CLINICAL GASTROENTEROLOGY AND HEPATOLOGY 2007;5:502–507

Effect of Prolonged Interferon Therapy on the Outcome of Hepatitis C

Virus–Related Cirrhosis: A Randomized Trial

LAETITIA FARTOUX,*,‡ FRANÇOISE DEGOS,§ CHRISTIAN TRÉPO,¶ ODILE GORIA,储 PAUL CALÈS,# ALBERT TRAN,**
CATHERINE BUFFET,‡‡ THIERRY POYNARD,§§ DOMINIQUE CAPRON,储 储 JEAN–JACQUES RAABE¶¶
DOMINIQUE ROULOT,*** SYLVIE NAVEAU,## JEAN–DIDIER GRANGE,‡‡‡ RENÉE E. POUPON,* RAOUL POUPON,*,‡ and
LAWRENCE SERFATY*,‡
*AP-HP, Hôpital Saint-Antoine, Paris, France; ‡Université Pierre et Marie Curie (UPMC-Paris 6), Faculté de Médecine Saint-Antoine, Paris, France; §AP-HP Hôpital
Beaujon, Clichy, France; ¶Hôpital Hôtel Dieu, Lyon, France; 储Hopital Charles Nicole, Rouen, France; #CHU d’Angers, Angers, France; **Hôpital de l’Archet, Nice,
France; ‡‡AP-HP Hôpital Bicêtre, Le Kremlin-Bicêtre, France; §§AP-HP Hôpital La Pitié-Salpétrière, Paris, France; 储 储CHU Hôpital Nord, Amiens, France; ¶¶Hôpital
Notre Dame du Bon Secours, Metz, France; ##AP-HP, Hôpital Avicenne, Bobigny, France; ***AP-HP Hôpital Antoine Béclère, Clamart, France; ‡‡‡AP-HP, Hôpital
Tenon, Paris, France

Background & Aims: The impact of interferon (IFN)
treatment on the occurrence of complications related to
hepatitis C virus (HCV)-related cirrhosis is debated because
the majority of studies are retrospective. We designed a
randomized controlled trial comparing the efficacy of pro-
longed IFN alfa-2a treatment vs nontreatment on compli-
cation-free survival in patients with compensated HCV
cirrhosis. Methods: A total of 102 patients (mean age, 60.5
ⴞ 9.5 y; male/female ratio, .82) with biopsy examination–
proven HCV cirrhosis, Child–Pugh score A, who were hep-
atocellular carcinoma (HCC) free, and had at least 1 risk
factor of complications were randomized to receive IFN or
no therapy for 24 months. Results: During the follow-up
evaluation, the complication rate was 24.5%: HCC occurred
in 12 and decompensation unrelated to HCC occurred in 13
patients. The number of HCC patients was similar in both
groups. The probability of complication-free survival was
not significantly different between treated and untreated
patients (98% and 72.3% vs 90% and 70.7% at 12 and 24 mo,
respectively, P ⴝ .59). The median time until complication
occurrence was 17.1 months in the treated group vs 13.6
months in the untreated group (P ⴝ .2). Conclusions:
This randomized controlled trial showed that a 2-year
course of IFN has little or no impact on complication-free
survival in patients with high-risk compensated HCV
cirrhosis.
patients with HCV-related cirrhosis, whereas in another con-
trolled trial, Valla et al15 did not find such an effect. A second
unanswered question is whether maintenance treatment with
IFN is capable of preventing complications in nonvirologic
responders through its antifibrotic and antiproliferative prop-
erties.16 –19 In a previous retrospective study, we showed that
IFN therapy was associated with better complication-free sur-
vival in patients with compensated HCV cirrhosis, independent
of the virologic response.3 However, this result was not con-
firmed in the meta-analysis by Camma et al,11 who found
beneficial effects of IFN therapy in HCC occurrence only in
sustained virologic responders. Finally, because of the lack of
randomized controlled trials with survival end points, it is still
a matter of controversy whether maintenance treatment with
IFN is beneficial or not in patients with HCV-related cirrhosis.
We therefore conducted a randomized controlled trial to
assess the effects of prolonged administration of IFN alfa-2a
(3 million units 3 times/wk) vs nontreatment on complication-
free survival in patients with compensated HCV cirrhosis. To
increase the number of events, we selected patients with at least
one risk factor of complications.
Patients and Methods
Patient Selection
Ten French centers participated in the study. The pro-
tocol was approved by the local ethics committees. Patients
prospectively seen between January 1999 and October 2002

