Network meta-analysis of adjuvant chemotherapy in biliary tract cancers: Setting the scene for new randomized evidence.

Abstract

The best adjuvant chemotherapy for resected biliary tract cancer (BTC) is under debate, with capecitabine supported by weak evidence. Aim of this network meta-analysis is to estimate the efficacy of different phase II/III regimens, comparing monotherapies (gemcitabine or fluoropyrimidines) head-to-head, against observation and combination regimens. A comprehensive literature search was conducted on PubMed and EMBASE for phase II/III randomized clinical trials (RCTs) available as of December 2023, reporting hazard ratios (HRs) of overall survival (OS) and event-free survival (EFS). A frequentist framework employing a random-effects model was applied; treatment rankings were outlined according to P-score, based on direct and indirect evidence. Exploratory subgroup analyses for OS were also performed (primary site, resected margin status and nodal involvement). Six RCTs (1979 total patients) were identified. Fluoropyrimidine monotherapy showed significantly better OS (HR .84 [.72-.97]) and EFS (HR .79 [.69-.91]) than observation, as any monotherapy did (HR .84 [.74-.96]; HR .79 [.70-.89]). In the head-to-head comparison for OS, only S1 confirmed to be superior to observation alone (HR .69 [.49-.98]) while fluoropyrimidines achieved the best P score (.81), similarly to any monotherapy (0.92). Combinations failed to prove superior to monotherapies with respect both to OS and EFS. Subgroup analyses were inconclusive due to results' inconsistency and limited sample size. Our work confirmed that adjuvant chemotherapy grants OS and EFS benefit for resected BTC patients. Fluoropyrimidines appeared the most effective option, confirming capecitabine as the preferred choice for the Western population.