Endgame for Epilepsy?

Researchers look toward a cure

On screen, a baby waves her arms and gurgles happily, then tenses, scrunches her eyes shut, and clutches her tiny fingers together. Her arm jerks to one side, stiffens, and finally relaxes. She cries. Then the shaking starts again.

"I just wanted to hold her," says Susan Axelrod, her voice intense as she recalls the film shown at a recent conference on epilepsy. "I always held my daughter when she was having seizures."

Axelrod raised a child with uncontrollable epilepsy and felt frustration with doctors who saw partial control of seizures and serious drug side effects as acceptable outcomes of their treatments. Her experience led her to found an organization called Citizens United for Research in Epilepsy in September 1998. The Chicago-based group enlisted Hillary Rodham Clinton's help and focused congressional attention, and new funding, on research aimed at a cure for epilepsy.

Despite many decades of research, improvements in anticonvulsant drugs, and advances in surgical therapy, seizures can't be completely controlled in 20 to 25 percent of the 2.3 million people in the United States with epilepsy. For these patients with so-called intractable epilepsy, current treatments at best lessen the occurrence of seizures and at worst cause debilitating side effects while having little or no benefit.

However, recent advances in genetics, molecular biology, imaging techniques, and bioengineering have paved the way for attacks on epilepsy's cause as well as its symptoms, says Daniel H. Lowenstein, an epilepsy researcher at the University of California, San Francisco.

"In the past, cure has been seen as a distant objective," says Lowenstein. "We are finally making some real headway."

Epilepsy produces brief disruptions in the brain's electrical functions; seizures are the outward sign of these disturbances. Normally, brain cells communicate with each other through electrical impulses, working together to control body movements and keep organs functioning properly.

In an epileptic attack, hundreds or thousands of electrical impulses, each more intense than usual, appear in unusually synchronous bursts. Like ripples in a pond, such disturbances spread through-out the brain. These interruptions in the brain's normal pattern of activity may result in seizures and can affect consciousness, sensations, and movement.

There are many different types of epilepsy. Some develop in childhood, probably as the result of genetic mutations that cause abnormalities in brain wiring, an imbalance in the chemicals that the brain uses to send signals, or a combination of the two. Epilepsy can also develop when the brain makes abnormal nerve connections while attempting to repair itself after a traumatic head injury, stroke, or damage from chronic conditions such as Parkinson's disease.

The National Institutes of Health in Bethesda, Md., defines a cure for epilepsy as "the prevention of epilepsy before it occurs in people at risk, and the cessation of seizures without therapy-associated side effects in those who develop the disease." Lowenstein helped organize a conference at NIH in March that considered how to find such a cure.

NIH's funding for basic and applied research into epilepsy increases from $81.7 million in fiscal year 1999 to a proposed $93.8 million in FY 2001. As announced at the conference, the latter amount includes about $1 million for research specifically looking into cures for the disorder.

Researchers say understanding epilepsy may shed light on how the normal brain functions. Since many genes seem to be involved, the disease may also be a good target for scientists trying to understand how malfunctions within a complex array of genes can lead to disease.

"There's almost a palpable excitement about the growing opportunities to treat epilepsy," says Timothy A. Pedley of Columbia-Presbyterian Medical Center in New York. "We're working to make the science fiction of today the reality of tomorrow."

The reality of today is that there's no cure for epilepsy. Current drugs prevent seizures by shutting down overactive nerve cells. Some boost the effects of chemicals in those cells that inhibit nerve signaling, while others block channels that control the flow of ions into and out of cells and thereby regulate nerve signals.

These drugs may stop seizures, but they don't prevent the abnormal cellular processes that cause epilepsy. Nor can these drugs prevent the development of epilepsy in people known to be at high risk, such as children with particular brain defects or adults who have suffered severe head injuries.

However, the basic science that's giving researchers a better understanding of epilepsy may yield preventive treatments, says Lowenstein. In the past few years, scientists have managed to tease out some of the chemicals in the brain that cause nerve cells to grow in a particular direction and to make connections with neighboring cells. These processes may be abnormal in people with epilepsy. Understanding the molecules that modify the growth of nerve cells, therefore, may offer insights into both inherited and acquired forms of epilepsy.

Finding genes linked to epilepsy may be one of the best ways of discovering which of these molecular signals play roles in the development of epilepsy, says Jeffrey L. Noebels, a neurologist at Baylor College of Medicine in Houston. Epilepsy genetics is getting a big boost from studies in mice, he says.

To find out what a gene does, scientists frequently create strains of mice in which they disable, or knock out, a single gene. For a surprisingly wide range of knocked-out genes, "epilepsy or seizures have been a common result," Noebels says.

Several of these two dozen genes have turned out to encode ion channels, as many researchers had expected, he says. Others have unknown functions or--as is the case for the gene for a protease enzyme inhibitor called cystatin B--functions that don't seem to have any relationship to epilepsy.

These unexpected connections "may teach us the most," says Noebels. "The beginning of the end [of the disorder] is finding out what these genes do in the brain and what targets we have for developing better drugs."

Developing new drugs, though, takes a lot of time. After the discovery of a gene involved in a disease, researchers may work 10 to 12 years to create a drug that's effective at targeting that gene's action in the body, says Raymond Dingledine, a pharmacologist at Emory University in Atlanta.

Moreover, drugs that interfere with nerve cell activity may have broad consequences. For example, one approach to calming brain activity is to target the so-called NMDA receptor. This ion channel permits calcium to enter nerve cells and thus trigger electrical signals between neurons. Drugs targeting this receptor have had severe side effects, presumably because the NMDA receptor is widespread and plays roles in normal brain development, nerve signaling, learning, and memory.

