Purpose Higher body mass index is linked to lower prostate specific antigen. This has given rise to concerns that prostate specific antigen may be less reliable for predicting prostate cancer among obese men. We tested the accuracy of prebiopsy prostate specific antigen for predicting prostate cancer across body mass index categories. Materials and Methods We used the REDUCE study, which tested dutasteride for prostate cancer risk reduction in men with a prostate specific antigen of 2.5 to 10.0 ng/ml and a negative pre-study biopsy. All men were required to have a biopsy at 2 and 4 years independent of prostate specific antigen. We assessed the performance of prebiopsy prostate specific antigen to predict overall and high grade prostate cancer (Gleason sum 7 or greater) in each body mass index group using AUC. Results Of 6,103 men who had a 2-year biopsy 1,646 (27%) were normal weight, 3,209 (53%) were overweight and 1,248 (20%) were obese. Mean adjusted prostate specific antigen for normal weight, overweight and obese subjects on placebo was 7.73, 7.17 and 6.79 ng/ml (p-trend=0.192), and on dutasteride 3.16, 2.93 and 2.62 ng/ml (p=0.008). AUC analysis using raw prostate specific antigen data for predicting prostate cancer ranged from 0.60 to 0.64 in the placebo arm and 0.58 to 0.66 in the dutasteride arm with no difference across body mass index categories (p-interactions a 0.212). Similar results were found for high grade prostate cancer with AUC ranging from 0.69 to 0.70 in the placebo arm and 0.65 to 0.75 in the dutasteride arm but no differences across body mass index categories (p-interactions a 0.157). Conclusions Among men with a previous negative biopsy the accuracy of prebiopsy prostate specific antigen to predict overall and high grade prostate cancer was independent of body mass index.
- prostate-specific antigen
- prostatic neoplasms