| |
A pilot study of therapeutic vaccination with envelope protein E1 in 35 patients with chronic hepatitis C;
|
| |
| |
EDITORIAL
Hepatology November 2003, Volume 38, Number 5
Nevens and colleagues
Department of Hepatology, UZ Gasthuisberg. Innogenetics NV, Ghent, Belgium
ABSTRACT
New treatments are needed for chronic hepatitis C patients in whom viral clearance cannot be achieved. Thirty-five chronic hepatitis C patients (genotype 1) were randomized to receive 20 µg of recombinant HCV E1 (E1) (n = 26) or placebo (n = 9) intramuscularly at weeks 0, 4, 8, 12, and 24. Thirty-four then received open-label E1 vaccine at weeks 50, 53, 56, 59, 62, and 65. Twenty-four patients (12 men, 12 women; mean age, 52 y; 18 interferon-based treatment failures; mean baseline alanine aminotransferase [ALT] level, 118 IU/L) underwent a biopsy before and after 2 courses of E1, 17 months later. Liver histology was scored by 2 blinded pathologists according to the Ishak and Metavir systems. Postinjection reactions were similar to placebo (alum only). Nine of 24 patients (38%) had improvement of 2 points or more, 10 (41%) remained stable, and 5 (21%) showed worsening in total Ishak score. Nine patients (38%) improved both on Ishak and Metavir fibrosis scores. Plasma HCV-RNA levels remained unchanged, whereas ALT levels showed a trend toward a decrease during treatment. All but 3 patients developed a significant de novo E1-specific T-cell response. The increase in anti-E1 antibody levels correlated with the decrease in total Ishak score and with the relative decreases in both Ishak fibrosis score and ALT level (all P .01). In conclusion, E1 therapeutic vaccination is well tolerated and the observed effects warrant further study.
The authors concluded: the encouraging results of this study were observed in a relatively small patient sample and over a limited period of time. E1 therapeutic vaccination is very well tolerated, which is important for treatment adherence. Despite its unresolved mechanism of action, this treatment modality deserves consideration as a novel way to possibly halt or reverse progression of liver fibrosis in the large number of hepatitis C patients who need an alternative to current antiviral therapies. For this reason, a further open-label extension has been started in which a second biopsy is planned for the placebo/E1 group once these patients have received a second course of E1. In addition, a larger European, placebo-controlled, randomized trial in chronic hepatitis C genotype 1 patients with no treatment alternatives has been initiated to confirm the effect of E1 treatment on liver histology.
At the AASLD meeting in Boston, October 24-29, the study authors reported immune responses seen in phase I study of 21 volunteers and patients. Link to read the AASLD abstract:
http://www.natap.org/2003/AASLD/day4_3.htm
EDITORIAL IN HEPATOLOGY
Marc G. Ghany, M.D.
David E. Kleiner, M.D., Ph.D.
National Institute of Diabetes and Digestive and Kidney Diseases and the National Cancer Institute, National Institutes of Health, Bethesda, MD
Worldwide, chronic hepatitis C virus (HCV) infection is a major cause of cirrhosis, end-stage liver disease, and hepatocellular carcinoma, and in the United States is the most common indication for orthotopic liver transplantation in adults. The natural history of the infection is not fully understood, but up to 25% of chronically infected individuals may develop these complications. Therapy is therefore advised for patients with advanced disease. The goals of therapy are straightforward: to eradicate viremia and as a consequence reverse hepatic fibrosis, reduce rates of hepatocellular carcinoma, and ultimately improve survival. Over the past decade therapeutic interventions have improved, and sustained rates of viral clearance from serum have increased from 10% with interferon monotherapy to 56% with pegylated interferon and ribavirin. More importantly, follow-up studies ranging from 1 to 15 years indicate that viral eradication is associated with improvement in hepatic fibrosis and reduction in the risk of liver cancer. Unfortunately, 13% of patients relapse and 31% fail to respond to pegylated interferon and ribavirin and thus may not benefit from therapy. The majority of relapsers and nonresponders are infected with genotype 1.
