Cost-effectiveness of testing for CYP2C19 loss-of-function carriers following transient ischemic attack/minor stroke: A Canadian perspective

Abstract

Background:

The CHANCE-2 study compared 3 weeks of aspirin–ticagrelor to aspirin–clopidogrel in CYP2C19 loss-of-function (LOF) allele carriers following a transient ischemic attack (TIA)/minor stroke and demonstrated a modestly lower risk of stroke recurrence with aspirin–ticagrelor. This stroke protection was largely for minor stroke and came at an increased risk of bleeding. The cost-effectiveness of implementing testing for LOF allele status to personalize antiplatelet regimen for secondary stroke prevention after a TIA/minor stroke in the Canadian health care context is unknown.

Methods:

Cost-effectiveness analysis using a decision-analytic Markov cohort model with a lifetime horizon was performed to determine the costs and health benefits of testing for LOF allele status compared with no testing (current standard of care). The population of interest was patients living in Canada who suffered a TIA/minor stroke. Outcomes of interest were life-years gained (LYG), quality-adjusted life years (QALY) gained, costs (reported in 2022 Canadian dollars), and the incremental cost-effectiveness ratio (ICER). We adopted the perspective of the Federal, Provincial, and Territorial Ministries of Health and used a 1.5% annual discount rate. Sensitivity analyses were performed to assess uncertainty.

Results:

Compared to standard of care, LOF allele testing leads to 0.14 LYG (undiscounted), 0.12 QALYs gained (undiscounted), and additional lifetime costs of CAD$432 (discounted) per patient. The ICER of the LOF allele testing strategy is CAD$4310 per QALY gained compared with standard of care. The probabilistic sensitivity analyses demonstrated that LOF allele testing was cost-effective in more than 99.99% of simulations using a willingness-to-pay threshold of CAD$50,000 per QALY.

Conclusion:

Based on available evidence, testing for LOF allele followed by short duration 3 weeks of aspirin–ticagrelor compared to standard-of-care aspirin–clopidogrel can lead to prolonged life and improved quality of life and can be considered very cost-effective when compared with other well-accepted technologies in health and medicine.

Keywords

Aspirin clopidogrel cost-effectiveness stroke ticagrelor transient ischemic attack

Introduction

There is an increased risk of early stroke recurrence (90 days) after an index transient ischemic attack (TIA) or minor stroke.^1–3 CHANCE (Clopidogrel in High-Risk Patients with Acute Nondisabling Cerebrovascular Events) and POINT (Platelet-Oriented Inhibition in New TIA and Minor Ischemic Stroke) trials showed that the use of dual antiplatelet therapy (DAPT) regimen with aspirin–clopidogrel for the initial high-risk period substantially decreased this risk in Chinese and less restricted ethnic populations, respectively.^4,5 This led to guideline recommendations for use of DAPT using aspirin–clopidogrel for 21–30 days followed by single antiplatelet alone.⁶ However, clopidogrel is a pro-drug which requires conversion into its active metabolite in a two-step process by hepatic cytochrome p450 (CYP) and is known to be less effective in carriers of CYP2C19 loss-of-function (LOF) alleles.^7,8 The prevalence of these alleles is much higher in Asian populations (approximately 60%) as compared to Caucasian populations (approximately 30%).^7,9

Ticagrelor is an alternative antiplatelet drug that does not require metabolic activation and has shown to have greater levels of platelet aggregation inhibition as compared to clopidogrel.¹⁰ The recently published CHANCE-2 trial tested if 3 weeks of DAPT with ticagrelor–aspirin was superior to clopidogrel–aspirin in those who were carriers of CYP2C19 LOF alleles. This trial showed a modestly lower risk of stroke recurrence with ticagrelor as compared to clopidogrel (7.6% vs 6%), and this effect was largest in those presenting with minor stroke (compared to moderate or severe stroke).¹¹ This comes with a trade-off of increased bleeding (albeit minor bleeding), increased rates of discontinuation (due to dyspnea and arrhythmia), and potential lack of generalizability to non-Asian populations where LOF allele prevalence is significantly less.

