Randomized, Double-Blind Trial Comparing Pegylated Interferon Alfa-2b to Interferon
Alfa-2b as Initial Treatment for Chronic Hepatitis C
HEPATOLOGY August 2001;34:395-403
Karen L. Lindsay,1 Christian Trepo,2 Tobias Heintges,3 Mitchell L. Shiffman,4 Stuart C. Gordon,5 John C. Hoefs,6 Eugene R. Schiff,7 Zachary D. Goodman,8 Mark Laughlin,9 Ruji Yao,9 And Janice K. Albrecht 9 For The Hepatitis Interventional Therapy Group 10
The full version of this article, including tables, is available in PDF format.
The primary objective of this international, randomized, active-controlled, parallel group, double-blind study was the evaluation of the safety and efficacy of peginterferon alfa-2b (0.5 ug/kg, 1.0 ug/kg, or 1.5 ug/kg QW) compared with interferon alfa-2b (3 MIU TIW) in adult patients with clinically compensated chronic hepatitis C who had not been treated previously with interferon.
As summarized in Table 2 and Fig. 1, the end-of-treatment virologic response (ETVR) for peginterferon alfa-2b was dose related. Increase in dose of peginterferon alfa-2b (0.5, 1.0, 1.5 ug/kg) was associated with an increase in the virologic response rate (33%, 41%, 49%, respectively). All 3 peginterferon alfa-2b dose groups had significantly higher proportions (P <.01) of patients with ETVR compared with the interferon alfa-2b group (24%). The viro-logic response rates were approximately twofold higher with 1.5 ug/kg peginterferon alfa-2b than with interferon alfa-2b at treatment week 48 (49% vs. 24%; P <.001). The higher ETVR rate in patients treated with 1.5 ug/kg peginterferon alfa-2b compared with those treated with 1.0 ug/kg peginterferon alfa-2b (49% compared with 41%, P =.049) is largely the result of a significantly higher response rate in HCV genotype 1 infected patients (87/223, 39% compared with 50/199, 25%, respectively, P =.002).
As with ETVR, the 3 doses of peginterferon alfa-2b had significantly higher SVR than interferon alfa-2b. The rate of SVR was approximately twofold higher with 1.0 ug/kg peginterferon alfa-2b (25% vs. 12%; P <.001) and 1.5 ug/kg peginterferon alfa-2b (23% vs. 12%; P <.001), as compared with interferon alfa-2b. However, unlike the virologic response at the end of treatment, there was not a dose response between the 1.0 ug/kg and 1.5- ug/kg peginterferon alfa-2b dose groups for SVR, 25% and 23%, respectively. This was related to a significantly higher relapse rate in the HCV genotype 1 patients treated with 1.5 ug/kg compared with those treated with 1.0 ug/kg peginterferon alfa-2b, 57/87 (66%) and 23/50 (46%), respectively (P =.025), whereas the relapse rate among patients infected with genotypes 2 or 3 was similar, 20/56 (36%) and 24/63 (38%), respectively.
Combined Virologic and Biochemical Response
When the virologic response was considered together with the biochemical response to treatment (combined response), the proportion of subjects with undetectable serum HCV RNA and normal ALT values was significantly higher for the 1.0- ug/kg (31%; P < .002) and 1.5- ug/kg (33%; P <.001) peginterferon alfa-2b dose groups than for the interferon alfa-2b group (20%) at the end of treatment (Table 2). The sustained combined response rate at week 72 was twofold higher in subjects treated with 1.0 ug/kg (24%; P <.001) or 1.5 ug/kg (23%; P <.001) of peginterferon alfa-2b than in subjects treated with interferon alfa-2b (12%). Although the combined virologic and biochemical response rate to the 0.5 ug/kg peginterferon alfa-2b dose was numerically higher than the response rate to interferon alfa-2b at 48 and 72 weeks, these differences were not statistically significant. Nearly all subjects who exhibited a sustained loss of HCV RNA had normal ALT values at week 72 (228/238). However, sustained normal ALT values were a poor predictor of sustained HCV RNA loss. Among subjects with normal ALT values after 24 weeks of follow-up, sustained virologic responses occurred in 67% of those treated with interferon alfa-2b, 68% of those given 0.5 ug/kg peginterferon alfa-2b, 80% of those who received 1.0 ug/kg peginterferon alfa-2b, and 82% of those treated with 1.5 ug/kg peg-interferon alfa-2b.
