Plasma pharmacokinetics and faecal excretion ofivermectin, doramectin and moxidectin following oraladministration in horses

C. GOKBULUT, A. M. NOLAN andQ. A. MCKELLAR*

Division of Veterinary Pharmacology, University of Glasgow, Bearsden Road, Glasgow G61 1QH and*Moredun Research Institute, Pentlands Science Park, Penicuik, Edinburgh EH26 OPZ, UK.

Keywords: horse; pharmcokinetics; ivermectin; doramectin; moxidectin

494 EQUINE VETERINARY JOURNALEquine vet. J. (2001) 33 (5) 494-498

Summary

The present study was carried out to investigate whether thepharmacokinetics of avermectins or a milbemycin couldexplain their known or predicted efficacy in the horse.

The avermectins, ivermectin (IVM) and doramectin(DRM), and the milbemycin, moxidectin (MXD), were eachadministered orally to horses at 200 g/kg bwt. Blood andfaecal samples were collected at predetermined times over80 days (197 days for MXD) and 30 days, respectively, andplasma pharmacokinetics and faecal excretion determined.Maximum plasma concentrations (Cmax) (IVM: 21.4 ng/ml;DRM: 21.3 ng/ml; MXD: 30.1 ng/ml) were obtained at (tmax)7.9 h (IVM), 8 h (DRM) and 7.9 h (MXD). The area undertheconcentration time curve (AUC) of MXD (92.8 ng.day/ml)was significantly larger than that of IVM (46.1 ng.day/ml)but not of DRM (53.3 ng.day/ml) and mean residence time ofMXD (17.5 days) was significantly longer than that of eitheravermectin, while that of DRM (3 days) was significantlylonger than that of IVM (2.3 days). The highest (dry weight)faecal concentrations (IVM: 19.5 g/g; DRM: 20.5 g/g;MXD: 16.6 g/g) were detected at 24 h for all molecules andeach compound was detected ( 0.05 g/g) in faeces between8 h and 8 days following administration.

The avermectins and milbemycin with longer residencetimes may have extended prophylactic activity in horses andmay be more effective against emerging and maturingcyathostomes during therapy. This will be dependent uponthe relative potency of the drugs and should be confirmed inefficacy studies.

Introduction

Avermectins and milbemycins have been used intensively tocontrol parasites of animals, humans and crops worldwide. Bothchemical groups are naturally derived 16-membered macrocycliclactones (Takiguchi et al. 1980) produced by the soil-dwellingactinomycetes, S t reptomyces spp. Milbemycins are structurallyrelated to the avermectins and share physicochemical propertieswith avermectins. The main structural difference between these 2groups is a disaccharide group found at the C-13 position of the

macrolide ring of avermectins. The milbemycins have nodisaccharide group at C-13 (Fig 1). Avermectins and milbemycinshave excellent activity against nematodes and have systemicactivity against several pathogenic ectoparasites of domesticanimals at low dosage rates. Due to the high activity ofavermectins and milbemycins against both nematodes andarthropods, they are now classified as endectocides ( M c K e l l a rand Benchaoui 1996). However, these drugs have no usefulactivity against trematodes or cestodes (Shoop et al. 1 9 9 5 ) .

Ivermectin (IVM) was the first macrocyclic lactoneanthelmintic, introduced as a veterinary antiparasitic agent inFrance in 1981 and is marketed as a mixture of 22,23 dihydro-B1a (>80%) and 22,23 dihydro-B1b (

C. Gokbulut et al. 495

Materials and methods

Twenty-four horses, weighing 490880 kg, were kept at pastureduring the experimental period. No invasive procedures wereinvolved beyond blood and faecal sampling procedures. Horseswere allocated into 3 groups of 8 such that the mean weight ofanimals in each group was similar, and the horses wereidentified by unique freeze brand or natural markings. Thecommercially available equine formulation of IVM (Eqvalanpaste, 1.87% w/v)1 and MXD (Equine gel, 2% w/v)2 and theinjectable cattle formulation of DRM (Dectomax 1% w/v)3were administered orally as a single bolus dose on the back ofthe tongue, each at 200 g/kg bwt.

