IST-3 Supports Wider Use of tPA in Ischemic Stroke Patients

Daniel M. Keller, PhD

October 31, 2012

BRASILIA, Brazil — An analysis of prespecified subgroups in the third International Stroke Trial (IST-3) shows that intravenous recombinant tissue plasminogen activator (rtPA) can be given safely in a wide variety of situations of acute ischemic stroke.

Richard Lindley, MD, professor of medicine at Sydney Medical School–Westmead Hospital and George Institute for Global Health at the University of Sydney, Australia, told delegates that the benefit of tPA did not differ by prerandomization blood pressure, blood glucose, or ischemic stroke subtype.

"Importantly, we have not been able to identify a subgroup with a particularly low risk of symptomatic intracranial hemorrhage, which, of course, is the main risk of intravenous [r]tPA," Dr. Lindley said.

He added that although patients with the most severe form of stroke, total anterior circulation infarction, had a greater risk for symptomatic intracranial hemorrhage, IST-3 still showed a benefit of rtPA at 6 months even for them. "Thrombolysis therefore appeared to be applicable across a wide group of patients in IST-3," he concluded.

Results were presented here at the 8th World Stroke Congress (WSC). The IST-3 results were published in the Lancet to coincide with their presentation at the XXI European Stroke Conference in May of this year, and were reported by Medscape Medical News at that time. The study was supported by Boehringer Ingelheim.

Increased Survival With Less Disability

IST-3 was an international, multicenter, prospective, randomized, open-label trial of intravenous rtPA 0.9 mg/kg (n = 1515) administered within 6 hours of the onset of stroke symptoms vs control (n = 1520). Brain imaging excluded intracranial hemorrhage and stroke mimics.

The primary outcome was the proportion of patients alive and independent (Oxford Handicap Scale [OHS], 0 - 2) at 6 months, as determined by a postal questionnaire or a blinded telephone interview. A secondary outcome of symptomatic intracranial hemorrhage (sICH) was assessed by a masked central review of imaging scans and a central adjudication panel.

An ordinal analysis of the OHS rank detected a favorable shift in the tPA group vs control participants, with an adjusted common odds ratio (OR) of 1.27 (95% confidence interval [CI], 1.10, 1.47; P = .001). "In simple English, it's the odds of surviving with less disability were 27% greater for patients treated with rtPA," Dr. Lindley reported.

A simple analysis of the number alive and independent at 6 months did not reach statistical significance (OR = 1.13; 95% CI, 0.95, 1.35; P = .181) because the trial was underpowered to show it. However, the ordinal analysis was prespecified in the statistical analysis plan and had more statistical power because it "looks at the entirety of the data," Dr. Lindley explained.

When the investigators looked at the OHS outcomes data by ordinal analysis in terms of prerandomization systolic blood pressure in the rtPA group (143 mmHg or less, 144 - 164 mmHg, 165 mmHg or greater), there was no effect of blood pressure (P = .525 for trend).

And blood pressure had no effect on sICH within 7 days of treatment. sICH occurred in 5.1%, 7.4%, and 7.9% of patients, respectively (P = .928 for trend). Even with the rising bleeding risks, there were no differences in OHS outcomes, suggesting that rtPA was still beneficial.

Dr. Lindley did not show the data for diastolic blood pressure but mentioned that the results were the same as for systolic pressure.

Similarly, there was no independent association of blood glucose (less than 5 mmol/L, 6 - 7 mmol/L, 8 mmol/L or greater) with the OHS outcome in the rtPA group at 6 months (P = .945 for trend) or with sICH (P = .45 for trend).

Finally, stroke subtype by the Oxford Community Stroke Project classification (total or partial anterior circulation infarct, lacunar infarct, or posterior circulation infarct) did not affect the outcome at 6 months (P = .863 for trend) or sICH in the rtPA group (P = .645).

For all the variables of prerandomization systolic blood pressure, glucose, or stroke subtype, the odds ratios for outcomes were greater than 1, indicating a benefit of using rtPA. Because there were no trends within each variable, it appeared that rtPA was similarly effective across the board.

Opening the Window Further

Session moderator Werner Hacke, MD, PhD, professor and chairman of the Department of Neurology at the University of Heidelberg, Germany, commented to Medscape Medical News that in these analyses, no subgroup benefited more than another. "You could put it the other way around," he added, that all subgroups benefited equally from rtPA.

He noted that more than 80% of the patients in IST-3 would be considered off-label in both the United States and Europe. "They were too old, they were too late, they were too severe," he said. "And it is very good news that even when you look at stroke subtypes, there is not a single subtype that would do worse. I mean, that was testing the limits in a way," Dr. Hacke added. "I would not have been surprised...if lacunar [strokes] would have had more hemorrhages. But that was not true, and this is in fact the reassuring thing about IST-3."

Dr. Hacke said the trial results give added support for the use of rtPA. "Before IST-3, we said, 'when you're in doubt, don't do it.' Now when someone comes at 5 hours and is 82 [years old] and has a [National Institutes of Health Stroke Scale] of 12, you can give it a try. And you are probably not hurting him," he suggested. This will lead to higher performance rates in thrombolysis within the label, he added.

He drew an analogy to the third European Cooperative Acute Stroke Study (ECASS 3), which widened the window for the use of rtPA from 3 hours to 4.5 hours after symptom onset. ECASS 3 was published in 2008 in the New England Journal of Medicine, and was reported by Medscape Medical News at that time. Dr. Hacke was a principal investigator of that trial.

"Because we now showed 4 and a half hours is also safe and efficacious, we saw the numbers rose in the 3 hour and the 3 to 4 and a half hour time window," Dr. Hacke said. "Both went up after that with an age limit that is in the label but not in the physicians' brain."

He raised the point that rather than having a single primary endpoint, more trials would have positive outcomes if 3 or 4 main endpoints would be allowed, to look for outcome patterns. In the IST-3 subgroup analyses, the primary endpoint missed significance, but the secondary endpoint and the shift analysis achieved significance.

"Nevertheless, the statistical purists will say this is a negative trial," Dr. Hacke said. "This is not a negative trial. It is neutral for the primary endpoint, and overall, it has very convincing effects that all go in the same direction, and that is exactly like ECASS 1 and ECASS 2, also considered to be negative trials. It's this gambling about the primary endpoint that the statisticians put upon us."

Boehringer Ingelheim donated tPA and placebo for the first 300 patients but had no other part in the study. Dr. Lindley has received payment as a conference Scientific Committee member and for occasional lectures from Boehringer Ingelheim, and he has attended national stroke meetings that were organized and funded by Boehringer Ingelheim. Dr. Hacke, who was not involved in the study, has disclosed no relevant financial relationships.

8th World Stroke Congress (WCS). Abstract 328. Presented October 11, 2012.