key: cord-0800828-uf4494ka
authors: Patel, Rashmi; Khare, Siddharth; Mahajan, Vinay S.
title: Alternative methods to detect SARS-CoV-2 antibodies
date: 2021-11-03
journal: Clin Lab Med
DOI: 10.1016/j.cll.2021.10.007
sha: 1079735e434809cfb9fa1d22909e7d74ac2d9c46
doc_id: 800828
cord_uid: uf4494ka

The coronavirus disease 2019 (COVID-19) pandemic has resulted in the development, validation, and rapid adoption of multiple novel diagnostic approaches across the world. As a result, hundreds of SARS-CoV2 serological assays have been developed and deployed across the world in the effort to contain the spread of the virus, as well as to supply timely and important health information. Drawing upon decades of experience, most of these serological assays were based on a conventional ELISA or the lateral flow assay format. The immunoassays that were developed were based on alternative technologies and are highlighted in this chapter with a brief discussion of the assay principle and the pros and cons for each assay. In addition, the principle and approach to the measurement of neutralizing antibodies is also discussed.

The SARS-CoV2 pandemic resulted in an intense demand for serological diagnostics, leading to the development of hundreds of assays across the world, especially lateral flow assays (LFAs). The COVID-19 test directory on the FindDx website includes an exhaustive catalogue of these assays 1 . By June 2021, the US Food Drug Administration (FDA) had issued Emergency Use Authorizations (EUA) for 80 serological tests for SARS-CoV2 2 . These ranged from rapid qualitative LFAs to semi-quantitative ELISAs and include assays that can be performed on fully-automated laboratory analyzers. In most COVID infected patients, antibodies are observed approximately 1-2 weeks following symptom onset or PCR positivity in symptomatic or asymptomatic individuals respectively 3, 4 . The observed timing of seroconversion also depends upon the sensitivity of the assay. Seroconversion may be picked up as early as the day of the first positive nucleic acid test after symptom onset with ultrasensitive single-molecule approaches (Quanterix Simoa) 5 . Unlike typical seroconversion profiles in other infectious contexts, near-simultaneous production of IgM, IgG and IgA has been observed in patients with confirmed SARS-CoV-2 6, 7 . While IgM titers may disappear within a month, IgG titers are detectable for much longer but also exhibit a gradual decline in the months following infection [8] [9] [10] . Higher antibody titers are seen following symptomatic or severe disease 11 .

The serological assays that have been developed have focused on the structural proteins of the virus, i.e., spike and nucleocapsid. Spike is a transmembrane glycoprotein comprising two parts, S1 and S2. The binding of the SARS-CoV-2 spike to its receptor, angiotensin-converting enzyme 2 (ACE2), is mediated by S1. S2 mediates fusion of the viral envelope to the cell membrane and cell entry during infection. The S1 receptor binding domain (RBD) binds ACE2 and is highly immunogenic. The spike protein contains sequences unique to SARS-CoV2 as well as shared with other beta-coronaviruses. Thus, assessing the serological response to antigens from other human coronaviruses can help exclude cross-reactive responses. Determining the antibody isotype (e.g., IgM, IgG or IgA) may provide additional information for determining immune status. Unlike natural infection, most vaccines induce anti-spike but not antinucleocapsid responses and IgG rather than IgA. These characteristics may help distinguish between vaccine-induced responses and natural infection. Vaccines are designed to elicit antibodies against the S1 RBD as antibodies to this region can neutralize the virus. Thus, in addition to antibody specificity, a complete serological characterization involves determining the isotype and neutralization potential.

There is a great interest in using serological parameters to determine infection risk, vaccine efficacy or vaccine prioritization. However, serological assays do not detect the presence of memory B or T cells, and it is possible that memory lymphocytes may offer some immunity in subjects with declining antibody titers 12 . The relationship between seropositivity or neutralization titers and protection remains unknown and is being evaluated in clinical and epidemiological studies. Given the rise of SARS-CoV2 variants as well as the variable magnitude of serological responses among convalescent individuals, the use of serology as a surrogate for protection is likely to remain a controversial and changing landscape. Due to the above uncertainties, the CDC and FDA have advised against the use of serological assays for assessing protection, and interpretive guidance is imperative while reporting a serological test result 13 . Given their utility in a point-of-care setting, LFAs for antibodies have been widely used in population serosurveys to estimate exposure rates and guide public health policy. Seroconversion indicates prior exposure but not active infection, and LFAs that detect viral antigens rather than antibodies should be used to determine infectious risk 14 .

