Blocking EGFR Pathway May Boost Tamoxifen Response

SAN ANTONIO—Adding the epidermal growth factor receptor (EGFR) inhibitor gefitinib mesylate (Iressa, also known as ZD1839) to tamoxifen (Nol-vadex) has improved initial antitumor response and delayed the development of acquired tamoxifen resistance in a xenograft model of human breast cancer, according to Rachel Schiff, PhD, of Baylor College of Medicine’s Breast Center. Dr. Schiff reported her group’s results at the 25th Annual San Antonio Breast Cancer Symposium (abstract 18).

Dr. Schiff said that resistance to selective estrogen-receptor modulators (SERMs) arises from an imbalance in their agonist and antagonist properties. Their studies support the concept that activation of the EGFR pathway en-hances tamoxifen’s agonist effect, which contributes to acquired resistance, and show that EGFR inhibition can eliminate this agonist effect.

In preclinical models of MCF-7 breast cancer tumors overexpressing HER-2, the researchers found that tamoxifen stimulated tumor growth, thereby promoting the development of de novo resistance. By blocking the EGFR/HER-2 pathway with gefitinib, however, they were able to restore tamoxifen sensitivity in this breast cancer model.

They then examined tamoxifen in MCF-7 tumors that did not overexpress HER-2, observing that tamoxifen resistance developed within 2 to 4 months. "This acquired resistance was caused by tamoxifen-stimulated growth," Dr. Schiff said. "Tamoxifen, over time, had become more of an agonist, as it is in the HER-2 tumors. So then the question becomes: Can targeting the EGFR/HER-2 pathway delay the development of acquired resistance to tamoxifen in tumors with normal levels of EGFR/HER-2?"

To answer that question, the researchers established non-HER-2-overexpress-ing MCF-7 tumors in nude mice that were receiving estrogen supplementation. The ovariectomized mice were then randomized to continued estrogen stimulation, with or without gefitinib; estrogen deprivation, with or without gefitinib; or tamoxifen, with or without gefitinib.

"As you might expect, ZD1839 had no effect in the estrogen-stimulated tumor host," Dr. Schiff said. "This suggests that the estrogen stimulation is not dependent on the EGFR/HER-2 pathway." Gefitinib also had little benefit in mice that were estrogen deprived.

Among tamoxifen-treated mice, however, Dr. Schiff reported that gefitinib appreciably improved tamoxifen response and markedly delayed the onset of tamoxifen resistance from 2 to 3 months to more than 6 months (see Figure). The researchers also found that gefitinib had only a modest positive effect when it was added later in the treatment course, suggesting that combining the two drugs from the outset would be the most effective treatment approach.

Dr. Schiff concluded that the concept of combining ER-targeted therapies with EGFR inhibitors should be tested in clinical trials designed to delay acquired resistance. "Maybe combinations of SERMs and tyrosine-kinase inhibitors or other signaling inhibitors would give us a new tool to improve the tissue profile of SERMs," she said. "We would devise a way to get rid of the adverse agonist activities in specific tissue such as breast and uterus while retaining the beneficial antagonist effects in other tissues."