Determine impurity level in relevant batches1

Determine mean + upper confidence limit for the impurity (Let this = A)

Acceptance criterion = A or B(as appropriate)

Isimpurity alsoa degradation

product?

IsA or B

greater than thequalifiedlevel?

Acceptance criterion = qualified levelor establish new qualified level2

Estimate maximum increase in impurityat retest date using data from relevant

accelerated and long-term stability studies

Determine maximum likely level as:A + increase in degradation product at

appropriate storage conditions.(Let this = B)

DECISION TREE #1: ESTABLISHING ACCEPTANCE CRITERION FOR A SPECIFIED IMPURITY IN A NEW DRUG SUBSTANCE

1 Relevant batches are those from development, pilot and scale-up studies.2 Refer to ICH Guideline on Impurities in New Drug Substances

Definition: upper confidence limit = three times the standard deviation of batch analysis data

YES

YES

NO

NO

DECISION TREE #2: ESTABLISHING ACCEPTANCE CRITERION FOR A DEGRADATION PRODUCT IN A NEW DRUG PRODUCT

Doesdegradation

occur during productmanufacture?

Estimate maximum increase in degradation product at shelf life using

data from relevant accelerated andlong-term stability studies.

(Let this = D)

Determine maximum likely level as drug substance acceptance criterion2.

((A or B) + C + D)

Ismaximum likely level

greaterthan thequalifiedlevel?

Estimate maximum increase in degradationproduct during manufacture from relevantbatches1. (Let this = C)

Acceptance criterion = maximum likely level.

Acceptance criterion = qualified levelor establish new qualified level3

or new storage conditionsor reduce shelf life.

NO

NO

YES

YES

1 Relevant batches are those from development, pilot and scale-up studies.2 Refer to Decision Tree 1 for information regarding A and B.3 Refer to ICH Guideline on Impurities in New Drug Products.

DECISION TREE #3: SETTING ACCEPTANCE CRITERIA FOR DRUG SUBSTANCE PARTICLE SIZE DISTRIBUTION

Is the drug product a soliddosage form or liquid containing undissolved

drug substance?

No drug substance particlesize acceptance criterion required for solution dosage forms.

1. Is the particle size critical to dissolution, solubility, or bioavailability?2. Is the particle size critical to drug product processability?3. Is the particle size critical to drug product stability?4. Is the particle size critical to drug product content uniformity? 5. Is particle size critical for maintaining product appearance?

Set Acceptance Criterion

No Acceptance Criterion Required

If YES to any

If NO to all

NO

YES

DECISION TREE #4: INVESTIGATING THE NEED TO SETACCEPTANCE CRITERIA FOR POLYMORPHISMIN DRUG SUBSTANCES AND DRUG PRODUCTS

Drug Substance

1.

2.

2.

3.

NO

YES

Conduct polymorphismscreen on drug substance. No further action

Candifferent polymorphs

be formed?

GO TO

NO

NO

YES

Do theforms have

different properties?(solubility, stability,

melting point)

Is drugproduct safety,performance or

efficacy affected?

GO TO

Characterize the forms:e.g., - X-ray Powder Diffraction - DSC / Thermoanalysis - Microscopy - Spectroscopy

No further test oracceptance criterionfor drug substance

YES

Set acceptance criterionfor polymorph content

in drug substance

3.

DECISION TREE #4: INVESTIGATING THE NEED TO SET ACCEPTANCE CRITERIA FOR POLYMORPHISMIN DRUG SUBSTANCES AND DRUG PRODUCTS

Drug Product - Solid Dosage Form or Liquid Containing Undissolved Drug Substance

N.B.: Undertake the following processes only if technically possibleto measure polymorph content in the drug product.

YES

NO

NO

YES

Doesdrug product

performance testingprovide adequate control if polymorph ratio changes

(e.g., dissolution)?

Establish acceptance criteriafor the relevant performance test(s).

Monitor polymorph form duringstability of drug product.

No need to set acceptance criteriafor polymorph change in drug product.

Does achange occur

which could affect

safety or efficacy?

Establish acceptance criteriawhich are consistent with

safety and/or efficacy.

DECISION TREE #5: ESTABLISHING IDENTITY, ASSAY AND ENANTIOMERIC IMPURITY PROCEDURES FOR CHIRAL

NEW DRUG SUBSTANCES AND NEW DRUG PRODUCTSCONTAINING CHIRAL DRUG SUBSTANCES

YES

AND RACEMIC

Consider the need forverifying chiral identity indrug substance releaseand/or acceptance testing.

