Clearance of Very Low Levels of Measurable Residual Disease with Blinatumomab Significantly Improves Outcomes in B-Cell Acute Lymphoblastic Leukemia

Background Persistent or recurrent measurable residual disease (MRD) is the most important predictor of relapse and survival in B-cell acute lymphoblastic leukemia (B-ALL). Blinatumomab improves overall survival when used as consolidation in the frontline setting, irrespective of MRD status. Conventional methods of MRD detection such as multiparameter flow cytometry (MFC) and quantitative PCR generally have a sensitivity of ~10-4, or 0.01%, and are insufficient to identify very low levels of MRD, whereas high-throughput next-generation sequencing (NGS)-based MRD assays can detect MRD at a sensitivity of 10-6. The efficacy of blinatumomab in patients (pts) with low-level MRD detected by NGS and the depth of responses obtained with blinatumomab are unknown. Methods This is a retrospective analysis of pts with B-ALL in complete remission who received blinatumomab for any level of MRD at our institution. Blinatumomab was given as monotherapy in pts with Philadelphia chromosome-negative (Ph-) ALL but was combined with a BCR::ABL1 TKI in all pts with Ph+ B-ALL. ClonoSEQ® (Adaptive Biotechnologies) which uses NGS assessment of B-cell and T-cell receptor gene rearrangements was used to quantitate MRD response in both retrospective banked and prospective clinical samples. Pts who converted from NGS MRD+ to MRD- with blinatumomab were considered responders. Pts who remained NGS MRD+ or who relapsed while receiving blinatumomab were considered non-responders. Results Between July 2015 and October 2023, 42 pts were treated with blinatumomab for MRD-positive B-ALL and were evaluable for NGS MRD response. The median age was 38.5 (range, 19-70). 31 pts (74%) were Ph- and 11 pts (26%) were Ph+. 33 pts (79%) were in first remission (CR1), and 9 pts (21%) were in CR2+. 22 pts (52%) had high-risk cytomolecular features (defined as complex, low hypodiploidy, or near triploidy cytogenetics, KMT2Ar, Ph-like ALL, or TP53 mutation). Overall, 41% (17/42 pts) achieved NGS MRD negativity following blinatumomab. The NGS MRD response rate was 31% (10/31 pts) in Ph- disease and 64% (7/11 pts) in Ph+ disease. The response rate for pts who were also MRD positive by MFC and/or PCR was 32% (9/28 pts) vs 57% (8/14 pts) in pts who had low-level MRD only detected by NGS (P=0.2). NGS MRD responses were only observed in CR1; the response rate in CR1 was 52% (17/33 pts) vs 0% (0/9 pts) in CR2+ (P=0.006). The NGS MRD response rate was higher in pts without high-risk features (60%; 12/20 pts) vs those with high-risk features (23%; 5/22 pts) (P=0.03). Clearance of NGS MRD with blinatumomab was associated with excellent long-term outcomes. The 2-year relapse-free survival (RFS) and overall survival (OS) rates in NGS MRD responders vs. non-responders were 71% vs 25% (P=0.001) and 100% vs 46% (P<0.001), respectively. Three relapses occurred among responders; 1 had CNS-only relapse (Ph+ disease) and 2 had marrow relapse (both Ph-like). In CR1, the 2-year RFS rate of responders vs non-responders was 71% vs 40% (P=0.04), respectively. The 2-year RFS rate of Ph- responders vs non-responders was 63% vs 33% (P=0.02), and the 2-year RFS rate of Ph+ responders vs non-responders was 86% vs 0% (P=0.03), respectively. Among responders with high-risk features, the 2-year RFS rate was 50% vs 35% (P=0.2), and for responders without high-risk features, the 2-year RFS rate was 89% vs 0% (P=0.002), respectively. Among responders who had MRD detected only by NGS (i.e. negative by MFC and/or PCR) prior to the start of blinatumomab, the 2-year RFS rate was 70% vs 50% in those with MRD detectable by NGS and by MFC and/or PCR (P=0.2), respectively. In a landmark analysis, non-responders who underwent allogeneic stem cell transplant (ASCT) consolidation (n=8) had better outcomes than those who did not (n=14), with 2-year RFS rate 63% vs 10% (P=0.02), respectively, suggesting that ASCT may salvage some pts who remain NGS MRD positive after blinatumomab. Conclusion To our knowledge, this is the largest analysis evaluating rates of NGS MRD responses with blinatumomab. Deep responses to blinatumomab as assessed by NGS MRD is associated with superior survival outcomes across clinical contexts and identifies pts who have excellent long-term outcomes. Non-responders to blinatumomab have poor outcomes (2-year RFS: 25%) but may be salvaged by ASCT. The relatively low rate of NGS MRD negativity with blinatumomab monotherapy (31% in Ph- B-ALL) highlights the need for combination therapies in B-ALL.