Exposure to per- and polyfluoroalkyl substances in residential settled dust and risk of childhood acute lymphoblastic leukemia.

Per- and polyfluoroalkyl substances (PFAS) are ubiquitous. Young children are commonly exposed to these chemicals via ingestion of settled dust. Several PFAS have been associated with cancers in adults, yet little is known about the risk in children. We investigated whether PFAS concentrations in residential dust were associated with childhood acute lymphoblastic leukemia (ALL). Vacuum bags were collected in homes of 178 children diagnosed with ALL and 204 healthy controls (age 0-7 years) residing in California (2001-2007). Dust samples were sieved and analyzed for 19 PFAS using targeted liquid chromatography mass spectrometry analysis. The effects of individual PFAS and PFAS mixtures were estimated for eight PFAS with at least 50% above the limit of quantification (LOQ) using logistic regression, G-computation, and generalized additive modeling (GAM). In the model mutually adjusting for eight PFAS, a statistically significant association was seen only for N-ethyl perfluorooctane sulfonamido acetic acid (EtFOSAA) (OR_continuous = 1.40, 95% CI = 1.05-1.86 and ${\text{OR}}_{4\mathrm{th}\mathrm{vs}.1\mathrm{st}\text{quartile}}=2.58$ , 95% CI = 1.16-5.71). Using G-computation, the eight PFAS mixture was positively associated with childhood ALL (OR = 1.60, 95% CI = 1.15-2.24), with positive weights for EtFOSAA, perfluoro-n-hexanoic acid (PFHxA), perfluoro-1-decanesulfonate (PFDS), and perfluoro-1-octanesulfonate (PFOS), and negative weights for perfluoro-1-hexanesulfonate (PFHxS) and bis(1H,1H,2H,2H-perfluorooctyl)phosphate (6:2 diPAP). Using GAM, the OR for the mixture reached a maximum of 2.24, at the highest value of log10 EtFOSAA and lowest value of log10 PFHxS. Exposure to a mixture of PFAS in settled dust was associated with an overall elevated risk of childhood ALL, with EtFOSAA and PFHxS being the main contributors to the positive and negative weights, respectively.