Multimodal evaluation of metabolic dysfunction–associated steatotic liver disease (MASLD)–related biliary tract cancer (BTC) and immunotherapy outcomes.

4091 Background: Rising incidence of BTC, particularly intrahepatic cholangiocarcinoma (iCCA), may be linked to increasing incidence of obesity, MASLD and type 2 diabetes mellitus (T2DM). Immunotherapy with immune checkpoint inhibitors (ICIs) has modestly extended survival in biliary tract cancer. However, the prevalence of MASLD in BTC, its immune microenvironment (TME) and outcome of MASLD-BTC with ICIs are unknown. Methods: Retrospective analysis of BTC patients (pts) treated with ICI between 5/2021-5/2024 with durvalumab, cisplatin, and gemcitabine. We used American Association for Liver Diseases (AASLD) criteria for MASLD: 1) metabolic dysfunction and 2) steatosis on imaging or biopsy. We calculated liver proton-density fat fraction (PDFF) in pre-treatment non-contrast CT scans (PDFF estimate 5% labelled as steatosis). We examined the statistical association between BMI and both tumor genotype and gene expression patterns using data from institutional genomic platforms (MAPP2 and RTI). Results: 179 BTC pts (65% of whom were iCCA) treated with durvalumab, cisplatin, and gemcitabine, 103 (57.5%) met AASLD MASLD criteria. In evaluable pts, the median overall survival (OS) was 18.4 months, and median follow-up time was 16.7 months. Non-MASLD pts had a median OS of 23.0 months (95% CI: 16.2, NA) versus 16.7 months (95% CI: 12.4, 21.2) with MASLD (p = .056). T2DM was associated with a worse OS (10.6 versus 21.2 months, p = .004). Multivariable cox model for OS demonstrated a hazard ratio (HR) of 1.45 (95% CI: 0.87, 2.4; p = .1564) for MASLD and 1.61 (95% CI: .99, 2.65; p = .0575) for T2DM. The median progression-free survival (PFS) was 8.5 months. MASLD BTC had a median PFS of 8.2 months (95% CI: 5.7, 10.1) versus 9.2 months (95% CI: 6.9, 16.2) without MASLD (p = .459). BTC pts with T2DM had median PFS of 5.8 months vs 13.0 months without T2DM (p = .001). Multivariable cox model for PFS had HR of 1.7 (95% CI: 1.1, 2.6; p = .014) for T2DM. Within our institutional database (n = 919), we observed depletion of KRAS (p = .008) and STK11 (p = .02) mutations in BTC pts with high BMI. RNA-seq (n = 77) suggests that elevated BMI was associated with low expression of pan-immune and epithelial-to-mesenchymal transition and increased expression of oxidative phosphorylation signatures. Conclusions: BTC is commonly associated with MASLD and may correlate with reduced OS and PFS with ICI, particularly in T2DM pts. Our findings suggest a distinct immunogenomic signature in MASLD-BTC and highlight the importance of further investigating the TME in this population.