Knowing as we all now do, COVID-19 is a highly contagious viral disease caused by the novel coronavirus designated SARS-CoV-2. Thus far, we find the median incubation period for COVID-19 is 5.1 days and 97.5% of individuals develop symptoms within 11.5 days after infection.1 The first case of COVID-19 in the United States was confirmed on January 20, 2020. On February 23, there were 14 confirmed COVID-19 cases in the US, plus an additional 39 repatriated US residents, most of whom were returning from a cruise.2 By mid-March, many states had canceled school and closed businesses to limit the spread of the virus. In April and May, the number of new COVID-19 cases began to decline, yet a surge in cases occurred in June and early July after many states reopened businesses in late May or early June. As of late-August, the US leads the world with over 5.4 million infected with SARS-CoV-2.
Many are aware of how these circumstances have developed in recent months and determining the disease prevalence is essential for local authorities to make informed decisions on opening businesses, returning students to schools, and resuming other economic activities. Much of the burden for enabling these activities has fallen to clinical laboratories.
Detect the Proper Protein
A nucleic acid amplification test (NAAT), such as RT-PCR, performed on a nasopharyngeal swab specimen is the preferred method for detecting SARS-CoV-2 infection at this point in our growing understanding of the virus and resulting disease state. Furthermore, serologic assays may detect active and past infection, and can be performed on serum, plasma, whole blood specimens, or other body fluids. Thus, rapid serologic assays are a convenient way to screen asymptomatic individuals who may have been infected recently or recovered without seeking medical attention.
Most SARS-CoV-2 serologic tests are designed to detect antibodies to one or more of several viral proteins, including a spike protein projecting from the surface of the nucleocapsid, the receptor binding domain (RBD) at the N-terminus of the spike protein, and a protein associated with the nucleocapsid. Notably, the nucleocapsid protein of SARS-CoV-2 has 90% sequence homology to SARS-CoV-1, the virus that caused the SARS outbreak in 2003, and therefore some cross-reactivity with anti-SARS-CoV-1 antibodies is possible. However, the spike protein and the RBD region of the SARS-CoV-2 virus display the lowest sequence homology to SARS-CoV-1 and are more commonly targeted in SARS-CoV-2 serologic assays.
Several studies have indicated that seroconversion occurs within about two weeks after the onset of symptoms and antibody concentrations reach a peak within six days of seroconversion.3 The titer of IgG antibodies generally correlates with viral neutralization activity.4 Although the antibody titer may decrease slightly in the three months following infection, a sensitive serologic assay should be able to detect individuals with history of SARS-CoV-2 infection.5
Evaluation of Serology Tests
The first serology tests that became available for anti-SARS-CoV-2 antibodies were rapid tests based on lateral flow immunochromatography (LFI), a technique used in many point-of-care (POC) immunoassay devices. These methods typically detect IgG and/or IgM antibodies to the virus. The SARS-CoV-2 nucleocapsid and spike proteins have been used as the antigen to capture anti-SARS-CoV-2 antibodies and are coated on a solid support. Anti-human IgG or IgM antibodies labeled with colloidal gold form complexes with captured antibodies and produce a blue line in the test region of the cartridge when antibodies are present. Typically, the result is available within 10 to 15 minutes. Most kits can test serum, plasma, or whole blood, and can be used in health care facilities or remote locations. Keep in mind, some of the available test kits do not specify which capture antigen is used or which antibodies react with the assay. That said, as an example, the product insert for the SD Biosensor IgM/IgG kit claims a sensitivity of 92.6% and a specificity 96.7% for specimens collected 8 days after the onset of COVID-19 symptoms. The following systems have since come into the market as well:
Both tests are qualitative and are run on the Vitros automated chemistry platform. There is a 48-minute cycle time for the first result. According to the product insert, the clinical agreement of positive specimens with confirmed viral infection is 83.3% for the total antibody assay and 87.5% for the IgG assay. The clinical agreement of negative specimens collected pre-December 2019 was 100% (95% CI, 99.1-100%) for both tests.