Heparinised blood samples were collected by jugularvenipuncture prior to drug administration, then at 1, 2, 4, 8, 12,20, 24, 32, 48, 72, 96 and 120 h and 8, 11, 25, 39, 66 and 80 days. Supplementary samples were collected on Day 197following administration of MXD to allow for the longerresidence time of this drug. Faecal samples (>10 g) were alsocollected per re c t u m throughout the blood sampling period,before drug administration and thereafter at 4, 8, 24, 32, 48 and120 h and 8, 11, 25 and 39 days in order to determine thepattern of faecal excretion. Blood samples were centrifuged at1825 g for 30 min and plasma transferred to plastic tubes. A l lplasma and faecal samples were stored at -20C untilestimation of drug concentration.

Analytical procedure

The parent compounds of IVM, DRM and MXD in plasma andfaeces were analysed by high performance liquid chromatography(HPLC) with a liquid phase extraction procedure adapted fromthat described by Scott and McKellar (1992).

Stock solutions (100 g/ml) of pure standard of IVM1,DRM2 and MXD3 were prepared using acetonitrile4 as thesolvent. These were diluted to give 5, 10, 100, 200 and 500 ng/ml and 0.5, 1, 5, 10 and 50 g/ml standard solutions forplasma and faecal samples, respectively, for calibration asstandard curves and to add to drug-free plasma and faecalsamples to determine the recovery.

Extraction

Drug-free plasma samples (1 ml) were spiked with either IVM,DRM or MXD standards to reach the following finalconcentrations: 0.5, 1, 10, 20 and 50 ng/ml. Acetonitrile (1 ml)was added and, after vortexing for 15 s, 5 ml chloroform4 wasalso added. The tubes were shaken on a slow rotary mixer for 15 min. After centrifugation at 1825 g for 15 min, thesupernatant was removed with a pasteur pipette. The organicphase (4 ml) was transferred to a thinwalled 10 ml conical glasstube and evaporated to dryness at 43C in a sample concentrator(model SC210A)5. The dry residue was dissolved in 100 l of N-methylimidazole6 solution in acetonitrile (1:1). To initiate thederivatisation, 150 l trifluoroacetic anhydride6 solution inacetonitrile (1:2) was added. Finally, 50 l of this solution wasinjected into the chromatographic system.

Wet-faecal material was mixed with a spatula to obtain ahomogeneous sample. Drug-free wet-faeces samples (0.5 g)were spiked with either IVM, DRM or MXD standards to reachthe following final concentrations: 0.05, 0.1, 0.5, 1.0 and 5.0 g/g. Water (1 ml) and 4 ml acetonitrile were added to the 10 ml ground glass tubes containing 0.5 g spiked andexperimental wet-faecal samples. After vortexing for 15 s, 6 mlchloroform was added. The tubes were shaken on a slow rotarymixer for 15 min. After centrifugation at 1825 g for 15 min, thesupernatant was removed with a pasteur pipette. The organicphase (5 ml) was transferred to a thinwalled 10 ml conical glasstube and evaporated to dryness at 43C in the sampleconcentrator. The dry residue was dissolved in 100 l of N-methylimidazole solution in acetonitrile (1:1). To initiate thederivatisation, 150 l trifluoroacetic anhydride solution inacetonitrile (1:2) was added. The derivatisated samples werediluted appropriately with acetonitrile and filtered using GF/Cglass microfibre filters7. The filtrate (50 l) was injected into thechromatographic system.

HPLC system

The mobile phase of 100% acetonitrile for IVM,acetonitrile:water (99.5:0.5) for DRM and acetonitrile:methanol(65:35) for MXD was delivered (model LC-10AS)8 at a flowrate of 1.8 ml/min for IVM and DRM and 1.2 ml/min for MXD.A Genesis C18 column9 (4 m, 150 x 4.6 mm) was used for IVMand MXD, and a Nova-Pak C18 column10 (4 m, 150 x 3.9 mm)for DRM. Fluorescence detection (model RF-10A)8 was at anexcitation wavelength of 365 nm and an emission wavelength of 475 nm.

