Implementation of ICH Q8, Q9, Q10

Regulatory
Assessment

International Conference on Harmonisation of Technical

Requirements for Registration of Pharmaceuticals for Human Use
 ICH Quality Implementation Working Group - Training Workshop

Regulatory Assessment

Presentation Overview
• Goal of Regulatory Quality Assessment
• Review of the case study
- Considerations during regulatory evaluation
- Areas of consideration by assessors will be presented in the form
of questions for the assessor
- The questions presented here are not necessarily the ones which
are finally communicated in regulatory deficiency letters
- API and Formulation
- Manufacturing Process Development
- Quality Risk Management
- Design Space
- Proposed Control Strategy
and Real Time Release Testing
- Assessors - Inspector Interaction
© ICH, November 2010 slide 3
 ICH Quality Implementation Working Group - Training Workshop

Regulatory Assessment

Goal of Regulatory Quality Assessment

• Assess
-That the product is capable of consistently meeting the
required quality
- That the manufacturing process is capable of producing
quality product
- That throughout product shelf life and life cycle commercial
batches will link to clinical batches in all relevant aspects
• These can be accomplished by
- Process development and control strategy according to
traditional standards
- Process development and control strategy according to
new paradigm

© ICH, November 2010 slide 4

 ICH Quality Implementation Working Group - Training Workshop

Regulatory Assessment

Principles of Assessment
• Assessment principles are the same regardless of the
development approach
• Meet Quality Target Product Profiles (QTPPs)
• Areas of assessment:
- API
- Formulation
- Manufacturing process
- Control strategy
- Analytical Procedures
- Stability
© ICH, November 2010 slide 5
 Implementation of ICH Q8, Q9, Q10

Regulatory Assessment

API and Formulation

International Conference on Harmonisation of Technical

Requirements for Registration of Pharmaceuticals for Human Use
 ICH Quality Implementation Working Group - Training Workshop

Regulatory Assessment

API General Considerations

• QbD principles apply to APIs
• QbD principles can guide manufacturing process
design and control strategy development
• Design space can be developed for API processes

© ICH, November 2010 slide 7

 ICH Quality Implementation Working Group - Training Workshop

Regulatory Assessment

API- Assessors’ Evaluation

• Have starting materials and process been
adequately described?
• Are there toxicity concerns with degradants and/or
related substances?
• Have adequate specifications and methods been
proposed?
• Have adequate process controls been described?
• Was the design space adequately developed and
data provided to support it?

© ICH, November 2010 slide 8

 ICH Quality Implementation Working Group - Training Workshop

Regulatory Assessment

Formulation - General Considerations

• Design space – formulation aspects
- Variable composition or component attributes
- Based on input raw material attributes
- Lot to lot variability
- Justified by data (Prior knowledge, DoE, etc)
• API attributes
- To be considered in the development of formulation and
choice of dosage form to meet QTPP
- Additional information may be needed for the
development of the formulation e.g. BCS, PK, stability,
excipient compatibility

© ICH, November 2010 slide 9

 ICH Quality Implementation Working Group - Training Workshop

Regulatory Assessment

Assessors’ Evaluation of the Formulation

• Is dosage form designed to meet QTPP?
• Are the roles of ingredients identified?
• Have the safety and compatibility of ingredients been
adequately addressed?
• Is the formulation adequately understood and
specified?
• Does the proposed formulation differ from the
formulation used in the pivotal clinical trials?
© ICH, November 2010 slide 10
 ICH Quality Implementation Working Group - Training Workshop

Regulatory Assessment
Assessors’ Evaluation of the Case
Study Formulation
• Why was Calcium Hydrogen Phosphate Hydrate
chosen with a water sensitive API?
- Concern about compatibility and stability

• Has material variability effects been understood?

- Adequacy of NIR testing
- Adequacy of dissolution model and method

• What is the function of D-mannitol in the formulation?

