C. W. Kimbrough1, K. M. McMasters1, A. R. Quillo1, N. Ajkay1 1University Of Louisville,Hiram C. Polk, Jr. Department Of Surgery,Louisville, KY, USA
Introduction: Although multiple retrospective studies suggest that occult metastases are a significant prognostic factor in breast cancer, the results of two prospective randomized trials have questioned the role of immunohistochemistry (IHC) in detecting occult disease for patients initially found to be node-negative. In this study, we sought to evaluate factors associated with overall survival in node-negative breast cancer patients staged by immunohistochemistry using the Surveillance, Epidemiology, and End Results (SEER) database.
Methods: The SEER database was queried for all patients between 2004 and 2011 with invasive lobular or ductal carcinoma and no evidence of distant metastases. Only patients with regional lymph nodes coded as negative on hematoxylin and eosin (H&E) stains that underwent additional studies using IHC were included for analysis. Patients were stratified by nodal involvement and overall survival was compared using Kaplan-Meier analysis with a log-rank test. Multivariate analysis controlling for patient and tumor characteristics was performed using a Cox-proportional hazards regression model.
Results: Overall, 93,070 patients were identified. Of these, 11,377 patients (12.2%) had occult metastases; 4657 with isolated tumor cells (N0(i+)) and 6720 with micrometastases (N1mi). On Kaplan-Meier analysis, occult metastases were associated with a small but significant decrease in overall survival (p<0.001). The 5-year survival approached 92.2% in patients without occult disease, while 5-year survival for occult metastases was 89.6%. Once further stratified by N-stage, there was no difference in overall survival observed between N0(i-) and N0(i+) patients (p<0.449), although N1mi patients demonstrated worse survival compared to both N0(i-) and N0(i+) groups (p<0.001). On multivariate analysis, micrometastasis remained an independent predictor for decreased survival compared to IHC-negative patients (HR 1.40, 95% CI 1.28–1.53), while isolated tumor cells were not a significant predictor (HR 1.05, 95% CI 0.92-1.20). Other negative prognostic factors included male sex, age at diagnosis, African-American ethnicity, increasing tumor grade, increasing T-stage, and negative hormone receptor status.
Conclusions: Patients with occult metastases found via IHC demonstrated a significant, but relatively small 2.6% overall survival difference at 5-years compared to patients with no evidence of nodal disease. Most of this survival difference is attributable to micrometastases, as isolated tumor cells have no prognostic significance in this study. Discontinuing the classification of isolated tumors cells as a separate subgroup of N0 disease warrants further consideration.
