Volume 8: Variant CJD
2. History of CJD surveillance up to 1990
Surveillance in the UK before the emergence of BSE
Concern about exposure to scrapie and risk of CJD
Human growth hormone-derived CJD

2.3 The first detailed study of the epidemiology of CJD in the UK was published by Professor Bryan Matthews, University Department of Neurology, Oxford in 1975. ¹ Experimental transmission of CJD to primates, ² and therefore the possibility of natural transmission in humans, prompted this project. Professor Matthews's study aimed to detect any evidence of natural transmission of CJD and to report on the incidence of CJD in England and Wales between 1964 and 1973. Neuropathologists were asked to notify Professor Matthews of cases of CJD and possibly related disorders, which had been verified by histological examination. ³ Information was obtained from hospital notes, questionnaires, and interviews with near relatives.

2.4 Forty-six cases of CJD were identified for the relevant time period, 30 of whom were women and 16 men. The average age of onset was 57 with an age range from 34 to 71. Using the 1971 census figures, the annual incidence of CJD over the period of the study was 0.09 per million.

2.5 The results of the search for sources of natural transmission were largely negative. Two cases were identified that had undergone intracranial surgery (ie, surgery within the skull) and further investigation uncovered 'strong circumstantial evidence' of iatrogenic transmission. ⁴ Another two cases had had some contact with ferrets, and it was known that transmissible mink encephalopathy (TME) was transmissible to ferrets, ⁵ but this was later thought to be insignificant as none of the ferrets kept by the cases were reported to have TSEs. ⁶ In addition, two small clusters of cases were detected, one in a small rural community in the Midlands and one in eastern England. Contact between each of the cases in the Midlands could not be established, but further study of the cases in eastern England suggested that iatrogenic transmission or dental procedures might have been involved. ⁷

2.6 Professor Matthews concluded that natural transmission from one 'overt' case to another case could not account for the occurrence of CJD within the population. He suggested that transmission from non-fatal cases that might remain infective after recovery was a possibility. (There are, however, no known forms of non-fatal CJD.) ⁸

Concern about exposure to scrapie and risk of CJD

2.7 In the 1970s, concern was raised by the ARC about the widespread occurrence of scrapie and the similarities between this disease and diseases of the human central nervous system, such as CJD. ⁹ Further concern about the risks to humans from scrapie arose in 1976 when the US Department of Agriculture decided that the carcasses of sheep and goats affected with, or exposed to, scrapie should not be used for human or animal food. ¹⁰

2.8 These concerns prompted MRC discussions, in 1978, on the value and feasibility of further CJD surveillance, including identification of any increased risk posed by contact with scrapie. ¹¹ As a result of these discussions, an MRC Working Group on CJD was set up later that year chaired by Professor (now Lord) Walton, Professor of Neurology at the University of Newcastle. ¹² Other members of this group included Dr Katherine Levy of the MRC and Professor Ingrid Allen, Professor of Neuropathology in Queen's University, Belfast. The MRC Working Group on CJD agreed that retrospective analysis of CJD cases from the period 1970-79 and prospective analysis of CJD cases were important. ¹³

2.9 Consequently, Professor Matthews was awarded several grants between February 1979 and February 1986, in order to conduct the studies discussed and to employ Dr Robert Will as an assistant. ¹⁴

2.10 The main results from Dr Will's and Professor Matthews's work are found in the following three studies:

The retrospective study (1970-79)

2.11 This study involved the analysis of data from Professor Matthews and the Office of Population Censuses and Surveys (OPCS) during the period 1970-79. The main findings were:

1. The study yielded 121 confirmed and 31 probable cases of CJD deaths. The average annual incidence was 0.31 per million. ¹⁵
2. One striking finding was the sex ratio of 1.68:1 with 99 female and 59 male cases. The significance of this finding was unknown but a similar ratio was emerging in a case-control study ¹⁶ being carried out at the time of publication of this study (see paragraph 2.16).
3. The possibility of case-to-case transmission was investigated by examining the 'space-time' clustering of cases. If transmission is occurring in rare diseases, it would be expected that disease onset in cases which are close in space might occur at similar times, provided that the incubation period between infection and disease onset is short. In this study, no significant difference was identified between observed and expected numbers of those dying within 1 to 3 years of each other and within 20 km. However, only the address at the date of onset of disease had been recorded for each case.
4. It was highlighted that 28 per cent of cases had a past history of major surgery, but no association between these cases could be made which might suggest iatrogenic transmission. However, it was noted that only a case-control study could say if there was an increased risk associated with major surgery and this was not conducted.
5. Occupational data were limited to occupation at date of death with the only noteworthy result being the absence of the disease among farm workers, ie, no increased risk was detected with possible increased exposure to scrapie.

