Blinatumomab consolidation for adult B-cell acute lymphoblastic leukemia in first and second complete remission

TO THE EDITOR:

Although the outcome of adult patients with B-cell precursor acute lymphoblastic leukemia (BCP-ALL) has dramatically improved in the last decades, patients with adverse genetic subtypes and/or persistence of minimal residual disease (MRD) are still at high risk of relapse. When treated with standard chemotherapy, adult patients with Philadelphia (Ph)^– BCP-ALL in first relapse have a dismal prognosis, with a median overall survival (OS) of ∼6 months and a 5-year OS rate of <20%.^1,2 

Blinatumomab is a CD3/CD19 bispecific T-cell engager that induces T-cell cytotoxicity against CD19⁺ B cells. The drug was first approved in relapsed/refractory (R/R) Ph^– BCP-ALL based on the results of the phase 3 TOWER study showing a higher overall response rate (44% vs 25%, P < .001) and a longer median OS (7.7 vs 4.0 months, P = .01) than those with standard of care.³ Blinatumomab was further approved for patients with BCP-ALL with persistent or reappearing MRD based on the results of the phase 2 BLAST study.⁴ 

More recently, 2 phase 3 studies in children, adolescents, and young adults with BCP-ALL in the first relapse suggested that blinatumomab used as consolidation after salvage therapy was more efficient and better tolerated than chemotherapy. Patients who received blinatumomab also showed a better MRD response and a greater likelihood of proceeding to allogeneic hematopoietic stem cell transplantation (alloHSCT) compared with chemotherapy alone.^5,6 Moreover, the use of blinatumomab after an attempt to reduce the leukemic burden rather than in overt relapse is now supported by many real-world studies or retrospective analyses of prospective trials showing the prognostic impact of bone marrow tumor infiltration before blinatumomab administration.^7-10 In the present real-world study, we aim at describing the outcome of adult patients who received blinatumomab in first or second complete remission (CR). As an internal comparator, we also reported on patients from the same institutions who received blinatumomab in overt relapse.

The study was conducted at Saint-Louis University Hospital in Paris and at Città della Salute e della Scienza University Hospital in Turin. One hundred fifteen patients with BCP-ALL consecutively treated between April 2012 and June 2021 were retrospectively included. We included patients who received commercial blinatumomab 1) in first CR (CR1), 2) in second CR (CR2), or 3) in overt relapse or refractory (R/R). Among the 115 patients included, 68 patients (59%) were treated in CR1, 31 patients (27%) in CR2 after chemotherapy-base salvage therapy usually combined with tyrosine kinase inhibitor (TKI) in Ph⁺ ALL (supplemental Table 1), and 16 (14%) patients in overt relapse or refractory. The median age was 37 years (range, 16-84), and 28 of 115 (24%) were Ph⁺. Patients in CR1 mostly received blinatumomab for MRD persistence (n = 59/68, 87%) or because of their inability to receive standard consolidation (n = 9/68, 13%, off-label use) because of chemotherapy toxicity (including 1 specifically related to asparaginase) in 8 cases and to low performance status in 1 case (Table 1). Patients in CR2 received blinatumomab for R/R disease and underwent a debulking strategy before blinatumomab. The number of blinatumomab cycles, along with the use of chemotherapy and/or alloHSCT after blinatumomab, was up to physician choice. Most patients with Ph⁺ ALL (26 of 28) received a combination of blinatumomab and TKI (14 ponatinib, 4 nilotinib, 4 imatinib, and 4 dasatinib). In most patients, MRD was assessed by immunoglobulin/T-cell receptor (IG/TR) clonospecific quantitative polymerase chain reaction (qPCR) (99/115, 86%). The other patients were monitored by BCR::ABL1 qPCR (8/115, 7%), flow cytometry (6/115, 5%), and KMT2A-r transcript qPCR (2/115, 2%). The study was authorized by the ethical committees of both institutions and conducted in accordance with the Declaration of Helsinki.

