Spatial radiation dose influence on xerostomia recovery and its comparison to acute incidence in head neck cancer patients

• Published in: Advances in radiation oncology 2019
• Main Authors: Guo, Yue, MD, MHS, Jiang, Wei, PhD, Lakshminarayanan, Pranav, MS, Han, Peijin, MD, MHS, Cheng, Zhi, MD, MPH, Bowers, Michael, BS, Hui, Xuan, MD, ScM, Shpitser, Ilya, PhD, Siddiqui, Sauleh, PhD, Taylor, Russell H., PhD, Quon, Harry, MD, MS, McNutt, Todd, PhD
• Format: Article
• Language: English
• Subjects: Hematology, Oncology, and Palliative Medicine
• ISSN: 2452-1094; 2452-1094
• DOI: 10.1016/j.adro.2019.08.009

AbstractPurposeRadiation-induced xerostomia is one of the most prevalent symptoms during and after head and neck cancer (HNC) radiotherapy (RT). We aimed to discover the spatial radiation dose-based (voxel dose) importance pattern in the major salivary glands in relation to the recovery of xerostomia 18 months after RT, and to compare the recovery voxel dose importance pattern to the acute incidence (injury) pattern. Methods and materialsThis study included all patients within our database with xerostomia outcomes following completion of curative intensity-modulated RT. CTCAE xerostomia grade was used to define recovered vs. non-recovered group at baseline, between end of treatment and 18 months post-RT, and beyond 18 months respectively. Ridge logistic regression was performed to predict the probability of xerostomia recovery. Voxel doses within geometrically defined parotid glands (PG) and submandibular glands (SMG), demographic, and clinical factors were included in the algorithm. We plotted the normalized learned weights on the 3-dimentioanl PG and SMG structures to visualize the voxel dose importance for predicting xerostomia recovery. ResultsA total of 146 HNC patients from 2008-2016 were identified. The superior region of the ipsilateral and contralateral PG was the most influencial for xerostomia recovery. The area under the receiver operating characteristic curve (AUC) evaluated using 10-fold cross-validation for ridge logistic regression was 0.68 ± 0.07. Compared to injury, the recovery voxel dose importance pattern was more symmetric and was influenced by lower dose voxels. ConclusionsThe superior portion of the two PGs (low dose region) are the most influential on xerostomia recovery and appear to be equal in their contribution.The dissimilarity of the influence pattern between injury and recovery suggests different underlying mechanisms. The importance pattern identified by spatial radiation dose and machine learning methods can improve our understanding of normal tissue toxicities in RT. Further external validation is warranted.