Abstract
INTRODUCTION: The cystatin family is particularly relevant in lung cancer research due to its links to inflammation, protease balance, and tumor progression. Although population-based studies have documented associations between cystatin and lung cancer, causal relationships remain undetermined. METHODS: Based on genomic statistics of seven different cystatins and three subtypes of lung cancer, we conducted a two-sample Mendelian randomization (MR) study. The inverse-variance weighted (IVW) method was the main approach for causality estimation. The weighted median, simple mode, weighted mode, and MR-Egger regression methods were further employed to validate the main findings. In the sensitivity analysis, horizontal pleiotropy was assessed by MR-Egger regression and Cochran's Q test. MR-PRESSO and Radial MR methods were used to identify heterogeneity and remove outliers. RESULTS: Genetically predicted Cystatin 8 was causally associated with squamous cell lung carcinoma (OR = 1.062, 95% CI: 1.004-1.124, p = 0.035). No causal relationships were found for genetically predicted cystatin 8, -B, -D, -F, or -M with squamous cell lung carcinoma, lung adenocarcinoma, and NSCLC. However, outliers were identified between Cystatin D, -M, and -F using MR-PRESSO and Radial MR. After the removal of outliers, the association between Cystatin D and lung adenocarcinoma turned significant (OR = 1.178, 95% CI: 1.023-1.358, p = 0.023). Sensitivity analyses confirmed the robustness of main results after outliers removal. CONCLUSION: Genetically predicted Cystatin 8 was causally associated with squamous cell lung carcinoma. Future population-based studies are required to substantiate these results.