Immunological Insights: A Multicenter Longitudinal Study on Humoral Response to COVID-19 Vaccines in Greece

: Vaccination has emerged as the most effective tool in the battle against COVID-19. To optimize vaccination protocols, a deeper understanding of the immune response to vaccination, including influential factors and its duration, is essential. This study aimed to assess the humoral response in vaccinated individuals with or without prior SARS-CoV-2 infection. A prospective observational study was conducted across 14 private healthcare structures in Greece. Anti-spike IgG titers were measured at different timepoints following the initial vaccination and booster doses of the BNT162b2, mRNA-1273, ChAdOx1 nCoV-19, and Ad26.COV2.S vaccines. A total of 505 participants were included in the first phase, evaluating the humoral response after the initial vaccination, and 311 participants were involved in the second phase, assessing the effects of booster vaccination. All vaccines elicited high anti-S IgG titers initially, followed by a subsequent decline that was addressed by the booster vaccination. The humoral response was sustained up to one year after the booster vaccination. mRNA vaccines induced higher anti-S IgG titers compared to vector vaccines, with mRNA-1273 eliciting higher titers than BNT162b2. Vaccination resulted in higher antibody titers than natural infection alone; however, convalescent patients who received vaccination had significantly higher anti-S IgG titers compared to those who received the booster vaccine without previous SARS-CoV-2 infection. Lower antibody titers were observed in men and older patients (>51.5 years old), as well as smokers, although the decline rate was lower in these subgroups. These results underscore the importance of booster doses and reveal the potential influence of age, gender, smoking habits, and vaccine type on varying humoral responses. Long-term monitoring of antibody persistence, evaluation of cellular immune responses, and assessment of vaccine efficacy against emerging variants should be considered to enhance our understanding of immunity dynamics and inform vaccine development and deployment strategies.


Introduction
The emergence and rapid global spread of COVID-19 (Coronavirus Disease of 2019) since late 2019 has posed an unprecedented Public Health challenge worldwide.Prompting the World Health Organization (WHO) to declare a global pandemic in March 2020 [1] this viral outbreak has necessitated urgent efforts to develop effective vaccines and treatments against the causative agent, SARS-CoV-2.Undoubtedly, vaccination has emerged as the cornerstone of preventive strategies to curb morbidity and mortality from SARS-CoV-2 infection [2][3][4][5].Antibodies binding to the receptor-binding domain of the SARS-CoV-2 spike protein prevent attachment to the host cell and neutralize the virus [6].
There are currently several types of vaccines against SARS-CoV-2 in the market or in pre-clinical development: pioneering messenger ribonucleic acid (mRNA) vaccines, adjuvanted recombinant protein vaccines, DNA vaccines, viral vector vaccines, bacterial vector vaccines, and whole virus vaccines, including inactivated vaccines and live attenuated vaccines.The WHO keeps in development a database of vaccines which is continuously updated [7].Most COVID-19 vaccines have the large surface spike protein as an antigenic target, a protein binding to the angiotensin-converting enzyme 2 (ACE2) receptor on host cells [4,5], and whole virus vaccines have multiple antigenic targets, as is expected [7].
But one size does not fit all, and not all vaccines are equally immunogenic to all patients.By analyzing cohorts, we learned that vaccine-induced protection wanes over time and that mRNA vaccines prevent more hospitalizations and severe infections from SARS-CoV-2 compared to vector vaccines [8].Moreover, some limited observational data have shown that the mRNA-1273 vaccine might be slightly more effective than the BNT162b2, but it is not certain whether this difference is clinically relevant [9][10][11].
Serological investigations aim to detect virus-specific antibodies, which serve as indicators of past infection or vaccination [12].The analysis of vaccine trials provides evidence that the key indicators of immune protection against symptomatic infection are the levels of binding and neutralizing antibodies (NAbs) targeting the spike protein and its receptor-binding domain (RBD) and that higher levels of these antibodies are associated with greater vaccine efficacy [13,14].In such studies, it was shown that mRNA vaccines offer better humoral response compared to vector vaccines [15,16].Moreover, factors that are known to influence humoral response to SARS-CoV-2 vaccination are patient characteristics such as age, sex [17,18], previous SARS-CoV-2 infection [19], comorbidities such as hypertension [17], obesity [18], diabetes [20], hemodialysis [21,22], hematological conditions [23,24] and/or treatments used for such conditions such as rituximab or bruton kinase inhibitors [24], solid organ transplantation [25], other immunosuppressive conditions such as Acquired Immune Deficiency Syndrome [26], and several immunosuppressive treatments [27].
We hereby report humoral response against SARS-CoV-2 through antibody measurements after the initial vaccination scheme and booster vaccination with the BNT162b2, the mRNA-1273, the ChAdOx1 nCoV-19, and the Ad26.COV2.S vaccines in individuals with or without prior history of natural SARS-CoV-2 infection.