Although manufacturers monitor performance using traceability to recognized standards, independent monitoring of assay performance by clinical laboratories was especially vital in ensuring that these assays were effectively utilized during a time that witnessed widespread supply chain disruptions with the potential to impact assay manufacturing and distribution. Indeed, when concerns were raised about the reliability of certain LFAs, the EUA for some assays were revoked by the FDA. Independent vigilance of serological assays by clinical laboratory practitioners continues to be very important as the antigen formulation used in the serological tests has remained unchanged despite the emergence of SARS-CoV2 variants. In particular, the interpretation of neutralization assays may be significantly impacted by SARS-CoV2 variants. Variants that may have an enhanced capacity to evade neutralizing antibodies are referred to as variants of concern (VoC) 15 . Such variants are likely to keep emerging as the virus continues to evolve in the face of immune pressure when herd immunity builds up in a population 16 . Although the FDA has issued an EUA for one ELISA-based qualitative neutralization assay (Genscript cPass), it is only intended to assess recent infection and its clinical applicability to determine the degree of immunity is not known. Most neutralization assays are high complexity tests and are performed by specialized laboratories. They have primarily been used by vaccine developers and need nuanced interpretation in the light of concerns about VoC's. The continued emergence of variants raises concerns about the need to update the assays to assess neutralization capacity against emerging variants. We have described neutralization assays in detail in a separate section.

There is a need for the development of reliable and clinically scalable multiplexed serological assays, suitable for point-of-care use. Some of these challenges may be addressed by adopting emerging alternate technologies. This chapter highlights a few novel and alternative approaches that rely upon electrochemical analysis, luminescent signals or label-free optical detection. Ultrasensitive and quantitative approaches, such as Simoa (Quanterix Technologies), have now entered the market, which has opened a range of novel possibilities in serological diagnostics, but these instruments are not widely available. However, they have also been used as a quantitative reference method to compare the performance of other serological assays across a broad range of antibody dilutions 17 . We emphasize that many of the approaches described in this chapter are still in development and are yet to receive regulatory clearance. This chapter is also not meant to be an exhaustive review of alternative approaches to serology; it is divided into sections based on the physical principle used for assay signal generation.

Broadly, a serological assay involves a mechanism for signal generation, signal amplification, and signal detection (Figure 1 ). If the unbound label can generate a signal on its own, it needs to be separated by washing prior to measurement. Such an assay is called "heterogeneous". If the label generates a signal only in presence of the analyte, the immunoassay can be performed in a washless or "homogeneous" format. If the signal can be quantitatively measured across a sufficiently wide dynamic range, the assay can be calibrated with a range of accepted standards to generate a quantitative result. Some approaches are based upon label-free or direct measurement of the physical or chemical changes induced by antibody binding, and do not require the use of a labeled secondary antibody as a detection probe.

After incubation of patient serum with the specimen and removal of unbound antibodies, the antigen-bound antibodies are measured using a labeled detection probe such as anti-IgG, or anti-IgM. Such a setup is called a sandwich immunoassay ( Figure 1 ). When the detection probe is labeled with an enzyme to amplify a signal-generating chemical reaction, it is called a sandwich ELISA (Enzyme-linked immunosorbent assay). ELISAs are typically coupled to an optical signal. Test sensitivity and specificity are established based upon comparison against established positive and negative samples based on PCR positivity. Prepandemic specimens have been used as negative controls. For all assays that were issued an EUA, the FDA has published estimates of sensitivity and specificity on its website 18 . However, it is important to note that these numbers are estimates with 95% confidence intervals and positive and negative predictive values depend upon the prevalence. The kinetics of seroconversion as well as the potential loss of titers upon convalescence should be considered while interpreting negative results.

The design of each assay step influences the overall signal-to-noise ratio and consequently the sensitivity and specificity of the test. SARS-CoV2 antigens, typically spike or nucleocapsid proteins, are used as capture probes for antibodies. Some assays include other human coronaviral antigens as specificity controls. Blocking of non-specific binding is especially important for minimizing the background signal in assays where antibodies bound to an antigen-coated surface are measured. The choice and source of the antigens may also influence their antigenicity and contribute to differences in performance between assays that detect the same antigen. For instance, the spike protein may be used as a capture probe in its fulllength form, or just the S1 domain or receptor-binding domain alone. However, the full-length spike protein is less stable than the S1 domain or RBD domain. The approach used for surface coupling may also impact antigenic stability and access to antigenic epitopes 19 . The antigens need to be stabilized especially for use J o u r n a l P r e -p r o o f in a point of care setting without refrigerated storage. Antigenicity is also influenced by glycosylation, and thus fully glycosylated antigens expressed in a mammalian host are likely to best capture the full breadth of the serological response.

