Evaluating flow cytometry (FC) and next generation sequencing (NGS) for assessing minimal residual disease (MRD) and B-cell aplasia (BCA) in children, adolescents and young adults (CAYA) with B-cell acute lymphoblastic leukemia (B-ALL)

Abstract Background: Several biomarkers are utilized to monitor for B-ALL relapse. For minimal residual disease (MRD), these include flow cytometery (FC) and next-generation sequencing (NGS). NGS is increasingly being used due to higher sensitivity than FC. For patients receiving CAR T-cells, CD19+ peripheral blood (PB) FC-BCA, and less often, bone marrow (BM) FC-BCA are utilized to assess persistence. How these assays cross-compare with outcomes is unknown. Objective: To cross-compare BM NGS-MRD, BM FC-BCA and PB FC-BCA in CAYA patients with B-ALL who achieved a BM FC-MRDneg complete remission (CR) on D+28 following CAR T-cells. Additionally, we compare these to the use of NGS to assess BCA (NGS-BCA) in the BM. Secondary objectives included analysis of the association of these biomarkers with relapse risk and CAR T-cell persistence in the BM. Methods: A multi-institution retrospective study of patients who received investigational CD22 or CD19/22 CAR T-cell constructs at the National Cancer Institute or commercial CD19 CAR T-cells at Johns Hopkins Children's Center from 2/2013-4/2025 was conducted. All patients had disease restaging at approximately D+28 post CAR T-cell infusion and were included for analysis if they achieved an FC-MRDneg remission, defined as no disease detected at a level of >0.01% of mononuclear cells. PB BCA was defined as <10 B-cells/uL. Patients with Adaptive clonoSEQ® reports from bone marrow specimens at D+28 were assessed for BM NGS-MRD and BM NGS-BCA. Those with any NGS identified disease-associated clone >0 were considered as NGS-MRD+. BM BCA was defined as <1% B-cells. BM FC-BCA was based on the percentage of CD19+ B-cells in relationship to total cells or total lymphocytes. BM NGS-BCA was calculated by dividing total immunoglobulin heavy chain sequence(s) by total nucleated cells. CAR T-cell persistence was evaluated by FC in the BM for investigational constructs and was defined as >0% CAR T-cells. Results: There were 43 patients who were FC-MRDneg with concurrent PB FC-BCA at D+28. The median age at CAR infusion was 13.6 years (range 0.8-38.1). Twenty-four subjects (55.8%) were male. Thirty-two patients (74.4%) received an investigational CAR-T cell construct and 11 (25.6%) received tisagenlecleucel. Sixteen of these 43 patients (37.2%) were NGS-MRD+, of whom 14 (87.5%) had NGS-BCA. Across 27 patients with NGS-MRD negativity, 23 (85.2%) had NGS-BCA. Thus, 37 (86.0%) patients had NGS-BCA regardless of NGS-MRD status. Forty-two patients had BM FC-BCA assessed, of which 36 (85.7%) had FC-BCA. BM FC-BCA and NGS-BCA were fully concordant. Of the 6 patients who did not have BM FC-BCA or BM NGS-BCA, 4 had non-CNS extramedullary disease (EMD). Of the 29 patients with available CAR T-cell persistence data, 22 (75.9%) had detectable BM CAR of which 16 (72.7%) were also NGS-MRDneg. All patients with detectable CAR T-cells had both PB FC-BCA and BM FC-BCA. Eighteen patients (41.9%) relapsed, of which 9 (52.9%) were NGS-MRD+ and 13 (76.5%) had NGS-BCA at D+28. Out of 16 patients who were NGS-MRD+ at D+28, 9 (56.3%) relapsed. However, of the 7 NGS-MRD+ patients who did not relapse, six received a post-CAR consolidative stem cell transplant. Out of 27 who were NGS-MRDneg, 9 (33.3%) relapsed. Across 16 patients with CD19+ disease pre-CAR and who experienced relapse, 12 had available immunophenotypic analysis across whom 2 were confirmed to have CD19 negative (CD19-neg) relapse. Both patients were NGS-MRD+ at D+28; and only one had BM BCA at D+28. The median time to relapse for patients who were NGS-MRD+ at D+28 was 4.8 months (range 1.5-11.6). This compared to 9.6 months (range 4.5-20.9) for those who were NGS-MRDneg (p = 0.02). Of 36 patients with BM FC-BCA and BM NGS-BCA, 13 (36.1%) relapsed; of 6 patients without BM FC-BCA and BM NGS-BCA, 5 (83.3%) relapsed. Nine of 12 patients (75%) with non-CNS EMD relapsed. Conclusion: We identified several discrepancies amongst BM NGS-MRD status and BM BCA by either FC or NGS. Interestingly, >85% of patients who were NGS-MRD+ still had BCA, by either FC or NGS. Regardless of NGS-MRD status, most patients had NGS-BCA. BM FC-BCA and BM NGS-BCA were fully concordant. While additional analysis of post-CAR T-cell surveillance metrics is underway, these results raise awareness of discrepencies in disease detection methodologies. Further investigation utilizing these biomarkers in B-ALL surveillance is warranted.