GAIN Americas trial again shows no benefit from neuroprotectant agent in stroke

Susan Jeffrey

April 04, 2001

Wed, 04 Apr 2001 17:00:00

Chicago, IL - Results of a randomized trial of the neuroprotective agent gavestinel (Glaxo Wellcome Inc, Research Triangle Park, NC) in patients with acute stroke have shown no improvement in functional outcome with treatment. The findings add this to a long list of potentially neuroprotective agents that have shown no benefit when put to the test in clinical trials. The study is published this week in the April 4, 2001 issue of JAMA.

Description of the cascade of cellular metabolic events that follow brain injury in stroke had raised some hope of interrupting this process, limiting the ultimate size of the infarction and its impact on function. However, despite extremely encouraging results in animal models, a variety of agents have failed to show a benefit in randomized clinical trials, and some have even raised safety issues. In this study, gavestinel, an antagonist of the glycine site of the N-methyl-D-aspartate receptor did not improve functional outcomes at 3 months in patients with acute ischemic stroke.

Nevertheless, the researchers, led by principal investigator Dr Ralph L Sacco (Columbia University and the New York-Presbyterian Hospital) conclude, "We still believe neuroprotection remains a viable strategy for acute stroke treatment and should continue to be studied."


The results of the Glycine Antagonist in Neuroprotection (GAIN) Americas with 1367 randomized patients mirror those seen in the GAIN International trial, as previously reported in heartwire . In that trial, 1455 patients were randomized to receive gavestinel or placebo, within 6 hours of symptom onset. At 3 months, Barthel Index scores, measuring functional outcomes, were identical between the groups.

At that time, Dr Kennedy Lees (University of Glasgow, Glasgow, UK), principal investigator of GAIN International concluded, "We are satisfied we have a clear and unequivocal result; that the drug had no effect on stroke outcome in this trial, but that we had also run a trial that had given every opportunity for the drug to be shown to be positive if it was going to be." He added though, that this was only a single trial, and conclusions would await the GAIN Americas trial.

GAIN Americas included 1367 patients from 132 centers in the US and Canada, randomly assigned to receive either placebo, or gavestinel, given as an IV loading dose (800 mg) plus 5 maintenance doses (500 mg every 12 hours. The main outcome measure was functional capability at 3 months using the Barthel Index, with scores trichotomized as dead/0-55 (dead or dependent), 60-90 (assisted independence), or 95-100 (independent). Like the GAIN International trial, results between the treatment groups were virtually the same.

Main outcome measure: Barthel Index score at 3 months

Barthel Index score



290 (41%)
278 (42%)
141 (21%)
144 (22%)
270 (39%)
244 (37%)
To download table as a slide, click on slide logo below

No differences were seen in any secondary endpoint, nor did gavestinel appear to improve outcomes among the 333 patients who also received thrombolytic therapy. No safety issues were raised with treatment.

The researchers examined a variety of reasons that the trial may have failed to show a benefit of treatment, acknowledging for example, that no testing has been done to show that these agents actually get into the vicinity of the infarct core or ischemic penumbra. However, they point out that the trial had some "unique design features, and was well executed.

"This glycine antagonist joins the growing list of neuroprotectants that have not shown improved outcomes for patients with acute stroke, despite promising preclinical results," Sacco et al conclude.

Ischemic penumbra new target of interest

In an editorial accompanying the publication, Dr Fred Plum (Weill Medical College of Cornell University, New York-Presbyterian Hospital, New York, NY) discusses the evolving understanding of the importance of the ischemic penumbra that develops around the infarct core; ie tissue that is not actually dead, but retains some functional activity that might be recovered if blood flow were restored. The penumbra had been recently described with the use of newer imaging techniques, perfusion and diffusion-weighted MRI.

