John Collinge 3 hour briefing of science journalists
Strain types and prion glycosylation: commentary on Collinge briefing
Kuru: Collinge account disputed
Collinge position on mechanically separated meat questioned
Review of TSE genetic factors
CJD rates increasing in Austria

Leak: Cow to calf BSE risk is lower than thought
Science Published in newspapers -- a few minor problems

Sheep-brain rabies vaccine in India may cause CJD
New fears over cow milk and diabetes

Cow to calf BSE risk 'is lower than thought'

By David Brown, Agriculture Editor Daily Telegraph ... Monday 7 October 1996

The risk of cows passing BSE to their calves is lower than scientists feared, according to findings by Government scientists. They are expected to dismiss maternal transmission as a factor of minimal importance and to strengthen forecasts that the epidemic will have ended by 2001. Experts at the Central Veterinary Laboratory in Weybridge, Surrey, are keeping full details of the results confidential until the findings of a research project, which will end in November, are published.

Germany and other EU countries are pressing Britain to go ahead with the selective cull of up to 147,000 cattle, which was suspended by the Government, partly because some of these animals are offspring of BSE victims. The Government's Spongiform Encephalopathy Advisory Committee highlighted evidence this summer that BSE could be passed from cow to calf.

This was based on interim results of research co-ordinated by the Central Veterinary Laboratory. The work involved the study of 303 calves born from cows that fell victim to BSE and 303 others born to healthy cows. All calves in the study have to be slaughtered to enable their brains to be inspected for signs of the disease. The last batch of calves will be slaughtered in November.

Some calves have fallen victim to the deadly brain illness - but now scientists believe that the genetic make-up of the animals may have had more to do with their deaths than maternal transmission. A senior source close to the work said yesterday that individual animals may be more genetically susceptible to contracting BSE than others particularly if they were exposed to the food suspected of causing BSE in the first place

Science Published in Newspapers -- a few minor problems

Listserve Comment ... 7 Oct 96

Below is flaming example of the defectilve scientific publication system we actually have _instead_ of Roland Heynkes' sensible proposal for a prompt dissemination of scientific results even if in draft form.

The cow-to-calf 'findings' above is a deliberate leak from the Central Veterinary Laboratory. I would characterize it as in extremely poor scientific taste, done out of obvious economic and political motivations. No sign of an actual document written up or submitted, much less published, no author identified, no outside review, no data cited, no opportunity for opposing interpretations, no details on which gene even [if one was identified, rather than, say, a common bull]. Indeed, the experiment won't even be completed for a month. And the affected animals "may be" more genetically susceptible to contracting BSE than others -- not a strong claim even. Some of this could be the fault of the newspaper's account.

However, the story is certainly provocative. 'Maternal transmission' has always included maternal inheritance, as pointed out by Stuart Neilson. Perhaps cows with this genotype are able to transmit maternally through placenta. The study would seemingly only provide correlations, not causation. Herds may be severely inbred if artificially inseminated, hard to see how they could be all that different. While the leak seeks to reassure us, I wonder instead how prevalent that genetic susceptibility is in English herds. The last sentence in the article quotes the 'senior source close to the work' as implying that some animals may be contracting BSE even if they are _not_ exposed to the tainted food, a la sporadic CJD.

Strain types and prion glycosylation: commentary on Collinge briefing

Listserve 10.7.96

There is one passage that has some unintended potential for being confusing:
 "...shape of the PrP protein: Collinge said that the position of various
glycosolated sugars was a "good candidate" for
 the cause of different strains of various prion diseases."
I think what Collinge intended here is 'heterogeneity' of the overall glycoslylation end product. The glycosylation apparently always takes place at the asparagine 'positions' 181, 197 in the amino acid sequence. These regions have been highly invariant in eutheran and metatheran mammals for some 135,000,000 years. In fact, even the chicken has DCFNITV for N=181, only a V to F change.

The biochemistry issue here is that dozens of enzymes are needed to attach [and on turnover degrade] the highly complex branched and bulky carbohydrate moieties to the prion protein. These might reasonably be expressed at different levels in different brain regions of different genotypetic backgrounds for different neural functions at different ages in different species. No one knows how the basic glycoslyation end product varies across species, in either composition or linkages, never mind the subtleties. Few labs want to do nanomolar glycolyslation analysis.

