ESTRO 36 Abstract Book

S185 ESTRO 36 2017 _______________________________________________________________________________________________

the HYBRID trial in November 2013 and then utilised for a further 14 RAIDER centres in June 2015. The Mann-Whitney U test and Kruskal-Wallis test were used to investigate if there were any significant difference in assessment scores between the two trials and individual staff grading respectively. Results The POD assessment was completed by 244 individuals (HYBRID=73, RAIDER=171) from 24 recruiting centres. The median assessment score was 92% (range: 25-100%) and 86% of individuals achieved the score required for QA approval on their first attempt. The distribution of assessment scores in RAIDER was found to be significantly different to HYBRID (p=0.034). Individuals would be more likely to achieve a score ≥83% for RAIDER (90%) than HYBRID (77%). Each centre had an average of 10.2 (Range 3-23) individuals complete the POD assessment. There was no statistically significant difference in assessment scores between different staff grades (p=0.713). The median assessment scores varied between recruiting centres, with a range between 83% and 100%. Conclusion A high median score was achieved by the individuals that completed the assessment indicating consistent POD selection with the expert consensus answers across individuals and centres. The POD assessment proved to be a feasible way of credentialing multiple individuals across all recruiting centres. Differences in assessment scores between HYBRID and RAIDER trials likely reflect the increased experience with pelvis CBCT in UK centres between 2013 and 2015. Individuals of all grades were able to successfully complete the assessment indicating the importance of appropriate local IGRT training rather than staff seniority when choosing the POD. OC-0352 CBCT-guided evolutive library for cervix adaptive IMRT B. Rigaud 1,2 , A. Simon 1,2 , M. Gobeli 3 , C. Lafond 1,3 , D. Williaume 3 , J. Leseur 3 , J. Castelli 1,2,3 , P. Haigron 1,2 , R. De Crevoisier 1,2,3 1 INSERM, U1099, Rennes, France 2 Université de Rennes 1, LTSI, Rennes, France 3 Centre Eugene Marquis, Radiotherapy Department, Rennes, France Purpose or Objective In the context of locally advanced cervix carcinoma adaptive radiation therapy (ART), this study aimed to simulate five treatment strategies, including an original CBCT-guided evolutive library. We compared geometrically the strategies by considering the coverage by the simulated PTVs for both CTV and OARs. Material and Methods Sixteen patients having received a total dose of 45 Gy by IMRT for locally advanced cervix carcinoma were included. Each patient had: three planning CTs corresponding to three bladder volumes (empty (EB), intermediate (IB) and full (FB)), a CT scan at 20 Gy and bi-weekly CBCTs for 5 weeks. The CTV and the OARs were manually delineated on each CT and CBCT. Five radiotherapy (RT) strategies were investigated (Figure 1): (1) “Standard RT” based on one planning CT with IB and considered as reference, (2) “ITV-based RT” with an ITV built from the three planning CT scans, (3) “RT with one midtreatment replanning” corresponding to the standard RT with replanning at 20 Gy, (4) “Pretreatment library ART” using the three planning CTs (EB, IB, FB) to define the plan of the day by a CTV overlapping criteria, and (5) “Evolutive library RT” corresponding to the 4 th strategy enriched by including CBCTs anatomy in the pretreatment library if the daily CTV shape was highly different (overlap scores between the library and the daily CTV). For each strategy, two PTV margins were used (7 and 10 mm). All the strategies were geometrically compared by considering, on the CBCTs, the percentage of coverage of the CTV or the OAR by the

strategy PTVs. The under-coverage of the CTV by the PTV was investigated using elastic registration.

Results The “Evolutive library RT” corresponded to a mean number of one per-treatment replanning (up to 3). For 50% of the cohort, no per-treatment replanning was needed. The table shows the CTV and OARs coverage by the PTV for all strategies. The evolutive library strategy provided the highest CTV coverage compared to the other strategies corresponding to a mean CTV coverage (min – max) of 98.3 % (96.4 – 100%) with 10mm margins and to 96.0 (93.0 – 99.7) with 7mm margins (p<0.05). Moreover, this strategy significant decreased the bowel-PTV overlapping.

Conclusion The “Evolutive library RT” strategy increased the CTV to PTV coverage, while not increasing the PTV bladder overlapping and even decreasing the bowel to PTV overlapping. OC-0353 Implementation of RTT led ‘plan of the day’ adaptive radiotherapy in cervical cancer