| |
EDITORIAL: Rescue therapy for drug resistant hepatitis B: Another argument for combination chemotherapy?
|
| |
| |
Gastroenterology
January 2004, Volume 126, Number 1
Tim Shawa 1, Scott Bowdena 1, Stephen Locarnini* a 1
aVictorian Infectious Diseases Reference Laboratory, North Melbourne, Victoria, Australia. 1Dr. Locarnini is a member of the Gilead Sciences advisory board and has received a research contract from Gilead.
"...study by Peters et al--After randomization, patients in the ADV monotherapy arm experienced hepatitic flares, which were not observed in the combination arm, suggesting the need for more effective virologic suppression in this setting... Combination chemotherapy has already proven successful in suppressing HIV infection, with consequent improvement in quality of life for those treated, and there seems little doubt that it will soon become more routinely used to treat patients with CH-B... Should LMV and ADV be used sequentially or simultaneously, and will their simultaneous use accelerate development of dual or multidrug resistance?... Other nucleosidic drugs, including emtricitabine, entecavir, and telbivudine, have progressed to advanced phase II or phase III clinical trials, and may become available in the near future...Optimism must be tempered by recognition that even long-term combination chemotherapy with safe and efficacious nucleosidic drugs will probably be insufficient to eliminate HBV mini-chromosomes from the nuclei of infected cells, which will require additional drugs and almost certainly immunomodulators... ongoing development.... should make...optimization of combination chemotherapy for controlling CH-B."
Chemotherapeutic control of chronic hepatitis B (CH-B) infection has remained problematic since the hepatitis B virus (HBV) was identified as the major cause of serum hepatitis more than 30 years ago. Until recently, interferon was the only drug available for use against CH-B, but its use frequently causes undesirable side effects, its efficacy is limited, and it must be administered by injection, so the need for more efficacious anti-HBV agents has long been evident.1 Before the advent of antiretroviral drugs, most attempts to treat CH-B by using nucleoside analogues had to be aborted because of toxicity and/or lack of efficacy.2 The discovery that the unusual replication strategy used by HBV entails an obligatory reverse transcription step catalyzed by the viral polymerase3 prompted the investigation of antiretroviral nucleoside and nucleotide analogues for potential anti-HBV activity. Lamivudine (LMV) and adefovir (ADV), both of which had been previously identified as having antiretrovirus activity, were among the first of such compounds shown to possess more potent antihepadnaviral activity. Unfortunately, their long-term efficacy is compromised by the almost inevitable emergence of drug-resistant mutant HBV populations, a scenario analogous to the antiretroviral experience.4
LMV ([-]--L-2´,3´-dideoxy-3´-thiacytidine) is a synthetic deoxycytidine analogue originally developed to treat human immunodeficiency virus (HIV) infection. It was found to inhibit HBV replication in patients who were coinfected with HIV and HBV, and, in 1998, it became the first oral treatment for CH-B to gain Food and Drug Administration approval.5 Treatment of CH-B patients with LMV results in a rapid decrease in viral load and reversal or arrest of existing liver disease in the majority of cases. Unfortunately, these improvements are rarely sustained because of the development of drug resistance. HBV resistance to LMV has been well characterized, both molecularly and clinically.4,6--8 Mutations that result in replacement of methionine (M) in the tyrosine-methionine-aspartate-aspartate catalytic site motif of HBV reverse transcriptase (rt) by valine (V), isoleucine (I), or serine (S), which have been designated rtM204V, rtM204I, and rtM204S, respectively,8 are sufficient to confer resistance to LMV. The rtM204I substitution has been detected in isolation, but rtM204V and rtM204S are found in association with other changes that may partially compensate for replicative defects imposed by rtM204V/I/S.9 LMV resistance results mainly from steric hindrance. Molecular modeling shows that, when rtM204 is replaced by valine or isoleucine, the side-chains of the substituted amino acids project into the deoxynucleoside triphosphate (dNTP) binding site, reducing the binding affinity for natural dNTPs and preventing LMV binding by occupying the space needed to accommodate its 3´ sulfur atom.10 Mutations that confer LMV resistance decrease sensitivity to LMV from 20 to >1000-fold in in vitro assays.
