PAT Implementation-Pharmanexa-NR PDF 2024

18/07/2024 Assuring Product Quality and Improving Production Efficiency through PAT Implementation No...

18/07/2024 Assuring Product Quality and Improving Production Efficiency through PAT Implementation Noviani Rustanto Seminar Farmasi Nasional Pharmanexa 2024 20th July 2024 1 Process Analytical Technology (PAT) Design of Experiments “….a system for designing, analyzing, and controlling (DoE) manufacturing through timely measurements (i.e. during Modern processing) of critical quality and performance attributes of Control Rational Risk Assessment raw and in-process materials and processes, with the goal of Systems ensuring final product quality…” Quality cannot be tested into products; it should be built-in or QbD should be by design ” Process Multivariate Analytical Analysis Technology (MVA) (PAT) Modern Data Handling Systems Source: Guidance for Industry PAT – A Framework for Innovative Pharmaceutical Development, Manufacturing and Quality Assurance, US FDA, Sep 2004 2024 2 2 1  18/07/2024 PAT : A Paradigm Shift in Pharma Industry To design and develop well understood processes that will consistently ensure a pre- defined quality at the end of the manufacturing process. Such procedures would be PAT is Process Centric consistent with the basic tenet of quality by design and could reduce risks to quality (not product centric) and regulatory concerns while improving efficiency. Reference: Guidance for Industry PAT — A Framework for Innovative Pharmaceutical Development, Manufacturing, and Quality Assurance, US FDA, Sep 2004 Measuring Product Understanding The Controlling The Process Quality in Real Time Process Based On Product Quality 2024 3 3 Process Analytical Technology (PAT) QC Limits Traditional Approach The process is developed, frozen and validated using 3 commercial batches Process safeguarding with SOPs, training, etc Subsequent quality control via off-line investigation in the laboratory Hard to know what is occurring in the process Process is a black box Variability IN + Fixed Process = Variability OUT PAT parallel quality measurement QC Limits PAT Approach At-line, on-line and in-line analysis for identifying critical source of variability and establishing process understanding Important: to have an “eye” to look inside starting material and process Real time process control and quality assurance based on statistics Enables continuous process verification Multivariate analysis for understanding complex relationships Process is understood and well controlled Variability IN + Variable Process Condition = Consistency OUT  reduce burden for validating the system Reference: Guidance for Industry PAT — A Framework for Innovative Pharmaceutical Development, Manufacturing, and Quality Assurance, US FDA, Sep 2004 2024 4 4 2  18/07/2024 Key Advantages of PAT - Better process understanding : - faster product development/scale-up to reduce time to market - maintaining robust and reliable processes - product quality assurance at lower costs - preventing out-of-spec outputs (rejects, scrap and rework) - Generating cost savings : - reducing production cycle times thus increasing capacity and improving productivity - less energy consumption - Reduce reagent and consumable consumption - Improving safety : process automation - Enable continuous processing and real time release to shorten production cycle time and achieve process efficiency - Facilitating continuous improvement - Eco-friendly : reduced waste, less footprint, less energy consumption 2024 5 5 PAT Implementation in Product Life Cycle Early Development Late Development Stage Continuous Improvement Product Process Scale Up & Commercial design design Transfer Manufacture Process Continuous Process Validation Verification PAT IMPLEMENTATION PAT is often associated with in-process control testing in commercial manufacturing. However, in order to successfully implement PAT, sufficient knowledge of the process/product has to be gained in the early stages of development. The advantage of using PAT principles and tools during development is to create opportunities to improve the mechanistic basis for establishing regulatory specifications 2024 6 6 3  18/07/2024 PAT Application in Solid Dosage Manufacturing Line Purity, identity (molecular structure) and Generate blends with the lowest conformity to predict how to process ‘Residual Heterogeneity’ material in QbD Environment NIR ensures that powders are not (e.g. crystallinity/polymorphous, moisture under or over blended. level, particle size) Over blending generates ‘fines’ that can result in poor tablet compression. RMIQ Ensure that powder flow is consistent for reliable tablet compression. In-line PSA Monitor feed-frame to ensure.  