Abstract
Ribavirin (RBV) is an integral part of standard-of-care hepatitis C virus (HCV) treatments and many future regimens under investigation. The pharmacokinetics (PK), safety, and tolerability of RBV in chronically HCV-infected patients with renal impairment are not well defined and were the focus of an open-label PK study in HCV-infected patients receiving RBV plus pegylated interferon. Serial RBV plasma samples were collected over 12 h on day 1 of weeks 1 and 12 from patients with moderate renal impairment (creatinine clearance [CLCR], 30 to 50 ml/min; RBV, 600 mg daily), severe renal impairment (CLCR, <30 ml/min; RBV, 400 mg daily), end-stage renal disease (ESRD) (RBV, 200 mg daily), or normal renal function (CLCR, >80 ml/min; RBV, 800 to 1,200 mg daily). Of the 44 patients, 9 had moderately impaired renal function, 10 had severely impaired renal function, 13 had ESRD, and 12 had normal renal function. The RBV dose was reduced because of adverse events (AEs) in 71% and 53% of severe and moderate renal impairment groups, respectively. Despite this modification, patients with moderate and severe impairment had 12-hour (area under the concentration-time curve from 0 to 12 h [AUC0–12]) values 36% (38,452 ng · h/ml) and 25% (35,101 ng · h/ml) higher, respectively, than those with normal renal function (28,192 ng · h/ml). Patients with ESRD tolerated a 200-mg daily dose, and AUC0–12 was 20% lower (22,629 ng · h/ml) than in patients with normal renal function. PK modeling and simulation (M&S) indicated that doses of 200 mg or 400 mg alternating daily for patients with moderate renal impairment and 200 mg daily for patients with severe renal impairment were the most appropriate dose regimens in these patients.
INTRODUCTION
Infection with hepatitis C virus (HCV) has been estimated to affect more than 170 million patients worldwide (1, 2). Chronic HCV infection may be an important contributor to renal disease and has been associated with a higher risk for, and a shorter time to, the development of chronic kidney disease (CKD) (3, 4). The prevalence of patients with end-stage renal disease (ESRD) undergoing hemodialysis infected with the virus has been shown to be approximately 14% (range, 2.6% to 22.9%) (5). Taken together with projections of an estimated 775,000 patients with ESRD needing dialysis in the United States by 2020, approximately 109,000 of these patients will be infected with HCV (6).
The treatment of chronic HCV infection in patients with renal impairment is difficult because the standard of care, as well as many new regimens under investigation, requires combination therapies that include ribavirin (RBV), an agent whose plasma exposure is affected by renal function (5, 7, 8). RBV exposure has been shown to influence both clinical response and toxicity to therapy (9–14). Current treatment guidelines recommend RBV doses ranging from 200 to 800 mg/day in patients with moderate to severe kidney disease not undergoing hemodialysis and markedly reduced RBV daily doses in patients on dialysis (5). It should be recognized that the recommendations in ESRD are based upon small case series from clinical practice using RBV doses of 200 to 400 mg/day (15–17). Given that the inclusion of RBV in current and future HCV treatment regimens will be required for the foreseeable future and given the large number of HCV-infected patients with markedly impaired renal function and the limited data on dosing of RBV in this patient population, the current study was undertaken to evaluate the safety, tolerability, and pharmacokinetic (PK) profile of RBV in patients with various degrees of renal impairment who received RBV plus pegylated interferon (IFN) for treatment of HCV infection.
MATERIALS AND METHODS
Study population and design.
Male and female patients who met the following criteria were eligible for enrollment in the study: aged 18 to 65 years; in good health based upon medical history, physical examination findings, and laboratory testing; able to understand and sign the consent form; current documented HCV infection as evidenced by enzyme-linked immunosorbent screening plus a supplemental specificity test with a radioimmunoblot assay or alternatively by HCV RNA quantifiable at >2,000 copies/ml (Amplicor HCV monitor test, version 2.0; Roche Molecular Diagnostics, Pleasanton, CA); and, for women of childbearing potential, a negative serum β-human chorionic gonadotropin pregnancy test documented within 24 h before the first dose of test drug and continued use of 2 reliable forms of contraception during the study and up to 6 months after receiving the last dose of RBV (the latter for both men and women). Among patients with impaired renal function, baseline 24-hour urine creatinine collection was required that documented moderate (creatinine clearance [CLCR], 30 to 50 ml/min) or severe (CLCR, <30 ml/min) renal impairment or ESRD requiring hemodialysis (patients with ESRD must have been undergoing hemodialysis for ≥2 months before study start).
Patients were excluded if they were pregnant or lactating; had received RBV or IFN therapy within 3 months before study start; had serology positive for HAV, HBV, or HIV; had documented cirrhosis or history or evidence of bleeding from esophageal varices; had evidence of a medical condition (other than HCV) associated with chronic liver disease (e.g., hemochromatosis, autoimmune hepatitis, alcoholic liver disease, or toxin exposures) or serology consistent with an increased risk of metabolic liver disease (e.g., low serum concentrations of ceruloplasmin or α1-antitrypsin); had significantly altered hematologic parameters at screening (neutrophil count of <1,500 cells/mm3, hemoglobin of <11 g/dl, and platelet counts of <90,000 cells/mm3) or a medical condition at baseline for which anemia would be problematic; had a history of severe psychiatric disease, which, in the opinion of the investigator, would preclude the patients from complying with the study procedures; had a history of immunologically medicated disease (e.g., inflammatory bowel disease, rheumatoid arthritis, idiopathic thrombocytopenia purpura, lupus erythematosus, autoimmune hemolytic anemia, or severe psoriasis); had a history of or active severe cardiac disease, chronic pulmonary disease with functional limitations, severe seizure disorder (or current anticonvulsant medication use), neoplasms (or a history with risk of recurrence of ≥20% within 2 years), solid organ transplantation, thyroid disease, or severe retinopathy; had documented allergy or hypersensitivity to RBV and/or IFN or related compounds; and had a positive urine test result or evidence of alcohol or illicit drug abuse within 1 year of entry.
