Journal of Infectious Diseases Advance Access published September 28, 2016

DAA-induced HCV clearance does not completely restore the altered

cytokine and chemokine milieu in patients with chronic hepatitis C

Hengst J.1,*, Falk C.S.2,3,4,*, Schlaphoff V.1,*, Deterding K.1, Manns M.P.1,2,3, Cornberg M.1,3,&,

ipt
Wedemeyer H.1,2,3,&

cr
1
Department of Gastroenterology, Hepatology and Endocrinology, Hannover Medical School,

Downloaded from http://jid.oxfordjournals.org/ at University of Hong Kong Libraries on October 2, 2016

Hannover, Germany

us
2
IFB-Tx
3
German Center for Infection Research (DZIF)
4
an
Institute of Transplant Immunology, IFB-Tx, Hannover Medical School, Hannover, Germany
M
Corresponding author: Heiner Wedemeyer, wedemeyer.heiner@mh-hannover.de, Phone: +49 511

532 6814, Hannover Medical School, Carl-Neuberg-Straße 1, D-30625 Hannover, Germany

ed

*
JH, CSF and VS contributed equally to this work
&
MC and HW contributed equally and share senior authorship
pt
ce
Ac

© The Author 2016. Published by Oxford University Press for the Infectious Diseases Society
of America. All rights reserved. For permissions, e-mail journals.permissions@oup.com.
 2

Acknowledgement

We thank Kerstin Daemen and Jana Keil for excellent technical assistance.

Funding statement

ipt
This work was supported by the International Research Training Group 1273 supported by the

German Research Foundation (DFG), the Center Research Grants 738 (project B2, B3) and 900

cr
(project A5) supported by the DFG, and the IFB-Tx (BMBF 01EO1302) and the German Center for

Downloaded from http://jid.oxfordjournals.org/ at University of Hong Kong Libraries on October 2, 2016

Infectious Diseases (DZIF) TTU Hepatitis and TTU-IICH.

us
Contributorship Statement
an
JH study concept and design, acquisition of data, analysis and interpretation of data, statistical

analysis, drafting and final approval of the manuscript.

M
CSF acquisition of data, final approval and revision of the manuscript.

VS study concept and design, interpretation of data, statistical analysis, drafting and final approval

of the manuscript.
ed

KD acquisition of data, final approval of the manuscript.

MPM drafting and final approval of the manuscript.

pt

MC interpretation of data, drafting and final approval of the manuscript.

HW study concept and design, interpretation of data, drafting and final approval of the manuscript.
ce

Conflict of Interest
Ac

JH has nothing to disclose.

VS has nothing to disclose.

KD has received lecture fees from Gilead, AbbVie and Merck.

CSF has nothing to disclose.

 3

MPM has received grants and personal fees from AbbVie, Biotest, Boehringer Ingelheim,

BristolMyers Squibb, Gilead, GlaxoSmithKline, Janssen, Merck (MSD), Novartis, Roche and

personal fees from Achillion.

MC has received lecture fees from AbbVie, Bristol-Myers Squibb, Boehringer Ingelheim Pharma

ipt
GmbH, Gilead, Janssen-Cilag, MSD Sharp & Dohme/Merck, Roche Diagnostic, Roche Pharma and

Siemens. MC has received advisory board fees from AbbVie, Bristol-Myers Squibb, Boehringer

cr
Ingelheim Pharma GmbH, Gilead, Roche Diagnostic and Roche Pharma as well as data safety board

Downloaded from http://jid.oxfordjournals.org/ at University of Hong Kong Libraries on October 2, 2016

fees from Janssen-Cilag.

us
HW has received grants from AbbVie, Gilead, Roche, Roche Diagnostics, Abbott, Myr GmbH and

Eiger, consulting fee or honorarium from AbbVie, Abbott, BMS, Boehringer Ingelheim, Eiger,
an
Gilead, Janssen, NovartisMSD/Merck, Roche, Roche Diagnostics and Transgene for work under

consideration for publication. Regarding financial activities outside the submitted work HW has
M
received money for board memberships from AbbVie, Abbott, BMS, Boehringer, Eiger, Gilead,

Myr GmbH, Novartis and Roche as well as for consultancy from Eiger, Janssen and Siemens as

well as payment for lectures including service on speakers bureaus from Falk Foundation and
ed

OmnisMed.
pt

Meeting where this information has been presented

European Association for the Study of the Liver (EASL2016), April 2016, Barcelona, Catalonia,
ce

Spain
Ac
 4

Abstract

Background. Persistent infection with the hepatitis C virus (HCV) causes profound alterations of

the cytokine and chemokine milieu in peripheral blood. However, it is unknown to what extend

these alterations affect the progression of liver disease and whether HCV clearance normalizes

ipt
soluble inflammatory mediators.

cr
Methods. We performed multianalyte profiling of 50 plasma proteins of 28 patients with

Downloaded from http://jid.oxfordjournals.org/ at University of Hong Kong Libraries on October 2, 2016

persistent HCV infection and advanced stages of liver fibrosis or cirrhosis and 20 controls with fatty

us
liver disease. The patients were treated for 24 weeks with sofosbuvir and ribavirin and were

sampled longitudinally. Ten patients experienced a viral relapse after treatment cessation.
an
Results. The cytokine and chemokine expression pattern was markedly altered in chronic
M
HCV patients as compared to healthy controls and NASH patients. Distinct soluble factors were

associated with the level of fibrosis/cirrhosis, viral replication or treatment outcome. The baseline

expression level of twelve cytokines distinguished SVR from relapse patients. While the majority of
ed

up-regulated analytes declined during and after therapy, HCV clearance did not lead to a restoration

of parameters that were suppressed.

pt

Conclusions. Chronic HCV infection appears to disrupt the milieu of soluble inflammatory
ce

mediators even after viral clearance. Thus, HCV cure does not lead to complete immunological

restitution.
Ac
 5

Introduction

Viral infections are controlled by a tightly regulated immune network. In addition to cellular

components of the immune system various soluble mediators are essential for eliminating the virus.

