PURPOSE: In the current pTNM classification system, nodal status of breast cancer is based on the number of involved lymph nodes and does not account for the total number of lymph nodes removed. In this study, we assessed the prognostic value of the lymph node ratio (LNR; ie, ratio of positive over excised lymph nodes) as compared with pN staging and determined its optimal cutoff points. PATIENTS AND METHODS: From the Geneva Cancer Registry, we identified all women diagnosed with node-positive breast cancer between 1980 and 2004 (n = 1,829). The prognostic value of LNRs was calculated for values ranging from 0.05 to 0.95 by Cox regression analysis and validated by bootstrapping. Based on maximum likelihood, we identified cutoff points classifying women into low-, intermediate-, and high-risk LNR groups. RESULTS: Optimal cutoff points classified patients into low- (< or = 0.20), intermediate- (> 0.20 and < or = 0.65), and high-risk (> 0.65) LNR groups, corresponding to 10-year disease-specific survival rates of 75%, 63%, and 40%, and adjusted mortality risks of 1 (reference), 1.78 (95% CI, 1.46 to 2.18), and 3.21 (95% CI, 2.54 to 4.06), respectively. In contrast to LNR risk categories, survival curves of pN2 and pN3 crossed after 15 years, and their adjusted mortality risks showed overlapping CIs: 2.07 (95% CI, 1.69 to 2.53) and 2.84 (95% CI, 2.23 to 3.61), respectively. CONCLUSION: LNR predicts survival after breast cancer more accurately than pN classification and should be considered as an alternative to pN staging.