This was an open-label, nonrandomized, parallel-group, multicenter, international study to evaluate the PK and safety profile of RBV in HCV-infected patients with moderate (CLCR, 30 to 50 ml/min [group A]) or severe (CLCR, <30 ml/min [group B]) renal impairment or ESRD requiring hemodialysis (group C). A control group with normal renal function was enrolled for comparison (CLCR, >80 ml/min [group D]). Renal function was determined from medical history, clinical laboratory test results (e.g., serum creatinine), and 24-hour urine collection. All study patients provided written informed consent to participate and were given a copy of the signed informed consent for their records.
Patients underwent screening assessments within 35 days of dosing to determine their eligibility. If enrolled, patients were assigned to treatment groups and polyethylene glycol-IFN (PEG-IFN) alfa-2a and RBV doses according to estimated CLCR. Patients received PEG-IFN alfa-2a and RBV for 12 weeks, and full PK sampling was performed on day 1 of weeks 1 and 12 (see below for sampling schedule and procedures). A 12-week time period was selected for this PK study to ensure steady-state levels at the time of multiple-dose sampling. Upon completion of the 12-week PK study, patients could continue treatment for an additional 12 to 36 weeks, based upon HCV genotype, to complete the full approved course of PEG-IFN alfa-2a and RBV.
Patients with moderate renal impairment (CLCR, 30 to 50 ml/min) and severe renal impairment (CLCR, <30 ml/min) received RBV twice daily in total doses of 600 mg and 400 mg, respectively, for 12 weeks. Patients with ESRD received daily 200-mg doses of RBV for 12 weeks. The recommended approved doses of RBV were used for patients with normal renal function (800 mg daily for genotype 2 or 3 and 1,000 or 1,200 mg daily for patients weighing <75 kg or ≥75 kg in patients with genotype 1, respectively). All patients were instructed to take their ribavirin doses with ≥200 ml of water and with food. All patients who received ≥1 dose of study medication and adhered to the study protocol were eligible for inclusion in the PK analysis.
PK assessments and methods.
Serial PK sampling was performed on day 1 of weeks 1 and 12. Plasma concentrations of RBV were collected into heparinized tubes at predose and at 0.5, 1, 3, 5, 8, and 12 h postdose. To determine whether RBV was cleared via hemodialysis, additional blood (arterial and venous) and dialysate samples were collected on 1 day between days 2 and 7 of week 12 from patients undergoing hemodialysis.
Plasma RBV concentrations were determined using tandem liquid chromatography-mass spectrometry (LC-MS/MS) by Huntingdon Life Sciences, Inc. (East Millstone, NJ). The detection range was 5 to 1,280 ng/ml with an interassay precision range of 5% to 9.5%. Dialysate samples were analyzed after 1:1 dilution with plasma-dialysate with an interassay accuracy range of 3.6% to 6.0%.
A noncompartmental PK analysis of the plasma RBV concentration data was performed using WinNonlin version 5.0.1 (Pharsight Corporation, Mountain View, CA); this analysis utilized 686 ribavirin plasma concentrations from 63 subjects. The calculated PK parameters were the maximum observed plasma concentration (Cmax); apparent total body clearance (CL/F); and total drug exposure, expressed as the area under the plasma concentration-time curve from 0 to time of interest (t; 12 h for RBV) after dosing (AUC0–t), calculated using the linear trapezoidal method.
Statistical analysis of noncompartmental data.
All PK parameters were summarized using descriptive statistics, including arithmetic mean, standard deviation, coefficient of variation, median, and range. Descriptive statistics were used to summarize individual plasma concentrations and drug dose data, AUC, dose-normalized AUC, Cmax, time to maximum observed plasma concentration, and CL/F. A 1-way analysis of variance model with the factor renal impairment was used to evaluate the effect of renal impairment on apparent clearance CL/F, as well as the log-transformed AUC0–t and Cmax using the general linear model procedure in SAS, version 8.2 (SAS Institute, Cary, NC).
Population PK analysis.
Owing to extensive dose modifications in the patients with moderate and severe renal impairment, a population PK analysis was performed using nonlinear mixed-effect modeling (NONMEM version VII, level 1.0) to describe the RBV PK data and simulate alternate dosing regimens for these patient populations. Simulations were performed to identify a dosing regimen for patients with chronic kidney disease (CKD) that provides RBV exposure similar to that in patients with normal renal function. The RBV “target” concentration for patients with impaired renal function was defined as being within 20% of the average predicted concentration at steady state (Cssavg) in patients with normal renal function from a previously developed population PK model (18) (reference population) receiving doses of 1,000 or 1,200 mg daily based upon body weight.
Development of the population PK model was performed primarily using the NONMEM objective function value (OFV) to discriminate between candidate models. For a single degree of freedom, an OFV decrease of 3.84 is equivalent to an improvement of the model fit, which is significant at a P value of <0.05. Goodness-of-fit plots and scientific plausibility were also used to guide model development. NONMEM model fitting used the first-order conditional estimation method (FOCE) with interaction.
Two- and three-compartment structural models were considered, based on exploratory graphical analysis and previous reports describing ribavirin PK (18). Given the probable dependence of ribavirin PK on renal function, clearly evident during graphical exploration of the data, the relationship between apparent clearance and 24-h urine creatinine clearance (CLCR) was included very early in structural model development in order to minimize the likelihood of model misspecification. Linear, proportional, and power models were considered for describing population apparent clearance. Different coefficients for patients with CLCR values at screening of 50 ml/min or below for defining patients with moderate or severe renal impairment were also explored. Absorption was highly variable: several alternative absorption models were investigated, including absorption with a lag time, sequential zero and first-order absorption, Weibull-like and Emax-like models for Ka, and the multiple-dose transit-compartment model (19).
For covariate model development, the influence of continuous (age, weight, lean body weight, body mass index, and height) and categorical (sex and race) covariates was tested on apparent clearance (CL/F), apparent volumes of distribution (V2/F, V3/F), relative bioavailability (F), and absorption-related parameters. Forward covariate selection was performed using a P value of <0.05 as the selection criterion. Subsequently, backwards deletion was performed using a P value of <0.01. If inclusion of a given covariate produced an improvement in objective function value (OFV) that was significant with respect to these P values during forward and backward covariate model building, it was retained in the model.
