Can ultrasound or MRI correctly determine the axillary response after neoadjuvant chemotherapy in patients with breast cancer and nodal invasion?

Purpose: Current routine imaging techniques (ultrasound and magnetic resonance imaging [MRI]) have a low sensitivity and specificity to determine the axillary response after neoadjuvant chemotherapy in breast cancer patients with nodal spread and are thus not reliable to guide surgical decision-making concerning the need for or the extent of the axillary lymph node dissection. In breast cancer patients, nodal spread is linked to a worse prognostic outcome. For invasive carcinoma, axillary lymph node dissection is indispensable, as it guides adjuvant treatments. However, this surgical act also imposes major morbidity, more specifically lymphedema and paraesthesia in 10 to 20% of the cases. The purpose of this study was to evaluate whether ultrasound and magnetic resonance imaging correctly determine the axillary response after neoadjuvant chemotherapy in patients with breast cancer and nodal invasion. Materials and methods: This is a retrospective study conducted in a secondary referral hospital between January 2010 and December 2016 including 78 breast cancer patients with solely axillary lymph node invasion and no further metastases (M0), treated with neoadjuvant chemotherapy followed by surgery including an axillary lymph node dissection after which radiotherapy was applied (with or without hormone or immune therapy depending on the immunohistochemical characteristics of the tumour). The positive and negative predictive value of axillary imaging with ultrasound and MRI after neoadjuvant chemotherapy was studied for its ability to predict the lymph node invasion as determined by histological analysis after surgery. Results: Seventy-eight patients and 80 carcinomas were analysed as two patients presented a bilateral pathology. The mean age of the patients was 59 years (± 13 years), the mean size of the tumour determined by ultrasound at the start was 22 mm (± 20 mm), all carcinomas were high-grade (2 or 3) lesions and in 97.5% the tumour was of the invasive ductal histological subtype. Thirty-one axillary lesions showed a complete response as determined by imaging after the neoadjuvant chemotherapy (i.e. 38.8%). However, 15 of them had a positive histology post-surgery (i.e. 48.4%), encompassing a negative predictive value of 51.6%. Furthermore, of the 49 axillary lesions that did not show a complete response on imaging following neoadjuvant chemotherapy (i.e. 61.2%), 17 had a negative histology post-surgery (i.e. 34.7%), corresponding with a positive predictive value of 65.3%. Conclusion: Despite its accompanying morbidity, axillary lymph node dissection after neoadjuvant chemotherapy remains recommended, even if imaging suggests a complete axillary response.