key: cord-0332251-293ib7js authors: Remy, L.; Tomomori-Sato, C.; Conkright-Fincham, J.; Wiedemann, L. M.; Conaway, J. W.; Unruh, J. R. title: Comparison of Antibody Levels in Response to SARS-CoV-2 Infection and Vaccination Type in a Midwestern Cohort date: 2021-08-18 journal: nan DOI: 10.1101/2021.08.16.21262036 sha: f225cc7a9bf302fd5002a754b130a5249973c302 doc_id: 332251 cord_uid: 293ib7js We present preliminary data in an ongoing observational study reporting SARS-CoV-2 spike protein reactive antibody levels from a convenience cohort of over 200 individuals in Kansas City. We observe stable antibody levels over 11 months in individuals who recovered from COVID19 infection caused by SARS-CoV-2. Our data revealed higher-than recovered levels from naive individuals vaccinated with Pfizer or Moderna vaccines and similar-to recovered levels from Johnson & Johnson (J&J) recipients. For all vaccines, inoculation after recovery resulted in higher antibody levels than vaccination alone. Responses to Pfizer and Moderna vaccines decreased over time from high initial levels but at the time of publication remain higher than those for recovered or J&J recipients. Within our limited cohort we did not see strong demographic trends other than higher antibody levels in recovered female individuals. The SARS-CoV-2 epidemic in the United States stretched the limits of health care informatics, demanding real-time in-depth information about viral load and antibody responses across the country. This challenge was felt especially in the American heartland, where access to accurate and timely testing had been limited. Testing was often performed in response to suspected infections, rather than proactively across time, and antibody testing was rarely routinely performed. We present preliminary data from a convenience sample in the Midwest over 11 months. Subjects were routinely polymerase chain reaction (PCR) tested for SARS-CoV-2 and tested over time for antibodies to the SARS-CoV-2 spike protein. Our objectives were to monitor levels of anti-SARS-CoV-2 Spike IgG antibody monthly in serum samples from a population with access to routine testing and to correlate those levels with infection, vaccination, and, where possible, demographic information. The Stowers Institute for Medical Research is a non-profit biomedical research institute located in Kansas City, Missouri. Stowers has approximately 550 employees including graduate students, post-doctoral trainees and support staff. At the beginning of COVID lockdowns in 2020, a routine surveillance PCR SARS-CoV-2 testing program was initiated to ensure safe return to work. In August 2020, our team began regular monthly collection of serum plasma for antibody detection on a convenience sample of employees and their household members. Data collection is ongoing. Titration based ELISA testing was performed on volunteered whole blood samples once per month for 11 months. The target of the ELISA test was a synthetic purified spike protein (original strain modified for stability) Robbiani et al., 2020; Wu et al., 2020) , which was detected using HRP labeled anti-human IgG secondary antibodies (Conkright-Fincham et al., 2021; Stadlbauer et al., 2020; Stadlbauer et al., 2021) . The test was initially validated on commercial negative, active and recovered samples (Conkright-Fincham et al., 2021) . For each sample, raw ELISA signals were plotted as a function of dilution, and results were reported as area under the curve (AUC) as calculated from a best fit to the data, normalized to the AUC obtained with the same set of dilutions performed on a sample of 20 ng/µL of a commercial anti-SARS-CoV-2 antibody (AM001414, Active Motif), such that a signal level of 20 represents equal area under the curve to the commercial antibody. Such a signal approximates a logarithmic transform of the titer value (Conkright-Fincham et al., 2021 ). Participants self-collected 100-150 µl fingertip blood specimens in capillary blood collection containers and serum samples were pre-processed as described in Conkright-Fincham et al (2021) . All rights reserved. No reuse allowed without permission. (which was not certified by peer review) 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 August 18, 2021. ; https://doi.org/10.1101 https://doi.org/10. /2021 Biases include cohort educational, economical, and racial bias (academic research center) relative to our community. Please see limitations section for detailed discussion. All error bars shown are standard error of the mean. In cases with two groups, t-tests were used for significance testing. For other groups, ANOVA was used. The study was approved by the MRIGlobal Institutional Review Board (registration IRB00000067). Written informed consent for study participation was obtained from all participants. In total, 295 individuals inquired about this study of which 289 were enrolled. Study enrollment was limited to institute members and immediate family members over the age of 18. A total of 33 participants withdrew from the study for various reasons, including sample collection discomfort. At the time of this