The prospective study (1980-84)

2.12 This study involved identification of CJD cases and comparison of each case with two controls, one with a neurological condition impossible to misdiagnose as CJD, and the other with a general 'medical' complaint. ¹⁷ These controls were matched by age, sex and hospital (not community-based controls). Significant findings were:

1. According to the clinical criteria used for diagnosis in the study, there were 83 definite, 32 probable and 25 possible CJD cases identified. The annual incidence of CJD was 0.49 per million. This higher incidence compared to the retrospective study where 0.31 cases per million were observed, was attributed to under-ascertainment in the earlier study.
2. As with the retrospective study, there was a marked preponderance of females with CJD. This sex difference was significant even allowing for the different age distributions of the female and male populations. However, this finding had not been observed in a similar study carried out in France ¹⁸ and again no explanation was offered.
3. There was no marked difference between CJD patients and controls with respect to surgical history, occupation or animal contact.

2.13 The possibility of carrying out tests for CJD susceptibility in patients was considered during discussions on the design of the prospective study by members of the Working Group on CJD in 1978. ¹⁹ For example, it was thought valuable to determine the HLA status of patients. Human Leukocyte Antigens (HLAs) are particular proteins present on the surface of white blood cells, and different types of HLA are associated with specific diseases.

2.14 In the eventuality, it was considered that there was sparse evidence for the involvement of susceptibility factors, and so general genetic screening would not be profitable. ²⁰ Whilst HLA typing was considered important, considerable difficulty was encountered in finding laboratories which were prepared to handle blood samples from CJD patients in view of the potential hazard of infection. Consequently, it does not appear that this work was ever carried out.

Amalgamated study of CJD (1970-84)

2.15 This study was an amalgamation of the data from both of the previous epidemiological investigations to cover the 15 years between 1970 and 1984. ²¹ Two age-and sex-matched controls were selected for 72 of the 122 cases diagnosed in the period 1980-84. Life histories of places of residence for all subjects were obtained.

2.16 The study identified:

1. 267 patients diagnosed as definite or probable cases of CJD. No evidence of space-time clustering (dates and places of onset) was found.
2. A relative excess of cases in women was noted again in the age range between 60 and 74 years, although the rates for both sexes appeared similar for those people under 60 years of age. The authors were unable to provide an explanation for this observation.

Human growth hormone-derived CJD

2.17 The studies described above did not identify an increased risk of CJD with exposure to scrapie nor any evidence of an iatrogenic cause. However, in 1985 it became apparent that several children treated with human growth hormone (hGH) extracted from cadavers in the United Kingdom had later developed CJD. ²² Of 2,000 children treated between 1959 and 1985, approximately 1-1.5 per cent have died from CJD as a result of the treatment. The methods used to prepare the hGH varied over the years, but generally involved maceration of pooled cadaveric pituitary glands ²³ in a blender, after storage in acetone. ²⁴ It became apparent that the procedures used had not inactivated the CJD agent.

2.18 The resistance of the scrapie agent to standard disinfection procedures was well documented. Experiments since the late 1950s had shown that it was resistant to treatment with chloroform, phenol, ether, and heating to 100°C. ²⁵ Likewise, the transmission of human spongiform encephalopathies from person to person was well documented in kuru. ²⁶ Thus, it was accepted that the cause of hGH-derived CJD was the inclusion and subsequent failure of inactivation of agent from CJD infected pituitaries. ²⁷ The link between CJD and hGH was considered by the Committee on Safety of Medicines (CSM), of which Dr David Tyrrell was a member, and was the subject of a High Court Action in 1996. ²⁸

2.19 In December 1995, Dr Robert Milner (Department of Obstetrics, Catholic University of Leuven) put a paper forward to the Spongiform Encephalopathy Advisory Committee (SEAC), proposing that CJD development in recipients of hGH might have been the result of infection with sheep scrapie agent. ²⁹ Evidence for this was derived from reports of the visits by the Medicines Inspectorate to the Pituitary Hormone Laboratory in Cambridge in September 1975 and January 1978. It appeared that equipment used to mince human pituitaries prior to extraction of growth hormone had also been used for the maceration of animal pituitaries potentially exposing the human growth hormone to scrapie agent.

2.20 The theory was considered by SEAC though several difficulties were identified. Firstly, there was no evidence of scrapie ever having crossed the species barrier to humans and secondly, pituitary hormones prepared in other countries where scrapie was not endemic had also caused CJD in recipients. Thus, transmission from a human source remained a more likely explanation for the emergence of these cases. ³⁰ SEAC does not appear, however, to have considered the possibility that CJD could have been transmitted through contaminated animal hormones to cattle. The CJD agent could have contaminated animal pituitaries in the laboratory practices described above. Unfortunately, laboratory inspection reports for the period up to 1975 are no longer available, so it is not possible to ascertain exactly what the procedures used were.