Long-term interleukin 10 therapy in chronic hepatitis C patients has a proviral and anti-inflammatory effect.

Abstract

An imbalance in Th1 and Th2 cytokine production is implicated in disease progression of HCV. Our aim was to determine the effect of IL-10 administration in patients with HCV-related liver disease. Thirty patients with advanced fibrosis who had failed antiviral therapy were enrolled in a 12-month treatment regimen with SQ IL-10 given daily or thrice weekly. Liver biopsies were performed before and after therapy. Serum and PBMC were collected for HCV RNA, ALT, and functional T-cell analysis. IL-10 led to significant improvement in serum ALT (mean ALT: day 0 = 142 +/- 17 vs. month 12 = 75 +/- 10; P <.05). Hepatic inflammation score decreased by at least 2 in 13 of 28 patients (mean decrease from 4.6 +/- 0.3 to 3.7 +/- 0.3, P <.05) and 11 of 28 showed a reduction in fibrosis score (mean change from 5.0 +/- 0.2 to 4.5 +/- 0.3, P <.05). Serum HCV RNA levels increased by 0.5 log during therapy (mean HCV RNA day 0: 12.3 +/- 3.0 Meq/mL; 12 months: 38 Meq/mL; P <.05) and returned to baseline at the end of follow-up (11.0 +/- 2.4 Meq/mL). Five patients developed viral loads of greater than 120 Meq/mL and two of these developed an acute flare in serum ALT. IL-10 caused a decrease in the number of HCV-specific CD4+ and CD8+ IFN-gamma secreting T cells and alterations in PBMC cytokine production towards a Th2 dominant profile. These changes parallel the improvement in ALT and rise in HCV RNA. In conclusion, long-term rIL-10 therapy appears to decrease disease activity, but also leads to increased HCV viral burden via alterations in immunologic viral surveillance.

Full text

Page 1

Long-Term Interleukin 10 Therapy in Chronic Hepatitis
C Patients Has a Proviral and Anti-inflammatory Effect
David R. Nelson,1Zhengkun Tu,1Consuelo Soldevila-Pico,1Manal Abdelmalek,1Haizhen Zhu,2
Yi Ling Xu,1Roniel Cabrera,1Chen Liu,2and Gary L. Davis1
An imbalance in Th1 and Th2 cytokine production is implicated in disease progression of
HCV. Our aim was to determine the effect of IL-10 administration in patients with HCV-
related liver disease. Thirty patients with advanced fibrosis who had failed antiviral therapy
were enrolled in a 12-month treatment regimen with SQ IL-10 given daily or thrice weekly.
Liverbiopsieswereperformedbeforeandaftertherapy.SerumandPBMCwerecollectedfor
HCV RNA, ALT, and functional T-cell analysis. IL-10 led to significant improvement in
serum ALT (mean ALT: day 0 ? 142 ? 17 vs. month 12 ? 75 ? 10; P < .05). Hepatic
inflammationscoredecreasedbyatleast2in13of28patients(meandecreasefrom4.6?0.3
to 3.7 ? 0.3, P < .05) and 11 of 28 showed a reduction in fibrosis score (mean change from
5.0?0.2to4.5?0.3,P<.05).SerumHCVRNAlevelsincreasedby0.5logduringtherapy
(mean HCV RNA day 0: 12.3 ? 3.0 Meq/mL; 12 months: 38 Meq/mL; P < .05) and
returned to baseline at the end of follow-up (11.0 ? 2.4 Meq/mL). Five patients developed
viral loads of greater than 120 Meq/mL and two of these developed an acute flare in serum
ALT. IL-10 caused a decrease in the number of HCV-specific CD4?and CD8?IFN-?
secreting T cells and alterations in PBMC cytokine production towards a Th2 dominant
profile. These changes parallel the improvement in ALT and rise in HCV RNA. In conclu-
sion, long-term rIL-10 therapy appears to decrease disease activity, but also leads to in-
creased HCV viral burden via alterations in immunologic viral surveillance. (HEPATOLOGY
2003;38:859-868.)
T
infected. Although the pathogenesis of hepatocellular in-
jury in HCV infection is not fully understood, there is
increasing evidence to suggest immune-mediated mecha-
nisms.1CD4?and CD8?T cells, as well as their inflam-
matory and regulatory cytokines, have been implicated in
he importance of hepatitis C virus (HCV) infec-
tion lies in its ability to cause insidious and pro-
gressive liver damage in the majority of those
both the hepatocellular damage and the perpetuation of
chronic HCV infection. CD4?cell responses are polar-
izedintoTh1andTh2types.Th1cellssecreteinterleukin
2 (IL-2) and interferon gamma (IFN-?), which are re-
quiredforhostantiviralimmuneresponses.Th2cellspro-
duce IL-4, IL-5, IL-13, and IL-10, which facilitate
antibody production and limit Th1 response. Patients
with chronic HCV infection have an activated T-cell re-
sponse cytokine pattern, with elevated levels of serum
IL-2, IL-4, IL-10, tumor necrosis factor ?, and IFN-?.
Patients with severe chronic hepatitis C or cirrhosis have
an enhanced expression of both IFN-? and IL-2 com-
pared with healthy controls and those with mild disease.2
In contrast, controls and patients with milder inflamma-
tion have a higher expression of IL-10 compared with
patients with overt liver disease. In addition, the intrahe-
patic messenger RNA levels of IFN-? and IL-2 correlate
with fibrosis and portal inflammation, suggesting that
Th1 cytokines might play a role in mediating hepatocel-
lular damage. Thus, an imbalance in Th1 and Th2 cyto-
kine production is implicated in disease progression of
chronic HCV.
Abbreviations: HCV, hepatitis C virus; IL-2, inteleukin 2; IFN-?, interferon
gamma; ALT, alanine aminotransferase; PBMC, peripheral blood mononuclear
cell; CTL, cytotoxic T lymphocyte; MHC, major histocompatibility complex; AFP,
?- fetoprotein.
From the Departments of1Medicine and2Pathology, University of Florida,
Gainesville, FL.
Received February 10, 2003; accepted July 30, 2003.
D.R.N.wassupportedbyNIHgrantsDK2595andHL64817andGCRCgrant
RR00082. Research support also provided by Schering Plough Research Institute
and Integrated Therapeutics Group.
G.L.D. now resides at Baylor University Medical Center, Dallas, TX.
Address reprint requests to: David R. Nelson, M.D., Section of Hepatobiliary
Diseases, University of Florida, 1600 S.W. Archer Rd., P.O. Box 100214, Gaines-
ville, FL 32610-0214. E-mail: nelsodr@medicine.ufl.edu; fax: 352-392-7393.
Copyright © 2003 by the American Association for the Study of Liver Diseases.
0270-9139/03/3804-0011$30.00/0
doi:10.1053/jhep.2003.50427
859