For faecal samples, a solvent delivery system (SpectraPhysics SP4000)11 connected to a Nemesis C18 column12 (4 m,

Fig 1: Chemical structures of a) ivermectin (IVM), b) doramectin (DRM)and c) moxidectin (MXD).

a)

b)

c)

HO

HO

HO

OH

OH

HOC

C

HO

OH

C

13

2225

23

22

1325

23

2225

23

150 x 4.6 mm) and a fluorescence detector (FL3000)13 at anexcitation wavelength of 365 nm and an emission wavelength of475 nm were used. The mobile phase was 100% acetonitrile forIVM, acetonitrile:water (97:3) for DRM and acetonitrile:water(96.5:3.5) for MXD.

Recovery and precision

Recovery of the 3 parent molecules under study was measured bycomparison of the peak areas from spiked plasma samples withthe areas resulting from direct injections of standards inacetonitrile carried through the derivatisation procedure. T h einterassay precision of the extraction and chromatographyprocedures was evaluated by processing replicate aliquots ofdrug-free horse plasma or faecal samples containing knownamounts of the drugs on different days. The limit of quantificationof the assay was 0.25 ng/ml for plasma and 0.05 g/g for faecalsamples, and was determined as 5 times the area of the peak atlimit of detection. Mean recoveries for plasma samples were87.5% (interassay CV = 9.04%) for IVM, 89.8% (interassay C V = 4.79%) for DRM and 95.3% (interassay CV = 8.01%) forMXD. Mean recoveries for faecal samples were 96.0% (interassayC V = 5.12%) for IVM, 97.5% (interassay CV = 9.13%) for DRMand 88.4% (interassay CV = 8.85%) for MXD.

To determine the dry proportion of wet faecal samples, 1.0 gof wet faeces from each sample was weighed exactly into anevaporating bowl and heated in an oven at 70C for 10 h. Theweight of each was determined and the percentage of each drysample calculated.

Pharmacokinetic and statistical analysis of data

The plasma concentration vs. time curves obtained after eachtreatment in individual animals were fitted with theWinNonlin software programme1 4. Pharmacokineticparameters for each animal were analysed usingnoncompartmental model analysis with extravascular input.The maximum plasma concentration (Cm a x) and time to reachmaximum concentration (tm a x) were obtained from the plottedconcentration-time curve of each drug in each animal. T h elinear trapezoidal rule was used to calculate the area under the

plasma concentration time curve (AUC):

where C represents the plasma concentration, i - 1 and i areadjacent data point times. The area under the first movementcurve (AUMC) was calculated using the equation:

Therefore, the mean residence time (MRT) was calculated as:

The pharmacokinetic parameters are reported as medianwith the interquartile range (Q1Q3). The median ofpharmacokinetic parameters for IVM, DRM and MXD obtainedfollowing oral administration to horses were statisticallycompared by the Mann-Whitney U test. The values wereconsidered significantly different at P

C. Gokbulut et al. 497

8 h and 4 days. Mean dry-faecal concentrations are shown inFig 3. The faecal excretion patterns of IVM, DRM and MXDwere similar and no significant difference was observed forCm a x and AUC values of any of the molecules in faeces. T h ehighest faecal excretions (19.5 g/g for IVM, 20.5 g/g forDRM and 16.6 g/g for MXD) were determined at 24 h for all molecules.

