The human prion diseases or transmissible spongiform encephalopathies (TSEs) are a group of fatal neurodegenerative disorders believed to be caused by a post-translational conformational change in cellular prion protein from its soluble form (PrPC) to a pathogenic protease resistant isoform PrPSc . The most common of these is sporadic Creutzfeldt-Jakob disease (sCJD) but a variant form of CJD (vCJD) was identified in the UK in 1996  and has been linked to human infection by the bovine spongiform encephalopathy (BSE) agent. Existing clinical tests for Creutzfeldt-Jakob disease (CJD) can establish a diagnosis of probable variant or probable sporadic CJD during the clinical phase of disease, but a definitive diagnosis depends on post-mortem examination of the brain. Moreover there is currently no practical way of determining whether an individual who is not manifesting symptoms is incubating the disease.
The presence of PrPSc in the peripheral tissues of patients with vCJD [3, 4] and experimental transmissions of BSE and natural scrapie between sheep by blood transfusion [5, 6] raised the possibility that iatrogenic transmission in humans by blood transfusion could occur. This risk has unfortunately been confirmed by recent reports of transmission of vCJD by blood transfusion [7–10]. The annual number of deaths from variant CJD has been declining since a first peak in 2000. However, the prevalence of infected individuals in the general UK population, as judged by a retrospective tonsil and appendix tissue study, appears to be higher than would be expected from the mortality rates, pointing to substantial numbers of sub-clinically infected individuals in the general population [11, 12]. To date all reported clinical cases of vCJD have been homozygous for methionine at codon 129 of the prion protein gene (PRNP), and there has been uncertainty over the susceptibility of the heterozygous and valine homozygous genetic sub-groups. There are now good reasons to believe that all PRNP codon 129 genotypic groups are susceptible to infection [9, 13] and it is possible that heterozygous and valine homozygous individuals may exhibit prolonged incubation periods or might remain in a sub-clinical state throughout life. In the absence of a screening test for vCJD these individuals could be a major source of iatrogenic transmission. Current methods that might identify these pre- or sub-clinical individuals, such as lymphoreticular biopsy, are highly invasive and clearly impractical as screening measures for blood donations. A screening assay that could be applied to a routine blood sample is urgently required.
Although there is a presumption that PrPSc is the infectious agent, the association of infectivity with blood is poorly understood. There is growing evidence from studies in mice and hamsters that initial TSE replication occurs in lymphoreticular tissues [14, 15] prior to invasion of the central nervous system (CNS) via the sympathetic nervous system [16–19]. Recent reports of the transmission of natural scrapie and experimental BSE between sheep by whole blood and buffy coat transfusion support the theory that infectivity is associated with, but not restricted to, the white cell component . Given that levels of detectable PrPSc and infectivity in peripheral lymphoreticular tissues such as spleen and tonsil in patients with vCJD are 2–3 logs lower than levels detected in the CNS [4, 21], it is likely that PrPSc is present at very low concentrations in peripheral blood. Attempts to detect PrPSc in human buffy coat by Western blot have thus far proven unsuccessful . Intracerebral inoculation of human buffy coat from clinical vCJD cases into susceptible mouse models has also failed to demonstrate infectivity , although this could be a reflection of species barrier and/or the small numbers of animals used in addition to the limited volume of inoculum that can be delivered intracerebrally. Because of its evident very low level, pre-mortem tests designed for the detection of any PrPSc present in blood/body fluids would require a high level of sensitivity, probably several logs greater than those diagnostic tests already approved by the European Commission and in place for post mortem TSE disease confirmation in slaughterhouse cattle and sheep. Many of these tests rely upon proteolytic treatments to digest PrPC and may lack the sensitivity required for blood screening as they may remove protease sensitive isoforms of disease associated prion protein. There are a number of emerging diagnostic tests in development for pre-mortem diagnosis which may possibly be capable of detecting PrPSc in peripheral blood, amongst which the protein misfolding cyclic amplification (PMCA) appears particularly promising [23, 24], but none of these has yet been validated in extensive study of human blood samples. These have recently been reviewed by Brown .
The difficulties of achieving the sensitivity needed for the detection of PrPSc has shifted efforts to an alternative approach of screening for biomarkers other than prion protein. Potential candidates assays have been recently reviewed by Parveen et al. . However most lack the specificity and sensitivity to be useful as diagnostic screening tests and would need to be employed with an array of confirmatory assays. At present many do not have data on studies of CJD in humans. Despite this there exist a small number of promising alternate approaches. Amongst these, emerging spectroscopic methods, for example the use of Fourier transform infrared spectroscopy (FT-IR) as a test for BSE in cattle or preclinical scrapie in hamsters, have been described [27, 28]. Analysis of blood serum by FT-IR supported by artificial neural network or multivariable pattern recognition analysis allowed the sensitive and specific discrimination between healthy and infected animals [28, 29]. In this paper we report on the development and assessment of atomic dielectric resonance (ADR) spectroscopic techniques in an analysis of clinical blood samples from patients with CJD. This also employs Fourier transformation and complex pattern analysis.
ADR is a new technology exploiting novel properties of resonance of radio waves from X-band and C-band radar developed for aerospace and ground remote sensing, imaging, and materials detection and recognition through surface boundaries. This technology has been recently described in a patent . ADR spectroscopy (ADRS) exploits dielectric properties of materials which determine reflectivity and penetration of radio/micro waves, allowing the use of electromagnetic energy as a means of remotely detecting and measuring target characteristics. Different materials have unique energy absorption and reflection properties and provide spectra in the broad range of frequencies employed that provide signatures for materials recognition and matching (typecasting). ADRS has been developed for a broad range of applications predominantly in geophysical, engineering or security fields which have not been published because of commercial or other sensitivities. Several small scale biological laboratory or clinical (patient or tissue) studies have indicated potential utility of ADR in biomedical investigation both in subsurface imaging and in materials characterisation, and the apparatus has been approved as a medical device for use in human subject studies as well as in the laboratory. However, interest in application of ADR in this area is novel and as yet no formal studies in a biomedical context have been published. ADR employs very low energy emission, and is a reagent free non-destructive non-consuming test. Unlike infrared spectroscopy, there is no requirement for drying or other special treatment of test materials. Preliminary (unpublished) studies by us of histology slides (remaining from a separate histology study) of brain sections from experimental scrapie infected mice and uninfected controls illustrated that it was possible to clearly distinguish scrapie infected mice at a range of stages of scrapie infection from normal controls by match ranking of ADR spectra taken from these slides . Results are presented here which demonstrate how characteristic ADR signatures can be exploited to discriminate between the blood of patients with clinical sCJD, those with clinical vCJD, those from non-CJD neurological control patients and those from healthy adult (blood donor) controls.