Pharmacokinetics of the Antimalarial Drug,AQ-13,in Rats and Cynomolgus Macaques

Abstract

The purpose of this study was to evaluate the bioavailability and pharmacokinetics of a new antimalarial drug, AQ-13, a structural analog of chloroquine (CQ) that is active against CQ-resistant Plasmodium species, in rats and cynomolgus macaques. Sprague-Dawley rats (n = 4/sex) were administered a single dose of AQ-13 intravenously (i.v.) (10 mg/kg) or orally (20 or 102 mg/kg). Blood and plasma samples were collected at several timepoints. AQ-13 achieved C _max after oral administration at approximately 3 to 4 h and could be detected in blood for 2 to 5 days after oral administration. The ratio of area under the curve (AUC) values at the high and low dose for AQ-13 deviated from an expected ratio of 5.0, indicating nonlinear kinetics. A metabolite peak was noted in the chromatograms that was identified as monodesethyl AQ-13. Oral bioavailability of AQ-13 was good, approximately 70%. The pharmacokinetics of AQ-13 was also determined in cynomolgus macaques after single (i.v., 10 mg/kg; oral, 20 or 100 mg/kg) and multiple doses (oral loading dose of 50, 100, or 200 mg/kg on first day followed by oral maintenance dose of 25, 50, or 100 mg/kg, respectively, for 6 days). The AUC and C _max values following single oral dose administration were not dose proportional; the C _max value for AQ-13 was 15-fold higher following an oral dose of 100 mg/kg compared to 20 mg/kg. MonodesethylAQ-13 was a significant metabolite formed by cynomolgus macaques and the corresponding C _max values for this metabolite increased only 3.8-fold over the dose range, suggesting that the formation of monodesethyl AQ-13 is saturable in this species. The bioavailability of AQ-13 in cynomolgus macaques following oral administration was 23.8% for the 20-mg/kg group and 47.6% for the 100-mg/kg group. Following repeat dose administration, high concentrations of monodesethyl AQ-13 were observed in the blood by day 4, exceeding the AQ-13 blood concentrations through day 22. Saturation of metabolic pathways and reduced metabolite elimination after higher doses are suggested to play a key role in AQ-13 pharmacokinetics in macaques. In summary, the pharmacokinetic profile and metabolism ofAQ-13 are very similar to that reported in the literature for chloroquine, suggesting that this new agent is a promising candidate for further development for the treatment of chloroquine-resistant malaria.

Malaria, a public health threat in more than 90 countries of the world, has an incidence of 300 to 500 million cases, including an estimated one million deaths per year, many of which are young children (Myrvang and Godal 2000). Chloroquine (CQ), a 4-aminoquinoline, is a widely used antimalarial drug that was synthesized over 60 years ago, but its usefulness is becoming increasingly limited because of the emergence of CQ-resistant Plasmodium strains. Cases of resistance of Plasmodium falciparum to treatment with antimalarials such as CQ, mefloquine, and pyrimethamine-sulfadoxine have been reported (Asindi et al. 1993). It is therefore important to identify analogs that are effective against these resistant strains and change the existing treatment regimens. A series of aminoquinolines that is structurally similar to CQ but has modifications in the aminoquinoline side chain has been synthesized and tested for antimalarial activity in vitro (De et al. 1996). The three-carbon side chain analog, AQ-13, which exhibits activity against parasites resistant to CQ, mefloquine, and other antimalarial agents, was selected for preclinical studies.

In humans, CQ binds extensively to a variety of tissues and has a large apparent volume of distribution, approximately 200 to 800 L/kg when calculated from plasma concentrations and 200 L/kg when estimated from whole blood data (Titus 1989; Krishna and White 1996; Ducharme and Farinotti 1996). CQ is 50% to 60% bound to plasma proteins (Ofori-Adjei et al. 1986) and has a very long terminal elimination half-life (t _½) of 1 to 2 months, with a relatively high total body clearance (CL) of about 1 L/h/kg as calculated from the plasma data (Frisk-Holmberg et al. 1984). Redistribution processes from the tissues to intravascular compartments are believed to be the factors primarily responsible for maintaining CQ blood concentrations for such long periods of time (Frisk-Holmberg, Bergqvist, and Termond 1985).

