Safety, Tolerability and Immunogenicity of Oxford-AstraZeneca ChAdOx1 Vaccine Among Polish School-Teachers

Methods Teachers were invited for serological testing ≥ 8 weeks after second vaccination. Quantitative post-vaccination anti-spike antibody responses were measured using the Abbott SARS-CoV-2 IgG II Quant assay (detection threshold: ≥ 7.1 BAU/ml). Multivariable logistic regression methods were used to identify predictors of immunogenicity.

Mild or moderate local and systemic reactions to the ChAdOx1 nCoV-19 vaccination were reported in numerous randomized control trials (RCTs) [7,[13][14][15][16]. For instance, in the report of the phase 2 of a singleblind RCT on the ChAdOx1 nCoV-19 vaccine delivered by Ramasamy et al. local and systemic reactions were more common in participants given the vaccine than in those given the control, meningococcal vaccine [13]. However, there is presently limited information available outside RCTs regarding the safety and tolerability of ChAdOx1 nCoV-19 vaccine [17,18].
The assessment of vaccine-generated immune responses to SARS-CoV-2 spike antigens, has mostly focused on the development of antibodies targeting the S1 domain of the viral spike protein. Anti-spike antibody titres, associated with neutralizing activity, provide a potential surrogate marker of protection [14,[19][20][21]. A key bene t of vaccine regimens is that anti-S IgG titres are higher than for natural infection. More than 99% of boosted participants of the RCT had neutralizing antibody responses by 14 days after the second dose [13]. Eyre et al. also reported that vaccination with the ChAdOx1 nCoV-19 vaccine led to detectable anti-spike antibodies in nearly all adult HCWs [20]. Immunogenicity data for the vaccine, reported from the community trials, are rather scant.
Understanding the time-dependent dynamics of post-vaccine anti-spike antibody measurements, and assessing how they differ between individuals e.g., by age, gender, body mass index (BMI), comorbidities etc. is also increasingly important; however, not much information is currently available regarding ChAdOx1 nCoV-19 vaccine [7,13,18,20].
Considering the important need of real-life data on safety, tolerability and immunogenicity data for the ChAdOx1 nCoV-19 vaccine and being concerned that such data have not been assessed yet in Polish population, we decided to assess reported adverse events following immunization, as well as to measure anti-spike IgG responses in teachers following two vaccine doses. How responses vary between those with and without previous evidence of infection, as well as other determinants of SARS-CoV-2 anti-spike IgG responses have also been evaluated.

Population and Setting
Post-vaccine antibody responses were studied between June-July 2021 in consecutive teachers recruited from primary, secondary and high-schools through local teacher networks. Convenience sampling design was adapted to recruit teachers with a goal of 200 participants. All schools were located in the capitals of 2 Polish provinces: Zielona Gora and Szczecin. Inclusion criteria were as follows: employed as a teacher; being immunized with 2 doses of the ChAdOx1 nCoV-19 vaccine at least 8 weeks before the survey. Teachers contacted the research team directly via a dedicated phone line if they wanted to participate.
Following initial contact with the study team, participants were then informed about the phlebotomy time and the health care facility address. At the phlebotomy time point, the participant information sheet was given to each teacher and written consent was then obtained.

Study instrument
The short questionnaire was developed by the authors after intensive literature research [7,13,17,18,20] and then given to the participants before the phlebotomy. The questions concerned sociodemographic data: age, gender, school location, and core health risk factors (body-mass index [BMI], smoking status, and presence of comorbidities, including type 2 diabetes, cancer, heart/lung/ kidney disease). Participants were then asked to record the date of ChAdOx1 nCoV-19 vaccine administration and dosing interval; this was then checked by a research team member in the national COVID-19 vaccination data base. Teachers were also asked whether they had experienced adverse effects, including both systemic and local effects. Systemic side-effects included symptoms such as: fatigue, malaise, headache, chills, fever, arthralgia, myalgia, nausea and diarrhea; local side-effects included injection site pain, tenderness, redness and swelling [7,13]. Participants could also tick "no symptoms". Data on previous SARS-CoV-2 infection were available for the research team from the national patients' data base after logging in. Due to the fact that some patients did not test themselves for SARS-CoV-2 infection despite medical history of COVID-19 like symptoms and/or contacts with infected patients, the information of previous infection reported by participants in the study ques-tionnaires was taken into account while assessing individuals previously infected with SARS-CoV-2.
Laboratory assays Blood samples were collected by a quali ed nurse/physician. Samples (5 mL) were centrifugated (15 min/4500 r. p. m.), stored at 4-8°C and then transported to the Synevo laboratory in Cracow, Poland where they were tested. Brie y, post vaccination anti-spike IgG responses were assessed using the Abbott SARS-CoV-2 IgG II Quant antibody assay, an automated, two-step chemiluminescent micro-particle immunoassay (CMIA), targeting the spike receptor binding domain (RBD).The assay is used for the qualitative and quantitative determination of IgG antibodies to SARS-CoV-2 in human serum and plasma on the ARCHITECT i System. The assay cut-off is ≥7.1 binding antibody units/ml (BAU/ml) reported by the manufacturer. The sensitivity (based on ≥14-day post-positive reverse transcription-PCR samples) and speci city of the Abbott anti-nucleocapsid assay has been previously evaluated as 98.3% (90.6-100.0%) and 99.5% (97.1-100%), respectively [22].
Each study participant was given a code number, placed both on the questionnaire and on a test tube. From July 11th 2021, teachers were able to obtain information about their vaccination serological test results.

