ALS EXPERTS EXPLORE DRUGS, STEM CELLS

by Margaret Wahl

TARRYTOWN, N.Y., June 13, 2003 – Some 100 ALS experts began an international discussion today to set plans for speeding human trials in amyotrophic lateral sclerosis.

At MDA’s ALS Clinical Trials: The Challenge of the Next Century, co-sponsored by the National Institutes of Health (NIH) and several pharmaceutical companies, the first of more than three dozen speakers reviewed progress to date in developing drug compounds or stem cell therapies to target ALS, among other topics.

MOLECULAR TARGETS
Serge Przedborski, co-director of the Eleanor and Lou Gehrig MDA/ALS Center at Columbia University in New York, discussed the need to zero in on specific molecular targets in deciding which compounds to test in ALS trials.

Przedborski, co-director of the conference, began the morning session by recapping progress since the 1993 discovery that a mutation in a gene called SOD1 leads to a highly toxic protein that kills motor neurons, the muscle-controlling nerve cells of the brain and spinal cord.

Shortly thereafter, it was found that mutations in the SOD1 gene are responsible for some 2 percent of human ALS cases. By the mid-1990s, researchers had developed a mouse and later a rat model of ALS by inserting genes with SOD1 mutations into these animals during their embryonic development.

Thanks in large part to the SOD1 rodent studies, it’s now established that motor neurons die in ALS because of a molecular cascade, a set of coordinated actions that results in the death of the cell. These actions, called “programmed cell death,” or “apoptosis,” can be divided into three phases, Przedborski reported.

“At the top of the cascade,” he noted, “we find the ‘initiators.’” These factors, which are both inside and outside the motor neurons, play the role of sensors. When they get signals that seem to warrant activating the apoptosis program, these sensors, which include compounds known as “prostaglandins,” initiate a death signal. Blocking such death signals can prevent the launch of the cell death program, studies have shown.

The current trial of celecoxib (Celebrex) in ALS is directed at blocking the prostaglandin known as PGE2 by interfering with another compound, COX-2, which is needed for PGE2 manufacture.

The next phase, Przedborski said, involves the cell death “regulators,” which include the Bcl-2 family of proteins. Some of these regulators help the cell survive, while others support the death program.

Genetic manipulations to increase the cell-survival protein Bcl-1 in mice with SOD1 mutations improved survival of the mice.

In the third phase of the cascade, cell death “executioners,” mainly a family of proteins known as “caspases,” become active.

In ALS, methods to block caspases could be beneficial, Przedborski said. These are being studied.

DRUG SCREENING
Jill Heemskerk, program director of Technology Development at the National Institute of Neurological Disorders and Stroke, part of NIH, spoke about her group’s aggressive drug screening effort to identify compounds that may slow neurodegeneration.

Heemskerk emphasized that out of every 1,000 compounds screened by today’s automated and sophisticated molecular methods, only 10 qualify as “hits” -- eligible to be moved to the next phase of research.

The “hits” then become “leads,” she said, and they’re next tested to compare their potency against their toxicity. Those in which potency exceeds toxicity actually will become drug candidates.

Tests of potency are based on whether the compound can alleviate signs of cell damage known to exist during the process of neurodegeneration. Some of the screens are based on decreasing signs of SOD1-related damage, while others are based on increasing the production of a substance that clears away excess glutamate, thought to be a factor in ALS and perhaps other neurodegenerative conditions.

Some 300 “hits” have so far been identified by the NINDS group, Heemskerk said. At least a few will likely go all the way to clinical trials.

STEM CELLS AND NEURORESTORATION
Robert Brown, director of the MDA/ALS Center at Massachusetts General Hospital in Boston, summarized progress in stem cell research as it may apply to repairs in the nervous system, as are needed in ALS.

He noted that he and other physicians were taught in medical school that there’s no regeneration in the nervous system after it has fully developed. Only recently this maxim has been proven false, he said.

Citing the work of Tom Jessell at Columbia Presbyterian Medical Center in New York, Brown noted with excitement that a gene called Bmi1 appears to allow the maintenance of a pool of stem cells, at least among cells that populate the blood. So far, the gene’s role in maintaining nervous system stem cells hasn’t been established, although it’s suspected to operate there in a similar way.

