Presenter(s)
J. L. Tan, E. Kaza, D. Ferguson, D. Brivio, T. C. Harris, D. A. O'Farrell, M. T. King, A. C. Smart, I. I. Franco, M. A. Dyer, R. A. Cormack, and I. Buzurovic; Department of Radiation Oncology, Brigham and Women’s Hospital, Dana-Farber Cancer Institute, Harvard Medical School, Boston, MA
Purpose/Objective(s): MRI-only treatment planning (MRTP) for interstitial high-dose-rate (HDR) gynecologic brachytherapy has been limited by poor catheter visibility on conventional anatomical MRI. We optimized a Pointwise Encoding Time Reduction with Radial Acquisition (PETRA) sequence to enhance catheter visualization and enable MRTP. This study evaluates the geometric accuracy and dosimetric performance of this MRI-only workflow compared with standard CT/MRI-based planning (CTP).
Materials/Methods: Twenty-four patients with gynecologic malignancies undergoing interstitial HDR brachytherapy were prospectively imaged with helical CT and 3T MRI (MAGNETOM Vida, Siemens Healthineers, Erlangen, Germany). MRI included 3D axial T2-SPACE and 3D PETRA (TR 3.32 ms, TE 0.07 ms, isotropic resolution 0.90 mm, matrix 416 × 416, acquisition time 5.55 min). High-risk clinical target volume (HR-CTV) and organs at risk (OAR; rectum, bladder, sigmoid) were delineated on T2-SPACE MRI. MRTP and CTP plans were independently generated in Oncentra Brachy (Elekta Brachytherapy, Netherlands). For MRTP, catheters were reconstructed exclusively on PETRA images and additionally guided by a predefined applicator pointset, which defined the fixed trajectories of the catheters through the obturator and into tissue. Reconstruction accuracy was assessed by calculating the Target Registration Error (TRE) between inactive dwell coordinates from MRTP and CTP. Dosimetric comparisons included dose–volume histogram (DVH) and equivalent dose in 2-Gy fractions (EQD2) metrics for HR-CTV D90 and OAR D2cc. Spatial dose agreement was quantified using Dice similarity coefficients (DSC) for volumes encompassed by 150% to 65% isodose levels. Statistical comparisons were performed using the Wilcoxon signed-rank test.
Results: MRTP catheter reconstruction using PETRA and the applicator pointset was successful along the entire catheter length in all patients. The mean TRE between MRTP and CTP dwell positions was 2.26 ± 1.3 mm, primarily attributable to anatomical/applicator shifts and CT–MRI registration uncertainty inherent to the conventional workflow. MRTP achieved higher HR-CTV D90 compared to CTP (EQD2: 81.8 ± 9.8 Gy vs 78.3 ± 9.0 Gy; DVH: 118.5 ± 12.4% vs 110.1 ± 11.0%; p < 0.0001), indicating improved target coverage. Differences in OAR D2cc EQD2 between MRTP and CTP were within ±0.5 Gy. DSC values exceeded 0.70 across the 125%–65% isodose levels, demonstrating strong spatial concordance between dose distributions.
Conclusion: An MRI-only workflow integrating PETRA imaging with an applicator pointset enables precise interstitial HDR gynecologic brachytherapy without CT, achieving geometric and dosimetric results comparable to conventional planning. This approach streamlines the workflow, reduces patient radiation exposure, and overcomes uncertainties inherent in CT/MRI registration, marking a significant advancement in MRI-guided brachytherapy.