Advertisement

A randomized placebo-controlled trial of elafibranor in patients with primary biliary cholangitis and incomplete response to UDCA

Open AccessPublished:January 20, 2021DOI:https://doi.org/10.1016/j.jhep.2021.01.013

      Highlights

      • In a phase II trial in PBC, the dual PPARα and δ agonist elafibranor achieved the primary endpoint of reducing ALP.
      • Compared to placebo, the 80 mg and 120 mg dose had positive effects on prognostically important composite endpoints.
      • In patients with pruritus at baseline, an improvement was noted in both elafibranor arms.
      • Elafibranor was generally well-tolerated over the duration of treatment.

      Background & Aims

      Patients with primary biliary cholangitis (PBC) who have an incomplete response to ursodeoxycholic acid remain at risk of disease progression. We investigated the safety and efficacy of elafibranor, a dual PPARα/δ agonist, in patients with PBC.

      Methods

      This 12-week, double-blind phase II trial enrolled 45 adults with PBC who had incomplete response to ursodeoxycholic acid (alkaline phosphatase levels ≥1.67-fold the upper limit of normal (ULN). Patients were randomly assigned to elafibranor 80 mg, elafibranor 120 mg or placebo. The primary endpoint was the relative change of ALP at 12 weeks (NCT03124108).

      Results

      At 12 weeks, ALP was reduced by -48.3±14.8% in the elafibranor 80 mg group (p <0.001 vs. placebo) and by -40.6±17.4% in the elafibranor 120 mg group (p <0.001) compared to a +3.2±14.8% increase in the placebo group. The composite endpoint of ALP ≤1.67-fold the ULN, decrease of ALP >15% and total bilirubin below the ULN was achieved in 67% patients in the elafibranor 80 mg group and 79% patients in the elafibranor 120 mg group, vs. 6.7% patients in the placebo group. Levels of gamma-glutamyltransferase decreased by 37.0±25.5% in the elafibranor 80 mg group (p <0.001) and 40.0±24.1% in the elafibranor 120 mg group (p <0.01) compared to no change (+0.2±26.0%) in the placebo group. Levels of disease markers such as IgM, 5’-nucleotidase or high-sensitivity C-reactive protein were likewise reduced by elafibranor. Pruritus was not induced or exacerbated by elafibranor and patients with pruritus at baseline reported less pruritic symptoms at the end of treatment. All possibly drug-related non-serious adverse events were mild to moderate.

      Conclusion

      In this randomized phase II trial, elafibranor was generally safe and well tolerated and significantly reduced levels of ALP, composite endpoints of bilirubin and ALP, as well as other markers of disease activity in patients with PBC and an incomplete response to ursodeoxycholic acid.

      Lay summary

      Patients with primary biliary cholangitis (a rare chronic liver disease) that do not respond to standard therapy remain at risk of disease progression toward cirrhosis and impaired quality of life. Elafibranor is a nuclear receptor agonist that we tested in a randomized clinical trial over 12 weeks. It successfully decreased levels of disease activity markers, including alkaline phosphatase. Thus, this study is the foundation for a larger prospective study that will determine the efficacy and safety of this drug as a second-line therapy.

      Clinical trial registration number

      Clinical Trials.gov NCT03124108

      Graphical abstract

      Keywords

      Introduction

      Primary biliary cholangitis (PBC) is a rare cholestatic liver disease affecting approximately 20-40 per 100,000 people with a strong predominance in women.
      • Selmi C.
      • Bowlus C.L.
      • Gershwin M.E.
      • Coppel R.L.
      Primary biliary cirrhosis.
      ,
      • Lindor K.D.
      • Gershwin M.E.
      • Poupon R.
      • Kaplan M.
      • Bergasa N.V.
      • Heathcote E.J.
      • et al.
      American association for study of liver diseases: primary biliary cirrhosis.
      In a large cohort of patients with PBC, elevation of ALP was associated with a 2.0–2.5-fold higher risk of liver transplantation or death.
      • Lammers W.J.
      • van Buuren H.R.
      • Hirschfield G.M.
      • Janssen H.L.A.
      • Invernizzi P.
      • Mason A.L.
      • et al.
      Levels of alkaline phosphatase and bilirubin are surrogate end points of outcomes of patients with primary biliary cirrhosis: an international follow-up study.
      Although occurring later during the natural history of the disease, elevated levels of bilirubin are associated with a 5–10-fold higher risk of liver events and death relative to patients with normal values.
      • Lammers W.J.
      • van Buuren H.R.
      • Hirschfield G.M.
      • Janssen H.L.A.
      • Invernizzi P.
      • Mason A.L.
      • et al.
      Levels of alkaline phosphatase and bilirubin are surrogate end points of outcomes of patients with primary biliary cirrhosis: an international follow-up study.
      Ursodeoxycholic acid (UDCA) is the first-line therapy for PBC since it has been shown that UDCA-induced decreases in ALP levels correlate with a delay in the development of liver-related outcomes and death.
      • Poupon R.E.
      • Lindor K.D.
      • Cauch-Dudek K.
      • Dickson E.R.
      • Poupon R.
      • Heathcote E.J.
      Combined analysis of randomized controlled trials of ursodeoxycholic acid in primary biliary cirrhosis.
      However, 15 to 40% of patients respond incompletely to UDCA or – less frequently – are intolerant.
      • Kaps L.
      • Grambihler A.
      • Yemane B.
      • Nagel M.
      • Labaenz C.
      • Ploch P.
      • et al.
      Symptom burden and treatment response in patients with primary biliary cholangitis.
      ,
      European Association for Study of Liver
      EASL Clinical Practice Guidelines: the diagnosis and management of patients with primary biliary cholangitis.
      Importantly, persistent ALP elevation translates into an increased risk of end-stage liver disease and death compared to patients with a complete response to UDCA therapy.
      • Carbone M.
      • Mells G.F.
      • Pells G.
      • Dawwas M.F.
      • Newton J.L.
      • Heneghan M.A.
      • et al.
      Sex and age are determinants of the clinical phenotype of primary biliary cirrhosis and response to ursodeoxycholic acid.
      The GLOBE PBC and UK-PBC scores,
      • Carbone M.
      • Sharp S.J.
      • Flack S.
      • Paximadas D.
      • Spiess K.
      • Adgey C.
      • et al.
      The UK-PBC risk scores: derivation and validation of a scoring system for long-term prediction of end-stage liver disease in primary biliary cholangitis.
      ,
      • Lammers W.J.
      • Hirschfield G.M.
      • Corpechot C.
      • Nevens F.
      • Lindor K.D.
      • Janssen H.L.A.
      • et al.
      Development and validation of a scoring system to predict outcomes of patients with primary biliary cirrhosis receiving ursodeoxycholic acid therapy.
      which include liver biochemistry following treatment with UDCA as well as surrogate measurements of disease stage, perform better than previously described response criteria, namely, Paris (1 and 2), Barcelona, Toronto, and Rotterdam, at estimating the risk of mortality or liver transplantation in patients with PBC. Considering the increased risk of mortality in patients with an inadequate response to UDCA, second-line therapies are required. The selective steroidal farnesoid X receptor (FXR) agonist obeticholic acid (OCA) has been approved for this indication after demonstrating improvement of ALP levels in patients with incomplete response to UDCA.
      • Nevens F.
      • Andreone P.
      • Mazzella G.
      • Strasser S.I.
      • Bowlus C.
      • Invernizzi P.
      • et al.
      A placebo-controlled trial of obeticholic acid in primary biliary cholangitis.
      In the 12-month pivotal study, 46–47% of patients treated with the 2 doses of OCA reached the primary endpoint (ALP <1.67x upper limit of normal [ULN] with >15% ALP reduction and normal bilirubin) while 11–16% of patients experienced adverse events, in part related to treatment-emergent worsening of pruritus.
      • Nevens F.
      • Andreone P.
      • Mazzella G.
      • Strasser S.I.
      • Bowlus C.
      • Invernizzi P.
      • et al.
      A placebo-controlled trial of obeticholic acid in primary biliary cholangitis.
      More recently, the pan-peroxisome proliferator-activated receptor (PPAR) agonist bezafibrate used for 24 months in patients with incomplete response to UDCA was shown to produce a response rate of 31% on a more stringent primary endpoint (complete normalization of ALP, aminotransferases, total bilirubin and albumin and normal prothrombin index) without exacerbating pruritus.
      • Corpechot C.
      • Chazouilleres O.
      • Rousseau A.
      • Le Gruyer A.
      • Habersetzer F.
      • Mathurin P.
      • et al.
      A placebo-controlled trial of bezafibrate in primary biliary cholangitis.
      The favourable efficacy to safety ratio of bezafibrate has raised interest in PPAR agonists as new second-line therapies in PBC and supports further exploration of this class of drugs.
      Activation of PPARα by specific agonists modulates bile acid metabolism through inhibition of bile acid synthesis (inhibition of CYP7A1 expression), detoxification of bile acids (induction of CYP3A4 and SULT2A1 or UGT2B4), increased bile acid output (increased expression of BSEP and MRP2) and decreased bile toxicity (induction of MDR2/3 and ABCG5/G8) by favouring formation of phospholipid-rich micelles.
      • Ghonem N.S.
      • Assis D.N.
      • Boyer J.L.
      Fibrates and cholestasis.
      ,
      • Li F.
      • Patterson A.D.
      • Krausz K.W.
      • Tanaka N.
      • Gonzalez F.J.
      Metabolomics reveals an essential role for peroxisome proliferator-activated receptor alpha in bile acid homeostasis.
      Furthermore, both activation of PPARα and PPARδ are associated with anti-inflammatory effects by inhibiting the NF-κB and AP-1 pathways,
      • Pawlak M.
      • Bauge E.
      • Bourguet W.
      • De Bosscher K.
      • Lalloyer F.
      • Tailleux A.
      • et al.
      The transrepressive activity of peroxisome proliferator-activated receptor alpha is necessary and sufficient to prevent liver fibrosis in mice.
      while PPARδ activation has also been shown to have anti-inflammatory effects through the BCL6-mediated pathway.
      • Lee C.H.
      • Chawla A.
      • Urbiztondo N.
      • Liao D.
      • Boisvert W.A.
      • Evans R.M.
      • et al.
      Transcriptional repression of atherogenic inflammation: modulation by PPARdelta.
      Thus, activation of PPARα and PPARδ modulates complementary pathways and may confer a therapeutic benefit in patients with PBC. PPAR activation might also be effective in reducing the disease burden by ameliorating pruritus,
      • Reig A.
      • Sesé P.
      • Parés A.
      Effects of bezafibrate on outcome and pruritus in primary biliary cholangitis with suboptimal ursodeoxycholic acid response.
      which is supported by data derived from the BEZURSO trial.
      • Corpechot C.
      • Chazouilleres O.
      • Rousseau A.
      • Le Gruyer A.
      • Habersetzer F.
      • Mathurin P.
      • et al.
      A placebo-controlled trial of bezafibrate in primary biliary cholangitis.
      Considering the pharmacological profile of the drug and its liver targeting,
      • Staels B.
      • Rubenstrunk A.
      • Noel B.
      • Rigou G.
      • Delataille P.
      • Millatt L.J.
      • et al.
      Hepatoprotective effects of the dual peroxisome proliferator-activated receptor alpha/delta agonist, GFT505, in rodent models of nonalcoholic fatty liver disease/nonalcoholic steatohepatitis.
      the present phase II trial aimed to assess the therapeutic efficacy and safety of 2 doses of elafibranor in patients with PBC who had an incomplete response to UDCA.