The Siemens anti-SARS-CoV-2 IgG antibody assay uses biotinylated SARS-CoV-2 RBD and streptavidin-coated microparticles to capture IgG antibodies. An acridinium ester-labeled, mouse monoclonal anti-human IgG antibody is used as the detection antibody and the reaction time is approximately 20 minutes. The Atellica instrument does not perform onboard dilution, but specimens may be manually diluted up to 8-fold. The result is reported as an index value, with reactive defined as an index ≥ 1.0. The measuring interval is in the range of 0.5 to 20 (index). This semi-quantitative assay is not currently recommended for screening convalescent plasma donors.
There have been several reports describing in-house developed, semi-quantitative ELISA assays for anti-SARS-CoV-2 antibodies.6 In a typical ELISA, wells are coated with SARS-CoV-2 spike or nucleocapsid protein as the capturing antigen. For an IgG assay, HRP conjugated anti-human IgG antibodies may be used to detect IgG antibodies against the virus. After a wash step to remove unbound material, the HRP substrate is added and optical density is measured by a UV-Vis spectrophotometer. The cut-off value is usually established using negative specimens collected prior to December 2019. The antibody titer can be reported up to 1:10,000. These in-house developed and validated, semi-quantitative ELISA assays may be established to support surveillance studies and to screen convalescent plasma donors.7 A limitation of the ELISA method is a relatively long incubation time, but the methods can be adapted to automated liquid-handling systems to achieve high-throughput analysis.
EUA and Umbrella EUA
Traditionally, it has taken anywhere from several months to years to develop a clinical assay and collect sufficient data to submit an application for 510k approval by the FDA as an in vitro diagnostic test that may be used in clinical laboratories. However, given the urgent need to drastically scale up COVID-19 testing capacity earlier this year, the FDA began issuing emergency use authorizations (EUAs) for tests with minimal validation in order to allow for rapid access to viral and antibody testing resources. An EUA requires significantly less validation data than a full 510k application and the FDA usually evaluates EUA applications within several days of submission. Many new diagnostic tests for COVID-19 received EUAs in the US. One such example was the Abbott Diagnostics ID Now product, which uses isothermal NAAT technology. Although the assay sensitivity is not equivalent to the sensitivity of an RT-PCR test, the test is attractive because it can produce results in 15 minutes.
On April 28, 2020, the FDA issued an “umbrella” EUA for SARS-CoV-2 antibody tests (including those using LFI or lab-developed ELISAs) that had been validated in studies performed at the NIH and NCI. This authorization gave hospitals and clinical laboratories access to many serologic tests during the first wave of the pandemic. However, on July 21, 2020, the FDA revoked the umbrella EUA based on additional data and concerns about the reliability of the tests. Manufacturers are now required to submit an individual EUA application for each test. In addition, the FDA revoked the EUA for a few of the LFI assays due to the lack of scientific evidence of their reliability and the potential risks involved in their use. Ultimately, EUAs for these tests are only in effect for the duration of the COVID-19 pandemic and may be terminated or revoked when the pandemic is over or at the discretion of the FDA.
Potential Use of COVID-19 Serological Assays
RT-PCR can detect patients with active infection within 2 to 3 days following the onset of symptoms. However, the viral load usually, but not always, falls to undetectable levels within 10 to 14 days following symptoms. Serologic tests may be able to identify individuals who had asymptomatic infections or have recovered from the disease. Both serologic testing and NAAT are needed to accurately assess the disease prevalence in a community, as their results are complementary. Moving forward, serologic assays will be needed to help characterize the immune response following SARS-CoV-2 infection, and perhaps to verify immunity when a vaccine becomes available. In addition, serologic assays may be used to screen potential plasma donors for convalescent plasma therapy to determine whether sufficient antibodies are present for the treatment to be effective.
