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.

# 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 *The Lancet 2026* Articles 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 Yunyun Xiong*, Fana Alemseged*, Zhixin Cao, Lee H Schwamm, Si Zhang, Mark W Parsons, Marc Fisher, Yahui Hao, Aoming Jin, Jinfeng Yin, Yong Jiang, Fengyuan Che, Lihua Wang, Li Zhou, Hongguo Dai, Yutie Zhao, Chunmiao Duan, Shuangzhe Wu, Ganghua Feng, Lixia Zong, Wanxing Ye, Ziran Wang, Ziqi Xu, Hao Wang, Manjun Hao, Yujie Ma, Xia Meng, Hao Li, Zixiao Li, Yilong Wang, Liping Liu, Xingquan Zhao, Bruce C V Campbell†, Yongjun Wang†, for the TRACE-5 investigators‡ Summary Background The ecacy and safety of intravenous thrombolysis with tenecteplase within 24 h after stroke onset due Lancet 2026; 407: 763–72 to basilar artery occlusion are not well studied. We aimed to assess whether intravenous tenecteplase administered Published Online within 24 h after symptom onset improved functional outcome compared with standard medical treatment in patients February 5, 2026 with basilar artery occlusion. https://doi.org/10.1016/ S0140-6736(25)02633-9 See Comment page 734 Methods TRACE-5 was a prospective, randomised, open-label, blinded-endpoint, superiority, phase 3 trial conducted *Contributed equally at 66 stroke centres in China. We included patients aged 18 years or older with stroke due to basilar artery occlusion †Joint senior authors who were eligible for intravenous thrombolytics within 24 h of stroke onset or the time they were last known to be well and had a pre-stroke modified Rankin scale (mRS) score of 3 or less (scores range from 0 to 6, with higher scores ‡TRACE-5 investigators are listed in the appendix (pp 5–7) indicating greater disability). Patients were randomly assigned to receive a single intravenous bolus of tenecteplase Department of Neurology, (0·25 mg/kg; maximum 25 mg) within 24 h after symptom onset or standard medical treatment (which could include Beijing Tiantan Hospital, intravenous alteplase at 0·9 mg/kg, maximum 90 mg, within 4·5 h of symptom onset; anticoagulation; or Capital Medical University, antiplatelets), with or without endovascular thrombectomy. The primary outcome was a score of 0–1 on the mRS or Beijing, China (Prof Y Xiong MD, return to the baseline mRS score (if the baseline pre-stroke mRS score was 2–3) at 90 days. Safety outcomes were Z Cao MD, S Wu MD, L Zong MD, M Hao MD, Y Ma MD, symptomatic intracranial haemorrhage and death. The primary outcome and safety outcomes were assessed in all Prof Z Li MD, Prof Yi Wang MD, randomly assigned participants included in their originally assigned groups. This trial is registered with ClinicalTrials. Prof L Liu MD, Prof X Zhao MD, gov, NCT06196320. Prof Yo Wang MD); Department of Neurology, Royal Melbourne Hospital, University of Findings Between Jan 24, 2024, and June 20, 2025, 452 patients were enrolled (mean age 66·4 years [SD 11·2], Melbourne, Parkville, VIC, 321 [71%] males, and 131 [29%] females), of whom 222 (49%) subsequently underwent thrombectomy; 221 were randomly Australia (F Alemseged MD, assigned to receive tenecteplase and 231 to receive standard medical treatment. Alteplase was used in 80 (35%) of the Prof B C V Campbell PhD); patients in the standard medical treatment group. An mRS score of 0–1 or return to the baseline mRS score occurred in Department of Neurology, Yale School of Medicine, New 83 (38%) patients in the tenecteplase group and 66 (29%) patients in the standard medical treatment group (adjusted Haven, CT, USA relative rate 1·50 [95% CI 1·09–2·08], p=0·014). Symptomatic intracranial haemorrhage within 36 h occurred in (Prof L H Schwamm MD); four (2%) patients in the tenecteplase group and seven (3%) patients in the standard medical treatment group (adjusted Department of Clinical relative rate 0·58 [95% CI 0·17–1·99]). All-cause mortality at 90 days was similar between groups (65 [29%] patients in Research Centre, Chinese Stroke Association, Beijing, the tenecteplase group and 71 [31%] patients in the standard medical treatment group; adjusted relative rate 0·87 [95% CI China (S Zhang BEng, 0·62–1·22]), as was the proportion of patients with an mRS score of 5–6 at 90 days (82 [37%] vs 89 [39%]; 0·87 [0·65–1·18]). Y Hao BSc); Department of Neurology, Liverpool Hospital, Interpretation In this trial involving Chinese patients with ischaemic stroke due to basilar artery occlusion, University of New South Wales, Sydney, NSW, Australia tenecteplase within 24 h after stroke onset improved functional outcome compared with standard medical treatment. (Prof M W Parsons PhD); The incidence of symptomatic intracranial haemorrhage and death was similar. Department of Neurology, Beth Israel Deaconess Medical Funding Noncommunicable Chronic Diseases-National Science and Technology Major Project, Beijing Municipal Centre, Harvard Medical School, Boston, MA, USA Science Fund for Distinguished Young Scholars, National Natural Science Foundation of China, and China (Prof M Fisher MD); China Shijiazhuang Pharmaceutical Company Recomgen Pharmaceutical (Guangzhou). National Clinical Research Centre for Neurological Diseases, Beijing Tiantan Copyright © 2026 Elsevier Ltd. All rights reserved, including those for text and data mining, AI training, and similar Hospital, Beijing, China technologies. (Prof Y Xiong, A Jin PhD, J Yin PhD, Y Jiang PhD, W Ye PhD, Introduction of disability and mortality (up to 80–90%) if recanalisation Prof X Meng MD, H Li PhD, Prof Yi Wang, Prof L Liu, Basilar artery occlusion is a rare but devastating subtype does not occur.1,2 An individual patient data meta-analysis Prof Yo Wang); Department of of large-vessel occlusion ischaemic stroke with high rates of four randomised controlled trials supports the robust Neurology, Linyi People’s Articles Hospital, Linyi, China Research in context (Prof F Che MD, Z Wang MD, H Wang MD); Department of Evidence before this study patients with mild strokes, did not require basilar artery Neurology, Second Affiliated Hospital of Harbin Medical Endovascular thrombectomy is established as an effective occlusion, and excluded patients intended for thrombectomy. University, Harbin, China treatment for basilar artery occlusion within 24 h of stroke Added value of this study (Prof L Wang MD); Department onset according to four phase 3 randomised controlled trials. of Neurology, Weifang People’s TRACE-5 is the first phase 3 trial to compare tenecteplase with However, access to endovascular thrombectomy remains Hospital, Weifang, China standard medical treatment in patients with basilar artery (L Zhou MD); Department of limited worldwide, particularly in low-income and middle- occlusion presenting within 24 h of symptom onset. Our