Differences in the Temporal Patterns of Volumetric Regression and Recovery for Spared and In-Field Submandibular Glands Among Patients Treated By Radiation for Oropharynx Cancer

Abstract

Submandibular gland sparing (nodal level IB) is increasingly recognized as having an important role in xerostomia reduction for patients undergoing radiation for head and neck cancer. Mean dose thresholds of 30 to 40 Gy have been proposed to optimize salivary sparing. While anatomical studies have shown that the submandibular glands shrink modestly in response to radiation during and after treatment, it is unknown how sparing this organ at risk (OAR) may affect its volume. We aimed to study volumetric changes in the spared and unspared submandibular gland using each patient as his or her own control. Eligible patients (n=15) received definitive chemoradiation for squamous cell carcinoma of the oropharynx between 2012 and 2015. Median follow-up was 26 months (range: 15 to 43 months). All patients received radiation to the primary and bilateral neck using intensity modulated radiation therapy with prescription doses ranging from 54 to 70 Gy. Ipsilateral level IB was always included in the high dose volume and the contralateral level IB was designed as an OAR and intentionally spared (with objective of mean less than 39 Gy). Patients were excluded if they had radiographic evidence of level IB nodal disease. The submandibular glands were delineated on the simulation computed tomography (CT) scans as well as follow up imaging from at least two different time points, starting at 3 months after the completion of treatment and until last follow up. Mean submandibular doses were recorded and volumetric changes were calculated using their CT simulation as a baseline measurement. A total of 30 submandibular glands were used for this analysis. The mean radiation doses to the treated ipsilateral and spared contralateral submandibular glands were 69 Gy (range: 63 to 72) and 36 Gy (range: 25 to 45), respectively. Each of the in-field submandibular glands showed a reduction in volume except for one patient. The mean volume reduction was 40% at 3 months. Each of the treated ipsilateral glands continued to regress in volume even after three month follow up. The average maximum change in volume was 57%. The spared contralateral submandibular gland experienced a similar reduction of volume (mean 43%) at the first 3 month time point. In contrast to the treated side, however, 12 of 15 spared submandibular glands displayed a recovery in volume after 3 months at their next follow up scan. The three spared glands that continued to decrease in volume were all treated to doses over 35 Gy. Notable differences in the temporal pattern of regression and recovery were observed for in-field and spared submandibular glands, with the latter demonstrating measurable and earlier volumetric gains that may be indicative of improved salivary flow and preservation. The clinical implications of these findings warrant further investigation.