Point-of-care LFAs as well as serological assays performed on central laboratory analyzers have been described in concurrent chapters in this volume. The rapid assay from NanoEntek that received an FDA EUA resembles a lateral flow assay but uses a microfluidic cartridge for precise control of fluid flow 20 . The fluorescent signal in the cartridge is read using a dedicated instrument with connectivity to the laboratory information system. ELISA kits which can yield quantitative results are typically manufactured as microparticle-based or 96-well microtiter-based immunoassays and require manual set up, washing and a plate reader. This involves significant infrastructure and operator skill and is suitable for medium throughput applications. Several COVID-19 serological assays for automated central laboratory analyzers have now become available and should be considered for high throughput operations. In this chapter, we have focused on serological assays that utilize alternate or unconventional detection methodologies. These assays may help bridge the gap between LFAs and central lab analyzers in terms of both assay performance and throughput. Although many of the approaches described in this chapter are proof-ofprinciple demonstrations and are yet to be commercialized or implemented in a clinical setting, a few have received an FDA EUA. The assays from Genaltye (Maverick Multi-Antigen Serology Panel) and Genscript (cPass) are two such examples and are among those described in this chapter 21, 22 . Among other serological assays that have received an EUA and rely upon an unconventional detection approach, the assays from MosaicQ and Luminex are noteworthy; they are solid-phase ELISAs that use light scattering or fluorescence properties of nanoparticles for sensitive or multiplexed detection 23, 24 . Both platforms can be combined with antigen tests. However, a specialized instrument is required for analysis.

This approach relies upon the functionalization of the surface of a working electrode in an electrochemical cell with desired antigens 25 . Antibody binding to the electrode surface results in quantitative changes in electrical properties such as impedance, which can be measured in a label-free and washless setup ( Figure  2) . Alternatively, antibody binding can be coupled to a reduction-oxidation (redox) reaction such as one catalyzed by the widely used ELISA label, horse-radish peroxidase (HRP), and the generated current can be measured in a low-cost setup (Figure 2) . A few such assays have been implemented in the context of COVID-19 serology as a proof-of-concept and are listed below.

The Amperial TM assay platform (Liquid Diagnostics) is designed to implement an ELISA assay with an electrochemical readout 26 . It uses microtiter plates fabricated with gold electrodes at the bottom of each well. First, the surface of the electrode is functionalized with the target antigen by ectropolymerizing a pyrrole solution containing native antigens. This entraps antigens in a native state on the surface of the electrode within a conductive polypyrrole hydrogel. Next a sandwich ELISA is performed with a HRPlabeled secondary antibody for detection (Figure 2 ). Upon addition of a redox substrate and the application of a voltage, the peroxidase reaction produces an electric current that serves as a quantitative assay readout. A sensitivity >88% and specificity >99.85% was observed with S1 antigen. The immobilization of native antigen in an electropolymerized polypyrrole hydrogel preserves its antigenicity and offers the flexibility to test various types of antigens. The assay is quantitative, and the current in nanoamperes can be read for the whole plate within 3 minutes. However, it is a conventional heterogeneous sandwich immunoassay involving multiple wash steps.

A scalable low-cost laser-engraving technique was used to fabricate a miniaturized arrangement of disposable printed graphene electrodes individually functionalized with S1 domain, nucleocapsid antigen, CRP and an anti-nucleocapsid antibody for simultaneously assessing the serological response, an inflammatory biomarker and virus detection respectively 27 . A graphene counter electrode and an Ag/AgCl reference electrode was also included to complete the biosensor circuit (Figure 2) . The low cost, high charge mobility, surface area and ease of bioconjugation make graphene an ideal material for biosensor electrodes. The biosensors are subjected to a conventional ELISA and the signal is amplified by HRPlabeled detection antibodies that produce an amperometric readout in the presence of a redox substrate. The device is linked to a compact battery-powered circuit that transmits the signal to a cellphone via bluetooth. Analytical sensitivity of 1pM was achieved with only 10 minutes of specimen incubation. The electrode was manually rinsed with wash buffers and incubated with the redox substrate, but this can be easily automated. Comparable sensitivity was seen with both saliva and serum. The study showed the potential of this approach in a quantitative, multiplexed assay that is suitable for point-of-care settings 26 .

(iii) Repurposed cellular impedance monitoring platform The Agilent xCELLigence system is an instrument designed to detect cellular impedance for real-time monitoring of cell cultures with label-free measurement of cellular function such as cell growth, shape or toxicity etc. The cells are grown and monitored in specially fabricated plates with an electrode at the base of each well. If, however, the same multiwell cell culture plates are used to perform an ELISA with coated antigens, the change in impedance as a result of antibody binding can be measured in a wash-less format 28 . This was tested with S1 and RBD antigens. Antibody binding led to a sharp increase in impedance followed by a gradual decay over several minutes. Antibody binding is rapidly detected in a washless format, within minutes. This was an experimental demonstration of electrochemical measurement of SARS-CoV2 serology by repurposing a device already in use in some research labs.

