icon star paper   Hepatitis B Articles (HBV)  
Back grey_arrow_rt.gif
Entecavir Therapy Induces de Novo HIV Reverse-Transcriptase M184V Mutation in an Antiretroviral Therapy-Naive Patient BRIEF REPORT
  Clinical Infectious Diseases May 1 2008;46:e88-e91
Martin R. Jakobsen,1 Hanne Arildsen,1 Henrik B. Krarup,2 Martin Tolstrup,1 Lars Ostergaard,1 and Alex L. Laursen1
1Department of Infectious Diseases, Aarhus University Hospital, Skejby, Aarhus, and 2Department of Clinical Biochemistry, Arhus University Hospital, Aalborg Sygehus, Aalborg, Denmark
Recently, entecavir was introduced as a potent drug against hepatitis B virus infection. Initially, it was suggested not to have any effect on human immunodeficiency virus (HIV) infection. This guideline was revised in 2007 because of a report showing that the M184V mutation was selected in an hepatitis B virus and HIV-coinfected patient previously treated with lamivudine. Our investigation revealed findings similar to those preveiously reported but in an antiretroviral therapy-naive patient coinfected with HIV and hepatitis B virus. After 26 weeks of entecavir therapy, the M184V mutation dominated the plasma viral population. Thus, entecavir should only be used for coinfected patients who simultaneously receive suppressive therapy against HIV infection.
Received 2 October 2007; accepted 13 December 2007; electronically published 26 March 2008.
Hepatitis B virus (HBV) infection is a leading cause of chronic hepatitis, cirrhosis, and hepatocellular carcinoma and is a serious health problem worldwide, with an estimated 370 million persons chronically infected [1]. The primary aim in the treatment of chronic HBV infection is to reduce the risk of cirrhosis and hepatocellular carcinoma. Treatment with nucleoside analogues suppresses the HBV DNA to low levels, reducing inflammation and the risk of liver damage and hepatocellular carcinoma [2]. Two nucleoside analogues have been used-lamivudine and adefovir dipivoxil, both of which are well tolerated. Resistance against these drugs is, however, relatively easily developed. For lamivudine monotherapy, resistance is developed in 70% of persons after 5 years of treatment [3]. For adefovir, 29% of persons developed resistance during a 5-year follow-up period [4]. Entecavir is the latest drug approved for the treatment of chronic HBV infection. Compared with the other nucleoside analogues, development of entecavir resistance is significantly limited, possibly because of the higher potency of the drug [5, 6]. In addition, entecavir retains efficacy against lamivudine- and adefovir-resistant viruses [7, 8].
Entecavir treatment has been stated not to have any effect on HIV infection with the dosages used to treat chronic HBV infection [9]. Recently, however, McMahon et al. [10] showed that entecavir had an effect on HIV replication in vitro and selected the reverse-transcriptase M184V mutation in virus in 2 HIV-infected patients. Both patients had been treated with lamivudine; thus, the M184V mutant could possibly be a reintroduced mutant from archived viruses selected during prior treatment. As a consequence, the recommended guidelines were recently changed. In our study, we demonstrated that entecavir can induce the M184V mutation in an HIV-infected patient who is naive to antiretroviral therapy (ART), suggesting that entecavir itself may induce the development of HIV drug resistance in vivo.
Plasma samples, stored at -80°C, were analyzed retrospectively. The samples were obtained before the initiation of HBV treatment with adefovir, during treatment, before entecavir was added to the treatment regimen, and at subsequent intervals (figure 1). HIV RNA was extracted from plasma using the QIAamp viral RNA kit (QIAamp), according to the manufacturer's procedure. A segment of the pol gene was amplified in a 50-uL 1-step RT-PCR (SuperScript III-Taq HIFI polymerase; Invitrogen), as described elsewhere [11].
Analysis of PBMC samples was included for the last 2 time points. Genomic DNA was extracted from cells/mL with use of the QIAamp blood mini kit (QIAamp), and proviral DNA was PCR amplified using a method similar to that described above.
HBV DNA quantification was performed by an in-house method using molecular beacons on an Mx3005P Real-Time PCR System (Stratagene). All samples were analyzed in 2 independent extractions and runs. Total DNA was isolated from 200 μL of plasma with use of the QIAmp Virus BioRobot 9604 kit (Qiagen) according to the manufacturer's instruction and was eluted with 125 μL of water. Primers HBV-Ab and HBV-Bb were selected from the precore region, amplifying a product of 189 base pairs (position 1779-1967). Ten microliters of sample DNA or external standard was added to 20 μL of reaction mix containing 50 mmol/L tris hydrochloride (pH, 8.8), 75 mmol/L potassium chloride, 10 mmol/L magnesium chloride, 0.15% Tween 20, 250 μmol/L deoxyribonucleotide triphosphate, and 0.25 U Taq DNA polymerase (Bioline). PCR was performed using 45 cycles for 10 s at 95°C, 30 s at 57°C, and 30 s at 72°C. The limit of detection was 100 copies/mL (30 IU/mL).
All PCR samples were diluted to a concentration of about 105 viral copies and were used in an amplification refraction mutation system described elsewhere [11]. In brief, a nonspecific forward primer and a reverse primer, specific for either M184V wild type or mutant, were mixed with the PCR product, TaqMan probe, and Fast Start DNA Master Hybridization mix (Roche Diagnostics). Samples were analyzed on LightCycler, version 2 (Roche Diagnostic), and relative quantification of either M184V wild-type or mutant populations was performed.
