Elafibranor, an Agonist of the Peroxisome Proliferator-Activated Receptor-α and -δ, Induces Resolution of Nonalcoholic Steatohepatitis Without Fibrosis Worsening

BACKGROUND & AIMS

Elafibranor is an agonist of the peroxisome proliferator-activated receptor-α and peroxisome proliferator-activated receptor-δ. Elafibranor improves insulin sensitivity, glucose homeostasis, and lipid metabolism and reduces inflammation. We assessed the safety and efficacy of elafibranor in an international, randomized, double-blind placebo-controlled trial of patients with nonalcoholic steatohepatitis (NASH).

METHODS

Patients with NASH without cirrhosis were randomly assigned to groups given elafibranor 80 mg (n = 93), elafibranor 120 mg (n = 91), or placebo (n = 92) each day for 52 weeks at sites in Europe and the United States. Clinical and laboratory evaluations were performed every 2 months during this 1-year period. Liver biopsies were then collected and patients were assessed 3 months later. The primary outcome was resolution of NASH without fibrosis worsening, using protocol-defined and modified definitions. Data from the groups given the different doses of elafibranor were compared with those from the placebo group using step-down logistic regression, adjusting for baseline nonalcoholic fatty liver disease activity score.

RESULTS

In intention-to-treat analysis, there was no significant difference between the elafibranor and placebo groups in the protocol-defined primary outcome. However, NASH resolved without fibrosis worsening in a higher proportion of patients in the 120-mg elafibranor group vs the placebo group (19% vs 12%; odds ratio = 2.31; 95% confidence interval: 1.02-5.24; P = .045), based on a post-hoc analysis for the modified definition. In post-hoc analyses of patients with nonalcoholic fatty liver disease activity score ≥4 (n = 234), elafibranor 120 mg resolved NASH in larger proportions of patients than placebo based on the protocol definition (20% vs 11%; odds ratio = 3.16; 95% confidence interval: 1.22-8.13; P = .018) and the modified definitions (19% vs 9%; odds ratio = 3.52; 95% confidence interval: 1.32-9.40; P = .013). Patients with NASH resolution after receiving elafibranor 120 mg had reduced liver fibrosis stages compared with those without NASH resolution (mean reduction of 0.65 ± 0.61 in responders for the primary outcome vs an increase of 0.10 ± 0.98 in nonresponders; P < .001). Liver enzymes, lipids, glucose profiles, and markers of systemic inflammation were significantly reduced in the elafibranor 120-mg group vs the placebo group. Elafibranor was well tolerated and did not cause weight gain or cardiac events, but did produce a mild, reversible increase in serum creatinine (effect size vs placebo: increase of 4.31 ± 1.19 μmol/L; P < .001).

CONCLUSIONS

A post-hoc analysis of data from trial of patients with NASH showed that elafibranor (120 mg/d for 1 year) resolved NASH without fibrosis worsening, based on a modified definition, in the intention-to-treat analysis and in patients with moderate or severe NASH. However, the predefined end point was not met in the intention to treat population. Elafibranor was well tolerated and improved patients' cardiometabolic risk profile. ClinicalTrials.gov number: NCT01694849.

Human antigen R contributes to hepatic stellate cell activation and liver fibrosis

