Guidance for Industry

Bioavailability and Bioequivalence

Studies for Orally Administered
Drug Products - General
Considerations

U.S. Department of Health and Human Services

Food and Drug Administration
Center for Drug Evaluation and Research (CDER)
October 2000
BP
 .
I

‘. .

G u i d a n c e fo r In d u stry
B ioavailability a n d B i o e q u iva lence
S tu d ies fo r O rally A d m inistered
D r u g P r o d u c ts - G e n e ral
C o n sid e ratio n s

Additional copies a r e available from the.

D r u g Information B r a n c h (HFD-210),
Centerfor D r u g Evaluation a n d R e s e a r c h (CDER),
5 6 0 0 Fishers L a n e , Rockville, M D 2 0 8 5 7 , (Tel) 3 0 1 - 8 2 7 - 4 5 7 3

Internet at http://www.fda.gov/cder/guidance/index.htm

U .S . D e p a r tm e n t o f Health a n d H u m a n Services
F o o d a n d D r u g A d m inistration
C e n ter for D r u g E v a l u a tio n a n d R e s e a r c h( C D E R )
O c to b e r 2 0 0 0
.

Table of Contents

L fNTRODUCTfON . .. . . . . .. . . . .. . .. . .. . . . . .. . . . . . .. .. . . . . .. .. . . . . .. . . . . . . . . .. . . .. . . . . . .. . .. . . . ... . . . .. . . . . .. . . . .. . . . . . .. . .. . . . . . . . . . .. . . . . .. . . . . . .. . . . .. . . . .. . . . . .. . . . . .. . .. . . . .. . . . 1

IL BACKGROUND.. 1
..................................................................................................................................................................
A. 2
GENERAL ...........................................................................................................................................................................
B. BIOAVAILABILITY ........................................................................................................................................................... 3
C. BIOEQUIVALENCE 4
............................................................................................................................................................
. METHODS TO DOCUMENT BA AND BE ......................................................................................................................
6
A. PHARMACOKINETIC STUDIES ........................................................................................................................................ 6
B. PHARMACODYNAMIC STUDIES ................................................................................................................................... lo
C. COMPARATIVE CLINICAL STUDIES ............................................................................................................................ lo
D. 10
IN VITRO STUDIES .........................................................................................................................................................
Iv. COMPARISON 11
OF BA MEASURES IN BE STUDIES ................................................................................................

V. DOCUMENTATION OF BA AND BE ............................................................................................................................

11
A. 12
SOLUTIONS ......................................................................................................................................................................
B. 12
SUSPENSIONS..................................................................................................................................................................
C. IMMEDIATE-RELEASE PRODUCTS CAPSULES AND TABLETS 12
...............................................................................
D. MODIFIED-RELEASEPRODUCTS ................................................................................................................................. 14
E MISCELLANEOUS DOSAGE FORMS .............................................................................................................................. 17
VL 17
SPECIAL TOPICS ..............................................................................................................................................................
A. 17
FOOD-EFFECT STUDIES .................................................................................................................................................
B. MOIETIESTOBEMEASURED ...... . ............................................................................................................................... 18
C. LONG HALF-LIFE DRUGS...............................................................................................................................................19
D. 20
FIRSTPOINTCMAX .......................................................................................................................................................
E ORALLY ADMINISTERED DRUGS INTENDED FOR LoCAL ACTION .......................................................................20
F. NARROW THERAPEUTIC RANGE DRUGS.................................................................................................................... 20

APPENDM 1: List of Guidances That Will Be Replaced .....................................................................................................

22

APPENDIX 2: General Pharmacokinetic Study Design and Data Handling . .. . .. . .. . . . . . . .. . .. . . . . . . .. . .. . . . . .. . . . . .. . . . . .. . .. . . . . .. . . . .. .. .. . 23

 GUIDANCE FOR INDUSTRY’

Bioavailability and Bioequivalence Studies for Orally Administered

Drug Products - General Considerations

I
This guidance represents the Food and Drug Administration’s current thinking on this topic. It does
not create or confer any rights for or on any person and does not operate to bind FDA or the public. An
alternative approach may be used if such approach satisfies the requirements of the applicable statutes
and regulations.

I. INTRODUCTION

This guidanceis intendedto provide recommendationsto sponsorsand/or applicantsplanning to include

bioavailability(BA) and bioequivalence(BE) infomration for orally administereddrug productsin
investigationalnew drug applications(IN&), new drug applications@IDAs), abbreviatednew drug
applications(ANDAs), and their supplements.This guidanceaddresseshow to meet the BA and BE
requirementsset forth in 21 CFR part 320 as they apply to dosageforms intended for oral
. .
admrm&ation2 The guidanceis also generally applicableto non-orally administereddrug products
where reliance on systemicexposuremeasuresis suitableto documentBA and BE (e.g., transdermal
delivery systemsand certainrectal and nasal drug products). The guidanceshould be useful for
applicantsplanning to conductBA and BE studiesduring the IND period for an NDA, BE studies
intendedfor submissionin an ANDA, and BE studiesconductedin the postapprovalperiod for certain
changesin both NDAs and ANDAs.~

This guidanceis designedto reducethe need for FDA drug-specificBABE guidances.As a result,this
guidancereplacesa number of previously issuedFDA drug-specificBE guidances(seethe list in
Appendix 1). On rare occasions,FDA may decide to provide additional BABE guidancesfor specific
drug products.

II. BACKGROUND

’ This guidance has been prepared by the Biopharmaceutics Coordinating Committee in the Center for Drug
Evaluation and Research (CDER) at the Food and Drug Administration (FDA).

’ These dosage forms include tablets, capsules, solutions, suspensions, conventional/immediate release, and
modified (extended, delayed) release drug products.

3 Other Agency guidances are available that consider specific scale-up and postapproval changes (SUPAC) for
different types of drug products to help satisfy regulatory requirements in both 2 1 CFR part 320 and 2 1 CFR 3 14.70.
k General

Studiesto measureBA and/or establishBE of a product are important elementsin supportof

INDs, NDAs, ANDAs, and their supplements.As part of INDs and NDAs for orally
administereddrug products,BA studiesfocus on determiningthe processby which a drug is
releasedfrom the oral dosageform and moves to the site of action. BA dataprovide an
estimateof the traction of the drug absorbed,as well as its subsequentdistriiution and
elimination. BA can be generallydocumentedby a systemicexposureprofile obtainedby
measuringdrug and/ormetaboliteconcentrationin the systemiccirculationover time. The
systemicexposureprofile determinedduring clinical trials in the IND period can serveas a
benchmarkfor subsequentBE studies.

Studiesto establishBE betweentwo products are important for certain changesprior to

approval for a pioneer product in NDA and ANDA submissions,and in the presenceof certain
postapprovalchangesin NDAs and ANDAs. In BE studies,an applicantcomparesthe
systemicexposureprofile of a test drug product to that of a referencedrug product. For two
orally administereddrug productsto be bioequivalent,the active drug ingredientor active
moiety in the test product should exhibit the samerate and extent of absorptionas the reference
drug product.

