ELSEVIER
`
`Veterinary Parasitology 86 (1999) 203-215 T
`wwwelsevier. com/1ocate/vetpar
`
`veterinary
`parasitology
`
`Ivermectin disposition kinetics after subcutaneous and
`
`intramuscular administration of an oil-based formulation
`
`to cattle
`
`A. Lifschitza, G. Virkel 3, A. Pis a, F. Imperialea, S. Sanchez 3, L. Alvareza,
`R. Kujanek b, C. Lanusse a”‘
`3 Laboratorio de Farmacologia, Departamento de Fisiopatoiogia, Facultad de Ciencias Veterinarias,
`UniVersidadNaciona1 del Centm, Campus Universitario, (7000) Tandil, Argentina
`1’ Bayer Argentina SA, Buenos Aires, Argentina
`Received 8 March 1999; accepted 7 June 1999
`
`Abstract
`
`Slight differences in formulation may change the plasma kinetics and ecto—endoparasiticide
`activity of endectocide compounds. This work reports on the disposition kinetics and plasma avail-
`ability of ivermectin (IVM) after subcutaneous (SC) and intramuscular (IM) administration as an
`oil-based formulation to cattle. Parasite-free Aberdeen Angus calves (n=24', 240-280 kg) were
`divided into three groups (n= 8) and treated (200 ug/kg) with either an IVM oil-based pharmaceu-
`tical preparation (IVM-TEST formulation) (Bayer Argentina S.A.) given by subcutaneous (Group
`A) and intramuscular (Group B) injections or the IVM-CONTROL (non-aqueous formulation)
`(lvomec®, MSD Agvet) subcutaneously administered (Group C). Blood samples were taken over
`35 days post-treatment a1 d the recovered plasma was extracted a11d analyzed by HPLC using fluo-
`rescence detection. IVM was detected in plasma between 12 h and 35 days po st-administration of
`IVM-TEST (SC and IM injections) and IVM-CONTROL formulations. Prolonged IVM absorption
`half—life (p < 0.05) and delayed peak plasma concentration (p < 0.001) were obtained following
`the SC administration of the IVM-TEST compared to the IVM-CONTROL formulation. No differ-
`ences i11 total plasma availability were observed among treatments. However, the plasma residence
`time and elimination half-life of IVM were significantly longer after injection of the IVM-TEST
`formulation. lVM plasma concentrations were above 0.5 ng/ml for 20.6 (CONTROL) and 27.5 days
`(lVM—TEST SC), respectively (p < 0.05). The modified kinetic behaviour ofIVM obtained after the
`administration of the novel oil-based formulation examined in this trial, compared to the standard
`
`* Corresponding author. Fax: +54—2293—426667
`E—Inai1 address.‘ clanussc@Vct.unicen.cdu.ar (C. Lanussc)
`
`0304-4017/99/$ - see front matter @1999 Hlsevier Science I-3.V. All rights reserved.
`PII: S0304-4017(99)00142-9
`
`AstraZeneca Ex. 2120 p. 1
`Mylan Pharrns. Inc. v. AstraZeneca AB IPR20l6-01325
`
`

`
`204
`
`A. Lifschirz et a1. / Vete1“1'nary Parasitology 86 {I 999) 203—2I5
`
`preparation, may positively impact on its strategic use in cattle. ©1999 Elsevier Science B.V. All
`rights reserved.
`
`Keywords: Iverrnectin; Oil-based formulation; Pharmacokinetics; Cattle
`
`1. Introduction
`
`The averrnectin family includes a series of natural and sernisynthetic molecules, such as
`abamectin, iver1r1ectir1 (IVM), dora1r1ectir1 and eprir1or11ectir1, which share sor11e structural
`and physicochernical properties. The excellent spectrum of activity of avermectins and
`rnilbemycins against several nematode and arthropod species resulted in the all-embracing
`name ‘endectocide’, with which they are now classified (McKellar and Benchaoui, 1996).
`They exhibit endectocide activity at extremely low dosage rates based on a common mode
`of action. IVM is commercially available as injectable and pour-on formulations for use in
`cattle. IVM is highly effective against adults as well as developing and hypobiotic larvae of
`most gastrointestinal nematodes, lungworms (Egerton et al., 1981) and many arthropods in
`cattle (Campbell et al., 1983).
