Prostate volume and prostate-specific antigen levels in men enrolled in a large screening trial

Objectives. Prostate volume correlates both with prostate-specific antigen (PSA) values and with the presence of benign prostatic hyperplasia (BPH). Here we examine the relationship between prostate volume and PSA level in a large, geographically diverse sample of men undergoing prostate cancer screening. Methods. We followed 35,323 men enrolled in the Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial. Each man was screened with digital rectal examination (DRE) and PSA levels for up to 4 years. Prostate volume was estimated by DRE performed by trained examiners at the PLCO sites. Linear and logistic regression was used to assess the effect of prostate volume and age on PSA levels. Regression coefficients were adjusted for the effect of prostate volume measurement error. Results. Prostate volume estimated by DRE showed considerable measurement error. Averaging volume over screening visits and accounting for examiner bias greatly reduced this error. Linear regression analysis showed a slope of 0.030/cm(3) of log PSA on prostate volume when correcting for measurement error (95% confidence interval [CI], 0.029 to 0.031). Age was independently associated with (log) PSA, with a slope of 0.022 per year. Logistic regression analysis of the risk of having an elevated PSA value (exceeding 4 ng/mL) showed an odds ratio of 1.9 (95% CI, 1.8 to 2.0) associated with a 10 cm(3) increase in prostate volume. The correlation of log PSA and prostate volume was 0.37. Prostate volume was not correlated with body mass index and showed weak correlation (r = 0.14) with age. Conclusions. Prostate volume and age are independently associated with increased PSA levels in a population of men undergoing screening.