Early monitoring of ribavirin serum concentration is not useful to optimize hepatitis C virus treatment in HIV-coinfected patients

Transcription

1 Antiviral Therapy 12: Early monitoring of ribavirin serum concentration is not useful to optimize hepatitis C virus treatment in HIV-coinfected patients Manuel Crespo 1 *, Leonor Pou 2, Juan I Esteban 3, Vicenç Falcó 1, Esteban Ribera 1, Rosa Lopez 4, Silvia Sauleda 5, Adriá Curran 1, Sara Villar del Saz 1, María Feijoo 1, Inma Ocaña 1 and Albert Pahissa 1 1 Infectious Diseases Department, Hospital Universitari Vall d Hebron, Universitat Autònoma de Barcelona, Spain 2 Clinical Biochemistry Department, Hospital Universitari Vall d Hebron, Universitat Autònoma de Barcelona, Spain 3 Liver Unit, Department of Medicine, Hospital Universitari Vall d Hebron, Universitat Autònoma de Barcelona, Spain 4 Pharmacy Department, Hospital Universitari Vall d Hebron, Spain 5 Centre de Transfusió i Banc de Teixits, Institut Catala de la Salut, Spain *Corresponding author: Tel: ; Fax: ; mcrespo@vhebron.net Background: Emerging data suggest that higher ribavirin (RBV) exposure could improve early hepatitis C virus (HCV) response. Furthermore, interindividual RBV bioavailability shows high variation, and dose-limiting haemolytic anaemia is a common adverse event. Therefore, it has been suggested that monitoring RBV serum levels could be used to drive dose modification and to optimize management of HCV-infected patients receiving combination treatment. Methods: To assess the effect of RBV serum levels on HCV RNA clearance at week 4 and 12 of treatment, and to determine the correlation between RBV serum concentration and haemoglobin decrease, RBV trough levels were measured by HPLC in stored serum samples obtained from 94 HCV HIV-coinfected patients at week 4 and 12 of treatment with peginterferon-α2b (1.5 μg/kg/weekly) plus ribavirin (800 1,200 mg/day). Results: The median RBV levels increased from 1.70 μg/ml at week 4 to 1.97 μg/ml at week 12 of treatment (P=0.001) and were independently predicted by weight-adjusted dose of RBV and co-administration of tenofovir. Haemoglobin drop was higher among patients who received zidovudine and weakly correlated with RBV level. Although RBV concentration was lower in genotype 1 or 4 HCV-infected patients who cleared the virus at treatment week 4, the ability of this parameter to discriminate between responders and non-responders at treatment week 4 and 12 was poor. Conclusion: Intracellular RBV accumulation early in treatment might improve the kinetics of HCV response in difficult to treat patients. Although this hypothesis and the potential interaction between RBV and tenofovir warrant further research, our data do not support RBV serum monitoring as a tool to optimize treatment in HCV HIV-coinfected patients. Introduction Monitoring early hepatitis C virus (HCV) response allows discontinuation of treatment in HCVmonoinfected patients showing serum HCV RNA decreases <2 log 10 at 12 weeks of treatment as compared to baseline [1]. This stopping rule is also useful for identifying non-response among HCV HIVcoinfected patients [2 5]. Emerging data have shown that rapid viral response (for example, HCV RNA undetectable at 4 weeks) is the strongest predictor of sustained viral response, both in HCV-monoinfected [1,6 8] and HCV HIV-coinfected patients [5,9]. The efficacy of ribavirin (RBV) to reduce relapses in patients receiving anti-hcv therapy has been clearly proven [10 12]. Nevertheless, data supporting the role of RBV together with interferon during the first phases of HCV clearance are still controversial [13 15] and the mechanism of action of this agent remains largely unknown [16]. Recent findings in HCV HIV-coinfected patients treated with pegylated interferon-α plus RBV suggest that a high RBV plasma concentration could improve early viral response rates [17], yet the data are discordant across different HCV genotypes [18]. The optimal dose of RBV is unknown and lower rates of sustained virological response observed among coinfected patients in randomized clinical trials [2 4] could have been related to RBV underdosing. Furthermore, RBV displays great interindividual variability [17,19,20] and RBV-related anaemia often 2007 International Medical Press

