Uncovering Therapeutic Resistance in BPDCN through Single-Cell Transcriptome Profiling

Background:

Blastic plasmacytoid dendritic cell neoplasm (BPDCN) is a rare, aggressive hematologic malignancy arising from plasmacytoid dendritic cells. The most commonly affected sites include the skin (65%), bone marrow (51%), and blood (45%). Tumor cells typically exhibit immunophenotypes of CD4+, CD45+, CD56+, CD123+, HLA-DR+, and TCL-1+. Despite initial treatment responsiveness, BPDCN frequently relapses, posing significant challenges in patient management. Although Next-Generation Sequencing (NGS) and genome-wide DNA methylation analysis have elucidated the molecular characteristics of BPDCN at the genetic and epigenetic levels, understanding the mechanisms of therapeutic resistance remains elusive. Comprehensive molecular studies are urgently needed. Single-cell RNA sequencing (scRNA-seq) provides unprecedented resolution to dissect cellular ecosystems, revealing insights into cellular heterogeneity, clonal evolution, and treatment responses.

Methods:

This study performed single-cell analysis on bone marrow and skin samples from BPDCN patients, collected pre- and post-treatment. Using GEXSCOPE platforms, scRNA-seq generated high-resolution transcriptomic data. Data preprocessing involved CeleScope for alignment and quantification, followed by Scanpy for data integration, dimensionality reduction, clustering, and differential expression analysis. Pathway enrichment analysis was conducted with Gene Ontology and Kyoto Encyclopedia of Genes and Genomes. Based on single-cell expression profiles and cell marker gene expression, we employed automated annotation combined with manual review to assign identity labels to each cell. Due to the complexity and diversity of cell types, primary annotation for major cell types was followed by secondary subtype clustering and annotation to construct and present a detailed cellular atlas.

Results:

A total of 6 patient samples were analyzed, including 2 from skin tissue and 4 from bone marrow. Bone marrow samples included 1 from an untreated patient at initial diagnosis, 1 from a patient after second-line therapy, 1 after third-line therapy, and 1 after multiple lines of therapy. This project constructed a cellular atlas of 51,611 cells, categorized into 13 cell types, including Cancer cells, Keratinocytes, Sweat gland cells, Granulocyte-monocyte progenitor cells (GMP), Erythroid progenitor cells (ProEryth), Megakaryocytic progenitor cells (ProMek), Neutrophil progenitor cells (ProNeutrophil), Endothelial cells (ECs), Fibroblasts, Plasma cells (PlasmaCells), T and NK cells (TandNK), Neutrophils, and Mononuclear phagocytes (MPs). Due to the significant impact of tumor microenvironment differences, we separately analyzed the subtypes of skin and bone marrow samples. Skin samples (5,190 cells) were divided into seven subtypes, and bone marrow samples (46,421 cells) into nine subtypes. Top 10 differentially expressed genes in BPDCN skin samples by log fold changes were AFF3, AC109466.1, NCAM1, IRF8, ARPP21, LINC01374, GNLY, RBFOX1, AC073091.3, CNTN4. In bone marrow samples, they were TCL1A, IRF8, IGLL1, CD1E, ITM2C, STMN1, CD74, SOX4, AL138899.1, BLNK. Further analysis based on treatment conditions identified significant differentially expressed genes (DEGs) and pathway alterations associated with therapeutic resistance. Clonal evolution analysis revealed the emergence and expansion of resistant subclones during treatment. New therapeutic targets were identified, highlighting potential strategies to overcome resistance and enhance treatment efficacy. Subsequent studies will validate the findings through bulk RNA sequencing and qPCR on additional patient samples and perform functional assays on BPDCN cell lines or primary cells. Additionally, in vitro drug sensitivity screening tests were conducted on tumor cells from 2 patients, with clinical application further validating the efficacy.

Conclusion:

Our study elucidates the molecular mechanisms of therapeutic resistance in BPDCN, offering critical insights into clonal evolution and transcriptomic alterations post-treatment. These findings pave the way for developing targeted interventions, potentially improving clinical outcomes for BPDCN patients. The application of scRNA-seq in this context underscores its utility in understanding cancer resistance mechanisms, guiding future therapeutic advancements.

No relevant conflicts of interest to declare.