USE OF RESPIRATORY-CORRELATED FOUR-DIMENSIONAL COMPUTED TOMOGRAPHY TO DETERMINE ACCEPTABLE TREATMENT MARGINS FOR LOCALLY ADVANCED PANCREATIC ADENOCARCINOMA

Purpose: Respiratory-induced excursions of locally advanced pancreatic adenocarcinoma could affect dose delivery. This study quantified tumor motion and evaluated standard treatment margins. Methods and Materials: Respiratory-correlated four-dimensional computed tomography images were obtained on 30 patients with locally advanced pancreatic adenocarcinoma; 15 of whom underwent repeat scanning before cone-down treatment. Treatment planning software was used to contour the gross tumor volume (GTV), bilateral kidneys, and biliary stent. Excursions were calculated according to the centroid of the contoured volumes. Results: The mean +/- standard deviation GTV excursion in the superoinferior (SI) direction was 0.55 +/- 0.23 cm; an expansion of 1.0 cm adequately accounted for the GTV motion in 97% of locally advanced pancreatic adenocarcinoma patients. Motion GTVs were generated and resulted in a 25% average volume increase compared with the static GTV. Of the 30 patients, 17 had biliary stents. The mean SI stent excursion was 0.84 +/- 0.32 cm, significantly greater than the GTV motion. The xiphoid process moved an average of 0.35 +/- 0.12 cm, significantly less than the GTV. The mean SI motion of the left and right kidneys was 0.65 +/- 0.27 cm and 0.77 +/- 0.30 cm, respectively. At repeat scanning, no significant changes were seen in the mean GTV size (p = .8) or excursion (p = .3). Conclusion: These data suggest that an asymmetric expansion of 1.0, 0.7, and 0.6 cm along the respective SI, anteroposterior, and medial-lateral directions is recommended if a respiratory-correlated four-dimensional computed tomography scan is not available to evaluate the tumor motion during treatment planning. Surrogates of tumor motion, such as biliary stents or external markers, should be used with caution. (C) 2010 Elsevier Inc.