Exploratory evidence of mitigated immune imprinting following concomitant COVID-19 and influenza vaccination Authors: Min Joo Choi, WonSeok Gwak, Jae Young Yun, Hae Jeon Ju, Eunjoo Choi, Hee Jin Cheong, Woo Joo Kim, Chulwoo Kim, Hwa Jung Kim, A-Yeung Jang, SunKyung Yoon, June-Woo Lee, Hye-Sook Jeong, Joon Young Song Published: 2025 ABSTRACT AbstractConcomitant administration of coronavirus disease (COVID-19) and influenza vaccination schedules has been widely adopted. However, long-term data on the immunological impact, particularly on variant-specific responses, remain limited. In this 10-month prospective study, 80 participants (40 per group) received either a concomitant bivalent COVID-19 mRNA booster and quadrivalent influenza vaccination (Group C) or separate vaccinations (Group S) at least 4 weeks apart. Immunogenicity was evaluated using an anti-S IgG electrochemiluminescence immunoassay, anti- receptor-binding domain (RBD) IgG enzyme-linked immunosorbent assay, focus reduction neutralisation test (FRNT), and hemagglutination-inhibition assay. Both Groups C and S elicited robust immune responses for up to 10 months with no immune interference, as antibody responses against wild-type (WT) and Omicron BA.5 remained comparable between the groups. Notably, neutralising antibody titre against Omicron BA.5 were higher in Group C at 1 month post-vaccination, with significantly higher neutralising potency (FRNT50/anti-RBD IgG titer ratio) against Omicron BA.5 and a higher BA.5-to-WT ratio in FRNT. These findings may suggest alleviated immune imprinting, although this remains exploratory and requires further validation. For influenza, the overall antibody responses were comparable between the two groups. These findings support the feasibility of concomitant influenza and COVID-19 vaccination, without evidence of immune interference. Moreover, the enhanced variant-specific immune responses with concomitant administration suggest a potential immunological benefit, warranting future investigation.Subject terms: Infection, Vaccines Introduction Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) continues to cause substantial morbidity and mortality worldwide1. Annual vaccination is recommended to manage the recurrent coronavirus disease (COVID-19) epidemic and reduce disease burden2. As COVID-19 vaccination timing closely aligns with that of influenza vaccination, and with early data indicating that co-administration of these vaccines does not significantly impact immunogenicity, concomitant vaccination is now strongly recommended in many countries3–5. Coformulated vaccines are also being developed6. However, as more data accumulate, it has become evident that co-administration could lower short-term immunity against the SARS-CoV-2 wild-type (WT) strain, although its clinical significance remains unclear7–13. Our previous study conducted during the 2022–2023 season using an Omicron BA.5-containing bivalent mRNA vaccine also demonstrated a decrease in antibody response against the WT in the co-administration group compared to separate administration14. Notably, however, antibody responses against the Omicron BA.5 variant were numerically higher in the concomitant vaccination group, suggesting that the co-administered influenza vaccine may mitigate SARS-CoV-2 immune imprinting through a redirected immune response or T-helper cell immune enhancement. Further investigation is warranted to better understand its effects on immune imprinting and assess its long-term impact. Long-term data on concomitant vaccination immunogenicity are scarce. While a few recent studies follow-up to six months, they are limited in scope10,15. A UK study (2021–2022) assessed only WT spike antibody responses, whereas a US study (2022–2023) included variant-specific evaluations but had notable limitations, including small sample size, absence of pre-vaccination titre, and short influenza-COVID-19 vaccinations intervals (1–28 days), requiring cautious interpretation. Our previous findings from the 2022–2023 season showed reduced WT but enhanced Omicron BA.5 responses with concomitant vaccination14; expanding on these results, herein, we extended the follow-up period to 10 months and incorporated an in-depth evaluation of variant-specific immune responses. This allowed us to assess the durability and quality of humoral immunity, including neutralising potency and surrogate markers of immune imprinting, in individuals receiving COVID-19 and influenza vaccines either concomitantly or separately. Results Baseline characteristics Figure 1 illustrates the cohort formation and participants included in each analysis. Of 80 participants, 40 received concomitant vaccinations, and 40 received separate vaccinations. The median interval between influenza and COVID-19 vaccinations was 32 days (interquartile range [IQR], 