2 M Crespo et al. occurs rapidly during the first 4 weeks of therapy [5], leading to dose reductions at a time when it might be crucial to maintain RBV exposure to maximize the chance of achieving a sustained viral response [1]. Hence, monitoring RBV levels early in treatment could provide a helpful basis for dosing adjustments to improve efficacy and prevent severe adverse events. The main aim of this study was to assess the effect of serum RBV levels on viral response at week 4 and week 12 of treatment across different HCV genotypes. In addition, the predictive factors for RBV serum concentration at week 4 and 12 of treatment and the correlation between RBV concentration and haemoglobin decrease at these time points were analysed. Materials and methods Patients All HCV HIV-coinfected patients consecutively treated in our hospital with pegylated interferon-α2b (peg- INF-α2b) plus RBV were included in the analysis, provided that serum samples recovered at week 4 and week 12 of treatment were available from the storage facilities. All the patients included had been enrolled in one of two trials: a randomized trial comparing the efficacy and safety of peg-inf-α2b (1.5 μg/kg/week) versus standard interferon-α2b, both in combination with a flat dose of RBV (800 mg/day) [5]; or an ongoing exploratory study aimed at individualizing the duration of therapy according to the viral response at treatment week 4 and 12, in which patients received peg-inf-α2b (1.5 μg/kg/week) in combination with a weight-adjusted dose of RBV (800 1,200 mg/day). In this latter study, patients with HCV RNA <50 IU/ml at week 4 were scheduled to receive 24 weeks of treatment, whereas those showing HCV RNA clearance between treatment weeks 4 and 12 and between treatment weeks 12 and 24 completed 48 weeks and 60 weeks of treatment, respectively. Patients showing no response, defined as an HCV RNA decrease <2 log 10 at treatment week 12 or detectable HCV RNA at week 24, were considered non-responders and stopped treatment. The main eligibility criteria were chronic HCV infection, HCV-treatment-naive, increased alanine aminotransferase (ALT) levels, no cirrhosis decompensation, and stable HIV infection with CD4 + T-cell counts 200 cells/mm 3. Monitoring Monitoring of HCV RNA levels at treatment weeks 4 and 12 was performed with a commercially available real-time PCR test combined with an automated platform for RNA extraction (COBAS AmpliPrep-COBAS TaqMan 48; Roche Molecular Diagnostic, Pleasanton, CA, USA) with a detection limit of 15 IU/ml and a dynamic range of 43 to 68,000,000 IU/ml. Undetectable HCV RNA levels at treatment week 4 and 12 were defined as rapid viral response and early viral response, respectively. RBV serum concentration was analysed by high-performance liquid chromatography (HPLC) using a modification of a previously validated method [21]. The solid-phase extraction was made on phenyl boronic acid cartridges and the chromatographic separation on C18-bonded silica column (Symmetry 300 TM mm, particle size of 3.5 μm). The assay was calibrated by using nonzero calibrators containing 0.25, 0.5, 1, 2.5 and 5 μg/ml of RBV, and 3 methylcytidine (2.5 μg/ml) as internal standard (IS). The retention times of RBV and IS were 3.28 and 4.65 min, respectively. The ratio of the peak area of the drug to that of the IS was plotted against the drug concentration. The limit of quantification (0.25 μg/ml), limit of detection (0.10 μg/ml), recovery rate (88 ±3%), range of linearity of the calibration curve ( μg/ml), and intra- and interassay precision (5.8% and 9.3%, respectively) of the present method are in keeping with the values recommended for validation of bioanalytical methods [22]. Blood samples were collected before the first daily dose of RBV in Vacutainer tubes without additives. Following centrifugation, serum was stored at -80 C until analysis. To assess RBV stability in samples stored at -80 C, 15 paired frozen samples were tested a median of 6 months apart. Baseline creatinine clearance was estimated using the Cockroft Gault equation [23]. Statistical analysis Student s t-test or