INDIVIDUAL PATIENT-DATA META-ANALYSIS: MECHANICAL THROMBECTOMY FOR LARGE ISCHEMIC STROKE (ATLAS COLLABORATION)

covered by Julieta Rosales

(Wednesday, 22 October 2025, Barcelona, Spain)

Background

• Six recent randomized controlled trials (RCTs) examined endovascular thrombectomy (EVT) versus best medical management (MM) in patients with large ischemic cores.
• Evidence supports EVT efficacy and safety, though study designs, imaging criteria, and outcomes varied.
• Individual trials lacked power for certain subgroup analyses.

Objectives

• Conduct a systematic review and individual patient data (IPD) meta-analysis to quantify the benefits and risks of EVT in large-core stroke and explore treatment effects across clinical and imaging subgroups.

Methods

• Included RCTs of acute anterior circulation stroke with large ischemic core: ASPECTS <6 or core volume ≥50 mL, within 24 hours.
• 6 trials included: RESCUE-Japan LIMIT, ANGEL-ASPECT, SELECT2, TENSION, TESLA, LASTE.
• Primary outcome: modified Rankin Scale (mRS) at 90 days.
• Secondary outcomes: functional independence (mRS 0–2), ambulation (mRS 0–3), mortality, sICH, and neurological worsening.
• Two-stage random-effects meta-analysis used.

Results (n = 1,886).

• EVT (n=944) vs MM (n=942).
• Primary outcome:
  • Shift analysis: aGenOR 1.63 (95% CI 1.42–1.87, p<0.001)
• Secondary outcomes:
  • mRS 0–2: 19.5% EVT vs 7.5% MM (aRR 2.61, p<0.001)
  • mRS 0–3: 36.6% EVT vs 19.8% MM (aRR 1.95, p<0.001)
  • Mortality: 31% EVT vs 37% MM (aRR 0.82, p=0.018)
  • sICH: 5.5% EVT vs 2.7% MM (aRR 2.02, p=0.017)

Subgroup Findings

• EVT benefit consistent across:
  • Age, sex, stroke severity, occlusion site, and time window (<6h and 6–24h).
  • ASPECTS strata (0–5) — benefit even for very low ASPECTS (0-2) only in early window.
  • Ischemic core volume <150 mL — maintained benefit; no benefit ≥150 mL.
  • Mismatch profiles — benefit present with or without mismatch.
• Late window (6–24h): EVT still effective, though absolute gains smaller.

Safety

• Higher rate of symptomatic intracranial hemorrhage (EVT 5.5% vs 2.7%), but overall mortality lower and net clinical benefit positive.

Conclusions

• EVT is superior to medical management in large-core ischemic strokes.
• Benefit is consistent across subgroups, including those with very low ASPECTS or large core volumes (up to 150 mL).
• Improved functional outcomes and lower mortality, despite a modest increase in hemorrhage risk.
• Confirms EVT as the standard of care for a broad spectrum of large-core stroke patients.

TENECTEPLASE CONFIRMED NON-INFERIOR TO ALTEPLASE IN POOLED META-ANALYSIS: CATALYST-TNK INTERIM RESULTS

covered by WSO FSL Nishita Singh

(Wednesday, 22 October 2025, Barcelona, Spain) A large, pooled individual-patient data (IPD) meta-analysis presented today at the World Stroke Congress 2025 demonstrated that the thrombolytic agent tenecteplase (TNK) is non-inferior to the current standard treatment alteplase (tPA) for acute ischemic stroke when administered within 4.5 hours of onset. The findings, from the CATALYST-TNK project led by Professor Keith Muir (University of Glasgow, UK), add further weight to growing evidence supporting tenecteplase as a practical and effective alternative for intravenous thrombolysis worldwide.

Tenecteplase, a genetically engineered variant of alteplase with greater fibrin specificity and single-bolus administration, has been evaluated in multiple international randomized controlled trials since 2010. CATALYST-TNK was established to harmonize and analyze individual patient data from all trials directly comparing tenecteplase and alteplase to refine estimates of efficacy and safety, and to formally test both non-inferiority and superiority hypotheses.

