Preclinical assessment of the feasibility of applying controlled release oral drug delivery to a lead series of atypical antipsychotics

Preclinical assessment of the feasibility of applying controlled release oral drug delivery to a lead series of atypical antipsychotics
of 13
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  Preclinical Assessment of the Feasibility of ApplyingControlled Release Oral Drug Delivery to a LeadSeries of Atypical Antipsychotics G. EICHENBAUM, 1 C. POLLOCK-DOVE, 3 J. NGUYEN, 3 S. LI, 3 J. EVANS, 3 H. BORGHYS, 2 L. KENNIS, 2 L. DONG, 3 W. VAN OSDOL, 3 W. DAI, 3 J. SCICINSKI, 3 J. CHEN, 3 Y. XU, 3 D. ASHTON, 2 C. MACKIE, 2 A. MEGENS 2 1  Johnson & Johnson Pharmaceutical Research & Development. 1000 Route 202 South, Raritan, New Jersey 08869 2  Johnson & Johnson Pharmaceutical Research and Development, Division of Janssen Pharmaceutica N.V., Beerse, Belgium 3 ALZA Corporation, 1010 Joaquin Road, Mountain View, California 94043  Received 24 August 2005; revised 23 October 2005; accepted 24 October 2005  Published online in Wiley InterScience ( DOI 10.1002/jps.20550 ABSTRACT:  In this paper, we present a preclinical approach for evaluating thefeasibilityofapplyingcontrolled-release(CR)oraldrugdeliverytoincreasethedurationof exposure and lower the  C  max  of compounds in a lead series of short half-life atypicalantipsychotics. Three lead compounds in the series had demonstrated potentialpharmacological benefitsforthe treatment ofpsychosis, inpreclinical studies.However,the compounds showed evidence of insufficient half-lives to enable a once-a-day (QD)product using immediate-release (IR) oral delivery. To evaluate and compare thepotential for oral CR delivery to extend the duration of action and thereby enable QDadministration, the  in vitro  solubility and permeability, and the duodenal and colonicabsorption of three compounds in the series were measured. Based on the results, onecandidatewasselectedforadvancementthatshowedmoderate invitro solubility,buthadthe highest  in vitro  permeability and ratio of colonic to duodenal bioavailability (0.9) inthe rat. The results from this study provided evidence that a CR drug delivery systemcould be used to extend the duration of exposure of the compounds in the series and ascientificbasisforselectingoneofthethreecompoundsasacandidate.   2006Wiley-Liss,Inc. and the American Pharmacists Association J Pharm Sci 95:883–895, 2006 Keywords:  site-specific delivery; site-specific absorption; oral absorption; ADME;pharmacokinetics; colonic drug delivery; intestinal absorption INTRODUCTION Controlled release (CR) drug delivery techno-logies have generally been applied to rescuepost-Phase II clinical candidates and to enablelifecycle management of marketed compounds. 1–5 However, CR drug delivery technologies haveseen limited application in the optimization of leadcompoundsatadiscoverystage. 6 Compoundsidentified during lead optimization that havephysicochemical or pharmacological propertiesbelieved to be unsuitable for immediate-release(IR) oral delivery (e.g., short half-lives or max-imum serum drug concentration [ C  max ]-relatedside effects) are sometimes not progressed, des-pite having promising and in some cases superiorpotency, selectivity, and/or efficacy. 6,7 The testing for feasibility and application of CR drug deliveryat an earlier stage offers an approachfor advance-ment of lead compounds and reducing attrition.Biopharmaceutical, pharmacokinetic (PK), orpharmacodynamic (PD) properties that make com-poundsincompatiblewithoralIRdrugdeliverymay  JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 4, APRIL 2006  883 Correspondence to : G. Eichenbaum (Telephone: 908-707-3574; Fax: 908-253-0448; E-mail:  Journal of Pharmaceutical Sciences, Vol. 95, 883–895 (2006)  2006 Wiley-Liss, Inc. and the American Pharmacists Association  makethemwellsuitedforCRdeliverytechnologies.These compound properties include short half-life,moderate first-pass effect, poor solubility, limitedupper gastrointestinal (GI) permeability, a narrowtherapeutic index, side effects related to  C  max  orrate-of-rise, or a high rate of metabolism in thesmall intestine. 