Re-excision rates among older breast cancer patients undergoing breast-conserving surgery (BCS): Impact of the SSO-ASTRO consensus guideline on margins.

Abstract

576 Background: BCS has been historically associated with a high re-excision rate, driven in part by the need of obtaining negative margins. The SSO-ASTRO consensus guideline on invasive margins, defined a negative margin as no ink on tumor. In this large population-based study of older breast cancer patients undergoing BCS for invasive breast cancer we evaluate the guideline impact on re-excision rates. Methods: Female patients diagnosed with stage I-II breast cancer between 2012-2015 were identified in the SEER-Medicare database. Patients were >66 years and underwent BCS. Patients treated with neoadjuvant chemotherapy were excluded. We defined the following time periods: pre-guideline (January 2012-September 2013); peri-guideline (October 2013-March 2014) and post-guideline (April 2014-December 2016). Re-excision was defined as a resection, BCS or mastectomy identified using ICD-9 or CPT codes between 4 and 90 days after initial BCS. Overall re-excision rates and 95%CI were calculated and groups compared using X2test. Within subgroups we calculated re-excision rates for the pre and post-guideline periods and report the relative percent change. Regression model evaluated the association between time periods and re-excision while adjusting for important covariates, risk ratios (RRs) and 95%CI are presented. Results: 17001 patients were included. 6762 of them had BCS in the pre, 1786 in the peri, and 8453 in the post-guideline periods. Overall 22.6% of the patients had a re-excision. The rate decreased from 24.8% pre-guideline to 20.3% post-guideline (P < 0.001). The relative change in re-excision varied according to region (Midwest 3-7.3%, Northeast -20%, West -16.5% and South -13.5%) Differences in the relative change according to race/ethnicity were also seen (Whites -19.2, Blacks -15.3% and Hispanics -9.9%). In the multivariable model, BCS in the post-guideline period was associated with a decreased risk of re-excision (RR = 0.83; 95%CI 0.79-0.88). Lobular histology was associated with a higher risk of re-excision (RR = 1.31; 95%CI 1.21-1.42); greater surgeon volume was associated with lower risk of re-excision (RR = 0.89; 95%CI 0.82-0.95). Conclusions: There has been a statistically significant decrease in the re-excision rate after BCS associated with the dissemination of the SSO-ASTRO consensus guideline on invasive margins. Our study confirms the impact that guidelines have modifying patterns or practice, reducing the frequency of unnecessary interventions.