Antiviral Therapy 14:501–511
Original article
Virological response at 4 weeks to predict outcome
of hepatitis C treatment with pegylated interferon
and ribavirin
Michelle Martinot-Peignoux1, Sarah Maylin1, Rami Moucari1,2, Marie-Pierre Ripault2, Nathalie Boyer 2,
Ana-Carolina Cardoso1,2, Nathalie Giuily 2, Corinne Castelnau 2, Michèle Pouteau2, Christiane Stern1,2,
Anne Aupérin3, Pierre Bedossa1,4, Tarik Asselah1,2 and Patrick Marcellin1,2*
INSERM, U-773, Centre de Recherche Biomédicale Bichat–Beaujon CRB3, Université Paris VII, Hôpital Beaujon, Clichy, France
Service d’Hépatologie, Hôpital Beaujon, Clichy, France
3
Service de Biostatistique et Epidémiologie, Institut Gustave Roussy, Villejuif, France
4
Service d’Anatomie-Cytologie Pathologique, Hôpital Beaujon, Clichy, France
1
2
*Corresponding author: E-mail: [email protected]
Background: Viral kinetics during therapy provides
information on how to individualize treatment. To
determine the benefit of assessing positive predictive
values (PPVs) and negative predictive values (NPVs) of
rapid virological responses (RVRs) and early virological
responses (EVRs), on-treatment outcomes in chronic
hepatitis C patients were examined.
Methods: A total of 408 patients (221 treatment-naive)
treated with pegylated interferon-α2b and ribavirin were
included. Hepatitis C virus (HCV) RNA was measured at
baseline, 4 weeks and 12 weeks. RVR was defined as undetectable HCV RNA at 4 weeks and EVR as ≥2 log10 decrease
in HCV RNA at 12 weeks. The additive value of RVR on predicting sustained virological response (SVR) was assessed
with receiver operating characteristic (ROC) curves.
Results: SVR, RVR and EVR were observed in 46%, 23%
and 78% of patients, respectively. PPVs of RVR were 96%,
100% and 100% in treatment-naive patients, relapsers
and non-responders, respectively. NPVs of failure to
achieve EVR were 97%, 75% and 91%, in treatment-­naive
patients, relapsers and non-responders, respectively. At
4 weeks, patients with RVR had the highest probability
to achieve SVR (odds ratio 44.98 in the entire population and 32.95 in treatment-naive patients). ROC curves
showed the area under the ROC curve to be 0.758 versus
0.832 in the entire population and 0.795 versus 0.858
in treatment-naive patients at baseline versus week 4,
respectively.
Conclusions: RVR is a strong predictor of SVR (PPV>96%)
and failure to achieve EVR is a strong predictor of nonSVR (NPV>75%), independent of patients’ pretreatment
status. Added to baseline characteristics, RVR increased
the accuracy to predict SVR. The combination of RVR and
EVR provided complementary information, and thus provides a key opportunity to individualize treatment and
improve the benefit/risk ratio of therapy.
Introduction
The goal when treating patients with chronic hepatitis
C is to obtain a sustained virological response (SVR),
which in most patients results in the eradication of hepatitis C virus (HCV) infection and improvement of histological outcome [1]. On the basis of three pivotal clinical trials [2–4], the current treatment for HCV infection
is a combination of pegylated interferon and ribavirin
[5]. Tailoring treatment by accurately predicting SVR
before therapy, or early during therapy, could encourage compliance in patients who are likely to respond
or allow discontinuation of therapy in patients who
©2009 International Medical Press 1359-6535 (print) 2040-2058 (online)
Marcellin.indd 501
will not benefit from treatment. It is well established
that baseline viral load is a useful predictor of response
to therapy [6–12]. Recently, it has been suggested that
therapy should only be continued in patients with a
2 log10 decrease in serum HCV RNA at 12 weeks of
pegylated interferon and ribavirin therapy, leading to a
decision to stop therapy at 12 weeks on the basis of this
criteria [9,13,14]. These patients are classified as having
early virological response (EVR). More recently, studies
have suggested that undetectable serum HCV RNA at
week 4 of therapy, called a rapid virological response
501
17/6/09 16:25:18
M Martinot-Peignoux et al.
(RVR), could be a predictor of an SVR [10,15–21]
so that therapy could be shortened. The development
of these new algorithms show that serum HCV RNA
detection has become a key tool in clinical practice,
although it is not standardized. Indeed, previous studies
[22–24] have reported that the assays now used in most
algorithm-based studies [3,4,10,12,14,15,17–20] overestimate baseline serum HCV RNA levels, and lack sensitivity for detecting serum HCV RNA during therapy
[25,26]. Furthermore, most of these algorithms were
developed from randomized controlled trials evaluating treatment-naive patients [3,4,6–15,17–20].
The aim of the study was to determine the value
of assessing early viral kinetics by compiling positive
predictive values (PPVs) and negative predictive values
(NPVs) of an RVR and an EVR for obtaining an SVR,
to support an individualized patient algorithm.
Methods
Patient selection
A total of 565 patients with chronic hepatitis C, seen
in everyday clinical practice and treated with pegylated
interferon-α2b and ribavirin prospectively followed in
our centre from January 2002 to December 2004, were
included in this non-randomized study. All patients had
increased serum alanine aminotransferase levels, detectable anti-HCV antibodies, detectable serum HCV RNA
and a liver biopsy performed within 12 months prior to
beginning treatment. Patients were only included in the
study if they completed a full course of therapy and complied with the follow-up schedule to undergo viral measurements at baseline, week 4 and/or week 12, at the end
of treatment and at week 24 of follow-up, performed in
our centre. In total, 408 patients (221 treatment-­naive
and 187 treatment-experienced) fulfilled these criteria
and were studied. Of the 187 treatment-experienced
patients, 94 received standard pegylated interferon-α2b
monotherapy and 93 received pegylated interferon-α2b
and ribavirin. Overall, 125 patients were previous nonresponders and 62 were previous relapsers to these prior
therapies (Table 1). A total of 157 patients were not
included, either because of virological measurements
performed externally or because of inadequate adherence to therapy (5 patients had to stop therapy because
of intolerance and 21 patients were lost to follow-up).
