ESTRO 37 Abstract book

S644

ESTRO 37

2 Tata Medical Centre, Department of Radiodiagnosis, Kolkata, India Purpose or Objective Carotid artery damage post radiotherapy can be prevented using dosimetric sparing of the carotids using helical tomotherapy techniques. CARSREL study assessed the feasibility of implementing such a technique and investigates if dosimetric benefits translate into clinically documented benefits of reduction of wall thickness of the carotid arteries. Material and Methods In this non-randomised prospective study, 18 patients with T1-2N0M0 glottic cancer in the study arm and another 18 patients receiving standard radiotherapy to the neck in the control arm, were recruited. Patients in the study arm received 55 Gy in 20 fractions over four weeks to the entire larynx. Bilateral carotid arteries were delineated and dose was reduced to as less as possible without compromising the dose to the planning target volume (PTV). Carotid dopplers were done in all patients to measure the carotid intimal thickness (CIT) at 7 pre- specified levels of the carotid artery at pre-treatment and at 6, 18 and 36 months’ post-treatment. Primary endpoints were feasibility of CSR and the difference in percent change of mean CIT between the arms. Secondary endpoints were to identify any dose- relationship with CIT and if only a 6 th month carotid doppler would predict long-term changes in CIT. This study was registered with the Clinical Trials Registry- India (CTRI/2012/12/003243). Results Patients were followed up to 36 months following treatment. Median age in both the arms were 60.5 years. In the CSR arm, mean dose to the right and left carotids were 12.98 and 12.39 Gy respectively. Laryngectomy free survival in the CSR arm was 72% at 3 years. Percent change in CIT at 6 months was negligible in both the arms (0.09% vs. 0.10%). They did not vary significantly even after 36 months of follow-up. One patient in the control arm developed a sclerotic plaque following standard radiation therapy. Maximum/mean/median carotid doses had no association with the change in CIT. Conclusion Carotid sparing in early glottic cancers is feasible with helical tomotherapy. Carotid wall thickness did not vary significantly with standard irradiation techniques and we found no clinical significance for the routine use of CSR. EP-1147 Prognostic value of neutrophil–lymphocyte ratio in head-neck cancer: Mono-institutional experience F. Martucci 1 , A. Richetti 2 , S. Cima 2 , K. Yordanov 2 , C. Azinwi 2 , M. Valli 2 , G. Pesce 2 , E. Cekani 2 , D. Bosetti 2 , I. Donegani 2 , I. Maddalena 2 1 Martucci Francesco, Radiation Oncology- Oncology Institute of Southern Switzerland Bellinzona-Lugano- Switzerland, Bellinzona, Switzerland 2 Oncology Institute of Southern Switzerland Bellinzona- Lugano- Switzerland, Radiation Oncology, Bellinzona, Switzerland Purpose or Objective The aim of this study was to investigate the relationship between pre and post-treatment inflammatory markers and the prognosis of patients with locally advanced oral cavity, hypopharyngeal and laryngeal carcinoma treated with concomitant chemoradiotherapy (CRT). Material and Methods Patients with primary squamous cell carcinoma treated with curative intent were analysed retrospectively. Chemoradiotherapy consisted of external beam radiotherapy with VMAT/RapidArc technique to a dose of 66-70 Gy with concurrent chemotherapy platinum based. Pre and post-treatment neutrophil-to-lymphocyte ratio

(NLR) were registered as inflammatory markers from blood samples obtained within 14 days before and after CRT. The primary endpoint was the 3-year overall survival (OS) and secondary endpoint was the 3-year disease-free survival (DFS). Actuarial Kaplan–Meier statistics and log rank test were used. Results Fifty-seven patients were considered with a median follow up of 37 months. Median pre-treatment NLR was 2.9 and median post-treatment NLR was 5.8. The 3-year estimated OS for pre-treatment NLR <2.9 vs >2.9 was 66% vs 67 (p = 0.541), the DFS 74% vs 69% (p = 0.743). The 3- year estimated OS for post-treatment NLR <5.8 vs >5.8 was 67% vs. 66% (p = 0.923), the DFS 69% vs. 75% (p = 0.663). Conclusion In this cohort of patients with locally advanced oral cavity, hypopharyngeal and laryngeal carcinoma treated with concomitant chemoradiotherapy, pre and post- treatment NLR was not a significant prognostic factor for both overall and disease-free survival. Our experience does not confirm the results of other series reported in the literature. Additional data could be necessary to investigate the prognostic value of inflammatory markers in head and neck cancers. EP-1148 Evaluation of atlas-based auto segmentation (ABAS) of breast cancer and head and neck cancer cases B. Tamaskovics 1 , J. Haussmann 1 , J. Brünger 1 , D. Hartmann 1 , E. Boelke 1 , P. Gerber 2 , F. Djiepmo Njanang 1 , P. Kröpil 3 , W. Budach 1 , C. Matuschek 1 1 University Hospital Düsseldorf Heinrich Heine University Düsseldorf, Radiation Oncology, Düsseldorf, Germany 2 University Hospital Düsseldorf Heinrich Heine University Düsseldorf, Dermatology, Düsseldorf, Germany 3 University Hospital Düsseldorf Heinrich Heine University Düsseldorf, Department of Diagnostic and Interventional Radiology, Düsseldorf, Germany Purpose or Objective Contouring target structures and organs at risk is a time intensive part of the clinical practice in radiotherapy. There are several solutions for automation this procedure. The aim of this study was to evaluate the feasibility of Atlas-based Auto segmentation (ABAS ® 2.0; Elekta, Stockholm, Sweden) for contouring of clinical target volumes and organs at risk Material and Methods The study was performed on two cohorts. One cohort consisted of 70 cases of postoperative whole breast irradiation for breast cancer. The second cohort was 70 patients with head and neck cancer. Clinical target volumes (CTV) and the organs at risk were contoured with ABAS and manual according to actual intern clinical standards. The included atlases were used; no further reference atlases sets were uploaded. The volume of organs at risk and lymph node levels in the head and neck cases, as well as the clinical target volume CTV in the breast cases was compared with the manual contouring performed by an experienced radiation oncologist. The Dice similarity coefficient (DSC) was used to quantify the amount of similarity. Craniocaudal deviations of the individual structures were analysed. Time of auto segmentation and subsequent manual correction of all structures was compared with manual contouring time of an experienced physician and correlated with recent published papers. Results DSCs were between 0.2261 (lymph node level V right) and 0.99 (outer contour). Mean craniocaudal deviations for CTV breast, heart, lungs, lymphatic levels II, III and IV were 7.29 mm, 5.58 mm, 1.59 mm, 3.27 mm, 9.55 mm, and 8.35 mm, respectively. Practically all of the segmented volumes needed a revision. Mean time for