Background and Objectives: Recently, there has been much interest in identifying primary breast cancer characteristics which have predictive value for axillary metastases. We studied breast cancer patients to determine variables associated with the incidence/extent of axillary involvement and to construct a modeled analysis. Methods: Patients with invasive ductal, lobular, and tubular breast cancer (group 1, n โซืกโฌ 15,719) were analyzed by tumor size and histology for the probability/extent of axillary metastases. A subgroup of patients was analyzed separately for any association of axillary involvement and other variables (group 2). Results: In group 1, the incidence and extent (number of positive lymph nodes) of axillary metastases correlated significantly with histology and increasing tumor size of ductal and lobular histologies. Significant associations for เด10% axillary involvement in group 2 were age and S phase for tubular histology and differentiation for ductal histology. In a multivariate analysis, increasing tumor size was the only statistically significant correlate for axillary involvement (group 2) and for increasing number of positive nodes (group 1). Conclusions: A multivariate model of tumor size and age combined with staging techniques can successfully confirm or assess extent of axillary metastases in breast carcinoma.