If sustained virologic eradication cannot be achieved, then an alternate objective is to slow or reverse progression of hepatic fibrosis, which usually parallels the decline in hepatic function and is probably the best predictor of long-term outcome. As we come to understand the molecular basis of hepatic fibrosis, specific anti-fibrogenic drugs may be developed. In the interim, antiviral or immunomodulatory agents are being evaluated as antifibrotics. Several studies have shown improvement or stabilization of histologic fibrosis following interferon therapy in patients who fail to eliminate virus from serum. Long-term interferon therapy in nonresponders is currently being evaluated for this role in a large multicenter National Institutes of Health (NIH) funded trial, the Hepatitis C Antiviral Long-term Treatment against Cirrhosis (HALT-C) trial, and the results of this and similar trials are eagerly awaited. Long-term ribavirin monotherapy may also be beneficial in selected patients. Other experimental strategies such as interleukin 12 have been shown to be ineffective. Clearly new therapies and different approaches are needed for relapser and nonresponder patients to combination therapy.
Aside from antiviral drugs, another approach has been to manipulate the immune response via active or passive vaccination. The failure of an adequate immune response from a variety of causes, such as inhibition of a specific immune response, the emergence of viral escape mutants, lack of neutralizing antibodies, and host tolerance, have been suggested as possible mechanisms of viral persistence. There is preliminary evidence that boosting the antibody response by active or passive immunization may be beneficial. It has been shown that vaccination of naive chimpanzees with recombinant E1 and E2 glycoproteins was able to prevent experimental infection after challenge with homologous virus but not heterologous virus. However, the course of infection following rechallenge with heterologous virus was attenuated and the majority of animals did not develop chronic infection. This suggests that vaccination with an envelope protein might alter the course of infection and be beneficial for chronically infected patients. In this issue of HEPATOLOGY, Nevens et al. have taken this approach by using a therapeutic viral E1 envelope vaccine and have claimed some startling results.
The original study was designed as a randomized placebo controlled trial to assess the tolerability and safety of a truncated E1 protein in patients with chronic hepatitis C. Interim analysis after 24 weeks suggested a favorable immune and biochemical response in vaccinated patients, and so the study was converted to open label with a follow-up liver biopsy performed in those initially randomized to the vaccine group (E1/E1). The investigators were able to demonstrate induction of a humoral response after the second course of vaccinations and a significant T-cell response in 21 of 24 patients, but this was not associated with a reduction in HCV RNA levels. Although the E1/E1-treated patients had a 23% reduction in alanine aminotransferase (ALT) levels, no change was observed in the placebo/E1 group. The vaccine was well tolerated. The surprising finding was that fibrosis improved in 9 (38%), remained unchanged in 5 (21%), and worsened in 10 (42%) patients when pre- and post-treatment biopsies were scored blindly using the Ishak and Metavir scoring systems. For comparison, one can look at the change in fibrosis scores in nonresponders to interferon and ribavirin therapy. In a large retrospective analysis of patients treated with different regimens of interferon and ribavirin, 17% of nonresponders improved and 21% worsened at least one point in fibrosis using the Metavir scale. In contrast, 25% of responders showed improvement whereas only 7% worsened. Thus, on the surface the results of the current study (which included 18 nonresponders) appear to be very encouraging.
How should we interpret these results? Are we to believe the investigators claim that vaccination with truncated E1 protein can reverse or halt progression of liver fibrosis? Should we recommend E1 vaccination for genotype 1 patients who fail peginterferon and ribavirin? Before we address these questions we must answer two more basic questions: How should we define disease improvement in uncontrolled clinical trials and how well has the current study met accepted parameters of improvement? At its simplest level, improvement is frequently defined as resolution of symptoms or elimination of the cause of disease. In chronic hepatitis C, viral and serum aminotransferase levels (typically ALT) and liver histology are followed to assess treatment response. For the first two, the accepted standard of improvement is clear: viral clearance to undetectable levels and normalization of serum ALT, with absence of viremia for 6 months after stopping therapy—a so-called sustained virologic response—being the best predictor of a long-term response. In the study by Nevens et al., no patient achieved viral clearance. Furthermore, it is difficult to discern whether any patient had a biochemical response because the data are presented as reduction in ALT levels and surprisingly the placebo/E1 group showed no improvement in ALT level at all after a course of E1 vaccination. Thus, the study failed to show improvement on the first two accepted parameters of response although a histologic response would still be a desirable outcome.