Study Design
This is a prospective observational study.Fourteen private laboratories from all over Greece (5 laboratories from the region of Attica, 2 laboratories from Evia, 5 from Peloponnese, 1 from Epirus and 1 from Western Macedonia) participated in the study.To enroll patients, we utilized convenience sampling.Consequently, we included all patients who sought laboratory tests at the aforementioned private laboratories, those undergoing treatment at hemodialysis units collaborating with these laboratories, or healthcare workers associated with those laboratories who agreed to participate.Questionnaires were filled out by all participants upon enrollment and blood samples were collected.The study consisted of two phases.In the first phase, which took place from 03.2021 to 09.2021, we measured anti-SARS-CoV-2 anti-spike antibodies at 3 weeks, 3 months, 6 months, and 9 months after the initial vaccination scheme.During the second phase of the study, conducted from 10.2021 to 11.2022, we measured anti-SARS-CoV-2 anti-spike antibodies at 3, 6, and 12 months after the booster vaccination.The vaccines used were the BNT162b2, the mRNA-1273, the ChAdOx1 nCoV-19 and the Ad26.COV2.S vaccines, according to the prevailing vaccination plan in Greece at that time.Convalescent patients were identified either through history of proven SARS-CoV-2 infection or by measurement of anti-nucleocapsid antibodies.Of note, patients with COVID-19 were outpatients with mild illness.

Study Population
All adults over the age of 18, regardless of their economic and social characteristics, who reside and/or work in Greece and are registered in the files of private primary health care structures were proposed enrollment in the study.An informative brochure was given to all patients and informed consent was obtained.Exclusion criteria were refusal to participate, patients lost to follow-up (≤2 visits), and contraindication for blood sampling.All participants completed a questionnaire that included demographic characteristics, somatometric measurements, lifestyle factors, medical history, previous history of COVID-19, contact with a COVID-19 patient, and vaccination history.

Serum Sampling
Serum specimens were collected and stored at −80 • C until testing.As mentioned, blood samples were collected at 3 weeks, 3 months, 6 months, and 9 months after the initial vaccination scheme for the first phase of the study and 3, 6, and 12 months after booster vaccination for the second phase of the study.

Quantitative Methods
Anti-S (anti-spike protein) IgG testing: The SARS-CoV-2 IgG II Quant Assay by Abbott Diagnostics was used, and a CMIA was used for the quantitative detection of IgG antibodies against the spike receptor-binding domain (RBD) of SARS-CoV-2 [29].

Definitions
The initial vaccination scheme includes the first two doses at a 3-4 week interval for the BNT162b2 and mRNA-1273 vaccines and at a 4-12 week interval for the ChAdOx1 nCoV-19 vaccines or the first dose in the case of the Ad26.COV2.S vaccine.The booster vaccination defines the third dose for the BNT162b2, mRNA-1273, and ChAdOx1 nCoV-19 vaccines or the second dose in the case of initial vaccination with the Ad26.COV2.S vaccine.For the first phase of the study, the participants were categorized based on information obtained from questionnaires and serological test outcomes and were grouped in a convalescent group, a vaccinated group without history of infection, and a vaccinated group with history of infection.The convalescent group was defined either by history of proven infection, as seen in patients' questionnaires or by the presence of anti-nucleocapsid antibodies.For the second phase of the study (10.2021-11.2022),vaccinated patients were categorized in two groups: booster vaccination with history of infection and booster vaccination without history of infection.The mRNA vaccine group defines patients having received mRNA vaccines (BNT162b2, mRNA-1273).The vector vaccine group defines patients having received vector vaccines (ChAdOx1 nCoV-19, Ad26.COV2.S).Normal weight participants were those with a BMI of 18.5-25 kg/m 2 .Overweight participants were participants with a BMI between 25 kg/m 2 and 29.9 kg/m 2 .Obese participants were participants with a BMI ≥ 30 kg/m 2 .According to smoking habits, participants were divided into current smokers, ex-smokers, and non-smokers.Smokers were patients with ≥10 cigarettes per day.Medical history refers to all diagnoses filled in the questionnaire by patients themselves and includes cardiovascular diseases, hypertension, dyslipidemias, diabetes, chronic pulmonary disease (e.g., asthma, chronic obstructive pulmonary disease), chronic kidney disease (including hemodialysis).

