Study design and patient population

The TASTE-2 is a multicentre, double-blind, randomised, placebo-controlled trial aiming to evaluate the efficacy and safety of edaravone dexborneol in AIS individuals who will receive EVT. All enrolled patients will be randomised with 1:1 to edaravone dexborneol or placebo within 24 hours of symptom onset and receive continuous application for 10–14 days. All these participants will receive standard medical care afterwards and be followed up for 3 months (figure 1).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

TASTE-2 study design. EVT, endovascular therapy; mRS, modified Rankin Scale; NIHSS, National Institute of Health Stroke Scale; TASTE-2, Treatment of Acute Ischaemic Stroke with Edaravone Dexborneol II.

This trial will include patients who are diagnosed with AIS, aged between 18 and 80 years, with a National Institutes of Health Stroke Scale (NIHSS) score between 6 and 25 (range 0–42, with lager values suggesting severe neurological dysfunction) and an Alberta Stroke Program Early Computed Tomography Score (ASPECTS) of 6–10 (range 0–10, with smaller values suggesting larger infarction). All these patients should be characterised with an anterior LVO at intracranial internal carotid artery, T-shaped bifurcation or the M1 segment of middle cerebral artery and are to undergo treatment with EVT within 24 hours after symptom onset, however, prior or concomitant IVT with alteplase is not prohibited if indicated. Additionally, an imaging evaluation of mismatch should be performed through an automated software and the criteria of DEFUSE-320 or DAWN21 should be met for those within 6–24 hours after symptom onset. This software will be used to calculate the infarct core volume, mismatch ratio and mismatch volume (figure 2), of which rationale and details (Patents no. ZL202110266924.6 and ZL202110226735.6) has been described elsewhere.22

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

The inclusion and exclusion criteria of the TASTE-2 trial. TASTE-2, Treatment of Acute Ischaemic Stroke with Edaravone Dexborneol II.

Patients will be excluded if CT scan indicating intracranial haemorrhagic diseases, being given any IVTs other than alteplase before EVT, being allergic to any components of edaravone dexborneol injection, having a history of congenital or acquired haemorrhagic disease, with an elevated systolic blood pressure surpassing 180 mm Hg or diastolic blood pressure exceeding 110 mm Hg after therapy, being in cases where the serum levels of alanine aminotransferase or aspartate transaminase are found to exceed three times the upper limit of normal, or if the current serum creatinine exceeds 1.5 times the upper limit of normal, or if the estimated glomerular filtration rate falls below 60 mL/min, etc (figure 2). Patients who have recently been administered or are currently receiving any Chinese traditional medicine containing borneol should also be excluded. Prior to enrolling patients, the trial must receive the endorsement from the Institutional Review Board at Beijing Tiantan Hospital, Capital Medical University and it is imperative to obtain written informed consent from every participant or their legally designated representative.

Randomisation

The randomisation sequence will be produced by SAS V.9.4 software at the Statistical and Data Center of the National Clinical Research Center for Neurological Diseases in China prior to the trial’s commencement, using a fixed-size block randomisation method. Each unique sequence number will be uniformly labelled on all the packages of the drug thereafter. This number will serve as the identifier of the drug in the smallest packaging unit and does not contain group information. All packaged drugs will be mailed to each site in randomisation sequence order with a minimum of one block. The eligible patients will be stratified based on the participating hospitals and randomly allocated to either the edaravone dexborneol group or the placebo group in equal proportions. After randomisation, the interventional drug should be given as soon as possible before the EVT is performed. Blinding will be maintained throughout the study.

Intervention

Patients allocated to the intervention group will be given intravenous edaravone dexborneol 37.5 mg, comprising 30 mg of edaravone and 7.5 mg of (+)-dexborneol, two times a day over a consecutive period of 10–14 days lasting for 30 min each time. Meanwhile, patients allocated to the control group will be given intravenous placebo of edaravone dexborneol in the same manner. The placebo contains all ingredients except the active edaravone and (+)-dexborneol. The first dose injection of edaravone dexborneol or placebo must precede the treatment of EVT and no later than the puncture of femoral artery. IVT with alteplase will not be prohibited neither before nor after randomisation if indicated. For this trial, EVT will be performed as a fundamental treatment and the reperfusion status of occluded large vessels will be evaluated using the 7-point extended Thrombolysis in Cerebral Infarction (eTICI) scoring system, and a higher eTICI score suggests a more significant reperfusion. Successful reperfusion is determined by reaching an eTICI score of 2b50 or higher, indicating a minimum of 50%–66% restoration of blood flow.23 Other standard of care can be used in both groups except those drugs that seem to be brain cytoprotective.