Page 2

We hypothesized that in vivo administration of IL-10
may tilt the balance away from Th1 cytokine dominance,
thereby reducing the hepatocellular injury that character-
izes HCV infection. Our aim was to determine the im-
mune-modulating properties of IL-10 in patients with
progressiveliverdisease,withthehopeofinterruptingthe
progression of liver damage. A previous pilot study had
suggestedthatIL-10waswelltolerated,normalizedserum
alanine aminotransferase (ALT), decreased hepatic in-
flammation, and reduced liver fibrosis in patients with
chronic hepatitis C who had not responded to previous
IFN-based therapy.3These data suggested that IL-10 has
an important role in chronic inflammation and fibrogen-
esis in HCV and that IL-10 might be useful in the treat-
ment of chronic hepatitis C and other chronic liver
diseases.
Patients and Methods
Patients. To assess the effect of rIL-10 in chronic
HCV, adult subjects with extensive fibrosis or cirrhosis
who had previously failed to respond to IFN-based treat-
ment were enrolled. Nonresponse to IFN-based treat-
mentwasdefinedasdetectableHCVRNAattheendof6
months of therapy. Exclusion criteria included decom-
pensated cirrhosis, hemoglobin less than 12 g/dL, white
blood count less than 3,500 per cubic milliliter, platelets
less than 75,000 per cubic milliliter, human immunode-
ficiencyvirusinfection,pregnancy,andseverenonhepatic
illnesses.
Study Design and Treatment Regimens. The study
was a randomized, 3-dose trial. Thirty subjects were ran-
domized (1:1:1; 10 subjects per group) to receive recom-
binant human IL-10 (rIL-10; Schering Plough Research
Institute, Kenilworth, NJ) subcutaneously at a daily or
thrice weekly dose of 8 ?g/kg or a thrice weekly dose of 4
?g/kg for at least 12 months. Each subject had a liver
biopsy before treatment and at the end of the 12-month
treatmentperiod.Subjectswhoshowednohistologicpro-
gression during therapy were offered the opportunity to
remain on rIL-10 (8 ?g/kg thrice weekly dosing) for an
additional 12 months. Participants were monitored as
outpatients on a weekly basis during therapy, and 2 and 6
months after treatment. Serum HCV RNA level was as-
sessed by branched chain DNA assay (Quantiplex HCV
RNA version 2.0; Bayer Diagnostics, Emeryville, CA)
with a lower limit of detection of 200,000 genome equiv-
alents(Eq)/mL.Negativesampleswerereevaluatedwitha
more sensitive reverse-transcription “nested” polymerase
chain reaction, as previously described.4
ThisstudywasconductedbetweenFebruary1999and
September 2000. The protocol was approved by the In-
stitutional Review Committee and the Clinical Research
Center Scientific Advisory Board, and all study subjects
provided written informed consent.
Immunologic Assays. Peripheral blood mononuclear
cells (PBMCs) were evaluated for their specific response
to HCV antigens (core, NS3, NS4, NS5) and HCV pep-
tides (panel of HLA A2 epitopes) by IFN-? ELISPOT,
tetramer, and3H proliferation assays. PBMCs were iso-
lated from 30 mL of fresh heparinized blood by Ficoll-
Hypaque gradients through centrifugation at 1,000g
followed by 3 washes in phosphate-buffered saline. Typ-
ical yield was 30 to 60 ? 106cells. PBMCs were imme-
diately frozen and stored in liquid nitrogen until T-cell
analysis was performed. The extent of immunologic anal-
ysis of individual patients was dictated by the number of
cells available at various time points.
HCV Antigens and Peptides. The following HCV
antigenswereobtainedfromcommercialsources(Chiron
CorporationLtd.,Emeryville,CA):core(c22,aminoacid
2-120), NS3 (c33, amino acid 1192-1457), NS4 (c100,
amino acid 1569-1931), or NS5 (amino acid 1569-
1931).Apanelofpeptidesthatrepresentpreviouslyiden-
tified HLA A2-restricted HCV cytotoxic T lymphocyte
(CTL) epitopes were selected for this study.5Five of these
peptides that are highly conserved across genotype and
express cross-genotype CTL reactivity were targeted for
ELISPOT assays and construction of major histocompat-
ibility complex (MHC) class I tetramers.6These peptides
are (1) core 35, (2) core 131, (3) NS3 1073, (4) NS3
1406, and (5) NS5B 2727. In addition, influenza flu ma-
trix peptide (GILGFVFTL) was used as a positive control
peptide and a human immunodeficiency virus–specific
peptide was used as a negative control. All HCV peptides
were synthesized by the University of Florida Peptide
Synthesis Core Laboratory using Fmoc solid phase meth-
ods. All peptides contained a free amine N-termini and
free acid C-termini. Lyophilized peptides were reconsti-
tuted at 20 mg/mL in dimethyl sulfoxide and diluted to 1
mg/mL with RPMI 1640 medium.
Proliferation Assays. PBMCs (2.5 ? 105) were
plated in triplicate in 96-well plates and stimulated with
either HCV antigen at 5 ?g/mL, or PHA (Sigma, St
Louis, MO). After 4 days, 1 ?Ci of [3H] thymidine (3H-
TdR; Amersham, Arlington Heights, IL) was added to
eachwell.Wellswereharvestedontoglassfiberfilterpaper
18 hours later and incorporated nuclide was measured by
liquid scintillation counting. The mean counts per
minute (cpm) of triplicate cultured wells were calculated
and proliferation was considered positive when the stim-
ulation index (stimulation index ? cpm incorporated in
response to antigen divided by cpm incorporated in the
absence of antigen) was greater than 3.
860NELSON ET AL.HEPATOLOGY, October 2003

