Predicting non-sentinel lymph node metastasis in Australian breast cancer patients: are the nomograms still useful in the post-Z0011 era?

Abstract

Axillary lymph node dissection (ALND) can be avoided in breast cancer patients with low-volume disease in the sentinel lymph nodes (SLNs) according to Z0011 trial. We believe that nomograms developed for predicting non-sentinel lymph node (NSLN) metastases can guide the axillary treatment in patients who do not fully match the criteria of Z0011 study. We identified risk factors and evaluated the performance of three nomograms to predict NSLN status in patients with positive SLNs. Data from 526 breast cancer patients with positive SLNs who underwent ALND at two Australian hospitals from 2002 to 2015 were studied. Univariate and multivariate associations for NSLN metastasis were analysed. Predictive models evaluated were MD Anderson Cancer Centre (MDA), Helsinki University Hospital and Memorial Sloan Kettering Cancer Centre. Thirty-nine per cent of patients demonstrated NSLN metastasis. The multivariate analysis identified extranodal extension (OR 3.2, 95% CI 2.07-4.80), tumour size >2 cm (OR 2.5, 95% CI 1.66-3.89), macrometastasis (OR 1.9, 95% CI 1.09-3.47), positive SLN ratio >0.5 (OR 1.7, 95% CI 1.16-2.60) and lymphovascular invasion (OR 1.6, 95% CI 1.09-2.44) as independent predictors for NSLN metastasis. MDA nomogram showed the best discrimination (area under the curve of 0.74) and a 9% false negative rate for predicted probability of NSLN metastasis ≤10%. Our results suggest that presence of extranodal extension and tumour size >2 cm may influence the need of further axillary treatment. Conversely, ALND can be safety spared in low risk patients identified by MDA nomogram.