Using A PQRI Approach in
Process Validation
Analysis of Powder Blends and Stratified
In-Process samples to demonstrate Unit
Dose Blend Uniformity
Darlene McKay
McNeil Consumer Healthcare
Requirements
FDA 21 CFR 211.110 (a)(3)
Demonstrating the adequacy of mixing to ensure
uniformity of in-process powder blends and finished
dosage units.
Health Canada Guidelines on Good Manufacturing
Practices (GMP), Division 2, Part C of Regulation
to the Food & Drugs Regulations require that:
All critical production processes must be validated
Powder Blend Uniformity
Powder blends should be tested for uniformity
based on samples that represent the final dosage
amount (1-3X weight of the final dosage unit)
Sounds simple enough,
obtain a sample thief capable of taking powder samples
in the weigh range of your product
Take samples from predefined locations (minimum 10)
Test samples according assay methodology to meet
blend uniformity specification
Pass Blend is uniform
Fail Blend is not uniform
Challenges
Problem is not all blends are alike
Some blends may fail blend criteria but when
compressed easily meet tablet content uniformity
testing
Particle dynamics are complex and although blends
may be adequately mixed and uniform in the blender or
tote, sampling may cause localized segregation
Sampling is therefore a critical factor and may provide
opportunity for variability or bias in your results
To adequately characterize a blend you need a
combination of both blend and finished tablet data
PQRI Approach
PQRI is the Product Quality Research Institute,
[Link]
2000 PQRI formed a Blend Uniformity Working Group
2002 submitted recommendation to FDA proposing an
alternate strategy to assessing blend and dosage unit
uniformity
2003 FDA issued Draft Guidance for Industry Powder
Blends and Finished Dosage Units Stratified InProcess Dosage Unit Sampling and Assessment
From blend, sample at least 10 locations (Small totes) or 15
locations (Large totes), with at least 3 replicates from each location
Verification of Manufacturing Criteria
Assay one sample per location
Adapted from Guidance for Industry: Powder Blends and Finished
Dosage Units Stratified In-Process Dosage Unit Sampling and
Assessment, Revised Attachments --- U.S. Department of Health and
Human Services Food and Drug Administration Center for Drug
Evaluation and Research (CDER), November 2003
Blend sample criteria:
RSD is NMT 5.0% and all individuals are within
+/-10% of the mean (absolute) 1
Assay 2nd and 3rd blend
samples from each location
No
Meet
Criteria?
Yes
Dosage Units:
During compression collect at least 7
tablets from each of at least 20 locations
Assay at least 3 dosage units per each
location, weight correct each result
Investigate original
criteria failure
Yes
Is mixing
problem
identified?
Blend is not uniform
RSD of all individuals NMT 6.0%. Each location mean is
within 90.0%-110.0% of target potency, and all
individuals are within 75.0%-125.0% of target potency. 2
No
Investigation points to blend
sampling error or some other
attributable cause
No
Assay at least 7 dosage units per
location, weight correct each result
Assay remaining dosage
units from each location,
weight correct each result
Meet
acceptance
Criteria?
Yes
Adequate powder mix
RSD of all individuals NMT 6.0%. Each
location mean is within 90.0%-110.0% of
target potency, and all individuals are within
75.0%-125.0% of target potency. 2
Go back to development
Blend is not uniform or post
blending practices are
causing segregation
No
Meets
acceptance
Criteria?
Yes
Examples of mean =/- 10% (absolute) are If mean strength = 103.0% then the interval is 103.0% =/- 10%, thus individual must fall within 93.0% to 113.0%
When comparing individual dosage units to 75.0%-125.0% of target strength, use the as is results (not weight corrected)
Why Use the PQRI approach?
Provides a reasonable, overall approach to
evaluating blend and product uniformity
Recognizes and allows for investigation of
sampling error
Relatively straight forward and easy to
apply
Allows for consistency
Blend Sampling
Sampling is critical. It is important to minimize
variability
Getting the right sample thief
Several models with multiple die sizes are commercially
available
Training
Consistency!
Consistency in technique, operator
Consistency in weight (use appropriate weight range for the
product)
Compression Sampling
Stratified Sampling:
Is the process of sampling dosage units at predefined intervals and
collecting representative samples from specifically targeted
locations in the compression/filling operation that have the greatest
potential to yield extreme high and lows in test results
Minimum 20 Locations
Focus on problem areas like initial start-up, end, or
critical points in the process
May also need to identify end-of-batch with
appropriate samples
Remaining locations can be equally distributed
across the batch
Testing
Consider having specific instructions in your assay
method for blend sample handling (i.e. weight
ranges, powder transfer technique)
Triplicate samples
Interpretation of results
Understanding variation and what it means to the
overall uniformity of the blend (are their trends that
may suggest a non-uniform blend or segregation?)
