Introduction: The Enhanced Control of Hypertension and Thrombolysis Stroke Study (ENCHANTED) showed that a low-dose alteplase was safe but not clearly non-inferior to standard-dose alteplase in acute ischemic stroke (AIS). Given the significant cost of this medicine, we undertook a cost-effectiveness analysis to determine the probability that low-dose is cost-effective relative to standard-dose alteplase in China. Methods: For ENCHANTED participants in China with available health cost data, cost-effectiveness and cost-utility analyses were undertaken in which death or disability (modified Rankin scale scores 2–6) at 90 days and quality-adjusted life-years (QALYs) were used as outcome measures, respectively. There was adherence to standard guidelines for health economic evaluations alongside non-inferiority trials and according to a health-care payer’s perspective. The equivalence margin for cost and effectiveness was set at USD 691 and −0.025 QALYs, respectively, for the base-case analysis. Probabilistic sensitivity analyses were used to evaluate the probability of low-dose alteplase being non-inferior. Results: While the mean cost of alteplase was lower in the low-dose group (USD 1,569 vs. USD 2,154 in the standard-dose group), the total cost was USD 56 (95% confidence interval [CI]: −1,000–1,113) higher compared to the standard-dose group due to higher hospitalization costs in the low-dose group. There were 462 (95% CI: 415–509) and 410 (95% CI: 363–457) patients with death or disability per 1,000 patients in the low-dose and standard-dose groups, respectively. The low-dose group had marginally lower (0.008, 95% CI: −0.016–0.001) QALYs compared to their standard-dose counterparts. The low-dose group was found to have an 88% probability of being non-inferior based on cost-effectiveness versus the standard-dose group. Conclusions: This health economic evaluation alongside the ENCHANTED indicates that the use of low-dose alteplase does not save overall healthcare costs nor lead to a gain in QALYs in the management of Chinese patients with AIS compared to the use of standard dose. There is little justification on economic grounds to shift from standard-of-care thrombolysis in AIS.

© 2022 The Author(s). Published by S. Karger AG, Basel

Introduction

Acute ischemic stroke (AIS), which accounts for approximately 84% of incident strokes worldwide [1], can be effectively treated with intravenous alteplase (or recombinant tissue plasminogen activator) according to strict time and tissue-based eligibility criteria [2]. There is now strong evidence for improved odds of a favorable functional outcome in patients who received alteplase treatment within 4.5 h of the onset of symptoms despite the increased risk of intracranial hemorrhage [3, 4]. Guidelines recommend use of the licensed-approved dose of 0.9 mg/kg of body weight of alteplase in AIS [5], except in Japan where a low dose of 0.6 mg/kg was approved after a single-arm study showing comparable efficacy and safety to the standard-dose [6]. The latter situation has influenced clinical practice more broadly in Asia but created uncertainty over the superiority of low-dose intravenous alteplase treatment [6-9], particularly of an apparent lower risk of symptomatic intracerebral hemorrhage compared to the standard-dose [6, 10-12].

The Enhanced Control of Hypertension and Thrombolysis Stroke Study (ENCHANTED) was undertaken to determine whether a low dose of intravenous alteplase provides non-inferior clinical efficacy compared to the standard dose [13]. The study found that the primary outcome of death or disability at 90 days was 53.2% for patients in the low-dose group, which was not clearly non-inferior to 51.1% for patients in the standard-dose group: the odds ratio was 1.09 (95% confidence interval [CI]: 0.95–1.25), but the upper boundary exceeded the non-inferiority margin of 1.14 and was not statistically significant (p = 0.51) [13].

Identifying therapies with best value for money is critical to developing economies where health-care budgets are more constrained compared to those in the developed world. Although many developing countries have established local health technology assessment processes to make treatment publicly available [14], many proven treatments do not receive government funding or co-payment such that a patient’s ability to pay affect the choice of treatment. For example, a report from the Chinese National Stroke Registry indicates that 44% of AIS patients received low dose (0.6–0.8 mg/kg) in hospital, with economic considerations deemed a key underlying reason [15]. In fact, variable-dose regimes of intravenous alteplase are utilized across Asia due to the high cost of treatment [7]. As cost of medications generally accounts for a large proportion of the total costs of hospital care for stroke patients [16], it is important to identify suitable cost-effective treatment options.