This interim analysis incorporated 12 randomized trials, ranging from early phase studies such as Haley (2010) and Parsons (2012) to recent large trials including AcT, TRACE-2, ATTEST-2, TASTE, and ORIGINAL, all using the now-standard dose of 0.25 mg/kg tenecteplase within 4.5 hours of last-known-well. Individual patient data were available for all trials except for ACT and the two TRACE trials. Hence, only interim results were presented.

The pooled dataset included 4,582 patients (2,295 tenecteplase; 2,287 alteplase) with a mean age of 68 years, median NIHSS 7 (IQR 5–12), and median onset-to-treatment time of 155 minutes.

The primary endpoint, functional independence defined as modified Rankin Scale (mRS) 0–1 at 90 days, was achieved in 56.0% of patients treated with tenecteplase and 53.7% with alteplase, yielding an absolute difference of +2.3 percentage points (95% CI –0.6 to +5.3). Both pre-specified non-inferiority margins (–3% and –1.3%) were met (p < 0.001 for non-inferiority), confirming tenecteplase’s non-inferiority to alteplase. Superiority analysis was not significant (risk ratio 1.04 [0.99–1.10]; p = 0.12), although the numerical trend favored tenecteplase. Further, secondary analysis that included IPD and pooled data of ACT and TRACE trials, were significant for both pre-specified non-inferiority margins (–3% and –1.3%) and for superiority, favoring tenecteplase.

For secondary outcomes, no significant differences were observed in mRS 0–2 (RR 1.02 [0.97–1.08]; p = 0.214) or overall mRS distribution (cOR 1.105 [0.971–1.258]; p = 0.112), though directional consistency again favored tenecteplase. Mortality was almost identical (8.8% vs 8.7%; RR 0.94 [0.81–1.10]), and symptomatic intracerebral hemorrhage (SICH) occurred in 2.2% of tenecteplase-treated patients versus 2.6% of alteplase-treated patients (RR 1.11 [0.85–1.47]). Fatal ICH within 7 days was 1.2% in both groups.

“These results reaffirm the safety and efficacy of tenecteplase at the 0.25 mg/kg dose and demonstrate its comparability to alteplase across a broad spectrum of acute stroke patients presenting within 4.5 hours,” said Professor Muir.

Professor Muir noted that this analysis remains interim and unadjusted, as individual-patient datasets from several recent trials (AcT, TRACE, TRACE-2) are still being integrated. The final, fully adjusted IPD meta-analysis, accounting for baseline severity, age, sex, and onset-to-treatment time, is expected to provide definitive conclusions.

In summary, CATALYST-TNK confirms tenecteplase’s non-inferiority to alteplase for intravenous thrombolysis in acute ischemic stroke. These findings strengthen the case for broader adoption of tenecteplase in global stroke guidelines.

Funding and acknowledgements:

CATALYST-TNK is supported by a grant from Boehringer Ingelheim to the University of Glasgow.

Reference: Muir K. et al. Pooled Individual-Patient Data Meta-analysis of Trials of Tenecteplase Compared to Alteplase in Acute Ischemic Stroke (CATALYST-TNK): Interim Analyses. Presented at the World Stroke Congress; 22 October 2025; Barcelona, Spain.

RESULTS FROM OPTIMAL TRIAL SHOW THAT MULTI-COMPONENT INTERVENTION HELPS REDUCE BOTH DOOR-TO-PUNCTURE TIME AND BETTER 3-MONTH OUTCOMES

covered by Ivan Koltsov

(Wednesday, 22 October 2025, Barcelona, Spain) The OPTIMAL cluster-randomized controlled trial, presented by Yaode He, showed the applicability and efficacy of a multi-component intervention in a hyperacute setting. It was based on an assumption that door-to-puncture time and outcomes in patients undergoing EVT may be optimized using behavior change theory. A multi-component behavior-change package (PEITER) consisted of persuasion, enablement, incentivization, training, and education.

The OPTIMAL trial randomized 16 hospitals (1:1) to deploy PEITER or maintain routine care to test whether team-focused measures could shorten EVT delivery time and enhance 90-day outcomes. PEITER hospitals received a 6-component intervention. The intervention was based on behavior change theory and included monthly video conferences and annual onsite surveillance.

