Monitoring strategy of COVID‐19 vaccination in dialysis patients based on a multiplex immunodot method: The CovidDial study

COVID‐19 vaccine was demonstrated to be effective in dialysis patients, but boosters are mandatory due to a rapid waning of anti‐spike antibodies. A vaccination strategy based on immunologic response might be useful to maintain a favorable risk–benefit balance in this vulnerable population.

dialysis patients, with a 28-day related mortality of 25%. 1 When multiple severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) vaccines were put on the market in early 2021, vaccination program in high-risk populations including end-stage kidney disease (ESKD) patients became a priority for healthcare authorities in many countries. 2As recommended by the manufacturers, a second dose of mRNA-based vaccines was mandatory 3 (BNT162b2 vaccine) or 4 weeks (mRNA-1273 vaccine) after the first injection to strengthen the humoral protection. 3sed on the experience with influenza and hepatitis B vaccines, 4,5 there was concern about the SARS-CoV-2 vaccine effectiveness in ESKD patients since their innate and adaptative immune responses are compromised.This assumption was confirmed in a study showing an insufficient humoral protection with a lower seroconversion rate and a faster waning of antibodies titers when compared to a healthy cohort. 6Therefore, additional vaccine doses (boosters) were strongly recommended for at-risk populations, without taking into account the circulating anti-spike (S) antibodies levels. 2 Nevertheless, functional assays revealed that neutralizing anti-S antibodies levels were highly correlated with clinical protection from SARS-CoV-2 infection in healthy participants. 7In addition, the emergence of variants with numerous spike protein mutations results in a lower efficiency of vaccines 8 along with the humoral immunity waning that significantly increases the risk of COVID-19 related hospitalization or death. 9This suggests that a reliable follow-up of immunologic response after vaccination is required, and particularly to maintain a favorable risk-benefit balance in vulnerable population.
Therefore, to further refining the vaccine strategy based on individual serological responses to SARS-CoV-2 vaccination, we prospectively monitored four different anti-spike antibodies titers in patients receiving maintenance dialysis.The secondary objective consisted in determining potential correlation between the anti-spike antibodies levels induced by the initial two-dose vaccination regimen and the risk of subsequent COVID-19 infection in dialysis patients.

| Study design
The CoviDial study is a 1-year observational prospective study including patients on hemodialysis or peritoneal dialysis who received three doses of mRNA-1273 vaccine (Spikevax ® Moderna) for prevention of COVID-19 (Figure S1).

| Study population
Patients aged >18 years, admitted for chronic dialysis (hemodialysis or peritoneal dialysis) in our three dialysis units and able to give informed consent, were eligible to participate in the CoviDial study.
The demographic and clinical data, dialysis-specific parameters, medical comorbidities, COVID-19 infection status, information about medications, laboratory findings, and humoral response against SARS-CoV-2 were recorded at baseline.Patients were reassessed after 1 month (just before the second injection of the vaccine), 3, 6, 9 (before the 3rd injection of the vaccine), and 12 months (Figure S1).Based on a previous study testing the performance of this assay, values ≥10 AU were considered as positive, between 5 to 9 AU as equivocal, and <5 AU as negative. 10Details about the technique as well as its clinical specificity and sensitivity are published elsewhere. 10

| Electrochemiluminescence immunoassay (ECLIA) for SARS-CoV-2 antibodies detection
The Elecsys ® Anti-SARS-CoV-2 method (Roche ® ) which was performed in clinical routine for anti-nucleocapsid and anti-S protein RBD antibodies determination, was used as a comparator.The technique is based on a double-antigen sandwich assay format with a quantification by electrochemiluminescence technique (ECLIA).For the anti-nucleocapsid antibody assay, a cut-off index ≥1 E/S was considered as positive with a clinical sensitivity of 99.5% (95% confidence interval [CI]: 97%-100%) and specificity of 99.8% (95% CI: 99.69%-99.88%)at 14 days post-PCR confirmation.The results of the SARS-CoV-2 spike protein receptor binding domain antibodies determination were expressed as U/mL and considered to be positive for values ≥0.8 U/mL.According to the manufacturer's product information, clinical sensitivity and specificity were 98.8% (95% CI: 98.1-99.3%)and 99.98% (95% CI: 99.91-100%), respectively.

