A case-cohort study of the association between per- and polyfluoroalkyl substances (PFAS) and breast cancer among participants in the American Cancer Society's Cancer Prevention Study-II.

Background: Per- and polyfluoroalkyl substances (PFAS) are hypothesized to increase breast cancer risk; findings from prior studies are inconsistent, with many not considering subtypes.

Objectives: To evaluate associations between serum PFAS concentrations and breast cancer incidence, overall and by selected characteristics.

Methods: We conducted a case-cohort study among Cancer Prevention Study-II LifeLink Cohort female participants (mean age = 68) who developed breast cancer (n = 781) during follow-up (1998-2015) or were in a randomly selected subcohort (n = 498). PFAS concentrations [perfluorooctanoic acid (PFOA) perfluorooctane sulfonate (PFOS), perfluorohexane sulfonic acid (PFHxS), and perfluorononanoic acid (PFNA)] were measured in blood samples collected at baseline (1998-2001). Multivariable Cox proportional hazards models with Prentice weighting were used to estimate hazard ratios (HR) and 95 % confidence intervals (CI) for associations between PFAS concentrations and breast cancer risk.

Results: PFAS concentrations were not associated with breast cancer overall but demonstrated heterogeneity by participant characteristics. For example, associations for the highest compared to lowest PFNA quartiles were in opposite directions for estrogen receptor positive (ER+) HR (95 % CI) = 0.88 (0.61-1.27) and estrogen receptor negative (ER^-) tumors (HR (95 % CI) = 2.01, 95 % CI: (0.91-4.42). Associations also varied by family history of breast cancer, age at blood collection, menopausal hormone therapy (HRT) use, and parity status.

Conclusion: PFAS were not associated with overall breast cancer risk in this population of predominantly older women; however, the findings suggest that certain subgroups may be more susceptible to PFAS-related risk. These preliminary results warrant replication in future prospective studies.