Open session of the standing technical committee of the EUFMD- 2004

Appendix 42 Prospects for improved laboratory diagnosis of FMD using real-time RT-PCR Nigel Ferris*, Scott Reid, Donald King, Geoff Hutchings and Andrew Shaw Pirbright Laboratory, Institute for Animal Health, Ash Road, Woking, Surrey GU24 0NF, UK Abstract: Definitive diagnosis of FMD requires the detection of virus, antigen or genome in clinical material. The aim was to evaluate the performance of an automated real-time RT-PCR procedure for this purpose.

Vesicular epithelia from eighteen countries were examined by ELISA, VI and RT-PCR. Retrospective analysis by RT-PCR was also performed on available material of two sample subsets collected from ‘confirmed’ cases during the 2001 UK FMD outbreak : firstly, samples which were negative by both ELISA and VI and secondly, others which were negative by ELISA on epithelial suspension but positive by VI. There was broad agreement between RT-PCR and VI for 79% and VI/ELISA combined for 82% of the overseas epithelial samples tested. There were no false negative results obtained with RT-PCR since all samples assigned negative by RT-PCR were also negative by VI/ELISA. However, the RT-PCR was able to detect FMDV in an additional 18% of the samples tested. Additionally, there was good agreement between the RT-PCR and ELISA/VI for the UK outbreak samples save for a group of related virus isolates from Wales. These viruses had evidently evolved during the epidemic and had a nucleotide substitution in the RT-PCR probe site, which prevented detection by RT-PCR using the routine diagnostic probe.

The ELISA and VI are deficient for specimens of poor quality where concentrations of infectious FMDV may either be low or absent. The features that influence sample quality appear to be less important for the RT-PCR as it can detect a small fragment of FMDV genomic RNA, not just live virus. Real-time RT-PCR provides an extremely sensitive and rapid procedure that contributes to improved laboratory diagnosis of FMD. However, the failure to detect the mutant FMDVs from the UK 2001 epidemic illustrates the importance of constant monitoring of representative field FMDV strains by nucleotide sequencing to ensure that the primers/probe set selected for the diagnostic RT-PCR is fit for purpose.

Introduction: Control of outbreaks of foot-and-mouth disease (FMD) is dependent upon a system of monitoring and early detection, which requires basic familiarity with clinical signs and the ability to characterise the strain of virus responsible by laboratory tests. Definitive diagnosis of FMD requires the detection of virus, antigen or genome in clinical material. Ideally, the sample of choice should be vesicular epithelium from clinically affected animals since, during the acute stage of the disease, it is rich in virus. The World Reference Laboratory (WRL) for FMD typically receives between 400 and 700 samples annually from overseas countries (Ferris and Donaldson, 1992; Table I), including other sample types besides epithelia, e.g. epithelial suspensions, cell culture antigens, blood, throat swab (probang) and milk samples. For almost twenty years, the WRL for FMD has used an indirect, sandwich enzyme-linked immunosorbent assay (ELISA) (Roeder and Le Blanc Smith, 1987; Ferris and Dawson, 1988; OIE, 2004) to identify FMDV. However, the ELISA is not 100% sensitive.

Consequently, suspensions of each specimen are also propagated in sensitive cell cultures (Ferris and Dawson, 1988) and the specificity of any isolated virus confirmed by the ELISA. Whilst such virus isolation (VI) methods are highly sensitive, they require four days before a negative result can be concluded (and reported as ‘no virus detected’ [NVD]). It is evident from Table I that FMDV antigen cannot be detected by ELISA and VI in approximately half the submitted samples. This has given cause for concern as to the efficiency of sample collection and dispatch and also with respect to the adequacy of the laboratory test procedures employed for their examination.

In emergencies, speed of diagnosis (clinical and laboratory confirmation) is of paramount importance to control spread and eradicate disease. Approximately, 90% of positive epithelial samples received during the 2001 UK FMD outbreak were so defined by the antigen ELISA on prepared suspensions, the remainder being serotyped after amplification and isolation of virus following cell culture passage.

Negatives could only be classified following double passage in cell culture, which took 4 to 5 days.

Consequently, the introduction by the UK Government towards the end of March of a 24/48 hour culling policy (all animals to be slaughtered on infected premises within 24 hours of diagnosis and those on neighbouring premises within 48 hours) meant that confirmation of disease was subsequently made by clinical judgement alone. This policy, although playing a critical role in controlling disease, caused huge controversy and provoked much debate on the likelihood of many animals being slaughtered unnecessarily, fuelled by the finding that neither virus nor antibody could be detected in samples received from many of the confirmed cases.

261