1. Process Analytical Technology (PAT) Incorporating Process Analytical Technologies Framework into the plant to manage risk

2. Challenge Pharmaceutical Analysis: Fall 20142  Worlds most complex bioreactor: Homosapien  Small variations can have catastrophic effects DevelopmentClinicalDiscovery Pharmaceutical Analysis Tools help maintain safety and quality throughout

3. Pharmaceutical CGMPs for the 21 st Century: A Risk-Based Approach.  An initiative started in August 2002, recognizing the need to eliminate the hesitancy to innovate, the Food and Drug Administration (FDA) launched a new initiative entitled  CGMP’s = Current Good Manufacturing processes  The most up-to-date concepts of risk management and quality systems approaches are incorporated into the manufacture of pharmaceuticals while maintaining product quality  Manufacturers are encouraged to use the latest scientific advances in pharmaceutical manufacturing and technology  The Agency's submission review and inspection programs operate in a coordinated and synergistic manner  Regulations and manufacturing standards are applied consistently by the Agency and the manufacturer  Management of the Agency's Risk-Based Approach encourages innovation in the pharmaceutical manufacturing sector  Agency resources are used effectively and efficiently to address the most significant health risks

4. Pharmaceutical CGMPs for the 21st Century: FDA Guidance on PAT  Released in September 2004.  Ajaz S. Hussain, Ph.D.  Previously Deputy Directory Office of Pharmaceutical Science CDER, FDA  Purpose:  Help remove confusion in industry about the PAT initiative  Historically called Process analytical chemistry (PAC), name change did not help implementation  Companies were not willing to “risk” gathering more data for fear of what to do with the data on non-validated PAT technologies

5. The “Scope”  The framework is founded on process understanding to facilitate innovation and risk-based regulatory decisions by industry and the Agency  The framework has two components:  a set of scientific principles and tools supporting innovation  a strategy for regulatory implementation that will accommodate innovation  The regulatory implementation strategy includes creation of a PAT Team approach to chemistry manufacturing and control (CMC) review and current good manufacturing practice (CGMP) inspections as well as joint training and certification of PAT review and inspection staff.  Note: Team = Agency and corporations ensuring a common understanding

6. The “Background”  Conventional pharmaceutical manufacturing is generally accomplished using batch processing with laboratory testing conducted on collected samples to evaluate quality.  This conventional approach has been successful in providing quality pharmaceuticals to the public.  However, today significant opportunities exist for improving pharmaceutical development, manufacturing, and quality assurance through innovation in product and process development, process analysis, and process control.  Efficient pharmaceutical manufacturing is a critical part of an effective U.S. health care system. The health of our citizens (and animals in their care) depends on the availability of safe, effective, and affordable medicines.  Unfortunately, the pharmaceutical industry generally has been hesitant to introduce innovative systems into the manufacturing sector for a number of reasons

7. PAT Framework  The goal of PAT is to enhance understanding and control the manufacturing process, which is consistent with our current drug quality system: quality cannot be tested into products; it should be built-in or should be by design.  Consequently, the tools and principles described in this guidance should be used for gaining process understanding and can also be used to meet the regulatory requirements for validating and controlling the manufacturing process The Agency considers PAT to be a system for designing, analyzing, and controlling manufacturing through timely measurements (i.e., during processing) of critical quality and performance attributes of raw and in-process materials and processes, with the goal of ensuring final product quality.

8. Building quality into products  The intended therapeutic objectives; patient population; route of administration; and pharmacological, toxicological, and pharmacokinetic characteristics of a drug  The chemical, physical, and biopharmaceutic characteristics of a drug  Design of a product and selection of product components and packaging based on drug attributes listed above  The design of manufacturing processes using principles of engineering, material science, and quality assurance to ensure acceptable and reproducible product quality and performance throughout a product's shelf life Process Understanding

9.  A process is generally considered 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 over the design space established for:  Materials used  Process Parameters  Manufacturing unit operations  Environmental impacts  “Other considerations”  Although retrospective process capability data are indicative of a state of control, these alone may be insufficient to gauge or communicate process understanding.

