CoChat

SARS-CoV-2 antibody magnitude and detectability are driven by disease severity, timing, and assay

Open access

Published version of a preprint

The DOI matches its Crossref record; 30 of the 30 references it cites were found in a registry or an index.

Abstract

Antibody detection of prior SARS-CoV-2 infection depends on severity, assay, and timing with implications for serosurveillance.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1126/sciadv.abh3409

Cites

30 of 30 references verified0 citation errors · 0 not checkable

Statistics

Figures

Fig. 1 Longitudinal antibody kinetics. Time since symptom onset is shown on the x axis versus the measured antibody response for each assay. For asymptomatic individuals, the time since the first positive polymerase chain reaction (PCR) test was used. Black points indicate individual time points, and longitudinal samples are connected with gray lines. y axes are transformed as indicated in Table 2. Assay units are as follows: S-Lum (conc, relative concentration), RBD-Lum (conc, relative concentration), RBD-LIPS (LU, light unit), RBD-Split Luc (RLU, relative light unit), S-Ortho IgG (S/C, sample result to calibrator result index), S-Ortho Ig (S/C, sample result to calibrator result index), S-DiaSorin (AU/ml, arbitrary unit per milliliter), N(full)-Lum (conc, relative concentration), N(frag)-Lum (conc, relative concentration), N-LIPS (LU, light unit), N-Split Luc (RLU, relative light unit), N-Abbott (S/C, sample result to calibrator result index), N-Roche (COI, cutoff index), and Neut-Monogram (ID50, 50% inhibitory dilution). Red dotted lines indicate cutoff values for positivity, as indicated in Table 2.
Fig. 1 Longitudinal antibody kinetics. Time since symptom onset is shown on the x axis versus the measured antibody response for each assay. For asymptomatic individuals, the time since the first positive polymerase chain reaction (PCR) test was used. Black points indicate individual time points, and longitudinal samples are connected with gray lines. y axes are transformed as indicated in Table 2. Assay units are as follows: S-Lum (conc, relative concentration), RBD-Lum (conc, relative concentration), RBD-LIPS (LU, light unit), RBD-Split Luc (RLU, relative light unit), S-Ortho IgG (S/C, sample result to calibrator result index), S-Ortho Ig (S/C, sample result to calibrator result index), S-DiaSorin (AU/ml, arbitrary unit per milliliter), N(full)-Lum (conc, relative concentration), N(frag)-Lum (conc, relative concentration), N-LIPS (LU, light unit), N-Split Luc (RLU, relative light unit), N-Abbott (S/C, sample result to calibrator result index), N-Roche (COI, cutoff index), and Neut-Monogram (ID50, 50% inhibitory dilution). Red dotted lines indicate cutoff values for positivity, as indicated in Table 2.CC BY · PubMed Central
Fig. 2 Correlation of responses between assays. (A) Spearman correlation of random intercepts derived from a mixed-effects model, representing responses at 21 days after symptom onset for each individual from the longitudinal data. Assays are sorted by hierarchical clustering using average distance clustering. Darker blue indicates higher correlation; colored label box indicates antigen for each binding assay and the neutralizing assay. (B) Pairwise scatterplots showing the random intercepts for the neutralizing assay (x axis) versus the random intercepts for each of the other assays (y axis). Assay units are indicated in Table 2.
Fig. 2 Correlation of responses between assays. (A) Spearman correlation of random intercepts derived from a mixed-effects model, representing responses at 21 days after symptom onset for each individual from the longitudinal data. Assays are sorted by hierarchical clustering using average distance clustering. Darker blue indicates higher correlation; colored label box indicates antigen for each binding assay and the neutralizing assay. (B) Pairwise scatterplots showing the random intercepts for the neutralizing assay (x axis) versus the random intercepts for each of the other assays (y axis). Assay units are indicated in Table 2.CC BY · PubMed Central
Fig. 3 Severity-stratified antibody response at baseline. Swarm plot of antibody response at the baseline visit for each study participant by assay, stratified into individuals who experienced no symptoms, individuals who experienced symptoms but were not hospitalized, and those who experienced symptoms and were hospitalized. y axes are transformed as indicated in Table 2.
Fig. 3 Severity-stratified antibody response at baseline. Swarm plot of antibody response at the baseline visit for each study participant by assay, stratified into individuals who experienced no symptoms, individuals who experienced symptoms but were not hospitalized, and those who experienced symptoms and were hospitalized. y axes are transformed as indicated in Table 2.CC BY · PubMed Central
