JOURNAL OF INTERFERON & CYTOKINE RESEARCH FULL MANUSCRIPT

Volume 37, Number 6, 2017

ª Mary Ann Liebert, Inc.
DOI: 10.1089/jir.2016.0078

Both Hepatitis C Virus-Specific T Cell Responses

and IL28B rs12979860 Single-Nucleotide Polymorphism
Genotype Influence Antihepatitis C Virus Treatment
Outcome in Patients with Chronic Hepatitis C

José M. Benito,1,2 Javier Garcı́a-Samaniego,3,4 Marcial Garcı́a,1,2 Antonio Madejón,3,4

Luz Martı́n-Carbonero,3 Alfonso Cabello,5 Beatriz Álvarez,5 Miguel Górgolas,5 and Norma Rallón1,2

Despite new treatments for hepatitis C virus (HCV) infection, IFNa-based regimens still have clinical relevance
in special populations of patients and remain the only therapeutic option for many patients. We sought to
elucidate the interplay between two relevant factors (IL28B polymorphism and T cell immune responses)
involved in the outcome of this therapy in HCV-infected patients. We evaluated 38 patients infected with HCV
genotype 1—17 coinfected with HIV—who were undergoing a full course of pegIFNa/RBV therapy. The
interdependence and roles of T cell-mediated immune responses and IL28B rs12979860 single-nucleotide
polymorphism genotype as predictors of virological response to anti-HCV treatment in patients with chronic
hepatitis C were evaluated using nonparametric tests. Factors associated with rapid virological response (RVR)
in univariate analysis were presence of CD4 T cell response against NS3 HCV protein, low baseline HCV-
RNA, and IL28B CC genotype. Factors associated with sustained virological response (SVR) in univariate
analysis were IL28B CC genotype, low baseline HCV-RNA, and presence of CD4 response against NS2. In the
multivariate analysis, low baseline HCV-RNA and NS3-specific CD4 response showed a clear trend toward
association with RVR (P = 0.09 and P = 0.07, respectively). Regarding SVR, IL28B CC genotype was the
strongest predictor (P = 0.02), with presence of NS2-specific CD4 response showing a clear trend (P = 0.09).
HCV-specific T cell response influences the outcome of pegIFNa/RBV therapy regardless of IL28B genotype.
HCV-specific T cell responses (adaptive immunity) seem to influence viral clearance both in the short and long
term during therapy (RVR and SVR), whereas the influence of the IL28B genotype (innate immunity) may be
more relevant to the long-lasting therapeutic effect (SVR).

Keywords: HCV-specific T cell responses, rs12979860 IL28B SNP, pegIFNa/RBV, treatment outcome

Introduction world. Moreover, although current treatment guidelines are

based on IFNa-free regimens (www.hcvguidelines.org),

N owadays, eradication of hepatitis C virus (HCV)

infection seems plausible due to the significant prog-
ress made in the development and incorporation of new
IFN-based therapies may still have clinical significance in
special populations of patients with high incidence of HCV-
related hepatocellular carcinoma (HCC), since IFNa therapy
therapeutic tools, especially direct acting antivirals targeting reduces the risk of HCC (Zhang and others 2011).
different HCV enzymes (Rockstroh and Bhagani 2013; Treatment with pegylated IFNa and ribavirin (pegIFNa/
Barth 2015). However, there are important economic and RBV) has been shown to have limited effectiveness and
social obstacles to be surmounted before this goal can be serious side effects (Fried and others 2002). These side ef-
reached. New antiviral drugs are extremely expensive, thus fects and suboptimal effectiveness are exacerbated in pa-
restricting availability, especially in certain parts of the tients coinfected with HIV, who make up a difficult-to-treat

1
IIS-Fundación Jiménez Dı́az, Universidad Autónoma de Madrid, Madrid, Spain.
2
Hospital Universitario Rey Juan Carlos, Móstoles, Spain.
3
Hepatology Unit, Hospital Universitario La Paz, Madrid, Spain.
4
CIBERehd, Madrid, Spain.
5
Infectious Diseases Unit, Hospital Universitario Fundación Jiménez Dı́az, Madrid, Spain.

1
2 BENITO ET AL.