(iv) Label-free electrochemical immunoassay using a gold micropillar array electrode A 3D gold microelectrode array was fabricated and decorated with reduced graphene oxide, and then conjugated to the SARS-CoV2 spike protein; the 3D geometry of the electrode results in a stronger current and also permits immobilization of the antigen at a higher density, resulting in greater antibody capture 29 . S1 and RBD were tested and yielded a quantitative measurement range of 1 pM -10 nM. A change in impedance was detected within 3 seconds of incubation. The signal is saturated over 10nM concentration when all available binding sites on the sensor surface are occupied by the specific antibodies. A low-cost coin-sized portable electrochemical impedance spectroscopy (EIS) analyzer (Palmsens Sensit) connected to a smartphone was used for measurements. This approach offers label-free detection with a few microliters of blood. The biosensor can be regenerated and reused after removing the bound antibodies with pH 2.5 formic acid. The sensor has a high regeneration capability with a good signal output even after 9 regeneration cycles. The electrodes require specialized techniques for fabrication, but the assay characteristics are suitable for point of care use.

A simple label-free rapid paper-based electrochemical serology sensor was implemented with RBD antigen immobilized on printed graphene electrodes in a vertical flow assay format 30 . Antibody binding was detected using an electrochemical spectroscopy technique called square wave voltammetry using a portable PalmSens analyzer. In this technique, the signal-to-noise ratio increases by the square root of the scan rate. The improved signal is a function of the time between pulse application and the current measurement, and the change in the faradaic current is measured as peaks on the voltammogram. The test was performed with patient sera and compared with a home-made colorimetric LFA device, prepared using the same batch of RBD. It exhibited better sensitivity for detection of IgM compared to IgG due to its larger size. It is a low-cost, rapid qualitative test suitable for point-of-care use with greater sensitivity compared to conventional lateral flow assays, but requires an EIS analyzer for readout. Being a label-free approach, unlike conventional immuno-sandwich lateral flow assays, it is not impacted by the risk of false negatives due to the high-dose "hook effect" or prozone phenomenon.

Luminescent labels, which catalyze a light-emitting chemical reaction, are widely used for sensitive detection using ELISAs. Unlike colorimetric or fluorescence detection, a light source or excitation is not required, and this approach offers greater sensitivity, lower background, and a wide dynamic range using relatively inexpensive instrumentation. The collected photons are amplified into a current signal that is reported as relative light units. Several luminescent substrates for the widely used immunoperoxidase or J o u r n a l P r e -p r o o f alkaline phosphatase labels have been developed and they form the basis of most modern ELISAs. Bioluminescence refers to analogous reactions observed in living organisms that are catalyzed by specialized enzymes called luciferases acting on substrates called luciferins. The luciferase-luciferin systems have been adapted to build highly sensitive immunoassays. These assays have been implemented in a fluid-phase ELISA as well as homogeneous washless formats using antigen-luciferase fusion proteins, split luciferases or designer proteins. Their broad dynamic range allows measurement across decades of concentration without the need for sample dilutions. The reagent systems for luminescent reactions are more complex than those used for spectrophotometric and fluorometric analyses and require more careful control and low temperature storage. Furthermore, luminescence measurements must follow the kinetics of the reaction. Although rapid luciferin oxidation can produce a very bright signal, signal detection may need to be timed with the injection of the luciferin substrate.

Luciferase immunoprecipitation (LIPS) assays are heterogeneous liquid phase immunoassays (Figure 3A ) 31, 32 . Unlike ELISA, which is typically a solid phase assay, antibody-antigen binding takes place in the fluid phase in a LIPS assay, thus better maintaining the native antigen conformation. Luciferase-labeled antigens are expressed as recombinant fusion proteins. The ability to use crude cell lysates of mammalian cells transfected with an antigen-luciferase fusion protein expression vector without further purification greatly simplifies assay development. This was particularly helpful for rapid development of these assays in the early days of the pandemic when antigens were not readily available. It will also allow the rapid development of serological tests for antigens from variants of concern without the need for purified antigens. A single preparation of crude lysates containing antigen-luciferase fusion protein can be used for thousands of assays. However, various configurations of antigen-luciferase fusions need to be tested and optimized during assay development. Luminescence is measured upon addition of the luciferase substrate using a microplate reader, yielding a quantitative result with high analytical sensitivity and a wide dynamic range without the need for sample dilution. LIPS assays have been implemented using luciferase fused to nucleocapsid as well as spike proteins 31 . For IgG against nucleocapsid, the sensitivity and specificity were 100% at 14 days after the onset of symptoms (n=35). For IgG against spike, sensitivity of 94% and specificity of 100% was observed.

