Back

Using genetics to understand shared mechanisms between hypertension and type 2 diabetes accounting for adiposity

Voller, B.; Woodward, R.; Gu, Q.; Murrin, O.; Delgado, J.; Violan, C.; Khalid, S.; Türkmen, D.; Fox, C.; Lamb, S. E.; Mancini, M.; Farmer, L.; Boddy, K.; Dudbridge, F.; Bowden, J.; Frayling, T. M.; Masoli, J. A.; Pilling, L. C.; the GEMINI Consortium,

2025-10-07 genetic and genomic medicine
10.1101/2025.10.03.25337275 medRxiv
Show abstract

Hypertension and type 2 diabetes (T2D) frequently co-occur, but their shared mechanisms are not fully understood. We investigated diagnostic patterns in 3.6 million primary care records, finding that hypertension precedes T2D by a median of 7.5 years (IQR 3.1-13.4), suggesting a crucial intervention window. Using data from large genome-wide studies we found substantial genetic overlap between hypertension and T2D (genetic correlation=48%, 95%CI 45-52). This overlap was partially attenuated after accounting for the genetic contributions of BMI, waist-hip ratio (WHR), and triglycerides (genetic correlation=29%, 95%CI 24-34), highlighting the role of adiposity but also independent shared mechanisms. Bidirectional causal effects were supported in Mendelian randomisation analysis. In colocalisation analysis using genetic data after adjusting for BMI, WHR and triglycerides, we identified four independent shared genomic regions, including an allele within PCSK7 that is associated with risk of both T2D and hypertension and circulating PCSK7 protein levels. Our findings highlight a key opportunity to intervene and reduce hypertension and T2D co-occurrence, and identify shared causal mechanism for both conditions independent of BMI, waist-hip ratio, and triglycerides. The potential novel mechanisms identified, in particular PCSK7, could uncover novel interventions or opportunities for prevention of this co-occurring condition pair.

Matching journals

The top 6 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.