Identification of natural SARS-CoV-2 infection in seroprevalence studies among vaccinated populations

Demmer, R. T.; Baumgartner, B.; Wiggen, T. D.; Ulrich, A. K.; Strickland, A. J.; Naumchik, B. M.; Bohn, B.; Walsh, S.; Smith, S.; Kline, S.; Stovitz, S. D.; Yendell, S.; Beebe, T.; Hedberg, C.

Importance: Identification of SARS-CoV-2 infection via antibody assays is important for monitoring natural infection rates. Most antibody assays cannot distinguish natural infection from vaccination. Objective: To assess the accuracy of a nucleocapsid-containing assay in identifying natural infection among vaccinated individuals. Design: A longitudinal cohort comprised of healthcare workers (HCW) in the Minneapolis/St. Paul metropolitan area was enrolled. Two rounds of seroprevalence studies separated by one month were conducted from 11/2020-1/2021. Capillary blood from round 1 and 2 was tested for IgG antibodies against SARS-CoV-2 spike proteins with a qualitative chemiluminescent ELISA (spike-only assay). In a subsample of participants (n=82) at round 2, a second assay was performed that measured IgGs reactive to SARS-CoV-2 nucleocapsid protein (nucleocapsid-containing assay). Round 1 biospecimen collections occurred prior to vaccination in all participants. Vaccination status at round 2 was determined via self-report. Setting: The Minneapolis/St. Paul, Minnesota metropolitan area. Participants: HCW age 18-80 years. Exposures: Round 1 recent SARS-CoV-2 infection assessed via a spike-only assay and participant self-report. Outcomes: Round 2 SARS-CoV-2 infection assessed via the nucleocapsid-containing assay. Area under the curve (AUC) was computed to determine the discriminatory ability of round 2 IgG reactivity to nucleocapsid for identification of recent infection determined during round 1. Results: Participants had a mean age of 40 (range=23-66) years, 83% were female, 46% reported vaccination prior to the round 2 testing. Round 1 seroprevalence was 9.5%. Among those not recently infected, when comparing vaccinated vs. unvaccinated individuals, elevated levels of spike 1 (p<0.001) and spike 2 (p=0.01) were observed while nucleocapsid levels were not statistically significantly different (p=0.90). Among all participants, nucleocapsid response predicted recent infection with an AUC(95%CI) of 0.93(0.88,0.99). Among individuals vaccinated >10 days prior to antibody testing, the specificity of the nucleocapsid-containing assay was 92% and while the specificity of the spike-only assay was 0%. Conclusions and Relevance: An IgG assay identifying reactivity to nucleocapsid protein is an accurate predictor of natural infection among vaccinated individuals while a spike-only assay performed poorly. In the era of SARS-CoV-2 vaccination, seroprevalence studies monitoring natural infection will require assays that do not rely on spike-protein response alone.