key: cord-0706689-n7gp24yq authors: Vogl, Thomas; Leviatan, Sigal; Segal, Eran title: SARS-CoV-2 antibody testing for estimating COVID-19 prevalence in the population date: 2021-01-14 journal: Cell Rep Med DOI: 10.1016/j.xcrm.2021.100191 sha: dc8146e7df3a4b47d60bc5ec18a5ea19fbed2bfa doc_id: 706689 cord_uid: n7gp24yq Reliable antibody testing against SARS-CoV-2 has the potential to uncover the population wide spread of COVID-19, which is critical for making informed healthcare and economic decisions. Here, we review different types of antibody tests available for SARS-CoV-2 and their application for population scale testing. Biases due to varying test accuracy, results of ongoing large-scale serological studies, and the use of antibody testing for monitoring the development of herd immunity are summarized. While current SARS-CoV-2 antibody testing efforts have generated valuable insights, the accuracy of serological tests and the selection criteria of the tested cohorts need to be carefully evaluated. have been accurately accounted for. Knowledge of the IFR is critical to assess the severity of 19 and to gain insights into its transmission. 72 Antibody responses against SARS-CoV-2 as basis for serological testing 73 The human adaptive immune system typically mounts a distinct response against SARS-CoV-2 74 including the production of specific IgM, IgG, and IgA antibodies 16 . In general, detectable amounts 75 of IgM appear after ca. 5 days after infection and level off at ca. 10 days, whereas IgG production is 76 at first delayed but surpasses IgM concentrations after ca. 10 days. This pattern is reflecting the 77 canonical role of IgM as first line of antibody responses with low affinity but high avidity 78 transitioning to the production of high affinity IgG 16 . However, beyond this general trend, timings of 79 seroconversion can vary substantially between SARS-CoV-2 infected individuals leading to possible 80 biases in determining the accuracy of diagnostic serology tests during the first two to three weeks 81 after infection. In some studies also simultaneous (rather than consecutive detection) of IgM and IgG 82 have been reported 20, 21 . Depending on the severity of infection, also the magnitude of antibody 83 responses can vary 9 , which can ultimately also impact the results of serology studies. 84 Serological tests for SARS-CoV-2 frequently test for both IgM and IgG antibody isotypes to cover 85 early (IgM) as well progressing (IgG) immune responses. Given that SARS-CoV-2 infects the 86 nasopharynx and lungs, mucosal IgA likely also contributes to containing infections. Yet, most 87 research has focused on blood IgG/IgM, possibly also owing to the high infectivity of mucosal 88 saliva/sputum samples of COVID-19 patients (blood on the other hand appears to be even in acute 89 patients about potential detrimental effects of antibody-dependent enhancement have been raised, which 94 warrant careful consideration in vaccine development and applications 26, 27 . 95 Regarding antigens, serological tests for SARS-CoV-2 typically detect antibodies against single 96 antigens such as the spike glycoprotein or the nucleocapsid protein yielding robust results 28 . The 97 genome of SARS-CoV-2 encodes ca. 30 proteins in total 29 , four of which encode structural proteins 98 forming the virion with the spike (S), membrane (M), envelope (E) proteins representing the viral 99 envelope and the nucleocapsid protein binding to the RNA genome. Testing for multiple SARS-CoV-2 100 antigens in parallel may improve diagnostic power 25 , although proteins other than spike tend to 101 show higher conservation between different coronaviruses (CoVs, see below) potentially providing 102 less discriminatory power. For example the nucleocapsid protein is more conserved between SARS-103 CoV-1 and SARS-CoV-2 than is the spike glycoprotein 29 . Non-structural proteins (NSPs) can also show 104 substantial conservation 30 and yielded little discriminatory power in a high resolution SARS-CoV-2 105 immunoassay 25 . Current antibody tests do not allow the discrimination between SARS-CoV2 strains 106 having accumulated different mutations and only genome sequencing can resolve phylogenetic 107 relationships (Tab. 1). 