Sunday Parallel Session 13, October 29, 2006
Understanding and Enhancing HCV Treatment
M. Trippler;
S. Bein; G. Gerken; J. F. Schlaak.
Introduction:
We have previously shown that the
transcriptional response to IFN-α varies between different donors in vitro
(JBC, 277[51], 49428-37, 2002) which led us to hypothesize that the analysis of
the in vivo IFN response in HCV patients might reveal specific profiles that
are associated with response and non-response to therapy.
Aims:
To identify those profiles, a total
of 39 Caucasian HCV patients were treated with standard combination therapy
consisting of either pegylated IFN-α2b (Pegintron
(PI), 1.5µg/kg body weight once weekly) for 12 months (HCV genotype 1, n=14) or
6 months (HCV genotype 2/3, n=1/6), or pegylated IFN-α2a (Pegasys (PS),
180µg once weekly) for 12 months (HCV genotype 1, n=14) or 6 months (HCV genotype
2/3, n=1/3), both in combination with ribavirin (800-1200 mg daily).
Methods:
Peripheral blood was collected 12h
before and 12h after the first injection of IFN-α and the transcriptional
profile was analysed using human genomic microarrays
(Affymetrix HG U133A) and quantitative real-time
RT-PCR. Three data sets were obtained and analysed:
a)
basal expression (BE): expression level 12h before
the first injection of IFN,
b)
induced expression (IE): expression level 12h after
the first injection and,
c)
fold change (FC): fold difference between BE and IE.
Then, class prediction was
performed using the PAM software package. 18 patients were non-responders (NR)
to therapy while 20 patients had a sustained virological response (SVR) and one
patient was lost to follow-up.
Results:
The data indicate that the
immediate antiviral response at 36 h after IFN (IVR) and the SVR can be
predicted in all three data sets with different levels of accuracy. FC was the
best parameter to predict IVR and SVR. IVR for PI and PS could be predicted
with 39 or 19 genes and 80% or 95 % accuracy while SVR for PI and PS could be
predicted with 25 or 11 genes and 80% or 95% accuracy, respectively. When PI
and PS were combined, 75 genes were needed to predict an SVR with 87% accuracy.
Interestingly, all responders were classified correctly while 30% of NR were
classified incorrectly. The ISGs that predicted SVR
did not overlap with those associated with the immediate antiviral activity of
IFN.
Conclusion:
These data indicate that a sustained
virological response in HCV patients can be predicted by analysis of the transcriptome before or at the very beginning of
combination therapy. It suggests that the SVR is largely dependent upon the
genetic background of the patient provided she or he is adherent to therapy.
Finally, the functional analysis of the differentially regulated genes that
were identified in this study could lead to the discovery of novel drug targets
to improve the efficacy of HCV therapy.
H.
Conjeevaram; J. E. Everhart; N. Afdhal;
R. S. Brown; M. W. Fried; C. D. Howell; L. J. Jeffers; N. A. Terrault; T. E. Wiley; J. H. Hoofnagle; A. Wahed.
Introduction
Chronic hepatitis C is associated
with an increased prevalence of insulin resistance (IR), which may be related
to hepatitis C virus (HCV) itself or to metabolic factors such as obesity. The
effect of antiviral treatment on IR is not well known.
Aims:
To examine metabolic responses
(weight and insulin sensitivity) during and after peginterferon and ribavirin
therapy of chronic hepatitis C.
Methods:
Virahep-C was a
prospective, multi-center study of combination antiviral therapy among treatment
naïve African American (AA) and Caucasian American (CA) genotype 1 infected
pts. 196 AA and 205 CA pts were treated with peginterferon alfa-2a (180 mcg/wk)
and ribavirin (1000-1200 mg/day) for up to 48 wks. Pts who were HCV RNA
positive at wk 24 were considered non-responders and therapy stopped; those who
were RNA negative were treated for an additional 24 wks. IR was determined by
HOMA (homeostasis model assessment)-index based upon fasting serum glucose and
insulin.
