Concern about the virus SARS-COV-2 that causes the disease called COVID-19 has been largely due to reported mortality rates. However, errors in reporting these rates, as well as unresolved differences between rates in different countries has led to confusion regarding the true health impacts. Because all rates currently being reported are dependent on the test used to identify infected patients, understanding that test and its history could lead to much needed clarity.
Errors in reported mortality rates have come from simple mistakes in calculation. An example has been equating the measured case fatality rate (deaths divided by patients testing positive for virus) with the number of people who died who had ever been infected. The latter number is unknown and will not be known until antibody titers can be produced for a much larger segment of the population. But that actual mortality rate is expected to be much lower, perhaps around 0.3% as estimated by an epidemiologist from Stanford University.
Another common error has been attributing the deaths of all infected people to COVID-19, regardless of other pre-existing illnesses. This error has become much worse recently due to governments mandating that all deaths of presumptive patients be listed as death from COVID-19, even if the patient was never tested for SARS-COV-2 at all.
These mortality rate errors would be worsened if there were errors in testing for presence of the virus. What is becoming increasingly clear is that there have been, and in some cases still are, serious questions regarding the reliability of the testing.
The test that is the focus of this discussion is the reverse transcriptase quantitative polymerase chain reaction (RT-qPCR) test to identify presence of nucleotide sequences attributed to the RNA of SARS-COV-2. For this test, which follows different protocols in different countries, serious problems have been seen. The first of these problems relates to the testing performed in China, the reported origin of the virus.
The Chinese Mystery
A scientific study was performed in China that targeted the testing of subjects who had been in close contact with SARS-COV-2 infected patients. The results were peer-reviewed and published in the Chinese Journal of Epidemiology on March 5th, 2020. The data-driven conclusion of the study was that “nearly half or even more” of patients testing positive for SARS-COV-2 did not actually have the virus. In other words, half of the results were false positives.
For perspective, this study was peer-reviewed and published in a Chinese state journal a month after COVID-19 was said to have surpassed the 2003 SARS epidemic and just as the World Health Organization (WHO) declared the outbreak to be a pandemic. This was a full month after China had ordered a lockdown affecting over 36 million people.
Mysteriously, this peer-reviewed study was withdrawn a few days after publication and is no longer available for review. In response, one investigative team asked a Chinese graduate student to contact the lead author of the study, Dr. GH Zhuang, for explanation. Dr. Zhuang responded by email but did not cite a reason for withdrawal of the paper, only saying that it was “a sensitive matter.” Others then made the false assumption that the author had identified some kind of mistake in the science despite the fact that no such mistake was ever identified.
As reported by the investigative team that contacted Dr, Zhuang, “Without access to the paper, nobody can assess the value of the work or determine whether it suffers from a scientific flaw. It’s also unknown if the paper was retracted for political reasons.”
At about the same time the paper was withdrawn, it was reported that the White House had ordered all meetings regarding COVID-19 to be classified. This meant that the U.S. Centers for Disease Control (CDC) would filter all information and decision making through the National Security Council (NSC). This unprecedented decision led to meetings being held in secret facilities that were “usually reserved for intelligence and military operations.”
Today, all decisions about COVID-19 in the U.S. are coordinated by secretive agencies like the CIA and DIA and the information behind them is classified. This fact is concerning given that U.S. senators who were members of intelligence committees engaged in insider trading related to the COVID-19 weeks before stock markets crashed due to the virus.
Regardless of politics, to understand the concept of a false positive one should realize that analytical test methods need to be balanced with respect to quality considerations like sensitivity and specificity. If a test is not sensitive enough, the analyte of interest will not be found when it is there, giving a false negative. If a test is not specific enough, something else in the test sample will be identified as being the target analyte when it is not, giving a false positive.
In this case, false positives could mean that the test is reacting to another virus or genetic source. Alternatively, the test could be detecting the presence of SARS-COV-2 residues after a previously infected individual is no longer sick. That would mean results are falsely attributing infection to a patient who has recovered. In other words, the test can detect the presence of viral material in a sample that does not represent a replicating (i.e. active) virus. Lastly, even very small amounts of contamination in the laboratory can cause a false positive. No matter the cause, false positives mean higher reported mortality rates, more confusion, more fear, and more bad decisions.
