Fear, uncertainty, doubt about sheep BSE
Prusiner study escalates sheep concerns
What is a "validated test" for TSE anyway?
She said, he said: is the Rubenstein western blot validated?
Who, what, where is Staten Island lab?
CJD epidemic waiting to happen
Vermont sheep get another week: courtroom scene confused
Sweetening the pot: government offers more money in Vermont
Other US BSE risks: the imported products picture
Former elite athletes could carry CJD
nvCJD suspected in woman's death
Jonathan Leake, Science & Environment Editor The Sunday Times 22 Jul 00 Tel 020 7782 5333 (Outside UK .. 44 20 7782 5333) Fax 020 7782 5731 Mbl 07957 420334 (Outside UK .. 44 7957 420334)A leading scientist has warned that BSE, so-called mad-cow disease is endemic in British sheep - and that people could be at a low but real risk from eating lamb.
New research by Professor Stanley Prusiner strongly suggests that the infective prion agent that causes BSE is found in sheep but at levels which have, until now, been undetectable. Prusiner, professor of neurology at the University of California in San Francisco, won the the Nobel prize for discovering prions. His laboratories are among the world leaders in such research.
Last week he said: "The implication of our latest work is that BSE is endemic throughout the British national sheep flock."
Prusiner's original work described how prions form when proteins that occur naturally in the brains of all mammals become deformed. The altered protein then acts as a template, changing other molecules in a chain reaction that devastates the brain. If an infected animal is eaten the prions start a similar chain reaction in the animal or person that ate it. However, although this mechanism has been well described the BSE prion has only been found in cattle, never in sheep - until now. Some British researchers argue that BSE prions arose through a genetic mutation in cattle and spread because material from dead cows was put in cattle feed.
Last week Prusiner said he and Professor Mike Scott, a Scottish researcher, appeared to have produced BSE in mice by infecting them with material from sheep suffering from scrapie, another prion disease. Prusiner, who has given up eating all sheep products because of his work, said: "Our initial data suggests that these sheep were producing more than one type of prion. One was the scrapie prion which killed them - but we believe at least some were also making the BSE prion."
The research uses "bovinised" mice in which the gene that makes prion proteins is replaced by the same gene from a cow. Such mice react to BSE prions just like a cow - but take less than 10 months to develop the disease, compared with more than three years for cows - so speeding up research. When such mice were innoculated with BSE prions the resulting disease was identical to that caused by variant CJD prions from humans - powerful evidence that the prions are the same.
By contrast, material from animals and humans with different prion diseases took longer and damaged the brain differently. The mice were then injected with material from sheep with scrapie - and again the incubation and symptoms were very close to those of BSE. This was powerful evidence that sheep can produce BSE prions. If they had died of scrapie the disease should have looked different.
Fred Cohen, professor of pharmacology at UCSF, said this was strong evidence that cattle got BSE because of changes in the way sheep carcases were rendered into animal feed. These changes involved cheaper solvents and lower temperatures - which Cohen believes still killed scrapie prions - but had no effect on the tougher BSE prions. In their latest work, said Cohen, he, Prusiner and Scott were replicating the changes in the rendering process in the laboratory - and injecting the resulting material into mice.
Cohen said that early results, so far unpublished, suggested that the theory was correct. Material subjected to the earlier, tougher regime, seemed not to produce disease.
"When scrapie-infected sheep were slaughtered the rendering process destroyed the scrapie prions but left behind the tougher BSE prions - to which cattle were vulnerable," he said. Such theories have been advanced before. Research by Moira Bruce at the Neuropathogenesis Unit in Edinburgh has confirmed that sheep can produce a range of prion particles but finding the one that causes BSE has eluded researchers until now.
If the results are confirmed - and many more experiments are needed to be sure - it will be a devastating blow to the British sheep industry. There is no evidence that people can catch BSE dirctly from eating sheep but most research has focussed on cattle so the possibility cannot be ruled out. Such a discovery would also devastate consumer confidence.
Opinion (webmaster): First, some technical clarifications are in order. The experiments did not exactly produce BSE in mice by infecting them with material from sheep suffering from scrapie. In transmitting sheep scrapie and BSE to knockout mice with a knockin bovine prion gene, the mice proved equally susceptible to sheep scrapie and BSE, with sheep scrapie actually producing the disease more rapidly than BSE. One explanation for this is that low levels of the BSE "strain" are actually endemic in the scrapie sheep but that it never surfaces as such because its presence is masked by the more rapidly growing sheep scrapie strain, a scenario calling to mind a similar situation with hyper and drowsy strains in mink [Bessen RA, Marsh RF. J Virol. 1992 Apr;66(4):2096; J Gen Virol. 1992 Feb;73 ( Pt 2):329]
The support for this: sheep scrapie can infect the "bovinized" mice very efficiently (no species barrier at all by comparison with BSE) and that sheep scrapie prions propagate in them much more rapidly than BSE. Hypothetically, an unusual selective treatment, such as an altered rendering process could remove the less dangerous sheep scrapie strain and allow the BSE strain to accumulate and spread to the cattle population.
In other words, it is said an oversimplification to think of scrapie, even within a single sheep brain, as a pure strain. Instead multiple strains propagate at different net efficiencies. The BSE strain has not been isolated as yet as the sole or dominant strain in any individual sheep scrapie sample; that search continues. The hypothesis is that the BSE strain may be endemic at a very low level in sheep and that its presence may be masked by conventional scrapie sheep prions that are present in much larger quantitities because they propagate more rapidly. This implies sheep scrapie may pose a threat to humans contrary to older epidemiological conclusions. <
This is a first, for Prusiner to show strong public leadership on a prion public health issue. Even though the data has not yet been written up for publication, Prusiner has to be taken seriously with 358 earlier prion papers. His team must be very excited by the implications of their research to make such a statement potentially affecting 40 million sheep, though nowhere do they suggest the data is strong enough yet to mandate a cull. As a medical doctor, Prusiner could not very well delay the announcement if further human exposure to BSE would occur during the months of nit-picking prior to print publication -- the UK has full access to the data already since Prusiner made the statement in England and one of the key team members is Scottish. Having the information, English authorities have to decide what to do with it.
Prusiner also submitted a sworn affidavit in the Vermont case, calling for carcass incineration to mitigate any possibility of risk from the imported sheep as well as their American-born progeny. Scrapie in the United States was first noticed in 1947 in descendants of imported British sheep. Since the argument is that a long-endemic strain of UK scrapie gave rise to BSE (back-transfer from cattle to sheep could have occurred more recently) , this raises the question of whether Prusiner also thinks American sheep are a risk. How would the consumer know which strain of scrapie they are getting? (Clinical animals are generally kept out of the marketplace but an animal may be infectious without showing signs.)
A paper due out shortly in EMBO Journal by Raymond, Bossers, Caughey, et al. supports some aspects of Prusiner's position. The in vitro conversion test shows scrapie, BSE, and CWD have low but non-zero risks of converting human prion to the abnormal form; this has been a reliable proxy in the past. While it isn't clear yet whether the same strain of scrapie mentioned by Prusiner was among those studied, taken together it would seem that some strains of scrapie might be of special concern in regards to direct sheep scrapie to human transmission. The new data is much stronger than, and trumps, older epidemiological questionaires about scrapie and CJD that the industry has relied on.
The morning of the Nobel Prize, Prusiner announced he had eaten a Porterhouse steak for breakfast -- at very time the English were banning meat on the bone and especially cuts like this containing dorsal root ganglia. Few people may care whether Prusiner eats lamb chops, tofu pate, or green salad but something has changed. Of course, the experiment itself, not individual assessments of dietary risks from eatin sheep, is the serious issue here.
Not everyone will be persuaded by the data or approach of the Prusiner group; "bovinized" mice have one cow gene interacting, or misinteracting, with 38,000 mouse genes. This may be our best model system but it is still a long way from infecting a real non-experimental cow with real sheep scrapie in a pasture situation or replicating long-discarded rendering conditons.
Note the experiment itself addresses forward infection, in a scenario where an original endemic form of scrapie infected cows through rendered feed. The English have mainly been fretting over backward infection, a scenario in which the cows started the BSE epidemic either through bad genetics or sporadic BSE and sheep subsequently became exposed to BSE (ie, endemic scrapie is held harmless).
No one can say at this time which sheep harbor scrapie, much less the dangerous scrapie strains; indeed, no one has any idea if these latter strains are common or even still exist in current British flocks. Are the British really going to cull 40 million sheep based on this? Are the US and Canada really going to cull an additional 5-8 million sheep based on Prusiner's work plus the EMBO journal article? The call will surely go out that "more studies are needed", but who will eat the sheep or cheese in the interim? In the mind of the consumer, there is already guilt by association. The choices are panicking the public, patronizing them with equally unwarranted assurances, or leaving things in a muddle with no recommendations.
There has been good progress lately on improved diagnostics that could identify TSEs earlier on, but few are validated and a diagnose-and-cull policy could not now make a dent in the TSE infectious load. Another approach is to forget eradication and concentrate on therapy for people and livestock exhibiting clinical effects. This is not a realistic option as no therapeutic compounds are currently in the pipeline.
Richard Lacey proposed long ago to remove all livestock from the British Isles, wait 10 years and start over. This was extremely unpopular; the respected professor, who had earlier blown the whistle on salmonella poisoning, was forced into retirement. The US came very close to this as well with its sheep flock a few years back -- total elimination with restocking from New Zeeland. It may well come to this in the end, as half-measures so far have not been effective and risks and questions remain about safety of the human food supply. While nominal CJD numbers in the US remains at low levels (there is no reportability requirement), the long incubation period makes this a very poor barometer of how many infected people are in the pipeline.
Government agencies, livestock producers, and consumers are left facing some very difficult decisions in a very incomplete informational environment.
Sat, 22 Jul 2000 Conference Reporter by Laura SpinneySurgical instruments contaminated with abnormal prions could be catalyzing the transmission of Creutzfeld-Jakob disease (CJD) and providing a perpetual source of infection, say Charles Weissmann and his colleagues from the MRC Prion Unit at Imperial College in London.
In 1977, the case of a 69-year-old woman - who had a steel electrode inserted into her brain as part of a treatment for epilepsy - came to light as a possible source of CJD infection. When her treatment was complete, the electrode was sterilized with a routine combination of benzene, ethanol and formaldehyde vapor for 48 hours and subsequently used to treat first a 23-year-old woman and then a 17-year-old boy.
The 23-year-old woman developed CJD 20 months after treatment and the boy succumbed to the disease after 16 months. The elderly woman also developed the disease, proving her to be the source of the infection.
Weissmann and his colleagues decided to try and reproduce this case experimentally, to see if the abnormal prions that cause CJD were somehow binding to the steel electrode. After exposing a steel wire to suspended extracts from scrapie-infected sheep brains for 16 hours and then washing it they exposed healthy mice to either the wash, the wire itself, or the wire following its sterilization with formaldehyde.
Mice exposed to the wash alone remained healthy. But those exposed to the steel wire developed disease symptoms after 72 days. A control group of mice that had the scrapie prions injected directly into their brains developed symptoms 65 days after exposure.
The incubation period can be used as a measure of infectivity, says Weissmann. The prions injected directly into the brains of the control group were slightly more infectious than those associated with the steel wire - but not by much.
When the steel wire sterilized with formaldehyde was used, the mice tested contracted the disease after 87 days. Even though the formaldehyde deformed the disease protein so that it could no longer be detected by silver staining, it was still able to cause symptoms in the mice.
When the researchers then looked at how long the wire had to be in contact with the scrapie agent to pass on the disease, they got a nasty shock. Five minutes exposure was enough to render it as infectious as 30 microliters of a 1% suspension of brain extract contaminated with scrapie.
It is possible, says Weissmann, that by binding to the steel the abnormal prion gains some resistance to the potent defensive mechanisms that the brain raises against infection. The question now is, does the wire bind the prions and then slowly release them to infect host cells as normal? Or are they irreversibly bound to the wire, and acting in situ? If the latter turns out to be true, that would suggest that the steel was somehow catalyzing the infectious process.
John Collinge, also of the MRC Prion Unit, comments that steel surgical instruments could be acting as a perpetual source of infection. And that is a huge concern for clinicians. It means that if those instruments are exposed to enough preclinical cases of CJD, they alone could give rise to an epidemic.
