Prenatal Exposure to Poly- and Perfluoroalkyl Substances (2009-2014) and Vaccine Antibody Titers of Measles, Mumps, Rubella, and Varicella in Children Four to Eight Years Old from the Healthy Start Cohort

Abstract

Background: Prenatal exposures to certain poly- and perfluoroalkyl substances (PFAS) are associated with reduced humoral responses to some childhood immunizations.

Objective: We estimated associations between prenatal PFAS exposure and child antibody titers for measles, mumps, rubella (MMR), and varicella after immunization.

Methods: We measured serum antibody titers of 145 children (4-8 y old) enrolled in the Healthy Start cohort in Colorado, whose mothers had PFAS quantified mid-pregnancy (2009-2014). We used linear and logistic regression models to assess the relationship between five PFAS detected in $>65\%$ of mothers and continuous or non-high-censored ("low") antibody titers and quantile g-computation to evaluate the overall effect of the PFAS mixture.

Results: Median concentrations of individual PFAS were at or below the median reported among females in the United States. After receiving two vaccine doses, seropositive levels of antibodies were detected among most (93%-100%) children. Each log-unit increase in perfluorononanoate was associated with 2.09 [95% confidence interval (CI): 1.13, 3.87] times higher odds of a low measles titer, and each log-unit increase in perfluorooctanoate was associated with 2.46 (95% CI: 1.28, 4.75) times higher odds of a low mumps titer. Odds ratios for all other PFAS were elevated, but CIs included the null. Each quartile increase in the PFAS mixture was associated with 1.35 (95% CI: 0.80, 2.26) times higher odds of a low measles titer and 1.44 (95% CI: 0.78, 2.64) times higher odds of a low mumps titer. No significant associations were observed between PFAS and varicella or rubella antibodies. In stratified analyses, associations were negative among female children, except for perfluorohexane sulfonate and varicella, whereas they were positive among males.

Discussion: Some prenatal PFAS were associated with lower antibody titers among fully immunized children. The potential for immunotoxic effects of PFAS requires further investigation in a larger study, because exposure is ubiquitous globally. https://doi.org/10.1289/EHP12863.

Figure 1.

Flow diagram for selection of mother–child pairs for analysis from the Healthy Start Cohort (2009–2014).

Figure 2.

Quantile-based g-computation mixture ORs (95% CIs) and positive or negative weights for each PFAS for non–high-censored antibody titer concentrations among children ($n=143$) with completed vaccinations at early childhood visit. (A) Unadjusted measles model; (B) measles model adjusted for child’s sex and time since immunization; (C) unadjusted mumps model; (D) mumps model adjusted for child’s sex and time since immunization. Note: CI, confidence interval; PFAS, poly- and perfluoroalkyl substances; OR, odds ratio; PFDA, perfluorodecanoate; PFHxS, perfluorohexane sulfonate; PFNA, perfluorononanoate; PFOA, perfluorooctanoate; PFOS, perfluorooctane sulfonate.

Figure 3.

Quantile-based g-computation mixture linear regression beta estimates for the difference in antibody titer per quartile increase in PFAS (95% CIs) and positive or negative weights for each PFAS for continuous antibody titer concentrations among children with completed vaccinations at early childhood visit. (A) Unadjusted rubella model ($n=143$); (B) rubella model adjusted for time since immunization ($n=143$); (C) unadjusted ln(varicella) model ($n=130$); (D) ln(varicella) model adjusted for time since immunization ($n=130$). Note: CI, confidence interval; PFAS, poly- and perfluoroalkyl substances; PFDA, perfluorodecanoate; PFHxS, perfluorohexane sulfonate; PFNA, perfluorononanoate; PFOA, perfluorooctanoate; PFOS, perfluorooctane sulfonate.