Comprehensive analysis of plasma cell heterogeneity and immune interactions in multiple myeloma.

This study focused on the role of plasma cells in multiple myeloma (MM) and the associated potential mechanisms. Transcriptomic data of MM and various gene sets from several public databases were downloaded for subsequent analyses. Through single-cell sequencing, 10 major cell types were identified and annotated. The differential gene expression and pathway enrichment between different plasma cell subtypes as well as cell communication analysis, transcriptional regulation analysis, and enrichment analysis in conjunction with the malignant subpopulation were performed. Next, the samples were clustered into two groups by applying non-negative matrix factorization (NMF). Additional analysis revealed notable disparities in survival between the two clusters, correlation with genes involved in classical metabolic pathways and pathway dysregulation, thus confirming the stability and validity of the clustering. Subsequently, Weighted Gene Co-expression Network Analysis was performed and hub genes from the modules most strongly associated with the clustering groups were extracted. We then constructed a prognostic prediction model using Least Absolute Shrinkage and Selection Operator and multiCox regression analysis. The predictive accuracy of the model was evaluated and robustness were confirmed in a separate validation cohort. The gene and pathway dysregulation for the two risk groups was analyzed. Ultimately, an investigation was conducted into the association between the risk model and various immunological features, in terms of antitumor immunotherapy, the tumor microenvironment, and immune checkpoints. This study provides an in-depth investigation into the potential mechanisms underlying MM development and offers new directions to improve therapeutic approaches and enhance patient outcomes.