RNA-binding proteins (RBPs) play a major role in the control of messenger RNA (mRNA) turnover and translation rates. We examined the role of the RBP, human antigen R (HuR), during cholestatic liver injury and hepatic stellate cell (HSC) activation. HuR silencing attenuated fibrosis development in vivo after BDL, reducing liver damage, oxidative stress, inflammation, and collagen and alpha smooth muscle actin (α-SMA) expression. HuR expression increased in activated HSCs from bile duct ligation mice and during HSC activation in vitro, and HuR silencing markedly reduced HSC activation. HuR regulated platelet-derived growth factor (PDGF)-induced proliferation and migration and controlled the expression of several mRNAs involved in these processes (e.g., Actin, matrix metalloproteinase 9, and cyclin D1 and B1). These functions of HuR were linked to its abundance and cytoplasmic localization, controlled by PDGF, by extracellular signal-regulated kinases (ERK) and phosphatidylinositol 3-kinase activation as well as ERK/LKB1 (liver kinase B1) activation, respectively. More important, we identified the tumor suppressor, LKB1, as a novel downstream target of PDGF-induced ERK activation in HSCs. HuR also controlled transforming growth factor beta (TGF-β)-induced profibrogenic actions by regulating the expression of TGF-β, α-SMA, and p21. This was likely the result of an increased cytoplasmic localization of HuR, controlled by TGF-β-induced p38 mitogen-activated protein kinase activation. Finally, we found that HuR and LKB1 (Ser428) levels were highly expressed in activated HSCs in human cirrhotic samples.

CONCLUSION

Our results show that HuR is important for the pathogenesis of liver fibrosis development in the cholestatic injury model, for HSC activation, and for the response of activated HSC to PDGF and TGF-β.

S-adenosylmethionine prevents hepatic tocopherol depletion in carbon tetrachloride-injured rats

In various experimental models, S-adenosylmethionine (SAMe) has been shown to reduce liver injury by preventing depletion of glutathione, one of the antioxidant systems that plays a critical role in defence against oxidative stress. On the other hand, alpha-tocopherol may be decreased in liver diseases, and treatment with this vitamin reduces liver injury in CCl(4)-treated rats. Since there is a close relationship among the different antioxidant systems (mainly glutathione, alpha-tocopherol and ascorbic acid), we have assessed whether, as well as restoring hepatic glutathione content, SAMe has any effect on liver alpha-tocopherol and ascorbic acid levels in CCl(4)-injured rats. Four groups of seven male Wistar rats treated for 9 weeks were studied: rats induced to cirrhosis with CCl(4), rats induced to cirrhosis plus SAMe administration (10 mg x kg(-1) x day(-1)) and their respective controls. Liver samples were obtained for measuring levels of glutathione, alpha-tocopherol, ascorbic acid and thiobarbituric acid-reactive substances (TBARS), and hydroxyproline concentration as an index of collagen content. The hydroxyproline content was higher in CCl(4)-injured rats than in the control group (4.4+/-1.8 and 1.1+/-0.3 micromol/g respectively; P<0.05). In CCl(4)-injured rats, SAMe administration decreased collagen content (2.7+/-1.0 microl/g; P<0.05) and TBARS, and corrected glutathione depletion. alpha-Tocopherol was significantly lower in CCl(4)-injured rats than in controls (17.3+/-4.9 and 23.0+/-4.0 micromol/g respectively; P<0.05). By contrast, alpha-tocopherol levels were similar (23.8+/-5.1 micromol/g) in CCl(4)-injured rats receiving SAMe and in controls. In CCl(4)-injured rats, liver ascorbic acid was decreased in comparison with controls (4.9+/-1.8 and 8.2+/-1.0 micromol/g respectively; P<0.05), levels which were not replenished by SAMe (4.6+/-0.4 micromol/g). In conclusion, SAMe not only decreases fibrosis and protects against hepatic glutathione depletion, but has a further antioxidant effect of preventing alpha-tocopherol depletion in CCl(4)-injured rats.

Enhanced DNA binding and activation of transcription factors NF-kappa B and AP-1 by acetaldehyde in HEPG2 cells

