Journal of Hepatology 38 (2003) 419425

www.elsevier.com/locate/jhep

Anti-tumor necrosis factor-alpha monoclonal antibody therapy in

severe alcoholic hepatitis q
Herbert Tilg 1,,*, Rajiv Jalan 2,, Arthur Kaser 1, Nathan A. Davies 2, Felix A. Offner 3, Stephen J. Hodges 2,
Othmar Ludwiczek 1, Deborah Shawcross 2, Heinz Zoller 1, Akeel Alisa 4, Rajeshwar P. Mookerjee 2,
Ivo Graziadei 1, Christian Datz 5, Michael Trauner 6, Detlef Schuppan 7, Peter Obrist 8,
Wolfgang Vogel 1, Roger Williams 2
1

Department of Medicine, Division of Gastroenterology and Hepatology, University Hospital Innsbruck, Anichstrasse 35, 6020 Innsbruck, Austria
2
Institute of Hepatology, University College London, London, UK
3
Institute of Pathology, Academic Teaching Hospital, Feldkirch, Austria
4
Cromwell Hospital, London, UK
5
Department of Medicine, General Hospital Salzburg, Salzburg, Austria
6
Department of Medicine, Karl-Franzens-University, Graz, Austria
7
Medical Department I, University of Erlangen-Nurnberg, Erlangen, Germany
8
Department of Pathology, University Hospital Innsbruck, Anichstrasse 35, 6020 Innsbruck, Austria

See Editorial, pages 518520

Background/Aims: Severe alcoholic hepatitis (AH) is associated with high mortality. Tumor necrosis factor-alpha
(TNFa) has been demonstrated to play an important role in its pathophysiology.
Methods: Twelve patients with biopsy-confirmed AH and a Maddrey discriminant factor .32 were treated with a
single infusion of the anti-TNF monoclonal antibody Infliximab at a dose of 5 mg/kg body weight. Serial measurements
were made for various cytokines using specific enzyme-linked immunoassays (ELISA). In four patients, liver biopsy
samples were available pretreatment and on day 1 28 of therapy.
Results: Ten of the 12 patients are alive at a median of 15 (1220) months. Two patients died within 30 days from
septicemia. Serum bilirubin levels, Maddrey score, neutrophil count and C-reactive protein fell significantly within the
first month. There was an early, though not significant, decrease in plasma levels of proinflammatory cytokines
(interleukins (IL)-1b, IL-6, IL-8, interferon-gamma), whereas plasma levels of TNFa remained near the sensitivity
limit of the assay throughout the treatment course. While TNFa mRNA expression in the liver did not change,
expression of IL-8, a cytokine regulated mainly by TNFa, was almost absent on day 1 28.
Conclusions: Our data suggest that randomized controlled trials of anti-TNF antibody in severe AH are warranted.
q 2003 European Association for the Study of the Liver. Published by Elsevier Science B.V. All rights reserved.
Keywords: Cytokines; Tumor necrosis factor-alpha; Infliximab; Alcoholic hepatitis; Inflammation

1. Introduction
Received 25 July 2002; received in revised form 3 December 2002;
accepted 13 December 2002
q
The authors who have taken part in this study have not a relationship
with the manufacturers of the drugs involved either in the past or present
and did not receive funding from the manufacturers to carry out their
research.
* Corresponding author. Tel.: 143-512-504-3255; fax: 143-512-5043317.
E-mail address: herbert.tilg@uibk.ac.at (H. Tilg).

Dr Tilg and Dr Jalan share the principle authorship of this article.

Acute alcoholic hepatitis (AH) in its severe form is associated with high morbidity and mortality. It is often superimposed on cirrhosis and 1-month mortality rates of about
3040% have been reported among patients hospitalized
with acutely decompensated liver disease due to an AH
with the histological features of steatohepatitis [1]. The
intrahepatic infiltration of polymorphonuclear leukocytes
is commonly associated with peripheral leukocytosis and

0168-8278/03/$30.00 q 2003 European Association for the Study of the Liver. Published by Elsevier Science B.V. All rights reserved.
doi:10.1016/S0 168-8278(02)00 442-7