Discussion

The pharmacokinetics and activity of avermectins andmilbemycins are particularly influenced by the physicochemicalproperties of the active molecules. It has been reported thatMXD is 100 times more lipophilic than IVM (Hayes 1994) andthat the water solubility of MXD (4.3 mg/l) (Lanusse andPrichard 1993) is much higher than that of IVM (0.0060.009 mg/l) (Fisher and Mrozik 1989). The watersolubility and lipophilicity of MXD are unusual for a drug, sinceincreased water solubility is usually directly associated withdecreased lipophilicity. In cattle, the concentration of MXD infat tissue has been shown to be 90-fold higher than that detectedin plasma 28 days following treatment (Zulalian et al. 1994).Although IVM, DRM and MXD showed similar absorptionpatterns, the plasma decline of MXD was initially faster incomparison to IVM and DRM following oral administration in

horses in the current study (Figs 2a,b). In contrast, the MRTvalue of MXD (17.5 days) was significantly longer than thevalues for DRM (3 days) and IVM (2.3 days). These results may also beassociated with greater proportions of MXD accumulating in fattissue than IVM and DRM. The higher fat tissue reservoir ofMXD may explain the extended persistence of that moleculecompared to that of the avermectins and could confer persistentefficiency against equine parasites, due to its longer retentiontime in plasma and excretion into the gastrointestinal tract.

There is a high correlation (r2 = 0.922) between bodyweightand time until MXD is no longer detectable in sheep (Shoop et al.1997). A similar correlation could not be determined for IVM orMXD in the current study, due to similar bodyweight of animalsin the groups. However, a correlation (r2 = 0.703) was foundbetween bodyweight and time until concentrations fell below thelimit of detection of DRM. The lightest horse (490 kg)demonstrated zero detectable plasma concentration of DRM atDay 8, whereas the heaviest horse (880 kg) reached zerodetectable plasma concentration at Day 39. These results could berelated to the amount of fat tissue in the animals.

The results of the present study differ substantially fromthose previously reported for IVM and MXD in horses (Perez etal. 1999). In the previous study, MXD was administered at 400 g/kg bwt, whereas in this study it was administered at 200 g/kg bwt. Nevertheless, the Cmax of 70.35 ng/ml obtainedby Perez et al. (1999) was more than double and the AUC (363.6 vs. 92.8 ng.day/ml) approximately 4 times that obtainedin the present study. Differences in the 2 studies may not beassociated simply with dosage rates, since IVM was given at thesame dosage rate of 200 g/kg bwt in both studies but producedd i fferent pharmacokinetic parameters. Therefore, the Cm a x43.99 ng/ml, AUC 132.7 ng.day/ml and MRT 4.78 daysobtained by Perez et al. (1999) were all substantially larger thanin the present study (Cmax 21.4 ng/ml, AUC 46.1 ng.day/ml andMRT 2.3 days). These differences may be due, in part, todifferences in methodology, although they appear to be too largeto be wholly attributed to such differences. It is unlikely thatformulation differences could be responsible, since both studiesused Eqvalan1. In the present study, horses were yardedfor theperiod during and immediately (4 h) after drug administrationand were then returned to a grass paddock. Feeding could,therefore, have caused differences in drug absorption, althoughfeeding is not considered to cause such dramatic differences inthe oral absorption of anthelmintics in horses (Q.A. McKellar,

20

15

10

5

00 20 40 60 80 100 120

Time (days)

MXDDRMIVM

Fig 3. Mean ( s.e.) dry-faecal concentration (g/g) of ivermectin(IVM), doramectin (DRM) and moxidectin (MXD) following oraladministration to horses (n = 8) at 200 g/kg bwt.

TABLE 1: Pharmacokinetic parameters of ivermectin (IVM), doramectin (DRM) and moxidectin (MXD) following their oraladministration to horses (n = 8) at a dose rate of 200 g/kg bwt

Pharmacokinetic parameters IVM DRM MXD

Cmax(ng/ml) 21.4 (13.236.3) 21.3 (14.245.3) 30.1 (19.633.2)

Tmax(h) 7.9 (5.011.0) 8.0 (4.117.9) 7.9 (5.07.9)

AUClast (ng.day/ml) 46.1 (25.264.1) 53.3 (43.1114.5) 92.8* (61.9110.8)

AUMClast (ng.day2/ml) 111.4 (51.7178.1) 151.0 (122.0352.0) 1716.2 (585.02077.0)

MRTlast(days) 2.3 (2.12.7) 3.0 (2.73.3) 17.5 (8.722.1)

