Single-cell profiling uncovers PTPRG-driven stemness in malignant plasma cells and signatures of treatment failure in multiple myeloma.

Background: Multiple myeloma (MM) is characterized by extensive intratumoral heterogeneity and complex interactions within the bone marrow microenvironment, yet the cellular and molecular drivers of treatment resistance remain poorly defined. Protein tyrosine phosphatase receptor gamma (PTPRG) has emerged as a candidate tumor suppressor in various malignancies by antagonizing proliferative and survival signaling, but its functional and prognostic relevance in MM has not been established. Methods: We analyzed 103,171 single-cell transcriptomes from 18 MM samples (10 optimal responders [OR] and 8 suboptimal responders [SOR] to bortezomib-melphalan-prednisone) to investigate cell-type composition, malignant plasma cell subclusters, and tumor-microenvironment crosstalk. InferCNV was used to distinguish malignant plasma cells, which were further reclustered and correlated with bulk prognostic phenotypes. Differential expression, pathway enrichment, transcription-factor activity, pseudotime trajectory, and ligand-receptor interaction analyses were performed. Finally, bulk datasets (GSE9782, GSE2658, MMRF-CoMMpass) and Results: Eleven major cell types were annotated, with plasma cells, T/NK cells, and CD14 Conclusions: This single-cell multi-omic dissection implicates a proliferative, stem-like MalPlasma3 subcluster and identified PTPRG as a key mediator of drug resistance and poor outcome in MM, offering novel prognostic biomarkers and therapeutic targets.