420

H. Tilg et al. / Journal of Hepatology 38 (2003) 419425

fever as well as signs of hepatic inflammation [2]. The

pathogenesis of this severe steatotic inflammation in the
liver remains unclear but various proinflammatory cytokines may be involved as proposed in a number of new
studies [35]. Among these cytokines, the proinflammatory
cytokine tumor necrosis factor-alpha (TNFa) has emerged
as a key factor in the inflammatory process. AH was one of
the first diseases shown to exhibit increased circulating
TNFa levels [68]. In addition, serum concentrations of
various TNF-inducible cytokines such as interleukins 1
and 8 (IL-1, IL-8) are increased in patients with AH
[9,10]. Furthermore, plasma levels of both TNFa and soluble TNF receptors have been correlated with endotoxemia
and stage of liver disease [11]. These findings, coupled with
evidence that long-term ingestion of alcohol increases
intestinal permeability and that patients with the highest
serum concentrations have the highest in-hospital mortality
[12] indicate that intestinally derived endotoxin and cytokines such as TNFa, have an important role in the pathogenesis of alcoholic steatohepatitis [12,13]. These human
data support a key role for TNFa in this disorder and are
further substantiated by data from animal experiments
[14,15]. Further evidence comes from studies in TNF receptor type-I deficient mice in which exposure to alcohol does
not induce steatohepatitis [15].
Infliximab, a chimeric mouse-human anti-TNF monoclonal antibody, has recently been shown to be an effective
therapy in various TNF-related diseases such as Crohns
disease or rheumatoid arthritis [16,17]. The international
pilot study reported here was initiated to assess the safety,
tolerability and clinical effects of Infliximab in severe AH.
2. Methods
2.1. Patients
This was a collaborative study between centers in Innsbruck and London.
Twelve patients (11 males, one female, 50.6 ^ 9.8 years; median ^ SD)
were recruited for the study after obtaining written and informed consent
from each subject with the approval of the local research ethics committees,
and in accordance with the Declaration of Helsinki 1951 of the World
Medical Association. Each patient had a history of severe alcohol abuse
in conjunction with clinical and laboratory stigmata of acute AH. The
patients had withdrawn from alcohol consumption immediately prior to
admission to the hospital in all cases. Inclusion criteria were: (i) increasing
bilirubin despite standard medical therapy; (ii) Maddrey discriminant
factor . 32 and (iii) histologically proven AH. Biopsy was performed
percutaneously (study center 1) in six patients, (in six patients, pretreatment
and in four on day 1 28; one patient died on day 1 21 and one refused
control biopsy) and by the transjugular route in the other six patients
pretreatment (study center 2). Exclusions criteria were: concurrent infection, gastrointestinal hemorrhage, severe anemia (hemoglobin ,8 mg/dl),
renal failure (creatinine .200 mmol/l) malignancy or pregnancy. Prior or
concomitant steroid therapy was not allowed in this study.

2.2. Therapy
The patients received 5 mg/kg body weight Infliximab (Schering-Plough,
USA; AESCA GmbH, Austria) as a single 2 h infusion. All our patients

were hospitalized and received standard treatment including treatment of

alcohol withdrawal with benzodiazepines, administration of fluid, calories,
vitamins and minerals and management of ascites. Standard diet contained
at least 25 kcal/kg/day administered either orally or through a nasogastric
tube. Patients were carefully observed for potential concomitant infection
and medical management was individualized according to each patients
condition and complication. Patients with low-protein (,10 g/l) ascites
received prophylactic norfloxacin. Laboratory assessment included
measurements of complete blood cell counts, serum bilirubin, electrolytes,
transaminases, albumin, and prothrombin time (PT) twice weekly during
hospital stay and then weekly until day 1 28. Thereafter, patients were seen
every 2 weeks until 1 year.

2.3. Cytokine assessment

Ethylenediaminetetraacetic acid (EDTA)-anticoagulated samples were
collected on the day of treatment (day 0) and thereafter on days 1 1, 13,
17, 114, 121 and 128. Samples were separated within 30 min after blood
drawing and then kept frozen at 2708C until analysis. All assays were
performed as a single batch using commercially available sets (BioSource
International, Nivelles, Belgium). Briefly, capture antibodies for TNFa,
interferon-gamma (IFNg), IL-1beta (IL-1b), IL-4, IL-6, IL-8, IL-10, IL12p40, IL-1 receptor antagonist (IL-1ra), TNF soluble receptors I and II
(TNFsRI, TNFsRII ) were coated at 1 mg/ml onto 96 well Maxisorb (Nunc)
enzyme-linked immunoassay (ELISA) plates. The plates were blocked with
bovine serum albumin (Fraction V, Sigma) before the addition of 100 ml
standard or sample together with biotin-conjugated detection antibody
(100 ml; 0.4 mg/ml). After incubation at room temperature for 2 h and
washing, streptavidin was added before incubation with substrate (0-phenylenediamine HCl). The reaction was stopped with 1.8 M sulfuric acid and
the optical density measured at 450 nm referenced against 630 nm. Individual standard curves were fitted using Excel and individual values extrapolated. The lower limit for the detection of TNF was 5 pg/ml. All the
samples were analyzed in a single batch and the intra-assay coefficient of
variation was 5.4%.