Emergency, Linfen Central income countries. As the first-line therapy for ischaemic stroke Hospital, Linfen, China within 4·5 h, intravenous thrombolysis is widely used and findings showed that tenecteplase administered within 24 h of (H Dai MD); Department of symptom onset improved excellent functional outcomes might offer additional benefits for patients with basilar artery Emergency, Puyang Oilfield without increased safety concerns for patients with basilar General Hospital, Puyang, occlusion presenting within 24 h after onset, especially with the artery occlusion. This pragmatic trial allowed the use of China (Y Zhao MD); use of the newer-generation thrombolytic, tenecteplase. Department of Neurology, We searched PubMed and MEDLINE for randomised trials alteplase in the control group and thrombectomy in both Daxing Teaching Hospital, groups, closely reflecting real-world clinical practice. published between Jan 1, 2000, and Oct 31, 2025, using the Capital Medical University, Beijing, China (C Duan MD); terms “tenecteplase”, “basilar artery occlusion”, and “study” or Implications of all the available evidence Department of Neurology, First “clinical trial”. To date, no phase 3 randomised controlled trial The TRACE-5 trial supports tenecteplase as an effective option for People’s Hospital of Chenzhou, has been published evaluating the efficacy and safety of treating basilar artery occlusion within 24 h, both in centres that Chenzhou, China (G Feng MD); tenecteplase in patients with confirmed basilar artery occlusion can provide endovascular thrombectomy and those that cannot. Department of Neurology, First Affiliated Hospital, Zhejiang within 24 h of symptom onset. One phase 3 trial compared Using single-bolus tenecteplase to complement endovascular University School of Medicine, alteplase with control, administered between 4·5 h and 24 h in thrombectomy without advanced imaging for patient selection Hangzhou, China (Z Xu MD); patients with posterior circulation stroke, but it mainly included enhances its feasibility for future implementation. Advanced Innovation Centre for Human Brain Protection, Beijing Tiantan Hospital, Beijing, China (Prof Yo Wang); benefit of endovascular thrombectomy for basilar artery endovascular thrombectomy versus alteplase in basilar Research Unit of Artificial occlusion within 24 h of stroke onset.3–7 However, artery occlusion, potentially oering additional benefit Intelligence in Cerebrovascular immediate access to endovascular thrombectomy is for these patients.16,17 Disease, Chinese Academy of currently limited, with a median global thrombectomy The Tenecteplase Reperfusion Therapy in Acute Medical Sciences, Beijing (Prof Yo Wang) access rate of about 3% and access rates of less than Ischaemic Cerebrovascular Events-5 (TRACE-5) trial was Correspondence to: 1% in 27% of countries.8,9 Therefore, intravenous designed to test whether intravenous tenecteplase at a Prof Yongjun Wang, Department thrombolytic therapies within 4·5 h of onset remain the dose of 0·25 mg per kilogram of bodyweight, of Neurology, Beijing Tiantan standard of care in basilar artery occlusion, particularly administered within 24 h after symptom onset, improved Hospital, Capital Medical in patients who initially present to stroke centres without functional outcome compared with standard medical University, Beijing 100070, China yongjunwang@ncrcnd.org.cn the capacity to perform endovascular thrombectomy.10,11 treatment in patients with basilar artery occlusion. The or Given the dismal prognosis of basilar artery occlusion, use of endovascular thrombectomy was at the discretion Prof Bruce C V Campbell, further research on whether the time window for of the treating physician. Department of Neurology, Royal intravenous thrombolysis can be extended up to 24 h is Melbourne Hospital, Parkville, warranted. Methods VIC 3050, Australia A previous trial showed the possibility of extending the Study design and participants bruce.campbell@mh.org.au time window for intravenous thrombolysis to 24 h in TRACE-5 was a prospective, randomised, open-label, See Online for appendix patients with posterior circulation stroke using alteplase.12 blinded-endpoint, superiority, phase 3 trial conducted at However, the trial population primarily included patients 66 stroke centres in China. The trial included patients with mild strokes, basilar artery occlusion was not with acute ischaemic stroke due to basilar artery required, and those scheduled for thrombectomy were occlusion who were aged 18 years or older and eligible for excluded. Therefore, evidence for thrombolytics, with or intravenous thrombolytics within 24 h of stroke onset or without endovascular thrombectomy, in the extended the time they were last known to be well. Basilar artery time window for patients with basilar artery occlusion is occlusion was defined as a potentially retrievable scarce. Tenecteplase, a modified form of human tissue occlusion of the basilar artery, which could be a near plasminogen activator with higher fibrin specificity and (defined as 99% stenosis with string sign) or complete longer half-life,11 was shown to be non-inferior to alteplase occlusion of the basilar artery on CT angiography (CTA) in randomised trials and even superior to alteplase by a or magnetic resonance angiography (MRA). There was study-level meta-analysis,13 and proven to be eective for no restriction on the National Institutes of Health Stroke anterior circulation large-vessel occlusion in patients not Scale (NIHSS) score (which ranges from 0 to 42, with receiving endovascular thrombectomy within 24 h after higher scores representing increased neurological onset with a numerically higher rate of symptomatic deficits). Eligible patients had a pre-stroke modified intracranial haemorrhage.14,15 Tenecteplase was also Rankin scale (mRS) score of 3 or less (scores range from associated with increased reperfusion before 0 to 6, with higher scores indicating greater disability). 