Powder blend is not segregating during compression.  Potency is consistent over entire Engineer particles so that they blend uniformly. process. Monitor and control process using NIR and in-line Particle Size Analyzer (PSA) to  Generate high quality granules  Dry granules to desired state.  Ensure over-drying does not occur. 2024 7 7 PAT Application in Liquid Dosage Manufacturing Line Purity, identity (molecular structure) and conformity to predict how to process material in QbD Environment (e.g. crystallinity/polymorphous, moisture Monitor and control filling process using NIR level, particle size) to ensure that content uniformity/assay is consistent over the entire process Monitor purging performance and residual RMIQ gas content in the bottle’s headspace In-line In-line Viscometer Headspace Gas Analyzer Monitor and control process using NIR to Generate dispersion blend with the Monitor and control process using NIR to lowest “residual heterogeneity” Ensure that assay after filtration meets Monitor and control liquid viscosity in real specification before filling time 2024 8 8 4  18/07/2024 Is PAT Tool Required For Each Unit Operation? Critical Process Parameter (CPP) : A process parameter whose variability has an impact on a critical quality attribute. Should be monitored or controlled to ensure the process produces the desired quality. PAT is used to monitor and control the CPPs Considerations for identifying and documenting CPPs Risk assessment and experimentation to establish the linkage between potential PAT provides information: CPPs and CQAs. This could be based on prior knowledge, scientific first - What is occurring in the process principles, quality risk management (QRM), design of experiment (DoE), and - What process parameters impact product quality other appropriate experimentation - Impact level of each parameter - CPPs interdependencies Interdependencies of the different CPPs Selected control strategy and the residual risk. Select PAT tool fits with your control strategy ! Reference: Guidance for Industry, Q8(R2) Pharmaceutical Development, US FDA, Nov 2009 2024 9 9 PAT : Major Techniques NIR spectroscopy Raman Spectroscopy  USP , - USP ,  EP 2.2.40 - EP 2.2.48  Guideline on the use of near infrared spectroscopy by the pharmaceutical industry and the data requirements for new submissions and variation [EMEA, 27 Jan 2014]  Development and Submission of Near Infrared Analytical Procedures, Guidance for Industry [US FDA, Aug 2021]  Guideline on Validation of Analytical Procedures [ICH Q2(R2) Annex 2, 14 Dec 2023] 2024 10 10 5  18/07/2024 Near InfraRed Spectroscopy (NIRS) 2024 11 11 Near Infrared Spectroscopy (NIRS) Short Wave Region: Long Wave Region: (700 – 1100 nm) (1100-2500 nm) Typically used for transmission Typically used for measurements diffuse reflectance measurements 2024 12 12 6  18/07/2024 Overtones and Combinations in NIR Spectra Each overtone describes the information of the previous one, but one order of magnitude less in intensity Spectrum of biscuit dough with the main areas of absorption identified 2024 13 13 Near Infrared Spectroscopy (NIRS) Components: Light source Slit Mirror Diffraction grating (Prism) Order sorter Detector Setup & Calibration: Slit opening Mirror distance Grating position Grating Components: o Grating Light source Beam splitter o Fourier Transform 2 Mirror: moving & fixed Mirror o Filter Laser Detector Setup & Calibration: Beam splitter setting Mirror distance Fourier Transform 2024 14 14 7  18/07/2024 Near Infrared Spectroscopy (NIRS) Linear Variable Filter (LVF) is one dimensional continuously varying bandpass filter This configuration offers: Compact, lightweight, robust, low power No moving parts or free-space optics No misalignment, long-term stability 100% transferability between units No fiber optics: Nothing to damage or wear out Insensitive to temperature variation No calibration is required Lamps >40,000 hours lifetime Maintenance free Components: Light source Linear Variable Filter Detector VIAVI supplied the Linear Variable Filter for the OSIRIS-REx Bennu mission. The OSIRIS-Rex spacecraft reached asteroid Bennu in Oct 2020 and collected samples from the near-Earth asteroid’s surface. OSIRIS-Rex’s asteroid sample return capsule landed in Utah in Sep 2023 2024 15 15 Near Infrared Spectroscopy (NIRS) Applications  For identification, qualification and assay of starting materials, intermediates and finished products and verification of physicochemical properties  For gaining process understanding  For