Introduction: We assessed the performance of our newly developed polygenic risk score (PRS) to predict microvascular and macrovascular complications of type 2 diabetes (T2D) in men and women. The PRS is composed of 600 common genomic variants associated to diabetes, cardiovascular and renal diseases and their key risk factors selected from summary statistics of meta-analyses of published genome-wide association studies performed in over 1.2 million of individuals. The performance of the polygenic model was assessed by c-statistics in 4098 genotyped participants of European descent of the ADVANCE trial (46.4% women) followed during a period of five years.

Methods: The logistic regression model that included the PRS adjusted for the principal component (PC1) of genetically determined ancestry, age at diagnosis and T2D duration, and treatment assignment, did not include any clinical or outcome data.

Results: The discrimination between cases (having a specific complication) from controls (free of this complication) at entry in ADVANCE had AUCs for microvascular complications of 0.63 (0.60-0.65) in men and 0.66 (0.63-0.69) in women, sex differences p = 0.07. AUCs for macrovascular complications were 0.56 (0.54-0.58) in men and 0.57 (0.54-0.61) in women; p = 0.41. The AUCs for prediction of incident cases, defined as having an outcome during the ADVANCE trial (free of outcome at baseline) compared to controls that did not have a specific outcome at any time during the study, were for microvascular events 0.66 (0.63-0.70) in men and 0.71 (0.66-0.77) in women; p = 0.15. AUCs for macrovascular events were 0.65 (0.62-0.68) and 0.72 (0.68-0.76) respectively, p = 0.01. AUC for prediction of cardiovascular death occurring during the trial was 0.71 (0.67-0.75) in men and 0.77 (0.72-0.82) in women; p = 0.04.

Conclusion: Our polygenic model demonstrated an overall better performance in women than in men and a better prediction capacity in individuals free of previous events in both sexes.


J. Tremblay: Research Support; Self; Servier. Stock/Shareholder; Self; OPTITHERA. R. Attaoua: None. M. Haloui: None. R. Tahir: None. C. Long: None. C. Hizel: None. J. Chalmers: None. S. Harrap: None. M. Woodward: Consultant; Self; Amgen, Kyowa Hakko Kirin Co., Ltd. P. Hamet: Research Support; Self; Servier. Stock/Shareholder; Self; OPTITHERA.


Genome Quebec; Canadian Institutes of Health Research; MEIE; CQDM; Opti Thera; Servier, Canada Research Chair in Predictive Genomics

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at