Both BA and BE studiesare requiredby regulations,dependingon the type of applicationbeing

submitted. Under 21 CFR 3 14.94,BE information is requiredto ensuretherapeutic
equivalencebetweena pharmaceuticallyequivalenttest drug product and a referencelisted
drug. Regulatoryrequirementsfor documentationof BA and BE are provided in 21 CFR part
320, which containstwo subparts. SubpartA covers generalprovisions,while SubpartB
contains18 sectionsdelineatingthe following generalBABE requimrnents:

Requirementsfor submissionof BA and BE data (320.21)

Criteria for waiver of an in vivo BA or BE study (320.22)
Basis for demonstratingin vivo BA or BE (320.23)
Types of evidenceto establishBA or BE (320.24)
Guidelinesfor conductof in vivo BA studies(320.25)
Guidelineson designof single-doseBA studies(320.26)
Guidelineson designof multiple-dosein vivo BA studies(320.27)
Correlationsof BA with an acutepharmacologicaleffect or clinical evidence (320.28)
Analytical methodsfor an in vivo BA study (320.29)
Inquiries regardingBA and BE requirementsand review of protocolsby FDA (320.30)
Applicability of requirementsregardingan IND application(320.31)
Proceduresfor establishingand amendinga BE requirement(320.32)
Criteria and evidenceto assessactual or potential BE problems(320.33)
Requirementsfor batch testing and certification by FDA (320.34)

2
. .

0 Requirementsfor in vitro batch testingof eachbatch (320.35)

0 Requirementsfor maintenanceof recordsof BE testing (320.36)
l Retentionof BA samples(320.38)
l Retentionof BE samples(320.63)

B. Bioavailability

Bioavailability is definedin 21 CFR 320.1 as ‘the rate and extentto which the activeingredient
or activemoiety is absorbedfrom a drug product and becomesavailableat the site of action.
For drug products that are not intendedto be absorbedinto the bloodstream,bioavailability
may be assessed by measurementsintendedto reflect the rate and extentto which the active
ingredientor activemoiety becomesavailableat the site of action.” This definition focuseson
the processesby which the active ingredientsor moieties are releasedfrom an oral dosageform
and move to the site of action.

From a pharmacokineticperspective,BA data for a given formulationprovide an estimateof the

relative traction of the orally administereddosethat is absorbedinto the systemiccirculation
when comparedto the BA data for a solution, suspension,or intravenousdosageform (21 CFR
320.25 (d) (2) and (3)). In addition, BA studiesprovide other useful pharmacokinetic
informationrelatedto distribution,elimination,the effectsof nutrientson absorptionof the drug,
doseproportionality,linearity in pharmacokineticsof the activemoietiesand, where
appropriate,inactivemoieties. BA datamay also provide information indirectly aboutthe
propertiesof a drug substanceprior to entry into the systemiccirculation,suchas permeability
and the intluenceof presystemicenzymesand/or transporters(e.g., p-glycopmtein).

BA for orally administereddrug productsmay be documentedby developinga systemic

exposureprofile obtainedfrom m easuringthe concentrationof activeingredientsandor active
moietiesand, when appropriate,its activemetabohtesover time in samplescolk&ed from the
systemiccirculation. Systemicexposurepatternsreflect both releaseof the drug substancefrom
the drug product and a seriesof possiblepresystemic/systemic actionson the drug substance
after its releaset?om the drug product. Additional comparativestudiesshouldbe performedto
understandthe relative contribution of theseprocessesto the systemicexposurepattern.

One regulatoryobjective is to assess,through appropriatelydesignedBA studies,the

performanceof the formulationsusedin the clinical trials that provide evidenceof safetyand
efficacy (21 CFR 320.25(d)(l)). The performanceof the clinical trial dosageform may be
optimized,in the contextof demonstratingsafetyand efficacy,before marketinga drug product.
The systemicexposureprofiles of clinical trial materialcan be usedas a benchmarkfor
subsequentformulation changesand may thus be useful as a referencefor Uure BE studies.

Although BA studieshavemany pharmacokineticobjectivesbeyond formulationperformance

as describedabove,it shouldbe noted that subsequentsectionsof this guidancefocus on using

3
relative BA (referredto as product quality BA) and, in particular, BE studiesas a meansto
documentproduct quality. In vivo performance,in terms of BABE, may be consideredto be
one aspectof product quality that provides a link to the performanceof the drug product used
in clinical trials and thus to the databasecontainingevidenceof safetyand efficacy.

C. Bioequivalence

Bioequivalenceis defined at 21 CFR 320.1 as “the absenceof a significant differencein the rate
and extentto which the active ingredientor activemoiety in pharmaceuticalequivalentsor
pharmaceuticalalternativesbecomesavailableat the site of drug action when administeredat the
samemolar doseunder similar conditionsin an appropriatelydesignedstudy.” As noted in the
statutorydefinitions,both BE and product quality BA focus on the releaseof a drug substance
from a drug product and subsequentabsorptioninto the systemiccirculation. For this mason,
similar approachesto m easuringBA in an NDA shouldgenerallybe followed in demonstrating
BE for an NDA or an ANDA. Establishingproduct quality BA is a benchmarkingeffort with
comparisonsto an oral solution, oral suspension,or an intravenousformulation. In contrast,
demonstratingBE is usually a more formal comparativetest that usesspecifiedcriteria for
comparisonsand predeterminedBE limits for the criteria.

1. IiVDhJDAs

BE documentationmay be useful during the INDNDA period to establishlinks

between(1) early and late clinical trial formulations;(2) formulationsusedin clinical trial
and stability studies,if different; (3) clinical trial formulationsand to-be-marketeddrug
product; and (4) other comparisons,as appropriate. In each comparison,the new
formulation or new method of manufactureis the test product and the prior formulation
or method of manufactureis the referenceproduct. The need to redocumentBE during
the IND period is generallyleft to the judgment of the sponsor,who may wish to use the
principlesof relevantguidances(in this guidance,seesectionsII.C.3, Postapproval
Changes,and IILD, In Vitro Studies)to determinewhen changesin components,
composition,and/or method of manufacturesuggesta needto perform further in vitro
andor in vivo studies.

A test product may fail to meet BE limits becausethe test product has higher or lower
measuresof rate and extent of absorptioncomparedto the referenceproduct or
becausethe performanceof the test or referenceis more variable. In some cases,
nondocumentationof BE may arisebecauseof inadequatenumbersof subjectsin the
study relativeto the magnitudeof inttasubjectvariability, and not becauseof eitherhigh
or low relative BA of the test product. Adequatedesign and executionof a BE study
will facilitateunderstandingof the causesof nondocumentationof BE.

Where the test product generatesplasmalevels that are substantiallyabovethoseof the

referenceproduct, the regulatory concernis not therapeuticfailure, but the adequacyof

4
the safety databasefrom the test product. Where the test product has levels that are
substantiallybelow those of the referenceproduct, the regulatory concernbecomes
themlxutic efficacy. When the variability of the test product rises,the regulatory
concernrelatesto both safety and efficacy, becauseit may suggestthat the test product
does not perform as well as the referenceproduct, and the test product may be too
variableto be clinically useful.

Propermapping of individual dose-responseor concentration-response curvesis useful

in situationswhere the drug product has plasmalevelsthat are either higher or lower
than the referenceproduct and am outsideusual BE limits. In the absenceof individual
data, population dose-responseor concentration-response data acquired over a range
of doses,including dosesabove the recommendedtherapeuticdoses,may be sufficient
to demonstratethat the increasein plasmalevels would not be accompaniedby
additionalrisk. Similarly, population dose-or concentration-response relationships
observedover a lower range of doses,including dosesbelow the recommended
therapeuticdoses,may be able to demonstratethat reducedlevels of the test product
comparedto the referenceproduct are associatedwith adequateefficacy. In either
event,the burden is on the sponsorto demonstratethe adequacyof the clinical trial
dose-responseor concentration-responsedata to provide evidenceof therapeutic
equivalence.In the absenceof this evidence,a failure to documentBE may suggesta
need for a reformulation,a changein the method of manufacturefor the test product,
and/or a repeatof the BE study.