Whether any of these soluble immune mediators (SIMs) can serve as a biomarker for predicting

ipt
disease severity or response to anti-viral treatment is so far unclear. It is well established that

chronic viral infections cause major changes in the inflammatory cytokine and chemokine milieu

cr
[1]. Still, limited data is available to what extent imprints in the immune network are reversible

Downloaded from http://jid.oxfordjournals.org/ at University of Hong Kong Libraries on October 2, 2016

once chronic infections are cleared. We address these questions exemplary in patients who were

us
persistently infected with the hepatitis C virus (HCV) and who received novel interferon-free

therapy with direct-acting antivirals (DAAs).

an
HCV is a human pathogenic virus and viral replication takes place mainly in hepatocytes

[2]. In the majority of cases HCV causes a persistent infection. It is estimated that 64-102 million
M
people are infected with HCV worldwide [3]. Chronic hepatitis C can lead to liver fibrosis, cirrhosis

and hepatic decompensation as well as hepatocellular carcinoma [2]. Persistent HCV infection has

been associated with suppression and exhaustion of HCV-specific immune responses [4].
ed

Moreover, HCV infection leads to induction of excessive ISG expression [5] and alterations in the

overall systemic inflammatory milieu [6, 7] which subsequently has been linked to alterations of
pt

CMV- and EBV-specific T cell responses as well as NK cell phenotype and function [8-10].

During the last few years several new DAAs were approved for the treatment of chronic
ce

hepatitis C (cHC) which interfere with viral replication or assembly [11-13]. These new treatment

options are very potent, leading to cure rates of 90-100% for most HCV genotypes [12]. The novel
Ac

therapies enable for the first time direct investigations of effects of cure from a persistent viral

infection on the human immune system. Of note, the novel DAAs do not exert direct immune-

modulatory activity in contrast to the previous treatment, which was based on administration of type

I interferons, mainly pegylated interferon alfa-2 (pegIFN-α2) [14, 15, 15, 16].
 6

Until now only a limited number of studies have been performed to investigate the

inflammatory cytokine and chemokine milieu in acute [6] and chronic HCV infections [17-19]. So

far, a single study investigated inflammatory mediators during DAA treatment of cHC but only four

parameters were investigated [18]. It therefore remains unclear if potential alterations in the

ipt
systemic inflammatory cytokine and chemokine milieu in cHC patients are restored upon clearance

of infection.

cr
The aim of this study was to investigate how the inflammatory milieu, including cytokines,

Downloaded from http://jid.oxfordjournals.org/ at University of Hong Kong Libraries on October 2, 2016

chemokines, growth factors and adhesion molecules, is altered in cHC and whether these changes

us
are associated with stage of liver disease. Moreover, we aimed to analyze whether distinct SIMs

differ between patients who achieve a sustained viral response (SVR) and patients who experience a
an
viral relapse after treatment cessation. Finally, we aimed to study whether clearance of the infection

by DAA treatment would restore the immunological imprints established upon chronic HCV
M
infection.

Material and methods

ed

Patient Material

In this study, a total of 53 subjects were studied including 5 healthy controls, 20 patients with
pt

nonalcoholic steatohepatitis (NASH) and 28 patients persistently infected with HCV. Chronic HC

patients were monitored before, during and after DAA treatment in the outpatient clinic of the
ce

Department of Gastroenterology, Hepatology, and Endocrinology at Hannover Medical School in

Germany. Patients with cHC were treated for 24 weeks with sofosbuvir (400 mg, qd) and ribavirin
Ac

(RBV) (weight-based starting dose between 800-1200 mg daily). Peripheral blood samples were

collected at baseline (BL; before therapy start), week 4 (w4), week 12 (w12), week 24 (w24; end of

treatment) during therapy and twelve weeks after treatment cessation (fu12). 18 patients achieved a

SVR, whereas ten patients experienced a viral relapse. All patients except two were anti-HIV

negative, and none of the patients was co-infected with the hepatitis B virus. The HIV-positive
 7

patients received antiretroviral therapy and had controlled HIV infection with negative HIV-RNA

values. Blood plasma was collected from EDTA-treated peripheral blood samples and stored at -

80°C for later analysis.

Patient characteristics are presented in table 1. For all cHC patients clinical data on liver

ipt
inflammation, liver fibrosis and cirrhosis were collected from routine clinical diagnostics, as

previously described in detail (PMID: 26250762). The diagnosis of NASH was based on liver

cr
histology in all but one patient who had evidence of steatosis and elevated liver enzymes in the

Downloaded from http://jid.oxfordjournals.org/ at University of Hong Kong Libraries on October 2, 2016

absence of any other cause of liver disease. Patients gave informed written consent for the study of

us
immunological parameters. The protocols for the sample collection and investigations were

reviewed and approved by the local ethics committee of Hannover Medical School (Study number
an
2148-2014). In this cohort, 18 patients achieved a SVR and ten patients experienced a viral relapse,

however, baseline samples were available for eight relapse patients only.
M
Cytokine and Chemokine Measurements