We examined 35 clones, including an 874-nucleotide PCR-generated fragment of the HIV-1 envelope gene, from 4 different time points. Each fragment had been cloned into the TOPO-pCR2.1 plasmid using the TOPO vector system by Invitrogen, according to the manufacturer's protocol. Each purified clone was sequenced using BigDye, version 1.1 (Applied Biosystems), on an ABI3130. A neighbor-joining phylogenetic tree was made using PHYLIP, version 3.66. The robustness of the tree was evaluated by bootstrap analysis with 100 resamplings.
The patient was a 66-year-old white man with HIV-HBV coinfection. He tested positive for hepatitis B surface antigen and hepatitis B early antigen in 1985 and for HIV infection in 1994. From 1997 through 2006, the patient's CD4+ T cell count was high (mean CD4+ T cell count, 1180 cells/μL; range, 910-1660 cells/μL), and his plasma HIV load was stable (mean HIV load, 29,330 copies/mL; range, 6531-42,700 copies/mL) (figure 1). Thus, the patient did not meet the criteria for starting ART. A liver biopsy was performed in 1986 and revealed only minor activity related to chronic HBV infection. The patient declined a second liver biopsy before September 2002, when he developed cirrhosis. Adefovir therapy was initiated in March 2004, and regular measurement of serum α fetoprotein levels and ultrasound of the liver were used to monitor the patient for the development of hepatocellular carcinoma. The patient's HBV DNA level decreased by 2.5 log10 copies/mL to 390 x 10-3rd copies/mL over the next year. In late 2005, however, his HBV DNA level increased by 1.5 log10 copies/mL, and genome sequencing detected the adefovir mutation N236T. Because there was still no need for starting ART, it was decided to treat the HBV infection alone by adding entecavir to the adefovir regimen in November 2006. After this addition to the patient's treatment regimen, his HBV DNA level decreased to an undetectable level. Simultaneously, his plasma HIV viral load decreased from 9500 copies/mL in November 2005 to 5476 copies/mL and 3296 copies/mL in March and June 2007, respectively (figure 1).
Recent results [10] have suggested that the M184V mutation could be induced or reactivated by entecavir therapy. Therefore, a routine HIV genotype test (ViroSeq, version 2.0; Celera Diagnostics) was performed in April 2007. The test identified the M184V mutation in the HIV reverse-transcriptase gene. To determine the exact relationship between the development of the M184V mutation and entecavir treatment, stored plasma samples were analyzed for the presence of the M184V mutation before, during, and after the introduction of entecavir treatment, using a highly sensitive amplification refraction mutation system. This study showed that viruses with the M184V mutation rapidly emerged in plasma after entecavir therapy had been initiated in an ART-naive patient and dominated the virus population in plasma after just 6 months of therapy [12].
Before entecavir treatment was initiated, only wild-type HIV populations could be detected. After 21 and 26 weeks of entecavir treatment, the proportion of plasma HIV populations containing the M184V mutation increased to 93% and 100%, respectively (figure 2A). These results supported the results of genotyping performed in April 2007. Furthermore, analysis of PBMC samples from the last 2 points (weeks 26 and 35 after entecavir initiation) revealed that the percentage of proviral genomes carrying the M184V mutation increased from 32% to 92% (figure 2A).

We constructed a phylogenetic tree of envelope isolates cloned from 4 temporally spaced samples. Viral isolates from weeks 21 and 26 (with a prevalence of M184V mutations of 93% and 100%, respectively) were distinct from early pretreatment isolates (figure 2B), although they clustered intermittently with isolates tested at entecavir therapy initiation (figure 2B).
This study showed that the M184V mutation rapidly emerged in HIV in plasma after 5 months of entecavir therapy in an ART-naive patient and dominated the virus population in plasma after 6 months of treatment. The same development, with some delay, was seen in proviral DNA.
Our patient differed from those studied in detail by McMahon et al. [10]. In that study, the HIV M184V mutant was recovered from virus in serum samples from 2 patients, both of whom had been receiving ART including zidovudine and lamivudine for 1 or 5 years before entecavir therapy was initiated. Thus, in those patients, previously archived viruses with the M184V mutation were most likely reactivated.
For the patient described in our study, HIV infection including the M184V mutation seemed to be very unlikely, because he had never been exposed to ART and was infected years before lamivudine was introduced as a drug for treatment of HIV infection. Therefore, entecavir appears to have induced the M184V mutation de novo. The only other explanation is a reactivation of an acquired superinfection, and this was not supported by the phylogenetic relationship (figure 2B).
The phylogenetic analysis supports the observation that entecavir had induced the M184V mutation. There was a high degree of homology between the virus isolates at initiation and during entecavir treatment and a clear distinction from early pretreatment isolates. This suggests that viral populations with the M184V mutation evolved during the 6-month period after commencing entecavir therapy and not as a reselection of a latently archived species. Furthermore, we observed a rapid introduction of genomes carrying the M184V mutation in the proviral DNA pool, suggesting a selective pressure by entecavir therapy in vivo.
We noted a decrease in HIV RNA levels after entecavir therapy was initiated. This correlated with the occurrence of the M184V mutation. This decrease could have occurred because of either an inhibitory effect of entecavir on the replication capacity of HIV or a loss of fitness by the M184V virus mutant, as described elsewhere [13].
In conclusion, we have demonstrated the induction of the M184V mutation in an ART-naive patient infected with HIV who was treated with entecavir for HBV coinfection. The results indicate that the M184V mutation was introduced de novo. Thus, entecavir treatment should not be used to treat HBV infection in ART-naive, HIV-infected patients.
  icon paper stack View Older Articles   Back to Top   www.natap.org