Vaccination immunogenicity assessment
On the basis of the results obtained after sero-testing, relevant parameters were calculated to assess the immunogenicity of the ChAdOx1-S vaccine. The following parameters were evaluated: GMT (geometric mean titers) calculated at ≥ 2 weeks after vaccination, PR (protection rate) -the proportion of subjects with an IgG antibody titer ≥7.1 BAU/ml.

Statistical analysis
Data were analyzed using a customized program STATISTI-CA PL, Version 12.5 (StatSoft, Cracow, Poland, 2016). Categorical data were presented as frequencies with percentages and continuous data as means and ranges. Teachers were grouped into those with evidence of prior infection ( i.e. those who reported having any positive anti-spike or anti-nucleocapsid antibody test or positive PCR prior to rst/second vaccination and those who reported having COVID-19 without con rmation by any diagnostic test) and those without (including participants with no previous serology or PCR testing).
Categorical variables were compared by the chi square test, while continuous variables were compared by a Student's t-test. Correlations were calculated using standard Pearson's correlation. The occurrence of adverse effects was studied for the rst and the second dose of ChAdOx1 n CoV-19 vaccine. To assess determinants of the occurrence of adverse effects we used the following strata: age (≤55 years vs >55 years), gender, teacher status regarding previous SARS-CoV-2 infection (binary variable), obesity (BMI <30 kg/m 2 vs ≥30 kg/m 2 ), and comorbidities (binary variable, with/without comorbidities).
Proportions of anti-spike positive teachers were estimated by checking anti-spike IgG antibody titer at least 8 weeks post-second vaccination. The primary endpoints, anti-spike IgG antibody titer and seroprotection rate, were analyzed. Positive anti-spike IgG antibody titers were analyzed using geometric mean titer (GMT). Multiple linear regression was applied to determine the predictors of immuno-genicity measured by anti-S antibody titers ≥8 weeks post-second vaccination. We modelled quantitative IgG antibody titres using several multi-variable logistic regression models, and tting separate models by prior infection status. All models were reduced by the use of the stepwise backward elimination method [23]. Non standardized regression coe cients in the regression model were used to evaluate any changes in the model. Regression results are presented together with 95% con dence intervals (CIs). A p-value was statistically signi cant if ≤0.05.

Results
Overall, 200 teachers were invited to participate of whom 8 were disquali ed due to the misinformation regarding their vaccination status. Finally, 192 teachers were tested for anti-spike SARS-CoV-2 IgG. Their demographic details and the SARS-CoV-2 infection status prior to immunization are given in Table 1.

Predictors of immunogenicity
Multiple logistic-regression analysis regarding an association of immunogenicity (measured by the level of anti-spike IgG titers) with selected variables revealed that previous infection with SARS-CoV-2 and longer dose interval were independent positive predictors of higher immunologic response (p<0.0001 and p=0.01 respectively); Table 3.

Results overview
In the present survey we sought to better understand the safety, tolerability and immunogenicity of the ChAdOx1 nCoV-19 vaccine among Polish teachers. Interestingly, two out of three teachers reported they would change the product for another vaccine -mostly a mRNA vaccine -if there was such an opportunity; this clearly highlights the necessity of this study.
The vast majority of teachers receiving ChAdOx1 nCoV-19 experienced more than one side effect at the same time. Adverse events were signi cantly more common after the rst dose and were mild in severity.
Injection site pain was the most common type of local side effects. Common systemic symptoms were: feverishness, followed by headache, malaise fatigue and chills. Reactions were signi cantly less common in males and in older teachers.
In this cohort of teachers, predominantly healthy adults of working age, all developed a positive anti-spike IgG antibody test by ≥ 8 weeks post vaccination. Mean GMTs remained relatively high in all age groups.
Previous infection with SARS-CoV-2 and longer dose interval were both positive predictors of higher immunologic response.