P atients with hepatitis C virus (HCV)-related cirrhosis are at

risk for progression of the disease and clinical deteriora-
tion.1– 6 The estimated yearly incidences for decompensation or
hepatocellular carcinoma (HCC) vary from 3.6% to 6.0% and
from 1.4% to 3.3%, respectively. Known risk factors for compli-
cations include age, liver function, portal hypertension, and
serum ␣-fetoprotein (AFP) level.2,3,7–9
The effectiveness of interferon (IFN) therapy in the preven-
tion of complications in patients with HCV-related cirrhosis is
still a matter of debate. Because of the need for a long-term
follow-up period to monitor complication occurrences, the ef-
fectiveness of IFN has been assessed mainly in multiple retro-
spective and uncontrolled studies, with conflicting results.10 –12
Nishigushi et al,13,14 in a randomized controlled trial, found
that IFN therapy may prevent or delay progression to HCC in
were enrolled if they met the following criteria: (1) age 18 –75
years; (2) informed written consent given before randomization
into the study; (3) histologic diagnosis of cirrhosis made within
2 years of enrollment; (4) positive HCV RNA in serum by
quantitative polymerase chain reaction assay; (5) no history or
clinical signs of complications of cirrhosis (ie, ascites, jaundice
[serum bilirubin level ⬎51 ␮mol/L], encephalopathy, or variceal
bleeding) on entry into the study; (6) no previous history of
Abbreviations used in this paper: AFP, ␣-fetoprotein; ALT, alanine
aminotransferase; HCC, hepatocellular carcinoma; HCV, hepatitis C
virus; IFN, interferon; NS, not significant.
© 2007 by the AGA Institute
1542-3565/07/$32.00
doi:10.1016/j.cgh.2006.10.016
April 2007
sclerotherapy or esophageal varices; (7) no evidence of HCC at
entry into the study on the basis of ultrasonography and AFP
level less than 50 ng/L; (8) absence of serum hepatitis B surface
antigen; (9) absence of co-existing chronic liver disease;
(10) absence of human immunodeficiency virus antibodies; (11)
absence of excessive alcohol consumption (⬎40 g/day); and (12)
at least 1 risk factor of complications: age older than 55 years,
presence of esophageal varices, serum bilirubin level greater
than 17 umol/L (and ⬍51 umol/L), prothrombin time less than
80%, serum albumin level less than 35g/L, platelet count less
than 130,000/mL, or AFP level greater than 20 ng/mL (and ⬍50
ng/L).
Study Design
Patients were assigned randomly by each center to re-
ceive either IFN-alfa2a (Roche Laboratory, Neuilly Sur Seine,
France), 3 millions units 3 times/wk for 2 years, or no treat-
ment. All patients were examined every 3 months for 2 years.
Routine evaluation included physical examination, standard
biochemical tests, AFP assay, and liver ultrasound examination
every 6 months. All patients were screened for varices at the
time of study entry, and then at 12 and 24 months. Patients
with medium and large esophageal varices were treated with
nonselective ␤-blockers. Compliance with the treatment regi-
men was assessed by monitoring adherence to the schedule of
visits.
Methods
Clinical parameters. The following data were re-
corded from all patients at the time of inclusion: age, sex, body
mass index, history of transfusion or intravenous drug use,
duration of HCV infection in patients with known risk factors,
␤-blocker intake, and a history of previous antiviral treatment.
Biochemical parameters. At randomization and at
each follow-up visit, blood samples were taken for serum AFP
level, alanine aminotransferase (ALT) level, albumin level, bili-
rubin level, prothrombin time, and platelet count. The Child–
Pugh score also was assessed.
Virologic parameters. All subjects were reactive for
anti-HCV antibodies and were tested for HCV RNA by reverse-
transcription polymerase chain reaction (Amplicor HCV; Roche
Diagnostic Systems, Branchburg, NJ) at the time of inclusion,
and then at 12 and 24 months. HCV genotypes were deter-
mined using the Inno-Lipa II HCV (Innogenetics, Ghent, Bel-
gium).
Histologic evaluation. Paraffin-embedded biopsy
specimens were re-analyzed by a single pathologist. Liver biopsy
