Increased tumor growth and drug uptake predicts for response to continuous infusion high-dose cytarabine.

One hundred twenty-four patients with recurrent or metastatic breast cancer were randomized to receive megestrol acetate 40 mg orally, four times daily, or tamoxifen 10 mg orally twice daily. If therapy failed patients were crossed over to the alternate treatment. Eligibility required that either the estrogen or progesterone receptor be positive or that both values be unknown, and that patients be at least 2 years postspontaneous menopause or over 50 years of age. Pretreatment characteristics were similar for both groups. Three patients had had previous hormonal therapy while one third had had chemotherapy. Objective response for evaluable patients based on strict UICC criteria was 29% with megestrol acetate and 31% with tamoxifen. Responses in patients with bone and soft tissue disease were similar for both regimens; however, 7 of 19 (37%) patients with visceral disease responded to tamoxifen but none of 18 (0%) responded to megestrol acetate. Response did not correlate with amount of estrogen or progesterone receptor. Unadjusted analysis of time to progression and survival showed no significant differences between regimens. With adjustment for pretreatment characteristics, patients on tamoxifen had a statistically significant prolongation of both of these parameters. Crossover data show 3 of 24 patients responding to tamoxifen after failure on megestrol acetate and 1 of 24 responding to megestrol acetate after failure on tamoxifen. However, crossover data should be viewed cautiously, as patients who are currently responding to initial treatment are those who would be most likely to respond to crossover therapy.