therapiestandard-verändernde Erkenntnis zu gewinnen ist oder für den Patienten ein wesentlicher Nutzen entspringt, untersucht ein Drittel aller präsentierten Studien die Integration von Substanzen mit neuen Targets. Hierzu gehören die Inhibitoren des epidermalen Wachstumsfaktor-Rezeptor-Pathways, des Neo-

© 2004 S. Karger GmbH, Freiburg Fax +49 761 4 52 07 14 E-mail [email protected] www.karger.com

Accessible online at: www.karger.com/onk

angiogenese-Pathways, des Insulin like growth factor-Mechanismus, sowie den Inhibitoren weiterer Signaltransduktionswege wie z.B. raf-Kinase. Die Daten sind viel versprechend und lassen auf mögliche therapeutische Fortschritte bei einer ganzen Reihe zentraler Tumorarten hoffen. Dies gilt insbesondere für die kolorektalen und HNO-Tumoren, das Ösophaguskarzinom, eventuell das Magenkarzinom, aber auch für das Mamma- und Ovarialkarzinom. Oxaliplatin hat sich mittlerweile durchgesetzt als wichtiger Kombinationspartner für diverse First-line- und Salvage-Protokolle bei einer Vielzahl von Tumoren, vom kolorektalem Karzinom über das Magenkarzinom, Pankreaskarzinom, Gallenblasen-/Gallengangskarzinom, Mamma- und Ovarialkarzinom und vielen weiteren Tumoren bis hin zum hochmalignen B-Zell-Lymphom und der Hämoblastose. Wegen seiner verbesserten Verträglichkeit und leichteren Applizierbarkeit gegenüber Cisplatin ist Oxaliplatin immer häufiger Partner in Kombinationstherapien. Noch wichtiger hierfür ist bei einer Vielzahl von Tumorentitäten allerdings die ausgeprägte Synergie zwischen Oxaliplatin und Fluoropyrimidinen und anderen Substanzen, insbesondere den EGF-Inhibitoren. So verwundert es nicht, dass die auf neue molekulare Targets zielenden Substanzen, insbesondere die Kombinationen mit Oxaliplatin-haltigen Protokollen, geprüft werden. Das Beispiel des kolorektalen Karzinoms mit einer Remissionsrate von 81% bei der Kombination von FU/FA, Oxaliplatin plus Cetuximab oder Iressa zeigt, dass der Synergismus zwischen FU und Oxaliplatin durch EGF-Inhibitoren weiter verstärkt werden kann und damit ein großes Entwicklungspotenzial für Kombinationstherapien vorhanden ist. In einer weiteren, sehr wichtigen Arbeit konnte D. J. Sargent von der Mayo Clinic Rochester zeigen, dass DFS ein valider Endpunkt für die Durchführung adjuvanter Studien beim Kolonkarzinom ist (mittlerweile auch von der U.S. Food and Drug Administration anerkannt). Dadurch können nun neue und effektive Therapiekonzepte den Patienten schneller zur Verfügung gestellt werden. Die in diesem Heft zusammengestellte Auswahl an Arbeiten gibt einen aktuellen Überblick und ermöglicht eine schnelle Orientierung in diesem sich dynamisch entwickelnden Feld. H.-J. Schmoll, Halle

Einleitung Onkologie 2004;27(suppl 1):2–11

Oxaliplatin beim Kolorektalkarzinom: Von der Palliation zur Kuration

Mit jährlich fast 60 000 Neuerkrankungen in Deutschland ist das kolorektale Karzinom eine der häufigsten Neoplasien. Zwar kann bei etwa 70% der betroffenen Patienten eine potenziell kurative Tumorresektion durchgeführt werden. Bei etwa der Hälfte der Patienten rezidiviert die Erkrankung jedoch und führt letztlich zum Tod. Doch kann eine adjuvante Chemotherapie auf Basis von 5-Fluorouracil (5-FU) die Prognose dieser Patienten deutlich verbessern: Im Vergleich zur alleinigen Resektion wird eine Mortalitätsreduktion um rund ein Drittel erreicht. Seit 1990 gilt daher die 6-monatige Therapie mit 5-FU/Folinsäure (5-FU/FS) im Stadium UICC III als Standard [20]. Bislang noch kontrovers diskutiert wird die adjuvante Chemotherapie für Patienten im Stadium II, da der Überlebensvorteil hier nur moderat ist. Daher beschränkt man die adjuvante Therapie in diesem Stadium meist auf Patienten mit erhöhtem Rezidivrisiko.

FOLFOX4 auf dem Weg in die adjuvante Therapie Erst 2003 wurde auf der Jahrestagung der American Society of Clinical Oncology (ASCO) mit MOSAIC (Multicenter International Study of Oxaliplatin/5-FU/LV in the Adjuvant Treatment of Colon Cancer) eine adjuvante Therapiestudie vorgestellt, die die klinische Routine verändern wird: Die gerade im New England Journal of Medicine publizierte Studie [1] untersuchte die zusätzliche Gabe von Oxaliplatin (Eloxatin) zu einem 5-FU/FS-Infusionsregime (FOLFOX4) – ein Protokoll, das sich bereits in der First-line-Therapie beim metastasierten Kolorektalkarzinom aufgrund der Steigerung von Ansprechraten und Überleben bewährt hat. In der multizentrischen MOSAIC-Studie wurden 2246 Patienten mit Kolonkarzinomen der Stadien II (T3 oder T4, N0, M0) und III (jedes T, N1 oder N2, M0) nach R0-Resektion randomisiert entweder dem Protokoll nach de Gramont (200 mg FS über 2 Std. i.v., 400 mg/m2 5-FU als Bolus, 600 mg/m2 5-FU i.v. über

© 2004 S. Karger GmbH, Freiburg Fax +49 761 4 52 07 14 E-mail [email protected] www.karger.com

Accessible online at: www.karger.com/onk

22 Std., Tag 1 und 2) oder dem gleichen Schema plus Oxaliplatin (85 mg/m2 an Tag 1) zugeteilt [1]. Die Therapie wurde 14-tägig über insgesamt 12 Zyklen verabreicht. Im FOLFOX4-Arm erhielten 74,7%, im Kontrollarm 86,5% der Patienten alle geplanten Therapiezyklen. In beiden Gruppen konnten über 80% der geplanten Dosis appliziert werden. Die Patientencharakteristika beider Studienarme waren ausgewogen: Jeweils 60% der Teilnehmer hatten ein Kolonkarzinom im Stadium III, je 40% einen Tumor im Stadium II.

Zusätzliche Oxaliplatin-Gabe verhindert ein Viertel aller Rezidive Nach einem medianen Follow-up von knapp 38 Monaten waren im Kontrollarm 293 Rezidive und Todesfälle (26,1%) aufgetreten, im FOLFOX4-Arm dagegen nur 237 Ereignisse (21,1%). Daraus berechnet sich eine Hazard Ratio für Rezidive von 0,77 für den Oxaliplatin-haltigen Arm im Vergleich zur Kontrolle (p = 0,002). Anders ausgedrückt: In der MOSAICStudie konnte das Rezidivrisiko durch die zusätzliche Gabe von Oxaliplatin zur Standardtherapie mit 5-FU/FS um 23% verringert werden (Abb. 1). Die Rate des krankheitsfreien Überlebens als primärem Endpunkt wurde von 72,9% im Kontrollarm um absolut 5,3% auf 78,2% im FOLFOX4-Arm erhöht. Da die meisten Rezidive innerhalb der ersten 3 Jahre nach kurativer Therapie auftreten, gehen die Autoren jedoch davon aus, dass die Verbesserung des krankheitsfreien Überlebens auch in einer signifikanten Überlebensverlängerung resultieren wird.

Effektivitätssteigerung nicht auf Kosten der Sicherheit Die zusätzliche Gabe von Oxaliplatin zum 5-FU/FS-Regime wurde generell gut toleriert. Häufigste Nebenwirkungen vom

3 Jahre krankheitsfreies Überleben 1

FOLFOX 4 (n = 1123) 78,2 % LV5FU2 (n = 1123)

0,9 Wahrscheinlichkeit

Tab. 1. MOSAIC-Studie: Verbesserung des krankheitsfreien Überlebens nach 3 Jahren um absolut 5% bei Hochrisikopatienten im Stadium II [16]

72,9 %

0,8 0,7 0,6 Hazard Ratio: 0,77 (0,65–0,92)

p < 0,002

0 0

10

20

30

40

50

Parameter

FOLFOX4

FU/FS

Patientenzahl Zahl der Ereignisse Wahrscheinlichkeit des ereignisfreien Überlebens nach 3 Jahren Hazard Ratio Relative Risikoreduktion

286 40 (14,0%)

290 54 (18,6%)

84,9%

79,8%

DFS (Monate)

Abb. 1. MOSAIC-Studie: Signifikante Reduktion der Ereignisrate um 23% nach 3 Jahren durch zusätzliche Gabe von Oxaliplatin zur 5-FU/FSTherapie [1]. DFS = Krankheitsfreies Überleben. FOLFOX4-Arm: 23% Risikoreduktion gegenüber LV5FU2.

Tab. 2. MOSAIC-Studie: Schwere Sicherheitsereignisse waren bei Patienten im Stadium II seltener als bei Patienten im Stadium III [16] Stadium

Patienten, %

100 90

Grad 0: keine Veränderung, keine Symptome

80

Grad 1: leichte

70

Parästhesie, Verlust der tiefen Sehnenreflexe

60

Grad 2: leichter oder

50

mittelschwerer objektiver sensorischer Verlust, mittelschwere Parästhesie

40 30

objektiver sensorischer Verlust oder Parästhesie mit Funktionsbeeinträchtigung

10 0

Patientenzahl Patienten mit SAE Therapieabbruch wegen Toxizität

Tod unter Therapie

Grad 3: schwerer

20

Während 1 Monat der Therapie

0,72 28%

Schwere Sicherheitsereignisse

6 Monate 12 Monate 18 Monate

Follow-up

II

III

FOLFOX4 5-FU/FS

891 18,8% 13,5%

1328 25,4% 15,3%

FOLFOX4 5-FU/FS

12,1% 5,8%

16,0% 5,3%

FOLFOX4 5-FU/FS

0,2% 0,7%

0,8% 0,5%

FOLFOX4 5-FU/FS

24,2% 15,3%

33,7% 17,6%

SAE = Serious adverse events. Abb. 2. MOSAIC: Periphere sensorische Neuropathie.

Grad 3/4 waren im FOLFOX4-Arm Neutropenie, Erbrechen und Diarrhoe. Eine Neutropenie vom Grad 3/4 war bei zusätzlicher Oxaliplatin-Gabe zwar häufiger als im Kontrollarm (41,1 vs. 4,7%), die Rate daraus resultierender Komplikationen (Fieber, Infektionen) war jedoch mit beiden Regimen vergleichbar (1,8 vs. 0,2%). Häufigste Nebenwirkung unter FOLFOX4 war erwartungsgemäß die durch Oxaliplatin bedingte periphere Neuropathie 12,4% (Grad 3). Sie war jedoch in der Regel rasch reversibel (Abb. 2). Die therapiebedingte Mortalität in der MOSAIC-Studie war mit je 0,5% in beiden Gruppen niedrig.

Ein weiteres wichtiges Ergebnis der MOSAIC-Studie ist die Tatsache, dass die Teilnehmer unabhängig vom Tumorstadium von der adjuvanten Therapie mit FOLFOX4 profitierten. Das belegt eine auf der diesjährigen ASCO-Tagung vorgestellte Subgruppenanalyse der 899 Patienten im Stadium II [16]: Während sich die Reduktion des Rezidivrisikos bei Patienten im Stadium III auf 24% belief, fiel sie

bei Patienten im Stadium II mit 20% etwas geringer aus. Die Autoren weisen jedoch darauf hin, dass Kolonkarzinome im Stadium II eine heterogene Gruppe darstellen und daher neben dem Nodalstatus weitere Faktoren für die Risikostratifizierung herangezogen werden sollten. Identifizierte man Patienten im Stadium II mit mindestens einem weiteren Risikofaktor (z.B. T4, Obstruktion oder geringer Differenzierungsgrad), so konnte die Wahrscheinlichkeit für das krankheitsfreie Überleben nach 3 Jahren durch FOLFOX4 um absolut 5% im Vergleich zum Kontrollarm gesteigert werden (Tab. 1). Das entspricht einer relativen Risikoreduktion durch die zusätzliche Oxaliplatin-Gabe um 28%, die die für Patienten im Stadium III noch übertrifft. Auch im Stadium II sollte daher anhand des individuellen Rezidivrisikos durchaus eine adjuvante Therapie mit dem FOLFOX4-Regime in Erwägung gezogen werden, resümieren Hickish und Mitarbeiter [16]. Diese Empfehlung ist auch deswegen gerechtfertigt, weil schwere Beeinträchtigungen der Sicherheit im Studienverlauf (Inzidenz schwerwiegender Nebenwirkungen, Abbruch und/oder Tod) im Stadium II noch seltener registriert wurden als im Stadium III (19,8 vs. 25,6%; Tab. 2).

Einleitung

Onkologie 2004;27(suppl 1):2–11

Vorteil von FOLFOX4 auch im Stadium II

3

Das FOLFOX4-Regime ist damit auf dem besten Weg, sich als neuer Standard in der adjuvanten Chemotherapie des Kolonkarzinoms zu etablieren. Auch der Einwand, dass bislang keine langfristigen Gesamt-Überlebensdaten vorliegen, ist wenig stichhaltig, wie eine von D.J. Sargent, Mayo Clinic Rochester, vorgestellte Analyse großer Phase-III-Studien zur adjuvanten Therapie des Kolonkarzinoms belegt [23]. Zwar war die 5-Jahres-Überlebensrate bislang traditioneller Endpunkt in den großen adjuvanten Therapiestudien. Würde dieser Parameter jedoch zukünftig durch die Rate für das krankheitsfreie Überleben nach 3 Jahren ersetzt, so ließen sich therapeutische Fortschritte zum Wohle betroffener Patienten rascher in die klinische Praxis einführen. Voraussetzung ist allerdings, dass beide Parameter korrelieren, dass also die Rezidivfreiheit 3 Jahre nach der Operation prädiktiv für das weitere Überleben ist, erläuterte Sargent. Um diese Hypothese zu überprüfen, analysierte seine Arbeitsgruppe die individuellen Daten von mehr als 17 000 Patienten mit Kolorektalkarzinomen aus 17 adjuvanten Therapiestudien. Sie waren in insgesamt 39 Armen mit auf 5-FU basierenden Regimen behandelt worden oder gehörten dem Kontrollarm ohne Therapie an.

Verlässlicher Parameter: Krankheitsfreies Überleben Die ungeheure Informationsbasis macht deutlich, dass drei Viertel aller Rezidive beim Kolorektalkarzinom innerhalb der ersten 3 Jahre nach Tumorresektion auftreten. «Das Maximum der Rezidivrate ist bereits nach 1,5 Jahren erreicht. Danach sinkt die Wahrscheinlichkeit für eine Rezidivierung sehr rasch ab», berichtete Sargent. Und wenn Rezidive verhindert werden, verbessert sich auch die Heilungschance: Entsprechend konnte eine sehr enge Korrelation zwischen der Rate für das krankheitsfreie Überleben nach 3 Jahren und der 5-Jahres-Gesamt-Überlebensrate nachgewiesen werden, sodass Sargent den ersten Parameter als «exzellenten Prädiktor» für das Gesamt-Überleben qualifizierte. In 22 vergleichenden Analysen von krankheitsfreiem und Gesamt-Überleben ergab sich in 19 Fällen für beide Parameter eine identische Aussage. Lediglich bei 3 Analysen zeigte sich ein signifikanter Vorteil nur für das krankheitsfreie Überleben. Gerade in diesen nicht übereinstimmenden Fällen war der Benefit der adjuvanten Therapie jedoch nur von marginaler Signifikanz. «Ein gering ausgeprägter Nutzen beim krankheitsfreien Überleben resultiert demnach nicht zwangsläufig in einem verbesserten Gesamt-Überleben», kommentierte Sargent. Insgesamt erlaubt das von seiner Arbeitsgruppe entwickelte Modell anhand der Therapieeffekte auf das krankheitsfreie Überleben eine gute Vorhersage des Gesamt-Überlebens. Seiner Meinung nach ist diese Variable daher als geeigneter Endpunkt anzusehen, der bevorzugt eingesetzt werden sollte, um die Erkenntnisse klinischer Studien rascher in die tägliche Praxis zu implementieren. Auf Basis dieser stringenten Analyse hat das Oncology

4

Onkologie 2004;27(suppl 1):2–11

Drug Advisory Committee (ODAC), ein Beratungsgremium der amerikanischen Zulassungsbehörde FDA (Federal Drug Administration), kürzlich die einstimmige Empfehlung ausgesprochen, das krankheitsfreie Überleben als validen Parameter bei der Bewertung neuer Zulassungsanträge anzusehen.

Irinotecan-Regime: Kein Benefit in der Adjuvanz Im Gegensatz zur zusätzlichen Gabe von Oxaliplatin zu 5-FU/FS haben Kombinationsregime mit Irinotecan bislang keinen Vorteil in der adjuvanten Therapie gezeigt. So fiel die von Leonard Saltz, Memorial Sloan-Kettering Cancer Center, vorgestellte Studie CALBG C89803 der Cancer and Leukemia Group B, in der das 5-FU/FS-Bolusregime nach Roswell Park über 4 Zyklen mit dem IFL- oder Saltz-Regime über 6 Zyklen verglichen wurde, in doppelter Hinsicht negativ aus [21]. In die Studie waren zwischen April 1999 und April 2001 insgesamt 1264 Patienten im Stadium III aufgenommen worden. Die Rekrutierung war dann aber wegen der erhöhten Toxizität im experimentellen IFL-Arm vorzeitig gestoppt worden: Therapiebedingte Todesfälle (2,8 vs. 1,0%; p = 0,008) und febrile Neutropenien (4 vs. 1%; p = 0,0005) wurden mit dem Saltz-Regime signifikant häufiger registriert als im Kontrollarm. Auch hinsichtlich der Effektivität konnte die IrinotecanKombination nicht überzeugen: Nach 2,6-jährigem Follow-up zeigte sich beim ereignisfreien Überleben kein Vorteil im Vergleich zum Kontrollarm. «Die jetzt vorgestellten Ergebnisse sind endgültig, und C89803 kann daher nicht als positive Studie betrachtet werden», resümierte Saltz die Ergebnisse. Eric van Cutsem, University Hospital Gasthuisberg, Leuven, kam deshalb in seinem Diskussionsbeitrag zu dem eindeutigen Schluss: «Das IFL-Bolusregime sollte in der adjuvanten Therapie des Kolorektalkarzinoms nicht eingesetzt werden. Derzeit hat von den Kombinationsregimen nur FOLFOX4 seine Effektivität in der adjuvanten Situation belegt». Um die Rolle von Irinotecan zu klären, muss laut van Cutsem die Studie PETACC 3 (Pan European Trials in Adjuvant Colorectal Cancer) mit einem Irinotecan-haltigen 5-FU/FS-Infusionsregime abgewartet werden.

Die Zukunft der adjuvanten Therapie Nach der Präsentation von X-ACT (Xeloda Adjuvant Colon Cancer Trial) werden 5-FU/FS-Regime zukünftig wohl ohnehin ihren Platz zugunsten der oralen Fluoropyrimidine räumen müssen. Für die Studie wurden 1987 Patienten mit einem Kolonkarzinom im Stadium III nach Resektion zum 5-FU/FSBolusregime der Mayo-Klinik oder zu Capecitabin randomisiert [5]. Nach median 3-jähriger Beobachtungszeit zeigte sich hinsichtlich des krankheitsfreien Überlebens als primärem Endpunkt zumindest eine Äquieffektivität beider Regime, tendenziell sogar ein Vorteil zugunsten des oralen Fluoropyri-

Einleitung

midins gegenüber dem Kontrollarm (64,2 vs. 60,6%), berichtete J. Cassidy, Universität Glasgow. Darüber hinaus überzeugte die Substanz aufgrund der besseren Verträglichkeit. Aufgrund dieser Ergebnisse sollte Capecitabine zukünftig 5-FU/FS in der adjuvanten Therapie des Kolonkarzinoms ersetzen, resümierte van Cutsem. Er sieht die Zukunft der adjuvanten Therapie in der Kombination oraler Fluoropyrimidine mit Zytostatika wie Oxaliplatin und mit neuen «targeted therapies» wie Bevacizumab, Cetuximab und Tyrosinkinase(TK)-Inhibitoren des EGFR (epidermal growth factor receptor).

Tab. 3. Intergroup-Studie N9741: Über 70% der Patienten beider Studienarme erhielten eine Second-line-Therapie, davon über die Hälfte mit der jeweils anderen Substanz [13]

Zahlreiche neue Kombinationsstudien mit molekularen Therapien

hinsichtlich Gesamt-Überlebenszeit (19,5 vs. 14,8 Monate) als auch hinsichtlich Ansprechraten (45 vs. 31%) und Zeit bis zur Progression (8,7 vs. 6,9) schnitt das Oxaliplatin-Regime bei gleichzeitig besserer Verträglichkeit günstiger ab und avancierte damit zum neuen Standard in der First-line-Therapie des metastasierten Kolorektalkarzinoms. Auf der diesjährigen ASCO-Tagung wurden die Endresultate von 355 Patienten der NCI-Studie vorgestellt, die zu FOLFOX4 oder dem modifizierten R-IFL-Regime randomisiert wurden. Sie sprechen ebenfalls für die Oxaliplatin-Kombination [13]. So konnte die mediane Überlebenszeit nach 22,5monatigem Follow-up mit dem FOLFOX4-Regime bis auf 18,8 Monate verlängert werden – gegenüber nur 16,3 Monaten mit R-IFL. Dieser Unterschied ist umso bemerkenswerter, kommentieren Goldberg et al., als dass jetzt in der Secondline-Therapie die Mehrzahl aller Patienten das zuvor in der First-line-Therapie nicht eingesetzte Zytostatikum (d.h. auch Oxaliplatin im R-IFL-Arm) erhielt (Tab. 3), während der Zugang zu Oxaliplatin in früheren Studienphasen noch limitiert und eine effektive Second-line-Therapie daher nur eingeschränkt möglich war. Auch bei Zeit bis zur Progression (10,0 vs. 6,9 Monate; p<0,0001) und Responseraten (47 vs. 32%; p = 0,0006) ergaben sich signifikante Vorteile zugunsten von FOLFOX4. Die Toxizität war in beiden Studienarmen nur mäßig ausgeprägt; allerdings waren Neutropenie und Parästhesien mit dem FOLOFOX4-Regime signifikant häufiger. Auch wenn die Verträglichkeit der Irinotecan-haltigen Kombiation nach Dosisreduktion deutlich verbessert werden konnte, ist das RIFL-Regime wegen seiner gegenüber FOLFOX4 geringeren Effektivität für die First-line-Therapie des fortgeschrittenen Kolorektalkarzinoms nicht als optimal zu betrachten, resümieren die Autoren.

Derartige Studien sind bereits angelaufen: So wird in der Studie 5202 der Eastern Cooperative Oncology Group (ECOG) bei Hochrisikopatienten im Stadium II das FOLFOX-Regime mit der Kombination aus FOLFOX und dem antiangiogenetisch wirkenden Bevacizumab verglichen. Gleichzeitig findet in dieser Studie mithilfe molekularer Prädiktoren (z.B. Mikrosatelliteninstabilität) eine Risikostratifizierung statt, wobei Patienten niedrigen Risikos einem Kontrollarm mit alleiniger Beobachtung zugeteilt werden. Für die Studie C-08 des National Surgical Adjuvant Bowel and Breast Project (NSABP) sollen insgesamt 2300 Patienten im Stadium II und III zu einem Kontrollarm mit einem modifizierten FOLFOX7-Regime (mFOLFOX7) oder zu einem experimentellen Arm mit mFOLFOX7 plus Bevacizumab randomisiert werden. Die dreiarmige Studie BO17920 schließlich vergleicht FOLFOX4 mit FOLFOX4 plus Bevacizumab und mit der Kombination Capecitabin plus Oxaliplatin (XELOX). Weitere Studien evaluieren die Effektivität von Cetuximab in der adjuvanten Therapie, ebenfalls in Kombination mit FOLFOX4 oder mit Capecitabin/Oxaliplatin.

Oxaliplatin-Regime: Standard in der First-line-Therapie Die hohe Toxizität des Saltz-Regimes war bereits 2001 in der Intergroup-Studie N9741 des National Cancer Institutes (NCI) bei Patienten mit inoperablem, histologisch gesichertem fortgeschrittenem Kolorektalkarzinom aufgefallen [22]. In der ursprünglich sechsarmigen, mehrfach modifzierten Studie hatten sich schließlich nur 2 Vergleichsarme mit FOLFOX4- und Saltz-Regime behaupten können. Allerdings hatte auch letzteres wegen inakzeptabler Toxizität und hoher 60-Tage-Mortalitätsrate (4,5%) durch Reduktion der Irinotecan-Dosis von 125 mg/m2 auf 100 mg/m2 und der 5-FU-Dosis von 500 mg/m2 auf 400 mg/m2 (R-IFL) geändert werden müssen. Auf der ASCOTagung 2003 waren die Studienergebnisse nach 20-monatigem Follow-up vorgestellt worden, die die signifikante Überlegenheit des FOLFOX4-Arms gegenüber dem IFL-Regime bei fast allen Studienendpunkten deutlich gemacht hatten [12]: Sowohl

Einleitung

Patienten mit 2nd-line-Therapie, % Irinotecan, % Oxaliplatin, % 5-FU, %

R-IFL (n = 151)

FOLFOX (n = 154)

72 21 54 34

71 53 18 21

OPTIMOX-Studien: Oxaliplatin-Regime weiter verbessern Das FOLFOX4-Regime hat sich als hoch effektive und insgesamt verträgliche Kombination bewährt. Allerdings kann die mit Oxaliplatin mögliche kumulative sensorische Neuropathie einen Behandlungsabbruch bei respondierenden Patien-

Onkologie 2004;27(suppl 1):2–11

5

FOLFOX: Günstig auch für gefährdete Patienten 526 Patienten mit unbehandeltem MCRC

Randomisation

FOLFOX7 (n = 264)

FOLFOX4 (n = 262)

Erneute Gabe FOLFOX7 (n = 125)

Abb. 3. Design der OPTIMOX1-Studie [6].

ten erforderlich machen. Da die Lebenserwartung von Patienten mit fortgeschrittenem Kolorektalkarzinom mittlerweile deutlich verlängert wurde und Patienten daher sehr viel länger behandelt werden, ist diese kumulative Nebenwirkung laut Paulo M. Hoff, M.D. Anderson Cancer Center, durchaus als relevant anzusehen. Limitierend kann weiterhin die durch den 14-tägig applizierten 5-FU-Bolus induzierte Neutropenie sein. OPTIMOX1, eine Phase-III-Studie an 526 Patienten mit inoperablem oder metastasiertem Kolorektalkarzinom, hatte daher das Ziel, die Effektivität des bewährten Regimes weiter zu steigern und gleichzeitig Nebenwirkungen zu minimieren. In der zweiarmigen Studie wurde das bekannte FOLFOX4Regime (Arm A) mit dem FOLFOX7-Schema verglichen, bei dem eine höhere Oxaliplatin-Dosis (130 mg/m2) mit einem 5-FU/FA-Regime ohne 5-FU-Bolus kombiniert wird. Nach 6 Gaben FOLFOX7 wird dann mit einem vereinfachten 5-FU/FS-Regime (s5LV5FU2) als Erhaltungstherapie weiterbehandelt, nach 12 Zyklen – bei Progression bereits früher – wieder auf FOLFOX7 umgestellt (Arm B). Laut den von de Gramont und Mitarbeitern vorgestellten Ergebnissen sind beide Regime ähnlich effektiv [6]: Die Gesamt-Überlebenszeiten sind mit 20 Monaten für FOLFOX4 und 21,6 Monaten für FOLFOX7 vergleichbar; progressionsfreies Überleben (9,2 vs. 9,0 Monate), Dauer der Tumorkontrolle (9,2 vs. 9,7 Monate) und Gesamtansprechraten (58,5 vs. 58,3%) nahezu identisch. Das Gesamtüberleben war in den Zentren, die die höchsten Anteile an Reinduktionen im FOLFOX7-Arm hatten, am längsten. Das angestrebte Ziel, die Therapieergebnisse durch Erhöhung der Oxaliplatin-Dosis weiter zu verbessern, wurde daher leider nicht erreicht, bedauerte Hoff in seiner Diskussion der Studienergebnisse. Allerdings hat die Studie klar gezeigt, dass eine Oxaliplatin-Therapiepause bei Erhalt der Wirksamkeit und verbessertem Toxizitätsprofil möglich ist, betonte er. So überzeugt das FOLFOX7-Regime durch eine bessere Verträglichkeit mit geringeren Neutropenie- und Neuropathieraten, und es wird von den Autoren als patientenfreundlicher beschrieben.

6

Onkologie 2004;27(suppl 1):2–11

Auch für ältere Patienten (>75 Jahre), die oft aus klinischen Studien ausgeschlossen werden, sind die Oxaliplatin-haltigen Kombinationen geeignet, wie eine exploratorische Analyse im Rahmen von OPTIMOX 1 belegt. Von den 37 über 75-jährigen Patienten in OPTIMOX1 erhielten 20 das FOLFOX4Standardregime, 17 FOLFOX7 [7]. Die Gesamtansprechraten waren mit 65% für FOLFOX4 und 53% für FOLFOX7 vergleichbar. Ähnliches gilt für das progressionsfreie Überleben mit 7,6 bzw. 9,4 Monaten (p = 0,96) und das Gesamt-Überleben mit 14,9 bzw. 25,1 Monaten (p = 0,15). Die Verträglichkeit des FOLFOX4-Regimes bei über 75-jährigen Patienten ist ähnlich günstig wie bei den jüngeren Studienteilnehmern; lediglich Neutropenien treten deutlich häufiger auf. FOLFOX4 und FOLFOX7 unterscheiden sich hinsichtlich ihres Toxizitätsprofils bei den über 75-Jährigen kaum, wobei die Rate an Neutropenien vom Grad 3/4 mit FOLFOX7 niedriger ist. Insgesamt bleibt das günstige Nutzen-Risiko-Profil der Oxaliplatin-haltigen Regime damit auch bei älteren Patienten erhalten, sodass die Autoren diese in ausgewählten Kollektiven mit gutem Performance-Status als sicher erachten. Ebenfalls effektiv sind FOLFOX-Regime bei Patienten mit ungünstiger Prognose, definiert anhand eines stark erhöhten Spiegels an alkalischer Phosphatase – so eine weitere exploratorische Analyse von OPTIMOX1 mit insgesamt 67 Teilnehmern [26]. Die Gesamtresponseraten waren mit 48,5% im FOLFOX4-Arm und 63,4% im FOLFOX7-Arm hoch und unterschieden sich nicht von denen im Gesamtkollektiv. Gesamt-Überlebenszeit und progressionsfreie Zeit waren mit 14 versus 11 Monaten bzw. 6,7 versus 6,5 Monaten in beiden Armen vergleichbar. Demnach kann Oxaliplatin das Überleben im Vergleich zu einer reinen 5-FU/FS-Therapie auch in diesem Kollektiv verlängern. Es ist jedoch mit rund 12 Monaten deutlich kürzer als der sonst mögliche Medianwert von rund 21 Monaten.

Mehr Lebensqualität durch therapiefreies Intervall? In OPTIMOX2 soll der Stellenwert der zwischen den FOLFOX7-Zyklen verabreichten LV5FU2-Erhaltungstherapie geklärt werden. Zu diesem Zweck erfolgt eine Randomisierung zu dem gleichen Regime wie in Arm B von OPTIMOX 1 oder zu einem «Stop-and-go»-Verfahren mit 6 Zyklen FOLFOX7, anschließender Therapiepause und erneuter Gabe von 6 Zyklen FOLFOX7 (Abb. 3). Davon verspricht man sich eine mit bessere Lebensqualität und eine Minimierung von Resistenzrisiko und Kosten. Im Rahmen einer Phase-II-Studie untersuchten T. André, Hôpital Tenon, Paris, und Mitarbeiter gleichzeitig als neue Strategie den Einsatz des antineoplastisch wirkenden, selektiven Cyclooxygenase-2-Hemmers Celecoxib im chemotherapiefreien Intervall des experimentellen Arms [2]. Bei bislang 42 evaluierbaren

Einleitung

Patienten überzeugte dieses Konzept durch ein gutes Sicherheitsprofil. Die Responserate liegt mit 47% – davon 2 Komplettremissionen (CR) – im üblichen Bereich. Das progressionsfreie Überleben in dieser Phase-II-Studie ist mit 6 Monaten jedoch enttäuschend kurz, kommentieren die Autoren. Zum Vergleich: In OPTIMOX1 betrug dieses Intervall im FOLFOX7-Arm 9 Monate. Ob die intermittierende Therapie mit chemotherapiefreien Intervallen eine Zukunft hat, muss die Phase III der OPTIMOX2-Studie zeigen. Grundsätzlich bietet es sich nach Ansicht der Autoren an, die Therapiepausen zur Applikation neuer molekularer Substanzen zu nutzen. Dagegen bezeichnete Hoff ein chronomoduliertes Regime mit Oxaliplatin und 5-FU/FS (FFL4–10), das in der Studie 05963 der European Organization for Research and Treatment of Cancer zur Verbesserung von Effektivität und Verträglichkeit erprobt wurde, als für den Routineeinsatz wenig geeignet [10]. Das chronomodulierte Regime, das höhere Dosen bei weniger Toxizität ermöglichen soll, war bei 554 Patienten mit metastasiertem Kolorektalkarzinom mit dem FOLFOX2-Regime verglichen worden. Die Autoren hatten eine 10%-ige Steigerung der 2-Jahres-Überlebensrate als primären Endpunkt durch die Chronomodulation erhofft. Mit Gesamt-Überlebenszeiten von 19,6 Monaten für FOLFOX2 und 18,7 Monaten für das chronomodulierte Regime hatten sich jedoch beide Therapien als vergleichbar effektiv herausgestellt. Progressionsfreies Überleben und Ansprechraten waren mit 8,4 Monaten bzw. gut 40% in beiden Armen praktisch identisch. Einziger Vorteil der Chronomodulation war die deutliche Reduktion der Neutropenierate von 25,3% mit FOLFOX2 auf 7,3%; ansonsten war die Toxizität in beiden Armen akzeptabel. Für das FOLFOX-Regime spricht jedoch die sehr viel einfachere Applikation, bei der die Infusionszyklen nicht an den zirkadianen Rhythmus angepasst werden müssen.

Tab. 4. Phase-III-Studie der AIO: Nebenwirkungsraten unter FUFOXversus CAPOX-Regime [3] Patienten, %  Capox (n = 164) FUFOX (n = 145)   Grad 1/2 Grad 3/4 Grad 1/2 Grad 3/4 Durchfall Übelkeit Erbrechen Neuropathie Stomatitis HFS Schmerz Fieber Infektion

33 49 27 52 12 29 30 17 12

12 7 5 16 1 1 5 0 3

35 62 31 45 16 27 33 18 14

14 9 7 25 3 0 4 1 4

In der adjuvanten Therapie des Kolorektalkarzinoms ist die Prüfung oraler 5-FU-Prodrugs in Kombination mit Oxaliplatin gerade angelaufen. Anders beim metastasierten Tumor, für den bereits vielversprechende Ergebnisse zu derartigen Kombinationen vorliegen. So zeigt eine deutsche randomisierte Phase-II-Studie, dass Kombinationen von Capecitabin (2 × 1000 mg/m2/Tag, Tage 1–14) mit Oxaliplatin (70 mg/m2 i.v. Tage 1 und 8; CAPOX) oder mit Irinotecan (100 mg/m2 i.v. Tage 1 und 8; CAPIRI) machbar sind und sich durch gute Verträglichkeit und hohe Aktivität auszeichnen. Wie auf der letztjährigen ASCO-Tagung berichtet, werden in der First-lineTherapie Gesamtresponseraten von 51,3%, davon 6,6% CR, mit CAPOX und von 42,6% (CR: 2,9%) mit CAPIRI erreicht [14]. Auch in der Second-line-Therapie nach dem protokollarisch vorgesehenen Cross-over bei Progress waren beide Regime weiterhin effektiv und gut verträglich [15]. Die Re-

sponserate betrug mit CAPIRI 21%, mit CAPOX 13%; bei jeweils etwa der Hälfte der Patienten wurde eine Stabilisierung erreicht. Schwere Nebenwirkungen (Grad 3/4) waren insgesamt selten. Diese Studie war die Basis für eine deutsche Phase-III-Studie der Arbeitsgemeinschaft internistischer Onkologen (AIO), in der die CAPOX-Kombination mit dem 5-FU/FS-Infusionsregime der AIO plus Oxaliplatin (FUFOX) bei 399 Patienten mit metastasiertem Kolorektalkarzinom als First-line-Therapie verglichen wird. Der auf der ASCO-Tagung 2004 vorgestellten Sicherheitsanalyse zufolge ist das Toxizitätsprofil der Zweierkombination mindestens ebenso günstig wie das des gut verträglichen FUFOX-Regimes [Arkenau et al.]. Häufigkeit und Schweregrad gastrointestinaler Nebenwirkungen und des Hand-Fuß-Syndroms sind in beiden Gruppen vergleichbar. Eine schwere sensorische Neuropathie vom Grad 3/4 tritt unter CAPOX sogar deutlich seltener auf als mit dem FUFOX-Regime (Tab. 4). Eine Hämatotoxizität vom Grad 3/4 war insgesamt selten; häufiger wurden lediglich Neutropenien im FUFOX-Arm dokumentiert. Drei verschiedene Oxaliplatin-Kombinationen werden in der Studie TREE 1 bei 150 Patienten mit metastasiertem Kolorektalkarzinom untersucht. Hier wird das klassische FOLFOX4-Regime mit CapeOx (130 mg/m2 Oxaliplatin i.v. Tag 1 plus 1000 mg/m2 Capecitabin 2x/Tag, 14 Tage alle 3 Wochen) und dem bFOL-Regime mit Oxaliplatin (85 mg/m2, Tag 1, 15) plus 5-FU-Bolusregime (500 mg/m2 Tag 1, 8, 15 alle 4 Wochen) verglichen [27]. Die Gesamtansprechraten liegen zwischen 38% (bFOL) und 52% (FOLFOX) ohne signifikante Unterschiede zwischen den drei Armen. Hinsichtlich Verträglichkeit erwies sich das bFOL-Regime als am günstigsten; im CapeOx-Arm mussten 38% der Patienten die Therapie toxizitätsbedingt in den ersten 12 Wochen unterbrechen. Entsprechend wurde die Capecitabin-Dosis für die Studie TREE 2 auf 850 mg zweimal täglich abgesenkt. Diese Studie, für die 223 Patienten rekrutiert wurden, gehört bereits einer neuen Generation an, da die 3 Therapieregime aus TREE 1

Einleitung

Onkologie 2004;27(suppl 1):2–11

Orale Prodrugs ersetzen 5-FU

7

Tab. 5. Hohe Responseraten in der Phase-II-Studie mit der Kombination aus FOLFOX4 plus Cetuximab [24]. Bei 7 Patienten erfolgte eine Resektion von Lebermetastasen. Gesamtansprechraten

N/%

95% CI

Komplettes Ansprechen Partielles Ansprechen Stabilisierung Progression

2 / 5% 32 / 76% 7 / 17% 1 / 2%

1–16 61–88 7–32 1–13

mit Bevacizumab kombiniert werden. Die jetzt präsentierten Sicherheitsdaten aus TREE 2 sind sehr beruhigend und sprechen für das akzeptable Toxizitätsprofil aller drei Arme – einschließlich CapeOx. Mit Spannung werden jetzt die Effektivitätsdaten dieser Studie erwartet, da eine weitere Überlebensverlängerung auf bis zu 24 Monate angenommen wird. Damit hätte man die Lebenserwartung von Patienten mit fortgeschrittenem Kolorektalkarzinom in den vergangenen Jahren etwa vervierfacht, betrug doch das mittlere Überleben noch in den achtziger Jahren bei alleiniger Supportivtherapie nur rund 6 Monate.

Vielversprechende Ergebnisse mit molekularen Therapieansätzen Welche beeindruckenden Ergebnisse mit Kombinationen der klassischen Chemotherapie plus Biologicals zu erreichen sind, zeigt eine internationale Phase-II-Studie, in der das FOLFOX4-Regime zusammen mit dem monoklonalen, gegen EGFR gerichteten Antikörper Cetuximab als Firstline-Therapie verabreicht wurde [24]. In die Studie wurden bislang 43 Patienten mit fortgeschrittenem Kolorektalkarzinom und Nachweis einer EGFR-Expression in Tumorgewebe oder Metastasen aufgenommen. Sie erhielten als First-line-Therapie zusätzlich zum FOLFOX4-Regime Cetuximab in einer Dosis von 400 mg/m2 i.v. an Tag 1; anschließend wurde Cetuximab wöchentlich in einer Dosis von 250 mg/m2 verabreicht. Die Effektivität der innovativen Kombination bei den 42 evaluierbaren Patienten ist ausgesprochen hoch: Insgesamt sprachen 34 Patienten (81%) an, zwei davon komplett. Werden auch Stabilisierungen berücksichtigt, so steigt die Tumorkontrollrate auf 98% an (Tab. 5). Das Sicherheitsprofil der Kombination beschreiben die Autoren als akzeptabel: Als häufigste Nebenwirkungen vom Grad 3/4 wurden Diarrhöen (26%), akneformer Ausschlag (21%) und Neutropenien (14%) registriert. In seiner Diskussion der Studienergebnisse bezeichnete Saltz die erreichten Responseraten als sehr ermutigend und als gute Basis für eine Phase-III-Studie. Der routinemäßige Einsatz der FOLFOX4-Cetuximab-Kombination außerhalb klinischer Studien sei jedoch bislang noch nicht gerechtfertigt. Hier seien die Ergebnisse einer CALBG-Studie an 1100 Pa-

8

Onkologie 2004;27(suppl 1):2–11

tienten abzuwarten, in der die Kombination von mFOLFOX6 mit oder ohne Cetuximab mit dem FOLFIRI-Regime, ebenfalls mit oder ohne Antikörper, in der First-line-Therapie des fortgeschrittenen Kolorektalkarzinoms verglichen werde.

Geprüft werden Antikörper und «small molecules» Ebenfalls durch sehr hohe Ansprechraten fällt eine weitere Phase-II-Studie auf, in der das FOLFOX4-Regime mit dem EGFR-TK-Inhibitor Gefitinib (IFOX) kombiniert wurde [8]. Hier erfolgte eine Stratifizierung der insgesamt 63 Patienten: 31 Teilnehmer waren nicht vortherapiert (Gruppe A), weitere 27 hatten bereits eine Irinotecan-haltige First-line-Therapie erhalten (Gruppe B). In Gruppe A sprachen 23 Patienten (77%) partiell auf die Therapie an, bei weiteren 20% gelang eine Tumorstabilisierung. In Gruppe B lag die Responserate bei 29%, die Stabilisierungsrate bei 58%. Die Zeit bis zur Progression beträgt 9,5 bzw. 5,2 Monate. Das IFOX-Regine ist allerdings durch eine höhere Rate gastrointestinaler Nebenwirkungen (Diarrhöe, Übelkeit, Erbrechen) im Vergleich zu FOLFOX4 charakterisiert. Die hohe Diarrhörate von 54% sollte laut Saltz auf jeden Fall zur Vorsicht bei der weiteren Erprobung von IFOX mahnen. Bislang ist noch unklar, ob die bei Patienten mit Bronchialkarzinom nachgewiesenen aktivierenden Mutationen im EGFR, die zu einer sehr starken Tumorschrumpfung und lang anhaltenden Remissionen führen, auch beim Kolorektalkarzinom auftreten. Die Hypothese, dass eine EGFR-Expression im Tumor Voraussetzung für das Ansprechen auf EGFR-Antagonisten ist, kann heute laut Saltz als widerlegt gelten. Er bezeichnete es als falsch und zynisch, Patienten eine gegen EGFR gerichtete Therapie nur aufgrund einer fehlenden EGFR-Expression vorzuenthalten.

Oxaliplatin ermöglicht Resektion von Lebermetastasen Bei mehr als der Hälfte aller Patienten mit kolorekatalem Karzinom treten im Erkrankungsverlauf Lebermetastasen auf. Lediglich die chirurgische Resektion, die allerdings primär nur in maximal einem Viertel der Fälle möglich ist, bietet den Betroffenen eine Heilungschance. Als effektiv hat sich jedoch die neoadjuvante Chemotherapie erwiesen, die zu Downstaging und verbesserter Resektabilität führt. So gelang in der Sequenzstudie von Tournigand et al. mit dem FOLFOX-Regime als First-line-Therapie bei 21 der 59 Responder eine chirurgische Entfernung zuvor inoperabler Lebermetastasen; 13 von ihnen konnten R0-reseziert werden [25]. Nach einer First-line-Therapie mit dem FOLFIRI-Regime war dagegen nur bei 8 von 61 Respondern eine Resektion möglich. Laut einer retrospektiven Analyse entspricht das Überleben R0-resezierter Patienten mit rund 70 Monaten dem von Patienten, deren Lebermetastasen primär chirurgisch entfernt werden konnten [9].

Einleitung

Problematisch ist allerdings das Vorgehen bei metastasierten Patienten nach Progression unter einer First-line-Therapie, da die Chancen für ein Ansprechen auf Oxaliplatin in der Second line geringer sind. In solchen Fällen könnte eine lokoregionäre Infusionstherapie über die Arteria hepatica (HAI) erfolgreicher sein, da das Tumorgewebe so einer höheren Zytostatikakonzentration ausgesetzt wird [17]. Die Autoren erprobten dieses Vorgehen bei 44 Patienten mit alleinigen, aber inoperablen Lebermetastasen, die auf eine First-line-Chemotherapie, meist mit Irinotecan-Kombinationen, nicht mehr angesprochen hatten. Die HAI wurde mit einer systemischen Chemotherapie kombiniert, um so eine effektive Kontrolle von Lebermetastasen und extrahepatischer Erkrankung zu erreichen. 21 Patienten erhielten neben der HAI mit Floxuridin und Dexamethason eine systemische Chemotherapie mit Oxaliplatin und 5-FU/FS, 23 weitere Patienten zusätzlich zur HAI die Kombination aus Oxaliplatin/Irinotecan. Im Gesamtkollektiv war die Ansprechrate mit 82% hoch, obwohl fast drei Viertel der Teilnehmer unter der vorherigen Therapie progredient geworden waren. In 7% der Fälle gelang eine pathologische Komplettremission. Bei 9 Patienten (20%) konnten die Lebermetastasen bereits erfolgreich reseziert werden; bei weiteren 7 Patienten ist eine Operation möglich. Nach einem medianen Follow-up von jetzt 21 Monaten beträgt die progressionsfreie Zeit bei den systemisch mit Oxaliplatin/5-FU behandelten Patienten 19 Monate, in der Oxaliplatin/IrinotecanGruppe 15 Monate. Die Gesamt-Überlebenszeiten erstrecken sich über mittlerweile 23 bzw. 36 Monate und sind damit als exzellent für dieses Patientenkollektiv zu werten. Die Autoren empfehlen jetzt eine randomisierte Vergleichsstudie, in der ihr Ansatz mit einer alleinigen neoadjuvanten systemischen Chemotherapie verglichen wird.

Pankreaskarzinom: GemOx verbessert Therapieergebnisse

Tab. 6. Therapieergebnisse der Phase-III-Studie von GERCOR/GISCAID mit der Gemcitabin-Monotherapie versus GEMOX-Kombination [19]

Patienten Ansprechrate* Überleben Progressionsfreies Überleben 1-Jahres-Überlebensrate 8-Monats-Überlebensrate

GO

G

157 26,8% 9,0 Monate 5,8 Monate 34,7% 56,5%

156 17,3% 7,1 Monate 3,7 Monate 27,8% 45,3%

P-Wert

0,04 0,13 0,038 0,048

*Remissionen wurden nicht zentral begutachtet, schließen geschätzte partielle Remissionen ein.

Überlebenszeit von immerhin 9,2 Monaten erreicht. Die 1-Jahres-Überlebensraten von 47% bei Patienten mit lokal fortgeschrittenem und von 26% beim metastasierten Pankreaskarzinom lagen in dieser Studie deutlich über denen der Gemcitabin-Monotherapie. Diese ermutigenden Daten rechtfertigten die Initiierung einer Phase-III-Studie, in der die neue Kombination mit Gemcitabin als Monotherapie verglichen wurde. Für die von der französischen GERCOR- und der italienischen GISCAD-Gruppe durchgeführte Studie wurden zwischen März 2001 und Februar 2003 313 Patienten mit histologisch gesichertem Pankreasadenokarzinom rekrutiert [19]. 156 Patienten wurden randomisiert dem Kontrollarm mit Gemcitabin (1 g/m2 wöchentlich als 30-minütige Infusion), 157 Patienten dem experimentellen Arm mit der GEMOX-Kombination zugeteilt. Rund 30% der Patienten in beiden Armen wiesen einen lokal fortgeschrittenen, 70% einen metastasierten Tumor auf. Primäres Ziel der Studie war eine 20%-ige Zunahme der Überlebensrate von 30 auf 50% nach 8 Monaten. Sekundäre Wirkparameter waren progressionsfreies Überleben, Ansprechrate, klinischer Benefit, Verträglichkeit und Lebensqualität.

Überlebensverlängerung um 2 Monate

Zur Sitzung «Gastrointestinal (Noncolorectal) Cancer» unter Leitung von Emily K. Bergsland, University of California, San Francisco Cancer Center, und Charles D. Blanke, Oregon Health and Science University, und zur Postersitzung «Gastrointestinal (Noncolorectal) Cancer» Seit Einführung von Gemcitabin – derzeit immer noch Standard in der Behandlung des Pankreaskarzinoms – versucht man eine Steigerung der unbefriedigenden Ansprech- und Überlebensraten durch Kombination mit weiteren Zytostatika zu erreichen. Am erfolgversprechendsten stellte sich in einer Phase-II-Studie die als GEMOX bezeichnete Kombination mit Gemcitabin (1000 mg/m2 i.v. als 100-min-Infusion, Tag 1) und Oxaliplatin (100 mg/m2 i.v., Tag 2) alle 2 Wochen heraus [18]: Bei guter Verträglichkeit wurden eine hohe Ansprechrate von 30%, eine klinische Benefitrate von fast 40%, ein medianes progressionsfreies Intervall von 5,3 und eine Gesamt-

Die bereits in der Phase-II-Studie beobachtete eindrucksvolle Aktivität der GEMOX-Kombination bestätigte sich auch in dieser randomisierten Vergleichsstudie anhand der signifikant höheren Responserate im Vergleich zur Gemcitabin-Monotherapie (26,8 vs. 17,3%; p = 0,04; Tab. 6). Auch bei der klinischen Benefit-Rate zeigte sich ein signifikanter Vorteil zugunsten des Kombinationsregimes (38,2 vs. 26,9%; p = 0,03), wobei Patienten mit lokal fortgeschrittenem Tumor besonders stark profitierten (45,1 vs. 34,0%). Darüber hinaus gelang eine erhebliche Verlängerung des progressionsfreien Überlebens von nur 3,7 Monaten unter Gemcitabin auf 5,8 Monate mit der Oxaliplatin-Kombination (p = 0,038). Auch die Gesamt-Überlebenszeit konnte durch die Kombinationstherapie um rund 2 Monate verlängert werden (9,0 vs.

Einleitung

Onkologie 2004;27(suppl 1):2–11

9

791 Patienten

R A N D O M I S I E R T

Arm A: Gemcitabin 1000 mg/m2/30 min qw x 3 alle 4 Wochen

Arm B: Gemcitabin 1500 mg/m2/150 min qw x 3 alle 4 Wochen

Arm C: Gemcitabin 1000 mg/m2/100 min Oxaliplatin 100 mg/m2/120 min q 14 Tage

Abb. 4. Design der dreiarmigen ECOG-Studie.

