Robustness of three external beam treatment techniques against inter‐fractional positional variations of the metal port in breast tissue expanders

INTRODUCTION: Temporary breast tissue expanders contain a metal port that varies in position throughout the course of radiation treatments. The purpose of this study was to quantify the robustness of the three most common external beam treatment techniques (tangential three‐dimensional conformal radiation therapy [3DCRT], volumetric modulated arc therapy [VMAT], and helical tomotherapy) against our measured inter‐fractional positional variations of the port. METHODS: For eight breast cases, a clinical plan was created for each of the three techniques. The dosimetric effect of our previously measured inter‐fractional port errors was evaluated for two classes of error: internal port errors (IPEs) and patient registration errors (PREs). For both classes of error, daily variable and systematic errors were modeled, and their cumulative effects were compared against the originally planned doses. RESULTS: For systematic IPE, the 1%–99% range in point dose differences inside a 5‐mm target abutting the implant was the highest for tangential 3DCRT, and it was within 6% and 9% when calculated with Monte Carlo and collapsed cone calculation engines, respectively. Daily variable PRE resulted in mean changes of −3.0% and −3.5% to V(100%Rx) of the target for VMAT and tomotherapy, respectively. For nearby organs, daily variable PRE resulted in changes to V(20Gy) of the ipsilateral lung of less than 2% in all three techniques, while V(5Gy) of the heart increased by as much as 6% in VMAT and 10% in tomotherapy. CONCLUSIONS: When IPEs were modeled, dose variability was the largest in tangential 3DCRT, leading to areas of underdosage in the shadow of the port. When PREs were modeled, the target coverage and nearby organs were affected the most in VMAT and helical tomotherapy. In reality, port positional errors result from a combination of IPE and PRE, suggesting that VMAT and tomotherapy are more robust when patient registration errors are minimized, despite the presence of IPE.