Predicting the extent of nodal disease in early-stage breast cancer.

Background: The role of regional nodal ultrasound (US) has been questioned since publication of the American College of Surgeons Oncology Group (ACOSOG) Z0011 data. The goal of this study was to determine if imaging and clinicopathologic features could predict the extent of axillary nodal involvement in breast cancer.

Study Design: Patients with T1-T2 tumors who underwent regional nodal US and axillary lymph node dissection from 2002 to 2012 were identified from a prospective database excluding those who received neoadjuvant chemotherapy. Patients whose metastases were identified by US confirmed by needle biopsy were compared with those identified by sentinel lymph node dissection (SLND) after a negative US.

Results: Metastases were identified by US in 190 patients, and by SLND in 518 patients. SLND patients had fewer positive nodes (2.2 vs. 4.1; p < 0.0001), smaller metastases (5.3 vs. 13.8 mm; p < 0.0001), and a lower incidence of extranodal extension (24 vs. 53 %; p < 0.0001) than the US group. Even when US identified ≤2 abnormal nodes, patients were still more likely to have ≥3 positive nodes (45 %) than SLND patients (19 %; p < 0.001). After adjusting for tumor size, receptor status, and histology, multivariate analysis revealed that metastases identified by US [odds ratio (OR) 4.01; 95 % confidence interval (CI) 2.75-5.84] and lobular histology (OR 1.77; 95 % CI 1.06-2.95) predicted having ≥3 positive nodes.

Conclusions: Imaging and clinicopathologic features can be used to predict the extent of nodal involvement. Patients with US-detected metastases, even if small volume, have a higher burden of nodal involvement than patients with SLND-detected metastases and may not be comparable with patients in the ACOSOG Z0011 trial.