You are here

Gemox versus surveillance following surgery of localized biliary tract cancer: Results of the PRODIGE 12-ACCORD 18 (UNICANCER GI) phase III trial

Julien Edeline, Franck Bonnetain, Jean Marc Phelip, Jérôme Watelet, Pascal Hammel, Jean-Paul Joly et al.

J Clin Oncol 35, 2017 (suppl 4S; abstract 225)

Editorial comment from Prof. Florian Lordick: 
Survival outcomes for patients with resected biliary tract cancers are dismal. This French multicenter study is the first adequately powered prospectiverandomized controlled trial assessing the efficacy of adjuvant chemotherapy in resected biliary tractcancers. Unfortunately, the primary endpoint of this study, an improvement in relapse-free survival, wasnot met. Longer follow-up and survival results should be awaited. Moreover, other studies investgatingother, maybe more adequate treatment regimens than gemcitabine-oxaliplatin, are investigated inother national networks.



No standard post-surgery adjuvant treatment is currently recommended inlocalized biliary tract cancer (BTC). Gemcitabine combined with platinum is the standardchemotherapy for advanced BTC. The aim of this phase III randomized trial was to assesswhether GEMOX would increase relapse-free survival (RFS) while maintaining health-relatedquality of life (HrQoL).


We performed a multicenter randomized phase III trial. Patients were randomized,within 3 months of R0 or R1 resection of a localized BTC (intra-hepatic, perihilar, extra-hepaticcholangiocarcinoma or gallbladder cancer), to receive either GEMOX 85 for 12 cycles(Experimental Arm A) or surveillance (Standard Arm B). Co-primary objectives were RFS andHrQoL. 190 patients and 126 RFS events were required to show an increase of median RFS from18 to 30 months.


Between July 2009 and February 2014, 196 patients were included in 33 Frenchcenters. Baseline characteristics were balanced, with similar primary sites, R0 resection rateswere 86.2% (Arm A) vs 87.9% (Arm B), lymph node invasion present in 37.2% vs 36.4%. InArm A, a median of 12 cycles was delivered (mean: 9.3, range: 0-12). Maximal grade of adverseevents were grade 3 in 57.5% vs 22.2%, and grade 4 in 17.0% vs 9.1%. During treatment one
patient died in each arm. The main grade ≥ 3 adverse events in the year following inclusion were peripheral neuropathy (50.0% vs 1.1%), and neutropenia (22.3% vs 0%). Median follow-up was44.3 months, with 54 and 64 RFS events in arms A vs B. There was no significant difference inRFS between the arms (log-rank p = 0.31), with a hazard ratio of 0.83 [95% CI: 0.58-1.19], p =0.31 (futility boundaries were crossed). Median RFS was 30.4 [95% CI: 15.4-45.8] vs 22.0months [95%CI: 13.6-38.3] in arms A & B respectively, and 4-years RFS was 39.3% [95%CI:28.4%-50.0%] vs 33.2% [95%CI: 23.1-43.7%]. Global Health HrQoL scores were not differentat 12 months (70.8 vs 83.3, p = 0.18) and at 24 months (75.0 vs 83.3, p = 0.50).


Adjuvant chemotherapy in BTC with GEMOX was feasible and associated withexpected toxicities and no deterioration of HrQoL. There was no significant difference in RFSbetween GEMOX and surveillance. Clinical trial information: NCT01313377

All content © 2017 American Society of Clinical Oncology. All rights reserved.

Search this site

Search form

ECCO2017 symposium webcast: Treatment evolution in advanced GI malignancies

Welcome and introduction - Florian Lordick


Subscribe to our E-Alert to keep up to date with the new items in the Resource Centre

Featured Content

New Book Content

This online Resource Centre has been sponsored by Lilly Oncology

Note that Lilly Oncology has no editorial control over the content of this Resource Centre. The Resource Centre and all content therein has been subject to an independent editorial review.