Determining the Optimal Dose of Tenecteplase Before Endovascular Therapy for Ischaemic Stroke (EXTEND-IA TNK Part 2)

Overview

Patients presenting to the emergency department with acute ischemic stroke, who are eligible for standard intravenous thrombolysis within 4.5 hours of stroke onset will be assessed for major vessel occlusion to determine their eligibility for randomization into the trial. If the patient gives informed consent they will be randomised 50:50 using central computerised allocation to either 0.4mg/kg or 0.25mg/kg intravenous tenecteplase before all participants undergo endovascular thrombectomy. The trial is prospective, randomised, open-label, blinded endpoint (PROBE) design.

The study will be a multicentre, prospective, randomized, open- label, blinded endpoint (PROBE), controlled phase 2 trial (2 arm with 1:1 randomization) in ischemic stroke patients. Randomized patients will first be stratified by both the setting of treatment: metropolitan hospital vs regional hospital (\>1 hour transfer to endovascular centre) vs mobile stroke unit; and by site of baseline arterial occlusion: Intracranial internal carotid artery (ICA) and Basilar artery versus Middle cerebral artery (MCA - M1 and M2); overall resulting in six strata. Imaging is performed with CT or MR (magnetic resonance) acutely as part of standard care with imaging follow-up at 18-30 hours. The sequences and the parameters used follow the STIR (Stroke Imaging Research) roadmap guidelines, but imaging takes place acutely and at 18- 30hrs only, as previously validated. The sample size estimation was based on the proportion of pre-endovascular reperfusion observed in the 0.25mg/kg group from Part 1 of EXTEND-IA TNK (22%). An estimated total sample size of 188 patients (with 94 patients in each of treatment and control arms) yielded 80% power to detect a significant difference of 20% in strata-weighted angiographic reperfusion (mTICI 2b/3) at initial angiogram (22% in 0.25mg/kg vs 42% in 0.4mg/kg arm) at two-sided statistical significance threshold of p=0.05 for superiority. Adaptive increase in sample size will be performed if the result of interim analysis using data from the first 150 patients is promising, as per the methodology of Mehta and Pocock. During the trial, blinded analysis of operational characteristics revealed a 20% reduction in the time from thrombolysis to arterial access versus part 1 due to improved workflow (In the first 150 patients in part 2 median 37min \[IQR 19-54\] versus 46min \[IQR 28-63\] in part 1). This directly impacts the time for thrombolysis to have an effect. A 20% reduction in the hypothesized rate of reperfusion at initial angiogram (18% vs 33%) would require 145 patients per group. Allowing for potential further improvements in workflow the sample size re-estimation was postponed from 150 to 240 patients with a revised minimum sample of 300 patients. Adaptive increase in sample size will be performed if the result of interim analysis using data from the first 240 patients is promising, as per the methodology of Mehta and Pocock. The maximum sample size is capped at 656 patients.