A systematic review has shown that alteplase was generally a cost-effective, even cost-saving, strategy compared with conventional treatment for AIS [17], and the economic merit was greater with longer time horizons after treatment [18]. Given the high costs of alteplase, a low-dose regimen might have economic merit compared to standard-dose, but there have been few direct cost-effectiveness dose-comparison studies. In this study, we aimed to conduct a within-trial health economic evaluation alongside ENCHANTED to evaluate whether the use of low-dose alteplase could be economically non-inferior compared to standard-dose in Chinese patients with AIS.

Materials and MethodsDesign and Population

In the alteplase-dose comparison arm of ENCHANTED, two doses of intravenous alteplase were compared in AIS patients who were eligible for thrombolytic therapy. Details of the trial design and outcome measures are documented elsewhere [13]. In brief, study participants were randomly assigned to receive intravenous alteplase, either at a standard dose (0.9 mg/kg of estimated, or measured, body weight; 10% as a bolus and 90% as an infusion over a period of 60 min with a maximum dose of 90 mg) or a low dose (0.6 mg/kg, 15% as a bolus and 85% as an infusion over a period of 60 min with a maximum dose of 60 mg) [13]. The primary outcome was the combined endpoint of death or disability at 90 days. Death or disability was defined as a score of 2–6 on the modified Rankin scale, where a score of 2–5 indicates varying degrees of disability (“poor health outcome”) and 6 indicates death [19]. Health-related quality of life was measured using the 3-level version of the EuroQoL group 5-dimension self-report questionnaire (EQ-5D-3L) at 28 days and 90 days. Participants (age ≥18 years) were from China, the United Kingdom, continental Europe, Australia, South America, and several Asian countries; approximately two-thirds were from Asia, and 43% from China [13]. A within-trial economic alteplase-dose comparison evaluation was only conducted for Chinese patients in which cost data were available. We considered whether the potential savings from low-dose treatment set against the observed marginal reduction in health benefit nonetheless establishes a case for it being economically non-inferior. The evaluation time frame was in line with the ENCHANTED primary outcome endpoint at 90 days. As the aim was to test the non-inferiority of low-dose compared to the standard-dose alteplase, we followed a guideline for health economic evaluations alongside non-inferiority trials [20].

Costs

The health economic evaluation was conducted using the healthcare payer’s perspective and included only direct health service costs. Costs included were those related to cost of alteplase and use of other health-care resources: the cost of alteplase applied to the patient during the first admission to hospital and recorded after close-out of the trial; costs related to outpatient visits, re-admission to hospital within 90 days of the study period, and all medications were captured. All costs data were converted into 2020 US dollars, accounting for inflation and purchasing power parity [21].

There is no consensus on how to determine the equivalence margin for costs in an economic evaluation of a non-inferior trial. We chose the equivalence margin for costs (θ) between the dose groups as the difference in alteplase cost, on the assumption that hospital costs were similar and that the difference was related to the cost of alteplase. The cost of a 50-mg vial of alteplase was 5,500 Renminbi Yuan (or USD 1,706) in the participating hospitals. As the difference in the dose of alteplase was 0.3 mg/kg of body weight, the expected difference in alteplase cost was 2,228 Renminbi Yuan (or USD 691), with an average body weight of 67.5 kg in the low-dose group. In the base-case analysis, we set the equivalence margin for costs at USD 691. Discounting of costs was not considered as the trial period was only 90 days.

Outcomes

The economic evaluation comprised a cost-effectiveness analysis, in which death or disability at 90 days was the outcome measure, and a cost-utility analysis in which quality-adjusted life-years (QALYs) were used as the outcome measure. The health state utility (HSU) was calculated using the time trade-off value set for EQ-5D-3L in the Chinese population [22]. QALYs for each study participant were calculated as the arithmetic product of life expectancy combined with the HSUs. As the minimal clinically important difference of EQ-5D-3L in stroke patients was estimated at 0.1 [23], the equivalence margin (δ) for QALYs between the dose groups was determined pro rata to be −0.025 (i.e., −0.1 × 90 ÷ 365.25) QALYs for the trial period of 90 days.

Statistical Analysis

A generalized linear regression model was used to compare the numbers of patients who died or were disabled at 90 days and costs between the two treatment groups. Probabilistic sensitivity analyses with 5,000 bootstrap replications were conducted to address uncertainties around costs and outcomes differences. Incremental cost-effectiveness ratios were calculated, and a cost-effectiveness plane was presented to illustrate the paired incremental costs and effectiveness from bootstrapping.

