Prognostic indicators of relapse-free and overall survival in patients with high-risk primary breast cancer following high-dose chemotherapy

Recent data suggest an overall decrease in breast cancer mortality. However, patients (pts) with high-risk primary breast cancer (HRBC [stage II disease with 10 axillary lymph nodes involved, stage IIIA, or IIIB inflammatory neoplasm]) have a high risk of relapse, ≥ 10% per year from the time of diagnosis. Administration of high-dose chemotherapy and stem cell rescue (HDCT) as a component of adjuvant therapy is one method of potentially improving outcome; it is unclear, however, which subset of patients with HRBC may benefit from this modality. Methods: We assessed relapse-free (RFS) and overall survival (OS) in HRBC pts treated with either single cycle or tandem HDCT. To define predictors of outcome immunohistochemistry (IH) and H&E stain-analysis was carried out by the same pathologist (P.C.) on the primary tumors. Age at diagnosis, stage of disease, tumor grade, number of axillary nodes involved, ER and PR status, proliferation rate by Ki-67, ErbB-2 overexpression, markers of cell cycle regulation/apoptosis (p21, p27, p16, p53), and the presence of dendritic cells/degree of microvascular density (by CD1a and CD34 staining) were analyzed as potential predictors of RFS and OS using both univariate and multivariate Cox regression. Results: Paraffin blocks were procured from 135 of the 222 HRBC patients who received HDCT between 8/94 and 8/98. The median follow-up is 53 months (range, 27-77 months). Sixty-four pts (47%) with stage II, 50 (37%) with stage IIIA, and 21 (16%) with stage IIIB HRBC were analyzed. Five-year projected RFS is 64% (95% CI, 56%-73%) and OS is 74% (95% CI, 67%-83%). After adjusting for age and stage, preliminary data from multivariate analysis suggested that p16 negativity/weak expression, p27, p21, and ER/PR positivity, and low/intermediate grade were associated with favorable RFS; OS was longer in patients with ER/PR, and p27 positive, and p16 negative/weakly positive tumors. Conclusion: The effect of biological tumor markers should be included as stratification factors and in the analysis of current and future trials of HDCT.