Date of Graduation
Doctor of Philosophy (PhD)
Laurence E. Court, Ph.D.
Lei Dong, Ph.D.
Steven J. Frank, M.D.
Kenneth R. Hess, Ph.D.
Rajat J. Kudchadker, Ph.D.
X. Ronald Zhu, Ph.D.
The main advantage for using protons in radiotherapy is their finite range in patients, allowing for potential improved sparing of normal tissues. However, this comes at a cost of increased sensitivity to range uncertainties. Density changes along the beam path will affect the proton range and the resultant dose distribution, making it difficult to estimate the impact of visible anatomic changes to the patient dose distribution. In order to better understand the effect of anatomy change on proton dose, some form of treatment-time verification is required and methods to correct for observed changes would be beneficial. Therefore, this project aims to develop image-guidance techniques for proton therapy that incorporates proton range changes to allow for accurate treatment-time dose verification and corrective actions to ensure proper dose delivery.
A method for quick estimation of the treatment-time dose based on CT-imaging using prior dose information was developed and validated. This technique uses changes in calculated radiological pathlength on CT images to remap prior dose distributions on new anatomy or new setup position. We assessed the accuracy of this technique compared to full dose calculation and found the average passing rate of 3D gamma analysis (3% dose-difference, 3-mm distance-to-agreement) were 96% and 89% for setup errors and severe anatomy changes, respectively. The average (maximum) of RMS deviation of the DVHs under the weekly anatomical change was 0.6% (2.7%) for all structures considered.
Using the quick dose estimation tool, we developed a method to position the patient based on dose information instead of simply using anatomic information. This would allow for dose-based optimization to be included in the patient setup process. We found a statistically significant improvement in target coverage and normal tissue sparing using our method when compared to anatomy-based setup.
Finally, we assessed a potential method to adapt spot scanning proton treatment plan beam parameters to account for anatomical changes. This range-adaptive method adjusts the proton beam directly to match the new range to anatomy in the treatment-time image. Using this technique, we were able to reduce normal tissue dose but ended up with increased target heterogeneity and reduced target coverage.
adaptive radiation therapy, proton therapy, dose-guided radiation therapy, lung cancer, image-guided radiation therapy, dose calculation, spot scanning proton therapy, anatomical changes in radiation therapy