PDF
Citation
Print
INTRODUCTION
Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) Omicron has rapidly become the dominant variant since its emergence in November 2021. Two doses of the primary vaccine series have shown waning effectiveness in patients, and extra doses of vaccinations were recommended worldwide [1]. Additional doses of vaccinations are prioritized in immunocompromised patients, including solid organ transplant patients, considering their increased risk of morbidity and mortality reported during the early years of the pandemic [2]. Hence, for the transplant population, a third vaccination dose is highly recommended [3].
While enhanced immunogenicity in the transplant population after the third vaccine dose has been reported [4-11], whether the elicited immunity can protect against Omicron infection remains unclear. Considering that Omicron, with multiple variants, can partially escape immunity acquired from coronavirus disease (COVID-19) vaccines targeted towards the wild-type strain [12-14], immunogenicity elicited through vaccination may not correlate with the actual clinical infection outcome. Recent studies on the association between anti-SARS-CoV-2 antibody levels and breakthrough infections have shown variable results and variable cutoff values of antibodies for predicting breakthrough infection [6, 15-19]. This is partially because many studies were conducted retrospectively, and in prospective studies, the number of subjects was not sufficient. We prospectively evaluated the association between anti-spike antibody levels after completing a three-dose COVID-19 vaccine series and the incidence and severity of subsequent breakthrough infection in 287 kidney transplant recipients (KTRs) during the Omicron surge.
MATERIALS AND METHODS
This was a noninterventional prospective study that examined the efficacy of a third dose of COVID-19 vaccination in KTRs. The study has been registered at clinicaltrials.gov (NCT05156086). All participants provided written consent, and the study was approved by the Institutional Review Board of the Seoul National University Hospital (IRB number: H-2110-175-1266).
Study population
Between November 15, 2021 and February 10, 2022, we enrolled 287 patients who had received a kidney transplant at least six months before the first dose of vaccination and two doses of a COVID-19 vaccine (61 patients with two doses of a viral vector vaccine [ChAdOx1 of Astra Zeneca], 143 with two doses of an mRNA vaccine [126 with BNT162b2 of Pfizer-BioNTech and 17 with mRNA-1273 of Moderna], and 83 with a heterologous combination of ChAdOx1/BNT162b2) and who planned to receive a third mRNA monovalent vaccination (Supplemental Data Fig. S1). The exclusion criteria were as follows: patients with a prior history of SARS-CoV-2 infection, those who had received B cell-depleting therapy (e.g., rituximab and bortezomib) or T cell-depleting therapy (anti-thymocyte globulin) within the last six months, and those who restarted dialysis due to graft failure. All patients were followed up every one to two months after the third dose of vaccination.
Serologic measurements of anti-SARS-CoV-2 antibody
Vaccine-induced SARS-CoV-2-specific antibodies were quantified using an Abbott SARS-CoV-2 IgG II Quant test (Abbott, Chicago, IL, USA), which is a two-step chemiluminescent microparticle immunoassay that detects IgG antibodies against the receptor-binding domain (RBD) of the S1 subunit of the spike protein of SARS-CoV-2. Measurements were between 6.8 and 40,000 arbitrary units (AU)/mL, and the seropositivity cutoff, as per the manufacturer’s instruction, was ≥50 AU/mL. Serum samples were collected from all patients within three weeks before the planned date of the third vaccine dose and at four weeks (± one week) post-third vaccination. Additionally, 126 patients who were seropositive post-third vaccination and consented to additional tests underwent additional antibody tests three months (± one week) post-third vaccination.
Assessment of symptomatic breakthrough infections
Symptomatic SARS-CoV-2 breakthrough infection cases were identified through self-reporting or relevant symptoms with proof of a positive real-time reverse transcription–PCR test or rapid antigen test [20], performed by medical personnel between two weeks and four months post-third vaccination. Asymptomatic infections could not be detected because periodic SARS-CoV-2 confirmation tests were not performed in all subjects. Data on the types and duration of symptoms, types of treatments received, and treatment outcomes were collected through a questionnaire and electronic medical records.
Statistical analysis
Descriptive statistics are expressed as means (SDs) or medians (interquartile range, IQR), as appropriate. Between-group differences in dichotomous variables were assessed using the chi-square test or Fisher’s exact test, and those in continuous variables were assessed using an independent sample t-test or the Mann–Whitney U test. Anti-RBD titers before the third dose of vaccination were compared with those assessed four weeks after the third dose using Wilcoxon’s signed-rank tests. Stepwise multivariable logistic regression was used to identify factors potentially associated with breakthrough infection and hospitalization during infection. Age, sex, and factors associated with P< 0.2 in univariable logistic tests were included in the multivariable model. Cutoff values of anti-RBD IgG, used to group patients according to infection risk, were determined through ROC tests with the Youden index. The difference in the cumulative incidence of breakthrough infection was demonstrated using a Kaplan–Meier curve and analyzed using the log-rank test. A two-tailed P<0.05 was considered significant. Statistical analyses were performed using SPSS (version 21.0; IBM Corp., Armonk, NY, USA) and R 4.1.1 (http://www.r-project.org).