specimens were stained with hematoxylin-phloxin-safran and
picrosirius red, and scored according to Knodell et al.20
Randomization. The coordinating center for the trial
was the Clinical Research Unit at the Hôpital Saint-Antoine.
Patients who enrolled in the study were assigned randomly in a
1:1 ratio to the IFN group or to the control group by the
controller. Randomization was stratified according to the cen-
ter.
Study end points. The primary end point was com-
plication-free survival. Complications were defined by the occur-
rence of at least 1 of the following criteria during the 24-month
follow-up period: HCC, variceal bleeding, ascites, hepatic enceph-
alopathy, or jaundice (defined by bilirubinemia ⬎51 ␮mol/L).
IFN AND HCV CIRRHOSIS 503
Decompensation was considered to be related to HCC if it was
diagnosed at the same time as HCC. The diagnosis of HCC was
always proven with a biopsy examination.
Secondary end points also were recorded during the 24-
month follow-up period: death or liver transplantation; changes
in serum ALT activities, serum albumin level, serum bilirubin
level, and prothrombin; esophageal varices grade increase
higher than 1 (grade 0 –2/3 or grade 1–3); end-of-treatment
virologic and biochemical responses; and adverse effects of
treatment. Safety was assessed by means of physical examina-
tions, laboratory tests, and spontaneous reports of clinical ad-
verse events during visits. A step-wise reduction in the dose of
IFN alfa-2a to 2 or 1 million units 3 times/wk was allowed when
necessary to manage clinically significant adverse events or
laboratory abnormalities according to a defined protocol. A
3-member Outcomes Committee, consisting of external experts
from the study, reviewed data from each outcome reported to
confirm that outcome criteria had been met.
Statistical Analysis
Sample size considerations. Sample size was based
on the primary clinical end point of time to treatment failure.
To estimate power, the rate of complications at 2 years was
assumed to be 25% without IFN in high-risk patients whereas a
reduction of 20% for the IFN group was taken as a clinically
relevant treatment effect.2,3 For 80% power at the 5% level with
a 1:1 randomization the required sample size was estimated to
be 100 subjects (50 patients per arm).
Intent-to-treat population. The intent-to-treat pop-
ulation consisted of all subjects randomized into the study,
regardless of whether they took study medication or completed
the planned duration of the study (ie, the entire randomized
population).
Treatment comparisons. The planned treatment
comparison was between IFN and nontreatment. All efficacy
analyses were performed on the intent-to-treat population. De-
scriptive statistics were provided as means ⫾ SD or percentages.
The Fisher exact test or the Mann–Whitney test were used for
statistical comparisons. Courses of biochemical parameters and
of esophageal varices grade during follow-up evaluation were
compared using repeated-measures analysis. The Kaplan–Meier
method was used to estimate the complication-free survival
curves, and the log-rank test for trend with 1 degree of freedom
was used to test for linear trend according to the treatment
group. The effects of treatment after adjusting for prognostic
factors such as age, bilirubin level, albumin level, prothrombin
time, platelet count, serum AFP level, the presence of esopha-
geal varices, and previous antiviral treatment was estimated
using the Cox proportional hazards regression model. A P value
of less than .05 was considered to denote a significant differ-
ence. Analyses were performed using the BMDP package (SAS
software version 8.1; SAS Institute Inc, Cary, NC).
Results
Baseline Patient Characteristics
A total of 102 patients with biopsy examination–proven
HCV cirrhosis, Child–Pugh score A, fulfilled the inclusion cri-
teria and were randomized between January 1999 and October
2002. Fifty-one of these patients were treated with recombinant
IFN alfa-2a for 24 months; the remaining 51 did not receive
504 FARTOUX ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 5, No. 4

