BACKGROUND: The absolute number of involved axillary lymph nodes (LNs) is considered the most important prognostic factor in breast cancer. Over the last decade, several studies indicated that the lymph node ratio (LNR) might predict outcome better than the number of positive LNs. In this study we test the applicability of earlier published LNR cutoff values and study the prognostic value of the LNR on a nationwide level. METHODS: A nationwide population-based study was performed, using data from the Netherlands Cancer Registry, including all women diagnosed with node-positive breast cancer between 1999 and 2005 (N = 25,315). Patients were divided into 3 LNR risk groups (low, ≤0.20; intermediate, 0.21-0.65; and high, >0.65). Kaplan-Meier survival analysis was performed. In order to evaluate whether LNR was associated with overall survival (OS), Cox proportional hazards modeling was used. RESULTS: For the entire cohort, 5- and 10-year OS rates were 78 % and 62 %, respectively. The number of positive LNs correlated with OS (5-year OS 84 %, 72 %, and 55 % for patients with 1-3, 4-9, and 10 or more positive LNs, respectively, P < .001). LNR also correlated with OS (5-year OS 86 %, 75 %, and 54 % for low-, intermediate-, and high-risk groups, respectively, P < .001). In the multivariable analysis, the risk of death increased with increasing LNR (P < .001). CONCLUSIONS: The LNR has an important prognostic value in node-positive patients, independent of traditional clinicopathological factors. LNR should be added as an independent prognostic variable to the current staging system.
BACKGROUND: The absolute number of involved axillary lymph nodes (LNs) is considered the most important prognostic factor in breast cancer. Over the last decade, several studies indicated that the lymph node ratio (LNR) might predict outcome better than the number of positive LNs. In this study we test the applicability of earlier published LNR cutoff values and study the prognostic value of the LNR on a nationwide level. METHODS: A nationwide population-based study was performed, using data from the Netherlands Cancer Registry, including all women diagnosed with node-positive breast cancer between 1999 and 2005 (N = 25,315). Patients were divided into 3 LNR risk groups (low, ≤0.20; intermediate, 0.21-0.65; and high, >0.65). Kaplan-Meier survival analysis was performed. In order to evaluate whether LNR was associated with overall survival (OS), Cox proportional hazards modeling was used. RESULTS: For the entire cohort, 5- and 10-year OS rates were 78 % and 62 %, respectively. The number of positive LNs correlated with OS (5-year OS 84 %, 72 %, and 55 % for patients with 1-3, 4-9, and 10 or more positive LNs, respectively, P < .001). LNR also correlated with OS (5-year OS 86 %, 75 %, and 54 % for low-, intermediate-, and high-risk groups, respectively, P < .001). In the multivariable analysis, the risk of death increased with increasing LNR (P < .001). CONCLUSIONS: The LNR has an important prognostic value in node-positive patients, independent of traditional clinicopathological factors. LNR should be added as an independent prognostic variable to the current staging system.