29–33). Baseline characteristics at enrolment are shown in Supplementary Table 1, and participant inclusion in the immunogenicity analyses at each time point is shown in Supplementary Tables 2 and 3. At enrolment, the median age was 34.5 years (IQR, 29–41 years), 50% were female, and 58.8% had prior SARS-CoV-2 infection, with well-balanced groups. Breakthrough infections of SARS-CoV-2 (from 2 weeks to 10 months post-booster: Group C, 10/40 [22.5%] vs. Group S, 13/40 [32.5%]; p = 0.459) and influenza (Group C, 10/40 [25.0%] vs. Group S, 8/40 [20.0%]; p = 0.592) did not differ significantly between groups. Study flow diagram. The diagram shows participant allocation and follow-up, with sample sizes included in each immunogenicity analysis at baseline, 1, 3, 6, and 10 months after vaccination. For influenza vaccination, analyses were performed at baseline, 1, 7, and 10 months post-vaccination. Immunogenicity of the COVID-19 vaccine The antibody response to SARS-CoV-2 peaked at 1 month post-vaccination and gradually declined thereafter, but remained above pre-vaccination levels at 10 months, particularly for the Omicron BA.5 variant, which had a lower baseline titre. There were no significant differences between Groups C and S in anti-S, anti-RBD, and neutralising antibody geometric mean titres (GMTs) against either WT or Omicron BA.5 variants at any time point (pre-vaccination, 1, 3, 6, and 10 months) (Fig. 2, Supplementary Table 4, and Supplementary Fig. 2). Although the fold reduction in neutralising antibody titres against WT at 6 months compared with the peak at 1 month was significantly higher in Group C, this difference was not significant at 10 months (Fig. 2B–D). The fold-changes in neutralising antibody responses against Omicron BA.5 over time were similar between groups (Fig. 2G–I). Comparison of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) neutralising antibody kinetics and neutralisation potency by vaccine strategy. Kinetics of neutralising antibody titre against SARS-CoV-2 wild-type (WT) (A) and Omicron BA.5 (F) are shown for each group. Mean fold rise (MFR) in neutralising antibody titre is presented for the following time points: post-vaccination 1 month compared to baseline (B, G), 6 months compared to 1 month (C, H), and 10 months compared to 6 months (D, I). Numbers above each group represent geometric mean fold changes with corresponding 95% confidence intervals. Neutralisation potency, defined as the ratio of FRNT to RBD IgG titre, is illustrated in panels E (WT) and J (BA.5), where median and fold change vs. D0 are annotated above each group. Error bars represent the mean with 95% confidence intervals, and neutralisation potency is presented as the median value. All between-group comparisons were performed using the Mann–Whitney U test. Within-group comparisons over time (i.e., baseline versus post-vaccination in E and J) were analysed using the Wilcoxon signed-rank test. P values are indicated as follows: p < 0.05 (*), p < 0.01 (**), p < 0.001 (***). Abbreviations: GMT, geometric mean titer; ND50, 50% neutralising dilution; MFR, mean fold rise; D0, day 0 (baseline); 1 M, 1 month post-vaccination; 6 M, 6 months post-vaccination; 10 M, 10 months post-vaccination; group C, concomitant vaccination group; group S, separate vaccination group; ns, not significant. The geometric mean ratio (GMR, Group C/Group S), adjusted for age, sex, and pre-vaccination titres, was similar between groups at each time point, except for a significantly higher neutralising antibody response against Omicron BA.5 in Group C at 1 month post-vaccination. This difference was most pronounced at 1 month but lost statistical significance at later time points (Fig. 3). Subgroup analyses based on SARS-CoV-2 infection history revealed antibody responses largely consistent with the overall findings, with a trend toward higher adjusted GMTs in Group C for BA.5, particularly at 1 month post-vaccination, although the differences were not statistically significant (Supplementary Figs. 3 and 4, Supplementary Table 5). Geometric mean titer (GMT) ratio for the severe acute respiratory syndrome coronavirus 2 and influenza virus strains, after adjusting for age, sex, and pre-vaccination GMT: concomitant vaccination with influenza vaccine (group C) versus separate vaccination (group S). WT wild type, CI confidence interval. Before vaccination, the neutralising potency (FRNT50/anti-RBD IgG titre ratio) against the WT was significantly higher in Group C than in Group S, but this difference disappeared 1 month post-vaccination, with a substantial increase in Group S (Fig. 2E). After no difference at one month, Group C again showed higher neutralising potency against the WT than Group S at 6 months post-vaccination. Neutralising potency against Omicron BA.5 showed notable patterns. At 1 and 6 months