• Winnie Li, Thomas G Purdie, Mojgan Taremi, Sharon Fung, Anthony Brade, B C John Cho, Andrew Hope, Alexander Sun, David A Jaffray, Andrea Bezjak, and Jean-Pierre Bissonnette.
• Radiation Medicine Program, Princess Margaret Hospital, and Department of Radiation Oncology, University of Toronto, Toronto, Ontario, Canada. winnie.li@rmp.uhn.on.ca
• Int. J. Radiat. Oncol. Biol. Phys. 2011 Dec 1; 81 (5): 1568-75.

PurposeTo assess intrafractional geometric accuracy of lung stereotactic body radiation therapy (SBRT) patients treated with volumetric image guidance.Methods And MaterialsTreatment setup accuracy was analyzed in 133 SBRT patients treated via research ethics board-approved protocols. For each fraction, a localization cone-beam computed tomography (CBCT) scan was acquired for soft-tissue registration to the internal target volume, followed by a couch adjustment for positional discrepancies greater than 3 mm, verified with a second CBCT scan. CBCT scans were also performed at intrafraction and end fraction. Patient positioning data from 2047 CBCT scans were recorded to determine systematic (Σ) and random (σ) uncertainties, as well as planning target volume margins. Data were further stratified and analyzed by immobilization method (evacuated cushion [n=75], evacuated cushion plus abdominal compression [n=33], or chest board [n=25]) and by patients' Eastern Cooperative Oncology Group performance status (PS): 0 (n=31), 1 (n=70), or 2 (n=32).ResultsUsing CBCT internal target volume was matched within ±3 mm in 16% of all fractions at localization, 89% at verification, 72% during treatment, and 69% after treatment. Planning target volume margins required to encompass residual setup errors after couch corrections (verification CBCT scans) were 4 mm, and they increased to 5 mm with target intrafraction motion (post-treatment CBCT scans). Small differences (<1 mm) in the cranial-caudal direction of target position were observed between the immobilization cohorts in the localization, verification, intrafraction, and post-treatment CBCT scans (p<0.01). Positional drift varied according to patient PS, with the PS 1 and 2 cohorts drifting out of position by mid treatment more than the PS 0 cohort in the cranial-caudal direction (p=0.04).ConclusionsImage guidance ensures high geometric accuracy for lung SBRT irrespective of immobilization method or PS. A 5-mm setup margin suffices to address intrafraction motion. This setup margin may be further reduced by strategies such as frequent image guidance or volumetric arc therapy to correct or limit intrafraction motion.Copyright © 2011 Elsevier Inc. All rights reserved.

39 keywords...

hide…