Science: How many more will die?

Scientists attempting to estimate the numbers likely to die from the human version of `mad cow disease' have some important new clues that could help direct their research.

Thirty-five people have so far died of the human version of "mad cow" disease. The question everyone wants answered is: how many more are destined to suffer from the new variant form of Creutzfeldt-Jakob disease (nvCJD), the debilitating and invariably fatal brain disorder? Several teams of researchers are working on the problem, but it is proving to be one of the most difficult projects in predictive science.

There are at least four things we need to know before making even a reasonable guess at the likely size of the nvCJD epidemic. The first is an estimate of the total amount of beef products contaminated with bovine spongiform encephalopathy (BSE) that went into food destined for human beings during the Eighties, when the cattle epidemic was at is peak and before the government introduced the first control measures to protect human health.

The second is the incubation period in humans, between being exposed to BSE and showing the first symptoms of nvCJD. If the average incubation period is long, and evidence from other forms of spongiform encephalopathies indicates that it could be up to 20 years or more, then the relatively small numbers of people who have died so far could be just the beginning of an epidemic of "biblical proportions", as a senior government adviser once described the potential danger.

Next we need to understand the nature of the "species barrier" between man and cow. This indicates how difficult it is for an infective agent to pass from one species to another. If the species barrier is effective it could have substantially limited the transmission of BSE to humans, and the final death toll could be mercifully small.

Finally, scientists would like to know how people with different genetic constitutions react after they have eaten food contaminated with BSE. It is known, again from work on diseases similar to nvCJD, that whatever the agent is that causes a transmissible spongiform encephalopathy in human beings or other animals, its effects can be radically altered by the genes of its host organism. Some people infected with BSE may end up developing a disease quite unlike nvCJD purely because of their genes, or some people may be born with genes that confer resistance to BSE altogether.

A central problem for any scientist trying to estimate the likelihood of a future nvCJD epidemic is that none of these four pillars of knowledge has yet been built. There are only hints of they may look like, and even this information may be inaccurate when it comes to a disease that no doctor had heard of before 1996, when nvCJD was formally described for the first time in the medical literature.

Take the amount of BSE that people ate in the Eighties. The most likely bovine material to harbour the infective agent is brain and spinal cord, and it would be useful for scientists to know how much of this offal was offered for consumption.

Unfortunately, the only attempt so far to carry out a "food audit" of BSE material has produced data of limited value. The Leatherhead Food Research Agency, an industry-funded organisation commissioned to investigate the problem by the Government, estimated that 270,000 bovine brains a year were sold during the early Eighties. It also calculated that a tiny fraction of mechanically recovered meat - scraps from the carcass used in cheap meat products - contained spinal cord and brain.

"The problem with the Leatherhead research is that few of us quite believed it. It was not so much the conclusions, but how they came up with them," said one member of the government's Spongiform Encephalopathy Advisory Committee. As a result of this disquiet, the Government is now attempting to find more reliable information about the way the food industry used brains, spinal cord and bovine offal before it was banned for human consumption at the end of the Eighties.

Another of the four pillars of knowledge concerns the genetics of nvCJD. All the people who have so far died of the disease share a common genetic trait. Technically it is known as homozygous methionine; they have inherited versions of the prp gene (responsible for the protein associated with CJD) which produce double doses of the amino acid methionine at a site on the prp protein known as codon 129.

There are two other versions of the human prp gene that occur naturally in the population. In one, homozygous valine, the prp protein contains double doses of the amino acid valine at codon 129. In the other, the heterozygous version, the protein contains a mixture of valine and methionine. Normally there is nothing to distinguish these three different genetic "types" in the human population - they are all normally healthy. However, when someone is infected with BSE it is possible that the versions of the codon 129 on the prp gene which they have inherited may result in either a different incubation period, or a different clinical condition, or both.

Pioneering research by Alan Dickinson at the Neuropathogenesis Unit in Edinburgh established nearly 30 years ago the importance of genes in incubation periods for spongiform encephalopathies in animals. More recent work by Moira Bruce has defined the precise combination of genes in mice that result in differences in incubation periods for BSE. It is this work on animals that has led scientists to suspect that something similar might happen in humans.

Evidence from human research that this could be the case comes from a study of the victims of kuru, a CJD-like disease that affected New Guinea tribespeople who engaged in cannibalistic behaviour, who caught the disease by eating or handling contaminated brains of dead relatives.

Paul Brown and colleagues from the National Institutes of Health in Bethesda, Maryland, studied 92 frozen blood samples stored since the Fifties from kuru victims to compare the genetic make-up of the codon 129 site on the prp gene with clinical features of the disease, such as the estimated incubation period. They found that people who were homozygous at codon 129 developed the illness earlier than those victims who were heterozygous.

The scientists concluded that what they found may have implications for the future course of nvCJD in Britain. It may be that the reason we have seen only people who are methionine homozygous with the disease is that others - the heterozygotes or homozygous valine types - may have a longer incubation period.

"If nvCJD behaves like kuru, future cases (with longer incubation periods) may begin to occur in older individuals with heterozygous codon 129 genotypes and signal a maturing evolution of the nvCJD `epidemic'," the scientists concluded in their research report.

Another concern is that people who are heterozygous at codon 129 - about half the population - may, if they are infected with BSE, end up developing a disease quite unlike nvCJD. There is a precedent for this in a bizarre and extremely rare inherited condition known as fatal familial insomnia (FFI), a brain disorder in which people eventually die from lack of sleep.

It is now known that FFI occurs because of an inherited defect in the prp gene, this time at a site on the gene known as codon 178. But what is striking is that a defect at this site results in FFI only if patients are also homozygous for methionine at 129. If they have valine at codon 129, the patients who inherit the codon 178 defect develop a more classical form of CJD, with symptoms quite unlike those of FFI.

What worries scientists such as Bob Will, director of the National CJD Surveillance Unit in Edinburgh, is that BSE could cause a disease quite unlike nvCJD in people who are not methionine homozygotes. "We have no evidence that it will, but we cannot rule it out," he says.

With no way of accurately estimating the typical incubation period of nvCJD and almost no knowledge of the species barrier that may have protected humans against infection, scientists trying to estimate how many will die of human BSE are facing formidable problems.

The only real insight into the future course of the epidemic will come about when a reliable test is developed to detect people incubating nvCJD at an early stage. Today's announcement of a tonsil test for people with nvCJD could be the breakthrough scientists have been waiting for.