- Described only as excipient in the case study
- Needs to be further explained

© ICH, November 2010 slide 11

 Implementation of ICH Q8, Q9, Q10

Regulatory Assessment
Manufacturing Process
Development

International Conference on Harmonisation of Technical

Requirements for Registration of Pharmaceuticals for Human Use
 ICH Quality Implementation Working Group - Training Workshop

Regulatory Assessment

Assessment of Manufacturing Process

Development
• Production process description needs to have
sufficient detail to enable assessment
• Assessment should evaluate
- Process design
- Use of risk management processes
including risk assessments
- Design space
- Robustness

© ICH, November 2010 slide 13

 ICH Quality Implementation Working Group - Training Workshop

Regulatory Assessment

Initial Quality Risk Assessment

Tablet Manufacturing Operation
Dr ug Moist u re
su bst a n ce con tent in Blen din g Lu br ica t ion Com pression Coa t in g P a cka gin g
pa rt icle size m a nu fa ct u re
in vivo perfor m a n ce
Dissolu t ion
Assa y
Degr a da t ion
Con t en t u niform it y
Appea r a nce
F ria bility
St a bilit y-chem ica l
St a bilit y-ph ysica l

- Low risk
• Aids assessor in understanding how different aspects of the
- Mediu m risk
- H igh risk
process can affect product quality
• Incorporates known risk factors of drug product – degradation
pathways (e.g., moisture sensitivity), solubility factors, etc.
• Includes effects of unit operations and starting materials
(including excipient properties)
• Atypical or unusual findings should be explained in greater deta

© ICH, November 2010 slide 14

 ICH Quality Implementation Working Group - Training Workshop

Regulatory Assessment

Assessors’ Evaluation of the Risk

Assessment
• Assessors to evaluate methodologies and outcome
- Explanation of risk ranking and score
- Setting of risk threshold
- Assurance that relevant factors have been considered
• Are results consistent with scientific principles and
prior knowledge?
• Was there a linkage of results to the development of
design space and control strategy?

© ICH, November 2010 slide 15

 ICH Quality Implementation Working Group - Training Workshop

Regulatory Assessment

DoE to Support Design Space

• Multifactorial DoE study of
Exp No Run Order API MgSt LubT Hard Diss
1 1 0.5 3000 1 60 101.24
2 14 1.5 3000 1 60 87.99
3 22 0.5 12000 1 60 99.13
4
5
6
8
18
9
1.5
0.5
1.5
3000
12000
12000
10
10
10
60
60
60
86.03
94.73
83.04
variables affecting dissolution
7 15 0.5 3000 1 110 98.07
8 2 0.5 12000 1 110 97.68
9 6 1.5 12000 1 110 85.47
10
11
16
20
0.5
1.5
3000
3000
10
10
110
110
95.81
84.38
• Use an appropriate experimental
design (e.g., some screening designs
12 3 1.5 12000 10 110 81
13 10 0.5 7500 5.5 85 96.85
14 17 1.5 7500 5.5 85 85.13
15
16
17
19
21
7
1
1
1
3000
12000
7500
5.5
5.5
1
85
85
85
91.87
90.72
91.95
cannot determine interactions)
18 4 1 7500 10 85 88.9
19 5 1 7500 5.5 60 92.37
20
21
11
12
1
1
7500 5.5 110 90.95
Scaled & Centered Coefficients for Diss at 60min
7500 5.5 85 91.95
• Provide more relevant
22 13 1 7500 5.5 85 90.86
experimental data and statistical
0

23 23 1 7500 5.5 85 89
-1

-2

-3
analysis for critical unit
operations
%

-4

• Address what parameters

-5

-6

were not varied in the design

MgSt*LubT
MgSt

Hard
API

LubT

API Mg Lubricati Tablet Mg

on St*LubT
Particle Stearate Hardnes
Size
N=23
DF=17 SSA
R2=0.986
Q2=0.981
Blending
R2 Adj.=0.982
RSD=0.725
time
s
Conf. lev.=0.95

MODDE 8 - 2008-01-23 10:58: 52

space experiments

© ICH, November 2010 slide 16

 ICH Quality Implementation Working Group - Training Workshop

Regulatory Assessment

Assessors’ Evaluation of Design Space

• Was a clear description of design space and its intended use
provided?
• Has the proposed design space been appropriately established?
- Demonstrated by data, supporting models and statistical evaluation
- Understanding of interactions of variables
- Multivariate vs univariate studies
- Justified for the intended scale
- Prior knowledge adequately summarised and/or referenced
• How could a design space built around one CQA (e.g particle size),
affect other CQAs?
• Is the design space consistent with the control strategy?

