SAN ANTONIO – A reassuring message regarding the cardiovascular safety of adjuvant aromatase inhibitor therapy for early breast cancer is provided by a large, population-based study of patients who were treated in real-world community settings rather than in the rarified world of randomized clinical trials.
The case-control study included 44,463 women diagnosed with early breast cancer in 2001-2007 and an equal number of closely matched controls obtained from a database of 30 million privately insured individuals, Dr. Jennifer A. Ligibel explained at the San Antonio Breast Cancer Symposium.
The study was undertaken because randomized trials tend to enroll a younger and otherwise healthier population of breast cancer patients than those encountered in community practice. Plus, the randomized trials nearly always compare aromatase inhibitors (AIs) vs. tamoxifen, making it difficult to sort out whether the increased risk of cardiovascular disease that is often reported with AIs in the randomized trials is a true drug side effect, or is instead due to a possible cardioprotective effect of tamoxifen, noted Dr. Ligibel of Harvard Medical School, Boston.
During the study period, 16% of the breast cancer patients filled prescriptions for an AI, 11% for tamoxifen, and 5% for both.
With a mean follow-up of about 1,000 days, 1.3% of the nearly 90,000 women in the case-control study had an MI, 2.9% had an ischemic stroke, 1.5% had a hip fracture, and 9.5% had any fracture.
The study showed a substantial reduction in the adjusted risks of ischemic stroke (P = .03) and hip fracture (P = .005) in tamoxifen-treated patients, compared with breast cancer patients who were not on either tamoxifen or an AI.
The AI-treated patients also had a significant reduction in risk of stroke (P = .004), compared with breast cancer patients who were on no endocrine treatment.
The AI-treated women had an adjusted risk of hip fracture that was similar to that in breast cancer patients on no endocrine treatment and in controls. However, the adjusted risk of all fractures was significantly greater in women who were on an AI than in breast cancer patients on no treatment (P = .02).
Intriguingly, breast cancer patients who were not on tamoxifen or an aromatase inhibitor had significantly lower risks of MI (P = .003) and all fractures (P less than .001) than were matched controls without breast cancer. This novel observation warrants further study, Dr. Ligibel continued.
Because patients on an AI had a slightly – but not significantly – lower risk of MI than did breast cancer patients not on endocrine therapy, the AI-treated women also had a significantly lower MI risk than did the controls without breast cancer.
The breast cancer patients and controls were exceptionally closely matched: The mean age was 67 years in both groups, 87% had no comorbidities, 38% in both groups were taking drugs from more than four classes, 21% were on statins, and 8% in both groups were on a proton pump inhibitor. In all, 84% of women in each group had a high school education. Mean household income in both groups was roughly $63,400 annually.
Dr. Alan Coates
"This is basically reassuring," Dr. Alan Coates said of the Harvard findings in a conference-closing wrap-up of the year’s top developments in early breast cancer.
"What’s different about this one is it compares a baseline of breast cancer patients not receiving endocrine therapy," noted Dr. Coates of the University of Sydney. "The conclusions are quite different from those you would reach by looking at the randomized trials [that] looked at a comparison of an aromatase inhibitor vs. tamoxifen."
A meta-analysis of seven such randomized trials with follow-up of 28-100 months that was presented at the conference by Dr. Eitan Amir concluded that AI therapy was associated with a highly significant 26% increased risk of the combined cardiovascular end point comprising MI, heart failure, or angina, compared with tamoxifen therapy (P less than .001).
He noted that the absolute increase in risk associated with AIs was quite small (0.8% in excess of the 3.4% rate with tamoxifen). The number of women who would need to be treated with an AI instead of tamoxifen in order to have one additional patient experience a cardiovascular event was 132.
This small absolute risk may be magnified in women who are at increased cardiovascular risk. In 2008, the Food and Drug Administration ordered a black box warning for anastrozole (Arimidex) based upon data from the ATAC (Arimidex, Tamoxifen, Alone or in Combination) study showing a 17% incidence of ischemic cardiovascular events with anastrozole in women with preexisting heart disease, compared with a 10% incidence with tamoxifen. ATAC was one of the randomized trials included in the new meta-analysis, noted Dr. Amir, a senior fellow in medical oncology and hematology at Princess Margaret Hospital, Toronto.