2.4. Hepatic histology

In four patients, liver biopsies were performed both before treatment with
Infliximab and 28 days afterwards (day 1 28). Biopsies were fixed in
buffered formalin, embedded in paraffin, cut into thin sections and stained
with hematoxylin and eosin. Histopathological examination was performed
by a single pathologist (F.A.O.), an expert in liver diseases and blinded on
patients treatment data. Histological features of AH (fatty changes,
ballooning degeneration of hepatocytes, cholestasis, neutrophils and hepatocyte necrosis) and degree of fibrosis were semi-quantitatively graded [18].

2.5. RNA preparation and reverse transcription

Total RNA was prepared using a modified guanidinium thiocyanate
method according to manufacturers instructions (RNA Clean, HybaidAGS, Heidelberg, Germany). Because of the low amount of mRNA recovered from liver biopsy specimens, we established a real-time reverse transcription-polymerase chain reaction (RT-PCR) for quantification of TNFa
and IL-8 mRNA. Reverse transcriptase reaction was performed with 1 mg
of total RNA, random hexamers (Roche, Mannheim, Germany) and MMLV
reverse trancriptase (GIBCO, Gaithersburg, MD) according to manufacturers directions.

2.6. TaqMan real-time polymerase chain reaction

For quantification of TNFa and IL-8 mRNA levels in liver biopsy specimens, the TaqMan real-time PCR (Perkin Elmer, Vienna, Austria) method
was used. TaqMan real-time PCR primers and TaqMan probe were
designed using the Primer Express Software (PE Applied Biosystems,
Vienna, Austria). Primers and probes were purchased from Microsynth
(Baglach, Switzerland) and were designed intron spanning to avoid co-

H. Tilg et al. / Journal of Hepatology 38 (2003) 419425

421

amplification of genomic DNA. For generation of the standard curve, serially diluted mRNA prepared from human monocytes was subjected to RT
as described above, and 2.5 ml of cDNA was incubated with 12.5 ml of 2
TaqMan Master Mix (Perkin Elmer; 8% glycerol, 1 TaqMan buffer,
200 mmol/l deoxyadenosine triphosphate, 200 mmol/l deoxyguanosine
triphosphate, 200 mmol/l deoxycytidine triphosphate, 400 mmol/l deoxyuridine triphosphate, 0.05 U/ml AmpErase uracil N-glycosylase, 5 mmol/l
MgCl2, 0.01 U/ml Amplitaq Gold), 2.5 ml TNFa/IL-8 TaqMan probe,
TNFa/IL-8 sense and reverse primers at a final concentration of
900 mmol/l. The amplification was carried out in the AbiPrism 7700
Sequence Detector (Perkin Elmer). Amplification conditions were 2 min
at 508C and 10 min at 958C, followed by 40 cycles of 15 s at 958C, followed
by 1 min at 608C. Analysis of data was done using the Sequence Detector
1.6 software. All reactions were performed in triplicate, and DRn and CT
were calculated from fluorescence activity data collected during the PCR. A
standard curve was generated upon blotting CT against RNA quantity.
To minimize intra- and inter-assay variability caused by differences in
RT efficiency, TNFa/IL-8 quantities were normalized to the amount of
glyceraldehyde-3-phosphate dehydrogenase (GAPDH) cDNA in liver
biopsy specimens. For GAPDH quantification, the GAPDH reaction mix
(Perkin Elmer) was used.

2.7. Statistical analysis

Data are expressed as median and range or mean and standard error.
Statistical analyses for laboratory parameters and cytokine measurements
were performed by Friedmans test checking for differences during the
observation period. Post-hoc analysis was performed by Wilcoxons signed
rank test to compare individual time-points with pretreatment levels (day
0).

3. Results
3.1. Clinical and laboratory assessments
The mean period from hospital admission to Infliximab
treatment was ,5 days in all patients. All patients were
deeply jaundiced when treatment was started. The infusions
were well tolerated with no acute side effects observed
(allergic episodes including rashes). Two of the 12 patients
died within the first 4 weeks of receiving Infliximab. Both
deaths were due to septicemia (one Candida albicans, one
Staphylococcus aureus). One of those two patients developed renal failure within 2 weeks after Infliximab administration and died on day 1 21 due too Candida septicemia.
Despite initial improvement in liver function, the other
patient developed staphylococcal septicemia and died
subsequently from multiorgan failure after 2 weeks. Ten
patients are still alive with a median observation period of
15 months (1220). Four patients had Grade II encephalopathy prior to administration of Infliximab and this had
improved to Grade I by day 7. Eleven of the 12 patients
had ascites, which was managed with a combination of
paracentesis and diuretics. Serum bilirubin levels fell significantly within 4 weeks of infusion P , 0:001. There was
also a significant decrease in white blood cell count P ,
0:05 and the C-reactive protein decreased P , 0:05.
Hemoglobin (pre: 107 (^5.1), post: 116 (^4.6) g/l) and
platelets (pre: 165.2 (^27), post 170.4 (^31.3) 10 9/l)
values showed no significant changes within the first 4