RESULTS
The enrolled patients had a median age of 49 yrs (IQR 38.5–60.0), and 51.9% were male. The median time after kidney transplantation was 5.6 yrs (IQR 1.9–10.7) (Table 1). The median intervals from the second dose of vaccination to the third dose were 111 (IQR 92–128) days for ChAdOX1-S/ChAdOX1-S primary vaccination, 94 (IQR 74–118) days for mRNA/mRNA vaccination, and 147 (IQR 133–159) days for ChAdOX1-S/BNT162b2 vaccination. Approximately three quarters of the patients were being administered a triple regimen comprising tacrolimus, antimetabolite, and steroids.
Anti-RBD IgG levels after third vaccination
All patients received an mRNA vaccine as the third dose of vaccination (BNT162b2 [N=247] or mRNA-1273 [N=40]). The seropositive rate of anti-RBD IgG increased from 57.1% (164/287) to 82.2% (236/287) after the third dose, with seroconversion in 58.5% (72/123) of seronegative patients after two doses of vaccination. The antibody levels also significantly increased from a median of 71.7 (IQR 7.2–402.8) AU/mL to 1,612.1 (IQR 153.9–5,489.1) AU/mL (P<0.001). The anti-RBD IgG levels after the third vaccination were 546.8 (IQR 25.4–1,832.5) AU/mL for ChAdOX1-S/ChAdOX1-S primary vaccination, 2,730 (229.5–7,847.1) AU/mL for mRNA/mRNA vaccination, and 1,689 (227.8–3,843.9) AU/mL for ChAdOX1-S/BNT162b2 vaccination (Fig. 1).
Through multivariable logistic regression, pre-third dose anti-RBD IgG seronegativity was identified as a risk factor for seronegative anti-RBD IgG after the third vaccination dose (hazard ratio [HR]=158.56) (Supplemental Data Table S1), even when including the type of primary vaccination instead of pre-third dose anti-RBD IgG seronegativity (because these two variables showed collinearity; data not shown). The use of mycophenolic acid or mycophenolate mofetil (HR=5.98 for a dose ≤500 mg/day and 19.01 for a dose <500 mg/day), the tacrolimus trough level (HR= 1.16), and primary vaccination with ChAdOX1-S/ChAdOx1-2 (HR=3.34) were significant risk factors for anti-RBD IgG seronegativity after the third vaccination dose (Supplemental Data Table S1). At three months after the third vaccination, the median anti-spike IgG titers of 124 patients (seropositive at four weeks after the third dose) were significantly decreased from 2,245.0 (IQR 657.8–6,292.3) AU/mL to 914.2 (IQR 148.9–3,751.2) AU/mL (P<0.001). Among them, 11/126 (8.9%) had become seronegative (<50 AU/mL) and 35/126 (23.2%) had anti-spike levels <200 AU/mL.
Symptomatic breakthrough infections and anti-spike IgG titers
Symptomatic SARS-CoV-2 breakthrough infection occurred in 65 patients within four months after the third vaccination dose. Among them, excluding one patient who was infected within two weeks post-third vaccination, 64 patients had symptomatic breakthrough infection. The median time to breakthrough infection was 83.5 (IQR 58.0–105.5) days. Post-third vaccination anti-RBD IgG levels significantly differed among the patients with and without symptomatic breakthrough infections or hospitalization (median±quartile, 1,779.6±5,936.7 for the no infection group, 722.2±4,509.4 for the symptomatic breakthrough infection without hospitalization group, and 33.5±119.5 for the symptomatic breakthrough infection with hospitalization group; P=0.00015) (Fig. 2). When the patients were regrouped according to an anti-spike IgG level of 400 AU/mL (determined based on an ROC curve; data not shown), the risk of developing symptomatic breakthrough infection differed significantly between the two groups (log-rank test, P<0.0001; Fig. 3A). In multivariable logistic regression tests, a post-third vaccination anti-spike IgG level <400 AU/mL was the only risk factor for symptomatic breakthrough infection (HR=3.46, P<0.001; Table 2). When the patients were grouped into four groups according to quartile values of anti-spike IgG after the third vaccination (Q1, 6.8–152.3; Q2, 152.3–1,624.5; Q3, 1,624.5–5,495.6; Q4, >5,495.6 AU/mL), the risk of symptomatic breakthrough infection did not differ among the Q2, Q3, and Q4 groups (P<0.0001) (Supplemental Data Fig. S2A).
