OA126] Intrafractional imaging during volumetric modulated arc therapy (VMAT) prostate treatment in combination with gold anchor fiducials

Abstract

Purpose Triggered imaging is the use of kV image acquisition during treatment delivery to monitor intrafractional motion of implanted fiducials. An automatically identified fiducial point on the kV image is compared to its expected position predetermined from the treatment planning CT, and the treatment can be paused if the difference exceeds a user-definable tolerance. The purpose of this work was to evaluate triggered imaging in combination with Gold Anchor fiducial markers. Methods Triggered imaging was performed using a spherical homogeneous phantom with images acquired in 0° and 270° on a Truebeam STX linac (Varian Medical Systems, USA). The phantom contained three Gold Anchor fiducials (Naslund Medical, Sweden) and was moved 1–6 mm using HexaMotion (Scandidos, Sweden) during triggered imaging. Since Gold Anchor fiducials are flexible and may end up in various shapes in the patient, the tests were performed with both compressed (around 5 mm) and more elongated fiducials (up to 15 mm). To evaluate how well the system could locate fiducials in patients, triggered imaging was performed during one fraction in five patients undergoing 2-arc VMAT prostate treatment, all with three Gold Anchor fiducials implanted in the prostate. Images were acquired in intervals of 30°, resulting in 24 images per patient. The fiducial tracking accuracy was visually inspected in all images. In both the phantom and patient evaluation a position difference tolerance of 5 mm was used. Results During the phantom tests, the system correctly identified all fiducial positions as being inside or outside their tolerance for movements ≤ 4 or ≥ 6 mm. Negligible prostate motion were observed during the five prostate treatment deliveries. The system correctly identified 316/360 fiducials as being inside tolerance, 36/360 could not be located and 8/360 were falsely identified as another fiducial or bone structure. The main reason for tracking failure was elongated and/or insufficiently separated fiducials. In both the phantom and the patient situation, the system did not always identify the same point on a fiducial. This tracking variation was larger for elongated fiducials. Conclusions Triggered imaging seems to perform satisfactorily when used in combination with well-separated and compressed Gold Anchor fiducials.