key: cord-0732733-rg4wze40
authors: Montague, Brian T; Wipperman, Matthew F; Hooper, Andrea T; Hamon, Sara C; Crow, Rowena; Elemo, Femi; Hersh, Lisa; Langdon, Shaun; Hamilton, Jennifer D; O’Brien, Meagan P; Simões, Eric A F
title: Anti-SARS-CoV-2 IgA Identifies Asymptomatic Infection in First Responders
date: 2021-10-12
journal: J Infect Dis
DOI: 10.1093/infdis/jiab524
sha: 446bad5b3217560aebbdc408c57ec5676ca0544b
doc_id: 732733
cord_uid: rg4wze40

BACKGROUND: IgA is an important component of the early immune response to SARS-CoV-2. Prior serosurveys in high-risk groups employing IgG testing alone have provided discordant estimates. The potential added benefit of IgA in serosurveys has not been established. METHODS: Longitudinal serosurvey of first responders (police, emergency medical service providers, fire fighters, and other staff) employing three serologic tests: anti-spike IgA, anti-spike IgG, and anti-nucleocapsid IgG correlated with surveys assessing occupational and non-occupational risk, exposure to COVID-19 and illnesses consistent with COVID-19. RESULTS: Twelve percent of first responders in Colorado at baseline and 22% at follow-up were assessed as having SARS-CoV-2 infection. Five percent at baseline and 6% at follow-up were seropositive only for IgA. Among those IgA positive only at baseline, the majority 69% had a positive antibody at follow-up. 45% of those infected at baseline and 33% at follow-up were asymptomatic. At all time points, the estimated cumulative incidence in our study was higher than that in the general population. CONCLUSIONS: First responders are at high risk of infection with SARS-CoV-2. IgA testing identified a significant portion of cases missed by IgG testing and its use as part of serologic surveys may improve retrospective identification of asymptomatic infection.

A c c e p t e d M a n u s c r i p t 3

Infection with SARS-CoV-2 is associated with a spectrum of illness from asymptomatic infection to severe pneumonia with acute respiratory distress syndrome . First responders are a recognized high-risk group for SARS-CoV-2 infection. Serologic surveys of first responders have shown variable results with some showing rates similar to the general population and others showing higher rates (see Table 1 ). [1] [2] [3] [4] [5] [6] [7] Some of this variability may relate to differences in access to personal protective equipment (PPE) and infection control policies. [8] Because high-risk groups such as older individuals and individuals with significant comorbidities may be underrepresented among first responders, the proportion of infections which are asymptomatic in this group may be higher.

Sensitive serologic assays have the potential to provide a measure of cumulative incidence of infection including both symptomatic and asymptomatic cases. A wide array of serologic tests have received Emergency Use Authorization (EUA) by the United States Food and Drug Administration (FDA). [9] These EUAs are based on validation studies conducted with confirmed positive COVID-19 cases. [10] In a multisite comparison of multiple serologic tests, the sensitivities and specificities of IgG tests ranged from 80% to 89% and 97% to 100%, respectively. [11] Given the potential for false positive tests when applied to low-risk populations, serosurvey strategies for these populations requiring concordant positivity of two tests with different antigenic targets have been recommended. [12, 13] In high-risk populations, however, the risk for false positive results will be lower and a single positive test may be sufficient to confirm infection. infection. [14] [15] [16] [17] [18] Persons with asymptomatic or mild infection in prior studies may have lower levels of IgG and may show a more rapid decay in antibody titers. [19] [20] [21] Given that the upper respiratory tract is a primary point of infection for SARS-CoV-2, mucosal and systemic IgA plays an important role in host defense. Yu et al showed that seroconversion from IgM to IgA occurred 2 days after the onset of symptoms, and before the seroconversion from IgM and IgG. [22] IgA has also been shown to play an early role in virus neutralization. [23] In a longitudinal study, the median time to seroreversion for IgA was 70.5 days and 25% had seroreversion at 104 days or greater. [24] This finding aligns with other published longitudinal studies using assays for IgA targeting the S1 domain of Spike. [25, 26] To characterize the risk of SARS-CoV-2 infection among first responders and the relative importance of IgA versus IgG testing in serologic monitoring, we conducted a longitudinal serosurvey among first responders in Colorado employing three serologic tests: anti-S1 domain of spike protein IgA, anti-S1 domain of spike protein IgG, and anti-nucleocapsid IgG and we correlated the serologic testing with participant surveys identifying times of exposure to persons with COVID-19 and illness potentially consistent with COVID-19.

