Pat-qbd
This document was uploaded by user and they confirmed that they have the permission to share it. If you are author or own the copyright of this book, please report to us by using this DMCA report form. Report DMCA
Overview
Download & View Pat-qbd as PDF for free.
More details
- Words: 5,938
- Pages: 82
Integrating process analytical technology (PAT)/quality by design in pharma manufacturing and development
Düsseldorf September, 2008
Ingrid Maes Siemens Competence Centre Pharma
Confidential / Copyright © Siemens AG 2008. All reserved. © Siemens AG,rights All rights reserved
Presentation outline: The Changing Pharma landscape in R&D and Manufacturing How can PAT/QbD be part of the manufacturing and development architecture? How can PAT/QbD be a continuous process understanding and improvement tool? How can PAT/QbD be an enabler for increasing profitability & productivity? How can PAT/QbD be an enabler for time-to-market?
Page 2
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
The changing pharma landscape and gap between R&D and Manufacturing
Confidential / Copyright © Siemens AG 2008. All reserved. © Siemens AG,rights All rights reserved
Changes and impacts on the pharmaceutical manufacturing marketplace
Social change
Regulations
Technology change
Demographic
Industrial IT
Life style
Advanced control
Patients
Integrated solutions
Pharmaceutical Industry
Economical pressure
driven by; Cost Patient safety
Market change
Low pipe line
New therapies
Manufacturing
Delivery form
Supply Chain
Personalize
Page 4
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Pharmaceutical companies have to change Price
Healthcare system Business
R&D low pipeline high costs
Page 5
Sept. 2008
Manufacturing inefficient quality problems
Sales / Marketing Expensive sales inefficient supply chain
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
The change requirement Price
Healthcare system Business
R&D Smarter More efficient
Page 6
Sept. 2008
Manufacturing Better Higher responsiveness
Sales Marketing more control
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Changes that affect R&D and manufacturing: Integrated and continuous R&D and approaval Now: Discovery
Lead
Pre-clinical
Screaning
Develop.
Evaluation
Phase I
Phase III
limited launch
Future:
Patho – physiology
Molecule Develoment
Page 7
Phase II
Sept. 2008
Submission
in-life testing
Submission
Phase IV
Manufacturing
=> Approval on a real-time basis, with live licences contingent on the performance of extensive in-life testing
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Personalized Medicine
Batch size Targeted / personalized
Smaller batches Shortening time to market Bio markers
Smaller and shorterrun Clinical Trials due to increased statistical relevance on selected patients groups.
Discovery & Research Page 8
Sept. 2008
Drug screening
Bioinformatics
Toxicology
Clinical Trials
upscale
produce
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
The change of the regulations
Now Now
Before
Co-operative regulation Goal fixed and route free
Based on three reference batches no change allowed
Based on risk analysis process understanding
regulations
manufacturing
Page 9
Sept. 2008
re gu lat io ns
Repressive regulation Goal and route fixed
manufacturing
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
The regulatory changes: FDA’s 21st Century Initiatives Today
Vision
New initiatives to: improve manufacturing quality accelerate development Lower the regulatory burden
FDA new principles: Quality by design & design space Quality systems approach Reflecting product & process understanding and knowledge
Page 10
Sept. 2008
FDA’s future focus: Keynote address at IFPAC February 2007, by FDA's Chief Medical Officer, Dr. Janet Woodcock, on
Development & manufacturing should be integrated Development of quality surrogates for clinical performance (link critical product attributes to clinical outcomes) Rigorous, mechanistically based and statistically controlled processes
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Moving Pharmaceutical Manufacturing into the 21st Century
“The pharmaceutical industry could be wasting more than $50 billion a year in manufacturing costs alone…” - Macher, Georgetown University, and Nickerson, Washington University,
2006
“…we’re using the evaluation tools and infrastructure of the last century, and, in some cases, tools from very early in the last century to develop this century’s medical advances.”
- Dr. Janet Woodcock, M.D., Deputy Director and Chief Medical Officer of FDA, 2007
Page 11
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Current manufacturing status:
Large inefficient batch equipment Low utilization 30 -40 % on average Low product yields Excessive amounts of product non-conformances Long lead-times due to stage and final product testing Capital and labour intensive High operating costs High inventories and excessive warehouse capacity Cycle time improvement perceived to be limited by regulatory constraints 1% yield improvement ~ $400M in savings
Page 12
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
A Changing Landscape
The Pharma industry is just at the beginning of a dramatic change Relative separation between: R&D and Manufacturing Healthcare services and pharma industry (except for clinical trials & sales activities) The Pharma industry is seeking for new technologies / concepts and alternative business models
Page 13
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
How things will change and what will be supporting technologies for the future vision? On short, medium and longer term
Confidential / Copyright © Siemens AG 2008. All reserved. © Siemens AG,rights All rights reserved
Technologies shaping the pharma R&D and manufacturing future Process Analytical Technology
E-labelling
Miniaturized Manufacturing Concepts Demand driven manufacturing
Organizational Network Analysis Condition based maintenance Modular Plant design
Disposable manufacturing technologies
Semantic Web
E-licensing
RFID
Lab on a chip Master Data Management
Electronic Data Capture
PLM
Simulation and Virtual Prototyping
Continuous ManufacturingConcepts
Cell on a chip
Electronic Batch Management Clinical Trial Management Page 15
Sept. 2008
Recipe driven manufacturing
SPC
E-patient diaries R&D Workflow Management Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
New technologies and alternative strategies that are shaping future pharma New technologies
What is the strategic response?
PAT/ Quality by Design PLM SPC
Build-in quality
Continuous Mftg. Concepts Miniaturized manufacturing PAT / real-time product release Condition based maintenance
Increase manufacturing performance & throughput
Demand driven manufacturing Modular plant design Recipe driven manufacturing Disposable manufacturing
Flexible manufacturing concepts
Clinical trial management WFM, Data portals e-CTD PLM Bioinformatics
Speed-up development & closing the gap
Page 16
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Pharma Manufacturing 2020
Confidential / Copyright © Siemens AG 2008. All rights reserved.
How will companies cope with the changing environment? What will be future challenges?
What is the strategic response?
Changes due to new production technologies Increase operating efficiency Reduce product costs / achieve competitive pricing (e.g. response to biogenerics) Produce individual products / address niche markets (Personalized Medicine) Accelerate time-to-market
Scenario 1: Modernize within existing facility But, essentially same approach & scale
Scenario 2: Continuous processing, RTPR, JIT production
5 year
Page 18
Sept. 2008
What will be the implications on the way we manufacture in future?
Manufacturing strategic leaps to the medium and longer term: Scenario 3 = Speciality Niche products: Small scale pilot centers, Integration of R&D and production: Small batches 24/7 running Scenario 4 = Gross / mass market: Large-scale highly flexible plants, with high throughput
10 year
Move to personalized medicines Clinical and patient feed-back loops Continuous optimization and improvement
5 year
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Short term future (within 5 years) What will be future challenges? What will be the implications on the way we manufacture in future?
