BSE Offers Lessons in Risk Assessment

One of the greatest challenges in drug development is performing adequate assessment of biological and microbiological risks that can compromise patient safety. Discovery of new infectious diseases, which occurs on average one to two times every year, influences how we view food and drug safety as well as hygiene.¹ The measures taken to contain and evaluate the risks of bovine spongiform encephalopathy (BSE or "mad cow disease") illustrate how new risks are evaluated to determine when potential changes to regulations, practices, and disclosures to consumers are needed. The lessons apply to both the pharmaceutical and biopharmaceutical industries, where risk assessment and mitigation is a regulatory requirement.

BSE'S BEGINNINGS

The 1986 discovery of BSE in the UK — a previously unknown disease in cattle — set off a chain of events that continues to affect the food and drug industries.

^2-12

Despite efforts to impose feeding bans, limit exports from the UK, and destroy at least four million diseased cattle, BSE prevalence continued to grow for years after protective measures were implemented (see Table 1).

^13,14

At the peak of the UK BSE epidemic, about 1,000 infected cattle were detected each week. BSE has since appeared in about 30 countries, apparently spread by UK cattle or feed.

^13,15

In 1994, a new disease, variant Creutzfeld-Jakob disease (vCJD), appeared in humans.

^16,17

vCJD likely was caused by a cross-species infection, or zoonosis, from BSE-infected cattle parts that entered the human food chain. It is thought that BSE itself originated from a similar transfer of scrapie infection from goats and sheep to cattle.

The advent of BSE and the discovery of similar transmissible spongiform encephalopathies (TSEs) in several other animal species forced the reevaluation of risks for organ and tissue grafts, vaccines, and drugs. CJD was already known in humans-as both a spontaneously arising disease and as an iatrogenic infection from human tissues and pituitary-derived growth hormone. However, little was known about how TSEs are transmitted, diagnostic tests were not available, and conventional thinking assumed high barriers to zoonosis.^2-4

RELATIVE RISK

Given existing knowledge and prevailing scientific opinion from 1986 to 1991, we can examine how the risks and unknowns were managed, starting with the following questions:

• What safeguards were appropriate to control BSE incidence in cattle before the first known case of human vCJD?

• What government controls on food are needed after a very rare food-borne risk is identified?

• How do we decide whether raw materials derived from the same animals pose a safety risk when used to manufacture pharmaceuticals?

• How safe is safe enough for food? And for drugs?

• How central a concern is BSE, or vCJD, for food and drug consumers?

Prior to 2003, the regulatory bodies and TSE scientific advisory committees that assess BSE risk in different countries placed the US and Canada in an intermediate category not assumed "BSE-free."^4-12,15 During 2003, officials discovered a BSE-infected animal in Canada and one in the US.^18,19 In April 2004, FDA announced that — against USDA regulations — a cow with central nervous system symptoms similar to BSE was slaughtered and rendered for use in animal feed without being tested for BSE.²⁰ The rendered protein was seized, and an ongoing investigation will determine whether it will be used solely for swine feed or destroyed.

What factors led to the emergence of BSE in North America? One hypothesis is that it developed spontaneously, independent of events in the UK. Another hypothesis suggests BSE occurs naturally in animal populations but it previously was not diagnosed.

Records show that 173 cattle imported into the US from the UK during the 1980s entered the food chain. There is a high probability that some were BSE positive.^13-15

Two TSEs already existed in North America, one in deer and one in elk. Perhaps one of these TSEs crossed the species barrier, infecting cattle.

Until 1997, US rendering, feed, and slaughtering practices matched those used in the UK in the 1980s. Feeding rendered mammalian parts to ruminants allowed scrapie in goats and sheep the opportunity to cross into cattle as BSE and possibly into humans as vCJD. Feeding high-risk tissues to ruminants is safe only if no infected animals are present in the population. North America lagged well behind the UK in imposing the 1997 partial feed ban. A full ban was not imposed until after the December 2003 discovery of BSE in the US.²³

Why weren't feed and rendering practices changed sooner? Perhaps regulatory officials and industry thought BSE infection could never happen here, that it cost too much to do anything about it, or that there was no reason to worry until BSE was found. Further, a formal risk assessment indicated that even the importation of 500 BSE-infected cattle would, in theory, not be enough to establish BSE in North America. Of course, many assumptions made in the model could not be tested, including infectivity, incubation times, and compliance with feed regulations.^21,22

