Background
Phase III open-label RCT, N=196 patients with biliary tract cancer (intrahepatic cholangiocarcinoma, extrahepatic cholangiocarcinoma, gallbladder cancer) following curative-intent R0 or R1 resection. Stratified by resection status (R0 vs R1) and tumor site. Conducted across 25 French centers. Enrolled 2009–2014.
Interventions and follow up
Arm A: Gemcitabine 1000mg/m² + oxaliplatin 100mg/m² (GEMOX) q2wk ×12 cycles (6 months)
Arm B: Observation
Primary endpoint: Relapse-free survival (RFS)
Median follow up: 46.5mo
Arm B: Observation
Primary endpoint: Relapse-free survival (RFS)
Median follow up: 46.5mo
Results
RFS: 30.4 vs 18.5mo, HR 0.88 (95% CI 0.62–1.25), P=.48 — not significant
OS: HR 0.99 (95% CI 0.66–1.48), P=.99
OS: HR 0.99 (95% CI 0.66–1.48), P=.99
Adverse events
Neurologic/constitutional: Grade ≥3 toxicities in 47% of Arm A, including peripheral neuropathy 8% and fatigue 6%.
Hematologic/exposure: Grade ≥3 neutropenia 5%; treatment completion rate ~57% of planned 12 cycles.
Hematologic/exposure: Grade ≥3 neutropenia 5%; treatment completion rate ~57% of planned 12 cycles.
Conclusions
Adjuvant GEMOX chemotherapy did not improve RFS or OS compared with observation in resected biliary tract cancer (HR 0.88, P=.48). The trial was underpowered and GEMOX did not prove effective as an adjuvant regimen in this disease.
Key Limitations
Underpowered trial (N=196) with a pre-specified HR assumption of 0.65 that was likely optimistic; the actual HR of 0.88 would have required several hundred more patients to detect. Mixed population including R0 and R1 resections across biologically heterogeneous BTC subsites (gallbladder, intra/extrahepatic CCA). Low cycle completion rate (~57%) may have diluted treatment effect. GEMOX is not a biliary-specific standard — gemcitabine+cisplatin became the dominant doublet after ABC-02 in the metastatic setting, and subsequent adjuvant trials (BILCAP, ASCOT) used fluoropyrimidines. Enrolled before routine molecular profiling (FGFR2, IDH1, HER2).
Clinical Context
A negative result in adjuvant biliary tract cancer, contrasting with BILCAP (capecitabine; positive per-protocol OS) and ASCOT/JCOG1202 (S-1; positive OS). The negative result, combined with positive fluoropyrimidine-based adjuvant trials, shifted practice away from GEMOX in the adjuvant setting. PRODIGE 12 has largely been superseded as a practice reference by BILCAP and ASCOT. GEMOX is not FDA-approved for adjuvant BTC.
References