Intra-Arterial Tenecteplase After Successful Reperfusion in Large Vessel Occlusion Stroke

Clinical Question (PICO)

In adults aged ≥ 18 with acute anterior-circulation large vessel occlusion (LVO) ischemic stroke (NIHSS 6–24; ASPECTS ≥ 6; eTICI 2b–3 after successful endovascular thrombectomy, EVT) presenting within 24 hours of last known well — does adjunctive intra-arterial (IA) tenecteplase (0.0313 or 0.0625 mg/kg), infused through a catheter proximal to the original occlusion site within 10 minutes of randomization, improve the proportion of patients achieving no disability (modified Rankin Scale [mRS] 0–1) at 90 days compared with placebo/sham, with an acceptable symptomatic intracranial hemorrhage (sICH) rate?

Bottom Line

Adjunctive IA tenecteplase at 0.0313 mg/kg or 0.0625 mg/kg after successful EVT was safe enough to advance to a phase 3 trial; the higher 0.1250 mg/kg dose crossed the prespecified sICH safety threshold. In the randomized phase 2a comparison (n = 157), the proportion achieving mRS 0–1 at 90 days was numerically higher with 0.0625 mg/kg (43.5% [20/46]) than control (33.8% [22/65]) — adjusted RR 1.15 (95% CI, 0.73–1.80; P = .55) — but the trial was not powered to demonstrate efficacy and the difference was not statistically significant.

Design

Trial type: Prospective, multicenter, open-label, blinded-outcome assessment, phase 1b (nonrandomized dose-escalation with a 14 + 8 design) and phase 2a (exploratory randomized dose-expansion).
N: 205 analyzed (phase 1b: 48; phase 2a: 157)
- Phase 1b dose tiers:
  - 0.0313 mg/kg: 14
  - 0.0625 mg/kg: 22 (initial 14 + 8 expansion)
  - 0.1250 mg/kg: 12
- Phase 2a randomized:
  - Tenecteplase 0.0625 mg/kg: 46
  - Tenecteplase 0.0313 mg/kg: 46
  - Control (no IA thrombolysis): 65
Randomization: Phase 2a used stratified 1:1:√2 allocation (probabilities 0.293 / 0.293 / 0.414), stratified by age (<70 vs ≥70) and admission NIHSS (<15 vs ≥15). Phase 1b was nonrandomized.
Setting: 30 sites in China (phase 2a at 23 sites).
Enrollment: July 2023 – August 2024; 90-day follow-up through late 2024.
Mean follow-up: 90 (±14) days.
Analysis: Modified Poisson regression with prespecified covariates (age, NIHSS, occlusion site) for adjusted risk ratios; generalized odds ratios (GenOR) and win ratio (WR) for ordinal/continuous secondary outcomes; inverse probability of treatment weighting for adjusted GenOR/WR. No formal adjustment for multiplicity on secondary/safety outcomes (exploratory).
Primary outcome (phase 1b): Proportion with symptomatic intracranial hemorrhage (sICH) within 24 hours by ECASS III criteria, central imaging core lab blinded assessment.
Primary outcome (phase 2a): No-disability outcome (mRS 0–1) at 90 (±14) days, assessed by two blinded independent central raters (video/audio recorded when possible).

Population

Inclusion Criteria (phase 2a; full criteria in Supplement 1 / eMethods 2)

Age ≥ 18 years
Acute ischemic stroke in the anterior circulation presenting within 24 hours of last known well
Baseline NIHSS 6–24
Baseline ASPECTS ≥ 6 on noncontrast CT
Occlusion of intracranial ICA, M1, or M2
eTICI 2b–3 after EVT

Exclusion Criteria

Prior intravenous thrombolysis before EVT
Any other contraindication to intra-arterial thrombolysis (full list in Supplement 1)

Baseline Characteristics (phase 2a overall)

