Blinatumomab binds to CD19 (B cells) and CD3 (T cells) to facilitate lysis of CD19-positive target B-lineage cells. Based on a single-arm phase 2 study, blinatumomab received approvals in the United States (accelerated) and European Union (conditional) for the treatment of Philadelphia-negative relapsed/refractory (r/r) B-cell precursor (BCP) acute lymphoblastic leukemia (ALL). This phase 3, randomized, open-label study investigated the effect of blinatumomab on overall survival (OS) compared with standard-of-care (SOC) chemotherapy in adult patients with r/r BCP-ALL.1
Patients with r/r BCP-ALL (if first relapse, within 1 year) were randomized 2:1 to receive blinatumomab or SOC chemotherapy (investigator’s choice of 1 of 4 defined regimens). Randomization was stratified by age, prior salvage therapy, and prior allogeneic stem-cell transplant (alloSCT). Blinatumomab was given in 6-week cycles of 4 weeks on (continuous infusion of 9 µg/day in week 1 of cycle 1, then 28 µg/day) and 2 weeks off. Dexamethasone was given predose to prevent cytokine release syndrome (CRS). Patients in remission after 2 induction cycles were eligible to continue therapy until relapse.
OS was the primary efficacy end point. Complete remission (CR) and combined CR or CR with partial or incomplete hematologic recovery (CR/CRh/CRi) were secondary efficacy end points. The primary analysis was scheduled to occur after 330 deaths had accrued. This prespecified interim analysis for an independent Data Monitoring Committee (DMC) occurred after 248 deaths (75%).
A total of 405 patients were randomized to blinatumomab (n = 271) or SOC (n = 134) and analyzed for efficacy. Baseline characteristics were balanced between treatment groups (blinatumomab, SOC): median age (37 years, 37 years); median bone marrow blasts (80%, 79%); prior salvage therapy (58%, 51%); and prior alloSCT (35%, 34%).
Based on the DMC analysis, median OS was 7.7 months for blinatumomab (95% confidence interval [CI], 5.6-9.6) and 4.0 months (95% CI, 2.9-5.3) for SOC (stratified log-rank test P = 0.012; hazard ratio, 0.71 [0.55-0.93]), which surpassed the prespecified boundary P value of 0.0183. Improvement in OS was consistent among subgroups based on age, prior salvage therapy, or prior alloSCT. Response rates were also higher for blinatumomab versus SOC, including CR (34% vs 18%; P <0.001) and CR/CRh/CRi (44% vs 25%; P <0.001).
A total of 376 patients received ≥1 doses of blinatumomab (N = 267) or SOC (N = 109; mostly FLAG ± anthracycline, followed by high- or intermediate-dose cytarabine-based, high-dose methotrexate-based, and clofarabine-based regimens) and were analyzed for safety. Adverse events in the blinatumomab arm were consistent with previous studies. Rates of grade 3+ neutropenia and infection were lower in the blinatumomab arm (52%, 58%) compared with SOC (34%, 38%). Grade 3+ neurologic events occurred at similar rates (9%, 8%).
TOWER is the first randomized study of an immunotherapy to demonstrate an OS benefit in this difficult-to-treat patient population.