In healthy human volunteers, approximately 77% of administered CQ is excreted in the urine as the parent compound and as the metabolite, desethylchloroquine (Walker et al. 1987). The major metabolites of CQ formed by humans after single and multiple doses are monodesethyl CQ and didesethyl CQ (McChesney, Fasco, and Banks, 1962; Price-Evans, Fletcher, and Baty, 1979; Essein and Afamefuna 1982). Monodesethyl CQ has antimalarial activity and represents about 20% to 35% of unchanged CQ in human plasma (Tracy and Webster 1996). This metabolite is formed rapidly and has been detected in blood and plasma samples during the absorption phase for CQ. The didesethyl CQ is a relatively minor metabolite and constitutes only 1% of the total amount excreted in urine and about 10% of CQ plasma concentrations (Frisk-Holmberg et al. 1984). CQ is metabolized to these N-dealkylated metabolites, possibly by the cytochrome P450 isoforms, CYP3A4 and CYP2D6, although the exact enzymes involved have not been definitively proven (Ducharme and Farinotti 1996).

CQ is well absorbed when administered orally and bioavailability in humans is approximately 80% (Gustafsson et al. 1983). It is equally well absorbed when administered by other extravascular routes, for example, subcutaneous (s.c.), intramuscular (i.m.), and rectal routes (Minker and Ivan 1991; Tjoeng et al. 1991).

Variability in CQ plasma pharmacokinetics has been related to difficulty in separating plasma from various blood cell components, particularly platelets, with which the drug interacts significantly. CQ concentrations in platelets are approximately 400-fold higher than in erythrocytes, and the exact procedure used to prepare plasma from whole blood greatly influences the concentration of CQ found in the resulting sample; thus whole blood, or platelet-free plasma, is recommended for pharmacokinetics studies of CQ (Rombo et al. 1985). In a pharmacokinetic study performed in healthy human volunteers, whole blood concentrations of CQ were about 8- to 10-fold higher than the plasma concentrations, and volume of distribution and Cl were approximately 10-fold lower when calculated from whole blood concentrations than when calculated from plasma concentrations (Frisk-Holmberg et al. 1984).

AlthoughAQ-13 is structurally very similar to CQ(Figure 1), prior to the current study the pharmacokinetics profile of AQ-13 was not known. Therefore, this study was designed to assess the oral bioavailability and pharmacokinetics of AQ-13 in male and female Sprague-Dawley rats and in cynomolgus monkeys following intravenous (i.v.) and oral administration. Both single-and multiple-dose regimens were examined and the major metabolite found in the blood was characterized.

MATERIALS AND METHODS

Materials

AQ-13, [N ¹-(7-chloro-quinolin-4-yl)-3-(N ³,N ³-diethylamino) propylamine] dihydrochloride trihydrate, was supplied by the Division of AIDS, National Institute of Allergy and Infectious Diseases (NIAID), via McKesson BioServices, Rockville, MD.CQ was purchased from Fluka Chemical Co., Ronkonkoma, NY. Dulbecco’s phosphate-buffered saline (PBS) was purchased from Gibco BRL (Grand Island, New York). Male and female Sprague-Dawley rats were purchased from Charles River Laboratories, Raleigh, NC. Rat plasma and blood was purchased from Pel-Freez (Rogers, AK), cynomolgus macaque plasma and blood from Sierra Biomedical, Inc. (Sparks, NV), and human plasma from Peninsula Blood Bank (Burlingame, CA). Centrifree ultrafiltration devices were from Millipore Corporation (Bedford, MA). All other reagents and solvents used were either analytical or high-performance liquid chromatograph (HPLC) grade.

