Can genetic evidence help us to understand the fetal origins of type 2 diabetes?

Lower birthweight is consistently associated with a higher risk of type 2 diabetes in observational studies, but the mechanisms underlying this association are not fully understood. Animal models and studies of famine-exposed populations have provided support for the developmental origins hypothesis, under which exposure to poor intrauterine nutrition results in reduced fetal growth and also contributes to the developmental programming of later type 2 diabetes risk. However, testing this hypothesis is difficult in human studies and studies aiming to do so are mostly observational and have limited scope for causal inference due to the presence of confounding factors. In this issue of Diabetologia, Wang et al (doi: 10.​1007/​s00125-016-4019-z) have used genetic variation associated with birthweight in a Mendelian randomisation analysis to assess evidence of a causal link between fetal growth and type 2 diabetes. Mendelian randomisation offers the potential to examine associations between exposures and outcomes in the absence of factors that would normally confound observational studies. This commentary discusses the results of the Mendelian randomisation study carried out by Wang et al, in relation to the study design and its limitations. Challenges and opportunities for future studies are also outlined.