This approach uses rationally designed antibody biosensors based on split nanoluciferase fragments (SmBiT and LgBiT) fused to SARS-CoV2 antigens (Figure 3B) 33 . Because an antibody has two fragment antigen-binding (Fab) arms, incubating the specimen with a 1:1 mix of SmBiT and LgBiT biosensors will result in half of the antiviral antibodies binding LgBiT with one Fab arm and SmBiT with the other Fab arm. If the antigen binding orientation is such that it brings the LgBiT and SmBiT fragments into close proximity, it results in the reconstitution of an intact, enzymatically active nanoluciferase enzyme. This can be used for luminescence-based detection of antigen-specific antibodies in a homogeneous assay format. Sensors for both RBD and nucleocapsid were designed using this approach and shown to require less than 30 minutes to result, suitable for point of care applications 33 . It is sensitive enough for testing serum or plasma (>99% sensitivity) and to a lesser extent saliva (79% sensitive).

Although once considered impossible, the explosion in the knowledge of protein structure and folding has led to the realization of de novo designed proteins 34 . One such rationally designed modular protein biosensor (lucCage and lucKey) was built such that antibody binding to the biosensor is thermodynamically coupled to switching from a closed dark state and an open luminescent state ( Figure 3C) 35 . This biosensor can be used in a homogeneous assay format and was built with RBD as the antigen. Antibodies against RBD could be detected at a sensitivity of 15 pM with a signal over background of over 50-fold. The lucCage biosensor is based on thermodynamic coupling between defined closed and open states of the system, thus, its sensitivity depends on the free energy change upon the binding of the sensing domain to the target but not the specific binding geometry. This enables the incorporation of various binding modalities, including small peptides, globular miniproteins, antibody epitopes and de novo designed binders, to generate sensitive sensors for a wide range of protein targets with little or no optimization. For point of care (POC) applications, the system has the advantages of being homogeneous, no-wash, and gives a nearly instantaneous readout; the quantification of luminescence can be carried out with inexpensive and accessible devices such as a cell phone camera. The ability to modularly design sensors with identical readouts for diverse antigens could enable multiplexed serological assays using an array of different sensors. However, there is considerable variation between different sensors in the level of activation at saturating target concentrations.

Nanoscale changes resulting from the binding of a protein to an optical surface can be amplified and detected by specialized optical techniques or devices such as surface plasmon resonance, biolayer interferometry, optical cavities or resonators 36, 37 . They have been commercialized into benchtop instruments and we highlight two approaches that have been applied to serological diagnosis of SARS-CoV2. Semiconductor-based photonic technologies are rapidly evolving, and such devices may soon become more commonplace.

Bio-layer interferometry is used to measure the binding of a protein in solution to an immobilized ligand on a biosensor tip ( Figure 4A) . The bio-layer interferometry instrument (Octet ®️ , Sartorius) is increasingly used in the research and biotechnology space. Protein binding produces an increase in optical thickness at the biosensor tip, altering the interference pattern of reflected light. This approach was used to build a proofof-concept assay for the rapid and semi-quantitative measurement of SARS-CoV-2 antibodies in saliva as well as plasma and is called bio-layer interferometry immunosorbent assay (BLI-ISA) 36 . Rapid real-time and quantitative antibody binding data can be obtained in <20 min in a washless and label-free fashion. Various sample types like saliva and serum can be tested on the same platform. The sensitivity and specificity were not assessed with clinical specimens.

The diagnostic applications of optical ring resonators have been commercialized by Genalyte. An optical ring resonator is a type of optical wave-guide based biosensor that traps light passing along an adjacent linear waveguide at its resonance frequency (Figure 4B ) 37 . The light makes multiple passes in the resonator allowing for larger effective interaction length (several centimeters) and improving the sensitivity of detection. This also significantly reduces the physical size of the sensor. Antibody binding to antigen coated on the ring resonator results in a detectable shift in its resonant frequency proportional to the mass of bound biomolecules. Primary and secondary antibodies are flowed over the biosensor and detected in real-time with a tunable laser. The assay protocols take less than 15 minutes and are run on an automated platform (Maverick TM , Genalyte) which supports a large number of immunoassays in addition to Sars-CoV2 serology 38 . A SARS-CoV-2 Multi-Antigen Serology Panel was developed by Genalyte on this platform and was issued an EUA by the FDA. The assay performance is similar to other fully automated platforms. It includes an array of antigens, including spike proteins from other human coronaviruses as specificity controls 21, 39 .

