New NMDA Antagonist Increases Brain-Related Death Rate

Source: Davis SM, et al. Termination of acute stroke studies involving Selfotel treatment. Lancet 1997;349:32.

This terse statement in Lancet exemplifies the considerable health and financial risks of introducing new drug therapy for currently dangerous neurologic illnesses. Recent clinical trials of brain derived neurotrophic factor (BDNF) for treating amyotrophic lateral sclerosis (ALS) conducted by Amgen and Regeneron (NY Times, 1/19/97) turned out to have no measurable benefit, resulting in profound disappointment for the patients and major declines in stock values for the pharmaceutical houses. Also, large European trials measuring the potential value of rtPA in stroke failed only after studies of hundreds of patients showed that severe or fatal brain hemorrhages significantly increased the risk for treated vs. untreated patients. (That trial provided a treatment window of six hours post onset rather than the three-hour post-onset U.S. trial; even so, although the latter program provided a better cohort outcome, it also increased risk of fatal brain hemorrhage).

The Selfotel trial echoes the above problems. Acute stroke is a devastating disease with a high outcome of severe disability or death. Selfotel, a new competitive NMDA antagonist produced by Ciba-Geigy, Ltd., appeared to be useful when analyzed by a phase II study reported by Grotta et al (Stroke 1995;26:602-605) and Clark et al (Neurology 1994;44 [Supp 2]:A270). With this encouragement, two phase III trials were initiated, one in the United States and Israel, the other in four foreign countries. By December 1995, 328 patients had reached a three-month outcome, the results of which indicated an overall mortality rate of 24% among Selfotel treats and 19% in controls (non-significant). Nevertheless, 13% of Selfotel patients died of brain-related events (mainly increased stroke and brain edema) compared to 5% among controls (P = 0.003). Overall, three-month outcome of survivors was too little improved to justify the study’s continuation.


These good-intentioned but disappointing results deserve neurologists’ serious consideration. The immediate disappointment is that yet another failure has occurred in treating progressive, disabling neurologic diseases. Perhaps even more important, however, is that pharmaceutical companies gamble expensively in trying to bring to market new agents that may ameliorate, but not cure, progressive neurologic diseases. Severe setback in these efforts comes at a time when most of the world has its mind set on bringing health care costs down. This will undoubtedly inhibit both academic and pharmaceutical research. Clinical neuroscience, including psychiatry, has a number of more-or-less helpful drugs in its pharmacopoeia, but outside the anticonvulsants for seizures, lithium for manic-depressive illness, and immune globulins for inflammatory polyneuropathies, only a few agents bring total, long-term relief for neurologic disease. Compare, for example, the medically induced longevity that can follow first hints of serious heart disease, against the rather feeble protection that can be achieved against first contact with evolving or recurrent strokes. Somehow, a balance must be struck between the probability of good result and expense of trial. Unless such cooperation is developed, drugs that do work will be excessively expensive because of the need to counterbalance the very costly failures. —fp