Early-life Exposure to Tobacco Smoke, Metabolomic Perturbation, and Metabolic Dysfunction-Associated Steatotic Liver Disease Incidence Risk in Later Life.
Abstract
Background
The association of early-life (in utero, childhood, and adolescence) tobacco smoke exposure (ETSE) with adulthood metabolic dysfunction-associated steatotic liver disease (MASLD) incidence risk and the roles of metabolomic perturbation in explaining the above association and serving as biomarkers for assessing MASLD risk remain unknown that warrant exigent investigation.
Methods
We included 168814 subjects from the UK Biobank. Relationships of in utero tobacco smoke exposure (IUTSE) and age of smoking initiation (ASI) with adulthood MASLD incidence were examined by Cox proportional hazards models. Elastic net regression (ENR) was exploited to construct the metabolic signatures associated with ETSE. Potential roles of individual metabolites and metabolic signatures were probed by mediation analyses. The MASLD risk prediction model was performed to exploratorily evaluate the possible value of extra addition of ENR-screened metabolites.
Results
Subjects with IUTSE presented a higher MASLD risk (hazard ratio=1.23, 95% confidence interval [CI]: 1.12∼1.36) than those without. Compared to never-smokers, subjects starting smoking in adulthood, adolescence, and childhood had 1.25 (1.10∼1.42), 1.28 (1.14∼1.45), and 1.42 (1.20∼1.66) times higher MASLD risk, respectively (Ptrend<0.001). Individual metabolites and metabolic signatures accounted for 0.15%∼38.18% of the ETSE-MASLD association, with fatty acids accounting for the greatest proportion among individual metabolites. Compared to the basic model, the model with extra inclusion of even 8 ENR-screened metabolites significantly linked to MASLD risk as exploratory metabolic biomarkers (Pdifference<0.001).
Conclusions
ETSE was linked to elevated risk of adulthood MASLD incidence, which was explained by perturbated metabolome, while the metabolites identified may have the potential to serve as exploratory biomarkers for MASLD.