Venetoclax in combination with cytarabine with or without idarubicin in children with relapsed or refractory acute myeloid leukemia: A multi-center phase II expansion study

Abstract Background: The prognosis for pediatric patients with relapsed or refractory acute myeloid leukemia (AML) remains poor, with 5-year survival rates of approximately 35%. Venetoclax, a selective BCL-2 inhibitor has been approved for use in adults unfit for intensive therapy and has shown promise in phase I pediatric data. However, limited published pediatric data are available. Methods: This multicenter Phase I/II trial (NCT03194932) combined venetoclax with cytarabine, with or without idarubicin or azacitidine, in pediatric patients with relapsed/refractory AML. Venetoclax was administered at the recommended phase II dose (RP2D; 360 mg/m2 daily) alongside conventional chemotherapy starting on day 8, including 1,000 mg/m2of cytarabine with or without 12 mg/m2of idarubicin for patients with limited prior anthracycline exposure, and/or azacitidine 75 mg/m2for 7 days. BH3 profiling was conducted to assess anti-apoptotic dependencies. Primary endpoints included safety and clinical response at the end of cycle 1. Secondary endpoints assessed response to venetoclax monotherapy and the correlation of BH3 profiling to treatment response. Results: Results of the phase I portion of this study have been previously published and showed that all dose levels were well tolerated. Venetoclax 360 mg/m2 (max 600 mg) with high-dose cytarabine (HDAC), with or without idarubicin or azacitidine, was identified as the RP2D. Forty of 45 participants treated at the RP2D were evaluable for response. After cycle 1, 25 (62.5%) participants achieved a complete response (CR), with 20/25 becoming minimal residual disease (MRD)-negative. Participants with favorable genetic profiles had the highest CR rate (7/8; 87.5%), while those with adverse genetic alterations including FLT3, TP53, and WT1 mutations, demonstrated variable responses. Among participants refractory to prior therapy, 6/11 (54.5%) achieved CR, with one additional partial response. Of four primary refractory participants, one achieved CR and another a partial response. The most common grade 3-4 adverse events included febrile neutropenia (32.8%), infections (20.9%), and gastrointestinal disorders (16.4%). A 50% reduction in peripheral blast count during venetoclax monotherapy was associated with a higher likelihood of CR with or without count recovery at the end of cycle 1 (18/23 [78.3%] vs 2/8 [25.0%], p=0.022). One treatment-related death occurred due to colitis and sepsis. BH3 profiling identified BCL-2 dependence in 23 patients but did not strongly predict response differences (p=0.266). Conclusion: Venetoclax combined with high-dose cytarabine-containing conventional chemotherapy was well-tolerated and active against pediatric relapse/refractory AML. A good response to venetoclax monotherapy therapy predicted complete response to combination therapy. These findings support further investigation of venetoclax-based regimens in this patient population, including the ongoing randomized pediatric phase 3 APAL2020D trial, which uses a backbone containing high-dose cytarabine and fludarabine.