DMD Fast Forward. Published on February 14, 2011 as DOI: 10.1124/dmd.110.034827 DMD FaTshti sF aortircwle ahrasd n. oPt buebenl icsohpyeeddi toedn a nFde fborrmuaattreyd. 1Th4e, f2in0al1 v1e rasison d moaiy: 1di0ff.e1r 1fr2om4/ tdhims vder.s1io1n0..034827 DMD/2010/34827 Pharmacokinetic interaction of the antiparasitic agents ivermectin and spinosad in dogs Stewart T. Dunn, Laura Hedges, 1Kathleen E. Sampson, 2Yurong Lai, Sean Mahabir, 2Larissa Balogh, and Charles W. Locuson Pfizer Animal Health, Veterinary Medicine Research & Development, Metabolism & Safety, D 333 Portage Street, KZO-300-421NW, Kalamazoo, MI 49001 o w n lo a d e d fro Pfizer Animal Health, Metabolism & Safety (STD, LH, and CWL) m d m d Pfizer Animal Health, Biometrics (SM) .as p e tjo u Pfizer Global Research & Development, Pharmacokinetics, Dynamics, and Metabolism, rn a ls .o Chesterfield, MO (KES, YL, and LB) rg a t A S P E T J o u rn a ls o n J a n u a ry 1 3 , 2 0 2 3 1 Copyright 2011 by the American Society for Pharmacology and Experimental Therapeutics. DMD Fast Forward. Published on February 14, 2011 as DOI: 10.1124/dmd.110.034827 This article has not been copyedited and formatted. The final version may differ from this version. DMD/2010/34827 a) Running title: Interaction of ivermectin and spinosad in dogs b) Corresponding author: Charles W Locuson Pfizer Animal Health, Veterinary Medicine Research and Development, Metabolism & Safety, 333 Portage Street, KZO-300-421NW, Kalamazoo, MI 49001 T: 269-833-2505 D o w F: 269-833-7721 n lo a d e d [email protected] fro m d m d .a s p c) Text pages : 20 etjo u rn a Tables 2 ls .o rg a Figures 2 t A S P E References 42 T J o u rn Words in Abstract 242 als o n J Words in Introduction 684 an u a ry 1 Words in Discussion 1591 3 , 2 0 2 3 d) Non-standard abbreviations: area-under-the-curve, AUC; central nervous system, CNS; blood- brain barrier, BBB; N-(3,4-dimethoxyphenethyl)-4-(6,7-dimethoxy-3,4-dihydroisoquinolin- 2[1H]-yl)-6,7-dimethoxyquinazolin-2-amine, CP100356; maximum plasma drug concentration observed, C ; drug-drug interaction, DDI; gamma aminobutyric acid, GABA; liquid max,obs chromatography tandem mass spectrometry, LS-MS/MS 2 DMD Fast Forward. Published on February 14, 2011 as DOI: 10.1124/dmd.110.034827 This article has not been copyedited and formatted. The final version may differ from this version. DMD/2010/34827 Abstract Neurological side effects consistent with ivermectin toxicity have been observed in dogs when high doses of the common heartworm prevention agent ivermectin are coadministered with spinosad, an oral flea prevention agent. Based on numerous reports implicating the role of the ATP binding cassette drug transporter P-glycoprotein (P-gp) in ivermectin efflux in dogs, an in vivo study was conducted to determine whether ivermectin toxicity results from a D pharmacokinetic interaction with spinosad. Beagle dogs were randomized to three groups o w n lo treated orally, in parallel: Treatment Group 1 (T01) received ivermectin (60 μg/kg), Treatment ad e d fro m Group 2 (T02) received spinosad (30 mg/kg), and Treatment Group 3 (T03) received both d m d .a ivermectin and spinosad. While spinosad pharmacokinetics were unchanged in the presence of s p e tjo u ivermectin, ivermectin plasma pharmacokinetics revealed a statistically significant increase in rn a ls .o AUC (3.6-fold over the control) when coadministered with spinosad. The majority of the rg a t A S interaction is proposed to result from inhibition of intestinal and/or hepatic P-gp-mediated P E T J o u secretory pathways of ivermectin. Furthermore, in vitro transwell experiments with a human rn a ≤ ls o MDR1-transfected MDCKII cell line showed polarized efflux at concentrations 2 µM, n J a n u a indicating that spinosad is a high affinity substrate of P-gp. In addition, spinosad was a strong ry 1 3 , 2 inhibitor of the P-gp transport of digoxin, calcein AM (IC50 = 3.2 µM), and ivermectin (IC50 = 023 2.3 µM). The findings suggest spinosad, acting as a P-gp inhibitor, increases the risk of ivermectin neurotoxicity by inhibiting secretion of ivermectin to increase systemic drug levels, and by inhibiting P-gp at the blood-brain barrier. 