Causal association between tobacco use, fibroblast growth factors, and pulmonary fibrosis: A two-sample Mendelian randomization study

Abstract

Introduction Cigarette smoking is a risk factor for pulmonary fibrosis, but its underlying mechanisms remain unclear. Fibroblast growth factors (FGFs) have been implicated in pulmonary fibrosis, yet their role in smoking-induced pulmonary fibrosis remains unclear. This study employed Mendelian randomization (MR) analysis to investigate the causal relationships among FGFs, smoking, and pulmonary fibrosis. Methods FGFs, smoking behavior, and pulmonary fibrosis data were collected from the IEU Open GWAS Project and the GWAS Catalog. Single-nucleotide polymorphisms were used for instrumental variable selection. Inverse-variance weighted (IVW) was the primary MR method, supplemented by Weighted mode, Weighted median, and MR-Egger regression. Sensitivity analyses included MR-PRESSO, MR-Egger, Cochran’s Q, and leave-one-out. Results IVW analysis showed that FGF10 (OR=0.74; 95% CI: 0.57–0.96, p=0.02), FGFR1 (OR=1.38; 95% CI: 1.04–1.84, p=0.03), and FGF19 (OR=0.78; 95% CI: 0.64–0.94, p=0.01) were statistically associated with pulmonary fibrosis risk. MR analysis of smoking behaviors as exposures and FGF10/FGFR1/FGF19 as outcomes showed no significant associations across all MR approaches. MR-Egger regression showed that horizontal multiplicity did not impact analysis except for FGFR3. MR-PRESSO did not identify any outliers. Sensitivity analysis showed that the association results were robust. Conclusions FGF10, FGF19, and FGFR1 may be causally associated with pulmonary fibrosis. However, the MR analysis did not support a causal relationship between smoking and FGFs, suggesting that smoking contributes to pulmonary fibrosis independently of FGFs. This highlights the need for well-designed prospective studies to minimize bias in observational studies and for larger GWAS datasets to enhance future MR analyses.