In order to assess the final model's predictive performance, a predictive check was carried out to compare pharmacokinetic metrics (AUC0–t and Cmax, after 12 weeks) derived by noncompartmental analysis (NCA) with the model predictions. The final model was used to simulate 100 new data sets based on the study patients' data. For each simulated data set, AUC0–t and Cmax were calculated for every patient, and the resulting distributions were compared with the NCA-derived values for each renal function group.
Safety assessments.
All patients who received ≥1 dose of either study drug and had ≥1 postbaseline safety assessment were included in the safety analysis. Adverse events, including laboratory data and time of onset, severity, and potential relationship to the study drug, were assessed at weeks 1, 2, 3, 5, 8, 10, and 12 and at the follow-up examination at week 13. Study patients had a complete physical examination, including vital signs, at screening and at study completion. The following clinical laboratory tests were performed at screening; study day 1; weeks 2, 4, 6, 8, 10, and 12; and follow-up: hematology and serum chemistries (sodium, potassium, chloride, carbon dioxide, creatinine, albumin, glucose, total bilirubin, alanine aminotransferase, aspartate aminotransferase, alkaline phosphatase, uric acid, blood urea nitrogen, fractionated cholesterol, and triglycerides) and dipstick urinalysis. In addition, thyroid function tests were performed upon screening, IFN antibodies were evaluated at screening and at the follow-up visit, and 24-hour urine collections for CLCR evaluation were performed at study day 1 and week 12.
During the study, RBV was to be discontinued if a patient experienced a confirmed hemoglobin decline to <8.5 g/dl or a decrease from baseline of >3 g/dl. In patients with anemia who had a baseline hemoglobin concentration of <10 g/dl and experienced a decrease from baseline of <3 g/dl in their hemoglobin concentrations, initiation or increase of erythropoietin therapy and RBV decrease or discontinuation were allowed at the discretion of the investigator.
RESULTS
Study population.
A total of 63 patients were enrolled from 16 centers worldwide; 62 patients had ≥1 postbaseline assessment and were included in the safety population; Table 1 lists the key demographic variables for all 62 patients who were evaluable for the safety analysis and the week 12 demographic variables for the remaining patients who were evaluable for the PK analysis. Across all groups, the majority of patients were male, with the highest percentage being observed in the ESRD group (83%). The majority of patients with ESRD were black (61%), whereas the other 3 treatment groups were predominantly white. Patients with normal renal function were slightly younger and weighed slightly less than the patients with renal impairment. The majority of patients across all treatment groups were infected with HCV genotype 1, and those with normal renal function had lower HCV RNA titers than did patients with renal impairment.
Table 1.
Patient demographics at baseline and week 12 for ribavirina
| Patient characteristic | Group A, moderate renal impairment |
Group B, severe renal impairment |
Group C, end-stage renal disease |
Group D, normal renal function |
||||
|---|---|---|---|---|---|---|---|---|
| Baseline (overall) (n = 17) | Wk 12 (RBV) (n = 9) | Baseline (overall) (n = 14) | Wk 12 (RBV) (n = 10) | Baseline (overall) (n = 18) | Wk 12 (RBV) (n = 13) | Baseline (overall) (n = 13) | Wk 12 (RBV) (n = 12) | |
| Male, n (%) | 10 (59) | 5 (56) | 10 (71) | 8 (80) | 15 (83) | 12 (92) | 7 (54) | 7 (58) |
| Race, n (%) | ||||||||
| Black | 8 (47) | 4 (44) | 4 (29) | 3 (30) | 11 (61) | 7 (54) | 3 (23) | 3 (25) |
| White | 8 (47) | 4 (44) | 10 (71) | 7 (70) | 7 (39) | 6 (33) | 10 (77) | 9 (75) |
| Age, yrs, mean ± SD (range) | 53 ± 7 (39–65) | 54 ± 6 (47–65) | 52 ± 6 (41–58) | 53 ± 6 (41–58) | 52 ± 8 (33–62) | 52 ± 9 (33–62) | 48 ± 11 (23–62) | 49 ± 11 (23–62) |
| Weight, kg, mean ± SD (range) | 86 ± 20 (53–127) | 86 ± 22 (53–127) | 82 ± 20 (53–112) | 83 ± 18 (58–107) | 84 ± 18 (62–136) | 83 ± 21 (62–136) | 80 ± 17 (52–99) | 82 ± 15 (52–99) |
| HCV GT 1, n (%) | 15 (88) | 8 (89) | 13 (93) | 9 (90) | 14 (78) | 9 (69) | 9 (69) | 8 (67) |
| Baseline HCV RNA, ×103 IU/ml, mean ± SD (range) | 6,101 ± 11,245 (0.7–46,500) | 3,975 ± 4,171 (0.7–12,400) | 5,212 ± 5,219 (260–18,000) | 6,031 ± 5,639 (260–18,000) | 5,009 ± 8,082 (14.8–24,200) | 6,542 ± 9,103 (226–24,200) | 3,171 ± 5,918 (6.7–21,800) | 3,140 ± 6,180 (6.7–21,800) |
| ALT, U/liter, mean ± SD (range) | 53.2 ± 30.8 (15–127) | 53.4 ± 42.6 (15–127) | 42.6 ± 18.8 (21–94) | 45.8 ± 20.0 (23–94) | 42.5 ± 29.4 (7–127) | 45.5 ± 33.5 (7–127) | 76.8 ± 52.2 (26–168) | 69.3 ± 46.4 (26–148) |
| CLCR from serum creatinine, ml/min, mean ± SD (range) | 47 ± 18 (31–108) | 43 ± 9 (31–58) | 29 ± 13 (12–53) | 28 ± 13 (12–53) | 13 ± 4 (8–20) | 12 ± 4 (8–19) | 110 ± 29 (71–162) | 114 ± 27 (74–162) |
| CLCR from 24-h urine collection, ml/min, mean ± SD (range); n | 48 ± 18 (25–90); n = 16 | 40 ± 8 (28–52); n = 7b | 27 ± 10 (11–40); n = 13 | 24 ± 8 (10–38); n = 9b | 6 ± 5 (0–12); n = 11 | 7 ± 6 (0.3–16); n = 7b | 117 ± 37 (45–165); n = 12 | 117 ± 29 (77–181); n = 12b |
Baseline values for evaluable time point. ALT, alanine aminotransferase; CLCR, creatinine clearance; GT, genotype; HCV, hepatitis C virus; RBV, ribavirin.