The study was approved by the Université Paris VII (Clichy, France) ethics committee and conducted according
to the guidelines of the declaration of Helsinki.
Treatment regimens
Patients were treated with pegylated interferon-α2b
(PegIntron®; Schering–Plough Corporation, Kenilworth,
NJ, USA) at a dosage of 1.5 µg per kg body weight
per week, and ribavirin (Rebetol®; Schering–Plough
502
Marcellin.indd 502
Corporation) at a dosage of 800–1,200 mg/day according
to body weight, in genotypes 1 or 4 and 800 mg/day in
genotypes 2 or 3. Treatment-naive patients infected with
genotypes 1, 4 or 5 and all previously treated patients were
treated for 48 weeks; treatment-naive patients infected
with genotypes 2 and 3 were treated for 24 weeks [5].
Virological, biochemical and histological evaluation
Serum HCV RNA was quantified by the VERSANT®
HCV RNA 3.0 (bDNA) Assay (Siemens Medical Solutions, Puteaux, France) with a quantification range of
615–7,690,000 IU/ml [27]. Serum samples <615 IU/ml
were evaluated with the VERSANT® HCV RNA Qualitative Assay (transcription-mediated amplification
[TMA]; Siemens Medical Solutions) with a detection
limit of ≤9.6 IU/ml [28], HCV genotypes were identified using the VERSANT® HCV LiPA 2.0 (Siemens
Medical Solutions) [29]. Serum HCV RNA and alanine
aminotransferase levels were determined at baseline,
weeks 4 and 12, at the end of treatment and 24 weeks
post-treatment. Liver biopsies were assessed using the
Metavir scoring system [30].
Definition of virological responses
RVR was defined at 4 weeks of treatment as undetectable serum HCV RNA. EVR was defined at 12 weeks
as ≥2 log10 decrease in serum HCV RNA from baseline
viral load. SVR was defined as undetectable serum HCV
RNA after 24 weeks of post-treatment follow-up.
Statistical analyses
The PPV was defined as the probability that the
­outcome (that is, SVR) would occur in patients fulfilling the criteria for RVR (at week 4) and EVR (at
week 12). The NPV was defined as the probability
that the outcome (that is, SVR) would not occur in the
absence of RVR or EVR.
Continuous variables were summarized as mean ±sd
and categorical variables as frequency and percentage.
Continuous variables were compared using the Student’s t-test or Mann–Whitney U test. Categorical variables were compared using the χ2 or Fisher’s exact test.
All P-values were based on a two-tailed test of statistical significance. Significance was accepted at P<0.05.
Univariate analyses were performed to identify baseline patient characteristics, which included age, gender,
source of infection, pretreatment status, serum alanine
aminotransferase, genotype, viral load and liver histology associated with SVR. Continuous variables were
transformed into dichotomous variables using their
median values as follows: age ≤45 versus >45 years;
serum alanine aminotransferase ≤2 versus >2× the
upper limit of normal range; viral load ≤400,000 verus
>400,000 IU/ml. Categorical variables with several outcomes were transformed into dichotomous variables
©2009 International Medical Press
17/6/09 16:25:18
Rapid virological response in patients treated with pegylated interferon and ribavirin
according to AASLD guidelines: genotype (1, 4, 5 and
6 versus 2 and 3) and advanced (Metavir F3–F4) versus
non-advanced (Metavir F0–F2) liver fibrosis. To identify independent factors associated with SVR, a first
model of multiple logistic regressions was constructed
including baseline factors that were associated with
SVR in univariate analyses. To determine the additive effect of RVR, a model was constructed with the
addition of RVR at week 4. Results were described as
odds ratios (ORs) with 95% confidence intervals (CIs).
Finally, receiving operator characteristic (ROC) curves
were constructed to compare the accuracy of these
models to predict SVR. Analyses were performed with
SPSS software for windows, version 12.0 (SPSS, Inc.,
Chicago, IL, USA).
Results
The epidemiological, clinical, virological and histological characteristics of the 408 included patients and
the 157 non-included patients are shown in Table 1.
The two groups of patients were similar according to
all baseline characteristics. SVR was observed in 46%
of the overall population. Response rates according to the patients’ pretreatment status are shown in
Table 1.
