Abstract
Early-life exposures could affect the risk of type 2 diabetes (T2D) in adulthood, but the mechanisms remain unclear. We included 319,951 (54.5% women) UK Biobank participants free of T2D at baseline and within the first year of follow-up to investigate whether early-life exposures contribute to T2D through biological aging estimated using the Klemera-Doubal (KDM) and PhenoAge methods. We first examined associations of exposures with biological age and T2D, respectively, using multivariable-adjusted linear models and Cox regression models. Next, we investigated the role of biological age acceleration in exposure-T2D associations using mediation analyses. Overall, the mean (standard deviation) chronological age of participants was 56.3 (8.1) years, KDM age was 40.9 (13.1) years, and PhenoAge was 44.4 (10.0) years. 17,062 T2D cases developed during a median (interquartile range) follow-up of 14.3 (13.5-15.0) years. Maternal smoking around birth, being part of a multiple birth, earlier puberty, and being relatively plumper or thinner at age ten were associated with both a higher biological age and T2D risk, while having a higher birth weight and being breastfed were associated with a lower T2D risk. Biological age acceleration partly mediated the exposure-T2D associations, ranging from a proportion of 8.3% (95%CI: 6.0-13.1%) of the association between birthweight-T2D being mediated by PhenoAge to a proportion of 27.7% (95%CI: 15.3-48.5%) of the association between breastfeeding-T2D being mediated by KDM. In conclusion, early-life exposures were associated with biological age acceleration that partly mediated the exposure-T2D associations, highlighting the importance of addressing early-life risks and biological aging in prevention strategies.</p>