A randomized controlled trial of high-dose ursodesoxycholic acid for nonalcoholic steatohepatitis Vlad Ratziu 1,⇑ , Victor de Ledinghen 2,, Fréderic Oberti 3,, Philippe Mathurin 4,, Claire Wartelle-Bladou 5,, Christophe Renou 6 , Philippe Sogni 7 , Marianne Maynard 8 , Dominique Larrey 9 , Lawrence Serfaty 10 , Dominique Bonnefont-Rousselot 11 , Jean-Philippe Bastard 12 , Marc Rivière 13 , Jean Spénard 13,14 on behalf of the FRESGUN 1 Université Pierre et Marie Curie, Assistance Publique – Hôpitaux de Paris, Hôpital Pitié Salpêtrière, Inserm UMR_S 938, Paris, France; 2 Hôpital du Haut Leveque, Pessac Bersol, France; 3 Centre Hospitalier Universitaire d’Angers, Angers, France; 4 Hôpital Claude Huriez, Lille, France; 5 CHG du Pays d’Aix, Aix en Provence, France; 6 CHG d’Hyères, Hyères, France; 7 Hôpital Cochin, Paris, France; 8 Hôpital Hotel-Dieu, Lyon, France; 9 CHUMontpellier, Montpellier, France; 10 Hôpital Saint-Antoine, Paris, France; 11 Service de Biochimie Métabolique Hôpital Pitié Salpêtrière, France, and EA 4466, Faculté de pharmacie, Université Paris Descartes, France; 12 Hôpital Tenon, Paris, France; 13 Axcan Pharma, Inc. Quebec, Canada; 14 Université de Montréal, Quebec, Canada Back groun d & Aims: Nona lcoholic steat ohep atiti s (NAS H) is a prevalent liver disease associated with increased morbidity and mortality. Ursodeoxycholic acid (UDCA) may have antioxidant, anti-inflammatory, and antifibrotic properties and may reduce liver injury in NASH. To date, no studies have assessed the effi- cacy and safety of high-dose UDCA (HD-UDCA) in patients with NASH. Methods: We conducted a 12-month, randomize d, double-blind, placebo-controlled multicenter trial to evaluate the efficacy and safety of HD-UDCA (28–35 mg/kg per day) in 126 patients with biopsy-proven NASH and eleva ted alan ine aminotransferase (ALT) levels. The primary study end point was reduction in ALT levels from baseline in patients treated with HD-UDCA compared with placebo. Secondary study end points were the proportion ofpatients with ALT normalization, relative reduction in the scores of serum markers of fibrosis and hepatic inflammation, and safety and tolerability. Results: HD-UDCA significa ntly reduced mean ALT levels À28.3% from baseline after 12 months compared with À1.6% with pla- cebo ( p <0.001). At the end of the trial, ALT levels normalized (635 IU/L) in 24.5% of patients treated with HD-UDCA and in 4.8% of patients who received placebo ( p = 0.003). Both results were not accounted for by changes in weight during the trial. HD-UDCA sign ifican tly reduc ed the Fibro Test Ò seru m fibros is marker ( p <0.001) compared with placebo. HD-UDCA also signif- icantly improved markers of glycemic control and insulin resis- tance. There were no safety issues in this population. Conclusions: Tre atment with HD-UDC A was saf e, imp roved aminotransferase levels, serum fibrosis mark ers, and sele cted meta bolic paramet ers. Studies with hist ologic end point s are warranted. Ó 2011 Publis hed by Else vier B.V. on behalf of the European Association for the Study of the Liver. Introduction Nonalcoholic fatty liver disease is rapidly emerging as the most prevalent hepatic disorder in the Western world [13], where it has become the most frequent cause of newly diagnosed cases of chronic liver disease [58]. This increase in prevalence is mostly attributable to the worldwide epidemics of obesity and diabetes mellitus. Nonalcoholic steatohepatitis (NASH), which is part ofthe spectrum of nonalcoholic fatty liver disease, is a progressive condition that, in a minority of patients, can progress to cirrhosis, end-stage liver disease, and hepatocellular carcinoma [7,46]. Sev- eral long-term observational studies have shown that NASH sig- nificantly reduces overall survival in affected patients compared with age- and gender-matched individuals in the general popula- tion and incre ases live r-rel ated morta lity 10-f old [1,12,19,53] . Currently, there are no approved medications