Methods Cases were 1933 men with histologically-confirmed incident prostate cancer (PCa), aged ≤75 years and diagnosed between 2005 and 2009. Concurrently, 1994 population controls from the same residential area and age distribution were randomly selected from electoral lists. Detailed lifestyle and work histories were collected during in-person interviews. Chemists used job specific information from questionnaires to provide semi-quantitative evaluations of concentration and frequency (% of work-day exposed) of exposure to chemical agents, including toxic metals, and a measure of confidence in the evaluation. Logistic regression was used to estimate odds ratios (OR) and 95% confidence intervals for PCa risk associated with exposure to Cr, CrVI, Ni and Cd.

Results Ever exposure to Cr (9%), CrVI (7%), Ni (5%) or Cd (2%) was similar among cases and controls. The risk for non-aggressive PCa (Gleason score < 7 or [3+4]) did not vary according to duration of exposure to any level or to substantial level of Cr, CrVI, Ni or Cd. However, the risk for aggressive PCa (Gleason score > 7 or [4+3]) was significantly increased with ≥25 years of exposure to any level of Cr (OR = 1.68 [1.05–2.69]) or Ni (OR = 2.25 [1.28–3.94]), and to a substantial level of Cr (OR = 2.77 [1.34–5.76]), CrVI (OR = 2.69 [1.28–5.66]) or Ni (OR = 3.80 [1.36–10.57]). The positive associations observed with Cr exposure disappeared when excluding subjects ever exposed to Ni.

Conclusion Our results suggest that long exposure to nickel, or to both nickel and chromium, may increase the risk for aggressive PCa.