University of Washington-Fred Hutchinson Cancer Center Poulsbo, WA
R. A. Hsi1,2, M. D. Greer3, A. Pham2, H. Parsai1, M. Montague4, S. Reith4, J. Bell1, and R. Mangibin1; 1EvergreenHealth Department of Radiation Oncology, Kirkland, WA, 2University of Washington, Department of Radiation Oncology, Seattle, WA, 3Department of Radiation Oncology, University of Arizona Cancer Center, Tucson, AZ, 4Fred Hutchinson Cancer Center Peninsula, Poulsbo, WA
Purpose/Objective(s): To assess the performance of multiparametric (mp) MRI in identification of intraprostatic tumor deposits in patients previously treated with local radiation therapy (RT) using a systematic and targeted MR-guided transperineal prostate biopsy technique. Materials/
Methods: Thirty-six patients with a rising PSA and negative metastatic work up after prior external beam RT, brachytherapy, or combined external beam RT and brachytherapy for localized prostate cancer underwent a combined systematic and targeted MRI-guided prostate biopsy using a transperineal approach. For each patient, a pre-biopsy mpMRI scan was obtained and imported into a prostate biopsy planning system. Transverse T2-weighted images were then reoriented from the supine to dorsal lithotomy position. Dividing the prostate into an apical and base section, a systematic array of transperineal biopsies spaced approximately 10mm apart was planned with additional biopsies targeting any mpMRI-identified PI-RADS 3, 4, or 5 lesion. Biopsy procedures were carried out under general anesthesia in the dorsal lithotomy position using a transrectal ultrasound with stepper-stabilizer and template grid. Matching of the planning mpMRI images to live ultrasound images was achieved using the template grid as a reference. Results: All patients successfully underwent biopsy without post-procedure urinary obstruction or infection. The median prostate volume was 32cc (interquartile range IQR 24-46). The median PSA prior to biopsy was 4.38 ng/mL (IQR 2.77-7.63). The median number of biopsy specimens obtained per patient was 20 (IQR 17-24). A total of 39 PI-RADS lesions were identified in 36 patients. Overall, the positive predictive value (PPV) of any PI-RADS 3-5 lesion for prostate cancer (based on pathologic confirmation) was 85% (33/39). The individual PPV of PI-RADS 3, 4 and 5 lesions for any cancer were 50% (3/6), 87% (20/23), and 100% (10/10), respectively. The PPV of PI-RADS 3, 4 and 5 lesions for GS=7 disease were 17% (1/6), 78% (18/23) and 100% (10/10). Thirty-nine percent (26/36) of patients harbored mpMRI-unidentified prostate cancer. Twenty-five percent (9/36) of patients harbored mpMRI-unidentified GS=7 disease. Conclusion: Using prostate mpMRI to identify intraprostatic tumor deposits after local radiation therapy results in a high PPV, particularly for PI-RADS 4 and 5 lesions. However, given the high rate of mpMRI-unidentified prostate cancer, the addition of systematic transperineal prostate biopsy information appears necessary to most accurately identify all intraprostatic tumor deposits.