for the treatment of NASH, and the effectiveness of nonpharmacologic measures in this patient population is insufficient [20]. Thus, there is an unmet medical need for pharmacologic agents that are able to halt or reverse disease progression. Urso deoxy choli c acid (UDCA) is a natu ral, hydrophi lic bile acid that norma lly const itut es $3% of t he human biliar y pool [37]. The pharmacologic attributes, immunomodulatory functions, and direct antiapoptotic properties of UDCA may interfere with the progr essi on of nona lcoho lic fatty live r dise ase/ NASH [6,24,29, 51,52]. UDCA reduces the mitochondrial membrane permeability Journal of Hepatology 2011 vol. xxx j xxx–xxx Keywords: Nonalcoholic fatty liver disease; Nonalcoholic steatohepatitis; Ami- notransferases; Steatosis; Fibrosis; Randomized clinical trials; Insulin resistance. ⇑ Corresponding author. Fax: +33 142161049. E-mail address: [email protected](V. Ratziu). VDL and FO contribu ted equally in the study. PM and CWB contributed equally in the study. Members of the FRESGUN (French Study Group for Urso in NASH) are listed in the Acknowledgements section. Abbreviations: ALT, alanine aminotransferase; AST, aspartate aminotransferase; BMI, body mass index ; HBA1 c, glyco sylat ed hemo glob in; HCC, hepatoc ellula r carcinoma; ITT, intent-to-treat; NAFLD, nonalcoholic fatty liver disease; NASH, nonalcoholic steatohepatitis; PP, per-protocol; HD-UDCA, high-dose ursodeoxy- cholic acid. Research Article Please cite this article in press as: Ratziu V et al. A randomized controlled trial of high-dose ursodesoxycholic acid for nonalcoholic steatohepatitis. J Hepatol (2011), doi:10.1016/j.jhep.2010.08.030
10
Embed
A Randomized Controlled Trial of High-dose Ursodesoxycholic Acid
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
8/3/2019 A Randomized Controlled Trial of High-dose Ursodesoxycholic Acid
A randomized controlled trial of high-dose ursodesoxycholic acidfor nonalcoholic steatohepatitis
Vlad Ratziu1,⇑, Victor de Ledinghen2,, Fréderic Oberti3,, Philippe Mathurin4,,Claire Wartelle-Bladou5,, Christophe Renou6, Philippe Sogni7, Marianne Maynard8,
Dominique Larrey9, Lawrence Serf aty10, Dominique Bonnefont-Rousselot11, Jean-Philippe Bastard12,Marc Rivière13, Jean Spénard13,14 on behalf of the FRESGUN
1Université Pierre et Marie Curie, Assistance Publique – Hôpitaux de Paris, Hôpital Pitié Salpêtrière, Inserm UMR_S 938, Paris, France; 2Hôpitaldu Haut Leveque, Pessac Bersol, France; 3Centre Hospitalier Universitaire d’Angers, Angers, France; 4Hôpital Claude Huriez, Lille, France; 5CHG du
Pays d’Aix, Aix en Provence, France; 6CHG d’Hyères, Hyères, France; 7Hôpital Cochin, Paris, France; 8Hôpital Hotel-Dieu, Lyon, France; 9CHU Montpellier, Montpellier, France; 10Hôpital Saint-Antoine, Paris, France; 11Service de Biochimie Métabolique Hôpital Pitié Salpêtrière, France, and
EA 4466, Faculté de pharmacie, Université Paris Descartes, France; 12Hôpital Tenon, Paris, France; 13 Axcan Pharma, Inc. Quebec, Canada;14Université de Montréal, Quebec, Canada
Background & Aims: Nonalcoholic steatohepatitis (NASH) is aprevalent liver disease associated with increased morbidity and
mortality. Ursodeoxycholic acid (UDCA) may have antioxidant,
anti-inflammatory, and antifibrotic properties and may reduce
liver injury in NASH. To date, no studies have assessed the effi-
cacy and safety of high-dose UDCA (HD-UDCA) in patients with
NASH.
Methods: We conducted a 12-month, randomized, double-blind,
placebo-controlled multicenter trial to evaluate the efficacy and
safety of HD-UDCA (28–35 mg/kg per day) in 126 patients with
biopsy-proven NASH and elevated alanine aminotransferase
(ALT) levels. The primary study end point was reduction in ALT
levels from baseline in patients treated with HD-UDCA compared
with placebo. Secondary study end points were the proportion of patients with ALT normalization, relative reduction in the scores
of serum markers of fibrosis and hepatic inflammation, and safetyand tolerability.