post-vaccination, the neutralising potency of Group C was significantly higher than that of Group S at all time points. Conversely, Group S showed a decrease in neutralising potency at 1 month compared with baseline, followed by a slight increase at 6 months post-vaccination (Fig. 2J). The BA.5-to-WT ratio in FRNT, a potential surrogate marker of immune imprinting, was significantly higher in Group C than in Group S at 1 month post-vaccination. Although the ratio remained higher in Group C at later time points, the difference was not statistically significant (Fig. 4). Subgroup analysis revealed that this trend was primarily driven by participants with prior SARS-CoV-2 infection, in whom Group C showed a significantly higher BA.5-to-WT ratio at 1 month. In contrast, infection-naive individuals exhibited similar ratios between groups across all time points (Supplementary Fig. 5). Comparison of the ratio of the BA.5/wild-type titre in the focus reduction neutralisation test by vaccine strategy. The line and error bars represent the median and interquartile range (IQR). All statistical comparisons were performed using the Mann–Whitney U test. P values are indicated as follows: p < 0.05 (*), p < 0.01 (**), p < 0.001 (***). Abbreviations: WT, wild type; group C, concomitant vaccination group; group S, separate vaccination group; ns, not significant. Immunogenicity of influenza vaccination The post-vaccination seroprotection rates, seroconversion rates, and GMTs of HI antibodies against the four influenza strains did not differ significantly among groups, except in two observations (Fig. 3, Fig. 5 and Supplementary Tables 6 and 7). First, at 1 month post-vaccination, Group C had lower antibody titres (40.0 vs. 55.6, p = 0.015) and seroprotection rate (69.2 vs. 97.5%, p = 0.001) against influenza B/Victoria. However, this difference was not observed in the adjusted GMR, which accounted for pre-vaccination titer, age, and sex. Moreover, Group C showed somewhat higher seroprotection and GMR values at later time points. Second, Group C had a lower seroprotection rate against influenza A/H3N2 at 1 month post-vaccination (87.2% vs. 100%, p = 0.026), although the GMTs and seroconversion rates were similar. Notably, the adjusted GMR at 1 month post-vaccination favoured Group C and this trend was maintained over time. Comparative seroconversion rates and hemagglutination inhibition (HI) titre after influenza vaccination in concomitant (Group C) and separate (Group S) vaccination groups. Left: Seroconversion rates (%) against each influenza strain (A/H1N1, A/H3N2, B/Victoria, B/Yamagata) at baseline (pre) and at 1, 7, and 10 months post-vaccination. Right: Corresponding geometric mean titre (GMTs) of HI antibodies. Statistical comparisons were performed between Group C and Group S at each time point. Seroconversion rates were compared using the chi-square or Fisher’s exact test, and GMTs using the Mann–Whitney U test. Asterisks indicate statistically significant differences between groups: *p < 0.05, **p < 0.01, ***p < 0.001. Discussion In this prospective study, 80 young adults were followed up for over 10 months to evaluate the long-term impact of concomitant vaccination with an Omicron-containing SARS-CoV-2 bivalent vaccine and quadrivalent influenza vaccine (QIV). Both Groups C and S elicited robust immune responses for up to 10 months, with no evidence of immune interference, as antibody responses against both the WT strain and Omicron BA.5 variant were not reduced in Group C compared to those in Group S at any time point. Notably, neutralising antibody titres against Omicron BA.5 were 1.4-fold higher in Group C at 1 month post-vaccination. This group exhibited significantly higher neutralising potency against Omicron BA.5 and a higher BA.5 to WT ratio, raising an exploratory signal of possible immunologic benefit. For influenza, overall antibody responses were comparable between the two groups. Growing evidence suggests that concomitant mRNA COVID-19 booster and influenza vaccination may impact immunogenicity in the immediate post-vaccination period against the WT strain compared to separate administration7–12. However, its clinical significance remains unclear. Moreover, a recent UK study (2021–2022) reported that while anti-S IgG levels against the WT were lower 1 month after concomitant vaccination, this difference disappeared by 6 months, supporting concomitant vaccination as a feasible strategy10. The present study assessed the antibody response through neutralisation capacity and is the first to conduct long-term follow-up for nearly a year post-vaccination. In this study, at 1 month, neutralisation potency against the WT increased more in Group S, and the decline in antibody titres from 1 to 6 months was less pronounced in this group. However, the differences between groups