What is the strategic response? Increase operational efficiency (OEE) PAT (Process Analytical Technology) / QbD Electronic data acquisition and Data Management EBR MES PLM Real-time enterprise (data integration and management, dashboards / cockpits, facilitating real-time decision making)
Scenario 1: Modernize within existing facility But, essentially same approach & scale
5 year
Page 19
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
PAT/QbD a key enabling technology for the future
Confidential / Copyright © Siemens AG 2008. All reserved. © Siemens AG,rights All rights reserved
Taking pharma manufacturing into the 21st century through PAT / quality by design: the PAT journey PAT / Quality by Design is about starting an expedition. Taking pharma manufacturing into the 21st century Offers the pharmaceutical industry the possibility to increase product quality consistency and to reduce product risk through
Basecamp
increased process knowledge & understanding optimized process control
with the use of PAT tools
Page 21
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
PAT / QbD
New regulations FDA’s new initiatives to: improve manufacturing quality accelerate development lower the regulatory burden
FDA’s new principles: risk based approach scientific approach, based on process understanding
PAT (Process Analytical Technology) PAT = Understanding + controlling the manufacturing process A process is well understood when: All critical sources of variability are identified and explained Variability is managed by the process Product quality attributes can be accurately and reliably predicted Process Understanding is inversely proportional to risk
Page 22
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
The PAT guidance Process control trough new technologies (innovations), focus on manufacturing science A system for designing (process development), analyzing and controlling manufacturing processes, based on timely measurements of critical Q & performance attributes of raw-materials, in-process materials and processes with the goal of ensuring final product Q. Processes to assure acceptable end-product Q at the completion of the process (quality by design)
September 2004
Focus of PAT is understanding
Page 23
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
PAT & QbD: Quality-by-Design: Design of Product and Manufacturing Process Novel Technologies Platform Technologies Process Analytical Technology Self-regulating systems
Page 24
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Conventional Q approach Raw mats./ conditioning
Formulation
Finishing
Packaging
QC-system: lab based (End of phase testing of Q)
Page 25
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
PAT-based approach Raw mats./ conditioning
Bioreactor / Downstream
Formulation
Packaging
QC-system: PAT based (in-process QC / process based QC)
Adding value with Q by decision: Intelligent use of process data Advanced supervisory control Faster cycle times / Real-time product release Page 26
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
PAT the skill set and architecture fit PAT Toolbox / Skill set
Product & process design (Advanced) Process Controls
PLM
Data Collection storage & retrieval
Process Analytics
Sept. 2008
regulations
PAT
Data Analysis & mining
Page 27
Fit in development & manufacturing landscape
Information management tools
DoE
Process Automation
Field equipment
PAT
Data Portals/ Knowledge Mgt.
MES
Process technology
Process Analytics
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
PAT shifts monitoring and control from process data to product quality
PAT Quality build in by design Right first time
Classic control Closed loop control
monitoring product quality
Real-time release
mathematical translation
Lab
monitoring process data
Temp., pH, pO2, pressure, …
LIMS
Hold / release
Advanced Control
Process feed Process Analyzer Sample Process output
Page 28
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Time-based information Management Manufacturing Architecture
Time based information management (PAT system, MIS):
LIMS SPC
Quality
Quality Suite
ERP
Quality Suite
MES
Data is collected throughout the
SCADA
manufacturing time period
PAT
Captures process parameter measurements
Local Operation
Captures quality attribute measurements Data export by time period or by “batch”
Automation System
PAT
as required
Sensor/ Field Equipment Unit A
Page 29
Unit B
Sept. 2008
Unit Y
Unit Z
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Quality Suite Manufacturing Architecture PLM
LIMS SPC
Quality
Quality Suite
SPC
Overall Dashboard
ERP
Quality Suite, contains: On shopfloor (in manufacturing): PAT LIMS M-SPC/SQC Business intelligence suite (Quality cockpit) On boardroom level SPC Dashboards Central data warehouse + integration layer to R&D and manufacturing PLM and Knowledge Management
SCADA / MES
PAT
Local Operation Automation System
PAT
Quality Suite
Sensor/ Field
All integrated from shopfloor to boardroom Equipment Unit A
Page 30
Unit B
Sept. 2008
Unit Y
Unit Z
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Where we are today ? Challenge: Continuous Optimization and improvement Continuous manufacturing Just-In-Time Production Demand driven manufacturing Regulatory strategy
Today 2004 Investment in individual PAT tools
ROI
First PAT System implementations Page 31
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
What is the true gain of PAT?
Time-To-Market
Supply Chain Improvement
Improved market responsiveness Reduction of production cycle times Decrease warehousing
Fast process development Fast product release (real-time product release)
Project
Production
Asset valuation profit
Heart of the supply chain
Revenue
Development
cost costreduction reduction
reduction reduction time timeto tomarket market
primary
formulation
packaging
warehousing
by
Improving Improving manufacturing manufacturing performance performance
end of patent life
Deliver years
0
Drug innovation & approval
Page 32
5
Sept. 2008
make
deliver source
manufacturing
20
10
Project
source
Production
Revamp
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Business Drivers for PAT/QbD
Company Image Reduced risk via technology platform, anticounterfeiting Improved product tracking Reverse poor image Improved quality system throught audits Reduced risk fo recall, warning letter, consent decree
Validation Optimization Validation needs understanding Integral part of project Built validation into process
PAT/QbD Improve Existing Process Gain new process understanding Process optimization Reduced cost of quality Raw material specifications Know product availability + yield Real Time Release
Page 33
Sept. 2008
End of Life-cycle Transferability of process Scale down
Site to site transfer Accelerate transfer Reduce validation effort Reduce project time Mitigate transfer risk Move manufacturing to most effective site
New Product Development Real Time Release (RTR) Fast time to market Fast scale-up Clinical batches Process optimization Reduced cost of quality
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Improve Existing Process
Manufacturing accounts for the highest cost factor:
Research Driven Pharma Companies
31 %
Generic Manufacturers
51 %
Contract Manufacturers
62 %
Source: St Gallen report, 2006 Page 34
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
New Product Development The vast majority of investigational products that enter clinical trials fail PAT/QbD in manufacturing of clinical trial batches will reduce clinical trial failure, because:
Production more consistent products
Page 35
Sept. 2008
Reduced clinical batch product variability and hence “cleaner” clinical trial results
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
PAT / QbD Benefits
Impact on KPI’s
Business value Reduce Work In Process (WIP) costs: 33% Reduced warehouse space Reduced scrap/rework: 25 % Real-time product release Reduced labor costs: 30% Increased & consistent product quality Reduced quality costs: 13% Reduced regulatory compliance costs: 70% Reduced clinical batch product variability and hence “cleaner” clinical trial results Faster time to market: scale-up & tech transfer Continuous process understanding
Page 36
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Case: Results of Business Case Background
Results
A vaccine plant is seeking to achieve cost savings through modernisation of manufacturing infrastructure and implementing Quality by Design principles
Page 37
Sept. 2008
The result was a calculation of the optimal scenario (scenario 3): Labour: ¼ of operations people could be re-allocated Manufacturing throughput time: throughput time decreased with 1/3 Quality: 13% of the cost of QA and QC are eliminated Waste reduction: 3.5% Inventory: inventory could be reduced by 1/3
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
PAT strategy development
Page 38
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
PAT/QbD implementation A step-wise approach
Confidential / Copyright © Siemens AG 2008. All reserved. © Siemens AG,rights All rights reserved
PAT implementation roadmap Philosophy of Recent FDA guidances: Not a typical guidance Description of the desired state Path to the desired state is up to the industry Desired state: Process understanding
PAT Framework
Standards & best practices
Page 40
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
The PAT Implementation Roadmap
Page 41
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Secondary process
Primary process
Process assessment (CTQ steps)
Page 42
Buffer + Medium
Raw mats./ Conditioning
Sept. 2008
Critical process steps, in terms of critical to end-product quality
Bioreactor
Separation
Formulation
Conditioning/ Finishing
Purification
Packaging
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Primary process
Optimisation over the complete process Buffer + Medium
Synthesis / Bioreactor
Separation
Purification
Secondary process
Optimizing throughput, yield, quality, efficiency
Page 43
Raw mats./ Conditioning
Sept. 2008
Formulation
Conditioning/ Finishing
Packaging
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Case: PAT at NVI (Netherlands Vaccine Institute)
Customer process and backgrounds: Pertusses (Whooping coughing disease) 50 year old process quality test unchanged 50 year waste 25% Procedure (simplified): Step 1 - definition critical to quality parameter Step 2 - design experiments Step 3 - execute tests Step 4 - modeling Step 5 - implement new strategy Resulted in: better process understanding improvement of quality reduce waste by 20% increase production yield by a factor 3 potentially eliminate animal testing
Page 44
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Group / Region Department Ingrid Maes Industry/ CC Pharma
Search for Critical To Quality (CTQ) process steps What affects product safety, quality and efficacy ? Critical product specifications …
… which CTQ attributes ? What are the CTQ attrributes that describe the quality and performance of the product?
? y lit a ic t i Cr Where is product sensitive for manufacturing variation (root cause)?