`The avermectins are closely related 16-membered macrocyclic lactones, with a disaccha-
`ride substituent at C13 (Fisher and Mrozik, I989). IVM, a semisynthetic derivative of the
`avermectin family, contains a minimum of 80% 22—23 dihydroaverrnectin B1,, and a maxi-
`r11ur11of 20% 22-23 dihy droavermectin B11, IVM is a large and highly lipophilic molecule
`that dissolves in most organic solvents; despite possessing two sugar rings and two hydroxyl
`groups, it is relatively insoluble in water (Jackson, 1989).
`The pharmacokinetic behaviour of IVM has been studied in different species (Prichard et
`al., 1985; Fink and Porras, 1989; Bogan and McKellar, 1988; Alvinerie et al., 1993; Toutain
`et al., 1997). The pharmacokinetic behaviour of the drug differs according to the route of
`administration, formulation and animal species (Fink and Porras, 1989). The comparative
`plasma disposition kinetics of IVM, moxidectin and doramectin subcutaneously injected
`into cattle, have been characterized recently (Lanusse et al., 1997). The high lipophilicity
`of these molecules accounts for a wide tissue distribution and long residence in plasma,
`which was clearly reflected in the pharmaeokinetic results obtained in those studies.
`The antiparasitic spectrum and efficacy pattern of the different endectocide molecules are
`sirnilarj. however, differences ir1 physico-chemical properties among them may account for
`differences in formulation flexibility, kinetic behaviour, and in the potency and persistence
`of their antiparasitic activity. It has been demonstrated that plasma availability of IVM (Lo
`et al., 1985) and doramectin (Wicks et al., 1993) in cattle is profoundly affected by the
`solvent vehicle in which the drug is formulated.
`Since the antiparasite activity of endectocide molecules depends on drug concentrations
`and time of parasite exposure to them, an evaluation of the comparative pharrnacokinetic
`profiles may help to estimate and optimize drug effieacy. Small dilferenees in fomiulation
`can alter disposition kinetics, and may result in important changes in ecto—endoparasiticide
`activity in livestock. The goal of the study reported here was to evaluate the disposition
`kinetics and plasma availability of IVM following subcutaneous and intramuscular admin-
`istration as a novel oil-based formulation to cattle.
`
`AstraZeneca Ex. 2120 p. 2
`
`

`
`A. L1'fsc1u'IZ et a1. / Veterinary Parasitology 86 (1999) 203—215
`
`205
`
`2. Materials and methods
`
`2.1. Experimental design
`
`2.1.1. Animals
`
`The trial was conducted in 24 parasite-free Aberdeen Angus male calves, weighing
`240-280 kg. All the animals were purchased from the same cattle ranch (area of Tandil,
`Province of Buenos Aires, Argentina). The health of the animals was monitored prior to
`and throughout the experimental period. Animals were ir1 optimal nutritional condition and
`grazing on a luceme/red clover pasture during the entire experimental period. They had free
`access to water.
`
`2.1.2. Treatments
`
`Calves were randomly allocated into three groups of eight animals each. Treatments were
`given as follows:
`
`2.1.3. Croup A
`Experimental animals were treated with an oil-based pharmaceutical preparation of iver-
`mectin (IVM-TEST formulation) (10 mg/rnl) (Baymec Prolong®, Batch 005) provided by
`Bayer Argentina SA. The treatment was givenby subcutaneous (SC) injection in the sho11l-
`der area at 200 ptg/kg body weight. This formulation contains IVM formulated in an oily
`vehicle and was considered as the test product iii the pliarmacokinetic trial.
`
`2.1.4. Group B
`Exerimental animals were treated with the IVM-TEST formulation (10 mg/n1l) (Baymec
`Prolong ®, Batch 005) provided by Bayer Argentina S.A by intramuscular (IM) injection
`in the isquio-tibial region at 200 ptg/kg body weight.
`
`2.1.5. Group C
`Eperirnental animals were treated with a commercially available formulation of iver-
`mectin (Ivomec®, MSD Agvet, NJ, Batch PRl08) by subcutaneous injection in the shoul-
`der area at 200 pug/kg body weight. This non-aqueous preparation contains IVM (10 mg/n11)
`formulated in a propylene glycol/glycerol formal (60 : 40) vehicle and was considered as
`the reference formulation (IVM-CONTROL) in the trial.