Results: HD-UDCA significantly reduced mean ALT levelsÀ28.3%
from baseline after 12 months compared with À1.6% with pla-
cebo ( p <0.001). At the end of the trial, ALT levels normalized
(635 IU/L) in 24.5% of patients treated with HD-UDCA and in
4.8% of patients who received placebo ( p = 0.003). Both results
were not accounted for by changes in weight during the trial.
HD-UDCA significantly reduced the FibroTestÒ serum fibrosis
marker ( p <0.001) compared with placebo. HD-UDCA also signif-icantly improved markers of glycemic control and insulin resis-
tance. There were no safety issues in this population.
Conclusions: Treatment with HD-UDCA was safe, improved
aminotransferase levels, serum fibrosis markers, and selected
metabolic parameters. Studies with histologic end points are
warranted.
Ó 2011 Published by Elsevier B.V. on behalf of the European
Association for the Study of the Liver.
Introduction
Nonalcoholic fatty liver disease is rapidly emerging as the most
prevalent hepatic disorder in the Western world [13], where ithas become the most frequent cause of newly diagnosed cases
of chronic liver disease [58]. This increase in prevalence is mostly
attributable to the worldwide epidemics of obesity and diabetes
mellitus. Nonalcoholic steatohepatitis (NASH), which is part of
the spectrum of nonalcoholic fatty liver disease, is a progressive
condition that, in a minority of patients, can progress to cirrhosis,
end-stage liver disease, and hepatocellular carcinoma [7,46]. Sev-
eral long-term observational studies have shown that NASH sig-
nificantly reduces overall survival in affected patients compared
with age- and gender-matched individuals in the general popula-
tion and increases liver-related mortality 10-fold [1,12,19,53].
Currently, there are no approved medications for the treatment
of NASH, and the effectiveness of nonpharmacologic measuresin this patient population is insufficient [20]. Thus, there is an
unmet medical need for pharmacologic agents that are able tohalt or reverse disease progression.
Ursodeoxycholic acid (UDCA) is a natural, hydrophilic bile
acid that normally constitutes $3% of the human biliary pool [37].
The pharmacologic attributes, immunomodulatory functions, and
direct antiapoptotic properties of UDCA may interfere with the
progression of nonalcoholic fatty liver disease/NASH [6,24,29,
51,52]. UDCA reduces the mitochondrial membrane permeability
Please cite this article in press as: Ratziu V et al. A randomized controlled trial of high-dose ursodesoxycholic acid for nonalcoholic steatohepatitis. J
The sensitivity of this assay was 0.25 ng/ml and the inter- and intra-assay varia-
tions were <10%.
Statistical analysis
Clinically significant expected mean differences from baseline ALT, in the HD-
UDCA and placebo arms at the end of treatment, were estimated to be 50% and
20%, respectively. Assuming a = 0.05 and power = 0.91, the estimated sample size
for each study arm, according to the Yates correction method, was 54 patients.
Assuming a 10% withdrawal rate by treatment arm, the inclusion of 60 patients
per treatment arm was planned. Descriptive statistics were provided for each
treatment arm and for the entire population, according to the nature of the vari-
ables (i.e., qualitative or quantitative). Qualitative variables were presented as
actual values and percentages, while quantitative variables were presented as
actual, mean, standard deviation (SD), median, first and third quartiles, maxi-
mum, and minimum. Observations at screening/study entry and demographic
data were provided for the intent-to-treat and safety populations. Differences
from baseline within each treatment arm were calculated and compared for
HD-UDCA versus placebo using Fisher’s exact test (qualitative parameters or
parametric observations that did not meet the assumptions for parametric test-
ing) or Student’s t -test (quantitative parameters).