This situation is further compounded in PrP mutants, polymorphisms, the resulting heteroligomers, and especially the transgenic animal. While the human prion gene is placed in these knockout mice, _none_ of human glycoslyation enzymes are, so mice glycosylation enzymes [which might not even be homologous to those of humans] are working across a species barrier. The end product is regretably chimeric: protein sequence human, CHO pattern more or less mouse.

Also worth mentioning are possibilities involving the extent and character of glycosylation in uptake, transmission, aggregation, and stability. The latter two involve interaction with heat shock chaperone protein and possibly RAGE (receptor for advanced glycation endproducts, new role in Alzheimer's).

On the subject of heat shock protein and prion aggregation equilibria, I call everyone's attention to an interesting paper in JBC [July 1996 pg 16856] in which NOS (nitric oxide synthase, maker of the well-known neurotransmitter) is completely and specifically knocked out in scrapie infection. The authors suggest an indirect effect, whereby malfunctional prion aggregate ties up hsp necessary for NOS to stay in a working configuration. This fits well with the emerging story of endgoenous prion disease in yeast. Obviously, when the prion is grown up in E coli for the Swiss 3D studies, glycosylation is completely missing [or dead wrong]. However we can be confident in the normals hosts that the sugars are fully exposed to solvent and contribute [perhaps ominously] to solubility.

Prusiner's group did some excellent studies on glycosylation and their position has gone back and forth on its significance (currently it's forth). They found an absolutely appalling mixture of glycosylations.

I myself am also of the opinion that heterogeneity of the overall glycoslylation end product is a "good candidate" for the cause of different strains of various prion diseases. One can imagine a prion from species X with such-and-such a gylcosylation infects species Y but can only convert host prion efficiently in this-or-that region of the brain, which produces new infectious prion with yet-another glycosylation that then is self-recognizing in later transmissions. In this sense, there may a memory of the origin of the infection retained in the brain regious exhibiting susceptible prion glycosylations. References {suggested by Roland Heynkes]:

Kocisko,D.A.; Priola,S.A.; Raymond,G.J.; Chesebro,B.; Lansbury,P.T.; Caughey,B.
Species-specificity in the cell-free conversion of prion protein to protease-resistant forms - a model for the scrapie species barrier - PNAS 1995 Apr 25; 92(9): 3923-7.

Lehmann,S.; Harris,D.A.
Effect of 2 pathogenic mutations on the cellular processing of the mouse prion protein
Journal of Cellular Biochemistry 1995; 1995(S21B): 104

John Collinge briefing of science journalists

7.10.96 Charles Arthur, The Independent

Last week Charles Arthur and other science writers from British national dailies had a three-hour conversation with Professor John Collinge, of St Mary's Medical School. He is a member of SEAC but probably better known as one of the British leaders in the field of prion diseases.

This is long (very, very long), but I think the information is interesting. One point: Professor Collinge was uncomfortable at the idea of getting a million requests for more information from people around the world about his work. Maybe readers could refrain unless he covers something that is very specific to their work from inundating him. (I leave this to the scientists to know where standard scientific enquiries tip over an edge.)

We covered a fair amount of old ground, but some new, certainly to me. I'll just report the interesting things as they arose, since it might be hard to organise otherwise. I've left out a lot of standard stuff about CJD (some is sporadic, some inherited) which would be in any standard text. Quotes are transcribed from my shorthand. Remarks in [square brackets] are my comments. I'll call the classic form of CJD "N-CJD" to distinguish it.

The main focus was on the transgenic mice which he is working on. These have human PrP genes. They are meth-meth homozygotes; by coincidence - rather, I think, than planning - this is the same gene pairing as all the British vctims of v-CJD.

Collinge emphasised that the SEAC scientists had worried when they identified v-CJD that it was not new, and that scientists from other countries would tell them so. The WHO meeting in the summer put them clear on that. "The meeting showed that nobody had seen it before. Once that worry [that v-CJD was not novel] was discounted then we really could feel it was different."

There is now a clinical case in a 50-year-old [to be confirmed after death; this extends the previously understood age range of v-CJD.] Collinge emphasised the strong consistency of pathology between cases of v-CJD, contrasting N-CJD which he said "varies a little bit".

He made the point about his 1991 paper showing that meth-val heterozygote humans are less - possibly not - susceptible to prion diseases. This has been shown in human growth hormone cases. But he said that in 1991 he had spoken to a vet: "Cows generally have no variations in that part of the gene [comparable to the meth-val codon variation in humans].. It looks as though all cows are equally susceptible. This is not true for sheep, mice and humans."

