Abstract

Given the cost and difficulties of bringing a new drug to market, repurposing of existing drugs presents an attractive proposition for new therapies. Metformin is an established treatment for diabetes which also appears to be associated with improved cardiovascular outcome. The exact mechanisms for this are unclear, but may be in part due to reductions in insulin resistance (IR) and inflammation. Both of these processes are linked with the development of adverse left ventricular remodelling, in particular, left ventricular hypertrophy (LVH), an independent cardiovascular risk factor. The aim of this series of studies was to utilise a translational approach involving ‘big data’, record linkage to a bio-resource, genomics and a proof-of-concept randomised controlled trial to determine whether metformin could be repurposed as a treatment for LVH. Three studies were conducted, first, a genomic study to determine whether genetically-determined IR was independently associated with echocardiographic LVH in patients with AS. Second, a large cohort study was analysed to determine whether metformin use was associated with improved cardiovascular outcome in diabetic patients with moderate/severe AS compared to non-metformin users. Finally, in a proof-of-concept study to assess the potential of metformin to cause regression of LVH, a randomised controlled trial of metformin vs. placebo was conducted in non-diabetic patients with IR and coronary artery disease without hypertension.

After adjustment for AS severity, age, sex and systolic blood pressure (SBP), patients in the highest tertile of IR genetic risk were significantly more likely to have LVH than those in the lowest tertile (OR 1.71; 95% CI 1.13–2.61, p=0.012). In the longitudinal cohort study, although diabetic patients overall had a worse outcome than non-diabetics, metformin users actually had similar outcomes to non-diabetic patients (HR=1.24, 95% CI 0.84–1.82, p=0.28) whereas diabetic patients on drugs other than metformin had significantly higher risk than non-diabetic patients (HR 1.70; 95% CI 1.23–2.34, p=0.0012) after adjustment for relevant clinical variables including AS severity, duration of diabetes, medication use and SBP. Finally, in the randomised trial, 12 months of metformin treatment reduced indexed LVMI compared to placebo (intention-to-treat: −2.7±2.3 g/m 1.7 vs. −1.4±2.7 g/m 1.7; p=0.051; per-protocol: −3.1±1.9 g/m 1.7 vs. −1.2±2.7 g/m 1.7; p=0.005).

By using a comprehensive translational approach including ‘big data’, genomics and a randomised clinical trial, these results suggest that metformin may have the potential to be repurposed as a treatment for cardiovascular disease and warrants further prospective trials to evaluate its use.