`
`2.1.6. Sampling
`Blood samples were taken into heparimzed vacutainer tubes prior to, and at, 0.5, 1, 2, 3,
`4, 5, 7, 9, ll, 15, 20, 25, 30 and 35 days post-treatrnent. Blood samples were centrifuged
`at 2000 >< gfor 20 min and the recovered plasma was kept in labelled vials at —20°C until
`analyzed within 2 to 3 weeks of collection.
`
`AstraZeneca Ex. 2120 p. 3
`
`

`
`206
`
`A. L1'fsc111'IZ et a1. / Vete1“inary Parasitology 86 {I 999) 203-215
`
`2.2. Analytical procedures
`
`2. 2.1. Chemical extraction and derivatization
`
`The extraction of I\/1\/I from spiked and experimental plasma samples was carried out
`following adaptations ofthe technique describedby De Montigny et al. (1990) and Alvinerie
`et al. (1993); Alvinerie et al. (1995). Briefly, a 1-ml aliquot of plasma sample was combined
`witl1 100 ptl of i11ternal standard (abamectin, 100 ng/n1l) and then mixed witl1 1-n11 aceto11i-
`trile. After mixing for 20 min, the solvent-sample mixture was centrifuged at 2000 >< gfor
`15 min. The supernatant was injected into a Supelclean LC13 cartridge (Supelco, Bellfonte,
`PA) previously conditioned with 2 ml methanol and 2 ml deionized water. The cartridge was
`flushed with 2 ml of water/methanol (3 : 1). The analytes were eluted with 1 ml of methanol
`and concentrated to dryness under a stream of nitrogen. The reconstitution was done using
`100 pd of a solution of N-methylirnidazole (Sigma, St. Louis, MO) in acetomtrile (1 : 1).
`Derivatization was imtiated by adding 150 pl trifluoroacetic anhydride (Sigma, St. Louis,
`MO) solution in acetonitrile (1 :2). After completion of the reaction (<30 s), an aliquot
`(100 pd) of this solution was injected directly into the chromatograph.
`
`2.2.2. Drug analysis
`IVM plasma concentrations were determined by high performance liq11id chromatogra-
`phy (HPLC) using a Shimadzu 10 A HPLC system (Shimadzu, Kyoto, Japan). HPLC anal-
`ysis was undertaken using a reverse phase C13 column (Phenomenex, 5 um, 4.6 x 250 mm)
`kept in a column oven at 3 0” C (Shimadzu, Kyoto, Japan) and an acetonitrile/methanol/water
`(55/40/5) mobile phase at a flow rate of 1.5 n1l/mm. IVM was detected with a fluorescence
`detector (Spectrofluorometric detector RF-10, Shimadzu, Kyoto, Japan), reading at an ex-
`citation wavelength of 365 nm and an emission wavelength of 475 run. IVM concentrations
`were determined by the ir1ter11al standard method using the Class LC 10 Software version 1.2
`(Shimadzu, Kyoto, Japan) on an IBM compatible AT 486 computer. The IVM/abamectin
`peak area ratio was used to estimate the IVM concentration in spiked (validation of the
`analytical method) and experimental samples. There was no interference of endogenous
`compounds in the chromatographic determinations. The solvents (Baker, Phillipsburg, NJ)
`used during the extraction ar1d drug analysis were HPLC grade.
`
`2.2.3. Validatiozipivcedures
`A complete validation of the analytical procedures for extraction and quantification of
`IVM was performed before starting the analysis of experimental samples from the phar-
`rnacokinetic trial. IVM (Batch # 95051 ZBEROI, purity 97.5%) and abamectin (Batch
`# 9505 25BER01, purity 97.4%) reference standards provided by Bayer Argentina S.A.,
`were used to prepare calibration curves in a range between 0.25—10ng/ml and 10-100
`ng/ml, respectively. The analytical method was validated according to the following
`criteria:
`
`(a) Linearity. IVM and the internal standard were identified by comparison with the re-
`tention times of pure reference standards. Linearity was established to determine the
`concentration-detector response relationship, as determined by injection of spiked IVM
`standards in plasma at different concentrations (triplicate determinations). Calibration
`
`AstraZeneca Ex. 2120 p. 4
`
`

`
`A. L1'fsc111'IZ et a1. / Veterinary Parasitology 86 {I 999) 203-215
`
`207
`
`curves were established using least-squares linear regression analysis and correlation
`cocfficicnts (r) and cocfficicnt of variation (CV) calculated.