Experimental Design

Rats

Male and female rats (n = 4/sex) were administered a single dose of AQ-13 formulated in PBS either i.v. (10 mg/kg) or orally (20 and 102 mg/kg). Rats were 7 weeks old and weighed 198 to 233 g (male) and 158 g to 178 g (female) at the start of the study. Animals were fasted for 16 to 19 h before drug administration. Body weights were recorded prior to dose administration, once weekly, and at sacrifice. Clinical signs were noted at approximately 1, 3, and 6 h post dose on the day of dosing and at least once daily on other days. Blood samples (~0.5 ml) were obtained from anesthetized rats (60:40 mixture of CO₂:O₂, to effect) at specified time intervals over 14 days. Blood samples from the 20-mg/kg treatment group were also processed for plasma by centrifugation at 2000 × g for 15 min. All blood and plasma samples were analyzed for AQ-13. Due to sampling volume limitations as required by animal care and use guidelines serial blood samples at each time point could not be collected from the same rat. Only four samples were collected from each rat.

Primates

The in-life phases of the single- and multiple-dose cynomolgus monkey studies were performed at the California Regional Primate Research Center (Davis, CA) and Sierra Biomedical, respectively.

For the single-dose study, 6 male cynomolgus macaques (2/dose group) were administered AQ-13 formulated in PBS either i.v. (10 mg/kg) or orally (20 and 100 mg/kg). Blood samples (~ 1 ml) were collected over a period of 7 days at specified time intervals by percutaneous femoral venipuncture. Primates ranged from 4 years, 6 months to 9 years, 8 months old, and weighed from 3.83 to 4.91 kg at the start of the study.

As part of a multiple dose toxicology study of AQ-13, blood samples were collected for toxicokinetic analysis. Each dose group consisted of 10 macaques (5/sex/dose group) that were orally administered a loading dose (LD) of 50, 100, or 200 mg/kg at time 0 on the first day of treatment. The corresponding maintenance doses (MD) of 25, 50, or 100 mg/kg (half the LD) were administered 12 and 24 h after the first dose and every 24 h thereafter through day 7. Blood samples were obtained by percutaneous femoral venipuncture from 5 animals/sex/timepoint on days 1 through 8 and from 3 animals/sex/timepoint on days 11 and 22. Blood samples were collected 2 h postdose on days 1 (LD), 2, 4, and 7 and on days 8, 11, and 22.

The protocol for each in-life study was approved by the performing institution’s Animal Care and Use Committee, and the studies were carried out in accordance with the National Research Council’s Guide for the Care and Use of Laboratory Animals as adopted and promulgated by the U.S. National Institutes of Health.

Analytical Method for Quantitation of AQ-13

Because previous studies have demonstrated that CQ concentration is higher in blood than plasma (Rombo et al. 1985), a comparison of plasma and whole blood concentrations of AQ-13 in samples collected in the rat study was conducted (using the 20-mg/kg oral treatment group only). Blood:plasma ratios were similar for both males and females, with most values ranging from 1.3 to 2.7. Therefore, whole blood was selected instead of plasma for determining the pharmacokinetic parameters for AQ-13 in the current study.

For all evaluations, the collected blood samples were stored frozen at approximately −20°C until analysis. AQ-13 levels were measured by HPLC with ultraviolet (UV) or fluorescence detection following liquid/liquid sample extraction, using a modification of a method previously reported for analysis of CQ and desethyl CQ in plasma (Karim et al. 1992). In the current study, UV detection was used for the majority of samples with higher blood concentrations of drug. Fluorescence detection was used for later time point samples expected to contain low levels of AQ-13 in order to improve the detection limit, but was not suitable for high-concentration samples because the linear range of the assay was too narrow. CQ was used as an internal standard for AQ-13, added prior to extraction.