Table 1. Baseline Characteristics of Patients According to Treatment Arm

Treatment (n ⫽ 51) No treatment (n ⫽ 51) Pa

Age, yb 60.5 ⫾ 9.7 (35–75) 60.5 ⫾ 9.4 (31–75) NS

Sex ratio, M/F 23/28 23/28 NS
Body mass index, kg/m2b 26.6 ⫾ 4.8 27.5 ⫾ 3.7 NS
Duration of contamination, yb,c 21.6 ⫾ 10.6 22 ⫾ 11.6 NS
Transfusion/intravenous drug use 27/2 22/1 NS
Genotype 1 (%) 78 80 NS
Platelet count, ⫻106/mm3b 133 ⫾ 53 (67–286) 117 ⫾ 39 (60–221) NS
Prothrombin time, %b 82 ⫾ 13 (53–108) 84 ⫾ 11 (61–106) NS
Albumin, g/Lb 39.6 ⫾ 5 (29–50) 40.2 ⫾ 4 (31–52 NS
Bilirubin, ␮mol/Lb 16 ⫾ 2 (6–45) 16 ⫾ 8 (4–39) NS
ALT, ⫻Nb 2.9 ⫾ 2.2 2.7 ⫾ 1.6 NS
Knodell scoreb 10.7 ⫾ 2.7 11.1 ⫾ 2.3 NS
Esophageal varices grade (0/1/2/3) 30/13/6/2 30/10/10/1 NS
␤-blocker intake 13 11 NS
Previous antiviral treatment (%) 73 86 NS
IFN monotherapy 35 36
IFN ⫹ ribavirin 38 50
AFP, ng/mLb 12 ⫾ 10 (2–45) 13 ⫾ 11 (3–44) NS

NS, not significant.

aMann–Whitney or Fisher exact tests.
bMean ⫾ SD (range).
cIn patients with known risk factor.

treatment during the same period. Baseline characteristics of
the patients were similar in both treatment arms (Table 1). All
patients had stopped alcohol consumption more than 6
months before randomization. Eighty-two percent of patients
had more than 1 risk factor of complications as previously
defined (Figure 1). The majority of patients in both arms had
received previous antiviral treatment (IFN monotherapy in 36%
and IFN ⫹ ribavirin combination in 44%) and all had positive
HCV RNA at the time of inclusion.
Follow-up Period
The follow-up period was 24 months and no patient
was lost to follow-up evaluation. Among the 51 patients in the
IFN group, 26 had treatment discontinuation or dose reduction
as a result of the occurrence of complications (9 patients) or
side effects (17 patients). Therefore, the median dose of IFN
that was received was 711 million units (range, 70 –965 million
units). At the end of the follow-up period, 9 (18%) treated
patients had normalization of ALT activity and 2 (4%) had a
virologic response (vs 4% and 0% of untreated patients). During
follow-up evaluation, the complication rate was 24.5%: HCC
occurred in 12 patients or decompensation unrelated to HCC
occurred in 13 patients (ascites and/or jaundice in 7, variceal
bleeding in 5, and encephalopathy in 1). HCC was diagnosed at
an early stage in all patients; HCC was unifocal in 11 patients
and bifocal in 1 patient, with a median size of 27 mm (range,
15–35 mm). All HCCs but one were diagnosed more than 12
months after inclusion. Four patients died during the follow-up