Median dosing interval
Median dosing interval in this study was 69 days, with the range of 25-111 days; only two teachers reported a shortened interval between ChAdOx1 doses. This means that the vast majority of the study participants did not follow the Polish government recommendation, announced on May 17th 2021, which endorsed 35 days as a between-doses interval [5]. It is worth emphasizing that this regulation was in opposite to the United Kingdom (UK) and several other countries policy to employ an 'extended interval' vaccine regimen in which the booster dose of ChAdOx1 vaccine is delayed for 10 to 12 weeks following the rst dose [6,24].

Frequencies and intensity of adverse reactions
We found that any systemic adverse effects affected almost three in four participants and local side effects -more than every second teacher. Furthermore, adverse effects observed in this study were similar in nature to those previously reported (feeling feverish, headache, malaise, chills, injection-site tenderness and pain) [13,16,17]. Severe systemic allergic reactions following immunization, such as anaphylactic shock, were not reported by study participants, however, the small sample size reduced the chance to detect such adverse events. Recent data imply an incidence of anaphylactic shock as very infrequent (1 per 200,000 -1 per million doses) [25].
We found fewer adverse events among the studied teachers after the boost vaccination than after the prime vaccination. Similar results have been seen in other studies which assessed the ChAdOx1 nCoV-19 vaccine safety [13,16,17]. Of note, some other studies found that individuals vaccinated with the ChAdOx1 nCoV-19 vaccine were more likely to experience systemic side-effects than those who had been given the BNT162b2 vaccine [26].
Side-effects were more prevalent in teachers aged <50 years than older teachers. Our results provide evidence to support other studies, both the RCTs and conducted in the community, of lower occurrence of side-effects in older than in younger individuals [13,16,17,26]. We also found that adverse effects were less common in men than women which is consistent with previous studies [17].
Local and systemic side effects were reported to be higher in individuals previously infected with SARS-CoV-2 than in those without known past infection for BNT162b2 and ChAdOx1 nCoV-19 COVID-19 vaccines [17,[27][28][29]. Given evidence from these studies, we investigated the extent to which previous SARS-CoV-2 infection was associated with reports of adverse effects. No consistent difference in occurrence of systemic/local adverse effects between individuals who reported a previous infection and those who did not might be due to the small sample size, as well as to the information bias; although in 55% of teachers SARS-CoV-2 infection was con rmed by an adequate test, in the rest of cases the analyses relied on participants reports regarding previous infection.

Anti-spike antibody titres
We found 100% seroconversion rate among the study population after receiving two doses of ChAdOx1 vaccine, however, 4.2% of participants were low responders. As such, these ndings show that vaccination is highly effective for induction of spike-speci c immune responses in the working-age population, such as teachers. Some other studies reported similar rates of low responses among participants, however, in contrast with our results, this was independently associated with several longterm health conditions [18].
Antibody levels after two vaccine doses were signi cantly higher in individuals with prior SARS-CoV-2 infection compared to those achieved by teachers without infection. Previous infection with SARS-CoV-2 was the biggest positive predictor of the magnitude of quantitative antibody response post second vaccine dose, with median readings 2.6-fold higher with previous infection compared to without infection. This was also reported by other authors who explored this trend regarding ChAdOx1 nCoV-19 vaccine [7,18], as well as other vaccines [18, [30][31][32]. For instance, a study conducted among Israeli healthcare workers (21 days post-dose 1 of the BNT162b2 mRNA COVID-19 vaccine) found that those with prior infection had antibody titres one magnitude order higher than naïve individuals [29]. These ndings may suggest prioritization on uninfected persons in regions where COVID-19 vaccine-sparing strategies are required [31,32]; such prioritization may also refer to the third vaccination in previously infected healthy individuals.
Concerning our second main nding, longer dose interval led to a greater immune response to ChAdOx1 nCoV-19 vaccine. Other studies con rmed that dosing interval is one of the most signi cant factor in determining the e cacy of this vaccine [7,8]. For instance, an analysis of four RCTs, conducted in South Africa, Brazil and the UK, found that e cacy was higher if the booster ChAdOx1 nCoV-19 vaccine dose was received 8-11 weeks after the rst; furthermore, it was increased if participants received the booster dose more than 11 weeks after the rst [7]. This incremental VE with increasing prime-boost intervals positively correlated with GMTs of anti-SARS-CoV2 spike IgG binding antibody. A similar boost to antibody responses was seen with a longer duration in some other studies on the BNT162b2 COVID-19 vaccine [21,33].
Among demographic factors, older age has been repeatedly reported to associate with reduced antibody responses after COVID-19 vaccination due to age-related decline in immune functions [18,34]. No statistically signi cant differences in the antibody response between age groups were found in this survey, possibly due to a de ned study population, i.e. teachers in a relatively young age (median of 50.5 years). However, a relatively low correlation was found between age (r 0.144, p=0.046) with anti-S antibody titres. Some studies found [7,13,21] that two vaccine doses achieved high responses across all age groups, which supports our ndings. By contrast, another two-dose mRNA vaccine candidates have shown immunogenicity in older adults, but absolute neutralizing antibody responses in adults aged 65-85 years were lower than in those aged 18-55 years [35]. A two-dose inactivated virus vaccine has also shown lower absolute neutralizing antibody titres in adults aged 60 years and older than in adults aged 18-59 years [36]. Additional studies on post-vaccination antibody responses in different age groups would be of value.
In our survey of teachers no association was found between female gender and better ChAdOx1 vaccine antibody response. This may be due to the small sample size and relatively small representation of males in the sample, as well as to the fact that the study population consisted of the middle-age individuals. Some previous studies found that between-gender differences in antibody responses become more marked above 60 years of age [20]. Our ndings are consistent with some previous observations [21], however, other authors which conducted larger population studies reported that females generate stronger humoral immunity than males [18,37].
Although many long-term health conditions, such as rheumatoid arthritis, chronic liver disease, type 2 diabetes, obesity, asthma, hypertension, as well as taking corticosteroids and immunosuppressants were independently associated with low responses [18,38], we did not detect any signi cant differences in antibody responses regarding teachers with comorbidities and healthy individuals. This may be attributable to the relatively low fraction of teachers reporting any long-term health conditions.

