Going Gangbusters on Stroke

The field of acute ischemic stroke is highly promising but enormously risky and fraught with disappointment. With the exception of tPA, no one has ever demonstrated a pharmaco-mechanism which produces efficacy in the clinic. For now, the key to success for companies will be good efficacy and safety data. Start-ups will need partners, but even alliances with major pharmaceutical companies don’t represent a short-term validation of their efforts, since even big companies are stumbling in this field.

Research on stroke is at the same stage as heart attack was twenty years ago, with many drugs just starting to reach the market. In a field notorious for disappointment, companies have to ask themselves not just what looks promising, but why current attempts will succeed where past ones didn't.

By Wendy Diller

• The field of acute ischemic stroke is highly promising but enormously risky and fraught with disappointment. With the exception of TPA, no one has ever demonstrated a pharmaco-mechanism which produces efficacy in the clinic.

• The approval of TPA for treatment of acute ischemic stroke last June was an important milestone in neurology, underscoring the significance of timely treatment for stroke and providing hope to researchers that something could get through the FDA.

• Although big companies are focusing on stroke, there's plenty of room for start-ups, at least at the early stages. Two key areas of research are clot busters and neuroprotectives.

• For now, the key to success for companies will be good data regarding efficacy and safety. Start-ups will need partners, but even alliances with major pharmaceutical companies don't represent a short-term validation of their efforts, since even big companies are stumbling in this field.

The June 1996 FDA approval of Genentech Inc. 's clot-busting drug TPA (Activase) for the treatment of acute stroke was an important milestone in neurology. TPA was the first drug approved for stroke, and it came after a series of dramatic disappointments with other compounds, which resulted in the abandonment of high-profile efforts by several large pharmaceutical companies. In some cases, the failures didn't show up until fairly far along in the clinical trial process.

TPA, on the other hand, despite severe limitations, appears to have improved outcomes of a group of 634 carefully selected patients by as much as 30% after three months, according to a pivotal trial conducted by the National Institute of Neurological Disorders and Stroke (NINDS). The impact of its approval goes well beyond its ability to help a small number of patients, however. It serves as a model for a successful trial design in a notoriously difficult field, and more significantly, its approval is leading to a sea-change in the way the medical system thinks about stroke. Previously, doctors couldn't do much for acute stroke victims. Often, patients didn't get to hospitals until well after the attack occurred because no one realized they were having a stroke. And once in the hospital, they weren't treated with a sense of urgency because of the lack of effective therapies.

But TPA's short window of treatment—it must be given within three hours after a stroke occurs—gives stroke treatment a new sense of urgency. The patient has to be diagnosed, transported and differentiated by stroke type within that time frame in order to qualify for treatment. This includes giving the patient a CT scan to determine if the stroke is ischemic (caused by a clot) or hemorrhagic (caused by a broken blood vessel). TPA is contraindicated if used on the latter cases because it increases bleeding. Thus, only 1% of all stroke patients are currently getting TPA, experts estimate.

TPA's success has bolstered the confidence of researchers in this risky field. Its very weaknesses—the increased risk of bleeding, the short time frame for treatment—are motivating others to look for even better approaches to treating acute ischemic stroke. Efforts are focused on two approaches: busting clots in order to restore blood flow to the injured area and protecting neurons from further damage. Pharmaceutical companies are at the forefront of work in the field, but recently device researchers, frustrated by the slow pace of drug development and bolstered by the success of Target Therapeutics Inc. 's Guiglielmi Detachable Coil(GDC) system in preventing the rupture of brain aneurysms, are also turning their attention to stroke (Target is now a subsidiary of Boston Scientific Corp. [See Deal]).

"What got my attention was the results of the TPA trial," says Charles Hadley, PhD, a general partner at Hillman Medical Ventures, a venture capital group which has recently invested in several device start-ups focusing on stroke. "Prior to that, you were not only making a leap of faith that the technology would succeed, but even if it did, you had to make an additional leap that it would make a difference. What got me more enthusiastic about stroke was the revelation that you had a reasonable window of time in which if you could clear the clot you could save brain." Subsequent to TPA's approval, he says, Hillman invested in LATIS , a six month-old company, which is applying a laser catheter to bust clots in the brain.

Difficult to Diagnose

Stroke is a leading cause of death in the US, and the major cause of neurological disability in adults. Of the 500,000 people affected each year, it kills 150,000 and consumes more than $26 billion in direct medical costs for survivors. Following a stroke, 71% of victims can't work at full capacity, 31% need help caring for themselves and 20% need help walking.

Although stroke is devastating, the medical community, both clinicians and academics, don't give it the same attention as some other major diseases like coronary artery disease and cancer. That's because stroke doesn't always have obvious symptoms, the disease is highly complex and little understood, and the performance of experimental interventional therapies has been so disappointing.

