key: cord-0892222-vpxgdvjw
authors: McAulay, K.; Bryan, A.; Greninger, A. L.; Grill, F.; Lake, D. F.; Kaleta, E. J.; Grys, T. E.
title: Retrospective Clinical Evaluation of Four Lateral Flow Assays for the Detection of SARS-CoV-2 Antibodies
date: 2020-07-03
journal: nan
DOI: 10.1101/2020.07.01.20129882
sha: 6aa9a1bc6ab0b316afe61aeb05b07343a4ebe7f8
doc_id: 892222
cord_uid: vpxgdvjw

Coronavirus disease 2019 (COVID-19) is a potentially life-threatening respiratory infection caused by severe acute respiratory coronavirus 2 (SARS-CoV-2), for which numerous serologic assays are available. In a CLIA laboratory setting, we used a retrospective sample set (n = 457) to evaluate two lateral flow immunoassays (LFIAs; two iterations of Rapid Response COVID-19 Test Cassette, BTNX Inc.) and a subset of to evaluate SARS-COV-2 IgG/IgM Rapid Test, ACON Laboratories (n = 200); and Standard Q COVID-19 IgM/IgG Duo, SD BIOSENSOR (n = 155) for their capacity to detect of SARS-CoV-2 IgG. In a cohort of primarily hospitalized patients with RT-PCR confirmed COVID-19, the BTNX assays demonstrated 95% and 92% agreement with the Abbott SARS-CoV-2 IgG assay and sensitivity was highest at [≥] 14 days from symptom onset [BTNX kit 1, 95%; BTNX kit 2, 91%]. ACON and SD assays demonstrated 99% and 100% agreement with the Abbott assay at [≥] 14 days from symptom onset. Specificity was measured using 74 specimens collected prior to SARS-CoV-2 circulation in the United States and 31 cross-reactivity challenge specimens, including those from patients with a history of seasonal coronavirus infection and was 98% for BTNX kit 1 and ACON and 100% for BTNX kit 2 and SD. Taken with data from EUA assays, these results suggest that LFIAs may provide adequate results for rapid detection of SARS-CoV-2. Replicating these results in fingerstick blood in outpatient populations, would further support the possibility that LFIAs may be useful to increase access to serologic testing.

