SAN ANTONIO—There was no difference between the aromatase inhibitors (AI) anastrozole and exemestane when used as initial adjuvant therapy in postmenopausal women with hormone receptor-positive primary breast cancer, according to the final analysis of the MA.27 trial. The study’s original hypothesis had been that the steroidal exemestane would show better efficacy and end-organ safety than the nonsteroidal anastrozole.
“Exemestane was not shown to be superior to anastrozole,” said presenter Paul Goss, MBBCH, director of breast cancer research and codirector of the breast cancer disease program at Massachusetts General Hospital in Boston. “There were no significant differences found in event-free survival (EFS), distant disease-free survival, disease-specific survival, or overall survival. Importantly, [adverse] event rates were low in both arms of the trial, probably a reflection of this low-risk population and improvements in breast cancer care, with 91% of patients disease free at median follow-up of 4.1 years.”
Paul Goss, MD, PhD, MBBCH
Anastrozole (Arimidex) is currently approved as upfront adjuvant, five-year monotherapy in hormone receptor-positive early breast cancer. Exemestane (Aromasin) is approved for use in sequence after two to three years of initial tamoxifen therapy. Some of the benefits of exemestane are that it is the more potent AI, does not induce intratumoral estrogens, and, through its mild androgenic activity, may exert a second antitumor effect; it also has a more favorable lipid metabolism profile, Dr. Goss said.
For this study, 7,576 women were randomized, after local treatment, to five years of adjuvant anastrozole (1 mg/day) or exemestane (25 mg/day). The primary endpoint was an improvement in EFS from 87.5% for anastrozole to 89.9% for exemestane. Of the total patient population, 1,620 women were randomized to receive the COX-2 inhibitor celecoxib (Celebrex) along with the AI. However, about 18 months into the trial, concerns about cardiovascular toxicities with COX-2 inhibitors caused this randomization to be discontinued, Dr. Goss said. Patient stratification factors included prior chemotherapy, lymph node status, and trastuzumab (Herceptin) use (SABCS 2010 abstract S1-1).
According to the results, the EFS came in with a hazard ratio (HR) of 1.02 (P = .85) in both treatment arms. The EFS was similar in both arms regardless of nodal status: In lymph node-negative patients, the HR was 1.04 (P = .726) and in lymph-node positive patients, the HR was 0.99 (P= .896). The HR for overall survival was 0.93 (P = .64). For distant disease-free survival, the HR was 0.93 (P = .46), and for disease-free survival, the HR was 0.93 (P = .62).
With regard to adverse events, 70% were grade 1 or 2. Vaginal bleeding, hypertriclyceridemia and hypercholesterolemia, and self-reported cases of a new osteoporosis diagnosis were less common with exemestane. Cardiovascular events were not different between the two arms, although atrial fibrillation was less common with anastrozole.
SABCS attendee Charles Vogel, MD, asked about the cost difference between the two AIs. He pointed out that in the state of New York, anastrozole costs about $16/month while exemestane is around $400/month. “The no-brainer is to give the cheaper drug,” Dr. Vogel said.
Dr. Goss responded that all three AIs, including letrozole (Femara), are scheduled to go off-patent in the coming years, which should bring the costs of the drugs in line with each other. “Access to these three AIs will appear and generic versions will, or have, become available,” he said. “These results provide patients with an additional choice depending on their local access to care.” For patients who develop symptoms on anastrozole or exemestane, the study results confirm that switching to the other agent appears to confer no harm, he said.
The drugs for this study were provided by the developers, but that is no longer the case, Dr. Goss said. As a result, there will be no long-term follow up in these patients. However, his group is conducting subset analysis on AIs and bone density and psychosexual side effects, with some results slated for presentation at the American Society of Clinical Oncology (ASCO) 2011 meeting, Dr. Goss said. “The goal of all these substudies is to personalize therapy for AIs,” he added.