Was regulatory oversight effective and properly resourced? Enforcing the 1997 feed ban required resources, education, and time. FDA studies show that compliance lagged years behind changes in the law — similar to the lag that occurred earlier in the UK. Also, confusion between different agencies (FDA, USDA, the Joint Institute for Food Safety and Applied Nutrition, as well as several European agencies) over jurisdiction and approach created initial difficulties. The differing charters of these agencies (to protect local meat industries, to protect public health, and so on) may have created challenges to collaboration. Despite these difficulties, a series of international guidance documents was successfully developed, approved, and is in use today. Since both BSE-infected cows in North America were born around 1997, they could have become infected just before the feed ban became law, or they could have been infected during the compliance lag period.^18,19

TESTING METHODS

BSE has an estimated incubation period of three to six years.

²¹

Young infected animals often test negative, complicating surveillance efforts. Therefore, any sampling plan should target older animals, where BSE-infection is easier to detect. Tests are expensive, and the biological test requires months. More rapid tests, including a Western blot method, are faster but may require confirmation to prevent false positives.

²⁵

In 2002 and 2003, about 20,000 US cattle (both downer cattle and randomly selected healthy cattle) were tested for BSE as part of USDA surveillance. With 36 to 45 million cattle slaughtered annually, the 20,000 cattle represented a sampling rate of approximately 1 in 1,700 cattle.^24,26 The number of samples tested was predicated on the assumption that the frequency of infection was less than 1 in 10⁶. Now that at least two cases have been identified in North America, the infection rate could be as high as 1 in 10⁵ or as low as 1 in 10⁸.

In February 2004, USDA announced plans to test 40,000 animals annually, a twofold increase over the previous year. In March, after further consideration by the TSE Advisory Committee and the Harvard Risk Analysis Team, USDA announced its intent to greatly increase sampling over the next two years, especially of downer cattle. If as many as 268,000 cattle were tested, an accurate estimation of BSE prevalence in cattle could be derived, and this proposal is under discussion.^22,23

PHARMACEUTICAL LINKAGE

Sporadic CJD, a TSE whose etiology is unknown and is unrelated to BSE in cattle, occurs in about one in one million persons annually. Iatrogenic CJD has occurred in people treated with human pituitary gland-derived growth hormone (pitHGH) and also in tissue-and organ-graft recipients.

^27,28

These findings suggested that infection could be passed by grafts and possibly also by pharmaceuticals. However, patients such as hemophiliacs who use blood or blood products frequently and patients treated with recombinant pharmaceuticals did not develop CJD at increased rates during the 1980s and 1990s, suggesting that any increased risk is quite low.

²⁸

Because of the experience with pitHGH and tissue grafts, US blood donors are prescreened for CJD risk as a cautionary measure, and raw materials must be assessed for risk.

Concerns about BSE contamination of pharmaceuticals became serious in the early 1990s when the first cases of vCJD were discovered. At one point, the European Parliament considered banning the use of all raw materials derived from animals. At the time, more than 90% of approved pharmaceuticals contained such materials, notably gelatin and stearate derivatives. To remove these pharmaceuticals from use without a single case of transmission to humans appeared excessively cautious. Instead, regulators issued guidance documents urging firms to evaluate animal-derived raw materials for suitability and safety, and a list of high-risk tissues was developed.^5-12 Many firms devoted significant process-development resources to replacing high-risk, animal-derived raw materials with items perceived to be safer. Serum, serum-derived proteins, even tallow-derived compounds (such as polysorbate 80, glycerol, and stearates), and bovine insulin (replaced with recombinant insulin produced in bacteria), were replaced in many pharmaceutical products.

For biological products, such as vaccines or proteins produced in cell culture, a clear distinction is made between one-time risks and recurrent additive risks.^8,9 One-time risks include the exposure of the original master cell bank to serum or other raw materials, while repeated risks involve using specified risk materials in each production lot. Based on the current data, blood and blood-derived products, although a potential concern, appear lower in risk than human organs or tissues for CJD and probably for vCJD as well. To date only one person has developed vCJD after exposure to donor blood of a person who later developed vCJD.³⁰ It cannot be determined whether the transfusion caused the infection.