Table 1. Patient characteristics at baseline
Characteristic
Non-included,
n=157
Included,
n=408
Treatment-naive,
n=221
Relapsers,
n=62
Male, n (%)
100 (64)
267 (65)
135 (64)
41 (66)
Age, yearsa
48 (11)
49 (10)
47 (10)
49 (10)
Age <45 years, n (%)
66 (42)
177 (43)
109 (49)
27 (43)
ALT, IU/mla
118 (72)
125 (88)
113 (76)
112 (91)
Source of infection
Blood transfusion, n (%)
39 (25)
97 (24)
44 (20)
14 (22)
Intravenous drug use, n (%)
42 (27)
101 (25)
61 (27)
23 (37)
Sporadic, n (%)
48 (30)
122 (30)
65 (29)
19 (30)
Unknown, n (%)
28 (18)
88 (21)
51 (14)
6 (11)
Baseline viral load
Log10 IU/mlb
5.613 (3.803–6.598) 5.528 (2.585–6.816) 5.474 (2.883–6.816) 5.554 (2.585–6.697)
≤400,000 IU/ml, n (%) 60 (38)
193 (47)
112 (51)
29 (47)
HCV genotype 1, n (%)c
68 (45)
228 (57)
11 (7)
2 (3)
2, n (%)
18 (11)
38 (9)
27 (12)
3 (5)
3, n (%)
40 (26)
89 (23)
53 (25)
24 (39)
4/5, n (%) 27 (17)
43 (11)
21 (10)
3 (4)
Liver histologyd
Fibrosis stage
1, n (%)
36 (31)
115 (29)
72 (35)
24 (40)
2, n (%)
50 (36)
122 (32)
69 (33)
17 (28)
3, n (%)
29 (21)
80 (21)
40 (19)
10 (17)
4, n (%)
22 (16)
69 (18)
27 (13)
9 (15)
Activity grade
1, n (%)
59 (43)
151 (40)
87 (46)
29 (50)
2, n (%)
68 (49)
193 (51)
100 (53)
26 (45)
3, n (%)
10 (7)
33 (9)
1 (1)
3 (5)
Previous therapy
Interferon, n (%)
37 (23)
94 (23)
–
24 (39)
Interferon + RBV, n (%)
25 (16)
93 (23)
–
38 (61)
Response to therapy SVR, n (%)
62 (40)
188 (46)
117 (53)
40 (65)e
Relapsers, n (%)
11 (7)
59 (15)
25 (11)
10 (16)
Non-responders, n (%)g
84 (53)h
161 (39)
79 (36)
12 (19)
Non-responders,
n=125
91 (73)
51 (10)
41 (33)
155 (91)
39 (31)
17 (14)
38 (31)
31 (24)
5.611 (3.390–6.679)
52 (42)
4 (3)
8 (6)
12 (10)
18 (14)
20 (17)
36 (30)
30 (25)
33 (28)
35 (31)
67 (59)
11 (10)
70 (56)
58 (44)
31 (25)f
22 (17)
72 (58)
Mean (±sd). bMedian (range). cHepatitis C virus (HCV) subtype 1a n=75 (18%) and 1b n=136 (33%). dMetavir scoring system. eEight patients had been previously
treated with pegylated interferon and ribavirin (RBV). f Two patients had been previously treated with pegylated interferon and RBV. gNon-responders to ongoing
therapy. hFive patients who had to discontinue therapy because of intolerance and 21 patients who were lost to follow-up were considered non-responders. ALT,
alanine aminotransferase; SVR, sustained virological response.
a
Antiviral Therapy 14.4
Marcellin.indd 503
503
17/6/09 16:25:19
M Martinot-Peignoux et al.
Baseline
On-treatment
SVR rates according to baseline viral level thresholds
of ≤400,000 versus >400,000 IU/ml, ≤600,000 versus
>600,000 UI/ml and ≤800,000 versus >800,000 UI/ml
and patients’ pretreatment status in patients infected
with genotype 1 are shown in Figure 1. The SVR rates
were significantly different (63% versus 43%; P=0.004)
in treatment-­naive patients with a baseline threshold of
≤400,000 versus >400,000 IU/ml. This difference was
not observed with the other thresholds or in experienced and genotype 1 patients (Figure 1).
Predictive value of rapid virological response: week 4
At 4 weeks of therapy, 320/408 (78%) patients underwent a visit with a blood collection performed in our
centre: 73/320 (23%) showed RVR. The 88 patients
that did not undergo testing at 4 weeks of therapy
were not different from those who underwent testing
(data not shown).
RVR was observed in 43%, 31% and 7% of treatmentnaive patients, relapsers and non-responders to previous
therapy, respectively (P<0.001). RVR was observed in
Figure 1. Sustained virological response rates according to basal viral load levels in treatment-naive patients, relapsers and
non-responders to previous therapy and in genotype 1 patients
Treatment-naive
90
80
70
P=0.004
63
57
60
50
56
47
46
43
Relapser a
100
40
30
20
10
Sustained virological response, %
Sustained virological response, %
100
0
90
80
76
70
70
55
60
50
40
30
20
10
400,000
800,000
600,000
Baseline viral load, IU/ml
70
60
50
40
23
22
24
26
20
10
0
Sustained virological response, %
Sustained virological response, %
80
28
800,000
Genotype 1
100
90
27
600,000
Baseline viral load, IU/ml
Non-responder a
100
90
80
70
60
50
40
30
40
36
29
32
36
31
20
10
0
400,000
600,000
800,000
Baseline viral load, IU/ml
≤ Baseline viral load
a
63
0
400,000
30
65
59
400,000
600,000
800,000
Baseline viral load, IU/ml
> Baseline viral load
Relapser or non-responder to previous therapy.
504
Marcellin.indd 504
©2009 International Medical Press
17/6/09 16:25:19
Rapid virological response in patients treated with pegylated interferon and ribavirin
5%, 47%, 36% and 12% of patients with genotypes 1,
2, 3 and 4/5, respectively (P<0.001). SVR was observed
in 97% and relapse in 3% of patients with RVR.
The PPVs of RVR according to patients’ pretreatment
status are shown in Figure 2A. The PPVs were 96%,
100% and 100% for treatment-naive patients, relapsers
and non-responders to previous therapy, respectively.
The NPVs of RVR (51–75%) were poor predictors of
failure to achieve SVR (Figure 2B).
The PPVs of RVR were 100%, 100%, 94% and
100% for genotypes 1, 2, 3 and 4/5, respectively.
