Positional reproducibility of pancreatic tumors under end-exhalation breath-hold conditions using a visual feedback technique.

PURPOSE: To assess positional reproducibility of pancreatic tumors under end-exhalation (EE) breath-hold (BH) conditions with a visual feedback technique based on computed tomography (CT) images. METHODS AND MATERIALS: Ten patients with pancreatic cancer were enrolled in an institutional review board-approved trial. All patients were placed in a supine position on an individualized vacuum pillow with both arms raised. At the time of CT scan, they held their breath at EE with the aid of video goggles displaying their abdominal displacement. Each three-consecutive helical CT data set was acquired four times (sessions 1-4; session 1 corresponded to the time of CT simulation). The point of interest within or in proximity to a gross tumor volume was defined based on certain structural features. The positional variations in point of interest and margin size required to cover positional variations were assessed.
RESULTS: The means ± standard deviations (SDs) of intrafraction positional variations were 0.0 ± 1.1, 0.1 ± 1.2, and 0.1 ± 1.0 mm in the left-right (LR), anterior-posterior (AP), and superior-inferior (SI) directions, respectively (p = 0.726). The means ± SDs of interfraction positional variations were 0.3 ± 2.0, 0.8 ± 1.8, and 0.3 ± 1.8 mm in the LR, AP, and SI directions, respectively (p = 0.533). Population-based margin sizes required to cover 95th percentiles of the overall positional variations were 4.7, 5.3, and 4.9 mm in the LR, AP, and SI directions, respectively.
CONCLUSIONS: A margin size of 5 mm was needed to cover the 95th percentiles of the overall positional variations under EE-BH conditions, using this noninvasive approach to motion management for pancreatic tumors.