Mandibular Cortical Bone Loss Early after Definitive Chemoradiotherapy for Oropharyngeal Squamous Cell Carcinoma

Abstract

Patients receiving radiotherapy (RT) for oropharyngeal carcinoma are at risk of mandibular toxicity. Risk factors include RT dose exceeding 60 Gy, local trauma, and uncontrolled periodontal disease. Treatment-related osteonecrosis is likely due to changes in bone remodeling. We aim to identify longitudinal changes in mandibular cortical thickness (C.Th) of patients treated with definitive RT for oropharyngeal cancer using quantitative computed tomography (CT). Forty-four patients with squamous cell carcinoma of the oropharynx treated with definitive chemoradiotherapy between 2016-2017 were included in this Institutional Review Board-approved study. A validated bone mapping technique was used to examine the cortical bone changes in the mandible. Briefly, the mandible was segmented from baseline and follow-up clinical CT data using Mimics and C.Th was calculated using Stradwin at thousands of points across the mandibular surface. Iterative closed point (ICP) registration was performed in a computer algorithm software and cortical changes between baseline and follow-up were calculated for corresponding points. The significance of the changes for each point on the surface between the baseline and follow-up scans was calculated. Statistical analyses on the surface maps were performed using a computer software toolbox for the statistical analysis of univariate and multivariate data using linear mixed effects models. The majority of patients had stage III-IV disease (98%) and a tonsil or base of tongue primary (93%). The median age was 60 years and 91% were male. All patient received concurrent chemotherapy and 11% received additional induction chemotherapy. Dental extractions were performed prior to RT in 63%; 4 patients were edentulous. Median time between baseline and follow-up imaging was 90.5 days (range, 25-190). Median volumes receiving 60 Gy for the symphysis, body, angle, and ramus were 0.8%, 29%, 76%, and 43% respectively. Mean C.Th at baseline and follow-up were 2.6 mm (95% CI 2.5-2.7) and 2.4 mm (95% CI 2.3-2.5) respectively, with an overall mean difference of 0.2 mm (95% CI 0.1-0.2, p<0.001). Significant C.Th loss was observed bilaterally at all anatomic locations, however symphyseal and proximal regions (-8.9%) incurred the greatest loss across patients. Greater C.Th loss was observed in patients that underwent pre-RT dental extractions (-7.0%, p<0.001) than those that did not (-4.6%, p=0.11). This study provides new insights into the early effects of head and neck RT on the mandible and suggest an increased bone remodeling following treatment throughout the mandible. Dental extractions prior to RT increase the extent of bone loss. No dose response was observed in this study, however further evaluation of a dose-response relationship are warranted.