A virus neutralization test (VNT) is a serological test used to quantify the subset of antibodies that can prevent viral infection (Figure 5) . Such antibodies are called neutralizing antibodies, and in the case of SARS-CoV2 infection or vaccination, they interfere with binding to its cellular receptor ACE2 and inhibit viral entry. Conventional VNTs are used alongside an infectivity assay (e.g., plaque assay) to assess the ability of antibodies to inhibit viral replication or neutralize viral infection, which takes 2-4 days to complete. Surrogate VNTs that measure the ability of antibodies to block the interaction between the spike and ACE2 proteins have also been devised. Seropositivity against spike protein measured by commercial assays does correlate with neutralization activity 40, 41 . Interestingly, the specimens that did not correlate with neutralization activity also exhibited greater discordance among serological assays from different manufacturers, suggesting that a neutralization assay could also be used for improving the specificity of a conventional serological assay as other common human betacoronaviruses do not use ACE2 as a receptor. Due to the dimeric nature of secreted IgA in saliva, it has been shown to be 15x more potent at neutralization than its monomeric form in plasma 42 . This highlights the potential value of measuring isotype-specific J o u r n a l P r e -p r o o f neutralization, which is typically not done. Neutralization assays have primarily been used in research, epidemiological studies, or vaccine development; their role in routine clinical diagnostics is yet to be defined.

Conventional VNTs measure the infection of a susceptible cell line with a defined amount of a specific replication competent SARS-CoV2 strain in the presence of varying dilutions of the plasma. Multiple viral strains may be used to assess neutralization breadth. The resulting infectious virions are quantified using a plaque assay that can take an additional 2-4 days. Alternatively, cytopathic effects (CPE) are observed and the estimated dilution at which 50% of the wells show a CPE, is reported as tissue-culture infectious dose (TCID50). The neutralizing titer is reported as the dilution required to produce a 50% reduction in infectious virions (PRNT50). While this is the gold standard approach, it is a labor-intensive protocol that takes several days and needs to be performed in specialized biosafety facilities, as SARS-CoV2 culture requires a higher biosafety level (BSL3). Recently, a high throughput label-free optical approach called laser force cytology that examines cellular deformability using optical tweezers in a microfluidic channel (Lumacyte Radiance TM ) has been adapted to count virally infected cells in order to automate the readout of VNTs 43 . To overcome the BSL3 requirements, pseudotyped retroviruses or replication-defective VSV particles have been engineered that utilize the SARS-CoV2 spike for cell entry. The pseudotyped viruses can be used in neutralization assays in a conventional BSL2 laboratory and show good agreement with assays using replication-competent SARS-CoV2 44 . Furthermore, pseudotyped viruses have also been engineered to express a fluorescent or luciferase reporter for ease of measurement and scalability in a clinical setting. The IMMUNOCOV TM assay that uses pseudotyped VSV-G engineered with a luciferase reporter is commercially available; sufficient virus reagent has been banked to test 5 million clinical samples 45 . VNTs based on surrogate engineered viruses can be readily adapted to study neutralization of variants by incorporating spike mutations from variants of interest.

Surrogate VNTs that assess the ability of antiviral antibodies to inhibit the interaction between the viral receptor (ACE2) and the spike protein have been devised (Figure 5C ). sVNT assays may miss neutralizing antibodies that interfere with downstream steps of cell entry following ACE2 receptor binding involving membrane fusion and cell entry. Thus, the full spectrum of neutralizing capacity is most reliably measured using neutralization assays that rely upon a "live" virus. ELISA-format surrogate VNTs require the lowest biosafety level and yield a result within hours but may miss samples with lower neutralizing capacity. Once such assay cPass (Genscript) has received an FDA EUA 22 . Updated ELISA assays that assess the neutralization of emerging variants are under development (Axim Biotechnologies). Surrogate VNTs can also be implemented using other rapid approaches including lateral flow assays (e.g., NeuCOVIX, Axim Biosciences).

We have highlighted selected alternative approaches to the serological diagnosis of SARS-CoV2. A broad variety of biosensors harnessing nanoscale phenomena, nanopore physics, oligonucleotide chemistry or next-generation sequencing have been proposed in the literature, some of which have been applied to SARS-CoV2 serology [46] [47] [48] . Nanomaterial phenomena have also been exploited to enhance the performance of lateral flow assays 49 . While such approaches can be applied towards improving the characterization of the antibody response, it is important to note that a serological assay does not provide complete information about humoral immunity. For instance, persistent antigen-specific memory B cells are not directly assessed by serological tests. Furthermore, the cellular immune response consisting of CD4 + and CD8 + T cells is integral to the immune response, and a comprehensive assessment may be required to better assess infection risk. The T-Detect COVID test (Adaptive Biotechnologies) based on the analysis of the T cell repertoire by next-generation sequencing is a step in this direction 50 . Although the bulk of the serological diagnostics have focused on blood specimens, it has been shown that the antibodies to SARS-CoV2 found in the saliva do correlate well with levels in the blood 51 . Serological monitoring of saliva may offer a non-invasive alternative to monitor seropositivity at a population scale 52 . Rapid advances have been made in the field of SARS-CoV2 serology, but concomitant measurement of cellular immunity will be critical in obtaining a comprehensive picture of immunity in the context of natural infection, vaccine-induced J o u r n a l P r e -p r o o f protection or population surveys of immunity. Use of novel or alternate technologies will be required to develop assays for clinically scalable and multiplexed assessment of immune function in COVID-19.

