Real-Time PCR (RT-PCR).

The front-end technology for extracting nucleic acids is the same as in regular PCR the only major difference with this technology is that once the amplification has taken place instead of having to take the product and run it on a gel, the gel phase of the testing has been eliminated, which provides much faster turn-around times for results. The detection of the amplified sequence is done using a computer program and melting point curves. . It is well beyond the scope of this article to describe fully how the system works but in general it works as follows: once the nucleic acids have been extracted and prepared from a clinical sample, a sequence template, for the pathogen to be detected is added.

The technology of cycles of heating and cooling are still used to allow amplification to occur. What is different is that as the cycling is in progress the formation of copies of the sequence in question is monitored and once the machine has sense that a sufficient amount of similar product has being generated it will indicate a positive result.

The complementary sequences are in a double-stranded state and when they are heated they will separate at a specific melting point, unique to that sequence. The computer program then compares the melting point obtained from the sequence generated from the clinical sample with the known melting points of the double-stranded sequence that was expected to be detected and confirms that the amplified sequence obtained is in fact the one that was being looked for.

While this is amazing technology and also much faster than regular PCR there are of course still issues. The technology is expensive for a start but perhaps more importantly the technologist needs to gain confidence in the technology itself -- there are no gels or any other backup to actually see what is going on. What is seen is basically a series of computer-generated graphs (see examples) and the technology is essentially a "black box". Once the technologists gain confidence however, the results of this new technology are very valuable from clinical point of view.

Apart from being able to make a diagnosis with this technology there are two other features that make it appealing from a clinical point of view, one of which we currently make use of, and one feature that we do not, as yet, use. The two features are the ability to measure viral load, which we do use, the other is for detecting sequence mutations.

RT - PCR Viral load.

Because the amplification of the sequence in question is monitored and graphed it is possible to extrapolate backwards, using a series of known standards, to determine how many sequence copies were in the original sample. Very basically it relates to the number of sequences initially present and the number of cycles it takes to amplify the sequences to a sufficient level to be detected. Obviously if there are large number of sequences in the original sample then the time to detection will be faster than if there are only a few sequences present in the original sample. A standard curve can be constructed using known copy numbers and this can then be compared with the sample result. The report then comes out as a copy number per unit of blood. The value of this is in infections were antiviral chemotherapy is being used. This technique is most commonly used in HIV treatment or in our case the treatment of CMV or EBVinfections in transplant patients. In Ontario only public-health labs are licensed to do HIV viral loads but we are able to do viral load for CMV and also for EB virus. The CMV viral load test has now supplanted the CMV antigenemia test that we used to provide.

It works something along these lines: A patient, post transplant, is suspected as having a CMV infection let's say. A blood sample is taken and assayed by RT PCR for CMV. Not only the presence of CMV but the copy number can be calculated. Let’s say this turns that to be 10,000 copies. The patient is started on antiviral chemotherapy and a few days later the assay is performed again. Let’s say the copy numbers is now 1000 copies. Clearly this reduction indicates that the therapy is working. If however the patient deteriorates a few days later another sample could be sent. Let's say that this time the copy number is now 20,000. This signifies that the antiviral agent is no longer working and it may be because the virus has mutated, developing a resistance to the antiviral. It is rarely is clearly cut it is clear cut as this example but at least the principle of using viral load to follow to monitor antiviral therapy can be appreciated.

Sometimes viral load is used to monitor patients post transplant to see if they are developing a infection such as CMV, in this case, antivirals can be started as soon as the copy number begins to increase. CMV antigenemia used to be used (and still is in many centers where RT-PCR is not available) to monitor CMV disease activity. It is important to understand that the correlation between RT-PCR viral load and CMV antigenemia is poor, particularly in situations where low copy numbers are present. Clinicians familiar with using CMV antigenemia results need to become experienced in using the RT- PCR results before they feel comfortable. Remember that lab results still need to be interpreted in light of the clinical picture.

The other advantage of RT-PCR technology is that when a mutation is present in the sequence (such as in the development of resistance) the sequence still anneals with the template and the melting point falls within the acceptable range but because the melting point is so exquisitely sensitive even a one nucleotide difference in the sequences can be detected. This then allows for the recovery of the mutated sequence which can be sequenced (using different technology) to determine exactly what nucleic acid base change has occurred as the result of the mutation. This is a very powerful research tool but is not used much at the present time in routine clinical applications.

Some issues with the technology.
Despite the accuracy, sensitivity and specificity of this methodology there are still some concerns that need to be appreciated. The first is obviously the expense, an RT-PCR test costs roughly $150 (a blood sugar costs about 1 to 2 cents). There must be a reasonable clinical suspicion that the person has the disease. In our institution the most common request is for the detection of herpes simplex virus in cases of confusion or ?encephalitis. We detect HSV in less than 5% of the tests requested, meaning that clinically we are wrong 95% of the time. It should be possible to bring the positivity rate up to about 30% using good clinical judgment and being more selective in ordering the test. Too often the test is ordered immediately the patient is seen but within a few hours it is already clinically determined that the etiology of the condition is not HSV however the HSV test is not canceled and valuable laboratory resources are wasted. Please make the effort to inform the lab as soon as possible if the test is not required. The second issue is that patient samples may contain nonspecific inhibitors that make the test invalid. When this happens the lab reruns the sample after diluting it, in the hope that the dilution will reduce the inhibitory effect and allow for a result to be obtained. If this does not work then the result is reported as indeterminate and it means that other diagnostic tests need to be used to confirm the diagnosis.

Lastly, none of these nucleic acid detection tests are licensed for clinical use as yet. We always have to include a comment on the report that the test is a research test only and should not be used for clinical management. Everyone recognizes that these tests are in fact the gold standard, eventually the licensing bodies would get around to approving them but until that time the rider on the report will still appear.

Reference. Clinical Virology Manual. Steven Specter, Richard L. Hodinka, Stephen A. Young. 3rd Edition. ASM Press. 2000.

The tests currently available by RT-PCR are:

Quantitative EB virus.
Quantitative CMV.
Qualitative West Nile Virus (Transplant donors only).
Qualitative HSV 1 &2.