Occlusion site and outcomes in intra-arterial tenecteplase after successful endovascular therapy: a secondary analysis of the ANGEL-TNK trial Authors: Shuo Li, Xin Tong, Huachen Zhang, Gang Luo, Ligang Song, Dapeng Sun, Wenhuo Chen, Xiaoxi Yao, Qi Zhou, Xuelei Zhang, Yiwen Xu, Thanh N Nguyen, Zhongrong Miao, Xiaochuan Huo Published: 2026 ABSTRACT AbstractBackgroundEndovascular thrombectomy achieves macroreperfusion in large vessel occlusion (LVO) stroke, but only one-quarter of patients had excellent functional outcome. Adjunct intra-arterial (IA) tenecteplase could further improve the treatment effect, yet its efficacy across internal carotid artery (ICA) versus middle cerebral artery (MCA) M1/M2 occlusion remains unclear.ObjectiveTo evaluate whether IA tenecteplase improves 90-day functional outcomes in LVO patients stratified by occlusion site (ICA, MCA M1, MCA M2).MethodsPrespecified secondary analysis of the ANGEL-TNK trial (multicenter, randomized, open-label, blinded endpoint) in 19 Chinese stroke centers. Participants were enrolled who had anterior circulation LVO, 4.5–24 hours from symptom onset, and CT angiography (CTA)/magnetic resonance angiography (MRA)-proven ICA, M1, or M2 occlusion.InterventionRandomization (1:1) to IA tenecteplase (0.125 mg/kg) or standard medical management after expanded Thrombolysis in Cerebral Infarction (eTICI) 2b50-3 reperfusion.Main outcome measureThe primary outcome was the rate of 90-day modified Rankin Scale (mRS) 0–1.ResultsThere were 256 patients in the trial, including 71 (27.7%) ICA, 122 (47.6%) MCA M1, and 62 (24.2%) MCA M2. ICA occlusion patients treated with IA tenecteplase had higher rates of 90-day mRS 0–1 (39.4% vs 13.2%; relative risk (RR) 2.99; 95% CI 1.51 to 5.96; p=0.002) and mRS 0–3 (54.5% vs 42.1%; RR 1.30; p=0.048) versus controls, with a significant shift toward better mRS scores (median 3 (IQR 1–4) vs 4 (2–6); odds ratio (OR) 2.26; p<0.001). Post hoc analysis showed higher eTICI progression in ICA patients (51.5%) versus MCA M1 (37.9%) and M2 (25.7%). IA tenecteplase had a lower any intracranial hemorrhage within 48 hours in ICA patients (15.2% vs 39.5%; RR 0.38; p<0.001) compared with standard medical management. No significant benefits were observed in the MCA M1/M2 subgroups, and interaction effects were significant between ICA and MCA segments for functional outcomes and safety.Conclusions and relevanceIA tenecteplase had a better functional outcome and lower hemorrhage risk in ICA occlusion compared with standard medical management but not in MCA M1/M2. Occlusion site is a critical determinant of response to IA tenecteplase. A pooled analysis of IA tenecteplase stratified by occlusion site strata is warranted. WHAT IS ALREADY KNOWN ON THIS TOPIC Intra-arterial (IA) tenecteplase shows potential for large vessel occlusion (LVO) stroke, but prior research did not stratify its efficacy/safety by occlusion site (internal carotid artery (ICA) vs middle cerebral artery (MCA) M1/M2)—a gap given their differing anatomy and stroke severity, which may affect treatment response. Intra-arterial (IA) tenecteplase shows potential for large vessel occlusion (LVO) stroke, but prior research did not stratify its efficacy/safety by occlusion site (internal carotid artery (ICA) vs middle cerebral artery (MCA) M1/M2)—a gap given their differing anatomy and stroke severity, which may affect treatment response. WHAT THIS STUDY ADDS In this prespecified secondary analysis of the ANGEL-TNK trial, IA tenecteplase improved 90-day disability-free outcomes in ICA occlusion (39.4% vs 13.2%, p=0.002) but not in MCA M1/M2. IA tenecteplase had lower hemorrhage risk in ICA patients compared with standard medical management. In this prespecified secondary analysis of the ANGEL-TNK trial, IA tenecteplase improved 90-day disability-free outcomes in ICA occlusion (39.4% vs 