A multimodal atlas for immunotherapeutic targeting of AML surface heterogeneity

Acute myeloid leukemia (AML) is a hematologic malignancy with high relapse rates and limited treatment options due to extensive intra-tumor heterogeneity across patients. To characterize this heterogeneity, we profiled matched bone marrow mononuclear cell (BMMC) samples from 26 patients with adult AML at diagnosis and relapse using the cellular indexing of transcriptome and epitope sequencing (CITE-seq) and quantitative flow cytometry. These data together represent a comprehensive multimodal and longitudinal single-cell resource that reveals the transcriptomic and immunophenotypic landscape of AML. Data integration of CITE-seq and flow cytometry surface antigen readouts enabled systematic quantitation of surface antigen co-expression across individual leukemic cells, providing a granular framework for the design of immunotherapeutic strategies to target heterogeneous AML. With this resource, we identified CD33, CLL-1, LAIR1, ITGA4, DEC-205, and CD244 as antigens that induced cytotoxicity in AML cell lines in vitro when co-targeted by antibody drug conjugates (ADCs) or chimeric antigen receptor T (CAR-T) cells, demonstrating the exploitation of AML heterogeneity for immunotherapeutic innovation.