Two hundred ninety-three patients were randomly assigned to three treatment regimens following mastectomy for operable but prognostically unfavorable breast cancer: L-PAM, CFP, or CFP with radiation therapy. For premenopausal patients an increased risk of recurrence was associated with the presence of unfavorable local signs, large number of lymph nodes involved, greater body weight, younger age, and L-PAM treatment. For the postmenopausal patients only three factors were associated with an increased risk of recurrent disease: large tumor size, large number of lymph nodes involved, and inner/central location of the primary lesion. Specifically, the treatment employed has shown no effect. Of particular importance is the fact that for neither group of patients does our experience presently demonstrate clear association of recurrent disease with the level of drug dose administered. Furthermore, evidence suggests that although patients who experience little or no myelosuppression have significantly worse disease-free intervals than patients who experience moderate or severe myelosuppression, here is no benefit for severe myelosuppression over moderate, myelosuppression.