3 DMD Fast Forward. Published on February 14, 2011 as DOI: 10.1124/dmd.110.034827 This article has not been copyedited and formatted. The final version may differ from this version. DMD/2010/34827 Introduction Ivermectin is one of the most effective and widely used antiparasitic agents ever discovered because of its broad spectrum activity against numerous endo- and ectoparasites, especially nematodes and arthropods (Geary, 2005; Omura, 2008). Chemically, ivermectin is a high molecular weight natural product macrocyclic lactone produced by the actinomycete, Streptomyces avermitilis (Campbell et al., 1983). Originally developed for veterinary use, ivermectin is commonly used to eliminate gastrointestinal nematodes in livestock and D o w heartworms in companion animals. In several instances ivermectin has also proven to be an n lo a d e d effective treatment for worm infections, as well as mites, lice, and scabies in human medicine fro m d (Omura, 2008). A number of companies currently market ivermectin and have adapted a range m d .a s p of formulations for dosing by the oral, parenteral, and topical routes. etjo u rn a ls .o rg a Much of ivermectin’s success could be attributed to its high therapeutic index. For instance, in t A S P E dogs, the effective heartworm prevention dose is 6 µg/kg once-monthly, but the more difficult to T J o u rn treat dermatological Demodex infections are often treated daily with extra-label ivermectin doses als o n J in excess of 50-fold the heartworm prevention dose (Mueller, 2004). The primary factors an u a ry 1 contributing to the therapeutic index of ivermectin appear to be: (1) the high affinity of 3 , 2 0 2 3 ivermectin for its primary pharmacological targets in parasites, the glutamate-gated chloride ion channels (Yates et al., 2003; Wolstenholme and Rogers, 2005); (2) the absence of the glutamate- gated chloride (anion) channels in mammalian hosts (Raymond and Sattelle, 2002); and (3) “protection” from ivermectin binding to its secondary target, γ-aminobutyric acid (GABA)-gated chloride channels, because the expression of these GABA channels in mammals is mostly limited to the central nervous system (CNS) (Campbell et al., 1983). When the limits of the 4 DMD Fast Forward. Published on February 14, 2011 as DOI: 10.1124/dmd.110.034827 This article has not been copyedited and formatted. The final version may differ from this version. DMD/2010/34827 safety margin are exceeded in the treatment of difficult to treat infections, the toxicological syndromes appear to be consistent with GABA receptor modulation caused by ivermectin- stimulated neuronal GABA release (Lovell, 1990). Despite the well-deserved superdrug status held by ivermectin, drug interaction concerns have recently arisen in dogs for use of extra-label doses in conjunction with the oral flea preventative, spinosad. According to a warning statement issued by the Food and Drug Administration, D o w increased incidence of ivermectin toxicoses have been observed during co-administration of the n lo a d e d two antiparasitic agents in dogs fro m d (http://www.fda.gov/AnimalVeterinary/NewsEvents/CVMUpdates/ucm047942.htm). The m d .a s p authors are not aware of any study detailing the rate of occurrence of adverse events with extra- etjo u rn a label ivermectin doses. When