Evaluation of magnetic resonance imaging-based prostate-specific antigen density of the prostate in the diagnosis of prostate cancer.

Objective: We evaluated prostate-specific antigen (PSA) density of the prostatic volume (PSAD) estimated using transrectal ultrasonography (TRUS; TRUS-based PSAD), magnetic resonance imaging (MRI; MRI-based PSAD), and PSA density of the transition zone (TZ) volume (PSATZD) estimated using MRI (MRI-based PSATZD) in the diagnosis of prostate cancer (PCa).

Methods: One hundred and twenty patients, who were suspected to have PCa based on PSA, ranged between 4.1 and 20.0 ng/mL were enrolled in this study.

Results: The prostatic volume estimated using TRUS was smaller than the volume estimated using MRI by 11.4% in the patients with PSA levels ranging 4.1-20.0 ng/mL, 7.2% in those 4.1-10.0 ng/mL, and 15.7% in those 10.1-20.0 ng/mL, respectively. PSA levels were correlated with the prostatic volume estimated using TRUS and MRI, and TZ volume estimated using MRI in the patients without PCa; however, the level was not correlated with them in the patients with PCa. The area under the receiver operating characteristic curve of MRI-based PSAD was higher than that of TRUS-based PSAD; however, there was no statistical difference. Stepwise logistic regression analysis for the prediction of PCa by using PSA-related parameters confirmed that MRI-based PSATZD was the most significant predictor in patients with PSA levels in the range of 4.1-20.0 ng/mL (P < 0.001), the range of 4.1-10.0 ng/mL (P = 0.002), and the range of 10.1-20.0 ng/mL (P < 0.001), respectively.

Conclusions: The prostatic volume estimated using TRUS was smaller than the volume estimated using MRI. MRI-based PSATZD is the most significant predictor in the four parameters.