disappeared later. After adjusting for key factors, including pre-vaccination titres, antibody levels did not differ significantly at any time point, suggesting no evident immune interference with concomitant vaccination. Previous studies on concomitant vaccination primarily focused on the immunogenicity of the WT SARS-CoV-2 monovalent vaccine7,8,10–12. However, with new variants emerging and the original strain disappearing, vaccines have incorporated variant strains since 2022–2023, making assessment of immune responses to emerging variants increasingly relevant16,17. A PubMed review (up to 1 February 2025) identified only three studies, excluding ours, that evaluated immune responses to variant strains9,15,18. A Dutch study (2021–2022) reported reduced Delta and Omicron neutralisation on day 42 with concomitant vaccination, whereas a US study (2022–2023) found increased BA.5 and XBB neutralisation at 1 and 6 months9,15. However, the absence of pre-vaccination titres in both studies limits interpretability. Additionally, the second study had a small sample size and relatively short interval between influenza and COVID-19 vaccinations (1–29 days), necessitating cautious interpretation. Another study, a prospective analysis from Israel (2022–2023) that measured pre- and post-vaccination changes, had a small sample size (18 per group) and found no significant differences in WT, BA.5, or BQ.1.1 neutralisation at 1 month post-vaccination18. Notably, the mean fold increase in WT antibody titres from pre- to post-vaccination was higher in the separate vaccination group (17-fold vs. 9-fold), whereas the BQ.1.1 titres increased more with concomitant vaccination (9-fold vs. 11-fold), suggesting a shift toward variant-specific immunity with concomitant vaccination. These conflicting findings may reflect differences in participant characteristics, vaccine composition, or SARS-CoV-2 exposure history. However, despite limited data, studies on the Omicron-containing bivalent COVID-19 vaccine suggest that immune responses to variants are comparable or higher with concomitant administration15,18. This aligns with our findings, where Group C showed a numerically higher BA.5 neutralising response and a significantly higher BA.5-to-WT ratio, suggesting that the co-administered influenza vaccine may enhance variant-specific immune responses. Although our study was not designed to assess vaccine effectiveness, the incidence of SARS-CoV-2 infection during the BA.5-dominant period was numerically lower in the concomitant group (22.5 vs. 32.5%, p = 0.459). However, given the small sample size and lack of statistical significance, interpretation should be cautious, and further studies are needed to determine the clinical relevance of enhanced variant-specific responses. Another notable finding of this study is that, unlike neutralisation potency against the WT, that against the Omicron BA.5 variant was significantly enhanced with concomitant vaccination, suggesting qualitative improvement in variant-specific responses. Unexpectedly, BA.5 neutralisation potency (FRNT50/anti-RBD IgG titre ratio) exceeded 1 in a few participants (three in Group S and four in Group C). This may stem from lower BA.5 anti-RBD IgG titres compared to those in previous reports, possibly influenced by ELISA calibration differences19, or involvement of non-anti-RBD antibodies in BA.5 neutralisation. While recent animal studies have suggested that enhanced NTD-targeted responses could contribute to improved cross-neutralisation20,21, their role remains unclear. Given their rarity and similar distribution across groups, they are unlikely to affect comparisons between the groups in this study. The immunological mechanism underlying the enhanced variant-specific response in Group C remains unclear. However, given the increased neutralising antibody titres against the variants relative to the WT, immune imprinting might be attenuated. Given the exploratory nature of these surrogate markers and the lack of direct mechanistic evidence, these observations remain speculative. Nonetheless, several immunological mechanisms may be postulated. One possible hypothesis is that co-administered influenza antigens may have stimulated innate immune pathways, leading to improved adaptive responses to SARS-CoV-2 variants. Alternatively, structural similarities or shared epitopes between influenza and SARS-CoV-2 proteins may contribute to cross-protective effects by enhancing B and CD4 + T cell responses against SARS-CoV-2 variants22–24. Given that immunogenicity is likely influenced by multiple factors, including prior exposure history and viral components, rather than solely by the co-administered influenza vaccine, further studies are needed to determine whether similar findings will be observed following the concomitant administration of recently