administered with ivermectin at its heartworm prevention dose, ls .o rg a the safety of spinosad has never been questioned based on field experience. Recently, the safety t A S P E of extra-label doses of the anthelmintic milbemycin oxime has also been confirmed with T J o u rn concurrent spinosad treatment (Sherman et al., 2010). als o n J an u a ry 1 Interestingly, a clue to the potential mechanism behind the ivermectin-spinosad drug interaction 3 , 2 0 2 3 in dogs comes from a well-characterized P-glycoprotein (P-gp) mutation in dogs found predominantly in the collie breed, but also other breeds (Mealey et al., 2001; Mealey and Meurs, 2008). Ivermectin is a substrate and inhibitor of the transmembrane drug transporter P-gp (Didier and Loor, 1996). A four base pair deletion in the ABCB1 gene encoding P-gp results in a truncated, non-functional protein (Mealey et al., 2001). This P-gp variant is associated with increased sensitivity to ivermectin due to the enhancement of BBB drug penetration (Pulliam et 5 DMD Fast Forward. Published on February 14, 2011 as DOI: 10.1124/dmd.110.034827 This article has not been copyedited and formatted. The final version may differ from this version. DMD/2010/34827 al., 1985; Mealey et al., 2001). Clearly, clinician knowledge of the mutation is relevant during ivermectin treatment, which is associated with neurotoxicity at high doses (≥100 µg/kg) typically tolerated by wild-type allele carriers (Hopkins et al., 1990; Mealey, 2004; Fecht and Distl, 2008). Because the safe use of high dose ivermectin therapies is related to P-gp function, it has been conjectured that spinosad may inhibit ivermectin efflux by P-gp in dogs with wild-type P-gp alleles. In such an instance, ivermectin would be predicted to have greater central nervous system (CNS) penetration and GABA-related pharmacology. Since ivermectin is a P-gp D o w substrate that is primarily eliminated in feces/bile and not extensively metabolized, it is also nlo a d e d reasonable to envision how P-gp inhibition could increase circulating ivermectin via decreased fro m d hepatic or intestinal secretion. The goal of this work was to test whether the ivermectin-spinosad m d .a s p e interaction in dogs is related to P-gp inhibition via pharmacokinetic studies and in vitro P-gp tjo u rn a efflux and inhibition experiments. ls .o rg a t A S P E T J o u rn a ls o n J a n u a ry 1 3 , 2 0 2 3 6 DMD Fast Forward. Published on February 14, 2011 as DOI: 10.1124/dmd.110.034827 This article has not been copyedited and formatted. The final version may differ from this version. DMD/2010/34827 Methods Chemicals and Reagents. Dosing material was commercially available ivermectin solution (Ivomec® (1%) in sterile solution 40% glycerol formal and propylene glycol, q.s. ad 100%. ™ Spinosad tablets (Comfortis ) were obtained from commercial retail resources for research use. Ivermectin was purchased from Sigma-Aldrich (St. Louis, MO). An internal standard for ivermectin, 23-hydroxy-doramectin, was prepared by Pfizer Inc. (Groton, USA). Spinosyn A was purchased from Chembiotek Research International Pvt. Ltd (Worcestershire, UK), and an D o w n internal standard consisting of spinosad containing a substituted amine was prepared by Pfizer lo a d e d Inc (Sandwich, UK). Methanol and acetonitrile were obtained from Honeywell Burdick and fro m d m Jackson (Muskegon, MI). N-methylimidazole, trifluoroacetic anhydride, and human albumin d .a s p e were obtained from Sigma-Aldrich (St Louis, MO). Acetic acid was obtained from Mallinckrodt tjo u rn a Laboratory Chemicals (Phillipsburg, NJ). Formic acid was obtained from EMD Chemicals Inc. ls.o rg a (Darmstadt, Germany). The solid phase extraction cartridges (C18 Isolute, 2 ml capacity