Fig. 1. Laboratory parameters in 12 patients with severe AH treated

with a single infusion of Infliximab (5 mg/kg body weight). Data are
presented as median (horizontal line) and 75% confidence interval
(box) with whiskers indicating minimal and maximal values; extreme
values are indicated with (8) and (*). (A) Total bilirubin levels (mmol/l),
(B) prothrombin time (PT) in seconds, (C) white blood cell count
(WBC, g/l). The symbol # indicates P , 0.05 and ## indicates P , 0.01.

H. Tilg et al. / Journal of Hepatology 38 (2003) 419425

Results from last samples taken.

Patients who died during follow-up.

Analyses based on surviving patients.

9.3
28.1
7.9
8.9
9.8
9.1
4.9
6.1
6.6
3.6
12
20
10.5
8.5
1.9
29.2
19
12
17
17.3
23
5.5
12.6
7.4
4.5
28
16
16
16.5
2.3
, 0.05
16.7
27
57.3
37.2
40.7
41
17.3
41.4
34.7
32.2
29.7
66.6
37.7
37.2
14.9
32.4
51.9
114.8
58.5
61.1
77.8
41.4
49.9
47.4
36.7
34.5
42.5
54.1
48.7
23
, 0.05
14
15.8
22
17
18
19
14.9
18.4
16.7
18.1
17.4
22.8
18.9
18.9
4.1
89
164 a
154
203
184
112
29
166
185
31
43
250 a
131
154
75
55
55
55
58
62
40
55
35
37
44
44
67
50.6
55
10.3
1
2
3
4
5
6
7
8
9
10
11
12
Mean c
Median c
SD c
P 0 vs. 28 c

216
415
272
331
298
583
166
279
465
147
212
310
308
288
127
, 0.001

15.8
17.5
33
20
21
21
18.4
18.8
15.9
17.6
16.3
16.8
19.5
19
4.8
NS

0
28
0

Days

28

28

(68)
(82)
(83)
(55)
(69)
(83)
(63)
(74)
(79)
(68)
(64)
(77)
(72)
(71)
(9)
(,0.05)

28

(73)
(78)
(73)
(68)
(74)
(75)
(42)
(45)
(86)
(40)
(60)
(60)
(57)
(60)
(14)

82
132
121
78
111
134
54
51
26
10
79
107
82.1
80.7
40.8
, 0.05

White cell count 10 9/l (%Neutrophils)

Maddrey index
Prothrombin time
Age

Bilirubin (mmol/l)

17
179
42
79
21
46
10
8
16.7
7
14
166
38
19
47.8

28

C-reactive protein (mg/l)

3.2. Cytokine analysis

Patient

Table 1
Clinical characteristics of the patients prior to and after treatment with Infliximab

weeks. There were no significant changes in the electrolytes

and the serum creatinine decreased from 110.5 (^7.3) to 91
(^5.5) mmol/l P 0:4 in the patients who improved and
deteriorated in the two patients who died (patient 2: 91
104 mmol/l and patient 12: 91314 mmol/l). Albumin
increased significantly in the surviving patients from 28.9
(^2.8) to 33.4 (^3.2) g/l P , 0:05. No significant
changes were observed in any of the liver enzymes (alanine
transaminase: pre, 28.6 (^3.9) to 28 days post, 34 (^5.4) U/
l; aspartate transaminase: pre, 54 (^7.6) to 28 days post 28
(^3.6) U/l; gamma glutamyl transpeptidase: pre, 192
(^45.8) to 28 days post, 93 (^34.4) U/l). Patient details
are presented in Fig. 1 and Table 1.

