Single dose radiotherapy during surgery for breast cancer patients where external beam radiation was not feasible - results after 3 years of follow-up

ABSTRACTS leaving 71 with normal mammograms to be reviewed. The reason for recall, further investigations and diagnosis after assessment were noted. Results: The symptoms, prompting recall were, a breast lump (33/ 71,47%), nipple changes (25/71,35%), breast pain (12/71,17%) and skin dimpling (1,1%). Of these ladies 48(68%) were considered normal on clinical examination. Further imaging was performed in 12(17%) ladies, 1 MRI and 11 ultrasound scans. Ten patients had samples sent for histology, 4 core and 6 punch biopsies. After full assessment 70(99%) of patients had benign or normal results and were returned to the NHSBSP. Only 1(1%), with a nipple abnormality, had a diagnosis of cancer and underwent a mastectomy. Conclusion: These results show that when the screening mammogram is normal the detection rate for breast cancer based on symptoms is only 1%. Recall should be selective based on the symptoms mostly likely to represent breast cancer.

P129. Sentinel Lymph node biopsy in breast cancer: How many nodes should we remove? Roslyn Pursell, J. Gill, R. Lovegrove, K. Naessens, T. Cominos, C. McCormick Royal Berkshire Hospital, London Road, Reading, Berkshire, RG1 5AN Aims: Sentinel node biopsy (SNB) is now the standard of care in assessment of T1-2, N0 breast cancers. This study investigates whether there is a maximum number of sentinel lymph nodes (SLN) that need to be excised without compromising the false negative (FN) rate of this procedure. Methods: Data was prospectively collected for 319 patients undergoing SNB between February 2001 and December 2006 at our institution. This data was analysed, both in terms of the order of SLN retrieval and relative isotope counts of the SLNs, in order to determine the maximum number of SLNs that need to be retrieved without increasing the FN rate. Furthermore we investigated the relationship between SLN blue dye intensity and the presence of SLN metastases. Results: The SLN identification rate was 97.2% with no false negative cases. In patients with SLN metastases, excision of the first 4 SLNs encountered results in the identification of a metastatic SLN in all cases. Although the majority (85.6%) of SNB metastases are in the hottest node, the SLN containing the metastasis is in the first four hottest nodes in 98.8% of patients with nodal metastases. The remaining 1.2% of SLN metastases were identified by blue dye. There was no statistically significant association between the SLN blue dye concentration and the presence of SLN metastases. Conclusion: A policy to remove a maximum of 4 SLNs does not result in an increased false negative rate of detection of SLN metastases.

P130. Tumour suppressor function of CHIP & LOX in human breast cancer Neill Patania, W. Jiangb, K. Mokbelc a The London Breast Institute, The Princess Grace Hospital, 45 Nottingham Place, London, W1U 5NY b University Department of Surgery, Wales College of Medicine, Cardiff University c St. George’s University, London Breast Institute, Brunel Institute of Cancer Genetics Background: Ubiquitin modification influences diverse cellular processes including protein stability. CHIP (carboxyl terminus Hsc70-interacting protein) is a ubiquitin ligase implicated in the modulation of oestrogen receptor and Her-2/neu stability. Lysyl-oxidase (LOX) is a catalyst for cross-linking collagens and elastin within the extra-cellular matrix. Altered expression has been demonstrated in several malignancies and the peri-tumoural stroma. In this study, mRNA expression of CHIP and LOX were assessed in a cohort of breast cancers and correlated to clinico-pathological parameters over a 10 year follow-up period.