with t A S P E T 100 mg separating medium) were obtained from Biotage (Charlottesville, VA). Minimum J o u α α rn a Essential Medium- (MEM ), fetal bovine serum (FBS), 10 mM MEM non-essential amino ls o n J a acids, 10,000 units/ml penicillin-10,000 μg/ml streptomycin, 0.25% Trypsin-EDTA, Dulbecco’s nu a ry 1 3 Phosphate Buffered Saline (D-PBS) and Hank’s Balanced Salt Solution (HBSS), pH 7.4, and , 2 0 2 3 customized HBSS containing 25 mM glucose and 20 mM Hepes, pH 7.4 (transport buffer), were ™ purchased from Invitrogen Corp. (Carlsbad, CA). Vybrant Multidrug Resistance Assay Kit and calcein-acetoxymethyl ester (calcein AM) in anhydrous DMSO were purchased from Molecular Probes (Eugene, OR). The calcein AM was aliquoted into single use vials and stored at –20oC under nitrogen gas in a dessicator. 7 DMD Fast Forward. Published on February 14, 2011 as DOI: 10.1124/dmd.110.034827 This article has not been copyedited and formatted. The final version may differ from this version. DMD/2010/34827 Study animals and treatment group randomization. All procedures in this study were conducted in compliance with Animal Welfare Act Regulations. The use of research animals was performed under the guide of an Institutional Animal Care and Use Committee approved Animal Use Protocol. Study subjects were purpose bred beagle dogs ranging in age from 3-7 years. Animals were allocated to treatments and runs by weight according to a randomized block design with one-way treatment structure. Blocking was based on run location within the study room. The experimental unit for treatment was the animal. Nine dogs were assigned to one of three D o w groups for oral dosing with either ivermectin (T01), spinosad (T02), or concurrent doses of n lo a d e d spinosad and ivermectin (T03). fro m d m d .a s p Dosing and blood draws. Ivermectin (69 µg/kg, average) and spinosad (>30 mg/kg) were etjo u rn a administered orally via gavage and by chewable tablet, respectively. The dose of ivermectin was ls .o rg a 10-fold higher than used for routine heartworm prevention to enable pharmacokinetic analysis. t A S P E Spinosad was administered at the recommended dose using the dose banding guidelines T J o u rn provided. Each dog was fed 30 min prior to dosing to stimulate digestive processes. After als o n J dosing, all subjects were administered 15 mL of water by syringe to ensure delivery of dose. An an u a ry 1 intensive blood sampling regimen followed the dose with samples taken at pre-dose, and 1, 2, 3, 3 , 2 0 2 3 4, 6, 10, 24, 48, and 72 h, and 5, 7, 10, and 14 d. Two mL blood samples were collected via jugular venipuncture with vacutainer tubes containing K -EDTA anticoagulant. Plasma was 2 prepared by centrifuging and stored at -20 °C until analysis. Clinical signs were recorded at 2, 4, and 10 h, and once daily for the remainder of the study. No adverse events were observed in any study subjects. 8 DMD Fast Forward. Published on February 14, 2011 as DOI: 10.1124/dmd.110.034827 This article has not been copyedited and formatted. The final version may differ from this version. DMD/2010/34827 Plasma analysis. Ivermectin in plasma was assayed by liquid chromatography with fluorescence detection after solid phase extraction and chemical derivatization. Standards and quality control samples (prepared from separate weighings) were prepared by the addition of appropriate volumes of standard solutions of ivermectin to aliquots of control dog plasma. Internal standard solution (75:25 water:acetonitrile) was added to each standard and sample, mixed, and then centrifuged. An Isolute extraction plate was primed with 2 mL methanol per well and allowed to drain under gravity. Wells were then washed with 2 mL water. Plasma standards and