Is the newdrug substance

chiral1?

Chiral identity, assay and impurity procedures

are not needed.

YESAND ONE ENANTIOMER

Needed for drug substance specification:2 -chiral identity3

-chiral assay4

-enantiomeric impurity5

Needed for drug product specification6: -chiral assay4

-enantiomeric impurity5

1 Chiral substances of natural origin are not addressed in this Guideline.

2 As with other impurities arising in and from raw materials used in drug substance synthesis, controlof chiral quality could be established alternatively by applying limits to appropriate starting materials or intermediates when justified from developmental studies. This essentially will be the case when there are multiple chiral centers (e.g., three or more), or when control at a step prior to production of the final drug substance is desirable.

3 A chiral assay or an enantiomeric impurity procedure may be acceptable in lieu of a chiral identity procedure.

4 An achiral assay combined with a method for controlling the opposite enantiomer is acceptable in lieu of a chiral assay.

5 The level of the opposite enantiomer of the drug substance may be derived from chiral assay data or from a separate procedure.

6 Stereospecific testing of drug product may not be necessary if racemization has been demonstrated to be insignificant during drug product manufacture and during storage of the finished dosage form.

NO

Is t

he d

rug

subs

tanc

e/ex

cipi

ent

ster

ile?

Doe

s dr

ug s

ubst

ance

/exc

ipie

ntsy

nthe

sis/

proc

essi

ng in

volv

est

eps

whi

ch in

here

ntly

re

duce

mic

roor

gani

sms?

Doe

s sc

ient

ific

evid

ence

dem

onst

rate

tha

t r

educ

tion

step

s re

sult

in m

icro

orga

nism

leve

ls

< a

ccep

tanc

e cr

iteria

lim

its (

and

the

abse

nce

of

com

pend

ial i

ndic

ator

org

anis

ms)

in t

he d

rug

subs

tanc

e/ex

cipi

ent?

Pro

vide

sup

port

ing

data

. M

icro

bial

lim

its a

ccep

tanc

e cr

iteria

and

tes

ting

may

not

be

nece

ssar

y.

No

furt

her

mic

robi

al li

mits

tes

ting

or

acce

ptan

ce c

riter

ia a

re n

eces

sary

.

Tes

t ea

ch lo

t fo

r m

icro

bial

lim

its

and

free

dom

fro

m c

ompe

ndia

l in

dica

tor

orga

nism

s.

Tes

t lo

ts o

n a

skip

-lot

basi

s fo

r m

icro

bial

lim

its a

nd f

reed

om f

rom

com

pend

ial i

ndic

ator

org

anis

ms.

Est

ablis

h m

icro

bial

lim

it ac

cept

ance

crit

eria

as p

er t

he h

arm

oniz

ed p

harm

acop

eial

m

onog

raph

.

DE

CIS

ION

TR

EE

#6:

MIC

RO

BIO

LOG

ICA

L Q

UA

LIT

Y A

TT

RIB

UT

ES

OF

DR

UG

SU

BS

TA

NC

E A

ND

EX

CIP

IEN

TS

Is t

he d

rug

subs

tanc

e/ex

cipi

ent

capa

ble

of s

uppo

rtin

g m

icro

bial

gr

owth

or

viab

ility

?

Are

mon

itorin

g m

icro

orga

nism

/indi

cato

r le

vels

co

nsis

tent

ly b

elow

acc

epta

nce

crite

ria

leve

ls?

Pro

vide

sup

port

ing

data

. M

icro

bial

lim

its a

ccep

tanc

e cr

iteria

and

tes

ting

may

not

be

nece

ssar

y.

NO

NO

NO

NO

NO

YE

S

YE

S

YE

S

YE

S

YE

S

Est

ablis

h m

icro

bial

lim

it ac

cept

ance

cr

iteria

as p

er t

he h

arm

oniz

ed p

harm

acop

oeia

l m

onog

raph

.

DECISION TREES #7: SETTING ACCEPTANCE CRITERIA FOR DRUG PRODUCT DISSOLUTION

1. What type of drug release acceptance criteria are appropriate?

Is the dosageform designed to produce

modified release?

YES

Establish drug release acceptance criteria.Extended release: multiple time pointsDelayed release: two stages, parallel

or sequential

Is drug solubilityat 37 ± 0.5°C high throughoutthe physiological pH range?(Dose/ solubility < 250 mL

(pH 1.2 - 6.8))

NO

Continued on next page.

