Paving the Road Toward Setting Clinically Relevant Drug Product - PowerPoint Presentation

Slide1

Paving the Road Toward Setting Clinically Relevant Drug Product Specifications: Biopharmaceutics Perspective on Information Needed, Approach, and Criteria

Sandra Suarez Sharp, Ph. D.FDA/OPQ/ONDP/Division of BiopharmaceuticsDISSOLUTION AND TRANSLATIONAL MODELING STRATEGIES ENABLING PATIENT-CENTRIC PRODUCT DEVELOPMENT Baltimore, MD May 17, 2017Slide2

OUTLINEDefinition of Clinically Relevant Drug Product Specifications (CRDPS)Data neededThe role of QbD implementation as a platform for enabling the establishment of CDRPSThe importance

of stablishing a link between product quality attributes, dissolution and in vivo performanceThe role of biopharmaceutics tools: dissolution and BA/BEApproaches/Criteria in setting CRDPSCurrently used approaches and emerging toolsAdvantages and disadvantagesConcluding remarksSlide3

Clinically Relevant SpecificationsClinically relevant means that the proposed criteria/limits of the identified critical quality attributes/parameters have been justified/set based on the understanding of its in vivo impact (systemic exposure, safety and/or efficacy)Slide4

Quality Target Product Profile (QTPP)A prospective summary of the quality characteristics of a drug product that ideally will be achieved to ensure the desired quality (in vitro/in vivo performance)Guide to establish product design strategy and keep product development effort focused and efficientSlide5

What Does QTPP Include?Intended use in clinical settingRoute of administration, dosage form (delivery systems), and container closure system

Drug substance quality attributes PSD, polymorphism, aerodynamic performance, etc.Drug product quality attributesAppearance, Identity, Strength, Assay, Uniformity, Purity/Impurity, Dissolution, Stability

, and othersSlide6

Implementation of QbD

Prior Knowledge

Risk Assessment

DOE

Process Analytical Technology

QbD Tools

Constitute the basis for enhanced

product and process understanding

including

process

control

.Slide7

QTTP

Risk Assessment

DoE

CMAs, CPPs

Drug

Product

Specifications

CQAs

Design Space

Dis

s

Paving the RoadSlide8

Setting CRDPS: Implies the Establishment of a Bridge

Dissolution

(CQA)

IN VIVO

PERFORMACE

Water

content

Particle

size

compression

blending

Without the use of dissolution it would be rather impractical to verify every variation in the CMAs/CPPs through the conduct of efficacy and safety and/or BA/BE studies.

The development of a dissolution method that is not only discriminating but also biopredictive becomes criticalSlide9

What Are the Consequences of the Lack of an In Vivo Link? CRDPS can still be established in the absence of a direct link (i.e. via PK studies)

However, without understating the relationship between variations in the quality attributes and clinical outcome for BCS class 2/4 drugs, drug product acceptance criteria limits may be overly wide, unnecessarily tight or completely irrelevant to clinical performanceAs such, variations in the identified critical quality attributes, which are within

the approved limits may pose unknown

, but

potentially significant

risks

to patients

Adapted from S. Suarez

,

The Relevance of Biopredictive Dissolution Testing

. DIA 2016,

Annual meeting

D

rug

products comprising BCS class

1/3 compounds that do not meet the criterion of very rapid/rapid dissolution may also pose a riskSlide10

….. “determining

the relationship between critical manufacturing variables

and a response surface derived from an

in vitro

dissolution

profile and an

in vivo bioavailability

data set.

……. the

goal is to

develop product

specifications that will ensure bioequivalence of future

batches”……

Recognition of Linking Quality

to

In

V

ivo Performance:

B

ack in 1997

Suarez, S.

FDA Experience on the Application of Modeling and Simulation in Setting Clinically Relevant Drug Product Specifications, PQRI 2017Slide11

Applicant's Proposed Dissolution Acceptance Criterion, Another StoryItems

Dissolution at 30 minTotal number of batches34aAverage

91.2 %

Standard

Deviation (SD)

3.7%

x-k*s

(average -2.549* SD)

81.8%

Calculated Q value

Q(%)=x-ks-5

76.8%

Proposed

Q value

75%

Example of dissolution acceptance criterion setting

proposal

for a BCS 2 drug based on a parametric approach rather than a response surface methodology linking in vitro to in vivo performanceSlide12

What Are the Challenges in Setting CRDPS?Lack of efficient strategies for the development of

in vitro dissolution methods that are predictive of in vivo performanceNo inclusion of dissolution testing in DoE studies for the selection of CMAs and CPPs and for the verification of design spaces rangesChallenges in conducting dedicated in vivo PK studies with aberrant product variantsAn ideal approach in setting CRDPS is via IVIVC or IVIVR implementation. However, the

overall acceptance rate of the

IVIVC

submissions is about 40%. Moreover, the number of IVIVC studies seen in the submissions per year is not

increasing*

S. Suarez, et. al, Regulatory Experience with

In Vivo In Vitro

Correlations (IVIVC) in New Drug Applications, AAPS J. Aug. 2016.Slide13

Approaches for Establishing CRDPS: Where Are We N

ow?