Because transcription factors NF-kappaB and activator protein-1 (AP-1) are known to regulate gene expression, we have analyzed the role of acetaldehyde in the activation of NF-kappaB and AP-1 in HepG2 cells. Binding activity and transactivation of NF-kappaB and AP-1 were determined by gel retardation assays and transfection of a luciferase reporter construct controlled by kappaB and AP-1 binding sites, respectively. Acetaldehyde enhanced the DNA binding of NF-kappaB and AP-1 by 1 and 4 h, respectively, increasing the kappaB- and AP-1-dependent luciferase expression. Supershift assays revealed the presence of NF-kappaB heterodimers p65/p50 and p50/p52, whereas nuclear c-Jun levels correlated with the DNA binding of AP-1. The enhanced binding of NF-kappaB to DNA by acetaldehyde in intact cells was accompanied by the proteolytic degradation of IkappaB-alpha. However, the addition of acetaldehyde to cytostolic extracts from untreated Hep G2 cells did not affect the DNA binding of AP-1 but activated the NF-kappaB heterodimer p65/p50 in the absence of IkappaB-alpha degradation. Preincubation of HepG2 cells with protein kinase C inhibitors abolished the enhanced DNA binding of NF-kappaB and AP-1 caused by acetaldehyde. Hence, these findings uncover a previously unrecognized role for acetaldehyde in the activation of NF-kappaB and AP-1, which may be of relevance in the alcohol-induced liver disease.

Differential role of ethanol and acetaldehyde in the induction of oxidative stress in HEP G2 cells: effect on transcription factors AP-1 and NF-kappaB

The oxidative metabolism of ethanol by the cytochrome P450 2E1 (CYP2E1) has been recognized to contribute to the ethanol-induced deleterious effects through the induction of oxidative stress. This study compared the effect of ethanol and acetaldehyde in the induction of oxidative stress and activation of transcription factors nuclear factor-kappaB (NF-kappaB) and activating protein 1 (AP-1) in HepG2 cells, which do not express CYP2E1, and HepG2 cells transfected with CYP2E1 (E47 cells). Neither ethanol (80 mmol/L) nor acetaldehyde (25-200 micromol/L) caused oxidative stress in HepG2 cells, an effect that was independent of blocking reduced glutathione (GSH) synthesis with buthionine-L-sulfoximine (BSO). However, BSO preincubation caused an overproduction of peroxides and activation of NF-kappaB and AP-1 in E47 cells even in the absence of ethanol. Furthermore, the incubation of E47 cells with ethanol (80 mmol/L for up to 5 days) depleted cellular GSH stores in both cytosol and mitochondria, reflecting the induction of oxidative stress. Ethanol activated NF-kappaB and AP-1 in E47 cells, an effect that was prevented by 4-methylpyrazole, potentiated by cyanamide, and attenuated by trolox C. Interestingly, however, despite the inability of acetaldehyde to induce oxidative stress in HepG2, acetaldehyde activated NF-kappaB and AP-1; in contrast, ethanol failed to activate these transcription factors in HepG2. Thus, our findings indicate that activation of NF-kappaB and AP-1 by ethanol and acetaldehyde occurs through distinct mechanisms. CYP2E1 is indispensable in the induction of oxidative stress from ethanol, whereas the activation of NF-kappaB and AP-1 by acetaldehyde is independent of oxidative stress.

Pulmonary abscess as a complication of transarterial embolization of multinodular hepatocellular carcinoma

Hepatic transarterial embolization has been widely used in the treatment of hepatocellular carcinoma and other liver tumors. Despite being a safe treatment, some complications associated with this procedure have been described. The most frequent adverse effect of transarterial embolization is postembolization syndrome, whereas infections account for only 3.4% of complication. We report the first case of lung abscess as a complication of transarterial embolization used in the treatment of multinodular hepatocellular carcinoma.

