|Home | About | Journals | Submit | Contact Us | Français|
Nevirapine is a nonnucleoside reverse transcriptase inhibitor used as part of combination therapy for human immunodeficiency virus (HIV) infection. Nevirapine may be prescribed for patients with hepatic fibrosis and cirrhosis. Significant autoinduction of cytochrome P450 3A4 and 2B6 following multiple dosing prompted an assessment of the metabolic profiles in patients with liver disease receiving chronic nevirapine therapy. HIV-infected patients with hepatic fibrosis who were receiving a stable antiretroviral regimen containing nevirapine for ≥6 weeks had liver biopsy specimens assessed by Ishak histologic scoring and were grouped by severity (group 1, Ishak scores of 1 and 2; group 2, Ishak scores of 3 and 4; group 3, Ishak scores of 5 and 6). Steady-state trough nevirapine levels were determined for all patients, and additional measurements were obtained at 1, 2, and 4 h following nevirapine dosing for a subset of patients. The pharmacokinetics of nevirapine and its five metabolites were characterized, and a comparison of the results for the different Ishak groups was performed. Among 51 patients with hepatic fibrosis, the majority of whom were coinfected with hepatitis C virus or hepatitis B virus, differences between the maximum and the minimum observed plasma concentrations demonstrated a statistically significant flattening of the systemic exposure curves with progression from Ishak group 1 to Ishak group 2 or 3, suggesting a decrease in systemic clearance with the progression of liver disease. However, there were no significant differences in the trough and the maximum nevirapine concentrations between the Ishak groups. The metabolite profiles were also comparable across the Ishak groups. In HIV-infected patients who were chronically treated with nevirapine and who had various degrees of hepatic fibrosis, including cirrhosis, trough plasma nevirapine concentrations were not significantly increased, and thus, no dose adjustment is warranted.
Nevirapine (Viramune; Boehringer Ingelheim Pharmaceuticals, Inc., Ridgefield, CT) is a nonnucleoside reverse transcriptase inhibitor (NNRTI) used in combination with other antiretroviral (ARV) drugs for the treatment of human immunodeficiency virus (HIV) type 1 (HIV-1) infection. Nevirapine may be administered to patients with normal liver function, patients with liver fibrosis, cirrhotic patients with mild hepatic impairment (Child-Pugh score A), and those with viral hepatitis.
The assessment of systemic exposure to any drug is important to determine the pharmacological effects relating to efficacy and safety. The aim of HIV drug therapy is to maintain the systemic exposure at or above the 50% inhibitory concentration, the level needed to inhibit viral replication by at least 50%. The level of exposure is typically determined by measuring a trough level (minimum concentration [Cmin]), i.e., the plasma drug concentration prior to administration of the next dose. Significantly lower Cmins would be expected to reduce drug efficacy, while toxicity might be a concern in patients with elevated systemic drug exposures or peak concentrations.
Patients coinfected with HIV and hepatitis C virus (HCV) present a greater therapeutic challenge, as liver impairment alters the metabolism of many drugs. In the majority of patients with severe hepatic impairment (Child-Pugh score C), metabolism occurs at a slower rate and the systemic exposure to hepatically metabolized drugs accumulates with the duration of therapy (4). It would be a matter of clinical concern requiring dose adjustment if cirrhotic patients experienced difficulty metabolizing nevirapine, which would lead to elevated exposures. Although the causal link between elevated nevirapine exposure and toxicity is unproven (8), long-term exposure to high plasma nevirapine concentrations has not been rigorously studied in clinical trials.