Pretreatment and posttreatment liver biopsies were analyzed in 61% (744/1219) of the subjects. The percentages of subjects with histologic improvement and the mean degrees of improvement are shown in Table 3. Overall, sustained improvement in hepatic inflammation (Knodell HAI categories I, II, III) was observed in all treatment groups. Mean changes in Knodell scores (-1.2 to -1.8) and the proportion of subjects with improved scores (47-50%) were similar across all 3 peginterferon alfa-2b dosing groups and the interferon alfa-2b group. When treatment groups were stratified according to virologic response (sustained response, relapse after an end-of-treatment response, and no response), improvement in hepatic inflammation scores, defined as at least a two-point decrease in the inflammatory score, was highly associated with sustained virologic response. The proportion of subjects in all treatment groups who showed an improvement in hepatic inflammation was approximately twofold higher among subjects who had a sustained response (77%-90%) than among those who relapsed after an end-of-treatment response (33%-46%) or did not respond at all (33%-41%). In addition, larger decreases in the group mean change from baseline scores were reported among subjects with a sustained re-sponse (-4.0 to -5.4) than among those with a relapse fol-lowing response (-0.2 to -0.9) or no response (-0.3 to -0.7). Hepatic fibrosis scores improved in 13% to 20% of subjects (Table 3). Similar to the observations with respect to hepatic inflammation scores, the proportion of subjects who showed improvement in hepatic fibrosis score (Knodell HAI category IV) was higher among subjects who had a sustained response (21%-37%) than among those who either relapsed or did not respond (4%-17%).
Factors Associated With Virologic Response
Logistic regression analysis of baseline disease characteristics and demographic variables identified only 2 covariates associated (P <.001) with sustained virologic response: an HCV genotype other than 1 and baseline HCV RNA viral levels of < 2 million copies/mL serum. Subgroup analysis of end-of-treatment and sustained viro-logic response by HCV genotype and baseline virus levels (Table 4) revealed that subjects with HCV genotype 2 or 3 and < 2 million HCV RNA copies/mL serum had the highest sus-tained response rates in all treatment groups (36%-68%). Among them, the 1.5-ug/kg peginterferon alfa-2b dose group attained the highest sustained response rate (68%), nearly twofold higher than that seen in the interferon alfa-2b group (36%). The end-of-treatment response rate in patients with genotype 2/3 infection was similar, however, regardless of the baseline level of HCV RNA.
Among genotype 2/3 infected patients, end-of-treatment virological response to treatment with 1.0 or 1.5 ug/kg peginterferon alfa-2b was nearly identical, 76% and 77%, in all 4 subgroups regardless of viral level. Virologic relapse (Table 5) occurred less frequently in patients with lower baseline levels of HCV RNA, however, in that 19% and 12% of patients with lower levels of HCV RNA relapse after treatment was discontinued, but 45% and 46% of those with higher levels of HCV RNA developed virologic relapse. The most common viral profile, HCV genotype 1 and > 2 million RNA copies/mL serum was present in 652/1219 (53%) subjects. In this subgroup, end-treatment virologic response rates occurred in 20%, 20%, and 35% of subjects in the pegin-terferon alfa-2b 0.5-, 1.0-, and 1.5- ug/kg treatment groups. End-of-treatment response in the group treated with peginterferon 1.5- ug/kg was nearly fourfold higher than that in the group treated with interferon alfa-2b (35% compared with 9%). On discontinuation of treatment after 48 weeks, 63% to 80% of patients developed virologic relapse, however, resulting in a sustained virological response rates of 5% to 8% in the peginterferon-treated groups and 2%in the group treated with interferon alfa-2b. In contrast to genotype 2/3 infected pa-tients, end-of-treatment response among genotype-1 infected patients treated with peginterferon 1.0- or 1.5- ug/kg varied considerably as a function of baseline HCV RNA level, in that 43% and 50% of patients with lower levels of HCV RNA experienced an end-of-treatment response, whereas only 20% and 35% of patients with higher levels of HCV RNA responded. Virologic relapse after discontinuation of treatment also varied by baseline viral level, in that 17% and 36% of patients with lower HCV RNA levels relapsed after treatment was discontinued, whereas 63% and 80% of those with higher HCV RNA levels developed virologic relapse.