Statistical Analysis
All analyses were performed with STATA 13.0.The Kolmogorov-Smirnov criterion was used to test the normality of the distributions of quantitative variables.Mean and standard deviation (SD) were used for variables that followed a normal distribution.For variables that did not follow a normal distribution, additional measures such as the median and interquartile range (IQR) were used.Absolute (N) and relative (%) frequencies were used to describe qualitative variables.To compare antibodies between two groups, the nonparametric Mann-Whitney test was employed.For comparisons between mRNA and nonvaccinated individuals, the difference in antibodies was investigated using linear regression models, taking into account the age and gender of the participants.For comparison between more than two groups, the non-parametric Kruskal-Wallis test was used.The Bonferroni correction was used to counteract the type I error due to multiple comparisons.The Wilcoxon signed-rank test was used to compare antibody measurements at 6 months after the 2nd and 3rd doses.Spearman's correlation coefficient (rho) was used to examine the relationship between two quantitative variables.The change in antibodies over time was assessed using mixed linear models, from which regression coefficients (β) and their standard errors (SE) were obtained.Moreover, the above method was used to estimate if the degree of change in the studied parameters over time differed according to their demographic characteristics.The mixed linear models were performed using logarithmic transformations.The significance level was set with two-sided p < 0.05.

Ethics
Our study strictly adhered to the principles set forth in the Declaration of Helsinki, ensuring the ethical conduct of research and the protection of participants' rights and wellbeing.The study was approved from an independent Ethics Committee of the University of West Attica N 13317/22-02-2021.