Outcomes

The primary efficacy outcome is defined as achieving favourable functional independence, which is indicated by a modified Rankin Scale (mRS, a scale of 0–6, scores 0–2 representing independence and scores 3–6 indicating dependence or death) score of 0–2 at 90 days. This score will be assessed in person, or by a real-time video if a face-to-face visit is not feasible, by a certified rater who is blind to the group assignments. This rater will be trained periodically under a theoretical guidance.24 The secondary efficacy outcomes encompass shift analysis of mRS, excellent functional outcome characterised by an mRS score of 0–1 at 90 days, the change in NIHSS score from 10 to 14 days to baseline, a decrease of NIHSS score ≥4 from 10 to 14 days to baseline, and the occurrence of new ischaemic or haemorrhagic stroke within 90 days. The primary safety outcome is the proportion of serious adverse events (SAEs) that occur throughout the entire study period. The secondary safety outcomes include all-cause mortality within 90 days, symptomatic intracranial haemorrhage within 36 hours, deterioration in neurological deficit defined as a ≥4-point increase in the NIHSS score from day 1 to baseline, and the proportion of AEs within 90 days. All SAEs will be evaluated by an independent clinical events committee. The study will also involve the collection of blood samples for further investigation into potential mechanisms (table 1).

Data Safety and Monitoring Board

To uphold ethical standards and prioritise patient safety, the data safety and monitoring board (DSMB) will convene according to the established protocol to carefully monitor the trial’s progress. The DSMB, consisting of non-trial participants, will be entrusted with specific responsibilities outlined by the trial executive committee before its commencement. Following each meeting, DSMB will furnish written recommendations to the trial steering committee chairs.

Sample size

According to the findings of the TASTE trial,19 the edaravone dexborneol group showed a 13.9% relative increase in neurological improvement compared with the edaravone alone group. Taking into account the positive effect of edaravone as a control in the TASTE trial, it is assumed that there will be a 20% relative increase in neurological improvement when comparing edaravone dexborneol with placebo in the upcoming TASTE-2 trial. In ANGLE-ACT, a large cohort study including Chinese patients treated with EVT, 45.4% of patients achieved mRS scores of 0–2 within 90 days.25 Therefore, it is estimated that in TASTE-2, this proportion in the interventional group will be approximately 54% and 45% in the control group. Based on these assumptions, an estimated total sample size of 1362 patients (with a dropout rate of 5%) and an equal distribution of 681 patients per group will provide at least a power of 90% to detect superiority of edaravone dexborneol over placebo in improving neurological function at a two-sided alpha significance level of p=0.05.

Interim analysis

Given the safety report of edaravone dexborneol in clinical studies19 26 as well as the anticipated pace of trial inclusion in China, this study will not conduct interim analyses.

Statistical analyses

The primary efficacy evaluation will be conducted based on an intention-to-treat analysis, while the analysis of safety will use the safety dataset. Differences in the proportion of mRS 0–2 at 90-day between study groups will be assessed through a logistic regression model, with trial centres considered as a random effect. The results will be presented as OR with a 95% CI. Additionally, we will examine whether the treatment effects vary across predefined subgroups by testing the interaction effect between treatment and subgroup using logistic regression models. Similar strategies to those used for analysing the primary outcome will be applied to other categorical outcomes. For shift analysis, an ordinal logistic regression model will calculate a common OR along with its 95% CI for the ordinal 90-day mRS score. All hypothesis tests will adopt a two-sided alpha level of significant set at 0.05 (α=0.05).

A prespecified subgroup analysis will be conducted to assess heterogeneity of treatment effect, with subgroups defined based on the following variables: gender (male or female), age (≤65 or >65), medical history (hypertension, hypercholesterolaemia, diabetes mellitus, stroke and coronary heart disease status, separately) (yes or no), time interval between last known well and randomisation (≤6 hours or >6 hours), NIHSS score range (6–15 or >15), ASPECTS score range (8–10 or 6–7), mismatch(yes or no), bridging therapy administration before EVT procedure (with alteplase or without alteplase), location of LVO site within the intracranial internal carotid artery versus M1 segment of middle cerebral artery, and reperfusion status of occluded vessels after EVT procedure according to eTICI scale categories 2b-3 vs 0–2 a.

The statistical analyses will be conducted by using SAS V.9. 4 or a higher version of the statistical software, in accordance with a predetermined statistical analysis plan that we will provide before the database is unmasked and locked.

Study organisation

This study will be carried out within approximately a hundred centres around mainland China from February 2022 and last about 24 months. To control centre effect, each centre must be experienced in EVT and has performed at least 50 thrombectomy therapies in the past calendar year and the median door-to-puncture time must be ≤120 min. During the study, an exit mechanism will be performed. All patients included will be observed to check their reactions to the treatment and can withdraw the informed content at any time for any reason based on the principles of Good Clinical Practice. Responsibility for the trial’s organisation lies with the executive and steering committees. The executive committee, led by the principal investigator and assisted by the steering committee, will supervise all the trial activities including protocol development and amendment during the study period. The executive committee is situated at the TASTE-2 Trial Coordinating Centre, which is part of the China National Clinical Research Centre for Neurological Diseases. Meanwhile, the steering committee will consist of recognised experts in neurology and statistics, providing professional supervision and strategic advice. The committee is scheduled to convene semiannually. Trained assessors, who are unaware of the treatment assignment, will adjudicate outcomes, SAEs and brain imaging.

Version modification

The protocol V.1.0 underwent a minor modification, resulting in the current version being 1.1. In this updated version, we included shift analysis of mRS at 90 days as an additional measure of secondary efficacy outcome without any disruption or impact on the trial’s executive processes.