introduced variant-specific vaccines, such as the XBB.1.5 and KP.3 monovalent vaccines, used from the 2023–2024 season onward. Recent studies suggest that immune imprinting to the WT SARS-CoV-2 strain may be attenuated by repeated exposures to variant strains25. Given the importance of prior exposure history in shaping vaccine response, we conducted subgroup analyses based on SARS-CoV-2 infection history. In our study, participants with prior SARS-CoV-2 infection—representing an additional antigenic exposure compared to infection-naive individuals—had all been infected with Omicron variants (BA.1, BA.2, or BA.5)(Supplementary Table 8), suggesting that repeated exposure to novel antigens may have led to relatively attenuated immune imprinting. While neutralising titre and potency showed similar trends across subgroups, the BA.5-to-WT neutralisation ratio was significantly higher in Group C only among previously infected participants. Regardless of co-administration, a single initial exposure to a novel antigen may be insufficient to overcome immune imprinting from prior vaccine antigens26. However, repeated exposures can help mitigate this imprinting, and co-administration may contribute to alleviating immune imprinting by facilitating such repeated exposure. Previous studies on the co-administration of COVID-19 and influenza vaccines have consistently demonstrated that influenza immunogenicity remains comparable and meets non-inferiority criteria across most strains3,4,7,9–11,13,18,27–30. Only two studies have reported failure to meet non-inferiority for B/Yamagata27 and B/Victoria28; however, the differences were marginal and likely of limited clinical significance, especially considering that the COVID-19 vaccines used in those studies were inactivated platforms rather than mRNA-based vaccines. In contrast, all studies involving co-administration of mRNA COVID-19 vaccines have shown no detrimental impact on the immunogenicity of influenza vaccines and, in some cases, even modestly enhanced immune responses3,4,7,9–11,29,30. Notably, two studies reported improved responses to A/H3N2 in the co-administration group: while one attributed the findings to difference in baseline titre, warranting cautious interpretation30, the other—uniquely assessing 6-month immunogenicity—found significantly less antibody decline between 1 and 6 months post-vaccination in the co-administration group, as measured by mean fold reduction10. Collectively, these findings suggest that the co-administration with mRNA COVID-19 vaccines does not compromise influenza immunogenicity, with one study indicating a potential for slower decline in antibody levels against A/H3N2 over time. In our study, consistent with previous reports, influenza immunogenicity was comparable between groups. Although seroprotection rates for A/H3N2 and B/Victoria were lower in Group C at 1 month post-vaccination, these differences were no longer significant after adjustment for baseline titre. From 6 months onward, adjusted GMRs for both A/H3N2 and B/Victoria tended to favour the co-administration group, indicating that concomitant vaccination was not inferior—and potentially immunologically favourable—compared to separate administration over time. Although these observations are based on a limited number of studies and should be interpreted with caution, the favourable trends—particularly with respect to long-term antibody persistence—warrant further validation in larger and more diverse cohorts. Future investigations should also explore the underlying immunological mechanisms, including the potential adjuvant like effects of mRNA vaccine—such as dendritic cell maturation and innate immune immune activation—on the responses to co-administered, antigenically unrelated vaccines31,32. Unexpectedly, in our study, despite excluding individuals with influenza virus infection (defined as self-reported diagnosis or a ≥ 4-fold HI titer rise), the seroprotection rate against B/Victoria in the co-administration group increased from 69.2% at 1 month to 92.6% at 10 months, suggesting possible immune boosting from exposure to influenza virus without disease development. However, the number of individuals who exhibited modest increases in HI titre over time (a 2-fold increase in HI titre without laboratory-confirmed influenza) was similar between the two groups, suggesting that asymptomatic exposures are unlikely to introduced differential bias in the observed immunogenicity outcomes. This study has several limitations. First, the non-randomised, preference-based design may introduce residual confounding despite balanced baseline characteristics. Differences in behavioural factors—such as exposure risk, health-seeking behaviour—as well as undetected prior infections could have influenced immunologic outcomes. However, we attempted to minimise