Availability of Critical To Quality (CTQ) measurements ? Are CTQ parameters continuously measured? Are CTQ parameters measured by lab assay at end or during batch? Can CTQ parameters be inferred from PAT sensor with calibration model?
What to do to design out risk? Case: Whole cell vaccine against B. pertussis infection: Inactivated cells are the actual product Outer membrane proteins are presented to immune system Outer membrane composition crucial for vaccine efficacy
Page 45
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Case: CTQ for pertussis vaccine CTQ for pertussis vaccine …
… How is CTQ determined ? Microarrays
Protein analysis
Vaccine performance is determined by presence of relevant outer membrane proteins Virulence activated genes (VAGs) are responsible for outer membrane proteins (Adhesins, toxins) Virulence gene expression : protective vaccine Virulence gene expression : non-protective vaccine = product quality market genes VAG expression depends on extracellular conditions
DNA microarrays for expression profiling Protein analysis (Western Blotting, ELISA) Supernatant analysis with 1H and 13C NMR
Classical (animal) tests do not provide enough discriminative power for assessment of CQA’s
Page 46
Sept. 2008
Protein profiling (nano LC-MS, new)
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Monitoring Critical Attributes CTQ for pertussis vaccine …
… what do we have ? pH, DO and T are timely but insufficient
Objective : understanding and controlling the variance that affect product quality during
Nutrients & metabolites with NMR
cultivation Explore (process & product) state of the art assays for product and process Define critical process parameters How can we control the process to yield the desired end product quality? On line monitoring A signal that reflects the critical process parameters
Page 47
Sept. 2008
mRNA and protein analysis are sufficient but not timely
Animal testing for end-product release
Online monitoring of critical attributes necessary. But which technology?
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Select appropriate process analyser technology Case: Bioreactor Which real-time monitoring technology?
Laser Laser Diode Diode Spectrom. Spectrom. Flow Flow
Temperature Temperature
Pressure Pressure
Level Level
Mass Spectroscopy: composition, ...
Liquid Liquid Analytics Analytics
Weighing Weighing Technology Technology
Positioners Positioners
Selected technologies for bioreactor monitoring
M
Gas Gas ChromaChromatography tography Mass Mass Spectroscopy Spectroscopy
NIR NIR Spectroscopy Spectroscopy Gas Gas Analytics Analytics
Medium /Substrate Buffer/Acid /Lye
IR IR Spectroscopy Spectroscopy Raman Raman Spectroscopy Spectroscopy
NMR NMR Spectroscopy Spectroscopy
Off gas
Inoculum Bioreactor
Laser Laser diffraction diffraction Spectroscopy Spectroscopy
Product NIR Spectroscopy: composition, Biologic performance, ...
Page 48
Sept. 2008
Air/O2/CO2
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Case: Real-time & in-situ Bioreactor monitoring with NIR
In situ NIR transmission probe
Page 49
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Use of NIR as quantitative tool for bioreactor monitoring Physical, Chemical Status
Biomass concentration monitoring:
Early detection of endpoint
Loga r ithm of B iom a s s
Monitoring of nutrient (glutamate, lactate) consumption and metabolite formation with NIR:
Biological Status
Biomass
1
2
3
4
5
6
7
Time
2. Lag Phase 3. Transient Acceleration 4. Exponential Phase 5. Stationary Phase 6. Death Phase / necrosis
Page 50
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Special case: Real-time infection detection with NIR
Early detection of disturbances
Example: Yeast culture
For control or fault detection: Contaminations by foreign microorganisms
Infected (102 CFU/ml)
B. subtilis, S. aureus, Ps. Aeruginosa,
Non infected
C. albicans, E. coli, A. niger and L. Brevis
Biological contamination
Page 51
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Real-time Product Release = measure & control “ultimate quality” From critical product specifications …
critical product specification
… to critical process specifications
critical process specification (requirements to be met to consider the process “under control”)
Page 52
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Use of NIR as qualitative tool for bioreactor monitoring Physical, chemical and biological status
Total bioreactor monitoring
No quantification of analytes, the overall behavior is assessed Output is a “process fingerprint” representing the behavior of the batch process MVDA (PCA)
NIR spectral data
Start
Bioreactor pathway monitoring
End
Change in process status during batch progress = Process trajectory / fingerprint / signature (multidimensional profile) Control correct bioreactor characteristics Early detection of disturbances For control or fault detection
Scores plot for the two main principal components on collected spectra during the batch process
Page 53
Sept. 2008
Compare different batches
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Use of NIR as qualitative tool for mixing monitoring Physical, chemical status
Total blending monitoring
No quantification of analytes, the overall behavior is assessed Output is a “process fingerprint” representing the behavior of the batch process MVDA (PCA)
NIR spectral data
0.003
PC2
Scores
26 0.002
19 20
24
23
31 27 33 25 28
22 21
36 35
3439
38 37 42 46
18
0.001
40 43 41
17
44
16 0
2 7 13 10 6 5 3 12 14 4 8
45 47 50 51 53 49 48 70 52 55 68 5654 90 71 7588 8972 74 62 76 8077 60 59 8587 66 5783 65 6486 82 69 7873 63 7981 58
End
1 -0.001
15
9
Start
-0.002
Change in process status during batch progress = Process trajectory / fingerprint / signature (multidimensional profile)
30 32
29
84
Control correct mixing characteristics
67
61
11
-0.003 -0.06 -0.05 RESULT5, X-expl: 100%,0%
PC1 -0.04
-0.03
-0.02
-0.01
0
0.01
0.02
Early detection of disturbances
0.04
For control or fault detection
Scores plot for the two main principal components on collected spectra during the batch process
Page 54
0.03
Sept. 2008
Compare different batches
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Batch to batch comparison Batch Start
Batch Endpoint (out of spec) Batch Endpoint (in spec)
Batch Endpoint (Product quality reached)
Batch 1 Batch 5
Batch 4
Batch 2
Batch 3 Page 55
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Case: Another example of batch to batch comparison
Page 56
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Correlation CTQ and NIR fingerprint Micro-array analysis
Process monitoring and control
Measurement of presence of virulence factors (outer membrane proteins)
Does an identical mRNA expression profile correlate with an identical NIR profile? Does a disturbed mRNA expression profile correlate with a disturbed NIR profile? Is NIR suitable for on line monitoring of our batch cultivation process???
Expression analysis of virulence gene expression (process understanding): Microarray analysis Which genes respond to which process conditions? Expression profiling (process fingerprinting)
Page 57
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Process fingerprint as an advisory / supervision tool Process trajectory an operating limits Driving a process… to desired operating conditions within operating
Process control Process data
Temp., pH, pO2, pressure, …
MVDA (PCA)
constraints on the basis of best available knowledge NIR spectral data
on process characteristics
MVDA (PCA)
Start
Relationship between controllable process parameters and process trajectory End
Allows to steer the process on optimal trajectory
How does this relate to process parameters?
Page 58
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Design space Operating limits ?
Design space exploration
Relationship between process parameters and end-
Process data
Temp., pH, pO2, pressure, …
MVDA (PCA)
product quality / performance ?
