Role of axillary lymph node dissection after tumor downstaging with induction chemotherapy for locally advanced breast cancer

Background: Induction chemotherapy has become the standard of care for patients with locally advanced breast cancer (LABC) and currently is being evaluated in prospective clinical trials in patients with earlier-stage disease. To better gauge the role of axillary lymph node dissection in patients with LABC this study was performed to assess initial axillary status on physical and ultrasound examination, axillary tumor downstaging following induction chemotherapy, and the accuracy of physical examination compared with axillary sonography in predicting which patients will have axillary lymph node metastases found on pathologic examination. Methods: Between 1992 and 1996, 147 consecutive patients with LABC were registered in a prospective trial of induction chemotherapy using 5-fluorouracil, doxorubicin, and cyclophosphamide. Physical and ultrasound examinations of the axilla were performed at diagnosis and after induction chemotherapy. Segmental resection with axillary lymph node dissection or modified radical mastectomy was performed, followed by postoperative chemotherapy and irradiation of the breast or chest wall and regional lymphatics. Results: Following induction chemotherapy, 43 (32%) of the 133 patients with clinically positive lymph nodes on initial examination had axillary tumor downstaging as assessed by physical and ultrasound examination. The sensitivity of axillary sonography in identifying axillary metastases was significantly higher than that of physical examination (62% vs. 45%, P = .012). The specificity of physical examination (84%) was higher than that of sonography (70%), but the difference did not reach statistical significance. Among the 55 patients in whom the findings of both physical and ultrasound examination of the axilla were negative following induction chemotherapy, 29 patients (53%) were found to have axillary lymph node metastases on pathologic examination of the axillary contents. However, 28 (97%) of these patients had either 1 to 3 positive lymph nodes or only micrometastases 2 to 5 mm in diameter. Conclusions: Preoperative clinical assessment of the axilla by physical examination combined with ultrasound examination is not completely accurate in predicting metastases in patients with LABC following tumor downstaging. However, patients with negative findings on both physical and ultrasound examinations of the axilla may be potential candidates for omission of axillary dissection if the axilla will be irradiated because minimal axillary disease remains. Patients who have positive findings on preoperative physical or ultrasound examinations should receive axillary dissection to ensure local control. A prospective randomized trial of axillary dissection versus axillary radiotherapy in patients with a clinically negative axilla following induction chemotherapy is currently underway.