To ward off such an epidemic, the MRC Prion Unit is currently looking at recipes of enzymes and denaturants that could remove the prions from surgical instruments. An alternative might be to use disposable instruments. But, says Collinge, the best way would be to devise an efficient blood test for CJD that could detect the infection at an early stage. That work is also ongoing.
by Laura SpinneyJohn Collinge from the Medical Research Council Prion Unit at Imperial College in London, UK says that it is more likely that it was the cannibalistic practice of feeding cows to other cows that was at the root of the epidemic. Further, he believes that the reason the epidemic happened when it did probably had little to do with the rendering process and more to do with the number of infectious cases reaching a critical threshold....
Prion diseases are highly conserved across mammals, says Collinge. He argues that it is therefore likely that these diseases also occur sporadically in other animals and have done since time immemorial. We just don't see it because it is so rare and we only found it in cattle because we amplified the number of sporadic cases by feeding them to other cows. "We turned cows into cannibals," he says.
The scenario is similar to that of kuru, a CJD-like disease that spread due to the cannibalistic rituals of the Fore tribe of New Guinea. Collinge notes that in both kuru and BSE, a group from a single species contained in a relatively small geographical area gave rise to a "hothouse" of recycling of the disease agent - the abnormal prion - creating the ideal conditions for an epidemic.
"Cannibalism is bad news," says Collinge. In 1988, the feeding of MBM to sheep and cattle was banned in the UK. Since then, the feeding of chickens to chickens and pigs to pigs has also been stopped [in the UK that is, not in the US. -- webmaster]
In the meantime, however, BSE has crossed the species barrier to humans in the form of nvCJD, which differs clinically from classical CJD mainly in the age of onset. The latter affects elderly adults while nvCJD first shows itself in adolescence or the early 20s. To date 76 cases of nvCJD have been recorded in the UK, and for the first time the upward trend is statistically significant, Collinge says. Importantly, the number of cases of the disease is rising at a rate of 30% a year. "This is not going away," he warns.
Sat, 22 Jul 2000 Conference Reporter Live daily coverage from top conferences by Alison Jorda...[The rise in nvCJD] is very worrying, says Prusiner, as the number of new cases of vCJD continues to rise. Recently, a cluster of cases around a small English village was identified. But Prusiner believes that despite public pressure, as much effort should be expended on finding a means to achieve early diagnosis and treatment as on understanding its cause.
According to studies presented by Prusiner at the 18th International Congress of Biochemistry and Molecular Biology, both may become a reality within the next few years.
"We have developed a new assay, called a Conformation Dependent Immunoassay, which can measure even minute prion particles in the brain," says Prusiner. "This, coupled with techniques that allow us to identify the different prion strains, may eventually provide us with a very rapid diagnostic test for Creutzfeldt-Jakob disease." But the availability of a diagnostic test for prion diseases does nothing other than satisfy scientific curiosity unless treatments become available in tandem.
Not surprisingly, Prusiner and his team at the University of California have also been working on this, and he believes they have already identified at least three therapeutic approaches that look promising. [protein x, Compound 60, site-specific antibodies to inhibit prion replication].
22 Jul 00 by John Dillon Times Argus Staff United States District Court for the District of Vermont: no web site Previous ruling by Judge Murtha on bovine growth hormone labelling Judge Murtha previously against USDA's proposed Vermont clear-cut at Lamb Brook in 1995BURLINGTON ñ Two Vermont sheep flocks won a temporary stay of execution Friday as a federal judge ordered a hearing next week on a controversial test the government says shows the animals harbor a version of mad cow disease.
Lawyers for the sheep owners went to court to challenge the U.S. Department of Agriculture's order that the flocks be destroyed because of concerns they could spread the always-fatal brain disorder to U.S. livestock.
The lawyers challenged the validity of the test the government used to justify its July 14 order, saying the test cannot distinguish between mad cow disease or a far more common sheep ailment called scrapie. U.S. District Judge J. Garvan Murtha scheduled a hearing for late next week for the two sides to offer experts on the test and its results.
"It appears to me the plaintiffs are looking for an opportunity to demonstrate whether the government is overreacting," Murtha said. "The overreaction centers around the test."
Under last week's USDA order, the sheep were scheduled to be seized and destroyed by Friday. However, the government has now agreed to wait while Murtha considers the legal challenge.
At next week's hearing, flock owners Houghton Freeman of Stowe and Larry and Linda Faillace of East Warren will each be able to call an expert scientific witness. The hearing is tentatively scheduled for Thursday at U.S. District Court in Brattleboro.
Larry Faillace, who holds a doctorate in animal science, was elated after Friday's hearing. "At least we're given the opportunity to question the (test) results and have due process here."
But lawyers representing the USDA made it clear in court Friday that they believed Murtha did not have the power to overrule the government's finding that the sheep pose a dire threat to livestock. The USDA ordered the sheep destroyed after ruling that an "extraordinary emergency" exists that "constitutes a real danger to the national economy and a potential serious burden on interstate and foreign commerce."
Assistant U.S. Attorney Joseph Perella told Murtha that the USDA emergency declaration cannot be reviewed by a federal court. Congress gave the USDA broad authority to seize and destroy livestock in emergency cases and did not mean to hobble the government when it must act quickly, he argued. "The reason judicial review is inappropriate is the very nature of the .. emergency."
But lawyers for the flock owners said the government lacked sufficient evidence to show the animals are sick. They questioned the test the USDA used to determine if the sheep had a form of transmissible spongiform encephalopathy ñ a family of degenerative, incurable illnesses that includes mad cow disease.
A New York lab hired by the government used a version of a test that has not been reviewed by other scientists, said Thomas Higgins, Freeman's lawyer. "We're advised by the USDA that this is not a regular tests, that it was enhanced somehow to give heightened sensitivity," he said. "It's suspect scientifically and would be more likely to give false positives."
The USDA contends the test was scientifically valid. USDA officials have said the sheep or their ancestors may have contracted mad cow disease from feed before they were exported from Europe. An outbreak of mad cow disease in Great Britain in 1995 devastated that nation's beef industry and killed about 50 people.
The USDA has monitored the Vermont flocks for years and last week said tests on brain tissue taken from four sheep culled from Freeman's flock showed the animals had some form of transmissible spongiform encephalopathy.
"We know this is a fatal disease and that it may have catastrophic results," said Assistant U.S. Attorney Melissa Ranaldo.
Freeman and the Faillaces counter there's no evidence of the mad cow disease passing to sheep. They say the Vermont sheep are healthy and show no sign of the degenerative brain disease.
John Buckley, the Faillaces' attorney, argued that destroying the sheep would ruin the business they had started. "They are irreplaceable," Buckley said of the East Friesian stock. "There are no comparable animals that can be obtained."
Freeman and the Faillaces imported the animals because of their high milk production. Both of the farmers have been making cheese from the milk, but the Vermont Health Department has advised consumers not to eat it because of the latest tests.
There are 376 sheep that the USDA wants to destroy, although a farmer in Lyndonville has voluntarily agreed to sell his flock to the government so they can be destroyed.
Ranaldo said the court should not delay the USDA seizure. "The longer the sheep are out in the pastures, there's the possibility that one of the sheep could escape" and introduce the disease broadly, she said.
Murtha also said he believed the case should be settled quickly. He cautioned lawyers for Freeman and the Faillaces that he would not hold a "de novo" hearing ñ meaning he would not review the entire USDA decision to seize and destroy the animals ñ but would limit next week's session to evidence on the diagnostic test.
"It seems to me the court has an obligation to move forward with this case," he said.
21 Jul 00 Valley Reporter by Cory Hatch (reprinted with permission)The science behind the US Department of Agriculture is attempts to seize and destroy the Faillace sheep flock in East Warren is still questioned by experts who find inconsistencies in the agency is testing and enforcement policies. According to the USDA, the Faillace sheep are one of three flocks imported from Belgium and the Netherlands that might have been exposed to feed contaminated with mad cow disease, otherwise known as bovine spongiform encephalopathy (BSE), prior to their shipment to the US.
According to Dr. Larry Faillace, the owner of the sheep, documents exist that prove his family is sheep were never given the feed. However, in a July 18 interview, Dr. Linda Detwiler from the U.S. Department of Agriculture argues that, while the sheep are certified as having receive BSE free feed, cross contamination might have occurred in some of the food processing facilities.
The USDA condemned the Faillace flock after sheep in Houghton Freeman is flock, located in Greensboro, VT, tested positive for a transmissible spongiform encephalopathy (TSE). BSE is one type of TSE. According to an outside scientist, the two of the tests used by the USDA, a Prp-Sc western blot test (protein analysis) and a recently developed and, a yet, unvalidated procedure developed by M.J. Schmeer gave conflicting results. "The sheep with the least dramatic Prp-Sc score got the highest ++(positive) rating on the blot" [sheep #3701 scored negative according to Rubenstein, positive according to Schmeer -- webmaster]
Detwiler said the Schmeer tests were not used in the decision to confiscate the sheep. [The Schmeer test were featured prominently as confirmatory in many USDA press releases. -- webmaster]
BSE is one form of a class of diseases called transmissible spongiform encephalopathies (TSE is) that produce neurological degeneration in the infected animal. This degeneration is similar to the effects of Alzheimer is disease in humans where a protein is reconfigured in the brain. This protein is called a prion and accompanies a gradual chemical process where the brain, normally firm, is reduced to a sponge-like consistency.
The types of TSE include BSE, found in cows; scrapie, found in sheep; chronic wasting disease (CWD), found in deer and elk; and Creutzfeldt-Jakob Disease (CJD) found in humans. In the literature, the distinctions between these diseases are species specific; in other words, scientists define each disease by what type of animal it infects.
Scientists have never found BSE in sheep, other than in laboratory induced conditions. The USDA was unable to classify the TSE they claim to have found in the sheep and said it might be scrapie.
"Unlike BSE, there is no evidence that scrapie poses a risk to human health. Further testing, which will take several years, is required to determine which type of TSE has infected these sheep," said the USDA. CONTROLS AND BLINDS The lack of blinds and control specimens used in the USDA studies could be questioned. The scientific method accepted around the world as standard procedure is to perform blind or double blind tests to rule out any experimenter bias. USDA testers had full knowledge of which samples came from the Vermont sheep according to their reports to Detwiler. Schmeer also knew before hand the samples came from the controversial Vermont sheep, stating "Dear Dr. Detwiler, These are the results from some of the samples of sheep that were submitted from the flocks in Vermont."
Standard practice in the scientific community is the inclusion of a control group in the study. A control group would be specimens of sheep unrelated to the test sheep, where the result of the test is known before hand. Controls help establish the validity test by providing a base line to compare the experimental specimens.
The Faillaces have disclosed USDA documents to The Valley Reporter from the Rubeinstein western blot test. Control specimens were not included in the test results. Detwiler denied these allegations saying that the proper controls and blinds were used for the tests.
Faillace has raised additional questions about the extremely high incidence of the disease in the Freeman flock. He cites the previous occurrence of TSE is in both cattle and sheep herds which is normally one in thousands. To find four out of seven sheep infected from the same small flock, he says, would be extremely unlikely.
When asked about the test results, one commentator responded, "It suggests that no controls were run or that the lab was finding scrapie everywhere that day." These conclusions seem to echo a 1999 letter to the USDA, from a Dr. Janice M. Miller regarding a sheep scrapie survey. "A test of 3000 brains was never carried out because we determined that a prevalence estimate indicated the need for a much larger sample number (10-11,000) than we were prepared to undertake with our resources," said Miller in the letter.
Under the supervision of the USDA, the sheep have undergone constant testing since their arrival in the United States. Four of these most recent tests were performed this year. The USDA has run several histopathology tests. In these tests a slice of brain is analyzed for the appearance of spongiosis, an increase in cell organelles (vacuoles) that might indicate disease. The Faillaces claim all previous tests show a negative result. The most recent USDA test shows plaques, but Detwiler admits the evidence is inconclusive.
The USDA has run immunohistochemistry tests where an antibody to the prion is attached to a fluorescent dye. The antibody would attach itself to any of the abnormal TSE proteins. All of the USDA tests have shown negative results.
The USDA had a scientist M.J. Schmeer run a blood test that is unproven in the science community. The methods used in this test are unknown. This test did indicate disease in five of the Freeman flock sheep. But, again, Detwiler said the USDA did not use this test in their decision to confiscate and destroy the sheep. Schmeer herself admits the test should not be used until validated.
Finally, the USDA has run several western blot tests. In this test, the brain tissue is ground up, dissolved with enzymes, and then tested against antibodies that would attach themselves to a prion protein. Again, the USDA had negative test results for all previous trials until the most recent. According to Faillace, the USDA used a new laboratory, Dr. Rubenstein is, for the most recent hay mite type western blot test.