Wilcoxon signed rank test were used to compare continuous data and the Fisher exact test was used to compare categoric data. A multivariate linear regression analysis with a backward stepwise approach was used to assess RBV serum concentration at week 4 and week 12 of treatment. The initial analysis included all baseline covariables and the final model retained only those variables that were statistically significant. A similar approach was used to assess haemoglobin decline from baseline, including RBV concentration at the two time points as a covariable. Potential predictive factors for viral response at week 4 and 12 of treatment were assessed by univariate and multivariate logistic regression analysis. Correlations were studied with Spearman s non-parametric correlation coefficient. Receiver operating characteristic (ROC) curves were constructed with RBV levels at week 4 and 12 of treatment to determine the cut-off point that best discriminated between virological responders and non-responders at these time points. All statistical analyses were performed using SPSS for Windows, version 12.0 (SPSS, IL, USA) International Medical Press

3 Independent predictors of ribavirin serum concentration Results From January 2001 to June 2006, 94 HCV HIVcoinfected patients, from a total of 110 who started treatment with peg-inf-α2b plus RBV during this period, had available serum samples obtained at week 4 and 12 of treatment. Fifty-three (56.4%) and 41 (43.6%) patients included in the analysis had been enrolled in the above-mentioned randomized trial [5] and in the exploratory study, respectively. Baseline characteristics of the patients are summarized in Table 1. Most were male, former intravenous drug users; almost 90% were on antiretroviral treatment and 85% had HIV RNA <50 copies/ml. Weightadjusted dose of RBV was 13.5 mg/kg (interquartile range [IQR] mg/kg). Although RBV dose was reduced because of anaemia (n=6) or hyperlactataemia (n=1) in 7 (7.4%) patients during the first 12 weeks of treatment, median RBV concentration increased from 1.7 μg/ml (IQR μg/ml) at week 4 to 1.97 μg/ml (IQR μg/ml; P=0.001) Table 1. Baseline characteristics of 94 patients included in the study Characteristic* Age, years (IQR) 39.7 ( ) Male, n (%) 67 (71) Intravenous drug use, n (%) 74 (78.7) BMI, kg/m 2 (IQR) 22.9 ( ) Creatinine clearance, ml/min (IQR) ( ) Serum HCV RNA Median, log 10 IU/ml (IQR) 6.3 ( ) 800,000 IU/ml (%) 71 (75.5) HCV genotype 1, n (%) 41 (43.6) 3, n (%) 32 (34.0) 4, n (%) 21 (22.3) Liver cirrhosis (F5 F6) 16/38 (42.1) Ribavirin dose, mg/kg (IQR) 13.5 ( ) Peg-IFN-α2b dose, μg/kg (IQR) 1.45 ( ) ALT, IU/l (IQR) 80 (53 121) CD4 + T-cell count, cells/mm 3 (IQR) 515 ( ) Current antiretroviral treatment 82 (87.2) NNRTI, n (%) 48 (51.1) Protease inhibitor, n (%) 26 (27.7) Zidovudine, n (%) 13 (13.8) Stavudine, n (%) 30 (31.9) Lamivudine/Emtricitabine, n (%) 48 (51.1) Abacavir, n (%) 26 (27.7) Didanosine, n (%) 10 (10.6) Tenofovir, n (%) 39 (41.5) *Continuous variables are expressed as median (interquartile range; IQR). Body mass index (BMI), calculated as the weight in kilograms divided by the square of the height in metres. Thirty-eight patients underwent liver biopsy and samples were assessed by modified ISHAK score, a score of F5 F6 indicates cirrhosis. ALT, alanine aminotransferase; HCV, hepatitis C virus; peg-ifn-α2b, pegylated interferon-α2b; NNRTI, non-nucleoside reverse transcriptase inhibitor. Table 2. Predictors of ribavirin serum concentration at week 4 and 12 of treatment with peginterferon-α2b and ribavirin Multivariate analysis* Parameter Beta (95% CI) P-value Week 4 RBV dose, mg/kg 0.21 ( ) <0.001 Tenofovir 0.55 ( ) Week 12 RBV dose, mg/kg 0.14 ( ) Tenofovir 0.64 ( ) *Backward stepwise linear regression analysis. The initial model also included age, gender, RBV dose, creatine clearance, body mass index and individual antiretroviral drugs as covariates. CI, confidence interval; RBV, ribavirin. at week 12 of treatment. A similar increase was observed when RBV levels (μg/ml) were adjusted by RBV dose (mg/kg): 0.14 μg/ml/dose (IQR μg/ml/dose) at week 4 and 0.16 (IQR μg/ml/dose) at week 12 of treatment (P=0.001). Coefficients of variation of adjusted