​

On May 5, 2020, ASM hosted a Serology Testing Webinar for COVID-19. Close to 800 experts from around the world tuned in, making it abundantly clear that the demand for factual information about this topic is high. Here we provide an in-depth look at the principles of serology testing – protocols, platforms, authorization guidelines, interpretation of results, accuracy concerns and overcoming roadblocks to success. 
 
When serology testing is used for diagnostic, surveillance or epidemiological purposes, antibodies and antigens are the 2 serum proteins of interest. Antibodies are specialized Y-shaped proteins, also known as immunoglobulins, that recognize foreign particles (antigens) located on microbial surfaces. Antibodies either mark microbes for destruction by immune cells or the complement system, or target and eliminate them directly. This process is specific. A given antibody recognizes and binds to its corresponding antigen in a manner analogous to a lock and key mechanism. Seroconversion is the development of detectable antibodies in the blood against a particular antigen. It takes 1-2 weeks post symptom onset for patients to seroconvert to SARS-CoV-2. 

Protocols and Platforms for COVID-19 Serology Testing 

COVID-19 serology testing relies on targeted antibodies binding to SARS-CoV-2-specific antigens. Blood serum is collected and applied to a testing platform that contains copies of viral antigen. Capillary action draws the blood through the device where it mixes with the antigens. If the patient has developed antibodies in their blood against SARS-CoV-2, the corresponding antibodies will recognize and bind to the antigens, indicating past exposure to SARS-CoV-2.
 
The platforms for COVID-19 serology tests on the market today include lateral flow assays, ELISA (enzyme-linked immunosorbent assays) and chemiluminescent immunoassays. These assay types differ in how they detect antibody-antigen binding.  
 
The U.S. Food and Drug Administration (FDA) has recently released updated performance characteristics for Emergency Use Authorized (EUA) serologic tests.

Antigens

Accurate interpretation of serology tests depends on antigen specificity. Antigens can be proteins, polysaccharides or lipids, but if they are not pathogen specific, the possibility of cross-reactivity increases and the reliability of test results decreases. The following viral antigens have been used to detect antibodies for SARS-CoV-2. 

• Spike Protein – Spike proteins (S proteins) are unique, mushroom-shaped surface proteins that bind host cells and mediate virus entry. Each monomer of S protein contains two subunits, S1 and S2, which facilitate attachment and membrane fusion, respectively. S1 and S2 subunits may be used individually or combined as antigens for serology testing. 

• Nucleocapsid – The nucleocapsid protein (N protein) is a basic RNA-binding protein that plays structural and nonstructural roles in infection. In complex with genomic RNA, N protein forms the viral capsid of SARS-CoV-2, and data suggest it plays a number of additional roles in pathogenesis. 

• Receptor Binding Domain (RBD) – RBD represents the portion of the S1 protein that binds angiotensin-converting enzyme 2 (ACE2), the human receptor for SARS-CoV-2. 3D animation depicting major elements of SARS-CoV-2, including the Spike S protein, N protein, viral envelope and helical RNA.

Source: https://www.scientificanimations.com/wiki-images/

Antibodies

Accurate interpretation of serology testing depend on antigen specificity, but also on the type of antibody being detected. Humans have 5 different classes of antibodies, and each plays a unique role in immunity. IgM, IgG, IgA and total antibody count are the primary targets of COVID-19 serology tests. The biological properties of these isotypes are distinguished below. Research to define the temporal kinetics of antibodies against SARS-CoV-2 is ongoing. 

• IgM is one of the first antibodies produced during infection. It can be expressed in monomeric form on the surface of B lymphocytes or found circulating in the blood and lymphatic fluid in pentameric form. The IgM pentamer consists of 5 antibodies joined together to form a ring-like-structure. This structure, coupled with the fact that the antigen-binding site of IgM is not highly specific, allows for simultaneous binding of multiple antigens and rapid clearing from the bloodstream during primary infection. Although IgM is the largest antibody by size, its relative abundance in the blood is only about 10% of total antibody count. 

