Case Study: Implementation of Design Space Concepts in Development of an ActiveCoated Tablet

1
Case Study Implementation of Design Space
Concepts in Development of an Active-Coated Tablet

• Robert A. Lipper, Ph.D.,Divyakant Desai, Ph.D.,
  San Kiang, Ph.D.
• Bristol-Myers Squibb Pharmaceutical Research
  Institute

Real World Applications of PAT and QbD in Drug
Process Development and Approval September 11,
2006 Arlington, VA
2
Outline

• Philosophy/Things to Ponder
• Technical Challenge and Approach
• Analysis of Variables
• Formulation
• Coating Process
• Uniformity of Input
• Spray Characterization
• DOE
• Control of Coating
• Results
• Scale-Up and Technology Transfer
• Model Development
• Summary

3
Design Space Ponderables

• The multidimensional combination and interaction
  of input variables. . .and process parameters
  that have been demonstrated to provide assurance
  of quality. from ICH Q8
• How many dimensions?
• How many combinations and interactions?
• How demonstrated?
• Homogeneous vs. heterogeneous systems
• Likelihood of stealth variables

4
QbD Philosophy

• QbD is about connecting the molecule and the
  patient.
• Science-based product/process design begins with
  the API molecular entity and is geared to meet
  patient needs for pharmacotherapy which is safe,
  effective, convenient and of consistently high
  quality.
• All products are designed and developed to be of
  high quality QbD provides a structured framework
  for documenting and presenting development
  rationale, experience and knowledge of the
  formulation and the process, and to ensure
  manufacture of products consistently fit for
  patient use.

5
Patient Requirements

• Content Uniformity
• Potency
• Stability
• Purity
• Consistent Bioavailability
• Cover Range of Potencies
• Readily Available in Distribution Channels
• Convenient and Pharmaceutically Elegant

6
Properties of the Molecule Technical Challenge

• pKa near neutral
• BCS Class III
• Hydrochloride Salt
• Acidic pH favors stability
• Dose 10 mg
• Primary degradation reaction occurs both in
  solid state and in solution
• Accelerates in presence of commonly used tablet
  excipients
• Accelerates with common processing conditions
  (roller compaction, wet granulation, compression)

7
QbD Approach--Tablet FormulationActive Film
Coating
SCHEMATIC
Inert tablet core
Inner layer seal coat of coating material

• Middle layer drug same coating material

Outer layer with same coating material

• This approach avoids
• compaction process
• granulation process
• direct drug contact with excipients
• Protects drug from environmental moisture
• Acidic environment used for all three layers

8
Manufacturing Process


9
Variables Considered to Define Design Space

• Spray Nozzle optimization
• Optimize air flow of spray gun for droplet size
  and spatial distribution
• Angle and distance of nozzles

• Formulation optimization
• API-to-polymer ratio
• Suspension pH
• Suspension Uniformity
• Suspension viscosity and solids content
• Coat thickness

CQAs Content Uniformity Potency

• Thermodynamic optimization
• Spray rate
• Inlet temperature
• Air flow

• Tablet bed optimization in coater
• Baffle configuration
• Pan load
• Pan speed

10
Formulation Factors

• HPMC- and PVA-based coating formulations were
  evaluated.
• Opadry II, a PVA-based coating formulation,
  provided the best stability
• Tablets were most stable when pH of the coating
  suspension was adjusted to around 2 for all three
  layers
• Three layer-coated tablets were more stable than
  those coated with two layers (i.e., omission of
  either inner or outer layer decreased stability)

11
Elements of QbD Preliminaries to Studying the
Coating Process

• Placebo cores are subjected to 100 on-line
  weight check controlled to 200 mg 2.
• A round biconvex tablet shape was chosen for
  durability, ease of coating and improved content
  uniformity.
• The process for dissolution of API in the coating
  vehicle was qualified using a UV fiber-optic
  probe.
• A re-circulation loop was designed in the tank to
  prevent sedimentation of pigments in the coating
  dispersion. A Raman probe was used to confirm
  the homogeneity of the coating dispersion.
  (Mixing design was optimized before undertaking
  spray characterization.)
• Brooks air flow controllers are used for
  monitoring atomization air feed into the spray
  guns.

