Abstract Book

ESTRO 37

S513

6MV photon beam with and without respiratory motion surrogates. A Scandidos Hexamotion™ motion platform was programmed with a periodic trajectory with 15mm amplitude and 4s period. The detected motion was measured with two RPM systems, one wall-mounted and one couch-mounted, using both surrogates. Clinical performance was evaluated retrospectively for 20 breast cancer patients (10 monitored with each surrogate) treated between November 2016 and February 2017 with deep inspiration breath hold radiotherapy to 50Gy in 25 fractions. For each patient, MV verification images of the medial field were acquired during treatment every 5 th fraction. The last two fractions with MV verification images were selected (Typically the 16 th and 21 st ) and rigidly registered to DRR. Distance to agreement was measured and compared between the two patient groups. Student’s t-test was used for all comparisons.

Conclusion The novel respiratory motion surrogate has replaced the Varian two-marker box in clinical use in our clinic due to its superior versatility in positioning, improved skin sparing and identical clinical performance. It has so far been used on (>200 breast, lymphoma and lung cancer patients) regardless of anatomy or reconstructive surgery. PO-0943 QA of lung cancer patients treated in deep- inspiration breath-hold using intra-treatment imaging M.L. Schmidt 1 , L. Hoffmann 1 , P. Poulsen 1 , M. Knap 1 , N. Voldby 2 , J. Bertholet 1 , D. Møller 1 1 Aarhus University Hospital, Department of Oncology, Aarhus C, Denmark 2 Aarhus University Hospital, Department of Pulmonology, Aarhus C, Denmark Purpose or Objective Respiratory target motion results in geometrical uncertainties in lung cancer radiotherapy. The target motion can be reduced by gated treatment in deep- inspiration breath-hold (DIBH), but uncertainties due to lack of intra-fractional DIBH reproducibility are introduced. This study uses imaging of internal implanted markers in involved lymph nodes (LN) in setup cone-beam CT (CBCT), intra-treatment fluoroscopy and post-CBCT to investigate the time-resolved DIBH position of the markers throughout the radiotherapy treatment. Material and Methods Five non-small cell lung cancer patients had 1-2 fiducial markers implanted in LN targets by EBUS and were treated with IMRT in DIBH, using an external gating box (Varian RPM) as a surrogate of the target position during DIBH. A daily setup CBCT for online soft tissue match on the primary tumor was acquired with a full gantry rotation in DIBH divided into three parts, i.e. acquired in 3x20 seconds. At fractions 1-3 and afterwards once a week, 5Hz fluoroscopic kV images were acquired orthogonal to the MV treatment beam during treatment delivery. Each treatment beam could be delivered within 20 seconds during a single breath-hold. A post-CBCT was acquired at the fractions with intra-treatment imaging. Offline, the marker positions were segmented in each CBCT projection and fluoroscopic kV image. From the segmented marker positions in the 2D images, the 3D marker trajectories were estimated with a sample rate of 11 Hz during CBCT acquisition and 5 Hz during treatment delivery. The marker position was compared during setup CBCT, treatment delivery and post-treatment CBCT, and the marker position reproducibility was investigated.

Results Surface dose was enhanced by up to 45% and 235% (Figure 2) and the area of at least 25% surface dose enhancement was 9.7cm 2 and 71.7cm 2 under the NRS and TMB, respectively. Mean detected motion of the phantom was 14.75±0.05mm and 15.05±0.05mm respectively for the NRS and TMB. Mean distance to agreement of the MV verification images was 3.7±1.1mm and 3.6±0.7mm (Table). No statistically significant differences were found for intersystem reproducibility or clinical performance. Vrt [mm] Lng [mm] Lat [mm] 3D [mm]

TMB -0.9 (1.6) -1.2 (2.0) 0.1 (1.1) 3.6 (0.7) NRS 0.4 (1.6) -2.9 (1.4) 0.6 (0.9) 3.7 (1.1)

Table: Mean values and standard deviations of the distance to agreement of MV verification images are shown for the vertical, longitudinal, lateral directions and total distance. No significant difference were found between the surrogates (p>0.5).