monitoring and controlling processes : off-line, At-line, On-line, In-line  For continuous improvement and life cycle management NIRS constitutes one of the major techniques in Process Analytical Technology (PAT) and be used as part of Real Time Release Testing (RTRT) strategy. NIRS enables enhanced assessment of the quality of materials and processes by extensive and more representative sampling Source: Guideline on the use of near infrared spectroscopy by the pharmaceutical industry and the data requirements for new submissions and variations, EMEA, 27 Jan 2014 Development and Submission of Near Infrared Analytical Procedures, Guidance for Industry, US FDA 2021 2024 16 16 8  18/07/2024 Are All NIR Spectrometers Suitable For PAT Analysis? Ensure that NIR Spectrometers capture critical material and process information throughout the process cycle within the very short time frame so that it provides real-time process control, process understanding and prediction that are statistically reliable. NIR Spectrometer Integration Time #Spectra per minute Grating system ~ 30 - 60 seconds 1-2 FT NIR system ~ 30 - 60 seconds 1- 2 Linear Variable Filter System 10 ms typical, minimum 10 6000 µsec Other considerations : Stability (misalignment potential due to the vibration effect, temperature effect, etc.), Reliability (spectra quality, spectra acquisition, wavelength accuracy, etc.), Robustness (spectra reproducibility, model transferability), Ruggedness (IP certified), Mobility (dimension & weight to fit the application), Ease of installation and operation 2024 17 17 LVF NIR PAT Solutions Real time process Rugged, IP65 housing monitoring Internal, rechargeable battery Field, at-line, in-line with continuousrun time >10 analysis hrs USB powered Bluetooth or USB connection IP65 and IP 67 rated IP65 and IP 67 rated Weight: 318 g Weight 250 g Real time process monitoring Rechargeable Li-ion battery Continuous liquid process (>8 hrs of continuous run monitoring in vessels, tanks, time) and pipelines WiFi or ethernet connection USB powered via hardwired Multi-axis sensor for cable positional triggering in tumble IP65 and IP 67 rated blender IP65 and IP 67 rated Weight: < 1.4 kg 2024 18 18 9  18/07/2024 MicroNIR Cloud-Based and On-Device Analysis Login Connect Select Method Analyze 2024 19 19 Where Does PAT Start? Dispensing Granulation Drying Blending Tabletting Coating 2024 20 20 10  18/07/2024 PAT For Raw Material Quality Monitoring 21 Raw Material Quality Test : Current Facts Other than chemical attributes (identity and purity), physical and mechanical attributes are critical for production process (=process-ability). Minor differences in physical attributes of raw material may affect process robustness. Processing difficulties can arise that result in the failure of a product to meet specifications, even if certain raw materials conform to established pharmacopeial specification, which generally address only chemical identity and purity. Inherent, undetected variability of raw material may be manifested in the final product. Sample preparation of offline wet lab analysis often causes loss of valuable information pertaining to the formulation matrix. Certain attributes (particle size and shape variations within a sample) of raw and in-process materials may pose a significant challenge in collecting representative samples because of their complexities and difficulties. It is well known that powder sampling procedures can be erroneous. Establishing effective processes for managing physical attributes or raw and in-process materials requires a fundamental understanding of attributes that are critical to product quality. Appropriate use of PAT principles and tools can provide relevant information relating to physical, chemical, and biological attributes. Several new technologies are now available that can acquire information on multiple attributes with minimal or no sample preparation. Source: Guidance for Industry PAT – A Framework for Innovative Pharmaceutical Development, Manufacturing and Quality Assurance, US FDA, Sep 2004 2024 22 22 11  18/07/2024 NIR and Raman Spectroscopy in Raw Material Application There are similarities between NIR and Raman Spectroscopy – Non destructive – Handheld devices – No sample preparation – User friendly for non-skilled personnel – Work through plastic – Required multiple standards for defining statistical threshold However there are many differences between both techniques….. 