COVID-19 Vaccination Imperatives in People on Maintenance Dialysis

The terrible effect of coronavirus disease 2019 (COVID-19) on people on maintenance dialysis is now well recognized. The case fatality rate of 20%–30% combined with the exposure risk, particularly for those on hemodialysis (HD), has led to a high overall death rate (1,2). In Ontario, Canada, 162 people on maintenance HD have died from COVID-19, equivalent to about 1.7% of the entire Ontario HD population (1). Few other populations have suffered such a high death rate from this virus.

Prevention of infection with vaccination in those on dialysis is, therefore, a high priority. However, the initially limited vaccine supply has made this problematic in many countries. Two questions have arisen. First, should those on dialysis be prioritized for vaccination ahead of the general population? Second, how effective is vaccination in those on dialysis, and what strategies might maximize efficacy?

With regard to prioritization, the story has been very varied. In Canada, Ontario put those on dialysis in the second-highest priority level, but most other provinces mainly prioritized by age. In the United States, there was also variation from state to state. The United Kingdom, Sweden, Italy, France, and Belgium prioritized patients on dialysis, but most European countries did not do so initially or explicitly, although this sometimes happened at the local level. The decision in each jurisdiction involved an interplay between vaccine supply, public health, and ethical considerations, but there may also have initially been insufficient awareness of the high vulnerability of the dialysis population to COVID-19 (1,2).

A further complication has arisen in those countries, including the United Kingdom, Germany, France, and Canada, where strategies to maximize the number of people who could be vaccinated quickly included delaying second mRNA vaccine administration from the recommended 3–4 weeks to 2–4 months. This raised the question of whether less immunocompetent populations, such as those on dialysis, should be exempted from the extended interval. In Canada, a number of provinces, including Ontario and Quebec, have now exempted some or all patients on dialysis from the extended interval, but delays in the second dose occurred prior to this policy change. Some UK regions exempted patients on dialysis who were on transplant waiting lists but not those on dialysis in general.

What do the data show? The relevant findings come from an increasing number of recently published studies looking at the serologic responses of these patients to vaccination, and we will focus on five of these (3–7). Hopefully, these studies will soon be supplemented by clinical data on incidence and severity of COVID-19 infections in patients on dialysis who are vaccinated once versus twice versus not at all, because clinical outcomes are the “bottom line.” Serologic responses may or may not correlate with actual protection from the virus, but resolving this will take more time.

Before discussing the findings, some points need to be emphasized. Most studies in patients on dialysis involve the BNT162b2 Pfizer BioNtech vaccine, but one looks at Moderna’s mRNA-1273 (6). Extrapolation of conclusions to other vaccines may not be appropriate. The studies use different antibody ELISA assays, and although almost all measure antibodies to the spike receptor binding protein (RBD) of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), their relative sensitivities may not be the same. It is unclear which antibody and what titer best protect against the virus, although anti-RBD is thought to be most important (4). Control subjects vary, with some studies having none and others using age-matched people, health care workers, or even convalescent serum from infected people not on dialysis—a more demanding comparator (4). Not all studies measure baseline serology to determine whether subjects had infection-induced antibodies prior to vaccination. New follow-up data on SARS-CoV-2 infection–induced anti-RBD antibodies in 2215 US patients on dialysis show that 93% still had detectable antibody after 6 months, and in 60%, it remained at a high level; so, inclusion of such patients will make the effect of vaccination look misleadingly good (8).

The first question is how well does a single BNT162b2 vaccine work? Goupil et al. (4) from Quebec looked at anti-RBD antibodies in 135 patients on HD with no prior SARS-CoV-2 infection. Only 43% had a detectable response 4 weeks after a single dose, and this did not improve after 8 weeks. Those who responded had much lower titers than healthy controls. Yau et al. (unpublished data) from Toronto also looked at anti-RBD in 142 patients on HD 4 weeks after a single BNT162b2 dose and found that just 47% responded. Again, titers were low, with <10% having levels equal to those in convalescent serum from previously infected nondialysis controls (Yau et al., unpublished data). Similarly, Chan et al. (6) from New York found only a weak antibody response and not until 4 weeks after an initial dose of mRNA 1273 in 40 patients on HD.