Page 3

Cytokine Assays. Separate cultures of 106PBMCs in-
cubated in the absence or presence of 5 ?g of HCV anti-
gen in 1 mL of complete medium were utilized to
determinecytokineproductioninculturesupernatantsof
those who had a stimulation index of greater than 3. Ali-
quots of the supernatant were removed after 4 days and
assayed for IL-4, IL-10, and IFN-? by ELISA (per the
manufacturer’s instructions).
ELISPOT Assay. A total of 2.5 ? 105PBMCs to-
gether with the respective individual HCV peptides (20
?g/mL), influenza peptide, antigens, or PHA (5 ?g/mL)
were evaluated by IFN-? ELISPOT assay by standard
methodology.7Allresultsreportedareexpressedasmeans
oftriplicatesinthepresenceorabsenceofsinglepeptides/
antigens. For statistical analysis, the mean number of
spots of 3 wells without peptide/antigen (control) was
subtracted from the mean of 3 wells obtained in the pres-
ence of HCV peptide/antigen. Specific spots greater than
themeancontrol(2standarddeviations)forthepeptides/
antigens in normal donors (a panel of 5 healthy controls
were evaluated) were required to be considered positive.
Tests with greater than 10 spots in the control wells indi-
catedactivatedTcellsandnewaliquotswereretestedafter
1weekinculturetoallowthemtoreturntoarestingstate.
Toexcludeartifacts,onlytestswithinthe95%confidence
interval for negative and positive controls were consid-
ered.
Class I Peptide Tetramers. PE-labeled HLA-A2 tet-
ramercomplexesweresynthesizedaspreviouslydescribed
for testing in select HLA A2 patients.8Briefly, a 15-aa Bir
A substrate peptide for Bir A–dependent biotinylation
has been engineered onto the COOH terminus of HLA-
A*0201. The A2-BirA substrate peptide fusion protein
and human ?2-microglobulin were expressed in Esche-
richia coli and folded in vitro with the specific peptide
ligand. The properly folded MHC-peptide complexes
were extensively purified and biotinylated on a single ly-
sine within the BirA substrate peptide by using the BirA
enzyme (Avidity; L.L.C., Denver, CO). Tetramers were
producedbymixingthebiotinylatedMHC-peptidecom-
plexes with phycoerythrin-streptavidin (Leinco Technol-
ogies Inc., St. Louis, MO) at a molar ratio of 4:1. The
detectionlimitwas0.02%fortheNS3-1073,NS4-1406,
and NS5-2727 tetramers, while it was 0.03% for the
Core-35 and -131 tetramers. This was determined using
the background signal plus 3 SDs after staining PBMCs
from HLA2?, HCV-negative healthy controls (n ? 5).
Flow Cytometry. For immunofluorescence staining,
cells were washed and stained for 20 minutes at 4°C with
the optimal dilution of each antibody. Cells were washed
again and analyzed at the University of Florida Flow Cy-
tometry Core Laboratory using FACSort and FACScan
flow cytometers (Becton Dickinson, San Jose, CA). Anti-
CD8, anti-CD4, anti-CD45 RO, anti-45 RA, anti-
CD25, anti-CD62L, and anti-CD38 were obtained from
Beckton Dickinson and used per the manufacturer’s in-
structions.
IL-10 Effect on HCV Replication in HCV Replicon
Cell Culture System. The subgenomic replicon cell line
FCA1(agiftofDr.Seeger,FoxChaseCancerCenter)was
used. The cells were maintained in Dulbecco’s modified
Eagle’s medium (Gibco-BRL) supplemented with 10%
fetal bovine serum, L-glutamine, nonessential amino ac-
ids, penicillin, and streptomycin. For cells with HCV
replicons, the culture medium was additionally supple-
mented with 500 ?g/mL of G418. HCV replicon–con-
tainingcellswerecoculturedwithorwithoutthepresence
of IL-10 (10 ng/mL and 20 ng/mL) for 48 hours. In
addition, a long-term culture (2 months) was established
withIL-10inthemedium.IFN-?2b(50U/mL)wasused
as a positive control. Effect of IL-10 on HCV RNA was
determined by the methodology previously described.9
Statistical Analysis. Adverse events were analyzed on
anintention-to-treatbasis.Analysisoftreatmentresponse
included the 30 randomized subjects who completed the
24 weeks in the study and therefore had pre- and end-of-
treatment samples for comparison. Baseline characteris-
tics of the 3 dose groups were compared using the ?2or
Wilcoxon nonparametric rank sum test. The 3 dose
groups were analyzed together for changes in ALT, HCV
RNA,histology,andserumcytokine/fibrosismarkersbya
paired sample t test. T-cell responses at each time point
were compared with pretreatment values by using non-
parametric Wilcoxon ranked sign test. All P values were
2-tailed.
Results
Patient Enrollment and Treatment Duration.
Thirty patients (17 men, 13 women) were randomized to
1 of 3 doses of rIL-10 with the intent of 12 months of
therapy. Previous treatment regimens included IFN-?2b
monotherapy (n ? 9) and IFN-?2b in combination with
ribavirin (n ? 21). Most patients were genotype 1 (n ?
27), while 3 patients were genotype 2. Serum ALT was
elevated in most patients with a mean ALT of 142 ? 17
(range, 34 to 443). Patients had well-advanced fibrosis
(mean fibrosis stage 5.0 ? 0.2; range, 3 to 6), active hep-
atitis (mean hepatic inflammation 4.6 ? 0.3; range, 2 to
7), and high viral loads (mean serum HCV RNA level
12.3 ? 3.0 Meq/mL).
Twenty-eightpatientscompleted12monthsofrIL-10
therapy and 25 met criteria (improved ALT and no his-
tologic progression) to continue with an additional 12
HEPATOLOGY, Vol. 38, No. 4, 2003 NELSON ET AL.861