`(b) Recovery. Drug recovery was estimated from calibration lines prepared with different
`IVM—fortified plasma samples using abamectin as internal standard. Percentages of IVM
`recovery from plasma samples were obtained in the range between 0.5 and 50 ng/ml. The
`mean percentage of recovery and the coefficient of variation (CV) were calculated. The
`CV was obtained as the ratio between the standard deviations and the mean recovery
`values of at least three determinations.
`
`(c) Precision: Inter—assay precision of the extraction and chromatography procedures was
`evaluated by processing replicate aliquots of pooled cattle plasma samples co11tai11i11g
`known amounts of lVM (2 and 20 ng/ml) on different days.
`((1) Detection and quantification Iimits: The limit of drug detection was established With
`injection and HPLC analysis of plasma blanks fortified with the internal standard, mea-
`suring the baseline noise at the time of retention of the IVM peak. The mean baseline
`noise at the IVM retentiontime plus three standard deviations was defined as the detection
`lin1it. The meanbaseline noise plus six standard deviations was defined as the theoretical
`quantification limit.
`
`2.3. Plzarmacokinetic and statistical analyses of the data
`
`The plasma concentrations vs. time curves obtained after each treatment in each indi-
`vidual animal were fitted with PK Solutions 2.0 (Ashland, OH) computer software. Phar-
`macokinetic parameters were determined using a model-independent method. The peak
`concentration (Cmax) and time to peak concentration (Tmax) were read from the plotted
`concentration—time curve for each individual animal. The terminal (elimination) half-life
`(71,261) and absorption half-life (T1/zab) were calculated as In 2/15 and ln 2/Icab, respectively,
`where I} is the terminal slope (h‘1) and [fa], the rapid slope obtained by feathering which rep-
`resents the first-orderabsorptionrate constant (h‘ 1). The areas underthe concentration—time
`curves (AUC) were calculated by the trapezoidal rule (Gibaldi and Perrier, 1982) and further
`extrapolated to infinity by dividing the last experimental concentration by the terminal slope
`(13). Statistical moment theory was applied to calculate the mean residence time (MRT) for
`IVM as follows
`
`T _ AUMC
`_ AUC
`
`where AUC is as defined previously and AUMC the area under the curve of the product of
`time and drug concentration vs. time from zero to infinity (Gibaldi and Perrier, 1982).
`IVM plasma concentrations are presented as mean :: SD. The pham1acokinetic parame-
`ters are reported as mean :: SD. Mean pharrnacokinetic parameters for IVM obtained after
`the administration of the different formulations were statistically compared by ANOVA.
`Where Fvalues were significantly different, the Tukey—Kramer multiple comparisons test
`was applied to indicate order of sigmficance. A Value ofp < 0.05 was considered significant.
`
`AstraZeneca Ex. 2120 p. 5
`
`

`
`208
`
`A. L1'fsc1r1'IZ er a1. / Veterinary Parasitology 86 {I 999) 203—2I5
`
`II-I IVM-TEST SC
`m IVM-TEST lM
`:I IVM-CONTROL
`
`V
`
`AUC 0-Tmax
`(ng.d/ml)
`
`T 1/2 ab
`(days)
`
`Fig. 1. Comparative partial area ur1der the ivermectin (IVM) concentration VS. time curves obtained between treat-
`ment and the time oftl1e peak plasma concentration (partial AUC0_ Tm” ), and absorption half-lives (T1/gab) obtained
`after the administration ofthe IVM-TEST (subcutaneous and intramuscular) ar1d IVM-CONTROL (subcutaneous)
`fonnulations to cattle (200 p.g/kg). Values are significantly difierent from those obtained for the IVM-CONTROL
`forrnulation at (*) p < 0.05.
`
`3. Results
`
`The analytical procedures, including chemical extraction, derivatization and HPLC anal-
`ysis of IVM were Validated. The linear regression lines for IVM in the range between
`0.25—l0 ng/rr1l and 10-100 ng/rr1l showed correlation coefficients of 0.999 and 0992, re-
`spcctivcly. The mean recovery of IVM from plasma was 89.5%. The detection limit of the
`analytical technique was 0.03 ng/ml;the theoretical quantification limit was 0.05 ng/ml. The
`inter—assay precision of the analytical procedure obtained after HPLC analysis of spiked
`standards of IVM (2 and l0ng/n11) on different days showed a CV of 1.84%.