Median age: 71 (IQR 60–77) years
92 female (44.9%); 113 male (55.1%) overall (phase 2a: 68 female / 43.3%, 89 male / 56.7%)
All 205 patients self-reported Asian race (phase 2a: 144 Han, 7 Bouyei, 1 Yi, 5 Zhuang)
Median baseline NIHSS: similar across arms (see Table 1)
Median baseline ASPECTS: 8 (6–9) tenecteplase 0.0625; 9 (8–9) tenecteplase 0.0313; 8 (7–9) control
Median time from last known well to randomization: 410 min (0.0625); 404 min (0.0313); 384 min (control)
Median time to study drug: 417 min (0.0625); 414 min (0.0313)
A higher proportion of intracranial ICA occlusions was present in the control group (imbalance; mitigated by adjusted analyses)

Interventions

Tenecteplase 0.0625 mg/kg (1/4 the standard IV dose): intra-arterial bolus/infusion through a distal access catheter or microcatheter positioned proximal to the initially occluded artery (ICA terminus for terminal ICA; M1 or M2 segment for those occlusions), within 10 minutes of randomization, under real-time angiographic guidance.
Tenecteplase 0.0313 mg/kg (1/8 the standard IV dose): identical procedure at half the dose.
Control: procedure terminated without further intra-arterial thrombolysis after successful EVT.

All patients received standardized medical and secondary preventive treatment per the Chinese Guidelines for Endovascular Treatment of Acute Ischemic Stroke 2018.

Outcomes

Phase 1b primary safety outcome — sICH within 24 hours (ECASS III):

0.0313 mg/kg: 1/14 (7.1%)
0.0625 mg/kg: 2/22 (9.1%) — 2 of the initial 14 patients had dose-limiting toxicity; the subsequent 8 patients had no sICH
0.1250 mg/kg: 3/12 (25.0%) — exceeded prespecified safety threshold (P = .04 vs expected); dose deemed not tolerated

Phase 2a primary efficacy outcome — mRS 0–1 at 90 days (prespecified):

Group	n/N (%)	Adjusted RR (95% CI) vs control	P
Tenecteplase 0.0625 mg/kg	20/46 (43.5%)	1.15 (0.73–1.80)	.55
Tenecteplase 0.0313 mg/kg	17/46 (37.0%)	0.85 (0.54–1.35)	.50
Control	22/65 (33.8%)	—	—

Unadjusted RRs: 0.0625 mg/kg 1.28 (0.80–2.06; P = .30); 0.0313 mg/kg 1.09 (0.66–1.81; P = .73). No prespecified subgroups had a statistically significant interaction.

Secondary Outcomes (phase 2a, exploratory):

mRS 0–2 at 90 days: 56.5% (0.0625), 50.0% (0.0313), 50.8% (control); adjusted RR 1.12 (0.80–1.56; P = .52) and 0.91 (0.64–1.32; P = .63)
Ordinal mRS at 90 days, median (IQR): 2 (1–5), 3 (1–4), 2 (1–5); adjusted GenOR 1.18 (0.70–1.99) and 0.95 (0.55–1.62)
Improved angiographic reperfusion (eTICI shift post-thenication): 3/46 (6.5%) at 0.0625; 4/46 (8.7%) at 0.0313; NA control (no post-randomization angiogram in control arm)
Excluding baseline eTICI 3: 3/26 (11.5%) at 0.0625; 4/30 (13.3%) at 0.0313
Change in NIHSS at 5–7 days or discharge from baseline, median (IQR): −7 (−12 to 0), −6 (−12 to −2), −6 (−13 to 0)

Safety / Adverse Events (phase 2a):

sICH within 24 hours: 6.5% (0.0625), 4.3% (0.0313), 3.1% (control)
Mortality at 90 days: 19.6% (0.0625; adjusted RR 0.78, 95% CI 0.36–1.66; P = .52), 15.2% (0.0313; adjusted RR 0.77, 95% CI 0.34–1.79; P = .50), 21.5% (control)
No significant between-group difference in any safety outcome

Figures

DATE trial patient flowchart — *Figure 1.* Flowchart of Patients Through the Successful Thrombectomy in Patients With Large Vessel Occlusion (DATE) Trial. ASPECTS indicates Alberta Stroke Program Early CT Score; CT, computed tomography; eTICI, Extended Treatment in Cerebral Infarction; ICA, internal carotid artery; IV, intravenous; MCA, middle cerebral artery; NIHSS, National Institutes of Health Stroke Scale. (a) Eligible patients were randomized with stratification allocation according to 2 strata: patient age (<70 vs ≥70 years), and admission NIHSS score (<15 vs ≥15).

