key: cord-0913610-jr2j4qmd
authors: Parikh, Bijal A.; Farnsworth, Christopher W.
title: Laboratory evaluation of SARS-CoV-2 in the COVID-19 pandemic
date: 2021-01-12
journal: Best Pract Res Clin Rheumatol
DOI: 10.1016/j.berh.2021.101660
sha: eda8128e95f27fa1bf11018f348a9b6bca833df4
doc_id: 913610
cord_uid: jr2j4qmd

Laboratory evaluation of SARS-CoV-2 involves the detection of viral nucleic acid, viral protein antigens, and the antibody response. Molecular detection of SARS-CoV-2 is the only diagnostic test available currently in acutely or recently infected individuals. In contrast, serological testing is typically performed once viral RNA has been cleared and symptoms have resolved. This leads to some confusion amongst clinicians of which test to order and when each is appropriate. While SARS-CoV-2 assays can suffer from poor sensitivity, all FDA authorized assays to date are intended to be qualitative. Serological tests have multiple assay formats, detect various classes of immunoglobulins, and have a distinct role in seroprevalence studies; however, the association with long-term protection remains unclear. Both molecular and serological testing for SARS-CoV-2 have complementary roles in patient management, and we highlight the challenges faced by clinicians and laboratorians alike in evaluation and interpretation of the currently available laboratory assays.

Once the EUA process allowed more commercial labs entry to clinical testing, a number of assays soon became widely available. The first commercial lab with an FDA approved EUA Kit was ThermoFisher's TaqPath COVID-19 Combo Kit on March 13, 2020 . While like the CDC method, this kit required specific instrumentation for extraction and analysis, it broadened the number of labs able to participate in testing. This was followed quickly by another manual method, the Quidel Lyra SARS CoV-2 assay on March 17, 2020 . Meanwhile labs around the country were developing their own workflows that were not intended to be marketed as test kits for distribution. In March 2020, Labcorp, Quest, the Wadsworth Center, Avellino Labs, and the Yale Clinical Virology Lab were among the first to successfully earn EUA status for their in-house testing. The first two sample-to-answer (no independent extractions required) were the Panther Fusion SARS-CoV-2 assay by Hologic and the Abbott RealTime SARS-CoV-2 EUA test, by Abbott Molecular. Finally, testing that could be performed outside CLIA accredited labs in the point-of care (POC) setting began to appear, further increasing access to diagnostic testing. The first notable entry into the field was Abbott's ID Now. As all of these testing options began to exponentially increase, clinicians were beginning to feel pressure to provide testing to more and more patients who met epidemiological criteria. However, understanding the challenges and limitations of molecular testing was underestimated by ordering physicians and the clinical laboratory alike.

The molecular detection of SARS-CoV-2 requires an initial conversion of RNA to DNA as the substrate for the polymerase chain reaction (PCR) dependent amplification. A reverse transcriptase or RT enzyme catalyzes this step prior to PCR. Following the RT step, a fluorophore-dependent system (i.e. Taqman, Molecular Beacons, Scorpion probes, etc) is used to determine whether viral nucleic acid is present. The strategy initially employed by the CDC assay was to target three different regions of the N gene along with a region of the human transcriptome, specifically the RNase P gene. This latter gene served as a specimen adequacy control (SAC), meaning that if RNase P was not detected, either the PCR was inhibited by some unknown substance within the patient's sample or that the patient was inadequately sampled. As a result of the limited fluorophores that can be multiplexed, many tests have been developed that replace the SAC in favor of an extraction or processing control (PRC). Specifically, a known (non-human) RNA template is spiked into the patient sample prior to extraction and then detected with a unique molecular probe. In all molecular platforms, SARS-CoV-2 (or a control target) is detected if the cycle threshold or CT is crossed prior to the completion of 40 PCR cycles. However, when two targets are used to diagnose SARS-CoV-2, specific instructions must indicate how the discrepancy should be interpreted, as described below. Importantly, all EUA assays for the diagnosis of SARS-CoV-2 are independent of the actual CT value and more simply qualitative in nature. While a great deal of attention has been placed upon understanding the details of molecular detection methods, diagnosis of SARS-CoV-2 RNA is highly comparable to many of the available assays for other clinically important respiratory pathogens.