NIR spectral data
Systematic approach to explore and document the
MVDA (PCA)
MVDA (PLS)
Qualitative Fingerprint
End-product Quality data LIMS
design space: Multi-factorial DoE
Golden batch trajectory Design space limits = Control limits
Page 59
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
The next challenge: the real-time PAT environment PAT System
Product & process design (Advanced) Process Controls
PAT
Data Collection storage & retrieval
Data Analysis & mining
Process Analytics
Page 60
Sept. 2008
PAT system requirements
Information management tools
DoE
All linked together into one environment / user interface Data collection: Synchronization of process data and analyser data Gathered of all analyser + sensor data in one database with aligned timestamps Data processing: On-line data processing, based on multivariate data analysis tools On-line comparison of processed data with with historic data Utilisation of the processed data for controlling the process online (closing the loop)
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
The PAT/QbD solution PAT Tools
Product & process design (Advanced) Process Controls
PAT system All linked together into one PAT/QbD system architecture, easy to integrate with development & manufacturing systems
PAT
Data Collection storage & retrieval
Data Portal
Information management tools
PAT Model builder MES
Data Analysis & mining
DoE
BATCH
SCADA
Process Analytics
Analyzers
Configuratio n
Data archive
ERP R&D ChemoChemometrics LIMS
Execute
Process
Offering one common user interface for all PAT tools
Page 61
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
PAT shifts monitoring and control from process data to product quality From design space to process control Predicting end-product quality Process data
PAT as part of the process control environment
Temp., pH, pO2, pressure, …
MVDA (PCA) NIR spectral data
MVDA (PCA)
MVDA (PLS) Predictive Model
Qualitative Fingerprint
End-product Quality data
Prediction Golden batch trajectory Control limits Actual batch progress
Based on real-time monitoring of NIR trajectories and process data, predict end product quality and perfomance CTQ parameter: Continuously measured OR Aperiodically measured OR Real time value Inferred from calibration model OR End-point value inferred from calibration model OR Scores of calibration model are CTQ parameters
Page 62
Sept. 2008
Early detection of process disturbances Process advisory Control correct bioreactor characteristics
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Example of a batch progress visualisation across different unit operations Offline Batch Monitoring Model 3-1-2004 - Predicted Scores [comp. 1]
10
0
ProdFerm
SeedFerm
LabFerm
-5
SeedFerm Hold and Trans
0
PreSeedFerm
tPS[1]
5
100
200 Num
300
Process is deviating SIMCA-P+ 10.5 - 5/10/2004 1:48:52 PM
: Actual batch trajectory : “golden” batch trajectory : Control limits
Page 63
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Example of a batch progress visualisation across different unit operations
520095_GR_A.M4:5 - Scores [comp. 1] (Aligned)
+3 Std.Dev t[1] (Avg) -3 Std.Dev t[1] (Aligned): 1000016_A
Unit Operation 1
t[1]
t[2]
520095_CO_A - batch level.M2 (PCA-X), FinalMonitor t[Comp. 1]/t[Comp. 2]
4
4.00
3
3.00
2
2.00
1
1.00
0
0.00
-1
-1.00
-2
-2.00
-3
-3.00
-4
-4.00
-5
-5.00
-6
-6.00
-7
-7.00
20
20.00
10
10.00
0
0.00
-10
-10.00
-20
-20.00
-40
-30
-20
-10
0
10
20
30
40
t[1]
Ellipse: Hotelling T2 (0.95)
520095_CO_A - batch level.M3 (PLS), Assay YPred[Comp. 8](YVar A1110-L20-O105-Assay_-_Liquid)/YVar(YVar A1110-L20-O105-Assay_-_Liquid)
101
1
2
3
4
5
6
7
8
9
10
Time ($Time) 520095_DR_A.M1 - Scores [comp. 1] (Aligned)
+3 Std.Dev t[1] (Avg) -3 Std.Dev t[1] (Aligned): 665002T_A
t[1]
Unit Operation 2 6
6.00
5
5.00
4
4.00
3
3.00
2
2.00
1
1.00
0
0.00
-1
-1.00
-2
-2.00
-3
-3.00
-4
-4.00
-1
0
1
2
3
4
5
6
7
8
9
10
11
12
13
520095_CO_A.M1 - Scores [comp. 1] (Aligned) 14 15 16 17 18 19 20 21 22
Unit Operation 3
23
24
100
100.00
99
99.00
98
98.00
98
99
100
101
YPred[8](A1110-L20-O105-Assay_-_Liquid)
RMSEE = 0.118192
Release +3 Std.Dev t[1] (Avg) -3 Std.Dev t[1] (Aligned): 665001T_A
25
Time ($Time)
t[1]
101.00
11 YVar(A1110-L20-O105-Assay_-_Liquid)
0
y=1.001*x-0.127 R2=0.9799
5
5.00
4
4.00
3
3.00
2
2.00
1
1.00
0
0.00
-1
-1.00
-2
-2.00
-3
-3.00
-4
-4.00
-5
-5.00
Raw Materials + LIMS
-6.00
-6
-7.00
-7 -1
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Time ($Time)
Page 64
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
The PAT/QbD system architecture
Confidential / Copyright © Siemens AG 2008. All reserved. © Siemens AG,rights All rights reserved
PAT shifts monitoring and control from process data to product quality
PAT Quality build in by design Right first time
Classic control Closed loop control
monitoring product quality
Real-time release
mathematical translation
Lab
monitoring process data
Temp., pH, pO2, pressure, …
LIMS
Hold / release
Advanced Control
Process feed Process Analyzer Sample Process output
Page 66
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
General communication between EBR, SIPAT, SPC, APC and Controller/HMI
EBR
SPC MES central realtime database
APC
PAT
Controller / HMI
Page 67
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
PAT / QbD in Manufacturing
Manufacturing ERP
MES LIMS Historian
Batch Execution Process Automation
Page6868 page
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Case: PAT/QbD in Manufacturing Example architecture
PAT on Manufacturing level
PAT on machine level
The overall Architecture is based on a distributed approach with a PAT/QbD software solution per process area
Page 69
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
PAT as part of the Development Architecture
PAT supports process development Collects process knowledge on: equipment/product interaction equipment behavior impact on final product quality To explore the design space
Allows to fasten process up-scaling and transfer (to manufacturing) Page 70
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
PAT / QbD in R&D
R&D Research
Development
Workflow Manager LIMS Workflow Manager MES
Lab automation DoE Tools
Page7171 page
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Case: PAT/QbD in Development Example architecture
PAT on Manufacturing level
PAT on machine level
Page 72
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
New Product Development
PAT/QbD can close the gap between development and manufacturing as:
a continuous process understanding and improvement tool to collect knowledge on: product performance (therapeutic) process / product interaction part of the knowledge hierarchy
Page 73
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Closing the gap: R&D and Manufacturing integration PLM
Knowledge Management System
Team Centre
Transform Knowledge Generator
Data Portal TeamCentre TeamCentre R&D Research
Manufacturing ERP
Development
Workflow Manager
MES Historian
MES
Batch Execution
LIMS
Workflow Manager
Lab automation
Dashboarding
LIMS
Process Automation
DoE Tools
Page 74
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Case: PAT / QbD as part of the overall system architecture R&D
Example: ERP
Knowledge Mgt
LIMS
Data Warehouse PLM
Page 75
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
The Regulatory changes
Today New initiatives to: improve manufacturing quality accelerate development Lower the regulatory burden FDA new principles: Quality by design & design space Quality systems approach Reflecting product & process understanding and knowledge
Page 76
Sept. 2008
Vision FDA’s future focus: Keynote address at IFPAC February 2007, by FDA's Chief Medical Officer, Dr. Janet Woodcock, on
Development & manufacturing should be integrated Development of quality surrogates for clinical performance (link critical product attributes to clinical outcomes) Rigorous, mechanistically based and statistically controlled processes Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Live-Licensing and e-CTD
Pharma company
Page 77
Sept. 2008
E-Submission
Regulatory bodies
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Conclusion
Confidential / Copyright © Siemens AG 2008. All reserved. © Siemens AG,rights All rights reserved
Conclusion: A multidisciplinary approach is the only way to success Required infrastructure
Required disciplines
MES
(Advanced) Controls
regulatory
Modeling
Chemometrics / MVDA
Process development
Process understanding
Process Analytics
Page 79
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Key Learning Points
Develop the future vision for manufacturing and development Select the PAT /QbD strategy that is supporting the vision Develop the PAT / QbD architecture that supports the vision Develop the PAT / QbD implementation strategy (roadmap)
Page 80
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Conclusions
PAT accelerates development and improves manufacturing PAT enables closing the gap between development and manufacturing Knowledge management tools will play a prominent role
Page 81
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Questions?
