Routine Axillary Ultrasound for Patients with T1-T2 Breast Cancer Does Not Increase the Rate of Axillary Lymph Node Dissection Based on Predictive Modeling

Since publication of the American College of Surgeons Oncology Group Z0011 trial results, demonstrating that many patients with nonpalpable axillary lymph nodes and one or two positive sentinel nodes do not require axillary lymph node dissection (ALND), preoperative axillary ultrasound (AUS) has become controversial. Clinicians are concerned that AUS may lead to unnecessary ALND. The authors developed an algorithm (Algorithm 1) in which the number of AUS-suspicious nodes and tumor biology direct management. For estrogen receptor-positive (ER+)/human epidermal growth factor receptor 2-negative (HER2-) breast cancer with a single AUS-suspicious node and a positive lymph node needle biopsy (LNNB), sentinel lymph node biopsy (SLNB) is performed with a specimen X-ray documenting retrieval of the clipped node. Other patients with positive LNNB receive neoadjuvant chemotherapy. The authors hypothesized that routine AUS and this algorithm could decrease ALND compared with a strategy of no preoperative AUS.

Decision-tree analysis and Monte Carlo simulation were used to assess the expected number of ALNDs under two strategies (routine AUS vs no AUS). Probabilities were drawn from a literature review and an institutional database. The authors assumed nodal pathologic complete response rates as reported in the literature. Four additional algorithms were created to assess whether any other treatment model could decrease the rate of ALND.

Using the routine AUS and the authors' algorithm, the predicted ALND rate was 9%, versus 10% for a strategy of no AUS, with overlapping uncertainty intervals. The remaining treatment algorithms showed similar results.

Use of AUS may help to tailor patient care without leading to overutilization of ALND, as long as neoadjuvant chemotherapy is administered when appropriate.