      Patients and methods

      The study protocol and amendments were reviewed by national authorities and Ethics committees at each investigational centre. The trial was registered on www.clinicaltrials.gov (NCT03124108).

      Study population

      This study included adult patients (age 18 to 75 years) with PBC as demonstrated by the presence of at least 2 of the following 3 diagnostic factors: i) a history of elevated ALP levels for at least 6 months prior to randomization, ii) positive anti-mitochondrial antibody titre (>1/40 on immunofluorescence or M2 positive by ELISA) or positive PBC specific anti-nuclear antibodies, iii) liver biopsy consistent with PBC. All patients were treated with UDCA for at least 12 months and were at a stable dose for at least 6 months prior to randomization. At inclusion, patients were required to have ALP levels ≥1.67x ULN (ULN = 104 U/L for females; 129 U/L for males). No minimum baseline pruritus was required for inclusion. The main exclusion criteria were i) other liver diseases including viral hepatitis (HBV and HCV), primary sclerosing cholangitis (PSC), alcohol-related liver disease, autoimmune hepatitis or overlap, non-alcoholic steatohepatitis (NASH), or history of alpha 1-antitrypsin deficiency; ii) alanine aminotransferase (ALT) or aspartate aminotransferase (AST) >5xULN, total bilirubin >2xULN, platelet count <150x103/μl, albumin <3.5 g/dl; iii) moderate or severe hepatic impairment (Child-Pugh B/C); iv) history of liver transplantation, current placement on a liver transplant list, current model for end-stage liver disease (MELD) score ≥15, signs and symptoms of cirrhosis/portal hypertension including ascites, oesophageal varices, history of variceal bleeding, hepatic encephalopathy, history of bacterial peritonitis, hepatocellular carcinoma or hepatorenal syndrome. Excluded medications were defined as follows: fibrates, obeticholic acid and glitazones within 2 months prior to screening, azathioprine, colchicine, cyclosporine, methotrexate, mycophenolate mofetil, pentoxifylline; budesonide and other systemic corticosteroids within 3 months prior to screening, or immunotherapy directed against interleukins or other cytokines or chemokines within 12 months prior to screening visit.

      Study design

      This was a randomized, double-blind, placebo-controlled clinical trial with 3 parallel groups. Eligible patients who had signed the informed consent were randomized, using an Interactive Response Technology centralized randomization system, in a 1:1:1 ratio to receive elafibranor 80 mg, elafibranor 120 mg, or placebo once daily for 12 weeks. UDCA treatment was continued throughout the study and maintained thereafter. The randomized treatment allocation was performed by permuted block randomization. During the study, investigators, patients and study personnel were blinded to the treatment allocation. A total of 45 patients were recruited at 21 investigational centres in the US and Europe between 2 May 2017 and 23 July 2018. Assessment visits occurred at randomization (day 0), week 2, week 4, week 8 and week 12. An end-of-study visit was planned following a protocol amendment implemented after the study start and was performed in a subset of patients after an off-study drug period of 16 to 30 days in 23 patients. At each visit, safety was assessed clinically, and blood samples were collected for measurement of efficacy and safety markers by a central laboratory. ALP, liver enzymes and 5’-nucleotidase levels and safety markers were measured at each visit. The bile acid precursor 7α-hydroxy-4-cholesten-3-one (C4), high-sensitivity C-reactive protein (hsCRP), IgM, and other inflammatory markers were measured at randomization (day 0) and at week 12. Safety assessment included physical examination, vital signs, arterial pressure, electrocardiogram and clinical laboratory testing with haematology, plasma lipids and renal function markers. Pruritus was evaluated at each visit using a visual analogue scale (VAS), PBC-40 quality of life questionnaire
      • Jacoby A.
      • Rannard A.
      • Buck D.
      • Bhala N.
      • Newton J.L.
      • James O.F.W.
      • et al.
      Development, validation, and evaluation of the PBC-40, a disease specific health related quality of life measure for primary biliary cirrhosis.
      and 5 D-itch questionnaire.
      • Elman S.
      • Hynan L.S.
      • Gabriel V.
      • Mayo M.J.
      The 5-D itch scale: a new measure of pruritus.

      Primary and secondary efficacy endpoints:

      The primary endpoint was the relative change (%) in serum ALP levels from baseline (day 0) to end-of-treatment period at week 12. Secondary endpoints were the percentages of patients achieving pre-defined therapeutic responses. Notably 2 main composite definitions were pre-defined: i) ALP <1.67xULN and total bilirubin <ULN and ALP reduction >15%, ii) ALP <1.5xULN and total bilirubin <ULN and ALP reduction >40%. Response rate was also assessed according to ALP reduction >10%, >20% and >40%, Paris I, Paris II, Toronto I, Toronto II criteria and GLOBE score.
      • Lammers W.J.
      • Hirschfield G.M.
      • Corpechot C.
      • Nevens F.
      • Lindor K.D.
      • Janssen H.L.A.
      • et al.
      Development and validation of a scoring system to predict outcomes of patients with primary biliary cirrhosis receiving ursodeoxycholic acid therapy.
      Other secondary endpoints included change from baseline in gamma-glutamyltransferase (GGT), ALT, AST, 5’-nucleotidase, total and conjugated bilirubin, albumin, IgM, hsCRP and other inflammatory markers, bile acid precursor C4, fibroblast growth factor-19 (FGF19) and plasma lipids (total cholesterol, LDL-cholesterol, HDL-cholesterol and triglycerides). Change in symptoms was evaluated using PBC-40 Quality of Life questionnaire.

      Safety

      Safety and tolerability assessment included adverse events, laboratory variables, a VAS for pruritus, the 5-D pruritus questionnaire (measuring the degree, duration, direction [improvement or worsening], disability [effect on daily activities], and distribution of itching), electrocardiography, physical examination, and vital signs.

      Sample size and statistical analysis

      Sample size was estimated assuming a standard deviation for the primary endpoint of 18% in each elafibranor arm and 15% for the placebo arm. Fifteen patients per arm (45 in total) were calculated to achieve at least 80% power to detect, for each dose-placebo comparison, an absolute difference of -20% in the ALP relative change from baseline with alpha risk of 0.05 using a 2-sided, 2-sample, unequal-variance t test. The 20% difference was chosen based on the effect of OCA in the phase II study in PBC.
      • Hirschfield G.M.
      • Mason A.
      • Luketic V.
      • Lindor K.
      • Gordon S.C.
      • Mayo M.
      • et al.
      Efficacy of obeticholic acid in patients with primary biliary cirrhosis and inadequate response to ursodeoxycholic acid.
      Efficacy analyses were performed on the modified intention-to-treat (mITT) population which was pre-defined as all randomized patients who received at least 1 dose of treatment, with available baseline value and at least 1 post-baseline value under treatment for the primary endpoint. A per-protocol population excluding patients with major deviations to protocol was used for confirmatory analysis on the primary and key secondary efficacy endpoints. All tests of hypotheses were 2-sided and conducted at the 5% significance level, and all CIs were 2-sided at the 95% level. No adjustment for multiplicity was made for the primary and secondary efficacy endpoints. For the main analysis, the primary efficacy endpoint was compared between each elafibranor dose and placebo using a non-parametric randomization-based analysis of covariance (ANCOVA) with baseline value as a covariate. A sensitivity analysis was also performed using an ANCOVA model with baseline value as a covariate. For continuous secondary efficacy endpoints, the differences in each dose-placebo comparison were assessed on the relative change from baseline using the same method as for the main analysis of the primary efficacy endpoint, and on the absolute change from baseline using an ANCOVA model with baseline value as a covariate. For secondary efficacy endpoints involving binary outcomes, the differences in proportions were assessed independently for each elafibranor dose-placebo comparison using a Fisher exact test. The safety analyses were performed in the safety population pre-defined as all randomized patients who received at least 1 dose of study medication. All statistical analyses were performed using SAS® (Version 9.3 or higher, SAS Institute Inc., Cary, NC, USA).

