ESTRO 37 Abstract book

ESTRO 37

S549

Purpose or Objective For brain metastases, we have shown previously that significant tumor shifts occur over time. For fractionated stereotactic radiosurgery (SRS), not only time but also the treatment itself may cause changes resulting in underdosage of tumor volume and an unintended dose to normal tissue. Therefore, we investigated whether the initially defined target volume covers the target volume during fractionated treatment by repeating the MRI halfway through the course of fractionated SRS. Material and Methods Twenty patients received a repeat-MRI during fractionated SRS for brain metastases in 3 or 5 fractions (overall treatment time 5-12 days), and were analyzed (SRS only n=8, SRS postoperative n=12 patients). The target volume (i.e. T1 contrast enhancing tumor or postoperative cavity) on both the initial MRI and the repeat-MRI were delineated on the T1-weighted sequence (TE= 2.4 ms, TR= 5.1 ms) post gadolinium contrast by one observer. The target volume (V outside ) on the repeat-MRI that was outside the initial target volume and the maximum distance (d outside ) of the target volume on the repeat-MRI outside the initial target volume were analyzed. Results

"underdosing" the tumor. We are currently analyzing the consequences on the tumor- and normal tissue dose, and continue to include patients in our repeated MRI study. PO-0988 Automatic fiducial tracking on 4D cine-MRI for MR-guided prostate radiotherapy A. Pathmanathan 1 , A. Andreychenko 2 , D. De Muinck Keizer 2 , L. Kerkmeijer 2 , A. Tree 1 , C. Van den Berg 2 , D.B. JCJ 2 1 The Institute of Cancer Research and The Royal Marsden NHS Foundation Trust, Radiotherapy and Imaging, Sutton, United Kingdom 2 University Medical Centre Utrecht, Department of Radiotherapy, Utrecht, The Netherlands Purpose or Objective Our ultimate goal is to develop optimal real-time MR- guided radiotherapy (RT) for extreme hypofractionation for prostate cancer. We have therefore collected an extensive, realistic 4D cine-MR dataset to study intrafraction motion of the prostate over a hypofractionated schedule. Here, we present the first step in our analysis: the fast automatic tracking of fiducial markers (FM) to establish ground truth motions. Material and Methods Thirty patients undergoing hypofractionated prostate RT, with four implanted cylindrical gold FMs (1 mm diameter, 5 mm length), had repeated MR imaging sessions in a multicenter Medical Ethics board approved study. These sessions included cine-MRI examinations prior to each of five weekly fractions. Patient positioning and immobilization was similar to that during RT. A balanced 3D gradient echo sequence was used, optimised for anatomical and FM contrast. Each cine-MR examination consisted of 55 3D datasets (‘dynamics’) acquired sequentially at the start of each imaging session. Each dynamic was acquired over a 11 second period, with a voxel size 0.96×0.96×2 mm 3 and a 384×384×120 mm 3 field of view, yielding a total acquisition time per examination of 10 minutes, similar to the duration of a RT fraction. A clinician manually identified the top and bottom of each FM on the first dynamic of each cine-MR dataset, to determine the FM centre precisely. FM centres in subsequent dynamics were determined automatically using in-house developed MATLAB code (MathWorks Inc., Natick, MA, USA) by means of the watershed transform applied locally around each FM (radius 7 mm). FM patterns from the first dynamic were correlated by template matching with subsequent dynamics. Rigid transformations of the FM template between first and subsequent dynamics were calculated, yielding the center of mass (COM) translation and template rotations. FM registration errors (FRE) and rotational angle errors, the difference between automatic detected FM and FM positions derived from rigid transformation, were calculated based on the assumption of FM template In total, 141 cine-MR examinations (7,755 dynamics) were acquired. The average absolute fiducial registration error was 0.4 mm whereas the average absolute rotation angle error was 0.7°, confirming both a high rigidity as well as high automatic registration accuracy. Intrafraction motion was strongly patient and session dependent, affected by factors including bladder and rectal filling. Figure 1 is an example of COM translations in a single cine-MR exam. A clear but small (~ 1mm) overall group rigidity. Results

Change in volume between initial MRI and

Volume initial

Volume repeat-

V outside (cc)

d outside (mm)

MRI (cc)

MRI (cc)

repeat- MRI (cc)

2.5 (0.1– 12.6 cc) 0.5 (0.1–

2.5 (1.4– 7.3 mm) 1.8 (0.8– 5.9 mm)

30.4 (6.3– 62.0 cc)

19.3 (4.5– 68.5 cc)

2.0 (-6.5– 12.5 cc)

SRS only

(n=8)

15.1 (2.8– 84.1 cc)

15.5 (2.6– 72.9 cc)

-0.6

(-

SRS postoperative (n=12)

11.2–3.3 cc)

3.8 cc)

Table 1. Results are given in median and range The median time between the initial and repeat-MRI was 8 days (5-14 days). The target volumes determined on the initial and repeat-MRI are given in Table 1. The initial volume of the target did not change significantly towards the second MRI for both the SRS only and SRS postoperative group (Wilcoxon Signed Ranks Test p =0.16 and p =0.16). For SRS only , a median V outside of 2.5 cc (0.1-12.6 cc) was observed, which was 24.0% (0.7-72.2%) of the initial target volume. The median d outside was 2.5 mm (1.4-7.3 mm). There was no correlation between d outside and the time between the two MRIs ( r s = 0.012, p =0.98). For SRS postoperative a median V outside of 0.5 cc (0.1-3.8 cc) was observed, which was 4.7% (0.3-27.7%) of the initial target volume. The median d outside was 1.8 mm (0.8-5.9 mm), and was also for this group not correlated with the time between the two MRIs ( r s = -0.046, p =0.89). For 50% of the patients in the whole group the maximum distance between the two target volumes exceeded the 2 mm. Conclusion For all patients, we found that a part of the target volume was outside the initial defined target volume halfway the treatment. For 50% of the patients this volume reached a distance exceeding commonly used margins in SRS. This phenomenon (which was not correlated with time) leads to the risk of partial