Usefulness of a Nanoparticle Formulation to Investigate Some Hemodynamic Parameters of a Poorly Soluble Compound

Usefulness of a Nanoparticle Formulation to Investigate Some Hemodynamic Parameters of a Poorly Soluble Compound

Usefulness of a Nanoparticle Formulation to Investigate Some Hemodynamic Parameters of a Poorly Soluble Compound 2 ¨ KALLE SIGFRIDSSON,1,3 JAN-ARNE BJ...

342KB Sizes 0 Downloads 52 Views

Usefulness of a Nanoparticle Formulation to Investigate Some Hemodynamic Parameters of a Poorly Soluble Compound 2 ¨ KALLE SIGFRIDSSON,1,3 JAN-ARNE BJORKMAN, PIA SKANTZE,1 HELEN ZACHRISSON2 1

¨ ¨ Pharmaceutical Development, AstraZeneca R&D Molndal, S-431 83 Molndal, Sweden

2

¨ ¨ Bioscience, AstraZeneca R&D Molndal, S-431 83 Molndal, Sweden

3

¨ Department of Chemical and Biological Engineering, Chalmers University of Technology, SE-412 96 Goteborg, Sweden

Received 10 June 2010; revised 30 September 2010; accepted 18 November 2010 Published online 22 December 2010 in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/jps.22440 ABSTRACT: Drug solubility is an important issue when progressing investigational compounds into clinical candidates. The present paper describes the development and characterization of a nanosuspension that was formulated to overcome problems with poor water solubility and possible adverse events caused by cosolvent mixtures, using ticagrelor as a model compound. A homogeneous nanosuspension of ticagrelor was formed using a wet milling approach, which yielded particle sizes around 230 nm. The nanosuspensions were chemically stable for at least 10 months at both room temperature and when refrigerated, and physically (i.e., particle size) stable for at least 10 months under refrigeration, and approximately 3 years at room temperature and when frozen. One rat model and two dog models were used to assess the pharmacokinetics and hemodynamic-related effects following intravenous administration of nanoparticles. There were no biologically consistent or dose-dependent effects of the nanoparticles on the hemodynamic parameters tested, that is, heart rate, mean aortic pressure, cardiac output, left femoral artery blood flow, or cardiac inotropy (measured as max dP/dt). In conclusion, a stable ticagrelor nanosuspension formulation was developed, suitable for intravenous administration. At the doses evaluated, this formulation was without hemodynamic effects in three sensitive preclinical models. © 2010 Wiley-Liss, Inc. and the American Pharmacists Association J Pharm Sci 100:2194–2202, 2011 Keywords: ticagrelor; intravenous; nanoparticles; injectables; stability; pharmacokinetics; preclinical

INTRODUCTION Drug solubility is an important issue when progressing investigational compounds into clinical candidates. More than 40% of compounds identified through combinatorial screening programs have poor aqueous solubility, thereby reducing bioavailability.1 Molecules with poor aqueous solubility are difficult to formulate using conventional approaches. Typical methods to enhance solubility include: Salt generation,2 micronization,3 conversion to amorphous material,4 use of cosolvents, surfactants, complexing agents, or various emulsions.5–10 However, such methods often involve significant amounts of adCorrespondence to: Kalle Sigfridsson (Telephone: +46-31776-2246; Fax: +46-31-776-3768; E-mail: carl-gustav.sigfridsson @astrazeneca.com) Journal of Pharmaceutical Sciences, Vol. 100, 2194–2202 (2011) © 2010 Wiley-Liss, Inc. and the American Pharmacists Association

2194

ditives to increase solubility to the millimolar range, which may induce unwanted side effects both in preclinical and clinical studies.11 An alternative method is to formulate drugs as nanometer-sized particles in suspension.12–16 Nanosuspensions are submicron colloidal dispersions of drug particles which are stabilized by surfactants.17,18 Increased dissolution velocity and saturation solubility of a drug can be achieved by using nanosuspensions,19 thereby improving oral absorption of poorly soluble drugs and increasing bioavailability compared with traditional formulations.20,21 Furthermore, a particular advantage of nanosuspensions is that, due to a sufficiently small drug particle size, they can be administered via other routes, for example, ocular,22 pulmonary,23 and intravenously.24 Inhibition of platelet aggregation is now a standard treatment for patients with acute coronary

JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 100, NO. 6, MAY 2011

USEFULNESS OF A NANOPARTICLE FORMULATION

2195

The current paper describes the development of a ticagrelor nanosuspension. This formulation was then rigorously assessed in three new sensitive models for investigating hemodynamic parameters in anesthetized rats and dogs.