The first assays with EUA targeted a limited subset of genes on SARS-CoV-2, as specificity from other coronaviruses was important to rule in or out infection with the pandemic strain. The World Health Organization (WHO) and countries outside the US also maintained a list of target sites and some clinical and commercial labs adopted these target sequences [9] . However, due to intellectual property protections, the exact target sequence of many commercial assays is currently unknown. A list of SARS-CoV-2 genes targeted by EUA assays is shown in Figure 2 . Notably, the most commonly targeted gene is the N gene followed by the long ORF1ab gene. The differences in genes targeted and the region of those genes targeted amongst the individual assays complicates direct comparison of their performance. PCR efficiency, which is determined by the logarithmic increase in amplified products (amplicons), can markedly differ even within a single assay with two unique targets. Often, manufacturers will include highly sensitive gene targets that are less specific for SARS-CoV-2 (and can amplify SARS-CoV and MERS-CoV) along with highly specific but often less sensitive targets, however, as described earlier, given the lack of circulating SARS-CoV and MERS-CoV, during the current pandemic, a presumptive positive is clinically equivalent to a positive result.

Rheumatologists in private practice have a variety of testing options to select from if they suspect their patient may have COVID-19. Testing that can be performed at the bedside is often preferred for a variety of reasons, including patient satisfaction, rapid turnaround time, and immediate guidance for clinical management. This type of testing parallels the efforts previously achieved with rapid Group A strep and Influenza testing, performed within minutes of sample collection. One of the first offerings in such POC testing included the EUA approved Abbott ID Now and Cepheid GeneXpert tests for SARS-CoV-2. A laboratorian's visceral reaction to POC testing is often supported by difficulties encountered with the safe and effective deployment of testing outside a controlled laboratory environment. For molecular testing, these issues are "amplified" as even a slight lapse in strict controls can lead to disaster. The foremost issue is environmental contamination and the adherence to necessary mitigation strategies. The Abbot ID Now assay is an open system where the NP swab is placed directly into the instrument, mixed vigorously, and nucleic acid amplified. The vigorous mixing has a risk of generating aerosols that can lead to contamination of work surfaces, personnel, or the instrument itself. A dedicated biosafety cabinet can be used to minimize these risks, however, this is not common equipment found in physician offices. With these issues aside, the next obstacle is receiving a steady stream of testing equipment and reagents. The high demand for such testing limited and continues to limit access to POC instrumentation. Finally, the sensitivity of the POC assays have been scrutinized and widely debated, especially for the Abbott ID Now [10] [11] [12] [13] . If used correctly, though, the sensitivity of the Abbott ID Now was still one of the poorest for all devices on the market with FDA approval (Table 1) . For patients with low viral loads that lack the potential to mount sufficient protective responses (e.g. immunocompromised [14] ) or those that can potentially spread an undiagnosed infection to others, such testing may be inadequate. The challenge, of course, is to be able to predict which patients these are. While such low sensitivity has limited implementation of these assays in the hospital setting, it is clear that private practice groups may not be aware of these shortcomings or choose to simply ignore them as often any test is better than no test.

Until very recently, it was unclear how individual assays performed against each other in terms of sensitivity. Case reports emerged early regarding specific platforms but a systematic analysis was only recently performed [15] . For this comparison, the US FDA mailed out SARS-CoV-2 reference panels to labs who had been approved for clinical testing through the EUA process. As shown in Table 1 , stated limits of detection (LODs) found in package inserts ranged in terms of units (copies/mL, copies/reaction, genome equivalents (GE)/mL, or TCID 50 /mL) and value, while the FDA-confirmed LODs provided the first indication of relative sensitivity. The implications for these studies are still not fully understood and whether manufacturers or labs will be required to cease testing because of lower sensitivity is also unclear. The clinician is generally unaware of the multiple analytical platforms their patient could have been tested on and will sometimes ask about the general sensitivity of the assay when a patient they are treating seems to have symptoms consistent with COVID-19 yet test negative. Often, clinicians will request the CT value to better understand the viral load in their patient. It is important to reiterate that there are no quantitative tests for SARS-CoV-2 currently approved by the FDA. Unlike blood-borne viral pathogens, including HIV-1 and HCV, it is impossible to standardize the specimen collection for any respiratory viral pathogen. While a high CT value (low viral load) may prompt a clinician to treat a patient less aggressively, the correlation between CT and severity has not been adequately established [16, 17] . In fact, recent reports of asymptomatic individuals with high viral loads complicates our ability to predict severity from CT values [18] . Transmissibility, however, may be linked to viral load and studies are underway to investigate this further [19, 20] . Until the clinical impact of acting upon CT values is better understood, clinicians are cautioned against using such information to guide patient management.