Results
In the first phase of the study, which took place from March 2021 to November 2021, the analysis included a final sample of 505 participants.Among these patients, 78 (15.4%) were in the convalescent group, 427 were in the vaccinated group without a history of infection, and 30 were in the vaccinated group with a history of infection.Within the vaccinated group, 306 (60.6%) patients received the BNT162b2 vaccine, 62 (12.3%) patients received the mRNA-1273 vaccine, 48 (9.5%) patients received the ChAdOx1 nCoV-19 vaccine, and 11 (2.2%) patients received the Ad26.COV2.S vaccine.Overall, 229 participants were male (45.35%), and 276 were female (54.65%).
Three weeks after completing the first two doses (or a single dose in the case of the Ad26.COV2.S vaccine), we observed a peak in anti-S IgG titers with all four vaccines.Moreover, we observed a decline in antibody titers over time (see Figure 1).The mean titer after 3 weeks was 10,907 (SD 10,246.6)[median 8039.5, IQR (2859.7-15,578.1)]IU/mL, which decreased to 3148.5 (SD 3834.5)[median 2101.5 IQR (896.5-3939.1)]IU/mL after 3 months, then further dropped to 1242.4 (SD 1893.5)[median 812.7 IQR (375.6-1373.7)]IU/mL at 6 months and 1207.9 (SD 2580.9)[median 733.7 IQR (385.9-1051.4)]IU/mL at 9 months.The rate of decrease was significantly lower in patients over 51.5 years old, although these patients had lower baseline antibodies at 3 weeks (see Figure 2).
Regarding the comparison between different types of vaccines, mRNA vaccines (BNT162b2, mRNA-1273) induced higher antibody titers compared to vector vaccines (ChAdOx1 nCoV- 3) IU/mL for the vector vaccine group, p < 0.001)] (see Figure 3).The antibody titers induced by mRNA vaccines were further compared to those of convalescent participants (either by history of proven infection, as noted in patients' questionnaires or by presence of anti-nucleocapsid antibodies).The mRNA vaccinated group presented significantly higher antibody titers compared to the convalescent group at 3 weeks [8334.7 (3044.1-16,032)IU/mL vs. 2068.6(743.6-4771)IU/mL, p < 0.001] and 3 months [2275.8(999.6-4149.9)IU/mL vs. 2036.3(2599.2) IU/mL, p < 0.001].We found no significant difference in antibody titers between mRNA vaccinated and convalescent participants at 6 and 9 months after vaccination.
In the second phase of the study, which took place from October 2021 to November 2022, only 311 patients participated.Among them, 269 (86.5%) patients received the booster vaccine, and 42 (13.5%)patients both received the booster vaccine and were convalescent.We note a peak in the humoral response after the booster vaccination [6 months after booster 5156 (2874.7-15,758.1)IU/mL vs. 811.3(354.5-1427.9)IU/mL after initial vaccination scheme, p < 0.001].The anti-S IgG were significantly higher in patients who were convalescent and received vaccination compared to those who received the booster vaccine and had never experienced SARS-CoV-2 infection [11,668.7 (5646.4-28,732.2) IU/mL vs. 8715.5 (5106.3-17,112)IU/mL, p < 0.001] at 3 months.This difference was not statistically significant at 6 months.A similarly strong humoral response was also seen in disease after vaccination.Contrarily to humoral response after the initial vaccination scheme, we did not observe a difference in anti-S IgG between women and men after the booster vaccination.In the second phase of the study, which took place from October 2021 to November 2022, only 311 patients participated.Among them, 269 (86.5%) patients received the booster vaccine, and 42 (13.5%)patients both received the booster vaccine and were convalescent.We note a peak in the humoral response after the booster vaccination [6 months after booster 5156 (2874.7-15,758.1)IU/mL vs. 811.3(354.5-1427.9)IU/mL after initial vaccination scheme, p < 0.001].The anti-S IgG were significantly higher in patients who were convalescent and received vaccination compared to those who received the booster vaccine and had never experienced SARS-CoV-2 infection [11,668.7 (5646.4-28,732.2) IU/mL vs. 8715.5(5106.[3][4][5][6][7][8][9][10][11][12][13][14][15][16][17]112) IU/mL, p < 0.001] at 3 months.This difference was not statistically significant at 6 months.A similarly strong humoral response was also seen in disease after vaccination.Contrarily to humoral response after the initial vaccination scheme, we did not observe a difference in anti-S IgG between women and men after the booster vaccination.
There was no statistically significant difference in anti-S IgG according to BMI.Even though smoking habits influenced the humoral response after the initial vaccination scheme, no statistically significant difference was found after the booster vaccination.Similarly, anti-S IgG after the booster vaccination did not differ significantly at 3 and 6 months for patients with no medical history (see Table 1).Again, mRNA vaccines induced higher antibody titers compared to vector vaccines at 3 months [mRNA booster: 8838.7 (5244.3-17,944.9)IU/mL vs. 2963.9(1216.1-4464.8)IU/mL for the vector booster, p < 0.001] and 6 months [mRNA vaccine group: 5205.5 (3037.4-12,516.4)IU/mL vs. 832.3(659.1-1477)IU/mL for the vector booster, p < 0.003] after booster vaccination.
There was no statistically significant difference in anti-S IgG according to BMI.Even though smoking habits influenced the humoral response after the initial vaccination scheme, no statistically significant difference was found after the booster vaccination.Similarly, anti-S IgG after the booster vaccination did not differ significantly at 3 and 6 months for patients with no medical history (see Table 1).Lastly, prolonged persistence of anti-S IgG is demonstrated 12 months after the booster dose with mean values 3592.2 (SD 3344.1)[median 2789.4 (IQR 1038.6-4116.4)]IU/mL, at levels higher than those measured 6 months after the initial vaccination scheme with mean values 2383.2 (SD 4978.5)[median 925.8 (IQR 345.6-1474.1)IU/mL].