There is a question of "a lack of reproducibility" in the western blot test. The hay mite data from this lab have not seen any follow-up. Why would the USDA send the samples to this lab out of all the labs that do western blots?"
According to Faillace, Dr. Emanuel Vanopdenbosch, head of the Belgian Animal Testing Laboratories "looked at the references and never heard of this test." Detwiler, however, said the protein test used is widely accepted and used in the scientific community but admitted that Dr. Rubenstein "tweaked" the test." Detwiler maintains the test's validity. "The western blot is pretty much the standard peer review," she said
For the TSE supposedly found in the Freeman sheep to be confirmed as either scrapie or BSE, a laboratory must inject mice with brain matter from the sheep. The mice must then go through several generations before the TSE can be identified. This process can take two to three years. The USDA is unwilling to run this test. One scientist called the USDA decision a "rush to judgement, saying the USDA does not have its scientific ducks in a row. Anyone could support the USDA in taking necessary draconian action. but not willy-nilly nor as a publicity stunt to impress EU (European Union) regulators."
A much greater threat may exist from CWD in deer and elk and echoed the Faillace is sentiment that the USDA might be receiving pressure from the beef and pharmaceutical industries. "The cattle industry has long wanted to slaughter these three particular sheep herds, because the development of BSE could jeopardize cattle, cosmetic, and pharmaceutical exports to the EU."
"There is a potential contamination to the environment," said Detwiler. It is our job to protect the public." Detwiler denied receiving any pressure from the beef or pharmaceutical industries. H4>Sheep case is a puzzle for scientists
July 23, 2000 By JOHN DILLON Staff WriterEven the lawyers were confused.
As an assistant U.S. attorney tried to convince a federal judge Friday that the government should be allowed to destroy two Vermont sheep flocks because of suspicion they harbor a form of mad cow disease, she stumbled through a lengthy description of the illness and admitted she erred in explaining whether the deadly bovine disease could spread to sheep.
Finally, Assistant U.S. Attorney Melissa Ranaldo stopped and apologized to the judge. "This, your honor, is the difficulty in presenting science in the courtroom," she said.
As Ranaldo acknowledged, courts are not always suited to interpreting cutting-edge science. Judges look for solid facts or make strict legal rulings. Scientific research is a shifting ground of experimentation and discovery where new conclusions are often drawn.
Yet the fate of the two Vermont flocks of East Friesian sheep and the livelihood of the farmers who tend them now rest on a federal judge's analysis of a controversial test used to discern the presence of a poorly understood but always-fatal class of infectious diseases.
The U.S. Department of Agriculture announced nine days ago it would seize and destroy 376 sheep because a test had shown four on one farm were infected with a form of transmissible spongiform encephalopathy, or TSE. The government has agreed not to kill the sheep while the matter is before the court. U.S. District Judge J. Garvan Murtha, who held a preliminary hearing Friday, has scheduled an additional session later this week to examine the reliability of the test the USDA used.
TSEs are a family of degenerative brain diseases that include mad cow disease, a mysterious ailment that ravaged the British beef industry and led to more than 70 deaths in the United Kingdom. An extremely rare human form of TSE is called Creutzfeldt-Jakob disease. A new strain of that disease appeared after bovine TSE leaped the species barrier to infect people; the form caused by ingesting infected cows is known as variant Creutzfeldt-Jakob disease, or vCJD.
The cause of the diseases has yet to be fully explained by researchers, although the leading theory is that they are spread by abnormal proteins, dubbed prions. Like no other life form, prions are virtually indestructible, surviving radiation and temperatures of up to 600 degrees Celsius. They apparently lack RNA or DNA nucleic acids, considered to be an essential building block of all life.
The USDA says the Vermont sheep or their forebears were exposed to mad cow disease through contaminated feed - made from meat and bone meal derived from infected animals - before they were imported from Europe in 1996. The flocks' owners, however, argue that the sheep are healthy, that the animals never ate the questionable feed, and that the parent flocks in Europe have shown no sign of mad cow disease or a similar sheep ailment called scrapie.
The stakes are enormous. If the Vermont animals do in fact have mad cow disease, it would be the first time that sheep have contracted the bovine version of TSE outside of laboratory experiments.
And if mad cow disease, or some other, new strain of TSE, gains a foothold in the United States, it could devastate the cattle industry, threaten the production of cosmetics and pharmaceuticals - many of which are derived from cattle - and endanger public health, the government says.
Indeed, the last time the USDA invoked its broad emergency powers to destroy livestock was back in 1983, according to court testimony. In its emergency order, the government said the Vermont outbreak "constitutes a real danger to the national economy and a potential serious burden on interstate and foreign commerce."
Both sides are preparing to offer dueling scientists at the court hearing later this week. Murtha cautioned that he would only take limited testimony on the tests the government relied on for the seizure order.
Yet the fundamental question is whether the government should act decisively when weighing unanswered questions of still-evolving science against the possible spread of a deadly, incurable disease. Or should the USDA simply wait until more tests are done on the Vermont sheep so researchers can learn conclusively if the animals are sick, as the USDA claims, or if the original test was somehow flawed.
Aligned with the USDA are leading TSE researchers, including 1997 Nobel laureate Stanley Prusiner, director of the Institute for Neurological and Degenerative Diseases in San Francisco. Prusiner was awarded the Nobel Prize for his theory that the TSE infectious agent was not a virus but the abnormal prion proteins. The renowned researcher "feels that any possibility of risk from these sheep should be mitigated by purchasing the animals and incinerating the carcasses," according to a sworn statement the USDA filed in court.
But the Vermont sheep owners - Linda and Larry Faillace of East Warren and Houghton Freeman of Stowe - say their animals are healthy and question the diagnostic test used by the USDA's expert. Judge Murtha said Friday he would allow the government and the flocks' owners to present testimony on the narrow question of whether the test was valid.
In court Friday, lawyers for the owners repeatedly challenged the "western blot" test used by Richard Rubenstein at the New York state Institute for Basic Research in Development Disabilities in State Island, N.Y. Rubenstein's July 7 analysis found that the brain tissue of four of the animals culled from Freeman's flock tested positive for TSE.
But the Faillaces said a USDA official told them the test was an "enhanced" version of the standard western blot that is more sensitive, and therefore potentially more prone to yielding false positives. Their lawyer and those representing Freeman said Rubenstein's version of the test is not widely accepted in the research community.
"We contend the USDA is relying on a testing methodology that may not be the best available and might be suspect in some quarters," said John Buckley, attorney for the Faillaces.
Rubenstein is likely to be a witness this week, and would not comment, except to say "there are variations in sensitivity" of western blot tests. A government lawyer said Friday the test was not "enhanced" but is the standard western blot analysis that is recognized internationally. [Rubbish: see OIE international test validation procedures below -- webmaster]
The test results have also been criticized for lacking a blind test -- in which the source of the samples are concealed. The samples were identified as coming from Vermont sheep suspected of harboring a TSE disease, which could have prejudiced the results.
Results of the western blot run apparently did not match results of a still-experimental blood test from the same animals conducted at a USDA lab in Ames, Iowa. One animal showing a high likelihood of TSE in the blood test tested negative in the USDA's western blot. [Sheep #3701 -- webmaster]
The quarantine imposed two years ago on the Vermont sheep may make sense, but not their immediate extermination, said one observer. "There's a rush to judgment. Why can't they wait two to three weeks and have the western blot repeated by a neutral lab?" he asked.
In its emergency declaration used to justify the sheep seizure, the USDA said their test found a TSE "of foreign origin" that is "different from TSE's previously diagnosed in the United States."
But USDA officials have since acknowledged that the western blot test used on the Vermont sheep tissue cannot distinguish whether the animals have scrapie - a relatively common brain disease in sheep that is not considered harmful to humans - or mad cow disease, or some new form of TSE.
"We haven't ruled out scrapie," said USDA spokesman Andrew Solomon. "We don't know in fact if it's scrapie, variant scrapie, or BSE." (BSE refers to bovine spongiform encephalopathy, the technical term for mad cow disease.) The only way to reliably distinguish between scrapie and BSE is to inject the infected brain tissue into mice, then wait months to see the results, scientists say.
Pierluigi Gambetti, a TSE researcher at Case Western University in Ohio, said the western blot test is the "gold standard" used to determine the presence of a TSE. But he said a western blot alone "would not provide a definitive answer" about which strain of TSE the animals may have.
And a European veterinarian familiar with the Vermont flocks and their ancestors in Belgium has challenged the USDA's conclusion.
"Experts in Europe don't believe the tests," said Dr. Bernard Carton, a vet in Belgium who has examined the parent sheep as well as the flocks in Vermont. "The tests from Rubenstein has not been validated yet," meaning no one knows how many false negatives or false positives they yield for each accurate diagnosis, he said in an interview.
Carton said all the parent flocks are healthy and free of scrapie or other TSEs. If four of the Vermont sheep had tested positive for a TSE, the flocks would likely be exhibiting signs of illness, since sheep with scrapie usually show symptoms within a few years, said Carton, who visited the Vermont farms late last year and has maintained close contact with the Faillaces. "But we don't see anything. I don't think they have any problems there," he said.
However, one danger with TSEs is their extremely long incubation period. The Vermont flocks could be infected with an unusual strain of TSE or BSE and not show outward signs of illness, according to Dr. Linda Detwiler, the USDA's lead veterinarian on TSE issues.
Sheep culled from the Vermont flock have also shown microscopic changes in their brain tissue, "funky" signs that could be an indication of TSE, she said. These changes have "become increasingly more significant over the years," she said. [an unusual reading of lab report from England -- webmaster]
Alarmed by these changes, the USDA ordered the experimental blood tests, which also indicated the disease, and then sent brain samples for follow-up work by Rubenstein's lab. "We took the blood positives and tested the brains. Lo and behold, they had the abnormal form of prion protein (associated with TSEs). That is considered a confirmatory marker," Detwiler said.
The Vermont sheep may be serving as sacrificial lambs so the USDA and the U.S. beef industry can assure Europe - which has banned U.S. beef due to concerns about the use of hormones in the animals - that this country is doing everything possible to control the spread of TSE.
"You don't have a consistent story about what's going on with these sheep, whether they have TSE or not, or atypical TSE. There's certainly no support for `foreign origin' (of the disease) and zero support for BSE (mad cow disease)," said a longtime observer.
Scrapie, the common form of sheep TSE, is endemic in the United States; deer and elk in western states are also infected with a version of TSE called chronic wasting disease. Yet the government has not been as aggressive in controlling those outbreaks..
"From the point of view of the beef industry, I believe these sheep farms had to go," he said. "They were pawns in a larger game."
July 23, 2000 By JOHN DILLON Staff Writer tel 802 229-4096, Times ArgusVermont sheep owners whose animals face possible slaughter over concerns they may carry a form of mad cow disease may be able to get more money from the U.S. Department of Agriculture to compensate for the loss of their livelihood.
The U.S. Department of Agriculture wants to seize, slaughter and incinerate 376 sheep after a test on four of the animals indicated suspected infection with a form of transmissible spongiform encephalopathy (TSE), a family of fatal neurological disorders that includes mad cow disease. One farmer in Lyndonville has turned his flock of 21 animals over to the government.
Owners of the two other flocks, Linda and Larry Faillace of East Warren and Houghton Freeman of Stowe, are fighting the seizure order in court. They claim their animals are healthy and that the test the government used is not scientifically valid. The USDA is required to pay fair market value for the sheep. But the agency is also considering mandating that the farmers be barred from using the land until the officials are certain there are no risks of other animals catching the disease through an infectious agent that might be present in the soil.
If the land is removed from production, the farmers may not be able to raise other breeds on the property. But an amendment backed last week by Sen. Patrick Leahy, D-Vt., and Sen. James Jeffords, R-Vt., would allow the USDA to offer an additional $4 million to help compensate for the loss of income, said staff members for the senators. The amendment was included late last week in the annual agriculture spending bill. The Senate-passed measure must now be reconciled with a House version of the USDA funding bill.
Vermont's congressional delegation last week said they backed the USDA seizure and slaughter order as the most prudent way to deal with a possible outbreak of the TSE disease. Leahy said in a statement Friday that the additional funds should help the farmers recover. "USDA wants to do the right thing by these owners, but they need this flexibility to make sure the compensation package considers all of their losses," said Leahy.
