Abstract Book

ESTRO 37

S520

close to the diaphragm base-line shifts of 11 mm were incidentally observed in the cranial caudal direction. The lateral and AP tumor displacements were generally smaller, however two patients with a medium-distal tumor exhibited large lateral tumor shifts during the treatment course (e.g. Figure 1). To account for these uncertainties CTV-to-PTV margins varying between 9 mm and 12 mm have to be applied (Table 1).

2.

Rigid registration between the planning CT and each CBCT was performed based on bony anatomy. CTV for the vaginal vault and upper vagina (CTVv), bladder and rectum were manually contoured on each image set. Each CBCT was deformed to the planning CT by computed deformable image registration (DIR) on a commercial treatment planning system (RayStation v5, Raysearch Laboratories). CTVv motion using non-rigid approximation was determined by deformation vector field (DVF) displacement of voxels within a 1mm internal wall of the CTVv. This was divided into halves to create structures representing the anterior (A), posterior (P), right (R) and left (L) walls (figure 1a and 1b). Rigid approximation of target motion was determined by translation of the most extreme R, L, A and P positions of the CTVv structure, relative to the mid femoral head. These measurements were taken at each 2.5 mm slice on which the CTVv was defined (figure 1c). Mean target motion using both methods was calculated for each patient and compared using paired t-test. Patient and population systematic (Σ) and random (σ) error were calculated for internal motion. For the rigid approximation data, the van Herk formula (2.5Σ + 0.7σ) was used to calculate the margins required to account for observed errors (1) and the PTV to encompass the population target motion and population set-up error was determined (2) .

3.

4.

5.

6.

7.

8.

9.

Conclusion This MRI study in a population of 15 esophageal cancer patients revealed substantial tumor displacements with respect to the bony anatomy that are unlikely to be picked up on CBCT images. To properly account for these uncertainties CTV-to-PTV margins of 9, 8 and 12 mm in LR, AP and CC directions respectively need to be incorporated. PO-0952 Validation of margin recipes in endometrial cancer IMRT using deformable image registration I. White 1 , D. McQuaid 1 , A. Dunlop 1 , S. Court 2 , N. Hopkins 1 , M. Caputo 3 , H. McNair 1 , S. Lalondrelle 1 1 The Institute of Cancer Research and Royal Marsden Hospital NHS Trust, Academic Radiotherapy, Sutton, United Kingdom 2 St George's Hospital NHS Trust, Physics, London, United Kingdom 3 European Institute of Oncology, Physics, Milan, Italy Purpose or Objective To quantify internal motion of the vaginal vault using both rigid and non-rigid approximation of cone beam CT (CBCT) and to determine the validity of current margin recipes in patients treated with IMRT for endometrial cancer. Material and Methods 1. CBCTs were acquired in 17 post-hysterectomy patients receiving adjuvant IMRT for endometrial cancer.

Results A total of 169 CBCTs were analysed. Results are illustrated in table 1. There was a significant difference between mean patient motion using the different registration strategies (p <0.05). DVF displacements of up to 30 mm occurred in the A/P direction which was not accounted for in PTV margins using the rigid margin recipe. No consistent relationship was found between rectal and bladder volume and time, or rectal and bladder volume and CTVv motion.