NGS-MRD negativity in post-HSCT ALL spares unnecessary therapeutic interventions triggered by borderline qPCR results without an increase in relapse risk

Abstract

Monitoring of minimal residual disease (MRD) after hematopoietic stem cell transplantation (HSCT) in patients with acute lymphoblastic leukemia (ALL) is vital for timely therapeutic intervention planning. However, interpreting low-positive results from the current standard method, quantitative PCR (qPCR) of immunoglobulin and T-cell receptor gene rearrangements (IG/TR), poses challenges due to the risk of false positivity caused by non-specific amplification. We aimed to improve MRD detection specificity using the next-generation amplicon sequencing (NGS) of IG/TR rearrangements for better relapse prediction. In pediatric and young adult ALL patients undergoing sequential post-HSCT MRD monitoring, we prospectively re-tested positive non-quantifiable qPCR results with NGS-MRD using the EuroClonality-NGS approach. We were able to confirm 13 out of 47 (27.7%) qPCR positive results using the more specific NGS-MRD method. Out of 10 patients with at least one MRD positivity confirmed by NGS, six relapsed (60%) 1-3.7 months after testing. Among 25 patients with all NGS-MRD results negative, two relapses occurred (8%) after 5.1 and 12.1 months. One-year RFS was 40% versus 96% and 3-year OS was 33.3% versus 94.4% for the NGS-positive and NGS-negative groups, respectively. The difference was not attributable to a varying rate of therapeutic interventions. Six patients out of 14 who had immunosuppressive treatment tapered or received donor lymphocyte infusion in response to MRD positivity developed significant graft versus host disease, leading to one fatality. This underscores the importance of enhancing the post-HSCT relapse risk prediction accuracy through NGS-MRD testing to avoid unnecessary interventions.

Figure 1

Flow diagram of the study design. AYA, adolescents and young adults; BM, bone marrow; neg, negative; PnQ, positive non‐quantifiable; pos, positive; post‐HSCT, post‐transplant.

Figure 2

Relapse prediction by NGS‐MRD. (A) Relapse‐free survival; (B) Overall survival. NGS positive, patients with at least one positive NGS‐MRD result (n = 10); NGS negative, patients with all NGS‐MRD results negative (n = 25); with intervention, NGS negative patients who received an MRD‐based therapeutic intervention (n = 7); no intervention, NGS negative patients without an MRD‐based therapeutic intervention (n = 18); non‐relapse death censored in RFS.

Figure 3

GvHD following therapeutic interventions. Swimmer plot showing post‐transplant course of MRD monitoring, therapeutic interventions, GvHD, and events of patients with GvHD development or flare following IST tapering or DLI in response to detected MRD positivity. First, IG/TR marker is displayed above, and second IG/TR marker below, MRD by fusion genes at the center of the patient bar with labels to the right. Therapeutic interventions are specified next to symbols (triangles) above patient bars.