Association between anti-spike IgG titer and hospitalization
Twelve patients (18.8%) with symptomatic breakthrough infections required hospital admission. Reasons for in-hospital care included oxygen therapy for pneumonia (N=3), severe gastrointestinal symptoms (e.g., diarrhea and vomiting; N=6), fever and myalgia for more than four days (N=2), and acute thrombotic occlusion of previous arteriovenous access (N=1). Among the nine patients who were admitted for non-pulmonary symptoms, seven showed poor oral intake and evidence of dehydration, of whom five showed acute kidney injury on admission. One patient with pneumonia required intensive care unit care, but the symptoms alleviated without mechanical ventilatory support.
The 52 patients who did not require admission had mild symptoms, which mostly alleviated with oral medication. Among the patients with symptomatic breakthrough infections, those who required hospital admission had lower post-third anti-RBD IgG levels than those who did not require admission (33.5± 119.5 vs. 722.2±4,509.4 AU/mL, P=0.016; Fig. 2). When the patients were regrouped according to an anti-RBD IgG level of 200 AU/mL (determined based on an ROC curve; data not shown), the risk of hospitalization was higher in patients with anti-RBD IgG <200 AU/mL than in those with anti-RBD IgG ≥200 AU/mL (log-rank test, P<0.0001; Fig. 3B). In multivariable models, a low post-V3 antibody level (<200 AU/mL) was a significant risk factor for hospitalization (HR=36.4, P=0.007; Table 3). When the patients were grouped into Q1, Q2, Q3, and Q4 groups, the risk of symptomatic breakthrough infection requiring hospitalization did not differ among the Q2, Q3, and Q4 groups (P<0.0001; Supplemental Data Fig. S2B).
DISCUSSION
The third COVID-19 vaccination dose significantly increased the seropositive rate and anti-spike antibody levels, in line with the results of previous studies [4-11]. However, 17.8% (51/287) of patients were seronegative despite being administered the third vaccination dose. As has been suggested by Manothummetha, et al. [21], mycophenolic acid use, tacrolimus levels, and the type of priming vaccination were risk factors for seronegativity post-third vaccination. However, these factors were unrelated to the incidence of breakthrough infection in both the univariable and multivariable models. Clinical risk factors for seronegativity were not risk factors for symptomatic breakthrough infection or hospitalization. Therefore, to reduce the infection risk, patients should be more accurately stratified using serological tests rather than known clinical factors of a poor vaccine response.
The anti-spike antibody level after the third dose of COVID-19 vaccination was associated with the rate of symptomatic breakthrough infections and hospitalization. Studies on the correlation between anti-spike antibody levels after the third vaccination dose and outcomes of SARS-CoV-2 breakthrough infections in KTRs are still limited and have shown variable results [6, 16-19, 22]. In contrast to the results of the present study, Lammert, et al. [6] and Kemlin, et al. [17] did not observe a significant association between post-third vaccination anti-SARS-CoV-2 antibody levels and breakthrough infections. However, the number of patients analyzed in these studies was relatively small (103 and 65, respectively), limiting the ability to reach statistical significance, and the use of different modeling methods can also have an influence. We considered antibody levels as a dichotomous variable using cutoff values from ROC curves because of the assumed nonlinear relationship between antibody levels and infection or hospitalization (Fig. 2 and Supplemental Data Fig. S2). Patients with antibody levels in the second (Q2), third (Q3), and fourth (Q4) quartiles did not show differences in the risk of symptomatic breakthrough infection or hospitalization. Moreover, when we analyzed antibody levels as a continuous variable, the results did not reach statistical significance.
In a US multicenter observational cohort study of 666 solid organ transplant recipients (including 351 KTRs) who had three or more vaccinations, lower anti-RBD antibody levels of 0.8–250 IU/mL (based on the Roche kit, similar to the WHO standard unit) were associated with an increased risk of breakthrough infection during the Omicron surge [16]. In France, Bertrand, et al. [18] reported that lower anti-spike antibody levels (<264 binding antibody units/mL, WHO standard) were associated with an increased risk of breakthrough infection, hospitalization, and death. A prospective cohort study in Italy, including 614 solid organ transplantations (including 275 KTRs), reported that antibody levels <817 IU/mL were associated with breakthrough infection [19]. These three studies presented results similar to ours, i.e., that anti-RBD antibody levels <400 IU/mL (based on the Abbott anti-SARS-CoV-2 IgG II Quant) were associated with an increased risk of symptomatic breakthrough infection. The antibody levels associated with an increased risk of breakthrough infection varied between the studies, which is thought to be due to differences in the test kits used (antibody units are still not standardized), patient immunosuppression status based on different drug regimens among hospitals, exposure to and amount of virus from the person of care, facilities, and the prevalence of Omicron infection in the community.
