ESTRO 36 Abstract Book

S468 ESTRO 36 _______________________________________________________________________________________________

Purpose or Objective Stereotactic body radiotherapy (SBRT) for prostate is a cost-effective treatment option with improved patient comfort and maintained excellent clinical outcomes. However, to ensure low levels of toxicity very accurate delivery is imperative, especially when combined with integrated focal boosts as in the Hypo-FLAME clinical trial methodology. Within this context, intra-fraction organ motion management becomes even more relevant. The novel BioXmark® (Nanovi A/S) biodegradable radio- opaque liquid fiducial marker was studied as alternative for current markers used in prostate motion management. The marker can be injected with very thin needles (down to 25G) and the injection procedure allows to vary the marker-size by altering the injected volume. In this study the automatic detectability of BioXmark® in 2D kV X-ray imaging was determined. Additionally, as Hypo-FLAME involves a multi-modality delineation of the boost foci, visibility/artefacts in different types of volumetric imaging was investigated. Material and Methods BioXmark® consists of sucrose acetate isobutyrate (SAIB), iodinated-SAIB and ethanol solution. Upon injection, ethanol diffusion out of the solution causes a viscosity increase and formation of a gel-like marker. A total of 8 markers (size 5-300 µL) organized in a rectangular grid were injected into a gelatin phantom. X-ray projection images using the Varian TrueBeam STx OBI were obtained by putting the gelatin phantom on top of an anthropomorphic pelvic phantom. A total of 120 images of each marker were acquired varying the positions of the marker relative to pelvic bony structures and using 24 clinically relevant X-ray kVp/mAs settings. Volumetric imaging was performed with CT, CBCT and MRI using a CIRS pelvic phantom. Automated marker detection was based on the normalized cross-correlation (NCC) of the projection image with a marker template retrieved from the CT image. Prior to detection, single markers were artificially isolated to minimize interference between detection of the different markers. Reference marker positions were manually determined on the image with highest exposure settings. A detection was successful if the optimal NCC value lied within a 1 mm (3 pixels) tolerance of the reference position. The tolerance was extended to 4 pixels to deal with the uncertainty of manual delineation. Results Detection success rates augmented with increasing marker-size obtaining a maximum for intermediate size (25-75 µL) markers (Figure 1). Larger marker sizes (>75 µL) had decreased detection success rates due to higher susceptibility for interference with the bony structure edges. Volumetric image artefacts were minimal whilst the markers itself were clearly visible (Figure 2). Conclusion Intermediate size (25-75 µL) BioXmark® liquid fiducial markers showed high detectability and minimal image artefacts making them a patient friendly alternative (thin needles) for the current markers used in fiducial-marker- based intra-fraction organ motion monitoring in prostate SBRT.

PO-0861 Geometric validation of a 4D-MRI guided correction strategy on the MR-Linac T. Van de Lindt 1 , R. Koopman 1 , A. Van de Schoot 1 , I. Torres-Xirau 1 , U. Van der Heide 1 , J.J. Sonke 1 1 Netherlands Cancer Institute Antoni van Leeuwenhoek Hospital, Radiation Oncology, Amsterdam, The Netherlands Purpose or Objective Currently in radiotherapy, respiratory motion correction strategies are performed by the use of 4D-(CB)CT. However, moving targets in for example the upper abdomen are not (clearly) visible on these images because of low soft-tissue contrast. The introduction of an integrated MRI and linear accelerator (MR-Linac) will allow for daily MRI-guidance of the tumor. Therefore, the aim of