RBV levels were 28.5% and 36.2%, respectively. The weight-adjusted RBV dose, but not the crude RBV dose, and tenofovir treatment, co-administered to 39 (41.5%) of the participating patients, were independently associated with higher RBV levels at treatment week 4 and week 12 (Table 2). No significant changes were observed in RBV median levels between 15 paired frozen samples analysed a median of 6 months apart (1.68 μg/ml versus 1.53 μg/ml; P=0.186). Maximum haemoglobin decrease was observed at treatment week 4: mean 2.22 mg/dl, 95% confidence interval [CI] mg/dl, and stabilized thereafter. Multivariate regression analysis showed that the only factor associated with greater haemoglobin drop at treatment week 4 was zidovudine treatment, co-administered to 13 participating patients (13.8%) (odds ratio [OR] 1.74; 95% CI ; P<0.001). In patients not on zidovudine (n=81), a weak correlation was found between RBV concentration and haemoglobin drop at week 4 (r=0.32; P=0.006) and week 12 of treatment (r=0.35; P=0.004), respectively. However, neither RBV concentration nor any baseline variable predicted haemoglobin decrease among these patients. Overall, HCV clearance was observed in 36.2% (34/94) of patients at week 4 and 75.5% (71/94) of patients at week 12 of treatment. Rates of virological clearance were determined for individual genotypes at treatment week 4 and 12: genotype 1, 21.4% (9/42) and 64.3% (27/42), respectively; genotype 3, 74.2% (23/31) and 100% (31/31), respectively; genotype 4, 9.5% (2/21) and 61.9% (13/21), respectively. It is noteworthy that median RBV concentration at treatment week 4 was lower in patients with rapid viral Antiviral Therapy 12:8 1219

4 M Crespo et al. Table 3. Predictors* of HCV RNA clearance at week 4 and 12 of treatment with peginterferon-α2b plus ribavirin among patients with HCV genotype 1 or 4 Univariate analysis Multivariate analysis Parameter OR (95% CI) P-value OR (95% CI) P-value Week 4 HCV RNA 800,000 IU/ml 0.22 ( ) ( ) RBV level 0.12 ( ) ( ) Week 12 HCV RNA 800,000 IU/ml 0.11 ( ) ( ) RBV level 0.93 ( ) ( ) *Age, gender, body mass index, pre-treatment hepatitis C virus (HCV) RNA, HCV genotype, ribavirin (RBV) dose (mg/kg), aspartate aminotransferase, alanine aminotransferase, γ-glutamyl transpeptidase, liver cirrhosis and antiretroviral treatment, as well as RBV serum concentration at treatment week 4 and 12 (each one in independent logistic regression models) were used in the univariable and multivariable analyses. Undetectable HCV RNA by a real-time PCR test with a detection limit of 15 IU/ml (COBAS Ampliprep-COBAS TaqMan 48; Roche Molecular Diagnostics, Pleasanton, CA, USA). CI, confidence interval; OR, odds ratio. response (1.43 μg/ml; IQR μg/ml) than in those with a detectable viral load at this time point (1.87 μg/ml; IQR μg/ml; P=0.054), mainly among patients with genotype 1 or 4 HCV (1.22 μg/ml; IQR μg/ml (Figure 1) versus 1.87 μg/ml; IQR μg/ml; P=0.003). Among genotype 1 or 4 HCV-infected patients, the chance to achieve viral clearance at week 4 of treatment was higher in those with low pre-treatment HCV RNA (<800,000 IU/ml) and lower RBV concentration, whereas low pre-treatment HCV RNA was the only predictor of viral clearance at treatment week 12 (Table 3). Treatment with abacavir (n=26, 27.7%) or any other single antiretroviral drug or combination regimen was not associated with virological clearance at week 4 or 12 of Figure 1. Correlation between ribavirin concentration and virological clearance at treatment week 4 (RVR) and 12 (EVR) in genotype 1 or 4 HCV HIV-coinfected patients Ribavirin trough level, µg/ml RVR n=11 Week 4 P=0.003 No RVR n=52 EVR n=40 Week 12 P=0.88 EVR, early virological response; RVR, rapid virological response. No EVR n=23 treatment. Among patients with genotype 3, neither RBV concentration nor any pre-treatment variable was associated with rapid viral response or early viral response (data not shown). Overall, ROC curve analysis showed that the ability of RBV concentration to identify viral responders was poor at week 4 (area under the curve 0.63, 95% CI ) and week 12 of treatment (area under the curve 0.54, 95% CI ). However, at treatment week 4, the performance