Abstract
The number of positive axillary lymph nodes (LNs) is the only node-related factor for prognostic evaluation of breast cancer recognized by AJCC (TNM staging). However, N staging may not completely reflect LN tumor involvement due to the erroneous count of LNs in the presence of matted LNs and different tumor volume in LNs. Additionally, the positive/total LN ratio (LNR) has been shown to outperform N staging in survival prediction. In our study, to better quantify the tumor involvement of axillary LNs, we measured the cross-sectional cancer area (CSCA) of the positive LNs in 292 breast cancer patients diagnosed between 1998 and 2000 in our institution and compared its prognostic value to that of number of positive LNs (metLN)/N stage and LNR. Statistical analyses of these three LN-related factors were performed by Kaplan-Meier method and multivariate Cox's regression model. Patients were divided into three groups based on the different LN CSCA (<50, 50-500, and >500 mm(2)), or LNR (<0.1, 0.1-0.65, and >0.65), or N stage (N1-N3). Multivariate analysis demonstrated LNR was the most significant LN-related survival predictor with hazard ratio (HR) 25.0 (P = 0.001), compared to the metLN (HR 0.09, P = 0.052) and CSCA (HR 2.24, P = 0.323).