Because previous studies have shown that an early response to interferon treatment is associated with sustained virologic response in patients with chronic hepatitis C,2,16-19 the time at which HCV RNA was first negative (<100 copies/mL serum) was analyzed in relationship to subsequent development of sustained virologic response. In each treatment group, the likelihood of an SVR occurring was highest in subjects whose first negative HCV RNA occurred at treatment week 4 (77%-86%), compared with those in whom HCV RNA was first negative at treatment week 12 (32%-52%), and those whose HCV RNA was first negative at treatment week 24 (13%-20%). For the interferon alfa-2b group as well as the peginterferon treatment groups, nearly all (93% to 100%) patients who eventually became sustained responders had developed undetectable serum HCV RNA by treatment week 24. These data confirm previous observations that early virologic response during treatment is associated with a higher likelihood of sustained response. The sensitivity of a negative serum HCV RNA result at treatment week 4 was low, however (49% in the interferon alfa-2b group and 46% to 65% in the peginterferon alfa-2b groups). Negative predictive values (the likelihood that a sustained response would occur if HCV RNA was not detected) for treatment week 4 were calculated for subjects treated with 1.0 ug/kg and those treated with 1.5 ug/kg, and were 85% and 77% respectively. The positive predictive value (the likelihood that a sustained response would not occur if HCV RNA was detected) for HCV RNA at treatment week 4 in these 2 treatment groups was 84% and 90%.
|Relapse Rate in Patients Treated With Peginterferon Alfa-2b by Baseline HCV RNA Level (Million Copies/mL) and Genotype|
|Number (%) of Subjects|
1.0 ug/kg QW
1.5 ug/kg QW
|Genotype 1 (All)||23/50 (46)*||57/87 (66)*|
|< 2 million||3/18† (17)||10/28 (36)|
|> 2 million||20/32 (63)||47/59 (80)|
|Genotypes 2/3 (All)||24/63 (38)||20/56 (36)|
|< 2 million||3/16 (19)||2/17 (12)|
|> 2 million||21/47 (45)||18/39 (46)|
|*P =.026 for the comparison with 1.0 ug/kg QW and 1.5 ug/kg QW.|
|†Of the 18 patients who had an end-of-treatment response, 15 had undetectable HCV RNA at Week 72; an additional patient who had detectable HCV RNA at Week 48 had undetectable HCV RNA at Week 72.|
|Of the 28 patients who had an end-of-treatment response, 18 had undetectable HCV RNA at Week 72; an additional patient who had detectable HCV RNA at Week 48 had undetectable HCV RNA at Week 72.|
With respect to safety and tolerability, peginterferon alfa-2b (all doses) was comparable to interferon alfa-2b. No new or unexpected adverse events specific to peginterferon alfa-2b were reported. The incidence and severity of adverse events were similar between peginterferon alfa-2b 0.5 ug/kg and 3 MIU interferon alfa-2b, whereas pro-files for the higher doses of peginterferon alfa-2b (1.0 ug/kg and 1.5 ug/kg) were similar to each other (Table 6) and associated with a somewhat higher frequency of fever and chills. As has previously been seen with interferon alfa-2b, the most adverse events with peginterferon alfa-2b were flu-like symptoms including headache, myalgia, fatigue, rigors, and fever. Compared with interferon alfa-2b, the incidence of injection site reactions was increased approximately twofold in patients receiving peginterferon alfa-2b. In all cases, the characteristics of the reaction were similar for both interferon alfa-2b and peginterferon alfa-2b: the event was generally mild, not treatment-limiting, and characterized by localized erythema. As previously reported for interferon alfa-2b, leukocyte, neutrophil, and platelet counts decreased from baseline during the first few weeks of treatment in all groups, stabilized during the remainder of treatment, and reverted to baseline levels upon cessation of treatment (data not shown).
