BOSTON — Despite high hopes for success, a phase III trial of the first medical therapy for intracerebral hemorrhage showed that the drug was no better than placebo in improving rates of death or disability 3 months after the bleed occurred.
Although recombinant activated factor VII (rFVIIa) reduced hematoma growth by up to 50% more than placebo, neither of the doses tested in the Recombinant Factor VII in Acute Intracerebral Hemorrhage Trial (FAST) improved rates of mortality or severe disability, Dr. Stephan Mayer said at the annual meeting of the American Academy of Neurology.
The results are an enormous disappointment to researchers and clinicians who felt that the drug's stellar phase IIB trial results showcased its potential to stop intracerebral bleeding and improve the dismal chances of patients who experience such an event. “In its phase IIB study, the drug created such a tremendous improvement that it almost seemed too good to be true,” with those in the active groups 38% less likely to die than those taking placebo, Dr. Mayer said. “And in fact, in this larger phase III trial, we found out that it was.”
The phase III FAST trial was conducted in 26 countries and randomized 841 patients to placebo or rFVIIa in doses of 20 or 80 mcg/kg. As in the earlier trial, patients had to receive the drugs no more than 4 hours after the onset of symptoms. The FAST demographics were fairly well matched, Dr. Mayer said, with two notable exceptions. There were significantly fewer intraventricular hemorrhages in the placebo group (29% vs. 35% in the 20-mcg and 41% in the 80-mcg groups). Left ventricular hypertrophy was also more prevalent in the treatment groups than in the placebo group, said Dr. Mayer, director of the neurologic intensive care unit at the Columbia-Presbyterian campus of the New York Presbyterian Hospital, and a principal investigator in the FAST study.
However, the mean time from symptom onset to treatment was identical in all groups (109 minutes), and almost all patients were treated within the 4-hour time frame (20% within 2 hours and 75% within 3 hours of symptom onset).
Safety was good, he said. The overall prevalence of thromboembolic events was no different among the groups (11% in placebo and the 20-mcg group, and 13% in the 80-mcg group). The frequency of venous thromboembolic events was similar (6% in placebo and 5% in each of the active groups), but arterial events were slightly more common in the 80-mcg group (10% vs. 6% in the 20-mcg group and 5% in the placebo group).
The increase in arterial events in the 80-mcg group was driven by a slight increase in myocardial ischemia and cerebral infarction during the first few days of dosing, Dr. Mayer said. “However, the total number of deaths among these patients was similar between groups, with two in the placebo group, four in the 20-mcg group, and five in the 80-mcg group.”
Thromboembolic events occurred slightly sooner in the 80-mcg group as well, he said.
The drug successfully decreased bleeding by 24 hours in the active groups. The placebo group had a mean increase of 26% in hematoma volume, whereas the increase was 18% in the 20-mcg group and 11% in the 80-mcg group. The absolute increase in hematoma volume by 24 hours was 7.6 mL in the placebo group, 4.7 mL in the 20-mcg group, and 3.8 mL in the 80-mcg group. “In terms of bleeding we essentially replicated our findings from the earlier trial,” Dr. Mayer said. “There was a highly statistically significant difference between placebo and the 80-mcg/kg group.” The study also determined that patients treated earlier experienced greater reductions in hematoma volume growth than those treated later.
However, the changes in bleeding did not translate into any significant improvements in clinical outcome. At 3 months, mortality was 19% in the placebo group, 18% in the 20-mcg group, and 21% in the 80-mcg group.
The combined end point of death or severe disability (a score of 5 or 6 on the Modified Rankin Scale) was not different among groups (24% in the placebo group, 26% in the 20-mcg group, and 29% in the 80-mcg group). “Actually, the 80-mcg group had slightly, but not statistically significant, higher odds of having a bad outcome than the placebo group did,” Dr. Mayer pointed out.
At the end of the trial, only 20% of the patients in each group were free from disability.
“In the first 2 weeks (of the phase III study), we did see greater mortality in the placebo group, but by the completion of the study, this effect was lost. The early deaths were directly tied to the neurologic event, including withdrawal of life support and brain death. After 15 days, the deaths we saw were more often related to comorbid illnesses, and they occurred mostly in our more elderly patients.”