10
0.5 b
24
15
21
20
23
23
20
17
17
1b
16
18.5
8.1

Duration of follow-up (months)

422

Serial cytokine assessment was performed in 11/12

patients (Fig. 2). The measured levels of the bioactive
form of TNFa was widely variable amongst the patients.
TNFa was undetectable in six patients and in the other five,
the values were 13, 18, 133, 283 and 370 pg/ml. Of the two
patients who died, the levels of TNFa were undetectable in
one and 18 pg/ml in the other. Plasma levels of the proinflammatory cytokines IL-6, IL-8 and IFNg showed a
decrease 1 day after Infliximab administration, as did the
level of the putative anti-inflammatory cytokine IL-10.
Conversely, levels of IL-4, demonstrated an increase after
treatment (Fig. 2). The shed TNFa receptor I and II levels
were substantially higher than found in healthy volunteers,
as shown previously[19], and remained largely unchanged
over the first 7 days following Infliximab administration
(TNFR-I: pre: 1.14 (^0.2), post: 0.97 (^0.3) ng/ml;
TNFR-II: pre: 1.03 (^0.3), post: 0.86 (^0.3) ng/ml on
days 0 and 17, respectively). Indeed, these high receptor
levels may offer an explanation, at least in part, for our
observation of low circulating TNFa. It is probable that
changes in cytokine concentrations were not statistically
significant due to the small sample population and large
inter-individual variability.
3.3. Hepatic morphology
All patients were cirrhotic and showed histopathological
evidence of severe AH. In four of the treated patients, posttreatment biopsies (day 1 28) were available. Substantial
reduction of hepatic steatosis and inflammatory infiltrate
were observed in three of these four patients (Table 2).
3.4. Liver TNFa and IL-8 mRNA expression
Infliximab treatment did not affect TNFa/GAPDH cDNA
(pre: 0.05 (range 0.000.15), 28 days post: 0.08 (range 0.00
0.18)), whereas IL-8/GAPDH cDNA (pre: 0.06 (range 0.01
0.27), 28 days post: (0.01 (range 0.000.02); P , 0:07)) was
markedly reduced on day 128 compared to day 0 (Fig. 3).

H. Tilg et al. / Journal of Hepatology 38 (2003) 419425

Fig. 2. Changes in circulating cytokine levels measured by ELISA

assay, before and following (days 0, 1, 3 and 7) treatment with Infliximab (5 mg/kg body weight). Data are presented as mean ^ standard
error. None of the changes were statistically significant.

4. Discussion
Hospitalized patients with severe AH are often severely
ill, with high mortality, particularly when the calculated
Maddrey discriminant factor is .32 [1]. Our pilot study
demonstrated that Infliximab might be safely administered
to this patient group. In addition, Infliximab treatment was
accompanied by a continuous decrease in serum bilirubin
levels and DF score over 34 weeks, suggesting a potential
beneficial effect in this disorder. The clinical improvement
was paralleled by evidence of histological improvement
seen in the follow-up liver biopsies that were obtained
within 4 weeks after Infliximab administration. Although
the improvement in liver function tests, inflammatory
markers and histological appearances following administration of Infliximab were impressive, the lack of a control
group does not allow us to conclude whether some of the
observed changes were due to spontaneous improvement in
the clinical condition. However, in each instance there had
been deterioration in liver function following hospital
admission with a high Maddrey score (.32), indicative of
poor prognosis, at the time of inclusion into the study.
Our results are also consistent with experimental data
demonstrating the key role of TNFa in AH [4]. Anti-TNF
antibody treatment has also been successfully used to

423

prevent liver injury in alcohol-fed rats [14]. The observation

that levels of proinflammatory cytokines are increased in
patients with severe AH provided also the rationale for a
study with pentoxifylline, an inhibitor of TNF transcription,
in AH [20]. In this large placebo-controlled study, pentoxifylline improved survival by 40% and furthermore, reduced
the incidence of hepatorenal syndrome by 6570%.
Although glucocorticoid treatment has been reported to
improve survival, meta analysis of the clinical trials to
date has failed to show a clear benefit of such therapy
[2124]. Individual data analysis of the last three randomized placebo-controlled double-blind trials, however,
revealed evidence that corticosteroids might improve
short-term survival in patients with severe AH [22].
TNFa is able to induce many of the observed clinical
aspects of AH such as anorexia, fever, wasting, hypoalbuminemia and neutrophilia [25]. Serum/plasma concentrations of TNF-inducible cytokines, such as IL-1, IL-6 and
IL-8 have been shown to be increased initially in hospitalized patients with alcoholic steatohepatitis and decline
during recovery [69,26]. These inflammatory cytokines
(IL-6 and IL-8) correlate with the clinical course of
AH[4]. Levels of TNFa tend to be variable and were
shown not to decrease during pentoxifylline treatment
[18]. This is consistent with the present findings. Plasma
concentrations of several proinflammatory cytokines such
as IL-1b, IL-6, IL-8 and IFNg decreased within the first
days after Infliximab administration. TNF levels were
very variable in our patients with undetectable values in
six of the subjects. Indeed the data for the TNF levels in
patients with AH in the literature are widely variable
ranging from values as low as 5 pg/ml to as high as
400 pg/ml [11,20,2730] suggesting that other factors
such as concomitant infection, degree of endotoxemia and
genetic factors may be important in determining the
measured levels of TNF. We did, however, find substantial
serum levels of TNF receptors, which did not change with
treatment, suggesting that the TNFa binding proteins are
constantly being shed, and presumably sequester the bioactive TNFa. We were unable to detect any correlations
between the changes in the cytokine levels and the improvement in the liver function tests. Nevertheless, as is evident
from Fig. 1 and Table 1, the improvement in the liver func-

