2402-P: MicroRNAs Have Potential Utility for Predicting Risk for Type 2 Diabetes in the Context of Underlying Genetic Characteristics

Introduction and Objective: Health research has historically conflated genetic ancestry with race and ethnicity, limiting precision in risk prediction for multifactorial conditions like type 2 diabetes (T2D). MicroRNAs (miRs) are dynamic molecular regulators that respond to environmental and behavioral exposures, potentially capturing extrinsic risk. The objective of this study was to determine whether miR expression patterns vary by genetic ancestry in individuals with prediabetes. Methods: We included genetic and miR data derived from plasma samples collected at the baseline visit for the Diabetes Prevention Program (DPP) (n=854). We also used a previously established miR risk score derived from random forest and lasso models. Genetic ancestry was estimated using the 1KG reference populations using k-means clustering. Logistic regression was used to determine the odds of developing T2D after 2 years based on ancestral group, miR risk score, or a combination of these two predictors. Results: Seven ancestral groups were selected based on cross-validation stability. Participants exhibited admixed ancestry, with a predominant European component followed by African. Ancestral composition was not associated with the odds of developing T2D. In contrast, the miR risk score was associated with increased odds (1.24, p<0.05). When the miR risk score and ancestral group were both included as predictors, the magnitude of the odds ratios increased for all ancestral groups. Conclusion: These findings suggest that integrating ancestry-informed genetic measures with miRs may improve precision in distinguishing intrinsic genetic risk from extrinsic social, behavioral and environmental influences. This combined approach may enhance risk prediction and inform targeted prevention strategies. Further studies are needed to determine if there are distinct sets of miRs that predict risk within ancestral groups and race and ethnic groups. Disclosure A. Jimenez: None. K.J. Lindquist: None. B.M. Stroebel: None. K.D. Longoria: None. E. Flowers: None. Funding R01DK124228K26DK138247