© ICH, November 2010 slide 17

 ICH Quality Implementation Working Group - Training Workshop

Regulatory Assessment

Example from the Case Study:

Crystallization Design Space
• Goals of Crystallization Process
- D90 between 5 – 20 microns
- Target set by dissolution and formulation DoE
- Degradant < 0.3% (qualified)
• Developmental knowledge
- Water during crystallization causes degradation
- Multiple parameters likely to influence PSD during
crystallization

© ICH, November 2010 slide 18

 ICH Quality Implementation Working Group - Training Workshop

Regulatory Assessment

Example from the Case Study:

Crystallization Design Space – Cont.
• Univariate studies explored water content of solvent
at max addition time and max temp
• DoE of 4 parameters established model for PSD:
- PSD D90 = 19.3 - 2.51*A - 8.63*B + 0.447*C -
0.0656*A*C + 0.473*A^2 + 1.55*B^2
- where A = Seed wt%, B = Agitator Tip Speed (m/s)
and C = Temperature (C)
- Statistical analysis shows that crystallization feed
time does not impact PSD across the tested range.

© ICH, November 2010 slide 19

 ICH Quality Implementation Working Group - Training Workshop

Regulatory Assessment

Assessors’ Evaluation of the

Crystallization Design Space
• Was the use of risk management processes acceptable?
- Was adequate information provided?
- Was there an appropriate use of prior knowledge?
- Did the application include the risk assessments for the most
important CQA/process parameter pairs e.g.
Degradation/Crystallization?
• Was it appropriate to do separate studies on formation of
degradant and PSD?
• Are the process parameters ‘scale independent’?
• How can the proposed model be confirmed?
- Case study relied on center point runs at scale

© ICH, November 2010 slide 20

 ICH Quality Implementation Working Group - Training Workshop

Regulatory Assessment

Assessors’ Evaluation of the

Crystallization Design Space – Continued
• Is it appropriate to split out API PSD and impurity profile in risk
assessment (Overall Risk Assessment for Process) ?
- Presented in the case study combined as “In Vivo Performance”
• Should crystallization have been classified as high risk in the
risk assessment for degradation?
• How was process and/or method uncertainty accounted for in
the model?
• Did the design space presented illustrate the interaction of
parameters?
- Case study showed two separate response surfaces for the two
CQAs evaluated

© ICH, November 2010 slide 21

 Implementation of ICH Q8, Q9, Q10

Regulatory Assessment
Proposed Control Strategy
and Real Time Release
Testing

International Conference on Harmonisation of Technical

Requirements for Registration of Pharmaceuticals for Human Use
 ICH Quality Implementation Working Group - Training Workshop

Regulatory Assessment

Assessors’ Evaluation Of the Control

Strategy
• Do the CQAs provide assurance that the QTPP will be met?
• Is the control strategy based on appropriate risk management?
• Is the placement of proposed controls maximally effective?
• Does the description of control strategy include down stream
tests?
• Are the Specifications adequate?
• What functional tests for excipients are needed? Were these
included?
• Assessing some elements of control strategy such as RTRT,
PAT, etc. may require assessors and inspectors with
specialized training

© ICH, November 2010 slide 23

 ICH Quality Implementation Working Group - Training Workshop

Regulatory Assessment

Blending Process Control Options

• Purpose – to assure that the blend is uniform
• Conventional control (option 1)
• RTRT (PAT based) control (option 2)

© ICH, November 2010 slide 24

 ICH Quality Implementation Working Group - Training Workshop

Regulatory Assessment

Blending Control Option 1

• Perform DoE to develop the design space
• CPPs involved – blender type, blending speed,
blending time, API particle size
• Assessors’ evaluation
- Were all CPPs properly identified during QRA?
- Are the reference method and sampling procedure
used to assess the blend uniformity adequate?
- Is the design space developed from the DoE
applicable at commercial scale?