The extraction method was as follows: The internal standard solution (15 μl of CQ in 50% methanol), saturated sodium chloride solution (150 μl), isopropanol (60 μl), and 2 M sodium hydroxide (150 μl) were added sequentially to the blood or plasma (150 μl), with brief vortex mixing between additions. The samples were extracted with hexane and after evaporation, reconstituted in mobile phase. They were then filtered through an Acrodisc Nylon 0.45-μm 13-mm syringe filter (Pall Gelman Sciences, Ann Arbor, WI) that had been prerinsed with mobile phase to remove extractables.

HPLC analysis was performed using a Beckman Ultrasphere ODS 5-μm C18 column (4.6 × 250 mm) at a flow rate of 1.0 ml/min, with a mobile phase of 13% acetonitrile, 87% buffer (0.1 M phosphoric acid with 0.2% triethylamine, pH = 3.0). UV detection was at 343 nm, fluorescence detection at excitation = 330 nm, emission = 370 nm. Sample injection volume was 100 μl. HPLC run times were 13 min, and AQ-13 and CQ eluted at approximately 5.5 and 10 min, respectively.

Standard curves (triplicate replicates of at least five concentrations of rat or cynomolgus macaque blood spiked with AQ-13) ranged from 0.1 to 10 μg/ml for UV detection and from 0.02 to 1.0 μg/ml for fluorescence detection. The limit of detection was 0.03 μg/ml using UV detection and 0.004 μg/ml using fluorescence detection. For all sample analyses, linearity of the standard curves, as measured by the correlation coefficient (r ²), was ≥.996, and back-calculated mean accuracy and precision of the calibration standards at each concentration were within ±15%. Blood standards were stable upon storage at −20°C for 6 months.

Metabolite Identification by LC-MS

Liquid chromatography–mass spectrometry (LC-MS) analysis was performed using a Vestec Model 201 mass spectrometer with Vestec Thermospray interface in full-scan mode from 150 to 400 m/z. The column used was a Beckman Ultrasphere ODS 5-μm, C18 column (4.6 × 250 mm). The mobile phase consisted of 0.05% trifluoroacetic acid in water (A) and 0.05% trifluoroacetic acid in 90% acetonitrile (B), with a gradient elution performed from 10% B to 25% B over 25 min at a flow rate of 1.0 ml/min. The AQ-13 metabolite peak and AQ-13 eluted at approximately 14 and 16 min, respectively. Blood samples collected at 8 h from the 100-mg/kg single-dose group were selected for the LC-MS analysis.

Analytical Method for Quantitation of Monodesethyl AQ-13

The metabolitewas quantitated by determining the ratio of the peak area of the metabolite to that of AQ-13 at each time point. This ratio was then multiplied by the concentration of AQ-13 in the sample to derive a concentration for the metabolite.

Plasma Protein Binding

For the protein binding experiment, six concentrations of AQ-13 ranging from 0 to 50 μg/ml (0 to 127 μM) were incubated in triplicate with plasma (rat, monkey, or human) at 37°C for 2 h. An aliquot of the medium was transferred to a Centrifree device and centrifuged at 1200 × g for 25 min. AQ-13 and protein (Bradford 1976) concentrations were determined before and after ultrafiltration. The average protein concentrations in the rat, monkey, and human plasma incubations were 62.76 ± 4.69, 64.21 ± 6.40, and 69.16 ± 6.01 mg/ml, respectively.

Data Analysis

All data were expressed in terms of μg/ml and then converted to micromolar concentration (μM). The single dose i.v. and oral data were analyzed by standard noncompartmental methods using RSTRIP version 2.02 (Micromath Scientific Software, Salt Lake City, UT) and WinNonlin version 1.5 (Pharsight Corporation, Mountain View, CA). Relevant pharmacokinetic parameters and constants were calculated on a molar basis. The following parameters were determined: t _½, terminal elimination half-life; trapezoidal AUC (area under the concentration-time curve); C ⁰, maximum concentration in blood immediately after i.v. administration; C _max, maximum concentration in blood after oral administration; t _max, time for C _max to be achieved; F, bioavailability of AQ-13 alone and AQ-13 plus monodesethyl AQ-13, calculated using AUC after oral and i.v. doses; CL, total body clearance after i.v. and oral administration; V _d , volume of distribution after i.v. and oral administration; and dose proportionality, ratio of AUC at high and low doses after both i.v. and oral administration.