`IVM was detected in plasma between 12 h and 35 days post-adrninistration of the IVM-
`TEST (SC and IM injections) and IVM-CONTROL formulations. The partial AUC Val-
`ues measured between treatment and the time of the peak concentration (AUC0,Tnm) and
`the absorption half—liVes obtained for the three treatments are shown in Fig. l. The mean
`plasma concentrations of IVM obtained after the SC administration of the IVM-TEST and
`IVM-CONTROL formulations are shown in Fig. 2. IVM concentration profiles, obtained af-
`ter the IM administration of the IVM-TEST fonnulatiorr are compared with those obtained
`after the SC administration of the control formulation in Fig. 3. Mean IVM plasma con-
`ccntrations obtained at 20, 25, 30 and 35 days post-adrninistration of the three treatments
`are shown in Fig. 4. The mean parameters that summarize the overall kinetic behaviour
`of the drug following the three treatments are compared in Table 1. The number of days
`
`AstraZeneca Ex. 2120 p. 6
`
`

`
`A. L1'fsc111'IZ er a1. / Vete1“1'nary Parasitology 86 {I 999) 203—2I5
`
`209
`
`0.1
`
`IVMconcentration(ng/ml)
`
`-A
`
`0
`
`5
`
`1 0
`
`20
`1 5
`Days post-treatment
`
`25
`
`30
`
`35
`
`Fig. 2. Mean (::SD) plasma concentrations of iveimectin (IVM) obtained after the subcutaneous administration
`of the IVM-TEST and IVM-CONTROL formulations to cattle (200 |.Lg/kg).
`
`10
`
`e
`
`D-‘
`
`
`
`IVMconcentration(ng/ml)
`
`0.1
`
`-0- IVM-TEST IM
`
`\ -V—lVM-CONTROL
`\E\;
`
`y
`
`0
`
`5
`
`10
`
`20
`15
`Days post-treatment
`
`25
`
`30
`
`35
`
`Fig. 3. Mean (:2 SD) plasma concentrations ofivermectin (IVM) obtained after the administration ofthe IVM-TEST
`(intramuscular) and IVM-CONTROL (subcutaneous) formulations to cattle (200 ug/kg).
`
`post-treatment in which plasma concentrations of IVM were greater than 0.5 and lng/ml
`after the administration of the IVM-TEST (SC and IM treatments) and IVM-CONTROL
`(SC) formulations are shown in Table 2.
`
`AstraZeneea Ex. 2120 p. 7
`
`

`
`210
`
`A. Lifschilz er a1. / Vete1“1'nary Parasilology 86 {I 999) 203—2I5
`
`
`
`IVMconcentration(ng/ml)
`
`I IVM-TEST SC
`IVM-TEST IM
`[3 IVM—CONTROL
`
`|—‘
`
`Days post-treatment
`
`Fig. 4. Comparative mean plasma. concentrations of iVern1eeti11 (IVM) obtained between 20 and 35 days after
`tl1e administ1‘atio11 of tl1e IVM-TEST (subcutaneous and i11t1‘a111useula1') and IVM—CONTROL (subcutaneous)
`fonnulations to cattle (200 p.g/kg). Values are significantly different from tl1ose obtained for the IVM—CONTROL
`foimulation at (*) p< 0.05 a11d (**) p< 0.01.
`
`Table 1
`Mean pharmacokinetic parameters for ivermectin (IVM) obtained after administration of the IVM-TEST (subcu-
`taneous and intramuscular) and IVM—CONTROL (subcutaneous) formulations to cattle at 200 pg/kg
`
`Kinetic parametersa
`
`IVM-TEST scf
`
`IVM-TEST 1Mf
`
`IVM—Controlf
`
`0.50 :: 0.17
`0.69 :: 0.30
`l.48 :: 0.92b’e
`T1/gab (days)
`35.4 :: 9.67
`22.6 :: 4. 56'“
`19.9 :: 8. 84°
`Cm (ng/ml)
`1.63 :: 0.51
`2.25 :: 0.88
`4.00 ::1.41d>*=
`Tm (days)
`206 :: 35.8
`188 :: 23.9
`203 :: 43.9
`AUC .:0,35day5)(I1g day/ml)
`207 :: 35.7
`189 :: 24.0
`206 :: 41.3
`AUCt(,ta1 (ng day/ml)
`5.86 :: 2.26
`7.24 :: 2.66
`9.49 :: 3.89b
`MRT (days)
`3.99 :: 0.76
`5.20 ::1.11b
`5.90 :: 3.36b
`T1/2.1 (days)
`a T1/2-ab, absorption half —life; C},W: peak olasma concentration; Tmax, time to peak plasma concentration;
`AUC(g_35day), area under the concentration VS. time curve between drug administration and 35 day post—treatment;
`ALTCLULH1, area under the concentration VS. time curve extrapolated to infinity; MRT, mean residence time; T1/M
`elimination half—life.