We performed multianalyte profiling of 50 cytokines, chemokines, adhesion molecules and growth

ed

factors in the blood plasma of all samples using the LUMINEX-based multiplex bead technology

(BioPlex Pro Human Cytokine Panel, Bio-Rad, Hercules, CA, USA). The assay was conducted
pt

following the manufacturer’s recommendations [8] and according to optimized protocols applied in

various previous studies, e.g. [8, 20-22]. Of note, all samples were run in one run. The beads were
ce

acquired on the LUMINEX instrument using the BioPlex Manager 6.0 software.
Ac

Statistical analyses

Data were analyzed using GraphPad Prism v6.0b (Graph Pad Software, La Jolla, CA, USA). All

data were evaluated for their statistical distribution using the Kolmogorov-Smirnov-Test or the

D’Agostino-Pearson test. In general, quantitative comparisons were performed using parametric

Student’s t-test for normally distributed values and non-parametric Wilcoxon test or Mann-Whitney
 8

test for values that did not show normal distribution. The statistical test used for each analysis is

mentioned in the respective figure legend. Principal component analysis (PCA) was performed

using Qlucore Omics Explorer v3.2 (Qlucore AB, Lund, Sweden). Regarding PCA analyses,

repetitive t-testing was applied to compare groups of patients with each other. For analysis values

ipt
were set to P= 0.05 and Q <0.2. Multiple test correction was applied by multiple t-test and the

False Discovery Rate (FDR) approach with a desired FDR (Q)=10%.

cr
Downloaded from http://jid.oxfordjournals.org/ at University of Hong Kong Libraries on October 2, 2016
Results

us
Altered inflammatory cytokine and chemokine milieu in cHC patients

The expression pattern of all SIMs analyzed in the plasma differed strongly between patients with
an
cHC, NASH patients, and healthy individuals (Figure 1A). The degree of biochemical disease

activity was similar between NASH and cHC patients with a median ALT value of 100 U/L but
M
cHC patients had higher liver stiffness values than NASH patients (Table 1). In both patient cohorts

of liver disease (cHC and NASH) a specific and unique prolife of inflammatory mediators was

observed with a distinct pattern of proteins up- or down-regulated in comparison to healthy

ed

controls. In total, the expression level of 25 analytes was significantly altered in cHC patients as

compared to healthy individuals (Table 2) with 17 SIMs being increased in cHC. Exemplary, the
pt

expression level of the chemokine IP-10 (CXCL10) and the cytokines IL-12p40, IFN-α2, LTA, IL-

18 and TRAIL are shown in Figure 1B, which were also up-regulated in comparison to NASH
ce

patients (Figure 1B). Interestingly, only six of the eight reduced SIMs were significantly lower in

cHC when performing additional individual t-testing including the four cytokines IL-17, IL-1β,
Ac

IFN-γ and IL-4 as well as the two growth factors FGF-basic and PDGF-bb. Of note, only the

expression level of PDGF-bb was decreased in NASH patients to the same extent as in patients with

cHC while reductions of IL-17, IL-1β, IFN-γ, IL-4 and FGF-b were largely specific for cHC

patients (Figure 1C).

 9

Thus, patients persistently infected with HCV showed a disrupted milieu of cytokines and

chemokines as compared to healthy individuals and patients with a non-viral inflammatory liver

disease, NASH.

ipt
The severity of liver damage affects the inflammatory milieu in cHC patients

During cHC also liver fibrosis and cirrhosis may influence levels of SIMs. Accordingly, we aimed

cr
to investigate whether the expression levels of SIMs correlated with liver stiffness values. All cHC

Downloaded from http://jid.oxfordjournals.org/ at University of Hong Kong Libraries on October 2, 2016

patients were grouped based on the stage of liver damage into patients with fibrosis (<14.5 kPa),

us
mild cirrhosis (14.5-25 kPa) and severe cirrhosis (>25 kPa). The expression levels of the soluble

form of the adhesion molecules VCAM-1 (CD106, sVCAM-1) and ICAM-1 (CD54, sICAM-1)
an
positively correlated with the fibroscan values, whereas the expression of the growth factor PDGF-

bb correlated negatively (Figure 2A). Expression levels of the two adhesion molecules sVCAM-1
M
and sICAM-1 increased gradually with the severity of cirrhosis (Figure 2B). In contrast to this, the

expression of PDGF-bb, FGF-basic, IL-17 and IL-4 decreased in cHC patients compared to healthy

individuals (Figure 1C) and showed a step-wise reduction correlating with the severity of liver
ed

stiffness (Figure 2B). Thus, out of the six SIMs, which were reduced in cHC as compared to healthy

individuals, four were associated with the cirrhosis stage.

pt

In summary, not only HCV infection but also the stage of liver cirrhosis affects the

expression of distinct analytes. Notably, only the expression of the adhesion molecules correlated
ce

positively with the severity of liver stiffness.