The risk of TSEs in pharmaceuticals is thought to lie in the use of high-risk, animal-derived raw materials, such as brain, spinal cord, and some other organs. These raw materials have been used as excipients, starting materials, or in-process materials that are not present in the final product.^29-31 Accordingly, the industry has been mandated since the 1990s to reduce or eliminate exposure to high-risk materials. Risk is assessed according to an animal's country of origin, tissue or organ sourced, and the proximity of the raw material to the patient.³¹

Table 2 lists some of the most commonly used raw materials and excipients in the pharmaceutical industry, and their relative TSE risk as currently understood.^4-12 Ruminants and primates are the highest-risk species, while TSEs have not been detected in poultry and pigs, so they are assumed to be safer. Brain and spinal cord tissue has the highest risk, based on tissue infection studies. Country of origin of the animals is another key risk factor having four categories, ranging from no BSE detected (not necessarily BSE-free), low incidence of BSE suspected, low incidence of BSE known, and highly infected with BSE. Sourcing raw materials from low-risk countries is one approach to reducing BSE risk. As the recent cases of BSE in North America do not significantly change the risk classification of the US or Canada, little direct impact is expected on most pharmaceutical products at this time. However, dietary supplements made using high-risk tissue such as bovine brains could be greatly affected. Similarly, firms with mature product lines, especially certain vaccines that require high-risk materials for growth, may have more serious regulatory concerns. Substituting complex, biologically active raw materials for "safer" materials may affect product quality and it requires significant process development. Relying on sourcing from a "safer" country of origin also can be risky, as the status of a country may change at any time and supplies are limited.

In contrast to the conservative steps applied to pharmaceutical products, dietary supplements were not placed under similar regulation until February 2004.

CONTROLS AND PRACTICES

What steps did industry and regulators take, and when did they take them relative to the emergence of BSE? Table 3 provides a brief chronology of some of the main events. It also shows that other countries did not immediately implement feed controls that the UK enacted starting in 1988, despite the devastation experienced in the UK cattle industry as a result of the BSE epidemic. For example, Canada implemented some rules only in July 2003, after finding a BSE-positive cow in May, and the US fully implemented these controls in February 2004, after finding a BSE-positive cow in December 2003.

^18,23-24

The US response lag is notable, since cattle move freely across the US-Canadian border.

To guide public policy and government spending, the meaning of this data must be evaluated in a broader context. Logically, relative risks should guide resource allocations and decisions, but not all diseases are created equal: some affect small children, some are highly infectious, and some, such as vCJD, create much pain and suffering. Each of these factors is a reason to increase the relative investment in surveillance and prevention.

Food-borne diseases cause an estimated 76 million illnesses each year, including 325,000 hospitalizations, and 5,000 deaths, according to CDC.^32,33 As of March 2004, about 155 vCJD cases have occurred worldwide. Society must decide how much money to allocate to improving meat safety as regards TSE compared to E. coli 0157: H7, Listeria, and other infections that have already caused hospitalizations and thousands of deaths in the US alone.³⁴ It can be argued that preventing the entry of a new disease into North America is worth expending disproportionate resources; likewise, it is necessary to focus limited resources to maximize the impact on public health. Policy makers have difficult choices to make, and the stakes are high. What should be done, and what can we afford?

What happens when we approach the TSE conundrum from the viewpoint of reducing risk to consumers? The following factors appear to have increased BSE and vCJD risk:

• There are large-scale movements and intermingling of livestock, feed, and semen, permeating the national food supply. For example, in tracing the peers of one cow, hundreds of animals at more than 15 locations had to be destroyed.^26,35

• Meat is processed in large batches, meaning that any consumer eating one hamburger comes in contact with meat from a large number of cattle, increasing the per-event risk of exposure.

• Automation of meat processing has increased the risk that spinal cord or brain could contaminate ground meat.

What actions should be considered, in addition to those already mandated by the EU, US, and Canadian authorities, to reduce potential health risk in humans? There are a number of options:

• Set up smaller or closed herds in defined geographic areas to contain any infections in a smaller population, and decrease shipping of animals and mixing of herds.

• Stop all animal rendering and all cross-species feeding, the activities that are thought to have created the risks. But what should be done with the waste?

• Test all cows older than 30 months for BSE before allowing them into the food chain, thus focusing resources on the highest-risk animals. This practice is used in the UK and Japan and is under discussion in the US. The difficulty lies in storing large quantities of meat to wait for test results.

