Abstract Book

ESTRO 37

S421

Purpose or Objective Evaluation of the efficacy and toxicity of primary chemoradiation with image-guided brachytherapy for cervical cancer. To identify possible predictors of treatment failure and toxicity. Material and Methods Retrospective analysis of consecutive patients with stage IB-IVA cervical cancer treated by primary chemoradiation and brachytherapy with curative intent from 2008 to 2016. Treatment consisted of external beam radiotherapy (EBRT: 46-48.6 Gy in 1.8-2 Gy fractions; 3D conformal or IMRT); with concurrent weekly cisplatin (40 mg/m 2 ); and image(MRI)-guided brachytherapy (3-4x 7 Gy HDR) using intracavitary (IC) or combined interstitial (IC+IS) techniques, according to GEC-ESTRO recommendations. In case of enlarged para-aortal (PAO) lymph nodes, EBRT included the PAO region, often with a boost to macroscopic nodes. Outcomes were local (LC), pelvic (PC), PAO nodal control; metastasis-free (MFS) and overall survival (OS) and adverse events (>3 months after therapy; according to CTCAE 3.0). Kaplan-Meier’s methodology was used to report incidences of all outcomes. To identify predictors of treatment failure and toxicity, Cox’s proportional hazards model and the Kruskal-Wallis test were used respectively. Results 155 patients were included for analysis. Median follow-up was 56.7 months (interquartile range (IQR) 27.8-79.3). Median age of patients was 56 (IQR 43-72). FIGO stage was IA-IIA in 34.1%, IIB in 48.4%, IIIA-IV in 13.6%. Pathologically enlarged lymph nodes were observed in 47.7%. Median cumulative EQD2 were: D90 HR-CTV 83.8 Gy (IQR 80.3-86.6), D2cc bladder 78.6 Gy (IQR 73.0-82.1), D2cc rectum 67.7 Gy (IQR 61.2-71.6), D2cc sigmoid 62.5 Gy (IQR 56.2-67.0) and D2cc bowel 55.1 Gy (49.9-62.1). At 3 and 5 years of follow-up LC was 91.3% and 91.3% respectively; PC was 86.3% and 84.7%; PAO nodal control was 87.9% and 86.7%; MFS was 75.4% and 72.2%, and OS was 74.8% and 65.9% . Tumor size was an independent predictor for LC (Table 1). Tumor size and N stage were predictors for PAO nodal control. Tumor size, FIGO stage and N stage were predictors for MFS. At 3 and 5 years of follow-up, the cumulative incidence of severe bladder, gastrointestinal and vaginal toxicity combined was 6.8% (Table 2). The only significant factor associated with an increased risk of toxicity was using IC technique alone (vs IC+IS) for vaginal mucositis (p=0.027). A trend towards significance was observed for: using IC technique alone (p=0.060) and vaginal tumor invasion (p=0.097) for vaginal stenosis and D2cc sigmoid >70 Gy (p=0.065) and comorbidity (p=0.091) for sigmoid toxicity.

Conclusion Chemoradiation with image-guided brachytherapy for locally advanced cervical carcinoma is a highly effective treatment with 91.3% LC at 5 years. Tumor size and lymph node metastasis were strong predictors of treatment failure. The incidence of severe adverse events was 6.8 % at 5 years. Toxicity may be further reduced by increasing conformity, using combined intracavitary-interstitial techniques and lowering the dose to organs at risk. PO-0809 Implementing a novel online education programme to support RTQA – the EMBRACE-II experience S. Duke 1 , N. Jensen 2 , T. Rumpold 3 , A. De Leeuw 4 , J. Lindegaard 2 , K. Tanderup 2 , R. Pötter 3 , R. Nout 5 , I. Jurgenliemk-Schultz 4 , L. Tan 1 1 Cambridge University Hospitals, Department of Oncology, Cambridge, United Kingdom 2 Aarhus University Hospital, Department of Oncology, Aarhus, Denmark 3 Medical University of Vienna, Department of Radiotherapy, Vienna, Austria 4 University Medical Center, Department of Radiation Oncology, Utrecht, The Netherlands 5 Leiden University Medical Center, Department of Radiation Oncology, Leiden, The Netherlands