      Results

      Sixty-eight patients from 21 centres in Europe and the US were screened for potential inclusion in the trial (Fig. S1). Forty-five were randomized to placebo, elafibranor-80 mg or elafibranor 120 mg groups (1:1:1; 15 patients per group). Demographics and baseline clinical characteristics were generally similar between treatment groups (Table 1). In the ITT population, 96% of patients were women with a PBC diagnosis based on positive anti-mitochondrial antibody test and elevated ALP (1 patient had a diagnosis of PBC based on a liver biopsy). No biopsy proven cirrhotic patient was included. The mean age was comparable in all groups: 56.5 years in the elafibranor-80 mg, 60.4 years in the elafibranor 120 mg and 60.5 years in placebo. On average, patients in the elafibranor 80 mg group had a higher baseline level of ALP compared to the other groups. Similarly, baseline GGT and 5’-nucleotidase as well as ALT and AST levels were numerically higher in the elafibranor 80 mg group. In all groups, patients had mild to moderately advanced PBC, according to Rotterdam criteria,
      • Zhang Y.
      • Li S.
      • He L.
      • Wang F.
      • Chen K.
      • Li J.
      • et al.
      Combination therapy of fenofibrate and ursodeoxycholic acid in patients with primary biliary cirrhosis who respond incompletely to UDCA monotherapy: a meta-analysis.
      with no sign of major liver dysfunction as illustrated by bilirubin, albumin or platelet levels within normal ranges (1 patient in the 80 mg group had a total bilirubin >ULN at baseline). All randomized patients completed the study except for 1 in the elafibranor 120 mg group who stopped participation after only 1 dosing because of a non-drug-related SAE (ischaemic stroke). This patient did not have a post-baseline ALP value under treatment and was not included in the mITT population. In each group 1 patient was not included in the per-protocol data set because of major protocol deviation.
      Table 1Demographic and baseline characteristics of the study population (ITT).
      Placebo (n = 15)Elafibranor

      80 mg (n = 15)
      Elafibranor

      120 mg (n = 15)
      All (n = 45)
      Gender (M/F)1/141/140/152/43
      Age, years60.5 (8.6)56.5 (8.7)60.4 (6.9)59.1 (8.15)
      UDCA dose, mg/kg14.80 (2.14)14.30 (4.01)13.50 (3.05)14.19 (3.13)
      ALP
      Upper limit of normal of ALP was 104 IU/L for women and 129 IU/L for men.
      , IU/L
      296.2 (115.5)350.6 (152.1)263.7 (137.6)303.5 (137.7)
      GGT, IU/L229.6 (115.9)282.3 (215.7)161.8 (139.9)224.6 (166.7)
      ALT, IU/L48.5 (22.3)57.6 (24.7)40.5 (16.9)48.9 (22.2)
      AST, IU/L46.7 (15.8)54.3 (18.7)39.4 (15.3)46.8 (17.4)
      Bilirubin mg/dl0.651 (0.259)0.577 (0.389)0.585 (0.303)0.605 (0.316)
      Direct bilirubin mg/dl0.301 (0.122)0.320 (0.290)0.240 (0.070)0.288 (0.185)
      Albumin, g/L42.5 (2.9)41.0 (3.3)41.3 (2.8)41.6 (3.0)
      Platelets, 109/L251.9 (74.6)271.7 (68.3)234.0 (88.6)252.6 (77.4)
      5’-Nucleotidase, IU/L14.4 (11.6)21.2 (16.3)14.3 (20.4)16.6 (16.4)
      Creatinine, μmol/L66.90 (13.56)66.61 (10.21)65.15 (11.53)66.22 (11.60)
      Cystatin C, mg/L1.03 (0.25)0.97 (0.20)0.96 (0.27)0.99 (0.24)
      Creatine kinase, IU/L109.9 (84.9)50.7 (26.2)75.5 (40.2)78.7 (60.2)
      IgM, g/L4.60 (2.70)2.95(1.01)3.36 (2.16)3.64 (2.15)
      hsCRP mg/L5.30 [0.5;12.3]6.40 [2.4;29.3]3.90 [0.4;35.0]5.30 [0.4;35.0]
      C4, nmol/L34.0 (51.6)38.6 (37.6)42.5 (29.6)38.7 (39.8)
      Descriptive statistics. Values are number of patients for gender, mean ± (SD) for continuous variables except for hsCRP which is noted as median with [min; max].
      ALP, alkaline phosphatase; ALT, alanine aminotransferase; AST, aspartate aminotransferase; C4, 7α-hydroxy-4-cholesten-3-one; GGT, gamma-glutamyltransferase; hsCRP, high-sensitivity C-reactive protein; UDCA, ursodeoxycholic acid.
      Upper limit of normal of ALP was 104 IU/L for women and 129 IU/L for men.

      Primary efficacy endpoint and effects on ALP levels

      The primary efficacy endpoint of the trial was met at both the 80 and 120 mg doses of elafibranor (Fig. 1A). Compared to placebo, the relative reduction in ALP levels from day 0 to the end-of-treatment period (week 12) was statistically significant in the 2 elafibranor-treated groups. The mean relative changes ±SD were -48.3±14.8% in the elafibranor-80 mg group, and -40.6±17.4% in the elafibranor-120 mg group compared to +3.2±14.8% in the placebo group. The resulting elafibranor effects vs. placebo (Mean [95%CI]) were -52.0% [-62.5% to -41.5%] ( p <0.001) and -43.9% [-55.7% to -32.1%] (p <0.001) for the elafibranor 80 mg and elafibranor 120 mg arms, respectively (Fig. 1A). These significant effects on ALP (p <0.001) were confirmed in the per-protocol population and in the sensitivity analysis. Compared to the placebo arm, which exhibited stable levels of ALP throughout the study, ALP declined starting at the first on-treatment visit (week 2) in the 2 elafibranor groups, and continued to decline at a slower rate until the end of the treatment period. During follow-up, ALP increased in the 2 elafibranor groups after stopping study drug. ALP at baseline was highest in the elafibranor 80 mg group, however the relative changes vs. baseline were comparable in the 2 elafibranor-treated groups (Fig. 1B and 1C).
      Figure thumbnail gr1
      Fig. 1Effects of elafibranor on ALP levels in the mITT population.
      (A) Primary efficacy endpoint: treatment effect (±95%CI) of elafibranor 80 mg and 120 mg vs. placebo on relative change in ALP at end-of-treatment (week-12); (B) Time course of ALP levels in placebo and elafibranor-treated groups (mean±95%CI); (C) Time course of relative change vs. baseline in placebo and elafibranor-treated groups (mean±95%CI). ∗∗∗p <0.001 vs. placebo according to non-parametric ANCOVA with baseline value as covariate. An end-of-study visit was planned following a protocol amendment implemented after the study start and was performed only in a subset of patients (n = 23) after an off-study drug period of 16 to 30 days. ALP, alkaline phosphatase; mITT, modified intention-to-treat.

      Response rates on ALP defined efficacy targets and composite endpoints

      In contrast to placebo-treated patients, almost all individuals exposed to elafibranor experienced a sustained drop in ALP (Fig. S2). More than 90% of patients treated with elafibranor 80 mg and 120 mg doses had reductions of ALP ≥20% (14/15 and 13/14, respectively) compared to only 6.7% (1/15) in the placebo arm (p <0.001) (Fig. 2A). None of the placebo-treated patients (0/15) showed ALP reductions ≥40%, contrasting with high proportions observed in the elafibranor-treated groups (13/15 or 86.7% in the 80 mg arm and 8/14 or 57.1% in the 120 mg arm). Finally, the proportion of patients with normalized ALP levels at the end of the 12-week treatment period was 13.3% (2/15) in the elafibranor 80 mg and 21.4% (3/14) in the elafibranor 120 mg group, but 0% in the placebo group (Fig. 2A). Both doses of elafibranor showed significant effects vs. placebo on a composite endpoint that is used in pivotal phase III trials. Significantly more patients achieved the composite endpoint of ALP <1.67 ULN, bilirubin <ULN and >15% ALP reduction in the elafibranor-treated groups (10/15 or 66.7% at 80 mg and 11/14 or 78.6% at 120 mg) vs. placebo group (1/15 or 6.7%). Similarly, 53.3% (8/15) in the elafibranor 80 mg group and 35.7% (5/14) in the elafibranor 120 mg group reached the more stringent objective of ALP <1.5xULN, total bilirubin <ULN and ALP reduction ≥40%, while no patient in the placebo group was able to achieve this objective (Fig. 2A). Using the composite GLOBE score, both doses of elafibranor increased the estimated transplant-free survival rates at 5, 10 or 15 years (Fig. 2B). In addition, the efficacy of both doses of elafibranor was also demonstrated on composite scores that define the risk of PBC-associated complications including liver transplant or death, like Paris II, Toronto I, Toronto II and Barcelona criteria (data not shown).
      Figure thumbnail gr2
      Fig. 2Effects of elafibranor on secondary efficacy endpoints in the mITT population.
      (A) Response rate on composite endpoints proposed as surrogate of liver outcomes and death and on proportion of patients with a 20%, 40% reduction or ALP normalization in elafibranor and placebo-treated groups. ∗p <0.05; ∗∗p <0.01; ∗∗∗p <0.001 vs. placebo according to Fisher exact test. (B) Treatment effect (±95% CI) of elafibranor 80 mg and 120 mg vs. placebo on absolute change in transplant free survival rate according 5-year, 10-year and 15-year GLOBE score from day 0 to the end-of-treatment period (Week 12). ∗p <0.05 vs. placebo according to ANCOVA with baseline value as covariate. ALP, alkaline phosphatase; mITT, modified intention-to-treat.