7,1 Monate), was einer 27%-igen Zunahme des medianen Überlebens entspricht. Dieser Unterschied zwischen beiden Studienarmen ist allerdings nicht signifikant, was sich laut Christophe Louvet, Hôpital Saint Antoine, Paris, durch die überraschend positiven Studienresultate, insbesondere auch des Kontrollarms erklären lässt: Erwartet worden waren Überlebenszeiten von 8 bzw. 6 Monaten. Das angestrebte Ziel – die 20%-ige Steigerung der 8-Monats-Überlebensrate – wurde zwar verfehlt (56,5 vs. 45,3%); der Unterschied zwischen beiden Regimen zu diesem Zeitpunkt ist jedoch signifikant, betonte Louvet. Über die Hälfte der Patienten erhielt nach Progression eine Second-line-Therapie. Louvet wies darauf hin, dass diese im Gemcitabin-Arm – entgegen dem Studienprotokoll – in fast drei Viertel der Fälle Platinderivate (einschließlich Oxaliplatin) beinhaltete. Dieser Faktor ist seiner Ansicht nach ein entscheidender Grund für den nur tendenziell ausgeprägten Überlebensvorteil der GEMOX-Kombination. Inwieweit die zusätzliche Gabe von Oxaliplatin oder die Applikation von Gemcitabin mit fixierter Dosierungsgeschwindigkeit zu der Überlebensverlängerung beiträgt, ist laut Louvet schwer auszumachen. Hedy Lee Kindeler, University of Chicago, kritisierte in ihrem Diskussionsbeitrag, dass die Gemcitabin-Regime in den beiden Studienarmen nicht vergleichbar waren. Diese Frage wird jedoch zur Zeit in einer ECOG-Studie der Phase III mit fast 800 Patienten geklärt: In 3 Armen werden Gemcitabin/ Oxaliplatin, die Monotherapie mit Gemcitabin als 30-minütiger Bolus und die Gemcitabin-Infusion verglichen (Abb. 4). Laut Kindler ist es jetzt an der Zeit, auch beim fortgeschrittenen Pankreaskarzinom neue molekulare Substanzen in die Behandlung zu integrieren, da die alleinige konventionelle Chemotherapie den ungünstigen Verlauf bei diesem Tumor voraussichtlich nicht ändern wird. «Solche Ansätze werden die Therapie vermutlich mehr bereichern als die Erprobung weiterer Zytostatika-Doubletten», betonte sie. «Die moleku-

10

Onkologie 2004;27(suppl 1):2–11

laren Therapieansätze sollten zuammen mit der derzeit besten Chemotherapie geprüft werden, und das ist sicherlich die GEMOX-Kombination», so die Meinung von Louvet. Neben der höheren Effektivität zeichnet sich GEMOX auch durch eine gute Verträglichkeit aus; die Toxizität ist im Vergleich zur Gemcitabin-Monotherapie kaum erhöht.

Oxaliplatin-Regime bei Magen- und Ösophaguskarzinom Schlecht ist die Prognose auch beim Magenkarzinom: Höchstens ein Drittel der betroffenen Patienten kann R0-reseziert werden; die langfristige Heilungsrate ist niedrig. Der palliativen Chemotherapie kommt daher große Bedeutung zu, weil sie die Lebensqualität im Vergleich zu alleinigen Supportivtherapie verbessert und eine Lebensverlängerung ermöglicht. Referenztherapie ist derzeit das ECF-Regime mit Epirubicin/Cisplatin/5-FU, mit dem Responseraten zwischen 40 und 45% und ein medianes Überleben von rund 9 Monaten erreicht werden. Aber: Das Regime gilt wegen der dauerhaften Versorgung der Patienten mit einer Pumpe als wenig patientenfreundlich, sodass nach therapeutischen Alternativen gesucht wird. L. Cavanna und Mitarbeiter erprobten bei 35 Patienten mit fortgeschrittenem oder metastasierten Magenkrebs die Kombination von Oxaliplatin (85 mg/m2 i.v., Tag 1) mit einem 5-FU/FS-Infusionsregime [4]. Auf die Therapie sprachen 17 Patienten (48,6%) an, einer von ihnen mit einer Komplettremission. Bei 10 weiteren Teilnehmern (28,6%) wurde eine Tumorstabilisierung erreicht. Die hohe Aktivität ging mit einer nur leicht ausgeprägten Toxizität einher: Schwere Nebenwirkungen vom Grad 3/4 waren insgesamt selten. Am häufigsten wurden Neutropenien (21,3%) und Thrombozytopenien registriert. Beim inoperablen Ösophaguskarzinom wird derzeit eine Radiochemotherapie mit 4 Zyklen 5-FU/Cisplatin und einer Bestrahlung mit 50 Gy (RTOG-Regime) als Referenztherapie betrachtet, mit der eine Langzeit-Überlebensrate von 25% erreicht wird. Giovani et al. erprobten in einer Phase-I-Studie ein Radiochemotherapie-Regime mit Ersatz von Cisplatin durch Oxaliplatin [11]. Die Dosisfindungsstudie an 33 Patienten zeigt, dass die Chemotherapie in einer dem FOLFOX4Regime entsprechenden Dosis sicher zusammen mit einer Bestrahlung von 50 Gy verabreicht werden kann. Bei den 28 evaluierbaren Patienten wurde eine Responserate von 61% erzielt: 3 Patienten sprachen komplett, 14 weitere partiell an. Die mediane Zeit bis zur Progression beträgt 5, die GesamtÜberlebenszeit 9 Monate. Jetzt ist geplant, die neue Kombination in einer kontrollierten Studie der Phase II/III mit dem RTOG-Referenzregime zu vergleichen. Bericht: Dr. Katharina Arnheim, Berlin

Einleitung

Literatur 1 André T, Boni C, Mounedji-Boudiaf L et al.: Oxaliplatin, Fluorouracil, and Leucovorin as Adjuvant Treatment for Colon Cancer. N Engl J Med 2004; 350:2343–2351. 2 André T, Maindrault-Goebel F, Mineur L et al.: Phase II study of an optimized 5FU/oxaliplatin strategy (OPTIMOX2) with celecoxib in metastatic colorectal cancer: A GERCOR study. Proc ASCO 2004;23:abstr 3554 (poster presentation). 3 Arkenau HT, Schmoll HJ, Kubicka S et al.: Phase III trial of capecitabine plus oxaliplatin (CAPOX) versus infusional 5-FU/LV plus oxaliplatin (FUFOX) in first-line metastatic colorectal cancer (MCRC): Interim safety analysis. Proc ASCO 2004; 23:abstr 3546 (poster presentation). 4 Cavanna L, Zaniboni A, Artioli F et al.: Oxaliplatin (OXA), 5-fluorouracil (5-FU) and leucovorin (LV) in patients with advanced or metastatic gastric cancer (A/MGC). Proc ASCO 2004;23:abstr 4068 (poster presentation). 5 Cassidy J, Scheithauer W, McKendrick W et al.: Capecitabine (X) vs bolus 5-FU/leucovorine (LV) as adjuvant therapy for colon cancer (the X-ACT study): Positive efficacy results of a phase III trial. Proc ASCO 2004;23:abstr 3509 (oral presentation). 6 De Gramont A, Cervantes A, André T et al.: OPTIMOX study: FOLFOX7/LV5FU2 compared to FOLFOX4 in patients with metastatic colorectal cancer. Proc ASCO 2004;23:abstr 3525 (poster presentation). 7 Figer A, Perez N, Carola E et al.: 5-fluorouracil, folinic acid and oxaliplatin (FOLFOX) in very old patients with metastatic colorectal cancer. Proc ASCO 2004;23:abstr 3571 (poster presentation). 8 Fisher GA, Kuo T, Cho CD et al.: A phase II study of IFOX (gefitinib with FOLFOX-4) in patients with metastastic colorectal cancer. Proc ASCO 2004;23:abstr 3514 (poster presentation). 9 Giacchetti S, Itzhaki M, Gruia G et al.: Long-term survival of patients with unresectable colorectal liver metastases following infusional chemotherapy with 5-fluorouracil, leucovorin and surgery Ann Oncol 1999;10:663–669. 10 Giacchetti S, Bjarnason G, Garufi C et al.: First line infusion of 5-fluorouracil, leucovorin and oxaliplatin for metastatic colorectal cancer: 4-day chronomodulated (FFL4–19) versus 2-day FOLFOX2. A multicenter randomized phase III trial of the chronotherapy group of the European Organization for Research and Treatment of Cancer (EORTC 05963). Proc ASCO 2004;23:abstr 3526 (poster presentation).

Einleitung

11 Giovani M, Conroy T, Francois E et al.: Phase I study of first line radiochemotherapy with oxaliplatin (Ox), 5-fluoruracil (5-FU) and folinic acid in inoperable locally advanced or metastatic esophageal cancer. Proc ASCO 2004;23:abstr 4044 (poster presentation). 12 Goldberg RM, Morton RF, Sargent DJ et al.: N9741: Oxaliplatin (oxal) or CPT-11 + 5-fluorouracil (5-FU)/leucovorin (LV) or oxal + CPT-11 in advanced colorectal cancer (CRC): Updated efficacy and quality of life (QOL) data from an intergroup study. Proc ASCO 2003;22:abstr 1009 (oral presentation). 13 Goldberg RM, Sargent DJ, Morton RF et al.: N9741: FOLFOX (oxaliplatin (Oxal)/5-Fluoruracil (5-FU)/leucovorin (LV)) or reduced dosis R-IFL (CPT-11 + 5-FU/LV) in advanced colorectal cancer (CRC): Final efficacy data from an intergroup study. Proc ASCO 2004;23:abstr 3621 (poster presentation). 14 Grothey A, Jordan K, Kellner O et al.: Randomized phase II trial of capecitabine plus irinotecan (CapIri) vs capecitabine plus oxaliplatin (CapOx) as first-line-therapy of advanced colorectal cancer. Proc ASCO 2003;22:abstr 1011 (poster presentation). 15 Grothey A, Jordan K, Kellner O et al.: Capecitabine (Xeloda) plus either weekly irinotecan (CAPIRI) or weekly oxaliplatin (CAPOX) are active second-line protocols in patients with metastatic colorectal cancer (MCRC) after failure of firsttime combination therapy: Results of a randomized phase II study. Proc ASCO 2004;23:abstr 3534 (poster presentation). 16 Hickish T, Boni C, Navarro M et al.: FOLFOX 4 as adjuvant treatment for stage II colon cancer: Subpopulation data from the MOSAIC trial. Proc ASCO 2004;23:abstr 3619 (poster presentation). 17 Leonard GD, Fong Y, Jarnagin W et al.: Liver resection after hepatic arterial infusion (HAI) plus systemic oxaliplatin (Oxal) combination in pretreated patients with extensive unresectable colorectal liver metastases. Proc ASCO 2004;23:abstr 3542 (poster presentation). 18 Louvet C, André T, Lledo G et al.: Gemcitabine combined with oxaliplatin in advanced pancreatic adenocarcinoma: Final results of a GERCOR multicenter phase II study. J Clin Oncol 2002;20:1512– 1518.

19 Louvet C, Labianca R, Hammel P et al.: GemOx (gemcitabine + oxaliplatin) versus gem (gemecitabine) in non resectable pancreatic adenocarcinoma: Final results of the GERCOR/GISCAID Intergroup Phase III. Proc ASCO 2004;23:abstr 4008 (oral presentation). 20 NIH consensus conference: Adjuvant therapy for patients with colon and rectal cancer. JAMA 1990; 264:1444–1450. 21 Saltz LR, Niedzwiecki D, Hollis D et al.: Irinotecan plus fluorouracil/leucovorin (IFL) versus fluorouracil/leucovorin alone (FL) in stage III colon cancer (intergroup trial CALBG C89803). Proc ASCO 2004;23:abstr 3500 (oral presentation). 22 Sargent DJ, Niedzwiecki D, O’Connell MJ et al.: Recommendation for caution with irinotecan, fluoruracil, and leucovorin for colorectal cancer. N Engl J Med 2001;345:144–146. 23 Sargent DJ, Wieand S, Benedetti R et al.: Diseasefree survival (DFS) vs. overall survival (OS) as a primary endpoint for adjuvant colon cancer studies: Individual patient data from 12,195 patients on 15 randomized trials. Proc ASCO 2004;23:abstr 3502 (oral presentation). 24 Tabernero JM, van Cutsem E, Sastre J et al.: An international phase II study of cetuximab in combination with oxaliplatin/5-fluorouracil (5-FU)/folinic acid (FA) (FOLFOX-4) in the first-line treatment oft patients with metastatic colorectal cancer (CRC) expressing epidermal growth factor receptor (EGFR). Preliminary results. Proc ASCO 2004;23:abstr 3512 (poster presentation). 25 Tournigand C, Louvet C, Quinaux E et al. FOLFIRI followed by FOLFOX versus FOLFOX followed by FOLFIRI in metastatic colorectal cancer (MRC): Final results of a phase III study. Proc ASCO 2002;22:abstr 494 (oral presentation). 26 Tournigand C, André T, Chirivella I et al.: 5-fluorouracil, folinic acid and oxaliplatin (FOLFOX) in poor prognosis. Proc ASCO 2004;23:abstr 3565 (poster presentation). 27 Welles L, Hochster H, Ramanathan R et al.: Preliminary results of a randomized study of the safety and tolerability of three oxaliplatin-based regimens as first-line treatment for advanced colorectal cancer (CRC) (‘Tree’ Study). Proc ASCO 2004;23: abstr 3537 (poster presentation).

Onkologie 2004;27(suppl 1):2–11

11

Dokumentation und Evaluation der Weiterbildung ASCO-Abstracts Onkologie 2004;27(suppl 1):12–58

Breast Cancer 660

A multistep randomized phase II/III trial comparing oxaliplatin (OXA)+5 fluorouracil (FU) to vinorelbine (VIN)+FU (FUN) after taxane (T)/anthracycline (A) failure in advanced/metastatic breast cancer (MBC) patients (pts): Final results. S. Delaloge, M. Tubiana-Hulin, A. Wardley, L. Del Mastro, A. Santoro, A. Zambelli, A. Bonetti, C. Boni, F. Habib, S. Brienza Institut Gustave Roussy, Villejuif, France; Centre René Huguenin, St Cloud, France; Christie Hospital, Manchester, United Kingdom; Istituto Nazionale per la Ricerca sul Cancro, Genova, Italy; Istituto Clinico Humanitas, Rozzano (MI), Italy; Fondazione Salvatore Maugeri, Pavia, Italy; Azienda Ospedaliera di Legnago, Legnago, Italy; Arcispedale Santa Maria Nuova, Reggio Emilia, Italy; Centre d’oncologie Al Azhar, Rabat, Morocco; Debioclinic S.A., Charenton le Pont, France

Background: OXA is a new DACH platinum (Pt) with preclinical evidence of activity in vivo and in mammary cell lines. Previous clinical study has shown promising activity both for OXA alone (Ann Oncol 2001 Feb 12(2):179–82) and with FU (JCO 2002 15;20:2551–8) in pretreated MBC. Methods: In order to compare tolerance and activity of OXA/FU to reference regimen FUN in pts pretreated with both T and A, we conduct a randomized phase II–III trial. Pts are assigned either to OXA 130 mg/m2 2 hours IV on day (D) 1 plus FU 750 mg/m2 daily by continuous IV infusion D1 to D5 q 3 wks or VIN 25 mg/m2 IV bolus plus the same FU. Eligibility criteria included relapsing progressive MBC, at least one previous chemotherapy (CT) for MBC, no more than 3 prior CT, previous A, previous T therapy given for MBC in an appropriate dose and schedule, at least one measurable target lesion (RECIST criteria), WHO PS 0–2, adequate baseline organs functions. Stratification factors: institution, best response

Median age (range) ≥2 previous CT PD as best response to previous T PS 1–2 Nb of cy Neutropenia Mucositis Neurosensory Toxic death RR PFS (wks) OS (wks)

OXA ARM N=68

VIN ARM N=69

p-value

53 (30–78) 21 (31%) 24 (35%)

52 (35–73) 18 (26%) 24 (35%)

0.7809 0.5340 0.9500

28 (42%) 6 (1–21) 9 (13%) 8 (12%) 5 (7.5%) 0 16 (23.5%) 19.1 (16.7–23) 61.7 (44.7–91)

27 (39%) 6 (1–28) 53 (78%) 22 (32%) 0 1 (1.5%) 19 (27.5%) 22.9 (18.6–30.1) 71 (49.3–85)

0.7519 0.5371 <0.001 0.0043 0.0279 – 0.5908 0.2640 0.7392

© 2004 S. Karger GmbH, Freiburg Fax +49 761 4 52 07 14 E-mail [email protected] www.karger.com

Accessible online at: www.karger.com/onk

to T: PR or CR vs no response, visceral site: yes vs no, number of previous chemotherapy lines in metastatic disease: 1 vs 2. Efficacy was evaluated by radiological assessment every 6 wks; responses are confirmed at least 4 wks later. Results: Between 7/00 and 7/02 137 Pts have been randomized in 33 centers (68 in OXA arm and 69 in VIN arm). Main pts characteristics, safety (G 3–4 AE by pts) and efficacy (RR, PFS, OS) results are shown in the table below. The study was prematurely discontinued due to accrual difficulty related to competitive drugs introduced (Capecitabine®) in the same clinical setting. Conclusion: OXA-5FU combination achieve similar results in term of RR, PFS and OS and favorable safety profile when compared to VIN-5FU combination which is one of the most active CT regimen in post taxane MBC.

705

A phase 2 study of oxaliplatin (O) – capecitabine (C) chemotherapy in metastatic breast cancer (MBC) patients pretreated with anthracyclines and taxanes N. Gebbia, F. Fulfaro, G. Badalamenti, C. Arcara, G. Cicero, G. Rinaldi, L. Vizzini, F. Spinnato, A. Russo, M. R. Valerio University of Palermo, Department of Oncology, Palermo, Italy

Background: O and C have showed an interesting activity as single agents in breast cancer patients. We carried out a phase II study to evaluate the efficacy and safety of combined O and C in anthracyclines and taxanes pretreated MBC patients. Methods: Eighteen MBC patients were treated with C 1,000 mg/sm bis in die per os, days 1 to 14 with a week of rest and O 130 mg/sm (2-hour intravenous infusion), day 1 every 3 weeks. Inclusion Criteria were: age <75, Performance Status (PS)≤2, normal kidney and liver function, absence of brain symptomatic metastases, at least two prior chemotherapy lines with anthracyclines and taxanes, informed consent. Mean age was 58 years (range, 33 to 74 years), with a median of two involved organs. Ten patients had a PS =0, 5 patients had a PS=1, 3 patients had a PS=2. The most frequent sites of disease were: liver (10 pts), skin (4 pts), bone (8 pts), pleura-lung (5 pts), lymphnodes (2 pts). Results: Fifteen patients were assessable for response and toxicity. According RECIST criteria 1 CR, 4 PR, 4 SD and 6 PD were obtained with an overall response rate of 33%. The partial responses occurred in patients with metastatic liver disease and the complete response occurred in a patient with cutaneous metastatic nodules and liver metastases. Median time to progression was 5.5 months (range 3–8 months) and median survival was 11.5 months (range 6–17 months). Hematotoxicity was prevalent but rarely severe, with only 2 patients (13.3%) with G3 neutropenia and one (6.6%) patient with G3 thrombocytopenia. G1–2 neutropenia, anemia and thrombocytopenia occurred in 5 (33%), 3(20%), and 5 (33%) patients, respectively. One third of patients developed grade 2–3 peripheral neuropathy, with grade 3 in only 2 (13.3%) patients. Nausea & vomiting G1–2 and diarrhea G1–2 were present in 6 patients. Hand-foot syndrome G2 occurred in 4 (26.6%) patients. Conclusions: From these preliminary data O/C combination seems to be effective with a favorable safety profile in anthracycline/taxane-pretreated MBC patients.

Developmental Therapeutics – Cytotoxic Chemotherapy 2008

Modeling the response of colorectal cancer (CRC) to common chemotherapy using multivariate gene expression S. C. Glasgow, J. Yu, D. C. Linehan, M. A. Watson, H. L. McLeod Washington University School of Medicine, St. Louis, MO

Background: Expression of key enzymes in chemotherapy drug pathways correlates with CRC response following treatment. No prior research has simultaneously described the patterns of expression in CRC for 5-fluorouracil, irinotecan and oxaliplatin pathways. Characterizing relative expression levels of these enzymes is a first step towards definitively predicting tumor responsiveness to chemotherapy. Methods: Twelve genes known to influence tumor response to chemotherapy were selected from the pharmacogenetic pathways of 5-fluorouracil, irinotecan and platinum agents. Genes included intracellular targets, DNA repair enzymes, drug transporters and metabolizers. Gene expression was examined in primary colorectal adenocarcinoma and adjacent normal mucosa in 51 patients with stage III or IV disease using TaqMan RT-PCR. Relative expression levels were calculated with β-actin as control using a method to account for differing polymerase amplification efficiencies. Results: TYMS, ERCC1 & 2, GSTP1, TOP1, and UGT1A1 were significantly over-expressed in tumor cells relative to normal mucosa, while CES2, DPYD and ABCG2 were under-expressed (p <0.05, Wilcoxon matched pairs test). Expression of ABCB1, ECGF and CYP3A4 did not differ significantly between tumor and normal tissues. Expression of enzymes regulating platinum-agent metabolism and DNA repair were co-regulated. Using differential expression relative to normal mucosa as a marker for tumor resistance, predicted tumor response rates were calculated. 19.6% of tumors were predicted responsive to single-agent irinotecan, 7.8% to 5-FU, and 15.7% to oxaliplatin. Two-thirds were predicted resistant to all three drugs. Conclusions: Significant differences in gene expression were found between CRC and normal mucosa. Distinct sensitivity subgroups existed among the 51 patients with pathologically homogenous disease. These preliminary observations allow further research to correlate multivariate gene expression with clinical outcomes in order to develop a model for predicting response to chemotherapy, eventually resulting in patient-specific treatment.

hort 1 – GW 1250 mg/day + FOLFOX4 at –20% of the standard dose; cohort 2 – GW 1250 mg/day + FOLFOX4 at standard dose; cohort 3 – GW 1500 mg/day + FOLFOX4 at standard dose. OTR was the dose level at which ≤ 1/6 pts experiences a DLT. DLT was based on hematological and non-hematological toxicity in cycle 1. Assessments of anti-tumor activity and left ventricular ejection fraction (LVEF) were performed every 4 cycles. Results: 13 pts were treated (cohort 1 – 3; cohort 2 – 3; cohort 3 – 7). Median age was 57 years (39–70). Histology: 2 colon, 1 ovary, 3 rectum, 2 cholangio, 2 cervix, 2 pancreas, and 1 esophagus. Total of 57 cycles were delivered: median 4 (2–9). Main toxicities are summarized in the table. 2 Pts had grade ≥ 3 hematological toxicities, which resolved after delay of the next cycle. No DLTs have been reported. LVEF values were relatively unchanged compared to pre-study values. 7 pts are evaluable for response. 2 PR, 2 SD and 3 PD. Conclusion: Standard dose FOLFOX4 in combination with GW 1500 mg/day is well tolerated. Enrollment continues at this dose level including pharmacokinetic assessments.

Cohort 1 Number of pts Adverse events Leucopenia Neutropenia Thrombopenia Diarrhea Nausea/vomiting Neurotoxicity Rash

3 G1–2 0 0 1 3 2 1 1

Cohort 2

G≥3 0 0 0 0 0 0 0

3 G1–2 0 0 0 3 2 2 1

Cohort 3

G≥3 0 0 1 0 0 0 0

7 G1–2 3 1 2 4 6 4 3

G≥3 0 1 0 0 0 0 0

2108

Pharmacokinetic study of oxaliplatin (lohp) and capecitabine combination in advanced colorectal cancer M. Gil-Delgado, G. Bastian, B. Paule, S. Urien, J.-P. Spano, G. Deguetz, K. Chouahnia, P. Saintini, J.-L. Breau, D. Khayat Pitie Salpetriere Hospital, Gif sur Yvette, France; Pitie Salpetriere Hospital, Paris, France; Paul Brousse Hospital, Villejuif, France; Avicenne Hospital, Bobigny, France

Background: EGFR and erbB2 receptor expression in tumors predicts a poor outcome. GW572016 (GW) is a potent EGFR/erbB2 RTK inhibitor. The FOLFOX4 is an effective therapy in colorectal cancer. It is hypothesized that inhibition of kinase activation by GW will contribute to the cytotoxicity of FOLFOX4. This phase I trial aimed to determine the safety and optimally tolerated regimen (OTR) of FOLFOX4 with daily oral GW in patients (pts) with solid tumors. Methods: Pts were sequentially enrolled into 3 cohorts: co-

Purpose: This trial has been designed in order to establish potential pharmacokinetic (PK) interaction between Oxaliplatin (Oxa)and Xeloda (Xel) when administrated in combination and to monitor neurological toxicity when Oxa is administered over 6 hours infusion. Treatment schedule: Oxa 130 mg/m2 6 hours perfusion at day 1, Xel 1000 mg/m2 twice daily at day 2 to 15. Treatment repeated every 22 days if >1000 neutrophiles and >100 000 platelets and no other G 3–4 non hematological toxicity. PK analysis were performed at day 1 and 15; receptor for IL 6 at day 1. Blood sampling was done – on day 1 (30min before the end of infusion and 1h30min after the end of infusion in order to validate the Pk population model we recently published for Oxa). On day 2, 30min, 1h, 2h and 4h after Xel. Xel, dFdC, dFdU, 5FU and FBAL were monitored by HPLC and GC-MS in order to determine potential metabolic interaction with Oxa. On day 14, 30min, 1h, 2h and 4h after Xel in order to compare adsorption and PK of Xel and metabolites with and without Oxa. Results: Between April 2003, and December 2003, 10 patients with advanced colo-rectal cancer were included, in first line chemotherapy for metastatic disease, 3 males and 7 females, 7 colon cancer and 3 rectum, medium age 56 (46 to 66): PS 0=7, PS 1=2, PS 2=1; liver involvement in 8 patients (80%) and lung in 4 patients (40%); median cycles 4 (1 to 9). Toxicity = 3 pts Grade III diarrhea, 1 heart pain at day 2 of first cycle and 1

Abstracts

Onkologie 2004;27(suppl 1):12–58

2044

Phase I trial to determine the safety and tolerability of GW572016 in combination with oxaliplatin (OX)/ 5-fluorouracil (5-FU)/leucovorin (LV) [FOLFOX4] in patients with solid tumors W. S. Lakhai, J. H. Beijnen, S. S. Den Boer, A. M. Westermann, M. Versola, K. Koch, P. Ho, L. Pandite, D. J. Richel, J. Schellens Netherlands Cancer Institute/Antoni van Leeuwenhoek Hospital, Amsterdam, Netherlands; Slotervaart Hospital, Amsterdam, Netherlands; Academic Medical Center, Amsterdam, Netherlands; GlaxoSmithKline, Research Triangle Park, NC

13

unexplained death. No hand-food syndrome was observed in this population. Only mild paresthesia in 2 Pts (20%)after 7 cycles Efficacy: 2 CR, 2 PR and 2 stabilizations. PK and IL6 receptor results will be presented. So far, 2 plasma samples are enough to extrapolate AUC of ultrafilterable Oxa according our model. Pk of Xel and metabolites are currently evaluated and Pk interaction between Oxa and Xel will be analysed. Conclusion: This combination is active, easy to administer and very well tolerated especially without significant neurological toxicity when Oxa is infused over 6 hours. Moreover, 2 blood samples are sufficient to monitor Oxa Pk according to the population Oxa parameters and to consider posology adaptation in further trials

2111

Phase I study of oxaliplatin (Ox), irinotecan (Irino), and capecitabine (Cape) in patients with solid tumors S. S. Krishnamurthi, J. M. Brell, K. M. Coviello, A. Dowlati, M. J. Egorin, C. L. Hoppel, S. T. Ingalls, S. P. Ivy, X. Li, S. C. Remick Case Comprehensive Cancer Center, Cleveland, OH; University of Pittsburgh Cancer Center, Pittsburgh, PA; National Cancer Institute, Bethesda, MD

Background: The combination of Ox, Irino and Cape has high potential for activity in gastrointestinal malignancies. Methods: Patients (pts) had solid tumors not amenable to curative therapy. Treatment: Ox intravenous (IV) days 1,8,15,22, followed by Irino IV days 1,8,15,22, and Cape orally bid Monday–Friday weeks1–4, repeated every 6 weeks. Primary endpoints: maximum tolerated dose (MTD), dose-limiting toxicity (DLT) and pharmacokinetic (PK) analysis. Response assessed q 2 cycles. Results: (see table) 22 pts enrolled: 13 males; median age 60, range 38–80; 11 colorectal cancer (ca); median 2 prior regimens. DLT at initial dose level 1A led to a protocol amendment lowering Cape dose. Peak SN38 concentrations of 10 ng/ml exceeded the IC50 doses reported for KB and L1210 tumor cells (0.37 and 3.6 ng/ml) (Kaneda, Cancer Res 1990). The mean peak ultrafilterable platinum was 0.75 +/– 0.17 2g/ml, exceeding the 0.5 2g/ml Ox concentration at which antitumor activity was observed against colon cancer using a human tumor cloning assay (Von Hoff, Clin Cancer Res 1998). Of 16 evaluable pts there was 1 complete response (gastric ca, prior 5-FU/radiation) and 8 with stable disease, including 6 with freedom from progression for ≥ 6 months. Conclusions: The recommended phase II dose is dose level 3. Preliminary PK parameters appear consistent with cytotoxic concentrations in pre-clinical models. This regimen is active in a pre-treated population. (Supported by NIH grants UO1 CA 62502–10 and MO1 RR000080, General Clinical Research Center.) Dose Escalation and Toxicity Dose Level

# Pts

Ox mg/m2

Irino mg/m2

Cape mg DLT bid

1A 1 2 3 = MTD

4 3 6 6

60 60 60 60

40 40 40 50

1000 300 450 450

4

3

60

60

450

14

2 gr 3 diarrhea 0 1 gr 3 hyponatremia 1 gr 3 nausea/ vomiting/ diarrhea 3 gr 3 diarrhea

Onkologie 2004;27(suppl 1):12–58

2128

Early results of a phase I study of weekly topotecan (T) in combination with weekly platinum compounds L. A. White, N. N. Sjak-Shie, C. Kimminau, A. M. Schmidt, A. O. Greco, J. R. Eckardt The Center for Cancer Care and Research, St Louis, MO

Background: Topotecan is synergistic in combination with platinum compounds. Unfortunately, the combination has demonstrated significant toxicities when administered on a q 3-wk schedule. Therefore, we initiated a phase I study of weekly topotecan in combination with the 3 commercially available platinum compounds to try to decrease toxicity and maintain activity. Methods: Patients were treated with platinum compounds given at a fixed dose of either: cisplatin (C) 30 mg/m2/wk, carboplatin (Cb) AUC 2/wk or oxaliplatin (OX) 40 mg/m2/wk on D 1, 8, and 15 of a 28 day cycle. The starting dose of topotecan was 2.5 mg/m2/wk for 3 of 4 wks with dose escalation in increments of 0.5 mg/m2. Results: To date, 16 pts have been entered. Patient eligibility criteria were typical for a phase I study in patients with refractory solid tumors, sex (11F/5M), Median age of 63 yrs (range 40 – 79). Patients have been treated as follows: T/C = 6 pts in dose level (DL)1, 1 in DL 2. T/OX = 7 pts in DL 1, 1 in DL 2. T/Cb = 1 pt in DL 1. Toxicity on the 1st DL of T/C include 3 pts. with gr 3 neutropenia, 2 pts with gr 3 thrombocytopenia, 1 pt with gr 3 anemia, 1 pt with gr 3 infection. One pt in DL 1 had a gr 2 thrombocytopenia requiring dose delay for 2 weeks and a dose reduction. Therefore the DL 1 cohort was expanded to 6 pts. Two patients in the 1st DL had a delay to start the 2nd cycle. One patient has been enrolled at the 2nd DL and it is too early to evaluate. T/OX toxicity on DL 1: 1 pt with gr 3 anemia and gr 4 fatigue (DLT) requiring hospitalization, and dose delay, 1 pt with gr 3 thrombocytopenia, 1 pt with gr 3 infection. One patient has been enrolled at the 2nd DL and it is too early to evaluate. T/Cb toxicity on DL 1: 1 pt with gr 3 neutropenia and anemia but no DLT in the 1st cycle. Responses on T/C include: 1 pt with ovarian had a CR after 4 cycles, 1 pt with CUP had a PR after 4 cycles. Responses on T/OX include: 1 pt with small cell carcinoma of the abdomen has maintained a PR after 8 cycles. Conclusions: Platinum compounds when given with topotecan and administered weekly appear to have activity in a variety of tumor types. The safety profile appears to be acceptable. Enrollment will continue until the MTD is identified for each topotecan doublet.

Developmental Therapeutics – Molecular Therapeutics 3017

Bowel perforation and fistula formation in colorectal cancer patients treated on Eastern Cooperative Oncology Group (ECOG) studies E2200 and E3200 B. J. Giantonio, H. X. Chen, P. J. Catalano, N. J. Meropol, P. J. O’Dwyer, A. I. B. Benson Abramson Cancer Center, University of Pennsylvania, Philadelphia, PA; The National Cancer Institute, Rockville, MD; Dana-Farber Cancer Institute, Boston, MA; Fox Chase Cancer Center, Philadelphia, PA; Northwestern University, Chicago, IL

Background: The addition of bevacizumab (rhu-anti-VEGF) to frontline chemotherapy for advanced colorectal cancer (CRC) improves survival. Bowel perforation and fistula formation have been reported in colorectal cancer patients treated with bevacizumab. Methods: ECOG conducted two trials of bevacizumab in patients with metastatic CRC: as frontline chemotherapy with IFL (E2200) and second line therapy with 5-fluorouracil/leucovorin/oxaliplatin (FOLFOX4)

Abstracts

(E3200). Toxicity forms and severe adverse event reports were reviewed for bowel perforation and fistula formation. Results: In E2200, 86 patients were treated. There were 2 reports of bowel perforation and 2 reports of enterocutaneous fistulae. The bowel perforations occurred in cycles 1 and 10; both occurred at anastomotic sites. One perforation occurred 27 days after hemicolectomy. The fistulae occurred in cycles 1 and 6. In E3200, 824 patients were treated (290 chemotherapy alone, 292 chemotherapy with bevacizumab, and 242 bevacizumab alone). There were 5 bowel perforations; all occurred in patients treated with bevacizumab (<1%). There were 3 fistulae, one in each arm of the study. All but one of the bowel perforations occurred in cycle 1 of therapy; one occurred in cycle 3. Three patients with perforation had invasive procedures to the gastrointestinal tract within 3 weeks of starting bevacizumab. The fistulae were noted at cycles 2, 3, and 10. Two were enterocutaneous and one was enterovaginal. All bowel perforations presented with pain: abdominal pain in 5, hip/lower extremity pain in 2, and rectal pain in one. Overall, 2 patients with perforation died of this complication. Conclusions: Use of bevacizumab in colorectal cancer patients may be associated with an increased risk of bowel perforation and fistula formation. Although an uncommon occurrence, bowel perforation should be considered in colorectal cancer patients treated with bevacizumab who develop symptoms consistent with this diagnosis.

3056

Results of a phase I trial of BAY 43-9006 in combination with oxaliplatin in patients with refractory solid tumors P. Kupsch, K. Passarge, H. Richly, K. Wiesemann, R. A. Hilger, C. G. Haase, B. Schwartz, M. E. Scheulen, S. Seeber, D. Strumberg University Medical School of Essen, Essen, Germany; Bayer AG, Clinical Pharmacology, Wuppertal, Germany; Bayer Corp., Westhaven, CT

Background:BAY 43-9006 (BAY) is a novel signal transduction inhibitor that prevents tumor cell proliferation and angiogenesis through blockade of the Raf/MEK/ERK pathway at the level of Raf kinase and the receptor tyrosine kinases VEGFR-2 and PDGFR-β. Previous single-agent phase I trials show BAY is well tolerated with manageable and reversible side effects, most commonly hand-foot skin (HFS) reaction, rash, fatigue, and diarrhea. This study was initiated to determine safety and pharmacokinetics (PK) of BAY in combination with oxaliplatin (OXAL). Methods:Patients (pts) received continuous BAY at 200 mg bid (cohort 1) followed by escalation to 400 mg bid (cohort 2) given with a fixed dose of OXAL at 130 mg/m2 q3wks. Results: 24 pts (M:F ratio =16:8; median age 60.5 yrs range 32–73) entered the study. Common tumor types were CRC (21%), Mel (21%) and RCC (13%). Drug-related dose-limiting toxicities to either drug were gr 3 thrombocytopenia and gr 3 diarrhea in cohort 1 (thrombocytopenia 1/8 pts; diarrhea 1/8 pts) and in cohort 2 (thrombocytopenia 2/16 pts; diarrhea 1/8 pts). Other toxicities consisted of non-hematologic: gastrointestinal - 92% (diarrhea - 58%, nausea 50%), constitutional - 71% (fatigue - 38%), neurological - 67% (sensory neuropathy - 46%), pain - 50%, dermatological - 46%. Hematologic AEs were uncommon. Single infusions of 130 mg/m2 OXAL did not significantly alter the steady state PK profile of BAY following multiple oral doses of 200 and 400 mg bid BAY. Similarly, PK profiles of total and unbound platinum were not significantly influenced by concomitant multiple doses of BAY. Prolonged stabilization of previously progressive disease was obtained in 2/8 pts in cohort 1 and 5/16 pts in cohort 2. Two additional pts with gastric cancer (1 in each cohort) had partial responses according to modified WHO criteria. Conclusions:Continuous 400 mg bid BAY in combination with 130 mg/m2 OXAL (q3week) appears to be well tolerated with no clinical-

Abstracts

ly relevant PK interaction. Clinical activity has been observed and an extension in CRC pts is ongoing. The recommended dose of BAY for phase II studies in combination with OXAL is 400 mg bid BAY.

3143

Dual role of p21waf1/cip1 in effects of oxaliplatin (L-OHP) against colon cancer cells in p53 dependent and independent pathway T. Hata, H. Yamamoto, M. Ikeda, M. Yasui, I. Seshimo, O. Takayama, H. Hukunaga, M. Ohue, M. Sekimoto, M. Monden Graduate School of Medicine, Osaka University, Suita Osaka, Japan

Backgrounds: L-OHP, an active agent for colorectal carcinoma (CRC) is a platinum derivative such as Cisplatin, but its mechanism in antitumor activity is different from Cisplatin. The aim of this study was to investigate the mechanism of L-OHP on cell growth inhibition in terms of p53 status and its down stream effector, p21waf1/cip1 in the CRC cell lines. Methods: p53 wild type cell lines, HCT116, and p53 mutant cell lines, DLD1 and SW480 were subjected to L-OHP for 48 hours to find out IC50. After 48-hour-exposure of L-OHP (IC50), expression of a cell cycle-related factor, p21waf1/cip1, was examined by western blot analysis. Flowcytometric analysis of these cell lines was performed after L-OHP exposure. Results: IC50 of L-OHP in HCT116, DLD1, and SW480 were 0.4, 9.1, and 2.1 µM, respectively, and maximal growth inhibitory effect on DLD1 was 40% or less even the dose of L-OHP was increased up to 50µM. suggesting p53 status may enhance L-OHP activity. In p53 wild type cell lines, p21waf1/cip1 expression was enhanced by western blot analysis, and increase in G0–G1 phase fraction and decrease in S phase fraction were observed after 12h exposure (G1 arrest pattern) by flowcytometric analysis. In contrast, p21waf1/cip1 expression was decreased, and rapid decrease in G1 phase fraction and increase in S phase was observed after 12h or 24h exposure of L-OHP (G2 arrest pattern) in the p53 mutant type cell lines. Although mRNA expression level of p21waf1/cip1 was not decrease after L-OHP exposure, protein expression was decreased. This decrease in protein expression was inhibited by Clasto-Lactacystin β-lactone, proteasome inhibitor in HCT116 and SW480 cell lines. Conclusion: The antitumor mechanisms of L-OHP were 1) enhancement of expression of p21waf1/cip1 via p53, and 2) direct degradation of p21waf1/cip1 independent of p53.

3145

Flavopiridol potentiates effect of oxaliplatin in vitro and in vivo and requires intact p53 M. V. Motwani, A. Racanelli, G. K. Schwartz Memorial Sloan Kettering Cancer Institute, New York, NY

Background: Oxaliplatin (O) as part of the FOLFOX regimen has been approved as second line therapy in the treatment of metastatic colon cancer. However, response rates, remain quite poor (<10%). We have shown that the CDK inhibitor flavopiridol (F) potently enhances the effect of a wide range of chemotherapy agents (Motwani M, et al CCR 7: 4209, 2001). This has been translated into a series of phase I clinical trials with encouraging clinical results. In view of this we evaluated the combination of F with O to determine whether we could improve the effect of O. Methods: HCT-116 colon cancer cell lines with wild type (p53+/+) and absent (p53-/-) p53 were treated with F and O as single agents, together or sequentially in vitro and in vivo and examined for induction of apoptosis by quantitative fluorescent microscopy and PARP cleavage, inhibition of colony formation

Onkologie 2004;27(suppl 1):12–58

15

units (CFU), and delay in tumor growth. These findings were correlated to changes in p53 and p21 expression by western blot. Results: As single agents F and O induced apoptosis in 12 ±2% and 2 ±1% of the p53 +/+ cells, respectively. This increased to 35 ±5% when both drugs were given together. When F was given before O, 40 ±4% of the cells underwent apoptosis. The reverse sequence of O followed by F induced apoptosis in only 14% of the cells. CFUs decreased from 50% and 70% with O and F alone to 25% with the two drugs given together. PARP cleavage was also higher with concurrent treatment compared to the single agents. O induced p53 and p21 in the p53 +/+ cells, and the combination therapy suppressed p21 expression with no decrease in p53. Treatment of p53-/- cells indicated no potentiation with the O/F combination under any of the conditions tested. Treatment of the p53 +/+ xenografts with O and F together resulted in an 80% decrease in tumor growth. This was compared to 47% with O alone and 69% when O was given before F Conclusions: 1) F potently enhances the effect of O, 2). O+F or F followed by O are superior to O followed by F, 3) p53 expression and the suppression of p21 may be a critical determinant in the response to therapy. 4) Based on these results, a phase I study of F with FOLFOX is planned in which F will be given with O. This work was supported by RO1 CA67819.

3171

A phase 1 trial of motexafin gadolinium and docetaxel for advanced solid tumors N. Ramnath, G. Chatta, M. Egorin, S. Phan, P. J. Creaven Roswell Park Cancer Institute, Buffalo, NY; University of Pittsburgh Cancer Institute, Pittsburgh, PA; Pharmacyclics, Sunnyvale, CA

Background: Motexafin gadolinium (MGd) is a redox mediator that selectively targets tumor cells, producing reactive oxygen species. In preclinical models, MGd enhances the activity of several chemotherapy drugs including taxanes. Based on this, a multi-center Phase 1 study with MGd and docetaxel was carried out in patients (pts) with advanced solid tumors, with normal bone marrow, hepatic and renal function. Experimental design: Cohorts of 3–6 pts were treated with MGd, starting at 2.5 mg/kg weekly with docetaxel starting at 30 mg/m2 on a weekly X 3 schedule every 28 days. The primary objective was to determine the maximum tolerated dose (MTD) of the combination on this schedule and the dose limiting toxicities (DLT). Results: Eight pts were treated at MGd (2.5–5 mg/kg weekly) along with docetaxel (30 mg/m2). The MGd was infused over 0.5h, 30 minutes prior to docetaxel (given IV over 0.5h). There were 8 males with a median age of 61yrs (range 54 –76). Diagnoses were NSCLC (3); SCLC (2); Gastric (1); bladder (1); and penile (1). (No of pts in parenthesis) Pts had had a median of 3 prior chemotherapy regimens (range 1–5); 3 pts had had prior radiation. Two pts received 1.5 cycles. At dose level (DL) 1, toxicities >/= 2 were fatigue, diarrhea, leucopenia, neutropenia, hyperglycemia, elevated APTT, seen in 3 pts. At DL 2 (MGd 5.0 mg/kg with docetaxel 30 mg/m2), 1 pt had a DLT (gr 3 rash). Other toxicities >/= gr 2 at DL 2 were nausea, vomiting, diarrhea, fatigue, and dehydration in 2 pts. Data on the last 3 pts is pending. Accrual is ongoing to determine the MTD. Conclusion: MGd in combination with weekly docetaxel is feasible and preliminary results show that there were no gr 4 toxicities.

16

Onkologie 2004;27(suppl 1):12–58

Gastrointestinal Cancer – Colorectal/Liver A. Adjuvant Colorectal Cancer 3619

FOLFOX4 as adjuvant treatment for stage II colon cancer (CC): Subpopulation data from the MOSAIC trial T. Hickish, C. Boni, M. Navarro, J. Tabernero, C. Topham, A. Bonetti, P. Clingan, A. Figer, T. Andre, A. De Gramont Dorset Cancer Center, Bournemouth, United Kingdom; Arcispedale Santa Maria Nuova, Reggio Emilia, Italy; Hospital Duran i Reynals, Llobregat, Spain; Val d’Hebron University Hospital, Barcelona, Spain; Royal Surrey County Hospital, Guildford, United Kingdom; Azienda Ospedaliera, Verona, Italy; Southern Medical Day Care Center, Wollongong, Australia; Tel Aviv Medical Center, Tel Aviv, Israel; Hopital Tenon, Paris, France; Hopital Saint Antoine, Paris, France

Background: Currently, stage II (Dukes B2) CC patients (pts) are included in clinical trials testing adjuvant chemotherapy as benefit of such treatment remains to be confirmed for these pts. Methods: FOLFOX4 [oxaliplatin/5-Fluorouracil (5FU)/leucovorin (LV)] is the first combination which improves 3-year disease free survival (DFS) (78.2%) when compared with 5-FU/LV (LV5FU2) alone (72.9%) in stage II/III CC as demonstrated in a large phase III randomized study (MOSAIC). The decrease in the risk of recurrence (RRR) is 23% in the overall population (p=0.002) (de Gramont, ASCO 2003). Stage II sub population data are presented here. Results: 40% of the 2246 patients were stage II: 451 in FOLFOX4, and 448 in LV5FU2; main characteristics were well balanced between the 2 groups with respectively for FOLFOX4/LV5FU2 (%): Male 57/52, median age (years) 61/60, colon 57/56, sigmoid 32/34, recto-sigmoid 11/9, KPS ≥ 80%: 88/ 90, Bowel obstruction: 16/19, Tumor perforation: 8/10, T4 (%)19/19, venous invasion: 8/10.

Number of treated patients Median N° of cycles received/planned Patients with any serious adverse events (SAE)* TOTAL FOLFOX4 LV5FU2 Treatment discontinuation for toxicity* TOTAL FOLFOX4 LV5FU2 Death on treatment* TOTAL FOLFOX4 LV5FU2 % RRR

Stage II (N = 899)

Stage III (N = 1347)

891 12/12 144 (16.2) 84 (18.8) 60 (13.5)

1328 11/12 270 (20.3) 168 (25.4) 102 (15.3)

80 (9.0) 54 (12.1) 26 (5.8)

141 (10.6) 106 (16) 35 (5.3)

4 (0.4) 1 (0.2) 3 (0.7)

8 (0.6) 5 (0.8) 3 (0.5)

20%

24%

The observed reduced risk of recurrence with FOLFOX4 was consistent in both stage II and stage III as shown by a non-significant treatment-by-stage interaction (p=0.77). Overall per patient major safety events on study [incidence of SAE, treatment discontinuation and/or death] were less frequent in stage II (19.8%) than in stage III (25.6%). Conclusion: In the FOLFOX4 arm of MOSAIC, stage II patients benefited from a 20% relative reduction in the risk of recurrence versus those of the LV5FU2 arm with a limited incidence of major safety events. This reflects the favorable benefit-risk of FOLFOX4 in this population.

Abstracts

Gastrointestinal Cancer – Colorectal/Liver B. Previously Untreated Advanced Colorectal Cancer 3503

A case for time to tumor progression (TTP) as the primary (1o) efficacy endpoint in 1st-line metastatic colorectal cancer (MCRC) therapy: Correlation of TTP and overall survival (OS) L. L. Miller, G. L. Elfring, G. Gruia, S. Hirawat, J. Y. Douillard, L. B. Saltz PTC Therapeutics, South Plainfield, NJ; Elfring Statistical Consulting LLC, Wilmington, NC; Pfizer Corporation, New York, NY; Centre R Gauducheau, Saint Herblain, France; Memorial Sloan-Kettering Cancer Center, New York, NY

Background: OS has been the 1o endpoint for phase III MCRC registration trials. However, increasing numbers of efficacious agents have extended OS and added therapeutic complexity. Evaluation of new drugs based on OS may be impeded by larger sample sizes, longer study durations, and crossovers. Thus, FDA, ASCO and AACR, in a review of surrogate and clinical benefit endpoints, have asked if TTP could replace OS as the 1o endpoint because it measures an early treatment effect and is not confounded by later therapies. Methods: As part of this effort, we assessed the relationship of TTP to OS in the 1o patient data from the 2 large, FDA-reviewed phase III trials (NEJM, 343:905, 2001; Lancet 355:1041, 2001) that form the basis of irinotecan/5-FU/leucovorin (IFL) approval in previously untreated MCRC. Patients (pts) were randomized to: Study 1 (bolus IFL or FL or I) and Study 2 (infusional IFL or FL) and could receive 2nd-line I or FL; 2nd-line oxaliplatin was not widely available. Results: The linear regression of OS vs TTP (mo) was OS=1.2xTTP+8.3 (r=.56), showing an ≈1:1 correlation between TTP and OS improvement and that pts lived an average of ≈8 more mo after progression. The OS:TTP relationship (slope) was not altered by treatment or prognostic factors (PS, LDH), but post-study survival (y-intercept) was affected. Factor

Category

N

Slope

y-intercept

Total Treatment

All pts Bolus IFL Bolus FL Bolus I Infusional IFL Infusional FL 0 1 2 > Normal £ Normal

1068 231 226 226 198 187 484 471 109 440 480

1.2 1.3 1.3 1.1 1.1 1.0 0.9 1.2 1.3 1.0 1.3

8.3 6.5 7.1 8.7 10.0 9.8 11.8 7.2 3.1 11.2 6.1

PS

LDH

3512

An international phase II study of cetuximab in combination with oxaliplatin/5-fluorouracil (5-FU)/folinic acid (FA) (FOLFOX-4) in the first-line treatment of patients with metastatic colorectal cancer (CRC) expressing Epidermal Growth Factor Receptor (EGFR). Preliminary results J. M. Tabernero, E. Van Cutsem, J. Sastre, A. Cervantes, J.-L. Van Laethem, Y. Humblet, P. Soulié, S. Corretgé, M. Mueser, A. De Gramont Vall d’Hebron University Hospital, Barcelona, Spain; University Hospital Gasthuisberg, Leuven, Belgium; Hospital Clínico San Carlos, Madrid, Spain; Hospital Clínico Universitario, Valencia, Spain; Erasme Hospital, Brussels, Belgium; Cliniques Universitaires St. Luc, Brussels, Belgium; Centre Paul Papin, Angers, France; Merck Farma y Química, SA, Barcelona, Spain; Merck KGaA, Darmstadt, Germany; Hôpital Saint-Antoine, Paris, France

Background: Cetuximab (Erbitux) is an IgG1 monoclonal antibody targeting the EGFR, shown to be effective in patients (pts) with EGFR-expressing metastatic CRC refractory to prior irinotecanbased chemotherapy (Cunningham ASCO 2003). Cetuximab has demonstrated high efficacy with an acceptable safety profile in combination with different regimens of irinotecan/FA/5-FU as 1st-line treatment (Rosenberg ASCO 2002; Van Laethem and Folprecht ASCO 2003). This trial was designed to evaluate the efficacy and safety of cetuximab with the FOLFOX-4 regimen as 1st-line treatment. Methods: Pts received cetuximab (400 mg/m2 week 1 and 250 mg/m2 weekly thereafter) plus FOLFOX-4 (every 2 weeks: oxaliplatin 85 mg/m2, day 1; FA 200 mg/m2 IV 2h and 5-FU 400 mg/m2 IV bolus followed by 600 mg/m2 IV for 22h, days 1 and 2) until PD or unacceptable toxicity. Main inclusion criteria included pts with non-resectable metastatic CRC, EGFR expression on tumor, no previous chemotherapy for metastatic CRC, and ECOG PS = 2. Results: Of the 61 pts screened 50 (82%) showed EGFR expression on the tumor. 43 pts have finally been enrolled. Currently, baseline data from 25 enrolled pts are available: M/F 16/9, median age 65 years (range 43–75), colon/rectum 17/8, ECOG PS=0/1 16/9, number of organs affected=1/2/3: 13/9/3. The observed best overall response was based on 20 evaluable pts: 14 (70%) partial response (of these, 7 responses have been confirmed so far), 5 (25%) stable disease, 1 not evaluable who withdrew his consent after 1st infusion. Two pts underwent curative resection of their hepatic metastasis after confirmed response. So far, 8 serious adverse events were observed (4 5-FU related diarrhea, 1 pulmonary embolism, 1 pneumonia, 1 PD with obstructive jaundice, 1 gout) none of them related to cetuximab. Conclusion: These preliminary data suggest that cetuximab may be safe and effective when combined with FOLFOX-4 in first-line treatment of patients with metastatic CRC.

