Clinical use of polygenic scores in type 2 diabetes: challenges and possibilities

Resulting from a combination of genetic and environmental factors, type 2 diabetes is highly heterogeneous in manifestation and disease progression, with the only common feature being chronic hyperglycaemia. In spite of vigorous efforts to elucidate the pathogenetic origins and natural course of the disease, there is still a lack of biomarkers and tools for prevention, disease stratification and treatment. Genome-wide association studies have reported over 1200 variants associated with type 2 diabetes, and the decreased cost of generating genetic data has facilitated the development of polygenic scores for estimating an individual’s genetic disease risk based on combining effects from most—or all—genetic variants. In this review, we summarise the current knowledge on type 2 diabetes-related polygenic scores in different ancestries and outline their possible clinical role. We explore the potential applicability of type 2 diabetes polygenic scores to quantify genetic liability for prediction, screening and risk stratification. Given that most genetic risk loci are determined from populations of European origin while other ancestries are under-represented, we also discuss the challenges around their global applicability. To date, the potential for clinical utility of polygenic scores for type 2 diabetes is limited, with such scores outperformed by clinical measures. In the future, rather than predicting risk of type 2 diabetes, the value of polygenic scores may be in stratification of the severity of disease (risk for comorbidities) and treatment response, in addition to aiding in dissecting the pathophysiological mechanisms involved.