| |||||||||||||||||||||||||||||||||||||||||||||||
The reverse-transcriptase inhibitor lamivudine (Zeffix, GlaxoSmithKline) is often used to treat chronic infection with hepatitis B virus (HBV) until resistance develops. Treatment may then be switched to the reverse-transcriptase inhibitor adefovir (Hepsera, Gilead), which has a lower frequency of resistance. Here, we describe three cases of primary adefovir resistance that were sensitive to tenofovir (Viread, Gilead). All three cases involved a rare HBV variant with a valine at position 233 of the reverse-transcriptase domain instead of isoleucine (rtI233V), as in the wild-type virus. This HBV variant also displayed resistance to adefovir and sensitivity to tenofovir in vitro.
Case Reports
Patient 1
Patient 1, a 52-year-old man, had had chronic hepatitis B for more than five years. He had undergone unsuccessful therapy with pegylated interferon. Subsequent treatment with lamivudine (100 mg per day) had initially suppressed the viremia, resulting in fewer than 200 molecules of HBV DNA per milliliter. However, within three months, the level of viremia increased again, to 2.5x107 molecules of HBV DNA per milliliter (Figure 1), and a mutation conferring resistance to lamivudine9 — namely, the substitution of isoleucine for methionine at position 204 in the reverse-transcriptase domain (rtM204I) — was detected by direct sequencing of the polymerase-chain-reaction (PCR) product in the first available sample from September 9, 2002. After an interval without therapy, during which this mutation reverted to wild-type, the reverse-transcriptase inhibitor tenofovir was administered for nine months (300 mg per day) as an experimental drug for HBV. Tenofovir is licensed for treatment of the human immunodeficiency virus (HIV) and has been used successfully to reduce levels of both HIV and HBV in patients infected with both viruses.7,10,11 Within the first three months, the level of viremia decreased from 5.5x107 molecules of HBV DNA to 740 molecules of HBV DNA per milliliter and remained at similar levels for another six months. Thereafter, tenofovir was replaced by adefovir (10 mg per day) because this drug had become licensed for therapy of chronic HBV. However, during the therapy with adefovir, the level of viremia increased immediately and within seven months had reached 1.5x106 molecules of HBV DNA per milliliter. Replacement of adefovir with tenofovir again resulted in a substantial decrease in the level of viremia. The decrease was not as rapid as it had been before, but after 100 weeks of the second course of tenofovir therapy, the level of viremia was down to the detection limit of the PCR assay — at about 30 molecules of HBV DNA per milliliter. Throughout the observation period, the patient had normal or slightly elevated levels of alanine aminotransferase. He was positive for antibody against hepatitis Be antigen (HBeAg); sequencing showed that the virus had the normal precore sequence as well as the core promoter mutations T1753C, A1762T, and G1764A known to be associated with HBeAg negativity.12
|
Patients 2 and 3 were a married couple. Patient 2, a 56-year-old man, had received a liver transplant in 1993 because of alcoholic liver disease and turned out to be infected with HBV during follow-up. He had had continuous therapy with immunosuppressive agents since his liver transplantation. Patient 3, his wife, 52 years of age, was found to have chronic hepatitis B in 1994. The genomic sequences of their viruses were very similar, but in Patient 2, the predominant viral variant had a normal precore sequence and the stop codon mutation (G1896A), whereas Patient 3 had only wild-type virus. Both patients were HBeAg-positive. Both husband and wife received lamivudine, which initially reduced their levels of viremia from 107 molecules of HBV DNA per milliliter to less than 105 molecules of HBV DNA per milliliter. Resistance subsequently developed, and the level of viremia increased again (Figure 1). This increase was presumably caused by the rtM204I mutation associated with resistance to lamivudine, which had been identified in the viral population of both patients. Patient 3, in addition, had a virus variant containing the mutation rtV173L (Figure 2). Lamivudine therapy was replaced by adefovir therapy, but the level of viremia did not change significantly in either patient. The substitution of tenofovir therapy for adefovir therapy precipitated a rapid decrease in HBV DNA levels, from greater than 5x108 molecules of HBV DNA per milliliter in Patient 2 and from 4.5x107 molecules of HBV DNA per milliliter in Patient 3 to fewer than 100 molecules per milliliter in both patients at the end of the study (November 24, 2005).
