Reply to “Implications of abnormal preoperative axillary imaging in the post Z011 era”

We would like to take the opportunity to address a number of points brought up in the accompanying commentary by Drs. Selleck and Senthil. The objective of the paper under discussion is to examine whether the routine practice of axillary imaging is beneficial in selecting patients who will require an axillary lymph node dissection (ALND) when managed according to ACOSOG Z0011 eligibility criteria. The study specifically refers to women presenting with cT1-2N0 invasive breast cancer undergoing breast-conserving surgery. Our results highlight that regardless of axillary imaging findings, the majority of patients presenting with early-stage disease in the absence of palpable adenopathy do not have ≥3 positive sentinel lymph nodes (SLNs) and therefore can be spared the morbidity of ALND. The authors of the commentary question the results by comparing our findings to papers with different methodologies and patient populations. The referenced study by Reyna and colleagues reports on a single-institution experience with routine axillary ultrasound and reflex fine needle aspiration (FNA) for any abnormal-appearing lymph node. Among a cohort of 384 women with cT1-2N0 invasive breast cancer, they report a false-negative rate of axillary ultrasound with FNA of 48%. It is well known that the positive predictive value of axillary imaging is improved with the addition of FNA. At the present time, the relevant clinical question is not simply the ability of ultrasound with or without FNA to identify an axillary metastasis, but how frequently 3 or more nodes containing metastases are identified, since this is the population who would be spared sentinel node biopsy. At our institution, we do not routinely perform axillary imaging or FNA for clinically node-negative women undergoing upfront breast-conserving surgery, as ALND is only performed in women found to have 3 or more positive SLNs