Randomized evaluation of combination chemotherapy vs. observation alone following response or stabilization after oophorectomy for metastatic breast cancer in premenopausal women.

Premenopausal patients with progressive measurable metastatic breast cancer who demonstrated either stable or responsive disease 12 weeks following oophorectomy were randomized either to receive cyclophosphamide, methotrexate, and 5-fluorouracil (CMF) combination chemo-therapy (20 patients) or to continue under observation alone (14 patients). Since the study began in 1974, data on receptor status were not required for entry into the study. Stratification for randomization was based on the nature of the oophorectomy response (stable vs. response), dominant metastatic site (visceral vs. osseous vs. soft tissue), and disease-free interval (greater than 2 years vs. lesser than 2 years). Three (21%) of the 14 patients under observation alone continued to improve whereas 6 of 18 (33%) of the patients given CMF improved further, an insignificant difference. The median time to failure from oophorectomy was 17.5 months for the CMF group and 6.1 months for the observation group (p = 0.01). Using a multivariate proportional hazards model, visceral disease (p = 0.05) and breast involvement (p = 0.001) were also associated with significantly shorter times to failure. After the randomization, the fraction of observation patients progressing within 8 weeks was significantly greater than that of the CMF patients (5/14 vs. 0/21, p = 0.01). With 9 of the 14 observation patients and 11 of the 20 CMF patients dead, the estimated median survivals are similar at 40.4 and 41.3 months, respectively. We conclude that the addition of CMF chemotherapy to patients with stable-disease or objective response following oophorectomy significantly increases the median duration to treatment failure, whereas there appears to be no survival advantage for such therapy.