Real-world comparative outcomes of FOLFOX vs FOLFIRI as second-line therapy in metastatic biliary tract cancer.

Abstract

582 Background: Biliary tract cancers (BTCs), including intrahepatic (IH) and extrahepatic (EH) cholangiocarcinoma and gallbladder cancer, are rare aggressive malignancies with <20% five-year survival. Most patients present with advanced disease. First-line gemcitabine–cisplatin ± durvalumab is standard; however, progression is common and limits second-line options. The ABC-06 trial established FOLFOX as the standard second-line regimen; however, both FOLFOX and FOLFIRI are widely used with differing toxicity profiles. The absence of head-to-head data makes real-world comparative analyses clinically relevant. Methods: We used de-identified data from TriNetX (2010–2025) to identify adults (≥18 years) with metastatic BTC who progressed after first-line therapy and received second-line FOLFOX or FOLFIRI. Propensity score matching (1:1) adjusted for age, sex, comorbidities, lines of treatment, and genomic alterations (KRAS, IDH1, BRAF, BRCA1/2, FGFR2). Overall survival (OS) was estimated using Kaplan–Meier methods, with comparisons by log-rank tests and Cox proportional hazards models. Hazard ratios (HRs) with 95% confidence intervals (CIs) were reported; two-sided P ≤ 0.05 defined significance. Results: A total of 647 patients with metastatic BTC who received second-line therapy following first-line gemcitabine–cisplatin (± durvalumab) were identified. Of these, 408 received FOLFOX only and 239 received FOLFIRI only. The mean age was 67.4 ± 11.5 years in the FOLFOX cohort vs 67.1 ± 11.4 years in the FOLFIRI cohort (p = 0.78). Male patients comprised 50% in the FOLFOX cohort vs 56% in the FOLFIRI cohort (p = 0.428). Most patients received gemcitabine–cisplatin alone. In the univariable analysis, OS was higher with FOLFOX compared to FOLFIRI, with 1-year OS rates of 40.0% vs 37.2% and 5-year OS rates of 16.3% vs 12.8%, respectively. After 1:1 propensity score matching (n = 200 per group), the well-balanced cohorts showed 1-year OS of 45.0% with FOLFOX vs 37.6% with FOLFIRI, and 5-year OS of 16.2% vs 13.1% (P = 0.039). Multivariable Cox regression confirmed a statistically significant reduction in mortality risk with FOLFOX compared to FOLFIRI (HR = 0.91; 95% CI: 0.85–0.97; P = 0.007). Subgroup analyses showed consistent trends toward improved survival with FOLFOX in EH BTC (HR = 0.85; P = 0.07) and IH BTC (HR = 0.91; P = 0.84), though not statistically significant, likely reflecting limited subgroup size. Conclusions: In this real-world analysis, second-line FOLFOX was associated with improved OS compared with FOLFIRI in metastatic BTC. Toxicity data and objective response rates were not available and require future prospective validation. Overall survival with FOLFOX vs FOLFIRI in second-line BTC. Regimen 1-Year OS (%) 5-Year OS (%) FOLFOX 45.0 16.2 FOLFIRI 37.6 13.1 Propensity-matched comparison: P = 0.039. OS = overall survival.