A multicentric randomized trial was performed in premenopausal women with node-positive, estrogen-receptor-negative breast tumors to assess the potential superiority of alternating adjuvant chemotherapy over 'standard' CMF chemotherapy. Between January 1989 and June 1992, 107 patients were entered into the study and randomly allocated to receive either cyclophosphamide 100 mg/m2 per as on days 1-14, methotrexate 40 mg/m2 and 5-fluorouracil 600 mg/m2 intravenously (i.v.) on days 1, 8 (CMF), every 4 weeks for a total of 6 cycles, or the following regimens: CMF as previously; epidoxorubicin 75 mg/m2 i.v. on day 1 and vincristine 0.75 mg/m2 i.v. on days 1, 8 (EV); mitomycin-C 10 mg/m2 i.v. on day 1 and vindesine 2 mg/m2 i.v. on days 1, 8 (MVs). The three regimens were given every 4 weeks for a total of 6 cycles according to the following schedule: CMF, EV, MVs, CMF, EV, MVs. At a median follow up of 48 months (range 30-72), 40 patients have relapsed and 17 have died overall. More patients in the triple-combination arm have relapsed and more have died, the latter difference tending toward statistical significance (p = 0.06). There was no statistical difference in the site of relapse between the two groups. Total duration of adjuvant therapy was similar in the two arms (312 chemotherapy cycles in the triple arm and 308 in the CMF arm). Treatment toxicity was also comparable, although more patients in the triple-combination arm were still regularly menstruating 6 months after the completion of chemotherapy. This study failed to show any advantage ensuing from the use of alternating chemotherapy in patients with early breast cancer.