Ingrid Maes Siemens Competence Center Pharma Phone: +32 2 536 98 39 [email protected]
Page 82
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All reserved. © Siemens AG,rights All rights reserved Ingrid Maes Industry CC Pharma
Düsseldorf September, 2008
Ingrid Maes Siemens Competence Centre Pharma
Confidential / Copyright © Siemens AG 2008. All reserved. © Siemens AG,rights All rights reserved
Presentation outline: The Changing Pharma landscape in R&D and Manufacturing How can PAT/QbD be part of the manufacturing and development architecture? How can PAT/QbD be a continuous process understanding and improvement tool? How can PAT/QbD be an enabler for increasing profitability & productivity? How can PAT/QbD be an enabler for time-to-market?
Page 2
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
The changing pharma landscape and gap between R&D and Manufacturing
Confidential / Copyright © Siemens AG 2008. All reserved. © Siemens AG,rights All rights reserved
Changes and impacts on the pharmaceutical manufacturing marketplace
Social change
Regulations
Technology change
Demographic
Industrial IT
Life style
Advanced control
Patients
Integrated solutions
Pharmaceutical Industry
Economical pressure
driven by; Cost Patient safety
Market change
Low pipe line
New therapies
Manufacturing
Delivery form
Supply Chain
Personalize
Page 4
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Pharmaceutical companies have to change Price
Healthcare system Business
R&D low pipeline high costs
Page 5
Sept. 2008
Manufacturing inefficient quality problems
Sales / Marketing Expensive sales inefficient supply chain
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
The change requirement Price
Healthcare system Business
R&D Smarter More efficient
Page 6
Sept. 2008
Manufacturing Better Higher responsiveness
Sales Marketing more control
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Changes that affect R&D and manufacturing: Integrated and continuous R&D and approaval Now: Discovery
Lead
Pre-clinical
Screaning
Develop.
Evaluation
Phase I
Phase III
limited launch
Future:
Patho – physiology
Molecule Develoment
Page 7
Phase II
Sept. 2008
Submission
in-life testing
Submission
Phase IV
Manufacturing
=> Approval on a real-time basis, with live licences contingent on the performance of extensive in-life testing
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Personalized Medicine
Batch size Targeted / personalized
Smaller batches Shortening time to market Bio markers
Smaller and shorterrun Clinical Trials due to increased statistical relevance on selected patients groups.
Discovery & Research Page 8
Sept. 2008
Drug screening
Bioinformatics
Toxicology
Clinical Trials
upscale
produce
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
The change of the regulations
Now Now
Before
Co-operative regulation Goal fixed and route free
Based on three reference batches no change allowed
Based on risk analysis process understanding
regulations
manufacturing
Page 9
Sept. 2008
re gu lat io ns
Repressive regulation Goal and route fixed
manufacturing
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
The regulatory changes: FDA’s 21st Century Initiatives Today
Vision
New initiatives to: improve manufacturing quality accelerate development Lower the regulatory burden
FDA new principles: Quality by design & design space Quality systems approach Reflecting product & process understanding and knowledge
Page 10
Sept. 2008
FDA’s future focus: Keynote address at IFPAC February 2007, by FDA's Chief Medical Officer, Dr. Janet Woodcock, on
Development & manufacturing should be integrated Development of quality surrogates for clinical performance (link critical product attributes to clinical outcomes) Rigorous, mechanistically based and statistically controlled processes
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Moving Pharmaceutical Manufacturing into the 21st Century
“The pharmaceutical industry could be wasting more than $50 billion a year in manufacturing costs alone…” - Macher, Georgetown University, and Nickerson, Washington University,
2006
“…we’re using the evaluation tools and infrastructure of the last century, and, in some cases, tools from very early in the last century to develop this century’s medical advances.”
- Dr. Janet Woodcock, M.D., Deputy Director and Chief Medical Officer of FDA, 2007
Page 11
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Current manufacturing status:
Large inefficient batch equipment Low utilization 30 -40 % on average Low product yields Excessive amounts of product non-conformances Long lead-times due to stage and final product testing Capital and labour intensive High operating costs High inventories and excessive warehouse capacity Cycle time improvement perceived to be limited by regulatory constraints 1% yield improvement ~ $400M in savings
Page 12
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
A Changing Landscape
The Pharma industry is just at the beginning of a dramatic change Relative separation between: R&D and Manufacturing Healthcare services and pharma industry (except for clinical trials & sales activities) The Pharma industry is seeking for new technologies / concepts and alternative business models
Page 13
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
How things will change and what will be supporting technologies for the future vision? On short, medium and longer term
Confidential / Copyright © Siemens AG 2008. All reserved. © Siemens AG,rights All rights reserved
Technologies shaping the pharma R&D and manufacturing future Process Analytical Technology
E-labelling
Miniaturized Manufacturing Concepts Demand driven manufacturing
Organizational Network Analysis Condition based maintenance Modular Plant design
Disposable manufacturing technologies
Semantic Web
E-licensing
RFID
Lab on a chip Master Data Management
Electronic Data Capture
PLM
Simulation and Virtual Prototyping
Continuous ManufacturingConcepts
Cell on a chip
Electronic Batch Management Clinical Trial Management Page 15
Sept. 2008
Recipe driven manufacturing
SPC
E-patient diaries R&D Workflow Management Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
New technologies and alternative strategies that are shaping future pharma New technologies
What is the strategic response?
PAT/ Quality by Design PLM SPC
Build-in quality
Continuous Mftg. Concepts Miniaturized manufacturing PAT / real-time product release Condition based maintenance
Increase manufacturing performance & throughput
Demand driven manufacturing Modular plant design Recipe driven manufacturing Disposable manufacturing
Flexible manufacturing concepts
Clinical trial management WFM, Data portals e-CTD PLM Bioinformatics
Speed-up development & closing the gap
Page 16
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Pharma Manufacturing 2020
Confidential / Copyright © Siemens AG 2008. All rights reserved.
How will companies cope with the changing environment? What will be future challenges?
What is the strategic response?
Changes due to new production technologies Increase operating efficiency Reduce product costs / achieve competitive pricing (e.g. response to biogenerics) Produce individual products / address niche markets (Personalized Medicine) Accelerate time-to-market
Scenario 1: Modernize within existing facility But, essentially same approach & scale
Scenario 2: Continuous processing, RTPR, JIT production
5 year
Page 18
Sept. 2008
What will be the implications on the way we manufacture in future?
Manufacturing strategic leaps to the medium and longer term: Scenario 3 = Speciality Niche products: Small scale pilot centers, Integration of R&D and production: Small batches 24/7 running Scenario 4 = Gross / mass market: Large-scale highly flexible plants, with high throughput
10 year
Move to personalized medicines Clinical and patient feed-back loops Continuous optimization and improvement
5 year
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Short term future (within 5 years) What will be future challenges? What will be the implications on the way we manufacture in future?
What is the strategic response? Increase operational efficiency (OEE) PAT (Process Analytical Technology) / QbD Electronic data acquisition and Data Management EBR MES PLM Real-time enterprise (data integration and management, dashboards / cockpits, facilitating real-time decision making)
Scenario 1: Modernize within existing facility But, essentially same approach & scale
5 year
Page 19
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
PAT/QbD a key enabling technology for the future
Confidential / Copyright © Siemens AG 2008. All reserved. © Siemens AG,rights All rights reserved
Taking pharma manufacturing into the 21st century through PAT / quality by design: the PAT journey PAT / Quality by Design is about starting an expedition. Taking pharma manufacturing into the 21st century Offers the pharmaceutical industry the possibility to increase product quality consistency and to reduce product risk through
Basecamp
increased process knowledge & understanding optimized process control
with the use of PAT tools
Page 21
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
PAT / QbD
New regulations FDA’s new initiatives to: improve manufacturing quality accelerate development lower the regulatory burden
FDA’s new principles: risk based approach scientific approach, based on process understanding
PAT (Process Analytical Technology) PAT = Understanding + controlling the manufacturing process A process is well understood when: All critical sources of variability are identified and explained Variability is managed by the process Product quality attributes can be accurately and reliably predicted Process Understanding is inversely proportional to risk
Page 22
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
The PAT guidance Process control trough new technologies (innovations), focus on manufacturing science A system for designing (process development), analyzing and controlling manufacturing processes, based on timely measurements of critical Q & performance attributes of raw-materials, in-process materials and processes with the goal of ensuring final product Q. Processes to assure acceptable end-product Q at the completion of the process (quality by design)
September 2004
Focus of PAT is understanding
Page 23
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
PAT & QbD: Quality-by-Design: Design of Product and Manufacturing Process Novel Technologies Platform Technologies Process Analytical Technology Self-regulating systems
Page 24
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Conventional Q approach Raw mats./ conditioning
Formulation
Finishing
Packaging
QC-system: lab based (End of phase testing of Q)
Page 25
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
PAT-based approach Raw mats./ conditioning
Bioreactor / Downstream
Formulation
Packaging
QC-system: PAT based (in-process QC / process based QC)
Adding value with Q by decision: Intelligent use of process data Advanced supervisory control Faster cycle times / Real-time product release Page 26
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
PAT the skill set and architecture fit PAT Toolbox / Skill set
Product & process design (Advanced) Process Controls
PLM
Data Collection storage & retrieval
Process Analytics
Sept. 2008
regulations
PAT
Data Analysis & mining
Page 27
Fit in development & manufacturing landscape
Information management tools
DoE
Process Automation
Field equipment
PAT
Data Portals/ Knowledge Mgt.
