`
`ENDO - Ex. 2037
`
`Amneal v. Endo
`
`|PR2014-00360
`
`ENDO - Ex. 2037
`Amneal v. Endo
`IPR2014-00360
`
`

`

`.38:25292.823
`
`
`
`megaw_omm<mxm_w_.mgamma.2genommazn
`..wOEZCEQHOznw"
`
`
`
`
`
`
`m<mIm_Um1:-<<m_mmm~v3.0.
`
`
`
`
`
`><$3m3<moocr.urb.
`
`magmaU<
`
`535:0:
`
`@8meman—
`
`
`
`magmas.
`
`93:2a.25.352comma—ma...”9:23.35:2a
`
`
`
`1.26930:1:30*amigo.583%
`
`¥ODE.0503256
`
`DEL—Sam
`
`@
`
`3&1m2?292<0;
`
`>318:93335833:553%533.818.
`
`
`
`
`
`
`

`

`U.S.A.
`
`U.K.
`
`PEOPLE’S REPUBLIC
`OF CHINA
`
`FEDERAL REPUBLIC
`OF GERMANY
`
`BRAZIL
`
`AUSTRALIA
`
`JAPAN
`
`CANADA
`
`Pergamon Press, Maxwell House, Fairview Park,
`Elmsford, New York 10523, U.S.A.
`
`Pergamon Press, Headington HiII Hall,
`Oxford OX3 OBW, England
`
`Pergamon Press, Room 4037, Qianmen Hotel, Beijing,
`People’s Republic of China
`
`Pergamon Press, Hammervveg 6,
`D~6242 Kronberg, Federal Republic of Germany
`
`Pergamon Editora, Rua Eca de Oueiros, 346,
`CEP 04011, Paraiso, Séo Paulo, Brazil
`
`Pergamon Press Australia, PO. Box 544,
`Potts Point, N.S.W. 2011, Australia
`
`Pergamon Press, 8th Floor, Matsuoka Central Building,
`1-7-1 Nishishinjuku,- Shinjuku-ku, Tokyo 160, Japan
`
`Pergamon Press Canada, Suite No. 271,
`253 College Street, Toronto, Ontario, Canada M5T 1R5
`
`Copyright © 1988 American Pharmaceutical Association
`All Rights Reserved. No part of this publication may be
`reproduced, stored in a retrieval system or transmitted in any
`form or by any means: electronic, electrostatic, magnetic tape,
`mechanical, photocopying, recording or otherwise, without
`permission in writing from the copyright holders.
`First edition 1988
`
`British Library Cataloguing in Publication Data
`Oral sustained release formulations : design
`and evaluation.
`1. Controlled release preparations
`I. Yacobi, Abraham II. HaIperin-Walega, Eva
`615’.6
`R8201.C64
`ISBN 0—08—036075—0
`
`Library of Congress Cataloging-in-Publication Data
`Oral sustained release formulations: design and evaluation/edited by
`Avraham Yacobi, Eva Halperin-Walega.
`”Published in cooperation with the American Pharmaceutical
`Association"——Cover.
`Includes index.
`
`I. Yacobi,
`2. Oral medication.
`1. Drugs—Controlled release.
`Avraham.
`II. Halperin-Walega, Eva.
`Ill. American Pharmaceutical
`Association.
`
`[DNLM: 1. Biopharmaceuticals.
`3. Dosage.
`OV 785063]
`R8201.C64073 1987
`615’.19——dc19
`
`87-29283 GIP
`
`2. Delayed-Action Preparations.
`
`Printed in Great Britain by A. Wheaton & Co. Ltd., Exeter
`
`
`
`

`

`Drug Studies Unit and the Departments of Medicine and Pharmacy, University of California,
`San Francisco, CA 94143, USA
`
`INTRGDUCI‘ION
`
`Chapter 9
`Controlled Release Drug Delivery: Pharmacodynamic
`Consequences
`
`J. Mordenti and R. L. Williams
`
`metabolism.
`
`Drugs are given via several routes and at varying
`ranging
`from administration
`as
`a
`bolus
`to
`rates,
`administration at a constant rate over several hours. days,
`or even longer. Although most drugs have been given as
`immediate release oral fornulations (capsules, tablets, or
`liquids) ,
`clinicians
`are
`accustomed
`to administering
`certain drugs such as antiarrhythmic agents or anticancer
`drugs
`intravenously at constant
`infusion rates.
`This
`method of drug administration is particulary applicable for
`drugs with low therapeutic indices. short half-lives, or
`high
`first pass hepatic metabolism.
`Intraarterial,
`intragastric,
`intrapulrmnary
`(e.g. ,
`inhalant),
`and
`intracystic administration of drugs at constant rates have
`also been errployed therapeutically (Table 1).
`These
`standard form of drug administration at constant rates are
`fornulation—independent
`and
`do
`not
`rely
`on
`a
`biopharmaceutical configuration of drug and excipients to
`control
`the release rate of
`the drug. More recently,
`formulation dependent controlled release drug products have
`permitted delivery of drugs at controlled rates to any of
`several
`sites
`in
`the
`body,
`including
`the
`skin.
`gastrointestinal tract, and eye (Table 1). This capability
`has arisen because of rapid advances in the biotechnology
`of controlled release formlations and
`because of the
`clinical advantages of controlled delivery of certain drugs
`to the body. Formlation dependent controlled release drug
`delivery,
`like
`formulation
`independent methods,
`is
`particularly valuable
`for
`drugs with low therapeutic
`indices,
`short half-lives or high first pass hepatic
`
`

