key: cord-0869010-psrxh1b6
authors: Higgins, Victoria; Fabros, Anselmo; Wang, Xiao Yan; Bhandari, Maria; Daghfal, David J.; Kulasingam, Vathany
title: Anti-SARS-CoV-2 IgM improves clinical sensitivity early in disease course
date: 2021-01-19
journal: Clin Biochem
DOI: 10.1016/j.clinbiochem.2021.01.003
sha: 02af6d07526e1eb31a8fa7f4e04c240b319e2ff3
doc_id: 869010
cord_uid: psrxh1b6

INTRODUCTION: Coronavirus disease 2019 (COVID-19), caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), is diagnosed by molecular-based detection of SARS-CoV-2 RNA. Serologic testing detects antibodies specific to SARS-CoV-2 and IgM specifically may serve as an adjunct test to PCR early in disease. We evaluated the Abbott anti-SARS-CoV-2 IgM and IgG assays along with DiaSorin anti-SARS-CoV-2 IgG and Roche anti-SARS-CoV-2 Total. METHODS: Specimens from 175 PCR-positive patients and 107 control specimens were analyzed using Abbott IgM and IgG, DiaSorin IgG, and Roche Total (IgA, IgG, IgM) assays. Sensitivity, specificity, cross-reactivity, concordance between assays, trends over time, positive predictive value (PPV), and negative predictive value (NPV) were determined. RESULTS: Abbott IgM sensitivity was 63.6% at 0 days post-PCR positivity, 76.5% at 1-5d, 76.3% at 6-14d, 85.2% at 15-30d, and 63.6% at >30d. All assays exhibited highest sensitivity 15-30d post-PCR positivity (83.3-85.2%). Combining Abbott IgM and IgG improved sensitivity by 22.7% compared to IgG alone when tested 0d post-PCR positivity. All assays had a specificity of 100% and only Abbott IgG exhibited cross-reactivity (anti-dsDNA). Cohen’s kappa varied between 0.86-0.93. Time to seroconversion from PCR positivity was lowest for Abbott IgM and highest for Abbott IgG. NPV was highest for Abbott IgM <14 days post-PCR positivity and Abbott IgG ≥14 days. CONCLUSION: The Abbott IgM assay exhibited the earliest response and greatest signal in most patients evaluated for serial sampling and had the highest NPV <14 days post-PCR positivity, suggesting its potential utility as an adjunct test to PCR early in disease course.

The Abbott IgM assay exhibited the earliest response and greatest signal in most patients evaluated for serial sampling and had the highest NPV <14 days post-PCR positivity, suggesting its potential utility as an adjunct test to PCR early in disease course.

Coronavirus disease 2019 , caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), emerged from Wuhan, China in late 2019 (1) . COVID-19 was first declared a Public Health Emergency of International Concern in January 2020 by the World Health Organization (WHO) and has infected over 83 million people globally, causing over 1.8 million deaths as of 3 January 2021 (2) . Clinical manifestations of COVID-19 illness vary in severity between patients from asymptomatic to severe pneumonia, acute respiratory distress syndrome, sepsis, and/or multisystem organ failure (3) . COVID-19 is diagnosed by molecularbased detection of SARS-CoV-2 RNA, most commonly by reverse transcription-polymerase chain reaction (RT-PCR) in nasopharyngeal and/or oropharyngeal specimens (4) . Viral RNA can be detected in these specimens as early as the first day of symptom onset, peaks within the first week, and can remain positive beyond three weeks in severe cases (5, 6) . However, PCR positivity only reflects viral RNA detection, not necessarily the presence of viable virus (7) , and its predictive value varies with time from exposure and symptom onset (8) . For example, one study reported the probability of a false negative result to be 100% on day 1 of infection, 67% on day 4, 38% on day 5 (symptom onset), 20% on day 8, and 66% on day 21 (8) .