MES
Process technology
Process Analytics
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
PAT shifts monitoring and control from process data to product quality
PAT Quality build in by design Right first time
Classic control Closed loop control
monitoring product quality
Real-time release
mathematical translation
Lab
monitoring process data
Temp., pH, pO2, pressure, …
LIMS
Hold / release
Advanced Control
Process feed Process Analyzer Sample Process output
Page 28
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Time-based information Management Manufacturing Architecture
Time based information management (PAT system, MIS):
LIMS SPC
Quality
Quality Suite
ERP
Quality Suite
MES
Data is collected throughout the
SCADA
manufacturing time period
PAT
Captures process parameter measurements
Local Operation
Captures quality attribute measurements Data export by time period or by “batch”
Automation System
PAT
as required
Sensor/ Field Equipment Unit A
Page 29
Unit B
Sept. 2008
Unit Y
Unit Z
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Quality Suite Manufacturing Architecture PLM
LIMS SPC
Quality
Quality Suite
SPC
Overall Dashboard
ERP
Quality Suite, contains: On shopfloor (in manufacturing): PAT LIMS M-SPC/SQC Business intelligence suite (Quality cockpit) On boardroom level SPC Dashboards Central data warehouse + integration layer to R&D and manufacturing PLM and Knowledge Management
SCADA / MES
PAT
Local Operation Automation System
PAT
Quality Suite
Sensor/ Field
All integrated from shopfloor to boardroom Equipment Unit A
Page 30
Unit B
Sept. 2008
Unit Y
Unit Z
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Where we are today ? Challenge: Continuous Optimization and improvement Continuous manufacturing Just-In-Time Production Demand driven manufacturing Regulatory strategy
Today 2004 Investment in individual PAT tools
ROI
First PAT System implementations Page 31
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
What is the true gain of PAT?
Time-To-Market
Supply Chain Improvement
Improved market responsiveness Reduction of production cycle times Decrease warehousing
Fast process development Fast product release (real-time product release)
Project
Production
Asset valuation profit
Heart of the supply chain
Revenue
Development
cost costreduction reduction
reduction reduction time timeto tomarket market
primary
formulation
packaging
warehousing
by
Improving Improving manufacturing manufacturing performance performance
end of patent life
Deliver years
0
Drug innovation & approval
Page 32
5
Sept. 2008
make
deliver source
manufacturing
20
10
Project
source
Production
Revamp
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Business Drivers for PAT/QbD
Company Image Reduced risk via technology platform, anticounterfeiting Improved product tracking Reverse poor image Improved quality system throught audits Reduced risk fo recall, warning letter, consent decree
Validation Optimization Validation needs understanding Integral part of project Built validation into process
PAT/QbD Improve Existing Process Gain new process understanding Process optimization Reduced cost of quality Raw material specifications Know product availability + yield Real Time Release
Page 33
Sept. 2008
End of Life-cycle Transferability of process Scale down
Site to site transfer Accelerate transfer Reduce validation effort Reduce project time Mitigate transfer risk Move manufacturing to most effective site
New Product Development Real Time Release (RTR) Fast time to market Fast scale-up Clinical batches Process optimization Reduced cost of quality
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Improve Existing Process
Manufacturing accounts for the highest cost factor:
Research Driven Pharma Companies
31 %
Generic Manufacturers
51 %
Contract Manufacturers
62 %
Source: St Gallen report, 2006 Page 34
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
New Product Development The vast majority of investigational products that enter clinical trials fail PAT/QbD in manufacturing of clinical trial batches will reduce clinical trial failure, because:
Production more consistent products
Page 35
Sept. 2008
Reduced clinical batch product variability and hence “cleaner” clinical trial results
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
PAT / QbD Benefits
Impact on KPI’s
Business value Reduce Work In Process (WIP) costs: 33% Reduced warehouse space Reduced scrap/rework: 25 % Real-time product release Reduced labor costs: 30% Increased & consistent product quality Reduced quality costs: 13% Reduced regulatory compliance costs: 70% Reduced clinical batch product variability and hence “cleaner” clinical trial results Faster time to market: scale-up & tech transfer Continuous process understanding
Page 36
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Case: Results of Business Case Background
Results
A vaccine plant is seeking to achieve cost savings through modernisation of manufacturing infrastructure and implementing Quality by Design principles
Page 37
Sept. 2008
The result was a calculation of the optimal scenario (scenario 3): Labour: ¼ of operations people could be re-allocated Manufacturing throughput time: throughput time decreased with 1/3 Quality: 13% of the cost of QA and QC are eliminated Waste reduction: 3.5% Inventory: inventory could be reduced by 1/3
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
PAT strategy development
Page 38
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
PAT/QbD implementation A step-wise approach
Confidential / Copyright © Siemens AG 2008. All reserved. © Siemens AG,rights All rights reserved
PAT implementation roadmap Philosophy of Recent FDA guidances: Not a typical guidance Description of the desired state Path to the desired state is up to the industry Desired state: Process understanding
PAT Framework
Standards & best practices
Page 40
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
The PAT Implementation Roadmap
Page 41
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Secondary process
Primary process
Process assessment (CTQ steps)
Page 42
Buffer + Medium
Raw mats./ Conditioning
Sept. 2008
Critical process steps, in terms of critical to end-product quality
Bioreactor
Separation
Formulation
Conditioning/ Finishing
Purification
Packaging
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Primary process
Optimisation over the complete process Buffer + Medium
Synthesis / Bioreactor
Separation
Purification
Secondary process
Optimizing throughput, yield, quality, efficiency
Page 43
Raw mats./ Conditioning
Sept. 2008
Formulation
Conditioning/ Finishing
Packaging
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Case: PAT at NVI (Netherlands Vaccine Institute)
Customer process and backgrounds: Pertusses (Whooping coughing disease) 50 year old process quality test unchanged 50 year waste 25% Procedure (simplified): Step 1 - definition critical to quality parameter Step 2 - design experiments Step 3 - execute tests Step 4 - modeling Step 5 - implement new strategy Resulted in: better process understanding improvement of quality reduce waste by 20% increase production yield by a factor 3 potentially eliminate animal testing
Page 44
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Group / Region Department Ingrid Maes Industry/ CC Pharma
Search for Critical To Quality (CTQ) process steps What affects product safety, quality and efficacy ? Critical product specifications …
… which CTQ attributes ? What are the CTQ attrributes that describe the quality and performance of the product?
? y lit a ic t i Cr Where is product sensitive for manufacturing variation (root cause)?
Availability of Critical To Quality (CTQ) measurements ? Are CTQ parameters continuously measured? Are CTQ parameters measured by lab assay at end or during batch? Can CTQ parameters be inferred from PAT sensor with calibration model?
