ESTRO 38 Abstract book

S212 ESTRO 38

We have developed a fast, robust and accurate ST tracking algorithm in cine‐MR data which was validated against marker tracking. The presented method for soft‐tissue contrast based tracking obleviates the need for surgically implanted fiducial markers during MR‐guided prostate RT on an MR‐Linac. OC-0411 Geometric efficacy of breath-hold gated MR- guided SABR for adrenal metastases J. Van Sornsen de Koste 1 , M. Palacios 1 , A. Bruynzeel 1 , F. Spoelstra 1 , B. Slotman 1 , F. Lagerwaard 1 , S. Senan 1 1 Amsterdam UMC- Vrije Universiteit Amsterdam- VU Medical Center, Department of Radiation Oncology, Amsterdam, The Netherlands Purpose or Objective Magnetic Resonance (MR) guided stereotactic ablative radiotherapy (SABR) improves target coverage and reduces organ‐at‐risk doses. We studied target coverage and breath‐hold performance in patients undergoing MR‐ Breath‐hold SABR data from 18 patients treated on a MRIdian unit (ViewRay Inc) was analyzed. Real‐time tumor tracking during SABR was performed using repeated fast planar MR imaging in a single sagittal plane, at 4 frames‐ per‐second with 3.5mm x 3.5mm in‐plane resolution via deformable image registration. An in‐room MR‐compatible monitor projects sagittal images to allow patient visualization of tracked GTV (GTV t ) and PTV (GTV+3mm) contours. Simulation MR‐imaging is performed in quiet inspiration. A delivery threshold‐ROI% determines the maximum permitted percentage of GTV t ‐area that can be outside the PTV‐area, before a beam‐hold is triggered. Breath‐hold related tumor coverage during SABR was analyzed for 82 fractions (40 hours of MR‐cine series). For each fraction, we analyzed: [1] geometric coverage of GTV t within PTV; [2] duty‐cycle efficiency; [3] stability of breath‐holds during each session [4] beam‐off latency effects on target coverage using a 500msec system‐latency time as a worst case scenario. Results Different threshold‐ROI% settings were used (range 7‐ 20%), but 75% of fractions used a 15% threshold‐ROI. Mean geometric GTV t coverage was 94.2% (5 th ‐95 th Percentile range: 90.6%–96.7%); corresponding mean duty‐cycle efficiency was 71.7% (range: 45.5%–99.5%). Average duty‐ cycle efficiency increased from 70.5% during the first fraction, to 76.7% during the last fraction (Figure), and treatment delivery times decreased from 32.4 to 28.6 minutes, respectively (p<0.04). Gating efficacy in patients during the initial 10 minutes of SABR delivery correlated with efficacy during the full SABR session (~ 31 minutes). On average, beam latency effects had marginal impact on GTV t coverage during repeated inspiration breath‐holds, leading to a reduction in mean GTV t coverage by only ‐ 0.9%. Increased latency effects were seen when (i) relatively short breath‐hold gates occurred in combination with mean large tumor motion (>15mm, Peak‐to‐Peak), (ii) both very short and high frequent gates occurred, and (iii) different breathing phases were used during gating (e.g. mid‐ventilation), leading to system‐latency effects causing up to ‐3.0% in mean GTV t coverage. Analysis of the first, middle and last fractions revealed that some patients used different breath‐hold phases than light inspiration in 10/54 fractions; in those patients latency effects increased for tumors showing an initial mean motion >12mm. guided adrenal SABR. Material and Methods

which in‐house developed Python code performed soft‐ tissue (ST) tracking of the prostate in subsequent dynamics using a mutual information metric and rigid transformations. We validated the performance of the ST algorithm with previously obtained, ground truth marker tracking (MT) data of the same dataset. Results The algorithm was applied to 7645 dynamics from 139 sessions with a mean processing time of 5.47±0.77 sec (mean±stdev) per dynamic. The success rate (difference between MT and ST result < 1 mm) was 98.93%. We found group translations after 10 minutes of 0.05±0.81mm for X (LR) , 0.83±1.90mm for Y (AP), ‐0.90±1.85mm for Z (CC) and corresponding rotations of ‐0.49±2.13° about X, 0.09±0.58° Y and 0.09±0.73° for Z. After 10 min, 12% of all sessions had demonstrated a 3D displacement > 5 mm. Linear regression and Pearson correlation analysis indicated a good correlation and non‐significant difference with all p‐values < 1e ‐5 between the ST and MT algorithm in the X (R=0.934), Y (R=0.966) and Z (R=0.953) directions as shown in Fig. 1. An overview of the ST intrafraction visualization tool is provided in Fig. 2, showing the rotation, translation and full 3D segmentation of the prostate at a specific time point.

Figure 1 : Overview of the linear regression analysis between the ST and MT algorithm results in the three translation directions.

Figure 2: Overview of the prostate intrafraction visualization tool, showing the different slices of the prostate, the found rotations and translations and the current 3D segmentation of the prostate. Conclusion