3514

In a Cox regression analysis, longer OS was better explained by longer (≥6 mo) TTP (p<0.0001, hazard ratio [HR] 0.31) than by PS=0 (p<0.0001, HR 0.47) or normal LDH (p<0.0001, HR 0.51). Conclusion: An irinotecan-induced improvement in TTP translates into a similar improvement in OS, supporting the use of TTP as the 1o endpoint in phase III MCRC trials.

A phase II study of gefitinib in combination with FOLFOX-4 (IFOX) in patients with metastatic colorectal cancer G. A. Fisher, T. Kuo, C. D. Cho, J. Halsey, C. N. Jambalos, E. J. Schwartz, R. V. Robert, R. H. Advani, H. A. Wakelee Stanford University, Palo Alto, CA

Background: Gefitinib (ZD1839, Iressa) is an inhibitor of epidermal growth factor receptor-1 and has been shown to produce growth inhibition and shrinkage in human tumors. Here we report the preliminary results of a trial to evaluate the efficacy and toxicity of Gefitinib

Abstracts

Onkologie 2004;27(suppl 1):12–58

17

(I) with FOLFOX-4 in patients (pts) with advanced colorectal cancer. Methods: Each cycle consisted of 14 days. Cycle 1 consisted of FOLFOX-4 while cycle 2 and subsequent cycles consisted of FOLFOX-4 with gefitinib at 500 mg PO qd. Pts were stratified by prior therapy for metastatic disease; Group A with no prior therapy and Group B with prior therapy Results: 56 pts have been enrolled (Group A: N=32; Group B: N=24) to date. Median age 54 years (range, 37–79) and median ECOG PS 0 (range 0–1). No patient had prior oxaliplatin. 359 cycles have been administered to date. All pts were assessable for toxicity and 49 were assessable for response. Two pts withdrew prior to completing 4 cycles due to unacceptable toxicity. Grade 3/4 toxicities included neutropenia (53%), diarrhea (49%), nausea (28%), and vomiting (21%). There was one treatment-related death (E. coli sepsis). In Group A, 21 of 27 pts (78%) achieved a partial response (PR). Nine patients in Group A went on to have liver metastases resected; only four of whom were deemed to have resectable disease at study entry. In Group B, 8 of 22 pts (36%) achieved PR. 18 of 22 Group B pts (82%) had received both Irinotecan and 5-FU. Conclusions: Gefitinib with FOLFOX 4 represents an active regimen with response rates that are thus far encouraging compared with previously reported results using FOLFOX alone. Accrual continues to date.

3517

Post-progression therapy (PPT) effect on survival in AVF2107, a phase III trial of bevacizumab in first-line treatment of metastatic colorectal cancer (mCRC) E. E. Hedrick, H. Hurwitz, S. Sarkar, S. Griffing, W. Novotny, A. Grothey Genentech, Inc, South San Francisco, CA; Duke Univ., Durham, NC; Mayo Clinic, Rochester, MN

Background: Bevacizumab [Avastin (BV)], a recombinant humanized monoclonal antibody against VEGF, significantly prolongs overall survival (OS) when added to 1st-line chemotherapy for mCRC. BV joins 5-Fluorouracil/leucovorin (5-FU/LV), irinotecan,(I) and oxaliplatin (Ox) as agents with demonstrated survival benefit in the treatment of mCRC. A recent analysis of phase III trials in mCRC, where treatment with 5-FU/LV, I, and Ox was available, demonstrates a correlation between the percentage of patients receiving all 3 active agents and OS (Grothey, J Clin Oncol, in press). We analyzed the effect of use of active agents over the treatment course on OS in AVF2107. Methods: The findings of AVF2107 have been reported. In AVF2107, use and composition of PPT (Ox-containing vs non-Oxcontaining) was at the discretion of the investigator. OS was determined in the two randomized treatment arms [IFL/placebo (Arm 1, n=412) and IFL/BV (Arm 2, n=402)] according to PPT use and composition. Baseline characteristics were compared between these subgroups. Results: 231 (56.2%) and 222 (55.2%) patients received PPT in Arms 1 and 2, respectively. Table 1 summarizes OS by PPT composition. For Arm 1 patients who received PPT, the group receiving Ox PPT was younger (median age 57 vs 64 yrs), had better performance status (ECOG 0, 64 vs 52%) and a longer 1st-line PFS (7.4 vs 5.8 months) compared to those receiving non-Ox PPT. For Arm 2, differences between Ox and non-Ox PPT subgroups were not observed. Conclusion: BV prolongs survival in patients with mCRC when added to first-line chemotherapy. In addition, this subset analysis suggests that a treatment strategy incorporating all active agents over the course of disease optimizes OS. These results are consistent with an analysis of other recent phase III trials in mCRC.

Arm/Subgroup

1st-line PFS (mos)

OS (mos)

IFL/pbo (n=411) No Ox PPT (n=122) Ox PPT (n=109) IFL/BV (n=402) No Ox PPT (n=125) Ox PPT (n=97)

6.2 (a) 5.8 7.4 10.6 (a) 10.1 9.4

15.6 (b) 15.8 (c) 22.2 (c) 20.3 (b) 19.6 (d) 25.1 (d)

3519

Molecular profiling predicts clinical outcome in patients with metastatic colorectal cancer treated with 5-FU/oxaliplatin J. Yun, W. Zhang, D. Park, D. Yang, O. Press, M. Gordon, N. Mallik, H. J. Lenz University of Southern California, Los Angeles, CA

Background: Genetic polymorphisms have been shown to predict clinical outcome to 5-Fluorouracil/Oxaliplatin chemotherapy in patients with colorectal cancer. We have shown in retrospective studies on patients with metastatic colorectal cancer that polymorphisms in the DNA repair pathway genes XPD751, ERCC1, metabolizing gene GSTP1, and target gene TS5’, TS3’UTR are predictive to clinical outcome to 5-FU/Oxaliplatin. In a prospective study, we analyzed these polymorphisms along with XPD312, XPD156, and TS5’SNP to confirm identified patterns. We also analyzed the polymorphism in the COX-2 gene, which is associated with angiogenesis in colorectal cancer. Methods: 168 patients with stage IV metastatic colorectal cancer were enrolled and received 5-FU/Oxaliplatin after failing CTP-11. 130 patients (68
,62 ) with median age 60 years (25 to 87) had sufficient follow up time for analysis. Median follow up time was 16.6 months. Genomic DNA was extracted and analyzed by a PCR-based RFLP technique. Results: A univariate analysis of the genotypes showed XPD156 was significant for response to chemotherapy (p=.04). Patients with the A/A genotype in XPD156 had greater response compared to patients with the C/C genotype. GSTP1 (p=.08), TS5’ (p=.10), and TS3’UTR (p=.07) showed a trend for response. From the table below, GSTP1 and TS5’SNP were significant for survival and XPD312 showed a trend. Survival

XPD312 Asp/Asp Asp/Asn Asn/Asn GSTP1 Ile/Ile Ile/Val Val/Val TS5’SNP 2R/2R 2R/3R 3R/3R

N

Median, mo

39 66 25

10.3 8.0 10.8

60 55 14

10.0 9.0 5.7

57 60 12

7.4 9.2 13.4

P Value 0.06

0.02

0.04

Conclusion: The data confirms that genes identified retrospectively GSTP1, TS, and XDP may predict survival and response in patients with colorectal cancer who received 5-FU/Oxaliplatin. The data on

18

Onkologie 2004;27(suppl 1):12–58

Abstracts

the polymorphisms are in contrast to our retrospective study, indicating that further functional analysis is critical to understanding their role in chemosensitivity.

3521

Insulin-like growth factor-I (IGF-1), IGF binding protein-3 (IGFBP-3), and outcome in metastatic colorectal cancer (CRC): Results from Intergroup Trial N9741 C. Fuchs, M. Pollak, D. J. Sargent, J. A. Meyerhardt, R. K. Ramanathan, S. Williamson, B. Findlay, E. Green, R. M. Goldberg Dana-Farber Cancer Institute, Boston, MA; McGill University, Montreal, PQ, Canada; NCCTG, Rochester, MN; ECOG, Pittsburgh, PA; SWOG, Kansas City, KS; NCIC, St. Catherines, ON, Canada

Background: IGF-1 has been shown to stimulate neoplastic cell growth and inhibit apoptosis, whereas IGFBP-3 inhibits the bioavailability of IGF-1 and has independent proapoptotic activity. In epidemiologic studies, higher levels of plasma IGF-1 increase the risk of CRC whereas higher levels of IGFBP-3 have been associated with decreased risk. Methods: We examined plasma levels of IGF-1, IGFBP-3, and C-peptide among 527 patients participating in a randomized trial of first-line chemotherapy for metastatic CRC (Intergroup trial N9741). Three chemotherapy arms were evaluated in this multicenter study: 5-FU/leucovorin/CPT-11, 5-FU/leucovorin/oxaliplatin, and oxaliplatin/CPT-11. Data on prognostic factors and body size as well as plasma specimens were collected at baseline. Results: Controlling for IGF-1 levels, treatment arm, and other prognostic factors, increasing IGFBP-3 levels were associated with a higher response rate (RR) to chemotherapy (P = 0.01; RR = 51% vs. 34% comparing highest to lowest quartiles). Compared to subjects in the lowest quartile of IGFBP-3, the hazard ratio (HR) for tumor progression for those in the highest quartile was 0.71 (95% CI, 0.53–0.95) (median TTP = 302 vs. 218 days, comparing highest to lowest quartiles; P for trend = 0.003). Neither IGF-1 nor C-peptide were significantly related to RR or TTP after adjusting for other covariates. Moreover, there was no statistical interaction between IGFBP-3 levels and treatment assignment. In preliminary analyses, higher plasma IGFBP-3 was positively associated with overall survival with an over 6 month difference in median survival comparing highest to lowest quartiles (HR = 0.50, 95% CI, 0.35–0.73). Mature data on TTP and overall survival will be available for presentation. Conclusions: Higher baseline levels of circulating IGFBP-3 predict an improved outcome among patients with metastatic CRC. The IGF pathway may represent an important target for future treatment strategies.

3525

OPTIMOX study: FOLFOX 7/LV5FU2 compared to FOLFOX 4 in patients with advanced colorectal cancer

conventional inclusion criteria (526 pts), and two exploratory studies in pts not included in previous studies (37 pts >75 yrs, 63 pts Alk Ph >3xUNV) of a new strategy with ox stop and go: FOLFOX7 (highdose ox and 5FU infusion, 6 cy) followed by 12 cy simplified (s) LV5FU2 and later FOLFOX7 reintroduction. Methods: 526 pts ≤ 75 yrs and/or Alk Ph ≤ 3xUNL were randomized Arm A FOLFOX4 (Ox 85 mg/m2 d1, LV 200 mg/m2, 5FU bolus 400 mg/m2 and 22h 600 mg/m2 d1–2, q2w) or Arm B FOLFOX7 x 6 cy (ox 130 mg/m2 d1, LV 400 mg/m2, 5FU 46 h 2.4 g/m2, q2w) followed by sLV5FU2 x 12 cy (LV 400 mg/m2, 5FU bolus 400 mg/m2 d1 and 46-h infusion 2.4 g/m2, q2w) then FOLFOX7 reintroduction. Results: Arm A, 262 pts: M/F 58/42%, PS 0/1–2 55/45%, median age 63 yrs; Arm B, 264 pts: M/F 62/38%, PS 0/1–2 56/44%, median age 63 yrs. 3514 cy (2938 with ox, median DI 41.9mg/m2/wk) were administered in arm A and 4107 cy (2076 with ox, median DI 62.2mg/m2/wk) in arm B. Grade 3–4 toxicity (pts %) was in arm A/B: neutrophils 33.2/21.9, platelets 3.1/10.6, nausea 5.7/9.4, mucositis 3.1/6.5, diarrhea 11.1/11.8, neuro 18.7/13.3, maximal 54.6/49.2. RR was 58.8% in arm A and 59.5% in arm B. 28 pts arm A and 26 pts arm B had complete resection of metastasis. In arm A and B, median PFS were 9.2 and 9.0 months (P=0.47), median OS were 20.7 and 21.4 months (P=0.75), respectively. The primary endpoint is time of disease control (TDC=PFS first FOLFOX+PFS reintroduction if PR or SD), median TDC was 9.9 arm A and 11.3 months arm B (P=0.61). Ox was reintroduced in 8.4% pts arm A and 42% arm B. Reintroduction achieved 11.6% RR, 33.9% SD. Post study, in arm A 17.6% pts received ox-based and 58.3% irinotecan-based regimen, in arm B 13.3% received ox-based and 58.3% irinotecan-based regimens. Conclusions: The optimox stop and go strategy is a convenient alternative to FOLFOX4 in first-line therapy. Further studies will use a lower oxaliplatin dose-intensity to improve the reintroduction rate and evaluate lack of sLV5FU2 maintenance.

3526

First line infusion of 5-fluorouracil, leucovorin and oxaliplatin for metastatic colorectal cancer: 4-day chronomodulated (FFL4-10) versus 2-day FOLFOX2. A multicenter randomized Phase III trial of the Chronotherapy Group of the European Organization for Research and Treatment of Cancer (EORTC 05963) S. Giacchetti, G. Bjarnason, C. Garufi, N. Tubiana-Mathieu, S. Iacobelli, L. Dogliotti, R. Smaaland, C. Focan, B. Coudert, F. Lévi Hopital Paul Brousse and INSERM, Villejuif, France; Toronto-Sunnybrook Regional Cancer Centre, Toronto, ON, Canada; Istituto Regina Elena, Roma, Italy; Centre Hospitalier Universitaire Dupuytren, Limoges, France; Universita G d’Annunzio di Chieti, Chieti, Italy; Ospedale San Luigi, Orbassano, Italy; Haukland Hospital, Bergen, Norway; Les Cliniques Saint-Joseph, Liège, Belgium; Centre Georges-François Leclerc, Dijon, France; Hopital Paul Brousse & INSERM, Villejuif, France

Background: FOLFOX4 has shown superiority over LV5FU2 and IFL in first line therapy. Limiting toxicities are neutropenia mainly due to 5FU bolus and cumulative sensory neurotoxicity which imposes to stop oxaliplatin (ox). OPTIMOX is a phase III study in pts with

Background: The tolerability and anticancer activity of a 5-day (d) infusion of 5-fluorouracil (FU), leucovorin (LV) and oxaliplatin (lOHP) were improved with chronomodulated (CM) delivery as compared with constant rate infusion in metastatic colorectal cancer (MCC) (Lévi, Lancet Oncology, 2, 2001). We investigated whether FFL4-10 - schedule (sch) B - (Berthault et al. JCO 1996) improves survival as compared with FOLFOX2 - sch A - (De Gramont et al. EJC 1997) in MCC patients (pts). Methods: 554 pts had to be randomized to receive either sch at the same dose levels (g/m2) per course (c) q14 d: FU (c1, 3; c2, 3.4; c3: 3.6; LV, 1.2; l-OHP, 0.1. For sch A, l-OHP was infused for 2 h at daytime (d1), LV for 2 h (d1, d2) and FU for 22 h (d1, d2). For sch B, FU-LV and l-OHP were infused daily from 10 pm to 10 am (peak at 4 am) and from 10 am to 10 pm (peak

Abstracts

Onkologie 2004;27(suppl 1):12–58

A. De Gramont, A. Cervantes, T. Andre, A. Figer, G. Lledo, M. Flesch, L. Mineur, G. Russ, E. Quinaux, P.-L. Etienne Hopital Saint Antoine, Paris, France; Hospital Clínico de Valencia, Valancia, Spain; Hopital Tenon, Paris, France; Tel Aviv Sourasky Medical Center, Tel Aviv, Israel; Clinique Saint Jean, Lyon, France; Hopital Drevon, Dijon, France; Clinique Sainte Catherine, Avignon, France; Landerskrankenhaus Salzburg, Paris, Austria; IDDI, Bruxelles, Belgium; Clinique Armoricaine, St Brieuc, France

19

at 4 pm) using a programmable pump. Pts were evaluated q14 d for toxicity and q4 c for response. The primary endpoint was Overall Survival (OS), 430 events being required. Results: 564 pts (sch A/B: 282/282 pts) from 36 centers with median follow-up >3 years; sch A/B %pts: women, 40.1/40.4; Colon primary, 59.9/58.5; PS 0/1/2, 49.3/40.4/9.2 vs 47.5/40.4/11.7; single M site, 50.7/50; synchronous M, 73/74.1; adjuvant chemo, 16.7/19.1. N of courses: sch A/B: 2916/2783. Toxic deaths: 1/2. OS (Hazard Ratio = 0.94, 95% CI:0.73–1.21) and Progression Free Survival (PFS) (HR=1.01, 95% CI: 0.81–1.26) did not differ significantly. Conclusions: This is the first trial that has compared the two most active treatment schedules as 1st line chemotherapy of MCC. No difference in OS and PFS were observed with a median survival >18 months in both regimens and a different but low toxicity profile (<7% of courses). Supported in part by an educational grant from SANOFI-SYNTHELABO.

FOLFOX2 (sch A) % neutropenia gr 3–4 per pt (per c) % diarrhea gr 3–4 per pt (per c) % mucositis gr 3–4 per pt (per c) % sensory neuropathy (per pt) % ORR (% best response), based on extramural review Median PFS (months) [95% CL] Median survival (months) [95% CL]

CM FFL 4-10 (sch B)

p-value

25.0 (6.5)

7.0 (2.5)

10.8 (1.2)

28.3 (5.1)

6.8 (0.8)

13.8 (1.9)

26.6 45.3 (52.8)

26.0 42.2 (51.1)

<0.001 (<0.001) <0.001 (<0.001) 0.002 (<0.001) 0.87 0.45 (0.67)

8.3 [7.9–9.3] 18.7 [17.7–21.0]

8.3 [7.4–9.4] 19.6 [18.2–21.2]

0.88 (Logrank) 0.53 (Logrank)

Pooled Arm A and Arm B (n=38)  Overall Incidence (%) Grade 3 and 4 (%) Asthenia Neuropathy Diarrhea Nausea Vomiting Stomatitis Thromboembolism Skin Toxicity

66 66 55 71 42 18 5 55

8 0 5 3 5 0 3 3

3532

Oxaliplatin (L-OHP) combined with irinotecan (CPT-11), leucovorin (LV) and fluorouracil (5-FU) compared with irinotecan, leucovorin and fluorouracil as first-line treatment for metastatic colorectal cancer (MCC): Preliminary results of a multicenter randomized phase III trial J. Souglakos, N. Ziras, A. Polyzos, A. Athanasiadis, S. Kakolyris, T. Giannakakis, E. Tselepatiotis, K. Kalbakis, N. Vardakis, V. Georgoulias For the Colon Cancer Working Group, of the Hellenic Oncology Research Group (HORG), Greece

3531

Cetuximab plus FOLFOX for colorectal cancer (EXPLORE): Preliminary safety analysis of a randomized phase III trial S. Badarinath, E. P. Mitchell, A. Jennis, C. D. Graham, V. L. Hansen, C. A. Henderson, T. T. Chen, C. Langer Florida Oncology Associates, Jacksonville, FL; Thomas Jefferson U., Philadelphia, PA; Northern New Jersey Cancer Association, Hackensack, NJ; Charleston Cancer Center, Charleston, SC; Northern Utah Associates, Ogden, UT; Peachtree Hematology Oncology Consultants, Atlanta, GA; Bristol Myers-Squibb, New York City, NY

Background: The EXPLORE study is a randomized phase III study comparing cetuximab (Erbitux) plus FOLFOX4 (5-FU, leucovorin, oxaliplatin), to FOLFOX4 in 2nd-line metastatic, EGFR-positive colorectal cancer (CRC) patients (pts). Since this study is the first experience combining cetuximab with FOLFOX4, an early safety assessment was performed. Methods: The first stage of this randomized phase III trial restricted accrual to 40 pts at 10 centers to obtain an early assessment of safety. Pts with metastatic EGFR-positive CRC who had progressed on prior first-line irinotecan therapy with an ECOG performance status ≤2 were randomized to either Arm A (cetuximab 400 mg/m2 initial dose followed by 250 mg/m2 weekly dose and FOLFOX4 d1, d2 q 2 weeks) or to Arm B (FOLFOX4 q 2 weeks). A pooled analysis of safety is presented. Results: Forty patients were randomized from March to November, 2003. They included 20 women and 20 men with a mean age of 61 years. Two patients

20

did not receive study therapy and are excluded from the analysis. A total of 178 cycles were administered to the 38 pts with a median of 3 cycles per pt (range 1–15). There have been 4 disease-related deaths and one severe non-fatal hypersensitivity reaction (HSR). A summary table of adverse events associated with chemotherapy and cetuximab is below. The incidence and severity of these events appear comparable to previous reports of cetuximab with chemotherapy or of FOLFOX4. Conclusions: In this pooled analysis the characteristic toxicities of cetuximab and FOLFOX4 do not appear to be increased. The study is continuing to accrue to its target of 1100 patients.

Onkologie 2004;27(suppl 1):12–58

Background: In a phase II trial conducted from our group the 4-drug regimen (5FU+LV+CPT-11+L-OHP) was highly active and well tolerated (JCO 2002;20:2661–2667). A randomized trial was conducted to compare in terms of overall survival (OS) the 4-drug regimen (Arm-A) with the ”standard” 3-drug (5FU+LV+CPT-11) FOLFIRI regimen (Arm-B) as first-line treatment in MCC Methods: Assuming OS 17months for Arm-B and 21 months for Arm-A, 276 pts were needed to show a significant difference (two-tailed log-rank test; a=0.05, b=0.8). For Arm-A CPT-11 150 mg/m2 30–90 min infusion was given on d1, L-OHP 65 mg/m2 2h infusion on d2, simultaneously but in different lines with LV 200 mg/m2 on days 2+3 followed by 5-FU 400 mg/m2/d iv bolus and 600 mg/m2/d 22h infusion on days 2+3. For Arm-B CPT-11 180 mg/m2 30–90 min infusion was given on d1, LV 200 mg/m2 on days 1+2 followed by 5-FU 400 mg/m2/d iv bolus and 600 mg/m2/d 22h infusion on days 1+2. Cycles were repeated every 2 weeks until disease progression, unacceptable toxicity or consent withdrawal. Results: As of today 203 pts and 1673 cy of therapy were evaluable for safety and efficacy. 101 pts on Arm-A: M/F: 62/39, PS 0/1/2: 59/30/12, median age 65 (25–75).102 pts on Arm-B: M/F: 57/45, PS 0/1/2: 47/39/16, median age 66 (42–80). In an intention-to-treat analysis the response rate was 45% (95%CI 35.45–55.27%) in Arm-A and 31% (95%CI 22.68–41.24%) in Arm-B (p=0.055). In addition 32% of pts presented stable disease in Arm-A and 23% in Arm-B (p=0.147). Disease progression was observed in 23% of pts in Arm-A and 45% in Arm-B(p=0.001). Median TTP was 8.9 and 6.1 months (p=0.0958) and OS 21.1 and 16.5 months in Arm-A and Arm-B, re-

Abstracts

spectively (Log rank=0.0956). Conclusions: These preliminary results show promising efficacy with acceptable safety for the 4-drug combination. Updated data will be presented.

Patients with grade 3–4

ARM-A

ARM-B

GRADE 3

GRADE 4

GRADE 3

GRADE 4

25(24.8%)

10(9.9%)

16(15.7%)

11(10.8%)

toxicity (NCI-CTC)

Neutropenia Febrile neutropenia

12(11.8%)

Toxic Deaths

5(4.8%)

1(1%)

2(2%)

Diarrhea

20(19.8%)

6(5.9%)

9(8.8%)

Neurotoxicity

2(2%)

2(2%)

1(1%)

4(3.9%) –

Anemia

3(3%)

–

–

–

Thrombocytopenia

1(1%)

–

4(3.9%)

2(2%)

Fatigue

5(4.9%)

–

5(4.9%)

2(2%)

Toxicities (gr. 3/4) (NCI CTC)

FOLFOX n=36

bFOL n=38

Nausea Vomiting Dehydration Diarrhea Neutropenia Hand-foot syndrome Neurotoxicity (gr. 3 only) Number (%) withdrawn due to toxicities

3 (8.3) 2 (5.6) 2 (5.6) 6 (16.7) 10 (27.8) 3 (8.3)

3 (7.9) 2 (5.3) 4 (10.5) 7 (18.4) 4 (10.5) 1 (2.6)

4 (11.1) 2 (5.6%)

3 (7.9) 6 (15.8%)

CapeOx n=32

5 (15.6) 5 (15.6) 6 (18.8) 7 (21.9) 2 (6.3) 5 (15.6) 4 (12.5) 11 (34.4%) p=0.004

3539

3537

Preliminary results of a randomized study of the safety and tolerability of three oxaliplatin-based regimens as first-line treatment for advanced colorectal cancer (CRC) (”Tree” study) L. Welles, H. Hochster, R. Ramanathan, L. Wong, L. Hart, A. Shpilsky, G. Jirau-Lucca, D. Emanuel Sanofi-Synthelabo, Inc, New York, NY; New York University Kaplan Cancer Center, New York, NY; University of Pittsburgh Hillman Cancer Institute, Pittsburgh, PA; Center for Cancer Prevention & Care, Temple, TX; Florida Cancer Specialists, Fort Myers, FL

Background: Oxaliplatin (Eloxatin)/fluoropyrimidine (O/F) combinations are highly effective in the treatment of CRC, but there is debate about the merits of different fluoropyrimidine (F) administration schedules & formulations. This randomized, multicenter study assessed the safety & tolerability of three O/F regimens (bolus, infusional and oral). Methods: Eligibility: Age ≥18, measurable metastatic CRC (mCRC), no prior chemotherapy for mCRC, ECOG performance status ≤1. The primary study endpoint is the incidence of Grade (gr.) 3/4 toxicities (tox) on each arm; secondary endpoints include standard efficacy measures. Regimens: FOLFOX: O 85 mg/m2, leucovorin (LV)350 mg, 5-FU bolus 400 mg/m2 and infusional 2400 mg/m2 over 46 hours, every (q) 2 weeks (wks); bFOL: O 85 mg/m2 days 1&15, LV 20 mg/m2 and bolus 5-FU 500 mg/m2 days (d) 1,8,15, q 4 wks; CapeOx: O 130 mg/m2 d 1, Capecitabine 1000 mg/m2 orally twice daily for 14 d q 3 weeks. Treatment is continued until disease progression or unacceptable tox. Results: Safety data are available for 106 of 150 patients (pts). Overall incidence of gr.3/4 tox was similar on FOLFOX and bFOL, with more gr. 3 neutropenia (n) on FOLFOX (but only 1 episode febrile n), & more withdrawals on bFOL (p=ns). CapeOx had a higher incidence of gr. 3/4 non-hematologic tox, & more pts required study discontinuation due to tox (p=0.004, Fisher’s exact test). A protocol Amendment with bevacizumab added to each regimen, reduced Cape dose, & planned accrual of 210 pts has begun enrollment. Data are not yet available. Conclusions: Preliminary data indicate that the 3 O/F regimens are well tolerated. More patients on CapeOx arm were withdrawn for tox, & cape dose has been reduced by 15%.

Abstracts

Phase III study of bolus 5-fluorouracil (5-FU)/folinic acid vs high dose 24h 5-FU infusion/folinic acid (FA) + oxaliplatin (OXA) in metastatic colorectal cancer (MCRC) G. Hospers, M. Schaapveld University Hospital Groningen, Groningen, Netherlands; Comprehensive Cancer Center North Netherlands, Groningen, Netherlands

Background: Oxaliplatin in combination with 5-FU and FA has demonstrated improved outcome as first line and second line treatment in patients (pts) with MCRC. Methods: From July 1999 till August 2002, 302 pts with MCRC were randomized in a phase III study and received as first line treatment bolus 5-FU 425 mg/m2 day 1–5, FA 20 mg/m2 day 1–5, q 4 wk (MA) or OXA 85 mg/m2, 2h-infusion, FA 200 mg/m2, 1h-infusion, 5-FU 2600 mg/m2, 24h-infusion day 1, q 2 wk (OX). The primary endpoints were response rate (RR) and toxicity. Courses were to be continued to progression. Results: 300 pts have been randomized and analyzed for toxicity and efficacy. Treatment arms were balanced for age, sex, PS, tumor sites, prior adjuvant chemotherapy. Number of courses (median) MA vs OX: 6 vs 8. The median follow up is 15.8 month, 70% of the pts have died. An updated analysis will be presented at the meeting. Table 1 shows the grade 3/4 toxicity and efficacy data. In the MA arm there was one TD due to neutropenic septicaemia. Immunological events occurred only in the OX arm; 2 classical anaphylaxis, 5 atypical with; hemolysis, thrombocytopenia, renal and pulmonary dysfunction, with one TD. During treatment with OX, fever (>38 C) without leucopenia of infection, occurred more often than in the MA arm (odds ratio 3.048). Second line treatment was not specified in the protocol, cross-over of MA to OX occurred in 13%. Conclusions: OX in the first line resulted in an increased RR and PFS with less grade 3/4 mucositis/diarrhea compared with MA. Immunological side effects deserve attention with OX. Despite a low treatment cross-over rate, OS in both groups was comparable. (Supported by Sanofi-Synthelabo).

Stomatitis/diarrhea Vomiting Leuco/thrombocytopenia Immunological Sensory neuropathy Toxic Death (TD) RR (% CR and PR confirmed) Med. Progr. Free Survival (PFS) Med. Overall Survival (OS)

Onkologie 2004;27(suppl 1):12–58

MA (%)

OX (%)

p

22 5 8 0 0 0.7 14 5.6 months 13.5 months

6 4 7 5 10 0.7 31 6.6 months 14.1 months

<0.001 0.748 0.828 0.015 <0.001 – <0.001 0.018 0.721

21

3545

3546

UFT/LV combination based regimens with oxaliplatin or irinotecan as first line treatment for patients (pts) with non resectable metastatic colorectal cancer(MCRC): Results of two multicentric phase II trials

Phase III trial of infusional 5-fluorouracil/folinic acid plus oxaliplatin (FUFOX) versus capecitabine plus oxaliplatin (CAPOX) as first line treatment in advanced colorectal carcinoma (ACRC): Results of an interim safety analysis

J.-Y. Douillard, J. Bennouna, P. Artru, H. Perrier, B. Paillot, F. Husseini, F. Priou, K. Imadalou, J. P. Delord, R. Bugat Centre R Gauducheau, Saint Herblain, France; Clinique Saint Jean, Lyon, France; Hopital Saint Joseph, Marseille, France; CHU Charles Nicolle, Rouen, France; Hôpitaux civils, Colmar, France; Centre Hospitalier Departemental, La Roche sur Yon, France; Bristol-Myers squibb Compagny, Rueil-Malmaison, France; Institut Claudius Regaud, Toulouse, France

Background: UFT®/LV has demonstrated comparable efficacy to 5-FU/LV iv regimens with clinical safety advantages [Douillard, Carmichael JCO 2002]. 5-FU based regimens containing irinotecan or oxaliplatin are the standard of care for pts with first line MCRC [Douillard, lancet 2000, de Gramont JCO 2000]. UFT®/LV is an orally available treatment, and is a potentially valuable alternative to infusional regimens. Based on these results, we conducted two phase II trials of UFT®/LV in combination with oxaliplatin and irinotecan. Methods: Eligibility criteria were similar in both studies. Pts with measurable non-resectable MCRC were included to evaluate efficacy (Objective Responses Rate) and safety of UFT®/LV when combined with oxaliplatin (Tegafox trial) and irinotecan (Tegafiri trial) as first line treatment of MCRC. UFT® was administered at 300 mg/m2/day d1–14, LV 90 mg/day d1–14 with oxaliplatin 130 mg/m2/day d1 every 3 wks and at 250 mg/m2/day d1–14, LV 90 mg/day d1–14 with irinotecan 250 mg/m2 d1 every 3 wks. Results: A total of 64 and 56 pts were enrolled in Tegafox and Tegafiri, respectively. Patients characteristics: Tegafox: median age: 68 y (range: 38–82); ECOG PS 0/1(%): 63/37; colon 77%, rectum 23%; liver metastasis: 65%; prior (neo)adjuvant CT: 27%. Tegafiri: median age: 66 y (range: 42–88); ECOG PS 0.1(%): 57/41; colon 68%, rectum 30%; liver metastasis: 50%; prior (neo)adjuvant CT: 27%. Safety and efficacy results as follows:

Median nb of cycle (range) G3/4 related AEs (% pts) neutropenia diarrhea vomiting sensory neuropathy Efficacy parameters evaluable pts ORR% 95 CI TTP (mo) % 95CI

Tegafox, n=64

Tegafiri, n=56

6 (1–14)

5 (1–10)

12 11* 8 16

22 22 11 0

n=58 34[22–47] 5.9[4.34; 8.21]

n=49 34[21–48] 5.7[4.44;5.98]

* no grade 4

22

Onkologie 2004;27(suppl 1):12–58

H.-T. Arkenau, H. Schmoll, S. Kubicka, T. Seufferlein, P. Reichardt, W. Freier, U. Graeven, A. Grothey, R. Porschen Central Hospital Bremen East, Bremen, Germany; Martin-LutherUniversity-Halle, Halle, Germany; Medical School Hanover, Hanover, Germany; University Clinic Ulm, Ulm, Germany; Charité Berlin, Berlin, Germany; Private Oncology Clinic Hildesheim, Hildesheim, Germany; Ruhr-University-Bochum, Bochum, Germany; Mayo Clinic, Rochester, MN

Background: In a previous phase III study, the FUFOX regimen has shown superior response rates and progression-free survival compared with bolus 5-FU/FA (Mayo Clinic protocol) in patients with ACRC (Grothey, ASCO 2002). The combination of capecitabine (CAP) and oxaliplatin (OX) has demonstrated good efficacy and safety results in recent phase II studies. In August 2002 we initiated a phase III trial to compare FUFOX and CAPOX as first line therapy in patients with ACRC. Here, we present the results of a planned interim safety analysis. Patients and Methods: From August 2002 to October 2003 275 (of target 420) patients (m:f = 176:99; median age 66 (range 32–85)) have been randomized to receive either FUFOX (135 pts. arm A: 5-fluorouracil 2000 mg/m2 24h infusion, folinic acid 500 mg/m2, oxaliplatin 50 mg/m2 d1,8,15,22; q5 wks) or CAPOX (140 pts. arm B: capecitabine 1000 mg/m2 bid d1–14, oxaliplatin 70 mg/m2 d1 and 8; q3 wks). All patients had measurable metastatic disease, ECOG performance status 0–2, normal renal and hepatic function. Results: To date 783 treatment cycles (320 FUFOX, 463 CAPOX) are evaluable (median number of cycles per patient: arm A: 4, range 1–8; arm B: 4, range 1–12. Seven deaths occurred within the first 60 days (3 (2.2%) in arm A: 1x bowel obstruction, 1x severe diarrhea, 1x unknown cause) and 4 (2.8%) in arm B: 1x stroke, 1x bowel obstruction, 1x severe diarrhea and sepsis, 1x unknown cause). Major toxicities grades 2–4 (NCI-CTC) are listed in the table below. Conclusions: These data show that both FUFOX and CAPOX have a comparable toxicity profile as first line therapy in ACRC and are both well tolerated. Updated toxicity results will be reported at the meeting. At that time, the planned enrollment of 420 patients will have been completed.

NCI-CTC (grade) Toxicity Arm A/B (% of pts.) Diarrhea Hand-foot syndrome Stomatitis Thrombocytopenia Leukopenia Neutropenia Neurotoxicity

2 15/21 6/11 2/3 5/9 5/10 13/6 14/5

Abstracts

3 18/21 0/0 2/0 1/4 0/1 5/1 4/5

4 2/1 – 0/1 0/0 0/0 6/3 0/0

3554

3555

Phase II study of an optimized 5FU-oxaliplatin strategy (OPTIMOX2) with celecoxib in metastatic colorectal cancer: A GERCOR study

XELOX (capecitabine plus oxaliplatin), a safe and active first-line regimen for elderly patients with metastatic colorectal cancer (MCRC): Post-hoc analysis of a large phase II study

T. Andre, F. Maindrault-Goebel, L. Mineur, F. Michel, M. Mabro, G. Ganem, M. Hebbar, D. Avenin, R. Moukhtar, A. De Gramont Hopital Tenon, Paris, France; Hopital Saint Antoine, Paris, France; Clinique Sainte Catherine, Avignon, France; Hopital Drevon, Dijon, France; Hopital Foch, Suresnes, France; Clinique Victor Hugo, Le Mans, France; CHRU, Lilles, France; GERCOR, Paris, France

Background: In the OPTIMOX study (André et al, ASCO 2003), 6 cycles of Folfox7 followed by simplified LV5FU2 maintenance were administered to decrease the neurotoxicity and to later allow FOLFOX7 reintroduction. Celecoxib is an anti-cox2 compound with antineoplastic properties such as angiogenic inhibition and apoptosis induction. We evaluated a new strategy exploring 1) the stop and go procedure (6 cycles of Folfox7 then stop and reintroduction Folfox7 at progression); 2) the addition of celecoxib during Folfox7 and chemotherapy-free interval (CFI). Methods: Patients (pts) with non resectable metastatic colorectal adenocarcinoma were treated in firstline by Folfox7: oxaliplatin 130 mg/m2 day (d1), and folinic acid 400 mg/m2 d1, and 5FU 46h 2400 mg/m2 every two weeks for 6 cycles. Chemotherapy was then interrupted and reintroduced at progression. Celecoxib 400 mg BID per os was administered during Folfox7 and CFI intervals until Folfox7 discontinuity for progression or limiting toxicity. 44 pts included, 42 eligible: Performans status (%) 0/1/2=45/43/12, median age 60 [31–76]. Half of patients had one metastatic site. Results: 42 pts are evaluable for safety and 41 for response. Grade 3–4 toxicity per pts (%) during the 6 first cycles of Folfox7 + celecoxib were neutropenia 5%, thrombocytopenia 17%, diarrhea 5%, nausea-vomiting 5%, neuropathy 2%, (maximal gr 3–4 toxicity 29%). During follow-up period (median 37 weeks), 2 pts presented grade 1–2 epigastralgia without ulcers and 1 digestive hemorrhage with gastritis. 1 CR, 15 PR (RR 39%; 95% CI: 24–54%), 18 SD (44%) and 7 PD (17%) were obtained (intent to treat). Median PFS was 28 weeks. 28/42 had a CFI (14 pts not evaluable for CFI: 5 metastasis resection after 3 months of therapy, 7 PD at 3 months of therapy, 2 out of protocol for toxicity). Median CFI duration was 72 days (range 29–147+) with 43% with PD after 2 months of CFI. Conclusions: In this phase II study, celecoxib combined with FOLFOX 7 is associated with a good safety and a response rate of only 39%. Pharmacokinetics and preclinical studies are needed to investigate potential interaction between celecoxib and FOLFOX 7 drugs.

C. Twelves, C. Butts, J. Cassidy, T. Conroy, F. De Braud, E. Díaz-Rubio, J. Tabernero, P. Schöffski, A. Figer, E. Van Cutsem, On Behalf Of The Xelox Investigators Tom Connors Cancer Centre, University of Bradford, Bradford, United Kingdom; Cross Cancer Institute, Edmonton, AB, Canada; Glasgow University, Glasgow, United Kingdom; Centre Alexis Vautrin, Vandoeuvre Les Nancy, France; European Institute of Oncology, Milan, Italy; Hospital Clínico Universitario San Carlos, Madrid, Spain; Hospital General d’Hebrón, Barcelona, Spain; Medizinische Hochschule Hannover, Hannover, Germany; Tel-Aviv Medical Center, Tel-Aviv, Israel; University Hospital Gasthuisberg, Leuven, Belgium

Background: Tumor-activated capecitabine (Xeloda) has superior response rates and equivalent progression-free and overall survival compared with bolus 5-FU/LV in first-line MCRC, with favorable safety and fewer hospitalizations. This safety advantage is confirmed in the adjuvant setting, where improved safety is also seen in older patients. XELOX is safe and active in first-line MCRC and could also be suitable for older patients. Methods: In this post-hoc analysis of mature safety data in older vs younger patients, 96 patients with MCRC received first-line XELOX: oxaliplatin 130 mg/m2 i.v. day 1 followed by oral capecitabine 1,000 mg/m2 twice daily, evening day 1– morning day 15, every 3 weeks. All patients are evaluable: 64% male; median age 64 years (34–79); 52 younger patients (<65 years), 44 older patients (≥65 years); median KPS 100 (80–100). Primary tumor: 64% colon; 33% rectal; 3% both. Metastases: 54% >1 metastatic site; 77% liver, 38% lymph nodes, 32% lung. Prior adjuvant fluoropyrimidines: 28%. Results: Patients received a median nine cycles of XELOX. Safety profiles are similar in younger and older patients (table). The incidence of dose reductions and withdrawals for adverse events was similar. The activity of XELOX was also similar in both groups, response rates: 58% (95% CI, 43–71; younger) vs 52% (95% CI, 37–68; older).

Related grade 3/4 adverse events

<65 years n=52  n %

≥65 years n=44  n %

All Diarrhea Nausea Vomiting Hand-foot syndrome Neutropenia Thrombocytopenia Hyperbilirubinemia

33 7 4 6 3 4 3 2

26 8 2 5 0 3 1 2

64 14 8 12 6 8 6 4

59 18 5 11 0 7 2 5

Conclusions: XELOX has a favorable safety profile with no clinically relevant differences between younger and older patients. Antitumor activity is preserved in older vs younger patients. These results imply that up-front dose reductions are not required when XELOX is prescribed in older patients. Ongoing phase II/III trials are prospectively evaluating XELOX in this population.

Abstracts

Onkologie 2004;27(suppl 1):12–58

23

3556

Expanded phase I/II study of PTK787/ZK 222584 (PTK/ZK), a novel, oral angiogenesis inhibitor, in combination with FOLFOX-4 as first-line treatment for patients with metastatic colorectal cancer W. P. Steward, A. Thomas, B. Morgan, B. Wiedenmann, C. Bartel, U. Vanhoefer, T. Trarbach, U. Junker, D. Laurent, D. Lebwohl Leicester Royal Infirmary, Leicester, United Kingdom; HumboldtUniversitat Berlin, Charite, Germany; University of Essen Medical Center, Essen, Germany; Schering AG, Berlin, Germany; Novartis Pharmaceuticals Corporation, East Hanover, NJ

Background: Increased angiogenesis correlates with poor prognosis in patients with metastatic colorectal cancer (CRC). PTK/ZK is a novel, oral angiogenesis inhibitor that potently inhibits all known VEGF receptor tyrosine kinases, which are important mediators of the formation of new blood vessels that contribute to tumor growth and metastasis. Methods: This phase I/II study assessed the safety and preliminary response profile of oral PTK/ZK in combination with chemotherapy. Previously untreated patients (N = 35) with advanced CRC were treated with escalating doses of PTK/ZK plus oxaliplatin/5-fluorouracil (5-FU)/leucovorin (FOLFOX4). PTK/ZK was administered once daily at doses of 500 mg (n = 4), 1,000 mg (n = 3), 1,250 mg (n = 23), 1,500 mg (n = 3), and 2,000 mg (n = 2). FOLFOX-4 was administered q 2 weeks, and consisted of oxaliplatin (85 mg/m2) infusion on day 1, leucovorin (200 mg/m2 via 2-hour infusion) on days 1 and 2, and 5-FU (400 mg/m2 bolus followed by 600 mg/m2 over 22 hours) on days 1 and 2. Results: The pharmacokinetics and toxicity profiles of both PTK/ZK and FOLFOX4 were unaffected by coadministration. PTK/ZK was well tolerated at doses ≤ 1,250 mg/day; no DLTs occurred at 1,250 mg/day in the first 6 patients evaluated for MTD. Adverse events at 1,250 mg/day included ataxia (grade [G] 3), neutropenia (G4), thrombocytopenia (G3), and 2 episodes of dizziness (G3). CNS adverse events (G3/4) were dose-limiting in 2 patients at 2,000 mg/day, and expressive dysphasia (G3) and intermittent dizziness (G3) were dose-limiting at 1,500 mg/day. Best response data by SWOG criteria for 28 evaluable patients to date showed 1 (4%) CR, 14 (50%) PR, and 9 (32%) patients had SD. For 34 evaluable patients, median PFS is 11 months (95% CI = 6.8, 12.0 months). Estimated median overall survival for 35 patients is 16.6 months (95% CI = 12.9, 21.0 months). Conclusions: The early efficacy and safety of PTK/ZK+FOLFOX4 for metastatic CRC has been encouraging, and phase III randomized controlled trials of PTK/ZK+FOLFOX4 in metastatic CRC (CONFIRM-1 & -2) are ongoing.

3561

Biweekly administration of irinotecan (CPT-11), oxaliplatin (OX) and 5-fluorouracil (5-FU) as first-line treatment of advanced or metastatic colorectal cancer (MCRC) J. J. Reina, P. Sánchez-Rovira, R. Bernabé, M. Bolaños, E. González, J. Belón, M. J. Gómez-Reina, P. Gallurt, J. A. Moreno-Nogueira Hospital Juan Ramón Jiménez, Huelva, Spain; Hospital Ciudad de Jaén, Jaén, Spain; Hospital Virgen del Rocío, Sevilla, Spain; Hospital San Pedro de Alcántara, Cáceres, Spain; Hospital Virgen de las Nieves, Granada, Spain; Clínica del Doctor Belón, Granada, Spain; Hospital Puerta del Mar, Cádiz, Spain; Hospital de Puerto Real, Cádiz, Spain

Background: Triple combination chemotherapy of CPT-11, OX, and 5-FU has been shown to be active and safe in MCRC. We design the present study based on the assumption that the biweekly administration of the three agents could improve the safety profile of this sched-

24

Onkologie 2004;27(suppl 1):12–58

ule. Secondary objectives were to determine ORR, TTP and OS achieved. Methods: Patients with histological confirmation of MCRC, measurable disease, ECOG PS ≤2 and adequate bone marrow, renal and hepatic functions were included. Previous adjuvant chemotherapy was allowed if finished at least 6 months before starting study treatment. Patients received OX (85 mg/m2, iv, D1) followed by CPT11 (150 mg/m2, iv, D1) and 5-FU (1000 mg/m2/day, ci, D1–4). Cycles were repeated every 2 weeks until progressive disease, excessive toxicity or consent withdrawal. Results: Fifty patients were enrolled (M/F, 34/16), median age 60 (30–74) and 94% had ECOG 0–1 PS. Primary tumour sites were colon (64%), rectum (34%) or both (2%). Median number of metastatic lesion sites was 1 (29% with ≥2 sites), mainly located in liver (73%), lung (27%) and lymph nodes (10%). 14% of patients received previous adjuvant chemotherapy. A total of 385 cycles (median 8, range 1–16) were administered. Median RDI was 86%, 86% and 84% for OX, CPT-11 and 5-FU, respectively. All patients were evaluable for toxicity. Main grade 3/4 toxicity per cycle was neutropenia (5%), diarrhea (5%), stomatitis (4%), asthenia/fatigue (4%) and nausea/vomiting (3%). Febrile neutropenia was observed in 1 patient and in 1 cycle. There was 1 treatment related death due to a septic shock after the first cycle. Efficacy: In 43 evaluable patients, 9 achieved CR, 22 PR, 8 SD and 4 PD, resulting in an ORR of 72% (95% CI: 59–86) and a tumour growth control (RR + SD) of 91%. With a median follow-up of 12.8 months, TTP and OS were 11.3 months (95% CI: 8.2–13.9) and 18.4 months (95% CI: 14.9–26.9), respectively. Conclusions: Biweekly administration of OX, CPT-11 5-FU is an active regimen with a manageable toxicity profile as firstline treatment in patients with advanced or metastatic CRC.

3563

Biweekly alternating FOLFOX and FOLFIRI in patients with previously untreated, advanced colorectal cancer (ACC): Updated results C. Fernandez-Martos, J. Aparicio, J. M. Vicent, I. Maestu, C. Llorca, I. Busquier, J. M. Campos, D. Asensio, R. Romero Instituto Valenciano de Oncologia, Valencia, Spain; Hospital Universitario de la Fe, Valencia, Spain; Hospital General Universitario de Valencia, Valencia, Spain; Hospital Virgen de los Lirios, Alicante, Spain; Hospital General de Elda, Alicante, Spain; Hospital Provincial de Castellon, Castellon, Spain; Hospital Arnau de Vilanova de Valencia, Valencia, Spain; Prasfarma/Almirall, Barcelona, Spain; Recerca Clinica, Barcelona, Spain

Background: Irinotecan (CPT-11) and oxaliplatin (L-OHP) are active agents in ACC. Combining each with continuous infusion (CI) 5-fluorouracil (5-FU) provides an objective response rate approaching 50%, with acceptable toxicity in first line treatment. The objective of this study was to assess the efficacy and tolerance of a protocol that alternates Gramont’s LV5FU2 schedule plus L-OHP (FOLFOX-4) with the same 5-FU regimen plus CPT-11 (FOLFIRI) on a biweekly basis. Methods: Patients with previously untreated, histologically confirmed, unresectable ACC were included if they had measurable disease; adequate bone marrow, liver, and renal functions; PS (ECOG) 0–2; life expectancy >3 months, and written informed consent. Treatment consisted of 2 hours CI folinic acid (FA) 200 mg/m2, bolus 5-FU 400 mg/m2 and 22 hours CI 5-FU 600 mg/m2, all on days 1 and 2 every 14 days. CPT-11 (180 mg/m2) and L-OHP (85 mg/m2) were given on day 1 of alternating chemotherapy administrations. 1 chemotherapy cycle (FOLFOX-FOLFIRI) takes 28 days. Response evaluation was scheduled every 3 cycles. Results: 79 patients were enrolled (42 males, 37 females). Median age was 61 years (range 39–77). On PS, patients were distributed as follows: 0 (43.8%), 1 (47.5%), and 2 (8.8%). The number of metastatic sites was 1 (39.7%) or >1 (60.3%). 745 infusions were administered in 77 patients who were evaluable for toxici-

Abstracts

ty (median 5 cycles per patient, range 1–9 cycles). Response evaluation was performed in 45 patients. 8.9% of patients showed a complete response and 51.1% showed a partial response. Disease was stabilized in 15.6% of patients and in 24.4% of patients disease progression was observed. Of the 77 patients evaluated for toxicity, 47.5% had grade 3–4 toxicity. The most common grade 3–4 adverse events were: neutropenia 28.9%, diarrhea 21.1%, pain 6.6%, leucopenia 5.3%, anemia 5.3%, asthenia 3.9%, infection 2.6%, febrile neutropenia 2.6%, nausea/vomiting 2.6%, neurotoxicity 1.3% and transaminase increase 1.3%. Conclusions: The updated results of this study suggest that this regimen is effective and shows a good safety profile.

3565

5-Fluorouracil, folinic acid and oxaliplatin (FOLFOX) in poor prognosis patients with metastatic colorectal cancer C. Tournigand, T. Andre, I. Chirivella, A. Figer, G. Lledo, M. Flesch, J. R. Mel Lorenzo, E. Achille, B. Landi, M. Hebbar Hopital Saint Antoine, Paris, France; Hopital Tenon, Paris, France; Hospital Clinico de Valencia, Valencia, Spain; Tel Aviv Sourasky Medical Center, Tel Aviv, Israel; Clinique Saint Jean, Lyon, France; Hopital Drevon, Dijon, France; Hospital Xeral Calde, Lugo, Spain; Clinique de l’Orangerie, Strasbourg, France; HEGP, Paris, France; CHRU, Lille, France

Background: High alkaline phosphatases (Alk Ph) level is an adverse prognostic factor in patients (pts) with metastatic colo-rectal cancer (CRC). Pts with Alk Ph level over 3-time the upper normal value (UNV) were excluded from previous studies. OPTIMOX trial consisted in a phase III study for pts with conventional inclusion criteria (526 pts), comparing FOLFOX4 to FOLFOX7 x 6 cycles, followed simplified LV5FU2 x 12 cycles and FOLFOX7 reintroduction; and two exploratory studies in pts >75 yrs (37 pts) and in pts with Alk Ph level >3-time the UNV (63 pts), treated according to the same regimens. Methods: This report concerns the tolerance and the efficacy observed for the 63 pts with an initial Alk Ph level over 3-time the UNV. 33 pts were treated with FOLFOX4 (arm A), and 30 with FOLFOX7 - sLV5FU2 - FOLFOX7 (arm B). Characteristics of these pts were: PS 0/1–2=26/74%, median age 63 yrs; LDH (normal / >normal) 6%/94%, metastatic site (1/ >1) 70/30%. Results: 60 pts (766 cycles) and 54 pts are evaluable for safety and response, respectively. 2 pts benefited from metastasis removal. Grade 3–4 toxicities (% of pts, arm A / arm B) were: neutrophils 15.2/10.0, platelets 9.1/3.3, hemoglobin 9.1/3.3, nausea-vomiting 0/10.0, diarrhea 6.1/6.7, neurotoxicity 9.1/10.0. Maximal toxicity per patient was grade 3–4 in 42.4% of arm A pts, and 33.4% of arm B pts. Response rate (intent-to-treat) was 55.6% (arm A 48.5%; arm B 63.3%). Median PFS was 28 weeks (arm A 29 w; arm B 28 w) and median OS was 50 weeks (arm A 61w and arm B 48 w). Conclusion: These safety and efficacy results incite to propose FOLFOX regimens (either FOLFOX4 or FOLFOX7) in poor prognosis pts with metastatic colorectal cancer.