10. Continuous Learning  Structured product and process development on a small scale, using experimental design or in-line process analyzers to collect data in real time, can provide increased insight and understanding for:  process development  Optimization  Scale-up  Technology transfer  Control  Process understanding continues in the production phase when other variables may be encountered  Supplier changes

11. Principles and Tools  Pharmaceutical manufacturing processes often consist of a series of unit operations, each intended to modulate certain properties of the materials being processed.  To ensure acceptable and reproducible modulation, consideration should be given to the quality attributes of incoming materials and their process-ability for each unit operation Unit Operation A Unit Operation B Unit Operation C Independent Variables Material Attributes Equipment Settings Independent Variables Dependent Variables (Material Attributes) Dependent Variables (Material Attributes) Dependent Variables (Material Attributes)

12. Process Understanding and Recipes  Process understanding leads to a set of instructions that if followed will provide materials of defined quality attributes  These set of instructions are developed in the R&D phases and transferred to plant ideally as a recipe  Recipe defines how resources come together in a sequence of steps to make a product or measurement  Recipe provides a proven framework to facilitate technology transfers  A recipe typically defines a unit operation and/or analytical quality testing process  Both have material genealogy at it’s core to ensure process understanding is transferred into a control strategy for the plant

13. Principles and Tools  Pharmaceutical manufacturing processes often consist of a series of recipes, each intended to modulate certain properties of the materials being processed.  To ensure acceptable and reproducible modulation, consideration should be given to the quality attributes of incoming materials and their process-ability for each recipe Recipe ARecipe BRecipe C Independent Variables Material Attributes Equipment Settings Independent Variables Dependent Variables (Material Attributes) Dependent Variables (Material Attributes) Dependent Variables (Material Attributes)

14. PAT Tools  There are many tools available that enable process understanding for scientific, risk-managed pharmaceutical development, manufacturing and quality assurance.  These tools when used within a system can provide effective means for acquiring information to facilitate process understanding, continuous improvement and development of risk-mitigation strategies.  In PAT framework, these tools can be categorized by:  Multivariate tools for design, data acquisition and analysis  Process analyzers  Process control tools  Continuous improvement and knowledge management tools  An appropriate combination of some, or all, of these tools may be applicable to a single-unit operation, or to an entire manufacturing process and its quality assurance

15. Multivariate tools  From a physical, chemical, or biological perspective, pharmaceutical products and processes are complex multi- factorial systems. Equipment Process Environment Materials Measurement People Output y = f(x) y Variability - source of the big risks to the product Inputs Variables (X) Inputs to the process control the variability of the output

16. Importance of Knowledge Management  The knowledge acquired in these development programs is the foundation for product and process design  It is noted in other sections of the guidance that continuous learning and scale up scale down models can help facilitate the overall process knowledge.  A knowledge base can be of most benefit when it consists of scientific understanding of the relevant multi-factorial relationships (e.g., between formulation, process, and quality attributes), as well as a means to evaluate the applicability of this knowledge in different scenarios  This benefit can be achieved through the use of multivariate mathematical approaches, such as statistical design of experiments, response surface methodologies, process simulation, and pattern recognition tools, in conjunction with knowledge management systems.  The applicability and reliability of knowledge in the form of mathematical relationships and models can be assessed by statistical evaluation of model predictions.

17. Building Knowledge based on Data  Capture data in a structured format: Recipe  Aggregate that data in a common location: Data Warehouse  Understand relationships: Multivariate Analysis Ref:http://www.systems-thinking.org/dikw/dikw.htm S88/S95 Data Warehouse Multivariate Analysis Recipe Framework

18. Process Analyzers  Process analyzers are embedded analytical technologies into the manufacturing process to :  Monitor – collect data passively to aid in process understanding  Control – collect and analyze data to control the outcome of the process  Release product (“real time release”) – collect, analyze and use data to release the product in real time  Optimization - variation of input parameters to achieve a desired outputs  Typically require Multivariate tools to analyze  Requires knowledge management system to manage quantity of data/information

19. How to deploy PAT Framework  Hardware:  PAT probes are well developed and deployed in food and chemical production  People  In drug development technologists that are not afraid of learning curve both for the tools and the equipment  Learning what a probe can measure may be product dependent and/or independent.  pH, Temperature, Pressure, p0 2  Chemical composition – requires model building  Production – must be practical and open minded  Software  Helps build complex models to aid scientists in correlating signals to process parameters

20. Traditional PAT Analyzers  Don’t forget the basics – you may already have the probes in place to measure your process!  Temperature  pH  Pressure  Humidity  Dew Point  Air flow  Common to use more “advanced” PAT probes to help understand fundamentals of your process and leverage traditional analyzers  Cost effective, helps minimize plant impacts

21. Advanced PAT Analyzers  Near Infrared  Inexpensive, probes molecular absorption (chemical bonds)  Raman Spectroscopy  to observe vibrational, rotational, and other low-frequency modes in a system (molecular vibrations)  Focused Beam Reflectance Measurements  Measure particle size in system (API or formulation/suspension)  Fluorescence  Acoustic  Others