      Investigation of PBC-related markers, inflammatory markers, aminotransferases and plasma lipids.

      GGT level remained stable throughout the treatment period in placebo-treated patients (+0.2±26%), while significant reductions were observed in both elafibranor-treated groups (at week-12: -37.1±25.5%; p <0.001 vs. placebo with 80 mg and -40.0±24.1%; p <0.01 vs. placebo with 120 mg) (Table 2 and Fig. 3). The GGT change over time (Fig. 3A and C) was similar to the changes in ALP observed in the elafibranor-treated groups (Fig. 1A and C). Additionally, a reduction of 5’-nucleotidase at both doses of elafibranor vs. placebo was observed at week 12 (Table 2; Fig. 3). Finally, significant decreases in IgM and inflammatory markers, including C-reactive protein and haptoglobin, were observed in the elafibranor-treated groups compared to the placebo-treated group (Table 2, Fig. S3). ALT and AST were only moderately elevated at baseline and remained stable throughout the treatment period. At week 12, the change vs. baseline value was comparable in the 2 elafibranor and the placebo-treated groups (Table 2, Fig. S4). Baseline values of bilirubin, platelets and albumin were all within the normal range and no significant changes occurred during the study period except for albumin that increased significantly by 1.8 g/dl in the elafibranor 120 mg group vs. placebo. As expected, patients had features of PBC-related dyslipidaemia, notably high HDL-cholesterol at baseline. Compared to placebo, elafibranor-treated groups showed decreases in total cholesterol, LDL-cholesterol and triglycerides (Table 2, Fig. S5). Finally, circulating levels of the bile acid precursor C4 were decreased in the elafibranor-treated groups, but not in the placebo group (Fig. S4 and Table S2). Other bile acids did not show statistically significant changes in elafibranor-treated groups vs. placebo, although there was a trend toward decreased secondary bile acid levels, with no apparent effect on primary bile acid levels (Table S2). Circulating levels of FGF19 decreased in all groups without any significant difference between elafibranor-treated groups and placebo (Table S2).
      Table 2Changes in standard laboratory values and exploratory biomarkers.
      Baseline mean (SD)End-of-treatment mean (SD)Absolute change mean (SD)Treatment effect vs. placebo LSmean [95% CI]
      The treatment effects of each dose compared to placebo (LSmeans [95% CI]) were estimated and the corresponding p values provided using an ANCOVA model with baseline value as a covariate.
      p value vs. placebo
      Ela-80 mgEla-120 mgPlaceboEla-80 mgEla-120 mgPlaceboEla-80 mgEla-120 mgPlaceboEla-80 mgEla-120 mgEla-80 mgEla-120 mg
      GGT, IU/L282.3 (215.7)158.5 (144.6)229.6 (115.9)190.8 (171.2)96.6 (90.8)230.2 (125.3)-91.5 (95.3)-61.9 (70.8)0.6 (54.4)-77.7[-123.4;-31.2]-82.0 [-128.8;-35.1]0.0010.001
      ALT, IU/L57.6 (24.7)40.9 (17.5)48.5 (22.3)57.1 (60.0)48.1 (30.4)47.3 (21.9)-0.5 (57.4)7.3 (29.1)-1.2 (8.6)2.6 [-25.7;31.0]6.8 [-21.9;35.5]0.8510.634
      AST, IU/L54.3 (18.7)40.1 (15.6)46.7 (15.75)60.3 (61.5)51.2 (27.2)42.4 (12.9)6.0 (55.3)11.1 (28.00)-4.3 (8.0)10.5[-17.1;38.1]15.3 [-12.6;43.3]0.4460.274
      Bilirubin, umol/L9.9 (6.7)9.7 (5.3)11.1 (4.4)9.7 (6.3)9.2 (5.1)11.1 (5.7)-0.2 (3.4)-0.5 (2.8)-0.0 (3.6)-0.4 [-2.8;2.0]-0.7 [-3.1;1.8]0.7390.569
      Direct bilirubin, umol/L5.5 (5.0)4.0 (1.2)5.2 (2.1)5.8 (5.3)4.0 (1.5)5.6 (2.8)0.3 (2.2)-0.1 (0.6)0.5 (1.5)-0.1 [-1.3;1.1]-0.5 [-1.8;0.7]0.8480.409
      Albumin, g/L41.0 (3.3)41.1 (2.7)42.5 (2.9)43.2 (3.6)43.4 (3.6)42.5 (1.8)2.2 (2.5)2.3 (2.7)0.0 (2.2)1.7 [0.0;3.5]1.8 [0.1;3.6]0.0540.044
      5’-Nucleotidase, IU/L21.2 (16.3)14.2 (21.1)14.4 (11.6)13.4 (12.9)9.6 (11.7)13.9 (12.3)-7.8 (8.3)-4.6 (13.1)-0.5 (3.5)-4.8 [-9.8;0.2]-4.2 [-9.2;0.8]0.0580.097
      Creatinine, μmol/L66.61 (10.21)65.38 (11.93)66.90 (13.56)68.0 (13.8)72.9 (13.2)66.4 (12.6)1.4 (8.8)7.6 (6.6)-0.5 (5.4)1.9 [-3.4;7.1]8.0 [2.6;13.3]0.4700.005
      Cystatin C, mg/L0.97 (0.20)0.95 (0.28)1.03 (0.25)1.01 (0.23)0.97 (0.23)1.09 (0.26)0.04 (0.12)0.02 (0.12)0.06 (0.13)-0.0 [-0.1;0.1]-0.0 [-0.1;0.0]0.5790.337
      Creatinine kinase, IU/L50.7 (226.2)77.9 (40.6)109.9 (84.9)51.0 (24.1)82.2 (26.1)88.0 (59.8)0.3 (15.3)4.4 (29.2)-21.9 (65.7)-9.1[-34.0;15.8]9.3 [-14.5;33.0]0.4640.433
      Cholesterol, mmol/L5.9 (1.5)6.0 (1.6)5.5 (1.0)5.4 (1.3)5.6 (1.5)5.6 (1.0)-0.5 (0.7)-0.4 (0.6)0.0 (0.4)-0.4[-0.8;-0.0]-0.3 [-0.8;0.1]0.0400.099
      HDL-C, mmol/L1.9 (0.5)2.0 (0.6)2.0 (0.7)1.9 (0.5)2.0 (0.6)2.0 (0.7)-0.2 (0.4)0.1 (0.3)-0.0 (0.3)0.0 [-0.3;0.2]0.1 [-0.2;0.3]0.8300.635
      LDL-C, mmol/L3.4 (1.2)3.4 (1.3)2.9 (0.9)3.0 (1.0)3.1 (1.2)3.0 (0.7)-0.4 (0.6)-0.3 (0.5)0.1 (0.3)-0.3 [-0.6; 0.0]-0.3 [-0.6;0.0]0.0440.086
      Triglycerides, mmol/L1.22 (0.43)1.22 (0.36)1.31 (0.56)1.06 (0.38)0.96 (0.28)1.29 (0.65)-0.16 (0.35)-0.25 (0.21)-0.02 (0.38)-0.20 [-0.40;0.07]-0.30 [-0.5;-0.02]0.1720.032
      IgM, g/L2.95 (1.0)3.50 (2.16)4.60 (2.70)2.61 (0.86)3.03 (1.91)4.53 (2.45)-0.34 (0.58)-0.47 (0.55)-0.08 (0.72)-0.50 [-0.90;-0.07]-0.60 [-1.00;-0.13]0.0240.012
      hsCRP, mg/L
      Geometric means [95%CI] at baseline and at the end-of-treatment period. The treatment effects of each dose compared to placebo (Geometric mean ratio [95% CI]) were estimated after back transformation and the corresponding p values provided using an ANCOVA model on the log-transformed hsCRP value at the end-of-treatment period with log-baseline hsCRP as a covariate.
      5.95 [4.16;8.49]4.20 [2.33;7.58]3.98 [2.60;6.12]2.74 [1.81;4.14]2.84 [1.68;4.78]4.01 [2.52;6.37]---0.49 [0.33;0.72]0.68 [0.46;1.00]<0.0010.049
      Haptoglobin, g/L1.45 (0.64)1.29 (0.44)1.15 (0.55)1.19 (0.42)1.04 (0.46)1.17 (0.60)-0.27 (0.43)-0.25 (0.11)0.03 (0.22)-0.20 [-0.40;-0.00]-0.20 [-0.40;-0.00]0.0310.017
      Fibrinogen, g/L4.86 (1.00)4.94 (1.02)4.27 (0.69)4.00 (1.00)4.48 (1.00)4.21 (1.04)-0.87 (0.95)-0.47 (0.60)-0.07 (1.09)-0.6 [-1.2;0.1]-0.1 [-0.8;0.5]0.0940.679
      ALT, alanine aminotransferase; AST, aspartate aminotransferase; Ela, elafibranor; GGT, gamma-glutamyltransferase; hsCRP, high-sensitivity C-reactive protein.
      The treatment effects of each dose compared to placebo (LSmeans [95% CI]) were estimated and the corresponding p values provided using an ANCOVA model with baseline value as a covariate.
      ∗∗ Geometric means [95%CI] at baseline and at the end-of-treatment period. The treatment effects of each dose compared to placebo (Geometric mean ratio [95% CI]) were estimated after back transformation and the corresponding p values provided using an ANCOVA model on the log-transformed hsCRP value at the end-of-treatment period with log-baseline hsCRP as a covariate.
      Figure thumbnail gr3
      Fig. 3Effects of elafibranor on serum levels of inflammation markers GGT and 5'-nucleotidase in the mITT population.
      (A,B) Treatment effect (±95% CI) of elafibranor 80 mg and 120 vs. placebo on absolute changes in GGT and 5'-nucleotidase serum levels from day 0 to the end-of-treatment period (Week 12); (C,D) Time course of median relative changes vs. baseline in GGT and 5'-nucleotidase serum levels in placebo and elafibranor-treated groups (median±IQR). ∗∗∗p <0.001 vs. placebo according to ANCOVA with baseline value as covariate. An end-of-study visit was planned following a protocol amendment implemented after the study start and was performed only in a subset of patients (n = 23) after an off-study drug period of 16 to 30 days. GGT, gamma-glutamyltransferase; mITT, modified intention-to-treat.