samples D o w were then applied to the extraction plate, which was then washed with 2 mL water, followed by 1 n lo a d e d mL 75:25 water:methanol, and dried under vacuum. The extraction plate was then eluted with 2 fro m d mL methanol. The eluate was collected in a 96-well plate (2 mL capacity), and evaporated to m d .a s p dryness under a stream of nitrogen gas at 60 oC using a sample concentrator (TurboVap 96, etjo u μ rn a Caliper Life Sciences, Hopkinton, MA). N-methyl imidazole (60 L of 50 % v/v in acetonitrile) ls .o μ rg a was added to each well in the plate, followed by trifluoroacetic anhydride (100 L of 50 % v/v in t A S P E acetonitrile) to derivatize the samples. Detection was accomplished using a Waters Acquity T J o u rn ultra-performance liquid chromatography system and fluorescence detector (Waters Corp., als o n J Milford, MA) with an excitation wavelength of 365 nm, and an emission wavelength of 475 nm. an u a ry 1 Chromatographic separations were performed using an Acquity BEH C18 column, 1.7 um, 2.1 x 3 , 2 0 2 3 100 mm, and gradient elution. The flow rate was 500 uL per min, with a column temperature of μ 45 ºC and an injection volume of 10 L. Solvent A was water and solvent B was 0.2 % acetic acid in water, methanol, and acetonitrile (4:32:64 % v/v). Gradient conditions began at 50 % B, increasing to 99 % B over 3.5 min, holding at 99% B for 3.5 min, then returning to initial conditions over 2.5 min. Ivermectin eluted at 6.08 min, and internal standard at 5.45 min. 9 DMD Fast Forward. Published on February 14, 2011 as DOI: 10.1124/dmd.110.034827 This article has not been copyedited and formatted. The final version may differ from this version. DMD/2010/34827 Standard curves were shown to be linear in dog plasma over the concentration range covering the unknown samples and no weighting was applied to the fit (r = 0.9967). Spinosyn A was assayed by liquid chromatography separation and tandem mass spectrometry (LC-MS/MS) methods. Standards, quality controls, and samples were prepared using the Hamilton MicroLab Star (Reno, NV). Appropriate volumes of standard solutions of spinosyn A were added to aliquots of control dog plasma in a 96-well format. Similarly, appropriate D volumes of control solvent were added to sample plasma, to maintain a common matrix. Internal o w n lo standard was then added to each standard and sample. The plate was thoroughly vortex mixed ad e d fro and centrifuged and aliquots of supernatant transferred to an injection plate. Mass spectrometry m d m d detection of test compounds was accomplished with an Applied Biosystems MDS/Sciex API .a s p e tjo 4000 triple quadrupole mass spectrometer (Foster City, CA). Chromatographic separations were urn a ls .o performed using a Zorbax Eclipse XDB C8 column, 3.5 um, 2.1 x 50 mm, and gradient elution. rg a μ t A S The flow rate was 600 µL per min, with an injection volume of 10 L. Solvent A was 0.1 % P E T J o formic acid and solvent B was methanol. Gradient conditions began at 10 % B, increasing to 90 u rn a ls o % B over 1 min, holding at 90% B for 1.4 min, then returning to initial conditions over 0.5 min. n J a n u a Spinosad eluted at 1.89 min, and internal standard at 2.06 min. Analysis was performed in ry 1 → 3, 2 positive ion mode at unit resolution for the spinosad forosamine transition of 732.5 142.3. 0 2 3 Calibration curves were shown to be quadratic up to 1000 ng/mL in dog plasma. 1/X weighting was used to provide the fit (r = 0.9982). Pharmacokinetic analysis. Pharmacokinetic calculations were performed using the noncompartmental approach (linear trapezoidal rule for AUC calculation) with the aid of Watson (v7.2, Innaphase, Inc; Wayne, PA). The ivermectin and spinosad plasma concentrations used to 10