ESTRO 36 Abstract Book

S989 ESTRO 36 2017 _______________________________________________________________________________________________

Systems, 3 fields, multi-field optimization, NUPO/PCS V13.7.14) and compared with the clinical VMAT plans (Varian Medical Systems, RapidArc, 2 arcs, 6MV photons, Acuros V13.6.15). All patients had a simultaneous integrated boost with a fractionation scheme of 35 x 1,55/2 Gy for PTV-elective/PTV-boost. For all plans, the focus was on sparing of all individual salivary and swallowing structures. Patients were categorized according to the amount of additional OAR sparing achieved with proton plans represented by the sum of mean doses to (1) all swallowing structures, (2) parotid gland(s) and (3) submandibular gland(s), if the mean dose was <45Gy for any of these structures using IMPT and/or VMAT. Results The range for the sum of the three mean doses for photon plans was 37.2-180.8Gy (median 115.4), and for proton plans 21.5-161.9Gy (median 92.1). Differences in total mean dose between photon and proton plans ranged from 1.2-37.2Gy (median 16.9). Assuming a threshold of 10Gy sparing in sum of mean doses (i.e. proton plan achieved at least 10Gy reduction in total mean dose): 82.5% of proton plans achieved this; threshold 15Gy: achieved by 67.5% proton plans; threshold 20Gy: achieved by 40.0% proton plans; threshold 25Gy: achieved by 17.5% proton plans. Conclusion The vast majority of patients had a dosimetric benefit with protons as assessed by mean dose to swallowing and salivary gland structures. The results show that the number of patients who would be selected/referred for protons is highly sensitive to the choice of threshold. This has implications for activity levels in proton and photon departments. EP-1833 Bowel doses in cervical cancer patients treated with a full bladder during radiotherapy. A. Schouboe 1 , E. Kjaersgaard 1 , N. Jensen 1 , L. Fokdal 1 , L. Nyvang 2 , M. Assenholt 2 , J. Lindegaard 1 , K. Tanderup 2 , A. Vestergaard 2 1 Aarhus University Hospital, Department of Oncology, Aarhus N, Denmark 2 Aarhus University Hospital, Department of Medical Physics, Aarhus N, Denmark Purpose or Objective Large inter-fractional organ motion is a challenge in cervix cancer radiotherapy. To reduce inter-fractional organ motion, a bladder filling instruction aiming for a comfortably full bladder has been introduced with daily Cone Beam CT (CBCT). The purpose of the study is to evaluate the correlation between bladder filling variation and bowel dose. Material and Methods Eight consecutive patients with locally advanced cervical cancer treated with chemo radiotherapy, were included in the study. Dose planning and treatment was performed in supine position. Prescribed dose was 45-50 Gy in 25-30 fractions to the pelvis, 5 fractions per week. Daily CBCT with bony fusion and couch correction was performed. A bladder filling protocol was applied with verbal and written instructions, advising patients to drink 450 ml of water in 15 minutes post micturition, 1 hour prior to each treatment. All patients were retrospectively re-planned with a uniform dose of 45 Gy in 25 fractions using VMAT. Target and organs at risk were delineated on CT images for dose planning and on every CBCT (n=210). The clinical target volume (CTV) encompassed the gross tumour volume (GTV), cervix, parametria, uterus, upper vagina and the nodal CTV. A PTV margin of 1.5cm was applied for the GTV, cervix, and uterus and 5mm for the nodal CTV and parametria. Outer extension of bowel loops were delineated including sigmoid in one volume (Bowel) on each of the CBCTs. Based on a bony fusion, all structures were transferred to the planning CT to asses bowel V30 Gy and V43Gy for each fraction. Changes in bowel V30 and

V43 of bowel were evaluated as a function of changes in bladder volume compared to the parameters on the planning CT. Results A large variation in bladder filling was observed and was found to be patient dependent. Large variation in bladder volume and uterus position resulted in large variation in V30 (Table 1). A linear correlation was found between bowel V30 and V43 and bladder volume. Inter-fractional bowel motion was observed due to changes in bladder volume. Linear regression showed that with an increase of 100 cc in bladder volume, the bowel V30 is decreased by a mean of 58 cc (range: 10 – 87 cc). For one patient, the position of the uterus was not affected by bladder filling changes and there was no clear correlation between bladder volume changes and bowel irradiation (Fig 1). Conclusion A correlation between increased bladder filling volume and dose to the bowel were established. Based on these results it is recommended to focus on a uniform bladder filling during planning CT and radiotherapy in order to reduce the dose to the bowel during treatment.