Fig. 4 Clinical predictors of antibody responses. Rank of variable importance (1 = highest rank; 50 = lowest rank) in a random forest classifier of top half versus bottom half of responders for each assay, based on random intercepts, including (A) all individuals (n = 128) and (B) only individuals who were not hospitalized (n = 97), as hospitalization is a strong predictor of antibody response. Variable importance was determined as the reduction in classification error averaged across 10 runs of the algorithm. Variables only relevant to hospitalized individuals (i.e., whether the individual was hospitalized, whether oxygen was required, whether the individual was in the ICU, and whether the individual required a ventilator) were omitted from the classifier in (B) and shown in gray. In addition, HIV status is excluded as a predictor for the Neut-Monogram assay for reasons described in the main text. The dependent variable (individual-level random intercepts derived from a mixed-effects model) is dichotomized into “high” and “low,” determined by the random intercept being in the upper or lower half of all random intercepts for that assay, respectively. Full labels of the predictor variables are provided in table S10.
Fig. 4 Clinical predictors of antibody responses. Rank of variable importance (1 = highest rank; 50 = lowest rank) in a random forest classifier of top half versus bottom half of responders for each assay, based on random intercepts, including (A) all individuals (n = 128) and (B) only individuals who were not hospitalized (n = 97), as hospitalization is a strong predictor of antibody response. Variable importance was determined as the reduction in classification error averaged across 10 runs of the algorithm. Variables only relevant to hospitalized individuals (i.e., whether the individual was hospitalized, whether oxygen was required, whether the individual was in the ICU, and whether the individual required a ventilator) were omitted from the classifier in (B) and shown in gray. In addition, HIV status is excluded as a predictor for the Neut-Monogram assay for reasons described in the main text. The dependent variable (individual-level random intercepts derived from a mixed-effects model) is dichotomized into “high” and “low,” determined by the random intercept being in the upper or lower half of all random intercepts for that assay, respectively. Full labels of the predictor variables are provided in table S10.CC BY · PubMed Central
Fig. 5 Estimated time to seroreversion and assay sensitivity by time and hospitalization status. (A) Mean time to seroreversion for individuals tested on each assay, stratified by hospitalization status, with 95% confidence intervals derived from bootstrapping. The symbol “‡” indicates increasing antibody responses over time (95% confidence interval for time to seroreversion was negative and did not cross 0), and the symbol “*” indicates antibody responses for which 95% confidence interval of time to seroreversion crossed 0. (B) Estimated sensitivity of each assay (showing posterior median estimates and 95% credible intervals), stratified by hospitalization status at 2-month intervals, from 0 to 6 months after seroconversion. Seroconversion was assumed to occur (if at all) 21 days after symptom onset (if symptomatic) or 21 days after positive PCR test (if asymptomatic).
Fig. 5 Estimated time to seroreversion and assay sensitivity by time and hospitalization status. (A) Mean time to seroreversion for individuals tested on each assay, stratified by hospitalization status, with 95% confidence intervals derived from bootstrapping. The symbol “‡” indicates increasing antibody responses over time (95% confidence interval for time to seroreversion was negative and did not cross 0), and the symbol “*” indicates antibody responses for which 95% confidence interval of time to seroreversion crossed 0. (B) Estimated sensitivity of each assay (showing posterior median estimates and 95% credible intervals), stratified by hospitalization status at 2-month intervals, from 0 to 6 months after seroconversion. Seroconversion was assumed to occur (if at all) 21 days after symptom onset (if symptomatic) or 21 days after positive PCR test (if asymptomatic).CC BY · PubMed Central
Fig. 6 Negative predictive values of the commercial assays. Negative predictive values shown are based on the estimated assay sensitivities for nonhospitalized individuals in Fig. 5B, for a range of prevalence between 5 and 50% (x axis). Bottom panels show the same data with a smaller range in the y axis to visualize small differences.
Fig. 6 Negative predictive values of the commercial assays. Negative predictive values shown are based on the estimated assay sensitivities for nonhospitalized individuals in Fig. 5B, for a range of prevalence between 5 and 50% (x axis). Bottom panels show the same data with a smaller range in the y axis to visualize small differences.CC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 30 references