In rodent models of spinal cord injury, investigators have been able to embed stem cells into the gap between segments of healthy cord tissue where tissue has been damaged. Brown said these interventions have resulted in “enormous benefit to the cord” and at least moderate benefits to the animals’ walking ability.

In the “shiverer mouse,” which has a genetic inability to make myelin (a sheath around nerve fibers) and constantly shakes and shivers, stem cells delivered into the brain migrated throughout the brain, allowed for myelin production, and apparently caused the animals to have more normal movement patterns.

In ALS, Brown said, it may not be possible, at least in the near future, to replace motor neurons that have been lost. However, he noted that having “good neighbors” may help motor neurons that are on the edge to survive instead of dying. Stem cells implanted in the nervous system could provide those “good neighbor” cells, he said.

A few experiments using human fetal stem cells implanted into the spinal cords of rodents with SOD1 mutations have shown improved survival. In humans, umbilical cord stem cells may be the first candidates for such a strategy, Brown hinted.

OF MICE AND MEN
The leap from mouse or rat studies to studies of humans with ALS was the subject of a talk by Jeffrey Rothstein, director of the MDA/ALS Center at Johns Hopkins University in Baltimore.

Rothstein noted that many substances that have looked promising in mice haven’t lived up to that promise when tested in humans.

There could be many reasons for this, he said. Among them is that mice with SOD1 mutations are generated differently from the way humans with SOD1 mutations are born. Mice have two normal SOD1 genes and are given human mutant SOD1 genes, while humans with this genetic type of ALS have one normal SOD1 gene and one with a genetic flaw.

He also noted that the vast majority of people with ALS -- some 98 percent -- don’t have SOD1 mutations at all. Therefore, the model, while it looks very much like human ALS, may have some important differences.

Rothstein said that mice are usually given experimental ALS treatments before they even show symptoms, while humans with the disease aren’t identified so early and usually get these treatments when their disease is well under way.

Finally, he noted, mice may be given an experimental drug once a day because it’s too hard to give it to them more often, while humans may be asked to take the drug several times a day.

Humans and mice may also be given the drug by different routes; for instance, mice may get it by injection and humans by mouth. Before moving to human trials, Rothstein noted, it’s important to examine the mice carefully to see whether the drug entered the nervous system, and to try to keep conditions for the mice and humans as similar as possible with respect to the drug experiment.

Rothstein emphasized that the SOD1-mutated mouse isn’t the only way to study ALS. He says there’s still a use for some methods that were relied on before the mice were developed, such as sections of the nervous system -- “organotypic models” -- that can be examined to see the precise effects of a treatment on the nerve cells and surrounding tissues.

STATISTICS, REGULATIONS AND STRATEGIES FOR CLINICAL TRIALS
Experts from the NIH, the Food and Drug Administration and Columbia University’s Biostatistics Department reviewed several topics relating to conducting human trials: how to conduct a trial that’s large enough to demonstrate whether a drug is truly effective, how to ensure against errors of logic in selecting participants or analyzing data, and how to meet complex regulations set up to protect patients and at the same time advance research in neurodegenerative diseases.

LESSONS FROM OTHER FIELDS
Researchers who have conducted clinical trials in AIDS, cancer and heart disease suggested some parallels between the challenges their fields faced several years ago and those faced by ALS investigators today.

Chief among these is the search for biological indicators -- “markers” -- of disease progression or improvement (and their corollary, drug effectiveness) that are easily measured, objective and can be standardized. For instance, in AIDS, the level of HIV in the blood and the number of remaining CD4+ cells in the damaged immune system have been and continue to be powerful tools for drug testing and patient care.

The main method of assessing the progress of ALS is strength testing, which is highly variable from day to day and from investigator to investigator. It also can be influenced by factors that have nothing to do with the medication being studied.

Taking a leaf from the cancer researchers’ book was also suggested. Cancer progression is measured by a standardized staging system, and such a system could perhaps be developed for ALS.

Participants suggested that ALS investigators attempt to determine whether the current El Escorial criteria by which patients are judged to have possible, probable or definite ALS are still appropriate or whether new criteria need to be developed.