Limitations
Potential limitations did exist in this study. The sample size of our study population was relatively small, which could possibly hinder some of the associations. Second, our cohort was predominately females.
This echoes the demographic pro le of Polish teachers with males comprising only 17.8% [39]. However, for the gender-discrepancy reason, the cohort may not be representative for the entire population of working adults. Third, self-reported data were used for variables such as BMI and comorbidities, as well as (in 45% of participants) regarding the previous SARS-CoV-2 infection. This can introduce information bias, together with misclassi cation. The time of evaluation ranged between 1 and 4 weeks following the second immunization, introducing time as a possible bias. However, when the evaluations were controlled for this factor, they still yielded the same results. Furthermore, we assessed the intensity of the overall side-effects following the ChAdOx1 nCoV-19 immunization, no matter which dose had been taken. Although the vast majority of participants reported adverse events as more intense after the rst vaccination than after the second one, the frequency assessed in this study could have been slightly higher than while assessed separately for each dose of the vaccine. Furthermore, T-cell response was not assessed in this study. Although the correlation between antibody response and vaccine e cacy is high, which suggests that the neutralizing antibody response is important, T-cell responses may contribute to protection from COVID-19 even in the presence of lower neutralizing antibody titers [20]. Finally, the study design (a cross-sectional study) does not allow an inference of causality [17].

Conclusions
While robust conclusions from the survey results are limited due to the small sample size, this report demonstrates that the ChAdOx1 nCoV-19 vaccine appears safe, well tolerated and immunogenic in the studied population of teachers. Although mild adverse effects affected the majority of teachers, more commonly after the rst dose, immunization led to detectable anti-spike antibodies in all teachers.
Signi cantly higher immunogenicity was observed in participants who reported previous SARS-CoV-2 infection. The latter nding suggests that in immunocompetent vaccine recipients with an evidence of previous infection a delay regarding the second dose could be considered any time when careful management in the use of vaccine resources is needed. Further large-scale studies are urgently required to better assess the duration of antibody responses regarding the ChAdOx1 nCoV-19 vaccine.
The timing between priming and boosting has emerged as a critical aspect of the COVID-19 vaccination programs. Our study justi es the longer dose interval as an important factor to enhance higher antibody response post ChAdOx1 nCoV-19 vaccination.
The results may help to remove the odium, commonly expressed by the Polish teachers, of ChAdOx1 nCoV-19 vaccine being "the worse" product. The ndings may also encourage general practitioners, nurses and other healthcare providers involved in immunization process, to promote the ChAdOx1 nCoV-19 vaccination among other patients. This is particularly important as the ChAdOx1 nCoV-19 vaccine is likely to be one of the least expensive among of all the currently authorized Covid-19 vaccines. Notably, further investigation on the effectiveness of the ChAdOx1 nCoV-19 vaccine need to be made in the context of the currently circulating variants of interest and variants of concern and the possible evolution of other SARS-CoV-2 lineages.   Antibody responses among 192 teachers following second vaccination with ChAdOx1 by prior infection.