2. ANDAs

BE studiesare a critical componentof ANDA submissions.The purposeof these

studiesis to demonstrateBE betweena pharmaceuticallyequivalentgenericdrug
product and the correspondingreferencelisted drug (21 CFR 314.94 (a)(7)). Together
with the determinationof pharmaceuticalequivalence,establishingBE allows a
regulatoryconclusionof therapeuticequivalence.

3. Postapproval Changes

Information on the types of in vitro dissolutionand in vivo BE studiesthat shouldbe

conductedfor immediate-releaseand modified-releasedrug products approvedas
eitherNDAs or ANDAs in the presenceof specifiedpostapprovalchangesis provided
in the FDA guidancesfor industry entitled SUPAC-IR: Immediate Release Solid Oral
Dosage Forms: Scale-Up and Post-Approval Changes: Chemistry,
Manufacturing, and Controls, In Vitro Dissolution Testing, and In Vivo
Bioequivalence Documentation (November 1995); and SUPAC-MR: ModiJied
Release Solid Oral Dosage Forms: Scale- Up and Post-Approval Changes:
Chemistry, Manufacturing, and Controls, In Vitro Dissolution Testing, and In
Vivo Bioequivalence Documentation (September1997). In the presenceof certain

5
 major changesin components,composition,and/or method of manufactureafter
approval,in vivo BE shouldbe redemonstrated. For approvedNDAs, the drug
product after the changeshould be comparedto the drug product before the change.
For approved ANDAs, the drug product atler the changeshould be comparedto the
referencelisted drug. Under section 506A(t)(2)(B) of the FederalFood, Drug, and
Cosmetic Act, postapprovalchangerequiring completionof studiesin accordancewith
21 CFR part 320 must be submittedin a supplementand approvedby FDA before
distriiuting a drug product made with the change.

III. METHODS TO DOCUMENT BA AND BE

As noted at 21 CFR 320.24, severalin vivo and in vitro methodscan be used to measureproduct
quality BA and establishBE. In descendingorder of preference,theseinclude pharmacokinetic,
pharmacodynamic,clinical, and in vitro studies. Thesegeneralapproachesare discussedin the
following sectionsof this guidance. Productquality BA and BE frequentlyrely on pharmacokinetic
measuressuch as AUC and Cmax that are reflective of systemicexposure.

A. Pharmacokinetic Studies

1. General Considerations

The statutorydefinitions of BA and BE, expressedin terms of rate and extent of

absorptionof the active ingredientor moiety to the site of action,emphasizethe use of
pharmacokineticmeasuresin an accessiblebiological matrix such as blood, plasma,
and/or serumto indicatereleaseof the drug substancet?omthe drug product into the
systemiccirculation.4This approachrestson an understandingthat measuringthe active
moiety or ingredientat the site of action is generallynot possibleand furthermore,that
somerelationshipexistsbetweenthe efficacy/safetyand concentrationof active moiety
and/or its irnIwtant metaboliteor metabolitesin the systemiccirculation. To measure
product quality BA and establishBE, relianceon pharmacokineticmeasurementsmay
be viewed as a bioassaythat assessesreleaseof the drug substancefrom the drug
product into the systemiccirculation. A typical study is conductedas a crossoverstudy.
In this type of study, clearance,volume of distribution,and absorption,as determined
by physiologicalvariables(e.g. gastricemptying,motility, pH), are assumedto have less
interoccasionvariability comparedto the variability arisingt?omformulation
performance. Therefore,differencesbetweentwo productsdue to formulation factors
can be determined.

4 If serial measurements of the drug or its metabolites in plasma, serum, or blood cannot be accomplished,
measurement of urinary excretion may be used to document BE.

6
2. Pilot Study

If the sponsorchooses,a pilot study in a small number of subjectscan be carried out

before proceedingwith a full BE study. The study can be usedto validateanalytical
methodology,assessvariability, optimize samplecollectiontime intervals,and provide
other information. For example,for conventionalimmediate-release products,careful
timing of initial samplesmay avoid a subsequentfinding in a full-scalestudy that the first
samplecollection occursafler the plasma concentrationpeak. For modified-release
products,a pilot study can help determinethe samplingscheduleto assesslag time and
dosedumping. A pilot study that documentsBE may be acceptable,provided that its
designand executionare suitableand a sufficient number of subjects(e.g., 12) have
completedthe study.

3. Pivotal Bioequivalence Studies

Generalrecommendationsfor a standardBE study basedon pharmacokinetic

measurementsare provided in Appendix 2.

4. Nonreplicate Study Designs

Nonreplicatestudy designsare recommendedfor BE studiesof most orally

administered,immediate-releasedosageforms. However, sponsorsand/or applicants
have the option of using replicatedesignsfor BE studiesof thesedrug products. These
studiesare descrii in sectionIRA.5 below. The recommendedmethod of analysis
of nonmplicateor replicatestudiesto establishBE is discussedin sectionIV. General
recommendationsfor nonreplicatestudy designsare provided in Appendix 2.

5. Replicate Study Designs

Replicatestudy designsare recommendedfor BE studiesof modified-releasedosage

forms and highly variabledrug products(within-subjectcoefficientof variation 2 30%),
including thosethat are immediaterelease,modified-release,and other orally
administereddrug products. The recommendedmethod of analysisof replicatestudies
to establishBE is discussedin sectionIV.

Replicatestudy designsoffer severalscientific advantagescomparedto nonreplicate

designs. The advantagesof replicatestudy designsare that they (1) allow comparisons
of within-subject variancesfor the test and referenceproducts; (2) indicate whether a
testproduct exhibits higher or lower within-subjectvariability in the bioavailability
measureswhen comparedto the referenceproduct; (3) suggestwhether a subject-by-
formulation (S*F) interactionmay be present;(4) provide more information about
factorsunderlying formulationperformance;and (5) reducethe number of subjects
neededin the BE study.

7
6. Study Population

Unless otherwise indicated by a specific guidance, subjects recruited for in vivo BE

studies should be 18 years of age or older and capable of giving informed consent. This
guidance recommends that iu vivo BE studies be conducted in iudividuals representative
of the general population, taking into account age, sex, and race factors. If the drug
product is inteuded for use iu both sexes,the sponsor should attempt to include similar
proportions of males and females in the study. If the drug product is to be used
predominantly iu the elderly, the sponsor should attempt to include as many subjects of
60 years of age or older as possible. The total number of subjects in the study should
provide adequatepower for BE demonstration,but it is not expectedthat there will be
sufficient power to draw conclusions for each subgroup. Statistical analysis of
subgroups is not recommended. Restrictionson admissioninto the study shouldgenerally
be basedsolely on safety considerations.In someinstances,it may be useful to admit
patientsinto BE studiesfor whom a drug product is intended. In this situation, sponsors
and/or applicantsshould attempt to enterpatientswhose diseaseprocessis stablefor the
durationof the BE study. In accordance with 21 CFR 320.3 1, for some products that
will be submitted in ANDAs, au IND may be required for BE studies to ensurepatient
safety.

