ESTRO 37 Abstract book

ESTRO 37

S608

Results The best PoD was chosen by 60% of the users (Figure 2). The choice of the best PoD was however correlated to the Dice range (r= 0.55 p=0.005): Higher the DR was, the best was the choice (Figure 2). User’s agreement of choice was significantly correlated with the DSC range (r =0.53, p<0.01) (Figure 2). When considering a Dice range value > 0.12 (median value), the best PoD was chosen by 72% of the users.

Purpose or Objective Endorectal balloon (ERB) has been recently applied to prostate radiotherapy (RT) to reduce intrafractional prostate motion and anorectal toxicities. However, the interfractional reproducibility of ERB is not well investigated. This study aimed to evaluate the reproducibility of planned target volume (PTV) and organ at risks (OARs) when ERB was applied, by comparing the dose discrepancy between plan simulation and treatment, as well as comparison with conventional techniques. Material and Methods 17 patients diagnosed with localized prostate cancer (T1- T2b) were retrospectively recruited. 9 patients were treated with ERB (ERB group) using Stereotactic Body Radiotherapy (SBRT) within 3 weeks; 8 patients were treated with conventional Volumetric Modulated Arc Therapy (VMAT) without ERB (control group) for approximately 7 weeks. For the control group, SBRT plans were generated with the same planning techniques and dose constraints of the ERB group as adopted from the RTOG 0938. The first 3 sets of weekly cone beam computed tomography (CBCT) from both groups were exported to represent the actual patient position for dosimetric analysis. With deformable registration of the planning CT to CBCT, the deformed planning CT sets were used for dose re-calculation under the original treatment plan to obtain actual delivered dose to the patient. The dose discrepancies were expressed as the absolute difference of PTV and OARs dose parameters (D max , D min , V 36.25 and V 34.4 for PTV; D max , V 34.4 , V 32.625 , V 29 and V 18.125 for rectum) between delivered treatment dose and original planning dose. These dose discrepancies were compared between ERB group and control group in accordance to the corresponding treatment weeks. Results Treatment with ERB application showed less PTV and rectal dose discrepancies, which means the prostate and rectal reproducibilities are better in ERB group. For PTV, D min and V 34.4 discrepancies of the second week were significantly lower in ERB group (p < 0.05). For rectum, the mean rectal dose discrepancies of all weeks were lower in ERB group. Among them, V 34.4 and V 29 discrepancies of first week, V 32.625 and V 29 discrepancies of second week, and V 32.625 discrepancy of third week were significantly lower in ERB group. No significant difference was found in bladder or femoral head dose discrepancies. Conclusion Dose discrepancy highly depends on the reproducibility of treatment positioning. Treatment with ERB application showed a reduction in PTV and rectum dose discrepancies from planning CT. The utilization of fixed shape ERB could minimize the interfractional variation of rectal shape, and exert constant pressure to keep the prostate in a fix shape and position. As a result, ERB improves the reproducibility of treatment positioning and therefore facilitates a more accurate treatment to be delivered. PO-1080 4DCT oesophageal tumour delineation in SCOPE2 – how is radiotherapy quality assurance beneficial? S. Cox 1 , G. Jones 2 , G. Radhakrishna 3 , S. Mukherjee 4 , M. Hawkins 5 , T. Crosby 6 , S. Gwynne 1 1 South West Wales Cancer Centre, Oncology Department, Swansea, United Kingdom 2 Velindre Cancer Centre, Radiotherapy Physics Department, Cardiff, United Kingdom 3 The Christie Hospital, Oncology Department,

Figure 1: Exemple of the case number 6.

The CTV was manually contoured on CBCT images to define the best PTV choice using the Dice similarity coefficient (DSC). The 3 PTV corresponding to 3 the bladder filling (empty, intermediate, and full) were bone registered on the CBCTs. The Die score is defined by the overlapping between the PTV and the CTV. The Dice range is defined by the difference between the highest and the lowest Dice values.

Figure 2: Representation of the PPL PTVs choices for all cases.

The colors represent different bladder filling, such as empty (blue), intermediate (red) and full bladder (green). The values in the colours bars correspond to the Dice Score. The best PoD (highest Dice score) is identified by the start (*). Conclusion The choice of the best PoD appears difficult, particularly in case of superposed PTVs (fixed uterus) or CBCT artefacts. Such ART strategy cannot therefore be proposed in routine treatment for cervix cancer, but within prospective trial (ongoing phase II NCT02937948). PO-1079 Study of SBRT Planning Dose Reproducibility With Endorectal Balloon In Localized Prostate Cancer Y. Poon 1 , M.C. Ng 2 1 Prince of Wales Hospital, Department of Clinical Oncology, New Territories, Hong Kong SAR China 2 St. Teresa's Hospital, Oncology Centre, Kowloon, Hong Kong SAR China