MATERIALS AND METHODS Materials and Chemicals Figure 1. Structure of Ticagrelor. All chiral centers are in the S-configuration except those marked ∗, which are in the R-configuration.

syndromes, including P2Y12 receptor antagonists, R such as clopidogrel (Plavix , marked by BristolMyers Squibb and Sanofi-Aventis), plus the use of aspirin.25 Ticagrelor (formerly known as AZD6140, Fig. 1) is the first, oral, reversibly binding antagonist of the P2Y12 receptor, which blocks adenosine-5 diphosphate-induced platelet aggregation.26 For ticagrelor, the development of an intravenous formulation was hampered by its poor aqueous solubility. The solubility of crystalline ticagrelor in water is approximately 18 :M at 25◦ C and 14 :M in pH 7.4 buffer. As ticagrelor is not ionizable, pH shifting was not an option to improve solubility. Furthermore, the achievement of millimolar solutions of ticagrelor, required for in vivo studies, was only possible by using cosolvents or solubilizing mixtures. Previous work with another poorly soluble compound (AZ77) indicated that a mixture of various solvents resulted in hemodynamic changes (e.g., fluctuations in aortic pressure), whereas a nanosuspension was without such effects (Fig. 2).27 Thus, a nanosuspension was the method of choice for the development of formulation suitable for intravenous administration, with ticagrelor as a model compound because such a formulation is composed mainly of water and the compound.

Ticagrelor (100% pure) was supplied by AstraZeneca R&D (M¨oIndal, Sweden). Polyvinylpyrrolidone K30 (PVP) was purchased from BASF (Gothenburg, Sweden). The disodium salt of Aerosol OT (AOT) was obtained from Cytec Industries Inc., (Kalamazoo, Michigan). Mannitol was supplied by Sigma–Aldrich (Stockholm, Sweden). All other reagents were of analytical grade. Zirconium oxide milling beads (0.6–0.8 mm diameter) were obtained from Glen Creston Ltd. (Stanmore, Middlesex, UK). Before use, milling beads were washed by continuous stirring in sodium hydroxide (1 M) for 24 h, then rinsed four times with approximately 2 L of ultrapure water ¨ (ELAG, AB Ninolab, Uplands Vasby, Sweden). The beads were then washed in sodium dodecyl sulfate (1%, w/v in water) by continuous stirring for 24 h, and rinsed four times with ultrapure water. Finally, the beads were dried in a drying cabinet at 60◦ C for 48 h. Preparation of Crystalline Ticagrelor Nanosuspensions Ticagrelor (60 mg) was weighed into a vial and a stabilizer solution consisting of PVP (1.33%, w/w)/AOT (0.066%, w/w) in water was added to a final weight of 600 mg. The resulting 10% (w/w) crude suspension was stirred and sonicated for 10 min to give a welldispersed slurry. The vessel was sealed and the slurry milled (Fritsch Planetary Micromill P7, Fritsch, IdarOberstein, Germany) at 700 rpm, for four periods of 30 min with 15 min intervals between each milling phase, to prevent temperature rises above 50◦ C. The resulting formulation was either used for administration (after appropriate dilution-–see below) within 24 h, or stored frozen at −20◦ C before use to prevent growth of microorganisms. The bulk substance was analyzed for microbiological contamination and met AstraZeneca specifications for intravenous administration. Milling beads were only used once and discarded. All equipment in contact with the formulation was washed four times in water and then stored in 70% of ethanol (v/v in water) for more than 24 h. Characterization of Nanosuspensions

Figure 2. Example of two formulations of a poorly soluble compound (AZ77) with (cosolvent) and without (nanosuspension) a hemodynamic effect in an animal model (n = 1), similar to dog model 2. DOI 10.1002/jps

The volume-averaged particle size (diameter) of the crystalline ticagrelor suspensions was measured by laser diffraction (Malvern Mastersizer 2000, Malvern Instruments Ltd., Malvern, Worcestershire, UK) JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 100, NO. 6, MAY 2011

2196

SIGFRIDSSON ET AL.

