Background: Both SARS-CoV-2 mRNA-based vaccines [BNT162b2 (Pfizer-BioNTech) and mRNA-1273 (Moderna)] have shown high efficacy, with very modest side effects in limiting transmission of SARS-CoV-2 and in preventing the severe COVID-19 disease, characterized by a worrying high occupation of intensive care units (ICU), high frequency of intubation and ultimately high mortality rate. At the INT, in Naples, only the BNT162b2 / Pfizer vaccine has been administered to cancer patients and healthcare professionals aged 16 and over.In the present study, the antibody response levels and their decline were monitored in an interval of 6-9 months after vaccine administration in the two different cohorts of workers of the INT - IRCCS "Fondazione Pascale" Cancer Center (Naples, Italy): the group of individuals previously infected with SARS-CoV-2 and vaccinated with a single dose; and that of individuals negative for previous exposure to SARS-CoV-2 vaccinated with two doses 21 days apart.Methods: Specific anti-RBD (receptor-binding domain) titers against trimeric spike glycoprotein (S) of SARS-CoV-2 by Roche Elecsys Anti-SARS-CoV-2 S ECLIA immunoassay were determined in serum samples of 27 healthcare workers with a previously documented history of SARS-CoV-2 infection and 123 healthcare workers without, during antibody titers’ monitoring. Moreover, geometric mean titers (GMT) and relative fold changes (FC) were calculated.Results: Bimodal titer decline was observed in both previously infected and uninfected SARS-CoV-2 subjects. A first rapid decline was followed by a progressive slow decline in the 6/9 month-period before the further vaccine boost. The trend was explained by 2 different mathematical models, exponential and power function, the latter revealing as predictive of antibody titer decline either in infected or in not previously infected ones. The value of the prolonged lower vaccine titer was about 1 log below in the 6/9-month interval after the single dose for previously infected individuals with SARS-CoV-2 and the two doses for those not previously infected. The titer change, after the boost dose administration, on the other hand, was ≥ 1.5 FC higher than the titers at the 6/9-month time-points in both cohorts. A similar quantitative immune titer was observed in both cohorts 8 days after the last boost dose. The subsequent immunoresponse trend remains to be verified. Discussion: The results show that a very rapid first decline, from the highest antibody peak, was followed by a very slow decline which ensured immune protection lasting more than six months. The apparent absence of adverse effects of the rapid decline on the vaccine's immune protective role has been related to a large majority of low avidity antibodies induced by current vaccines. High avidity antibodies with prolonged anti-transmission efficacy show a longer half-life and are lost over a longer interval period. The cellular immunity, capable of preventing severe clinical diseases, lasts much longer. The unbalanced dual activity (cellular vs humoral) while effective in limiting ICU pressure and overall mortality, does not protect against transmission of SARS-CoV-2, resulting in high circulation of the virus among unvaccinated subjects, including the younger population, and the continuous production of variants characterized by changes in transmissibility and pathogenicity. The high mutation rate, peculiar to the RNA virus, can however lead to a dual opposite results: selection of defective and less efficient viruses up to extinction; risk of more efficiently transmitted variants as the current omicron pandemic.Conclusions: In conclusion the current bimodal antibody-titer decline, following BNT162b2 mRNA anti-SARS-CoV-2 vaccination, needs a further extended analysis to verify the protective borderline levels of immunity and the optimal administration schedule of vaccine boosters. Our current results can contribute to such goal, besides a direct comparison of other FDA-approved and candidate vaccines.