The relationship between tumor mutational burden (TMB) and microsatellite instability (MSI) in biliary tract tumors and molecular background changes at different TMB thresholds.

Abstract

e15167 Background: Pembrolizumab obtained FDA approval for the treatment of metastatic tumors with high mutational burden, regardless of histopathology. Subsequently, the efficacy of TMB and immunotherapy was confirmed in biliary tract cancer (BTC). Presently, the relationship between TMB and MSI-H, as well as the molecular background under different TMB statuses in BTC, remains unclear. Methods: This study investigated the formalin-fixed paraffin-embedded (FFPE) tissues of 2200 patients with BTC who underwent next-generation sequencing (NGS) between November 2017 and January 2024 at 3DMed Clinical Laboratory Inc. These patients are currently under follow-up. Statistical analysis was conducted using Microsoft Office Excel (2013). The study classifies TMB into 10 muts/MB, 20 muts/MB, and 50 muts/MB, and explores their correlation with MSI-H and alterations in the molecular background. Results: Among patients with TMB values of ≥10, 20, and 50, the proportions were 280/2200, 88/2200, and 24/2200, respectively. Notably, as the TMB threshold increased, the corresponding MSI-H proportions rose to 41/280, 40/88, and 15/24. For patients with TMB < 10 and MSS type, mutations were predominantly in TP53 (913/1917), along with KRAS, ARID1, SMAD4 (29.1%, 15.0%, and 12.0%). In patients with TMB > 10 or > 20 and MSI-H, mutations in KMT2C, TGFBR2, ACVR2A, TP53, KMT2D, ARID1A, LRP1B, PBRM1 exceeded 50%, with ACVR2A, KMT2C, TGFBR2 mutation rates reaching up to 80%. For all patients with TMB ≥50 and MSI-H, ACVR2A mutations were present, with KMT2C and KMT2D mutation rates above 90%, and ARID1A, TGFBR2, RNF43 above 80%. Conclusions: As TMB values increase, there is a corresponding elevation in the proportion of patients with MSI-H. Additionally, there are distinct differences in gene mutations between TMB-H and MSI-H patients with TMB-L and MSS patients in the Chinese BTC population, particularly ACVR2A, KMT2C, KMT2D, ARID1A, TGFBR2, and RNF43, which hold significant value for precise patient stratification and treatment. Molecular subsets of TMB-L/MSS and TMB-H/MSI-H could contribute to BTC prognosis evaluation in future clinical studies.