1239 Methods: Breast cancer (n ¼ 115) and normal background tissues (n ¼ 31) were collected after excision. Following RNA extraction, reverse transcription was carried out and transcript levels were determined using real-time quantitative PCR (normalized against CK19). Results: CHIP expression decreased with increasing Nottingham Prognostic Index (NPI): NPI-1 vs. NPI-3 (12.2 vs. 0.2, p ¼ 0.0264), NPI-2 vs. NPI-3 (3 vs. 0, p ¼ 0.0275), TNM stage: TNM-1 vs. TNM-2 (12 vs. 0, p ¼ 0.0639), TNM-1 vs. TNM-2,3,4 (12 vs. 0, p ¼ 0.0434) and tumour grade: grade 1 vs. grade 3 (17.7 vs. 0.3, p ¼ 0.0266), grade 2 vs. grade 3 (5 vs. 0, p ¼ 0.0454). The overall survival (OS) for tumours classified as ‘low level expression’ was better than ‘high level expression’ (118.1 vs. 152.3 months, p ¼ 0.039). LOX expression decreased with increasing NPI: NPI-1 vs. NPI-2 (3 vs. 0, p ¼ 0.0301) and TNM stage: TNM1 ¼ 3854639, TNM-2 ¼ 908900, TNM-3 ¼ 329, TNM-4 ¼ 1.2 (p ¼ NS). Conclusion: CHIP expression is significantly associated with favourable parameters, including tumour grade, TNM stage, NPI and OS. LOX expression is associated with an improved NPI. P131. Audit of Blue dye guided axillary surgery Sankaran ChandraSekharan, S. Marsh, I. Anwar Essex County Hospital, Lexden Road, Colchester, CO3 3NB Aim: The aim of the audit was to look at the accuracy of using blue dye alone for identifying Sentinel nodes and to look at false negative rates. Background: We are currently in the audit phase of the Almanac study. All patients having axillary surgery (who are not in the Almanac study) will have blue dye injected at time of operation. The MDT prior to operation will have decided whether patients are having a sample or full dissection. Results: We have looked at 527 patients (2003-2008) who have had only blue dye guided axillary surgery. In 507 (96.2%) the blue node was identified. Out of these 225 had positive nodes and 282 had negative nodes. Of the 282, 14 had false negative nodes (5.85%). In the fourteen patients 8 had heavy nodal disease of more than 5 nodes positive, 3 had 2 nodes positive and 3 had one positive node. In the 20 patients where no blue node was identified 13 had negative axillary nodes and 7 had positive nodes Discussion and conclusion: The combined method is considered as the gold standard for SLNB. In our unit we found that using blue dye alone our Sentinel node pick up rate (96.2%) and the false negative rate (5.85%) is very acceptable and compares well with the combined technique. In district general hospitals where getting gamma camera time may be a problem using blue dye alone may be a good alternative in the hands of experienced breast surgeons. P132. Single dose radiotherapy during surgery for breast cancer patients where external beam radiation was not feasible - results after 3 years of follow-up Mohammed Keshtgara, J. Vaidyab, C. Staceyc, J. Tobiasc, N. Williamsd, M. Baumd a Royal Free Hospital, Pond Street, London, NW3 2QG b Whittington Hospital, London c UCLH, London d Clinical Trials Group, UCL Medical School, London Introduction: Intra-operative radiotherapy (IORT) with IntrabeamÒ (Carl Zeiss, Germany) has been used since 2000 in the international randomised TARGIT Trial to determine if there is equivalence between the novel IORT technique and conventional external beam radiotherapy (EBRT) in women with early breast cancer. Some patients were unsuitable for inclusion in our trial for a number of reasons and were given IORT as a single treatment off-trial. Methods: Patients with invasive breast cancer underwent wide local excision followed by IORT (n ¼ 74). Low energy x-rays are emitted from the point source, delivering 20 Gy radiation dose to the breast tissue at the surface of the tumour bed. Radiotherapy is delivered directly into the area of interest, following which women can then proceed to have chemotherapy and/ or adjuvant hormonal therapy as required. In addition, patients who were

1240 deemed unfit for surgery (n ¼ 3) received interstitial radiotherapy alone under local anaesthetic. Women were followed-up for local recurrence. Results: Over the past 7 years in centres in 3 countries (UK, Germany and Australia), 77 patients have been treated in this way, with median age of 66 (56-77 IQR) years and a median follow-up of 37 (25-54 IQR) months. To date there have been two local recurrences, which gives an estimated annual local recurrence rate of 0.78% (95%CI 0.09% to 2.77%). Conclusion: This cohort adds to the evidence that targeted radiotherapy using IORT could offer a safe and effective method of delivering radiotherapy to breast cancer patients in whom EBRT is not feasible or is not an option.

P133. BORIS (Brother of Regulator of Imprinting Sites) as a possible new blood marker for breast cancer Sankaran ChandraSekharana, N. Porea, E. Klenovab a Essex County Hospital, Lexden Road, Colchester, CO3 3NB b Essex University Introduction: The CTCF- paralogous transcriptional factor, BORIS, is a member of the Cancer- Testis gene family specifically expressed in the testis, but abnormally expressed in various malignancies including breast cancer. The aim of the study was to investigate the expression of BORIS in the leukocyte fractions from Breast cancer patients and healthy donors. Methods: With Ethical Committee consent, whole blood samples were collected from breast cancer patients (n ¼ 87) and from healthy donors (n ¼ 52). The healthy donors were classified depending upon any history of infections, inflammations or recent surgery to check if conditions with increased leukocyte counts affected BORIS expression. The leukocyte fractions were isolated and immunostained with the anti-BORIS antibodies. The immuno-reactivity was scored depending on the percentage of positive cells and the intensity of staining. The IRS ranged from 0 (no staining) to 12 (maximum staining). The results were analysed statistically using Student’s t- test. Results: In the healthy donor group 19 % (n ¼ 10/52) of the samples showed very weak expression with a mean IRS of 0.25  0.009. There were no intra group variations. However 88.5% of leukocyte fractions from breast cancer patients were BORIS-positive with a mean IRS of 4.25  0.034. The IRS for BORIS in this cohort increased with tumour size. Conclusions: The high incidence of BORIS expression in breast cancer patients suggests that BORIS may have a role as an early marker for breast cancer. Key words: BORIS- Brother of Imprinting sites.

