Abstract Book

S158

ESTRO 37

OC-0303 Treatment plan comparison between a 1.5 T MRI-Linac and a standard Linac for esophageal cancer M. Nachbar 1 , D. Mönnich 1 , P. Kalwa 1 , D. Zips 2,3 , D. Thorwarth 1,2 , C. Gani 3 1 University Hospital Tübingen, Section for Biomedical Physics- Department of Radiation Oncology, Tuebingen, Germany 2 German Cancer Research Center DKFZ, German Cancer Consortium DKTK partner site Tübingen, Heidelberg, Germany 3 University Hospital Tübingen, Department of Radiation Oncology, Tuebingen, Germany Purpose or Objective This study compares radiotherapy treatment plans for a 1.5 T MRI-Linac and a state of the art clinical treatment system for thoracic esophageal squamous cell carcinoma patients. Material and Methods Twelve patients were included in this planning study. Patients were replanned with an initial plan to a total of 50 Gy covering the primary tumor, affected lymph nodes and elective nodal areas. A subsequent boost plan covered the primary tumor to a total of 60 Gy (2 Gy / fraction). The first plan was created for the Versa HD treatment system (Elekta AB, Stockholm, Sweden) with volumetric modulated arc therapy (VMAT) without the influence of a magnetic field. The second plan was created for the hybrid 1.5 T MRI-Linac using a nine beam step-and-shoot technique with optimized beam angles. Realistic MR-linac plans were generated including the following key parameters. Source-to-isocentre distance: 143.5 cm, leaf width at isocentre: 7.15 mm, fixed isocentre 13 cm above the table. Furthermore, the MR- linac table as well as the RF-coil needed for MR imaging were considered during plan optimization. All plans were generated with Monaco Research 5.19.03 (Elekta AB, Stockholm, Sweden). Treatment plans were optimized to achieve similar target volume coverage. Differences between nine relevant DVH parameters were analysed with a Wilcoxon signed rank test in R 3.3.2. Results Treatment planning was feasible for ten patients. For two patients the cranio-caudal extension of the target volume exceeded the maximum field size of 22 cm of the MRI- Linac. For all patients, both generated treatment plans met dose volume histogram (DVH) criteria with respect to organ at risk sparing and target coverage. Three out of nine DVH parameters showed significant differences (p<0.05). Significantly higher values for the MRI-Linac plans were observed for lung V 5 , the mean lung dose and the mean heart dose. The absolute differences of all parameters between the conventional and the MR-linac plan are shown in Fig. 1.

Conclusion NTCP-model based selection of prostate patients for hypofractionation is feasible. The number of patients eligible for hypofractionation depends on applied thresholds for NTCP increases. Model-based selection could be used in shared decision making. OC-0302 Endorectal balloon for prostate SBRT: dosimetric impact on erectile function preservation? G. Lamanna 1 , A. Dubouloz 1 , M. Rouzaud 1 , R. Miralbell 1 , T. Zilli 1 1 HUG, Radiotherapy, Genève, Switzerland Purpose or Objective One of the most common side effects from radiotherapy treatment of localized prostate cancer remains erectile dysfunction. Vessel-sparing radiotherapy has shown promising results on preserving erectile function when compared to historical series. We hypothesized that the use of an endorectal balloon (ERB) for prostate stereotactic body radiotherapy (SBRT) might improve erectile function preservation by decreasing the dose to the internal pudendal arteries (IPA). The aim of this dosimetric study was to compare the IPA dose in 10 prostate cancer patients simulated with and without ERB before SBRT of the prostate. Material and Methods Ten prostate cancer patients simulated with and without ERB were planned to receive 36.25Gy (5x7.25Gy) to the planning treatment volume (PTV) while the dose to urethra (urethral planning risk volume, uPRV) was reduced to 32.5 Gy. IPAs were delineated on both planning CT scans with MRI registration. Plans with and without ERB were optimized using a volumetric modulated arc (VMAT) technique within well defined dose constraints. No optimization was undertaken on IPA. SBRT doses to IPA (D max and D mean ) were compared between groups using non-parametric tests. Results IPA volumes were similar in the two CT datasets, with a median volume of 4.7 cc (range, 3.8-7.2) versus 5.1 cc (range, 3.8-7) in patients with and without ERB (p=NS), respectively. No differences in target coverage, rectal, and bladder walls doses were observed between the two groups. The dose delivered to IPA was significantly lower in patients simulated with ERB compared to patients planned without. Median IPA D max decreased from 26.8 Gy (range, 16.2-37.7) to 21.1 Gy (range, 13.5-36.9) with ERB (p=0.013). Similarly, the median IPA D mean decreased from 15.4 Gy (range, 8.8-23.0) to 11.5 Gy (range, 8.7-17.1) with ERB (p=0.009). Thus, with ERB the mean decrease in IPA dose was a 16% and 22% for D max and D mean , respectively, compared to patients planned without ERB. Conclusion The use of ERB to treat prostate cancer patients with SBRT showed a significant sparing of IPA compared to plans generated with an empty rectum. As no specific dose constraints are available for vessel-sparing radiotherapy, maximal reduction on IPA doses should be attempted in SBRT treatments to maximize erectile function preservation.

Fig.1 Differences between the plans generated for the conventional Versa HD Linac with a VMAT technique and the MRI-Linac with a step-and-shoot technique. Positive differences mark an increase in the respective metric for the MRI-linac plans. Conclusion