Integration of peripheral blood lymphocyte subsets and cytokines enhances prognostic stratification in newly diagnosed multiple myeloma

Summary

Immune dysregulation is involved in multiple myeloma (MM) pathogenesis and progression yet current staging systems do not incorporate indicators of immune status. We retrospectively included 260 newly diagnosed multiple myeloma (NDMM) patients with baseline data of peripheral blood (PB) lymphocyte subsets and cytokines from Tianjin Medical University General Hospital. Through majority voting across seven survival prediction algorithms and subsequent Cox analysis, cluster of differentiation (CD)3 ⁺ CD8 ⁺ T‐cell absolute count and interleukin (IL‐10) were identified as immune prognostic predictors to construct the immune risk score. A comprehensive prognostic nomogram was subsequently established by combining this score with high‐risk cytogenetic abnormalities, serum calcium and serum lactate dehydrogenase. Shapley additive explanation analysis ranked the immune risk score as the leading contributor to the final model. The nomogram achieved a cross‐validated 3‐year time‐dependent area under the curve for overall survival (OS) of 0.825. The model was further evaluated in a temporal validation cohort. In exploratory analyses within our cohort, the nomogram showed better prognostic performance than the established staging systems and further stratified outcomes among patients receiving anti‐CD38 antibody‐containing regimens. The model was deployed as an exploratory online calculator for convenience. These findings require independent external multicentre validation before implementation.