|
HEPATOLOGY, August 1998, p. 341-346, Vol. 28, No. 2
Sequential Versus Concomitant Administration
of Ribavirin and Interferon Alfa-n3 in Patients With Chronic
Hepatitis C Not Responding to Interferon Alone: Results of a
Randomized, Controlled Trial
Raffaello Sostegni1, Valeria Ghisetti2,
Fabrizia Pittaluga2, Giovanna Marchiaro2,
Giuseppe Rocca1, Elisabetta Borghesio1, Mario
Rizzetto1, and Giorgio Saracco1
From the 1 Dipartimento di Gastroenterologia, Ospedale
Molinette, Università di Torino, and the 2 Laboratorio di
Microbiologia, Ospedale Molinette, Torino, Italy.
ABSTRACT
We conducted a three-arm, randomized trial in 96 patients with
chronic hepatitis C who did not respond to interferon alfa to
compare treatments. Group 1 (33 patients) received ribavirin alone
(1,000 mg/daily for 6 months) followed by interferon alfa n-3 alone
(3 MU thrice weekly for 6 months); group 2 (33 patients) received
ribavirin plus interferon alfa n-3 for 6 months at the above doses;
and group 3 (30 patients) received interferon alfa n-3 alone (3 MU
thrice weekly for 6 months). At the end of treatment, 3 patients
(10%) in group 1, 13 (41%) in group 2, and 5 (17%) in group 3 had
normal alanine transaminase (ALT) levels (group 2 vs. groups 1 and
3, P = .008). After 6 months of follow-up, only 4 patients (12.5%)
in group 2 still had normal ALT values (P = .03). At the end
of therapy, hepatitis C virus (HCV) RNA was no longer detectable by
polymerase chain reaction in 4 (13%), 9 (27%), and 2 (7%) patients,
respectively, in groups 1, 2, and 3 (P = NS). Six months
posttherapy, only 5 (15%) patients in group 2 were still HCV RNA
negative (P = .02). At the time of follow-up liver biopsy,
performed 6 months after the end of treatment, a significant
improvement of the necroinflammatory scores was observed among
group 2 patients (P = .01) but not in the other two groups.
Side effects reflected the profile of each drug as monotherapy;
mild hemolytic anemia was the most frequent side effect caused by
ribavirin. In conclusion, concomitant administration of ribavirin
and interferon alfa n-3 was significantly superior to the
sequential schedule or interferon alfa n-3 monotherapy in inducing
a sustained response in patients with chronic hepatitis C who had
not responded to interferon alone. However, combination therapy at
the dose and duration adopted in this study is capable of modifying
the natural course of the disease in only a minority of these
patients. (HEPATOLOGY 1998;28:341-346.)
INTRODUCTION
Interferon (IFN) monotherapy induces a permanent remission in a
minority of treated patients with chronic hepatitis C.1
Patients in whom IFN monotherapy is not effective do not respond to
either initial or repeat IFN therapy. Their alanine transaminase
(ALT) levels do not normalize, even during IFN
administration.2,3 Thus, retreatment of nonresponders
with IFN alone generally is considered ineffective.4
Pilot studies5,6 using ribavirin (a broad-spectrum
antiviral drug given by mouth) with IFN have reported sustained
responses in approximately 40% of patients who had a relapse or no
response to IFN alone. Further studies have shown that the response
to combination therapy is significantly different between patients
who had no response to IFN and those who had a response while
undergoing therapy but then relapsed; with combination therapy a
sustained response is observed in most relapsers but in only few
nonresponders.7,8 A recent meta-analysis9 has
shown that the estimated probability of sustained response after
IFN-ribavirin combination therapy is approximately 16% for previous
IFN nonresponders; the standard schedule currently proposed is
concomitant administration of 1,000/1,200 mg of ribavirin orally
each day and 3,000,000 IU (3 MU) of IFN thrice weekly for 6 months.
The aim of this trial was to establish the efficacy and
tolerability of IFN-ribavirin combination therapy administered
according to two different temporal schedules (ribavirin
monotherapy followed by IFN vs. concomitant administration of
ribavirin and IFN) compared with IFN therapy alone in patients with
chronic hepatitis C that did not respond to previous IFN
therapy.
Pretherapy features were also evaluated by univariate analysis
to determine whether associations exist between baseline variables
and a sustained response.
