The natural history of breast carcinoma in patients with > or = 10 metastatic axillary lymph nodes before and after the advent of adjuvant therapy: a multiinstitutional retrospective study.

BACKGROUND

The majority of patients with breast carcinoma with > or = 10 metastatic axillary lymph nodes (ALNs) develop recurrent disease within 5 years from diagnosis. The purpose of the current study, performed retrospectively, was to characterize the natural history of this subset of patients, both before and after the advent of adjuvant anthracycline-based chemotherapy and tamoxifen.

METHODS

Retrospectively, patients with primary breast carcinoma (N = 882) with > or = 10 metastatic ALNs, treated between 1954 and 1998, were selected from 3 institutions: The University of Texas M. D. Anderson Cancer Center (Houston, TX); the Institut Gustave Roussy (Villejuif, France); and Hospital Clinico Universitario (Valencia, Spain). All patient data had been registered prospectively in clinical databases. One group consisted of 314 patients treated with locoregional therapy alone (no adjuvant therapy) from 1954 to 1983. The second group included 568 patients who received adjuvant anthracycline-based chemotherapy between 1974 and 1998 with or without adjuvant tamoxifen.

RESULTS

The median follow-up time was 140 months. Disease-free survival rates at 15 and 20 years for the no adjuvant therapy and adjuvant therapy groups were 17% and 16% versus 26% and 24%, respectively. The overall survival rates at 20 years for the no adjuvant therapy and the adjuvant therapy groups were 9% and 21%, respectively. By multivariate analysis, the independent factors associated with survival in the adjuvant therapy group were tumor size and the number of metastatic lymph nodes.

CONCLUSIONS

The retrospective analysis suggested that adjuvant anthracycline-based chemotherapy and hormonal therapy have altered the natural history in this high-risk group of patients. However, despite such improvements, survival rates remained low, and innovative therapeutic approaches are, therefore, needed to improve clinical outcomes.