Predictive value of early virological response: week 12
At 12 weeks of therapy, 300/408 (74%) patients underwent a visit with a blood collection performed in our
centre: 234/300 (78%) showed EVR. The 108 patients
that did not undergo testing at 12 weeks of therapy
were not different from those who underwent testing
(data not shown). EVR was observed in 80%, 91%
and 61% of treatment-naive patients, relapsers and
non-responders to previous therapy, respectively. EVR
was observed in 81%, 95%, 97% and 67% of patients
with genotypes 1, 2, 3 and 4/5, respectively. SVR was
Figure 2. Positive predictive value and negative predictive value of virological response at week 4 and week 12
A
B
Week 4
96
100
100
100
90
80
70
60
50
40
30
20
10
0
100
90
Negative predictive value, %
Positive predictive value, %
100
Week 12
Treatmentnaive
Relapser a
75
80
70
60
51
50
40
30
20
10
0
NonGenotype 1
responder a
Treatmentnaive
Relapser a
NonGenotype 1
responder a
Non-RVR c
RVR b
C
D
100
90
80
70
70
70
55
60
50
41
40
30
20
10
0
Treatmentnaive
Relapser a
NonGenotype 1
responder a
EVR d
Negative predictive value, %
100
Positive predictive value, %
68
58
97
93
93
90
75
80
70
60
50
40
30
20
10
0
Treatmentnaive
Relapser a
NonGenotype 1
responder a
Non-EVR e
(A) Positive predictive value and (B) negative predictive value at 4 weeks of therapy. (C) Positive predictive value and (D) negative predictive value at 12 weeks of
therapy according to the type of virological response and patients’ pretreatment status. aRelapser or non-responder to previous therapy. bRapid virological response
(RVR) was defined as undetectable serum hepatitis C virus (HCV) RNA by transcription-mediated amplification (TMA) at 4 weeks of therapy. cNon-RVR was defined as
failure to achieve an RVR, that is, detectable serum HCV RNA with TMA at 4 weeks of therapy. dEarly virological response (EVR) was defined as ≥2 log10 drop in viral load
from baseline at 12 weeks of therapy. eNon-EVR was defined as failure to achieve an EVR, that is, <2 log10 drop in viral load from baseline at 12 weeks of therapy.
Antiviral Therapy 14.4
Marcellin.indd 505
505
17/6/09 16:25:20
M Martinot-Peignoux et al.
observed in 64% of patients with EVR and in 6% of
patients who failed to achieve EVR. A relapse was
observed in 17% of patients with EVR.
The NPVs of EVR according to patients’ pretreatment status are shown in Figure 2D. The NPVs of EVR
were 97%, 75% and 93% for treatment-­naive patients,
relapsers and non-responders to previous therapy,
respectively. PPVs of EVR (41–79%) were poor predictors of SVR (Figure 2C). The NPVs of EVR were 93%,
80%, 100% and 94% for HCV genotypes 1, 2, 3 and
4/5, respectively.
independently associated with SVR were younger
age (≤45 years), genotype 2 or 3, lower baseline viral
load (≤400,000 IU/ml) and non-advanced liver fibrosis (fibrosis stage ≤2), and were similar to the overall
population (Table 6). The ROC curve revealed an area
under the curve of 0.795 (Figure 3B).
At 4 weeks, the addition of RVR to the regression
model showed the highest probability of achieving
SVR for this factor (OR 32.95, 95% CI 6.857–158.3;
Table 7). The ROC curve showed an increased accuracy
with an area under the curve of 0.858 (Figure 3B).
Predictors of sustained virological response:
regression models
Discussion
Characteristics associated with SVR were identified at
baseline in univariate analyses (Tables 2 and 3).
In our study performed in a large patient population
(n=408) seen in everyday clinical practice (not included
in randomized studies), including treatment-naive and
-experienced patients infected with the entire HCV
genotype spectrum (1 to 5), the SVR rate was 53% in
treatment-naive patients (similar to that observed in
controlled trials) and 25% in non-responders to previous therapy. The results of this study demonstrate that
rapid clearance of serum HCV RNA (RVR), achieved in
23% of the patients, is a strong predictor of SVR (PPVs
96–100%). Inversely, failure to achieve an EVR (<2 log10
decrease at week 12), observed in 22% of the patients,
is a strong predictor of non-SVR, (NPVs 93–97%). The
regression models and ROC curves constructed (for the
entire population and for treatment-naive patients) at
week 4 showed that adding RVR increased the accuracy
to nearly 85% compared with 76% at baseline.
Although they are highly accurate, regression models
[18,19] remain unsuitable for clinicians in everyday clinical practice with regards to the assessment of PPV and
Treatment-naive and -experienced patients
The baseline characteristics independently associated
with SVR were younger age (≤45 years), genotype 2 or
3, lower baseline viral load (≤400,000 IU/ml) and nonadvanced liver fibrosis (fibrosis stage ≤2; Table 4). The
ROC curve of this model showed an area under the
curve of 0.758 (Figure 3A).
At 4 weeks, the addition of RVR to the regression
model (Table 5) showed the highest probability of
achieving SVR for this variable (OR 44.98, 95% CI
10.48–193.1). The ROC curve showed an increased
accuracy with an area under the curve of 0.832
(Figure 3A).
