S S37 Background: Neoadjuvant chemo radiation therapy (CRT) is the standardmanagement strategy for patientswith rectal cancer. The conventionally accepted interval between CRT& surgery is six to eight weeks. A number of retrospective studies have postulated that a delay in surgery beyond six weeks may result in further tumour downgrading and may improve surgical outcome. However there remains a concern that longer interval may result in tumour progression compromising the oncological safety. It may also result in greater fibrosis and make the surgery difficult. Since the effect of delayed interval has never been evaluated adequatelya randomized control trial was designed to evaluate the effect of delay in surgery after CRT. Material and methods: A total of 134 cases (67 in each arm) of locally advanced carcinoma rectum will be recruited from 1/August/ 2012 to 31/July/2015(Three years). Eligible patients after receiving CRT undergo response assessment and are randomized into two groups, control group (short arm, SA) & the test group (long arm, LA). Patients in SA undergo surgery at 6-8 weeks while patients in LA undergo surgery at 10-12 weeks post CRT. Total duration of the study will becompletion of follow up of three years from the date of recruitment of the last patient. The primary end point is to assess the long term impact of delaying surgery after CRT in locally advanced rectal cancers in terms of Disease free survival & overall survival. A planned interim analysis was performed after recruitment of the first sixty cases to evaluate pathological response & perioperative morbidity and mortality. Results: A total of sixty patients (31-long arm, 29-short arm) were recruited out of which 57 underwent final analysis (29-LA, 28-SA). The mean time interval between CRT and surgery was 51.93 days in the SA and 74.38 days in the LA.No significant differencewas observed in the duration of surgery (SA-180min, LA-194min), perioperative wound infections (SA-17.24%, LA-20.68%) or urinary complications. Three patients in each arm developed postoperative anastomotic leak for which a diversion was required. Pathological analysis does not reveal any significant difference in terms of mean tumour regression grading (LA-2.89, SA-2.88), Pathological complete response and circumferential resection margin positivity. Conclusion: The interim analysis reveals that Delay in surgery post CRT is not associated with increased perioperative morbidity & mortality. However no further down staging of tumor was observed after prolonging the interval for surgery. The effect of increased time interval in terms of long term loco regional control and distant metastasis remains to be evaluated and will be available at the completion of study No conflict of interest. http://dx.doi.org/10.1016/j.ejso.2014.08.070 74. GIST may have a higher incidence of second primary cancers which negatively impact upon survival M. Smith, A.L. Mahar, C.H. Law, Y.J. Ko 1 The Royal Marsden NHS Foundation Trust, Academic Surgery, London, United Kingdom Queen’s University, Public Health Sciences, Kingston, Canada Odette Cancer Centre Sunnybrook Health Sciences Centre, Surgical Oncology, Toronto, Canada Odette Cancer Centre Sunnybrook Health Sciences Centre, Medical Oncology, Toronto, Canada Background: A higher incidence of second primary cancer (SPC) has been reported in patients diagnosed with gastrointestinal stromal tumors (GIST). We aimed to identify patients with GIST who develop a SPC, quantify the risk of additional malignancy and evaluate the impact upon survival. Material and methods: Individuals diagnosed with GIST from 20012009 were identified from the SEER database. Standardized incidence ratios (SIR) were calculated using SEER*Stat software (V.7.1.0). Cox-proportional hazards and logistic regression identified predictors of survival and SCP. Results: 1705 cases of GISTwere identified, with 181 (10.6%) patients developing SPC overall. The risk of SCP was significantly higher in our cohort than expected (SIR 1.36; 1.08-1.7 95% CI). Older age (p<0.0001) and extra-oesophagogastric GIST (p1⁄40.0027) were significant predictors of SPC. Colorectal cancer was the commonest synchronous cancer (30%). Median time to diagnosis of metachronous SPC was 21.9 months, the commonest sites being urinary system in men (SIR 2.39; 1.03-4.72 95% CI) and colon (SIR 2.96 1.19-6.09; 95% CI) in women. Overall 5-year survival was 65%. A synchronous SCP associated with reduced overall survival (HR 1.55 1.05-2.3 95% CI, p1⁄40.04). Conclusion: Patients with GIST have a higher incidence of SPC when compared with to the general population. Older age and primary disease site are predictors of SPC. 10.6% of our cohort developed a SPC, with urinary tract being the commonest site overall. Colon cancer was the commonest synchronous cancer, and women were at an increased risk of metachronous colon cancer. Synchronous SCP was associated with a poorer overall survival. At diagnosis and during surveillance of GIST, screening for colon and urinary tract cancers may be considered. No conflict of interest. http://dx.doi.org/10.1016/j.ejso.2014.08.071 75. Endoscopic sentinel node biopsy and axillary dissection for early breast cancer can be navigated by the fusion image of 3D-CT lymphography and SPECT-CT K. Yamashita, K. Yanagihar, H. Takei 1 Nippon Medical School, Dept. of Breast Surgery, Tokyo, Japan Background: 3D-CT mammary lymphography (LG) can image the detail lymphatic map from the breast gland to the axillary nodes (AN) and can enable us to detect the precise sentinel node (SN) and to perform the endoscopic SN biopsy easily. It can also detect the second and the third SN on the lymphatic map. These nodes biopsy will contribute to omit AN dissection on the SN-positive patients. Then we try to overlay the fusion image of SPECT on the endoscopic view, and evaluate its real-time navigation of endoscopic surgery. Methods: 3D-CT LG was performed by the subcutaneous injection of 2 ml Iopamidol 300 above the tumor and near areola, and by taking CT images at 1 minute. They were reconstructed to produce a 3D image of lymph ducts and lymph nodes by thevolume renderingmethod. SPECTwas performed by injection of 99mTc phytate 74mBq and taking images after 2 hours. SN biopsy and AN dissection were performed by dye and RI method using endoscope with the optical trocar Visiport through only 1 cm long skin incision, and overlaying 3D-CT image on the endoscopic view with the SPECT. Results:We have performed the endoscopic SN biopsy on 350 patients. The SNmetastasis was found on 80 patients: single SNmetastasis on 32, the second SN on 8, and the third SN on 7. The fusion image of 3D-CT LG and SPECTenable us to identify the anatomical position of SN onANmap by RI detector probe, and to introduce the endoscope by the overlay system. It also enable us to get the second SN easily. The axillary node metastasis was not observed on the patients without second SN metastasis. There was no false negative study. There is no local recurrence at 10 years after surgery. Conclusions: The overlay of 3D-CT LG on the endoscopic view improves the identification rate of SN and the manipulation of the endoscopic surgery. We are developing this visual processing technique to apply the real-time navigation for the endoscopic partial mastectomy. No conflict of interest. http://dx.doi.org/10.1016/j.ejso.2014.08.072 76. Optimal treatment of the axilla after positive sentinel lymph node biopsy in primary invasive breast cancer patients (surgery versus radiotherapy)eOTOASOR trial: 5 years follow-up of a randomized clinical trial A. Savolt, Z. Matrai, C.S. Polgar, N. Udvarhelyi, G. Rubovszky, E. Kovacs, P. Musonda, G. Peley 1 National Institute of Oncology, Department of Breast and Sarcoma Surgery, Budapest, Hungary