`

`J. Mordenti and R. L. Williams
`
`m1
`
`
`
`mmmmmmmmmmnmc
`DELIVERY
`
`1, bottom panel).
`
`An optimally formlated and bioavailable controlled
`release product can produce effective concentrations of
`drug in blood or plasma over an extended time period.
`It
`can also avoid the wide oscillation between near toxic or
`toxic and subtherapeutic drug concentrations that can occur
`with immediate release formlations of drugs with short
`half-lives (Figure 1, top panel). An added advantage of a
`controlled release product is the improvement in conpliance
`that may occur with once or twice daily dosing, as opposed
`to the four or more doses that nust sometimes be given when
`the drug is prepared as an immediate release preparation.
`Attributes of a controlled release product can become
`disadvantages in certain circunstances. A poorly available
`or
`inadequately dosed controlled release product may
`produce no effective concentrations throughout a dosing
`interval, whereas an immediate release product with the
`same degree of bioavailability or dosing limitation may
`still
`produce
`intermittently effective
`concentrations
`(Figure 1, middle panel).
`Analogously,
`a controlled
`release formlation that delivers too nuch drug may produce
`toxic or near toxic concentrations for an extended period,
`whereas an imnediate release formlation, even at too high
`a dose, can permit periods of respite from toxicity (Figure
`
`

`

`
`
`
`
`Controlled Release Drug Delivery
`
`197
`
`Toxic
`
`I
`Effective
`
`%:
`
`<I
`
`I:
`*—
`
`ZL
`
`Uo
`
`5
`0
`
`g
`,:
`:
`i.—
`
`<L
`
`2m g
`
`0
`
`Z 9’
`
`—<D
`
`:
`p—
`
`Zw0ZO0
`
`EFFECT OF FORMULATION AND BIOAVAIBILITY ON
`GURE 1:
`“4 ‘ OR BLCXD CONCENTRATIONS FM HYPO‘IHEYI‘ICAL DMEDIA'IE
`CONIROIJID RELEASE
`FORMULATIONS:
`I)
`OPI‘IMAILY
`’
`OAVAIIABLE
`FORMULATIONS;
`II)
`PWRLY
`BIOAVAILABLE
`ULTIONS} and III) OPI'IMALLY BIOAVAILABLE FORMULATIONS
`IN EXCESSIVE DOSE (SHADE) AREA = CONTROLLED RELEASE
`; OPEN AREA = M’IEDIA'IE RELEASE W).
`
`
`
`
`
`
`.
`Toxuc
`
`» Effective
`
`
`
`

`

`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`198
`
`J. Mordenti and R. L. Williams
`
`
`
`controlled
`of
`consequences
`pharmacokinetic
`The
`' release dosage form are generally well understood and ‘
`described.
`The pharmacodynamics changes associated with '
`administration of a controlled release drug system are less -'
`well
`studied.
`In this review, we will discuss
`the
`pharmacokinetic and
`pharmacodynamic
`factors
`that
`are
`pertinent
`to the clinical use of drugs
`in controlled
`release fornulations.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`PHARMACOKINETIC AN) mammmrc FACTORS:
`W
`
`General principles of action for many drugs are based
`on the assunption that a pharmacologic effect is produced
`by interaction of unbound drug or metabolite with one or
`more receptors (Figure 2).
`If excess drug or metabolite
`reaches these primary receptors or if a different set of
`receptors
`that produce an undesirable side effect are
`stinulated, toxicity can ensue. The relationships shown in
`Figure 2 assume that the concentration of unbound drug and '
`‘ metabolite
`(Cu)
`in the blood or plasma
`is in direct
`equilibrium with the concentration of drug and metabolite
`at the receptor site. Description of the time course of a
`drug
`in
`blood
`or
`plasma
`can
`be
`described with
`phannacokinetic methods. A description of the effect of a
`drug relative to this concentration (or drug dose) may be
`described using pharmacodynandc methods.
`Pharmacokinetic
`methods
`of
`analysis
`based
`on
`conpartmental,
`nonconpartmental or physiologic models have been well
`described and widely applied.
`Pharmcodynanfic methods are
`less generally utilized, perhaps because current methods of
`assessing
`pharmacologic
`effects
`in
`vivo
`are
`less
`sophisticated
`than
`the
`ability to measure
`unbound
`concentration of drug or netabolite in biologic fluids.
`Despite the difficulty in assessing the pharmacologic
`effect of a drug, correlations between drug concentration .
`(or dose) with a desired pharmacologic response are useful
`in the design and selection ‘of optimal dosing regimens
`(including those for controlled release drug products).
`With this informtion, and in conjunction with a conpetent
`analytical
`laboratory, more rational drug dosing regimens
`can be developed. Drug concentration and effect relation-
`ships
`can
`also be used to explore the consequence of
`formlating a drug as a controlled release product.
`
`
`
`
`
`
`
`and
`investigators
`provided
`have
`reviews
`Recent
`clinicians with technques to define the effect of a drug
`
`
`in relation to its concentration in a biologic fluid of
`
`
`interest
`(1) .
`These techniques utilize any of several
`
`
`pharmacokinetic methods to describe the concentration-tine
`
`