Results:
After 24 and 48 wks of treatment,
average BMI decreased by 1.22 kg/m2 and 2.24 kg/m2 (p<.001 for both) and
HOMA levels decreased by 0.94 (p=.006) and 1.00 (p=.10). Average weight loss
was 3.6 kg (p<.001) at 24 and 6.4 kg (p<.001) at 48 wks, with similar
decreases regardless of initial weight in both sexes and races. Mean reduction
in HOMA-index at 24 and 48 wks was significant among males (-1.44, p=.007 and
-1.42, p=.01) but not females (0.01 and -0.17) and among CA (-0.95, p=.04 and
-1.15, p<.005) but not AA (-0.93 and -0.71). Sustained virological response
(SVR) occurred in 40% and relapse in 15% of pts. Weight loss did not correlate
with virological response at 24 weeks. After therapy, BMI returned towards
baseline in both SVR and relapse subjects and was not significantly different
from baseline values in any group by wk 72. Changes in HOMA paralleled those of
BMI and at 72 wks were comparable among non-responders, relapsers and pts with
SVR. There was no association of change in HOMA with viral response at 24 weeks
or with SVR and similarly no significant relationship with change in glucose
and viral response indicating a lack of association of viral clearance with IR
or glucose levels.
Summary/Conclusions:
In pts with HCV genotype 1
infection, treatment with peginterferon and ribavirin is associated with
significant decreases in BMI and improvement in insulin sensitivity. These
effects are transient and reversed once treatment is discontinued. Eradication
of HCV does not appear to result in sustained improvement in IR.
Y.
Asahina; N. Izumi; N. Komatsu; N. Umeda;
T. Hosokawa; K. Ueda; K. Tsuchiya; H. Nakanishi; J. Itakura;
M. Kurosaki; M. Uchihara; S. Miyake.
Introduction:
Mechanisms involving resistance to interferon (IFN) and
ribavirin (RBV) have not been fully elucidated, and it is difficult to predict
treatment responses before initiation of PEG-IFN plus RBV therapy. The
ISG15/USP18 system, a novel IFN-regulatory pathway, may play an important role
in innate immune functions, suggesting that the balance between ISG15 and USP18
can contribute to the IFN-regulated antiviral state. Also, a helicase, RIG-I,
and its adaptor molecule, CARDIF, may play a key role in innate immunity.
Aim:
To elucidate the mechanisms underlying resistance of to
PEG-IFN plus RBV, and to determine whether analysis of ISG15/USP18 and
RIG-I/CARDIF are useful in predicting responses to therapy.
Methods:
Fifty-seven patients chronically infected with HCV-1b were
treated with PEG-IFN alfa-2b plus RBV for 48 weeks. mRNA expression levels in
their liver and PBMC were measured using real-time PCR before therapy was
initiated. G3PDH was used as an internal control. Patients with non-viral liver
disease were also analyzed. Serum HCV dynamics were analyzed by real-time PCR.
Results:
Final outcome was as follows: 27 were sustained viral
responders (SVR), 17 were relapsers, and 12 were non-responders (NR) who were
positive for HCV-RNA throughout the treatment period. Hepatic expression levels
of ISG15, USP18 and RIG-I in patients with HCV were 9-, 5- and 3-fold higher
than in non-HCV patients. Hepatic ISG15 expression was significantly higher in NR
compared with SVR or relapser (NR=1.0±0.10;
SVR=0.25±0.25; relapser=0.52±0.40 copies/G3PDH,
p=0.001, 0.02). Similarly, hepatic USP18 expression, which was significantly
correlated with that of ISG15 (r2=0.88, p<0.0001), was
significantly higher in NR compared with SVR or relapser
(NR=1.1±0.06; SVR=0.43±0.21; relapser=0.78±0.34
copies/G3PDH, p=0.001, 0.02). Expression levels of both genes were
significantly correlated with viral decline rates at the 1st and 2nd phases of
HCV dynamics (ISG15: r=-0.66, p=0.005; r=-0.75, p=0.0009, USP18: r=-0.75,
p=0.004; r=-0.71, p=0.002). Both ISG15 and USP18 expression levels in PBMC were
correlated with hepatic expression and were significantly higher in NR compared
with SVR or relapser. Expression of RIG-I was similar
to those of ISG15/USP18, but negatively correlated with CARDIF expression.
