A predictive model and mechanistic study of treatment effectiveness in patients newly diagnosed with small cell lung cancer

Introduction Extensive-stage small cell lung cancer (ES-SCLC) is an aggressive malignancy with a poor prognosis. This study aimed to identify and validate clinical and laboratory biomarkers for predicting treatment response and overall survival (OS) in ES-SCLC patients. Methods We retrospectively analyzed 101 ES-SCLC patients receiving first-line treatment. Logistic and Cox regression analyses identified independent factors influencing treatment efficacy and OS. Subgroup analysis was performed to compare white blood cell (WBC) changes between chemotherapy-alone and chemo-immunotherapy groups. Predictive models were constructed and evaluated via cross-validation, ROC, and calibration curves. Differential expression of key proteins (neuron-specific enolase (NSE), fibrinogen (FIB), and gastrin-releasing peptide precursor (ProGRP)) was validated using GEO database data. Results Pre-chemotherapy tumor size and post-cycle 2 FIB levels were independent predictors of treatment efficacy. Pre-chemotherapy WBC count, pre-chemotherapy D-dimer, and post-cycle 2 ProGRP were independent risk factors for OS. The predictive models demonstrated strong performance. Subgroup analysis showed no significant difference in WBC changes between treatment regimens (mean change: -2.30 ± 2.47 vs. -2.08 ± 2.45, p=0.659). GEO data confirmed the differential expression of FIB and ProGRP. Discussion Our findings establish robust and validated models based on readily available clinical metrics (tumor size, WBC, D-dimer, FIB, ProGRP) to predict outcomes in ES-SCLC, which could aid in personalizing treatment strategies. The stability of WBC trends across therapies strengthens the prognostic value of baseline WBC.