Room: AAPM ePoster Library
Purpose: The present study aims to analyze the difference in dose distribution between different dose calculation algorithms in clinical cases of Stereotactic Ablative Body Radiotherapy (SABR) for lung cancer.
Methods: 96 patients received SABR for lung cancer at Kyushu University Hospital from 2013 to 2017. All treatment plans were calculated by Acuros XB in Eclipse TPS (AX). We recalculated these plans using Collapsed Cone Convolution/Superposition in RayStation (CC) with the identical MU values and beam arrangements. DVH parameters (Maximum Dose (Dmax), Minimum Dose (Dmin), Homogeneity Index (HI), Conformity Index (CI), and D95 of the PTV) were calculated for each plan. We investigated the difference between the two calculations and also examined the impact of tumor location by comparing the above 5 DVH parameters. Moreover, we measured the irradiated dose using a thorax phantom. Then, we assessed the calculation accuracy of the two algorithms by comparing the measured dose and the calculated dose by AX or CC for each fraction.
Results: Overall, CC overestimated the dose to PTV, compared to AX. The mean differences of Dmax, Dmin, and D95 were 1.17, 1.95, and 1.85 Gy, respectively. For HI, CC is higher than AX, whereas AX is higher than CC in CI. The mean differences of HI and CI were 0.02 and 0.06. Furthermore, all values were significantly different between the two algorithms (p<0.05). Dmin, HI, D95 had a significant correlation with the tumor location, and the difference was greater when the PTV was attached to the chest wall. Regarding the calculation accuracy of the two algorithms, the dose difference of AX and CC for each fraction were 0.02, 0.34 Gy; consequently, AX was more accurate.
Conclusion: The present study has demonstrated that there was a significant difference in dose distribution between different dose calculation algorithms in clinical cases of lung SABR.