Judging improvement in histology is a more difficult yet fundamentally very important problem. Histologic evaluation is beset by a number of problems that are not shared by virologic or biochemical evaluation. Both viral and biochemical evaluations can be done on a nearly continuous basis, allowing investigators to sample disease activity until the end point is reached. Inadequate or unsatisfactory samples can be easily replaced. In contrast, liver biopsies are only done on a limited basis and rebiopsy for an inadequate or unsatisfactory sample is almost never done. This problem is compounded by observer variation, which even in the hands of the best observer results in 10% to 30% of second reading values being either higher or lower than the first reading, and by sampling issues that arise from disease quantitation based on a needle biopsy 1/50,000 the size of the organ. Nevertheless, it is possible to demonstrate changes in histology.
There are two accepted methods of defining histologic improvement in chronic liver disease. One method is to predefine a threshold change that will be considered significant, e.g., a 2-point reduction in the total inflammation on the Ishak scale, or a 1-point improvement in fibrosis. The proportion of patients who meet the criteria is reported but it is not possible to judge significance without a control group. This is the method that was chosen by Nevens et al. when they reported the central finding of their study. However, in an uncontrolled trial, the amount of change required should be chosen so that it is clearly beyond what would be expected by chance. In a natural history study of fibrosis progression in untreated patients, almost 20% of patients had a 1-point or greater reduction in fibrosis.28 Only 2% had a 2-point or greater change for the better, so a 2-point difference in fibrosis might be a better standard for uncontrolled trial.18 It is unfortunate that the investigators altered the study design and did not rebiopsy the placebo group.
A better method to detect histologic improvement in an uncontrolled study is to look for change in the cohort as a whole using an appropriate statistical test for paired samples. In this way one can define a shift for better or worse even if the average change is small. The disadvantage is that one cannot say anything about an individual case, nor can one define a percentage of the population as “improved.” However, when combined with an appropriate cutoff point for individual change, this can be a powerful tool for defining improvement in a characteristic such as histology. Given that the mean fibrosis and inflammation scores were essentially unchanged in the study by Nevens et al., the cohort would not have shown a statistically significant shift using paired sample analysis.
The study also raises a perplexing issue regarding the possible mechanism of action of the vaccine in reducing fibrosis. The expected clinical result after vaccination should have been a reduction in HCV RNA levels. We can only speculate that vaccination may have led to a down-regulation of the inflammatory response although this response would not be predicted after vaccination with a recombinant protein. Unfortunately the investigators did not perform additional immunologic studies to elucidate the mechanism of action of the vaccine, and this deserves further investigation.
In the final analysis we have an intriguing report on a novel agent that has shown some promise in achieving its immunological goals. With regards to the treatment response, although the data are presented in such a way as to suggest histologic and biochemical response, the investigators have failed to show that the variation they observe is due to a therapeutic response and not due to chance. In the future, we would recommend that uncontrolled studies use more stringent criteria than what can be used for controlled trials and that paired statistical analysis be used to show that a trend toward improvement is present in the cohort as a whole.
At present we cannot recommend the use of this strategy either as initial therapy for untreated patients or maintenance therapy for relapsers or nonresponders to peginterferon and ribavirin until the results of controlled trials are available. An intriguing question is whether this approach may have a role in combination with antiviral therapy. This may lead us to the therapeutic pot of gold.
|
|
| |
| |
|
|
|