Values are expressed as median with the interquartile (IQ) ranges (Q1Q3). *MXD significantly different (P

unpublished data) as it does in ruminants (Ali and Hennessy1996) and dogs (McKellar et al. 1993). Parasitism could havehad an effect, since the horses used by Perez et al. (1999) wereknown to be infected with gastrointestinal parasites and suchinfections may have modest effects on absorption ofanthelmintics (McKellar et al. 1991). Unfortunately, theparasitological status of the horses in the present study wasunknown, although cyathostomes were seen in the faeces ofthese animals and clearly they were exposed to parasites. Themost probable factor affecting the pharmacokinetics ofivermectin in the present study was the breed and size of theanimals used. Perez et al. (1999) used Chilean Criollo horsesweighing 390446 kg, whereas a mixed group of Thoroughbredsand hunters weighing between 560 and 690 kg wereadministered IVM in the present study.

The mode of activity of avermectins and milbemycins is notspecific to parasitic nematodes and arthropods and when theseagents reach the environment they may affect nontarg e to rganisms, which play an important role in the decomposition offaeces (McKellar 1997). Ivermectin has been shown to beexcreted in high concentration in the bile of ruminants (Bogan andMcKellar 1988) and eliminated primarily in faeces, with less than2% of the total dose being excreted in urine (Chiu et al. 1990). Ithas been shown that wet-faecal concentrations of IVM as low as0.001 ppm are toxic to some dung-breeding insects (Strong andJames 1993). Other avermectins and milbemycins have similarecotoxicological effects, since they share similar broad-spectrumantiparasitic activity, although the potency of different agents maymake them less of a risk for specific nontarget species (McKellar1997). The present study indicated that highest concentrations ofall the macrocyclic lactones are found in the faeces until 48 h afterdrug administration; by 120 h after administration, concentrationsof the anthelmintics were below the limit of detection (0.05 g/g).This suggests that the period of greatest environmental risk is for2 days following administration of these drugs to horses, althoughit is known that very low concentrations of ivermectin (0.001 ppm) have deleterious effects on some dung-breedingo rganisms (Strong and James 1993).

In conclusion, the results from this study show that thepersistence of MXD in plasma is significantly greater than IVMand DRM and this may have a positive effect on its efficacy. Nosignificant difference was observed for the faecal excretionpatterns of IVM, DRM and MXD following their oraladministration in horses.

Acknowledgements

The technical assistance of Mr I. Gibson is gratefullyacknowledged.

Manufacturersaddresses

1Merck, Rahway, New Jersey, USA.2Pfizer Inc., Groton, New Jersey, USA.3American Cyanamid, Princeton, New Jersey, USA.4Rathburn Chemical Ltd., Paisley, Strathclyde, UK.5Savant Instrument Inc., Holbrook, New York, USA.6Sigma-Aldrich Co. Ltd., Gillingham, Dorset, UK.

7Whatman International Ltd., Maidstone, Kent, UK.8Shimadzu, Kyoto, Japan.9Crawford Scientific, Strathaven, Strathclyde, UK.10Waters, Milford, Massachusetts, USA.11Burke Electronics Ltd., Glasgow, UK.12Phenomenex, Macclesfield, Cheshire, UK. 13Spectra Physics, Hemel Hempstead, Hertfordshire, UK.14Scientific Consulting Inc., Cary, North Carolina, USA.

References

Ali, D.N. and Hennessy, D.R. (1996) The effect of level of feed intake on thepharmacokinetic disposition and efficacy of ivermectin in sheep. J. vet.Pharmacol. Therap. 19, 89-94.

Bogan, J.A. and McKellar, Q.A. (1988) The pharmacodynamics of ivermectin insheep and cattle. J. vet. Pharmacol. Therap. 11, 260-268.

Chiu, S.H., Green, M., Baylis, F.P., Eline, D. and Rosegay, A. (1990) Absorption,tissue distribution and excretion of tritium-labeled ivermectin in cattle, sheepand rat. J. Agric. Food Chem. 38, 2072-2078.