Creatinine clearance from 24-hour urine creatinine collection at week 12.
During the first 12 weeks of the study, 11 patients (17%) prematurely discontinued treatment; 5 patients (8%) prematurely discontinued for safety reasons (see discussion below; 2 patients with severe renal impairment and 3 with ESRD), and 6 patients (10%) prematurely discontinued for non-safety-related reasons. The RBV dose was modified because of adverse events for approximately 75% of the patients with severe renal impairment and one-half of the patients with moderate renal impairment.
RBV PK. (i) Noncompartmental analysis.
Following a single dose, plasma RBV concentrations rose rapidly, with peak concentrations occurring within the first 2 to 3 h, and steadily declined over the remaining 9 to 10 h. Mean plasma RBV AUC0–12 on day 1 of week 1 was highest in patients with normal renal function (3,928 ng · h/ml) and was 28% lower in patients with moderate renal impairment (2,835 ng · h/ml), 14% lower in patients with severe renal impairment (3,364 ng · h/ml), and 24% lower in patients with ESRD on hemodialysis (2,989 ng · h/ml), reflecting the lower starting doses.
Table 2 lists the key steady-state RBV PK parameters observed at week 12; Fig. 1 shows the mean, median, and individual RBV AUC values for each of the treatment groups. At week 12, RBV plasma concentrations remained relatively flat in contrast to the sharp increase and gradual decrease observed after RBV administration on day 1. At week 12, the multiple-dose RBV mean AUC0–12 was 28,192 ng · h/ml in patients with normal renal function and was 36% higher in patients with moderate renal impairment (38,452 ng · h/ml), 25% higher in patients with severe renal impairment (35,101 ng · h/ml), and 20% lower in patients with ESRD on hemodialysis (22,629 ng · h/ml). The dose-normalized RBV mean AUC0–12 values were similar among renally impaired treatment groups (191 to 206 ng · h/ml) and were approximately 4-fold higher than those in patients with normal renal function (57 ng · h/ml). It is important to note that the RBV dose was modified for a large percentage of patients during the first 12 weeks of treatment, especially those in groups A and B with moderate and severe renal impairment (see safety section below). For this reason, the RBV dose administration in these patients before the week 12 PK assessments is likely to be different than the protocol-defined dose. The differences in dose may have been a change in dose (mg) level, dose interval, or both. Consequently, the actual RBV dose levels and dosing intervals were used to calculate the multiple-dose RBV PK parameters. As a result of the considerable RBV dose modification in groups A and B, RBV exposure at week 12 may not have reached steady state; therefore, the AUC0–12 and Cmax values reported for these 2 treatment groups are likely an underestimation of the anticipated steady-state values of these parameters at the protocol-defined dosing regimens. The extent of the difference between the multiple-dose RBV AUC0–12 and Cmax values and the anticipated steady-state RBV AUC0–12 and Cmax depended on the level, duration, and timing of the dose modification for each patient.
Table 2.
Week 12 steady-state ribavirin pharmacokinetic parameters
| Parameterb | Group A, moderate renal impairment (n = 9) | Group B, severe renal impairment (n = 10) | Group C, end-stage renal disease (n = 13) | Group D, normal renal function (n = 12) |
|---|---|---|---|---|
| Cmax, ng/ml, mean ± SD (range; CV) | 4,218 ± 1,801 (2,070–7,080; 43%) | 3,677 ± 1,828 (1,360–6,770; 50%) | 2,367 ± 560 (1,700–3,560; 24%) | 2,907 ± 757 (1,870–3,920; 26%) |
| Ratio of geometric least-squares mean (90% CI)a | 1.38 (1.06, 1.80) | 1.15 (0.85, 1.57) | 0.82 (0.69, 0.97) | |
| AUC0–12, ng · h/ml, mean ± SD (range; CV) | 38,452 ± 16,553 (18,835–75,327; 43%) | 35,101 ± 14,706 (n = 9) (14,695–50,435; 42%) | 22,629 ± 5,649 (n = 12) (14,836–37,437; 25%) | 28,192 ± 7,444 (18,303–38,326; 26%) |
| Ratio of geometric least-squares mean (90% CI)a | 1.31 (1.00, 1.70) | 1.17 (0.87, 1.57) | 0.81 (0.68, 0.97) | |
| Dose-normalized AUC, ng · h/ml, mean ± SD (range; CV) | 191 ± 93 (91–333; 49%) | 196 ± 53 (n = 8) (111–250; 27%) | 206 ± 60 (n = 7) (143–333; 29%) | 57 ± 20 (33–100; 35%) |
| CL/F, liters/h, mean ± SD (range; CV) | 6 ± 3 (3–11; 43%) | 6 ± 2 (n = 8) (4–9; 33%) | 5 ± 1 (n = 7) (3–7; 20%) | 20 ± 6 (10–30; 30%) |
| Ratio of geometric least-squares mean (90% CI)a | 0.34 (0.13, 0.55) | 0.28 (0.07, 0.49) | 0.26 (0.04, 0.49) |
Ratio of renal impairment group to group with normal renal function.
AUC, area under the concentration-time curve; CV, coefficient of variance; CI, confidence interval.
Fig 1.
Ribavirin AUC0–12 values for individual patients after multiple-dose administration. Open circles, individual patient values; solid lines, mean; dashed lines, median. A, group A (moderate renal impairment; CLCR, 30 to 50 ml/min); B, group B (severe renal impairment; CLCR, <30 ml/min); C, group C (end-stage renal disease with hemodialysis; CLCR, <10 ml/min); D, group D (normal renal function; CLCR, >80 ml/min). AUC, area under the concentration-time curve; RBV, ribavirin.
