This study assessed the humoral response to COVID-19 vaccination in HD patients by measuring the IgG sRBD titres and the NAbs percentage of inhibition at four points of time: before vaccination (V1), 1 month after the primary COVID-19 vaccination (V2), 8 months after the primary vaccination (V3), and 1 month after the booster dose (V4).

At 8 months after the primary vaccine (V3), the median NAbs percentage of inhibition and the IgG sRBD antibodies titre increased significantly compared to 1 month after the primary vaccine (V2). One possible explanation is the exposure to COVID-19 infection. However, this hypothesis cannot be proven since we did not measure the antibodies to nucleocapsid (anti-N IgG) in this study. Another plausible hypothesis is that the humoral response might just have been maintained for months after the primary vaccination without any new exposure or infection of COVID-19. Among subjects with increasing NAbs percentage of inhibition and IgG sRBD antibodies titre from V2 to V3, four (12.5%) subjects were confirmed positive for COVID-19 with RT-PCR, and 12 (37.5%) experienced COVID-19 symptoms but did not get tested. This finding is consistent with a previous study by Zhong et al., which demonstrated that vaccinated subjects with previous COVID-19 infection had a higher increase in antibody levels compared to individuals without previously confirmed COVID-19 infection [16]. Another study by Fuëssl et al. revealed a similar finding i.e., previous COVID-19 infection was associated with higher antibody response after the first booster vaccination (p = 0.001) [17]. A systematic review and meta-analysis also demonstrated a stronger humoral response in haemodialysis patients with previous COVID-19 infection compared to those without. In this study, Peiyao et al. showed that the seropositive conversion rate in patients without prior infection (82.9%) was significantly lower than in patients with prior infection (98.4%) (p < 0.00001). In addition, when the antibody titres were compared among the two groups, the mean difference was 1.14, indicating that patients with prior infection are more likely to develop antibodies [18]. Not only in haemodialysis patients, this pattern of response has also been demonstrated in the general population. Prendecki et al. found that anti-S titres were significantly higher in individuals with previous natural infection (16,535 AU/mL [IQR 4,741–28,581]) than in infection-naïve individuals (615.1 AU/mL [IQR 286.4–1,491]; p < 0.0001) [19].

Sixteen (50%) subjects with increasing antibody titres at V3 never experienced any symptoms of COVID-19. This might imply that primary vaccination can reduce the severity of COVID-19 infection. Similarly, a previous study by Ranzani et al. also showed that the effectiveness of the Sinovac vaccine against symptomatic COVID-19 infection was 46.8% [20]. Another previous study Kaabi et al. demonstrated an efficacy of 78.1–72.8% in reducing the risk of symptomatic COVID-19 after the primary vaccination with inactivated COVID-19 vaccines [21].

This study demonstrated that the administration of the COVID-19 booster dose significantly increased IgG sRBD antibody titres in ESKD patients undergoing routine haemodialysis. A previous study by Patyna et al. also revealed a significant increase in antibody titres to 4,560 BAU/mL (646.7 – 7,272.5) (p < 0.001) among haemodialysis patients at 1 month after the booster vaccine [22]. Likewise, Shashar et al. also demonstrated that the IgG sRBD antibody titres increased significantly to 16,336.8 ± 15,397.3 AU/mL (p < 0.0001) in patients undergoing routine haemodialysis [12]. Bruminhent et al. evaluated the immunogenicity of a booster dose of ChAdOx1 nCoV-19 after primary vaccination of inactivated SARS-CoV-2 vaccines among dialysis patients. The study also showed that among haemodialysis patients, the anti-RBD IgG level and % neutralization by sVNT increased significantly from 85.3 BAU/mL (IQR 1106–3762.3) and 47.9% (IQR 13.5–85.4), respectively, to 1740.9 BAU/mL (1106–3762.3) and 99.4% (IQR 98.8–99.7), after the booster dose [23]. Another study by Cheng et al. also revealed a similar finding i.e., a significant increase of anti-RBD IgG after a heterologous booster dose with mRNA vaccine (AZ followed by Moderna) among haemodialysis patients. The anti-RBD IgG level increased from 1342.0 ± 1894.0 AU/mL to 22,011 ± 1016 AU/mL [24].

In addition, Quiroga et al. also revealed that a booster dose induced a strong humoral response as indicated by a significant increase in antibody titres after booster vaccines (142 [29–1666] IU/mL vs 6021 [1405–10 000] IU/mL] (p < 0.001) [25]. Another study by Agur et al. exhibited a significant increase in the S1-RBD antibody titre from 2.15 ± 0.75 to 3.99 ± 0.83 AU/mL among haemodialysis patients who received BNT162b2 as the booster dose [3].

