Blinatumomab for minimal residual disease in adults with B-cell precursor acute lymphoblastic leukemia

Abstract

Approximately 30% to 50% of adults with acute lymphoblastic leukemia (ALL) in hematologic complete remission after multiagent therapy exhibit minimal residual disease (MRD) by reverse transcriptase-polymerase chain reaction or flow cytometry. MRD is the strongest predictor of relapse in ALL. In this open-label, single-arm study, adults with B-cell precursor ALL in hematologic complete remission with MRD (≥10^-3) received blinatumomab 15 µg/m² per day by continuous IV infusion for up to 4 cycles. Patients could undergo allogeneic hematopoietic stem-cell transplantation any time after cycle 1. The primary end point was complete MRD response status after 1 cycle of blinatumomab. One hundred sixteen patients received blinatumomab. Eighty-eight (78%) of 113 evaluable patients achieved a complete MRD response. In the subgroup of 110 patients with Ph-negative ALL in hematologic remission, the Kaplan-Meier estimate of relapse-free survival (RFS) at 18 months was 54%. Median overall survival (OS) was 36.5 months. In landmark analyses, complete MRD responders had longer RFS (23.6 vs 5.7 months; P = .002) and OS (38.9 vs 12.5 months; P = .002) compared with MRD nonresponders. Adverse events were consistent with previous studies of blinatumomab. Twelve (10%) and 3 patients (3%) had grade 3 or 4 neurologic events, respectively. Four patients (3%) had cytokine release syndrome grade 1, n = 2; grade 3, n = 2), all during cycle 1. After treatment with blinatumomab in a population of patients with MRD-positive B-cell precursor ALL, a majority achieved a complete MRD response, which was associated with significantly longer RFS and OS compared with MRD nonresponders. This study is registered at www.clinicaltrials.gov as #NCT01207388.

Figure 1.

Disposition of patients in the study. *Reasons for not meeting eligibility criteria included: MRD level lower than the required ≥10⁻³ (therefore, inclusion criterion not fulfilled because disease burden too low; n = 48); not in hematologic CR (ie, overt relapse; n = 31); technical (n = 5); central nervous system relapse (n = 2); active infection (n = 2); alternative therapy (n = 2); neurologic disorder (n = 2); CD19⁻ (n = 1); hepatic disorder (n = 1); and consent withdrawn (n = 1). †Reasons for discontinuation (n = 33) included: adverse event (n = 20 [17.2%]), disease relapse (n = 10 [8.6%]), physician decision (n = 2 [1.7%]), and other (n = 1 [0.9%]). DOR, duration of hematologic remission; LLOQ, lower limit of quantitation; RFS, relapse-free survival.

Figure 2.

Complete MRD response after cycle 1 by clinical characteristics at baseline and conduct of therapy in cycle 1 (primary end point efficacy set). Three (75%) of 4 patients (95% CI, 19% to 99%) with Ph⁺ ALL and 2 (50%) of 4 patients (95% CI, 7% to 93%) with t(4;11) and/or MLL-AF4⁺ disease had a complete MRD response during cycle 1. MRD complete response rates were similar for patients with or without treatment interruptions during cycle 1. CR1, first CR; CR2/3, second or third CR.

Figure 3.

RFS and OS among Ph-positive patients in hematologic CR at start of treatment (key secondary end point full analysis set). (A) RFS without censoring at allogeneic HSCT or postblinatumomab chemotherapy. Median follow-up, 29.9 months. (B) RFS by remission status at screening without censoring at allogeneic HSCT or postblinatumomab chemotherapy. Complete MRD response was defined as MRD negativity with minimum sensitivity of 10⁻⁴. (C) OS without censoring at allogeneic HSCT or postblinatumomab chemotherapy. Median follow-up, 30.0 months. (D) OS by complete MRD responder status in cycle 1 among evaluable patients (landmark analysis, excluding patients who were censored or had relapsed or died within 45 days of beginning treatment), without censoring at allogeneic HSCT or postblinatumomab chemotherapy. (E) RFS without censoring at allogeneic HSCT or postblinatumomab chemotherapy by complete MRD responder status in cycle 1 and salvage status among evaluable patients (landmark analysis, excluding patients who were censored or had relapsed or died within 45 days of beginning treatment). NR, not reached.