Abstracts

3571

5-fluorouracil, folinic acid and oxaliplatin (FOLFOX) in very old patients with metastatic colorectal cancer A. Figer, N. Perez, E. Carola, T. Andre, I. Chirivella, G. Lledo, M. Flesch, F. Rivera, P. Colin, A. De Gramont Tel Aviv Sourasky Medical Center, Tel Aviv, Israel; Hôpital Saint Antoine, Paris, France; CH de Senlis, Senlis, France; Hopital Tenon, Paris, France; Hospital Clinico de Valencia, Valencia, Spain; Clinique Saint jean, Lyon, France; Hopital Drevon, Dijon, France; Hospital de Valdecilla, Santander, Spain; Clinique Courlancy, Reims, France

Background: Pts with age over 75 years are usually excluded from randomised studies. OPTIMOX trial consisted in a phase III study for pts with conventional inclusion criteria (526 pts), comparing FOLFOX4 to FOLFOX7 x 6 cycles, followed simplified LV5FU2 x 12 cycles and FOLFOX7 reintroduction; and two exploratory studies in pts >75 yrs (37 pts) and in pts with Alk Ph level >3-time the UNV (63 pts), treated according to the same regimens. Methods: This report concerns the tolerance and the efficacy observed for the 37 pts aged 75 years or older. 20 pts were treated with FOLFOX4 (arm A), and 17 with FOLFOX7 - sLV5FU2 - FOLFOX7 (arm B). Characteristics of these pts were: PS 0/1–2=49/51%, LDH (normal / >normal / unknown) 51%/27%/22%, metastatic site (1/ >1) 73/27%, alk phosp (nl/elevated/unknown) 59%/35%/6%. Results: 37 pts are evaluable for safety and 35 pts for response. Grade 3-4 toxicities (% of pts, arm A / arm B) were: neutrophils 55/24, platelets 5/0, hemoglobin 5/0, nausea-vomiting 0/12, diarrhea 5/6, neurotoxicity 20/24. Maximal toxicity per patient was grade 3-4 in 80% of arm A pts, and 59% of arm B pts. Response rate (intent-to-treat) was 59.4% (arm A 65%; arm B 53%). Median PFS was 37 weeks (arm A 33 w; arm B 41 w) and median OS was 79 weeks (arm A 65w and arm B 90w). Moreover, in OPTIMOX study, age does not appear as a prognosis factor in a multivariate analysis. Conclusion: These results indicate that efficacy is maintained with Folfox regimens in patients >75years.

3579

Phase 1/2a study of EKB-569, an irreversible inhibitor of epidermal growth factor receptor, in combination with 5-fluorouracil, leucovorin, and oxaliplatin (FOLFOX-4) in patients with advanced colorectal cancer (CRC) S. Tejpar, E. Van Cutsem, E. Gamelin, D. Machover, P. Soulie, A. Ulusakarya, S. Laurent, J. M. Vauthier, S. Quinn, C. Zacharchuk University Hospital Gasthuisberg, Leuven, Belgium; Centre Paul Pipin, Angers, France; Hopital Paul Brousse, Villejuif, France; Wyeth Research, Paris, France; Wyeth Research, Cambridge, MA

Background: EKB-569 is an orally active, low molecular weight, selective, irreversible inhibitor of the epidermal growth factor receptor tyrosine kinase. In animal models, EKB-569 in combination with different chemotherapy agents shows at least additive effects on growth inhibition of colorectal and other tumors. Methods: This open-label, dose-escalation study evaluates the maximum tolerated dose (MTD), safety, pharmacokinetics, and anti-tumor activity of EKB-569 in combination with the FOLFOX-4 regimen in patients (pts) with previously untreated advanced CRC. Pts (3–6 per cohort) were to receive EKB-569 (25, 35, 50, or 75 mg/day orally) starting on day 3 in combination with the FOLFOX-4 regimen (day 1: oxaliplatin 85 mg/m2 and leucovorin [LV] 200 mg/m2, 120 min, simultaneously, followed by 5fluorouracil [FU] 400 mg/m2, bolus, and FU 600 mg/m2, 22-h continuous infusion; day 2: LV and FU as on day 1). Dose escalation was based on safety evaluations after 2 treatment cycles (14 days/cycle).

Onkologie 2004;27(suppl 1):12–58

25

Treatment continues until evidence of disease progression or unacceptable toxicity. Results: Twenty nine pts were enrolled; enrollment is complete. Median age was 61 yrs (range, 46–70); ECOG performance status was 0: 71% and 1: 29%. Dose-limiting toxicities within the first 2 treatment cycles were grade 3 diarrhea (1 pt) and grade 3 febrile neutropenia (1 pt) in the 35-mg cohort. Thus, the MTD was 25 mg EKB-569, FOLFOX-4. For 25 pts, treatment-emergent adverse events (TEAEs), all grades, with frequency ≥30% were diarrhea (76%), paresthesia (60%), nausea (48%), and neutropenia (40%). Grade 3 or 4 TEAEs with frequency ≥5% were neutropenia (32%) and diarrhea (8%). Of 11 pts who completed ≥4 treatment cycles, 4 had partial response (PR), 6 had stable disease (2 pts for ≥6 mos, 1 pt with unconfirmed PR), and 1 had progressive disease (RECIST guidelines). Conclusions: EKB-569 at an oral dose of 25 mg in combination with the FOLFOX-4 regimen is generally well tolerated and shows promising antitumor activity in pts with previously untreated advanced CRC.

Cohort 2

12 (8:4) 57 (46–71) 10 3 2:8:2 G1–2 G3–4 12 0 11 0 9 0 11 0 11 0 5 0

11 (10:1) 59 (47–71) 11 4 3:6:2 G1–2 G3–4 4 7 5 5 4 2 6 2 7 0 3 0

3599

3585

A phase 1b dose-escalation trial of erlotinib, capecitabine and oxaliplatin in metastatic colorectal cancer (MCRC) patients J. P. Delord, P. Beale, E. Van Cutsem, S. Clarke, C. Verslype, R. Bugat, A. Rakhit, S. Fettner, U. Brennscheidt, A. Feyereislova Institut Claudius Regaud, Toulouse, France; Sydney Cancer Centre, Sydney, Australia; University Hospital Gasthuisberg, Leuven, Belgium; F. Hoffmann-La Roche, Nutley, NJ; F. Hoffmann-La Roche, Inc, Basel, Switzerland; Roche Products Ltd, Welwyn Garden City, United Kingdom

Background: Epidermal growth factor receptor (HER1/EGFR) overexpression occurs in 45–80% of CRC patients (pts). Erlotinib (Tarceva™) is a potent HER1/EGFR inhibitor. Capecitabine and oxaliplatin have proven activity in MCRC pts. This trial determined the safety, the maximum tolerated dose (MTD) and pharmacokinetics (PK) of erlotinib in combination with capecitabine and oxaliplatin. Methods: MCRC pts were previously untreated or had received firstline treatment with 5-FU ± irinotecan. Patients were to receive escalating doses of daily oral erlotinib, capecitabine (days 1–14) (100 mg + 1,650 mg/m2 [Cohort 1]; 100 mg + 2,000 mg/m2 [Cohort 2]; 150 mg + 2,000 mg/m2 [Cohort 3]) and i.v. oxaliplatin 130 mg/m2, every 21 days. Dose-limiting toxicity (DLT) was defined according to NCI-CTC criteria (2.0) in cycle 1 or 2. The MTD was the dose below which >1/3 pts had a DLT. Results: The first 6 pts in Cohort 1 had no DLTs. In Cohort 2, 2/6 pts had DLTs: one with G3 diarrhea and G3 intolerable rash; one with G3 diarrhea. Cohort 2 was expanded to 11 pts and two more pts experienced DLTs: one pt had G4 diarrhea and one pt had G3 pruritic rash. As 4/11 pts had DLTs, this cohort exceeded the definition of MTD. Thus, Cohort 1 was expanded to 12 pts and there were no DLTs in the additional pts. The most common adverse events (AEs) were diarrhea and rash. Preliminary evidence of antitumor activity was seen in both cohorts. Overall, 5 pts had a partial response (PR) and 14 had stable disease (SD). There was no evidence of PK interaction between erlotinib, capecitabine and oxaliplatin. Conclusions: These data indicate that capecitabine 1,650 mg/m2 and oxaliplatin 130 mg/m2 in combination with erlotinib 100 mg/d is the recommended dose for phase II trials in MCRC pts.

26

Number of pts (M:F) Median age (range) Prior surgery Prior chemotherapy Best response PR:SD:PD AEs No. pts with AEs GI disorders Skin General disorder Nervous system Musculoskeletal

Cohort 1

Onkologie 2004;27(suppl 1):12–58

TOMOX compared to FOLFOX4 as first-line treatment in patients (pts) with advanced colorectal cancer (ACRC): Preliminary results of a multicenter randomized phase II trial C. Grávalos, C. García-Girón, A. I. León, A. Salud, B. Esteban, I. Sevilla, J. Maurel, A. Murias, E. García-Rico, H. G. Cortés-Funes Hospital Universitario 12 de Octubre, Madrid, Spain; Hospital General Yagüe, Burgos, Spain; Fundación Jiménez Díaz, Madrid, Spain; H. Arnau de Vilanova, Lleida, Spain; Hospital General de Segovia, Segovia, Spain; Hospital Virgen de la Victoria, Málaga, Spain; Hospital Clinic i Provincial, Barcelona, Spain; Hospital Insular de las Palmas, Las Palmas, Spain; Hospital de Montepríncipe, Madrid, Spain

Background: FOLFOX-4 has shown to be safe and active (response rate (RR): 50.7%) in chemotherapy-naive treated pts (De Gramont, JCO 2000). TOMOX is a promising new combination with RR 61.9% (Douillard J, ASCO 2000). We designed a randomized phase II trial to compare the activity (RR), tolerance, time to progression (TTP), overall survival (OS), quality of life and day-hospital frequentation between the two arms. Methods: From Jan/2002 to Nov/2003, 166 pts without previous treatment for ACRC have been randomized to arm A: FOLFOX-4 or arm B: TOMOX (raltitrexed 3 mg/m2 + OXA 130 mg/m2 d1, q3w). Results: Ninety-five pts are evaluable for efficacy and toxicity. 43/95 pts are still receiving treatment. Baseline data: 56% men; median age 64; ECOG 0/1: 50/45; 38% rectal cancer; 87.4% pts had previous primary tumor resection. Median number cycles were: 6.5 (1–16) in A; 5 (1–13) in B. Efficacy: The RR were: TOMOX 39.1% (95%CI 25–54.6%) and FOLFOX-4: 49% (95%CI 34.4–64.6%). The TTP and OS were similar. Tolerance: G3–4 neutropenia was more frequent in FOLFOX-4 (39.1% vs 2%, p <0.0001). Transaminitis was more higher with TOMOX (2.1% vs 22.4%, p= 0.003). Compliance: Patients under TOMOX regimen have less frequency attendant at day-hospital for chemotherapy administration (median 12 days vs 6, p <0.0001). Conclusions: Our preliminary results suggest that FOLFOX4 and TOMOX, as administered in this trial, show similar high activity (RR, TTP and OS) as first line treatment. The grade 3–4 toxicity profile varies depending on the arm: neutropenia is more common in FOLFOX4 and transaminitis in TOMOX.

Abstracts

3600

Effect of celecoxib on neurotoxicity in patients with metastatic colorectal cancer treated with 5-FU/oxaliplatin (CIFOX) R. D. Agafitei, S. Schneider, S. Iqbal, D. Yang, S. Groshen, H. J. Lenz University of Southern California, Los Angeles, CA

Background: Neuropathy is the dose limiting toxicity for patients with metastatic colorectal cancer treated with 5-FU/oxaliplatin. 5FU/oxaliplatin was administered on a compassionate use protocol at USC/Norris Cancer Center from November 1998 until May 2001. Methods: The majority of patients (84%) received continuous infusion 5-FU 200 mg/m2/d for 10 weeks with a 2 week break and oxaliplatin 130 mg/m2 every 3 weeks (CIFOX). In October 2000, celecoxib 200 mg bid was added to the regimen, because of the emerging preclinical data revealing a potential increase in efficacy and decrease in side effects. We reviewed the reported neurologic toxicities and compared patients who received celecoxib to those who did not. Results: A total of 263 patients were entered onto the trial; 11 were excluded due to lack of toxicity data. 73 patients received celecoxib and 179 did not. Patients on both regimens had similar characteristics. Of the 73 patients who received celecoxib, there were 46(63%) males and 27(37%) females; 12(16%) Asian, 6(8%) African American, 48(66%) Caucasian, and 7(9%) Hispanic; 54(74%) colon cancer and 19(26%) rectal cancer; median age was 61 (range 26–80) years. Of the 179 patients who did not receive celecoxib, there were 120(67%) males and 59(33%) females; 20(11%) Asian, 5(3%) African American, 132 (74%) Caucasian, and 22(12%) Hispanic; 153(85%) colon cancer and 26(14%) rectal cancer; median was age 60 (range 24–83). The definition of neuropathy included cold sensitivity, numbness/tingling, paresthesia and jaw discomfort. None of the 73 patients who received celecoxib experienced any grade 3 or 4 neurotoxicity, compared to 10 of the 179 patients who did not receive celecoxib (p=0.024). The patients who received celecoxib were less likely to experience grade 3 or 4 neuropathy, (0% vs. 6%) but more likely to experience grade 1 neuropathy (67% vs. 48%) than those who did not receive celecoxib. Conclusions: These findings are compatible with the hypothesis that celecoxib, when added to 5-FU/oxaliplatin, may reduce the neuropathy associated with this regimen. Prospective clinical trials should be done to confirm these results.

20/5FU bolus 400 weekly x 4, q 6 week. Results: Pts/arm were: FOLFOX-154, R-IFL-151. 60 day all-cause mortality was FOLFOX 2.0%, R-IFL 2.7%. Common grade ≥3 toxicities are below. With median follow-up of 18.0 months, median time to progression (TTP) (primary endpoint) for FOLFOX is significantly better than for R-IFL: 10.1 vs 6.5 months (mo) respectively (Logrank p<0.0001). The median survival (OS) is significantly greater for FOLFOX than R-IFL: 20.5 vs 16.3 mo (p = 0.026). Confirmed response rates (RR) were improved for FOLFOX vs. R-IFL (47% vs 32%, p=0.007). Sixty-five percent of pts received second line Rx with 75% in each arm receiving the agent not used in their first line treatment (in contrast to earlier phases of the study when access to Oxal was limited by availability). Conclusions: R-IFL lessens the toxicity burden of IFL, and results in efficacy similar to full dose IFL (median TTP 6.9 mo, RR 32%, median OS 15 months (Goldberg, JCO 2004)). FOLFOX improves RR, TTP, and OS compared to R-IFL. These results suggest that R-IFL, while not the best primary treatment option, retains activity with less toxicity than IFL, and may have utility as a platform for adding targeted therapies to IFL. Supported by NIH Grant CA25224, Pharmacia and Sanofi-Synthelabo. Table. Grade 3 or Worse Toxicity R-

FOL- P-

R-

FOL-

P-

IFL

FOX

IFL

FOX

Value

Value

Nausea

10%

6%

0.20

Neutropenia

35%

59%

<0.001

Vomiting

8%

6%

0.49

F.Neutropenia

7%

12%

0.16

Dehydration

6%

4%

0.60

Paresthesia

1%

14%

<0.001

Diarrhea

17%

11%

0.12

Infection

1%

5%

0.03

3624

First line tailored chemotherapy in advanced colorectal cancer (ACRC) with 5-fluouracil/leucovorin (5FU/LV) or oxaliplatin/irinotecan chosen by the expression of thymidylate synthase (TS) and dihydropyrimidine (DPD) C. H. Smorenburg, G. J. Peters, C. J. Van Groeningen, P. Noordhuis, J. M. Van Riel, H. M. Pinedo, G. Giaccone VU Medical Center, Amsterdam, Netherlands

Background: After reducing doses of CPT-11 from 125 to 100 (doses in mg/m2) and 5FU from 500 to 400 (LV unchanged, 20) due to increased toxicity & early mortality (Rothenberg, JCO 2001), Intergroup N9741 concurrently randomized 355 of a planned 600 patients (pts) with CRC to FOLFOX-4 or R-IFL. Randomization was terminated early due to superior results on FOLFOX. Methods: The regimens were: FOLFOX 4 (Oxal 85 d 1 + LV 200/5-FU 400 bolus + 600 as a 22 hour infusion d 1,2 q 2 week); R-IFL (CPT-11 100 + LV

Background: Treatment of ACRC with 5FU/LV has a response rate (RR) <25%. In a retrospective study both low TS and low DPD gene expression levels in metastases of ACRC were positive predictive factors of a response to 5FU (Salonga et al: CCR 2000). In contrast, the efficacy of oxaliplatin and irinotecan is supposedly independent of TS and DPD. Methods: Gene expression levels of TS and DPD using a quantitative RT-PCR assay were determined in pretreatment needle biopsies of metastatic tissue in previously untreated pts with ACRC. The cut-off values for TS and DPD relative to B-actin were 5 x 10 –3 and 45 x 10 -3, respectively. Chemotherapy consisted in arm A (low TS and low DPD) of weekly bolus 5FU/LV 500 mg/m2 and in arm B (high TS and/or DPD) of 3-weekly oxaliplatin 85 mg/m2 and irinotecan 200 mg/m2. After progression, cross-over to the other regimen was allowed. Toxicity and response were evaluated every 3 and 9 weeks, respectively. Statistics using X2 test (one-sided, p<.05, 80% power) required 20 evaluable pts in arm A to detect a significant increase of RR >50%. Results: 47 out of 54 pts with a biopsy were treated in arm A (n=26) or arm B (n=21). 7 pts were not allocated to arm A or B due to deteriorating condition (3), other chemotherapy (2) and metastectomy (2). All pts in arm A were evaluable for response, and partial response (PR) was observed in 11 (RR 42%) and stable disease (SD) in 9 pts. All but 1 pt in arm B were evaluable, with a PR in 5 (RR 25%) and SD in 9 pts. No responses were observed in

Abstracts

Onkologie 2004;27(suppl 1):12–58

3621

N9741: FOLFOX (oxaliplatin(Oxal)/ 5-fluorouracil (5-FU)/ leucovorin (LV) or reduced dose R-IFL (CPT-11 + 5-FU/LV) in advanced colorectal cancer (CRC): Final efficacy data from an intergroup study R. M. Goldberg, D. J. Sargent, R. F. Morton, C. Fuchs, R. K. Ramanathan, S. K. Williamson, B. P. Findlay, H. R. Pitot, S. R. Alberts, NCCTG, CALGB, ECOG, SWOG, NCIC University of North Carolina at Chapel Hill, Chapel Hill, NC; Mayo Clinic, Rochester, MN; Iowa Oncology Research Association CCOP, Des Moines, IA; Dana-Farber Cancer Institute, Boston, MA; University of Pittsburgh, Pittsburgh, OR; University of Kansas, Kansas City, KS; St Catharine’s Hospital, St Catherines, ON, Canada

27

11 pts with a high TS/DPD after crossover to 5FU/LV, while crossover to the other regimen (n=10) resulted in 1 PR and 4 SD. No neutropenic fever or toxic deaths occurred. Conclusions: This study shows a higher efficacy of monotherapy with 5FU/LV in pts with a low level of TS and DPD than expected in unselected pts. As the need for fresh metastatic tumor tissue hampers this approach, investigation of alternative methods for TS and DPD assessment and other predictive markers is warranted.

3631

Sequential capecitabine-based treatment of patients with metastatic colorectal cancer (MCRC) combined with oxaliplatin first-line: A retrospective analysis H.-G. Hoeffkes, M. Arland, A. Freidank, O. Ranze Cancer Center, Klinikum Fulda, Fulda, Germany

Background: A favorable outcome of first-line therapy of capecitabine (XEL) and oxaliplatin (XELOX) was recently documented for patients with MCRC. Retrospectively, we analyzed MCRC patients treated with sequential therapy of XELOX, XEL+irinotecan, and XEL+mitomycin in progressive disease (PD). Methods: Inclusion criteria: 18–75 yrs of age, serum creatinine <1.5 mg/dL, informed consent. Therapy was capecitabine-based with 1,000 mg/m2 bid from day1–14 (XEL), first-line with oxaliplatin 130 mg/m2 (XELOX), d1 q3 wks for max of 6 cycles, if progress or toxicity, a second-line therapy with XEL and irinotecan 250 mg/m2 (XELIRI), d1 q3 wks for max of 6 cycles, if progress or toxicity, a third-line therapy with XEL and mitomycin 10 mg/m2 (XELMIC), d1 q3 wks for 6 cycles. CT scans performed every 12 wks to assess response according to the RECIST criteria. Primary objectives were the number of responses, time-to-progression (TTP), and overall survival. Secondary objectives were NCI-CTC Grad 3+4 toxicity. Results: All 16 pts, median age 61.2 yrs underwent prior tumor surgery. Sites of metastases were: liver (63%), lung (47%), peritoneum (33%), local (20%), others (33%). 7 pts (44%) received prior adjuvant chemotherapy. After a median of 5 cycles XELOX (range: 2–10) 1 CR (6%), 8 PR (50%), 3 SD (19%) and 4 PD (25%) were observed. In 1 pt XELOX dosage was reduced to 50% because of diarrhea. Median time of response was 13.1 wks (range: 0–26) with a median TTP (Kaplan-Meier estimation, 9 events, 7 censored) of 26 wks. 13 pts actually receive second-line XELIRI and 2 pts FOLFIRI. After 6 cycles of XELIRI, 4 pts have a PD and 1 pt a SD, yet receiving XELMIC. 3 pts died in week 29, 34 and 60 due to concomitant diseases. After a median of 33.4 wks (range: 20–60), 13 pts (81%) are still alive and median survival has not been reached yet. Grade 3 + 4 toxicity were: 6 diarrhea, 1 neutropenia, 1 hand-and-foot syndrome. Conclusions: First-line XELOX is safe and effective in MCRC with an overall response of 56% and a TTP of 28 weeks. Toxicities rarely occurred and are manageable with standard supportive care.

3669

A preliminary report of biweekly oxaliplatin plus continuous ufur therapy for metastatic colorectal cancer patients C.-P. Lin, I. Y. Chang, C. Cheng, W.-Y. Wang Taipei Municipal Jen-Ai Hospital, Taipei, Taiwan Republic of China

Background: The aim of this study is to define the safety and efficacy of biweekly oxaliplatin therapy combined with continuous oral Ufur therapy in colorectal cancer. Biweekly therapy in outpatient is more acceptable and supplemented with Ufur(tegafur+uracil)) may maintain the dose intensity of drug therapy. Methods: patients of metasta-

28

Onkologie 2004;27(suppl 1):12–58

tic colorectal cancer with one or more measurable metastatic lesion, who have ECOG performance status ≤2, and with adequate bone marrow, liver, renal function were enrolled. Treatment plans: Oxaliplatin 85 mg/m2 IV in 2 hrs with premedication of MgSO4 (2 gm) Calcium gluconate (1 gm) were given biweekly in out-patient. Ufur 200 mg p.o. BID daily, until disease progress or totally 12 course of oxaliplatin. Results: 21 patients were enrolled in this trial (M/F=12/9), median age was 58 years (33–71), primary tumor site: colon 14 cases, rectum 7 cases, medium number of metastatic site was 2, and 57% of the patient had 2 or more sites (Liver 71%, LN 48%, Lung 14%, Peritoneum 14%, skin 4%). The grade III/IV toxicity of nausea 24%, neutropenia 14%, thrombocytopenia 19%, peripheral neuropathy 24%. Efficacy: 2 patients dropped out due to intolerance to continuous Ufur therapy, but still received oxaliplatin, and more than 76% (16 patients) received more than 6 courses(3 months) oxaliplatin therapy. One patient reached CR and continued oral Ufur therapy, she remains CR for more than 11 months. Two patients had PR, and one received liver segmental resection, remains healthy. Five patients (26%) had stable disease, and 11 patients(58%) had progress disease. Conclusions: our data suggest that biweekly oxaliplatin plus Ufur therapy is a active and tolerable regimen in treating metastatic colorectal cancer patients.

3677

XPD, XPA, ERCC1 and XPG/ERCC5 single nucleotide polymorphisms (SNPs) in oxaliplatin-treated colorectal cancer (CRC) J. M. Vila, I. Moreno, M. Monzo, R. Ibeas, J. Moreno, M. Pinuaga, F. Martinez, A. Navarro, E. Pou, J. Sole Monne Surgery, Hospital Municipal de Badalona, Badalona, Spain; Oncology, Hospital Municipal de Badalona, Badalona, Spain; Departamento de Anatomia Universidad de Barcelona, Barcelona, Spain; Radiotherapy, Hospital General de Catalunya, Barcelona, Spain

Background: Oxaliplatin damages DNA, leading to apoptosis. XPD, XPA, ERCC1 and XPG/ERCC1 are involved in DNA repair, and polymorphic variants in these genes can influence the efficacy of oxaliplatin. We analyzed SNPs in these genes and correlated the results with toxicity and response to oxaliplatin in CRC. Methods: 33 CRC patients (pts) recruited between February 2002 and September 2003 were treated with oxaliplatin-based chemotherapy. DNA was obtained from peripheral blood cells at baseline, and allelic discrimination assay with ABI Prism 7700 was used to analyze SNPs at XPD Lys751Gln, XPA 5’utr A/G, ERCC1 codon 2 Thr/Lys and XPG/ ERCC5 codon 3 His/His. Results: Pts characteristics: age 62.5 (44–78); all pts, PS 0–1. Pts with XPD C/C showed greater hematological toxicity (44%) than those with A/A (9%) (p=0.08). Ptes with XPD A/C and C/C (30%) showed better objective response than those with A/A (18%) (p=ns). Pts with XPG C/C showed better objective response (70%) than those with C/T and T/T (8%) (p=0.002). Pts with XPG C/C achieved longer time to progression (8.5 mos) than those with C/T and T/T (4 mos) (p=0.002). Ptes with XPG C/C achieved better median survival (+11 mos) than those with C/T and T/T (8.4 mos) (p=0.01). Conclusions: XPD C/C may be related to greater hematological toxicity, and XPG C/C may be a predictive marker of response, time to progression and survival in oxaliplatintreated CRC pts. Studies with a larger number of patients should be carried out to confirm these results.

Abstracts

3680

Predictive value of single nucleotide polymorphisms (SNPs) in DNA repair genes in non-selected advanced colorectal cancer (CRC) patients (p) treated with oxaliplatin (OXA) plus 5-Fluorouracil (5-FU) first-line chemotherapy (CT) E. Martinez-Balibrea, J. L. Manzano, S. Catot, A. Martinez-Cardus, L. Nuñez, M. Taron, A. Abad Hospital Universitari Germans Trias i Pujol, Badalona, Spain

Background: OXA cause DNA cross-linking and oxidative damage. SNPs in DNA repair genes are associated with inter-individual differences in DNA repair capacity and can explain the heterogeneity in OXA response. We assessed SNPs in four DNA repair genes belonging to three of the major DNA repair pathways: nucleotide excision repair (XPD Lys751Gln; ERCC1 C118T); base excision repair (XRCC1Arg399Gln) and homologous recombination repair (XRCC3Thr241Met) and correlated the results with outcome in OXA/5FU-treated advanced CRC p. Methods: 39 consecutive advanced CRC p were included. TaqMan 5’ nuclease assay was used to examine SNPs from DNA isolated from peripheral blood lymphocytes. Results: p characteristics: median age: 59 years (33–72); 59% male. Overall outcome: median time to progression (TTP): 7 months (m); overall response rate (OR): 46%. Homozygosity for wild type alleles (Lys, C, Arg and Thr, respectively) was considered a favorable genotype for OXA response. OR according to genotype (favorable/ unfavorable): XPD23 47.8/43.8%; ERCC1 33,3/47,2%; XRCC1 47.4/45% and XRCC3 44.4/38.1% (P=ns). No differences in TTP according to single genotype were observed. In contrast, when we analyzed the impact of allele combination, p carrying 4 or more unfavorable alleles (47% p) had a shorter TTP than those carrying less than 4 (53% p) (4.5m vs 7.5m; P=0.07). Conclusions: The impact of allele combination in DNA repair genes may be a key predictive marker of chemosensitivity. The prospective inclusion and genotyping of advanced CRC p is needed to clarify the role of these SNPs in predicting response to OXA/5FU.

TXPD 751 Lys/Lys Lys/Gln Gln/Gln

59% 36% 5%

ERCC1 118

XRCC1

C/C C/T T/T

Arg/Arg Arg/Gln Gln/Gln

8% 46% 46%

XRCC3 49% 44% 8%

Thr/Thr 30% Thr/Met 50% Met/Met 20%

3681

A phase II trial of oxaliplatin (L-OHP) and uracil-tegafur (UFT)/folinic acid (FA) for advanced colorectal cancer (ACC) in elderly patients G. Rosati, S. Cordio, A. Tucci, G. Blanco, R. Bordonaro, C. Pizza, G. Reggiardo, L. Manzione San Carlo Hospital, Potenza, Italy; San Luigi Hospital, Catania, Italy; Civil Hospital, Nola, Italy; Data Management Unit, Medi Service, Genova, Italy

equate renal, hepatic and hematologic function and no previous treatment for advanced disease (adjuvant chemotherapy more than 6 months and/or radiotherapy out of the target lesion were allowed) were included. Pts were followed by a geriatric and a quality of life assessment with specific scales and EORTC-QLQ-C30 questionnaire Results: A total of 46 pts were enrolled. All pts were evaluable for toxicity and 40 (too early) for response, M/F 26/20, age 70 to 89, performance status 0/1/2 25/20/1, colon 29 and rectum 17. Metastases locations: liver 32, local (recurrent) 15, lung 11, peritoneal carcinomatosis 6. 14/46 had had prior adjuvant chemotherapy and 7 radiotherapy outside on target lesions. Only one case of unacceptable cardiotoxicity and one case of grade 4 diarrhea were registered. Grade 3 diarrhea and neutropenia occurred in 13%, and 2%, respectively. Most common grade 2 toxicities were thrombocytopenia and nausea/vomiting 15%, anemia 11%. On evaluable pts, 20 (50%) objective responses were observed; 15 pts (37.5%) had stable disease and progressive disease was diagnosed in 5 cases (12.5%). Median overall survival and median time to progression have not yet been reached Conclusions:: These preliminary results confirmed that this tested chemotherapy combination is active and tolerated in elderly pts with ACC. The updated analysis will be presented at the meeting

3688

XELOX (capecitabine and oxaliplatin) as 1st line treatment for elderly patients (pts) with advanced/metastatic colorectal cancer (MCRC) L. López-Gómez, P. Escudero, A. Yubero, J. Feliú, A. Salud, A. Galán, M. Bolaños, J. M. Vicent, F. Losa, M. González-Barón Hospital Virgen de la Salud, Toledo, Spain; Hospital Clínico Lozano Blesa, Zaragoza, Spain; Hospital Obispo Polanco, Teruel, Spain; Hospital La Paz, Madrid, Spain; Hospital Arnau de Villanova, Lleida, Spain; Hospital de Sagunto, Valencia, Spain; Hospital San Pedro de Alcántara, Cáceres, Spain; Hospital General Universitario de Valencia, Valencia, Spain; Hospital de la Creu Roja, Barcelona, Spain

Background: XELOX is a highly active combination in 1st line MCRC, comparable to FOLFOX with little neutropenia and a convenient 3-weekly cycle length. As a well-tolerated, more home-based therapy, XELOX merits investigation in a more elderly pt population. Methods: Objectives of this phase II study were response rate (RR), safety profile, time to progression, overall survival and clinical benefit after XELOX chemotherapy as 1st line treatment in pts aged ≥70 years with histologically proven MCRC. Selection criteria included no prior chemotherapy (except adjuvant therapy), measurable disease according to RECIST, ECOG PS ≤ 2 and adequate bone marrow, renal and hepatic function. Patients received oxaliplatin 130 mg/m2 i.v. D1 followed by oral capecitabine 1000 mg/m2 twice daily for 14 days (750 mg/m2 if Cr Cl=30–50 ml/min) every 3 weeks. Toxicity was evaluated every cycle using WHO toxicity criteria. Results: 32 pts are evaluated for safety to date: M/F, 23/7, median age 75 years (70–82), ECOG PS 0/1: 47%/53%. 96% pts had no comorbidity, 92% had mild dependence on help (Barthel Index) and most (M/F

Background: Colorectal cancer is usually diagnosed in elderly patients (pts). Since the increase of life expectancy makes chemotherapy more appealing in this setting, we test the activity and toxicity of L-OHP and oral UFT/FA as first-line therapy in pts with ACC aged 70 or older Methods: A two-stage trial was planned with an accrual goal of 44 pts. The treatment included L-OHP 65 mg/m2 as an intravenous 3-hour infusion on day 1 and 8 plus UFT 300 mg/m2 and FA 90 mg in three divided doses given orally on days 1–14 of each threeweek cycle. Pts with ACC with age ≥70 years, measurable lesions, ad-

Grade 3–4 Adverse Events per patient (%)

Abstracts

Onkologie 2004;27(suppl 1):12–58

Thrombocytopenia Neutropenia Diarrhea Asthenia Nausea Stomatitis

10 7 3 13 13 3

Fever Hand-foot syndrome Paresthesia Anorexia Abdominal pain

3 3 3 3 3

29

77%/43%) were autonomous (Lawton Index). Tumor sites were colon (n=21) and rectum (n=10). Median number of metastatic sites was 1 (1 site 86%, ≥ 2 sites 14%), in liver (79%), lung (25%) and nodes (11%). Previous treatment included surgery (87%), chemotherapy (32%) and radiotherapy (10%). Currently a total of 95 cycles have been administered: median 3 (1–8). Median relative dose intensity is 95% for oxaliplatin and 97% / 91% (Cr Cl ≤50 / Cr Cl >50 ml/min, respectively) for capecitabine. There was 1 treatment-related death: G4 diarrhea. Efficacy data will be presented. Conclusions: XELOX appears well tolerated in 1st line treatment of elderly pts with MCRC.

3712

Weekly oxaliplatin (OXA), 5-fluorouracil (FU) and leucovorin (LV) as first line treatment for advanced colorectal cancer (CRC) – A phase II study D. Tsavdaridis, A. Athanasiadis, N. Pissanidis, M. Saleh, A. Moshidis, A. Chatzichristou, N. Makrantonakis, P. Kamentsidis, I. Natsiopoulos, C. Christakis General Hospital IKA, Thessaloniki, Thessaloniki, Greece; General Hospital Larissa, Larissa, Greece; General Hospital Papanikolaou, Thessaloniki, Thessaloniki, Greece

Background: OXA has previously shown high efficacy in metastatic colorectal cancer when combined with 5FU/LV, mainly as a FOLFOX schedule. This phase II clinical study was designed to evaluate the efficacy and toxicity of a different OXA-5FU/LV schedule, as first line treatment of advanced colorectal cancer. Methods: Forty-five patients with advanced CRC (median age 69 years, range 50–80, 34 males, 11 females were included. Sites of metastases included liver 25 (55.6%), lymph nodes 14 (31.1%), pelvis 12 (26.7%), peritoneum 11 (24.4%), other 9 (20.0%). Eleven patients had received prior adjuvant chemotherapy, and two radiotherapy outside of the target lesions. The regimen consisted of 200 mg/m2 LV (2h i.v.), 450 mg/m2 5-FU (bolus), and 45 mg/m2 OXA (2h i.v.), weekly for 6 weeks as induction therapy and then weekly for 3 weeks with 1 week rest until progression or unacceptable toxicity. Results: All patients were evaluable for response and toxicity. Grade 3–4 toxicities were: neurotoxicity (4%), diarrhea (4%), leucopenia (2%), nausea/vomiting (2%) and allergy (2%). Dose reductions were necessary in 14 of 45 patients (31.1%) and treatment was early interrupted in 3 of 45 patients (2 progressive disease and 1 death due to acute myocardial infarction). Complete response was observed in 4/45 (8.9%), partial response in 11/45 (24.4%) for a response rate of 33.3%. At a median follow-up of 19 months, 29 patients had relapsed (64.5%). Median time to progression was 13.4 months [95% CI: 10.6–18.2], and median survival 20.4 months [95% CI: 17.2–28.2]. Conclusions: This schedule is highly effective as first line treatment for advanced CRC. It is a feasible, welltolerated regimen, with low incidence of G3–4 toxicities. Further evaluation is warranted.

3715

Management of acute oxaliplatin-related peripheral sensory neuropathy with magnesium and calcium S. Pederiva, C. B. Caspar Kantonsspital Baden, Baden, Switzerland

Background: Oxaliplatin is an effective anticancer agent in the treatment of colorectal cancer. However, it is associated with chronic peripheral sensory neuropathy, an often dose-limiting side effect, which occurs after 4–6 cycles of chemotherapy. Furthermore, an acute, coldinduced syndrome with dysesthesia and sometimes myoclonia occurs

30

Onkologie 2004;27(suppl 1):12–58

in most patients (85–95%) which is reversible but possibly very painful. Recent data showed that magnesium- and calcium-infusions (mg/ca-) before oxaliplatin reduce the acute neuropathy1. Methods: We administered magnesiumsulfat (mg) 1.25g (= 5 mmol = 24.3mg ionised Mg) and calciumgluconat (ca) 1.375g (= 90 mg ionised Calcium) infusions (mixed in 100ml normal saline over 30 min before chemotherapy) in 11 patients with acute sensory neuropathy receiving an oxaliplatin-based chemotherapy. Results: 11 patients received 3–18 doses of oxaliplatin (median =10) preceeded by mg/ca-infusions. There were no side effects nor changes in electrolytes caused by the mg/ca-infusions. Only minimal neuropathy was reported by the patients with mg/ca-infusions. In patients with marked acute neuropathy during previous cycles, the pretreatment with mg/ca-infusions dramatically reduced the dysesthesias (e.g. during exposure to cold in the refrigerator). In one of the patients the infusion of oxaliplatin had to be interrupted because of severe acute neuropathy but could be resumed without any problems after application of a mg/ca-infusion (up to now for a total of 18 doses of oxaliplatin). However, the mg/ca-infusions do not influence the cumulative and often dose-limiting chronic neuropathy. Discussion: Our data indicate that the prophylactic treatment of acute sensory neuropathy with mg/ca-infusions is safe, inexpensive and effective. There is no change in blood electrolytes. However, this treatment does not seem to prevent the oxaliplatin-related chronic neuropathy. 1 Lainé-Cessac P et al. Acute oxaliplatin neurotoxicity dramatically improved with intravenous calcium and magnesium salts. Ileme Congrés Annuel de la Société Francaise de Pharmacologie, Nancy, France, 1998.

3717

Combination of oxaliplatin (OXA) and 5-fluorouracil, leucovorin (5-FU/LV), as first line therapy, in elderly patients >70years, with metastatic colorectal cancer H. G. Botto FOI, Buenos Aires, Argentina

Background: In a previous trial the efficacy of OXA-5-FU/LV combination was reported in Metastatic Colo-Rectal Cáncer (MCRC) patients (pts), refractory to 5-FU/LV, (H.G.Botto ASCO, 2000, 1050). The purpose of this study was to evaluate the tolerance and effectiveness, of the combination of OXA-5-FU/LV given every two weeks, as first line therapy, to elderly pts ≥70 years (y) with MCRC. Methods: Fourteen elderly pts. ≥70 y previously untreated with MCRC, and with good hematological, cardiac and renal functions, compose the sample to this trial; median age was 73 years (range 70–83 y), M/F 8/6; PS, 0=6,1=5, 2=3; sites of disease: liver 8, locoregional 4, lung 2. Treatments: OXA 70 mg/m2 IV, 5 FU 500 mg/m2/LV 20 mg/m2, IV, every 2 weeks, until disease progression or toxicity; response was proved every 8 weeks, the median nº of cycles was 6 (range 4–12 cycles), the total nº of cycles was 86. In all pts the response to treatment, toxicity and survival have been evaluated. Results: PR 5/14 (35.71%), SD 5/14 (35.71%), PD 4/14 (28.58%); OR, 5/14 (35.71%), all pts with hepatic metastasis; and SD 2/14 (14.28%) pts with hepatic metastasis, 2/14(14.28%) pts with locoregional disease, and 1/14 (7.14%) pt with lung metastasis. Toxicity Gr 2/3: diarrhea 6/14 (42.85%), neutropenia 7/14 (50%), thrombocitopenia 4/14 (28.58%), paresthesia 4/14 (28.58%). Time to disease progression, median 7 months (4–12 m). Overall survival, median 16 months (10–36 m). One pt underwent tumoral hepatic resection after PR with Oxa-5FU/LV. Conclusions: The every two weeks combination of OXA-5-FU/LV, showed in elderly pts ≥70 y with metastatic colo-rectal cancer, positive antitumoral response, acceptable toxicity and good survival time. The most important antitumoral activity was detected in the hepatic metastasis.

Abstracts

3726

A combination of oxaliplatin and UFT-l,-leucovorin as first line treatment in advanced colorectal cancer. An ONCOPAZ phase II study J. De Castro, J. Vicent, C. García-Girón, M. Constela, E. Fonseca, J. Montalar, M. Lomas, L. Antón-Aparicio, F. Dorta, M. González-Barón Hospital Universitario La Paz, Madrid, Spain; Hospital General Universitario, Valencia, Spain; Hospital General Yagüe, Burgos, Spain; Complejo Hospitalario, Pontevedra, Spain; Hospital Universitario, Salamanca, Spain; Hospital La Fe, Valencia, Spain; Hospital Infanta Cristina, Badajoz, Spain; Hospital Juan Canalejo, La Coruña, Spain; Hospital Nta. Sra. De La Candelaria, Santa Curz de Tenerife, Spain

Background: To assess the efficacy and safety of the oxaliplatin-UFTl,LV combination as first-line treatment in patients with advanced CRC. Methods: Between 04/99 and 01/00, 81 patients with recurrent or metastatic CRC were included. Mean age was 63 years (40-77), male/female ratio 46/35, ECOG 0/1/2 39/36/6. The study regimen consisted of Oxaliplatin 85 mg/m2 in 120-minute intravenous infusion on days 1 and 15; intravenous l,LV 250 mg/m2 given in two hours on day 1, followed by oral UFT 390 mg/m2 on days 1 to 14, and oral l,LV 7.5 mg/12 hours on days 2 to 14. Pills were taken before meals to favor absorption (for instance, at 8 a.m. and 8 p.m.). Cycles were repeated every 28 days. Results: A total of 470 cycles of chemotherapy were delivered with a median of six cycles per patient (range 1–12). A prescheduled preliminary analysis was performed after inclusion of 16 patients. High gastro-intestinal toxicity was observed: G 3–4 diarrhea 9/16 (56%), G 3–4 nausea-vomiting 3/16 (18%). This lead to reduction of the UFT dose to 300 mg/m2. With this new dosage G 3–4 diarrhea and G 3–4 nausea/vomiting were present only in 24 and 14% respectively. Other grade 3–4 toxicities were stomatitis in 1 (1%), anemia in 3 (5%), neutropenia in 2 (3%), thrombocytopenia in 1(1%), fatigue in 7 (11%). Oxaliplatin-associated peripheral sensory neuropathy was observed in 84% of patients, with grade 3 neurotoxicity occurring in 10 (16%) patients. Grade 3–4 laryngopharyngeal dysestesia was observed in two patients (3%). All patients were evaluable for response: one patient achieved complete response, 27 had partial response, ORR 35% (95% CI 24–46%) with 44% stable disease and 20% PD. The median time to progression was 7.3 months and the median overall survival was 16.8 months. Conclusions: Oxaliplatin and UFT-1,LV is an active combination in advanced colorectal cancer and demonstrate response rate and survival similar intravenous regimens. The toxicity is moderate at 300 mg/m2 dose of UFT.

(sinusoidal 12 hour infusion with flow rate peak at 4:00pm); 5FU 900 mg/m2 and L-Folinic Acid (LFA) 150 mg/m2 d1–4, q14 (sinusoidal 12 hour infusion with flow rate peak at 4:00am. Results: 147 cycles (6.11 average per pt.) were performed: average dose intensity was 48.21 mg/sqm/w, 319.28 mg/sqm/w 1630,41 mg/sqm/w for OHP, LFA and 5FU respectively. Average dose according to body surface was 501 mg/sqm, 3184.17 mg/sqm, 16705,74 mg/sqm, respectively. No cycle was delayed due to toxicity. Haematological and hepatic toxicity weren’t observed, neither was any gr. 4(WHO) toxicity. Radiological Objective Response evaluation has been performed in 23 pt (table 2). ORR was 65.22%. Median survival reached 13 months (only 3 pt died). Conclusions: These data seem confirm that FFL4/10 schedule has very interesting activity in metastatic colorectal cancer. Moreover, chronomodulated infusion is considerably less toxic than conventional flat or bolo infusion.

min max min median

Age

N pts

27

TAB2

ORR

41 77 63 64

M F Cycles Mean

14 13 147 6

RC RP SD PD

17.39% 47.83% 13.04% 21.74%

3748

A feasibility study of gefitinib in association with capecitabine (CAP) and oxaliplatin (OXA) as first-line treatment in patients with advanced colorectal cancer (ACRC) M. Zeuli, A. Gelibter, C. Nardoni, T. Gamucci, A. Gabriele, C. F. Pollera, F. Di Costanzo, C. Signorelli, V. Ferraresi, F. Cognetti Regina Elena National Cancer Institute, Rome, Italy; Ospedale SS. Trinità, Sora (FR), Italy; Medical Oncology, Belcolle Hospital, Viterbo, Italy; Azienda Ospedale Careggi, Firenze, Italy

Background: Oxaliplatin (OHP) and 5Fluorouracil (5FU) efficacy and toxicity strictly depend on circadian rhythms in cell cycle regulation, reduced glutathione content, dehydropyrimidine and thymidilate synthase activities, topoisomerase I activity, membrane permeability as well as liver and renal blood flow. Methods: From August 2002 to December 2003 (table 1) 27 untreated colorectal cancer patients(pt) were treated with OHP based chronomodulated delivery schedule using Melodie® infusion system: OHP 25 mg/m2 /d1–4q14

Background: EGF-receptor (EGF-R) is commonly overexpressed in epithelial tumors including ACRC and has thus become the target of various molecular therapeutic approaches. Gefitinib (Iressa), an EGF-R tyrosine kinase inhibitor, demonstrated activity in lung and head and neck cancer, but its clinical activity in ACRC is still unknown. We designed this trial to assess the safety and efficacy of Iressa in association with CAP-OXA regimen as first-line treatment in pts affected by metastatic colorectal cancer Methods: From October 2002 21 pts were included in a two steps trial (m:f:12/9, median age 66 yrs (54–73). Actually the twelve pts planned for first safety analysis are evaluable for toxicity and response. CAP was orally administered (1000 mg/m2) twice a day continuously for 15 days and OXA was administered (120 mg/m2) as a 2-hour infusion on day 1, repeated every 3 weeks for six courses. Gefitinib has been given orally, once daily, at 250 mg, continuously from day 1.Pts on response at the end of the treatment continued to receive gefitinib until progression or unacceptable toxicity. Results: 40 cycles were totally administered. The most common side-effect was Diarrhea (14/40 cycles: one G4, two G3 and five G2).Acneiform skin lesion (NCI-CTC grade 1–2 in four and one pts respectively) was frequent, but often disappeared or decreased under continued therapy with Iressa. Further major side-effects included Nausea/Vomiting G1–2 in 12/40 cycles, thrombocytopenia G1–2 6/40 cycles, neutropenia G2–3 4/40 cycles. 82.5% of cycles were given at full dose, while 17.5 were given at full dose but delayed; no reduction of dose was performed. 12 pts are evaluable for efficacy

Abstracts

Onkologie 2004;27(suppl 1):12–58

3738

Chronomodulated delivery schedule of oxaliplatin, 5 fluorouracil and folinic acid (FFL4/10) as first line treatment in metastatic colorectal cancer: Preliminary data of a phase II study M. Pirovano, A. Quintè, S. Masseroni, A. Romanelli, A. Biffi, F. Scaglione, F. Fraschini, S. Sestini, R. Valsecchi, D. Tabiadon San Carlo B.Hosp, Milan, Italy; Università degli Studi di MilanoPharmacology Dep, Milan, Italy

31

according to an intent-to-treat analysis: 1 CR, 5 PR, 1 SD, 3 PD and 2 withdrawn for SAE (asthenia, diarrhea, and vomiting). Conclusions: Iressa in combination with Cape and Oxa is a feasible regimen in ACRC with an acceptable toxicity profile and an interesting, although preliminary, activity. Further studies are required to assess the potential of Iressa as treatment option in ACRC. Accrual will continue up to 35 patients as planned.