The mean RBV CL/F was 3 to 4 times faster in patients with normal renal function (20.0 liters/h) than in patients with renal impairment (≈5 to 6 liters/h). However, it should be noted that this parameter may also have been overestimated for patients with moderate and severe renal impairment due to extensive dose modifications, as previously described. Nevertheless, the potentially overestimated mean CL/F in the patients with moderate and severe renal impairment (6 liters/h) was only 30% of the mean CL/F in patients with normal renal function (20 liters/h) and was similar to the mean CL/F in patients with ESRD on hemodialysis (5 liters/h). The relationship between RBV CL/F and CLCR appeared linear only when the estimated CLCR was >50 ml/min.
In patients who received hemodialysis (group C), the mean venous RBV concentrations from samples collected 15 min after the start of, at the middle of, and near the end of hemodialysis were slightly more than one-half of the mean arterial RBV concentrations (901 and 1,615 ng/ml, respectively), suggesting an extraction coefficient of approximately 50%.
(ii) Population PK analysis and simulations.
RBV PK in patients with CKD was best described by a 2-compartment model with first-order elimination and parallel zero- and first-order absorption. Clearance was defined as having nonrenal and renal components, with the latter being wholly dependent on the 24-hour urine CLCR. The effect of 24-hour urine CLCR on CL/F was adjusted to allow different power coefficients for patients with CLCR values, >50 ml/min versus ≤50 ml/min. Clearance in patients with ESRD was estimated as a separate, independent, model parameter, necessary since such patients underwent dialysis during the course of the study. Interindividual variability was included on clearance, central volume of distribution, absorption rate constant, duration of zero-order absorption, and intercompartmental clearance. Residual variability was a combination of additive and proportional terms. An additional covariate effect of lean body weight was applied linearly to the relative bioavailability parameter. The parameter estimates for the final pharmacokinetic model are presented in Table 3.
Table 3.
Parameter estimates for the final model for ribavirin PKe
| Parameter | Estimate | SE (%) | Shrinkage (%) |
|---|---|---|---|
| Structural model | |||
| Relative bioavailabilitya | 1.00 | ||
| Clearance in ESRD (CLESRD, liters/h)b | 4.00 | 18.5 | |
| Nonrenal clearance (CLH, liters/h) | 2.49 | 16.8 | |
| Effect of CRAV on CL, CLCR ≤ 50 ml/min (θCL1)c | 1.10 | 27.1 | |
| Effect of CRAV on CL, CLCR > 50 ml/min (θCL2)c | 1.42 | 5.01 | |
| Central vol of distribution (V2, liters) | 380 | 8.22 | |
| Peripheral vol of distribution (V3, liters) | 2,230 | 8.11 | |
| Intercompartmental clearance (Q, liters/h) | 84.3 | 17.6 | |
| Absorption rate constant (Ka, per h) | 2.62 | 30.3 | |
| Duration of zero-order absorption (D, h) | 0.931 | 19.2 | |
| Effect of LBW on F (θLBW)d | −0.0111 | 27.3 | |
| Statistical model | |||
| IIV in clearance (ωCL2, %) | 34.2 | 30.8 | 17.2 |
| IIV in central vol of distribution (ωV22, %) | 34.1 | 49.4 | 29.9 |
| IIV in intercompartmental clearance (ωQ2, %) | 95.5 | 32.8 | 9.26 |
| IIV in absorption rate constant (ωKa2, %) | 82.2 | 113.5 | 47.1 |
| IIV in duration of zero-order absorption (ωD2, %) | 82.6 | 46.7 | 20.7 |
| Additive residual error (σadd2, SD) | 37.4 | 15.6 | 13.3 |
| Proportional residual error (σprop2, SD) | 0.239 | 2.92 |
Fixed.
Patients with ESRD: CL = CLESRD.
Patients without ESRD: CL = CLH + CRAV(θCL1|θCL2).
Fi = TVF · [1 + θLBW · (LBWi − LBWmed)]; LBWmed = 71 kg.
CRAV, average 24-h urine creatinine clearance, ml/min; LBW, lean body weight, kg; IIV, interindividual variability; TVF, total volume of flow.
With this model, the predicted RBV concentrations in patients with normal renal function in this study were similar to those obtained using the previously developed reference model (18), providing confirmation that the population PK model developed using the data from this renal impairment study was behaving as expected (Fig. 2). The predictive checks confirm that the model is able to reproduce the values of AUC0–12 (Fig. 3) and Cmax (data not shown) observed with the noncompartmental analysis.
Fig 2.
Predicted concentration-time profiles for patients with normal renal function. The green lines represent the predicted concentration-time profiles for patients in the study treated with the recommended dose of ribavirin. The black lines represent the median predicted concentration-time profile in the reference population receiving the recommended dose. The light and dark gray areas represent the 80% and 90% prediction intervals of the reference population, respectively. Patients treated with the 1,000-mg daily dose (400 mg in the morning and 600 mg in the evening) appear in the left panel, and patients treated with the 1,200-mg daily dose (600 mg twice daily) appear in the right panel. The dashed line represents the predicted profile for a single patient; this patient was enrolled in the normal renal function group (D) but had consistently low estimates of creatinine clearance (<80 ml/min) and was excluded from the analysis.
Fig 3.
Predictive check for week 12 AUC0–12. Histograms represent the distribution of simulated week 12 AUC0–12 values for the study population. The dashed blue line is the mean of the simulated AUC0–12, and the solid red line is the mean AUC0–12 obtained by noncompartmental analysis.
Simulations using the model predicted that dosing with 200 mg of RBV daily would result in a median decrease in RBV exposure of 16% and 9.5% in patients with ESRD or severe renal impairment, respectively, relative to the Cssavg target. In patients with moderate renal impairment, compared with the Cssavg target, dosing with 200 mg daily was predicted to result in a median decrease in RBV exposure of 26%. Dosing with 200 mg twice daily was predicted to result in a median increase in RBV exposure of 48%. Alternate daily dosing of RBV 200 mg or 400 mg was predicted to result in a median increase in RBV exposure of 11% relative to the Cssavg target (Fig. 4).