`

`Controlled Release Drug Delivery
`
`199
`
` PHARMACOKINETIC
`
`Ran-MW.“mam-m
`
`
`
`ABSORPTION
`
`-—-—> Cu ,pare nt
`Cu,metabolite
`
`CXNCENIRATION OF DRUG.
`
`Cu,metabolite
`
`!! I!
`
`ELIMINATION
`
`———->
`
`DISTRIBUTION
`
`
`
` hm «my «
`
`
`
`Cu,parent
`
`EFFECT
`
`RELATIONSHIPS
`PHAR’IACOKJNETIC/PHAMACGDYNAMIC
`FIGURE 2:
`BE‘IWEEN DRUG DOSE AND DRUG EFFECT.
`Cu IS THE UNBOUI‘D
`
`

`

`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`4
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`200
`
`J. Mordenti and R. L. Williams
`
`‘Ihrough the application of
`course of a drug in the body.
`one or more pharmacodynamic models
`relating effect
`to
`concentration, conbination pharmcokinetic/pharm-codynamic
`models may
`describe
`the
`relationship
`between
`drug
`concentration and effect.
`Primary pharmacokinetic and
`pharmacodynamic models
`and
`their
`interrelationship are
`shown in Table 2.
`
`Knowledge of- concentration/effect relationships can
`be important
`for predicting the change in pharmacologic
`effect that may occur when a patient is transferred from an
`mmediate release to a controlled release preparation.
`From a scrutiny of the plasna concentration versus time
`curves depicted in Figure 1, it might be supposed that a
`marked increase in pharmacologic effect will occur at the
`high
`concentrations
`of
`drug
`that
`are
`achieved
`
`Information
`relative to a controlled release product.
`about
`the concentration/response relationship can confirm
`or
`refute this suppositio .
`(he possible relationship
`between drug concentration and effect
`is the me or
`sigmoid Em}, model
`(Table 2). As discussed by Holford and
`Sheiner
`(1) ,
`this relationship is inherently attractive
`because it postulates no effect when no drug is present and
`suggests
`that
`a
`naxinum effect
`is
`attained as
`the
`concentrations of drug increases.
`One characteristic of
`the me model is that at concentrations above the EC50
`(the concentration at which effect
`is half maximal)
`increments
`in
`concentration
`are
`associated
`with
`disproportionally smaller increments in effect
`(Table 2).
`In this instance,
`the higher concentrations of drug that
`occur with an immediate release product may not necessarily
`be associated with proportional
`increments
`in effect.
`Alternately,
`if the drug concentrations following dosing
`are
`below the E050,
`increments
`in blood or plasna
`concentration my result in alrrost proportional increments
`in pharmacologic effect.
`'Ihe pharmacologic consequences of
`formulating a drug as a controlled release product may be
`miniIral or profound depending on the concentration achieved
`relative to the EC50.
`
`Clinical data document these observations for a num-
`nunber of drugs. Carruthers (2) studied the dose-response
`relationships of pindolol when given orally as inmediate
`release or controlled release preparations at different
`doses. Data in this study indicated conparable reductions
`in exercise heart rate after doses of immediate release
`pindolol
`ranging between 5 and 20 mg and a controlled
`release fornulation of either 20 or 30 mg (Figure 3).
`In
`
`this study,
`a Inaxinum pharmcologic response to pindolol
`
`
`
`
`