How does the performance of two-dose vaccination compare? Grupper et al. (3) looked at serology responses to conventional-interval two-dose BNT162b2 in 56 patients on HD at a single Israeli center a median of 30 days after the second dose. They measured anti-RBD antibodies, and impressively, 54 of the 56 recipients, equivalent to 96%, showed a positive response, although titers were less than half of those in healthy vaccinated controls. Older patients were particularly likely to be low responders. Yau et al. (unpublished data) found that 88% of 76 recipients had an anti-RBD response, and an impressive 60% had antibodies levels above those seen in convalescent serum. The response was even better for antispike antibodies, with 96% seroconverting. Similar findings come from Yanay et al. (5) in Israel with BNT162b2 and Chan et al. (6) with mRNA-1273, who showed response rates of 90% and over 80% in 160 and 40 patients, respectively, although again, titers were low (5,6). Overall, these studies show that the proportion of patients on dialysis with a detectable response is much higher and that the response is notably stronger in those receiving two vaccines versus one.

T cell–mediated immunity may be more crucial for responding to viral infection but is more challenging to study. Schrezenmeier et al. (7) from Berlin examined T cell response to SARS-CoV-2 using an IFN-γ release assay 3–4 weeks after double-dose BNT162b2 vaccination. Responses were present in 68% of 31 patients on HD compared with 93% of similarly aged controls, and the IFN-γ levels were lower. T cell and antibody responses were correlated in the patients on HD. Responses after one vaccination were not measured. One small study may be insufficient to draw robust conclusions.

What about the response of previously infected patients on dialysis to vaccination? Goupil et al. (4) looked at 19 previously infected patients on HD and found that prevaccination antibody levels were similar to those of previously infected nondialysis controls and that there was a further serologic response to a single BNT162b2 vaccination in 16 of these 19, with antibody levels higher at 8 than at 4 weeks but still lower than in the controls (4). Similarly vigorous responses were seen by Chan et al. (6) after even one mRNA-1273 vaccine in 20 patients with prior infection (6). This suggests that one vaccine dose elicits a significant serology response in most patients on HD with previous infection.

What are the implications of these findings? First, using serologic criteria, vaccination done correctly can work in the large majority of people on dialysis. The serologic data support the early clinical impression that vaccinated patients on dialysis have less frequent and less severe subsequent SARS-CoV-2 infection. There is no justification for therapeutic nihilism and no evidence to support vaccine hesitancy or refusal in this vulnerable population. Second, two doses work much better than one, both in the frequency and in the intensity of the serologic response. This strongly supports the argument for patients on dialysis to be exempted from the extended interval in jurisdictions using this policy for their general population. The extended interval essentially delays the achievement of a much better serologic response in this mortality-prone population. As vaccine supply increases, the issue will become less relevant in many high-income countries, as is already the case in the United States, but it may be important for some time to come in less fortunate countries. Third, in the subgroup of patients on HD with prior infection, one vaccine may, at least in the short term, be serologically as good as two doses in those not previously infected, although two are still recommended (4,6).

A few other points should be made. Only one of these studies included people on home dialysis; although home patients have lower risk of infection than their center HD counterparts, their case fatality rates are equally high, and the serologic findings are similar (1,2,5,9). Therefore, the same imperative for vaccination and exemption from delayed second doses should apply. Future research should focus on which antibodies, what titers, and which T cell responses best correlate with clinical protection; whether postvaccination antibody titers in people on dialysis will persist as long as infection-induced levels; and whether these levels should be monitored regularly and used to trigger booster doses of vaccine. Interestingly, France has already recommended a third vaccine dose for patients on dialysis and patients with transplants. Also, how will all this be influenced by the new SARS-CoV-2 δ-variant for which double vaccination seems especially important?

These data are therefore a “call to action” for the nephrology community internationally to advocate for people on dialysis to receive both urgent vaccination and exemption from extended intervals. Time is short, and fatality rates in this vulnerable population are high.

Disclosures

P.G. Blake reports consultancy agreements with Ontario Renal Network; receiving honoraria from Baxter Global for speaking engagements; and serving on the editorial board of American Journal of Nephrology and as Medical Director of Ontario Renal Network. M.A. Hladunewich reports consultancy agreements with Alnylam Pharmaceuticals; receiving research funding from Calliditas Therapeutics, Chemocentryx, Ionis, Pfizer, and Roche; serving as a scientific advisor or member of Kidney International and UpToDate; and serving as Medical Lead for the Glomerular Disease Ontario Renal Network. M.J. Oliver is the sole owner of Oliver Medical Management Inc., which is a private corporation that licenses the Dialysis Measurement Analysis and Reporting (DMAR) software system. Oliver Medical Management Inc. is co-owner of a Canadian patent for DMAR systems. M.J. Oliver reports receiving honoraria from Amgen, Baxter Healthcare, and Janssen; other interests/relationships with Ontario Health; and contracted Medical Leads at Ontario Renal Network.

Funding

None.