PF246 PERSONALIZED FOLLOW‐UP OF ACUTE MYELOID LEUKEMIA THROUGH MULTIPARAMETER FLOW CYTOMETRY: A PROOF OF CONCEPT

Abstract

Background:Standardization, machine learning techniques and comparison to normality are concepts that are changing the landscape of multiparameter flow cytometry data analysis in clinical hematology. We applied these concepts to the follow‐up of acute myeloid leukemia (AML), which remains challenging due to the heterogeneity of the disease.Aims:To evaluate a new flow cytometry data analysis approach for the follow‐up of measurable residual disease in AML.Methods:Based on retrospective flow cytometry data, we developed a strategy for personalized monitoring of AML. This strategy is an evolution of the “leukemia associated immunophenotype” or “LAIP”‐approach that has been used the last decades in specialized flow cytometry laboratories. It relies on new concepts, comparing simultaneously the patient's follow‐up and diagnosis results to control group samples. We named these concepts the “leukemic cloud” and the “abnormality ratio”. To evaluate this approach, flow cytometry data of 6 AML patients and 20 control patients was used. All samples were analyzed with the same 10‐parameter (FSC‐A, SSC‐A, CD45, CD34, CD117, CD13, CD33, CD19, CD3, HLA‐DR) FACS Canto II flow cytometry assay. The “abnormality ratio” results over time were compared with morphology and molecular biology results.Results:The “Abnormality Ratio” results over time were in line with the clinical evolution of the patients (see example on image below) and were overall more sensitive for detecting residual disease than morphology and molecular biology, excepted for NPM1‐mutated AML. The “Abnormality Ratio” did not consider phenotypic switches of the malignant cell population identified at diagnosis. However, the discrepancy between the “abnormality ratio” results and morphology or molecular biology results could be used to objectify these phenotypic switches.Summary/Conclusion:We developed a quasi‐fully automatable flow cytometry data analysis algorithm. It could be used as a basis for the development of “next generation” flow cytometry assays for monitoring of hematological malignancies such as AML. Prospective studies are needed to confirm the applicability and limitations of our concepts.image