Response to Neoadjuvant Chemotherapy Predicts Axillary Node Involvement in Breast Cancer Management


When therapy is given to a patient before definitive surgical management of the primary tumor, that therapy is called neoadjuvant, preoperative, or primary. Neoadjuvant therapy is being used in a wide variety of tumor types and following early successes in locally advanced breast cancer, head and neck cancer, and bladder cancer. Its use is now being extended to earlier stages of disease. The advantages associated with the use of neoadjuvant therapy have been impressive. Not only may it improve disease-free and overall survival, but it also may influence the surgeon's options for control of the primary tumor site, making less radical surgical procedures just as effective as the more aggressive and disfiguring procedures necessary to control larger tumors.

The term "locally advanced," when applied to breast cancer, has generally included patients with stage IIIB disease (which includes any size tumor with extension to the chest wall and/or inflammation) plus the subset of stage IV patients with ipsilateral supraclavicular lymph node involvement. In a recent series of 156 evaluable patients reported by Kuerer and colleagues from the M.D. Anderson Cancer Center, the definition of locally advanced disease was extended to include patients with stage IIA (T2 > 4 cm), IIB, and IIIA disease. Any patient was included in whom the primary mass was sufficiently large (and/or the breast sufficiently small) that an appropriate primary surgical procedure would have been mastectomy.

Patients were staged by a battery of tests that included abdominal computed tomography or ultrasonography, bone scan, bilateral mammograms, and breast and axilla ultrasonography before and after four cycles of chemotherapy. Chemotherapy consisted of 5-fluorouracil, doxorubicin, and cyclophosphamide (FAC). Responses to induction chemotherapy were assessed clinically and confirmed by histologic examination of the specimens removed at surgery. Post-operatively, patients with residual tumor less than 1 cm3 received four additional cycles of FAC; those with larger residual tumor burden and those with more than four positive lymph nodes in the surgical specimen received four additional cycles of FAC plus four cycles of methotrexate and vinblastine. Within six weeks of completing chemotherapy, patients received radiation therapy to the chest wall, internal mammary lymph nodes, and supraclavicular/high axillary lymph nodes. The median follow-up of these patients is about three years.

Thirty patients (19%) had a histologic complete response to four cycles of FAC chemotherapy. Nineteen of the 30 (63%) had no axillary node involvement at dissection compared to 13 of the 40 patients (33%) who had a major response to chemotherapy just short of complete response (< 1 cm3 residual disease) and 15 of the 86 patients (17%) who had larger residual tumor burden. Of the 22 patients who were rendered clinically free of disease and had a clinically negative axilla before surgery, four patients (18%) had positive nodes on dissection of the surgical specimen. Only one or two nodes were involved in these four patients. Of the 133 patients who had axillary node involvement clinically at the time of presentation, 36 (27%) had a complete pathologic response to neoadjuvant chemotherapy. Patients with N1 disease were more likely to have a complete response than those with N2 disease. Ninety-four percent of patients had some measurable response to the preoperative chemotherapy. The 10 patients who did not respond were treated with radiation therapy before surgery. Larger primary tumors were less likely to have a complete response than smaller primary tumors.

Kuerer et al conclude that patients who have a complete response to neoadjuvant chemotherapy do not need to undergo axillary dissection as a component of the surgical treatment because the likelihood of finding disease is low in those with negative clinical and ultrasound examinations. Features of the tumor, such as estrogen receptor status and tumor nuclear grade, did not predict complete response. Performance of surgery on the breast was an important component of defining patients who did not need axillary dissection. On clinical grounds, 19 patients appeared to have a complete response to neoadjuvant chemotherapy; however, 10 of these (53%) were found to have residual disease at the primary site on microscopic examination of resected tissue, and nine (47%) had axillary lymph node involvement. Thus, resection or biopsy of the primary site of disease is important. However, in light of the fact that patients generally receive axillary radiation therapy following chemotherapy, it is possible that the subset of patients with clinical complete response confirmed by biopsy of the primary site and clinically negative axilla can forego axillary dissection. (Kuerer HM, et al. Cancer J Sci Am 1998;4:230-236.)


Women with locally advanced breast cancer comprise about 10-15% of all cases of breast cancer in the United States and up to perhaps 30% in medically underserved populations. It seems clear from the available literature that neoadjuvant chemotherapy often causes significant tumor reductions and permits less extensive surgical procedures to achieve at least comparable and probably superior disease control. Over the last 30 years, the focus of breast cancer treatment has appropriately been refined. Before that time, major efforts centered on the control of local and regional disease. However, clinical investigation has now clearly identified breast cancer as a systemic disease, and multimodal approaches are being defined that are increasing the fraction of cured patients.

What is the optimal timing of locoregional and systemic therapy to improve outcome? From studies of neoadjuvant chemotherapy, first in women with inflammatory breast cancer, but more recently in studies of women with smaller primary tumors,1 it is apparent that beginning with systemic treatment has certain advantages. It permits local control with a more limited subsequent surgical approach. And perhaps, if this M.D. Anderson study is confirmed, it may define a subset of patients in whom breast biopsy alone is the sole surgical procedure following neoadjuvant chemotherapy with locoregional control being assured by post-chemotherapy radiation therapy.

It is possible to look down the road and see additional questions that may be addressed if even more effective chemotherapy regimens, such as those incorporating paclitaxel, were employed in the neoadjuvant setting. If patients with stage II and III large tumors benefit from this approach, is it possible that neoadjuvant therapy should be used as first-line therapy regardless of the size of the primary? One can envision a study in which the treatment variable is the order of treatment modalities. In women whose tumors respond completely to neoadjuvant chemotherapy, would even segmental resection or lumpectomy be considered more extensive surgical procedures than necessary to achieve local control? Bonadonna and his colleagues used primary chemotherapy in operable breast cancer (most patients with locally advanced disease are termed inoperable).2 They found that more than 70% of women treated with neoadjuvant chemotherapy followed by local radiation therapy had local tumor control without a surgical procedure. Furthermore, among the 14 patients who achieved a complete response to neoadjuvant chemotherapy and went on to axillary dissection, 13 (93%) had no pathologic axillary node involvement. In the M.D. Anderson series of patients with larger primary tumors, 63% of the complete responders had negative axillae.

Of course, one could imagine unfortunate consequences to the "flying blind" approach that would be necessary without adequate surgical staging. The treatments are not without their late sequelae and thus, one should not attempt to make up for a lack of detailed staging information by simply giving larger field radiation therapy. However, the slow extension of neoadjuvant approaches to patients who cannot reasonably be called inoperable and, thus, probably do not really have locally advanced disease, has opened the way for addressing additional questions that may lead to improved survival and quality-of-life in women with breast cancer.


    1. Bonadonna G, Valagussa P. Semin Oncol 1996;23: 464-474.

    2. Bonadonna G, et al. J Clin Oncol 1998;16:93-100.