Room: AAPM ePoster Library
Purpose: Stereotactic body radiation therapy (SBRT) for prostate cancer is becoming more prevalent due to its comparable effectiveness and shorter treatment time. However, its associated GU and GI toxicities still have room for improvements. Among efforts on lowering toxicity, urethra sparing is essential but remains challenging, mainly due to nontrivial prostatic urethra delineation. In this study, we sought to optimize urethra MRI sequences with a MR-guided RT (MRgRT) system and compared the urethra contours in different workflows.
Methods: Seven prostate cancer patients were scanned on a 0.35T MRgRT system using T2-weighted (T2w) 3D HASTE and 3D TSE sequences with acquisition times between 7-8 minutes. Scan parameters common to both were: resolution=1.5 mm isotropic, TE=250ms, and TR=1900ms. HASTE and TSE were acquired with 6 and 4 averages, respectively. A radiation oncologist contoured the urethra on the patient’s planning CT, registered pre-treatment diagnostic MRI, and both MRgRT MRIs. The three image sets were acquired on different days but MRgRT MRIs and CTs were acquired with patients in treatment position. The oncologist qualitatively scored the urethra visibility on a 4-point scale. Using MRgRT planning as baseline, urethra contours were quantitatively evaluated relative to the urethra contour on 3D HASTE using Hausdorff distance (HD), mean-distance-to-agreement (MDA), and DICE coefficient.
Results: Qualitatively, HASTE scored best for urethra visibility with 3s for all patients. The average HD/MDA/DICE for CT, CT + diagnostic MRI, and 3D TSE contours were 14.68±3.20mm/5.30±2.11mm/0.12 ± 0.11, 11.47± 4.15mm/3.98±1.99mm/0.15 ± 0.18, and 6.64±2.01mm/1.47±0.59mm/0.45±0.15, respectively.
Conclusion: The contours’ qualitative scores showed optimized urethra MRI sequence was superior for urethra visualization. Quantitatively, the contours’ significant variability demonstrated using the right tools is crucial for urethra delineation and sparing. Urethra contouring uncertainty introduced by cross-modality registration and sub-optimal imaging contrast may lead to significant treatment error when cold dose area was created for protecting urethra.