`
`bMean kinetic parameters are significantly different from those obtained for the IVM—CONTROL formulation at
`p < 0.05.
`°Mean kinetic parameters are significantly different from those obtained for the IVM—CONTROL formulation at
`p < 0.0 1 .
`d Mean kinetic parameters are significantly different from those obtained for the IVM—CONTROL formulation at
`p < 0.00 l.
`eMean kinetic parameters are significantly different from those obtained for IVM-TEST IM at p < 0.05.
`fData Values are expressed as mean :: SD (I1: 8).
`
`Astrazeneca Ex. 2120 p. 8
`
`

`
`A. L1'fsc111'IZ er a1. / Veter“r'r1ary Parasitology 86 {I 999) 203—2I5
`
`21 1
`
`Table 2
`Comparison of the number of days post-treatment during which IVM plasma concentrations were greater than
`0.5 or 1 ng’ml, after the administration ofthe IVM-TEST (subcutaneous and intramuscular) and IVM-CONTROL
`formulations to cattle
`
`IVM-TEST SC
`IVM-TEST IM
`IVM-CONTROL
`
`Days >O_5 ng/ml
`27.5 :: 5.97?‘
`23.8 :: 3.53
`20.6 :: 3.20
`
`Days >1 ng/ml
`22.5 :: 5.343
`20.0 :: 2.67
`16.3 :: 2.32
`
`“Values are significantly different from those obtained for the IVM-CONTROL fonnulation at p < 0.05.
`
`4. Discussion
`
`Phannaceutical technology l1as been applied to develop different drug fonnulations and
`delivery systems to optimize the pharmacological potency of IVM and other endectocide
`molecules currently available. Alternative IVM formulations for use in several different
`species have been introduced to the market or are under development since the expiration
`ofthe original patent for the first approved IVM formulation (Ivomec®, MSD AgVet.). The
`persistence of the broad-spectrurn antiparasitic activity of endectocide compounds relies on
`their disposition kinetics and pattern of plasma/tissues exchange in the host (Lanusse et al.,
`1997). Even slight modifications to their plasma/tissue exchange pattern and/or disposition
`kinetics may result in substantial changes in their concentration and residence time at the
`site of parasite location which, ir1 tum, would alter the potency and persistence of their
`antiparasitic activity. The influence of drug fonnulation and route of administration on the
`disposition and plasma bioavailability of IVM in cattle has been confirmed in the trial
`reported here.
`The aqueous solubility of an active ingredient and the features of its pharmacotechnical
`preparation may influence the systemic availability (bioavailability), which depends on
`the rate and extent of absorption of a drug from the site of injection to the bloodstream
`(Baggot and McKellar, 1994). The vehicle in which endectocides are fonnulated may play
`a role in their absorption kinetics and resultant plasma availability (Lo et al., 1985; Lanusse
`et al., 1997). As demonstrated by Lo et al. (1985), the absorption of IVM is markedly
`influenced by its pharmaceutical preparation. Following parenteral administration, the low
`solubility ofIVM inwater and its deposition in subcutaneous tissue favour a slow absorption
`from the injection site and provide prolonged duration ir1 the bloodstream (Lanusse et al.,
`1997)
`The determination of the incremental AUC (partial AUCs) has been proposed as an ade-
`quate way to estimate the comparative rates of absorption of different dnrg formulations in
`bioequivalence trials (Chen, 1992). A significantly higher partial AUC0_Tm,X was obtained
`following the SC administration of the IVM-TEST compared to the IVM-CONTROL for-
`mulation (Fig. 1). In fact, a delayed absorption of IVM from the site of SC injection was
`observed following the administration of the IVM-TEST formulation. This slower absorp-
`tion process correlated with a peak plasma concentration attained at a delayed Tmax (4
`days), compared to those obtained after either the 1M administration of the same formu-
`lation (2.25 days) or the SC treatment with the IVM-CONTROL (1.63 days) fonnulation.