Page 4

months of IL-10 therapy. Twenty-four patients signed an
informedconsenttocontinuewiththeextensionprotocol
and were between 2 and 12 weeks on additional rIL-10
treatmentwhenwestoppedfurtherdosingduetoconcern
regarding long-term proviral effects of IL-10. Two pa-
tientsdevelopedviralloadsabove120millioninconjunc-
tion with marked elevation of ALT. Figure 1 shows the
altered protocol for the overall group, including those
who continued beyond 12 months.
Serum ALT. Twenty-eight patients completed at
least 12 months of rIL-10 therapy. IL-10 was well toler-
ated during the initial 12-month treatment regimen and
led to significant improvement in serum ALT (all treat-
ment groups at day 0: mean ALT 142 ? 17 and median
135; month 6: mean 62 ? 6 and median 57; month 12:
mean 75 ? 10 and median 73, and end of follow-up:
mean 110 ? 16 and median 104; P ? .05). This effect
wasnoticedacrossall3treatmentgroupsascanbeseenin
Fig.2.Eighteenofthe28patientswhocompletedtherapy
maintained a normal ALT at the 12-month period. Two
patients showed a late flare of ALT between 12 to 13
months that led to the discontinuation of the extension
study. ALT returned to baseline after rIL-10 was discon-
tinued.
SerumHCVRNA. MeanserumHCVRNAlevelsfor
all groups increased by 0.5 log during the 12 months of
therapy (mean HCV RNA levels at day 0: 12.3 ? 3.0
Meq/mL;6months:25.6?6.6Meq/mL;12months:38
Meq/mL; P ? .05) and returned to baseline at the end of
follow-up (11.0 ? 2.4 Meq/mL). This effect was seen at
all treatment doses (see Fig. 3). Five patients developed
viral levels of greater than 100 Meq/mL, and 2 of these
subjects developed an acute flare in serum ALT (10 ?
baseline) that was then associated with marked reduction
inviralload(greaterthan2logdrop),suggestingimmune
activation. With discontinuation of rIL-10 therapy, the
viral levels and serum ALT of these two patients returned
to baseline.
Liver Histology. Paired pre- and 12-month liver bi-
opsies are shown in Fig. 4. Fourteen patients had im-
provement in inflammatory scores, with 13 showing an
improvementof?2(meandecreaseforallpairedsamples
from 4.5 ? 0.3 to 3.7 ? 0.3, P ? .05). Four patients had
an increase of inflammatory scores from baseline. Two of
these patients (HAI increase by 3) correspond to the 2
who had a marked ALT elevation at month 12, which led
to discontinuation of the study. Eleven of 28 patients had
a reduction in fibrosis score, 14 had no change, and 3 had
an increase of 1 (mean change from 5.0 ? 0.3 to 4.5 ?
Fig. 1. IL-10 study design. Under an open label, randomized design
patients were treated with rIL-10 8 ?g/mL subcutaneously on either a
daily or thrice weekly schedule or 4 ?g/kg thrice weekly. Twenty-four of
the 30 patients dosed were continued beyond the initial 12-month
dosing (2-12 weeks).
Fig. 2. Mean serum ALT for each treatment group during 12 months
of rIL-10 therapy and follow-up. Y-axis represents ALT in IU/mL.
Fig. 3. Mean serum HCV RNA for each treatment group during 12
months of IL-10 therapy and follow-up. Y-axis represents HCV RNA in
Meq/mL.
862NELSON ET AL.HEPATOLOGY, October 2003

Page 5

0.3, P ? .05). No patient with a pretreatment fibrosis
scoreof6showedanyimprovement(i.e.,allimprovement
was observed in noncirrhotic patients).
Serum ?-Fetoprotein. As a group, the mean serum
?-fetoprotein (AFP) fell from 13.3 ? 3.5 to 6.2 ? 1.2
(P ? .02) by 6 months of rIL-10 therapy and remained
normal at 12 months, 6.3 ? 1.3 (P ? .02). One patient
had a dramatic rise from an elevated pretreatment level
and was found to have a hepatocellular carcinoma (see
below).AllotherabnormalAFPsdecreasedornormalized
by6monthsoftherapyandmaintainedthatlevelthrough
the end of therapy.
Adverse Events. Safety analysis was performed on all
30 patients who received study medication. Symptoms
experiencedduringtreatmentarelistedinTable1.rIL-10
was well tolerated with minimal side effects. The most
frequent clinical adverse events seen during treatment
with rIL-10 were headache, fatigue, and some myalgias/
fatigue.Thesewerenotdoselimitinginanypatient.Mild
anemia occurred during the first 2 months of therapy in
most subjects with an average decrease in hemoglobin of
1.6 ? 0.5 g/dL. Hemoglobin levels stabilized without
dose reduction in any patients. The platelet count was
unchanged during 12 months of therapy (180,000 vs.
179,000/dL; P ? not significant), and there was no effect
on neutrophil counts.
No patients required dose adjustments but 2 patients
withdrew from the initial 12-month study. One patient
experienced a serious adverse event. A hepatocellular car-
cinomawasdiscoveredduetoarisingAFP4monthsafter
study entry. Abdominal imaging revealed a 3-cm hepato-
cellularcarcinomaintherightlobe.Thiscirrhoticpatient
had an elevation of AFP (78) and a negative abdominal
computed tomography scan prior to entry into the study.
The patient underwent successful cadaveric liver trans-
plantation at the University of Florida 2 months later. Of
note, he is now undergoing IFN-based antiviral therapy
for recurrent HCV with extensive bridging fibrosis. The
other patient was withdrawn due to noncompliance that
developed when the patient noted a decrease in his sexual
drive that he thought was rIL-10 related.
Immunologic. IL-10 caused a decrease (2-fold) in the
number of HCV-specific CD4?IFN-?–secreting T cells
by month 12 of treatment (Fig. 5). Of interest, the pro-
liferation assays showed no overall changes for the group,
either when evaluated as mean responses or number of
patientswithsignificantresponses.Proliferativeresponses
to HCV-specific antigens could be detected before treat-
ment in 16 of 30 patients, all at a low level (mean stimu-
lation index 4.6, range 3-8). As shown in Fig. 6, these
responses did not undergo any significant changes during
rIL-10 therapy. But, a phenotypic change in PBMC cy-
tokine production upon antigen stimulation from Th0/
Th1 to a Th2 dominant profile (high IL-10 and low
Fig. 4. Changes in Ishak (A) histologic activity index and (B) fibrosis scores between the first and second liver biopsies.
Table 1. Adverse Events Summary
Symptom
4 ?g
TIW
n ? 10
8 ?g
TIW
n ? 10
8 ?g QD
n ? 10
All Patients
n ? 30
Headache (mild)
Fatigue
Arthralgia/myalgia
Abdominal pain
Depression
GI symptoms
Sleep disturbance
7
7
3
3
2
2
2
6
5
3
2
1
1
0
7
3
2
3
3
1
1
20 (66%)
15 (50%)
8 (27%)
8 (27%)
6 (20%)
4 (13%)
3 (10%)
Abbreviations: TIW, three times a week; QD, everyday; GI, gastrointestinal.
HEPATOLOGY, Vol. 38, No. 4, 2003NELSON ET AL.863

