Primary Supervisor: Dr Elias Allara, HDR UK Clinical Research Fellow, ea431@cam.ac.uk.
Supervisory Team: Professor Emanuele Di Angelantonio, Professor of Clinical Epidemiology, Professor Angela Wood, Professor of Biostatistics and Health Data Science.
Project Outline:
Healthy ageing is the maintenance of functional ability and quality of life across the life course. As populations age rapidly worldwide, promoting healthy ageing has become essential to ensure the sustainability of healthcare systems. A key determinant of healthy ageing is the avoidance or delay of complications arising from existing long-term conditions. In high-income settings, cardiovascular diseases (CVDs) are among the leading causes of morbidity and mortality and frequently complicate common non-cardiovascular long-term conditions (non-CVDs) such as diabetes and chronic kidney disease. Although effective CVD prevention could substantially improve healthy ageing in individuals with pre-existing non-CVDs, current prevention strategies may insufficiently address those non-CVDs for which causal associations to CVD risk are less well established – including several chronic inflammatory conditions and severe mental illnesses – an uncertainty that is especially pronounced in minority ethnic groups.
An illustrative example is rheumatoid arthritis, where patients show an increased risk of CHD and stroke not fully explained by traditional cardiovascular risk factors such as dyslipidaemia, hypertension and thrombosis. Research into the aetiology of this association has identified interleukin-6 (IL-6) as a proatherogenic factor, leading to the investigation on the cardiovascular benefits of ziltivekimab, a novel IL-6 inhibitor, in a large-scale trial (ZEUS, NCT05021835). Some IL-6 inhibitors such as tocilizumab and sarilumab are already approved for the treatment of rheumatoid arthritis in the UK, raising the prospect of potentially prioritising them in rheumatoid arthritis patients who are at risk for CHD and stroke. This could (i) reduce translation time, potentially providing preventative cardiovascular benefits to at-risk rheumatoid arthritis patients without needing separate clinical trial; and (ii) reduce the need for polypharmacy, potentially lowering the chances of harmful drug-drug interactions while maintaining the benefits of cardiovascular prevention.
These benefits can be further amplified by leveraging multi-ethnic and multi-ancestry datasets, enabling the discovery of pathways that are particularly beneficial in specific population groups. This has the potential to ultimately help address the marked cardiovascular health inequalities noted in contemporary whole-population datasets (e.g. Allara, Lancet Public Health, 2025). Together, these strategies have the potential to prevent CVD complications, preserve cardiovascular function, and thereby promote healthy ageing in a large and increasingly diverse patient population.
Project Plan:
In this PhD project, the student will harness the power of whole-population and multi-omics data to systematically assess associations and mediation pathways between >300 non-CVDs and two major CVDs – coronary heart disease (CHD) and stroke.
The project will address two key research aims:
- Identify non-CVDs causally associated with CHD and stroke.
Using large-scale, multi-ancestry population datasets from >70 million people (i.e. OpenSAFELY, SAIL, Human Technopole/Regione Lombardia) and multi-omic datasets from >2 million participants (e.g. Our Future Health, All of Us, the Million Veteran Program, UK Biobank), the project will integrate observational and Mendelian randomization (MR) analyses to establish putative causal associations between >300 non-CVDs and incident CHD and stroke. The inclusion of diverse ancestry groups will ensure equitable representation and enhance generalisability. - Characterise molecular mediators linking non-CVDs with CHD and stroke. Building upon findings from the first aim, the project will perform systematic mediation analyses to identify molecular and metabolic traits – spanning routinely-used clinical biomarkers and -omics traits – that explain the associations between specific non-CVDs and either CHD and stroke. Structural equation modelling and multivariable MR will be employed to disentangle complex causal pathways and prioritise modifiable mediators for therapeutic targeting.
The student will be able to build on the supervisory team’s expertise analysing whole-population datasets (e.g. Allara, Lancet Public Health, 2025 and Kerr, Lancet, 2024), conducting multi-omics analyses (e.g. Karjalainen, Nature, 2024, and Gaziano, Nature Medicine, 2021) and investigating the genetic, biological and behavioural determinants of ageing (e.g. Codd, Nature Genetics, 2021, and Bountziouka, Lancet Healthy Longev, 2022).
Main Methods to be Used:
We plan to use three main methods:
- Observational analyses in whole-population datasets to quantify associations between >300 non-CVDs and either CHD or stroke, using linked electronic health records for >70 million adults from England, Wales and Italy, adjusted for potential confounders.
- Mendelian randomization, which uses genetic variation as a "natural experiment" to test whether a risk factor is causally associated with disease, will explore potential causality using summary statistics from multi-ancestry genome-wide association studies in >2 million participants from Our Future Health, All of Us, the Million Veteran Program, and UK Biobank.
- Mediation analyses, a statistical approach for quantifying the pathways connecting an exposure (e.g. a non-CVD) with an outcome (e.g. CHD or stroke), will leverage clinical biomarkers, proteomic, and metabolomic data from UK Biobank and other studies to uncover biological pathways underpinning causal associations.
Whenever feasible, analyses will be stratified by ethnicity/ancestry, sex and age.
Key References:
- Allara E et al. Lancet Public Health. 2025 Nov;10(11):e943-e954
- Kerr S et al. Lancet. 2024 Feb 10;403(10426):554-566
- Karjalainen MK et al. Nature. 2024 Apr;628(8006):130-138
- Gaziano L et al. Nat Med. 2021 Apr;27(4):668-676.
- Codd V et al. Nat Genet. 2021 Oct;53(10):1425-1433
- Bountziouka V et al. Lancet Healthy Longev. 2022 May;3(5):e321-e331