There are many reasons why a patient may test negative unexpectedly. While the actual analytical sensitivity is an obvious reason (patient's viral load may be too low to be detected by the chosen platform), other reasons include more common preanalytical issue that challenge all areas of laboratory medicine (Fig 3) . First, specimen source can affect the outcome of testing. Various testing options available to the clinician for SARS-CoV-2 molecular testing may or may not include all possible specimen types ( Table 2 ) [21] . We have observed patients in which the only positive specimens are retrieved from the lower respiratory tract despite multiple attempts at testing NP swabs [22] . This could reflect the biology of the virus in certain individuals or the clinical disposition of the patient. The emergence of saliva testing as an alternative to the uncomfortable NP swab procedure may impact the rate of false positives as well [23] . Inadequacy of specimen collection [24] cannot always be determined by the assay depending on whether a specimen adequacy control was included (see Molecular Detection of SARS-CoV-2). Additionally, if patients self-medicate nasal passages with ointments or topical creams, the molecular testing may be inhibited, leading either an invalid result or possibly a false negative. Other clinical factors impacting pre-test probability include whether the patient was symptomatic for COVID-19 and/or had close contact with someone who was infected. The timing of collection relative to the onset of symptoms also can have an effect on test results, as can the time it takes from collection to lab testing [25] . Most SARS-CoV-2 molecular tests require refrigeration for no more than 72 hrs from collection and room temperature storage for even less time. Beyond these timeframes, if the specimen is not frozen, degradation of the viral RNA can occur. Finally, as supply chain issues have plagued all areas of laboratory testing, collection devices are no exception. The lack of universal transport media (UTM) has necessitated the use of alternative transport media including viral transport media (VTM) containing various antibiotics, normal saline, and phosphate buffered saline (PBS) [26] . The equivalency of these alternative transport media needs to be confirmed by the individual lab if not already included in the manufacturer's instructions for use. Even swabs have been in short supply, requiring clinicians to become creative with choice of swabs that are flexible enough to be used in NP sampling, without risking additional harm to patient or being incompatible with molecular testing altogether. In summary, negative results can occur even if the patient has an ongoing SARS-CoV-2 infection, and the laboratory testing component is but one factor in the entire process.

One constant premise laboratories have leaned on is that molecular diagnostics for SARS-CoV-2 changes weekly, if not daily. For example, as regulatory guidance is developed and subsequently modified, the role of the EUA process and FDA oversight has become less clear [27] . While relevant, the complexities of regulatory bodies governing laboratory testing is beyond the scope of the current review, yet heavily influences the availability of quality testing products and suppliers [28] . While it is too early to predict their effectiveness, creative solutions are being considered for a variety of testing issues. These include surveillance testing where individual patient reports are not generated in favor of summary statistics [29] . Additional developments include the ability to pool patient testing when prevalence is low, thus helping to conserve valuable and often limited testing resources [30] . Finally, as we approach the first influenza season, molecular diagnostics incorporating analytes for both COVID-19 and other respiratory infections are just now entering the market [31, 32] . Depending on what the flu season will look like, these may become critical factors in discriminating the types of infections present and triaging to specific therapies. Ultimately, once vaccination is available and broadly dispersed, the role of molecular testing may be less critical, whereas sensitive serological or antigen tests may play a unique role in patient management. In summary, molecular detection of SARS-CoV-2, a single stranded RNA virus, has faced and will continue to be confronted by numerous logistical challenges unseen by the practicing clinician that rapidly impact options for patient testing.

In contrast to molecular testing which has had clear diagnostic utility from the onset of the COVID-19 pandemic, the utility of serological assays has been more convoluted [33, 34] . This is primarily due to the rapid emergence of serological assays, which has outpaced scientific understanding of their clinical utility. The rapid proliferation of serological assays for SARS-CoV-2 was due at least in part to the FDA's initial decision to not require EUA for their distribution. The reasoning behind this was that these assays were not meant to be diagnostic and that the assays would be used primarily by high complexity laboratories [35] . However, many of the assays distributed were from relatively unknown vendors using lateral flow devices that resemble pregnancy tests, many of which possessed poor performance characteristics [36] . However, potential for widespread misuse was perpetuated by calls from several entities including the White House Coronavirus Task Force [37] and from health policy experts in high profile journals [38] arguing for widespread serological testing for SARS-CoV-2 to demonstrate immunity and allow for return to work, etc. However, other major entities including the CDC, WHO, Infectious Diseases Society of America (IDSA), and several laboratory experts have strongly cautioned against rapid and widespread implementation given the unknowns of serological testing including test performance, clinical utility, association with protection, mechanism of protection, and the duration of protection [33, 34, [39] [40] [41] . As a result, the FDA promptly reversed course on May 4 th , 2020 requiring all serological assays to have EUA and meet certain performance characteristics [42] .

