The pharmacokinetic profile of raltegravir-containing antiretroviral therapy in HIV-infected individuals over 60 years of age

Abstract

Background:

Antiretroviral safety and efficacy and may differ in older versus younger HIV-infected patients. The objective of this study was to assess the pharmacokinetic (PK) profile in older HIV-infected subjects (>60 years) switching combination antiretroviral therapy (cART) to a raltegravir (RAL) containing regimen.

Methods:

Nineteen HIV-infected patients over 60 years of age on effective cART (HIV-RNA < 50 copies/ml) were enrolled in this prospective 24-week study. On day 1, patients switched to tenofovir/emtricitabine (245/200 mg once daily) and RAL (400 mg twice daily). On day 28, intensive PK sampling was undertaken in a fasted state and RAL plasma concentrations determined. Neurocognitive function was assessed at baseline and week 24 using a neuropsychological battery. RAL PK parameters were compared to those of two younger historical HIV-infected control groups that received twice-daily RAL co-administered with darunavir/ritonavir (DRV/r) 800/100 once daily by nonlinear mixed effects modelling.

Results:

In HIV-infected subjects over the age of 60 (mean ± SD age: 66 ± 3.4 years, n = 19) switching to a RAL containing regimen, we observed no safety concerns, no plasma virological rebounds, and no differences in RAL apparent oral clearance when compared to younger HIV-infected populations (mean ± SD age: 41 ± 9.2 years, n = 38) based on population pharmacokinetic analysis. After 24 weeks of study therapy a decline in cognitive function was observed [change in (SD) global score of (0.91 (1.3), P = 0.018].

Conclusions:

No significant changes in RAL exposure associated with age were observed.

Keywords:

• Raltegravir
• Pharmacokinetics
• Aging

Introduction

Life-expectancy for people living with HIV (PLWH) has dramatically increased leading to an ageing HIV-infected population receiving life-long combination antiretroviral therapy (cART).^1–4

Comorbidities such as cardiovascular, metabolic, renal, bone disease, and cognitive decline become more prevalent with age.^3,5,6 The management of these comorbidities in PLWH is complex involving the use of multiple medications in conjunction with antiretroviral therapy.^7–9 Important pharmacokinetic changes occurring with age include reductions in renal and hepatic clearance of drugs and an increase in the volume of distribution of lipid soluble drugs.¹⁰ These changes lead to a potential prolongation of the elimination half-life and increases in plasma drug exposure, thereby increasing the risk of drug-associated toxicities. These effects may occur in older HIV-infected individuals on cART.

As the HIV integrase inhibitor raltegravir is not a substrate for the CYP450 iso-enzymes, it represents an attractive treatment option for older HIV-infected patients due to a reduced likelihood of drug–drug interactions.^11,12 The aim of this study was to assess the plasma pharmacokinetic profile, safety, and antiviral efficacy of raltegravir in HIV-infected subjects aged 60 years of age or older.

Methods

This prospective pharmacokinetic study was conducted at St Mary's Hospital (London, UK) and Chelsea and Westminster NHS Foundation Trust (London, UK) between January 2011 and October 2012. Adult subjects of 60 years of age or greater on cART were eligible to participate. Inclusion criteria included no previous exposure to any HIV-integrase inhibitors, no evidence of HIV drug-resistance mutations and plasma HIV RNA < 50 copies/ml (Quantiplex assay^TM; Bayer, Emeryville, CA, USA). Subjects with active opportunistic infection or concomitant medications, which could interact with raltegravir were excluded.

On day 1, subjects switched cART to tenofovir/emtricitabine 245/200 mg once daily and raltegravir 400 mg twice daily for the duration of the study period. Intensive pharmacokinetic (PK) sampling was performed at steady state on day 28. Blood samples for raltegravir quantification were collected under fasting conditions 10 minutes before dosing (pre-dose) and every hour for 12 hours and then 24 hours post-dosing. Raltegravir concentrations were determined by HPLC-MS/MS as previously described.^13,14 The lower limit of quantification for raltegravir was 4.7 ng/ml. PK parameters were obtained and expressed as geometric mean, 95% confidence interval, using WinNonlin (version 6.1; Pharsight Corporation, Mount View, CA, USA). Inter-patient variability in parameters were expressed as coefficients of variation (standard deviation/mean) × 100. CD4⁺ lymphocyte count, plasma HIV RNA and haematology and chemistry panels were undertaken throughout the study period.