Nevirapine is readily absorbed (>90%) after oral administration (10); freely partitions to all tissues, including the brain, because of its low level of protein binding (~60%); and is eliminated primarily via hepatic metabolism. Cytochrome P450 (CYP) oxidation, glucuronide (ether) conjugation, and urinary excretion of the ether-glucuronidated-oxidized metabolites represent the primary routes of nevirapine biotransformation and elimination in humans. Renal excretion of unchanged drug plays a minor role (13). The chronic administration of nevirapine induces hepatic CYP enzymes 3A4 and 2B6 by approximately 20% to 25% each (9). Autoinduction of CYP 3A4 and CYP 2B6 isozymes leads to an approximately 1.5- to 2-fold increase in the apparent oral clearance of nevirapine as treatment continues from a single dose to 2 to 4 weeks of 200 to 400 mg/day. Autoinduction also results in a corresponding decrease in the terminal-phase half-life of nevirapine in plasma, from approximately 45 h (single dose) to approximately 25 to 30 h (multiple dosing with 200 to 400 mg/day). Induction is complete within 28 days (9, 13), and the resultant steady-state plasma nevirapine Cmin of 4.7 μg/ml (interquartile range, 3.6 to 6.4 μg/ml) is stable for at least 1 year of therapy (7).
Because of this significant autoinduction of CYP 3A4 and CYP 2B6 following multiple dosing and the general observation that CYP 3A4 activity but not CYP 2B6 activity is affected by hepatic impairment (4), metabolic profiling of patients with hepatic fibrosis or cirrhosis receiving chronic (>6 weeks) nevirapine therapy was performed. If clearance was decreased in hepatically impaired patients due to reduced metabolic autoinduction, the difference between the nevirapine Cmin and peak concentrations (maximum plasma concentrations [Cmax]) would be smaller, as the Cmins would increase and produce a flatter pharmacokinetic curve over the 12-h dosing interval. Assessment of each plasma sample for nevirapine and the five oxidative metabolites following glucuronidase treatment evaluated whether potentially increased nevirapine exposures were due to an inability of the CYP 3A4 or CYP 2B6 isozyme to metabolize the parent drug, were a result of impaired hepatic metabolism (4), or were a result of the prevention of the excretion rate transfer of the glucuronidated product of oxidation (15).
This study was designed to evaluate the steady-state trough clearance of nevirapine and its five oxidative metabolites among HIV-1-infected patients with hepatic fibrosis, including patients with cirrhosis, receiving a stable ARV regimen which included nevirapine at 200 mg twice daily and to examine the effect of the degree of hepatic impairment on systemic nevirapine clearance.
This open-label, phase IV study was conducted at eight sites in the United States, France, and Spain (www.clinicaltrials.gov, NCT00144248). The patients were at least 18 years old and had HIV-1 infection and hepatic fibrosis, as documented by liver biopsy within 24 months of enrollment (no time restriction was used for cirrhotic patients). Patients were receiving a stable ARV regimen which included nevirapine at 200 mg twice daily for ≥6 weeks prior to enrollment. Alternatively, patients receiving a dose of nevirapine at 400 mg once daily for ≥6 weeks as part of a stable ARV regimen who were willing to switch to nevirapine at 200 mg twice daily for ≥14 days prior to collection of samples for Cmin determination were also enrolled. Patients who were receiving other NNRTI therapies within 4 weeks of study enrollment or who were concurrently receiving clarithromycin, rifampin (rifampicin), or St. John's wort were excluded from the study. No non-study-related treatments were administered.
All patients were monitored for approximately 4 weeks; and a baseline CD4+ cell count, HIV-1 RNA load determination, and laboratory tests were performed. Targeted physical examinations and documentation of concomitant medications and all adverse events occurred throughout the study period. Laboratory tests were performed to provide a baseline profile of the patient population and to calculate a Child-Pugh score for cirrhotic patients (5). A central pathologist read all slides with liver biopsy specimens and categorized them by using the Ishak fibrosis grading system (6), and patients were assigned to one of three groups (mild, Ishak group 1; moderate, Ishak group 2; severe, Ishak group 3) on the basis of their Ishak score (Table (Table1).1). The Child-Pugh score was used as a secondary measure of hepatic impairment for cirrhotic (Ishak group 3) patients.