12
In-Line Raman Monitoring for Coating Suspension
System Design
Proprietary Information
13
Spray Characterization and Design

• Goal
• Identify spray system hardware, configuration and
  operating parameters to produce
• Flat, focused spray pattern with uniform droplet
  size and intensity (narrow RSD)
• Uniform and stable spray cone
• Approach
• Characterize spray with two-camera imaging system
  (Off-line)
• Profile camera measures spray cone angle, spray
  intensity, and spray axis angle
• Droplet camera measures droplet size distribution
  (DV50 and DV90), mean droplet speed, and droplet
  density
• Effect of various parameters (Off-line)
• Nozzle types
• Nozzle operation under controlled air flow rates
• Solid content in coating suspension
• Air/liquid ratio

14
Effect of Nozzle Type
Comparison of Nozzle Types I and II
Selection criteria flat, focused spray pattern
with uniform droplet size and intensity.
Cone angle of (A) Type I and (B) Type II nozzles
15
Spray Characterization Summary

• Cone angle, droplet size, and droplet density
  are affected by suspension formulation
  (API-to-polymer ratio) and flow rate. Spray
  parameters are customized to assure content
  uniformity at different API-to-polymer ratios.
• Droplet size and cone angle decrease with
  increased air flow rate
• Too high a ratio of pattern air to atomizing air
  can create a hollow spray cone which (in this
  case) would adversely affect content uniformity
• Air volume is much preferred over air pressure
  to control the droplet size distribution and
  pattern of the spray, and is independent of
  coater scale or plumbing

16
Coating Control Use of Raman to Monitor the
Inner Layer
A U
Wave number (cm-1)
17
Setup of Raman Probe for In-Line Coating
Monitoring

• Pan size 36
• Pan speed 12 rpm
• Distance from the bed 4 inches
• Scan time 24 sec

18
Coating Kinetics of the Inner Layer Followed by
In-Line Raman Spectroscopy
El Hagrasy, A., Chang S-Y., Desai, D. and Kiang,
S. American Pharmaceutical Review 9(1)40-45
(2006)
19
Middle (Active) Layer Monitoring (2.5 mg API
Coated tablets)
20
Raman Prediction of the Inner Layer from
Different Spatial Locations
48 (I)
60(I)
5.5
5.5
5.0
5.0
4.5
4.5
4.0
4.0
3.5
3.5
Weight Gain
Weight Gain
3.0
3.0
2.5
2.5
2.0
2.0
1.5
1.5
1.0
1.0
1
2
3
4
5
6
1
2
3
4
5
6
Location
48 (II)
Location
60 (II)
5.5
5.5
5.0
5.0
4.5
4.5
4.0
4.0
3.5
3.5
Weight Gain
Weight Gain
3.0
3.0
2.5
2.5
2.0
2.0
1.5
1.5
1.0
1.0
1
2
3
4
5
6
1
2
3
4
5
6
Location
Location
El Hagrasy, A., Chang S-Y., Desai, D. and Kiang,
S. Journal of Pharmaceutical Innovation. Accepted
(2006)
21
Monitoring/Control of Coating Summary

• First coating layer can be monitored using a
  Raman probe (feasible to do at-line as well)
  and/or weight gain
• Second coating layer (active layer) API
  deposition can be monitored using weight gain
  and/or an off-line rapid HPLC or UV fiber-optic
  method
• Third coating layer can be monitored using an
  off-line Raman probe and/or weight gain
• The process has been successfully scaled using 24
  inch, 36 inch, 48 inch, and 60 inch Compu-Lab
  coaters (Batch sizes 14 to 215 kg)

22
Classification of Process Variables
PAR Proven Acceptable Range NOR Normal
Operating Range EOF Edge of Failure If frank
failures are observed, EOF should be estimated
23
Parameters Fixed for DOE