Figure 1. Distribution of the number of risk factors of complication at

baseline in the study population. Risk factors are defined in the Patients Figure 2. Cumulative probability of complication-free survival accord-
and Methods section. ing to treatment arm (bold line, treatment; thin line, no treatment).
April 2007
Table 2. Number of Actual Events According to Treatment
Arm During the 24-Month Follow-up Period
Treatment No treatment
(n ⫽ 51) (n ⫽ 51)
HCC 6 6
Variceal bleeding 2 3
Ascites and/or jaundice 3 4
Hepatic encephalopathy 1 0
Liver-related death 1 2
HCC, hepatocellular carcinoma.
period. The causes of death were HCC in 2 patients, liver failure
in 1 patient, and a nonhepatic cause in 1 patient.
Outcomes According to Treatment Group
The probability of complication-free survival was not
significantly different between treated and untreated patients
(98% and 72.3% vs 90% and 70.7% at 12 and 24 mo, respectively,
P ⫽ .59) (Figure 2). The number of actual events according to
treatment arm is shown in (Table 2). Overall complications
occurred in 22% of treated vs 27% of untreated patients (not
significant [NS]). Considering the 31 treated patients received
more than 80% of the IFN dose, the difference still was not
significant (23% vs 27%). The median time until complication
occurrence was 17.1 vs 13.6 months in treated and untreated
patients, respectively (NS). The probabilities of HCC occurrence
and of decompensation unrelated to HCC (ie, ascites, jaundice,
variceal bleeding, or hepatic encephalopathy) were both similar
between treated and untreated patients (0% and 18% vs 2% and
16% at 12 and 24 mo, respectively, P ⫽ .96; 2% and 10% vs 8%
and 14.5% at 12 and 24 mo, respectively, P ⫽ .33) (Figure 3). The
median time until HCC occurrence was 22.9 months (range,
17.1–24 mo) in the treated group vs 15.3 months (range, 4.6 –24
mo) in the untreated group (P ⫽ .07), with a yearly incidence of
5.7% in both groups. The median time until decompensation
occurrence was 15.2 months in the treated group vs 11.9
months in the untreated group (NS). During the follow-up
period there was no difference in terms of the percentage of
Figure 3. Cumulative probabilities of HCC occurrence (A) and of HCC-free decompensation (ascites and/or jaundice, variceal bleeding, enceph-
alopathy) (B) according to treatment arm (bold line, treatment; thin line, no treatment).
IFN AND HCV CIRRHOSIS 505
patients with variceal bleeding (3.9% vs 5.9%), clinical decom-
pensation (ascites and/or jaundice and/or encephalopathy)
(7.8% vs 7.8%), or isolated esophageal varices grade increase
(7.8% vs 9.8%) unrelated to HCC between treated and untreated
groups, respectively. Time courses of the changes in the main
biologic variables are shown in Figure 4. Although the mean
ALT levels during follow-up evaluation tended to be lower in
treated than in untreated patients, the serum bilirubin level,
albumin level, and prothrombin time were not modified signif-
icantly.
Prognostic Factors of Complication-Free
Survival
There was a positive relationship between the number
of baseline risk factors and the rate of events during the fol-
low-up evaluation (P ⫽ .004) (Figure 5). Potential risk factors
affecting complication-free survival were analyzed by Cox pro-
portional hazards regression (Table 3). The platelet count was
the only independent factor predictive of complication-free
survival.
Discussion
This randomized controlled trial showed that a 2-year
course of IFN has little or no impact on complication-free
survival in patients with high-risk compensated HCV cirrhosis.
In this population, the only independent prognostic factor of
complication-free survival was low platelet counts.
This was a randomized controlled trial assessing the effects
of prolonged IFN therapy vs nontreatment on complication-
free survival in patients with HCV-related cirrhosis. Because
pegylated IFN was not available at the time of this trial, our
patients were treated with a standard regimen of IFN (3 million
units 3 times/wk). As in previous retrospective studies con-
ducted in Western countries, our cirrhotic patients were con-
taminated predominantly by transfusion.2,3 It has been sug-
gested that transfusion-transmitted HCV may be associated
with more severe liver disease.21 We have no clear explanation
for the female predominance in our cohort.
506 FARTOUX ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 5, No. 4

At baseline, all patients had compensated biopsy examina-

tion–proven cirrhosis with positive HCV RNA despite previous
antiviral treatment in the majority of the patients. To obtain
enough power for statistical analysis we included patients with
at least 1 risk factor of complications, such as age, a slight
alteration of liver function tests, portal hypertension, or a slight
increase in AFP level (⬍50 ␮g/L). In this high-risk population,
the overall complication rate was 24.5% at 2 years, reaching 39%
in patients having more than 3 baseline risk factors. Yearly
incidences of HCC or decompensation unrelated to HCC were
5.7% and 6.5%, respectively. Liver-related death occurred in 3
(3%) patients during the follow-up evaluation. These complica-
tion rates were higher than those retrospectively observed in
unselected European patients with HCV-related cirrhosis.2,3
In the absence of viral eradication, it is important to consider
the possibility that IFN therapy results in alternative benefits to
patients with HCV-related cirrhosis. Only 2 patients in the
treated group had negative polymerase chain reaction at the
end of the follow-up evaluation. Those 2 patients had no
complications but an analysis was not performed because of a Figure 5. Rate of events according to the number of risk factors at
lack of power. Therefore, our study mainly evaluated properties baseline. P ⫽ .004.

of IFN other than antiviral (ie, anti-inflammatory or antiprolif-

erative).16 –19 This study evaluated maintenance treatment with
a 2-year course of IFN in patients with HCV-related cirrho-
sis.11–15
Considering the intent-to-treat population, we did not ob-
serve any significant effects of IFN treatment vs nontreatment
on complication-free survival. Similar results were obtained
when considering patients who received more than 80% of the
IFN dose. Our results do not confirm the prospective study of
Nishigushi et al,13,14 who found a preventive effect of a short
course of IFN on HCC occurrence in cirrhotic patients despite
the absence of virologic response. Indeed, although the median
time until occurrence was shorter in the treated group,
the probabilities of HCC occurrence were similar in both
groups. The discrepancy between our results and those of the
Nishigushi et al13,14 studies could be owing to the far higher
incidence of HCC in Japanese patients. Finally, our conclusions
confirm the meta-analysis of Camma et al11who showed that
the frequency of HCC in cirrhotic patients was similar between
nonvirologic responders and untreated patients. However, we
cannot exclude a delayed effect of IFN on HCC occurrence, and
longer follow-up time is needed before drawing definitive con-
clusions.

