Concentrations of Per-and Polyfluoroalkyl Substances (PFAS) and Pancreatic Cancer: A Case-Control Study in New York

The aim was to investigate the concentrations of some per- and polyfluoroalkyl substances (PFAS) in patients with pancreatic cancer from New York, and to compare them with a group of controls from the general population of the United States. We selected 50 cases of pancreatic cancer from donors to the New York University Pancreatic Biorepository. Controls were selected from the 2017-18 National Health and Examination Survey sample (n = 167), matched to cases on age, sex, and race and ethnicity. Six PFAS were analyzed in serum samples using high performance liquid chromatography in conjunction with mass spectrometry. PFAS concentrations were categorized into tertiles to explore non-linear associations, and odds ratios (OR) were estimated using conditional logistic regression, adjusting by BMI. Most PFAS were not associated with pancreatic cancer risk. Serum perfluorohexane sulfonic acid (PFHxS) was associated with a decreased risk (OR for upper tertile = 0.24, 95% CI: 0.09, 0.67). In contrast, participants with the highest tertile of perfluoroundecanoic acid (PFUnDA) had a higher risk (OR = 2.60, 95% CI: 1.11, 6.09). Adjusting for BMI did not materially change the results. Study limitations include: in pancreatic cancer patients, blood used to measure PFAS was collected around the time of diagnosis; cases and controls could not be sampled from the same geographic location; slightly different laboratory methods were used to analyze PFAS in cases and controls. Most PFAS studied were not significantly associated with pancreatic cancer, except for PFHxS and PFUnDA, which exhibited opposite trends. Findings and limitations of the present study warrant further investigation with improved study designs and data on complex PFAS mixtures.