13.2%, p=0.002) but not in MCA M1/M2. IA tenecteplase had lower hemorrhage risk in ICA patients compared with standard medical management. HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE, OR POLICY Occlusion site determines response to IA tenecteplase, supporting targeted use in ICA LVO to optimize outcomes. Occlusion site determines response to IA tenecteplase, supporting targeted use in ICA LVO to optimize outcomes. Introduction Endovascular thrombectomy (EVT) remains the optimal treatment for patients with large vessel occlusion (LVO) acute ischemic stroke (AIS).1 While previous randomized trials show that up to 90% of LVO-AIS patients achieve successful reperfusion (defined as an expanded Thrombolysis in Cerebral Infarction (eTICI) score of 2c-3) after mechanical thrombectomy, only 26.9% are disability-free and 46.0% achieve functional independence at 90 days—a phenomenon termed “futile thrombectomy”.2 Currently, successful reperfusion after mechanical thrombectomy is defined as reperfusion of at least 50% of the affected vascular territory (scores on the eTICI scale of 2c to 3). No-reflow, hypothesized as a key mechanism of futile thrombectomy, is defined by Tmax>6 s on CT perfusion and occurs in up to 80% of cases despite successful reperfusion.3 4 Persistent microcirculatory hemodynamic impairment post-thrombectomy underscores the need for adjunctive therapies. Adjunct intra-arterial (IA) fibrinolytic agent after thrombectomy is assumed to dissolve the residual microthrombi in the cerebral microcirculation. Recent trials have yielded conflicting results: the CHOICE (Chemical Optimization of Cerebral Embolectomy) trial showed that IA alteplase improved 90-day disability-free outcomes without increasing symptomatic intracranial hemorrhage (sICH) but was prematurely terminated due to placebo supply shortages.5 In contrast, POST-TNK (Intra-Arterial Tenecteplase Following Endovascular Reperfusion for Large Vessel Occlusion Acute Ischemic Stroke) and POST-UK (Intra-Arterial Urokinase After Endovascular Reperfusion for Acute Ischemic Stroke) trials found no benefit of IA tenecteplase or urokinase after near-complete reperfusion (eTICI 2c-3),6 7 while ATTENTION-IA (Intra-Arterial Tenecteplase After Successful Endovascular Recanalization in Patients with Acute Posterior Circulation Arterial Occlusion) showed no improvement with IA tenecteplase in posterior circulation occlusions.8 The ANGEL-TNK trial, however, demonstrated that IA tenecteplase improved 90-day functional independence (modified Rankin Scale (mRS) 0–1) in anterior circulation LVO patients treated 4.5–24 hours after stroke onset.9 These divergent findings highlight the need to identify optimal patient subgroups for IA fibrinolysis. The ANGEL-TNK trial enrolled patients with internal carotid artery (ICA), middle cerebral artery (MCA) M1, or dominant M2 occlusions. Anatomical differences across these sites underlie distinct clinical profiles: ICA occlusion typically involves a larger thrombus burden and affects a broader vascular territory with fewer collaterals, leading to more severe ischemic injury. Mechanistically, ICA occlusion is associated with higher rates of distal microembolization.10 In contrast, MCA M1/M2 occlusions often involve smaller thrombi but may disrupt perfusion to critical motor/sensory cortices. These structural and hemodynamic variations may influence response to IA fibrinolytics: ICA lesions might require more aggressive thrombolysis to resolve residual microthrombi, while M1/M2 sites could exhibit differing reperfusion dynamics. Previous trials have shown that occlusion site modulates outcomes after endovascular therapy, yet few have explicitly evaluated adjunctive fibrinolysis across anatomical subgroups.11 This prespecified secondary analysis of the ANGEL-TNK trial aimed to address this gap by systematically