Fig 4.
RBV Cssavg in chronically HCV-infected patients with CKD. bid, twice daily; CKD, chronic kidney disease; Cssavg, average predicted concentration at steady state; ESRD, end-stage renal disease; HCV, hepatitis C virus; qd, once daily; RBV, ribavirin.
Safety and tolerability.
Table 4 summarizes the safety, tolerability, and key laboratory changes for each treatment group over the 12-week study period and 1-week poststudy follow-up examination. Almost all patients in each of the 4 treatment groups reported ≥1 adverse event, and most patients had ≥1 adverse event that was considered by the investigator to be remotely, possibly, or probably related to study drug. The most frequently reported adverse events were those known to be associated with RBV and PEG-IFN alfa-2a, including fatigue, anemia, headache, nausea, pyrexia, diarrhea, chills, and arthralgia. Patients in the severe renal impairment group had a greater incidence than the other 3 treatment groups of severe or serious adverse events, RBV dosage adjustments or discontinuations, and laboratory abnormalities leading to discontinuation. The most common serious adverse events were anemia (1 moderate impairment group patient and 2 patients with severe renal impairment) and mental status change (reported in 2 patients with severe renal impairment). No adverse events considered to be life threatening were observed in any patient. The RBV dose was modified for adverse events or laboratory abnormalities for approximately 75% of patients with severe renal impairment, as well as one-half of the patients with moderate renal impairment. The 3 most frequently reported laboratory abnormalities potentially related to RBV dose modification in patients with severe or moderate renal impairment were hemoglobin concentration of ≤10 g/dl or a decrease from baseline in hemoglobin concentration of ≥3 g/dl (43% to 47%), followed by abnormal white blood cells (36% to 41%) and platelet counts (24% to 29%). Modification of PEG-IFN alfa-2a doses for adverse events was less common and occurred in 14% to 31% of patients across all 4 treatment groups (Table 4).
Table 4.
Adverse events during first 12 weeks
| Event typed | No. (%) |
|||
|---|---|---|---|---|
| Group A, moderate renal impairment (n = 17) | Group B, severe renal impairment (n = 14) | Group C, end-stage renal disease (n = 18) | Group D, normal renal function (n = 13) | |
| Any AE | 17 (100) | 14 (100) | 16 (89) | 13 (100) |
| Related AEa | 17 (100) | 14 (100) | 16 (89) | 12 (92) |
| Severe AE | 2 (12) | 6 (43) | 5 (28) | 4 (31) |
| Serious AEb | 2 (12) | 3 (21) | 3 (17) | 1 (8) |
| Related serious AEa | 1 (6) | 3 (21) | 2 (11) | 0 |
| Total no. of SAEs | 2 | 7 | 6 | 1 |
| Any ribavirin dose modification for AEs or lab abnormalities | ||||
| Any | 9 (53) | 10 (71) | 4 (22) | 3 (23 |
| For Hgb | 8 (47) | 6 (43) | 1 (6) | 0 |
| For platelets | 4 (14) | 4 (29) | 1 (6) | 1 (8) |
| Premature withdrawals for AEs or lab abnormalities | ||||
| Ribavirin | 2 (12) | 5 (36) | 3 (17) | 0 |
| PEG-IFN alfa-2a | 0 | 2 (14) | 3 (17) | 0 |
| Laboratory changes | ||||
| Hgb > 8.5 to < 10.0 g/dl | 7 (41) | 5 (36) | 5 (28) | 1 (8) |
| Hgb < 8.5 g/dl | 4 (24) | 4 (29) | 0 | 0 |
| Neutrophils < 1.0c | 8 (47) | 1 (7) | 4 (22) | 5 (38) |
| Neutrophils < 0.5c | 3 (18) | 1 (7) | 0 | 1 (8) |
| Platelets < 100c | 3 (18) | 5 (36) | 6 (33) | 4 (31) |
| Platelets < 50c | 0 | 2 (14) | 1 (6) | 1 (8) |
| White blood cells | 7 (41) | 5 (36) | 1 (6) | 1 (8) |
Events judged by the investigator to be remotely, possibly, or probably related to treatment.
SAEs by group: moderate renal impairment, anemia (1), auricular perichondritis (1); severe renal impairment, anemia (2), hypotension (1), mental status change (2), renal failure (1), fluid overload (1); end-stage renal disease, vasculitis (1), vascular occlusion (1), acute renal failure (1), pyloric stenosis (1), pyrexia (1), hemoptysis (1); normal renal function, constipation (1).
Multiplied by 109/liter.
AE, adverse event; Hgb, hemoglobin; PEG-IFN, pegylated interferon; SAE, serious adverse event.
DISCUSSION
The management of HCV-infected patients involves combinations that include RBV. Given that a large proportion of patients with HCV have altered renal function and that more than 40% of RBV elimination following intravenous administration is accounted for by renal clearance (20), the current study was undertaken to assess the safety, tolerability, and PK profile of RBV in HCV-infected patients with various degrees of renal impairment who received treatment with RBV plus pegylated IFN alfa-2a. Since this was a controlled clinical study that included PK assessments, it was necessary to define fairly extensive inclusion and exclusion criteria to ensure that the population was as homogeneous as possible so that data could be properly interpreted by limiting as many confounding variables as possible. This should be taken into consideration by practitioners when dealing with individual patients in daily practice, a population that is much more heterogeneous. In the present study, patients with renal impairment received decreased doses of RBV estimated to reflect concentrations seen in patients with normal renal function receiving RBV. The data from the current study revealed that doses of 600 and 400 mg for patients with moderate and severe renal impairment, respectively, were too high. The dose-normalized RBV exposure for these groups was approximately 4 times greater, and the systemic clearance 30% lower, than that observed for patients with normal renal function. In addition, patients in these 2 treatment groups had difficulties tolerating their assigned RBV doses, with approximately 75% of patients with severe and one-half of patients with moderate renal impairment requiring RBV dose modification during the first 12 weeks of therapy. Anemia, a common RBV-related adverse event associated with increased RBV exposure (12–14, 21), was the most frequently reported adverse event in patients with moderate and severe renal impairment, reported in 65% and 64% of patients, respectively. Only 20% of patients with moderate and severe renal impairment received their full course of assigned RBV doses for the first 12 weeks. Furthermore, patients with severe renal impairment had the highest frequency of treatment discontinuation for safety reasons (36%). Taken together, these results suggest that in patients with moderate or severe renal impairment (CLCR, <50 ml/min), the RBV adjusted dose should be ≤30% of the standard dose administered to patients with normal renal function (i.e., ≈200 to 400 mg daily).
