A single dose ChAdOx1 nCoV-19 vaccine elicits high antibody responses in individuals with prior SARS-CoV-2 infection comparable to that of double dose vaccinated SARS-CoV-2 infection naïve individuals

Background A single dose COVID-19 vaccines, mostly mRNA-based vaccines, are shown to induce robust antibody responses in individuals who were previously infected with SARS-CoV-2, suggesting the sufficiency of a single dose to those individuals. However, these important data are limited to developed nations and lacking in resource-limited countries, like Ethiopia. Methods We compared receptor-binding domain (RBD)-specific IgG antibodies in 40 SARS-CoV-2 naïve participants and 25 participants previously infected with SARS-CoV-2, who received two doses of ChAdOx1 nCoV-19 vaccine. We measured the antibody response in post-vaccination blood samples from both groups of participants collected at four different post-vaccination time points: 8- and 12-weeks after each dose of the vaccine administration using an in-house developed ELISA. Results We observed a high level of anti-RBD IgG antibodies titers 8-weeks after a single dose administration (16/27; 59.3%) among naïve participants, albeit dropped significantly (p<0.05) two months later, suggesting the protective immunity elicited by the first dose ChAdOx1 nCoV-19 vaccine will likely last for a minimum of three months. However, as expected, a significant (p<0.001) increase in the level of anti-RBD IgG antibodies titers was observed after the second dose administration in all naïve participants. By contrast, the ChAdOx1 nCoV-19 vaccine-induced anti-RBD IgG antibody titers produced by the P.I participants at 8- to 12-weeks post-single dose vaccination were found to be similar to the antibody titers seen after a two-dose vaccination course among infection-naïve participants and showed no significant (p>0.05) increment following the second dose administration. Conclusion Taken together, our findings show that a single ChAdOx1 nCoV-19 dose in previously SARS-CoV-2 infected individuals elicits similar antibody responses to that of double dose vaccinated naïve individuals. Age and sex were not associated with the level of vaccine-elicited immune responses in both individuals with and without prior SARS-CoV-2 infection. Further studies are required to assess the need for a booster dose to extend the duration and amplitude of the specific protective immune response in Ethiopia settings, especially following the Omicron pandemic.


Results
We observed a high level of anti-RBD IgG antibodies titers 8-weeks after a single dose administration (16/27; 59.3%) among naïve participants, albeit dropped signi cantly (p<0.05) two months later, suggesting the protective immunity elicited by the rst dose ChAdOx1 nCoV-19 vaccine will likely last for a minimum of three months. However, as expected, a signi cant (p<0.001) increase in the level of anti-RBD IgG antibodies titers was observed after the second dose administration in all naïve participants. By contrast, the ChAdOx1 nCoV-19 vaccine-induced anti-RBD IgG antibody titers produced by the P.I participants at 8-to 12-weeks post-single dose vaccination were found to be similar to the antibody titers seen after a two-dose vaccination course among infection-naïve participants and showed no signi cant (p>0.05) increment following the second dose administration.

Conclusion
Taken together, our ndings show that a single ChAdOx1 nCoV-19 dose in previously SARS-CoV-2 infected individuals elicits similar antibody responses to that of double dose vaccinated naïve individuals. Age and sex were not associated with the level of vaccine-elicited immune responses in both individuals with and without prior SARS-CoV-2 infection. Further studies are required to assess the need for a booster dose to extend the duration and amplitude of the speci c protective immune response in Ethiopia settings, especially following the Omicron pandemic.