At time of writing, there are 41 serological assays available approved under the EUA in the United States [43] . For most clinical purposes, the assays that are available have comparable performance for the detection of antibodies to SARS-CoV-2. Despite this, the assay design varies considerably between manufacturers. Supply chain issues that have at times plagued molecular testing have not affected J o u r n a l P r e -p r o o f serological assays for SARS-CoV-2. This may be due to the lessened demand for serology relative to diagnostic testing.

Commercially available assays may take several formats; 1) chemiluminescent immunoassays that are generally high throughput and run on large, commercially available analyzers available in most clinical laboratories, 2) enzyme-linked immunosorbent assays (ELISA) which are typically performed on 96-well plates in a manual or semi-automated fashion and 3) lateral flow assays which resemble pregnancy tests. There are potential benefits and detriments to each assay design. For example, chemiluminescent immunoassays generally have high reproducibility and larger throughput than other methods. However, they do not generate titers, which are the gold standard when assessing protection and response to infection [44] . ELISAs can be easily titered, but their lower throughput, especially when generating a titer is a major limitation in most clinical laboratories. Finally, lateral flow-based assays may potentially be useful, particularly in seroprevalence studies or in low resource areas. However, these assays frequently lack sufficient clinical sensitivity and specificity for detection of antibodies to SARS-CoV-2 [36] . Importantly, only one serological assay has been cleared for use at POC at time of writing [43] . Hospitals may use EUA lateral flow serological methods that are considered moderately complex at the bedside under a CLIA license. However, sites such as physicians' offices that perform testing under a CLIA Certificate of Waiver are only authorized to perform lateral flow testing on devices authorized for POC testing [45] . Finally, neutralization assays, which detect the presence of neutralizing antibodies to SARS-CoV-2 have also been developed [46] . However, neutralization assays require either Biosafety level 3 facilities using relatively low throughput methods or the creation of a pseudovirus-based assay. No neutralization assays have been approved by the FDA for clinical use at this time.

Available assays detect anti-SARS-CoV-2 IgG, IgM, IgA or total antibody. When assessing previous exposure to SARS-CoV-2, there is currently no known benefits to any one specific assay design [47] . Frequently after exposure to a virus, seroconversion with IgM occurs several days to weeks before IgG leading some to speculate that IgM would provide enhanced sensitivity for detection of acute SARS-CoV-2 infections. However, this has not been proven to be true in patients with COVID-19. In a study of 26 patients tested longitudinally, 9/26 patients seroconverted IgG and IgM simultaneously and 10/26 seroconverted IgG prior to IgM [48] . Similarly, Zhao et al. observed near simultaneous median time to positivity of 12 days for IgM and 14 days for IgG using assays not yet available in the US [49] . Furthermore, total antibody assays which identify several classes of antibodies have similar clinical sensitivity for detection of anti-SARS-CoV-2 antibodies relative to assays which only detect IgG. Tang et al. demonstrated comparable performance between the Roche anti-SARS-CoV-2 total immunoglobulin assay, the Abbott anti-SARS-CoV-2 IgG assay, and the EUROIMMUN anti-SARS-CoV-2 IgG assay at >14 days post symptom onset and at <14 days post symptom onset [50, 51] . Similarly, Harb et al. observed comparable sensitivities in convalescent plasma between assays that target total anti-SARS-CoV-2 immunoglobulin and those that target IgG. These studies imply that neither total immunoglobulin nor the detection of IgM improve the sensitivity of serological assays for early detection of patients with COVID-19 infections. It is important to note that the IDSA recommends against the use of assays for SARS-CoV-2 IgA. This is primarily due to the low observed sensitivity and specificity [52] . The IDSA also recommends against the use of combination IgG or IgM tests which are assessed separately, but wherein only one of the two are positive [53] . This is primarily due to concerns of enhanced cross reactivity of IgM and lower specificity relative to IgG, reducing overall specificity. Manufacturers' claims for sensitivity and specificity can be found on the FDA website [43] , but independent studies have demonstrated lower sensitivities in hospitalized patients likely due to differences in patient populations.

Commercially available serological assays for SARS-CoV-2 also detect antibodies to different viral antigens. The vast majority of assays detect antibodies directed against the nucleocapsid protein, the spike protein, or the receptor-binding domain (RBD) region of the spike protein. While the nucleocapsid protein tends to be more highly immunogenic than the spike protein [47] , studies have yet to demonstrate conclusive differences between clinical assays that target antibodies to different SARS-CoV-2 proteins [50, 54] . Although the nucleocapsid protein is more highly conserved across coronaviruses than the spike protein, clinical assays have demonstrated similar specificity. As a result, the IDSA and the CDC make no recommendation at this point regarding the antigenic target used [53, 55] .

