To assess the predictive value of a nomogram model constructed based on serum electrolytes levels and related parameters for extensive-stage small cell lung cancer (SCLC).

A retrospective analysis was conducted on the baseline data, serum electrolytes levels, and related parameters of 231 SCLC patients initially diagnosed at the First Hospital of Lanzhou University between September 2016 and August 2019. Receiver operating characteristic (ROC) curve analysis was performed to screen indicators with diagnostic value. Binary logistic regression analysis was utilized to identify independent risk factors for extensive-stage SCLC, and a nomogram model was constructed. The model was evaluated using calibration curve and decision curve analyses.

Seven predictive indicators were identified by ROC curve analysis: lactate dehydrogenase (LDH), lymphocytes, serum magnesium (Mg²⁺), neutrophil-to-lymphocyte ratio (NLR), lymphocyte-to-monocyte ratio (LMR), lactate dehydrogenase-to-lymphocyte ratio (LDHLR), and mean platelet volume-to-lymphocyte ratio (MPVLR). Univariate logistic regression analysis showed that sex, LDH, lymphocytes, serum Mg²⁺, LDHLR, and MPVLR were significantly associated with the disease stage (limited-stage vs extensive-stage) at initial diagnosis in SCLC. After multivariate analysis adjustment, serum Mg²⁺ and sex were identified as independent predictors of limited-stage/extensive-stage classification in initially diagnosed SCLC patients. The fusion model constructed based on these factors yielded an area under the curve of 0.654 (95% confidence interval: 0.580–0.727), demonstrating a sensitivity of 69.2% and specificity of 59.1%. The calibration curve indicated good consistency between the model-predicted probability of extensive-stage SCLC and imaging examination results. Decision curve analysis further demonstrated the model's high clinical utility.

Sex and serum magnesium level at admission are independent predictive factors for extensive-stage disease in initially diagnosed SCLC patients.