Integrated Method Development and Validation

Report
Integrated Method Development and
Validation
RACI Conference - Chemical Analyses
Dr. Ludwig Huber
[email protected]
Today’s Agenda
• Lifecycle management of analytical
procedures: development,
validation and routine use
• Using principles of Quality by
Design to get most robust
methods
• Defining validation parameters,
acceptance criteria and test
procedures
• Templates and examples for
efficient and consistent
documentations
© Copyright
Ludwig Huber - LabCompliance
FDA 2014
FDA 2013
NATA 2013
Slide 2
FDA Guide – Bioanalytical Method Validation
Major differences to the 2001 Guide
• Section on System Suitability testing
• Inclusion of incurred sample
reanalysis
• Level of details on LBA similar to
chromatographic methods
• Concentrations below the LLOQ should
be reported as zeros
• Sample Analysis Reporting should
include: All accepted and rejected
analytical runs
© Copyright
Ludwig Huber - LabCompliance
Slide 3
FDA Guide – Analytical Method Validation
• Components of Quality by Design
(QbD)
– Begin with an initial risk
assessment and follow with
multivariate experiments (design
of experiments).
– Lifetime management
• Requires submission of method
development data
– You should submit development
data within the method validation
section if they support the
validation of the method.
© Copyright
Ludwig Huber - LabCompliance
Slide 4
Official Guidelines for Method
Validation
QbD
• ICH - Guidance for Industry - Q2 (R1)
components
Text and Methodology
Must be followed in US and Europe
• FDA: Analytical Procedures and Methods
Validation for Drugs and Biologics (Draft, Feb 2014)
• FDA - Industry Guidance
Bioanalytical Method Validation (Draft, Nov 2013)
• USP <1225>: Validation of Compendial Methods
• USP <1226>: Verification of Compendial Procedures
• USP <1224>: Transfer of Analytical Procedures
ICH = International Conference for Harmonization
USP = United States Pharmacopeia
© Copyright
Ludwig Huber - LabCompliance
Slide 5
Method Validation
• The accuracy, sensitivity, specificity, and
reproducibility of test methods have not been
established and documented (W-187)
• Failure to validate analytical test methods used
for API for potency testing. (W-259)
• For example, your firm failed to validate the xxx
compound to quantify Peak A for potency and
robustness.
• Your firm has been unable to determine why the
chromatographic columns of the same make and
model had variability and could not provide
adequate separation (W-259)
www.fdawarningletter.com
© Copyright
Ludwig Huber - LabCompliance
Slide
Slide6
6
Method Validation Parameters for
different Method Tasks (ICH Q2)
Identification
Impurity
Quantitative
Impurity
Qualitative
Assay
No
yes
No
Yes
No
No
No
Yes
Yes
Yes
No
No
No
Yes
Yes
Yes
Specificity
Yes
Yes
Yes
Yes
Limit of detection
No
No
Yes
No
Limit of quantitation
No
Yes
No
No
Linearity
No
Yes
No
Yes
Range
No
Yes
No
Yes
Analytical Task
Accuracy
Precision
Repeatability
Intermediate
Reproducibility
Robustness
© Copyright
Expected to be done during Method Development
Ludwig Huber - LabCompliance
Slide 7
Parameters and Tests (ICH Q2)
Parameter
Accuracy
Tests (examples)
Minimum at 3 concentrations, 3 replicates
Precision
Repeatability
Intermediate
Reproducibility
Specificity
Limit of detection
Limit of Quantitation
Minimum of 9 determinations over the specified range
Over 3 days, 2 operators, 2 instruments,
Only required if testing is done in different laboratories
Prove with specific methods: HPLC, DAD, MS, dif. columns
Visual approach, S/N >= 3
S/N >= 10, Standard deviation of response
Linearity
Min 5 concentrations: visual, correlation coefficient (r)
Range
80 to 120% of test concentration, from linearity tests
© Copyright
Ludwig Huber - LabCompliance
Slide 8
Why Should we Change the Traditional
Way
• Problems in routine use, too many failures
• Developers not end-users
• Low emphasis on method robustness and
ruggedness
• Poor knowledge on critical parameters –
– problems during method transfer
• No or inadequate use of risk assessment
• Invested time not very efficient
© Copyright
Ludwig Huber - LabCompliance
Slide 9
Possible Conflict of Interests
• Development chemist
– Shortest time possible
• Routine User / QC Director
– No problem during routine use
– No out-of-specification situations
• Quality Assurance
– Enough documentation for inspections
• Regulatory Affairs
– Enough documentation for registration
• Finance
– Lowest development and validation cost
© Copyright
Ludwig Huber - LabCompliance
Slide 10
Objectives of the New Approach
• Efforts for method development and validation
should be value adding: building knowledge
• Method will work consistently within its design
space
– Changing people
– Changing material (e.g., chromatographic
column)
– Environment (transfer)
• Focus on critical parameters using a risk based
approach
Compliance is still important !!!