Despite the availability of serological assays for SARS-CoV-2 exceeding several months, the clinical utility is still relatively narrow. The proposed utilities include 1) diagnosis of acute infection, 2) seroprevalence studies and 3) determining protection after previous exposure.

Early in the course of the pandemic, serological assays were proposed to be an important supplement to diagnostic testing [56] . This was due to supply chain issues cited previously for diagnostic molecular methods and numerous appeals that serology may play a role for diagnosing acute infection [49, 57] However, serological assays have poor sensitivity for detection of antibodies to SARS-CoV-2 early after symptom onset. Theel et al. previously demonstrated sensitivities of <50% and <11% at 8-14, and <8 days post symptom onset respectively using the Abbott, Epitope, EUROIMMUN, and Ortho-Clinical SARS-CoV-2 serological assays [54] . Similarly, Tang et al. demonstrated sensitivities of ~42% with Roche, ~31% with Abbott, and 33% with the EUROIMMUN SARS-CoV-2 assays in patients with <14 days postsymptom onset. As previously noted, detection of IgM antibodies to SARS-CoV-2 does not seem to provide enhanced clinical sensitivity. At our institution, patients with SARS-CoV-2 typically present to the ED within 3-4 days from symptom onset, well before seroconversion would be anticipated. Therefore, it is not advised to use SARS-CoV-2 serological assays for assessing the presence of acute infection [53, 55] . However, some symptomatic patients may present a week or later after symptom onset and are persistently negative by PCR. The CDC recommends that serological testing is used as an adjunct to diagnostic molecular testing if a patient presents more than 9 days from symptom onset [55] . However, the IDSA recommends against the use of serology until 14 days after symptom onset, citing a pooled sensitivity of 68% in patients days 7-14 days [53] . Pediatric patients with multisystem inflammatory syndrome, a disease presenting with Kawasaki-like features including fever and shock, may benefit from the use of serological assays to confirm diagnosis. In a study of 95 confirmed cases in NY, 47% were diagnosed by positive serology in the absence of positive molecular testing [58] . Thus, while serological assays should not be used for acute diagnosis, they have limited clinical utility in a subset of patients.

Unsurprisingly, patients that are immunocompromised often fail to seroconvert or have longer time to seroconversion relative to immunocompetent patients after SARS-CoV-2 infection. A study of 21 patients with chronic lymphoblastic leukemia revealed only 67% seroconverting to IgG at a minimum of 28 days following symptom onset [59] . Similarly, studies have demonstrated lower seroconversion rates among SARS-CoV-2 infected cancer patients relative to health care workers [60] . However, little is available in the peer-reviewed literature among patients with rheumatic diseases. A case study of two patients with MS treated with ocrelizumab both failed to mount an immune response at 6 and 7 weeks after symptom onset [61] . Importantly, the presence of autoimmune diseases and potentially cross reacting antibodies does not seem to effect the specificity of serological assays for SARS-CoV-2 [50, 54, 62] . Further studies are needed to confirm the time to seropositivity in this potentially at-risk population and the sensitivity of commercially available assays in immunosuppressed patients. However, given the likelihood of a lower test sensitivity, negative results from a serological assay should be interpreted with caution in patients receiving immune modulating therapies.

Seroprevalence studies SARS-CoV-2 serological assays have been used to assess the seroprevalence within a population. These studies have important ramifications for public health policy including understanding the burden of disease particularly due to limited early testing by molecular methods and to quantitate the mortality rate among infected individuals. Several studies have demonstrated that the seroprevalence of SARS-CoV-2 is 2-10 times greater than the number of patients that have been reported positive by molecular methods [63, 64] . For seroprevalence studies, it is crucial for the test to have high specificity and positive predictive value (PPV), particularly for an emerging virus with a relatively low prevalence [34] . For example, in a population of 1,000,000 people with a prevalence 1%, a test with a sensitivity and specificity of 99% respectively would yield PPV of ~50%. Thus, 1 in 2 positive results would be a false positive. As a result, the CDC advocates for the use of an assay with a specificity >99.5% or an orthogonal approach by which all positive results are tested again using secondary method [55] . While seroprevalence studies have limited clinical application for individual patients, many patients are interested in their serostatus. Likely as a result of marketing from manufacturers and underlying public interest, the vast majority of serological tests at our institution are performed in outpatient settings. However, in settings with low pre-test probability and low prevalence (i.e. outpatient physician office in a patient with limited previous symptoms), a method with high specificity is required, similarly to seroprevalence studies. Therefore, it is important for clinicians to understand the limitations of the assay (i.e. known specificity) and the approximate prevalence in the area when serological assays are used for assessing previous exposure in asymptomatic patients.