Serum amino-terminal propeptide of type III procollagen levels in chronic pancreatitis

The severity of pancreatic fibrosis, a characteristic feature of patients with chronic pancreatitis (CP), can be assessed only by direct histologic analysis of pancreatic tissue. Since serum levels of the amino-terminal type III procollagen propeptide (PIIIP) can reflect the degree of fibrogenic activity in several diseases associated with fibrosis, the current study was aimed at investigating whether PIIIP are increased in chronic pancreatitis, the relationship between PIIIP and pancreatic fibrogenic activity, and the influence of pancreatectomy, pancreatic exocrine function, and duration of disease on PIIIP levels. Serum PIIIP was measured in 18 patients with CP (15 without liver disease and three with cholestasis) and in 21 healthy controls. The effect of pancreatectomy on PIIIP was evaluated in seven patients, in whom PIIIP was measured immediately before and 2 months after surgery. Prolylhydroxylase (PHase) activity as an index of pancreatic fibrogenesis was evaluated in pancreatic tissue from 11 patients who had undergone subtotal pancreatectomy and from 11 organ donors. The bentiromide (BT)-PABA test as an index of exocrine pancreatic function was measured in all patients. PIIIP was significantly higher in patients who had or had not undergone pancreatectomy (17.3 +/- 4.0 and 25 +/- 11.4 ng/ml, respectively) than in controls (12.3 +/- 3.1 ng/ml) (p < 0.001). PIIIP decreased significantly after pancreatectomy (before, 32.0 +/- 9.3 ng/ml; after, 18.4 +/- 4.8 ng/ml; p = 0.005). PHase was significantly higher in patients (773 +/- 250 cpm/mg protein) than in controls (405 +/- 121 cpm/mg protein) (p < 0.001). PIIIP was correlated with pancreatic PHase (r = 0.7, p = 0.001) but not with BT-PABA or with the duration of the disease. In conclusion, serum PIIIP levels are increased in patients with CP and reflect the severity of pancreatic fibrogenic activity. No relationship between the serum PIIIP levels and the pancreatic exocrine function and duration of disease was found.

The prognosis of patients with alcoholic liver disease. An international randomized, placebo-controlled trial on the effect of malotilate on survival

The aim of this study was to examine the effect of malotilate on survival in patients with alcoholic liver disease and to determine prognostic variables for survival. Four hundred and seven patients with alcoholic liver diseases, from seven European liver units, entered a randomized placebo-controlled, double-blind trial: 140 patients received malotilate 1500 mg/day, 133 patients received 750 mg/day, and 134 patients received placebo. The patients were included in the study over a period of 3 1/2 years, and the study was closed 1 year after the entry of the last patient. Eighty-four patients died (35, 19, 30 patients in groups 1500 mg/day, 750 mg/day, and placebo, respectively). Survival was slightly better in the 750 mg/day group than in the two other treatment groups, when tested by conventional log-rank tests (p = 0.06). However, a treatment effect was supported by a highly significant (p = 0.006) non-proportionality of the death intensity in patients receiving 750 mg/day against those receiving either 1500 mg/day or placebo. Prognostic variables for survival were evaluated using the multiple Cox regression analysis of clinical and laboratory variables and with or without liver histology variables, as determined at entry into the study. The analysis was stratified for the three treatment regimens. In the analysis including liver histology variables, independent significant prognostic variables were: years of high alcohol intake, prothrombin index, alkaline phosphatases, creatinine, immunoglobulin M, white blood cell count, and liver cell steatosis. In the analysis without liver histology variables, prognostic variables were: years of high alcohol intake, prothrombin index, alkaline phosphatases, creatinine, and immunoglobulin M.

Relationship between hepatic lipid peroxidation and fibrogenesis in carbon tetrachloride-treated rats: effect of zinc administration

1. Lipid peroxidation and hepatic fibrogenesis were investigated in 25 carbon tetrachloride-treated rats and in 25 control animals. Rats were further divided into two groups to receive either a standard diet or one supplemented with zinc. From each group, animals were killed at weeks 3 and 18 of the experiment for histological and biochemical assessments which included hepatic lipid peroxide and collagen concentrations and plasma zinc concentration as well as the hepatic activities of proline hydroxylase and collagenase. 2. Results indicated that oral zinc supplementation was associated with a decrease in lipid peroxidation (mean 51%; P < 0.05), collagen deposition (mean 32%; P < 0.001) and proline hydroxylase activity (mean 30%; P < 0.05) at week 18, together with an increase in collagenase activity (mean 208%; P < 0.01) at week 3, in carbon tetrachloride-treated rats. 3. There was a significant direct correlation between lipid peroxidation and proline hydroxylase activity in carbon tetrachloride-treated rats (r = 0.52; P < 0.01) and also a significant inverse correlation between lipid peroxidation and plasma zinc concentration in these animals (r = -0.62; P < 0.001). 4. These findings are consistent with the hypothesis that hepatic lipid peroxidation plays an important role in the aetiology of hepatic fibrogenesis and that zinc mitigates the process.