The RT-qPCR test for SARS-COV-2 is being used a qualitative test, despite the technique name including the word quantitative. This means that the actual amount of virus in a sample is not considered important, only the presence of even a small amount of virus. This approach is odd for a pandemic screening test because a replicating virus should be present in large quantities in an infected patient. That is, if you’re trying to identify who is infected, you don’t need more sensitivity, you need more specificity because a replicating virus should not be hard to detect. This concern would be lessened if the actual test results showing levels of virus detected were available. Unfortunately all the public sees is numbers of positive or negative determinations.
WHO Guidelines and the Virus
The World Health Organization (WHO) originally based testing on a relatively simple kit developed in Germany, not on the Chinese protocol. WHO has since developed general guidance for testing SARS-COV-2. This guidance requires some understanding of terminology so it’s helpful to understand the virus and the principle of RT-qCPR.
The SARS-COV-2 virus genome was first sequenced in early January 2020.. Like other coronaviruses, it is a single-stranded, positive sense RNA, as opposed to double-stranded RNA or DNA viruses. The viral genome is approximately 29.9 kilobases long, making it, like other coronaviruses, among the largest single-stranded RNA viruses. And like other such viruses, it evolves quickly, meaning the genetic sequence changes over time.
RT-qPCR involves multiple steps. The sample is first lysed (i.e. the cells are cut) to release any viral material. Then the RNA present is converted into complementary DNA (cDNA) using an enzyme called reverse transcriptase. This is sometimes called the “extraction” step. After this, the cDNA is used as a template for amplification using qPCR, allowing the original quantity of target RNA to be determined.
The amount of virus present is determined by the number of PCR cycles the sample required above the threshold, or background noise. The number of cycles required is called the Cq and the threshold is called the Ct although many people refer to both as Ct. The Cq is inversely related to the amount of virus present so large Cqs mean less virus and typically levels above a Cq of 38 are considered negative.
The amplification is not done on the entire cDNA sequence but on segments that are expected to be representative of the specific genome of interest and, correspondingly, not representative of other genetic materials that could be present. Segments of the SARS-COV-2 genetic code that are usually targeted by primers correspond to sections of the original RNA named ORF1a, ORF1ab, S, M, E, and N. The first two are Open Reading Frames (ORFs), the next three are genes coding for surface proteins, and the last (N) is a gene that codes for what is called the nucleocapsid, a protein shell immediately surrounding the genetic material but beneath the surface proteins.
Synthetic primers and fluorescent probes are identified to match up with the target genetic segments to facilitate amplification and detection. The primers are small nucleotide sequences that bind to the targeted segments of the cDNA genetic sequence. The number of primers used is recommended to be at least two but can be three. Sometimes there is only one if there are other confirmatory tests performed on presumptive positives. Protocols from different countries target different genes in the SARS-COV-2 sequence. China used Orf1ab and then confirmed with a test for N. In any case, the target sequence within the viral genome, and therefore the primers used, are critical decisions for success because issues with primer design and efficiency can lead to variation in results.
As described in an article in The Scientist, the WHO-recommended primers first target the E gene of SARS-COV-2. The E gene is considered highly divergent and therefore more specific to the different coronaviruses. If a lab following WHO guidance obtains a positive screening test, it will do confirmatory testing targeting other areas of the virus genome. To avoid false positives, “every positive test has been confirmed with whole genome sequencing, viral culture, or electron microscopy.”
The U.S. Test
Unfortunately, the U.S. did not follow the WHO recommended test guidance and decided to follow its own rules for testing of SARS-COV-2. In fact, WHO and CDC never discussed the U.S. using the same test as being done internationally. Those who questioned the decision learned that “It’s not clear why the CDC chose to develop a different assay to that selected by the WHO and taken up by other countries. The CDC declined to respond to questions from The Scientist.”
Due to problems later found with the CDC tests, the U.S. was later severely criticized for its decision. Although CDC has been secretive about the details, the problems with its test appear to have included both test design issues and contamination.
