Interim Results of the Phase II Study Investigating CPX-351 in Combination with Ivosidenib for Patients with IDH1-Mutated Acute Myeloid Leukemia and High-Risk Myelodysplastic Syndrome

Background: Isocitrate dehydrogenase 1 (IDH1) mutations are present in 7-14% of patients with acute myeloid leukemia (AML) and 2-4% of patients with myelodysplastic syndrome (MDS). Mutations in the IDH1 gene interfere with normal myeloid differentiation through the production of the oncometabolite 2-hydroxyglutarate (2HG). Ivosidenib is a potent, oral, and highly selective IDH1 inhibitor that reduces tumor 2HG levels and is currently approved for the treatment of relapsed/refractory (R/R) or newly diagnosed AML in older patients ineligible for intensive chemotherapy. Ivosidenib has also been evaluated in combination with standard 7+3 chemotherapy (Stein EM, Blood 2021) and was shown to be safe and lead to robust remission rates. CPX-351 is a nano-scale liposome that includes a fixed molar ratio of cytarabine and daunorubicin and has demonstrated superior response rate and overall survival compared to 7+3 in patients with newly diagnosed secondary AML (Lancet JE, Lancet 2021). Based on the data showing the safety and efficacy of 7+3 and ivosidenib and the superiority of CPX-351 to 7+3 in a randomized study, we designed this phase II, investigator-initiated trial of CPX-351 and ivosidenib for adults with newly diagnosed or R/R IDH1-mutated AML and high-risk MDS (NCT04493164).

Methods: Adult patients 18 years or older with newly diagnosed (Arm A) or R/R (Arm B) AML or high-risk MDS harboring an IDH1 mutation were eligible. All patients received induction therapy with CPX-351 (daunorubicin 44 mg/m² and cytarabine 100 mg/m²) administered according to its FDA-approved induction (day 1, 3, and 5 of a 28-day cycle for first induction and day 1 and 3 of a 28-day cycle for second induction if complete remission (CR) has not been attained) and consolidation (day 1 and 3 of a 28-day cycle for up to 2 cycles) schedules. Ivosidenib was administered orally at a dose of 500 mg daily starting on day 1 of induction and continued through consolidation, then subsequently continued as an oral maintenance therapy for a duration of up to 2 years. Key endpoints included efficacy and safety, as well as time-to-event outcomes of overall survival (OS), event-free survival (EFS), and duration of response (DOR).

Results: As of 7/10/2024, 11 patients were enrolled (4 in Arm A and 7 in Arm B) and are response-evaluable. The median age was 57 years (range 45-72). The median prior lines of therapy in Arm B was 3 (range 1-4); 5 patients (71%) had received a prior hypomethylating agent, 7 patients (100%) had prior venetoclax, and 2 patients (29%) had a prior HSCT. The most common all-grade treatment related adverse events (TRAEs) were rash (45%), thrombocytopenia (18%), and ECG changes (QTc prolongation in 9% and bradycardia in 9%). No episodes of differentiation syndrome were reported. Grade 3 or higher TRAEs were rash (27%) and thrombocytopenia (18%). There were 2 occurrences of dose-limiting toxicity (DLT), consisting of grade 4 prolonged thrombocytopenia in the setting of marrow remission, which resolved without dose hold on day 45 in both patients with attainment of CR and complete remission with incomplete hematologic recovery (CRi). There were no deaths due to TRAEs. Median count recovery following induction occurred at 39 days (range 30-55).

In Arm A, the overall response rate (ORR) was 100% (4/4), with 1 patient achieving CR, 2 patients achieving CRi, and 1 patient achieving a morphologic leukemia free state (MLFS). Measurable residual disease (MRD) was undetectable by flow cytometry (sensitivity 10^-4) in all 4 patients (100%). The median DOR was 16.3 months with a median EFS of 17.5 months and median OS of 29.3 months. In Arm B, the ORR was 43% (3/7), with 1 patient achieving CR and 2 patients achieving CRi. MRD was undetectable in all 3 responding patients. Median EFS was 4.2 months and median OS was 12.0 months. Four responding patients proceeded to HSCT; 3 remain in remission and 1 newly diagnosed patient experienced relapse at 17.5 months post HSCT. IDH1 variant allele frequency (VAF) was measured in follow-up in 9/11 patients, with a decrease in VAF observed in 8/9 patients (89%) and VAF becoming undetectable (sensitivity of 2%) in 6/9 patients (67%).

Conclusion: Interim results suggest acceptable safety and high efficacy of CPX-351 in combination with ivosidenib in newly diagnosed or R/R IDH1-mutated AML. This study is ongoing and updated data will be presented as available.