Prognostic nomogram for extensive-stage small-cell lung cancer in the immunotherapy era: a multicenter study of first-line platinum-etoposide with or without a PD-1/PD-L1 inhibitor.

Background: In the immunotherapy era, most patients with extensive-stage small-cell lung cancer (ES-SCLC) still experience early progression, whereas only a minority derive more durable benefit. Most existing prognostic models were developed primarily with chemotherapy-treated cohorts and have not been adequately validated in patients receiving first-line chemoimmunotherapy. We aimed to develop and internally validate a robust nomogram integrating metastatic topography and inflammatory biomarkers to improve baseline risk stratification in contemporary ES-SCLC. Methods: We analyzed a multicenter retrospective cohort of 605 treatment-eligible ES-SCLC patients treated in routine clinical practice with first-line platinum-etoposide (EP) with or without programmed cell death protein 1/programmed death-ligand 1 (PD-1/PD-L1) inhibitors. Patients were randomly split into a training cohort (n=421) and a validation cohort (n=184). A multivariable Cox proportional hazards model incorporating systemic immune-inflammation index (SII), lung immune prognostic index (LIPI), tumor marker status, metastatic sites, and treatment modality was used to develop a nine-variable nomogram and an integer-based scoring system. Model performance was evaluated by concordance index (C-index), time-dependent area under the curve (tAUC), calibration, and decision curve analysis (DCA), with head-to-head comparisons against three published prognostic models. Results: Independent prognostic factors indicating worse prognosis included high neuron-specific enolase (NSE), elevated SII, unfavorable LIPI, malignant pleural effusion (MPE), and specific metastatic sites (liver, bone, brain, and adrenal). Chemoimmunotherapy was independently associated with longer overall survival (OS) after adjustment for baseline prognostic factors. The nomogram showed consistent discrimination with C-indices of approximately 0.75 in both cohorts, and tAUCs generally exceeding 0.70 across 12-24 months. Calibration indicated close agreement between predicted and observed survival probabilities, and DCA demonstrated meaningful net clinical benefit over default strategies. Compared with the Dang, Gao, and Ge models, our nomogram achieved higher tAUCs with better stability across training and validation cohorts. Conclusions: This nine-variable nomogram provides a clinically accessible tool for baseline risk stratification in treatment-eligible patients with ES-SCLC receiving contemporary EP-based therapy in routine practice. It may support individualized prognostic assessment, determination of follow-up intensity, and supportive-care planning. However, its applicability to patients with poorer performance status or heavier disease burden requires further external validation.