EXTENSIVE MOLECULAR CHARACTERIZATION FOR PROGNOSTIC AND PREDICTIVE TRANSLATIONAL ANALYSES IN ADVANCED BILIARY TRACT CANCERS

BACKGROUND: The prognosis of advanced or metastatic biliary tract cancers (BTC) is dismal with a 5-years expected survival rate of 17% in men and 15% in women, respectively. This is mainly due to limited treatment options and poor responses to conventional therapies coupled with a poor understanding of the biological background and lack of specific targeted drugs. Extensive genetic profiling allowed to distinguish among these heterogeneous category of cancers different molecular entities thus paving the way for targeting specific genomic alterations (e.g., FGFR2 fusions and IDH1/2 mutations). Notwithstanding, comprehensive, precise and reliable characterizations of clinical and pathological features of BTCs according to their molecular background are still being studied. On that basis, we planned a large monocentric translational study to describe the incidence and the association of selected molecular features with survival outcome and main clinical and pathological characteristics in advanced BTC patients. MATHERIAL AND METHODS: This is a retrospective study conducted at Istituto Oncologico Veneto in Padua in collaboration with the Surgical Pathology Unit, University Hospital of Padua. All consecutive patients with a confirmed diagnosis of advanced/metastatic intrahepatic cholangiocarcinoma (iCCA) and availability of formalin-fixed, paraffin-embedded tissue samples were considered eligible. From January 2008 to May 2023 they were retrospectively selected, and then prospectively followed until November 2023. Targeted alterations included: FGFR2/3 aberrations, IDH1/2 mutations, HER2 alterations, BRAF alterations, NTRK alterations and mismatch repair proteins deficiency (MMR). These alterations were identified by means of immunohistochemistry, RT-PCR, next-generation sequencing, and fluorescence in situ hybridization. Data collected included demographic and pathological characteristics, treatments, 1st line progression free survival, overall survival. Response to treatments was evaluated by RECIST criteria, version 1.1. RESULTS: A total of 323 patients with adequate clinical information constituted our clinical dataset. All cases with available tissue samples were re-evaluated by two pathologists to assess the quality of samples for molecular analyses, to confirm the primary tumor site and pathological characteristics. Two hundred forty-seven patients with iCCA were eligible and constituted our molecular dataset. Twenty-three patients (9.3%) presented FGFR2 fusions/amplifications, 224 patients (90.7%) were wild type (wt). IDH1/2 resulted mutated in 50 patients (20.2%). A total of 18 patients (7.3%) had a HER2 gene alteration, 9 patients (3.6%) had a BRAF gene alteration, 3 patients (1.2%) NTRK1amplifications and 6 patients (2.4%) had a MMR deficit. Patients harboring FGFR2 fusions/amplifications had a lower risk of death compared to wt patients (HR=0.48, 95% CI 0.28 - 0.80, p-value=0.005). FGFR2 altered patients had a median OS of 24.5 months (95% CI 19.4 – NA) compared to 15.7 moths (95% CI 13.4 – 18.6) of wild type patients. This advantage in OS was maintained also at multivariate analysis (p-value = 0.009). CONCLUSION: Our data provide a strong proof - challenged with a robust and detailed multivariate model - that FGFR2 alterations can be prognostic for better survival. The recognition and the measurement of their prognostic impact could be of primary importance for the correct interpretation of currently available data. Our data should be considered in the design of new trials, for example as adjusting or stratification factors in prospective clinical studies.