|
Methods and Results
All three patients provided written informed consent for testing. To determine whether the rtI233V mutation alone is necessary and sufficient to render HBV resistant to adefovir, we employed site-directed mutagenesis to replace the isoleucine at position 233 with valine in the wild-type HBV genotype D15 using the plasmid vector pTHBV1.3 and the primer pair 5'CTTTTGTCTTTGGGTgTACATTTAAACCCTAAC3' (sense) and 5'GTTAGGGTTTAAATGTAcACCCAAAGACAAAAG3' antisense; the lowercase letter indicates the mutated nucleotide. The successful introduction of the mutation was confirmed by sequencing. The HepG2 subclonal cell line C3A (American Type Culture Collection) was plated in six-well cell culture plates at a confluence of 50 percent (approximately 1.5x106 cells) and then transfected with 12 µg of wild-type or mutated plasmid. On the next day, the cells were passaged into 12-well cell-culture plates. To eliminate input DNA, cells were incubated with 100 µg per milliliter of DNaseI (Roche) at 37°C for two hours. Two days after transfection, cells were treated with the free, unmodified drugs adefovir, tenofovir (Moravek), or lamivudine at concentrations ranging from 0.1 to 3 µM. The medium used to supplement the drugs was renewed daily. Cell-culture supernatants and cells were harvested on day 6. Replicative intermediates of HBV DNA were extracted as described previously,16 subjected to Southern blotting,17 and hybridized with a phosphorus 32–labeled probe generated with the Rediprime random prime labeling system (Amersham) from full-length HBV DNA. The quantity of HBV DNA was significantly decreased in treated cells at each concentration of adefovir (Figure 3A). With the wild-type plasmid, a concentration of adefovir of approximately 0.5 µM resulted in 50 percent inhibition, a value slightly higher than that described by Yang et al.18 but lower than that found by Brunelle et al.19 The rtI233V variant required concentrations of adefovir that were about six times as high for a similar level of inhibition (Figure 3A). This factor exceeds that of previously described variations between naturally occurring strains.18 In contrast, the inhibition efficacy of tenofovir or lamivudine did not differ significantly between the wild-type and the variant HBV DNA. Lamivudine was by far the most efficacious drug in this in vitro system. The inhibitory effect of all three drugs was most clearly demonstrated by the amount of open circular HBV DNA (Figure 3A); this is the only form leading to secreted infectious virus particles.
|
Discussion
In a large clinical study of the efficacy of adefovir as a treatment for chronic hepatitis B, the rtI233V mutation was not observed, but the rtN236T or rtA181V mutation developed in 5.9 percent of the patients within three years. The in vitro efficacy14 against these adefovir-induced mutants decreased by a factor of 2.5 to 13.8. Thus, the decrease of in vitro efficacy by a factor of 6 to 10 caused by the naturally occurring variant at position rt233 may indicate that this variation in vivo is even more relevant than previously identified adefovir-induced resistance mutations.
The rtI233V mutation was already present before adefovir therapy was initiated. It was independent of lamivudine resistance; as was seen in Patient 1, no lamivudine resistance mutation was detected at the beginning of adefovir therapy. Neither HBeAg status nor the immune status seems to be important for adefovir resistance of this variant, since these two factors differed in the three patients. Furthermore, the variant rtI233V was stable during an observation period of up to 220 weeks in these patients, even without selective pressure from adefovir therapy.