Because of the extensive dose modifications in patients with moderate and severe renal impairment, the PK data were difficult to interpret and noncompartmental analysis was not considered an optimal method. Population PK modeling has the advantage of including information on individual dose and dosing time over the course of the study and therefore was considered more appropriate for this data set. Results of this analysis indicated that in patients with severe renal impairment (CKD stage 4) or ESRD (CKD stage 5), RBV 200 mg daily produced levels within 20% of the target concentration and therefore was considered an appropriate dose for these patient populations. However, among patients with moderate renal impairment (CKD stage 3), 200 mg daily produced insufficient RBV exposure (estimated 26% decrease from target value) and 200 mg twice daily produced simulated target levels that were increased by 48% and deemed too high. In contrast, alternating 200 mg or 400 mg daily resulted in RBV exposure only 11% greater than target and hence met the criterion of being ±20% of target concentrations derived from patients with normal renal function. It is recognized that alternate daily dosing of 200 mg or 400 mg will require patients with moderate renal impairment to adhere to instructions on alternate-day dosing changes; however, this dosing strategy would ensure sufficient RBV exposure for efficacy while minimizing potential toxicity. As a result of these data, the Food and Drug Administration (FDA) requested revision of the Copegus product label to reflect the dosing recommendations contained in Table 5.
Table 5.
Recommended ribavirin dosing in chronically HCV-infected patients with CKD
| Renal impairment and CLCR, ml/min | Renal severity | RBV dose (mg/day) |
|---|---|---|
| Normal (>80 ml/min) | None | 1,000/1,200 |
| CKD stage 3 (30–50 ml/min) | Moderate | 200 alternating with 400 |
| CKD stage 4 (<30 ml/min) | Severe | 200 |
| CKD stage 5 (dialysis) | ESRD | 200 |
There is limited experience with RBV in HCV-infected patients with ESRD receiving dialysis. A number of small studies using RBV plasma trough concentrations and/or hemoglobin measurements to guide therapy have shown average RBV doses between 133 and 200 mg daily to be optimal (16, 17, 22). The current study used RBV doses of 200 mg daily and showed plasma levels approximately 20% lower than those for patients with normal renal function. At this dose, RBV was well tolerated and generally safe, with dosage modifications being relatively infrequent (22%). As such, the current results suggest that RBV 200 mg once daily in patients with ESRD on dialysis is well tolerated and provides RBV exposure similar to that in patients with normal renal function.
In terms of RBV extraction following hemodialysis, our results suggest an extraction coefficient of approximately 50%. However, it is important to consider that dialysis clearance and extraction coefficient measure the ability of dialysis to remove drug from the blood but do not represent how effectively the drug is cleared from the body. Only free drug, in the circulation, is available for removal by hemodialysis. Drugs with large volumes of distribution are not readily removed by hemodialysis because of the small fraction of drug in the body that is present in the central compartment (23). RBV has a large apparent volume of distribution (>2,000 liters), which reflects the extensive distribution to sites outside the plasma (primarily intracellular) (24). Due to the large volume of distribution, RBV is not efficiently cleared by hemodialysis because only a very small portion of the total amount of drug in the body is available to be cleared by hemodialysis. This translates to a negligible effect of hemodialysis on the plasma exposure of RBV, similar to that observed in previous reports showing little removal by hemodialysis (25).
In conclusion, in HCV-infected patients with moderate or severe renal impairment receiving RBV at doses of 600 mg and 400 mg, respectively, RBV plasma exposure was approximately 30% higher and clearance 3- to 4-fold lower than that observed among patients with normal renal function. In addition, RBV was poorly tolerated in these patients, with more than 70% having to change or modify the dose due to adverse events. Subsequently, PK modeling and simulation were performed to identify the optimal RBV dosing regimen for patients with CKD (26), resulting in a change to the RBV U.S. product label to doses of 200 mg or 400 mg alternating daily for patients with moderate renal impairment and 200 mg daily for patients with severe renal impairment or ESRD.
ACKNOWLEDGMENTS
Support for third-party writing assistance for the manuscript, furnished by Andrew Luber, Health Interactions, was provided by Genentech, Inc., and F. Hoffmann-La Roche Ltd.
