← Back
NeuroTrials.ai
Neurology Clinical Trial Database

TRACE-5

Tenecteplase versus standard medical treatment for basilar artery occlusion within 24 h (TRACE-5): a multicentre, prospective, randomised, open-label, blinded-endpoint, superiority, phase 3 trial

Year of Publication: 2026

Authors: Yunyun Xiong, Fana Alemseged, Zhixin Cao, ..., Zhihua Hao

Journal: The Lancet

Citation: Lancet 2026; available online February 5, 2026

Link: https://doi.org/10.1016/S0140-6736(25)02633-9


Clinical Question

Does intravenous tenecteplase (0.25 mg/kg) administered within 24 hours of symptom onset improve functional outcome compared with standard medical treatment in patients with ischaemic stroke due to basilar artery occlusion?

Bottom Line

Tenecteplase within 24 hours significantly improved the rate of excellent functional outcome (mRS 0–1) in basilar artery occlusion (38% vs 29%; adjusted RR 1.50, p=0.014) without increasing symptomatic ICH or mortality. This is the first phase 3 RCT supporting extended-window thrombolysis specifically for confirmed basilar artery occlusion, with benefit preserved regardless of thrombectomy use, and without requiring advanced perfusion imaging for patient selection.

Major Points

  • First phase 3 RCT demonstrating benefit of tenecteplase up to 24 hours specifically in confirmed basilar artery occlusion
  • Primary outcome (mRS 0–1 or return to baseline at 90 days): 38% tenecteplase vs 29% standard treatment (adjusted RR 1.50, 95% CI 1.09–2.08; p=0.014)
  • Ordinal mRS shift analysis was also significant (adjusted common OR 1.51, 95% CI 1.06–2.15; p=0.022)
  • Substantial pre-thrombectomy reperfusion (eTICI 2B–3) was significantly higher with tenecteplase: 24% vs 13% (adjusted RR 1.90, 95% CI 1.04–3.49; p=0.038)
  • Symptomatic ICH within 36 hours was low and similar: 2% vs 3% (adjusted RR 0.58, 95% CI 0.17–1.99)
  • 90-day mortality was similar: 29% vs 31% (adjusted RR 0.87, 95% CI 0.62–1.22), reflecting the severity of basilar artery occlusion
  • 49% of patients underwent endovascular thrombectomy; 35% of controls received alteplase within 4.5 h
  • Subgroup analyses showed consistent benefit across age, sex, NIHSS severity, pc-ASPECTS, occlusion site, time windows, TOAST classification, and thrombectomy status
  • Notably significant benefit in the 6–24 h time window subgroup (adjusted RR 2.51, 95% CI 1.40–4.50) and in patients not undergoing thrombectomy (adjusted RR 1.63, 95% CI 1.04–2.54)
  • Pragmatic design used non-contrast CT/CTA or DWI without advanced perfusion imaging — facilitates implementation in resource-limited settings
  • Single-bolus tenecteplase administration offers practical advantage over 1-hour alteplase infusion
  • Predominantly atherothrombotic mechanism (84%) reflecting Chinese population with high intracranial atherosclerosis prevalence

Design

Study Type: Multicentre, prospective, randomised, open-label, blinded-endpoint (PROBE), superiority, phase 3 trial

Randomization: 1

Blinding: Open-label with blinded endpoint assessment (PROBE design). Outcome assessors and the Endpoint Adjudication Committee were masked to treatment allocation. Randomisation used Central Interactive Management System with computerised allocation sequences.

Enrollment Period: January 24, 2024 to June 20, 2025

Follow-up Duration: 90 days

Centers: 66

Countries: China

Sample Size: 452

Analysis: Intention-to-treat for primary analysis. Modified Poisson regression with robust SE adjusted for age, baseline NIHSS score, and onset-to-randomisation time (<6 h vs 6–24 h). Ordinal logistic regression for mRS shift analysis (mRS 5–6 combined) with proportional odds assumption verified by Brandt test (p=0.40). Per-protocol analysis prespecified as secondary population. No interim analyses. No adjustment for multiple testing. Sample size provided 80% power to detect 12% absolute difference (33% vs 21%) at one-sided alpha 0.025 with 5% attrition. SAS version 9.4.