© ICH, November 2010 slide 25

 ICH Quality Implementation Working Group - Training Workshop

Regulatory Assessment

Blending Control Option 2

• Control of blending end-point by NIR
• Includes a chemometric model to predict the end-
point of the process
• Assessors’ evaluation
- Is the model properly developed and validated?
- Do the model predictions correlate with standard blend
uniformity measurements?
- Are all sources of variation (e.g., excipients) included
in the model?
- Is the probe location adequate?
© ICH, November 2010 slide 26
 ICH Quality Implementation Working Group - Training Workshop

Regulatory Assessment

Real Time Release Testing – Assessors’

Evaluation General Considerations
• Have tests been verified at full scale?
• Have analytical procedures been validated? If the
procedure contains a model, has it been validated
and has an adequate maintenance plan been
proposed?
• Have alternate traditional testing procedures been
provided for any RTRT? To be used for
- Stability testing
- Regulatory testing
- Break down of equipment when specified in dossier
© ICH, November 2010 slide 27
 ICH Quality Implementation Working Group - Training Workshop

Regulatory Assessment

Example from Case Study: RTRT for

Dissolution
• Quality Risk Assessment shows that API particle size, lubrication
and compression have potential to impact dissolution
• Analysis of in-vivo data also shows that API particle size impacts
bioavailability
- Larger particles have lower Cmax and AUC
• Multi factorial DoE carried out to estimate impact of factors on
dissolution
- Factors investigated: API particle size, magnesium stearate specific
surface area, lubrication time and tablet hardness
- Response measured: % dissolved at 20 min
- DoE data analyzed to identify statistically significant factors affecting
dissolution

© ICH, November 2010 slide 28

 ICH Quality Implementation Working Group - Training Workshop

Regulatory Assessment

Example: RTRT for Dissolution

• Predictive model for dissolution defined from DOE data
Prediction algorithm:
Diss = 108.9 – 11.96 × API – 7.556×10-5 × MgSt – 0.1849 × LubT –
3.783×10 -2 × Hard – 2.557×10-5 × MgSt × LubT

• Model verified by comparing predicted data with measured dissolution data for 3
batches
Graphical Representation of Dissolution Design Space
Diss (% at 20 min)

Area of potential risk

Design for dissolution failure
Space
Graph shows interaction
between two of the
variables: API particle size
and Mg Stearate Specific
Surface Area

© ICH, November 2010 slide 29

 ICH Quality Implementation Working Group - Training Workshop

Regulatory Assessment

Dissolution Model Based on RTRT –

Assessors’ Evaluation
• Has a robust and discriminatory reference procedure (e.g. dissolution by HPLC)
been provided?
• Has the dissolution model been validated with an independent data set (i.e. not
just the DoE data)?
• Has model applicability been demonstrated across all variability proposed in the
design space (e.g. change in scale, change in equipment type etc)
• Has process and/or method uncertainty been incorporated in the model?
- Has a process been described for revision of design space on basis of
prediction intervals?
• Has the applicant considered multivariate trend monitoring for the CQA and/or
CPP that impact dissolution (e.g. API particle size, compression parameters
etc)?
• Have plans been provided for model maintenance throughout the product life
cycle?
- Plans to revise the model (e.g. with change in API PSD outside the range that was
evaluated via the DoE)
- To be done under the company’s quality system and subject to GMP inspection

© ICH, November 2010 slide 30

 ICH Quality Implementation Working Group - Training Workshop

Regulatory Assessment

Dissolution Model based on RTRT -

Assessors’ Evaluation Continued
• Is the model prediction compared with the reference method for a
statistically significant number of batches?
• Is the proposed acceptance criteria for dissolution appropriate?
• Given that there are more than 2 parameters that impact
dissolution, should the dissolution design space be represented
graphically as an interaction of more than one response surfaces?
• How capable is the model:
- For taking into account variation in tablet hardness throughout the
run?
- For predicting failed batches?

