4. INFECTIVITY IN CNS TISSUE
A base assumption for the study is that the BSE infective agent can cause new variant Creutzfeldt Jakob disease (vCJD) in humans. This hypothesis has been strengthened by the results of research that has shown that the agents which cause BSE and vCJD are the same (Collinge et al, 1997; Bruce et al, 1997).
The infectivity (i.e. the potential to cause infection) of tissue from cattle with BSE is expressed in terms of its Infectious Dose 50 (ID50) value. This is the dose (i.e. the quantity which each person would need to consume) to cause infection of 50% of the exposed population. This term acknowledges that some people may become infected from much smaller doses, while others may be uninfected after consuming much larger doses.
The infectivity of the BSE agent has been considered in detail by the Scientific Steering Committee (SSC) of the European Commission and their assessment presented in their opinion adopted at their meeting on the 13-14 April 2000 “Oral Exposure of Humans to the BSE Agent: Infective Dose and Species Barrier”. This opinion is used as the basis for this risk assessment.
The estimates of infectivity levels in this section all relate to infectivity in central nervous system (CNS) tissues, i.e. the brain and spinal cord. DRG are part of the peripheral nervous system and the level of infectivity present may well be less than that in the CNS. The Pathogenesis experiment referred to previously did not provide data on infectivity levels. It is therefore assumed that the infectivity in DRG would be the same as that in the brain or spinal cord. Infectivity has not been identified in other peripheral nervous system tissues.
4.1 Infectious dose for cattle
The SSC concluded that the various approaches to assessing the infectivity from a clinically infected brain yielded a range of values from 101 to 103 cattle oral ID50/g. They noted that the higher value may represent a worst case scenario if the oral route is more efficient than data suggests and a particularly high titre of infected brain is sampled. They conclude that such a high dose cannot be ruled out. The lower value is based in part on the results of the attack rate experiment carried out by the UK MAFF. It is noted that this experiment is incomplete and that it is not possible to obtain a final value for the infectious dose. The SSC gives some weight to the calculations of Diringer (1999) using the results of published and peer reviewed experiments. This results in an estimated infectious dose of 50 cattle oral ID50/g.
From this data it is proposed to adopt a distribution of values ranging from 10 to 103 cattle oral ID50/g with a best estimate value of 50 cattle oral ID50/g.
The infectivity of BSE for humans is believed to be lower than in cattle due to the species barrier. The species barrier in this context is defined as the factor by which the effective infectivity in one species is reduced when given to a second species. Thus, if the cattle-human species barrier was 100, it would mean that 100 times more infective material would be required to infect a man than a bovine.
In their opinion, the SSC concluded that the size of the species barrier between BSE in ruminants and BSE in humans (vCJD) is not known. They considered that a worst case scenario considering no (=1) species barrier should be included, although available evidence indicates that values greater than one are likely to be more realistic. They recommended that, until more scientific data are available, for risk assessments of human exposure to potentially BSE infected products, a species barrier of about 1 should be considered as a worst case scenario and that the range from 104 to 101 be considered. This supports the assumptions made by DNV in previous risk assessments in which the species barrier was represented as a distribution using values of 10, 100, 1000 and 10,000 with equal probabilities, and a 1% probability of it being 1 (DNV, 1997 a & b). It is proposed to use the same distribution in this assessment.
4.3 Infectious Dose for Humans
The infectivity density of CNS tissue from an infected bovine to humans is obtained from the product of the infectious dose for cattle and the cattle human species barrier. Combining the distributions given above in the probabilistic assessment, results in an estimate of the median value of the infectivity density for humans of 0.26 human oral ID50 per gram, with a 95 percentile range of 0.002 to 44.