Quantitative measurement of fibrosis in pancreatic tissue. Evaluation of a colorimetric method

A colorimetric method has been used to quantify the collagen contained in 23 specimens of pancreatic tissue (11 controls and 12 chronic pancreatitis). The method takes advantage of the selective capacity of Sirius red to stain collagen protein and of rapid green to stain noncollagen protein. The results obtained by this method were compared with those of standard morphometry to determine tissue fibrosis. With the morphometric method, the values of the control group were 6.6 +/- 4.0% (fiber area/total area), and those of chronic pancreatitis 66.0 +/- 19.0% (difference 59.4, 95% confidence interval for difference: 47.2-71.6, P less than 0.001). The values obtained with the colorimetric method were 89.1 +/- 11.6 micrograms collagen/mg total protein in the control group, and 132.7 +/- 25.3 micrograms collagen/mg total protein in the chronic pancreatitis group (difference 43.6, 95% confidence interval for difference: 26.3-61.0, P less than 0.001). A highly significant correlation (r = 0.847; p less than 0.001) was observed between the amount of collagen measured colorimetrically and the degree of fibrosis determined morphometrically. These results demonstrate that the colorimetric method is a reproducible, simple, and rapid technique to quantitate fibrosis in histological preparations of pancreatic tissue.

Human gastric alcohol dehydrogenase: its inhibition by H2-receptor antagonists, and its effect on the bioavailability of ethanol

Two types of alcohol dehydrogenase isoenzymes (differing in their affinity for ethanol, sensitivity to 4-methylpyrazole, and electrophoretic migration) have been identified in the human stomach. At the high ethanol concentrations prevailing in the gastric lumen during alcohol consumption, the sum of their activities could account for significant oxidation of ethanol. In vitro, these activities were inhibited by cimetidine and ranitidine, but not by famotidine. In vivo, therapeutic doses of cimetidine (but not of famotidine) increased blood ethanol levels when ethanol was given orally, but not when it was given intravenously, indicating a significant contribution of the gastric ADH to the bioavailability and thereby the potential toxicity of ethanol.

Gastric origin of the first-pass metabolism of ethanol in humans: effect of gastrectomy

The areas under the curve (AUCs) of blood ethanol concentrations are much smaller after oral than after intravenous administration of a small dose of ethanol. To study whether this difference is due to ethanol oxidation in the stomach, in the small intestine, or during first pass through the liver, we compared the AUCs after random administration of the same ethanol dose by the intravenous, oral, and intraduodenal routes to 5 abstaining alcoholics and via the two former routes to 10 subjects with Billroth II subtotal gastrectomy. In the nonoperated subjects, the AUCs after an ethanol dose (0.15 g/kg) given orally were 17% (p less than 0.01) of those achieved intravenously, in spite of the fact that greater than 99% of the dose had disappeared from the stomach at the completion of the AUC. By contrast, the AUCs after intraduodenal administration did not differ from those produced intravenously, indicating that neither the intestine nor the liver make a detectable contribution to this first-pass metabolism. Moreover, gastrectomy completely abolished the first-pass metabolism of ethanol. Gastric metabolism decreases the bioavailability of the ingested alcohol and thus attenuates its systemic toxicity. The abolition of this "protective barrier" in gastrectomized patients may increase their vulnerability to ethanol.