Multivariate analysis revealed that higher expression of USP18 and older age
were predictive of NR.
Conclusion:
Higher expression levels of ISG15/USP18 and RIG-I in PBMC and
liver correlated with virological non-responsiveness to PEG-IFN plus RBV.
Therefore, measurement of these gene expression levels in PBMC and in liver may
be useful to predict the final response to therapy.
D.
Sylvestre; A. Smith; L. Barrett; J. B. Cohen.
Background:
Active drug injectors are the group most likely to be
affected by HCV and are responsible for transmitting a majority of new cases.
Strategies to simultaneously reduce drug use and treat HCV have the best chance
of reducing the scope and consequences of this disease.
Methods:
The aim of this prospective pilot safety study was to
initiate HCV treatment in 50 street-recruited heroin users after stabilizing
their drug use for 12-24 weeks using the sublingual partial opioid
receptor agonist buprenorphine (BPN). Subjects were
screened for active HCV at sites of syringe exchange. BPN therapy was offered
at a community-based medical clinic and weekly attendance was required. Subjects
who maintained at least 75% of program attendance for 12-24 weeks, regardless
of intercurrent drug use, could choose to undergo
integrated BPN/HCV treatment or instead elect a 12 week BPN taper. HCV
treatment was undertaken with concurrent BPN maintenance and was followed by a
24-week BPN taper. Drug use was measured by self report as well as random
monthly urine toxicology tests.
Results:
Of 415 screened, 272 (65%) were eligible for study, and of
those 188 (69%) completed enrollment. 143 (76%) of enrollees initiated BPN treatment. 65 patients (45%)
remained on BPN for at least 12 weeks and were given the option of initiating
HCV treatment: 55 chose HCV treatment and 10 chose a BPN taper. To date, 53
have completed study: 36 (68%) completed treatment and there were 10 (19%)
early terminations due to side effects or failure to show, 3 (6%) were
incarcerated, and 4 (8%) had medical terminations. There are 19 (36%) sustained
responders to date, even though 44 (80%) had at least one episode of illicit drug
use during HCV treatment.
Conclusion:
Offering HCV treatment in the context of concurrent drug
treatment for heroin use appears to be an effective means of engaging
challenging street-recruited active drug users, and may even be an interesting
strategy to improve motivation toward drug treatment. Retention in treatment
for 12-24 weeks, as assessed by >75% of weekly program attendance, is a
reasonable strategy for identifying those who can safely and effectively
undergo HCV therapy, and the majority who thus stabilize will elect to do so.
Although virologic outcomes may be modestly lower than those of nonaddicted individuals, the majority will successfully
complete treatment even in the setting of intercurrent
drug use. These results argue for increased involvement of addiction
specialists in hepatitis C diagnosis and treatment.
Sponsor:
NIH 1R01DA015629
Y.
Zhou; U. Müh; D. Bartel; B.
Hanzelka; G. Rao; T. Kieffer; A. D. Kwong; C. Lin.
Background:
Telaprevir (TVR,
VX-950) is a highly selective HCV NS3-4A protease inhibitor that has
demonstrated strong antiviral activity in patients infected with genotype 1 HCV
when dosed alone or in combination with pegylated interferon alfa (PEG-IFN)
with or without ribavirin (RBV). Continuous decline and a median 5.5-log10
reduction in the plasma HCV RNA load was observed in patients dosed with telaprevir + PEG-IFN and the viral load dropped below the
limit of detection (LOD, <10 IU/mL) in 4 of 8 patients by the end of the
14-day dosing period. In 12 of 12 patients who received telaprevir
+ PEG-IFN + RBV for 28 days, the plasma HCV load dropped below the LOD.
Although NS3 protease variants were observed in some patients during the first
week of dosing with telaprevir + PEG-IFN +/- RBV,
subsequently HCV RNA became undetectable. Thus, NS3 protease variants
conferring in vitro resistance to telaprevir appear
to be suppressed by PEG-IFN, a hypothesis we tested in vitro in the studies
reported here.
Methods:
Variant proteases containing
substitutions at residues 36, 54, 155, or 156 of the HCV NS3 protease domain
were expressed and purified to homogeneity, and the potency of telaprevir against these variant enzymes was determined.
