COMPARISON OF ADJUVANT CHEMOTHERAPY WITH METHOTREXATE AND FLUOROURACIL WITH AND WITHOUT CYCLOPHOSPHAMIDE IN BREAST-CANCER PATIENTS WITH ONE TO 3 POSITIVE AXILLARY LYMPH-NODES

Background: Alkylating agents administered as single agents or in combination with antimetabolites or anthracyclines delay the appearance of metastases and prolong the survival of breast cancer patients after surgery. Purpose: This phase III clinical trial was designed to evaluate the therapeutic efficacy and toxicity of the alkylating agent cyclophosphamide in combination with the antimetabolites methotrexate and fluorouracil adjuvant to breast cancer surgery. Methods: This study consisted of 255 breast cancer patients (a) with one to three histologically positive axillary lymph nodes and either no detectable primary tumor or operable primary tumors 5 cm or less (T0-T2) (95% of the patients) or (b) with tumors larger than 5 cm (T3) and with negative axillary nodes. Patients were randomly allocated to receive either methotrexate (60 mg/m2) and fluorouracil (600 mg/m2) (MF) intravenously on days 1 and 8 every 28 days for eight cycles or cyclophosphamide (100 mg/m2) orally on days 1-14 plus MF (CMF) every 28 days for the same duration. Median follow-up was 7.8 years, and maximum follow-up was 13 years. Results: There were no statistically significant differences in time to treatment failure or overall survival for patients treated with MF or CMF. At 8 years after completion of treatment, time to treatment failure was 55% (95% confidence interval [CI] = 50%-60%) and 59% (95% CI = 54%-64%) and overall survival was 69% (95% CI = 65%-73%) and 67% (95% CI = 62%-72%) for MF- and CMF-treated patients, respectively. The hazard ratios (MF to CMF) for time to treatment failure and for survival, estimated with a proportional hazards model, were 1.02 (95% CI = 0.69-1.50) and 0.87 (95% CI = 0.56-1.34), respectively. Myelosuppression was significantly greater (P<.0001) in CMF-treated patients during cycles 1-6. Median white blood cell count nadirs were between 4.4 X 10(3)/muL and 3.5 X 10(3)/muL in patients receiving MF and between 3.0 X 10(3)/muL and 2.4 X 10(3)/muL in those receiving CMF. Dose reductions were more frequent in CMF-treated patients, which led to statistically significant differences (P<.0001) in amounts of methotrexate and fluorouracil administered. Primary cancers at other sites occurred in 14 patients (5.5%)-six treated with MF and eight treated with CMF. Conclusions: Our findings suggest that the addition of cyclophosphamide to adjuvant chemotherapy with MF offers no therapeutic advantage but results in increased myelosuppression. Implications: Future trials will define the possible advantages of antimetabolites in adjuvant therapy. Further information will also become available when results of the ongoing National Surgical Adjuvant Breast and Bowel Project trial comparing adjuvant MF to CMF in node-negative breast cancer patients are presented.