Factors predicting the nodal involvement in early breast cancer

Background: The nodal status is an important prognostic factor in early breast cancer. This study assessed association between clinicopathological factors and involvement of Sentinel Lymph Node (SLN) and Non Sentinel Lymph Node (NSLN) in the SNAC 1 trial. Accurate statistical models can assist surgical counselling and potentially avoid axillary surgery in selected cases. Method: This is a retrospective analysis of 1088 patients. Lymphoscintigraphy, blue dye injection and gamma probe were used for SLN mapping and retrieved nodes were examined with H&E and immunohistochemistry. Validations of the Memorial Sloan–Kettering Cancer Center (MSKCC) and Coombs’s equations to predict the status of SLN were performed and the Area Under Curve (AUC) was calculated. Results: Positive SLNs were detected in 291 out of 1024 patients. 162(55.7%) patients had macrometastases, whereas micrometastases and isolated tumour cells were identified in 100 and 29 women respectively. Univariate analysis revealed that the involvement of SLNs was correlated with age, palpability, primary tumour site, Peritumoural Vascular Invasion (PVI), extensive intraductal component and tumour grade. Multivariate analysis demonstrated that the PVI status (p<0.001), tumour size (p<0.001), tumour site (p<0.05) were significant predictors of the SLN status. The AUCs of the (MSKCC) and Coombs’s equations were 0.723 (95% CI 0.688–0.758) and 0.693 (95% CI 0.659–0.728) respectively. Conclusion: Primary tumour characteristics were significant predictors of SLN involvement. The validation of existing models revealed that those models are imperfect for clinical use. Creation of a nomogram with high predictive performance based on the analysis of local patients’ data may yield more accurate outcomes. Reference[1]Gill G. Sentinel-lymph-node-based management or routine axillary clearance? One-year outcomes of Sentinel Node Biopsy Versus Axillary Clearance (SNAC): A Randomized Controlled Surgical Trial. Ann Surg Oncol. 2009; 16:266-75.