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) emerged in 25 2019 as the causative agent of coronavirus disease 2019 (COVID-19), a pandemic 26 respiratory infection resulting in over 10 million cases and 500,000 deaths between 27 November 2019 and June 2020 (WHO, 2020). 28 Antibody detection is currently being implemented in many clinical centers to aid 29 in identification of recent disease and to investigate population seroprevalence. 30 Accurate laboratory tests impact clinical decision-making, and understanding 31 performance of a test is essential to determination of when to use the test and what the 32 results might mean. For example, specificity is of particular importance in a low 33 prevalence setting (Farnsworth and Anderson, 2020) . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted July 3, 2020. . https://doi.org/10.1101/2020.07.01.20129882 doi: medRxiv preprint 4 blood specimens (a further four unknown specimens were assumed to be either serum 47 or plasma), were received frozen, and underwent either one or two freeze-thaw cycles 48 prior to testing. Specificity specimens were obtained from two sources: 74 excess 49 clinical serum specimens collected and stored in 2018, and 31 "cross-reactivity 50 challenge" specimens collected between March and April 2020. Among these 105 51 specimens, there were 27 from individuals with a history of seasonal coronavirus 52 infection (as determined by a syndromic respiratory PCR test) within 3 years prior to 53 collection (HKU1, n = 13; NL63, n = 6; OC43, n = 6; 229E, n = 2), and 4 specimens 54 reactive for rheumatoid factor, HIV-1 antibody, HAV total antibody, HBV core total 55 antibody and surface antibody, HCV antibody and/or HSV2 antibody. These specimens 56 were tested after 0, 1, or 2 freeze thaw cycles. hereafter referred to as ACON, 10 µL serum or plasma, or 15 µL whole blood was 68 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted July 3, 2020. . https://doi.org/10.1101/2020.07.01.20129882 doi: medRxiv preprint 5 transferred to the specimen well and then two drops of buffer were added to the buffer 69 well; results were read after 10-15 min. 70 Standard Q COVID-19 IgM/IgG Duo (SD BIOSENSOR): This kit is supplied as 71 individual IgM and IgG cartridges; only the IgG cartridges were evaluated in this study 72 and are hereafter referred to as SD. For this assay, 10 µL serum, plasma, or whole 73 blood was transferred to the specimen well and then two to three drops of buffer were 74 added to the buffer well; results were read and interpreted after 10-15 minutes. 75 Results for all assays were interpreted by two readers (KM, TG) and 76 photographed for reference, with the exception of the SD assay, for which the first 30 CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted July 3, 2020. Sensitivity of the BTNX assays was evaluated using specimens from a primarily 93 hospitalized cohort of individuals with RT-PCR confirmed COVID-19 (Seattle cohort, n = 94 352) and stratified by the number of days since symptom onset. Sensitivity of BTNX kit 95 1 at <7 days since symptom onset (n = 154) was 16% (95% CI: 10-22%), at 7-13 days 96 (n = 103) it was 48% (38-58%), and ≥ 14 (n = 95) days it was 95% (88-98%). Sensitivity 97 of BTNX kit 2 at the same time points was 13% (8-19%), 50% (40-60%), and 91% (83-98 96%), respectively. 99 We then compared assay performance to that of the Abbott SARS-CoV-2 IgG 100 assay, which holds and Emergency Use Authorization (EUA) from the FDA and for 101 which optical density (OD) values and interpretations for 268 of these specimens. For a 102 number of samples, >1 sample result was available from the same patient on the same 103 day since symptom onset. When this occurred, the mean OD value was determined and 104 assigned to all samples collected that day. We reviewed the sample specific OD and 105 mean OD for 157 specimens for which both values were available and found that taking 106 the mean did not alter the interpretation in any case; therefore, we opted to use the 107 mean data for comparison with LFIAs. Overall agreement with the Abbott assay was 108 95% ( CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted July 3, 2020. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted July 3, 2020. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted July 3, 2020. . https://doi.org/10.1101/2020.07.01.20129882 doi: medRxiv preprint 9 CoV-2 have been primarily focused on automated assays. Here we evaluated four 161 LFIAs for their capacity to detect anti-SARS-CoV-2 IgG in retrospective serum, plasma, 162 and whole blood specimens, focusing on sensitivity at ≥ 14 days after symptom onset in 163 an inpatient cohort and on specificity. is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted July 3, 2020. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted July 3, 2020. One potential use for these assays might be to confirm antibody production in 218 patients with resolved symptoms, independent of disease detection by a SARS-CoV-2 219 PCR assay or not. In this case, after a 14 day self-quarantine, would these serological 220 tests detect a positive result? The results of this study of sera from a primarily 221 hospitalized population show that the sensitivity of IgG detection at 14 days or more 222 post-symptom onset was 95% in two cases (BTNX kit 1, and ACON) . When compared 223 directly with Abbott results, sensitivity increased to 99% for both of these assays; 224 similarly, BTNX kit 2, and SD sensitivity was 94% and 100%, respectively. These LFIA 225 tests show good performance for a 15-minute test that is very easy to perform; however, 226 BTNX kit 1 and ACON were the only assays to generate false positive IgG results, 227 supporting the theory that assays providing higher sensitivity may come with a 228 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted July 3, 2020. . https://doi.org/10.1101/2020.07.01.20129882 doi: medRxiv preprint 12 compromise in specificity. Nonetheless three of the kits tested did not show any false 229 positives in a sample set that included a diverse representation of potential cross-230 reactivity, and specificity for any assay did not fall below 98%. Additional studies will be 231 needed to determine if this measure of sensitivity holds true for more mild disease, and 232 whether sensitivity may increase (or decrease) past 14 days. Other studies have shown 233 improved sensitivity after 17 days from symptom onset (Bryan et al., 2020) . While no 234 serologic test is perfect, these results are encouraging that rapid and simple tests can 235 provide an adequate level of sensitivity and specificity. Importantly, several other LFIAs 236 tested by our group showed poor sensitivity and/or specificity (data not shown), 237 indicating the importance of rigorous validation prior to implementation in any setting. It 238 is likely that all tests will have a measurable false-positivity rate, but our results suggest 239 that a substantial number of samples from patients with a history of seasonal CoV or 240 even other viral infections will be required to better define the rate of false-positivity. 241 Even if some tests maintain a high (>99%) specificity, the individual patient may be best 242 served by an orthogonal approach to testing, whereby two methods that target different 243 antigens (whether two LFIAs or an LFIA followed by an EIA or CIA) are used to increase 244 positive predictive value for predicting true exposure to SARS-CoV-2. However, 245 manufacturers are only obliged to disclose the nature of their assay target(s) upon EUA 246 issuance, so the role of the many pending assays in this approach is currently unclear. 247 The sensitivity of the LFIAs characterized herein suggests that such an approach would 248 have only a minor impact on clinical sensitivity overall by using two assays. This 249 concept is supported by the fact that of the false negatives, four samples were not 250 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted July 3, 2020. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted July 3, 2020. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted July 3, 2020. Pathology staff for all their hard work supporting this study and the care of our patients 293 during this pandemic. We would also like to thank the manufacturers for supplying some 294 of the kits (ACON and BTNX kit 1). We also thank Safe Health Systems who supplied 295 some kits (SD and BTNX kit 2) as part of a joint partnership with Mayo Clinic. 296 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted July 3, 2020. This research did not receive any specific grant from funding agencies in the public, 303 commercial, or not-for-profit sectors

Performance 306 Characteristics of the Abbott Architect SARS-CoV-2 IgG Assay and Seroprevalence

Dynamic 309 profile for the detection of anti-SARS-CoV-2 antibodies using four 310 immunochromatographic assays

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

Authorized Serology Test Performance

Prevalence of antibodies to four human 317 coronaviruses is lower in nasal secretions than in serum

International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity

The copyright holder for this preprint this version posted July 3, 2020. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted July 3, 2020. . https://doi.org/10.1101/2020.07.01.20129882 doi: medRxiv preprint