AcT Trial: Protocol for a Pragmatic Registry‐Linked Randomized Clinical Trial

Abstract

Background

Intravenous thrombolysis with alteplase is widely used in patients with acute ischemic stroke presenting early after symptom onset. Recent phase II trials have suggested that intravenous tenecteplase may be safer and associated with higher early reperfusion rates as compared with alteplase. This study investigates whether intravenous tenecteplase is noninferior to intravenous alteplase for the treatment of acute ischemic stroke.

Methods

This is a pragmatic, registry‐linked, prospective, randomized (1:1) controlled, open‐label parallel group clinical trial (AcT [Alteplase Compared to Tenecteplase in Patients With Acute Ischemic Stroke]) with blinded end point assessment of 1600 patients to test if intravenous tenecteplase (0.25 mg/kg body weight, maximum dose 25 mg) is noninferior to intravenous alteplase (0.9 mg/kg body weight; maximum dose, 90 mg) in patients with acute ischemic stroke eligible for intravenous thrombolysis in clinical routine. Patients are recruited from comprehensive and primary stroke centers and enrolled using deferral of consent. The proposed sample has at least 90% power with a noninferiority margin of 5%, assuming incidence of the 90‐day modified Rankin Scale score of 0 to 1 is 38% in the tenecteplase and 35% in the alteplase groups, and a loss to follow‐up rate <5%.

Results

The blinded primary end point is the proportion of subjects achieving a 90‐day modified Rankin Scale score of 0 to 1. Key safety outcomes include 24‐hour symptomatic intracerebral hemorrhage and 90‐day all‐cause mortality. All serious adverse events within a 24‐hour period will be reported and coded using the Medical Dictionary for Regulatory Activities. Outcomes are collected either centrally (primary, key secondary, and safety end points) or through ongoing Canadian stroke registries. The primary analysis is a simple unadjusted comparison of proportions.

Conclusions

Results from the trial will provide real‐world evidence of the effectiveness of intravenous tenecteplase versus alteplase in patients with acute ischemic stroke presenting early after stroke onset.

Nonstandard Acronyms and Abbreviations

AcT

Alteplase Compared to Tenecteplase in Patients With Acute Ischemic Stroke

DSMC

Data Safety Monitoring Committee

EVT

endovascular thrombectomy

mRS

modified Rankin Scale

OPTIMISE

Optimizing Patient Treatment in Major Ischemic Stroke With EVT

QuiCR

Quality Improvement and Clinical Research

Clinical Perspective

• This article describes the design and methods used in the AcT (Alteplase Compared to Tenecteplase in Patients With Acute Ischemic Stroke) randomized controlled trial that investigates whether intravenous tenecteplase is noninferior to intravenous alteplase for the treatment of acute ischemic stroke.

• The AcT trial is a pragmatic, registry‐linked, prospective, randomized (1:1) controlled, open‐label, parallel‐group clinical trial with blinded end point assessment.

• If the AcT trial demonstrates noninferiority of intravenous tenecteplase when compared with alteplase, we believe that the trial will lead to a change in clinical practice.

Intravenous alteplase and endovascular thrombectomy (EVT) are widely used therapies in patients with acute ischemic stroke. ¹ , ² , ³ The common goal of both therapies is fast, effective, and safe reperfusion. Since publication of data from recent EVT trials, ⁴ , ⁵ , ⁶ physicians, hospitals, and health systems have focused their effort on implementing efficient triaging systems and workflow processes and improving device design and techniques to further improve the speed and efficacy of EVT. Despite evidence for efficacy of intravenous alteplase, extensive public and professional education campaigns, and enhanced quality of acute stroke facilities, rates of alteplase use remain low. ⁷ , ⁸ , ⁹ , ¹⁰ This has been attributed to the relative lack of efficacy of intravenous alteplase in patients with more severe stroke, increased risk of intracranial hemorrhage, and logistical delays in administering alteplase (bolus and infusion).

