Genomic risk prediction for type 2 diabetes in Australian individuals aged 70 years and older

Abstract Aims The utility of a polygenic score (PGS) for type 2 diabetes (T2D) has been demonstrated in the general adult population. However, while previous studies have included older adults within broader age ranges, the performance of PGS specifically in older individuals aged ≥70 years remains unclear. We aimed to evaluate the predictive utility of a PGS in an older cohort. Materials and Methods We derived a PGS in 12 174 Australian participants aged ≥70 years from the ASPREE trial, with a median follow‐up of 4.6 years. T2D was defined by self‐report, commencement of glucose‐lowering medication, or a fasting plasma glucose of ≥7.0 mmol/L. Multivariable logistic and Cox models examined associations between the PGS and baseline and incident T2D, adjusting for clinical risk factors. Risk prediction was evaluated using area under the curve (AUC), C‐index, and net reclassification improvement (NRI). Results At baseline, 1150 (9.4%) participants had prevalent T2D. During follow‐up, an additional 590 (4.8%) developed incident T2D. Per standard deviation, the PGS was significantly associated with baseline (odds ratio: 2.39 [95% CI: 2.19–2.61]) and incident (hazard ratio: 1.55 [1.40–1.71]) T2D. The PGS improved prediction over the clinical risk factors, increasing the AUC from 0.70 to 0.79, and C‐index from 0.67 to 0.71 (both p < 0.001). Adding the PGS to a clinical risk model resulted in NRI of 0.63 [0.57–0.71] for baseline T2D and 0.42 [0.30–0.51] for incident T2D within 5 years. Conclusion The PGS significantly enhances T2D risk prediction in older adults, supporting its potential as a clinical tool for risk stratification.