Impact of protease inhibitors on intracellular concentration of tenofovir-diphosphate among HIV-1 infected patients

Abstract

Intracellular nucleoside reverse transcriptase inhibitor (NRTI) concentrations are associated with plasma HIV-1 response. Coadministration of protease inhibitors with NRTIs can affect intra-cellular concentrations due to protease inhibitor inhibition of efflux transporters. Tenofovir-diphosphate (TFV-DP) concentrations within peripheral blood mononuclear cells were compared among individuals receiving either atazanavir or darunavir-based regimens. There was a trend towards higher TFV-DP concentrations among women and among participants receiving atazanavir. TFV-DP intracellular concentrations were positively associated with undetectable plasma HIV-1 RNA.

Combination antiretroviral therapy (cART) results in virologic suppression and robust immune reconstitution, leading to prolonged life expectancy in HIV-1 infection. Despite cART, eradication of HIV-1 has not been achievable, as the virus remains detectable in suspected reservoirs, including peripheral blood mononuclear cells (PBMCs), even with plasma virologic suppression [1].

Nucleoside reverse transcriptase inhibitors (NRTIs), the cART ‘backbone’, are prodrugs requiring intracellular phosphorylation to produce active metabolites. Clinical studies have shown significant associations between intracellular NRTI concentrations and virologic response [2–5]. Intracellular NRTI concentrations may be modulated by drug–drug interactions mediated by membrane transporter inhibition [6,7]. For example, PBMCs express efflux transporters, including p-glyco-protein (p-gp) and multidrug resistance associated proteins (MRPs), which can be inhibited by protease inhibitors. In particular, higher intracellular concentrations of tenofovir-diphosphate (TFV-DP), the active metabolite of tenofovir disoproxil fumarate (TDF), were achieved when coadministered with lopinavir/ritonavir (LPV/RTV) compared with a nonprotease inhibitor regimen [7].

Atazanavir (ATV) and darunavir (DRV), two protease inhibitors recommended as first line for ART-naive patients, are both p-glycoprotein (p-gp) inhibitors [8–10], although ATV is a more potent p-gp inhibitor than DRV [11]. To determine whether intracellular concentrations of TFV-DP differ in patients receiving ATV vs. DRV-based regimens, we compared TFV-DP concentrations in PBMCs of participants receiving RTV-boosted ATV vs. RTV-boosted DRV and assessed relationships with plasma HIV-1 RNA.

This was a substudy of a cross-sectional investigation of HIV-1 infected patients at the Ponce de Leon Center in Atlanta, Georgia, receiving regimens of daily TDF/emtricitabine (300 mg/200 mg) as well as RTV (100 mg)-boosted and either once-daily ATV (300 mg) or DRV (800 mg). Eligibility criteria included documented adherence to cART and undetectable plasma HIV RNA for at least 6 months before study entry. PBMCs were isolated from blood collected at trough times for 30 participants from May to December 2012 [12]. The Emory University Institutional Review Board and Grady Research Oversight Committee approved this study. All study participants gave written informed consent.

PBMCs were collected and prepared using a well documented method [13]. Two million PBMCs were suspended in 70% methanol; supernatants were dried and stored at −20°C until analysed by liquid chromatography-tandem mass spectrometry (LC-MS/MS) as previously described [14], with minor modifications. Intracellular TFV-DP was separated using a Kinetex XB-C18 column (100 × 2.1 mm) with 2.6 μm particle size (Phenomenex, Torrance, California, USA) at a flow rate of 200 μl/min. About 2 mmol/l NH₃H₂PO₄ with 3 mmol/l hexylamine as solvent A and acetonitrile as solvent B was used in a gradient elution programme as follows: 3–25% B from 0 to 18 min, 25–80% B from 18 to 22 min, 80% B from 22 to 25 min. Equilibration time between two injections was 10 min. An API5000 triple-quadrupole mass spectrometer in positive mode was used for detection, by multiple reaction monitoring (MRM): TFV-DP (448 → 176). Plasma HIV-1 RNA was measured with COBAS! Ampliprep/COBAS Taqman version 2.0 HIV-1 assay (Roche Molecular Systems, Inc) [15].