Linda and Larry Faillace could not be reached for comment Saturday night. But they have said earlier that their sheep business is worth about $11 million because of the high value of their East Friesian breeding stock. Freeman, who owns Skunk Hollow Farm in Greensboro, said he did not know of the amendment and could not comment. "I'll speak to my lawyer on Monday. They (USDA officials) haven't even offered me compensation (for the animals)," he said. "They spent five hours on my farm and haven't given me an offer."
Meanwhile, state agriculture officials and representatives of the Vermont cheese industry are seeking to reassure the public that Vermont cheese is safe to eat. The Vermont Health Department last week recommended that consumers avoid cheese made from milk of the sheep raised at the Greensboro and East Warren farms. The Health Department issued the warning, but did not recall the cheese.
Health Commissioner Jan Carney said there was a slight, "theoretical" risk of humans contracting TSE from the cheese, even though no cases of dairy-induced infection has been reported during a mad cow disease outbreak in the United Kingdom.
Vermont Agriculture Commissioner Leon Graves last week said the publicity over the USDA seizure order has caused an unwarranted overreaction. "Consumers need to know that only two of the three farms involved in the USDA action are currently involved in cheese production, and that their entire year 2000 cheese output is still on those farms," Graves said. "I want consumers across the country to rest assured that we are looking out of their best interest and that the quality and safety of Vermont cheese is above reproach."
24 Jul 00 Trade Statistics: UK to US Compiled by Terry S.Singeltary Sr of Bacliff, Texas[Opinion (webmaster): The US has focused for years on tracing, containing, and eradicating live animal imports from the UK or other countries with acknowledged BSE like Belgium, including some 499 cattle and the Vermont sheep. This strategy does not acknowledge imports of rendered bovine products from England during the BSE period nor secondary products such as surgical catgut, which is to say surgical cowgut, or dairy cattle embryos, vaccines for veterinarian and human medicines. What has become of these?
Mr. Singeltary, who lost his mother to CJD of unexplained origin a few years back and went on to became a well-known TSE activist, has tracked down voluminous pertinent import data through correspondence with UK officials and searches of government web sites. Imports of such products are frequently cited by Europeans in rating BSE risks in the US and in shutting out US exports.
Many people's eyes glaze over when reviewing reams of sometimes older trade statistics. There is no proof that any of the imported products was contaminated with BSE nor if so, any evidence that any BSE product lead to infection in US livestock, surgical patients, or what not. Nonetheless, the data obtained by Mr. Singeltary establish that an appalling variety and tonnage of products that were imported by the US from the UK and othr BSE-affected countries during the peak of the BSE epidemic years.]
10 January 1990 COMMERCIAL IN CONFIDENCE
NOT FOR PUBLICATION
COMMITTEE ON SAFETY OF MEDICINES
WORKING PARTY ON BOVINE SPONGIFORM ENCEPHALOPATHY
SURGICAL CATGUT SUTURES
2.1 At the first meeting of the Working Party on Bovine
Spongiform Encephalopathy on 6 September 1989, detailed
consideration was given to XXXXX Surgical Catgut. This
arose from the Company's response to the Letter to Licence
Holders, indicating that the bovine small intestine source
material was derived from UK cattle, unlike 8 other
licenced catgut sutures. In contrast XXXXX Surgical
Catgut was stated to hold over 90% share of the market for
catgut sutures, and to constitute approximately 83% of all
sutures used in U.K.
IMPORTS OF SUTURES FROM THE KNOWN BSE COUNTRY;
3006.10.0000: STERILE SURGICAL CATGUT, SIMILAR STERILE SUTURE MATERIALS
AND STERILETISSUE ADHESIVES FOR SURGICAL WOUND CLOSURE; AND SIMILAR
STERILE MATERIAL
U.S. Imports for Consumption: December 1998 and 1998 Year-to-Date
(Customs Value, in Thousands of Dollars) (Units of Quantity: Kilograms)
<--- Dec 1998 ---> <--- 1998 YTD --->
Country Quantity Value Quantity Value
=================================================================
WORLD TOTAL . . . . . . . 10,801 3,116 143,058 40,068
Belgium . . . . . . . . . --- --- 107 14
France . . . . . . . . . 81 49 2,727 1,132
Switzerland . . . . . . . --- --- 1,357 1,693
United Kingdom . . . . . 1,188 242 35,001 5,564
U.S. Imports for Consumption: December 1998 and 1998 Year-to-Date
Subheading 300210: ANTISERA AND OTHER BLOOD FRACTIONS, AND MODIFIED
IMMUNOLOGICAL PRODUCTS
3002.10.0010: HUMAN BLOOD PLASMA
U.S. Imports for Consumption: December 1998 and 1998 Year-to-Date
(Customs Value, in Thousands of Dollars) (Units of Quantity: Kilograms)
<--- Dec 1998 --- <--- 1998 YTD ---
Country Quantity Value Quantity Value
=================================================================
WORLD TOTAL . . . . . . . 25,740 1,827 270,357 20,476
Belgium . . . . . . . . . 14 8 145 60
France . . . . . . . . . --- --- 134 60
Netherlands . . . . . . . --- --- 11 5
Switzerland . . . . . . . 10,462 597 86,101 5,894
United Kingdom . . . . . --- --- 335 62
3002.10.0020: NORMAL HUMAN BLOOD SERA, WHETHER OR NOT FREEZE-DRIED
U.S. Imports for Consumption: December 1998 and 1998 Year-to-Date
(Customs Value, in Thousands of Dollars) (Units of Quantity: Kilograms)
<--- Dec 1998 --- <--- 1998 YTD ---
Country Quantity Value Quantity Value
=================================================================
WORLD TOTAL . . . . . . . 1,039 817 19,056 22,678
Austria . . . . . . . . . --- --- 9,194 18,707
Belgium . . . . . . . . . --- --- 22 15
Netherlands . . . . . . . 353 2 6,733 41
Switzerland . . . . . . . 374 218 1,084 440
United Kingdom . . . . . --- --- 1 4
3002.10.0030: HUMAN IMMUNE BLOOD SERA
U.S. Imports for Consumption: December 1998 and 1998 Year-to-Date
(Customs Value, in Thousands of Dollars)
(Units of Quantity: Kilograms)
<--- Dec 1998 --- <--- 1998 YTD ---
Country Quantity Value Quantity Value
=================================================================
WORLD TOTAL . . . . . . . 1,926 461 14,484 3,563
...
United Kingdom . . . . . 2 8 464 192
3002.10.0040: FETAL BOVINE SERUM (FBS)
U.S. Imports for Consumption: December 1998 and 1998 Year-to-Date
(Customs Value, in Thousands of Dollars) (Units of Quantity: Kilograms)
<--- Dec 1998 --- <--- 1998 YTD ---
Country Quantity Value Quantity Value
=================================================================
WORLD TOTAL . . . . . . . 2,727 233 131,486 8,502
...
Belgium . . . . . . . . . --- --- 17 32
United Kingdom . . . . . 329 82 743 756
3002.10.0090: OTHER BLOOD FRACTIONS NOT ELSEWHERE SPECIFIED OR INCLUDED
U.S. Imports for Consumption: December 1998 and 1998 Year-to-Date
(Customs Value, in Thousands of Dollars)
(Units of Quantity: Kilograms)
<--- Dec 1998 --- <--- 1998 YTD ---
Country Quantity Value Quantity Value
=================================================================
WORLD TOTAL . . . . . . . 88,467 27,343 944,412 309,947
...
United Kingdom . . . . . 1,887 2,300 26,823 23,585
===================================================================
http://www.ita.doc.gov/industry/otea/Trade-Detail/Latest-December/
Imports/30/300290.html
U.S. Imports for Consumption: December 1998 and 1998 Year-to-Date
Subheading 300290: HUMAN BLOOD; ANIMAL BLOOD PREPARED FOR THERAPEUTIC,
ETC. USES; TOXINS, CULTURES OF MICRO-ORGANISMS (EXCLUDING YEASTS) AND
SIMILAR PRODUCTS NESOI
<--- Dec 1998 --- <--- 1998 YTD ---
Country Quantity Value Quantity Value
=================================================================
WORLD TOTAL . . . . . . . 36,178 643 250,982 11,604
...
United Kingdom . . . . . 584 39 11,292 588
http://www.ita.doc.gov/industry/otea/Trade-Detail/Latest-Month/Imports/
05/051199.html
U.S. Imports for Consumption: March 1999 and 1999 Year-to-Date
Subheading 051199: ANIMAL PRODUCTS, NESOI; DEAD HORSES AND OTHER EQUINE
ANIMALS, BOVINE ANIMALS, SHEEP, GOATS AND POULTRY, UNFIT FOR HUMAN
CONSUMPTION, NESOI
0511.99.2000: PARINGS AND SIMILAR WASTE OF RAW HIDES OR SKINS; GLUE
STOCK, NOT ELSEWHERE SPECIFIED OR INCLUDED
U.S. Imports for Consumption: March 1999 and 1999 Year-to-Date
(Customs Value, in Thousands of Dollars)
(Units of Quantity: Kilograms)
0511.99.4024: DAIRY CATTLE EMBRYOS
U.S. Imports for Consumption: March 1999 and 1999 Year-to-Date
(Customs Value, in Thousands of Dollars)
(Units of Quantity: Number)
<--- Mar 1999 --- <--- 1999 YTD ---
Country Quantity Value Quantity Value
=================================================================
WORLD TOTAL . . . . . . . --- --- 53 16
Canada . . . . . . . . . --- --- 9 3
France . . . . . . . . . --- --- 44 13
0511.99.4050: ANIMAL PRODUCTS NOT ELSEWHERE SPECIFIED OR INCLUDED;
DEAD ANIMALS OF CHAPTER 1, UNFIT FOR HUMAN CONSUMPTION
U.S. Imports for Consumption: March 1999 and 1999 Year-to-Date
(Customs Value, in Thousands of Dollars)
(Units of Quantity: Kilograms)
<--- Mar 1999 --- <--- 1999 YTD ---
Country Quantity Value Quantity Value
=================================================================
WORLD TOTAL . . . . . . . 718,476 2,313 2,206,867 4,739
Belgium . . . . . . . . . --- --- 13 18
France . . . . . . . . . 1,088 14 1,489 20
United Kingdom . . . . . 11 3 38 9
http://www.ita.doc.gov/industry/otea/Trade-Detail/Latest-December/
Imports/30/300220.html
U.S. Imports for Consumption: December 1998 and 1998 Year-to-Date
Subheading 300220: VACCINES FOR HUMAN MEDICINE
3002.20.0000: VACCINES FOR HUMAN MEDICINE
U.S. Imports for Consumption: December 1998 and 1998 Year-to-Date
(Customs Value, in Thousands of Dollars)
(Units of Quantity: Kilograms)
<--- Dec 1998 --- <--- 1998 YTD ---
Country Quantity Value Quantity Value
=================================================================
WORLD TOTAL . . . . . . . 25,702 26,150 550,258 378,735
Belgium . . . . . . . . . 14,311 12,029 248,041 199,036
France . . . . . . . . . 3,902 4,859 87,879 92,845
Switzerland . . . . . . . 716 353 9,303 4,271
United Kingdom . . . . . 4,075 1,172 162,960 47,148
==================================================================
http://www.ita.doc.gov/industry/otea/Trade-Detail/Latest-December/
Imports/30/300230.html
U.S. Imports for Consumption: December 1998 and 1998 Year-to-Date
Subheading 300230: VACCINES FOR VETRINARY MEDICINE
List of (6-digit) Subheadings in this (2-digit) Chapter
Next (6-Digit) Subheading ... Descending ... Ascending
Latest Monthly Data
Switch from U.S. Imports to U.S. Exports
About These Trade Data Tables
3002.30.0000: VACCINES FOR VETRINARY MEDICINE
U.S. Imports for Consumption: December 1998 and 1998 Year-to-Date
(Customs Value, in Thousands of Dollars)
(Units of Quantity: Kilograms)
<--- Dec 1998 --- <--- 1998 YTD ---
Country Quantity Value Quantity Value
=================================================================
WORLD TOTAL . . . . . . . 6,528 237 87,149 2,715
Canada . . . . . . . . . --- --- 2,637 305
Federal Rep. of Germany --- --- 104 5
Netherlands . . . . . . . 138 64 472 192
New Zealand . . . . . . . 6,390 173 83,882 1,895
United Kingdom . . . . . --- --- 54 318
=================================================================
http://www.ita.doc.gov/industry/otea/Trade-Detail/Latest-December/
Imports/30/300610.html
U.S. Imports for Consumption: December 1998 and 1998 Year-to-Date
Subheading 300610: STERILE SURGICAL CATGUT, SIMILAR STERILE SUTURE
MATERIALS AND STERILE TISSUE ADHESIVES FOR SURGICAL WOUND CLOSURE;
STERILE HAEMOSTATICS, ETC.
