ABSTRACT Background Neoadjuvant chemotherapy (NAC) in up to 40%–60% of node‐positive (cN+) breast cancer patients allows nodal pathological complete response, particularly in Her2+ and triple‐negative subtypes. Accurate post‐NAC axillary restaging is therefore critical to identify candidates for surgical de‐escalation. Axillary ultrasound (AUS) remains the most widely used tool, but prior studies report highly variable performance, with false‐negative rates approaching 25%, raising concerns about its reliability as a standalone guide. Methods We retrospectively evaluated 413 patients with biopsy‐proven cN+ breast cancer treated with NAC (2000–2024). All underwent AUS before surgery. AUS results (ycN0 vs. ycN+) were compared with final pathology (ypN0 vs. ypN+). Diagnostic performance was assessed, and predictors of false‐negative AUS were identified using logistic regression. Results AUS classified 51.6% of patients as ycN0, whereas only 45% achieved nodal pathologic complete response. Overall, AUS demonstrated 68% sensitivity, 76% specificity, 72% accuracy, positive predictive value 78%, and negative predictive value 66%. Performance differed by subtype: accuracy was highest in ER‐/Her2 tumors (75%), but markedly lower in Her2+ (69%) and lobular carcinoma (67%). On multivariate analysis, absence of breast clinical complete response (OR 2.38, 95% CI 1.14–4.93) independently predicted false‐negative AUS findings. Conclusions AUS following NAC provides only moderate accuracy, with a clinically relevant risk in patients lacking breast response. AUS alone should not determine omission of axillary surgery. A multimodal, biology‐informed strategy combining AUS with other imaging techniques is needed to safely guide axillary de‐escalation and minimize both overtreatment and undertreatment.
Albasini et al. (Fri,) studied this question.