8 If recognized and evaluated dur-ing the lead optimization stage, CR drug deliverycould enable successful advancement of thesecompounds. One of the benefits of considering CRat the late lead optimization stage is the ability toselect compounds from a series based on physico-chemical properties that make a molecule suitablefor oral or transdermal CR delivery. Alternatively,consideration of CR at an earlier stage in leadoptimization could enable a priori design of physi-cochemical properties that are required for CR bystructural modification.To fully exploit the transit time of a typical oralCRdevice(12–20h), 9 adequatecolonicabsorption,solubility, and stability of the formulation to pre-cipitation, and a sufficiently low dose size are keyrequirements. 8 It is now recognized that themarkedly slower rate of material transit in thecolon—11 to 12 h in the proximal colon 9,10 com-pared with only 3–4 h in the small intestine 11  — permits drugs to stay in contact with the colonicmucosa for a longer time than in the small intes-tine, and thus greatly increases the total absorp-tion within this compartment. In addition, theabsorptive capacity of the colon hasbeen shown tobe high for highly lipid-soluble compounds 12,13 despite lower absorptive area, lower mesentericblood flow, lower paracellular pore size, andreduced carrier mediated transport capacity forimproving bioavailability. 14 Historically, two types ofapproacheshave beensuccessfully applied to assess the feasibility andbenefits of oral CR drug delivery for drug candi-dates that are in late-stage clinical trials or on themarket. The first approach is to directly test a CRdosage form in animals and then in humans. Thesecond approach is to directly measure the extentof colonic absorption using human colonic intuba-tion studies. 15 However, neither of these appro-aches is feasible during the lead optimizationstage. Direct evaluation of colonic absorptionin humans by colonic intubation is not possiblebecause of the unknown safety profile of thecompounds. In addition, limited quantities of material and time make it impractical to prepareand compare IR and CR dosage forms for multiplecompounds. Therefore, preclinical assays areneededforevaluatingoralCRtechnologiesduring lead optimization. The need to develop and applypreclinical assays(e.g.,toevaluatecolonicabsorp-tion) to assess the feasibility of CR oral delivery isalsomotivatedbytheobservationthatthereareGIcompartment-to-compartment differences (i.e.,duodenum vs. jejunum vs. ileum vs. colon) inabsorption and metabolism 13,16,17 that impact thefeasibility of CR. Aspartoftheresearcheffortsthatprecededthepresent work, studies of   in vitro  receptor binding profilesand invivo PDeffectswereperformedforaseriesofstructurallyrelateddopamineD2antago-nists. Based on the results, these compounds,which have structural similarities to risperidone,wereidentifiedashavingpotentialbenefitsforthetreatment of psychosis. The primary motivationfor considering CR delivery for this series of compounds at the lead optimization stage wasthe PK and PD properties that they displayed in adog apomorphine-induced emesis model. Theseproperties suggested they would have insufficienthalf-lives or durations of action to achieve QDdosing, whichwasconsideredessentialforfurtherdevelopment. A further motivation for selecting acompoundthatwaswellsuitedforCRdeliverywasto enable testing of the impact of a lower  C  max  onside effects, such as extra pyramidal side effects(EPSs) or prolactin release, which are common forother atypical antipsychotics. 