PATIENTS AND METHODS
Ninety-six patients with chronic hepatitis C that did not
respond to previous courses of recombinant IFN- or lymphoblastoid IFN administered at a
minimum dose of 9 MU/wk for a minimum of 12 weeks were enrolled in
an Italian randomized, controlled trial. All patients were
considered nonresponders because they did not have normal ALT
levels and clearance of viremia at the end of a single course of
IFN therapy. Patients were eligible if they had all the inclusion
criteria: age >18 years and <65 years; positive results for
antibody to hepatitis C virus (HCV); positive results for HCV RNA
by polymerase chain reaction (PCR); chronic hepatitis C at liver
biopsy; previous nonresponse to IFN- alone (9 MU/wk for 12 weeks at least); and evidence of
elevated ALT levels (at least 1.5 times upper limit of normal;
range, 0-40 IU). The exclusion criteria were relapses after a
previous course of IFN-
alone; positivity for hepatitis B serum antigen; positivity for
antibody to human immunodeficiency virus; alcoholic liver disease;
hemochromatosis; Wilson's disease; drug-related liver disease;
autoimmune hepatitis; inadequate hemoglobin level (<10 g/dL),
platelet count (<70,000/mm3), white blood cell count
(<3,000/mm3), or granulocyte count
(<1,500/mm3); decompensated cirrhosis; intravenous
drug abuse; abnormal serum uric acid level; presence of concomitant
significant medical illness; history of hemolytic anemia;
1-antitrypsin deficiency; obesity-induced liver disease; and
hemophilia. Patients were randomly assigned by computer coding to
one of the following study groups: group 1 (33 patients), 1,000
mg/daily ribavirin for 6 months followed by 3 MU natural human
leukocyte IFN- (IFN-n3,
Alfaferone; Alfa-Wasserman, Bologna, Italy) thrice weekly for an
additional 6 months; group 2 (33 patients), 1,000 mg ribavirin
daily plus 3 MU IFN-n3 thrice weekly for 6 months; and group 3 (30
patients), 3 MU IFN-n3 thrice weekly for 6 months. In groups 1 and
2, ribavirin was supplied as 200-mg capsules given orally in two
divided doses of 1,000 mg daily. Patients were seen every month
during therapy and follow-up. At each visit, blood was drawn for
routine hematologic and biochemical tests, and aliquots of serum
were stored at .gif) 80°C immediately after collection.
All details of the study were approved by the ethical committee,
and patients gave written informed consent for inclusion. Efficacy
was assessed by monitoring of serum ALT levels and plasma HCV RNA
levels. Safety was assessed by collection of adverse event data and
by periodic hematology and biochemistry tests. The biochemical
response to IFN-n3 was defined as complete when ALT levels were
normal at the end of therapy. The response was considered sustained
if ALT values were normal during the 6-month follow-up. A relapse
was defined as a return to abnormal ALT levels after a complete
response within the 6-month posttherapy period. All the patients
underwent liver biopsies within the 6 months before inclusion into
the study; a second liver biopsy was proposed to each patient at
the end of follow-up. Each liver biopsy was reviewed blindly by two
pathologists according to the Ishak's scoring
system.10
Virology
Serum for HCV RNA levels was measured at the start of treatment,
at the end of therapy, and at the end of follow-up; in group 1, an
additional HCV RNA quantitation was performed at the end of
ribavirin administration. HCV RNA was quantified by branched DNA
signal amplification assay (Quantiplex, version 2.0) according to
the manufacturer's instructions (Chiron Corporation, Emeryville,
CA). This assay has a sensitivity of 2 × 105 HCV
genome equivalents/mL (genEq/mL); serum samples below this cut-off
were evaluated for HCV RNA by a sensitive nested PCR. HCV RNA was
extracted from 50 µL of serum, reverse transcribed, and
amplified as previously described.11 The sensitivity of
this assay was 50 genEq/mL of serum. HCV genotyping was carried out
using a probe-specific hybridization assay (Line probe assay, LIPA
HCV; Innogenetics, Ghent, Belgium).
Statistical Analysis
Dichotomous variables were compared using the  2 test, and quantitative variables were
compared with Student's t test. To determine the strength of
association between pretreatment selected variables and sustained
response, a logistic regression model was used.
RESULTS
Two patients in group 1 did not receive therapy; therefore, the
intent population comprised 94 patients. No significant differences
were found between the groups regarding age, sex, mean baseline ALT
values, liver histology, source of infection, viremia, or genotypes
table 1).
