Lessons learned using an MRI-only workflow during high-dose-rate brachytherapy for prostate cancer.

PURPOSE: We report clinical observations of a technique using an MRI-only workflow for catheter insertion and treatment planning in patients receiving standard-care high-dose-rate brachytherapy before external beam radiotherapy for prostate cancer. METHODS AND MATERIALS: Forty patients with intermediate or high-risk prostate cancer were enrolled on a prospective clinical trial approved by our institution's research ethics board. Multiparametric MRI with stereotactic navigation was used to guide insertion of brachytherapy catheters, followed by MRI-based treatment planning. RESULTS: Sixty-two implants were performed. Median catheter insertion + imaging time was 100 minutes, and overall anesthesia time was 4.0 hours (range, 2.1-6.9 hours). MRI at the time of brachytherapy restaged 14 patients (35%) who were found to have a higher stage of disease. In 6 patients, this translated in directed insertion of brachytherapy catheters outside the prostate boundary (extracapsular disease [n = 2] or seminal vesicle invasion [n = 4]). Most patients (80%) had gross tumor visible on MRI, which influenced catheter insertion and treatment planning. MRI depicted postimplant anatomic boundaries clearly, with the exception of the apical prostate which was blurred by trauma after catheter insertion. Conventional dose-planning objectives for the rectum (V75 < 1.0 cc) were difficult to achieve, but toxicities were low (acute grade ≥ 2 genitourinary = 20%, late grade ≥ 2 genitourinary = 15%, and late grade ≥ 2 gastrointestinal = 7%). Urethral trauma visualized on MRI led to two transient Grade 3 events. CONCLUSIONS: Despite a standard-care approach, MRI acquired throughout the procedure altered catheter insertion and dose-planning strategies. An MRI-only workflow is feasible but must be streamlined for broader acceptance.

MR-guided prostate biopsy for planning of focal salvage after radiation therapy.

PURPOSE: To determine if the integration of diagnostic magnetic resonance (MR) imaging and MR-guided biopsy would improve target delineation for focal salvage therapy in men with prostate cancer. MATERIALS AND METHODS: Between September 2008 and March 2011, 30 men with biochemical failure after radiation therapy for prostate cancer provided written informed consent and were enrolled in a prospective clinical trial approved by the institutional research ethics board. An integrated diagnostic MR imaging and interventional biopsy procedure was performed with a 1.5-T MR imager by using a prototype table and stereotactic transperineal template. Multiparametric MR imaging (T2-weighted, dynamic contrast material-enhanced, and diffusion-weighted sequences) was followed by targeted biopsy of suspicious regions and systematic sextant sampling. Biopsy needle locations were imaged and registered to diagnostic images. Two observers blinded to clinical data and the results of prior imaging studies delineated tumor boundaries. Area under the receiver operating characteristic curve (Az) was calculated based on generalized linear models by using biopsy as the reference standard to distinguish benign from malignant lesions. RESULTS: Twenty-eight patients were analyzed. Most patients (n = 22) had local recurrence, with 82% (18 of 22) having unifocal disease. When multiparametric volumes from two observers were combined, it increased the apparent overall tumor volume by 30%; however, volumes remained small (mean, 2.9 mL; range, 0.5-8.3 mL). Tumor target boundaries differed between T2-weighted, dynamic contrast-enhanced, and diffusion-weighted sequences (mean Dice coefficient, 0.13-0.35). Diagnostic accuracy in the identification of tumors improved with a multiparametric approach versus a strictly T2-weighted or dynamic contrast-enhanced approach through an improvement in sensitivity (observer 1, 0.65 vs 0.35 and 0.44, respectively; observer 2, 0.82 vs 0.64 and 0.53, respectively; P < .05) and improved further with a 5-mm expansion margin (Az = 0.85 vs 0.91 for observer 2). After excluding three patients with fewer than six informative biopsy cores and six patients with inadequately stained margins, MR-guided biopsy enabled more accurate delineation of the tumor target volume be means of exclusion of false-positive results in 26% (five of 19 patients), false-negative results in 11% (two of 19 patients) and by guiding extension of tumor boundaries in 16% (three of 19 patients). CONCLUSION: The integration of guided biopsy with diagnostic MR imaging is feasible and alters delineation of the tumor target boundary in a substantial proportion of patients considering focal salvage.