Therapy

'IN THIS TOGETHER'

THERAPY

Preamble. A great deal of publicity has surrounded the recently announced therapeutic trial of the drug quinacrine. About a year ago, this drug was one of several compounds studied by scientists led by Dr. Byron Caughey at the Rocky Mountain Laboratories of the National Institutes of Health, who found that the drug strongly inhibited the accumulation of PrP in scrapie-infected tissue cultures, noted that it had been used as an anti-parasitic drug in humans, and suggested that it might have therapeutic promise in Creutzfeldt-Jakob disease (CJD). [Scrapie is a natural disease of sheep that is closely related to CJD, and often used as its experimental equivalent. PrP (‘prion protein’) is an altered form of a normal cellular protein that (apparently) causes both scrapie and CJD]. Dr. Stanley Prusiner’s group at the University of California San Francisco Medical Center recently confirmed Coughey’s tissue culture work and launched a campaign to initiate immediate clinical trials in patients with prion diseases.

Tissue culture experiments are extremely useful in screening potentially useful drugs. If a drug doesn’t work in culture, it almost never works in the much more complicated milieu of the body, and can be eliminated from further consideration. However, even if a drug does show promise in tissue culture experiments, it may still not be therapeutically useful because in the living body a drug is subject to the pharmacokinetics of absorption, distribution, inactivation or degradation, and excretion that do not exist under tissue culture conditions. Chloroquin (a drug that is chemically and pharmacologically similar to quinacrin) had no effect when tried several years ago in scrapie-infected mice, and preliminary results in new experiments in mice using both chloroquin and quinacrine are similarly unpromising.

Quinacrine has already been used in two patients with CJD: after several weeks, one is said to have improved, and the other has not. Further clinical trials are planned, and it is not difficult to imagine a mounting pressure to treat patients with different forms of CJD, individuals at increased risk of disease (mutation-positive members of TSE family, dura mater and growth hormone recipients), perhaps even the entire meat-eating population of Europe! If this short-cut approach to testing drugs on humans with an incurable disease turns out well, quinacrine will be the first useful drug in the treatment of CJD, and a remarkable achievement in translating rationally designed drug therapy from tissue cultures to humans. If it does not, then the extinction of hope not only among variant CJD families, but among all patients (and their families) with any form of CJD will have merely been the latest example of a premature human trial before adequate studies have been done in animals.

Early animal studies (Table 1)

Beginning around 1970, dozens of drugs and other pharmacologically active chemicals – either chosen at random, or for their effect on other kinds of infectious disease agents – were tested with a view to preventing or at least slowing the progression of disease in experimentally infected animals. In a typical experiment, groups of animals were begun on a drug at some point before, during, or after inoculation of the infectious agent, and then observed over a period of several months for signs of illness. When the drug was administered at the same time as the infection, some animals never became ill, or at least the incubation period was markedly prolonged. However, with increasing intervals between the times of drug administration and infection, the beneficial effects of the drug were found to diminish and then entirely disappear (Figure). In particular, no effect was ever observed once the illness had begun, and because we have no test to detect pre-symptomatic infection, drugs that only act before the onset of symptoms have no practical application in human disease. Moreover, because many of the drugs have toxic side effects, they cannot be given prophylactically over an entire lifetime.

Recent animal and in vitro studies (Table 2)

In 1985, following the discovery that PrP was closely associated with infectivity and could therefore be used as an index of infection, compounds began to be tested for their effect on PrP, either by blocking its entry into the brain (assuming it enters the body from the outside as happened, for example, in cases of iatrogenic and variant CJD, or by influencing its behavior within brain cells (which would be effective whatever the origin of the infection).

Most of these studies have been conducted in vitro, either in cultures of scrapie-infected cells, or in a cell-free test tube reaction mixture in which the normal precursor of PrP is converted into its abnormal amyloid polymer. Some of these studies reexamined drugs that had earlier been found to prolong incubation periods in treated animals, while others explored previously untested classes of drugs. Their modes of action are still not fully understood, and they appear to operate by different mechanisms: direct binding to the precursor, inhibition of precursor-to-polymer conversion, or interference with interactions of PrP with other cellular proteins. Several of the drugs share similar chemical structures that make them able to penetrate plasma cell membranes (where PrP is attached).

When considering the effects of these drugs, it is important to keep in mind the difference between measuring infectivity in treated cell cultures or brain extracts,
and measuring infectivity in treated animals. The first is like mixing together an acid and an antacid and then giving it to animals, the second is like treating animals with the antacid and then giving them the acid - a far more stringent (and practical) test of the drug.

With respect to blocking neuroinvasion of the infectious agent – and thereby preventing the disease, even if infection were to occur – it has been shown that the agent and/or PrP travels to the brain via at least two different pathways: one direct pathway via the vagus nerve, and one indirect pathway via the spleen, splanchnic nerve, and spinal cord. It is likely that when one route is blocked, the other is used, and it is possible that the bloodstream may also be involved. Therefore, any effective inhibition of neuroinvasion will need to block all of these potential routes. Several different compounds have been shown to prolong the incubation period in experimental rodents by inhibiting one or another pathway, but the strategy of using multiple drugs with different roadblocks to the routes of infection has not yet been attempted.

