|
NIH Consensus Development Conference on Management of Hepatitis C
Interferon Alfa-n 1 Trials
Geoffrey C. Farrell, M.D., F.R.A.C.P.
Background
Lymphoblastoid interferon (IFN-alpha-n1) is produced from a human lymphoid cell
line and consists of multiple IFN-alpha subtypes of which at least two are glycosylated.
(1) This differs from the recombinant IFNs, which are single unglycosylated
proteins produced from individual IFN-alpha genes (usually subtype 2) expressed
in E. coli. The differences between IFN-alpha preparations provide a potential
basis for varying clinical effects. A meta- analysis of earlier published trials
indicated that IFN-alpha-n1 was at least as efficacious against hepatitis C
as recombinant products (IFN-alpha2b, IFN-alpha2a). (2) One apparent difference,
however, was a lower rate of posttreatment relapse after IFN-alpha-n1 . Thus,
sustained response (SR) was 25 percent with IFN-alpha-n1 compared with 16 percent
for recombinant products. In order to provide sufficient power to detect a 10
percent difference in response rate between lymphoblastoid and recombinant IFN-alpha,
a randomized, clinical trial (the 096 study) was conducted at 63 centers in
13 countries.
Methods
Patients had previously untreated, chronic hepatitis C virus (HCV) infection,
defined by positive anti- HCV, consistently raised alanine aminotransferases
(ALT), appropriate liver histology, and exclusion of other disease. Patients
(n=1071) were randomized to receive lymphoblastoid interferon (IFN-alpha-n1)
or recombinant interferon-alfa2b (IFN-alpha2b), both 3 million units (MU) tiw
subcutaneously for 24 weeks. HCV genotype (by line probe assay) and serum HCV
RNA titer (by quantitative polymerase chain reaction [PCR]) were determined
at baseline.
Primary endpoints were: end-treatment response (ETR}two or more successively
normal serum ALT levels including the week 24 value; sustained response (SR)~continued
normal ALT after an ETR, with normal values at weeks 48 (week 48 SR) and 72
(week 72 SR). Secondary endpoints were serum HCV RNA at weeks 24, 48, and 72
, and quantitative histology (by Knodell's histologic activity index [HAI] score)
at weeks 24 and 72. Primary endpoints were analyzed by intent to treat, and
secondary endpoints were according to treatment received.
Results
Groups were well matched for demographic, viral, clinical, and histologic
variables; about 10 percent had cirrhosis. Predominant genotypes were la and
lb in the United States, and types lb, 2a, and 3a in Europe and Australia. At
end of treatment, ALT response was 35.3 percent for IFN-alpha-n1 and 37.9 percent
for IFN-alpha2b (NS). The type and frequency of reported adverse experiences
were similar; 26 patients (5 percent) receiving IFN-alpha-n1 and 19 (4 percent)
receiving IFN-alpha2b withdrew because of adverse experiences. Viral and histologic
responses were determined in the 970 evaluable patients. At 24 weeks, there
was no difference between the groups in the proportion who were HCV RNA negative
(IFN-alpha-n1, 37.9 percent vs. IFN-alpha2b, 42.0 percent, NS). More than 70
percent of ALT responders had cleared serum HCV RNA. Among patients who had
an ETR, liver histology at 24 weeks was improved (2-point or greater reduction
in HAI activity score) in 61 percent; there were no differences between treatment
groups. Among responders (ETR), posttreatment relapse was less frequent with
IFN-alpha-n1 than with IFN-alpha2b. Thus sustained ALT responses (SR) to IFN-alpha-n1
were more frequent than to IFN-alpha2b (12.0 percent vs 7.6 percent at 48 weeks,
P=O.OI; 10.3 percent vs 6.7 percent at 72 weeks, P=0.02).
Relapse was more likely if HCV RNA was present at end treatment. SR was associated
with viral loss, and more patients treated with IFN-alpha-n1 were HCV RNA negative
at week 72 compared with IFN-alpha2b (9.9 percent vs. 5.7 percent, p Sixteen
candidate variables were examined by multivariate regression analysis in relation
to treatment response. Heavier weight, cirrhosis at screen, and genotypes 1
were associated with lack of ETR, whereas larger body surface area and white
race were positively associated with ETR. Viral titer was not associated with
ETR, but was independently associated with SR.
SR varied from 3 percent with genotype 1, to 20 percent for genotypes 2, 3,
and mixed infections. SR (72 weeks) after treatment with IFN-alpha-n1 was superior
across all genotypes compared with IFN-alpha2b.