764 Articles Patients were ineligible if there was an intracranial on clinician judgement. Other management for haemorrhage or other diagnosis (eg, brain tumour) endovascular thrombectomy patients adhered to the identified by baseline imaging; posterior circulation Chinese Guideline for Endovascular Treatment of Acute Acute Stroke Prognosis Early CT Score (pc-ASPECTS)18 Ischemic Stroke 2023,21 which recommends endovascular of less than 6 (scores range from 0 [extensive ischaemia, thrombectomy for acute basilar artery occlusion within poor prognosis] to 10 [no ischaemia, excellent prognosis]19) 12 h of onset if the inclusion criteria of the Endovascular on diusion weighted imaging or non-contrast CT or Treatment for Acute Basilar-Artery Occlusion CTA source images; significant cerebellar mass eect or (ATTENTION)6 and Basilar Artery Occlusion Chinese acute hydrocephalus; established frank hypodensity on Endovascular (BAOCHE)7 trials are met (class I; level of non-contrast CT indicating subacute infarction; or evidence A). For acute basilar artery occlusion, bilateral extensive brainstem ischaemia. The complete endovascular thrombectomy within 12–24 h of onset is list of eligibility criteria is provided in the appendix ( p 9). recommended if the BAOCHE inclusion criteria are met The trial adhered to the principles outlined in the (class IIa; level of evidence B). Declaration of Helsinki and complied with the All outcome assessments were conducted by certified International Council for Harmonization Guidelines for clinicians masked to treatment allocation and trained in Good Clinical Practice. The study was approved by the their administration at each participating site. The ethics committee of Beijing Tiantan Hospital screening log systematically captured demographic data, (KY2023-227-02). There was no patient or public medical history, concomitant medications, laboratory involvement in the study design, conduct, and reporting. parameters, neurological evaluations (including mRS Before participant enrolment, the trial protocol received and NIHSS scores), and imaging findings. Non-contrast approval from the institutional review board of each CT or MRI at 24–36 h post intervention evaluated participating site and was subsequently published.20 All intracranial haemorrhage, and 90-day mRS scores were of the enrolled patients or their legally authorised determined through in-person visits or structured representatives provided written informed consent. This telephone interviews. trial is registered with ClinicalTrials.gov, NCT06196320. Study monitoring was managed by the TRACE-5 Coordinating Centre. Monitors conducted on-site visits Randomisation and masking to verify protocol compliance by auditing study records, Patients were randomly assigned (1:1) to intravenous cross-referencing source documentation, and engaging tenecteplase or standard medical treatment with the with investigators to assess trial conduct. Monitors were Central Interactive Management System for Clinical tasked with ensuring protocol adherence and verifying Research, which is a secure, Good Clinical Practice- accurate completion of electronic case report forms. compliant electronic web-based server that robustly preserves allocation concealment. The randomisation Outcomes system used computer-generated allocation sequences, The primary outcome, which was chosen on the basis of stratified by the investigator’s intention for endovascular precedent in previous thrombolytic trials, was the thrombectomy and intention for intravenous alteplase (if proportion of participants with no disability (defined as randomly assigned to standard medical treatment), and an mRS score of 0–1) or return to the baseline mRS score using covariate-adjusted minimisation to balance age at 90 days (if the baseline pre-stroke mRS score was 2–3). (≤70 years vs >70 years), NIHSS score (<10 vs ≥10), and Secondary outcomes were the proportion of participants time from onset to randomisation (<6 h vs 6–24 h). with functional independence (defined as an mRS score Qualified physicians and members of the Endpoint of 0–2) or return to the baseline mRS score at 90 days, Adjudication Committee, who were masked to treatment the proportion of patients with an mRS score of 0–3 at allocation, independently conducted study outcome 90 days, the ordinal distribution of mRS scores at 90 days assessments and endpoint event adjudication, (scores 5–6 were combined to avoid counting a shift from respectively. death [mRS score of 6] to severe disability requiring constant nursing care [mRS score of 5] as a treatment Procedures success), the proportion of patients with early clinical Immediately after randomisation, the intervention group improvement (reduction in acute 72 h NIHSS score of ≥8 received tenecteplase as a single intravenous bolus over points compared with the baseline score or 72 h NIHSS 5–10 s, dosed at 0·25 mg/kg (maximum 25 mg). The score of 0–1), and the proportion of patients with standard medical treatment group could receive complete occlusion at baseline who reached an expanded intravenous alteplase (0·9 mg/kg, maximum 90 mg; Thrombolysis in Cerebral Infarction22 grade of 2B–3 on starting with an initial bolus of 10% of the total dose, initial digital subtraction angiography run before followed by continuous infusion of the remaining 90% thrombectomy (grades range from 0 [no reperfusion] to 3 over 60 min) within 4·5 h of stroke onset, anticoagulation, [complete reperfusion]). such as heparin infusion, or antiplatelets as per standard Safety outcomes were symptomatic intracranial care, with or without endovascular thrombectomy based haemorrhage within 36 h (defined according to the Safe Articles Implementation of Thrombolysis in Stroke-Monitoring central review of all brain imaging performed during Study criteria as local or remote parenchymal admission). haemorrhage type 2, combined with a neurological The data and safety monitoring board monitored safety deterioration of ≥4 NIHSS points, or leading to death),23 outcomes after the first 50 patients were enrolled and all-cause mortality within 90 days, and the proportion of every subsequent 100 patients. Severe adverse events patients with an mRS score of 5–6 at 90 days. Adverse within 90 days were also recorded. eects were systematically assessed at 24 h, 7 days or discharge, and 90 days from randomisation (including Statistical analysis On the basis of previous studies,6,7,17,24 the sample size of 452 patients provided 80% power to detect a 12% dierence in the proportions of patients reaching an 1043 patients underwent screening mRS score of 0–1 (33% intervention vs 21% control), assuming a one-sided alpha level of 0·025, an attrition 591 excluded rate of 5%, and a 3-month follow-up period. No interim 330 did not meet inclusion criteria analyses were planned or performed. 