population and one with higher rates of chronicity and faster genotype as predictors of virological response to anti-HCV
liver disease progression (Hernandez and Sherman 2011). treatment in patients with chronic hepatitis C.
Therefore, despite its limitations, there is still much interest
in identifying aspects that allow for a better management of
Materials and Methods
patients treated with IFNa-based regimens. In addition, re-
search on genetics and immune mechanisms and the virus– Study population
host interaction in the context of this therapy could shed
light on correlates of immune protection, which are quite A longitudinal study was conducted in 38 patients
necessary for the development of an effective HCV vaccine. chronically infected with HCV genotype 1 (HCV-1), in-
This search, however, has proven to be much more complex cluding 21 HCV-monoinfected (HCV group) and 17 HCV/
than expected (Fauvelle and others 2013). HIV-coinfected (HCV/HIV group) patients. Twenty HCV-
To date, several host and viral factors have been de- and HIV-seronegative healthy volunteers were included as
scribed as predicting virological response to anti-HCV controls to establish reference values for a positive HCV-
therapy with IFN-based regimens (Shirakawa and others specific T cell response as previously described (Rallón
2008; Masarone and Persico 2011). Given the immuno- and others 2011a). All patients had completed a full course
modulatory effects of pegIFNa and ribavirin (Brinkmann of therapy with pegIFNa-RBV, had a validated treatment
and others 1993; Thomas and others 2003; Langhans and outcome, and had stored cell samples available for IL28B
others 2012) and the role of HCV-specific cellular immune rs12979860 SNP genotyping as previously described (Ral-
response in viral clearance among patients able to eradicate lón and others 2011b) and for HCV-specific T cell response
the virus after acute HCV infection (Brinkmann and others analysis. To participate in the study, written informed
1993), HCV-specific immune responses have been suggested consent for specific examinations was obtained from all
as a potential factor influencing IFNa-based treatment out- individuals, and the study protocol was evaluated and ap-
come, although with controversial results (Cramp and others proved by the hospital ethics committee.
2000; Thimme and others 2001; Aberle and others 2007; Capa The following on-treatment virological responses were
and others 2007; Pilli and others 2007; Rosen and others 2007; assessed: rapid virological response (RVR), defined as
Burton and others 2008; Caetano and others 2008; Arends and undetectable plasma HCV-RNA at week 4 after initiation
others 2010; Moller and others 2011; Pembroke and others of therapy; early virological response (EVR), defined as a
2012; Zhang and others 2012). decrease in HCV-RNA level of ‡2 log at week 12; end-of-
As a result of this, there is no conclusive evidence for treatment virological response (EOTVR), defined as un-
the role of HCV-specific T cell responses in treatment- detectable plasma HCV-RNA at the completion of therapy;
induced viral eradication, and this may be due to dif- and SVR, defined as undetectable plasma HCV-RNA 24
ferent reasons. Most studies analyzing the potential as- weeks after the end of treatment. Patients with poor drug
sociation between virus-specific T cell responses and compliance and/or who discontinued therapy due to side
virological response to treatment did not control for other effects were excluded from this analysis, as previously
important baseline factors predicting response, such as described (Rallón and others 2011b).
the single-nucleotide polymorphism (SNP) located near
the gene coding for IL28B or interferonl3 (IFNl3). This Viral load and HCV genotyping
SNP is a host factor that is strongly associated with both
sustained virological response (SVR) and on-treatment Plasma HCV-RNA was measured using a real-time PCR
viral kinetics (Ge and others 2009; Tanaka and others assay (COBAS TaqMan 48; Roche, Barcelona, Spain),
2009; Rallón and others 2010). which has a lower limit of detection of 10 IU/mL. HCV gen-
IFNl3 belongs to the family of type III interferons, shares otyping was performed on plasma using a commercial RT-PCR
functional characteristics with type I interferons (IFNa and hybridization assay (Versant HCV Genotype 2.0 LiPA; Sie-
IFNb), and has antiviral properties against HCV (Kotenko mens, Barcelona, Spain). Plasma HIV-RNA was measured
and others 2003; Robek and others 2005). A recent study using Versant HIV-1 RNA v 3.0 (Siemens), which has a lower
has shown that plasma levels of IFNl3 upregulate follow- limit of detection of 50 HIV-RNA copies/mL.
ing administration of exogenous IFNa as part of anti-HCV
therapy and that this upregulation is influenced by IL28B Clinical specimens
genotype, suggesting a tight in vivo relationship between
the two interferons (Rallón and others 2012). On the con- All immunological studies were done retrospectively us-
trary, in response to viral infections, the IFNl family ing baseline (before initiation of anti-HCV therapy) stored
seems to upregulate the production of IFNg (Ank and cryopreserved peripheral blood mononuclear cells (PBMCs)
others 2006), a cytokine with an important role in lym- samples. Viability of thawed PBMCs was always >85%.
phocyte recruitment into the liver tissue (Lalor and others
2002), and with a central role in regulating adaptive T cell IL28B genotyping (rs12979860 SNP)
responses (Schoenborn and Wilson 2007).
On the basis of this knowledge, it is important to clarify The characterization of IL28B allelic variants was con-
the relative role of HCV-specific T cell responses and of the ducted in a blinded manner on DNA specimens collected
IL28B genotype (as a factor related to antiviral innate re- from each individual. The SNP rs12979860 was assessed with
sponse) in the context of HCV clearance induced by IFNa- a custom TaqMan assay (Livak 1999) using the 5¢ nuclease
based anti-HCV treatment. The aim of this study was to assay with allele-specific TaqMan probes (ABI TaqMan allelic
analyze the interdependence and relative roles of T cell- discrimination kit and the ABI7900HT Sequence Detection
mediated immune responses and IL28B rs12979860 SNP System; Applied Biosystems, Carlsbad, CA).
ADAPTIVE IMMUNITY AND IL28B IN CHRONIC HCV PATIENTS 3