Table 3. Cox Proportional Hazards Regression Assessing

Prognostic Factors of Complication-Free Survival
P OR (95% CI)

Age .48 1.02 (0.97–1.07)

Prothrombin time .79 1.005 (0.96–1.05)
Bilirubin level .83 0.99 (0.94–1.05)
Albumin level .26 0.94 (0.84–1.05)
Platelet count .01 0.98 (0.96–0.99)
AFP level .4 0.91 (0.73–1.14)
Presence of esophageal varices .23 1.8 (0.68–4.94)
Previous antiviral treatment .14 0.497 (0.19–1.28)
Figure 4. Course of ALT level, serum bilirubin level, prothrombin time,
Treatment arm .87 0.931 (0.37–2.31)
and serum albumin level in treated () and untreated ( ) patients. Val-
ues are expressed as mean ⫾ SEM. OR, odds ratio; 95% CI, 95% confidence interval.
April 2007
The potential effect of IFN therapy on liver function and
portal hypertension in cirrhotic patients has been evaluated in
a few studies that were mainly retrospective. In a previously
published study, we found that a history of IFN therapy was
associated independently with a lower probability of decompen-
sation in patients with HCV-related cirrhosis.3 These results
were not confirmed in another retrospective European study
with a larger number of patients.2 In the present study, the
probabilities of HCC-free decompensation were similar between
treated and untreated patients. The lack of effect of IFN on liver
function is reinforced by the courses of prothrombin time,
serum albumin level, or bilirubin level, which were similar
between both groups. It has been suggested that IFN therapy
can decrease the portal gradient in cirrhotic patients without
precise data on the virologic response.15,22,23 In our study, the
nonvirologic response in treated patients may explain the lack
of benefit of IFN on portal hypertension complications such as
variceal bleeding or variceal increase size.
In terms of tolerance, 17 of 51 (33%) patients had treatment
discontinuation or dose reduction because of side effects. This
result is in accordance with published data in cirrhotic patients
treated with standard IFN.24,25 Maintenance treatment with
pegylated IFN at a lower dosage as recommended may be more
valuable in terms of tolerance in cirrhotic patients.24
In conclusion, our results do not support the concept of
maintenance treatment with standard IFN monotherapy for
preventing complications in patients with high-risk compen-
sated HCV-related cirrhosis. A longer duration of treatment
and/or the use of pegylated IFN, alone or combined with
ribavirin, should be evaluated further.
References
1. Fattovich G, Pantalena M, Zagni I, et al. European Concerted
Action on Viral Hepatitis (EUROHEP). Effect of hepatitis B and C
virus infections on the natural history of compensated cirrhosis:
a cohort study of 297 patients. Am J Gastroenterol 2002;97:
2886 –2895.
2. Fattovich G, Giustina G, Degos F, et al. Morbidity and mortality in
compensated cirrhosis type C: a retrospective follow-up study of
384 patients. Gastroenterology 1997;112:463– 472.
3. Serfaty L, Aumaitre H, Chazouilleres O, et al. Determinants of
outcome of compensated hepatitis C virus-related cirrhosis.
Hepatology 1998;27:1435–1440.
4. Hu KQ, Tong MJ. The long-term outcomes of patients with com-
pensated hepatitis C virus-related cirrhosis and history of paren-
teral exposure in the United States. Hepatology 1999;29:1311–
1316.
5. Benvegnù L, Gios M, Boccato S, et al. Natural history of compen-
sated viral cirrhosis: a prospective study on incidence and hier-
archy of major complications. Gut 2004;37:744 –749.
6. Khan MH, Farell GC, Byth K, et al. Which patients with hepatitis
C develop liver complications? Hepatology 2000;31:513–520.
7. Heathcote EJ. Prevention of hepatitis C virus-related hepatocel-