Serologic testing detects antibodies (e.g. IgG, IgM) specific to SARS-CoV-2 in blood, serum, or plasma. While serologic testing is not useful on its own for COVID-19 diagnosis (9), it may serve as an adjunct to molecular-based testing for COVID-19 diagnosis if used ≥15 days after symptom onset in cases with suggestive clinical presentation, but where RT-PCR results are negative or not available (9) . As IgM is a marker of acute infection, it may be a useful tool to combine with PCR to improve sensitivity and specificity early in the disease course (i.e. <14 days after symptom onset) (10) (11) (12) . Antibody response has been reported to correlate with disease severity, with more severe cases exhibiting immediate seroconversion (13) . Furthermore, antibody titers were found to be higher in severe compared to non-severe cases two weeks postsymptom onset (12) . Serologic testing may also have clinical utility for surveying asymptomatic infection in close contacts and population-level assessment of the prevalence of past SARS-CoV-2 infection (12, 14, 15) . While data are still limited, there is mounting evidence that antibodies detected by commercial serologic assays correlate with neutralization capacity (16) and confer some resistance to re-infection (17, 18) . Thus, serologic testing may also have clinical utility for international travel authorization, assessing reinfection risk in workplaces, and facilitating economic activity resumption.

In order to demonstrate an adequate positive predictive value, it is important for serologic assays to demonstrate high sensitivity and specificity, particularly when seroprevalence is low (19) . It has been suggested that laboratories should implement SARS-CoV-2 serologic tests that have manufacturer-claimed sensitivity ≥95% and specificity ≥99.5% based on specimens obtained ≥14 days after symptom onset or PCR positivity (20) . We evaluated the sensitivity, specificity, cross-reactivity, concordance, trends over time, positive predictive value, and negative predictive value for four serologic assays: Abbott anti-SARS-CoV-2 IgM serologic assay, Abbott anti-SARS-CoV-2 IgG, DiaSorin anti-SARS-CoV-2 IgG, and Roche anti-SARS-CoV-2 Total assays.

This work was exempt from Quality Improvement (QI) review and Research Ethics Board (REB) approval at the University Health Network (UHN; Toronto, Canada). Presence or absence of SARS-CoV-2 infections was determined by SARS-CoV-2 viral RNA detection in nasopharyngeal swabs tested at UHN microbiology lab on assays validated for clinical use (Seegene Allplex 2019-nCoV assay which has been approved by Health Canada for Emergency Use Authorization and verified by UHN microbiology lab). Deidentified residual patient serum and plasma samples were collected from UHN and analyzed using four anti-SARS-CoV-2 serologic assays at UHN, including SARS-CoV-2 IgG and SARS-CoV-2 IgM on the Abbott ARCHITECT® i (Abbott Diagnostics), SARS-CoV-2 S1/S2 IgG on the LIAISON® XL (DiaSorin), and Elecsys® anti-SARS-CoV-2 total (IgA, IgG, IgM) on the cobas e411 (Roche Diagnostics). Details and performance characteristics of these four serologic assays are described in Table S1 . The Abbott anti-SARS-CoV-2 IgG and IgM assays are qualitative chemiluminescent microparticle immunoassays (CMIA). The anti-SARS-CoV-2 IgG assay detects IgG antibodies to the nucleocapsid protein of SARS-CoV-2, while the anti-SARS-CoV-2

IgM assay detects IgM antibodies to the receptor-binding domain (RBD) of the spike protein (S1) of SARS-CoV-2. The DiaSorin anti-SARS-CoV-2 S1/S2 IgG assay is a qualitative chemiluminescent immunoassay (CLIA) that detects IgG antibodies against the spike protein (S1 and S2 subunits) of SARS-CoV-2. Lastly, the Roche anti-SARS-CoV-2 total assay is a qualitative electrochemiluminescence immunoassay (ECLIA) that detects IgA, IgM and IgG antibodies to the nucleocapsid protein of SARS-CoV-2. Preventative maintenance, function checks, calibration, and internal quality control were performed according to the manufacturer's instructions.

To determine the sensitivity of four anti-SARS-CoV-2 antibody assays, serum or plasma samples were collected from 175 patients that were confirmed positive for SARS-CoV-2 infection by PCR testing within the previous 0-73 days. Sensitivity was calculated as true positive / (true positive + false negative), where true positivity was defined as PCR positivity.

Total sensitivity and sensitivity for various categories of days post-PCR positivity were determined.

To determine the cross-reactivity of the four anti-SARS-CoV-2 antibody assays, serum or plasma samples were collected from 107 patients that were positive for viruses other than SARS- Specificity was assessed using 32 of these samples that were collected from patients in 2019, before SARS-CoV-2 was thought to be circulating in Ontario, Canada. Specificity was calculated as true negative / (true negative + false positive), where true negativity was defined as negative by PCR or sample collection prior to the circulation of SARS-CoV-2.