Results
Study participants
Fig. 1 shows the flow chart of patient participation in the trial.Among the 192 prescreened patients, the central pathologist
did not confirm the histological diagnosis of steatohepatitis in
26. Among the 146 screened patients, 19 no longer satisfied the
increased ALT criteria (>50 IU/L) on centralized measurement,
126 patients were randomized (62 in the HD-UDCA group and
64 in the placebo group) and 116 completed the trial. Baseline
patient characteristics are shown in Table 1. There were no signif-
icant differences between groups except for the proportion of
Research Article
Please cite this article in press as: Ratziu V et al. A randomized controlled trial of high-dose ursodesoxycholic acid for nonalcoholic steatohepatitis. J
Please cite this article in press as: Ratziu V et al. A randomized controlled trial of high-dose ursodesoxycholic acid for nonalcoholic steatohepatitis. J
Fig. 2. ALT levels during treatment. Results are expressed as means ± SD of fold
elevation over baseline values at each time point during treatment with HD-
UDCA (triangles) and placebo (circles). ALT, alanine aminotransferase; HD-UDCA,
high-dose ursodeoxycholic acid; M, month; SD, standard deviation.
Placebo
HD-UDCA
0
5
10
15
20
25
30
0 0
14.3
6.5
24.5
4.8
M0 M6 M12 M0 M6 M12
% o
f p a t i e n t s w i t h N A L T
Fig. 3. Proportion of patients with ALT normalization (<35 IU/L) at end of
treatment (M12). p <0.003 between groups. NALT, normalized alanine amino-
transferase; HD-UDCA, high-dose ursodeoxycholic acid; M, month.
Research Article
Please cite this article in press as: Ratziu V et al. A randomized controlled trial of high-dose ursodesoxycholic acid for nonalcoholic steatohepatitis. J
There was a moderate but not statistically significant ( p <0.09)
reduction in mean body weight (±SD) in patients treated with
HD-UDCA (À1.65 ± 3.82 kg) compared with placebo (À0.37 ±
4.14 kg), which amounted to a relative change from baseline
(±SD) of À1.75 ± 4.00% and À0.4 ± 4.0% in the HD-UDCA and pla-
cebo arms, respectively. In patients treated with HD-UDCA, the
change in weight did not correlate with changes in serum glucose
levels (r = 0.09; p = 0.54), serum insulin levels (r = 0.03; p = 0.83),
homeostatic model assessment scores (r = À0.08; p = 0.96), or
glycosylated hemoglobin levels (r = 0.11; p = 0.43), suggestingthat the metabolic effects associated with HD-UDCA were not
attributed to weight changes during treatment. Moreover, among
patients with normalized ALT levels at M12, weight loss tended
to be lower in the HD-UDCA arm (n = 13; mean relative change
from baseline ± SD, À3.02 ± 4.90%) than in the placebo arm
(n = 3; À9.26 ± 4.40%; p = 0.06). The same weight loss trend was
reported by patients without ALT normalization in the HD-UDCA
arm (n = 39; À1.34 ± 3.8%) compared with patients in the placebo
arm (n = 58; 0 ± 3.8%; p <0.08), suggesting that ALT normalization
was also not fully attributable to weight loss. There was a trend
toward a more profound reduction of serum adiponectin levelsfrom baseline to M12 in patients given placebo (À22%; 36
patients with paired M12 and M0 measurements; median
change, 0.78 [interquartile range, 0.58]; p = 0.12) than with HD-UDCA (À5%; 29 patients; median change, 0.95 [interquartile
range, 0.61]).
Predictors of response
Multivariate analyses showed that independent predictors of ALT
reduction were treatment with HD-UDCA ( p <0.0001), high base-
line ALT levels ( p = 0.0003), increased age ( p = 0.001), and male
gender ( p <0.02). When an alternate definition of treatment
response was used (i.e., >30% reduction in ALT levels at the end
of treatment compared with baseline), HD-UDCA was associated
with an odds-ratio of 6.2 (95% confidence interval, 2.6–14.8) for
meeting this end point. No histologic features (e.g., steatosis,
fibrosis, nonalcoholic fatty liver disease activity score) were asso-
ciated with ALT reduction.
Safety
There were no deaths, cirrhotic complications (including progres-
sion to liver failure), or reports of hepatocellular carcinoma in the
study. Two patients in the HD-UDCA arm discontinued the treat-
ment because of diarrhea and abdominal discomfort. There was
no difference in the incidence of severe adverse events (AEs) or
intensity of non-severe AEs between the two treatment arms.
Most AEs were emergent but mild. Three patients in the HD-
UDCA arm required dosage changes compared with none in the
placebo arm. The most commonly reported AEs were gastrointes-
tinal (Table 3). Diarrhea and abdominal discomfort occurred
more frequently in the HD-UDCA arm than in the placebo arm.
