Negative predictive value for cancer in patients with "gray-zone" PSA level and prior negative biopsy: preliminary results with multiparametric 3.0 Tesla MR.

PURPOSE: To investigate the negative predictive value (NPV) for malignancy of 3.0 Tesla (T) MRI in patients with "gray zone" PSA level and prior negative biopsies. MATERIALS AND METHODS: We enrolled 26 patients with PSA level between 2.5 and 10 ng/mL and no cancer at previous biopsies. Examinations were performed on a 3.0T system using T2-weighted imaging, diffusion-weighted imaging, spectroscopy, and postcontrast dynamic study. A regional scheme was used to record MRI findings and to perform subsequent transrectal-ultrasonography-guided biopsy. Based on the matching between imaging and biopsy findings we estimated MRI predictive values, sensitivity, specificity and accuracy on a per-patient and per-region basis. RESULTS: On a per patient basis, MRI had five true-positive (5/26; 19.2%), eight true-negative (8/26; 30.8%) and no false-negative cases, corresponding to a NPV and sensitivity of 100% each. Thirteen patients were assessed as false-positive cases (13/26; 50.0%) (specificity of 38.1%). Five of them (5/26; 19.2%) showed high-grade prostatic intraepithelial neoplasia (HGPIN) and atypical small-acinar proliferation (ASAP) at biopsy. T2-weighted imaging alone showed per-region NPV (96.8%). CONCLUSION: Because of the high NPV, MRI had the potential to avoid unnecessary biopsy in approximately one-third of "gray-zone" patients with a negative examination. Additionally, MRI was useful to address to biopsy more than one-third of patients with cancer or high-risk lesions as HGPIN and ASAP.