Overlapping HCV peptides 2011a). For each sample, a minimum of 50,000 CD4+ and
50,000 CD8+ events were acquired using an FC500 flow
A genotype-matched peptide array of 460 overlapping pep- cytometer, and data analysis was performed using CXP
tides (Biodefense and Emerging Infections Research Resources software (Beckman Coulter). Gating was done on CD8+
Repository, Manassas, VA) spanning all HCV proteins was bright and CD4+ T cells. The percentage of CD8+ bright or
used to assess HCV-specific CD4+ and CD8+ T cell immune CD4+ T cells expressing any combination of cytokines was
responses as previously described (Rallón and others 2011a). assessed using the negative control of each sample to es-
Each peptide was 12 to 19 amino acids (aa) long and overlapped tablish quadrant boundaries, as previously described (Rallón
the adjacent peptides by 11 to 12 aa. Overlapping peptides have and others 2011a). Figure 1 shows a representative example
been previously shown to be suitable for the measurement of of flow cytometry data with the gating strategy used.
both CD4+ and CD8+ T cell responses (Maecker and others On the basis of expression of the two cytokines examined,
2002). In the stimulation assays, peptides were grouped into three unique CD4+ and CD8+ T cell subsets were defined
eight pools according to HCV protein of origin: core (28 pep- (cells not producing either of the two cytokines were not
tides), E1 (29 peptides), E2/p7 (64 peptides), NS2 (32 peptides), considered). For each of these subsets, a threshold for a
NS3 (98 peptides), NS4A and NS4B (together 47 peptides), positive response was defined as the 95 percentile of 160
NS5A (71 peptides), and NS5B (91 peptides) as previously de- pooled values obtained from eight different stimulations
scribed (Rallón and others 2011a). The dimethylsulfoxide con- (with the eight HCV pools) in 20 HCV- and HIV-
centration in culture medium was <5% for all peptide pools. seronegative healthy donors, as previously described (Ral-
lón and others 2011a). This threshold was applied to each
Measurement of HCV-specific T cell responses sample after background subtraction (the response of the
negative control in each particular sample). Prevalence of
HCV-specific CD4+ and CD8+ T cell responses were response for each HCV protein was defined as the propor-
evaluated by measuring the production of two different tion of patients showing detectable (above the threshold
cytokines (IFNg and TNFa) in response to HCV genotype- level) response for that particular protein. Level of response
matched overlapping peptide pools, using multiparameter was defined as the proportion of T cells producing IFNg
flow cytometry as previously reported (Rallón and others and/or TNFa in response to each of the HCV proteins.

FIG. 1. Representative example of flow cytometry data. Patterns of CD4+ and CD8+ T cell production of IFNg and TNFa
in response to polyclonal stimulation (PMA plus ionomycin), CD28 and CD49d stimulation (unstimulated control), and
HCV core peptides (HCV core). Numbers in plots represent the percentage of positive events. HCV, hepatitis C virus; PMA,
phorbol 12-myristate 13-acetate.
4 BENITO ET AL.

Table 1. Characteristics of Patients Included in the Study

Patients (n = 38)
Characteristic HCV monoinfected (n = 21) HIV/HCV coinfected (n = 17) P
Median age (years) 48 (36–57) 40 (36–46) 0.19
Sex (% of males) 52 88 0.02
Median plasma HCV-RNA (Log IU/mL) 6.5 (6.1–6.9) 6.5 (6.4–6.8) 0.45
Median CD4 count (cells/mL) NA 442 (377–760) NA
Median CD4 count (%) 55 (46–60) 27 (22–36) <0.0001
Patients receiving HAART (%) NA 88 NA
Patients with IL28B CC genotype (%) 33 35 0.9
HAART, highly active antiretroviral therapy; HCV, hepatitis C virus.

Statistical analyses response to any of the proteins (global prevalence) was

significantly higher in the HCV group when compared to the
The main characteristics of the study population and the HCV/HIV group (62% versus 18%, P = 0.02). We observed
different parameters evaluated are expressed as median a trend toward higher prevalence of HCV-specific CD8 re-
(interquartile range, IQR). Differences between groups sponse against E1, NS2, and NS5a proteins in the HCV
were tested using nonparametric tests. Logistic regression
analysis was applied to test the influence of T cell re-
sponses and IL28B genotype on treatment outcomes ad-
justing for baseline HCV viremia and presence of HIV
coinfection. All statistical analyses were performed using
the SPSS software version 13 (SPSS, Inc., Chicago, IL).
All P values were two tailed and were considered as sig-
nificant only when below 0.05.

Results
Study population
Table 1 shows the baseline characteristics of the two
groups of patients included in the study. There were no
significant differences between HCV and HCV/HIV groups
in terms of age, baseline HCV-RNA, or IL28B genotype
distribution. There were significantly more males in the HCV/
HIV than the HCV group. As expected, the level (expressed
as proportion) of CD4 cells was significantly higher in the
HCV compared to the HCV/HIV group. Median (IQR) CD4
count in HCV/HIV patients was 442 (377–760) cells/mL. The
majority (88%) of HCV/HIV patients were receiving highly
active antiretroviral therapy and presented an undetectable
level of HIV viremia at the moment of study.

HCV-specific T cell immune response

In the whole population of patients, HCV-specific CD4 T
cell response was detectable in 60%, while CD8 T cell re-
sponse was detectable in only 37%. HCV core was the
protein with the highest prevalence of CD4 response (43%),
and HCV-E2/p7 proteins the most frequently targeted by
CD8 T cells (prevalence of 24%). Level of response was
very low overall, with median values below 0.5% in all
cases; the global level of response (the sum of responses to
the different proteins) was 0.39% (0.06–1.8) for CD4 and
0.50% (0.19–0.70) for CD8 response. Level of CD4 re-
sponse for each of the HCV proteins tended to be higher
FIG. 2. Prevalence (A) and level (B) of baseline HCV-
compared to CD8 response, although the differences were
specific CD4 and CD8 T cell response in the whole population
not statistically significant (Fig. 2). of patients. In graph A, ‘‘global’’ indicates the presence of
Prevalence of HCV-specific CD4 response tended to be detectable response to any of the proteins. In graph B,
higher in the HCV compared to the HCV/HIV group, al- ‘‘global’’ is the sum of responses against the different pro-
though the differences observed did not reach statistical teins. The scale located to the right of the y axis of graph B
significance. However, prevalence of HCV-specific CD8 applies only to global response. HCV, hepatitis C virus.
ADAPTIVE IMMUNITY AND IL28B IN CHRONIC HCV PATIENTS 5