| Statistical analysis
Demographic and clinical characteristics of patients were presented as frequencies and proportions for categorical variables.For numerical variables, we calculated the median and interquartile range (IQR) as well as the mean and standard deviation (SD) or standard error of the mean (SEM), as appropriate.
According to the variable distribution, the nonparametric oneway analysis of variance (ANOVA) Friedman test was applied to compare anti-spike antibodies titers at each time-point, followed by Dunn's multiple comparison test.We used the nonparametric Mann-Whitney test to compare anti-spike antibodies titers at month 3 between the subsequent SARS-CoV-2 infected patients versus the noninfected cohort.
All statistical analysis was performed by using GraphPad Prism version 9.5.0 for Windows (GraphPad Software, San Diego, California USA, www.graphpad.com).A p value <0.05 was considered significant.

| Baseline demographic, clinical and pathological characteristics of the patients
From 187 screened patients, 66 patients were excluded for the following reasons: no serological evaluation available at baseline (n = 52), patients not yet on chronic dialysis (n = 4), and patients who have already received a dose of vaccine (n = 10).Among the 121 participants who fulfilled the inclusion criteria, 102 (84.3%) were treated with hemodialysis (Figure 1).
During the 12-month follow-up period, 16.5% of patients died (n = 20).The cause of death was from unknown origin in four patients, severe sepsis in four patients, cardiac failure in six patients, neoplasia in five patients, and following a surgical procedure in one patient (Figure 1).COVID-  At baseline, the IgG antibodies against the RBD, the ectodomain (S1 + S2), the S1 subunit, and the S2 subunit of the S-protein were found in 21.5%, 23.1%, 19.8%, and 21.5% of patients, respectively.

| The time course of vaccine response
The anti-S protein RBD measured by the Elecsys ® Anti-SARS-CoV-2 method was positive in 73.6% patients 1 month after the first vaccine injection and in 98.1% 2 months after the second dose, showing an excellent response to the mRNA-1273 vaccine.At month 6, almost all patients (99%) developed anti-RBD antibodies.Following the Belgian health authorities' recommendation for the COVID-19 vaccination, we administrated a third dose in mid-October 2021 (month 7), regardless of antibody levels.Two and 5 months later (months 9 and 12), 99% and 100% of patients had antibodies against the spike protein.
The anti-RBD antibodies detected by the multiplex immunodot technique were found in 57.9%, 96.2%, 91.8%, 97.1%, and 96.8% of patients at months 1, 3, 6, 9, and 12, respectively.Similar results were observed for the anti-S1 unit of the spike protein.The seroconversion rates were lower for the anti-S (S1 + S2, ECD) and the anti-S2 antibodies in the first months but rose to 97.1% and 95.1% by the ninth month.

| The time course of anti-SARS-CoV-2 antibodies levels
For the study of vaccination response kinetic, we analyzed the time course of anti-SAR-CoV-2 antibodies levels in 68 patients who received the three vaccine doses and for whom data were available at all time points.One month after the first vaccinal dose, the four different epitopes anti-spike antibodies levels measured by the COVIDOT assay were increased but only significantly for the anti-S1 and anti-RBD antibodies (30.4 ± 4.3 vs. 6.9 ± 2.2, p = 0.0156 and 38.7 ± 4.7 vs. 10.1 ± 3.0, p = 0.0016, respectively; Figure 2A,D).The second dose significantly boosted the immune response with approximately a two-fold increase of the mean antibodies levels 2 months later (anti-S1: 68.2 ± 3.6 vs. 30.4± 4.2, p < 0.0001; anti-S2: 32.9 ± 4.2 vs. 18.7 ± 4.4, p < 0.001; anti-S1 + S2: 57.0 ± 3.9 vs. 23.6 ± 4.5, p < 0.0001 and anti-RBD: 85.6 ± 3.0 vs. 38.7 ± 4.7, p < 0.0001).However, this response was not sustained over time as evidenced by the significant decrease of the four different anti-spike antibodies (anti-

