Primary tumour response on breast MRI as a predictor of axillary pathologic response in breast cancer patients treated with neoadjuvant chemotherapy

Objectives To investigate whether breast radiologic response on breast MRI can predict axillary pathologic response (ypN0 vs. ypN+) in breast cancer patients treated with neoadjuvant chemotherapy (NACT) in the total study population and in subgroups according to clinical nodal (cN)-status and oestrogen receptor (ER)-expression. Materials and methods This retrospective single-centre study consecutively included breast cancer patients who, between 2012 and 2022, had undergone baseline and post-NACT breast MRI. Measures of predictive ability, including sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and diagnostic odds ratio (DOR) with 95% confidence intervals (CI), were calculated. Subgroup analyses were performed for cN-status and ER expression. Results Of 251 included patients, 48.6% were cN0 and 34.7% had ER-negative tumours. For axillary pathologic response prediction by breast MRI, the sensitivity was 84%, specificity 40.4%, PPV 48.3%, NPV 79.2%, and DOR 3.6 (95% CI: 1.83-7.00). The probability of achieving ypN0 was 60.2% in all patients, 82% in cN0 patients versus 39.5% in cN+ patients, and 78.2% in ER-negative tumours versus 50.6% in ER-positive tumours. High NPVs were observed in cN0 patients (92%) and ER-negative tumours (90.1%) and were lower in cN+ patients (62.2%) and ER-positive tumours (72.2%). Conclusions The probability of ypN+ given breast complete response on breast MRI was below 10% in cN0 patients and ER-negative tumours. Therefore, in these subgroups, breast complete response on breast MRI may be clinically useful to exclude ypN+. In cN+ patients and ER-positive tumours, due to lower NPVs, breast complete response on breast MRI may not be clinically useful to exclude ypN+.