capacity of RBV concentration improved among patients with genotype 1 or 4: area under the curve 0.82; 95% CI The cutoff value that best discriminated between rapid virological responders and patients with detectable HCV RNA at this time point was RBV serum trough level <1.63 μg/ml (sensitivity 1; specificity 0.64; positive predictive value 0.35; negative predictive value 1). Discussion In the present study, RBV trough serum level at week 4 and 12 of treatment showed broad interindividual variability, similar to previously reported data [17,19,20,24]. A plateau level has been reported to occur in RBV concentration after 28 days of administration [16,17,19,25]. In contrast, we found a significant increase in RBV concentration from week 4 to 12 of treatment, in agreement with Dahari et al. [26]. According to pharmacokinetic analysis, some authors have suggested that RBV dosage should be based on renal function rather than on patient s weight [24,27]. However, a study in healthy volunteers showed that <5% of RBV dose was eliminated in urine, suggesting that in patients with normal creatinine clearance RBV metabolism is the main route of elimination [28]. In keeping with previous results [17,19], we found that the weight-adjusted dose of RBV, but not the glomerular filtration rate or the crude RBV dose, was a International Medical Press

5 Independent predictors of ribavirin serum concentration main determinant of RBV concentration. Furthermore, in our study tenofovir co-administration was associated with higher RBV levels. The mechanism accounting for tenofovir RBV interactions has not yet been elucidated. In healthy subjects, co-administration of tenofovir and a single dose of RBV did not change the individual pharmacokinetic profiles of these drugs [29]. In vitro analysis has shown that the mono- and diphosphate forms of tenofovir are inhibitors of purine nucleoside phosphorylase dependent breakdown of didanosine [30], which may explain the increase in plasma didanosine concentration observed in clinical practice under co-administration of both drugs [31]. Durand-Gasselin et al. [32] have found high levels of phosphorylated metabolites of tenofovir in erythrocytes, further supporting a potential interaction of tenofovir on didanosine metabolism. RBV is also a purine nucleoside analogue, and extensively distributes into red blood cells following oral administration. Phosphorylated metabolites of tenofovir could interfere with the degradation pathway of RBV and might account for the higher RBV levels observed in our study among patients receiving both drugs. In addition, tenofovir can have a negative effect on creatinine clearance [33] and, hypothetically, could interfere with RBV renal clearance. However, in our study, the glomerular filtration rate showed no differences between patients who received tenofovir and those who did not. Hence, whether tenofovir interferes with RBV metabolism or RBV renal clearance remains to be elucidated. RBV monitoring has been proposed as a useful strategy for managing toxicity [17], but we found only a weak correlation between haemoglobin drop and RBV concentration. Furthermore, the haemoglobin decrease shows great interindividual variability and usually occurs during the first month of therapy [5]. Hence, RBVinduced anaemia can be rationally managed by erythropoietin supplementation and dose-reduction of RBV as necessary [19]. Also, in our study, co-administration of zidovudine was associated with a greater decrease in haemoglobin, which supports the recommendation to avoid zidovudine during anti-hcv therapy [34]. Another potential benefit of RBV serum monitoring would be the possibility to individually adjust RBV dose early in treatment to achieve a target concentration favouring viral response. Data obtained from small observational studies have shown that higher RBV concentrations increase sustained viral response rates in HCV-monoinfected patients [19,20,35]. In one study in HCV HIV-coinfected patients, an HCV RNA drop >2 log at treatment week 4 and week 12 was found to be more likely with higher plasma RBV levels [17], whereas another only found this correlation in genotype 1 or 4 HCV HIV-coinfected