Cardiovascular Profiles of Aromatase Inhibitors May Be Different, but Clinical Relevance Unknown

PARIS—Although aromatase inhibitors have gained acceptance as recommended adjuvant therapy for postmenopausal women with hormone-receptor positive breast cancer, in the absence of controlled head-to-head trials it is not clear whether there are clinically relevant differences among the three available agents—anastrozole, letrozole, and exemestane. A study presented here at the European Cancer Conference suggests that the three agents have different cardiovascular profiles and that anastrozole has improved cardiovascular safety compared with the other two aromatase inhibitors (AIs). However, the study was not a comparative trial between the three agents, but was instead based on separate analyses of three respective large randomized controlled landmark trials of each drug. “The differences in cardiovascular safety profiles between the AIs confirm that these drugs are not interchangeable. At present, anastrozole is the only AI in the adjuvant setting with a detailed benefit-risk profile from over five years of follow-up, from which no cardiovascular safety concerns emerged,” stated Jean-Marc Nabholtz, MD, of the Breast Cancer Research Institute in Paris, a principal investigator of the ATAC (Arimidex [anastrozole], Tamoxifen, Alone or in Combination) trial, which found that anastrozole significantly reduced the risk of recurrence in postmenopausal women with hormone-positive breast cancer. He analyzed safety data from the BIG (Breast International Group) 1–98 study, which compared the use of letrozole versus tamoxifen in about 8,000 women; IES (Intergroup Exemestane Study), which compared exemestane versus tamoxifen in about 5,000 women; and ATAC monotherapy arms, which compared anastrozole versus tamoxifen in about 6,000 women. At a median follow-up of 26 months in BIG I-98, letrozole-treated patients had a significantly greater incidence of moderate to severe cardiac events compared with tamoxifen: 2.1% vs 1.1 %, respectively, of Grade 3 to 5 adverse cardiac events. Seven cerebrovascular deaths occurred in the letrozole group compared with one in the tamoxifen group; cardiovascular deaths occurred in 13 and six patients, respectively. For IES, at 37.4 months median follow-up, a significantly greater incidence of myocardial infarction was found with exemestane compared with tamoxifen: 20 occurrences vs 8 occurrences, respectively. For ATAC, 68 months of follow-up showed that stroke was significantly reduced in the anastrozole arm compared with tamoxifen (62 vs 88 cases, respectively), and that tamoxifen was associated with more ischemic cardiovascular events than anastrozole was: 104 for anastrozole vs 127 for tamoxifen. Dr. Nabholtz said that the majority of cardiovascular events observed in ATAC were mild to moderate in severity and that the incidence of myocardial infarction was similar for anastrozole and tamoxifen. Also, the numbers of cardiovascular deaths were similarly low for both treatment groups, and the majority of these deaths occurred after treatment was stopped. ‘Contentious Issue’ “The situation of cardiovascular profiles of the different AIs is a contentious issue,” said the Discussant for the study, R. Charles Coombes, MD, of Imperial College School of Medicine, Hammersmith Hospital Cancer Center, in London. Although some differences between AIs were shown in the data reported by Dr. Nabholtz, Dr. Coombes said that the numbers of cardiac events in the different trials were quite small, and that the data were collected differently in each study. It is possible that there are differences between these drugs in terms of cardiovascular safety and it is possible that the incidence of cardiovascular and ischemic events is increased with letrozole, Dr. Coombes said, explaining that anastrozole may be associated with fewer cardiovascular events, but that these are noncomparative data. Longer follow-up of existing studies and head-to-head trials to compare AIs would shed more light on the cardiac safety of these agents, he concluded.