Evaluating the Prognostic Impact of IDH Mutations in Intrahepatic Cholangiocarcinoma
Rachel N Harvey1, Rebecca Gelfer1, Esther Drill2
1Department of Surgery, Memorial Sloan Kettering Cancer Center, New York, NY.
Purpose:
Isocitrate dehydrogenase 1 and 2 (IDH1 and IDH2) mutations are common in intrahepatic cholangiocarcinoma (ICC), but their prognostic value is unclear. Using a large data set, we assessed their impact in resected and nonresected ICC.
Methods:
Adults from two medical centers (MSKCC and Erasmus) with ICC treated with curative-intent resection (resected) or managed nonoperatively (unresectable) who underwent next-generation sequencing were analyzed retrospectively. Kaplan-Meier and Cox regressions assessed the impact of IDH status on outcomes.
Results:
Of the 795 patients analyzed, 25% had IDH1/2 mutations (IDHmut) and 43% underwent resection. Median overall survival (OS) of the cohort was 32 months in IDHmut and 28 months for IDHwt (P = .2). High-risk genetic alterations (TP53mut, KRASmut, and CDKN2Adel) were more frequent in IDH wild-type (IDHwt; odds ratio, 2.26; q < 0.001). OS was 19 months in patients with high-risk alterations versus 40 months in patients without (P < .001). In resected patients, recurrence-free survival (RFS) in IDHmut was 20 months versus 14 months for IDHwt (P = .018), and OS was 69 months versus 50 months, respectively (P = .2). However, after controlling for high-risk alterations, the potential benefit of IDHmut was no longer apparent (RFS: hazard ratio [HR], 0.78; P = .095; OS: HR, 0.88; P = .4). In unresectable IDHmut patients, progression-free survival was 9.4 months versus 9.1 months for IDHwt (P = .7), and OS was 22 months versus 18 months, respectively (P = .13). There remained no differences after controlling for high-risk alterations. IDH status was not a significant survival predictor in multivariable models.
Conclusion:
In this cohort of patients with ICC, IDHmut was not an independent predictor of survival, after controlling for high-risk alterations and clinical variables. IDH mutational status alone should, therefore, not be used to guide prognosis.

