Type 2 diabetes genetic risk and incident diabetes across diabetes risk enhancers

Abstract Aims To evaluate the predictive value of a contemporary type 2 diabetes (T2D) polygenic score (PGS) in detecting incident diabetes across a range of diabetes risk factors. Materials and Methods We analysed participants in the Further Cardiovascular Outcomes Research With PCSK9 Inhibition in Subjects With Elevated Risk (FOURIER) trial ( ClinicalTrials.gov , number NCT0176463), which compared the efficacy of proprotein convertase subtilisin/kexin type 9 (PCSK9) inhibitor evolocumab versus placebo in lowering cardiovascular outcomes in participants with stable atherosclerotic cardiovascular disease and LDL cholesterol levels of 70 mg/dL (1.8 mmol/L) or higher who were receiving statin therapy. Genetic risk was characterized using a previously validated T2D PGS based on ~1.2 million single‐nucleotide polymorphisms. PGS was analysed continuously and categorically as high (top 20% of the PGS) and low to intermediate (lower 80% of the PGS). The effect of evolocumab on incident diabetes in patients without diabetes at baseline was also assessed. HbA1c was measured at baseline and every 24 weeks thereafter, while FPG was measured at baseline, week 12, week 24 and every 24 weeks thereafter. Potential cases of incident diabetes were adjudicated centrally. Hazards ratios (HRs) for incident diabetes were adjusted for baseline characteristics and ancestry. Results Among 9388 participants, the mean age was 63 ± 9 years and 22.7% were women, with median HbA1c 39 mmol/mol (36 mmol/mol – 41 mmol/mol; 5.7% [5.4%–5.9%]) and mean body mass index (BMI) 28.7 ± 5 kg/m 2 . Diabetes developed in 690 participants (7.3%) during 2.3 years of median follow‐up. T2D PGS predicted incident T2D (HR per 1‐SD 1.22, 95% CI 1.14–1.32, p < 0.001). The rates of incident T2D in the high and low to intermediate genetic risk categories were 12.1% versus 6.8%, respectively (HR 1.43 95% CI 1.20–1.70, p < 0.001). Notably, high T2D genetic risk had greater predictive strength among individuals with lower HbA1c ( P ‐int = 0.0499) and lower BMI ( P ‐int = 0.004). Conclusions The T2D polygenic score serves as an independent predictor of incident diabetes, particularly among individuals with lower distribution of traditional risk factors.