Abstract

To determine whether the radiation dose to the parotid gland stem cell (SC) region is associated with long-term patient-reported xerostomia after definitive head and neck cancer (HNC) radiation therapy (RT), and whether mid-treatment dose assessment improves the association. The SC region of the parotid gland, defined as being located next to the dorsal edge of the mandible, near the Stensen’s duct at the anterior border based on pre-clinical investigations, was delineated on CT scans with a 0.5 cm isotropic margin for 65 HNC patients that had undergone definitive RT between 2009 and 2014. Prospectively collected EORTC QLQ-H&N35 quality-of-life questionnaires with minimum 9 months follow-up were used to score xerostomia on a 4-grade scale, where grade 3 and 4 was considered severe xerostomia in this analysis. The SC regions were delineated on pre-treatment as well as mid-treatment CT scans to determine the best model for predicting xerostomia. The association between the mean dose to the spared parotid gland or SC region of the spared parotid and the risk of severe xerostomia was examined using logistic regression and receiver operating characteristics (ROC). Increasing radiation dose to either whole parotid or the SC region was associated with an increased risk of patient-reported xerostomia (p=0.003 and p=0.005). Importantly, the mid-treatment analysis showed that the dose to the SC region was more predictive of xerostomia than that of the pre-treatment or using the whole parotid dose, as per the ROC areas under the curve (AUCs) in Table 1. For every 1 Gy increase in radiation dose to the SC region evaluated at mid-treatment, we observed an 8% increase in the odds of xerostomia. We furthermore found that the parotid volume of patients with xerostomia was on average 27% reduced at mid-treatment, compared to only 15% for patients without xerostomia. We found that increased radiation dose to the SC region of the spared parotid gland was associated with an increased risk of patient-reported xerostomia, especially when evaluated at mid-treatment. This supports the hypothesis that targeting the SC region reduces regenerative capacity of the gland, which is also supported by the large reduction in parotid size. These results can assist clinicians in adapting RT to optimize efficacy while improving patients’ post treatment quality-of-life.Abstract 2985; Table 1Logistic regression models predicting patient-reported xerostomia.OR per Gy (95% CI)p-valueROC AUC (95% CI)Spared parotid mean dose pre-treatment1.08 (1.03, 1.13)0.0030.72 (0.60, 0.85)Spared parotid SC mean dose pre-treatment1.07 (1.02, 1.12)0.0050.70 (0.57, 0.83)Spared parotid mean dose mid-treatment1.07 (1.01, 1.13)0.0260.72 (0.54, 0.90)Spared parotid SC mean dose mid-treatment1.08 (1.02, 1.14)0.0120.76 (0.60, 0.93) Open table in a new tab

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call