© ICH, November 2010 slide 31

 ICH Quality Implementation Working Group - Training Workshop

Regulatory Assessment

Dissolution Model based on RTRT -

Assessors’ Evaluation Continued
• Have details been provided on how the model would
be used as a feed forward control, to adjust process
parameters (e.g. compression parameters)
depending on API particle size and/or magnesium
stearate specific surface area?

• Could a routine in process disintegration test lower

the risk of implementing this RTRT?

© ICH, November 2010 slide 32

 ICH Quality Implementation Working Group - Training Workshop

Regulatory Assessment

Example from the Case Study:

RTRT for Tablet Assay and CU

• Based on in-process tablet weight control

- Part of compression operation
• Fill volume during compression adjusted by a
feedback loop from the tablet weight measurement

© ICH, November 2010 slide 33

 ICH Quality Implementation Working Group - Training Workshop

Regulatory Assessment

Example from Case Study: RTRT for

Assay and Content Uniformity
• Risk Assessments as part of the QRM process shows four
factors have potential to affect Assay and CU:
- API Particle Size
- Environmental moisture control
- Blending and Lubrication
- Absence of segregation before and during compression
• API Particle Size controlled by incoming materials testing and
release
• Blend uniformity and absence of down stream segregation are
key elements of control strategy

© ICH, November 2010 slide 34

 ICH Quality Implementation Working Group - Training Workshop

Regulatory Assessment

RTRT for Tablet Assay and CU:

Assessors’ Evaluation
• Are adequate data presented to demonstrate absence of
segregation?
- During compression, especially at beginning and end of run
- When blend is held prior to compression
• Does the NIR method predict % active content of the blend (vs.
indicating uniformity by variance change)?
• How is the use of the RTRT described in the specification?
• Is the information provided (e.g. data points, number of batches,
comparison of individual tablets) adequate, to compare the
assay calculated by weight to assay measured by HPLC?

© ICH, November 2010 slide 35

 Implementation of ICH Q8, Q9, Q10

Regulatory Assessment
Assessor – Inspector
Interactions

International Conference on Harmonisation of Technical

Requirements for Registration of Pharmaceuticals for Human Use
 ICH Quality Implementation Working Group - Training Workshop

Regulatory Assessment

Assessor - Inspector Interaction

• Certain aspects of the application may need to be
verified at site, such as

- Has a statistically based criterion for release (e.g.

acceptance limits, sample size, confidence intervals,
outliers) been defined and addressed by the PQS?

- Does the company’s quality system have procedures to

trend tablet weight during routine production and to
accept/reject batches on the basis of RTRT?

© ICH, November 2010 slide 37

 ICH Quality Implementation Working Group - Training Workshop

Regulatory Assessment
Assessor - Inspector Interaction
Continued
• Certain aspects of the application may need to be
verified at site, such as
- Implementation of commercial manufacturing process
- Implementation of design space, RTRT, control strategy.
- Management of design space and models
- Confirmation of data
- Input for batch release strategy
- Sampling plan especially for RTRT

• Communication between inspector and assesor is

important

© ICH, November 2010 slide 38

 ICH Quality Implementation Working Group - Training Workshop

Regulatory Assessment

Case Study Example of Interaction

Between Assessors and Inspectors
Points to Consider
• For Crystallization Design Space
- Conducting the inspection during the review period
- Communication between Inspector and assessor prior to inspection
- Including assessors and inspectors on inspection
- May require specialized training for things like models and RTRT
- Reviewing procedures for design space management within the
company’s quality system

• For future inspections after commercialization

- Did verification of design space for crystallization at commercial scale
support conclusion that the design space was scale independent?

© ICH, November 2010 slide 39

 ICH Quality Implementation Working Group - Training Workshop

Regulatory Assessment

Conclusions
• Use of ICH Q8, Q9, Q10 will facilitate regulatory
assessment
- Knowledge rich applications provide transparency and
facilitate assessment
- Systematic development described in regulatory submissions
will improve the regulatory assessment
- Improve the efficiency of the review / assessment
- Enable science and risk based regulatory decisions
- Improve communication
- Between Regulators and Industry
- Between Assessors and Inspectors

© ICH, November 2010 slide 40