A new pooled analysis of two smaller trials testing the new approach of dropping thrombolysis before thrombectomy in patients with large vessel occlusion stroke has shown similar outcomes with and without tissue plasminogen activator (tPA), but has also hinted at specific subgroups where one approach or the other may be preferred.
“I think the data show that bridging with thrombolysis before thrombectomy is neither dangerous or a panacea, and the overall results highlight the importance of departing from polarized decisions in favor of precision medicine,” said lead author Raul Nogueira, MD, Emory University, Atlanta, Georgia.
“The critical question should not be: ‘Should we or should we not be administering thrombolysis to all eligible patients presenting to a thrombectomy capable center?’ but rather: ‘Who are the specific patients who should and should not receive thrombolysis in a thrombectomy capable center?'” Nogueira concluded.
The Systemic Thrombolytic Randomization in Endovascular Stroke Therapy (SHRINE) analysis was presented at the virtual International Stroke Conference (ISC) 2021 on March 18.
In his presentation, Nogueira explained that three randomized trials have recently been completed in Asia to evaluate the noninferiority of endovascular treatment alone (primary/direct thrombectomy) versus intravenous thrombolysis followed by endovascular therapy (combined/bridging therapy) in anterior circulation large vessel occlusion stroke in patients eligible for thrombolysis within 4.5 hours from stroke onset. These are the DIRECT-MT trial, the SKIP trial, and the DEVT trial.
“Despite their individual merits, each of these trials had some limitations mostly related to their sample sizes and generous noninferiority margins,” Nogueira said.
The current SHRINE analysis was designed to pool patient-level data from two of these trials — SKIP from Japan and DEVT from China — performed between 2017-2020. Both trials were published in JAMA earlier this year.
The main results of the pooled analysis just narrowly missed showing noninferiority of direct thrombectomy over the standard practice of bridging therapy with thrombolysis.
“We found that the lower boundary for the adjusted odds ratio for the primary outcome fell just short of significance so we cannot fully confirm the noninferiority of direct thrombectomy in this pooled analysis. However, there was a strong trend (P = .06), and most point estimates, if not all, directionally favor direct thrombectomy,” Nogueira reported.
When looking at the subgroup analysis, the researchers found some interesting trends.
“While our subgroup analysis was frequently underpowered, I think we had significant contributions here with three groups that are likely to benefit from primary thrombectomy (interaction P < .10),” he said. “These are those presenting over 3 hours from symptom onset, those with an intracranial ICA occlusion, and those with atrial fibrillation.
“Also, there is a possibility of benefit (odds ratio > 1.5) for patients with more severe strokes, intracranial atherosclerotic disease, or hyperglycemia. It will be important to look at these groups carefully in further subgroup analyses of larger datasets,” Nogueira commented.
Introducing his presentation, he noted that the use of thrombolysis prior to endovascular therapy has benefits but also potential disadvantages, including additional risks and costs. So the current studies were designed to investigate whether thrombolysis could be omitted in patients going straight to thrombectomy.
“It is critical to understand that any comparisons of direct thrombectomy versus bridging with thrombolysis before thrombectomy only apply to certain situations. These are when patients are taken to thrombectomy capable centers with fast thrombectomy workflow and there is immediate availability of the neuro-endovascular team,” Nogueira cautioned.
“These studies do not apply to primary stroke centers, those centers with suboptimal thrombectomy workflow, or situations when the neuro-endovascular team is not immediately available. In these conditions, we must continue to provide thrombolysis to all eligible patients,” he added.
The SKIP and DEVT trials had a similar design, both randomizing patients to thrombectomy alone or thrombolysis plus thrombectomy. They also had similar inclusion criteria including premorbid modified Rankin Scale (mRS) score of 0-1, baseline National Institutes of Health Stroke Scale (NIHSS) score of ≥ 6, intracranial internal carotid artery (ICA) or middle cerebral artery (MCA) M1 occlusion, ASPECTS score ≥ 6 NCCT or ≥ 5 DWI, and eligible for thrombolysis within 4.5 hours of stroke onset.
One difference between the two trials was that SKIP used a reduced dose of alteplase (0.6 mg/kg) whereas DEVT used the standard dose.
The primary outcome for this pooled analysis was functional independence (mRS 0-2) at 90 days, with a lower boundary for noninferiority of 0.85.
