Associations between systemic immune-inflammation index and immune-related hypothyroidism in non-small cell lung cancer patients receiving immune checkpoint inhibitors: a retrospective study

Objective Immune-related hypothyroidism (irH) is one of the most common endocrine immune-related adverse events (irAEs) mediated by immune checkpoint inhibitors (ICIs), and its occurrence is influenced by the systemic immune-inflammatory status. This study investigated the association between baseline systemic immune-inflammation index (SII) and irH risk in patients with non-small cell lung cancer (NSCLC) receiving ICIs. Methods In this retrospective study, patients with driver gene-negative locally advanced or metastatic NSCLC were included. All patients underwent first-line treatment with PD-1 inhibitors combined with chemotherapy at Yancheng Third People’s Hospital between October 2019 and December 2024. SII was determined using the following formula: (platelet count × neutrophil count)/lymphocyte count. Univariate and multivariate logistic regression analyses were performed to determine independent factors associated with irH. A nomogram prediction model was subsequently constructed and assessed using calibration curves, receiver operating characteristic (ROC) curves, and decision curve analysis (DCA). Results 160 patients were enrolled, including 65 (40.6%) in the irH group and 95 (59.4%) in the non-irH group. Compared with the non-irH group, the irH group had significantly lower baseline log2-SII (8.95 ± 0.65 vs. 9.55 ± 0.80, P < 0.001). Multivariate regression analysis revealed that log2-SII (OR = 0.13, 95% CI: 0.05-0.27, P < 0.001) was an independent protective predictor for irH, while age, TG, LDL, TB, FIB, and D-dimer were independent predictors (all P < 0.05). The nomogram developed from these factors demonstrated potential discriminative ability (AUC = 0.884, 95% CI: 0.832–0.935) and calibration (absolute error = 0.025), and DCA revealed a net clinical benefit across a threshold probability range of 5%–90%. Subgroup analysis revealed that in both LUSC and LUAD patients, the irH group exhibited significantly lower log2-SII levels than the non-irH group (all P < 0.05). Conclusion SII is closely associated with irH in NSCLC patients receiving ICIs, with baseline SII serving as an independent predictor of irH. The nomogram incorporating SII along with age, lipid profiles, bilirubin, and coagulation parameters can effectively predict the risk of irH, providing a reference for identifying high-risk patients at an early stage and enhanced thyroid function monitoring in clinical practice.