Forum Post : Real-time RT-PCR tests the new gold standard: a review of a Defra-funded investigation
Real-time RT-PCR tests the new gold standard
An investigation, funded by DEFRA, of two real-time RT-PCR assays: "these rRT-PCR assays offer superior sensitivity over established diagnostic tests."
Review by Mary Marshall, from a stakeholders perspective. We invite comments from stakeholders, including livestock keepers, vets and scientists.
An investigation, funded by DEFRA, of two real-time RT-PCR assays:
Donald P King, Nigel P Ferris, Andrew E Shaw, Scott M Reid, Geoff H Hutchings, Angelica C Giuffre, John M Robida, Johnny D Callahan, William M Nelson, Tammy R Beckham. (2006). Detection of foot-and-mouth disease virus: comparative diagnostic sensitivity of two independent real-time reverse transcription-polymerase chain reaction assays. J. Vet Diagn Invest. 18 (1):93-97.
A joint study by US and UK scientists has concluded that real-time RT-PCR assays are the new gold standard for foot and mouth disease virus (FMDV) detection and control. The laboratory work was performed during 2004 by staff from the Foreign Animal Disease Diagnostic Laboratory (FADDL, USDA Animal Plant Health Inspection Service, Plum Island Animal Disease Center, Greenport, New York, USA) and the Institute for Animal Health (IAH, Pirbright, Woking, Surrey, UK), using samples of epithelial tissue submitted to the World Reference Laboratory for FMD (IAH, Pirbright) from suspect FMDV cases during a 40-year period (1964-2004) from 69 countries.
The investigation, funded by DEFRA, compared in parallel the ability of two real-time (r) RT-PCR assays to detect a diverse selection of FMDV isolates: the IAH Pirbright RT-PCR (5UTR) and the Vet Alert TM [Tetracore] RT-PCR assay (3D). The latter assay was devised by Tetracore and USDA Agricultural Research Service scientists in 2000.
The epithelial tissues had been previously assayed for FMDV by using the currently accepted gold standard tests: virus isolation (VI) and the antigen-enzyme-linked immunosorbent assay (Ag-ELISA). The archived samples used in the study covered all 7 FMDV serotypes: O, A, Asia 1, C, SAT 1, SAT 2 and SAT 3, and also included samples in which no FMDV had been detected by VI or Ag-ELISA.
Comparison of the two assays.
When all the samples were analyzed together, this analysis showed that the overall difference between the 2 assays was not significant. However, there were significant differences between the 2 assays at the serotype level. "Ten serotype O viruses were missed by the 5'UTR [Pirbright] assay. These consisted of closely related isolates collected from Wales during the 2001 epidemic in the United Kingdom.
In other words, the Pirbright PCR test does not now and would not in 2001 have identified all the FMD viruses found in the UK 2001 epidemic. The Tetracore test, on the other hand, did identify all of these viruses. It should be noted that the UK Chief Scientist, Dr. David King, was fully informed of this technology and it was demonstrated to him by an expert US group in 2001 (personal communication, Roger Breeze). While it might not be productive now to ask what might have been different had this test been used in 2001 to rapidly identify infected animals, it is important that this technology should play a central role in planning the detection and control of any future outbreak.
The results [of this investigation] show that rRT-PCRs can detect FMDV in samples that are negative by VI and Ag-ELISA and probably reflect that molecular methods have high analytical sensitivity and will detect FMDV even if the samples are of poor quality and do not contain live virus.
The authors conclude that the results emphasize the importance of creating depth and diversity in assay systems used to detect economically important disease such as FMD (and CSF and Avian Influenza). The study suggests that using both of these rRT-PCR assays in concert would decrease the likelihood of a false-negative result secondary to sequence variability and add additional validity to the final diagnosis. In other words, using both of these tests would eliminate any reasonable doubts, which would be of most significance in identifying an index case.
Parallel and multiplex rRT-PCRs exploiting these improved assay characteristics are currently under development. However even when used independently, these rRT-PCR assays offer superior sensitivity over established diagnostic tests. This report provides further confidence in the use of this method for routine diagnosis of FMD and re-emphasizes the effectiveness of rRT-PCR as a diagnostic tool for routine FMD detection and control.