DATE trial ICH and 90-day mRS distribution — *Figure 2.* Intracranial Hemorrhage (ICH) Within 24 Hours in Phase 1b and Distribution of Score on the Modified Rankin Scale at 90 Days in Phase 2a. (A) Proportion of ICH within 24 hours by European Cooperative Acute Stroke Study III (ECASS III) classification; aICH denotes any intracranial hemorrhage; sICH, symptomatic intracranial hemorrhage. (B) Distribution of mRS scores at 90 days among the three phase 2a arms.

Criticisms

Underpowered for efficacy. Phase 2a randomized 157 patients across three arms; the trial was explicitly designed to evaluate safety and explore efficacy signal — not to demonstrate efficacy. CIs on adjusted RRs span both substantial benefit and substantial harm (e.g., 0.0625 mg/kg RR 1.15, 95% CI 0.73–1.80).
Imbalance in occlusion site. A significantly higher proportion of intracranial ICA occlusions in the control group was mitigated by adjusted analyses, but residual unmeasured confounding remains.
Phase 1b dose-escalation used a nonrandomized design, so the dose-finding inferences are vulnerable to temporal/operator confounding even though the sICH threshold crossing at 0.1250 mg/kg was statistically prespecified.
No perfusion imaging required. Pragmatic design reduces explanatory power: tissue-level penumbra/core status was not characterized.
Generalizability: Asian-only population, anterior circulation, small ischemic core (ASPECTS ≥ 6), no prior IV thrombolysis. Findings may not extend to non-Asian populations, posterior circulation strokes, large-core LVO, or bridging IV thrombolysis recipients.
Open-label treatment — operators knew assignment. Outcome assessors were blinded.
Active comparator bias in angiographic endpoint: “improved angiographic reperfusion” only meaningful in the IA arms (no control post-randomization angiogram).

Funding

Funded by Recomgen Pharmaceutical (Guangzhou) and the China Shijiazhuang Pharmaceutical Company. The funders had no role in design, conduct, data collection/management/analysis/interpretation, manuscript preparation/review/approval, or the decision to submit for publication. Trial registration: ChiCTR2300073787 (phase 1b) and ChiCTR2400080624 (phase 2a).

The paper

Authors. Hou X, Huang J, Wang L, He Y, Song J, Guo C, Yang S, Shi X, Chen L, Liu Q, Su J, Zeng L, Jiang M, Chen B, Cheng X, Chen S, Pan H, Shen X, Wu Y, Tang X, Wang J, Han S, Pu T, Wu C, Li F, Qu L, Fu Z, Liu H, Li Y, Mei B, Cheng Y, Hu Z, Zhang H, Lv T, Wu M, Xu R, Ye Q, Kong L, Mi S, Wu J, Wang Y, Tian Z, Sun W, Ma J, Xu X, Wu Y, Wang D, Nogueira RG, Nguyen TN, Saver JL, Zi W, Zhou Z.
Title. Intra-Arterial Tenecteplase After Successful Reperfusion in Large Vessel Occlusion Stroke: A Randomized Clinical Trial.
Journal. JAMA Neurology.
Year. 2025.
DOI. 10.1001/jamaneurol.2025.2036
PMCID. PMC12228979

