A Gated IMRT Study Based On Comparative 2D and 3D Dose Measurements
P Dvorak*, S Stevens, M McQuaid, S Dawes, A Richmond, The London Clinic, London, UKMO-G-213AB-2 Monday 5:15:00 PM - 6:00:00 PM Room: 213AB
Purpose: To perform a quantitative study of Varian RPM gating modes and parameters on a gated IMRT of esophagus cancer case using both 2D and 3D dose measurements.
Methods: A 4DCT scan reconstructed into a 10-phase series was acquired. For each of 7 IMRT fields, the target trajectory, due to respiratory and also cardiac motion, perpendicular to the beam axis was recorded. An alternative set of smoothed trajectories was also processed to form closed hysteresis-like curves. Each trajectory was replicated using a MotionSim phantom in the plane perpendicular to beam axis. Motion was correlated to the position of a surrogate platform, monitored using a reflective marker block and RPM gating system. Dose planes were measured using a MapCheck2 detector placed upon the MotionSim phantom. Dynamic delivery was performed for non-gated, narrow window (2mm or 40-60% in amplitude or phase of breathing cycle, respectively) and wide window (5mm or 30-70% in amplitude or phase of breathing cycle, respectively) gated delivery. Corresponding 3D dose distributions were obtained using 2D maps of dose difference between reference and measurement, backprojected in the direction of the given field onto the original reference 3D dose distribution.
Results: Results based on 2D dose analysis do not always correlate with results based on reconstructed 3D doses, e.g. using narrow gating was found superior to wide gating based on conventional gamma 2D analysis. However, CTV DVH analysis did not show a significant difference. On other hand, both 2D and 3D approaches favour phase- over amplitude-based gating. Both methods also showed better results for adjusted smoother trajectories.
Conclusions: Interpreting dose measurements using three-dimensional reconstructed dose directly on patient model and volume structures provides more relevant information than conventional 2D dose-to-dose approaches and has potential to uncover different trends and the magnitude of a given effect as indicated in this study.