Discussion
HD-UDCA significantly reduced aminotransferase levels in
patients with histologically proven NASH in this randomized,double-blind placebo-controlled trial, and a quarter of the HD-
UDCA-treated patients normalized ALT during the course of ther-
apy. Although there was a modest, albeit not significant, decrease
in body weight in the HD-UDCA arm, this did not account for the
1.0
0.8
0.6
0.4
0.2
0.0
F
i b r o
T e s
t V a
l u e
% o f
f i b r o
T e s
t r e
d u c
t i o n
f
r o m
M 0
, m e a n
M0 M6 M12 M0 M6 M12
100
-100
75
-75
50
-50
25
-25
0
1.0
0.8
0.6
0.4
0.2
0.0
F i b r o
T e s
t V a
l u e
M0 M6 M12
1.0
0.8
0.6
0.4
0.2
0.0
F i b r o
T e s
t V a
l u e
M0 M6 M12
A B
C D
Fig. 4. Changes in FibroTest values during treatment. (A) Absolute mean
change. (B) Relative mean change. (C) Absolute mean change in patients with
baseline fibrosis. (D) Absolute mean change in patients without baseline fibrosis.
M, month.
PlaceboHD-UDCA
-60
-50
-40
-30
-20
-10
0
10
20
30
-28 (55)
-2 (35)-8 (59)
9 (37)
-51 (28)
19 (48)
ALTM12 vs M0
ASTM12 vs M0
GGTM12 vs M0
-44% vs. 0%Medianreduction
-19% vs. 0%
p <0.001 %
c h a
n g e ,
m e a n
( S D )
-62% vs. 15%
Fig. 5. Biochemical response (LFTs). Mean percent change at end of treatment
(M12) compared with baseline (M0) for ALT, AST, and GGT in the HD-UDCA arm
(black boxes) and placebo arm (white boxes). Negative numbers indicate a
reduction during the trial. ALT, alanine aminotransferase; AST, aspartate amino-
transferase; GGT, c-glutamyl transpeptidase; LFTs, liver function tests; M, month;
SD, standard deviation.
Table 2. Relative change in glycemic parameters and insulin resistance
surrogate markers at M12⁄.
-2.2
-2.3
-19
-20
Serum glucose level
HbA1c level
Insulin level
HOMA score
+3.9
+5.2
-0.2
+6
0.002
<0.001
0.038
<0.009
HD-UDCAParameter Placebo p
⁄Percent relative change between baseline median values and M12.
JOURNAL OF HEPATOLOGY
Please cite this article in press as: Ratziu V et al. A randomized controlled trial of high-dose ursodesoxycholic acid for nonalcoholic steatohepatitis. J
by eliciting survival signals through the stimulation of the MAPK
pathway [42], and pro-apoptotic diseases such as NASH [14]
could benefit from such antiapoptotic activity. A small pilot study
showed that UDCA reduces the number of apoptotic hepatocytes
by almost half [4]. UDCA could also improve hepatic inflamma-
tion by altering the metabolism of arachidonic acid [18]. Further,
effects such as membrane stabilization, changes in hepatic lipid
content [5], and immunomodulation may also result in cytopro-
tection from oxidative stress and cell death. Unfortunately, these
experimental data and the results from this study are not direct
evidence of a histologic benefit. While only well-conducted trialswith histologic end points can provide a formal demonstration of
the benefits of HD-UDCA, the data presented herein provide the
proof of principle on which future trials may be based.
Biochemical markers of liver fibrosis have been extensively
validated in patients with viral hepatitis including longitudinal
validation in therapeutic trials [38,40,41] where changes in Fibro-
Test scores correlated with changes in histologic stages of fibrosis
[39]. Specific validations in patients with NASH are available for
some of these markers [3,8,15,45]. Data from the current study
showed that patients treated with HD-UDCA (but not placebo)experienced a significant reduction in FibroTest scores. Arguably,
at least part of the FibroTest decrease could have been due to
reductions in GGT (one of the components of the FibroTest score),
although HD-UDCA reduced GGT levels by a median of 32% even
in patients with stable or increased FibroTest values. While this
suggests that the reduction in GGT did not fully account for thereduction in FibroTest scores, additional demonstration of the
potential antifibrotic effects of HD-UDCA is needed, using differ-
ent markers that are not based on GGT values. Serum hyaluronic
acid levels did not change during this trial. However, this marker
has insufficient diagnostic performance, in particular in the pre-
cirrhotic stage.