`

`Controlled Release Drug Delivery
`
`201
`
`131132
`
`H)
`PR
`NEE—WIRI—
`
`
`PHARMMICKINETIC MODELS
`
`CCMPARIMENI‘AL
`
`NONGMPARIMENI‘AL
`
`PHYSIOLOGIC
`
`PI-IZ-‘RMACCDYNAMIC MODELS
`
`memcr
`
`LINEARMGJEL
`
`E=SC
`
`IOGLINEARMCDEL
`
`E=SlogC+I
`
`EMAXMGDEE.
`
`E=EMAXC
`
`SIGMOID EMAX MODEL
`
`E = W1“
`
`EC50 + C
`
`EC50N + CN
`
`
`llllIIII
`
`E C S I
`
`effect.
`
`concentration of drug.
`slope of the line relating effect to concentration.
`arbitrary constant with no physical meaning.
`EMAX = naxinum effect attributable to the drug.
`EC50 = concentration producing 50% of EMAX-
`N = a number influencing the slope of the curve.
`
`less.
`was probably attained at oral doses of 5 mg or
`Administration of the drug at higher doses, whether as a
`controlled or immediate release product, produced less than
`proportional decrements in exercise heart rate.
`In terns
`of an me pharmcodynanfic model, the doses of pindolol in
`this study produced concentrations that were well beyond
`those at which a response was half maximl
`(the EC50) and
`probably close to or beyond the concentration where naxinal
`
`
`
`

`

`finedroMIv.
`
`dna
`
`LR.
`
`sm.m
`
`
`mmHofim—Uxm.mOZOHEAEUO2092~.Uyfinn—zfimmhammmHEmagEmma—manyE:”MmmRva.«NV.HQHOnZHmmmfiHmlem—ZHEWDW02¢QSOQAHQAfizOHBzm—DZBm0mmmon5305950mwm92~m.N27238mm
`
`
`
`
`
`
`
`
`(wdm 31w iWBH
`
`UZ.OZ<»w
`
`ozzhut
`
`359,:m2;
`
`

`

`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Controlled Release Drug Delivery
`
`203
`
`for nitroglycerin
`Comparable data
`effect occurred.
`administered intravenously were reported by Imhof et a1
`healthy volunteers,
`reduction in
`
`re than two fold increment in the
`rate- of administration of
`the drug
`(from 3.4 to 7.5
`meg/min).
`These data suggest
`that
`the EC50
`for
`this
`
`steady state
`(the mean
`administration of the drug at
`
`below approximately 0.5 ng/ml
`concentration
`produced
`by
`a rate of 3.4 meg/min).
`
`the
`information about
`Better
`rate of administration.
`phannacokinetic
`and
`pharmacodynamic behavior of drugs
`selected for controlled
`release
`fornulation
`suggest
`that
`the pharmacologic effect of
`a drug may differ
`depending on the route of administration and the release
`characteristics of its formlation. Variation in response
`between an J'erediate and a controlled release fornulation
`can depend on both pharnacokinetic and pharmacodynamic
`factors
`(Table 3).
`Pharmacokinetic
`factors
`that
`can
`contribute to differences in response between an immediate
`and a controlled release formulation include variation in
`the concentration ratio of active metabolite (s)
`to parent
`drug.
`as well
`as differences
`in drug or metabolite
`concentrations arising as a result of nonlinear metabolism
`or nonlinear protein binding.
`pharmoodynanfic factors
`causing variable response beth immediate and controlled
`release formilations
`relate to differences in receptor
`sensitivity
`produced
`by
`relatively
`constant
`concentration of drug at
`the receptor site following
`administration of a controlled release product, as opposed
`to rising and falling concentratio
`following
`administration
`of
`an mediate
`formlation.
`
`Pharmacokinetic Factors
`
`
`selected
`Drugs
`Formation of Active Metabolite (s) .
`for controlled release formlation frequently possess short
`half—lives and may exhibit substantial degrees of first
`
`pass hepatic metabolism.
`In both instances.
`the rapid
`elimination of the oonpound that makes it a candidate for
`
`
`fornulation as a controlled release product may lead to the
`
`

`

`J. Mordenti and R. L. Williams
`
`m3
`
`Fm mam-m; DRUG KW
`
`“M
`
`PHAM’IACOKINETIC FACTORS
`
`RATE OF DELIVERY
`
`ROUI‘E OF DELIVERY (FIRST PASS METAHDLIR'U
`
`FORMATION OF ACTIVE MEI‘AH)LI'I‘E(S)
`
`treatment of angina and related conditions (6) .
`
`rapidly
`'
`dinitrate,
`Isosorbide
`metabolized in nan to the 2- and 5-mononitrate metabolites.
`The clearance of the parent conpound is high, and the drug
`exhibits a half-life of less than 1 hour
`(5).
`The 5-
`mononitrate metabolite possesses substantial vasodilatory
`activity and
`is available in some countries
`for
`the
`
`SATURABLE METAELISW
`
`SATURABLE PROTEIN BINDING
`
`PHARMACGDYNAMIC FACTORS
`
`SENSITIVITY
`
`TOLERANCE
`M
`
`formation of one or more active metabolites. Examples
`include
`isosorbide
`dinitrate,
`nitroglycerin,
`and
`propranolol.
`The presence of an active metabolite can be
`detected using several methods.
`'
`
`concentration response curve with the measurements plotted
`sequentially (4). When this curve exhibits anticlockwise
`hysteresis,
`the presence of an active metabolite may be
`Differences in the rate of input of
`
`

`

`onON0*ONOw
`
`N.eiv.dD9urDesabeRdbmrtnoC
`mm;mmvfimmmmO72ZOHBEZMUZSPHEHZHQWBMMOmOmHflaw/EmzuszmmmHmmZOHBsmm3ENE
`
`
`
`
`
`
`
`
`
`
`SEEScozozcoocoozom.9:85
`
`BJnSSSJd onomfls
`
`bugpuozs u! abuoqg waxed
`
`.35wszon2:gUZHn—mbm923VEgg
`
`
`
`

`

`206
`
`J. Mordenti and R. L. Williams
`
`IMMEDIATE RELEASE
`
`
`
`4—- PARENT
`
`
`45...... METABOLITE
`
`
`
`Z0 E
`
`a;
`
`E
`
`0Z00
`
`TIME
`
`CONTROLLED RELEASE
`
`PARENT
`
`METABOLITE
`
`
`
`TIME
`
`DOSINCS.
`
`Z ,
`
`9;
`' E
`5
`g
`
`ZOU
`
`PARENT
`TO
`METAEDLITE
`FIGURE 5:
`REIATIQ‘ISHIPS AFTER
`IMMEDIATE
`AND
`
`CONCENIRATIG‘I
`DRUG
`CON'IROILED
`RELEASE
`
`

`

`Controlled Release Drug Delivery
`
`207
`
`
`
`in drug
`Nonlinear Pharmacokinetics. Nonlinearity
`disposition, characterized by disproportionality between
`dose of drug reaching the systemic circulation and clear-
`
`In recent studies, we assessed the absorption and
`" disposition of isosorbide dinitrate and the appearance and
`elimination of
`its two primary metabolites
`in healthy
`7; individuals after administration of the drug via several
`1,, routes and in various immediate and controlled release
`_ dosage formulations.
`The data in Table 4 provide ratios
`, of the 2— and 5- mononitrate metabolite
`concentrations
`.1 relative to isosorbide dinitrate at selected times after
`f administration of isosorbide dinitrate.
`Blank spaces in
`: the rows of
`the table indicate decline of
`the parent
`} compound
`(isosorbide dinitrate)
`to limits below assay
`: sensitivity.
`These data from a representative individual
`: document wide variability in metabolite-to—parent ratios
`1 after administration of isosorbide dinitrate via different
`‘ routes and at different rates.
`Intravenous administration
`‘, generally produces higher concentrations of
`isosorbide
`dinitrate initially (lower ratio of metabolite to parent),
`“ but
`the ratio of metabolite to parent compound increases
`rapidly as parent concentrations decline.
`In contrast, the
`high
`first pass metabolism of
`the drug after oral
`administration produces initially high metabolite—to—parent
`ratios.
`This pattern of metabolite to parent also occurs
`with oral controlled release and transdermal delivery
`systems
`(Table
`4).
`For most
`dosage
`formulations,
`isosorbide dinitrate metabolite-to—parent
`ratios become
`extremely high at later times after dosing owing to the
`longer half-lives of the metabolites relative to the parent
`compound. Part of the differences in metabolite to parent
`ratios displayed in Table 4 may also be attributed to
`differences in distribution between isosorbide dinitrate
`and
`its metabolites
`or
`to saturable metabolism of
`isosorbide dinitrate (see below).
`The data in Table 4
`document that the ratio of active metabolites to isosorbide
`dinitrate can vary over a wide range (by factors of 10 or
`more) at specific times after dosing, depending on the
`route and rate of administration. Concentrations of drug
`and metabolite in blood are thought to produce proportional
`concentrations at the receptor site (5) of action (Figure
`2).
`If true for isosorbide dinitrate, then widely varying
`concentration ratios of metabolite to parent drug ratios,
`such as those depicted in Table 4, can exist at isosorbide
`dinitrate receptors at varying times after dosing.
`If so,
`the pharmacologic response to isosorbide dinitrate may also
`vary widely at any given time after dosing if comparisons
`are made between formulations with different rates of drug
`delivery and following different routes of administration.
`
`_
`
`
`
`

`

`208
`
`J. Mordenti and R. L. Williams
`
` TIME IV CRAL
`
`
`-—-_——
`(hr)
`SOLN IR
`CR
`
`ED
`ORAL
`IV
`ID
`———- MM—m
`SOLN IR
`CR
`
`0.50 0.50 2.18 1.35 2.75 --
`2.34 16.0 4.68 6.44 --
`3.01 35.4 13.3 -- —-—
`1.0
`0.75 4.21 3.29 -— --
`2.0 -- 6.67 1.00 3.65 -- -- 63.2 4.65 20.7 -—
`4.0
`-— -- 0.57—-
`-— --- ~— 3.90--
`2.95
`6.0 —— —— l.96-- 0.84 -- -- 25.4—-—-
`4.83
`12.0 -- -—-
`-— -—- 1.64
`-—- —— -—— --
`6.65
`
`W D
`
`AEA.ARE EWJMAA REERESENTATTVE SUBJECT
`IR = IMMEDIAHE RELEASE
`CR = CONTROLLED RELEASE
`TD = TRANSDERMAL DELIVERY
`
`is observed for many drugs. Nonlinearity may be
`ance,
`attributable to saturation of one or more enzyme system
`responsible
`for
`drug metabolism
`(Michaelis-Menton
`kinetics),
`saturation of
`renal
`excretory mechanisms,
`saturation of blood or plasma drug binding sites, or
`saturation of active absorption processes.
`
`
`
`

`

`Controlled Release Drug Delivery
`
`209
`
`produce saturable first pass metabolism because of their
`slower
`rate
`of
`drug
`input
`and
`consequently
`lower
`intrahepatic drug concentration.
`For
`some drugs,
`the
`influence of saturable first pass metabolism and controlled
`release characteristics of a fornulation on bioavailability
`have
`been
`carefully defined.
`At
`low oral doses,
`propranolol AUCs are equivalent
`(7) irrespective of whether
`the drug is given as an innediate or controlled release
`fornulation (Figure 6). As the dose of drug increases, the
`ratio of the dose adjusted AUC of the controlled release
`formulation relative to AUC of
`the immediate
`release
`formulation falls significantly.
`This decline
`occurs
`because
`relatively more
`drug
`escapes
`hepatic bio-
`transformation at the higher rates of drug input associated
`with the Iimnediate release fornulation. At
`the highest
`doses, propranolol ADC ratios for the inmediate release and
`controlled release formulation again approach unity as
`conparable amounts of drug escape hepatic biotransfornation
`irrespective of the rate of administration. Data such as
`that presented in Figure 6 suggest that the pharmcologic
`response to propranolol may vary widely between imnediate
`and controlled release formulations even when they are
`given at conparable doses.
`
`Saturable metabolism has been documented for many
`that
`are
`administered
`in controlled
`release
`drugs
`preparations,
`including
`propranolol,
`nitroglycerin,
`theophylline, and isosorbide dinitrate. For nitroglycerin,
`the influence of saturable first pass metabolism after oral
`administration may be profound. After low oral doses of
`the drug, virtually no parent drug appears in the systemic
`circulcation. As the oral dose is increased, substantial
`amounts appear systemically as more of the drug escapes
`hepatic biotransformation (Table 5). A controlled release
`oral formlation of nitroglycerin
`is
`unlikely
`to
`produce effective nitroglycerin concentrations unless given
`at high doses.
`The same obser-vation my be true for
`isosorbide dinitrate and other drugs that are candidates
`for oral controlled release drug delivery.
`Efficacy
`studies to document the pharmacologic effects of controlled
`release relative to inmediate release formulations have in
`general not been performed. The astute clinician will
`realize that the efficacy of certain drugs when prepared in
`controlled release
`fornulations may
`be
`substantially
`altered.
`
`Saturable Protein Binding. Although all drugs
`2.
`are potentially capable of exhibiting saturable binding to
`blood or plasma binding sites, several drugs
`(naproxen,
`disopyramide)
`exhibit
`saturable binding at concentrations
`
`
`
`

`

`amoaéBEmmoo
`
`
`
`OOv00m00m00.O
`
`gfifigfimfii8m8§m§§§§8¢m89.0B<oHa<mmE6MERE
`
`
`
`
`
`
`.EaSozémomm92fififimms8mafi8385EH.mamas988mm
`
`.\.Oo2(R,mmd\\B0.na/m,.8wmmm.oJ.an=E<Q<mm>m0l_.
`nm552<EoEmo0mn.oW.c...\I.I“.
`
`WS
`
`mnuua«WxIIlUWU!.,u-n,O.s
`
`210
`
`
`
`.$fi?§‘1riir£v¥?i
`
`_.
`
`

`

`
`
`Controlled Release Drug Delivery
`
`
`
`TA ARE FROM
`4-
`’IMENTA
`w-
`'MMB
`,
`- mm
`c
`
`
`
`W A
`
`'IREA'IMENT B mm c
`
`(ng/ml)
`
`(ng/ml)
`
`(ng/ml)
`
`1.09
`
`—-——
`
`—-—-
`-——--
`——--—
`
`—-
`
`--—
`
`0.13
`
`0.07
`
`0.03
`—-—-—
`——-
`
`-———
`
`———
`
`0.33
`
`0.14
`
`0.07
`0.07
`0.04
`
`0.09
`
`-.—
`
`
`
`A 6
`
`REPRESENTATIVE SUBJECT
`.5 MG NI'lROGL-YCERIN SOLUTION
`9.0
`MG NI'IROGLKCERIN SOLUTION
`13.0 MG NITKJGLYCERIN SOLUTION
`
`211
`
`'1
`’3
`[
`
`Eli/r
`
`‘
`
`i
`
`-t occur after administration of standard therapeutic
`uses.
`The
`pharmacokinetic
`and
`pharmacodynamic
`
`.. onsequences for changes in fraction of unbound drug in
`
`golood (fb) or plasma (fp) depend on how rapidly a drug is
`
`leared from the blood.
`For a low clearance drug, an
`
`fncrease in fp or
`fb will
`result
`in a corresponding
`
`'ncrease in plasma clearance and a subseqaent decline in
`
`; otal plasma or blood drug concentration, with no change in
`
`{10 armacologic effect.
`For a highly cleared drug, changes
`
`'1'
`protein binding may be reflected more directly in
`
`'3
`- es in pharmcologic effect, with no change in total
`
`olasma or blood clearance. The general pharmacokinetic and
`
`9;.
`.
`codynamic consequences of changes
`in drug protein
`
`oinding have been reviewed elsewhere (8,9).
`
`
`[drugs
`
` i
`
`The censequences of saturable protein binding for
`that may be prepared in controlled release prepara-
`
`

`

`212
`
`J. Mordenti and R. L. Williams
`
`recently
`We
`study.
`tions has only received inital
`documented the bioequivalence of a generic formlation of
`disopyramide relative to a standard product and solution
`using unbound rather than total drug concentration (10).
`As part of this work, we demontrated that bioequivalence
`assessments
`based
`on total drug concentrations can be a
`relatively insensitive indicator of bioavailability for a
`drug
`that exhibits
`saturable protein binding.
`This
`indifference occurs because total plasma or blood AUC,
`which is used as an indicator of extent of bioavailability,
`can vary as a consequence of differences in rate of drug
`input.
`
`pharmcodynamic
`the
`studied,
`less well
`Although
`consequences of saturable protein binding are likely to be
`as inportant as the pharmcokinetic changes. Consider the
`relationships that might occur between unbound drug and
`total drug concentrations for a drug such as disopyramide
`after administration in an immediate versus a controlled
`release product
`(Figure 7).
`For
`the immediate release
`formulation,
`rapid increments and decline in total drug
`concentration may be associated with corresponding changes
`in unbound drug concentration as saturation of plasma or
`blood binding sites occurs.
`For a controlled release
`product,
`the fraction of unbound drug may not increase to
`the same degree or at
`the same rate, depending on the
`release characteristics of
`the product and the binding
`parameters of the drug.
`In Figure 7, the concentration of
`free drug is represented by the difference between the
`total
`and
`bound
`drug
`concentrations.
`This
`figure
`illustrates the difference in free drug concentrations that
`can
`occur
`between
`immediate
`and
`controlled
`release
`formulations
`for
`a
`drug
`that exhibits
`concentration
`dependent protein binding. Theoretically. the unbound drug
`concentration controls
`the pharmacologic effect of
`the
`drug.
`If changes in the unbound concentration occur for a
`drug that exhibits concentration—dependent protein binding,
`then marked differences in pharmacologic effect between an
`immediate and a controlled release product may occur. For
`disopyramide, we have documented that
`a pharmacologic
`effect of
`the drug correlates well with unbound drug
`concentration. Careful clinical studies will be necessary
`to document the pharmacodynamic consequences of fornulating
`drugs such as disopyramide as a controlled release product.
`
`ALTERATIONS IN PHARMACOLOGIC EFFECT
`
`Thus far, only the pharmacokinetic consequences of
`controlled release drug formlations have been discussed.
`A less well
`studied aspect of controlled release drug
`
`
`
`

`

`
`
`l
`
`l l
`
`Controlled Release Drug Delivery
`
`213
`
`CONCENTRATION
`
`
`
`TIME
`
`CONCENTRATION
`
`

`

`214
`
`J. Mordenti and R. L. Williams
`
`delivery pertains to changes in pharmacologic effect that
`can
`occur with
`a drug given at
`a
`constant versus
`intermittent rate of drug administration. Most of the work
`in this area has been devoted to studies of the development
`of tolerance and dependence to narcotic analgesics.
`In
`general,
`the likelihood of developing tolerance is reduced
`when a drug is administered at low doses and at irregular
`intervals which are longer than the time required for decay
`of
`its pharmacologic effect
`(11) .
`This situation is
`comparable to intermittent dosing of a drug at intervals
`longer than several half- lives.
`In contrast, the terflency
`to develop tolerance is enhanced with larger doses given at
`intervals shorter than the titre required for decay of the
`pharmacologic effect
`(11) .
`Presunably this situation
`occurs with the constant
`rate of drug input
`that
`is
`characteristic
`of
`a
`controlled
`release
`formlation.
`Although data is lacking,
`it
`appears
`that controlled
`release fornulations may thus promote the development of
`tolerance. Many of the drugs now fornulated as controlled
`release
`formlations
`have
`been
`associated with
`the
`development of
`tolerance.
`For exanple,
`the headache
`produced by nitroglycerin may decrease in both frequency
`and severity with continued administration of
`the drug
`(12).
`Clinical efficacy studies may be necessary to
`document the changes in pharmacologic effect that can occur
`with a drug that
`is formlated as a controlled release
`product.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`CONCLUSIONS
`
`
`
`
`
`
`
`the
`offers
`delivery
`drug
`release
`Controlled
`opportunity of delivering drug products more effectively
`
`and with greater safety and convenience.
`For many drugs,
`pharmacologic effects will be conparable,
`irrespective of
`
`the rate of drug release from a formlation.
`For other
`
`drugs, pharHacokinetic and pharmacodynamic factors may act
`
`in such a way as to alter drug effect depending on rate and
`
`route of input. Drugs that exhibit saturable absorption,
`saturable elimination, or saturable protein binding; drugs
`
`that
`form active netabolites; or drugs that demonstrate
`
`alterations in receptor sensitivity with different dosing
`
`patterns are likely to fall
`in this category.
`Careful
`clinical and pharmcokinetic studies will be required to
`
`assess the inportance of changes in drug input rate on the
`clinical pharmacology of inmediate and controlled release
`
`dosage formlations.
`
`

`

`
`
`
`Controlled Release Drug Delivery
`
`215
`
`REFERENCES
`
`Holford, N. , Sheiner, L., Understanding the dose-effect
`relationship: Clincal application of phamacokinetic—
`pharmcodynamic models. Clin. Pharmacokinet.
`_6_, 429
`(1981).
`
`Cbservations on three dosage
`Carruthers, S. George.
`forms of pindolol. Amer. Heart J., 103, 451—455 (1982).
`
`Imhof, P.R., Sieber, A., Hodler, J., Muller, P., Ott,
`B., Frankhauser, P., Chu, L. C., Gerardin, A. Plasma
`concentrations
`and
`haemodynamic
`effects
`of
`nitroglycerin during and after intravenous infusion in
`healthy volunteers. Eur. J. Clin. Pharmcol., _2_3_, 99-
`106 (1982).
`.
`
`Sheiner' Lu Bo
`meaZZi] R. L. I met, Lo Z. ’
`Relationship
`between
`the
`pharmacokinetic
`and
`pharmacodynamics of procainamide.
`Clin. Pharmacol.
`
`Then, 29, 278-289 (1976).
`
`Aronow, W. S. Glyceryl Trinitrate
`Elkayam, U.,
`(Nitroglycerin) Ointment and Isosorbide Dinitrate:
`A
`Review of
`their
`Pharmacological
`Properties
`and
`Therapeutic Use. Drugs, 2;, 165—194 (1982) .
`
`Taylor, T., Chasseaud, L. F., Major, R., Doyle, E.
`IsoSorbide S-Mononitrate Pharmcokinetics in Humans.
`Biomarm. and Drug Dispgition. , 2, 255—263 (1981) .
`
`Pooled Michaelis-Menton
`Propranolol:
`Wagner, J. G.
`of
`input
`rate
`on
`the
`effect
`parameters
`and
`bioavailability. Clin. Pharmacol.
`'Iher., 11, 481—487
`(1985) .
`
`Protein binding and kinetics of drugs
`Blaschke, T. F.
`in liver disease.
`Clin. Pharmcokinet.,
`_2_, 32-44
`(1977) .
`
`William, R. L. Protein Binding in Hepatic Disease:
`Pharmacokinetic and Clinical
`Inplications. M
`Medica Hoechst., Septen'ber 24-28, 1985.
`
`Thibonnier, N., Holford, N., Upton, R. A., Blume, C.
`D.,
`and
`William,
`R.
`L.
`Pharmcokinetic-
`of
`Pharmacodynamic
`analysis
`urbound
`disopyramide
`directly measured in serial plasma samples in man.
`g_._
`Pharmacokinet. Bimini!” 2, 559-573 (1984) .
`
`_ 2.
`
`3.
`
`7" 4.
`
`i
`
`s.
`
` 1.
`
`8.
`
`9.
`
`10.
`
`

`

`216
`
`11.
`
`J. Mordenti and R. L. Williams
`
`Muller] PI] mf’ Po Roy Burkart: Fa] Chu' L. Co,
`Gerardin, A.
`Human pharmacological studies of a new
`transderrral system containing nitroglycerin.
`Eur. J.
`Clin. Pharmacol., g,
`473-480 (1982).
`
`Morrison, R. A., Wiegand, U. W., Jahnchen, E., Hohmann,
`13.,
`Bechtold, H., Meinertz, T.,
`Fung, H.
`I.
`Isosorbide dinitrate kinetics
`and
`dynamics
`after
`intravenous, sublingual,
`and percutaneous dosing in
`
`Seevers, M. H., Deneau, G.A. Physiological aspects of
`tolerance and physical dependence.
`In Physiological
`Pharmacology, A Comprehensive Treatise, Volume 1. W. S.
`Root, F. G. Hofnann, Editors.
`New York: Academic
`Press” 565—640 (1963).
`
`angina. Clin. Pharmacol. Ther., _3_§, 747-756 (1983).
`
`