Transcript Lilly Wave Lilly Brand PowerPoint Template

Dissolution Profile Comparisons:
Approaches and Controversies
Discussant Comments
Stan Altan
Midwest Biopharmaceutical
Statistics Workshop
Ball State University
Muncie , Indiana
May 20, 2014
Role of the Discussant
• Draw some connections and identify
commonalities across the materials presented
by the speakers
• Questions the presenters might have left
unanswered
• Stimulate discussion among audience and
speakers
7/17/2015
© 2014 Eli Lilly and Company
2
Presentation 1 Tom Parks
Validation Acceptance Criteria on the Observed
Mean and Standard Deviation for 2-Sided
Dissolution Specifications Using Bayesian
Methodology and Simulation
• Objective : Determine acceptance criteria for 2-sided
dissolution specifications to provide assurance that
future dissolution results will meet the specification limits
• ASTM E2709 (CuDal)
• Bayesian posterior predictive distribution used to
define joint acceptance regions of 𝑥 and S
defined by desirable probability of passing the
USP <711> criteria
7/17/2015
© 2014 Eli Lilly and Company
3
Presentation 1 Tom Parks
• FDA 2011 guidance, “Process Validation:
General Principles and Practices” states,
“The number of samples [in the process
performance protocol] should be adequate to
provide sufficient statistical confidence of quality…”
• Comment – compare with USP <711>
• To provide assurance that the 2-sided
dissolution specifications would be met for any
other sample pulled randomly from the batch
• Comment – compare with cGMP regulations for
validating pharmaceutical manufacturing 21 CFR
211.100(a) and 211.110(a)
7/17/2015
© 2014 Eli Lilly and Company
4
Presentation 1 Tom Parks
Sampling and Statistics (21 CFR 211.165(d)):
• Acceptance criteria for the sampling and testing
conducted by the quality control unit shall be
adequate to assure that batches of drug
products meet each appropriate specification
and appropriate statistical quality control criteria
as a condition for their approval and release.
The statistical quality control criteria shall include
appropriate acceptance levels and/or
appropriate rejection levels.
7/17/2015
© 2014 Eli Lilly and Company
5
Presentation 1 Tom Parks
• What’s the inference space for the Bayesian
methodology for calculating the probability of
passing USP <711> criteria given specified 𝑥
and S
• a single lot
• 3 or more lots (say during PPQ of PV)
• a process (say during continued verification)
• How could we develop this into a manufacturing
control or release strategy to apply to a
manufacturing process?
7/17/2015
© 2014 Eli Lilly and Company
6
Presentation 2 Sutan Wu
Current Statistical Issues in Dissolution Profile
Comparison
• Examine the performance of bootstrapping f2
and f2 index
• Gain empirical knowledge of the values of Mdistance: is M-distance a good substitute? What
would be the “appropriate” cut-off point(s)?
7
Presentation 2 Sutan Wu
• Three Scenarios
• 1: similarity factor f2 “safe” case - for both batches
• %CV at earlier time points (within 15 mins) <= 20% and
%CV <= 10% at other time points;
• Only one measurement after 85% dissolution
• 2. large batch variability cases (f2 is not recommended
generally)
• %CV > 20% (<= 15 mins) or/and %CV > 10% (> 15mins)
• Different mean dissolution profile but same variability for
both batches
• Same mean dissolution profile but testing batch has large
variability
• 3: multiple measurements after 85% dissolution
•
•
•
“Safe” Variability cases: Dissolution Guidance
recommendations
Large Variability cases
Dissolution mean profile generated from 2 parameter
Weibull
8
Methods
Pros
•
Simple to
compute
• Clear Cut-off
Point: 50
Mahalanobis
Distance
•
Model-dependent
Approach
•
•
Only the mean dissolution profile
to be considered;
•
At least 3 same time point
measurements for the test and
reference batch;
Comments
•
Approximately
over 95%
applications
•
Bootstrapping
f2 is used for
data with
large
variability
•
Only one measurement should be
considered after 85% dissolution
of both products;
•
%CV <=20% at the earlier time
points and <=10% at other time
points.
Both the mean
profile and the
batch variability
to be considered
together
Simple stat
formula
•
Same time point measurements
for the test and reference batches;
•
A few
applications
•
Cut-off point not proposed
•
Hard to have
a common
acceptable
cut-off point
Measurements
at different time
points
•
•
Model selection
Cut-off point not proposed
•
Some internal
lab studies
Similarity factor
𝑓2
•
Cons
9
Presentation 2 Sutan Wu
Conclusions
• Bootstrapping f2 is recommended when the similarity factor f2
is around 50 or large batch variability is observed
• Comment - How should the final result then be reported? How
does one correct for the bias in the f2?
• Large batch variability cases, new cut-off points may be
proposed. Testing batches would be penalized by larger
batch variability.
• Comment - Not clear how this would mitigate the regulatory
requirement for f2 or the statistical issues with f2
• M-Distance is another alternative approach for dissolution
profile comparisons. Its values also depends on the batch
variability. The cut-off point is required for further deep
examinations, particularly, M-Distance values at different
batch variability and bootstrapping f2 around 50.
• Comment - Not clear how this would be incorporated into a
submission for example comparing new process with old?
10
Presentation 3 Fasheng Li
Statistical Evaluation of Dissolution for
Specification Setting and Stability Studies
• Objective
 Setting Extended Release Dissolution Specifications
 Number of time points needed
 Evaluation of possible specifications
 Dissolution on Stability
 No significant linear trend observed
 Non-linear trend observed
 Methodology – 2 or 3 parameter Weibull
 Comment – limitations of the Weibull?
Presentation 3 Fasheng Li
Evaluation of Dissolution Specifications
• Determine Number of Time Points (3)
 Evaluate proposed dissolution specifications against
USP <711> at each time point
• Simulations performed on individual dissolution data at
each of the specification time points to check the
probabilities of passing different stages (L1, L2, and L3)
of USP <711> dissolution test
• Comment – not clear how this would be done – Posterior
predictive or some other?
Presentation 3 Fasheng Li
Stability Trending
 Fit three-parameter Weibull model to mean or individual
dissolution profiles at each of the stability time points
 Follow the ICH Q1E guidance if linear trend
 If no trend, the risk of failing dissolution at a future
stability test can be quantified by Constructing prediction
limits
 if nonlinear trend, construct 95% confidence interval to
define the shelf life
13
13
General Comments
• If proposing a model independent approach,
should we be concerned about possible
correlated observations across time points?
• If proposing a model dependent approach
(nonlinear model)
• Account for heterogeneity, what variance
structures are reasonable?
• Extension to stability characterization
• Suitability for manufacturing control strategies
• Design considerations
• Choice and number of time points
• How many vessels
7/17/2015
© 2014 Eli Lilly and Company
14
General Comments
• Profile comparisons (comparison of 2 lots,
groups of lots, processes) is still an fraught with
controversy
• The science matter experts are moving towards
a more complex and sophisticated
understanding of biorelevant dissolution
experiments using media mimicking the
composition of gastrointestinal fluids – this could
be considered closer to in vivo conditions
compared with compendial media (especially for
BSC II and IV compounds). These could be
seen as more informative in understanding the
clinical implications of formulation changes.
7/17/2015
© 2014 Eli Lilly and Company
15