1288 patients were included (666 in the PEITER group, 622 in the control group). Patients treated in PEITER hospitals were more likely to achieve a more favorable outcome, defined as an mRS score of 0-3 (66.4% vs. 55.4%, adjusted odds ratio 1.646 [1.054-2.57], p = 0.028). Additionally, median door-to-puncture time was 103 minutes in the behavioral package group and 111 minutes in the control group (adjusted mean difference -22.4 [-42.0-2.8], p = 0.025).

Thus, the behavior-change bundle in the OPTIMAL trial helped reduce in-hospital delays and improve 90-day mRS outcomes in patients undergoing EVT. Optimizing team behaviors and pathway reliability can make a measurable difference for patients.

TENECTEPLASE SHOWS HIGHER EARLY REPERFUSION RATES THAN LOW-DOSE ALTEPLASE IN JAPANESE STROKE PATIENTS: T-FLAVOR TRIAL RESULTS

covered by WSO FSL Nishita Singh

(Wednesday, 22 October 2025, Barcelona, Spain) The T-FLAVOR trial, presented today at the World Stroke Congress 2025 by Professors Inoue, Hirano and Toyoda, demonstrated that tenecteplase (0.25 mg/kg), when given before mechanical thrombectomy in patients with large vessel occlusion (LVO)-type ischemic stroke, was associated with a higher rate of early substantial reperfusion compared with low-dose alteplase (0.6 mg/kg), the standard regimen currently approved in Japan.

Tenecteplase has gained international acceptance for acute ischemic stroke following a series of positive global trials. However, the drug is not yet approved for ischemic stroke use in Japan. The T-FLAVOR trial was therefore initiated as a key investigator-led study to support regulatory approval for tenecteplase in the Japanese population.

The study was a phase II, multicenter, prospective, randomized, open-label, masked-endpoint (PROBE) trial conducted at 18 sites across Japan, comparing intravenous tenecteplase (0.25 mg/kg) with alteplase (0.6 mg/kg) in patients with LVO-type stroke eligible for both intravenous thrombolysis (within 4.5 hours) and mechanical thrombectomy (within 6 hours).

Between December 2021 and March 2025, a total of 218 patients were randomised (107 tenecteplase; 111 alteplase). The median age was 77 years, and 79% had a pre-stroke mRS of 0–1. Median baseline NIHSS was 18, with internal carotid or proximal middle cerebral artery occlusion in most cases.

The primary efficacy outcome—substantial reperfusion defined as mTICI score ≥2b or no retrievable thrombus on the initial angiogram was achieved in 10.3% of patients receiving tenecteplase versus 3.6% with alteplase. The absolute difference of 6.5% (95% CI 0.9–12.1; p=0.06) in the full analysis set and 7.4% (95% CI 0.8–14.0; p=0.03) in the per-protocol set indicated a statistically significant higher rate of early reperfusion with tenecteplase.

Subgroup analyses revealed numerically higher reperfusion rates for tenecteplase particularly in M2 MCA (17.2% vs. 7.1%) and basilar artery occlusions (16.7% vs. 0%). The 90-day functional outcome did not differ significantly between groups (common OR 1.47, 95% CI 0.92–2.35; p=0.11), nor did the rate of early neurological improvement at 72 hours (67.3% vs. 55.9%; p=0.08). Safety outcomes were comparable. Major bleeding occurred in 1.9% of tenecteplase-treated patients versus 0% in the alteplase group (p=0.15), and 90-day mortality was 6.5% versus 9.9% (p=0.36).

“These findings suggest that tenecteplase achieves earlier substantial reperfusion than low-dose alteplase and could represent a more effective thrombolytic agent for patients with large vessel occlusion in Japan,” said Professor Toyoda. “Importantly, this study lays the groundwork for national regulatory approval and aligns Japan with international advances in acute stroke therapy.”

The authors acknowledged limitations including a relatively small sample size, COVID-19-related delays in patient recruitment, and a shorter median needle-to-angiography time that may have limited the ability to detect larger between-group differences.

In conclusion, T-FLAVOR provides the first randomized evidence from Japan that tenecteplase (0.25 mg/kg) prior to thrombectomy is safe and enhances early reperfusion compared with low-dose alteplase (0.6 mg/kg), with similar rates of bleeding and mortality.

Funding and acknowledgements:

The trial was funded by the Japan Agency for Medical Research and Development (AMED; JP24lk0221186), the Intramural Research Fund for Cardiovascular Diseases of NCVC (grants 21-4-2 and 24-B-6), and the NCVC Crowdfunding Project.