Footnotes
Published ahead of print 30 September 2013
REFERENCES
- 1. Chak E, Talal AH, Sherman KE, Schiff ER, Saab S. 2011. Hepatitis C virus infection in USA: an estimate of true prevalence. Liver Int. 31:1090–1101 [DOI] [PubMed] [Google Scholar]
- 2. Lavanchy D. 2009. The global burden of hepatitis C. Liver Int. 29(Suppl 1):74–81 [DOI] [PubMed] [Google Scholar]
- 3. Butt AA, Umbleja T, Andersen JW, Chung RT, Sherman KE. 2011. The incidence, predictors and management of anaemia and its association with virological response in HCV/HIV coinfected persons treated with long-term pegylated interferon alfa 2a and ribavirin. Aliment. Pharmacol. Ther. 33:1234–1244 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Perico N, Cattaneo D, Bikbov B, Remuzzi G. 2009. Hepatitis C infection and chronic renal diseases. Clin. J. Am. Soc. Nephrol. 4:207–220 [DOI] [PubMed] [Google Scholar]
- 5. Ghany MG, Strader DB, Thomas DL, Seeff LB. 2009. Diagnosis, management, and treatment of hepatitis C: an update. Hepatology 49:1335–1374 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Meyers CM, Seeff LB, Stehman-Breen CO, Hoofnagle JH. 2003. Hepatitis C and renal disease: an update. Am. J. Kidney Dis. 42:631–657 [DOI] [PubMed] [Google Scholar]
- 7. Welsch C, Jesudian A, Zeuzem S, Jacobson I. 2012. New direct-acting antiviral agents for the treatment of hepatitis C virus infection and perspectives. Gut. 61(Suppl 1):i36–i46 [DOI] [PubMed] [Google Scholar]
- 8. Schaefer EA, Chung RT. 2012. Anti-hepatitis C virus drugs in development. Gastroenterology 142:1340–1350.e1. 10.1053/j.gastro.2012.02.015 [DOI] [PubMed] [Google Scholar]
- 9. Lindahl K, Schvarcz R, Bruchfeld A, Ståhle L. 2004. Evidence that plasma concentration rather than dose per kilogram body weight predicts ribavirin-induced anaemia. J. Viral Hepat. 11:84–87 [DOI] [PubMed] [Google Scholar]
- 10. Reau N, Hadziyannis SJ, Messinger D, Fried MW, Jensen DM. 2008. Early predictors of anemia in patients with hepatitis C genotype 1 treated with peginterferon alfa-2a (40KD) plus ribavirin. Am. J. Gastroenterol. 103:1981–1988 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Maeda Y, Kiribayashi Y, Moriya T, Maruhashi A, Omoda K, Funakoshi S, Murakami T, Takano M. 2004. Dosage adjustment of ribavirin based on renal function in Japanese patients with chronic hepatitis C. Ther. Drug Monit. 26:9–15 [DOI] [PubMed] [Google Scholar]
- 12. Reddy KR, Shiffman ML, Morgan TR, Zeuzem S, Hadziyannis S, Hamzeh FM, Wright TL, Fried M. 2007. Impact of ribavirin dose reductions in hepatitis C virus genotype 1 patients completing peginterferon alfa-2a/ribavirin treatment. Clin. Gastroenterol. Hepatol. 5:124–129 [DOI] [PubMed] [Google Scholar]
- 13. Opravil M, Rodriguez-Torres M, Rockstroh J, Snoeck E, Chung RT, Tietz A, Torriani FJ. 2012. The dose-response relationship of peginterferon alfa-2a and ribavirin in the treatment of patients coinfected with HIV-HCV. HIV Clin. Trials 13:33–45 [DOI] [PubMed] [Google Scholar]
- 14. Snoeck E, Wade JR, Duff F, Lamb M, Jorga K. 2006. Predicting sustained virological response and anaemia in chronic hepatitis C patients treated with peginterferon alfa-2a (40KD) plus ribavirin. Br. J. Clin. Pharmacol. 62:699–709 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Bruchfeld A, Lindahl K, Ståhle L, Söderberg M, Schvarcz R. 2003. Interferon and ribavirin treatment in patients with hepatitis C-associated renal disease and renal insufficiency. Nephrol. Dial. Transplant. 18:1573–1580 [DOI] [PubMed] [Google Scholar]
- 16. Bruchfeld A, Lindahl K, Reichard O, Carlsson T, Schvarcz R. 2006. Pegylated interferon and ribavirin treatment for hepatitis C in haemodialysis patients. J. Viral Hepat. 13:316–321 [DOI] [PubMed] [Google Scholar]
- 17. van Leusen R, Adang RP, de Vries RA, Cnossen TT, Konings CJ, Schalm SW, Tan AC. 2008. Pegylated interferon alfa-2a (40 kD) and ribavirin in haemodialysis patients with chronic hepatitis C. Nephrol. Dial. Transplant. 23:721–725 [DOI] [PubMed] [Google Scholar]
- 18. Wade JR, Snoeck E, Duff F, Lamb M, Jorga K. 2006. Pharmacokinetics of ribavirin in patients with hepatitis C virus. Br. J. Clin. Pharmacol. 62:710–714 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Anderson B, Holford N. 2008. Mechanism-based concepts of size and maturity in pharmacokinetics. Annu. Rev. Pharmacol. Toxicol. 48:303–332 [DOI] [PubMed] [Google Scholar]
- 20. Lertora JJ, Rege AB, Lacour JT, Ferencz N, George WJ, VanDyke RB, Agrawal KC, Hyslop NE., Jr 1991. Pharmacokinetics and long-term tolerance to ribavirin in asymptomatic patients infected with human immunodeficiency virus. Clin. Pharmacol. Ther. 50:442–449 [DOI] [PubMed] [Google Scholar]
- 21. Morello J, Rodríguez-Novoa S, Jiménez-Nácher I, Soriano V. 2008. Usefulness of monitoring ribavirin plasma concentrations to improve treatment response in patients with chronic hepatitis C. J. Antimicrob. Chemother. 62:1174–1180 [DOI] [PubMed] [Google Scholar]
- 22. Rendina M, Schena A, Castellaneta NM, Losito F, Amoruso AC, Stallone G, Schena FP, DiLeo A, Francavilla A. 2007. The treatment of chronic hepatitis C with peginterferon alfa-2a (40 kDa) plus ribavirin in haemodialysed patients awaiting renal transplant. J. Hepatol. 46:768–774 [DOI] [PubMed] [Google Scholar]
- 23. Rowland M, Tozer T. 1995. Clinical pharmacokinetics: concepts and applications, 3rd ed, p 443–462 Lippincott, Williams and Wilkins, Baltimore, MD [Google Scholar]
- 24. Glue P. 1999. The clinical pharmacology of ribavirin. Semin. Liver Dis. 19(Suppl 1):17. [PubMed] [Google Scholar]
- 25. Kramer TH, Gaar GG, Ray CG, Minnich L, Copeland JG, Connor JD. 1990. Hemodialysis clearance of intravenously administered ribavirin. Antimicrob. Agents Chemother. 34:489–490 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Brennan B, Magnusson M, Wilkins J, Nieforth K, Solsky J, Witt C. 2012. Population pharmacokinetic (PK) analysis of ribavirin (Copegus) in chronic hepatitis C (CHC) patients with chronic kidney disease (CKD) leads to changes in recommended dosing, abstr 1093 47th Annu. Meet. Eur. Assoc. Study Liver, Barcelona, Spain [Google Scholar]