bias by excluding participants with confirmed or serologically inferred infections (based on anti-N IgG for COVID-19 and ≥4-fold HI titer rise for influenza). Furthermore, the influenza vaccination rate in the preceding season exceeded 85% and was similar between groups, suggesting comparable baseline health behaviours and awareness. Second, the study cohort primarily consisted of healthy younger adults (median age 34.5 years); therefore, the findings may not be generalisable to older adults or individuals with underlying health conditions. Third, the relatively small sample size limits the statistical power to detect subtle differences between groups and precludes robust subgroup analyses. In conclusion, this study demonstrated that antibody responses to the WT strain remained comparable between groups for up to 10 months, while concomitant vaccination enhanced the neutralisation potency against Omicron BA.5 and increased the BA.5-to-WT neutralisation titre ratio. Further research is needed to assess whether these differences influence clinical outcomes and whether recently introduced variant-based vaccines, such as XBB and JN.1, exhibit similar immunological patterns when co-administered, while also considering vaccination frequency and antigenic differences following the transition to new variant-based vaccines. Methods Study design This study was an extended follow-up of a previous open-label, non-randomised clinical trial (NCT05970887) conducted at the International St. Mary’s Hospital in Incheon, South Korea14. Of the 154 participants in the initial study, 80 (40 per group) consented to a long-term one-year follow-up. All patients had received three prior COVID-19 vaccine doses. All participants had received three prior COVID-19 vaccine doses. For those with a history of SARS-CoV-2 infection, the infection had occurred at least 3 months before enrolment, and thus they had a total of four antigen exposures. No participants had more than one prior infection (Supplementary Table 8). Participants were assigned to either concomitant administration of the BA.5-based bivalent mRNA COVID-19 vaccine and QIV (Group C) or separate administration with a 4-week interval (Group S), matched by age, sex, and prior SARS-CoV-2 infection. Demographic and clinical data were collected at the baseline. Blood samples were collected at six time points: pre-vaccination (for both vaccines), 1, 3, 6, and 10 months post-SARS-CoV-2 booster, and 7 months post-QIV (Supplementary Fig. 1)). Additionally, at each visit, the participants completed a survey to evaluate COVID-19 or influenza infection throughout the study period, and anti-nucleocapsid (N) antibody titre were measured using the SARS-CoV-2 IgG assay to detect evidence of prior SARS-CoV-2 infection (Abbott Laboratories). SARS-CoV-2 infection during the study period was defined as either a self-reported laboratory-confirmed diagnosis or seroconversion of anti-nucleocapsid (N) antibodies from negative to positive. Influenza infection was defined as either a self-reported laboratory-confirmed diagnosis or a ≥ 4-fold rise in hemagglutination inhibition (HI) titre against at least one of the four tested strains. Participants who met these predefined infection criteria were excluded from subsequent immunogenicity analyses. The study was approved by the Ethics Committee of the International St. Mary’s Hospital (IRB: IS22OISE0060) and conducted in compliance with the Declaration of Helsinki and Good Clinical Practice. Written informed consent was obtained from all participants. Further details, including vaccine formulation, are provided in the Supplementary Material. Vaccines The BA.5-based bivalent BNT162b2 mRNA vaccine was a combination of 15 μg of mRNA encoding the WT spike protein and 15 μg of mRNA encoding the spike protein of the Omicron BA.4/BA.5 subvariant. Each single-dose (0.5 mL) influenza vaccine contained 15 μg of hemagglutinin per strain, including two A strains—A/Victoria/2570/2019 (H1N1) pdm09-like virus and A/Darwin/9/2021 (H3N2)-like virus—and two B strains—B/Austria/1359417/2021 (B/Victoria lineage)-like virus and B/Phuket/3073/2013 (B/Yamagata lineage)-like virus. All the vaccines were administered intramuscularly to the upper arm by trained staff at the trial site. In group C, the COVID-19 vaccine was administered to one arm and the influenza vaccine was administered to the other arm. Immunogenicity assessment To evaluate the immune response to SARS-CoV-2, we measured anti-spike (S) and anti-receptor-binding domain (RBD) IgG antibodies against both the WT and Omicron BA.5 variant. Anti-S IgG levels were evaluated using an electrochemiluminescence immunoassay (Elecsys anti-SARS-CoV-2 spike ECLIA; Roche Diagnostics, Pleasanton, CA, USA), following the manufacturer’s protocol. Additionally, anti-RBD IgG responses against WT and