DNA Methylation Biomarkers Predict Offspring Metabolic Risk From Mothers With Hyperglycemia in Pregnancy.
Level 3 - non-randomized controlled study
Prospective longitudinal cohort study assessing cord blood biomarkers and long-term offspring outcomes with mechanistic cell validation.
PubMed 40834241 · doi:10.2337/db25-0105
What was done
Researchers investigated whether cord blood DNA methylation marks associated with maternal hyperglycemia during pregnancy could predict offspring metabolic risk. Offspring were evaluated for β-cell dysfunction at ages 7, 11, and 18 years, comparing the predictive ability of methylation markers against clinical factors alone. Mechanistic validation was conducted in human β-cells and pancreatic ductal epithelial cells exposed to hyperglycemia.
What was found
The abstract provides no numerical data, hazard ratios, or model performance metrics (e.g., AUC). It reports that 19 cord blood DNA methylation biomarkers associated with maternal glycemic status improved the prediction of offspring β-cell dysfunction at 7, 11, and 18 years of age compared to clinical factors alone, and that in vitro hyperglycemia influenced methylation-dependent gene expression in pancreatic cell lines.
Why it matters
This study indicates that epigenetic modifications present at birth can serve as early biomarkers of long-term diabetes risk and help clarify how in utero hyperglycemia contributes to metabolic disease later in life.
Limits
The abstract does not state the cohort sample size, patient population characteristics, or specific quantitative effect sizes and prediction statistics. Findings rely on surrogate cord blood tissue for β-cell risk, and the abstract does not report external cohort replication.