1University of Leeds, School of Medicine, Leeds, United Kingdom; 2Mellanby Centre for Bone Research, Academic Unit of Bone Metabolism, Sheffield, United Kingdom; 3St Vincent’s University Hospital, Dublin, Ireland; 4University of Leeds, Clinical Trials Research Unit, United Kingdom; 5Emerging Innovations Unit, Discovery Sciences, R&D, Astrazeneca, Cambridge, United Kingdom
The causative link between circulating glucocorticoid excess and osteoporosis is established. Although circulating cortisol levels do not change significantly with age, local tissue metabolism may be implicated in age-related bone loss. The enzyme11β-hydroxysteroid dehydrogenase type 1 (11β-HSD1) increases local cortisol production, is expressed in human osteoblasts and its activity increases with age leading to a decrease in bone formation. We hypothesised that selective 11β-HSD1 inhibition may enhance post-menopausal bone formation. A UK phase II randomised, double-blind, placebo-controlled trial of 90 days treatment with AZD4017 (a selective 11β-HSD1 inhibitor) was conducted across 2 centres. 55 post-menopausal women with osteopenia on bone density criteria were recruited. Participants received 400 mg twice daily of oral AZD4017 (active) compared to matched placebo over 90 days. The primary outcome measure was impact of AZD4017 on the bone formation marker, osteocalcin. Secondary objectives included correlation with 11β-HSD1 activity, safety, tolerability and reversibility. Baseline characteristics (age, BMI, blood pressure, bone density, cortisol secretion or metabolism) were matched between the active and placebo groups. At 90 days absolute osteocalcin measurements did not differ between active group (mean (SD): 22.3 (SD 8.59) ng/ml, n = 22) versus placebo (21.7 (SD 9.21) ng/ml, n = 24). Similarly, linear regression analysis of osteocalcin at 90 days adjusted for baseline osteocalcin demonstrated an estimated treatment effect of 0.95 (95% CI (–2.69, 4.60), not significant). Selective 11β-HSD1 inhibition was inferred through a large reduction in the urinary [THF+alloTHF]/THE ratio at 90 days (active group (mean 0.1 (SD 0.12), n = 19) versus placebo (mean 0.6(SD 0.26), n = 22). This returned to baseline by 180 days. Importantly urinary free cortisol/cortisone ratio as a surrogate marker of 11β-HSD2 activity, was unchanged at 90 days (estimated treatment effect −0.054 (95% CI (–0.21, 0.10)). 5α/5β reductase activity was also unchanged (estimated treatment effect −0.19 (95% CI (−0.34, −0.039)). Linear regression analysis of osteocalcin adjusted for change from baseline at 90 days in 11β-HSD1 activity, suggests that degree of 11β-HSD1 inhibition was a poor predictor of osteocalcin at 90 days. This phase II randomised controlled trial demonstrates that AZD 4017 selectively inhibits 11β-HSD1 activity in vivo in a safe and reversible manner. However, at 90 days there is no treatment effect on bone formation as measured by osteocalcin. Therefore, the relative impairment of bone formation in post-menopausal women, is not improved by changes to the intracellular production of cortisol, at least in subjects with normal glucocorticoid levels.