A fifth of ischemic stroke or transient ischemic attack (TIA) patients will have recurrent events within the first 3 months [Refs 1-3] despite aggressive medical therapy with antiplatelets and risk factor control.

Clopidogrel is one of the mainstays of antiplatelet secondary prevention therapy in patients with ischemic stroke. CYP2C19 loss of function (LOF) mutations impair the effectiveness of clopidogrel [Ref 4]. The prevalence of LOF mutations is approximately 60% in the local population [Ref 5], rendering the effectiveness of empiric clopidogrel treatment doubtful. For patients who have LOF mutations, other treatment options for secondary prevention of ischemic stroke need to be tested. This study aims to determine the feasibility and clinical impact of genetic testing guided antiplatelet therapy in ischemic stroke patients on the prevention of major adverse cardiovascular or cerebrovascular events.

Clopidogrel naive ischemic stroke or TIA patients aged 21 years and above will be randomised to genetic testing guided antiplatelet therapy or standard medical therapy within 7 days of their index event. Patients allocated to testing group will have blood sample drawn for diagnosis of CYP2C19 LOF mutations. Patients who test positive for an LOF mutation (intermediate and poor metabolisers) will be offered alternative antiplatelet therapy in the form of aspirn (for those who need monotherapy) or aspirin plus ticagrelor or dipyridamole (for those who need dual antiplatelet therapy) to be decided by the managing physician. Patients who test negative for LOF mutation will continue on clopidogrel. Platelet reactivity index (enables the identification of patients with an inadequate response to antiplatelet agents) will be measured at baseline.

This is a multicentre, prospective, randomised, open blinded endpoint trial.

The study will include three visits: Baseline (within 1 week of index event), 3 months, and 12 months. Randomisation will be performed for patients with unknown CYP2C19 status while no randomisation will be done for patients with known CYP2C19 status at the time of recruitment. Baseline visit will need to be in-person, whle the 3-month and 12-month follow-ups can either be inperson or via telephone. Data on antiplatelet regimen, MACCE outcomes, and GUSTO outcomes will be collected at the follow-up visits. Blood tests will be performed on the genetic testing guided antiplatelet therapy arm and standard medical therapy arm at baseline to determine the platelet reactivity index (PRI).

Patients with unknown CYP2C19 status will be randomised to genetic testing versus standard medical therapy in a 1:1 ratio, via a block randomisation process. The block randomisation sequence will be generated by a web-based randomisation service in blocks of four for each recruitment site. Blinding is not applicable for the treatment assigned.

Patients randomised to testing will have blood drawn for testing of the CYP2C19 LOF mutations. Physicians of patients who test positive for an LOF mutation (intermediate and poor metabolizers) will be notified of the mutation with recommendations of possible alternative antiplatelet regimens suggested. Aspirin will be the recommended monotherapy, and aspirin in combination with ticagrelor or dipyridamole will be the recommended dual antiplatelet regimen. The decision of alternative medications used in patients with LOF mutations will be left to the discretion of the primary physician. Patients in the standard medical therapy group, and patients in the genotype guided antiplatelet therapy group who do not have LOF mutations (normal metabolisers) will be left to continue the original intended antiplatelet regimen - this may be clopidogrel monotherapy, or in combination with aspirin. Patients who experience recurrent events during follow-up, or who have drug intolerance can be changed from the antiplatelet regimen as per primary physician's discretion. Information on all medications, including antiplatelet regimen at discharge and at subsequent reviews will be collected.

There will be no randomisation for patients with known CYP2C19 status prior to recruitment. Patients with known CYP2C19 status will be recruited as a comparison arm for outcomes measurements.

CYP2C19 genotyping will be performed by the Singapore General Hospital's clinical laboratory services. Turnaround time for CYP2C19 is approximately 7 days. Patient will be notified of CYP2C10 LOF mutation result.

Demographic characteristics that will be ascertained includes age, sex, ethnicity, education level, date of study randomisation. Past medical history will be recorded: hypertension, diabetes, hyperlipidemia, ischemic heart disease, atrial fibrillation, smoking, peripheral vascular disease, previous ischemic stroke/transient ischemic attack, intracranial bleeding, clinically significant bleeding, smoking and alcohol. Information pertaining to hospital admission, stroke characteristics, medications that the patients are on, particularly antiplatelet, anticoagulation, proton pump inhibitor, histamine H2 receptor antagonists, and cardiovascular medication prior to the admission and at discharge will be assessed. Use of single versus dual antiplatelet therapy, and duration of dual antiplatelet therapy will be considered confounders.

For subjects randomised to the standard medical therapy arm, genetic testing for CYP2C19 LOF mutations may be performed for clinical purpose after randomisation and before the end of subject's participation in the study, which could result in treatment changes. Such cases will not be considered as a protocol violation. These subjects will continue with treatment as per advised by the primary physician and may continue study follow ups till the end of the study.

If subjects are identified after randomisation to have cardioembolism or prothrombotic state necessitating the use of anticoagulation, the treating physician should manage as per standard medical care. These will not be considered as protocol violation as the diagnoses were made after randomisation. As these subjects will no longer be on antiplatelet therapy, there is no need to continue with study follow up and this will be deemed as withdrawn from study participation.

Subject may withdraw his/her consent from study participation voluntarily at any time without prejudice to him/her or effect on the subject's medical care.

The Principal Investigator may stop a subject's participation in the study at any time for justified reasons. If a subject withdraws voluntarily from participation in the study for any reason, the data that have been collected until the time of his/her withdrawal will be kept and analysed. The reason is to enable a complete and comprehensive evaluation of the study. Standard medical care will continue for all subjects who withdrew from the study.

Inclusion Criteria:

• Patients diagnosed with acute ischemic stroke or transient ischemic attack.
• Age 21 years to 100 years.
• Can be randomised within 7 days of onset of index event [Refer to footnote 1].
• Clopidogrel naive immediately prior to index event [Refer to footnote 2].

Footnote 1: Date of index event to be taken as Day 1. This means that randomisation must be done by Day 7. For unknown onset or wake-up stroke where the last seen well and symptoms discovery are on different days, Day 1 is taken to be the date of symptoms discovery. Footnote 2: Patients who were on short-term antiplatelets which included clopidogrel but are no longer on clopidogrel prior to stroke would fulfil for this inclusion criteria. In this context, short-term antiplatelets are defined as 21 days for minor stroke TIA and 3 months for large vessel disease.

Exclusion Criteria:

• Known diagnosis of dementia [Refer to footnote 3].
• Known diagnosis of a life limiting illness with life expectancy of less than 1 year.
• Known cardioembolism or prothrombotic state necessitating the use of anticoagulation, or having a contraindication to clopidogrel.

Footnote 3: "Known diagnosis of dementia" will be defined as clinical diagnosis of dementia prior to the index stroke event as indicated in the patient's medical records.

Patients will be randomised to genetic testing versus standard medical therapy in a 1:1 ratio, via a block randomisation process. The block randomisation sequence will be generated by a web-based randomisation service in blocks of four for each recruitment site. Blinding is not applicable for the treatment assigned.