Raltegravir pharmacokinetics before and during treatment with ombitasvir, paritaprevir/ritonavir plus dasabuvir in adults with human immunodeficiency virus‐1 and hepatitis C virus coinfection: AIDS Clinical Trials Group sub‐study A5334s

Abstract

Aims

AIDS Clinical Trials Group study A5334s evaluated the pharmacokinetics of raltegravir before and during combined administration of ombitasvir, paritaprevir/ritonavir, plus dasabuvir (OBV/PTV/r + DSV) and weight‐based ribavirin in human immunodeficiency virus (HIV) and hepatitis C virus (HCV) coinfected adults. The pharmacokinetics of OBV/PTV/r + DSV during raltegravir coadministration were also characterized.

Methods

Adults living with HIV/HCV coinfection receiving steady‐state raltegravir (400 mg twice daily) with 2 nucleos(t)ide analogues were enrolled. Pharmacokinetics of raltegravir were assessed prior to HCV therapy, and 4 weeks later following initiation of OBV/PTV/r (25/150/100 mg) once daily + DSV (250 mg) twice daily. Geometric mean ratios (GMRs) and 90% confidence intervals (CIs) were used to compare the following: raltegravir pharmacokinetics with HCV therapy (week 4) vs before HCV therapy (week 0); OBV/PTV/r and DSV pharmacokinetics vs historical healthy controls; raltegravir pharmacokinetics at week 0 vs historical control adults living with HIV.

Results

Eight of 11 participants had decreased raltegravir exposures after initiation of HCV therapy. The GMRs (90% CI) for maximum concentration and area under the concentration–time curve of raltegravir with vs without HCV therapy were 0.68 (0.38–1.19) and 0.82 (0.58–1.17), respectively. Comparing OBV/PTV/r pharmacokinetics in healthy controls, A5334s study participants demonstrated generally lower maximum concentration and area under the concentration–time curve values by 41–82% and 4–73%, respectively. Raltegravir exposures tended to be higher in A5334s study participants compared to adults living with HIV.

Conclusions

The majority of participants' plasma raltegravir exposures were lower after initiation of HCV therapy in coinfected adults; however, confidence intervals were wide.

What is already known about this subject

• Prior pharmacokinetic drug‐interaction studies of raltegravir with ombitasvir, paritaprevir/ritonavir plus dasabuvir (OBV/PTV/r + DSV) conducted in healthy seronegative volunteers report increases in raltegravir exposures during coadministration.

• Despite increases in raltegravir exposure, the use of raltegravir with OBV/PTV/r + DSV is considered safe and does not require dose adjustments.

What this study adds

• During coadministration of raltegravir with OBV/PTV/r + DSV in people living with human immunodeficiency virus and hepatitis C virus coinfection, we observed unexpected decreases in raltegravir exposures in the majority of participants, which are not considered clinically significant.

• Some drug interactions in people with human immunodeficiency virus/hepatitis C virus coinfection may behave differently compared to healthy individuals.

1. INTRODUCTION

Globally, there are >2 million people living with both human immunodeficiency virus (HIV)‐1 and hepatitis C virus (HCV) infections.1 Mortality rates due to liver‐related complications are >3 times greater among people living with HIV than the general population, and chronic infection with HCV has historically been a major source of liver disease due to shared routes of transmission of HIV and HCV, and reduced spontaneous clearance of HCV in those with HIV‐1 infection.2, 3, 4, 5 The arrival of all‐oral direct acting antiviral agents (DAAs), however, has led to high cure rates of chronic HCV infection, and these regimens are equally efficacious and tolerable in people with HIV/HCV coinfection.6 Therefore, treatment guidelines recommend that all HIV/HCV coinfected persons be considered candidates for curative HCV treatment, and that they should generally be treated the same as persons without HIV infection except with regard to the consideration of drug–drug interactions between antiretrovirals and DAAs.7, 8, 9

The all‐oral DAA combination regimen of ombitasvir, paritaprevir (identified by AbbVie and Enanta Pharmaceuticals) with ritonavir, plus dasabuvir (OBV/PTV/r + DSV) combined with or without ribavirin has demonstrated >90% sustained virological response in people with HIV and HCV genotype 1 coinfection.10, 11, 12 However, there are concerns of specific drug–drug interactions with the use of OBV/PTV/r + DSV and certain HIV antiretrovirals due to shared metabolic pathways. For example, the use of darunavir with ritonavir or cobicistat is not recommended in combination with OBV/PTV/r + DSV due to decreases in darunavir plasma trough concentrations.8, 9, 11, 13