3769

Oxaliplatin (OXA) with weekly bolus 5-fluorouracil (FU) and low-dose leucovorin (ld-LV) in advanced colorectal cancer (ACRC) patients (pts): Report of safety and efficacy M. Chacon, E. Wasserman, A. Guercovich, J. O¨connor, G. Giornelli, C. Coronado, M. Barugel, E. Roca, S. Carraro, R. Chacon Fleming Institute, Buenos Aires, Argentina; FUCA, Buenos Aires, Argentina; Udaondo Hospital, Buenos Aires, Argentina

Background: OXA and infused FU plus LV (FOLFOX) has proved to be one of the most active regimens in ACRC. On going trials, adjuvant and metastatic setting, have hypothesized that OXA given with bolus FU plus LV would be equally active and safe as OXA combined with infusional FU regimens. Since OXA registration in our country, constrains related to cost and convenience prompt to develop schedules of OXA/bolus FU/ld-LV. Patients and Methods: We assessed retrospectively a cohort of pts with ACRC treated with OXA 85 mg/m2 days 1 and 15 plus LV 20 mg/m2 followed by a 500 mg/m2 bolus dose of FU on days 1, 8, and 15 every 28 days. Data of safety and efficacy are presented. Results: Between 1996 and 2002, 273 pts were treated in two centers. Pts characteristics: median age 56 (22–79), male/female: 151/122, colon/rectum 199/74, median metastatic sites 1 (1–4), first line/second/third (%): 68/13/3 and 16% received after metastasis resection (R0). 1146 cycles (cy) were administered, median per patient 4 (1–11). Safety: the main grade 3–4 toxicities were (pts): gastrointestinal 28, hematological 18, allergies 2, sepsis 3 pts and severe neurotoxicity 12 pts (4,3%). Toxic deaths occurred in 3 pts (1%). Efficacy: Objective response rate (ORR) and time related parameters are summarized as follows: 17 pts initially unresectable ACRC pts were rendered resectable after a 6 (2–11) median cy of treatment. Conclusions: OXA combined with bolus FU and ld-LV seems to be safe and effective. Next multicentre randomized trials will show the comparative value between OXA combined with bolus or infusional FU.

n pts assessable OR (PR+CR) Stable disease (>3 months) TTP (months) Overall survival (months)

32

First line

Second line

Third line

168/185 29% 32% 9 (2–17) 11 (2–33)

30/35 13% 36% 4 (2–13) 8 (2–14)

8/8 12% 33% 3 8 (1–8)

Onkologie 2004;27(suppl 1):12–58

3774

Oxaliplatin, UFT and oral leucovorin combination chemotherapy in 5-fluorouracil refractory colorectal carcinoma: A phase II study H.-M. Ryoo, S. H. Bae, H. I. Lee, J. L. Lee, K. H. Lee, M. Hyun, K. U. Park Daegu Catholic Univ Hospital, Daegu, Republic of Korea; Daegu Catholic Univ Hospital, Daegu, Daegu, Republic of Korea; YeungNam Univ Hospital, Daegu, Republic of Korea; Dong Guk Univ Hospital, Gyeongju, Republic of Korea

Background: This phase II study evaluated the efficacy and safety of a combination of oxaliplatin and UFT/oral leucovorin in Korean Patients with advanced colorectal cancer (ACC) previously treated with a fluorouracil-based regimen. Methods: From December 1998 to December 2001, 28 patients [male:female 12:16, median age 61 (range 37–71)] with measurable disease and Easter Cooperative Oncolgy Group performance status of 0–2 were treated with a combination of oxaliplatin (130 mg/m2 i.v. for 2 hours) on day 1, UFT (300 mg/m2 p.o.) and 5-FU (20 mg/m2 p.o.) on day 1–21, every 3 weeks. Results: 10 out of 28 evaluable patients achieved complete (n=3) or partial (n=7) response, leading to an overall response rate of 35.7%. The estimated median time to progression and survival were 5.5 months and 8.5 months respectively. From the 129 cycles analyzed, sensory neuropathy was the most common toxicity (73%), grade 3,4 hematologic toxicities included neutropenia 1.5%, and thrombocytopenia 1.5%, other toxicities were mild. Conclusions: This combination of oxaliplatin and UFT/oral leucovorin is active and feasible in patients with advanced colorectal cancer previously treated with a fluorouracilbased regimen

4143

Oxaliplatin (OXA) plus folinic acid (FA) and 5-fluorouracil (FU) i.v. bolus (OXAFAFU) versus irinotecan (IRI) plus FA and FU i.v. bolus (IRIFAFU) in advanced colorectal carcinoma (ACC): Activity and toxicity results of the SICOG 0103 randomized trial P. Comella, B. Massidda, G. Filippelli, F. De Vita, D. Natale, A. Farris, F. Buzzi, L. Maiorino, S. Tafuto, S. Mancarella National Tumor Institute, Naples, Italy; University Medical School, Cagliari, Italy; City Hospital, Paola, Italy; Second University Medical School, Naples, Italy; City Hospital, Pende, Italy; University Medical School, Sassari, Italy; City Hospital, Terni, Italy; San Gennaro Hospital, Naples, Italy; City Hospital, Pozzuoli, Italy; City Hospital, Campi Salentina, Italy

Background: IRIFAFU regimen showed activity and toxicity similar to other IRI+FU based combinations in ACC (Comella et al, Ann Oncol 2002). Primary end-point of SICOG trial 0103 was to compare response rate (RR) of OXAFAFU with that of IRIFAFU in ACC. Methods: ACC patients (pts) with measurable disease randomly received: IRI 200 mg/m2 i.v. (90-min) on day 1, and l-FA 250 mg/m2 i.v. (2-hour) + FU 850 mg/m2 i.v. (bolus) on day 2 (IRIFAFU); or OXA 100 mg/m2 i.v. (2-hour) on day 1, and l-FA 250 mg/m2 i.v. (2-hour) + FU 1,050 mg/m2 i.v. (bolus) on day 2 (OXAFAFU high dose [HD]); both regimens were given every 2 weeks. After a planned interim analysis on the first 145 treated pts, OXA was reduced to 85 mg/m2, and FU to 850 mg/m2 (OXAFAFU low dose [LD]). Results: From January 2001 to June 2003, 274 (135 IRIFAFU, 71 OXAFAFUHD, 68 OXAFAFULD) pts were treated. Pre-treatment characteristics according to arms: PS=1–2: 54/38/26; adjuvant FAFU: 34/19/15; two or more disease sites: 63/40/31; liver mets: 99/56/52. WHO toxicity (% pts) according to arms: febrile neutropenia= 4/10/1.5; Grade 4 neu-

Abstracts

tropenia=14/35/19; Grade 3 or more diarrhea=25/10/10. Sixty-day mortality rate=4.4/4.2/4.3. Lévi Grade 3 neuropathy occurred in 11%/3% of pts treated with OXAFAFUHD/OXAFAFULD. RR according to arms: IRIFAFU=31%, OXAFAFUHD=41%, OXAFAFULD=44%. After a median follow-up of 19 months, 166 (60%) pts showed progression, and 118 (43%) pts died: overall median progression-free and survival times were 7.7 and 17.3 months. Numbers of events are still few for comparative analysis. Conclusions: OXAFAFULD regimen produced a significantly greater RR (Fisher’s test, p=0.0478), and a lower toxicity than IRIFAFU. Updated results on progression-free and overall survival according to arm of treatment will be presented at the time of the meeting.

8241

Activity of oxaliplatin and 5-fluorouracil/folinic acid (FOLFOX2) in elderly patients with simultaneous advanced colon (ACC) and Hormone refractory prostate cancer (HRPC) F. Campana, E. Raymond, J. L. Misset, E. Cvitkovic, F. Cvitkovic Hopital Saint Louis, Paris, France; Centre Rene Huguenin, Saint Cloud, France

Background. There is a likelihood of simultaneous colon and HRPC given the increased actual incidence of both cancers in elderly patients. HRPC has a high prevalence of epigenetic mismatch repair deficiency where oxaliplatin, unlike other alkylators, maintains activity. FOLFOX2 has been show active in HRPC (Droz JP, Ann Oncol 2003). We describe two cases of simultaneous advanced tumors treated with FOLFOX2. Methods. Two chemo-naïve patients with synchronous pathologically documented ACC and HRPC received the FOLFOX2 combination every 2 weeks (85 mg/m2 oxaliplatin with 400 mg/m2 bolus 5FU and 100 mg/m2 folinic acid day 1 followed by 2200 mg/m2 48h continuous infusion 5FU). Tumor response was evaluated clinically by ultrasound, CT-Scan and CEA. PSA was also used to assess the response of prostate cancer. Results. Two patients (pts) (PS 1; age:72 and 74) with measurable liver metastasis (colon cancer), non measurable but symptomatic bone metastasis (prostate cancer), and elevated baseline PSA values (12.5 and 255 ng/ml) were evaluable for toxicity and response. Pt 1 had 8 cycles and PT 2, 6 cycles; Grade 2 anemia, nausea, and weight loss were reported in both pts. Grade 3 asthenia was reported in one pt after 4 cycles. No peripheral neuropathy was reported. FOLFOX2 induced objective responses (60% and 70% tumor regression that lasted 8+ and 6 months) in colon cancer liver metastasis. Responses were associated with >50% sustained decrease in CEA levels in both pts. Chemotherapy was associated with pain relief and significant reductions of PSA values (90% decrease in one pt and normalization in the other). Duration of biological PSA responses was 3 months for one pt with partial PSA response and sustained for 8+ months in the pt with complete PSA response. Conclusion. Our clinical experience suggests that front-line FOLFOX2 chemotherapy is safe and active to control tumor progression of synchronous advanced colon and hormone-refractory prostate cancers in elderly males.

Gastrointestinal Cancer – Colorectal/Liver C. Previously Treated Advanced Colorectal Cancer 3510

Activity of cetuximab in patients with colorectal cancer refractory to both irinotecan and oxaliplatin H. J. Lenz, R. J. Mayer, P. J. Gold, B. Mirtsching, P. J. Stella, A. L. Cohn, A. W. Pippas, N. Azarnia, M. N. Needle, E. Van Cutsem USC Norris Cancer Center, Los Angeles, CA; Dana-Farber Cancer Institute, Boston, MA; Swedish Cancer Institute, Seattle, WA; Center for Oncology Research and Treatment, Dallas, TX; St. Joseph Mercy Hospital, Ann Arbor, MI; Rocky Mountain Cancer Center, Denver, CO; Lakeland Regional Cancer Center, Lakeland, FL; ImClone Systems, Somerville, NJ; University Hospital Gasthuisberg, Leuven, Belgium

Background: Cetuximab (Erbitux) has demonstrated activity in patients with epidermal growth factor receptor (EGFR) expressing colorectal cancer (CRC) refractory to irinotecan (CPT-11) both as a single agent and in combination with irinotecan. This phase II study was designed to explore the activity of cetuximab in patients with EGFR+ CRC refractory to both CPT-11 and oxaliplatin (ox). Cetuximab is a chimeric monoclonal antibody that binds specifically to the EGFR preventing homo and heterodimerization and signal transduction. Methods: Patients were treated with cetuximab at standard doses 400 mg/m2 loading dose over 2 hours, then 250 mg/m2 over 1 hour weekly. Results: 350 patients with EGFR + CRC who had failed both CPT-11 and ox were enrolled. Median age 59; M:F = 55:45; ECOG performance status 0:1 = 42:57. The most commonly encountered adverse events were an acne-like skin rash (84% any grade, 5% grade 3) and fatigue/malaise (40% any grade, 9% grade 3/4). Three patients experienced a grade 3/4 allergic reaction requiring discontinuation of study treatment. Partial responses (PR) were observed in 12% of patients (28/235; 95% CI 8–17). 34% of patients had stable disease for at least 6 weeks bringing the disease control rate to 46%. Prior to an amendment to the protocol 9 patients with EGFR negative CRC were enrolled, 2 of whom had a PR. Conclusions: Cetuximab was well tolerated in this patient population and produced major objective responses in 12% of patients who had exhausted all available treatment options for metastatic colorectal cancer. This result is consistent with results reported by Saltz (ASCO 2002) and Cunningham (ASCO 2003).

3515

Bevacizumab (BV) plus 5-FU/leucovorin (FU/LV) for advanced colorectal cancer (CRC) that progressed after standard chemotherapies: An NCI Treatment Referral Center trial (TRC-0301) H. X. Chen, M. Mooney, M. Boron, L. Grochow, J. Zwiebel, D. Vena, K. Mosby, C. Grandinetti, R. S. Kaplan, TRC Participating Investigators NCI, Cancer Therapy Evaluation Program, Bethesda, MD; EMMES Corporation, Rockville, MD

Background: The addition of bevacizumab (BV, Avastin) to IFL for untreated metastatic CRC produces survival benefit (Hurwitz et al, ASCO 2003), but the role of BV in other CRC settings is not established. The purpose of this trial is to evaluate the combination of BV with FU/LV for patients who have exhausted standard therapeutic options. Methods: This is a single arm, multicenter trial conducted nationwide through TRC. The 1st stage was to obtain the RR in 100 evaluable pts; if efficacy is observed, the 2nd stage would open as an expanded access protocol. Main eligibility criteria: advanced CRC

Abstracts

Onkologie 2004;27(suppl 1):12–58

33

progressing after irinotecan- and oxaliplatin-based regimens, ECOG PS 0–2. Treatment plan: BV 5 mg/kg q 2 wks, plus bolus or infusional FU/LV, in 8-wk cycles. All pts are followed for RR, safety, PFS and OS. Results: Due to rapid accrual, many patients were already in the pipeline awaiting study entry by the time the 1st stage accrual target was met. As a result, 350 pts were enrolled in the 1st stage between 7/03 and 11/03: 57% male, 93% ECOG PS 0–1, median age 61 (26–92), 78% with liver and 65% with lung metastases. To date, the median time from enrollment is 8 wks (5–22 wks). Based on partial follow-up at 8 wks, 2 pts had PR (pending confirmation), 19 had SD, 42 had PD; 264 pts are too early to assess. 10 pts have died (8 due to PD, 1 from pneumonia, 1 unknown); 8 pts were off therapy due to adverse events, including G3–4 GI bleeding (3), venous thrombosis (1), G3 CNS bleeding (1), G4 cerebral ischemia (1) and G4 atrial fibrillation (1). The study is currently closed to accrual while data collection for the 1st stage continues. By May 2004, all 350 pts will have had ≥6.5 months of follow-up; data on RR, PFS, OS and safety profile will be presented. Conclusion: It is clear that patients are interested in novel therapeutics such as BV. This trial will provide important safety and efficacy information on the BV/FU/LV regimen for CRC patients in the 3rd line setting. This TRC experience may also assist in planning nationwide trials with new agents in the future when similar situations arise. (Sponsored by NCI, in collaboration with Genentech, Inc.)

3534

Capecitabine/irinotecan (CapIri) and capecitabine/ oxaliplatin (CapOx) are active second-line protocols in patients with advanced colorectal cancer (ACRC) after failure of first-line combination therapy: Results of a randomized phase II study A. Grothey, K. Jordan, O. Kellner, C. Constantin, G. Dietrich, H. Kroening, L. Mantovani, C. Schlichting, H. Forstbauer, H.-J. Schmoll Mayo Clinic, Rochester, MN; University of Halle, Halle, Germany; Klinikum Lippe-Lemgo, Lemgo, Germany; Krhs BietigheimBissingen, Bietigheim-Bissingen, Germany; Altstadt Hospital, Magdeburg, Germany; Klinikum St. Georg, Leipzig, Germany; Diakoniekrankenhaus, Rotenburg/Wuemme, Germany; Private Practice, Troisdorf, Germany

Background: Previously, we reported the interim results of a randomized phase II trial with CapIri vs CapOx as first-line therapy in ACRC (Grothey et al., ASCO 2003). The study protocol included a recommended cross-over as second-line therapy. Here we present the results of CapIri and CapOx as second-line treatment in patients with ACRC, pretreated with a capecitabine-based combination regimen. Methods: Treatment protocols consisted of capecitabine 1000 mg/m2 bid d1–14 plus irinotecan 100 mg/m2 iv d1,8 or oxaliplatin 70 mg/m2 iv d1,8; qd 22. A total of 161 patients (pts) were initially randomized in the phase II trial, of which 68 received CapOx or CapIri as 2nd-line therapy within the protocol (33 pts CapOx after CapIri, 35 pts CapIri after CapOx). Results: NCI-CTC grade 3/4 toxicities for 2nd-line therapies (CapIri vs CapOx): anemia 2.7% vs 6.0%, thrombocytopenia 8.1% vs 3.0%, diarrhea 10.8% vs 0%, nausea/vomiting 0% vs

3.0%, infection 2.7% vs 3.0%, bilirubin 8.6% vs 6.4%; HFS grade 2/3 13.5% vs 3.0%, alopecia grade 2 21.6% vs 9.1%. Grade 3 sensory neurotoxicity was present in 10.8% of pts at onset of 2nd line CapIri, but decreased by at least one grade in all pts after 3 cycles CapIri. Efficacy (CapIri vs CapOx): RR 20.6% vs 12.7%, SD 50% vs 48.4%, PD 29.4% vs 38.7%; PFS 5.1 mos vs 4.3 mos; OS (after start of 2nd-line) 9.6 mos vs 10.6 mos. Overall survivals for pts with 1st- and 2nd-line sequential therapy were almost identical: CapOx->CapIri 17.8 mos, CapIri->CapOx 17.7 mos. Conclusions: CapIri and CapOx are both effective and tolerable second-line treatment regimens after capecitabine-based first-line combination therapy. Severe diarrhea was more frequently associated with second-line CapIri, oxaliplatinmediated sensory neurotoxicity after first-line CapOx was reversible under second-line CapIri. Overall efficacy parameters of our randomized phase II trial do not suggest superiority of a particular treatment sequence.

3542

Liver resection after hepatic arterial infusion (HAI) plus systemic oxaliplatin (Oxal) combinations in pretreated patients with extensive unresectable colorectal liver metastases G. D. Leonard, Y. Fong, W. Jarnagin, R. Harris, L. Schwartz, M. D’Angelica, P. Paty, R. Dematteo, L. Blumgart, N. Kemeny Memorial Sloan-Kettering Cancer Center, New York, NY

Background: Metastatic colorectal cancer has a 5-year survival <5% but increases to 30% in patients (pts) who undergo complete resection. Neoadjuvant systemic (SYS) chemotherapy may increase the proportion of patients suitable for resection, but this is unlikely to occur after first line chemotherapy failure, given the 10–15% response rates of second line SYS Oxal (Rothenberg et al, J Clin Oncol 2003:21;2059, Tournigand et al, Proc Am Soc Clin Oncol 2001;20:abstr 494). HAI plus SYS Oxal may be more effective in this regard, especially in previously treated patients. Methods: Forty-four patients with liver only metastatic colorectal cancer were deemed unresectable by experienced hepatobiliary surgeons. They were treated on either of two phase I trials consisting of HAI (floxuridine and dexamethasone) plus either SYS Oxal with irinotecan (CPT-11) or SYS Oxal with fluorouracil and leucovorin. Responding pts proceeded to surgery if deemed resectable. Results: The overall response rate using HAI plus SYS Oxal was 82% in the 44 unresectable patients despite 70% of pts having progressed on prior SYS CPT-11. Sixteen pts (36%) had minimal residual disease post HAI plus SYS Oxal, 9 pts (20%) have already had complete gross resection of liver metastases while the other 7 are being considered for resection. To illustrate the extent of disease in the 44 pts in this study, the characteristics of these pts are compared with the largest study in the literature that used SYS neoadjuvant chemotherapy alone in mostly chemotherapy naïve pts (table 1). Conclusions: HAI plus SYS Oxal combinations achieved a high response rate which allowed resection in at least 20% (to potentially 36%) of pts, who have already progressed on the latest generation of SYS chemotherapy and who have extensive liver metastases.

Author (#patients)

Age (median)

Diameter of metastases >5 cm

>4 metastases

Bilateral disease

Liver involvement with tumor >25%

CEA >10 ng/dl

Synchronous disease

Giacchetti et al* (n=151) Current study (n=44)

58 59

34% 41%

30% 86%

NA 91%

48% 91%

68% 84%

60% 75%

CEA=carcinoembryonic antigen *=Ann Oncol 1999; 10:663–669

34

Onkologie 2004;27(suppl 1):12–58

Abstracts

3551

Chronomodulated irinotecan (I), fluorouracil (F), leucovorin (L), and oxaliplatin (O) (chrono IFLO) as salvage therapy in oxaliplatin- and irinotecan-resistant advanced colorectal cancer (ACRC): Treatment results and toxicity profile D. Gholam, S. Giacchetti, C. Brezault Bonnet, M. Bouchahda, R. Sverdlin, R. Adam, C. Jasmin, F. Levi Hopital Paul Brousse-Hepato-Biliary Center, Dept of Oncology & Chronotherapy Center, Villejuif, France

Background: Combination of FLO or IFL showed superiority over FL in phase III studies in ACRC (Giacchetti, JCO, 2000 - Douillard, JCO, 2000). Chronotherapy showed an impact on tolerability and activity of oxaliplatin, fluorouracil, and irinotecan (Lévi, Lancet, 2001). Methods: 71 patients (pts) received 608 courses (c) of a 5-day chrono IFLO delivered with a multichannel pump (Melodie, Aguettant, F) q 3 weeks (starting dose: I:160, F:2800, L:1200, O:75 mg/m2/c). Thirtyseven pts were men and 34 women. Median age: 58 years (33–82); WHO performance status (PS) 0/I/II: 38/21/12 pts; number of organs involved 1/2/3: 22/36/13 pts; liver: 83% pts; lung: 28%; peritoneum: 28%; prior chemotherapy lines: 2/3/4+: 25%/45%/30% pts; prior FLO and IFL: all pts. Toxicity was assessed every 3 weeks. Response was evaluated every 3c with CT scan and abdominal ultrasound. Results: A median of 7 c (2–20) was administered. Main grade 3–4 toxicities were diarrhea (44% pts, 8% c), neutropenia (31% pts, 6%c), nausea or vomiting (28% pts, 4% c), thrombocytopenia (6% pts, 0,8% c), mucositis (3% pts, 0,5% c). Grade 3 sensory neuropathy occurred in 15% pts only despite extensive prior oxaliplatin. PS remained ≤ 1 in 77% of the pts. 53 pts had stable disease. 3 pts, all with initial PS= 0, had a partial response >50%. Median time to progression is 6 months. At 15 months, overall survival rate is 51% [95% CI, 39 to 63]. Conclusions: After failure of both FLO and IFL, Chrono IFLO offered survival benefit, with good tolerability and quality of life in heavily pre-treated ACRC patients. Chrono IFLO deserves randomized evaluation as a third-line treatment of ACRC.

3562

Phase II study of short-time infusion of oxaliplatin in combination with capecitabine (XELOX30) in patients with progressive colorectal cancer after irinotecan and 5-fluorouracil (FU) treatment P. Pfeiffer, B. Glimelius, H. Sørbye, J. P. Mortensen, L. Baltesgard, K. M. Tveit, D. Ogreid, H. Starkhammar Odense University Hospital, Odense C, Denmark; Karolinska sjukhuset, Stockholm, Sweden; Haukeland University Hospital, Bergen, Norway; Vejle Hospital, Vejle, Denmark; Tromso University Hospital, Tromso, Norway; Ullevaal University Hospital, Oslo, Norway; Rogaland Central Hospital, Stavanger, Norway; Linkoping University Hospital, Linkoping, Sweden

Background: Oxaliplatin (Ox) in combination with FU or Capecitabine (Xel) prolong median survival as first and second line therapy in patients with advanced colorectal cancer (ACRC). Ox induces acute, cold-induced neuropathy in most patients. The incidence is claimed to be infusion rate-dependent and therefore a 2 hour infusion is recommended, but in a prior pilot study (Pfeiffer et al, Acta Oncol 2003; 42: 832) a 30 minutes infusion could be given without additional neuropathy. Methods: For practical and economic reasons, but also for patient´s convenience, we performed this phase II study to examine XELOX30 (Xel 1000 mg/m2 orally twice daily days 1–14 and Ox 130 mg/m2 as 30 minutes infusion day 1) in patients with pro-

Abstracts

gressive colorectal cancer after therapy with irinotecan and FU. Neuropathy was graded 0–3 using the Ox-specific scale. Other adverse events were graded using NCI-CTC. Results: From Nov 2002 to Sept 2003, 69 patients with ACRC were included and treated with XELOX30. Median age was 62 years (33–73), median WHO PS 1 (0–2). Median number of courses was 6 (range 1–10), median cumulative dose of Ox was 830 mg/m2. Response rate was 16%. Median TTP was 5.6 (4.6–7.5) months and median survival was 9.2 (8.6–12.6) months. Neurotoxicity: Grade 0: 9%, grade 1: 68%, grade 2: 17% and grade 3: 5%. Other grade 3 toxicities were nausea/vomiting: 7%, diarrhea: 8% and PPE 8%. Fourty patients have stopped therapy due to PD (n=20), GI-toxicity (n=7), neurotoxicity (n=2), patients wish (n=5) and other (n=6). Conclusions: XELOX30 is an active, and very convenient 2. line regimen with an excellent safety profile similar to other Ox schedules.

3580

Phase II study of capecitabine, oxaliplatin and erlotinib in previously treated patients with metastatic colorectal cancer (MCRC) J. A. Meyerhardt, A. Xhu, P. C. Enzinger, D. P. Ryan, J. W. Clark, M. H. Kulke, A. Michelini, M. Vincitore, A. Thomas, C. S. Fuchs Dana-Farber Cancer Institute, Boston, MA; Massachusetts General Hospital, Boston, MA; Harvard Medical School, Boston, MA

Background: Erlotinib (Tarceva) is an orally active, selective inhibitor of the HER1/EGFR tyrosine kinase with clinical activity in various solid tumors. The combination of capecitabine (C) and oxaliplatin (O) is becoming an increasingly attractive option to use in the treatment of MCRC. We conducted a phase II study to evaluate the objective response rate (ORR) of O+C+erlotinib in previously treated patients (pts) with MCRC. Methods: Eligible pts had measurable MCRC, either 1 prior systemic regimen for MCRC or progression of disease within 12 months of adjuvant therapy, performance status 0–2. The treatment regimen consisted of 21-day cycles of C at 1000 mg/m2 BID x 14 days (reduced to 750 mg/m2 after first 13 patients experienced excess of grade (gr) 3/4 toxicities), O at 130 mg/m2 on day 1, and erlotinib 150 mg qd. Responses were based on RECIST criteria Results: To date, 22 pts have been enrolled with the following characteristics: male/female, 16/6; PS ECOG 0/1, 11/11; median age 59, range 32–73; 73% prior CPT-11 for metastatic disease. 20 patients are evaluable for response and 22 patients are evaluable for toxicity. The ORR was 20% (0 CR, 4 PR [2 confirmed, 2 unconfirmed]). 64% of patients had stable disease (8 pts >4 months). Prior to the dose reduction of C, 11 of 13 pts had grade 3/4 toxicities (primarily GI toxicities). Since the dose reduction of C (9 pts), there have only been 4 pts with gr 3/4 toxicities and no pts off study due to toxicity. Based on the current enrollment rate, we expect full accrual of 32 pts and updated results (including ORR and time-to-progression) by June 2004. Conclusions: In a population of previously treated patients with MCRC, capecitabine, oxaliplatin and erlotinib appears to be an active regimen. Initial toxicities of the regimen have been markedly improved upon with a modest dose reduction. Supported by: SanofiSynthelabo, Roche Laboratories, Genentech.

Onkologie 2004;27(suppl 1):12–58

35

3635

Evaluation of oxaliplatin, 5-FU and leucovorin maintenance treatment in refractory metastatic colon cancer C. Headlee, S. Beasley, D. Jackson, C. Teel, B. C. Mirtsching Center for Oncology Research & Treatment, Dallas, TX

Background: The response rate in ≥2nd-line therapy for Oxaliplatin as salvage therapy after failure of irinotecan (I)- or continuous dosing 5FU-containing regimens is low (8–11% PR) and median time-toprogression is short (3.5–4.5 months). Methods: 40 patients were treated with oxaliplatin, 5FU and leucovorin (OFL), or single-agent oxaliplatin (O). All patients who did not progress were treated continuosly until disease progression. Our observations of a fraction of patients who have protracted disease remission and/or stability are herein reported. Results: Prior treatment was 5FU-containing adjuvant therapy (21), or an I-containing regimen (40), capecitabine (15) or 5FU (bolus or infusion) +/- L (12) for metastatic disease. Baseline characteristics: 31 male (78%), 9 female (22%), 30 white (75%), 5 black (12.5%) and 5 Hispanic (12.5%). Treatment was 1st-line in 4 (10%), second-line in 11 (28%), third-line in 20 (50%) or ≥ to fourthline in 5 (12%). Responses: CR 0 (0%), PR 13 (33%), SD 11 (27%) and PD 16 (40%). The mean number of doses of oxaliplatin was 8 (range: 1–46). Response duration ranged from 109–861d (mean 277 d). SD duration ranged from 17 to 465d (mean 196d). 4 patients (10%) with SD and 6 patients (15%) with response that had non-progression for > than 240 d (8 months). CEA was elevated at baseline in 33 (83%), range 15–2283, mean 504. The mean time to maximum CEA reduction was 143d. 50% of SD patients had a stable or reduced CEA. All other stable patients had minor elevations of CEA ≤5 units changed. The patients with protracted non-progression had the following characteristics: 3 females, 7 males, 2 hispanics, 8 caucasians and the mean age of 56. 1st-line: 2, 2nd-line: 2, 3rd-line: 4, 4th-line: 2. Conclusions: Oxaliplatin-containing salvage therapy for colorectal cancer benefits patients with response or with stable disease similarly, with delayed progression, which may be substantial (greater than 8 months) in 25% of all patients treated, even though many were heavily pre-treated. Maintaining treatment until progression is important in achieving these results.

3690

Second line chemotherapy after first line irinotecan, oxaliplatin and 5-FU/LV (FOLFOXIRI) in metastatic colorectal cancer (MCRC) patients (pts) G. Masi, G. Allegrini, L. Marcucci, S. Cupini, I. Brunetti, E. Fontana, S. Ricci, E. Cerri, C. Barbara, A. Falcone Civil Hospital, Livorno, Italy; S. Chiara Hospital, Pisa, Italy

Background: We conducted a phase I–II study with first line FOLFOXIRI (irinotecan 125–175 mg/sqm on day (d) 1, oxaliplatin 100 mg/sqm d1, LV 200 mg/sqm d1, 5-FU 3800 mg/sqm 48-h IV chronomodulated continuous infusion starting on d1, repeated every 2 weeks) and a phase II study with first line simplified FOLFOXIRI (irinotecan 165 mg/sqm d1, oxaliplatin 85 mg/sqm d1, LV 200 mg/sqm d1, 5-FU 3200 mg/sqm 48-h flat continuous infusion starting on d1, repeated every 2 weeks) respectively in 42 and 32 MCRC pts. Overall response rate, median progression free (PFS) and overall survival (OS) were 71% and 72%, 10.4 months (mos) and 10.8 mos, 26.5 mos and 28.4 mos respectively. A possible concern about the use of all three main active agents up-front is the possibility that this might limit, after progression, disease control with second line treatments. Methods: Among the 70 pts so far progressed 53 (76%) have received second line chemotherapy (22: FOLFIRI, 16: FOLFOXIRI, 5: 5-FU protracted infusion, 3: FOLFOX, 3: 5-FU+MMC, 2: CPT-11, 1: CPT-

36

Onkologie 2004;27(suppl 1):12–58

11+LOHP, 1: raltitrexed). Seventeen pts (24%) did not receive second line treatments: 10 because of deterioration of performance status (PS), 4 because of patient refusal and 3 because of death. Patients characteristics at the time of second-line treatment were: M/F 35/18 pts, median age 64 yrs (range 44–75), ECOG PS ≥ 1 21 (40%) pts, multiple sites of disease 33 (47%) pts. Results: A median of 4.2 mos of second-line chemotherapy per patient were administered (range 1–8). Overall response rate (51 out of 53 evaluable pts) was 31% (2CR+14PR, 95%CI=19–46%), stable disease were 19 (38%) and 16 pts (31%) progressed. Median duration of response was 7.7 mos. At a median follow up of 13.5 mos from the start of salvage chemotherapy median PFS and overall survival were respectively 6.7 and 15.2 mos. Conclusions: These results suggest that first line FOLFOXIRI does not impair the possibility to obtain objective responses and delay tumor progression with second line treatments containing the same agents used in first-line.

3695

Hepatic artery infusion (HAI) of oxaliplatin (Ox) as third and fourth-line therapy for patients with hepatic metastases (M) from colorectal cancer (CRC) pretreated with systemic Ox H. Gollasch, B. Gebauer, I. Sturm, P. Thuss-Patience, D. Pink, C. Stroszczynski, A. Kretzschmar Charité Campus Buch Helios-Klinikum, Medicine & Radiology of Robert Rössle Klinik, Berlin, Germany; Charité Campus Virchow Klinikum, Berlin, Germany

Background: The liver is the predominant site of M from CRC. HAI of cytotoxic agents has a long tradition but due to technical complications and lack of a survival benefit compared to systemic treatment it did not become widely accepted as 1st-line treatment of patients (pts) with M from CRC confined to the liver. HAI might serve as a salvage option for pts in whom all available systemic therapies failed and in whom liver M are still dominant. Methods: We retrospectively analysed the outcome of all pts who were treated with HAI of Ox after systemic pretreatment with Ox in our institution. Results: All 10 pts (age 35–68) had received fluoropyrimidines, irinotecan and Ox in two (5 pts) or three previous palliative lines and were pretreated with liver resection (1) and local thermoablation (3). In 8 cases disease progressed while systemic Ox was given. 2 pts had small lung M beside advanced liver M; CEA median 67 (3–2511). HAI was introduced in median 14 months (8–31) from start of first line palliative therapy. Ox 100 mg/m2 (9 pts) and 130 (1) was infused over 2 h via interventionally placed inguinal port (2) or via repeated single-use catheters (8) (tip in the proper hepatic artery) and accompanied by i.v. 5-FU (FA-modulation/bolus/infusion; mg/m2: 400–400–2400/48 h)in 8 pts or oral capecitabine (2.5 g/m2 d1–14)in 2. In median 6 cycles (2–9), q 2–3 weeks, were applied in an outpatient setting. Toxicity was mild beside one port-infection, one case of prolonged but uncomplicated neutropenia IV° and repeated severe but reversible abdominal pain during HAI in 2 cases. No neurosensory toxicity was seen. Beside 3 PR we observed one MR accompanied by CEA and CA 19-9 drop of 94% and 98% resp. 4 pts had SD while in 2 pts disease in the liver progressed. Extrahepatic progression occured in 5 pts while they received HAI. Median TTP, TTP in the liver and survival from start of HAI were 4 (1–8), 6 (1–8) and 13 (2–15+) months. Conclusions: HAI of Ox via interventionally placed catheters in combination with systemic fluoropyrimidines is a feasible and effective outpatient option for heavily pretreated pts with hepatic M from CRC.

Abstracts

3727

Use of the combination of raltitrexed (R) and oxaliplatin (O) as a salvage treatment for patients with advanced colorectal cancer (ACRC) refractory to 5-fluorouracil (5-FU) based chemotherapy and with hyperexpression of thymidylate synthase (TS): Final report M. Murad, R. Guimaraes, A. O. Scalabrini-Neto, N. Ourives Hospital das Clinicas Federal University of Minas Gerais, Belo Horizonte, MG, Brazil

Background: TS is an important target for fluoropyrimidine based TS inhibitors and its hiperexpression has been correlated with resistance to 5-FU, but not necessarily to R, a folate-based TS inhibitor, which appears to be more correlated to FPGS activity (van Triest et al, 1999; Cheradame et al, 1999). O is a novel diaminocyclohexane platinum agent also active in ACRC and obviously not related to TS status. Methods: 21 patients with ACRC not responsive to a first line 5-FU based chemotherapy: 14(67%) to IFL(Saltz) regimen, 4(19%) to 5FU/LV- (Mayo Clinic) regimen and 3(14%) to capecitabine) and high expression of TS (determined by IHC technique-grades 2+ and 3+) and age less than 70 Y/O, measurable disease, WHO PS 0–2, life expectancy ≥ 2 months, adequate renal and liver functions and informed consent were accrued to receive: T–3.0 mg/m2 and 0–120 mg/m2 every 21 days until progression. Results: Thirteen were male and 8 female; median age: 58(39–67); median PS: 1 (1–2). Number of cycles administered: 98 (median per patient: 4). There were 7 PR (33%), 3 SD(14%) and 11(53%) PD. Clinical benefit based upon weight gain, PSand analgesic consumption was observed in all 7 responders (33%). The median PFS was 6.5 months (1–12) and the median overall SV was 8 months (2–12+). The 1 year SV was 17%. The regimen was well tolerated. WHO grade 4 diarrhea occurred in one patient, grade 3 in 12(12%) cycles; grade 2 vomiting in 10(10%) cycles; grades 3 and 4 neutropenia in 16(16%) cycles; reversible febrile neutropenia in 2 patients: (onedied due to sepsis); grade 3 thrombocitopenia in 6(6%) cycles; anemia grades 2 and 3 in 21(21%) cycles and grade 2 oral mucositis in 3 (3%) of the cycles. Grades 1 and 2 neuropathy was observed in 61% of the cycles. Conclusions: These favorable results suggest that this combination might be effective in the treatment of patients with ACRC with high expression of TS and should be also explored as a first line regimen for this subset of patients with predictable resistance to fluoropyrimidine based chemotherapy.

3729

FOLFOX4 in metastatic colorectal cancer (MCRC) patients who progressed after infusional 5-FU/irinotecan based combination regimens F. Dane, P. F. Yumuk, N. S. Turhal, M. Ekenel, G. Basaran, M. S. Iyikesici, C. Tecimer Marmara University, Istanbul, Turkey; Kadir Has University, Istanbul, Turkey

Background: Irinotecan based regimens are the standard 1st line treatment for MCRC because the other promising agent oxaliplatin is commercially unavailable in Turkey. Oxaliplatin, 5-FU and calcium leucovorine (FOLFOX4) has been reported to be one of the promising treatment options for irinotecan resistant tumors. We want to report our experience of the patients who received FOLFOX4 after progression on irinotecan based combination chemotherapy. Methods: All patients (n=17) with MCRC who received FOLFOX4 were collected retrospectively. Results: Nine (52.9%) patients were male and median age was 59 years (range 50–73). Two (11.8%) had stage II, 8 (47.1%) had stage III disease before relapse and 7 (41.2%) had stage IV disease at presentation. Median time to progression (TTP)

Abstracts

on 1st line chemotherapy for metastatic disease was 6 months. Twelve patients (70.6%) received FOLFOX4 as 2nd line and 5 patients (29.4%) as 3rd line treatment. The TTP was 3 months and median overall survival (OS) was 22 months. There was no statistically significant difference in TTP or OS in between the patients who got FOLFOX4 as 2nd versus 3rd line treatment. TTP was 3 months (95% CI: 2–4) and 4±1 months (95% CI: 3–5) respectively, p=0.48. OS was 32±5 months (95% CI: 23–41) and 19±3 months (95% CI: 14–24) respectively, p=0.146. Conclusion: The benefit of FOLFOX4 in metastatic colorectal cancer patients who progresses after infusional 5-FU and irinotecan containing combination regimens is limited.

Gastrointestinal Cancer – Colorectal/Liver D. Hepatic Metastases of Colorectal Cancer 3550

Resection of non resectable liver metastases after chemotherapy: Prognostic factors and long term results R. Adam, V. Delvart, G. Pascal, D. Castaing, D. Azoulay, S. Giachetti, B. Paule, F. Levi, F. Kunstlinger, H. Bismuth HepatoBiliary Center, Paul Brousse Hospital, Villejuif, France

Background: Chemotherapy allows an increased number of patients with primarily non resectable colorectal liver metastases (CRLM) to be further resected after tumor downstaging. However the benefit of this ”rescue surgery” is still debated and prognostic factors of outcome are lacking Methods: From February 1988 to December 2000, 1400 consecutive patients with CLRM were managed at a single institution. From these, 295 (21%) were primarily resected and 1105 (79%) initially non resectable were treated by systemic chemotherapy using either oxaliplatin or irinotecan regimens. Non resectable patients were routinely reexplored every 3–4 courses. Surgery was reconsidered when potentially curative after response to chemotherapy. Fifteen factors assumed to be predictive of survival were evaluated by uni- and multivariate analysis in patients secondarily resected Results: In the non resectable group, 139 ”good responders” (13%) could be secondarily submitted to hepatic resection after an average of 10 courses of chemotherapy. CRLM were initially non resectable because too large (7%), ill-located (15%), multinodular (55%) or associated to extrahepatic tumor (22%). Operative mortality within 2 months was 0.7% (1/139). Tumor recurrence was treated by repeat hepatectomy in 50 patients (36%) and pulmonary resection in 27 (19%). After a mean follow up of 5.3±3.2 years, overall 5-year survival was 36% (disease-free: 22%) compared to 47% for primarily resected patients (p= 0.02). At multivariate analysis, 4 preoperative factors were independently associated to decreased survival: rectal primary, number of metastases ≥3, CA 19-9 level >100 UI/L and concomitence of peritoneal or lymph nodes metastases. 5-year survival according to the presence of 0,1, 2, 3 or 4 of these factors was respectively 60%, 30–40%, 10–18% and 0–3%. Conclusions: Effective chemotherapy allows 13% of patients with irresectable CRLM to be rescued by liver surgery with a hope of long term remission. Although lower than that of initially resectable patients, the overall 5-year survival is 36% for a similar risk of mortality. 4 preoperative factors are able to predict long-term survival.

Onkologie 2004;27(suppl 1):12–58

37

3553

Surgical resection of metastases (mts) after biweekly chemotherapy with irinotecan, oxaliplatin and 5-fluorouracil/leucovorin (FOLFOXIRI) in initially unresectable metastatic colorectal cancer (MCRC) A. Falcone, G. Masi, S. Cupini, L. Marcucci, S. Ricci, I. Brunetti, E. Cerri, E. Fontana, M. Andreuccetti, G. Allegrini Civil Hospital, Livorno, Italy; S. Chiara Hospital, Pisa, Italy

Background: We conducted a phase I–II study with the triple drug combination FOLFOXIRI (irinotecan 125–175 mg/sqm day (d) 1, oxaliplatin 100 mg/sqm d1, l-LV 200 mg/sqm d1, 5-FU 3800 mg/sqm 48-h IV chronomodulated continuous infusion starting on d1, repeated every 2 weeks) and a subsequent phase II study with a simplified FOLFOXIRI regimen (irinotecan 165 mg/sqm d1, oxaliplatin 85 mg/sqm d1, l-LV 200 mg/sqm d1, 5-FU 3200 mg/sqm 48-h flat continuous infusion starting on d1, repeated every 2 weeks) respectively in 42 and 32 unselected MCRC pts with initially unresectable mts. Overall response rate, median progression free (PFS) and overall survival (OS) were 71% and 72%, 10.4 months (mos) and 10.8 mos, 26.5 mos and 28.4 mos respectively. Methods: In consideration of the high activity of these regimens 30 out of 74 pts (40%) could be revaluated for surgical resection of mts after chemotherapy. Among these 30 pts revaluated for surgery a R0 surgical resection could be performed in 19 pts (26% of the initial 74 pts). Characteristics of the 19 radically resected pts were: median age 66 years (45–73), ECOG PS ≥ 1 6 pts (32%), median CEA 10 ng/ml (1–288), median number of mts 3 (1–10), liver involvement ≥ 25% 12 pts (63%). Sites of disease were: liver only 14 pts, liver + lymphnodes 3 pts, liver + peritoneum 1 pt, liver + lung 1 pt. Mts were synchronous in 12 pts (63%) and metachronous in 7 pts with a median disease free interval of 12 mos. Results: After a median follow up of 28.8 mos median PFS is 18.6 mos and median OS is 39.6 mos. In 2 pts progressed after surgery a surgical re-resection and/or radiofrequency ablation was performed. Conclusions: First line FOLFOXIRI enables R0 surgical resection in about a quarter of pts with clearly initially unresectable MCRC. Median OS of resected pts is very promising. Prospective evaluation of FOLFOXIRI as neo-adjuvant treatment in initially unresectable colorectal cancer pts with liver and/or lung metastases is planned.

3592

The MIROX Regimen: Intensive systemic chemotherapy combined with surgical treatment of colorectal cancer metastases. Results of a phase II study J. Taïeb, P. Artru, F. Paye, M. Hebbar, F. Maindrault Goebel, T. André, C. Tournigand, C. Louvet, R. Parc, A. De Gramont Hopital ST Antoine, Paris, France; Clinique St Jean, Lyon, France; CHRU, Lille, France; Hopital Tenon, Paris, France

Background: This work was designed to evaluate the efficacy and tolerance of adjuvant FOLFOX-7 followed by FOLFIRI (the MIROX strategy) in patients with resectable metastatic colorectal cancer. Methods: Forty seven patients with resectable colorectal cancer metastases were prospectively enrolled in this study between September 1999 and October 2001. The treatment consisted in six cycles of FOLFOX-7 (leucovorin 400 mg/m2, Oxaliplatin 130 mg/m2 and 5-FU 2400 mg/m2 as a 46-hour infusion, every two weeks), followed by six cycles of FOLFIRI (leucovorin 400 mg/m2, CPT11 180 mg/m2 in 90 min infusion, 5-FU 400 mg/m2 and 5-FU 2400 mg/m2 as a 46-hour infusion, every two weeks). Surgery was performed before chemotherapy (26 patients) or after the 6 cycles of FOLFOX7 (21 patients), 6 cycles of FOLFIRI being administered thereafter. Results: All but one patient underwent curative surgery and two pts refused post-opera-

38

Onkologie 2004;27(suppl 1):12–58

tive chemotherapy. Tolerance was generally good. The most relevant toxicities were grade 3–4 neutropenia and thrombocytopenia, observed in 12 patients (25%), without any febrile neutropenia or bleeding event. No toxic death occurred. Two pathologically confirmed complete responses and 15 partial responses (overall response rate of 80%, CI: 69–88) occurred with FOLFOX-7 for the 21 patients receiving this regimen before surgery. The 2-year overall survival rate was 88%. The two year recurrence free survival was 51%. Conclusion: The MIROX strategy is feasible and well tolerated in patients with resectable metastatic colorectal cancer. Progression free and overall survivals, with a 25 months median follow up, are similar to those observed with the best therapeutic strategies currently available, though a longer follow-up is needed to confirm these results. This regimen is currently compared to a LV5-FU2 regimen in a randomized phase III study.

3593

Triplet therapy with oxaliplatin, irinotecan, 5-fluorouracil and folinic acid within a combined modality approach in patients with liver metastases from colorectal cancer J. De La Cámara, J. Rodriguez, F. Rotellar, A. Viudez, J. GarcíaFoncillas, F. Pardo, I. Gíl-Bazo, A. Chopitea, S. Martín-Algarra Clínica Universitaria.University of Navarra, Pamplona, Spain

Background: Although surgical resection remains the treatment of choice for liver metastases from colorectal cancer, only a minority of pts are suitable for surgery. Preoperative chemotherapy allows secondary liver resection and subsequent long-term survival in some patients (pts)with unresectable CCR liver metastases Methods: The aim of this study was to evaluate the safety and efficacy of oxaliplatin 120 mg/m2 day 1, CPT-11 150 mg/m2 day 1 and 14, fluorouracil 2600 mg/m2 days 1 and 14 and folinic acid 500 mg/m2 days 1 and 14, every four weeks, in metastasic liver CCR patients either technically unresectable or with known poor prognostic factors for long term survival after surgery. Results: 39 pts (M/F: 30/9), median age 55 (28–77), median ECOG 1 (0–2) were included. Mean number of liver nodes and mean size were 6.7 and 6.1 cm, respectively. 22 pts (56.4%) were initially unresectable while 17 (43.6%) had poor prognostic factors including large tumor size (52.9%), multinodular (23.5%), ill-located (23.5) and/or bilobar metastases (35.3%).Overall response rate to induction chemotherapy was 64%, 3 pts (7.7%) progressed while ontherapy. 23 pts underwent secondary surgery, including hepatectomy (52.2%) and wedge resections (47.8%), with (n=10) or without radiofrequency. An R0 resection was achieved in 84% of pts, with margins <1cm in 60.9% of them. Pathological complete response rate was 17.4%. After neoadyuvant chemotherapy, 11 out of 22 initially unresectable CCR pts underwent surgery with curative intent, including 9 of then with a R0 resection. After a median follow up of 15 months (4–40), median progression-free survival and overall survival have not been reached. Main grade 3–4 toxicity included anemia (7.7%), leucopenia (13.9%), neutropenia (30.8%), plaquetopenia (5.1%), vomiting (5.1%) and diarrhea (23%) 1 pt died due to postoperative complications. Conclusions: Induction therapy with this triplet regimen achieves a high resectability and pathological complete response rates.

Abstracts

3595

Individualizse the treatment of liver metastases from colorectal cancer using RAND appropriateness method: Oncosurge decision model. Results of a consensus detecting from the international taskforce group G. Poston, D. Haller, J. Kahan, M. Cornelis, S. Mahyaoui, OncoSurge International Taskforce Group Royal Liverpool University Hospital, Liverpool, United Kingdom; University of Pennsylvania, Philadelphia, PA; RAND corporation, Leyden, Netherlands; Evidis, Paris, France; Sanofi-Synthelabo, Paris, France

Background: Less than 20% of patients with colorectal liver metastases (CLM) are suitable for liver resection with curative intent. Recent studies suggested that the use of oxaliplatin based neoadjuvant chemotherapy may increase this figure to 30–40% of all patients. However, there is no consensus that determine which patients might benefit from such chemotherapy and subsequently have the chance of coming to curative liver surgery. It will be many years before clinical trials addressing this question are able to report. The aim of this project was to create a therapeutic decision model using the RAND Appropriateness Method. Methods: Following a comprehensive literature review, an international panel of liver surgeons, oncologists and radiologists (16 in total) provided both quantitative and qualitative expressions of opinion concerning the resectability of CLM. During 2 panel meetings a consensus was detected regarding specific patient characteristics in decision making for CLM resection. These characteristics are: contraindications; primary tumour characteristics and CLM characteristics. All characteristics were applied to appropriateness relating to resection and chemotherapy. The appropriateness of each therapeutic decision was assessed by the panel for all possible patient scenarios. Results: Consensus was agreed that features which totally pre-clude consideration for resection included not-treatable extra-hepatic disease, unfit for surgery, CLM involvement of >70% liver or 6 segments. Optimal type (5FU/LV, oxaliplatin, irinotecan) and sequences of chemotherapy pre and post resectability were evaluated. Other specific data relating to CLM (number and size of mets, resection margin, portal nodes) will be presented. Conclusions: This validated decision model tool would help healthcare providers to analyse the appropriateness of liver resection alone or in association with optimal chemotherapeutic sequence based upon individual patient and tumour characteristics. This tool will be available on academic website.

effects were seldom recorded in this 3 months regime and consisted of PNP III in 16%, diarrhea III and IV in 12%, nausea III in 8%; hematologic side effects grade III and IV were neutropenia in 2%, thrombocytopenia in in 6% and anemia in 2%. The achieved response rates were 2 complete responses, 33 partial responses, 11 stable diseases and 4 progressions; leading to an ORR of 70% and a progression under treatment of 8%. None of the progredient patients became inoperable after CTX. Tumor marker response correlated with radiologic response, 15% of the responding patients obtained a normalization of their tumor marker. The extent of the liver resection could be substantially reduced in responding patients. The perioperative mortality was 0% and the morbidity was 4% (1 bile leak, 1 biloma). At the time of this analysis 48% of the patients are alive without recurrence, 34% have recurrence and 18% died after tumor recurrence. Conclusions: Oxaliplatin based chemotherapy demonstrates a high response rate given preoperatively without increasing the perioperative morbidity. No patient demonstrated with nonresectable disease after CTX. The extend of the intended liver resection was considerably reduced in responders. Survival outcome measurements appear promising and further exploration of preoperative CTX is highly recommended.

3613

Resection of previously unresectable liver metastases from colorectal cancer (LMCRC) after chemotherapy (CT) with CPT-11/L-OHP/LV5FU (Folfirinox): A prospective phase II trial F. Quenet, B. Nordlinger, M. Rivoire, J. R. Delpero, G. Portier, D. Mery-Mignard, E. Magherini, S. Payrard, M. Ychou Centre Val d’aurelle, Montpellier, France; Hopital Ambroise Paré, Boulogne Billancourt, France; Centre Léon Bérard, Lyon, France; Institut Paoli Calmette, Marseille, France; Hopital Purpan, Toulouse, France; Laboratoire Aventis, Paris, France

Background: Curative liver resection remains the best treatment option for colorectal liver metastases (CRCLM). Overall survival is still limited due to high recurrence rates. Adjuvant chemotherapy was ineffective in reducing recurrence, therefore neoadjuvant treatment was explored. Methods: During the last two years 50 patients were treated with 6 cycles of neoadjuvant chemotherapy prior to liver resection for CRCLM. Chemotherapy consisted of Oxaliplatin/5FU/LV according to FOLFOX4 regimen or Oxaliplatin (130 mg/m2 d1) and Xeloda (2000 mg/m2 d1–7) q14 over a 3 month period. Primary endpoint of these early results were side effects, radiologic and tumor marker response (CEA, CA 19–9), extend of liver resection and perioperative morbidity/mortality. Results: Grad III and IV side

Background: 5 year-survival after LMCRC resection is 30–35%. Folfiri and Folfox are standard regimens in MCRC. The triple combination Folfirinox as induction CT could improve tumor shrinkage and therefore could allow resection of LM in unresectable patients (pts). Methods: Based on a multidisciplinary consensus, unresectability is defined as follows: LM in right or left lobe in contact of controlateral liver pedicle; LM in contact of veina cava or involving 2 hepatic veins and in contact with the third; liver remaining mass less than 25–30% of functional liver; no initial forseen optimal resection, but susceptible to become resectable after shrinkage. Biweekly 48h-cycles (Cy): L-OHP 85 mg/m2, CPT-11 180 mg/m2, LV5FU were administered. An Independent Committee (IC) reviewed pts eligibility and RR. The primary endpoint was resectability R0 defined as R0= margin ≥2mm, R1=margin <2mm, Ra= resection + either cryosurgery or radiofrequency. Results: Among 34 pts included, 25 were eligible as per IC. To date, 24 pts out of 25 were assessable for primary endpoint. Median age=57y [39–69], M/F=15/10, PS 0/1=17/8, metachronous LM=12 & synchronous LM=13. After a median number of 6 Cy [4–10] before surgery, the overall RR was 72% (CR=1, PR=17, MR=3, SD=1). Pathological status after surgery was: R0=9, R1=5, Ra=6. 2 pts had cryosurgery only and 2 pts were not operated upon. Thus 22 pts had complete local treatment and 14 pts could undergo surgical resection R0-R1. Post-surgical complication =2 biliary fistula, 2 infections leading to hospitalization were reported. After resection, the same CT was pursued in adjuvant setting till 12 Cy in 11 pts. Conclusions: In this population who initially could not undergo optimal resection, 37.5% of assessable pts underwent R0 resection and 91.7% could receive complete local treatment with resection and / or ablation procedures after active CT.