108 In addition to SARS-CoV-2, six more CoVs are known to infect humans and are potential candidates 109 to elicit cross-reactive antibodies, that could interfere with serological tests. RBD of SARS-CoV-1 or MERS-CoV 32 , but conserved regions of the SARS-CoV-2 RBD can also be 117 recognized by antibodies developed in the immune response against SARS-CoV-1 33 indicating some 118 cross-reactivity of RBD-specific antibodies between SARS-CoV-1 and SARS-CoV-2 34-36 . This cross-119 reactivity is impacted by different degrees of conservation of sections within the RBD of SARS-CoV-1 120 and SARS-CoV-2, related to epitopes overlapping with the ACE2 binding site. These finding suggest 121 that investigating antibody responses against protein segments at higher resolution can provide an 122 additional layer of information. Considering the much lower prevalence of SARS-CoV-1 (ca. 8,000 123 reported cases, no new cases after 2004) and MERS-CoV (less than 3,000 cases since 2012) 31 124 opposed to the spread of SARS-CoV-2 (>80 million cases diagnosed by RT-qPCR in December 2020), 125 cross-reactivities against SARS-CoV-1 and MERS-CoV are generally not expected to bias population 126 scale SARS-CoV-2 testing. 127 In addition to the highly pathogenic SARS- volume can impair the test's precision and the low total sample volume restricts the detection limit. 148 Furthermore, enzymatic signal amplification is not readily achievable in LFAs (as the enzymes and 149 substrate would inherently mix and react within the polymeric strip once substrate is added) 42 . 150 ELISAs on the other hand require laboratory equipment, trained personnel, and a longer workflow, 151 but can provide quantitative results and a higher accuracy than LFAs as they allow for enzymatic 152 signal amplification. Antibodies in the sample typically bind to antigens of interest immobilized on a 153 surface and the amount of bound antibodies is detected by addition of an enzyme linked detection 154 antibody specific for the isotype of interest (e.g. IgG, IgM, or IgA). Extensive washing between these 155 detection steps reduces background signals, making ELISAs highly accurate 43 . CLIAs are similar to 156 ELISAs, but rather rely on coated microparticles than immobilized antigens, yielding also high 157 accuracy with excellent amenability for automation. 158 Neutralization assays involve a more complex workflow detecting the inhibition of viral infection of 159 cultured target cells (Tab. 1). In detail, neutralizing antibodies block any step (typically entry) before 160 the first virally encoded synthetic event, whereas impeding the spread of infection in a culture by 161 blocking release would not represent neutralization 44 . Thereby neutralization assays represent the 162 only method capable of assessing patients' generation of neutralizing antibodies (NAbs), a key 163 J o u r n a l P r e -p r o o f 5/22 requirement for protection against reinfection (although also non-antibody-mediated cellular 164 immunity and non-neutralizing antibody-mediated immunity play distinct roles in the anti SARS-CoV-165 2 immune response 45 ). In typical neutralization assays patients' serum samples are serially diluted 166 and mixed with artificially produced SARS-CoV-2. These dilutions are then used to infect cell 167 cultures. If antibodies in the blood sample possess neutralizing capacity, lysis of the cultured cells is 168 reduced, which can be quantified as the number of plaques resulting from different dilutions. 169 Downsides of this approach involve a rather low throughput, the requirement for in part high safety 170 precautions (as the experiments require working with infectious SARS-CoV-2), and a rather long 171 incubation period until results are available (it takes several days for lysis of cells to become visible 172 as plaques) 43 . Several variations and alternatives exist for this plaque-reduction neutralisation assay 173 (PRNTs) recently reviewed by Khoury et al. 46 . These include replacing live SARS-CoV-2 with 174 replication-defective pseudoviruses, in which the SARS-CoV-2 spike protein is incorporated into the 175 surface of vesicular stomatitis virus or lentiviruses (single-cycle virus neutralization assays). 