Reports of the prevalence of antibodies to SARS-CoV-2 vary from a few percent to 24.7% in New York City.8 As with any screening test, the disease prevalence influences the clinical performance of serologic tests. A screening test must have high sensitivity to minimize false negative results, and a confirmatory test should have high specificity to limit false positive results. In a population with low disease prevalence, a screening test would require high specificity to minimize the number of false positive results. This is a limitation of all screening tests and most do not have sufficient specificity to justify their use in screening asymptomatic individuals. According to its website, the CDC does not generally recommend testing for SARS-CoV-2 infection in asymptomatic individuals unless they have been exposed to someone with confirmed COVID-19 or are advised by their health care provider to get tested.
Clinical Validation of Serologic Assays
Validation of serologic assays in a clinical setting involves consideration of several factors. Some assays measure the total amount of antibody, regardless of the isotype. IgG, IgM, and IgA antibodies against SARS-CoV-2 have all been identified, but these three emerge at different stages of the disease. Typically, the first to appear are IgM antibodies, followed by IgG and IgA. Therefore, total antibody tests would seem to have an advantage of greater sensitivity, but this has not been demonstrated. In fact, for screening potential convalescent plasma donors, the FDA has only approved the Ortho IgG assay, because the results correlate most closely with the results of viral neutralization studies, and the test is designed to be semi-quantitative.
Another consideration when comparing the results of serologic assays involves the antigen used to capture the antibodies. As mentioned, various serologic assays use nucleocapsid or spike proteins, and detect antibodies to specific domains within those proteins. This can lead to disagreement between two serologic assays that detect different antigens on the virus.
Finally, the differences in analytical sensitivity between various serologic tests—most notably between ELISA-based methods and LFI methods—can produce discrepant results. This has been our observation in validating serologic methods. There were instances of qualitative disagreement between the results of the ELISA and the results of LFI devices. Better correlation was observed between the ELISA and the Ortho total and IgG antibody methods. Studies to correlate the ELISA with the Roche total antibody assay, and the Siemens total and IgG methods, are ongoing.
In response to the COVID-19 pandemic, the FDA greatly simplified the application and approval procedure for SARS-CoV-2 diagnostic tests via the EUA process. While many serologic assays have become available to hospitals and clinical laboratories in the US, some of these assays do not provide reliable, actionable information and suffer from poor analytical performance. Thus, the selection of a serologic assay to be used as a surveillance tool needs to balance the hospital needs, test performance, available resources, and the regional disease prevalence.
Zhicheng Jin, PhD, received his doctorate in analytical chemistry with a focus on mass spectrometry. He is currently a Clinical Chemistry Fellow in the Department of Pathology and Genomic Medicine at Houston Methodist Hospital.
Xin Yi, PhD, DABCC, FAACC, is a board-certified clinical chemist, currently serving as the Co-director of Clinical Chemistry at Houston Methodist Hospital in Houston, Texas, and an Assistant Professor of Clinical Pathology and Laboratory Medicine at Weill Cornell Medical College.
David W. Bernard, MD, PhD, is the Medical Director of Clinical Pathology for Houston Methodist Hospital System. He is also Associate Professor of Clinical Pathology and Laboratory Medicine at Weill Medical College of Cornell University and is a member of the Houston Methodist Research Institute. As Medical Director of Clinical Pathology, David directs the strategic and day-to-day laboratory operations for all of Houston Methodist’s hospitals and clinical facilities, as well as provides onsite direction at the flagship Houston Methodist Hospital in the Texas Medical Center. He is also Medical Director for the Houston Methodist Research Institute Biorepository.
Roger L. Bertholf, PhD, is Medical Director of Clinical Chemistry at Houston Methodist Hospital, Professor of Pathology and Genomic Medicine at Houston Methodist Research Institute, and professor of clinical pathology and laboratory medicine at Weill Cornell Medical College. He is board-certified in clinical chemistry, toxicological chemistry, and point of care testing. Roger is also editor of Laboratory Medicine, published by the American Society for Clinical Pathology and Oxford University Press.
Enter our Sweepstakes now for your chance to win the following prizes:
Just answer the following quick question for your chance to win:
Entries are limited to one entry per person in any active sweepstakes.
Thank you for your entry.