Treatment-naive patients
The regression models and ROC analysis were repeated
for treatment-naive patients. The baseline characteristics
Table 2. Univariate analysis showing baseline factors associated with sustained virological response in treatment-naive and
–experienced patients
Factor
Sustained virological response
Yes (n=188)
No (n=220)
Male, %
68
66
Age ≤45 years, %
58
32
ALT>2×ULN, % 61
61
Source of infection Intravenous drug use, %
28
22
Blood transfusion, %
17
19
Other, %
13
11
Unknown, %
41
48
Pretreatment status, %a
84
59
HCV genotype 2 or 3, % 45
18
Baseline viral load ≤400,000 IU/ml, %
57
40
Liver inflammation, %b
48
34
Liver fibrosis, %c
77
50
P-value
0.719
<0.001
0.951
0.335
–
–
–
–
<0.001
<0.001
0.001
0.004
<0.001
Treatment-naive and relapsers to previous therapy. bActivity grade <2 according to the Metavir scoring system. cFibrosis stage ≤2 according to the Metavir scoring
system. ALT, alanine aminotransferase; HCV, hepatitis C virus; ULN, upper limit of normal.
a
506
Marcellin.indd 506
©2009 International Medical Press
17/6/09 16:25:20
Rapid virological response in patients treated with pegylated interferon and ribavirin
Table 3. Univariate analysis showing baseline factors associated with sustained virological response in treatment-naive patients
Factor
Sustained virological response
Yes (n=117)
No (n=103)
P-value
Male, %
65
62
Age ≤45 years , % 63
34
ALT>2×ULN, % 59 50
Source of infection
Intravenous drug use, %
26
28
Blood transfusion, %
14
16
Other, %
14
12
Unknown, %
46
44
HCV genotype 2 or 3, % 49
21
Baseline viral load ≤400,000 IU/ml, %
60
40
Liver inflammation, %a
50
35
Liver fibrosis, %b
82
54
0.673
<0.001
0.201
0.897
–
–
–
–
<0.001
0.002
0.014
<0.001
Activity grade <2 according to the Metavir scoring system. bFibrosis stage ≤2 according to the Metavir scoring system. ALT, alanine aminotransferase; HCV, hepatitis C
virus; ULN, upper limit of normal.
a
Table 4. Regression model showing factors associated with sustained virological response in the 408 patients: baseline
Factor
Regression coefficient b
Standard error
P-value
Odds ratio
95% CI
Age ≤45 years HCV genotype 2 or 3
Viral load ≤400,000 IU/ml
Liver fibrosisa
0.701
1.109
0.873
1.290
0.287
0.304
0.289
0.307
0.015
<0.001
0.003
<0.001
2.015
3.033
2.394
3.631
1.149–3.534
1.671–5.505
1.358–4.220
1.988–6.632
a
Fibrosis stage ≤2 according to the Metavir scoring system. CI, confidence interval; HCV, hepatitis C virus.
Figure 3. Receiver operating characteristic curves for prediction of sustained virological response at baseline and week 4
B
1.0
1.0
0.8
0.8
Baseline AUC=0.758a
0.6
Week 4 AUC=0.832b
0.4
Sensitivity
Sensitivity
A
Baseline AUC=0.795a
Week 4 AUC=0.858b
0.6
0.4
0.2
0.2
0
0
0.0
0.2
0.4
0.6
1-Specificity
0.8
1.0
0.0
0.2
0.4
0.6
0.8
1.0
1-Specificity
(A) Treatment-naive and -experienced patients. (B) Treatment-naive patients. aArea under the curve (AUC) at baseline. bAUC at baseline plus rapid virological response.
Antiviral Therapy 14.4
Marcellin.indd 507
507
17/6/09 16:25:20
M Martinot-Peignoux et al.
NPV, in comparison with convenient tools such as RVR
and EVR. Indeed, our results show that a single serum
HCV RNA measurement at 4 weeks therapy (RVR) can
predict SVR with a PPV>96% without taking into account
baseline characteristics and patients status. Moreover, in
patients who failed to achieve ≥2 log10 drop from baseline serum HCV RNA at 12 weeks of therapy (EVR), the
probability of SVR was very low (NPV>93%).
The accuracy of RVR and EVR are strictly dependent
upon the tools used to measure serum HCV RNA. HCV
RNA assays should be sensitive, specific and accurate,
and the quantitative assays should have a broad dynamic
range. No commercially available assay has all of these
qualities. Recently, two important technical issues have
been reported (overestimation of viral load in undiluted
serum samples and underestimation of viral load for genotypes 2 and 4) in the most commonly used assay, which
have implications for clinical practice [24], emphasizing that the choice of the assays is of crucial importance for the accuracy of the results. In our study, we
used the combination of two assays (bDNA and TMA).
The assay we used to measure the decrease in viral
load (VERSANT® HCV RNA 3.0 [bDNA]) is, so far,
the most reliable available assay for serum HCV RNA
quantification [22,24] and accurately assesses the 2 log10
decrease during therapy. In order to compensate for its
lack of sensitivity in evaluating serum HCV RNA eradication, we used the TMA assay that has been reported
to be the most sensitive available assay [23,25,26,31,32]
for qualitative detection of HCV RNA. As reported
[25,26], TMA allows a better identification of patients
with virological end-of-treatment response and, consequently, a better identification of non-responders who
were previously classified as relapsers. Thus, one can
argue that TMA allows a more sensitive and accurate
determination of RVR.
Guidelines recommend 24 weeks of treatment in treatment-naive patients infected with genotypes 2 or 3 and
48 weeks in those infected with genotypes 1 or 4 [5].
The dose and duration of therapy are fixed and do not
take into consideration differences in the host or virus.
However, on the basis of on-treatment viral kinetics,
several controlled studies suggest that certain patients
could be successfully treated with a shorter duration
than with the standard duration in combination therapy
[10–12,17,20,33–36] (Table 8). In these tailored therapy
Table 5. Regression model showing factors associated with sustained virological response in the 408 patients: the additive
impact of RVR
Factor
Regression coefficient b
Standard error
P-value
Odds ratio
95% CI
Age ≤45 years Liver fibrosisa
RVRb
0.813
1.084
3.806
0.309
0.336
0.743
0.008
0.001
<0.001
2.256
2.956
44.984
1.231–4.133
1.529–5.717
10.481–193.1
Fibrosis stage ≤2 according to the Metavir scoring system. bRapid virological response (RVR) was defined as undetectable serum hepatitis C virus (HCV) RNA at week 4.