• IgG is the smallest and most abundant circulating antibody. It exists in monomeric form, makes up approximately 80% of total antibody count and is primarily found in serum. IgG typically appears later in infection when mature B cells receive signals to switch from production of IgM to IgG. During the secondary immune response, IgG can have many potential roles, including direct neutralization of microbes and targeting of microbes for immune cell-mediated processes. As the most specific and long-lasting isotype, IgG is a key player in establishing post-infection immunity. 

• IgA is primarily responsible for protecting mucosal surfaces, exists in dimeric form and can be found in serum, mucosal secretions, saliva, tears, sweat and breast milk.  Illustration of IgM, IgG, and IgA antibody isotypes. Image adapted from OpenStax College.

Source: http://cnx.org/content/col11496/1.6/

Cross-Reactivity Concerns

COVID-19 serology assays are designed to be specific for SARS-CoV-2, but how do we know they will not cross react with other coronaviruses? Cross-reactivity with the common coronaviruses (cCoV's) would be especially detrimental since 60-75% of children have antibodies to one or more cCoV’s, and 90% of adults over 50 years of age have antibodies to all 4 cCoV's. Fortunately, there is not a lot of sequence identity shared between SARS-CoV-2 and cCoV’s (approximately 21-34% AA homology), but the FDA does require laboratories to include a note on all positive serology reports stating, “False positive results due to antibodies to cCoV’s may occur.” 
 
What about the closer relatives of SARS-CoV-2? SARS-CoV-2 and SARS-CoV share 90% amino acid identity for their respective N proteins and 77% for S proteins. SARS-CoV-2 and MERS-CoV share 49% amino acid identity for N proteins and 33% for S proteins. These data highlight the value of using S protein antigens to increase the specificity of serology tests.

Predictive Values and Disease Prevalence

Positive and negative predictive values are dependent on test performance characteristics (specificity and sensitivity), as well as disease prevalence. In regions with low COVID-19 disease prevalence, the risk of false positive results by serologic testing is higher, even with excellent specificity. Therefore, the rate of infection needs to be taken into account, and repeat testing may be useful to confirm results in populations with low disease prevalence. 

Protective Immunity Against SARS-CoV-2

A key question that continues to circulate is, if someone was exposed to SARS-CoV-2, develops antibodies and comes in contact with the virus again in the future, will they be protected? To answer questions about long-term immunity, it’s important to ensure serology tests target longer-lasting, highly specific IgG antibodies and not only the shorter acting, less specific IgM isotype. 
 
In addition, the mere presence of IgG antibodies does not guarantee protection from future infection. A recent study reported no reinfection of rhesus macaques who were re-exposed to SARS-CoV-2 nearly a month after primary infection. This research offers hope that the development of IgG may provide immunity against SARS-CoV-2; but the data is limited, and we do not yet have definitive proof that this is the case, nor do we know how long immunity will last if it exists.
 
The duration of antibody responses against other human coronaviruses may provide relevant background information about this topic. Previous studies have shown that IgG antibodies to common coronaviruses peak approximately 2 weeks post infection and return to baseline about a year post exposure. Reinfections have been observed for at least 3 of the 4 cCoV’s. The reason for reinfection is not fully understood, and could stem from declines in protective immunity or reexposure to genetically distinct forms of the virus. SARS-CoV antibodies peak about 3-4 months post infection and decline to undetectable levels about 6 years post exposure. And evidence indicates that MERS-CoV neutralizing antibodies may remain present as long as 3 years post exposure to the virus. 

SARS-CoV-2 Neutralizing Antibodies

Neutralizing antibodies (NAb) are a subset of antibodies produced against a virus that independently block viral entry into host cells and are primarily of the IgG isotype. How can we determine if the antibodies being detected by common serology assays are capable of neutralizing SARS-CoV-2? 
 