As the alteplase-dose arm of ENCHANTED was designed to test the non-inferiority of low-dose alteplase, we evaluated the probability of incremental cost-effectiveness pairs located in the non-inferiority range (see Fig. 1). The incremental cost and effectiveness in all bootstrap samples were scattered on the cost-effectiveness plane. The equivalence margin for costs was set as USD 691, and the equivalence margin for effectiveness set as −0.025 QALYs. Non-inferiority of cost-effectiveness can be demonstrated if 95% of the incremental cost-effectiveness pairs are located to the right of the equivalence margin for effectiveness and below the equivalence margin for costs (i.e., the area in gray in Fig. 1). We used Stata MP version 15.1 for the economic evaluation.

Fig. 1.

Cost-effectiveness plane of 5,000 bootstrap replications of the incremental costs in US dollars and incremental effectiveness in quality-adjusted life-years (QALYs) for the low-dose alteplase group compared to the standard-dose alteplase group. The x-axis denotes the predefined equivalence margin for costs (θ) at USD 691, and the y-axis is the predefined equivalence margin (δ) for effectiveness at −0.025 QALYs. The cost-effectiveness pairs in the gray area are non-inferior replications. The cost-effectiveness pairs in the oval represent the 95% CI of the bootstrap replications. With the predefined equivalence margin for costs and effectiveness, the low-dose alteplase had a probability of 88% of being non-inferior in cost-effectiveness compared to the standard-dose group.

Scenario Analysis

The first scenario analysis evaluated the cost difference between the trial groups if a theoretical dose was applied to participants (i.e., 0.9 mg/kg for the standard-dose group and 0.6 mg/kg for the low-dose group) [13] with cap doses set at 60 mg and 90 mg for the low-dose and standard-dose groups, respectively [13]. The costs of alteplase were calculated based on the unit price for the theoretical dose applied to patients. The cost of a 50-mg injection of alteplase was USD 1,706 at a trial hospital. We calculated the theoretical cost of alteplase for individual patients in the trial by multiplying the cost of alteplase per microgram by the theoretical dose in the first scenario analysis.

In the second scenario analysis, the equivalence margin for costs was set as an average inpatient cost of stroke in China, where the average hospitalization cost for a case of AIS was 17,731 Chinese Yuan (USD 5,153) [24]. The equivalence margin for cost was set at USD 5,153 and equivalence margin for effectiveness at −0.025 QALYs. We also determined the robustness of the cost-effectiveness result by varying the cost of alteplase by 20% for each study participant.

Reporting Quality

Reporting of this study is in line with the Consolidated Health Economic Evaluation Reporting Standards statement [25]. A quality check is presented in online supplementary Table 1 (see www.karger.com/doi/10.1159/000525869 for all online suppl. material).

Results

A total of 867 patients with cost data collected after completion of the alteplase-dose arm of the ENCHANTED trial were included, which accounted for 61% of participants in China. Table 1 shows that there were no significant differences in the characteristics of participants at baseline or their treatment except the dose of alteplase. In addition, comparisons between the characteristics of study participants in the economic evaluation and those who did not have available costs data in China are presented in online supplementary Table 2. The mean age of the patients was 64 years, and nearly a third were women in both groups, but the median National Institutes of Health Stroke Scale (NIHSS) scores were 8 (IQR 4–12) and 7 (IQR: 4–12) in the low-dose and standard-dose groups, respectively. Of note, the bolus dose of alteplase was not significantly different in the two groups.

Table 1.

Characteristics of patients at baseline and their treatment

Costs and HSUs are summarized in Table 2, where the mean cost of alteplase was USD 1,569 and USD 2,154 for the low-dose and standard-dose groups, respectively. While the costs of alteplase and outpatient visits were lower in the low-dose group, the average cost of inpatient visits and re-admission to hospital was higher compared to the standard-dose group. HSUs at 28 and 90 days for alive patients were lower in the low-dose group but the difference was not statistically significant.

Table 2.

Costs and health state utilities (HSUs) for the within-trial period

Results of the cost-effectiveness and cost-utility analyses are presented in Table 3. The low-dose group incurred USD 56 (95% CI: −1,000–1,113) additional costs but with more patients with death or disability, and lower QALYs, at 90 days. The average QALYs was 0.205 (0.199–0.211) years for the low-dose group and 0.212 (0.207–0.218) years for the standard-dose group. There were 462 (95% CI: 415–509) and 410 (95% CI: 363–457) patients with death or disability per 1,000 patients in the low-dose and standard-dose groups, respectively. However, none of these differences in costs, clinical effectiveness, or QALYs was statistically significant.