Many people, physicians included, don't know the warning signs of stroke, or if they do, they don't understand the importance of getting prompt treatment. Strokes often occur when people are sleeping. Victims usually don't realize what has happened until they wake up. Moreover, in some cases, symptoms may be so moderate that people dismiss them, as often happens with transient ischemic attacks. "Strokes are harder to diagnose than gunshot wounds. Some are clear and some are subtle," says Paul M. Pepe, MD, MPH, former director of EMS for Houston and now professor and chairman of emergency medicine at the Allegheny University of the Health Sciences and director of emergency services at Allegheny General Hospital in Pittsburgh, PA. To add to the complexity, severity of symptoms doesn't always correlate with the patient's recovery. People who appear to have severe strokes may regain their functions fully, while others who are admitted into the hospital with moderate symptoms ultimately die.

Nevertheless, over the past few decades researchers have learned a lot about stroke and what causes brain cells to die, advances which are crucial to research on new treatments. More than 80% of strokes are ischemic, that is, they occur when a blockage stops or reduces blood flow to an area of the brain. Three main kinds of ischemia exist: thrombosis, which occurs when a clot forms in a cerebral blood vessel; embolism, which is caused by a clot or particle traveling from another part of the body to the brain or neck, and stenosis, which results from a severe narrowing of an artery.

Hemorrhagic strokes, which occur when a cerebral blood vessel breaks or ruptures, are less common: subarachnoid hemorrhagic strokes result when a blood vessel in the brain ruptures and bleeds between the brain and the skull, and cerebral hemorrhages occur when a defective artery in the brain bursts and floods the surrounding tissue with blood. The underlying cause in this case is often hypertension.

Most ischemic strokes affect a core area in which blood flow is so drastically reduced that cells can't recover and die. This seems to occur, according to the National Stroke Association, when cerebral blood flow is 20% of normal or less. In such cases, nerve cells are irreversibly damaged within a few minutes. Surrounding the core ischemic tissue, i.e., the area where the blood flow is cut off, is another area of the brain in which cerebral blood flow is reduced by only 20-50% of normal. In this area, called the "ischemic penumbra," cells are endangered but not yet irreversibly damaged.

The Benefits of Rapid Response

Researchers believe there are many mechanisms at work causing brain damage from stroke. The series of biochemical processes leading to cell death, known as the ischemic cascade, usually begins as blood flow is cut off with the loss of the ability of the cells to produce energy, particularly adenosine triphosphate (ATP). While brain cell death and injury occur rapidly within the core area following the attack, it takes place more slowly in the penumbra.

Brain cells respond to energy failure by elevating the concentration of intracellular calcium. At the same time, they release excessive amounts of glutamate, a neurotransmitter, which drives calcium levels even higher. This, in turn, stimulates chemical and electrical activities in receptors on other brain cells, which leads to further degradation and destruction. The brain cells ultimately die as a result of the actions of calcium activated proteases (enzymes which digest cell proteins), lipases (enzymes which digest cell membranes), and free radicals formed from the ischemic cascade. Just how vulnerable a nerve cell is depends on the extent of the cell's calcium overload and its ability to generate ATP as the energy source needed to kick out the excessive calcium ions.

One of the key findings in the past decade is that cells at risk don't necessarily die quickly, even though they may lose their ability to function. The total brain damage from a stroke can take hours or even days to reach its maximum effect. Intervention at different points in the cascade, therefore, could result in effective treatment.

Researchers are looking not only at clot-busting drugs like TPA and urokinase, which are thrombolytics, as well as clot-busting drugs with other mechanisms of action, but also at neuroprotectants. The former use a variety of mechanisms to break up the clot and normalize blood flow, presumably preventing an ischemic cascade or greatly reducing its impact. The latter interfere at various points in the cascade to prevent or reduce further damage to neurons. Among the most widely researched neuroprotectants are glutamate antagonists, which block the rush of calcium into cells following a stroke; calcium channel blockers, which work to stop the intracellular build up of calcium through electrically charged mechanisms; and calpain inhibitors. These prevent the release of the protease calpain, which when activated by the rush of calcium, breaks down other proteins. Another approach is kinase inhibitors, which prevent the release of various kinases, which block enzymes needed for ATP production.

Opportunity and Frustration

Attracted by new understanding and the huge unfulfilled opportunity, a host of companies, large and small, are working in the field. Interneuron Pharmaceuticals Inc. , Janssen Pharmaceutica NV , a division of Johnson & Johnson , and Knoll AG a division of BASF AG , are all in Phase III trials with stroke drugs, the former with neuroprotectants, citicoline and lubeluzole, respectively, and the latter with a clot-busting drug, Ancrod, which, as a defibrinogenating agent, rather than a thrombolytic, may not have the potential to cause inter-cranial bleeding. Abbott Laboratories Inc. is also in Phase III with prourokinase, a thrombolytic like TPA, which is delivered directly to the site of the clot via intra-arterial catheter. The assumption is that delivered that way, the drug can work faster and requires smaller doses than TPA, which is administered intravenously. Therefore, it can be administered later than TPA and may not have as high a risk of bleeding. These experimental products show some promise with reduced side effects, say their sponsors, although the field's top investigators don't expect any of them to be more than a moderate advance.