The variants were also introduced into a HCV replicon and cell lines containing
the variant proteases were selected with neomycin. The anti-HCV activity
against these variant replicon cell lines was determined for telaprevir and IFN-alpha.
Results:
Single variants, including V36M/A,
T54A, or R155K/M, conferred low-level (<10-fold change in sensitivity) in
vitro resistance to telaprevir. The A156V/T single
variants or V36M/A+R155K/T double variants had high-level (>40-fold) in
vitro resistance to telaprevir. Computational modeling
analysis suggests that there are 3 distinct mechanisms of resistance for these
variants: (i) an increase in flexibility of the
binding pocket; (ii) a loss of interaction (van der
Waals contact, hydrogen bond, or salt bridge); (iii) or an increase in steric hindrance. However, the IFN-alpha IC50s of all these
variants were similar to that of wild-type replicon cells, suggesting that
these variants remain fully sensitive to IFN-alpha.
Conclusions:
These in vitro studies confirm that
NS3 variants observed during telaprevir dosing have
reduced sensitivity to telaprevir, but retain full
sensitivity to IFN-alpha. Thus, they provide a rationale for combination
therapy with telaprevir + PEG-IFN and are consistent
with the results of two clinical studies of telaprevir
+ PEG-IFN +/- RBV, in which continuous decline of HCV RNA was observed in all
patients without evidence of viral breakthrough.
T.
Kanto; I. Itose; M. Inoue; M. Miyazaki; H. Miyatake; M. Sakakibara; T. Yakushijin; T. Oze; N. Hiramatsu; T. Takehara; A. Kasahara; N. Hayashi.
Backgrounds and aim:
A combination of PEG-IFNα and
ribavirin has been used for the treatment of chronic hepatitis C. One of the
reasons for the treatment failure is a high relapse rate after the cessation of
therapy. In order to suppress HCV reappearance, the treatment extension from 48
up to 72 weeks needs to be considered. However, no reliable marker is currently
available for defining potential relapsers. It is well known that the
enhancement of Th1 response is closely related to HCV clearance. Dendritic cells (DC) are potent antigen presenting cells that
regulate Th1/Th2 differentiation, which are numerically and functionally impaired
in HCV infection. However, little is known whether the restoration of DC
function contributes to a successful outcome in the combination therapy. We
thus aimed to evaluate the feasibility of blood DC frequencies and their
function for the predictors of relapse.
Methods:
Twenty-five CHC patients with HCV genotype 1 and high titer were enrolled in this study. They received 48 weeks
of PEG-IFNα2b and ribavirin and their virological response was determined
at week 12 of therapy and 24 weeks after its completion. During the treatment,
frequencies of myeloid DC (MDC) and plasmacytoid DC
(PDC) and their changes from the beginning of the therapy were determined by
means of flow cytometoric analyses. The ability of
patient DC to stimulate allogeneic CD4+ T cells was
assessed at the end and after the therapy by determining the ratio of DC-primed
T cell proliferation to those in healthy counterparts.
Results:
Among 25 patients who completed 48-week treatment, 11
patients achieved sustained virological response (SVR), 11 were transient
response (TR) and 3 were non-response (NR), respectively. In comparison between
the SVR and TR groups, MDC and PDC frequencies did not differ throughout the
therapy. By tracing the changes from the beginning to week 12, MDC frequency
was not different between the groups. In contrast, the PDC frequency
significantly declined in TR group (p<0.05), whereas those in SVR did not.
At week 48 and thereafter, allostimulatory capacity
of DC in TR was sustained to be lower than those in SVR patients (p<0.05).
Even in patients who once attained negative serum HCVRNA at week 12, PDC
decrease and impaired DC function were more significant in TR than in SVR
(p<0.05). Negative predictive value of DC function at week 48 for SVR was
100%, when DC-primed CD4 T cell proliferation ratio was less than 0.7.
Conclusions:
The early decline of PDC frequency and sustained DC dysfunction at the end of treatment are served as predictors of relapse in 48 weeks of PEG-IFNα2b and ribavirin therapy.