• Increase general surveillance for BSE by testing more animals. This is being implemented in Canada and the US and will be helpful in defining risk and guiding public policy.

• Consider reduction in batch sizes of feed and meat products, and improve feed traceability. Reducing batch sizes would expose fewer people to infected meat if an infected animal entered the food chain. Additionally, since feed likely is the causative agent, knowing what feed is used where could allow more targeted sampling when an infected cow is found. At present, it is not possible to trace feed.

• Lastly, more effective cleaning procedures, combined with dedicated equipment, would reduce the opportunity for cross-contamination between high- and low-risk feeds.

These measures may increase costs, yet they also could decrease other microbial risks such as E. coli. The current industry drive for economy should be balanced with attention to safety. Cost pressure is intense in the beef industry, but the burgeoning market for organic foods suggests that many consumers are willing to pay more for high-quality foods.

Will BSE fears drive the beef and dairy industries to create new definitions of quality, with resultant impact on prices and profit margins? Careful scientific assessment of the risks and benefits, coupled with business modeling, will be crucial to good decision-making. Surveillance and inspection also will play a role, providing ongoing feedback on incidence and risk and maintaining vigilance for errors or accidents.

RESOURCE LIST

Centers for Disease Control and Prevention (CDC)www.cdc.gov/ncidod/diseases/cjd/cjd.htm

Food and Drug Administration (FDA)www.fda.gov/oc/opacom/hottopics/bse.html

The Animal and Plant Health Inspection Service (APHIS) www.aphis.usda.gov/lpa/issues/bse/bse.html

Department of Health, UKwww.oie.int/

REFERENCES

1. Centers for Disease Control. CDC Reports. Available at: www.cdc.gov/ncidod/diseases/dbmd/diseaseinfo

2. Wells GA, et al. A novel progressive spongiform encephalopathy in cattle. Veterinary Record 1987; 121: 419-420.

3. Baker HF and Ridley RM, eds. Prion Diseases. Totowa, NJ: Humana Press; 1996.

4. Institute of Food Science & Technology. BSE and vCJD in humans; Annexes 1 and 2. October 2001. Avaliable at: www.ifst.org/hottop5.htm

5. European Medicines Agency. CPMP position paper on production of tallow derivatives for use in pharmaceuticals. December 1997. Available at: www.emea.eu.int/pdfs/human/press/pp/116397en.pdf

6. European Medicines Agency. Explanatory note: gelatin for use in pharmaceuticals. Available at: www.emea.eu.int/pdfs/human/press/pus/430600en.pdf

7. European Medicines Agency. EMEA expert workshop on human TSEs and medicinal products derived from human blood and plasma. 1 December 2000. Available at: www.emea.eu.int/pdfs/human/regaffair/045001en.pdf

8. European Medicines Agency. CPMP position statement on new variant CJD and plasma-derived medicinal products. Available at: www.emea.eu.int/pdfs/human/press/pos/020198en.pdf (Superseded by CPMP/BWP/2879/02). See also Questions and answers on bovine spongiform encephalopathies (BSE) and vaccines. Available at: www.emea.eu.int/pdfs/human/bwp/081901en.pdf

9. European Medicines Agency. Public statement on the evaluation of bovine spongiform encephalopathies (BSE)- risk via the use of materials of bovine origin in or during the manufacture of vaccines. Available at: www.emea.eu.int/pdfs/human/press/pus/047601en.pdf

10. European Medicines Agency. Position paper on re-establishment of working seeds and working cell banks using TSE compliant materials. Available at: www.emea.eu.int/pdfs/human/press/pp/2231402en.pdf

11. European Medicines Agency. Discussion paper on the investigation of manufacturing processes for plasma-derived medicinal products with regard to vCJD risk. Available at: www.emea.eu.int/pdfs/human/bwp/513603en.pdf

12. CBER. Guidance for industry: revised preventive measures to reduce the possible risk of transmission of Creutzfeldt-Jakob disease (CJD) and variant Creutzfeldt-Jakob disease (vCJD) by blood and blood products. Available at: www.fda.gov/cber/gdlns/cjdvcjd.pdf

13. World Organisation for Animal Health. Number of cases of BSE reported in the United Kingdom. Available at: www.oie.int/eng/info/en_esbru.htm Updated 13 November 2003.