Table 2. Rates of On-Treatment Virological

Responses According to Presence of HIV
Coinfection, IL28B Genotype, and Baseline
Hepatitis C Virus-RNA Load
RVR EVR EOTVR SVR
Patients (%) (%) (%) (%)
HCV monoinfected 19 95 80 55
(n = 21)
HCV/HIV coinfected 7 56 47 40
(n = 15)
P value 0.38 0.01 0.04 0.4
IL28B CC genotype 27 100 91 82
(n = 11)
IL28B non-CC 8 68 54 33
genotype (n = 25)
P value 0.1 0.04 0.06 0.008
Low baseline 67 100 100 100
HCV-RNA (n = 3)
High baseline 9 77 61 42
HCV-RNA (n = 33)
P value 0.045 0.9 0.3 0.04
Low baseline HCV-RNA: <600,000 IU/mL. High baseline HCV-
RNA: ‡600,000 IU/mL.
EOTVR, end-of-treatment virological response; EVR, early
virological response; HCV, hepatitis C virus; RVR, rapid virolog-
ical response; SVR, sustained virological response.

The association of HCV-specific T cell response with vi-

FIG. 3. Prevalence of baseline HCV-specific CD4 and rological response was analyzed by stratifying patients ac-
CD8 T cell response against the different HCV proteins in cording to the presence or absence of CD4 or CD8 T cell
patients carrying IL28B CC or non-CC variants. ‘‘Global’’ response against each of the HCV proteins. Figure 4 shows
indicates the presence of detectable response to any of the the rate of RVR and of SVR according to the presence or
HCV proteins. HCV, hepatitis C virus.
absence of HCV-specific CD4 response. Overall, the rate of
RVR was higher in patients showing detectable HCV-specific
compared to the HCV/HIV group, although the differences CD4 response compared to those lacking this response. This
did not reach statistical significance. difference reached statistical significance when patients were
It has been described that a higher rate of SVR after stratified according to CD4 response against NS2 protein
pegIFNa-RBV therapy occurs in patients carrying the (44% versus 0% of RVR in patients with and without de-
rs12979860 IL28B CC genotype compared to patients car- tectable CD4 response, respectively, P = 0.006), and also
rying the non-CC genotype. Thus, we stratified our study when patients were stratified according to CD4 response
population into CC and non-CC IL28B carriers and com- against NS3 protein (36% versus 4% of RVR in patients with
pared the prevalence of HCV-specific T cell response ac- and without detectable CD4 response, respectively, P = 0.02).
cording to IL28B genotype. There were no significant RVR showed a trend toward higher rates when CD4 response
differences in the prevalence of CD4 or CD8 T cell response against NS4ab protein was present (P = 0.12). When patients
between CC and non-CC patients (Fig. 3). were grouped according to the presence or absence of CD8
Factors associated with virological response response, the rate of RVR tended to be higher in patients
to anti-HCV treatment showing detectable CD8 response, although the differences
were not statistically significant (data not shown).
In the whole population of patients, rates of RVR, EVR, Rates of EVR and EOTVR were similar between patients
EOTVR, and SVR were 14%, 78%, 66%, and 49%, re- independently of the presence or absence of HCV-specific
spectively. The influence of HIV coinfection, IL28B geno- CD4 or CD8 response (data not shown). There were no
type, and baseline HCV-RNA load on on-treatment differences in the rate of SVR according to the presence or
virological responses is shown in Table 2. HCV/HIV- absence of CD8 response. However, the rate of SVR tended
coinfected patients showed a significantly lower rate of EVR to be higher in patients showing detectable CD4 response.
and EOTVR compared to HCV-monoinfected patients. Specifically, CD4 response against NS2 and NS3 proteins
IL28B CC carriers showed a significantly higher rate of exhibited a clear trend (SVR: 78% versus 39%, P = 0.10 in
EVR and SVR compared to non-CC carriers. RVR and patients with and without detectable CD4 response against
EOTVR were also higher, although the difference did not NS2; SVR: 70% versus 40%, P = 0.15 in patients with and
reach statistical significance. Finally, patients with low without detectable CD4 response against NS3) (Fig. 4).
baseline HCV-RNA load (<600,000 IU/mL) showed higher A multivariate logistic regression analysis was performed to
rates of RVR and SVR compared to patients with high ascertain which factors were significantly and independently
baseline HCV-RNA load (‡600,000 IU/mL). associated with the probability of attaining RVR (Table 3).
6 BENITO ET AL.