| Initial vaccination response and further SARS-CoV-2 infection
Self-limited or asymptomatic COVID-19 infections were demonstrated by PCR testing in 15 patients (12.4%), the majority of which (10 subjects) occurred after January 1st, 2022, when the Omicron variant was predominant in Belgium 11 (Figure S2).
Based on the ECLI method for SARS-CoV-2 antibodies detection, a level below the upper limit of quantification (250 U/mL) was found at month 3 in 50% of patients (seven of 14, one patient died before the third month) who later tested positive for SARS-CoV-2 and 26.7% of patients (24 of 90) considered not to be contaminated by the virus during the 1-year observational period (p = 0.078).As shown in the Figure 4A, the mean antibodies level at month 3 was lower in SARS-CoV-2 positive patients compared to SARS-CoV-2 negative cohort, but without being statistically significant (162.3 ± 28.9 vs. 212.4± 8.1, p = 0.059).In COVI-DOT multiplex analyses (Figure 4B), the difference between the two groups was statistically significant for anti-RBD (71.57± 9.01 vs. 85.79 ± 2.61, p = 0.0131), anti-S1 + S2 (41.07 ± 7.96 vs. 61.68 ± 3.56, p = 0.0237), and anti-S2 (13.79 ± 5.03 vs. 39.70 ± 3.86, p = 0.0096), but not for the anti-S1 antibody (60.57± 8.27 vs. 69.70 ± 3.14, p = 0.158).In line with previously published studies, 12,13 we found a significant waning of anti-spike antibodies titers between months 3 and 6 confirming that the humoral response to the two-dose vaccination scheme is impaired in dialysis populations when compared to healthy patients. 14 a study evaluating immune response after the second dose of the seropositivity in the dialysis cohort, however with significant lower anti-spike antibody levels than in the healthy control group. 15Our study also brought to light that a third vaccine dose induced an immune response in 97% of patients with a statistically significant rebound of the four different anti-spike antibodies levels detected by the multiplex COVIDOT assay.Using a serologic test detecting IgG antibodies to the RBD of the S1 subunit of SARS-CoV-2 spike protein, Ducloux et al.

| DISCUSSION
showed that 93% hemodialysis patients had an antibody titer >50 AU 1 month after the third vaccine dose with a median increase in antibody titers of 580%. 16But no information was provided about the long-term antibodies kinetic and a possible lowering of antibodies levels as we observed in our prospective study 5 months after the booster vaccine.
Since there are numerous methods and lack of standardization for assessing the humoral and cellular immune response to SARS-CoV-2, 17,18 the comparison of studies is challenging especially to investigate potential relationship between antibodies titers and clinical outcomes.With this in mind, we decided to use two different methods consisting of an electrochemiluminescence immunoassay (Elecsys ® Anti-SARS-CoV-2 method) and a multiplex immunodot enzymatic method (COVIDOT).The Elecsys ® assay had the advantage to have a very high clinical sensitivity and specificity, but it evaluated only one single epitope with a relatively low upper limit of quantification (250 U/mL), whereas the COVIDOT allowed us to semi-quantify several Anti-SARS-CoV-2 antibodies in a single assay, however, with a lower sensitivity and specificity.
Interestingly, we found a statistically significant correlation between lower anti-spike antibodies titers after the two-dose regimen vaccination and the risk for de novo viral infection, remembering that the majority of infections occurred more than 6 months later when the Omicron variant was predominant.It is worth mentioning that studies with a follow-up longer than 6 months carried out in dialysis patients receiving three vaccine doses are relatively scarce. 19We can mention Ficiociello's work conducted in 124 hemodialysis patients who received three doses of mRNA-1273 vaccine resulting in a strong humoral response after 250-296 days. 12However, no relationship between antibody index value and ability to neutralize the virus or fight the infection has been demonstrated. 12e limitations of our study are mainly the small sample size demonstrated that previous mRNA-1273 two-dose regimen vaccination provided a marginal effectiveness against symptomatic Omicron infection whereas a booster vaccine increased the protection up to approximatively 50%. 20Compared to the initial SARS-CoV-2 virus, the lower protective efficacy of mRNA-based COVID-19 vaccine can be attributed to extensive mutations in the spike proteins, leading to a relative resistance to neutralizing monoclonal antibodies against the receptor-binding domain. 21Finally, our study did not investigate the potential relationship between in vitro neutralization levels and the anti-SARS-CoV-2 antibodies titers, although such a correlation was demonstrated in healthy subjects. 7 conclusion, our study confirms the importance of additional The protocol was specifically designed to monitor SARS-CoV-2 antibodies against the nucleocapsid protein (N) and four epitopes of the spike protein (S1, S2, Ectodomain S1 + S2 and Receptor Binding Domain) by using the multiplex D-tek COVIDOT 5 IgG immunodot assay during the vaccination period.In parallel, patients were tested for anti-nucleocapsid and anti-spike IgG antibodies with the clinically available Elecsys ® Anti-SARS-CoV-2 immunoassay (Roche ® ).According to the Belgian health authorities' recommendation for the COVID-19 vaccination schedule in the dialysis population, we started the immunization within our three dialysis units (CH EpiCURA, Ath, Baudour and Hornu, Belgium) in mid-March 2021.A second dose and a third dose of vaccine were administrated in mid-April and in mid-October 2021, respectively.The CoviDial study was conducted following the ethical principles of the Declaration of Helsinki and approved by the central ethical committee of CUB Erasme (Brussels, Belgium) as well as our local ethical committee (CH EpiCURA, Baudour, Belgium) on February 4, 2021 (CCB B4062021000037).
At different time points, a serum sample was collected by using a standardized operating procedure and immediately sent to our central lab.The BlueDiver COVIDOT 5 IgG (Alphadia D-tek, Mons, Belgium), a commercially available CE-marked enzyme immunoassay, was used for the semiquantitative detection of IgG antibodies against five main epitopes of the SARS-CoV-2: the nucleocapsid protein (N), the spike (S) protein (ECD, ectodomain, S1 + S2), the S1 subunit of the S-protein, the S2 subunit of the S-protein, and the receptor-binding domain (RBD) of the S1 subunit.The results are expressed as arbitrary units (AU) and range from 0 [negative result] to 100 [high positive result], depending on the color intensity of the enzymatic reaction.
19 infections were diagnosed by PCR testing in 15 patients (12.4%), mostly during the time when Omicron variant was predominant in Belgium 11 (10 out of 15 cases).The symptoms were mild or absent, and no patient died.F I G U R E 1 Patients flow chart.