If SARS-CoV-2 serological testing is performed, it is important to provide clear information to the patient regarding the utility of the result, particularly when positive. One of the most important discussions regarding COVID-19 serology is if a positive result equates to protection from future SARS-CoV-2 infections. This has been phrased as an "immunity passport," which would permit previously infected patients with the presence of antibodies to travel, return to work, and generally resume life as normal with the presumption of immunity. However, the degree of protection and duration of immunity offered by previous exposure to SARS-CoV-2 is still relatively unknown. As a result, the CDC, IDSA, and WHO all recommend against the use of serological assays for determining immune status [53, 55, 65] . Nonetheless, there is mounting evidence that infection with SARS-CoV-2 confers some degree of protection. Rhesus macaques infected with SARS-CoV-2 were protected from reinfection 35 days following the initial exposure [66] . Interestingly, three fisherman that were previously infected with SARS-CoV-2 had no evidence of viral infection and experienced no symptoms after subsequent exposure from an outbreak on a fishing vessel [67] . Specimens from the exposed fisherman that were drawn prior to re-exposure all demonstrated the presence of neutralizing antibodies. This implies that exposure and the generation of neutralizing antibody titers are sufficient for protection from reinfection. However, there are several limitations with presuming that positive serological results for anti-SARS-CoV-2 assays equates to long-term immunity. For example, Tang et al. previously demonstrated that hospitalized patients that died or had worse outcomes had higher neutralizing antibody titers than hospitalized patients with improved outcomes, implying that neutralizing titers may not associate with improved outcomes or protection [68] . Furthermore, a study of convalescent plasma from 149 individuals revealed neutralizing titers <1:50 in 33% of patients [69] . In contrast, the early recommendation from the FDA was a minimum neutralizing titer of 1:160 with an ideal of 1:320 for convalescent plasma donors [70] . Moreover, neutralizing titers may not be the mechanism of protection, with several studies demonstrating the importance of T cells [71, 72] . Another problem with inferring protection from future SARS-CoV-2 infection with serological assays is that they serve as an imperfect proxy for neutralization. One study found a negative percent agreement (NPA) of <55% or less using a neutralizing cutoff of 1:128 as positive compared to positive serological results from three high throughput, commonly used clinical assays [68] . Similarly other authors demonstrated a NPA of 32% with neutralizing assays relative to a commercially available ELISA [73] . While there is a modest correlation between the signal generated on commercially available serological assays and neutralizing assays, most assays are qualitative. While quantitative SARS-CoV-2 serological assays are emerging, correlations are required with these assays relative to neutralizing antibody titers. Another remaining question is the durability of the immune response to SARS-CoV-2. Several studies have demonstrated that circulating antibody concentrations decrease considerably within the first 90 days from symptom onset, particularly in patients with mild and asymptomatic infections [74] [75] [76] . Importantly, longitudinal studies of seasonal coronaviruses have found that patients are frequently reinfected with the same coronavirus, often within 12 months of the previous infection. Longitudinal studies assessing the durability of antibodies to SARS-CoV-2 are needed.

As the pandemic resumes, serological assays are being developed and used for various purposes. Serological assays have been implemented to identify convalescent plasma donors with presumably high titers of antibodies. Early results from the expanded access program for convalescent plasma have demonstrated that patients that receive convalescent plasma units with higher levels of antibody as identified by the Ortho-Clinical SARS-CoV-2 IgG assay have better outcomes relative to those with lower levels of antibodies [77] . Serological assays may also provide value in identifying vaccinated individuals and the sufficiency of the humoral response once a vaccine is available. Since vaccinations target portions of SARS-CoV-2 spike protein, it will be crucial for providers to know the antigenic target of the assay when used for this purpose. Finally, algorithm-based approaches using serological assays that target multiple antigens and immunoglobulin classes may be useful in future. An algorithmic approach is used for hepatitis B virus serology which allows for discrimination between recent infection, previous infection, and immunity [78] . However, no such algorithm has been proposed or endorsed by professional societies for SARS-CoV-2.