Deep Dive — click to expand

What this is

DATE is the first dose-finding randomized evaluation of adjunctive intra-arterial tenecteplase after successful mechanical thrombectomy for anterior circulation LVO. It is positioned as the natural successor to CHOICE (intra-arterial alteplase, JAMA 2022), substituting tenecteplase — the IV thrombolytic that has displaced alteplase in many stroke centers — and adding an explicit phase 1b dose-escalation that cleanly identifies 0.1250 mg/kg as too toxic while leaving 0.0625 mg/kg (one-quarter of the standard IV dose) and 0.0313 mg/kg (one-eighth) as candidates for further study. The headline signal is a 9.7-percentage-point absolute increase in 90-day no-disability (43.5% vs 33.8%) at the 0.0625 mg/kg dose — directionally consistent with CHOICE — but in a phase 2a sample (n = 157 randomized) explicitly not powered for efficacy. The trial’s primary contribution is therefore the dose-safety map, not the efficacy estimate.

1. Shadow Audit

The authors frame the sICH crossing at 0.1250 mg/kg as a clean safety failure that justifies advancing only 0.0313 and 0.0625 mg/kg. What’s understated is the gradient across the two retained doses. The 0.0625 mg/kg arm showed sICH of 6.5% versus 3.1% control — a doubling in absolute terms, in a 46-patient arm with a wide CI. The 0.0313 mg/kg arm’s 4.3% sICH looks more reassuring, but its adjusted RR for the primary mRS 0–1 endpoint was 0.85 (95% CI 0.54–1.35) — numerically worse than control, despite a 3.2-percentage-point absolute gain in mRS 0–1 (37.0% vs 33.8%). The favorable direction was entirely driven by the higher dose, but the higher dose also doubled sICH. There is no dose within the tested range that clearly decouples efficacy from bleeding — the shadow audit says the dose-response curve may be flat in efficacy while steeper in bleeding than the authors acknowledge.

A second omission: median time from last known well to study drug was 414–417 minutes — over six hours — far beyond the IV thrombolysis window. The trial therefore tested late-window adjunctive IA tenecteplase, not the early-window rescue that clinicians often invoke when discussing CHOICE-style strategies. The framing in the introduction (rescue residual distal occlusions) does not match the enrollment timeline.

2. Inversion Engine

For the 0.0625 mg/kg dose to be declared futile at phase 3, the true adjusted RR for mRS 0–1 would need to land below ~1.05 with the lower 95% CI excluding 1.15. Given the observed point estimate of 1.15 (95% CI 0.73–1.80), the trial would need a substantially larger sample and a tighter CI to either confirm or refute this. For the alternative conclusion — that IA tenecteplase does not improve functional outcome at any of the tested doses — to hold, one would need to see (a) the 0.0625 mg/kg arm’s true effect drift toward null as the phase 3 enrolls broader populations, and (b) no compensating signal in the eTICI-shift secondary outcome. The eTICI improvement rate of 6.5% (0.0625) and 8.7% (0.0313) is low — suggesting the drug is not meaningfully improving macrovascular reperfusion post-EVT, which is the mechanism on which its efficacy rests. If a phase 3 with n ≈ 800–1000 reproduces an adjusted RR near 1.05–1.10, the entire post-EVT adjunctive thrombolysis paradigm collapses.

3. Second-Order Catalyst

If a phase 3 trial confirms a clinically meaningful absolute benefit (≥ 7 percentage points in mRS 0–1), the first movers are high-volume comprehensive stroke centers in China already running 24/7 thrombectomy programs — the 23 phase 2a sites are operationally ready. Adoption would propagate through the Chinese Stroke Association guidelines within ~12 months, then to AHA/ESO via parallel trial replication (likely an Australasia/Europe-led confirmatory trial given the populations tested). The protocol-level change: every successful-reperfusion thrombectomy case receives an intra-arterial bolus of tenecteplase 0.0625 mg/kg infused through the existing microcatheter before sheath removal, extending the thrombectomy procedure by 5–10 minutes. Cost driver: tenecteplase is ~$1,000–2,000 per vial; the marginal cost per patient is trivial relative to the thrombectomy episode.