In our study, the subjects with increasing IgG sRBD antibody titres were younger compared to subjects with declining antibody titres. This is consistent with a previous study that demonstrated that age has a significant effect on antibody levels after vaccination [26]. There is an inverse correlation between age and IgG levels in the dialysis group (p = 0.004). In the dialysis group, older age is associated with a lower antibody response [26]. Likewise, the same finding has been demonstrated in the general population. There is an inverse correlation between post-vaccination anti-S titre and age, with individuals older than 50 years generating a significantly weaker serological response than those younger than 50 years [19].

Laboratory values of haemoglobin, lymphocyte count, platelet count, BUN, creatinine, ferritin, and albumin were not significantly different between the two groups. This is contrary to the results of a previous study by Agur et al. The study revealed that low albumin levels affect lower antibody responses in booster vaccines. This could be caused by the much smaller number of subjects in our study. The frequencies of comorbidities, including DM, hypertension, anaemia, and joint pain, did not differ significantly between the two groups. This finding is similar to the study by Agur et al., which showed that the proportion of diabetes mellitus did not affect the antibody response [3].

The IgG sRBD antibody titre before the booster dose was significantly lower in subjects with increasing antibody titres after the booster dose (p = 0.004). This finding is clinically significant for patients who experience antibody waning after the primary vaccination. We can conclude that for these patients, the administration of a booster dose is beneficial and able to mount an adequate immune response by significantly increasing the IgG sRBD antibody titres. Therefore, the administration of a booster dose is recommended in haemodialysis patients whose antibody levels decline over time after the primary vaccination. Meanwhile, a proportion of patients maintain high antibody titres 8 months after the primary vaccination, indicating good serological persistence and longevity among haemodialysis patients. Similarly, Bensouna et al. demonstrated that subjects with low antibody titre after the second dose underwent a higher increase in antibody titres after the booster dose [27]. This result suggests that booster dose administration is beneficial for the population that mounts a low antibody response after primary dose vaccination.

In the additional analysis, our study also demonstrated that patients who experienced COVID-19 symptoms between the second dose and the booster dose had a significantly greater increase in IgG sRBD titre (ratio of IgG sRBD at V4 compared to V3) than the asymptomatic ones. This finding implies that patients who experienced COVID-19 symptoms and are thus suspected to have been infected with COVID-19 underwent a more robust humoral response. Similarly, a previous study by Karakizlis et al. also demonstrated that haemodialysis patients with COVID-19 past infections sustained higher anti-SARS-CoV-2-spike levels compared to infection-naïve patients after the primary vaccination. However, both groups reached the upper detection limit of 40,000 AU/mL after the booster dose administration [28].

This study showed that 3 subjects (10.71%) did not experience an increase in the NAbs percentage of inhibition, but had an increase in IgG sRBD antibody titres after vaccination. Comparably, Nayak et al. showed that almost half of the study subjects did not induce neutralizing antibodies but experienced an increase in IgG sRBD antibody titres after COVID-19 infection. These conditions may be related to variations in immune responses between different individuals [29].

Regarding the IgG sRBD GMT, our study showed different results from Shashar et al. [12]. In the study by Shashar et al., the IgG sRBD titre after the booster dose reached 16,336.8 AU/mL. Meanwhile, in our study, the IgG sRBD titre increased to 23,050 U/mL after the booster dose. This finding might be caused by different methods of antibody measurement. Our study utilized the Electro Chemiluminescence Immunoassay (ECLIA) (sensitivity 96%, specificity 99.9%), whereas Shashar et al. employed the Chemiluminescent Microparticle Immunoassay (CMIA) (sensitivity 93%, specificity 99.5%) [30].

In this study, the most common AEs after the booster dose include mild pain at the injection site (55.26%), mild fatigue (10.53%), and swelling at the injection site (10.53%). This finding is similar to the BNT162b2 booster vaccine clinical trial that revealed pain at the injection site (12.9%) and fatigue (7.2%) as the most common AEs [31]. No serious or life-threatening adverse events were reported among the study subjects after the booster dose.

The main strength of this study is its cohort prospective design, in which patients are followed-up for four weeks after the booster dose vaccination. Also, in this study, the immunogenicity was evaluated by measuring the neutralizing antibody levels against sRBD with the sVNT method, one of the most reliable methods to estimate the neutralizing antibody levels in serum. The limitation of this study is the relatively small number of samples. In addition, this study only assessed the immunogenicity based on only the humoral responses, not along with the cellular responses. As is known, the cellular immune responses also play crucial roles in protecting against COVID-19 [32]. Moreover, past COVID-19 infections could be confirmed since not all patients underwent RT-PCR. Instead, past COVID-19 infections could only be inferred from patients’ history of symptoms.