      Safety and tolerability.

      Elafibranor was generally well tolerated and there were no deaths during the study. Treatment-emergent adverse events (TEAEs) were experienced by 80.0%, 86.7%, and 80.0% of patients in the elafibranor 80 mg, elafibranor 120 mg, and placebo groups, respectively, while treatment-related TEAEs were experienced by 13.3%, 33.3%, and 6.7% of patients, respectively. Severe TEAEs were experienced by 13.3%, 13.3%, and 13.3% of patients in the elafibranor 80 mg, elafibranor 120 mg, and placebo groups, respectively (Table 3, Table S1). No serious adverse events occurred in the placebo and elafibranor 80 mg treatment arm, while 2 patients in the elafibranor 120 mg group experienced at least 1 of severe intensity. One patient suffered an ischaemic stroke 24 hours after first intake of elafibranor 120 mg, which led to permanent withdrawal of the study drug. One patient in the 80 mg group with baseline values above 2x ULN showed increased aminotransferase activities (3x baseline) at the end-of-treatment visit. One patient in the 120 mg group experienced elevated aminotranferases at the last on-treatment visit possibly due to a suspected flare of autoimmune hepatitis (AIH). A previous history of another autoimmune disease (myasthenia gravis), in addition to PBC, was documented. Treatment was stopped as the patient had reached the end of the study and subsequent liver histology revealed portal inflammation expanding from the portal tract into the hepatic lobule, but no hepatocellular necrosis. The patient received high doses of steroids leading to ALT and bilirubin decrease. This episode was considered by the investigator as suspected AIH. Indeed, in patients with poly-auto-immune diseases, AIH consecutive to PBC or AIH-PBC overlap syndrome are not uncommon, occurring in up to 2.5% and 14% of patients with PBC, respectively.
      European Association for Study of Liver
      EASL Clinical Practice Guidelines: the diagnosis and management of patients with primary biliary cholangitis.
      While in this case, concomitant medications could be considered as confounding factors, a causal relationship to study drug could not be excluded. Severe non-serious adverse events were all reported as not drug-related in the elafibranor groups. Drug-related non-serious adverse events reported in the elafibranor-treated groups were of mild or moderate intensity and consisted of gastrointestinal events (nausea, diarrhoea), fatigue and headache. There were no significant changes in laboratory safety parameters except for a mild and reversible increase in plasma creatinine that was observed in the elafibranor 120 mg group (mean change vs. baseline of +7.6 μmol/L vs. -0.5 μmol/L in placebo). There were no clinically relevant changes in any other safety parameters, electrocardiogram, nor vital signs.
      Table 3Adverse events, treatment-emergent adverse events and serious adverse events in the elafibranor- and placebo-treated groups.
      CategoryPlacebo, n = 15 n (%)[# AEs]Elafibranor 80 mg, n = 15 n (%) [# AEs]Elafibranor 120 mg, n = 15 n (%) [# AEs]Overall, n = 45 n (%)[# AEs]
      Patients with at least 1 AE12 (80.0) [28]13 (86.7) [46]13 (86.7) [51]38 (84.4) [125]
      Patients with at least 1 TEAE12 (80.0) [25]12 (80.0) [41]13 (86.7) [46]37 (82.2) [112]
      Patients with at least 1 treatment-related TEAE1 (6.7) [1]2 (13.3) [6](33.3) [5]8 (17.8) [12]
      Pruritus[1][1][1][3]
       Dizziness[1][1]
       Lab abnormalities[4][4]
       Autoimmune hepatitis[1][1]
       Nausea[3][3]
      Patients with at least 1 serious TEAE0 [0]0 [0]2 (13.3)[3]2 (4.4)[3]
      Patients with at least 1 severe TEAE2 (13.3) [2]2 (13.3) [3]2 (13.3) [5]6 (13.3) [10]
      Patients with at least 1 serious treatment-related TEAE0 [0]0 [0]1 (6.7) [1]1 (2.2) [1]
      Patients with at least 1 TEAE leading to permanent withdrawal of study drug0 [0]0 [0]1 (6.7) [2]1 (2.2) [2]
      Deaths0 [0]0 [0]0 [0]0 [0]
      AE, adverse event; TEAE, treatment-emergent adverse event.
      Two patients in the placebo group experienced episodes of severe pruritus, while none in the elafibranor-treated groups did. Elafibranor treatment did not induce or exacerbate pruritus. In contrast, a favourable trend was evidenced in exploratory analyses assessing proportion of patients according to evolution of VAS (Fig. 4). In addition, a reduction of the VAS score in the subgroup of patients that reported pruritus (VAS >0 mm) at baseline could be observed (Table 4, Fig. S5). A similar trend was observed in the pruritus domain of PBC-40 quality of life questionnaire with a median change from baseline of -25% and -21% in the 80 mg and 120 mg group, compared to placebo, which remained unchanged (data not shown). Other domains of PBC-40 quality of life did not show any trend.
      Figure thumbnail gr4
      Fig. 4Effects of elafibranor on evolution of pruritus vs. baseline in the mITT population.
      Proportion of patients with improved, stable, or worsened VAS in each treatment groups. mITT, modified intention-to-treat; VAS, visual analogue scale.
      Table 4Effects on visual analogue scale in patients with pruritus at baseline (VAS >0).
      Placebo (n = 10)Elafibranor 80 mg (n = 10)Elafibranor 120 mg (n = 10)
      Median % change vs. baseline-7-24-49
      IQR-50 ; 790-57 ; 2-65 ; 6
      Min – Max-100 ; 3233-100 ; 300-100 ; 18
      Results are expressed as % change vs. baseline VAS.