RVR Table 3. Baseline Factors Associated

with Rapid Virological Response
and with Sustained Virological Response
50
p=0.006
Factor OR CI P
40 p=0.02 RVR
p=0.12 Low baseline HCV-RNA 26 0.64–1083 0.09
30 load (<600,000 IU/mL)
Detectable CD4 response 11.6 0.85–16 0.07
Rate (%)

against HCV-NS3 protein

20 SVR
IL28B CC genotype 15 1.5–140 0.02
Detectable CD4 response 7.3 0.75–70 0.09
10
against HCV-NS2 protein
CI, confidence interval; HCV, hepatitis C virus; OR, odds ratio;
0 RVR, rapid virological response; SVR, sustained virological response.
Core E1 E2/p7 NS2 NS3 NS4ab NS5a NS5b Global

SVR Discussion
To our knowledge, this is the first study to analyze the
100 interdependence and relative contribution of two different
factors involved in the HCV eradication induced by peg-
p=0.10
80 p=0.15
IFNa/RBV treatment: one related to adaptive antiviral im-
mune response and another related to innate antiviral
immune response. The main findings of our study are as
60 follows: first, IL28B genotype (as a factor related to innate
Rate (%)

immunity) is not associated with T cell mediated anti-HCV

40 responses; second, both factors contribute significantly and
independently to predictions of virological response after
pegIFNa/RBV treatment; third, HCV-specific T cell re-
20
sponses (adaptive immunity) seem to influence viral clear-
ance during the short (RVR) and the long (SVR) term of
0 therapy, whereas the influence of IL28B genotype (innate
Core E1 E2/p7 NS2 NS3 NS4ab NS5a NS5b Global immunity) may be more relevant in the long term (prelim-
inary results were reported by Rallón and others 2015).
Presence of CD4 T cell response
IL28B genotype is a strong predictor of treatment-
Absence of CD4 T cell response induced HCV eradication (Ge and others 2009; Tanaka and
others 2009; Rallón and others 2010), although the mech-
FIG. 4. Rate of RVR and SVR according to the presence anisms underlying this association are not yet understood.
or absence of hepatitis C virus-specific CD4 T cell response. A boosting of innate antiviral immunity through modula-
The term ‘‘global’’ refers to the presence of detectable re- tion of IFNl3 production, either at baseline and/or after
sponse to any of the proteins. RVR, rapid virological re- exogenous administration of IFNa, could be involved, al-
sponse; SVR, sustained virological response.
though the results surrounding this issue are controversial
(Langhans and others 2011; Rallón and others 2012).
Alternatively, IL28B genotype could influence the ability
Independent (explanatory) variables included in the analysis to mount an adaptive T cell-mediated anti-HCV response. In
were IL28B genotype (CC versus non-CC), baseline HCV- the simian model of AIDS, it has been shown that IFNl3
RNA load (low versus high), and HCV NS3-specific CD4 increases CD4 response to vaccination with HIV DNA, as
response (presence versus absence). Factors showing a clear well as the killing activity of CD8 cells (Morrow and others
trend toward an association with RVR were (odds ratio [95% 2009, 2010). These findings suggest that IL28B genetic
CI], P value) detectable CD4 response against HCV-NS3 variation could potentially impact cellular immune re-
protein (11.6 [0.85–160], P = 0.07) and low baseline HCV- sponses against viral pathogens. However, our study does
RNA load (26 [0.64–1083], P = 0.09), whereas IL28B CC not support this hypothesis since there was no correlation
genotype did not show this same trend (P = 0.2). The same between HCV-specific T cell responses and IL28B genotype
analysis applied to SVR and, including IL28B genotype (CC in HCV-infected patients. This finding of ours is in agree-
versus non-CC), baseline HCV-RNA load (low versus high), ment with a very recent study in HCV spontaneous resolvers
and HCV NS2-specific CD4 response (presence versus ab- and in chronically infected HCV patients (Bes and others
sence), revealed that IL28B CC genotype was the variable 2012). Nevertheless, a potential association of IL28B ge-
most strongly associated with SVR (15 [1.5–140], P = 0.02) notype with other characteristics of T cell responses not
followed by detectable CD4 response against HCV-NS2 measured in our study cannot be ruled out.
protein, which showed a clear trend (7.3 [0.75–70], P = 0.09), On the contrary, our study is useful in elucidating the role of
whereas HCV load was not associated (P = 0.9). HCV-specific adaptive immunity in controlling HCV infection
ADAPTIVE IMMUNITY AND IL28B IN CHRONIC HCV PATIENTS 7

