P1 Impact of SSO-ASTRO margin guidelines and intra-operative imaging on re-operation rates following breast conserving surgery: A single tertiary institution's six year experience

Abstract

Introduction Breast conserving surgery (BCS) offers favorable cosmesis but carries a re-excision (RE) risk. We assess the effectiveness of image-guided BCS considering updated guidelines regarding acceptable margins published by SSO-ASTRO in 2014 and 2016. We also compare the RE rate of radio-guided occult lesion localisation using iodine-125 seeds (ROLLIS) with hookwire-localisation (HWL), as well as assessing RE rate within our 2014 trial of intraoperative imaging (TRIDENT). Methods A retrospective single centre study was conducted with patients identified using the Sir Charles Gairdner Hospital (SCGH) Theatre Management System between 2012 and 2018. We included female patients aged 18 or over who underwent image-guided wide local excisions with subsequent RE for inadequate margins. Histology from core biopsy, initial excision and re-excision were recorded in conjunction with tumour characteristics. Results Of 710 eligible patients, 117 underwent RE. The RE rate for invasive cancer reduced from 19.44% to 11.73% (p <0.05); meanwhile there was no change in the RE rate for DCIS (23.75% vs 21.92%, p= 0.74) after the updated guidelines. 501 patients with invasive cancer and 143 patients with DCIS underwent HWL excision, while 57 patients with invasive cancer and 12 patients with DCIS had ROLLIS guided excision. The relative risk reduction for RE for HWL was 12.2% compared to ROLLIS guided excision. The RE rate reduced from 29.55% to 16.67% (p=0.18) with utilisation of TRIDENT. Conclusions We demonstrated a statistically significant improvement in RE rates for image localized invasive cancer after implementing the SSO-ASTRO guidelines. SCGH RE rates after BCS with radiological lesion localisation are lower than the rates described in the literature and we attribute this to increased experience in a tertiary multidisciplinary team setting. DCIS RE rate were not improved. TRIDENT use was associated with reduction in RE rates for DCIS but not enough to be statistically significant.