14. Anderson RM, et al. Transmission dynamics and epidemiology of BSE in British cattle. Nature 1996; 382: 779-788.

15. Center for Veterinary Medicine. CVM update. 6 July 2001. Available at: www.fda.gov/cvm/index/updates/bse72001.htm

16. Heye N, et al. Creutzfeldt-Jakob disease and blood transfusion. The Lancet 1994; 343: 298-299.

17. Will G, et al. A new variant of Creutzfeldt-Jakob disease in the UK. The Lancet 1996; 347:921-925.

18. FDA. Statement on BSE cow in Alberta, Canada. 20 May 2003. Available at: www.fda.gov/bbs/topics/NEWS/2003/NEW00908.html

19. Centers for Disease Control and Prevention. Bovine spongiform encephalopathy in a dairy cow-Washington State. Morbidity and Mortality Weekly Report 2004; 52(53):1280-1285. Available at: www.cdc.gov/mmwr/PDF/wk/mm5253.pdf

20. FDA. Statement on Texas cow with central nervous system symptoms. 4 May 2004. Available at: www.fda.gov/bbs/topics/news/2004/NEW01061.html

21. Harvard Center for Risk Analysis. Evaluation of the potential for BSE in the United States. Available at: www.hcra.harvard.edu/pdf/ResponsetoComments.pdf

22. Harvard Center for Risk Analysis. Comments on USDA bovine spongiform encephalopathy (BSE) surveillance plan. 15 March 2004. Available at: www.aphis.usda.gov/lpa/issues/bse/BSE_Harvard03-12-04.pdf

23. USDA. Veneman announces expanded BSE surveillance program. Available at: www.usda.gov/Newsroom/0105.04.html

24. USDA. Bovine Spongiform Encephalopathy (BSE) surveillance. Veterinary Record 2004; March 16, Available at: www.aphis.usda.gov/ipa/issues/bse/bse-surveillance.html

25. European Medicines Agency. EMEA workshop on applications to pharmaceuticals of assays for markers of TSE. 19-20 January 1999. Available at: www.emea.eu.int/pdfs/human/bwp/025799en.pdf

26. FDA. Statement by Lester M. Crawford, DVM, Ph.D, Deputy Commissioner of Food and Drugs, DHHS, before the committee on agriculture, nutrition, and forestry, United States Senate, January 27, 2004. Available at: www.fda.gov/ola/2004/bse0127.html

27. CDC. Questions and answers regarding Creutzfeldt-Jakob disease infection-control practices. Available at: www.cdc.gov/ncidod/diseases/cjd/cjd_inf_ctrl_qa.htm

28. EU Scientific Committee on Medicinal Products and Medical Devises. Opinion on the risk quantification for CJD transmission via substances of human origin. 16 Feb 2000. Available at: www.europa.eu.int/comm/dg24/health/sc/scmp/out28_en.pdf

29. TSE Advisory Committee. Presumptive transfusion transmission of vCJD. Possible implications for FDA policies. TSE Advisory Committee Meeting of January 2004. Available at: www.fda.gov/ohrms/dockets/ac/04/minutes/4019m1_quicksummary.pdf

30. Lubiniecki AS, Shadle PJ. Raw material considerations. In Brown F., Fernandez J, eds., Development of Specifications for Biotechnology Pharmaceutical Products. Basel, Switzerland: S. Karger Publishing; 1997. See also Shadle PJ. Qualification of raw materials for biopharmaceutical use. BioPharm International 2004; 17(2): 28-35.

31. EU Scientific Steering Committee. Listing of specified risk materials: a scheme for assessing relative risks to man. Opinion of the Scientific Steering Committee, adopted on 9 Dec 1997. Available at: www.europa.eu.int/comm/food/fs/sc/ssc/out22_en.pdf

32. CDC. Disease Information. Available at: www.cdc.gov/ncidod/dbmd/diseaseinfo

33. Young A. Other ills dwarf mad cow disease. Miami Herald, 4 January 2004. Available at: www.tallahassee.com/mld/miamiherald/living/health/7628869.htm

34. Moss, M, Oppel, RA, and Romero S. Mad cow forces beef industry to change course. The New York Times, Jan 5, 2004, ppA1, A14.

35. FDA. FDA statement on rendered products derived from BSE cow in Washington state. 27 December 2003. Available at: www.fda.gov/bbs/topics/NEWS/2003/NEW01000.html