Clinical outcomes of patients with IDH1/IDH2-mutated clonal hematopoiesis: A single-center retrospective study

Abstract Introduction: Clonal hematopoiesis is defined by the expansion of hematopoietic stem or progenitor cells carrying myeloid malignancy–associated somatic mutations and is increasingly recognized as a precursor for myeloid neoplasms. Mutations in IDH1 and IDH2 are of particular interest due to their early role in leukemogenesis and disease transformation in chronic myeloid malignancies, and frequent association with clonal hematopoiesis of indeterminate potential (CHIP) and clonal cytopenia of undetermined significance (CCUS), and targetability in myeloid neoplasms. These mutations promote clonal expansion by altering cellular metabolism and the production of oncometabolites. The natural history and clinical outcomes of individuals with isolated IDH1 or IDH2 clonal hematopoiesis remain poorly defined. We aimed to better understand the clinical course and outcomes of IDH-mutated (IDHm) clonal hematopoiesis patients (pts). Methods: We conducted a retrospective study at a tertiary cancer center to evaluate pts with IDHm CHIP or CCUS. Pts were included if IDH1 or IDH2 mutations of presumed hematopoietic origin were identified on next-generation sequencing performed on bone marrow (BM) or peripheral blood (PB). Mutations detected on liquid biopsy were considered hematopoietic in origin if they met one or more of the following criteria: (1) Absent in matched tumor tissue sequencing, (2) detected in PB mononuclear cell (buffy coat) controls, or (3) present in both PB and tumor with a higher variant allele frequency (VAF) in PB. Time-to-event analyses were performed using Kaplan-Meier estimates, and Cox proportional hazards models using JMP® statistical software. Results: A total of 22 pts with IDHm clonal hematopoiesis with VAF ≥2% were included in the analysis; 6 (30%) IDH1m and 14 (70%) IDH2m. Among 22 pts, 13 (59%) had CHIP, and 9 (41%) had CCUS. All IDH2m pts (n=16) had R140 mutations; IDH1m pts (n=6) had R132 (83%) and V152 (17%). The median age was 71 years (range, 46–91), and 9 (41%) were female. A history of smoking and alcohol use was noted in 9 (40%) and 14 (64%) pts, respectively. Common comorbidities, including hypertension (n=11, 50%) and coronary artery disease (n=2, 15%), were reported. Eighteen pts (82%) had a history of prior malignancy, with 12 (55%) and 6 (27%) receiving prior chemotherapy and radiation, respectively. The median number of co-mutations was 4 (range 1–9), and the median IDH VAF was 8% (range, 2–13). Based on the Clonal Hematopoiesis Risk Score (CHRS), 12 (55%) were classified as high-risk and 10 (45%) as intermediate-risk. The median follow-up time from clonal hematopoiesis detection was 28.5 months (range, 6.0–59.3). The median overall survival was 48.3 months (95% confidence interval [CI] 6.9–104.6). Seven pts (32%) progressed to a myeloid malignancy (myelodysplastic syndrome [MDS] in 5 [23%], acute myeloid leukemia [AML] in 2 [9%]) with a median time to progression of 20.9 months (range, 5.9–37.9). The median overall survival was 28.0 months (95% CI 0-99.3) from the time of myeloid transformation. Of 7 pts, 1 MDS and 2 AML pts (43%) received IDH inhibitors, 2 MDS pts were treated with hypomethylating agent-based therapy, and 2 MDS pts had supportive care only. Eight deaths (36%) occurred during follow-up related to their primary solid tumor (n=5, 63%), congestive heart failure (n=1, 13%), and transformation to a myeloid neoplasm (n=1, 13%). The cause of death was unknown in 1 patient. On univariate analysis, the presence of cytopenia (hazard ratio [HR] 4.2, 95% CI 1.12–15.4, p=0.02) was significantly associated with progression to MDS/AML. Age, gender, IDH mutation subtype, number of mutations (≥2), VAF, CHRS score, and prior history of malignancy and receipt of antineoplastic therapy were not statistically significant predictors. Conclusions: Pts with CCUS had a significantly higher risk of progression to MDS/AML compared to those with CHIP, underscoring the clinical relevance of cytopenia in IDHm clonal hematopoiesis. The lack of association between ≥2 mutations and progression suggest that even a single IDH1 or IDH2 mutation carries a higher risk in this cohort of predominantly cancer pts. These findings highlight the importance of early identification and risk-adapted monitoring. Although optimal interventions remain undefined, ongoing studies exploring mutation-specific approaches, such as IDH-targeted therapies, may be helpful in this high-risk population.