Mice. "It's very hard to transmit CJD [N-CJD] to ordinary mice. You can wait up to 18 months for it to occur. But if you put [N-CJD] into these transgenic mice, they all get sick." Though all the mice are presently meth-meth, he is also breeding val-val and meth-val strains. "We haven't tried oral transmission of [N-CJD] to these [meth-meth] mice."

We then discussed species barriers, and how you would show that a disease didn't pass. This quickly moved on.. "If you take ordinary mice and inject them with BSE, it kills them quite efficiently, they become sick from 350-700 days. If you then give their brains to a fresh set of mice, they become ill and die in about 140 days." The brains of the transgenic mice that were injected with BSE will be fed to another group of transgenic mice. "Maybe [the first group] haven't lived long enough [to show signs of the disease]." Brief discussion of the shape of the PrP protein: Collinge said that the position of various glycosolated sugars was a "good candidate" for the cause of different strains of various prion diseases. He noted that the 3-D structure of the mouse prion protein, published recently, doesn't include codons 106-126 "which we think are the most important". He added, "We're not far off having the structure of the human protein. We hope to have that by early next year."

Back to the BSE experiments on the transgenic mice. (It being a discussion with journalists, the questions leap around a lot.) "If the mice get sick next weeek, then it means that the species barrier is no better than between cows and mice. The other thing is that we have only injected a few dozen mice. But millions of the human population have been exposed to BSE.

"If you look at the [human] data and interpret it in the most optimistic way, then the barrier [between cows and humans] is at least as good as between cows and mice. But that's not such a good result."

[At this point we journalists started to sense a distinct lack of "feel-good" factor about what we were being told: the news, from Collinge's perspective, was not good and was probably going to turn bad. I had previously thought that he was fairly neutral in this debate, but this conversation showed me that he thinks that v-CJD *is* BSE in humans.]

"We have injected some v-CJD brain into transgenic mice. We've transmitted more than 20 cases of normal CJD [N-CJD] into transgenic mice. If v-CJD looks very different and resembles BSE [in the neuropathology] then we can draw the connection. "The Edinburgh Neuropathology Unit [Moira Bruce strain typing experiments] might not get a result at all because they're trying to do it across the species barrier." "We should get a result in the v-CJD transgenic mice fairly soon." He was reluctant at first to say how long ago they were injected. Eventually he said it was more than 200 days ago.

There was then a long discussion about possible treatments, and about the function of the PrP protein. I will omit this, as I don't think it sheds any light on the debate in hand. Apart from a couple of points of general interest, which I'll just quote. "PrP is normally turned over [destroyed] in the body in the course of a few hours." "If you increase production of the protein too far then you cause the disease." (Prusiner team experiments.) re treatments: there are two living v-CJD patients, but trying drugs out on them would not be ethical, he suggested. "The fact that they are going to die doesn't mean that you have the right to make their lives miserable."

Some discussion of kuru: the shortest incubation period was about 5 years, while cases are still occurring now - about 6-8 each year - with incubation periods of 40 years. Most occurred about 10-15 years after exposure at the feast.

BAck to BSE and v-CJD. "If it's BSE, then you have to add on the species barrier [to the time between exposure and clinical signs]. It looks as though this [barrier] is at least moderate. That can at leasts double it [incubation], so it might be 10 years before something happened. "If you suppose that v-CJD is BSE in humans: BSE appeared in 1985 and built up. Either way it's depressing. If v-CJD comes from the early part of the epidemic then that's 10 years ago. The bad news from that is that the number of cases of BSE built up to thousands a year in 1990. "Alternatively,if we're saying the few cases coming through are from an incubation of 5 years, that suggests there's a very small species barrier. "The level of exposure [of humans] has been so high. I wasn't expecting to see anything for 10 years." [Not clear if this is 10 years from BSE or 10 years from now.] He said that his concern was raised by the cases of v-CJD in two teenagers at the end of last year. "I rasied it at a very senior level that this could be the start of something." [He declined to say what level that was, or when exactly.] "When we sat down in 1990 and discussed this [possibility that BSE might transmit to humans] we said 'It will look like kuru'. People who received growth hormone treatments also got a kuru-like illness." [They were young, like the v-CJD victims.]