CI, confidence interval.
a
Table 6. Regression model showing factors associated with sustained virological response in 221 treatment-naive patients: baseline
Factor
Regression coefficient b
Standard error
P-value
Odds ratio
95% CI
Age ≤45 years HCV genotype 2 or 3
Viral load ≤400,000 IU/ml
Liver fibrosisa
0.832
1.353
1.056
1.757
0.399
0.429
0.411
0.474
0.037
0.002
0.010
<0.001
2.299
3.870
2.874
5.795
1.051–5.027
1.668–8.981
1.285–6.429
2.289–14.67
a
Fibrosis stage ≤2 according to the Metavir scoring system. CI, confidence interval; HCV, hepatitis C virus.
Table 7. Regression model showing factors associated with sustained virological response in 221 treatment-naive patients:
additive impact of RVR
Factor
Regression coefficient b
Standard error
P-value
Odds ratio
95% CI
Age ≤45 years Liver fibrosisa
RVRb
0.979
1.601
3.495
0.424
0.517
0.801
0.021
0.002
<0.001
2.663
4.960
32.95
1.160–6.112
1.801–13.66
6.857–158.3
Fibrosis stage ≤2 according to the Metavir scoring system. bRapid virological response (RVR) was defined as undetectable serum hepatitis C virus (HCV) RNA at week 4.
CI, confidence interval.
a
508
Marcellin.indd 508
©2009 International Medical Press
17/6/09 16:25:21
Rapid virological response in patients treated with pegylated interferon and ribavirin
studies, RVR was the key factor for the success of shorter
treatment [10–12,17,20,33– 35,37,38]. Of note, some
authors [10,15,17,34] report that patients with shorter
duration of therapy show a trend to higher rates of
relapse. By contrast, several large controlled trials comparing standard therapy to prolonged therapy report that
extending the duration of treatment to 48 or 72 weeks,
according to HCV genotype, could increase SVR rates
[20,39] (Table 8). It is important to note that the results
of early kinetics to identify patients who will have an
SVR, or of monitoring treatment duration, appear to
be identical for pegylated interferon-α2a and ribavirin
[10,12,14,16,20,33,34,36–38] and pegylated interferonα2b and ribavirin [11,13,15,17– 19,21,35,40,41]. All
these studies were controlled trials that were mostly performed in treatment-naive patients infected with either
genotype 1 or genotypes 2 and 3. Our study performed
in patients seen in everyday clinical practice, including
treatment-naive patients, relapsers and non-responders to previous therapy, who were infected with HCV
genotypes 1 to 5 and treated with the standard therapy
regimens, is consistent with the notion of individualized
therapy according to treatment-related viral kinetics.
Finally, it is important to keep treatment algorithm
as simple as possible. The main point of our study is
to show that SVR can be predicted, with a PPV≥96%,
in treatment-naive and -experienced patients, whatever
their HCV genotype, with a single serum HCV RNA
testing as early as 4 weeks of therapy. This cost-effective
approach allows clinicians to tailor treatment schedule
to the individual patient, with early prediction applicable in the clinical setting.
RVR is a robust predictor of SVR and provides a key
opportunity to individualize therapy. However, because
failure to achieve a RVR is not reliable as a negative predictor (the probability that SVR will not occur), EVR
assessment is still necessary. Assessment of RVR and EVR
are both needed to obtain the most accurate information
to predict treatment outcome. This information is clinically very useful to identify patients who will benefit from
treatment and from a shortened duration of therapy and
those patients who are most resistant in whom treatment
should be stopped or be extended. Shorter treatment
would reduce drug exposure avoiding unnecessary side
effects and reduce the cost for both patients and society.
Acknowledgements
The authors are grateful to Laurence Leclere for her
valuable technical assistance.
Table 8. Sustained virological response rates in patients with rapid virological response according to treatment regimen
Study
Country
Regimen
Regimen
duration
Participants, na Genotype RVR
Assay
Rate of
sensitivity, IU/ml SVR
Von Wagner
Germany
PEG-IFN-α2a 16 versus 153
2/3
Yes
600
82% et al. [10]
+RBV 24 weeks
versus 80%
Zeuzem
Multinational PEG-IFN-α2b 24 versus 237
1
Yes
29
89% et al. [11]
+RBV 48 weeks
versus 85%
Jensen
Multinational PEG-IFN-α2a 24 versus 729
1
Yes
<50
88% et al. [12]
+RBV 48 weeks
versus 91%
Mangia
Italian
PEG-IFN-α2b 12 versus 283
2/3
Yes <50 85% et al. [17]
+RBV 24 weeks
versus no versus 64%
Sanchez-Tapias Spanish
PEG-IFN-α2a 24 versus 510
1/4
Yes
<50 79% et al. [20]
+RBV 48 weeks
versus 64%
Yu et al. [33]
Taiwan
PEG-IFN-α2a 16 versus 150
2
Yes
<50 100% +RBV 24 weeks
versus no versus 57%; Yes
98% versus no
versus 77%
Shiffman Worldwide
PEG-IFN-α2a 16 versus 1,469
2/3
Yes
<50 78% et al. [34]
+RBV 24 weeks
versus 80%; 85% versus 85%
Dalgard
Sweden and PEG-IFN-α2b 14 versus 428
2/3
Yes
<50
81% et al. [35]
Norway
+RBV
24 weeks
versus 91%;
Mangia
South Italy
PEG-IFN-α2a 24 versus 997
1
Yes
<50
84% et al. [36]
+RBV
48 weeks
versus 77%
Study
design
Multicentre
randomized
controlled
Single-arm
open-label
Post hoc analysis;
randomized
Multicentre
randomized
Multicentre partially
randomized
Regional randomized
open-label
Randomized
Open-label
randomized
Multicentre
randomized
controlled
All participants were treatment-naive. PEG-IFN, pegylated interferon; RBV, ribavirin; RVR, rapid virological response; SVR, sustained virological response.
a
Antiviral Therapy 14.4
Marcellin.indd 509
509
17/6/09 16:25:21
M Martinot-Peignoux et al.