Concordance between the four anti-SARS-CoV-2 antibody assays was determined using all 175 samples from the sensitivity analysis and all 107 samples from the cross-reactivity and specificity analysis. Overall percent agreement, positive percent agreement (PPA), negative percent agreement (NPA), and Cohen's kappa were calculated for each pair-wise comparison.

Serial serum and plasma samples (n=6-20) were collected from five patients that were positive for SARS-CoV-2 by PCR. Data was expressed as a ratio of the response (i.e. assay signal) to the positivity cut-off of the assay. This ratio was examined over time since PCR positivity.

Sensitivity for <14 days and ≥14 days post-PCR positivity, as well as specificity for each of the four anti-SARS-CoV-2 serologic assays as determined in this study were used to estimate the positive predictive value (PPV) and negative predictive value (NPV) at seroprevalence values of 1%, 5%, and 10%. Anti-SARS-CoV-2 antibody sensitivity of four SARS-CoV-2 serologic assays is summarized in Table 1 and Fig. 1 . The total sensitivity (independent of days since PCR positivity) ranged from 69.0% to 74.9% across assays. All assays exhibited greater sensitivity ≥14 days post-PCR positivity (76.5%-82.4%) compared to <14 days (52.1%-64.4%), although Abbott anti-SARS-CoV-2 IgM exhibited the greatest sensitivity among the assays <14 days. All assays exhibited the highest sensitivity between 15-30 days post-PCR positivity, ranging from 83.3% to 85.2%,. As expected, the lowest sensitivity was observed for samples collected the same day as PCR positivity, ranging from 38.1% to 63.6% with the highest sensitivity observed for Abbott anti-SARS-CoV-2 IgM. Overall, 20.6% of patients positive for SARS-CoV-2 by PCR were negative across all four serologic assays, in-line with the large immunocompromised population at UHN.

When including both Abbott anti-SARS-CoV-2 IgG and IgM assays, total sensitivity improved from 74.9% (IgG alone) and 74.3% (IgM alone) to 77.7%, as shown in Table S2 and were negative for anti-SARS-CoV-2 antibodies by all four assays. All 32 control patient samples that were collected pre-COVID-19 were negative by all four serologic assays. Therefore, all four serologic assays exhibited an assay specificity of 100%. Table S3 and Fig. S2-S3 summarize the concordance between the four serologic assays. 

Trends in anti-SARS-CoV-2 antibody levels over time since PCR positivity, as measured by four serologic assays, are shown in Fig. 2-4, S4-S5 . It is evident that the antibody response to SARS-CoV-2 infection varies between individual patients and across serologic assay platforms and isotypes. Patients 1 and 2 ( Fig. 2-3 Table S4 summarizes days from PCR positivity to seroconversion as determined for each serologic assay. Overall, the time to seroconversion was lowest for Abbott anti-SARS-CoV-2 IgM (median: 0 days; mean: 2.2 days), followed by Roche anti-SARS-CoV-2 total, which also contains IgM (median: 2 days; mean: 3.2 days). Abbott anti-SARS-CoV-2 IgG (median: 3 days; mean: 5.2 days) and DiaSorin anti-SARS-CoV-2 IgG (median: 3 days; mean: 4.8 days) exhibited longer times to seroconversion.

PPV and NPV for the four serologic assays across different seroprevalence estimates (1%, 5%, 10%) <14 days post-PCR positivity and ≥14 days are shown in Table 2 . PPV for all four serologic assays was 100% across all seroprevalence estimates due to the specificity of 100% determined in this study. The NPV across all assays was higher for samples collected ≥14 days post-PCR positivity compared to <14 days, ranging from 97.7%-99.8% and 95.4%-99.7%, respectively. Notably, NPV decreased with increasing seroprevalence.

Here, we report analytical and clinical performance characteristics of the Abbott anti- Chemiluminescent anti-SARS-CoV-2 serologic assays, as used in this study, have been reported to exhibit high sensitivity, as summarized in a systematic review and meta-analysis (21) . The highest pooled sensitivity for anti-SARS-CoV-2 serologic assays was observed for chemiluminescent assays (i.e. 97.8%), followed by enzyme-linked immunosorbent assays (84.3%) and lateral flow immunoassays (66.0%) (21) , the latter most commonly used in point-ofcare testing systems. Furthermore, anti-SARS-CoV-2 antibodies were reported to have higher pooled sensitivity at least three weeks after symptom onset compared to the first week and there was no association between immunoglobulin class and pooled sensitivity or specificity (21) . The higher overall pooled sensitivity reported in this systematic review and meta-analysis compared to our study is likely due to its higher combined sample size of over 29,000 tests and the fact that 60% of our samples were analyzed within three weeks of PCR positivity. Perkmann, et al. also evaluated Abbott anti-SARS-CoV-2 IgG, DiaSorin anti-SARS-CoV-2 IgG, and Roche anti-SARS-CoV-2 total and reported sensitivities more similar to those observed in our study, ranging from 83.1% to 89.2% in 65 samples from COVID-19 patients evaluated ≥14 days after symptom onset (22) . However, as is apparent in our study and others (10, 23) , sensitivity of the different immunoglobulin classes differ depending on days post symptom onset or PCR positivity.