For all 3 hematological parameters, the profile over time for peginter-feron alfa-2b was similar to that of interferon alfa-2b. Dose reduction for neutropenia occurred infrequently with both peginterferon alfa-2b and interferon alfa-2b, but was more common in the 1.5 ug/kg treatment group (5%) compared with the other treatment groups (2% to 3%). Dose reduction for thrombocytopenia was more common in the peginterferon treatment groups (2% to 3%) compared with the interferon alfa-2b group (0.3%).
As shown in Table 6, the incidence of dose reduction in-creased with higher doses of peginterferon alfa-2b, and was higher than for interferon alfa-2b. However, discontinuation rates were comparable in all 3 peginterferon alfa-2b groups (9%-11%) and were slightly higher than in the interferon alfa-2b group (6%).
Peginterferon alfa-2b, a pegylated formulation of interferon alfa-2b, was developed to decrease the rapid clearance of interferon alfa-2b. Hepatitis C viral infection is an ideal setting in which to use peginterferon alfa-2b because maintaining constant antiviral pressure on hepatitis C virus is crucial during therapy because of the short virion half-life (2.7 hours) and estimated production and clearance rate of 10 (12) virions/ day.20 Nonpegylated interferon alfa-2b is undetectable after approximately 24 hours of administration (Laughlin, M., un-published data, 1995) thus permitting viral breakthrough and the development of resistance. Because the plasma half-life of peginterferon alfa-2b in humans is approximately 10-fold longer than that of interferon alfa-2b (Laughlin, M., unpub-lished data, 1995), the pegylated form intensifies interferon therapy by maintaining effective plasma drug concentrations over time. In addition, the weekly dosing schedule with pegin-terferon alfa-2b may positively affect patient compliance and quality of life.
This database further substantiates the profound influence of hepatitis C viral genotype and HCV RNA level on response to alpha interferon, and emphasizes the importance of further research to identify the mechanisms of viral resistance.21-23 Despite optimization of drug delivery, HCV genotype and viral load had a major impact on the sustained response rate. Patients receiving 1.0 or 1.5 ug/kg peginterferon alfa-2b had sustained virologic response rates ranging from 62% to 68% in those infected with genotype 2/3 and < 2 million copies HCV RNA/mL serum compared with 7% to 8% in those infected with genotype 1 and > 2 million copies HCV RNA/mL serum.
The influence of viral genotype was a significant factor in the finding that the virologic response rate at the end of treatment was greater in the 1.5- ug/kg peginterferon alfa-2b dose group than in the 1.0- ug/kg dose group (49% and 41%), but the proportion of subjects with sustained virologic responses was similar for the 2 groups (23% and 25%). The higher dose of peginterferon alfa-2b induced a response in a significantly greater proportion of patients infected with HCV genotype 1 but this response was not maintained following the end of treatment.
Results from our study are very useful in defining strategies for more effective antiviral therapy against hepatitis C virus. End-treatment response and sustained response rates have been conventionally used to describe results in clinical trials since they were proposed at the NIH Consensus Development Conference on Hepatitis C in 1997. 26 Since that time, it has been shown that sustained virological response (SVR) is associated with the greatest long-term clinical benefit.27-29 Sustained virological response is a function of both the end-treat-ment response rate and the degree of posttreatment relapse, each of which can be differentially affected by a treatment regimen. Based on antiviral pharmacodynamic studies, it appears that clearance of hepatitis C virus from serum and intracellular compartments is important during therapy.3,20
We have shown that the rate of virologic relapse varies considerably among patient subpopulations treated with the same peginterferon monotherapy regimens, and this may reflect a differential clearance rate from intracellular compartments. Relapse rates are lower and sustained virological response rates higher in patients treated with the combination of ribavirin and interferon alfa-2b for 48 weeks compared with those treated with interferon alfa-2b alone.1,2