© Copyright
Ludwig Huber - LabCompliance
Slide 11
What we really want
• Design a method and validation procedures
to ensure that the method works for the
intended routine use, independently from
– Where it is being used
– Who is using it
– Specified instrumentation
– Actual method parameters, as long as
they are in the defined operating range
Trouble free operation – transfer –
With no method specific OOS results
© Copyright
Ludwig Huber - LabCompliance
Slide 12
QbD - Background and regulatory Situation
• Principles widely applied in all industries, particularly in car
industry
• Adopted by FDA in the 21st Century cGMP initiative
Reference: Pharmaceutical Quality for the 21st Century: A Risk-Based
Approach (2003)
• Adopted by ICH in Q8: Product Development, 2005,
updated in Q8 (R2), 2009
• In 2006, Merck & Co.’s Januvia became the first product.
• Starting to be adopted to analytical laboratories, e.g., used
to design robustness into analytical methods with the
Analytical Target Profile (ATP) concept
• 2013: FDA/EMA Q&As on method validation by QbD
• 2014: New FDA method validation guide with QbD components
© Copyright
Ludwig Huber - LabCompliance
Slide 13
QbD in Laboratories: Key Applications
• Development and validation of analytical
methods
– HPLC and others
• Transfer or analytical procedures
• Verification of compendial methods
• Analytical instrument qualification
• Dissolution testing
• Near Infrared Spectroscopy (NIR)
method
• Water analysis
© Copyright
Ludwig Huber - LabCompliance
Slide 14
EFPIA Positioning Paper
• Establishment of Analytical Target Profile (method
performance criteria, acceptance criteria)
• ATP defines ‘what’ needs to be measured not ‘how’
• ATP is submitted to regulatory agencies and approved
instead of an analytical procedure
• Any analytical method conforming to the approved ATP
can be used
• Alternative methods, e.g., new technology, can be used
through internal change control procedure
In line with FDA‘s general approach for QbD (no re-approval
required as long as working in the approved design space)
Also in line with the European Variation Guideline and with ICH Q8
Reference: Ermer, European Pharmaceutical Review, Vol 10, Issue 3 (2011)
EFPIA = European Federation of the Pharmaceutical Industries and Association
© Copyright
Ludwig Huber - LabCompliance
Slide 15
QbD in Laboratories – Current Situation
and key Applications
Situation
• No formal regulations or guidelines, no FDA pilot project
• QbD can be used for all critical analytical quality
parameters
• Some laboratories are starting to adapt QbD for analytical
method validation
• FDA/EMA address methods in Q&As sessions and guide
• EFPIA Positioning Paper
Key Applications
• Development and validation of analytical methods
• Method transfer, disolution testing
EFPIA = European Federation of the Pharmaceutical Industries and Association
© Copyright
Ludwig Huber - LabCompliance
Slide 16
Regulated
Not regulated
Traditional Development & Validation of
Analytical Methods
Preparation
• Select preliminary method, scope &
specifications
• Assure performance of equipment
• Assure that operators are qualifified
Development
•
•
•
•
• Document final acceptance criteria
• Document final scope
• Perform validation tests, incl. robustness
Validation
Routine Operation
© Copyright
Select and optimize method & parameters
Robustness testing
Define operational limits and SST
Preliminary validation experiments
• Controlled transfer
• Regular review
• Controlled changes & Revalidation
Ludwig Huber - LabCompliance
Slide 17
Quality by Design for Analytical Methods
Specifications
Analytical Target
Profile, Quality
Target Method
Profile
Method development