using the Mie theory with a particle refractive index of 1.59. Larger particles were removed by sedimentation as appropriate, and the particle size reevaluated. Both the chemical and physical stabilities of ticagrelor nanosuspensions (1, 20, or 50 mM) were examined under different storage conditions. Nanosuspensions were stored at room temperature (up to 32 months), and under refrigeration (nominally 4◦ C) for up to 10 months. Samples were also stored frozen (−20◦ C) for up to 40 months and thawed prior to analyses. Ticagrelor concentrations and the particle size distribution were evaluated in the nanosuspensions on the day of preparation, and in all stored samples. Before evaluation of particle size, the formulations (1 and 50 mM) were diluted to approximately 0.1–0.6 mM within the Malvern Mastersizer 2000, according to the supplier’s instructions. Hemodynamic Assessments

Animals The animal studies were approved by the Ethical Committee for Animal Research at the University of Gothenburg, Sweden, and conducted in accordance with the US National Institute of Health Guide for the Care and Use of Laboratory Animals.28 R TM R TM   Male, (350–400 g) Hsd:Sprague–Dawley SD rats were supplied by Harlan Nederland (Horst, the Netherlands). The animals were housed under standard housing conditions and allowed to acclimatize for 1 week prior to use. Food and water were allowed ad libitum. Twenty-eight beagle dogs (both sexes, 11–18 kg) ¨ were provided by the Rååh¨ojden Kennel (Orkelljunga, Sweden). These dogs were group-housed under standard conditions of light and temperature for 1 week prior to use.

Preparation of Nanosuspensions for Dosing For all in vivo evaluations and stability studies, the stock nanosuspension of ticagrelor was diluted to the required concentrations with 5% mannitol (w/w) in water. The vehicle control was 5% of mannitol. Mannitol was used as a tonicity modifier in the present formulations. The nanosuspensions were ultrasonicated just prior to administration.

Rat Study The hemodynamic effects of a ticagrelor nanosuspension were evaluated in a ferric chloride (FeCl3 ) induced model of thrombosis. Rats were anesthetized R , Research with sodium thiobutabarbital (Inactin Biochemicals international, Natic, Massachusetts), intubated, and the body temperature maintained at 38◦ C by external heating. The left femoral artery was catheterized for both blood pressure and heart rate measurements, which were monitored continuJOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 100, NO. 6, MAY 2011

ously by a custom-made computer program (Pharmlab V4.0, AstraZeneca R&D). Thrombus induction and blood flow monitoring were conducted as previously described with one exception; the control period prior to drug infusion was 5 min instead of 10 min.29 Ticagrelor nanosuspensions were administered intravenously via a catheter in the left jugular vein at five dose levels (n = 4 rats per dose level). Bolus doses of ticagrelor nanosuspensions (2.25–225 :g/kg) were given, followed by infusion of 0.3 to 30 :g/(kg min) for 70 min. Each infusion started 15 min prior to FeCl3 exposure.

Dog Model 1 In a modified Folts model,29,30 12 dogs were anesthetized with sodium pentobarbital, intubated, and ventilated. The right carotid artery was catheterized for blood pressure recordings, and needle electrodes were used for electrocardiographic assessment. Blood flow monitoring under restrictive conditions was conducted as previously described,29 with an additional assessment of blood flow in the left femoral artery as a reference. Blood pressure, heart rate, and blood flow were recorded on a Grass Polygraph (7D, Grass Instruments, Quincy, Massachusetts) and were monitored continuously by a custom-made computer program (Pharmlab V4.0). Animals received either intravenous administration of vehicle [5% of mannitol w/w; infusion rate 0.01 mL/(kg min); n = 6] or increasing doses of ticagrelor nanosuspensions (n = 6) via the left jugular vein. Ticagrelor nanosuspension administration was preceded by an initial 30-min control period during which vehicle alone was infused, followed by five 30min drug infusion periods. During these dosing periods, ticagrelor nanosuspensions were administered at progressively increased dose levels, that is, a bolus intravenous dose over 1 min followed by a 29-min infusion. The doses evaluated were as follows: 0.75 + 0.1, 2.25 + 0.3, 7.5 + 1, 22.5 + 3, and 75 + 10 :g/kg and :g/(kg min) bolus + infusion, respectively.