Abstracts

Onkologie 2004;27(suppl 1):12–58

3598

Neoadjuvant chemotherapy for resectable colorectal cancer liver metastases: Impact on magnitude of liver resection and survival T. Gruenberger, B. Schuell, G. Kornek, W. Scheithauer University of Vienna, Vienna, Austria

39

3722

Hepatic artery chronomodulated infusion of irinotecan, 5-fluorouracil and oxaliplatin against liver metastases in heavily pretreated patients with colorectal cancer M. Bouchahda, K. Tanaka, R. Adam, S. Giacchetti, C. BrézaultBonnet, D. Gholam, F. Kunstlinger, D. Castaing, C. Jasmin, F. A. Lévi Hôpital P. Brousse and INSERM E0354, Villejuif, France; Hôpital P. Brousse, Villejuif, France

Background: Liver proliferation and detoxication are regulated by a molecular clock involving 12 circadian genes (Fu et al. Cell 2002, Matsuo et al. Science 2003). Chemotherapy adjustment to these rhythms may enhance antitumor efficacy (Lévi et al. Lancet Oncol. 2001). Methods: We assessed the therapeutic potential of 3-drug Hepatic Artery Chronomodulated Infusion (HACI) in 20 heavily pretreated non hospitalized patients (pts) with metastatic colorectal cancer (MCC). They received 5-day (d) q21 d courses (c) with d1 irinotecan (160 mg/m2 from 2 to 8 a.m. peak at 5 a.m.) and d2–5 oxaliplatin (20 mg/m2/d, from 10 a.m. to 10 p.m., peak at 4 p.m.) and 5-fluorouracil (600 mg/m2/d, from 10 p.m. to 10 a.m., peak at 4 a.m.). 110 courses (c) were given (median, 5; 2–15) using a multichannel pump (Mélodie, Aguettant, F). Toxicity was assessed q21 d and response q3 c with CT scan. Results: Pt characteristics: prior chemotherapy lines: 1/2/3/4+ : 3/1/5/11 pts; WHO Performance Status (PS) 0/1/2 : 10/7/3 pts; median age: 63 years (32–73); liver only: 13 pts; and lung: 7 pts. Treatment was withdrawn for thrombosis (4 pts after 2–7 c) and/or Grade (gr) 3 abdominal pain (3 pts after 3–5 c). Other gr 3–4 toxicities also occurred in less than 20% pts and 5% c (vomiting: 4 pts, 4 c; diarrhea: 4 pts, 5 c; leucopenia: 3 pts, 6 c; fatigue: 2 pts, 2 c; thrombocytopenia: 1 pt, 2 c, alopecia, 3 pts; sensory neuropathy, 2 pts). In 18 pts with measurable lesions, liver metastases progressed (Prog) in 5 pts, were stabilized in 6 (Stab, 33%) and responded >50% in 7 (PR, 39%). Lung Prog occurred in 8 pts, resulting in a 33% objective response rate. PS remained stable in all Stab or PR pts. Median Prog free survival is 6 months and median survival is not reached at 12 months (6 pts alive at 20+ months). Conclusions: Three-drug HACI displayed good tolerability and activity in heavily pretreated MCC patients. Avoidance of extrahepatic progression may require alternating systemic chemotherapy. Such combined modality deserves testing as neoadjuvant therapy of unresectable MCC confined to the liver. Support: ARTBC, Hôp P Brousse, Villejuif (F)

3776

Laparoscopic radiofrequency ablation and systemic chemotherapy for colorectal liver metastases D. A. McCain Hackensack University Medical Center, Hackensack, NJ

Background: Evaluation of 38 patients with colorectal liver metastases(164 lesions) from 11/2000 to12/2003 treated with systemic chemotherapy and laparoscopic radiofrequency ablation. Methods: All patients were treated with standardized systemic regimens of 5-FU, irinotecan and oxaliplatinum with laparoscopic radiofrequency ablation. Results: Median follow up was 16 months (range 3–36 months) with a 78.6% survival at 16 months. Survival by KaplanMeier actuarial survival analyis was 84.7% at 1 year and 52.3% at two yearswith improved disease free survival seen in patients with lesions <5cm in size and <5 in number. Conclusions: Laparoscopic radiofrequency ablation was safe and effective when combined with systemic chemotherapy in treating colorectal liver metastases with patients being able to maintain their chemotherapy treatment schedules. Patients with low volume disease had the highest survival benefit.

40

Onkologie 2004;27(suppl 1):12–58

Gastrointestinal Cancer – Colorectal/Liver E. Rectal Cancer 3552

A phase II study of preoperative oxaliplatin, capecitabine, and external beam radiotherapy in patients with locally advanced rectal adenocarcinoma L. R. Duck, C. Sempoux, B. Honhon, B. Coster, J.-C. Coche, J. Kerger, J.-L. Canon, Y. Humblet, P. Scalliet, J.-P. Machiels Université Catholique de Louvain, Brussels, Belgium; Hôpital Saint-Joseph, Gilly, Belgium; Hôpital Saint-Pierre, Ottignies, Belgium; Université Catholique de Louvain, Mont-Godinne, Belgium; Clinique Notre-Dame, Charleroi, Belgium

Background: Local recurrence after surgery is a cause of treatment failure in pts with locally advanced rectal cancer (LARC). Preoperative RT decreases local recurrence but not distant spread. Oxaliplatin (OX) and capecitabine (CAP) are both highly active cancer drugs for advanced colorectal cancer and have radiosensitizing properties. Therefore, adding CAP and OX to preoperative RT should improve its efficacy in terms of local control and prevention of metastases. Methods: This study was designed to investigate the efficacy (based on pathological response rate) and safety of preoperative chemo-RT in patients with LACR (T3-T4 and/or N+ staged by endorectal ultrasound). RT was administered for 5 weeks, 5 days a week (1.8 Gy/fraction, total dose 45 Gy, 3D conformation technique) in combination with OX 50 mg/m2 intravenously, weekly for 5 weeks and oral CAP 825 mg/m2 twice a day, each day of radiation. Surgery was performed 6–8 weeks after completion of RT. Results: Since December 2002, 37 out of the 40 planned pts were enrolled. 31 pts (ECOG 0–2; median age 60 y, ranging from 33 to 82; males/females 16/15) are evaluable for toxicity and 17 for pathological response after surgery. Neurotoxicity was very limited, as only 10 pts (32%) experienced mild (grade I) neurotoxicity. The most frequent grade 3–4 NCI CTC toxicity was diarrhea: 7/31 pts (22%). 1 of these pts died of candida septicemia 6 wks after the last chemotherapy infusion. Total OX and CAP dose received was 92% and 93% of the planned dose, respectively. The main reason to reduce chemotherapy dosages was diarrhea occuring at 4–5th week of the treatment. RT had to be interrupted in 2 pts due to diarrhea (total dose received 27 and 39.6 Gy). To date, 17 resected rectal specimens were reviewed by the same pathologist. 1/17 had a pathological complete remission and 5 had only microfoci of residual tumor. Conclusions: These results demonstrate that preoperative OX and CAP in combination with RT is feasible in pts with LARC. As expected, the only clinically relevant toxicity was diarrhea. Dose intensity of chemotherapy and RT was high. Updated data will be presented at the meeting.

3557

Neoadjuvant therapy with oxaliplatin (OXA) and 5-fluorouracil (5FU) continuous infusion (CI) combined with radiotherapy (RT)in rectal cancer: First results of the Bologna phase II study C. Pinto, A. L. Gentile, B. Iacopino, S. Neri, G. Ugolini, F. Minni, C. Ceccarelli, G. N. Martinelli, B. Cola, A. Martoni Policlinico S. Orsola-Malpighi, Depts of Radiotherapy, Emergency Surgery, Medical Oncology, Pathology & General Surgery, Bologna, Italy

Background: The aim of this study was to assess 1) the impact of preoperative RT combined with OXA/5FU CI chemotherapy (CT) on tumour response, sphincter preservation and disease control in rectal cancer; 2) the correlation between biopathological factors and thera-

Abstracts

py response. Methods: Pts entering the study had histologicallyproven rectal adenocarcinoma and stage uT3-T4 N-/+ or uT2 N-/+ with inferior location (<5 cm from the anal margin). CT consisted of a weekly schedule of OXA 60 mg/m2 IV for 6 times and 5FU 225 mg/m2/die CI IV d 1–38. RT was started the same day as CT. RT was delivered up to a dose of 50.4 Gy in daily fractions of 1.8 Gy (45 Gy in 25 fractions + a boost of 5.4 Gy in 3 fractions). Rectal surgery with TME was performed 6–8 weeks after the end of neoadjuvant treatment. 8 courses of adjuvant CT with LV5FU2 (de Gramont regimen) were given after surgery. The pathological examination of pre-treatment biopsy and operative specimen considered the immunohistochemical determination of Ki67, p53, bcl2, TS, EGFR, MLH1 and MSH2. Results: 26 pts were accrued between January 2002 and December 2003. The pt characteristics were: 18 men and 8 women; median age 64 years (41–82); median Karnofsky PS 100 (70–100); stage: 2 uT2N-M0, 1 uT2N+M0, 8 uT3N-M0, 12 uT3N+M0, 1 uT4N-M0, 2 uT4N+M0. So far 21 pts have completed the neoadjuvant therapy and surgery (median follow-up 10 months). No grade 4 toxicity was observed, while grade 3 toxicity was: neutropenia 4.8%, thrombocytopenia 4.8%, transaminases increase 4.8%, neurotoxicity 4.8%, diarrhoea 14.3%, epithelitis 4.8%. Pathological downstaging occurred in 12 pts (57.1%), including 3 pT0N0 and 8 pTmicN0. The median value of pathological determination in pre-treatment biopsy and the operative specimens was, respectively: Ki67 8% and 34%; p53 41% and 35%; TS 15% and 10%; MLH1 86% and 72%; MSH2 79% and 53%. In pT0N0 and pTmicN0 pts the biopsy media value of TS was 3.4% and MSH2 94.8%. Conclusions: The early results indicate that 1) this neoadjuvant treatment is active (52.4% pT0N0 + pTmicN0) and welltolerated; 2) major pathological responses were observed in tumours with a low TS and a high MSH2.

3560

A phase I/II study of preoperative oxaliplatin (O), 5-fluorouracil (5-FU), and external beam radiation therapy (XRT) in locally advanced rectal cancer: CALGB 89901 D. P. Ryan, D. Niedzwiecki, D. Hollis, B. E. Miedema, S. Wadler, J. E. Tepper, R. J. Mayer, CALGB MGH Cancer Center, Boston, MA; Duke University Medical Center, Durham, NC; CALGB Statistical Center, Durham, NC; University of Missouri, Columbia, MO; Weill Medical College of Cornell University, New York, NY; University of North Carolina, Chapel Hill, NC; Dana-Farber Cancer Institute, Boston, MA

Background: 5FU and XRT administered preoperatively is a standard regimen for patients with T3/T4 rectal cancer. Preclinical data demonstrate radiation sensitization for O and 5FU. The aims were to determine the MTD of weekly O with 5FU and XRT and to determine the pathological response at the MTD. Methods: Eligibility criteria included adenocarcinoma of the rectum within 12 cm of the anal verge; T3/T4 disease as demonstrated by endorectal USG or MRI; no metastatic disease; adequate bone marrow, liver, and kidney function. 5FU 200 mg/m2 was administered over 24 hours daily. XRT was delivered at 1.8 Gy per day, 5 days/week (28 fractions total). O was administered over 1 hour on day 1 of each week. Surgical resection was recommended 4–6 weeks after chemoradiation. DLTs included grade 4 neutropenia, grade 3/4 thrombocytopenia or grade 3/4 nonhematologic toxicity requiring >7 days interruption in therapy. A traditional phase I accrual design was followed. The MTD is the dose level below the level where 2 of 6 patients experience DLT or a maximum dose of O of 60 mg/m2. Results: 18 patients (7 women, 15 caucasian, median age 56.5) were enrolled to 4 dose levels of O (30, 40, 50, 60 mg/m2). No DLTs were observed in the first 3 dose levels. 2 patients were replaced in dose level 2 (1 hypersensitivity reaction, 1 colonic obstruction). The third patient enrolled at dose level 4 experi-

Abstracts

enced a DLT (diarrhea week 4) and a patient at this level was replaced due to disease related bowel perforation. Three additional patients were enrolled at dose level 4 without any DLTs. Two patients experienced grade 3 neutropenia. No patient experienced grade 3 thrombocytopenia or neuropathy. Grade 3 diarrhea occurred in one patient each at the 30 and 50 mg/m2 dose levels during weeks 4 and 1, respectively, and 3 patients at 60 mg/m2 during weeks 2, 3, and 4, respectively. An additional 19 patients have been accrued at dose level 4. Conclusions: The recommended phase II dose of weekly O when administered with 5FU and XRT is 60 mg/m2. Updated toxicity and pathological response rate for 25 patients treated at the MTD will be available in Spring 2004.

3575

Preliminary phase II SOCRATES study results: Capecitabine (CAP) combined with oxaliplatin (OX) and preoperative radiation (RT) in patients (pts) with locally advanced rectal cancer (LARC) R. Glynne-Jones, D. Sebag-Montefiore, A. McDonald, S. Falk, T. Maughan Mount Vernon Hospital, Northwood, United Kingdom; Cookridge Hospital, Leeds, United Kingdom; Beatson Oncology Centre, Glasgow, United Kingdom; Bristol Oncology Centre, Bristol, United Kingdom; Velindre Hospital, Cardiff, United Kingdom

Background: CAP, an oral fluoropyrimidine preferentially activated to 5-FU in tumors by the enzyme thymidine phosphorylase (TP), has superior antitumor activity and favorable safety compared with bolus i.v. 5-FU/LV in pts with metastatic colorectal cancer (MCRC). OX plus 5-FU/LV further improves outcomes in MCRC. Preclinical xenograft models show that RT and OX both upregulate tumor TP vs normal tissue. The aim of this study was firstly to determine the recommended dose of neoadjuvant CAP in combination with OX and pelvic RT in pts with LARC, and secondly to assess the efficacy and safety of this regimen. Methods: Pts with primary unresectable LARC took CAP (500, 650, 825 mg/m2 twice daily) 7 days/week during RT and received i.v. OX 130 mg/m2, days 1 and 29, and RT (25 fractions of 1.8 Gy; 45 Gy total over 25 days). After identifying the maximum tolerated dose (MTD) of CAP, the study continued into phase II for a total sample size of 80 pts. Results: Currently we have enrolled 52 patients; six in each phase I cohort, and another 34 in phase II. Median age is 55 years (24–75); most pts are male (76%), and Caucasian (97%). All pts were staged T3/T4 by pre-operative MRI. Dose-limiting toxicities (DLTs) occurred in 2/6 pts receiving CAP 825 mg/m2 (grade 3 diarrhea and moist desquamation), establishing the MTD at this dose. CAP 650 mg/m2 twice daily is being used in the ongoing phase II. To date, 42 pts have undergone surgery. Complete pathologic responses occurred in 9 (21%) of pts, a further 10 (24%) were downstaged to ypT1T2N0 and 36 (77%) had tumorfree circumferential resection margins. Among 34 pts evaluable for safety, the most common (≥20%) treatment-related adverse events (AEs) were paresthesia (79%), diarrhea (62%), neuropathy (38%), and nausea (38%). Most AEs were mild (64% G1–2). In phase II 2/34 pts had G3–4 diarrhea and 1/34 had G3–4 lethargy. Conclusions: The recommended dose of neoadjuvant CAP is 650 mg/m2 twice daily with OX 130 mg/m2 (days 1 and 29) and pelvic RT. The regimen is well tolerated and shows promising activity in LARC.

Onkologie 2004;27(suppl 1):12–58

41

3607

Preoperative chemoradiation with oxaliplatin and 5-fluorouracil in locally advanced rectal carcinoma V. Alonso, A. Salud, P. Escudero, J. Valencia, M. Mira, A. Ruiz de Lobera, J. Lambea, R. Grandez, A. Tres, A. Anton Hospital Miguel Servet, Zaragoza, Spain; Hospital Arnau de Vilanova, Lerida, Spain; Hospital Clinico, Zaragoza, Spain

Background: The aim of this phase II trial was to assess the impact of preoperative external radiation therapy combined with 5-Fluorouracil (5-FU) and oxaliplatin (LOHP) on pathologic tumor response, sphincter preservation and tumor control in patients with locally advanced rectal carcinoma. Methods: Fifty-three patients with locally advanced rectal carcinoma (T3/T4 or N+) received radiotherapy (50.4 Gy/28 fractions) and chemotherapy with continuous infusion of 5-FU 200 mg/m2/day on days 1–5 + LOHP 60 mg/m2 on day 1 every week (weeks 1–6). Surgery was performed between 5–7 weeks after the end of the treatment. Adjuvant chemotherapy with 5-FU + Folinic Acid (NCCTG) x 4 cycles was administered after surgery. Results: Patients have been recruited since May 2002 to November 2003. 35 male/18 female. Median age 66 years (range 32–78). Clinical stage (determined by EUS+CT or RMI): cT3 43p, cT4 10p, cN+ 32p. Tumor location (from anal verge): ≤ 5 cm in 26p, >5 cm in 27p. Surgery (performed in 43 patients) consisted of low anterior resection in 27p, abdominal perineal resection in 15 and 1 palliative colostomy. Tumor downstaging was observed in 33p (75.6%), 34 (79%) had negative lymph nodes, including 10p with complete pathological response (22.7%). 36% of patients with tumors in distal rectum had sphinctersaving surgery. Main adverse effects (NCI-CTC): diarrhea G3 7.5%, sensitive peripheral neurotoxicity G1 60%, skin G2 10%, skin G3 2%, nausea/vomiting G2 7.5%, asthenia G2 20%, cardiac (miocardial infarct) G4 2%. No G3–4 hematolgic toxicity was observed. Conclusions: Preoperative chemoradiotherapy with 5FU+LOHP is a feasible and well tolerated treatment. Our results are comparable to the best series of preoperative treatment.

3637

A phase I study of uracil/tegafur (UFT) and oral leucovorin (LV) + weekly oxaliplatin (OXA) and preoperative radiotherapy (RT) in locally advanced rectal cancer (LARC) C. Aschele, M. L. Vitali, G. A. Binda, F. Decian, G. Battistini, C. Caroti, M. D’ Amico, P. Ricci, A. Siragusa, L. Gallo E.O. Ospedali Galliera, Depts of Medical Oncology, Radiotherapy & Colorectal Surgery, Genova, Italy; E.O. Ospedali Galliera, Depts of Medical Oncology, Radiotherapy& Colorectal Surgery, Genova, Italy; Clinica Chirurgica - Università di Genova, Genova, Italy; Osp Evangelico Internazionale, Genova, Italy

Background: We have recently demonstrated that OXA, given weekly at the dose of 60 mg/m2, may be safely combined with infusional FU and preoperative pelvic RT for LARC resulting in rates of pathological complete responses (pCR’s) approaching 30%. Replacing infusional FU with an oral fluoropyrimidine might reduce the inconvenience and some of the potential complications of this regimen. Methods: Since June 2002, 15 patients (10 males, 5 females; median age 63, range 33–75 years) with LARC (cT3-T4 and/or N+; one patient also had a potentially resectable liver metastasis) were thus accrued in a phase I study aimed at defining the maximum tolerated dose (MTD) of UFT (200–250–300–350 mg/m2/day, Monday through Friday for all the duration of RT) combined with weekly OXA (60 mg/m2) and standard pelvic RT (50.4 Gy - 28 fractions). Oral LV (90 mg/day, fixed dose) was administered with the same schedule as UFT.

42

Onkologie 2004;27(suppl 1):12–58

Results: No grade III–IV toxicity was observed at the first dose level (n=4; UFT 200 mg/m2/day). At the second dose level (UFT 250 mg/m2/day), one patient discontinued treatment because of protracted severe obtundation. This cohort was thus expanded (n=6) without any other grade 3–4 toxicity or major deviation from the planned treatment. All the patients accrued at the third dose level (n=5; UFT 300 mg/m2/day) completed treatment as planned without any grade III–IV toxicity. Two patients were not operated due to progression of extrapelvic disease while one patient with a clinical complete response refused surgery and one has just completed chemoradiation and is waiting for operation. Among 11 patients for whom final pathologic reports are available, 2 pCR’s were reported and 3 patients were staged as pT1-2N0. Conclusions: These results demonstrate that oral UFT may replace infusional FU in combination with weekly OXA and standard pelvic RT with low toxicity and without reaching the MTD up to the dose of 300 mg/m2/day. pCR’s were obtained providing preliminary evidence of high antitumor activity. The study is now accruing patients at UFT 350 mg/m2/day (last dose level).

3646

Preoperative capecitabine, oxaliplatin and high-dose pelvic conformal radiotherapy in locally advanced rectal cancer (LARC) E. Tucci, R. Algeri, P. C. Giulianotti, A. P. Pecci, G. Schiaroli, F. Sbrana, F. Perillo Ospedale Misericordia, Grosseto, Italy

Background: Preoperative chemoradiotherapy has proven to improve resectability and local control in LARC. The aim of the present study is to evaluate activity and toxicity of oral capecitabine and oxaliplatin administered concurrently with high dose pelvic conformal radiotherapy as neo-adjuvant treatment of LARC. Methods: Capecitabine 825 mg/sqm bid (days 1 to 40) and oxaliplatin 60 mg/sqm iv every 2 weeks (3 administrations) were given concomitantly with pelvic conformal radiotherapy 1.8 Gy daily up to 45 Gy followed by a 5-fraction boost up to 54 Gy in 6 weeks. From February to November 2003, 16 patients (pts) entered the study: 11 males, 5 females, median age 70 years (range 61–78). Clinical stage, determined by CT, was T3N0-1 in 13, T4N1 in 3 pts. Results: To date 14 pts are fully evaluable (2 too early): a clinical response was observed in 11/14. All pts underwent a R0 surgical resection. Pathological complete response was verified in 4 pts, a minimal microscopic residuum was found in 2 pts. Tumor down-staging by TNM criteria was observed in 10/14 pts. A conservative surgical procedure was performed in 5/9 pts previously planned for abdominoperineal resection. The treatment was well tolerated: no G4 toxicity; G3 diarrhea occurred in 3 pts, G2 leucopenia in 3, G2 anemia in 1. Conclusions: These preliminary data show that the association of capecitabine and oxaliplatin with high dose pelvic conformal radiotherapy as preoperative treatment of LARC is well tolerated and seems active as regards pathological response rate and tumor down-staging, allowing sphincter sparing surgery in pts otherwise candidate to abdominoperineal resection.

Abstracts

3741

A phase I study of oxaliplatin (OXA), 5-fluorouracil (5FU) and concomitant preoperative irradiation in patients with locally advanced rectal cancer J. Balart, E. Marcuello, J. Garriga, J. Pernas, F. Sancho, D. González, L. Pétriz, J. Llaurado, A. Martínez, J. Guallar Hospital de Sant Pau, Barcelona, Spain; Sanofi-Synthélabo S.A., Barcelona, Spain

Background: OXA could act as a radiosensitizer and has a synergic effect with 5FU, increasing the efficacy of chemoradiotherapy in rectal cancer. This study was performed to determine the viability of concomitant preoperative irradiation, OXA and 5FU and to define maximum tolerated dose (MTD) in patients (pts) with rectal cancer. Methods: We included 9 pts with high-risk non-resected rectal adenocarcinoma in a single-centre phase I study over 16 months. Median age: 68 years (46–72); 8 men, 1 women; normal leukocytes and platelets; normal renal function; transaminases <2 UNL; WHO PS <2; written informed consent was obtained prior to study entry. Stages: T3 (8 pts), T4 (1 patient (pt)); N0 (3 pts), N1 (5 pts), N2 (1 pt). All 9 pts had voluminous tumours and 5/9 pts had risk of R1–R2 resection. Pts received pelvic irradiation 45 Gy; 1,8 Gy/fraction from Monday to Friday for 5 weeks, 5FU 225 mg/m2/day continuous infusion x 5 weeks and OXA 65, 75, 85 mg/m2 on day 1, 15 and 29. Dose escalation was allowed if 0/3 or 1/6 pts had dose limiting toxicity (DLT) at previous dose level. Results: Level 1: 6 pts treated. 2 pts had CTC-Grade 3 diarrhea and 1 pt grade 3 rectitis, considered as DLT and leading to chemoradiotherapy discontinuation. Level 0 (OXA 60 mg/m2 and 5FU 225 mg/m2): 3 pts treated and none had grade 3–4 toxicity, being this dose level considered as MTD. There was not neurologic toxicity greater than grade 1. All 9 pts completed radiotherapy, but 3 pts on dose level 1 did not complete third OXA dose and up to 2 weeks of 5FU. All 9 pts underwent radical surgery, without operative mortality. Overall response rate was >80%. 4 pts (44%) had a major down-staging (2 pT0N0M0 and 2 pT2N0M0). Conclusions: This combined treatment appears feasible and effective. OXA 60 mg/m2/15 days x 3 doses should be the recommended dose for phase II studies. 6 additional pts have been included in the trial for a better characterisation of the safety profile of the combination.

3744

Preoperative chemo/radiotherapy with oxaliplatin and 5FU-AF in rectal cancer M. Tomirotti, D. Asnaghi, R. Rovej, M. Botturi, B. Galassi, K. Borgonovo, A. Fasolo, F. Onida, P. Sburlati, A. Scanni IRCCS Ospedale Maggiore of Milan, Milano, Italy; Ospedale Niguarda, Milano, Italy; Ospedale Fatebenefratelli e Oftalmico, Milano, Italy

Background: Randomized trials demonstrated that pre-operative radiotherapy reduces the risk of local relapse and increases of more than 10% the overall survival at 5 years. European trials are ongoing in order to demonstrate the potential superiority of preoperative chemo-radiotherapy. Methods: We started a multicentric trial in order to evaluate efficacy and safety of preoperative chemo/radiotherapy in rectal cancer. The enlistment is ongoing. Endpoints: to obtain histologic complete response; to reduce incidence of local relapse; to preserve anal sphincter; to increase overall survival and disease-free survival. Patients main characteristics: histologically documented rectal cancer (stage pT3); PS
Abstracts

fusion and FA 100 mg/mq i.v. bolus days 1,2,3,4,5). Surgery is foreseen between 3 and 6 weeks by the end of treatment. Results: At present we enlisted 22 patients, 12 males, 10 females, median age 55 years (range 39–70). 16/22 patients have been already submitted to surgery. Responses: 1/16 NC, 15/16 OR (4 pathological CR+11 PR); anal sphincter was preserved in 7/16. Toxicity was mild; neutropenia was avoided by G-CSF. Conclusions: If our results and those of other European trials will be favorable, this approach could represent the treatment of choice for locally advanced rectal cancer.

3746

Phase II study of neoadjuvant treatment of rectal cancer with oxaliplatin, raltitrexed and radiotherapy E. Casado, J. De Castro, B. Castelo, B. Sanchez, J. Feliu, A. Escribano, D. García-Olmo, A. Redondo, C. Belda, M Sereno, M. Gonzalez Baron Hospital Universitario La Paz, Madrid, Spain

Background and objective: Preoperative chemoradiation can improve the outcome of surgery in rectal cancer. The aim of this phase II study was to evaluate the efficacy and toxicity of a preoperative radiochemotherapy combination including oxaliplatin, raltitrexed and radiotherapy in locoregionally advanced rectal cancer. Methods: Between January 2002 and July 2003, 43 consecutive patients (22 women, 21 men; median age 61 years, range 34–74) with stage II and III rectal adenocarcinomas have been enrolled. All patients underwent endorectal ultrasonography (uT2N1= 2, uT3N0= 14, uT3N1= 23, uT4N0= 1, uT4N1= 3), and CT scan for staging purposes. Radiotherapy was delivered with a three-field technique to a dose of 50Gy over 5 weeks with a concomitant boost approach. Chemotherapy was initiated concurrently with RT and consisted of three courses of raltitrexed 3 mg/m2 followed by oxaliplatin 130 mg/m2 every 21 days. Surgery was planned six to eight weeks after RT. Three additional cycles of chemotherapy were given after surgery. Results: 33 patients can be evaluated for response and 43 for toxicity. 25 patients were downstaged (76%), there were 6 pathological complete responses (18%) and sphincter sparing surgery was accomplished in 22 patients (67%). Grade 3 diarrhea developed in two patients, and grade 4 in one patient (7%). Grade 3 neutropenia was observed in three patients (7%), including one febrile neutropenia. Other adverse events were mild, including grade 2 hepatotoxicity (transaminases) in 5 patients (11%). Conclusion: The regimen is feasible, has a low toxicity profile and can be administered in a three weeks basis. This treatment achieves significant tumor downstaging with a remarkable rate of conservative surgeries and complete pathological responses.

3747

Compliance to chemotherapy with oxaliplatin and 5 fluorouracil in patients with advanced rectal cancer who have received prior high-dose pelvic radiotherapy A. Makris, R. Glynne-Jones, D. Sebag-Montefiore, L. Samuel, S. Myint, M. Harrison, A. Martin, J. Dixon, Colorectal Clinical Oncology Group Mount Vernon Hospital, Pinner, Middlesex, United Kingdom; Aberdeen Royal Infirmary, Aberdeen, United Kingdom; Clatterbridge Hospital, Wirral, United Kingdom

Background: Pelvic radiotherapy (RT) may compromise the compliance and effectiveness of future chemotherapy (CT). Patients and Methods: 20 patients (15 male, 5 female), mean age 69 yrs (range 51–75) were studied in a prospective, multicenter, phase II study. All patients had recurrent or metastatic diease. All had received previous

Onkologie 2004;27(suppl 1):12–58

43

pelvic RT: 45Gy 25 fx (15), 45 Gy 30 fx (3), 45 Gy 20 fx (2) and 25 Gy 5 fx (1) at a 16 months prior to this study. Most patients had received concurrent chemoradiation (CRT). Patients were treated with a schedule consisting of oxaliplatin 85 mg/m2 on day 1 and 5FU 400 mg/m2 iv bolus and 600 mg/m2 iv 22hour infusion with folinic acid on day 1 and 2. A maximum of eight cycles of CT were administered every two weeks. Results: A median number of 7 cycles of CT was given (range 3–8), with 132 cycles being given in total. There were 8 episodes of grade III neutropaenia (1 with sepsis). There were 4 episodes of grade III/IV diarrhoea, two of which required hospitalisation, but all resolved. In 18 evaluable patients patients the overall response rate was 33% (3 CR; 3 PR), with 2 patients having SD. Conclusion: This study shows that the combination of oxaliplatin, 5FU and folinic acid after high-dose pelvic RT is well tolerated and has comparable efficacy to that seen in other metastatic studies and a dose intensity equal to that achieved in adjuvant studies. Based on the results of this study, adjuvant oxaliplatin is being delivered in a randomised phase III study (CHRONICLE) following CRT.

9624

Neoadjuvant oxaliplatin (OXA), raltitrexed (TOM), 5-fluorouracil (FU) and folinic acid (FA) + radiotherapy followed by optimal surgery in locally advanced rectal cancer (LARC) A. Avallone, P. Delrio, E. Ravo, P. Marone, S. Lastoria, A. Petrillo, F. Tatangelo, A. Budillon, I. Carreca, P. Comella INT Pascale, Naples, Italy; Università degli studi di Palermo, Palermo, Italy

Background: Preoperative chemoradiation in LARC may produce downstaging and downsizing of the tumor, increases the change of better local control and, in selected cases, sphincter sparing procedures.OXA displayed radiosensitizing properties and significantly enhanced the antitumor activity of TS inhibitor, FU and TOM. In a previous dose-finding study we showed that OXA (100 mg/mq) +TOM (2.5 mg/mq) + FU (900 mg/mq) + LFA (250 mg/mq) given concomitantly to radiotherapy is a feasible approach for LARC in a preoperative setting (Avallone ASCO 2003). Therefore, a phase II study was started to determine the activity of this treatment in newly diagnosed patients with EUS- and MRI-staged LARC. Methods: Eligible pts received three biweekly courses of OXA + TOM on day 1 and FU + LFA on day 2 in combination with pelvic RT (45 Gy, 1.8 Gy/fraction x 25 fractions). Chemotherapy was always administered prior to radiotherapy and toxicity was evaluated weekly with NCI-CTC version 2. Total Mesorectal Excision was planned 6–8 weeks after chemoradiation. EUS and MRI were repeated after chemoradiotherapy. Pts with cT4, pN+ and pathological +ve circumferential resection margin (CRM) also received 4 months of adjuvant FU/LFA. Results: 21 pts (F, 11; M, 10) were recruited. Median age was 60 years, with PS (ECOG) 0–1. Clinical staging revealed T4, N+ and clinical +ve CRM (≤5 mm by MRI), in 3, 16 (10 N1, 6 N2), and 15 pts, respectively. So far, 18 pts are evaluable for EUS and MRI response, and 17 pts have proceeded to TME. Downsizing of the tumor was detected in all pts. At pathological evaluation, CRM >2 mm occurred in all but one pt, 8 pts showed a complete response (TRG 1), 7 pts a downstaging (5 TRG 2, and 2 TRG 3), and 2 pts had no stage variation (1 TRG 3 and 1 TRG 4). Safety findings were consistent with those observed in phase I with incidence of hematological grade 4 in only 3 pts (neutropenia, 3 pts), and non-hematological grade 3 in 3 pts (diarrhea, 2, and stomatitis,1). Conclusions: These preliminary results highlight the favorable safety profile and the efficacy of the combination. Updated results will be presented at the meeting.

44

Onkologie 2004;27(suppl 1):12–58

Gastrointestinal Cancer – Non-Colorectal/Liver 3672

A phase I trial of intraperitoneal (IP) chemohyperthermia (CH) with oxaliplatin (L-OHP) and irinotecan (CPT-11) in patients (pts) with peritoneal carcinomatosis (PC): Pharmacokinetic (PK), tissue distribution and tolerance M. Ducreux, V. Boige, M. Pocard, P. Lasser, D. Malka, M. Bonnay, E. Magherini, C. Oprea, D. Elias Institut Gustave Roussy, Villejuif, France; Aventis Pharmaceutical Company, Paris, France

Background: The complete resection of macroscopic PC followed by intraoperative IPCH to treat residual microscopic disease enables curing some patients (pts). Methods: Pts with PC from gastrointestinal or other origin underwent complete cytoreductive surgery followed by intraoperative IPCH with a fixed dose of L-OHP (460 mg/m2) and one of the 7 increasing doses of CPT-11 (from 300 to 700 mg/m2). IPCH was performed using an open procedure for 30 min at an IP temperature of 43°C with 2l/m2 of 5% dextrose instillation in a closed continuous circuit. Just before IPCH, pts received systemic IV leucovorin (LV) 20 mg/m2 and 5FU 400 mg/m2 to maximize the effect of L-OHP and CPT-11. Results: From May 2002 to June 2003, 39 pts were included with: M/F: 13/26; median age: 46y[26–60]; primary tumor was: 20 colorectal, 11 appendix pseudomyxoma, 3 endocrine tumors, 3 mesothelioma, 2 small bowel. The PK of IP L-OHP was not modified by increasing doses of IP CPT-11, and the drug concentration was 17.8 higher in bathed tumor tissue that in non-bathed tissue. The PK of IP CPT-11 showed an exponential decrease in the peritoneal perfusion and a peak of plasma concentration at 30 min. SN-38 was present in peritoneal instillation immediately after the beginning of IPCH. CPT concentration in tumor tissue increased until the step of 400 mg/m2, and then remained stable. It was 16 to 23 higher than in non-bathed tissues. Half of the L-OHP and CPT-11 was absorbed during the procedure. The hospital mortality rate was 2.5% and the non hematological grade 3 to 5 complication rate (abdominal and intraabdominal) was 25%. At the recommended dose of 400 mg/m2, 5/6 pts had grade 3–4 hematological toxicity: 2 anemia, 5 neutropenia, 1 thrombopenia. The hematological toxicity was correlated to the extent of PC (p=0.017) and duration of surgery (p=0.046). Conclusions: This trial demonstrated a dose related exposure and efficient intratumoral penetration of the 2 drugs. The recommended dose of IP CPT11 was 400 mg/m2, for phase 2. This phase 2 trial is ongoing.

Gastrointestinal Cancer – Non-Colorectal/Liver A. Pancreas Cancer 2059

Clinical and pharmacokinetic study of gemcitabine (GEM) - oxaliplatin (OXA) association in metastatic or locally advanced pancreatic adenocarcinoma M. Airoldi, L. Cattel, R. Passera, C. Micari, E. Ferrarato, C. Zanon Ospedale San Giovanni Antica Sede, Torino, Italy; Università di Torino, Facoltà di Farmacia, Torino, Italy

Background: Combination of gemcitabine (GEM) with oxaliplatin (OXA) is among the most active chemotherapy for advanced and metastatic pancreatic adenocarcinoma. Based on preclinical data, a supra-additive effect was observed when tumor cells were exposed to GEM first and to OXA 24 h later. Study aim was to evaluate clinical benefits and tolerability of GEM-OXA association in MTS/LA pan-

Abstracts

creatic carcinoma. By administering sequence-dependent combinations of the two drugs (GEM d1-OXA d2 or OXA d1-GEM d2), pharmacokinetic (PK) interactions were also evaluated. Methods: 30 patients (median age 60.7 yr; ECOG PS = 1) affected by MTS (67%) or LA pancreatic carcinoma (33%) were enrolled. Pts received GEM 1,000 mg/m2 as 10 mg/m2/min infusion and OXA 100 mg/m2 as 2 h infusion q 2 wks (median cycles/patient = 10). For PK study, 15 patients (28 cycles) were enrolled following two alternated protocols: A (GEM-OXA) (9 pts, 16 cycles) received 1st-3rd-5th cicle GEM on day 1 and OXA on day 2; B (OXA-GEM) (6 pts, 12 cycles) received OXA in 2nd-4th-6th cycle on day 1 and GEM on day 2. Results: 30 pts were evaluable for response and toxicity. Response rate was 30%, superimposable in LA and MTS pts; 50% of patients had clinical benefit. Median PFS and OS were 5.5 and 9.5 months, with 40% of pts alive at 1 year. Grade 3/4 toxicity per patient: 10% neutropenia and thrombocytopenia, 13% nausea/vomiting, 7% diarrhea and 10% peripheral neuropathy. By evaluating GEM and OXA PK parameters in arms A and B, no statistically significant difference was revealed between two protocols (Mann-Whitney test). GEM AUC and Cl (A vs B, p 0.84 and 0.87); OXA tot AUC and Cl (p 0.87 and 0.80); OXA ultr AUC and Cl (p 0.91 and 0.84). Pt though level progressively accumulated, from 1st to 6th cycle (Cmin 74.7–321.35 ng/ml). Conclusions: This phase II study showed that biweekly GEM-OXA as a first line chemotherapy in MTS/LA pancreatic carcinoma is well tolerated and has a excellent clinical activity. PK profile of both drugs does not show statistically significant differences between GEM-OXA and OXA-GEM alternate schedules.

4008 GemOx (gemcitabine + oxaliplatin) versus Gem (gemcitabine) in non resectable pancreatic adenocarcinoma: Final results of the GERCOR /GISCAD Intergroup Phase III C. Louvet, R. Labianca, P. Hammel, G. Lledo, F. De Braud, T. Andre, M. Cantore, M. Ducreux, A. Zaniboni, A. De Gramont Hopital St. Antoine, Paris, France; Ospidale Riuniti, Bergamo, Italy; Hopital Beaujon, Clichy, France; Clinique St. Jean, Lyon, France; Istituto Oncologico Europeo, Milano, Italy; Hopital Tenon, Paris, France; Ospedaliera C. Poma, Mantova, Italy; Institut Gustave Roussy, Villejuif, France; Poliambulanza di Brescia, Brescia, Italy

Background: The combination of gemcitabine 1 g/m2 as a 100 min infusion (10 mg/m2/mn) D1 and oxaliplatin 100 mg/m2 in 2h infusion D2 (GemOx) every two weeks showed interesting results in a multicenter phase II study (Louvet et al, JCO 2002) and was thus further explored in a phase III randomised study. Methods: Patients were stratified according to center, performance status and type of disease (locally-advanced (LA) versus metastatic (M)) and were randomly assigned to GemOx or standard treatment with gemcitabine alone (1 g/m2 in a 30 min infusion weekly). 300 patients were needed to demonstrate an absolute improvement of 20% in 8-month survival (30% to 50%). Results: 326 patients have been enrolled; 13 were non eligible, 156 were allocated to Gem arm and 157 to GemOx. Preliminary results presented last year (Louvet, ASCO 2003) were actualized and confirm a significant improvement for the GemOx arm in response rate (28.7% vs 16.7%, p = 0.02), median progression free survival (5.5 months vs 3.7, p = 0.04), and clinical benefit (38.9% vs 29.2%, p = 0.05), with a good tolerability of the GemOx combination. Final analysis (March 04) has shown median overall survival of 7.1 months (Gem) and of 9.0 months (Gemox), respectively (p = 0.13, HR 1.20 [.95–1.54]). The 8-month survival probability was 45% in the Gem arm and 56% in the Gemox arm. For LA pts (30% of total enrolled), median survivals were identical (10.3 months) in both arm, while for M pts (70%), median survival was 6.7 months (Gem) arm

Abstracts

and 8.5 months (Gemox), respectively (p = 0.17; HR 1.21 [.91–1.63]. Second-line chemotherapy was administered in 53.4% of the Gem pts (mainly platin-based) and in 52.3% of the Gemox pts (miscellaneous). Conclusion: Gemox achieved significantly better results than Gem in terms of RR, clinical benefit and PFS, but the difference did not reach the significant level for OS.

4096

Gemcitabine, oxaliplatin and weekly high-dose 5-FU as a 24-hr-infusion in chemonaive patients with advanced or metastatic pancreatic adenocarcinoma: Preliminary results of a multicenter phase II-study D. Wagner, P. Buechner-Steudel, A. Wein, R. Masri-Zada, H. Schmalenberg, G. Kleber, O. Kuss, M. Moehler, R. Behrens, W. E. Fleig Martin-Luther-University Halle/Saale, Halle/Saale, Germany; Friederich-Alexander-University, Erlangen, Germany; F.-Schiller-University, Jena, Germany; J.-Gutenberg-University, Mainz, Germany; Gastroenterology Practice, Halle/Saale, Germany

Background: Combinations of gemcitabine (GEM)/5-FU, GEM/oxaliplatin (LOHP) or 5-FU/LOHP work synergistically in pancreatic and/or colorectal malignancies, and have non-overlapping safety profiles. This phase II-study was designed to evaluate the efficacy and safety of the triple combination GEM/LOHP/5-FU in patients (pts) with advanced pancreatic adenocarcinoma. Methods: Eligibility criteria: chemonaive pts with histologically proven advanced, recurrent or metastatic pancreatic adenocarcinoma (ECOG 0–1: expected survival >3 months: measurable disease: adequate renal, hepatic and bone marrow function). According to the results of our phase I-study (Proc ASCO 2003, p 1298), pts were treated with GEM 900 mg/m2 as a 30min infusion, followed by LOHP (2-hr infusion) after a 30 min rest and 5-FU (24-hr-infusion) on d 1, 8, then every 3 weeks. The primary endpoint was tumor response, secondary endpoints were survival, time to progression, toxicity, clinical benefit and quality of life. Results: 45 pts enrolled, median follow up is 5.6 months. Pt. characteristics: m/f 31/14, median age 63 (range 45–74), ECOG 0/1 20/25, locally advanced/metastatic disease 18/27 pts. (40/60%). Median survival of all pts is 7.4 months (95% CI: 4.3–9.9). Currently 169 cycles in 32 pts. can be evaluated for toxicity: grade III/IV (NCI-CTC) toxicities occurred in 23.4/3% of cycles and were leucopenia 1.8/0%, thrombocytopenia 3/0%, anemia 2.4/0%, nausea 3.6/0%, emesis 0.6/0%, anorexia 3/1.8%, fatigue 1.2/0.6%, elevated bilirubin 2.4/0.6% or AP 2.4/0%, thrombosis/embolism 1.2/0%, somnolence 0.6/0%, hypoglycemia 0.6/0%. Analysis of response is still pending. Conclusions: The GEM/LOHP/5-FU combination is tolerated well. The current survival is promising, but has to be confirmed in a phase III study. (Supported by grants from Eli Lilly and Company, Indianapolis, IN, USA and Sanofi-Synthelabo, Paris, France)

Onkologie 2004;27(suppl 1):12–58

45

4108

Randomised trial of capecitabine plus oxaliplatin (CapOx) versus capecitabine plus gemcitabine (CapGem) versus gemcitabine plus oxaliplatin (GemOx) in advanced pancreatic cancer V. Heinemann, A. Golf, G. Seipelt, J. Bauer, T. Höhler, M. Clemens, J. M. Löhr, M. Stauch, F. Lordick, A. Hinke University of Munich, Munich, Germany; Medical Oncology, Bad Soden, Germany; Clinic Nuremberg North, Nuremberg, Germany; Johannes-Gutenberg University, Mainz, Germany; Mutterhaus der Borromaerinnen, Trier, Germany; University of Mannheim, Mannheim, Germany; Medical Oncology, Kronach, Germany; University of Technology, Munich, Germany; WiSP, Langenfeld, Germany

Objective: To compare the efficacy and tolerability of three different combination regimens, CapOx, CapGem, and GemOx for treatment of advanced pancreatic cancer. Patients: Of 165 planned pts, 135 pts have been enrolled between June 2002 and December 2003 from 37 centres. Patients (pts) received either capecitabine 2 x 1000 mg/m2 po d1–14 plus oxaliplatin 130 mg/m2 iv d1 (CapOx) or capecitabine 2 x 825 mg/m2 po d1–14 plus gemcitabine 1000 mg/m2 iv d 1+8 (CapGem) or or gemcitabine 1000 mg/m2 iv d1+8 plus oxaliplatin 130 mg/m2 d8 (GemOx). In all three treatment arms, cycles were repeated every 3 wks. Progression-free survival after 3 months was evaluated as a primary end-point. Results: At the interim analysis, 44, 45, and 46 pts had been included into the respective treatment arms. Median age was 63 yrs (range 40–74), 92% of pts presented with a KPS >70%, and 76% had metastatic disease. Stratification was performed with regard to KPS and stage of disease, and pts were accordingly well balanced between treatment groups. At the time of evaluation, 263 treatment cycles were evaluable for toxicity. Generally, hematotoxicity was low with regard to grade 3–4 anemia, leukopenia, and thrombopenia: 0%, 0%, 0% for CapOx, 18%, 0%, 0% for CapGem, and 0%, 5%, 15% for GemOx. Alopecia grade <2 occurred in 18%, 25%, and 32% of pts. Palmar-plantar erythrodysesthesia grade ≥ 2 was observed in 5% of patients each in the CapGem and the CapOx arm. Conclusion: At present time, all three treatment regimens CapOx, CapGem, and GemOx are comparably well tolerated and show a low profile of side effects. A comparative analysis of treatment efficacy will be presented at the meeting.

4119

Gemcitabine and oxaliplatine (GEMOX) in gemcitabinerefractory advanced pancreatic cancer: A phase II study J.-L. Van Laethem, M. Polus, R. Marechal, A. Demols, F. Gay, M. Marijken, A. Hendlisz, M. Peeters Erasme University Hospital, Brussels, Belgium; CHU Sart Tilman, Liege, Belgium; Universitair Ziekenhuis, Ghent, Belgium; Bordet Institute, Brussels, Belgium

Background/Aims: GEMOX regimen has been shown to be active in first line therapy of advanced pancreatic cancer; this study aims to evaluate the activity and tolerability of GEMOX in patients previously treated by standard administration of gemcitabine. Patients and Methods: Thirty patients (M/F: 16/17; median age: 57; locally advanced disease/ metastatic: n=12/21; WHO PS 0/1/2: n = 12/17/4, mesurable disease) were included. All patients had progressed under or after gemcitabine therapy (gem alone: n = 14, gem + other drug: n = 10; gem + radiotherapy: n = 9) administered as a 30 min IV infusion). The GEMOX regimen consisted of gem 1000 mg/m2 as a 100 min infusion D1 (10 mg/m2/min) and oxaliplatine 100 mg/m2 as a 2h infusion D2, q 2 weeks. Results: All patients received at least one

46

Onkologie 2004;27(suppl 1):12–58

cycle of GEMOX (median: 4; range: 1–21); dose reduction or delay was required in 6 and 8 patients (18 and 24%, respectively). Thirty patients were evaluable for response: PR = 7/30 (23.3%); SD = 8 weeks = 10/30 (30%); SD <8 weeks = 1/30; PD =11/30 (36%). Median duration of response was 5 months and TTP 4 months (range: 0.5–15 m). Median survival was 4 months. Clinical benefit was observed in 16/31 patients evaluable (51.6%). Grade III/IV toxicities occurred in 7/33 patients (21.2%) and grade I/II/III neurotoxicity was observed in 17(51%)/3 (9%) and 4 (12%) patients, respectively. There was one toxic death due to febrile neutropenia. Conclusions: The GEMOX regimen is active, well-tolerated and may be of benefit in patients who become refractory to standard gemcitabine therapy.

Gastrointestinal Cancer – Non-Colorectal/Liver B. Gastric and Esophageal Cancer 4014

Updated results of an exploratory gene expression analysis for primary esophageal cancer (PEC) patients (pts) treated with oxaliplatin (OXP), protracted infusion (PI) 5FU and radiation (XRT) L. P. Leichman, D. Lawrence, K. Clark, H. Nava, E. Nava, C. G. Leichman, G. Proulx, J. Berdzik, W. Greco, L. Pendyala Comprehensive Cancer Center of the Desert, Palm Springs, CA; Roswell Park Cancer Institute, Bufflao, NY

Background: In a study designed to explore the relationship between expression of TS, 2GCS, 2GT, ERCC-1 and XPA genes and clinical outcome for 38 pts with PEC, who received OXP (85 mg/m2 day (D)1, D15 and D29), PI 5FU (180 mg/m2 D8–D42) and XRT (5040 cGy beginning D8 for 28 fractions) TS appeared to be prognostic for survival (Leichman et al., Proc. ASCO, 22:263, 2003). Because recurrence and deaths have continued, an updated analysis for pts followed through October 2003 is presented including expression of DPD and MRP-2 genes. Methods: Pts had endoscopic biopsies of PEC prior to OXP (D1), prior to 5FU and XRT (D8) and at the end of treatment (ET). Gene expression was assayed typically in 2 specimens at each time point in 29 (ET) to 37 (D1) pts by real time QRTPCR and expressed relative to β-actin. The mean mRNA levels at each time were logarithmically transformed and using Cox proportional hazards model with time-dependent covariates explored for relationships to survival (S) and progression free survival (PFS). Results: Overall median S (n=38) and PFS (n=38) are 26.9 and 12.1 months (mo) respectively at a median follow-up of 31 and 30 mo for the censored pts. Stage 4 (n=22) versus Stages 2/3 (n=16) showed no differences in S (90% C.I. on hazard ratio: 0.9, 3.8) or PFS (90% C.I. on hazard ratio: 0.9, 3.2). Now, a year after our initial analysis, only D1 mean XPA expression showed relation to S in a model that included stage (XPA 90% C.I. on hazard ratio: 1.3, 6.9). XPA as a timedependent covariate at D1, D8 and ET for 28 pts with stage included also showed a relation to S (90% C.I. on hazard ratio: 1.0, 7.5). None of the biomarkers on D1 or as time-dependent covariates showed a relation to PFS. Conclusions: In successive exploratory analyses of PEC pts treated with OXP, 5FU and XRT, TS and XPA, gene expressions demonstrated variable time dependent relationships to survival. Expression of these genes, and their association with survival for PEC pts undergoing multimodality are appropriate for further larger studies. (Supported by CA 915490 and CA 16056)

Abstracts

4044

Phase I study of first line radiochemotherapy (RCT) with oxaliplatin (Ox), fluorouracil (FU) and folinic acid (FA) in inoperable locally advanced (LA) or metastatic (m) esophageal cancer (EC) M. Giovannini, T. Conroy, E. François, J. Seitz, V. Boige, M. Ducreux, M. Ychou, J. P. Metges, I. Sedeki, Y. Yataghene Inst Paoli Calmettes, Marseille, France; Centre Alexis Vautrin, Nancy, France; C. Antoine Lacassagne, Nice, France; CHU la Timone, Marseille, France; Inst Gustave Roussy, Villejuif, France; CRLCC, Montpellier, France; ICH CHU Morvan, Brest, France; Sanofi-Synthelabo, France, France

Background: RCT with 4 cycles (cy) FU/CDDP and 50 Gy (RTOG regimen) is the reference treatment (Tt) for patients (pts) with Inoperable EC, and is associated with 25% long-term survival.To improve these results, we assessed safety and activity of RCT using FU/FA/Ox in EC. Methods: pts with adenocarcinoma (AC), squamous cell (SC), adenosquamous carcinoma (ASC); no prior Tt; radiotherapy (RT) field <30 cm high; primary tumor ≥5 cm high or metastases; age 18–75 yrs; PS ≤ 2. Tt: 3 RCT cy followed by 3 cy Folfox 4.RCT: Ox d1, FU bolus (b) + continuous infusion (ci) + FA fixed dose 200 mg/m2 d1–2 q2 weeks (wks) with concurrent RT 50 Gy in 25 fractions/ 5 wks. 5 doses levels (mg/m2): Ox (50-75-85-100), FUb (300-400), FUci (400600), then Folfox 4, 3 cy d1–2 q2 wks. First endpoint: Maximal Tolerated Dose (MTD) of Ox/FU + RT 50 Gy (dose level at which at least 3/3 or 3/6 pts experience dose limiting toxicity defined as G4 hematotoxicity or G3–4 non-hematotoxicity except esophagitis, dysphagia, nausea and vomiting).Results: 33 pts treated; 12LA/21m; 26M/7F; 10AC/4ASC/19SC; median age 58 (35–76); median cy 6 (1–10); 56% pts completed therapy. MTD was observed at level 5 (mg/m2) Ox: 100 + FUb/ci: 400/600.Recommended doses are (mg/m2) Ox: 85 + FUb/ci: 400/600 (Folfox 4).Efficacy in 28 evaluable pts: ORR = 61% (3CR, 14PR). Median TTP and overall survival (months): 5 [95% CI: 3–6] and 9 [95% CI: 5–13].Conclusions: In EC Ox/FU/FA + RT 50 Gy is safe and active; a randomised phase II/III study comparing Folfox 4 + RT 50 Gy versus RTOG regimen is starting.

