The CACHE Study: Coronary Artery Care in HaEmophilia

Background and Rationale

Haemophilia, ageing and cardiovascular disease It is only relatively recently, due to the development of both effective and safe replacement therapies, that there is an aging population with haemophilia. This represents new challenges with respect to managing comorbidities. Cardiovascular disease (CVD) is the leading cause of death globally in the general population, and as people with haemophilia (PWH) age they are also at risk of developing CVD which poses particular challenges due to the opposing risks of bleeding, from haemophilia and antithrombotic treatments, versus thrombosis. Better understanding of the risks of atherosclerosis in PWH is crucial in order to best support aging PWH; and it has been identified as a research priority by patients and healthcare professionals.

Although mortality from cardiovascular disease in PWH has historically been about 40% less than the general population, it remains unclear whether this is because people with haemophilia develop atherosclerosis more slowly than the general population or whether deficiency of FVIII/FIX protects from formation of an occlusive thrombus at time of atherosclerotic plaque rupture. As FVIII is produced and stored within endothelial cells lining the blood vessels and FIX is produced in the liver, it may also be that FVIII and FIX deficiencies have different effects on the rates of atherosclerosis.

The most informative studies of atherosclerosis in PWH to-date used ultrasound to measure carotid and femoral intima-media thickness and computed tomography (CT) to derive coronary artery calcium scores. The first study measured carotid and femoral intima-media thickness (IMT) and brachial flow-mediated dilatation (FMD) as markers of atherosclerosis and endothelial dysfunction respectively in 51 obese PWH, and 42 obese and 50 matched non-obese male patients. All the PWH had haemophilia A (33% severe, moderate 16%, mild 50%), HIV was excluded, and the mean age was 50 +/-13 years. Carotid IMT was increased in obese as compared with non-obese subjects, but no difference was found in mean carotid and femoral IMT between obese PWH and obese control individuals. Thirty-five per cent of the obese PWH and 29% of the obese controls had an atherosclerotic plaque, irrespective of the severity of haemophilia. Brachial FMD was comparable between obese PWH and obese controls. The second study evaluated the presence and extent of atherosclerosis by coronary artery calcification score (CACS) derived from computed tomography and carotid IMT in 69 PWH (51 haemophilia A, 18 haemophilia B; 40% severe, 11% moderate, 49% mild). This again showed that CACS and carotid IMT were similar to controls and that the extent of atherosclerosis was related to traditional cardiovascular risk factors. Although these studies suggest PWH develop atherosclerosis at a similar rate to the general population; they are severely limited by the quality of the imaging techniques, patient numbers and relatively young patient age (average 50 years).

Cardiology guidelines recommend using a risk score to calculate cardiovascular risk over time and engaging patients in discussing modifiable risk factors. These scores have been used in PWH. A Dutch/UK cohort used both the QRISK(R)2 and SCORE algorithms. The predicted 10-year QRISK(R)2 risk was significantly higher in PWH than in the general population (8.9 vs. 6.7%), indicating more unfavourable cardiovascular risk profiles; and the increased risk became apparent after the age of 40 years. In contrast a study of 100 PWH in Europe compared to 200 aged-matched controls, found no significant difference in the 10-year cardiovascular mortality risk >10% between PWH and controls using SCORE. Other studies have looked at individual cardiac risk factors and PWH had higher rates of hypertension than expected for the general population. Critically whether the cardiac risk algorithm conveys the same actual CVD risk and is valid in PWH is not known.

Oxford University Hospitals has the ability to explore CVD risk and atherosclerosis in haemophilia with a proof of concept project due to having both a large haemophilia centre and the only UK photon-counting cardiac CT scanner.

Photon-counting cardiac CT

Coronary computed tomography angiography (CCTA) is a sensitive and widely used, non-invasive imaging modality for diagnosing coronary artery disease (CAD) and recent clinical guidelines have expanded its use as a first-line diagnostic tool in the management of chest pain of recent onset, suggestive of typical/atypical angina or non-anginal chest pain with electrocardiographic changes.