RESULTS

Blood and Plasma Levels of AQ-13 in Rats

Blood samples were extracted and analyzed for all animals. Plasma samples were only analyzed on a subset of animals (20 mg/kg AQ-13 dose group). Analysis for AQ-13 in both plasma and blood samples collected prior to i.v. and oral administration (predose) revealed no detectable drug. There were no interfering peaks in the regions in which AQ-13 eluted, by either UV or fluorescence detection.

Plasma levels of AQ-13 peaked at a concentration of 1.2 ± 0.4 μM at 3 h after oral administration in male rats and at 1.3 ± 0.3μM at 8 h after oral administration in female rats and were undetectable after 3 days (Figure 2). Pharmacokinetic parameters in male and female rats after both i.v. and oral administration of AQ-13 are presented in Table 1. After i.v. administration, blood AQ-13 levels declined gradually with a terminal half-life of 9 to 10 h. Following oral administration of 20 and 102 mg/kg AQ-13, maximum concentrations of 2 and 7 μM, respectively, were achieved in 3 to 4 h. Thereafter concentrations declined gradually with terminal half-lives ranging from 13 to 40 h. There was no apparent gender difference. In general with a fivefold increase in dose, a sevenfold increase in AUC and a threefold increase in C _max were observed. The mean oral bioavailability for AQ-13 in male and female rats was 69.3% for the 20-mg/kg dose and 87.7% for the 102-mg/kg dose.

Metabolite Identification

An unknown peak eluting earlier than AQ-13 was consistently observed in spiked blood standards and in many blood samples from both rats and primates, but not in blood blanks or neat standards. For blood standards, the area of this peak was constant at 2% to 4% of the AQ-13 peak area. However, the unknown peak was present at a larger percent in the actual experimental blood samples and tended to increase with time following dose administration, suggesting that the peak represented a metabolite. LC-MS analysis revealed the unknown peak in the blood samples to have a monoisotopic integer mass of 263 for the neutral molecule, which agreed with the structure of monodesethyl AQ-13 (Figure 3).

Following i.v. administration of AQ-13 to rats, the metabolite concentration was highest at 5 min and was 4% to 6% of the AQ-13 concentration at this initial timepoint. After low (20 mg/kg) and high (102 mg/kg) oral doses, the highest concentration of the monodesethyl AQ-13 was reached between 6 and 8 h; the respective values were 34% to 35% and 32% to 33% of the AQ-13 concentrations at the same time point. The relevant pharmacokinetic parameters and constants calculated for the metabolite alone and for AQ-13 plus the metabolite in rats are summarized in Table 1.

Single-Dose Pharmacokinetics in Cynologus Macaques

The molar AQ-13 and monodesethyl AQ-13 blood levels for individual macaques (identified with unique numbers) are presented in Figure 4A through C . The pharmacokinetic parameters and constants for AQ-13, monodesethyl AQ-13, and AQ-13 plus monodesethyl AQ-13 in individual primates are summarized in Table 2.

Following i.v. administration, maximum concentrations, averaging 10 μM, were observed at the first sampling time point. After single oral doses of 20 and 100 mg/kg, mean C _max values averaging 1 and 16 μM were observed at 8 and 2 h, respectively. Overall, the rate of decline was similar after i.v. and oral administrations and concentrations decreased biphasically with a mean terminal half-life of approximately 36 to 38 h.

