A phase I study of venetoclax in combination with multiagent cytotoxic chemotherapy for children and adolescents with relapsed or refractory acute lymphoblastic leukemia or mixed phenotype acute leukemia: Results of the dose determination cohort

Abstract Background Relapse is the most common cause of treatment failure for children and adolescents with acute lymphoblastic leukemia (ALL) and remains a major clinical challenge with long-term survival rates between 30-50% (Rheingold et al J Clin Oncol 2019). Venetoclax (VEN), a BCL-2 antagonist, has demonstrated promising efficacy in both pediatric and adult ALL, but has not been widely studied in combination with intensive multiagent chemotherapy in ALL (Luskin et al Blood Adv 2025; Place et al Pediatr Blood Cancer 2025). Herein, we report results from an investigator-initiated, multicenter phase 1 study of VEN in combination with intensive, multiagent chemotherapy for children and adolescents with relapsed/refractory ALL or mixed phenotype acute leukemia (MPAL). Design/Methods DFCI 21-757 is an investigator-initiated phase 1 study (NCT05292664) of VEN combination therapy in high-risk hematologic malignancies. Cohort C is testing VEN in combination with multiagent chemotherapy (dexamethasone, vincristine, doxorubicin, and calaspargase pegol (Cal-peg)) based on DFCI 16-001 high-risk Induction IA (Vrooman et al Blood 2024) in patients (pts) 1-21 years of age with relapsed/refractory ALL or MPAL. Part I is a dose finding cohort that follows a rolling-6 design with a primary objective to describe safety and tolerability of the regimen and to identify the maximally tolerated dose (MTD) and recommended phase 2 dose (RP2D) of VEN in combination with multiagent chemotherapy. The starting dose level (DL), DL1, of VEN is the 600 mg adult equivalent dose (AED). De-escalation to DL-1 (400 mg AED) occurs if a dose limiting toxicity (DLT) is seen in 2 or more pts. In the absence of excess toxicity, the dose may be escalated to DL2 (800 mg AED). The MTD is defined as the highest DL associated with 1 or fewer of 6 pts experiencing a DLT. A DLT is defined as: 1) grade (gr) 3 or greater non-hematologic adverse event (AE) that is possibly, probably or definitely related to VEN during the re-induction cycle or for 30-days following study treatment; 2) prolonged myelosuppression; or 3) inability to deliver ≥80% of planned doses due to study-related toxicity. Pts are monitored for AEs and graded for severity using the Common Terminology Criteria for Adverse Events (CTCAE) version 5.0. Complete remission (CR) is defined as <5% marrow blasts, no evidence of extramedullary leukemia, absolute phagocyte count (APC) of ≥1.0×103/μL, platelets of ≥75×103/μL and evidence of trilineage hematopoiesis. CR with incomplete count recovery (CRi) is defined as CR without adequate APC and/or platelet recovery, as defined above. Partial response (PR) is defined as M2 marrow and ≥50% decrease in marrow blast percentage, and without extramedullary disease. MRD is being assessed using next generation sequencing (NGS) by ClonoSEQ (Adaptive, Seattle WA). All statistical analyses are descriptive. Results Nine pts have enrolled onto Cohort C (8 B-ALL, 1 MPAL). Median age was 12.2 years (range 1.4-18.8 years). Four pts enrolled at DL1, and 2 DLTs (gr 4 duodenal hemorrhage and gr 3 gastritis) occurred, prompting de-escalation to DL-1. Of 5 pts enrolled at DL-1, there have been no DLTs. Serious adverse events (SAEs) not meeting DLT criteria but attributed to VEN included gr 3-4 infection (2 pts), gr 3 sepsis (1 pt), gr 3 mucositis (1 pt); 1 pt experienced gr 3 pancreatitis attributed to Cal-peg. The most common, clinically relevant gr 3 AEs included febrile neutropenia (6), hypertension (3), anorexia (2), hyperglycemia (2), mucositis (2), and sepsis (2). There were no recurrent gr 4 AEs and no gr 5 AEs. 7 pts achieved CR, 1 CRi, and 1 PR. NGS MRD was available for all pts. Of the 8 pts achieving CR/CRi, 5 had MRD below the limit of detection (<10-6), and 3 had MRD between 10-2 and 10-3. Conclusions Results from the dose finding cohort suggest that VEN at the 400 mg AED in combination with multiagent reinduction chemotherapy is tolerable and demonstrates promising efficacy in pts with relapsed ALL/MPAL. There was excess gastrointestinal toxicity at DL1, likely related to prolonged exposure to VEN in combination with dexamethasone, which has not been seen at DL-1. Importantly, overall observed toxicity is comparable to reported toxicity rates for multiagent reinduction regimens (Hogan et al Haematologica 2025), suggesting that VEN does not further increase toxicity of this regimen.