after treatment, which is an issue that has been widely debated ven by IL28B genetic variation (Rallón and others 2012)
over time, with some studies reporting an association (Cramp could be involved in this long-term antiviral effect. Further
and others 2000; Sreenarasimhaiah and others 2003; Mor- studies measuring the levels of IFNl3 longitudinally during
ishima and others 2006; Aberle and others 2007; Pilli and pegIFNa/RBV treatment are needed to test this hypothesis.
others 2007; Rosen and others 2007; Caetano and others 2008; The main strengths of our study are the prolonged obser-
Moller and others 2011; Zhang and others 2012) but not others vation period in a population of patients being treated for
(Lauer and others 2005; Capa and others 2007; Burton 2008; HCV infection, lasting the entire period of follow-up, and the
Barnes and others 2009; Arends and others 2010; Pembroke detailed analysis of immunological responses, IL28B poly-
and others 2012). Several reasons may account for these morphism, and several on-treatment outcomes with clinical
contradictory results, including differences in study design relevance. Moreover, we were able to test the contribution
(cross sectional or longitudinal), different patient cohorts (in- and interdependence of both HCV-specific T cell responses
cluding different HCV genotypes or restricted to only one and IL28B genotype to anti-HCV therapy outcome.
genotype), differences in ways of measuring treatment out- However, some limitations of our study merit comment,
come (either early or long-term viral kinetics), and different beginning with the relatively small number of patients ana-
methodological approaches to evaluate T cell responses. lyzed together with the fact that some other factors impacting
We used a longitudinal design in a cohort of patients infected on treatment outcome were not controlled in our study; these
with HCV genotype 1 (the most difficult-to-treat genotype) and likely prevented most of the results from reaching a level of
evaluated T cell responses using overlapping peptides that en- significance below 0.05. In addition, we did not measure
compassed the whole HCV proteome. Moreover, our cohort of other characteristics of T cell responses and thus cannot rule
patients also included patients coinfected with HIV (an addi- out the existence of other potential associations between T
tional factor of poor response to treatment). Using this approach, cell responses and virological response. However, it must be
we found that the presence of HCV-specific T cell responses at noted that even with these limitations, we found a clear trend
baseline is associated with achievement of virological response toward an association between T cell responses and treatment
to pegIFNa/RBV treatment. Interestingly, this association was outcome. These findings make the case for further studies
restricted to CD4 responses against nonstructural proteins, and with larger cohorts to validate our results.
was more apparent with viral clearance during the first weeks of In summary, our study is the first to explore the interde-
therapy (RVR), although it was also present with outcome in the pendence and roles of T cell-mediated immune responses
long term (SVR). and IL28B genotypes on several clinically relevant on-
Our results regarding the association between T cell re- treatment outcomes as predictors of virological response to
sponses and early viral kinetics (RVR) are in agreement pegIFNa/RBV therapy in patients with chronic HCV in-
with two previous studies (Aberle and others 2007; Pilli and fection. Our results suggest that HCV-specific T cell re-
others 2007); indeed, this association was independent of sponses (adaptive immunity) influence viral clearance
other well-known predictors of response, such as the IL28B throughout HCV treatment (RVR and SVR) independently
genotype, which is one of the most interesting and original of IL28B genotype. Within this context, the influence of
findings from our study. Our finding strongly supports the IL28B genotype (innate immunity) is more significant on the
role of virus-specific T cell response in viral kinetics during long-lasting effect of therapy (SVR).
the first weeks of therapy, corresponding to the second phase
of HCV viremia decline (Neumann and others 1998), which Acknowledgments
has been attributed to the death of infected hepatocytes
driven by HCV-specific immune responses (Neumann and The authors thank all the patients who participated in the
others 1998; Layden and Layden 2002), and is in agreement study. This study was funded by projects RD12/0017/0031
with the established role of this type of immune response in and PI14/00518 (part of the Plan Nacional I+D+I) and was
the spontaneous clearance of HCV during acute infection cofunded by ISCIII-Subdirección General de Evaluación
(Thimme and others 2001). and the European Regional Development Fund. Norma
A potential mechanism underlying this association could Rallón is a Miguel Servet investigator from the ISCIII
be the boosting of T cell response (present before the (CP14/00198), Madrid, Spain. Marcial Garcı́a is a predoc-
starting of pegIFNa/RBV therapy) after the very first re- toral student cofunded by the CP14/00198 project and In-
duction of HCV viremia induced by therapy, and this en- tramural Research Scholarship from IIS-FJD.
hanced T cell response may contribute to HCV elimination
thereafter. This hypothesis is supported by previous studies Author Disclosure Statement
showing that HCV-specific T cell responses are boosted in
parallel with a rapid reduction of HCV during the first weeks No competing financial interests exist.
of therapy (Kamal and others 2002; Aberle and others
2007). In fact, our results demonstrating a clear trend toward References
an association of RVR with T cell response at baseline Aberle J, Perstinger G, Weseslindtner L, Sinzinger U, Gurguta
(before the administration of treatment) provide empirical C, Steindl-Munda P, et al. 2007. CD4+ T cell responses in
support of the role of T cell responses as the cause of HCV patients with chronic hepatitis C undergoing peginterferon/
clearance during the first weeks of therapy. ribavirin therapy correlate with faster, but not sustained, viral
On the contrary, IL28B CC genotype was the strongest clearance. J Infect Dis 195:1315–1319.
predictor of SVR, suggesting a role played by innate im- Ank N, West H, Bartholdy C, Eriksson K, Thomsen A, Paludan
munity in long-term viral kinetics. Given the anti-HCV S. 2006. Lambda interferon (IFN-l), a type III IFN, is induced
action of IFNl3 (Kotenko and others 2003; Robek and by viruses and IFNs and displays potent antiviral activity
others 2005), a modulated expression of this molecule dri- against select virus infections in vivo. J Virol 80:4501–4509.
8 BENITO ET AL.