`Consistently, the absorption half—life of IVM given subcutaneously as an oil-based fonnu-
`
`AstraZeneca Ex. 2120 p. 9
`
`

`
`212
`
`A. L1'fsc111'IZ et a1. / Veterirrary Parasitology 86 {I 999) 203-215
`
`lation was significantly longer (1.48 days) than that obtained for the control formulation
`(0.50 days).
`IVM is absorbed relatively slowly from the site of injection, at rates which largely de-
`pend on the composition of the formulation vehicle. It has been previously shown that an
`inj eetable aqueous solution of IVM allows faster absorption and higher peak plasma concen-
`tration of the drug than administration of the drug in a non-aqueous vehicle (Lo et al., 1985).
`The commercially used vehicle contains propylene glycol/glycerol formal (60 :40) and is
`the formulation used as control preparation in the current trial, whereas the IVM-TEST
`formulation is prepared in an oil-based vehicle. The release of the drug from the subcu-
`taneous depot of the oily vehicle after SC administration of the IVM-TEST formulation
`delays the absorption process, resulting in a delayed peak concentration (delayed 7},,aX) and
`prolonged absorption half-life. Thus, the rate of absorption from the subcutaneous space
`appears to be the rate limiting step in the disposition of IVM after SC administration of the
`TEST formulation. This interpretation of the data is compatible with the lower peak plasma
`concentration (Cmax) obtained after SC administration of IVM in the oil-based vehicle com-
`pared with the IVM-CONTROL fonnulation. Field utility of long-acting formulations for
`endectocides may be more favoured by a delayed elimination with higher concentration
`profiles between 20 and 35 days post-treatrnent, instead of a high peak concentration; it
`is also likely that this consideration may be different for the control of different types of
`intemal/extemal parasites.
`The comparative analysis of the absorption phase suggests that the absorption process
`of the oil-based preparation may be faster from the site of M irgection compared to SC
`injection. This faster absorption was reflected in an earlier Znax and shorter T1/gab obtained
`after IM treatment compared to SC administration of the same TEST formulation (Table 1).
`Greater blood flow in muscle tissue, compared to the subcutaneous space, may favour faster
`absorption after IM treatment. However, a subcutaneously injected drug solution tends to
`take the shape of the space into which it is administered, resulting in an increased absorption
`surface compared to IM administration; this extended area of contact between the drug and
`blood vessels may compensate for the reduced blood flow i11 SC tissue (Nowakowski et
`al., 1995). The absorption of IVM injected as a non-oily preparation (IVM-CONTROL)
`from the SC tissue was consistently faster than that observed following the SC and IM
`administration of the TEST preparation. Both, the delayed absorption given by the oily
`vehicle and the increased absorption surface at the SC site of im ection may account for the
`‘flip-flop’ pl1er1or11er1or1, in which the disposition of the drug fror11 the body is controlled by
`the absorption process (Lanusse et al., 1997). In fact, the disposition of IVM given SC and
`IM as the TEST formulation was delayed compared to that obtained for the CONTROL
`preparation; both, the mean residence time (MRT) and T1/26,1 of the drug were prolonged
`after treatments with the oil-based formulation. A delayed elimination controlled by the
`rate of release of the drug from the depot at the site of injection, contributes to the greater
`persistence of IVM given as the TEST preparation. The influence of the absorption kinetics
`on the disposition of parenterally administered IVM has been demonstrated in this trial; an
`absorption rate-lirnited elimination of IVM may account for the long residence of high drug
`concentrations in the bloodstream following injection of the oil-based formulation to cattle.
`IVM is a highly lipophilic molecule that is extensively distributed into tissues (Bogan
`and MeKellar, 1988; Chiu et al., 1990). The distribution into adipose tissue is particularly
`
`AstraZeneca Ex. 2120 p. 10
`
`

`
`A. L1'fsc111'IZ er a1. / Veterinary Parasitology 86 {I 999) 203-215
`
`213
`
`relevant, since it may act as a drug reservoir that contributes to the long residence of the
`drug in the bloodstream. Higher IVM plasma concentrations were obtained between 7 and
`35 days after the SC administration of the IVM-TEST compared with the IVM-CONTROL
`formulation (Fig. 2). The plasma profiles of IVM after its IM administration were higher
`than the CONTROL from Day 11 to Day 35 post-treatment (Fig. 3). IVM concentration
`profiles in plasma reflect those achieved in different tissues, including those in which tar-
`get parasites are located (Lanusse and Prichard, 1993). A high correlation between drug
`concentrations in plasma and those achieved in different target tissues has been obtained
`after the SC administration of IVM, moxidectin and doramectin in cattle (Lifschitz et al.,
`1998). Therefore, the enhanced IVM plasma profiles obLai11ed between 20 ar1d 35 days
`after administration of the IVM-TEST formulation may result in an increased availability
`of the active drug in the tissues where target parasites are located, which may account for
`the prolonged anthelmintic persistence of the drug given to cattle in the TEST formulation.