But if the opportunities in stroke seem golden, it is also a field fraught with pitfalls and frustration. Peptides, anti-inflammatory steroids, and sedatives to calm overexcited brain neurons and slow down its biochemical reactions, have been studied to no avail. Some of the most respected pharmaceutical companies, facing unexpected failure, have stopped work on stroke; among them are Upjohn, now Pharmacia & Upjohn Inc. [See Deal] likewise, recently stopped trials of sefotel, a neuroprotective glutamate antagonist, because of side effects and lack of efficacy.

Perhaps no area of stroke research has seen more heartbreak than glutamate antagonists, in particular those focused on N-methyl D-aspartate (NMDA) receptors. Glutamate antagonists work early on in the cascade, when glutamate binds to NMDA receptors causing them to open channels which allow calcium to enter the nerve cell. When ischemia occurs, the massive levels of glutamate can overwhelm the NMDA channels, forcing them into an open position and allowing a highly destructive calcium flood into the cell. NMDA channel antagonists theoretically prevent glutamate from binding to NMDA receptors, blocking the influx of calcium and reducing calcium-mediated cell death.

In the late 1980s, following the discovery of NMDA receptors, pharmaceutical companies' interest in the field exploded. Animal studies showed convincing evidence that NMDA antagonists were effective in reducing the extent of injury, even if they were administered after the stroke had begun. But evaluation in humans was much less straightforward:

MK-801 was the first major setback. Followed closely in the scientific community, it appeared to be highly specific and very effective in preventing destruction of brain tissue in animals following stroke. But the drug caused bizarre psychotic and other toxic side effects in people, forcing Merck to abandon it; Synthelabo SA (which was working on eliprodil) and Ciba Geigy gave up work on similar drugs for the same reasons.

Stroke Treatment in the Year 2005

Exhibit 4

SOURCE: National Stroke Association

Looking Harder at NMDA Antagonists

Still, because NMDA antagonists appear to be so effective, interest in them continues; roughly half the stroke drugs in development fall into this category, according to some experts. While small and large companies are active, some cutting edge research is coming from start-up companies, which are betting that they can find compounds that retain the benefits but overcome the side effects of first generation NMDA agents. Small companies are more willing to explore combination therapies or multi-mechanism drugs (which act at several different points in the ischemic cascade), compound therapies, and other non-traditional approaches than large companies.

Among those furthest along in trials with an NMDA receptor antagonist is Cambridge NeuroScience Inc. , whose compound, Cerestat, is in Phase III clinical studies; as a testament to the fickleness of this field, this June the company suspended enrollment in the trial after an analysis of data collected on patients aroused concerns about the drug's safety (current participants will remain in the trial, which is ongoing). Pharmos Corp. , likewise, is conducting clinical trials on HU-211, a multi-mechanism drug which acts on NMDA receptors and is also an anti-oxidant and anti-inflammatory agent. The drug is in Phase II trials in head trauma and in Phase I for stroke. Cognetix Inc. , a four-year-old Salt Lake City based company is developing peptide extracts from marine snails which act on NMDA receptors. NPS Pharmaceuticals Inc. , also, is close to filing an NDA for an NMDA-antagonist, NP 1506.

Indeed, start-ups are generally increasingly active in work on stroke, both for pharmaceuticals and more recently, devices. Committing resources to a field which has seen so many failures is risky, but small companies seem ready to make the leap. Most, however, are reluctant to undertake the project alone and either have or are looking for partners. The partners provide not only financial support but also help with clinical trials, which are notoriously complex and expensive in this field. Perhaps the two biggest risks faced by stroke start-ups are whether the product works and figuring out how to design a trial which proves the agent's value to the FDA.

A New Generation of Contenders

While TPA's approval gave stroke therapy development a boost, it also set a standard of care, which companies seeking to get their drugs through the FDA have to surpass. In addition to improved neurological function, companies are looking for results which demonstrate longer windows of treatment or help extend the time limit for TPA. They are also looking for therapies which have fewer side effects and don't require doctors to distinguish between ischemic and hemorrhagic stroke, thus saving time. In their quest, they are taking different tacks.

Pharmos, for example, sees opportunity in going where large companies fear to tread—developing a multi-mechanism drug. Almost all of the compounds in clinicals today have one mechanism of action, because pharmaceutical companies find multi-mechanism compounds require such complex data, says Gad Riesenfeld, PhD, the company's COO and president. But one mechanism alone can't block stroke or head injury, he argues. Pharmos' drug, HU-211 has three mechanisms of action, making it more effective and allowing for a longer window of treatment, since it affects different steps in the cascade. First, by blocking the NMDA channel near the beginning of the cascade, it inhibits the deadly calcium influx into neurons. Next, it has antioxidation activity, and third, it has anti-inflammatory properties. Moreover, it doesn't have hallucinogenic side effects, since it doesn't bind as tightly to NMDA receptors as previous drugs in its class, Riesenfeld says.

