Bortezomib and vorinostat in combination with mitoxantrone, dexamethasone, and pegasparaginase during induction and reinduction for infants with acute lymphoblastic leukaemia: a multicentre single-arm phase 1/2 study.

backgroundAcute lymphoblastic leukaemia in infants remains a therapeutic challenge. In preclinical studies we previously identified bortezomib and vorinostat as active agents against KMT2A rearranged (KMT2Ar) leukaemia. The aim of this study was to determine the tolerability of incorporating bortezomib and vorinostat into an acute lymphoblastic leukaemia-chemotherapy backbone for newly diagnosed infants with acute lymphoblastic leukaemia both with and without KMT2Ar.

methodsIn this single-arm phase 1/2 study conducted at 18 hospitals in the USA and Canada, we enrolled patients aged 1 year or younger at the time of diagnosis with newly diagnosed acute lymphoblastic leukaemia or undifferentiated leukaemia (with or without extramedullary disease) who had 25% or greater blasts in the bone marrow. Patients with T-cell acute lymphoblastic leukaemia were eligible, and patients with mixed-phenotype acute leukaemia were eligible provided the morphology and immunophenotype were predominantly lymphoid. We evaluated the incorporation of bortezomib (0·043 mg/kg intravenously once per day on days 1, 4, 8, 11, 15, and 18) and vorinostat into a chemotherapy backbone containing dexamethasone (5 mg/m

findingsFrom Jan 29, 2016, to Nov 17, 2021, 50 patients were enrolled to the study of whom 35 had KMT2Ar mutations. Median age was 171 days (IQR 100-276); patients were mostly White (31 [62%]), 26 (52%) patients were male and 24 (48%) were female. Median follow-up for alive patients was 4·9 years (IQR 3·6-6·0). No DLTs occurred in the first two vorinostat dose levels, and 44 patients received vorinostat at dose level 3 (DL3) in the dose expansion phase. Four patients treated at DL3 had protocol-defined induction DLTs and one patient treated at DL3 had a protocol defined reinduction DLT. The most frequently occurring grade 3-4 adverse events during induction (defined as occurring in 20% or more of participants) inclusive of all three dose levels included hypertension (24 [48%]), infection (22 [44%]), fever and neutropenia (21 [42%]), anorexia (15 [30%]), alanine aminotransferase elevation (10 [20%]), hypokalaemia (17 [34%]), hypocalcaemia (17 [34%]), and hypoalbuminemia (10 [20%]); the most frequent adverse event during reinduction was hypertriglyceridemia (N=10 [23%] of 44). Serious adverse events occurred in four patients treated at DL3 during induction (three episodes of bacterial sepsis and one induction death) and in one patient treated at DL3 during reinduction (bacterial sepsis secondary to ileal perforation). Four patients had treatment-related deaths, all infectious events (parainfluenza [n=1; during induction], cytomegalovirus [n=2; one in consolidation and one in maintenance], and Aspergillus fumigatus [n=1; in maintenance]).

interpretationAdverse events during induction and reinduction were consistent with those reported in previous infant acute lymphoblastic leukaemia studies. Infections remain a substantial cause of morbidity and mortality in this immunocompromised patient population and extended beyond the induction and reinduction chemotherapy phases.

fundingAmerican Lebanese Syrian Associated Charities, Gateway for Cancer Research, Lucile Packard Foundation for Children's Health, National Institutes of Health.