Seroprevalence of SARS-CoV-2 - Specific antibodies among Victorian blood donors

Routine surveillance does not capture all people infected with SARS-CoV-2 because some are asymptomatic, not diagnosed, or not reported; therefore, estimating the proportion of the population with SARS-CoV-2 antibodies (i.e., seroprevalence) can improve understanding of the population-level incidence of infection.

This report uses de-identified blood donor specimens to examine trends in infection- and vaccine-induced SARS-CoV-2 seroprevalence before and throughout the first Omicron epidemic wave in Victoria.

Specimens were collected from Victorian blood product donations received during each of 4 time intervals: November 26–30, 2021 (n=1,000), December 24–28, 2021 (n=1,000), January 27–31, 2022 (n=1,000), and 24 February–March 2, 2022 (n=1,099) and tested using the Roche Elecsys anti-SARS-CoV-2 anti-spike and anti-nucleocapsid protein immunoassays. Crude seroprevalence estimates and 95% confidence intervals (CIs) were calculated.

The presence of anti-spike antibodies indicates prior vaccination against SARS-CoV-2 infection and/or natural infection. The presence of anti-nucleocapsid protein antibodies indicates previous natural infection, most likely within the recent past.

Prevalence of anti-spike antibody was very high (>98.0%) across all 4 time points, with little variation by age group and sex.

Prevalence of anti-nucleocapsid seroprevalence was very low in November (0.2%), and December (0.7%), increasing to 9.5% in January and 22.5% in February. For both the January and February time points, anti-nucleocapsid seroprevalence was highest among donors aged 18–29 years (22.2% and 34.2%, respectively), decreasing steadily with increasing age-group to 0.0% in January and 11.1% in February, among donors aged 70­­–89 years.

Seropositivity for anti-spike was modestly higher than in the general population based on vaccine coverage rates. This may reflect the presence of anti-spike antibodies induced by vaccination and those induced following infection, as it is not possible to distinguish between them. It may also be due to the behaviour of donors who may be more likely to be vaccinated.

Seroprevalence for anti-nucleocapsid across time points was consistent with the epidemiology of notified cases, with the Omicron wave in Victoria beginning in late December. Of note, however, the November and December anti-nucleocapsid prevalence estimates were lower than cumulative case notifications for that period in people 18¬89 years (1.6% and 1.9%, respectively), while the January and February time points were higher (7.7% and 9.8%, respectively).

Limited evidence suggests that as a marker of recent infection, anti-nucleocapsid antibodies may have lower sensitivity in vaccinated compared with unvaccinated persons who become infected. Additional work is underway to refine sensitivity estimates for the Roche anti-nucleocapsid immunoassay in this context and inform an analysis approach that accounts for a reduction in test sensitivity.