Grandjean · Journal of immunotoxicology 2017 · Prospective birth cohort study · n=490

Estimated exposures to perfluorinated compounds in infancy predict attenuated vaccine antibody concentrations at age 5-years.

Cited 235 times in the scientific literature.

Level 3 - non-randomized controlled study

Non-randomized prospective birth cohort study with longitudinal follow-up

PubMed 28805477 · doi:10.1080/1547691X.2017.1360968 · record verified 2026-08-27

What was done

Researchers evaluated 490 members of a Faroese birth cohort, with 275 and 349 children attending clinical examinations and providing blood samples at 18 months and 5 years of age, respectively. PFAS concentrations were measured at birth and at both follow-up visits. Using breastfeeding duration data, serum PFAS trajectories during infancy were modeled. Serum antibody concentrations against tetanus and diphtheria vaccines were measured at age 5 years. Data from an earlier cohort born 8 years prior were included for pooled analyses.

What was found

Prenatal exposure showed inverse associations with vaccine antibody concentrations at age 5, with declines up to approximately 20% per two-fold increase in exposure, while associations with measured serum levels at 18 months and 5 years were weaker. Modeled concentrations at 3 and 6 months showed the strongest inverse associations with 5-year antibody levels, particularly for tetanus. In pooled analyses, each doubling of early infancy PFAS exposure was associated with a statistically significant 19% to 29% decrease in tetanus antibody concentrations at age 5.

Why it matters

This study highlights the first 6 months of life as a potential critical window where PFAS exposure, substantially influenced by breastfeeding, may impair long-term humoral immune responses to routine childhood vaccinations.

Limits

Infancy exposure levels at 3 and 6 months were estimated via toxicokinetic modeling rather than direct blood sampling. Loss to follow-up occurred (490 enrolled vs. 349 evaluated at age 5). Findings from this Faroese cohort may have limited generalizability given specific background exposure sources, and residual confounding remains possible in observational designs.

Cited by