*2.3. Comparison of Income and Race between Infected and Vaccinated Groups*

Demographic comparisons were evaluated for age, income, and race in infected (*n* = 25) and vaccinated (*n* = 17) individuals from the February 2021 cohort. In both groups, the average age was approximately 50. The median household income of participants was significantly lower in the infected group than that of the vaccinated group (Supplementary Figure S2a). Three students were removed from this comparison since their income data were felt to be reflective of their guardians' income as their zip codes were out of state. The race distribution between infected and vaccinated groups in the February cohort revealed similar percentages of African American and White in the infected group but a lower percentage of African American in the vaccinated group (Supplementary Figure S2b). These data are consistent with other reports demonstrating the disproportionate effects of the pandemic and vaccine accessibility on lower-income and African American groups in PA as well as nationwide US [16–19].

#### **3. Discussion**

The ELISAs used in this study were shown to have a high degree of sensitivity and specificity. These assays are semi-quantitative while most commercially available antibody detection kits only offer qualitative results. The ability to perform functional, neutralizing assays of antibody activity remains limited to BSL-3 laboratory settings. Therefore, an assay such as the RBD ELISA, which is significantly correlated with neutralization titer, has utility for measuring virus-specific activity and could possibly be used to define a correlate of immune protection in clinical and vaccine studies. We observed a 4.5-fold increase in community immunity by RBD ELISA between Fall of 2020 and February of 2021 and were able to demonstrate that this increase was driven both by infection and vaccination.

The correlations between RBD, N, and neutralization were particularly interesting. When a serum sample had a high RBD titer (≥24,300), which indicated strong immune responses elicited by either infection or vaccination, that sample had the ability to neutralize the virus regardless of the N titer. Out of 12 samples with RBD titers at 900, only 2 were positive in the neutralization assay, both of which also had a positive N titer. Two samples with negative RBD titers but positive N titers (one at 900 and the other at 2700) were evaluated for neutralization and neither of them neutralized SARS-CoV-2 virus, suggesting these were false positives or possibly they represent cross reactivity with another coronavirus [20]. When a sample had an RBD titer between 900 and 24,300, it appeared more likely to neutralize the virus if it also had a positive N titer. A total of 12 out of 16 samples from infected individuals with an RBD titer at 2700 or 8100 demonstrated neutralization, whereas only 2 out of 8 samples from vaccinated individuals did. This could have several implications. First, the fact that 4 out of 16 samples with positive RBD and N titers failed to neutralize the virus was consistent with previous findings that people who had contracted COVID-19 could sometimes be reinfected [21,22]. Secondly, the complete vaccination history was not available for all samples, therefore it is possible that some vaccinated cases were not fully vaccinated, meaning that they either received one dose or received the second dose within 14 days at the time of sample collection. Lastly, two infected samples with RBD titer = 2700 and N titer = 300 demonstrated high neutralization titer at 160, suggesting that other factors may also play a role in the generation of neutralizing activity. For example, binding sites on the ACE2 receptor and the binding affinity of an antibody have both been shown to be critical for neutralization potential [23–25]. The SARS-CoV-2 immune response is different in infected versus vaccinated individuals [26]. This could result in a greater breath or alter the class switching and/or affinity maturation of S-specific antibodies in infected versus vaccinated individuals and is consistent with our observations.

The entire population was naïve to SARS-CoV-2 infection when this virus emerged. However, the pandemic has affected people from diverse social and economic backgrounds differently. By comparing income and race distribution between infected and vaccinated groups, we found that people with lower incomes were more likely to have been infected

than vaccinated and people of color were less likely to have been vaccinated at the time of sampling. The February 2021 cohort of specimens were obtained when PA restricted vaccination to health care workers, residents/care staff in skilled nursing facilities, individuals ages 65 and older and those ages 16–64 with certain underlying medical conditions [27]. Restrictions on access to vaccination could have influenced these results.

This study has several limitations. Although people of both genders and most age groups and races were included, it had more women, seniors (aged 60–69 and 70–79), and African Americans as compared to the representative distribution in Allegheny County, PA (Supplementary Table S2). In addition, whereas children under 15 make up 15.6% of the population, this study was focused on adults. Finally, patient information regarding previous COVID-19 diagnosis or vaccination was limited to data accessible by chart abstraction for the majority of participants. A major strength of this study is the use of two assays that permits classification of individuals as either vaccinated or infected. With many of the vaccine platforms containing the S antigen, the need for a non-vaccine antigen for community serosurveys of infection is of increased importance. In addition, estimating local seroprevalence using a small cohort of samples can be a valuable tool for quickly assessing the degree of immunity in a community and can inform public health decisions during a pandemic.

#### **4. Materials and Methods**