The impact of OBV/PTV/r + DSV coadministration on raltegravir exposures is not clear. There have been some reports demonstrating increases in exposures of raltegravir when administered with OBV/PTV/r + DSV.14, 15 In a study of healthy adults, raltegravir area under the plasma concentration–time curve (AUC_τ) and trough concentration increased by an average of 134% and 100%, respectively, during coadministration with OBV/PTV/r + DSV.14 Similarly, a case report of an adult liver transplant patient with HIV/HCV coinfection reported a 265% increase in raltegravir AUC_τ during treatment with OBV/PTV/r + DSV.15 In contrast, a single‐centre observational study of HIV/HCV coinfected adults reported no significant differences in raltegravir trough concentrations either with or without OBV/PTV/r ± DSV.16

Raltegravir is metabolized by hepatic glucuronidation primarily by uridine glucuronsyl‐transferase 1A1 (http://www.guidetopharmacology.org/GRAC/ObjectDisplayForward?objectId=2990) isoenzyme, which is inhibited by OBV, PTV, and DSV, and is induced by ritonavir.13, 17, 18 However, there have been no studies designed to prospectively assess the full pharmacokinetic profile of raltegravir when given with OBV/PTV/r + DSV in people living with HIV/HCV coinfection. Instead, prior studies have either enrolled only healthy volunteers, which ignore possible disease‐specific factors contributing to pharmacokinetic variability (e.g. concomitant medications, comorbidities, variable adherence), or have been limited in number of subjects or pharmacokinetic samples analysed. Although no dose adjustments are currently recommended for raltegravir when administered with OBV/PTV/r + DSV, there remains uncertainty whether raltegravir exposures are changed in people with HIV/HCV coinfection treated with the DAA regimen.

The aim of this study was to prospectively assess the pharmacokinetics of raltegravir in adults with HIV/HCV coinfection in the presence and absence of OBV/PTV/r + DSV. As a secondary aim, the pharmacokinetics of OBV, PTV/r and DSV were evaluated while coadministered with raltegravir for comparison to pharmacokinetic parameters reported in prior studies in HIV and HCV seronegative individuals (i.e. healthy volunteers), and in people with HCV monoinfection.

2. METHODS

2.1. Study design

AIDS Clinical Trials Group study A5334s was an intensive pharmacokinetic substudy conducted with participants who enrolled in parent study A5329 (NCT02194998). Study A5329 was a nonrandomized, open‐label, phase 2 study of interferon‐free HCV therapy for 24 or 12 weeks in sequentially enrolled cohorts of participants with HIV‐1 and HCV (genotypes 1a or 1b) coinfection and who were taking protocol‐defined antiretroviral regimens. HCV therapy consisted of coformulated ombitasvir/paritaprevir/ritonavir (12.5/75/50 mg) 2 tablets once daily, plus dasabuvir (250 mg) twice daily and weight‐based ribavirin (1000 or 1200 mg) in 2 divided doses twice daily. Qualifying antiretroviral regimens for participation in A5334s included raltegravir (400 mg) or darunavir/ritonavir (600/100 mg) every 12 h, plus tenofovir disoproxil fumarate (300 mg), and either emtricitabine (200 mg) or lamivudine (300 mg) every 24 h. Participants were required to be on the antiretroviral regimen for at least 14 days prior to study enrolment. The darunavir/ritonavir group closed early before reaching the target accrual, enrolling 1 of 12 participants, as there was a low likelihood that a sufficient number of participants would enrol prior to the closure of the A5329 study.

Blood samples were collected on study entry (i.e. week 0) for determination of HIV antiretroviral plasma concentrations in the absence of HCV therapy, and again 4 weeks later (i.e. week 4) for determination of HIV antiretroviral and HCV DAA plasma concentrations during coadministration. Study participants were required to keep a written record of dates and times of the last 3 doses of raltegravir or darunavir/ritonavir prior to the first pharmacokinetic visit for review by study site staff to ensure adherence. Adherence to antiretrovirals and HCV medications were similarly recorded for review at the week 4 visit. On the morning of each intensive pharmacokinetic sampling visit, participants were provided a standardized breakfast prior to sample collections. Breakfast consisted of a moderate‐fat meal (approximately 500–600 kCal, 18–21 g fat). Samples were collected by venipuncture prior to dosing (i.e. predose), and 1, 2, 3, 4, 5, 6, 8, 10 and 12 h after dosing. Lunch, dinner and an evening snack were provided approximately 4, 9 and 12 h after dosing, respectively.

