Lymph node ratio: a proposed refinement of current axillary staging in breast cancer patients.

摘要

BACKGROUND: The optimal method for classifying lymph node (LN) status in breast cancer patients is unknown. We sought to determine if LN ratio (LNR) improves axillary staging. STUDY DESIGN: Kentucky Cancer Registry data (1996 to 2007) were used to compare LN categorization schemas. Overall survival (OS) was evaluated using the Kaplan-Meier method and log rank tests. Schemas included: LN positive (+) vs negative (-) disease, current American Joint Committee on Cancer (AJCC) staging (0 vs 1 to 3 vs 4 to 9 vs >/=10 LN+), and LNR 0 vs 0.01 to 0.20 vs 0.21 to 0.65 vs >0.65 (LN- vs low, intermediate, and high risk LN+ groups). RESULTS: There were 1,436 patients who had complete LN evaluation data: 880 (61.3%) were LN- and 556 (39.6%) were LN+; 309 (21.5%) had 1 to 3 positive LNs, 138 (9.6%) had 4 to 9 positive LNs, and 109 (7.6%) had 10 or more positive LNs. For LN+ patients, the median number of positive LNs was 3; median LNR was 0.23. The median follow-up was 65 months. LN status was associated with 5-year OS (91.3% and 73.3% for LN- and LN+ groups, respectively, p < 0.001). Increasing AJCC pN stage was associated with worse OS (5-year OS 80.5%, 75.3%, and 49.8% for pN1 to N3, respectively, p < 0.001). LNR was also associated with OS (5-year OS of 83.1%, 72.7%, and 52.7% for the low, intermediate, and high risk LN+ groups, respectively, p < 0.001). In subgroup analyses of patients in the 1 to 3 and 4 to 9 LN+ groups, OS was statistically associated with LNR (p = 0.021 and p = 0.016, respectively). On multivariable survival analysis, LNR was associated with OS, independent of AJCC categorization, p = 0.003. CONCLUSIONS: LNR was associated with OS, regardless of AJCC LN categories.