The focus of this study was to evaluate (i) the degree of hepatic impairment among HIV-1-infected patients with hepatic fibrosis; (ii) steady-state nevirapine Cmins; and (iii) clearance estimates when evaluation of the Cmin was supplemented by measurement of the plasma levels at 1, 2, and 4 h after nevirapine administration in a subgroup of patients. Other evaluations included determination of the plasma levels of the nevirapine metabolites (2-hydroxynevirapine, 3-hydroxynevirapine, 8-hydroxynevirapine, 12-hydroxynevirapine, and 4-carboxynevirapine [Fig. [Fig.1])1]) in all patients and Child-Pugh scores among patients with cirrhosis (Ishak score, 5 or 6).
Steady-state nevirapine and metabolite Cmins were obtained for all patients at 12 h (±2 h) after administration of the evening oral dose of nevirapine. Additionally, pharmacokinetic measurements were obtained at 1 h (±15 min), 2 h (±30 min), and 4 h (± 30 min) after the morning nevirapine administration for a subset of patients (n = 34). Limited sampling schemes have shown that the accuracy of the results by analysis of a single sample of nevirapine at steady state is within 14% of the true area under the plasma concentration-time curve (AUC) because the drug has a half-life of >24 h and is administered twice daily (16). The pharmacokinetics of nevirapine and the five metabolites in all patients were characterized by noncompartmental methods with the pharmacokinetic and statistical software program WinNonlin (Pharsight Corp., Mountain View, CA).
A high-performance liquid chromatography method with UV detection for the analysis of nevirapine in human plasma was used as described previously (12). The nevirapine high-performance liquid chromatography method with UV detection was validated over the concentration range of 0.025 to 10.0 mg/liter; samples with concentrations above the upper limit of quantitation were reanalyzed after dilution. The average accuracy over two different concentration ranges was within 5% of the true value. The within- and between-day precisions were within 10% for quality control samples over both ranges. Repetitive thawing and freezing did not have an effect on the results (which were within 2% of the results for the control) through a minimum of three cycles. The results for thawed samples, which remained in plasma for 24 h before extraction, were within 4% of the theoretical results. The stability of the extracted samples on the autosampler at room temperature was evaluated for 31 h, and the results were observed to be within 4% of the results for a fresh analytical sample. Samples have been studied for over 3 years for freezer stability (−20°C) and were found to have no degradation (the results obtained with frozen samples were within 9% of the value for fresh samples), and stock solutions of nevirapine or the internal standard in methanol have been stable for nearly 2 years at −5°C.
To evaluate a potential shift in clearance, a liquid chromatography-tandem mass spectrometry (LC/MS/MS) assay was developed for the metabolites (14) after glucuronidase treatment. The LC/MS/MS assay was necessary because four of the five metabolites of nevirapine are regiospecific isomers with identical molecular weights. The LC/MS/MS method is applicable for the quantitation of nevirapine metabolites within a nominal concentration range of 0.010 to 1.0 mg/liter; samples with concentrations above the upper limit of quantitation were reanalyzed after dilution. The average accuracy at the lowest concentration for the quality control sample was 16% of the true value for 8-hydroxynevirapine; the average accuracies for all other metabolites were closer to the values for their known respective standards. Within- and between-day precisions were within 12% of the values for the quality control samples for all five metabolites. Repetitive thawing and freezing did not have an effect on any metabolite (the values for the metabolites were within 10% of the value for the control) through a minimum of three cycles. The results for thawed samples, which remained in plasma for 4 h before extraction, were within 5% of the theoretical results. The stability of the extracted samples on the autosampler at room temperature was evaluated for 48 h, and the values were observed to be within 12% of the value for a fresh analytical sample for 2-hydroxynevirapine and 3-hydroxynevirapine; the values for the other metabolites were within 6% of the theoretical values.
The primary parameters of interest included the steady-state observed Cmax, the observed Cmin, the absolute difference between those two concentrations (Cdiff), and the AUC over one dosing interval (AUCτ), estimated by using the linear trapezoidal method. Single missing pharmacokinetic concentration measurements or values below the limit of quantitation were not included in the calculations.
Prior to analysis, the values of the pharmacokinetic parameters were log transformed by using the natural log scale. For the primary pharmacokinetic analyses, descriptive statistics (including the 95% confidence intervals) and graphical displays were used to summarize and evaluate the pharmacokinetic parameters (Cmin, Cmax, AUCτ, and clearance).