evaluating outcomes of EVT with adjunct IA tenecteplase across distinct occlusion sites. Methods Study design This prespecified secondary subgroup analysis was derived from the ANGEL-TNK trial (NCT05624190), a multicenter, prospective, randomized, open-label, blinded endpoint trial conducted at 19 comprehensive stroke centers in China.9 A complete list of the ANGEL-TNK Investigators appears in online supplemental file 1. The parent trial investigated whether IA tenecteplase thrombolysis after successful endovascular therapy for anterior circulation LVO improves clinical outcomes in a Chinese population.12 Patients were randomized 1:1 to receive IA tenecteplase (recombinant human tenecteplase-tissue plasminogen activator, rhTNK-tPA) or standard medical management (MM). The trial protocol is detailed elsewhere11 and in online supplemental 2, with informed consent obtained from each patient or legal representative. Ethical approval was granted by institutional review boards at all participating centers, and the study adhered to the Consolidated Standards of Reporting Trials (CONSORT) guidelines. The current analysis aimed to determine whether rhTNK-tPA confers superior functional recovery compared with MM among LVO patients stratified by occlusion site (ICA, MCA M1, or dominant M2). Patients were categorized into three groups according to occlusion location, which was initially determined by CT angiography (CTA) or magnetic resonance angiography (MRA) for screening and subsequently confirmed during endovascular treatment. Patient Population Patients aged over 18 years and with last known well onset of symptoms to randomization time between 4.5 to 24 hours were enrolled. The other inclusion criteria were: (1) National Institutes of Health Stroke Scale (NIHSS) (range 0 to 42, higher scores indicating greater neurological deficit) score of ≥2, (2) pre-stroke score of 0 or 1 on the modified Rankin scale (mRS, range 0 to 6, higher scores indicating increasing disability, and 6 indicating death) before stroke onset, and (3) signed informed consent from the patient or their healthcare proxy. Imaging inclusion criteria were: (1) CTA- or MRA-proven intracranial occlusion of the ICA, M1 or dominant M2 segment of the MCA; (2) Alberta Stroke Program Early CT Score (ASPECTS) ≥6 on non-contrast CT (NCCT) scan or diffusion-weighted imaging (DWI) MRI; (3) CT perfusion (CTP) or MR perfusion: ischemic infarct core <70 mL, mismatch ratio ≥1.2, and mismatch volume ≥10 mL; and (4) treated with endovascular therapy resulting in an eTICI score of 2b50-3 at the end of the procedure. Patients who received intravenous thrombolysis on admission, intravenous heparin during treatment, or who would likely require antiplatelet therapy within the first 24 hours after endovascular treatment were excluded from enrollment. Patients were ineligible if they had midline shift or clinical signs of herniation, mass effect, high risk of hemorrhage, acute bilateral strokes, or multiple intracranial occlusions. Additional inclusion and exclusion criteria are provided in the study protocol.12 Intervention Following successful reperfusion (eTICI 2b50-3), eligible participants were randomized 1:1 to IA tenecteplase or MM. MM followed Chinese Stroke Association guidelines.13 For the IA tenecteplase group, recombinant human tenecteplase (rhTNK-tPA; 0.125 mg/kg, max 12.5 mg) was infused over 15 min via a microcatheter positioned in the horizontal segment of the MCA distal to the lenticulostriate branches. Post-infusion angiography (anteroposterior and lateral views) recorded final eTICI grades. The MM group underwent control angiography without pharmacological intervention. Study Outcomes The primary outcome was excellent outcome (mRS 0–1) at 90 days. Rankin Structured Interview (RSI) was used as a scoring tool to assess the mRS.14 Secondary outcomes