In the 3 subgroups, the median number of blinatumomab cycles given was 2 (range, 1-6). A cytokine release syndrome grade ≥3 was reported in 2 patients. An immune effector cell–associated neurotoxicity syndrome (ICANS) grade ≥3 was observed in 11 patients (supplemental Table 2). Nineteen patients (16.5%) had to temporarily interrupt blinatumomab because of adverse events, whereas in 4 patients (3.5%), blinatumomab was permanently discontinued (ICANS in 2 cases, ICANS and cytokine release syndrome in 1 case, and serious infection in 1 case). In R/R patients, a CR was reached after blinatumomab in 9 of 16 (56%), among whom 4 were bridged to alloHSCT in continuous CR (CCR). The median disease-free survival (DFS) and OS were 6.0 and 10.8 months, in line with the results of TOWER study (Figure 1).³ In the CR subgroups, a complete MRD response was achieved after 1 blinatumomab cycle in 83% of CR1 and 86% of CR2 patients (P = .99). Forty-six patients (42%) treated in CR (41% CR1, 45% CR2) were bridged to allo-HSCT in CCR after blinatumomab, with 68% of all patients who underwent transplantation, being in complete MRD response at the time of transplant. Twenty patients treated in CR relapsed, 12 from the CR1 cohort and 8 from the CR2 cohort. Eight patients had a CNS relapse (40%), among whom 7 patients had a combined relapse. After a median follow-up of 3.1 years (95% CI, 2.4-3.7), the 3-year cumulative incidence of relapse was 23% for CR1 and 26% for CR2 patients (P = 0.31), the 3-year DFS was 68% for CR1 and 67% for CR2 patients (P = 0.41), and the 3-year OS was 80% for CR1 and 71% for CR2 patients (P = 0.32) (Figure 1A,B). Among patients with Ph⁻ ALL aged <70 years who received blinatumomab in CR1 before any transplant, 25/47 (53%) were bridged to allo-HSCT in CCR. Among these 47 patients, 17 were alive in CCR after 3 years of follow-up, 7 without and 10 after allo-HSCT. A time-dependent analysis of allo-HSCT in these patients did not suggest a benefit of transplant neither for DFS (hazard ratio [HR] 2.9; 95% CI, 0.80-10.4, P = 0.11 by the Mantel-Byar test), nor for OS (HR 5.5; 95% CI, 0.82-37.6, P = 0.08 by the Mantel-Byar test) (supplemental Figure 3). Considering the 99 patients treated in CR, univariate analysis showed that MRD levels before blinatumomab and best MRD response after blinatumomab were significantly associated with DFS and OS (Figure 1C,D, supplemental Figures 1 and 2). Patient characteristics, disease-related features, and allo-HSCT performed in continuous complete remission (time-dependent analysis) did not predict DFS or OS in univariate analysis (supplemental Table 2). Both pre- and post-blinatumomab MRD levels retained significance in bivariate analysis for DFS and OS (Figure 1E).

Our observation underlines the efficacy of blinatumomab in consolidation after chemotherapy-based salvage, showing comparable outcomes between patients treated in CR2 and CR1. Similar proximity between CR1 and CR2+ patients’ outcome was already noted in the phase 2 BLAST study for patients with positive MRD.¹¹ Here, the median survival of CR1 and CR2+ patients with a MRD ≥10⁻³ before blinatumomab was 41.2 vs 23.1 months, respectively (P = .4). In CR2 patients, promising DFS and OS were observed as compared with those of historical cohorts of patients treated with chemotherapy alone or with blinatumomab in overt relapse.^1-3 Our results corroborate those reported in 2 phase 3 trials by Brown et al and Locatelli et al, performed in children, adolescents, and young adults in second CR of intermediate and high-risk Ph^– BCP-ALL. In the study by Brown et al, the 2-year DFS was 54% for the blinatumomab group vs 39% for the chemotherapy group (HR, 0.70; 95% CI, 0.47-1.03), and the 2-year OS was 71% vs 58%, respectively (HR, 0.62; 95% CI, 0.39-0.98).⁶ In the study by Locatelli et al, 2-year estimated event-free survival was 66% in the blinatumomab group and 27% in the chemotherapy group (HR, 0.33; 95% CI, 0.18-0.61), and OS was 85% and 70%, respectively (HR, 0.43; 95% CI, 0.18-1.01).⁵ Although these 2 randomized trials have led to important changes in pediatric clinical practice,¹² this study suggests that similar strategies using blinatumomab in consolidation and not in overt relapse should be considered in adults. Indeed, a second CR may be achieved in up to 60% to 70% of patients after standard chemotherapy for late relapse^1,2 or after inotuzumab ozogamicin.¹³ 

Moreover, our study suggests a strong benefit of consolidation with blinatumomab in patients who achieved a complete MRD response before exposure to blinatumomab. In our cohort, these patients were mostly CR1 patients who presented toxicity to prior chemotherapy or CR2 patients with favorable responses to salvage therapy. Such a benefit of consolidation with blinatumomab in CR1 MRD-negative patients was recently evidenced by the E1910 ECOG phase 3 study that showed a dramatic improvement of OS for patients exposed to blinatumomab (HR, 0.42; 95% CI, 0.24-0.75).¹⁴ It should be emphasized that the significant incidence of CNS relapses highlighted in our report underscores the critical necessity for adequate CNS prophylaxis among patients treated with blinatumomab.

Finally, our study underscores the significance of both pre- and post-blinatumomab MRD as predictors of patient outcomes. Although the role of alloHSCT is being strongly questioned because of the increased use of blinatumomab as a frontline treatment, it is important to consider high levels of pre-blinatumomab MRD and/or detectable post-blinatumomab MRD as cautionary indicators that can guide therapeutic decisions after blinatumomab treatment. These MRD assessments provide valuable insights to inform post-blinatumomab treatment strategies and optimize patient outcomes.

Contribution: I.U., E.L., and N.B. designed the research, analyzed the data, and wrote the manuscript; I.U., E.L., F.R., M.C., F.C., D.F., M.S., N.D., R.I., L.A., E.R., H.D., E.A., and N.B. managed the patients and provided clinical data; R.K., D.F., and E.C. performed disease characterization and MRD analyses; and all authors reviewed and approved the manuscript.

Conflict-of-interest disclosure: N.B. received honoraria and research funding from Amgen. The remaining authors declare no competing financial interests.

Correspondence: Nicolas Boissel, Hematology Adolescents and Young Adults, Saint-Louis Hospital, 1 Ave Claude Vellefaux, 75010 Paris, France; email: nicolas.boissel@aphp.fr.