Extended cell and plasma drug levels after one dose of a 3-in-1 nanosuspension containing lopinavir, efavirenz, and tenofovir in non-human primates

Abstract

Objective:

To characterize a drug-combination nanoparticle (DcNP) containing water-insoluble lopinavir (LPV) and efavirenz (EFV), and water-soluble tenofovir (TFV), for its potential as a long-acting combination HIV treatment.

Design:

Three HIV drugs (LPV, EFV, TFV) with well-established efficacy and safety were co-formulated into a single DcNP suspension. Two macaques were administered one subcutaneous injection and drug concentrations in plasma and mononuclear cells (in peripheral blood and lymph nodes) were analyzed over two weeks. Pharmacokinetic parameters and cell-to-plasma relationships of LPV, EFV, and TFV were determined.

Results:

This 3-in-1 nanoformulation provided extended concentrations of all drugs in lymph node cells that were 57–228-fold higher than those in plasma. Levels of all 3 drugs in peripheral blood mononuclear cells persisted for two weeks at levels equal to or higher than those in plasma.

Conclusion:

With long-acting characteristics and higher drug penetration/persistence in cells, this 3-in-1 DcNP may enhance therapeutic efficacy of these well-studied HIV drugs due to co-localization and targeting of this 3-drug combination to HIV host cells.

Introduction

About 2.1 of 37 million HIV-infected persons are children (<15 years) [1]. Fewer antiretroviral regimens are available for children than adults. Due to physio-pharmacological differences, pediatric treatment requires age-appropriate dosing. Current WHO guidelines for treatment-naïve children (3–10 years old) recommend the non-nucleoside reverse transcriptase inhibitor (NNRTI), efavirenz (EFV), and combinations of nucleoside reverse transcriptase inhibitors (NRTIs), including abacavir, zidovudine or tenofovir disoproxil fumarate (TDF) plus lamivudine (3TC) or emtricitabine [2]. For children <3 years, WHO guidelines suggest using a protease inhibitor (PI) [lopinavir (LPV)] plus two NRTIs [2]. To maximize the potential for HIV suppression, a drug combination inhibiting three HIV replication targets—protease and the nucleoside and non-nucleoside binding pockets of reverse transcriptase—is of interest. We studied three antiretroviral drugs not under patent protection—LPV, EFV, and tenofovir (TFV)—that have well-characterized efficacy and safety profiles and are still widely used in resource-limited settings.

In patients with durable viral suppression, residual HIV can be isolated from tissues, particularly lymph nodes [3]. With a report of lymph node drug insufficiency [4], and confirmation of its link to residual virus [3, 5, 6], efforts are underway to enhance drug exposure in these tissues and cells [7]. Lymph node drug insufficiency with oral treatment may be related to (1) variability in pharmacokinetics of the drugs commonly used (e.g., T_max for EFV, TFV, and LPV in plasma are 3–5, 1, and 6 hr, respectively) [8–10], (2) differing drug clearance and thus variable intracellular drug kinetics, (3) limited drug penetration from plasma into node tissue and cells, and (4) challenges in patient adherence to lifelong daily dosing [11].

In searching for a drug combination that could address lymph node drug insufficiency and provide consistent drug levels in cells within the blood and tissues that harbor HIV, we focused on developing a stable, scalable, and long-acting HIV drug-combination nanoparticle (DcNPs) capable of delivering a water-soluble and water-insoluble drugs in one injection. Based on a proven DcNP platform with an HIV drug combination intended for adults [12, 13], with consideration of the WHO recommended regimens for children [2], we investigated the effects of DcNPs on the pharmacokinetics of LPV, EFV, and TFV—three well-characterized HIV drugs targeted to three distinct antiviral regions: HIV protease (LPV) and non-nucleoside (EFV) and nucleoside (TFV) domains of HIV reverse transcriptase. With a single dose non-human primate study of DcNPs containing LPV, EFV, and TFV, we found intracellular LPV and EFV levels persisted at higher levels than those in plasma over 2-weeks, and intracellular TFV levels were approximately equivalent to plasma levels.

Materials and Methods

Chemicals

Lopinavir, EFV, and TFV were obtained from Waterstone (Carmel, IN). Excipient phospholipids DSPC and DSPE-methoxy polyethylene glycol (mPEG₂₀₀₀) were from Corden Pharma (Liestal, Switzerland). Acetonitrile, methanol, methylene chloride, and triflouroacetic acid were from Fisher (Waltham, MA).

Methods

Lopinavir, EFV, and TFV were stabilized in nanoparticles with lipid excipients DSPC and DSPE-mPEG₂₀₀₀ as reported [12]. Final LPV:EFV:TFV molar ratios were 0.8:1.0:1.5; DcNP-associated fractions were 76.6, 78.7, and 7.9%.

Pharmacokinetic study

Two male macaques (M. nemestrina) were given one subcutaneous injection of DcNPs (25.0, 16.7, 22.6 mg/kg LPV, EFV, TFV) with IACUC approval. Blood was collected at specified time points over 336 hours (14 days) [11]. At 24 and 192 hours, an inguinal lymph node was excised. Drug in plasma, peripheral blood mononuclear cells (PBMCs), and lymph node mononuclear cells (LNMCs) were analyzed using a validated mass-spectrometry assay similar to published methods [14, 15].