Here he said something which I think is very significant to a lot of the discussions that have been bouncing back and forth on BSE-L. I've capitalised it just to catch the eye of any drifters. It's worth considering the implications.


Were incubation periods shorter in younger kuru victims? "We suspect that there's age-specific sensitivity. I think it's quite likely that there's age-specific susceptibility. Kuru was a disease mainly of young children and young women. But it was women and children who were involved in the cannibalistic feasts and their preparation. The men ate muscle rather than brains. "The pathogenicity of the agent is extremely high in this situation. Everybody who sits down at one of those feasts dies of kuru." [The journalists were starting to feel spooked. Still, we asked, couldn't v-CJD be something else?]

"It's hard to believe that this disease [v-CJD] isn't due to a novel animal prion. First because it's a new strain. If you propose that it's sporadic CJD [N-CJD] then you have to assume that all these people separately developed this new strain spontaneously in a short time. "The probability against that is astronomical. [To explain it rationally] You simply have to look for a common source of exposure." On this [rather worrying] note he said that he thought the "BSE is scrapie in cattle" theory/hypothesis "was unfortunate" because it prevented people taking the idea that BSE could pass to humans seriously enough.

He pointed out that it was quickly shown that BSE could pass through food to cats - and suggested that many cases went unreported: those which were recorded tended to be near veterinary hospitals, which one would expect to be geared up to recognise the illness better than country vets.

Thus on the expected number of v-CJD cases we should expect: "I would be surprised if we were not looking at larger numbers. There were only 10 or so cases of BSE in its first year." The most likely culprit, he said, was mechanically-recovered meat because until last year this contained vertebrae - and those were not cleaned of spinal cord in an expert manner. Abbatoir workers use bandsaws, not scalpels.

MAFF (UK Ministry of Agriculture, Fisheries and Food) is trying to do an audit of which foods contained brain and spinal material in the past decade or so. It's not clear how much food manufacturers etc are cooperating. Spending more on BSE research earlier in the epidemic might not necessarily have produced more information. "I don't think you solve these things by throwing money at them. It's driven by a number of people who are intrinsically interested in the subject."

The questions jumped around some more. v-CJD might have some effect on the peripheral nervous system: "Patients have pains in their limbs." PrP also seems to be a "housekeeping" gene - that is, it's always switched on/expressed. "The level of expression doesn't change through the disease. There's no evidence that the [insoluble plaque] deposits cause problems." The cause of the "spongy" holes is still not clear either.

And while we do know a lot about CJD, we still don't know if it is triggered by a cumulative dose or a single incident. Normal people have absolutely no "prion" or "flipped" PrP. I asked about the cofactor hypothesis (that something else helps to flip the protein; that it's not just spontaneous.) Wouldn't this explain why the transgenic mice hadn't developed a TSE, whereas normal mice did? He said that the cofactor idea weasn't supported by his experiments, because you can pass N-CJD to normal and transgenic mice. The PrP gene doesn't seem to need a cofactor. [Either that, or it's a gene-expressed protein that humans and mice share? I still wonder.]

Finally: "It's depressing in that I think that v-CJD is BSE. It's hard to think what other explanation there might be. And we don't know if it's a cumulative dose which causes it. "On the good side, we know more about this than about any comparable brain disease, so we could get real therapies in the next 5-10 years - something that would stop it in its tracks. We'll have to wait and see about the epidemic in humans. "If we take the cat evidence at face value, then maybe the same will happen in humans. That would be the most optimistic outcome."

And on that jolly note, we said our goodbyes.

Collinge account of kuru disputed

Listserve 10.7.96

A comment on the mention of cannibalism, which has become deeply entrenched in this topic and which has little basis in recorded observation. This new version is certainly one of the most detailed I have seen:

"Kuru was a disease mainly of young children and young women. But it was women and children who were involved in the cannibalistic feasts and their preparation. The men ate muscle rather than brains. The pathogenicity of the agent is extremely high in this situation. Everybody who sits down at one of those feasts dies of kuru."
In reality, these feasts were not observed (by Gajdusek or any other visitor), but surmised. The nature of the feasts, lists of participants and other details were not collected. Despite this lack of any first hand evidence, these feasts are recorded in numerous scientific papers with a level of detail that increases with each repetition. They can not be relied on as accurate evidence of CJD risk factors.

