2774 - Retrospective Replanning of EBRT Regimen for Intended Avoidance of Bulboclitoris and Vaginal Canal for GYN Cancer Patients
Presenter(s)
A. Jones1, D. Neish1, J. P. Chino2, K. Light3, S. G. Meltsner4, O. I. Craciunescu4, and D. N. Ayala-Peacock5; 1Duke University, Durham, NC, 2Department of Radiation Oncology, Duke University Medical Center, Durham, NC, 3Duke University Medical Center, Durham, NC, 4Duke University Medical Center, Department of Radiation Oncology, Durham, NC, 5Vanderbilt University Medical Center, Nashville, TN
Purpose/Objective(s): This study aims to replan external beam radiation therapy (EBRT) treatments with the goal of minimizing dose to the bulboclitoris (BC) and vaginal canal (VC) while preserving planning target volume (PTV) coverage and similar sparing of standard of care organ-at-risk (OAR) (bladder, rectum, sigmoid, bowel).
Materials/Methods: This retrospective study performed under an IRB approved protocol included female patients treated with EBRT for several pelvic cancers (uterine, vaginal, cervical, urethral, vulvar). The BC and VC were contoured on T2-weighted MRI and the contours were transferred to planning CT via rigid registration using a technology company's treatment planning system. Additional structures were created to account for the volume of BC and VC that do not overlap with the PTV. Patients were replanned to spare these structures while maintaining PTV coverage and OAR sparing with the same technique used for the clinical plan (IMRT or VMAT). A software workflow extracted several dose metrics including the mean dose, V40, V30, and V20 Gy for the BC, VC, BC-PTV, and VC-PTV. Statistical analysis was performed on dose metrics for clinical and replanned treatments using Wilcoxon signed-rank test. For BC and BC-PTV, data was stratified by cancer type (vulvo-vaginal vs. non-vulvo-vaginal) and percent overlap with PTV (<1%, =1% & <35%, =35%). Median values were calculated for each metric and reported by stratum. A two-sided p value <0.05 was used for all tests.
Results: Thirty patients were included in the analysis: 15 uterine, 6 vaginal, 7 cervical, 1 vulvar, and 1 urethral. Due to >90% of patients having substantial VC overlap with the PTV, limited sparing was achieved. For the BC, significant statistical differences were seen post-replan for all sites, when separating the non-vulvovaginal patients, and when BC/PTV overlap was < 35%. For BC, V30 was identified as a potential dose metric constraint.
Conclusion: Intentional dose reduction to the BC can be achieved without losing PTV coverage for a variety of pelvic cancers and may inform the design of dose constraints in prospective studies. The high overlap between VC and PTV for these patients leads to minimal dose sparing, regardless of cancer type.
Abstract 2774 - Table 1
|
|
| Mean Dose (Gy) | V30 Gy (%) | ||
|
|
| Median Diff. | P-value | Median Diff. | P-value |
| Bulboclitoris | All (N=30) | -2.96 | <0.001 | -6.95 | <0.001 |
| Non-vulvovaginal (N = 23) | -2.96 | <0.001 | -6.96 | <0.001 | |
| Vulvovaginal (N = 7) | -1.68 | 0.052 | -2.76 | 0.059 | |
| PTV overlap <1% (N = 10) | -3.1 | 0.006 | -4.96 | 0.009 | |
| PTV overlap 1-35% (N = 10) | -3.12 | 0.006 | -15.79 | 0.009 | |
| PTV overlap 35-100% (N = 10) | -0.56 | 0.103 | -1.35 | 0.059 | |
| Bulboclitoris-PTV | All (N=30) | -3.48 | <0.001 | -7.8 | <0.001 |
| Non-vulvovaginal (N = 23) | -3.48 | <0.001 | -8.36 | <0.001 | |
| Vulvovaginal (N = 7) | -2.72 | 0.059 | -5.28 | 0.059 | |
| PTV overlap <1% (N = 10) | -3.10 | 0.006 | -3.95 | 0.009 | |
| PTV overlap 1-35% (N = 10) | -3.99 | 0.009 | -12.19 | 0.014 | |
| PTV overlap 35-100% (N = 10) | -3.56 | 0.019 | -16.92 | 0.024 | |