patients [18]. In contrast to previous data, Dahari et al. [26] in a detailed prospectively conducted study on RBV pharmacokinetics in 24 HCV HIV-coinfected patients (91% with genotype 1), found that RBV plasma concentration at week 4 of treatment was significantly lower among six patients who eventually obtained a sustained response than in 18 non-responders. The authors hypothesis was that intracellular RBV concentration may be enhanced early in treatment in sustained responders, a concept supported by the fact that haemoglobin decline at week 4 of treatment was significantly higher among viral responders. The differences between Dahari s results [26] and the above-mentioned data might be explained by several possible components of bias in the latter studies: RBV adherencedriven bias (low RBV levels and a poor virological response could be due to poor treatment adherence); sampling error (the samples used for analysis might have not actually been trough samples, but instead samples obtained after RBV dose among patients maintaining good adherence to treatment and, therefore, having a greater chance of virological response); and/or by haemolysis bias (because of high RBV accumulation in erythrocytes, haemolysis occurring during whole blood handling to obtain serum or plasma samples for analysis could falsely increase RBV serum/plasma levels). In keeping with Dahari s findings, RBV trough serum levels in the present study were lower among patients showing viral clearance at week 4 of treatment, mainly in patients bearing genotype 1 or 4. Theoretically, early intracellular accumulation can increase the mutagenic effect of RBV, especially among difficult to treat patients in whom interferon antiviral effectiveness is weak or moderate [36]. In the present study, however, the ability of RBV concentration to discriminate between patients who did or did not clear HCV RNA at week 4 and 12 of treatment was poor. Therefore, whether early intracellular RBV accumulation improves the kinetics of HCV response to combination treatment warrants further research. Meanwhile, RBV dose should be probably weight-based in HCV HIV-coinfected patients starting treatment [37], mainly among difficult-to-treat patients [38], and viral response at week 4 and 12 should be monitored to optimize treatment. A recent substudy of patients enrolled in the RIBAVIC trial raised concerns that abacavir co-administration might have a negative effect on early virological response to peg-inf-α2b plus RBV [39]. In the present study, virological response at weeks 4 and 12 of treatment was not influenced by abacavir or any other single antiretroviral drug or combined regimen. As a potential limitation of our study, we should note that patients were included only if stored serum samples obtained at week 4 and 12 of treatment were Antiviral Therapy 12:8 1221

6 M Crespo et al. available for analysis, which could potentially imply a selection bias in that patients who prematurely discontinued therapy because of intolerance or serious adverse effects were excluded. However, 85% of patients who started therapy with peg-inf-α2b plus RBV in our hospital in 2001 were included in the present analysis, and the rates of rapid viral response and early viral response were similar to those observed among patients included in a randomized trial [5]. Additionally, RBV concentration was analysed in serum samples stored for up to 5 years. Loregian A et al. [40] have recently reported that RBV remains stable in plasma samples stored at -20 C for up to 1 month. Although we cannot definitely exclude a gradual decrease in RBV levels over time, no significant changes were observed in 15 frozen samples analysed a median of 6 months apart. In conclusion, RBV serum trough level mainly depends on weight-adjusted RBV dose, and higher levels were associated with tenofovir co-administration. As previously reported, zidovudine was associated with higher haemoglobin decrease and should be avoided. In contrast with most previous reports, RBV concentration at treatment week 4 was lower in genotype 1 or 4 patients showing a rapid viral response, which suggests that early intracellular RBV accumulation might improve the