Seroresponse to SARS-CoV-2 Vaccines among Maintenance Dialysis Patients over 6 Months

Abstract

Background and objectives: Although most patients receiving maintenance dialysis exhibit initial seroresponse to severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) vaccination, concerns exist regarding the durability of this antibody response. This study evaluated seroresponse over time.

Design, setting, participants, & measurements: This retrospective cohort study included patients on maintenance dialysis, from a midsize national dialysis provider, who received a complete SARS-CoV-2 vaccine series and had at least one antibody titer checked after full vaccination. IgG spike antibodies (anti-spike IgG) titers were assessed monthly with routine laboratory tests after vaccination; the semiquantitative assay reported a range between zero and ≥20 Index. Descriptive analyses compared trends over time by history of coronavirus disease 2019 (COVID-19) and vaccine type. Time-to-event analyses examined the outcome of loss of seroresponse (anti-spike IgG <1 Index or development of COVID-19). Cox regression adjusted for additional clinical characteristics.

Results: Among 1870 patients receiving maintenance dialysis, 1569 had no prior COVID-19. Patients without prior COVID-19 had declining titers over time. Among 443 recipients of BNT162b2 (Pfizer), median (interquartile range) anti-spike IgG titer declined from ≥20 (5.89 to ≥20) in month 1 after full vaccination to 1.96 (0.60-5.88) by month 6. Among 778 recipients of mRNA-1273 (Moderna), anti-spike IgG titer declined from ≥20 (interquartile range, ≥20 to ≥20) in month 1 to 7.99 (2.61 to ≥20) by month 6. The 348 recipients of Ad26.COV2.S (Janssen) had a lower titer response than recipients of an mRNA vaccine over all time periods. In time-to-event analyses, recipients of Ad26.COV2.S and mRNA-1273 had the shortest and longest time to loss of seroresponse, respectively. The maximum titer reached in the first 2 months after full vaccination was associated with durability of the anti-spike IgG seroresponse; patients with anti-spike IgG titer 1-19.99 had a shorter time to loss of seroresponse compared with patients with anti-spike IgG titer ≥20 (hazard ratio, 15.5; 95% confidence interval, 11.7 to 20.7).

Conclusions: Among patients receiving maintenance dialysis, vaccine-induced seroresponse wanes over time across vaccine types. Early titers after full vaccination are associated with the durability of seroresponse.

Keywords: COVID-19; COVID-19 vaccines; SARS-CoV-2; chronic hemodialysis; end-stage renal disease; maintenance; peritoneal dialysis.

Graphical abstract

Figure 1.

Flow diagram of 1870 patients included in the study. Baseline defined as anti-spike IgG titer >1 Index before or within 10 days after first dose of vaccine; prior coronavirus disease 2019 (COVID-19) defined as positive severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) test before full immunity (at 14 days after completion of a vaccine series).

Figure 2.

Anti-spike IgG titers versus months after date of full immunization, comparing by vaccine type. (A) Among patients without prior COVID-19, anti-spike IgG titers waned over time. (B) Among patients with prior COVID-19, anti-spike IgG titers remained stable. Of note, the box plots shown here are bounded by the upper and lower limits of the assay used. For example, in (A), the median (interquartile range; IQR) titer in month 1 was ≥20 (≥20 to ≥20) among the recipients of mRNA-1273 and ≥20 (5.89 to ≥20) among recipients of BNT162b2. The dots represent the outliers, defined as >1.5×IQR above the third quartile or <1.5×IQR below the first quartile. The tables below the plots show the number of titers for each month, by vaccine type.

Figure 3.

Kaplan–Meier time-to-event curves for the outcome of antibody titer <1 Index or diagnosis of COVID-19, among those without a history of COVID-19, by vaccine type. Among patients without prior COVID-19, Ad26.COV2.S/Janssen recipients had the shortest time to anti-spike IgG <1 Index, and mRNA-1273/Moderna recipients had more durable anti-spike IgG titer levels. Data are shown beginning at day 30, at which time all patients have had at least one opportunity for assessment of the outcome of antibody titer <1 Index via monthly laboratory measures. The curves, therefore, start at the proportion of patients who had not experienced the outcome as of day 30. Patients were censored at death, transplantation, administration of an additional vaccine dose (third dose or booster dose), or last available titer assessment.

Figure 4.

Kaplan–Meier time-to-event curves for the outcome of anti-spike IgG titer <1 Index or diagnosis of COVID-19, among those without a history of COVID-19, by maximum initial anti-spike IgG titer. Among patients without prior COVID-19, those with higher maximum initial anti-spike IgG titer had more durable anti-spike IgG titer levels. Data are shown beginning at day 30, at which time all patients have had at least one opportunity for assessment of the outcome of anti-spike IgG titer <1 Index via monthly laboratory measures. The curves, therefore, start at the proportion of patients who had not experienced the outcome as of day 30. Patients were censored at death, transplantation, administration of an additional vaccine dose (third dose or booster dose), or last available titer assessment. Patients with maximum initial anti-spike IgG titer <1 Index are not shown because, given our definition of maximum initial titer, all had experienced the outcome by the end of month 2. (A) All patients; (B) recipients of Ad26.COV2.S only; (C) recipients of mRNA-1273 only; (D) recipients of BNT162b2 only. 95% CI, 95% confidence interval; HR, hazard ratio.

Figure 5.

In multivariable Cox proportional hazards regression, vaccine type was most strongly associated with loss of anti-spike IgG seroresponse (outcome of anti-spike IgG titer <1 Index or development of COVID-19). Inadequate dialysis defined by hemodialysis dose single-pool Kt/V <1.2 or peritoneal dialysis dose weekly Kt/V <1.7. Hepatitis B surface antibody (HBsAb) ≥10 mIU/ml signifies hepatitis B seroimmunity. Immunomodulating medications include anti-inflammatory medications, antineoplastic agents, corticosteroids, and certain anti-infective medications. BMI, body mass index; ref, reference.