Although, ticagrelor–aspirin is a reasonable early management for patients presenting with TIA and minor stroke, its clinical adoption in the Canadian population is subjective to its cost-effectiveness. Prior to adoption in routine stroke practice, various key issues need to be considered by clinicians and health policymakers. First, the prevalence of clopidogrel resistance is comparatively low in Caucasian populations and genetic testing for clopidogrel resistance is not routinely available and comes with an upfront cost. Second, the use of ticagrelor does not have insurance coverage in Canada for indication of stroke and thus implies the use of personal savings for an individual. Finally, the CHANCE-2 trial only showed a difference in risk of recurrence of minor strokes between the two groups.¹¹ Minor stroke have minimal hospitalization stay/costs when compared to recurrent moderate–severe stroke.

A cost-effectiveness analysis of LOF allele testing versus standard of care (no testing) incorporates the currently available clinical data and balances the potential benefits against the risks and cumulative costs over the lifetime of patients, following DAPT for TIA/minor stroke. This economic evaluation will aid Canadian health policy decision-makers in determining whether to adopt LOF allele testing at this time or seek further data.

Methods

Study design

We developed a probabilistic Markov state-transition cohort model over a patient lifetime horizon to evaluate the costs and health benefits of CYP2C19 LOF allele testing to personalize DAPT therapy for secondary prevention following a TIA/minor stroke, as per the CHANCE-2 study protocol. As per study protocol, rapid genotyping was performed for two single-nucleotide polymorphisms, CYP2C19*2 and CYP2C19*3. Only patients with at least one LOF allele (*2 or *3) were classified as LOF carriers and were enrolled in the study. Our simulated population of interest was patients living in Canada who suffered a TIA/minor stroke with average age of 65 years at the time of decision.

In the model, CYP2C19 LOF allele carriers are started on aspirin–ticagrelor × 3 weeks with clinical outcomes modeled on the CHANCE-2 study. Ticagrelor 180 mg loading dose is given on Day 1 followed by 90 mg twice daily × 3 weeks followed by single antiplatelet aspirin, as is routinely done clinically. CYP2C19 non-carriers were started on aspirin–clopidogrel, and the clinical outcomes were based on CHANCE-1 substudy.⁸ Clopidogrel 300 mg loading dose is given on Day 1 followed by 75 mg daily × 3 weeks followed by single antiplatelet aspirin. Patients in the no-testing arm (current standard of practice) were all started on aspirin–clopidogrel × 3 weeks, and the clinical outcomes modeled on the CHANCE-2 study. Our rationale for using the CHANCE-2 study outcomes for the non-testing population were as follows: (1) it is a direct methodological comparison to aspirin–ticagrelor in that study, (2) the proportional breakdown of stroke severity is not published in CHANCE-1 (an important clinical nuance for modeling outcomes), and (3) the POINT study, although predominately a North American population, has different methodology and patient population (higher 600 mg loading dose of clopidogrel, shorter time to randomization, 90 vs 21 days use of DAPT, to name a few) and importantly does not report on the breakdown of stroke severity.

Outcomes of interest were life-years gained (LYG), quality-adjusted life years (QALY) gained, lifetime costs (reported in 2022 Canadian dollars), and the incremental cost-effectiveness ratio (ICER) measured in CAD$ per QALY. An ICER threshold of < CAD$50,000 per QALY is often cited as a benchmark for cost-effectiveness.¹² We adopted the perspective of the Federal, Provincial, and Territorial Ministries of Health, who are the single third payers for the Canadian health care system, a publicly funded health care system. The cycle length of the model was 1 month. All health outcomes and costs were discounted at 1.5% per year as per recommendations from the Canadian Agency for Drug and Technology in Health (CADTH).¹³

Model structure

A simplified schematic of the decision model is seen in Figure 1. The model begins with patients < 24 h following a TIA/minor stroke. The two compared alternative interventions are (1) CYP2C19 LOF allele testing and (2) no LOF allele testing (current clinical standard in Canada). In the testing arm, if patients are found to be LOF allele carriers, then they are started on an aspirin–ticagrelor regimen × 3 weeks. If they are non-carriers, then they are started on aspirin–clopidogrel × 3 weeks. In the no-testing arm, all patients are started on aspirin–clopidogrel. The proportion of patients who are CYP2C19 LOF allele carriers was modeled as 30% (with an estimated standard error of 10%) based on prior reports of its prevalence in Caucasian populations.^11,14 This prevalence was further modified with one-way sensitivity analyses to reflect the diverse population in Canada.

