Accuracy of sentinel lymph node biopsy in invasive lobular carcinoma of the breast.

Abstract

e12604 Background: The safety of the sentinel lymph node biopsy procedure (SLNB) in the surgical management of breast cancer relies upon a false negative rate (FNR) being less than 10%. The accuracy of SLNB in invasive lobular carcinoma (ILC), the second most common type of breast cancer, has not been evaluated. Because of high rates of false negative imaging and the diffuse growth pattern in ILC, less accurate pre-operative staging and a potentially unreliable lymphatic drainage pattern may impact the accuracy of SLNB in this tumor type. We therefore sought to characterize the accuracy of SLNB in a cohort of patients with ILC. Methods: We queried an institutional database of 707 patients with ILC and identified 196 patients who underwent SLN mapping with excision of both sentinel and non-sentinel nodes. A false negative was defined as having negative sentinel lymph nodes and a positive non-sentinel node. We calculated the FNR and sensitivity of SLNB and evaluated clinicopathologic variables. Results: Of 196 cases, 183 were clinically node-negative, 9 were clinically node-positive, and 4 had unknown clinical node status. Of the 183 clinically node-negative patients, 69 (37.7%) patients had node-positive disease at surgery. Overall, 7 of 196 cases had false negative SLNB, yielding an FNR of 8.97%. The sensitivity of SLNB was 91%. Patients with a false negative SLNB were significantly older than patients without (mean age 63 versus 54.7 years, p = 0.041). Significantly fewer sentinel and non-sentinel nodes were removed in women aged 50 years or older compared to those under 50 (1.9 vs. 2.5 sentinel nodes, p = 0.0158; 4.7 vs. 7.9 non-sentinel nodes, p = 0.0077). There were no differences in tumor receptor subtype, grade, stage, presence of lymphovascular invasion, or receipt of neoadjuvant therapy in those with a false negative SLNB compared to those without. Conclusions: The high rate of nodal positivity in clinically node negative patients highlights the challenges of clinical nodal assessment in ILC. Despite this, the SLNB procedure had a FNR that fell within the acceptable range, supporting its use in ILC. The relationship between number of sentinel nodes removed and FNR deserves further study, particularly in older women where extent of nodal surgery continues to decline.