Skip to content
    Real-World Studies

    Real-World Effectiveness of Cisplatin–Gemcitabine–Durvalumab in Advanced Biliary Tract Cancer

    Camera et alJournal of Gastroenterology

    Real-World Effectiveness of Cisplatin–Gemcitabine–Durvalumab in Advanced Biliary Tract Cancer
    Although cisplatin–gemcitabine–durvalumab has established a new first-line standard of care in advanced biliary tract cancer, the generalizability of TOPAZ-1 trial outcomes to routine clinical practice required validation. To evaluate whether trial survival outcomes translate to real-world populations, a retrospective, international, multicenter study was conducted across 55 centers in 12 countries.
    The primary overall survival analysis evaluated 1,382 real-world patients treated with cisplatin–gemcitabine–durvalumab, comparing them against reconstructed individual patient-level data from the TOPAZ-1 trial arm. Baseline characteristics between the real-world and clinical trial cohorts were aligned using entropy balancing. Patients were further categorized by trial eligibility: 1,358 patients had evaluable trial eligibility status (912 TOPAZ-1-in and 446 TOPAZ-1-out), and 1,351 patients with complete required data were included in eligibility-stratified survival analyses (910 TOPAZ-1-in and 441 TOPAZ-1-out).
    Following calibration, baseline characteristics were well balanced between groups. Overall survival was longer in the real-world cohort compared to the TOPAZ-1 comparator in both unweighted (hazard ratio [HR] 0.83, p = 0.012) and calibrated analyses (HR 0.84, p = 0.018). When stratified by trial criteria, trial-eligible real-world patients had longer observed overall survival than the trial cohort (HR 0.76, p < 0.001), while trial-ineligible patients experienced similar survival outcomes (HR 0.98, p = 0.85), with calibrated models yielding consistent results.
    Restricted mean survival time analyses showed increasing absolute between-cohort survival differences over time. In exploratory subgroup evaluations, the largest observed survival differences favored real-world patients with an ECOG performance status of 0 (calibrated HR 0.61, p < 0.001) and those with locally advanced disease (calibrated HR 0.38, p < 0.001). The study concluded that TOPAZ-1 survival outcomes were broadly reproduced in routine practice, though subgroup findings in locally advanced disease and ECOG PS 0 represent exploratory results requiring cautious interpretation due to reduced effective sample sizes and limited empirical overlap.

    Key Takeaways

    • In a calibrated comparative analysis, patients with advanced biliary tract cancer receiving real-world cisplatin–gemcitabine–durvalumab had longer overall survival than observed in the TOPAZ-1 trial arm (calibrated HR 0.84, p = 0.018; unweighted HR 0.83, p = 0.012).
    • Real-world patients meeting trial eligibility criteria had longer observed overall survival than the TOPAZ-1 cohort (HR 0.76, p < 0.001), whereas trial-ineligible patients had similar survival outcomes to the trial cohort (HR 0.98, p = 0.85).
    • Exploratory subgroup analyses identified the greatest survival differences among patients with ECOG performance status 0 (calibrated HR 0.61, p < 0.001) and locally advanced disease (calibrated HR 0.38, p < 0.001).
    Read the original sourceJournal of Gastroenterology