Dog Model 2 Sixteen dogs were anesthetized with "-chloralose. A dual sensor pressure transducer (Millar MicroTip SPC-771, Millar Instruments, Houston, Texas) was inserted into the right carotid artery. The proximal transducer was placed in the ascending aorta and the distal transducer in the left ventricle, for the measurement of aortic blood pressure, and cardiac inotropy (max dP/dt), respectively. Aortic blood pressure, cardiac inotropy, and heart rate were measured continuously. These parameters were also assessed after (a) temporary (1 min) occlusion of the left anterior descending (LAD) artery and (b) direct infusion of adenosine at two concentrations (15 :g/ min at 0.3 mL/min and 30 :g/min at 0.6 mL/min) DOI 10.1002/jps

USEFULNESS OF A NANOPARTICLE FORMULATION

intracoronary. Data sampling and analysis were assessed using a custom-made computer program (Pharmlab V4.0). Dogs were randomized by a simple, manual randomization method to receive intravenous administration (via the right saphenous vein) of either vehicle (n = 8) or two doses of ticagrelor nanosuspensions [n = 8; 210 :g/kg bolus and 30 :g/(kg min) infusion followed by 700 :g/kg bolus and 100 :g/(kg min) infusion]. Each experiment consisted of a 75-min control period followed by two consecutive 75 min periods during which either vehicle or ticagrelor nanosuspension was given as bolus dose in the first minute followed by a continuous infusion for the next 74 min. During each 75-min period, the blood flow response to (a) LAD occlusion was tested three times and (b) adenosine infusion was evaluated once.

Statistical Analyses Mean values were calculated over 5-min intervals for individual animals. Hemodynamic data from the three animal models were then expressed as mean ± standard error of the mean (SEM) for each time point. Analytical Methods

Evaluation of Nanosuspensions Ticagrelor concentrations in nanosuspensions were evaluated using a reversed-phase liquid chromatography gradient system with ultraviolet detection. An XBridge Shield RP18 column (3.5 :m, 3.0 × 50 mm2 ; Waters, En Yvelines Cedex, France) was used with a mobile phase of water (90%, v/v)/acetonitrile (10% v/ v) with 0.03% (v/v) trifluoroacetic acid. The detection wavelength was 280 nm.

Plasma Concentrations of Ticagrelor Blood samples for the evaluation of ticagrelor were collected into trisodium citrate tubes (0.106 mol/L; 9 volumes of blood:1 volume of citrate), from the rats via the catheter in the left femoral artery at the end of the experiment. Dog blood samples were collected into potassium–EDTA tubes. In the Folts model, samples were taken from the left jugular vein after 20-min infusion for each dose level. In the second dog model, blood samples were collected from the right femoral artery immediately at the end of each infusion period. Plasma was separated by centrifugation (10,000 × g, 5 min, 4◦ C) and stored frozen (−20◦ C) until analyzed. After protein precipitation, ticagrelor plasma concentrations were evaluated using liquid chromatography with tandem mass spectrometry. The mean intrabatch accuracy and precision were 91.9%–109.0% and 4.0%–8.4%, respectively. The lower limit of quantification was 5 ng/mL. DOI 10.1002/jps

2197

RESULTS Characterization and Stability of Ticagrelor Nanosuspensions X-ray diffraction studies of 1 and 50 mM ticagrelor nanosuspension formulations showed the crystal structure to be unchanged by the milling process (data not shown). The volume-weighted mean size of the ticagrelor nanosuspension particles (1 and 50 mM) was approximately 230 nm. In both formulations, more than 90% of particles were less than 300 nm; d(0.9) (90th percentile) was approximately 300 nm; d(50) (50th percentile) was 130 nm; d (0.1) (10th percentile) was 80 nm. In addition, numberweighted mean, surface-weighted mean, and specific surface area were 60 and 120 nm, and 50 m2 /g, respectively. Dilution had no effect on the particle size. Ticagrelor was chemically stable in both nanosuspension formulations for at least 10 months at room temperature, under refrigeration (nominally 4◦ C), and when stored frozen at −20◦ C for 3 weeks. The particle size distribution was unaffected by storage for at least 10 months under refrigeration (nominally 4◦ C), and approximately 3 years at room temperature and when stored frozen at −20◦ C [Table 1, with no d(0.9) value above 330 nm, and no d(0.1) value below 60 nm]. Figure 3 shows that the particle size distribution in a 20 mM ticagrelor nanosuspension after milling and after storage at −20◦ C for 3 years were very similar. Volume-weighted mean size after 3 years was 201 nm: d(0.1) = 66 nm; d(0.5) = 131 nm; d(0.9) = 291 nm. The number-weighted mean, surface-weighted mean, and specific surface area were 60 and 113 nm, and 51 m2 /g, respectively. On very rare occasions, there was some tendency for particle aggregation and sedimentation during storage. However, such samples immediately redispersed to the original particle size distribution following ultrasonication for 10 s, and remained so for more than 24 h postsonication. Thus, the stored samples were routinely sonicated for at least 10 s prior to evaluation. Hemodynamic Measurements

