5060 Background: The Nutritional Prevention of Cancer Trial demonstrated a 52% lower incidence of prostate cancer in its selenium (Se) arm. Consequently, this trial was designed to assess if Se supplementation would reduce the progression of prostate cancer in subjects on active surveillance for their disease. Methods: A Phase 2b randomized, double-blind placebo-controlled clinical trial was conducted in men with localized prostate cancer who had elected to forego therapy for active surveillance. Subjects <85 years of age with biopsy proven non- metastatic prostate cancer, prostate specific antigen (PSA) levels <50ng/ml, Gleason score <8, no other treatment for prostate cancer, and life expectancy ≥3 years were recruited. A total of 140 men were randomized to placebo (n=46), 200μg/day (n=47) or 800μg/day (n=47) Se p.o. (as selenized yeast) and followed every 3 months for up to 5 years. The primary endpoint was PSA velocity, a surrogate marker of prostate cancer progression. Serum levels of PSA and Se were determined at baseline and at 3-month intervals thereafter. Variance-stabilizing transformation ln(PSA+1) was carried out and PSA velocity (slope of ln(PSA+1)) was estimated using mixed effects regression models. Results: Following adjustment for age, body mass index, serum selenium levels, pack-years of smoking at baseline, duration on study, race, PSA analysis method, and Gleason score; percent changes in PSA velocity (90% CI) for 200 μg/day and 800 μg/day treatment groups vs. placebo were -2.85% (-7.19, 1.70) and -0.70% (-5.16, 3.98), respectively. Estimated mean PSA velocities and PSA doubling times for placebo, 200 μg/day and 800 μg/day treatment groups were 0.09, 0.07, and 0.09 ng/mL/yr; and 4.92, 5.86 and 5.34 years, respectively. Upon secondary analyses, PSA velocity was statistically significantly higher in smokers (vs. non-smokers) (percent change in PSA velocity and 90% CI: 0.09 (0.03, 0.14)). Conclusions: Selenium supplementation did not significantly affect PSA velocity in subjects with non-aggressive disease. Longitudinal analyses also suggest that active surveillance was an appropriate choice for these subjects. Cigarette smoking may influence prostate cancer progression and needs further investigation. This work was supported by PHS CA079080 and CA023074. No significant financial relationships to disclose.
Read full abstract