Some reflections on the possible link between bovine spongiform encephalopathy and the new variant of Creutzfeldt–Jakob disease

Creutzfeldt-Jakob disease (CJD) has suddenly become the focus of scientific, epidemiologic and, not least, passionate political debate following the March 1996 statement on a possible link between bovine spongiform encephalopathy (BSE) and the new variant of CJD (nvCJD) by the Secretary of State for Health of the United Kingdom in the House of Commons. However, CJD is a rare disease; most physicians have never seen a single case and, hopefully, will not see any during their career. CJD is the commonest type of the transmissible spongiform encephalopathies (TSE) or prion disease; CJD occurs in inherited, acquired and sporadic forms. The inherited form is autosomal dominant and associated with coding mutations of the prion protein gene. The acquired form occurs after iatrogenic exposure to biological material drawn from infected patients (cadaveric-derived growth hormone, dura mater and cornea) or after neurosurgery with surgical tools contaminated during brain surgery of an infected patient. The Kuru epidemic form in Papua New Guinea was transmitted through cannibalistic rituals. The sporadic form which accounts for 85% of all CJD cases occurs worldwide with a comparable annual incidence rate from one country to another (about one case per one million and per year), a mean age of onset at around 65 years (very rare before 40 years) and a rapid progression to death (3–6 months). However, three clusters of sporadic CJD have been reported (two in former Czechoslovakia and one in Israel among Jews of Libyan origin). These three clusters are linked to a genetical susceptibility associated with anomalies of the gene that codes for PrP. Analytic epidemiologic studies, until now, have been unable to identify any consistent risk factors for the occurrence of the sporadic form of CJD (environmental, foodborne, or occupational exposure) other than a genetic susceptibility (methionine homozygotes at codon 129). In particular, no relationship between the incidence of CJD in humans and of scrapie in sheep has been documented.

BSE was first observed clinically in April 1985 in the UK and was identified as a spongiform encephalopathy in November 1986. The epidemic peaked in 1992 (36 681 cases) and to date more than 160 000 cases have been notified in the UK. Ruminant-derived meat and bone meal (MBM) were implicated as the mode of transmission of BSE in December 1987. As a consequence, a ban on ruminant feed was implemented in the UK in July 1988, followed more than a year later by a ban on the use of certain specified offals (brain, spinal cord, spleen, etc.) for human consumption and for inclusion in MBM in England and Wales. Other European countries have also implemented similar measures.

Current evidence suggests that BSE originated from MBM contaminated by a scrapie-like agent derived from sheep or cattle. The infectiousness of MBM was probably facilitated by changes in the early 1980s in the method of rendering of ruminant offal into MBM which made it less effective in neutralizing the BSE agent. Although the pattern of the BSE epidemic in the UK is consistent with a main route of transmission through contaminated feed, a vertical route of transmission of BSE from dam to calf has been discussed since 1995, and a 10% rate of transmission has even been reported recently [1]. In contrast to scrapie in sheep and goats and other TSEs in rodents, so far there is no evidence that cattle genotype influences the susceptibility to BSE.

The emergence of BSE and its epidemic spread in the UK have raised many questions and concerns about the transmissibility of this new non-conventional agent to other species and to humans through the consumption of infectious beef products. Since its emergence in UK cattle, BSE has been transmitted to other species, either naturally or experimentally. BSE has been transmitted naturally to cats probably through the consumption of contaminated beef-derived pet food (69 cases in the UK, one in Denmark and one in Ireland), and other wild feline species and antelopes in zoologic parks. Other species, such as mink in the United States, may also have been affected by the BSE agent. The BSE agent is transmissible experimentally by inoculation into the brain to mice, sheep, pigs and monkeys and by the oral route to sheep, goats and mice.

Several other European countries have reported cases of BSE: most of which were not among imported animals: the Republic of Ireland (188 cases), France (26 cases), Switzerland (231 cases) and Portugal (61 cases). Several other sporadic cases were reported among cows imported from the UK to Oman (two cases), the Falkland Islands (one case), Canada (one case), Denmark (one case), Germany (four cases) and Italy (two cases). BSE has been introduced into the cattle of these countries by importation of either infected animals [2] or of infected MBM [3]. Based on the incidence of BSE in the UK cattle, the number of BSE cases that should have occurred in the cattle of other European countries from the importation of infected cattle from the UK was estimated at 1668 cases [2]. So far a very small proportion of these cases have been officially reported. Therefore, BSE cannot be considered as a British problem alone, and potential exposure of humans to infectious beef may have occurred, although to a much lower extent than in the UK, in other countries; a European approach is therefore required.

Because of the occurrence of BSE, and to conduct epidemiologic studies on a larger sample, a surveillance and research project on CJD in five European countries (the UK, France, Germany, The Netherlands and Italy) was initiated in 1993 [4]. The goal of this project was to assess a possible change in the incidence of CJD in relation to the BSE epidemic and to develop etiologic studies of the risk factors associated with sporadic cases of CJD. The recognition of nvCJD in the UK in March 1996 about 10 years after the beginning of the BSE epidemic has changed dramatically the theoretical risk of transmission of BSE to humans through the consumption of infectious beef to a possible one [5]. At present, 15 confirmed cases have been identified in the UK and are characterized by a younger age at onset, a longer duration of illness and very specific unusual neuropathological lesions (PrP plaques) [5]. Some authors have questioned whether nvCJD has really emerged as a new entity. Although a detection bias cannot be ruled out completely, it is very unlikely that a progressive dementia affecting young adults that results in death within 6 to 24 months would have gone undetected for so long. In addition, the unusual neuropathologic features of nvCJD (PrP plaques) have not been found in all past CJD cases investigated in the UK and other European countries [5]. A case of nvCJD has also been reported in France, which suggests, according to the hypothesis of a link between nvCJD and BSE, that the potential impact of BSE on human health may not be limited to the UK.

