ESTRO 2020 Abstract Book

S316 ESTRO 2020

Kingdom ; 7 Institute of Oncology Ljubljana, Department of Radiotherapy, Ljubljana, Slovenia ; 8 The Norwegian Radium Hospital-Oslo University Hospital, Department of Oncology, Oslo, Norway ; 9 Postgraduate Institute of Medical Education and Research, Department of Radiotherapy and Oncology, Chandigarh, India ; 10 Cross Cancer Institute and University of Alberta, Department of Oncology, Edmonton, Canada ; 11 St James's University Hospital, Leeds Cancer Centre, Leeds, United Kingdom ; 12 Radiotherapiegroep Arnhem, Department of Radiotherapy, Arnhem, The Netherlands ; 13 UZ Leuven, Department of Radiation Oncology, Leuven, Belgium ; 14 Leiden University Medical Center, Department of Radiation Oncology, Leiden, The Netherlands Purpose or Objective After definitive chemoradiotherapy (CRT) and image guided brachytherapy (IGBT) for locally advanced cervical cancer (LACC), nodal failure (NF) negatively impacts survival and salvage options are limited. Data are scarce on risk factors for NF, which could help identify patients at risk and might allow for optimising primary treatment strategies. Material and Methods Data for NF located in pelvic, inguinal en para-aortic (PAO) regions and separately PAO failure (NF PAO ) were collected from the international multicentre EMBRACE study cohort (www.embracestudy.dk). Analysis was done on data from a 2017 extraction, building on previously published work * . Missing data was handled using multiple imputation. Univariable and subsequently multivariable Cox-regression were performed using clinical variables potentially associated with NF: FIGO-stage (alternated with parameteria involvement), tumour width, uterine corpus involvement, Hb(including-nadir), overall treatment time (≥49 vs. <49 days), PAO-elective radiotherapy, local recurrence and nodal risk groups at diagnosis (pelvic without common iliac, any common iliac without PAO, any PAO nodes). Candidate multivariable models were compared with Akaike’s information criterion (AIC) to assess the model with the best fit. Subgroup analyses were performed for both NF and NF PAO in patients with the following sites of nodal involvement at diagnosis (excluding PAO-positive cases): 1. internal/external iliac, common iliac and parametrial sites, 2. any common iliac. * Nomden et al. Radiotherapy and Oncology (2019) 134: 185-190. Results 1338 patients with 152 NF and 104 NF PAO were available for analysis. For NF, the final model showed that at diagnosis larger tumour width, nodal risk groups and local recurrence after treatment were associated with a significantly higher risk of NF. With increasing Hb-nadir during treatment, however, the risk on NF was less (table 1). For NF PAO the same factors were found, with the addition that elective PAO irradiation independently lowered the risk of NF PAO (table 2). For subgroup 1 “any common iliac nodes without PAO” (HR=1.67, 95%-CI 1.00- 2.78; p=0.05) and local recurrence (HR=7.31, 95%-CI 4.20- 12.73; p<0.0001) remained significant for NF. For NF PAO virtually the same relation was seen, with the addition that PAO elective radiotherapy had a protective effect for NF PAO (HR 0.38, 95%-CI 0.17-0.86; p=0.02). For subgroup 2 only local recurrence was significant for NF.

Conclusion In cervix cancer, positive nodes (pelvic and PAO) at diagnosis, tumour width, and local recurrence after treatment increase the risk of NF and NF PAO , while a higher Hb-nadir decreases this probability. Additionally, we found elective PAO-radiotherapy decreased the risk of NF PAO and in particular in patients with positive pelvic nodes at diagnosis. More detailed investigations regarding nodal management and nodal control after treatment are part of the on-going EMBRACE II study. OC-0567 T2 and Apparent Diffusion Coefficient MRI for Cervical cancer delineation: agreement between volumes C. Rosa 1 , L. Gasparini 1 , F.C. Di Guglielmo 1 , A. Delli Pizzi 2 , B. Seccia 2 , R. Cianci 2 , R. Basilico 2 , A. Augurio 1 , L.A. Ursini 1 , A. Porreca 3 , M. Di Nicola 4 , L. Caravatta 1 , D. Genovesi 1 1 Ospedale Clinicizzato S.S. Annunziata, Radiotherapy Oncology- Chieti, Chieti, Italy ; 2 Ospedale Clinicizzato S.S. Annunziata, Department of Radiology, Chieti, Italy ; 3 G. D’Annunzio University, Department of Economy, Chieti, Italy ; 4 G. D’Annunzio University, Laboratory of Biostatistics- Department of Medical- Oral and Biotechnological Sciences, Chieti, Italy Purpose or Objective External beam radiotherapy (EBRT), concurrent chemotherapy and brachytherapy represent standard treatments for locally advanced cervical cancer. It is well kwon as T2-weighted MRI is the gold standard for tumor volume delineation in this setting. Functional Diffusion- Weighted MRI (DWI) images, in combination with T2-MRI, as for other neoplasms, could improve lesion detection. Our study tries to assess the difference in GTV delineation based on T2 weighted and Apparent Diffusion Coefficient maps (ADC-MRI), in order to evaluate volumes and inter- observer agreement between two different specialists, a radiologist and a radiation oncologist, with the aim to improve volumes delineation in cervical cancer patients. Material and Methods Thirty-seven cervical cancer patients (6 adenocarcinoma, 31 epidermoid carcinoma) were evaluated. All patients performed a diagnostic 1.5T MRI, with both axial T2 weighted images and ADC maps, calculated with the use of minimum 2 different b values. Two observers, one radiologist and one radiation oncologist, blinded and independently from each other, delineated GTV on T2 (T2 GTV ) and ADC (ADC GTV ) sequences. GTV volume (cm 3 ) was measured for T2 and ADC for each observer, and evaluated using the t-test. DICE index, Bland-Altman analysis (mean difference, 95% limits of agreement), coefficient of