The AUC and C _max values for AQ-13 administered as an oral dose of 100 mg/kg were 10- to 15-fold higher than those obtained after a 20-mg/kg oral dose. In general, a twofold greater bioavailability was observed for AQ-13 after oral administration of 100 mg/kg as compared to the bioavailability obtained after 20 mg/kg, with mean values of 47.6% and 23.8%, respectively.

Following a single dose of AQ-13, the peak concentration for the metabolite was observed 8 to 12 h after 10 mg/kg i.v. and 20 mg/kg oral dose. In the animals that received an oral dose of 100 mg/kg AQ-13, high concentrations of the metabolite were observed as early as 20 min following dose administration, and its concentration remained nearly unchanged for 3 to 4 days, with the absolute values exceeding those for AQ-13 for nearly the entire course of the study. The levels of the metabolite were greater after oral administration than after i.v. administration. In general, the decline of the metabolite was similar to that of the parent drug after the 10 mg/kg i.v. and 20 mg/kg oral doses, but was much more prolonged after the 100-mg/kg oral dose.

The AUC for the metabolite after 100 mg/kg was approximately 16-fold higher than the value obtained after 20-mg/kg dose. However, the C _max for the metabolite after the higher AQ-13 oral dose was only about 3.8-fold higher. These results suggest that there is a lack of dose proportionality in AQ-13 pharmacokinetics that may be attributed to nonlinear bioavailability and saturable metabolite kinetics. The relationship between dose and pharmacokinetic parameters, AUC and C _max, in rats and monkeys are shown for AQ-13 and its major metabolite in Figure 5. The plot draws attention to the deviation from linearity. Furthermore, the ratios: (AUC)_AQ−13/(AUC)_{monodesethyl AQ−13} and (Cmax)_AQ−13/(Cmax)_{monodesethyl AQ−13} as summarized in Table 3 for rats and monkeys is also suggestive of saturable metabolite kinetics in monkeys.

Multiple-Dose Pharmacokinetics

Adverse clinical signs were observed in cynomolgus macaques that were administered a LD of 200 mg/kg; four animals in this dose group had repeated convulsions following administration of the LD. One female died approximately 1 h after the loading dose and one male died on day 5 of the study. Treatmentrelated adverse clinical signs following administration of AQ-13 included convulsions, emesis, decreased activity, hunched posture, and drooping eyelids; these effects were dose dependent in incidence and severity. Dose-dependent decreases in body weight and food consumption were also observed. All clinical signs, body weights, and food consumption returned to control levels by the end of the recovery period at the end of the study (day 22).

The highest AQ-13 levels in the blood were observed in both sexes on the same day—on day 1 after the 100-mg/kg MD, and on the day of final treatment (day 7) after the 25- and 50-mg/kg MDs. The highest AQ-13 levels (pooled means for both sexes) observed were 10.06 ± 0.52, 14.21 ± 2.50, and 29.29 ± 3.10μM after MDs of 25, 50, and 100 mg/kg, respectively. Figure 6 illustrates that these levels declined rapidly by day 22, to less than 1 μM. High concentrations of the metabolite, monodesethyl AQ-13, were observed by day 4, and these levels were consistently higher than the parent drug levels up to day 22. Both AQ-13 and metabolite levels were dose related, with higher levels observed in the animals that received larger doses, as shown for males and females in Figure 6. The maximum metabolite levels (mean values from males and females) observed in these three dose groups were in the range of 17 to 35 μM and were achieved on day 8, 1 day after treatment was stopped. Typically, the decline of the metabolite was similar to that of the parent drug in the animals that received the lowest MD. As observed in the single-dose study, the decline was more prolonged after higher MDs (50 and 100 mg/kg).