Five patients in group 1 (16%), 3 patients in group 2 (9%), and
2 patients in group 3 (7%) withdrew from the study because of
adverse events. At the end of treatment, 3 patients (10%) in group
1, 13 (41%) in group 2, and 5 (17%) in group 3 had normal ALT
levels (group 2 vs. groups 1 and 3, P = .008). After 6
months of follow-up, only 4 patients (12.5%) in group 2 had normal
ALT concentrations (P = .03). Of the patients treated with
the sequential therapeutic administration, 7 (22.5%) had normalized
ALT levels at the end of the ribavirin course, but 4 of these
relapsed during IFN-n3 administration. The ALT patterns observed in
the three groups are shown in Fig. 1.
.gif) |
Fig. 1. Mean ALT levels in the 3 groups. ( ) Group 3; ( ) group 2; and ( ) group 1. |
At the end of treatment, mean ALT levels were significantly
reduced among patients in group 2 compared with those in the other
two groups (P = .01); this difference was still present at
the end of follow-up (P = .007).
Virological Analysis
The mean quantitative levels of HCV RNA detected in the three
groups during therapy and follow-up are reported in Fig. 2; in each
group, an insignificant reduction in viremia was observed at the
end of treatment compared with baseline levels (group 1, 4.1 vs.
4.9 × 106; group 2, 1.3 vs. 3.1 ×
106; group 3, 2.4 vs. 5.1 × 106
genEq/mL), but the mean HCV RNA levels returned to pretherapy
values at the end of follow-up in all groups. No significant
difference in mean viremic levels among the three groups was
observed at the end of therapy and follow-up. In group 1, the
decrease in viremia was more pronounced at the end of ribavirin
administration than that measured at the end of the IFN-n3 course;
this effect was not limited to patients with favorable predictive
factors (low viremia, non-1 genotype, mild histological picture)
but was seen in almost all the patients.
At the end of treatment, HCV-RNA was no more detectable by PCR
in 4 (13%), 9 (27%), and 2 (7%) patients, respectively, in groups
1, 2, and 3 (P = NS). Six months later, only 5 (15%)
patients in group 2 were still HCV RNA-negative (P = .02).
Four of these 5 patients also had normal ALT levels.
Histology
The mean necroinflammatory score and the mean staging score were
similar in the three groups at the start of therapy (table 1). Of the 96 patients
included in this study, 28 agreed to undergo a second liver biopsy
6 months after the end of therapy: 8 in group 1, 12 in group 2, and
8 in group 3 (table 2).
All 4 patients in group 2 who had sustained responses underwent
second biopsies.
A significant reduction in necroinflammatory scores was found in
patients in group 2 (P = .03) compared with
those in the other two groups; no statistically significant
difference in staging was seen among the three groups.
Factors Predictive of Sustained Response
The 4 patients with sustained responses and the 29 with no
response or relapse in group 2 were stratified according to sex,
age, mean baseline ALT levels, source of infection, viremia,
genotype, and histology. Because of the limited number of patients
considered, only a univariate logistic regression analysis was
performed to correlate these baseline characteristics with the
long-term outcome of IFN therapy. Results are reported in table 3 Low levels of HCV RNA and
history of blood transfusions were the only significant factors
influencing sustained response.
| View This table |
table 3. Results of Univariate Logistic Regression Analysis
Correlating Initial Features With Long-Term Outcome of Patients in
Group 2 |
Side Effects
Overall, 11 (11%) patients discontinued treatment early because
of side effects: 5 in group 1, 4 in group 2, and 2 in group 3.
Three of the 5 patients who dropped out in group 1 left the study
during ribavirin administration: 1 for severe pruritus, 1 for
abdominal pain, and 1 for severe hemolytic anemia (7 g/dL); the
other 2 patients discontinued IFN-n3 treatment because of severe
mental depression. In group 2, 2 patients discontinued therapy for
headache, 1 for severe mental depression, and 1 for an ischemic
cardiac episode probably triggered by a mild decrease in hemoglobin
(10 g/dL). Finally, 2 patients in group 1 discontinued IFN-n3
therapy for headache and myalgias. These symptoms promptly reversed
after therapy was discontinued.