The incidence of LMV resistance in patients with CH-B rises progressively at rates between 14% and 32% annually, approaching 100% after 4 years. This relatively slow development (compared with HIV) is probably caused by constraints imposed by overlapping reading frames in the HBV genome and by reduced opportunities for complementation and recombination imposed by the genetic haploidy of HBV. High pretherapy serum HBV DNA and alanine aminotransferase (ALT) levels, longer duration of therapy, and incomplete suppression of viral replication are the main risk factors known to accelerate development of LMV resistance. These factors are consistent with the absolute dependence of mutagenesis on viral replication. In established CH-B, virtually all hepatocyte nuclei are infected with an apparently stable population of HBV mini-chromosomes. The emergence of a new mutant HBV as the dominant population requires that HBV mini-chromosomes containing mutant genomes must displace the existing population. During active liver disease, hepatocyte turnover and proliferation increases, providing what has been termed "replication space," that is, a fresh supply of infectible hepatocytes for colonization by mutant HBV.14 Under these conditions and appropriate selection pressure, a new HBV population may become predominant.
ADV, an acyclic deoxyadenosine monophosphate analogue, is the biologically active component of is its oral prodrug (9-{2-[bis[(pivaloyloxy)methoxy](phosphoinyl]methoxy]ethyl}-adenine) dipivoxil, which gained Food and Drug Administration approval for use against CH-B in September 2002. It belongs to a class of broad-spectrum antiviral agents, the nucleoside phosphonates, and was originally investigated as a potential antiretroviral drug. It underwent clinical trials for treatment of HIV infection but is not currently used as an antiretroviral drug because continued administration of the relatively high dose required to suppress HIV replication (100--120 mg/day; at least 10-fold greater than for HBV) was found to have the potential to cause nephrotoxicity. Clinical experience with ADV as an anti-HBV drug is limited but increasing. Large-scale multicenter clinical trials longer than 48 weeks have shown that its efficacy is comparable to that of LMV. Viral resistance did not develop during 60 weeks treatment with ADV; on the contrary, increasing suppression of viremia was observed as treatment progressed. However, an HBV mutant that exhibited a 6- to 15-fold reduction in sensitivity to ADV in vitro was isolated from 2 of 124 hepatitis B e antigen (HBeAg)-negative patients who had been treated with ADV for 96 weeks. Resistance to ADV in these isolates was caused by substitution of threonine for asparagine at rt236 (rtN236T), which appears to be sufficient to confer ADV resistance. The rtN236T change does not affect sensitivity to LMV. At the molecular level, ADV resistance appears to be caused by indirect perturbation in the triphosphate binding site within the dNTP binding pocket.21 The different mechanism may explain why HBV resistance to ADV is less common than resistance to LMV and does not confer as great a reduction in drug sensitivity. LMV and ADV are taken up and activated by different biochemical pathways: their activated products compete with different dNTPs (dCTP and dATP, respectively) and their toxicities do not appear to overlap. There is some evidence that ADV is active against HBV in cells other than hepatocytes and it has also been shown that LMV and ADV act at least additively in vitro. Finally, it has been reported that the rtL180M+rtM204V changes that confer LMV resistance fortuitously confer a 5-fold increase in sensitivity to ADV in vitro. Collectively, these data amount to a strong argument for the clinical use of ADV and LMV in combination.
Preliminary clinical trials have already shown that ADV is effective against LMV-resistant HBV in vivo, either when used alone or in combination with LMV. Two new reports in this issue of GASTROENTEROLOGY record the longer-term results of separate large-scale multicenter, controlled trials of ADV alone or in combination with LMV in cases of LMV resistant CH-B. Perrillo et al. studied a total of 135 cases of LMV resistant CH-B. All cases continued LMV therapy (100 mg/day); in addition, 86 patients received ADV (10 mg/day) whereas the remaining 49 were administered a placebo. Forty of those receiving combination treatment had signs of decompensated liver disease or had recurrent posttransplant hepatitis. One hundred twenty-six patients (80 and 46 in the ADV and placebo groups, respectively) completed the 52 week study, in which the primary endpoint was a decline in serum HBV DNA to 105 copies/mL or 2 log10 reduction from baseline at both weeks 48 and 52. The endpoint was achieved in an overwhelming majority (>85%) of patients after ADV treatment, whereas only a small minority (11%) of the placebo group responded, a difference that was highly statistically significant (P < 0.001). In addition, ALT normalization was observed in 30% of those receiving the drug combination but in only 6% of those who continued LMV monotherapy. A majority of patients (88%) were HBeAg positive pretreatment: HBeAg loss occurred in 15% and 2% of combination and monotherapy patients, respectively. Responses to combination therapy were similar regardless of disease status, and adverse side effects were not observed. The authors conclude that the addition of ADV to LMV treatment is beneficial to patients with compensated or decompensated liver disease caused by LMV-resistant CH-B. This raises 2 new issues: (1) whether ADV alone would be sufficient to control LMV-resistant HBV and (2) whether the combination of LMV and ADV would be more efficacious than monotherapy with either drug