Effects of cimetidine on gastric alcohol dehydrogenase activity and blood ethanol levels

Chronic use of cimetidine and alcohol are commonly associated, but studies on their interactions are the subject of controversy. To investigate this question, a small ethanol dose (0.15 g/kg body wt) was randomly administered on 2 consecutive days either orally or intravenously to 6 normal volunteers, before and after 1 wk of oral administration of 400 mg of cimetidine twice daily. Although cimetidine did not change the areas under the curve of blood ethanol concentrations after intravenous administration, those after oral alcohol intake were twice as large with cimetidine than without. Similar effects were reproduced in rats after intravenous administration of cimetidine (50 mg/kg body wt). In vitro, cimetidine was a noncompetitive inhibitor of gastric alcohol dehydrogenase activity at concentrations as low as 0.01 mM, 100-fold lower than those needed to inhibit the hepatic dehydrogenase. These results indicate that gastric alcohol dehydrogenase activity governs, in part, the systemic bioavailability of ethanol. Consequently, systemic effects of alcohol may be exacerbated in patients receiving cimetidine.

The contribution of the stomach to ethanol oxidation in the rat

To estimate the amount of ethanol that can be oxidized in the stomach, steady-state conditions were created in a group of fed rats by giving a loading dose of ethanol (2 g/kg body wt I.V.) followed by continuous infusion either intravenously or intragastrically. The rate of ethanol oxidation was calculated from the rate of infusion required to maintain steady blood levels of approximately 30 mM for at least 3 hours. Gastrointestinal ethanol concentrations and total contents also remained steady. The rate of ethanol oxidation was 19.3% faster during intragastric than during intravenous infusion (p less than 0.01). When measured at the prevailing luminal ethanol concentration (145 mM) and expressed per body weight, the gastric ADH activity represented 14% of the hepatic activity at 30 mM ethanol, suggesting that gastric ADH activity could account for most of the increased rate of oxidation when ethanol is given intragastrically. Thus, gastric ethanol oxidation by a high Km ADH in the rat represents a significant fraction of the total rate of ethanol oxidation and it is therefore one of the factors which determines the bioavailability of orally administered ethanol.

Hepatitis B virus infection and vaccination in children undergoing hemodialysis

Among 54 patients, between 2 and 18 years of age, submitted to hemodialysis due to severe chronic renal failure, the prevalence of hepatitis B virus markers was 66.7% and that of HBsAg was 13.0%. Eighteen children, with no evidence of hepatitis B virus infection, were vaccinated. Following three vaccine injections, only 2 patients did not respond and a third one developed low anti-HBs titers. The vaccine was well tolerated. No relationship was observed between the intensity of the humoral immune response and age, sex, type of renal disease and time on dialysis. Seroconversion rates (87.5%) and geometric means of anti-HBs titers (greater than 4 000 Ausab Units) of these patients are similar to those observed following vaccination of healthy adults.

Double blind controlled trial of d-penicillamine in patients with primary biliary cirrhosis

One hundred and eighty nine patients with primary biliary cirrhosis were entered into a double blind, placebo controlled randomised trial starting in January 1978 to assess the therapeutic value of d-penicillamine 1200 mg daily. Eighteen of the 98 patients receiving d-penicillamine and 22 of the 91 placebo treated patients died during the study. Thirty six per cent of those on d-penicillamine and 8% of those on placebo were withdrawn from the study. No difference in overall survival was noted between the two groups of patients whether the results were analysed for the entire period of observation or only during the period in which the patients were receiving therapy. The mortality rate of those receiving d-penicillamine in histological stage I to II, however, was one third of that of the placebo group although this difference did not reach statistical significance. Using the occurrence rate ratio as the statistical method of analysis, no effect of d-penicillamine was noted on any clinical, biochemical or histological features examined, except the serum alanine aminotransferase activity which was greater in those on active treatment. In this trial we have been unable to establish any therapeutic benefit from the drug.