Table 2
Histological changes in four patients with severe AH treated with Infliximab (pre, biopsy pretreatment; day 1 28, biopsy 4 weeks after Infliximab
administration)
Patient no

Steatosis %

Ballooning

Cholestasis

Neutrophils

Mononueclar cells

Ductal metaplasia

Fibrosis

1, pre
1, day 1 28
2, pre
2, day 1 28
3, pre
3, day 1 28
4, pre
4, day 1 28

60
,5
70
40
30
0
80
20

11
2
111
(1)
111
2
2
2

111
1
111
(1)
111
11
1
2

1
2
111
1
111
11
1
1

1
1
11
11
1
11
1
1

11
1
111
(1)
11
11
1
1

111
111
11
11
111
111
111
111

424

H. Tilg et al. / Journal of Hepatology 38 (2003) 419425

Fig. 3. Liver IL-8 and TNFa cDNA levels in four patients pre- and posttreatment (d 1 28). Complementary cDNA was synthesized from
messenger RNA obtained from liver biopsies. Quantification was
performed by TaqMan real-time PCR. Upper panel: IL-8/GAPDH
cDNA ratio and TNFa/GAPDH cDNA ratio are presented as median
(horizontal line), 75% confidence interval (box), and minimal and
maximal values (whiskers). Middle panel: representative TaqMan
real-time amplification plots for IL-8; fluorescence activity (DRn) is
plotted against cycle number; filled and open circle, day 0; rectangle
and cross, day 28. Lower panel: representative TaqMan amplification
plots for TNFa as described above. Please note that IL-8 and TNFa
have been normalized against respective GAPDH control amplifications as described in Section 2.

tion tests paralleled the reduction in the inflammatory

markers. The inflammatory markers (CRP) increased only
in the two patients who died (Patients 2 and 12) suggesting
that the severity of inflammation is likely to be important in
modulating the outcome following administration of Infliximab in AH patients.
IL-10, an anti-inflammatory cytokine showed a transient
decrease, but the levels of another anti-inflammatory cyto-

kine, namely IL-4, were found to remain elevated, possibly

pointing to a change in the balance of T-helper lymphocyte
populations towards the recovery process. The early reduction in IL-10 may represent a more general inhibition of the
activated white cells whereas the increase in IL-4 may be a
more specific response.
Since serum cytokine levels reflect whole body dynamics,
it was, therefore, of interest to find that the liver mRNA
levels for IL-8, an important neutrophil attractant chemokine, was substantially reduced at day 1 28 in the liver
tissue. Transcription of the genes for TNFa and other cytokines are barely detectable in the normal liver, whereas
induced TNFa initiates the regeneration of liver tissue
after injury [5,31]. Although the use of Infliximab might
thereby inhibit liver regeneration, the present data showed
a slight improvement in PT (as an indicator of hepatic synthesis), suggesting that, either residual TNFa and/or other
cytokines such as IL-6 could take over a role in the regeneration of the diseased liver [32]. More importantly, the
transcription of the TNFa-regulated chemokine IL-8 was
almost completely absent in the liver within 4 weeks of
treatment with Infliximab. IL-8 may be responsible for the
recruitment of neutrophils into the liver in this disorder [9]
and the observed reduction of IL-8 is consistent with the
reduction in neutrophil infiltration observed histologically
after treatment. The histological data must be interpreted in
light of the fact that the pre- and post-biopsies were available only from four patients.
The mechanism of beneficial effect of Infliximab in
Crohns disease, rheumatoid arthritis and psoriasis
[16,17,33] is not known although a recent observation
suggested that there is an induction of apoptosis (a pathologic feature of human AH [34,35]), and thereby an elimination of inflammatory cells which may be of importance [36].
In relation to the recent finding that patients treated with
Infliximab may be prone to develop infectious diseases such
as tuberculosis [37], this complication was not observed
over a median follow-up of 15 months in our patients who
did not receive any prophylactic treatment. On the basis of
the data presented here, a randomized placebo-controlled
clinical trial of anti-TNF antibody in severe AH is
warranted.