During the first 90 days (first three cycles), patients could have one of the four possible clinical transitions: (1) death, (2) survive (without a clinical event), (3) sustain an ischemic stroke (fatal, severe, moderate, or mild in severity) or (4) sustain a hemorrhage (fatal, intracranial (ICH), major, or minor in severity) (see Table 1). The estimated proportion of patients who sustained an ischemic stroke within 90 days in the aspirin–ticagrelor group is 6%,⁸ 6.7% in the non-carrier aspirin–clopidogrel group,⁸ and 7.6%¹¹ in the aspirin–clopidogrel non-testing arm. As per CHANCE-2, stroke severity breakdown in the aspirin–ticagrelor group was as follows: 2% sustained a fatal stroke (mRS 6), 16% severe (mRS 4 or 5), 33% moderate (mRS 2 or 3), and 49% mild (mRS 0–1). In the aspirin–clopidogrel group, 3% sustained a fatal stroke, 9% severe, 26% moderate, and 62% mild. The CHANCE-1 substudy⁸ does not report on functional outcome (i.e. stroke severity) of ischemic strokes, so it was assumed stroke severity and bleeding complication proportions were similar to that reported in CHANCE-2. The estimated proportion of patients that sustained a bleeding complication in the aspirin–clopidogrel arm is 2.5% (of which 86% were minor and 7.5% ICH),¹¹ and the estimated proportion in the aspirin–ticagrelor arm is 5.3% (of which 95% were minor and 2% were ICH).¹¹ Although myocardial infarction is reported clinically in the trials, we did not model this outcome as there were no statistically significant differences between aspirin–ticagrelor and aspirin–clopidogrel group in the CHANCE-2 study.

Table 1.

Model input parameters for the base-case analysis and ranges used in sensitivity analyses.

	Base-case value	Standard error	Distribution	Reference
Patient characteristics
Age, y	65	–	–	Wang et al.¹¹
Proportion of patients in the following health state (within 90 days of initial TIA/minor stroke)
LOF allele carriers	0.30	0.1	Beta	Wang et al.¹¹
Dead
Carriers	0.0028	0.000049	Beta	Wang et al.¹¹
Non-carriers	0.0056	0.000099	Beta	Wang et al.⁸
No LOF allele testing	0.0056	0.000099	Beta	Wang et al.¹¹
Survive without stroke/bleed
Carriers	0.88	0.14	Beta	Wang et al.¹¹
Non-carriers	0.90	0.27	Beta	Wang et al.⁸
No LOF allele testing	0.89	0.15	Beta	Wang et al.¹¹
Develop an ischemic stroke
Carriers	0.059	0.001	Beta	Wang et al.¹¹
Non-carriers	0.067	0.003	Beta	Wang et al.⁸
No LOF Allele testing	0.076	0.001	Beta	Wang et al.¹¹
Develop a fatal stroke
Carriers	0.021	0.0004	Beta	Wang et al.¹¹
Non-carriers	0.033	0.0006	Beta	Wang et al.⁸
No LOF allele testing	0.033	0.0006	Beta	Wang et al.¹¹
Develop a severe stroke
Carriers	0.16	0.0033	Beta	Wang et al.¹¹
Non-carriers	0.086	0.0017	Beta	Wang et al.⁸
No LOF allele testing	0.086	0.0017	Beta	Wang et al.¹¹
Develop a moderate stroke
Carriers	0.33	0.0087	Beta	Wang et al.¹¹
Non-carriers	0.26	0.0062	Beta	Wang et al.⁸
No LOF allele testing	0.26	0.0062	Beta	Wang et al.¹¹
Develop a minor stroke
Carriers	0.49	0.017	Beta	Wang et al.¹¹
Non-carriers	0.62	0.029	Beta	Wang et al.⁸
No LOF allele testing	0.62	0.029	Beta	Wang et al.¹¹
Develop a fatal bleed
Carriers	0.018	0.0003	Beta	Wang et al.¹¹
Non-carriers	0.038	0.0007	Beta	Wang et al.⁸
No LOF allele testing	0.038	0.0007	Beta	Wang et al.¹¹
Develop an ICH
Carriers	0.018	0.0003	Beta	Wang et al.¹¹
Non-carriers	0.075	0.0014	Beta	Wang et al.⁸
No LOF allele testing	0.075	0.0014	Beta	Wang et al.¹¹
Develop a major bleed
Carriers	0.018	0.0003	Beta	Wang et al.¹¹
Non-carriers	0.025	0.0005	Beta	Wang et al.⁸
No LOF allele testing	0.025	0.0005	Beta	Wang et al.¹¹
Develop a minor bleed
Carriers	0.95	0.316	Beta	Wang et al.¹¹
Non-carriers	0.86	0.111	Beta	Wang et al.⁸
No LOF allele testing	0.86	0.111	Beta	Wang et al.¹¹
Hazard ratios
Hazard ratio of death after ICH (months)
0–1	20.8	–	Log-normal	De Caterina et al.¹⁵
2–12	4.5	–	Log-normal	De Caterina et al.¹⁵
13–24	2.2	–	Log-normal	De Caterina et al.¹⁵
Hazard ratio of death after moderate ischemic stroke (months)
0–8	2.5	–	Log-normal	De Caterina et al.¹⁵
Hazard ratio of death after severe ischemic stroke (months)
0–8	8.2	–	Log-normal	De Caterina et al.¹⁵
9–54	2.17	–	Log-normal	De Caterina et al.¹⁵

