In Vitro-In Vivo Correlation:

Linking Drug Release to Clinical

Performance

Yihong Qiu, Ph.D.

Abbott Laboratories
51st Land O' Lakes Conference: Bridging Material and
Product Quality in Developing Tablet Dosage Forms

June 4, 2009

Outline
 Introduction
– In vitro-in vivo correlation (IVIVC)
 IVIVC and Product Quality
Q y
– Clinical linkage of in vitro drug release
– Product and process development, understanding, manufacturing,
control and changes
 Challenges and Considerations in Developing IVIVC
– In vitro vs. in vivo
– Drug, formulation and dissolution characteristics
• BCS and delivery technology
• Product design: IR, MR, FDC
– IVIVC and BA/BE studies
 Summary

June 4, 2009

1
Introduction
In Vitro-In Vivo Correlation (IVIVC)

• A predictive mathematical model describing the relationship between

an in
i vitro
it property t (usually
( ll extent
t t or rate
t off d
drug release)
l ) and
da
relevant in vivo response (e.g., plasma concentration or amount
of drug absorbed)
50
120

• Type of IVIVC 100 40

% Dissolved (or F in vitro)

Plasma Concentration (mg/L

80
30
60

Fast-A 20
40

– Level A: Profile correlation

Medium-B
20 Slow-C 10

0
0 4 8 12 16 20 24 0
Time (h) 0 12 24 36 48 60 72
Time (h)

– Level B: Summary parameter correlation

– Level
L l C:
C Characteristic
Ch t i ti parameter
t correlation
l ti
– Multiple Level C: Multiple parameter correlation

* FDA Guidance for Industry: Extended release oral dosage forms: Development,
evaluation, and application of in vitro/in vivo correlation. 09/1997

June 4, 2009

Illustration: Level A IVIVC

In vitro drug release Observed or predicted Cp profiles
120.0

100.0

80.0
Biostudy
eleased

60.0
% Re

A
40.0
B
20.0
C
0.0
0 5 10 15 20 25
T(hr)

Disposition Function
IV or IR Cp profiles
Deconvolution 1.4

Modeling 1.2 IR
Plasma Conc.

1 IV
0.8

relationship C(t) = f(t)*C(t) 0.6

0.4

IVIVC model 0.2

0
0 4 8 12 16 20 24

110.0
IVIVC plot
Convolution Time (hr)

100.0
90.0
80.0
Estimated in vivo input
70.0
In vivo
In vivo

60.0 A
1.000
50.0
40.0 B 0.800
% Released

30.0
20.0
C 0.600
10.0
0.0
0.400
0.0 10.0 20.0 30.0 40.0 50.0 A
In vitro
(1) Validation 0.200
B
C
-

(2) Prediction 0 5 10
Time (hr)15 20

Y. Qiu. In: Developing Oral Dosage Forms: Pharmaceutical Theory and Practice. Edited by Y. Qiu
June 4, 2009 et al.. Academic Press, San Diego, CA. 2009. pp-379-408

2
Illustration: Multiple Level C IVIVC
In vitro parameters: Q10, Q25, Q60
In vivo parameters: Cmax, AUC, (Cmax/AUC)
In vitro
120
IVIVC model
100

80
20.00 20.00
% Dissolved

18.00
60 18.00
16.00
16.00
A 14.00
40 14.00

Cmax, AUC
M1 12.00

Cmax, AUC
B 12.00
10.00
20 M2 10.00
8.00
C 8.00
6.00
- 6.00 Cmax
4.00
- 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0
4.00
Cmax
Tim e (hr)
2.00 AUC
2.00 AUC
-
- - 10 20 30 40 50 60 70 80 90 10
- 10 20 30 40 50 60 0
Biostudy Q(10min) Q(25min)

14
In vivo 20.00 0.75
18.00 Q10
12 0.70
16 00
16.00 Q25
10 14.00 0.65 Q60

Cmax, AUC

Cmax/AUC
12.00
0.60
8
10.00
Cp

A 8.00 0.55
6
M1
6.00 0.50
4 B Cmax
4.00
M2 AUC 0.45
2.00
2 C
- 0.40
30 40 50 60 70 80 90 100 110 - 20.00 40.00 60.00 80.00 100.00
0
0 4 8 12 16 Q(60min)
Q(10min), Q(25min), Q(60min)
Tim e (hr)

June 4, 2009

IVIVC and Product Quality

 Critical Quality Attributes (CQA)
– Physical, chemical, biological and microbiological property that must be
controlled to ensure product quality
• T i l ffor solid
Typical lid products:
d t P Purity,
it potency,
t stability
t bilit and
d drug
d release
l

 IVIVC: Clinical linkage

– Use in vitro test to predict in vivo performance of dosage forms
• Most common and feasible: Drug release (arbitrary  biorelevant  predictive)