included original mRS (shift analysis) at 90 days, functional independence (mRS 0–2) at 90 days, independent ambulation (mRS 0–3) at 90 days, early neurological improvement defined as an NIHSS score of 0 or 1 or improvement by ≥10 points at 36 hours, change in infarct volume defined as from baseline imaging (CT perfusion or DWI) to NCCT at 7 days or at discharge (whichever was earlier) or to MRI at 36 hours, time to maximum >6 s (Tmax; CTP parameters used to measure the perfusion status of tissues) and European Quality of Life Visual Analog Scale (EQ-VAS; range, 0 to 100, with lower scores indicating worse quality of life). Safety outcomes were sICH within 48 hours following randomization defined by the Heidelberg Bleeding Classification (an increase by ≥4 points or an increase by ≥2 points of a NIHSS subcategory is a relevant change in neurological status with any ICH on imaging),15 any ICH within 48 hours, and death within 90 days from the onset of stroke. Statistical Analysis Patients were analyzed according to their randomization group. All randomized subjects are included in the full analysis set. Categorical variables were reported as frequencies (%) and continuous variables as medians (IQR). Baseline comparisons used Mann–Whitney U tests (continuous) or χ²/Fisher’s exact tests (categorical). Confounders were selected as variables with p<0.05. For the primary outcome, a generalized linear model with log link and binomial error distribution with adjustment for study center and occlusion site was used to calculate the relative risk (RR) and the 95% confidence interval (95% CI) between the two treatment groups. Similar models were used for the secondary outcomes of mRS 0–2 at 90 days, mRS 0–3 at 90 days, NIHSS 0–1 or improved ≥10 points at 24 hours, and bleeding events within 48 hours. Wilcoxon–Mann–Whitney generalized odds ratio (OR) with its 95% CI was calculated to compare the distribution of mRS at 90 days as the proportional odds assumption for ordinal logistic regression was not met (p<0.001). Linear mixed model with adjustment for study center and occlusion site was used to compare the Tmax>6 s volume at 24 hours, 24 hours infarct core volume change from baseline, and EQ-VAS at 90 days between the two treatment groups. The Cox proportional hazard model with site as a random effect was used to compare the risk of death within 90 days between the two treatments, and the hazard ratio (HR) with its 95% CI was calculated. All statistics were two-sided with p<0.049 considered significant. Statistical analyses were performed with SAS version 9.4 (SAS Institute Inc., Cary, NC, USA). Results A total of 256 patients were enrolled, with 126 randomly assigned to the IA tenecteplase (TNK) group and 129 to the medical management (MM) group. Of these, 71 (27.7%) had ICA occlusion (MM: 33, TNK: 38), 122 (47.6%) had MCA M1 occlusion (MM: 58; TNK: 64), and 62 (24.2%) had M2 occlusion (MM: 30, TNK: 32). Baseline Characteristics Baseline characteristics by occlusion site are summarized in online supplemental table 1, with subgroup comparisons between TNK and MM provided in online supplemental table 3. ICA occlusion patients had a median age of 71.9 (IQR 61.7–80.7) years, with 30 (42.3%) female and 41 (57.7%) male. Median baseline NIHSS was 16.0 (14.0–21.0), ASPECTS 6.0 (6.0–8.0), and infarct volume 26.4 (8.8–54.6) mL. Most had eTICI 2b67 (43.7%) at randomization, followed by 2c (26.8%), 2b50 (23.9%), and 3 (5.6%). Median time from symptom onset to randomization was 9.4 (6.6–12.2) hours. MCA M1 occlusion patients had a median age of 70.9 (60.6–77.1) years, 58 (47.5%) female, median NIHSS 14.5 (12.0–18.0), ASPECTS 7.0 (6.0–8.0), infarct volume 18.0 (3.8–44.0) mL, and eTICI 2b67 (32.8%) predominant. Median time to randomization was 9.5 (6.9–14.0) hours. MCA M2 occlusion patients had a median age of 72.5 (62.8–80.2) years, 