Room: AAPM ePoster Library
Purpose: The purpose of this study is to quantify the robustness of the three most-common modalities for breast treatment against inter-fractional positional variations of the metal port in breast tissue expanders for postmastectomy patients treated with radiation. The three modalities are 3DCRT tangents, VMAT and Tomotherapy.
Methods: For eight breast cases (200 fractions), the daily deviation of the port in the image guidance MVCT from its reference position was measured in the three cardinal directions. A clinical plan was created for all patients for each modality: 3DCRT tangents, VMAT and Tomotherapy. The dosimetric effects of the measured positional errors were evaluated for two classes of error: port displacement relative to internal anatomy and patient displacement relative to the radiation beam. Dose accumulation was done for two scenarios: daily measured errors, and for the largest measured displacement as a systematic persistent error. All scenarios were simulated in the three modalities for each patient fraction, and the dose was accumulated per patient for all fractions.
Results: For the three modalities, measured daily internal port error do not affect target coverage statistically (p=0.05). A systematic internal port error leads to changes in target coverage of the order of 1% in all modalities. Measured daily patient registration errors have small effects on target coverage in all modalities with some outliers of 18% and 20% for VMAT and Tomotherapy. A systematic patient registration error reduces target coverage by a mean of 7%, 6% and 16% for tangents, VMAT and Tomotherapy.
Conclusion: The cumulative dosimetric effect of patient registration errors relative to the beam has a greater impact on target coverage than port positional errors relative to the internal anatomy in all modalities. This indicates that overall alignment of anatomical landmarks should be prioritized at the expense of reasonable misalignment of the metal port.
Funding Support, Disclosures, and Conflict of Interest: This work was funded by NSERC Discovery Grant #2017-06253 and the QEII-GSST.