Abstract Book

S813

ESTRO 37

twice daily, during the RT days. After a mean interval after completion of CRT of 54 days (range, 13-142), all patients underwent surgery (n = 15, abdominoperineal resection; n = 62, anterior resection). All but one patient presented a R0 resection (R1 in the remaining patient). Adjuvant chemotherapy was delivered in 43/77 patients. The tumor regression grade (TRG) was assessed using the Mandard score. Results Median follow-up period was 53.3 months (range, 3.9 – 89.7). Nineteen patients presented clinical stage II disease, while 7, 42, and 9 patients presented stage IIIA, IIIB, and IIIC disease, respectively. The overall rate of G2 or more toxicity was 22% (17/77 patients). Only 2 patients (2.6%) presented a G3 toxicity (dermatitis in 1, diarrhoea in 1). None of the patients presented a G3 (or more) hematologic toxicity and/or G4 non-hematologic toxicity. We observed only 2 patients (2.6%) who presented a G3-4 late toxicity. Two patients presented a late GI toxicity (2.6%). A total of 15% and 32% of patients presented a complete (TRG = 1) or an almost complete response (TRG = 2), respectively. Forty-seven per cent of the patients presented a TRG ≤ 2 (complete or almost complete response). The actuarial 5-year overall survival, local control, disease-free survival rates were 84% (95% confidence interval [CI]: 75 - 93%), 78% (95% CI: 68 - 88%), and 66% (95% CI: 54 - 77%), respectively. Conclusion SIB delivered with HT and daily IGRT is safe, and allows high rates of local control and survival. EP-1499 PET imaging for target volume delineation in rectal cancer radiotherapy: results of a phase II study B. Floreno 1 , C. Rinaldi 1 , M. Fiore 1 , P. Trecca 1 , C. Greco 1 , A. Iurato 1 , R.M. D'Angelillo 1 , L. Trodella 1 , S. Ramella 1 1 Campus Biomedico University, Department of Radiation Oncology, Roma, Italy Purpose or Objective The aim of this study is to evaluate if the integration of imaging techniques, such as CT, MRI and PET/CT, can allow the reduction of treatment volumes and avoid the elective nodal irradiation reducing the toxicity without compromising the effectiveness of the treatment. Material and Methods Eligible patients had histologically proven locally advanced rectal adenocarcinoma (stage T3 N0-N1, M0) and ECOG performance status 0-1. Staging procedures included CT, MRI and 18F-FDG PET scan. All patients received 3D- conformal radiotherapy, and the CTV included the rectal lesion and the corresponding mesorectum without elective nodal irradiation. The total dose delivered was 50.4 Gy in 1.8 Gy/fx . Toxicity was recorded and scored according to the CTCAE vers 3.0 scale. Results Between December 2007 and October 2016, 52 patients (median age 67 years, range 45-85 y) affected by rectal cancer with a II-III stage were enrolled in this trial. The concomitant CT was platinum-based with 5FU/Capecitabine in 32 patients (61,5%) and 5FU/capecitabine in 20 patients (38,5%). The patients showed an excellent compliance to the protocol. Toxicity was evaluated in 50 patients. G1-G2 hemathological toxicity was NEU/PLT/HB in 21/14/15 patients (42-28- 30%) and G3 anemia in only 2 patients (4%). G3 rectal toxicity/diarrhea occurred in 1 and 3 patients (2%- 6%). Pathological complete response (pT0N0) was in 44% patients with TRG 1 in 48.9% patients. 3y-OS was 89%, 5y-

OS 64% with a median survival of 7.5 years and median DFS was 19.6 months. Conclusion PET/CT can be an useful tool for target delineation in rectal cancer. This study shows that the reduction of treatment volume, without elective nodal irradiation can lead to a lower toxicity with promising clinical results. EP-1500 Prediction of anal carcinoma chemoradiotherapy outcome by a new PET-based biomarker E. Rusten 1 , B.L. Rekstad 2 , C. Undseth 3 , E. Hernes 4 , M.G. Guren 3 , E. Malinen 5 1 Rusten Espen, Dept of Medical Physics, Oslo, Norway 2 Oslo University Hospital, Department of Medical Physics, Oslo, Norway 3 Oslo University Hospital, Department of Oncology, Oslo, Norway 4 Oslo University Hospital, Department of Radiology and Nuclear Medicine, Oslo, Norway 5 University of Oslo, Department of Physics, Oslo, Norway Purpose or Objective To evaluate a new [18F]fluorodeoxyglucose positron emission tomography (FDG-PET) biomarker for predicting treatment outcome of squamous cell carcinomas of the anal canal treated with definitive chemoradiotherapy. The impact of the new metric on locoregional control is assessed by comparing it to other PET-based metrics and clinicopathological prognostic factors, such as TNM stage and human papilloma virus (HPV) status. Material and Methods Ninety-three patients with anal squamous cell carcinoma referred to chemoradiotherapy (54-58 Gy, concomitant Mitomycin c/ 5-fluorouracil) were prospectively included. Prior to treatment, FDG-PET/CT imaging was performed on a Siemens Biograph mCT and used to determine M stage (distant metastases) and N stage (lymphatic spread: N3 bilateral/perirectal + iliaca interna/inguinal). A new PET metric (ZMP) was formed by an equal Z-normalized weighing of the metabolic tumor volume (MTV) and peak standardized uptake value (SUVpeak). The significance of the classifier was evaluated in univariate Cox regression analysis and the dependency on other clinical factors in bivariate analysis. Results With a median follow up time of 25 months there were 11 (12 %) in-field local or locoregional recurrences. Lymph node stage (N3), HPV status, MTV, total lesion glycolysis (TLG) and ZMP were in univariate analysis highly significant parameters for predicting lack of locoregional control (p<0.01, HR range 5.1-7.5) while SUVmax/peak/mean was not significant (p>0.2). In bivariate analysis HPV status was found to be the most independent predictor. The best bivariate combinations with HPV included the categorical variable N3 (p<0.01) and the non-categorical variable ZMP (p=0.01). Conclusion The clinical factors HPV status and N3 stage were the most predictive of chemoradiotherapy treatment outcome for anal cancer, though their dichotomous nature limited their precision. The FDG-PET parameters MTV, TLG and ZMP were all predictive of outcome to a similar degree, but the proposed ZMP gave a slightly better prediction model together with HPV.