Integrated transcriptomics reveals a TNFRSF-metabolism axis and a context-dependent BCMA signaling state in myeloma progression

Abstract

Multiple myeloma (MM) evolves from monoclonal gammopathy of undetermined significance (MGUS) and smoldering multiple myeloma (SMM) with extensive metabolic reprogramming and immune remodeling. We investigated whether coordinated activation of the tumor necrosis factor receptor superfamily (TNFRSF) aligns with metabolic state and prognosis across this continuum. We integrated bone-marrow single-cell RNA sequencing from healthy donors, MGUS, SMM, and MM with bulk RNA sequencing from the Multiple Myeloma Research Foundation (MMRF) CoMMpass cohort. At single-cell resolution, per-cell metabolic activity scores increased with disease stage and were highest in malignant plasma cells. TNF/TNFRSF signaling was strongest in MM, and expression of TNFRSF17 (B-cell maturation antigen, BCMA) in plasma cells correlated with metabolic activity (Spearman ρ = 0.56 in MM). In the clinical cohort, a pooled TNFRSF score stratified overall survival (log-rank p < 0.0001), with individual associations for TNFRSF17 and TNFRSF1A/1B. Joint stratification by TNFRSF activation and metabolic activity revealed the poorest outcomes in tumors with low TNFRSF activation but high metabolic activity, indicating that metabolic state contributes to adverse risk in a receptor-context–dependent manner, while TNFRSF captured additional risk not explained by metabolism alone. When focusing specifically on BCMA, the prognostic ordering across BCMA x metabolism strata was not identical to the pooled TNFRSF model, underscoring that metabolic risk is modulated by the receptor state rather than acting as a universal driver. These findings define an integrated TNFRSF-metabolism axis and a context-dependent BCMA state in advanced MM, clinically relevant but adjustment-sensitive survival associations that warrant further validation and combination strategies that explicitly consider metabolic status.