      Discussion

      This international, randomized, double-blind, multicentre, placebo-controlled phase II clinical trial evaluated the efficacy and safety of elafibranor, a novel PPARα and PPARδ agonist, for the treatment of non-cirrhotic patients with PBC and an incomplete response to UDCA. Compared to placebo, both elafibranor doses tested produced highly significant decreases of ALP, a key surrogate marker of long-term liver outcomes in this cholestatic disease, and concomitantly improved markers of underlying immunological and inflammatory processes. Notably GGT, hsCRP, as well as IgM, the hallmark immunoglobulin reflecting disease activity in PBC, decreased significantly. All those markers increased again towards baseline values after the treatment was stopped. This data was available in a subgroup of 23 patients following an amendment that was implemented after the study had started. Although this phase II study is of limited size, no major safety issues were identified. Elafibranor did not exacerbate pruritus, which is a common, debilitating and difficult to treat adverse symptom in PBC. Secondary markers of PBC and inflammation were significantly improved at the end of the 12-week treatment period in the current trial, including GGT, 5’-nucleotidase, hsCRP and IgM, further supporting the anti-cholestatic and anti-inflammatory effects of elafibranor. In patients with mild to moderately advanced disease and normal bilirubin level at baseline, favourable trends in bilirubin serum concentrations were observed (-0.2±3.4 mg/dl at 80 mg and -0.5 mg/dl at 120 mg vs. 0.0±3.6 mg/dl), an important prognostic signal for liver outcomes.
      The beneficial effects of elafibranor in this study are reflected by favourable changes of surrogate markers of disease activity and prognostic scores, including ALP
      • Kuiper E.M.
      • Hansen B.E.
      • de Vries R.A.
      • den Ouden-Muller J.W.
      • van Ditzhuijsen T.J.M.
      • Haagsma E.B.
      • et al.
      Improved prognosis of patients with primary biliary cirrhosis that have a biochemical response to ursodeoxycholic acid.
      and the GLOBE Score. This is in line with a direct hepatic activation of PPARα and PPARδ as observed in other studies exploring PPAR activation, including a meta-analysis of 6 small proof-of-concept studies,
      • Zhang Y.
      • Li S.
      • He L.
      • Wang F.
      • Chen K.
      • Li J.
      • et al.
      Combination therapy of fenofibrate and ursodeoxycholic acid in patients with primary biliary cirrhosis who respond incompletely to UDCA monotherapy: a meta-analysis.
      where the pure PPARα agonist fenofibrate consistently improved ALP levels and other markers indicative of potential benefit in patients treated with UDCA. Similarly, 50 and 200 mg doses of the pure PPARδ agonist seladelpar have been shown to reduce ALP levels by 53-63% and improve GGT and 5'-nucleotidase in patients with PBC with an inadequate response to UDCA.
      • Jones D.
      • Boudes P.F.
      • Swain M.G.
      • Bowlus C.L.
      • Galambos M.R.
      • Bacon B.R.
      • et al.
      Seladelpar (MBX-8025), a selective PPAR-delta agonist, in patients with primary biliary cholangitis with an inadequate response to ursodeoxycholic acid: a double-blind, randomised, placebo-controlled, phase 2, proof-of-concept study.
      However, the trial was prematurely discontinued. Finally, a placebo-controlled 24-month study (BEZURSO) including 100 patients with PBC with incomplete response to UDCA has shown that bezafibrate, a non-selective PPAR agonist acting predominantly on PPARα and additionally on PPARγ and PPARδ isoforms, produced strong (60%) rapid and sustained reduction of ALP levels.
      • Corpechot C.
      • Chazouilleres O.
      • Rousseau A.
      • Le Gruyer A.
      • Habersetzer F.
      • Mathurin P.
      • et al.
      A placebo-controlled trial of bezafibrate in primary biliary cholangitis.
      Most importantly, in UDCA-treated patients with ALP or AST ≥1.5xULN or abnormal total bilirubin at baseline, 31% of patients in the bezafibrate group and 0% in the placebo group reached the primary composite endpoint of normalization of ALP levels with normal levels of ALT, AST, total bilirubin and albumin and normal prothrombin index – defined as complete biochemical remission.
      • Carbone M.
      • Sharp S.J.
      • Flack S.
      • Paximadas D.
      • Spiess K.
      • Adgey C.
      • et al.
      The UK-PBC risk scores: derivation and validation of a scoring system for long-term prediction of end-stage liver disease in primary biliary cholangitis.
      Similarly, the response rates obtained using other ALP-related composite endpoints were significantly higher in the bezafibrate group vs. placebo.
      The current study is the first to show the therapeutic efficacy of elafibranor, a dual PPARα and PPARδ agonist, in patients with PBC and incomplete response to UDCA. Administration of elafibranor results in a rapid reduction of ALP that is sustained until the end of the treatment period. This effect was associated with significant improvements of other important PBC- and inflammation-related markers such as GGT, 5'-nucleotidase, IgM and hsCRP. The observed beneficial effects in this phase II trial of elafibranor on cholestatic liver injury are likely multifactorial. Firstly, PPARα and PPARδ activation affect bile acid metabolism and recycling at several levels. Activation of PPARα has been shown to downregulate the expression of CYP7A1, a key enzyme of the classical pathway which catalyses hydroxylation of cholesterol on position 7 and constitutes the rate-limiting step of the classical bile acid synthesis pathway.
      • Post S.M.
      • Duez H.
      • Gervois P.P.
      • Staels B.
      • Kuipers K.
      • Princen H.M.
      Fibrates suppress bile acid synthesis via peroxisome proliferator-activated receptor-alpha-mediated downregulation of cholesterol 7alpha-hydroxylase and sterol 27-hydroxylase expression.
      In line with this – and as previously shown with bezafibrate
      • Corpechot C.
      • Chazouilleres O.
      • Rousseau A.
      • Le Gruyer A.
      • Habersetzer F.
      • Mathurin P.
      • et al.
      A placebo-controlled trial of bezafibrate in primary biliary cholangitis.
      and seladelpar
      • Jones D.
      • Boudes P.F.
      • Swain M.G.
      • Bowlus C.L.
      • Galambos M.R.
      • Bacon B.R.
      • et al.
      Seladelpar (MBX-8025), a selective PPAR-delta agonist, in patients with primary biliary cholangitis with an inadequate response to ursodeoxycholic acid: a double-blind, randomised, placebo-controlled, phase 2, proof-of-concept study.
      – both doses of elafibranor induced significant reductions of circulating levels of C4, a bile acid intermediate. Unlike with FXR agonists, the C4-lowering effects of elafibranor are independent of increased levels of FGF19.
      • Zweers S.J.
      • de Vries E.M.
      • Lenicek M.
      • Tolenaars D.
      • de Waart D.R.
      • Koelfat K.V.K.
      • et al.
      Prolonged fibroblast growth factor 19 response in patients with primary sclerosing cholangitis after an oral chenodeoxycholic acid challenge.
      Thus no concerns related to FGF19-associated is growth factor produced and secreted by enterocytes might be associated with long-term malignancy risk are expected.
      • Uriarte I.
      • Latasa M.U.
      • Carotti S.
      • Fernandez-Barrena M.G.
      • Garcia-Irigoyen O.
      • Elizalde M.
      • et al.
      Ileal FGF15 contributes to fibrosis-associated hepatocellular carcinoma development.
      In addition, the PPARα and PPARδ agonist elafibranor might also exert favourable effects through increased bile acid conjugation, increased bile acid output, and formation of non-toxic bile acid micelles in the bile ducts.
      • Ghonem N.S.
      • Assis D.N.
      • Boyer J.L.
      Fibrates and cholestasis.
      ,
      • Li F.
      • Patterson A.D.
      • Krausz K.W.
      • Tanaka N.
      • Gonzalez F.J.
      Metabolomics reveals an essential role for peroxisome proliferator-activated receptor alpha in bile acid homeostasis.
      Finally and beyond effects on bile acid metabolism, simultaneous activation of PPARα and PPARδ has been directly linked to anti-inflammatory effects through inhibition of the AP1 and NF-κB pathway,
      • Pawlak M.
      • Bauge E.
      • Bourguet W.
      • De Bosscher K.
      • Lalloyer F.
      • Tailleux A.
      • et al.
      The transrepressive activity of peroxisome proliferator-activated receptor alpha is necessary and sufficient to prevent liver fibrosis in mice.
      and modulation of the BCL6-regulated inflammation pathway.
      • Lee C.H.
      • Chawla A.
      • Urbiztondo N.
      • Liao D.
      • Boisvert W.A.
      • Evans R.M.
      • et al.
      Transcriptional repression of atherogenic inflammation: modulation by PPARdelta.
      Currently, OCA is the only approved therapy for patients with PBC and an inadequate response or intolerance to UDCA. While no head-to-head comparison is available, in patients with mild to moderately advanced disease and ALP levels >1.5 ULN at baseline, the mean ALP decrease from baseline after 12-week treatment with 10 mg OCA was about 2-fold less than the decrease obtained with 80 mg and 120 mg elafibranor (24% for OCA 10 mg compared to 40-50% for elafibranor).
      • Hirschfield G.M.
      • Mason A.
      • Luketic V.
      • Lindor K.
      • Gordon S.C.
      • Mayo M.
      • et al.
      Efficacy of obeticholic acid in patients with primary biliary cirrhosis and inadequate response to ursodeoxycholic acid.
      In the pivotal POISE trial in patients with ALP level >1.67xULN at baseline, OCA 10 mg induced a 30-35% decrease in ALP levels after 12 months of treatment.
      • Nevens F.
      • Andreone P.
      • Mazzella G.
      • Strasser S.I.
      • Bowlus C.
      • Invernizzi P.
      • et al.
      A placebo-controlled trial of obeticholic acid in primary biliary cholangitis.
      The primary composite endpoint adopted by regulatory authorities as a surrogate of liver outcomes was achieved in 66.7% of patients receiving elafibranor 80 mg and 78.6% receiving 120 mg, compared to 6.7% of those receiving placebo (p <0.001) (while the POISE trial reported 47% at 10 mg OCA vs. 10% in placebo).
      This phase II trial with elafibranor reinforces the potential use of PPAR agonists as second-line therapy in patients with PBC insufficiently controlled by UDCA. Although a direct comparison with the BEZURSO trial is difficult due to differences in patients’ baseline characteristics, treatment duration and trial endpoints, it is notable that numerically similar median reductions of ALP level from baseline were observed here at 12 weeks. ALP decreased by a median of 60% in bezafibrate-treated patients,
      • Corpechot C.
      • Chazouilleres O.
      • Rousseau A.
      • Le Gruyer A.
      • Habersetzer F.
      • Mathurin P.
      • et al.
      A placebo-controlled trial of bezafibrate in primary biliary cholangitis.
      compared to 51% and 41% median reductions in the 80 mg elafibranor and 120 mg -elafibranor groups, respectively.
      In the current study, a higher absolute change vs. baseline in ALP and GGT was observed in the 80 mg group compared to the 120 mg group. In the absence of stratification for ALP baseline levels, this could be related to the inclusion of patients with higher disease activity in the 80 mg group, which was reflected by the observed higher baseline values of cholestasis markers. However, the statistical analysis (ANCOVA) took potential baseline differences into account by analysing baseline values as covariates. Overall, the relative decline in ALP, GGT and 5’nucleotidase and other markers were comparable in the 2 elafibranor-treated groups. Such a difference in disease activity at baseline might also account for the apparent differences in response rates on composite endpoints and on ALP normalization. Consequently, the current study suggests that the 80 mg dose of elafibranor is equipotent to the 120 mg dose. Importantly, the current study did not assess markers of fibrosis and this will be part of the planned phase III trial
      In the current study, with 15 patients per group treated for 12 weeks, the 2 doses of elafibranor were generally safe and well tolerated with no imbalance between groups in total occurrence of TEAEs. Drug-related non-serious adverse events reported in the elafibranor-treated groups were of mild or moderate intensity, and mainly consisted of gastrointestinal side effects – in particular nausea and diarrhoea – fatigue and headache. Two patients in the 120 mg group experienced at least 1 SAE of severe intensity. One patient suffered an ischaemic stroke following the first dose of elafibranor, which was not thought to be treatment-related. One patient experienced an increase in serum transaminases at week 12 of drug exposure suggestive of AIH based on accompanying typical histology and response to steroids. Features of AIH can overlap with PBC and it has been suggested that up to 10% of patients with PBC develop overlapping pathologies as part of the natural history of the disease.
      European Association for Study of Liver
      EASL Clinical Practice Guidelines: the diagnosis and management of patients with primary biliary cholangitis.
      In this case, the patient’s pre-existing history of myasthenia gravis suggests a complex autoimmune trait and a constellation that may predispose to other organ involvement. On the other hand, it is very challenging to differentiate disease activity from injury and vigilance related to the use of drugs in patients with chronic liver disease is therefore warranted.
      • Hoofnagle J.H.
      • Bjornsonn E.S.
      Drug induced liver injury-types and phenotypes.
      While this condition and/or other concomitant medications could be considered as confounding factors, a causal relationship to study drug could not be excluded. In a large phase IIb trial in patients with NASH, treatment with elafibranor at the same doses of 80 mg and 120 mg for 52 weeks was generally safe and well tolerated, and there was no indication of elafibranor toxicity at examination of end-of-treatment liver biopsies.
      • Ratziu V.
      • Harrison S.A.
      • Francque S.
      • Bedossa P.
      • Lehert P.
      • Serfaty L.
      • et al.
      Elafibranor, an agonist of the peroxisome proliferator-activated receptor-alpha and -delta, induces resolution of nonalcoholic steatohepatitis without fibrosis worsening.
      In line with the mechanism of action of elafibranor on PPARα and PPARδ, elafibranor-treated patients had reduced total cholesterol, LDL-cholesterol and triglycerides, while HDL-cholesterol remained stable. Although PBC hyperlipidaemia has not been associated with increased risks of cardiovascular events, the activity profile of elafibranor on plasma lipids is reassuring compared to FXR agonists, which have been shown to decrease HDL-cholesterol and increase LDL-C levels in patients with PBC as a class effect.
      • Nevens F.
      • Andreone P.
      • Mazzella G.
      • Strasser S.I.
      • Bowlus C.
      • Invernizzi P.
      • et al.
      A placebo-controlled trial of obeticholic acid in primary biliary cholangitis.
      ,
      • Papazyan R.
      • Liu X.
      • Liu J.
      • Dong B.
      • Plummer E.M.
      • Lewis 2nd, R.D.
      • et al.
      FXR activation by obeticholic acid or nonsteroidal agonists induces a human-like lipoprotein cholesterol change in mice with humanized chimeric liver.
      A slight but significant and reversible increase in plasma creatinine was observed in the elafibranor 120 mg group. This is a well-known effect of fibrates
      • Bonds D.E.
      • Craven T.E.
      • Buse J.
      • Crouse J.R.
      • Cuddihy R.
      • Elam M.
      • et al.
      Fenofibrate-associated changes in renal function and relationship to clinical outcomes among individuals with type 2 diabetes: the Action to Control Cardiovascular Risk in Diabetes (ACCORD) experience.
      which is not linked to a decreased glomerular filtration but rather to increased release of creatine from muscle.
      • Camerino G.M.
      • Pellegrino M.A.
      • Brocca L.
      • Digennaro C.
      • Camerino D.C.
      • Pierno S.
      • et al.
      Statin or fibrate chronic treatment modifies the proteomic profile of rat skeletal muscle.
      In support of the renal safety of elafibranor in this trial, the circulating levels of cystatin C, a more accurate marker of glomerular filtration,
      • Shlipak M.G.
      • Matsushita K.
      • Arnlov J.
      • Inker L.A.
      • Katz R.
      • Polkinghorne K.R.
      • et al.
      Consortium CKDP. Cystatin C versus creatinine in determining risk based on kidney function.
      was not affected by elafibranor treatment (data not shown), consistent with the findings in patients with NASH after 12 months of treatment.
      • Ratziu V.
      • Harrison S.A.
      • Francque S.
      • Bedossa P.
      • Lehert P.
      • Serfaty L.
      • et al.
      Elafibranor, an agonist of the peroxisome proliferator-activated receptor-alpha and -delta, induces resolution of nonalcoholic steatohepatitis without fibrosis worsening.
      There were no clinically relevant changes in any other safety parameters.
      Finally, pruritus is a frequently observed and difficult to treat symptom of cholestatic liver diseases and represents a heavy burden on the quality of life of patients with PBC. Elafibranor did not worsen nor exacerbate pruritus in patients with mild to moderate disease. In contrast, a favourable trend was observed in patients reporting pruritus at baseline, with both VAS and PBC-40 quality of life scores improved. It can be speculated that this is related to a beneficial effect of PPARs – as observed in the BEZURSO trial
      • Corpechot C.
      • Chazouilleres O.
      • Rousseau A.
      • Le Gruyer A.
      • Habersetzer F.
      • Mathurin P.
      • et al.
      A placebo-controlled trial of bezafibrate in primary biliary cholangitis.
      – but this should be formally addressed in a larger scale, longer duration study. As the pathophysiology of pruritus is complex, the exact mechanism through which PPARα and/or PPARδ activation ameliorate this symptom merits further investigation.
      In conclusion, this phase II trial of elafibranor demonstrated significant improvements in markers of liver injury in patients with PBC and an incomplete response or intolerance to UDCA, which support the conduct of longer duration clinical trials with larger number of patients to further assess the efficacy to safety ratio of elafibranor in this patient population. Taken together, this study supports a phase III trial of elafibranor at the 80 mg daily dose, with the aim of assessing its efficacy, safety and tolerability relative to the currently approved second-line therapy for patients with PBC.

