Reports

This report is based on medical evidence presented at sanctioned medical congress, from peer reviewed literature or opinion provided by a qualified healthcare practitioner. The consumption of the information contained within this report is intended for qualified Canadian healthcare practitioners only.

59th Annual Meeting of the American Academy of Neurology

Boston, Massachusetts / April 28-May 5, 2007

Intracerebral hemorrhage (ICH) is the most fatal and least treatable form of stroke. Mortality at one month following symptom onset exceeds 30%. The volume of hematoma is associated with mortality and poor functional outcomes following ICH. Early hematoma expansion is a cause of early neurologic deterioration in patients with ICH. Therefore, therapies that reduce bleeding and arrest hematoma expansion would be promising candidates to use in the emergency department to improve outcomes after ICH.

Human factor VIIa (FVIIa) is a plasma protein involved in the coagulation cascade following blood vessel injury. Recombinant factor VIIa (rFVIIa) has been used as a treatment for bleeding in patients with hemophilia who have antibodies to factor VIII or IX. Because it binds to locally activated platelets, rFVIIa initiates coagulation locally at the site of blood vessel injury.

A phase IIb study of rFVIIa yielded promising results in patients with ICH. This study randomized 399 patients with ICH who presented within three hours after onset of symptoms to placebo or one of three doses (40 µg/kg, 80 µg/kg or 160 µg/kg) given within one hour after the baseline computed tomography (CT) scan. “The findings caused tremendous excitement in the stroke community,” commented principal investigator Dr. Stephan A. Mayer, Department of Clinical Neurology and Neurosurgery, Columbia University, New York, New York. Hematoma growth was decreased significantly in the groups assigned to rFVIIa compared with placebo and although the study was not designed to detect differences in clinical outcomes, the rate of mortality and severe disability (as defined by the modified Rankin scale) was also significantly lower in patients randomized to active treatment. Serious arterial thromboembolic adverse events were more common in patients assigned to rFVIIa compared with placebo, although this difference did not reach statistical significance.

Momentum for Phase III Trial

The favourable results obtained with rFVIIa in the phase IIb trial prompted the initiation of FAST (Factor VIIa for Acute Hemorrhagic Stroke Treatment), a phase III study in which the primary outcome measure was death and disability on the modified Rankin scale at three months.

In FAST, which was conducted at 160 centres in 26 countries, 821 patients with ICH documented by CT scan within three hours of symptom onset were randomly assigned to either placebo or rFVIIa 20 or 80 µg/kg. Treatment was given within one hour after the baseline CT scan (no later than four hours after symptom onset). “We completely reproduced the hemostatic effect; the mean increase in ICH volume was 26% in placebo compared to 11% in patients randomized to the 80-µg/kg dosage of rFVIIa,” reported Dr. Mayer. “The absolute reduction in bleeding was even greater when treatment was initiated within three hours.”

Unlike the phase IIb study, however, the hemostatic benefit with rFVIIa did not translate into a clinical benefit, as there was no difference in achievement of the main outcome between patients randomized to placebo or active treatment. The proportion of patients who died or were severely disabled (score of 5 or 6 on the modified Rankin scale) was 24% of those assigned to placebo, 26% of those receiving 20 µg/kg and 29% of those randomized to 80 µg/kg.

The safety of the coagulant factor was also verified in FAST. Venous thromboembolic events (deep-vein thrombosis or pulmonary embolism) occurred in 6% of the placebo group compared with 5% of patients receiving 80 µg/kg, a non-significant difference. “Recombinant FVIIa did cause an absolute 5% increase in arterial thromboembolic events, which were split evenly between minor troponin leaks and somewhat more disabling cerebral infarction as detected on follow-up brain imaging within three days of dosing. Most of these infarcts were asymptomatic but some had major neurologic deficits,” Dr. Mayer told delegates.

Potential Randomization Imbalance

Several potential explanations can be offered for the discordant clinical findings in FAST and its predecessor phase IIb trial, Dr. Mayer suggested. A randomization imbalance resulted in an excess of patients with ventricular bleeding being assigned to the 80-µg/kg dosage compared with placebo (41% vs. 29%, respectively). “Intraventricular hemorrhage is a well-established determinant of poor outcome after ICH,” he noted. In addition, the proportion of patients with left ventricular hypertrophy was also greater in those randomized to active treatment compared with placebo.

Even though the median scores on the Glasgow Coma Scale were similar in all three groups (14 in the placebo group compared with 13 in the active treatment groups), 6% of patients assigned to 80 µg/kg of active treatment were in deep coma compared with only 3% of patients assigned to placebo. “These randomization imbalances indicated that sicker people were randomized to active treatment, which may have affected outcome,” offered Dr. Mayer. Other factors may also have contributed to the lack of an effect on the primary outcome, he added. “For some reason, the placebo group did extraordinarily badly in the previous trial,” he remarked. Mortality at 90 days among placebo recipients in the phase IIb trial was 29% compared with 19% in FAST.

Early mortality actually favoured rFVIIa in FAST (although not significantly), but an excess number of deaths between day 15 and day 90 in the patients randomized to active treatment resulted in the final non-significant effect on death. “Most of the deaths in the early weeks were neurologic deaths,” Dr. Mayer reported. “A little more than half of the deaths between day 15 and day 90 had nothing to do with direct neurologic problems but rather medical comorbidities such as infections, renal failure, gastrointestinal bleeding and cancer syndrome. For some reason, an overweighting of these relatively random medical complications occurred in the active treatment groups.”

Finally, liberal inclusion criteria may have resulted in a significant number of patients who had little chance of benefiting from treatment being enrolled in FAST. “When trying to measure a drug’s ability to minimize a brain lesion and promote recovery, you want people who are more resilient and have an easier time bouncing back up the Rankin scale,” Dr. Mayer said. “You want a population whose death and disability will be a result of the stroke, not random comorbidities.”

Consequently, it was reasoned, any future trials of rFVIIa, if conducted, should probably restrict enrolment to patients younger than 70 years with active bleeding but a reasonable hematoma volume, and narrow the time window of treatment to less than three hours from symptom onset. New imaging modalities such as CT angiography may select those patients with active bleeding who may benefit from treatment and enhance the benefit:risk ratio, Dr. Mayer added.

Possibilities for ICH

Despite the lack of a favourable effect on clinical outcomes in FAST, the coagulant factor remains the most promising treatment for ICH, for which no treatments are currently available, noted Dr. Mayer. “Unlike failed trials in neuroprotection, there is a clearly demonstrated biologic effect with rFVIIa, and the drug is capable of doing remarkable things as we saw in the phase IIb study.”