However, conventional CCTA identifies only anatomically significant coronary artery stenoses which are found in \<50% of all patients referred for this test. Importantly, the majority of acute coronary syndromes are caused by unstable but non-obstructive atherosclerotic plaques, which cannot be identified by conventional CCTA or other non-invasive diagnostic tests detecting coronary luminal stenosis or stress-induced myocardial ischaemia. In the normal population, early detection of cardiovascular disease and subsequent optimisation has been associated with reduced cardiovascular mortality.

This has led to the development by our study investigators of the perivascular fat attenuation index (FAI or FAIPVAT), a novel imaging biomarker which detects vascular inflammation in the human coronary arteries using images from routine CCTA scans and is predictive of cardiovascular risk. The FAI relies on the discovery that inflammatory signals released by the diseased coronary artery, inhibit adipogenesis and trigger lipolysis in the perivascular adipose tissue (PVAT), shifting perivascular CT attenuation from the lipid towards that of the aqueous phase. Vascular inflammation is a driver of coronary atherosclerotic plaque formation, and a characteristic feature of atherosclerotic plaque rupture, leading to acute coronary syndromes including myocardial infarction. The causal role of inflammation in vascular disease pathogenesis has recently been documented by the CANTOS trial, which demonstrated that specific targeting of residual inflammatory risk improves clinical outcomes.

In the CRISP-CT study, recently published in the Lancet, our investigators demonstrated that FAIPVAT predicts all-cause and cardiac mortality over and above clinical risk factors and the current interpretation of CCTA. These findings are validated in two large and substantially different "real-life" prospective cohorts of patients undergoing clinically-indicated CCTA, one in Europe and the other in the United States, demonstrating similar predictive value.

Oxford University Hospitals has a newly-installed state of the art, photon-counting CT scanner, capable of performing CCTA to assess atherosclerotic burden and peri-vascular FAI. By incorporating an updated detector, single photons can be converted directly into electrical signals, removing the need for a two-step conversion process which is used in conventional CT scanners. This allows for higher resolution images than conventional CT at a given radiation exposure and therefore offers the opportunity to assess in detail the cardiovascular risk associated with haemophilia.

Oxford Haemophilia Centre

The Oxford Haemophilia Centre is one of the largest in the UK. The haemophilia team look after 190 PWH over the age of 40 years (range 40-88 years), of which 146 are over the age of 45 years, and 117 over the age of 50 years. Of PWH >45 years: 45 have severe haemophilia, 14 moderate haemophilia, 87 mild haemophilia; 77% have haemophilia A. Comorbidities are common and approximately 50% have hypertension and 7% atrial fibrillation. In those >65years, 6% have known ischaemic heart disease, 6% history of ischaemic stroke, and 4% peripheral vascular disease.

In order to help answer the question of whether PWH develop atherosclerosis at a similar rate to the general population with similar known cardiac risk factors, and whether there is a difference in the stability of the plaques, the investigators will use this state-of-the-art cardiac imaging technique to look in detail at atherosclerosis in aging patients with haemophilia and compare how this imaging data correlates with standard cardiovascular risk scores such as QRISK. As a comparison to general population, imaging data and standard cardiovascular risk factors will be compared to anonymised control scans from other studies at the University of Oxford with matching for age, gender, and exposure to cardiovascular risk factors (smoking, hypertension, high cholesterol, diabetes, family history). These data could pave the way for future multi-centre clinical studies with increased numbers of PWH (especially through close regional haemophilia centres and potentially reducing age of screening), extension to other common bleeding disorders such as von Willebrand disease, and ultimately impact on clinical care for people with bleeding disorders (cardiac screening); as well as further translational research studies into the role of factor VIII and factor IX in the pathogenesis of atherosclerosis.