What to do to design out risk? Case: Whole cell vaccine against B. pertussis infection: Inactivated cells are the actual product Outer membrane proteins are presented to immune system Outer membrane composition crucial for vaccine efficacy
Page 45
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Case: CTQ for pertussis vaccine CTQ for pertussis vaccine …
… How is CTQ determined ? Microarrays
Protein analysis
Vaccine performance is determined by presence of relevant outer membrane proteins Virulence activated genes (VAGs) are responsible for outer membrane proteins (Adhesins, toxins) Virulence gene expression : protective vaccine Virulence gene expression : non-protective vaccine = product quality market genes VAG expression depends on extracellular conditions
DNA microarrays for expression profiling Protein analysis (Western Blotting, ELISA) Supernatant analysis with 1H and 13C NMR
Classical (animal) tests do not provide enough discriminative power for assessment of CQA’s
Page 46
Sept. 2008
Protein profiling (nano LC-MS, new)
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Monitoring Critical Attributes CTQ for pertussis vaccine …
… what do we have ? pH, DO and T are timely but insufficient
Objective : understanding and controlling the variance that affect product quality during
Nutrients & metabolites with NMR
cultivation Explore (process & product) state of the art assays for product and process Define critical process parameters How can we control the process to yield the desired end product quality? On line monitoring A signal that reflects the critical process parameters
Page 47
Sept. 2008
mRNA and protein analysis are sufficient but not timely
Animal testing for end-product release
Online monitoring of critical attributes necessary. But which technology?
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Select appropriate process analyser technology Case: Bioreactor Which real-time monitoring technology?
Laser Laser Diode Diode Spectrom. Spectrom. Flow Flow
Temperature Temperature
Pressure Pressure
Level Level
Mass Spectroscopy: composition, ...
Liquid Liquid Analytics Analytics
Weighing Weighing Technology Technology
Positioners Positioners
Selected technologies for bioreactor monitoring
M
Gas Gas ChromaChromatography tography Mass Mass Spectroscopy Spectroscopy
NIR NIR Spectroscopy Spectroscopy Gas Gas Analytics Analytics
Medium /Substrate Buffer/Acid /Lye
IR IR Spectroscopy Spectroscopy Raman Raman Spectroscopy Spectroscopy
NMR NMR Spectroscopy Spectroscopy
Off gas
Inoculum Bioreactor
Laser Laser diffraction diffraction Spectroscopy Spectroscopy
Product NIR Spectroscopy: composition, Biologic performance, ...
Page 48
Sept. 2008
Air/O2/CO2
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Case: Real-time & in-situ Bioreactor monitoring with NIR
In situ NIR transmission probe
Page 49
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Use of NIR as quantitative tool for bioreactor monitoring Physical, Chemical Status
Biomass concentration monitoring:
Early detection of endpoint
Loga r ithm of B iom a s s
Monitoring of nutrient (glutamate, lactate) consumption and metabolite formation with NIR:
Biological Status
Biomass
1
2
3
4
5
6
7
Time
2. Lag Phase 3. Transient Acceleration 4. Exponential Phase 5. Stationary Phase 6. Death Phase / necrosis
Page 50
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Special case: Real-time infection detection with NIR
Early detection of disturbances
Example: Yeast culture
For control or fault detection: Contaminations by foreign microorganisms
Infected (102 CFU/ml)
B. subtilis, S. aureus, Ps. Aeruginosa,
Non infected
C. albicans, E. coli, A. niger and L. Brevis
Biological contamination
Page 51
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Real-time Product Release = measure & control “ultimate quality” From critical product specifications …
critical product specification
… to critical process specifications
critical process specification (requirements to be met to consider the process “under control”)
Page 52
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Use of NIR as qualitative tool for bioreactor monitoring Physical, chemical and biological status
Total bioreactor monitoring
No quantification of analytes, the overall behavior is assessed Output is a “process fingerprint” representing the behavior of the batch process MVDA (PCA)
NIR spectral data
Start
Bioreactor pathway monitoring
End
Change in process status during batch progress = Process trajectory / fingerprint / signature (multidimensional profile) Control correct bioreactor characteristics Early detection of disturbances For control or fault detection
Scores plot for the two main principal components on collected spectra during the batch process
Page 53
Sept. 2008
Compare different batches
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Use of NIR as qualitative tool for mixing monitoring Physical, chemical status
Total blending monitoring
No quantification of analytes, the overall behavior is assessed Output is a “process fingerprint” representing the behavior of the batch process MVDA (PCA)
NIR spectral data
0.003
PC2
Scores
26 0.002
19 20
24
23
31 27 33 25 28
22 21
36 35
3439
38 37 42 46
18
0.001
40 43 41
17
44
16 0
2 7 13 10 6 5 3 12 14 4 8
45 47 50 51 53 49 48 70 52 55 68 5654 90 71 7588 8972 74 62 76 8077 60 59 8587 66 5783 65 6486 82 69 7873 63 7981 58
End
1 -0.001
15
9
Start
-0.002
Change in process status during batch progress = Process trajectory / fingerprint / signature (multidimensional profile)
30 32
29
84
Control correct mixing characteristics
67
61
11
-0.003 -0.06 -0.05 RESULT5, X-expl: 100%,0%
PC1 -0.04
-0.03
-0.02
-0.01
0
0.01
0.02
Early detection of disturbances
0.04
For control or fault detection
Scores plot for the two main principal components on collected spectra during the batch process
Page 54
0.03
Sept. 2008
Compare different batches
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Batch to batch comparison Batch Start
Batch Endpoint (out of spec) Batch Endpoint (in spec)
Batch Endpoint (Product quality reached)
Batch 1 Batch 5
Batch 4
Batch 2
Batch 3 Page 55
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Case: Another example of batch to batch comparison
Page 56
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Correlation CTQ and NIR fingerprint Micro-array analysis
Process monitoring and control
Measurement of presence of virulence factors (outer membrane proteins)
Does an identical mRNA expression profile correlate with an identical NIR profile? Does a disturbed mRNA expression profile correlate with a disturbed NIR profile? Is NIR suitable for on line monitoring of our batch cultivation process???
Expression analysis of virulence gene expression (process understanding): Microarray analysis Which genes respond to which process conditions? Expression profiling (process fingerprinting)
Page 57
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Process fingerprint as an advisory / supervision tool Process trajectory an operating limits Driving a process… to desired operating conditions within operating
Process control Process data
Temp., pH, pO2, pressure, …
MVDA (PCA)
constraints on the basis of best available knowledge NIR spectral data
on process characteristics
MVDA (PCA)
Start
Relationship between controllable process parameters and process trajectory End
Allows to steer the process on optimal trajectory
How does this relate to process parameters?
Page 58
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Design space Operating limits ?
Design space exploration
Relationship between process parameters and end-
Process data
Temp., pH, pO2, pressure, …
MVDA (PCA)
product quality / performance ?
