Comparison of the Dosimetry of Spinal Fields in Craniospinal Irradiation using Two Dimensional, Three Dimensional and Intensity Modulated Radiation Therapy Planning Techniques

Abstract

Materials/Methods: The CT-based treatment planning was performed for the CSI volume using lateral cranial fields with a matched posterior single spine field. The prescription dose was 36 Gy to the CS axis delivered with 4 junction shifts. Target delineations were as follows: 2D plan field edge was placed 1 cm lateral to vertebral bodies; 3D plan had 1 cm expansion from the spinal canal for clinical target volume (CTV) for adults (0.5 cm for children) and an additional 0.7 cm to field edge for planning target volume (PTV). For IMRT plan, CTV was equal to PTV without additional expansion. To improve target coverage, a field in field technique was used for both 2D and 3D plans to act as compensators. The OAR evaluated included heart, thyroid, liver, lung, kidney, and small bowel. Dose volume histograms were used to assess the plans and time spent per plan was recorded. Results: The CT scans of 10 consecutive patients (4 pediatric and 6 adult) treated at our institution with CSI from 2004-2007 were selected. There was no statistical difference in PTV coverage when looking at the mean, minimum, maximum, and median doses andthepercentvolumecoveredby95%and107%ofthedose.MeandosetoOARwerealldecreasedwhencomparingIMRTto3D as follows: heart 11.3%, liver 12%, thyroid 2%, kidneys 22%, bowel 8% (p\0.05). Mean dose to OAR were all decreased when comparing 3D to 2D as follows: heart 5%, liver 13.5%, kidneys 31%, lungs 16.3% , bowel 28% (p\0.05). Total time spent planning (not including contouring) for each plan was approximately 4 hours per IMRT plan and 6-7 hours per 3D and 2D plans. Conclusions: Optimal treatment planning using either 2D or 3D plans with field in field technique results in similar PTV coverage when compared to IMRT plans. However, IMRT plans were superior to either 2D or 3D when looking at mean dose to most OAR. The IMRT planning took significantly less amount of time for treatment planning. These results make IMRT treatment planning more desirable in clinical practice. Author Disclosure: P.K. Parhar, None; C. Hitchen, None; T. Duckworth, None; K. DeWyngaert, None; A. Narayana, None.