Dosimetric impact of hollow intraoral stents for head and neck cancer radiotherapy: A phantom study.

Abstract

To investigate the dosimetric impact of the calculation boundaries and dose calculation algorithms of radiotherapy in head and neck cancer patients with an opened oral cavity connected to the exterior by a hollow intraoral positioning stent. A homemade silicone phantom with an opened oral cavity was placed in a CIRS head phantom to model head and neck cancer patients with a hollow intraoral positioning stent. 3D-CRT plans were designed on CT images of the phantom in Monaco and Pinnacle3 treatment planning systems (TPSs) with the same beam parameters. The default boundary and manually extrapolated boundary were both adopted in these two TPSs to explore the dosimetric impact on treatment plans. The nanoDot™ optically stimulated luminescence dosimeters (OSLDs) were chosen to measure the planned dose surrounding the oral cavity of the head phantom after calibration. The doses in the air cavity and two measuring points at the joint area were dramatically changed from 0.0, 92.4 and 148.8 cGy to 177.8, 244.2 and 244.1 cGy in Monaco after adopting the extrapolated boundary. While the calculated doses at the same place were changed from 61.2, 143.7 and 198.3 cGy to 175.4, 234.7 and 233.2 cGy in Pinnacle3 with a similar calculation boundary. For the Monaco TPS, the relative errors compared to the OSLD measured doses were 2.94±1.93%, 0.53±8.64%, 2.65±1.87% and 3.93±1.69% at 4 measuring positions. In contrast, the relative errors 4.03±1.93%, 4.85±8.64%, 7.61±1.87% and 5.61±1.69% were observed in Pinnacle3 . The boundary setting of an opened oral cavity in TPSs has a significant dosimetric impact on head and neck cancer radiotherapy. An extrapolated boundary should be manually set up to include the whole oral cavity in the dose calculation domain to avoid major dose deviations.