Background
Page 4/17 COVID-19 continues to be a major public health concern, causing severe illness and deaths in Ethiopia and the rest of the world alike. Mass vaccination against SARS-CoV-2 is the most effective public health intervention to protect against morbidity and mortality related to SARS-CoV-2 infection [1]. Safe, e cacious, and licensed COVID-19 vaccines, including ChAdOx1 nCoV-19 (AZD1222; Oxford-AstraZeneca) are available [2][3][4][5], despite being challenged by the recurrent emergence of new SARS-CoV-2 variants. Real-world vaccine e cacy studies from developed countries have shown that the current vaccines are able to generate effective humoral and cellular immunity, albeit differential responses are observed between vaccine-induced immunity and hybrid (vaccine-induced immunity combined with natural infection) immunity [6,7]. Several correlates of protection studies have demonstrated that higher antibody titers are associated with decreased risk of subsequent symptomatic SARS-CoV-2 infection [8][9][10][11][12], and several studies from developed countries have revealed the rapid waning of antibody levels among SARS-COV-2 infection naïve vaccine recipients compared to those individuals with hybrid immunity [7,[13][14][15][16]. Despite the importance of immune durability data for guiding national vaccination strategies, there is a dearth of studies from Ethiopia and other African countries looking at more locally relevant populations.
The ChAdOx1 nCoV-19 vaccine utilizes a replication-de cient adenoviral vector that induces expression of SARS-CoV-2 spike (S) protein in host cells, particularly in the skeletal muscle [17]. Vaccinated individuals subsequently generate antibodies against the spike protein, including those that target the receptor-binding domain (RBD), which contains many neutralizing epitopes. However, they do not generate antibodies against other SARS-CoV-2 structural and non-structural proteins, such as nucleocapsid (N) [18]. Studies have shown a strong correlation between anti-RBD IgG titers and SARS-CoV-2 neutralizing titers [15]. Therefore, in resource-limited countries, it is advantageous to use anti-RBD IgG testing as a proxy for virus neutralization to assess the protection offered by the ChAdOx1 nCoV-19 vaccine.
As part of the strategy to evaluate the Ethiopian national COVID-19 response through vaccination, we established a longitudinal cohort of healthcare professionals working at the Armauer Hansen Research

Study Design and Participants
We conducted a longitudinal prospective study constituting healthcare professionals from AHRI, who were also one of the priority target recipients of the ChAdOx1 nCoV-19 vaccine. Vaccination was offered through the Ethiopian Ministry of Health national COVID-19 vaccination campaign. In this analysis, only participants who were vaccinated with the ChAdOx1 nCoV-19 vaccine were included. The study protocol was reviewed and approved by the AHRI/ALERT Ethics Review Committee (PO/32/20) and only voluntary participants and who were able to give written informed consent were included in the study. About 5ml of venous blood was collected from a total of 65 (n=26 female and n=39 male) participants, aged 24-59 years (mean of 38.1 ± standard deviation of 8.36) before vaccination and four subsequent postvaccination follow-ups ( Figure 1) to monitor humoral responses to the vaccine. The rst (F1) and second follow-ups (F2) were conducted at 8-and 12-weeks after the rst dose, respectively, whereas the third (F3) and fourth (F4) follow-ups were done at 4 and 8 weeks after the second dose, respectively ( Figure 1). All participants received their rst dose between March 23, 2021 and March 31, 2021. The interval between the rst and the second doses was 12 weeks (range 83-97 days). Participants were then strati ed into two groups based on previous exposure to SARS-CoV-2 before vaccination and hereafter denoted as naïve and P.I. (Figure 1). The two groups: naïve (n=40) and P.I. (n=25) contained similar distributions of age and sex (Figure1 and Table 1). Table 1 summarizes the demographics of each group of study participants. In addition, participants completed a questionnaire at each visit regarding their history of RT-PCR con rmed SARS-CoV-2 infection.

ELISA Methods:
Prior to analyzing the presence of anti-RBD IgG antibodies, each serum sample was treated with Triton X-100 at the nal concentrations of 1.0% and incubated at room temperature (RT) for 30 minutes. This procedure was performed aiming to reduce risk from any potential virus in serum [19]. Detection of anti-RBD IgG antibodies in the sera was done using a validated in-house ELISA as described previously [20].
To determine the end titer in seropositive serum samples, a two-fold serial dilution starting at 1: 200 in a 96-well ELISA plate was done. The end titer was de ned as a serum dilution at which the observed optical density (OD) at 450 nm reads matched to the OD 450 readout for pre-COVID-19 serum sample diluted at 1:200 and included in each ELISA run as a negative control. Inter-assay variability was normalized by including a convalescent serum in each run of titration ELISA.
Since the magnitude and durability of SARS-CoV-2 antibody response has been shown to positively correlate with the severity level of COVID-19 [14,21] and yet we did not have clinical records of our P. I. participants. Thus, a convalescent sera panel (CSP; n=15) was collected from COVID-19 recovered individuals, three-months after being discharged from the Saint Paulo's Millennium College Hospital in 2020 (before the rollout of the vaccines) were included in the anti-RBD IgG antibodies titers analysis.
Based on the level of anti-RBD IgG titers in these CSP as a reference, we inferred whether our P.I. participants have had symptomatic or asymptomatic SARS-CoV-2 infection.