26 (41.9%) female, median NIHSS 15.0 (10.0–19.0), ASPECTS 6.0 (6.0–8.0), infarct volume 14.0 (3.3–35.0) mL, and eTICI 2b67 (35.5%) most common. Median time to randomization was 9.5 (6.5–13.4) hours. ICA occlusion patients had higher baseline NIHSS, lower ASPECTS, and larger infarct volumes versus the M1/M2 groups. No baseline differences existed between the TNK and MM subgroups within any occlusion site. Study Outcomes Table 1 presents a summary of the study outcomes categorized by occlusion site, with detailed data available in online supplemental table 2. For the primary outcome, the proportion of patients with mRS 0–1 at 90 days was significantly higher in the TNK group than in the MM group among ICA occlusion patients (39.4% vs 13.2%; relative risk (RR) 2.99; 95% CI 1.51 to 5.96; p=0.002). No significant differences were found in M1 (32.8% vs 26.6%; RR 1.23; 95% CI 0.87 to 1.74; p=0.235) or M2 occlusion groups (54.3% vs 44.4%; RR 1.22; 95% CI 0.70 to 2.14; p=0.484) (figure 1). Interaction effects were significant between ICA and M1 (p=0.017) as well as ICA and M2 (p=0.030) but not across all three occlusion sites (p=0.169). Table 1Study outcomes stratified by occlusion site (internal carotid artery, M1, M2) and treatment (intra-arterial tenecteplase versus medical management): effect sizes and statistical valuesOutcomeICAM1M2P-value for interactionEffect sizeP-valueEffect sizeP-valueEffect sizeP-valueAll groupsPrimary outcome mRS score 0–1 at 90 days, no. (%)2.99 (1.51–5.96)0.0021.23 (0.87–1.74)0.2351.22 (0.70–2.14)0.4840.1685Secondary outcome mRS score at 90 days2.26 (1.48–3.45)<0.0010.80 (0.52–1.25)0.3351.07 (0.36–3.17)0.9010.0848 mRS score 0–2 at 90 days, no.(%)1.50 (0.91–2.46)0.1130.80 (0.58–1.12)0.1950.95 (0.57–1.60)0.8550.3011 mRS score 0–3 at 90 days, no. (%)1.30 (1.00–1.67)0.0480.80 (0.60–1.05)0.1100.89 (0.60–1.32)0.5560.0579 Tmax>6 s volume at 24 hours*−2.73 (−49.22–43.76)0.906−17.49 (−50.65–15.68)0.297−5.15 (−42.84–32.54)0.7840.7620 NIHSS 0–1 or improved ≥10 points at 36 hours†1.15 (0.42–3.17)0.7850.88 (0.41–1.91)0.7510.64 (0.33–1.27)0.2030.6859 Infarct core volume change from baseline‡5.85 (−9.63–21.32)0.451−3.17 (−16.93–10.59)0.648−3.87 (−19.88–12.14)0.6280.6294 EQ-VAS at 90 days, median (IQR)20.38 (3.84–36.91)0.017−3.53 (−17.22–10.16)0.6100.38 (−18.94–19.7)0.9690.0943Safety outcomes sICH within 48 hours§0.46 (0.10–2.22)0.3342.76 (0.56–13.68)0.214––0.1080 Any ICH within 48 hours0.38 (0.24–0.63)<0.0011.23 (0.67–2.27)0.5001.35 (0.78–2.34)0.2840.0269 Death within 90 days, no. (%)1.00 (0.23–4.32)0.9970.34 (0.14–0.84)0.0201.25 (0.22–7.24)0.8030.0156*There were 83 patients who could not be assessed with perfusion because of unsuitable clinical practice (34 cases in the IA tenecteplase group and 49 in the MM group).†The NIHSS score is an ordinal scale to evaluate the severity of stroke ranging from 0 to 42, with higher scores indicating more severe deficits. Data were missing for one patient in the MM group because of early discharge.‡Infarct core volume change from baseline imaging (calculated by CTP or DWI) to NCCT at 7 days or at discharge (whichever is earlier) or MRI at 36 hours. There were two patients who could not be assessed because of serious illness or death (both in the MM group).§sICH was defined according to the Heidelberg Bleeding Classification (an increase in the NIHSS score of ≥4 points or an increase in the score for an NIHSS subcategory of ≥2 points with any ICH on imaging).CTP, computed tomography perfusion; DWI, diffusion-weighted imaging; EQ-VAS, European Quality of Life Visual Analog Scale; IA, intra-arterial; ICA, internal carotid artery; ICH, intracranial hemorrhage; M1, middle cerebral artery M1; M2, middle cerebral artery M2; MM, medical management; MRI, magnetic resonance imaging; mRS, modified Rankin scale; NIHSS, National Institutes of Health Stroke