R.P and S.K are affiliated with Vijna Labs, India, a technology incubator focused on artificial intelligence, healthcare and diagnostics.

A heterogeneous immunoassay for antigen-specific antibodies relies upon immobilized antigen as a capture probe and often uses enzyme labels for signal amplification; it requires step-wise reagent addition with multiple washes. A homogeneous immunoassay is washless and generates a signal only in the presence of the target analyte. An assay devised using antigens fused to split-luciferase domains is shown as an example.

J o u r n a l P r e -p r o o f

A) The binding of antibodies to antigens coated on the surface of a working electrode impedes electron transport, and results in a quantitative change in impedance. This can be measured using various electrochemical impedance spectroscopy techniques. For instance, as illustrated using Nyquist plots, the frequency dependence of the impedance is influenced by a change in the surface properties of the electrode when it is bound by an antibody. B) Two approaches to functionalize a working electrode are shown. A graphene electrode can be functionalized with non-covalently stacked pyrene-labeled antigens; such a graphene layer can also be coated on gold electrodes. Electrodes used in the Amperial TM platform are functionalized with native antigens entrapped in a conducting polypyrrole hydrogel using in situ electropolymerization. C) A typical electrochemical electrode design comprising a working electrode, a counter electrode and a reference electrode.

J o u r n a l P r e -p r o o f Figure 3 : Bioluminescence assays A) Luciferase is an enzyme that catalyzes a bioluminescent reaction by oxidizing a luciferin substrate. Luciferase systems that are well-characterized are widely used in bioassays (Renilla luciferase and coelenterazine substrate are shown). B) A luciferase immunoprecipitation assay (LIPS). Crude lysate bearing recombinant antigen-luciferase fusion protein is incubated with serum and antibodies are pulled down with anti-Ig beads. Antigen-specific antibodies are measured using a luciferase signal. C) Split nanoluciferase fragments (large BiT (LgBit) and small BiT (SmBiT)) are fused to RBD. An active nanoluciferase is assembled upon bivalent antibody binding, which can be sensitively detected with a luciferase substrate in a homogeneous assay format. D) A de novo designed lucCage:lucKey protein biosensor. The lucCage protein is built with a cage domain and a latch domain, which contains a target-binding motif and a split luciferase fragment (SmBit). The lucKey contains a key peptide that binds the lucCage cage domain and the complementary split luciferase fragment (LgBit). Binding of the analyte to the lucCage latch stabilizes the open conformation of lucCage, interaction with the lucKey and assembly of an intact luciferase. The thermodynamics of the system are designed such that the intact luciferase is reconstituted only in the presence of the target analyte.

J o u r n a l P r e -p r o o f Figure 4 : Label-free optics A) Biolayer interferometry relies upon measuring the interference pattern of white light reflected from an internal reference layer and the biosensor tip coated with antigen. Binding of antibodies to the test surface can shift the optical path by a few nanometers and this is detected by analyzing the interference pattern. This approach allows real-time monitoring of protein binding and dissociation and is compatible with crude samples as the surrounding medium does not influence these measurements. B) An optical ring biosensor is an optical cavity that is functionalized with antigen and coupled to a tunable laser. A dip in the signal intensity of a tunable laser is used to determine the resonant wavelength of the optical ring. Binding of an analyte results in a shift in the resonant wavelength, which can be monitored in real time.

A) A neutralization assay is typically set up using cells cultured in a multi-well plate. The ability of serial dilutions of test serum to interfere with virus-induced cytopathic effects (CPE) using a predetermined dose of infectious virus is measured. B) Neutralization assay detects the presence of antibodies which can prevent the infection or cell entry of infectious SARS-CoV2. Alternatively, reporter viruses pseudotyped with the SARS-CoV2 spike protein and rely on spike protein for cell entry can be used at a lower biosafety level. C) Surrogate neutralization assay relies upon measuring the ability of the specimen to interfere with the binding of surface-bound ACE2 receptor with spike protein or RBD in a competitive ELISA.