Discussion
The finding that IFN-alpha-n1 conferred an efficacy advantage compared
with a recombinant alphainterferon when given at 3 MU, tiw for 24 weeks was
not due to a greater number of patients having an ETR. Rather, it was attributable
to the lower proportion of patients who subsequently relapsed. This reduction
in relapse rate produced a higher SR for IFN-alpha-n1, and more patients treated
with IFN-alpha-n1 had sustained clearance of serum HCV RNA. The findings of
this prospective study are consistent with the earlier meta-analysis of interferon
trials for hepatitis C; i.e., there is no difference between lymphoblastoid
and recombinant alpha-interferon products in ETR, but 50 percent improvement
in SR with the lymphoblastoid interferon due to a lower relapse rate. (2) It
appears likely that the clinical difference in efficacy results from IFN-alpha-n1
exerting more potent anti-HCV activity across all HCV genotypes. The greater
diversity of alpha-interferon subtypes present in the Iymphoblastoid product
might be responsible for this.
The present results provide more extensive data concerning the posttreatment
course following 24 weeks of treatment with IFN-alpha than have been previously
available. The proportion of patients with normalization of ALT who were negative
with regard to HCV RNA in serum increased from 70 percent at end treatment,
to 80 percent at 48 weeks and 90 percent at 72 weeks. This indicates a relationship
between sustained ALT normalization and clearance of HCV RNA. The minority of
ALT responders who remain HCV RNA positive are likely to undergo clinical relapse
in the year following treatment. The present data also confirm that sustained
ALT and viral response are associated with histologic improvement of liver inflammation
and necrosis, but not of fibrosis.
Several studies with recombinant IFN-alpha have shown that treatment courses
of 12 months (or longer) reduce the relapse rate by around 50 percent, thereby
effectively doubling SR. (3) Data regarding the comparative efficacy of higher
interferon doses have been conflicting, and improvement in efficacy may be compromised
by a higher rate of intolerable adverse experiences. (3-4) The issue therefore
arises as to whether lymphoblastoid interferons also exhibit greater potency
when administered in higher doses or for longer. This was addressed in a large
European multicenter trial (091 ) of IFN-alphan l. (5) Entry and response criteria
were similar to the 096 study; SR was at 48 week followup. Patients (n=440)
were randomly assigned to receive one of the following four regimens (all tiw):
3 MU for 24 weeks, 3 MU for 48 weeks, 5 MU for 24 weeks, and 5 MU for 48 weeks.
ETR ranged from 38 percent to 50 percent (NS). There was no difference in SR
between the 24 weeks 3 MU and 5 MU arms [6 percent vs. 14 percent, NS] or between
the corresponding two 48- week arms (19 percent vs. 20 percent, NS). However,
at both dose increments, prolongation of therapy to 12 months effectively doubled
the SR (P=O.OO 1 ) by halving the relapse rate. Histological improvement, as
seen by reduction in the necroinflammatory activity of the HAI score, was most
effectively maintained in the group receiving 5 MU for 48 weeks
Conclusions
IFN-alpha-n1 and IFN-alpha2b have similar ETR rates and safety profiles,
but the SR rate is higher with IFN-alpha-n1. Sustained response was associated
with loss of HCV viremia and with histologic improvement, the latter being maximal
in those who exhibited clearance of HCV RNA. HCV genotype was the most powerful
determinant of SR, and the greater efficacy of Iymphoblastoid interferon appeared
to extend to all major genotypes. There is definite evidence that 12 months
of treatment is superior to 6 months in terms of ALT and HCV RNA as efficacy
endpoints, and whether there is truly benefit for histologic response at the
higher dose of IFN-alpha-n1 (5 MU vs. 3 MU tiw) needs to be considered in relation
to adverse experiences. Finally, the possibility that Iymphoblastoid interferons
could produce a better SR than recombinant alpha-interferons when given for
a 12-month (or longer) treatment course is an important issue that requires
further study.
References
1.Zoon K, et al. Purification and characterization of multiple components
of human lymphoblastoid interferon-alpha. J Biol Chem 1992;1267:15210-6.
2.Bardelli F, Messori A, Rampazzo R, Alberti A, Martini N. Effect of recombinant
or lymphoblastoid interferon-alpha on alanine aminotransferase in patients with
chronic hepatitis C or chronic non-A non-B hepatitis. A meta-analysis. Clin
Drug Invest 1995;9:239-54.
3.Poynard T, Leroy V, Cohard M, Thevenot T, Mathurin P, Opolon P, Zarski JP.
Meta-analysis of interferon randomized trials in the treatment of viral hepatitis
C: effects of dose and duration. Hepatology 1 996;24:778-9.
4.Lindsay KL, Davis GL, Schiff ER, et al. Response to higher doses of interferon
alfa-2b in patients with chronic hepatitis C: a randomized multicenter trial.
Hepatology 1996;24:1034-40.
5.Marcellin P, Hopf U, Trepo C, et al. A randomized, double-blind, controlled,
multicenter study of lymphoblastoid interferon alpha-nI in the treatment of
adults with chronic hepatitis C. J Hepatol. In press.
table Of Contents
|
