Does a Picture Make a Difference? Ultrasound Guidance in the Management of the Axilla After Neoadjuvant Chemotherapy.

Abstract

Tumor size and axillary nodal status are important prognostic indicators and guide management in early-stage breast cancer. Surgery in the form of a sentinel lymph node (SLN) biopsy (SLN surgery) or full axillary lymph node dissection (ALND) remains the standard approach, with SLN surgery now established as the standard initial procedure in patients with a clinically negative axilla in view of its lower morbidity. Consequently, there is interest in expanding the use of SLN surgery alone for some patients with low volume of axillary involvement. Recent findings from the American College of Surgeons Oncology Group (ACOSOG) Trial Z0011 showed that SLN surgery without follow-up ALND could be safely offered to patients with T1-2 tumors and low axillary burden (one or two SLNs involved) when their treatment included lumpectomy and radiation therapy for completion of breast conservation. While this study enrolled fewer patients and had fewer events than originally planned, SLN surgery demonstrated noninferiority for the overall survival end point based on a one-sided hazard ratio margin of 1.3. Despite these limitations, SLN surgery alone has been embraced in the United States as a standard for patients who meet Z0011 eligibility criteria. Of interest, and contrary to protocol requirements, a nontrivial number (in both arms) of trial Z0011 patients received direct nodal irradiation through a third field (high tangents), especially among patients with more nodes involved. The impact of this is, and will remain, unknown. ACOSOG also conducted trial Z1071, where the role of SLN surgery was compared with standard ALND after neoadjuvant chemotherapy in 756 patients with operable breast cancer and pre treatment biopsy proven axillary nodal involvement (cT0-T4 cN1-N2). While feasible, the observed 12.6% false negative rate (FNR) of SLN surgery after neoadjuvant chemotherapy was higher than the clinically accepted cutoff of 10%. Consequently, the Z1071 authors concluded that strategies were needed to optimally identify patients with biopsy proven node-positive disease treated with neoadjuvant therapy whom could then be offered SLN surgery alone as an alternative to ALND. In the article that accompanies this editorial, ACOSOG investigators report on a secondary analysis of trial Z1071. This analysis assessed axillary ultrasound (AUS) as a selection criterion to stratify women for risk of residual axillary involvement following neoadjuvant chemotherapy with the goal of identifying those who could be safely spared a full ALND in the setting of negative SLN surgery. In this report, Z1071 investigators observed that the use of AUS to assess the axilla following neoadjuvant chemotherapy and before surgery reduced the FNR of SLN surgery to 9.8%, below the cutoff of 10%. This new analysis of Z1071 describes a subset of 611 patients who underwent an AUS after neoadjuvant chemotherapy, and whom had similar characteristics (based on age, clinical node status at presentation, completion of chemotherapy, and numbers of nodes removed by ALND, but not race) of patients included in the primary analysis of in trial Z1071. Central review of all images was done by a single radiologist, though no standard imaging protocol or equipment was mandated, and the FNR with and without AUS was reviewed. Four hundred and thirty patients of them (70.4%) had normal nodal anatomy by AUS based on imaging criteria of nodal cortex smaller than 3 mm and a preserved fatty hilum. Among these 430 patients with normal AUS, 342 (56.5% of them) were ultimately found to have node involvement by ALND, while 130 of the 181 patients with abnormal AUS (71.8% of them) had node involvement. These results were statically significant showing that AUS does improve predictions of nodal status. An abnormal AUS after neoadjuvant chemotherapy was also associated with more positive nodes (75.4%) compared with patients with a normal AUS (63.9%). Lower FNR was seen in the group with suspicious nodes at AUS, but this did not reach statistical significance. However, a key point is that if combined, normal AUS and SLN had a FRN of 9.8%, under the 10% threshold for clinical care. Nonetheless, in the setting of normal AUS, only 39.0% of women had complete pathologic response. This high percentage, 61%, without complete pathologic response indicates that at this time, surgical staging remains critical to appropriate care. AUS has been used extensively and is now an established imaging tool to assess the axilla as part of the diagnostic evaluation of a suspicious breast mass and in patients with newly diagnosed breast cancer. Although the performance of AUS is operator dependent, historically it has outperformed other imaging modalities like magnetic resonance imaging or positron imaging. Data on the reported sensitivity (27% to 94%) and specificity (53% to 98%) of AUS varies widely, and the evaluation of its accuracy is JOURNAL OF CLINICAL ONCOLOGY E D I T O R I A L VOLUME 33 NUMBER 30 OCTOBER 2