As IgM is a marker of acute infection, anti-SARS-CoV-2 IgM sensitivity has been reported to be higher between 8-14 days (94.4%) compared to ≥15 days (89.5%), likely reflecting antibody class switching (24) . While PCR sensitivity was higher than anti-SARS-CoV-2 IgM ELISA before 5.5 days post-symptom onset, IgM sensitivity was higher than PCR after 5.5 days post-symptom onset (10) . Overall, sensitivity increased from 51.9% to 98.6% when testing PCR-negative patients with IgM ELISA compared to PCR testing alone (10), supporting its potential use as an adjunct test for early diagnosis of COVID-19. While we did not compare the performance of PCR and IgM testing, we did report serologic positivity the same day as positive molecular-based SARS-CoV-2 RNA detection, supporting its ability to detect antibody response while viral shedding is still occurring. Furthermore, we found that measuring anti-SARS-CoV-2 IgM and IgG on the same sample exhibits greater sensitivity than anti-SARS-CoV-2 IgG testing alone, especially when measured on the same day as PCR positivity. While anti-SARS-CoV-2 IgG assays exhibit good clinical performance, their greatest utility is seen two weeks after symptom onset (23, (25) (26) (27) . For example, sensitivity of the Abbott SARS-CoV-2 IgG assay was reported to reach 100% at day 17 after symptom onset (day 13 after PCR positivity) when assessing 689 serum specimens (25) and to exhibit an overall sensitivity of 78.3%, reaching 100% sensitivity 15 days after symptom onset when assessing 141 serum specimens (26) . We report a similar trend with Abbott SARS-CoV-2 IgG sensitivity 64.4% <14 days post- However, notable differences include that the antigen utilized in the Roche anti-SARS-CoV-2

Total assay is the nucleocapsid, similar to the anti-SARS-CoV-2 Abbott IgG assay but unlike the anti-SARS-CoV-2 Abbott IgM assay, which utilizes the RBD of the spike protein.

Furthermore, the Roche anti-SARS-CoV-2 Total assay detection method is electrochemiluminescence compared to chemiluminescence utilized by the Abbott assays and the Roche anti-SARS-CoV-2 Total assay also detects IgA antibodies. The potential benefit of measuring IgA may include its appearance earlier in disease course compared to other antibody subtypes (10, 28, 29) . However, we did not find any added sensitivity benefit of measuring IgA, which may have been observed if the addition of IgA was compared to an assay from the same manufacturer without IgA, rather than comparing to another manufacturer, and/or if we were able to analyze sensitivity by days post-symptom onset, rather than PCR positivity. IgA has also been reported to potentially associate with disease severity more strongly as compared to IgM and IgG (28) , however, we were unable to compare its association to disease severity due to the lack of clinical information available on our cohort.

Our study reported 100% specificity for all four serologic assays using 32 samples (22) . Other studies reported specificities of 99.9% using 1,020 serum specimens collected prior to SARS-CoV-2 circulation (25) and 99.3% using 107 serum specimens collected either before the emergence of SARS-CoV-2 or from PCR negative patients (26) for Abbott anti-SARS-CoV-2 IgG.