249 no near or complete basilar artery occlusion 69 no consent obtained Analyses, including safety, adhered to the intention-to- 12 time from onset to randomisation >24 h treat principle, with all randomly assigned participants 261 met exclusion criteria 70 both anterior and posterior circulation included in their originally assigned groups. Per-protocol large-vessel occlusion analysis was performed in the per-protocol population, 64 pc-ASPECTS <6 39 contraindications to intravenous thrombolysis which comprised all participants who completed the 28 established frank hypodensity on non-contrast CT assigned interventions as required and in whom there 22 bilateral extensive brainstem ischaemia were no major violations of the trial protocol. Major 17 based on investigator judgement 10 already received thrombolysis violations defined during data audit included substantial 5 already received endovascular therapy deviation from inclusion or exclusion criteria, use of 2 already received tirofiban or batroxobin 9 significant cerebellar mass effect or acute disallowed concomitant treatments that seriously hydrocephalus interfere with ecacy assessment after inclusion, 6 intracranial haemorrhage or tumour substantial non-compliance with the assigned treatment 1 contraindication to imaging with contrast agents 1 pre-stroke mRS score ≥4 regimen, excessive delay in drug administration, or non- 1 a clinically evident pregnancy adherence to the scheduled follow-up windows. 1 current participation in other interventional trial 1 previously enrolled in TRACE-5 Per-protocol analysis was prespecified and considered as 1 not expected to survive a year a secondary population for ecacy analysis. Outcomes and statistical analyses were prospectively predefined as primary, secondary, or exploratory in the trial protocol. 452 randomly assigned Treatment eect estimates are reported as point estimates with corresponding 95% CIs. Descriptive statistics (means and SDs, medians and IQRs, and numbers and 221 assigned to tenecteplase and included 231 assigned to standard medical proportions) were computed for all study variables, in intention-to-treat population treatment and included in intention- including baseline characteristics and outcome to-treat population measures. We did not have missing data for the primary or secondary outcomes. 221 included in safety population 231 included in safety population A prespecified statistical analysis plan was finalised before database lock. The primary outcome and 11 excluded 29 excluded dichotomous secondary outcomes were analysed using 6 enrolled inappropriately 5 enrolled inappropriately modified Poisson regression with robust SE25 adjusted 4 crossover to standard medical 4 crossover to tenecteplase for age, baseline NIHSS score, and onset-to- treatment group group 1 follow-up visits outside the 16 alteplase outside 4·5 h time randomisation time (dichotomised as <6 h vs 6–24 h). predefined time window window Treatment eects were expressed as adjusted relative 1 thrombolytic drug dose <80% standard dose rates accompanied by 95% CIs. Ordinal analysis of 1 follow-up visits outside the 90-day mRS outcomes, merging scores 5 and 6, used predefined time window ordinal logistic regression, adjusted for the same 2 used contraindicated medications within 24 h after covariates, contingent on validation of the proportional thrombolytic therapy odds assumption. Prespecified subgroups included age (≤70 years 210 included in per-protocol population 202 included in per-protocol population vs >70 years), sex (male vs female), baseline severity of stroke (NIHSS score <10 vs ≥10), pc-ASPECTS (≤7 vs 8–10), intention for intravenous alteplase (yes vs no), intention Figure 1: Trial profile mRS=modified Rankin scale. pc-ASPECTS=posterior circulation Acute Stroke Prognosis Early CT Score. for endovascular thrombectomy (yes vs no), early (≤60 min 766 Articles after randomisation) versus delayed (>60 min after Tenecteplase group Standard medical randomisation) endovascular thrombectomy versus no (n=221) treatment group endovascular thrombectomy, time from stroke onset to (n=231) randomisation (<6 h vs 6–24 h vs wake-up stroke), pre- Age, years 66·5 (10·9) 66·2 (11·6) stroke mRS score (0–1 vs 2–3), glucose concentration Sex (<7·8 mmol/L vs ≥7·8 mmol/L), site of basilar artery Male 159 (72%) 162 (70%) occlusion (distal vs mid vs proximal), stroke mechanism Female 62 (28%) 69 (30%) by Trial of Org 10172 in Acute Stroke Treatment (TOAST) Hypertension 165 (75%) 170 (74%) classification (cardioembolic stroke vs embolic stroke of Diabetes 55 (25%) 58 (25%) undetermined source vs atherothrombotic stroke), Basilar Atrial fibrillation 16 (7%) 21 (9%) Artery Tomography Angiography (BATMAN) score on mRS score before stroke* baseline CTA (<7 vs ≥7, scores range from 0 [extensive 0–1 211 (95%) 220 (95%) thrombus in the basilar artery, poor collaterals] to 10 [no 2–3 10 (5%) 11 (5%) thrombus in the basilar artery, robust collaterals]),26 and Baseline NIHSS score† 13 (7–25) 11 (6–21) Posterior Circulation Collateral Score (PC-CS) on baseline <10 86 (39%) 99 (43%) CTA (0–3 vs 4–5 vs ≥6; PC-CS scores range from 0 [no ≥10 135 (61%) 132 (57%) collateral flow] to 10 [robust collateral circulation]).27 Basilar artery occlusion‡ Unadjusted ecacy and safety analyses were performed as post-hoc analyses in the intention-to-treat and per- Complete occlusion 208/218 (95%) 220/230 (96%) protocol populations. We also did a post-hoc sensitivity Near-complete occlusion 10/218 (5%) 10/230 (4%) analysis of ecacy and safety outcomes, including Basilar artery occlusion site adjustment for patient or family refusal to consent for Proximal 126/218 (58%) 125/230 (54%) endovascular thrombectomy, along with additional Mid 64/218 (29%) 74/230 (32%) endovascular thrombectomy characteristics, recanali- Distal 28/218 (13%) 31/230 (13%) sation at end of procedure, haemorrhagic transformation pc-ASPECTS§ 8 (6–8) 8 (7–8) classifications, and exploratory post-hoc subgroup ≤7 91 (41%) 88 (38%) analysis stratified by imaging modality (CT vs MRI). 8–10 130 (59%) 143 (62%) Significance was defined as two-sided p value of less TOAST classification than 0·05. No adjustment for multiple testing was made, Large-artery atherosclerosis 191 (86%) 190 (82%) and CIs for secondary outcomes and subgroups should Cardioembolism 29 (13%) 40 (17%) not be used for hypothesis testing. Embolic stroke of undetermined source 1 (<1%) 1 (<1%) Analyses were conducted with SAS software Time from symptom onset to randomisation, h 6·1 (3·8–11·3) 6·5 (3·4–11·6) (version 9.4). <4·5 70 (32%) 85 (37%) 4·5–24 113 (51%) 102 (44%) Role of the funding source Wake-up stroke 38 (17%) 44 (19%) The funders of the study had no role in study design, Intravenous alteplase¶ 1 (<1%) 80 (35%) data collection, data analysis, data interpretation, or Intention at time of randomisation to perform 130 (59%) 125 (54%) writing of the report. No confidentiality agreements endovascular thrombectomy existed between the manufacturer and investigators, and Endovascular thrombectomy 112 (51%) 110 (48%) the manufacturer had no authority to delay or restrict Data are mean (SD), n (%), or median (IQR). mRS=modified Rankin scale. NIHSS=National Institutes of Health Stroke publication of trial findings. Scale. pc-ASPECTS=posterior circulation Acute Stroke Prognosis Early CT Score. TOAST=Trial of Org 10172 in Acute Stroke Treatment. *Scores on the mRS range from 0 to 6, with higher scores indicating greater disability. †Scores on Results the NIHSS range from 0 to 42, with higher scores indicating a greater deficit. ‡Four patients did not have near- complete or complete basilar artery occlusion on core laboratory assessment. §Scores on the pc-ASPECTS were Between Jan 24, 2024, and June 20, 2025, 1043 patients assessed on baseline CT angiography source images, non-contrast CT brain images, or diffusion weighted imaging, were screened across sites. Following eligibility ranging from 0 to 10, with higher scores indicating fewer early ischaemic changes. ¶Four patients assigned to tenecteplase crossed over to control (one received alteplase and three received no intravenous thrombolysis) but were assessment, 452 qualified for study inclusion, and analysed in the tenecteplase group as per the intention-to-treat principle. 