key: cord-0718458-f86jlksx authors: Hicks, S.; Pohl, K.; Neeman, T.; McNamara, H.; Parsons, K.; He, J.-S.; Ali, S.; Nazir, S.; Rowntree, L.; Nguyen, T.; Kedzierska, K.; Doolan, D.; Vinuesa, C.; Cook, M.; Coatsworth, N.; Myles, P.; Kurth, F.; Sander, L.; Gruen, R.; Mann, G.; George, A.; Gardiner, E.; Cockburn, I. title: A dual antigen ELISA allows the assessment of SARS-CoV-2 antibody seroprevalence in a low transmission setting date: 2020-09-14 journal: nan DOI: 10.1101/2020.09.09.20191031 sha: 414aea261d9ea2b6f95185cbc20a57b254de6578 doc_id: 718458 cord_uid: f86jlksx Estimates of seroprevalence of SARS-CoV-2 antibodies have been hampered by inadequate assay sensitivity and specificity. Using an ELISA-based approach to that combines data about IgG responses to both the Nucleocapsid and Spike-receptor binding domain antigens, we show that near-optimal sensitivity and specificity can be achieved. We used this assay to assess the frequency of virus-specific antibodies in a cohort of elective surgery patients in Australia and estimated seroprevalence in Australia to be 0.28% (0 to 0.72%). These data confirm the low level of transmission of SARS-CoV-2 in Australia before July 2020 and validate the specificity of our assay. responses to SARS-CoV-2 [4] . Further, serological testing is likely to be valuable in the 82 assessment of vaccine efficacy. However, analyses of seroprevalence, especially in low 83 prevalence settings are hampered by assays with inadequate sensitivity and specificity [5] . 84 85 Australia has reported low case numbers of COVID-19 per head of population compared to 86 other developed Westernized countries, especially before the July/August 2020 outbreak in 87 Melbourne, Victoria (Australian Department of Health). Efforts to control the spread of the 88 virus have likely been helped by relative geographical isolation and an advanced healthcare 89 system. However nucleic acid testing generally only reveals a fraction of the total numbers of 90 infections thus the overall numbers of previous infections is unknown [3, 6] . Nonetheless it is 91 likely that the total of previously infected individuals is low as a proportion of the population 92 (<1%) and thus assessment of seroprevalence requires the use of highly sensitive and specific 93 methodologies. In order to assess the seroprevalence of SARS-CoV-2 in Australia we 94 . CC-BY-NC-ND 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted September 14, 2020. . https://doi.org/10.1101/2020.09.09.20191031 doi: medRxiv preprint therefore developed a dual-antigen ELISA assay which gave superior sensitivity and 95 specificity compared to assays that rely on single antigens. 96 . CC-BY-NC-ND 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted September 14, 2020. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted September 14, 2020. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted September 14, 2020. . https://doi.org/10.1101/2020.09.09.20191031 doi: medRxiv preprint To optimize our assay we used a library of 184 serum samples collected pre-2020 as negative 147 controls, and a panel of 43 sera from individuals infected with SARS-CoV-2 as positive 148 controls. Initial optimization of assay conditions was carried out with defined pools of sera 149 from 5 positive donors and 5 negative donors. Noting that even small gains in specificity can 150 substantially reduce the number of false negatives in large sero-surveys, we optimized the 151 concentration and amount of antigen used for coating, blocking and washing conditions. 152 Overall, we found that the principal factors affecting assay performance were the coating 153 conditions and the necessity of stringent washing (Table S1 and Figure S1 ). 154 To handle large numbers of samples we optimized our ELISA assay for automation. We is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted September 14, 2020. . https://doi.org/10.1101/2020.09.09.20191031 doi: medRxiv preprint (Figure 1 A and B) . Surprisingly, the S1 protein gave very poor sensitivity and specificity 173 with many negative samples giving high values ( Figure S3 ). We next investigated the 174 possibility of combining data for both antigens. Plotting responses to the RBD and N 175 antigens revealed that even the less responsive positive control samples generally had at least 176 elevated responses to both antigens ( Figure 1C ). Thus, using the mean of the responses to the 177 RBD and N responses, we found that a cutoff of 1.302 gave 100% sensitivity and 98.91% 178 specificity ( Figure 1D ). Neither IgA nor IgM responses distinguished positive and negative 179 donors as well as IgG, and averaging IgA or IgM responses to both antigens did not 180 substantively improve the assay ( Figure S4) . In our initial screen 41/2991 were above our cutoff of 1.302 (Figure 2A) , correcting for the 191 specificity of our assay we calculated the seroprevalence to be 0.28% (0 to 0.71%). To 192 . CC-BY-NC-ND 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted September 14, 2020. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted September 14, 2020. . https://doi.org/10.1101/2020.09.09.20191031 doi: medRxiv preprint 11,190 cases/116 deaths by 17 th July when sample collection finished suggesting that testing 217 was capturing 10-15% of cases and that there was a low case fatality rate, similar to other 218 jurisdictions with high testing rates [10]. Note that due to the small number of positive 219 samples we have not attempted to stratify our analysis based on the demographic 220 characteristics of our cohort. A key caveat of our study is that the positive controls used for 221 assay validation are skewed to hospitalized individuals and thus we do not know with 222 certainty the performance characteristics of the assays for asymptomatic cases who are 223 known to have lower antibody levels [4, 11] . Moreover, a recent study has suggested that 224 asymptomatic cases may not always seroconvert, though the assays used there had lower 225 sensitivity than we report for our assay [5, 11] . Overall however, these data suggest that the 226 low case number seen in Australia was reflective of low community transmission not 227 inadequate testing. This is supported by the fact that the subsequent outbreak in Melbourne in 228 July/August 2020 emerged from breaches of hotel quarantine of overseas travelers rather than 229 undetected community transmission. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted September 14, 2020. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted September 14, 2020. . https://doi.org/10.1101/2020.09.09.20191031 doi: medRxiv preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted September 14, 2020. . https://doi.org/10.1101/2020.09.09.20191031 doi: medRxiv preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted September 14, 2020. . https://doi.org/10.1101/2020.09.09.20191031 doi: medRxiv preprint Clinical and immunological assessment of 259 asymptomatic SARS-CoV-2 infections An evaluation of COVID-19 serological 261 assays informs future diagnostics and exposure assessment Prevalence of SARS-CoV-2 in Spain 263 (ENE-COVID): a nationwide, population-based seroepidemiological study Studying the pathophysiology of coronavirus 266 disease 2019: a protocol for the Berlin prospective COVID-19 patient cohort (Pa-COVID-267 19) A serological assay to detect SARS-CoV-2 269 seroconversion in humans Robust T cell immunity in 276 convalescent individuals with asymptomatic or mild COVID-19. bioRxiv 13 Prevalence of 283 SARS-CoV-2 Antibodies Among Healthcare Workers at a Tertiary Academic Hospital in Performance Characteristics of the Abbott 286