lular carcinoma. Gastroenterology 2004;127:S294 –S302.
8. Niederau C, Lange S, Heintges T, et al. Prognosis of chronic
hepatitis C: results of a large, prospective cohort study. Hepatol-
ogy 1998;27:1435–1440.
9. Tsukyuma H, Hiyama T, Tanaka S, et al. Risk factors for hepato-
cellular carcinoma among patients with chronic liver disease.
N Engl J Med 1993;328:1797–1801.
10. Baffis V, Shrier I, Sherker AH, et al. Use of interferon for preven-
IFN AND HCV CIRRHOSIS 507
tion of hepatocellular carcinoma in cirrhotic patients with
hepatitis B or hepatitis C virus infection. Ann Intern Med
1999;131:696 –701.
11. Camma C, Giunta M, Andreone P, et al. Interferon and prevention
of hepatocellular carcinoma in viral cirrhosis: an evidence-based
approach. J Hepatol 2001;34:593– 602.
12. Papatheodoridis GV, Papadimitropoulos VC, Hadziyannis SJ.
Effect of interferon therapy on development of hepatocellular
carcinoma in patients with hepatitis C virus infection. Ann Intern
Med 1999;131:696 –701.
13. Nishiguchi S, Kuroki T, Nakatani S, et al. Randomised trial of
effects of interferon-␣ on incidence of hepatocellular carcinoma
in chronic active hepatitis C with cirrhosis. Lancet 1995;346:
1051–1055.
14. Nishiguchi S, Shiomi S, Nakatani S, et al. Prevention of hepato-
cellular carcinoma in patients with chronic active hepatitis C and
cirrhosis. Lancet 2001;357:196 –197.
15. Valla DC, Chevallier M, Marcellin P, et al. Treatment of hepatitis
C virus related cirrhosis: a randomized controlled trial of inter-
feron alfa 2b versus no treatment. Hepatology 1999;29:1870 –
1875.
16. Castilla A, Prieto J, Fausto N. Transforming growth factors b1 and
alfa in chronic liver disease. Effects of interferon therapy. N Engl
J Med 1991;324:933–940.
17. Manabe N, Chevallier M, Chossegros P, et al. Interferon-alfa 2b
therapy reduces liver fibrosis in chronic non-A, non-B hepatitis: a
quantitative histological evolution. Hepatology 1993;18:1344 –
1349.
18. Jimenez SA, Freundlich B, Rosenbloom J. Selective inhibition of
human diploid fibroblast collagen synthesis by interferon. J Clin
Invest 1984;74:1112–1116.
19. Baron S, Tyring SK, Fleischmann WR Jr, et al. The interferons.
Mechanisms of action and clinical applications. JAMA 1991;266:
1375–1383.
20. Knodell RG, Ishak KG, Black WC, et al. Formulation and applica-
tion of a numerical scoring system for assessing histological
activity in asymptomatic chronic active hepatitis. Hepatology
1981;1:431– 435.
21. Tong JM, El-Farra NS, Reikes AR, et al. Clinical outcomes after
transfusion-associated hepatitis C. N Engl J Med 1995;332:
1463–1466.
22. Burroughs AK, Groszmann R, Bosch J, et al. Assessment of
therapeutic benefit of antiviral therapy in chronic hepatitis C: is
hepatic venous pressure gradient a better end point? Gut 2002;
50:425– 427.
23. Garcia-Tsao G, Rodriguez-Perez F, Blei AT. Treatment with inter-
feron reduces portal pressure in patients with chronic hepatitis C.
A randomized placebo trial. Gastroenterology 1996;10:A1193.
24. Wright TL. Treatment of patients with hepatitis C and cirrhosis.
Hepatology 2002;36:S185–S194.
25. Marco V, Craxi A. Antiviral treatment of HCV cirrhosis. Hepatol
Rev 2005;2:35– 40.
Address requests for reprints to: Dr Lawrence Serfaty, Service
d’Hépatologie, Hôpital Saint-Antoine, 184 rue du Faubourg Saint-
Antoine, 75571 Paris Cedex 12, France. e-mail: lawrence.serfaty@
sat.aphp.fr; fax: (33) 1-49-28-21-07.
Supported by the Clinical Research Department, Assistance Pub-
lique-Hôpitaux de Paris (grant AOM97029), and the Fondation de la
Recherche Médicale.
The authors thank Fabrice Carrat and Claude Chastang for their
helpful statistical assistance.