221 patients were randomly assigned to the tenecteplase group and 231 to the standard medical treatment group Table 1: Baseline characteristics (figure 1). Protocol non-adherence was recorded in 40 (9%) patients, including crossover, alteplase use beyond 4·5 h, and inappropriate enrolment (figure 1). thrombolytic to arterial puncture for the 222 (49%) patients Baseline characteristics are shown in table 1 and the receiving endovascular thrombectomy was 45 min appendix (pp 11–12). Mean age was 66·4 years (SD 11·2); (IQR 25–80). The reasons for not proceeding with 321 (71%) patients were male and 131 (29%) were female. endovascular thrombectomy in the study overall and in the The median baseline NIHSS score was 12 (IQR 7–23). The subgroup in whom endovascular thrombectomy was median interval from symptom onset to randomisation intended at the time of randomisation are shown in the was 6·4 h (IQR 3·7–11·5). The median interval from appendix (pp 13–14). Alteplase was used in Articles Tenecteplase group Standard medical Adjusted effect size* p value (n=221) treatment group (95% CI)† (n=231) Primary outcome mRS score 0–1 or return to baseline mRS score at 90 days‡ 83 (38%) 66 (29%) 1·50 (1·09–2·08) 0·014 Secondary outcomes mRS score of 0–2 or return to baseline mRS score at 90 days 98 (44%) 99 (43%) 1·16 (0·88–1·54) 0·29 mRS score of 0–3 at 90 days 114 (52%) 123 (53%) 1·06 (0·82–1·37) 0·66 Ordinal distribution of mRS scores at 90 days ·· ·· 1·51 (1·06–2·15) 0·022 0 36 (16%) 30 (13%) ·· ·· 1 43 (19%) 34 (15%) ·· ·· 2 17 (8%) 34 (15%) ·· ·· 3 18 (8%) 25 (11%) ·· ·· 4 25 (11%) 19 (8%) ·· ·· 5–6§ 82 (37%) 89 (39%) ·· ·· Early clinical improvement¶ 65 (29%) 64 (28%) 1·08 (0·76–1·52) 0·68 Substantial reperfusion at initial angiogram|| 34/141 (24%) 16/126 (13%) 1·90 (1·04–3·49) 0·038 Safety outcomes Symptomatic intracranial haemorrhage within 36 h** 4 (2%) 7 (3%) 0·58 (0·17–1·99) 0·39 All-cause mortality within 90 days 65 (29%) 71 (31%) 0·87 (0·62–1·22) 0·41 mRS score 5–6 at 90 days 82 (37%) 89 (39%) 0·87 (0·65–1·18) 0·38 Data are n (%) or n/N (%) unless otherwise stated. mRS=modified Rankin scale. NIHSS=National Institutes of Health Stroke Scale. *The common odds ratio is shown for the ordinal score on the mRS, and the relative rate is shown for other outcomes. †The primary, secondary, and safety outcomes were adjusted for age, baseline NIHSS score, and onset-to-randomisation time (dichotomised as <6 h vs 6–24 h). ‡Scores on the mRS range from 0 to 6, with higher scores indicating greater disability; the primary outcome also included patients who returned to their baseline mRS score (if the baseline pre-stroke mRS score was 2–3) at 90 days. §mRS scores 5–6 were prespecified to be merged in ordinal analysis to avoid counting a shift from death (mRS score 6) to severe disability requiring constant nursing care (mRS score 5) as a treatment success. ¶Early clinical improvement defined by a reduction on NIHSS score of ≥8 compared with the initial deficit or a score of 0–1 at 72 h. ||Proportion of patients with complete occlusion at baseline who reached the expanded Treatment in Cerebral Infarction grade of 2B–3 (reperfusion of ≥50% of the affected territory) on initial digital subtraction angiography run before thrombectomy. Two patients did not go for digital subtraction angiography due to significant improvement of NIHSS scores, and repeated CT angiography showed complete recanalisation of occlusion (Arterial Occlusive Lesion score of 3). **Symptomatic intracranial haemorrhage within 36 h (defined as local or remote parenchymal haemorrhage type 2, combined with a neurological deterioration of ≥4 NIHSS points, or leading to death). Table 2: Efficacy and safety outcomes figure 2). In patients with a pre-stroke mRS score of 2–3, four (40%) of ten in the tenecteplase group versus two (18%) of 11 in the standard medical treatment Standard medical group returned to their baseline mRS scores. Secondary treatment group 13% 15% 15% 11% 8% 8% 31% outcome measures and safety data are shown in table 2. (n=231) The proportional odds assumption for ordinal logistic Tenecteplase regression (scores 5 and 6 were combined) was satisfied group 16% 19% 8% 8% 11% 8% 29% based on the Brandt test (p=0·40). Stratified subgroup (n=221) evaluations of treatment eects on the primary endpoint 0 25 50 75 100 are shown in figure 3. Results in the per-protocol analysis were aligned with those of the primary intention-to-treat Figure 2: Distribution of scores on the mRS at 90 days in the intention-to-treat population analysis (appendix pp 15–16, 26). The appendix shows Scores on the modified Rankin scale range from 0 to 6, with 0 indicating no symptoms, 1 indicting symptomatic post-hoc ecacy and safety outcomes using unadjusted but not disabled, 2 indicating disabled but independent, 3 indicating dependent but ambulatory, 4 indicating not analyses in the intention-to-treat and per-protocol ambulatory or capable of body self-care, 5 indicating requiring constant nursing care, and 6 indicating death. analysis (pp 17–20) and post-hoc analysis including an Percentages might not total 100 because of rounding. mRS=modified Rankin scale. additional adjustment for patient or family refusal to 80 (35%) of 231 patients in the standard medical treatment consent to endovascular thrombectomy (pp 21–22). group. The appendix (p 23) shows post-hoc endovascular The primary endpoint of no disability (mRS score 0–1) thrombectomy procedural characteristics, including final or return to the baseline mRS score at the 90-day recanalisation. Post-hoc exploratory subgroup analysis follow-up was reached in 83 (38%) of 221 patients in the stratified by imaging modality is provided in the appendix tenecteplase group versus 66 (29%) of 231 patients in the (p 27). standard medical treatment group (adjusted relative Symptomatic intracranial