ESTRO 36 Abstract Book

S670 ESTRO 36 2017 _______________________________________________________________________________________________

stage 3 disease. This approach has shown to reduce both local recurrence rates and increase the rate of sphincter preservation procedures. Up to 20% of patients 6 weeks post neoadjuvant CRT have a complete histological response (pCR). PCR has shown to correlate with better and sustained oncological outcomes. The feasibility of the emerging watch and wait management strategy for patients with pCR will depend on the reliability of restaging assessments post CRT. We looked the accuracy of pre-operative MRI in predicting the rectal cancer tumour stage, node status and complete clinical response in patients who have undergone neoadjuvant chemoradiotherapy using histopathologic analysis as the reference standard. Material and Methods We retrospectively identified all patients who underwent neoadjuvant CRT (50.4 Gy, 1.8 Gy/fraction, in 5.5 weeks, with continuous infusional fluorouracil 225 mg/m2daily) for rectal cancer and proceeded to standard TME at our institution over a 16 month period. Their initial cTNM staging was collected as was their restaging ycTNM post CRT (based on diffusion weighted MRI pelvis). The sensitivity and specificity of the latter at predicting tumour, nodal and complete clinical response compared to surgical histology was analysed. Results 43 patients underwent CRT and subsequent TME over the time period at our institution. Overall histopathological response rate was 93% with a pCR rate of 14%. MRI had a sensitivity of 58% and specificity of 94% at assessing compete clinical response, 95 CI 40-93%, 80-99% respectively. At predicting tumour response MRI had sensitivity of 53% and specificity of 85%, 95 CI 45-80%, 74- 94% respectively. Accuracy of predicting nodal response were lower with a sensitivity of 43% and specificity of 40% , 95 CI 30-88%,32-58% respectively. The average modal time interval between CRT and MRI was 5 weeks while the average modal time between CRT and surgery was 8 weeks Conclusion Our study suggests that MRI alone may not be accurate enough in assessing clinical stage post neoadjuvant CRT, and particularly the clinical node status. Imaging alone will likely be needed to be combined with clinical, biochemical and endoscopic assessments in order to improve reliability of post treatment rectal staging. EP-1258 High precision SIB-IMRT versus conventional radiotherapy in anal cancer: a propensity score analysis F. Arcadipane 1 , A. Lepinoy 2 , P. Franco 1 , M. Ceccarelli 3 , B. De Bari 2 , L. Lestrade 2 , G. Furfaro 1 , M. Mistrangelo 4 , G. Créhange 5 , U. Ricardi 1 1 Radiation Oncology, Oncology, Turin, Italy 2 Radiation Oncology, Radiation Oncology, Besançon, France 3 Cancer Epidemiology and CPO Piemonte, Epidemiology, Turin, Italy 4 Surgery, Surgical Sciences, Turin, Italy 5 Radiation Oncology, Radiation Oncology, Dijon, France Purpose or Objective To evaluate clinical outcomes of a simultaneous integrated boost- intensity modulated radiotherapy (SIB- IMRT) approach in patients with non-metastatic anal cancer compared to those of a set of patients treated with 3-dimensional conformal radiation and sequential boost (CRT). Material and Methods A retrospective cohort of 190 anal cancer patients consecutively treated between March 2007 and October 2015 at 2 academic centres with concurrent chemo- radiation employing either SIB-IMRT or CRT was analysed. The SIB-IMRT group consisted of 87 patients, treated with 2 cycles of Mitomycin and 5-Fluorouracil using a SIB-IMRT based schedule of 42-45 Gy/28-30 fractions to the elective

pelvic lymph nodes and 50.4-54 Gy/28-30 fractions to the primary tumor and involved nodes, based on pre- treatment staging. The CRT group comprised 103 patients, treated with Mitomycin or Cisplatin and 5-Fluorouracil or Capecitabine concurrent to CRT with 36 Gy/20 fractions to a single volume including gross tumor, clinical nodes and elective nodal volumes, and a sequential boost to primary tumor and involved nodes of 23.4 Gy/13 fractions. We determined colostomy-free survival (CFS) and overall survival (OS), loco-regional recurrence and distant metastases rates for each radiation modality. Cox proportional-hazards model addressed factors influencing OS and CFS. Propensity score-matched analyses were performed to compare SIB-IMRT and CRT. Results Median follow-up for the entire patient group was 32 months. Average overall treatment time was 42 days in the SIB-IMRT group and 59 days in the CRT group. Patients treated with CRT had significantly higher stage and lower grading. The overall survival at the time of analysis was 74%, similarly for the two groups. Three-year colostomy- free survival was 66% for all patients, with no significant difference between the two groups (61% for SIB-IMRT and 74% for CRT, Log-Rank 0.85). The cumulative incidence of colostomies showed that the majority of events occurred within 18 months in both groups. We found no significant difference in terms of outcomes by univariate analysis and a propensity score analysis adjusted for disparities between the groups.

NA:Not

Available

Tab. 1 Patient and treatment characteristics and pattern of failure