Real-world outcomes of myelodysplastic syndrome and chronic myelomonocytic leukemia treated with IDH1 and IDH2 inhibitors in the frontline setting

Abstract

Abstract Introduction: Isocitrate dehydrogenase 1 or 2 (IDH1/2) mutations occur in 3-5% of patients with myelodysplastic syndrome (MDS). Selective IDH1/2 inhibitors including ivosidenib (IDH1), olutasidenib (IDH1), and enasidenib (IDH2) are currently approved for IDH1/2-mutated (mIDH1/2) relapsed/refractory (r/r) acute myeloid leukemia; ivosidenib is also approved for mIDH1 r/r MDS based on clinical benefit observed in 18 patients on trial AG120-C-001. However, there are limited prospective clinical trial or real-world data for IDH inhibitors in treatment-naïve MDS and chronic myelomonocytic leukemia (CMML). Methods: This was a retrospective, single-institution analysis of mIDH1/2 MDS and CMML patients treated with an IDH inhibitor in the frontline setting between 05/2018-09/2024. Patients previously treated with venetoclax (VEN) or a hypomethylating agent (HMA) were excluded. The primary outcomes of interest were treatment response based on the 2023 modified International Working Group criteria for MDS and rates of peripheral blood (PB)-complete remission (CR) (Stampfl-Mattesberger, Blood, 2021). Secondary objectives included time to best response, evaluation of transfusion independence (TI) pre/post-IDH1/2 therapy, rates of IDH1/2 variant allele frequency (VAF) reduction, rates of differentiation syndrome (DS), and overall survival (OS). Results: Our cohort (n=16) included 7 mIDH1, and 9 mIDH2 patients (13 MDS, 3 CMML) of which 14 (87.5%) were male, and 13 (81.3%) were White. The median age at diagnosis was 78 years (range [r], 55-91), with 7 (43.8%) patients >80 and 3 were ≥90 years old. Eleven (68.8%) had high-risk disease by IPSS-R (>3.5) and IPSS-M (>0). The most common IDH1 and IDH2 variants were R132H (28.6%) and R140Q (44.4%), respectively. The median VAF of IDH1 was 29.3% (r, 9.4-46.6) and 35.8% (r, 18.1-42) for IDH2. Most (75%) of our cohort received an IDH1/2 inhibitor as monotherapy while 3 (18.8%) received it in combination with a hypomethylating agent (HMA). Among patients with lower risk disease, the most common reason for treatment initiation was neutropenia. The median follow-up was 16.5 months, and the median treatment duration was 10 months for both mIDH1 and mIDH2 patients. Of those who received at least one cycle, 9 (56.3%) had BM evaluations (BME), of which 44.4% achieved CR, 22.2% achieved CR with limited count recovery (CRl), 11.1% achieved CR with partial hematologic recovery (CRh), and 22.2% did not respond. Eight (50%) patients achieved PB-CR. Median times to best response (PB or BM) among mIDH1/2 patients were 100 days (r, 56-527) and 85 days (r, 27-240), respectively. Of the 9 patients who were TD prior to IDH1/2 therapy, 6 (66.7%) achieved TI. Nine mIDH1/2 patients had sequential next-generation sequencing (NGS) by BM, and 5 (55.5%) had VAF reduction, though none achieved mutation clearance. DS was suspected in 18.8%, most (66.7%) within the first 90 days of treatment. Only one patient discontinued treatment permanently despite resolution of symptoms without glucocorticoids due to a comorbid substance use disorder. Four (25%) patients underwent stem cell transplantation (HCT), in 1 of whom received IDH1 inhibitor as monotherapy. The median OS was 26 months for the entire cohort and the median OS for patients >80 years was also 26 months. There was a trend toward higher hazard of death for patients receiving IDH inhibitor monotherapy versus in combination with HMA (HR 1.87; 95% CI, 0.47–7.38), and a trend for improved OS when stratified by HCT (HR 0.46; 95% CI, 0.09-2.29). Conclusion: This study highlights safety and efficacy of IDH inhibitors, particularly among older adults with treatment-naïve MDS or CMML. Nearly half of our cohort were adults >80 years, a group often underrepresented in pivotal clinical trials. The overall response rate (CR + CRl + CRh) was 88.9% among 9 patients with BME; 50% of the entire cohort achieved PB-CR and median OS was 26 months. Of those who were previously transfusion dependent, 66.7% achieved TI and rate of DS (18.8%) was comparable to historical prospective data. We are continuing to examine days alive out of hospital as a patient-centered outcome.