Statistical analysis
GraphPad Prism version 8.0 for Windows (GraphPad Software, La Jolla California USA) was used for statistical analyses. We measured anti-RBD IgG end titers ranging from 1:200 to 1:51200. Graphs were plotted using log10-transformed anti-RBD IgG end titers values. Table 1 comprised the calculated geometric means (GM) titers. To compare the mean differences of anti-RBD IgG-antibodies titers across each time point of serum collection, we used the unpaired non-parametrical t-test with p <0.05 =*; p< 0.01=**; p<0.001=***; p<0.0001****). Because some participants dropped out and others joined, the number of blood samples provided after the rst dose varied among participants enrolled at baseline ( Figure 1 and Table 1). Participants who did not provide any post-vaccine samples were excluded from the analysis.

Results
Post-vaccination seropositivity

Comparison of anti-RBD IgG antibody titers
We rst compared the pre-vaccination anti-RBD IgG titers level in P.I. participants with that of the titers in CSP. A signi cantly (p-value <0.0001) higher levels of anti-RBD IgG titers were observed among recovered COVID-19 patients 3-months after natural infection than in those P.I. participants (Fig. 2a).
We also compared the levels of anti-RBD IgG titers between naïve and P.I. participants across the four post-vaccination follow-ups. P.I. had a four-fold increase at 8-weeks post-rst dose compared to their prevaccination anti-RBD IgG antibodies levels (p=0.0004; Table 1). We also observed higher anti-RBD IgG titer levels in naïve participants at 8-weeks post-rst dose, yet rapidly (30%) declined at 12-weeks postrst dose (p<0.05; Figure 2c). By contrast, high and sustainable post-rst dose antibody levels were observed in P.I. participants and were found to be comparable to those naïve participants post-second dose. This trend remained consistent when results were strati ed by sex and age ( Figure 3). Figure 2c, we observed a signi cant (p-value <0.01) increase in the level of anti-RBD IgG antibodies in naïve participants 8-weeks post-second dose compared to the level of the titer that was observed 12-weeks post-rst dose. On contrary, we did not observe a statistically signi cant (p-value > 0.5) difference in the level of anti-RBD IgG antibodies titers in P.I. participants from those observed postsingle dose (Figure 2d). Unlike the decline shown 12-weeks post-rst dose, we did not see a signi cant (pvalue >0.5) decline in mean anti-RBD IgG antibodies titers 12-weeks post-second dose in both naïve and P.I. participants (Table 1 and Figure 2b).

Discussion
Immune protection following either vaccination or natural infection with SARS-CoV-2 decreases overtime [22]. Although the minimum antibody titer that correlates with protection has not yet been established, a decreased antibody titer is shown to be associated with an increased risk of subsequent symptomatic SARS-CoV-2 infection [8, [10][11][12]. In the present study, immunologically naïve participants had relatively comparable anti-SARS-CoV-2 RBD IgG responses at 8-weeks post-rst dose ChAdOx1 nCoV-19 with those of participants with previous exposure to SARS-CoV-2. This nding is consistent with other studies on mRNA-based vaccines [7,13]. Notably, at three months post-single-dose, the level of anti-RBD IgG antibodies elicited by a single dose ChAdOx1 nCoV-19 vaccine dropped signi cantly in individuals who were not previously infected with SARS-CoV-2. This is consistent with the suggestion made by the UK Joint Committee on Vaccination and Immunization (JCVI) -that is protective immunity elicited by the rst dose ChAdOx1 nCoV-19 vaccine will likely last for a duration of 12 weeks [23]. As reported elsewhere [24], we also noted that four of our SARS-CoV-2 infection naïve participants experienced a delay in generating anti-RBD IG antibodies. Unexpectedly, we also noted seroconversion in 2 males P.I. participants aged 30 and 42 years who had higher anti-RBD IgG antibody titers 8-weeks post-rst dose, indicating evidence of a rare event of rapid waning of humoral response in single-dose vaccinated individuals with prior SARS-CoV-2 infection. This could be due to critical medical conditions such as immunosuppression, though in our study, such medical conditions were not systematically recorded. Similarly, seroreversion was observed after receiving two hepatitis A vaccination in HIV-positive patients [25].
As expected, individuals who naturally contracted SARS-CoV-2 prior to vaccination developed a more rapid and sustained response to the ChAdOx1 vaccine than immunologically naïve individuals. The vaccine-induced anti-RBD IgG antibody titers produced by participants with prior SARS-CoV-2 infection at three months after a single dose vaccination appeared to be comparable to the antibody titers levels seen after a two-doses for infection-naïve participants. Similar ndings have been previously reported for the ChAdOx1 nCoV- 19 [26] and BNT162b2 vaccines [7,14,15]. Similarly, SARS-CoV-2 immunologically naïve individuals developed relatively higher and durable anti-RBB IgG antibody titers after the second dose vaccination. However, we did not observe a statistically signi cant difference in antibody response after second dose between participants with and without evidence of previous SARS-CoV-2 infection. Our ndings, along with other studies [6, 15, 27, 28] suggest a single vaccine dose in previously infected individuals elicits similar antibody responses to that of double dose vaccination.
Participants with SARS-CoV-2 antibodies at baseline before the rst vaccine injection, regardless of their sex and age (ranged 21-59), have developed strong anti-RBD IgG antibodies to the COVID-19 vaccine with no statistically signi cant variability between the rst and the second dose. Given the age range (21-59 years with an average of 38.1 years) of our participants, this nding not surprising; however, an agedependent decreasing pattern of anti-RBD IgG antibodies titers was reported across similar age groups to our study [29,30]. On the other hand, the absence of statistically signi cant antibody titers difference between male and female participants is surprising and inconsistent with previous reports [27,30] yet agrees with the nding reported by Wheeler et al. [31] and Lee et al. [32].
Our ndings are the rst evidence from Ethiopia supporting the superiority of "hybrid immunity" in eliciting a strong immunity against SARS-CoV-2 infection, comparable to that of two doses of the ChAdOx1 nCoV-19 vaccine in infection naïve individuals. Recent studies have demonstrated that sera from mRNA-vaccinated individuals with prior infection provide broader cross-neutralizing antibodies against several SARS-COV-2 variants, including Delta variants [33]. Consistent with this, another recent study revealed the mechanism by which hybrid immunity improves B cells and antibodies against SARS-CoV-2 variants [34]. However, a study revealed that the anti-spike antibody titers of BNT162b2 recipients were remarkably higher than those of ChAdOx1 nCoV-19 recipients [13]. Additionally, recent studies showed increased risk of SARS-CoV-2 Omicron infection in both vaccinated and previously infected individuals through evasion of vaccine-or infection-induced immune response and suggested the need of the rapidly developing new, Omicron variant-speci c vaccine (refs). It also remains to be determined if the hybrid immunity elicited by the ChAdOx1 nCoV-19 vaccine will effectively protect vaccinees from the subsequent infections with different SARS-CoV-2 spike variants. Thus, further studies that aim to nd a correlation between the level of ChAdOx1 nCoV-19 or any vaccine-induced antibody titer and clinical outcome are required.