Scale; NNCT, non-contrast computed tomography; sICH, symptomatic intracranial hemorrhage. Study outcomes stratified by occlusion site (internal carotid artery, M1, M2) and treatment (intra-arterial tenecteplase versus medical management): effect sizes and statistical values There were 83 patients who could not be assessed with perfusion because of unsuitable clinical practice (34 cases in the IA tenecteplase group and 49 in the MM group). The NIHSS score is an ordinal scale to evaluate the severity of stroke ranging from 0 to 42, with higher scores indicating more severe deficits. Data were missing for one patient in the MM group because of early discharge. Infarct core volume change from baseline imaging (calculated by CTP or DWI) to NCCT at 7 days or at discharge (whichever is earlier) or MRI at 36 hours. There were two patients who could not be assessed because of serious illness or death (both in the MM group). sICH was defined according to the Heidelberg Bleeding Classification (an increase in the NIHSS score of ≥4 points or an increase in the score for an NIHSS subcategory of ≥2 points with any ICH on imaging). CTP, computed tomography perfusion; DWI, diffusion-weighted imaging; EQ-VAS, European Quality of Life Visual Analog Scale; IA, intra-arterial; ICA, internal carotid artery; ICH, intracranial hemorrhage; M1, middle cerebral artery M1; M2, middle cerebral artery M2; MM, medical management; MRI, magnetic resonance imaging; mRS, modified Rankin scale; NIHSS, National Institutes of Health Stroke Scale; NNCT, non-contrast computed tomography; sICH, symptomatic intracranial hemorrhage. Modified Rankin Scale distribution stratified by occlusion sites. ICA, internal carotid artery; M1, middle cerebral artery M1; M2, middle cerebral artery M2;MM, medical management; mRS, modified Rankin Scale; TNK, intra-arterial tenecteplase. Regarding secondary outcomes, the TNK group showed a significant shift toward better 90-day mRS scores in ICA occlusion patients (median score 3 (IQR 1–4) vs 4 (IQR 2–6); generalized OR 2.26; 95% CI 1.48 to 3.45; p<0.001), but not in M1 (median score 4 (IQR 1–6) vs 2 (IQR 1–5); generalized OR 0.80; 95% CI 0.52 to 1.25; p=0.335) or M2 occlusion groups (median score 1 (IQR 0–4) vs 2 (IQR 1–4); generalized OR 1.07; 95% CI 0.36 to 3.17; p=0.901). The TNK group was associated with higher rates of mRS 0–3 at 90 days (54.5% vs 42.1%; RR 1.30; 95% CI 1.00 to 1.67; p=0.048) and better EQ-VAS scores at 90 days (median 70.0 (IQR 45.0–95.0) vs 40.0 (IQR 0.0–80.0); β 20.38; 95% CI 3.84 to 36.91; p=0.017), while no significant difference was observed in mRS 0–2 (39.4% vs 26.3%; RR 1.50; 95% CI 0.91 to 2.46; p=0.113) in ICA occlusion patients, which was also not shown in the M1/M2 groups. Post hoc analysis revealed a higher proportion of eTICI progression in the ICA occlusion group (51.5%) compared with the M1 (37.9%) and M2 (25.7%) groups (figures 2 and 3). eTICI progression was defined as any improvement in the eTICI score between the initial angiographic result immediately after thrombectomy and the final angiographic assessment after AI TNK administration. Expanded Thrombolysis in Cerebral Infarction (eTICI scores) at randomization and final angiography. In the intra-arterial (IA) tenecteplase group, scores reflect angiographic assessment before and after IA tenecteplase administration; in the medical management group, scores are derived from control angiography without pharmacological intervention. ICA, internal carotid artery; M1, middle cerebral artery M1; M2, middle cerebral artery M2; MM. medical management; TNK, intra-arterial tenecteplase. Changes in expanded Thrombolysis in Cerebral Infarction (eTICI) scores between randomization and final angiography. eTICI progress is defined as an improvement of at least 1 eTICI score. EVT, endovascular thrombectomy; ICA, internal carotid