Objectives: Data have suggested benign prostatic hyperplasia, and not cancer, as the major reason for elevated prostate-specific antigen (PSA) values between 2.0 and 9.0 ng/mL. If this hypothesis were correct, within these ranges, a smaller prostate volume would be a stronger predictor of cancer than the PSA level itself (the relative contribution from cancer is greater in smaller glands).
Methods: We examined our institutional data set of transrectal ultrasound-guided procedures from 2000 to 2003. We studied patients who presented for their first prostate biopsy with a PSA level of 2.0 to 9.0 ng/mL. The indications for biopsy were elevated age-specific PSA level or abnormal digital rectal examination findings. Other covariates included patient age, abnormal transrectal ultrasound findings, transrectal ultrasound volume, and biopsy sampling scheme. Univariate analyses were used to assess the association between each variable and cancer diagnosis. Multivariate logistic regression modeling was then used to determine the adjusted risk factors for cancer at biopsy.
Results: On univariate analyses, all measured covariates were predictive of cancer. On multivariate modeling, the significant risk factors (in order of strength) for positive biopsy findings were smaller prostate volume (odds ratio [OR] 0.26, P <0.001), increasing age (OR 1.72, P <0.001), increasing PSA (OR 1.64, P <0.001), and the presence of hypoechoic lesions (OR 2.42, P <0.001).
Conclusions: When the PSA level is in the 2.0 to 9.0 ng/mL range, a smaller prostate volume is the strongest predictor of cancer detection. These data support previous studies suggesting the amount of benign prostatic hyperplasia, and not cancer, as the major factor responsible for elevated PSA.