Plasma Metabolomic Biomarkers of Middle-Adulthood to Late-Adulthood Atopic Dermatitis: A Prospective Cohort Study...

Although atopic dermatitis (AD) in adults has been prevalent, there has been a paucity of previous studies on the metabolic markers linked to its development. We investigated whether plasma metabolomic biomarkers are associated with incident middle-adulthood to late-adulthood AD and identified its predictors.

This study utilized data collected from the UK Biobank between 2006 and 2010. We prospectively analyzed metabolites in 79,414 participants aged 40 to 69 years, predominantly of European ancestry and free of AD, to assess the incidence of AD, which were collected using the International Classification of Diseases from hospital inpatient admission records from the baseline through 2022. The relationship between these metabolites and the incidence of AD was estimated using Cox proportional hazard models, and the predictive value of the metabolites for the incidence of middle-adulthood to late-adulthood AD was assessed.

The incidence rate of middle-adulthood to late-adulthood AD was 7.81 per 10,000 person-years, and the median follow-up was 161 months. A total of 35 plasma metabolites included in lipoprotein lipids, fatty acids, amino acids, and biomarkers related to fluid balance were significantly linked to the risk of middle-adulthood to late-adulthood AD incidence. Significant non-linear associations of middle-adulthood to late-adulthood AD incidence were observed with triglycerides in medium very-low density lipoprotein, leucine, and the total concentration of branched-chain amino acids. The incorporation of the identified 35 metabolites with the covariates for middle-adulthood to late-adulthood AD risk prediction achieved an area under the curve of 0.71 (95% confidence interval, 0.67-0.75) and reclassification ability (net reclassification improvement = 21.4%, P < 0.001).

Findings from this prospective study suggest that the identified plasma metabolites could better clarify the underlying metabolic vulnerabilities and improve risk prediction associated with the development of middle-adulthood to late-adulthood AD.