Ac

Principal Component Analysis (PCA) identifies distinct pre-treatment patterns of SIMs

distinguishing SVR from relapse patients

The patients were treated with sofosbuvir and RBV only in this study. Thus, we had the unique

chance to study whether inflammatory mediators may be predictive of viral clearance with

suboptimal interferon-free therapies. SVR and relapse patients clearly differed from each other in
 10

the expression levels of SIMs already before start of DAA therapy (Figure 3A). A distinct clustering

of both patient groups could be seen in the PCA plot, which represents the strength of difference

based on all significantly different values (P=.05, Q<.2). To identify on which SIMs the clustering

was based, a heatmap displaying all significantly different parameters was generated (Figure 3B).

ipt
Twelve parameters including ten cytokines, one chemokine SDF-1α (CXCL12) and one growth

factor (β-NGF) were significantly different between the patient groups based on the PCA analysis

cr
(Figure 3C). Ten of those were confirmed by multiple test correction (Figure 3C). The most

Downloaded from http://jid.oxfordjournals.org/ at University of Hong Kong Libraries on October 2, 2016

contrasting cytokine in both analyses was IL-12p40 (PCA: P=.0002, Q=.0102; multiple t-test:

us
P=.0002). Noticeably, the expression level of each of these analytes was higher in relapse as

compared to SVR patients (Figure 3C). IP-10 was not statistically different between these two

groups.
an
Thus, a pre-treatment identification of cHC patients who may not achieve a SVR with
M
suboptimal DAA treatment may be possible by studying expression levels of distinct inflammatory

mediators.
ed

DAA therapy restores some but not all mediators in SVR patients

As shown in Figure 4A, viral loads and alanine aminotransferase (ALT) levels, a marker for liver
pt

inflammation, declined rapidly upon DAA treatment initiation. Moreover, liver stiffness values also

improved upon HCV clearance. Due to these alterations, we investigated whether the imprints on
ce

the cytokine and chemokine milieu acquired during chronic HCV infection (BL) disappear upon

viral clearance.
Ac

Both groups differed not only at BL (Figure 3) but also during and after therapy as relapse

patients showed higher expression levels of most SIMs increased during cHC (Figure 4B). The

expression level of 22 analytes decreased significantly from BL to fu12 in SVR patients (Table 3)

including 17 SIMs that were significantly higher in all cHC patients compared to healthy controls

(Table 2). This is shown in Figure 4B exemplary for the six analytes shown in Figure 1B. Despite
 11

the drastic decline in SVR patients, expression levels similar to healthy controls were not reached

(Figure 4C). Four chemokines (GRO-α (CXCL), IL-8 (CXCL8), MIP-1β (CCL4) and RANTES

(CCL5)) and TNF decreased significantly during therapy (Table 3) even though their expression

was not enhanced in cHC (Figure 1A). Notably, only the expression of hepatocyte growth factor

ipt
(HGF) was lowered significantly in SVR and relapse patients at fu12. IP-10 was the only analyte

associated with viral relapse, as its expression level decreased rapidly and increased again in relapse

cr
patients upon viral re-occurrence (Figure 4C).

Downloaded from http://jid.oxfordjournals.org/ at University of Hong Kong Libraries on October 2, 2016

In contrast, the expression levels of all significantly decreased SIMs in cHC (Figure 1C)

us
were not restored during DAA treatment (Figure 4D) in neither patient group.

Recapitulating, the majority of inflammatory mediators that were increased in cHC patients
an
decreased significantly during and after therapy, however, not reaching levels comparable to

healthy individuals. The imprint on the six suppressed SIMs was not altered 36 weeks after
M
treatment initiation.

Discussion
ed

The introduction of IFN-free therapy of hepatitis C allows for the first time to investigate the

immunological consequences of clearance of a chronic viral infection in humans which has been
pt

persistent for decades in most individuals. We here show in large multianalyte profiling of 50

soluble immune mediators (SIM) (i) that the expression pattern of SIMs differed strikingly between
ce

cHC patients, healthy individuals and patients with a non-viral inflammatory liver disease; (ii) that

distinct analytes, in particular those which were down-regulated in HCV, correlated with the
Ac

severity of liver fibrosis and cirrhosis; (iii) that patients being able to clear HCV infection with the -

meanwhile suboptimal - treatment of sofosbuvir plus RBV could be distinguished from relapsing

patients after therapy based on their pre-treatment cytokine and chemokine profile; and maybe most

importantly (iv) that the altered inflammatory milieu did not normalize upon viral clearance even
 12

though most increased pro-inflammatory parameters partially declined while factors found to be

suppressed in cHC patients remained at low levels for up to 8 months after HCV RNA negativation.

It is well established that HCV infection is associated with a profound activation of the

interferon system by i.e. induction of interferon stimulated gene expression [5]. We here also found

ipt
an up-regulation of 22 SIMs in cHC compared to healthy controls. Most of those parameters were

also higher in cHC than in NASH patients. No study has yet performed a broad profiling of

cr
cytokines, chemokines, growth and adhesion factors in cHC patients versus individuals with fatty

Downloaded from http://jid.oxfordjournals.org/ at University of Hong Kong Libraries on October 2, 2016

liver disease. We identified distinct differences between cHC and NASH patients for various other

us
parameters including GM-CSF, IL-2, IL-16 and β-NGF. IL-12p40, which was the most

significantly increased cytokine in cHC compared to healthy controls, was not increased in NASH
an
patients. Notably, IL-12p40 needs to form a heterodimer with IL-12p35 to be functional active as

IL-12p70. However, IL-12p70 expression was not increased in cHC patients, suggesting that the
M
pro-inflammatory cytokine IL-23, the heterodimer between IL-12p40 and p90 may be responsible

for this effect. In this context it is important to note that the degree of liver inflammation was

similar between the NASH and cHC patients with median ALT values of approximately 100 U/L in
ed

both groups, however, FibroScan values were higher in cHC compared to NASH. Thus, both the

stage of liver disease and HCV infection may have contributed to the differences in cytokine and
pt

chemokine patterns.

