Multi-omics profiling reveals an immunosuppressive microenvironment and PRL-3 as a therapeutic target in myeloma extramedullary disease

Abstract Background Extramedullary disease (EMD), an aggressive manifestation of multiple myeloma, remains a significant adverse prognostic factor despite advances in novel immunotherapeutic approaches. Method Single-cell RNA sequencing of five matched pleural effusion or ascites and bone marrow pairs (including two from GSE230510), and whole-exome sequencing of seven extramedullary specimens were performed to characterize the tumor microenvironment and heterogeneity, aiming to identify potential therapeutic targets. Results CD8+ T-cell cytotoxicity was significantly reduced in extramedullary lesions versus bone marrow (p = 0.002). Macrophages exhibited higher M2 polarization scores in extramedullary sites (p = 0.008). Cell–cell communication analysis revealed that macrophages, functioning as signal senders, exerted significantly stronger signaling influences on myeloma cells in the extramedullary lesions compared to the bone marrow. This interaction primarily mediated through 2 pathways. One is FN1-related pathway (FN1–SDC1, FN1–CD44, FN1–ITGA4+ITGB1, and FN1–ITGA4+ITGB7). The other is SPP1-associated pathway (SPP1–CD44 and SPP1–ITGA4+ITGB1). Hotspot analysis combined with the WGCNA algorithm was employed to identify intra-tumoral expression modules associated with extramedullary infiltration. Modules related to cell proliferation, glycolysis and hypoxia, p53 signaling, and NF-κB signaling were preferentially expressed in myeloma cells of the extramedullary lesions. Validation using the MMRF CoMMpass dataset demonstrated that the identified modules, particularly those associated with cell proliferation, glycolysis and hypoxia, and p53 signaling—were significantly correlated with poorer overall survival (p < 0.001; p < 0.001; p < 0.001). The gene PTP4A3 (PRL3) was found to be upregulated in myeloma cells across all five pairs of extramedullary lesions compared to matched bone marrow. Analysis of the MMRF CoMMpass dataset revealed that elevated PTP4A3 (PRL3) expression was significantly associated with poor survival outcomes (p = 0.0024). Treatment of myeloma cell lines with a specific PRL-3 inhibitor markedly suppressed cell migration and proliferation while promoting apoptosis. In vivo studies using NCG mice further demonstrated the anti-myeloma effects of PRL-3 inhibition. Whole-exome sequencing revealed that the RTK–RAS signaling pathway was mutated in all ME samples, while mutations in the NOTCH and WNT pathways were detected in 86% and 71% of ME samples, respectively. Conclusion Extramedullary lesions of multiple myeloma exhibit an immunosuppressive microenvironment. PRL3 represents a potential therapeutic target in multiple myeloma with extramedullary infiltration.