2.2. Measurement of raltegravir and DAAs in plasma

Plasma concentrations of raltegravir were determined using a ultra‐performance liquid chromatography‐tandem mass spectrometry method validated in accordance with the Food and Drug Administration Bioanalytical Method Validation Guidance and approved by Division of AIDS Clinical Pharmacology Quality Assurance Method Review Program.19 Raltegravir was extracted from EDTA plasma via solid phase extraction using Waters Oasis HLB 30‐mg 96‐well plates. Chromatography was conducted with an Acquity BEH C18 2.1 × 50 mm, 1.7‐μm column and isocratic mobile phase of 35:65:0.1 (water:methanol:formic acid, v/v/v) at 0.4 mL/min. Standard curves ranged from 10 to 4000 ng/mL. Inter‐assay and intra‐assay accuracy and precision was <8.41% for quality control samples and <14.5% at the lower limit of quantitation.

Plasma concentrations of DAAs and ritonavir were determined simultaneously using a validated liquid chromatography method with tandem mass spectrometric detection. Ombitasvir, paritaprevir, dasabuvir, dasabuvir M1 metabolite and ritonavir were extracted from plasma samples using protein precipitation coupled with online solid phase extraction. Sample extracts were loaded on a solid phase extraction cartridge (Chromolith Guard Cartridge, RP‐18e, 10 × 4.6 mm) using 20/80 (v/v) methanol/water at a flow rate of 3.0 mL/min. Gradient elution starting with 0.1% formic acid in 40/60 (v/v) acetonitrile/water and increasing to 0.1% formic acid in 70/30 (v/v) acetonitrile/water over the course of 1.6 minutes at a total flow rate of 0.8 mL/min was utilized to elute the analytes off the solid phase extraction cartridge and onto the analytical column (Agilent Zorbax SB‐C18, 2.1 × 50 mm, 3.5 μm) for separation. Detection was achieved using a Sciex API5500 mass spectrometer in positive ion multiple‐reaction monitoring mode. Ombitasvir‐D₁₃, paritaprevir‐D₈, dasabuvir‐¹³CD₃, dasabuvir M1‐D₆ and ritonavir‐D₅ were used as the internal standards for the assay of ombitasvir, paritaprevir, dasabuvir, dasabuvir M1 metabolite and ritonavir, respectively. Human plasma with K₂ EDTA was used to prepare the calibration standards with a curve range of 0.472–311, 0.617–407, 4.59–3030, 4.85–3200 and 5.03–3320 ng/mL for ombitasvir, paritaprevir, dasabuvir, dasabuvir M1 metabolite and ritonavir, respectively. The accuracy (expressed as percent bias) at the lower limit of quantitation for ombitasvir, paritaprevir, dasabuvir, dasabuvir M1 metabolite and ritonavir was 3.9, 2.2, 2.4, 6.1 and 2.6%, respectively. The accuracy of the quality control samples prepared at concentrations distributed throughout the calibration curve range were between 0.7 and 4.7% for ombitasvir, −1.0 and 4.5% for paritaprevir, −0.2 and 3.3% for dasabuvir, −1.8 and 5.9% for dasabuvir M1 metabolite, and −4.3 and −0.1% for ritonavir. Their precision (coefficient of variation) for ombitasvir, paritaprevir, dasabuvir, dasabuvir M1 metabolite, and ritonavir was ≤5.3, ≤7.0, ≤3.5, ≤4.4 and ≤4.2%, respectively.

2.3. Pharmacokinetic analysis

Noncompartmental pharmacokinetic analysis of the plasma concentration‐time data for each drug at steady‐state was first performed using PKSolver (version 2.0, China Pharmaceutical University, Nanjing, China) and then verified using Phoenix WinNonlin (version 7.0, Certara, St. Louis, MO, USA).20 Pharmacokinetic parameters of interest included the maximum observed plasma concentration (C_max), the time to reach maximum concentration (t_max), the trough plasma concentration (C₂₄ for once daily [QD] drugs and C₁₂ for twice daily [BID] drugs), and the area under the concentration–time curve for the dosing interval (AUC_τ; AUC₂₄ for QD drugs or AUC₁₂ for BID drugs) determined using the linear trapezoidal rule. In calculating the AUC₂₄ for QD drugs (i.e. PTV/r and OBV), the predose concentration (C₀) was also used as C₂₄. Apparent clearance (CL/F) was calculated as dose divided by AUC_τ.

2.4. Statistical analysis

Primary objective. As mentioned previously, the darunavir/ritonavir group closed before reaching the target accrual of participants due to lack of enrolment. Thus, the primary objective was to estimate participant‐level plasma pharmacokinetic parameters of raltegravir for comparison before and after dosing to steady‐state of HCV DAA therapy. Descriptive statistics were calculated for pharmacokinetic parameters during each study period. Log‐transformed pharmacokinetic parameters of raltegravir with and without coadministration of HCV therapy were compared at the within‐participant level using the Wilcoxon signed‐rank test, with a type I error rate set to 5%. As raltegravir exposure was expected to increase with coadministration of the HCV DAA therapy, the change in C_max parameter was considered the outcome measure of greatest interest and was the basis of the study design.

