The Origin and Role of Thromboembolism in the Pathogenesis of Ischaemic Stroke

Stroke remains the second leading cause of death and the commonest cause of dependency in adults worldwide. The majority of strokes (80%) are due to cerebral ischaemia with a minority caused by subarachnoid (5%) or intracerebral (15%) haemorrhage.

Some stroke classification systems are based on the presumed mechanism, such as the Trial of Org 10172 in Acute Stroke Treatment (TOAST), the Causative Classification System (CCS) and the Atherosclerosis Small Vessel Disease Cardiac Source Other cause Dissection (ASCOD) classifications. These systems assign individual patients to cardioembolic, atherothromboembolic, lacunar (due to intrinsic disease of the cerebral perforating arterioles), other, or uncertain causes of stroke. However, a substantial proportion of patients (up to 25%) remain in the 'undetermined' category, because they have several potential overlapping causes, are incompletely investigated or are 'cryptogenic' where no cause has been identified.

Cryptogenic Stroke:

Cryptogenic stroke may have an embolic origin, supported by a pattern of brain infarction typically seen in patients with a definite embolic source: mainly cortical and affecting multiple vascular territories. In a systematic review, 1 in 6 ischaemic strokes were cryptogenic (\~17%; range 9 to 25%) and in the Oxford Vascular Study of 2555 patients with a first ischaemic stroke, 1 in 3 were cryptogenic. Unfortunately, these patients have a high risk of recurrent ischaemic stroke (\~25% at 5 years) that is comparable to those with known causes of stroke, indicating the need for better understanding and treatment of cryptogenic stroke. Two large randomised controlled trials compared direct oral anticoagulant therapies with aspirin to prevent stroke recurrence in patients with cortical cryptogenic strokes, testing the hypothesis that many of these strokes were due to an unidentified cardioembolic source (particularly paroxysmal atrial fibrillation). However, anticoagulant therapies were not superior to aspirin in preventing recurrent stroke and carried an increased risk of bleeding. Thus, the optimal preventative strategy in patients with cryptogenic stroke remains open for further exploration, and likely requires better patient stratification (i.e. selecting out those with occult thromboembolism) and understanding of its pathogenesis.

Lacunar Stroke:

As well as the uncertainty existing around causes and management of cryptogenic stroke, the aetiology and best management remains poorly understood in another common and important subtype of stroke, lacunar stroke. Lacunar stroke accounts for 25% of all ischaemic strokes and most haemorrhagic strokes in patients aged over 65 years, but the understanding of its pathophysiology remains limited. Epidemiological studies indicate that emboli are uncommon in lacunar ischaemic stroke. While it is physically possible for emboli to enter the lenticulostriate arteries, laboratory studies suggest that this is infrequent. Some larger lacunar infarcts in the basal ganglia may be due to atheromatous occlusion of a lenticulostriate artery or an embolus entering a lenticulostriate artery from the middle cerebral artery. However, in the few studies where embolic sources were actively sought, the percentage of patients with lacunar stroke and a source of emboli was only 11%. Cryptogenic stroke definitions sometimes exclude patients with brain imaging showing lacunar (small subcortical) infarctions, but this means that embolic sources may be missed through presumed non-embolic causes of lacunar stroke. Given that management of cardioembolic stroke is quite different to that for atherothromboembolic stroke and that intrinsic small vessel disease appears to need an alternative management strategy, there is an urgent unmet clinical need to determine the actual underlying cause of an ischaemic stroke in many patients.

Determining Stroke Aetiology and Treatment:

The mechanisms underlying ischaemic stroke can be divided into cardio-embolic (including paradoxical embolism, such as deep vein thromboses crossing patent foramen ovale), artery-to-artery atherothrombotic embolism, extra- or intra-cranial large artery atherosclerosis with acute superimposed occlusion, intrinsic small vessel disease (lacunar) and arterial inflammation.