haemorrhage within 36 h rate 1·50 [95% CI 1·09–2·08]; two-tailed p=0·014; table 2; post randomisation occurred in four (2%) of 221 patients 768 puorG mRS score 0 1 2 3 4 5 6 Proportion of patients (%) Articles Tenecteplase Standard medical Relative rate p interaction group treatment group (95% CI) Age (years) 0·98 ≤70 54/132 42/136 1·47 (0·98–2·20) >70 29/89 24/95 1·51 (0·87–2·61) Sex 0·49 Male 58/159 49/162 1·39 (0·94–2·04) Female 25/62 17/69 1·72 (0·92–3·24) NIHSS score* 0·74 <10 53/86 47/99 1·32 (0·89–1·97) ≥10 30/135 19/132 1·50 (0·84–2·67) pc-ASPECTS† 0·80 ≤7 29/91 20/88 1·60 (0·90–2·84) 8–10 54/130 46/143 1·42 (0·95–2·12) Intention for intravenous alteplase 0·74 Yes 28/83 28/95 1·36 (0·80–2·32) No 55/138 38/136 1·56 (1·03–2·36) Intention for endovascular thrombectomy 0·58 Yes 42/130 37/125 1·34 (0·86–2·10) No 41/91 29/106 1·68 (1·04–2·72) Timing of endovascular thrombectomy 0·88 Early (≤60 min after randomisation) 19/67 18/66 1·24 (0·64–2·41) Delayed (>60 min after randomisation) 16/45 15/44 1·50 (0·71–3·18) None 48/109 33/121 1·63 (1·04–2·54) Time from stroke onset to randomisation, h 0·08 <6 35/104 38/110 1·16 (0·73–1·85) 6–24 40/79 16/77 2·51 (1·40–4·50) Wake-up stroke 8/38 12/44 0·96 (0·37–2·47) Pre-stroke mRS score‡ 0·57 0–1 79/211 64/220 1·48 (1·06–2·06) 2–3 4/10 2/11 6·93 (0·38–126·23) Baseline glucose concentration, mmol/L 0·33 <7·8 42/99 36/111 1·35 (0·86–2·12) ≥7·8 34/97 21/92 2·02 (1·16–3·51) Occlusion site 0·77 Proximal 46/126 38/125 1·40 (0·91–2·16) Mid 27/64 24/74 1·61 (0·90–2·86) Distal 7/28 4/31 2·99 (0·78–11·44) Stroke mechanism by TOAST classification 0·91 Cardioembolic 10/29 8/40 2·03 (0·77–5·38) ESUS 0/1 1/1 ·· Atherothrombotic 73/191 57/190 1·46 (1·03–2·08) BATMAN score§ 0·49 <7 26/74 15/75 1·97 (1·03–3·73) ≥7 5/13 2/18 4·32 (0·74–25·21) PC-CS¶ 0·59 0–3 3/21 4/30 1·32 (0·27–6·31) 4–5 16/34 8/30 1·99 (0·82–4·80) ≥6 8/16 2/13 5·70 (0·96–33·83) Overall 83/221 66/231 1·50 (1·09–2·08) 0·5 1·0 2·0 4·0 6·0 8·0 10·0 Favours standard medical treatment Favours tenecteplase Figure 3: Subgroup analyses for the primary outcome of mRS score of 0–1 or return to baseline mRS score at 90 days BATMAN=Basilar Artery Tomography Angiography. ESUS=embolic stroke of undetermined source. mRS=modified Rankin scale. NIHSS=National Institutes of Health Stroke Scale. pc-ASPECTS=posterior circulation Acute Stroke Prognosis Early CT Score. PC-CS=Posterior Circulation Collateral Score. TOAST=Trial of Org 10172 in Acute Stroke Treatment. *Scores on the NIHSS range from 0 to 42, with higher scores indicating greater deficit. †The pc-ASPECTS was determined by use of baseline CT angiogram source images, non-contrast CT brain images, or diffusion weighted imaging, and ranges from 0 to 10, with higher scores indicating fewer early ischaemic changes. ‡Scores on the mRS range from 0 to 6, with higher scores indicating greater disability. §BATMAN scores were evaluated on baseline CT angiography, ranging from 0 to 10, with higher scores indicating better vascular perfusion and collateral circulation. ¶The PC-CS was calculated on baseline CT angiography, ranging from 0 to 10, with higher scores indicating better collateral blood flow. Articles in the tenecteplase group versus seven (3%) of 231 patients endovascular thrombectomy is not available or aordable, in the standard medical treatment group (adjusted including low-income countries. The 2024 European relative rate 0·58 [95% CI 0·17–1·99]; table 2). Among Stroke Organisation and European Society of Minimally the seven patients in the standard medical treatment Invasive Neurological Therapy guidelines for the acute group, all had endovascular thrombectomy; one had management of basilar artery occlusion explicitly alteplase before endovascular thrombectomy and two had recommended extending the intravenous thrombolysis complications during endovascular thrombectomy window in basilar artery occlusion up to 24 h, although (vessel perforation). Post-hoc haemorrhagic trans- this recommendation was based on expert consensus formation classifications are listed in the appendix (p 24). rather than randomised evidence.16 All-cause mortality at 90 days was similar between The safety profile of tenecteplase was favourable, with groups (65 [29%] patients in the tenecteplase group and no dierences observed in symptomatic intracranial 71 [31%] patients in the standard medical treatment haemorrhage or 90-day mortality compared with group; adjusted relative rate 0·87 [95% CI 0·62–1·22]), as standard medical treatment. Importantly, the risk of was the proportion of patients with an mRS score of 5–6 symptomatic intracranial haemorrhage in the at 90 days (82 [37%] vs 89 [39%]; 0·87 [0·65–1·18]; table 2). tenecteplase group was low (~2%), which is lower than Severe adverse events did not dier between the groups the risk reported in anterior circulation stroke.14 This (appendix p 25). finding is consistent with previous evidence suggesting that the risk of haemorrhagic transformation in posterior Discussion circulation stroke might be lower than in anterior The TRACE-5 trial showed that, in patients with circulation28 and further supports the safety of extending ischaemic stroke due to basilar artery occlusion, intravenous thrombolysis beyond 4·5 h in basilar artery treatment with intravenous tenecteplase within 24 h of occlusion. symptom onset led to a larger proportion of patients with A previous trial29 showed no benefit of tenecteplase in no disability at 90 days than that observed with standard the late time window for anterior circulation large-vessel medical treatment. Of note, the magnitude of treatment occlusion (when most patients underwent immediate eect in this trial is similar to that reported in patients endovascular thrombectomy). By comparison, the with anterior circulation large-vessel occlusion not TRACE-5 population was younger, had a longer interval receiving endovascular thrombectomy in the TRACE-3 between thrombolysis and arterial puncture (45 min trial.14 The magnitude of eect in secondary outcomes vs 15 min), and a lower rate of endovascular thrombectomy using mRS score 0–2 and mRS score 0–3 dichotomies because most sites in this trial did not have endovascular was smaller, but there was evidence of net benefit in the thrombectomy capability at all times. In contrast to the ordinal analysis of mRS (scores 5–6 were combined). The Extending the Time Window for Thrombolysis in Posterior proportion of patients with substantial reperfusion Circulation Stroke without Early CT Signs (EXPECTS) before endovascular therapy was significantly higher in trial,12 which focused on the use of intravenous alteplase in the tenecteplase group than in the standard medical an extended time window in a population of patients with treatment