Leukemic Stem Cell Measurable Residual Disease (LSC MRD) Status Is a Better Predictor of Relapse Than Multiparametric Flowcytometric Analysis in Acute Myeloid Leukemia Patients

Introduction: Relapse occurs in a substantial subset of multiparametric flow cytometry measurable residual disease (MFC MRD) negative acute myeloid leukemia (AML) patients. This relapse is thought to arise from leukemic stem cell-like populations (also called relapse-initiating cells), which are dormant, chemoresistant, and lack immunogenicity. Although the quantitation of these cells early into chemotherapy (LSC MRD) predicts outcomes, clinical algorithms have not incorporated this data into risk stratification schema. This creates a need for a more refined approach to monitor chemotherapy responses in adult AML. We compare LSC MRD against MFC MRD and demonstrate that LSC MRD is superior in predicting relapse in adult AML.

Methods: A total of 242 patients of adult (≥ 18 years) AML were accrued at a single center over 8 years (2015 – 2023). Patients were uniformly treated with "3 + 7" induction. Diagnostic samples were sequenced using a 135-gene capture-based myeloid panel and risk-stratified as per European LeukemiaNet 2022 (ELN-22) recommendations. MRD sampling for LSC and MFC was performed at the end of induction (PI) using either a 10-color 2-tube assay (n=143) or a 16-color single-tube assay (n=69). MRD analysis was restricted to patients in morphological remission (n = 207). Templates were standardized on MRD-negative BM samples of acute lymphoblastic leukemia patients. CD34+ CD38 dim to negative progenitors were analysed using a combination of different from normal (DfN) and leukemia-associated immunophenotype (LAIP) approaches. A cut-off value of 0.01% was used to define MFC MRD positivity. Depending upon the LSC and MFC MRD status, patients were divided into 4 groups: Group 1 – MFC+ /LSC+ (n = 59, 28.3%), Group 2 – MFC+/LSC- (n = 31, 15.1%), Group 3 – MFC-/LSC+ (n = 24, 11.5%), Group 4 – MFC-/LSC- (n = 93, 45.1%). Overall survival (OS) and relapse free survival (RFS) were used as the end points. Analysis of outcomes was performed using log-rank test and Cox proportional hazards regression model.

Results: Median age was 37.0 years (M:F; 2.11:1) with a median follow-up of 34.8 months. Median OS was not reached and median RFS was 35.3 months (95% CI 27.3 to 87.9). Of the 207 samples, 43.9% (n=91) of patients were MFC MRD positive. Similarly, 40.0% (n=83) samples were PI LSC MRD positive (0.0003%-4.3%). The most common immunophenotypic markers that the abnormal LSC populations expressed were CD45RA (50%), CD371 (41%), HLA DR (27.27%), and CD123 (19.3%). Patients were classified into favorable (n = 121, 58.4%), intermediate (n = 65, 31.4%), and adverse (n = 21, 10.1%) ELN-22 risk groups. As expected, MFC MRD and LSC MRD positivity were more prevalent in adverse and intermediate risk groups as compared to favorable risk. The presence of MFC MRD predicted an inferior RFS (HR = 1.7; 95% CI 1.0 to 2.9; p = 0.03) when compared to the MFC-MRD negative group. Patients that were LSC MRD positive had a significantly inferior OS [(HR = 2.2; 95% CI 1.2 to 3.9; (p = 0.002)] and RFS [(HR = 2.9; 95% CI 1.6 to 5.4; (p < 0.0001)] compared to the LSC MRD negative group. Within the ELN-22 subgroups, presence of LSC MRD (but not MFC MRD) had a significant impact on RFS in the favorable (HR = 2.9; 95% CI 1.4 to 6.4; p =0.0007) and intermediate (HR = 3.3; 95% CI 1.0 to 10.5; p = 0.006) ELN-22 risk groups. OS was found to be inferior (HR = 2.8; 95% CI 1.0 to 7.5; p = 0.00172) in LSC MRD positive cases only in the intermediate risk group. LSC MRD status did not impact outcomes in the adverse risk group. On combining the MRD results obtained with both modalities, patients that were positive by both techniques (MFC+/LSC+) had a significantly inferior outcome with respect to OS (HR = 2.4; 95% CI 1.3 to 4.5; p = 0.0054) and RFS (HR = 3.3; 95% CI 1.8 to 5.9; p = 0.0001) as compared to patients negative by both (MFC-/LSC-). On further analysis, MFC-/LSC+ group also had a significantly inferior RFS (HR = 2.8; 95% CI 1.1 to 7.0; p = 0.0026), compared to the MFC-/LSC- group. Groups with MFC+/LSC- and MFC-/LSC- had similar outcomes.

Conclusion: We conclude that patients with LSC MRD positivity had inferior outcomes irrespective of the MFC MRD status. Thus, LSC MRD status is an important parameter to evaluate during the PI time point and can be a better predictor of relapse compared to legacy methods.