publication, there were 252 participants enrolled. Samples from unvaccinated individuals reporting positive SARS-CoV-2 PCR tests were compared to those from unvaccinated individuals not reporting positive results. Those results are shown in Fig. 1 . Recovered individual samples displayed a range of signal levels. Recovered individual samples showed consistent and significant increases over those not reporting infection, with an average signal level of 22 for recovered individual samples compared to 1.2 for those not reporting infection. A small number of samples showed positive antibody signals without a positive PCR test. Those samples were ruled to be a result of asymptomatic infections after investigating contacts and symptoms. Positive values followed an approximately normal distribution with a minimum signal of 10. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) 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 August 18, 2021. ; https://doi.org/10.1101/2021.08.16.21262036 doi: medRxiv preprint Figure 1 . Anti-SARS-CoV-2 spike antibody (IgG) signals (relative to AM001414) for unvaccinated individuals reporting positive PCR tests and those not reporting positive PCR tests. N refers to number of independently collected samples. Samples from unvaccinated individuals reporting positive SARS-CoV-2 PCR tests were recorded as a function of time after reported symptom onset. For asymptomatic individuals, the date of the positive test was used as the zero-time point. This data set includes multiple samples from many individuals. Fig. 2 shows those signals as a function of time. It is important to note that the assay detects IgG representing long lasting antibody signals. No significant decrease in antibody signals was detected from the aggregate data set over the 11-month time span measured during this study, though certain individuals may have shown a decrease. Three individuals were asymptomatic accounting for six of the samples included. Previous studies have shown decreases in antibody levels over time (Long et al., 2020) . It is possible that we do not observe this for our population because of the nature of our cohort or because of the nature of our assay. Certainly, limiting detection to IgG eliminates well documented decreases in other immunoglobulin species observed in some studies. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Antibody signals as a function of time after reported vaccination are shown in Fig. 3 . Strong antibody responses are observed for Moderna and Pfizer vaccinated individuals starting around day 10 after the first dose of vaccine. After day 20, levels are variable but relatively stable with average signals slightly below 40. After 40 days, signal levels from Pfizer and Moderna decline significantly, trending towards the recovered individual level at 120 days after the first dose. Further data is necessary to understand if this trend will continue or level off at longer time points. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. 4 shows statistical comparisons of different groups including recovered individuals more than 20 days after the first vaccine dose. J&J vaccine represents the lowest signal levels followed by recovered unvaccinated (slightly higher), followed by Pfizer and Moderna and recovered vaccinated groups. Combining Pfizer and Moderna signals results in a statistical increase for recovered vaccinated individuals above naïve vaccinated individuals with P<0.0001. That result matches measurements from other studies as well (Bradley et al., 2021; Krammer et al., 2021) . The timeline of vaccination response for recovered individuals does not show significant trends over time (Fig. S1 ). These data clearly demonstrate the rapid increases in antibody binding in response to all vaccinations ten days after the first dose. The lower levels of antibody binding in response to J&J vaccination correlate with the lower reported efficacy of that vaccine. Nevertheless, is difficult to go beyond simple correlation given that our study measures binding only to one set of epitopes and does not measure neutralization or B cell production which are key factors in the immune response. At the time of publication, we do not have data on new variant binding which may differ from the binding of the original strain. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) 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 August 18, 2021. ; https://doi.org/10.1101/2021.08.16.21262036 doi: medRxiv preprint We explored the demographic diversity of our data set to look for potential differences in antibody generation with gender, age, and race. Out of these demographic factors, only gender significantly impacted the antibody production in response to infection with SARS-CoV-2. Gender did not affect the vaccination response for Pfizer and Moderna manufacturers. We did not have enough data from individuals vaccinated with the J&J vaccine to compare. We also did not have enough samples from recovered, unvaccinated individuals to measure the impact of race, but all other factors are presented for vaccination response