2123 - Assessing Skin Dose Robustness in Breast VMAT: A Comparison of Virtual Bolus and Robust Planning
Presenter(s)
H. Kim, D. Yoo, C. S. Hong, S. Choi, and J. S. Chang; Department of Radiation Oncology, Yonsei Cancer Center, Heavy Ion Therapy Research Institute, Yonsei University College of Medicine, Seoul, Korea, Republic of (South)
Purpose/Objective(s): Skin dose robustness in breast VMAT remains unclear due to over-fluence peaks. This study examines the impact of breast position/anatomy changes on VMAT skin dose vs. 3D-CRT and assesses whether virtual bolus (VB) or robust optimization (RO) stabilizes skin dose despite geometric variations.
Materials/Methods: An anthropomorphic phantom, with/without a breast attachment, simulated breast-conserving and mastectomy cases. Four plans (3D-CRT, VMAT, VMAT+VB, VMAT+RO) were created using treatment planning system. To simulate set-up errors during treatment, the phantom was intentionally shifted in the anterior-posterior direction from -6 mm to +6 mm in 2 mm increments, resulting in seven distinct set-up scenarios for each plan. Skin dose measurements were obtained using optically stimulated luminescence dosimeters (OSLDs), which were placed at five locations around and on the nipple of the phantom to assess dose variations relative to the original setup position.
Results: Under the exact setup scenario, the measured skin dose showed no significant difference between VMAT and 3D-CRT. VMAT exhibited greater variability in measured skin dose under setup perturbations (mean = 95.41%, s = 9.47) compared to 3D-CRT (mean=99.65%, s = 3.62), indicating less consistency (p=0.003). However, this variation significantly improved with RO (mean = 99.65%, s = 5.47) (p=0.001). The effect of RO in VMAT (mean = 99.60%, s = 2.03) was more pronounced in the phantom with a breast attachment, surpassing 3D-CRT (s = 2.68) in consistency (p=0.985). In contrast, VB resulted in the largest variations (s = 12.60), regardless of breast attachment, and generally led to lower overall skin dose (96.7%).
Conclusion: VMAT is less robust than 3D-CRT in maintaining skin dose stability under breast contour variations. Among the evaluated techniques, RO proved more effective than VB in improving skin dose robustness. Further research is needed to establish a threshold for when replanning is required or RO suffices to address clinically significant contour variations in VMAT.
Abstract 2123 - Table 1: Mean and standard deviation of measured dose with setup shifts from -6 mm to 6 mm in the anterior-posterior direction, relative to the original setup position| Mean (-2 mm ~ +2 mm) | Std. Dev. (-2 mm ~ +2 mm) | Mean (-4 mm ~ +4 mm) | Std. Dev. (-4 mm ~ +4 mm) | Mean | Std. Dev. | |||
| Total | 3D CRT | 99.44 | 4.39 | 99.93 | 3.84 | 99.65 | 3.62 | |
| VMAT | 99.50 | 4.74 | 97.98 | 5.77 | 95.41 | 9.47 | ||
| VMAT+VB | 99.50 | 5.31 | 98.04 | 9.07 | 96.67 | 12.60 | ||
| VMAT+RO | 100.83 | 4.09 | 100.71 | 4.08 | 99.60 | 5.47 | ||
| Breast-Conserving Surgery Simulation (With Breast Attachment) | 3D CRT | 100.21 | 3.24 | 100.61 | 2.86 | 100.35 | 2.68 | |
| VMAT | 101.08 | 2.14 | 100.24 | 2.63 | 98.68 | 4.46 | ||
| VMAT+VB | 100.94 | 5.24 | 100.69 | 8.68 | 100.56 | 10.33 | ||
| VMAT+RO | 100.87 | 1.67 | 100.96 | 1.73 | 100.34 | 2.03 | ||
| Mastectomy Simulation (Without Breast Attachment) | 3D CRT | 98.67 | 5.37 | 99.26 | 4.60 | 98.94 | 4.29 | |
| VMAT | 97.93 | 6.11 | 95.73 | 7.11 | 92.14 | 11.85 | ||
| VMAT+VB | 98.06 | 5.25 | 95.38 | 8.89 | 92.78 | 13.60 | ||
| VMAT+RO | 100.79 | 5.70 | 100.46 | 5.57 | 98.86 | 7.47 |