Table 3.

Cost-effectiveness and cost-utility analyses of the low-dose alteplase compared to the standard-dose alteplase groups

The cost-effectiveness plane of the 5,000 bootstrapped estimates of incremental costs and QALYs for the low-dose compared to standard-dose group are illustrated in Figure 1. With the predefined equivalence margin for costs of USD 691 and effectiveness of −0.025 QALYs, the low-dose group had an 88% probability of being non-inferior to the standard-dose group (Fig. 1).

Scenario Analysis

The average theoretical doses of alteplase were 40.4 mg and 61.4 mg, and the average costs of alteplase were USD 1,390 and USD 2,105 for the low-dose and standard-dose groups, respectively. The total cost was lower (USD 8,438 and USD 8,513 for the standard-dose and low-dose groups, respectively) in the low-dose group, with an incremental cost of USD −74 (95% CI: −1,130–981). The cost-effectiveness plane of the 5,000 bootstrapped estimates of the incremental costs and QALYs for this scenario analysis are presented in online supplementary Figure 1. With the predefined equivalence margin for costs of USD 691 and effectiveness of −0.025 QALYs, there were 4,609 simulations located in the non-inferiority domain. Thus, the low-dose group had a 92.2% probability of being non-inferior for cost-effectiveness compared to the standard-dose group. When the equivalence margin for costs was set as an average inpatient cost for a thrombolyzed AIS patient, the low-dose group had 100% probability of being non-inferior compared to the standard-dose group.

Finally, varying the cost of alteplase did not change the overall result of the cost comparison in the base-case analysis. The higher cost in the low-dose group persisted for a 20% increase (USD 8,715 vs. USD 8,734 for the standard-dose and low-dose groups, respectively) or 20% decrease (USD 8,424 vs. USD 8,506 for the standard-dose and low-dose groups, respectively) in the cost of alteplase.

Discussion

In this within-trial economic evaluation of the Chinese participants in the pragmatic ENCHANTED clinical stroke trial, who comprised approximately one quarter of the study population, we have shown that treating AIS patients with low-dose alteplase does not save overall healthcare costs or improve QALYs. Low-dose alteplase intervention is highly likely to be economically non-inferior compared to the standard-dose treatment within specified equivalence margins for costs and QALYs.

To our knowledge, this is the first economic evaluation of the cost-effectiveness of low-dose versus standard-dose alteplase, using Chinese AIS patients where the dose comparison is especially relevant. Although the direct cost of alteplase is lower with a lower dose, the overall cost is slightly higher at USD 56 (95% CI: USD −1,000 to USD 1,113) after accounting for higher hospitalization costs in patients receiving low dose. Moreover, the low-dose group does not have a statistically significant improvement in QALY, with a marginal incremental QALY of −0.008 (95% CI: −0.016 to 0.001) years compared to the standard-dose group. According to predefined equivalence margins for costs and effectiveness, the low-dose alteplase treatment strategy had a high (88%) probability of being non-inferior compared to the standard-dose group, although these measures used for these calculations were not significantly different between the dose groups.

Previous studies on the comparative efficacy of different doses of alteplase have not shown clinical non-inferiority of low-dose alteplase, neither for a primary clinical endpoint nor across patients subgroups with AIS [13, 26]. Our study provides further justification that low-dose alteplase is not a “superior” treatment to standard-dose alteplase from the perspectives of cost-saving or improvement in QALY. Any clinical decision over use of low-dose alteplase is, therefore, confined to the perception of a safety benefit in regard to the risk of ICH.

Our choice of using a cost-effectiveness rather than cost-minimization analysis for a non-inferiority trial is based on the uncertainties over the size of the clinical effect [20, 27]. As our goal was to gage the economic non-inferiority of low-dose alteplase treatment, we defined a priori equivalence margins for costs and effects. When the equivalence margins for cost and effects were defined as the difference in alteplase cost and as minimal clinically important difference of EQ-5D, respectively, in AIS patients over 90 days, the low-dose group appeared economically non-inferior in 4,393 out of 5,000 bootstrapping. Moreover, using scenario analyses with a higher equivalence margin for cost, as the average inpatient cost of stroke in China, there was further increase in the probability of the low-dose group being economically non-inferior [24].