Results

The three drugs with different hydrophobicities (LPV, EFV, TFV: LogP 4.7, 4.46, −3.6) were stabilized in a DcNP nanosuspension. No safety issues, including injection site reactions, were observed. Drug concentration time-courses in plasma and PBMCs were determined; LPV, EFV, and TFV were detectable in plasma throughout the 2-week study (Figure 1).

Figure 1. Time-course of plasma and PBMC intracellular concentrations of LPV, EFV, and TFV after a single SC dose of the drug combination nanoparticle (DcNP) suspension.

The squares (■) represent PBMC intracellular drug concentrations, and circles (●) represent plasma drug concentrations. Each time point represents arithmetic mean ± SD of plasma and PBMC intracellular drug concentrations from N=2 male pigtail macaques following a 25.0 mg/kg LPV, 16.7 mg/kg EFV, and 22.6 mg/kg TFV SC dose. At some timepoints, the SD is too small to be visible. PBMC cell volume was assumed to be 0.2829 pL/cell [32].

Intracellular drug levels for all drugs were equal to or higher than those in plasma; this is unexpected as oral and injectable dosage forms typically produce lower PBMC versus plasma drug concentrations [16–18]. Significantly higher intracellular concentrations versus plasma were more prominent for LPV and EFV (Figure 1A, 1B). By day 14, intracellular EFV levels in PBMCs were > 100-fold higher than in plasma. Based on the plasma time-course, the long-acting behavior of DcNPs with LPV, EFV, and TFV may extend beyond 2 weeks, particularly intracellularly (Figure 1A-1C).

Pharmacokinetics of intracellular and plasma concentration-time profiles of LPV, EFV, and TFV were analyzed. Compared to plasma drug exposure (expressed as area-under-the-curve, AUC_0–336h), the intracellular AUC was equivalent for TFV (3,146 / 3,004) and were 2.8-fold (59 / 21) and 7.1-fold (211 / 28) higher for LPV and EFV, respectively (Table 1). Enhanced intracellular exposures for LPV and EFV are reflected in parallel increases in C_max, with slight delays to peak plasma versus cell concentrations as reflected in t_max (8.0 v 6.5 h for LPV; 8.0 v 4.0 h for EFV). TFV exhibited two C_max peaks—an early smaller peak and a later higher peak for both plasma and PBMCs (Figure 1C). The TFV plasma peaks occurred at 1 and 48 hours, and in PBMCs at 1 and 12.3 hours, suggesting possible earlier cellular TFV availability.

Table 1.

Plasma versus PBMC intracellular PK parameters for LPV, EFV, and TFV after a single SC dose of the drug combination nanoparticle (DcNP) suspension as well as cell-to-plasma ratios at 24 and 192 hours.

SC Dose (mg/kg)	Lopinavir (LPV)		Efavirenz (EFV)		Tenofovir (TFV)
	25.0		16.7		22.6
Pharmacokinetic parameters
	Plasma	PBMCs	Plasma	PBMCs	Plasma	PBMCs
AUC0–336h (hr•μg/mL)	20.8 (19.9)	59.0 (90.0)	27.7 (11.0)	211.2 (35.2)	3,004.1 (3.3)	3,146.3 (6.1)
Cmax (μg/mL)	0.8 (22.5)	2.2 (62.4)	1.0 (0.7)	7.1 (67.1)	21.7 (1.7)	25.4 (17.8)
Tmax (hr)	8.0 (0.0)	6.5 (32.6)	8.0 (0.0)	4.0 (35.4)	48.0 (0.0)	12.3 (135.6)
t1/2 (hr)	91.9 (7.3)	308.9 (121.7)	61.6 (4.1)	418.0 (114.8)	75.4 (9.1)	38.8 (11.4)
Cell-to-plasma drug concentration ratio*
Time point	LNMC/Plasma	PBMC/Plasma	LNMC/Plasma	PBMC/Plasma	LNMC/Plasma	PBMC/Plasma
24h	3.5 (61.8)	1.6 (44.2)	29.6 (54.8)	4.7 (36.1)	1.2 (20.7)	1.2 (47.7)
192h	56.6 (130.6)	4.3 (101.7)	227.8 (31.6)	341.1 (38.1)	0.9 (33.9)	0.6 (1.8)

Arithmetic mean (% coefficient of variation).

LNMCs, lymph node mononuclear cells; PBMCs, peripheral blood mononuclear cells; SC, subcutaneous.

LNMC and PBMC cell volume assumed to be 0.2829 pL/cell [32].

^*Cell drug concentration of LNMC and PBMC divided by plasma drug concentration at the same time point.

As LPV, EFV, and TFV concentrations persisted even at day 14, the terminal half-lives (t_1/2) are estimated values. In this context, the t_1/2 in PBMCs for LPV and EFV were much higher than those in plasma. For TFV, the PBMC t_1/2 was about half that of plasma (Table 1).

