Résumé
CAALL-F01 is a French prospective multicentric cohort study focused on children and adolescents, aged 1-17 years, with acute lymphoblastic leukemia (ALL). Patients with BCP-ALL were initially stratified based on age, white blood cell count, central nervous system (CNS) and/or testis involvement, genetics, and response to a prednisone prephase (PP). This stratification defined three risk groups: standard (B-SR), medium (B-MR), and high risk (B-HR). T-cell ALL patients were stratified into two groups, T-SR and T-HR, depending on CNS status and response to PP.
As asparaginase is a major drug for childhood ALL and the pegylated form, pegaspargase (PEG), was not registered in the EU, a two-parallel-arm randomized clinical trial (RCT) was conceived and embedded in the cohort study. Randomization occurred before the first PEG infusion (D12), in the B-SR, B-MR, and T-SR groups. Patients were allocated (1:1) to either receive one infusion of PEG 2500 IU/m² on D12 (Arm A) or two infusions of 1250 IU/m² on D12 and D26 (Arm B) during induction. The dose assigned at randomization was used in subsequent treatment phases. Patients in the B-HR and T-HR groups received two infusions of PEG 2500 IU/m² during induction and 2500 IU/m² per infusion afterwards.
The study had two primary objectives: 1) to assess the superiority of the fractionated scheme in terms of pharmacokinetics, 2) to assess the equivalence in tolerance of the two schemes. The co-primary endpoints were: 1) the incidence of adequate (>100 IU/L) asparaginase activity (AA) at the end of induction (D33 ± 1 day), 2) the incidence of severe (CTCAE Grade ≥ 3) asparaginase-related toxicities (CNS thrombosis, pancreatitis, severe allergy/anaphylaxis, and bilirubin elevation) from D12 of induction to D49.
Results: From September 2016 to February 2022, 2032 patients (mean age, 5.4 years, 57% male) were included. Among these, 1624 patients in the B-SR (942 pts), B-MR (518 pts), or T-SR (164 pts) groups were randomly allocated to Arm A (814 pts) or Arm B (810 pts). The 2 arms were well balanced for baseline characteristics and prognostic features. The median follow-up was 44.6 months. The D33 AA was evaluable in 1475 (86.7%) randomized patients. The two-infusion arm was superior to the one-infusion arm, as D33 AA was ≥ 100 U/L in 87.9% of the patients in Arm A (653/743) versus 96.6% in Arm B (707/732) (p<.0001). The cumulative incidence of any of the four targeted toxicities was 15.9% in Arm A (129/814 pts) versus 14.1% in Arm B (114/810 pts) (difference, 1.78%, 95% confidence interval (CI) -1.70 to +5.26%); as the equivalence margin was set to 4%, equivalence was not demonstrated. Minimal residual disease levels assessed by clone-specific IG-TCR PCR at the end of induction and consolidation did not differ between the 2 arms, nor did the 48-month event-free survival (EFS, 95.5% CI 95 94.7-97.4 vs 97% CI 95.78-98.12). Post hoc analyses revealed that 48-month EFS in Arms A and B were not significantly different for patients with B-ALL (94.0%, 95CI 92.0-95.9 for Arm A vs 95.8%, 95CI 94.2-97.4 for Arm B, HR= 0.78, 95CI 0.51-1.20, p=0.26), nor for patients with T-cell ALL (88.0%, 95CI 81.2-95.2 for ArmA vs 87.2%, 95CI 80.1-94.9 for Arm B, HR=1.16, 96CI 0.49-2.72; p=0.74).
Conclusion: For non-high-risk ALL, representing 80% of the study population, a fractionated scheme of pegaspargase (2 doses of 1250 IU/m²) is at least as well tolerated as a one-infusion scheme (1 dose of 2500 IU/m²) and more effective in terms of maintaining adequate AA throughout induction. However, at the current MFU of 44 months, this did not translate into superior long-term outcomes.
Strullu:Clinigen: Membership on an entity's Board of Directors or advisory committees; Servier: Other: symposium. Boissel:Pfizer: Other; Sanofi: Other; Adveysa: Other; Amgen: Other. Baruchel:Jazz: Membership on an entity's Board of Directors or advisory committees; Wugen: Membership on an entity's Board of Directors or advisory committees; Astra Zeneca: Research Funding; Servier: Other: TRavel Grant, Symposium, Research Funding.