ICH: intracerebral hemorrhage, LOF: loss-of-function, TIA: transient ischemic attack.

We followed the International Society for Pharmaco-economics and Outcomes Research–Society for Medical Decision-Making (ISPOR–SMDM) guidelines for modeling¹⁶ and the Consolidated Health Economic Evaluation Reporting Standards (CHEERS) statement for reporting economic evaluation results¹⁷ (Supplementary Table 1).

Probability of death

After the incident of TIA/minor stroke, the baseline age-specific probability of death was applied and obtained from Canadian lifetables.¹⁸ These baseline probabilities were modified using hazard ratios (HR) to reflect the incremental mortality due to the presence of a stroke severity, ICH, or major bleed. The magnitude of this increased mortality declines as time elapsed from the incident event. This decline was modeled using a chain of time-specific states, so-called tunnel states (see Figure 1). For example, the HR for death after an ICH was 20.8 the first month after the event, 4.5 the following year, and 2.2 the year after, after which baseline age-specific probability of death applied.¹⁵ After ICH, it was assumed that patients would resume antiplatelet therapy after approximately 2 years corresponding to the end of the tunnel state. HR were also applied for moderate and severe ischemic strokes (Table 1). Patients were able to have further recurrent ischemic strokes over their lifetime. The severity of this additional stroke was modeled using a weighted average severity and had a one-time upfront cost.

Quality of life estimates

The baseline utility for TIA/minor stroke was applied to all patients and was 0.75.¹⁹ An additional utility decrement was applied based on whether a severe or moderate ischemic stroke, ICH, or major bleeding event occurred (see Table 2). A wide range was applied to the utilities reflecting the clinical variability.²⁰ The utility decrement was applied for the duration of the increased mortality hazard related to the clinic event, after which the baseline TIA/minor stroke utility was applied over the remaining time horizon.

Table 2.

Health utilities and costs for the base-case analysis with ranges used in sensitivity analyses.

	Base-case value	Low	High	Distribution	Reference
Health state utilities
TIA/minor stroke	0.75	0.68	0.83	Beta	Tengs and Wallace²⁰
Moderate ischemic stroke	0.39	0.33	0.45	Beta	Tengs and Wallace²⁰
Severe ischemic stroke	0.2	0.17	0.23	Beta	Tengs and Wallace²⁰
ICH	0.62	0.41	0.83	Beta	Micieli et al.²¹
Costs (2021 CAD$)
Cost of LOF allele testing	300	250	350	Gamma	NICE²²
Medication costs per month
Aspirin	2	–	–	Gamma	Pricing²³
Aspirin–clopidogrel	15	–	–	Gamma	Pricing²³
Aspirin–ticagrelor	110	–	–	Gamma	Pricing²³
Transitional one-time costs
Minor ischemic stroke	9961	6641	13,281	Gamma	Patient Cost Estimator²⁴
Moderate ischemic stroke	26,368	17,579	35,157	Gamma	Patient Cost Estimator²⁴
Severe ischemic stroke	45,934	30,623	61,245	Gamma	Patient Cost Estimator²⁴
ICH	34,091	22,730	45,460	Gamma	Patient Cost Estimator²⁴
Major bleed	6358	4239	8477	Gamma	Patient Cost Estimator²⁴
Minor bleed	3667	2445	4889	Gamma	Patient Cost Estimator²⁴
Ongoing monthly costs
ICH	9223	6149	12,297	Gamma	Patient Cost Estimator²⁴
Moderate ischemic stroke	2472	1648	3296	Gamma	Patient Cost Estimator²⁴
Severe ischemic stroke	4240	2827	5653	Gamma	Patient Cost Estimator²⁴

ICH: intracerebral hemorrhage; CAD: Canadian dollars; LOF: loss-of-function; TIA: transient ischemic attack.

