Background: We sought to determine whether lymph node ratio (LNR; defined as number of positive nodes/number of nodes dissected) provides additional prognostic information in node-positive breast cancer patients.
Methods: Data from a cohort of 319 node-positive breast cancer patients diagnosed between 1956 and 1982 were analyzed for overall survival (OS) on the basis of current American Joint Committee on Cancer (AJCC) nodal staging versus LNR.
Results: In terms of AJCC categorization, 157 patients (49.2%) were pN1 (1-3 positive nodes), 97 (30.4%) were pN2 (4-9 positive nodes), and 65 (20.4%) were pN3 (≥10 positive nodes). In terms of LNR, 90 (28.2%) were low risk (LNR = 0.01-0.20), 119 (38.3%) were intermediate risk (LNR = 0.21-0.65), and 110 (34.5%) were high risk (LNR > 0.65). The median follow-up was 68.7 months. AJCC nodal status correlated with OS (median OS 85.9, 70.4, and 48.4 months for pN1-3, respectively, P = 0.018). LNR also correlated with OS (median OS 105.8, 72.2, and 48.4 months for the low-, intermediate-, and high-risk groups, respectively, P < 0.005). On multivariate analysis, LNR predicted OS independent of pN status (P < 0.001). Stratifying by pN status, LNR could discriminate distinct subpopulations of patients with significantly different OS rates. In a multivariate model controlling for clinicopathologic factors (tumor size, grade, estrogen receptor, progesterone receptor, and her-2-neu status), LNR remained a significant predictor of OS (P < 0.001).
Conclusions: LNR has the ability to discriminate populations with different OS rates within traditional AJCC node classification groups and predicts OS independent of traditional clinicopathologic factors. These results should be validated and considered for future incorporation into the breast cancer staging system.