4068

Oxaliplatin (OXA), 5-fluorouracil (5-FU) and leucovorin (LV) in patients with advanced or metastatic gastric cancer (A/MGC) L. Cavanna, A. Zaniboni, F. Artioli, A. Lazzaro, A. Rizzi, M. Mazzocchi, P. Bernuzzi, R. Bertè, L. Bidin, M. Palladino Medical Oncology, Civic Hospital, Piacenza, Italy; Oncology Division, Casa di Cura Poliambulanza, Brescia, Italy; Oncology Division, Hospital of Carpi-Mirandola, Modena, Italy

Background: Treatment options for advanced or metastatic gastric cancer (A/MGC) are limited and therefore inclusion of novel substances is necessary. Few studies have confirmed the activity and tolerability of the combination of oxaliplatin (OXA) and 5-fluorouracil (5-FU) modulated with leucovorin (LV) administered to patients (pts) with A/MGC. Methods: Thirty-five patients with A/MGC treated with FOLFOX-4 regimen. All patients had a pathologically confirmed A/MGC and at least one measurable lesion, 14 pts (40%) were previously untreated and 21 (60%) received prior chemotherapy. The regimen included: LV 200 mg/sqm/d as 2-hour infusion followed by bolus 5-FU 400 mg/sqm/d and a 22-hour infusion of 5-FU 600 mg/sqm/d, repeated for 2 consecutive days every 2 weeks, OXA 85 mg/sqm on day 1, given as a 2-hour infusion in 250 ml of 5% dextrose concurrent with LV (FOLFOX-4 regimen). Pts characteristics: males-21 (60%), female-14 (40%), median age-67 years (range

Abstracts

28–78), 16 pts (45.7%) had a performance status (PS) of 0 or 1 and 19 (54.3%) had a PS of 2 (WHO/ECOH scale), histology: adenocarcinoma in all cases. The 35 pts reported in this study received 283 cycles of therapy (median number 8, range 3–16). Median follow-up time is 7 months. The most common toxicities was hematological: grade 3 hematological toxicity included anemia in 16%, neutropenia in 18.5% and thrombocytopenia in 10%. Only one patient had a grade 4 neutropenia, while no other grade 4 hematological toxicities were observed; none pts experienced neutropenic fever. Only 3 pts (8.6%) had a grade 3 diarrhea. No other grade 3 or 4 toxicities including peripheral neuropathy were reported in these pts and no toxic deaths occurred. Results: A complete response (CR) was achieved in 1 patient (2.9%), a partial response (PR) in 16 pts (45.7%), a stable disease (SD) for at least 12 weeks 10 pts (28.6%), and 8 pts (22.9%) showed tumor progression. Conclusions: This study, giving a 48.6% of best response rate (CR + PR) and a 28.6% of SD showing good activity with mild and acceptable toxicity of OXA/5-FU/LV regimen on patients with A/MGC; survival and response analyses will be presented.

4184

Oxaliplatin-based regimen as neoadjuvant chemotherapy for Chinese patients with advanced gastric cancer: Preliminary results of a phase II study J.-F. Ji, Z. Yu, X. N. Zhong, X. J. Wu, Q. Wu, Z. D. Bu, Y. Wang University School of Oncology, Beijing Cancer Hospital, Beijing, China; Beijing Cancer Hospital, Beijing, China

Background: Gastric cancer is the first cause of death per cancer in China, with more than 160,000 deaths/year. At time of resection, 75–85% of gastric cancer (GC) patients (pts) are expected to have nodal involvement and are at high risk of postoperative relapse. A Phase-II trial was conducted to evaluate the benefits of surgery combined with Oxaliplatin (OXA)-based regimen as neoadjuvant and adjuvant chemotherapy, on the survival of pts with locally advanced disease. Methods: 15 pts (Stage IIIb or IV) have been enrolled by now. All pts had histologically proven gastric adenocarcinoma and no previous palliative chemotherapy. Median age: 59 years (33–69 years), male/female ratio: 10/5, performance status: 0–2. Pts received OXA 130 mg/m2 3H-infusion day 1, leucovorin (LV) 200 mg/m2 (2H-infusion) followed by 5FU 400 mg/m2 (bolus) and 5FU 2.5g/m2 (22h-continuous infusion) day 1, repeated every 3 weeks. Efficacy was evaluated after 2 cycles. Results: All pts are evaluable for response with a more than 50% tumor reduction in 7 of 15 (46.7%)pts, SD was observed in 6 pts (40.0%) and PD in 2(13.3%).14 of 15 pts received a total 6 cycles (pre-op +or- post-op) of chemotherapy and all 15 pts came to surgery after receiving 2–6 cycles. OXA-5FU/LV was general well tolerated. The most common toxicity was Grade (Gr) 2 or 3 neutropenia and diarrhea or Gr 2 nausea/vomiting, No patients experienced Gr 4 toxicity. Neutropenic fever was not observed. An R0 curative resection was possible in 7 pts. There were no postoperative mortalities and no treatment related deaths; 14 of 15 pts are surviving (2 to 24 months) and one PD pt died of disease 2 months after surgery. Pathologic examinations of operative samples showed significant chemotherapy-induced changes in 6 pts. The trial is still open and more mature data will be available at the meeting. Conclusions: In view of the favorable response rate and toxicity profile, this protocol will be further assessed in a multicenter phase II trial.

Onkologie 2004;27(suppl 1):12–58

47

4206

A phase II study of oxaliplatin with ELF regimen in patients with advanced gastric cancer H.-M. Pan, N. Zhu, F. Lou, Y. Guo, W. Jin, H.-Z. Lou, Y. Zheng Sir Run Run Shaw Hospital - Zhejiang University, Hangzhou, China

Background: ELF (Etoposide, Leucovorin, 5-FU) regimen is widely used in gastric cancer. Oxaliplatin (Eloxatin) is a third generation platinum compound with activity against gastric cancer. A phase II study was conducted to evaluate the efficacy and safety of the combination of Oxaliplatin and ELF regimen in the treatment of patients with advanced gastric cancer. Methods: Patients received Oxaliplatin 100 mg/m2 2hr on day 1, calcium folinate 200 mg/m2 1/2 hr on day 1–3, 5-FU 500 mg/m2 2 hr on day 1–3 and etoposide 100 mg/m2 3 hr on day1–3. Cycles were repeated every 21 days. Efficacy and safety were evaluated every 3 cycles and responses were re-confirmed 4 weeks later. Forty-five patients were entered in the study: 46% male / 54% female, median age 44 years (range 28–72 yrs), 38 adenocarcinoma and 7 signet ring cell carcinoma, 64% newly diagnosed disease and 36% previously treated. Results: A total of 172 cycles were given, 39 patients were analyzed for response. There were 7 complete responses and 13 partial responses (response rate 51%), 11 stable disease (28%) and 8 progressive disease (21%). The median time to progression was 5.8 months and median overall survival was 8.2 months. All patients were analyzed for toxicity. The most common hematologic toxicity were anemia in 24.4% of patients (7/45 Gr1, 4/45 Gr2), leucopenia in 53.3% (16/45 Gr1, 4/45 Gr2, 4/45 Gr3) and thrombocytopenia in 22.2% (6/45 Gr1, 3/45 Gr2, 1/45 Gr3). No Gr4 anemia, leucopenia and thrombocytopenia were observed. The most common non-hematologic toxicity was nausea/vomiting: 48.9% of patients (15/45 Gr1, 4/45 Gr2, 3/45 Gr3). Peripheral neuropathy was noted in 44.4% of patients (12/45 Gr1, 5/45 Gr2, 3/45 Gr3; NCI scale). Grade 3 & 4 alopecia was noted in 13 cases (29%). Conclusions: This Oxaliplatin / ELF regimen shows good efficacy and an acceptable safety profile in advanced gastric cancer patients and may proved to be a suitable alternative regimen in this indication.

4213 Oxaliplatin (Ox) and raltitrexed (Ral) as first line treatment for locally advanced and metastatic gastric adenocarcinoma: Results of an ONCOPAZ phase II trial (OPHA 0141) J. I. Chacón, J. Feliú, J. A. G. Saenz, E. Fonseca, L. M. Anton, M. Bolaños, J. R. Mel, P. Escudero, J. M. Vicent, M. G. Baron Hospital Virgen de la Salud, Toledo, Spain; Hospital La Paz, Madrid, Spain; Hospital Fundación Alcorcón, Alcorcón, Spain; Hospital Universitario de Salamanca, Salamanca, Spain; Hospital Juan Canalejo, A Coruña, Spain; Hospital San Pedro de Alcántara, Cáceres, Spain; Hospital Xeral Calde, Lugo, Spain; Hospital Lozano Blesa, Zaragoza, Spain; Hospital General de Valencia, Valencia, Spain

Background: No data exist about efficacy of the combination of Ox and Ral for treatment of advanced gastric adenocarcinoma, a doublet that might be active in this disease. Methods: A multicenter phase II trial was launched in Aug 2002. Primary objectives were: response rate and toxicity. Eligible pts: locally advanced or metastatic gastric adenocarcinoma; at least 1 measurable lesion, ECOG 0–2, CrCl >65 ml/min. Therapy: Ral 3 mg/m2 IV day 1; Ox 130 mg/m2 IV day 1; cycles every 3 weeks. Treatment until progression or unacceptable toxicity. Results: 47 patients were enrolled. Data provided from 42 pts. 35 M (83.3%), 7 F (18.7%). Median age: 63 (range: 29–77). Stages: IIIA: 3 (7.5%); IIIB: 3 (7.5%); IV: 34 (85%). Total number of

48

Onkologie 2004;27(suppl 1):12–58

Cycles: 194. Median number of cycles: 4 (range: 1–13). RESPONSE RATE (by intention to treat): 36 patients valuables for response. CR: 0; PR: 15 (35,7%); EE: 8 (19%); P: 13 (30.9%); non evaluable: 6 (14%). TOXICITY: Grade 3–4 toxicities were (by number of cycles): Neutropenia: 3 (1.8%); Hemoglobin: 2 (1.2%); platelets: 1 (0.6%); AST/ALT increase: 12 (8.1%); nausea: 7 (4.2%); vomiting: 5 (3%); diarrhea: 4 (2.4%); asthenia: 10 (6.3%); mucositis: 0%; neuropathymotor: 1 (0.6%); neuropathy-neurosory: 3 (2.6%). Neurosensory toxicity grades 1–2 was present in 37% of cycles. 1 febrile neutropenia episode was produced in one patient, with septic shock and death (considered as toxic death). SURVIVAL: Median OS is of 5,9 months (IC 95%: 4.6–6.3 months). Conclusions: Oxaliplatin-raltitrexed is not enough active in gastric cancer to warrant further studies. General toxicity of the combination is very low, although unexpected severe, potentially lethal toxicity, may appear in some patients.

4237

Weekly oxaliplatin, fluorouracil and folinic acid (OXALF) as first-line chemotherapy for elderly patients with advanced gastric cancer (AGC) D. Santini, F. Graziano, V. Catalano, M. Di Seri, E. Testa, G. Catalano, S. Spalletta, B. Vincenzi, S. Cascinu, G. Tonini Campus Bio-Medico University, Rome, Italy; Medical Oncology Unit, Urbino, Italy; Medical Oncology Unit, Pesaro, Italy; Medical Oncology Unit, La Sapienza University, Italy; division of medical oncology, Ancona, Italy

Background: The weekly PLF regimen with routine filgrastim support showed high activity in elderly patients with AGC (Graziano et al, BJC, 2003). We investigated safety and efficacy of a similar weekly oxaliplatin-based regimen without filgrastim support. Methods: Chemotherapy-naive pts with AGC aged 70 or older were considered eligible for study entry. Inclusion criteria consisted of: ECOG performance status I or II, normal renal, liver, bone marrow functions and measurable disease. Frail elderly pts were excluded after baseline geriatric assessment. Pts received weekly oxaliplatin 40 mg/m2, fluorouracil 500 mg/m2 and folinic acid 250 mg/m2 (OXALF). All drugs were given intravenously on a day-1 schedule. After 8 weekly cycles, pts with complete response (CR), partial response (PR) and stable disease (SD) received six more weekly administrations. Results: 231 cycles were administered to 17 evaluable pts; median age was 75 years (70–81), M/F sex ratio was 12/5. 11 pts had two or more metastatic sites. According to intent-to-treat the overall response rate was 47% (CI: 27%–54%), with 2 CR, 6 PR, 7 SD and 2 progressions. In the whole group, time to disease progression and median survival were 5.9 months and 7.5 months respectively. Due to grade II toxicities, chemotherapy was delayed one or two weeks in 5 pts. No grade III–IV toxicities including peripheral neuropathy were reported in these patients and no toxic deaths occurred. Grade II nausea and fatigue resulted the commonest non-hematologic toxicity (respectively, 29% and 23.5% of pts). Other grade II non-hematologic toxicities were: diarrhea (11.7%), mucositis (5.8%) and transaminases increase (5.8%). No episodes of grade II neurological and hematologic toxicity were reported. Conclusions: Weekly OXALF chemotherapy showed palliative activity with favorable response rates and toxicity profile in elderly pts with AGC. More mature data will be presented during the meeting.

Abstracts

4243

A phase I study of radiation therapy (RT) with concurrent oxaliplatin (OXL), cisplatin (P) and protacted infusion 5-fluorouracil (FU) for locally advanced esophageal (ES) or gastroesophageal (GEJ) carcinoma R. Gallego, A. Cervantes, M. Pera, H. Manzano, R. Salazar, C. Conill, J. Maurel, X. Marfà, P. Gascón Hospital Clínic, Barcelona, Spain; Hospital Clínic, Valencia, Spain; Hospital Son Dureta, Palma de Mallorca, Spain; Hospital Vall d’Hebron, Barcelona, Spain; Sanofi-Synthélabo S.A., Barcelona, Spain

Background: Concurrent chemoradiotherapy with P and FU is superior to RT alone for locally advanced ES and GEJ carcinoma (Herskovic NEJM 1992) but at the cost of substantial toxicity. OXL has demonstrated activity in gastrointestinal tumors and overcome mismatch repair cisplatin-resistance in vitro. This study was undertaken to assess toxicity in patients with ES and GEJ treated with RT, OXL, P and FU. Methods: OXL, P and FU in 96 hour protracted infusion were given to 19 patients in two cycles every 4 weeks during RT (50.4 Gy in 28 fractions), using escalating doses outlined in the table. Results: Patients characteristic: 17 male, 2 female. Median age: 59 years (range 42–75); PS 0;4, PS 1;15. Loss weight (<10%;12p,>10%; 7p).Grade 3–4 neutropenia occurred in three patients (one at dose level 3 and two at dose level 4) and grade 3–4 thrombocytopenia occurred in 2 patients (one at level 3 and 1 at level 4). Non-hematological toxicity is shown in the table. The most common grade 3–4 toxicities were diarrhea (1p), mucositis (1p) and asthenia (1p). Dose level 3 was defined to be the maximum tolerated dose. 12/19 (63%) patients underway tumor resection with 2/19 pCR (10%). Conclusions: The 3rd level is feasible and well-tolerated. A phase II study in patients with resectable ES and GEJ is ongoing. Table. Non-Hematological Toxicity. Dose Level

OXL Dose (mg/sqm/IV)

P Dose (mg/sqm/IV)

FU Dose (mg/sqm/IV)

Number of p.

Grade 34/DLT

1 2 3 4

60 85 85 85

55 55 75 75

600 600 600 750

3 3 6 7

0/0 0/0 1/1 2/2

performance status was 70 (80–90)%. Therapy consisted of Ox (60/70 mg/m2) (2.5h), Gem 1000 mg/m2 (30 min), FA 200 mg/m2 (30 min) and 5-FU 750 mg/m2 (24h IV) given on day 1,8, q d 22. DLT was defined as WHO Grade IV neutropenia for ≥5 days, neutropenic fever, WHO grade 4 thrombopenia for more than 5 days or associated with bleeding or necessity of platelet transfusion, > WHO grade 2 nonhematologic toxicity except for nausea/emesis and alopecia or a delay in the subsequent cycle for more than 3 weeks. Further severe toxicities could be regarded as DLTs according to investigators decision. MTD was defined as appearance of DLTs in at least 2/6 pts. during the 1st cycle of therapy. Results: At a dose level (DL) of 70 mg Ox /m2, 1/6 pts. developed WHO grade 4 thrombopenia and urosepsis grade 4 during the 1st cycle and 2 other pts. showed WHO grade 3 neuropathy after 3 cycles which by decision of investigator was regarded as a DLT, too. In the DL of 60 mg/m2, 1/6 pts developed grade 3 neuropathy after 6 cycles. Therefore, MTD was reached at 70 mg/m2 and DL 60 mg/m2 will be subjected to further evaluation. 11 pts. were evaluable for response with 1 CR, 3 PR, 3 MR, 2 SD and 2 PD. Median duration of response was 10.4 weeks (4.7–14.7). Median survival was 7.1 months. Conclusions: The combination of 60 mg Ox, 1000 mg Gem, 200 mg/m2 FA and 750 mg/m2 5-FU is a feasible outpatient drug combination that should be further evaluated in a phase II setting.

4266

Phase I dosing-escalating study of oxaliplatin in combination with oral tegafur-uracil/leucorvin in patients with advanced gastric cancer J.-S. Chen, J.-S. Huang, T.-S. Yang, Y.-C. Lin, H.-M. Wang, C.-T. Liau, K.-M. Rau Chang Gung Memorial Hospital, Taipei, Taiwan Republic of China; Chang Gung Memorial Hospital, Keelung, Taiwan Republic of China; Chang Gung Memorial Hospital, Kaohsiung, Taiwan Republic of China

Background: In patients (pts.) with esophageal cancer (EC) chemotherapy is considered in the palliative situation or for downstaging. Cisplatin/5-FU combinations (partly combined with radiotherapy) are among the most commonly used drugs but do require hospitalization to guarantee sufficient hydration. Oxaliplatin (Ox) is a less nephrotoxic platin derivative which reduces the need for hydration. In this study we tried to define dose limiting toxicity (DLT) and the maximal tolerable dose (MTD) of Ox when combined with Gemcitabine (Gem), 5-FU and folinic acid (FA). Methods: 13 (11m/2f) pts. with inoperable EC (3 adenocarcinoma/10 squamous cell cancer) were included. Median age was 59 (46–77) years, median Karnowski

Background: The aim of this dosing-finding study was to determine the maximum-tolerated dose (MTD), the dose-limiting toxicities (DLTs) and to define the recommended dose of biweekly oxaliplatin (OXA) (Oxalip®, TTY Biopharm Company, Taipei, Taiwan) with a fixed dose of oral tegafur-uracil (UFUR®, TTY Biopharm Company, Taipei, Taiwan) / leucvorin (LV) in patients with advanced gastric cancer (GA). Secondary objectives included evaluation of the safety and anti-tumor activity. Methods: Between Oct 2001 to Aug 2003, 28 chemo-naive patients with advanced GA were enrolled. OXA was administrated as 2-hour infusion (dose from 55 mg/m2, 70 mg/m2, 85 mg/m2, 100 mg/m2, and 115 mg/m2) on day 1 and 15 every 28 days with tegafur-uracil 300 mg/m2/day and LV 60 mg/day orally three times a day from day 1 to day 21, followed by one-week rest. DLTs were defined as grade IV hematologic toxicity or grade III non-hematologic toxicity in the first course of treatment. Safety and efficacy were evaluated throughout treatment. Dose escalation to the next cohort was based upon standard criteria for phase I trial. Results: There were 11 females (39.7%) and 18 males (60.7%) with a median age of 58.5 (32–72). The dose level of OXA 100 mg/m2 was found to be MTD. The most common DLT is diarrhea. Major grade III/IV toxicities of the 28 patients were: vomiting 4 (14.3%), diarrhea 4 (14.3%), renal 2(7.1%), leukopenia 1(3.6%), and thrombocytopenia 2 (7.1%). There were two treatment-related deaths due to neutropenia infection and upper gastrointestinal bleeding, respectively. The response among 28 patients were partial response 13 (46.4%, 95% CI 26.74% to 66.12%), stable disease 9 (32.1%), and 6 (21.5%) progressive disease, respectively. By the end of Nov. 2003, 14 patients were expired. The median time to treatment failure, time to progression and overall survival were 124 days, 107 days and 407 days, respectively. Conclusions: The recommended dose for phase II study is OXA 100 mg/m2

Abstracts

Onkologie 2004;27(suppl 1):12–58

4250

Gemcitabine, oxaliplatin, 5-FU and folinic acid as an outpatient first-line therapy in patients with inoperable esophageal cancer: A phase I study J. M. Stieler, A. Hilbig, U. Pelzer, L. Roll, A. Helm, M. Gövercin, H. Riess, H. Oettle Charite Universitätsmedizin Berlin Campus Virchow Klinikum, Berlin, Germany

49

biweekly with fixed dose of oral tegafur-uracil 300 mg/m2/day and LV 60 mg/day three times a day for 21 days in cycle of 28 days. The most common DLT is diarrhea. The response was 46.4%.

4267

Raltitrexed plus oxaliplatin and leucovorin modulated continuous 5-FU infusion (ROLF) in metastatic (MGC) and local advanced (LAGC) gastrointestinal cancer. A phase I / II study E. J. Johnson-Buarque Cettro - Oncologia/Hematologia, Brasilia, DF, Brazil

Background: Raltitrexed (RX), oxaliplatin (ox) and 5-FU have shown synergistic and addictive activity against gastrointestinal cancer. The objective of the study was to evaluated the efficacy and toxicity of this combination and find the best oxaliplatin dose. Methods: Between April 2001 and April 2003 23 patients (pts) with MGC and LAGC (16 gastric and 7 colorectal cancer) were treated every 3 weeks with: day 1 RX 2,5 mg/m2, followed by OX 85 mg/m2 (after third cycle if no grade 3 or 4 toxicity increased to 100 mg/m2). 5-FU 1000 mg/m2 24hs continuous infusion, initiated after OX and lasted for 3 days modulated with leucovorin 50 mg each 6 hours. Pts: 13 male / 10 female median age 64, 19 pts with more than one site of metastasis (mets). (9 pts with carcinomatosis and 9 with liver mets. Results: Toxicity: 5 pts experienced grade 3/4 stomatites and/or diarrhea. Only 2 pts with grade 3/4 leucopenia and 1 pt with neurotoxicity grade 3. Response: 19 pts were evaluated for response. CR-10.5% PR 52.5% SD- 21% and Progr.16%. Median duration of response 7 months; median survival 16+ months, with 11 pts still alive. Conclusions: ROLF is an active scheme advanced gastric and colon cancer with low toxicity and in 40% pts a dose of 100 mg/m2 of OX is safe and could improve response.

Gastrointestinal Cancer – Non-Colorectal/Liver C. Hepatocellular and Biliary Cancer 4086

Gemcitabine plus oxaliplatin in advanced hepatocellular carcinoma (AHCC): Results of a phase II study J. Taieb, T. Mansourbakht, M. Ducreux, L. Bonyhay, T. De Baere, M. Delgado, T. Poynard, V. Boige Pitié Salpétrière Hospital, PARIS, France; Institut Gustave Roussy, Villejuif, France

Background: New therapies are needed to improve the prognosis of patients (pts) with AHCC. The rationale to develop the gemcitabine / oxaliplatin combination in AHCC is based on: (i) the synergy between these two drugs, (ii) the clinical activity of these drugs in HCC according to preliminary phase II results, (iii) the lack of renal or hepatic toxicity of oxaliplatin in cirrhotic pts. Recently, we conducted a pilot study to evaluate two distinct bimonthly schedules of gemcitabine plus oxaliplatin in advanced HCC. We have therefore selected the most promising schedule to perform a phase II trial in non pretreated AHCC patients. Methods: 30 pts with non pre-treated AHCC were prospectively enrolled. They received gemcitabine 1000 mg/m2 d1 and oxaliplatin 100 mg/m2 d2 (GEMOX). Treatment was repeated every two weeks until disease progression or limiting toxicity. Eligibility criteria were: pathologically proven advanced HCC or alpha-fetoprotein (AFP) levels over 250 ng/ml associated with a radiological liver tumor, PS (ECOG) 0–2, age ≥18, measurable disease, adequate hematological and renal functions, compensated Child A or B cirrhosis, and written informed consent. Results: Twenty six patients are

50

Onkologie 2004;27(suppl 1):12–58

currently evaluable for toxicity and efficacy. Their characteristics, treatment toxicity and clinical activity are summarized in the following table. No Toxic death occurred. Hematological grade 3–4 toxicities consisted in thrombocytopenia (18%), anemia (18%) and neutropenia (30%), without any febrile neutropenia or bleeding event. No grade3–4 non hematological toxicity was observed. Grade 1 and 2 neurotoxicity (specific scale) was observed in 14 and 6 pts, respectively. Four objective responses were observed, one leading to a surgical resection of the tumor. Conclusions: gemcitabine plus oxaliplatin seems to be an active regimen with manageable toxicity in non pretreated AHCC patients.

Age Sex PS (mean) (M/F) (0/1/2)

GEMOX (n=26)

60

24/2

5/15/6

N Cycles

211

Grade Disease Significant 3–4 control AFP toxicity (PR+SD) decrease

42%

69%

46%

4091

A multicenter phase II study of capecitabine plus oxaliplatin (CapOx) in advanced biliary system adenocarcinomas O. Nehls, H. Oettle, J.-T. Hartmann, R. Hofheinz, A. Hochhaus, M. Makowski, D. Arnold, C. Burkart, M. Gregor, B. Klump University Hospital Tuebingen, Tuebingen, Germany; Charite, Campus-Virchow-Clinic, Berlin, Germany; University Hospital Mannheim, Mannheim, Germany

Background: To evaluate the feasibility of capecitabine and oxaliplatin combination therapy (CapOx) in unresectable or metastatic adenocarcinomas of the biliary system. Patients and Methods: 56 pts (25M, 31F) were included (median age, 63 yrs). Major eligibility: histologic proven, measurable disease, age 18–75 yrs, ECOG PS 0–2. A total number of 250 cycles (median: 5; range 1–16) of oxaliplatin (130 mg/m2, d1) plus capecitabine (2000 mg/m2, d 1–14) were administered 3 weekly for gallbladder carcinoma (GBC) (20 pts), extrahepatic (20 pts), and intrahepatic (16 pts) cholangiocarcinoma (CCC). Pts were assessed for response according to WHO standard criteria. Clinical outcome was determined separately for pts with either GBC/extrahepatic CCC or intrahepatic CCC (mass-forming type), because these two distinct presentations have been supposed to differ substantially from each other according to clinical behaviour. Results: Grade 4 toxicities (WHO) were diarrhea in 1 pt (1% of cycles), thrombocytopenia in 1 pt (1%), leukopenia in 1 pt (1%), and fever in 2 pts (1%); grade 3 toxicities were: diarrhea in 2 pts (1%), thrombocytopenia in 3 pts (2%), and fever in 1 pt (1%). Grade 3/4 peripheral sensory neuropathy (Lévis scale) was found in 10 pts (15%). Two pts were removed from study due to oxaliplatin-related allergic reactions. One patient harboring intrahepatic CCC died with septic shock after the first treatment cycle. On 31 evaluable pts with GBC or extrahepatic CCC the response-rate was as follows: complete response (CR), 2 pts (6% of the pts); partial response (PR), 7 pts (23%); stable disease (NC) 15 pts (48%). Thus, the overall disease control rate (CR + PR + NC) was 77%, whereas 7 pts (23%) had progressive disease (PD). In contrast, in 16 evaluable pts with intrahepatic mass-forming CCC no CR or PR was observed, but 4 pts (25%) were found to have NC, and in 12 pts (75%) PD was encountered. Conclusions: Our preliminary data suggest the CapOx regimen to be well tolerable and highly active for advanced GBC and extrahepatic CCC (disease-control rate: 77%), whereas outcome might be poorer in intrahepatic CCC. Survival data will be presented at the meeting.

Abstracts

4133

Intra-hepatic lipiodol I 131 combined with oxaliplatin, infusional FUDR, leucovorin in hepatocarcinoma and hepatic metastasis F. P. Costa, R. Schemerling, O. F. Costa, P. A. Costa, C. Albertotti, S. Martins, A. C. Buzaid, F. Maluf, I. Mayer, R. Marques Hospital Sírio Libanês, São Paulo, Brazil

Background: We conducted a pilot study with L131 internal hepatic radiation combined with IA OHP, infusion FUDR and LV in 19 AHC and 20 HM pts, to assess feasibility, response rate, symptom relief, TTP and 1-year OS. Methods: Median age was 60 yo (20–76), 32 pts were male. All pts had predominant hepatic disease and symptomatic progression prior to inclusion. Portal vein thrombosis was documented in 3 pts. KPS was 70% in 4 pts and ≥80% in 35 pts. All HM and 10 out of 19 AHC pts have had prior treatment (range: 1–9). Out of 19 AHC pts, 12 were Child-Pugh A, 7 were B and 2 were C. Out of 20 HM, 14 were colorectal tumors. All pts were assigned to receive only one treatment. However, re-treatment was allowed after documented liver progression by CT or MRI in responders or to reach a tumor radiation dose ≥120 Gy. Therapy was a single L131 intra-hepatic infusion dose followed by OHP 100 mg/m2 IA over 2h on D1, and FUDR 60 mg/m2/d + LV 15 mg/m2/d IA by 96h infusion (D1–4). G-CSF SC was given from D5–15. The L131 dose was planned for each pt in order to deliver >120 Gy to tumor mass and <50 Gy to normal liver parenchyma. To date, 45 treatments were delivered (whole liver - 20; R lobe - 15; L lobe - 3). 6 pts got more than one treatment. The L131 median dose was 50 mCi (range: 25–80). Results: 10/2001 till 10/2003

AHC (n= 19)

HM (n= 20)

PR SD/ Clinical Benefit Median Tumor dose (range) Median TTP (range) Median Follow-up (range) 1-year OS

67% 6% / 52% 180 Gy (46–688) 6.9 mos. (1–17) 6.1 mos. (1–25) 22%

27% 27% / 39% 228 Gy (46–463) 3.4 mos. (1–8.3) 5,3 mos. (1–13) 5%

All pts developed severe fatigue post-procedure lasting 7 to 14 days. Other grade III/IV toxicity included thrombocytopenia (28%), elevation of liver enzymes (17%), neutropenia (15%), enteritis (11%), diarrhea (8%), hepatic encephalopathy (8%), mucositis (6%) and fever (4%). No clinical benefit was observed in Child B and C AHC pts. 2 AHC had treatment related death (liver failure and gastric bleeding). 3 other pts (1 HM, 1 Child B and 1 Child C AHC) died within 60 days of therapy due to PD. Conclusion: ICM is a feasible and active in refractory AHC and HM pts. An improved TTP trend was observed for AHC (p=0.057). A phase II study is ongoing for HM and AHC pts, restricted to Child-Pugh A.

4169

Phase II study of oxaliplatin in patients with unresectable, metastatic or recurrent hepatocellular cancer Y. Yen, J. Doroshow, L. Leong, D. Lim, L. Wagman, R. Morgan, P. Frankel, H. Lenz, D. Gandara, S. Shibata City of Hope Comprensive Cancer Center, Duarte, CA; USC, Los Angeles, CA; UC Davis, Sacramento, CA

Background: Prolonged survival for patients (pts) with unresectable hepatocellular cancer (HCC) has remained elusive, with median overall survival consistently less than 6 months. Oxaliplatin has re-

Abstracts

cently demonstrated promise activity in HCC. The aim of this study is to determine the response rate, survival, time to progression and toxicity in pts with poor prognosis HCC treated with single agent oxaliplatin. Methods: pts were required to have measurable recurrent, metastatic or unresectable HCC, exposed to no more than 2 prior chemotherapy regimens. Prior radiation, chemoembolization, and/or alcohol injections were permitted. Karnofsky performance of 70% or above and adequate organ and hematologic function were required. All patients (pts) received treatment with oxaliplatin 100 mg/m2 on day 1 and 15 as a two hour IV infusion and were pretreated with antiemetics. Treatment was repeated every 28 days. Results: Fifteen pts have been enrolled and fourteen eligible patients are evaluable for response; there are five Caucasians, one African-American, and eight Asians. 12 are males and 2 females. Median age at study entry was 62 years, (range 34–80). Karnofsky performance status was 70/80/90/100% in 1/1/3/9 pts; thirteen pts had previous surgery. Among the 14 evaluable pts, one PR was deserved and six had stable disease. The pt experiencing a PR normalized his AFP; has only a small residual liver mass remaining; and is free of progression at nine months. Median time to progression or death on this study is 2.7 months (mo); the overall median survival is 10 mo. Six mo survival is 51% (95% CI 28%–94%) and 6 mo event-free survival is 29% (95% CI 11%–72%). Five pts died from progressive disease, and one pt died of pneumonia. Grade III toxicities included neutropenia (1), anemia (1), fatigue (2), transaminitis (2), hyperbilirubinemia (1), elevated alkaline phosphatase (1), hyperglycemia (1), hyponatremia (1), and neuropathy (1). Conclusion: Oxaliplatin as a single agent has demonstrated potential activity in the treatment of poor prognosis HCC; accrual to this study will continue past the interim analysis. (Supported by CA62505 and N01-CM17101)

Genitourinary Cancer 4534

A phase II multicentre study of oxaliplatin (Ox) in combination with paclitaxel (Px) in patients (pts) who failed cisplatin (CDDP) based chemotherapy (CT) for germ cell tumors (GCT) C. Theodore, A. Flechon, K. Fizazi, C. Chevreau, J. P. Delord, J. P. Lotz, L. Geoffrois, P. Kerbrat, B. Bui, Y. Yataghene IGR, Villejuif, France; C. Leon Berard, Lyon, France; Inst Claudius Regaud, Toulouse, France; Hop Tenon, Paris, France; Centre Alexis Vautrin, Nancy, France; C. Eugene Marquis, Rennes, France; Inst Bergonie, Bordeaux, France; Sanofi-Synthelabo, France, France

Background: pts with either CDDP refractory (REF) or with a poor prognosis relapse (REL) GCT, defined as progressing less than 2 or between 2 and 6 months (mos) of CDDP administration respectively have dismal prognosis.Px induces 11 to 26% response rates (RR) in a similar pt population.Ox is a platinum compound with a lack of cross resistance with CDDP. Methods: REF and REL pts, who had received at least one line of CT containing Etoposide+CDDP. Treatment: combination of Px (175 mg/m2, 3h infusion) followed by Ox (130 mg/m2, 2h infusion) administered q3 weeks. Endpoints were RR (WHO criteria) and tumor marker decrease (TMD) >90% or >50%.The 2-step modified Gehan design was used for sample size determination. Results: 26 pts were enrolled; median age 31 yrs (range 18–46); PS 0/1:13/13; 18 had non seminoma (among whom 5 mediastinal primaries), 2 seminoma and 6 mixed GCT; 16 pts were REF and 10 pts were REL; 22, 3 and one pts had previously received 2, 3 and 5 lines of CDDP based CT. A total of 87 cycles (median 4, range 1–7) were administered. The worst toxicity (NCI-CTC) consist-

Onkologie 2004;27(suppl 1):12–58

51

ed in: grade (G) 3 and 4 neutropenia (N) in 3 and 4 pts respectively, G 3 anemia in 3 pts, G 3 thrombocytopenia in one pt, G 3 neuropathy in one pt and G3 infection without N in one pt. One Partial Response with normal TM (lasting 18+ mos) and 4 TMD >90% and 3 TMD >50% were observed. Median Time to Tumor Progression and Overall Survival (months) were 2.0 [95% CI:1.5–2.5] and 8.8 [95% CI: 5–12] respectively. However 2 pts were eventually salvaged by surgical exeresis of limited disease and are alive free of disease progression at 13+ and 23+ months respectively. Conclusions: Ox-Px Combination is safe and looks active in refractory and relapsed GCT, it warrants further developments within the treatment strategy in these patients.

4535

Oxaliplatin and irinotecan plus granulocyte-colony stimulating factor (G-CSF) as third-line treatment in relapsed or cisplatin-refractory germ-cell tumor (GCT) patients: A phase II study D. Farmakis, D. Pectasides, M. Pectasides, M. Nikolaou, M. Koumpou, A. Gaglia, A. Koumarianou, G. Aravantinos, T. Economopoulos Metaxas Memorial Cancer Hospital, Piraeus, Greece; Attikon University Hospital, Athens, Greece; Ag. Anargiri Cancer Hospital, Athens, Greece

Background: Oxaliplatin is an active agent in patients with cisplatinrefractory GCT, while the combination of irinotecan with platinum compounds has also induced responses in such patients. Given the synergistic activity of the two agents, we investigated the efficacy and tolerability of their combination in patients with relapsed or cisplatinrefractory GCT. Methods: Eighteen patients with relapsed or cisplatin-refractory GCT were treated with oxaliplatin 85 mg/m2 (days 1, 15) and irinotecan 80 mg/m2 (days 1, 8, 15) every 4 weeks for a maximum of 6 cycles. Results: All patients were assessable for response and toxicity. Seven patients (40%) achieved a favorable response (4 complete and 3 partial responses). One of the complete responders relapsed after 2.5 months and, despite treatment with high dose chemotherapy, he died 2 months later; the rest 3 patients are continuously disease-free for 11+, 14+ and 19+ months. The partial responders progressed subsequently and died after 2, 3 and 4.5 months, respectively. None of the patients with extragonadal mediastinal GCT responded to chemotherapy. The regimen had a good tolerance. Neutropenia (grade 3/4, 17%), neutropenic infections and sepsis were not common probably due to prophylactic G-CSF administration. Thrombocytopenia and anemia were not a serious problem. Grade 3/4 diarrhea and nausea/vomiting were noted in 22% and 28% of patients, respectively. Oxaliplatin-associated neurotoxicity was rather low; grade 3 peripheral sensory neuropathy was recorded in 11%. Conclusions: The combination of oxaliplatin and irinotecan is feasible and associated with significant clinical antitumor activity, mild and manageable toxicity and easy outpatient administration in patients with relapsed or cisplatin-refractory GCT. This combination offered a possibility for long-term disease-free status (17%), despite the poor prognostic features of the study group.

4544

Gemcitabine and oxaliplatin combination: A multicenter phase II trial in unfit patients with locally advanced or metastatic urothelial cancer A. Font, E. Esteban, J. Carles, M. A. Climent, J. L. Gonzalez-Larriba, A. Berrocal, J. Bellmunt, I. Garcia-Ribas, X. Marfa, X. Fabregat Hospital Germans Trias i Pujol, Badalona, Spain; Hospital General de Asturias, Oviedo, Spain; Hospital del Mar, Barcelona, Spain; Instituto Valenciano de Oncologia, Valencia, Spain; Hospital Clinico San Carlos, Madrid, Spain; Hospital General de Valencia, Valencia, Spain; Hospital Valle de Hebron, Barcelona, Spain; Eli Lilly & Co, Alcobendas, Spain; Sanofi Sinthelabo, Barcelona, Spain

Background: Although cisplatin is considered the cornerstone for the treatment of advanced bladder cancer, up to 50% of the patients (pts) cannot receive it because they are considered unfit mainly due to poor renal function. New non nephrotoxic regimens are needed. Gemcitabine and oxaliplatin are active, non-nephrotoxic and with non-overlapping toxicity and therefore being attractive to be studied in combination for unfit bladder cancer patients. Methods: Thirty-six pts unfit (mainly due to renal function, creatinine clearance <60 mL/min) to receive standard cisplatin-based treatment were included in this multicentric phase II. We describe here the first 27 patients and give response data from the first 12 evaluable pts. Median age was 69 years (range 60–81), 24 male and 3 female, median ECOG was 1 (range 0–2). Three pts had received previous adjuvant chemotherapy and 2 were previously treated with radiotherapy. Fifteen pts had one or two metastatic sites and 12 pts had more than two metastatic sites. Calculated median creatinine clearance was 48 mL/min (range 30–76). Patients were treated with gemcitabine 1200 mg/m2 d 1 and 8 and oxaliplatin 100 mg/m2 on day 8 every 21 days. Results: Median number of cycles was 3 (range 1–6). A total of 74 cycles are evaluable for toxicity. Grade 3–4 hematological toxicity was mild. Grade 3 anemia was reported in 19% of the cycles Grade 3 thrombopenia in 11.5% and Grade 3–4 neutropenia in 4.6% of the cycles. Hospitalization due to neutropenia and fever was required for 2 pts. So far 12 pts are evaluable for efficacy for an overall response rate of 50% (6 partial responses, 5 stabilizations and 1 progression).Conclusion: Gemcitabine-oxaliplatin is an active and tolerable combination in unfit patients with locally advanced or metastatic urothelial cancer.

Gynecologic Cancer 5020

A phase I–II trial of gefitinib in combination with vinorelbine and oxaliplatin as salvage therapy in women with advanced ovarian cancer (AOC) D. Mavroudis, E. Efstathiou, A. Polyzos, A. Athanasiadis, G. Milaki, E. Kastritis, A. Kalykaki, Z. Saridaki, A. Dimopoulos, V. Georgoulias University General Hospital of Heraklion, Crete, Greece; Alexandras Hospital, Athens, Greece; Laikon Hospital, Athens, Greece; General Hospital, Larissa, Greece

Background: To evaluate activity and tolerability of gefitinib (Iressa) in combination with vinorelbine and oxaliplatin as salvage therapy in women with AOC. Methods: Women with AOC recurrent or refractory after ≥ 1 previous line of platinum-containing chemotherapy who had measurable disease by RECIST criteria or assessable by Ca-125 received oral gefitinib 250 mg/day with vinorelbine 25 mg/m2 and oxaliplatin 50 mg/m2 on days 1 and 8 every 3 weeks without prophylactic growth factor support. Gefitinib was continued until disease progression. Results: So far 33 patients (pts) have been entered, (10 CDDP-sensitive; 23 CDDP-refractory disease). Among the first 10

52

Onkologie 2004;27(suppl 1):12–58

Abstracts

pts, 4 DLTs were observed in the first cycle (2 febrile neutropenias, 1 grade 4 neutropenia lasting >5 days, 1 grade 3 diarrhea) resulting in dose reduction for vinorelbine (20 mg/m2) and oxaliplatin (40 mg/m2). In the next 10 pts only 2 DLTs (febrile neutropenia and grade 4 neutropenia lasting >5 days) were observed and the study is ongoing as phase II. All pts were evaluable for toxicity and 29 (19 CDDP-refractory; 10 CDDP-sensitive) for response (2 too early, 1 protocol violation, 1 lost to follow up). Three complete (CR) and 2 partial responses (PR) were seen in the CDDP-refractory group (ORR 23.8%; 95%CI: 5.6–42.0%) and 4 CR with 5 PR in the CDDP-sensitive group (ORR 90%; 95%CI: 71.4–100%). Median duration of response was 5.2 months (range 1–6) for CDDP-refractory and 6.6 (range 1–10) for CDDP-sensitive; median time to progression was 4.1 months (range 1–14) and 8.6 (range 1–12), respectively. So far 8 pts have died (7 disease progression; 1 pulmonary embolism), 14 had disease progression, and 11 continue treatment. 134 cycles have been administered: median 3 (range 1–9) cycles per pt. 2oxicity included grade 3/4 neutropenia each in 8 (24%) pts, febrile neutropenia 4 (12%), grade 3 anemia 1 (3%), grade 3 diarrhea 3 (9%), neurotoxicity 1 (3%), rash 1 (3%) and transaminase elevation 1 (3%). Conclusions: The concurrent administration of gefitinib with vinorelbine and oxaliplatin is feasible and shows promising activity in pretreated pts with AOC.

5023

Phase II trial of oxaliplatin (OXA) combined with paclitaxel (P) inplatinum + taxanes sensitive advanced ovarian cancer (AOC) patients: Final results P. Viens, T. Petit, A. Yovine, P. Bougnoux, G. Deplanque, P. Cottu, R. Delva, J. P. Lotz, J. S. Van Belle, J. M. Extra Paoli-Calmettes, Marseille, France; Centre Paul Strauss, Strasbourg, France; CAC, Le Kremlin-Bicêtre, France; Hôpital Bretonneau, Tours, France; Institut Claude Bernard, Metz, France; Hôpital St Louis, Paris, France; Centre Paul Papin, Angers, France; Hôpital Tenon, Paris, France; UZ-Gent, Gent, Belgium; Institut Curie, Paris, France

Background: OXA is active in mismatch repair (MMR) deficient cells, a condition prevalent in pretreated AOC. OXA/P shows high activity in investigator initiated studies (Faivre et al, Ann Oncol 10:1125, 1999). We evaluated efficacy and safety of OXA/P in AOC pts with clinically evaluable disease pretreated with 1 platinum based regimen ± taxanes and a progression-platinum-free interval (PPFI) ≥6 months (mo). Methods: From Dec 99 to Apr 02, 105 pts, 14 centres received P 175 mg/m2, 3h iv followed by OXA 130 mg/m2, 2h iv, every 3 weeks, for up to 9 cycles (cy) without G-CSF. Median (m) age: 58 years (range 37–83), 96% PS 0/1. PPFI: 69 pts >12 mo, 34 pts 6–12 mo; 80 pts (76%) taxane-pretreated. Results: 708 cy administered; m 7 cy/pt (1–9). Relative dose intensity: OXA 0.95, P 0.97. Efficacy: 98 pts eligible, 7 noneligible (3 bowel occlusion, 3 >1 prior platinum-based chemotherapy, 2 PPFI not determinable). 98 pts evaluable for response (see table) ORR: 81% (95%CI 71–88), all but 3 responses third party reviewed; 90 pts evaluable for biological response (baseline CA125 ≥50 UI/ml): 89% response (decrease >75% or CA125 normalization >1 mo). Combined RECIST /biological response: 88% (95%CI 80–94). M followup: 27 mo (14–48); m PFS: 10.2 mo; projected OS >28 mo (44 events). Safety (NCI-CTC): % pts with grade (gr) 3/4: neutropenia 21/46; thrombocytopenia 15/1; nausea-vomiting 9/2; asthenia 7/1; allergic reaction: 1/1; cumulative reversible neurotoxicity gr 2/3 (OXA specific scale): 50/25 after a m 7 cy (3–9); febrile neutropenia: 9 episodes/8 pts; 4% cy required G-CSF; no treatment-related deaths. Conclusion: The OXA/P combination shows the highest level of efficacy in platinumsensitive third party reviewed large phase II studies in platinum/taxane pretreated AOC pts with PPFI >6 mo, and can be administered in an outpatient setting without special safety measures.

Abstracts

Response (RECIST)   Prior taxanes No prior taxanes Total N=75 (%) N=23 (%) PPFI (mo) CR+PR SD PD

6–12 N=28 24 (86) 2 (7) 2 (7)

>12 N=47 35 (74) 8 (17) 4 (9)

6–12 N=2 1 (50) 1 (50) –

>12 N=21 19 (90) 2 (10) –

79 (81) 13 (13) 6 (6)

5034

Gemcitabine and oxaliplatin followed by paclitaxel and carboplatin as first line therapy for patients with advanced epithelial ovarian cancer. A phase II trial of sequential doublets. (GO-FIRST) K. Chrystal, C. B. Steer, K. A. Cheong, E. Galani, A. Strickland, F. Lofts, C. Gallagher, H. Thomas, P. Harper Guys Hospital, London; St Georges Hospital, London; St Bartholomew’s Hospital, London; Royal Surrey County Hospital, London

Background: Gemcitabine and oxaliplatin are active in epithelial ovarian cancer with minimal overlapping toxicity, and in pre-clinical studies have shown synergy. The primary objective of this prospective phase II study is to evaluate the response rate (RR) and toxicity of gemcitabine and oxaliplatin (GO) administered sequentially with paclitaxel and carboplatin (PC). Secondary objectives were time to progression, quality of life outcomes and overall survival. Methods: Patients with histologically proven epithelial ovarain cancer, FIGO stage II–IV, measurable disease (RECIST), and PS 0–2 were eligible. Treatment consisted of gemcitabine 1250 mg/m2 D1, D8 and oxaliplatin 135 mg/m2 D8 q21days for 4 cycles followed by carboplatin AUC6 and paclitaxel 175 mg/m2 D1 q21days for 4 cycles. Interval debulking was permitted after 4 cycles. Results: 20 patients are enrolled. Median age 62 yrs (range 39–78), FIGO III–IV (16/4). 12/20 pts were only able to have an initial biopsy and of these, 7 were subsequently debulked. 4 cycles of GO were given prior to the planned 4 cycles of PC. The RR to 4 cycles of GO was 80% (CR4/PR12). With subsequent debulking surgery and further chemotherapy (PC) the final RR was 85% (13 CR/4 PR). One pt died of a PE and one progressed on GO and died. At a median F/U of 21 mths 15 pts are still alive, 6 without progression. The median TTP was 15.2 mths. Grade 3/4 toxicities for GO: nausea and vomiting 4pts (20%), neutropenia 4pts (20%). PC: alopecia 14 (70%), neutropenia 2 (10%), thrombocytopenia 1 (5%). Peripheral neuropathy was the major toxicity. 4 pts developed G2 and a further 1pt G3 on GO. With sequential PC, 8 pts developed G2 and 5 pts G3 neuropathy requiring dose reduction in 3 pts and cessation of paclitaxel in 6 pts. Conclusions: GO is a new doublet with activity in ovarian cancer, but sequential PC should not be used given the cumulative neurotoxicity.

Onkologie 2004;27(suppl 1):12–58

53

5075

Phase II trial of oxaliplatin (OXA) combined with paclitaxel (P) as first-line chemotherapy for patients (pts) with advanced ovarian cancer (AOC): Preliminary results G. Deplanque, A. Goupil, M. Fabbro, E. Levy, O. Rixe, B. Benettaib, E. Cvitkovic Clinique Cluade Bernard, Metz, France; Centre René Huguenin, St Cloud, France; CRLC Val d’Aurelle, Montpellier, France; Hôpital European Georges Pompidou, Paris, France; Hôpital de la PitiéSalpétrière, Paris, France; CAC, Le Kremlin-Bicêtre, France

Background: Despite the introduction of paclitaxel/platinum combinations in first line treatment of AOC, a small proportion of patients fail to respond. This has been linked to genetic or epigenetic deficiency in activity in mismatch repair (MMR; MLH1, MSH2). OXA has activity in MMR deficient cells and the OXA/ P combination has been shown to be safe and highly active in pre-treated AOC pts (Faivre et al. Ann Oncol 10:1125, 1999). This motivated the present phase II study in first line AOC. Methods: AOC pts with histologically proven stage IIc or IIIa-c after maximal debulking surgery (microscopic disease or macroscopic disease ≤2 cm) performed 2–7 weeks before study treatment start, received 175 mg/m2 P, 3-hour intravenous infusion, followed by 130 mg/m2 OXA, 2-hour intravenous infusion, every 3 weeks, on an outpatient basis, without G-CSF. A maximum of 6 cycles were administered before evaluation by second look laparoscopy or laparotomy. Pts with positive histology after second look surgery could receive up to 2 further cycles; pts with pathological complete response received no further treatment. Results: Between Aug 2001 and Jan 2003, 17 pts were entered in 9 centres (49 planned), 14 eligible, 3 ineligible (2 colorectal cancer with peritoneal presentation, 1 pre-treated with OXA/cisplatin/P). Median age 59 years (range 40–68); Performance status 0/1/2: 10/3/1; histology serous: 11, other: 3; FIGO stage IIIb: 5, IIIc: 9; microscopic residual disease: 3, macroscopic: 11. Patients received a median of 6 cycles (6–8). Efficacy: Of 13 pts evaluable for pathological response, 8 had complete response (62%) and 5 stable disease. With a median follow-up of 14 months (6–24), only 2 pts with macroscopic residual disease progressed. Median PFS: 14 months (5–22), 9 out of 12 patients have an ongoing PFS >12 months. Safety (NCI-CTC; 91 cycles) % by pts: neutropenia grade (Gr) 3/4: 7%/14%; febrile neutropenia: 7%, reversible neurosensory toxicity (OXA specific scale) ≥Gr2: 21%. Conclusion: The OXA/P combination shows high antitumor activity and a safe toxicity profile in first-line treatment for AOC.

