Abstract
4493
Background.Estrogens (E) and estrogen receptors (ER) are targets of breast cancer treatments. However, approximately 30% of ER+ tumors exhibit intrinsic hormone resistance. Methods.To address possible mechanisms underlying this, previously untreated post-menopausal patients with ER+ breast cancers were treated for 4 months in a neoadjuvant setting with the aromatase inhibitor (AI) exemestane alone, or in combination with the antiestrogen tamoxifen (TAM). Matched pre- and post-treatment tumor samples from the same patient, were analyzed by gene expression profiling using an Affymetrix U-133 2 plus platform containing about 47,000 probe sets. Independently, E and TAM regulated genes were defined using a human breast cancer xenograft model. Results. In general, tumors from patients treated with the AI+TAM combination responded less well than tumors treated with AI alone. Additionally we find that: 1. AI vs. AI+TAM treated tumors differ extensively in their overall gene expression profiles. 2. AI alone alters global gene expression ~5 times more than AI+TAM, and is 11 times more effective in modifying expression of E regulated genes; 3. Among E regulated genes, there is little overlap between AI and AI+TAM treatment groups. AI+TAM preferentially induce genes, like androgen receptors, expressing TAM “E-like” agonist activity, or genes uniquely regulated by TAM. 4. A list of 25 E regulated genes may predict response or resistance to neoadjuvant endocrine therapy in ER+ tumors. Conclusions. We have defined E and TAM regulated genes in a xenograft tumor model. In patient samples, we have identified E regulated genes that change in response to neoadjuvant endocrine therapy and are involved in estrogen-dependent tumor growth. A pre-treatment gene signature of ER+ tumors predicts response or intrinsic resistance to endocrine therapies. Under low estrogen conditions as generated by AI, the agonist properties of TAM are paradoxically exposed, diminishing the effectiveness of combination therapy.