Immune response to COVID-19 vaccination in a population with a history of elevated exposure to per- and polyfluoroalkyl substances (PFAS) through drinking water

Abstract

Background: Exposure to per- and polyfluoroalkyl substances (PFAS) has been linked to lower vaccine-induced antibody concentrations in children, while data from adults remains limited and equivocal. Characteristics of PFAS exposure and age at vaccination may modify such effects.

Objective: We used the mass administration of novel COVID-19 vaccines to test the hypothesis that prior exposure to environmentally-relevant concentrations of PFAS affect antibody response to vaccines in adolescents and adults.

Methods: Between April and June 2021, 226 participants aged 12-90 years with a history of exposure to PFAS in drinking water and who received an mRNA COVID-19 vaccine participated in our prospective cohort study. SARS-CoV-2 anti-spike and anti-nucleocapsid antibodies (IgG) were quantified before the first and second vaccine doses and again at two follow-ups in the following months (up to 103 days post dose 1). Serum PFAS concentrations (n = 39 individual PFAS) were measured once for each participant during baseline, before their first vaccination. The association between PFAS exposure and immune response to vaccination was investigated using linear regression and generalized estimating equation (GEE) models with adjustment for covariates that affect antibody response. PFAS mixture effects were assessed using weighted quantile sum and Bayesian kernel machine regression methods.

Results: The geometric mean (standard deviation) of perfluorooctane sulfonate and perfluorooctanoic acid serum concentrations in this population was 10.49 (3.22) and 3.90 (4.90) µg/L, respectively. PFAS concentrations were not associated with peak anti-spike antibody response, the initial increase in anti-spike antibody response following vaccination, or the waning over time of the anti-spike antibody response. Neither individual PFAS concentrations nor their evaluation as a mixture was associated with antibody response to mRNA vaccination against COVID-19.

Impact statement: Given the importance of understanding vaccine response among populations exposed to environmental contaminants and the current gaps in understanding this relationship outside of early life/childhood vaccinations, our manuscript contributes meaningful data from an adolescent and adult population receiving a novel vaccination.

Keywords: Emerging contaminants; Epidemiology; Health studies; PFAS.

Fig. 1. Schedule of blood sampling relative to vaccine administration.

Participants were asked to provide a blood sample before their first and second vaccine doses as well as blood samples during two follow-up visits, anchored around 30 and 60 days, respectively, after their second dose of the 2-dose vaccine series. Target windows depicted in this diagram are inclusive of both the different schedules for Pfizer-BioNTech (+21 day) and Moderna (+28) second doses as well as a 7-day buffer granted to participants for scheduling convenience, participants arriving outside of this schedule were included if their blood draws fell within the wider windows for follow-up depicted here. Baseline data are referred to as “day 0” throughout, including when the data are presented graphically.

Fig. 2. Log₂ anti-S antibody AUC as a function of days since first vaccination dose (Pfizer-BioNTech or Moderna) for participants recovered from a prior COVID-19 infection or naive to COVID-19.

A All data collected as a function of time (days from first vaccine are on the x-axis) for both the participants naive to (blue) and recovered from (red) prior COVID-19 infection. The longitudinal data by log₂ PFOA (B) and PFOS (C) quantile, boundaries of each quantile are displayed in gray shaded header above each panel.

Fig. 3. Change in log₂ anti-S antibody AUC from baseline to the two longest follow-up windows for participants recovered from a prior COVID-19 infection and those naive to COVID-19.

Box plots of the change in anti-S antibody AUC between baseline and third visit (A) and between the third (42–68 days after first vaccine) and fourth visits (70–103 days after first vaccine) (panel B) are shown for participants naive to and recovered from a prior COVID-19 infection. The box indicates interquartile range (IQR = Q–Q1); midline indicates the median. Whiskers are minimum or maximum without outliers. The outliers are the numbers that are below Q1 − 1.5*IQR or above Q3 + 1.5*IQR.