Following acute stroke, patients undergo diagnostic brain imaging with computed tomography (CT) or magnetic resonance imaging (MRI). CT or MRI angiography may identify a point of medium to large arterial obstruction or atheromatous ipsilateral carotid stenosis as a likely source of embolism, but they are commonly unable to show small emboli or to visualise directly the origin of the thromboembolus. Patients will also undergo clinical investigations to identify potential causes of stroke including blood pressure measurement, clinical biochemistry, electrocardiogram, echocardiography, and carotid and vertebral artery imaging to identify treatable symptomatic carotid stenosis or other pathologies such as dissection, usually with Doppler ultrasound, CT or MRI angiography or prolonged electrocardiogram monitoring. In younger patients (\<65 years) or where other aetiologies are suspected, additional investigations may include genetic or metabolic screens and echocardiography. Most patients will be commenced on lifelong anti-platelet, statin and antihypertensive therapies. If atrial fibrillation or flutter is detected, they will be considered for anticoagulant treatment with direct oral anticoagulants or warfarin therapy. Patients thought likely to have an embolic stroke but in whom no embolic or other relevant source is found, are considered at this point to have cryptogenic stroke, and currently, they receive the same treatment as for non-cardioembolic stroke. Given their high risk of stroke recurrence, it raises the question as to whether a new imaging technique could identify stroke aetiology and the origin of thromboembolism, thereby improving patient management.

Positron emission tomography and computed tomography:

Positron emission tomography combined with CT (PET-CT) fuses anatomical with functional imaging which can be tailored to a specific disease process depending on which radiotracer is used. Recently, the investigators have used the tracer 18F-sodium fluoride to study microcalcification in a range of cardiovascular disease processes, successfully highlighting disease activity in coronary artery plaque, carotid artery plaque, abdominal aortic aneurysms and calcific aortic stenosis. However, such techniques can be applied to other pathophysiological processes depending on the properties of the radiotracer. The great strength of PET is its exquisite sensitivity and ability to detect even small areas of disease activity.

Platelet Biology and 18F-GP1:

The activation and deposition of platelets are major contributors to human thrombus formation, especially in the arterial circulation. The glycoprotein IIb/IIIa receptor is expressed on activated platelets and is key in the fibrin-crosslinking process of platelet aggregation. It is also the target for anti-platelet therapy used in routine clinical practice for high-risk percutaneous coronary intervention. The thrombus tracer, 18F-GP1, is a derivative of elarofiban and has a high and specific binding affinity for the activated glycoprotein IIb/IIIa receptor. In studies of Cynomolgus monkeys, 18F-GP1 binds to acute venous and arterial thrombi. It has recently undergone preliminary human clinical studies confirming it is a highly sensitive method of identifying in vivo arterial and venous thrombosis. The investigators have undertaken preliminary studies using 18F-GP1 and have demonstrated that it has excellent in vivo binding properties which enable detection of intravascular thrombosis in a range of conditions including left ventricular thrombus following myocardial infarction, pulmonary thromboembolism, deep vein thrombosis and coronary thrombosis.

Study Aims:

The aims of this project are to establish the contribution of activated platelets to the mechanism of various subcategories of ischaemic stroke. The investigators will establish the frequency and distribution of activated platelets in thromboemboli of patients with ischaemic stroke including those with cryptogenic stroke and lacunar stroke. This will inform the pathophysiology and mechanism of stroke subtypes and establish sources and origins of platelet activation.

Research Hypothesis:

In patients with ischaemic stroke, the investigators hypothesise that non-invasive 18F-GP1 imaging will:

1. Identify the origin and contribution of activated platelets to thromboembolism in patients with ischaemic stroke.
2. Establish the frequency and distribution of activated platelets in subtypes of ischaemic strokes.
3. Have the potential to influence management of patients with stroke.

Rationale for Study:

1. Origin and source of thromboembolism: To date, some techniques and clinical assessments make assumptions about the source of thromboemboli. Moreover, an important proportion of patients have no clear source of embolism or cause for their stroke. For example, strokes that occur patients in atrial fibrillation are often assumed to have thrombus from the atrial appendage or a valve. However, it is unknown whether this is the situation in all cases and what is the contribution of activated platelets to thromboembolic events.
2. Cerebrovascular thrombosis in cryptogenic and lacunar strokes:In a substantial percentage of acute strokes, there is real uncertainty regarding the presence or contribution of activated platelets and thrombosis to cerebral infarction. This is the case for patients with cryptogenic stroke and especially for lacunar stroke. Indeed, there is debate regarding the contribution of activated platelets and embolic or in situ thrombus (the latter might be an endstage event on a damaged arteriolar endothelium) to lacunar strokes and this has been hard to resolve because of the difficulties of visualising activated platelets and thrombus at multiple locations with current non- invasive image techniques. This study will assess patients with cryptogenic and lacunar strokes to establish the evidence of whether activated platelets and thrombosis plays a role, and if so, in what proportion of patients this occurs and whether there are any dominant associated clinical features.