alone in treatment-naive patients. Results of the second study by Peters et al.29 provide partial clarification. The cohort of 59 patients in this study were all HBeAg positive with LMV-resistant CH-B, had compensated liver disease, and fulfilled entry criteria similar to those applied in the study by Perrillo et al.28 Patients were randomized into 3 groups: those who either continued LMV monotherapy, received a combination of LMV and ADV, or received ADV alone. The primary endpoint was the time-weighted average reduction in serum HBV DNA up to week 16, although treatment was continued and patients monitored for at least 48 weeks. Results and conclusions were similar to those of the first study: a rapid virologic response was observed only in ADV recipients and not in those continuing LMV monotherapy. At week 48, median serum HBV DNA decreases were 3.59 and 4.04 log10 from baseline in the groups receiving the ADV/LMV combination and ADV, respectively, with no change in the group on LMV monotherapy. In addition, ALT normalized in approximately half the patients who received ADV compared with only 5% of those given LMV alone. After randomization, patients in the ADV monotherapy arm experienced hepatitic flares, which were not observed in the combination arm, suggesting the need for more effective virologic suppression in this setting. Both studies used conservative primary endpoints based on serum viral load reductions from baseline measurements because no liver tissue was available for histologic or virologic analysis. Although assays of different sensitivity were used in the respective studies, the median decrease in viral load from baseline to week 48 was around 4 log10 copies/mL in the patients receiving ADV/LMV or ADV monotherapy. Patients with decompensated liver disease appeared to respond equally well to ADV as those patients with compensated disease. In HBeAg-positive patients, HBeAg loss was confined to patients receiving ADV, with the exception of 1 patient receiving LMV and placebo. Although it was shown from the studies that several patients lost HBeAg, the inadequacy of using this as an endpoint was reported by Peters et al., who showed that the response was not durable and the majority of their patients with HBeAg loss regained HBeAg during the posttreatment follow-up. One patient in this study underwent hepatitis B surface antigen (HBsAg) seroconversion and 2 patients in the Perillo Study lost HBsAg on treatment. However, both patients in the latter study had undergone liver transplantation and had received hepatitis B immunoglobulin posttransplant.
These 2 studies confirm that ADV dipivoxil is active against LMV-resistant HBV in vivo and also indicate that LMV provides little, if any, antiviral or clinical benefit for patients in whom LMV-resistant HBV is already present. Unfortunately, they provide few insights to the probable responses to ADV and LMV—given either sequentially in that order or simultaneously in combination—of treatment-naive cases. Combination chemotherapy has already proven successful in suppressing HIV infection, with consequent improvement in quality of life for those treated, and there seems little doubt that it will soon become more routinely used to treat patients with CH-B. Mutations that confer LMV resistance are different from those that confer ADV resistance, and cross-resistance between ADV and LMV has not been observed to date. Clearly, ADV can be used to rescue cases in which LMV resistance has developed and, presumably, vice versa. In such cases, continuation of LMV after ADV rescue therapy has been initiated appears to provide little benefit, but will the converse be true? Should LMV and ADV be used sequentially or simultaneously, and will their simultaneous use accelerate development of dual or multidrug resistance? Currently, the wholesale cost of treatment with LMV is less than two-thirds the cost of ADV treatment (approximately US$260 versus US$450 per month, respectively). Other nucleosidic drugs, including emtricitabine, entecavir, and telbivudine, have progressed to advanced phase II or phase III clinical trials, and may become available in the near future, but whether they will be active against ADV-resistant HBV in vivo is presently unknown. Monotherapy with LMV, ADV, or the newer nucleosidic agents consistently produces rapid and dramatic decreases in viremia. Unfortunately, only a small proportion (12%--25%) of treated individuals lose HBeAg and show normalization of serum ALT liver histology, accompanied by improvements in liver histology. Moreover, HBeAg seroconversion is uncommon and HBsAg seroconversion is a rare event. Evidence to date indicates that the dramatic initial response in viral loads are invariably followed by a much slower elimination of residual virus, during which the risk of appearance of drug-resistant virus is cumulative. It would not be overly optimistic to predict that acceleration of the rate of virus elimination during the second phase would be enhanced by rational combination chemotherapy. Optimism must be tempered by recognition that even long-term combination chemotherapy with safe and efficacious nucleosidic drugs will probably be insufficient to eliminate HBV mini-chromosomes from the nuclei of infected cells, which will require additional drugs and almost certainly immunomodulators. If elimination of viral mini-chromosomes is achievable, it will become necessary to adopt new endpoints to monitor progress. Fortunately, ongoing development of more sophisticated, accurate, and sensitive methods and approaches to monitor treatment responses and to detect developing drug resistance should make this possible and facilitate the development and optimization of combination chemotherapy for controlling CH-B.
|
|
| |
| |
|
|
|