      Abbreviations

      AIH, autoimmune hepatitis; ALP, alkaline phosphatase; ALT, alanine aminotransferase; AST, aspartate aminotransferase; C4, 7α-hydroxy-4-cholesten-3-one; FGF19, fibroblast growth factor-19; FXR, farnesoid X receptor; GGT, gamma-glutamyl transferase; hsCRP, high-sensitivity C-reactive protein; ITT, intention-to-treat; mITT, modified ITT; OCA, obeticholic acid; PBC, primary biliary cholangitis; PPAR, peroxisome proliferator-activated receptor; PSC, primary sclerosing cholangitis; TEAEs, treatment-emergent adverse events; UDCA, ursodeoxycholic acid; VAS, visual analogue scale.

      Financial support

      This study was funded by Genfit SA

      Authors’ contributions

      Study concept and design: SM, PB. Analysis and interpretation of data; JMS, AP, KVK, MAH, SC, DP, AB, GMH, CL, JV, DJ, BS, VL. Drafting of the manuscript: RH, PB, JMS. Critical revision of the manuscript for important intellectual content; JMS, AP, KVK, MAH, SC, DP, AB, GMH, CL, JV, DJ, BS, VL. Statistical analysis: DM

      Data availability statement

      Data is available from the Study Sponsor Genfit SA.

      Conflict of interest

      JMS reports consultancy: BMS, Boehringer Ingelheim, Echosens, Galmed, Genfit, Gilead Sciences, Intercept Pharmaceuticals, Madrigal, Nordic Bioscience, Novartis, Pfizer, Roche, Sanofi, Zydus. Research Funding: Gilead Sciences. AP has received grant funding, personal fees, and advisory board fees from Intercept Pharmaceuticals; advisory board fees and fees for teaching from Novartis; and personal fees from CymaBay Therapeutics and Inova Diagnostics. KVK: serves as consultant to or an advisory boards for Conatus, CymaBay, Gilead, Intercept, La Jolla, Merck and Novartis. He receives research support from Genfit, Gilead, High Tide, Intercept, NGM Biopharma and Novartis and serves as a speaker for Abbvie, Gilead Sciences and Intercept. MAH: Consultancy for Roche, Novartis, Falk and Intercept. SC received research support from Genfit, Gilead and Zydus. DP: nothing to disclose. AB: nothing to disclose. GMH has consulted for Intercept, Genfit, Novartis, GSK, Cymabay and Gilead. CL reports research grants: Gilead, Intercept, CymaBay, Genfit, Genkyotex, Enanta, GSK, Novartis, NGM, High Tide, Durect, Alnylam, Zydus, Cara Therapeutics, Target PharmaSolutions; Consulting fees/Advisory boards: CymaBay, GSK, Shire, Pliant, Target PharmaSolutions, Flashlight Therapeutics, Cara Therapeutics; Royalties: Up-to-date; other: Editorial board Liver Transplantation. JV Research Grants: Allergan, Arena, CymaBay, Enanta, Genkyotex, Intercept, Lilly, NGM Pharmaceuticals, Novartis, TaiwanJ, Scientific Advisor: Arena, BioIncept, Blade, CymaBay, Enanta, Genkyotex, Glaxo-Smith-Kline, Intercept, Lilly, Novartis, TaiwanJ, Authorship: Up-to-Date Immunosuppression in Liver Transplantation; AASLD Writing Committee AIH Guidance In Press, 2019. DJ reports consultancy and grant funding from Intercept and Consultancy from Novartis. AT has nothing to disclose. BS is consultant and president of the SAB of Genfit SA. SM was a former Genfit employee and has currently no COI. RH, DM and PB are Genfit employees. VL reports consultancy for Genfit.