3. 2 |
Anti-SARS-CoV-2 antibodies determination at baseline Based on the multiplex immunodot enzymatic immunoassay, 97 (80.2%) patients were seronegative for the anti-nucleocapsid antibody while 23 patients (19%) had a positive test.The remaining three patients had an equivocal result with values ranging between 5 to 9 AU.Nine of the 23 patients (39%) with a positive anti-nucleocapsid antibody test had a history of a symptomatic SARS-Cov-2 infection with a mean time interval of 5.3 ± 3.7 months [Min 0.7 -Max 11.7].

This 1 -
year observational prospective study demonstrated that the mRNA-1273 anti-SARS-CoV-2 vaccine was highly effective in patients receiving maintenance dialysis with responder rates of 73.6% 1 month after the first injection and 98.1% 2 months after the second dose.A booster dose administrated systematically 5 months later lead to a humoral response in 99% and 100% of patients at months 9 and 12, respectively.Alongside the humoral response, the vaccination resulted in effective clinical protection with a very low rate of SARS-CoV-2 infection accompanied by few or no symptoms.Interestingly, patients who were subsequently infected with SARS-CoV-2 initially had an inefficient humoral response after the primary two-dose vaccine regimen with significant lower titers of anti-RBD, anti-S1 + S2 and anti-S2 antibodies assessed by multiplex immunodot enzymatic assay at month 3.
attributable to exclusion of one third of the subjects at baseline and missing data regarding the anti-spike antibodies kinetics throughout the 1-year observational period.Conversely, the prospective nature of the protocol, the enrolment of a homogenous dialysis cohort vaccinated according to a uniform schedule with a narrow time range, the assessment of SARS CoV-2 antibodies titers based on an original multiplex immunodot enzymatic assay, and the evaluation of clinical efficacy strengthen the relevance of this study.Another limitation of our study is the lack of a systematic screening of COVID-19 infection, which may have resulted in underestimating its prevalence.According to the protocol applicable in our hospital at that time, SARS-CoV-2 PCR testing was only performed in patients with suspicious symptoms or in case of hospitalization for any reason.It should also be noted that most COVID-19 cases in our study were suspected to be related to the Omicron variant based on Belgian epidemiologic data and not on the virus sequencing (FigureS2).Interestingly, Altarawneh et al.
mRNA-based COVID-19 vaccines (boosters) to maintain a sufficient humoral response while ensuring an effective clinical protection in dialysis patients.The multiplex immunodot assay allowed us to highlight the association between a less robust immunization response to the initial vaccination and the risk of ensuing SARS-CoV-2 infection.Assessing circulating IgG antibodies titers against four main epitopes of the viral spike protein could be a valuable tool to personalize COVID-19 vaccine strategy and to maintain a favorable benefit (COVID-19 effective protection)-risk (side effects and vaccine-hesitancy) balance at the individual level.Further studies are required to prospectively validate this strategy in the context of emerging variant with mutations on SARS-CoV-2 spike protein and subsequent new mRNA-based vaccines.
Demographic and clinical characteristics of patients (n = 121).