While the majority of analytes studied here was increased in cHC patients as compared to
ce

healthy controls, six SIMs were down-regulated. These parameters were also lower in HCV

infected individuals than in NASH patients, except PDGF-bb. Of note, four of these decreased
Ac

analytes were inversely associated with the severity of liver cirrhosis (IL-4, IL-17, FGF-basic and

PDGF-bb). The role of some of these factors such as PDGF or FGF in liver fibrogenesis is well

established even though altered plasma levels may frequently be a consequence rather than the

cause for a pathological condition. E.g., intrahepatic expression of IL-17 promotes liver

fibrogenesis by activation of stellate cells [24]. On the first view our finding of reduced circulating
 13

plasma levels of IL-17, in particular in patients with advanced cirrhosis, is therefore surprising.

However, down-regulation of IL-17 production may be the result of a regulatory loop to prevent

further fibrosis progression. On the other hand, the strong correlation between distinct adhesion

molecules (sVCAM-1; sICAM-1) and the stage of liver fibrosis could indicate a direct

ipt
pathophysiological link as higher expression of these markers could recruit pro-inflammatory

immune cells to the liver which can promote fibrogenesis [25]. Importantly, associations between

cr
histological disease activity and serum levels of sICAM-1, sVCAM-1 and IL-4 have already been

Downloaded from http://jid.oxfordjournals.org/ at University of Hong Kong Libraries on October 2, 2016

described previously in hepatitis C patients [26].

us
Immune responses can contribute to control of HCV replication in patients receiving

interferon-free therapy with novel DAAs [27]. The role of innate and adaptive immunity is likely to
an
be the more important the weaker the antiviral drug regimen is. We here had the unique

opportunity to study potential immune correlates of viral response in cHC patients receiving only
M
sofosbuvir and RBV without combination with other DAAs. These patients were treated during the

first half of 2014 when only sofosbuvir was approved in Europe while other potent DAAs became

available later that year. As a consequence, ten patients relapsed after completing 24 weeks of
ed

sofosbuvir plus RBV, while 18 patients achieved a SVR. Interestingly, 10 SIMs were differently

expressed before therapy in the two groups of patients with all parameters being higher in
pt

individuals who did not clear HCV infection. These markers included various cytokines such as IL-

12p40, Il-2Ra (sCD25) and IFN-α2. Thus, a more activated immune system was predictive of
ce

treatment failure. Overall, these findings are in line with the concept of an over-exhausted interferon

system in cHC failing to control HCV which has been also associated with a lower response to the
Ac

previous pegIFN-α based therapies [5]. In contrast to pegIFN-α2 therapy, there was no difference in

the pre-treatment IP-10 expression level between SVR and relapse patients in interferon-free

therapies. This observation is also in line with a previous study investigating the role of IP-10 in

HCV genotype 2 or 3 infected patients treated with sofosbuvir und RBV [18]. In interferon-

containing therapies the relative increase of IP-10 during therapy was indicative for responsiveness
 14

to pegIFN-α2 – which was mainly observed in patients with low IP-10 levels before therapy. In

contrast, the interferon system including IP-10 is not boosted during IFN-free therapy of hepatitis C.

The intrahepatic and peripheral interferon-stimulated gene expression declines with HCV

eradication [30]. Thus, the lack of association between pretreatment IP-10 levels and treatment

ipt
outcome in sofosbuvir plus RBV therapy may not be surprising.

Successful treatment of hepatitis C is associated with complete virological cure and rapidly

cr
declining liver inflammation during therapy. It has been shown previously that cytokines and

Downloaded from http://jid.oxfordjournals.org/ at University of Hong Kong Libraries on October 2, 2016

chemokines decline during IFN-free treatment of hepatitis C including blood IP-10, MCP-1, MIP-

us
1β, and IL-18 levels [18] and intrahepatic IFN-stimulated genes [30]. Indeed, we here could

confirm the decline of the same plasma SIMs in SVR patients except for the chemokine MCP-1.
an
However and importantly, most of the parameters, which were elevated pre-treatment did not

normalize during further follow-up. HCV RNA becomes undetectable within 2-8 weeks of
M
sofosbuvir therapy and liver enzymes also normalize rapidly. Thus, viral replication was absent for

6-8 months in most patients in this study. Continuous inflammation despite viral suppression is a

major topic of debate in HIV infection [31]. Longer follow-up is required to answer the important
ed

question whether the inflammatory parameters will further decline over time in successfully treated

HCV patients and whether a long-term restoration of the interferon system is achieved after HCV
pt

clearance.
ce

Intriguingly, all six cytokines and growth factors that were significantly decreased in cHC

compared to healthy controls were similarly affected in SVR and relapse patients and did not
Ac

normalize upon DAA treatment. This phenomenon was also observed for one immune cell subset,

mucosal-associated invariant T (MAIT) cells [32, 33]. It has been shown that MAIT cells are

decreased in frequency and have an impaired functional capacity in cHC compared to healthy

individuals. This suggests that not only SIMs but also immune cell subsets that are down-regulated

upon cHC are not able to recover upon viral clearance, the exact mechanism for this remains
 15

unclear. The degree of liver cirrhosis decreases upon successful DAA treatment [34], however, the

decreased MAIT cell frequency does not recover even until one year after treatment cessation [32]

indicating that the improvement of liver stiffness does not seem to affect this immune cell subset.