Generally single-point dissolutionacceptance criteria with a lower limitare acceptable.

Is dosage formdissolution rapid?

(Dissolution > 80% in 15 minutesat pH 1.2, 4.0, and 6.8)

Has a relationship beendetermined between disintegration

and dissolution?

Generally disintegration acceptance criteria with an upper time

limit are acceptable.YES

NO

YES

YES

NO

NO

DECISION TREES #7: SETTING ACCEPTANCE CRITERIA FOR DRUG PRODUCT DISSOLUTION

2. What specific test conditions and acceptance criteria are appropriate? [immediate release]

Doesdissolution significantlyaffect bioavailability?

(e.g., have relevant developmentalbatches exhibited unacceptable

bioavailability?)

Attempt to develop test conditions and acceptance criteria which can distinguish batches

with unacceptable bioavailability.

YES

NO

YES

NO

YES

NO

Do changes informulation or

manufacturing variables affect dissolution?

(Use appropriate ranges.Evaluate dissolutionwithin pH 1.2 - 6.8)

Are these changes controlledby another procedure and acceptance

criterion?

Adopt appropriate test conditionsand acceptance criteria without

regard to discriminating power, topass clinically acceptable batches.

Adopt test conditions and acceptance criteriawhich can distinguish these changes.

Generally, single point acceptance criteria are acceptable.

DECISION TREES #7: SETTING ACCEPTANCE CRITERIA

FOR DRUG PRODUCT DISSOLUTION

3. What are appropriate acceptance ranges? [extended release]

YES

NO

NO

YES

YES

NO

YES

NO

Are bioavailabilitydata available for batches

with different drug release rates?Is drug release independent of

in vitro test conditions?

Can an in vitro / in vivorelationship be established?

(Modify in vitro test conditionsif appropriate.)

Use all available stability, clinical, andbioavailability data to establish appropriate acceptance ranges.

Use the in vitro / in vivocorrelation, along with

appropriate batch data, to establish acceptance ranges.

Are acceptanceranges >20% of the

labeled content?

Provide appropriatebioavailability data

to validate theacceptance ranges.

Finalize acceptance ranges.

Is t

he d

rug

prod

uct

a dr

y do

sage

for

m

(e.g

. so

lid o

ral o

r dr

y po

wde

r)?

Do

prod

uctio

n lo

ts c

onsi

sten

tly m

eet

mic

robi

al li

mits

acc

epta

nce

crite

ria?

Est

ablis

h pr

eser

vativ

e ch

emic

al a

ccep

tanc

e cr

iteria

and

pe

rfor

m p

rese

rvat

ive

effe

ctiv

enes

s va

lidat

ion

of p

rodu

ct

cont

aini

ng le

ss t

han

or e

qual

to

the

min

imum

spe

cifie

d pr

eser

vativ

e co

ncen

trat

ion,

or

dem

onst

rate

the

inhe

rent

an

timic

robi

al a

ctiv

ity o

f th

e dr

ug p

rodu

ct.

Est

ablis

h m

icro

bial

lim

it ac

cept

ance

crit

eria

as p

er t

he h

arm

oniz

ed p

harm

acop

oeia

l m

onog

raph

.

Per

form

mic

robi

al li

mits

tes

ting

on a

lo

t-by

-lot

basi

s.

Mic

robi

al li

mits

acc

epta

nce

crite

ria a

nd t

estin

g m

ay n

ot b

e ne

cess

ary.

Doe

s th

e dr

ug p

rodu

ct c

onta

in

antim

icro

bial

pre

serv

ativ

es o

r po

sses

s in

here

nt a

ntim

icro

bial

ac

tivity

?

Per

form

ski

p-lo

t te

stin

g fo

r m

icro

bial

lim

its,

or p

rovi

de s

cien

tific

just

ifica

tion

for

no r

outin

e m

icro

bial

lim

its t

estin

g.

DE

CIS

ION

TR

EE

#8:

MIC

RO

BIO

LOG

ICA

L A

TT

RIB

UT

ES

OF

NO

N-S

TE

RIL

E D

RU

G P

RO

DU

CT

S

YE

SN

O

YE

S

YE

S

NO

NO

Doe

s sc

ient

ific

evid

ence

dem

onst

rate

grow

th in

hibi

tory

pro

pert

ies

of t

he

drug

pro

duct

?

YE

S

NO