Costs

Costs reflected direct medical costs associated with medication, hospitalization, and physician services (see Table 2). The cost of aspirin–ticagrelor for 1 month (as per the province of Alberta) is CAD$110, CAD$15 for aspirin–clopidogrel for 1 month, and CAD$2 for single antiplatelet aspirin per month.²³ The cost of the machine to test for LOF allele is approximately CAD$7500 and will need to be replaced after approximately 3000 tests. There is an additional out-of-pocket cost patients pay for testing (~CAD$200). This resulted in an overestimated modeled cost per patient of CAD$300 (and a standard error of CAD$50 was used to reflect the uncertainty of this cost) and adjusted in the sensitivity analysis.¹⁶

Costs associated with clinical events were obtained from the Canadian Institute for Health Information (CIHI) patient cost database, a Canadian wide initiative reporting average cost data for acute inpatient events, complex continuing care, and rehabilitation (Table 2).²⁴ Costs were specifically taken from the province of Alberta, with wide standard errors to reflect the potential variability of costs across provinces. Costs not available in the CIHI patient cost database were taken from the literature (see Table 2). CIHI costs are reported in 2018 CDN dollars (most recent reported costs).

Sensitivity analysis

The base-case analysis was performed probabilistically with 10,000 Monte Carlo simulations, with the final outputs representing the expected (mean) values. Beta-distributions were applied to all probabilities and utilities, gamma-distributions to all costs, and log-normal distributions for all odds ratios and HR. Distributions were estimated using the parameters (e.g. mean and standard errors) from the respective published source documentation.

A series of deterministic one-way sensitivity analysis was completed on several variables including the cost of minor stroke, prevalence of LOF allele carriers, proportion of strokes in the non-testing arm, and proportion of ICH, which are important drivers of clinical outcomes. A multi-way probabilistic sensitivity analysis was performed to assess overall uncertainty, and a cost-effectiveness acceptability curve is reported to show the probability that a strategy is cost-effective given certain willingness-to-pay thresholds.

All analyses were performed using the TreeAge Pro Healthcare 2021 (TreeAge Software Inc., Williamstown, MA, USA). The local institutional review board waived the need for patient consent or ethics approval for this economic analysis.

Results

Clinical benefits

In the base-case analysis, LOF allele testing leads to 0.14 LYG or 0.12 QALYs gained (undiscounted) compared to standard of care.

Economic evaluation

After discounting for the economic base-case analysis, LOF allele testing leads to additional discounted lifetime costs of CAD$432 per patient with corresponding 0.10 discounted QALYs gained, which results in a discounted ICER of CAD$4310 per QALY for the LOF allele testing strategy when compared with standard of care. Detailed base-case figures are summarized in Tables 3 and 4.

Table 3.

Health economic outcomes of LOF allele testing including life years and QALY gained (undiscounted).

	Life expectancy, years (undiscounted)	LYG (undiscounted)	QALY (undiscounted)	QALY gained (undiscounted)
LOF allele testing	16.40	0.14	12.31	0.12
Standard of care	16.26	Reference	12.19	Reference

CAD: Canadian dollars; QALY: quality-adjusted life years; LOF: loss-of-function.

Table 4.

Health economic outcomes of LOF allele testing including QALY gained (discounted), costs (discounted), and ICER.

	QALY (discounted)	QALY gained (discounted)	Costs, CAD$ (discounted)	Incremental cost, CAD$ (discounted)	ICER (CAD$/QALY) (discounted)
LOF allele testing	10.61	0.1	7760	432	4310
Standard of care	10.51	Reference	7329	Reference	Reference

CAD: Canadian dollars; QALY: quality-adjusted life years; LOF: loss-of-function; ICER: incremental cost-effectiveness ratio.