The secondary pharmacokinetic analyses included a Kruskal-Wallis test to assess whether the values for patients in a higher hepatic fibrosis stratum were significantly different from the values for patients in a lower hepatic fibrosis stratum. Additionally, Fisher's exact test was used to determine whether there was an association between hepatic impairment and nevirapine Cmins (i.e.,<6,000 ng/ml and ≥6,000 ng/ml).
Descriptive statistics were also used to summarize and assess the baseline demographics, adverse events, and laboratory values.
Fifty-one HIV-1-infected patients with chronic liver disease, confirmed by the presence of fibrosis on liver biopsy, and taking nevirapine as part of a stable ARV regimen were entered into this pharmacokinetic study. Overall, the majority of patients were male (78%) and white (88%), and the mean age was 49 years. The demographic characteristics were comparable among the three Ishak groups (Table (Table1).1). The median time of exposure to nevirapine was 3.4 years. The nevirapine exposure time was fairly evenly divided: <2 years, 31.4% of patients; ≥2 to 4 years, 27.5%; and >4 to 6 years, 23.5%. The median HIV RNA load was 2.03 log10 copies/ml (range, 1.30 to 4.30 log10 copies/ml), and the median CD4+ cell count was 400 cells/mm3 (range, 141 to 861 cells/mm3). Thirty-four (66.7%) patients were coinfected with HCV, 4 (7.8%) were coinfected with hepatitis B virus (HBV), 1 was coinfected with both HCV and HBV, and 12 (23.5%) were not infected with either HBV or HCV. Among the 34 patients coinfected with HIV and HCV, 25 were classified in Ishak fibrosis group 2 or group 3.
A total of 20 patients were classified in Ishak group 1, 22 patients were classified in Ishak group 2, and 9 patients were classified in Ishak group 3. Eight of the nine cirrhotic patients classified in Ishak group 3 had mild hepatic impairment (Child-Pugh score A) (Table (Table1),1), on the basis of calculated Child-Pugh scores. A score could not be calculated for one cirrhotic patient because the prothrombin value was unavailable. There were no patients classified as having Child-Pugh score B or C.
A total of 51 patients were entered into the study; 5 patients were excluded from the pharmacokinetic analyses. Of these five patients, one patient was taking nevirapine at 400 mg once daily rather than at 200 mg twice daily, another patient did not take the required evening dose of nevirapine prior to sampling, a third patient's sample was lost in transit, and the remaining two patients had extremely low nevirapine and metabolite concentrations, consistent with nonadherence.
The primary pharmacokinetic endpoints were designed to investigate the magnitude and shape of the systemic exposure curves with increasing fibrosis severity, as presented in Table Table2.2. Differences in the values of the pharmacokinetic parameters were observed between Ishak group 1 and Ishak groups 2 and 3, while the values of the pharmacokinetic parameters in Ishak groups 2 and 3 were comparable. The difference between Cmax and Cmin (Cdiff) demonstrated a flattening of the systemic exposure curves with a progression from Ishak group 1 to Ishak group 2 or 3.
The average Cdiffs for Ishak groups 2 and 3 were 48% and 33% lower than the average Cdiff for Ishak group 1, respectively. The results of the Kruskal-Wallis test showed that patients in a higher hepatic fibrosis stratum had Cdiffs significantly different from those of patients in a lower hepatic fibrosis stratum (n = 33; P = 0.0404).
The geometric mean nevirapine Cmin was approximately 30% higher for patients in Ishak group 2 and 3 than for patients in Ishak group 1 (Table (Table2).2). However, patients in a higher hepatic fibrosis stratum did not have significantly different steady-state Cmins compared with those for patients in a lower hepatic fibrosis stratum (n = 46; P = 0.1365, Kruskal-Wallis test). In addition, the geometric mean nevirapine Cmaxs were comparable among the Ishak groups (n = 33; P = 0.9585, Kruskal-Wallis test; Table Table22).