Discussion
Our findings demonstrate a peak in antibody titers 3 weeks after the initial vaccination scheme with all four vaccines, but we observed a decline in antibody titers over time.This finding is expected and in accordance with previous studies [15,[30][31][32].
On the other hand, our results show a prolonged persistence of anti-S IgG 12 months after the booster dose at levels higher than those measured 6 months after the initial vaccination scheme.This finding supports the fact that the booster vaccination plays a crucial role in sustaining protective immunity against SARS-CoV-2 and suggests that periodic booster vaccinations may be necessary to maintain optimal immunity, especially in the face of emerging SARS-CoV-2 variants.However, the optimal vaccination schedule is still unknown.To determine the best vaccination protocol, it is essential to understand how long the immune response lasts.This requires regular measurements of anti-S IgG and/or neutralizing antibodies (NAbs) to guide the planning of the vaccination schedule.In line with other studies [31,33] we note an increase in anti-S IgG titers after the booster vaccination [31,[34][35][36].A recent proof-of-concept study provides a framework to determine the optimal time of a booster dose at the individual level based on the half-life of neutralization titers [37].
During the first phase of the study, the comparison between the convalescent group and the mRNA vaccinated group without a history of infection indicates that vaccination induces higher antibody titers than natural infection alone.This finding supports the recommendation of vaccination for individuals previously infected with SARS-CoV-2, as it provides an additional boost to the naturally acquired humoral response.In the second phase of the study, the anti-S IgG were significantly higher in patients who were convalescent and received vaccination compared to those who received the booster vaccine and had never experienced SARS-CoV-2 infection.A similarly strong humoral response is also seen in disease after vaccination.An explanation for that could be that recovery from infection could induce a more complex and complete immune response directed against various antigens of the virus other than the spike protein.Together with previous data, our findings lend support to the notion of utilizing previous infection history as a reference when contemplating future vaccination protocols [38][39][40][41].
As expected, and as previously described [15,16,[42][43][44], our study showed that mRNA vaccines induce higher anti-S IgG titers compared to vector vaccines, with the mRNA-1273 vaccine offering an increased humoral response compared to BNT162b2 [45].While it may be challenging to directly correlate antibody responses with clinical outcomes [46], the clinical data from vaccine cohorts consistently align on this matter.Indeed, observational data available show that mRNA vaccines offer greater protection against hospitalizations and severe SARS-CoV-2 infections when compared to vector vaccines [8], and it has also been reported that the mRNA-1273 vaccine might be slightly more effective [9] although the number needed to vaccinate was >290 in a study [11].This observation underscores the importance of considering vaccine platform selection in terms of optimizing humoral responses and potential vaccine effectiveness.
Our study reveals variations in antibody response and kinetics based on age.This could be attributed to immunosenescence.Notably, our research demonstrated that patients over 51.5 years old exhibited lower baseline antibody levels when compared to their younger counterparts.This observation aligns with findings from other studies [42,47,48].Inversely, a study of healthcare workers did not find a statistically significant relationship between age and antibody titers [49].In our study, even if older patients had lower initial antibody titers, the rate of decrease was significantly lower suggesting that continuous measurements are needed in order to establish the optimal vaccination protocol in this group of patients.
A disparity in humoral response was also observed according to sex.Men showed lower anti-S IgG titers compared to women at 3 weeks, 3 months, and 6 months, but the rate of decrease over time was significantly lower in men compared to women.Variations in humoral response with other vaccines such as influenza have already been observed but are not well understood and do not seem to be only attributed to sex hormones [45,47,50,51].Likewise, other studies on COVID-19 vaccines show greater humoral responses in women [45,50,51].Pharmacokinetic and pharmacodynamic parameters which could influence vaccine absorption and distribution and might be different between men and women could also play a role.
BMI is known to affect the development of post-vaccine humoral response [49].We observed lower anti-S IgG titers in obese patients, but the results were not statistically significant, which implies that this could be further investigated in larger scale studies.A prospective longitudinal case control study showed that more than half of individuals (55%) with severe obesity had unquantifiable titers of neutralizing antibody against SARS-CoV-2 compared to 12% of individuals with normal BMI.Antibody positivity is also lower in obese patients in a population antibody surveillance study of 212.102 individuals by Ward et al. [47].Other studies suggest similar results in NAbs or anti-S IgG antibodies [50,52,53].
