Aromatase inhibitors have demonstrated their superiority to tamoxifen as adjuvant therapy for early breast cancer in postmenopausal women, but are associated with an increased risk of fractures. The aim of our study was to analyze bone loss, bone turnover and their determinants in postmenopausal women treated with anastrozole. We investigated bone loss and bone turnover markers (BTM) in a prospective open cohort study of 118 postmenopausal women treated with anastrozole for an early hormone-dependent breast cancer. Women without osteoporosis were not treated and compared with an age-matched control group of 114 healthy women. Osteoporotic patients (T-score<or=-2.5 S.D.) received weekly risedronate. Bone mineral density (BMD), and the BTM serum osteocalcin and serum C-terminal cross linking telopeptide of type I collagen (CTX) and 17beta-estradiol were measured at baseline and 1 year later. In the surveillance group, anastrozole induced after 1 year of treatment a marked bone loss at the spine (mean+/-S.E.M., [95% confidence interval]) -3.3+/-0.4% [-4.1 to -2.5]), and hip (2.8+/-0.4% [-3.6 to -2]) that was significantly greater than in controls (p<0.0001). Anastrozole induced an increase in bone remodelling: osteocalcin (+36.6%, p<0.0001) and CTX (+34%, p<0.0001). In univariate models, a recent menopause, a low body mass index, a complete chemotherapy (>or=6 courses) and a marked antiestrogenic response--defined by a level of 17beta-estradiol<or=2 pg/ml at 1 year or a decrease >50% between baseline and 1 year--were associated with greater bone loss. In multivariate model, women in the highest quartile of bone loss at the spine (>5.6% at 1 year) and hip (>4.9%) had a marked antiestrogenic response with OR of 10.4 [95% C.I. 1.9-57.2] (p=0.007) and 5.7 [1.3-25] (p=0.024) respectively. Among patients in the surveillance group, those with a normal T-score at both sites (n=46) had also a significant bone loss at spine -3.3+/-0.5% [-4.3 to -2.3], p<0.0001 and at the hip -2.9+/-0.6% [-4.1 to -1.7] p<0.0001. In osteoporotic women treated simultaneously with anastrozole and risedronate, bone loss was prevented at hip, and increased at the spine (+4.1+/-0.9% [2.3 to 5.9], p=0.008), and BTM decreased (-24%, -39% for CTX, p=0.003 and 0.001 vs. changes in the untreated group). Anastrozole increases bone turnover and induces an accelerated bone loss that is significantly related to the suppression of 17beta-estradiol production induced by aromatase inhibitor. The bisphosphonate risedronate prevents anastrozole induced bone loss.