5078

Phase II study of oxaliplatin (OXA) and docetaxel (DTX) in recurrent platinum-sensitive ovarian cancer G. Ferrandina, M. Ludovisi, G. Dagostino, D. Lorusso, A. Naldini, A. C. Testa, I. Paris, V. Gallotta, C. Pozzo, G. Scambia Catholic University of the Sacred Heart, Roma, Italy; Catholic University of the Sacred Heart, Campobasso, Italy

Background: We conducted a phase II study in order to evaluate the efficacy and safety of the combination of OXA and DTX in recurrent platinum-sensitive ovarian cancer patients. Methods: Patients received DTX 75 mg/m2 (60-min i.v.) on day 1, followed by OXA 100 mg/m2 (120-min i.v.) on day 1 every 21 days. Results: Between October 2002 and December 2003, 14 Caucasian patients (median age: 55.5 yrs; range, 39–68) were enrolled. At first diagnosis 1 (7.1%) patient had FIGO stage I ovarian cancer, 12 (85.8%) had FIGO stage III and 1 (7.1%) had FIGO stage IV disease. Sites of relapse were as follows: liver/spleen = 3 cases (21.4%), pelvis = 3 cases (21.4%), lym-

54

Onkologie 2004;27(suppl 1):12–58

phnodes = 4 cases (28.6%), peritoneal = 3 cases (21.4%) and multiple sites = 1 case (7.2%). The median platinum-free-interval was 23.5 months (range 13–58). At recurrence the median CA125 was 324.5 U/ml (range 21–5596 U/ml). Of the 13 patients evaluable, 4 (30.8%) had complete response and 3 (23.1%) had partial response to OXA/DTX with an overall response rate of 53.9%. The median time to response was 10 wks (range 5–32) and the median duration of response was 16 wks (range 6–31). Five (38.5%) patients had stable disease (median duration of stabilization: 21 wks, range 12–40). One (7.7%) patient progressed while on treatment. Therefore, an overall clinical benefit was observed in 92.3% of patients. All patients were evaluable for toxicity. A total of 81 courses were given, with a median of 5.5 cycles per patient (range 1–10). Severe toxicities (Grade 3–4 NCI-CTC) included: neutropenia in 33.3% of cycles; severe anemia and thrombocytopenia were not observed. Grade 3–4 neurotoxicity and alopecia were detected in 4.9% and 24.7% of cycles respectively. Allergic reaction was observed only in one case. Doses were reduced by 20% in 13.5% of cycles due to neurological toxicity. Conclusions: In recurrent platinum-sensitive ovarian cancer patients the OXA/DTX combination is highly active with acceptable toxicity, thus making it an attractive regimen which warrants the prosecution of the study.

5104

Bi-weekly oxaliplatin (oxa) and gemcitabine (gem) in cisplatin pretreated patients with relapsed ovarian cancer (roc): Preliminary data of a phase II trial D. Germano, D. Bilancia, A. Di Nota, A. M. Dello Russo, R. Romano, G. Rosati, F. Giotta, L. Manzione San Carlo Hospital, Potenza, Italy; Oncology Institute, Bari, Italy

Background: Both OXA and GEM have shown single agent activity in pts with ROC. Responce rates to second-line therapies remain low and there is a need to develop more effective regimens. In view of the synergistic effect using the sequence GEM followed by OXA (Faivre S, Cancer Chemother Pharmacol 1999, 44(2): 117–23), we studied these agents in pts with ROC heavely pretreated. The aim of the study was to evaluate the efficacy and toxicity of combination of GEM 1000 mg/m2 D1 IV and OXA 100 mg/m2 in 2h infusion D2, treatment was repeated every 2 weeks until progression of desease or intollerable toxicity Methods: The study was monoinstitutional and started in November 2002.To date 15 patients, median age 66 years (range 52–82) have been treated. Median Karnofski Performance Status was 70% (range 60–90), all had at least 1 prior platinum based chemotherapy and 11 had received a taxane Results: At November 2003 of 15 patients entered onto study, 13 were assesable. Pts received a median of 6 cycles of treatment (range 4–11). There was 1 patient (7.7%) with complete response, 3 patients (23%) achieved a partial response. Using the technique of Rustin et al. (JCO 1996, 14:5; 1545–1551) Ca 125 responses occurred in 3 of the 4 responding patients. As concerns toxicity, a low profile (grade 1 –2, WHO criteria) was observed: nausea 38.5%, trombocytopenia 30.7%, neuropaty 38.5%. Conclusions: The GEMOX combination is extremely well tolerated and even in this small pts group encouraging responses have been documented. Recruitment continues and further data will be presented.

Abstracts

5111

Gemcitabine (GEM) + oxaliplatin (OX) in patients (pts) with stage III/IV ovarian cancer following 3 cycles of carboplatin (CB) + paclitaxel (PAC): Preliminary report of a phase II study R. Arca, J. Ibarra, G. Lerzo, M. Mandachain, E. Mickievicz, J. Perez, E. Richardet, J. Rodger, M. Van Kooten, M. Orlando Hospital Urquiza, Concepcion del Uruguay-Entre Rios, Argentina; COIR, Mendoza, Argentina; LUCEN, Buenos Aires, Argentina; Instituto Alexander Fleming, Buenos Aires, Argentina; Hospital Roffo, Buenos Aires, Argentina; Hospital Penna B. Blanca, Bahia Blanca, Argentina; Hospital Italiano de Cordoba, Cordoba, Argentina; Eli Lilly, Buenos Aires, Argentina

Background: GEM is active in recurrent epithelial ovarian cancer. OX activity in ovarian cancer is similar to cisplatin in first-line therapy and PAC in nonrefractory relapses. This multicenter, prospective phase II study combines GEM/OX as first-line sequential therapy for advanced ovarian cancer in pts after 3 cycles of CB/PAC. Primary endpoint is time to disease progression; secondary endpoints are CA 125 response rate, objective response, response duration, overall survival, correlation between expression of MMR and p53 and efficacy, and toxicity. Methods: Pts with confirmed histologic diagnosis of stage III/IV epithelial ovarian carcinoma, initial surgery including anexohysterectomy + omentectomy, 3 cycles of CB ≥AUC 5 and PAC ≥175 mg/m2, no other prior chemotherapy, and adequate organ and marrow function were enrolled. Three to 4 weeks after CB/PAC, pts received GEM 1000 mg/m2 on days (d) 1,15 (100-minute infusion) and OX 85 mg/m2 on d2,16 (120-minute infusion), q28d. Results: To date, 18 pts (median age 57 [range 42–73]; Zubrod performance status 0/1 [10/8 pts]) have been enrolled. 8 pts had stage IIIB; 7, stage IIIC; 3, stage IV. At baseline, median CA 125 level was 777 ng/ml (range 27–9330). Of 13 evaluable for CA 125 response, 12 had complete response and 1 progressed; 3 were evaluable for objective response (1 complete and 2 partial). Total number GEM/OX cycles was 51 (median 2.83 [range 1–3]). All intended doses given as planned; only 5.8% were delayed due to insufficient hematologic recovery. All were evaluable for toxicity. Grade (G) 3 anemia occurred in 2 cycles (3.9%); G3 neutropenia, in 1 cycle (1.9%); mild thrombocytopenia (G1), in 4 cycles (7.8%). G3 nausea occurred in 1 cycle (1.9%); G3 vomiting, in 2 cycles (3.9%). Reversible G1/2 sensory neuropathy occurred in 9 (17.6%) and 3 cycles (5.9%), respectively. No treatmentrelated deaths occurred. Conclusions: GEM/OX following CB and PAC is feasible, with minimal and manageable toxicity, in pts with ovarian cancer. Updated results of survival and pharmacology will be presented. Supported by Eli Lilly, Debiopharm.

well-tolerated salvage therapies with minimal toxicities are needed. Methods: We designed the R-GEMOX regimen, with rituximab 375 mg/m2 d1, gemcitabine 1000 mg/m2 d2 and oxaliplatin 100 mg/m2 d2 (recycling on d15). Between January 2002 and August 2003, 24 patients with refractory/relapsing B-cell lymphoma not eligible for HDT were enrolled in an open unicenter pilot study whose primary objective was to determine the overall response rate (ORR) after 4 cycles of R-GEMOX. Patients were planned to receive 8 cycles if a good response (at least PR) was observed after 4 cycles. Median age was 64 years (range: 43–77) and histological subtypes were: diffuse large Bcell lymphoma (n=16), follicular (n=6) and mantle cell (n=2). Prior treatment included anthracyclin (n=22), rituximab (n=11) and HDT (n=6). The median number of prior treatments was 2 (range: 1 to 5) and 8 patients had received at least 3 prior regimens. Results: 172 cycles were given. The dose administered was 100% of the intended dose for the three drugs in all patients but 4, for whom the dose of oxaliplatin was reduced due to neurotoxicity (n=3) or preexisting renal insufficiency (n=1). Two patients progressed after 3 cycles. After 4 cycles, observed responses were: 6 CR, 4 CRu, 11 PR and 1 failure resulting in an ORR of 88%. Among the 21 responders after 4 cycles, 17 achieved CR or CRu, 2 remained in PR after completion of 8 cycles. Treatment was stopped after 5 cycles in 2 patients because of physician decision. As of December 2003, 22 patients are alive, 15 in continuous complete remission, 3 in CRu, 2 in PR and 2 in progression. NCIC grade 3–4 neutropenia and thrombocytopenia were reported in 50% and 21% of cycles. Two patients developed a grade 4 infection. There was no renal toxicity. Conclusion: The R-GEMOX regimen shows promising activity with an acceptable toxicity. It is evaluated in an ongoing multicentric phase II study.

6681

Dexamethasone, high-dose cytarabine, and oxaliplatin (DHAOx), and rituximab plus DHAOx (R-DHAOx) for treatment of patients with B-cell non-Hodgkin’s lymphoma: Results from two consecutive phase II studies D. Machover, B. Delmas-Marsalet, Y. Gumus, S. C. Misra, A. Ulusakarya, N. Brahimi, E. Goldschmidt, N. Frenoy, C. Guettier Paul Brousse Hospital, Villejuif, France

Background: High-dose therapy with autologous stem cell support (HDT) is now an established treatment for chemosensitive relapse in aggressive lymphoma. However, not all patients are candidates for HDT because of age, comorbidities or previous HDT. Effective and

Background: To determine, from two consecutive phase II studies, the efficacy of oxaliplatin (L-OHP), cytarabine (ara-C), and dexamethasone (DHAOx), and DHAOx plus rituximab (R-DHAOx), in patients with non-Hodgkin’s lymphoma (NHL). Methods: In Study I, treatment consisted of DHAOx (dexamethasone, 40 mg/day, days 1 to 4; L-OHP, 130 mg/m2, day 1; and ara-C, 2,000 mg/m2, every 12 hrs, day 2). In Study II, treatment consisted of R-DHAOx (DHAOx plus rituximab, on day 1, 375 mg/m2). Courses were repeated every 21 days. Patients had failed to achieve a complete response (CR) with initial chemotherapy, were in relapse, or could not receive standard treatment. Results: Response to therapy: Study I: Eight patients (61.5%) achieved a CR, and 1 (8%) had a PR. None of the complete responders has relapsed; they had a median disease-free survival (DFS) of 44.6 months. Study II: Fifteen patients (68%) achieved a CR, and 3 (13.5%) had a PR. None of the complete responders has relapsed; they had a median DFS of 19.8 months. Responses were obtained in lymphomas of follicular, marginal-zone, mantle-cell, and diffuse large B-cell subtypes, and in patients with or without resistance to prior chemotherapy. Disappearance of molecular markers occurred in all of the complete responders whose tumor cells carried markers. Toxicity: Myelosuppression and dose-related peripheral neuropathy were the most prominent toxic effects. Conclusions: DHAOx and RDHAOx are highly active for salvage treatment of patients with Bcell NHLs, and possess toxicity characteristics which compare favorably to those reported with ara-C plus cisplatin-containing regimens.

Abstracts

Onkologie 2004;27(suppl 1):12–58

Adult Leukemia, Lymphoma and Transplantation 6592

Rituximab, gemcitabine and oxaliplatin (R-GEMOX): A promising regimen for refractory/relapsed B-cell lymphoma T. El Gnaoui, B. Joly, J. Dupuis, K. Belhadj, A. Rahmouni, C. Copie-Bergman, A. Allain, I. Tabah-Fisch, F. Reyes, C. Haioun Hôpital Henri Mondor, Créteil, France; Sanofi-Synthelabo, Paris, France

55

Lung Cancer 7250

Chemotherapy with gemcitabine and oxaliplatin in 13 malignant pleural mesothelioma (MPM): A monocenter experience J.-M. Brechot, J.-F. Morère, G. Des Guetz, K. Chouahnia, J.-L. Breau Hôpital Avicenne, Bobigny, France

Background: A wide range of chemotherapic agents used singly or in combination have been evaluated in MPM. Doublets with gemcitabine and a platinum salt emerge as one of the most promising regimen. The combination of gemcitabine and oxaliplatin (GEMOX) has a very good toxicity profile. This chemotherapy has been administered in clinical practice in our department for 2 years. Methods: We have rewieved the data of consecutive non pretreated MPM patients (pts) treated with GEMOX between October 2001 and October 2003 at our department. Results: 13 pts (9 males, 4 females), mean age 62y (50–74), PS 0–1 12 pts, PS 2 1 pt, with a MPM diagnosed by pleuroscopy (9), thoracotomy (1) or blind pleural biopsy (3), with epithelial (4), mixed (4) and unspecified (5) histological subtype, all stage III or IV, received as outpatient treatment an association of gemcitabine 1000 mg/m2 D1 and oxaliplatin 100 mg/m2 D2 every 2 weeks with a mean of 8 cycles. In 9 evaluable pts, 2 PR, 6 SD and 1PD were observed. Median time to progression (TTP) was 7,5 months. Median survival is 12 months. Seven pts died, all in progressive disease. No grade 3 or 4 hematologic toxicities were observed. In 5 cases, treatment was discontinued because of non hematologic toxicities (allergia grade 3 in 2 pts at the 6th and 7th cycles, peripheral neurotoxicity in 3 pts (at the 6th and 10th cycles). Conclusion: GEMOX is a well tolerated and active chemotherapy regimen in MPM. It could be recommended in clinical practice.

7263

A multicenter phase II trial with gemcitabine plus oxaliplatin as second line treatment in patients with advanced non-small cell lung cancer (NSCLC) C. Kouroussis, K. Syrigos, A. Potamianou, N. Ziras, P. Ziotopoulos, J. Stergiou, P. Ginopoulos, E. Tselepatiotis, V. Biozionelou, V. Georgoulias For the Lung Cancer Working Group, of the Hellenic Oncology Research Group (HORG), Greece

Background: Gemcitabine and Oxaliplatin have shown activity in NSCLC even in pretreated patients (pts). We conducted a phase II study to determine the activity of this combination as second line treatment in NSCLC. Methods: Patients were eligible if they had confirmed NSCLC, measurable or evaluable disease, performance status (ECOG) 0–2 and adequate hepatic, renal and bone marrow function. Gemcitabine dose was 1500 mg/m2 (30 minute infusion) on days 1 and 8, and Oxaliplatin 130 mg/m2 (2–6 hour infusion) on day 8. Cycles were administered every 3 weeks. Results: 24 pts were enrolled, 21 males and 3 females. Median age was 61.5 years (range 42–72); 83% had PS 0 or 1 and 17% PS 2; 12 pts had received 2 lines of treatment. All pts were evaluable for toxicity and 23 for response. Three pts achieved partial response (13%; 95% c.i 0.72%–26.81%) and 6 (26%) stable disease. The median time to progression was 3 months (range 1–12.7) and the median survival 5.6 months (range, 1–14). A total of 84 cycles were administered (median number/patient: 3) and the median dose administered was 85% for Gemcitabine and 81% for Oxaliplatin. No grade IV or febrile neutropenia occurred. Grade 3 neutropenia occurred in 4 (17%) pts, grade 3 anemia in 3 (12.5%) and grade 3 thrombocytopenia in 2 (8%). Non-hematological toxicity was

56

Onkologie 2004;27(suppl 1):12–58

mild and not clinically relevant. The only grade 3 non-hematological toxicity was neurotoxicity in 2 (8%) pts which was reversible with the discontinuation of treatment. Conclusions: The Gemcitabine plus Oxaliplatin is a well tolerated but marginally active regimen in pretreated pts with advanced NSCLC.

7348

Preliminary results of phase II trial with combination of gemcitabine and oxaliplatin in elderly patients with advanced NSCLC R. Buosi, S. Poletti, V. Cantele, S. Piazza, R. Torazzo, E. Negru, A. Vandone, O. Alabiso Ospedale Maggiore della Carità, Novara, Italy

Background: Oxaliplatin (L-OHP) is a novel antineoplastic platinum derivative with a DACH carrier ligand. In preclinical studies it appeared to have clinical activity in many tumor cell lines, including some that are resistant to cisplatin and carboplatin. Based upon in vitro studies demonstrating synergistic anti-tumor activity between LOHP and new citotoxyc drugs, as Gemcitabine, we have conduced a phase II study to evaluate the toxicity and potential synergistic effect by combining the two drugs in elderly patients with locally advanced or metastatic NSCLC. Methods: 14 chemotherapy naive patients with stage III B and IV NSCLC were enrolled in the study. The patients show these characteristics: 4 females, 10 males, median age 73 years, range 65–78, 12 patients with ECOG performance status 0, 2 with ECOG 1. Cancer istotype: 2 adenocarcinoma, 10 squamous cell, 1 large cell, 1 in determinated carcinoma. 3 patients have bone metastasis, 4 have lung metastasis and 3 with brain lesion. 4,9 cycles for patient are administered. Treatment schedule: L-OHP 65 mg/mq and Gemcitabine 1000 mg/mq, both drugs are administered at days 1–8, to be repeated every 3 weeks Results: 13/14 patient evaluable for response rate: 6 RO (43%) in which 1 RC and 5 RP, 1 PD (7%) and 6 SD (43%). Only 1 patient had a worsening of PS, 1 patient in RP after 3 cycles is die for unrelated therapy cardiac failure. We observed mild hematological toxicity: 2 pts show grade (Gr) 2 anemia; 3 pts Gr 2 trombocytopenia, and Gr 3 and Gr 4 trombocytopenia we relevate in only 1 pt respectively. Neither Gr 4 neutropenia nor febrile episodes were detected. Also non hematological toxicity is low: nobody experimented Gr >2 nausea and vomiting. Gr. 2 diarrhea and Gr 2 neurotoxicity are manifested in 2 pts and in 4 pts respectively; 1 pt have Gr 2 transaminases increase. The 2 pts with Hb level ≤10 g/dl are treated with recombinant human erythropoietin 10.000 UI s.c. 3 times/wk until to normal Hb level Conclusions: these encouraging results indicate that the GEMOX schedule could be used in treatment of advanced NSCLC and so we advise to use this combination in elderly patients.

Abstracts

8093

Patient Care

Potential prevention and treatment of oxaliplatin associated peripheral neuropathy

8042

Oxaliplatin (with 5FU and leucovorin) in patients above age 70 («elderly») suffering from metastatic colorectal cancer: An observational study B. Exquis, M. Aapro, S. Kohler Groupe Médical Onex, Onex, Switzerland; IMO Clinique de Genolier, Genolier, Switzerland; Sanofi-Synthélabo Suisse, Meyrin, Switzerland

Background: Data on standard chemotherapy in the real elderly are rare. Patients and Methods: We conducted, during 18 months, an observational study of Oxaliplatin in combination with 5-FU and leucovorin (FUFOL) for treatment of metastatic colorectal cancer, in order to assess safety, response rate and exact regimens used in patients below and above 70 years. Twenty-five centres documented 137 patients (57% male, median age 62 years, colon/rectum/other: 64%/30%/6%, ECOG PS 0/1/2/3: 34%/55%/10%/1%, previous chemotherapy: 69%). Total number of administered cycles was 816 (median 6, range 1–23). Results: Regimen used: 71% received Eloxatin at 85 mg/m2 every 2 weeks, 18% at 130 mg/m2 every 3 weeks, and 11% received other regimens. FUFOL regimens were 37% FOLFOX4, 27% simplified 46h infusion, 21% weekly bolus (Roswell Park), 8% daily bolus (Mayo), 7% others. Response rates: CR 4%, PR 21%, SD 37, PD 27%, unknown 11%. ECOG PS before vs. after therapy: 20% of patients improved, 62% stable, 18% deteriorated. Safety: Grade 3/4 toxicities (per cycle) were nausea (0.1%), vomiting (0.5%), diarrhea (2%), stomatitis (0.2%), thrombocytopenia (1%), leucocytopenia (1%), granulocytopenia (0.6%), lymphocytopenia (7.2%). 75% no hair loss, 18% had mild and 7% had grade 2 hair loss. Grade 3 neuropathy (Levy scale) was reported by 3%. These 4 patients had at least tumor stabilisation or partial response. Influence of age: 33 patients were older than 70. Overall response rate was 24% for patients up to 70 years and 27% for older patients. Grade 3/4 toxicities (per cycle) were (≤70/>70 years): nausea 0.3%/0.0%, vomiting 0.7%/0.0%, diarrhea 1.8%/2.8%, stomatitis 0.4%/0.0%, neuropathy 0.7%/0.0%. Conclusion: Combination therapy with Eloxatin as used here is safe and efficacious in daily practice. Elderly patients benefit equally from therapy without increased toxicity.

K. R. Hoffman M.P.H., Totowa, New Jersey, NJ

Background: The major toxicities of oxaliplatin are two different types of peripheral neuropathies, a reversible one related to the drug’s infusion and an irreversible one based on the cumulative dose received. The incidence of onset and the severity of the neuropathy increase as the number of cycles increase as does the prevalence of the toxicity. Gabapentin (Neurontin) and carbamazepine (Tegretol) are two drugs used in the treatment of autonomic dysfunction and peripheral neuropathic pain. Given the similar syndromes, a trial was first conducted to see if the use of these agents could prevent the progression of this oxaliplatin-associated toxicity. A second trial was then begun to see if these agents could totally prevent oxaliplatin-associated neurotoxicity. Methods: 34 consecutive patients receiving oxaliplatin-5-FU-leucovorin rescue chemotherapy for stage IV colon cancer were available for study. Treatment was determined by random drawing of treatment options. Results: In the initial study of 24 patients, grade 1 toxicity had to be reported before treatment was initiated with either drug, usually after the 2nd cycle. Treatment was begun a week before the 3rd cycle. Gabapentin was begun at 900 mg/day and escalated to 2700 mg/day, as needed. Carbamazepine was begun at 400 mg/day and escalated to 800 mg/day, as needed. A success rate for preventing any grade 3 or higher sensory symptomatology occurred in 75% (9/12) in both groups. No life-threatening nor persistent neuropathy occurred. To judge whether these agents could totally prevent the sensory neuropathy, the next 10 patients have been randomized at the start of their treatment to either gabapentin, carbamazepine or nothing. After 3 cycles the 6 treated patients have grade 1 sensory neuropathy while 2 of the 4 non-prophylaxed patients had grade 2 sensory neuropathy and were started on therapy before cycle 3. No drug toxicity associated with either carbamazepine or gabapentin was observed. Conclusions: While a prospective, doubleblinded randomized trial needs to confirm these findings, gabapentin or carbamazepine can be considered as adjuvant therapy to minimize the sensory neuropathic toxicity of oxaliplatin.

Grade 3/4 toxicity (% per cycle)

≤70 years N=104 >70 years N=33

Abstracts

CR + PR

Nausea

Vomiting

Diarrhea

Stomatitis

Haemoglobin

Platelets

Leucocytes

Granulocytes

Lymphocytes

Neuropathy (Levy scale)

24% 27%

0.3 0.0

0.7 0.0

1.8 2.8

0.4 0.0

0.0 0.0

0.0 1.6

1.6 0.0

0.5 0.0

6.3

0.7

Onkologie 2004;27(suppl 1):12–58

57

Tumor Biology / Immunobiology / Human Genetics 9506

Assessment of multiple markers for association with response rate (RR) and failure-free survival (FFS) in patients with advanced colorectal cancer (CRC) treated with chemotherapy in the MRC CR08 (FOCUS) randomized trial J. W. Adlard, S. D. Richman, P. Royston, J. M. Allan, A. Meade, M. Parmar, P. Selby, P. Quirke, M. T. Seymour, on behalf of the FOCUS Trial Collaborators Cancer Research UK, Leeds, United Kingdom; Medical Research Council, London, United Kingdom; University of Leeds, Leeds, United Kingdom

Background: We have assessed potential prognostic and predictive markers of response to fluorouracil (FU) in advanced CRC patients randomized 4:1:1 to FU alone, FU+oxaliplatin, or FU+irinotecan. Methods: Pathology specimens were retrieved from 846 patients in the FOCUS trial. 85% were from the primary site. Normal and tumor DNA and RNA were extracted and tissue microarrays were made for

immunohistochemistry (IHC). The following factors were assessed for effect on FFS and RR to first-line therapy: age; performance status; primary site; liver metastases; serum ALP; grade; mucoid status; IHC for hMLH1, hMSH2, P53, SMAD4, dUTPase, DPD, thymidine phosphorylase (TP), thymidylate synthase (TS); TS mRNA expression; DNA polymorphisms in TS (TSER, TS1494), hMLH1 (–93 G2A), MTHFR (677 C2T). Results: Overall RR and median FFS to first-line therapy were 42% and 8 months respectively. The table shows markers with significant relationships to RR and FFS. Together markers explained 6% of the variability in outcome, with the IHC/histology markers contributing most. DNA polymorphisms showed no clear association with outcomes. Similar associations of marker status were found for patients treated with FU alone or FU+oxaliplatin/irinotecan. Using logistic regression, a predictive algorithm could separate tumors into groups with RR from 15% to 50% for treatment with FU alone and 25% to 70% for treatment with FU+oxaliplatin/irinotecan. Conclusions: Significant differences in RR have been identified for mucoid status, TS IHC and dUTPase IHC scores allowing development of a predictive algorithm for response to FU-containing chemotherapy. Associations with modest differences in FFS were also identified. Ongoing work is looking at markers specific for oxaliplatin or irinotecan benefit.

Table. Markers associated with adverse RR/FFS. n/s=not significant. *=significant in multivariate analysis. Marker associated with adverse RR/FFS

p-value for RR

Odds ratio (OR) for RR

95% C.I. for OR

Absolute difference in RR (%)

p-value for FFS

Hazard ratio (HR) for FFS

95% C.I. for HR

Difference in median FFS (months)

Mucoid tumor* High TS IHC* High dUTPase (*for FFS only) High TS mRNA* Grade 3* Low DPD* Liver secondaries present* High ALP

0.0001 0.0001 0.052

2.64 1.85 1.39

1.58–4.42 1.38–2.49 1.01–1.91

21.3 15.0 8.0

n/s 0.0001 0.002

n/s 1.37 1.32

n/s 1.17–1.62 1.11–1.56

n/s 1.6 1.1

n/s n/s n/s n/s n/s

n/s n/s n/s n/s n/s

n/s n/s n/s n/s n/s

n/s n/s n/s n/s n/s

0.004 0.013 0.014 0.021 0.046

1.42 1.27 1.24 1.26 1.18

1.12–1.81 1.05–1.53 1.04–1.47 1.03–1.38 1.01–1.38

0.5 1.6 1.5 1.1 1.1

58

Onkologie 2004;27(suppl 1):12–58

Abstracts

Dokumentation und Evaluation der Weiterbildung Autorenindex Dokumentation und Evaluation der Weiterbil Onkologie 2004;27(suppl 1):59–60

Aapro, M. 57 Abad, A. 29 Achille, E. 25 Adam, R. 35, 37, 40 Adlard, J.W. 58 Advani, R.H. 17 Agafitei, R.D. 27 Airoldi, M. 44 Alabiso, O. 56 Albertotti, C. 51 Alberts, S.R. 27 Algeri, R. 42 Allain, A. 55 Allan, J.M. 58 Allegrini, G. 36, 38 Alonso, V. 42 André, T. 16, 19, 23, 25, 25, 38, 45 Andreuccetti, M. 38 Anton, A. 42 Anton, L.M. 48 Antón-Aparicio, L. 31 Aparicio, J. 24 Aravantinos, G. 52 Arca, R. 55 Arcara, C. 12 Arkenau, H.-T. 22 Arland, M. 28 Arnold, D. 50 Artioli, F. 47 Artru, P. 22, 38 Aschele, C. 42 Asensio, D. 24 Asnaghi, D. 43 Athanasiadis, A. 20, 30, 52 Avallone, A. 44 Avenin, D. 23 Azarnia, N. 33 Azoulay, D. 37 Badalamenti, G. 12 Badarinath, S. 20 Bae, S.H. 32 Balart, J. 43 Baltesgard, L. 35 Barbara, C. 36 Baron, M.G. 48 Bartel, C. 24 Barugel, M. 32 Basaran, G. 37 Bastian, G. 13 Battistini, G. 42 Bauer, J. 46 Beale, P. 26 Beasley, S. 36 Behrens, R. 45 Beijnen, J.H. 13 Belda, C. 43 Belhadj, K. 55 Bellmunt, J. 52 Belón, J. 24 Benettaib, B. 54 Bennouna, J. 22 Benson, A.I.B. 14 Berdzik, J. 46 Bernabé, R. 24 Bernuzzi, P. 47 Berrocal, A. 52 Bertè, R. 47 Bidin, L. 47 Biffi, A. 31 Bilancia, D. 54 Binda, G.A. 42 Biozionelou, V. 56 Bismuth, H. 37 Bjarnason, G. 19 Blanco, G. 29 Blumgart, L. 34 Boige, V. 44, 47, 50 Bolaños, M. 24, 29, 48 Bonetti, A. 12, 16 Boni, C. 12, 16 Bonnay, M. 44 Bonyhay, L. 50 Bordonaro, R. 29 Borgonovo, K. 43 Boron, M. 33 Botto, H.G. 30 Botturi, M. 43 Bouchahda, M. 35, 40

Bougnoux, P. 53 Brahimi, N. 55 Breau, J.-L. 13, 56 Brechot, J.-M. 56 Brell, J.M. 14 Brennscheidt, U. 26 Brezault Bonnet, C. 35, 40 Brienza, S. 12 Brunetti, I. 36, 38 Bu, Z.D. 47 Budillon, A. 44 Buechner-Steudel, P. 45 Bugat, R. 22, 26 Bui, B. 51 Buosi, R. 56 Burkart, C. 50 Busquier, I. 24 Butts, C. 23 Buzaid, A.C. 51 Buzzi, F. 32 Campana, F. 33 Campos, J.M. 24 Canon, J.-L. 40 Cantele, V. 56 Cantore, M. 45 Carles, J. 52 Carola, E. 25 Caroti, C. 42 Carraro, S. 32 Carreca, I. 44 Casado, E. 43 Cascinu, S. 48 Caspar, C.B. 30 Cassidy, J. 23 Castaing, D. 37, 40 Castelo, B. 43 Catalano, G. 48 Catalano, P.J. 14 Catalano, V. 48 Catot, S. 29 Cattel, L. 44 Cavanna, L. 47 Ceccarelli, C. 40 Cerri, E. 36, 38 Cervantes, A. 17, 19, 49 Chacón, J.I. 48 Chacon, M. 32 Chacon, R. 32 Chang, I.Y. 28 Chatta, G. 16 Chatzichristou, A. 30 Chen, H.X. 14, 33 Chen, J.-S. 49 Chen, T.T. 20 Cheng, C. 28 Cheong, K.A. 53 Chevreau, C. 51 Chirivella, I. 25, 25 Cho, C.D. 17 Chopitea, A. 38 Chouahnia, K. 13, 56 Christakis, C. 30 Chrystal, K. 53 Cicero, G. 12 Clark, J.W. 35 Clark, K. 46 Clarke, S. 26 Clemens, M. 46 Climent, M.A. 52 Clingan, P. 16 Coche, J.-C. 40 Cognetti, F. 31 Cohn, A.L. 33 Cola, B. 40 Colin, P. 25 Comella, P. 32, 44 Conill, C. 49 Conroy, T. 23, 47 Constantin, C. 34 Constela, M. 31 Copie-Bergman, C. 55 Cordio, S. 29 Cornelis, M. 39 Coronado, C. 32 Corretgé, S. 17 Cortés-Funes, H.G. 26 Costa, F.P. 51 Costa, O.F. 51

© 2004 S. Karger GmbH, Freiburg Fax +49 761 4 52 07 14 E-mail [email protected] www.karger.com

Accessible online at: www.karger.com/onk

Costa, P.A. 51 Coster, B. 40 Cottu, P. 53 Coudert, B. 19 Coviello, K.M. 14 Creaven, P.J. 16 Cupini, S. 36, 38 Cvitkovic, E. 33, 54 Cvitkovic, F. 33 D’Amico, M. 42 Dagostino, G. 54 Dane, F. 37 D’Angelica, M. 34 De Baere, T. 50 De Braud, F. 23, 45 De Castro, J. 31, 43 De Gramont, A. 16, 17, 19, 23, 25, 38, 45 De la Cámara, J. 38 De Vita, F. 32 Decian, F. 42 Deguetz, G. 13 Del Mastro, L. 12 Delaloge, S. 12 Delgado, M. 50 Dello Russo, A.M. 54 Delmas-Marsalet, B. 55 Delord, J.P. 22, 26, 51 Delpero, J.R. 39 Delrio, P. 44 Delva, R. 53 Delvart, V. 37 Dematteo, R. 34 Demols, A. 46 Den Boer, S.S. 13 Deplanque, G. 53, 54 Des Guetz, G. 56 Di Costanzo, F. 31 Di Nota, A. 54 Di Seri, M. 48 Díaz-Rubio, E. 23 Dietrich, G. 34 Dimopoulos, A. 52 Dixon, J. 43 Dogliotti, L. 19 Doroshow, J. 51 Dorta, F. 31 Douillard, J.-Y. 17, 22 Dowlati, A. 14 Duck, L.R. 40 Ducreux, M. 44, 45, 47, 50 Dupuis, J. 55 Eckardt, J.R. 14 Economopoulos, T. 52 Efstathiou, E. 52 Egorin, M.J. 14, 16 Ekenel, M. 37 El Gnaoui, T. 55 Elfring, G.L. 17 Elias, D. 44 Emanuel, D. 21 Enzinger, P.C. 35 Escribano, A. 43 Escudero, P. 29, 42, 48 Esteban, B. 26 Esteban, E. 52 Etienne, P.-L. 19 Exquis, B. 57 Extra, J.M. 53 Fabbro, M. 54 Fabregat, X. 52 Falcone, A. 36, 38 Falk, S. 41 Farmakis, D. 52 Farris, A. 32 Fasolo, A. 43 Feliú, J. 29, 43, 48 Fernandez-Martos, C. 24 Ferrandina, G. 54 Ferrarato, E. 44 Ferraresi, V. 31 Fettner, S. 26 Feyereislova, A. 26 Figer, A. 16, 19, 23, 25, 25 Filippelli, G. 32 Findlay, B. 19 Findlay, B.P. 27 Fisher, G.A. 17

Fizazi, K. 51 Flechon, A. 51 Fleig, W.E. 45 Flesch, M. 19, 25, 25 Focan, C. 19 Fong, Y. 34 Fonseca, E. 31, 48 Font, A. 52 Fontana, E. 36, 38 Forstbauer, H. 34 François, E. 47 Frankel, P. 51 Fraschini, F. 31 Freidank, A. 28 Freier, W. 22 Frenoy, N. 55 Fuchs, C.S. 19, 27, 35 Fulfaro, F. 12 Gabriele, A. 31 Gaglia, A. 52 Galán, A. 29 Galani, E. 53 Galassi, B. 43 Gallagher, C. 53 Gallego, R. 49 Gallo, L. 42 Gallotta, V. 54 Gallurt, P. 24 Gamelin, E. 25 Gamucci, T. 31 Gandara, D. 51 Ganem, G. 23 García-Foncillas, J. 38 García-Girón, C. 26, 31 García-Olmo, D. 43 Garcia-Ribas, I. 52 García-Rico, E. 26 Garriga, J. 43 Garufi, C. 19 Gascón, P. 49 Gay, F. 46 Gebauer, B. 36 Gebbia, N. 12 Gelibter, A. 31 Gentile, A.L. 40 Geoffrois, L. 51 Georgoulias, V. 20, 52, 56 Germano, D. 54 Gholam, D. 35, 40 Giacchetti, S. 19, 35, 40 Giaccone, G. 27 Giachetti, S. 37 Giannakakis, T. 20 Giantonio, B.J. 14 Gíl-Bazo, I. 38 Gil-Delgado, M. 13 Ginopoulos, P. 56 Giornelli, G. 32 Giotta, F. 54 Giovannini, M. 47 Giulianotti, P.C. 42 Glasgow, S.C. 13 Glimelius, B. 35 Glynne-Jones, R. 41, 43 Gold, P.J. 33 Goldberg, R.M. 19, 27 Goldschmidt, E. 55 Golf, A. 46 Gollasch, H. 36 Gómez-Reina, M.J. 24 González, D. 43 González, E. 24 González-Barón, M. 29, 31, 43 Gonzalez-Larriba, J.L. 52 Gordon, M. 18 Goupil, A. 54 Gövercin, M. 49 Graeven, U. 22 Graham, C.D. 20 Grandez, R. 42 Grandinetti, C. 33 Grávalos, C. 26 Graziano, F. 48 Greco, A.O. 14 Greco, W. 46 Green, E. 19 Gregor, M. 50 Griffing, S. 18

Grochow, L. 33 Groshen, S. 27 Grothey, A. 18, 22, 34 Gruenberger, T. 39 Gruia, G. 17 Guallar, J. 43 Guercovich, A. 32 Guettier, C. 55 Guimaraes, R. 37 Gumus, Y. 55 Guo, Y. 48 Haase, C.G. 15 Habib, F. 12 Haioun, C. 55 Haller, D. 39 Halsey, J. 17 Hammel, P. 45 Hansen, V.L. 20 Harper, P. 53 Harris, R. 34 Harrison, M. 43 Hart, L. 21 Hartmann, J.-T. 50 Hata, T. 15 Headlee, C. 36 Hebbar, M. 23, 25, 38 Hedrick, E.E. 18 Heinemann, V. 46 Helm, A. 49 Henderson, C.A. 20 Hendlisz, A. 46 Hickish, T. 16 Hilbig, A. 49 Hilger, R.A. 15 Hinke, A. 46 Hirawat, S. 17 Ho, P. 13 Hochhaus, A. 50 Hochster, H. 21 Hoeffkes, H.-G. 28 Hoffman, K. 57 Hofheinz, R. 50 Höhler, T. 46 Hollis, D. 41 Honhon, B. 40 Hoppel, C.L. 14 Hospers, G. 21 Huang, J.-S. 49 Hukunaga, H. 15 Humblet, Y. 17, 40 Hurwitz, H. 18 Husseini, F. 22 Hyun, M. 32 Iacobelli, S. 19 Iacopino, B. 40 Ibarra, J. 55 Ibeas, R. 28 Ikeda, M. 15 Imadalou, K. 22 Ingalls, S.T. 14 Iqbal, S. 27 Ivy, S.P. 14 Iyikesici, M.S. 37 Jackson, D. 36 Jambalos, C.N. 17 Jarnagin, W. 34 Jasmin, C. 35, 40 Jennis, A. 20 Ji, J.-F. 47 Jin, W. 48 Jirau-Lucca, G. 21 Johnson-Buarque, E.J. 50 Joly, B. 55 Jordan, K. 34 Junker, U. 24 Kahan, J. 39 Kakolyris, S. 20 Kalbakis, K. 20 Kalykaki, A. 52 Kamentsidis, P. 30 Kaplan, R.S. 33 Kastritis, E. 52 Kellner, O. 34 Kemeny, N. 34 Kerbrat, P. 51 Kerger, J. 40 Khayat, D. 13 Kimminau, C. 14

Kleber, G. 45 Klump, B. 50 Koch, K. 13 Kohler, S. 57 Kornek, G. 39 Koumarianou, A. 52 Koumpou, M. 52 Kouroussis, C. 56 Kretzschmar, A. 36 Krishnamurthi, S.S. 14 Kroening, H. 34 Kubicka, S. 22 Kulke, M.H. 35 Kunslinger, F. 37, 40 Kuo, T. 17 Kupsch, P. 15 Kuss, O. 45 Labianca, R. 45 Lakhai, W.S. 13 Lambea, J. 42 Landi, B. 25 Langer, C. 20 Lasser, P. 44 Lastoria, S. 44 Laurent, D. 24 Laurent, S. 25 Lawrence, D. 46 Lazzaro, A. 47 Lebwohl, D. 24 Lee, H.I. 32 Lee, J.L. 32 Lee, K.H. 32 Leichman, C.G. 46 Leichman, L.P. 46 Lenz, H. 51 Lenz, H.J. 18, 27, 33, León, A.I. 26 Leonard, G.D. 34 Leong, L. 51 Lerzo, G. 55 Lévi, F.A. 19, 35, 37, 40 Levy, E. 54 Li, X. 14 Liau, C.-T. 49 Lim, D. 51 Lin, C.-P. 28 Lin, Y.-C. 49 Linehan, D.C. 13 Llaurado, J. 43 Lledo, G. 19, 25, 25, 45 Llorca, C. 24 Lofts, F. 53 Löhr, J.M. 46 Lomas, M. 31 López-Gómez, L. 29 Lordick, F. 46 Lorusso, D. 54 Losa, F. 29 Lotz, J.P. 51, 53 Lou, F. 48 Lou, H.-Z. 48 Louvet, C. 38, 45 Ludovisi, M. 54 Mabro, M. 23 Machiels, J.-P. 40 Machover, D. 25, 55 Maestu, I. 24 Magherini, E. 39, 44 Mahyaoui, S. 39 Maindrault-Goebel, F. 23, 38 Maiorino, L. 32 Makowski, M. 50 Makrantonakis, N. 30 Makris, A. 43 Malka, D. 44 Mallik, N. 18 Maluf, F. 51 Mancarella, S. 32 Mandachain, M. 55 Mansourbakht, T. 50 Mantovani, L. 34 Manzano, H. 49 Manzano, J.L. 29 Manzione, L. 29, 54 Marcucci, L. 36, 38 Marcuello, E. 43 Marechal, R. 46 Marfà, X. 49, 52 Marijken, M. 46 Marone, P. 44 Marques, R. 51 Martin, A. 43 Martín-Algarra, S. 38 Martinelli, G.N. 40 Martínez, A. 43 Martinez, F. 28 Martinez-Balibrea, E. 29

60

Martinez-Cardus, A. 29 Martins, S. 51 Martoni, A. 40 Masi, G. 36, 38 Masri-Zada, R. 45 Masseroni, S. 31 Massidda, B. 32 Maughan, T. 41 Maurel, J. 26, 49 Mavroudis, D. 52 Mayer, I. 51 Mayer, R.J. 33, 41 Mazzocchi, M. 47 McCain, D.A. 40 McDonald, A. 41 McLeod, H.L. 13 Meade, A. 58 Mel Lorenzo, J.R. 25, 48 Meropol, N.J. 14 Mery-Mignard, D. 39 Metges, J.P. 47 Meyerhardt, J.A. 19, 35 Micari, C. 44 Michel, F. 23 Michelini, A. 35 Mickievicz, E. 55 Miedema, B.E. 41 Milaki, G. 52 Miller, L.L. 17 Mineur, L. 19, 23 Minni, F. 40 Mira, M. 42 Mirtsching, B.C. 33, 36 Misra, S.C. 55 Misset, J.L. 33 Mitchell, E.P. 20 Moehler, M. 45 Monden, M. 15 Montalar, J. 31 Monzo, M. 28 Mooney, M. 33 Moreno, I. 28 Moreno, J. 28 Moreno-Nogueira, J.A. 24 Morère, J.-F. 56 Morgan, B. 24 Morgan, R. 51 Mortensen, J.P. 35 Morton, R.F. 27 Mosby, K. 33 Moshidis, A. 30 Motwani, M.V. 15 Moukhtar, R. 23 Mueser, M. 17 Murad, M. 37 Murias, A. 26 Myint, S. 43 Naldini, A. 54 Nardoni, C. 31 Natale, D. 32 Natsiopoulos, I. 30 Nava, E. 46 Nava, H. 46 Navarro, A. 28 Navarro, M. 16 Needle, M.N. 33 Negru, E. 56 Nehls, O. 50 Neri, S. 40 Niedzwiecki, D. 41 Nikolaou, M. 52 Noordhuis, P. 27 Nordlinger, B. 39 Novotny, W. 18 Nuñez, L. 29 O’Connor, J. 32 O’Dwyer, P.J. 14 Oettle, H. 49, 50 Ogreid, D. 35 Ohue, M. 15 Onida, F. 43 Oprea, C. 44 Orlando, M. 55 Ourives, N. 37 Paillot, B. 22 Palladino, M. 47 Pan, H.-M. 48 Pandite, L. 13 Parc, R. 38 Pardo, F. 38 Paris, I. 54 Park, D. 18 Park, K.U. 32 Parmar, M. 58 Pascal, G. 37 Passarge, K. 15 Passera, R. 44

Onkologie 2004;27(suppl 1):59–60

Paty, P. 34 Paule, B. 13, 37 Paye, F. 38 Payrard, S. 39 Pecci, A.P. 42 Pectasides, D. 52 Pectasides, M. 52 Pederiva, S. 30 Peeters, M. 46 Pelzer, U. 49 Pendyala, L. 46 Pera, M. 49 Perez, J. 55 Perez, N. 25 Perillo, F. 42 Pernas, J. 43 Perrier, H. 22 Peters, G.J. 27 Petit, T. 53 Petrillo, A. 44 Pétriz, L. 43 Pfeiffer, P. 35 Phan, S. 16 Piazza, S. 56 Pinedo, H.M. 27 Pink, D. 36 Pinto, C. 40 Pinuaga, M. 28 Pippas, A.W. 33 Pirovano, M. 31 Pissanidis, N. 30 Pitot, H.R. 27 Pizza, C. 29 Pocard, M. 44 Poletti, S. 56 Pollak, M. 19 Pollera, C.F. 31 Polus, M. 46 Polyzos, A. 20, 52 Porschen, R. 22 Portier, G. 39 Poston, G. 39 Potamianou, A. 56 Pou, E. 28 Poynard, T. 50 Pozzo, C. 54 Press, O. 18 Priou, F. 22 Proulx, G. 46 Quenet, F. 39 Quinaux, E. 19 Quinn, S. 25 Quintè, A. 31 Quirke, P. 58 Racanelli, A. 15 Rahmouni, A. 55 Rakhit, A. 26 Ramanathan, R.K. 19, 21, 27 Ramnath, N. 16 Ranze, O. 28 Rau, K.-M. 49 Ravo, E. 44 Raymond, E. 33 Redondo, A. 43 Reggiardo, G. 29 Reichardt, P. 22 Reina, J.J. 24 Remick, S.C. 14 Reyes, F. 55 Ricci, P. 42 Ricci, S. 36, 38 Richardet, E. 55 Richel, D.J. 13 Richly, H. 15 Richman, S.D. 58 Riess, H. 49 Rinaldi, G. 12 Rivera, F. 25 Rivoire, M. 39 Rixe, O. 54 Rizzi, A. 47 Robert, R.V. 17 Roca, E. 32 Rodger, J. 55 Rodriguez, J. 38 Roll, L. 49 Romanelli, A. 31 Romano, R. 54 Romero, R. 24 Rosati, G. 29, 54 Rotellar, F. 38 Rovej, R. 43 Royston, P. 58 Ruiz de Lobera, A. 42 Russ, G. 19 Russo, A. 12 Ryan, D.P. 35, 41

Ryoo, H.-M. 32 Saenz, J.A.G. 48 Saintini, P. 13 Salazar, R. 49 Saleh, M. 30 Saltz, L.B. 17 Salud, A. 26, 29, 42 Samuel, L. 43 Sanchez, B. 43 Sánchez-Rovira, P. 24 Sancho, F. 43 Santini, D. 48 Santoro, A. 12 Sargent, D.J. 19, 27 Saridaki, Z. 52 Sarkar, S. 18 Sastre, J. 17 Sbrana, F. 42 Sburlati, P. 43 Scaglione, F. 31 Scalabrini-Neto, A.O. 37 Scalliet, P. 40 Scambia, G. 54 Scanni, A. 43 Schaapveld, M. 21 Scheithauer, W. 39 Schellens, J. 13 Schemerling, R. 51 Scheulen, M.E. 15 Schiaroli, G. 42 Schlichting, C. 34 Schmalenberg, H. 45 Schmidt, A.M. 14 Schmoll, H.-J. 22, 34 Schneider, S. 27 Schöffski, P. 23 Schuell, B. 39 Schwartz, B. 15 Schwartz, E.J. 17 Schwartz, G.K. 15 Schwartz, L. 34 Sebag-Montefiore, D. 41, 43 Sedeki, I. 47 Seeber, S. 15 Seipelt, G. 46 Seitz, J. 47 Sekimoto, M. 15 Selby, P. 58 Sempoux, C. 40 Sereno, M. 43 Seshimo, I. 15 Sestini, S. 31 Seufferlein, T. 22 Sevilla, I. 26 Seymour, M.T. 58 Shibata, S. 51 Shpilsky, A. 21 Signorelli, C. 31 Siragusa, A. 42 Sjak-Shie, N.N. 14 Smaaland, R. 19 Smorenburg, C.H. 27 Sole Monne, J. 28 Sørbye, H. 35 Souglakos, J. 20 Soulié, P. 17, 25 Spalletta, S. 48 Spano, J.-P. 13 Spinnato, F. 12 Starkhammar, H. 35 Stauch, M. 46 Steer, C.B. 53 Stella, P.J. 33 Stergiou, J. 56 Steward, W.P. 24 Stieler, J.M. 49 Strickland, A. 53 Stroszczynski, C. 36 Strumberg, D. 15 Sturm, I. 36 Sverdlin, R. 35 Syrigos, K. 56 Tabah-Fisch, I. 55 Tabernero, J. 16, 17, 23 Tabiadon, D. 31 Tafuto, S. 32 Taïeb, J. 38 Taieb, J. 50 Takayama, O. 15 Tanaka, K. 40 Taron, M. 29 Tatangelo, F. 44 Tecimer, C. 37 Teel, C. 36 Tejpar, S. 25 Tepper, J.E. 41 Testa, A.C. 54

Autorenindex

Testa, E. 48 Theodore, C. 51 Thomas, A. 24, 35 Thomas, H. 53 Thuss-Patience, P. 36 Tomirotti, M. 43 Tonini, G. 48 Topham, C. 16 Torazzo, R. 56 Tournigand, C. 25, 38 Trarbach, T. 24 Tres, A. 42 Tsavdaridis, D. 30 Tselepatiotis, E. 20, 56 Tubiana-Hulin, M. 12 Tubiana-Mathieu, N. 19 Tucci, A. 29 Tucci, E. 42 Turhal, N.S. 37 Tveit, K.M. 35 Twelves, C. 23 Ugolini, G. 40 Ulusakarya, A. 25, 55 Urien, S. 13 Valencia, J. 42 Valerio, M.R. 12 Valsecchi, R. 31 Van Belle, J.S. 53 Van Cutsem, E. 17, 23, 25, 26, 33 Van Groeningen, C.J. 27 Van Kooten, M. 55 Van Laethem, J.-L. 17, 46 Van Riel, J.M. 27 Vandone, A. 56 Vanhoefer, U. 24 Vardakis, N. 20 Vauthier, J.M. 25 Vena, D. 33 Verslype, C. 26 Versola, M. 13 Vicent, J. 31 Vicent, J.M. 24, 29, 48 Viens, P. 53 Vila, J.M. 28 Vincenzi, B. 48 Vincitore, M. 35 Vitali, M.L. 42 Viudez, A. 38 Vizzini, L. 12 Wadler, S. 41 Wagman, L. 51 Wagner, D. 45 Wakelee, H.A. 17 Wang, H.-M. 49 Wang, W.-Y. 28 Wang, Y. 47 Wardley, A. 12 Wasserman, E. 32 Watson, M.A. 13 Wein, A. 45 Welles, L. 21 Westermann, A.M. 13 White, L.A. 14 Wiedenmann, B. 24 Wiesemann, K. 15 Williamson, S.K. 19, 27 Wong, L. 21 Wu, Q. 47 Wu, X.J. 47 Xhu, A. 35 Yamamoto, H. 15 Yang, D. 18, 27 Yang, T.-S. 49 Yasui, M. 15 Yataghene, Y. 47, 51 Ychou, M. 39, 47 Yen, Y. 51 Yovine, A. 53 Yu, J. 13 Yu, Z. 47 Yubero, A. 29 Yumuk, P.F. 37 Yun, J. 18 Zacharchuk, C. 25 Zambelli, A. 12 Zaniboni, A. 45, 47 Zanon, C. 44 Zeuli, M. 31 Zhang, W. 18 Zheng, Y. 48 Zhong, X.N. 47 Zhu, N. 48 Ziotopoulos, P. 56 Ziras, N. 20, 56 Zwiebel, J. 33