kinetics of the HCV response to combination treatment. Although further studies are necessary to explore this hypothesis and the potential interaction between RBV and tenofovir, our data do not support the use of RBV level monitoring as a tool to optimize HCV therapy in HCV HIV-coinfected patients. Acknowledgements The authors thank Celine Cavallo for English language editing. Funding This study was supported in part by grants from Red Temática Cooperativa de Investigación en SIDA (Red de grupos 173) del FISS; Centro de Investigación Biomédica en Red en Enfermedades Hepáticas y Digestivas (CIBER 06/040028). References 1. Davis GL, Wong JB, McHutchison JG, Manns MP, Harvey J, Albrecht J. Early virologic response to treatment with peginterferon alfa-2b plus ribavirin in patients with chronic hepatitis C. Hepatology 2003; 38: Torriani FJ, Rodriguez-Torres M, Rockstroh JK, et al. Peginterferon alfa-2a plus ribavirin for chronic hepatitis C virus infection in HIV-infected patients. N Engl J Med 2004; 351: Carrat F, Bani-Sadr F, Pol S, et al. Pegylated interferon alfa- 2b vs standard interferon alfa-2b, plus ribavirin, for chronic hepatitis C in HIV-infected patients: a randomized controlled trial. JAMA 2004; 292: Chung RT, Andersen J, Volberding P, et al. Peginterferon alfa-2a plus ribavirin versus interferon alfa-2a plus ribavirin for chronic hepatitis C in HIV-coinfected persons. N Engl J Med 2004; 351: Crespo M, Sauleda S, Esteban JI, et al. Peginterferon alpha- 2b plus ribavirin vs interferon alpha-2b plus ribavirin for chronic hepatitis C in HIV-coinfected patients. J Viral Hepat 2007; 14: Ferenci P, Fried MW, Shiffman ML, et al. Predicting sustained virological responses in chronic hepatitis C patients treated with peginterferon alfa-2a (40 KD)/ribavirin. J Hepatol 2005; 43: Mangia A, Santoro R, Minerva N, et al. Peginterferon alfa- 2b and ribavirin for 12 vs. 24 weeks in HCV genotype 2 or 3. N Engl J Med 2005; 352: Jensen DM, Morgan TR, Marcellin P, et al. Early identification of HCV genotype 1 patients responding to 24 weeks peginterferon alpha-2a (40 kd)/ribavirin therapy. Hepatology 2006; 43: Crespo M, Esteban JI, Ribera E, et al. Utility of week-4 viral response to tailor treatment duration in hepatitis C virus genotype 3/HIV co-infected patients. AIDS 2007; 21: Fried MW, Shiffman ML, Reddy KR, et al. Peginterferon alfa-2a plus ribavirin for chronic hepatitis C virus infection. N Engl J Med 2002; 347: Manns MP, McHutchison JG, Gordon SC, et al. Peginterferon alfa-2b plus ribavirin compared with interferon alfa-2b plus ribavirin for initial treatment of chronic hepatitis C: a randomised trial. Lancet 2001; 358: Bronowicki JP, Ouzan D, Asselah T, et al. Effect of ribavirin in genotype 1 patients with hepatitis C responding to pegylated interferon alfa-2a plus ribavirin. Gastroenterology 2006; 131: Layden-Almer JE, Ribeiro RM, Wiley T, Perelson AS, Layden TJ. Viral dynamics and response differences in HCVinfected African American and white patients treated with IFN and ribavirin. Hepatology 2003; 37: Neumann AU, Lam NP, Dahari H, et al. Differences in viral dynamics between genotypes 1 and 2 of hepatitis C virus. J Infect Dis 2000; 182: Herrmann E, Lee JH, Marinos G, Modi M, Zeuzem S. Effect of ribavirin on hepatitis C viral kinetics in patients treated with pegylated interferon. Hepatology 2003; 37: Pawlotsky JM, Dahari H, Neumann AU, et al. Antiviral action of ribavirin in chronic hepatitis C. Gastroenterology 2004; 126: Rendon AL, Nunez M, Romero M, et al. Early monitoring of ribavirin plasma concentrations may predict anemia and early virologic response in HIV/hepatitis C virus-coinfected patients. J Acquir Immune Defic Syndr 2005; 39: Aguilar D, Gonzalez-Requena D, Veronese L, et al. Cut-off values of RBV through concentration for sustained virological response and for hematological toxicity in HIV/HCV co-infected patients treated with ribavirin and pegylated interferon. 14th Conference on Retrovirus and Opportunistic Infections February 2007, Los Angeles, CA, USA. Abstract Jen JF, Glue P, Gupta S, Zambas D, Hajian G. Population pharmacokinetic and pharmacodynamic analysis of ribavirin in patients with chronic hepatitis C. Ther Drug Monit 2000; 22: Larrat S, Stanke-Labesque F, Plages A, Zarski JP, Bessard G, Souvignet C. Ribavirin quantification in combination treatment of chronic hepatitis C. Antimicrob Agents Chemother 2003; 47: Svensson JO, Bruchfeld A, Schvarcz R, Stahle L. Determination of ribavirin in serum using highly selective solid-phase extraction and high-performance liquid chromatography. Ther Drug Monit 2000; 22: Shah VP, Midha KK, Findlay JW, et al. Bioanalytical method validation - a revisit with a decade of progress. Pharm Res 2000; 17: International Medical Press

