A study on the impact of patient-related parameters in the ability to spare parotid glands by intensity-modulated radiotherapy for head and neck squamous cell carcinomas.

Abstract

This study aims to study the effect of geometric- and patient-related variables in achieving the desired dose-volume constraints to parotid for patients undergoing definitive/adjuvant intensity-modulated radiotherapy (IMRT) for head and neck squamous cell carcinomas (HNSCC). This retrospective study considered HNSCC patients who underwent IMRT at our center between 2009 and 2014. Patients' details and dose-volume parameters were collected, and correlated with dose to parotids using Pearson's correlation test. Sixty-seven patients were found to be eligible. Oral primary predominated (37.3%) and 53 (79%) had locally advanced disease. The parotid volume (PV) had no impact on mean dose. There was a negative linear correlation between the ratio (Parotid-planning target volume [PTV] overlap)/PV and mean dose to the whole parotid (y = 1.14-0.011×; R2 = 0.509; P = 0.001), suggesting that a ratio of 0.7 resulted in mean dose ≥ 40 Gy and potentially irreversible xerostomia. No association was found between site of primary and parotid dose, but node-positive (N+) patients received significantly higher dose compared to N0 patients (34.5 Gy vs. 29.5 Gy; P = 0.001). N + patients persisted to receive higher dose even when T stage was accounted for (32.5 Gy vs. 28.9 Gy for </= T2, 35.4 Gy vs. 30.1 Gy for > T2; P = 0.003). On the last follow-up, 35 patients (52%) had >/=Grade II xerostomia. Average mean dose to combined PV was higher in patients with > grade I xerostomia compared to patients with </=Grade I (35 Gy vs. 29 Gy; P = 0.01). The degree of PTV overlap with parotid (≥30%) and N + disease are the strongest predictors of poor parotid sparing by IMRT.