7. Single-Dose/Multiple-Dose Studies

Instances where multipledose studies may be usetil are defiued at 21 CFR

320.27(a)(3). However, this guidauce generally recommends single-dose
pharmacokinetic studies for both immediate- and modified-release drug products to
demonstrate BE because they are generally more sensitive in assessingreleaseof the
drug substancefrom the drug product into the systemic circulation (see section V). If a
multiple-dose study design is necessary,appropriate dosageadministration and
sampling should be carried out to document attainment of steady state.

8. Bioanalytical Methodology

Bioanalytical methods for BA and BE studies should be accurate,precise, selective,

sensitive, and reproducible. A separateFDA guidance entitled Bioanalytical Methods
Validation for Human Studies (published in drawlin December 1998) will be available,
when f?nalized,to assist sponsorsiu validating bioanalytical methods.

9. Pharmacokinetic Measures of Systemic Exposure

Both direct (e.g., rate constar& rate profile) and indirect (e.g., Cmax, Tmax, mean
absorption time, mean residence time, Cmax normalized to AUC) pharmacokiuetic
measuresare limited in their ability to assessrate of absorption. This guidance,

8
therefore,recommendsa changein focus from thesedirect or indirect measuresof
absorptionrate to measuresof systemicexposure. Cmax and AUC can continueto be
used as measuresfor product quality BA and BE, but more in terms of their capacityto
assessexposurethan their capacity to reflect rate and extent of absorption. Relianceon
systemicexposuremeasuresshould reflect comparablerate and extent of absorption,
which in turn should achievethe underlying statutoryand regulatoryobjectiveof
ensuringcomparabletherapeuticeffects. Exposuremeasuresare defined relative to
early, peak, and total portions of the plasma,serum,or blood concentrationtime
profile, as follows:

a. Early Exposure

For orally administeredimmediate-releasedrug products,BE may generallybe

demonstratedby measurementsof peak and total exposure. An early exposure
measuremay be indicatedon the basisof appropriateclinical efficacy/safetytrials
and/or pharmacokinetic/pharmacodynamic studiesthat call for better control of drug
absorptioninto the systemiccirculation (e.g.,to ensurerapid onsetof an analgesiceffect
or to avoid an excessivehypotensiveaction of an antihypenensive).In this setting,the
guidancerecommendsuse of partial AUC as an early exposuremeasure. The partial
areashouldbe truncatedat the populationmedian of Tmax valuesfor the reference
formulation. At leasttwo quantifiablesamplesshouldbe collectedbefore the expected
peak time to allow adequateestimationof the partial area.

b. Peak Exposure

Peak exposureshould be assessedby measuringthe peak drug concentration(Cmax)

obtaineddirectly from the datawithout interpolation.

C. Total Exposure

For single-dosestudies,the measurementof total exposureshould be:

0 Area under the plasma/serum/bloodconcentration-timecurve from time zero to

time t (AU&), where t is the last time point with measurableconcentrationfor
individualformuation.

l Area under the plasma/scrumMoodconcentration-timecurve from time zero to

time i&nitty (AUC&,), where AU& = AU&, + C&, Ct is the last
measurabledrug concentrationand h, is the terminal or eliminationrate constant
calculatedaccordingto an appropriatemethod. The tem~inalhalf-life (t& of
the drug should also be reported.

9
 For steady-statestudies,the measurementof total exposureshould be the areaunder
the plasma,serumor blood concentration-timecurve t?omtime zero to time z over a
dosing interval at steadystate(AU&), where z is the length of the dosinginterval.

B. Pharmacodynamic Studies

Pharmacodynamicstudiesare not recommendedfor orally administereddrug productswhen the

drug is absorbedinto the systemiccirculation and a pharmacokineticapproachcan be used to
assesssystemicexposureand establishBE. However, in thoseinstanceswhere a
pharmacokineticapproachis not possible,suitably validatedpharmacodynamicmethodscan be
used to demonstrateBE.

C. Comparative Clinical Studies

Where them are no other means,well-controlledclinical trials in humansmay be useful to

provide supportiveevidenceof BA or BE. However, the use of comparativeclinical trials as an
approachto demonstrateBE is generallyconsideredinsensitiveand should be avoidedwhere
possible(21 CFR 320.24). The use of BE studieswith clinical trial endpointsmay be
appropriateto demonstrateBE for orally administereddrug productswhen measurementof the
active ingredientsor activemoietiesin an accessiblebiological fluid (pharmacokineticapproach)
or pharmacodynamicapproachis infeasible.

D. In Vitro Studies

Under certain circumstances,product quality BA and BE can be documentedusing in vitro

approaches(21 CFR 320.24). For highly soluble,highly permeable,rapidly dissolving,orally
administemddrug products,documentationof BE using an in vitro approach(dissolution
studies)is appropriatebasedon the biopharmaceuticsclassificationsystern5 This approach
may also be suitableunder somecircumstancesin assessingBE during the IND period, for
NDA and ANDA submissions,and in the presenceof certain postapprovalchangesto
approved NDAs and ANDAs. In addition, in vitro approachesto documentBE for
nonbioproblem drugs approvedprior to 1962 remain acceptable(21 CFR 320.33).

Dissolution testing is also usedto assessbatch-to-batchquality, where the approachmay

become one of the tests,with defined procedures,in a drug product specificationto allow batch
release. Dissolution testing is also used to (1) provide processcontrol and quality assurance,
and (2) assessthe need for further BE studiesrelative to minor postapprovalchanges,where
dissolutioncan functionas a signalof bioinequivalence.In vitro dissolutioncharacterizationis
encouragedfor all product formulationsinvestigated(includingprototypeformulations),

5 See the FDA guidance for industry on Waiver of In Vivo Bioavailability and Bioequivalence Studies for Immediate
Release Solid Oral Dosage Forms Based on a Biopharmaceutics Classijication System (August 2000). This
document provides complementary information on the BiopharmaceuticsClassification System (BCS).

10
 particularly if in vivo absorptionchamcteristicsare being defined for the different product
formulations. Such efforts may enablethe establishmentof an in vitro-in vivo correlation. When
an in vitro-in vivo correlationor associationis available(21 CFR 320.22),the in vitro test can
servenot only as a quality control specificationfor the manufacturingprocess,but also as an
indicator of how the product will perform in vivo. The following guidancesprovide
recommendationson the developmentof dissolutionmethodology,settingspecifications,and the
regulatoryapplicationsof dissolutiontesting: (1) Dissolution Testing of Immediate Release
Solid Oral Dosage Forms (August 1997); and (2) Extended Release Oral Dosage Forms:
Development, Evaluation, and Application of In Vitro/In Vivo Correlations (September
1997).

This guidancerecommendsthat dissolutiondata from threebatchesfor both NDAs and

ANDAs be usedto set dissolutionspecificationsfor modified-releasedosageforms, including
extended-releasedosageforms.