We identified these three rtI233V strains among the strains isolated from 80 patients whose HBV had been sequenced in past years at the Institute of Medical Virology in Giessen, Germany. Using the S gene sequence for genotyping, we identified 41 patients with genotype D (which has a worldwide distribution), 30 with A2, 2 with A1, 5 with C, 1 with E, and 1 with G plus A2. Our three patients had genotype D (subgroup D3)20 and had closely related but not identical reverse-transcriptase and S gene sequences. Remarkably, all three had the HBsAg subtype corresponding to an isoleucine at the polymorphic site 127 of the HBsAg protein, which is very rare (3 of 152 strains) among patients with the genotype D worldwide.20 The frequency of the 3 HBV strains with the ayw4 subtype and the rtI233V mutation among the 80 sequenced HBV strains may not reflect the real frequency of such variants in Germany. A GenBank search revealed that only 3 of about 500 previously reported HBV strains had the rtI233V mutation. Two of these strains were from Southeast Asia and were genotype C,21 and the third was a HBV strain found in gibbons.22 A lamivudine-resistant genotype C strain found in Chinese patients had a threonine at that position, but this variant may not have been viable because it had a truncated S gene.23 Thus, there is very little variation at rt233 and no obvious genotypic association. There is, however, at least one other example of primary adefovir resistance. Two of three cases of in vivo adefovir resistance that we discussed in a previous report7 involved HBV strains that had an rtL217R mutation before adefovir therapy. This mutation is present in most HBV strains with genotype A2, which are prevalent in Europe and the United States, but not in strains with other genotypes, including A1, which are found in Africa and Asia.20,24
Our observations demonstrate that some naturally occurring HBV strains are primarily resistant to adefovir. The rapid and strong effect of tenofovir in all three patients along with the in vitro data suggest that the patients' nonadherence to the medication was not the reason for the failure of adefovir therapy. Reactivation of HBV replication, as occurred in Patient 1, has also been reported in three patients by Van Bömmel and Berg,25 but sequencing data for HBV DNA have not been reported by these authors, and therefore it is not known whether these patients carried relevant HBV variants.
In agreement with a previous report,26 we found that the in vitro efficacy of tenofovir against wild-type HBV was similar to that of adefovir. Van Bömmel et al. reported that tenofovir was generally superior to adefovir in vivo,6 a finding that was most likely not caused by the presence of reverse-transcriptase variants. They ascribed this general superiority of tenofovir in vivo to the fact that the dose of the drug can be increased by a factor of 10, whereas the dose of adefovir is fixed.25 Since the dose of adefovir cannot be increased, the presence of naturally occurring HBV variants, as described here and in our earlier report,7 may decrease its efficacy.
Dr. Rockstroh reports having received consulting and lecture fees from Gilead and GlaxoSmithKline. The University of Bonn has submitted a patent application on a system for the detection of HBV resistance mutations. No other potential conflict of interest relevant to this article was reported.
Source Information
From the Institute of Medical Microbiology and Immunology and the Department of Medicine I, University of Bonn, Bonn (O.S., J.K.R.); the Heinrich-Pette-Institut, Hamburg (H.S., A.F., C.O., H.W.); the Institute of Medical Virology, University of Giessen, Giessen (U.C.W., W.R.W., W.H.G.); the Practice for Gastroenterology and Hepatology, Herne (H.H.); and the Practice Abelein/Helm, Nuremberg (M.H.) — all in Germany.
Drs. Schildgen and Sirma contributed equally to the article. Drs. Will and Gerlich contributed equally to the article.
Address reprint requests to Dr. Gerlich at the Institute of Medical Virology, Frankfurterstr. 107, D 35392 Giessen, Germany, or at wolfram.h.gerlich{at}viro.med.uni-giessen.de.
References
| |||||||||||||||||||||||||||||||||||||||||||||||
Related Letters:
Hepatitis B Virus with Primary Resistance to Adefovir
Chang T.-T., Lai C.-L., Schildgen O., Sirma H., Gerlich W.
Extract |
Full Text |
PDF
N Engl J Med 2006;
355:322-323, Jul 20, 2006.
Correspondence
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | TERMS OF USE | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2010 Massachusetts Medical Society. All rights reserved. |
Previous