This study was a longitudinal serosurvey of first responders conducted in Colorado between May of 2020 and March of 2021 with serologic data correlated with longitudinal assessment of exposures, periods of illness and confirmed cases of A c c e p t e d M a n u s c r i p t 5

Outreach occurred by email to agency staff as well as through direct communication via agency leadership. REDCap questionnaires were used for all data collection. [27] Interested persons were provided a link to an online screening questionnaire which included a link to review an electronic copy of the consent form. Subjects were consented, paper consents were signed, and phlebotomy was performed at scheduled events at each participating agency. Individuals who had not completed the online screening could complete the screening process on site.

All participants completed an intake questionnaire assessing their occupational and home risk for infection and any history of COVID-19. Individuals were categorized by their job role into the groups Police, Firefighters, EMS (emergency medical personnel), or Other, with those in the Other category being individuals who worked in ancillary job roles for participating agencies. Participants were surveyed regarding their prior exposures to persons with COVID-19, prior confirmed diagnosis of COVID-19, access to PPE, patterns of PPE utilization as part of their job role, and proportion of time spent in direct contact with the public. A follow-up survey was sent to assess periods of illness prior to baseline testing for which the participant took leave from work as well as periods of illness when the participant were either not tested or tested negative for SARS-CoV-2 infection. Individuals with reported illness or COVID-19 exposures after the baseline questionnaires were invited to complete supplemental questionnaires.

Blood samples were collected, serum was isolated on site and samples were frozen for serologic testing at baseline and at follow-up after 2-3 months. All samples were tested at ICON Laboratories using the Abbott Architect Anti-SARS-CoV2 chemiluminescent microparticle immunoassay (MIA) for IgG (anti-nucleocapsid protein, anti-N), Euroimmun Anti-SARS-CoV-2 ELISA for IgG (anti-S1 domain of Table 2 ). The presence of antibodies without report of prior COVID-19 diagnosis or periods of illness was defined as asymptomatic infection. For those with no follow-up assessment, categorization was based on baseline data. Given the potential for waning of antibody response, if an individual had a higherlevel categorization at baseline than follow-up, the baseline categorization was retained.

The proportion of participants with antibody positivity was assessed at baseline and follow-up, stratified by job role and history of exposure, illness, and COVID-19 diagnosis. Cumulative incidence Data management, statistical analyses, and figure generation were performed using SAS® Version 9.4 (SAS Institute Inc., Cary, NC, USA). Chi-square testing was used to assess differences in the proportion serologically positive across work setting and by exposure and illness categories.

Assessment of statistically significant differences between incidence rates by job role was performed using Poisson regression (PROC Genmod). Kaplan Meier estimates were generated using PROC Lifetest.

IRB approval was obtained from the Western IRB, protocol # 20201662. Across all roles, participants were predominantly male, Caucasian, non-Hispanic, with median age by occupational role between 34 (EMS) and 47 (Other) (see Table 3 ). Greater than 90% of Firefighters and EMS reported access to both N95 respirators and eye protection and over 80% reported access to surgical masks as well. Reported access was slightly lower among Police with 78% reporting access to N95 respirators, 80% to eye protection, and 82% to surgical masks. Those reporting Other roles did not report access to PPE. Firefighters reported being unable to socially distance at work 40% of the time compared to 50% for EMS and 38% for Police. Greater than 70% of Among those with no reported periods of illness or exposure, antibody testing was completely negative in 90% of cases at baseline at 86% of cases at follow-up (see Table 4 ). Fifteen percent before adjustment for false positive IgA and 12% after adjustment were positive for at least one antibody in the absence of either reported illness or exposure. Thirteen percent were positive by IgA and 6% were positive for at least one IgG. Positivity of at least one antibody was identified at baseline in 43% (41% adjusted) of those with prior exposure and 41% (36% adjusted) of those with prior illness. IgA positivity at baseline was identified in 36% of those with prior exposure and 35% of those with prior illness. Among those with a history of COVID-19, 75% (73% adjusted) had at least one positive antibody at baseline with 43% positive by all three tests, 68% for at least one IgG, and 