A strength of our study is the longitudinal sampling before, during and after DAA treatment

ipt
of cHC patients. Moreover, we had the unique possibility to investigate how SVR and relapse

patients differ in regard to their cytokine and chemokine profile as patients were treated with

cr
sofosbuvir and ribavirin only, a treatment regime that is meanwhile considered as being suboptimal

Downloaded from http://jid.oxfordjournals.org/ at University of Hong Kong Libraries on October 2, 2016

for most HCV genotypes. It has to be considered that most HCV infected patients had already

us
advanced stages of liver disease and that the mean age of the HCV patients is higher than that of the

five healthy donors. A larger control cohort would be preferable. Moreover, all patients were treated
an
with ribavirin and we cannot exclude distinct effects of RBV on SIMs.

In conclusion, persistent HCV infection is associated with profound alterations of soluble

M
inflammatory mediators, which are associated with liver disease progression, treatment outcome

and viral presence. Importantly, these changes were not fully reversible upon clearance of the viral

infection.
ed
pt
ce
Ac
 16

References

1. Beltra JC, Decaluwe H. Cytokines and persistent viral infections. Cytokine 2016;82:4-15.

2. Protzer U, Maini MK, Knolle PA. Living in the liver: hepatic infections. Nat Rev Immunol 2012;12:201-

13.

ipt
3. Gower E, Estes C, Blach S, Razavi-Shearer K, Razavi H. Global epidemiology and genotype distribution

of the hepatitis C virus infection. J Hepatol 2014;61:S45-57.

cr
Downloaded from http://jid.oxfordjournals.org/ at University of Hong Kong Libraries on October 2, 2016
4. Klenerman P, Thimme R. T cell responses in hepatitis C: the good, the bad and the unconventional. Gut

us
2012;61:1226-34.

5. Heim MH, Thimme R. Innate and adaptive immune responses in HCV infections. J Hepatol 2014;61:S14-

25.
an
6. Duffy D, Mamdouh R, Laird M, et al. The ABCs of viral hepatitis that define biomarker signatures of
M
acute viral hepatitis. Hepatology 2014;59:1273-82.

7. Marra F, Tacke F. Roles for chemokines in liver disease. Gastroenterology 2014;147:577,594.e1.

ed

8. Owusu Sekyere S, Suneetha PV, Hardtke S, et al. Type I Interferon Elevates Co-Regulatory Receptor

Expression on CMV- and EBV-Specific CD8 T Cells in Chronic Hepatitis C. Front Immunol 2015;6:270.
pt

9. Golden-Mason L, Waasdorp Hurtado CE, Cheng L, Rosen HR. Hepatitis C viral infection is associated

with activated cytolytic natural killer cells expressing high levels of T cell immunoglobulin- and mucin-
ce

domain-containing molecule-3. Clin Immunol 2015;158:114-25.

10. Rehermann B. Pathogenesis of chronic viral hepatitis: differential roles of T cells and NK cells. Nat Med
Ac

2013;19:859-68.

11. Gane EJ, Stedman CA, Hyland RH, et al. Nucleotide polymerase inhibitor sofosbuvir plus ribavirin for

hepatitis C. N Engl J Med 2013;368:34-44.

 17

12. Lawitz E, Mangia A, Wyles D, et al. Sofosbuvir for previously untreated chronic hepatitis C infection. N

Engl J Med 2013;368:1878-87.

13. Foster GR, Irving WL, Cheung MC, et al. Impact of direct acting antiviral therapy in patients with

chronic hepatitis C and decompensated cirrhosis. J Hepatol 2016.

ipt
14. Wiegand J, Cornberg M, Aslan N, et al. Fate and function of hepatitis-C-virus-specific T-cells during

peginterferon-alpha2b therapy for acute hepatitis C. Antivir Ther 2007;12:303-16.

cr
Downloaded from http://jid.oxfordjournals.org/ at University of Hong Kong Libraries on October 2, 2016
15. Barnes E, Harcourt G, Brown D, et al. The dynamics of T-lymphocyte responses during combination

therapy for chronic hepatitis C virus infection. Hepatology 2002;36:743-54.

us
16. Missale G, Pilli M, Zerbini A, et al. Lack of full CD8 functional restoration after antiviral treatment for
an
acute and chronic hepatitis C virus infection. Gut 2012;61:1076-84.

17. Jablonska J, Pawlowski T, Laskus T, et al. The correlation between pretreatment cytokine expression

patterns in peripheral blood mononuclear cells with chronic hepatitis C outcome. BMC Infect Dis
M
2015;15:556,015-1305-1.

18. Carlin AF, Aristizabal P, Song Q, et al. Temporal dynamics of inflammatory cytokines/chemokines
ed

during sofosbuvir and ribavirin therapy for genotype 2 and 3 hepatitis C infection. Hepatology

2015;62:1047-58.
pt

19. Wandrer F, Falk CS, John K, et al. Interferon-Mediated Cytokine Induction Determines Sustained Virus

Control in Chronic Hepatitis C Virus Infection. J Infect Dis 2016;213:746-54.

ce

20. Wranke A, Heidrich B, Ernst S, et al. Anti-HDV IgM as a marker of disease activity in hepatitis delta.

PLoS One 2014;9:e101002.

Ac

21. Gisa A, Suneetha PV, Behrendt P, et al. Cross-genotype-specific T-cell responses in acute hepatitis E

virus (HEV) infection. J Viral Hepat 2016;23:305-15.

 18

22. Lemke A, Noriega M, Roske AM, et al. Rat renal transplant model for mixed acute humoral and cellular

rejection: Weak correlation of serum cytokines/chemokines with intragraft changes. Transpl Immunol

2015;33:95-102.

