The objective of this multicenter phase III trial was to study the impact on time to treatment failure (TTF) and survival of cyclophosphamide, Adriamycin, and 5-fluorouracil (CAF) versus CAF/thiotepa, Adriamycin, vinblastine, and Halotestin (TsAVbH), a partially noncross-resistant regimen used in a rotating schedule in the treatment of hormone insensitive metastatic breast cancer in accordance with the Goldie and Coldman hypothesis.
Three hundred forty-three patients received 6 cycles of induction treatment with one of 2 regimens. Patients with estrogen receptor-negative tumors or those with estrogen receptor-positive or estrogen receptor-unknown tumors with demonstrated unresponsiveness to hormone treatment were eligible. Complete responders were randomized to either observation or maintenance therapy with cyclophosphamide, methotrexate, 5-fluorouracil, prednisone, tamoxifen, and Halotestin (CMF[P]TH). Patients with partial response or stable disease on completion of induction therapy were maintained on CMF plus Halotestin.
There were no differences in the primary end point of TTF (median 7.3 and 7.4 months, respectively). There was a significant difference in TTF and survival by duration of disease-free interval: a median of 8.8 and 21.2 months for those with a disease-free interval of ≥2 years versus 6 to 8 and 13.3 months for those with a disease-free interval <2 years (P
= 0.016 and <0.001), respectively. Toxicity of the 2 treatment regimens was similar.
There were no differences observed in TTF, survival, and toxicities between the 2 treatment arms, both of which contained doxorubicin (Adriamycin) as the most active agent. The results of observation versus maintenance in complete responders were reported separately.