artery; M1, middle cerebral artery M1; M2, middle cerebral artery M2; MM, medical management; TNK, intra-arterial tenecteplase. In terms of safety outcomes, no significant difference in sICH within 48 hours was found in the ICA or M1 occlusion groups, and no valid statistical result was obtained for the M2 group. The TNK group showed lower rates of any ICH within 48 hours in the ICA occlusion group (15.2% vs 39.5%; RR 0.38; 95% CI 0.24 to 0.63; p<0.001), with significant interaction effects between ICA and M1 (p=0.0009), ICA and M2 (p=0.001), and across all three occlusion sites (p=0.027). In the M1 occlusion group, patients in the TNK group had a significantly lower 90-day mortality risk (HR 0.34; 95% CI 0.14 to 0.84; p=0.020), with significant interaction effects between ICA and M1 (p=0.019) and across all occlusion sites (p=0.016). Discussion This prespecified analysis reveals a striking dichotomy in treatment response based on occlusion site: IA tenecteplase significantly improved 90-day disability-free outcomes (mRS 0–1), distribution of the 90-day mRS score (mRS shift), and 90-day functional recovery (mRS 0–3) in patients with ICA occlusion patients, but conferred no benefit in MCA M1/M2 segments. Notably, ICA occlusion patients treated with IA tenecteplase also had lower rates of any ICH, whereas no such safety signals were observed in the MCA subgroups. These findings suggest that anatomical occlusion site could be one of the critical determinants of response to IA fibrinolysis after EVT. Adjunct IA fibrinolysis is a promising strategy for managing no-reflow after successful thrombectomy, yet randomized trials have yielded conflicting results. The CHOICE trial demonstrated that IA alteplase improved 90-day functional independence without increasing sICH but was terminated early due to placebo shortages.5 In contrast, POST-TNK, POST-UK, and ATTENTION-IA found no benefit of IA fibrinolytics compared with thrombectomy alone.6–8 The ANGEL-TNK trial found IA tenecteplase after successful endovascular treatment resulted in a higher percentage of patients with a non-disabled outcome (mRS score of 0 or 1) at 90 days than standard medical treatment. The ANGEL-TNK trial differed from the previous trials. First, ANGEL-TNK restricted the use of periprocedural antithrombotic drugs via defined inclusion and exclusion criteria and thus administered a higher IA tenecteplase dose (0.125 mg/kg) compared with other trials. Second, ANGEL-TNK only included patients between 4.5 to 24 hours because these patients were not considered eligible for intravenous thrombolysis according to Chinese guidelines.16 Third, ANGEL-TNK included the same population of patients who were successfully recanalized (eTICI 2b-3, 50–100% reperfusion) after endovascular therapy as the CHOICE trial, while the POST-UK and POST-TNK trials excluded patients with eTICI 2b50-67 (50–89% reperfusion). The ANGEL-TNK, CHOICE, POST-TNK, and POST-UK trials included patients with the same occlusion site, including ICA, M1, and M2. ANGEL-TNK enrolled a higher proportion of ICA occlusions (27.8%) versus CHOICE (9.7%), POST-TNK (21.5%), and POST-UK (23.4%). This may explain the divergent results, as ICA occlusion is associated with greater thrombus burden and difficulty achieving eTICI>2b reperfusion. The median ASPECTS is low compared with many thrombectomy trials.2 This may be one reason for the lower proportion of mRS 0–1 in the control group, especially in the ICA occlusion subgroup. Several mechanisms underlie the observed site-specific efficacy. First, ICA occlusion imposes a greater microcirculatory burden, as evidenced by higher NIHSS scores, larger infarct volumes, and lower ASPECTS in our ICA subgroup.10 17 Second, the ICA could have more rates of medium-sized thrombus migration and embolization, causing secondary medium vessel occlusion (MEVO), which has