This study identified a high cumulative incidence of SARS-CoV-2 infection among first responders in Colorado with rates as high 12% at baseline and 22% at follow-up as measured by seropositivity of one or more antibody tests (IgA, Anti-S IgG, Anti-N IgG). Of high risk first responders, 8% were seropositive for only anti-spike IgA as compared to 6% positive for anti-S or anti-N IgG. Among those with positive IgA only, the majority remained positive at follow-up. Individuals with positive IgA were more likely to report prior respiratory illness.

A multisite serosurvey from the period between March and May of 2020 reported population rates of only 6.9% in the most highly impacted areas such as New York. [31] The prior serosurvey of Colorado first responders required concordance of two positive IgG tests with an overall proportion positive of 4%. This proportion is similar to the proportion observed in our study positive by all 3 assays. It is also similar to the results of a life-insurance applicant survey from September of 2020 in which 4.1% of applicants from Colorado tested positive for anti-N IgG and the estimates of positivity based on modeling. [30, 32] . All the contemporaneous first responder surveys were based on testing exclusively for IgG, suggesting that these prior surveys may have significantly underestimated the rate of SARS-CoV-2 exposure and infection.

The specificity of the Euroimmun IgA ELISA among 30 samples from laboratory confirmed SARS-CoV-2 infected persons and 80 pre-pandemic samples submitted to the US FDA for Emergency Use Authorization was 91.2%. [33] In a subsequent validation study based on 100 samples from laboratory confirmed SARS-CoV-2 infected persons and 300 pre-pandemic samples, the specificity of the Euroimmun IgA ELISA was found to be higher at 93.7% for IgA compared to 91.7% found for IgG (anti-S). [29] These validation studies are conducted with known positive cases and collected at a fixed interval after onset of symptoms typically within 2-4 weeks.

A c c e p t e d M a n u s c r i p t 12 The performance of these tests when applied in individuals who are well as part of serosurveys has not been well described. In a longitudinal study of 1013 German healthcare workers, IgA positivity was identified in 6.8% of workers as compared IgG positivity which was found in only 2.1%. [34] In this study, IgA remained more common at 10-week follow-up seen in 10.4% or participants as compared to IgG in 1.6% with 6% of those IgA positive at baseline developing IgG positivity by the time of follow-up. This study reported a lower specificity of IgA testing at 73% but treated all positives as true positives. In our analysis using the more specific manufacturer recommended cutoffs for positive for the IgA assay, we equally found significant numbers of individuals positive for IgA in the absence of IgG positivity with a slightly higher percentage (12%)

developing IgG positivity at follow-up. It is notable on our sample that the estimated 35% false positive rate for the IgA test is similar to the proportion of individuals (31%) who were positive by IgA alone at baseline and negative on all tests at follow-up suggesting that this may be a potential indicator of false positive tests.

The significant portion of cases identified by isolated IgA positivity in this cohort is notable and highlights the potential importance of IgA in longitudinal serosurveys. In prior surveys, as much as 45% of individuals infected with SARS-CoV-2 have been asymptomatic. [35] Much of this data derives from investigation of confined cohorts, however in a large serosurvey from Iceland 43% of those positive by IgG had had no symptoms of illness. [36] In our sample of high-risk first responders, 45% of those infected at baseline and 33% at follow-up were asymptomatic with 13% of those infected at baseline and 5% at follow-up identified by IgA positivity alone.