EUAs: Serology and Other Adaptive Immune Response Tests for SARS-CoV-2

Interpreting Diagnostic Tests for SARS-CoV-2

Longitudinal characterization of the IgM and IgG humoral response in symptomatic COVID-19 patients using the Abbott Architect

Ultrasensitive high-resolution profiling of early seroconversion in patients with COVID-19

Comparison of SARS-CoV-2 IgM and IgG seroconversion profiles among hospitalized patients in two US cities

IgA dominates the early neutralizing antibody response to SARS-CoV-2 n

Longitudinal follow-up of IgG anti-nucleocapsid antibodies in SARS-CoV-2 infected patients up to eight months after infection

The duration, dynamics and determinants of SARS-CoV-2 antibody responses in individual healthcare workers

Dynamics of neutralizing antibody titers in the months after SARS-CoV-2 infection

Defining the features and duration of antibody responses to SARS-CoV-2 infection associated with disease severity and outcome

SARS-CoV-2 infection induces long-lived bone marrow plasma cells in humans

SARS-CoV-2 serology: Test, test, test, but interpret with caution

Rethinking Covid-19 Test Sensitivity -A Strategy for Containment

SARS-CoV-2 variants of concern partially escape humoral but not T-cell responses in COVID-19 convalescent donors and vaccinees

SARS-CoV-2 variants, spike mutations and immune escape

Evaluation of serological lateral flow assays for severe acute respiratory syndrome coronavirus-2

19-emergency-use-authorizations-medical-devices/euaauthorized-serology-test-performance

COVID-19 Antibody Tests and Their Limitations

On-field evaluation of a ultra-rapid fluorescence immunoassay as a frontline test for SARS-CoV-2 diagnostic

Target specific serologic analysis of COVID-19 convalescent plasma

Validation of a commercially available indirect assay for SARS-CoV-2 neutralising antibodies using a pseudotyped virus assay

Agreement between commercially available ELISA and inhouse Luminex SARS-CoV-2 antibody immunoassays

Evaluation of the Quotient® MosaiQ TM COVID-19 antibody microarray for the detection of IgG and IgM antibodies to SARS-CoV-2 virus in humans

Electrochemical sensors and biosensors based on nanomaterials and nanostructures

Development and validation of a highly sensitive and specific electrochemical assay to quantify anti-SARS-CoV-2 IgG antibodies to facilitate pandemic surveillance and monitoring of vaccine response

SARS-CoV-2 RapidPlex: A Graphene-Based Multiplexed Telemedicine Platform for Rapid and Low-Cost COVID-19 Diagnosis and Monitoring

Rapid detection of SARS-CoV-2 antibodies using electrochemical impedance-based detector

Sensing of COVID-19 Antibodies in Seconds via Aerosol Jet Nanoprinted Reduced-Graphene-Oxide-Coated 3D Electrodes

Paper-based electrochemical biosensor for diagnosing COVID-19: Detection of SARS-CoV-2 antibodies and antigen

Sensitivity in Detection of Antibodies to Nucleocapsid and Spike Proteins of Severe Acute Respiratory Syndrome Coronavirus 2 in Patients With Coronavirus Disease

Antibody profiling by Luciferase Immunoprecipitation Systems (LIPS)

Engineering luminescent biosensors for point-of-care SARS-CoV-2 antibody detection

The coming of age of de novo protein design

De novo design of modular and tunable protein biosensors

Rapid and sensitive detection of SARS-CoV-2 antibodies by biolayer interferometry

Optical ring resonators for biochemical and chemical sensing

Detection in whole blood of autoantibodies for the diagnosis of connective tissue diseases in near patient testing condition

Evaluation of the Genalyte Maverick SARS-CoV-2 Multi-Antigen Serology Panel

Commercial Serology Assays Predict Neutralization Activity against SARS-CoV-2

Serological Assays Estimate Highly Variable SARS-CoV-2 Neutralizing Antibody Activity in Recovered COVID-19 Patients

Enhanced SARS-CoV-2 neutralization by dimeric IgA

Viral Infectivity Quantification and Neutralization Assays Using Laser Force Cytology

The role of pseudotype neutralization assays in understanding SARS CoV-2

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

Multiplex quantitative detection of SARS-CoV-2 specific IgG and IgM antibodies based on DNA-assisted nanopore sensing

Opportunities and Challenges for Biosensors and Nanoscale Analytical Tools for Pandemics: COVID-19

Viral immunology. Comprehensive serological profiling of human populations using a synthetic human virome

Strategies for developing sensitive and specific nanoparticle-based lateral flow assays as point-of-care diagnostic device

Clinical Validation of a Novel T-cell Receptor Sequencing Assay for Identification of Recent or Prior SARS-CoV-2 Infection

Persistence of serum and saliva antibody responses to SARS-CoV-2 spike antigens in COVID-19 patients

COVID-19 Serology at Population Scale: SARS-CoV-2-Specific Antibody Responses in Saliva