To investigate the impact of age on raltegravir exposure, two younger historical HIV-infected control groups on twice daily raltegravir (400 mg) co-administered with darunavir/ritonavir (DRV/r) 800/100 mg once daily were included in population PK modelling: control group 1 (n = 14, age: 25–55 years, administered drug in fasted state¹⁵) and control group 2 (n = 24, age: 26–51 years, administered drug with a low fat meal¹⁴). Bioanalysis of raltegravir concentrations were performed at the same laboratory for all studies, therefore mitigating cross-study comparisons. Nonlinear mixed effects modelling were applied to the data from the three studies using the SAEM algorithm of Monolix (version 4.2.0; Lixoft, Paris, France), allowing independent assessment of age (stratified < 60 years versus ≥ 60 years), food (fed versus fasted), and DRV/r use (present or absent). The influence of these covariates on raltegravir relative bioavailability (F) and apparent oral clearance (CL/F) was investigated and a change in objective function value (OFV) of at least 3.84 was required in order to accept the model with an additional parameter (P = 0.05, Chi-squared distribution, 1° of freedom).

Neurocognitive testing was performed at baseline and week 24 in the over 60 population using a computerized cognitive test battery (CogState; CogState Ltd, Melbourne, Australia), which has been previously validated in HIV-infected subjects.¹⁶ Mean scores from the eight tasks (NPZ-8) assessed were used to derive a global composite measure of neurocognitive function. Two tailed paired t-tests were used to evaluate intra-individual changes in global cognitive composite z-scores between baseline and 24 weeks.

Results

Nineteen subjects (18 male, 16 of white ethnicity) in the over 60 population completed all study procedures. Fifteen subjects (78%) were receiving at least one concomitant medication, median (range) 4 (0–5), none of which would be expected to interact with raltegravir exposure. Raltegravir was safe and well tolerated in all subjects. No laboratory safety concerns were observed and plasma HIV RNA remained < 50 copies/ml throughout the study in all subjects.

PK parameters for raltegravir (individual parameters for the over 60 group and population parameters for all subjects) are summarized in Table 1. No statistically significant difference in raltegravir CL/F was observed between the study group and the younger controls following univariate analysis (ΔOFV: (1.39; P = 0.12). Use of DRV/r did not influence raltegravir CL/F (ΔOFV: (0.71; P>0.29), whereas in the presence of food, the relative bioavailability of raltegravir (F) was reduced by approximately 53% from 1 to 0.468 (ΔOFV: (7.14; P = 0.003, Table 1).

Table 1. Subject characteristics and pharmacokinetic parameter results

Parameter	Cases	Control group 1	Control group 2	Estimate	RSE (%)
Cohort characteristics
Number	19	23	14
Age, mean (SD), years	66 (3.4)	37 (7)	40 (11)
Viral load < 50 copies/ml (%)	100	100	100
Baseline CD4 T-cell count, mean (SD), cells/μl	618 (278)	466 (252)	560 (228)
Raltegravir PK parameters [geometric mean (95% CI), CV]
AUC0–12 (ng h/ml)	6240 (3735–104 06), 75
Cmax (ng/ml)	1732 (969–3095), 81
Tmax (hour)	2.8 (2.1–3.9), 53
Ctrough (ng/ml)	73.7 (45.9–118), 101
Population modeling PK parameters
CL/F (l/h)				80.6	18
Vc/F (l)				313	41
Q/F (l/h)				14.3	42
Vp/F (l)				858	40
ka (h^− 1)				0.280	31
F				1 (fixed)	F
Covariates
ThetaFOOD F				− 0.758	34
FFOOD				0.468	26

Note: AUC_0–12 or AUC_0–24: area under the curve over the 12-hour dosing interval; C_max: maximum plasma concentration; T_max: time to maximum concentration; C_trough: trough plasma concentration; CL/F: apparent oral clearance; Vc/F: apparent volume of distribution of the central compartment; Q/F: inter-compartmental clearance; Vp/F: apparent volume of distribution of the peripheral compartment; k_a: absorption rate constant; F: relative bioavailability of raltegravir; theta_FOOD: factor associated with change in relative bioavailability of raltegravir in the presence of low fat meal; F_FOOD: relative bioavailability of raltegravir in the presence of low fat meal; RSE: relative standard error, RSE = (SE_ESTIMATE/ESTIMATE) × 100.

