Associations of drinking water per- and polyfluoroalkyl substances with cause-specific mortality in men with prostate cancer
Background: Over 3.5 million US men are living with prostate cancer (PCa). Certain per- and polyfluoroalkyl substances (PFAS) are carcinogenic and impair cardiometabolic health. Due to comorbidities and treatment, men with PCa may be uniquely susceptible to PFAS. Objectives: To develop and validate a model for drinking water perfluorooctanoic acid (PFOA), perfluorooctane sulfonate (PFOS), and perfluorononanoic acid (PFNA), and study associations of each PFAS with all-cause, PCa, and cardiovascular disease (CVD) mortality in PCa survivors. Methods: We conducted a cohort study of 51,289 men with PCa diagnosed from 2006 to 2017 and followed through 2024 from the New Jersey (NJ) cancer registry. PFAS monitoring data (n=26,161 results) were obtained from NJ government evaluations and Community Water Systems (CWS) compliance reports and were used to estimate PFAS concentrations using a two-phase XGBoost model with 67 predictors of PFAS contamination across 566 CWS. PFAS predictions between 2006-2024 were linked to participant address histories. Adjusted hazard ratios (aHR) and 95% confidence intervals (CI) were estimated from Cox models with sociodemographic, clinical and time-varying neighborhood covariates. Results: There were 19,749 deaths, including 3,778 PCa deaths, and 4,972 CVD deaths over 466,843 person-years. Higher PFOA exposure was associated with all-cause mortality in certain quintiles (aHRQ2vsQ1: 1.07, 95% CI: 1.07-1.12; aHRQ3vsQ1: 1.06, 95% CI: 1.01-1.11) and with CVD mortality (aHRQ5vsQ1: 1.12, 95% CI: 1.01-1.24). Higher PFNA exposure was suggestively associated with all-cause mortality (aHR≥3.8ng/L vs <2ng/L: 1.22, 95% CI: 0.98-1.50) and PCa mortality (aHR≥3.8ng/L vs <2ng/L: 1.42, 95% CI: 0.96-2.10). Associations were strengthened when applying five-year lags. For example, higher PFOA (aHRQ5vsQ1: 1.15, 95% CI: 1.06-1.25), PFOS (aHRQ5vsQ1: 1.16, 95% CI: 1.07-1.25) and PFNA (aHR≥3.8ng/L vs <2ng/L: 1.57, 95% CI: 1.06-2.31) were associated with higher all-cause mortality. Discussion: Exposure to CWS-level PFAS was associated with higher mortality in PCa survivors. Replication is needed to confirm findings.
Authors
- Antoinette M. Stroup (ORCID: https://orcid.org/0000-0003-3341-4018)
- Jaime E. Hart (ORCID: https://orcid.org/0000-0002-0826-1163)
- Panos G. Georgopoulos (ORCID: https://orcid.org/0000-0003-2119-0301)
- Elisa Victoria Bandera (ORCID: https://orcid.org/0000-0002-8789-2755)
- Robert J. Laumbach (ORCID: https://orcid.org/0000-0002-4350-3674)
- Emily S. Barrett (ORCID: https://orcid.org/0000-0001-9463-524X)
- Saum B. Ghodoussipour (ORCID: https://orcid.org/0000-0001-8901-0210)
- Verónica M. Vieira (ORCID: https://orcid.org/0000-0001-7153-4606)
- Gerald Harris (ORCID: https://orcid.org/0000-0003-3633-6078)
- Stefanie A. Joseph (ORCID: https://orcid.org/0000-0002-6482-6659)
- Mike Z. He (ORCID: https://orcid.org/0000-0003-2357-3883)
- Megan R. Shanahan (ORCID: https://orcid.org/0009-0008-7423-4674)
- Hari Subramaniam Iyer (ORCID: https://orcid.org/0000-0002-7596-9049)
- Julianne Varga (ORCID: https://orcid.org/0009-0005-1142-7741)
Institutions
- Rutgers, The State University of New Jersey (US)
- Brigham and Women's Hospital (US)
- New Jersey Department of Health (US)
- Harvard University (US)
- University of California, Irvine (US)
- Environmental and Occupational Health Sciences Institute (US)
- Rutgers Cancer Institute (US)
- Icahn School of Medicine at Mount Sinai (US)
Publication Details
- Journal
- medRxiv
- Published
- 2026-11-12
- DOI
- https://doi.org/10.64898/2026.11.10.26367726
- Primary Topic
- Per- and polyfluoroalkyl substances research
- Type
- preprint
Funders
- National Institute of Environmental Health Sciences