Precision medicine to improve the outcome of patients with heart failure with preserved ejection fraction

Research questions Is it possible to reclassify heart failure with preserved ejection (HFpEF) into distinct diagnoses and do these distinct groups respond differentially to treatments? Is it possible to identify phenotypic and genetic factors that could be used to predict who will develop HFpEF? Background Heart failure occurring in the presence of a normal, or preserved, left ventricular ejection fraction (HFpEF) is 'the single largest unmet need in cardiovascular medicine'. Rather than being a single diagnosis, HFpEF represents a heterogeneous syndrome involving a range of pathophysiological mechanisms, clinical factors and outcomes. However, to-date, HFpEF has been considered as a single disease entity. This 'one-size-fits-all' approach is a major reason why we do not know who will develop HFpEF, why there are no treatments, and why quality of life and prognosis remain so poor. Aims and objectives I aim to: Reclassify HFpEF into distinct diagnoses based on disease mechanisms, clinical factors and outcome, and evaluate whether the distinct groups respond differentially to treatments. Identify phenotypic and genetic factors that could be used to predict who will develop HFpEF. Establish and lead a UK National HFpEF registry that will be a platform for collaborative UK clinical and translational HFpEF research. Methods Single-centre cohort I will apply machine learning techniques to a prospective cohort of 1,500 deeply phenotyped patients with HFpEF in order to derive and validate subgroups of HFpEF and evaluate differences in treatment response. I will compare the subgroups with matched patients without HFpEF in order to identify phenotypic complexes that distinguish HFpEF and investigate the role of genetics in HFpEF development. National HFpEF registry Using Part 1 as a demonstrator, I will establish and lead a National HFpEF registry. The registry will provide deep phenotyping linked to outcomes at scale, be a platform for trials matching drug mechanism of action with HFpEF subgroup/anticipated treatment response, and provide simplified UK-wide access for industry. The NIHR-BHF Cardiovascular Partnership have agreed to make this a national priority study, and the BHF Data Science Centre will adopt it as a vanguard project. More than ten UK centres have agreed to participate. NIHR Health Informatics Collaborative (HIC) In parallel to Parts 1 and 2, I will apply similar machine learning techniques to routinely collected electronic health record data held in the NIHR Cardiovascular HIC, which provides less detailed patient characterisation but much greater numbers (>250,000 patients). Timelines for delivery Fourteen milestones including: National registry operational month 19, HIC analysis complete month 22, recruitment to single-centre cohort complete month 24. Anticipated impact and dissemination This fellowship is an exciting opportunity to deliver personalised care to patients with HFpEF that will provide the basis for tailored therapies, improved risk stratification and potentially preventative interventions. It will open new avenues of clinical and biological research, provide an industry platform and reduce strain on NHS and Government budgets. Dissemination will be via my Patient Advisory Group, lay written and video summaries, websites, patient group presentations, high-impact open-access publications, conference presentations, a dedicated symposium and media outlets.