– Bridge a critical gap between product CQAs and clinical performance

• Establish in vitro dissolution as one of the most important CQAs
– Ser e as a critical tool for prod
Serve product
ct and process understanding
nderstanding
– Aid product/process development, manufacturing and control

– Provide significantly increased assurance for consistent product quality and

performance under QbD
• Predict and control clinical performance within the life cycle of a product

June 4, 2009

3
 IVIVC and Product Quality
 With a validated IVIVC
– Prior to product submission
• Facilitate application of QbD principles in product development
• Guide product and process design and reduce development time/cost
• Aid scale-up, optimization and risk management
– Assess and define CPP, design space, risks, control strategy, etc
• Assure product quality by setting meaningful specifications

– Post product approval

• Justify waiver of in vivo BE studies
– Support changes (e.g., SUPAC), variations; Reduce regulatory burden
– e.g., IR products (BCS 2, 3, 4 drugs); MR products
• Ensure consistent quality and performance during commercial production
– Planned or unexpected changes/variations of raw materials, composition, process, site,
equipment, etc.

June 4, 2009

IVIVC and Product Quality

 Without a validated IVIVC
 With IVIVR Illustration of mapping in vitro
and in vivo results
• Map
p in vitro and in vivo p
performance 3.00
Formulation 1
Plasma Conc. (mcg/ml)

2.50 Formulation 2
• During development 2.00
Formulation 3
Formulation 4

o Guide formulation and process 1.50 Formulation 5

1.00
screening and understanding
0.50

o Justify biorelevant specifications 0.00

0 10 20 30 40
Time (hr)
• Post-approval
100
Bioequivalent
o Ensure consistent quality of 95 Non-bioequivalent

commercial products
Bioavailabilitty

90

85
o Justify biowaiver 80
Release

 Without IVIVR 75

70
Limits

50 60 70 80 90 100
• Lower confidence in using in vitro test In Vitro Drug Release

for assuring product quality

June 4, 2009

4
Developing IVIVC
 Methodology
– IVIVC modeling, evaluation and applications
• Primary focus of most IVIVC colloquiums and publications over the
years
• FDA guidance and EMEA NfG published in 1997 and 1999,
respectively (CPMP/QWP/604/96)
• Premise: available in vitro and in vivo data appropriate for
establishing IVIVC

 Data
– Obtaining suitable in vitro and in vivo data is not a given
– Understanding and appropriate use of the data

June 4, 2009

Challenges in Developing IVIVC

 In vitro vs. in vivo

Simple, static and controlled Complex, dynamic and variable

June 4, 2009

5
In Vivo Drug Release/absorption
from Dosage Forms in the GI tract
 Estimated from availability of the drug in systemic circulation
(high variability)
 Data = f ((drug
g properties,
p p formulation, biological,
g
physiological variables, and their interactions)
• Drug’s physicochemical and biopharmaceutics properties
o Solubility, dose, lipophilicity, permeability, ionization, physical and chemical
stability, pH-dependency, IDR, solid phase, surface area, wetability, etc.

• Biological and physiological variables

o Transport mechanism, metabolism, transporters (absorptive, secretive),
regional
g difference,, motility,
y, shear force,, residence time,, food,, lumen
contents, secretion, enterohepatic recycling, surface area, fluid volume,
microflora etc.

• Formulation design
o Dosage form type, size, release mechanism, sensitivity to environmental
changes, drug release kinetics and duration etc.

June 4, 2009

In Vitro Test
 Standard method
• Pharmacopoeial methods

 Attempt to match in vivo data/simulate one or more specific GI

conditions
• Modified standard methods (hydrodynamic, shear, food…)
– e.g., paddle+polystyrene beads; Milk; FeSSIF/FaSSIF; Two-phase;
Stationary basket+Paddle; Ex-vivo fluid (aspirated human IF);…
• New models (motility, transit, secretion, food, …)
– e.g., Multi-vessel; Multi-compartment (TNO); Rotating dialysis cell;
Flow-through cell drop method;…

 Data = f (test method and parameters, drug properties,

formulation)
June 4, 2009

6
Considerations in Developing IVIVC
 Critical in achieving IVIVC
– Feasibility
• Understand drug g properties,
p p formulation characteristics, and their
interplays with GI environment

– Essential Condition
• Apparent in vivo absorption
 Dissolution rate limited
• Relevant in vitro dosage form attribute
 Dissolution

• Formulations
 Different in vivo performance
• In Vitro Test
 Differentiating (IVIVA, IVIVR); Predictive (IVIVC)

June 4, 2009

Immediate Release (IR) Dosage Forms:

General Considerations
 Generally more difficult to achieve IVIVC

 Feasibility of IVIVC: API dependent

– Apparent absorption

• Mostly occurs in the upper intestine (often a function of many

potentially confounding variables)

• Short absorption phase in most cases (difficult to characterize for

L
Level
l A IVIVC)

• Parameters amenable to Level B, C, Multiple Level C IVIVC

June 4, 2009

7
IR Dosage Forms: Drug Property Considerations
 BCS Class II
– IVIVC Possible: Dissolution rate or solubility limited absorption; Impact of formulation/process
– In vitro data influenced by need for sink condition, surfactant, volume of test medium, etc.

