Optimal Applications of

1
Optimal Applications of In-Line or At-Line
Manufacturing Controls in Pharmaceutical
Production

• Ajaz S. Hussain, Ph.D.
• Deputy Director (Act.)
• Office of Pharmaceutical Science, CDER, FDA
• 19 July 2001, ACPS Meeting

2
ACPS Meeting Discussion Objectives

• Opportunity Significant public health and
  economic benefits may be realized through optimal
  applications of modern in-line or at-line
  process controls and tests in pharmaceutical
  manufacturing.
• Initiate public discussion on opportunities and
  challenges associated with the regulatory
  applications of modern Process Analytical
  Chemistry tools in pharmaceutical industry
• Discuss anticipated win-win opportunities for
  the U.S. public and the industry

3
Process Analytical Chemistry
Many technologies - real-time - without
sampling - multivariate - e.g., Near IR, Raman,.
Traditional
4
http//www.nirplus.com/
5
From Incoming Raw Material Inspection to Final
Product Release XXXXXX instruments, software,
and application support services meet a full
range of pharmaceutical testing needs
6
Impact on Product Quality

• Current situation
• Discovery Combinatorial chemistry plus high
  throughput screening
• Product development rate limiting?
• Black-box (inability to reliably predict product
  performance changes when formulation/process
  variables are varied)
• Variable physical functional attributes of raw
  materials that conform to compendial standards
• PCA Focus on both physics and chemistry at the
  same time

7
Product Development Evolution

• Over the last 100 years
• Art? Science Engineering based
• Trial-and-Error ? DOE ? CAD
• Dosage forms ? Drug Delivery Systems ?
  Intelligent Drug Delivery Systems
• Few creative options tested ? Many creative
  options tested
• Batch processing ? Continuous (automated)
  processing

8
Product Development Process

• Multi-factorial complex problem
• Significant reliance on personal knowledge
• Historical data likely to have been generated by
  a guided Trial-n-Error approach
• Many choices for achieving target specifications

• Without up-to-date information, high potential
  for
• misjudgments
• reinventing the wheel
• Mobile institutional memory
• Approved product may need frequent changes ..

9
Product Development Knowledge
Level of Sophistication HIGH MEDIUM LOW
Details Resolved
HIGH
MEDIUM LOW
Rules
MECHANISTIC MODELS
EMPIRICAL MODELS
HEURISTIC RULES Rules of Thumb
HISTORICAL DATA DERIVED FROM TRIAL-N-ERROR
EXPERIMENTATION
10
Controlling Unit Operations

• Blending
• Equipment (type, size, operating speed,)
• Process (time)
• (BUA)
• Wet granulation
• Equipment (type, size, operating speed,)
• Fluid addition (composition, volume, rate)
• Process (time)
• (Moisture Content)

11
Limitations of current approach

• Unit operations are intended to produce
  in-process materials that possess optimal
  attributes for subsequent manufacturing steps
• Do current controls always ensure consistent
  quality of in-process materials?
• Physical attributes of pharmaceutical raw
  materials (e.g., excipients) can be highly
  variable
• Adjustments in process parameters outside
  established (validated) range may require
  regulatory approval

12
Current Situation In-Process Tests

• In-process testing samples collected and sent to
  QC lab
• Processing parameters and specification are set
  based on limited data
• Raw material (e.g., excipients) specifications do
  not necessarily reflect their functionality
• In-process sample collection, testing,
  verification and Exceptions contribute to long
  (production ) cycle time

13
Process Validation Limitations?

• Harwood and Molnar. Using DOE techniques to avoid
  process problems. Pharm. Dev. Tech. 1998.
• ....well-rehearsed demonstration that
  manufacturing formula can work three successive
  times.
• It is authors experience that ... validation
  exercise precedes a trouble-free time period in
  the manufacturing area only to be followed by
  many hours (possibly days or weeks) of
  troubleshooting and experimental work after a
  batch or two of product fails to meet
  specifications. This becomes a never-ending
  task.
• Not a general observations but illustrates what
  happens when quality was not built-in

14
PROCESS D WITH QC TESTSCycle Times including
BULK ACTIVE
20 DAYS
15 DAYS
BLEND 2 PRE-BLEND
GRANULATION
MILLING
CHEMICAL WEIGHING
BLEND 1 MSD
COMPRESS
FINAL BLEND
PROCESSING
10 DAYS
15 DAYS
QC1
QC3
QC2
60 DAYS
21-90 DAYS
G.K. Raju, MIT.
15
Process Controls Evolution.......
16
Modern In-Process Controls

• Near-IR and other noninvasive spectroscopic and
  imaging techniques have been widely used in
  chemical and food industry for 10 years
• These technologies can provide real time
  control of processes without having to collect
  samples
• One can potently process materials until
  optimal attributes are achieved (as opposed to
  stopping at a predetermined time and then testing
  if quality is acceptable)
• Using pattern recognition tools one can relate
  the Near-IR spectrums to both physical and
  chemical attributes of the materials and hence be
  in a position to predict product performance (and
  improve quality!!)

17
Example Direct compression tablet formulation

• NIR
• Identification and characterization (moisture,
  particle size,.)
• On-line control of adequacy of mix with respect
  to all components
• At-line assurance of acceptable hardness and
  friability
• At-line assurance or control of content
  uniformity
• At-line assurance of dissolution rate

• Conventional
• Compendial tests for excipients
• Blending
• BUA Testing (drug only)
• Compaction
• Hardness, thickness, weight, friability
• Content uniformity
• Dissolution

18
Inhomogeneity in Experimental ProductHand
Blended Preparation (FDA)
blue
red
0.05
0
-0.05
-0.1
Scaled Data
-0.15
-0.2
-0.25
-0.3
1000
1100
1200
1300
1400
1500
1600
1700
Wavelength (nm)
Localized Spectra
Chemical Image
19
Commercial Grade Product
20
Experimental Product (FDA) Poorly Blended
Preparation
21
Distribution of Lubricant in Blend
Spectra obtained every 15mm
S. Hammond, Pfizer
22
(No Transcript)
23
Aspirin Tab.
Blister Pak.
Malik, Poonacha, Moses, and Lodder. AAPS Pharm
Sci. Tech. 2001
24
Malik, Poonacha, Moses, and Lodder. AAPS Pharm
Sci. Tech. 2001
25
Challenges

• Mind-set FDA will not accept it
• Method suitability and Validation approaches
• Chemometrics
• Mechanisms for regulatory introduction
• Investment cost

26
Summary

• Potential benefits of PAC
• Manufacturing and QC cycle time and cost
  reduction
• Improving product quality
• Provide information during processing for feed
  back-control
• Direct (without sample collections and
  manipulations) and rapid measurement of critical
  product attributes
• Facilitate establishment of causal-links between
  product/process variables and product performance
• Improve patient and operator safety
• A win-win opportunity that will require out of
  the box thinking and approach for optimal
  applications of PCA tools