D102-16 Smoking Status Is Not an Independent Prognostic Factor in Small-cell Lung Cancer: A Large-scale, Multicenter Cohort Study With Propensity Score-based Weighting

Abstract Rationale Small-cell lung cancer (SCLC) is an aggressive malignancy closely linked to smoking, yet a distinct subset arises in never-smokers. Whether smoking independently affects prognosis after controlling for baseline and treatment differences remains uncertain. This study examined the impact of smoking on overall survival (OS) and progression-free survival (PFS) in a large multi-center SCLC cohort and explored subgroup variations by age and comorbidities. Methods We retrospectively analyzed patients with pathologically confirmed SCLC treated at three tertiary hospitals in China between July 2018 and July 2024. Patients were classified as never-smokers or ever-smokers. OS and PFS were assessed using Kaplan-Meier and Cox proportional hazards models. Inverse probability of treatment weighting (IPTW) was applied to balance baseline characteristics. Causal mediation analysis evaluated indirect effects of all variants. Subgroup analyses examined the modifying effects of all factors within smoking strata. Results A total of 1,180 patients were included, of whom 29.1% were never-smokers (Table 1). Before adjustment, ever-smokers showed significantly worse OS than never-smokers (HR = 1.20; 95% CI 1.04-1.38; p = 0.013). After IPTW adjustment, the difference became non-significant (HR = 1.13; 95% CI 0.77-1.66; p = 0.54), suggesting that baseline clinical differences largely explained the crude disparity (Table 2). Mediation analysis indicated that sex, clinical stage, and surgical resection partially mediated the smoking-OS relationship (Table 3). In multivariable models (Table 4), chemotherapy remained strongly associated with improved OS (smokers, HR = 0.67, 95% CI 0.56-0.81; non-smokers, HR = 0.25, 95% CI 0.18-0.33; both p 0.001) in both groups and PFS (HR = 0.25; 95% CI 0.08-0.86, p = 0.0260.01) in never-smokers. Radiotherapy showed benefit only in univariable model, while immunotherapy did not significantly affect OS or PFS in any model.Subgroup analyses (Figure 2) showed that the influence of age, hypertension, and tuberculosis on survival differed between smokers and never-smokers, suggesting potential biological and clinical heterogeneity between smoking-related and non-smoking SCLC. Conclusions In this multi-center cohort, smoking status was not an independent predictor of survival after IPTW adjustment. The apparent survival disadvantage among smokers was largely mediated by differences in sex, stage, and surgical treatment. Chemotherapy significantly improved both OS and PFS across all groups. Distinct subgroup patterns indicate that smoking and non-smoking SCLC may represent different clinical subsets, highlighting the need for further studies integrating genomic and environmental data to clarify underlying mechanisms. This abstract is funded by: None