in Fig. S2 and infection response in Fig. S3 . All rights reserved. No reuse allowed without permission. (which was not certified by peer review) 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 August 18, 2021. ; https://doi.org/10.1101 https://doi.org/10. /2021 We have measured IgG in human serum via binding to spike protein in an automated ELISA assay over time for a small cohort in Kansas City undergoing frequent PCR testing. Our cohort showed sustained antibody signals after infection, extending to ten months after infection. Responses to vaccines from Moderna, Pfizer, and J&J were clearly seen ten days after the first dose and more consistently after 20 days. Antibody signals in individuals who received the J&J vaccine were significantly lower than in those who received vaccines from the other manufacturers and in recovered individuals, correlating with the reduced efficacy reported for the J&J vaccine (Sadoff et al., 2021) . After vaccination with Pfizer and Moderna vaccines, responses peaked around 40 days after the first dose and trended downwards over 120 days. Future testing is needed to determine whether antibody levels continue to decline or whether responses tend to remain at levels similar to those seen in recovered individuals. In agreement with previous studies, the J&J vaccine elicited a clear increase in signal for those vaccinated after recovery, with a similar trend from Pfizer and Moderna recovered samples. We observed little in the way of demographic impacts on antibody production, with only gender impacting infection recovery and no impact of gender, age, or race on responses elicited by the Pfizer and Moderna vaccines. Perhaps the biggest limitation of our data set is the focus on IgG binding to a single spike isoform. Variations in the spike protein as well as variations in the epitopes presented to serum antibodies are All rights reserved. No reuse allowed without permission. (which was not certified by peer review) 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 August 18, 2021. ; https://doi.org/10.1101 https://doi.org/10. /2021 known to cause significant changes in antibody binding. In addition, IgA and IgM antibodies represent a strong initial antibody response to vaccination and infection and are well known to decay following response. Those limitations likely explain our lack of decay after recovery as reported elsewhere. Our assay also does not measure immune cell populations or viral neutralization, which are well known components of the immune response. Therefore, our antibody measurement represents a biased indicator of immune response over time and should not be interpreted on its own. We also rely on self-reported vaccine data. Our data set suggests that that data is fairly accurate but when considering outliers reporting inaccuracies could have a significant impact. Another limitation is uncertainty in collection method and timing. Samples were self-collected during a 24-hour period prior to serum preparation and freezing. While check-in steps should eliminate poorly collected and preserved samples, there is a possibility of contamination, timing error, and unknown collection factors that should be considered when interpreting the data. Finally, it is important to note that our cohort, being volunteers from an academic research center, is a biased representation of our local community as a whole. As such, they are likely more educated and financially stable than our surrounding community. Racially, our cohort is more dominated by Asian and White participants than the general population. These factors dramatically affect access to and knowledge of health care and thus, our findings should be interpreted with those factors in mind. A serological assay to detect SARS-CoV-2 seroconversion in humans Antibody Responses after a Single Dose of SARS-CoV-2 mRNA Vaccine A High-Throughput Automated ELISA Assay for Detection of IgG Antibodies to the SARS-CoV-2 Spike Protein Antibody Responses in Seropositive Persons after a Single Dose of SARS-CoV-2 mRNA Vaccine Clinical and immunological assessment of asymptomatic SARS-CoV-2 infections Convergent antibody responses to SARS-CoV-2 in convalescent individuals Interim Results of a Phase 1-2a Trial of Ad26.COV2.S Covid-19 Vaccine SARS-CoV-2 Seroconversion in Humans: A Detailed Protocol for a Serological Assay, Antigen Production, and Test Setup Repeated cross-sectional sero-monitoring of SARS-CoV-2 in New York City A new coronavirus associated with human respiratory disease in China We thank John Pak, Ph.D from Chan Zuckerberg Biohub for tips on spike protein expression and purification. We thank the Florian Krammer lab for the gift of the Spike protein expression vector. We are grateful to the Kansas City Community Blood Center for negative samples. We thank the many individuals from Stowers who contributed to the assay development and execution to make this study possible and, in particular, discussions with Jennifer Gerton, Brent Kreider, and David Chao. All rights reserved. No reuse allowed without permission.(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.