ESTRO 2020 Abstract book

S1102 ESTRO 2020

whether the mean heart dose (MHD) reflects the doses to the other cardiac substructures. Material and Methods Fifty women with left-sided breast cancer were prospectively evaluated. Patients received 3D-conformal RT (40 Gy delivered in 15 fractions of 2.67 Gy +/- boost of 13.35 Gy). Whole heart, cardiac chambers and coronary arteries (left main (LM), left anterior descending (LAD), circumflex coronary artery (CX), right coronary (RCA)) were contoured on radiotherapy CT-planning scans according to the cardiac contouring atlas of F.Duane. Dose distribution were generated for all cardiac substructures. Dosimetric associations were calculated. Results All patients met the dose constraint of the heart (mean MHD=3.08 Gy ; mean V20 heart=3.8%). The mean D mean and D max to the LAD were 11.45 and 29.5 Gy, respectively. The D max to the LAD ranged from 2.33 to 39.8 Gy. Low doses were delivered to LM, CX, and RCA ( D mean ≤1.3 Gy; D max ≤1.6 Gy). Considering the other cardiac chambers, left ventricle (LV) was the most exposed structure with a D mean of 4.78 (1.35-7.8) Gy and a D max of 37 (7.6-45.4) Gy. Internal mammary node irradiation resulted in significantly higher doses to the LAD and LV (p=0.002; p=0.003). Pearson correlation coefficients between MHD and D mean for cardiac substructures were all statistically significant. The strongest correlation with MHD was found for D mean to the LAD (r=0.81). For every 1 Gy increase in MHD, D mean LAD rose by 3.4 Gy. Howerver, the proportion of the variance in D mean LAD predictable from MHD was moderate (R² = 0.65). For all other cardiac substructure, the predictive value of MHD was moderate to weak with no R² value above 0.7. Conclusion Our dosimetry study illustrates that the LAD and LV are the most exposed cardiac substructures in left-sided breast cancer radiotherapy. With poor predictive value, MHD was not precise enough to reflect with confidence doses to cardiac substructures, including LAD. For optimal heart sparing radiotherapy, it would be necessary to delineate LV and LAD and to assess doses within these substructures as well as to the whole heart. PO‐1881 Dosimetric impact of left anterior descending artery contouring in left‐sided breast irradiation Z. Naimi 1 , R. Moujahed 1 , M. Ben Rejeb 1 , S. Ghorbel 1 , J. Yahyaoui 1 , A. Hamdoun 1 , L. Kochbati 1 1 Abderrahmen Mami Hospital, Radiation Oncology, Ariana, Tunisia Purpose or Objective The aim of our study was to assess the impact of contouring the left anterior descending artery on treatment planning in left-sided breast cancer radiotherapy. Material and Methods The data of fifty women treated for left-sided breast cancer were prospectively evaluated. Patients were planned for 3D conformal RT with or without lymph node irradiation. The prescription dose was 40 Gy delivered in 2.67 daily fractions (+/- additional boost of 13.35 Gy). Delineation of the heart substructures and coronary arteries was performed using the cardiac contouring atlas of F.Duane. For each patient, two different treatment plans were generated by two separate dosimetrists. The first plan considered only the heart as organ at risk while the second plan considered the heart as well as the LAD. The following dose metrics were chosen for plan comparisons: The D mean and D max to the heart, LAD and left ventricle, the volume of the heart that received more than 20 Gy (V20 heart ) and the volume of the CTV that received more than 95% of the prescription dose (V95%). Student’s t -test was used to compare these parameters. Results The mean heart dose was significantly lower for the second plan (2.34 vs 3.08 Gy; p<0.001) as well as the D mean (7.1 vs

11.45 Gy ; p<0.001) and D max (18.9 vs 29.5 Gy; p<0.001) to the LAD. Despite no specific dose constraint, optimization on the LAD showed significant reduction in V20 heart (2 vs 3.83%; p<0.001), D mean (3.43 vs 4.78 Gy; p<0.001) and D max (34.5 vs 37 Gy; p<0.001) to the left ventricle. The coverage of the CTV was not substantially affected as there was no significant difference in the V95% between the two plans (96.46 vs 95.06%; p=0.06). Conclusion Despite no specific dose constraint, delineation of the LAD resulted in considerable cardiac dose-sparing with sufficient coverage of the clinical target volume. Since left-sided breast irradiation has been associated with an increased cardiac mortality, the LAD should be routinely contoured and considered as a separate organ at risk. PO‐1882 The risk of pelvic insufficiency fractures after radiotherapy using planned dose analysis. L. Devlin 1 , S. Millar 1 , S. Morris 1 , A. Sadozye 1 , R. Harrand 1 , K. Graham 1 , A. Kerr 1 , C. Duncanson 1 , A. Duffton 1 1 Beatson West of Scotland Cancer Centre, Radiotherapy, Glasgow, United Kingdom Purpose or Objective Pelvic insufficiency fractures (PIF) are fractures caused by normal or physiological stress on weakened bone. Radiotherapy (RT) increases the risk of insufficiency fractures in patients as a late effect. 1 Symptoms include debilitating pain, with limited management options. The dose-effect relationship has previously been defined with a risk of fracture at 22% at 35Gy and 45% at 40Gy at sacrum D50%. 2 Aim To determine in patients with PIF, if the sacrum D50% at radiotherapy planning exceeded the dose levels defined by Patients with confirmed radiological PIF were included in analysis. Radiotherapy technique was volumetric modulated arc therapy (VMAT), patients were treated with a library of 2 adaptive plans or with 1 plan. Retrospective dosimetric analysis was done on the planning CT, by delineation of the sacrum in Eclipse V15.5 (Varian Medical Systems, Palo Alto) . A structure was created using the boolean operator to define sacrum overlap with PTV. Dose volume histograms were exported for the sacrum structure and overlap structure and planning characteristics recorded. Results All Ca Cervix patients had 4 Brachytherapy insertions (24Gy) and Ca Endometrium patient had no insertions. 6 patients presenting symptom was lower back pain and 4 patients PIF were radiological incidental findings. All patients had concurrent weekly cisplatin. Dose prescribed 50Gy/25# (n=6) and 45Gy/25# (n=4). 10 gynaecological patients treated with external RT between March 2013-June 2019,had confirmed sacral insufficiency fractures on diagnostic magnetic resonance imaging (MRI),RT completed between March 2013- December 2018. Median age of patients was 57 (IQR 37.5- 64.75) with 7 (70%) patients > 50 years. 2 patients were pre menopausal on hormone replacement therapy (HRT) remaining patients not on HRT. Time to confirmed diagnosis of insufficiency fractures from radiotherapy completion was mean of 12 (IQR 5-20.75) months. The median sacrum volume was 189.1cm 3 (IQR 194.23-157.26). The median sacrum overlap with PTV volume was 42.66cm 3 (IQR 39.26-46.42). Figure 1 shows individual patients sacrum and sacrum-PTV volumes The max dose to the sacrum was 54.05Gy, min 16.22Gy and mean 34.16Gy (IQR 36.13-39.74). The median D50% of the sacrum was 39.03Gy (IQR 37.04-40.61). 9/10 (90%) of patients D50% were above 35Gy and 4/10 (40%) were above 40Gy . For the patients who were symptomatic the mean sacrum dose was 39.05Gy and the median D50% 40.52 Gy. Ramlov et al (2017). Material and Methods