Various classes of neuroprotective agents have been tested in humans, with some showing promising phase II results. However, with the exception of the National Institute of Neurological Disorders and Stroke (NINDS) rt-PA trial , none has been proven efficacious on the basis of a positive phase III trial. Notable failures include trials of the lipid peroxidation inhibitor tirilazad mesylate , the ICAM-1 antibody enlimomab , the calcium channel blocker nimodipine , the g-aminobutyric acid (GABA) agonist clomethiazole [118,119], the glutamate antagonist and sodium channel blocker lubelu-zole , the competitive NMDA antagonist selfotel , and several noncompetitive NMDA antagonists (dextrorphan, gavestinel, aptiganel and eliprodil) [122-124]. The high financial costs of these trials have raised questions about the commercial viability of continued neuroprotective drug development. How can we explain this apparent discrepancy between bench and bedside studies [125, 126]? The lack of efficacy can be related to several factors, some relating to the preclinical stage of drug development, and others to clinical trial design and methodology.
In the preclinical stage, therapies are often tested on healthy, young animals under rigorously controlled laboratory conditions, and, most often, the treatment is not adequately tested (for example, by multiple investigators in different stroke models) before it is brought to clinical trial. Whereas experimental animals are bred for genetic homogeneity, genetic differences and factors such as advanced age and co-morbidities (hypertension, diabetes) in patients may alter their therapeutic response. Moreover, despite similarities in the basic pathophysiology of stroke between species, there are important differences in brain structure, function, and vascular anatomy. The human brain is gyrated, has greater neuronal and glial densities, and is larger than the rodent brain. Some rodents (gerbils) lack a complete circle of Willis (gerbils), while others (rats) have highly effective collaterals between large cerebral vessels. As a result, there are important differences in the size, spatial distribution, and temporal evolution of the ischemic lesions between experimental models and humans. This is important, because the infarct volume is the standard outcome measure in animal models, whereas success in clinical trials is typically defined by clinical improvement. Finally, outcomes in animal models are usually assessed within days to weeks, whereas in humans, functional scores [National Institutes of Health Stroke Scale (NIHSS), Barthel index, etc.] are typically assessed after 3-6 months.
In the clinical trial stage, major problems include the relatively short therapeutic time window of most drugs; the difficulties in transporting patients quickly to the hospital; the imprecise correlation between symptom onset and the actual onset of cerebral ischemia; the high cost of enrolling patients for an adequately powered study; and the use of nonstan-dardized and relatively insensitive outcome measures. A recent review showed that of 88 stroke neuro protective trials, the mean sample size was only 186 patients, and the median time window for recent (1995-1999) neuroprotective trials was as late as 12 h
. Another major factor accounting for past failures is that patients with different stroke pathophys-iology and subtype are often combined in a trial, whereas the drug being tested might be more effective in a certain stroke subtype (e. g.,strokes with predominant gray matter involvement).
In addition to the above, delivery of the drug to target ischemic tissues poses unique challenges
. Pharmacokinetic properties of the drug, and alterations in cerebral blood flow after stroke need to be taken into account. Blood flow can drop to below 5-10% of normal levels in the infarct core, and to 30-40% of baseline in the surrounding penumbra
. In addition, the blood-brain barrier restricts direct exchange between the vascular compartment and the cerebral parenchyma, and post-stroke edema and raised intracranial pressure further impair efficient delivery. Strategies that have been explored to penetrate the blood-brain barrier include intracere-bral and intraventricular delivery, use of hyperosmolar substances (e.g., mannitol, arabinose) and pharmacological agents (bradykinin, mannitol, nitric oxide) to facilitate osmolar opening, and the development of carrier-mediated transport systems. These strategies appear promising; however, they remain limited by the prohibitively narrow time windows for effective stroke treatment.
Given these past failures, the focus has shifted towards expanding the therapeutic time window, improved patient selection, the use of brain imaging as a selection criterion, combination acute stroke drug treatments, use of validated rating scales to assess functional end points, and improved stroke trial design and organization [127,130]. A number of new neuroprotection trials are currently underway or in the planning stages. These include trials of the free radical spin trap agent NXY-059 (now in phase III trials), intravenous magnesium, the antioxidant ebse-len, the AMPA antagonist YM872, and the serotonin antagonist repinotan [131-133]. With the insights gained from prior neuroprotective trials, it is anticipated that one or more of the impending trials will prove successful.
Was this article helpful?