P134. Validation of Van Nuys classification in 183 patients with screen detected DCIS, a retrospective analysis Lona Jalinia, G. Mazdaib, I. Mokadema, H. Lucraftb, D. Hemmingc, D. Browella a Department of Surgery, Queen Elizabeth Hospital, Gateshead, Sherrif Hill, Gateshead, Tyne and Wear, NE9 6SX b Northern Centre for Cancer Treatment, Newcastle General Hospital c Department of Histopathology, Queen Elizabeth Hospital, Gateshead Introduction: The Van Nuys Prognostic Index (VNPI) is a numerical index designed to guide treatment following Breast Conserving Surgery for DCIS. It combines size, margin and grade and separates patients into low risk group suitable for treatment with excision only, moderate risk group suitable for excision with radiotherapy and high-risk group where mastectomy is recommended. The authors tested the validity of VNPI and the proposed benefit of adjuvant radiotherapy. Methods: All patients who had undergone surgery for DCIS between 1992 and 2004 were included. Data was collected using breast cancer database and case notes. The results were constructed according to VNPI and the outcome of each group evaluated. The risk factors for recurrence were analysed using c2 test.

ABSTRACTS Results: There were 183 patients. Forty-five patients were in the low risk group, fourteen in the high-risk group and one hundred and twenty four within the moderate risk group. The overall rate of recurrence was 7.1%. The moderate risk group benefited most from radiotherapy with risk reduction of 59% following radiation. The low risk subgroup showed no significant benefit from adjuvant radiotherapy. Seventy one percent of the patients in the high-risk VNPI group were treated with mastectomy with no recurrences over the 62 months median follow-up. Univariate analysis showed a significant association between risk of recurrence and high nuclear grade (p ¼ 0.018, c2 ¼ 5.57), comedo necrosis (p ¼ 0.032, c2 ¼ 4.63) and tumour size (P ¼ 0.0063, c2 ¼ 16.21). Conclusion: The VNPI can help the clinician and the patients evaluate their options allowing choice with an informed consent. P135. Use of sentinel lymph node biopsy halves the Lymphoedema rate overall Tahira Aslam, B. Chandrasekaran, R. Cochrane Wrexham Maelor Hospital, Croesnewydd Road, Wrexham, LL13 7TD Introduction: Lymphoedema following breast cancer treatment is a chronic, disabling condition. Following introduction of sentinel lymph node biopsy (SLNB) for axillary staging only the patients with positive axilla have completion clearance. How much has SLNB reduced the morbidity of lymphoedema in our DGH? Methods: All the patients with lymphoedema from Jan 2004 to July 2008 for a single surgeon were included. This continuous series was split into two groups, before and after introduction of SLNB. Prior to SLNB we used Axillary Node Sampling (ANS) for small tumors (20 mm and grade 1, 2) while in second half SLNB was used in all the patients as a staging tool (except patients with palpable lymph nodes or personal wishes). Arm circumferences were measured to determine the presence of lymphoedema. Results: A total of 120 patients were diagnosed with lymphoedema out of 467 cancers treated. The axillary clearance rate fell from 64.5 %(n ¼ 160) to 39 %(n ¼ 86). Lymphoedema decreased from 31 % (n ¼ 74) to 15 %(n ¼ 13) using two proportion Z test p < 0.001, 95% CI 0182-0.459. Pearson correlation analysis showed presence of positive lymph nodes correlates with the subsequent development of lymphoedema (r. ¼ 0.76, p < 0.001), but not the age of the patient, number of the axillary procedures and node count.Z test of two proportion showed that SLNB (p < 0.001) and ANS (p < 0.01) in comparison with ANC have a decreased risk of developing lymphoedema. Conclusion: Compared to ANC(axillary node clearance) and selected ANS, SLNB has halved our lymphoedema rate overall. P136. Use of multiple drains after mastectomy is associated with more discomfort and longer post-operative hospital stay Soni Soumian, A. Saratzis, R. Willetts, P. Stonelake Russells Hall Hospital, Dudley, West Midlands, DY1 2HQ Introduction: Seromas constitute a common complication following surgery for breast cancer and closed drainage is used routinely to reduce its incidence. The aim of this study was to evaluate the influence of number of drains on patient discomfort, seroma formation, and hospital stay after mastectomy for breast cancer. Methods: Based on a retrospective review of our clinical database, 110 consecutive patients from January 2004 - January 2006 who had undergone a mastectomy and axillary clearance for breast cancer were sent a simple postal questionnaire for data collection. Results: A total of 70 patients responded (all females; mean age: 69.4 years). 27 patients (38.57%) had 3 drains inserted unilaterally, 24 (34.28%) had 2 and 19 (27.14%) had 1 drain. They were divided into two groups; first group with 1 drain (19 patients) and the other with 2 or 3 drains (51 patients). Patients with one drain had a significantly shorter post-operative hospital stay [median: 2 days (range: 1-4 days) versus 2 days (1-8 days); Mann Whitney U test, p ¼ 0.02]. Fifteen patients (21.43%) complained of