Iv. COMPARISON OF BAMEASURES IN BE STUDIES

An equivalenceapproachhas been and continuesto be recommendedfor BE comparisons. The

recommendedapproachrelies on (1) a criterion to allow the comparison,(2) a confidenceinterval for
the criterion, and (3) a BE limit. Log-transformationof exposuremeasuresprior to statisticalanalysisis
recommended. BE studiesare performed as single-dose,crossoverstudies. To comparemeasuresin
thesestudies,data have been analyzedusing an averageBE criterion. This guidancerecommends
continueduse of an averageBE criterion to compareBA measuresfor replicateand nonreplicateBE
studiesof both immediate-and modified-releaseproducts.However, sponsorshave the option to
explain why they would use anothercriterion (e.g.,an individual BE criterion for replicatedesignstudies
of highly variable drug products). Sponsorsshould documentselectionof the criterion in the study
protocol. Sponsorsandor applicantswishing further informationon this approachshouldcontactthe
appropriateCDER review division. The criteria to allow comparisonof BE measureswill be provided
in a separateFDA guidancefor industry.6 When the individual or populationBE criterion is used,in
addition to meetingthe BE limit basedon confidencebounds,the point estimateof the geometric
test/referencemean ratio should fall within SO-125%.

V. DOCUMENTATION OF BA AND BE

An in vivo study is generallyrecommendedfor all solid oral dosageforms approvedafter 1962and for
bioproblem drug products approvedprior to 1962. Waiver of in vivo studiesfor different strengthsof
a drug product may be grantedunder 21 CFR 320.22 (d)(2) when (1) the drug product is in the same

‘Average, Population, and Individual Approaches to Establishing Bioequivalence (draft guidance published
August 1999). When finalized, this guidance will provide recommendations on criteria for comparison of BE
measures.

11
dosageform, but in a different strength;(2) this d&rent strengthis proportionally similar in its active
and inactive ingredientsto the sttengthof the product for which the samemanufacturerhas conducted
an acceptablein vivo study; and (3) the new strengthmeetsan appropriatein vitro dissolutiontest. This
guidancedefinesproportionally similar in two ways:

Definition 1: All active and inactive ingredientsare in exactly the sameproportion between
di&rent strengths(e.g.,a tablet of 50-mg strengthhas all the inactive ingredients,exactly half
that of a tablet of lOO-mgstrength,and twice that of a tablet of 25mg strength).

Definition 2: The total weight of the dosageform remainsnearly the samefor all strengths
(within f 5 percentof the total weight of the strengthon which a bio-study was performed),the
sameinactive ingredientsare usedfor all strengths,and the changein any strengthis obtainedby
altering the amount of the active ingredientand one or more of the inactiveingredients.7For
example,with respectto an approved5-mg tablet, the total weight of new l- and 2.5-mg
tabletsremainsnearly the same,and the changesin the amountof active ingredientare offset by
a changein one or more inactiveingmdients. This definition is generallyapplicableto high-
potency drug substanceswhere the amount of active drug substancein the dosageform is
relatively low (e.g.,_<5 mg).

k Solutions

For oral solutions,elixirs, syrups,tinctures,or other solubilizedforms, BA and/or BE can be

demonstratedusing nonclinical studies(21 CFR 320.22(b)(3)(i)). Generally,in vivo BE studies
are waived for solutionson the assumptionthat releaseof the drug substancefrom the drug
product is selfevident and that the solutionsdo not contain any excipientthat significantlyaffects
drug absorption (21 CFR 320.22 (b) (3) (iii)). H owever, there are certain excipients,such as
sorbitol or mannitol,that can reducethe bioavailabilityof drugswith low intestinalpermeability
in amountssometimesusedin oral liquid dosageforms.

B. Suspensions

BA and BE for a suspensionshould generallybe establishedas for immediate-releasesolid oral

dosageforms, and both in vivo and in vitro studiesare recommended.

C. Immediate-Release Products: Capsules and Tablets

1. General Recommendations

For product quality BA and BE studies,where the focus is on releaseof the drug
substancefrom the drug product into the systemiccirculation,a single-dose,fasting

’ The changes in the inactive ingredients should be within the limits defined by the SUPAC -IR and SUPAC-MR
guidances.

12
 study should be performed. In vivo BE studiesshouldbe accompaniedby in vitro
dissolutionprofiles on all skengthsof eachproduct. For ANDAs, the BE study should
be conductedbetweenthe test product and referencelisted drug using the strength
specifiedin Approved Drug Products with Therapeutic Equivalence Evaluations
(Orange Book).

2. Waivers of In Vivo BE Studies (Biowaivers)

a. INDs, NDAs, and ANDAs: Preapproval

When the drug product is in the samedosageform, but in a different strength,and is

proportionallysimilar in its active and inactiveingmdients,an in vivo BE demonstration
of one or more lower strengthscan be waived basedon dissolutiontestsand an in vivo
study on the higheststrength.*

For an NDA, biowaivers of a higher strengthwill be determinedto be appropriate

basedon (1) clinical safetyand/or efficacy studiesincluding data on the doseand the
desirabilityof the higher &err& (2) linear eliminationkineticsover the therapeuticdose
range;(3) the higher strengthbeing proportionallysimilar to the lower &en& and (4)
the samedissolutionproceduresbeing used for both strengthsand similar dissolution
resultsobtained. A dissolutionprofile shouldbe generatedfor all strengths.

The fi test should be usedto compareprofiles from the different strengthsof the
product. An & value_>50 indicatesa sufficiently similar dissolutionprofile suchthat
further in vivo studiesare not necessary.For an fi value < 50, further discussionswith
CDER review staff may help to determinewhetheran in vivo study is important(2 1
CFR 320.22 (d)(2)(ii)). The $ approachis not suitablefor rapidly dissolvingdrug
products (e.g., _>85% dissolvedin 15 minutesor less).

For an ANDA, conductingan in vivo study on a strengththat is not the highestmay be

appropriatefor reasonsof safety, subjectto approval by review staff. In addition, as
with an NDA, the Agency will considera waiver requestfor a recentlyapprovedhigher
strengthwhen an in vivo BE study was performedon a lower strengthof the samedrug
product submittedin an ANDA under the following circumstances:

l Linear elimination kinetics has been shown over the therapeuticdoserange.

l The higher strengthis proportionallysimilar to the lower strength.

’ This recommendation modifies a prior policy of allowing blowaivers for only three lower strengths on ANDAs.

13
 l Comparativedissolutiontestingon the higherstrengthof the test and referencedrug
productis submittedand found acceptable.

l The sponsorinitiatedthe BE studyon the lower strengthwithin five working daysof

the approvaldateof a higher strengthof the referencelisted drug. A studyis
consideredinitiated when the first subjectis dosed.

Sponsorsof ANDAs wishing to submita biowaiverrequestunderthesecircumstances

shouldfirst contactthe RegulatorySupportBranch,Office of GenericDrugs, for advice
on the proper filing procedure.

b. NDAs and ANLIAs: Postapproval

Informationon the typesof in vitro dissolutionand in vivo BE studiesfor irnmediate-

releasedrug productsapprovedas either NDAs or ANDAs in the presenceof
specifiedpostapprovalchangesareprovidedin an FDA guidancefor industryentitled
SUPAC-IR: Immediate ReleaseSolid Oral DosageForms: Scale-Up and Post-
Approval Changes: Chemistry, Manufacturing, and Controls, In Vitro Dissolution
Testing,and In Vivo BioequivalenceDocumentation(November 1995). For
postapprovalchanges,the in vitro comparisonshouldbe madebetweenthe prechange
and postchangeproducts. In instanceswheredissolutionprofile comparisonsare
recommended,an fi test shouldbe used. An t$ valueof 150 suggests a sufliciently
similar dissolutionprofile andno furtherin vivo studiesate needed.W h e n in vivo BE
studiesare recommended,the comparisonshouldbe madefor NDAs betweenthe
prechangeand postchangeproducts,and for ANDAs betweenthe postchangeand
referencelisted drug products.

