Classical prognostic factors were analyzed in patients with low-risk primary breast cancer, defined as absence of tumor-positive axillary lymph nodes, tumor size less than or equal to 5 cm in diameter, and no invasion into skin or deep fascia. The primary surgical treatment was total mastectomy and lower axillary dissection. None of the patients received adjuvant therapy. Between 1977 and 1990, 7315 patients entered the study, and at the time of this analysis (January 1, 1990), the median follow-up time is 5 years. In univariate analyses, the following variables were significantly related to recurrence-free survival: age in premenopausal patients; tumor size; number of negative nodes removed; histological grade; and in premenopausal patients, estrogen receptor and progesterone (PgR) status. In multivariate analyses, age in premenopausal patients was the most important factor, followed by tumor size and histological grade, whereas PgR status in premenopausal patients was just of borderline significance. These variables should be included in multivariate analyses testing the value of more recently introduced prognostic factors.