2012 - Treatment Robustness Assessment of Postoperative Whole Breast Radiotherapy with Locoregional Nodal (LN) Irradiation
Presenter(s)
G. Atkins1, L. Bell1, B. Shen1, A. W. LaVigne1, A. Liu2, A. P. Gurses3, F. Asrari1, and S. Han-Oh1; 1Department of Radiation Oncology and Molecular Radiation Sciences, Johns Hopkins University School of Medicine, Baltimore, MD, 2Department of Computer Science, Johns Hopkins University, Baltimore, MD, 3Armstrong Institute for Patient Safety and Quality, Johns Hopkins University School of Medicine, Baltimore, MD
Purpose/Objective(s): Postoperative whole breast radiotherapy (RT) can be designed using various treatment techniques, including 3D-conformal field-in-field (3D-FIF) technique, VMAT, or a combination of both (Hybrid). Multiple studies report that each technique generally achieves desirable target coverage and OAR sparing, each with its own advantages and disadvantages. However, the robustness of each technique against positioning uncertainties for intended dose delivery is not clearly understood. We sought to investigate the robustness of these 3 techniques to help clinicians determine the best treatment approach for patients.
Materials/Methods: Ten left-sided patients previously treated with 3D-FIF breast + LN RT were replanned with VMAT & Hybrid techniques. The robustness against positioning deviation in 1 degree of freedom (1-DOF) was evaluated by shifting the planning CT with our institution’s 1.5-cm translation and 3-degree rotation tolerance limits for patient repositioning. The robustness against 6-DOF deviations was studied using 55 pre-treatment CBCTs, which were converted to corrected CBCTs and virtual CTs for treatment dose calculations. The dosimetric effects of positioning deviations were analyzed on clinical target volumes, including chest wall (CW), axillary LNs (Ax_LNs), supraclavicular LNs (SCV_LNs), internal mammary LNs (IMNs), and organs-at-risk (OARs).
Results: With 1-DOF shifts, target coverage (V95%) of CW, Ax_LNs, SCV_LNs, and IMNs was seven times more affected by translational deviations than rotational ones for all 3 techniques. 3D-FIF was the most robust against translational shifts, showing a 10% reduction in V95% compared to the treatment plans, while Hybrid and VMAT showed reductions of 30% and 32%, respectively. However, 3D-FIF V95% was more sensitive to yaw, whereas Hybrid and VMAT were robust against all rotational directions. The average 6-DOF deviations from the pre-treatment CBCTs were 0.3 ± 0.2 cm in LR, 0.4 ± 0.3 cm in SI, 0.4 ± 0.3 cm in AP, 1.6 ± 1.1 deg in pitch, 1.1 ± 0.8 deg in roll, and 1.3 ± 1.0 deg in yaw. With these modest shifts within patient repositioning limits, all 3 techniques did not show significant differences in CW and Ax_LNs V95% (< 2%) from the treatment plans. 3D-FIF was the most stable for SCV_LNs and IMNs coverage with < 3% deviation, while Hybrid and VMAT showed decreases of 9.1 ± 6.8% and 7.9 ± 6.8% for SCV_LNs, and 11.6 ± 12.2% and 17.8 ± 17.3% for IMNs, respectively. All 3 techniques kept OAR doses below treatment planning limits, but Hybrid and VMAT showed less deviation from the plans compared to 3D-FIF.
Conclusion: This robustness assessment demonstrates different sensitivities of 3 standard breast planning techniques to patient setup uncertainty, suggesting the utility of technique-specific or patient-specific setup tolerance limits to minimize coverage variation.