2.4.1. Sample size considerations

Substudy A5334s planned to enrol 12 participants each in the raltegravir‐ and darunavir/ritonavir‐based antiretroviral treatment groups, in order to yield 10 evaluable participants per regimen with complete pharmacokinetic data on both study visits. The sample size of 10 analysis‐eligible participants was chosen in part based on the feasibility consideration that no more than 20–25% of the 50 individuals in each parent study treatment arm would be willing to participate in this pharmacokinetic substudy. For the within‐participant test of difference, a sample size of 10 participants would provide power of at least 89.4% to detect an increase in raltegravir C_max of 50% or larger, assuming the interindividual percent coefficient of variation for raltegravir C_max would be 70%. Bioequivalence was also assessed by comparing 90% confidence intervals around geometric mean ratios of pharmacokinetic parameters to no‐effect boundaries of 80–125%. Due to the high variability in raltegravir pharmacokinetics, if the true alteration in raltegravir C_max due to coadministration of OBV/PTV/r + DSV was an increase of 50%, the probability of declaring inequivalence was poor (38.2%); if the true alteration was small (−20 to +20%), the probability was high that bioequivalence tests would be inconclusive.

2.4.2. Secondary and exploratory objectives

Additional objectives included estimating pharmacokinetic parameters of ombitasvir, paritaprevir, ritonavir, dasabuvir and dasabuvir M1 metabolite while receiving antiretroviral therapy, and to compare pharmacokinetics with those reported in prior studies in healthy volunteers. The plasma pharmacokinetics of ombitasvir, paritaprevir, ritonavir and dasabuvir were compared to summary statistics for parameters reported in prior studies of healthy participants. These data were obtained from a drug–drug interaction study in healthy participants.21 That study enrolled 2 cohorts of 12 participants each to evaluate steady‐state pharmacokinetics of ombitasvir/paritaprevir/ritonavir (25/150/100 mg QD) and dasabuvir (250 mg BID) with and without either (i) dolutegravir or (ii) abacavir and lamivudine under nonfasting conditions. Geometric mean ratios and confidence intervals were calculated for AUC_τ, C₀, C_max as reported during study periods without (i) dolutegravir (healthy control group 1) and without (ii) abacavir plus lamivudine (healthy control group 2). As an additional exploratory objective, week 0 raltegravir exposures in A5334s were compared to historical data from studies evaluating raltegravir plasma pharmacokinetics in HIV monoinfected patients.22, 23, 24

All statistical tests and confidence intervals are 2‐sided, with type I error rates set to 5% and confidence interval coverage probabilities of 95%, except for confidence interval coverage of 90% around geometric mean ratios. Tests with P‐values ≤.05 or lower are considered statistically significant. Statistical analyses were performed using SAS version 9.4 (SAS Institute Inc., Cary, NC, USA).

3. RESULTS

3.1. Study participants

Eleven adults with HIV/HCV coinfection receiving raltegravir‐based antiretroviral therapy were enrolled in A5334s from September 2015 through April 2016 at 7 US clinic sites. All participants completed the full pharmacokinetic substudy. In assessing adherence to raltegravir and the DAAs, no doses were missed in the 3 days leading up to the study day visits according to self‐reported written records. The median (range) time between the first and second pharmacokinetic study days visits was 32 (28–35) days. Baseline characteristics are displayed in Table 1. The majority of participants were white (55%), male (82%) and without cirrhosis (73%).

Table 1.

Baseline characteristics of A5334s participants on raltegravir‐based antiretroviral therapy (n = 11)

	Mean ± standard deviation or n (%)
Age, y	44.8 ± 8.5
Sex
Male	9 (82%)
Female	2 (18%)
Race
White	6 (55%)
Black or African American	3 (27%)
Unknown	2 (18%)
Ethnicity
Not Hispanic or Latino	6 (55%)
Hispanic or Latino	5 (45%)
Cirrhosis status
Noncirrhotic	8 (73%)
Cirrhotic	3 (27%)
History of intravenous drug use
Never	5 (45%)
Previously	6 (55%)
Hepatitis C virus genotype
1a	9 (82%)
1b	2 (18%)
Weight, kg	88.8 ± 14.9
Body mass index, kg m–2	28.9 ± 4.6
Ribavirin dosing divided twice daily
1000 mg	2 (18%)
1200 mg	9 (82%)
Nucleos(t)ide reverse transcriptase inhibitors TDF/FTC	11 (100%)