Page 6

IFN-?) was noted (see Fig. 7). These changes appear to
parallel the improvement in ALT and rise in HCV RNA.
In 8 HLA A2 subjects tested, the number of IFN-?–
secreting HCV-specific CD8?cells decreased during
therapy (Fig. 8). Tetramer analysis was performed in lim-
ited remaining samples for 3 HLA A2–positive patients.
Where a tetramer-positive population was identified, no
changes in the frequency of HCV-specific CD8?cells
during rIL-10 therapy were observed (Table 2). In the
setting of decreased IFN-? response to HCV peptide
stimulation, this suggests impaired CD8?effector func-
tion.Limitednumberofcellsdidnotallowamoreformal
evaluation of this phenomena. The inhibitory effect on
viral-specific immune responses was reversible with either
discontinuationofrIL-10,orasin2ofthe5patientswith
viral load greater than 100 Meq, it occurred spontane-
ouslywithanALTflare.Importantly,thisacuteALTflare
corresponded with a 2 log decrease in HCV RNA (Fig.
9A) and a dramatic 10- to 100-fold rise in HCV-specific
CD4?(Fig. 9B) and CD8?IFN-? activity (Fig. 9C) that
was targeted toward NS3, NS4, and NS5.
The overall dynamics of the CD4 and CD8 cells dur-
ing rIL-10 therapy are shown in Table 2. There was no
significant change in percent of total CD4?or CD8?
cells, but the number of CD4?/CD25?regulatory cells
showed a decline during the first 12 months of therapy
followed by a return to baseline during the follow-up
period. The majority of CD4?CD25?T cells isolated
prior to therapy consisted primarily of CD45RO?/
CD28?27?cells, resembling a memory T-cell pheno-
type. Of note, rIL-10 therapy was associated with a sig-
nificant decease in CD45RO expression and an increase
in CD62L expression.
IL-10 Effect on HCV Replication in HCV Replicon
Cell Culture System. To determine whether IL-10 has a
directroleinHCVviralreplication,wetestedtheeffectof
rIL-10 using HCV replicon cell line FCA1.10IL-10 was
usedat10ng/mLand20ng/mLincellculture.Asshown
Fig. 7. Cytokine release from PBMC upon recognition of HCV antigens
at day 0, month 6, and month 12 of rIL-10 therapy and after the
follow-up period (with standard error of the mean).
Fig. 8. HCV-specific CD8?activity to HCV peptides as measured by
IFN-? ELISPOT assay in 8 HLA A2?patients. Number of spots in wells per
2.5 ? 105PBMCs with HCV peptide minus control wells at months 0, 6,
and 12 of therapy and for follow-up. *Significant change (P ? .05) from
baseline value.
Fig. 5. HCV-specific CD4?activity to HCV antigens as measured by
IFN-? ELISPOT assay. Number of spots in wells per 2.5 ? 105PBMCs
with HCV antigen minus control wells at months 0, 6, and 12 of therapy
and for follow-up. *Significant change (P ? .05) from baseline value.
Fig. 6. Number of patients who had detectable proliferative T-cell
responses to HCV-specific antigens derived from core, NS3, NS4, and
NS5 during IL-10 therapy and follow-up.
864 NELSON ET AL.HEPATOLOGY, October 2003

Page 7

in Fig. 8, while IFN decreased viral RNA replication,
IL-10 did not have a significant effect on viral RNA rep-
lication after 48 hours of incubation.
Discussion
IL-10 is an important cytokine in the pathogenesis of
both infectious and inflammatory processes. The central
and necessary role of IL-10 in protecting against severe or
inflammatorypathologyhasbeenclearlyshowninmodels
of experimental infection with intracellular pathogens us-
ing IL-10 genetically deficient mice.11,12In the process of
limiting inflammation, IL-10 likely plays a role in pre-
venting eradication in chronic infection by intracellular
pathogens. But the lack of a mouse model for HCV in-
fection has limited our understanding of the role of this
important cytokine in HCV infection. The results of this
trial provide unique insight into the role of IL-10 in the
pathogenesisofchronicHCVinfection.Itiscurrentlyfelt
that the hepatocellular injury seen in HCV infection re-
sults from host immune-mediated responses to infection.
Available data derived from studies on infected humans
and chimpanzees suggest that control of HCV infection
inthenormalhostisdependentonthevigor,breadth,and
the quality of the early cell-mediated immune response.13
Quantitatively, individuals who have a more vigorous
HCV-specific CD4?and CD8?response are more likely
to clear HCV during acute infection or clear chronic in-
fectionafterinterferontherapy.14,15TheseTcellscanalso
regulatetheintensityoftheimmuneresponsethroughthe
productionofamplifyingcytokines.Inacuteself-limiting
HCVinfection,circulatinghelperT-cellcloneshavebeen
shown to predominantly produce IFN-?, suggesting a
Th1-like pattern.16In contrast, those who develop a
chronicinfectionhavealimitedTh1responseandamore
exuberant Th2 response.17It has therefore been specu-
lated that this Th2 type of response may impair the pa-
Table 2. The Dynamics of the Frequency and Phenotype T Cells During rIL-10 Treatment
0 Mo6 Mo12 Mo 14 Mo 17 Mo19 Mo Note
Frequency
CD8?
CD4?
CD8?/CD4?
CD4?/CD25?
13.80
29.51
0.29
1.90
8.37
10.43
29.29
0.22
1.38
5.12
11.45
29.21
0.23
1.22*
4.93*
13.36
33.26
0.38
2.29
8.14
16.38
34.39
0.29
2.85
8.24
13.18
34.34
0.31
2.97
8.62
%lymphcytes
%lymphcytes
%lymphcytes
%lymphcytes
%CD4?
CD8?Tetramer
P1
P2
P3
P7
Phenotype
CD8?/CD69?
CD8?/CD25?
CD27?/CD4?
0.04
0.08
0.10
0.04
0.06
0.10
0.12
0.06
0.05
0.12
0.11
0.03
0.06
0.08
0.08
0.04
0.06
0.11
24.77
83.94
0.00
0.03
19.46
65.94
0.01
12.15
41.17
0.04
0.09
22.87
78.11
0.00
0.04
18.43
62.92
0.01
13.13
44.83
0.03
0.07
18.59
63.64
0.00
0.06
18.12
62.03
0.01
12.08
41.35
0.05
0.10
25.02
75.23
0.00
0.05
19.56
58.81
0.00
22.18
66.69
0.03
0.21
29.75
86.52
0.00
0.05
23.36
67.93
0.01
25.15
73.13
0.07
0.32
26.17
76.19
0.00
0.05
17.48
50.90
0.01
24.94
72.63
%lymphcytes
%lymphcytes
%lymphcytes
% CD4?
%lymphcytes
%lymphcytes
%lymphcytes
%CD4?
%lymphcytes
%lymphcytes
%CD4?
CD28?/CD4?
CD45RA?/CD4?
CD45RO?/CD4?
CD38?/CD4?
CD62L?/CD4?
Phenotype CD4?/CD25?
CD28?/CD27?
CD28?/CD27?
CD28?/CD27?
CD28?/CD27?
CD45RA?/CD45RO?
CD45RA?/CD45RO?
CD45RA?/CD45RO?
CD45RA?/CD45RO?
CD38?/CD62L?(Ts)
CD38?/CD62L?(Ts)
CD38?/CD62L?(Ts)
CD38?/CD62L?(Ts)
0.07
0.00
82.75
17.02
0.16
0.32
88.06
11.48
0.00
0.04
47.04
52.92
0.04
0.00
79.24
20.73
0.11
0.11
79.44
20.17
0.13
0.00
64.72
35.15
0.20
0.00
85.01
14.76
0.28
0.27
72.57
26.88
0.00
0.06
60.73
39.21
0.05
0.00
87.92
12.04
0.19
0.20
75.07
24.54
0.05
0.00
69.19
30.76
0.00
0.00
81.46
18.54
0.06
0.01
78.12
21.81
0.00
0.02
82.62*
17.36*
0.01
0.00
89.55
10.44
0.16
0.27
74.16
25.42
0.00
0.02
79.88
20.00
%CD4?/CD25?
%CD4?/CD25?
%CD4?/CD25?
%CD4?/CD25?
%CD4?/CD25?
%CD4?/CD25?
%CD4?/CD25?
%CD4?/CD25?
%CD4?/CD25?
%CD4?/CD25?
%CD4?/CD25?
%CD4?/CD25?
*Significant change from baseline value (P ? .05, paired t test).
HEPATOLOGY, Vol. 38, No. 4, 2003 NELSON ET AL. 865