The work was funded by a grant from Schering–
Plough, La Défense, France. The funding source had
no influence in the design and conduct of the study,
the drug supply, collection, management, analysis and
interpretation of the data, or preparation, review and
approval of the manuscript.
This study was presented at The 58th Annual AASLD,
The Liver Meeting, Boston, MA, USA, 2–6 November
2007, and The 43rd Annual European Association for
the Study of the Liver, Milan, Italy, 23–27 April 2008.
Disclosure statement
The authors declare no competing interests.
References
1.
Maylin S, Martinot-Peignoux M, Moucari M, et al.
Eradication of hepatitis C virus in patients successfully
treated for chronic hepatitis C. Gastroenterology 2008;
135:821–829.
2.
Manns MP, McHutchison JG, Gordon SC, et al.
Peginterferon alfa-2b plus ribavirin compared with
interferon alfa-2b plus ribavirin for initial treatment of
chronic hepatitis C: a randomised trial. Lancet 2001;
358:958–965.
3.
Fried MW, Shiffman ML, Reddy KR, et al. Peginterferon
alfa-2a plus ribavirin for chronic hepatitis C virus infection.
N Engl J Med 2002; 347:975–982.
4.
Hadziyannis SJ, Sette H, Morgan TR, et al. Peginterferon
alpha 2a and ribavirin combination therapy in chronic
hepatitis C. Ann Intern Med 2004; 140:346–355.
5.
National Institutes of Health. National Institutes of
Health Consensus Development Conference Statement:
management of hepatitis C: June 10–12, 2002. Hepatology
2002; 36 Suppl 1:S3–S20.
6.
Marcellin P, Pouteau M, Martinot-Peignoux M, et al. Lack of
benefit of escalating dosage of interferon alfa in patients with
chronic hepatitis C. Gastroenterology 1995; 109:156–165.
7.
Martinot-Peignoux M, Marcellin P, Pouteau M, et al.
Pretreatment serum HCV-RNA levels and HCV genotype
are the main and independent prognostic factors of
sustained response to alpha interferon therapy in chronic
hepatitis C. Hepatology 1995; 22:1050–1056.
8.
Martinot-Peignoux M, Boyer N, Pouteau M, et al.
Predictors of sustained response to alpha interferon therapy
in chronic hepatitis C. J Hepatol 1998; 29:214–223.
9.
Berg T, Sarrazin C, Herrmann E, et al. Prediction of
treatment outcome in patients with chronic hepatitis C:
significance of baseline parameters and viral dynamics
during therapy. Hepatology 2003; 37:600–609.
10. Von Wagner M, Huber M, Berg T, et al. Peginterferon-a-2a
(40KD) and ribavirin for 16 or 24 weeks in patients with
genotype 2 or 3 chronic hepatitis C. Gastroenterology
2005; 129:522–527.
11. Zeuzem S, Buti M, Ferenci P, et al. Efficacy of 24 weeks
treatment with peginterferon alfa-2b plus ribavirin in
patients with chronic hepatitis C infected with genotype 1
and low pretreatment viremia. J Hepatol 2006; 44:97–103.
12. Jensen DM, Morgan TR, Marcellin P, et al. Early
identification of HCV genotype 1 patients responding to 24
weeks peginteferon alpha-2a (40kd)/ribavirin. Hepatology
2006; 43:954–960.
13. Davis GL, Wong JB, McHutchison JG, Manns MP,
Harvey J, Albrecht J. Early virological response to treatment
with peginterferon alfa-2b plus ribavirin in patients with
chronic hepatitis C. Hepatology 2003; 38:645–652.
510
Marcellin.indd 510
14. Ferenci P, Fried MW, Shiffman ML, et al. Predicting SVR in
chronic hepatitis C patients treated with peginterferon alpha
2a (40KD)/ribavirin. J Hepatol 2005; 43:425–433.
15. Dalgard O, Bjoro K, Block-Hellum K, et al. Treatment with
pegylated interferon and ribavirin in HCV infection with
genotype 2 or 3 for 14 weeks: a pilot study. Hepatology
2004; 40:1260–1265.
16. Carlsson T, Reichard O, Norkrans G, et al. Hepatitis C
virus RNA kinetics during the initial 12 weeks treatment
with pegylated interferon-alpha 2a and ribavirin according
to virological response. J Viral Hepat 2005; 12:473–480.
17. Mangia A, Santoro R, Minerva N, et al. Pegintefereron
alfa-2b and ribavirin for 12 vs 24 weeks in HCV genotype 2
or 3. N Engl J Med 2005; 352:2609–2617.
18. Martinot-Peignoux M, Comanor L, Minor JM, et al.
Accurate models predicting sustained response with
pegylated interferon with ribavirin in patients with chronic
hepatitis C. J Viral Hepat 2006; 13:701–707.
19. Martinez-Bauer E, Crespo J, Romero-Gomez M, et al.
Development and validation of two models for early
predeiction of response to therapy in genotype 1 chronic
hepatitis C. Hepatology 2006; 43:72–80.