Recent phase II acute ischemic stroke trials have suggested that intravenous tenecteplase is safer and associated with higher early reperfusion rates as compared with intravenous alteplase. ¹¹ , ¹² , ¹³ , ¹⁴ A recently published individual patient‐level meta‐analysis ¹⁵ of 291 patients from 3 phase II trials showed that tenecteplase at a dose of 0.25 mg/kg body weight was associated with greater odds of achieving early neurological improvement and better 90‐day outcomes when compared with intravenous alteplase. In addition, evidence suggests that tenecteplase may achieve earlier and better recanalization of thrombus within large intracranial arteries than intravenous alteplase. ¹⁶ , ¹⁷ However, none of these trials have addressed the pragmatic primary question about the real‐world effectiveness of tenecteplase versus alteplase as standard thrombolytic therapy in all patients who are currently eligible for intravenous thrombolysis. This question is particularly relevant in the modern era where a subpopulation of patients receiving intravenous thrombolysis also undergo EVT. The AcT (Alteplase Compared to Tenecteplase in Patients With Acute Ischemic Stroke) trial will therefore seek to demonstrate the noninferiority of intravenous tenecteplase compared with intravenous alteplase on 90‐day functional outcome assessed using the modified Rankin Scale (mRS) score. The secondary objective of this study is to compare the safety of intravenous tenecteplase compared with alteplase.

Methods

Study Design

The AcT trial is a pragmatic, registry‐linked, prospective, randomized (1:1) controlled, open‐label parallel group clinical trial with blinded end point assessment. A sample of 1600 patients with acute ischemic stroke eligible for intravenous thrombolysis will be recruited to test whether intravenous tenecteplase (0.25 mg/kg body weight; maximum dose, 25 mg) is noninferior to intravenous alteplase (0.9 mg/kg body weight; maximum dose, 90 mg). A simplified minimal sufficient balance algorithm ⁵ , ¹⁸ is used to randomize subjects, which includes a single algorithm to assure approximately equal assignment to tenecteplase or alteplase within site. After each site has enrolled 5 subjects using simple randomization, if imbalance in the proportion of patients assigned to each group within site is detected at a threshold P value of 0.3 (from a chi‐square test), a weighted randomization distribution of 0.65 : 0.35 is used. If no imbalance is detected, simple randomization of 0.5 : 0.5 is used. Randomization is centralized, secure, and concealed using a real‐time web‐based server. Investigators can access the randomizer either through the Internet, secure text, or a local telephone. Given the pragmatic design of the trial and the time‐sensitive nature of acute stroke, blinding the enrolling health personnel to treatment allocation is not practical. The data that support the findings of this study are available from the corresponding author upon reasonable request.

Patient Population

The trial will recruit patients from the emergency departments of participating primary or comprehensive stroke centers across Canada. Inclusion criteria are pragmatic. All patients with acute ischemic stroke eligible to receive intravenous alteplase in clinical routine are eligible for enrollment in the trial. Standard care in Canada is informed by the Canadian Stroke Best Practices guidelines (Table). ¹⁹ Patients eligible for EVT in addition to intravenous thrombolysis are eligible for enrollment. Since the benefits of thrombolysis with intravenous alteplase in the pediatric population are unknown, patients <18 years of age are not eligible to be enrolled. Women with pregnancy known to the investigator by history or examination, without requiring pregnancy testing, may be enrolled only in consultation with an expert stroke physician (either in person or through telestroke). All patients will have standard‐of‐care medical management on an acute stroke unit. There are no additional trial‐specific management recommendations; management will be according to local standards of care, physician discretion, and guided by the Canadian Stroke Best Practice guidelines. ¹⁹

Table

i.Box 5B criteria for acute thrombolytic therapy with intravenous alteplase. 19 The decision to use alteplase in these situations should be based on the clinical judgment of the treating physician. The relative benefits of alteplase therapy vs any potential risks or contraindications should be weighed on an individual basis. IV alteplase.

Treatment inclusion criteria

i.Diagnosis of ischemic stroke causing disabling neurologic deficit in a patient who is ≥18 years of age. ii.For adolescents, decision to administer alteplase should be based on clinical judgment, presenting symptoms, and patient age; and, if possible, consultation with a pediatric stroke specialist. iii.Time from last known well (onset of stroke symptoms) <4.5 hours before alteplase administration.