23. Kumar A, Sharma A, Duseja A, et al. Patients with Nonalcoholic Fatty Liver Disease (NAFLD) have

ipt
Higher Oxidative Stress in Comparison to Chronic Viral Hepatitis. J Clin Exp Hepatol 2013;3:12-8.

24. Hammerich L, Tacke F. Interleukins in chronic liver disease: lessons learned from experimental mouse

cr
models. Clin Exp Gastroenterol 2014;7:297-306.

Downloaded from http://jid.oxfordjournals.org/ at University of Hong Kong Libraries on October 2, 2016

25. Williams MJ, Clouston AD, Forbes SJ. Links between hepatic fibrosis, ductular reaction, and progenitor

us
cell expansion. Gastroenterology 2014;146:349-56.

an
26. Patel K, Remlinger KS, Walker TG, et al. Multiplex protein analysis to determine fibrosis stage and

progression in patients with chronic hepatitis C. Clin Gastroenterol Hepatol 2014;12:2113,20.e1-3.

27. Rehermann B, Bertoletti A. Immunological aspects of antiviral therapy of chronic hepatitis B virus and
M
hepatitis C virus infections. Hepatology 2015;61:712-21.

28. Diago M, Castellano G, Garcia-Samaniego J, et al. Association of pretreatment serum interferon gamma
ed

inducible protein 10 levels with sustained virological response to peginterferon plus ribavirin therapy in

genotype 1 infected patients with chronic hepatitis C. Gut 2006;55:374-9.

pt

29. Askarieh G, Alsio A, Pugnale P, et al. Systemic and intrahepatic interferon-gamma-inducible protein 10

kDa predicts the first-phase decline in hepatitis C virus RNA and overall viral response to therapy in chronic
ce

hepatitis C. Hepatology 2010;51:1523-30.

30. Meissner EG, Wu D, Osinusi A, et al. Endogenous intrahepatic IFNs and association with IFN-free HCV
Ac

treatment outcome. J Clin Invest 2014;124:3352-63.

31. Klatt NR, Chomont N, Douek DC, Deeks SG. Immune activation and HIV persistence: implications for

curative approaches to HIV infection. Immunol Rev 2013;254:326-42.

 19

32. Hengst J, Strunz B, Deterding K, et al. Nonreversible MAIT cell-dysfunction in chronic hepatitis C virus

infection despite successful interferon-free therapy. Eur J Immunol 2016.

33. Spaan M, Hullegie SJ, Beudeker BJ, et al. Frequencies of Circulating MAIT Cells Are Diminished in

Chronic HCV, HIV and HCV/HIV Co-Infection and Do Not Recover during Therapy. PLoS One

ipt
2016;11:e0159243.

34. Deterding K, Schlevogt B, Port K, Cornberg M, Wedemeyer H. Letter: can persisting liver stiffness

cr
indicate increased risk of hepatocellular cell carcinoma after successful anti-HCV therapy? - authors' reply.

Downloaded from http://jid.oxfordjournals.org/ at University of Hong Kong Libraries on October 2, 2016

Aliment Pharmacol Ther 2016;43:546-7.

us
an
M
ed
pt
ce
Ac
 20

Figure 1. cHC causes an altered cytokine and chemokine milieu as compared to healthy

individuals and NASH patients. A, Heatmap showing the expression pattern of 48 cytokines and

chemokines normalized to the median value of all healthy individuals. Data are presented for five

healthy controls, 20 NASH patients and 26 cHC patients. Cytokines and chemokines are ordered

ipt
based on their expression level in cHC patients. IL-1α and IL-15 were excluded, as the median

value of healthy individuals for these cytokines was lower 0.1. B, Expression level of several

cr
cytokines and chemokines (pg/ml) which are significantly up-regulated in cHC and C, expression

Downloaded from http://jid.oxfordjournals.org/ at University of Hong Kong Libraries on October 2, 2016

level of all six cytokines and growth factors (pg/ml) which are significantly reduced in cHC

us
compared to healthy individuals. *P<.05; **P<.01; ***P<.001; ****P<.0001, multiple t-test with a

FDR (Q)=10%, horizontal bars represent mean and standard error of the mean (SEM).
an
Figure 2. The expression level of several analytes is affected by progression of liver fibrosis
M
and cirrhosis. A, Correlation analysis of sVCAM-1, sICAM-1 and PDGF-bb with fibroscan values.

B, Expression of adhesion molecules, cytokines and growth factors (pg/ml) that are significantly

different regulated in healthy (n=5) compared to fibrotic (fibroscan <14.5, n=6) or cirrhotic
ed

(fibroscan 14.5-25, n=8; fibroscan >25, n=12) cHC patients (n=26). *P<.05; **P<.01; ***P<.001,

multiple t-test with a FDR (Q)=10%, horizontal bars represent mean and SEM.
pt

Figure 3. The cytokine and chemokine milieu differs before start of DAA therapy between
ce

SVR and relapse patients. A, Principal component analysis (PCA) showed a distinct clustering of

cHC patients that have a sustained virologic response (SVR, n=18) upon DAA treatment and
Ac

patients that have a vial relapse (n=8). The PCA plot is calculated on baseline values of all

measured SIMs and shows how the patients cluster together due to their treatment outcome (SVR

and relapse). B, Heatmap summarizing the cytokines and chemokines that differ significantly in

their expression levels upon PCA between SVR and relapse patients. C, Expression level of

cytokines and chemokines (pg/ml) that were significantly different expressed upon PCA were
 21

analyzed by multiple t-test with a FDR approach (Q)=10%. cHC patients that experienced a viral

relapse after treatment cessation show higher expression levels of these analytes. *P<.05; **P<.01;

