Endocrine Abstracts

Glucocorticoids (GC) are commonly prescribed, but their use is associated with significant adverse metabolic and bone effects. 11β-hydroxysteroid dehydrogenase type 1 (11β-HSD1) regenerates active GC within tissues, tightly controlling the availability of GC to bind and activate the GC receptor; Our preclinical data have shown that 11β-HSD1 knock out mice resist the adverse effects of exogenous GC excess. We have now conducted a randomized, double-blind, placebo-controlled clinical study to determine if pharmacological inhibition of 11β-HSD1 (using AZD4017) can prevent the adverse effects of prescribed GC without compromise to their anti-inflammatory actions.

30 healthy male volunteers (age; 38 ± 12 years, BMI; 25.2 ± 2.3 kg/m²) were recruited and underwent metabolic assessments including a 2-step hyperinsulinemic euglycemic clamp incorporating stable isotopes to measure glucose and fatty acid flux and adipose tissue microdialysis. Immune-inflammatory responses were measured using an OX40 (CD134) assay. All participants received prednisolone 20 mg once daily for 7 days, and, in addition, were randomized to treatment with either AZD4017 (400 mg twice daily) or matched placebo. After 7 days, all investigations were repeated. The primary endpoint was assessed using a general linear model adjusting for baseline measures. Secondary endpoints were assessed using Wilcoxon (paired), and Mann-Whitney (unpaired) tests, where appropriate.

The predefined primary end-point (change in glucose disposal (Gd)) was threefold lower in the AZD4017 group, but this failed to reach statistical significance (−0.58 ± 2.12 vs −1.56 ± 1.99, AZD4014 vs placebo P = 0.17). As expected, prednisolone + placebo worsened metabolic phenotype; Gd decreased (P = 0.004), circulating triglyceride (P = 0.018) and glycerol levels (P < 0.0001) increased, as did adipose tissue interstitial fluid glycerol release (P = 0.011). Osteocalcin (as an index of bone formation) also decreased significantly (P < 0.0001). In contrast, when prednisolone was co-administered with AZD4017, there were no significant changes in any of these variables. Importantly, the OX40 (CD134) assay demonstrated a robust response to prednisolone (1.4 ± 0.3 vs 0.6 ± 0.2%, P = 0.011), which persisted in the prednisolone + AZD4017 treated group (1.1 ± 0.3 vs 0.6 ± 0.1%, P = 0.016). Circulating prednisolone and prednisone levels were not different between the 2 groups.

We have demonstrated that 11β-HSD1 inhibition limits the adverse metabolic and bone effects of prednisolone. Oral prednisolone (an ‘active’ GC) is reliant upon tissue-specific regeneration (from inactive prednisone) for many of its biological actions and these appear independent of circulating levels. In addition, we have provided the first clinical evidence to show that 11β-HSD1 inhibition has potential as a strategy for selectively limiting the adverse side effects of prescribed GC.