Absolute exclusion criteria

i.Any source of active hemorrhage or any condition that could increase the risk of major hemorrhage after alteplase administration. ii.Any hemorrhage on brain imaging. Relative exclusion criteria (requiring clinical judgment based upon the specific situation).

iii.Historical

iiia.History of intracranial hemorrhage. iiib.Stroke or serious head or spinal trauma in the preceding 3 months. iiic.Major surgery, such as cardiac, thoracic, abdominal, or orthopedic in the preceding 14 days. Risk varies according to the procedure.

iv.Arterial puncture at a noncompressible site in the previous 7 days.

v.Clinical

va.Symptoms suggestive of subarachnoid hemorrhage. vb.Stroke symptoms attributable to another nonischemic acute neurological condition such as seizure with postictal Todd's paralysis or focal neurological signs caused by severe hypo‐ or hyperglycemia. vc.Hypertension refractory to aggressive hyperacute antihypertensive treatment such that target blood pressure <180/105 cannot be achieved or maintained. Blood pressure should be treated rapidly and aggressively to minimize delays to thrombolysis. vd.Patient currently prescribed and taking a direct non–vitamin K oral anticoagulant.

CT or MRI findings

vi.CT showing early signs of extensive infarction

vii.Laboratory

viia.Blood glucose concentration <2.7 mmol/L or >22.2 mmol/L. viib.Elevated activated partial thromboplastin time. viic.International normalized ratio >1.7. viid.Platelet count <100 000/mm3.

CT indicates computed tomography; and MRI, magnetic resonance imaging.

Where approved by the local Research Ethics Board, to reduce time to treatment, patients will be enrolled using a deferral‐of‐consent procedure. The responsible treating physician will determine patient eligibility for the trial and initiate randomization and treatment administration if the patient is deemed eligible for the trial. An information sheet will be made available to the physician to help them inform the patient or their surrogate about the trial. Sharing of such information with the patient/surrogate decision maker may happen before, during, or after randomization. Consent is obtained from subjects or their surrogates within 7 days of randomization or before discharge, whichever is earlier. If a patient dies, and consent has yet to be obtained, reasonable efforts are made to obtain consent from their surrogates. If consent could not be obtained despite best efforts (no surrogate or surrogate cannot be contacted), the patient will continue in the trial. Subjects or, if incapacitated, their legal representatives will have the right to withdraw from further participation (including long‐term follow‐up by record linkage). Subjects may withdraw consent for further contact but continue follow‐up via record linkage. This process of consent was developed in consultation with an ethicist and a focus group involving patients/caregivers, respects the Tri‐Council Policy Statement–Ethical Conduct for Research Involving Humans guidelines, ²⁰ and reflects the imperative to treat patients quickly, so as not to bias results or disadvantage enrolled patients compared with patients not enrolled in the trial.

Participating Hospitals

All participating hospitals also participate in 1 of 2 Canadian acute stroke registries: the QuiCR (Quality Improvement and Clinical Research) or OPTIMISE (Optimizing Patient Treatment in Major Ischemic Stroke With EVT) registries. QuiCR (https://www.ucalgary.ca/quicr/) is a cloud‐based repository of data on all patients with ischemic stroke receiving thrombolysis and/or endovascular treatment in Alberta. OPTIMISE is a national registry supported by the Canadian Stroke Consortium to support the implementation and quality control for endovascular therapy and intravenous thrombolysis among patients with acute ischemic stroke. The data collection fields in OPTIMISE were designed to mirror the QuiCR registry. Data for the OPTIMISE registry are captured using a web‐based electronic data capture and reporting system and housed on a secure server at the Population Health Research Institute at McMaster University.

