The introduction of aromatase inhibitors as a new class of agents represents a further step in improving breast cancer treatment and possibly in preventing this disease. Although these agents are now used in the first- and second-line treatment of postmenopausal breast cancer, the heterogeneity of patients enrolled in clinical trials prevents a thorough assessment of the effectiveness of potential sequential and combination therapies. Such investigations are more easily performed in the laboratory, and to this end, a tumor model in nude mice was established to simulate several aspects of the postmenopausal breast cancer patient. This model showed that aromatase inhibitors are more efficient than tamoxifen at reducing tumor volume. Additionally, the combination of an aromatase inhibitor plus tamoxifen does not improve the antiproliferative results obtained with the aromatase inhibitor alone, a finding corroborated in the Arimidex, Tamoxifen Alone or in Combination adjuvant clinical trial. To investigate the effect that potential sequences of treatment have on tumor growth, letrozole was administered in sequence with tamoxifen to nude mice bearing human xenografts. Tumor growth was significantly reduced with the sequence compared with tamoxifen alone. Additionally, when agents were alternated every 4 weeks, mice started on letrozole fared better than those started on tamoxifen. Finally, letrozole alone provided the best and most sustained reduction in tumor growth. These experiments suggest the means to evaluate therapeutic combinations in the laboratory to guide potential trial designs and provide the best chance of success to the patients who enter these clinical trials. © 2003 Elsevier Inc. All rights reserved.