Data on the efficacy of UDCA in NASH are limited and contra-
dictory. In the largest randomized trial conducted to date, Lindor
et al. showed that treatment with UDCA over 2 years reduced ALTlevels to the same extent as placebo [27]. There was no histologic
benefit of treatment except for improvement in steatosis in 46%of UDCA-treated patients and 37% of patients receiving placebo.
Remarkably, the magnitude of the ALT reduction in patients trea-
ted with UDCA in that study was similar to data from the present
trial, although the high rate of ALT and steatotic improvement
reported in the placebo arm in the study by Lindor et al. stands
in sharp contrast to the results of the current and previous trials
[11,44], where these changes were minimal. It is unclear whether
methodologic issues or study conduct might explain the results
observed in the Lindor et al. trial. There was a high dropout rate
in the UDCA arm (only 56 of 80 patients had available end-of-
treatment data) and a quarter of randomized patients did not
complete the trial (half due to AEs, a percentage that is much
higher than in the present study). In a second randomized trial,
Dufour et al. did not show any significant benefit of UDCA over
placebo on ALT reduction, although the combination of UDCA
and vitamin E was beneficial [11]. While that study was under-
powered (17 or less patients per arm), patients in the UDCA
arm experienced a mean reduction of 35 IU/L even though base-line ALT levels were higher in the UDCA arm than in the placebo
arm. Finally, in an older, controlled, but not randomized study,Laurin et al. first suggested a beneficial effect of UDCA on ALT lev-
els and steatosis [25]. While these trials provide conflicting
results, they were all conducted with a standard dose of UDCA
(13 to 15 mg/kg). We hypothesized that a higher dose might
result in a more marked biologic effect based on results from tri-
als in diseases other than NASH. In patients with primary scleros-
ing cholangitis, bile enrichment with UDCA increases from 43%,
at doses of 10–13 mg/kg, to 58%, at doses of 26–32 mg/kg [49].
It remains to be demonstrated in vivo whether the other cytopro-
tective effects of UDCA suspected to be beneficial in NASH are
also dose dependent. Currently available data from trials of HD-
UDCA are conflicting. A benefit has been reported in some trials
in patients with primary sclerosing cholangitis [10,16,31] butnot in others [28]. Further, positive results have been reported
in patients with cystic fibrosis [9,56] but not with primary biliarycirrhosis [2]. Recently, Leuschner et al. presented the results of an
18-month randomized trial of higher doses of UDCA (23–28 mg/kg) without a beneficial histological effect compared with pla-
cebo [26]. However, the inclusion criteria were complex, mixing
both histological and clinical and biological features. The validity
of the histological features, as a case definition of NASH, was also
questionable because a non-validated composite score was used
Table 3. Emergent AEs occurring in >10% of patients.
24
17
16
15
10
5
3
Diarrhea, n
Abdominal cramps, n
Abdominal discomfort, n
Asthenia, n
Flatulence, n
Back pain, n
Pruritus, n
11
12
7
25
15
7
7
<0.01
0.29
<0.04
0.09
0.37
0.76
0.32
HD-UDCA(n = 62)
Parameter Placebo(n = 64)
p
Research Article
Please cite this article in press as: Ratziu V et al. A randomized controlled trial of high-dose ursodesoxycholic acid for nonalcoholic steatohepatitis. J
and steatosis was not a compulsory feature. Importantly, akin to
the Lindor trial, but in sharp contrast to the current and other tri-
als, UDCA was unable to reduce ALT levels significantly more
than placebo [43,44]. Therefore, besides non-standardized inclu-
sion criteria and an unusually long inclusion period (6 years), the
understanding of the differences in the course of the placebo arm
might be key to solving the discrepancies between different trial
results. The current trial adds to the data showing an efficacy of HD-UDCA, at least on biochemical parameters in NASH patients.