Data in Table 1 indicate LPV, EFV, and TFV levels in LNMCs persisted at day 8 (192 hours), and were generally higher than plasma. By day 8, LPV and EFV LNMC levels were 57- to 228-fold higher than plasma. For TFV, LNMC levels were about equivalent to those in plasma. Collectively, these data indicate that water-insoluble and water-soluble HIV drugs (LPV, EFV, TFV) could be combined in a single DcNP and provide persistent plasma and intracellular drug levels in LNMCs and PBMCs for 2 weeks.

Discussion

Most oral fixed-dose combination regimens target two HIV domains. Currently, oral HIV treatment formulations for weekly or monthly administration are unavailable. The ability to stabilize three HIV drugs (LPV, EFV, TFV) with drastically different hydrophobicities in one DcNP injectable formulation enabled persistent levels of all drugs in plasma and cells (PBMCs) for two weeks. By comparison, after administration of these drugs orally [19–21] or injection of TFV [22], these drugs clear from the body within 24–48h. These drugs in our DcNPs, targeted to three HIV sites, have well-established clinical safety and antiviral profiles and are among currently recommended first line agents for children by WHO. In two primates, all drugs in this 3-in-1 DcNP suspension persisted in LNMCs at levels higher than plasma (Table 1). Although more animals could improve confidence, we observed limited variation in the drug levels in the two monkeys and the data clearly indicate long-acting behaviors of the drugs given in this DcNP formulation. Compared to previous DcNP study that included ritonavir as a booster for LPV [13], we observed a shorter LPV plasma apparent half-life (92 vs 477 hours) but longer PBMC apparent half-life (309 vs 151 hours).

Development of HIV-drug resistance is a potential consequence of inadequate viral suppression. Current standard of care regimens use a combination of three antiretrovirals directed to two targets. Previous studies of initial ART using plasma viral load as a primary endpoint did not show benefit from four- or five-drug therapies targeting more than two viral sites [23, 24]. However, tissue penetration was not addressed in these studies. With our three-drug injectable combination targeted to three separate sites and with persistent drug levels in cells and plasma, this DcNP formulation may enhance viral suppression in tissues, promote treatment adherence, and reduce potential for drug resistance. Persons with HIV unable or unwilling to swallow oral tablets or capsules—including persons hospitalized with critical illness, after surgery, with “pill fatigue;” or children—may benefit from this strategy. A 3-in-1 injectable formulation also negates the need to inject individual drugs separately to create a multi-drug regimen.

Injectable long-acting (LA) cabotegravir LA and rilpivirine LA have been shown to provide up to 2 years of viral suppression with every 4–8 week intramuscular injections [25]. Providing sustained intracellular and tissue levels of these two hydrophobic drugs depends on drug distribution and pharmacokinetics of each. These particulate hydrophobic drugs deposit and are retained in muscle, and slowly release into nearby blood capillaries before redistributing intracellularly [26, 27]. This results in lower cellular than plasma levels. In contrast, we found intracellular drug levels of lipophilic EFV and LPV and hydrophilic TFV in PBMCs with our formulation were higher than or equal to plasma (Figure 1, Table 1). Preliminary data from other DcNP formulations suggested that the nanosuspension is first absorbed from the subcutaneous space into lymph vessels, which are more permeable than blood vessels [28]. The relatively stable DcNP-bound drugs subsequently distributed to and were retained in nodes throughout the lymphatic system. These data are consistent with the higher observed PBMC than plasma drug concentrations. A radiolabeled lymphocyte study in SHIV-infected primates indicated ~75% of lymphocytes are in lymph nodes, which can traffic to-and-from the blood [29]. These and other potential mechanisms leading to high and early peaks of cell and plasma drug concentrations after subcutaneous administration of DcNPs are under investigation and beyond the scope of this report.

Though TFV is not indicated for younger children (<3) due to renal and bone toxicity [30], the enhanced intracellular levels and lower plasma TFV level achieved with our DcNPs could potentially provide equivalent cell-dependent therapeutic effects with lower dose. Lower plasma TFV exposure with equivalent anti-viral potency has been documented in adults using TFV alafenamide (TAF) [31], a prodrug of TFV that enhances intracellular lymphocyte TFV. While it is of interest, an efficacy study in a pediatric primate model with this DcNP formulation is beyond the scope of this report.

In summary, one dose of a three-drug combination provided persistent LPV, EFV, and TFV levels in LNMCs, PBMCs, and plasma for 2 weeks; in contrast, conventional oral dosage forms of these drugs would require 14 consecutive doses over 2 weeks to achieve similar plasma levels, and intracellular drug levels would be less than those in plasma. In addition to the hydrophobic drugs LPV and EFV, this DcNP platform enabled incorporation of water-soluble TFV with a prolonged plasma exposure and enhanced intracellular drug levels. This higher intracellular drug exposure coupled with multi-week persistence may help improve adherence and overcome limited lymphatic drug exposure. This DcNP platform may be considered for developing lymphocyte-targeted, long-acting antiretroviral combinations with drugs that have diverse hydrophobic and hydrophilic characteristics.