D. Modified-Release Products

Modified-releaseproductsinclude delayed-release
productsand extended(controlled)-release
products.

As definedin the U.S. Pharmacopeia (USP), delayed-release drug productsare dosageforms

that releasethe drugsat a tim e later than immediatelyafter administration(i.e., thesedrug
productsexhibit a lag tim e in quantifiableplasmaconcentrations).Typically, coatings(e.g.,
entericcoatings)are intendedto delaythe releaseof medicationuntil the dosageform has
passedthroughthe acidic medium of the stomach.In vivo testsfor delayed-release drug
productsare similar thosefor to extended-release drug products. In vitro dissolutiontestsfor
theseproductsshoulddocumentthat they are stableunderacidic conditionsand that they
releasethe drug only in a neutralmedium (e.g.,pH 6.8).

14
Extended-releasedrug products are dosageforms that allow a reduction in dosing t?equencyas
comparedto when the drug is presentin an immediate-releasedosageform. Thesedrug
products can also be developedto reduce fluctuationsin plasma concentrations.Extended-
releaseproducts can be capsules,tablets,granules,pellets, and suspensions.If any part of a
drug product includesan extended-release component,the following recommendationsapply.

I. NDAs: BA and BE Studies

An NDA can be submittedfor a previously unapprovednew molecular entity, or for a

new salt, new ester,prodrug, or other noncovalentderivativeof a previously approved
new molecularentity, formulatedas a modified-releasedrug product. The first
modified-releasedrug product for a previously approvedimmediate-releasedrug
product should be submittedas an NDA. Subsequentmodified-releaseproductsthat
are pharmaceuticallyequivalentand bioequivalentto the listed drug product should be
submittedas ANDAs. BA recommendationsfor the NDA of an extended-release
product are consideredat 21 CFR 320.25(f). The purpose for an in vivo BA study for
which a controlled-release
claim is made is to determineif all of the following conditions
are met:

0 The drug product meetsthe controlledreleaseclaims made for it.

0 The BA profile establishedfor the drug product rules out the occurrenceof any
dosedumping.

0 The drug product’s steady-stateperformanceis equivalentto a currently

marketednoncontrolledreleaseor controlled-releasedrug product that contains
the sameactive drug ingredientor therapeuticmoiety and that is subjectto an
approvedfull new drug application.

0 The drug product’s formulation providesconsistentpharmacokinetic

performancebetweenindividual dosageunits.

As noted at 21 CFR 320.25 (f) (2), the reference material(s) for such a BA study
shall be chosen to permit an appropriate scientific evaluation of the controlled
release claims madefor the drug product, such as:

0 A solutionor suspensionof the active drug ingredientor therapeuticmoiety

0 A currently marketednoncontrolled-release drug product containingthe same

active drug ingredientor therapeuticmoiety and administeredaccordingto the
dosagerecommendationsin the labeling

15
a A currently marketed controlled-releasedrug product subjectto an approved
full new drug applicationcontainingthe sameactive drug ingredient or
therapeuticmoiety and administeredaccordingto the dosagerecommendations
in the labeling

To satisfy the CFR recommendationsfor BA studiesfor an extended-releasedrug

product submitted as an NDA, this guidancerecommendsthe following studies:

0 A single-dose,fasting study on all strengthsof tablets/capsulesand highest

strength of beaded capsules

0 A single-dose,f&-effect study on the highest strength

0 A steady-statestudy on the highest strength

When substantialchangesin the components/compositionand/or method of

manufacture for an extended-releasedrug product occur between the to-be-marketed
NDA dosageform and the clinical trial material, BE studiesare recommended.

2. ANDAs: BE Studies

For extended-releaseproducts submitted as ANDAs, the following studiesare

recommended: (1) a single-dose,replicate, fasting study comparing the highest strength
of the test and reference listed drug product; and (2) a food-effect, nonreplicate study
comparing the highest strengthof the test and referenceproduct (section VIA).
Becausesingle-dosestudiesare consideredmore sensitivein addressingthe primary
question of BE (i.e., releaseof the drug substancefrom the drug product into the
systemiccirculation), multiple-dose studiesam generally not recommended,even in
instanceswhere nonlinear kinetics are present

3. Waivers of In Vivo BE Studies (Biowaivers): NDAs and ANDAs

a. Beaded Capsules- Lower Strength

For extended-releasebeaded capsuleswhere the strengthdiffers only in the number of

beadscontaining the active moiety, a single-dose,fasting BE study should be carried out
only on the highest strength,with waiver of in vivo studiesfor lower strengthsbasedon
dissolutionprofiles. A dissolutionprofile should be generatedfor each strengthusing the
recommendeddissolutionmethod. The f2 test should be used to compareprofiles from
the different strengthsof the product. An G value of 2 50 can be used to confirm that
fbrther in vivo studiesare not needed.

16
 b. Tablets - Lower Strength

For extended-releasetablets,when the drug product is in the samedosageform but in a

different skngth, is proportionallysimilar in its active and inactiveingredients,and has
the samedrug releasemechanism,an in vivo BE determinationof one or more lower
strengthscan be waived basedon dissolutionprofile comparisons,with an in vivo study
only on the higheststrength. The drug productsshouldexhibit similar dissolutionprofiles
betweenthe highest strengthand the lower strengthsbasedon the 6 test in at leastthree
dissolution media (e.g.,pH 1.2,4.5 and 6.8). The dissolutionprofile shouldbe
generatedon the test and referenceproducts of all strengths.

4. Postapproval Changes

Information on the types of in vitro dissolutionand in vivo BE studiesfor extended-

releasedrug products approvedas either NDAs or ANDAs in the presenceof
specifiedpostapprovalchangesare provided in an FDA guidancefor industry entitled
SUPAC-MR: ModiJied ReleaseSolid Oral Dosage Forms: Scale-Up and Post-
Approval Changes: Chemistry, Manufacturing, and Controls, In Vitro Dissolution
Testing, and In Vivo Bioequivalence Documentation (September 1997). For
postapprovalchanges,the in vitro comparisonshould be made betweenthe pmchange
and postchangeproducts. In instanceswhere dissolutionprofile comparisonsare
recommended,an fi test should be used. An & valueof _>50 suggestsa similar
dissolutionprome. A failure to demonstratesimilar dissolutionprofilesmay result in the
needto perform an in vivo BE study. When in vivo BE studiesam conducted,the
comparisonshould be made for NDAs betweenthe prechangeand postchange
products, and for ANDAs betweenthe postchangeproduct and referencelisted drug.

E. Miscellaneous Dosage Forms

Rapidly dissolvingdrug products,such as buccal and sublingualdosageforms, should be tested

for in vitro dissolutionand in vivo BA and/orBE. Chewabletabletsshouldalso be evaluated
for in vivo BA and/or BE. Chewabletablets(as a whole) shouldbe subjectto in vitro
dissolutionbecausethey might be swallowedby a patient without proper chewing. In general,
in vitro dissolutiontest conditionsfor chewabletabletsshouldbe the sameas for nonchewable
tabletsof the sameactive ingredient/moiety.Infrequently,different test conditionsor acceptance
criteria may be indicatedfor chewableand nonchewabletablets,but thesedifferences,if they
exist, shouldbe resolvedwith the appropriatereview division.