Neutralizing antibodies are classically detected using Plaque Reduction Neutralization Tests (PRNTs). However, PRNTs require the use of live virus and must be performed in biosafety level 3 (BSL3) facilities for SARS-CoV-2. Furthermore, this type of test is challenging and time consuming to perform. As an alternative, BSL2 neutralization tests have been developed using Pseudotyped Vesicular Stomatitis Virus (VSV) expressing SARS-CoV-2 spike (S) protein. At this time, it appears that these tests are only available for use in academic institutions and research laboratories. 
 
A recent study in China used Pseudotyped VSV expressing S1, S2 or RBD of SARS-CoV-2 to test for neutralizing antibodies in samples from 175 hospitalized patients who had mild disease (were not admitted to the intensive care unit). Results showed that neutralizing antibodies were detectable 10-15 days after symptom onset, but the titers were highly variable amongst samples. While neutralizing antibodies were produced in some cases, 6% of patients did not develop NAb titers at all, and 30% only developed low NAb titers (>1:500). Additionally, the study reported a correlation between older age and increased neutralizing antibodies. 
 
A few additional studies have indicated a correlation between the antibodies detected by SARS-CoV-2 serology tests and neutralizing action, but the data is preliminary, and very few patients have been evaluated in total. Taken together, it’s important to consider the possibility that not everyone will develop neutralizing antibodies to SARS-CoV-2. When neutralizing antibodies are developed we still don’t know whether they are fully protective, what titer is needed to have full protective immunity or how long the titers will persist. More information is needed to determine the answers to these pressing questions.

Roadblocks to Success 

Accuracy concerns and questions about authorization guidelines have generated hesitancy amongst clinical laboratories performing serology tests. Much of this uncertainty is based on limited data. Ironically, the direct applications of serology testing create inherent roadblocks to answering these questions. When laboratories are making real-time decisions about the allocation of resources, time and expertise, the fact that serology testing is not intended for diagnostic use is impactful. Hospital labs must prioritize the identification of acute cases of illness in order to contain the spread of disease and provide effective treatment for critically ill patients. In short, surveillance studies are useful for epidemiologic purposes, but not the first priority in triage situations. That’s why serology tests are being predominately conducted in large clinical labs and research facilities at this time.
 
It’s therefore valuable to conclude with a summary of the current recommended uses for serology testing.

When COVID-19 Serology Testing Should Be Used

• The primary application of serology testing is the identification of individuals who have previously been infected with SARS-CoV-2. This knowledge can be used to guide epidemiology and seroprevalence studies, as well as facilitate contact tracing. 

• Serology tests may also be used to identify potential convalescent plasma donors and to evaluate the immune response to candidate vaccines.

• Finally, there is potential for serology tests to aid in the diagnosis of COVID-19 in RT-PCR negative patients who present later during disease course.

When COVID-19 Serology Testing Should NOT Be Used

• Serology testing should not be used to diagnose acute or recent cases of COVID-19.

• At this time, serology tests cannot be used to definitively determine if a patient has developed protective immunity.

• Because of the above limitations, SARS-CoV-2 serology testing should not be used to guide personal protective equipment (PPE) use or adherence to social distancing practices.

 A few weeks ago, Research!America hosted a webinar entitled, Understanding the Landscape of COVID-19 Vaccine and Treatment R&D, during which ASM CEO Stefano Bertuzzi; former FDA Commissioner Dr. Mark McClellan; and former CDC Director, Dr. Julie Gerberding, discussed the value and limitations of serology testing. 
 
“In absence of approved and proven therapies, diagnostics become particularly important,” said Dr. Bertuzzi. “70% of doctors’ decisions are based on tests performed in the lab.” But we need to continue to be cautious.
 
“We need to be clear about what testing is for, and we need to do the careful science to not unnecessarily waste the resources we have developed,” added Dr. Gerberding. “We’re nowhere near herd immunity, so we need to find other solutions to restart our economy.”  
 

​