There were no statistically significant differences among groups
regarding frequency of side effects, with the exception of anemia;
the mean ± SD decreases in hemoglobin in groups 1 and 2 were
2.3 ± 1.11 and 2.5 ± 1.14 g/dL, respectively. Because
of hemolytic anemia, a dose reduction (usually 20% of the total
dose of ribavirin) was necessary in 6 (24%) patients in group 1 and
8 (27.5%) patients in group 2. No change in IFN-n3 dosage was
observed. The pattern of mean hemoglobin levels and reticulocyte
counts of the three groups is shown in Fig. 3.
The decrease in hemoglobin was accompanied by a parallel decrease
in red blood cell count, but both of the biochemical abnormalities
reversed after suspension of ribavirin therapy.
 |
Fig. 3. Hemoglobin levels and reticulocyte
counts in the 3 groups
(means ± SD).
) Hemoglobin (g/dL);
and () reticulocytes
(%). |
DISCUSSION
Retreatment with IFN for patients who did not benefit from a
first therapeutic cycle with the cytokine has been attempted in
several studies,with disappointing results. There has been no
significant response to either initial or repeat IFN
administration; in particular, increasing the dose or the duration
did not confer a benefit. To date, the most promising alternative
therapy for these patients is the nucleoside analog ribavirin;
although this drug induces only a transient beneficial effect when
given as monotherapy,12-14 it seems to increase in
effectiveness when associated with IFN.4 There is
evidence that the response to combination therapy is significantly
different between nonresponders, i.e., those who did not have a
response during therapy, and relapsers, i.e., those who responded
during therapy but then rebounded after treatment withdrawal.
According to a recent meta-analysis,9 approximately 50%
of those who relapse after a first IFN course have a sustained
response with combination therapy, compared with only 19% of
nonresponders; in recent studies,15,16 combination
therapy was of no advantage compared with IFN alone for retreatment
of nonresponders. Our data confirm that only a minority (12.5%) of
IFN-resistant patients have a sustained response when treated with
the combination of ribavirin and IFN, indicating that sequential
administration of the two drugs is ineffective and has no impact in
reducing the frequency of side effects. Although modest, these
results represent some therapeutic gain for patients who are likely
to have a progressive clinical course. Patients treated with
simultaneous combination therapy had significantly lower ALT levels
during the follow-up than those in the other two groups. Mean HCV
RNA values were not different among the three groups at the end of
follow-up, but all 4 sustained responders in group 2 were no longer
viremic, in indication that they were likely to have attained
permanent viral eradication. The reduction of viremia in group 1 at
the end of the 6-month course with ribavirin alone was not
significant; a similar finding was reported by other authors. Di
Bisceglie et al.13 reported a decrease in HCV RNA titers
during therapy and for several months afterward; a viremic
decrement at the end of therapy was found also by Dusheiko et
al.14 whereas in a Swedish study,12 mean HCV
RNA levels decreased from 4.1 × 106 to 3.4 ×
106 genEq/mL at cessation of therapy. The surprising
progressive increase in viremia observed in group 1 during IFN
treatment after ribavirin withdrawal suggests that ribavirin may
play a detrimental role on the action of IFN, possibly by enhancing
the selection of viral strains resistant to IFN. The mechanism
whereby ribavirin is able to potentiate the effect of IFN when
administered given in combination is still unknown; the hypothesis
that the drug acts by eliminating the intracellular reservoirs of
HCV not accessible to IFN17 was not
confirmed.18 The hypothesis of an immunomodulatory
effect of ribavirin19 needs further evidence.
Ribavirin combined with IFN was well tolerated at a dose of
1,000 mg/d for 6 months. Only 4 patients (3 in group 1 and 1 in
group 2) discontinued therapy early because of side effects
probably related to the ribavirin administration. The drug was also
safe in the consistent proportion of cirrhotic patients who are
less tolerant to a wide variety of therapeutic agents.
In conclusion, with the schedule used in this study, the natural
history of HCV-positive nonresponders to IFN was not significantly
modified. Further trials with higher doses and/or longer duration
of the combination of ribavirin and IFN are needed to improve
efficacy.
Footnotes
Abbreviations: IFN, interferon; ALT, alanine transaminase; HCV,
hepatitis C virus; PCR, polymerase chain reaction.
Received January 7, 1998; accepted April 6, 1998.
Address reprint requests to: Giorgio Saracco, M.D., Department
of Gastroenterology, Molinette Hospital, Corso Bramante 88 10126
Torino, Italy. Fax : 39-11-6634213.
REFERENCES
Copyright © 1998 by the American Association for the Study
of Liver Diseases.
|
