Sex-specific associations of the bilirubin-to-albumin ratio with lung cancer risk: evidence from Cox regression and causal survival forest analysis

BACKGROUND: The bilirubin-to-albumin (BA) ratio may reflect the balance between circulating bilirubin and albumin-related binding capacity, but its association with lung cancer risk remains unclear. METHODS: We analyzed data from the Korean Cancer Prevention Study-II, a nationwide prospective cohort. After excluding participants with baseline cancer or missing key variables, 81,746 men and 49,421 women were included. Total, indirect, and direct BA ratios were calculated and standardized per 1-standard deviation increase. Sex-specific associations with lung cancer risk were assessed using Cox proportional hazards models, quartile analyses, restricted cubic splines, and causal survival forest models. RESULTS: In Cox models, higher total and direct BA ratios were inversely associated with lung cancer risk in men, whereas higher total and indirect BA ratios were positively associated with risk in women. Restricted cubic spline analyses supported inverse dose-response patterns for total and direct BA ratios in men. Causal survival forest analyses showed statistically significant positive average treatment effects for lung cancer-free survival probability only in men across 5-, 10-, and 13-year horizons. CONCLUSIONS: BA ratio showed sex-specific associations with lung cancer risk, with more consistent inverse associations among men.