Uncertainty assessment

Deterministic one-way sensitivity analyses were completed on specific model parameters. First, if the cost of minor stroke was set equal to the cost of a moderate stroke (i.e. CAD$26,368), the ICER is CAD$9500 per QALY gained. Second, if the LOF allele carrier prevalence was modeled at a slightly increased 40% (with a standard error of 10%), then the ICER slightly decreases (more cost-effective) to CAD$3020 per QALY. Third, if the proportion of ischemic stroke in the non-testing arm was modeling the same as the non-carriers (6.7% rather than base-case value 7.6%), then the ICER slightly increases to CAD$4790 per QALY. Finally, we investigated the effect of ICH occurrence on overall outcomes given the reduced prevalence in the aspirin–ticagrelor group (2.0% vs 7.5%). If the proportion of ICH in the aspirin–ticagrelor was modeled the same as the aspirin–clopidogrel group (i.e. 7.5% of hemorrhages), the ICER increases to CAD$10,400 per QALY still indicating strong cost-effectiveness.

The probabilistic sensitivity analyses demonstrated that LOF allele testing was cost-effective more than 99.99% of simulations using a willingness-to-pay threshold of CAD$50,000 per QALY (Figure 2). Even for an ICER threshold of CAD$30,000 (20,000) per QALY, the probability of LOF allele testing being cost-effective is 99.999% (99.1%).

Discussion

We evaluated the long-term clinical effectiveness and cost-effectiveness of testing for CYP2C19 LOF allele status to personalize DAPT for secondary prevention following a TIA/minor stroke. We found that LOF allele testing leads to significantly improved health benefits, that is, 0.12 additional undiscounted QALYs, when compared to no such testing. This comparative effectiveness relates to a number needed to test of around 8 to gain one full QALY. This health benefit comes with incremental lifetime costs of CAD$432 per patient, resulting in an ICER of CAD$4310 per QALY gained. There is very low sampling uncertainty, as demonstrated in the cost-effectiveness acceptability curve. At a willingness-to-pay threshold of CAD$50,000/QALY, LOF function testing is the preferred option in more than 99.9% of simulations.

There are clinical trade-offs associated with the use of aspirin–ticagrelor over aspirin–clopidogrel shown in CHANCE-2, and these nuances were incorporated in the model using quality of life indices (utilities). First, there is a modestly lower risk of stroke recurrence with aspirin–ticagrelor compared to aspirin–clopidogrel (5.9 vs 6.7%), and this protection was almost exclusively for minor strokes. Moreover, there is an increased incidence of bleeding with aspirin–ticagrelor (5.4 vs 2.5%), of which 95% is minor bleeding. In addition, the rates of ICH were lower with aspirin–ticagrelor (2 vs 7.5%) in CHANCE-2. ICH comes with a significant reduction in quality of life and associated upfront and long-term cost burden.²¹

We have observed that despite the increased upfront costs of testing, and the increased ticagrelor cost, the major driver of the overall results is due to the aspirin–ticagrelor stroke risk reduction in the LOF allele carriers and its associated long-term clinical benefit and cost reductions. Even mild stroke prevention has important clinical and health care cost implications. In addition, the reduced incidence of ICH likely contributes to this overall effect. This was further investigated in the one-way sensitivity analysis which modeled the same rates if ICH in the aspirin–ticagrelor group as in the aspirin–clopidogrel group, showing very robust decision results, with an ICER that still is considered very cost-effective.

The question remains regarding the generalizability of LOF allele testing in non-Asian populations. In the CHANCE-2 population, 59% of patients screened were LOF allele carriers, which is very similar to the estimated prevalence in the Chinese population of 62%.¹¹ In a Canadian population with higher incidence of Caucasians, many more patients will have to be screened to detect LOF allele function as the estimated prevalence in Caucasian population is around 30%.⁸ In our model, we used this prevalence of 30% (with a lower value of 20% and upper value of 40%). In our one-way sensitivity analysis, we modeled different prevalence’s of LOF function and found that the more prevalent the LOF allele status, the more favorable (cost-effective) point-of-care testing becomes. In addition, the rate of symptomatic large artery atherosclerosis is much higher in Asian populations compared with Caucasian predominant populations, and it seems that the benefit of DAPT for stroke prevention is higher in this population.²⁵