The drug is in Phase II trials for head injury and Phase I trials for stroke, for which it needs more elderly enrollees, says Anat Biegon, PhD, VP, R&D of Pharmos; the simultaneous clinical trials help Pharmos hedge its bets. The trial's six hour window of treatment (three hours longer than TPA)— is acceptable precisely because the drug's multi-mechanism features enable it to act at different points in the cascade. Pharmos expects the drug to go into Phase III trials next year for both head injury and stroke, enrolling as many as 1,000 patients for each indication, but it will need a partner, says Riesenfeld.

Cerestat, from Cambridge Neuroscience, is also an NMDA antagonist, but unlike many of the early ones which failed, it doesn't compete with glutamate by binding to the same sites as glutamate, says Robert McBurney, PhD, chief scientific officer and SVP, R&D, Cambridge Neuroscience. That is, it doesn't try to inhibit glutamate by binding to the same sites on a molecule, but rather binds to different sites on that molecule. The importance of this approach is that competitive NMDA antagonists have to fight with glutamate to bind and therefore require high dosing to make sure there is enough of them to overpower the glutamate. Cerestatand other non-competitive NMDA antagonists, on the other hand, can be dosed at lower, presumably less toxic levels. The drug's setback earlier this month, however, leaves lingering questions about its future.

Companies have different strategies for insulating themselves against risk, but ultimately, their data will drive success, at least at this early stage. No strategy can protect start-ups from poor results; witness the drop in Cambridge Neuroscience stock following its Cerestattrial findings. Cypros Pharmaceutical Corp. is trying to attack ischemia at its inception following injury, an approach that it believes is more comprehensive than interfering with any particular reaction during the cascade. The company's drug, CPC-211, or Ceresine, keeps ATP levels up and lactic acid levels down, theoretically at least, preventing ischemic damage, says the company's head of research, Anthony Fox, MD, PhD, the company's VP, drug development and regulatory affairs, and previously director of cardiovascular and anesthesiology clinical research at Glaxo Inc. (now part of Glaxo Wellcome PLC ).

But, like Cambridge Neuroscience, Cypros is hedging its bets by simultaneously conducting trials, one for head injury and one for stroke. The phase II data for head injury, involving 30 patients, looks positive and the phase II results for stroke should be ready this summer, Fox says. In head injury, CPC-211 lowered the level of brain lactic acid rapidly by up to 50%. The lactic acid which accumulates after a brain injury is a neurotoxin, which causes brain swelling and damage. The reduction lasted 24 hours, allowing for convenient, infrequent administration.

Trial design is paramount in Cypros' planning, and Fox believes this generation of researchers is learning from past mistakes. The clinical endpoints in stroke trials are getting standardized so that people know what to measure and what needs to be done to demonstrate the drug's efficacy, he explains. The use of a site management organization to coordinate the dozens of investigation centers involved in a Phase III study is invaluable, he says.

Mercury Therapeutics Inc. , on the other hand, is focusing on an entirely new mechanism, stress activated protein kinases (SAP), which were recently discovered by the company's co-founders. Formed in the summer of 1996, the company is developing inhibitors of SAP kinase, which mediates and induces many of the chronic inflammatory and immune activities set off by a stroke, including apoptosis, or cell death. A SAP inhibitor would work further down the cascade than NMDA antagonists, and would ameliorate the progressive tissue damage which follows stroke, says Neal Birnberg, PhD, CEO of the company. Blocking NMDA receptors may not be enough, given multiple mechanisms in cascade, he adds. Because it would work later in the cascade, an SAP kinase inhibitor would have a longer window of treatment, possibly hours or even days. Tests in animal models show that if SAP kinase is shut down, the viability of injured cells improves substantially.

Another company with a different approach is Medinox Inc. Its proprietary NOX compounds trap and neutralize nitric oxide (NO), a pivotal mediator of inflammation, which is overproduced following stroke and other health problems as well as surgical procedures, like transplantation, septic shock, and arthritis. Many companies have looked at NO enzyme inhibitors, but these generally eliminate all forms of the molecule. In doing so, they incur toxic side effects, since NO is an important vasodilator and intercellular messenger. The NOX compounds, on the other hand, are scavengers which seek to reduce, but not eliminate NO, allowing it to continue its critical housekeeping functions, says Ching-San Lai, PhD, founder and CEO of Medinox. In addition, they do not cross the blood brain barrier, thereby avoiding psychotic side effects of many drugs which get into the brain. Rather, the NOX compounds bind to NO molecules as they float by and inactivate them.

In the two years since its inception, Medinox has demonstrated in a rat stroke model, that its compound, NOX-151, reduced the size of infarct, or damaged area, by 30-40%, says Lai, a former professor of biophysics at the Biophysical Research Institute of the Medical College of Wisconsin. The company hopes to go into clinicals with NOX-151 next year.

NPS Pharmaceuticals is about to file an NDA for an NMDA antagonist, which in animal studies decreases the size of the damage to the brain by roughly 50%, says Doug Reed, MD, the company's VP, business development. Although the animal studies didn't continue long enough after the stroke to determine the impact of a reduction in the size of the damage, one can hypothesize that the outcome will be improved function, he says. Still, the industry has to grapple with an issue: getting neurological improvement data in animals is tough for everyone in the field, but using infarct size as a surrogate marker to show improved functional outcome isn't acceptable to the FDA. The company expects to start clinical trials within six months.