Page 8

tient’s ability to clear the virus. Antiviral therapy may
counteract this Th2 effect. IFN stimulates an HCV-spe-
cific CD4?and CD8?response, while ribavirin sup-
presses the Th2 response.18Patients who achieve a
sustained response to treatment have an active pre-
treatment T-cell response, which is enhanced during
therapy.19,20In contrast, those patients with a low
pretreatment T-cell response before therapy are unlikely
toaugmenttheirresponseduringtherapy,andtheythere-
fore do not clear the virus.21IL-10 probably plays a piv-
otal roleinthisantiviral
Administration of neutralizing monoclonal antibodies to
anti–IL-10 enhances in vitro IFN-? production in HCV
patients.22Furthermore, IL-10 levels differ widely be-
tween individuals, possibly due to polymorphisms in the
promoter region of the IL-10 gene.23For example, 3 sin-
glenucleotidepolymorphismsinthepromoteralterIL-10
expressionandappeartoaffectresponsetoIFNtherapy.24
Low-levels of IL-10 are associated with better response to
IFN therapy.25
In our study, long-term therapy with recombinant
IL-10decreasesdiseaseactivityasmeasuredbyinflamma-
tory activity within the liver and serum ALT levels, but
also leads to increased HCV viral levels via alterations in
the host’s immunologic viral surveillance. The mecha-
nismsresponsiblefortheseeffectsarelikelymultifactorial.
IL-10 is known to suppress the production of proinflam-
matorycytokines,suchastumornecrosisfactor?,IFN-?,
and IL-12 and has an overall suppressive effect on the
generation of a Th1 response.26,27We have shown here
that exogenous IL-10 administration suppresses IFN-?
production and leads to an increase in IL-10 production
from responding peripheral T cells. If a similar phenom-
enon occurred in the liver infiltrating T cells, this would
explain the decrease in hepatic inflammation, as reflected
in the decrease in HAI scores. It is also interesting to note
that IL-10 can inhibit proinflammatory activation of en-
dothelial cells, which play a crucial role in recruiting leu-
kocytes to sites of injury or infection (i.e., inhibit T-cell
recruitment to the liver parenchyma).28During longitu-
dinal analysis, ALT and viral level appear to reflect the
responsetotherapy.
Fig. 9. (A) Mean serum ALT and HCV RNA for the subject who
developed an ALT flare while on rIL-10 therapy. Y-axis represents ALT in
IU/mL and HCV RNA in Meq/mL. (B) HCV-specific CD4?activity to HCV
antigens in the same subject as measured by IFN-? ELISPOT assay. Log10
scale of number of spots in wells per 1 ? 106PBMCs with HCV antigen
minus control wells at months 0, 6, and 12 of therapy and for follow-up
months 13, 14, and 16. (C) HCV-specific CD8?activity to HCV peptides
in the same subject as measured by IFN-? ELISPOT assay. Log10scale of
number of spots per 1 ? 106PBMCs in wells with HCV peptide minus
control wells at months 0, 6, and 12 of therapy and for follow-up months
13, 14, and 16.
Fig. 10. Effect of IL-10 on HCV RNA replication. Northern blot analysis
of FCA1 cells treated with IL-10 and IFN. (Lane 1) FCA1 cell control;
(lane 2) FCA1 cells treated with 50 U/mL IFN; (lane 3) FCA1 cells
treated with 10 ng/mL IL-10; (lane 4) FCA1 cells treated with 20 ng/mL;
(lane 5) Huh-7 cell control.
866NELSON ET AL.HEPATOLOGY, October 2003