VI. SPECIAL TOPICS

A. Food-Effect Studies

17
 Coadministrationof food with oral drug productsmay influence drug BA and/or BE. Food-
effect BA studiesfocus on the effectsof food on the releaseof the drug substancefrom the drug
product as well as the absorptionof the drug substance.BE studieswith food focus on
demonstratingcomparableBA betweentest and referenceproductswhen coadministemdwith
meals. Usually, a single-dose,two-period, two-treatment,two-sequencecrossoverstudy is
recommendedfor both food-effect BA and BE studies.

B. Moieties to Be Measured

I. Parent Drug VersusMetabolites

The moietiesto be measuredin biological fluids collectedin BA and BE studiesare

eitherthe active drug ingredientor its activemoiety in the administereddosageform
(parent drug) and, when appropriate,its active metabolites(21 CFR 320.24(b)(l)(i)).9
This guidancerecommendsthe following approachesfor BA and BE studies.

For BA studies(seesectionIIB), determinationof moietiesto be measuredin

biological fluids should take into accountboth concentrationand activity.
Concentration refers to the relative quantity of the parent drug or one or more
metabolitesin a given volume of an accessiblebiological fluid suchas blood or plasma.
Activity refersto the relative contributionof the parentdrug and its metabolite(s)in the
biological fluids to the clinical safetyand/orefficacy of the drug. For BA studies,both
the parentdrug and its major active metabolitesshouldbe measured,if analytically
feasible.

For BE studies,measurementof only the parentdrug releasedf?om the dosageform,

ratherthan the metabolite,is generallyrecommended.The rationalefor this
recommendationis that the concentration-timeprofile of the parentdrug is more
sensitiveto changesin formulationperformancethan a metabolite,which is mom
reflectiveof metaboliteformation,distribution,and elimination. The following am
exceptionsto this generalapproach.

l Measurementof a metabolitemay be preferredwhen parent drug levels are too low

to allow reliable analyticalmeasurement in blood, plasma,or serum for an adequate
length of time. The metabolitedata obtainedfrom thesestudiesshouldbe subjectto
a confidenceinterval approachfor BE demonstration.If there is a clinical concern
relatedto efficacy or safety for the parentdrug, sponsorsand/or applicantsshould

9A dosage form contains active and, usually, inactive ingredients. The active ingredient may be a prodrug that
requires further transformation in vivo to become active. An active moiety is the molecule or ion, excluding those
appended portions of the molecule that cause the drug to be an ester, salt, or other noncovalent derivative of the
molecule, responsible for the physiological or pharmacological action of the drug substance.

18
 contactthe appropriatereview division to determinewhether the parentdrug should
be measuredand analyzedstatistically.

l A metabolitemay be formed as a result of gut wall or other presystemic

metabolism. Ifthe metabolitecontributesmeaningfullyto safetyand/orefficacy,the
metaboliteand the parent drug shouldbe measured.When the relative activity of
the metaboliteis low and doesnot contributemeaningfullyto safetyand/orefficacy,
it doesnot need to be measured. The parent drug measuredin theseBE studies
shouldbe analyzedusing a confidenceinterval approach. The metabolitedata can
be used to provide supportiveevidenceof comparabletherapeuticoutcome.

2. Enantiomers VersusRacemates

For BA studies,m easmment of individual enantiomersmay be important. For BE

studies,this guidancerecommendsm easumnentof the racemateusing an achiral assay.
Measurementof individual enantiomersin BE studiesis recommendedonly when all of
the following conditionsare met: (1) the enantiomersexhibit diffemnt phannacodynamic
characteristics;(2) the enantiomersexhibit difherentpharmacokineticcharacteristics;(3)
primary efficacy/safetyactivity resideswith the minor enantiomer,and (4) nonlinear
absorptionis present(as expressedby a changein the enantiomerconcentrationratio
with changein the input rate of the drug) for at leastone of the enantiomers.In such
cases,BE criteria should be applied to the enantiomersseparately.

3. Drug Products with Complex Mixtures as the Active Ingredients

Certain drug productsmay contain complex drug substances(i.e., active moietiesor

active ingtedientsthat are mixtures of multiple syntheticand/ornaturalsource
components).Some or all of the componentsof thesecomplex drug substancesmay
not be charac$er&dwith regardto chemicalstructureand/orbiological activity.
Quantificationof all active or potentiallyactivecomponentsin pharmacokineticstudies
to documentBABE is neither necessarynor desirable. Rather,BA and BE studies
shouldbe basedon a small number of markersof rate and extent of absorption.
Although necessarilya case-by-easedetermination,criteria for marker selectioninclude
amountof the moiety in the dosageform, plasmaor blood levels of the moiety, and
biological activity of the moiety relativeto othermoietiesin the complexmixture. Where
pharmacokineticapproachesare not feasibleto assessrate and extent of absorptionof
a drug substancefrom a drug product, in vitro approachesmay be preferred.
Phamracodynamicor clinical approachesmay be called for ifno quantifiablemoieties
are availablefor in vivo pharmacokineticor in vitro studies.

C. Long Half-Life Drugs

19
In a BA/pharmacokineticstudy involving an oral product with a long half-life drug, adequate
characterizationof the half-life calls for blood samplingover a long period of time. For a BE
determinationof an oral product with a long half-life drug, a nonreplicate,single-dose,crossover
study can be conducted,provided an adequatewashout period is used. If the crossoverstudy
is problematic,a BE study with a parallel design can be used. For either a crossoveror parallel
study, samplecollection time shouldbe adequateto ensurecompletionof gastrointestinaltransit
(approximately2 to 3 days) of the drug product and absorptionof the drug substance.Cmax,
and a suitably truncatedAUC can be used to characterizepeak and total drug exposure,
respectively. For drugs that demonstratelow intra-subjectvariability in distribution and
clearance,an AUC truncatedat 72 hours (AU&72 r,J may be usedin place of AU&, or
AU&,. For drugs demonstratinghigh i&a-subject variability in distriiution and clearance,
AUC truncationwarrantscaution. In such cases,sponsorsand/or applicantsshould consult the
appropriatereview staff.

D. First Point Cmax

The first point of a concentration-timecurve in a BE study basedon blood and/orplasma

measurementsis sometimesthe highestpoint, which raisesa questionaboutthe measurementof
true Cmax becauseof insufficientearly samplingtimes. A carefullyconductedpilot study may
avoid this problem. Collection of an early time point between5 and 15 minutesafter dosing
followed by additionalsamplecollections(e.g.,two to five) in the first hour aI?erdosing may be
sufficient to assessearly peak concentrations.If this samplingapproachis followed, data sets
should be consideredadequate,even when the highest observedconcentrationoccurs at the
first time point.

E. Orally Administered Drugs Intended for Local Action

Documentationof product quality BA for NDAs where the drug substanceproducesits effects
by local action in the gastmintestinaltract can be achievedusing clinical efficacy and safety
studiesand/orsuitably designedand validatedin vitro studies. Similarly, documentationof BE
for ANDAs, and for both NDAs and ANDAs in the presenceof certain postapprovalchanges,
can be achievedusing BE studieswith clinical efficacy and safetyendpointsand/or suitably
designedand validatedin vitro studiesif the latter studiesare eitherreflectiveof important
clinical effectsor are more sensitiveto changesin product performancecomparedto a clinical
study. To ensurecomparablesafety,additional studieswith and without food may help to
understandthe degreeof systemicexposurethat occursfollowing administrationof a drug
product intendedfor local action in the gastrointestinaltract.

