ESTRO 36 Abstract Book

S821 ESTRO 36 2017 _______________________________________________________________________________________________

Adding a single aperture per iteration yielded the lowest cost function value per aperture included in the treatment plan. However, adding one aperture per modality every iteration resulted in a plan of comparable quality in only 120 iterations of the column generation loop instead of 580 for the single-aperture scheme. Conclusion MBRT planning produced a clinically realistic chest wall plan combining the advantages of photon and electron radiotherapy. The final plan was robust to initial conditions despite the iterative nature of column generation. This work opens the door to robust multi- modality planning and delivery. EP-1546 MR-Linac based single fraction ablative radiotherapy for early-stage breast cancer: a planning study T. Van 't Westeinde 1 , K.R. Charaghvandi 1 , J.K. Horton 2 , S. Yoo 2 , V. Scholten 1 , B. Van Asselen 1 , H.J.G.D. Van den Bongard 1 1 UMC Utrecht, Radiotherapy, Utrecht, The Netherlands 2 Duke University Medical Center, Radiotherapy, Durham, USA Purpose or Objective Our department is currently working on the implementation of an MRI-linear accelerator (MR-linac) for several tumor sites. Dose distribution in the presence of a magnetic field can be affected by the electron return effect (ERE), which can occur at tissue boundaries like skin and lung. Other MRL settings such as the fixed collimator and isocenter position may also influence the RT plan. We investigated the dosimetric feasibility of single fraction ablative radiotherapy in the prone and supine position for early-stage breast cancer using an MR-Linac approach. Material and Methods Preoperative contrast-enhanced (CE) CT and MRI scans were used from 10 cT1-2N0(sn) breast cancer patients included in an ongoing clinical trial on preoperative ablative radiotherapy. The gross tumor volume (GTV) was delineated on matched CE MRI- & CT-scans in the supine position. The clinical target volume (CTV) was created by expanding the GTV with 2 cm, thereby excluding skin and chest wall. The planning target volumes PTV GTV and PTV CTV were created by expanding both GTV and CTV 3 mm, excluding the skin. Prescribed doses were 20 Gy for PTV GTV and 15 Gy for PTV CTV . Rationale for dose prescription and organs at risk (OAR) constraints for a single fraction ablative RT were previously defined (1). Adequate target coverage was defined as 99% of the PTV should receive ≥95% of the prescribed dose. Intensity modulated radiation therapy (IMRT) plans were made in the presence of a 1.5T magnetic field, using Monaco Research version 5.19.01 planning system. 7 beams with individually chosen beam angles were used for each plan. Dosimetry was evaluated in all simulated plans.

Results For supine positioning the median volume that received at least 95% of the prescribed dose was ≥99% for PTV GTV and PTV CTV . The median GTV volume was 1.1 cc, the median CTV volume 72.9 cc, the median PTV GTV volume 5.3 cc and the median PTV CTV volume was 104.9 cc. The median ratio PTV CTV to ipsilateral breast was 11.6%. The predefined OAR constraints were achieved in all plans (table 1).

Conclusion Single fraction ablative radiotherapy in supine position on the MR-Linac is dosimetrically feasible. The feasibility of prone MR-linac treatment will be available at the 36 th ESTRO conference. (1) Charaghvandi RK, den Hartogh MD, van Ommen et al. MRI-guided single fraction ablative radiotherapy for early- stage breast cancer: a brachytherapy versus volumetric modulated arc therapy dosimetry study. Radiother Oncol 2015 Dec;117(3):477-482. EP-1547 Optimal treatment planning for H&N: evaluation of a predict parotid glands sparing tool N. Delaby (France), S. Martin, O. Henry, E. Chajon, C. Lafond 3 1 Centre Eugène Marquis, Radiotherapy, Rennes CEDEX, France 2 INSERM, U1099, Rennes, France 3 University Rennes 1, LTSI, Rennes, France Purpose or Objective The complexity of the clinical objectives in IMRT yields a variability in treatment planning, especially between operators. A major difficulty is currently to appreciate the optimality of treatment plans. A previous published model