Axillary ultrasound for predicting response to neoadjuvant treatment in breast cancer patients—a single institution experience

Abstract

Abstract Background In node-positive breast cancer patients at diagnosis (cN +) that render node-negative after neoadjuvant systemic treatment (NAST), axillary lymph node dissection (ALND) can be avoided in selected cases. Axillary ultrasound (AUS) is most often used for re-staging after NAST. We aimed to determine sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of AUS after NAST for predicting nodal response at the Institute of Oncology, Ljubljana. Methods Biopsy-confirmed cN + patients consecutively diagnosed at our institution between 2008 and 2021, who received NAST, followed by surgery were identified retrospectively. Only patients that underwent AUS after NAST were included. AUS results were compared to definite nodal histopathology results. We calculated sensitivity, specificity, PPV and NPV of AUS. We also calculated the proportion of patients with false-positive AUS that results in surgical overtreatment (unnecessary ALND). Results We identified 437 cN + patients. In 244 (55.8%) AUS after NAST was performed. Among those, 42/244 (17.2%) were triple negative (TN), 78/244 (32.0%) Her-2 positive (Her-2 +), and 124/244 (50,8%) luminal Her-2 negative cancers. AUS was negative in 179/244 (73.4%), suspicious/positive in 65/244 (26.6%) (11/42 (26.2%) TN, 19/78 (24.4%) Her-2 + , and 35/124 (28.2%) luminal Her-2 negative cancers). On definite histopathology, nodal complete response (pCR) was observed in 89/244 (36.5%) (19/42 (45.2%) TN, 55/78 (70.5%) Her-2 + , and 15/124 (12.1%) luminal Her-2 negative cancers). Among patients with suspicious/positive AUS, pCR was observed in 20/65 (30.8%) (6/11 (54.5%) TN, 13/19 (68.4%) Her-2 + and 1/35 (2.9%) luminal Her-2 negative cancers). Sensitivity was 29.0%, specificity 77,5%, PPV 69.2%, NPV 38.5%. Specificity and PPV in TN was 68.4% and 45.4%, in Her-2 + 76.4% and 31.6%, in luminal Her-2 negative 93,3% and 97,1%, respectively. Conclusion In approximately half of the patients, AUS falsely predicts nodal response after NAST and may lead to overtreatment in 30% of the cases (ALND). However, AUS has to be interpreted in context with tumor subtype. In luminal Her-2 negative cancers, it has a high PPV and is therefore useful.