After 24 weeks of study therapy a decline in cognitive function was observed [change in (SD) global score of (0.91 (1.3), P = 0.018]. No associations between clinical factors including type of prior cART or type of concomitant medication and change in global cognitive scores at week 24 were observed (all P values >0.1).

Discussion

This is the first study to describe in detail the PK profile of raltegravir in older PLWH. No significant changes in raltegravir exposure associated with age were observed. These results are consistent with a previous composite analysis of phase I and II clinical studies in patients aged between 19 and 71 years where no clinically meaningful effect of age on raltegravir PK parameters were observed.¹⁷ On assessing the effects of food intake on raltegravir PK, intake of a low fat meal significantly reduced raltegravir relative bioavailability by approximately 50%. This is consistent with a previous report suggesting that raltegravir total plasma exposure and C_max are reduced by 46 and 52%, respectively following a low fat meal.¹⁸ In contrast to previous reports, use of DRV/r did not significantly influence raltegravir concentrations.¹⁴

The pharmacokinetics of raltegravir is known to be variable and characterized by large inter- and intra-patient variability.¹⁹ Similar to our study, in previously published results evaluating the pharmacokinetics of raltegravir, gender, age, Body Mass Index, and food intake did not have a clinically meaningful effect of raltegravir pharmacokinetics.^17,19,20

The observed decline in cognitive function after 24 weeks of commencing the study regimen is an unexpected finding. Before the baseline cognitive assessment, subjects underwent a practice cognitive assessment during study screening procedures. This test was included in the study protocol in order to reduce the impact of improvements in cognitive testing parameters over time with practice.^21,22 However, for the follow-up cognitive assessment (24 weeks after baseline), no further practice assessment was administered. We believe that the decline in cognitive function noted in our study is due to the lack of a practice test being administered before week 24 rather than being associated with any other specific study interventions, such as a modification of antiretroviral therapy to a raltegravir-containing regimen. This study has several limitations. First, the study has relative small sample size and second, a control group on raltegravir co-administered with tenofovir/emtricitabine was not available for analysis as we tried to minimize the effect of cross-study comparisons by selecting only control groups that had pharmacokinetic analyses performed at the same laboratory.

In summary, raltegravir was safe and well tolerated in older HIV-infected individuals. We observed no significant changes in raltegravir PK parameters associated with age.

Disclaimer Statements

Contributors AW and MB conceived the study. JHV, TB, BMP and AJ were involved in data gathering. JHV, LD, LE and DB were involved with data analysis and interpretation. JHV drafted the manuscript. All authors were involved in the critical revision of the manuscript and approved it.

Conflicts of interest

AW and MB have received honoraria or research grants, or been a consultant or investigator, in clinical trials sponsored by Abbott, Boehringer Ingelheim, Bristol-Myers Squibb, Gilead Sciences, GlaxoSmithKline, Janssen Cilag, Roche, Pfizer, and ViiV Healthcare. JHV is supported by the Wellcome Trust Translational Medicine and Therapeutics Fellowship. He has received honoraria by Merck, Janssen Cilag, and sponsorship to attend scientific conferences from Janssen Cilag, Merck, Gilead Sciences, ViiV Healthcare, and AbbVie. LD is supported by PreDiCT-TB.

All remaining authors declare no competing interest.

Funding

This study was funded by an investigator initiated grant from Merck UK.

Ethics approval

Ethical approval was given by the Imperial College London NHS health care trust Ethics Committee and London REC office EudraCT number 2010-022907-23.

Acknowledgements

The authors would like to thank all the study participants, the nursing and pharmacy staff, and the Clinical Trials Centre, Imperial College London (St Mary's Campus), UK, and St Stephen's Centre at Chelsea and Westminster Hospital.