Mammographic density change with 1 year of aerobic exercise among postmenopausal women: a randomized controlled trial

Abstract

Background: The Alberta Physical Activity and Breast Cancer Prevention (ALPHA) Trial examined the influence of aerobic exercise on biological factors that are associated with breast cancer risk. Mammographic density, a secondary outcome, is reported here.

Methods: The ALPHA Trial was a parallel group randomized controlled trial conducted between May 2003 and July 2007. Postmenopausal, sedentary women ages 50 to 74 years (n = 320) were evenly randomized to aerobic exercise (45 minutes, 5 days per week) or control (usual life-style) for 1 year. Dense fibroglandular tissue and nondense fatty tissue were measured from mammograms at baseline and 1 year using computer-assisted thresholding software for area measurements and a new technique that relies on the calibration of mammography units with a tissue-equivalent phantom for volumetric measurements.

Results: Nondense volume decreased in the exercise group relative to the control group (difference between groups = -38.5 cm(3); 95% confidence interval, -61.6 to 15.4; P = 0.001). Changes in total body fat accounted for this decrease. Changes in dense area and dense volume, measures that have previously been associated with breast cancer risk, were not significantly different between the groups (P > or = 0.26).

Conclusions: Achieving changes in mammographic measures may require more exercise or a study population with higher baseline levels of sex hormones or a wider range of mammographic density. The data from this study, however, suggest that the protective effect of exercise on breast cancer risk may operate through a mechanism other than mammographic density.

Trial registration: ClinicalTrials.gov NCT00522262.

Conflict of interest statement

Disclosure of Potential Conflicts of Interest: No potential conflicts of interest were disclosed.

Figures

Figure 1

Participant recruitment, randomization, and follow-up. * From 1284 responses from mailings to 4543 screening program participants and 2170 responses from media campaigns. † Excludes subjects with missing values of covariates (n = 3 controls, n = 2 exercisers).

Figure 2

Distribution of mammographic measures at baseline.

Figure 3

Change in mammographic measures before and after adjustment for change in percent body fat, by intervention group. * P-value for difference in mean change between exercisers and controls, from general linear models. † Adjusted for baseline mammographic density, age, age at menarche, age at first birth, and years since menopause.

Figure 4

Change in nondense area and nondense volume before and after adjustment for change in percent body fat, by mean duration of exercise. * Difference between mean change in exercisers in that exercise duration group and mean change in controls. † P-value for difference in mean change between exercisers in that exercise duration group and controls. ‡ Adjusted for baseline mammographic density, age, age at menarche, age at first birth, and years since menopause. § Low duration: < 150 min/week. For analysis of nondense area, n = 38, and nondense volume, n = 33 exercisers. ∥ Moderate duration: 150– 225 min/week. For analysis of nondense area, n = 65, and nondense volume, n = 60 exercisers. ** High duration: > 225 min/week. For analysis of nondense area, n = 47, and nondense volume, n = 47 exercisers.