SARS-CoV-2 is an RNA betacoronavirus that is responsible for the current COVID-19 pandemic. Laboratory evaluation for diagnosis and surveillance has been rapidly developed to assist treatment public health efforts to prevent its spread. However, with the heightened awareness that laboratory testing is crucial for this function, hundreds of platforms have been made available in relatively short period of time. Many of these platforms have distinct advantages to offer, including turnaround time, sensitivity, types of specimens accepted, ease of use, availability and cost. False negative results can occur due to a number of factors. These can analytical but also pre-analytical, before the specimen even arrives in the lab. Molecular-based testing is the only diagnostic assay format authorized by the FDA, but it is still only qualitative, and can vary greatly in assay sensitivity. Serological testing is more useful for monitoring seroprevalence, while its role in assessing protective immunity still under investigation. In addition to laboratory-based testing options, clinicians will continue to be presented with numerous point-of-care assays to choose from, and these will suffer from similar challenges face by the central laboratory. While we have made tremendous progress in making effective testing available to those most in need of it, we continue to face unanswered questions related to the future role of molecular and serological testing. For example, we do not fully understand how viral load is associated with clinical outcomes and if a quantitative test would therefore be useful. Still unknown is how the durability of the serologic response differs in natural infection compared to vaccination. Importantly, the humoral immune response to SARS-CoV-2 in patients with rheumatologic diseases and on immune modulating therapies requires further evaluation. Laboratory testing for SARS-CoV-2 will continue to be confronted by numerous logistical challenges that influence options for patient testing.

• The sensitivity of FDA EUA molecular diagnostic assays varies widely and depending upon the platforms and may not accurately capture individuals with low viral loads. • False negative test results can derive from issues within the laboratory or outside the clinical testing environment, such as inadequate sample quality.

• The clinical utility of serological assays for SARS-CoV-2 is currently relatively limited but may aid in diagnosis in rare cases.

• Prospective studies describing the association between quantitative molecular diagnostic results and outcomes are needed to interpret the significance of the viral load. • Association between protection from future SARS-CoV-2 infections and serological assays positive for the presence of antibodies to SARS-CoV-2 require further studies, particularly those that assess the durability of the serologic response.

• The development of algorithm-based approaches to serological testing to assess for acute infection, chronic infection, and vaccination may provide enhanced value to serological testing to SARS-CoV-2.

• Further studies are required assessing the humoral immune response to SARS-CoV-2 in patients with rheumatologic diseases and on immune modulating therapies. Table 1: EUA Assay Sensitivity and Instructions for SARS-CoV-2 Molecular Assays 

The species Severe acute respiratory syndrome-related coronavirus : classifying 2019-nCoV and naming it SARS-CoV-2

SARS: how a global epidemic was stopped

Comparing SARS-CoV-2 with SARS-CoV and influenza pandemics

Nidovirales: Evolving the largest RNA virus genome

Emerging coronaviruses: Genome structure, replication, and pathogenesis

Emerging SARS-CoV-2 mutation hot spots include a novel RNA-dependent-RNA polymerase variant

Testing For SARS-CoV-2: The Day the World Turned its Attention to the Clinical Laboratory

FDA

Protocol-World Health Organization

Comparison of Cepheid Xpert Xpress and Abbott ID Now to Roche cobas for the Rapid Detection of SARS-CoV-2

Performance of Abbott ID Now COVID-19 Rapid Nucleic Acid Amplification Test Using Nasopharyngeal Swabs Transported in Viral Transport Media and Dry Nasal Swabs in a New York City Academic Institution

point-of-care antigen and molecular-based tests for diagnosis of SARS-CoV-2 infection

How many are we missing with ID NOW COVID-19 assay using direct nasopharyngeal swabs? Findings from a mid-sized academic hospital clinical microbiology laboratory

COVID-19 in Immunocompromised Hosts: What We Know So Far

CoV-2 Reference Panel Comparative Data. The FDA SARS-CoV-2 Reference Panel

Clinical Course and Molecular Viral Shedding Among Asymptomatic and Symptomatic Patients With SARS-CoV-2 Infection in a Community Treatment Center in the Republic of Korea

SARS-CoV-2 viral load predicts COVID-19 mortality

Unclear but present danger: An asymptomatic SARS-CoV-2 carrier

Culture-Based Virus Isolation To Evaluate Potential Infectivity of Clinical Specimens Tested for COVID-19

Predicting infectious SARS-CoV-2 from diagnostic samples

The SARS-CoV-2 Test & Controls Database

The Brief Case: "Not Positive" or "Not Sure"-COVID-19-Negative Results in a Symptomatic Patient

Saliva in the Diagnosis of COVID-19: A Review and New Research Directions

SARS-CoV-2 Nasopharyngeal Swab Testing-False-Negative Results From a Pervasive Anatomical Misconception