3006.10.0000: STERILE SURGICAL CATGUT, SIMILAR STERILE SUTURE MATERIALS
AND STERILETISSUE ADHESIVES FOR SURGICAL WOUND CLOSURE; AND SIMILAR
STERILE MATERIAL
U.S. Imports for Consumption: December 1998 and 1998 Year-to-Date
(Customs Value, in Thousands of Dollars) (Units of Quantity: Kilograms)
Belgium . . . . . . . . . --- --- 107 14
Federal Rep. of Germany 1,795 356 16,878 3,741
France . . . . . . . . . 81 49 2,727 1,132
Subject: Re: exports from the U.K. of it's MBM to U.S.???
Date: Tue, 8 Feb 2000 14:03:16 +0000
From: S.J.Pearsall@esg.maff.gsi.gov.uk
To: flounder@wt.net (Receipt Notification Requested) (Non Receipt
Notification Requested)
Terry
meat and bonemeal is not specifically classified for overseas trade
purposes. The nearest equivalent is listed as "flours and meals of meat
or offals (including tankage), unfit for human consumption; greaves". UK
exports of this to the US are listed below:
Country Tonnes
1980
1981 12
1982
1983
1984 10
1985 2
1986
1987
1988
1989 20
1990
Subject: Re: Imports of MBM or Ruminants to the U.S. from foreign
Countries with the potential risk of BSE...
Date: Tue, 28 Dec 1999 17:19:15 -0500
From: Linda Detwiler
To: flounder@wt.net (Receipt Notification Requested)
I have attached the file ibov96.txt containing all of the bovine imports for 1996.
Subject: [Fwd: IMPORTED UK AND NETHERLANDS BEEF?] -Reply
Date: Thu, 3 Sep 1998 6:54:00 -0400
From: Linda Detwiler
To: flounder@wt.net (Receipt Notification Requested)
I will check on this as I had not heard about the UK. The Netherlands
would not have suprised me as they did not have a case until March
1997.
...
now my question would be, how many of these animals that fed on MBM's
from these countries, were imported to the United States,
via 3rd country routes??? i will give you that answer below...TSS
Marva Thompson
Foreign Trade Reference Room
202/482-2185
"The U.S. is apparently still importing beef, pork, sheep, and lamb
from countries in which BSE is found [this is probably
completely legal under regulations applicable at time of import--
webmaster]:
Bovine anmls bnlss ex prcssd frozen/U.S. Imports for Consumption 1997
year to date (custom value, in thousands of dollars)
(units of quantity: kilograms)
United Kingdom 37,122 kilograms, 43 thousand dollars
Netherlands 56,260 kilograms, 413 thousand dollars
Canada 18,141,481 kilograms, 23,914 million dollars
Livers of bovine animals, edible, frozen. U.S. Imports for consumption
Netherlands 19,230 kilograms, 25 thousand dollars
Canada 160,632 kilograms, 147 thousand dollars
Tongues of bovine animals, edible, frozen U.S. Imports for consumption
Netherlands 1,047 kilograms, 4 thousand dollars
Canada 767,859 kilograms, 2,028 million
Hi-qulty beef cuts w/bone in prcssd f/c u.S. Imports for consumption
Canada 25,332 kilograms, 37 thousand dollars
Beef cuts w/bone in excpt prcdssd fr/ch u.S. Imports for consumption
Netherlands 5,276 kilograms, 30 thousand dollars
Canada 117,142 kilograms, 353 thousand dollars
Meat bovine anmls cuts w/bone ex prrocssd fr us imports for consumption
Netherlands 51,836 kilograms, 444 thousand dollars
Canada 120,955,010 kilograms, 253,199 million
Cattle hides, whole, fresh or wet-salt u.S. Imports for consumption
Belgium 1,270 pieces, 112 thousand dollars
United kingdom 36 pieces, 3 thousand dollars
Ireland 12,797 pieces, 839 thousand dollars
Italy 50 pieces, 10 thousand dollars
Fr germany 2,500 pieces, 36 thousand dollars
Canada 1,405,430 pieces, 67,320 million dollars
Hides/skins
bovine anmls nesoi whole frh/wet-saltd u.S. Imports for consumption
United kingdom 13 pieces, 1 thousand dollars
Italy 4 pieces, 4 thousand dollars
Germany 9,455 pieces, 139 thousand dollars
Canada 567,816 pieces, 17,196 million dollars
Cattle hides, whole, fresh or wet-salted u.S. Imports for consumption
1998 year to date
Italy 7 pieces, 2 thousand dollars
Ireland 1,408 pieces, 85 thousand dollars
France 25 pieces 2 thousand dollars
Canada 965,355 pieces, 37,244 million dollars
Hides and skins of bovine animals, whole, nesoi, fresh or wet-salted
U.S. Imports for consumption
United kingdom 18 pieces, 3 thousand dollars
Sweden 1 pieces, 1 thousand dollars
Italy 2 pieces, 2 thousand dollars
Germany 5,565 pieces, 72 thousand dollars
Canada 84,327 pieces, 2,257 million dollars
Sheep, lamb skins, no wool, nesoi, pickled not split, u.S. Imports for
Consumption
United kingdom 9,504 pieces, 88 thousand dollars
Sheep, lamb skins, no wool, nesoi, pickled, split u.S. Imports for
Consumption
United Kingdom 149,580 pieces, 1,212 million dollars
Netherlands 50,400 pieces, 267 thousand dollars
Italy 4,175 pieces, 64 thousand dollars
France 13,644 pieces, 57 thousand dollars
Canada 131,642 pieces, 241 thousand dollars
Flawed inspection of food is a danger, senate panel told
9-11-98 Knight Rider Tribune News
The government's current system to check food imports for possible
health dangers is dangerously flawed, experts in the food
business told a Senate subcommittee Thursday. U.S. inspectors
check only 2 percent of all foreign shipments and consistently
issue low penalties to importers who break the rules, experts
said. Unscrupulous importers typically import large amounts of
products that will not pass (Food and Drug Administration)
inspection, said a former West Coast customs broker.
He said importers easily bypass inspections by docking at
high-volume ports, such as Los Angeles-Long Beach and New
York, where the inspection force is stretched thin.
Inspections are so low there they virtually pass right through.
Subject: MBM/U.K. imports of MBM to the U.S./BSE Inquiry
http://www.bse.org.uk/dfa/dfa25.htm
Date:Mon, 10 Apr 2000 15:14:21 -0700
From: "Terry S. Singeltary Sr."
To: flounder@wt.net
69. On 14 February 1990, Mr Meldrum wrote a letter to the
Chief Veterinary Officers of a number of countries. [76] On 15
February 1990, Mrs Attridge and other officials were sent a
copy of the letter of 14 February 1990 and a list of the
countries to which it had been sent. They were stated to be
the countries which had imported ruminant based meat
and bone meal from the United Kingdom. The countries listed
were Norway; Sweden, Switzerland, Czechoslovakia,
Hungary, Nigeria, Thailand, South Africa, Malaysia, Taiwan,
Hong Kong, South Korea, Japan, Canada, USA,
Turkey, Kenya, Malta, Libera, Lebanon, Saudi Arabia, Sri
Lanka, Puerto Rico, Curacao, Finland.[77] The letter from
Mr Meldrum included the following:
Although we have kept the Office Internationale des Epizooties (OIE)
fully informed about this new disease, and they will
shortly be disseminating information and recommendations to member
countries, I am writing to you on a personal basis to
ensure that you are aware of all the developments in relation to BSE,
including its likely cause. The majority of our findings
have now been published in the Veterinary Record.í[78]
70. On 20 February 1990, Dr Pickles wrote to Ms Verity
(APS/CMO). Dr Picklesí minute included the following:
1. Mr Meldrum is arguing that MAFF have already taken all the
necessary and responsible steps to warn importing countries
of the BSE dangers in UK meat and bone meal. Yet the action taken
so far overseas suggest the message has not got
through, or where it has this has been late. The first nation
that woke up to the danger did so a year after our own feed
ban. It seems even now several EC countries neither ban our
imports or the general feeding of ruminant protein. It also
seems the OIE and CVO have yet to inform the rest of the world.
2. I do not see how this can be claimed to be responsibleí. We
do not need an expert group of the Scientific Veterinary
Committee to tell us British meat and bone meal is unsafe for
ruminants. I fail to understand why this cannot be tackled
from the British end which seems to be the only sure way of doing
it, preferably by banning exports. As CMO says in his
letter of 3 January surely it is short sighted for us to risk
being seen in future as having been responsible for the
introduction of BSE to the food chain in other countries.íí[79]
Fri, Jul 21, 2000 By Harriet Tolputt, PA NewsTests were being carried out today to determine whether the human form of mad cow disease claimed another victim. Kirsty Garven, 20, of Waverton, Chester, was unable to walk or talk when she died after 14 months of suffering. Scientists are analysing her brain tissue to confirm whether her death was caused by new variant CJD.
Her parents, Alex and Jennifer Garven, said today that more could have been done to prevent their daughter's death. Mrs Garven, 53, told the Chester Chronicle: "It's greed and profit - that's why we cannot let them get away with it." Referring to the former Conservative government, she added: "I always remember listening on the radio when they said there was no connection with BSE.
"They said nobody can contract it. I hear the same things now about these GM crops. It's all wrong."
Consultant neurologist Nick Fletcher, of the Walton Centre for Neurology in Liverpool, said Miss Garven had all the classical symptoms of new variant CJD when he diagnosed her last October. He notified the National CJD Surveillance Unit in Edinburgh that she was a suspected case and following her death two weeks ago, the centre began carrying out tests.
News of Miss Garven's death comes days after it emerged that there has been a cluster of deaths in the village of Queniborough, Leics. The latest Government figures show there have been 75 known or probable cases of new variant CJD since 1996.
23 Jul 00 Office of International Epizootics (OIE)Opinion (webmaster): What exactly is a validated test for TSE?
The short answer to the common questions:
-- what exactly are the criteria? (see OIE validation manual),
-- who sets the definition? (Office of International Epizootics),
-- is this internationally agreed upon, did US participate, does US help fund OIE? (yes, yes, yes),
-- who determines that a specific test is OIE-validated? (internal decision of agency, country, or continent),
-- is the western blot per se a validated test? (no),
-- must each different antibody and protocol be validated all over again? (yes, for reasons given below),
-- how can a polyclonal antibody be used in a validated test if supplies could be permanently exhausted? (a definite shortcoming),
-- must a validated test be reproducible and transferable to a another lab anywhere in the world? (yes after training),
-- does peer-reviewed scientific publication constitute validation of a test? (no, but have a place in the process),
-- how does a validated test for TSE differ from a validated test for prostate cancer? (TSEs are infectious, animal to human),
-- why are TSE tests difficult to validate? (vexatious disease agent leads to many false positives and false negatives),
-- why is a validation process necessary? (vexatious disease agents lead to many false positives and false negatives)
In the case of the Vermont sheep, the western blot was the general method used. No two western blot protocols are the same. Each western blot protocol needs to be separately validated to control for the many variables inÝtissue preparation, storage, digestion, running conditions, transfer conditions, primary antibody, washing conditions, second antibody selection, and detection method.
The specific western blot used by Rubenstein has not even begun the process of validation. The Schmeer test, capillary electrophoresis not western blot, is also nowhere with validation. A company in Switzerland called Prionics has the most exhaustively evaluated test, known as Prionic-Check. Researchers there have optimized and standardized experimental parameters to minimize false positives and false negatives. The EU-approved and EU-validated BSE and scrapie test is approaching stage III validation (3,000 samples checked against a gold standard).
APHIS/USDA has internal conflicts concerning the validation approach, with some saying that the TSEs are sufficiently novel that standard test validation should not apply (while offering no alternative). Others say that as the normal protein (the "analyte" in OIE terms) is just an isoform of a normal protein co-localizing with the abnormal disease protein marker PrP-Sc, the rigorous standard validation methods used in other diseases is fully applicable. This rigorous standard is precisely what makes the validation protocol suitable for objective and consistent regulatory measures.Ý
USDA will argue at the hearing that "a western blot is a western blot" but this is a falsehood.Ý There are good and bad diagnostic tests of every type, be they western blots, immunohistocompatibility (IHC), complement fixation, or enzyme-linked immunosorbent assays (ELISAs) tests. Any test has an inherent false positive and false negative rate -- none are perfect.
