Sitagliptin in patients with non-alcoholic steatohepatitis: A randomized, placebo-controlled trial

AIM To evaluate the effect of sitagliptin vs placebo on histologic and non-histologic parameters of nonalcoholic steatohepatitis (NASH). METHODS Twelve patients with biopsy-proven NASH were randomized to sitagliptin (100 mg daily) (n = 6) or placebo (n = 6) for 24 wk. The primary outcome was improvement in liver fibrosis after 24 wk. Secondary outcomes included evaluation of changes in NAFLD activity score (NAS), individual components of NAS (hepatocyte ballooning, lobular inflammation, and steatosis), glycemic control and insulin resistance [including measurements of glycated hemoglobin (HbA1C) and adipocytokines], lipid profile including free fatty acids, adipose distribution measured using magnetic resonance imaging (MRI), and thrombosis markers (platelet aggregation and plasminogen activator inhibitor 1 levels). We also sought to determine the correlation between changes in hepatic fat fraction (%) [as measured using the Iterative Decomposition of water and fat with Echo Asymmetry and Least-squares estimation (IDEAL) MRI technique] and changes in hepatic steatosis on liver biopsy. RESULTS Sitagliptin was not significantly better than placebo at reducing liver fibrosis score as measured on liver biopsy (mean difference between sitagliptin and placebo arms, 0.40, P = 0.82). There were no significant improvements evident with the use of sitagliptin vs placebo for the secondary histologic outcomes of NAS total score as well as for the individual components of NAS. Compared to baseline, those patients who received sitagliptin demonstrated improved HbA1C (6.7% +/- 0.4% vs 7.9% +/- 1.0%, P = 0.02), and trended towards improved adiponectin levels (4.7 +/- 3.5 mu g/mL vs 3.9 +/- 2.7 mu g/mL, P = 0.06) and triglyceride levels (1.26 +/- 0.43 mmol/L vs 2.80 +/- 1.64 mmol/L, P = 0.08). However, when compared with placebo, sitagliptin did not cause a statistically significant improvement in HbA1C (mean difference, -0.7%, P = 0.19) nor triglyceride levels (mean difference -1.10 mmol/L, P = 0.19) but did trend towards improved adiponectin levels only (mean difference, 0.60 mug/mL, P = 0.095). No significant changes in anthropometrics, liver enzymes, other adipocytokines, lipid profile, thrombosis parameters, or adipose distribution were demonstrated. The MRI IDEAL procedure correlated well with steatosis scores obtained on liver biopsy in both groups at baseline and post-treatment, and the Spearman correlation coefficients ranged from r = 0.819 (baseline) to r = 0.878 (post-treatment), P = 0.002. CONCLUSION Sitagliptin does not improve fibrosis score or NAS after 24 wk of therapy. The MRI IDEAL technique may be useful for non-invasive measurement of hepatic steatosis.