Dual BRAF/MEK inhibition in <i>BRAF</i> ^V600E –mutated biliary tract cancer with exploratory histology-dependent response: A systematic review and meta-analysis.

Abstract

4108 Background: The BRAF V600E mutation defines an actionable subset of rare cancers. Metastatic BTC carries a poor prognosis, with median survival under one year on standard chemotherapy. Given the rarity of BRAF V600E-mutated BTC and the absence of RCTs, rigorous evidence synthesis is needed; existing data for dabrafenib plus trametinib come only from single arm basket trials such as ROAR and NCI-MATCH. The low prevalence of this mutation limits the feasibility of large randomized studies. This SRMA aimed to quantify the efficacy of dual BRAF/MEK inhibition in BTC and explore potential histology dependent differences in response. Methods: We performed a systematic review and meta-analysis (SRMA) pooling single arm Phase II basket trial cohorts (N = 53 BTC; N = 17 CNS). Seven case reports were summarized qualitatively and excluded from pooled analyses. The primary endpoint was pooled overall response rate (ORR). A random effects model was used. Subgroup and heterogeneity analyses were performed given the tissue-specific behavior of V600E-driven tumors. Histology was evaluated as an exploratory moderator and heterogeneity was quantified. Results: In the pooled cohort of 70 patients (53 BTC and 17 CNS), the overall ORR was 45.3% (95% CI: 24.5-66.9%). Heterogeneity for the BTC subgroup was moderate (I 2 = 62.6%), while overall pooled heterogeneity across all studies was low (I 2 = 14%). ORR in BTC was numerically higher than in Central nervous system (CNS) cohorts (34.9%) but subgroup differences did not reach statistical significance (P = 0.3386). The observed median OS of 13.5 months in the ROAR BTC cohort compares favourably with historical outcomes commonly reported in the 6-12 month range for advanced BTC. The consistency of pooled estimates, despite non-randomized designs, suggests a stable treatment signal while acknowledging that causality cannot be inferred from uncontrolled data. Median progression-free survival was 9.0 months in BTC and 5.5-8.0 months in CNS cohorts. Median OS in the ROAR BTC cohort was 13.5 months (95% CI: 10.4-17.6), exceeding historical expectations. The pooled rate of Grade ≥3 AEs was 59.2%, consistent with the known safety profile of dabrafenib plus trametinib; no treatment-related deaths were reported. Qualitative data from case reports confirmed regression of CNS metastases and feasibility in patients with severe hepatic dysfunction. Conclusions: Dual BRAF/MEK inhibition shows clinically meaningful activity in BRAF V600E-mutated BTC, with survival exceeding historical expectations. Exploratory analyses indicate histology-dependent differences, supporting tumor-specific therapeutic stratification. These findings align with growing tissue-agnostic evidence for MAPK inhibition and reinforce the value of routine molecular profiling, though conclusions are limited by small cohorts and single arm studies.