We aimed to understand how SARS-CoV-2 antibody titer decrease following SARS-CoV-2 mRNA vaccination and to estimate the timing of booster vaccination. Six hundred sixty-two healthcare workers were administered with total of three doses of SARS-CoV-2 mRNA vaccine during the same short period. Of them, three volunteers were enrolled to measure anti-receptor binding domain (RBD) antibody titers (IgG) monthly following the second and the third doses. Serum anti-RBD antibody titers were measured monthly and the decay curve of the antibody was analyzed. We estimate the timing of the third and fourth vaccine based on the observed antibody titer decrease and the period of breakthrough infections in the vaccine recipients. Anti-RBD antibody decreased exponentially following the 2nd dose. Between 108 and 117 days following the second dose, breakthrough infection of SARS-CoV-2 occurred in 11 out of the 662 vaccine recipients. Based on the decrease in anti-RBD antibody and the timing of the breakthrough infections, we estimate that the optimal timing of a third dose would be at earliest 108 days after the second dose, when anti-RBD antibody titers are less than 338 BAU/mL. The anti-RBD antibody titers were sustained relatively higher for 161 days following the third dose (416 days following the second dose). We estimate that the optimal timing of a third dose would be at earliest 108 days after the second dose, or anti-RBD antibody titers are less than 338 BAU/mL. We suggest that a fourth dose should be administered later than 161 days following the third dose.