Rat Model Mean heart rate was unaffected by increasing doses of ticagrelor nanosuspensions (Fig. 4a). Overall, the mean aortic pressure was also not affected by the doses of ticagrelor nanosuspension evaluated (Fig. 4b). Initially, there appeared to be a slight transient rise in mean aortic pressure (∼10 mmHg) that peaked approximately 5 min after removal of the FeCl3 -saturated filter paper. The mean aortic pressure returned to baseline values within approximately 20 min (Fig. 4b). This event was not dose related and was due to the technique. Heart rate and JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 100, NO. 6, MAY 2011

2198

SIGFRIDSSON ET AL.

Table 1.

Stability of Particle Size in Ticagrelor Nanosuspensions Under Various Storage Conditions Particle Sizea (nm) After Storage for

Ticagrelor Concentration (mM)

Particle Sizea Storage (nm) on Day 0b Temperature

1

234

20 50

202 236

Room ◦ 4 C ◦ −20 C ◦ −20 C Room ◦ 4 C ◦ −20 C

3 weeks − − 240 − − − 230

3 months − − − − 256 − −

10 months

>30 months

230 235 − − 263 243 243

− − − 201c 229d − −

a Volume-weighted

mean size. the day of preparation. c At 40 months. d At 32 months. –, Not determined. b On

mean aortic pressure were not affected by ticagrelor nanosuspensions.

Dog Model 1 In the modified Folts model, increasing doses of ticagrelor nanosuspension had no effect on mean heart rate in anesthetized dogs compared with vehicle controls (Fig. 5a). Mean aortic pressure was also unaffected by rising plasma concentrations of ticagrelor nanosuspensions in comparison with vehicle (Fig. 5b). Infusion of either vehicle or ticagrelor nanosuspensions had no effect on left femoral artery blood flow (Fig. 5c). Heart rate, mean aortic pressure, and left femoral artery blood flow were not affected by ticagrelor nanosuspensions.

Dog Model 2 Figure 6a shows that two dosing regimens of ticagrelor nanosuspensions had no effect on mean heart rate versus vehicle control in the second dog model. Neither dosing scheme affected the mean aortic pressure versus control (Fig. 6b). Compared with vehicle control, cardiac inotropy (evaluated by max dP/ dt) was not affected by either regimen of ticagrelor

nanosuspensions (Fig. 6c). Heart rate, mean aortic pressure, and dP/dt were not affected by ticagrelor nanosuspensions. Plasma Concentrations of Ticagrelor In the rat model, increasing doses of ticagrelor nanosuspension resulted in corresponding linear increases in plasma concentrations. At the end of the experiment, the mean ± SEM plasma levels of ticagrelor were 12 ± 0.8, 48 ± 1.8, 161 ± 4.1, 565 ± 28, and 1634 ± 50 nM, respectively, for the 2.25/0.3, 7.5/ 1, 22.5/3, 75/10, and 225/30 [:g/kg bolus dose and :g/ (kg min) infusion dose] ticagrelor nanosuspensions. Increasing doses of ticagrelor nanosuspensions also resulted in increases in ticagrelor plasma concentrations in the modified Folts dog model. For example, the mean ticagrelor ± SEM plasma concentration immediately following 0.75 :g/kg + 0.1 :g/(kg min) was 17.9 ± 0.4 nM. The mean ± SEM concentration of ticagrelor was 762 ± 59 nM after 75 :g/kg + 10 :g/ (kg min). In the second dog model, the mean ± SEM plasma levels of ticagrelor were 4.1 ± 1.5 and 13.4 ± 2.1 :M, at the end of the dosing intervals with 210 :g/kg

Figure 3. Distribution of particle size in a 20 mM of ticagrelor nanosuspension after milling, and after storage at −20◦ C for 3 years. JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 100, NO. 6, MAY 2011

DOI 10.1002/jps

USEFULNESS OF A NANOPARTICLE FORMULATION

2199

Figure 4. Effect of ticagrelor on (a) heart rate and (b) aortic pressure in anesthetized rats (mean ± SEM, n = 4).