What WAS observed was the ritual preparation of deceased tribal members. The skin, flesh and contents were removed from the skull with a range of stone and wooden (bamboo?) implements. By custom this was carried out by women and pre-pubescent boys. The gruesome black-and-white footage I have seen shows the process in some detail on what appears to be a disinterred corpse, not a fresh corpse, and in a manner that is not in the least reminiscent of food preparation - the object of attention was the skull, whilst the flesh and brain fell unheeded to the ground.

It is clear that women and children (both shown in the footage) were exposed to raw brain through lacerations to their hands, inhalation of spray and ingestion from their hands. It appeared to me that the flesh was re-interred and the bones taken elsewhere, as is common in Swat, Tibet, Venice or Greece (all of which I have also observed) and any number of other contemporary burial customs. Whilst it is possible that cannibilistic rituals took place, there is no hard evidence for it and none whatsoever of the "men ate the flesh and women at the brain" variety.

It is my belief that this is a misinterpetation of burial custom and nothing more. As this is not an area I have any particular expertise in, I will add that there is a growing body of literature that disputes the existence of any ritual human cannibalism, some of which details this particular example. Reports of cannibalism (viz. Ceausescu, Idi Amin, the Aztecs, Native Americans) tend to serve a political purpose.

The last sentence, "everybody who sits down at one of these feasts dies of kuru" is unsupportable. Rates of kuru were high, but it was not the leading cause of death. Precise epidemiological evidence linking kuru to particular forms of exposure is not available, but certainly indicates that the majority of those exposed to ritual preparation did NOT die from kuru.

Sheep-brain rabies vaccine in India may cause CJD

Sheep-brain rabies vaccines

Arya, S. C.
Vaccine 1991 9 1 70 India

The author recommends a mandatory histological screen for scrapie in a proportion of sheep (10-20%) in all flocks used to produce sheep-brain rabies vaccine to prevent the spread of viruses which can cause spongiform encephalopathies in India.

Aquisition of CJD in India through sheep-brain rabies vaccines

Arya, S. C. [Correspondence]. National Medical Journal of India 1991 4 6 311-312 Because of the widespread use of Semple-type rabies vaccines produced from sheep brains there is a possibility of the number of spongiform encephalopathy cases in man increasing due to transmission of the scrapie agent. The author suggests a number of measures that should be adopted to monitor prevalence of Creutzfeldt-Jakob disease in those vaccinated with the Semple-type vaccine.

Will, R. G. Prion disease. [Correspondence].
Lancet (British Edition) 1990 336 8711 369-370

A letter followed by five others on the same subject by H. Diringer, L.R. Bridges, H. Fraser, J. Hardy and A. Goate, and S.C. Arya. The correspondence by S.C. Arya suggests the cases of Creutzfeldt-Jakob disease in India may have resulted from scrapie in rabies vaccine made from sheep's brain.

New fears over link between cows' milk and child diabetes

Chris Mihill, Medical Correspondent

The Guardian ... Friday 4 October 1996

A controversial theory that cows' milk can trigger diabetes< in some children was given further weight today with the results of a laboratory study.

Insulin dependent diabetes develops when the body's immune system destroys the beta-cells that make insulin. One theory is that cows' milk can make children's immune systems destroy the insulin-producing cells.

Researchers from St Bartholomew's Hospital, London, and the University of Rome, say in the Lancet that people with insulin dependent diabetes seem to have immune systems that are primed to attack cows' milk protein, called beta-casein, which resembles a beta-cell.

In tests 51 per cent of those with diabetes had cells that reacted to beta-casein, but only 2.7 per cent of healthy people reacted to the milk. "The association between insulin dependent diabetes and early consumption of cows' milk may be explained by the generation of a specific immune response to beta-casein," the researchers conclude.

However, other researchers in the Lancet urged caution, saying the link is plausible but unproved. Leonard Harrison, from Melbourne University, Australia, points out that only a small proportion of people exposed to cows' milk in the first year of life develop diabetes. He said: "Much more research is needed."
Scrapie epidemic in Norway

Scrapie epidemic in Norway

Ulvund MJ. Department of Sheep and Goat Research, Norwegian colllege of Vet Med. Box 264, 4301 Sandnes, Norway. 