Study Outcomes

Trial Data

Duration of follow‐up will be up to 90–120 days. The primary outcome (mRS) is collected at the end of follow‐up (targeted at 90 days but for practicality, up to 120 days after stroke onset). The primary outcome (assessed using the mRS) is determined by the Rankin Focused Assessment method using centralized telephone interview by trained study personnel blinded to treatment allocation. ²¹ Secondary outcomes including the actual 90‐day mRS score, the 5‐item EuroQol Scale, EuroQol visual analogue scale, and percentage of patients returning to baseline level of functioning are also collected by the same blinded study personnel at the same telephone interview. Key safety outcomes collected directly by the trial include 90‐day all‐cause mortality and symptomatic intracerebral hemorrhage occurring within 24 hours of thrombolytic administration. Symptomatic intracranial hemorrhage is defined as intracranial hemorrhage that is temporally related to and directly responsible for worsening of the patient's neurological condition and, in the investigator's opinion, be the most important factor for the neurological worsening. Given the short half‐life of both thrombolytic agents (tenecteplase and alteplase) and their known safety profile, serious adverse events that occur beyond 24 hours are considered unrelated to the study drug. Data on any serious adverse events that occur within the first 24 hours of randomization will be collected directly through the trial database. These also include 3 serious adverse events of special interest for the study: (1) symptomatic intracerebral hemorrhage, (2) orolingual angioedema (even if reported as an adverse event), and (3) peripheral bleeding requiring blood transfusion. All available 24‐hour imaging per standard care will be assessed for infarct extent (Alberta Stroke Program Early CT Score) ²² and hemorrhage (using the Heidelberg classification) ²³ by imaging core lab personnel blinded to treatment allocation. Any serious unexpected adverse drug reactions occurring at any time during follow‐up will be reported directly through the trial database. A trial safety committee (constituted by the sponsor and chaired by physicians outside of the trial steering committee) will review unblinded safety data throughout the duration of the trial to ensure that events are being reported and fall within accepted norms for routine stroke care. The independent Data Safety Monitoring Committee (DSMC) will periodically review unblinded overall safety data to determine patterns and trends of events, or to identify safety issues, which would not be apparent on an individual case basis. Details of the DSMC review are outlined in the DSMC charter.

Registry Data

Other secondary outcomes, including ambulatory status at discharge, discharge destination, door‐to‐needle time, door‐in–door‐out time at primary stroke centers, proportion of patients administered EVT and, for those getting EVT, door‐to‐arterial‐puncture time, computed tomography‐to‐arterial‐puncture time, and recanalization status at first angiographic acquisition, are all collected through the registries. Serious adverse events or adverse events collected through the registries will be reported separately from those collected directly by the trial. The trial sponsor will work with the participating hospitals to perform regular monthly quality checks on data entry and logic.

Administrative Data

The Canadian Institute for Health Information collects and analyzes information on health and health care in Canada. Such information includes administrative, clinical, and demographic information on patient hospital admissions and discharges. These data are received directly from hospitals or from their respective health authorities or the provincial ministry/department of health and will be provided to the trial in a fully anonymized manner to the trial. Processes for access to these administrative data and the timelines for availability vary provincially. Home time as estimated by the number of nights a patient is back at their premorbid living situation without an increase in level of care within 90 days of the stroke will be obtained from these administrative data. In addition, wherever available, these data will be used to tabulate preexisting conditions (comorbidities) that may influence care or the hospital stay on conditions that arose after admission and that may thus represent complications of care and preexisting conditions (comorbidities) that do not influence care or the hospital stay. Other data elements including interventions offered during admission will also be collected. Such data may inform additional analyses including economic analyses.

Sample Size

A total of 1600 subjects will be randomly assigned to receive either intravenous tenecteplase or alteplase in a 1:1 ratio, assuming a loss to follow‐up rate <5%. Based on prior literature, the incidence of primary outcome (mRS score, 0–1) 90 days after randomization is assumed to be 38% and 35%, respectively. ¹ , ² , ³ , ⁴ Assuming a 1‐sided noninferiority margin of 5%, a 1‐sided significance type I error of 2.5% and 90% power to show that tenecteplase is noninferior to alteplase, 759 subjects are needed in each arm of the trial. The choice of 5% as a noninferiority margin represents 50% of the estimate of effect size (10%) for intravenous alteplase administered within 3 hours of stroke symptom onset versus control for the outcome mRS score 0 to 1 measured at 90 days obtained from the largest patient level pooled meta‐analysis of such data. ³ The choice of 5% as the noninferiority margin in this trial means that at least half of the point estimate of effect for intravenous alteplase versus control will be preserved. Additionally, the choice of 5% is strictly less than the lower 95% CI bound of ≈6% on the same point estimate in data from Emberson et al. ³ Hence, the noninferiority margin is guaranteed to be less than the lowest reasonable estimate of alteplase versus control (placebo) effect size.