Inclusion Criteria

  • Age ≥18 years
  • Ischaemic stroke due to basilar artery occlusion (near [99% stenosis with string sign] or complete occlusion on CTA or MRA)
  • Eligible for intravenous thrombolytics within 24 hours of stroke onset or last known well
  • Pre-stroke modified Rankin Scale score ≤3
  • No restriction on NIHSS score

Exclusion Criteria

  • Intracranial haemorrhage or other diagnosis (e.g., brain tumour) on baseline imaging
  • Posterior circulation ASPECTS (pc-ASPECTS) <6 on DWI, non-contrast CT, or CTA source images
  • Significant cerebellar mass effect or acute hydrocephalus
  • Established frank hypodensity on non-contrast CT indicating subacute infarction
  • Bilateral extensive brainstem ischaemia
  • Both anterior and posterior circulation large-vessel occlusion
  • Contraindications to intravenous thrombolysis
  • Already received thrombolysis, endovascular therapy, or tirofiban/batroxobin
  • Pre-stroke mRS score ≥4
  • Clinically evident pregnancy
  • Current participation in another interventional trial
  • Previously enrolled in TRACE-5
  • Not expected to survive a year
  • Contraindication to imaging with contrast agents

Arms

FieldTenecteplase groupControl
InterventionSingle intravenous bolus of tenecteplase 0.25 mg/kg (maximum 25 mg) over 5–10 seconds within 24 hours of symptom onset, with or without endovascular thrombectomy at clinician discretion. 51% underwent thrombectomy.Standard medical treatment including intravenous alteplase (0.9 mg/kg, max 90 mg; 10% bolus then 90% over 60 min) within 4.5 h of onset (used in 35%), anticoagulation (e.g., heparin infusion), or antiplatelets per standard care, with or without endovascular thrombectomy at clinician discretion. 48% underwent thrombectomy.
DurationSingle bolus; 90-day follow-upPer standard of care; 90-day follow-up

Outcomes

OutcomeTypeControlInterventionHR / OR / RRP-value
Proportion of participants with mRS score 0–1 or return to baseline mRS score (if pre-stroke mRS was 2–3) at 90 daysPrimary66/231 (29%)83/221 (38%)p=0.014
mRS 0–2 or return to baseline at 90 daysSecondary99/231 (43%)98/221 (44%)1.1695% CI 0.88–1.54; p=0.29
mRS 0–3 at 90 daysSecondary123/231 (53%)114/221 (52%)1.0695% CI 0.82–1.37; p=0.66
Ordinal distribution of mRS at 90 days (scores 5–6 combined)SecondarymRS 0: 30 (13%), 1: 34 (15%), 2: 34 (15%), 3: 25 (11%), 4: 19 (8%), 5–6: 89 (39%)mRS 0: 36 (16%), 1: 43 (19%), 2: 17 (8%), 3: 18 (8%), 4: 25 (11%), 5–6: 82 (37%)1.5195% CI 1.06–2.15; p=0.022
Early clinical improvement (NIHSS reduction ≥8 or NIHSS 0–1 at 72 h)Secondary64/231 (28%)65/221 (29%)1.0895% CI 0.76–1.52; p=0.68
Substantial reperfusion (eTICI 2B–3) on initial angiogram before thrombectomySecondary16/126 (13%)34/141 (24%)1.995% CI 1.04–3.49; p=0.038
Symptomatic intracranial haemorrhage within 36 hAdverse7/231 (3%)4/221 (2%)0.5895% CI 0.17–1.99; p=0.39
All-cause mortality at 90 daysAdverse71/231 (31%)65/221 (29%)0.8795% CI 0.62–1.22; p=0.41
mRS 5–6 at 90 daysAdverse89/231 (39%)82/221 (37%)0.8795% CI 0.65–1.18; p=0.38