      Supplementary data

      References

        • Selmi C.
        • Bowlus C.L.
        • Gershwin M.E.
        • Coppel R.L.
        Primary biliary cirrhosis.
        Lancet. 2011; 377: 1600-1609
        • Lindor K.D.
        • Gershwin M.E.
        • Poupon R.
        • Kaplan M.
        • Bergasa N.V.
        • Heathcote E.J.
        • et al.
        American association for study of liver diseases: primary biliary cirrhosis.
        Hepatology. 2009; 50: 291-308
        • Lammers W.J.
        • van Buuren H.R.
        • Hirschfield G.M.
        • Janssen H.L.A.
        • Invernizzi P.
        • Mason A.L.
        • et al.
        Levels of alkaline phosphatase and bilirubin are surrogate end points of outcomes of patients with primary biliary cirrhosis: an international follow-up study.
        Gastroenterology. 2014; 147 (1338-1349 e5; quiz e15)
        • Poupon R.E.
        • Lindor K.D.
        • Cauch-Dudek K.
        • Dickson E.R.
        • Poupon R.
        • Heathcote E.J.
        Combined analysis of randomized controlled trials of ursodeoxycholic acid in primary biliary cirrhosis.
        Gastroenterology. 1997; 113: 884-890
        • Kaps L.
        • Grambihler A.
        • Yemane B.
        • Nagel M.
        • Labaenz C.
        • Ploch P.
        • et al.
        Symptom burden and treatment response in patients with primary biliary cholangitis.
        Dig Dis Sci. 2019; https://doi.org/10.1007/s10620-019-06009-3
        • European Association for Study of Liver
        EASL Clinical Practice Guidelines: the diagnosis and management of patients with primary biliary cholangitis.
        J Hepatol. 2017; 67: 145-172
        • Carbone M.
        • Mells G.F.
        • Pells G.
        • Dawwas M.F.
        • Newton J.L.
        • Heneghan M.A.
        • et al.
        Sex and age are determinants of the clinical phenotype of primary biliary cirrhosis and response to ursodeoxycholic acid.
        Gastroenterology. 2013; 144 (560-569 e7; quiz e13-4)
        • Carbone M.
        • Sharp S.J.
        • Flack S.
        • Paximadas D.
        • Spiess K.
        • Adgey C.
        • et al.
        The UK-PBC risk scores: derivation and validation of a scoring system for long-term prediction of end-stage liver disease in primary biliary cholangitis.
        Hepatology. 2016; 63: 930-950
        • Lammers W.J.
        • Hirschfield G.M.
        • Corpechot C.
        • Nevens F.
        • Lindor K.D.
        • Janssen H.L.A.
        • et al.
        Development and validation of a scoring system to predict outcomes of patients with primary biliary cirrhosis receiving ursodeoxycholic acid therapy.
        Gastroenterology. 2015; 149 (1804-1812 e4)
        • Nevens F.
        • Andreone P.
        • Mazzella G.
        • Strasser S.I.
        • Bowlus C.
        • Invernizzi P.
        • et al.
        A placebo-controlled trial of obeticholic acid in primary biliary cholangitis.
        N Engl J Med. 2016; 375: 631-643
        • Corpechot C.
        • Chazouilleres O.
        • Rousseau A.
        • Le Gruyer A.
        • Habersetzer F.
        • Mathurin P.
        • et al.
        A placebo-controlled trial of bezafibrate in primary biliary cholangitis.
        N Engl J Med. 2018; 378: 2171-2181
        • Ghonem N.S.
        • Assis D.N.
        • Boyer J.L.
        Fibrates and cholestasis.
        Hepatology. 2015; 62: 635-643
        • Li F.
        • Patterson A.D.
        • Krausz K.W.
        • Tanaka N.
        • Gonzalez F.J.
        Metabolomics reveals an essential role for peroxisome proliferator-activated receptor alpha in bile acid homeostasis.
        J Lipid Res. 2012; 53: 1625-1635
        • Pawlak M.
        • Bauge E.
        • Bourguet W.
        • De Bosscher K.
        • Lalloyer F.
        • Tailleux A.
        • et al.
        The transrepressive activity of peroxisome proliferator-activated receptor alpha is necessary and sufficient to prevent liver fibrosis in mice.
        Hepatology. 2014; 60: 1593-1606
        • Lee C.H.
        • Chawla A.
        • Urbiztondo N.
        • Liao D.
        • Boisvert W.A.
        • Evans R.M.
        • et al.
        Transcriptional repression of atherogenic inflammation: modulation by PPARdelta.
        Science. 2003; 302: 453-457
        • Reig A.
        • Sesé P.
        • Parés A.
        Effects of bezafibrate on outcome and pruritus in primary biliary cholangitis with suboptimal ursodeoxycholic acid response.
        Am J Gastroenterol. 2018 Jan; 113: 49-55
        • Staels B.
        • Rubenstrunk A.
        • Noel B.
        • Rigou G.
        • Delataille P.
        • Millatt L.J.
        • et al.
        Hepatoprotective effects of the dual peroxisome proliferator-activated receptor alpha/delta agonist, GFT505, in rodent models of nonalcoholic fatty liver disease/nonalcoholic steatohepatitis.
        Hepatology. 2013; 58: 1941-1952
        • Jacoby A.
        • Rannard A.
        • Buck D.
        • Bhala N.
        • Newton J.L.
        • James O.F.W.
        • et al.
        Development, validation, and evaluation of the PBC-40, a disease specific health related quality of life measure for primary biliary cirrhosis.
        Gut. 2005; 54: 1622-1629
        • Elman S.
        • Hynan L.S.
        • Gabriel V.
        • Mayo M.J.
        The 5-D itch scale: a new measure of pruritus.
        Br J Dermatol. 2010; 162: 587-593
        • Hirschfield G.M.
        • Mason A.
        • Luketic V.
        • Lindor K.
        • Gordon S.C.
        • Mayo M.
        • et al.
        Efficacy of obeticholic acid in patients with primary biliary cirrhosis and inadequate response to ursodeoxycholic acid.
        Gastroenterology. 2015; 148: 751-761 e8
        • Kuiper E.M.
        • Hansen B.E.
        • de Vries R.A.
        • den Ouden-Muller J.W.
        • van Ditzhuijsen T.J.M.
        • Haagsma E.B.
        • et al.
        Improved prognosis of patients with primary biliary cirrhosis that have a biochemical response to ursodeoxycholic acid.
        Gastroenterology. 2009; 136: 1281-1287
        • Zhang Y.
        • Li S.
        • He L.
        • Wang F.
        • Chen K.
        • Li J.
        • et al.
        Combination therapy of fenofibrate and ursodeoxycholic acid in patients with primary biliary cirrhosis who respond incompletely to UDCA monotherapy: a meta-analysis.
        Drug Des Devel Ther. 2015; 9: 2757-2766
        • Jones D.
        • Boudes P.F.
        • Swain M.G.
        • Bowlus C.L.
        • Galambos M.R.
        • Bacon B.R.
        • et al.
        Seladelpar (MBX-8025), a selective PPAR-delta agonist, in patients with primary biliary cholangitis with an inadequate response to ursodeoxycholic acid: a double-blind, randomised, placebo-controlled, phase 2, proof-of-concept study.
        Lancet Gastroenterol Hepatol. 2017; 2: 716-726
        • Post S.M.
        • Duez H.
        • Gervois P.P.
        • Staels B.
        • Kuipers K.
        • Princen H.M.
        Fibrates suppress bile acid synthesis via peroxisome proliferator-activated receptor-alpha-mediated downregulation of cholesterol 7alpha-hydroxylase and sterol 27-hydroxylase expression.
        Arterioscler Thromb Vasc Biol. 2001; 21: 1840-1845
        • Zweers S.J.
        • de Vries E.M.
        • Lenicek M.
        • Tolenaars D.
        • de Waart D.R.
        • Koelfat K.V.K.
        • et al.
        Prolonged fibroblast growth factor 19 response in patients with primary sclerosing cholangitis after an oral chenodeoxycholic acid challenge.
        Hepatol Int. 2017; 11: 132-140
        • Uriarte I.
        • Latasa M.U.
        • Carotti S.
        • Fernandez-Barrena M.G.
        • Garcia-Irigoyen O.
        • Elizalde M.
        • et al.
        Ileal FGF15 contributes to fibrosis-associated hepatocellular carcinoma development.
        Int J Canc. 2015; 136: 2469-2475
        • Hoofnagle J.H.
        • Bjornsonn E.S.
        Drug induced liver injury-types and phenotypes.
        N Eng J Med. 2019; 31: 264-273
        • Ratziu V.
        • Harrison S.A.
        • Francque S.
        • Bedossa P.
        • Lehert P.
        • Serfaty L.
        • et al.
        Elafibranor, an agonist of the peroxisome proliferator-activated receptor-alpha and -delta, induces resolution of nonalcoholic steatohepatitis without fibrosis worsening.
        Gastroenterology. 2016; 150: 1147-1159 e5
        • Papazyan R.
        • Liu X.
        • Liu J.
        • Dong B.
        • Plummer E.M.
        • Lewis 2nd, R.D.
        • et al.
        FXR activation by obeticholic acid or nonsteroidal agonists induces a human-like lipoprotein cholesterol change in mice with humanized chimeric liver.
        J Lipid Res. 2018; 59: 982-993
        • Bonds D.E.
        • Craven T.E.
        • Buse J.
        • Crouse J.R.
        • Cuddihy R.
        • Elam M.
        • et al.
        Fenofibrate-associated changes in renal function and relationship to clinical outcomes among individuals with type 2 diabetes: the Action to Control Cardiovascular Risk in Diabetes (ACCORD) experience.
        Diabetologia. 2012; 55: 1641-1650
        • Camerino G.M.
        • Pellegrino M.A.
        • Brocca L.
        • Digennaro C.
        • Camerino D.C.
        • Pierno S.
        • et al.
        Statin or fibrate chronic treatment modifies the proteomic profile of rat skeletal muscle.
        Biochem Pharmacol. 2011; 81: 1054-1064
        • Shlipak M.G.
        • Matsushita K.
        • Arnlov J.
        • Inker L.A.
        • Katz R.
        • Polkinghorne K.R.
        • et al.
        Consortium CKDP. Cystatin C versus creatinine in determining risk based on kidney function.
        N Engl J Med. 2013; 369: 932-943