TDF, tenofovir disoproxil fumarate; FTC, emtricitabine

3.2. Pharmacokinetic results

3.2.1. Raltegravir pharmacokinetics

Summaries of participant‐specific plasma pharmacokinetic parameter estimates for raltegravir pre‐HCV therapy and during HCV therapy are provided in Table 2 and concentration–time summary profiles are displayed in Figure 1. The geometric mean raltegravir plasma C_max was 32% lower when coadministered with HCV therapy but differences were not statistically significant and the wide 90% confidence interval reflects the variability in the individual changes in C_max (Table 2). Similarly, while raltegravir plasma AUC_τ, C₀ and C₁₂ tended to be lower when coadministered with HCV therapy, compared to without HCV therapy, confidence intervals were wide. No participants in this cohort experienced HIV‐1 virologic failure as measured and defined by the parent A5329 study (increase in HIV‐1 RNA ≥ 200 copies/mL at any time after study entry until 4 weeks after permanent discontinuation of HCV therapy). Therefore, these changes in raltegravir exposure were unlikely to be clinically significant.

Table 2.

Summary statistics of raltegravir pharmacokinetic parameters at weeks 0 and 4, (n = 11)

Parametera	Week 0 (pre‐HCV therapy)	Week 4 (with HCV therapy)	P‐valueb	GMR week 4/week 0 (90% CI)
C max, ng mL−1	2726 (62%) [1562, 4758]	1840 (73%) [1032, 3283]	.15	0.68 (0.38, 1.19)
t max, h median (range)	5 (1–6)	4 (2–10)	–	–
AUC τ, ng*h mL−1	10407 (52%) [6924, 15642]	8575 (64%) [5388, 13646]	.21	0.82 (0.58, 1.17)
C 0, ng mL−1 c	347 (113%) [125, 961]	221 (94%) [128, 382]	.16	0.60 (0.20, 1.79)
C 12, ng mL−1	191 (228%) [79, 462]	148 (97%) [86, 257]	.90	0.78 (0.36, 1.69)
CL/F, L h−1	38.4 (64%) [25.6, 57.8]	46.6 (72%) [29.3, 74.2]	.24	1.21 (0.86, 1.72)

^aValues are geometric mean (% coefficient of variation) [95% confidence interval] unless otherwise indicated.

^bWilcoxon signed rank test.

^cOne participant did not have predose concentration (C₀) collected at week 4.

AUC_τ, area under plasma–concentration time curve to infinity; CI, confidence interval; C₀, predose plasma concentration at time 0 h; C₁₂, postdose plasma concentration at time 12 h; C_max, maximum observed plasma concentration; CL/F, apparent oral clearance; GMR, geometric mean ratio; HCV, hepatitis C virus; t_max, time to maximum observed plasma concentration.

Figure 1.

Geometric mean (95% confidence intervals) plasma concentration vs time profile for raltegravir before (week 0) and during (week 4) treatment with ombitasvir, paritaprevir/ritonavir and dasabuvir, plus ribavirin. Inset displays data plotted on a semilogarithmic scale

Raltegravir plasma C_max and AUC_τ values were increased in only 3 (27%) participants (all noncirrhotic) when HCV therapy was coadministered, whereas 8 individuals had decreases in C_max and AUC_τ values (Figure 2). Among participants with decreases in raltegravir exposure during HCV therapy, the average change in plasma raltegravir C_max and AUC_τ values from week 0 were − 44% and − 36%, respectively. Among participants with increasing values during HCV therapy, the average change in plasma raltegravir C_max and AUC_τ values were 125% and 94%, respectively.

Figure 2.

Participant‐level changes in steady‐state raltegravir A, maximum concentration (C_max) and B, area under the concentration–time curve (AUC) values before (week 0) and during (week 4) treatment with ombitasvir, paritaprevir/ritonavir and dasabuvir, plus ribavirin. Blue circles represent those participants who experienced a decrease in C_max or AUC from week 0 to week 4, while green squares represent individuals with increases

Raltegravir plasma pharmacokinetic parameters measured at week 0 in the A5334s HIV/HCV coinfected population were higher compared to historical data from HIV monoinfected patients; however, the confidence intervals were wide because of the substantial variability associated with raltegravir pharmacokinetics. The point estimates for C₀ and C₁₂ observed in A5334s were significantly higher (geometric mean ratios of 5.51 [90% CI, 2.22–13.66] and 3.03 [90% CI, 1.34–6.82], respectively) than those reported in 1 of the historical raltegravir pharmacokinetic studies (Table S1).