CDC began manufacturing its test kit in January and shipped it on February 5th to state labs and to 30 other countries including 191 international labs. A February 12th briefing at the CDC discussed problems with the CDC test. Although the statements made were unclear, it appears that states were complaining that the test was “inconclusive” and therefore CDC was going to focus on “redoing the manufacturing.”
It was reported that, “in addition to originally sending out flawed tests, the CDC added to the confusion by providing limited information to labs in the weeks that followed. There was a period of time after the tests were recalled where there was near silence. It was about two weeks.” It was only after an open letter to Congress on February 28th, from more than 100 virologists and other specialists, that the CDC responded by allowing independent labs that had validated their own tests to begin testing.
The CDC test originally included three primers, all targeting one gene—the N gene of SARS-COV-2 that encodes for the nucleocapsid. The primers were denoted N1, N2, and N3. Nucleocapsids of RNA viruses “are fairly simple structures that contain only one major structural protein, often referred to as the nucleocapsid protein or core protein. This protein is usually basic or has a basic domain.”
Although the CDC test might have provided good sensitivity, it appears that it did not provide high specificity as it targeted only one basic gene of the coronavirus. CDC admitted this in a disclaimer noted in the method, saying, positive results “do not rule out bacterial infection or co-infection with other viruses. The agent detected may not be the definite cause of disease.”
At first, due to CDC secrecy, problems with the test kit were difficult to understand. As the Washington Post reported, “The trouble with the CDC test arose because the third attempt at a match, known as the N3 component, produced an inconclusive result even on known samples of the coronavirus.” But that was not the whole story.
On February 28th, as the open letter to Congress was being recognized, it was reported that the N3 primer of the CDC kit was contaminated. This caused the negative control within the kit, containing DNA that was unrelated to SARS-COV-2, to react as if it was a positive hit for SARS-COV-2. In other words, the kits were generating false positives for negative controls.
How much contamination was present was not clear because, again, the actual test results giving amounts of virus found are not available to the public. And CDC has not been open with communications about the problems found. Oddly enough, in April, test kits in the UK were also found to be contaminated with COVID-19.
What did CDC do to correct the problems with the kit? Instead of re-manufacturing the N3 primer as originally planned, on March 15th the CDC simply told everyone who had the kit to remove the N3 primer and use the kits without it. Additionally, CDC changed its method requirements to eliminate the need to confirm positive results. This made the test kit that was based on detection of only one basic gene in SARS-COV-2 even less specific and told users that results didn’t need to be confirmed. These changes made the test less reliable in terms of identifying SARS-COV-2 and therefore made any subsequent estimates of mortality rates less reliable as well.
The history of testing for SARS-COV-2 infection has involved mysteries and problems that have led to delays in testing and reporting of rates of infection than are falsely higher than actual. Exacerbating these issues are government mandates for medical professionals to list COVID-19 as cause of death for patients that have inconclusive causes of death and, in some cases, were never tested for SARS-COV-2 at all.
These facts add to the growing concerns among citizens around the world about government response to the COVID-19 pandemic. These concerns include that the response is estimated to generate at least as many deaths as the virus will cause and that mortality rates have been reported in error. Moreover, the introduction of alarming government actions that are reminiscent of a totalitarian police state must be considered as well as the unprecedented decision in the U.S. that all public health information and decisions be controlled by the NSC in secret meetings.
Only complete transparency of the details behind the decisions made and issues seen can help maintain the public trust and lead to mutual agreement about the path forward. There are many questions that will need answers. For example: Are reported rates for other diseases like influenza dropping in proportion to the rise in reported infection by SARS-COV-2? What did the Chinese study that was mysteriously retracted actually say? What has investigation into the CDC kit contamination revealed? What other countries have based their mortality figures on test kits that could have been unreliable?
Citizens can help by calling on authorities and test facilities to publicly share the details of testing including the actual results of the RT-qPCR tests, including Cq (or Ct) values. In addition to information sharing, an international investigation into the problems seen with testing, starting with Chinese results and U.S. test kits, should be conducted. Such an investigation can lead to preventing the reporting of false positives and the ensuing panic that comes from artificially high estimated mortality rates.