Second-line chemotherapy (CT) with or without bevacizumab (BV) in metastatic colorectal cancer (mCRC) patients (pts) who progressed to a first-line treatment containing BV: Updated results of the phase III “BEBYP” trial by the Gruppo Oncologico Nord Ovest (GONO).

Abstract

3615 Background: Retrospective data suggested that the continuation of BV with second-line CT beyond progression (PD) in pts who received BV in first-line can improve the outcome. Recently, results of the AIO/AMG ML18147 study demonstrated an improved overall survival (OS) by continuing BV beyond PD. Methods: This phase III study randomized pts with measurable mCRC treated in first-line with BV plus fluoropyrimidine, FOLFIRI, FOLFOX or FOLFOXIRI, to receive in second-line mFOLFOX6 or FOLFIRI (depending on first-line CT) with or without BV. The primary end-point was progression free survival (PFS).To detect a HR for PFS of 0.70 with an α and β error of 0.05 and 0.20 respectively, the study required 249 events. Assuming an accrual time of 24 months (mos) and a follow up of 12 mos we planned to randomize 262 pts. Results: Considering the results of the AIO/AMG ML18147 trial, the study accrual was stopped prematurely. A total of 185 pts were randomized and 184 pts were included in the ITT analysis (1 pt randomized in error). Pts characteristics were (arm A/arm B): number 92/92, gender M75%-F25%/M57%-F43%, median age 66 (38-75)/62 (38-75) years, PS=0 82%/82%, multiple site of disease 76%/77%. At a median follow up of 18 mos the study met its primary endpoint by improving PFS in the BV arm. We updated results and at a median follow up of 22 mos the improvement in PFS for the experimental arm was confirmed with a median PFS of 5.2 mos for arm A and 6.7 mos for arm B (HR=0.66; 95% CI 0.49–0.90; unstratified p=0.0072). Subgroup analyses showed a consistent benefit in all subgroups including gender (F: HR=0.63; M: HR=0.72) and first-line PFS (≤10 mos: HR=0.57; >10 mos: HR=0.71). Response rates (RECIST) were 18% and 21% (p=0.71). Toxicity profile was consistent with previously reported data. The OS data are still immature, with 56 events in arm A and 54 in arm B and the median OS is 16.0 mos and 16.5 mos respectively (HR=0.83; 95% CI 0.57-1.22; unstratified p=0.34). Conclusions: This study demonstrates an improvement in PFS by continuing BV in second-line in pts who had received CT+BV in first-line. Clinical trial information: NCT00720512.