Single-cell RNA Sequencing Identifies Prognostic Biomarkers in Extramedullary Multiple Myeloma

Multiple myeloma (MM) is the second most common hematologic malignancy, accounting for approximately 10% of all hematological cases, with higher morbidity and mortality.

This study aimed to investigate the clonal evolutionary characteristics to identify novel prognostic biomarkers associated with extramedullary progression in MM.

We downloaded transcriptomic profiles and single-cell microarray (scRNA-seq) data from public databases. Then, we used the LASSO method to develop a prognostic signature and validated its efficacy using external MM cohorts. We evaluated the differences in the immune microenvironment and drug sensitivity (IC50) between the different risk score groups. scRNA-seq analysis identified key cell types through AUCell scores, cell communication, and differentiation trajectory analyses.

In total, 126 DEGs were identified as crucial genes associated with extramedullary and intramedullary MM. After LASSO analysis, seven signature genes were selected to develop a risk score model, and high-risk patients showed worse outcomes. Subsequently, the nomogram incorporating age, albumin, b2m, LDH, and RiskScore predicted 1-, 3-, and 5-year outcomes with high AUCs. Immune analyses showed that 25 immune cell types, 35 immune checkpoints, 27 chemokines, 20 MHC molecules, and 14 receptor- related genes differed significantly between the two risk groups. We also identified 116 drugs (roscovitine and JNK inhibitor VIII) with significantly different IC50 values between the two risk groups. CD4+ T cells exhibited the highest signature gene activity. CellChat analysis demonstrated enhanced communication between CD4+, NK, and CD8+ T cells.

Our study has proposed a risk score model based on seven identified signature genes for MM prognosis and revealed CD4+ T cells to be a major immune cell type associated with MM progression, contributing to personalized treatment decision-making and precise risk stratification of MM.