[Polygenic risk score combined with traditional risk factors for assessing the risk of type 2 diabetes mellitus among Chinese adults].

OBJECTIVE

To evaluate the predictive ability of a polygenic risk score(PRS) combined with traditional risk factors for type 2 diabetes mellitus(T2DM) in Chinese adults.

METHODS

Using data from the "China Adult Chronic Disease and Nutrition Surveillance(2015)" project, we conducted a case-control study comprising 2041 people with T2DM and 1149 healthy adults of same age as control group. Total DNA was extracted from cryopreserved leukocyte samples separated from blood, and genotyping was performed using the Illumina ASA platform. Two PRS models were constructed based on East Asian-specific and trans-ancestry genome-wide association study(GWAS) summary statistics, and their predictive performance was compared. T2DM-related risk factors were selected using forward stepwise regression. The better-performing PRS was then integrated with traditional risk factors into a combined prediction model. The receiver operation curve(ROC) was used to evaluate and compare the predictive performance of models containing traditional factors alone, PRS alone, and the combined model.

RESULTS

Our analysis revealed that the trans-ancestry GWAS-derived PRS provided more reliable genetic stratification for T2DM risk in this population compared to an East Asian model. Beyond genetic factors, multivariable analysis confirmed significant associations between T2DM incidence and several traditional risk factors, including lipid profiles, blood pressure, urban-rural distribution, education, and drinking status. While models containing only traditional risk factors or PRS alone showed moderate predictive capability(area under the curve(AUC)=0.656 and 0.666, respectively), their integration significantly enhanced performance, elevating the AUC to 0.724.

CONCLUSION

These findings demonstrate that incorporating a PRS derived from a trans-ancestry GWAS covering extensive SNPs alongside conventional risk factors substantially improves T2DM risk prediction in Chinese adults.