The link between BSE and the nvCJD agent has not been formally scientifically demonstrated; however, we should already be thinking according to the hypothesis of a link between BSE and nvCJD, about the potential consequences on human health in the decades to come! The question should be dealt with in two parts: (1) is the BSE agent the same as the nvCJD one and (2) how does transmission to humans occur? Although no definitive answer can be given to question 1, the geographic and temporal correlation between nvCJD and BSE and recent scientific findings are strongly supportive of the BSE–CJD link. The scientific findings in favor of such a link include the similarities of the neuropathologic lesions observed on the brain of a monkey that developed TSE after experimental inoculation with BSE-contaminated brain [6] and the fact that the disease-related isoform of the prion protein of nvCJD has strain characteristics distinct from those of other types of CJD (called type 4) but which resemble those of BSE transmitted to mice, domestic cats and macaques [7]. Studies on the transmission of the nvCJD strain to non-transgenic mice and transgenic mice that express human PrP will probably shed new light on this debate within the next few months. Demonstration that the agent of BSE and nvCJD is the same will not prove the transmission to humans by the consumption of infectious beef but will be very much in favor of it. Additional epidemiologic evidence will be, however, necessary to demonstrate this route of transmission.

In contrast to scrapie in sheep, and based on current knowledge, the infectious tissues of BSE-infected animals are limited to the brain and the spinal cord. The risk of transmission to humans, should, in theory, be limited. However, brain and spinal cord tissues may have been introduced into meat prior to the 1988 and 1989 offal ban. As for the AIDS epidemic, one should distinguish BSE disease from BSE infection: because most cows infected with BSE are slaughtered for human consumption before BSE signs occur (the incubation period is about 5 years) most of the infected cows entered the human food chain before 1990. Under this assumption, Anderson et al [8] have estimated, through a very comprehensive modeling of the UK BSE epidemic (based on back calculation as for AIDS), that 903 000 animals had been infected of which 446 000 entered the food chain prior to the 1989 ban and 283 000 more before the end of 1995. Although these results indicate that humans have been widely exposed in the UK, nothing is known of the actual risk (probability) of transmission to humans. Therefore, no serious projection can be made until we know more about the temporal trend of nvCJD (the shape of the epidemiologic curve, the time interval for the number of cases of the nvCJD to be multiplied by 2, the distribution of the incubation period, etc.). Therefore, a risk analysis is very difficult or impossible because many of the necessary parameters, routes of transmission to humans, infective dose, probability of infection when exposed etc.) are unknown and will remain so for some time. In this context and following intense media attention in the UK, the president of the Royal Statistical Society answered the question of the predicted size of the nvCJD epidemic with the answer ‘zero to million’[9]. However, attempts were made to estimate the potential size of a future nvCJD epidemic based on the 14 first UK cases [10]. Although this study did not make any real prediction, it underlined several important points: first it is very much too soon to conclude that because of the limited number of confirmed nvCJD cases there will not be a large epidemic in the future; second, although the number of cases over the next years will provide valuable information, it could take a few more years before any firm indication will be available on the number of cases of nvCJD to expect [10]. In addition, the authors showed clearly that no serious prediction will be possible without improving the current estimates of the distribution of the incubation period of nvCJD [10].

In the countries other than those which participated in the European surveillance scheme, surveillance of CJD is very heterogenous and appears to be not very suitable for timely detection of any change in the incidence or CJD or the nvCJD [11]. Moreover, current data do not allow meaningful comparisons. This problem and the potential for a change in the incidence of CJD in the years to come stress the need for a European consolidated surveillance of CJD which should not be limited to five countries and should include non-EU countries such as Switzerland [11]. Because CJD and nvCJD are rare events, at least for the time being, a multicenter approach at the EU level will facilitate epidemiologic studies that could not be done in a single country. A similar European collaboration is necessary for the surveillance of BSE in other European countries. It has been shown recently that the BSE agent can be transmitted by the oral route to sheep with a disease and tissue infectivity distribution similar to the one observed in the natural scrapie [12]. Because the contamination of the ovine herd by the BSE agent through MBM is a plausible hypothesis, an outbreak of scrapie caused by the BSE agent could occur among sheep without being detected by the usual clinical and pathologic methods. Such an outbreak in sheep linked to the BSE agent would be much more difficult to control than an outbreak among cattle (it is assumed that its transmission among the sheep flock would be similar to the transmission of natural scrapie) and could add a new threat to humans. The surveillance of scrapie should therefore be widely reinforced to detect a possible occurrence in good time. New tests that allow the typing of the protease prion PrP (type 4) from tonsil biopsies in nvCJD patients [13] may be very helpful in the future to characterize TSE agents and monitor their occurrence in humans and animals.