3.2.2. OBV/PTV/r + DSV plasma pharmacokinetics

Pharmacokinetic parameters for ombitasvir, paritaprevir, ritonavir, dasabuvir and the M1 metabolite of dasabuvir from the HIV/HCV coinfected study participants while receiving raltegravir are shown in Table 3.

Table 3.

Summary statistics of ombitasvir/paritaprevir/ritonavir, dasabuvir and dasabuvir M1 metabolite pharmacokinetic parameters (n = 11)

Parametera	Ombitasvir	Paritaprevir	Ritonavir	Dasabuvir	M1
C max, ng mL−1	114 (33%) [91, 143]	640 (133%) [264, 1556]	866 (60%) [574, 1308]	582 (40%) [449, 754]	269 (72%) [190, 380]
t max, h median (range)	5 (4, 8)	4 (3, 6)	5 (3, 8)	3 (1, 6)	4 (3, 6)
AUC τ, ng*h mL−1	1328 (27%) [1111, 1587]	4575 (114%) [2049, 10 213]	6820 (71%) [4511, 10 311]	3967 (35%) [3137, 5016]	1670 (66%) [1203, 2318]
C 0, ng mL−1	28 (36%) [22, 35]	23 (138%) [10, 53]	51 (172%) [24, 108]	145 (38%) [112, 188]	46 (46%) [32, 64]
C 12, ng mL−1	–	–	–	153 (38%) [115, 202]	56 (51%) [41, 76]
CL/F, L h−1	18.8 (26%) [15.8, 22.5]	32.8 (119%) [14.7, 73.2]	14.7 (51%) [9.7, 22.2]	63.0 (34%) [49.8, 79.7]	–

^aValues are geometric mean (% coefficient of variation) [95% confidence interval] unless otherwise indicated.

AUC_τ, area under plasma concentration–time curve to infinity; CI, confidence interval; C₀, predose plasma concentration at time 0 h; C₁₂, postdose plasma concentration at time 12 h; C_max, maximum observed plasma concentration; CL/F, apparent oral clearance; t_max, time to maximum observed plasma concentration.

Geometric mean ratio point estimates with 90% confidence intervals were constructed to compare pharmacokinetic parameters measured in HIV/HCV coinfected A5334s study participants vs historical control data from healthy volunteers (Figure 3). For paritaprevir, decreased C_max values were observed in HIV/HCV coinfected A5334s participants compared to healthy historical control values by 69% in healthy control group 1, and 82% in healthy control group 2. Ritonavir C_max, as well as AUC_τ values, were decreased in coinfected participants compared to healthy controls by as much as 59 and 43%, respectively. Dasabuvir C_max, C_0, and AUC_τ were decreased in coinfected individuals in comparison to healthy controls (55, 65 and 57% respectively, for study group 1). None of the participants from the A5334s cohort experienced HCV virologic failure as defined and measured in A5329 (two consecutive HCV RNA measurements of >1 log 10 IU/mL; or, failure to achieve HCV RNA <lower limit of quantitation by week 6; or, 2 consecutive HCV RNA measurements ≥lower limit of quantitation). The pharmacokinetic parameters of ombitasvir measured in HIV/HCV coinfected A5334s study participants did not differ significantly compared to historical healthy control data.

Figure 3.

Estimated geometric mean ratios of hepatitis C virus direct acting antiviral agent plasma pharmacokinetics when comparing A5334s data (numerator) to healthy control data (denominator) from 2 external control groups who were human immunodeficiency virus‐1 and hepatitis C virus seronegative reported by Khatri et al., 2016.²¹ Healthy external control group 1 data were individuals before receiving dolutegravir, and healthy external group 2 data were individuals before receiving abacavir plus lamivudine. CI, confidence interval

4. DISCUSSION

In this pharmacokinetic study, we report no statistically significant differences of raltegravir plasma pharmacokinetics in the absence vs presence of OBV/PTV/r + DSV combined with ribavirin in adults living with HIV/HCV coinfection. These results are consistent with the findings of a single‐centre observational study reporting no differences in raltegravir plasma trough concentrations before and during treatment with OBV/PTV/r + DSV among 14 patients with HIV/HCV coinfection.16 However, contrary to expectations based on a prior drug–drug interaction study in healthy individuals, we observed decreases in raltegravir exposures during coadministration with HCV therapy in the majority of our study cohort.14 Raltegravir C_max values decreased in 8 out of 11 study participants at week 4, with changes from baseline ranging from approximately −1 to −90% (mean, −44%). Magnitudes of change in AUC_τ were similar.