Omicron BA.5 were quantified using enzyme-linked immunosorbent assay (ELISA) as described previously33. Neutralising antibody titre against SARS-CoV-2 variants were measured using the focus reduction neutralisation test (FRNT) with a 50% neutralisation threshold (FRNT50). FRNT was performed against the WT strain (bCoV/Korea/KCDC03/2020, NCCP No. 43326) and the Omicron BA.5 variant (GRA: BA.5, NCCP No. 43426) as previously described34. Serum samples were serially diluted from 1:20 to 1:43,740 and neutralisation titre were calculated using a four-parameter logistic regression model. For influenza-specific immunity, the responses to four QIV antigens (A/H1N1, A/H3N2, B/Yamagata, and B/Victoria) were evaluated using standard HI assays. Serum samples were pretreated with a receptor-destroying enzyme and diluted from 1:10 to 1:1280. An HI titer of ≥40 was considered indicative of seroprotection. Immunogenicity assessments of the bivalent COVID-19 vaccine and QIV as indicators of the immune response included seroconversion rates, geometric mean titre (GMTs), and mean fold rise (MFR). Seroconversion was defined as a ≥four-fold rise from baseline for the COVID-19 vaccine, and for the influenza vaccine, as either a post-vaccination HI titer ≥40 (if baseline <10) or a ≥four-fold increase (if baseline ≥10). Statistical analysis For each time-point analysis, participants were excluded from the corresponding immunogenicity analyses if there was evidence of SARS-CoV-2 or influenza virus infection during the study period. Descriptive statistics were presented as frequencies and percentages of categorical variables. Antibody titre below the detection limit were assigned a value of half the detection threshold (1:200 and 1:20 for SARS-CoV-2 binding and neutralising antibodies, respectively, and 1:10 for the influenza virus). GMTs with 95% confidence intervals (CIs) were computed using the log10-transformed data. The GMR was determined by calculating the mean difference on a logarithmic scale. Categorical variables were analysed using the chi-squared test or Fisher’s exact test, whereas continuous variables were compared using the Mann–Whitney U test for between-group comparisons and the Wilcoxon signed-rank test for within-group comparisons over time. Statistical significance was set at p < 0.05. Subgroup analyses of COVID-19 vaccine immunogenicity were conducted based on prior SARS-CoV-2 infection status, which was verified using anti-nucleocapsid (N) antibody testing. All statistical analyses were conducted using SPSS Statistics for Windows (version 24.0; IBM Corp., Armonk, NY, USA), and the data were visualised using GraphPad Prism (GraphPad Software, San Diego, CA, USA). Supplementary information Supplementary Information Acknowledgements This work was supported by a National Research Foundation of Korea grant funded by the Korean government (RS-2022-00165946), Korea University Guro Hospital (KOREA RESEARCH-DRIVEN HOSPITAL) funded by Korea University Medicine (grant number: K2512581) and the Korea Disease Control and Prevention Agency (grant number: 6634-319-260-01). We are grateful to the staff of the International St. Mary’s Hospital infection control team for their assistance in recruiting and managing study participants and facilitating the study process. We also thank all individuals who volunteered to participate in this study. Author contributions M.J.C. and J.Y.S. obtained funding and designed the study. M.J.C., J.Y.Y., H.J.J., and J.Y.S. contributed to the acquisition of clinical and laboratory data. W.G., A.J., S.Y., J.L., and H.J. performed laboratory analyses. M.J.C., W.G., E.C., C.K., H.J., W.J.K., H.J.C., and J.Y.S. interpreted the data. M.J.C., J.Y.Y., E.C., and H.J.K. performed statistical analyses. M.J.C. and J.Y.S. draughted the manuscript. All authors reviewed the manuscript critically for important intellectual content and approved the final version for submission. Data availability The study data are not publicly available owing to ethical and regulatory restrictions on participant privacy. However, a pseudonymized individual study dataset will be made available from the corresponding author upon request. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. These authors contributed equally: Min Joo Choi, WonSeok Gwak. Contributor Information Hye-Sook Jeong, Email: jeongnih@korea.kr. Joon Young Song, Email: infection@korea.ac.kr. Supplementary information The online version contains supplementary material available at 10.1038/s41541-025-01347-y. References Associated Data Supplementary Materials Supplementary Information Data Availability Statement The study data are not publicly available owing to ethical and regulatory restrictions on participant privacy. However, a pseudonymized individual study dataset will be made available from the corresponding author upon request.