An unexpected finding from the current trial is the significant
improvement in some metabolic parameters, including glycemic
control, fasting insulin levels, and surrogate markers of insulin
resistance. The magnitude of this benefit could not be explained
by the minimal, non-significant weight loss experienced by
patients in the HD-UDCA arm. If this effect is confirmed, it mayoffer an explanation for the improvements in steatosis that were
reported in some UDCA trials. Interestingly, a much smaller trialdid not find any effect of UDCA alone on serum glucose, serum
insulin, or adiponectin [4], which directly challenges the results
obtained here. More data are necessary but the results from the
present trial open the field of the metabolic actions of UDCA to
further investigation. A biologic explanation for our results is
not readily available. UDCA reduces serum levels of tumor necro-
sis factor-a [32] a pro-oxidant cytokine that promotes insulin
resistance [17,55]. Standard-dose UDCA is not a potent modulator
of TGR5 [21,50], farnesoid X receptor-a [30,35], or the small hete-
rodimeric partner [57], which are all important regulators of lipid
and glucose metabolism. It seems improbable, although it has not
yet been disproved, that higher doses of UDCA may result in a sig-
nificant modulation of these nuclear/cellular receptors. On the
other hand, taurine-conjugated UDCA, acting as a chemical chap-
erone, alleviates the endoplasmic reticulum stress [59] that occurs
in obesity, improves insulin action in the liver, restores systemic
insulin sensitivity, and clears fatty liver in rodents [34]. Whether
HD-UDCA results in similar actions remains to be proven. Data
presented here certainly provide a rationale for a further explora-
tion of beneficial metabolic effects of HD-UDCA.An important finding of this study is the excellent safety pro-
file of HD-UDCA in patients with NASH. Only two patients had tostop taking the drug (due to digestive AEs) and no worsening of
liver function was reported. Previous studies testing high doses
of UDCA (20–32 mg/kg) did not show any new or severe AEs,
and the overall tolerance to HD-UDCA was similar to that of reg-
ular doses [2,10,16]. However, a recent trial of high-dose UDCA
did show an unanticipated and high rate of progression of liver
disease in patients with primary sclerosing cholangitis [28].
Moreover, in alcoholic cirrhosis with jaundice, the standard-dose
UDCA treatment increased the rate of cirrhosis complications
over placebo, although part of this deleterious effect may not
have been independent of the baseline severity of liver disease
[36]. These adverse outcomes were not observed in the current
study, where participants did not have jaundice, hepatic insuffi-ciency, decompensated cirrhosis, or obstructive biliary disease.
HD-UDCA appears safe for use in patients with NASH without
jaundice or decompensated cirrhosis.
In conclusion, this study demonstrates that HD-UDCA, 28–35 mg/kg per day, is safe and well-tolerated in patients with
NASH. Treatment with HD-UDCA resulted in a strong and sus-
tained reduction of ALT levels, suggesting that the numerous
hepatoprotective actions of this molecule, which were previously
documented in experimental/preclinical models, may result in an
improvement of hepatic cell injury and inflammation in people
with NASH. The potency of HD-UDCA in reducing biochemical
surrogates of liver fibrosis in NASH awaits confirmatory data with
different noninvasive markers. Unexpectedly, HD-UDCA resulted
in beneficial effects on glycemic control and insulin sensitivity
contingent to putative, although yet uncharacterized metabolic
actions and not mere weight loss. The benefits reported with
HD-UDCA in this trial support the initiation of larger studies withhistologic end points in patients with NASH.
Conflict of interest
This trial was sponsored and funded by Axcan Pharma S.A. V. Rat-
ziu is a consultant to Astellas Pharma, Axcan Pharma, Gilead Sci-
ences, Genentech, Intercept Pharmaceuticals, and sanofi-aventis.
Acknowledgments
The authors wish to thank the following practitioner and other
staff involved in the FRESGUN (French Study Group for Urso in
NASH):Si Nafa Si Ahmed (CHR La Source, Orléans), S. Ben Ali (Hôpital
Saint-Joseph, Marseille), R. Anty (Hôpital de l’Archet II, Nice), F.
Bailly (Hôpital HOTEL DIEU, Lyon), C. Bernhardt (Hôpital Pitié-
Salpêtrière, Paris), M. Bourlière (Hôpital Saint-Joseph, Mareseille),
J. Boursier (CHU d’Angers, Angers), J.P. Bronowicki (Hôpital Bra-
bois, Nancy), C. Bureau (Hôpital PURPAN, Toulouse), V. Canva
(Hôpital Huriez, Lille), X. Causse (CHR La Source, Orléans), F.