Thirty-five percent of the patients (17/48) had elevated Cmins of ≥6,000 ng/ml (Table (Table3).3). The proportion of patients with elevated Cmins was higher in Ishak groups 2 and 3 than in Ishak group 1. Specifically, elevated Cmins of ≥6,000 ng/ml were present in 44% of the patients in Ishak group 3, 40% of the patients in Ishak group 2, and 26% of the patients in Ishak group 1. The results of Fisher's exact test revealed that the degree of hepatic fibrosis (i.e., Ishak group 1, Ishak group 2, and Ishak group 3) and nevirapine Cmins (P = 0.5731) were not statistically significantly associated. Additionally, no serious adverse events were observed for patients who had elevated nevirapine Cmins of ≥6,000 ng/ml.
The nevirapine metabolite profiles were comparable across the Ishak groups (Table (Table4);4); the geometric mean concentrations of the 2-hydroxy, 3-hydroxy, and 12-hydroxy metabolites of nevirapine were 186 ng/ml, 646 ng/ml, and 483 ng/ml, respectively. The geometric mean concentrations of the 8-hydroxy and 4-carboxy metabolites were 29 ng/ml and 18 ng/ml, respectively.
During the study period, a total of eight patients in the trial experienced adverse events, but these were unrelated to nevirapine use. No single event was reported more than once in any treatment group. Most of the events were of mild intensity; however, four events in Ishak group 1 patients were of moderate intensity. There were no serious adverse events or deaths reported for any of the patients in the study.
HIV-infected patients with chronic liver disease present a greater therapeutic challenge, as hepatic impairment alters the metabolism of many drugs. In the majority of patients with severe hepatic impairment, metabolism occurs at a slower rate and systemic exposures to hepatically metabolized drugs accumulate with time on therapy. If patients with hepatic impairment experience difficulty metabolizing ARV agents, leading to elevated exposures, a dose adjustment might be necessary.
Nevirapine is primarily cleared through oxidative drug metabolism in the liver, followed by glucuronidation to water-soluble conjugates for rapid elimination from the body. The loss of a substantial number of hepatocytes due to fibrotic tissue replacement compromises the liver's ability to clear drugs by oxidation and glucuronidation. Since nevirapine uses these two pathways for systemic clearance, a change in the steady-state hepatic clearance of nevirapine in the setting of fibrosis might result in a change in systemic nevirapine exposure, as well as a shift in the metabolite profile. Recently, Barreiro et al. (1) postulated that nevirapine clearance was not affected by hepatic fibrosis and potentially by cirrhosis because there are multiple pathways (CYP 2B6 and CYP 3A4) for oxidation and subsequent glucuronidation. However, only an elastometry method (FibroScan) was used to evaluate the degree of fibrosis without comparative histological data. The authors concluded that more profound hepatic damage would be required before a significant increase in plasma nevirapine concentrations was observed.
In our study, the majority of HIV-infected patients who had various degrees of histologically proven hepatic fibrosis (including cirrhosis) and who were chronically treated with nevirapine had Cmins within the expected therapeutic range. Thirty-four percent of the patients had elevated nevirapine Cmins (>6,000 ng/ml), and no safety issues related to nevirapine were observed.
Comparison of the pharmacokinetic data for the three Ishak groups showed that the geometric mean nevirapine trough concentration was higher for Ishak groups 2 and 3 than for Ishak group 1, although the difference was not statistically significant. For the proportion of patients (35%) who had Cmins of ≥6,000 ng/ml, the Cmins were approximately 30% higher than the expected level of 4,700 ng/ml that has been observed in HIV-1-infected patients with normal hepatic function. There were no differences in the geometric means of the nevirapine Cmaxs among the Ishak groups.
As the hepatic fibrosis progressed, the magnitude of the difference between Cmax and Cmin diminished (flatter curves), suggesting an increase in the nevirapine half-life. Increasing nevirapine Cmins approximated nevirapine Cmaxs. Therefore, with the gradual progression of liver impairment, clinically significantly elevated nevirapine levels did not occur. The proportional quantity of the nevirapine metabolites produced was not affected in any of the hepatically impaired groups, indicating that none of the pathways were altered by hepatic fibrosis or impairment.