`This may be relevant to control arrested larvae of developmental stages of gastrointestinal
`nematodes. Additionally, the duration of effective IVM plasma concentrations may be par-
`ticularly important in the treatment of tick infestations, since these organisms may feed on
`blood over several days.
`The drug concentration of an endectocide molecule required at the target tissues to inhibit
`either the development of larval stages or the establishment of dilferent internal and external
`parasites has not been determined. While IVM plasma concentrations as low as 0.48 ng/ml
`have been shown to control Hypoderma flies (Alvinerie et al., 1994), it is not clear what
`drug concentration is necessary to inhibit the development ofi11cor11ir1g L301" gastrointestinal
`nematodes. However, the correlation between the available data on the persistence of the
`anthehnintic activity of IVM and the concentration profiles determined in different pharma-
`cokinetic trials may provide useful information to estimate the minimal drug concentration
`above which larval development does not occur. It seems likely that IVM plasma concen-
`trations between 0.5 and lng/n11 would be indicative of the minimal drug level required
`for optimal anthelmintic activity for most gastrointestinal/lung nematodes. This theoretical
`assumption may assist in the indirect estimate of the impact of modified IVM formulations,
`such as the TEST preparation under study in the current trial, on the persistence of an-
`thelmintic activity. A comparative estimation of the period of time post-treatment in which
`IVM plasma concentrations were greater than either 0.5 or 1 ng/ml after administration
`of the different treatments is shown in Table 2. Consistent with the longer residence time
`of IVM administered subcutaneously as an oil-based formulation, there were significantly
`longer time periods during which IVM concentrations were >0.5 (27.5 days) and >1 ng/n1l
`(22.5 days) after SC administration of the IVM-TEST formulation, compared with those
`obtained for the standard IVM-CONTROL formulation (20.6 and 16.3 days, respectively).
`Although IM administration of the TEST formulation resulted in apparently greater ‘protec-
`tionperiods’ than the SC treatment with the IVM-CONTROL formulation, these differences
`did not reach statistical significance.
`Pharmaceutical preparations ca11 be designed to modify systemic pharmacokinetic pro-
`files through manipulation of the rate of absorption from the SC space (Wicks et al., 1993).
`Since the discovery and development of new molecules are long and expensive processes,
`the improvement of pharmaceutical preparations and delivery systems for existing dnrgs
`has been proposed as a reasonable alternative for the future of parasite control in livestock
`
`AstraZeneca Ex. 2120 p. 11
`
`

`
`214
`
`A. L1'fsc111'IZ et a1. / Veter“1'nary Parasitology 86 {I 999) 203—2I5
`
`(McKellar, 1994; Hennessy, 1997). The modified pharrnacokinetic pattern obtained for IVM
`after the administration of the novel oil-based formulation examined in this trial, compared
`to the standard preparation, may positively impact on the strategic use of the dnig in parasite
`control.
`
`Acknowledgements
`
`Adrian Lifschitz is a recipient of a fellowship from the Consejo Nacional de Ir1vestiga-
`ciones Cientificas y Ténicas (CONICET), Argentina. The technical advice of Dr. Michel
`Alvinerie (INRA, Toulouse, France) in the development of the analytical techmques is ac-
`knowledged. The authors gratefully acknowledge the cattle facilities provided by Dr. Mario
`Nardello. Research at the Laboratorio de Farmacologia, Departamento de Fisiopatologia,
`Facultad de Ciencias Veterinarias, Universidad Nacional del Centro (Tandil, Argentina)
`is partially supported by the Consejo Nacional de Investigaciones Cientificas y Ténicas
`(CONICET), Argentina, Comision de Investigaciones Cientificas de la Provincia de Buenos
`Aires (CICPBA) and Fu11daci()nAr1torcl1as (Argentina).
`
`References
`
`Alvineiie, M., Sutra, J.F., Galtier, P., 1993. Ivermectin in goat milk after subcutaneous injection. Vet.