176 Pseudoviruses can also facilitate detection of infection, as fluorescent reporters genes can be 177 included, that will be expressed once infecting host cells. Multi-cycle virus neutralization assays rely 178 on replicating pseudoviruses (or native SARS-CoV-2) and allow to follow viral replication over longer 179 periods. The use of pseudoviruses mitigates some downsides of PRNTs both relating to biological 180 safety precautions and duration of the workflow, although spike folding and presentation may vary 181 from its native state on SARS-CoV-2. Also replication dynamics of pseudoviruses may deviate from 182 native SARS-CoV-2 46 . 183 Molecular diagnostics for SARS-CoV-2 can also rely on the use of antibodies specific to viral antigens. 184 Rather than detecting the presence of immune responses in the patient's blood, these tests allow to 185 detect the presence of viruses in specimens (yielding information similar to RT-qPCR testing). For 186 such rapid antigen testing approaches, the applied detection antibodies need to be carefully 187 selected, as cross-reactivity, especially to seasonal common cold CoVs, could bias results. To this 188 end, a panel of monoclonal antibodies highly specific for SARS-CoV-2 could improve accuracy. 189 Researchers of the Johns Hopkins University have compiled a valuable overview on commercially 190 available SARS-CoV-2 antibody tests including the two most important criteria for examining 191 serological tests specificity and sensitivity (as self-reported by the manufacturers) 40 An antibody test with 95% sensitivity and 95% specificity may appear intuitively highly reliable for 231 executing such testing, yet its real-world applicability depends on the true prevalence of the 232 infection in the population 54 . Generally, high specificity of a test is critical at a low prevalence in the 233 population (ca. <10%), otherwise the false positives may outweigh the true positives. High sensitivity 234 is critical at high prevalences in the population, otherwise the false negatives may outweigh the true 235 negatives (Fig. 1a) . Positive/negative predictive value (PPV, NPV) are metrics representing test 236 performance at a certain prevalence in the population (reviewed in detail by 43 ). 237 For example, testing a population with 1% prevalence of a disease, with a test of 100% sensitivity 238 and 99% specificity would report almost 2% positives (Fig. 1b) . Under the same conditions, a test 239 with 95% specificity would report almost 6% positives, an overwhelming overestimation of disease 240 prevalence. At a high prevalence of 50% in the population, the reported prevalence by two tests of 241 99% and 95% specificity and 100% sensitivity would report 50.5% and 52.5% positives, which 242 represents only a slight overestimation. 243 In contrast, at a prevalence of 90% in the population (for antibody responses possibly only 244 achievable by vaccination), a test with 100% specificity and 90% sensitivity would report an 81% 245 prevalence which may lead to an underestimation of the effect of a vaccination. The same test (with 246 100% specificity and 90% sensitivity) would however report only a minor difference at a true 247 prevalence of 5% (reporting a value of 4.5%). 248 Hence, the sensitivity of a test does not strongly impact results at a low prevalence in the population 249 (ca. <10%), whereas specificity does not dramatically affect results at high prevalence (Fig. 1) . This 250 notion has direct implications for selecting suitable testing methods such as LFAs or ELISAs 55 , 251 depending on the expected prevalence in the population. Beyond technical aspects of the accuracy 252 of antibody testing, also biological factors affect the reported population scale prevalence. The 253 above considerations rely on the assumption that every infected individual mounts detectable 254 antibody response after infection (seroconversion). A complete lack of seroconversion or the rapid 255 loss of detectable antibody responses would lead to an underestimation of the total prevalence. In a 256 recent study 56 , seroconversion was reported in nearly all (621/624) individuals with confirmed SARS-257 CoV-2 infection. Seroconversion in individuals with suspected SARS-CoV-2 infection showed a lower 258 J o u r n a l P r e -p r o o f 7/22 rate (37% positive) and could be both due to lack of actual infection or seroconversion. A small scale 259 study 57 has reported lack of seroconversion in some individuals likely exposed to SARS-CoV-2, 260 whereas T-cell responses (non-antibody-mediated cellular immunity) were readily detected 261 However, it is unclear how frequent such lack of seroconversion occurs and to which extend 262 population scale profiling may be affected. 