Individual patient data meta-analysis of adjuvant gemcitabine-based chemotherapy for biliary tract cancer: combined analysis of the BCAT and PRODIGE-12 studies

Although gemcitabine-based chemotherapy is the standard of care for advanced biliary tract cancers (BTCs), adjuvant phase III studies (BCAT in Japan, PRODIGE 12 in France) failed to show benefit, possibly owing to fewer patients (n = 225 and n = 194) compared with the adjuvant capecitabine BILCAP trial (n = 447). We performed a combined analysis of both gemcitabine-based chemotherapy adjuvant studies.We performed individual patient data meta-analysis of all patients included in BCAT and PRODIGE 12. BCAT study randomised patients with extrahepatic cholangiocarcinoma to single-agent gemcitabine or observation. PRODIGE 12 randomised patients with all BTC subtypes to gemcitabine-oxaliplatin combination or observation. Combined analysis was performed using Kaplan-Meier curves and a Cox regression model stratified on the trial.Two hundred and twelve versus 207 patients were randomised in the gemcitabine-based chemotherapy versus observation arms. Baseline characteristics were balanced between arms. The median follow-up was 5.5 years. After 258 relapse-free survival (RFS) events, there was no difference in RFS (log-rank p = 0.45; hazard ratio [HR] = 0.91 [95% confidence interval [CI] 0.71-1.16]; p = 0.46). RFS rates at five years were 40.8% (95%CI: 33.9%-47.5%) for gemcitabine-based chemotherapy versus 36.6% (95%CI: 29.8%-43.4%) for observation. After 201 deaths, there was no difference in overall survival (OS) (log-rank p = 0.83; HR = 1.03 [95%CI: 0.78-1.35]; p = 0.85). OS rates at five years were 50.5% (95%CI: 43.1%-57.4%) for gemcitabine-based chemotherapy versus 49.3% (95%CI: 41.6%-56.5%) for observation.With 419 patients included, this analysis did not show significant improvement in RFS and no trend in improvement in OS. Gemcitabine-based chemotherapy should not be used as an adjuvant treatment for BTC.