The present study looked at nevirapine metabolic pathways and, within the limitations of this patient population, could not detect a difference in metabolite concentrations between Ishak groups and healthy volunteers. The results of this study were consistent with the finding of a previous study with healthy volunteers (13). In a radiolabel study with healthy volunteers, the 2-, 3-, and 12-hydroxy metabolites of nevirapine were excreted in the urine at 22.9%, 33.1%, and 29.7%, respectively, whereas the 8-hydroxy and 4-carboxy metabolites were excreted to a lesser extent: 3.2% and 0.7%, respectively. In the current study, the 8-hydroxy and 4-carboxy metabolites of nevirapine were also detected in plasma to a far lesser extent. In summary, there were no apparent alterations in the metabolic pathways or shifts to another pathway as a result of hepatic fibrosis or cirrhosis.
Additionally, nevirapine was found to be safe and well tolerated in HIV-1-infected patients with various degrees of hepatic fibrosis. There were no nevirapine safety or tolerability events reported in this patient population. Similarly, the 2NN trial observed no relationships between pharmacokinetic parameters and adverse events when adjustments for known covariates of gender, baseline CD4+ count, and hepatitis virus coinfection were made (8).
The patients in our study had been taking nevirapine for a median of 3.4 years as part of their ARV regimen. While there were no clinically significant observations related to pharmacokinetics, safety, or tolerability for this patient population, conclusions cannot be drawn for patients initiating nevirapine therapy.
Barreiro et al. (1) found that efavirenz levels were statistically significantly higher in cirrhotic patients (n = 16), were correlated with the degree of liver fibrosis, and were more frequently above the toxic threshold in cirrhotic patients. Also, in the HEPADOSE study, Dominguez et al. (3) demonstrated that patients who were coinfected with HIV and a hepatitis virus and who were receiving efavirenz had significantly higher Cmins than those infected with HIV alone. In contrast, there was no significant difference in Cmins among coinfected patients taking nevirapine.
In contrast to the finding by Macias et al. (11) that highly active ART (HAART) regimens that included nevirapine were associated with the faster progression of liver fibrosis in HIV-infected patients with chronic hepatitis, Berenguer et al. (2) recently evaluated a cohort of 201 patients coinfected with HIV and HCV who had a liver biopsy and who were receiving HAART. They found that an additional year each of HAART and NNRTI-based therapy was associated with a reduction in the progression of liver fibrosis. Notably, nevirapine exposure was more consistently associated with a reduction in fibrosis than efavirenz.
This study demonstrated that HIV-infected patients with histologically proven hepatic fibrosis, including cirrhosis with mild hepatic impairment, chronically treated with nevirapine did not experience substantial increases in nevirapine Cmins that would warrant a dose adjustment. Given the significant number of coinfected patients and the potential antiviral and antifibrotic benefits of HAART, the pharmacokinetic findings in this study support the use of nevirapine in patients with various degrees of hepatic fibrosis, including cirrhosis (Child-Pugh score A). Patients should be carefully monitored for evidence of drug-induced toxicity. Given the lack of pharmacokinetic data on nevirapine in cirrhotic patients with moderate to severe hepatic impairment (Child-Pugh scores B and C), nevirapine should not be used in these patients.
P.M., J.M., and F.F. contributed to the study by enrolling patients; A.M.C., T.R.M., J.M.W., and P.J.P. were involved in data analysis, data interpretation, and writing the manuscript. The following study investigators contributed to the study by enrolling patients: Richard Cazen, Maurizio Bonacini, Ralph Liporace, and Lynn Taylor (United States).
Financial support was provided by Boehringer Ingelheim Pharmaceuticals, Inc.
Envision Pharma (Southport, CT) provided assistance with editing the manuscript.
A.M.C., T.R.M., J.M.W., and P.J.P. are all employees of Boehringer Ingelheim Pharmaceuticals, Inc. J.M. is a consultant for Boehringer Ingelheim Pharmaceuticals, Inc. F.F. and P.M. are study investigators for Boehringer Ingelheim Pharmaceuticals, Inc.
Published ahead of print on 20 July 2009.