Follow-up was only available for 783 participants, limiting the sample for evaluation of serostatus over time. Despite this, the associations between antibody response and exposure and illness were consistent between baseline and follow-up. Early in the pandemic access to SARS-CoV-2 testing was limited. It is possible that some of those with reported illness may have had infections A c c e p t e d M a n u s c r i p t 13 with other circulating respiratory viruses. There was less circulation of these viruses between the period of the 1st and 2nd sample and the consistency of the identified associations between baseline and follow-up suggests that confounding by other viral infections in these analyses is limited.

Based on serologic testing, first responders are at higher risk of infection with SARS-CoV-2 relative to the general population. In this serosurvey, anti-spike IgA identified a significant number of cases missed by anti-S and anti-N IgG testing. Testing for IgA as part of serologic surveys is therefore important and may improve retrospective identification of asymptomatic or mildly symptomatic cases. Additional information is needed to clarify the level of protection conferred to those with isolated IgA positivity following infection with SARS-CoV-2.

M a n u s c r i p t 18 Table 1 

Risk Factors of SARS-CoV-2 Antibodies in Arapahoe County First Responders-The COVID-19

Arapahoe SErosurveillance Study (CASES) Project

Prevalence of SARS-CoV-2 Antibodies in First Responders and Public Safety Personnel

SARS-CoV-2 Seroprevalence among Healthcare, First Response, and Public Safety Personnel

Prevalence of COVID-19 IgG Antibodies in a Cohort of Municipal First Responders

Prevalence of Coronavirus Antibody Among First Responders in Lubbock

COVID-19 Exposure Among First Responders in Arizona

SARS-CoV-2 Infection among Serially Tested Emergency Medical Services Workers

Occupational exposures and programmatic response to COVID-19 pandemic: an emergency medical services experience

Evaluation of Six Commercial Mid-to High-Volume Antibody and Six Point-of-Care Lateral Flow Assays for Detection of SARS-CoV-2 Antibodies

Multi-center nationwide comparison of seven serology assays reveals a SARS-CoV-2 non-responding seronegative subpopulation

Orthogonal SARS-CoV-2 Serological Assays Enable Surveillance of Low-Prevalence Communities and Reveal Durable Humoral Immunity

Estimating COVID-19 Antibody Seroprevalence

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

Longitudinal observation and decline of neutralizing antibody responses in the three months following SARS-CoV-2 infection in humans

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

Seroprevalence of anti-SARS-CoV-2 antibodies in COVID-19 patients and healthy volunteers up to 6 months post disease onset

Immunological memory to SARS-CoV-2 assessed for up to 8 months after infection

Rapid Decay of Anti-SARS-CoV-2 Antibodies in Persons with Mild Covid-19

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

Serological follow-up of SARS-CoV-2 asymptomatic subjects

Distinct features of SARS-CoV-2-specific IgA response in COVID-19 patients

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

Persistence and decay of human antibody responses to the receptor binding domain of SARS-CoV-2 spike protein in COVID-19 patients

Temporal Course of SARS-CoV-2 Antibody Positivity in Patients with COVID-19 following the First Clinical Presentation

IgA-Ab response to spike glycoprotein of SARS-CoV-2 in patients with COVID-19: A longitudinal study

Research electronic data capture (REDCap)--a metadata-driven methodology and workflow process for providing translational research informatics support

ELISA (IgG) Instruction for Use

Performance of nucleocapsid and spike-based SARS-CoV-2 serologic assays

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

Seroprevalence of SARS-CoV-2 Antibodies in the US Adult Asymptomatic Population as of

SARS-COV-2 ELISA (IgA)" from Euroimmun. accessdata.fda.gov: FDA

Prevalence and Course of IgA and IgG Antibodies against SARS-CoV-2 in Healthcare Workers during the First Wave of the COVID-19 Outbreak in Germany: Interim Results from an Ongoing Observational Cohort Study

Prevalence of Asymptomatic SARS-CoV-2 Infection

Spread of SARS-CoV-2 in the Icelandic Population

A c c e p t e d M a n u s c r i p t A c c e p t e d M a n u s c r i p t 28 Figure 2