Ray Tracing and Mote Carlo Based Dose Computation for Chest Wall and Ribs for Lung Cancer Patients Treated with Cyberknife
T Podder*, T Biswas, M Yao, S Lo, M Machtay, University Hospitals and Case Western Reserve University, Cleveland, OHSU-E-T-517 Sunday 3:00PM - 6:00PM Room: Exhibit Hall
Purpose: Chest wall and ribs are considered organs at risk while treating lung tumors adjacent to chest wall. This study was to evaluate the dosimetric differences in chest wall and ribs irradiation for Cyberknife stereotactic body radiation therapy (SBRT) when treatment plans were computed using Ray Tracking Algorithm (RTA) and Monte Carlo Algorithm (MCA).
Methods: Twenty eight lung cancer patients who had tumors abutting the chest wall were selected for this study. A uniform 2cm thickness of chest wall and appropriate portions of rib were contoured for evaluating dose distribution using RTA and MCA of Cyberknife MultiPlan software (version-3.5.2). Treatment plans were first computed using RTA and then were recomputed using MCA without changing the total monitor unit (MU). Range of prescription dose was 50Gy-60Gy (mean=54.6Gy). Studied dosimetric parameters were maximum point dose and doses to 1cc, 5cc, 10cc, 30cc, 50cc, 70cc, and 100cc of chest wall. Maximum point dose and doses to 1cc, 2cc, 5cc, and 10cc of ribs were also studied. These parameters were considered as clinically more relevant.
Results: Dosimetric computation results revealed that RTA overestimated dose to both the chest wall and the ribs for all the studied parameters. Increase of average dose was in the range of 7.5-12.1% (p-value<0.05, except 50-100cc p-value>0.05), and 10.3-13.4% (p-value <0.05) for chest wall and ribs, respectively. The lower the volume of interest, the larger was the difference in dose for the two algorithms. Average difference in maximum point dose for chest wall was 11.2% (p-value <0.001) and that was 13.4% for ribs (p-value <0.002).
Conclusion: As compared to Monte Carlo algorithm, Ray Tracing algorithm significantly overestimates dose to chest wall and ribs. Although treatment planning with Monte Carlo is more involved, it should be used because of accuracy and significant difference in dose distribution in chest wall and ribs.