group. The incidence of symptomatic very mild posterior circulation stroke with a median intracranial haemorrhage within the first 36 h after NIHSS score of 3, our trial focused on expanding treatment and 90-day all-cause mortality were similar intravenous thrombolysis eligibility to patients with basilar between the two groups. artery occlusion, a devastating condition with a dismal TRACE-5 was designed to evaluate the ecacy and prognosis. However, as a thrombolytic trial without a lower safety of tenecteplase in an extended time window in limit on the NIHSS, a lower baseline NIHSS score was patients with basilar artery occlusion. Endovascular observed in this trial than in the BAOCHE and thrombectomy within 24 h of symptom onset is highly ATTENTION trials.6,7 eective for basilar artery occlusion,3 but access is limited Importantly, subgroup analyses demonstrated a in many countries. Consistent with the majority of consistent treatment eect, regardless of the underlying previous thrombolytic trials, we chose an mRS score cause of stroke or NIHSS score, with a benefit of of 0–1 as the primary outcome. A study-level meta- tenecteplase preserved in patients with basilar artery analysis of 11 randomised controlled trial results showed occlusion and NIHSS scores of less than 10, a subgroup superiority of tenecteplase over alteplase, with the in which the ecacy of thrombectomy remains strongest signal also observed at mRS score 0–1.13 Our uncertain.3 As suggested by the results of the Basilar trial supports wider use of thrombolytic therapies up to Artery International Cooperation Study (BASICS) trial,4 24 h after onset of basilar artery occlusion, analogous to intravenous thrombolysis beyond 4·5 h of symptom trials in anterior circulation stroke14 but without the need onset might be a preferred treatment option for patients for perfusion mismatch imaging. This could greatly with basilar artery occlusion with NIHSS scores of less expand access to reperfusion therapies for patients than 10.16 The trial was not powered to show a significant requiring long transfers to centres with endovascular benefit in the subgroup of patients who were intended thrombectomy capability or in settings in which for endovascular thrombectomy. However, there was no 770 Articles evidence of a dierential treatment eect. The ongoing comments for the manuscript. YoW and BCVC were principal POST-ETERNAL trial (NCT05105633) and the planned investigators of the study. SZ and YH contributed to the trial execution individual patient data pooled analysis with TRACE-5 and participant recruitment. JY calculated the sample size, formulated the statistical plan, and performed the statistical analysis. All other will shed more light on this question. The predefined authors served as local investigators or co-investigators, with subgroup analyses showed significantly higher rates of responsibilities encompassing participant recruitment, data collection, excellent functional outcome with tenecteplase in and the critical review, revision, and final approval of the manuscript before submission. The steering committee (YoW, BCVC, LHS, MF, and patients with basilar artery occlusion who did not MWP) was responsible for the overall design, protocol development, undergo endovascular thrombectomy and did not receive interpretation, and supervision of the trial. YoW, YX, and JY had access alteplase in the extended time window (>4·5 h). to and verified all the underlying data reported in the manuscript, and all TRACE-5 was a pragmatic trial and predominantly authors had final responsibility for the decision to submit for publication. used non-contrast CT brain images, CTA source images, or diusion weighted imaging to assess baseline Declaration of interests LHS serves as a scientific consultant regarding trial design and conduct ischaemic changes, with no requirement for advanced for Genentech and is a member of the steering committee of the imaging to select patients with basilar artery occlusion TIMELESS trial (NCT03785678); is a consultant on user interface design for treatment in the late time window. This approach and usability to LifeImage; and is a member of a data safety monitoring facilitates the implementation and translation of the trial board for Penumbra (MIND; NCT03342664). MF serves as a consultant for Simcere USA, Revalesio, and Lumosa, and is a member of a data results in settings in which advanced imaging might not safety monitoring board for Moleac. All other authors declare no be available. Patients enrolled based on MRI versus CT competing interests. had similar treatment eect in post-hoc analysis. Data sharing Diusion weighted imaging–FLAIR mismatch was not Data collected for the study, including de-identified individual used in selection of patients. The convenience of single- participant data and a data dictionary defining each field in the set, can bolus administration of tenecteplase in patients with be made available to others on reasonable request to the corresponding author and after signing appropriate data sharing agreements. Such basilar artery occlusion is a major practical advantage requests must be approved by all the respective ethics boards and that will further facilitate implementation of the trial appropriate data custodians. results in metropolitan and rural settings. Acknowledgments This trial has several limitations. First, the trial was open We thank the Noncommunicable Chronic Diseases-National Science label. However, outcomes were assessed by clinicians who and Technology Major Project (2024ZD0527600, 2024ZD0527605), were unaware of the treatment assignments. Second, we Beijing Municipal Science Fund for Distinguished Young Scholars (JQ24058), and the National Natural Science Foundation of China restricted enrolment to patients who did not have large (82171272). China Shijiazhuang Pharmaceutical Company Recomgen baseline infarcts in the posterior circulation (consistent Pharmaceutical (Guangzhou) supplied tenecteplase and an unrestricted with previous trials in patients with basilar artery grant for infrastructure funding. occlusion), so the generalisability of results to patients References with larger established infarction is uncertain.6,7 Further 1 Mattle HP, Arnold M, Lindsberg PJ, Schonewille WJ, Schroth G. Basilar artery occlusion. Lancet Neurol 2011; 10: 1002–14. studies are needed to address these questions. Third, the 2 Alemseged F, Nguyen TN, Alverne FM, Liu X, Schonewille WJ, trial was performed in Chinese patients, an ethnic group Nogueira RG. Endovascular therapy for basilar artery occlusion. characterised by a higher prevalence of intracranial Stroke 2023; 54: 1127–37. atherosclerosis than other ethnicities, which might limit 3 Nogueira RG, Jovin TG, Liu X, et al. Endovascular therapy for acute vertebrobasilar occlusion (VERITAS): a systematic review and the generalisability of the trial results. Fourth, the individual patient data meta-analysis. Lancet 2025; 405: 61–69. tenecteplase used in this trial was the same formulation 4 Langezaal LCM, van der Hoeven EJRJ, Mont’Alverne FJA, et al. used in TRACE-2,30 which demonstrated similar results Endovascular therapy for stroke due to basilar-artery occlusion. N Engl J Med 2021; 384: 1910–20. compared with alteplase as other formulations of 5 Liu X, Dai Q, Ye R, et al. Endovascular treatment versus standard tenecteplase available internationally. However, formal medical treatment for vertebrobasilar artery occlusion (BEST): comparability data between tenecteplase formulations an open-label, randomised controlled trial. Lancet Neurol 2020; 19: 115–22. have not been published. 