CELL INTELLIGENCE · 肿瘤细胞治疗研究

The clinical heterogeneity of multiple myeloma (MM) remains incompletely captured by existing staging systems. To determine whether baseline immune profiles could refine prognostication, we conducted a large-scale analysis of 703 newly diagnosed MM patients. Peripheral blood immune subsets and serum cytokines were quantified before treatment via flow cytometry and multiplex immunoassays. Time-dependent ROC analysis identified optimal prognostic thresholds for each parameter. Univariate analysis associated inferior overall survival (OS) with low CD19 B-cell counts, a low CD4 /CD8 ratio, high NK cell percentage, elevated levels of IL-1 , sIL-2R, IL-6, IL-8, IL-10, and TNF, and low complement C3. A multivariate Cox model integrated the most robust predictors into an immune risk score (IM): IM = 0.107 (CD4 /CD8 ) + 0.001 sIL-2R + 0.003 IL-6 + 0.006 IL-8 1.238 C3. Using the optimal cut-off (0.394), patients were stratified into high-risk (n = 231) and low-risk (n = 472) groups. The low-risk group exhibited significantly longer median OS (64.5 months vs. 32.2 months; p < 0.0001), and the IM score remained an independent prognostic factor after adjusting for clinical variables. Subgroup analysis confirmed its predictive value across treatment backgrounds. These results establish the pre-treatment systemic immune state as a powerful prognostic determinant and provide a clinically applicable immune-based scoring system for improved risk stratification in MM.