Designing a New Planning Target Volume Based on Rotation in Patients With Oropharyngeal Cancer

Abstract

Studies show that rotation is a significant source of target displacement. The standard isotropic expansion of Clinical Target Volume (CTV) to Planning Target Volume (PTV) neglects rotation. While imaging and 6D couch corrections perform a single shift through isocenter, they poorly track off-axis rotations through the spine or spine shifts that produce twists or arching. This study aims to determine dosimetric differences in head and neck patients when a standard expansion PTV is replaced by one that tracks the space occupied by target due to uncorrected rotational error. To perform off-axis rotation, a hypothetical point was placed in the center of the cervical spinal canal on Computed Tomography (CT) scan and the image was rotated around the longitudinal axis ±5°. Comparison plans were generated using a standard 3mm isotropic CTV expansion against the new PTV formed of the rotated CTV to evaluate mean dose reduction to superior Pharyngeal Constrictor Muscle (PCM). A published Normal Tissue Complication Model (NTCP) was used to predict dysphagia probability, based on the dose to superior PCM and supraglottis. The average percent change for PTV volume and overlap with superior, middle and inferior PCMs in a pool of 20 patients was -19%, -37%, -59% and -45.2%. The new rotational PTV causes more reduction in the superior PCM overlap in base of tongue lesions compared to tonsillar lesions (57.8% versus 25.8%, p = 0.01 as well as middle PCM overlap (73% versus 49%). The smallest isotropic expansion that covers the target when the spine rotates ±5° was 5 mm. 35% of patients required an expansion more than 3 mm to account for off-axis rotation. In 10 patients randomly selected for VMAT planning, the average change to the volume overlap of PTV70 with superior PCM using the standard and new PTV was 45% (8-100%), with an average mean dose reduction of 2.7 Gy. Based on the NTCP model, 50% of cases had a 30% or more probability of swallowing complications. Using the new technique, the NTCP decreased to less than 30% in 80% of these patients. Considering other Organs at Risk (OARs) constraints, there was not a difference between the new plan versus the standard plan. Designing a new PTV based on spinal rotation result in less volume overlap between PTV and PCM and can ultimately lead to less mean superior PCM dose and risk of dysphagia. A standard 3mm isotropic expansion is insufficient to cover the target volume due to spine rotation, and a 5mm margin is needed.