Process and Analytical Validation

1
Process and Analytical Validation Working Group
Arthur H. Kibbe, Ph.D. Chair June 13, 2002
2
Process Analytical Technologies Systems for the
analysis and control of manufacturing processes
based on timely measurements during processing of
critical quality parameters and performance
attributes of raw and in-process materials and
processes, to assure acceptable end-product
quality at the completion of the
process. Definition of validation Three lots
and done? Follow the C in CGMP. The science
continues to move forward. In many cases,
existing validation processes are adequate for
PAT Internal buy-in may be most difficult. Case
studies (examples) will help educate inside the
company
3
PAT Validation Background - Existing validated
measurements invariably correlate poorly with
process performance - Univariate measurements
used to infer compliance of multivariate dynamic
systems - Measure what we can measure, not what
needs to be measured - Measurement has not been
seen as process related - Measurement needs to
respond to process need over the product life
cycle - Need to understand the process -
Recognize that the conventional approach to
validation might be limiting
4
Understand the Process - Break down into unit
operations - Assess risk potential for each unit
individually and collectively using techniques,
e.g. experimental design - Design systems to
manage risk Univariate measurement
Multivariate systems - Develop systems -
Establish proof of concept - Challenge
validation Objective - confirm processes and
measurement validity in real time across life
cycle
5
Validation Postulates 1. Validation Protocols
will be different for new products associated
with well-designed understood manufacturing
processes and existing products where PAT is
applied retrospectively 2. The validation plan
will reflect the total system design concept
since an RTQC/QA manufacturing process,
statistically based, essentially revalidates
itself on every manufacturing batch 3. The
rationale for model validation incorporating
pass/fail criteria must be clearly defined
thereby establishing the authenticity of
predictions in routine manufacturing and ensuring
compliance
6
A Checklist for Sensor and Chemometrics
Validation Sensor validation Software validation
including multivariate algorithms Sensor
calibration and calibration transfer
validation Process monitoring protocol (batch
vs.continuous) Process modeling (fundamental
process) Process control protocol Data management
and storage protocol
7
A Checklist of Targets for Validation Method
Types Primary method (no other method required,
e.g., endpoint detection) Secondary Method
(requires primary method to validate) Analyte
Types Direct measure (e.g., active
ingredient) Indirect measure (e.g.,
dissolution) Virtual measure (e.g. customer
satisfaction or quality index value) Dimensionali
ty univariate multivariate
8
PAT Implementation Questions What information is
needed and why? Where are the appropriate
measurement points? When (how often) are the
measurements needed? How is PAT provided
information to be used? Who will interpret this
information? (i.e., who receives and interprets?)
9
It is possible to define three distinct ways of
analyzing unit operations and releasing products
that are being developed and manufactured.
Condition 1 This is generally the current
operating scenario. Product is manufactured
according to fixed process conditions set during
development and confirmed during initial process
and product validation. Release is conducted by
physical and chemical testing subsequent to
manufacture. Condition 2 Product is
manufactured according to process conditions that
have been shown during development and
manufacturing to infer product performance and is
confirmed during initial process and product
validation. Relationships are developed and
confirmed with physical and chemical testing
subsequent to manufacturing runs. Release is
conducted by review of process conditions during
each batch manufacture.
10
Condition 3 Product is manufactured according
to process conditions that are responding to
direct measurement of in-process product quality
or unit dosage forms as they are being
manufactured. Relationships are developed
between process and product performance that are
optimized and bounded by data obtained in
development and manufacturing runs. Release is
conducted by data collected from in-process
product or each dosage form during manufacture.
Release specification form and validation
criteria can be defined for each condition based
on the nature of their release.
11
Questions to Move Forward on Improving PAT
Validation Should there be a difference in
expectations between developmental product
released for PI, PII, PIII than for routine
manufactured lots? Current validation practice
requires each unit operation in a manufacturing
process to be tested individually. Is it more
appropriate to design validation testing around
a total systems approach that includes multiple
unit operations? If so, what level of process
and product understanding is required during
design, development, scale-up and
manufacture? Could and should there be official
designations for products and processes that are
inherently capable of being appropriately
measured and controlled that would allow for
predicting product release characteristics?
12
Content Recommendations for Guidance Document 1.
Suitable for intended purpose 2. General
validation criteria 3. References to existing
guidance documents (ICH Q2 a and b Analytical
Validation, Q6 a and b, Specifications, FDA
Analytical Procedures and Methods Validation) 4.
Research exemption 5. OOT/OOS 6. Encourage use
of PAT. FDA should have a mechanism to institute
these new technologies and methods
13
Example PAT Method Validation Issues
Ex. 1. How to handle validation of PAT method
for a process monitoring or set-up parameter that
will be used for learning or decision-making? As
for other analytical methods use scientific
judgement to develop appropriate validation and
long-term maintenance protocol. New parameters
appear that need to be measured during a product
life cycle, especially for research purposes like
process optimization (the safe harbor issue)
14
Ex. 2. A validated laboratory method exists for a
regulatory parameter across an NDA range. How do
we replace this with a PAT method? Validate PAT
method, including long-term maintenance program
to monitor performance (both instrument function
and calibration model), and submit supplemental
regulatory method change documentation to
FDA. PAT method should correlate to validated lab
method as appropriate for intended use, and
should provide the same or increased assurance of
identity, strength, quality, purity, or potency
of the material being tested as the analytical
procedure described in the approved application.
15
Ex. 3. PAT Method is used in pilot plant for
process monitoring and development of a new drug.
When and how to validate? Steps 1. Calibrate
PAT method for use in pilot plant 2. Calibrate
and validate internally for factory process
monitoring 3. Validate and file method with FDA
if it ends up as a regulatory parameter.
16
Ex. 4. PAT method has been validated but probe
has a mechanical problem or sampling problem
(e.g., window fouling). An equivalent validated
laboratory method exists. State up front in
internal process control documentation the
preferred method and circumstances in which the
equivalent method could be used.
17
Ex. 5. A PAT method can only be calibrated
during factory processing. Process is optimally
run within an operating range narrower than the
fully validated range, so it is not always
possible to easily calibrate at the ends of the
range. Validate PAT method and use hybrid
approach. Specify the validated range for the
PAT method. When operating outside of this
range, must use laboratory method until PAT
method is revalidated to include the new range.
18
Ex. 6. PAT method is validated for typical
batches over a fixed range. Want to expand
calibration range/incorporate process change or
reworks. Include provision in PAT method
validation for controlled updating of calibration
model and revalidating as needed
19
Ex. 7. Accuracy of PAT method is different than
previously validated laboratory method.
Validate analytical method to the accuracy
required BY THE PROCESS rather than to the limits
of the method. A long-term maintenance program
to monitor performance (both instrument function
and calibration model) should be in place.