Acknowledgements
This work was supported by the Austrian Science Fund (P
14641 and 15783) and the Foundation for Liver Research
UK.

References
[1] Maddrey WC, Boitnott JK, Bedine MS, Weber Jr FL, Mezey E, White
Jr RI. Corticosteroid therapy of alcoholic hepatitis. Gastroenterology
1978;75:193199.
[2] Orrego H, Blake JE, Blendis LM, Medline A. Prognosis of alcoholic

H. Tilg et al. / Journal of Hepatology 38 (2003) 419425

[3]

[4]
[5]
[6]
[7]

[8]

[9]
[10]

[11]

[12]

[13]

[14]

[15]

[16]

[17]

[18]
[19]

[20]

cirrhosis in the presence and absence of alcoholic hepatitis. Gastroenterology 1987;92:208214.

McClain CJ, Barve S, Barve S, Deaciuc I, Hill DB. Tumor necrosis
factor and alcoholic liver disease. Alcohol Clin Exp Res
1998;22:248S252S.
McClain CJ, Barve S, Deaciuc I, Kugelmas M, Hill D. Cytokines in
alcoholic liver disease. Semin Liver Dis 1999;19:205219.
Tilg H, Diehl AM. Cytokines in alcoholic and nonalcoholic steatohepatitis. N Engl J Med 2000;343:14671476.
McClain CJ, Cohen DA. Increased tumor necrosis factor production
by monocytes in alcoholic hepatitis. Hepatology 1989;9:349351.
Bird GL, Sheron N, Goka AK, Alexander GJ, Williams RS. Increased
plasma tumor necrosis factor in severe alcoholic hepatitis. Ann Intern
Med 1990;112:917920.
Khoruts A, Stahnke L, McClain CJ, Logan G, Allen JI. Circulating
tumor necrosis factor, interleukin-1 and interleukin-6 concentrations
in chronic alcoholic patients. Hepatology 1991;13:267276.
Hill DB, Marsano LS, McClain CJ. Increased plasma interleukin-8
concentrations in alcoholic hepatitis. Hepatology 1993;18:576580.
Sheron N, Bird G, Koskinas J, Portmann B, Ceska M, Lindley I,
Williams R. Circulating and tissue levels of the neutrophil chemotaxin interleukin-8 are elevated in severe acute alcoholic hepatitis,
and tissue levels correlate with neutrophil infiltration. Hepatology
1993;18:4146.
Hanck C, Rossol S, Bocker U, Tokus M, Singer MV. Presence of
plasma endotoxin is correlated with tumour necrosis factor receptor
levels and disease activity in alcoholic cirrhosis. Alcohol Alcohol
1998;33:606608.
Felver ME, Mezey E, McGuire M, Mitchell MC, Herlong HF, Veech
GA, Veech RL. Plasma tumor necrosis factor alpha predicts decreased
long-term survival in severe alcoholic hepatitis. Alcohol Clin Exp Res
1990;14:255259.
Adachi Y, Moore LE, Bradford BU, Gao W, Thurman RG. Antibiotics prevent liver injury in rats following long-term exposure to ethanol. Gastroenterology 1995;108:218224.
Iimuro Y, Gallucci RM, Luster MI, Kono H, Thurman RG. Antibodies to tumor necrosis factor alfa attenuate hepatic necrosis and
inflammation caused by chronic exposure to ethanol in the rat. Hepatology 1997;26:15301537.
Yin M, Wheeler MD, Kono H, Bradford BU, Gallucci RM, Luster MI,
Thurman RG. Essential role of tumor necrosis factor alpha in alcoholinduced liver injury in mice. Gastroenterology 1999;117:942952.
Targan SR, Hanauer SB, van Deventer SJ, Mayer L, Present DH,
Braakman T, et al. A short-term study of chimeric monoclonal antibody cA2 to tumor necrosis factor alpha for Crohns disease. Crohns
disease cA2 study group. N Engl J Med 1997;337:10291035.
Maini R, St Clair EW, Breedveld F, Furst D, Kalden J, Weisman M, et
al. Infliximab (chimeric anti-tumour necrosis factor alpha monoclonal
antibody) versus placebo in rheumatoid arthritis patients receiving
concomitant methotrexate: a randomised phase III trial. ATTRACT
study group. Lancet 1999;354:19321939.
Alcoholic liver disease: morphological manifestations. Review by an
international group. Lancet 1981;1:707711.
Van Zee KJ, Kohno T, Fishcer E, Rock CS, Moldawer LL, Lowry SF.
Tumor necrosis factor soluble receptors circulate during experimental
and clinical inflammation and can protect against excessive tumor
necrosis factor a in vitro and in vivo. Proc Natl Acad Sci USA
1992;89:48454849.
Akriviadis E, Botla R, Briggs W, Han S, Reynolds T, Shakil O.
Pentoxifylline improves short-term survival in severe acute alcoholic

