Exemestane significantly decreases bone mineral density in healthy postmenopausal women who take it to reduce their risk of breast cancer, according to a study published online Feb. 7 in The Lancet Oncology.

Two years of daily oral treatment with the steroidal aromatase inhibitor was found to worsen age-related decreases in BMD "by about three times, even in the setting of adequate calcium and vitamin D intake," said Dr. Angela M. Cheung of the University Health Network in Toronto and her associates.

Dr. Cheung and her colleagues performed a safety substudy as part of the National Cancer Institute of Canada’s Mammary Prevention Trial 3 (MAP.3), a double-blind randomized study showing that exemestane significantly reduced the development of breast cancer in high-risk women, compared with placebo. Previous studies had suggested that exemestane exerted a small protective effect and caused less bone loss than other, nonsteroidal aromatase inhibitors such as anastrozole and letrozole.

This substudy focused on the drug’s effect on bone health and involved 351 subjects with T-scores above 2.0 at the lumbar spine, total hip, and femoral neck at baseline. The women had adequate dietary intake of calcium and vitamin D but received supplements as well. This was not a large enough sample to assess fracture risk directly, so the researchers instead assessed the intermediate end points of bone density and structure.

Theirs is the first study to use high-resolution peripheral quantitative CT scanning as well as the usual dual-energy x-ray absorptiometry to examine bone. Unlike the latter method, high-resolution peripheral quantitative CT captures aspects of bone fragility, including microarchitectural, geometric, and material properties that heavily influence fracture risk, the investigators noted.

The study subjects were treated and followed at three medical centers in Canada and two in the United States; median age was 61 years. A total of 176 were randomly assigned to receive daily oral exemestane and 175 to receive placebo.

The primary end point, percent change in total volumetric BMD at the distal radius after 2 years, was significantly greater with exemestane (mean loss of 6.1%) than placebo (mean loss of 1.8%). Exemestane also decreased cortical volumetric BMD at the distal radius by 4.6%, compared with 1.9% for placebo.

The drug also decreased total volumetric BMD at the distal tibia (loss of 5.0%) and cortical volumetric BMD at the distal tibia (loss of 5.3%), compared with placebo (losses of 1.3% and 2.3%, respectively).

In addition, exemestane was associated with a loss of 7.9% in mean cortical thickness at the distal radius (compared with –1.1% with placebo) and a loss of 7.6% in cortical thickness at the distal tibia (compared with –0.7% with placebo). Exemestane also cut mean total area and mean cortical area at both sites, compared with placebo, Dr. Cheung and her associates said (Lancet Oncol. 2012 [doi:10.1016/S1470-2045(11)70389-8]).

With exemestane, more women had clinically significant areal BMD loss at the lumbar spine (45%), total hip (26%), and femoral neck (22%), compared with placebo (20%, 19%, and 6%, respectively). Moreover, 65% of women taking exemestane showed clinically significant areal BMD loss at one of these three sites, compared with only 35% of women taking placebo.