While the company isn't yet sure what window of treatment it will establish, its agent acts at a different site than Cerestat, notes Reed. For all companies now, he says, the hurdle is to show additional benefits over TPA within a three hour time frame.

Enter Device Proponents

As uncertainty about the ability of pharmaceuticals to make a significant dent in the way stroke is treated lingers, more researchers are taking a hard look at device-oriented approaches. Most neurosurgical techniques have attempted to refine coronary procedures but they've been ineffective and have high complication rates. Moreover, neurologists, who are primarily responsible for stroke patients, are more comfortable with pharmaceutical therapies.

Just as TPA had a big impact on acute ischemic stroke companies, however, the approval in September 1995 of Target Therapeutic's GDC coil for treatment of inoperable aneurysms was critical for advancing the use of devices in the brain. Aneurysms, which are a common cause of hemorrhagic stroke, were once the province of neurosurgeons, but the availability of the GDC coil has given rise to a new medical subspecialty, the interventional neuroradiologist. And having someone comfortable with devices and familiar with stroke has helped create momentum in the field.

Some venture capital firms which steer away from investing in pharmaceutically-oriented stroke companies, are willing to look at device start-ups. "We won't do pharmaceuticals because they are too risky," says Sigrid Van Bladel, PhD, a partner in New Enterprise Associates, which has three device investments in the stroke area. With devices, the "FDA process is shorter and there are more straightforward endpoints, like did you bust clot or not?," she says. "You can do initial trials in the peripheral arteries to get proof of principle, which is easier than with pharmaceuticals. You aren't messing as much with the clinical balance of the body. Thus, the risk is mechanical, i.e., can you do it, rather than what biochemical balance are you upsetting."

Device therapies also have advantages over pharmaceuticals, because once in place, they work almost instantly, unlike drugs, which may take several hours to break up clots, says Horst Adam, CEO of Endovasix Inc. , one of the companies NEA is backing. Thus, the time limit for when patients are eligible for treatment could be extended. And clot-busting devices, unlike drugs, don't carry a systemic risk of hemorrhage, he points out.

But while devices have been used to break up clots in the coronary and peripheral arteries, getting them into the tortuous, fragile arterial passages of the brain is a different matter, Adam says. The ability to get past the carotid artery, which leads from the neck into the brain, will distinguish those who succeed in this field, he believes. Because the cerebral arteries are not intertwined in tissue and are much thinner they don't have the resistance of coronary arteries and so are more fragile; thus, it makes sense to design a device specifically for them. EndoVasix is in animal trials with a small (350 microns in diameter) guide-wire type device which clears out occlusions in the brain and is compatible with existing catheters. The clot is dissolved through acoustic energy, but the device doesn't use vibration because that could damage the artery. The direct application of such a device to a clot might extend the eligible treatment window from three hours to six or more, Adam believes.

"We think we can overcome two most prominent failures of early development, size and flexibility," Adam notes. While initially the company conceived of the device as an adjunct to TPA, it now believes that the device breaks up the clots into small enough particles for it to stand on is own. The approval of some adjunctive neuroprotectants could be of benefit to it if they stretch the window of intervention from 6 to 12 hours.

The field of neuroprotectives is even more focused on pharmaceuticals than that of clot busting. But one company, Neuroperfusion Inc. , founded 1992, has a system for keeping brain cells alive post-stroke by maintaining a blood flow in the injured part of the brain. The system pumps oxygenated blood backwards from the femoral artery up into the brain through the venuous system, bypassing the area of the clot, and then recycles the blood—in effect, keeping the brain frozen to prevent further damage. This preserves the brain for further treatment, extending the window of opportunity. While the procedure has aroused much interest and has been publicized, it has, however, so far only been done in eight patients. The company and its researchers at the Medical Center at the University of California at Los Angeles (UCLA) just received FDA approval to expand the trial to 30, but UCLA investigators say they are having a hard time finding qualified patients due to the long time before stroke victims show up at hospitals and the tight criteria for eligible enrollees.

Results are extremely encouraging says William Worthen. President and CEO of Neuroperfusion. Patients have been kept on the system for 4-6 hours, potentially widening the window in which additional therapies can be administered. But the system doesn't break up the clot; once it is turned off, in a few cases, patients have gone on to have a stroke. The interventional neuroradiologists, who are likely to be at the forefront of using this device, are already familiar with the GDC coil, which is harder to use, says Worthen. Guidant Corp. recently invested $6.1 million in the company [See Deal], but doesn't receive exclusive rights to the company's technologies.

