The goals of this clinical trial involving postmenopausal women with metastatic breast cancer were to: (a) examine the effects of letrozole on tamoxifen (TAM) pharmacokinetics; (b) examine estrogen suppression in patients receiving TAM plus letrozole; and (c) evaluate tolerability, toxicity, objective response, and time to progression for the combination. Postmenopausal women with measurable or evaluable metastatic breast cancer received TAM (20 mg daily) for 6 weeks, and then letrozole (2.5 mg daily) was added. To examine for any effect of letrozole on the levels of TAM and two metabolites [N-desmethyl-TAM and 4-hydroxy-TAM], serum samples were obtained at 6, 12, 18, and 24 weeks. To examine for aromatase inhibition, serum samples were obtained before treatment and at 6, 12, 18, and 24 weeks for estradiol, estrone (E1) E1 sulfate, and sex hormone-binding globulin. A total of 34 patients were entered on this trial, and 23 patients were still on study at week 24, 18 of whom had blood samples available at both week 6 and week 24. The 95% confidence interval for the mean difference between levels at week 24 and levels at week 6 was -34 to 15 ng/ml for TAM, -35 to 45 ng/ml for N-desmethyl-TAM, and -1 to 2 for 4-hydroxy-TAM. For estradiol, a significant decrease (median, 88.5%; range, 73.7-95.2%) was identified after 6 weeks of letrozole, which was maintained for an additional 12 weeks. Similar significant reductions were identified for E1. E1 sulfate levels increased after 6 weeks of TAM alone but then decreased significantly after the addition of letrozole. Sex hormone-binding globulin levels were significantly elevated after 6 weeks of TAM alone and remained elevated after the addition of letrozole. Six of the 34 patients (17.6%) achieved an objective response (95% confidence interval, 6.8-34.5%), with a median time to disease progression of 7.6 months. There was no indication of a systematic decrease in TAM, N-desmethyl-TAM, or 4-hydroxy-TAM after the additional of letrozole. Estrogen suppression induced by letrozole was substantial despite the concomitant administration of TAM. The antitumor effect of TAM plus letrozole was less than expected.