The pooled data include a total of 438 patients, 221 of whom received bridging therapy with tPA before thrombectomy and 217 who went straight to thrombectomy without thrombolysis.
Baseline characteristics showed no differences between the two groups. The median age was 72 years; median NIHSS score was 16-17; and median ASPECTS score was 8; around 25% had ICA occlusions and 75% had MCA occlusions. Time from stroke onset to puncture was a median of about 3 hours.
Results showed a trend towards a favorable effect for direct thrombectomy on the primary endpoint (mRS 0-2) at 90 days with an adjusted odds ratio of 1.27 (95% CI, 0.84 – 1.92).
“This was just short of the prespecified lower noninferiority boundary of 0.85,” Nogueira noted.
“Secondary outcomes were pretty flat, with all odds ratios slightly above 1.0, so directionally favoring direct thrombectomy but far from any significance,” he added.
These secondary outcomes were 90-day mRS shift analysis (OR, 1.01 [0.72 – 1.40]); 90-day mRS 0-1 (OR, 1.03 [0.67 – 1.58]), and 90-day mRS 0-3 (OR, 1.11 [0.70 – 1.77]).
Safety outcomes showed a trend towards an increase in symptomatic ICH in the bridging group (9.0%) vs 6.5% in the direct thrombectomy group, but this did not reach significance, with an adjusted OR of 0.77 (0.37 – 1.61).
Any ICH was significantly different, with a rate of 39.4% in the tPA group vs 26.3% in the direct thrombectomy group (adjusted OR, 0.54 [0.35 – 0.82]; P = .03)
“This endpoint of any ICH may not have such a large importance in terms of outcome, but we may want to think about this in terms of patient selection,” Nogueira commented.
Puncture site complications were, as expected, higher in the bridging group (4.1% vs 0.5%).
There was no difference in 90-day day survival; which was high at about 87% in both groups, Nogueira reported.
In terms of subgroups, Nogueira said, “I really think there are a lot of important findings here.”
He pointed out a significant interaction by time to treatment (P = .02) with patients treated beyond 3 hours doing better with direct thrombectomy (90-day mRS 0-2; OR 2.28) versus those treated earlier doing better with bridging (OR, 0.84).
There was also a suggestion of an interaction by occlusion site (P = .08) with patients having an intracranial ICA occlusion doing better with direct thrombectomy (OR, 3.04) versus those with MCA occlusion (OR, 1.05).
In addition, Nogueira highlighted certain patient categories that had a tendency to do better with direct thrombectomy, but these were not significant in terms of interaction on the primary outcome of mRS 0-2 at 90 days.
These included patients with NIHSS scores > 15, those with atrial fibrillation, and those with intracranial atherosclerotic disease. “We need to watch these groups in larger future analyses,” Nogueira said.
However, 90-day survival did show a significant interaction (P = .03) for the presence of atrial fibrillation, with AF patients doing better with direct thrombectomy (OR for 90-day survival, 1.75) and those without AF appearing to do better with bridging (OR, 0.33).
“A potential explanation for this could be the rate of symptomatic ICH which was higher in AF patients in the bridging group,” Nogueira noted.
Commenting on the study for Medscape Medical News, Yvo Roos, MD, PhD, Amsterdam Medical Center, the Netherlands, cautioned that these subgroups were probably too small for definitive answers. “They are certainly interesting but should just be considered hypothesis generating at this point,” he said.
Roos presented the results of another similar trial, MR CLEAN NO IV, at the meeting, which also showed similar outcomes with the two approaches.
He pointed out that other trials are ongoing, and there are plans for a larger meta-analysis of all the data when they become available, which will give much more reliable information on individual patient characteristics and suitability for one approach or the other.
Nogueira has reported receiving consultancy fees for advisory roles with Anaconda, Biogen, Cerenovus, Genentech, Imperative Care, Medtronic, Phenox, Prolong Pharmaceuticals, and Stryker Neurovascular and having stock options for advisory roles with Astrocyte, Brainomix, Cerebrotech, Ceretrieve, Corindus Vascular Robotics, Vesalio, Viz-Al, and Perfuse.
International Stroke Conference (ISC) 2021. Late-breaking abstract 4. Presented March 18, 2021.