Our work provides important and novel insights. The decision analysis suggests that based on the currently available clinical data, LOF allele testing is effective and very cost-effective. It could lead to longer and quality-improved survival at a cost of less than CAD$4310 per QALY gained. As the price of ticagrelor becomes generic over time, and the cost per patient of point-of-care LOF allele testing will even be reduced with advancements in technology, which will only strengthen the economic argument for its implementation. For this point-of-care testing to be clinically helpful in this patient population of TIA/minor stroke, it requires a fast turn-around time (GMEX point-of-care turn-around time was around 85 min in CHANCE-2). Ideally, when a patient is seen in the emergency department or rapid TIA clinic/stroke prevention clinic, testing for LOF allele can be completed and the results known the same day resulting in immediate initiation of a personalized antiplatelet treatment regimen.

As all decision-analytic modeling studies, our study has several limitations. The decision-analytic model uses input parameters from the available clinical studies, specifically the CHANCE substudy for the clinical parameters of LOF allele non-carriers. This approach requires assumptions minimizing clinical and methodological differences between this smaller population and the overall patient demographic enrolled in CHANCE-2. Second, in our model, we assumed that patients are either carriers or non-carriers for the CYP2C19 allele; however, in clinical practice, there is likely a small percentage that will have false-positive/negative results.²² The clinical implications of this small population of patients are uncertain and unlikely to affect the overall results. Third, the CHANCE substudy or CHANCE-1 does not report on ischemic stroke or bleeding severity (an important clinical nuance that needed to be included in the model), so this analysis assumed that stroke severity and bleeding proportions were the same as CHANCE-2. Non-carriers may have slightly different clinical event rates; however, the authors feel this is unlikely to have meaningful clinical impact on the overall results. Fourth, the results are presented in Canadian dollars and reflective of medical practice of Canada, where a publicly funded universal health care system exists, which may not be directly applicable to other jurisdictions. Finally, the clinical outcomes seen in CHANCE-2 may not be generalizable to the very elderly population where these results are often extrapolated to in clinical practice.

Our model can provide directions for further research in this area. This cost-effectiveness analysis can be replicated in other patient populations, countries, and health care systems. It can also target the areas within this field that would most benefit from further research through partial perfect information analyses.

In conclusion, based on the available clinical data and our evidence-based decision analysis, testing for LOF allele status is expected to lead to prolonged life and improved quality of life, and can be considered very cost-effective when compared with other well-accepted technologies in health and medicine.

Supplemental Material

sj-pdf-1-wso-10.1177_17474930221111898 – Supplemental material for Cost-effectiveness of testing for CYP2C19 loss-of-function carriers following transient ischemic attack/minor stroke: A Canadian perspective

Supplemental material, sj-pdf-1-wso-10.1177_17474930221111898 for Cost-effectiveness of testing for CYP2C19 loss-of-function carriers following transient ischemic attack/minor stroke: A Canadian perspective by Andrew Micieli, Nishita Singh, Beate Jahn, Uwe Siebert, Bijoy K Menon and Andrew M Demchuk in International Journal of Stroke

Footnotes

Acknowledgements

The authors thank Dr Mohammed Almekhlafi for his edits to the final article.

Declaration of conflicting interests

The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article.

Disclosures

Dr Micieli: None relevant to this study.

Dr Menon: None relevant to this study.

Dr Demchuk: None relevant to this study.

Dr Jahn: None relevant to this study.

Dr Siebert: None relevant to this study.

ORCID iD

Andrew Micieli

Supplemental material

Supplemental material for this article is available online.

References

Hill

Yiannakoulias

Jeerakathil

, et al The high risk of stroke immediately after transient ischemic attack: a population-based study. Neurology 2004; 62: 2015–2020.

Johnston

Gress

Browner

, et al Short-term prognosis after emergency department diagnosis of TIA. JAMA 2000; 284: 2901–2906.

Wang

Zhao

, et al Clopidogrel with aspirin in acute minor stroke or transient ischemic attack. N Engl J Med 2013; 369: 11–19.

Pan

Elm

, et al Outcomes associated with clopidogrel-aspirin use in minor stroke or transient ischemic attack: a pooled analysis of Clopidogrel in High-Risk Patients With Acute Non-Disabling Cerebrovascular Events (CHANCE) and Platelet-Oriented Inhibition in New TIA and Minor Ischemic Stroke (POINT) Trials. JAMA Neurol 2019; 76: 1466–1473.

