Clinical outcomes for IDH1 mutant biliary tract cancer (BTC) treated with contemporary systemic therapy.

513 Background: Isocitrate dehydrogenase 1 and 2 (IDH1/2) play a key role in cellular metabolism and epigenetic regulation. Conserved missense IDH1 mutations lead to an accumulation of the onco-metabolite 2-hydroxygluterate, which drives oncogenesis and inhibits cellular differentiation. Ivosidenib is now approved for IDH1 mutant BTC following the results of the phase III ClarIDHy trial. It remains unclear what are the long-term outcomes for patients (pts) with IDH1 mutant BTC treated with chemotherapy, targeted therapy, and immunotherapy. Methods: This was a retrospective analysis of BTC pts who underwent prospective, clinical grade, next generation sequencing by MSK-IMPACT 341, 410, 468 or 505. The primary objective was to define the clinical outcomes of systemic treatment for those pts with IDH1 mutant BTC. Secondary objectives included description of co-occurring genomic alterations. Progression-free survival (PFS) was calculated from the start date of treatment to the date of progression or death. Overall survival (OS) was calculated from the date of unresectable/metastatic disease. This study was approved by the MSKCC Institutional Review Board (NCT01775072). Results: 1124 pts with BTC underwent somatic genomic sequencing with MSK-IMPACT, 143 (12.7%) of which had IDH1 mutations. 78 (55%) were female and median age at diagnosis was 54 (range 32-94). Almost all were intrahepatic cholangiocarcinoma (139; 97%), with 2 (1.5%) gallbladder, and 2 (1.5%) perihilar. The most common co-occurring alterations were in ARID1A (33; 21.7%), PBRM1 (29;20.3%) and BAP1 (19;13.3%). Median TMB was 2.6 mut/Mb (0.8-68.5). 2 pts had microsatellite instability and 1 had a co-occurring IDH2 mutation. 112 (78%) had unresectable/metastatic disease at diagnosis. The median number of lines of therapy was 2 (0-9). With a median follow up time of 18.4 months (mos) (range 1.5 - 184.2), median OS was 23.8 mos (95% CI 20.4-29.1) for those with unresectable and metastatic disease. When only accounting for pts with distant metastatic disease, median OS was 20.7 mos (95% CI 17.6-28). In those who had first line platinum-based therapy (86/133, 65%), median PFS (mPFS) was 8.3 mos (95% CI 6.4-11.2). 49 (37%) pts were treated with an IDH1 inhibitor, with 46/49 (94%) receiving ivosidenib, 29/49 (59%) in the second line. mPFS for those treated with ivosidenib in second line was 4.6 mos (95% CI 3.6-10.0) vs. 2.6 mos (95% CI 1.8-6.7) for 5-fu based chemotherapy (p=0.032). There was no difference in OS for those treated with IDH1 inhibitors compared to those that were not (25.7 vs. 20.7 mos; p=0.5). 11 (8%) pts received immunotherapy-based treatments with a mPFS of 2.7 mos (95% CI 2.2-NR). Conclusions: Our retrospective data indicate that IDH1 mutant BTC appears to exhibit similar PFS to first-line cytotoxic chemotherapy compared to historic unselected populations with favorable outcomes to second line IDH1 inhibition.