Page 9

relative activity of the IFN-? HCV-specific CD4?T-cell
responsetomainlynonstructuralproteins.rIL-10admin-
istration decreased the number of HCV-specific CD4?
and CD8?T cells secreting IFN-? within the peripheral
blood. This may be related to an effect on either virus-
specific T-cell frequency or the function of viral-specific
cells. However, the lack of significant changes in the pro-
liferation assays, along with the Th1 to Th2 dominant
cytokine profile change, suggests that rIL-10 altered the
phenotypicresponseofHCV-specificCD4?cells.Infact,
tetramer analysis of HCV-specific CD8?cells in 3 HLA
A2 subjects did not show a decrease in frequency of
HCV-specific CD8?T cells, despite a decrease in the
IFN-? responses. This would suggest that the HCV-spe-
cific CD8?cells were present but had impaired effector
function. Recently, HCV-specific CD8?T cells have
been described in chronic HCV patients that display im-
paired proliferative activity and IFN-? production.29
Whether this reduction in effector CTL function led to
theincreasedvirallevelsobserved,or,conversely,whether
the increased HCV viral antigen production led to the
induction of anergy/tolerance, is still uncertain. To eval-
uatethepossibilitythatIL-10mayhavedirectlyenhanced
viral replication and viral antigen production, we cocul-
tured rIL-10 in the HCV replicon system. In this in vitro
model, short-term culture was not associated with in-
creased viral replication. While not conclusive, this data
suggests that the proviral effect of IL-10 is immunologi-
cally mediated.
We have recently determined that CD4?CD25?T
regulatory cells play an important role in HCV persis-
tence (manuscript submitted). These regulatory T cells
have a major role in controlling the immune response to
self and foreign antigen and are able to suppress CD8?
T-cell proliferation and IFN-? production induced by
antigen-specificstimuli.30,31Thesecellsinhibitbothnaı ¨ve
and memory T cells in an antigen-specific manner via a
mechanism that is partially dependent on the production
of immunoregulatory cytokines IL-10 and transforming
growth factor ?.32We therefore wanted to evaluate the
effect of exogenous rIL-10 on this regulatory subset. We
found that rIL-10 administration led to an initial decline
in CD4?CD25?cells that then began to increase back to
baselineovertime.Sincetheseregulatorycellsarethought
tobeessentialforinductionoftolerance,theirdecreaseon
therapywouldhavebeenpredictedtoleadtoanenhance-
mentofTh1responses,whichwasnotseen.Thisparadox
might be explained by the fact that others have previously
shown that T regulatory cells are constitutively capable of
responding to cytokines and that IL-10 may have a sup-
pressive autocrine effect.33Our study clearly indicates
that this poorly understood population of cells is a dy-
namic subset that was influenced by rIL-10 administra-
tion (directly or indirectly via increased HCV RNA
levels).
Lastly, IL-10 may play a critical role in tumor surveil-
lance. One patient with cirrhosis and a baseline elevated
AFP had a hepatocellular carcinoma detected within 4
monthsofIL-10therapy,despiteabaselinenegativecom-
puted tomography scan. Of note, all other patients that
had abnormal AFPs at baseline had a significant decrease
in levels during therapy. It is our hypothesis that this
patient had a small undetectable tumor at entry into the
study and that the IL-10 administration lead to a more
rapid tumor growth. Recent investigation has shown that
IL-10playsamajorroleinsuppressingdendriticcellfunc-
tion that is critical to a tumor-specific immune response
and that IL-10 can result in an impaired T-cell immune
response to tumor.34The role of this important cytokine
in tumor development associated with chronic HCV and
cirrhosis remains to be determined.
Insummary,ourdatasuggeststhatrecombinantIL-10
improves hepatic inflammation but leads to increases in
viral load via immune modulation. These data emphasize
the critical importance of Th1/Th2 cytokines and the
interplay with HCV-specific T-cell response in both viral
control and disease progression. Our findings also high-
lightthecomplexityofthehostimmuneresponseandthe
limitations in our current understanding of the mecha-
nisms by which cytokine therapies effect responses in
chronic viral-induced liver disease. Multiple trials aimed
at blocking proinflammatory cytokines and their media-
tors are currently in the planning phase. Our data would
send a note of caution to those interested in long-term
anti-cytokinetherapy,especiallyincirrhoticpatients,and
would suggest that these trials be accompanied by close
viral and immunologic analysis.
References
1. Nelson DR. The Immunopathogenesis of hepatitis C virus infection. In:
Raymond Koff, ed. Clin Liver Dis 2001;5:931-953.
2. Napoli J, Bishop A, McGuinness P, Painter DM, McCaughan GW. Pro-
gressive liver injury in chronic hepatitis C infection correlates with in-
creased intrahepatic expression of Th1-associated cytokines. HEPATOLOGY
1996;24:759-765.
3. Nelson DR, Lauwers GY, Lau JYN, Davis GL. Interleukin-10 treatment
reduces fibrosis in patients with chronic hepatitis C: A pilot trial in inter-
feron nonresponders. Gastroenterology 2000;118:655-660.
4. Lau JYN, Davis GL, Brunson ME, Qian KP, Lin HJ, Quan S, DiNello R,
et al. Hepatitis C virus infection in renal transplant recipients. HEPATOL-
OGY 1993;18:1027-1031.
5. Cerny A, McHutchison JG, Pasquinelli C, Brown ME, Brothers MA,
Grabscheid B, Fowler P, et al. CTL response to HCV-derived peptides
containing the HLA-A2.1 binding motif. J Clin Invest 1995;95:521-530.
6. Chang K, Thimme R, Melpolder J, Oldach D, Pemberton J, Moorhead-
Loudis J, McHutchison JG, et al. Differential CD4? and CD8? T-cell
responsiveness in hepatitis C virus infection. HEPATOLOGY 2001;33:267-
276.
HEPATOLOGY, Vol. 38, No. 4, 2003NELSON ET AL. 867

