The Role of Treatment Plan Adaptation During the Course of Proton Radiotherapy for Patients with Head and Neck Cancer
N Shusharina*, A Chan, J Adams, G Chen, G Sharp, Massachusetts General Hospital, Boston, MATH-A-BRA-10 Thursday 8:00:00 AM - 9:55:00 AM Room: Ballroom A
Purpose: To evaluate the advantages of adaptive replanning associated with proton radiotherapy treatment of patients with head and neck cancer.
Methods: Six patients with tumors in the base of skull were treated with proton radiotherapy. Two CTs of the same patient, planning and repeat, were first registered rigidly in relation to bony structures of the skull to remove the setup error. A deformable B-spline based registration was performed between the two CTs. The resulting deformation field was used to automatically transfer structure contours from the planning-CT to the repeat-CT. The original treatment plan was applied to the repeat-CT and the dose distribution was recalculated. The differences between new and original dose distributions were evaluated by comparing isodose lines and by performing distance-to-agreement dose distribution analysis. The automatically transferred contours were used to compute dose-volume histograms for the repeat-CT and these were compared to the original DVHs.
Results: Autotransferred contours were investigated by a physician and found satisfactory for the most critical structures. The targets were delineated manually. Anatomic changes caused by patient response to the treatment were primarily due to tumor shrinkage in the nasal cavity and in the paranasal sinuses. As a result the areas around target became highly heterogeneous with respect to density distribution. In 2 cases these heterogeneities resulted in hot spots within the tumor larger than 1 cm³ with a dose increase of 6%. In one case a cold spot with a dose decrease of 7% was observed.
Conclusions: Evaluation of six cases shows no significant dosimetric changes that can favor adaptive replanning at the mid-course of proton radiotherapy of head and neck cancer. Analysis of a larger patient set is needed to establish robust patterns of tissue deformation and dosimetric changes, with a focus on variations of the dose distribution in the presence of heterogeneities.