An anti-RBD antibody level <200 IU/mL (based on Abbott anti-SARS-CoV-2 IgG II Quant) was associated with an increased risk of hospitalization. In a Spanish study that included 965 KTRs [22], patients with an anti-RBD antibody level <100 AU/mL (based on Abbott anti-SARS-CoV-2 IgG II Quant) were at increased risk of breakthrough infection and pneumonia, and patients with an antibody level <20 AU/mL had an increased risk of death during the Delta or Omicron surge after two or three vaccinations. The severity of Omicron infection is markedly attenuated compared with that of the previous variant, and deaths from COVID-19 are decreasing. However, immunosuppressed patients, such as KTRs, have a higher chance of acquiring severe disease than healthy people.
Identifying patients with severe outcomes, such as hospitalization and death, and taking additional preventive measures will be essential for the management of COVID-19 in the transplant population. The modulation of immunosuppressive regimens during vaccination [23], vaccinations using novel vaccines of different platforms [24] or vaccines targeting multiple antigens, and pre- or post-exposure passive immunization using monoclonal antibodies or convalescent plasma [25, 26] have been suggested as additional measures to enhance immunogenicity in KTRs. Multiple studies [27], including our own, have demonstrated a positive association between mycophenolic acid use or tacrolimus levels and the vaccine response in KTRs. While reducing immunosuppression, e.g., with mycophenolic acid or tacrolimus, appears to be a potential strategy to enhance the vaccine response [28], its efficacy and safety must be critically evaluated. Specifically, recent investigations have shown that mycophenolic acid withdrawal does not significantly improve the vaccine response or only yields a short-lived response with an increase in alloimmunity, as indicated by elevated levels of donor-specific antibodies [29]. Therefore, although reducing immunosuppression may be a promising approach to enhance the vaccine response, its potential benefits and risks should be carefully assessed in future studies. The optimal method for preventing symptomatic infection in KTRs with low anti-SARS-CoV-2 antibody levels is yet to be determined. Although additional booster shots are effective in inducing seroconversion [28], the current evidence does not strongly support bivalent vaccines over monovalent vaccines for protection against the Omicron variant in the transplant population. Pre-exposure prophylaxis with monoclonal antibodies, such as tixagevimab and cilgavimab, has demonstrated neutralizing activity against Omicron, but recent evidence suggests that a higher dose is required to prevent breakthrough infection, and neutralizing antibody activity commonly wanes by three months post-injection [30, 31]. Given the complex nature of immunosuppression in KTRs and the evolving nature of the COVID-19 pandemic, continued studies are crucial for developing effective prevention and treatment strategies.
There are some limitations when stratifying KTRs using widely used automated binding anti-RBD antibody tests, such as the Abbott anti-SARS-CoV-2 IgG II Quant. First, as tests have been usually developed against the ancestral strain, they may not accurately reflect the neutralizing capacity against the Omicron variant. However, studies have demonstrated that such tests exhibit a good correlation with the results of live virus neutralization tests against both the ancestral strain [32] and the Omicron variant [33, 34]. It should be noted that T cell responses also play a role in preventing SARS-CoV-2 infection, and weak correlations between T and B cell responses have been reported, particularly in cases of weak immune responses [35-37]. Nevertheless, automated binding antibody reagents are useful in identifying immunocompromised patients who require additional measures at a low cost. In some countries, antibody testing in transplant patients is used as a criterion for recommending monoclonal antibody therapy in cases of low antibody titers [18].
There are some precautions in applying the results of this study to the current COVID-19 situation or healthy people. First, most of the population worldwide has had a prior history of SARS-CoV-2 infection and probably has sufficiently high antibody levels. The association between anti-RBD IgG levels and symptomatic breakthrough infection was weak in patients with higher anti-RBD IgG levels. Second, immune responses to new SARS-CoV-2 variants that continue to emerge will differ from those to Omicron. Third, social and behavioral factors, which are essential in explaining the risk of infection, should be considered. Breakthrough infections can even occur in healthy individuals with very high antibody titers to SARS-CoV-2 [38]. The extent of virus exposure likely plays a critical role in breakthrough infections. Nevertheless, this study is meaningful in that it prospectively confirmed the association between anti-RBD binding antibody levels and the rate of symptomatic breakthrough infections and particularly very low antibody levels with hospitalization, in KTRs who are vulnerable to severe complications from COVID-19. These results can be useful as a basis for identifying patients who could benefit from additional protective measures among KTRs.
SUPPLEMENTARY MATERIALS
Supplementary materials can be found via https://doi.org/10.3343/alm.2024.44.1.64
XML Download