(bolus) + 30 :g/(kg min) and 700 :g/kg (bolus) + 100 :g/(kg min), respectively.

DISCUSSION A wet milling method using two stabilizers was developed to prepare nanosuspensions of ticagrelor, with adequate chemical and physical stability. This formulation administered intravenously did not exhibit hemodynamic effects in several sensitive preclinical models. Developing formulations suitable for oral and parenteral administration routes is an integral part of drug discovery and development. Both the physical and chemical properties of ticagrelor, resulting in poor aqueous solubility, necessitated the development of a nanosuspension suitable for intravenous administration. Several key criteria need to be met for a nanosuspension to be a successful drug formulation, including both chemical and physical stabilities, adequate particle size, and type and amount of additives to minimize toxicity and maintain tonicity.

DOI 10.1002/jps

In developing pharmaceutical formulations, one challenge is achieving acceptable chemical stability to ensure adequate storage time before the amount of active constituent falls below an acceptable threshold and/or the concentration of degradation products increase to levels that may be toxic. Generally, one advantage of suspensions versus solutions is improved chemical stability because the chemical stability of a drug in suspension is more closely related to that of the solid-state molecule.31 In the present study, ticagrelor was shown to be chemically stable in the nanosuspension formulation for the duration of the period investigated, that is, at least 10 months at room temperature and under refrigeration. These results are in accordance with those obtained for room temperature stability of the bulk substance.27 In contrast, ticagrelor concentrations in phosphate buffered solution at pH 7.4 declined after 1 month under normal laboratory conditions of light and temperature.27 Thus, the nanosuspension of ticagrelor prepared by the method described meets the criteria of adequate chemical stability.

JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 100, NO. 6, MAY 2011

2200

SIGFRIDSSON ET AL.

Figure 5. Effect of ticagrelor on (a) heart rate, (b) aortic blood pressure, and (c) left femoral artery blood flow in anesthetized dogs (Folts model; mean ± SEM, n = 6).

Figure 6. Effect of ticagrelor on (a) heart rate, (b) aortic pressure, and (c) cardiac inotropy in anesthetized dogs (mean ± SEM, n = 8).

Physical stability of the formulation is also an essential requirement for a nanoparticulate system. Poorly formulated nanoparticles can aggregate over time forming larger particles, thereby creating a potential safety concern. Ostwald ripening (difference in local solubility as a function of particle size leads to material transport from smaller to larger particles) will result in an increase in mean particle size with time. Minimizing the presence of large particles is critical for intravenous administration, in particular, the presence of particles larger than 5 :m should be avoided as these could block capillaries.32 Nanoparticles may also sediment over time in storage, and may be difficult to resuspend if not adequately stabilized. Aggregation, Ostwald ripening, and sedimentation issues can be controlled during processing and storage by carefully selecting excipients and stabilizers. The physical stability and adequate control of particle size were demonstrated for ticagrelor nanosuspensions. The large crystals present in the bulk prepara-

tion of pure ticagrelor were milled in the presence of minimal amounts of surface active agents. Two stabilizers were used; PVP for steric stabilization and AOT for electrostatic stabilization of the ticagrelor particles. Through their use, the particle size distribution of ticagrelor (approximately 230 nm) remained unchanged throughout the storage conditions evaluated, meeting the criteria for physical stability and adequate particle size. Maximizing exposure to a drug is important in assessing its toxic potential. An advantage of nanosuspensions, consisting essentially of pure drug, is their rapid dissolution (in order of seconds to minutes), thereby achieving pharmacokinetic properties similar to those of a solution. For example, Clement et al.33 demonstrated that a nanosuspension of flurbiprofen had the same exposure and in vivo distribution as a solution of the drug after intravenous injection to rats. Similarly, the pharmacokinetics of AZ68, a poorly soluble compound, when administered intravenously as

JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 100, NO. 6, MAY 2011

DOI 10.1002/jps

USEFULNESS OF A NANOPARTICLE FORMULATION

a solution or as a nanosuspension was similar; AZ68 is approximately five times less soluble than ticagrelor at physiological pH.34 In the present study, the plasma concentrations after intravenous administration of ticagrelor nanosuspensions were less than the exposures obtained after oral administration to rats and dogs.27 This observation suggests that the ticagrelor particles dissolved upon dilution in blood allow for a solution-like distribution. A potential risk with intravenous administration is the precipitation of the drug within the circulatory system either due to pH shift (e.g., as for a pHadjusted solution), or to a dilution effect (e.g., as for a cosolvent formulation). However, ticagrelor is unlikely to form larger particles because it is neutral at physiological pH, and apparently dissolves rapidly upon dilution in blood. The development of a suitable nanosuspension formulation allowed for the robust evaluation of ticagrelor on hemodynamic parameters in several in vivo models, including animal models of thrombosis.29 In the rat model, no effect on heart rate was observed, and, although there was a transient increase in aortic pressure with administration of ticagrelor nanoparticles, this effect was not dose related and resolved quickly with continued infusion of the drug. In the dog, there was also no obvious effect of the ticagrelor nanosuspension, at the doses evaluated, on the various hemodynamic parameters assessed in either model. In addition, as required for nanosuspensions, the low levels of additives did not result in hemodynamic effects. Such a nanosuspension formula of ticagrelor has been used intravenously to study the effects of this compound on adenosine-mediated hyperemia responses in a canine preclinical model.35

CONCLUSIONS Ticagrelor nanosuspensions were successfully prepared using a simple wet milling technique with minimal amounts of stabilizers. Such formulations were chemically and physically stable during storage. Furthermore, this nanosuspension formulation of ticagrelor was without effect on hemodynamic parameters in preclinical models.

ACKNOWLEDGMENTS Financial support for the conduct of the research was provided by AstraZeneca. The authors thank Anders Lundqvist for evaluating plasma concentrations of ticagrelor. The authors acknowledge Jackie Phillipson, PhD (Gardiner-Caldwell Communications, Macclesfield, UK) for assistance in editing the manuscript, funded by AstraZeneca. Financial disclosure: All authors are employees of AstraZeneca. DOI 10.1002/jps

2201

REFERENCES 1. Lipinski CA. 2002. Poor aqueous solubility—an industry wide problem in drug discovery. Am Pharm Rev 5:82–85. 2. Serajuddin ATM. 2007. Salt formation to improve drug solubility. Adv Drug Deliv Rev 59:603–616. 3. Hargrove JT, Maxson WS, Wentz AC. 1989. Absorption of oral progesterone is influenced by vehicle and particle size. Am J Obstet Gynecol 161:948–951. 4. Hirasawa N, Ishise S, Miyata H, Danjo K. 2003. An attempt to stabilize nilvadipine solid dispersion by the use of ternary systems. Drug Dev Ind Pharm 29:997–1004. 5. Loftsson T, Brewster ME. 1996. Pharmaceutical applications of cyclodextrins. J Pharm Sci 85:1017–1025. 6. Stella VJ, Rajewski RA. 1997. Cyclodextrins: Their future in drug formulation and delivery. Pharm Res 14:556– 567. 7. Floyd AG. 1999. Top ten considerations in the development of parenteral emulsions. Pharm Sci Tech 4:134–143. 8. Lawrence MJ, Rees GD. 2000. Microemulsion-based media as novel drug delivery systems. Adv Drug Deliv Rev 45:89– 121. 9. Nakano N. 2000. Places of emulsions in drug delivery. Adv Drug Deliv Rev 45:1–4. 10. Akers MJ. 2002. Excipient–drug interactions in parenteral formulations. J Pharm Sci 91:2283–2300. 11. Pifferi G, Restani P. 2003. The safety of pharmaceutical excipients. Farmaco 58:541–550. 12. Jia L, Wong J, Cerna C, Weitman S. 2002. Effect of nanonization on absorption of 301029: Ex vivo and in vivo pharmacokinetic correlations determined by liquid chromatography/mass spectrometry. Pharm Res 19:1091–1096. 13. Merisko-Liversidge E, Liversidge GG, Cooper ER. 2003. Nanosizing: A formulation approach for poorly-water-soluble compounds. Eur J Pharm Sci 18:113–120. 14. Lindfors L, Skantze P, Skantze U, Rasmusson M, Zackrisson A, Olsson U. 2006. Amorphous drug nanosuspensions. 1. Inhibition of Ostwald ripening. Langmuir 22:906–910. 15. Lindfors L, Forssen S, Skantze P, Skantze U, Zackrisson A, Olsson U. 2006. Amorphous drug nanosuspensions. 2. Experimental determination of bulk monomer concentrations. Langmuir 22:911–916. 16. Sigfridsson K, Lundqvist AJ, Strimfors M. 2009. Particle size reduction for improvement of oral absorption of the poorly soluble drug UG558 in rats during early development. Drug Dev Ind Pharm 35:1479–1486. ¨ 17. Muller RH, Peters K. 1998. Nanosuspensions for the formulation of poorly soluble drugs I. Preparation by size-reduction technique. Int J Pharm 160:229–237. 18. Rabinow BE. 2004. Nanosuspensions in drug delivery. Nat Rev 3:785–796. ¨ 19. Mauludin R, Muller RH, Keck CM. 2009. Kinetic solubility and dissolution velocity of rutin nanocrystals. Eur J Pharm Sci 36:502–510. 20. Liversidge GG, Conzentino P. 1995. Drug particle reduction for decreasing gastric irritancy and enhancing absorption of naproxen in rats. Int J Pharm 125:309–313. 21. Liversidge GG, Cundy K. 1995. Particle size reduction for improvement of oral bioavailability of hydrophobic drugs: I. Absolute oral bioavailability of nanocrystalline danazol in beagle dogs. Int J Pharm 125:91–97. 22. Pignatello R, Bucolo C, Ferrara P, Maltese A, Puleo A, Puglisi G. 2002. Eudragit RS100 nanosuspensions for the ophthalmic controlled delivery of ibuprofen. Eur J Pharm Sci 16:53– 61. ¨ 23. Jacobs C, Muller RH. 2002. Production and characterization of a budesonide nanosuspension for pulmonary administration. Pharm Res 19:189–194. JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 100, NO. 6, MAY 2011

