Monte Carlo dose verification for lungSBRTwith CMS/XiO superposition algorithm

Purpose. Superposition dose calculation algorithms are known to suffer compromised accuracy in regions of electronic disequilibrium. This has particular relevance for treatment of lung nodules with electronic disequilibrium in their periphery using stereotactic body radiotherapy (SBRT). This study assessed the accuracy of a commercial superposition dose calculation algorithm within the context of SBRT in order to quantify the impact of these potential dosimetric uncertainties particularly in the periphery of the target. Method and materials. Selected lung SBRT 3D-CRT plans with 10-13 beams of 6 MV were originally created on the CMS/XiO treatment planning system with the superposition algorithm. The corresponding Monte Carlo (MC) plans were produced using the same patient geometry and plan parameters and compared with the CMS plans. Electronic disequilibrium was investigated by constructing a 2 mm thick 'peel' of the gross target volume (GTV) and planning target volume (PTV). An 18 MV plan was created corresponding to each 6 MV plan. Beam reduction was studied to investigate the dosimetric effect of beam arrangement. Results. The doses calculated by CMS agreed with MC in a water phantom but overestimated in very low density areas and the buildup and builddown regions. In patient plan comparison, for GTVs and GTV peels, CMS agreed with MC in mean dose within 2.4% for the 6 MV plans, and 3.3% for the 18 MV plans. For PTVs and PTV peels, CMS agreed with MC in mean doses within 2.5% for both 6 and 18 MV beams. The ratio in D₉₅ between CMS and MC was close to 1.0 for all 6 MV plans. Reducing beams could result in underdoses. Conclusions. The CMS superposition algorithm is able to perform accurate dose calculations for SBRT lung plans and ensure adequate peripheral dose. Dosimetric uncertainties are correlated to many factors such as density, energy, number of beams, tumor size/location, and density variation of the target and surrounding tissues.