Plasma Protein Binding

AQ-13 concentration in all incubates, including controls (to evaluate nonspecific binding to the membranes), were within ± 11% of the theoretical concentration. Less than 2% of the total plasma protein was recovered in the ultrafiltrate, indicating that AQ-13 recovered in this fraction is almost free of plasma proteins. The percentage of AQ-13 bound to rat, primate, and human proteins decreased marginally with an increase in total AQ-13 concentration. At the lowest concentration, the values of AQ-13 bound were 65.6% ± 3.1%, 74.2% ± 6.5%, 55.2% ± 0.0% for rat, primate, and human, respectively. The plots of the plasma protein binding data (total versus bound) for the species were linear in the tested concentration range, with slopes of 0.47, 0.58, and 0.48 for rat, primate, and human, respectively (Figure 7).

DISCUSSION

In the present study, measurement of AQ-13 levels in rat blood and plasma samples collected from the low-dose groups (20 mg/kg) revealed that the drug levels in blood were consistently greater than in plasma, for both sexes. The detection limits of the assay method were low enough to allow measurement of the drug over a prolonged period and therefore accurate estimates of AUC and pharmacokinetic parameters were obtained. By using HPLC with fluorescence detection for the later time point samples, blood levels as low as 4 ng/ml could be detected. These methods not only demonstrated the approximately twofold higher AQ-13 levels in blood versus plasma, but also avoided the reported problems associated with plasma separation for CQ (Rombo et al. 1985). Interestingly, AQ-13 blood levels in the current study were not elevated over plasma levels to as great an extent as has been reported for CQ. Frisk-Homberg and colleagues (1984) reported that the blood:plasma ratios for CQ was as high as 10; however, the CQ doses administered in their study were much lower (2.08 to 8.3 mg/kg) than the doses of AQ-13 administered in the current study, and blood:plasma ratios for CQ decreased at the higest dose studied. Therefore, a possible explanation for the observed lower blood:plasma ratios for AQ-13 is that saturation of binding in whole blood and other tissues occurs as the dose administered is increased.

An unknown peak, identified by LC-MS as monodesethyl AQ-13, was observed in many sample chromatograms at an earlier retention time than AQ-13. This peakwas significantly larger in rat and cynomolgus macaque blood samples after oral than after i.v. administration, which is consistent with a metabolite formed by the gastrointestinal tract and/or liver. It is well established that the major metabolite of CQ, the structural analog of AQ-13, is monodesethyl chloroquine (Tracy and Webster, 1996; Ducharme and Farinotti 1996; Bergqvist and Frisk-Holmberg, 1980; Frisk-Holmberg et al. 1984; McChesney, Fasco, and Banks 1962). The data generated in the present studies indicate that AQ-13 is metabolized similarly in vivo.

After a single dose to macaques, peak AQ-13 concentrations were observed earlier than the peak metabolite concentrations, consistent with a lag time associated with the formation of the metabolite. The metabolite levels were four- to ninefold higher than the AQ-13 concentrations at the time of peak metabolite concentration after the 20- and 100-mg/kg single oral doses. However, in rats after administration of similar doses (20 and 102 mg/kg), the peak metabolite concentration was only 32% to 35%of the AQ-13 concentration at that time. Thus, the formation of the metabolite appears to be markedly higher in macaques than in Sprague-Dawley rats. The low bioavailability of AQ- 13 in macaques, compared to the bioavailability in rats, may be attributed to the rapid metabolism of the drug prior to its reaching the systemic circulation. A comparison of the fraction absorbed by macaques (F) calculated using the AUC of the parent drug plus metabolite supports this idea, as the value is much higher than when F is calculated with the AUC of the parent drug alone. This is not true for the rat data, which show little impact on the F determined based on the AUC of AQ-13 only, compared to F calculated using the AUC for parent drug and metabolite combined. The bioavailability of AQ-13 plus metabolite may be biologically relevant, as monodesethyl chloroquine has been shown to have antimalarial activity, contributing substantially to the therapeutic efficacy of CQ (Ducharme and Farinotti 1996).