Statistical Analysis Plan

Primary analysis will be by intention to treat (as randomized) and unadjusted. First, noninferiority will be established if the lower boundary of the 95% CI of the percentage difference in subjects achieving excellent outcome (mRS score, 0–1) in the tenecteplase versus the alteplase arm is >–5% (the noninferiority margin). If noninferiority is demonstrated, then a test of superiority of tenecteplase versus alteplase will be performed as part of secondary analysis. A secondary analysis will be conducted on a per‐protocol population, repeating the approach for the as‐randomized population. Since the trial has pragmatic eligibility criteria, patients who may have been inadvertently enrolled and received thrombolysis beyond 4.5 hours from stroke onset and any treatment crossovers are defined as protocol deviations to be excluded from such analyses. These unadjusted analyses will be supported by fitting a multivariable logistic regression model to provide adjusted estimate of the effectiveness of tenecteplase over alteplase for the primary outcome. The odds ratio of excellent 90‐day outcome (mRS score, 0–1) associated with the treatment groups will be estimated using a logistic regression model after adjusting for age, sex, baseline stroke severity, stroke onset‐to‐needle time, and registry (QuiCR versus OPTIMISE) and include site as a random‐effects variable.

Secondary analyses will evaluate key safety and secondary outcomes using relevant tests of association. Frequency tables will be used to summarize categorical variables by treatment group. Descriptive statistics will be used to summarize continuous data variables by treatment group. In addition, binary, ordinal, and continuous secondary outcomes will be analyzed using generalized mixed‐effects regression models with the appropriate link function. These analyses will be adjusted for age, sex, baseline stroke severity, stroke onset‐to‐needle time, and registry (QuiCR versus OPTIMISE) and include site as a random‐effects variable.

Subgroup analyses will also be secondary and include analyses of primary and relevant secondary and safety outcomes by registry (QuiCR versus OPTIMISE), type of enrolling hospital (primary stroke centers versus comprehensive stroke centers), age (continuous and as <80 years versus ≥80 years), sex (male versus female), baseline stroke severity as measured by the National Institute of Health Stroke Scale (<8, 8–15, and >15), and presence of large‐vessel occlusion on baseline computed tomography angiography. Evidence of a treatment‐by‐subgroup variable interaction will be tested by including a multiplicative interaction term in the model. Subgroup analyses will remain exploratory. A detailed statistical analysis plan will be developed and reviewed by the trial steering committee.

Schedule for interim analyses (at every one‐third of total patients enrolled) will be finalized in consultation with the DSMC. Since tenecteplase is the investigational drug, the overall principle of interim analyses is to determine early if tenecteplase causes more mortality or is significantly inferior to alteplase. Early stopping of the trial for efficacy of tenecteplase is generally to be avoided. The guidance on stopping for safety pertains to a substantial mortality difference favoring alteplase at interim. This may be met if the observed P value for mortality comparing the 2 randomized groups is below a threshold defined using a power family approach to alpha‐spending using φ=1, and if the numeric rate of mortality favors alteplase (eg, if it is found that tenecteplase is substantially and significantly inferior to alteplase in terms of mortality at interim). Details are provided in the AcT trial DSMC charter.

Monitoring

The emphasis in trial execution is on making sure that the “right” patient receives the appropriate intervention (ie, correct randomization, treatment assignment) with adequate assessment of primary outcome (ie, complete, correct, and timely blinded event ascertainment). To align the requirements of good clinical practice with the considerations in a pragmatic randomized clinical trial, a risk‐based approach to monitoring will be used. Central monitoring will be the primary focus with limited on‐site risk‐based monitoring (if required) in coordination with the registry coordinators. The AcT trial portal facilitates such monitoring (Figure 1). The following steps will be in place and monitored: enrollment and randomization, consent, conduct and reporting of data including regular safety outcome monitoring, maintenance of delegation, training and personnel logs, fidelity, accuracy and quality of intervention, and quality of registry data and of data linkages and that of the blinded primary outcome assessments.

Figure 1.

The various features of the AcT trial portal (electronic trial master file).

A, The portal face used for randomization, validation, consent, SAE entries, and referral to documents (customized for each site); (B) consent page showing various flexible consent options that can be accessed via the trial portal; (C) validation page to log any protocol deviations; (D) day 90 assessment summary page (restricted access to blinded assessors only) to record D90 information from the patient/caregiver; and (E) monitoring page for the central site to monitor consents, SAE, deaths, protocol deviation, and data completion status. AcT indicates Alteplase Compared to Tenecteplase in Patients With Acute Ischemic Stroke; and SAE, serious adverse event.

