Nelarabine, intensive L-asparaginase, and protracted intrathecal therapy for newly diagnosed T-cell acute lymphoblastic leukaemia in children and young adults (ALL-T11): a nationwide, multicenter, phase 2 trial including randomisation in the very high-risk group.

BACKGROUND: T-cell acute lymphoblastic leukaemia has distinct biological characteristics and a poorer prognosis than B-cell precursor acute lymphoblastic leukaemia. This trial aimed to reduce the rate of radiation and haematopoietic stem-cell transplantation (HSCT) while improving outcomes by adding nelarabine, intensified L-asparaginase, and protracted intrathecal therapy in the Berlin-Frankfurt-Münster (BFM)-type treatment.

METHODS: In this nationwide, multicenter, phase 2 trial, we enrolled patients with newly diagnosed T-cell acute lymphoblastic leukaemia (age <25 years at diagnosis) conducted by Japan Children's Cancer Group and Japan Adult Leukemia Study Group. Patients were stratified into standard-risk, high-risk, and very-high-risk groups according to prednisolone response, CNS status, and end-of-consolidation minimal residual disease. We used the Associazione Italiana di Ematologia Oncologia Pediatrica (AIEOP)-BFM-ALL 2000-backbone chemotherapy. Nelarabine (650 mg/m2 per day for 5 days) was given to high-risk and very high-risk patients. All patients received, until the measurement of end-of-consolidation minimal residual disease, an identical therapy schedule, which included the prednisolone pre-phase remission induction therapy with dexamethasone (10 mg/m2 per day, for 3 weeks [for patients <10 years] or for 2 weeks including a 7-day off interval [for patients ≥10 years]) instead of prednisolone, and consolidation therapy added with Escherichia coli-derived L-asparaginase. On the basis of the stratification, patients received different intensities of treatment; L-asparaginase-intensified standard BFM-type therapy for standard risk and nelarabine-added high risk BFM-type therapy for high risk. In the very high-risk group, patients were randomly assigned (1:1) to group A (BFM-based block therapy) and group B (another block therapy, including high-dose dexamethasone) stratified by hospital, age (≥18 years or <18 years), and end-of-induction bone marrow blast percentage of M1 (<5%) or M2 (≥5%, <25%)+M3 (≥25%). Cranial radiotherapy was limited to patients with overt CNS disease at diagnosis (CNS3; >5 white blood cells per μL with blasts) and patients with no evidence of CNS disease received protracted triple intrathecal therapy. Only very high-risk patients were scheduled to receive HSCT. The primary endpoint was 3-year event-free survival for the entire cohort and the proportion of patients with disappearance of minimal residual disease between randomly assigned groups A and B in the very high-risk group. Secondary endpoints were overall survival, remission induction rate, and occurrence of adverse events. 3 years after the completion of patient accrual, a primary efficacy analysis was performed in the full analysis set and the per-protocol set. This study is registered with the Japan Registry of Clinical Trials, jRCTs041180145.

FINDINGS: Between Dec 1, 2011, and Nov 30, 2017, of 349 eligible patients (median age 9 years [IQR 6-13]), 238 (68%) were male, and 28 (8%) patients had CNS3 status. 168 (48%) patients were stratified as standard risk, 103 (30%) as high risk, 39 (11%) as very high risk, and 39 (11%) as no risk (patients who had off protocol treatment before risk assessment. The composite complete remission (complete remission plus complete remission in suppression) rate after remission induction therapy was 89% (298 of 335 patients). HSCT was performed in 35 (10%) of 333 patients. With a median follow-up of 5·2 years (IQR 3·6-6·7), 3-year event-free survival was 86·4% (95% CI 82·3-89·7%) and 3-year overall survival was 91·3% (87·7-93·8%). The proportion of minimal residual disease disappearance was 0·86 (12 of 14 patients; 95% CI 0·57-0·98) in group A and 0·50 (6 of 12 patients, 0·21-0·79) in group B. Grade 3 peripheral motor neuropathy was seen in 11 (3%) of 349 patients and sensory neuropathy was seen in 6 (2%) patients. The most common grade 3 or worse adverse event was febrile neutropenia (294 [84%] of 349 patients). Treatment-related death occurred in three patients due to sepsis, gastric perforation, or intracranial haemorrhage during remission induction.

INTERPRETATION: The ALL-T11 protocol produced encouraging outcomes with acceptable toxicities despite limited cranial radiotherapy and HSCT use.

FUNDING: Ministry of Health, Labor and Welfare of Japan, and Japan Agency for Medical Research and Development.

TRANSLATION: For the Japanese translation of the abstract see Supplementary Materials section.