Abstract Book

ESTRO 37

S412

Conclusion In patients with poor pathological responses the local recurrence rate was high. The success rate of local recurrence salvage was poor. Whenever TME was performed, either because of a need for completion or because of local recurrence, it was associated with a high APR rate. The extent of clinical tumour regression after radiotherapy might be considered as a guide for further management. PO-0796 Presence of High Risk HPV decreases odds of APR in patients with anal squamous cell cancer J. Jiang 1 , P. Wu 2 , M. Tam 2 , A. Lee 3 , K. Du 2 1 Montefiore Medical Center/Albert Einstein College of Medicine, Radiation Oncology, Bronx, USA 2 New York University Medical Center, Radiation Oncology, New York City, USA 3 SUNY Health Science Center at Brooklyn, Radiation Oncology, Brooklyn, USA Purpose or Objective Definitive chemoradiotherapy is the upfront treatment of choice in anal squamous cell carcinoma while surgery is reserved to refractory cases. Anterior peroneal resection (APR) of anal cancers often necessitates life-long a life- long ostomy, leading to lower quality of life for the patient. High risk human papillomavirus HPV subtypes have been shown to have improved outcomes in head and neck cancer. We analyze data from a large hospital based database to evaluate the impact of high risk HPV status on APR in patients with anal cancer. Material and Methods The National Cancer Database (NCDB) was queried for patients with anal SCC between the years of 2004-2014. Those with in situ disease or metastatic disease, unknown HPV subtype, and unknown treatment data were excluded. Logistic regression was used to examine the cross-sectional relationship between HPV status and the odds of undergoing abdominoperineal resection (APR). Overall survival (OS) and time to APR were estimated using the Kaplan-Meier and multivariate Cox regression From 2004 to 2014, 2,880 patients with anal cancer were identified to have known HPV subtypes. Of this, 1,710 (59.4%) were non-high risk and 1,170 (40.6%) were high risk HPV subtypes. 86 (2.4%) of 3,614 patients received APR. The median overall survival was 28.7 months (IQR 18.9 – 40.4). Univariable and multivariable logistic regression showed that individuals with high risk HPV had a 0.39 (95% CI: 0.21 – 0.74) odds of having APR compared to those who had non-high risk HPV (p = 0.004) after adjustment for T stage, N stage, age, sex, race, and Charlson-Deyo score. High risk HPV had a lower risk of APR (univariate: HR = 0.36, p < 0.001; multivariate: HR = 0.42, p = 0.007). Univariable and multivariable Cox regression showed no relationship between OS and high risk HPV (univariate: HR = 0.89, p = 0.266; multivariate: HR = 0.94, p = 0.592). The median time to APR for those with high risk HPV subtypes was 161 months and 126 months for those without high risk HPV. APR was not found to be associated with OS in univariate Cox regression analysis (HR = 1.52, p = 0.079). Conclusion This is the first reported analysis of the effect high risk HPV subtype on anal cancer OS and time to APR using the NCDB. While HPV subtype does not influence OS, those with high risk HPV subtypes have a lower risk of receiving APR. PO-0797 Diffusion-weighted MRI for early assessment of tumor response in rectal cancer patients. C. ROSA 1 , R. Cianci 2 , L. Caravatta 1 , M. Di Tommaso 1 , A. Delli Pizzi 2 , S. Di Biase 1 , F. Patani 1 , D. Genovesi 1 models. Results

Table 1. Median Visual Analogue Scale (VAS) and interquartile range stratified by baseline characteristics.

Conclusion This study suggests that there is a wide disparity in preference concerning organ-sparing approaches for rectal cancer in both patients with a history of rectal cancer and volunteers. Wait-and-see is often the highest preferred treatment, but is also among the least preferred treatment options. These findings may give new insights in how patients value current rectal cancer treatment options and can be used for cost- or comparative-effectiveness analyses. PO-0795 Preoperative radiotherapy and local excision of rectal cancer: Results of a prospective study K. Wisniowska- Polish Colorectal Study Group 1 1 The Maria Skodowska-Curie Memorial Cancer Center, Department of Radiotherapy, Warsaw, Poland Purpose or Objective Performing local excision after preoperative radiotherapy is of doubtful value for poor pathological responders. In order to guide optimal management after radiotherapy, appropriate clinical tumour response criteria to predict poor pathological responders are thus required. Material and Methods 89 patients with cT1-3aN0 <3–4 cm rectal adeno- carcinoma received 5 × 5 Gy plus a 4 Gy boost (71.9%) or 55.8 Gy in 31 fractions with a 5-FU bolus and leucovorin (28.1%). Local excision was performed 6–8 weeks later. Patients with good pathological responses (ypT0–1) were observed. Completion total mesorectal excision (TME) was recommended for poor responders (ypT1 with surgical margin+ and ypT2-3). Results Good pathological responses to radiation were observed in 63 (71%) patients. The remaining 26 (29%) had poor responses; 8 underwent completion TME and 18 either refused TME or were deemed unfit. Clinical complete responses were seen in 24% (group 1) and residual tumours in the remaining patients; tumour size being ≤1 cm in 30% (group 2) , >1 cm - ≤2 cm in 24% (group 3), and >2 cm in 19% (group 4). Good pathological responses were observed in 86%, 83%, 58% and 47% of patients from respective groups 1, 2, 3, and 4; p=0.02. Median follow- up was 8.9 years (IQR 7.4-11.3 years). Recurrence rate after local excision was 13% for good pathological responders and as high as 56% for poor responders. In total, the success rate of local recurrence salvage was only 44% (R0 resection and no subsequent local re- recurrence or distant metastases). Three out of five patients who were suitable at baseline for anterior resection and who underwent completion or salvage TME, were subjected to abdomino-perineal resection (APR).