Arends JE, Claassen M, van den Berg C, Nanlohy N, van Er- ferential effects on functions of CD4+ Th1, Th2, and
pecum KJ, Baak BC, et al. 2010. T-cell responses at baseline regulatory T cell clones in hepatitis C. PLoS One 7:e42094.
and during therapy with peginterferon-a and ribavirin are not Lauer GM, Lucas M, Timm J, Ouchi K, Kim AY, Day CL, et al.
associated with outcome in chronic hepatitis C infected pa- 2005. Full-breadth analysis of CD8+ T-cell responses in acute
tients. Antiviral Res 87:353.360. hepatitis C virus infection and early therapy. J Virol 79:12979–
Barnes E, Gelderblom HC, Humphreys I, Semmo N, Reesink 12988.
HW, Beld M, et al. 2009. Cellular immune responses during Layden JE, Layden TJ. 2002. Viral kinetics of hepatitis C: new
high-dose interferon-a induction therapy for hepatitis C virus insights and remaining limitations. Hepatology 35:967–970.
infection. J Infect Dis 199:819–828. Livak K. 1999. Allelic discrimination using fluorogenic probes
Barth H. 2015. Hepatitis C virus: is it time to say goodbye yet? and the 5¢nuclease assay. Genet Anal 14:143–149.
Perspectives and challenges for the next decade. World J Maecker H, Dunn H, Suni M, Khatamzas E, Pitcher CJ, Bunde T,
Hepatol 7:725–737. et al. 2002. Use of overlapping peptide mixtures as antigens for
Bes M, Sauleda S, Campos-Varela I, Rodriguez-Frias F, Ca- cytokine flow cytometry. J Immunol Methods 255:27–40.
samitjana N, Homs M, et al. 2012. IL28B genetic variation Masarone M, Persico M. 2011. Antiviral therapy: why does it
and hepatitis C virus-specific CD4+ T-cell responses in anti- fail in HCV-related chronic hepatitis? Expert Rev Anti Infect
HCV-positive blood donors. J Viral Hep 19:867–871. Ther 9:535–543.
Brinkmann V, Geiger T, Alkan S, Heusser CH. 1993. Interferon Moller JF, Moller B, Wiedenmann B, Berg T, Schott E. 2011.
alpha increases the frequency of interferon gamma-producing CD154, a marker of antigen-specific stimulation of CD4 T
human CD4+ T cells. J Exp Med 178:1655–1663. cells, is associated with response to treatment in patients with
Burton JR, Klarquist J, Im K, Smyk-Pearson S, Golden-Mason L, chronic HCV infection. J Viral Hep 18:e341–e349.
Castelblanco N, et al. 2008. Prospective analysis of effector Morishima C, Polyak SJ, Ray R, Doherty MC, Di Bisceglie
and regulatory CD4+ T cells in chronic HCV patients under- AM, Malet PF, et al. 2006. Hepatitis C virus-specific immune
going combination antiviral therapy. J Hepatol 49:329–338. responses and quasi-species variability at baseline are asso-
Caetano J, Martinho A, Paiva A, Pais B, Valente C, Luxo C. ciated with nonresponse to antiviral therapy during advanced
2008. Differences in hepatitis C virus (HCV)-specific CD8 T- hepatitis C. J Infect Dis 193:931–940.
cell phenotype during pegylated alpha interferon and ribavirin Morrow MP, Pankhong P, Laddy DJ, Schoenkly KA, Yan J,
treatment are related to response to antiviral therapy in patients Cisper N, et al. 2009. Comparative ability of IL-12 and IL-
chronically infected with HCV. J Virol 82:7567–7577. 28B to regulate Treg populations and enhance adaptive cel-
Capa L, Soriano V, Garcı́a-Samaniego J, Nuñez M, Romero M, lular immunity. Blood 113:5868–5877.
de Mendoza C, et al. 2007. Evolution of T-cell responses to Morrow MP, Yan J, Pankhong P, Shedlock DJ, Lewis MG,
hepatitis C virus (HCV) during pegylated interferon plus ri- Talbott K, et al. 2010. IL-28B/IFN-lambda 3 drives granzyme
bavirin treatment in HCV-monoinfected and in HCV/HIV- loading and significantly increases CTL killing activity in
coinfected patients. Antivir Ther 12:459–468. macaques. Mol Ther 18:1714–1723.
Cramp ME, Rossol S, Chokshi S, Carucci P, Williams R, Neumann AU, Lam NP, Dahari H, Grecht DR, Wiley TE, Layden
Naoumov N. 2000. Hepatitis C virus-specific T-cell reactivity TJ, et al. 1998. Hepatitis C viral dynamics in vivo and the anti-
during interferon and ribavirin treatment in chronic hepatitis viral efficacy of interferon-a therapy. Science 282:103–107.
C. Gastroenterology 118:346–355. Pembroke T, Rees I, Gallagher K, Jones E, Mizen P, Navruzov
Fauvelle C, Lepiller Q, Felmlee DJ, Fofana I, Habersetzer F, T, et al. 2012. Rapad early innate control of hepatitis C virus
Stoll-Keller F, et al. 2013. Hepatitis C virus vaccines— during IFN-a treatment compromises adaptive CD4+ T-cell
progress and perspectives. Microb Pathog 58:66–72. immunity. Eur J Immunol 42:2383–2394.
Fried MW, Shiffman ML, Reddy KR, Smith C, Marinos G, Pilli M, Zerbini A, Penna A, Orlandini A, Lukasiewicz E,
Goncales FL, et al. 2002. Peginterferon alfa-2a plus ribavirin Pawlotsky JM, et al. 2007. HCV-specific T-cell response in
for chronic hepatitis C virus infection. N Engl J Med 347: relation to viral kinetics and treatment outcome (DITTO-
975–982. HCV project). Gastroenterology 133:1132–1143.
Ge D, Fellay J, Thompson A, Simon J, Shianna K, Urban T, Rallón NI, Soriano V, Naggie S, Restrepo C, McHutchison J,
et al. 2009. Genetic variation in IL28B predicts hepatitis C Vispo E, et al. 2012. Impact of IL28B gene polymorphisms
treatment-induced viral clearance. Nature 461:399–401. on interferon-l3 plasma levels during pegylated interferon-a/
Hernandez MD, Sherman K. 2011. HIV/hepatitis C coinfection ribavirin therapy for chronic hepatitis C in patients coinfected
natural history and disease progression. Curr Opin HIV AIDS with HIV. J Antimicrob Chemother 67:1246–1249.
6:478–482. Rallón N, Garcı́a-Samaniego J, Madejón A, de la Vega E, Cor-
Kamal SM, Fehr J, Roesler B, Peters T, Rasenack JW. 2002. tejoso L, Romero M, et al. Adaptive immunity and rs12979860
Peginterferon alone or with ribavirin enhances HCV-specific IL28B polymorphism in chronic hepatitis C patients treated
CD4+ T helper 1 responses in patients with chronic hepatitis with pefIFN/RBV. Abstract P.A. 26.21. 4th European Con-
C. Gastroenterology 123:1070–1083. gress of Immunology, Vienna, September 6–9, 2015.
Kotenko S, Gallager G, Baurin V, Lewis-Antes A, Shen M, Rallón N, Naggie S, Benito JM, Medrano J, Restrepo C,
Shah NK, et al. 2003. IFN-lambdas mediate antiviral pro- Goldstein D, et al. 2010. Association of a single nucleotide
tection through a distinct class II cytokine receptor complex. polymorphism near the interleukin-28B gene with response to
Nat Immunol 4:69–77. hepatitis C therapy in HIV/hepatitis C virus-coinfected pa-
Lalor P, Shields P, Grant A, Adams D. 2002. Recruitment of tients. AIDS 24:F23–F29.
lymphocytes to the human liver. Immunol Cell Biol 80:52–64. Rallón N, Soriano V, Naggie S, Restrepo C, Goldstein D,
Langhans B, Kupfer B, Braunschweiger I, Arndt S, Schulte W, Vispo E, et al. 2011b. IL28B gene polymorphisms and viral
Nischalke HD, et al. 2011. Interferon-lambda serum levels in kinetics in HIV/hepatitis C virus-coinfected patients treated
hepatitis C. J Hepatol 54:859–865. with pegylated interferon and ribavirin. AIDS 25:1025–1033.
Langhans B, Nischalke HD, Arndt S, Braunschweiger I, Nat- Rallón N, Soriano V, Restrepo C, Garcia-Samaniego J, Labarga
termann J, Sauerbruch T, et al. 2012. Ribavirin exerts dif- P, López M, et al. 2011a. HCV-specific T-cell responses in
ADAPTIVE IMMUNITY AND IL28B IN CHRONIC HCV PATIENTS 9