Reference: Inoue M. et al. Tenecteplase versus alteplase for large vessel occlusion recanalization (T-FLAVOR): A randomized, multicenter phase II trial. Presented at the World Stroke Congress; 22 October 2025; Barcelona, Spain.

ANGEL-COAST TRIAL RESULTS SHOW THAT CONTACT ASPIRATION IN VERTEBROBASILAR ARTERY OCCLUSION IS ASSOCIATED WITH HIGHER RATES OF FIRST-PASS EFFECT, BUT LOWER RATES OF FINAL RECANALIZATION

covered by Ivan Koltsov

(Wednesday, 22 October 2025, Barcelona, Spain)

A multicenter, prospective, randomized, open-label, blinded-endpoint (PROBE) trial, ANGEL-COAST, compared contact aspiration with stent retriever thrombectomy as first-line strategies for acute basilar artery occlusion (BAO). Investigators hypothesized that first-line contact aspiration may be more efficient compared to stent retriever due to the vertebrobasilar anatomy.

ANGEL-COAST enrolled 338 patients across 22 actively enrolling sites across China: 170 were randomized to first-line contact aspiration and 168 to first-line stent retriever. The study population was well-balanced across groups.

Key inclusion criteria were a <24 h time window, age of 18 years or older, basilar artery occlusion, baseline NIHSS score of 10 or higher, baseline PC-ASPECTS of 6 or higher, and pre-stroke mRS score of 2 or less. The trial was designed to test whether the device choice could translate into better technical quality (first-pass effect, FPE), defined as eTICI 2c/3 after the first thrombectomy pass without rescue treatment.

Using an intention-to-treat analysis, investigators found that the first-pass effect was achieved in 43.5% of patients treated with contact aspiration versus 27.4% with stent retriever, corresponding to a treatment effect of 2.04 (95% CI 1.30-3.22; p=0.002). A bit more significant results were demonstrated in the per-protocol analysis (49.7% vs. 27.2%, treatment effect 2.65, 95% CI 1.65-4.24, p<0.001).

However, a series of secondary outcome analyses showed lower final recanalization quality in the contact aspiration group (87.6% of eTICI≥2b50 patients in the first-line contact aspiration group vs. 94.0% in the first-line stent retriever group, treatment effect 0.45, 95% CI 0.21-0.99, p=0.046).

While contact aspiration achieved a higher first-pass effect (eTICI 2c/3) than stent retrievers, this early advantage was offset by lower final reperfusion rates.

TIROFIBAN TO PREVENT RECURRENT STROKE IN HIGH-RISK TIA/MINOR STROKE WITH INTRACRANIAL ATHEROSCLEROSIS

covered by Gianluca De Rubeis

(Wednesday, 22 October 2025, Barcelona, Spain) This randomized trial evaluated tirofiban, an intravenous antiplatelet agent, as an acute strategy to prevent recurrent stroke in patients with high-risk TIA or minor ischemic stroke and intracranial atherosclerosis. In this large study of 4,724 patients, tirofiban did not reduce recurrent stroke compared to placebo and was associated with more any-bleeding events, while severe bleeding remained rare.

CHANCE-4 was a multicenter, randomized, double-blind, placebo-controlled trial. Adults with high-risk TIA/minor ischemic stroke and intracranial atherosclerosis were randomly assigned to receive intravenous tirofiban or placebo, in addition to standard medical therapy. This trial was conducted in China, and the primary efficacy outcome was recurrent stroke. Safety outcomes included severe bleeding (including symptomatic intracranial hemorrhage per protocol) and bleeding during follow-up.

A total of 4,724 patients were included in the study. The primary analysis was neutral: Odds Ratio 1.00 (95% CI 0.83–1.22) for recurrent stroke with tirofiban versus placebo. Severe bleeding occurred at 0.4% with tirofiban and 0.3% with placebo (similar rates), while bleeding was more frequent with tirofiban (3.8% vs. 1.5%).

Conclusion. In patients with high-risk TIA/minor stroke and intracranial atherosclerosis, early IV tirofiban did not reduce recurrent stroke and increased any-bleeding events, with rare severe bleeding, similar to placebo. These findings do not support the routine use of early tirofiban for recurrence prevention in this population.