An alternative strategy now being explored uses gene therapy to coax rats to develop antibodies to a specific part of the NMDA receptor. These antibodies may block the receptors when a seizure is occurring but not otherwise, thus limiting side effects, explains Matthew J. During of Jefferson Medical College at Thomas Jefferson University in Philadelphia.

He and his colleagues vaccinated rats with a virus that had been genetically engineered to produce a fragment of the NMDA receptor, and the rats developed antibodies to this protein. Normally, the so-called blood-brain barrier keeps antibodies produced outside the brain from seeping in. However, after trauma such as stroke or at the beginning of an epileptic seizure, the blood-brain barrier temporarily leaks.

Just two of the nine treated animals developed seizures when given a chemical called kainate, which normally induces seizures, During and his colleagues reported in the Feb. 25 SCIENCE. In contrast, 13 of 17 unvaccinated rats and 6 of 8 given gene therapy against an unrelated protein had severe seizures.

The researchers also reported no behavioral differences in the rats they studied, suggesting that the vaccination didn't harm other brain functions.

"We don't know exactly how it works," During admits. "I think we have an approach with a lot of potential to prevent the development of epilepsy," he says. He and his colleagues plan to study the effect of NMDA antibodies in dogs and, eventually, in people.

"This is a really interesting and unique approach," says Oswald Steward, a neuroscientist at the University of California, Irvine. However, transferring therapies from mice to people is never a sure thing, he says.

During's team can't say yet whether people will develop an immune response that will keep the injected antibodies from working. Nor can the researchers be sure vaccination won't cause an autoimmune disease or long-term damage to NMDA receptors in the central nervous system.

"Many of us feel that ... it will take a while to move into therapies that prevent epilepsy," says Dennis Spencer, a neurosurgeon at Yale University School of Medicine in New Haven, Conn. "In the meantime, we need treatments that are more effective with fewer side effects."

Physicians also need better ways of visualizing the brain's activity to understand where epileptic attacks begin, says Spencer. Before spreading to other parts of the brain, abnormal electrical signals arise in tiny regions called focal points.

In some people, surgical removal of these focal points can reduce or even eliminate seizures. Surgery is far from a general cure for epilepsy, however. Not all areas of the brain can be easily accessed, not all patients with epilepsy have well-defined focal points, and not all focal points are expendable.

New brain-imaging technologies are providing better pictures of epileptic focal points, says Brian Litt, a neurologist and biomedical engineer at the University of Pennsylvania in Philadelphia.

Scientists are also using the imaging techniques and implanted electrodes to carefully monitor the brain's electrical activity to predict when a seizure will occur (SN: 5/23/98, p. 326).

Taking advantage of faster, more powerful computer, Litt's team has monitored electrical activity in the brains of 17 people with a type of epilepsy called temporal lobe epilepsy. "Preliminary data suggest that seizures don't start as abrupt events," says Litt. Rather, the brain changes leading up to a seizure can start "perhaps days or hours ahead of time," he says.

At a conference in Atlanta last October, he presented data from three patients with temporal lobe epilepsy. His computer analysis of electrical activity could predict seizure 20 to 50 minutes before they occurred.

If these patterns hold true for people with other types of epilepsy, says Litt, it should be possible to develop an implanted device that can identify these abnormal electrical signals in plenty of time to abort the seizure by releasing small quantities of a drug or electrically stimulating focal points to ward off synchronized nerve impulses. Such an appliance would be a sort of pacemaker for the brain, say researchers.

Figuring out how to prevent epilepsy may hinge on understanding the reasons why the disease takes so long to appear in some people. Epileptic seizures may begin as long as five years after a head injury, for example. Researchers believe that nerve death and repair during this latent period may cause "a whole set of events that don't in and of themselves cause a seizure, but they predispose the brain to hyperexcitability and seizures," says Dingledine.

Since current antiseizure drugs don't seem to prevent epilepsy from developing, he suspects that the mechanisms that set the stage for seizures differ from the mechanisms that cause epileptic attacks. With the growth in animal models for epilepsy, tools are becoming available to use in searching for changes in the brain that bring about epilepsy, he says.

Using DNA chips (SN: 3/8/97, p. 144), researchers have tracked the activity of hundreds of genes in mouse brains during this latent period. They've found 29 genes expressed more strongly and 9 expressed more weakly in mice that later show epilepsy than in normal mice. However, the gene chips scanned only about 5 percent of the total mouse genome, leaving scientists many more genes to examine. Moreover, the function of most of these genes has yet to be unraveled.

In people, the interactions of genes and the environmental factors that cause epilepsy are likely to be even more complicated than they are in mice, says Samuel F. Berkovic of the University of Melbourne in Australia.

"Epilepsy is perhaps the best paradigm [to use] to understand diseases with complex causes," he says.

The complicated nature of epilepsy may, in part, explain why this disease--mentioned in ancient Babylonian writings more than 3,000 years ago--has seemed supernatural to some people. In many countries, the condition is surrounded with mystical beliefs and social taboos, which make it difficult for researchers to study.

"Of all the disorders [recognized] in the brain, epilepsy is one of the oldest," notes Noebels. Yet he cites the recent NIH conference as "the very first meeting with scientists talking about a cure." He adds, "The idea of curing epilepsy seemed too remote ... but now all of a sudden it seems like the right time to put things together."

It's none too soon for Axelrod and her daughter. "I'm sick of `living well with epilepsy,'" she says. "I don't want to live with epilepsy at all."