Encrypted login | home

Program Information

First Cardiac Radiosurgery MLC Tracking Results

no image available
S Lydiard

S Lydiard1,2*, V Caillet3, S Ipsen4, R O'Brien1, R Bruder4, O Blanck5, J Booth3, P Keall1, (1) University of Sydney, Camperdown, (2) Auckland District Health Board, Auckland, (3) Royal North Shore Hospital, Saint Leonards, NSW, (4) University of Luebeck, Luebeck, Schleswig-Holstein, (5) University Medical Center Schleswsig-Holstein, Kiel, Schleswig-Holstein


MO-AB-FS4-10 (Monday, July 31, 2017) 7:30 AM - 9:30 AM Room: Four Seasons 4

Purpose: Cardiac Radiosurgery has recently been proposed as a treatment option for Atrial Fibrillation (AF). However, it faces possibly the most challenging target tracking scenario: both respiratory and cardiac motion. MLC tracking is clinically used for respiratory motion compensation in lung cancer treatments but it’s capabilities with cardiac motion is unknown. We report the first results of MLC tracking for cardiac targets.

Methods: The following steps were followed: i) Four dual arc AF Radiosurgery plans of varying complexity were created; three VMAT with target expansions of 0mm, 3mm and 5mm respectively, and one Dynamic Conformal Arc (DCA). All delivered 5x10Gy to both pulmonary vein antra. ii) Human target motion trajectories were acquired with ultrasound from three healthy volunteers and programmed into a motion platform. iii) Plans were delivered with a standard linac to a biplanar diode array dosimeter placed on the motion platform. For each motion trace, each plan was delivered without MLC tracking, with MLC tracking, and with MLC tracking and motion prediction. Real-time target position was acquired from Calypso beacons attached to the dosimeter. Static dosimetric measurements with no motion platform movement were also acquired for each plan. iv) Dosimetric accuracy was assessed with γ-tests using the static measurement as reference.

Results: MLC tracking improved dosimetric accuracy in all measurements. The average 2%/2mm γ-fail rate was improved from 15% without MLC tracking to 7.5% with MLC tracking (p=0.06) and 6.4% with MLC tracking and motion prediction (p=0.05). The average 3%/3mm γ-fail rate was improved from 4.6% without tracking to 2.8% with MLC tracking (p=0.06) and 2.7% with MLC tracking and motion prediction (p=0.05).

Conclusion: MLC tracking was successfully applied to compensate for cardiac target motion during AF Radiosurgery. Significant dosimetric accuracy improvements were seen despite challenging target motion and complex treatment plans, warranting further development of this technique.

Funding Support, Disclosures, and Conflict of Interest: NHMRC Program Grant

Contact Email: