Hepatitis B virus (HBV) infection is a leading cause of chronic liver disease worldwide [approximately 350 million people are hepatitis B surface antigen (HBsAg)-positive] and is responsible for approximately 80% of cases of hepatocellular carcinoma . Because HBV and HIV share similar modes of transmission and because immunodeficiency is a risk factor for chronic HBV infection, the prevalence of chronic HBsAg carriers is approximately 10% in HIV-infected patients . The advent of highly active antiretroviral therapy (HAART) has led to an increase in the contribution of chronic liver diseases to overall morbidity and mortality among such patients . HIV–HBV co-infection complicates patient management and might critically interfere with HBV viral kinetics, given the key role of the immune system in the clearance of free virions and infected cells .
Interferon alpha is the first effective treatment for patients with chronic hepatitis B , but had major side-effects and a low response rate, particularly among HIV-infected patients . Lamivudine, a potent antiretroviral drug, is the first nucleoside analogue effective against HBV reverse transcriptase to be marketed . Despite inducing an initial reduction in the HBV viral load and liver injury, the long-term use of lamivudine has been questionned by the rapid appearance of resistance mutations in the YMDD motif of the pol gene, especially in HIV-co-infected patients .
Tenofovir disoproxil fumarate (TDF) is a new nucleotide analogue licensed in 2001 for HIV infection . Its chemical structure and its remarkable activity on HBV replication in vitro, makes this drug a very attractive candidate for inclusion in HAART regimens of HIV–HBV-co-infected patients. In retrospective studies, TDF showed potent and sustained activity on the HBV viral load in HIV-infected patients , including those infected with lamivudine-resistant strains  and pre-core mutants . One 24-week and two 48-week prospective studies confirmed these results in a small number of patients [13–15], but none provided detailed dynamics of HBV decay in HIV–HBV-co-infected patients. Many factors, such as the degree of liver injury (as reflected by histological status and liver biochemistry), the co-administration of lamivudine or HBV characteristics (baseline viral load, genotype, hepatitis B e status and pol mutants) are suspected to influence antiviral efficacy in HBV-monoinfected patients, but their influence remains to be studied in HBV–HIV-co-infected patients.
We therefore conducted a prospective open-label study of HIV–HBV-co-infected patients starting a TDF-containing antiretroviral regimen, in order to assess the long-term HBV dynamics and the influence of baseline parameters, including HBV genetic variability, on HBV dynamics.
Patients and data collection
From May 2002 to May 2003, 326 HBV–HIV-co-infected patients were enrolled in the French multicentre prospective HIV–HBV Cohort Study designed to identify risk factors for liver fibrosis progression . All patients gave their written consent to participate in the cohort, and the study received ethical approval in accordance with the Helsinki declaration. A subgroup of patients, all monitored in the same HIV clinic (Saint-Antoine University Hospital, Paris) and starting a new antiretroviral regimen including TDF 300 mg a day, were included for the present study. Inclusion criteria were HIV enzyme-linked immunosorbent assay (ELISA) positivity confirmed by a complete Western blot pattern, HBsAg seropositivity, an available pre-TDF serum sample, and a baseline HBV-DNA load of at least 1000 copies/ml.
Demographic and clinical characteristics were collected at baseline and during follow-up. The Child–Pugh score was determined in cirrhotic patients at the beginning and end of follow-up, in order to evaluate the prevalence and incidence of decompensated cirrhosis during the study. At each HBV load assay, biochemical data (liver enzymes, creatinemia, calcemia and phosphatemia) were determined using standard methods. The histological status (METAVIR score)  was obtained from the HIV–HBV Cohort Study database. The HIV immunovirological status was determined prospectively: HIV viral load was measured using the branched DNA technique (bDNA Quantiplex 3.0; Bayer Diagnostics, Cergy Pontoise, France; detection limit 50 copies/ml) and CD4 and CD8 cells were counted by flow cytometry. Markers of hepatitis C virus (HCV) and hepatitis delta virus (HDV) were tested by using commercial ELISA techniques as recommended by the manufacturers.
Hepatitis B virus marker assays
HBsAg and hepatitis B e antigen (HBeAg) and antibodies were detected using commercial ELISA tests, as recommended by the manufacturer (Dia Sorin, Antony, France). The HBV load in serum was measured at least once during the first month, then at least once every 3 months using a commercial quantitative polymerase chain reaction (PCR) assay (PCR-Amplicor; Roche Diagnostic Systems, Meylan, France; detection limit 200 copies/ml). If the viral load was above the working range of the assay, the sample was diluted and retested. DNA sequencing or DNA chip technology (bioMerieux, Marcy l’Etoile, France) were used to detect HBV-DNA polymerase gene mutations and for genotyping.
Frequencies, means or medians of variables at inclusion were expressed with a 95% confidence interval or standard error and range, as appropriate. Pearson χ2 tests and paired-sample T tests were used as appropriate to compare changes in the different parameters during follow-up.
The time to HBV undetectability and the factors influencing it (concomitant treatment with lamivudine, initial HBV viral load, degree of immunodeficiency, HBV genotype, METAVIR fibrosis score, preC and pol gene mutations) were analysed using the Kaplan–Meier method and the Cox proportional hazards model. We initially considered the results of the METAVIR score as a variable with three categories: F0–F1, F2 and F3–F4. However, the results of preliminary models showed that the threshold was just before F2: data of F2 patients were similar to data of F3–F4 patients (P = 0.02) and different to data of F0–F1 patients (P = 0.3). That is why we presented the general result as ‘METAVIR score ≥ F2’ in Table 1.
Because the event is interval-censored, we imputed its date to the middle of the censoring window. A check using imputation of either the start or the end of the censoring window showed that the results of the proportional hazards regression were not affected. Significance was assumed at P < 0.05.
Linear mixed models were adjusted to the time course of the HBV viral load decline from its baseline value, using the following equation:
where CVi(t) is the viral load of the ith subject at time t adjusted on the value of HBV viral load at baseline, (.)+ denotes the positive part, b1i and b2i are zero-centered random effects, and e is gaussian. With this parametrization, fixed effect b1 corresponded to the slope during the first 100 days, and b1 + b2 to the slope thereafter. Fixed effect b1 could depend on a baseline characteristic, allowing us to investigate the determinants of the initial decrease (these factors are the same as those analysed in the study of the time to HBV undetectability and are listed above). We took into account the fact that samples below the detection level were censored, but the results were unaffected when censored measures were attributed the value of the detection limit.
Analyses were carried out using the following software: SPSS release 11.5.0, R (version 1.9.1) and WinBUGS (version 1.4).
From April 2002 to May 2004, 28 HBV–HIV-co-infected patients with a median age of 41.2 years were enrolled and prospectively studied for a median duration of 517 days (SD 168, range 147–802 days) after starting TDF therapy (Table 2). The median estimated durations of HIV and HBV infections were 11.3 and 7.2 years, respectively. Immunodeficiency was mild, with a median CD4 lymphocyte count of 440 cells/mm3 and a median HIV load of 3.81 log copies/ml.
Previous exposure to antiretroviral drugs included protease inhibitors (PI) in 71% of patients, nucleoside reverse transcriptase inhibitors (NRTI) in 93% and non-nucleoside reverse transcriptase inhibitors (NNRTI) in 68%. Twenty-four patients had previously received lamivudine, for a mean of 55 months (SD 19.6 months, range 1.5–87.4 months). TDF was combined with PI (two patients), NRTI (two patients), NNRTI (two patients), PI plus NNRTI (two patients), PI plus NRTI plus NNRTI (one patient), NRTI plus NNRTI (nine patients) or PI plus NRTI (10 patients) at treatment initiation. Sixteen patients received the lamivudine–TDF combination in the antiretroviral regimen. Figure 1 shows the distribution of patients with former treatment or co-treatment with lamivudine at inclusion. Patients on TDF or TDF–lamivudine did not differ in terms of their baseline characteristics.
Twenty-four patients were infected with an HBeAg-positive HBV strain, whereas the other four patients were HBeAg negative at baseline, despite ongoing viral replication. The mean HBV load at baseline differed according to the hepatitis B e status: 5.45 log copies/ml (SD 1.66) in hepatitis B e-negative patients versus 7.82 log copies/ml (SD 1.90) in hepatitis B e-positive patients (P = 0.02). Co-infection by HCV was noted in four patients (among whom two presented with replicative HCV hepatitis) and one patient had serological markers for HDV but undetectable HDV DNA. Six patients presented at baseline with an elevated rate of alanine aminotransferase (ALT; > 3 N, i.e. 95 IU). HBV was genotyped in 23 patients, showing a predominance of genotype A (n = 16), followed by genotypes D (n = 3), G (n = 3), and E (n = 1). The pol gene was sequenced in 24 patients previously exposed to lamivudine, showing a lamivudine resistance mutation in 18 cases (M204I/V in six patients, L180 plus M204I/V in seven and V173 plus L180 plus M204I/V in five).
The results of a liver biopsy performed within 18 months of inclusion were available for 17 patients. The METAVIR scores were as follows: fibrosis: F0 = 2, F1 = 2, F2 = 7, F3 = 3, F4 = 3; activity: A0 = 3, A1 = 6, A2 = 3, A3 = 5. None of the patients had decompensated cirrhosis at inclusion (Child–Pugh stage B in two patients).
Changes in biological and virological parameters
The mean duration of TDF treatment was 71 weeks (SD 168 days, range 177–802 days). During this period, a total of 152 HBV tests were performed (i.e. five to eight tests per patient). The mean reduction in HIV-1 viral load was 1.4 log10 copies/ml (3.4 versus 2.0, P < 0.0001), whereas the proportion of patients with undetectable HIV-1 viral loads (< 50 copies/ml) increased from 22% at baseline to 75% at the end of the study. The mean CD4 cell count increased from 414 to 438 cells/mm3 (not significant).
The HBV viral load declined by a mean of 4.6 log10 copies/ml on TDF-containing therapy (7.5 versus 2.9, P < 0.001), and fell below the detection limit in 21 patients. This was accompanied by a significant decrease in ALT activity (mean 125 versus 50 UI, P < 0.05). Creatinemia increased slightly (84.9 versus 78.8 mmol/l, P < 0.01), but no case of severe renal impairment occurred. Other biochemical markers, including bilirubinemia, γ-glutamyl transferase, calcemia and phosphatemia, did not change. Overall, TDF was well tolerated with no decompensated cirrhosis during follow-up (one patient became Child–Pugh C but was also hepatitis C infected). A combination of lamivudine with TDF (16/28 patients) did not influence the clinical or biological outcome.
At the end of follow-up, four patients had lost HBeAg and seroconverted to hepatitis B e antibodies, that is 16.7% of the 24 patients who were HBeAg positive at baseline. HBV DNA became undetectable in all four patients who were HBeAg negative at baseline.
Hepatitis B virus DNA kinetics
Time to undetectability (hepatitis B virus DNA < 200 copies/ml)
The median time taken for HBV DNA to become undetectable in serum was 272.5 days [95% confidence interval (CI) 203.5, 416.0; Fig. 2). This time increased with the magnitude of HBV-DNA load at the onset of TDF treatment: it was 150 days in patients with less than 108 log HBV copies/ml, and 316 days in those with viral loads above this value; this corresponds to a hazards ratio of 0.6 (95% CI 0.47, 0.78) per increase of one log of HBV load at baseline. Other factors associated with this endpoint in univariate analysis were HIV infection lasting more than 10 years, ALT levels above 50 UI and HBeAg seropositivity. HCV co-infection did not influence the time to undetectability. After adjustment for HBV viral load, only HBeAg positivity at baseline remained associated with a longer time to reach HBV-DNA undetectability (hazard ratio 0.21, 95% CI 0.05–0.95).
Hepatitis B viral load decline
Mathematical linear mixed models were adjusted to the observed values in order to model the effect of TDF on the HBV viral load decline over time. A model with two slopes provided a better fit than a model with one slope (P < 0.001). A slope change after the 100th day of treatment yielded a satisfactory fit with observed data (Fig. 3a). Overall, during the first 100 days, the decline was 0.040 ± 0.004 log/day, corresponding to a 1 log decrease per 25 days (95% CI 20, 30). After this initial phase, the slope flattened out to only 0.003 ± 0.005 log/day (Fig. 3b). The slope of the initial phase increased with the initial HBV viral load, by 0.002 ± 0.0001 log/day per increase of one log unit (P < 0.0001). Therefore, a subject with a baseline viral load of 105 copies per ml would lose one log unit in 29 days, compared with only 23 days for a subject with a baseline viral load of 1010 copies per ml.
The initial rate of decline was also slightly more pronounced in patients with baseline ALT values higher than 50 UI (0.003 ± 0.001 log/day, P = 0.0048), HBeAg positivity (0.004 ± 0.001 log/day, P = 0.001), and a viral mutation in the YMDD motif (0.003 ± 0.001 log/day, P = 0.012). The initial rate of decline was slower in patients with anti-hepatitis B e antibodies (0.003 ± 0.001 log/day, P = 0.0013), and in patients with a METAVIR fibrosis score greater than 2 at baseline (0.004 ± 0.001 log/day P = 0.0004). Each of these characteristics led to a difference of 1 or 2 days in the time required for the viral load to decline by 1 log unit. Concomitant lamivudine treatment did not affect the initial rate of decline (0.001 ± 0.001 log/day, P = 0.41), neither did long-standing HIV infection (0.002 ± 0.001 log/day, P = 0.07), nor the METAVIR activity score (0.001 ± 0.001 log/day, P = 0.46) or the presence of an HCV co-infection (0.0001 ± 0.001 log/day, P = 0.51).
TDF is a recently licensed nucleotide analogue of HIV reverse transcriptase that demonstrated a potent activity against HBV in HIV–HBV-co-infected individuals [13,14]. The prospective study presented herein included 28 co-infected patients starting a HAART regimen including TDF, and is the first study to analyse the long-term dynamics of HBV in HIV–HBV-co-infected patients treated with TDF.
Most of the patients studied had been exposed to lamivudine, and the HBV strain was often resistant to this drug. The mean decline in HBV viral load was 4.6 log copies/ml after 71 weeks of TDF therapy, in keeping with the results of a recent clinical trial (−4.5 log at 48 weeks in pretreated patients) . HBV-DNA PCR became undetectable in 75% of patients (including all the HBeAg minus patients), and 16.7% of patients who were HBeAg-positive patients seroconverted to hepatitis B e antibodies. These results confirm, with longer follow-up, the activity of TDF on HBV in HIV–HBV co-infection, even in patients with lamivudine-resistant or presumed pre-C mutant strains. This antiviral activity also had a marked impact on liver function, as ALT activity fell significantly in all 28 patients studied.
Concerning TDF tolerability, no grade III or more adverse events occurred. Creatinemia increased slightly in the treated patients but none developed severe renal impairment. This is in keeping with the results of previous clinical trials , although cases of Fanconi syndrome in patients regularly followed in HIV clinics have been published [18,19].
Hepatitis B virus dynamics
We modeled HBV-DNA decay based on serial samples during a mean treatment period of 71 weeks, thus bringing new data on long-term HBV dynamics in HIV–HBV patients treated with TDF. The short-term kinetics of HBV decay have previously been studied only in HBV-monoinfected patients treated with lamivudine [20,21], lamivudine plus famciclovir , entecavir  or adefovir . Most of the studies showed an early biphasic pattern of HBV clearance, reflecting the clearance of free virions followed by infected cells. In our study, long-term HBV clearance was also biphasic, mimicking that of the early HBV kinetics in HBV-monoinfected patients. This long-term biphasic clearance has been also observed with HIV .
Importantly, long-term HBV dynamics was not influenced by baseline parameters such as co-treatment with lamivudine, genomic characteristics of HBV or immunosuppression caused by HIV.
Concomitant lamivudine therapy did not influence the time taken for HBV DNA to become undetectable in our TDF-treated patients. Lamivudine–TDF combination therapy has previously been compared with lamivudine monotherapy in terms of viral load in treatment-naive patients, with respective declines of 4.7 and 3.0 log copies/ml after 48 weeks (P = 0.055) . The large proportion of patients with lamivudine resistance mutations or previous lamivudine exposure in our study population may explain why combination therapy was not more potent than TDF monotherapy. Surprisingly, the presence of lamivudine resistance mutations was significantly associated with a steeper initial decline in the HBV load during TDF therapy. However, this paradoxical effect has already been observed in a similar setting of HIV–HBV co-infection . One may suggest that the reduced fitness of lamivudine-resistant mutants may increase their sensitivity to other DNA polymerase inhibitors, even if in-vitro studies suggest a similar activity of TDF on wild-type and lamivudine-resistant isolates [26,27]. Nevertheless, YMDD mutations probably do not clinically impact the long-term activity of TDF because the time to undetectability was not influenced by the presence of these mutations.
Baseline HBV genomics characteristics such as the genotype did not influenced the HBV decay either. Similar results reporting the lack of influence of HBV genotypes on antiviral response were noted with lamivudine  and adefovir dipivoxil , but data are more contradictory regarding the antiviral reponse to interferon. A meta-analysis of 10 controlled clinical trials showed that the therapeutic response to interferon was independent of baseline parameters, including the HBV genotype , whereas one study  reported a poorer response to interferon in HBV-monoinfected patients with genotype C.
In our study, the magnitude of immunosuppression conferred by HIV infection did not significantly influence the response to TDF. However, the low level of immunosuppression present in our study population might account for this negative result, and we cannot draw a definitive conclusion on the real impact of a low CD4 cell count on HBV viral dynamics.
The factor most strongly associated with a longer time to undetectability is the initial HBV viral load. This result might have a clinical impact because a twofold difference in initial HBV load leads to a difference of one week in the loss of one log of viral load. We also found that the slope of the first decay phase was steeper when the baseline HBV viral load was high. The explanation for this unexpected result probably relies on intrahepatic events such as the kinetics of the death of infected hepatocytes or the efficacy of local immune responses. An analysis of the intrahepatic HBV parameters would have given a better insight into this prospect, as it has been shown that cccDNA levels are strongly correlated with levels of total intracellular HBV DNA, and serum HBV DNA in patients treated with adefovir .
The hepatitis B e status also had an influence, although less marked, on HBV decay. The time to reach an HBV-DNA level of less than 200 copies/ml was significantly shorter in hepatitis B e-negative patients, and this independently of the baseline HBV load. A previous study of five HIV–HBV-co-infected patients harbouring preC mutant strains also showed the efficacy of TDF in this setting, but did not offer a comparison with HBeAg-positive patients . However, the clinical implication of such data is weak because the time difference in reaching the undetectability level as measured in our study was only of 1 or 2 days between HBeAg-positive and HBeAg-negative patients.
Elevated ALT activity at baseline was also associated with a more rapid HBV decay, whereas a METAVIR fibrosis score above 2 was associated with slower HBV decay. This suggests that TDF is more efficient in patients with active inflammatory disease, as previously shown in HBV-monoinfected individuals treated with lamivudine or adefovir [7,32]. Again, the time difference to reach undetectability according to initial levels of ALT or liver fibrosis was only of 1 or 2 days, thus without true clinical impact on the response to TDF.
Overall, in this prospective study of 28 pretreated HIV–HBV-co-infected patients, TDF had a potent and durable effect on HBV replication in all patients despite the viral diversity of the HBV strains. Larger prospective studies are needed to determine the precise impact of HIV-induced immunodeficiency and HBV genomic variability on the response to TDF.
The authors are grateful to V. Massari, F. Carrat and G. Pannetier for their help in the HIV–HBV Cohort data management, to Dr L. Morand-Joubert for her technical advice on HIV, to S. Charpentier and Y.J. Beyer for their participation to the data collection.
Sponsorship: This study has been sponsored by the institut de Médecin et d'Epidémiologie Appliquée (IMEA) and supported by grants from Ensemble contre le Sida (SIDACTION) and the Agence Nationale de Recherche sur le Sida (ANRS). It also received an additional grant from Gilead Sciences.
1. Lok AS. Chronic hepatitis B. N Engl J Med 2002; 346:1682–1683.
2. Ockenga J, Tillman H, Trautwein C, Stoll M, Manns M, Schmidt R. Hepatitis B and C in HIV-infected patients. Prevalence and prognostic value. J Hepatol 1997; 27:18–24.
3. Thio CL, Seaberg EC, Skolasky R Jr, Phair J, Visscher B, Munoz A, Thomas DL, Multicenter AIDS Cohort Study. HIV-1, hepatitis B virus, and risk of liver-related mortality in the Multicenter Cohort Study (MACS). Lancet 2002; 360:1921–1926.
4. Ganem D, Prince AM. Hepatitis B virus infection – natural history and clinical consequences. N Engl J Med 2004; 350:1118–1129.
5. Krogsgaard K, Bindslev N, Christensen E, Craxi A, Schlichting P, Schalm S, et al
. The treatment effect of alpha interferon in chronic hepatitis B is independent of pre-treatment variables. Results based on individual patient data from 10 clinical controlled trials. European Concerted Action on Viral Hepatitis (Eurohep). J Hepatol 1994; 21:646–655.
6. Di Martino V, Thevenot T, Colin J, Boyer N, Martinot M, Degos F, et al
. Influence of HIV infection on the response to interferon therapy and the long-term outcome of chronic hepatitis B. Gastroenterology 2002; 123:1812–1822.
7. Lai CL, Chien RN, Leung NW, Chang TT, Guan R, Tai DI, et al
. A one-year trial of lamivudine for chronic hepatitis B. Asia Hepatitis Lamivudine Study Group. N Engl J Med 1998; 339:61–68.
8. Benhamou Y, Bochet M, Thibault V, Di Martino V, Caumes E, Bricaire F, et al
. Long-term incidence of hepatitis B virus resistance to lamivudine in human immunodeficiency virus-infected patients. Hepatology 1999; 30:1302–1306.
9. Schooley RT, Ruane P, Myers RA, Beall G, Lampiris H, Berger D, et al
. Tenofovir DF in antiretroviral-experienced patients: results from a 48-week, randomized, double-blind study. AIDS 2002; 16:1257–1263.
10. Piketty C, Pellegrin I, Katlama C, Rozenbaum W, Neau D, Le Teuff G, et al. Efficacy of tenofovir disoproxil fumarate (TDF) in hepatitis B virus in HIV-1 coinfected patients: the TECOVIR study.
In: 11th Conference on Retroviruses and Opportunistic Infections
. San Francisco, USA, 8–11 February 2004 [Poster 834].
11. Nunez M, Perez-Olmeda M, Diaz B, Rios P, Gonzalez-Lahoz J, Soriano V. Activity of tenofovir on hepatitis B virus replication in HIV-coinfected patients failing or partially responding to lamivudine. AIDS 2002; 16:2352–2354.
12. Bruno R, Sacchi P, Zochetti C, Ciappina V, Puoti M, Filice G. Rapid hepatitis B virus-DNA decay in coinfected HIV-hepatitis B virus ‘e-minus’ patients with YMDD mutations after 4 weeks of tenofovir therapy. AIDS 2003; 17:783–784.
13. Benhamou Y, Tubiana R, Thibault V. Tenofovir disoproxil fumarate in patients with HIV and lamivudine-resistant hepatitis B virus. N Engl J Med 2003; 348:177–178.
14. Nelson M, Portsmouth S, Stebbing J, Atkins M, Barr A, Matthews G, et al
. An open-label study of tenofovir in HIV-1 and hepatitis B virus coinfected individuals. AIDS 2003; 17:F7–F10.
15. Dore GJ, Cooper DA, Pozniak LP, DeJesus E, Zhong L, Miller MD, et al
. Efficacy of tenofovir disoproxil fumarate in antiretroviral therapy-naive and -experienced patients coinfected with HIV-1 an hepatitis B virus. J Infect Dis 2004; 189:1185–1192.
16. Lacombe K, Massari V, Serfaty L, Faure-Chounina V, Gozlan J, Massot O, et al
. HIV–HBV coinfection a prospective cohorte study for modelling fibrosis progression: data at inclusion.
