Maturing progenitor-like immunophenotypic cluster and megakaryocytic repopulating ability predict measurable residual disease in Acute Myeloid Leukemia

Abstract

Abstract Introduction: Regenerative bone marrow ability after myeloablative agents is directly correlated with Measurable Residual Disease [MRD] and quality of remission. Indeed, robust platelet count recovery after induction is independently associated with survival. Hemopoietic Stem Cells (HSC) may follow an alternative non-classical differentiation pathway in which megakaryocytes branch directly from HSC during marrow stress. Here, we examined whether the magnitude of megakaryocytic repopulating ability correlates with the Acute Myeloid Leukemia (AML) immunophenotypic differentiation arrest, as defined by multiparameter flow cytometry (MFC). Additionally, we also evaluated laboratory and genomic features to enhance prediction of Measurable Residual Disease (MRD) response. Methods: After IRB approval, 125 AML cases were examined for post-induction megakaryocyte repopulating capacity in high power field (HPF). Megakaryocytes in post-induction bone marrow specimens were scored on a 0-4 [0= absent; 1= rare; 2= decreased; 3= adequate; and 4=increased]. AML cases were identified according to reported arrested differentiation algorithms based on flow cytometry antigen expression (e.g., CD34, CD117, HLADR, CD33/CD13 etc.). HSC-like, Common Myeloid Progenitor (CMP-like), Multipotent (MPP-like), Granulocytic Macrophage Progenitor (GMP-like) were assigned to hemopoietic progenitor cell group (HPC like), whereas Monocytic and Granulocytic progenitors (MP and GP, respectively) were assigned to Maturing progenitor like subgroup. Descriptive statistics, correlation coefficients, and logistic regression models were used to assess predictors of MRD. All analysis were conducted in SAS. Results: Median age was 68.5 years (25-93). By the European Leukemia Network 2022 (ELN22) classification, 9.3%, 21.2% and 69.3% of patients were favorable, intermediate and adverse risk. Superior complete remission (CR) plus complete remission incomplete (CRi) was observed in 53.4% of maturing-like vs 46.5% of HPC-like AML subtypes, p=0.06. Interestingly, in maturing like vs HPC-like subtypes, statistically significant CR+CRi was restricted to AML cases younger than 60 years,14/19 (74%) vs 5/19 (26%), respectively p=0.01. Given improved chemotherapy efficacy in maturing like subtypes, we examined potential predictors for MRD response. MRD was positively correlated with LDH (r=0.31, p=0.0034) and CD34 expression at diagnosis (r=0.26, p=0.02) but negatively correlated with post-induction megakaryocytic repopulating ability (r= 0.36, p=0.0048). In maturing-like vs HPC-like AML subtypes, FLT3, RAS, NPM1 and P53 mutations were observed in 34/49 (79%) vs 9/43 (21%), p=0.0002; 15/18 (83%) vs 3/18 (17%), p=0.006; 21/22 (95%) vs 1/22 (5%), p=<.0001; 10/29 (34%) vs 19/29 (66%), p=0.01. In multivariate analysis, higher megakaryocytic repopulating ability was associated with lower MRD (HR=-5.8, CI 95% -8.9, -1.12, p=0.01) and higher CD34 expression with higher MRD (HR=0.15, CI 95% 0.02, 0.28, p=0.03), even after accounting for ELN22 risk and age (ANOVA, p=0.0021). Conclusions: Our findings suggest that post-induction megakaryocytic reconstitution may serve as useful indicator of lower MRD burden, particularly among younger AML patients with maturing-like AML subtypes. Adequate megakaryopoiesis following chemotherapy may indicate the capacity of these cells to circumvent conventional lineage checkpoints to enhance hemopoietic reconstitution. Additional studies investigating the correlation between post-induction megakaryopoiesis, bone marrow regenerative ability and leukemia eradication are needed. Single cell multiomics analysis could unveil important insight in AML patients achieving robust post-induction megakaryopoiesis. This could allow improvement in MRD monitoring.