In the present study, smokers exhibited lower antibody titers in response to vaccination compared to non-smokers and ex-smokers at 3 and 6 months The immune response is acknowledged to be impacted by smoking habits through poorly understood mechanisms [54].Watanabe et al. report a strong association between smoking and lower antibody titers after COVID-19 vaccination [18].Comparable observations have been yielded from other studies as well [45,49,52,54].Further research is needed to better understand the underlying mechanisms and implications of smoking on vaccine-induced immune responses, and studies should also focus on clinical outcomes such as hospitalization and mortality rates after vaccination in smokers.This is even more interesting given the fact that smoking has been associated with severe COVID-19 [55].
It is noteworthy that humoral response disparities in gender, age, and smoking habits were not observed in the post-booster phase.This could be explained by the smaller population size in the second phase of our study, but curiously, even in the RENNAISSANCE study, no differences in age and gender were observed in post-booster titers [45].
Our study has a few limitations.Firstly, there was an important drop-out rate in the second phase of the study, with only 311 patients participating, and this could affect the study's results, making it an important limitation to note.Additionally, due to the nature of the study, which is observational, there is a notable imbalance between different vaccines groups.Moreover, since data on medical history were collected from questionnaires, no specific information on underlying disease was available.Thus, medical history was binarily defined by presence or absence of underlying disease and no information on immunosuppression or other comorbidities such as hemodialysis was available, which could have impacted the humoral response as expected and as shown in other studies [10,43,56].Therefore, even if we have the information that 39 hemodialysis patients participated in the study, they are grouped among patients with "positive medical history".Similarly, smoking habits were arbitrarily noted in questionnaires by patients.Hence, smoking definitions are based on participants' responses, and no specific information on quantity of smoking is available.Lastly, clinical outcomes were not evaluated either.
Still, correlates of protection to SARS-CoV-2 are undetermined [13,46].Hence, relying solely on the presence of antibody titers is just one aspect of a complex immune response.The development of vaccines to prevent SARS-CoV-2 infection has mainly relied on the induction of NAbs to the spike protein of SARS-CoV-2, but there is growing evidence that T-cell immune response can contribute to protection as well.Other studies have already tried to focus on cellular response and its correlation to vaccine-induced protection showing promising results [57,58].Future studies should be multifaceted and focus on the correlation between cellular and humoral immune response and include neutralization assays and clinical outcomes as well.
The results of this study contribute to the growing body of knowledge on COVID-19 vaccination and immunity.We showed variation in antibody response and kinetics based on age, gender, and smoking habits as well as that mRNA vaccines induce higher antibody titers compared to vector vaccines, and we highlight the importance of booster doses on maintaining higher antibody titers.Long-term monitoring of antibody persistence, evaluation of cellular immune responses, and assessment of vaccine efficacy against emerging variants should be considered to further understand the dynamics of immunity and guide vaccine development and deployment strategies in order to ensure long-term protection against SARS-CoV-2 and mitigate the risk of future outbreaks.

Figure 1 .
Figure 1.Mean titers of anti-S IgG over time after initial vaccination scheme for all vaccines.

Figure 2 .
Figure 2. Comparison of antibody kinetics between age groups (>51.5 years old and <51.5 years old).

Figure 1 .
Figure 1.Mean titers of anti-S IgG over time after initial vaccination scheme for all vaccines.

Figure 2 .
Figure 2. Comparison of antibody kinetics between age groups (>51.5 years old and <51.5 years old).

Figure 2 .
Figure 2. Comparison of antibody kinetics between age groups (>51.5 years old and <51.5 years old).

Figure 3 .
Figure 3.Comparison of antibody kinetics between mRNA and vector vaccines.

Figure 3 .
Figure 3.Comparison of antibody kinetics between mRNA and vector vaccines.

Figure 4 .
Figure 4. Comparison of antibody kinetics between mRNA vaccines.

Figure 4 .
Figure 4. Comparison of antibody kinetics between mRNA vaccines.

Table 1 .
Antibody kinetics according to medical history.

Table 1 .
Antibody kinetics according to medical history.