Evolution of the Use of Completion Axillary Lymph Node Dissection in Patients with T1/2N0M0 Breast Cancer and Tumour-Involved Sentinel Lymph Nodes Undergoing Mastectomy: A Cohort Study

Background In breast cancer, completion axillary lymph node dissection (cALND) was previously recommended for patients with at least one tumour-affected sentinel lymph node (SLN). Several prospective trials predominantly in patients undergoing breast-conserving surgery showed no benefit and increased arm morbidity with this procedure. We report the influence of these trials on clinical practice of patients undergoing mastectomy. Methods We analysed prospectively collected data from patients with primary invasive breast cancer treated at German breast cancer units between January 2008 and December 2015. Time trends of cALND rates were analysed in patients undergoing mastectomy for T1/2N0M0 breast cancer with one or two tumour-involved SLNs. Multivariable logistic regression was used to determine factors influencing the decision not to perform cALND. Results Among the entire study cohort of 166,074 patients treated at 179 breast cancer units, 4093 patients (2%) had T1/2N0M0 breast cancer with one or two tumour-involved SLNs and underwent mastectomy. cALND rates decreased from 89.9% in 2010 to 55.5% in 2015 ( p < 0.001). Rates decreased from 82% to 8% in patients with micrometastatic SLN disease and from 93% to 63% in those with macrometastasis ( p < 0.001). In multivariable analysis, factors associated with omission of cALND were treatment at a general, nonacademic hospital, pT1 status, older age, higher number of removed SLNs, fewer tumour-affected SLNs, and SLN micrometastasis (all p < 0.001). Conclusions Despite limited evidence from prospective trials relating to the omission of cALND specifically in patients undergoing mastectomy, our nationwide data show that use of cALND decreased in these patients in routine clinical practice.