Raystation treatment planning system

A total of 80 SBRTs, using the volumetric modulated arc therapy (VMAT) technique, was designed. The patients’ original prescriptions were 33 Gy in 5 fractions. According to our institution’s RT planning protocol for pancreatic cancer, the plans were optimized to achieve 33 Gy in five fractions to 85% of the PTV. The initial treatment plans were generated using volumetric modulated arc therapy (VMAT), for which the PTV definition included breath-hold variation and a 2 mm setup margin. The beam arrangements included one or two full arcs per each patient’s breath-hold constraints. Pancreas, duodenum, and bowel objectives were 1 cc below 33 Gy and 20 cc below 20 Gy. Each plan was reviewed by our institution’s radiation oncology group and approved by the attending physician before delivery. Evaluation doses were generated using the RayStation treatment planning system (TPS; RaySearch Laboratories, AB, Stockholm, Sweden) by propagating the initial radiation fields onto co-registered breath-hold CTs acquired at the time of treatment simulation.

Radiotherapy contours delineation was performed by a radiation oncologist using the software MIM following EORTC guidelines^{31 ,32 (link)} and contours were reviewed by a second senior radiation oncologist. The gross tumor volume (GTV) was defined from the T1CE and included the resection cavity and any contrast-enhancing margins. The clinical target volume (CTV) was determined by 1.5 cm expansion of the GTV and modified to anatomical boundaries, and planning target volume (PTV) by 0.3 cm expansion of the CTV (Figure 3). T1CE images with associated radiotherapy structures were imported into RayStation treatment planning system (v.10B, RaySearch Laboratories). Organs at risk (OARs) were delineated in RayStation from the T1CE image by a radiation therapist with the aid of a MRI-based OARs atlas, following EORTC guidelines.^{31 ,32 (link)}

The irradiation plan was carried out as a step and shoot intensity-modulated radiotherapy (IMRT) plan, describing the different dose levels as separate target regions. Planning was done with Raystation treatment planning system (RaySearch Laboratories, Stockholm, Sweden) based on a CT scan of the hypoxia chamber containing 96-well plates filled with water. Irradiation was performed on a Siemens Artiste (6 MV) (Siemens, München, Germany).

The RayStation treatment planning system (versions 9A/10A, RaySearch Laboratories, Stockholm, Sweden) was used for this study. Proton dose engines in RayStation require a complete description of the material composition, including mass density, mass fraction of atomic elements, and mean ionization energy for each voxel of the patient. With the use of CT images as input, this was implemented by converting HU to mass density with the HU-D table. Then, the mass fraction of atomic elements and mean ionization energy of the voxel were determined from a number of well-established core materials through the interpolation of mass density (13 ).
A stoichiometric calibration method (14 (link), 15 (link)) was used to establish the HU-D table for the CT simulator used in this study (Siemens SOMATOM Definition Edge plus, Siemens Healthcare, Forchheim, Germany). Separately, a patient group-based method (11 (link)) was used to create the HU-D table for the CBCT system used in this study (integrated into the Hitachi Probeat CR proton delivery system, Hitachi, Ltd., Tokyo, Japan). CT and CBCT image datasets of brain, head and neck, and thorax patients were used to establish the HU-D relationship on the CBCT images for typical materials such as air, brain, bone, or lung. The HU-D tables for CT and CBCT were then used for proton dose calculation on images acquired with respective modalities.