• 36 inch Compu-Lab coater
• Batch size 50 kg (250,000 tablets)
• Large baffles
• Type I nozzles
• Nozzle distance from tablet bed
• Ratio of pattern air to atomizing air
• Coating pan speed
• Tablet bed temperature
• Dew point ? 10C

24
Formulation and Process Optimization DOE
Design API Film Coated Tablets 2(5-1)
Fractional Factorial with 3 Center Points Designs
19 Runs
25
Creating the Process Design Space Process
Parameters Studied
Air Volume
525
600
Critical Parameters API application rate was
most criticalfor content uniformity.
300
50
Atomizing/Pattern Air Volume
75
5
Inlet Air Temperature
Screening for Parameter Ranges for Optimal
Content Uniformity
Spray rate
55
API in suspension
200
18
DOE
60
Spray rate
60
0.75
API/Opadry Ratio
0.75
8
11
105
API in suspension
26
Summary of Potency and Content Uniformity Results
of Second Layer Coated Tablets
27
Towards Process Understanding

• Based on risk assessment around the chosen
  formulation and processing approach, content
  uniformity and potency of tablets are considered
  to be the most critical product quality
  attributes
• Variables were systematically analyzed for their
  potential to influence the critical quality
  attributes
• Controlled experiments including DOE were
  conducted around the key variables to establish
  reliable operating ranges
• The process has been shown to be capable of
  consistently achieving content uniformity RSD of
  2.8-3.9

28
Utilization of Design Space for Tech Transfer

• Optimal design of mixing tank
• Raman spectroscopy

• fast-HPLC
• UVFO
• Raman / NIR

• Real-time Imaging Technology

• Raman Spectroscopy

Nozzle and tablet flows Characterization
Coating Process
Coating Process
Nozzle Characterization
Coating Suspension Homogeneity
Coating Suspension Homogeneity
Final Product
Final Product

• Optimization of spray pattern
• Support scale-up
• Specify process parameters to enhance tech
  transfer

• Real-time monitoring of the coating kinetics
• Effect of process variables on coating uniformity

• Fast check of coating uniformity
• Develop an index of mixing efficiency
• Determination of coating end point

• Minimize risk of sedimentation
• Continuous verification of TiO2 content

DEM and PBE models are being developed to predict
coating uniformity and coating weight in
production coater
29
Workflow for Coating Process Model
Nozzle optimization Feed tank optimization and
scale-up At-line uniformity analysis tablet
velocity characterization
PAT applications
Formulation
1.Predict RSD 2.Reduce DoE batches 3.Provide
added insight to design space for CMC
RSD model for Production Coater
Thermodynamics mass transfer
DEM-1M model
PBE-2 zone model
30
Summary

• A product design approach was chosen to address
  the chemical instability of the API in
  traditional formulations
• A clear and complete process understanding is
  being created during product development to
  assure process robustness
• Several PAT techniques, some with potential for
  in-line process control, are being utilized to
  develop a deeper process understanding
• Reliable operating ranges have been established
  for key process variables
• Process understanding gained through the QbD
  approach is being leveraged in scale-up and
  technology transfer

31
Acknowledgements
Project Leads Divyakant Desai San
Kiang Formulation and Drug Product
Process William Early Charles Van Kirk Howard
Stamato Srinivasa Paruchuri Sanjeev Kothari API
Process Steven Chan John Korzun Analytical
RD Harshad Patel Leon Liang Xujin Lu
PAT Arwa El-Hagrasy Don Kientzler Wei
Chen Shih-Ying Chang Technical
Operations Howard Miller Megan Schroeder DEM
modeling Fernando Muzzio (Rutgers
University) Regulatory Sciences Steve
Liebowitz
32
(No Transcript)
33
Backup slides
34
Formulation and Process OptimizationDOE - API
versus Polymer Amounts
-- Each batch was coated up to 10-mg potency.
Tablets corresponding to 2.5-mg and 5-mg
were collected at the appropriate times
(theoretical weight gain) and results were
treated separately. -- Effectively, three
separate DOEs were performed for the three
strengths 2.5 mg, 5 mg and 10-mg,
respectively.
35
Representative Tablet Formulations