Getting the Message Out

Many experts believe small companies will find it virtually impossible to commercialize stroke drugs on their own, given the complexity of trials and lack of a well-established marketing infrastructure. Cambridge Neuroscience, for example, has a co-development agreement with Boehringer Ingelheim GMBH [See Deal], to jointly develop and market Cerestat for stroke and head injury. The agreement, signed in 1995, included a $10 million upfront equity investment by Boehringer Ingelheim in Cambridge Neuroscience, plus a series of milestone payments. Boehringer Ingelheim, which has worldwide rights to TPA outside of the US and a strong interest in central nervous system disorders, sponsored the TPA stroke trials in Europe and is organizing the Cerestat stroke trials. Cambridge Neuroscience is responsible for the head injury trials, which are still ongoing, despite the stroke setback. Pharmos is actively looking for partners, as are several other companies.

Cambridge Neuroscience is hardly alone in seeking partners to help commercialize their products. Start-ups are finding that, far from shying away from stroke, large companies are actively pursuing it and looking for licensing partners. A lot will depend on results. Even Boehringer's expertise didn't insulate Cambridge Neuroscience from disappointment and potential failure. Still, investor interest isn't flagging. "I don't think stroke is an area people are shying away from because of past difficulties," says Birnberg. "The pharmaceutical companies we interact with are mainly interested in stroke."

All of these companies, if they are successful will be pioneers of sorts. But their work is being made somewhat easier by the massive effort underway by Genentech and nonprofit groups to educate the public and medical community about stroke and change the way stroke patients are handled in hospitals. One of the findings of the NINDS study was that even patients in the control group fared better than typical patients because of the close attention they received in the hospital. Most hospitals have yet to set up stroke teams with neurologists, emergency room physicians, and possibly interventional neuroradiologists, but that is changing slowly.

As the medical community's approach to stroke changes, so will the kinds of marketing and educational efforts which product companies undertake. The challenge will be as more kinds of doctors get involved in treatment, how to address the different groups, some of which are engaged in turf wars to treat these patients. With the entry of devices, the lines between the physician specialties who treat stroke victims are blurring, says Jerry Williamson, director of marketing at Cambridge Neuroscience. The availability of effective devices and drugs for acute stroke will require neurologists to think differently than they have in the past, he adds. Neurology has been focused on the treatment of chronic, not acute, diseases, but that is changing. Now, neurologists are more available to the emergency room at all times than they have been in the past, and emergency room physicians are getting better versed in stroke.

What the Future Holds

The approval of TPA not only gave researchers hope but provided a practical model of how to do a stroke trial which will meet regulatory standards. Still, for all of its impact on the stroke world at large, TPA is likely to help only 1% of stroke victims, given its current limitations, says Hsu and others. Genentech is currently conducting a trial which increases the window of treatment from 3 to 5 hours, which may broaden TPA's impact in real terms, but its limitations leave the field open to new ideas.

Ultimately, researchers believe that compound therapy is likely to be the most appropriate approach to treatment. This could be TPA plus any number of neuroprotectives, assuming they get approval. The neuroprotectives, for example, might be used to help extend the time limits for administering the clot-busting drugs.

In fact, combination therapy may help show that a drug which isn't suitable on its own, either because of lack of efficacy or side effect, may be useful if taken with another drug or merged into one agent; a study published this May in Strokeof combined use of citicoline and Merck's MK-801 in rats, found that together the drugs had a significant effect on the size of the damaged area of the brain following a stroke even though each drug was given in sub-optimal doses. This is of interest as one is trying to reduce the unacceptable side effects of a potent NMDA antagonist. The problem is that it is hard enough in this field to do single drug trials successfully, let alone multi-drug trials.

In the short-term, several key Phase III trials are nearing completion and their results again impact approach to stroke. Disappointment at Cambridge Neuroscience lends a cautionary note, but hot on its heels are several other contenders. Interneuron Pharmaceuticals expects to publish results this summer of its second pivotal Phase III trial of citicoline for stroke and it's hard to know what to expect, despite long experience with the drug outside of the US. A first Phase III trial of 259 patients, who received placebo or one of three dose levels of the drug found patients who received the minimal dose showed significantly more neurological improvement than the control group, but those who received the mid-dose level didn't have a better outcome. The company believes this occurred because the patients in the mid-dose level group had more underlying pathology.

On one hand, because citicoline has been used in Japan and Europe for stroke for years, it is known to be relatively safe, says Takashi Kiyoizumi, MD, PhD, VP, business development and strategic planning at Interneuron. And because it has multiple mechanisms of action, which act at different points of the ischemic cascade, it doesn't have to be administered within three hours after a stroke. But some skeptics caution that the 24-hour treatment window permitted in its Phase III US trial is suspicious. The company says that if the drug is taken within 24 hours after a stroke, it limits the extent of the damage by preventing the accumulation of toxic free fatty acids. In animal models, the company says, it has been shown to significantly reduce the size of the damaged area. But, others note that if the 24-hour window proves to be acceptable, that will fly in the face of many current theories regarding the mechanisms of stroke.