Discussion

The AcT trial is designed as a pragmatic, registry‐linked, prospective, randomized (1:1) controlled, open‐label parallel group clinical trial with blinded end point assessment to evaluate real‐world noninferiority of intravenous tenecteplase as compared with intravenous alteplase in the management of patients with acute ischemic stroke. The trial is an example of a large, simple trial enrolling patients in a real‐world acute stroke care setting. The pragmatic nature of the AcT trial allows for the investigation of the effectiveness of intravenous tenecteplase in comparison with alteplase, where timely delivery of treatment for acute stroke patients is of utmost importance. Recent findings have demonstrated the advantages of embedding pragmatic trials within clinical and administrative registries. ²⁴ , ²⁵ Through linking of data collected directly by the trial with data from ongoing Canadian acute stroke registries that are focused on improving acute stroke care delivery, the AcT trial leverages the high‐quality rigor of a traditional randomized controlled trial design while achieving more efficient data collection, better acceptability, and faster recruitment. Direct costs to the trial are reduced. Moreover, although implemented in an acute care setting, the trial rates high on pragmatism on every Pragmatic Explanatory Continuum Indicator Summary dimension of the explanatory‐pragmatism continuum, thus attesting to the generalizability of the trial results (Figure 2).

Figure 2.

Pragmatism of the AcT trial using the PRECIS‐2 Framework and the other unique features of the AcT trial as assessed at the beginning of the trial.

AcT indicates Alteplase Compared to Tenecteplase in Patients With Acute Ischemic Stroke ; CSC, comprehensive stroke center; QOL, quality of life; PRECIS, Pragmatic Explanatory Continuum Indicator Summary; PSC, primary stroke center; and RCT, randomized controlled trial.

The AcT trial was designed not to significantly affect the routine care of patients with acute stroke. An important feature of the trial therefore was its adoption of deferral of consent for patient enrollment at the time of randomization. This is particularly important in emergency trials, such as in acute stroke, where time is of the essence and most patients lack capacity to provide their own consent or, even if capable, may not understand the complexities of a trial to quickly provide truly informed consent. Patients’ caregivers are also often not available to give timely consent. The process of deferred consent within the AcT trial was designed in consultation with ethicists and patients/caregivers and using a prespecified framework that satisfies ethics principles as outlined in the Tri‐Council Policy Statement on Research Involving Humans (2018). ²⁶ Whenever possible, patients or their families are informed about the trial at presentation. They may decide to opt out at this stage. Formal consent, however, is obtained only after randomization and drug administration and preferably within 7 days of admission. In keeping with the pragmatic nature of the trial, multiple approaches to documenting consent (including in person, electronic consent, and telephone consent) are available, which is especially important to facilitate substitute decision‐maker support. While designed before the COVID‐19 pandemic, this approach has been essential to facilitate trial continuation during times of restricted access for families and caregivers. Another key element of the AcT trial is the need for seamless enrollment of patients into treatment arms in busy acute stroke care settings. To this end, the trial has adopted a fast centralized randomization process using real‐time web‐based servers and with web, text messaging, and phone‐based mechanisms for randomization and drug allocation. The trial's adoption of remote risk‐based monitoring of key trial elements centrally and its use of blinded assessors to assess 90‐day primary and secondary outcomes using a telephone helps limit loss to follow‐up, especially during the ongoing pandemic when face‐to‐face patient visits have become difficult.

In conclusion, results from the AcT trial will provide strong evidence about the effectiveness of intravenous tenecteplase in patients with acute ischemic stroke eligible for intravenous thrombolysis as per the current standard of care in Canada, that is, the Canadian Stroke Best Practice guidelines (Box 5B). ¹⁹ These standards are similar to other national and international guidelines. ²⁷ , ²⁸ , ²⁹ If the AcT trial demonstrates noninferiority of intravenous tenecteplase when compared with alteplase, we believe that the trial will lead to a change in clinical practice.

Sources of Funding

This study was supported by the Canadian Institute for Health Research, Heart and Stroke Foundation of Alberta/University of Calgary, Alberta SPOR SUPPORT Unit.

Disclosures

BKM and MDH report stock in Circle NVI and patents for systems of triage in acute stroke. Others: none.

Supporting information

SVIN.121.000447