2202

SIGFRIDSSON ET AL.

24. Wong J, Brugger A, Khare A, Chaubal M, Papadopoulos P, Rabinow B, Kipp J, Ning J. 2008. Suspensions for intravenous (IV) injection: A review of development, preclinical and clinical aspects. Adv Drug Deliv Rev 60:939–959. 25. Bassand JP, Hamm CW, Ardissino D, Boersma E, Budaj A, Fernandez-Aviles F, Fox KA, Hasdai D, Ohman EM, Wallentin L, Wijns W. 2007. Guidelines for the diagnosis and treatment of non-ST-segment elevation acute coronary syndromes. Eur Heart J 28:1598–1660. 26. Husted S, van Giezen JJ. 2009. Ticagrelor: The first reversibly binding oral P2Y receptor antagonist. Cardiovasc Ther 27:259–274. 27. AstraZeneca, data on file to be published. 28. US National Institute of Health Guide for the Care and Use of Laboratory Animals (NIH publication no. 85–23, revised 1985). 29. van Giezen JJ, Berntsson P, Zachrisson H, Bj¨orkman JA. 2009. Comparison of ticagrelor and thienopyridine P2Y12 binding characteristics and antithrombotic and bleeding effects in rat and dog models of thrombosis/hemostasis. Thromb Res 124:565–571.

JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 100, NO. 6, MAY 2011

30. Folts J. 1991. An in vivo model of experimental arterial stenosis, intimal damage, and periodic thrombosis. Circulation 83(6 suppl):3–14. 31. Huang LF, Tong WQ. 2004. Impact of solid state properties on developability assessment of drug candidates. Adv Drug Deliv Rev 56:321–334. ¨ 32. Muller RH, Jacobs C, Kayser O. 2001. Nanosuspensions as particulate drug formulations in therapy: Rationale for development and what we can expect for the future. Adv Drug Deliv Rev 47:3–19. 33. Clement M, Pugh W, Parikh I. 1992. Tissue distribution and plasma clearance of a novel microcrystalline-coated flurbiprofen formulation. Pharmacologist 34:204. ¨ 34. Sigfridsson K, Forsen S, Hollander P, Skantze U, de Verdier J. 2007. A formulation comparison, using a solution and different nanosuspensions of a poorly soluble compound. Eur J Pharm Biopharm 67:540–547. 35. Bj¨orkman J-A, Kirk I, van Giezen JJ. 2007. AZD6140 inhibits adenosine uptake into erythrocytes and enhances coronary blood flow after local ischaemia or intracoronary adenosine infusion. Circulation 116(suppl):II-28.

DOI 10.1002/jps