Value of swab types and collection time on SARS-COV-2 detection using RT-PCR assay

Comparison analysis of different swabs and transport mediums suitable for SARS-CoV-2 testing following shortages

Department of Health & Human Services. Rescission of Guidances and Other Informal Issuances

Regulation of Laboratory-Developed TestsA Clinical Laboratory Perspective

Information for Laboratories about Coronavirus (COVID-19)

Evaluation and Comparison of the Hologic Aptima SARS-CoV-2 and the CDC 2019 nCoV real-time RT-PCR Diagnostic Panel using a Four-Sample Pooling Approach

Clinical evaluation of AusDiagnostics SARS-CoV-2 multiplex tandem PCR assay

Evaluation of the QIAstat-Dx Respiratory SARS-CoV-2 Panel, the First Rapid Multiplex PCR Commercial Assay for SARS-CoV-2 Detection

The Role of Antibody Testing for SARS-CoV-2: Is There One?

SARS-CoV-2 Serology: Much Hype, Little Data

Emergency Use Authorizations for Medical Devices 2020

Test performance evaluation of SARS-CoV-2 serological assays

The White House 2020

Waiting for Certainty on Covid-19 Antibody Tests -At What Cost?

Tests for COVID-19 -Letter to Health Care Providers

Insight into FDA's Revised Policy on Antibody Tests: Prioritizing Access and Accuracy

Authorized Serology Test Performance

Effectiveness of convalescent plasma therapy in severe COVID-19 patients

CAP Responds to Your COVID-19 Questions

Development and validation of IMMUNO-COV TM : a high-throughput clinical assay for detecting antibodies that neutralize SARS-CoV-2

ORF8 and ORF3b antibodies are accurate serological markers of early and late SARS-CoV-2 infection

Antibody responses to SARS-CoV-2 in patients with COVID-19

Antibody responses to SARS-CoV-2 in patients of novel coronavirus disease 2019

Clinical Performance of Two SARS-CoV-2 Serologic Assays

Clinical Performance of the Roche SARS-CoV-2 Serologic Assay

Evaluation of the EUROIMMUN Anti-SARS-CoV-2 ELISA Assay for detection of IgA and IgG antibodies

Infectious Diseases Society of America Guidelines on the Treatment and Management of Patients with COVID-19 2020

Performance Characteristics of Four High-Throughput Immunoassays for Detection of IgG Antibodies against SARS-CoV-2

Information for Laboratories about Coronavirus (COVID-19)

Serology assays to manage COVID-19

Evaluation of Nucleocapsid and Spike Protein-Based Enzyme-Linked Immunosorbent Assays for Detecting Antibodies against SARS-CoV-2

Multisystem Inflammatory Syndrome in Children in New York State

Anti-SARS-CoV-2 antibody response in patients with chronic lymphocytic leukemia

Lower detection rates of SARS-COV2 antibodies in cancer patients versus health care workers after symptomatic COVID-19

Negative SARS-CoV-2 antibody testing following COVID-19 infection in Two MS patients treated with ocrelizumab

Detection of IgM and IgG antibodies against SARS-CoV-2 in patients with autoimmune diseases

Seroprevalence of Antibodies to SARS-CoV-2 in 10 Sites in the United States

Humoral Immune Response to SARS-CoV-2 in Iceland

SARS-CoV-2 infection protects against rechallenge in rhesus macaques

Neutralizing antibodies correlate with protection from SARS-CoV-2 in humans during a fishery vessel outbreak with high attack rate

Association between SARS-CoV-2 neutralizing antibodies and commercial serological assays

Convergent antibody responses to SARS-CoV-2 in convalescent individuals

FDA. Recommendations for Investigational COVID-19 Convalescent Plasma. FDA 2020

Antigen-specific adaptive immunity to SARS-CoV-2 in acute COVID-19 and associations with age and disease severity

Robust T cell immunity in convalescent individuals with asymptomatic or mild COVID-19

Sex, age, and hospitalization drive antibody responses in a COVID-19 convalescent plasma donor population

Antibody responses to SARS-CoV-2 short-lived

Clinical and immunological assessment of asymptomatic SARS-CoV-2 infections

Longitudinal evaluation and decline of antibody responses in SARS-CoV-2 infection

Effect of Convalescent Plasma on Mortality among Hospitalized Patients with COVID-19: Initial Three-Month Experience

CDC. Recommendations for Routine Testing and Follow-up for Chronic Hepatitis B Virus (HBV) Infection | CDC 2019