NIR spectral data
Systematic approach to explore and document the
MVDA (PCA)
MVDA (PLS)
Qualitative Fingerprint
End-product Quality data LIMS
design space: Multi-factorial DoE
Golden batch trajectory Design space limits = Control limits
Page 59
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
The next challenge: the real-time PAT environment PAT System
Product & process design (Advanced) Process Controls
PAT
Data Collection storage & retrieval
Data Analysis & mining
Process Analytics
Page 60
Sept. 2008
PAT system requirements
Information management tools
DoE
All linked together into one environment / user interface Data collection: Synchronization of process data and analyser data Gathered of all analyser + sensor data in one database with aligned timestamps Data processing: On-line data processing, based on multivariate data analysis tools On-line comparison of processed data with with historic data Utilisation of the processed data for controlling the process online (closing the loop)
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
The PAT/QbD solution PAT Tools
Product & process design (Advanced) Process Controls
PAT system All linked together into one PAT/QbD system architecture, easy to integrate with development & manufacturing systems
PAT
Data Collection storage & retrieval
Data Portal
Information management tools
PAT Model builder MES
Data Analysis & mining
DoE
BATCH
SCADA
Process Analytics
Analyzers
Configuratio n
Data archive
ERP R&D ChemoChemometrics LIMS
Execute
Process
Offering one common user interface for all PAT tools
Page 61
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
PAT shifts monitoring and control from process data to product quality From design space to process control Predicting end-product quality Process data
PAT as part of the process control environment
Temp., pH, pO2, pressure, …
MVDA (PCA) NIR spectral data
MVDA (PCA)
MVDA (PLS) Predictive Model
Qualitative Fingerprint
End-product Quality data
Prediction Golden batch trajectory Control limits Actual batch progress
Based on real-time monitoring of NIR trajectories and process data, predict end product quality and perfomance CTQ parameter: Continuously measured OR Aperiodically measured OR Real time value Inferred from calibration model OR End-point value inferred from calibration model OR Scores of calibration model are CTQ parameters
Page 62
Sept. 2008
Early detection of process disturbances Process advisory Control correct bioreactor characteristics
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Example of a batch progress visualisation across different unit operations Offline Batch Monitoring Model 3-1-2004 - Predicted Scores [comp. 1]
10
0
ProdFerm
SeedFerm
LabFerm
-5
SeedFerm Hold and Trans
0
PreSeedFerm
tPS[1]
5
100
200 Num
300
Process is deviating SIMCA-P+ 10.5 - 5/10/2004 1:48:52 PM
: Actual batch trajectory : “golden” batch trajectory : Control limits
Page 63
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Example of a batch progress visualisation across different unit operations
520095_GR_A.M4:5 - Scores [comp. 1] (Aligned)
+3 Std.Dev t[1] (Avg) -3 Std.Dev t[1] (Aligned): 1000016_A
Unit Operation 1
t[1]
t[2]
520095_CO_A - batch level.M2 (PCA-X), FinalMonitor t[Comp. 1]/t[Comp. 2]
4
4.00
3
3.00
2
2.00
1
1.00
0
0.00
-1
-1.00
-2
-2.00
-3
-3.00
-4
-4.00
-5
-5.00
-6
-6.00
-7
-7.00
20
20.00
10
10.00
0
0.00
-10
-10.00
-20
-20.00
-40
-30
-20
-10
0
10
20
30
40
t[1]
Ellipse: Hotelling T2 (0.95)
520095_CO_A - batch level.M3 (PLS), Assay YPred[Comp. 8](YVar A1110-L20-O105-Assay_-_Liquid)/YVar(YVar A1110-L20-O105-Assay_-_Liquid)
101
1
2
3
4
5
6
7
8
9
10
Time ($Time) 520095_DR_A.M1 - Scores [comp. 1] (Aligned)
+3 Std.Dev t[1] (Avg) -3 Std.Dev t[1] (Aligned): 665002T_A
t[1]
Unit Operation 2 6
6.00
5
5.00
4
4.00
3
3.00
2
2.00
1
1.00
0
0.00
-1
-1.00
-2
-2.00
-3
-3.00
-4
-4.00
-1
0
1
2
3
4
5
6
7
8
9
10
11
12
13
520095_CO_A.M1 - Scores [comp. 1] (Aligned) 14 15 16 17 18 19 20 21 22
Unit Operation 3
23
24
100
100.00
99
99.00
98
98.00
98
99
100
101
YPred[8](A1110-L20-O105-Assay_-_Liquid)
RMSEE = 0.118192
Release +3 Std.Dev t[1] (Avg) -3 Std.Dev t[1] (Aligned): 665001T_A
25
Time ($Time)
t[1]
101.00
11 YVar(A1110-L20-O105-Assay_-_Liquid)
0
y=1.001*x-0.127 R2=0.9799
5
5.00
4
4.00
3
3.00
2
2.00
1
1.00
0
0.00
-1
-1.00
-2
-2.00
-3
-3.00
-4
-4.00
-5
-5.00
Raw Materials + LIMS
-6.00
-6
-7.00
-7 -1
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Time ($Time)
Page 64
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
The PAT/QbD system architecture
Confidential / Copyright © Siemens AG 2008. All reserved. © Siemens AG,rights All rights reserved
PAT shifts monitoring and control from process data to product quality
PAT Quality build in by design Right first time
Classic control Closed loop control
monitoring product quality
Real-time release
mathematical translation
Lab
monitoring process data
Temp., pH, pO2, pressure, …
LIMS
Hold / release
Advanced Control
Process feed Process Analyzer Sample Process output
Page 66
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
General communication between EBR, SIPAT, SPC, APC and Controller/HMI
EBR
SPC MES central realtime database
APC
PAT
Controller / HMI
Page 67
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
PAT / QbD in Manufacturing
Manufacturing ERP
MES LIMS Historian
Batch Execution Process Automation
Page6868 page
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Case: PAT/QbD in Manufacturing Example architecture
PAT on Manufacturing level
PAT on machine level
The overall Architecture is based on a distributed approach with a PAT/QbD software solution per process area
Page 69
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
PAT as part of the Development Architecture
PAT supports process development Collects process knowledge on: equipment/product interaction equipment behavior impact on final product quality To explore the design space
Allows to fasten process up-scaling and transfer (to manufacturing) Page 70
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
PAT / QbD in R&D
R&D Research
Development
Workflow Manager LIMS Workflow Manager MES
Lab automation DoE Tools
Page7171 page
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Case: PAT/QbD in Development Example architecture
PAT on Manufacturing level
PAT on machine level
Page 72
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
New Product Development
PAT/QbD can close the gap between development and manufacturing as:
a continuous process understanding and improvement tool to collect knowledge on: product performance (therapeutic) process / product interaction part of the knowledge hierarchy
Page 73
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Closing the gap: R&D and Manufacturing integration PLM
Knowledge Management System
Team Centre
Transform Knowledge Generator
Data Portal TeamCentre TeamCentre R&D Research
Manufacturing ERP
Development
Workflow Manager
MES Historian
MES
Batch Execution
LIMS
Workflow Manager
Lab automation
Dashboarding
LIMS
Process Automation
DoE Tools
Page 74
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Case: PAT / QbD as part of the overall system architecture R&D
Example: ERP
Knowledge Mgt
LIMS
Data Warehouse PLM
Page 75
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
The Regulatory changes
Today New initiatives to: improve manufacturing quality accelerate development Lower the regulatory burden FDA new principles: Quality by design & design space Quality systems approach Reflecting product & process understanding and knowledge
Page 76
Sept. 2008
Vision FDA’s future focus: Keynote address at IFPAC February 2007, by FDA's Chief Medical Officer, Dr. Janet Woodcock, on
Development & manufacturing should be integrated Development of quality surrogates for clinical performance (link critical product attributes to clinical outcomes) Rigorous, mechanistically based and statistically controlled processes Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Live-Licensing and e-CTD
Pharma company
Page 77
Sept. 2008
E-Submission
Regulatory bodies
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Conclusion
Confidential / Copyright © Siemens AG 2008. All reserved. © Siemens AG,rights All rights reserved
Conclusion: A multidisciplinary approach is the only way to success Required infrastructure
Required disciplines
MES
(Advanced) Controls
regulatory
Modeling
Chemometrics / MVDA
Process development
Process understanding
Process Analytics
Page 79
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Key Learning Points
Develop the future vision for manufacturing and development Select the PAT /QbD strategy that is supporting the vision Develop the PAT / QbD architecture that supports the vision Develop the PAT / QbD implementation strategy (roadmap)
Page 80
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Conclusions
PAT accelerates development and improves manufacturing PAT enables closing the gap between development and manufacturing Knowledge management tools will play a prominent role
Page 81
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All rights reserved. Ingrid Maes Industry CC Pharma
Questions?
Ingrid Maes Siemens Competence Center Pharma Phone: +32 2 536 98 39 [email protected]
Page 82
Sept. 2008
Confidential / Copyright © Siemens AG 2008. All reserved. © Siemens AG,rights All rights reserved Ingrid Maes Industry CC Pharma
More Documents from "Narendrakumar"
Thirunellaivenkateshwaranichq8andq9areview
December 2019 19
Ichq8, 9 & 10 The History And Overview
December 2019 26
Implementation Of Ichq8 Q9 & Q10 Guidelines
December 2019 28
Gmp In Pharmaceutical Processing
December 2019 27
2 1six Sigma
December 2019 19