← Back

EXTEND-IA TNK Part 2

Effect of Intravenous Tenecteplase Dose on Cerebral Reperfusion Before Thrombectomy in Patients With Large Vessel Occlusion Ischemic Stroke

Share Share Copied! Compare

Year of Publication: 2020

Authors: Bruce C. V. Campbell, Peter J. Mitchell, Leonid Churilov, ..., Bernard Yan

Journal: JAMA

Citation: JAMA. 2020;323(13):1257–1265. doi:10.1001/jama.2020.1511

Link: https://jamanetwork.com/journals/jama/fullarticle/2762902

Clinical Question

Does a higher dose (0.40 mg/kg) of tenecteplase improve cerebral reperfusion before thrombectomy compared to the standard dose (0.25 mg/kg) in patients with large vessel occlusion ischemic stroke?

Bottom Line

Tenecteplase 0.40 mg/kg did NOT improve cerebral reperfusion prior to EVT vs 0.25 mg/kg (19.3% vs 19.3%; adjusted RR 1.03; P=0.89) in LVO stroke. No differences in functional outcomes, mortality, or sICH. Pooled analysis with original EXTEND-IA TNK confirmed tenecteplase superiority over alteplase (20.0% vs 9.9% reperfusion; RR 1.90; P=0.04). Rural patients had 2x higher reperfusion (34% vs 17%; P=0.001) due to longer lysis-to-puncture time. ICA occlusions: 0% reperfusion.

        Major Points
        No dose-response: identical reperfusion 19.3% each (adjusted RR 1.03; P=0.89). 0.25 mg/kg is sufficient for LVO.
sICH numerically higher with 0.40 mg/kg: 4.7% vs 1.3% (RR 3.50; P=0.12, NS). 4/7 were wire perforation–related.
Rural patients: 34% vs 17% reperfusion (RR 2.15; P=0.001) — longer lysis-to-puncture (152 vs 41 min) allows more thrombolytic dwell time.
ICA occlusions: 0/66 achieved substantial reperfusion. 15% had partial ACA territory recanalization.
Pooled with original EXTEND-IA TNK: tenecteplase 20.0% vs alteplase 9.9% (RR 1.90; P=0.04). Ordinal mRS common OR 1.50 (P=0.04).
mRS 0-2 at 90d: 59% vs 56% (adjusted RR 1.08; P=0.40). Death: 17% vs 15% (P=0.35).
Adaptive re-estimation after 240 patients: conditional power <1% for 15% effect — confirmed futility of expanding.
300 patients, 28 AU/NZ sites, PROBE design. Randomized 1:1 within 4.5h of onset.
Lysis-to-puncture overall: 45 vs 48 min. Rural 167 vs 146 min. Metropolitan 44 vs 41 min.
Supports standardizing tenecteplase at 0.25 mg/kg for EVT-eligible LVO stroke.

      

Design

Study Type: Multicenter, randomized, open-label, blinded-endpoint trial

Randomization: 1

Blinding: Blinded endpoint assessment

Enrollment Period: December 2017 – July 2019

Follow-up Duration: 90 days

Centers: 28

Countries: Australia, New Zealand

Sample Size: 300

Analysis: Modified Poisson regression for primary/secondary outcomes, intention-to-treat; prespecified adaptive sample size re-estimation

Inclusion Criteria

Age ≥18 years
Acute ischemic stroke with large vessel occlusion (ICA, MCA, basilar)
Eligible for IV thrombolysis within 4.5 hours of symptom onset
Planned endovascular thrombectomy

Exclusion Criteria

Pre-stroke mRS >3
Extensive early ischemic change on noncontrast CT
Contraindications to thrombolysis or thrombectomy

Arms

Field	Tenecteplase 0.40 mg/kg	Control
Intervention	Tenecteplase 0.40 mg/kg IV bolus (max 40 mg) before thrombectomy	Tenecteplase 0.25 mg/kg IV bolus (max 25 mg) before thrombectomy
Duration	Single dose	Single dose

Outcomes

Outcome	Type	Control	Intervention	P-value
Substantial reperfusion (>50% of involved territory) prior to thrombectomy	Primary	19.3%	19.3%	0.89
mRS score at 90 days \| 95% CI: 0.74–1.24	Secondary	Median 2	Median 2	0.73
Functional independence (mRS 0–2 or no change) \| 95% CI: 0.90–1.29	Secondary	56%	59%	0.40
Freedom from disability (mRS 0–1 or no change) \| 95% CI: 0.84–1.29	Secondary	49%	49%	0.69
Early neurological improvement \| 95% CI: 0.91–1.27	Secondary	62%	68%	0.39
Death	Adverse		17% (0.40 mg/kg) vs 15% (0.25 mg/kg); P=0.35
Symptomatic intracranial hemorrhage	Adverse		4.7% vs 1.3%; P=0.12
Parenchymal hematoma	Adverse		2.7% vs 4.0%; P=0.52

Subgroup Analysis

Reperfusion benefit seen in rural patients likely due to longer time between thrombolysis and puncture; no dose-by-location interaction

Criticisms

Study not powered to detect minimal clinically important differences (~3–5%)
Wide confidence intervals due to small sample size
Generalizability may be limited to LVO population
Slight imbalance in core volume between arms (non-significant)

Funding

National Health and Medical Research Council of Australia, National Heart Foundation of Australia, iSchemaView (provided RAPID software)

Based on: EXTEND-IA TNK Part 2 (JAMA, 2020)