Weight corrected versus as is
Conclusions
This approach is still in the draft form with FDA
Health Canada has no published guidance specific
to blend uniformity
FDA draft identified some additional requirements
around application to routine manufacture
Standard Criteria (Readily Pass) versus
Marginal Criteria (Marginally Pass)
It is important that you characterize your product
in development
Even though you may meet the PQRI criteria, you may
decide to go back and modify the process
Marginally pass may mean more routine testing and/or
potential for failure
Case Study
Unit Dose Blend Uniformity and
Stratified Sample Case Study
Case 1: Readily Pass
From blend, sample at least 10 locations (Small totes) or 15
locations (Large totes), with at least 3 replicates from each location
Assay one sample per location
Blend sample criteria:
RSD is NMT 5.0% and all individuals are within
+/-10% of the mean (absolute) 1
Blend Sample Criteria
Meet
Criteria?
RSD NMT 5.0%?
PQRI Acceptance Range 88.4-108.4%
Yes
Blend Uniformity Results
Average = 98.4%
RSD = 2.5%
Range = 94.3-102.1%
Dosage Units:
During compression collect at least 7
tablets from each of at least 20 locations
Assay at least 3 dosage units per each
location, weight correct each result
Compression Results
Location Averages = 95.2 to 101.8%
RSD (n=60) = 3.4%
Range (as is) = 92.5-105.5%
RSD of all individuals NMT 6.0%. Each location mean is
within 90.0%-110.0% of target potency, and all
individuals are within 75.0%-125.0% of target potency. 2
Meet
acceptance
Criteria?
Yes
Adequate powder mix
Case 2: Marginal Pass
From blend, sample at least 10 locations (Small totes) or 15
locations (Large totes), with at least 3 replicates from each location
2nd , 3rd
Average = 103.9%, 102.8%
Assay one sample per location
RSD = 9.2%, 4.4%
Blend sample criteria:
Range = 81.0-125.8%,
RSD is NMT 5.0% and all individuals are within
+/-10% of the mean (absolute)
91.4-109.8%
1
Assay 2nd and 3rd blend
samples from each location
No
Investigate original
criteria failure
Is mixing
problem
identified?
All analytical
results verified
and No trends
were noted by
location
No
Investigation points to blend
sampling error or some other
attributable cause
Meet
Criteria?
Blend Uniformity Results
Average = 101.9%
RSD = 5.9%
Range = 93.7-119.2%
Blend Sample Criteria
RSD NMT 5.0%?
PQRI Acceptance Range 91.9-111.9%
Compression Results
Location Averages = 95.1 to 102.1%
RSD (n=140) = 3.2%
Range (as is) = 90.1-106.7%
Adequate powder mix
Assay at least 7 dosage units per
location, weight correct each result
RSD of all individuals NMT 6.0%. Each
location mean is within 90.0%-110.0% of
target potency, and all individuals are within
75.0%-125.0% of target potency. 2
Meets
acceptance
Criteria?
Yes
Good to Go?
Need to evaluate in relation to other batches
Need to have solid investigation to support
sampling error
This may be considered Marginal based
on FDA guidance
Could mean additional sampling/testing on
routine production
References/Acknowledgements
PQRI is the Product Quality Research Institute, [Link]
The Use of Stratified Sampling of Blend and Dosage Units to
Demonstrate Adequacy of Mix for Powder Blends, Garth Boehm, Jon Clark, John
Dietrick, Laura Foust, Thomas Garcia, Muralidhara Gavini, Loren Gelber, Jean-Marie Geoffroy, Pedro Jimenez,
Gerald Mergen, Fernando Muzzio, Jerry Planchard, James Prescott, Jozef Timmermans, Neeru Takiar
FDA 21 CFR 211.110 (a)(3)
Guidance for Industry (Draft), Powder Blends and Finished Dosage
Units Stratified In-Process Dosage Unit Sampling and Assessment,
U.S. Department of Health and Human Services Food and Drug Administration Center for Drug Evaluation and
Research, November 2003
Health Canada Guidelines on Good Manufacturing Practices (GMP),
Division 2, Part C of Regulation to the Food & Drugs Regulations