The trend observed in anti-SARS-CoV-2 IgG (Abbott) index values measured in serial samples over time were previously shown to vary among eight COVID-19 patients confirmed with PCR positivity (25) . Some patients reached the positivity threshold by 12 days postsymptom onset, others plateaued after 12-16 days, while others continued to rise after 16 days (25) . One study of sixty-three patients with confirmed COVID-19 collected samples at 3-day intervals until discharge and reported seroconversion in 96.8% of patients with the median time to seroconversion of 13 days (12) . Interestingly, three seroconversion types were observed: synchronous IgG and IgM seroconversion, IgM seroconversion earlier than IgG, or IgM seroconversion later than IgG (12) . The trends observed in our study were similarly heterogeneous between patients and the average time to seroconversion was 5. Our study is not without limitations. Firstly, we did not collect information on the day of symptom onset due to confidentiality of personal health information, and thus all temporal data is only depicted as days since PCR positivity. This limits the accuracy of our assessment of sensitivity and antibody trends, as they are dependent on when the individual was tested for SARS-CoV-2 infection by PCR. However, this measure does reduce the potential subjectivity of using day of symptom onset as defined by each individual patient. Furthermore, as we did not collect information on clinical presentation and/or symptom severity, we are unable to make correlations between the assay signal and/or duration of antibody positivity with disease severity and/or patient outcome.

In conclusion, we report a sensitivity of 87.5% 15-30 days post-PCR positivity, an overall specificity of 100%, and no cross-reactivity with patient samples containing autoantibodies, viral antigens, or viral antibodies for the Abbott anti-SARS-CoV-2 IgM assay. Combining Abbott anti-SARS-CoV-2 IgM and IgG testing improved sensitivity by 22.7% compared to IgG testing alone when tested the same day as PCR positivity. The Abbott anti-SARS-CoV-2 IgM assay exhibited the earliest response and greatest signal in the majority of patients evaluated with serial sampling and had the highest NPV <14 days post-PCR positivity, suggesting its potential utility as an adjunct test to PCR early in disease course. Zoomed in view to better visualize the response around the positivity cut-off (i.e. ratio =1) Table 1 . Sensitivity comparison among four SARS-CoV-2 serologic assays by days post-PCR positivity

A Novel Coronavirus from Patients with Pneumonia in China

Weekly Epidemiological Update: Coronavirus disease 2019 (COVID-19) [Internet]. World Health Organization (WHO)

Characteristics of and Important Lessons From the Coronavirus Disease 2019 (COVID-19) Outbreak in China: Summary of a Report of 72 314 Cases From the Chinese Center for Disease Control and Prevention

Assay Techniques and Test Development for COVID-19 Diagnosis. ACS central science

Interpreting Diagnostic Tests for SARS-CoV-2

Viral load dynamics and disease severity in patients infected with SARS-CoV-2 in Zhejiang province, China

Virological assessment of hospitalized patients with COVID-2019

Variation in False-Negative Rate of Reverse Transcriptase Polymerase Chain Reaction-Based SARS-CoV-2 Tests by Time Since Exposure

Antibody tests for identification of current and past infection with SARS-CoV-2. The Cochrane Database of Systematic Reviews

Profiling Early Humoral Response to Diagnose Novel Coronavirus Disease (COVID-19)

Combination of serological total antibody and RT-PCR test for detection of SARS-COV-2 infections

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

Different longitudinal patterns of nucleic acid and serology testing results based on disease severity of COVID-19 patients. Emerging Microbes & Infections

Ten reasons why immunity passports are a bad idea

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

SARS-CoV-2 seroprevalence and neutralizing activity in donor and patient blood from the San Francisco Bay Area. medRxiv

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

Serological assays for emerging coronaviruses: Challenges and pitfalls. Virus Research

Canadian Society of Clinical Chemists, Forgie SE, Association of Medical Microbiology and Infectious Disease Canada, et al. SARS-CoV-2 (COVID-19) serology: implications for clinical practice, laboratory medicine and public health

Diagnostic accuracy of serological tests for covid-19: systematic review and meta-analysis

Side by side comparison of three fully automated SARS-CoV-2 antibody assays with a focus on specificity

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

Longitudinal Monitoring of SARS-CoV-2 IgM and IgG Seropositivity to Detect COVID-19

Performance Characteristics of the Abbott Architect SARS-CoV-2 IgG Assay and Seroprevalence in

Assessment of SARS-CoV-2 serological tests for the diagnosis of COVID-19 through the evaluation of three immunoassays: Two automated immunoassays (Euroimmun and Abbott) and one rapid lateral flow immunoassay (NG Biotech)

Clinical Performance of Two SARS-CoV-2 Serologic Assays

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

A serological assay to detect SARS-CoV-2 seroconversion in humans

We would like to thank the UHN Core Lab Specimen Management team, Tech IVs, and biochemists for their help with sample retrieval and input. VK serves as a consultant for Abbott Diagnostics. DD and MB are employed by Abbott. Reagent kits were provided by AbbottDiagnostics. This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.