Cited by

See all 130

Metadata

Authors
Peluso, M. J., Takahashi, S., Hakim, J., Kelly, J. D., Torres, L., Iyer, N. S., Turcios, K., Janson, O., Munter, S. E., Thanh, C., Donatelli, J., Nixon, C. C., Hoh, R., Tai, V., Fehrman, E. A., Hernandez, Y., Spinelli, M. A., Gandhi, M., Palafox, M.-A., Vallari, A., Rodgers, M. A., Prostko, J., Hackett, J., Trinh, L., Wrin, T., Petropoulos, C. J., Chiu, C. Y., Norris, P. J., DiGermanio, C., Stone, M., Busch, M. P., Elledge, S. K., Zhou, X. X., Wells, J. A., Shu, A., Kurtz, T. W., Pak, J. E., Wu, W., Burbelo, P. D., Cohen, J. I., Rutishauser, R. L., Martin, J. N., Deeks, S. G., Henrich, T. J., Rodriguez-Barraquer, I., & Greenhouse, B.
Affiliations
  • University of California, San Francisco (US)
  • Abbott (United States) (US)
  • Monogram Biosciences Inc., South San Francisco, CA, USA.
  • Vitalant Research Institute (US)
  • Vitalant Research Institute, San Francisco, CA, USA.
  • Chan Zuckerberg Biohub San Francisco (US)
  • National Institutes of Health (US)
  • Division of HIV, Infectious Diseases, and Global Medicine, University of California, San Francisco, San Francisco, CA, USA.
Corresponding author
  • Peluso, Michael J.
  • Takahashi, Saki
  • Hakim, Jill
  • Henrich, Timothy J.
  • Greenhouse, Bryan
Venue
Science Advances
Published
2021-07-30
Volume
7
Issue
31
ISSN
2375-2548 (ISSN-L)
OpenAlex
W3192055536
Cited by
130
References
29
Citations by year
2026: 4 · 2025: 7 · 2024: 18 · 2023: 23 · 2022: 52 · 2021: 26
FWCI
9.41
Type
journal article
Language
en
Open access
Locations
  • Science Advances (publishedVersion, cc-by)
  • medRxiv (acceptedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • eScholarship (California Digital Library) (submittedVersion, cc-by)
  • PubMed Central (submittedVersion, cc-by)
  • eScholarship (California Digital Library) (submittedVersion)
Funders
  • National Institutes of Health (T32 AI60530-12)
  • National Heart, Lung, and Blood Institute (R01 HL105704, R33- 29077)
  • National Institute of Allergy and Infectious Diseases (3R01AI141003-03S1, R01AI158013)
  • National Institute of Dental and Craniofacial Research
  • Chan Zuckerberg Biohub Investigator Fund
  • MIDAS Coordination Center (MIDASNI2020)
  • Instituto Nacional de Ciência e Tecnologia Midas (MIDASNI2020)
  • Damon-Runyon Cancer Research Fellowship
Topics
  • Health Sciences › Medicine › Infectious Diseases › SARS-CoV-2 and COVID-19 Research
  • Health Sciences › Medicine › Infectious Diseases › SARS-CoV-2 detection and testing
  • Health Sciences › Medicine › Infectious Diseases › COVID-19 Clinical Research Studies
Keywords
SARS-CoV-2, COVID-19 pandemic, antibody response, serological surveillance, disease severity, seroreversion, neutralizing antibodies, spike protein antibodies, nucleocapsid antibodies, mild infection, assay variability

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: SARS-CoV-2 antibody magnitude and detectability are driven by disease severity, timing, and assay. https://app.cochat.ai/papers/10.1126/sciadv.abh3409
The paper
Peluso, M. J., Takahashi, S., Hakim, J., Kelly, J. D., Torres, L., Iyer, N. S., Turcios, K., Janson, O., Munter, S. E., Thanh, C., Donatelli, J., Nixon, C. C., Hoh, R., Tai, V., Fehrman, E. A., Hernandez, Y., Spinelli, M. A., Gandhi, M., Palafox, M.-A., . . . Greenhouse, B. (2021). SARS-CoV-2 antibody magnitude and detectability are driven by disease severity, timing, and assay. Science Advances. https://doi.org/10.1126/sciadv.abh3409

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.