263 In addition to the lack of seroconversion after SARS-CoV-2 infection, also temporal dynamics of 264 antibody responses and cross-reactivity ( population scale testing at current prevalences (although some ELISAs/CLIAs have also lacked 381 sufficient sensitivity/specificity). Point of care rapid LFAs could still benefit from improvements 382 (depending on the application) 49,51,52 . Some LFAs have displayed excellent specificity, although these 383 may not have been widely applicable, given the large availability of LFAs with varying accuracy, 384 especially early during the pandemic. Hence, laboratory ELISAs/CLIAs appear currently as the 385 method of choice for monitoring the population wide spread in a post-lockdown world. While this 386 approach is viable in high-income countries with the required molecular diagnostic infrastructure in 387 place, carrying out such testing efforts in low-and middle-income country may be challenging. 388 Therefore, improving low-cost, rapid diagnostic kits will be critical and could be achieved by testing 389 for multiple antigens and isotypes (including IgA) in a single assay, or combining different low-cost 390 tests with independent systematic biases to reduce the overall error rates of LFAs. LFAs provide also 391 the key advantage of enabling self-testing, which is highly relevant in any country imposing 392 restrictions on individual movement to contain case numbers. In any setting, the applied serological 393 tests should be thoroughly independently validated for the situation in which it will be employed. 394 Even leaving possible testing inaccuracies aside, estimated thresholds for herd immunity were not 395 achieved in any region of the world, thus measures to prevent the spread of COVID-19 should 396 J o u r n a l P r e -p r o o f 10/22 continue. Given the relatively short time of the current outbreak, longitudinal studies will need to 397 assess long-term antibody responses of recovered patients. Once vaccines have been broadly 398 deployed, analogous studies will need to assess the duration of vaccines' protection against SARS-399 CoV-2 reinfection. The first large scale studies on thousands of random individuals have deepened 400 our understanding of the true prevalence of COVID-19 and these efforts will narrow down the IFR as 401 a key metric for the severity of COVID-19. A large denominator identified by serology testing (i.e. 402 many people infected and having developed antibodies without diagnosis during the acute phase), 403 could reveal that SARS-CoV-2 is less severe than estimated from RT-qPCR testing. These findings will 404 increase our understanding of the transmission dynamics of SARS-CoV-2, will help to improve 405 modelling efforts, and can thereby guide preparations for possible future outbreaks. 406 Table 1 732 Tab. 1: Testing approaches for SARS-CoV-2 and types of antibody assays. The airborne lifetime of small speech droplets and 420 their potential importance in SARS-CoV-2 transmission Estimating the asymptomatic proportion of 423 coronavirus disease 2019 (COVID-19) cases on board the Diamond Princess cruise ship Asymptomatic and Presymptomatic SARS-CoV-2 Infections in Residents 427 of a Long-Term Care Skilled Nursing Facility Hospitalization Rates and Characteristics of Patients 435 Hospitalized with Laboratory-Confirmed Coronavirus Disease 2019 -COVID-NET, 14 States Clinical characteristics of asymptomatic 439 and symptomatic patients with mild COVID-19 Extrapulmonary 442 manifestations of COVID-19 Prevalence of Asymptomatic SARS-CoV-2 Infection Clinical and immunological assessment 448 of asymptomatic SARS-CoV-2 infections The Incubation Period of 450 COVID-19) From Publicly Reported Confirmed Cases: Estimation 451 and Application Asymptomatic Transmission, the Achilles' Heel of Current 454 Strategies to Control Covid-19 Substantial undocumented infection 456 facilitates the rapid dissemination of novel coronavirus (SARS-CoV-2). Science (80-) The effect of large-scale anti-460 contagion policies on the COVID-19 pandemic COVID-19 diagnostics in context Temporal profiles of viral load 466 in posterior oropharyngeal saliva samples and serum antibody responses during infection by 467 SARS-CoV-2: