WHY RESEARCHERS LOOK BEYOND ANIMAL MODELS IN HUMAN TRIALS
San Antonio, TX — “There is a big argument as to whether animal models can replicate the human stroke with respect to treatment effect,” said Wolf-Deiter Heiss, MD, Professor of Neurology at Cologne University in Germany, at a scientific symposium here at the 27th Annual Meeting of the American Stroke Association (ASA) in February.
Dr. Heiss was referring to the fact that stroke investigators have seen experimental drugs work in lab rats, then fail in human trials.
Despite these challenges, he said, “We are all convinced that animal experiments are important, and that they are the only way to understand the pathophysiology of ischemic cell damage and stroke.”
Dr. Heiss suggested that investigators cannot let the success or failure of a trial depend on the outcomes of a rat study. They need to look beyond the rat study for different trial endpoints, he said.
Tissue plasminogen activator (tPA) trials were not successful in animal studies, but were in human trials, he pointed out, explaining that rats have lower concentrations of plasminogen activator in their blood than humans.
In another study presented at the ASA meeting, he said, citicoline was shown to reduce infarct volume by 50 percent and more in humans and in animals, but this did not translate into any clinical difference on the National Institutes of Health Stroke Scale (NIHSS), an 11-item scale used in practice and trials to assess neurologic outcomes and degree of recovery after stroke.
ROLE OF ANIMAL MODELS
So why then are animal models still important? Marc Fisher, MD, a staff neurologist at Belmont State Hospital in Worcester, MA, said animal modeling is needed, and will continue to be useful for more traditional purposes such as assessing safety and evaluating infarct size. For thrombolytic therapy, models can be used to evaluate perfusion effects. He added that animal models have been developed to allow investigators to assess functional outcome measures.
Another important consideration, he said, is that animal models can be used to assess and compare aspects of clinical trials between labs, between models, and even between species. Dr. Fisher added that this function has not been used as closely as it should.
Exciting new breakthroughs are coming along in the field of imaging that may show results outside functional scores like the NIHSS or a Rankin score. At this time, he said, diffusion/perfusion studies in animals have shown some value in terms of predictive response, but this has not yet been seen in human studies.
“We need to validate the diffu-sion/perfusion responses in clinical trials that we have seen in animals, and we really don't have that yet,” Dr. Fisher said. “We have considered adding MR angiography for thrombolytic drug development, and perhaps SPECT when that technology comes along.”
James Grotta, MD, Professor of Neurology and Director of the Stroke program at the University of Texas School of Medicine in Houston, said another problem with taking developments from bench to bedside has to do with study design. Investigators can induce or recreate a particular kind of stroke in rats and lab animals. In an animal model, they can, for example, design a trial to look only at cortical strokes or specifically exclude other kinds of stroke. Human studies have yet to become so precise. Consequently, a drug that tested well among rats with cortical infarcts will get tested in humans with several different kinds of infarcts and will not perform as well as it did in the laboratory.
In the future, MRIs and other forms of scanning might be used to “standardize” strokes in humans, so studies can be conducted among a more uniform type of stroke.
Also, to date, he said, combined strategies using reperfusion therapy to open up the blood vessels followed immediately by neuroprotectant drugs have yet to be tested in any large trials.
Dr. Grotta was one of the investigators in a study of lubeluzole, a glutamate and nitric oxide inhibitor. In the trial, patients were given tPA within two hours of stroke symptom onset, and then lubeluzole within an hour of that. The company, Janssen Pharmaceuticals of Piscataway, NJ, pulled the plug on the trial early before even half of the planned subjects had been enrolled because a simultaneous larger trial with lubeluzole did not show benefit when it was started up to eight hours after stroke onset (Stroke 2000; 31: 2543–2551; Cerebrovascular Disease 2001; 12: 258–263).
The problems might have been due to dosing, Dr. Grotta said. In rat studies with lubeluzole, the optimum blood level was 40 to 100 ng/mL, which was also tried in the human study. However, too many human subjects developed cardiovascular side effects such as arrhythmias at the highest dose, and when doses in the lowest range were used, there was no effect at all (Stroke 1996; 27: 76–81).
“It may well be that the failure of this drug had to do with the fact that we didn't get enough of this drug on board. And the side effects of our drugs might limit the doses, and prevent us from getting the doses up where we want them. We might not have been pushing the drugs up to the doses that we need in our clinical trials.”
Commenting on the ASA session by phone, Dr. Vladimir Hachinski, Professor of Neurology at the University of Western Ontario, in London, Canada, said there are inherent weaknesses in any animal study, simply because rat brains are vastly different from human brains.
“Rats are lissencephalic – they have no gyra. They have flat brains. You have to test a drug in animals that have gyri. So some experimental drugs have not been tested on the appropriate animal models.”
Dr. Hachinski continued: “There are probably nearly 120 drugs that have worked in labs, but not in people. We have all of these good things that work in rats, that don't work in a higher species. I'm afraid that unless we do something about it, companies will stop work on neuroprotectants.”
STUDIES EXAMINE CITICOLINEAS NEUROPROTECTIVE THERAPYFOR STROKE
Over the last few years, citicoline has been studied for its possible use as a neuroprotective therapy for stroke. Researchers believe citicoline may reduce central nervous system ischemic injury by stabilizing cell membranes and reducing free radical generation. Following are references for studies that have been conducted in the last two years.
- ♦ “Citicoline: Neuroprotective Mechanisms in Cerebral Ischemia,” Journal of Neurochemistry 2002; 80 (1): 12–23.
- ♦ “A Phase III Randomized Efficacy Trial of 2000 mg Citicoline in Acute Stroke Patients,” Neurology 2001; 57 (9): 1595–1602.
- ♦ “Effects of Citicoline on Phospholipid and Glutathione Levels in Transient Cerebral Ischemia,” Stroke 2001; 32 (10): 2376–2381.
- ♦ “Effect of Citicoline on Ischemic Lesions as Measured by Diffusion-Weighted Magnetic Resonance Imaging,” Annals of Neurology 2000; 48 (5): 713–722.
- ♦ “Evaluating the Efficacy of Citicoline in Embolic Ischemic Stroke in Rats: Neuroprotective Effects When Used Alone or in Combination with Urokinase,” Experimental Neurology 2000; 161 (2): 733–739.
- ♦ “A Randomized Efficacy Trial of Citicoline in Patients with Acute Ischemic Stroke,” Stroke 1999; 30 (12): 2592–2597.