Charlotte (Hôpital Pitié-Salpêtrière, Paris), O. Chazouillères
(Hôpital Saint-Antoine, Paris), S. Darhancy (Hôpital Huriez, Lille),
Y. Deugnier (Hôpital Pontchaillou, Rennes), N. Dib (CHU d’Angers,
Angers), M. Doffoel (Hôpital Civil-Strasbourg), J. Foucher (Hôpital
du Haut-Levêque, Bordeaux), F. Habersetzer (Hôpital Civil, Stras-
bourg), F. Lainé (Hôpital Pontchaillou, Rennes), P. Lebray (Hôpital
Pitié-Salpêtrière, Paris), S. Métivier (Hôpital PURPAN, Toulouse),
J.B. Nousbaum (CHU La Cavale Blanche, Brest), V. Oulès (HôpitalSaint-Joseph, Marseille), J.M. Péron (Hôpital PURPAN, Toulouse),M. Picon-Coste (CHG du Pays d’Aix, Aix en Provence), P. Podevin
(Hôpital COCHIN, Paris), F. Tanné (CHU La Cavale Blanche, Brest),
A. Tran (Hôpital de l’Archet II; Nice).
The authors also wish to thank P. Colin and G. Chauvière for their
input in the design of this trial, M. Mouret for statistical expertise,
and J. Wert for editorial assistance in the production of this man-
uscript, which were sponsored by Axcan Pharma Inc.
References
[1] Adams LA, Lymp JF, St. Sauver J, Sanderson SO, Lindor KD, Feldstein A, et al.
The natural history of nonalcoholic fatty liver disease: a population-based
[2] Angulo P, Dickson ER, Therneau TM, Jorgensen RA, Smith C, DeSotel CK, et al.Comparison of three doses of ursodeoxycholic acid in the treatment of
primary biliary cirrhosis: a randomized trial. J Hepatol 1999;30 (5):830–
835.
[3] Angulo P, Hui JM, Marchesini G, Bugianesi E, George J, Farrell GC, et al. The
NAFLD fibrosis score: a noninvasive system that identifies liver fibrosis in
patients with NAFLD. Hepatology 2007;45 (4):846–854.
[4] Balmerm ML, Siegrist K, Zimmermann A, Dufour JF. Effects of ursodeoxy-
cholic acid in combination with vitamin E on adipokines and apoptosis in
patients with nonalcoholic steatohepatitis. Liver Int 2009;29:1184–1188.
[5] Bellentani S, Chao YC, Ferretti I, Panini R, Tiribelli C. Chronic administration
of ursodeoxycholic and tauroursodeoxycholic acid changes microsomal
JOURNAL OF HEPATOLOGY
Please cite this article in press as: Ratziu V et al. A randomized controlled trial of high-dose ursodesoxycholic acid for nonalcoholic steatohepatitis. J
et al. Diagnostic value of biochemical markers (FibroTest-FibroSURE) for the
prediction of liver fibrosis in patients with non-alcoholic fatty liver disease.
BMC Gastroenterol 2006;6:6.
[46] Ratziu V, Poynard T. Assessing the outcome of nonalcoholic steatohepatitis?
It’s time to get serious. Hepatology 2006;44 (4):802–805.
[47] Rodrigues CM, Fan G, Ma X, Kren BT, Steer CJ. A novel role for ursodeoxy-
cholic acid in inhibiting apoptosis by modulating mitochondrial membrane
perturbation. J Clin Invest 1998;101:2790–2799.
[48] Rodrigues CM, Fan G, Wong PY, Kren BT, Steer CJ. Ursodeoxycholic acid mayinhibit deoxycholic acid-induced apoptosis by modulating mitochondrial
transmembrane potential and reactive oxygen species production. Mol Med
1998;4 (3):165–178.
[49] Rost D, Rudolph G, Kloeters-Plachky P, Stiehl A. Effect of high-dose
ursodeoxycholic acid on its biliary enrichment in primary sclerosing
cholangitis. Hepatology 2004;40 (3):693–698.
[50] Sato H, Macchiarulo A, Thomas C, Gioiello A, Une M, Hofmann AF, et al. Novel
potent and selective bile acid derivatives as TGR5 agonists: biological
screening, structure–activity relationships, and molecular modeling studies.
J Med Chem 2008;51 (6):1831–1841.
Research Article
Please cite this article in press as: Ratziu V et al. A randomized controlled trial of high-dose ursodesoxycholic acid for nonalcoholic steatohepatitis. J
Please cite this article in press as: Ratziu V et al. A randomized controlled trial of high-dose ursodesoxycholic acid for nonalcoholic steatohepatitis. J