an observational cohort study Immunology of COVID-19: 470 current state of the science Antibody tests suggest that coronavirus infections vastly exceed official counts How deadly is the coronavirus? Scientists are close to an answer The many estimates of the COVID-479 19 case fatality rate Antibody responses to SARS-CoV-2 in 482 patients with COVID-19 Longitudinal observation 485 and decline of neutralizing antibody responses in the three months following SARS-CoV-2 486 infection in humans Virological 489 assessment of hospitalized patients with COVID-2019 Chronological evolution of IgM, IgA, IgG and 492 neutralisation antibodies after infection with SARS-associated coronavirus Distinct features of SARS-CoV-2-specific IgA 496 response in COVID-19 patients Viral epitope profiling of COVID-19 498 patients reveals cross-reactivity and correlates of severity SARS-CoV-2 Associated with Cellular Immune Response without Seroconversion Emerg Infect Dis Population-596 based surveys of antibodies against SARS-CoV-2 in Southern Brazil Seroprevalence of 598 anti-SARS-CoV-2 IgG antibodies in Geneva, Switzerland (SEROCoV-POP): a population-based 599 study Humoral Immune Response to SARS-CoV-2 in Iceland Infection fatality rate of SARS-CoV-2 infection in a German community 605 with a super-spreading event, preprint Performance 613 Characteristics of the Abbott Architect SARS-CoV-2 617 dozens-of-times-higher-than-the-number-of-confirmed-618 cases?redirect=%2Fen%2Fweb%2Finfectious-diseases%2Fwhat-s-new%2Fcoronavirus-covid-619 19-latest-updates Estimates of the severity of 629 coronavirus disease 2019: a model-based analysis Using Early Data to Estimate the Actual 631 Infection Fatality Ratio from COVID-19 in France the infection and case fatality ratio for coronavirus disease (COVID-19) using age-adjusted 635 data from the outbreak on the Diamond Princess cruise ship Herd Immunity: Understanding COVID-19 Population biology of infectious diseases: Part I Population biology of infectious diseases: Part II The role of herd immunity in control of measles Projecting the transmission 646 dynamics of SARS-CoV-2 through the postpandemic period Herd immunity -estimating the level required 651 to halt the COVID-19 epidemics in affected countries Neutralizing antibody responses to SARS-CoV-2 in a COVID-19 recovered 654 patient cohort and their implications, preprint Detection of SARS-CoV-2-Specific Humoral 657 and Cellular Immunity in COVID-19 Convalescent Individuals SARS-CoV-2 660 infection protects against rechallenge in rhesus macaques Primary exposure to SARS-CoV-2 protects 664 against reinfection in rhesus macaques SARS-CoV-2 vaccines: status report. Cell Rapid COVID-19 vaccine development Duration of Antibody 672 Responses after Severe Acute Respiratory Syndrome. Emerg Infect Dis The time course of the immune response to 676 experimental coronavirus infection of man The behaviour of recent isolates of human respiratory coronavirus in vitro and in 681 volunteers: Evidence of heterogeneity among 229E-related strains Disappearance of Antibodies to SARS-684 Associated Coronavirus after Recovery Two-Year Prospective Study of the 687 Humoral Immune Response of Patients with Severe Acute Respiratory Syndrome Persistence of Antibodies 691 against Middle East Respiratory Syndrome Coronavirus. Emerg Infect Dis Lack of Peripheral Memory B Cell 695 Responses in Recovered Patients with Severe Acute Respiratory Syndrome: A Six-Year Follow-696 Up Study An efficient 699 method to make human monoclonal antibodies from memory B cells: potent neutralization 700 of SARS coronavirus Long-Term Persistence of IgG Antibodies in SARS-CoV Infected Healthcare 703 Workers, preprint Identifying SARS-CoV-2 related 705 coronaviruses in Malayan pangolins Real-time tracking of 707 self-reported symptoms to predict potential COVID-19 Genomic characterisation and epidemiology of 710 2019 novel coronavirus: implications for virus origins and receptor binding Seroprevalence of immunoglobulin M and G 715 antibodies against SARS-CoV-2 in China This article reviews different types of antibody tests available for SARS-CoV-2 and their application for population scale testing. Biases due to varying test accuracy, results of ongoing large-scale serological studies, and the use of antibody testing for monitoring the development of herd immunity are summarized.