TENECTEPLASE FOR BASILAR ARTERY OCCLUSION (± MECHANICAL THROMBECTOMY) (TRACE-V)

covered by Gianluca De Rubeis

(Wednesday, 22 October 2025, Barcelona, Spain) This phase 3, multicenter PROBE trial (prospective, randomized, open-label with blinded endpoint) compared tenecteplase (TNK) with alteplase (tPA), each with or without mechanical thrombectomy (MT), for patients with basilar artery occlusion (BAO) treated within 24 hours of onset and imaging showing near or total occlusion. The tenecteplase group achieved a higher rate of excellent functional outcomes and better early angiographic reperfusion, with similar safety to alteplase.

Intravenous thrombolysis is a key therapy for posterior circulation stroke, and single-bolus delivery of tenecteplase offers workflow advantages in systems where MT is variably available. This trial tested whether TNK (± MT) was superior to tPA (± MT) for the treatment of BAO.

Across 452 randomized patients, participants received either tenecteplase or alteplase, and MT was performed according to the local practice/availability. The primary efficacy outcome was the proportion with a modified Rankin Scale (mRS) score of 0–1 at 90 days. The key outcome was reperfusion on the initial angiography. The safety outcomes included symptomatic intracranial hemorrhage, major bleeding, and mortality.

The tenecteplase arm achieved an mRS of 0–1 in 37.6% of patients versus 28.6% with alteplase (p = 0.014). Early reperfusion on initial angiography was 15.4% with tenecteplase and 7.3% with alteplase. Safety outcomes were similar between the groups, with no signs of increased symptomatic intracranial hemorrhage, major bleeding, or death. Approximately 50% of the patients underwent MT, a limitation that may affect the interpretation of MT-specific effects.

Conclusion. In BAO treated within 24 hours, tenecteplase (± MT) was associated with greater odds of excellent recovery and higher early reperfusion than alteplase (± MT), without a safety penalty.

FREE COMMUNICATION 1 – REHABILITATION 1

covered by Shamala Thilarajah and Kwah Li Khim

(Wednesday, 22 October 2025, Barcelona, Spain) Wonderful session with 7 presentations from Canada and the UK, moderated by Natan Bernstein from Israel and Marit Jensen from Hamburg.

1. Kevin Moncion (UK) presented results from a secondary analysis of a trial testing the effects of high intensity (HI) aerobic exercise on sleep quality in patients with subacute stroke. While the HI aerobic exercise improved subjective sleep quality, it did not improve sleep efficiency.
2. Suzanne Ackerley (UK) presented results on the evaluation of an online group-based telerehabilitation program aimed at providing more therapy (in terms of duration and intensity) to patients. Program was well received by both patients and staff, despite challenges with infrastructure and staffing constraints.
3. Alexander Thiel (Canada) presented the feasibility outcomes of a trial testing the effects of rTMS on upper limb outcomes. Implementation of the protocol at multiple sites was found to be feasible and safe. Improvement in FM-UE was seen. While three serious adverse events were reported, they were not related to the rTMS.
4. Lara Boyd (Canada) shared findings from a trial testing the effects of high-intensity interval training (HIIT) on cognition post-stroke. While HIIT did not enhance skilled motor learning, it improved cognitive-motor processing speed and response inhibition in stroke patients. The study suggests HIIT as a feasible and beneficial intervention for cognitive recovery post-stroke.
5. Linda Tremain (UK) explored enablers and barriers to exercise referral schemes post-stroke using the COM-B framework. Stroke survivors and communication partners were interviewed. Study findings identified the national exercise referral scheme as a learning environment for recovery and discovery, and emphasized the importance of community support and personal recovery in encouraging exercise participation.
6. Hatem Lazem (UK) discussed the psychometric properties of a markerless motion capture system (Media-Pipe) for assessing upper limb range of motion post-stroke. Compared with a 2D motion analysis software (Kinovea), the system showed excellent validity and reliability, offering a low-cost and quick alternative for motion analysis in both clinical and home settings.
7. Serafeim Perdikis (UK) presented findings from a trial testing the effects of brain-computer interface based functional electrical stimulation (BCI-FES) on upper limb outcomes in chronic stroke. The trial demonstrated a significant improvement in motor function (FM-UE), particularly in patients with severe baseline deficits.