been reported to occur in one in four patients following intravenous thrombolysis,18 and in up to 9% of EVT cases.19 Clot fragility in secondary MEVO may be higher because clot result from fragmentation of a more proximal occlusion. Adjunct IA alteplase could be more effective in managing the secondary MEVO than further EVT. These are supported by our post hoc analysis showing greater eTICI score progression and larger eTICI 3 score proportion in ICA patients treated with IA tenecteplase than the M1 and M2 occlusion subgroups. Although successful reperfusion is defined as eTICI 2b-3, several studies found that eTICI 3 had better functional outcome than eTICI 2b.20 Third, the ICA supplies a larger territorial territory with fewer collaterals compared with the MCA segments, making microvascular reperfusion more dependent on pharmacological intervention. We observed that in the ICA occlusion group, the TNK group had trends of lower rates of any ICH within 48 hours than the M1 and M2 occlusion subgroups. Although without statistical difference, the sICH is also lower in the TNK group in ICA occlusion patients. There could be several explanations. First, MR CLEAN Med trial results showed a higher risk of sICH in patients who received periprocedural intravenous aspirin and unfractionated heparin during endovascular stroke treatment.21 Therefore, the ANGEL-TNK trial restricted the use of periprocedural antithrombotic drugs. Second, administration of tenecteplase in the ANGEL-TNK trial was through a microcatheter in the horizontal segment of the MCA artery distal to the origin of the lenticulostriate branches. This avoids the potential bleeding risk associated with high local drug concentrations in the lenticulostriate branches area. Third, in ICA occlusion patients, more than 50% had eTICI progression after tenecteplase administration and 57.5% of patients had an eTICI 2c-3 in the final angiography, which correlates with lower hemorrhage risk.22–24 This study included patients with M2 occlusion, which was considered one of the MEVO sites and widely discussed. M2 segment of MCA was the largest proportion of occlusion sites in the DISTAL (Endovascular Treatment for Stroke Due to Occlusion of Medium or Distal Vessels) and ESCAPE-MeVO (Endovascular Treatment of Stroke Due to Medium-Vessel Occlusion) trials.25 26 The rate of excellent functional outcome (mRS 0–1) at 90 days of the MM group in the ANGEL-TNK trial was comparable with these two trials. The TNK group of the M2 occlusion group had a higher mRS 0–1 proportion than the MM group (54.3% vs 44.4%) but without a significant difference. Future trials including TECNO (Safety and Efficacy of Intra-Arterial Tenecteplase for Noncomplete Reperfusion of Intracranial Occlusions, NCT05499832) and RESCUE-TNK (Rescue Thrombolysis for Medium Vessel Occlusion, NCT05657470) will further define IA tenecteplase’s role in MEVO. Limitations This study has several limitations. First, subgroup sample sizes were small, potentially limiting the detection of true treatment effects in the M1/M2 groups, though trends favored IA tenecteplase. Second, the underpowered design increased the risk of type II error. Third, results are specific to Chinese patients with cardioembolic stroke, limiting generalizability. Fourth, exclusion of patients receiving intravenous thrombolysis, heparin, or antiplatelets may affect external validity. Conclusions IA tenecteplase improves functional outcomes and reduces ICH in ICA occlusion patients after EVT but shows no benefit in MCA M1/M2 segments. These findings support targeted use of IA tenecteplase in ICA occlusion and highlight the need for MCA-specific trials to optimize treatment strategies. Future research is needed to further clarify the efficacy and safety of IA tenecteplase in application at different occlusion sites. Supplementary material