In this study, carried out in a nursing home, the humoral immunogenicity elicited by the BNT162b2 vaccine and the durability of such response along the time were conditioned by age and age-related factors. While the magnitude of the initial response to the vaccine was independently associated with the inflammatory status of the participants, the durability of such response in the long-term (up to 8 months), was only predicted by the previous levels of thymosin-α1, a thymic-derived immunomodulatory hormone.

Immunity against emerging viruses and de novo vaccine responses depend on the efficacy of mounting primary responses, which is known to be impaired in older adults, mainly due to immunosenescence [12]. Thus, the benefits of vaccination in the prevention of infectious diseases are limited in older people, who normally show defective humoral responses to several vaccines [4], such as Influenza [5, 6], Hepatitis B virus [17], and SARS-CoV2 [1, 18,19,20,21]. In fact, despite the considerable high efficacy of the available SARS-CoV-2 mRNA vaccines, aged people, even vaccinated, maintain increased mortality rates due to COVID-19 than younger subjects [22].

Preclinical studies in animal models, as well as vaccine efficacy studies and the early use of convalescent plasma, suggest that neutralizing antibodies are the best correlators of protection against the more severe outcomes of COVID-19 [1, 4, 8, 9, 23,24,25,26]. In our study, although the seroconversion rate was 100% for all vaccinees, the magnitude of the initial humoral response was significantly lower in the oldest participants, which was in accordance with previous reports [1, 18]. We also observed a negative association between anti-S IgG titers 1 month after vaccination with the age of the participants, a correlation that was maintained 4 months later (data not shown). In addition to the diminished magnitude of the initial response among older adults, current evidence indicates the increasing risk of SARS-CoV-2 symptomatic infection with time due to the waning of humoral immunity [1, 10]. We found that anti-S IgG titers rapidly decayed early after vaccination in all participants without exception, but especially among aged vaccinees, in accordance with previous reports [19]. On the other hand, the loss of anti-S IgG titers did not correlate with the age of the participants. Consequently, the identification of reliable age-related factors associated with the immunogenicity and durability of the response over time is mandatory and could help to improve personalized vaccination/boosting strategies.

It is known that the distribution of immune cell subsets affects the immune responses against vaccine antigens [27,28,29]. Moreover, age-dependent differences in T and B lymphocytes, as well as in NK populations distribution and functions have been well-defined [30]. Thus, their relationship with antibody responses following vaccination and the durability of the humoral response over time could provide valuable information for understanding protective long-term immunity against COVID-19. As expected, in our cohort, the distribution of the peripheral immune cell subset depended on age. Thus, older participants showed lower levels of total lymphocytes, CD3, CD4, CD8, and B lymphocytes, but higher levels of NK cells, an immune profile tightly related to immunosenescence [30,31,32,33]. Interestingly, all immune subsets analyzed correlated positively to the magnitude of the humoral response to the vaccine, except NK cells, which were not associated with such a response.

Another age-related immune defect is inflammaging, a low grade chronic persistent inflammatory phenotype found in older populations [34], that contributes to the decline in the immune response and has been shown to inhibit immune responses to vaccination [17, 35, 36]. Accordingly, older participants of our study showed significantly higher levels of inflammatory markers, including homocysteine, β2-microglobulin, and D-dimers, and a trend to higher levels of hsCRP. All of these markers correlated negatively with the magnitude of the response, but only the levels of hsCRP correlated with the short-term loss of anti-S antibody titers. These results point out that pharmacological strategies aiming at blocking baseline inflammation can be potentially used to boost COVID-19 vaccine responses, but would not be probably enough to avoid the intense waning of humoral protection.

The aging of the immune system is proposed to impact COVID-19 vaccination efficiency [37]. One of the main factors underlying immunosenescence is age-related thymic involution [15]. Hence, we hypothesized that deficient thymic function related to aging might impact COVID-19 vaccination efficiency and should also be considered a key player in aged populations. In our study, older participants showed a reduced thymic output, determined by a lower sj/β-TRECs ratio in the periphery. Interestingly, the RTL of immune cells from the older group tended to be shorter than that from the other groups, suggesting an age-dependent shift from naïve to memory phenotype induced by homeostatic T-cell proliferation to compensate for the diminished T-cell thymic output in older participants. Interestingly, the sj/β-TRECs ratio tended to correlate positively with the magnitude of the response to the vaccine, also being associated with the short-term loss of anti-S IgG titers. In accordance, diminished responsiveness to vaccination against other viruses as yellow fever virus has been previously associated with lower thymus activity [38]. Remarkably, thymus activity does not rely only on cellular output, but also on its production of immunomodulatory hormones. We found that although the levels of thymosin-α1 at T1 did not correlate with anti-S IgG titers at T1, they did at T8. Moreover, and despite there being no differences in the levels of thymosin-α1 between age groups, we found that older people with a higher magnitude of response at T1 had higher levels of thymosin- α1. Furthermore, levels of thymosin- α1 predicted the short- and long-term loss of anti-S IgG titers, being the unique variable independently associated in the multivariable analysis with the loss of antibodies at both the short- and the long-term. Our results are of relevance since thymosin-α1, a peptide with immunomodulatory properties that induces differentiation of B and T lymphocytes has already been used in several clinical trials (phase I/II/III/IV) and research settings, such as different types of cancer, sepsis, and infectious diseases, including COVID-19 [39,40,41,42,43,44]. Notably, its use as an adjuvant to the Influenza and hepatitis B vaccines improved the immunogenicity and durability of the humoral responses, showing to be safe and well tolerated [39, 40, 42]. Future research addressing the potential benefit of using this immunomodulator as an adjuvant of COVID-19 vaccination would be desirable since it could also improve the durability of responses. Also remarkable, the potential role of the thymus in the severity of COVID-19 has been discussed [45,46,47,48,49], and thymosin-α1 has also been used as a treatment for severe COVID-19, although with controversial results [44, 50, 51].

Our study has several limitations. We present an exploratory study with a relatively small sample size but fully representative of a nursing home, allowing us a close monitor follow-up. On the other hand, all participants from the study were men, and consequently, our results might not be extrapolated to cohorts including women. However, this approximation allowed us to avoid the potential influence of gender in age-related factors measured and its relation with the magnitude and loss of antibodies over time. Finally, the vaccination of residents of nursing homes was considered a priority, and the urgent vaccination of this especially vulnerable population did not allow us to collect samples before vaccinations. Nevertheless, our results support and extend previous observations from COVID-19 vaccine responses following vaccination in older people and raise novel potential approaches to optimize COVID vaccination/boosting protocols in this vulnerable population.