[21]

[22]

[23]
[24]

[25]
[26]

[27]

[28]

[29]

[30]

[31]

[32]

[33]

[34]

[35]

[36]

[37]

425

hepatitis: a double-blind, placebo-controlled trial. Gastroenterology

2000;119:16371648.
Christensen E, Gluud C. Glucocorticoids are ineffective in alcoholic
hepatitis: a meta-analysis adjusting for confounding variables. Gut
1995;37:113118.
Mathurin P, Mendenhall CL, Carithers RL, Ramond MJ, Maddrey
WC, Garstide P, et al. Corticosteroids improve short-term survival
in patients with severe alcoholic hepatitis (AH): individual data analysis of the last three randomized placebo controlled double blind trials
of corticosteroids in severe AH. J Hepatol 2002;36:480487.
Christensen E. Alcoholic hepatitis glucocorticosteriods or not? J
Hepatol 2002;36:547548.
Imperiale TF, McCullough AJ. Do corticosteroids reduce mortality
from alcoholic hepatitis? A meta-analysis of the randomized trials.
Ann Intern Med 1990;113:299307.
Bazzoni F, Beutler B. The tumor necrosis factor ligand and receptor
families. N Engl J Med 1996;334:17171725.
Hill DB, Marsano L, Cohen D, Allen J, Shedlofsky S, McClain CJ.
Increased plasma interleukin-6 concentrations in alcoholic hepatitis. J
Lab Clin Med 1992;119:547552.
Rodriguez-Rodriguez E, Gonzalez-Reimers E, Santolaria-Fernandez
F, Milena-Abril A, Rodriguez-Moreno F, Oramas-Rodriguez J, Martinez-Riera A. Cytokine levels in acute alcoholic hepatitis: a sequential
study. Drug Alcohol Depend 1995;39:2327.
Fujimoto M, Uemura M, Nakatiani Y, Tsujita S, Hoppo K, Tamagawa
T, et al. Plasma endotoxin and serum cytokine levels in patients with
alcoholic hepatitis: relation to severity of liver disturbance. Alcohol
Clin Exp Res 2000;24:48S54S.
Spahr L, Rubbia-Brandt L, Pugin J, Giostra E, Frossard JL, Borisch B,
Hadengue A. Rapid changes in alcoholic hepatitis histology under
steroids: correlation with soluble intercellular adhesion molecole-1
in hepatic venous blood. J Hepatol 2001;35:582589.
Spahr L, Rubbia-Brandt L, Frossard J, Giostra E, Rougemont A,
Pugin J, et al. Combination of steroids with infliximab or placebo in
severe alcoholic hepatitis: a randomized controlled pilot study. J
Hepatol 2002;37:448456.
Akerman P, Cote P, Yang SQ, McClain C, Nelson S, Bagby GJ, Diehl
AM. Antibodies to tumor necrosis factor-alpha inhibit liver regeneration after partial hepatectomy. Am J Physiol 1992;263:G579G585.
Cressman DE, Greenbaum LE, DeAngelis RA, Ciliberto G, Furth EE,
Poli V, Taub R. Liver failure and defective hepatocyte regeneration in
interleukin-6-deficient mice. Science 1996;274:13791383.
Chaudhari U, Romano P, Mulcahy LD, Dooley LT, Baker DG,
Gottlieb AB. Efficacy and safety of infliximab monotherapy for
plaque-type psoriasis: a randomised trial. Lancet 2001;357:1842
1847.
Natori S, Rust C, Stadheim LM, Srinivasan A, Burgart LJ, Gores GJ.
Hepatocyte apoptosis is a pathologic feature of human alcoholic hepatitis. J Hepatol 2001;34:248253.
Ziol M, Tepper M, Lohez M, Arcangeli G, Ganne N, Christidis C, et
al. Clinical and biological relevance of hepatocyte apoptosis in alcoholic hepatitis. J Hepatol 2001;34:254260.
Lugering A, Schmidt M, Lugering N, Pauels HG, Domschke W,
Kucharzik T. Infliximab induces apoptosis in monocytes from
patients with chronic active Crohns disease by using a caspasedependent pathway. Gastroenterology 2001;121:11451157.
Keane J, Gershon S, Wise RP, Mirabile-Levens E, Kasznica J,
Schwieterman WD, et al. Tuberculosis associated with infliximab, a
tumor necrosis factor alpha-neutralizing agent. N Engl J Med
2001;345:10981104.