Approach 1:

Limits established based on the characteristics of batches tested in pivotal phase 3 clinical trials

Approach 2:

L

imits established based on a range of release characteristics resulting in BE

Approach 3

Limits

established based on predictive and robust in vivo in vitro correlations

S. Suarez, Establishing clinically relevant drug product specifications. AAPS 2012, Annual meetingSlide14

14

Design

Space (DS)

Operating

Space

How are the DS/ CRDPS ranges established using the approaches described before?

The QbD Paradigm

All the batches within the design space are

BE,

and PK studies to

demonstrate

BE are not necessary. Slide15

Approaches for Establishing CRDPS

APPROACH 1

No PK data available

linking variations in CMAs/CPPs to dissolution/clinical performance

Ranges

established based on batches tested in pivotal phase 3 clinical trials

and similarity testing

Clinically relevance

assured

for

drug products comprising

BCS class 1 /

3

compounds

Clinically relevance

not always assured

for

drug products comprising

BCS class

2 /4

compounds

No regulatory flexibility for BCS Class 2/4 drugs as ranges are established based on ranges/ in process controls from batches tested in Pivotal Phase 3 TrialsSlide16

Approaches for Establishing CRDPS, Cont.

APPROACH 2

(IVIVR/Response surface Methodology)

PK data available

linking variations in CMAs/CPPs to dissolution/clinical performance

rank order relationship

Regulatory flexibility limited to the

extremes of dissolution profiles for batches that were

BESlide17

17

Manufacture product variants with different release characteristics

Select Optimal dissolution method

with adequate discriminating power

Determine dissolution rates resulting in

similar in vivo performance

CRDP limits established based on a range of release characteristics resulting in bioequivalence

Determine bioavailability for

product variants

Approach 2, Cont.Slide18

Approaches for Establishing CRDPS, Cont.

APPROACH 3

(IVIVC)

PK data available

linking variations in CMAs/CPPs to dissolution/clinical performance

rank order relationship resulting in either:

Level A IVIVC

Multiple Level C IVIVC

CRDP limits established

based on

model predictions

Regulatory flexibility limited

by

ranges in release rate used in the construction of the

model (no extrapolation)

Multiple Level C IVIVC may not be applicable to waive major CMC changes requiring BE, but could be useful in supporting the establishment of CRDPS Slide19

Drug

Product

Specifications

Relevant

Clinically

Are We

T

here

Y

et? What

Are

the

Current

E

merging

T

rends

?

19Slide20

The Utility of In Silico PBPK M&S

Prior knowledge (product properties and API physicochemical properties), BA/BE data from phase 1/Initial Risk Assessment

Conduct deconvolution of concentration-time profiles

Use

in silico

PBPK model to predict concentration-time profile with

dissolution

profile as an input

Develop a dissolution method (DM) that mimics the deconvoluted profile

Use the DM as an endpoint in the DoE studies

Determine BA for all product variants (e.g. changes in PSD)

Established IVIVC/R

CRDPS Ranges chosen to ensure similar (BE) product performance

Adapted from Suarez, S.

Strategies for developing dissolution test methods fitted for purpose

. AAPS Annual Meeting 2015Slide21

Time (min)

Dissolution (%)

CRDP ranges: Pass

similarity testing (Approach 1)

Clinical batch

CRDP

ranges:

Slowest releasing

Profile batch and clinical batch are BE

(Approach 2)

Failed BE and virtual BE

CRDP

ranges:

Slowest

releasing

profile

batch and

clinical

batch pass

virtual BE (PBPK/extrapolation Approach)

In the presence of acceptable IVIVC,

CRDP

limits established based

on

model

predictions (Approach 3)

INCREASED

REGULATORY

FLEXIBILITY

CRDPS and Regulatory FlexibilitySlide22

Interdisciplinary Integration: Key in Setting CRDPSSlide23

23

Dissolution

Model

IVIVC/R

Model

Convolution

Model

Multivariate

Model

Impact on

BA/BE

Risk assessment

to identify

Manufacture

Parameters,

Material

Attributes

PK Model

PD Model

PK/PD model

Inefficacy

PK/PD model

toxicity

PK/PD Model

Range of specifications

chosen to ensure

targeted performance

RTRT

Impact on

Dissolution

PBPK

Model

DoE studies

API

physicochemical

properties

Interdisciplinary Integration: Key in Setting CRDPSSlide24

Concluding RemarksAre we there yet?

Setting CRDPS for BCS class 2/4 drugs requires establishing a link to in vivo performanceChallenges with the conduct of dedicated PK studiesThe success rate of IVIVC is lowPBPK absorption M&S is a promising but underused tool for risk

assessment

and

establishing

CRDPS

What

is the benefit for Industry?

What to expect in the near future?

Advancement

in vivo predictive

dissolution and IVIVC

combined with

PBPK absorption models

provides an opportunity for taking a major step in model based drug

development

and to support regulatory flexibility in drug product specifications

Use of new approaches in the development of IVIVC

Use

of mechanistic PBPK

absorption

models

to:

Guide the development of a biopredictive dissolution method

Increase

the rate of success of IVIVC

Risk assessment

for the

selection

of CMAs and CPPs

Verification

of the

Design Spaces

With the ultimate goal to setting CRDPSSlide25

OPQ Values

Put

patients first

by balancing risk and

availabilitySlide26

AcknowledgmentsPaul Seo, Ph.D. Division Director, Division of Biopharmaceutics, ONDP/OPQRamesh Sood, Senior Scientific Advisor, ONDP/OPQ

Sarah Pope-Miksinski, Office Director, ONDP/OPQSlide27