Purpose: Anatomical changes in a patient during radiation therapy can cause imprecise delivery of the irradiation dose, resulting in an increased dose to normal tissue and a reduced dose to tumor tissue. We have developed a method that uses the deviation in the water equivalent length (ΔWEL) along the spot beam path to quickly check for anatomical changes and then estimate the resulting dose differences (ΔDose) from the planned dose distribution. It is designed for use in spot-scanning proton beam therapy (SSPT).
Methods: To determine whether the WEL can be used to predict dose changes due to respiration, we investigated the correlation between the ΔWEL obtained with the proposed method and the ΔDose obtained from freely available 4D-computed tomography (CT) data. Three ΔWEL analysis methods were evaluated: 1) ΔWEL map, 2) ΔWEL spot, and 3) monitor-unit (MU)-weighted ΔWEL spot. Treatment plans were prepared for a simulated target implanted into a chest defined by a 4D CT dataset provided by DIR-Lab. The ΔWELs between the planned CT paths and the comparison target CT paths were calculated spot by spot using DICOM-Radiation Therapy software. Linear regression fits between the ΔWEL histogram (median and 95% width) and ΔDose (root mean square deviation (RMSD) and 3% dose pass rate) were calculated for the three analysis methods
Results: The combination of RMSD and ΔWEL showed a relatively high correlation expressed as the coefficient of determination (0.3631< R^2< 0.8894) while the combination of the 3% dose pass rate and ΔWEL showed a low correlation (0.0002
Conclusion: These results show that the ΔWEL spot histogram can provide information for predicting the dose deviation to the target and for judging whether to continue or terminate the treatment course.
Funding Support, Disclosures, and Conflict of Interest: Fujii and Hirayama are employees of Hitachi, Ltd.