In a submission to the Royal Society inquiry, Roger Breeze wrote: Fortunately, we were able to take the [USDA/Tetracore rRT-PCR] devices and test system into the field in Uruguay in November 2001, where they performed splendidly on farm in a remote area. I must re-emphasize that the ARS test is intended to be used as an on-farm test (although it can be a central laboratory test) because time and timeliness are the critical issues in stamping out an epidemic. ( ttp://www.royalsoc.ac.uk/inquiry/388.pdf ).
The conclusion must be that real-time RT-PCR tests are the new gold standard and should be routinely used for rapid detection and control of FMD, in the lab and, more importantly, in the field. The USDA/Tetracore test is available for use on a portable device near the site of a suspected infection. Why then, is rRT-PCR absent from DEFRA FMD contingency plans?
Please see the related forum discussion: Can real-time RT-PCR be used for a different approach to eradicate and control CSF? (http://www.fmd-and-csf-action.org/forums/csf ).
We invite your comments on this paper.
In a personal communication, Roger Breeze wrote:
When Dr. Dan Rocks group (which had not worked on FMDV) began to develop the real time FMDV PCR assay with Tetracore in 1999/2000, they made the surprising observation that at that time there was not a complete genetic sequence of even one representative from each of the seven FMDV serotypes in the literature. In order to acquire the intellectual basis for rational scientific design of a real time PCR test, Rocks group sequenced all of the 103 FMDVs then stored at Plum Island: this work was an instant turning point in the history of FMD research (Carillo, C. et al, Comparative genetics of foot-and-mouth-disease virus, Journal of Virology, 79, 6487-6504, 2005). It was a turning point because it demonstrated the means to sequence FMD viruses very quickly so the task was trivialized and the bioinformatics capacity to compare many different genetic sequences of the same organism simultaneously, to look for similarities and differences. For the first time, we were able to make definitive statements about FMDVs based upon their complete genetic sequences.
The world does not need a farrago of real time PCR tests for FMDV it needs one that works and has a rational design. Fortunately, that test has been available from Tetracore for some years. The next generation of test whether based on PCR or future technologies - will not be discovered by serendipitously rooting through old case files but be designed as a result of knowing the complete genetic sequences of all the FMDVs collected over the generations at FMD World Reference Centers and in other repositories. Exploiting the genetic potential long languishing in the FMD reference centers will be of enormous benefit to public and private sector research and enterprise all over the world, particularly for those with radically new ideas and technologies that do not have access to live viruses. Thats why the data from Carillo et al are in the open literature.
Do I think that things would have been different in 2001 if the USDA/Tetracore test had been used in the UK then? My answer is no. Of course, Fred Brown and a US team demonstrated the technology at the time to the UK Chief Scientist, Dr. David King, in his office and offered him the data. But he had no interest in new technology; I think he was already committed to his course of action.
Would it be different in 2006? The answer sadly is still no. To understand why, DEFRA would need to answer the question: What will you do differently in the future now that you have re-discovered the fact that RT-PCR is superior to the technology you have been using to detect FMD virus? I don't want to guess what the DEFRA answer might be or to put words into DEFRAs mouth.
How does DEFRA plan to use PCR?
Will every herd destined for slaughter in an FMD outbreak be tested by PCR to confirm that it is actually infected before being killed? Or will slaughter still depend upon being in the wrong place at the wrong time (that is, within 3 miles of an FMD herd confirmed by PCR)?
How the test will be used has always been the critical issue, not whether it is better than another type of test: that was a red herring that has nevertheless delayed answering the critical question for 5 years.
Here's an example of how I think it could be used. Let's say FMD virus is detected on an index farm in Cheshire. Immediately (that day) we now want to know if there are any other farms in Cheshire with infected animals (known or unknown to their owners). Cheshire is a major dairy county and every day farms send their milk to processing plants. FMD virus is present in the milk of infected cows. Immediately (the same day) FMD is detected in Cheshire, I would dispatch teams to every plant processing milk from Cheshire and they would test every tanker load of milk as it arrives for FMD virus using the PCR test on portable machines set up in a van or tent - in this way one can test continuously throughout the outbreak every herd in Cheshire without ever going on the farm - the sources of each tanker of milk are known.
I see PCR being deployed extensively and as close as possible to the problem to get close to real time continuous information. Will this take a lot of machines, manpower, training and preparedness? Of course it will. Can it be done by taking samples from the milk processing plants to Pirbright? No.