Critical Method
Parameters and
Critical Attributes, Risk
Assessment
Continuous
Monitoring and
Improvements
QC Tracking
Control Strategy
for CMAs
System Suitability
© Copyright
Design Space,
Method
Operational
Ranges
Method
Qualification
(ICH Q2)
Ludwig Huber - LabCompliance
Slide 18
QbD Terms in Method Development
and Validation
Product Development
Method Development
Method Validation
Examples for Methods
Target product profile
(TPP)
Analytical target profile
(ATP)
Accurate quantitation of
impurities in drugs
Quality target product
profile (QTPP)
Quality target method
profile (QTMP)
LOQ <0.05%, precision
and accuracy at LOQ
better than 15%
Critical process
parameters (CPP)
Critical method
parameters (CMP)
Flow rate, temperature,
pH of mobile phase
Critical quality attributes
(CQA)
Critical Method
attributes (CMA)
Resolution, peak tailing
Proven acceptable
range (PAR)
Method operational
design range (MODR)
pH ± 1,
col temperature ± 2
© Copyright
Ludwig Huber - LabCompliance
Slide 19
Define the Analytical Target Profile
• Method operational intent (what the method has to measure)
Inputs from end-user department
– Ease of use, analysis cycle time, acceptable solvents,
analysis cycle time
• Method performance characteristics, e.g., precision,
accuracy, specificity, LOD/LOQ, linearity
• Acceptance criteria for method performance characteristics
• Which instruments will be used, where will the method be
used (specific lab, specific site, global)?
Example (incomplete): Quantitative impurity analysis compound at ≥0.05% with
an accuracy and precision of 15% RSD at the limit of quantitation and 5% at
20x LOQ.
© Copyright
Ludwig Huber - LabCompliance
Slide 20
Determine Factors Impacting Critical
Method Attributes (CMA)
• Test Conditions
Factor3
Factor2
– HPLC Mobile phase composition, pH
– Column Temperature, detector wavelength Factor1
MCA
– Sample extraction time
• Material attributes
Factor1
– Matrix, sample stability, sample solubility,
Factor2
Factor3
column batch
Use Fishbone diagrams
– Reference standards, quality of reagents
and risk assessment
• Environmental conditions
– Humidity, room temperature, electromagnetic interference
• Random effects
– Analysts, e.g., skill level, thoroughness
– Timing, e.g., day and night shift
– Instrument, e.g., performance, maintenance
© Copyright
Ludwig Huber - LabCompliance
Slide 21
Apply Risk Assessment to Support
Defined Criticality of Method Attributes
• Identify parameters with impact on the method’s performance (Risk
Identification)
– Rely on subject matter experts, Brainstorming meeting
– May also go back to development experiments
• Develop a prioritization matrix (Risk Evaluation)
– Look at factors with highest impact on method performance
– Link at specified instrument
functionality, performance and qualification
– Rank, e.g., in three categories: high (3), medium (2), low (1)
• Determine risk priority numbers for individual parameters
Severity
Probability
Detectability
Risk Number
Factor 1
3
3
2
18
Factor 2
2
1
3
6
© Copyright
Ludwig Huber - LabCompliance
Slide 22
Example Risk Prioritization Matrix
Impact of Method Parameters on Performance
pH
%organic
phase
UV
Wavelength
2
1
3
2
2
1
1
3
2
Repeatability
2
1
1
2
3
Accuracy
2
3
1
3
1
Specificity
3
1
1
3
2
10
8
5
14
10
Method
Parameter
Method Attributes
col.
temp.
flow rate
LOQ
1
Linearity
Risk Priority
Number (RPN)
1 = low, 2= medium 3 = high impact
© Copyright
RPN ≥ 9 included in DOE study
Ludwig Huber - LabCompliance
Slide 23
Determine the Method Operational Design
Range (MODR) through DOE
• From the „Critical Method Attribute“ exercise, select factors that
based on the risk assessment will impact method performance.
