Durvalumab or pembrolizumab plus gemcitabine and cisplatin versus gemcitabine and cisplatin alone as first-line therapy for advanced cholangiocarcinoma: A meta-analysis.

Abstract

4160 Background: Cholangiocarcinoma (CCA) is a rare, aggressive malignancy of the bile duct epithelium with rising incidence and mortality. Curative surgery is feasible in few cases; for unresectable or metastatic disease, systemic therapy is standard. Gemcitabine plus cisplatin (GemCis) has been the first-line backbone, but phase 3 trials (TOPAZ-1 and KEYNOTE-966) demonstrated overall survival (OS) benefit with added immune checkpoint inhibitors (durvalumab or pembrolizumab). This systematic review and meta-analysis evaluates the efficacy and safety of immunotherapy (IO) plus GemCis versus GemCis alone in advanced CCA. Methods: We systematically searched PubMed, Embase, Scopus, and Google Scholar for English-language studies (2009–2025) including randomized controlled trials (RCTs) and cohort studies of first-line GemCis alone or with durvalumab or pembrolizumab in unresectable/metastatic CCA. Primary endpoint: OS. Secondary: progression-free survival (PFS) and grade 3-4 adverse events (AEs) per CTCAE. Kaplan-Meier curves were digitized, and individual patient data reconstructed using IPD tools for pooled estimates. Random-effects models generated hazard ratios (HRs) and pooled medians; heterogeneity assessed via I². Results: We included 3,703 patients from TOPAZ-1, KEYNOTE-966, ABC-01/02, and six cohort studies: 1,737 on GemCis alone, 1,433 on GemCis + durvalumab, and 533 on GemCis + pembrolizumab. In the full meta-analysis, GemCis + durvalumab showed the best outcomes (median OS 14.1 months, PFS 7.38 months) versus GemCis + pembrolizumab (OS 12.8 months, PFS 6.62 months) and GemCis alone (OS 9.27 months, PFS 5.81 months). Restricting to RCTs only, IO + GemCis improved OS versus GemCis alone (median 12.82 vs. 11.1 months; HR 0.78, 95% CI 0.71–0.86, p<0.001), with no significant PFS difference (p=0.20). Grade 3-4 AEs were infrequent across arms; thrombocytopenia was higher with pembrolizumab (p=0.0002), but other severe toxicities were comparable. Conclusions: Adding immunotherapy to GemCis significantly enhances survival in unresectable/metastatic CCA, with durvalumab + GemCis providing the most pronounced benefit in this pooled analysis incorporating RCTs and real-world cohorts. Regimens were well-tolerated with manageable toxicity. These findings reinforce chemoimmunotherapy as first-line standard and highlight potential differential efficacy between PD-L1 and PD-1 inhibitors warranting further study.