Polygenic scores used to predict type 2 diabetes outcome in largest study to date

New genetic markers that drive the development of type 2 diabetes and other vascular diseases have been identified.

An international group of researchers, as part of the Type 2 diabetes Global Genomics Initiative, conducted the largest genome-wide association study (GWAS) for type 2 diabetes. They used data from a highly diverse set of individuals to locate 1289 genetic markers associated with type 2 diabetes, of which 145 were not known previously.

Co-senior author Dr Cassandra Spracklen, assistant professor of biostatistics and epidemiology at the University of Massachusetts, Amherst, said: 'Although environment and lifestyle are well-established risk factors for type 2 diabetes, heritability has been estimated to be 69 percent amongst individuals of 35-60 years of age.'

Publishing their findings in Nature, the researchers collected data from more than 2.5 million people, 428,452 of whom had type 2 diabetes. They organised the GWAS data into six subsets of genetically similar studies, according to participants' ancestries.

Type 2 diabetes is a common condition that can lead to increased blood glucose levels, cardiovascular problems, and other complications due to impaired insulin secretion and resistance. The disease affects more than 400 million adults worldwide and several different factors contribute to its development, including lifestyle, obesity and genetics. Type 2 diabetes is also known to vary considerably in prevalence between people of different ancestries.

The researchers identified 1289 genetic variants linked with type 2 diabetes, mapping to 611 different positions (loci) in the genome, of which 145 loci had not been seen before.

Within the GWAS data, the researchers identified eight clusters with distinct profiles of association with 37 cardiometabolic phenotypes. They discovered that the associations between individual clusters and diabetes complications, such as coronary artery disease, peripheral artery disease and end-stage neuropathy, existed across ancestry groups, highlighting the importance of obesity-related processes in the development of vascular outcomes.

The researchers then used the data from these clusters to build polygenic risk scores and tested their association with type 2 diabetes-related vascular outcomes in another diverse ancestry group of more than 250,000 people, of which over 30,000 had type 2 diabetes.

'We're trying to understand how diabetes develops,' said Dr Spracklen, 'And we're trying to better understand how these genetic variants are actually working within a biological tissue or at the cellular level, which can ultimately lead to new drug targets and treatments.'

While the research covers a highly diverse population, the authors also acknowledge that the dataset is far from complete and caution in their paper that: 'Bolstering GWAS collections in underrepresented populations remains an urgent priority for the human genetics research community and highlights the need for careful interpretation of results that does not generalise findings across ancestry groups that are sensitive to biased representation.'