22. NIR/Raman Comparison Near Infrared Raman  Absorption process promotes electron from ground state to higher energy level state  Electromagnetic radiation is “perturbed” (i.e., shifted up or down) due to stokes or anti-stokes raman scattering

23. NIR/Raman Spectra Water insensitive High S/N Water absorbs Broad spectra

24. NIR Advantages/Disadvantages  Advantages  No dilution of samples  No need of matrices such as KBr or mineral oil  No sample preparation  Eliminates sample preparation errors, analyst labor and sample destruction  Excellent for on-line, in-line applications, wireless technology available  Disadvantages  Low structural selectivity (no structure elucidation)  Not recommended for studying wetted samples unless interested in monitoring moisture (water bands are too strong)

25. Raman Advantages/Disadvantages  Advantages  Rich information content (fundamental vibrational modes)  Suitable for solids, liquids and gases  Requires minimal if any sample preparation  Excellent for aqueous systems (Water is not a strong Raman scatterer)  Remote sampling capability  Fiber optics  Non-invasive  Long working distance objectives  Non destructive  Disadvantages  Small amount of fluorescence can swamp signal  Leverage higher wavelength lasers to reduce chances (lower energy) but even 785nm lasers can have this effect on some samples

26. Tools used for PAT Framework Standard mechanism to describe the manufacturing risks of a medicinal product Drives an overarching control strategy Defines a development strategy to reduce risk to patients Risk Management Planning, conducting, analyzing and interpreting controlled tests to evaluate the factors that control the value of a parameter or group of parameters. Factorial Designs Optimizations Design of Experiments The science of extracting information from chemical systems by data-driven means from controlled or uncontrolled tests Multivariate analysis and data mining Production of predictive mathematical models Chemometrics

27. Tools used for PAT Framework Standard mechanism to describe the manufacturing risks of a medicinal product Drives an overarching control strategy Defines a development strategy to reduce risk to patients Risk Management Planning, conducting, analyzing and interpreting controlled tests to evaluate the factors that control the value of a parameter or group of parameters. Factorial Designs Optimizations Design of Experiments The science of extracting information from chemical systems by data-driven means from controlled or uncontrolled tests Multivariate analysis and data mining Production of predictive mathematical models Chemometrics

28. Risk Management Approach  Failure Mode and Effects Analysis (FMEA)  Developed by the US military back in the 1940’s  Some Basic Steps  Break down the product and process into its components or steps  Identification and assessment of the following for every item listed: function(s), potential failure mode(s), failure mode effect(s), failure mode cause(s), and controls for detecting or preventing the failure mode(s);  Evaluation of the risks associated with the failures modes and prioritizing them according to importance;  Implementation of corrective actions to minimize the occurrence of the more significant failure modes; Risk Management is a methodology to derive a process and product independent measure of performance related to the clinic

29. Dog and Pony Show FMEA Process in Laymen Terms 1. Tell me the manufacturing process to make the product 2. Tell me how great the process is 3. TELL ME HOW THE PROCESS MAY FAIL 4. TELL ME HOW I CAN PULL THE PROCESS BACK INTO CONTROL AFTER “FAILURE” OBSERVED How to Fix it?

30. Classically Risk in quantified in terms of:  Severity  What will happen if this exposure occurs?  Probability  What is the likelihood of the exposure?  Detectability  Is the exposure easily detected?

31. Documentation of Knowledge in FMEA  Defining the rationale behind the decisions as a function of current knowledge enables risk management to steer the appropriate direction  Transparent communication helps decision making

32. Classic Risk mitigation strategy  Quantify the risks  How input parameters (independent variables) are influencing output parameters (dependent variables)  Pareto the risks  Ranks risks high to low to prioritize  Establish counter measures and mitigation routes to:  Minimise  Severity  Probability  Maximise  Detectability

33. Evolution of Knowledge leading to Reduced Risk API (CMA's)RPN API Bulk Density84 API bulk Morphology210 API interfacial forces140 API Moisture Content72 API Particle Size192 API Polymorphic form200 API purity72 API rheological properties75 API Surface Area192 API solubility84 Excipients & Media (CMA's)RPN Binder Viscosity150 Coating material identification96 Packaging materials (CMA's)RPN Process parameters (CPP's)RPN Granulating solvent desired weight280 FIller 2 desired weight320 Filler desired weight320 "Inlet Air dew point-spraying"75 "Inlet Air Temperature profile-spraying"75 Blending time105 Disintigrant desired weight320 Lubricant desired weight320 Binder desired weight320 Coating Nozzles cleanliness72 "Air Flow Rate-profile-spraying"125 Campaign size84 Granulation Spray Rate profile160 Phase 1 Development  Phase 2 Development