18,19 Finally, thepotency of the compounds compared with those of antipsychotics on the market suggested that thedose would not prohibit the use of an oral CRdelivery system.Herein we describe an approach for preclinicalassessment of the feasibility of oral CR deliveryleadingtotheselectionofaleadmoleculefromthepresent series. MATERIALS AND METHODS Materials  JNJ16558711  (TRANS 3-[2-(8-bromo-1,3,4,4a,9b-hexahydro-2H-pyrido[4,3-b]indol-2-yl)ethyl]-2,8-dimethyl-4H-pyrido[1,2-a]pyrimidin-4-one),  JNJ16601650  (TRANS 3-[2-(8-chloro-1,3,4,4a, 5,9b-hexahydro-2H-pyrido[4,3-b]indol-2-yl)ethyl]-2,8-dimethyl- 4H-pyrido[1,2-a]pyrimidin-4-one),  JNJ16514537   (TRANS 3-[2-(8-chloro-1,3,4,4a,5,9b-hexahydro-2H-pyrido[4,3-b]indol-2-yl)ethyl]-2-methyl-4H-pyrido[1,2-a]pyrimidin-4-one),and  ris- peridone  (3-[2-[4-(6-fluoro-1,2-benzisoxazol-3-yl)-1-piperidinyl]ethyl]-6,7,8,9-tetrahydro-2-methyl- 884  EICHENBAUM ET AL.  JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 4, APRIL 2006 DOI 10.1002/jps  4H-pyrido[1,2-a]pyrimidin-4-one) were all obtainedfrom the internal Johnson & Johnson Pharmaceu-tical Research & Development compound collection.Purity of the compounds was confirmed by elemen-tal analysis and found to be  > 99%. Other reagentsused were of the highest grade available. Aqueous Solubility Versus pH Standard United States Pharmacopoeia (USP)buffer solutions of artificial intestinal fluid (AIF)(pH 6.8 and 7.5) and artificial gastric fluid (AGF)(pH 1.2) were used. Stock solutions of AIF wereprepared by adding 3.4 g of potassium phosphatemonobasic (KH 2  PO 4 ), and either 0.8 g (pH 7.5) or0.45 g (pH 6.8) of sodium hydroxide (NaOH) to0.50 L of purified deionized water. A stock solu-tion of AGF was prepared by adding 1 g of sodiumchloride (NaCl) and 3.10 mL of hydrochloric acid(36.5–38.0% HCl) to 0.5 L of purified deionizedwatertoachievea pHof1.2. Additionally,abufferof pH 4.5 using potassium monobasic and asolution of NaOH was prepared at the same ionicstrength of 0.15. These four buffer solutions weremixed to create a set of buffer solutions withvarious pH values (pH 1.2, 3, 4.5, 6.8, 7.5, 8.5, 10).Saturated solutions were prepared by adding sufficient compound to 0.5-g buffer solutions,vortexing to disperse the compound, and thensonicating for 10 min at room temperature. Thesolutions were then agitated at room temperaturefor at least 24 h. After centrifugation, samples of the supernatant were taken for high-performanceliquid chromatographic (HPLC) analysis. Thefinal pH was measured and shown. Due to limitedavailability of compound, only one sample wasprepared and measured at each pH. HPLC for Solubility Assays Formic acid, ammonium hydroxide, and ammo-niumformatewerepurchasedfromSigmaAldrich(St. Louis, MO). Methanol was American Chemi-cal Society (ACS) grade and was purchased from VWR (West Chester, PA). Water was preparedwith a Milli-Q 1 water purifier (Billerica, MA). Allchemicals used for buffer preparation were of reagent grade.High-performance liquid chromatography wascarriedoutwithanAgilentSeries1100comprising the modular components: quaternary pump, avacuum solvent microdegasser, an autosamplerwith a 96-well plate tray, and an on-line diodearray detector. The column was a Zorbax 300SB-C8 (4.6  150 mm, Agilent, Foster City, CA)maintainedat50 8 C.Themobilephaseswerewaterwith 0.4% ammonium formate (pH 3.1) andmethanol. A gradient method was used starting from90%Ato10%B.After30min,theportionofA was linearly decreased to 10%. The initial mobile-phase composition was restored within 1 min andmaintained for column regeneration (equilibra-tion)foranother9min,resultinginatotalruntimeof 40 min. The flow rate was 1 mL/min, and theinjection volume was 50  m L. The diode arraydetector was programmed to acquire the ultra-violet (UV) spectra at 246 nm. For solubilitytests, an isocratic method was used. The mobilephase composition was 0.4% ammonium formate(pH 3.1): methanol, 59:41. The total run time was10 min.Stock solutions were prepared by dissolving 1 mg of compound in 20 mL of methyl alcohol(MeOH)andsonicatingfor10min.Serialdilutionsfrom the stock solution resulted in concentrationsof 3.125, 6.25, 12.5, and 25  m g/mL. Artificial Membrane Permeability Permeability studies were performed in thesame manner as described previously. 20,21  A PIONDouble-Sinkprotocolwasusedforthepermeabilitymeasurement. A 96-well microplate (acceptor com-partment) was completely filled with 200  m L of  Acceptor Sink Buffer with a pH of 7.4 (NaPO4buffer containing 5% dimethyl sulfoxide [DMSO]). A hydrophobic filter plate was fixed on the bufferfilled plate. The filter surface was impregnatedwith 4  m L of lipid solution, which was composed of PC (0.8% w/w)/PE (0.8% w/w)/PS (0.2% w/w)/PI(0.2%w/w)/CHO (1.0% w/w), and 1,7-octadiene(97.0%w/w). A 96-well microplate (donor compart-ment)wasfilledwith200 m L of 50- m M sample stocksolution at specified pH (3–8). The donor plate andthe acceptor plate were put together forming aparallel artificial membrane permeability assay(PAMPA) sandwich (acceptor/lipid/donor). ThePAMPA sandwich was allowed to incubate atambient conditions without mixing for 4 h. Thecompound concentrations in the acceptor compart-ment and donor compartment were quantified byUV spectroscopy. Lipid retention of the compoundswas taken into consideration when PAMPA P eff  was calculated. Verapamil was included as apositive control because it has both high PAMPA flux (60  1/10 6 cm/s) and high absorption inhumans (95%). 20 Unless otherwise indicated, thedata point at each pH is an average of eight PRECLINICAL CONTROLLED RELEASE  885 DOI 10.1002/jps JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 4, APRIL 2006  measurements that were made on two separateplateswithfoursamplesperplate.Exceptionswereas follows: five, six, and seven replicates wereobtained for verapamil at pH 7.4, 6.2, and 5.0,respectively; seven replicates were obtained forrisperidone at pH 7.4; and seven replicates wereobtained for JNJ16558711 at pH 7.4 and pH 5.0. Dosing Solution Preparation  A summary of the dosing schedule is shownin Table 2. For IV dosing, JNJ16558711, JNJ16601650, and JNJ16514537 were added to0.9% saline solution at 0.25 mg/g, vortexed, andthen sonicated for 10 min. The acidity of thefumaric salts allowed for complete solubilizationof the compounds. Risperidone (free base) requiredtitration with an acidic buffer (AGF at pH 1.2) tolower the pH and solubilize the compound. Forintestinaldosing,deionizedwaterwasusedinsteadof0.9%salineandtheconcentrationwasloweredto0.16 mg/g. Artificial gastric fluid was again usedto acidify the risperidone solution. The final pH of all of the aqueous dosing solutions was 3.0.For the formulation dosing, a 50/50 blend of Transcutol/Tween 80 with 0.5% butylated hydro-xyltoluene was used. Concentrations of 0.16 mg/g or 1.6 mg/g for all four compounds were prepared,vortexed, and then sonicated for 10 min. Standard Solution Preparation The JNJ16558711, JNJ16601650, and JNJ16514537 stock solutions (25.8, 37.4, and 33.2  m g/mL)were prepared by dissolving the corresponding compounds in methanol. Intermediate standard Table 1.  Structures and Physicochemical Properties of Three Gamma Carbolines and RisperidoneCompound Structure Mol wt. pKas a LogP b Human JejunalP eff   (  1/10 4 cm/s)Risperidone 410.5 7.91 3.75 4.46 JNJ-16514537 394.9 8.48,3.68 3.55 2.51 JNJ-16558711 453.4 8.55,3.55 3.87 2.92 JNJ-16601650 408.9 8.49,3.68 3.84 2.78 a  ACD pKa DB (Advanced Chemistry Development, Inc., Toronto ON, Canada,, 2003). b QMPRplus (Simulations Plus, Inc.,, Lancaster, CA, 2004). Table 2.  Dosing Schedule for Rat In Vivo StudiesName Site(s) VehicleDose(mg/kg)Concentration(mg/mL)Dosing volume(mL/kg)IV Intravenous 0.9% saline, pH 3 0.10 0.25 0.4 Aqueous Duodenum/colon Deionized water, pH 3 0.10 0.16 0.625Formulation Duodenum/colon Transcutol/Tween 0.10 0.16 0.625 886  EICHENBAUM ET AL.  JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 4, APRIL 2006 DOI 10.1002/jps  solutions were obtained by appropriate dilutionof each stock solution with deionized water.Standards for calibration curves were preparedby serial dilutions of the intermediate standardsolutions in rat plasma to achieve concentrationsof 0.2, 0.5, 2.0, 10, 50, 100, and 250 ng/mL of eachcompound. Quality control samples with concen-trations of 0.6, 60, and 200 ng/mL were preparedin the same manner. A risperidone stock solution (25  m g/mL) wasprepared by dissolving the pure compound in 50/ 50 (v/v) acetonitrile/deionized water. Intermediatestandard solution was obtained by appropriatedilutionofthe25- m g/mLstocksolutionswithdeioni-zed water. Plasma standards for calibration curveswere prepared by serial dilutions of the inter-mediate standard solution in rat plasma to achieveconcentrations of 0.2, 0.5, 1.0, 5.0, 10, 50, 100, and200 ng/mL of risperidone. Plasma standards withrisperidoneconcentrationsat0.6,60,and160ng/mLwere also prepared in the same manner. Protein Precipitation The plasma sample cleanup procedure for JNJ16558711, JNJ16601650, JNJ16514537, andrisperidone consisted of several steps: transfer-ring 50  m L of plasma sample to 50  m L of buffer (pH6.2); mixing well; adding 300  m L of methanolcontaining corresponding internal standard; vor-texing for 3 min; transferring 300  m L to a proteinprecipitation filter plate with collection; filtering under the pressure; and adding 100  m L of 0.1%formic acid solution to above collected solution.The 10  m L of the final solution was injected intothe liquid chromatography/tandem mass spectro-metry (LC/MS/MS) system. Mass Spectrometry Mass spectrometry was performed on a PE-Sciex API 4000 triple quadrupole mass spectrometerequipped with a Turbo Ion Spray interface. High-performance liquid chromatography eluate wasintroduced to the stainless steel capillary sprayerheld at 5.5 kV through a 50- m m inner diameter(ID)stainlesssteeltubing.Themassspectrometerwas optimized in the multiple reaction-monitor-ing mode (MRM). The parent/product ion pairs(m/z) focused were 453.1/201.1 for JNJ16558711,409.1/201.1 for JNJ16601650, 395.1/187.1 for JNJ16514537, and 411.1/191.1 for risperidone.The HPLC system included Shimadzu HPLCpumps and a Leap autosampler. Chromatographyseparation was achieved on a phenomenex 1 C18BDSHyperClonecolumn(50  2.0mm,5 m mparti-cle size). Two mobile phases were used for thisHPLC method. Solution A consisted of 0.4% aque-ousformicacid(pH2.9)whileSolutionBconsistedof methanol. Sample was eluted isocraticallyat 40% B for the analysis of JNJ16558711, JNJ16601650, and JNJ16514537; and at 35% Bforrisperidone.Theretentiontimeswere1.89minfor JNJ16558711, 1.34 min for JNJ16601650,1.01 min for JNJ16514537, and 1.02 min forrisperidone. The mobile phase was delivered at aflow rate of 600  m L/min. Apomorphine-Induced Emesis in Dogs  Antagonism of apomorphine-induced emesis indogs is a simple and reliable measure of the func-tional D 2  antagonism of neuroleptics in speciesthat—more than rodents—is predictive for thepharmacokinetic properties of compounds inman. 22,23 It enables evaluation of the oral bioacti-vity and the onset and duration of action of neuro-leptics. Apomorphine (0.31 mg/kg, sc)-inducedemesis was assessed up to 1 h after challenge indogs of both sexes and various body weights. 23 Test compound orsolvent was given 1 h before theapomorphine injection. Criterion for drug-induced protection: complete absence of emesis(not observed in controls;  n  > 1,000). ED50’s andcorresponding 95% confidence limits were calcu-lated according to the method of Finney (1962) forcategorical data. 24 Dopamine antagonists protectagainst apomorphine-induced emesis in dogs. Thetest is predictive for potency, oral absorption, andonset and duration of action of dopamine antago-nists. Rat Site-Specific Dosing Male Sprague Dawley rats weighing 300–400 g were purchased from Charles River Labs. All ratexperiments were conducted under animal proto-cols approved by the ALZA Institutional AnimalCare and Use Committee (IACUC). The rats werecatheterized at Charles River Labs and had oneor more of the following chronic catheters fordosing and sampling: jugular/carotid, duodenal,ileal, and colonic. Rats were limited to only one GIcatheter and one vascular catheter; triple-cathe-terized rats were not used. For IV dosing, ratswere purchased with jugular and carotid cathe-ters for dosing in the jugular catheter and bloodsampling through the carotid catheter. For PRECLINICAL CONTROLLED RELEASE  887 DOI 10.1002/jps JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 95, NO. 4, APRIL 2006
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks