Blinatumomab Added to Chemotherapy Improves Disease-Free Survival in Newly Diagnosed NCI Standard Risk Pediatric B-Acute Lymphoblastic Leukemia: Results from the Randomized Children’s Oncology Group Study AALL1731

Introduction: Acute lymphoblastic leukemia (ALL) is the most common childhood cancer. Children with newly-diagnosed, National Cancer Institute (NCI) standard risk (SR) B-ALL have high survival rates when treated with traditional chemotherapy (chemo), yet some relapse and die. Relapsed ALL is a leading cause of pediatric cancer mortality and about half of relapses occur in SR B-ALL. AALL1731 (NCT03914625) is a phase 3 randomized trial to determine if 2 non-sequential cycles of the bi-specific T-cell engager blinatumomab (15 mg/m²/day IV continuous x 28 days) added to chemo improves disease-free survival (DFS) in children with NCI SR B-ALL with average or higher relapse risk.

Methods: AALL1731 enrolled newly diagnosed NCI SR (age >1 and <10 years with initial white blood cell count (WBC) <50,000/uL) B-ALL patients (pts), BCR::ABL1 negative, without testicular or CNS3 disease. After a 3-drug induction, pts were assigned to 3 risk groups based on tumor genetics, CNS status, and multiparameter flow cytometry (mpFC) defined minimal residual disease (MRD) at induction day 8 in peripheral blood (PB) and end of induction (EOI) in bone marrow (BM). Pts with favorable cytogenetics [ETV6::RUNX1 or double trisomies of chromosomes 4 and 10 (DT)], day 8 PB mpFC MRD <1% and EOI BM mpFC MRD <0.01% were categorized as SR-Favorable and non-randomly received chemo alone given known outstanding outcomes. Pts with unfavorable cytogenetics (iAMP21, KMT2A rearrangement, t(17;19), hypodiploidy), EOI mpFC MRD ≥0.1% for DT and ≥0.01% all others, or neutral cytogenetics with CNS2 status were categorized as SR-High. All others were considered SR-Average (Avg).

SR-Avg pts were further stratified based on EOI BM high-throughput sequencing of immunoglobulin loci (HTS) MRD. Those with undetectable EOI HTS MRD were non-randomly assigned to standard-intensity chemo alone (Arm A); all others were randomized to Arm A or standard-intensity chemo plus 2 cycles of blinatumomab (Arm B).

Post induction, SR-High pts received augmented BFM-based chemo. SR-High pts with end consolidation BM mpFC MRD <0.1% were randomized to chemo (Arm C) or chemo plus 2 cycles of blinatumomab (Arm D). Blinatumomab cycles were inserted before and after interim maintenance I. Planned accrual included 2245 pts with a minimum follow-up (FU) of 3 years.

Results: Accrual began June 28, 2019. At the first planned interim efficacy analysis (data cutoff June 30, 2024), 1440 (63%) of the 2245 SR-Avg/SR-High eligible and evaluable pts had been randomized. Median age was 4.3 years [interquartile range (IQR) 2.8-6.4]; 52.6% were boys, 26% were Hispanic and 5% were non-Hispanic Black. Median FU was 2.5 years (IQR=1.6-3.2). 722 pts were randomized to control Arms A (418) and C (304), and 718 to blinatumomab Arms B (417) and D (301). In intent-to-treat analyses, the 3-year DFS (± standard error) was 96.0±1.2% for pts randomized to blinatumomab arms vs 87.9±2.1% for those randomized to control arms. Using a stratified log-rank test, adding blinatumomab significantly improved DFS [Hazard Ratio (HR) 0.39, 95% confidence interval (CI) 0.24-0.64, 1-sided p<0.0001], substantially exceeding pre-specified interim efficacy stopping criteria and prompting the COG data safety and monitoring committee to recommend early termination of randomization.

Among SR-Avg patients, 3-year DFS for Arm B (blinatumomab) was 97.5±1.3% vs 90.2±2.3% for Arm A (control) (HR 0.33, 95%CI 0.15-0.69). For SR-High patients, 3-year DFS was 94.1±2.5% for Arm D (blinatumomab) vs 84.8±3.8% for Arm C (control) (HR 0.45, 95%CI 0.24-0.85). Overall, there were 6 deaths in remission, all among SR-High pts (Arm C=2, Arm D=4), none during blinatumomab cycles. Of 56 relapses on control arms, 10 were isolated CNS (iCNS), 34 isolated BM (iBM) and 5 combined CNS/BM. Of 19 relapses on blinatumomab arms, 9 were iCNS, 9 iBM, and 1 combined. Blinatumomab was well tolerated, with 0.3% of first courses associated with Grade 3+ cytokine release syndrome and 0.7% with seizures.

Conclusions: This randomized clinical trial definitively establishes that adding blinatumomab to chemo significantly improves DFS in newly diagnosed childhood SR B-ALL, both of average and higher risk, resulting in outcomes similar to those previously achieved in only the most favorable risk subsets. The addition of blinatumomab represents a major breakthrough and is a new treatment standard, with implications for children with newly-diagnosed B-ALL.