The RT-PCR “test”

 

 A common method used when studying Deoxyribonucleic Acid (DNA) is Polymerase Chain Reaction (PCR). This is a method that repeatedly duplicates strands of DNA, like using a photocopier to amplify the quantity present in a sample to a level where there is enough material to work reliably with. The method relies on mixing the DNA with an enzyme to aid in the “chain reaction part”, specifically DNA polymerase, “primer” fragments of DNA (fragments characteristic of the DNA we want to duplicate), “probe” fragments (fragments of DNA that react to an excitation source by fluorescing, the glow of which can be captured by a photo detector), and DNA nucleotides, or “building blocks”. The latter consist of 4 specific nucleotides: adenine, cytosine, guanine, and thymine, typically referred to by the letters A, C. G and T. The DNA nucleotides combine in pairs to form the ladder rungs of the DNA double helix. Each pair is a specific sequence as for example G only pairs with C and A with T.

But a virus does not contain DNA but instead a single strand of Ribonucleic Acid (RNA). This is where the RT (Reverse Transcription) comes in. By using another enzyme, reverse transcriptase, we can create a complimentary DNA strand which then can be multiplied using the above Polymerase Chain Reaction method.

When a piece of DNA, in a fluid that includes the polymerase enzyme, is heated to 94-96°C the DNA helix separates into two individual strings where the enzyme acts as a catalyst. When cooled the primers and probes combine with the matching segments of the individual strings and the DNA nucleotides fill in the blanks thus resulting in a doubled number of DNA double helix strands. The heat/cool cycle is repeated until there are enough pieces of DNA that the light emitted when fluorescence is activated can be measured by the photo detector.

So, how many cycles are needed before there is enough material to generate a readable signal? Obviously the more of the target DNA the sooner a measurable fluorescence can be detected. Just to set the table so to speak, keeping in mind the process is exponential, a single strand in our input sample will result in 1.07 billion strands after 30 cycles, 34.36 billion after 35 cycles and 1.1 trillion after 40 cycles. All from a starting single strand!

According to a report by the UK Public Health England titled “Understanding cycle threshold (Ct) in SARS-CoV-2 RT-PCR” the threshold line (Ct – cycle threshold) should be 26 cycles (67 million strands using our base case of 1 starter strand). Unfortunately, while it could be used as a means of measuring how much viral material was in the sample it would need to be calibrated using standards which is typically not done. So, an arbitrary cycle threshold is used and that varies from lab to lab and country to country.

From that same report a direct quote: “Ct values cannot be directly compared between assays of different types due to variation in the sensitivity (limit of detection), chemistry of reagents, gene targets, cycle parameters, analytical interpretive methods, sample preparation and extraction techniques. Additionally, Ct values are not provided for all SARS-CoV-2 molecular detection methods. Some commercial RT-PCR techniques are closed 'black box' systems whereby the operator cannot observe the reaction in real-time and the result is interpreted by software into a qualitative non-interrogatable positive or negative result.“

Still, while the more material being amplified in the original sample the sooner (a lower Ct) that a fluorescence signal can be detected. That still does not answer the question of how much material is required to get a “positive” result. Obviously the higher the number of cycles done the more likely a signal will be received even though the initial quantity was quite small. This is important as one should realise with our base case example, what if that single strand was from contamination? As indicated from the quote above there are a lot of other factors that come into play.

Personally, I have learned a lot about the RT-PCR process and much better understand what is being done, it still leaves a lot of unanswered questions. While proponents assert it is a very species specific “test” they do not provide, or at least I have not found, any definitive proof that the template used is only present in the SARS-CoV-2 virus. Thus, are the “primers” being used appropriate? What is the maximum threshold being used? Never reported. What care is taken to ensure no contamination? For other true testing methods, it is expected that duplicates, standards, and blanks be run at the same time as the regular samples to catch various sources of error. I have yet to see any documentation that these kinds of controls are included. Similarly, it is common with other testing methods to do a repeat analysis of an anomalous result, in this case a positive, to show reproducibility and therefore establish confidence in the result. I have yet to see any evidence that this is done on any basis.

Note in the previous paragraph in referring to RT-PCR I used “test”. This is because it really is not a true test, a “measure” of the viral load. It is, as previously used here merely a method of amplification. If anything can be called a “test” it is the evaluation of fluorescence which is merely used to determine if an artificial threshold is reached; if at any time, there is measurable fluorescence up to and including the maximum number of cycles used then it is classed as “positive” with no other apparent validation and checking done.

Things may be a lot better now but back in the spring there were two case histories involving large ships quarantined and virtually all tested using RT-PCR: the Diamond Princess cruise ship, and the US aircraft carrier Theodore Roosevelt. In both cases approximately 20% of all personnel tested “positive” yet only half of those positives ended up with symptoms. The other half were written off as being “asymptomatic” with not a thought given that some or even all were false positives as a result to the issues identified above. As a result, this is one “test” I personally do not put much faith in. Too bad it is being used indiscriminately for purposes it should not be.