In: XVth International AIDS Conference
. Bangkok, Thailand, 11–16 July 2004 [Poster B3325].
17. Bedossa P, Poynard T. An algorithm for the grading of activity in chronic hepatitis C. The METAVIR Cooperative Study Group. Hepatology 1996; 24:289–293.
18. Verhelst D, Monge M, Meynard JL, Fouqueray B, Mougenot B, Girard PM, et al
. Fanconi syndrome and renal failure induced by tenofovir: a first case report. Am J Kidney Dis 2002; 40:1331–1333.
19. Peyriere H, Reynes J, Rouanet I, Daniel N, de Boever CM, Mauboussin JM, et al
. Renal tubular dysfunction associated with tenofovir therapy: report of 7 cases. J Acquir Immune Defic Syndr 2004; 35:269–273.
20. Lewin SR, Ribeiro RM, Walters T, Lau GK, Bowden S, Locarnini S, et al
. Analysis of hepatitis B viral load decline under potent therapy: complex decay profiles observed. Hepatology 2001; 34:1012–1020.
21. Herz AVM, Bonhoefer S, Anderson RM, May RM, Nowak MA. Viral dynamics in vivo: Limitations on estimates of intracellular delay and virus decay. Proc Natl Acad Sci U S A 1996; 93:7247–7251.
22. Lau GK, Tsiang M, Hou J, Yuen S, Carman WF, Zhang L, et al
. Combination therapy with lamivudine and famciclovir for chronic hepatitis B-infected Chinese patients: a viral dynamics study. Hepatology 2000; 32:394–399.
23. Wolters LM, Hansen BE, Niesters HG, DeHertogh D, de Man RA. Viral dynamics during and after entecavir therapy in patients with chronic hepatitis B. J Hepatol 2002; 37:137–144.
24. Tsiang M, Rooney JF, Toole JJ, Gibbs CS. Biphasic clearance kinetics of hepatitis B virus from patients during adefovir dipivoxil therapy. Hepatology 1999; 29:1863–1869.
25. Wu H, Zhang JT. The study of long-term HIV dynamics using semi-parametric non linear mixed-effects models. Statist Med 2002; 21:3655–3675.
26. Delaney WE IV, Yang H, Miller MD, Gibbs CS, Xiong S. Combinations of adefovir with nucleoside analogs produce additive antiviral effects against hepatitis B virus in vitro
. Antimicrob Agents Chemother 2004; 48:3702–3710.
27. Ying C, De Clercq E, Nicholson W, Furman P, Neyts J. Inhibition of the replication of the DNA polymerase M550V mutation variant of human hepatitis B virus by adefovir, tenofovir, L-FMAU, DAPD, penciclovir and lobucavir. J Viral Hepatitis 2000; 7:161–165.
28. Nafa S, Ahmed S, Tavan D, Pichoud C, Berby F, Stuyver L, et al
. Early detection of viral resistance by determination of hepatitis B virus polymerase mutations in patients treated by lamivudine for chronic hepatitis B. Hepatology 2000; 32:1078–1088.
29. Westland C, Delaney W IV, Yang H, Chen SS, Marcellin P, Hadziyannis S, et al
. Hepatitis B virus genotypes and virologic response in 694 patients in phase III studies of adefovir dipivoxil. Gastroenterology 2003; 125:107–116.
30. Wai CT, Chu CJ, Hussain M, Lok AS. HBV genotype B is associated with better response to interferon therapy in HBeAg(+) chronic hepatitis than genotype C. Hepatology 2002; 36:1425–1430.
31. Werle-Lapostolle B, Bowden S, Locarnini S, Wursthorn K, Petersen J, Lau G, et al
. Persistence of cccDNA during the natural history of chronic hepatitis B and decline during adefovir dipivoxil therapy. Gastroenterology 2004; 126:1750–1758.
32. Marcellin P, Chang TT, Lim SG, Tong MJ, Sievert W, Shiffman ML, et al
. Adefovir dipivoxil for the treatment of hepatitis B e antigen-positive chronic hepatitis B. N Engl J Med 2003; 348:808–816.