HIV/hepatitis C virus-coinfected patients on highly active Thimme R, Oldach D, Chang K, Steiger C, Ray S, Chisari F.
antiretroviral therapy are comparable to those observed in 2001. Determinants of viral clearance and persistence
hepatitis C virus-monoinfected individuals. J Acquir Immune during acute hepatitis C virus infection. J Exp Med 194:
Defic Syndr 57:1–8. 1395–1406.
Robek MD, Boyd BS, Chisari FV. 2005. Lambda interferon Thomas H, Foster G, Platis D. 2003. Mechanisms of action of
inhibits hepatitis B and C virus replication. J Virol 79:3851– interferon and nucleoside analogues. J Hepatol 39:S93–S98.
3854. Zhang CH, Xu GL, Jia WD, Li JS, Ma JL, Ge YS. 2011. Effects
Rockstroh J, Bhagani S. 2013. Managing HIV/hepatitis C co- of interferon treatment on development and progression of
infection in the era of direct acting antivirals. BMC Med hepatocellular carcinoma in patients with chronic virus in-
11:234–242. fection: a meta-analysis of randomized clinical trials. Int J
Rosen HR, Weston SJ, Im K, Yang H, Burton JR, Erlich H, Cancer 129:1254–1264.
et al. 2007. Selective decrease in hepatitis C virus-specific Zhang Y, Liu Y, Zhao Y, Shi L, Ma L, Tan H, et al. 2012.
immunity among African Americans and outcome of antivi- Hepatitis C virus nonstructural protein specific T cells are
ral therapy. Hepatology 46:350–358. associated with virological responses to combination therapy
Schoenborn JR, Wilson CB. 2007. Regulation of interferon- in chronic HCV patients. Liver International 32:102–109.
gamma during innate and adaptive immune responses. Adv
Immunol 96:41–101.
Shirakawa H, Matsumoto A, Joshita S, Komatsu M, Tanaka N,
Umemura T, et al. 2008. Pretreatment prediction of viro- Address correspondence to:
logical response to peginterferon plus ribavirin therapy in Dr. José Miguel Benito
chronic hepatitis C patients using viral and host factors. Laboratorio de Investigación en VIH y Hepatitis Vı́ricas
Hepatology 8:1753–1760. IIS-Fundación Jiménez Dı́az
Sreenarasimhaiah J, Jaramillo A, Crippin J, Lisker-Melman M, Universidad Autónoma de Madrid
Chapman WC, Mohanakumar T. 2003. Concomitant aug- Avenida Reyes Católicos, 2
mentation of type 1 CD4+ and CD8+ T-cell responses during Madrid 28040
successful interferon-a and ribavirin treatment for chronic Spain
hepatitis C virus infection. Hum Immunol 64:497–504.
Tanaka Y, Nishida N, Sugiyama M, Kurosaki M, Matsuura K, E-mail: jbenito1@hotmail.com;
Sakamoto N, et al. 2009. Genome-wide association of IL28B jose.benito@hospitalreyjuancarlos.es
with response to pegylated interferon-alpha and ribavirin
therapy for chronic hepatitis C. Nat Genet 41:1105–1109. Received 24 August 2016/Accepted 16 December 2016