False positives are precisely the issue in the overloaded gels at Staten Island. Dr. Rubenstein will be asked what the sources of false positives and false negatives are in his system and what the rates of these are. If he replies, as does Dr. Schmerr, that there are no false positives or negatives, the Court will want to see the (non-existent) evidence for this. Note sheep #3701 scored positive for disease at Schmeer's lab but a definite negative at Rubenstein's. One or both of the tests therefore has a very high false response rate of 20% or more (5 sheep were tested at both).
The Prionics validation publication describes their false positives and negatives and how they are dealt with.Ý The test uses a monoclonal(pure) antibody 6H4, which recognizes both PrP-C and PrP-Sc; the two isoforms are readily distinguished by their size following PK digestion. Dr. Oesch discusses the finer points of differentiation and controls for the protease K digestion in a second validation paper.
Polyclonal sera, such as the rabbit polyclonal used by Rubenstein, are difficult to use in validations because quantities are limited.Ý Also,Ýreactivity of irrelevant antibodies (the ones the rabbit made in the course of its life in addition to the ones made in response to the inoculated prions) are hard to control for when testing large diverse populations: Ten flocks of sheep might test out perfectly, then one flock will produce a lot of false positives because that flock carries a harmless intestinal or respiratory bug that the rabbit also was exposed to, triggering cross-reacting spurious antibody reactions. Validation, to be meaningful, always include testing a large variety of animals, always comparing test results to a gold standard of carefully characterized reference animals.
APHIS will argue at the trial that polyclonal sera are better because a monoclonal is too specific, binding to only one epitope of 5-6 amino acids that may not be present in the diverse populations under considerations. This is readily circumvented by using two or more monoclonals to different epitopes and sequencing the prion gene of sheep in question (to show epitope conservation). Prionics uses a monoclonal antibody. A consistent, indeed identical, monoclonal antibody reagent can be produced in any quantity; when supplies of a polyclonal run out, the validation process must start all over with a new rabbit. Ý
Clearly there is a need in TSE to use tests validated at the highest level possible or risk sinking into a morass of false negatives and positives that make a shambles of TSE monitoring and eradication efforts. One could ask whether the courtroom is the appropriate forum for these arcane matters -- the alternative for USDA to first validate a test to international standards, then determine using it which flocks and herds need Extraordinary Emergency Seizure Orders.
PRINCIPLES OF VALIDATION OF DIAGNOSTIC ASSAYS FOR INFECTIOUS DISEASES
Validation is the evaluation of a process to determine its fitness for a particular use. A validated assay
yields test results that identify the presence of a particular analyte (e.g an antibody) and allows
predictions to be made about the status of the test subject. However, for infectious disease diagnostic
assays, the identity and definition of the criteria required for assay validation are elusive, and the
process leading to a validated assay is not standardised.
By considering the variables that affect an assay¼s performance, the criteria that must be addressed in
assay validation become clearer. The variables can be grouped into three categories: (a) the sample -
host/organism interactions affecting the analyte composition and concentration in the serum sample;
(b) the assay system - physical, chemical, biological and technician-related factors affecting the
capacity of the assay to detect a specific analyte in the sample; and (c) the test result - the capacity of a
test result, derived from the assay system, to predict accurately the status of the host relative to the
analyte in question.
Factors that affect the concentration and composition of analyte in the serum sample are mainly
attributable to the host and are either inherent (e.g. age, sex, breed, nutritional status, pregnancy,
immunological responsiveness) or acquired (e.g. passively acquired antibody, active immunity elicited
by vaccination or infection). Nonhost factors, such as contamination or deterioration of the sample, may
also affect the analyte in the sample.
Factors that interfere with the analytical accuracy of the assay system are instrumentation and
technician error, reagent choice and calibration, accuracy of controls, reaction vessels, water quality,
pH and ionicity of buffers and diluents, incubation temperatures and durations, and error introduced by
detection of closely related analytes, such as antibody to cross-reactive organisms, rheumatoid factor,
or heterophile antibody.
Factors that influence the capacity of the test result to predict accurately the infection or analyte status
of the host1 are diagnostic sensitivity (D-SN), diagnostic specificity (D-SP), and prevalence of the
disease in the population targeted by the assay. D-SN and D-SP are derived from test results on samples
obtained from selected reference animals. The degree to which the reference animals represent all of
the host and environmental variables in the population targeted by the assay has a major impact on the
accuracy of test result interpretation. For example, experienced diagnosticians are aware that an assay,
validated by using samples from northern European cattle, may not give valid results for the distinctive
populations of cattle in Africa.
The capacity of a positive or negative test result to predict accurately the infection status of the animal
is the most important consideration of assay validation. This capacity is not only dependent on a highly
precise and accurate assay and carefully derived estimates of D-SN and D-SP, but is heavily influenced
by prevalence of the infection in the targeted population. Without a current estimate of the disease
prevalence in that population, the interpretation of a positive or negative test result may be
compromised.
Many variables obviously must be addressed before an assay can be considered ¼validated¼ (8).
However, there is no consensus whether the concept of assay validation is a time-limited process
during which only those factors intrinsic to the assay are optimised and standardised, or whether the
concept includes an ongoing validation of assay performance for as long as the assay is used.
Accordingly, the term ¼validated assay¼ elicits various interpretations among laboratory
diag-nosticians and veterinary clinicians. Therefore, a working definition of assay validation is offered
as a context for the guidelines outlined below.
A. DEFINITION OF ASSAY VALIDATION
A validated assay consistently provides test results that identify animals as positive or negative for an analyte or process (e.g.
antibody, antigen, or indur-ation at skin test site) and, by inference, accurately predicts the infection status of animals with a
predetermined degree of statistical certainty. This chapter will focus on the principles underlying devel-opment and maintenance of
a validated assay. Guidelines for the initial stages in assay develop-ment are included because they constitute part of the validation
process. How this early process is carried out heavily influences the capacity of the eventual test result to provide diagnostic
accuracy.
B. STAGES OF ASSAY VALIDATION
Development and validation of an assay is an incre-mental process consisting of at least five stages: 1) Determination of the
feasibility of the method for a particular use; 2) Choice, optimisation, and stand-ardisation of reagents and protocols;
3) Deter-mination of the assay¼s performance characteristics; 4) Continuous monitoring of assay performance; and 5) Maintenance
and enhancement of validation criteria during routine use of the assay (Figure 1). Although some scientists may question the
rele-vance of the fourth and fifth stages as validation criteria, they are included here because an assay can be considered valid only to
the extent that test results are valid, i.e. whether they fall within statis-tically defined limits and provide accurate infer-ences.
An indirect enzyme-linked immunosorbent assay (ELISA) test for detection of antibody will be used in this chapter to illustrate the
principles of assay validation. It is a test format that can be difficult to validate because of signal amplification of both specific and
nonspecific components (2). This methodology serves to highlight the problems that need to be addressed in any assay validation
process. The same basic principles are used in validation of other complex or simple assay formats.
Because of space limitations, this introductory chapter provides only basic guidelines for the principles concerned with assay
validation. It is derived from a more detailed treatment of the subject (6).
STAGE 1. FEASIBILITY STUDIES
In the ELISA example, feasibility studies are the first stage in validating a new assay. They are carried out in order to determine if the
selected reagents and protocol have the capacity to distinguish between a range of antibody concentrations to an infectious agent while
providing minimal background activity. Feasibility studies also give initial estimates of repeatability, and of analytical sensitivity
and specificity.
1.
Control samples
It is useful to select four or five samples (serum in our example) that range from high to low levels of antibodies against the
infectious agent in question. In addition, a sample containing no antibody is required. These samples will be used to optimise the
assay reagents and protocol during feasibility studies, and later as control samples. The samples ideally should represent
known infected and uninfected animals from the population that eventually will become the target of the validated assay. The
samples should have given expected results in one or more serological assays other than the one being validated. The samples
are preferably derived from individual animals, but they may represent pools of samples from several animals. A good practice
is to prepare a large volume (e.g. 10 ml) of each sample and divide it into 0.1-ml aliquots for storage at ‚20ƒC. One aliquot of
each sample is thawed, used for experiments, and held at 4ƒC between experiments until depleted. Then, another is thawed for
further experimentation. This method provides the same source of serum with the same number of freeze-thaw cycles for all
experiments (repeated freezing and thawing of serum can denature antibodies so should be avoided). Also, variation is reduced
when the experimenter uses the same source of serum for all experiments rather than switching among various sera between
experiments. This approach has the added advantage of generating a data trail for the repeatedly run samples. After the initial
stages of assay validation are completed, one or more of the samples can become the serum control(s) that are the basis for data
expression and repeatability assessments both within and between runs of the assay. They may also serve as standards if their
activity has been predetermined; such standards provide assurance that runs of the assay are producing accurate data (8).
Fig. 1. Stages in the incremental process of assay validation. Shadowed boxes indicate action
points within each stage in the process.
2.
Selection of method to achieve normalised results
Normalisation adjusts raw test results of all samples relative to values of controls included in each run of the assay (not to be
confused with transformation of data to achieve a ¼normal¼ [Gausian] distribution). The method of normalisation and expression
of data should be determined preferably no later than at the end of the feasibility studies. Comparisons of results from day to
day and between laboratories are most accurate when normalised data are used. For example, in ELISA systems, raw optical
density (absorbance) values are absolute measurements that are influenced by ambient temperatures, test parameters, and
photometric instrumentation. To account for this variability, results are expressed as a function of the reactivity of one or
more serum control samples that are included in each run of the assay. Data normalisation is accomplished in the indirect
ELISA by expressing absorbance values in one of several ways (8). A simple and useful method is to express all absorbance
values as a percentage of a single high-positive serum control that is included on each plate. This method is adequate for most
applications. More rigour can be brought to the normalisation procedure by calculating results from a standard curve
generated by several serum controls. It requires a more sophisticated algorithm, such as linear regression or log-logit
analysis. This approach is more precise because it does not rely on only one high-positive control sample for data
normalisation, but rather uses several serum controls, adjusted to expected values, to plot a standard curve from which the
sample value is extrapolated. This method also allows for exclusion of a control value that may fall outside expected confidence
limits.
For assays that are end-pointed by sample titration, such as serum (viral) neutralisation, each run of the assay is accepted or
rejected based on whether control values fall within predetermined limits. Because sample values usually are not adjusted to a
control value, the data are not normalised by the strict definition of the term.
Whatever method is used for normalisation of the data, it is essential to include additional controls for any reagent that may
introduce variability and thus undermine attempts to achieve a validated assay. The normalised values for those controls need to
fall within predetermined limits (e.g. within ±2 or ±3 standard deviations of the mean of many runs of each control).
STAGE 2. ASSAY DEVELOPMENT AND STANDARDISATION
When the feasibility of the method has been established, the next step is to proceed with devel-opment of the assay and standardise the
selected reagents and protocols.
1. Selection of optimal reagent concentrations and protocol parameters
Optimal concentrations/dilutions of the antigen ad-sorbed to the plate, serum, enzyme-antibody conju-gate, and substrate
solution are determined through ¼checkerboard¼ titrations of each reagent against all other reagents, following confirmation of
the best choice of reaction vessels (usually evaluation of two or three types of microtitre plates, each with its different binding
characteristics, to minimise back-ground activity while achieving the maximum spread in activity between negative and
high-positive samples). Additional experiments determine the optimal temporal, chemical, and physical variables in the
protocol, including incubation temperatures and durations; the type, pH, and molarity of diluent, washing, and blocking
buffers; and equipment used in each step of the assay (for instance pipettes and washers that give the best reproducibility).
Optimisation of the reagents and protocol includes an assessment of accuracy by inclusion, in each run of the assay, one or more
serum standards of a known level of activity for the analyte in question. An optimised assay that repeatedly achieves the same
results for a serum standard and the serum controls may be designated as a standardised assay.
2. Repeatability ‚ preliminary estimates
Preliminary evidence of repeatability (agreement between replicates within and between runs of the assay) is necessary to
warrant further development of the assay. This is accomplished by evaluating results from replicates of all samples in each
plate (intraplate variation), and interplate variation using the same samples run in different plates within a run and between
runs of the assay. For ELISA, raw absorbance values are usually used at this stage of validation because it is uncertain whether
the results of the high-positive control serum, which could be used for calculating normalised values, are reproducible in
early runs of the assay format. Also, expected values for the controls have not yet been established. Three-to-four replicates of
each sample run in at least five plates on five separate occasions are sufficient to provide preliminary estimates of
repeatability. Coefficients of variation (standard deviation of replicates/mean of replicates), generally less than 20% for raw
absorbance values, indicates adequate repeatability at this stage of assay development. However, if evidence of excessive
variation (>30%) is apparent for most samples within and/or between runs of the assay, more preliminary studies should be
done to determine whether stabilisation of the assay is possible, or whether the test format should be abandoned. This is
important because an assay that is inherently variable has a high probability of not withstanding the rigours of day-to-day
testing on samples from the targeted population of animals.
