Polygenic risk scores predict diabetic complications and their response to therapy

Johanne Tremblay, et al.

Go to the profile of Nature Communications
Nov 06, 2019
0
0

Received Date: 3rd October 19

Johanne Tremblay, Mounsif Haloui, Francois Harvey, Ramzan Tahir, Francois-Christophe Marois-Blanchet, Carole Long, Redha Attaoua, Paul Simon, Lara Santucci, Candan Hizel,  John Chalmers,  Michel Marre,  Stephen Harrap, Renata Cifkova,  Alena Krajcoviechova,  David Matthews,  Bryan Williams,  Neil Poulter,  Sophia Zoungas,  Stephen Colagiuri,  Giuseppe Mancia,  Diederick E Grobbee,  Anthony Rodgers, Lisheng Liu, Mawusse Agbessi,  Vanessa Bruat, Marie-Julie Fave, Michelle Harwood,  Philip Awadalla,  Mark Woodward,  Pavel Hamet

Type 2 diabetes increases the risk of cardiovascular and renal complications, but early risk prediction can lead to timely intervention and better outcomes. Through summary statistics of meta-analyses of published genome-wide association studies performed in over 1.2 million of individuals, we combined 9 PRS gathering genomic variants associated to cardiovascular and renal diseases and their key risk factors into one logistic regression model, to predict micro- and macrovascular endpoints of diabetes. Its clinical utility in predicting complications of diabetes was tested in 4098 participants with diabetes of the ADVANCE trial followed during a period of 10 years and replicated it in three independent non-trial cohorts. The prediction model adjusted for ethnicity, sex, age at onset and diabetes duration, identified the top 30% of ADVANCE participants at 3.1-fold increased risk of major micro- and macrovascular events (p=6.3x10-21 and p=9.6x10-31, respectively) and at 4.4-fold (p=6.8x10-33) increased risk of cardiovascular death compared to the remainder of T2D subjects. While in ADVANCE overall, combined intensive therapy of blood pressure and glycaemia decreased cardiovascular mortality by 24%, the prediction model identified a high-risk group in whom this therapy decreased mortality by 47%, and a low risk group in whom the therapy had no discernable effect. Patients with high PRS had the greatest absolute risk reduction with a number needed to treat of 12 to prevent one cardiovascular death over 5 years. This novel polygenic prediction model identified people with diabetes at low and high risk of complications and improved targeting those at greater benefit from intensive therapy while avoiding unnecessary intensification in low-risk subjects.

Read in full at medRxiv.

This is an abstract of a preprint hosted on an independent third party site. It has not been peer reviewed but is currently under consideration at Nature Communications.

Go to the profile of Nature Communications

Nature Communications

Nature Research, Springer Nature