A key strength of our study is the use of reliable and systematic individual level data from a clinical trial which provides high internal validity [28]. However, there are several potential limitations, many of which are germane to using data from a clinical trial. First, while the cost saving for standard-dose alteplase were related to higher inpatient costs in the low-dose group, we did not have information on what specific inpatient services drove these differences. Future studies should include more details of cost categories to allow determination of specific components of such economic analysis. Second, cost data were not available at all ENCHANTED Chinese sites, which introduces further selection bias. While the study characteristics were balanced between the dose groups, the sample size for economic evaluation reduced by approximately 40% of all Chinese participants in the alteplase-dose arm of ENCHANTED. Finally, all relevant costs and health outcomes might not have been fully captured within the conventional trial follow-up period of 90 days.

In summary, this health economic evaluation alongside ENCHANTED has shown that treating Chinese AIS patients with low-dose alteplase does not result in either cost savings or improvement in QALYs. With predefined equivalence margins for costs and QALYs, the low-dose alteplase intervention has an 88% probability of being economically non-inferior compared to the standard-dose strategy. The study finds little justification on economic grounds for shifting standard care to a low-dose regimen.

Statement of Ethics

This study complied with the guidelines for human studies. The research was conducted ethically in accordance with the World Medical Association Declaration of Helsinki. This study protocol was reviewed and approved by Ethics Review Committee (RPAH Zone), Sydney Local Health District, NSW, Australia, with Protocol Nos. X11-0123 and HREC/11/RPAH/176. At each participating center, the study protocol was approved by the appropriate ethics committee, and written informed consent was obtained from each patient or an appropriate surrogate.

Conflict of Interest Statement

Craig S. Anderson reports honoraria and travel reimbursement from Takeda outside of this study. The other authors have no disclosures to report.

Funding Sources

The ENCHANTED study received grants from the National Health and Medical Research Council (NHMRC) of Australia (Project Grant Nos. 1020462 and 1101113), the Stroke Association of the UK (TSA 2012/01 and 2015/01), Ministry of Health and the National Council for Scientific and Technological Development of Brazil (CNPQ: 467322/2014-7, 402388/2013-5), and the Ministry for Health, Welfare and Family Affairs of the Republic of Korea (HI14C1985). Lei Si is funded by a National Health and Medical Research Council (NHMRC) of Australia Early Career Fellowship (GNT1139826), and Craig S. Anderson is funded by an NHMRC Senior Investigator Fellowship (APP1175861). Thomas Lung is supported by a NHMRC Early Career Fellowship (APP1141392) and National Heart Foundation of Australia Postdoctoral Fellowship (award 101956). The funders have no role in the identification, design, conduct, and reporting of the analysis.

Author Contributions

Xiaoying Chen and Lei Si contributed to the concept and rationale for the study, undertook statistical analyses, and wrote the first draft of manuscript. Guofang Chen, Yong-jun Cao, Guojun Wu, Jinli Zhang, Jingfen Zhang, YuKai Liu, Shihong Zhang, and Lili Song contributed to data collection, revision, and approval of the final version of the manuscript for publication. Menglu Ouyang, Xia Wang, Candice Delcourt, Hisatomi Arima, Lidan Wang, Thomas Lung, Mingsheng Chen, Craig S. Anderson, and Stephen Jan commented upon and approved the final version of the manuscript for publication.

Data Availability Statement

Individual participant data used in these analyses can be shared by formal request with protocol from any qualified investigator to the Research Office of The George Institute for Global Health, Australia. All data generated or analyzed during this study are included in this article and its online supplementary material. Further inquiries can be directed to the corresponding author.

This article is licensed under the Creative Commons Attribution-NonCommercial 4.0 International License (CC BY-NC). Usage and distribution for commercial purposes requires written permission. Drug Dosage: The authors and the publisher have exerted every effort to ensure that drug selection and dosage set forth in this text are in accord with current recommendations and practice at the time of publication. However, in view of ongoing research, changes in government regulations, and the constant flow of information relating to drug therapy and drug reactions, the reader is urged to check the package insert for each drug for any changes in indications and dosage and for added warnings and precautions. This is particularly important when the recommended agent is a new and/or infrequently employed drug. Disclaimer: The statements, opinions and data contained in this publication are solely those of the individual authors and contributors and not of the publishers and the editor(s). The appearance of advertisements or/and product references in the publication is not a warranty, endorsement, or approval of the products or services advertised or of their effectiveness, quality or safety. The publisher and the editor(s) disclaim responsibility for any injury to persons or property resulting from any ideas, methods, instructions or products referred to in the content or advertisements.