Subgroup Analysis

Prespecified subgroup analyses showed consistent treatment effect across all subgroups with no significant interactions. Notable subgroup findings: Age ≤70 (RR 1.47, 0.98–2.20) vs >70 (RR 1.51, 0.87–2.61), p-interaction=0.98. NIHSS <10 (RR 1.32, 0.89–1.97) vs ≥10 (RR 1.50, 0.84–2.67), p-interaction=0.74. Time <6 h (RR 1.16, 0.73–1.85) vs 6–24 h (RR 2.51, 1.40–4.50) vs wake-up (RR 0.96, 0.37–2.47), p-interaction=0.08. Intention for thrombectomy yes (RR 1.34, 0.86–2.10) vs no (RR 1.68, 1.04–2.72), p-interaction=0.58. Thrombectomy performed — no thrombectomy subgroup (RR 1.63, 1.04–2.54). Alteplase intention yes (RR 1.36, 0.80–2.32) vs no (RR 1.56, 1.03–2.36), p-interaction=0.74. BATMAN score <7 (RR 1.97, 1.03–3.73) vs ≥7 (RR 4.32, 0.74–25.21), p-interaction=0.49. Proximal occlusion (RR 1.40, 0.91–2.16), mid (RR 1.61, 0.90–2.86), distal (RR 2.99, 0.78–11.44), p-interaction=0.77.


Criticisms

  • Open-label design, though mitigated by blinded endpoint assessment (PROBE) and masked adjudication committee
  • Conducted entirely in Chinese patients with high prevalence of intracranial atherosclerosis (84% atherothrombotic mechanism) — generalisability to other ethnicities with different stroke mechanisms (e.g., cardioembolic-predominant populations) is uncertain
  • Heterogeneous control arm: 35% received alteplase, others received anticoagulation or antiplatelets only, complicating interpretation of the comparator
  • 49% of patients underwent thrombectomy in both groups, making it difficult to isolate the independent effect of tenecteplase from combined thrombolysis + thrombectomy
  • Patients with large established posterior circulation infarcts (pc-ASPECTS <6) were excluded, limiting generalisability to those with extensive ischaemia
  • High mortality in both groups (~30%) reflects disease severity but means a large proportion of patients did not contribute to the functional outcome assessment
  • No lower limit on NIHSS resulted in lower baseline severity (median NIHSS 12) than prior basilar artery thrombectomy trials (BAOCHE, ATTENTION), potentially enriching for milder strokes
  • 9% protocol non-adherence rate including crossovers and alteplase use beyond 4.5 h window, though per-protocol results were consistent
  • Tenecteplase formulation used was the same as TRACE-2 (Chinese manufacture); formal comparability data with internationally available formulations have not been published
  • mRS 0–2 and mRS 0–3 secondary endpoints did not reach significance, suggesting benefit was concentrated at the excellent outcome end of the disability spectrum
  • Secondary outcomes and subgroup analyses were not adjusted for multiple testing
  • Median thrombolytic-to-arterial puncture time was 45 min, longer than some trials (e.g., 15 min in TIMELESS), which may have allowed more time for tenecteplase to achieve reperfusion

Funding

Noncommunicable Chronic Diseases-National Science and Technology Major Project (2024ZD0527600, 2024ZD0527605), Beijing Municipal Science Fund for Distinguished Young Scholars (JQ24058), National Natural Science Foundation of China (82171272), and China Shijiazhuang Pharmaceutical Company Recomgen Pharmaceutical (Guangzhou) supplied tenecteplase and an unrestricted grant for infrastructure funding

Based on: TRACE-5 (The Lancet, 2026)

Authors: Yunyun Xiong, Fana Alemseged, Zhixin Cao, ..., Zhihua Hao

Citation: Lancet 2026; available online February 5, 2026

Content summarized and formatted by NeuroTrials.ai.