***P<.001, multiple t-test was used, horizontal bars represent mean and SEM.

ipt
Figure 4. The cytokine and chemokine milieu improves but does not normalize upon

successful DAA treatment. A, HCV RNA levels (IU/ml) and alanine transferase (U/L) levels

cr
during and after DAA treatment are shown for all 28 cHC patients before, during after treatment

Downloaded from http://jid.oxfordjournals.org/ at University of Hong Kong Libraries on October 2, 2016

with DAAs. B, Heatmap showing the expression of all SIMs during DAA treatment of cHC patients

us
normalized to the median value of all healthy individuals. The heatmap summarizes five healthy

individuals, 20 NASH patients and 28 cHC patients of which 18 cleared the infection upon therapy
an
(SVR) and 10 experienced a viral relapse. The expression of IL-12p40, MCP-3, M-CSF and LTA

was extrapolated as the expression level was more than 15-fold increased in cHC. IL-1α and IL-15
M
were excluded as the median value of healthy individuals was lower 0.1. C-D, Patients that clear the

infection are shown as grey diamonds (SVR, n=18) and patients that experienced a viral relapse are

shown as black diamonds (n=8). C, Mean expression level of SIMs (pg/ml) that are significantly
ed

elevated in cHC compared to healthy controls and significantly decrease upon successful DAA

treatment (from BL to fu12). *P<.05; **P<.01; ***P<.001; ****P<.0001, Wilcoxon test, mean and
pt

SEM are presented. D, Mean expression level of analytes (pg/mL) that are significantly reduced in

cHC.
ce
Ac
 Downloaded from http://jid.oxfordjournals.org/ at University of Hong Kong Libraries on October 2, 2016
22

ipt
cr
us
an
M
ed
pt
ce
Ac
 Downloaded from http://jid.oxfordjournals.org/ at University of Hong Kong Libraries on October 2, 2016
23

ipt
cr
us
an
M
ed
pt
ce
Ac
 Downloaded from http://jid.oxfordjournals.org/ at University of Hong Kong Libraries on October 2, 2016
24

ipt
cr
us
an
M
ed
pt
ce
Ac
 Downloaded from http://jid.oxfordjournals.org/ at University of Hong Kong Libraries on October 2, 2016
25

ipt
cr
us
an
M
ed
pt
ce
Ac
Table 1. Patient characteristics at baseline.

Healthy NASH cHC

patient number 5 20 28

HCV RNA 1,624,000

- -

ipt
(IU/mL) (1,600-7,600,000)

gender (m/f) 4/1 12/8 15/13

cr
age (years) 38.8 (27-58) 48.5 (23-72) 56.8 (41-72)

Downloaded from http://jid.oxfordjournals.org/ at University of Hong Kong Libraries on October 2, 2016

Outcome

us
- - 18/10
(SVR/Relapse)

Fibroscan (kPa) - 8.0 (3.6-23.5) 22.4 (5.8-48)

ALT (U/L) -
an
101.6 (29-242) 106.3 (31-415)

HCV Genotype
- - 12/1/13/2
M
(1/2/3/4)
ed
pt
ce
Ac
Table 2. SIMs that are differently expressed between cHCV (BL) and healthy

individuals (Mann-Whitney test).

P<.0001 P<.001 P<.01 P<.05 n.s.

IFN-α2 SCF M-CSF IL-7 GRO-α

ipt
MCP-3 IL-17 PDGF-bb IFN-γ IL-1α

β-NGF FGF-b sICAM-1 HGF IL-1RA

cr
CTACK LIF IL-18 MIF IL-2

Downloaded from http://jid.oxfordjournals.org/ at University of Hong Kong Libraries on October 2, 2016

SDF-1α LTA IL-5

us
IL-3 IL-2Ra IL-6

IL-4 IL-8

IL-1β
an IL-9

IP-10 IL-10
M
IL-12p40 IL-12p70

TRAIL IL-13

IL-15
ed

IL-16

Eotaxin
pt

G-CSF
ce

GM-CSF

MCP-1

MIG
Ac

MIP-1α

MIP-1β

RANTES

SCGF-β
 Downloaded from http://jid.oxfordjournals.org/ at University of Hong Kong Libraries on October 2, 2016
ipt
cr
us
sVCAM-1

an
VEGF
TNF

M
ed
pt
ce
Ac
Table 3. Analytes that decrease from BL to fu in SVR patients (Wilcoxon-test).

P<.0001 P<.001 P<.01 P<.05 n.s.

IL-18 IP-10 M-CSF IL-8 IL-1β

MIP-1β LIF RANTES IL-1RA

ipt
GRO-α MIG IFN-α2 IL-2

SCGF-β sICAM-1 TNF IL-3

cr
HGF SCF TRAIL IL-4

Downloaded from http://jid.oxfordjournals.org/ at University of Hong Kong Libraries on October 2, 2016

MCP-3 sVCAM-1 IL-5

us
IL-12p40 IL-6

IL-2Ra IL-7

LTA
an IL-9

IL-1α IL-10
M
IL12p70

IL-13

IL-15
ed

IL-16

IL-17
pt

β-NGF
ce

CTACK

FGF-b

G-CSF
Ac

GM-CSF

IFN-γ

MIF

MCP-1
 Downloaded from http://jid.oxfordjournals.org/ at University of Hong Kong Libraries on October 2, 2016
ipt
cr
us
PDGF-bb

an
SDF-1α
MIP-1α

VEGF

M
ed
pt
ce
Ac