Raltegravir is metabolized by hepatic glucuronidation primarily by UGT1A1. Inflammation associated with chronic HCV infection accelerates hepatocyte damage, and could affect liver uptake and reduce drug metabolizing enzyme activity, resulting in decreased clearance of drugs that undergo hepatic metabolism.25, 26 HCV infection rapidly clears from the plasma and liver with DAA therapy.27, 28 In another substudy of A5329, the early plasma and intrahepatic kinetics of HCV were assessed over the first week of starting OBV/PTV/r + DSV. Seven days after treatment initiation, mean plasma HCV RNA decreased by −4.1 log₁₀ IU/mL from baseline, and >90% of HCV‐infected hepatocytes were cleared upon measurement of intrahepatic HCV RNA in liver biopsies.29 This rapid and marked reduction in infected hepatocytes with DAA therapy may prompt a swift cascade towards normalization of hepatic function, a reduction in inflammation, and ultimately restoration of metabolizing enzymes. In liver transplant recipients treated for recurrent HCV infection with DAAs, significant decreases in liver transaminase levels and increases in maximal liver function capacity (measured via the LiMAx test) were observed within the first 4 weeks of treatment.30 Furthermore, plasma concentrations of tacrolimus (a CYP3A4 substrate) and mycophenolate acid (a UGT substrate) are known to decrease within weeks after starting DAA therapy, possibly as a result of the normalization of drug metabolizing enzymes.31, 32 Altogether, our findings of raltegravir exposure changes may reflect the rapid clearance of HCV induced by potent DAAs, and thus, the initial normalization of hepatocyte function including restoration of drug metabolizing enzyme activities.

Also noteworthy is that raltegravir exposures were highest in people with HIV/HCV coinfection before receiving HCV DAA therapy, and lowest in comparison to people with HIV‐1 monoinfection (external data). Elevated antiretroviral concentrations in people with HIV and viral hepatitis compared to those with HIV monoinfection have been previously reported.33, 34 Raltegravir is a substrate of the efflux transporter http://www.guidetopharmacology.org/GRAC/ObjectDisplayForward?objectId=768, which is expressed at the apical membrane of enterocytes and the bile canalicular surface of hepatocytes. Viral infections and associated inflammatory responses and tissue damage are known to alter gene expression of P‐gp.35 For instance, in the presence of HCV‐induced cirrhosis, hepatic protein expression of P‐glycoprotein is downregulated.36 In contrast, the expression of P‐gp in the upper intestinal tract is significantly higher in people living with HIV while receiving antiretroviral therapy compared to those without HIV infection.37 Increases or decreases in the expression of P‐gp could lead to lower or higher concentrations of P‐gp substrates, respectively, which may explain some of the observed differences in raltegravir concentrations across patient populations and time points. However, the effects of HIV/HCV coinfection and HCV eradication with DAAs on the expression of drug metabolizing enzymes and transporters over time warrants further investigation.

From a drug–drug interaction perspective, a strong effect on UGT1A1 is required for a significant drug interaction with raltegravir to be evident.38 Ombitasvir, paritaprevir and dasabuvir inhibit UGT1A1 but with varying degrees of inhibitory potencies.13 In a prior drug interaction study of 12 healthy, HIV‐1 and HCV seronegative volunteers, raltegravir C_max and AUC_τ values were increased approximately 130% when coadministered with OBV/PTV/r + DSV; however, administration with the 2D regimen (PTV/r + OBV without DSV), resulted in only a 22% increase in the raltegravir AUC.14 This difference was likely due to weaker UGT1A1 inhibition by paritaprevir (50% inhibitory concentration [IC₅₀] 3.6 μM), and ombitasvir (IC₅₀ 2.1 μM), compared with dasabuvir (IC₅₀ 0.9 μM).39 Paritaprevir is a weak competitive inhibitor of UGT1A1 and is unlikely to exhibit inhibitory effects at clinically relevant concentrations.40 Ombitasvir is a weak‐to‐moderate inhibitor of UGT1A1 with minor drug interaction potential. Dasabuvir is the most potent inhibitor of UGT1A1, and has the potential to alter the disposition of UGT1A1 substrates.41 Although our study included dasabuvir as part of the HCV DAA treatment regimen, differences in dasabuvir concentrations could partially explain the observed discordant differences in raltegravir pharmacokinetic changes. Overall, dasabuvir plasma exposures appeared to be considerably lower in our HIV/HCV coinfection population compared to values reported in seronegative individuals, perhaps diminishing the expected interaction involving UGT1A1. Furthermore, average dasabuvir plasma C_max and AUC values were 1.4‐fold and 1.2‐fold higher, respectively, among the 3 participants whose raltegravir exposures increased at week 4, compared to those whose raltegravir exposures decreased. Thus, observed increases and decreases in raltegravir exposures may have been driven by varying dasabuvir exposure and inhibition of UGT1A1.