• Choose levels of each factor (two or higher)
• Select range over which factors will be varied,
e.g., in two level study there will be a high and low level value
– Requires good knowledge of the method
• Use the multivariate experimental design approach
• Define and perform experiments
• Perform statistical analysis of data
• Interpret the data
• Perform follow-up runs (if necessary)
© Copyright
Ludwig Huber - LabCompliance
Slide 24
Design of Experiments (DoE)
• Early DoE strategies began in 1920s
• Part of QbD
• Helps to understand the cause and effect relationship
between input factors and output (e.g., test parameters vs.
method performance)
• Most important to determine a method’s robustness
• Typically implemented through simultaneously changing two
or more parameters, reducing the number of experiments
• Facilitated through availability of software, e.g., Design
Expert (www.statease.com), Minitab (www.minitab.com)
Ludwig Huber
© Copyright
Ludwig Huber - LabCompliance
Slide 25
Simple DoE Example for HPLC Method
- Impact on Selectivity Run #
%
Org.Phase
pH
Col. Temp
Column
Flow Rate
1
-1
-1
+1
+1
+1
2
-1
+1
+1
-1
+1
3
-1
+1
-1
+1
+1
4
+1
-1
+1
-1
-1
5
+1
-1
-1
+1
-1
6
+1
+1
-1
-1
-1
-1 =
40 % ACN
4.0
25 ºC
10 cm
2.0 ml/min
+1 =
60% ACN
40% Water
6.0
40 ºC
20 cm
2.5 ml/min
FDA: Need sufficient statistical power to support analytical “Design Space”
© Copyright
Ludwig Huber - LabCompliance
Slide 26
Validate the Method for Intended Use
• Formally validate the method following ICH Q2
• Develop a method qualification plan
• Assure that equipment is formally qualified
(specifically spelled out in the new FDA guide)
• Assure that personnel is formally trained
• Perform qualification experiments, including robustness
testing
• Evaluate data and document results
• Write a validation report
© Copyright
Ludwig Huber - LabCompliance
Slide 27
Examples for HPLC Robustness Testing
• Deliberately change critical operational limits and evaluate
impact on performance: precision, accuracy
• Include sample preparation and testing parameters
Sample preparation (accuracy)
• Extraction time (-20% of target)
• Extraction temperature (± 5 ºC)
HPLC
• Col Temperature (± 3 ºC)
• Mobile phase composition (± 2%)
• Buffer concentration (± 2%)
• Flow rate (± 0.3 mL/min)
• Detection wavelength (± 1 nm)
• Column Lot (quality, selectivity)
Ambient temperature/humidity
Stability of samples, standards
© Copyright
Define acceptable ranges !
Ludwig Huber - LabCompliance
Slide 28
Examples for Acceptance Criteria
Quantitative Impurities in Finished Drugs
Parameter
Accuracy
Test
90 – 110%, 80 – 120% at specifications limit
Precision
Repeatability
Intermediate
Reproducibility
Specificity
Limit of Detection
Limit of Quantitation
Linearity
Range
© Copyright
<4 % RSD (up to 15% at LOQ)
<5.0 % RSD (higher at LOQ)
< 6% RSD (higher at LOQ)
Peak resolution >1.5 (related substances)
or >2 (main peak)
Peak purity check with UV DAD or MS
N/A
0.05%
visual inspection of linearity curve, r>0.9900
o.k. if accuracy, precision, linearity criteria are met
Ludwig Huber - LabCompliance
Slide 29
Example: Report Summary Table
Validation
Parameter
Accuracy
Method
Precision
Intermediate
Precision
Specificity
Linearity
Range
Robustness
© Copyright
Measure
Acceptance criteria
Results
Recovery – Conc1
97 – 103 %
99%
Recovery – Conc2
97 – 103 %
100%
Recovery – Conc3
97 – 103 %
100%
RSD
≤ 1.5 %
0.4%
RSD
≤ 2.0 %
0.8%
Peak Resolution Factor R
R for all peaks >1.5
All peaks >2.0
Correlation Coefficient
≥ 0.9900
0.9900
Visual inspection of plot
Linear response plot
Shows linearity
Correlation Coefficient
≥ 0.9900
0.9900
Precision at 3 concentrations
≤ 1.5 %
<1%
Recovery at 3 Conc.
