Combination chemotherapy for metastatic or recurrent carcinoma of the breast--a randomized phase III trial comparing CAF versus VATH versus VATH alternating with CMFVP: Cancer and Leukemia Group B Study 8281.

PURPOSE We sought to compare three doxorubicin-based therapies for metastatic breast cancer for response frequency, time to treatment failure (TTF), and survival. MATERIALS AND METHODS Women with metastatic breast cancer who had measurable disease, required laboratory tests, had received no prior chemotherapy for metastases, had a Cancer and Leukemia Group B (CALGB) performance status < or = 2, and provided informed consent were eligible. Treatment included the following: arm I--cyclophosphamide, doxorubicin, and fluorouracil (CAF); arm II--vinblastine, doxorubicin, thiotepa, and halotestin (VATH); and arm III--VATH alternating with cyclophosphamide, methotrexate, fluorouracil, vincristine, and prednisone (CMFVP) on cycles 3, 5, 7, 9, etc. Doses were modified for toxicities. Standard CALGB response and toxicity criteria were used. RESULTS Between August 1982 and February 1987, 497 women were entered and 491 were treated on study. Pretreatment characteristics were well balanced and the median follow-up duration was 79 months. There were no significant differences in response (complete [CR] plus partial [PR]) at 50% on arm I, 57% on arm II, and 51% on arm III. The median TTFs were 8, 8, and 9 months, respectively, in favor of arm III when compared with arm I (P = .028). The median survival times for treatment arms I, II, and III were 15, 17, and 17 months, respectively. After multivariate regression analyses, only estrogen receptors (ER), performance status, and number of metastatic sites influenced TTF and survival. Leukopenia was the most common grade 3 or 4 toxicity, occurring in 90%, 80%, and 92% of patients per arm, respectively. Lethal toxicities were seen in four, five, and six women, respectively. Overall, there were more grade > or = 3 toxicities on arm II than I, and most occurred on arm III (P = .02). CONCLUSION The VATH regimen appears similarly effective to the CAF regimen as initial therapy. Alternating CMFVP with VATH did not improve response rate or survival. After accounting for other variables, treatment arm was not related to outcome. New therapeutic regimens are still needed.