Click here to


Are you sure ?

Yes, do it No, cancel

Method to Quantify the PTV Margin Required for Patients Treated with MLC Tracking for Lung SABR

V Caillet1,2*, B Swan4,5 , A Briggs2 , N Hardcastle3,7 , D Jayamanne2 , T Eade 2, P Greer5,6 , R O'Brien1 , P Keall1 , J Booth2,(1) ACRF Institute X, University Of Sydney, NSW,(2) Royal North Shore Hospital, St Leonards, NSW, (3) Peter MacCallum Cancer Centre, Melbourne, Victoria, (4) Calvary Mater Newcastle, Newcastle, (5) Central Coast Cancer Centre, Gosford, Australia,(6) Faculty of Science and IT, School of Mathematical and Physical Sciences, The University of Newcastle, Newcastle, Australia, (7) Institute of Medical Physics, University of Sydney, Sydney, NSW


(Thursday, 8/2/2018) 7:30 AM - 9:30 AM

Room: Davidson Ballroom A

Purpose: To estimate the minimum GTV-PTV margins for Multi-Leaf Collimator (MLC) tracking treatment of lung SABR from retrospective geometric analysis.

Methods: Thirteen patients were treated in our institution as part of the first-in-world MLC tracking for lung SABR (LIGHTSABR, NCT02514512) using electromagnetic beacons implanted around the tumor. GTV-PTV margins were determined from a combined probability density function (PDF) of each source of uncertainties, including measured MLC tracking specific uncertainties such as finite leaf speed, width, prediction algorithm and beacon-to-tumor differential motion (surrogacy); and general radiotherapy uncertainties including inter-observer contouring variability and end-to-end spatial accuracy of the linear accelerator obtained from the literature. The random and systematic components were combined to estimate the margins required to provide coverage 95% of the time and compared to the Van Herk methodology for mid-ventilation.

Results: The minimum GTV-PTV margins (mean ± standard deviation) required for MLC tracking were 5.5 ±1.2mm, 5.8 ±1.9mm and 5.4 ±1.1mm in the left-right, superior-inferior and anterior-posterior direction, respectively. Contouring variability had the strongest contribution to displacements between the GTV and the beam with margins of 4.6 mm in all directions. Margins required for only the MLC tracking uncertainties were quantified in each direction to be 2.5 ±0.9mm, 3.4 ±1.1mm and 1.9 ±0.7mm. Margins required for mid-ventilation approach were 7.9 ±0.7mm, 8.3 ±1.1mm, 8.3 ±0.9mm Calculation of margin for each patient showed no correlation between the required margins and the tumor amplitude (r=0.21).

Conclusion: MLC tracking for lung SABR requires smaller margins than the mid ventilation approach for safe tumour targeting. Our method allows the isolation of factors contributing to beam to target displacements towards a generalized approach to margin setting for personalized real-time adaptive radiotherapy.

Funding Support, Disclosures, and Conflict of Interest: Keall and Booth are investigators on MLC tracking clinical trials partially supported by Varian Medical Systems. PJK is an inventor on one licensed patent related to MLC tracking. Keall, O'Brien and Caillet gratefully acknowledge funding from the ACRF. Keall acknowledges funding from an Australian Government NHMRC Senior Professorial Research Fellowship.


MLC, Treatment Techniques, Clinical Trials


TH- External beam- photons: Motion management (intrafraction)

Contact Email