Is Ki-67 expression prognostic for local relapse in early-stage breast cancer patients treated with breast conservation therapy (BCT)?

Purpose Ki-67 is a human nuclear protein whose expression is strongly up-regulated in proliferating cells and can be used to determine the growth fraction in clonal cell populations. Although there are some data to suggest that Ki-67 overexpression may be prognostic for endpoints such as survival or postmastectomy recurrence, further elucidation of its prognostic significance is warranted. Specifically after breast conservation therapy (BCT) (defined in this setting as breast-conserving surgery and adjuvant radiation therapy), whether Ki-67 predicts for locoregional recurrence has not been investigated. The purpose of this study was to assess Ki-67 expression in a cohort of early-stage breast cancer patients to determine whether a significant independent association between Ki-67 and locoregional relapse exists. Methods and Materials Ki-67 staining was conducted on a tissue microarray of 438 patients previously treated with BCT, and expression was analyzed with clinicopathologic features and outcomes from our database. Results Ki-67 expression was more prevalent in black patients (37% of black patients vs 17% of white patients, P<.01), younger patients (27% of patients aged ≤50 years vs 15% of patients aged >50 years, P<.01), estrogen receptor (ER)-negative tumors (25% of ER-negative tumors vs 17% of ER-positive tumors, P=.04), human epidermal growth factor receptor 2 (HER2)/neu-positive tumors (35% of HER2-positive tumors vs 18% of HER2-negative tumors, P=.01), and larger tumors (26% of T2 tumors vs 16% of T1 tumors, P=.03). On univariate/multivariate analysis, Ki-67 did not predict for overall survival (74.4% vs 72.6%), cause-specific survival (82.9% vs 82.1%), local relapse-free survival (83.6% vs 88.5%), distant metastasis-free survival (76.1% vs 81.4%), recurrence-free survival (65.5% vs 74.6%), and locoregional recurrence-free survival (81.6% vs 84.7%): P>.05 for all. Conclusions Ki-67 appears to be a surrogate marker for aggressive disease and significantly correlates with known prognostic features such as age, race, hormone receptor status, and HER2 status but independently does not predict for locoregional outcomes after BCT when these other prognostic clinicopathologic features are taken into consideration. The independent associations of Ki-67 with race and age appear to be novel to our study.