20. Sanchez-Tapias JM, Diago M, Escartin P, et al.
Peginterferon alpha 2a plus ribavirin for 48 versus 72 weeks
in patients with detectable hepatitis C virus RNA at week 4
of treatment. Gastroenterology 2006; 131:451–460.
21. Moucari R, Ripault MP, Oulès V, et al. High predictive
value of early viral kinetics in retreatment with
peginterferon and ribavirin of chronic hepatitis C patients
non-responders to standard combination therapy. J Hepatol
2007; 46:596–604.
22. Pisani G, Cristiano K, Wirz M, Bisso GM, Gentili G.
Overestimation of the hepatitis C virus RNA content of
reference preparations by the AMPLICOR HCV Monitor
test, version 2.0. J Clin Microbiol 2002; 40:4765–4767.
23. Morishima C, Chung M, Ng KW, Brambilla DJ, Gretch DR.
Strengths and limitations of commercial tests for Hepatitis
C virus RNA quantification. J Clin Microbiol 2004;
42:421–425.
24. Chevaliez S, Bouvier-Alias M, Brillet R, Pawlotsky JM.
Overestimation and underestimation of hepatitis C virus
RNA levels in a widely used real-time polymerase chain
reaction-based method. Hepatology 2007; 46:22–31.
25. Sarrazin C, Teuber G, Kokka R, Rabenau H, Zeuzem S.
Detection of residual hepatitis C virus RNA by
transcription-mediated amplification in patients with
complete virological response according to polymerase
chain reaction-based assays. Hepatology 2000; 32:818–823.
26. Gerotto M, Dal Pero F, Bortoletto G, et al. Hepatitis C
minimal residual viremia [MRV] detected by TMA at the
end of pegylated-interferon plus ribavirin therapy predicts
post-treatment relapse. J Hepatol 2006; 44:83–87.
27. Elbeik T, Surtihadi J, Destree M, et al. Multicenter
evaluation of the performance characteristics of the Bayer
VERSANT HCV-RNA 3.0 assay [bDNA]. J Clin Microbiol
2004; 42:563–569.
28. Comanor L, Anderson F, Ghany M, et al. Transcriptionmediated amplification is more sensitive than conventional
PCR-based assays for detecting residual serum HCV-RNA at
end of treatment. Am J Gastroenterol 2001; 96:2968–2972.
29. Bouchardeau F, Cantaloube JF, Chevaliez S, et al.
Improvement of hepatitis C virus [HCV] genotype
determination with the new version of the INNO-LiPA
HCV assay. J Clin Microbiol 2007; 45:1140–1145.
30. Bedossa P, Poynard T. An algorithm for the grading of
activity in chronic hepatitis C. The METAVIR Cooperative
Study Group. Hepatology 1996; 24:289–293.
31. Morishima C, Morgan TR, Everhart JE, et al. HCV RNA
detection by TMA during the hepatitis C antiviral long-term
treatment against cirrhosis (Halt-C) trial. Hepatology 2006;
44:360–367.
32. Berg T, Weich V, Teuber G, et al. Importance of minimal
residual viremia for relapse prediction in HCV type 1
patients receiving standard or individualized treatment.
Hepatology 2007; 46 Suppl 1:317A.
©2009 International Medical Press
17/6/09 16:25:21
Rapid virological response in patients treated with pegylated interferon and ribavirin
33. Yu ML, Dai CY, Huang JF, et al. A randomized study
of peginterferon and ribavirin for 16 versus 24 weeks in
patients with genotype 2 chronic hepatitis C. Gut 2007;
56:553–559.
34. Shiffman ML, Suter F, Bacon BR, et al. Peginterferon
alfa-2a and ribavirin for 16 or 24 weeks in HCV genotypes
2 or 3. N Engl J Med 2007; 357:124–134.
35. Dalgard O, Bjoro K, Ring-Larsen H, et al. Pegylated
interferon alfa and ribavirin for 14 versus 24 weeks in
patients with hepatitis C virus genotype 2 or 3 and rapid
virological response. Hepatology 2008; 47:35–42.
36. Mangia A, Minerva N, Bacca D, et al. Individualized
treatment duration for hepatitis C genotype 1 patients: a
randomized controlled trial. Hepatology 2008; 47:43–50.
37. Berg T, Von Wagner M, Nasser S, et al. Extended treatment
duration for hepatitis C virus type 1: comparing 48
versus 72 weeks of peginterferon alfa-2a plus ribavirin.
Gastroenterology 2006; 130:1086–1097.
38. Ferenci P, Laferl H, Scherzer TM, et al. Peginterferon alfa
2a and ribavirin for 24 weeks in hepatitis C type 1 and 4
patients with rapid virological response. Gastroenterology
2008; 135:451–458.
39. Marcellin P, Heathcote EJ, Craxi A. Which patients with
genotype 1 chronic hepatitis C can benefit from prolonged
treatment with ‘accordion’ regimen? J Hepatol 2007;
47:580–587.
40. Saracco G, Olivero A, Ciancio A, et al. A randomized 4-arms
multicenter study of interferon alfa-2b plus ribavirin in the
treatment of patients with chronic hepatitis C relapsing after
interferon monotherapy. Hepatology 2002; 36:959–966.
41. Pearlman BL, Ehleben C, Saife S. Treatment extension to 72
weeks of peginterferon and ribavirin in hepatitis C genotype
1-infected slow responders. Hepatology 2007; 46:1688–1694.
Accepted for publication 26 January 2009
Antiviral Therapy 14.4
Marcellin.indd 511
511
17/6/09 16:25:21