4. Asymmetric Leverage

The asymmetry is unfavorable. Tenecteplase at 0.0625 mg/kg delivered a 9.7-percentage-point absolute improvement in mRS 0–1 over control — large in trial terms — but on a denominator of patients already successfully reperfused (the most favorable substrate). If you scale that to all-comer LVO thrombectomy recipients (where ~75% achieve successful reperfusion and ~27% are disability-free at 90 days per the HERMES meta-analysis), the absolute increment shrinks: even a 10% relative improvement in mRS 0–1 would shift the disability-free fraction from ~27% to ~30% population-wide. The denominator is large (millions of LVO strokes globally per year), but the per-patient leverage is small. The bleeding-cost is asymmetric in the other direction: sICH doubling from 3% to 6.5% is a real mortality/morbidity signal in absolute terms, even if not statistically significant in a 46-patient arm.

5. Paradigm Destroyer

The reflex DATE challenges is the implicit assumption that adjunctive IA thrombolysis after successful EVT is biologically meaningful. The eTICI-shift rate of 6.5–8.7% in the active arms means ~92% of patients got no additional angiographic reperfusion from the IA tenecteplase — and yet the higher dose still produced a 9.7-point absolute gain in mRS 0–1. That is biologically implausible if you believe the drug’s only mechanism is thrombus dissolution. The paradigm-killing interpretation is that tenecteplase at microdoses is doing something other than mechanical reperfusion — perhaps microcirculatory/no-reflow rescue (the no-reflow phenomenon hypothesis in the introduction), or perhaps a placebo/nocebo effect on unblinded post-procedure care. Until a phase 3 includes perfusion-weighted MRI to demonstrate tissue-level reperfusion changes, the 0.0625 mg/kg signal remains mechanistically unanchored.

MVP — Minimum Viable Proof

If a phase 3 randomized trial of IA tenecteplase 0.0625 mg/kg vs control in n ≥ 800 anterior-circulation LVO thrombectomy recipients shows adjusted RR for mRS 0–1 of ≥ 1.20 (95% CI lower bound > 1.05) with sICH not exceeding 7%, that is sufficient to update bedside practice: every thrombectomy suite adds tenecteplase to its formulary for the 5-minute post-reperfusion bolus.

Best Combination

DATE complements CHOICE (JAMA 2022, IA alteplase) by suggesting that tenecteplase — which has largely replaced alteplase for IV thrombolysis — may be deliverable intra-arterially at lower mg doses (0.0625 vs CHOICE’s 0.125 mg/kg alteplase equivalent) with comparable safety. The two trials together support a class effect of adjunctive microdose IA thrombolysis; the next phase 3 should likely use tenecteplase at 0.0625 mg/kg as the lead candidate, with alteplase at quarter-dose as the comparator arm to settle which agent is preferred. Combining the DATE 0.0625 mg/kg data with the CHOICE alteplase data via Bayesian borrowing would compress the phase 3 timeline by ~30%.

Overvalue Warning

First, the +9.7-point absolute mRS 0–1 improvement at 0.0625 mg/kg is from a 46-patient arm with a CI that crosses 1 — readers should not quote it as evidence of efficacy. Second, the safety conclusion (no significant sICH increase at 0.0313 or 0.0625 mg/kg) is also powered thinly; the 6.5% vs 3.1% doubling at the higher dose is a real signal that a larger trial will either confirm or refute. Third, the “all-comer LVO” extrapolation is unsupported: this was an Asian, anterior-circulation, small-core, no-prior-IV-thrombolysis population. Adopting the protocol in a posterior-circulation or large-core LVO thrombectomy suite extrapolates beyond the data.

Educational use notice. The Clotbust Review is dedicated to medical education. Figures from published articles are embedded for teaching purposes, with attribution to the original publication. We do not claim copyright over any embedded figure; all rights remain with the original publisher and authors under the article's published license. If you are a rights holder and would like a figure removed, please contact the editorial team. See our editorial notice for the full policy.