Page 10

7. Takaki A, Wiese M, Maertens G, Depla E, Seifert U, Liebetrau A, Miller
JL, et al. Cellular immune responses persist, humoral responses decrease
twodecadesafterrecoveryfromasinglesourceoutbreakofhepatitisC.Nat
Med 2000;6:578-582.
8. Altman JD, Moss PAH, Goulder PJR, Baroch DH, McHeyzer-Williams,
Bell JI, McMichael AJ, et al. Phenotypic analysis of antigen-specific T
lymphocytes. Science 1996;274:94-96.
9. ZhuH,ZhaoH,CollinsCD,EckenrodeSE,RunQ,McIndoeRA,Craw-
ford JM, et al. Gene expression associated with interferon alfa antiviral
activity in an HCV replicon cell line. HEPATOLOGY 2003;37:1180-1188.
10. GuoJT,BichkoVV,SeegerC.EffectofalphainterferononthehepatitisC
virus replicon. J Virol 2001;75:8516-8523.
11. Deckert M, Soltek S, Geginat G, Lutjen S, Montesinos-Rongen, Hof H,
Schluter D. Endogenous interleukin-10 is required for prevention of a
hyperinflammatory intracer immune response in Listeria monocytogenes
meningoencephalitis. Infect Immun 2001;69:4561-4571.
12. Kane MM, Mosser DM. The role of IL-10 in promoting disease progres-
sion in leishmaniasis. J Immunol 2001;166:1141-1147.
13. Cooper S, Erickson AL, Adams EJ, Kansopon J, Weiner AJ, Chien DY,
Houghton M, et al. Analysis of a successful immune response against
hepatitis C. Immunity 1999;10:439-449.
14. Hoffman RM, Diepolder HM, Zachoval R, Zwiebel FM, Jung MC,
Scholz S, Nitschko H, et al. Mapping of immunodominant CD4? T
lymphocyte epitopes of hepatitis C virus antigens and their relevance dur-
ing the course of chronic infection. HEPATOLOGY 1995;21:632-638.
15. GrunerNH,GerlachTJ,JungMC,DiepolderHM,SchirrenCA,Schraut
WW,HoffmannR,etal.AssociationofhepatitisCvirus-specificCD8?T
cells with viral clearance in acute hepatitis C. J Infect Dis 2000;181:1528-
1536.
16. Diepolder HM, Zachoval R, Hoffmann RM, Wierenga EA, Santantonio
T, Jung MC, Eichenlaub D, et al. Possible mechanism involving T-lym-
phocyte response to non-structural protein 3 in viral clearance in acute
hepatitis C infection. Lancet 1995;346:1006-1007.
17. Tsai SL, Liaw YF, Chen MH, Huang CY, Kuo GC. Detection of type
2-like T-helper cells in hepatitis C virus infection: implications for hepa-
titis C virus chronicity. HEPATOLOGY 1997;25:449-457.
18. Cramp ME, Rossol S, Chokshi S, Carucci P, Williams R, Naoumav NV.
Hepatitis C virus-specific T-cell reactivity during interferon and ribavirin
treatment in chronic hepatitis C. Gastroenterology 2000;118:346-355.
19. LohrHF,SchmitzD,ArenzM,WeyerS,GerkenG,MeyerzumBuschen-
felde KH. The viral clearance in interferon-treated chronic hepatitis c is
associated with increased cytotoxic T cell frequencies. J Hepatol 1999;31:
407-415.
20. Nelson DR, Marousis CG, Davis GL, Lau JYN. Intrahepatic HCV-spe-
cific CTL activity and response to interferon therapy in chronic HCV.
HEPATOLOGY 1998;28:225-230.
21. LasarteJJ,Garcia-GraneroM,LopezA,CasaresN,GarciaN,CiveiraMP,
Borras-Cuesta F, et al. Cellular immunity to hepatitis C virus core protein
and the response to interferon in patients with chronic hepatitis C. HEPA-
TOLOGY 1998;28:815-822.
22. Piazzolla G, Tortorella C, Schiraldi O, Antonaci S. Relationship between
interferon-gamma, interleukin-10, and interleukin-12 production in
chronichepatitsCandin-vitroeffectsofinterferon-alpha.JClinImmunol
2000;20:54-61.
23. Eskdale J, Keijsers V, Huizinga T, Gallagher G. Microsatellite alleles and
single nucleotide polymorphisms combine to form four major haplotype
families at the human interleukin-10 locus. Gene Immun 1999;1:151-
155.
24. Edwards-Smith CJ, Jonsson JR, Purdie DM, Bansal A, Shorthouse C,
Powell EE. Interleukin-10 promoter polymorphism predicts initial re-
sponse of chronic hepatitis C to interferon alfa. HEPATOLOGY 1999;30:
526-530.
25. Cramp ME, Rossol S, Chokshi S, Carucci P, Williams R, Naoumov NV.
Hepatitis C virus-specific T-cell reactivity during interferon and ribavirin
treatment in chronic hepatitis C. Gastroenterology 2000;118:346-355.
26. Moore KW, de Waal Malefyt R, Coffman RL, O’Garra A. Interleukin-10
and the interleukin-10 receptor. Annu Rev Immunol 2001;19:683-765.
27. Trinchieri G. Regulatory role of T cells producing both interferon gamma
and interleukin 10 in persistent infection. J Exp Med 2001;194:F53-F57.
28. LisinskiTJ,FurieMB.Interleukin-10inhibitsproinflammatoryactivation
of endothelium in response to Borrelia burgdorferi or lipopolysaccharide
but not interleukin-1beta or tumor necrosis factor alpha. J Leukoc Biol
2002;72:503-511.
29. Wedemeyer H, Xiao-Song H. Nascimbeni M, Davis AR, Greenberg HB,
Hoofnagle JH, Liang TJ, et al. Impaired effector function of hepatitis C
virus-specific CD8? t cells in chronic hepatitis C virus infection. J Immu-
nol 2002;169:3447-3458.
30. Sakaguchi S. Regulatory T cells: key controllers of immulogic self-toler-
ance. Cell 2000;101:455-458.
31. Piccirillo CA, Shevach EM. Cutting edge: control of CD8? T cell activa-
tion by CD4?CD25? immunoregulatory cells. J Immunol 2001;167:
1137-1140.
32. Groux H, O’Garra A, Bigler M, Rouleau M, Antonenko S, de Vries JE,
Roncarolo MG. A CD4? T-cell subset inhibits antigen-specific T cell
responses and prevents colitis. Nature 1997;389:737-742.
33. BacchettaR,SartiranaC,LevingsMK,BordignonC,NarulaS,Roncarolo
MG. Growth and expansion of human T regulatory type 1 cells are inde-
pendent from TCR activation but require exogenous cytokines. Eur J Im-
munol 2002;32:2237-2245.
34. Yang AS, Lattime EC. Tumor-induced interleukin 10 suppresses the abil-
ityofspleniccellstostimulateCD4andCD8T-cellresponses.CancerRes
2003;63:2150-2157.
868 NELSON ET AL.HEPATOLOGY, October 2003