Also completing Phase III trials are lubeluzole from Janssen Pharmaceuticals, which blocks sodium channels in nerve cells, and Knoll's Ancrod. While both may be of benefit to stroke victims, neither is expected to set the world on fire. Still, even if this latest round of drugs doesn't live up to expectations, the field will continue to make advances. "Right now we don't know how to get drugs to the right patients in time to help them," says Sidney Starkman, MD, one of the UCLA investigators working with Neuroperfusion. "Well, a drug is harmful in people. Eventually, we will figure out how to modify it so that it won't make a person sick. We will have drugs for treating stroke. In stroke we never knew we had any opportunity. There is no question that this field will be an enormous."

The Headache of Trial Design

Design of stroke trials is particularly troublesome, since scientists still don't understand the mechanisms behind the disease that well. Too many stroke drugs have failed as late as Phase II-III trials, due largely to the difficulty of designing and organizing trials in the field and the lack of good, predictable animal models. Researchers have trouble coming up with optimal dosing, for example, because they don't know how to practically measure biochemical markers in the brain. No one knows how long it takes for glutamate levels in stroke victims to peak and therefore how much glutamate antagonist to deliver. "In animals it peaks quickly and disappears," says Anat Biegon, PhD, VP, R&D, of Pharmos. "In people it's more difficult to get data. Some evidence exists that it takes longer to peak but you can't go into the brain to measure glutamate. In animals, it's a terminal experiment."

Many of the companies which have failed in their efforts to develop stroke therapies didn't have solid pre-clinical programs, says Dr. Chung Y. Hsu, MD, PhD, professor and head, cerebrovascular disease section, department of neurology, Washington University School of Medicine. Hsu, who consults with pharmaceutical companies on pre-clinical and clinical trial designs, says that while some, like Merck, were cautious, others jumped too quickly from animal into clinical studies. "Researchers in industry are under pressure to come up with compounds. They need to show productivity, and they don't go through what they ought to before spending $20 million on a clinical trial." The pressure to get good animal data starts early as small companies try to put their best foot forward as they look for partners, which is inevitable in this difficult field.

The problem is that animal studies are idealized and based upon homogenous populations. They are generally conducted in healthy young male rats, who all undergo the same procedure (females usually aren't used because their reproductive cycles add an extra variable). In humans, however, adequate recruitment is another matter because stroke is random, tough to predict and generally occurs in elderly people with underlying cardiovascular disease or other pathologies, observes Robert McBurney, PhD, chief scientific officer and SVP, research, of Cambridge Neuroscience.

Thus, the studies progress slowly and take a long time to complete. In past animal studies, researchers administered drugs quickly following the stroke, but in clinicals they delayed, giving those same compounds for as long as six hours. They needed the extra time to get more patients into their trials and assumed that such delays didn't matter since the human brain is larger and reactions take longer than in animals with small brains, says Biegon. The result was the animal trial results didn't correlate well the clinical trial data and all too often the clinical trial data came up short.

The importance of trial design—and the difficulties surrounding it—are exemplified by the TPA study, which involved 634 patients. The study enrolled enough patients to meet its highly exclusionary criteria only because the communities and hospitals in which it took place reorganized the way they approach stroke, so that patients could be diagnosed faster, get to the hospital faster and be identified as fitting the study protocol faster than in a typical setting, says Paul Pepe, MD, who was responsible for organizing the EMS in Houston, where one of the lead clinical investigators for TPA was based.

In contrast, a European study of TPA and several others of streptokinase, a competing thrombolytic, all failed, researchers speculate, perhaps, because the windows of treatment were six hours, which may be too long, researchers speculate. These studies found that the drugs didn't benefit stroke patients and in some cases increased the risk of hemorrhage.

Improving Animal Studies

Whether or not current contenders in the field are learning from past failures and TPA's success won't be known until the completion of some of the trials now underway. Whatever the answer, experts believe the learning curve will be slow. Many companies still aren't addressing pre-clinical trial design as carefully as they should, observes Hsu, although he wouldn't be specific. This adds an element of risk to their program, one which is unnecessary, given their already precarious standing.

The solution may be to challenge new compounds in a number of new kinds of animal models; although this is expensive and more complex, it ultimately reduces the chances of letting an unsuitable compound get to advanced clinical trials, says Hsu. Some companies are seeking to avoid the complexity of stroke in the short term by focusing their initial trials on head injury, then moving to stroke. Head injury trials are somewhat more straightforward because they more closely resemble animal models: the injury occurs at a known time and usually is the result of one impact. Most head injury victims, like most animal models, are healthy young males. Pharmos, for example, which has a drug for stroke in Phase I trials, is simultaneously conducting Phase II trials on the drug for head injury in order to alleviate some risk, says Biegon. Cypros Pharmaceuticals, likewise, is simultaneouslyconducting two Phase II trials for one of its two lead compounds, CP-211, for stroke and head injury. The aim, in part, is to spread the risk, says Anthony Fox, MD, the company's head of R&D.

The pitfall is that while head injury and stroke may share some pathological similarities, they are also likely to result in different biochemical reactions, which researchers have yet to distinguish. The question is just how much do these differences matter?

