Molecular and clinical determinants of targeted therapy (TT) outcomes in biliary tract cancer (BTC): Analysis of a prospectively maintained next generation sequencing (NGS) biorepository.

555 Background: BTCs are genomically diverse. TT have established antitumor activity in a subset of pts that harbour actionable alterations; however, most patients (pts) do not respond to or experience acquired resistance on TT. We sought to explore genomic mechanisms of response across the spectrum of BTCs. Methods: We leveraged a prospectively maintained cohort of pts with histologically confirmed BTC who underwent molecular profiling using an FDA authorized, targeted, NGS assay. Additional clinicopathologic information was retrospectively extracted with an aim to describe the genomic profile and clinical actionability using OncoKB, to explore outcomes for pts treated with TT, and to nominate innate and acquired genomic mechanisms of resistance. Results: N = 1254 pts had molecular profiling: 61.2% intrahepatic (iCCA), 16.8% extrahepatic cholangiocarcinoma (eCCA), 22.1% gallbladder cancer (GBC). MSI high was observed in 2.1% and the median TMB was 3.3 (range: 0-74.6 muts/mb). Among MSS tumors, the most frequently altered genes were TP53 (36%), CDKN2A (21%), ARID1A (18%), KRAS (17%), and IDH1/2 (16%). Genomic profiles differed significantly based on anatomic subsite. Oncogenic drivers were mostly clonal except for ERBB2 amplifications and mutations, which were subclonal in 19%, and 28% of cases respectively. OncoKB level 1/2 events observed in 32.2% (iCCA 40%, eCCA 15%, and GBC 22%). Progression-free survival (PFS) for pts with these oncogenic drivers: IDH , FGFR2 , ERBB2, BRAF V600E , MSI high , and TMB high from start of matched TT (N=164) is listed in the table. Clinical factors or co-occurring genomic alterations did not impact outcomes. Profiling of paired pre- and post-progression samples (N=25) nominated pathway reactivation in FGFR2 , BRAF , NTRK, and ERBB2 driven tumors with emergence of recurrent oncogenic FGFR2 , RAS , MEK , and MET alterations, respectively, as well as off target MYC amplification and CDKN2A loss. Loss of the oncogenic driver at progression was only observed for ERBB2 driven tumors. Conclusions: Acknowledging the limitations of the study design, matched TT in a real-world, prospectively maintained cohort offers meaningful PFS, most notably in those pts with MSI high BTC treated with immunotherapy. In those with acquired resistance to TT, alterations predicted to reactivate the primary oncogenic pathway were identified. ERBB2 heterogeneity and loss is predicted to be a mechanism of resistance to HER2 targeted therapy. Outcomes for BTC treated with matched TT. Target Median PFS, months (95% CI) IDH1/2 (N = 72) 4.2 (2.9-6.8) FGFR2 (N=37) 7.4 (6.5-9.4) ERBB2 (N =22) 8.2 (2.95-16.7) MSI high (N = 14) NR (10.15-NR) TMB high (N=12) 2.69 (1.64-NR) BRAF V600E (N=7) 16.6 (3.7-NR)