Prostate Cancer and PFOA

To the Editor: Recently, there were two articles published in Journal of Occupational and Environmental Medicine that have examined issues surrounding prostate cancer and perfluorooctanoic acid (PFOA—measured in serum as perfluorooctanoate).1,2 We offer this letter to identify important additional information that was not addressed in these articles. Ducatman et al1 examined serum PFOA and prostate-specific antigen (PSA) test results among 25,412 participants (≥20 years) from the cross-sectional C8 Health Project study conducted in 2005 to 2006 in the mid-Ohio river area near the DuPont Washington Works plant. Ducatman et al did not observe an association between PFOA and PSA levels. As part of the series of C8 Science Panel findings related to the geographic area, Steenland et al2 reported on the incidence of 18 diseases among 3713 workers at this Washington Works plant where PFOA was used as a processing aid in the polymerization of tetrafluoroethylene. Steenland et al suggested there was a modest evidence of a positive exposure trend for prostate cancer on the basis of 129 cases. Relative risks by increasing quartiles of modeled cumulative serum PFOA (ng/mL years) were 1.00 (reference), 1.81 (95% confidence interval [CI], 0.69 to 4.78), 2.45 (95% CI, 0.96 to 6.25), and 1.88 (95% CI, 0.72 to 4.88). Both Ducatman et al and Steenland et al referred to the same 3M Company cohort mortality study (Cottage Grove, Minnesota) as evidence for a positive association between PFOA and prostate cancer mortality.3 Nevertheless, Lundin et al cautioned about their internal analyses because of the few prostate cancer deaths (n = 16) and that the exposed categories in the stratum-specific standardized mortality ratios were modestly above unity whereas the least exposed were markedly below. Sociodemographic differences may have played a role.3,4 Most importantly, prostate cancer mortality is a poor reflection of its incidence. Unfortunately, Ducatman et al and Steenland et al did not mention the prostate cancer incidence analysis of these 3M employees.4 This study combined two cohorts resulting in 9027 employees who were either at the Cottage Grove plant that manufactured PFOA (ammonium salt) or in a nearby plant with similar production workers who were not exposed to PFOA on the job. On the basis of state cancer registry data, a total of 431 prostate cancer cases (188 employed at the Cottage Grove plant) were diagnosed between 1988 and 2008. A job, department, and task-based exposure matrix was developed regarding PFOA with historic area and personal air monitoring data.4 No association was observed between quartiles of cumulative PFOA exposure (μg/m3 years) and prostate cancer incidence. The hazard ratios by increasing cumulative PFOA exposure were 1.00 (reference), 0.80 (95% CI, 0.57 to 1.11), 0.85 (95% CI, 0.61 to 1.19), 0.89 (95% CI, 0.66 to 1.21), and 1.11 (95% CI, 0.82 to 1.49). Serum PFOA concentrations have been measured since the mid-1990s among these PFOA production workers and known to be higher compared with nonoccupational populations.5,6 Finally, Ducatman et al and Steenland et al did not mention the International Agency for Research on Cancer June 2014 workgroup evaluation for the human carcinogenicity of PFOA.7 PFOA was classified as a "possible" (group 2B) human carcinogen on the basis of limited experimental and human (testicular, kidney) evidence. Prostate cancer was not a basis for this classification. Geary W. Olsen, DVM, PhD 3M Company, Medical Department St Paul, Minn. Carol A. Ley, MD, MPH 3M Company, Medical Department, St Paul, Minn.