97 – 103%
99.6%
Column Temp. ±2 C
R for all peaks >1.5
R for all peaks >2.0
Mobile Phase ±2 %
R for all peaks >1.5
R for all l peaks >2.0
Sample extraction time -20 %
Recovery in spec.
Recovery in spec
Compound stability 6 days
<3% degradation
<2% degradation
Ludwig Huber - LabCompliance
Slide 30
Assure that the Method Remains in a
State of Control
• Run system suitability tests
– Select critical test parameters based on risk
assessment and design space experiments
• Track quality control sample test results
• Thoroughly look at OOS results, and if method
specific, implement a corrective action plan
• Apply rigorous change control procedures
© Copyright
Ludwig Huber - LabCompliance
Slide 31
Change Control
• Follow change control procedure
• Assess the impact of each change and
perform risk assessment
• Take advantage of knowledge gained during
robustness testing
• Evaluate if the method parameter change is
within the defined and tested design space
and boundaries (method operational design
ranges)
• If not may have to revalidate the method
© Copyright
Ludwig Huber - LabCompliance
Slide 32
Coninually monitor and improve the
Method
• Actively collect inputs from operators on reliability and performance of
the method
• Evaluate customer complaints
• Conduct regular method review, e.g., yearly
• Track and trend system suitability
• Respond to adverse trends before they become problems
• Continually improvement through
– Problem solving and corrective action
– Preventive action
– Verification of correctve and preventive actions
© Copyright
Ludwig Huber - LabCompliance
Slide 33
Benefits of QbD for Laboratories
- Example: Analytical Method Validation • Facilitates technology innovation (new technology can be
used without FDA re-approval, as long as the Analytical
Target Profile (ATP) is the same (future thinking)
• Technology changes can be implemented without loss of
time – facilitates continuous improvement
• Less analytical method related Out-of-Specification and
failure investigations
• Lower failure rates for method transfer
• Allowed method changes without revalidation well defined
through design space and robustness testing
Ludwig Huber
© Copyright
Ludwig Huber - LabCompliance
Slide 34
Industry Barriers to QbD
• Current guidelines not in line with QbD approaches
• Registration currently not based on method performance but
on method conditions
• Low motivation to change
• Only little experience in the industry
• Requires new tools and skills for analysts
• Implementation Challenges
– Collaboration between functions
– Experience with new concepts
– Workload and resource limitations
Ludwig Huber
© Copyright
Ludwig Huber - LabCompliance
Slide 35
FDA-EMA Collaborative Research on
QbD for Analytical Methods: Q&As
Question
• What are the Agencies’ views with respect to the use of
analytical target profile (ATP) for analytical methods?
Answer
• In general, an analytical process profile (ATP) can be
acceptable as a qualifier of the expected method performance
by analogy to the QTPP as defined in ICH Q8 (R2).
• However, the Agencies would not consider analytical
methods that have different principles (e.g.,HPLC to NIR)
equivalent solely on the basis of conformance with the ATP. An
applicant should not switch between methods without
appropriate regulatory submission and approval
© Copyright
Ludwig Huber - LabCompliance
Slide 36
FDA-EMA Collaborative Research on
QbD for Analytical Methods: Q&As
Question
• What are the Agencies expectations in regulatory submissions for
Method Operational Design Ranges (MODR)?
Answer
• For example, data to support an MODR could include: (a)
appropriately chosen experimental protocols to support the
proposed operating ranges/ conditions; and (b) demonstration of
statistical confidence throughout the MODR.
• Issues for further reflection include the assessment of validation
requirements as identified in ICH Q2(R1) throughout the MODR
and confirmation of system suitability across all areas of the
MODR
© Copyright
Ludwig Huber - LabCompliance
Slide 37

similar documents