A dose proportionality ratio of 6.6 was obtained for AQ-13, which deviated from the expected value of 5; studies of CQ pharmacokinetics in rats performed by our laboratory showed a dose proportionality ratio of 9.4 after oral administration (Ramanathan-Girish et al. 1998), indicating nonlinear kinetics for both drugs. Evidence for CQ dose-dependent kinetics, probably due to changes in CQ disposition and elimination, has been published by Frisk-Holmberg et al. (1979, 1985). Half-lives of 3.1, 42.9, and 312 h were observed for CQ doses of 250, 500, and 1000 mg, respectively, administered to human volunteers. In contrast, a study by Gustafsson et al. (1983) does not support capacity-limited CQ elimination.

After 10 mg/kg i.v. and 20 mg/kg oral doses to macaques, the metabolite-time profile may be described as showing a formation rate limitation. The declining slope and the terminal half-life for monodesethyl AQ-13 is approximately the same as parent AQ-13. However, after a higher oral dose of AQ-13, the profile was described as showing an elimination rate limitation. The metabolite levels in plasma exceeded AQ-13 levels within 20 min after dosing and the declining phase for monodesethyl AQ-13 was observed only after a time period equivalent to 5- half-lives of the parent AQ-13. A significant fraction of AQ-13 is probably converted to metabolite because the concentration of monodesethyl AQ-13 exceeded AQ-13 plasma levels quite rapidly after dosing. Alternatively, the volume of distribution of the more polar metabolite may be lower than AQ-13, thus resulting in higher blood concentrations of metabolite. The C _max for the metabolite after a single 100-mg/kg dose of AQ-13 was approximately 3.8-fold higher than that after a 20-mg/kg dose, a value not directly proportional to the observed 15-fold increase in the peak AQ-13 levels. This lack of dose proportionality suggests that the metabolic pathway is saturated after a single 100-mg/kg dose.

Collectively, our results strongly indicate that the overall blood disposition profile of AQ-13 in rats is similar to that of CQ (Ramanathan-Girish et al. 1998). Absorption was fairly rapid, and the highest AQ-13 levels were observed within 1 to 5 h after oral administration and declined gradually thereafter. The oral bioavailability of AQ-13 after a low dose was approximately 70% and was slightly higher at the higher dose level; these values are similar to those observed for CQ in rats (Ramanathan-Girish et al. 1998).

Although the bioavailability of AQ-13 was lower in nonhuman primates than in rats, the toxic effects of the drug in cynomolgus macaques were more severe. In a 7-day oral repeatdose toxicity study performed in Sprague-Dawley rats (Riccio et al. 1998), no treatment-related deaths occurred at any dose (25.4, 50.8, 101.5 mg/kg), although adverse reactions including ataxia, hypoactivity, and hindlimb weakness were observed. In the repeat-dose study in cynomolgus macaques, the rapid increase to and maintenance of elevated blood levels of AQ-13 and its metabolite are consistent with the onset and persistence of clinical signs, particularly in the high-dose group that received the MD of 100 mg/kg. With this regimen, significant clinical effects were observed, including seizures and lethality after administration of the 200-mg/kg LD. Effects were less pronounced at lower doses. Although AQ-13 levels had declined to less than 1 μM in all the three dose groups by the end of the study, and the metabolite levels were still elevated in the two higher dose groups, suggesting involvement of the metabolite in the toxic effects that were observed.

In summary, AQ-13 is a new antimalarial agent that is active against CQ-resistant organisms. The current study indicates that AQ-13 is converted to monodesethyl AQ-13 by rats and cynomolgus macaques following i.v. and oral administration. The concentration of monodesethyl AQ-13 exceeds the AQ-13 drug levels at many time points. The bioavailability of AQ-13 is lower in macaques, compared to rats, which is probably due to its greater first-pass metabolism of AQ-13 to monodesethyl AQ-13. There was evidence of saturation of the metabolism pathway at the high oral dose and accumulation of the metabolite after repeat dosing. Overall, the pharmacokinetics and metabolism of AQ-13 appears very similar to the disposition of CQ that has been reported extensively in the literature, and AQ-13 is a promising new candidate for malaria therapy.