UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

On the regional dose susceptibility of parotid gland function loss and recovery : an effort toward amelioration of radiotherapy-induced xerostomia Clark, Haley


The introduction of intensity-modulated radiotherapy treatment has produced a small surplus of treatment planning flexibility compared to conventional techniques. Target volumes containing cancerous tumours are given a prescribed dose, but the surrounding normal tissue can sometimes be selectively irradiated. Therefore, as tissue-sparing techniques improve, the knowledge of complication risk in normal tissue becomes increasingly important. Xerostomia is one of the most common normal tissue complications in head-and-neck cancer patients. It refers to the non-distinct symptom of dry mouth. In the case of radiotherapy-induced xerostomia, it is generally due to the loss of salivary function resulting from radiation damage to the parotid parenchyma. In severe cases it can drastically reduce oral hygiene and is known to strongly detract from a patient's quality of life. We investigate the regional dose susceptibility of salivary function loss and recovery in the parotid gland with the intent of more precisely quantifying the risk of xerostomia. Reports have indicated regional dose susceptibility of loss in rat parotid glands. Similar results have been seen in human, though they seem to indicate primarily a morphological dependence on the shape of the dose distribution, not specific regional dependence. Further, they consider only subjective xerostomia, not objective salivary function loss nor recovery. The quantification of the regional dependence of loss and recovery of salivary function would substantially benefit our understanding of the complication risk of xerostomia. Immediate improvements in patient outcomes may follow. To this end, we investigate the functional relation between dose delivered to sub-segments of the parotid and whole mouth saliva measurements. To enable the investigation, we have developed a contour recognition system which is able to identify embedded planar organ contours with minimal human effort. Additionally, we have developed a sub-segmentation system capable of partitioning organ contours into arbitrarily-complex sub-segments.

Item Media

Item Citations and Data


Attribution-NonCommercial-ShareAlike 2.5 Canada