MRD Response Predicts Survival in Blinatumomab-Treated Patients With Relapsed/Refractory ALL

In patients with relapsed/refractory B-cell precursor acute lymphocytic leukemia (ALL) who are treated with blinatumomab, achieving minimal residual disease (MRD) was associated with improved overall survival (OS) and relapse-free survival (RFS), according to a study published in Blood Advances.

“MRD response after initial standard chemotherapy is the most important prognostic factor in firstline treatment of ALL, but its prognostic significance is less clear in the treatment of relapse,” lead study author Nicola Gôkbuget, MD, of the University Hospital in Frankfurt, Germany, told ASH Clinical News. “In our study, we were able to demonstrate a significant correlation of MRD response with survival, suggesting that physicians can potentially make decisions based on MRD response, such as including additional targeted therapies.”

In this single-arm, open-label, phase II trial, Dr. Gôkbuget and investigators enrolled patients with Philadelphia chromosome–negative B-cell ALL that was refractory to first salvage therapy or recurred within a 12-month period of first remission or following allogeneic hematopoietic cell transplantation (alloHCT).

Participants were given daily continuous intravenous infusions of blinatumomab (9 µg on days 1-7, 28 µg/day in cycle 1, and a target dose of 28 µg thereafter). Each 4-week treatment cycle was followed by a 2-week period free from treatment. Per study protocol, those who achieved a complete response (CR) or CR with partial hematologic recovery (CRh) within the first 2 treatment cycles could receive up to 3 additional cycles.

Out of 225 patients, a total of 90 patients achieved CR or CRh. Approximately three-quarters of these patients (74.6%) had received at least 1 prior salvage therapy, and most were younger (between ages 18 and 35). All patients had either early relapse (within 12 months of diagnosis) or refractory disease.

Most patients (83.3%) experienced an MRD response during the first 2 treatment cycles (primary endpoint):

  • 64 patients (71.1%) achieved a complete MRD response (defined as no detectable patient-specific re-arrangements of immunoglobulin/T-cell receptor genes)
  • 11 patients (12.2%) achieved an MRD response (defined as low positive MRD, or detectable leukemic cells <10-4)

Also, participants who achieved CR appeared to have a higher MRD response rate than those with CRh (88.2% [n=60/68] vs. 68.2% [n=15/22]; p value not provided). “A majority of MRD-evaluable patients with CR achieved MRD response regardless of the number of previous lines of therapy, and the proportion of patients achieving MRD response was similar for patients with (78.6%) and without (85.5%) prior [alloHCT],” the authors reported.

To determine whether achieving MRD affected survival, the investigators then assessed OS and RFS, comparing median survival rates between MRD responders and MRD nonresponders. For both OS and RFS, survival was longer in patients with MRD response than those without MRD response, although no p values were reported (TABLE).

The authors noted that MRD response “was shown to be an important factor in maintaining remission in patients with [relapsed/refractory] ALL, irrespective of prior [transplant] or number of lines of prior therapy.”

Based on the finding that CR and MRD response rates were potentially lower in patients with ≥50% bone marrow blasts at baseline, Dr. Gôkbuget added that blinatumomab should be administered after a cytoreductive treatment in patients with high tumor burden.

“It is certainly interesting that patients with a blast-free marrow, which might be considered as ‘failure’ to respond to blinatumomab, achieved MRD responses as well,” she said. “This underlines the fact that future response assessment in ALL should combine cytology and MRD measurement.”

Prognostic indicators in ALL include and the number of lines of salvage therapy, Dr. Gôkbuget noted, as well as T cell expansion or the proportion of regulatory T cells. For those treated with blinatumomab or other new immunotherapies, however, there is limited information about the value of these prognostic indicators.

“MRD response, however, may help in times where more and different targeted therapies are fortunately becoming available for ALL,” she commented, “allowing clinicians to design sequential therapies for patients with lack of MRD response.”

The study is limited by the highly heterogenous patient population, as well as the open-label, single-arm design. Dr. Gôkbuget emphasized the need for “more standardized approaches and a follow-up treatment to avoid relapse in patients without a transplant option.” Identifying better predictors of response to blinatumomab, as well as mechanisms contributing to resistance, also is of great interest for future research, she concluded.

Several study authors report relationships with Amgen, which sponsored the trial.


Gôkbuget N, Kantarjian HM, Brüggemann M, et al. Molecular response with blinatumomab in relapsed/refractory B-cell precursor acute lymphoblastic leukemia. Blood Adv. 2019;3:3033-7.

This paper is a follow-up analysis from a phase II study that reported on the safety, response rate, and survival of patients with relapsed/refractory Philadelphia chromosome–negative ALL who received blinatumomab therapy.

Dr. Gôkbuget and authors evaluated the impact of MRD measurement as a prognosticator for survival in the responding patients. The authors demonstrated that the patients with complete or near complete MRD responses had significantly improved relapse-free and OS, compared with MRD nonresponders. For example, median RFS was 12.5 months among responders, and only 2.3 months for MRD nonresponders.

These findings, while perhaps not surprising, provide additional support for the importance of standardized, sensitive MRD measurements in refining prognosis for patients with ALL who achieve morphological remission following treatment. This report is particularly useful because it focuses on the impact of MRD following successful therapy for relapsed disease; most studies demonstrating the value of MRD measurements are conducted in patients with previously untreated disease.

The study, however, also raises some important questions. For example, while MRD responders had better outcomes than nonresponders, most still relapsed. Why? Was the timepoint for MRD measurement the optimal timepoint for prediction? The study only reported on MRD following 2 cycles of treatment. Would multiple timepoints have further refined the prognostic accuracy? Does this suggest that more sensitive MRD assays are needed to further refine RFS, particularly when helping to predict outcomes of patients with relapsed disease? Or that the clone that gave rise to relapse was one that was not detected by the clone-specific polymerase chain reaction assay used?

More sensitive, next-generation sequencing methods using MRD that are now available also can monitor multiple clones simultaneously and may help to answer some of these important questions about MRD prognostication.

Wendy Stock, MD
University of Chicago Medicine
Chicago, IL