6 Tao C, Nogueira RG, Zhu Y, et al. Trial of endovascular treatment of In conclusion, tenecteplase administered within 24 h acute basilar-artery occlusion. N Engl J Med 2022; 387: 1361–72. of symptom onset improved functional outcomes in 7 Jovin TG, Li C, Wu L, et al. Trial of thrombectomy 6 to 24 hours patients with basilar artery occlusion compared with after stroke due to basilar-artery occlusion. N Engl J Med 2022; 387: 1373–84. standard medical treatment, including alteplase and 8 Otite FO, Saini V, Sur NB, et al. Ten-year trend in age, sex, and thrombectomy in those eligible, without increasing the racial disparity in tPA (alteplase) and thrombectomy use following haemorrhagic risk. Our results support the use of stroke in the United States. Stroke 2021; 52: 2562–70. 9 Asif KS, Otite FO, Desai SM, et al. Mechanical Thrombectomy tenecteplase up to 24 h after stroke onset as an eective Global Access For Stroke (MT-GLASS): a Mission Thrombectomy therapeutic option in patients with basilar artery (MT-2020 Plus) study. Circulation 2023; 147: 1208–20. occlusion in both endovascular thrombectomy-capable 10 Powers WJ, Rabinstein AA, Ackerson T, et al. Guidelines for the early management of patients with acute ischemic stroke: 2019 and endovascular thrombectomy-limited settings. Update to the 2018 Guidelines for the Early Management of Acute Contributors Ischemic Stroke: A Guideline for Healthcare Professionals From YX and FA prepared the first draft of the report. YoW, BCVC, LHS, MF, the American Heart Association/American Stroke Association. Stroke 2019; 50: e344–418. and MWP conceptualised the study design and provided critical Articles 11 Alamowitch S, Turc G, Palaiodimou L, et al. European Stroke 21 Chinese Stroke Association, Chinese Interventional Neuroradiology Organisation (ESO) expedited recommendation on tenecteplase for Society, Intervention Group of Committee of Stroke Prevention and acute ischaemic stroke. Eur Stroke J 2023; 8: 8–54. Control of Chinese Preventive Medicine Association. Chinese 12 Yan S, Zhou Y, Lansberg MG, et al. Alteplase for posterior circulation guideline for endovascular treatment of acute ischemic stroke 2023. ischemic stroke at 4.5 to 24 hours. N Engl J Med 2025; 392: 1288–96. Chinese J Stroke 2023; 18: 684–711. 13 Palaiodimou L, Katsanos AH, Turc G, et al. Tenecteplase vs 22 Liebeskind DS, Bracard S, Guillemin F, et al. eTICI reperfusion: alteplase in acute ischemic stroke within 4.5 hours: a systematic defining success in endovascular stroke therapy. J Neurointerv Surg review and meta-analysis of randomized trials. Neurology 2024; 2019; 11: 433–38. 103: e209903. 23 Wahlgren N, Ahmed N, Dávalos A, et al. Thrombolysis with 14 Xiong Y, Campbell BCV, Schwamm LH, et al. Tenecteplase for alteplase for acute ischaemic stroke in the Safe Implementation of ischemic stroke at 4.5 to 24 hours without thrombectomy. Thrombolysis in Stroke-Monitoring Study (SITS-MOST): N Engl J Med 2024; 391: 203–12. an observational study. Lancet 2007; 369: 275–82. 15 Palaiodimou L, Papageorgiou NM, Safouris A, et al. Ecacy and 24 Zi W, Qiu Z, Wu D, et al. Assessment of endovascular treatment for safety of intravenous thrombolysis in the extended time window for acute basilar artery occlusion via a nationwide prospective registry. acute ischemic stroke: a systematic review and meta-analysis. JAMA Neurol 2020; 77: 561–73. J Clin Med 2025; 14: 5474. 25 Zou G. A modified Poisson regression approach to prospective 16 Strbian D, Tsivgoulis G, Ospel J, et al. European Stroke studies with binary data. Am J Epidemiol 2004; 159: 702–06. Organisation and European Society for Minimally Invasive 26 Alemseged F, Shah DG, Diomedi M, et al. The basilar artery on Neurological Therapy guideline on acute management of basilar computed tomography angiography prognostic score for basilar artery occlusion. Eur Stroke J 2024; 9: 835–84. artery occlusion. Stroke 2017; 48: 631–37. 17 Alemseged F, Ng FC, Williams C, et al. Tenecteplase vs alteplase 27 van der Hoeven EJ, McVerry F, Vos JA, et al. Collateral flow predicts before endovascular therapy in basilar artery occlusion. Neurology outcome after basilar artery occlusion: the posterior circulation 2021; 96: e1272–77. collateral score. Int J Stroke 2016; 11: 768–75. 18 Puetz V, Khomenko A, Hill MD, et al. Extent of hypoattenuation on 28 Keselman B, Gdovinová Z, Jatuzis D, et al. Safety and outcomes of CT angiography source images in basilar artery occlusion: intravenous thrombolysis in posterior versus anterior circulation prognostic value in the Basilar Artery International Cooperation stroke: results from the safe implementation of treatments in stroke Study. Stroke 2011; 42: 3454–59. registry and meta-analysis. Stroke 2020; 51: 876–82. 19 Alemseged F, Van der Hoeven E, Di Giuliano F, et al. Response to 29 Albers GW, Jumaa M, Purdon B, et al. Tenecteplase for stroke at late-window endovascular revascularization is associated with 4.5 to 24 hours with perfusion-imaging selection. N Engl J Med collateral status in basilar artery occlusion. Stroke 2019; 50: 1415–22. 2024; 390: 701–11. 20 Xiong Y, Alemseged F, Cao Z, et al. Tenecteplase versus standard 30 Wang Y, Li S, Pan Y, et al. Tenecteplase versus alteplase in acute care in patients with acute basilar artery occlusion: a multi-centre, ischaemic cerebrovascular events (TRACE-2): a phase 3, prospective, randomised, open-label, blinded endpoint, phase 3, multicentre, open-label, randomised controlled, non-inferiority controlled trial. Stroke Vasc Neurol 2025; published Oct 28. https:// trial. Lancet 2023; 401: 645–54. doi.org/10.1136/svn-2025-004432. 772 --- [PDF原文](https://sci-net.xyz/storage/7932541/c13a9030a56e9c3229600ae1e269db02335c529578ba703705c056af783bd5df/Tenecteplase-versus-standard-medical-treatment-for-basilar-artery-occlusion-within-24-h-TRACE.pdf) DOI: 10.1016/S0140-6736(25)02633-9