Integration of clinical features and sociodemographic factors: a simplified prognostic model for patients with multiple myeloma based on a double-center retrospective analysis.

This study aimed to develop and validate a prognostic nomogram that combines clinical and sociodemographic factors of patients with newly diagnosed multiple myeloma (MM). A total of 257 newly diagnosed patients with MM from two independent medical centers in China were included in this retrospective cohort study. Univariate and multivariate Cox regression models were used to identify independent risk factors and to construct the nomogram. The predictive ability of the nomogram was evaluated using the areas under the curve (AUCs) and calibration curves. K-fold cross-validation was employed for internal validation of the nomogram performance. Moreover, a stratification system to determine risk level was generated based on the nomogram. Hemoglobulin, creatinine, rurality, and marital status were significantly associated with overall survival (OS) and were incorporated into the nomogram for OS prediction. The prognostic nomogram showed good discrimination and accuracy, and its predictive capability was superior to the International Staging System. The AUC values predicting the 1-, 3-, and 5-year OS probabilities of the nomogram were 0.775, 0.755, and 0.754, respectively. Subsequently, patients were classified into high- and low-risk subgroups based on the median total points of the nomogram; this risk stratification clearly distinguished between high- and low-risk MM patients with significantly different clinical outcomes (median OS: 27 months vs. 84 months). We established a novel prognostic prediction model by comprehensively incorporating clinical and sociodemographic variables, which can effectively predict the survival outcomes in patients with MM.