Fisher, M.H. and Mrozik, H. (1989) Chemistry. In: Ivermectin and Abamectin, Ed:W.C. Campbell, Springer-Verlag, New York. pp. 1-24.

Goudie, A.C., Evans, N.A., Gration, K.A., Bishop, B.F., Gibson, S.P., Holdom,K.S., Kaye, S.R., Wicks, S.R., Lewis, D. and Weatherley, A.J. (1993)Doramectin - a potent novel endectocide. Vet. Parasitol. 49, 5-15.

Hayes, P.W. (1994) Understanding the unique persistent anthelmintic activity ofthis second-generation macrocyclic lactones. In: Proceedings of the AustralianVeterinarians in Industry Confere n c e, Australian Veterinary A s s o c i a t i o n ,Melbourne, Australia. pp 1-6.

Lanusse, C.F. and Prichard, P. (1993) Relationship between pharmacologicalproperties and clinical efficacy of ruminant anthelmintics. Vet. Parasitol. 49,123-158.

Lanusse, C., Lifschitz, A., Virkel, G., Alvarez, L., Sanchez, S., Sutra, J.F., Galtier ,P. and Alvinerie, M. (1997) Comparative plasma disposition kinetics ofivermectin, moxidectin and doramectin in cattle. J. vet. Pharmacol. Therap.20, 91-99.

McKellar, Q.A. (1997) Ecotoxicology and residues of anthelmintic compounds.Vet. Parasitol. 72, 413-426.

McKellar, Q.A. and Benchaoui, H.A. (1996) Avermectins and milbemycins. J. vet.Pharmacol. Therap. 19, 331-351.

McKellar, Q.A., Galbraith, E.A. and Baxter, P. (1993) Oral absorption andbioavailability of fenbendazole in the dog and the effect of concurrentingestion of food. J. vet. Pharmacol. Therap. 16, 189-198.

McKellar, Q.A., Jackson, F., Coop, R.L., Jackson, E. and Scott, E. (1991) Effect ofparasitism with Nematodirus battus on the pharmacokinetics of levamisole,ivermectin and netobimin. Vet. Parasitol. 39, 123-136.

Perez, R., Cabezas, I., Garcia, M., Rubilar, L., Sutra, J.F., Galtier, P. and Alvinerie,M. (1999) Comparison of the pharmacokinetics of moxidectin (Equest) andivermectin (Eqvalan) in horses. J. vet. Pharmacol. Therap. 22, 174-180.

Scott, E.W. and McKellar, Q.A. (1992) The distribution and some pharmacokineticparameters of ivermectin in pigs. Vet. Res. Commun. 16, 139-146.

Shoop, W.L., Ostlind, D.A., Rohrer, S.P., Mickle, G., Haines, H.W., Michael, B.F.,Mrozik, H. and Fisher, M.H. (1995) Avermectins and milbemycins againstFasciola hepatica: in vivo drug efficacy and in vitro receptor binding. Int. J.Parasitol. 25, 923-927.

Shoop, W.L., Michael, B.F., Haines, H.W., Murphy, T. P., Faidley, T. P., Hajdu, R. andThompson, D.R. (1997) Moxidectin and ivermectin in lambs: plasma depletionand efficacy against helminths. J. vet. Pharmacol. Therap. 2 0, 10-19.

Strong, L. and James, S. (1993) Some effects of ivermectin on the yellow dung fly,Scatophaga stercoraria. Vet. Parasitol. 48, 181-191.

Takiguchi, Y., Mishna, H., Okuda, M. and Terada, M. (1980) Milbemycines, a newfamily of macrolide antibiotics: fermentation, isolation and physico-chemicalproperties. J. Antibiot. 33, 1120-1127.

Zulalian, J., Stout, S.J., Da Cunha, A.R., Garces, T. and Miller, P. (1994)Absorption, tissue distribution, metabolism and excretion of moxidectin incattle. J. Agric. Food Chem. 42, 381-387.

498 Plasma pharmacokinetics of IVM, DRM and MXD

Received for publication: 17.4.00Accepted: 7.11.00