Concerns about the toxicity of selenium has limited the doses used in chemoprevention. Based on previous studies, intakes of 400 microg/day and plasma selenium of 1000 ng/ml (Dietary Reference Intakes, Academy Press, New York, 2000, p. 384) were established as the no observed adverse effect level (NOAEL). This investigation summarizes the plasma response and toxicity reports from 24 men with biopsy-proven prostate cancer who were randomized to either 1600 or 3200 microg/day of selenized yeast as part of a controlled clinical trial testing selenium as a chemopreventive agent for prostate cancer progression. Subjects were on these doses for averages of almost 12 months. Plasma selenium levels were monitored throughout the course of follow-up. Symptoms of selenium toxicity were assessed by patient interview with specific questions regarding breath, hair and nail changes. Several liver and kidney function tests and hematology were measured at 6-month intervals. 8 subjects were randomized to the 1600 microg/day and 16 to the 3200 microg/day group. The mean plasma selenium levels achieved with supplementation were 492.2 ng/ml (SD = 188.3) and 639.7 ng/ml (SD = 490.7) for the 1600 and 3200 microg/ day doses, respectively. The 3200 microg/day group reported more selenium-related side effects. Blood chemistry and hematology results were all within normal limits for both treatment groups. More subjects on 3200 microg/day reported symptoms of selenium toxicity; however, these reports did not correspond to peaks in plasma selenium levels. We observed no obvious selenium-related serious toxicities. As selenium is used in more chemoprevention and therapeutic settings, additional information on selenium species, sequestration of selenium in specific organs, excretion, and toxicities is needed.