Presented at the 47th Annual Meeting of the European Assn for Animal Production, Lillehammer, Norway, 25-29 August. 1996
Scrapie was very rare, with the first case reported in 1981. There has been a pronounced increase in cases over the past 3 years, with 21 new flocks affected up to 29th August 1996 i.e. it is rising epidemically. The reason for this is unclear. Farms are emptied of sheep, the meat not used, the sheep slaughtered, and farms left empty for 2 years. Certain breeds are particularly affected and an affect has been seen due to the anxiety resulting from BSE in the UK. The rise is not felt to be purely due to BSE anxiety and the author things that something else has happened to cause the epidemic, so far unknown.

Collinge Follow-up

Listserve Commentaries 10.9.96

Someone else has mentioned the "kuru" issues, so I will mention the Collinge conjecture that mechanically-separated meat is the most infectious bovine food product, because it may have contained bits of spinal cord.

Note that MSM could not be responsible for the BSE epidemic, since cows are not fed this meat (it would require double rendering). Therefore the conjecture of infectivity would relate only to human food. This may seem obvious, but I want to make sure a witchhunt for MSM does not start as a result of this conjecture.

Also note that brains themselves would most likely be the most infective food product for either cows or humans. Although bovine MSM normally contains some bits of bone and marrow, spinal tissue would only be possible if MSM were prepared from vertebrae. However, poultry MSM is normally made from whole body skeletons and could contain such material routinely (although we're not worrying about chicken TSE's currently!).

I have seen and heard this "200,000 times" infectivity statement many times in the past, and have used it myself in talking to others. But please note there is NO scientific basis for this particular number. It could just as easily be as low as 1,000 times or as high as 1,000,000,000 times.

In order to quanitify the infectivity ratio of this size, an experiment would have had to be run on at least 200,000 animals to arrive at a score of "Oral: 1, Injected: 200,000". Obviously our knowledge is based on scores or hundreds of animals, not millions. Consequently the "200,000" is a "plausible" conjecture, based on the scientific evidence of an infectivity rate at least 100x higher in injection than in the oral route.

As a point of comparison, if the actual infectivity rate were an expectation of 1 case in 1443 tests, then, in an experiment involving 1000 animals, there would be a 50% chance of finding NO cases developed.

The TSEs ... a review of genetic factors

Lev Goldfarb and Paul Brown ... Ann. Rev. Med 1995 pages 57-65
Listserve Review 10.8.96

I came across this article yesterday by Lev Goldfarb and Paul Brown in the Ann. Rev. Med. 1995 pages 57-65. Being in a book, it is not indexed by Medline [though 35 other articles on prions by these two oauthors are returned].

It has a nice review of iatrogenic CJD, including some incidence rate data by country for growth hormone CJD. 58 cases, incidence 0.1% US, 0.7% UK, 1.9% France. 4 cases from infertility treatment with gonadotrophin, all in Australia. There is also a good table on genetic CJD, showing for each mutation, age of onset, duration, and clincal pathology [highly variable]. Most striking quotes: "In Central Slovakia, CJD is the fifth most common cause of death." [Recall that while E200K M129 is the genetic basis there, it seems of recent origin and is in a high scrapie area.] "Thus far, CJD transmission via blood or blood derivatives has not been proven, despite the fact that infectivity can be regularly shown in the blood of experimentally infected animals and occasionally in the blood of CJD patients." Reference given for this is Paul Brown. 1993 "Infectious cerebral amyloidosis: clinical spectrum, risks and remedies." Dev Biol. Stand. 80:91-101 [puzzling, Dev Biol. Stand is indexed for 1993, but this abstract is not there.]

CJD rates increase in Austria

The Lancet September 14, 1996

Worldwide the incidence of Creutzieldt-Jakob disease (CJD) is about 1 per million population per year. The rate in Austria was once much lower than this (average 0 18 per million for 1969-85) but in the last 10 years it has risen to one of the highest in Europe, 1 25 per million in 1995 (figure; g Neurol Neurosurg Psychiatry 1996; 61: 139-42).

Prof Herbert Budka and colleagues from the Institute of Neurology in Vienna attribute the increase to more effective case finding: "Austria is a small country with a high necropsy rate and well developed neurology and neuropathology services". There has been a similar increase in the incidence of CJD in the UK which has also been attributed to increased awareness of the illness, particularly in older people. None of the Austrian patients reported had the new variant CJD described earlier this year in the UK, although two women died unusually young (at the ages of 27 and 30) but without any of the other features associated with the new variant.