 BCS Class I
– IVIVC is less likely, except for:
• Dissolution rate limiting due to formulation/process
• BCS Borderline API

 BCS Class III

– IVIVC rare: Gastric emptying and/or permeability is usually the rate-controlling step

 BCS Class IV
– IVIVC is less likely: Significant competing or rate-limiting processes other than dissolution:
parallel pathways, metabolism, non-linearity, etc.
– Opportunity for IVIVR or IVIVC may exist, e.g.,
• Both dissolution and permeability may limit the rate of in vivo absorption
• BCS borderline API (e.g., metabolism)

June 4, 2009

Extended-Release (ER) Dosage Forms:

General Considerations
 Drug release from dosage form controls drug input in the GI
tract
– Generally desirable and more likely to obtain IVIVC
– Feasibility of IVIVC: drug molecule dependent

 Apparent absorption
– Mostly occurs in small intestine and ascending colon or
throughout
– Longer absorption phase amenable to developing Level A, B, C or
Multiple Level C IVIVC
– Generally
G ll hi
higher
h variability
i bilit ddue tto
• A wide range of drug release and absorption environment during the
traverse of the dosage form through the GI tract
– Often influenced (confounded) by multiple variables and their
interactions that are different from the “unit impulse input”

June 4, 2009

8
ER Dosage Forms: Drug Property Considerations
 IVIVC: Possible
– BCS Class I
• Dissolution rate limiting
– BCS Class II
• Dissolution rate limiting:
• Release (metering) vs. dissolution (dose,/solubility, mechanism)

 IVIVC: Rare
– BCS Class III and IV
• Permeability rate limiting, competing processes, absorption window etc.
• Relative rate of release to permeation (release duration, region dependent)
• Not feasible for ER development in most cases (e.g., due to region-dependent
transport)

June 4, 2009

ER System: Delivery Technology Considerations

Dosage form behavior and IVIVC depends on drug property, technology
and formulation design
• Osmotic Pump system
– In vitro release generally insensitive to test conditions
– Higher probability to obtain IVIVC
– However
• In vivo results depend on APIs (e.g., food effect)
• Lack of flexibility to adjust test condition to match in vivo performance
• Reservoir system
– In vitro release typically sensitive to in vitro test conditions
– In vivo results depend on drug property and formulation design (e.g., food effect)
– Possible to adjust test condition for obtaining IVIVC
• Matrix system
– Hydrophilic matrix: Gel strength and system integrity also affect rate and mechanism of drug release
– In vitro release sensitive to in vitro test conditions
– In vivo results depend on drug property and formulation design (e.g., food effect)
– Possible to adjust test condition for obtaining IVIVC

June 4, 2009

9
Additional Considerations: Drug Property,
Product Design and Release Characteristics
 Release characteristics
• e.g., IR/ER, ER/DR, DR/ER, FDC, …
 Formulation design
g and drug
gpproperty
p y ((Pchem,, Biopharm,
p , PK))
• Feasibility and IVIVC may vary with different data segment and/or API
• Setting specification (impact on Cmax and AUC)
Illustration: IR+ER
120 Illustration: DR+ER or ER+DR
120

100
100

80
80
eased

ased
60
% Rele

% Relea
60
IR+ER
40 LL1
40 DR+ER
UL1
LL2 ER+DR
20
UL2 20 LL
ER UL
0 0
0 2 4 6 8 10 12 14 16 18 20 0 2 4 6 8 10 12 14 16 18
Tim e (hr) Tim e (hr)

June 4, 2009

IVIVC and BA/BE Studies

 Development Approach
 Retrospective
– Review historical data from development BA/BE studies
– Evaluate IVIVC
– Determine the need and timing of an IVIVC study if necessary
 Prospective (or concurrent, proactive)
– Plan for IVIVC investigation at the start of a project
• e.g., for BCS II drugs, MR delivery
– Utilize data from the development BA/BE studies and start
exploring
l i IVIVC
• If IVIVC exists, establish and validate as early as possible to facilitate
product development (time, cost and resource)
• If IVIVC doesn’t exist, adjust strategy of development (risk assessment,
activities and timeline)

June 4, 2009

10
Summary
 IVIVC
– Linking drug release to in vivo performance: Highly desirable
• ER product: A regulatory expectation
• IR product: Very useful if exists
– Facilitate rational development of product and process
– Ensure product quality
– Offer opportunities for regulatory flexibility

 IVIVC needs to be explored and developed on a case-by-case basis

– There is no universal in vitro model

 IVIVC development
d l t
– Modeling is only part of the key IVIVC components
– Understanding API, formulation and biopharmaceutics is equally or more important
– Proactive development can maximize opportunity and development efficiency

June 4, 2009

11