Principal Investigator: Mr Adam Mitchell
Uppsala UniversityTags: 30223, Amylin, Bone, diabetes, fracture, genetics
Type 2 diabetes mellitus (T2DM) is associated with an increased risk of fracture despite an unchanged or even increased bone mineral density. Islet amyloid polypeptide (IAPP) is a gastro-intestinal hormone implicated T2DM pathophysiology and with increased bone mass. Using UK Biobank data, our aims are:
- To confirm previous results of variations in the IAPP gene in relation to fracture risk, bone mineral density and bone size.
- To estimate both the observational and the unconfounded, potentially causal association between T2DM and bone measures and risk of fracture using a Mendelian randomization approach.
Our proposed project deals with different aspects of T2DM and fracture risk, two common health problems in the general population with suffering on the individual level and high health care costs on the societal level. Persons with T2DM have a higher risk of fracture although the mechanisms remain largely unknown. By using data from the UK Biobank and applying regular epidemiological as well as molecular epidemiological methods, we aim to provide knowledge on the diabetes-fracture association that may in the future lead to better identification of persons at risk and ultimately prevention. The research will be conducted using standard statistical methods for the association between prevalent T2DM or genetic variation in the IAPP gene measured as single nucleotide polymorphisms (SNPs) and fracture risk and bone health. We will further use information on SNP variation in genes related with diabetes, glucose and insulin metabolism as instrumental variables in a ‘Mendelian randomization’ approach to assess the potentially unconfounded and causal effect of them on fracture risk and bone health. We will also analyse the reverse: whether genetic variants related with fracture risk and bone health are causally related with diabetes. We would require the full cohort for our analysis on T2DM and incident fractures. We would also require data from all individuals with genetic information (for the SNPs of interest) and from all individuals with DXA measurements.