3. Determination of analytical sensitivity and specificity
The analytical sensitivity of the assay is the smallest detectable amount of the analyte in question, and analytical specificity is
the degree to which the assay does not cross-react with other analytes. These parameters are distinguished from diagnostic
sensitivity and specificity as defined below. Analyt-ical sensitivity can be assessed by end-point dilution analysis, which
indicates the dilution of serum in which antibody is no longer detectable. Analytical specificity is assessed by use of a panel of
sera derived from animals that have experienced related infections that may stimulate cross-reactive anti-bodies. If, for
instance, the assay does not detect antibody in limiting dilutions of serum with the same efficiency as other assays, or
cross-reactivity is common when sera from animals with closely related infections are tested, the reagents need to be
recalibrated or replaced, or the assay should be abandoned.
STAGE 3. DETERMINING ASSAY PERFORMANCE CHARACTERISTICS
If the feasibility and initial development and stand-ardisation studies indicate that the assay has potential for field application, the
next step is to identify the assay¼s performance characteristics.
1. Diagnostic sensitivity and specificity
a) Principles and definitions
Estimates of diagnostic sensitivity (D-SN) and diagnostic specificity (D-SP) are the primary parameters obtained
during validation of an assay. They are the basis for calculation of other parameters from which inferences are made
about test results. Therefore, it is imperative that estimates of D-SN and D-SP are as accurate as possible. Ideally, they
are derived from testing a series of samples from reference animals of known history and infection status relative to the
disease/infection in question. Diagnostic sensitivity is the proportion of known infected reference animals that test
positive in the assay; infected animals that test negative are considered to have false-negative results. Diagnostic
specificity is the proportion of uninfected reference animals that test negative in the assay; uninfected reference animals
that test positive are considered to have false-positive results. The number and source of reference samples used to
derive D-SN and D-SP are of paramount importance if the assay is ever to be properly validated for use in the general
population of animals targeted by the assay.
It is possible to calulate the number of reference samples, from animals of known infection/exposure status, required for
deter-minations of D-SN and D-SP that will have statistically defined limits. Formulae for these calculations, their
limitations, and a discussion of the selection criteria for standard sera are detailed elsewhere (6). Because of the many
variables that must be accounted for, at least 300 reference samples from known-infected animals, and no fewer than
1,000 samples from known-uninfected animals, should be included to determine initial estimates of D-SN and D-SP,
respectively. The number of samples should be expanded to approximately 1,000 and 5,000, respectively, to establish
precise estimates of D-SN and D-SP (6).
b) Standards of comparison for the new assay
In serology, the ¼standard of comparison¼ is the results of a method or combination of methods with which the new assay is
compared. Although the term ¼gold standard¼ is commonly used to describe any standard of comparison, it should be
limited to methods that unequiv-ocally classify animals as infected or unin-fected. Some isolation methods themselves
have problems of repeatability and sensitivity. Gold standard methods include unequivocal isolation of the agent or
pathognomonic histo-pathological criteria. Because the gold standard is difficult to achieve, relative stand-ards of
comparison are often necessary; these include results from other serological assays and from experimentally infected or
vaccinated animals. Calculations of D-SN and D-SP are most reliable when the gold standard of comparison is available.
When only relative standards of comparison are available, estimates of D-SN and D-SP for the new assay may be
compromised because the error in the estimates of D-SN and D-SP for the relative standard is carried over into those
estimates for the new assay.
c) Precision, repeatability, reproducibility, and accuracy
Repeatability and reproducibility are estimates of precision in the assay. Precision is a measure of dispersion of results
for a repeatedly tested sample; a small amount of dispersion indicates a precise assay. Repeat-ability in a diagnostic assay
has two elements: the amount of agreement between replicates (usually two or three) of each sample within a run of the
assay, and the amount of between-run agreement for the normalised values of each control sample. Reproducibility is the
amount of agreement between results of samples tested in different laboratories. Accur-acy is the amount of agreement
between a test value and the expected value for an analyte in a standard sample of known activity (e.g., titre or
concentration). An assay system may be precise, but not accurate, if the test results do not agree with the expected value
for the standard.
Reliable estimates of repeatability and accuracy, both within and between runs of the assay, can be obtained by use of
normalised results from the many runs of the new assay that were required to assess the sera of reference animals (less
reliable estimates were obtained from preliminary data using raw absorbance values). At least 10, and preferably 20
runs of the assay will give reasonable initial estimates of these parameters. Methods for evaluating these parameters
have been described in detail (6).
Accuracy can be assessed by inclusion of one or more standards (samples of known titre, concentration, etc.) in each run
of the assay. The standards may be control sera provided that the amount of analyte (e.g. titre, concentration) in each one
has been previously determined by comparison with primary or secondary reference standards (8), and the control sera
are not used in the data normalisation process.
Reproducibility of the assay is determined in several laboratories using the identical assay (protocol, reagents, and
controls) on a group of at least 10 samples, preferably duplicated to a total of 20 samples. These samples need to
represent the full range of expected analyte concentrations in samples from the target population. The extent to which the
collective results for each sample deviate from expected values is a measure of assay reproducibility. The degree of
concordance of between-laboratory data is one more basis for determining whether the assay¼s performance
characteristics are adequate to constitute a validated assay.
2. Selection of the cut-off (positive/ negative threshold)
To achieve estimates of D-SN and D-SP of the new assay, the test results first must be reduced to posi-tive or negative
categories. This is accomplished by insertion of a cut-off point (threshold or decision limit) on the continuous scale of test
results. Although many methods have been described for this purpose, three examples will illustrate different approaches,
together with their advantages and disadvantages. The first is a cut-off based on the frequency distributions (6) of test results
from uninfected and infected reference animals. This cut-off can be established by visual inspection of the frequency
distributions, by receiver-operator characteristics (ROC) analysis (4, 9), or by selection that favours either D-SN or D-SP,
whichever is required for a given assay (7). A second approach is establishing a cut-off based only on uninfected reference
animals; this provides an estimate of D-SP but not D-SN. The third method provides an ¼intrinsic cut-off¼ based on test results
from sera drawn randomly from within the target population with no prior knowledge of the animals¼ infection status (3).
Although no estimates of D-SN and D-SP are obtained by this method, they can be determined as confirmatory data are
accumulated.
If considerable overlap occurs in the distributions of test values from known infected and uninfected animals, it is difficult to
select a cut-off that will accurately classify these animals according to their infection status. Rather than a single cut-off, two
cut-offs can be selected that define a high D-SN (e.g. inclusion of 99% of the values from infected animals), and a high D-SP
(e.g. 99% of the values from uninfected animals). The values that fall between these percentiles would then be classified as
suspicious or equivocal, and would require testing by a confirmatory assay or retesting for detection of seroconversion.
3. Calculation of diagnostic sensitivity and specificity
The selection of a cut-off allows classification of test results into positive or negative categories. Calcu-lation of D-SN and
D-SP are aided by associating the positive/negative categorical data with the known infection status for each animal using a
two-way (2 x 2) table (Table 1). After the cut-off is established, results of tests on standard sera can be classified as true
positive (TP) or true negative (TN) if they are in agreement with those of the gold standard (or other standard of comparison).
Alter-natively, they are classified as false positive (FP) or false negative (FN) if they disagree with the standard. Diagnostic
sensitivity is calculated as TP/(TP + FN) whereas diagnostic specificity is TN/(TN + FP); the results of both calculations are
usually expressed as percentages (Table 1).
Table 1. Calculations of D-SN and D-SP aided by a 2 x 2 table that associates infection status with test results from 2,000
reference animals
STAGE 4. MONITORING VALIDITY OF ASSAY PERFORMANCE
1. Interpretation of test results ‚ factors affecting assay validity
An assay¼s test results are useful only if the inferences made from them are accurate. A common error is to assume that an
assay with 99% D-SN and 99% D-SP will generate one false-positive and one false-negative result for approximately every
100 tests on animals from the target popu-lation. Such an assay may be precise and accurate, but produce test results that do
not accurately predict infection status. For example, if the prev-alence of disease in a population targeted by the assay is only
1 per 1,000 animals, and the false-positive test rate is 1 per 100 animals (99% D-SP), for every 1,000 tests on that
population, ten will be false positive and one will be true positive. Hence, only approximately 9% of positive test results will
accurately predict the infection status of the animal; the test result will be wrong 91% of the time. This illustrates that the
capacity of a positive or negative test result to predict infection status is dependent on the prevalence of the infection in the
target popu-lation (5). Of course, the prevalence will probably have been determined by use of a serological test.
An estimate of prevalence in the target population is necessary for calculation of the predictive values of positive (PV+) or
negative (PV-) test results. When test values are reported without providing estimates of the assay¼s D-SP and D-SN, it is
not possible to make informed predictions of infection status from test results (5). It is, therefore, highly desirable to
provide an interpretation statement with test results accompanied by a small table indicating PV+ and PV- for a range of
expected prevalences of infection in the target population. Without provision of such information, test results from the assay
may have failed to accurately classify the infection status of animals, and thus do not reflect a fully validated assay.
STAGE 5. MAINTENANCE AND ENHANCEMENT OF VALIDATION CRITERIA
A validated assay needs constant monitoring and maintenance to retain that designation. Once the assay is put into routine use, internal
quality control is accomplished by consistently monitoring the assay for assessment of repeatability and accuracy (1).
Reproducibility between laboratories should be assessed at least twice each year. It is useful to volunteer membership in a
consortium of labor-atories that are interested in evaluating their output. In the near future, good laboratory practice, inclu-ding
implementation of a total quality assurance programme, will become essential for laboratories seeking to meet national and
international certifi-cation requirements (see Chapter I.2.).
Proficiency testing is a form of external quality control for an assay. It is usually administered by a reference laboratory that
distributes panels of samples, receives the results from the laboratories, analyses the data, and reports the results back to the
laboratories. If results from an assay at a given laboratory remain within acceptable limits and show evidence of accuracy and
reproducibility, the labor-atory may be certified by government agencies or reference laboratories as an official laboratory for that
assay. Panels of sera for proficiency testing should contain a full representation of an analyte¼s concentration in animals of the target
population. If the panels only have high-positive and low-positive sera (with none near the assay¼s cut-off), the exer-cise will only
give evidence of reproducibility at the extremes of analyte concentration, and will not clarify whether routine test results on the
target pop-ulation properly classify infection status of animals.
Because of the extraordinary set of variables that impact on the performance of serodiagnostic assays, it is highly desirability to
expand the number of standard sera from animals of known infection status because of the principle that error in the estimates of
D-SN and D-SP is reduced with increasing sample size. Furthermore, when the assay is to be transferred to a completely different
geographic region, it is essential to re-validate the assay by subjecting it to sera from populations of animals that reside under local
conditions.
When serum control samples are nearing depletion, it is essential to prepare and repeatedly test the replacement samples. When
other reagents, such as antigen for capture of antibody, must be replaced, they should be produced using the same criteria as for the
original reagents, and tested in at least five runs of the assay. Whenever possible, it is important to change only one reagent at a time
to avoid the compound problem of evaluating more than one variable at a time.
C. VALIDATION OF ASSAYS OTHER THAN ENZYME LINKED IMMUNOSORBENT ASSAY
Although the example used has been an indirect ELISA test, the same principles apply to the validation of any diagnostic assay. It is of
utmost importance not to stop after the first two stages of assay validation ‚ that does not constitute a validated assay for diagnostic
use. Although reagent and protocol refinement are important, the selection of the reference populations is probably the most critical
factor. It is no surprise when reviewing the literature to find a wide range of estimates for D-SN and D-SP for the same basic assay.
Although part of the variation may be attributed to the reagents chosen, it is likely that the variation in estimates of D-SN and D-SP
is due to biased selection of sera on which the test was ¼validated¼. This stage in assay validation needs more attention than it has been
given previously. This is particularly true in the current atmosphere of international trade agree-ments and all their implications
with regard to movement of animals and animal products.