34. Tools used for PAT Framework Standard mechanism to describe the manufacturing risks of a medicinal product Drives an overarching control strategy Defines a development strategy to reduce risk to patients Risk Management Planning, conducting, analyzing and interpreting controlled tests to evaluate the factors that control the value of a parameter or group of parameters. Factorial Designs Optimizations Design of Experiments The science of extracting information from chemical systems by data-driven means from controlled or uncontrolled tests Multivariate analysis and data mining Production of predictive mathematical models Chemometrics

35. Design of experiments (DOE)  Design of Experiments (DOEs) refers to a structured, planned method, which is used to find the relationship between different factors (let's say, x variables) that affect a project and the different outcomes of a project (let's say, y variables).  The method was coined by Sir Ronald A. Fisher in the 1920s and 1930.  DOEs are mainly used in the research and development department of an organization where majority of resources goes towards optimization problems.  In order to minimize optimization problems, it is important to keep costs low by conducting few experiments. Design of Experiments is useful in this case, as it only necessitates a small number of experiments, thereby helping to reduce costs.

36. Approaches  Allows for optimization of several variables simultaneously  Can observe effects of individual factors as well as combinations  Factorial Design (screening)  An experiment with at least two factors, each with a high and low  Experiment eventually takes on every possible combination of these factors  Response Surface Design (optimization)  An experiment that takes the significant factors and uses several intermediate volumes in its generated experiments

37. Process  Step 1: Define risk of overall process to define what parameters should be prioritized and studied  Risk assessment will have identified the input parameters that need to be studied  Parameters should be measured in a quantitative method to obtain best results  Step 2:  Incorporate replication through your design to measure variability in design space. This will determine what is noise versus signal.  Step 3: Randomize the Run Order:  In order to evade uncontrollable influences such as changes in raw material and equipment variation, it is necessary to run experiments in a randomized order.  Step 4:  Block out Known Sources of Variation: Through blocking one can screen out the effects of known variables such as shift changes or machine differences.  One can divide the experimental runs into homogenous blocks and then mathematically remove the differences. This increases the sensitivity of the design of experiment. However it is important to not block out anything one wants to study.  Step 5: Know Which Effects (if any) Will be Aliased:  An alias means that one has changed one or more things in the same way at the same time.  Step 6: Always Confirm Critical Findings: At the end of a design of experiment it is easy to assume that the results are accurate.  However it is important to confirm one's findings and to verify the results. This validation can be done using many other management tools available.

38. Reality: Minitab to the rescue!

39. Minitab to rescue Continued:

40.  Ask to “create design”  Run the design in order prescribed  Enter corresponding data  Analyze data  independent to dependent variables

41. DOE in Automated Sample Prep  Used to optimize automated derivatization/sample preparation method  Responses Number of GC peaks Area of each peak/area of the external standard peak  Factors: EtOH Pyridine ECF1 NaOH ECF2 Levels: Reagent volumes Cell Culture Media Characterization by GC-MS with Automated Sample Preparation EAS 2011: Ingersol et. al.

42. Design Process Factor (reagent) Level (amt  L) EtOH25-333 Pyridine25-85 ECF125-45 NaOH25-80 ECF225-45 Minitab Gilson Software Experimental Design sample GC-MS data Minitab

43. Data Analysis Cell Culture Media Characterization by GC-MS with Automated Sample Preparation EAS 2011: Ingersol et. al.

44. Data Analysis ECF2 Component 1 Component 2 Component 3 Component 4 Component 5 ECF1 EtOH Cell Culture Media Characterization by GC-MS with Automated Sample Preparation EAS 2011: Ingersol et. al.

45. Tools used for PAT Framework Standard mechanism to describe the manufacturing risks of a medicinal product Drives an overarching control strategy Defines a development strategy to reduce risk to patients Risk Management Planning, conducting, analyzing and interpreting controlled tests to evaluate the factors that control the value of a parameter or group of parameters. Factorial Designs Optimizations Design of Experiments The science of extracting information from chemical systems by data-driven means from controlled or uncontrolled tests Multivariate analysis and data mining Production of predictive mathematical models Chemometrics

46.  The science of extracting information from chemical systems by data-driven means.  Highly interfacial discipline, using methods frequently employed in core data-analytic disciplines such as:  Multivariate statistics  Applied mathematics  Computer science  in order to address problems in  Chemistry  Biochemistry  Medicine  Biology and  Chemical engineering