Reperfusion therapy is the first step in rescuing this tissue, Plum writes, but interventions such as intra-arterial thrombolysis and ancrod (Viprex? - BASF) have been reported to stretch the 3-hour time window for thrombolysis, perhaps to 6 hours or longer. Tissue pressure effects within the penumbra, caused as cells swell and burst in response to continued ischemia, pose a risk to the amount of tissue that might ultimately be salvageable, he notes. It may be that surgical decompression of the infarct could offer a bridge solution to reduce mortality and improve functional outcomes.

Whether neuroprotective agents can improve outcomes specifically by reducing or arresting damage within the penumbra, "remains to be seen," he writes. "The GAIN Americas trial adds to the already substantial number of apparently promising neuroprotective agents that have proven ineffective when evaluated in rigorous clinical studies," Plum added. "Increased understanding of the dynamic physiology and molecular aspects of the penumbra, along with identification of its potential capacities for improvement by using prompt and novel diagnostic methods, may prove useful for developing new and much-needed approaches for neuroprotection therapy in acute stroke."

Neuroprotectants: Why don't they work?

Ft Lauderdale, FL - The failure of neuroprotective agents to show a benefit was the topic of some debate at the recent American Heart Association 26th International Stroke Conference.

In one report, Dr Antoni Davalos (Hospital Doctor, Barcelona, Spain) and colleagues investigated the hypothesis that in many patients treated with neuroprotectants, the substances supposedly generated by the deterioration that were meant to be addressed by treatment may not even be present. They measured admission levels of glutamate, GABA, interleukin-6, and nitric oxide metabolites in the blood and spinal fluid of 258 patients with initial stroke, and found specific levels of these markers were predictive of early neurologic deterioration over the following 2 to 3 days.

However, these predictive levels were present in only 40% of the population they examined. In a trial setting, however, all would receive a neuroprotectant. "Probably the failure of neuroprotection is due to the fact that two thirds of the patients with an acute stroke do not show toxic levels of these substances in their fluids, the substances that neuroprotectants are trying to block," Davalos said.

In a separate paper, Drs David Matchar and Gregory Samsa (Center for Health Policy Research, Duke University Medical Center, Durham, NC) suggest the possibility that if neuroprotectants provide marginal but clinically relevant benefits, the trials to date may be underpowered to detect them.

Matchar and Samsa used computer simulations to explore this possibility, and found that even small overestimates of the efficacy of a given drug can result in substantial reductions in the statistical power of clinical studies. However, the methods by which trial designers estimate potential efficacy is not precise, leading in some cases to overly optimistic estimates, Samsa said. Perhaps "worse still," he added, even small benefits might have a large public health impact in the setting of stroke.

"I don't think researchers have entered into these trials, which by some measures may be underpowered, to find these modest benefits," Matchar said. "There are a lot of constraints on how research is done ._._. and looking for the 'big win' has been effectively the strategy taken in the neuroprotective field for the most part."

A better sense is needed in society and in the research community of what constitutes a minimally important improvement, and then trials should be designed on that basis, he added. "One of the implications of designing trials for identifying smaller benefits is that they are larger trials, and more expensive trials, trials that will challenge the research infrastructure."

In an interview with heartwire , Dr Philip Gorelick (Center for Stroke Research, Chicago, IL), like the GAIN Americas researchers, raised the issue of whether these agents actually reach the ischemic penumbra. The agents have been measured in spinal fluid and plasma, but it's not clear that they are reaching the tissue it is hoped they will preserve - given that no blood is reaching it either. Studies that label the drugs with a radioactive tag such as gadolinium, then use an imaging technique such as MRI to determine their deposition in the brain could technically be done, Gorelick said, but to date, they have not for reasons that are not clear.

Gorelick authored an editorial in a recent issue of the Lancet, exploring more fully the potential reasons for the continuing failure of these agents.


Related links

1. [Heartwire > News; Mar 30, 2001]

2. [Heartwire > News; Jun 2, 2000]

3. [Heartwire > News; Feb 11, 2000]


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: