`
`
`Parmacokineticl
`Pharmacodynamic
`Correlation
`
`Eds'fed b3;
`
`Harmut Derenderf, Ph. D.
`
`Gfinter Hochhaus, Phfl.
`
`Ufiizws'sify 0f Harm
`Gainesmfiea E‘iorfdfl
`
`
`
`CRC Press
`
`BocaRath London Tokyo
`
`InnoPharma Exhibit 1027.0001
`
`

`

`
`
`library of Congress Cataioging-inuPublitatitm Data
`
`Handhook of pimmmcokinticfphzmnaeotlymmie corl'elmioii 1’ edited by Hortmm Derendorl', Giinlhet‘ Hochhotts,
`p.
`cm.
`includes bibliographical references and index.
`lSEN ii~8493—‘§303—X
`l. Pimt‘maeokineties. 2. DrugsiPhysiologienl effect.
`I. Det‘endm’f, Hin‘tnntt.
`ii. Hochhmts, Giintheix
`iDNLM: L Pharnmeokinetics.
`2., Pharmacology. 3. Doxe~Response Relationship Drug,
`(2%" 38 H2365 1995}
`RM30i5.l-i36
`1995
`615’.7"—-{le2{)
`DNLMKDLC
`for Libraiy of Congress
`
`«‘1. Models, Theoretical.
`
`9422365
`(TI?
`
`This book contains information obtained from authentic and highly regarded sources. Reprinted material is quoted with
`petmission,m1d sources are indicated. A wide variety of references are listed. Reasonable efforts have been made to publish
`reliable data and infommtion, but the author and ihe publisher cannot assume responsibility for the validin of all materials
`or for the consequences of their use.
`Neither this; book not any part may be reproduced or transmitted in any form or by any means, electronic or mechanical,
`including photocopying miei'ofiiming‘ and recording, or by any information storage or retrier syxtem. without prior
`permission in writing from the publisher.
`All rights reserved. Authorization to photocopy items for internal or persotiai use, or the pemonal or internal use of
`specific clients. may be granted by CRC Press. Erie, provided that $30 per page photocopied is paid directly to Copyright
`Ciearance Centctl 27 Congi‘esss Street, Salem‘ MA 0 [MO USA. The fee code For users of the ’i‘mnsactiomil Reporting Service
`is [SEN 0~84§3—8303«XX?5X$0.00+$.50. The fee is subject to Change without notice. For orgzniizaiions that have been granted
`a photocopy license by the (ICC, a separate syxiem of payment has been arranged.
`CRC Presst Inch: cement does not extend to copying for general distribution. for promotion. for creating new works,
`or for resale, Specific permission must be obtained in writing from CRC Pl'cfib for such copying.
`Direct all inquiries; to CRC Press, lnc.‘ 200i} Corporate Bli‘ti.. NW“ 80m Rziton. Eitn‘idn 3343i.
`
`© 19953 by CRC i’t'ess. Inc.
`
`'
`
`No claim to original 1,3,3, Government works
`International Standard Book Number i}»l§493-$3()3—X
`Library of Congress Cart! Number 94—22355
`Printed in iitC United States of America 1
`2 3 ii
`Printed on acid-free paper
`
`3 6 Ti 8 {J 0
`
`InnoPharma Exhibit 1027.0002
`
`

`

`n—mw ‘
`
`Table of Content$
`
`
`Chapter i
`Pharmacokinetic~Pharmacodymimic Modeling of Reversible Drug Effects ............... ..
`Jiirgen Vanity,
`
`................ ..I
`
`Chapter 2
`PharmacokincticFilmmacodynamit; Modeling of Irma-'ersil'fle Drug Eff‘ccls
`Steven C. Ebert
`
`Chapter 3
`Pharmzwcakinetic—Pharmacociytmmic Modeling in Dyng Development:
`Comments and Applications ........................................................................................ ..
`Joseph C. Fleishakflr and James .1. Ferry
`
`............... .5?
`
`Chapter 4
`Dose Optimizatian Based on thu‘macokinetic-Pharmacodynamic Modeling
`Giimher Hucitlmus and Hartmut Deremlerf
`
`’39
`
`Cimpte‘ 5
`Pharmacekinetic-Pharmacadynamic Correlations of Anemhciic
`Virgina D. Scilmith and Keith ’I‘. Muir
`
`
`
`Chapter 6
`g,
`P11311112 (20E;inetEC«Phat:nacodynamic Correlations of Anatgcsics ...................................
`
`......... .. 14%
`
`Judith S.‘Walker
`
`Chapte‘ 7‘
`Pharmz cokEnetEC~Pharmamdyzmmit: Correlations of BERHXUEMENRCS
`Samir K. Gupta and Everett H. Ewan-‘00:!
`
`(Shame: ‘ 8
`PbammwkineficfPimrmacmiy:mmic Correlations of Anticmvulsalm ....................................... .. 185
`Meindert Dankwf and Rub A. Vimkufl
`
`Chapter 9
`Pharmacokincrir:—Pharmacodynzm'uEC Relationahipr; of Cardiovasmtim‘ Drugs; ........................... .. 192?
`Richard L. Lalonde
`
`InnoPharma Exhibit 1027.0003
`
`

`

`Chaptm‘ 10
`PharmacokinetiCXPharmacodwmnic Correlations of Selected Vasodilators ............................. “227
`Tsang—Bin 'l‘zeng and I‘Io-Leung Fang
`
`Chapter 11
`Phazmamdynamics of Anticoagulants ....................................................................................... .241
`Dennis Mungall and Richard H. White
`
`Chapter £2
`PharmacokinetiC~Pharmacodynamic Correlatiom 0f Amihixmmines ........................................ .275
`Eric Smack, Achie! Van Peer, and .103 Heykams
`
`Chapter 13
`{SJ—Agonists: Tm‘buialine, Albuiemh and Peneteml .................................................................. “299
`Giinfher Hacithaus and Helmut Mi’illnmml
`
`Chapter 14
`PharmacokimeEicmPharmacodynamic Correlations 0f Corticosteroids ........................................ .323
`Hehnut Malimann, Stefan Balhach, Giinther Hochhaus,
`Jiit‘gen Barth, and Hartmut Derendm‘f
`
`Chapter 15
`Pharmacokineticipharmacodynamit Modeling 0f Antibiotics;
`Arne Nolting and Hartmuf Derendorf
`
`Chapter 16
`Clinical Pharmacodynamics of Anticancer Drugs ..................................................................... “389
`Howard L. McLeod and Winiam E. Evans
`
`Chapter 1?
`Computer Applications in Clinical Phaumicokinetics and Pharmacodynamics ....................... “415
`Dennis Mungali, Joe Heissler, and Matthieu Kaitenbach
`
`{Mex ............................................................................................................................................ . . 465
`
`InnoPharma Exhibit 1027.0004
`
`

`

`
`
` This material may be protected by Copyright law (Title 17 U.S. Code)
`
`Chapter 2
`
`i’HARMACOKINETIC~PHARMACODYNAMIC MODELING
`OF IRREVERSIBLE DRUG EFFECTS
`
`Steven C. Ebert
`
`INTRODUCTION
`
`The mathematical relationship hetween drug concentration and response generally assumes drug
`binding to a given population of receptors. Greater occupation of these receptors will result in a
`greater effect, and vice vet'sa, The variability in receptor affinity accounts for the nonlinear relation—
`ship between concentration and effect.
`Most research involved in the characterization of phai‘rnacodynamic variables has been ( one with
`drugs exhibiting reversible effects. With reversible drug effects, the number and drug a il‘inity ol'
`receptors remain relatively constant, allowi g reproducible effects with repeated drug exposures. In
`certain instances, sensitization or tolerance to the drug effect may occur,
`in contrast, the goal oi‘ antimicrobial therapy is to eliminate the very target at which tl e drug is
`directed,
`i.e., the “receptors” are actually he pathogenic organisms. The desired ei’t‘ee of drug
`therapy,
`the death of pathogens,
`is there ore irreversible. The replication oi" organisms, with
`subsequent replenishment of the “receptor” anther may occur, however, and result in an a maternity
`reversible effect as measnred by the total number of organisms. In this system, the development of
`tolerance or frank resistance oi‘ the remain'ng “receptors” is more likely to'oeenr, since the most
`susceptihle organisms are killed. Therefore, it would he expected that sustained expos ire to an
`antimicrobial would result in it gradually di ninishing effect over time, unless a sufficient drug—free
`period exists to enable the organism population to fully recover its susceptibility.
`Figure 1 shows the relationship between harmacoltinetic variables, pharmacodynninie neasures
`of antimicrobial activity, and drug effect. lntuitively, increasing the dose of an antimicrobial should
`increase its el‘l'eet. B}; understanding the nharinaeodynatnie features of a given class of a tiimicr‘o»
`hials, one may, however, be able to i‘urther enhance outcome by modifying dosing paratnte ers other
`than dose size, e.g., dosing l‘i‘equenev. The pliarmaeodynamic properties of antittiierohials may be
`estimated at little, whereas for other drug classes, response must be estimated in rive. This chapter
`will address the pharmacokinetic and pharmaeodynznnic properties of antimicrobials when little»
`once drug response, will characterize tirttg eli’eets, anti will then examine techniques that have been
`used to link these phenomena.
`
`
`
`
`
`PHARl‘r’lACOKINE't‘iC VARIABLES
`
`The pharmaeekinetic variables of a (lrtig determine the time course of drug concentration in
`serum and, ultimately, at the site of infection. ’t‘wo different categories of pharmacoltinetie variables
`exis:. Some phnrmacokinetie variables may be altered by adjusting the (losing regimen, whereas
`others resttlt from the chemical properties of a particular drug, and are minimally inl‘iuencetl by
`dosing regimen. in the clinical setting, modification of regirnen—dependent variables is the goal of
`pharniaeokinetic therapeutic drug monitoring, in order to achieve the desired concentration pro
`file?“
`
`REGIMEN-DEI’ENDENT VARIABLES
`
`Most aiitimicrohials are administered intermittently, i.e., a given total daily dose is divided and
`the fractions given at fixed intervals. The three phat‘mncokinetie variables that may he used to
`
`,
`(LM‘JK-Xfit
`
`mg by t‘RC) Pit-st luv.
`
`InnoPharma Exhibit 1027.0005
`
`35
`
`

`

`36
`
`Handbook of Pelleti'iiracokinefin/’33}:rri'iiiacorlwiciiiiit Correlation
`
`
`Pharmacokinctic Parametch
`
`D038 RegimemRelnted:
`' Area under the: concentratiomime curve (AUC)
`’ Peak concentration
`a Duration of time concentratiotm exceed in critical value
`
`Other:
`' Extent ot" binding to proteins;
`’ Extent of delivery to less accessible tissue sites
`- intracellular penetration
`
`Ding Concentration at
`the Site of Infection
`
`
`
`Pharinnggglynnmic Parameters
`’ Minimum inltibitoq‘i’cidnl ccncentmtinn
`(MlCIMBC)
`' Relationship between conccntantion and
`cirlnl activity
`* Postantibictic effect {FAB}
`- Emergence ()f wsistmtcc
`a mencefabsencc of 355803, other host
`factors
`
`Ovemn Amimicmbia! Adm“).
`of Dmg Regina!
`
`
`
`‘
`Primary:
`‘ RE‘lUflloll 1“ m"le 0f Palllllgcmc
`organisms:
`
`Secondary:
`e Reduction in morbidityfmottality
`
`FIGURE It Relationship between plmrrrmcukinetir: parmnctcm, pharmamdynumic parameters, and drug eit‘cctsfoutccmc
`for antiniicmbials.
`
`describe the tinic course of antimicrobial activity are thc 24—h cumulative area under tltc concentration-
`timc curve (AUC), pcnk conccntrrttion, and duration of timc above a particular concentration
`(T > (I).
`
`AUC
`
`The ADC is. a function ol~ the total daily dose of antibiotic and drug clearance. With some
`cxccptions (cg, certain beta-lactams),
`the AUC is not influenced by tin: length of the closing
`interval. in ctltcr words, administration of a given daily dos 1, as a singlc done or as smaller fractions
`at fixed intervals should yield the same 24—h AUC. The AUC best reflects overall drug exposure.
`
`Peak
`
`The peak scrum concentration is largely determined by tlrc magnitude (if any given individual
`dose. Single administration of a daily dose of antibiotic will yield the highest peak concentration;
`divided doses will yield smaller peak concentrations. Peak scrum concentration may rcflcct maxi-
`mum drug effect nnrlfor a “driving force" For cxtrnvnscular drug delivery.S
`
`T > C
`
`A third rcgimcnfldcpcmicnt variable is; the duration of time over which conccritt'ntimis exceed 21
`certain value {T > C). Astrnnting that (in: mean Stcanfiynrtatc concentration exceeds rim target
`concentration, dividing a given daily dose into many fractions ardministcrccl at short intervals will
`increase “I” > C. The ultimate extension of this; would be continuous infusion“ Administration of
`
`larger fractions; of this daily rinse at longer intervals will tlrcrct‘orc dccrcasc T > C. T > C may be
`important for characterizing the length of effect for plicnorncna requiring a tlncnhold concentration
`for rcsponsn.
`A summary of the impact 01‘ rinsing regimen modification (daily close, interval) on the rcgimcw
`dependent variables is shown in Tnblc l. Thcnc variables an: usually lntcrmrclntcd. An increase in
`the total daily dose: of drug without changing the intch will increase AUC, peak, and T > C,
`making it impnnsiblc to determine which of the three variables is mmt important in increasing {ling
`cl‘l'cct. By modifying the cloning interval as Well, lllBSC variables may be altered ii‘idcpnnrlcntly, As;
`will be discussed later, maximizing drug cxgvrtntn‘n through rlillcrcnt variables; inay bc important l‘or
`
`InnoPharma Exhibit 1027.0006
`
`

`

`Phili’itlt’lc‘okl!iKile”Philf‘fttétc‘t’klyttClinic Mort'eiiug cg’lrrttnei‘sioie Qi‘itg figfii‘kets
`
`3’?
`
`TABLE 1
`
`Impact of Dust: Regimen Modification
`on Pharmacokinctic Parameters
`
`increase daily dose
`‘
`
`Same daily dose
`
`Decrease
`interval
`
`nuc: T
`Pouk: -
`T>C:ll
`
`AUC: -
`
`Peak: ti»
`“not
`
`AUG. t
`
`Decrease daily dose
`
`Peak: ll
`T>C:~or’l‘
`
`Same
`interval
`
`Increase
`interval
`
`AUC; T
`Peak i
`T>Czi
`
`Auo: t
`Wait Tl
`T>C1~orl
`
`AUC: —
`
`ABC: m
`
`Peak: «
`'l‘>C:-
`
`AUC: t
`
`Peak: 1»
`T>C:~l»
`
`Peak: l
`T>c:l
`
`ABC: t
`
`Peak: —
`T>C:l~l;
`
`Note:
`
`Impact of dose modification of antibiotics {total daily dose, dose
`interval) on phnrniaeokitmtio variables (AUG. peak T > C): T,
`in—
`crease; —~. little or no change: l, decrease. The table assumes first-order
`drug elimination and modest accumulation with multiple dosing.
`
`if maximining the 24—h AUC is the single most
`different elnsses of antibiotics. For example,
`important method to enhance efficacy of a certain antibiotic, increasing or decreasing the dosing
`interval for it given daily dose of drug should not influence effect. if peak concentration is: most
`important, lengthening the dosing interval and administering larget infrequent doses would be best.
`Finally, if T > C determines drug effect, smaller doses administered frequently would be the most
`efficient means of drug administration.
`
`REGIMEN—INDEPENDENT VARIABLES
`
`Certain other pharmoookinetio properties of antiiniorobials also influence the concentration time
`course at the site of infection. Howeyen because they cannot be appreciably altered by modifying
`the {losing regiment they play only a minor role in pharmacokineticfpharmacodynan‘tie modeling.
`
`Set-om Protein Binding
`Most ontimicrobials exhibit some degree of binding to serum proteins. Thin binding is reversible
`in nature and primarily concerns hydrophobic and hydrophiiic binding mechanisms? Most antimi-
`crobials bind to albumin, although some (primarily basic compounds) may bind to alpha—l-acitl
`glycoprotoin. In general, the extent of protein binding within a class of compounds increases with
`increasing molecular weight and hydrophobioity. Protein binding in serum is not generally consid-
`ered to significantly affect the phormucokinetic profile of a particular drug unless the amount ol‘drug
`that is bound in serum exceeds 80%.?
`Protein binding decreases the activity of antimicrobials in serum. Only the free. unbound portion
`of the drug in serum is active against bacteria.“ For example, if a drug were 87.5% bound to serum
`proteins, tho concentration of drug in serum required to inhibit growth would he expected to be
`cightfold higher than that of orotoin-fi‘ce fluid. This phenomenon generally holds true for most
`untitniorobials.
`in some instances,
`the presence of other endogenous substances may enhance
`antimicrobial effect so that the activity in serum is greater than expected." in general, however, one
`should assume that, for highly bound {>80%) antiinierobials, total serum drug concentrations will
`ottoi‘ostiittttte antimicrobial activity.
`Protein binding will impair passage ol’antii‘nicrobinls into oxtravasoular fluid, which is often the site
`of bacterial
`infection. Because most capillary pores do not permit passage of serum albumin into
`oxtravasculur sites, only the frees unbound component of drug in serum diffuses extrayasculztrly.
`
`InnoPharma Exhibit 1027.0007
`
`

`

`38
`
`Handbook of Pharmaco/tirtetir‘r’l’hmrrracodynarnie Correirttt'ort
`
`Consequently, the unbound serum drug concentration appears to he a more conservative (and accurate)
`measure of the drug concentration at sites of infection that are reached by passive diffusionm-‘i
`
`Tissuefii‘luid Penetration
`
`In other eascn, antimicrobial deliver}r into less accensible sites is required. These include penetra»
`tion into the ccrebroepinal fluid, eye, and prostate gland. In addition, penetration of some antiini~
`crobials into intracellular sites is important in treating pathogens that multiply within mammalian
`cells. All of these sites demonstrate a lipophilic barrier which restricts passive diffusion of many
`drugs.'2 *l‘herei‘ore, drugs that are more liponhilie are more likely to be able to enter these restricted
`sites by diffusion across this; lipid barrier. Parndoxieally, many ot‘these same drugs are highly bound
`to serum albumin Another mechanism by which drugs may enter intracellular fluid is through ion
`“trapping”.l3 The quinolones are a good example of this. At physiologic pH, the quinolones are un~
`ionized and more lipid soluble. Once hey enter the more acid intracellular fluid, they become
`ionized and are less able to diffuse out of cells.
`
`Serum protein binding and the extent of delivery of drugs to extravascttlnr Sites are important
`plmrmacokinetic properties in determining the active concentration oi" drug at the infected site. In
`get eral, these properties are drugepec’lia and not appreciably al ered by modifying the dosii g
`regimen. One possible exception concerts drugs that may exhibit aaturahle protein binding. Admin—
`iatering these drugs; at target infrequent doses will yield high peak concentrations, which may rest it
`in
`ransiently lower protein binding, hereby enhancing the extort andx‘or rate of extravascular
`delivery." Whether this practico in fact leads to greater drug et’fictey is, however, unproven.
`In summary, pharmacokinetic variables determine the time cou
`of concentrations in new "it
`ant, ultimately, the site 01’ infection. l igh protein binding and limited extt‘nvaseular delivery ol‘
`drugs; may mandate that serum concentrations greatly in excess of ti e desired trite concentration he
`act loved. By adjusting the dosing frequency, one may achieve either a serum concentration profi c
`with high peak concentrations and low minimum (trough) concentrations, or one with relatively
`coistttnt concentrations over time. Which regimen is prei‘erab C will be influenced by tl e
`pharrnaeodynaniit; propertiett of the agent to be used.
`
`
`
`
`
`
`
`
`
`PHARMACODYNAMIC PARAMETERS
`
`Pharinacodynamic parameters of antimicrobials re ate drug concentration and the desired effect.
`in most cases, these nhnrrnaco yntttnic parameters are determined in nine, and then extrapolated to
`the in vivo situation. The reason for this is that repeated incaaurcs ol‘ the typical endpoint for
`antimicrobial et’l‘cet, the number of pathogenic; organisms,
`is difficult to perform in vivo (luring
`therapy. The ittaiority oi“ these aharmacotiynan ic parameters may be lit to a standard sigmoid Bum
`model. B}; examining the relationship between concentration and effect, one can design dosing
`regimens that will optimize overall response wl ile uni 1g the minimum amount ol’dtug. This section
`will discuss various pharmacot ynaznic naramc crs, their relationship to concentrations achieved in
`hive} and differences between observation}: in ain't? a rd in Vii-’0.”
`
`
`
`
`
`
`
`The minimum nntirnierobia concentration MAC), minimum inhibitory concentration (MIC),
`and the minimum bactericidal concentration {MBC} are estimates of" the intrinsic activity of an
`antibiotic against a particular organism. These partnne ers may be quantified by incubating a known
`inoculum of bacteria in the appropriate meditt n with various concentrations; of antihiotic added.
`Broth is the medium used most commonly 'or testing, although agar may be used for MIC
`determination. A schematic diagram relating drt gconeentration to the numberoi‘hneteria remaining
`in the medium at 24 h is shown in Figure
`
`MIC
`
`zero not change in the number
`The MIC it; the minimum concentration that prevents growth,
`Glorgttttisms over time. The MIC is; the most commonly used measure of antibiotic activity. it may
`
`InnoPharma Exhibit 1027.0008
`
`

`

`Pttrn‘ntamr’rraeirc-Pttrtt‘rrtrrc‘or{ttnonn'c Modeling Qflrrettersihtc {Drag Effects
`
`39
`
`2
`
`0
`
`6)
`
`ChangeinLogCFU@24hrinii;
`
`I
`
`Log Concentration
`
`FIGURE 2. Relationship between antibiotic concentration and the number oforganistns remaining after 24 it of incubation.
`MAC, minimum antimicrobial concentration, the minimum concentration resulting in any reduction in organism growth;
`MIC. minimum inhibitory concentration. the concentration preventing, any net growth of organisms; and MBC, minimum
`bactericidal concentration, the minimum concentration resulting in a reduction in the number of organisms by 99.9%?
`
`be estimated by counting actual numbers of organisms over time, at; is shown in Figure 2, or by
`visually inspecting the media for changes in optical density caused by growth. In general, the lower
`the MIC, the greater the susceptibility of the bacteria or, conversely, the greater the activity of a
`particular antibiotic. When comparing Mle of various; antibiotics against a particular bacterial
`strain, it should he noted that MIC values must be assessed relative to the achieyable antibiotic
`concentrations in vino
`
`Bach antimicrobial has an established MIC “breakpoint”, This breakpoint is used to classify
`susceptible ys. resistant isolates. Bacterial strains with an MIC in excess of the breakpoint are
`considered “renietant” to the antibiotic, whereas those with bdle less than the breakpoint are
`
`“susceptible”.
`Determination of MIC breakpoints is a complicated and highly subjective process. Hypt)ll‘tt§ll~
`cally, thebreakpoint should be determined based on the clinical efficacy of antimicrobials; for
`example, it the majority of infections due to organisms with an MIC of lo mgr’l or greater to a
`particular antibiotic Fail to respond to therapy with that drug and infections caused by organisms with
`Mle of 4 mgt‘l or less respond favorably, a breakpoint of 8 mg?! would be appropriate. Unfortu~
`natelyt breakpoints are rarely determined based on clinical success. Most commonly, breakpoints are
`determined based on achievable serum concentrations in humans following standard doses, with
`little or no correlation to efficacy. Future efforts will he directed at determining the breakpoint based
`on in vino efficacy and its correlation to one or more pltarmaookineticmttharmacodynamic parameters
`(see later in chapter).‘3
`
`MBC
`Because the MIC is only an estimate of growth inhibition, the use of this measure to predict
`efficacy in viva relies on the host‘s immune system to assist in the eradication ol" the pathogen. The
`MBC is the minimum antibiotic concentration that kills 99.9% of the original number of bacteria,
`i.c., the number of bacteria is reduced by 3 log it) units. This parameter reflects the ability of an
`antibiotic not only to inhibit growth, but substantially reduce the number of bacteria. MBC deter-
`mination is; a more labor-intensive process which requires direct pathogen quantification. It is used
`to assess antimicrobial activity in clinical situations where host immune factors are less helpful in
`
`InnoPharma Exhibit 1027.0009
`
`

`

`40
`
`Handbook of Pizarmocokinct{CXPr’tarrnacorfynonrit: Correlation
`
`helping to eradicate pathogens, such as endocarditis, ostcomyclitis, meningitis, and infections in
`neutropenic patients.
`For some drugs such as the beta-lactam antibiotics, aminoglycosides, and quinolones, the MBC
`is often similar in magnitude to the MIC. These agents are often referred to as “bactericidal”
`antibiotics. As such, the MIC is frequently used as an estimate of the MBC for these drugs. Some
`bacteria may, however, be deficient in the important biochemical processes required for a bacteri-
`cidal effect to occur. In these situations, the NBC may far exceed the MIC. These organisms are
`often referred to as “tolerant”, and other treatment options, such as combination antimicrobial
`therapy, may be required to produce a bactericidal effect.
`Other antimicrobial classes, such as the inacrolidcs, tetracyclines, and chloramphcnicol, do not
`reliably produce a bactericidal effect at concentrations close to the MIC. These agents are termed
`“bacteriostatie” drugs, and MBC testing is not usually done. Consequently, these agents are typically
`avoided when treating infections where bactericidal activity is required. An exception to this is the
`use of chloramphenicol
`for meningitis
`‘ausetl by susceptible Streptococcus paemnottine or
`Hoemophfttts itgfhtenzoe, where a bactericidal effect is observed.
`The MIC and MBC are time-honored measures of antimicrobial activity against pathogens. How-
`ever, these measures are limited as to the information they provide. First, MIC and MBC determination
`imos are made after a fixed incubation period, and therefore only reflect a specific time point: they
`do not prt‘iyide information on the time course of antimicrobial activity. Second, bdle and MBCs are
`treasured using a standard inoculuni of bacteria, usually 10* to 10“ organisms. Infectious diseases
`commonly inyolve much greater numbers of bacteria than are used in susceptibility testing. This higher
`‘noculum of organisms may require antimicrobial concentrations in excess of the nominal MIC to
`‘nhibit growth. Third, MIC and MBC testing are done on bacteria in rapid, logarithmic phase growth.
`it the in nine situation, bacteria causing infection at certain sites (cerebrospinal fluid, cardiac yegeta»
`ions, abscesses, intracellular environment) may be growing much more slowly, if at, all, due to lack
`of nutrients andfor oxygen. The beta—lacuna antibiotics exert a markedly reduced activity against
`slowly growing or nongrosying bacterial‘i-‘i’ Consequently, much higher concentrations may be
`equired to exert an inhibitory or bactericidal effect in him. Fourth, the antibiotic concentrations used
`o measure MICs and MBCs remain constant throughout the incubation period. In the in nine setting,
`a: tibiotic concentrations often vary widely due to drug elimination, which may result in diminished
`effect andror regrowth. Fifth, MlCr’MBC determinations are done using,
`twofold drug dilutions,
`al owing the possibility of as great as a twofold error in precise estimation, Finally, MICKMBC
`determinations do not account for the presence of host factors such as white blood cells, complement,
`a: d antibodies. These host factors may enhance antimicrobial activity in vivo.
`in summary, while MICS and MBCS provide useful information regarding the intrinsic activity
`of antimicrobials against pathogens, the information is inadequate for designing dosing regimens
`ai tied at ottirnizing drug effect in nine.
`
`
`
`
`
`AC
`
`
`
`It is co monly assumed that, given the nominal MIC of an antibiotic against a pathogen, drug
`concentration below the MIC will allow unimpeded organism growth. This is in fact often not true.
`Tie MAC '3 the smallest concentration found in vitro to exhibit any influence on the rate of growth
`ol‘ bacteria when compared to control cultures with no antibiotic. The MAC may be many times
`lower than the MlC, and may be observed with beta-lactams, aminoglycosirles, quinolones, and
`other antib'otics.m Subinhibitory concentrations of antibiotics will not only slow growth of organ-
`isms, but may decrease adherence to membranes and increase phagocytosis of organisms by white
`blood cells.” In addition, antibiotic concentrations above the MAC may help to delay regrowth of
`bacteria tlu *ing periods in which serum concentrations drop below the MIC. This may lengthen the
`postamibio ic effect {see below)” The precise role of the MAC as a pharmacotlynt‘tmic variable has
`not been defined, but it likely contributes to enhanced eradication ol‘patltogens in immunocompetent
`hosts, own when serum antibiotic concentrations exceed the MIC for only brief periods of timc.‘it
`
`InnoPharma Exhibit 1027.0010
`
`

`

`Pills:rnmcnkinen’cwPf;mmnmrfynnnniC illocieiing offr'rnveizrihle Drug Ejfiecb‘
`
`41
`
`
`
`BaclericidalRene(claim)
`
`Cnnnentration
`
`Thetapautlc Range
`— Drug A “Drug 8
`
`FIGURE 3. Relationship between bactericidal ran: (BR) of two nntihiotlcsl A and B, and concentration. The BR for each
`drug increases initially, then becomes fixer! 211 highor concentrations. For {lrug A, the BIZ-concentration rnlntionship is linear
`throughout the range of concentrations obserth in rim. In contrast, tho relationship is primarin nonlinear {fixed} for drug
`8 because much higher concentrations an: achieved in viva compared with the point of saturation.
`
`BACTERICIDAL RATE
`
`As wan stated previously, the MK: and MBC do not provide information on the time course of
`antimicrobial activity. For example, we know that a not bactericidal effect occurs at concentrations
`at or above the MBC, but not whether the rate of such a bactericidal called is further enhanced at
`concentrations; above the MBC. It is likely that, over a specific range, increased concentrations; will
`l‘CSllll in a greater bactericidal rate (BR). This phenomenon is; termed cancanIrwin:e-deyemlenr 13R.
`As conccnlt‘niions an: increased further, the BR becomes; maximal and is not influenced by higher
`
`conmnlrntions, At this point cancannation-i:m’€;;tmt(fenf ER is observed. Clinically, whether a drug
`exhibits concotitration—dependent or —in€lt:pondonl BR depends on the relationship of the concentra-
`tion~BR curve to cnnccntt’ntions observed in vino. Figure 3 show; the relationship between BR and
`concentration for two nntimicrobials, A and B. Both exhibit connontration-dependenl BR at low
`concentrations and concentrationindependent BR at high concentrations. In the clinical setting,
`achiovzthlc concentrations of drug A are restricted to the: lower end of the concentration range (solid
`lino}. Thorclore‘ drug A would Clinically he considered to exhibit Gonnantrntion—dependent BR. In
`contrast‘ achievable concentrations; of (hug B an: much higher, well into the satumbln BR range.
`Drug B would theml‘ore he considered clinically to exhibit conccnlrntintnindopondnni BR.
`Amimicrohials that inhibit call'wnll synthesis, such as the beta—lactams (penicillins, cephalospor—
`ins) and glycopnntidcs (varicomycin, toicoplanin), are examples of drugs that clinically exhibit
`conceittintinmindopcndent BKWEI At concentrations greater than four times the MEG, BR is only
`minimally increased. These higher concentrations are easily achieved in lawn.
`In contrant antibacterials, such as the lluornquinolones, aminogiycosidcsi and mnh‘onidnzolo, are
`drugs that exhibit concerttratim-dependonl BR203E BR continues to incroasn with concentrations up
`to 16-32 times the MBC. The relatively low concentrations; of these drugs; achieved in viva in
`relation to their MBCS results in their activity approximating the lower end of the concentration»
`effect curve, where a linear relationship hotwcen concentration and BR existst
`The effect being discussed here, BR, is in l'act tho slope of tin: curve describing the number
`ol‘surviving bacteria vs.
`time. The endpoint for therapy, the resultant number of bacteria: may
`be estimated by calculating the integral of the slnpn of this curve. For a drug with
`
`InnoPharma Exhibit 1027.0011
`
`

`

`42
`
`ifatto’imok of Pleat?nacolrinetio/Phot‘ntrtcorfivnomEC Correlation
`
`concentratiotnindependent BR, the BR {slope} will be essentially a constant, K. The integral of this
`will then he the constant times time, K X t. Therefore, the overall antimicrobial effect for a drug
`exhibiting concentrationvimi’epetirtent bactericidal activity will he proportionate to the duration of
`time over which the effect is observed,
`
`In contrast, for a drug with concentrationwdependent BR, the slope (BR) will be related to the
`concentration, C. Integrating this equation will yield C X t, which is, in fact, the area under the
`concentration—time curve {AUC}. Therefore, the overall antimicrobial effect for a drug exhibiting
`concentratitin-dependent bactericidal activity will reflect the drug AUG. The application of these
`principles to designing dosing regimens will he discussed later.
`
`POSTAN'I‘IBID'I‘IC EFFECT
`
`
`
`
`
`
`
`The pharinacmlvnamic parameters discussed so far have described the fate of bacteria for the time
`during which they are exposed to an antimicrobial. Concentraticit-response relationships for revers-
`ihle effects imply that when i
`0 drug is present, zero effect should he observed. In this case, that
`would suggest that in the ahse
`ice of antibiotic, unimpeded bacterial growth should occur. This led
`many antimicrobial pharmaco
`ogists to conclude that continuous exposure of pathogens to active
`concentrations of antimicrobial is necessary to achieve maximal net effect?2
`in the lQElOs, investigators working, with staphylococci noted that the organisms did not imme-
`diate v resume growth alter tr nsient exposure to pt—micillin.23 The implications of this finding were
`not p irsued further until the 19?0s, when McDonald