ESTRO 2020 Abstract Book

S522 ESTRO 2020

Results Median GLS was -20.5% [-16 ; -28.5] before RT and -20.2% [-8.2 ; -29.4] 6 months after with a mean variation of -1.1% [-18.3 ; +6.1.]. As defined, 6 patients (7.8%) had GSL alteration at 6 months. AB, SB, AM and A segments showed significantly more alterations than the other segments. Regional alterations are summarized in table 1. Mean Dmean in altered segments was significantly higher the non-altered segments 7.7Gy vs 5.4Gy (p=0.02). Conclusion The results of our study show that radiation dose to the different heat segments is correlated with their subclinical alteration. Global heart delineation seems to be insufficient during the planning of breast radiotherapy. Segmental delineation of the LV may be an interesting alternative to limit segmental doses to reduce the risk of subclinical alterations. A mean dose of 5 Gy could be proposed in exposed heart segment. PO-0979 Variability of axillary nodal CTV contouring in breast cancer (PART I - geometric evaluation) M.C. Leonardi 1 , F. Cattani 2 , M. Pepa 1 , R. Luraschi 2 , S. Vigorito 2 , S.G. Gugliandolo 1 , D. Ciardo 1 , R. Orecchia 3 , B.A. Jereczek-Fossa 1 , A. Breast Cancer Study Group 4 1 IEO- European Institute of Oncology IRCCS, Division of Radiotherapy, Milan, Italy ; 2 IEO- European Institute of Oncology IRCCS, Department of Medical Physics, Milan, Italy ; 3 IEO- European Institute of Oncology IRCCS, Scientific Directorate, Milan, Italy ; 4 Associazione Italiana di Radioterapia e Oncologia Clinica, Airo, Milan, Italy Purpose or Objective To determine the interobserver variation in axillary lymph node definition in breast cancer (BC) by comparing the nodal single centre CTV (SCCTV, drawn by radiation oncologists, ROs, of each participating institution) with the nodal gold-standard CTV (GSCTV). Material and Methods This analysis was an expansion from a previous study, where the axillary nodal variability between institutions and observers was investigated for three representative patients, P1, P2, P3, with increasing level of anatomic complexity, and it formed the basis for further dosimetric assessment. In the present analysis, a GSCTV, created in DICOM format by the median of all the axillary nodes CTVs drawn by the BC experts of the radiotherapy (RT) centres originally involved, was reviewed by 3 independent BC experts and one radiologist to reach a consensus and validated using the STAPLE outline. For each case patient, the nodal CTVs (SCCTV) drawn by 18 ROs, all of them already participating in the previous work, were chosen to make comparison with GSCTV. For each patient and each centre, contours of the three considered node levels (L2, L3 and L4) and the CTV, both as SC and GS, were retrieved from MIM software version 6.1.7 (MIM Software, Cleveland, OH) and subsequently imported into ImSimQA software (v4.2, Oncology Systems Limited, Shrewsbury, UK), to compute the following metrics: conformity index (CI), mean distance to conformity (MDC), volume difference (Vol.diff(%)), DICE similarity coefficient (DICE), sensitivity index (Se.Idx) and inclusion index (Incl.Idx). Results A total of 18 RT centres partecipated in the study. For each centre, SCCTV and the GSCTV of each corresponding representative patient were compared (Table 1). Overall, in P1 and P2 the CTV volume was significantly underestimated by the ROs (-29.25% and -27.83%, respectively), whereas in P3 it was only slightly overestimated (+12.53%). Nevertheless, the degree of superposition of contours, and therefore the quality of Segment AB AM SB SM PB PM A Alteration 29 19 23 8 7 4 18

contouring, was higher in P1 and P2 than in P3 - having P1 and P2 higher values of CI, DSC and Incl.Idx and lower values of MDC. The same analysis was carried out on each nodal level (Table 2). Considering the average on all patients, L4 showed the highest concordance with the corresponding GS both in terms of shape and volume, both in terms of position.

Table

1

Table

2

Conclusion Substantial differences were observed between the SCCTV and the GSCTV of the 3 representative patients, especially for the one with altered arm set-up (P3). The central levels (L2 and L3) were the most penalized. Overall, the results show a very high variability in lymph nodes contouring, which raises some issues in the era of highly conformal techniques. PO-0980 Preliminar evaluation of a new bra for large or pendular breasts irradiation. F. Castaño 1 , J. Gomez 1 , M. Árquez 1 , J. Acosta 1 , V. Hernandez 2 , A. Camarasa 2 , L. Torres 1 , P. Araguas 1 , M. Arenas 1 1 Hospital Universitari Sant Joan de Reus, Department of Radiation Oncology, Reus, Spain ; 2 Hospital Universitari Sant Joan de Reus, Physics Department, Reus, Spain Purpose or Objective Treating patients with large or pendulous breasts can be problematic. The aim of the present study is to investigate the possible dosimetric benefitof a new bra (Chabner XRT® Radiation Bra CIVCO) for large or pendulouswhole breast irradiation (WBI), as well as assess its impact on reproducibility and toxicity. Material and Methods Prospective study of patients with large or pendulous breast treated with conservative surgery and sentinel node biopsy candidates to WBI. Mammary ptosis classified according to Regnaul scale: GI: mild, GII: moderate and GIII: severe. Two CT studies were carried out in 14 patients, with one study usingthe breast bra. Radiation plans were produced with bra and without bra and the following parameters were compared: Dmeanlungdose (ipsilateral)and V16Gy, Dmeanheartdoseand V10Gy and Dmeanliverdose. Dailyverification of patient positioning was done using ConeBeam CT.Toxicity was evaluated weekly according to RTOG scale. Statistics: difference of means in the cohorts, paired Student's T-test, Pearson's correlation coefficient. Significant probability: p<0.05.