Johnston

Easton

Farrant

, et al Clopidogrel and aspirin in acute ischemic stroke and high-risk TIA. N Engl J Med 2018; 379: 215–225.

Powers

Derdeyn

Biller

, et al 2015 American Heart Association/American Stroke Association Focused Update of the 2013 guidelines for the early management of patients with acute ischemic stroke regarding endovascular treatment: a guideline for healthcare professionals From the American Heart Association/American Stroke Association. Stroke 2015; 46: 3020–3035.

Pereira

Rihal

DYF

, et al Clopidogrel pharmacogenetics. Circ Cardiovasc Interv 2019; 12: e007811.

Wang

Zhao

Lin

, et al Association between CYP2C19 loss-of-function allele status and efficacy of clopidogrel for risk reduction among patients with minor stroke or transient ischemic attack. JAMA 2016; 316: 70–78.

Paré

Mehta

Yusuf

, et al Effects of CYP2C19 genotype on outcomes of clopidogrel treatment. N Engl J Med 2010; 363: 1704–1714.

10.

Wang

Chen

Lin

, et al Ticagrelor plus aspirin versus clopidogrel plus aspirin for platelet reactivity in patients with minor stroke or transient ischaemic attack: open label, blinded endpoint, randomised controlled phase II trial. BMJ 2019; 365: l2211.

11.

Wang

Meng

Wang

, et al Ticagrelor versus clopidogrel in CYP2C19 loss-of-function carriers with stroke or TIA. N Engl J Med. 2021; 385: 2520–2530.

12.

Neumann

Cohen

Weinstein

. Updating cost-effectiveness—the curious resilience of the $50,000-per-QALY threshold. N Engl J Med 2014; 371: 796–797.

13.

Guidelines for the Economic Evaluation of Health Technologies: Canada. 4th ed. https://www.cadth.ca/sites/default/files/pdf/guidelines_for_the_economic_evaluation_of_health_technologies_canada_4th_ed.pdf

14.

Pan

Chen

, et al Genetic polymorphisms and clopidogrel efficacy for acute ischemic stroke or transient ischemic attack: a systematic review and meta-analysis. Circulation 2017; 135: 21–33.

15.

De Caterina

Connolly

Pogue

, et al Mortality predictors and effects of antithrombotic therapies in atrial fibrillation: insights from ACTIVE-W. Eur Heart J 2010; 31: 2133–2140.

16.

Caro

Briggs

Siebert

, et al Modeling good research practices—overview: a report of the ISPOR-SMDM modeling good research practices task force-1. Med Decis Making 2012; 32: 667–677.

17.

Husereau

Drummond

Petrou

, et al Consolidated health economic evaluation reporting standards (CHEERS) statement. Value Health 2013; 16: e1–e5.

18.

Government of Canada SC. Life tables, Canada, provinces and territories. https://www150.statcan.gc.ca/n1/en/catalogue/84-537-X (2022, accessed 15 February 2022).

19.

Pan

Wang

Liu

, et al Cost-effectiveness of clopidogrel-aspirin versus aspirin alone for acute transient ischemic attack and minor stroke. J Am Heart Assoc 2014; 3: e000912.

20.

Tengs

Wallace

. One thousand health-related quality-of-life estimates. Med Care 2000; 38: 583–637.

21.

Micieli

Demchuk

Wijeysundera

. Economic evaluation of andexanet versus prothrombin complex concentrate for reversal of factor Xa-associated intracranial hemorrhage. Stroke 2021; 52: 1390–1397.

22.

NICE. Spartan RX and point-of-care CYP2C19 test to guide treatment in acute coronary syndrome. https://www.nice.org.uk/advice/mib223/resources/spartan-rx-pointofcare-cyp2c19-test-to-guide-treatment-in-acute-coronary-syndrome-pdf-2285965510334917

23.

Pricing. Pharmaceutic pricing. https://pricingdoc.acfp.ca/pricing/ (accessed 22 December 2021).

24.

CIHI. Patient Cost Estimator. https://www.cihi.ca/en/patient-cost-estimator (accessed 19 December 2021).

25.

Yaghi

de Havenon

Rostanski

, et al Carotid stenosis and recurrent ischemic stroke: a post-hoc analysis of the POINT trial. Stroke 2021; 52: 2414–2417.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.44 MB