Even if effective against environmentally-acquired infection, such an approach would have no usefulness for the vast majority of patients who have or are incubating sporadic and familial forms of CJD, in which the disease process apparently arises within the brain itself. For these, the only logical approach will be to attack the infectious agent already inside the cells of the brain, and as difficult as this might seem, a great deal of progress has been made within the past few years in the study of drugs which show this kind of intracellular activity. Some have been tested in tissue cultures, others in animals, measuring either a reduction in the amount of ‘prion’ protein, or (in animals) a prolongation of the incubation period. None has resulted in a cure, but two drugs merit special mention: MS-8209, a polyene antibiotic related to the anti-fungal drug Amphotericin B, that prolongs the incubation period even when given to animals long after they have been experimentally infected; and iododoxyrubicin, an anthracine derivative that slows disease progression in experimentally-infected hamsters, and has already been used with some success in human patients with other forms of amyloidosis.

If efforts to negate the infectious process should prove fruitless, it may still be possible to treat the disease by interfering with the biochemical consequences of PrP polymerization that are thought to be directly responsible for the brain damage in TSE. Evidence from tissue culture and immunohistologic studies of brain tissue indicates that the normal precursor of PrP probably contributes to the repertoire of cellular responses to ‘oxidative stress’ from the free radical formation that is common to many neurodegenerative diseases. The biochemistry of oxidative stress is complicated and still incompletely understood, but inter-relationships between metal ions (especially copper), ‘excitatory’ amino acids, enzymes, and ion channel proteins all appear to be involved, and in some cases imbalances can be reversed by drugs. However, the principle of blocking the effects, rather than the causes, of TSE remains untested.

Human studies (Table 3)

Only a few drugs have been used in humans with CJD. Most are antiviral compounds used at time when the disease was thought most likely to be caused by an ‘unconventional virus’ , and a few were tried because they had been shown to prolong incubation periods in experimental animals (HPA-23 and amphotericin B). Early reports of successes with amantadine, vidarabine, and acyclovir were followed by disappointment when further patients showed no improvement, and the other antiviral drugs used in early studies of just one or two patients were similarly unsuccessful. The recently announced human trial of quincacrine does not, on the basis of animal experiments, appear any more promising than these earlier attempts to treat CJD.

Strategies for prevention

If a source of infection is environmental, it can be avoided and eliminated, but this always happens after the fact, and because the incubation period is so long, many people will have already become infected before the source is identified. One pre-emptive approach depends on progress in the field of genetic engineering. In mice, the gene that codes for the normal precursor of PrP appears to be redundant, and if made dysfunctional protects the mice from experimental infection with scrapie. If the human gene is similarly redundant, it should be possible to ablate or otherwise neutralize it and in consequence protect humans against CJD. This strategy would have obvious application to familial CJD, and would presumably work as well in all other forms of disease (although it may be objected that its universal application would be impractical as a preventive measure for a disease as rare as sporadic CJD).

Another possible approach would be the continuous prophylactic administration of one or more of the drugs that have been shown to prolong the incubation period in animal models. Because almost all such drugs are effective only at or very close to the time of infection, this approach would not be applicable to individuals whose infection had occurred in the past (e.g., human growth hormone or BSE infections), but might be useful for very recent suspected exposures to tissue grafts or surgical instruments when the likelihood of potential contamination is quickly discovered. Long-term prophylactic drug administration might also be considered in mutation-positive members of TSE families.

The ideal preventive strategy would be the development and use of a vaccine, but because PrP is a misshapen rather than chemically different form of a normal body protein, the immune system does not recognize it as ‘foreign’, and so would not be expected to produce antibodies to vaccines made with the protein. However, antibodies against short amino acid sequences of PrP have recently been reported to cure infected tissue culture cells, and mice immunized with very small amino acid fragments develop antibodies that react only with the abnormal form of the protein. Very recently, by means of some unusually clever genetic engineering, a healthy transgenic mouse has been created that produces antibodies to its own PrP precursor, and totally resists an infectious challenge with scrapie. The method is far too complicated to envisage its use in humans, but nevertheless proves the point that with appropriate genetic manipulations, an effective vaccine against PrP need not result in a generalized autoimmune reaction. There is a vast gap between work in tissue cultures or mice and work in humans, but these discoveries open doors to the possibility of creating a practical and potent vaccine against the TSE disease agents.

Legend for figure. Prolongation of the incubation period of experimental scrapie in rodents as a function of the interval between the start of treatment and the time of infection. Drugs shown: Dextran sulfate 500 (red triangles), HPA-23 (blue circles), and amphotericin B (green squares). No drug had any benefit when used in already symptomatic animals.