Therapeutic Approaches To Treating Stroke: Definition of Terms

Exhibit 2

Stroke Therapy Approach Therapy/Company

Clot-Dissolving Drugs

Thrombolytics:Help re-establish cerebral circulation by dissolving clots which obstruct blood flow. In June 1996 the clot buster Activase (TPA) became the first ischemic stroke treatment to receive FDA approval. •Tissue Plasminogen Activator (TPA) (Activase)/ Genentech: An enzyme found naturally in the body which converts plasminogen into the enzyme plasmin to dissolve a blood clot. Results of five-year trial conducted by the National Institute of Neurological Disorders and Stroke found that carefully selected stroke patients who received Activase within three hours of onset of stroke symptoms were at least 33% more likely than placebo patients to recover from their stroke with little or no disability after three months (NEJM, Vol. 333, No. 24, pp 1581-1587). Two limitations: short window of treatment and risk of hemorrhage. •Recombinant Prourokinase/Abbott Laboratories: Single-chain urokinase-type plasminogen activator appears to generate plasmin primarily at the site of the clot and does not tend to produce much free plasmin. This lessens the risk of generalized fibrinolysis and bleeding. Abbott is conducting a Phase III US trial involving the direct delivery of recombinant prourokinase via intra-cranial catheter to the site of the stroke.

Defibrinogenating Agents:These drugs work by reducing levels of fibrinogen in the blood to dissolve a clot. •Ancrod/Knoll Pharmaceuticals: Currently in Phase III clinical trials, Ancrod is derived from the venom of a Malayan pit viper. It is a proteolytic enzyme, which breaks down fibrinogen, a protein essential to clotting. It acts as a blood thinner by reducing blood viscosity and allowing better blood flow in the affected artery.

Antithrombotic Agents:Help prevent clots from forming. •Nadroparin Calcium (Fraxiparine)/Sanofi: Fraxiparine is a low molecular weight heparin. Both heparin and its LMW compounds are anticoagulants and are typically used to prevent deep vein thrombosis, characterized by clots that form in the extremities, which may lead to pulmonary embolism. Study indicates that treatment with Fraxipatine, within 48 hours after ischemic stroke improves outcome (reduces risk of death or dependency at 6 months).

Neuroprotectives Neuroprotective drugs work to reduce damage caused by the effects of the ischemic cascade following stroke. While no neuroprotective drugs are available commercially, several types are in clinical trials for treatment of acute ischemic stroke

Glutamate Antagonists:Glutamate is an important neurotransmitter, or chemical, which helps brain cells communicate with each other. During a stroke, excessive amounts of glutamate are released into the brain. The glutamate opens up the calcium channels in the nerve cells, allowing calcium to flood the cells and kill them. Glutamate attaches itself only to neurons with glutamate receptors. Drugs that interfere with the progression of glutamate into neurons are known as glutamate antagonists. •ACPC/Symphony Pharmaceuticals: While some neuroprotective agents block the NMDA receptor, others work on the receptor's binding sites. ACPC binds to the receptor and regulates ion flow. It is in Phase I of clinical trials. •Aptiganel HCI, Cerestat/Cambridge Neurosciences and Boehringer Ingelheim collaboration: Cerestat is a type of glutamate antagonist which works by blocking the NMDA ion channel. Glutamate and other neurotransmitters open or close ion channels on cell membranes. By blocking the NMDA ion channel, Cerestat may reduce the toxic effect of calcium and therefore minimize death of affected neurons. The drug is in Phase III clinical trials. •ACEA 1021/CoCensys and Novartis collaboration: An antagonist of the glycine site on the NMDA receptor complex. Glycine acts as a co-transmitter with glutamate. Blocking glycine would lessen glutamate's effect and help protect healthy cells. Phase II US clinical trial scheduled to start in 1997.

Calcium Antagonists:These attempt to interrupt the ischemic cascade and minimize cell death after a stroke. They block the intracellular build-up of calcium through electrical charges, possibly enhancing cerebral blood flow and slowing the entry of calcium into nerve cells.

Other Types of NeuroprotectiveAgents: •Enlimomab/Boehringer-Ingelheim Pharmaceuticals: Enlimomab acts on white blood cells, which flock to the blood clot, and add to the injury by further blocking blood flow. It reduces white blood cell adhesion and migration, minimizing ischemic injury. It just completed Phase III clinical trials. •Citicoline/Interneuron Pharmaceuticals: May have several neuroprotective qualities, including limiting the extent of the infarct by preventing the accumulation of toxic free fatty acids, promoting recovery of brain function by providing cytidine and choline needed in the formation of nerve cell membranes, and promoting the synthesis of acetylcholine, a neurotransmitter associated with cognitive function. A first Phase III trial demonstrated clinically significant neurological improvement in some stroke patients and a second trial is ongoing. •Lubeluzole/Janssen: Acts on glutamate, which is released into the brain after a stroke, rather than on a receptor site. It inhibits the glutamate-activated nitric oxide pathway. A European clinical trial showed no significant benefits but a North American trial indicated improved outcome. Currently in Phase III clinical trials.

SOURCE: National Stroke Association