47. Multivariate Statistics  Principal component Analysis (PCA)  PCA projects high dimensional data to a lower dimension (X’s only are classified)  PCA projects the data in the least square sense– it captures big (principal) variability in the data and ignores small variability  Conceptually how it works:  Find a component (dimension vector) which explains as much x-variation as possible  Find a second component which:  is orthogonal to (uncorrelated with) the first  explains as much as possible of the remaining x-variation  Process continues until researcher satisfied or increase in explanation is judged minimal  Projection to Latent Structures (PLS) a.k.a. “Partial Least Squares”  PLS finds a set of orthogonal components that :  maximize the level of explanation of both X and Y  provide a predictive equation for Y in terms of the X’s  This is done by:  fitting a set of components to X (as in PCA)  similarly fitting a set of components to Y  reconciling the two sets of components so as to maximize explanation of X and Y

48. Correlation to Dissolution  PLS2 Model for Dissolution

49. How does all this come together? Mitigation of risk through PAT framework

50. Today’s Validated Process

51. Tomorrow’s Controlled and Monitored Process

52. PAT and Process Monitoring and Control

53. Basic Tablet Manufacturing Process

54. Example 1: How to use NIR to help monitor qualitatively material identity

55. Risk – Raw Materials  Correct Material?  Correct Specification?  Purity  Water content  Etc.  Manufacturability?  Chemical  Physical properties

56. Qualitative Analysis Application - Raw Material ID Pharmaceutical users want to apply NIR to inspect incoming raw materials quickly at the loading dock. They often would like to equip the receiving area with an NIR to be used by technicians who are not trained as scientists. The instrument will be used to confirm the identity of each container of material that is received. This would otherwise have to be done in the QC lab by wet chemical techniques Bunch of X’s = PCA Analysis

57. Library Samples  One lot each of the following substances was provided for the construction of the raw material library  D-Glucose  D-Fructose  Sucrose  D-Mannitol  D-Sorbitol  D-Lactose Monohydrate  Acetylsalicylic Acid  Acetaminophen  L-Ascorbic Acid  Citric Acid Why test all these if they are not all in your formulation?

58. Data collected and library built  Library is a collection of spectra with correlating material IDs and Lots in this case  Principal component analysis was done on the library to define a model for predicting each excipient  It worked (Trust me ) GlucoseFructose

59. Basic Tablet Manufacturing Process Example 2: Using FTIR to determine blend mixture time

60. Risk – Blended Materials  During formulation the API and excipients are dispensed then blended into a homogenous mixture.

61. Blend Uniformity in Tablets  Achieving proper blend uniformity for tablets prior to pressing is critical to achieve proper distribution of actives and excipients.  Under-blending results in inhomogeneous distribution of active ingredients in tablet  Over-blending leads to unrecoverable material  Example shows how blend uniformity can be successfully monitored in seconds using FT-NIR technology

62. Region of Interest for Component #2 Component #2 Component #1 How long would you wait?

63. Basic Tablet Manufacturing Process Example 3: Using FT NIR to determine content uniformity and formulation composition

64. Show Concentration Differences in Tablets  Can use both Transmission and Reflection measurements to analyze tablets. Can be done without removing the tablets from the system  Distinguish different clinical tablet formulations by amount of active ingredient  Finished product is a tablet and manufacturing protocols need independent verification of amount of active ingredient.

65. Reflectance Spectra for Clinical Tablets

66. Tablet Reflectance Results 0 10 20 30 0102030 Label Claim (Rel. Units) NIR Calculated (Rel. Units) Calibration Validation R = 0.99742 RMSEC = 0.606

67. Tablet Transmittance Results 0 10 20 30 0102030 Label Claim (Rel. Units) NIR Calculated (Rel. Units) Calibration Validation R = 0.99965 RMSEC = 0.0561

68. Tablet Transmission Data Classification Using Discriminant Analysis

69. Ideally now you know the defintion  PAT Framework  PAT is a system for designing, analyzing, and controlling manufacturing through timely measurements (i.e., during processing) of critical quality and performance attributes of raw and in-process materials and processes with the goal of ensuring final product quality  Process requires a set of tools greater than the physical analyzers  Risk Based Analysis (FMEA)  DOEs to help build models  Chemometrics to help monitor and build complex relationships

70. What/When/Why PAT?  What?  PAT is the incorporation of process measurements within the manufacturing process  Goals: Monitor, Control, Release Product, Optimize  When?  Should be part of the development process – helps optimize product and monitor as a function of scale up  Trend to include them as part of equipment trains  Deployment to the plant must always be kept at forefront of design. If it can not be deployed to the plant, must translate knowledge to tools that can be deployed  Why?  Ultimate goal is to gain enough process understanding to reduce overall risk of poor product quality