In our coinfected population, paritaprevir, ritonavir and dasabuvir plasma exposures were generally lower than in cross‐study comparisons with healthy volunteer external groups. The greatest difference was a 69–82% decrease in paritaprevir C_max geometric mean ratios in the A5334s study population. Dasabuvir and ritonavir C_max values were as much as 55 and 59% reduced, respectively. Raltegravir does not inhibit or induce metabolic enzymes or drug transporters known to affect the disposition of OBV/PTV/r + DSV. Thus, it is unlikely that any differences in DAA exposures between populations were due to metabolic‐ or transporter‐mediated drug interactions with raltegravir. Rather, physiological factors affecting drug absorption, transport, metabolism and elimination may contribute to the pharmacokinetic differences noted between HIV/HCV coinfected and seronegative populations. In particular, the noted changes in C_max values reflect a reduction in the extent of drug absorption, possibly due to differences in intestinal drug absorption processes. For instance, ritonavir and all of the HCV DAAs are substrates of P‐gp and http://www.guidetopharmacology.org/GRAC/ObjectDisplayForward?objectId=792 transporters that vary in expression in the liver and gastrointestinal tract depending on liver disease and inflammatory stimuli.35

Our study had several limitations. Although raltegravir is a recommended first‐line antiretroviral according to treatment guidelines, its use has diminished in favour of newer integrase strand transfer inhibitors. Likewise, the use of OBV/PTV/r + DSV for the treatment of HCV has been supplanted by newer treatment regimens that have shorter courses of treatment duration and less drug–drug interaction potential. Another limitation to our study was the reliance on external historical data as comparators for raltegravir pharmacokinetics in HIV monoinfected patients, and pharmacokinetics of DAAs in healthy individuals. Although we made every effort to identify studies that matched the study design of A5334s (i.e. similar dosing and collection of intensive samples), inherent differences across studies may limit the validity of some observations regarding the pharmacokinetic differences across study populations. Finally, we were unable to enrol the targeted sample size for individuals receiving darunavir/ritonavir‐based antiretroviral therapy; however, the safety and pharmacokinetics of OBV/PTV/r + DSV with darunavir/ritonavir have been well‐established.11 Furthermore, in the parent A5329 study, longitudinal sparse pharmacokinetic samples are collected and will be analysed from all participants, thus providing an opportunity to assess antiretroviral (and DAA) concentrations over time to understand how drug exposures may change with DAA therapy and HCV eradication.

In conclusion, the present study did not find significant elevations in raltegravir plasma pharmacokinetics when administered with OBV/PTV/r + DSV and ribavirin in adults living with HIV/HCV coinfection. Rather, we observed decreases in raltegravir C_max and AUC values in the majority of study participants; however, raltegravir has a wide therapeutic index and these changes were not associated with virologic failure. Our findings provide evidence that expected drug–drug interactions in populations with HIV/HCV may not be the same as those observed in studies with healthy participants, and exposures to HIV antiretrovirals and HCV antivirals may be altered in coinfected populations.

CONTRIBUTIONS

Conceived and designed the study: C. Venuto, S. Rosenkranz, M. Sulkowski, D. Wyles, D. Cohen, B. Alton‐Smith, G. Morse.

Data collection: all authors.

Analysed the data: C. Venuto, Y. Cramer, S. Rosenkranz, J. Schmidt.

Wrote the paper: C. Venuto, Y. Cramer, S. Rosenkranz, J. Schmidt.

Reviewed and approved the paper: all authors.

COMPETING INTERESTS

M.S. is an AbbVie Advisory Board member and receives AbbVie research support to Johns Hopkins University; M.S. is a Gilead Advisory Board member and receives Gilead research support to Johns Hopkins University.

D.E.C. and J.S. are employees of AbbVie and own AbbVie stock.

Supporting information

TABLE S1 Estimated geometric mean ratios for raltegravir plasma pharmacokinetics when comparing A5334s week 0 parameters (numerator) to external control data of adults with human immunodeficiency virus monoinfection (denominator).

Venuto CS, Cramer YS, Rosenkranz SL, et al. Raltegravir pharmacokinetics before and during treatment with ombitasvir, paritaprevir/ritonavir plus dasabuvir in adults with human immunodeficiency virus‐1 and hepatitis C virus coinfection: AIDS Clinical Trials Group sub‐study A5334s. Br J Clin Pharmacol. 2020;86:132–142. 10.1111/bcp.14148

DATA AVAILABILITY STATEMENT

The authors confirm that all data underlying the findings are fully available without restriction. Due to ethical restrictions, study data are available upon request from sdac.data@sdac.harvard.edu with the written agreement of the AIDS Clinical Trials Group.