REFERENCES
1. Cembrowski G.S. & Sullivan A.M. (1992). Quality control and statistics. In An Introduction to Clinical Chemistry. Bishop M.L.,
Duben-Engelkirk J.L. & Fody E.P., eds. Lippincott, Philadelphia, USA, 63-101.
2. Crowther J.R. (1995). ELISA theory and practice. In Methods in Molecular Biology. Humana Press, Totowa, NJ, USA, 1-256.
3. Greiner M., Franake C.R., Bohning D. & Schlattmann P. (1994). Construction of an intrinsic cut-off value for the
sero-epidemiological study of Trypanosoma evansi infections in a canine population in Brazil: a new approach towards
unbiased estimation of prevalence. Acta Trop., 56, 97-109.
4. Greiner M., Sohr D. & Gobel P. (1995). A modified ROC analysis for the selection of cut-off values and the definition of
intermediate results of serodiagnostic tests. J. Immunol. Methods, 185, 123-132.
5. Jacobson R.H. (1991). How well do serodiagnostic tests predict the infection of disease status of cats. J. Am. Vet. Med. Assoc.,
199, 1343-1347.
6. Jacobson R.H. (1997). Validation of serological assays for diagnosis of infectious diseases (submitted for publication).
7. Smith R.D. (1991). Clinical Veterinary Epidemiology. Butterworth-Heinemann, Stoneham, MA, USA, 1-223.
8. Wright P.F., Nilsson E., Van Rooij E.M.A., Lelenta M. & Jeggo M.H. (1993). Standardization and validation of enzyme-linked
immunosorbent assay techniques for the detection of antibody in infectious disease diagnosis. Rev. sci. tech. Off. int. Epiz., 12,
435-450.
9. Zweig M.H. & Campbell G. (1993). Receiver-operating characteristic (ROC) plots: a fundamental evaluation tool in clinical
medicine. Clin. Chem., 39, 561-577.
1
In this chapter, the terms 'positive' and 'negative' have been reserved for test results and never refer to infection or
antibody/antigen status of the host. Whenever reference is made to 'infection' or 'analyte', any method of exposure to an
infectious agent that could be detected directly (e.g. antigen) or indirectly (e.g. antibody) by an assay, should be inferred.
APHIS Emergency Programs ProMed posting of 24 July 00
" The following is a summary of the current situation relative to a finding of an atypical TSE in sheep in Vermont. On 10 Jul 2000, 4 sheep from Vermont flocks tested positive for a transmissible spongiform encephalopathy (TSE). These 4 positive results, plus 2 suspicious results, were obtained on a Western blot analysis. The samples were processed as described in the following publication: Rubenstein, et al., Arch. Virol. (1994) 139:301-311. The antibody used for the analysis was the polyclonal antisera derived from the rabbit anti-C57BL mouse PrP antibody, referred to as rabbit anti-CMPrP8 in the following publication: Kascsak, et al., Brown, F. editor, Transmissible Spongiform Encephalopathies - Impact on Animal and Human Health, Dev. Biol. Stand. Basel, Karger (1993) 80:141-151. This is a validated recognized test, performed at a USDA cooperating laboratory...."Lisa A Ferguson Sr. Staff Veterinarian USDA, APHIS, Veterinary Services Riverdale, MD Joseph F. Annelli, DVM, MS Chief Staff Veterinarian Emergency Programs Staff Veterinary Services Animal and Plant Health Inspection Services United States Department of Agriculture
Opinion (webmaster): It does not matter whether one uses the lay definition of validated test (APHIS) or the validation manual of Office of International Epizootics -- neither the Rubenstein nor Schmeer tests used in Vermont is validated. Rubenstein apparently will not be able to point to a published page that shows prior experience with western blot of sheep Prp-Sc (the abnormal conformer at issue in Vermont).
The samples were processed as described in Rubenstein, et al., Arch. Virol. (1994) 139:301-311.Ý The method in this publication refers to the classic preparation protocol of Hilmert and Diringer with only slight modification. The publication describes preparation of hamster and mouse brain.
Preparation of sheep tissue is not described. Do we assume that sheep brains are the same as rodents?
The antibody used for the analysis was the polyclonal antisera derived >from the rabbit anti-C57BL mouse PrP antibody, referred to as rabbit anti-CMPrP8 in the following publication: Kascsak, et al., Brown, F. editor, Transmissible Spongiform Encephalopathies - Impact on Animal and Human Health, Dev. Biol. Stand., Basel, Karger (1993) 80:141-151.
This 1993 publication refers to a rabbit polyclonal antiserum directed against gel-purified mouse PrP-Sc (rabbit anti-CMPrP).Ý Methods for production of this antibody are actually found in the cited earlier publication (Kaszsak, RJ, Rubenstein, R, Merz, PA, et al. Immunological comparison of scrapie-associated fibrils isolated from animals infected with four different scrapie strains. 1986. J Virol 59: 676-683.)Ý
The 1993 publication is found in the proceedings from a meeting, not exactly a scientific venue.Ý The rabbit antibody identified as rabbit anti-CMPrP was tested against extracts from sheep brain.Ý The extraction technique in that publication is erroneously referenced to a 1991 publication describing immunohistochemistry techniques with no mention of PrP-Sc extraction.Ý The brief methods section does not provide details on the PK treatment and the centrifugation steps are different than those described in the 1994 publication (above) so the PrP-Sc preparation method cannot be determined from these references.
Further, Table 1 in the 1993 publication indicates that the rabbit serum identified as anti-MCPrP reacts with sheep PrP-C, the normal cellular form of the protein.Ý No data on reactivity with sheep PrP-Sc are shown in this publication although the authors refer to their ability to diagnose sheep scrapie with western blotting.Ý Variations in the concentration of PrP-Sc in different parts of the brain of sheep are mentioned as unpublished data. This is July, 2000.
Is this is a validated recognized test? No data are presented in these publications to show the reactivity of the rabbit antiserum with PrP-Sc from sheep.Ý No data on the sensitivity and specificity of the western blot with this rabbit antiserum are shown in the publications cited.Ý Negative antigen or antibody controls are not described.
Additional misinformation can be found on APHIS web site:
23 Jul 00 Institute web site Background on lab that did western blot tests on Vermont sheep"The Institute for Basic Research in Developmental Disabilities (IBR) is the research division of the New York State Office of Mental Retardation and Developmental Disabilities (OMRDD). Its primary mission is to conduct basic and clinical research. In conjunction with the research program,
The Institute provides extensive services and educational programs. Services include the George A. Jervis Clinic, a tertiary-level diagnostic and research clinic and the Comprehensive Genetic Disease Program of Richmond County, which provides services to Staten Island. The educational programs are embodied in the University Affiliated Program (UAP). The UAP encompasses the Department of Education and Technology Transfer, the College of Staten Island/IBR Center for Developmental Neuroscience and Developmental Disabilities and the State University of New York (SUNY) Health Science Center at Brooklyn/IBR Developmental Disabilities Center.
The Institute works with Alzheimer Disease (IBR was one of four leading laboratories to pinpoint the gene for Alzheimer disease);Batten Disease (IBR scientists have developed the first noninvasive test for this group of rare genetic neurodegenerative disorders), Fetal Alcohol Syndrome, Fragile X Syndrome (IBR scientists first developed tests for fragile X syndrome), Phenylketonuria (PKU) (the institute's first director developed the PKU test), Prenatal Cocaine Exposure, and Taurine deficiency (IBR patented the discovery that taurine is not found in cow milk and some premature infants were at risk for developmental delays if fed taurine-deficient formulas.Ý Now, all government-approved infant formulas have taurine. The patent has brought millions of dollars to the state of New York)."
Dr. Rubenstein currently has an extramural NIH grant.
Grant Number:ÝÝ1R01HL63837-01 Project Title:Ý Creutzfeldt Jakob Disease Diagnosis By Immunomultispectral UV Fluorescence, 1999-2003.
Dr. Rubenstein has had 5 previous NIH grants:
1R03AG14200-01 TRANSGENIC SAM MICE--AGING AND NEURONAL BAPP EXPRESSION (1996-1998) 5R29NS25308-01-04 SCRAPIE AGENT REPLICATION IN AN IN VITRO NEURONAL MODEL (1988-1993) 5R29NS25308-05 SCRAPIE AGENT REPLICATION IN AN IN VITRO NEURONAL MODEL (1988-1994)Current and past grants summaries are retrievable from NIH. There may be additional USDA/APHIS contracts with IBR.
Past publication abstracts may be viewed free at PubMed using search term:
Rubenstein AND (((((((((prion[All Fields]) OR scrapie[All Fields]) OR bse[All Fields]) OR spongiform[All Fields]) OR cjd[All Fields]) OR gss[All Fields]) OR ffi[All Fields]) NOT breast[All Fields]) NOT electron[All Fields])
22 Jul 00 webmaster opinion (for comprehensive coverage of the public relations industry, see PR Watch)The purpose of various USDA and Vermont Dept of Health press releases was to spread fear, uncertainty, and doubt (shortened to FUD by professionals). This is taught early on in journalism school; today, most students go on to be public relations specialists, not journalists.
They start with a standardized form and fill in the F, U, D according to the particular issue at hand. Invariably fatal neurological disorder -- this was the F section. One piece of this cheese and you die from a horrible dementia!!!
The U section stressed that so many uncertainties surround this mysterious disease that several more years of study (transgenic mice) would be necessary to even determine whether it was BSE, so they had to act now in an abundance of caution. Who could blame them!!!
For the D section, they focused on doubts about whether the sheep were pastured on grass as the documentary record indicated or if in fact the sheep's ancestors were possibly affected by shadowy traffic in smuggled UK bone meal in the shady country of Belgium, where they feed the cattle dioxin whereas an upright country like US only feeds it to chickens. Maybe they only ate grass during the day!!!
It is also deemed ethical to malign the target using guilt-by-association. The idea is to mention BSE in the every breath, even though the Vermont sheep only had a "diffuse astrocytosis... commonly seen in adult sheep" unlike real sheep with experimental BSE. These sheep might be mad cows!!!
For school assignments, they must color a press release (newspaper story) in MS Word, red for F, blue for U, green for D. When all the words are assigned the correct color, they can turn in their assignment.
Color code: FEAR UNCERTAINTY DOUBT MALIGNA minor mad sheep press release mystery was cleared up today by a reporter who witnessed an over-heated discussion in the corridors at the Atlanta Emerging Infections Conference. The webmaster was baffled by stupid and unsupportable comments the Center for Disease Control was making about risks from sheep milk and cheese. Surely they knew the scientific literature better. it turned out they hadn't actually made the comments. Vermont public health had simply made them up out of whole cloth to please USDA and attributed them to CDC for credibility, thus enraging CDC.
Below are two interesting, indeed somewhat confused, articles to color. The first starts off with a scary headline that illustrates the collateral damage to lievstock producers and consumer confidence about BSE in the US that the high profile USDA action brought. Then there is some reassuring misinformation about import restrictions that doesn't hold up, seeding further apprehensions in the reader. Then, "there's a chance that the Vermont sheep don't have TSE at all but rather a disease called "scrappie," that is not dangerous to humans."
The second article, about killing 21 sheep from a small neighboring flock, gets off to a bad start, "USDA removes first Vermont sheep flock linked to virus" and later stresses industry support of the magnaminious USDA decision to set the fair market value of the Level 3 Biohazard sheep at $3000 per animal, some 30x over Friday's auction prices.
Are these to be understood as public relations journalism or simply as well-intentioned efforts to write on an inherently difficult and confusing subject?
Four sheep were confirmed positive on July 10 for possible transmissible spongiform encephalopathy -- the sheep's form of the fatal disease. However, it could take up to two years for scientists to be sure due to the lengthy incubation period of the disease. Consequently, fears were raised about the possibility that the disease might constitute a possible health risk for humans, Creutzfeldt-Jakob Disease, the human form of the disease, and the U.S. government has chosen to err on the side of caution.
The ongoing bovine brain tissue surveillance system has been in place in the U.S. for many years with no evidence of BSE seen. Severe restrictions have been in place on the importation of certain products from countries where BSE was known to exist, such as Britain. Even people who have traveled to Britain and that have spent a total of six months or more on that side of the pond from 1980 to 1996 are banned from donating blood in the United States.
Infected cattle in Europe has claimed the lives of some 67-69 victims between 1985 and March of this year who most likely had the fatal condition; an estimated 180,000 cows have been infected with mad cow disease in the last 20 years - apparently through feed that included the rendered brains of animals that carried the virus.
Officials are presently trying to link the eating of beef from infected cattle decades ago to the current mini-epidemic that is going on in Britain; investigators are looking at a cluster of