Relationship between systemic immune-inflammation index and all-cause mortality in stages IIIB-IV epidermal growth factor receptor-mutated lung adenocarcinoma.

Background: This study investigates the relationship between the systemic immune-inflammation index (SII) and all-cause mortality (ACM) risk in individuals with stages IIIB-IV epidermal growth factor receptor (EGFR)-mutated lung adenocarcinoma. Methods: The clinical data of 187 individuals with stages IIIB-IV EGFR-mutated lung adenocarcinoma from Anhui Chest Hospital, collected from June 2017 to December 2023, were retrospectively analyzed. SII was calculated as platelet count × neutrophil count/lymphocyte count. The receiver operating characteristic (ROC) curve was employed to determine the optimal threshold SII, and individuals were classified as low and high SII groups. ACM serves as the primary endpoint. Univariate and multivariate analyses were conducted using Cox proportional hazards models. The robustness of the findings was tested by subgroup and sensitivity analyses. Results: The ACM risk was notably elevated in the high SII group (p = 0.001) compared with the low SII group. Multivariate Cox analysis demonstrated that SII can independently predict poor prognosis. In the fully adjusted model, compared with the low SII group, the ACM risk was 1.985 times higher in the high SII group (hazard ratio [HR] = 1.985; 95% confidence interval [CI] = 1.216-3.240; p = 0.006). Subgroup analyses showed that SII was more strongly associated with ACM risk in men (HR = 3.245; p = 0.005), and this relationship was also significant among female patients (HR = 2.036; p = 0.048). In individuals aged ≥ 65 years, a high SII was significantly associated with an elevated ACM risk (HR = 2.675; p = 0.004). No such relationship was observed in individuals aged under 65. Sensitivity analyses indicated that high SII remained significantly correlated with elevated ACM risk after excluding individuals with special types of adenocarcinoma, stage III lung adenocarcinoma, or diabetes (all p< 0.05), supporting its potential as an independent prognostic indicator. ROC curve analysis demonstrated that SII had a moderate predictive ability for ACM, with an AUC of 0.669 (95% CI = 0.527-0.812; p = 0.021). Conclusion: Elevated SII is an independent biomarker for predicting ACM in individuals with stages IIIB-IV EGFR-mutated lung adenocarcinoma, with a stronger predictive value in male and older populations.