A shorter induction corticosteroid reduces acute lymphoblastic leukaemia induction deaths in the icicle-all-14 randomised trial

Abstract

Background and Aim - Treatment-related toxicity poses an obstacle to improving acute lymphoblastic leukaemia (ALL) outcomes. The ICiCLe-ALL-14 randomised clinical trial investigated the toxicity impact of a shortened induction corticosteroid course in young children with non-high-risk ALL. Method - Between October 2016 and August 2022, patients 1-10 years old with newly diagnosed standard-risk (SR) and intermediate-risk (IR) B-precursor ALL were `randomised to receive a long (28 days, continuous with 5-day taper; Long-Pred) or short (21 days, pulsed, no taper; Short-Pred) course of prednisolone 60 mg/m2/day during induction as part of the ICiCLe-ALL-14 trial. End-induction toxicity (deaths and Grades 3-4 NCI-CTCAE toxicities) was the primary endpoint. Treatment response (including minimum residual disease [MRD] levels), relapse, and survival outcomes represented secondary objectives. All analysis was by intention-to-treat. Results - Of 1,307 eligible patients, 1,246(95%) received randomised induction corticosteroid (623/arm, matched for conventional risk factors). Thirty (2·4%) died in induction, all sepsis-related. Significantly more deaths were recorded in the Long-Pred group: 22/623 (3·5%) versus 8/623 (1·3%) (p=0·0095). More deaths were similarly observed in Long-Pred SR [7(88%)/8] and IR [15(68%)/22] sub-groups. Using Cox multivariable regression, induction mortality was significantly lower with the Short-Pred regimen (HR[hazard ratio], 3·12; 95% CI[confidence interval]1·38-7·04; p=0·0062), especially in younger patients (HR decrease of 0·83/unit increase in age; 95%CI, 0·68-0·99; p=0·0474). Rates of Grades 3-4 toxicity, predominantly sepsis and drug-induced hypertension, were comparable between the randomised groups (Long-Pred:44% versus Short-Pred:45%). Rates of induction-remission and MRD response were similarly comparable. So were treatment-related deaths post-induction (overall, 4%) and relapses (overall, 20%). At a median of 30 months, the estimated 3-year event-free survival was similar between Short-Pred (72%, 95%CI 68-76%) and Long-Pred (73%, 95%CI 68-77%) groups. Conclusions - Reducing corticosteroid duration during induction therapy in young children with non-high-risk acute lymphoblastic leukaemia significantly decreases treatment-related deaths without compromising treatment response or relapse risk (Clinical Trials Registry India CTRI/2015/12/006434). Cited by (0)