Prognostic impact and clinical characteristics of KRAS mutations and CDKN2A loss in IDH1 -mutant intrahepatic cholangiocarcinoma

Background: In the ClarIDHy trial, ivosidenib significantly improved progression-free survival (PFS) compared with placebo in patients with previously treated IDH1 -mutant cholangiocarcinoma (CCA) and is currently awaiting insurance approval in Japan. However, approximately 40% of patients in both arms experience disease progression within 2 months, suggesting underlying biological heterogeneity within IDH1 -mutant CCA. We hypothesized that genomic co-alterations may serve as potential determinants of this heterogeneity. Methods: We analyzed patients with intrahepatic cholangiocarcinoma (iCCA) in the Center for Cancer Genomics and Advanced Therapeutics (C-CAT) database between June 2019 and June 2025. Point mutations were annotated using OncoKB, and copy number variants and rearrangements were evaluated using C-CAT (approval number CDU2021-001N). Additionally, we reviewed institutional patients with IDH1 -mutant iCCA diagnosed between October 2001 and October 2025, with detailed pathological and radiological assessments (approval number 2018-149). Results: Among 2484 patients with iCCA, 353 had IDH1 mutations (14.2%). IDH1 -mutant iCCA tended to show longer overall survival (OS) and time to treatment failure (TTF) for first-line therapy than IDH1 -wild type (median OS, 22.4 vs. 20.1 months; HR, 0.83; P =0.054; median TTF, 7.8 vs. 6.4 months; HR, 0.87; P =0.068). In IDH1 -mutant iCCA, KRAS mutations and CDKN2A loss were less frequent than IDH1 -wild type ( KRAS mutations, 11.6% vs. 27.6%, P <0.001; CDKN2A loss, 17.8% vs. 27.5%, P <0.001). Among IDH1 -mutant iCCA, KRAS mutations showed a trend toward shorter OS (median, 17.0 vs. 23.2; HR, 1.60; P =0.069) and were associated with shorter TTF (median, 5.7 vs. 8.5 months; HR, 1.86; P =0.0039). CDKN2A loss was correlated with shorter OS (median, 17.6 vs. 23.7 months; HR, 1.58; P =0.037). IDH1 -mutant iCCA without KRAS mutations or CDKN2A loss demonstrated significantly longer OS (median, 24.9 vs. 17.6 months; HR, 0.61, P =0.010) and TTF (median, 8.4 vs. 6.1 months; HR, 0.70, P =0.020) than the other subsets. In multivariate analysis, KRAS mutations predicted shorter OS (HR, 1.75; P =0.046) and TTF (HR, 1.77; P =0.012). In an institutional cohort of 37 patients, only one case was pathologically classified as large-duct type and harbored KRAS mutation, with all other cases classified as small-duct type. Radiologically, IDH1 -mutant iCCA without KRAS mutations and CDKN2A loss tended to present in a peripheral location (60.0% vs. 33.3%) and intratumoral transversing vessels (84.0% vs. 55.6%). Conclusions: IDH1 mutations less frequently co-occur with KRAS mutations and CDKN2A loss. IDH1 -mutant iCCA without these co-alterations exhibited favorable clinical outcomes and distinct radiopathological characteristics, highlighting the biological heterogeneity associated with genomic co-alterations in IDH1 -mutant iCCA.