7 Independent predictors of ribavirin serum concentration 23. Cockcroft DW, Gault MH. Prediction of creatinine clearance from serum creatinine. Nephron 1976; 16: Maeda Y, Kiribayashi Y, Moriya T, et al. Dosage adjustment of ribavirin based on renal function in Japanese patients with chronic hepatitis C. Ther Drug Monit 2004; 26: Lindahl K, Schvarcz R, Bruchfeld A, Stahle L. Evidence that plasma concentration rather than dose per kilogram body weight predicts ribavirin-induced anaemia. J Viral Hepat 2004; 11: Dahari H, Markatou M, Zeremski M, et al. Early ribavirin pharmacokinetics, HCV RNA and alanine aminotransferase kinetics in HIV/HCV co-infected patients during treatment with pegylated interferon and ribavirin. J Hepatol 2007; 47: Bruchfeld A, Lindahl K, Schvarcz R, Stahle L. Dosage of ribavirin in patients with hepatitis C should be based on renal function: a population pharmacokinetic analysis. Ther Drug Monit 2002; 24: Lin CC, Philips L, Xu C, Yeh LT. Pharmacokinetics and safety of viramidine, a prodrug of ribavirin, in healthy volunteers. J Clin Pharmacol 2004; 44: Ramanathan S, Cheng A, Mittan A, Ebrahimi R, Kearney BP. Absence of clinically relevant pharmacokinetic interaction between ribavirin and tenofovir in healthy subjects. J Clin Pharmacol 2006; 46: Ray AS, Olson L, Fridland A. Role of purine nucleoside phosphorylase in interactions between 2,3 -dideoxyinosine and allopurinol, ganciclovir, or tenofovir. Antimicrob Agents Chemother 2004; 48: Kearney BP, Sayre JR, Flaherty JF, Chen SS, Kaul S, Cheng AK. Drug-drug and drug-food interactions between tenofovir disoproxil fumarate and didanosine. J Clin Pharmacol 2005; 45: Durand-Gasselin L, Da SD, Benech H, Pruvost A, Grassi J. Evidence and possible consequences of the phosphorylation of nucleoside reverse transcriptase inhibitors in human red blood cells. Antimicrob Agents Chemother 2007; 51: Accepted for publication 1 August Gallant JE, Parish MA, Keruly JC, Moore RD. Changes in renal function associated with tenofovir disoproxil fumarate treatment, compared with nucleoside reversetranscriptase inhibitor treatment. Clin Infect Dis 2005; 40: Alberti A, Clumeck N, Gerlich W, Lundgren J, Palù G, Reiss P. Short statement of the first European Consensus Conference on the treatment of chronic hepatitis B and C in HIV co-infected patients. J Hepatology 2005; 42: Tsubota A, Akuta N, Suzuki F, et al. Viral dynamics and pharmacokinetics in combined interferon alfa-2b and ribavirin therapy for patients infected with hepatitis C virus of genotype 1b and high pretreatment viral load. Intervirology 2002; 45: Dixit NM, Layden-Almer JE, Layden TJ, Perelson AS. Modelling how ribavirin improves interferon response rates in hepatitis C virus infection. Nature 2004; 432: Ramos B, Nunez M, Rendon A, et al. Critical role of ribavirin for the achievement of early virological response to HCV therapy in HCV/HIV-coinfected patients. J Viral Hepat 2007; 14: Hadziyannis SJ, Sette H, Jr., Morgan TR, et al. Peginterferon-alpha2a and ribavirin combination therapy in chronic hepatitis C: a randomized study of treatment duration and ribavirin dose. Ann Intern Med 2004; 140: Bani-Sadr F, Denoeud L, Morand P, et al. Early virologic failure in HIV-coinfected hepatitis C patients treated with the peginterferon-ribavirin combination: does abacavir play a role? J Acquir Immune Defic Syndr 2007; 45: Loregian A, Scarpa MC, Pagni S, Parisi SG, Palu G. Measurement of ribavirin and evaluation of its stability in human plasma by high-performance liquid chromatography with UV detection. J Chromatogr B Analyt Technol Biomed Life Sci 2007; 856: Antiviral Therapy 12:8 1223

8