Although not quite meeting its primary endpoint,
And a new meta-analysis (MAGNA) of previous studies in a similar population has provided more detailed estimates of the treatment benefit of thrombectomy in these patients.
The TESLA trial, which included patients with large-core infarcts (ASPECTS score 2-5) within 24 hours of symptom onset, showed encouraging trends towards a benefit with thrombectomy for the primary outcome of 90-day utility-weighted scores on the modified Rankin scale (mRS), but this did not reach the prespecified Bayesian superiority threshold.
Several secondary efficacy endpoints also showed suggestions of benefits with thrombectomy.
“The interventional group had higher mean or average utility-weighted mRS scores than the control group which means that their functional recovery at 90 days was trending for better outcome and less disability,” lead TESLA investigator, Osama Zaidat, MD, neuroscience & stroke director at Mercy St. Vincent Medical Center, Toledo, Ohio, said in an interview. “They also showed better neurological improvement and a higher chance of achieving a good outcome (mRS 0-3).”
These patients with large-core infarct strokes were not included in the initial trials of endovascular therapy in patients presenting in the late time window, up to 24 hours, as it was thought they would not benefit. However, three recent trials (RESCUE-Japan LIMIT; ANGEL ASPECT; and SELECT 2) have shown that patients with large core infarcts can still benefit from endovascular thrombectomy.
While these three previous trials used sophisticated imaging techniques (MRI or CT perfusion) to select patients, and restricted patients included to those with an ASPECTS score of 3-5, the TESLA study had a more pragmatic design, using just noncontrast CT scan evaluation without advanced imaging to select patients, and extending the inclusion criteria to patients with an ASPECTS score of 2.
“Noncontrast CT scans are available at all stroke centers so this study is more practical, highly generalizable, and more applicable globally,” Dr. Zaidat commented.
“However, our results suggest that when using noncontrast CT only to select patients, the gain or treatment effect of thrombectomy seems to be smaller than when using sophisticated advanced imaging to make the decision to go for thrombectomy or not as in the other trials,” he added.
The TESLA trial results were presented at the recent European Stroke Organisation Conference, held in Munich.
The study included 300 stroke patients with anterior circulation large‐vessel occlusion (NIHSS of 6 or more) with a large‐core infarction (investigator read ASPECTS Score 2-5), selected on the basis of noncontrast CT scan, who were randomized to undergo intra-arterial thrombectomy or best medical management (control) up to 24 hours from last known well.
The trial had a Bayesian probabilities design, with a primary endpoint of the 90-day utility-weighted mRS (uw-mRS), a relatively new patient-centered outcome used in stroke trials, which includes a quality-of-life measurement. Utilities represent preferences for mRS health states and range from 0 (death) to 1 (perfect health), so in contrast to the traditional mRS scores, a higher uw-mRS score is better.
The 90-day uw-MRS scores were 2.93 in the thrombectomy group vs. 2.27 in the control group.
The Bayesian probability of thrombectomy superiority was 0.957, which Dr. Zaidat said was “similar” to a P value of .043, but this was less than the prespecified superiority probability of > .975 to declare efficacy.
A separate analysis in a population of patients selected by core-lab read noncontrast CT scan, showed a Bayesian probability of benefit with thrombectomy of 0.98, “similar” to one-sided P value of .02.
In terms of secondary endpoints, there were also some encouraging trends, including a suggestion of benefit in the 90-day mRS ordinal shift (odds ratio 1.40; P = .06).
The number of patients achieving functional independence (mRS 0-2) was 14% in the thrombectomy groups vs. 9% in the control group (P = .09); and a good functional outcome (mRS 0-3) was achieved in 30% of thrombectomy patients vs. 20% of those in the control group (P = .03).
Major neurological improvement (NIHSS scale of 0-2 or improvement of 8 points or more) occurred in 26% of thrombectomy patients vs. 13% of controls (P = .0008).
Quality of life, measured by the EuroQol 5-Dimension 5-Level survey, also showed a trend towards improvement in the thrombectomy group with mean scores of 53 vs. 46 (P = .058).
In terms of safety, all-cause mortality was similar in the two groups (35% thrombectomy and 33% control) and symptomatic intracerebral hemmorhage (ICH) occurred in 3.97% of thrombectomy vs. 1.34% of control patients (relative risk, 2.96).
“Cost-effective analysis and additional subgroup studies will provide more insight about the training needs to read the CT scan and if there is any value to treat patients with an ASPECTS score of 2,” Dr. Zaidat concluded.
“Larger pooled analysis will also be very useful in understanding the threshold of brain volume with irreversible damage beyond which thrombectomy wouldn’t be helpful,” he added.