Intracellular drug concentrations were log transformed; geometric means and 95% confidence intervals (CIs) were compared for each arm using a two-sided, two-sample t-test. Relationships between intracellular TFV-DP and detection of plasma HIV-1 RNA were assessed using point biserial correlation tests for each arm.

Study population demographic and clinical characteristics were previously reported [12]. Thirty patients were enrolled (n = 15 on ATV and n = 15 on DRV): 23/30 (76.7%) male, 26/30 (86.7%) black and median age 46.9 years (interquartile range, IQR, 37.9–51.9). Five women were in the ATV group and two in the DRV group (P = 0.4).

Higher geometric mean intracellular concentrations of TFV-DP were seen in the ATV arm than in the DRV arm (0.60 pmol/10⁶ cells, 95% CI 0.41–0.89 vs. 0.39 pmol/10⁶ cells, 95% CI 0.21–0.71), and in women than in men [0.65 pmol/10⁶ cells (95% CI 0.33–1.29) vs. 0.44 pmol/10⁶ cells (95% CI 0.29–0.68)], although neither reached statistical significance (Fig. 1a).

Fig. 1. Scatterplots of intracellular tenofovir-diphosphate concentrations in pmol/10⁶ cells are shown with brackets indicating geometric mean and 95% confidence intervals.

(a) Intracellular concentrations are compared between atazanavir (ATV, circles) vs. darunavir (DRV, squares) arms, and between women (up-pointing triangles) vs. men (down-pointing triangles). (b) Intracellular TFV-DP concentrations are compared between participants in each treatment arm stratified by HIV viral load. ATV arm with plasma HIV viral load ≥40 copies/ml (plasma HIV RNA+, circles) vs. plasma HIV viral load <40 copies/ml (plasma HIV RNA-, squares); DRV arm with plasma HIV RNA+ (up-pointing triangle) vs. plasma HIV RNA- (down-pointing triangle). *^P < 0.05.

Ten of 15 patients in the ATVarm had a detectable plasma viral load [median 225 copies/ml (IQR 140–310)] compared with four of 15 in the DRV arm [median 200 copies/ml (IQR 95–3520)], P = 0.03. A significant positive correlation was seen between intracellular TFV-DP concentration and undetectable plasma HIV RNA (<40 copies/ml) in the ATVarm (r = −0.56, P=0.03) but not in the DRV arm (^r=−0.02, P = 0.94), Fig. 1b.

TFV-DP intracellular concentrations were slightly higher than prior published reports: Pruvost et al. [7] found TFV-DP C_trough median of 0.222 pmol/10⁶ cells (IQR 0.160–0.376 pmol/10⁶ cells), while Dumond et al. [16] found TFV-DP mean concentration to be 0.112 pmol/ 10⁶ cells (SD 0.077 pmol/10⁶), although significant interindividual variability has been documented with intracellular TFV concentrations. Consistent with Pruvost et al. [7], women trended towards higher TFV-DP concentrations than men. This phenomenon has been described with other phosphorylated NRTIs, including zidovudine-triphosphate (TP) and lamivudine-TP [5]. Although biologic mechanisms for these sex differences remain unclear, women may phosphorylate NRTIs differently than men and hormonal differences may play a role. A significant positive correlation between intracellular TFV-DP and undetectable plasma HIV-1 RNA was seen with ATV but not DRV. One mechanistic explanation for this is that DRV has greater antiviral activity than ATV [17], making plasma HIV-1 RNA suppression less dependent upon intracellular TFV-DP concentrations for patients on DRV.

In conclusion, TFV-DP concentrations in PBMCs were similar in both treatment arms. The trend towards higher intracellular concentrations in the ATV arm could be explained by the fact that five of seven women were in that group: further studies should explore this sex effect. In addition, TFV-DP concentrations predicted detectable plasma HIV-1 RNA in the ATV arm, emphasizing the importance of adequate antiretroviral penetration into PBMCs for achieving virologic control in patients receiving this protease inhibitor.

Future prospective studies should evaluate intracellular ART concentrations throughout the dosing interval coupled with longitudinal HIV-1 RNA measurements. These data would elucidate the virologic impact of PBMC drug concentrations over time. Finally, studies must be adequately powered to investigate sex differences in antiretroviral pharmacology with a goal of informing clinical practice.