In hope of reducing the duration of adjuvant treatment in premenopausal patients with operable breast cancer and histologically positive axillary lymph nodes, a prospective controlled study was started in September 1975. A total of 160 patients were randomized to receive 12 cycles of adjuvant CMF, while 165 were allocated to receive 6 cycles. At 3 years from radical mastectomy, the relapse-free survival was 85.4% in the 12-cycle group compared to 82.6% for the 6-cycle subset (P = 0.29). In both treatment groups, the incidence of treatment failure was directly proportional to the number of involved axillary nodes and to the clinical tumor size. Drug-induced amenorrhea as well as estrogen receptor status failed to significantly affect the results obtained. Treatment failures were more often documented in distant sites, while only 4--5% relapse was observed in local-regional areas. Overall survival was also similar for both treatment arms (86.2% vs. 85.1%, P = 0.49). Toxicity was moderate and reversible and no drug-induced neoplasms were so far observed. Present results achieved with 6 cycles of adjuvant CMF appear encouraging. However, they are still too preliminary to recommend the routine use of 6 cycles instead of the classical 12 cycles.