Conclusions
Our ndings demonstrate that administrating only one dose vaccine to individuals previously infected with SARS-CoV-2 infection could elicit antibody response comparbale to that of two doses of ChAdOx1 nCoV-19 vaccinated SARS-CoV-2 naïve infection and an effective strategy to stretch the supply of vaccines. Age and sex were not associated with the level of vaccine and hybrid immune responses.
However, our ndigs should not be translated to the Omicron variant as our study was not designed to determine the effectivenes of vaccine-elicited antibodies against Omicron. Thus further studies that represents participants with SARS-CoV-2immunologically naïve and blue color (P.I.) represents participants with likely previous SARS-CoV-2 infection; BL =baseline or prevaccination; F1= 8-weeks after the rst dose; F2=12-weeks after the rst dose; F3 =8-weeks after the second dose; and F4 =8-weeks after the second dose. The rst dose and second dose were given to participants at BL and F3, respectively prior to blood sample collection.

Figure 2
Analysis of ChAdOx1 nCoV-19 vaccine-induced antibody response in naïve and previously infected (P.I.) participants pro le at different time points. Comparison of anti-RBD IgG titers pro le (a) between P.I. participants (red dots) before they received vaccination and convalescent serum panel (CSP: blue dotes); (b) between naïve (blue dots) and P.I. (red dots) participants across different time points before and after vaccination; among naïve participants (c) and (d)P. I. participants across different points. CSP=convalescent sera panel included as a reference to indirectly assess whether P.I. participants had previously either asymptomatic or symptomatic SARS-CoV-2 infection. Each colored dot corresponds to an individual participant. Horizontal bars represent mean with 95% CI of anti-RBD IgG titer levels (transformed to Log 10 value) within the indicated groups. The broken line denotes the assay detection limit. Unpaired non-parametrical t-test with p <0.05 =*; p< 0.01=**; p<0.001=***; p<0.0001=****, ns=nonsigni cant) was used compare the mean differences of anti-RBD IgG-antibodies titers across each time point of serum collection. BL=baseline or prevaccination; F1= 8-weeks after the rst dose; F2=12-weeks after the rst dose; F3 =8-weeks after the second dose; and F4 =8-weeks after the second dose. The rst dose and second dose were given to participants at BL and F3, respectively prior to blood sample collection. Figure 3