This piece was originally published in the Worcester Telegram & Gazette on Monday, May 4, 2020.
By Sean Rollins
Associate Professor of Biology
Department of Biology/Chemistry
With regards to COVID-19, much of the information we receive is an over simplification of a very complex issue. Diagnostic testing is a critical component of any effective COVID-19 response. Before we can get back to any sense of normalcy and open society for business, effective testing is needed.
The bad news is that no common medical diagnostic test is perfect. Test results are not simply positive or negative. People who are truly positive for the infection could be incorrectly identified as negative. Good diagnostic tests might be 95% accurate. That means, if we let 20 people identified as negative into our place of business, one of those people could be infected.
There are two general types of tests being used. Direct testing, looking for the virus “directly” within a nasal swab; and antibody or “indirect” testing, looking for an immune response in the form of antibodies. The assumption is that if you have antibodies against the virus, you have the infection. There are advantages and disadvantages for each type of test.
Coronaviruses use RNA as their genetic material. HIV, influenza and Ebola are examples of other RNA viruses. RNA is an infamously unstable molecule; it breaks down easily. Direct tests are looking for this unstable RNA that breaks down quickly, making it harder to find. To complicate matters, human skin contains enzymes that breakdown RNA as part of your innate immune system, protecting you from viruses. How you collect your swab and how long it sits before processing become factors in your testing accuracy.
Dr. Birx, the coronavirus response coordinator for the Trump Administration, recently commented that antigen testing would be a “breakthrough.” This type of testing is similar to a flu swab or a rapid strep test; instead of detecting the RNA, look for a protein that is unique to the virus. The caveat is that we all have been previously infected with other coronaviruses; it is a frequent cause of the common cold.
In immunology, this is called cross-reactivity. Antibodies can respond to protein that looks similar to other proteins that we have seen before. You have to find the protein (or at least a small region of a protein) that is unique to COVID-19 but looks significantly different from other coronaviruses.
Another critical factor is how much RNA is present at the time of testing; referred to as viral load. Right after infection, the amount of virus in a nasal swab is low, unlikely to be detected. It takes several days for the number of virus particles to accumulate to detectable levels. People are likely to be walking around in the early stages of infection, potentially shedding virus, with a negative diagnostic test. A mouse model indicates that the virus replicates within 1-2 days of infection suggesting the virus could be shed early during human infection.
Antibody production also takes time and there are different types of antibodies. The earliest antibody response is typically 3-5 days after infection. IgM antibody is the body’s equivalent of a first responder but it isn’t the most prevalent blood antibody. IgG is the most prevalent antibody but takes longer to develop. IgG antibodies can start showing up five days after infection and peak between 7-21 days post-infection. Again, there is a significant lag between initial infection and diagnostic detection. A recent study suggests COVID-19 patients produce significant quantities of antibodies 11-12 days after infection. This same study also indicated that only 80% of patients produced IgG antibodies, although 100% of patients produced some type of antibody.
One advantage of an IgG antibody test is that it could serve as an indicator for future immunological protection. The immune system produces cells that retain memory from past infections and respond quickly to subsequent exposures. IgG is a critical neutralizing antibody; it coats the virus and blocks it from infection. It would be terrific to know that if you produce a high titer of IgG, you are protected from future infection but the science is not there.
Injecting COVID-19 patients with antibodies from a patient with a previous infection (plasma therapy), has been shown to be an effective experimental treatment. For this reason, protective immunity looks promising, but the length of time for protection is unclear and based on studies of other coronaviruses, immunological memory is not as long as seen with most infections. Scientists are also attempting to produce synthetic antibodies, which is a promising therapy option.
There are a number of additional complications associated with antibody testing. There are reports of patients experiencing reinfection. If you already have antibodies from a previous infection, how do you know if a patient is re-infected? Another significant population of patients are immunocompromised. We are all familiar with AIDS but there are a number of conditions that reduce one’s ability to produce antibodies. Cancer treatments, being a transplant recipient or even pregnancy can suppress the immune system.
Finally, and this applies to antibody and direct testing, diagnostic testing is a one-size-fits-all approach; which is not reality. We all have different diets, sleep, exercise, age, weight, other illnesses such as diabetes, asthma, heart disease, etc. We are also infected with a different number of virus particles. Every infection is different; a patient’s immune response and the virus’s ability to reproduce varies. Immunity passports have been proposed as a mechanism to allow individuals to go back to work. These passports will only be good as the diagnosis accuracy and frequency of testing.
Testing right now is kind of the wild-west; states and regions are doing their own thing. Standardization across state lines will help in the collection and analysis of data to make the most informed decisions moving forward. The ideal scenario would be to use both nasal swabs and antibody testing, but testing capabilities are limited. It is critical to understand that good testing is not the golden ticket to end this pandemic. Testing, vaccination, herd immunity and therapies are needed to fully deal with the pandemic. The best option to resolve this crisis as quickly and painlessly as possible, will require partnerships between the federal government, international organizations, states, commercial entities and end users.
Sean Rollins, Ph.D, is an associate professor of microbiology at Fitchburg State University and a adjunct assistant professor of biology at MCPHS University. He taught microbiology labs at Harvard Medical School for 10 years and earned a Ph.D. in microbiology from The Ohio State University. He did his post-doctoral fellowship in infectious diseases at Massachusetts General Hospital and a second post-doctoral fellowship in biological chemistry and molecular pharmacology at Harvard Medical School.