Rifaximin-α reduces circulating ceramides and increases phosphatidylcholines in patients with ALD: A lipidome analysis of the GALA-RIF Trial

Background: Therapeutic targets to halt alcohol-related liver disease (ALD) are highly needed. The GALA-RIF trial showed that rifaximin-α reduced progression of liver fibrosis in patients with ALD, but the mechanism of action remains unknown. The lipid metabolism plays an important role in ALD, and animal models have shown that a pharmacologic reduction of hepatic ceramides can reduce alcohol-induced liver injury. We performed plasma lipidomics from patients of the GALA-RIF trial to assess the effect of rifaximin-α on the lipid profile. Methods: We used plasma EDTA from the GALA-RIF trial assessing oral rifaximin-α or placebo for 18 months in a double-blind placebo-controlled 1:1 RCT in patients with biopsy-proven ALD and no previous hepatic decompensation. Plasma was sampled at baseline, after 1 month and after 18 months. Samples were analyzed using an untargeted UHPLC-QTOF/MS lipidomics platform. We performed liver biopsies at baseline and after 18 months and assessed liver fibrosis stage according to the Kleiner fibrosis score. Histological treatment response was classified as “progression”, “stable” or “regression”. Here, we report the results based on the per-protocol population (n=108) for the completeness of the lipidomic data. Results: All 108 patients were Caucasian, 91 (84%) were men and the mean age was 59 (SD 9) years. Distribution of fibrosis stages at baseline was F0/F1/F2/F3/F4=5/31/48/18/6. Significantly fewer patients progressed in fibrosis in the rifaximin-α group (n=13, 24%) compared to placebo (n=23, 43%). We identified 231 lipid species from 16 lipid classes. After 1 month treatment, rifaximin-α significantly decreased the levels of 4 ceramides (Cers) and 2 phosphatidylcholines (PCs) and increased the levels of 2 phospholipids and 1 sphingomyelin (SM) compared to placebo. At 18 months, rifaximin-α had increased the levels of 7 PCs and 2 SMs while Cers were still decreased but not significant. In the rifaximin-α group, we could stratify patients who progressed in liver fibrosis by a distinctly different lipidome at baseline with significantly elevated levels of lysophosphatidylcholines, sphingomyelins and phosphatidylcholines. Conclusion: The beneficial effects of rifaximin-α in patients with ALD may be mediated by the modulation of the lipid metabolism, including reducing the levels of circulating hepatotoxic ceramides. Further, rifaximin-α increases the levels of fibrosis-protective phosphatidylcholines.