F. Narrow Therapeutic Range Drugs

20
 This guidancedefinesnarrow therapeuticrange” drug productsas thosecontainingcertain drug
substancesthat are subjectto therapeuticdrug concentrationor pharmacodynamicmonitoring,
and/or where product labeling indicatesa narrow therapeuticrange designation.Examples
includedigoxin,lithium,phenytoin,theophylline,and warfarin. Becausenot all drugs subjectto
therapeuticdrug concentrationor pharmacodynamicmonitoring are narrow therapeuticrange
drugs, sponsorsand/or applicantsshould contactthe appropriatereview division at CDER to
determinewhether a drug should or should not be consideredto have a narrow therapeutic
range.

This guidancerecommendsthat sponsorsconsideradditionaltesting and/orcontrols to ensure

the quality of drug productscontainingnarrow therapeuticrangedrugs. The approachis
designedto provide increasedassuranceof interchangeabilityfor drug productscontaining
specifiednarrow therapeuticrange drugs. It is not designedto influencethe practice of
medicine or pharmacy.

Unlessotherwiseindicatedby a specificguidance,this guidancerecommendsthat the traditional

BE limit of 80-125% for non-narrowtherapeuticrange drugs remain unchangedfor the
bioavailability measures(AUC and Cmax) of narrow therapeuticrange drugs.

lo This guidance uses the term “narrow therapeutic range” instead of “narrow therapeutic index” drug, although the
latter is more commonly used.

21
 APPENDIX 1

List of Guidances That Will Be Replaced

1. Guidelinesfor the Evaluation of Controlled ReleaseDrug Products (April 1984).

2. Statistical Procedures for Bioequivalence Studies Using a Standard Two-Treatment

Crossover Design (July 1992).

3. Oral Extended (Controlled) ReleaseDosage Form: In Vivo Bioequivalence and In Vitro

Dissolution Testing (September 1993).

4. Drafi Guidancefor Industry In Vivo Bioequivalence Studies Based on Population and

Individual Bioequivalence Approaches (October 1997).

5. Drug specificbioequivalenceguidanceshrn the Division of Bioequivalence,Office of Generic

Drugs, Office of PharmaceuticalScience,Centerfor Drug Evaluationand Research,FDA.

22
 APPENDIX 2

General Pharmacokinetic Study Design and Data Handling

For both replicate and nonreplicate,in vivo phamracokineticBE studies,the following general
approachesare recommendedrecognizingthat the elementsmay be adjustedfor certain drug
substancesand drug products.

Study conduct:

0 The test or referenceproducts should be administeredwith about 8 ounces(240 ml) of

water to an appropriatenumber of subjectsunder fasting conditions,unlessthe study is
a foodeffect BABE study.

0 Generally,the highestmarketedsnengthshouldbe administeredas a single unit If

necessaryfor analyticalreasons,multiple units of the higheststrengthcan be
administeredproviding the total single-doseremainswithin the labeleddoserange.

l An adequatewashoutperiod (e.g.,more than 5 half lives of the moietiesto be

measured)should separateeach treatment.

0 The lot numbersof both test and referencelisted productsand the expiration date for
the referenceproduct should be stated. The drug contentof the test product should not
differ fi-omthat of the referencelisted product by more than 5 percent. The sponsor
should include a statementof the compositionof the test product and, if possible,a
side-by-sidecomparisonof the compositionsof test and referencelisted products. In
accordancewith 21 CFR 320.38, samplesof the test and referencelisted product must
be retainedfor 5 years.

0 prior to and during each study phase,subjectsshould (1) be allowed water as desired
except for one hour before and after drug administration;(2) be provided standard
mealsno lessthan 4 hours afler drug admi&ratiom (3) abstainfrom alcohol for 24
hours prior to each study period and until after the last samplefrom eachperiod is
collected.

Samplecollectionand samplingtimes:

a Under normal circumstances,blood, rather than urine or tissue,should be used. In most

cases,drug, or metabolitesare measuredin serumor plasma. However, in certain
caseswhole blood may be more appropriatefor analysis. Blood samplesshouldbe

23
 drawn at appropriatetimes to describethe absorption,distribution, and elimination
phasesof the drug. For most drugs, 12 to 18 samples,including a predose sample,
should be collectedper subjectper dose. This samplingshould continue for at least
threeor more terminal half lives of the drug. The exacttiming for samplecollection
dependson the natureof the drug and the input from the administereddosageform.
The samplecollection shouldbe spacedin sucha way that the maximum concentration
of the drug in the blood (Cmax) and terminal eliminationrate constant(Q can be
estimatedaccurately. At leastthree to four samplesshouldbe obtainedduring the
terminal log-linearphaseto obtain an accurateestimateof h, fkom linear regression.
The actualclock time when samplesare drawn as well as the elapsedtime relatedto
drug administrationshould be recorded.

Subjectswith predoseplasma concentrations:

0 If the predoseconcentrationis less than or equal to 5 percentof Cmax value in that

subject,the subject’sdata without any adjustmentscan be included in all
pharmacokineticmeasurementsand calculations. If the predosevalue is greaterthan 5
percentof Cmax, the subjectshouldbe droppedfrom all BE study evaluations.

Data deletiondue to vomiting:

l Data fi-om subjectswho experienceemesisduring the courseof a BE study for

immediate-release productsshouldbe deletedftom statisticalanalysisif vomiting occurs
at or before 2 times median Tmax. In the caseof modified-releaseproducts,the data
from subjectswho experienceemesisany time during the labeleddosinginterval should
be deleted.

The following pharmacokineticinformationis recommendedfor submission:

0 Plasmaconcentrationsand time points

0 Subject,period, sequence,treatment
l AUC,,,AUC,,,Cmax,Tmax,h,,andt1/2
0 Intersubject,inttasubject,and/ortotal variability, if available
0 Subject-by-formulationinteractionvariancecomponent(on’), if individual BE criterion
is used
0 Cmin (concentrationat the end of a dosing interval),Cav (averageconcentrationduring
a dosing interval), degreeof fluctuation [(Cmax-Cmin)/Cav],and swing [(Crnax-
Cmin)/Cmin] if steady-statestudiesare employed
0 Partial AUC, requestedonly as discussedin sectionIII. A.9.a.

In addition,the following statisticalinformation shouldbe provided for AU&, AU&,, and Cmax:

24
.

l Geometricmean
l Arithmetic mean
l Ratio of means
l Confidenceintervals

Logarithmic transformationshouldbe provided for measuresusedfor BE demonstration.

Roundingoff of confidenceintervalvalues:

l Confidenceinterval (CI) valuesshouldnot be roundedoff; therefore,to passa CI limit

of 80-125, the value should be at least 80.00 and not more than 125.00.

25
MEMORANDUM DEPARTMENT OF HEALTH AND HUMAN SERVICES
PUBLIC HEALTH SERVICE
FOOD AND DRUG ADMINISTRATION
CENTER FOR DRUG EVALUATION AND RESEARCH

DATE:

FROM: Director
Division of OTC Drug Products, HFD-560

SUBJECT: Material for Docket No. /y 74fl-Q52N.

TO: Dockets Management Branch, HFA-305

m The attached
display under
material should be placed
the above referenced
on public
Docket No.

q This material
Comment No.
should be cross-referenced to

Charles J.