2024 23 23 NIR and Raman Spectroscopy in Raw Material Application Dipole moment variation  active MEASUREMENT PRINCIPLE Polarizability variation  active on on all kind of pharma ingredients limited pharma ingredients Max. 2 seconds DURATION PER ANALYSIS Vary from 5 seconds to minutes Fingerprint SIGNAL OUTPUT Sharp peaks Raman Spectroscopy NIR Spectroscopy No issue FLUORESCENCE ISSUE Fluorescence interference may occur No issue ANALYSIS OF COLOR SAMPLE May damage/burn ABLE TO DETECT: Yes moisture level Not sensitive Yes particle size difference Not sensitive Affect processability ! Yes Crystallinity Not sensitive Need sampling LIQUID ANALYSIS Without sampling Eye safe SAFETY FOR OPERATOR Eye injury risk  safety google is mandatory Yes PAT SUITABILITY No 2024 24 12  18/07/2024 Handheld NIR and Raman Spectroscopy : Performance Comparison API Excipients 99.5 99.5 Raman active 100 chemicals 90 80 Fluorescence interference 70 Weak signal and longer time 60 53 50 47 Total 40 34 30 25 22 19 20 10 0 NIR Raman Good Raman Bad Raman Uncertain (1) Test done over 200 raw materials (50% API and 50% excipients) (2) Raman using laser 785 nm 2024 25 25 Raw Material Identification is Not PAT This is Raw Material Qualification PAT Physical RM Identification Characteristics This is To confirm chemical not To differentiate between structure of a molecule grades of the same PAT substance [in terms of moisture level, particle size, or polymorphs) Raman, NIR NIR The same chemical structure but different physical characteristics may have different processability and/or efficacy that leads to production and/or bioavailability issue 2024 26 26 13  18/07/2024 NIRS in Incoming Raw Material Identification and Qualification  Based on the comparison of the spectral data of the substance with the spectral reference library  Qualification is often performed after the sample identification, with the qualification spectral library derived from samples chosen to represent the defined variability of the different grades of the same substance  If identification and/or qualification are based on more than one analytical method, it should be clear which reference method(s) will be replaced by the proposed NIR procedures  When NIRS on its own is not sufficient to identify or qualify a substance, it should be supplemented by other different analytical procedure(s) so that the tests taken together ensure specificity.  As we have more and more spectra collected for each raw material, define design space for each material and establish the control strategy and perform continuous improvement Source.: Guideline on the use of near infrared spectroscopy by the pharmaceutical industry and the data requirements for new submissions and variations. EMEA, 27 Jan 2014 2024 27 27 NIR Library Development Recommendation  Spectral reference library is established by several samples of several batches of difference substances  The library should be constructed from materials deemed acceptable according to registered or compendia testing regimes. All materials must have Quality Control (QC) documentation associated with the raw material lots used. However, do not use reference standard spectra for the library as the spectra of reference standard tells you nothing about the processability dan the batch-to-batch variations of your raw materials  Samples from all relevant suppliers used should be incorporated into the library  Samples should exhibit the variation that is typical for the substance to be analyzed (e.g., variation in solid-state form, particle size, crystallinity, synthesis route, storage time) For example, MCC 101 vs MCC 102, Paracetamol regular vs Paracetamol micronized : different particle sizes Paracetamol form I and form II : require different granulation and compression process to meet QTPP Atorvastatin Form I vs amorphous : different release profile and stability Source.: Guideline on the use of near infrared spectroscopy by the pharmaceutical industry and the data requirements for new submissions and variations. EMEA, 27 Jan 2014 USP Near Infrared Spectroscopy 2024 28 28 14  18/07/2024 Spectral Match Value (SMV) Spectral Match Values (SMV) is a simple, yet highly flexible and powerful RM Qualification method Reference spectrum (Mean library spectrum) Based on correlation to a mean spectrum, SMV Test spectrum performs a pairwise wavelength correlation across a defined spectral region The mean correlation value is the SMV and this is compared to all libraries in the set for unique classification a ,b a b When Ɵ is close to zero, cost (Ɵ ) is close to 1 and the correlation is high 2024 29 29 Raw Material Qualification: Source Difference  Evaluation on two different sources of riboflavin sodium phosphate  Both materials met compendial specification  Source B showed different NIR spectra compared to source A although both have the same molecular structure.  NIR with SMV method is able to differentiate source A from source B 2024 30 30 15  18/07/2024 Raw Material Qualification: Particle Size Difference Spectral Match Value (SMV)  SMV model was developed using MicroNIR to differentiate Cefixime powder vs Cefixime micronized  Spectra shift is observed due to different particle size. Cefixime Powder vs Micronized Cefixime powder with bigger particle size gives higher absorption value.  SMV model with 0.9800 threshold is able to differentiate Cefixime micronized to the Cefixime powder in real time 2024 31 31 Raw Material Qualification: Batch Consistency Evaluation on different batches of Citicoline, a highly hygroscopic powder. Citicoline BN 1B1600 is an approved batch that is successfully used in our client’s production line. Result for the other 2 batches of incoming material  BN 1B2953 showed similar spectra to the approved reference batch. This was used successfully in the production.  BN 1B1300 showed an absorbance shift in all wavelength and higher absorbance within 1450 – 1500 nm for H2O vibration. This indicated the batch had bigger particles thus lower solubility as well as higher moisture content. All batches met Compendial Specifications and were The batch produced with this material failed to meet the released for routine production dissolution specification. 2024 32 32 16  18/07/2024 PAT For Process Monitoring 33 Drying Monitoring 34 17  18/07/2024 Fluid Bed Drying Monitoring and Control In Line process monitoring using MicroNIR PAT-U Quantitative calibration model: PLS for LOD determination 2024 35 35 Fluid Bed Drying Monitoring and Control Model Validation RMSECV: 0.52% RMSEP: 1.10% Verification Batch showed high predictability 2024 36 36 18  18/07/2024 Blending Monitoring 37 Thief Sampling is Considered Non Representative Physically sampling by “Grab Sampling” is to be avoided. The sample thief disturbs the local composition of the powder bed (see the image on right hand side) and results in a non-representative sample. Multiple sampling of the same bed results in the entire blend “less homogenous”  U.S. Department of Health and Human Services, F.D.A., Guidance for Industry: Powder Blends and Finished Dosage Units-Stratified In-Process Dosage Unit Sampling and Assessment, October 2002 (withdrawn on August 7,2013). Image from Muzzio et al. Chapter 15, Solids Mixing Handbook of  U.S. Department of Health and Human Services, F.D.A., Industrial Mixing: Science and Practice, Edited by Edward L. 2002. ANDA’s: Blend Uniformity Analysis; Withdrawal of Paul, Victor A. Atiemo-Obeng, and Suzanne M. Kresta Copyright 2004 John Wiley & Sons, Inc. Draft Guidance. 2024 38 38 19  18/07/2024 We Need to Rethink Our Sampling Approaches Homogeneity is the limiting case of ‘Zero Heterogeneity’. We should only talk about ‘Residual Heterogeneity’ Sampling devices typically contribute 10-50 times the errors contributed by analytical techniques. Heterogeneity cannot be corrected for ‘statistically’ as the sampling error will change each time a new ‘grab’ sample is extracted. The latest US FDA process validation guidance of 2011 states that all sources of variability should be understood. In this case, why is thief sampling still used, even though the draft guidance promoting its use has been withdrawn? PAT instruments also lead to ‘grab’ samples, however, based on time averaging of ‘composite’ samples, it is a better approach than thief sampling. 2024 39 39 Golden Rules of Powder Sampling 3D Spatial Lot Heterogeneity. It is impossible to 1. A powder should always be sampled when in representatively sample using physical devices motion 2. The whole of the stream of powder should be taken for many short increments of time in preference to part of the stream being taken for the whole of the time Allen, T., 2003.1 – Powder Sampling. In: Allen, T. (Ed.), Powder Sampling and Particle Size Determination. Elsevier, Amsterdam, pp. 1-55 1D Dynamic Lot Heterogeneity. This is the domain of PAT 2024 40 40 20  18/07/2024 NIRS For Blend End Point Monitoring Delay Time Effective Scan Time 2024 41 41 Blend End Point Analysis 3 4 2024 42 42 21  18/07/2024 Critical Attributes Affect Blend Uniformity Rotation Time API concentration Excipients Particle Size and Shape Blender Capacity Rotation Speed Fill level Loading Sequence 2024 43 43 Effect of Critical Attributes on Blending Profile 0.1 Blender capacity 0.09 Loading sequence 0.08 0.07 0.06 0.05 0.04 Active Layered - Centered 0.03 Active Spot - Off Center 0.02 0.01 0 0 5 10 15 20 25 30 35 40 Rotation Number Rotation speed Fill level 2024 44 44 22  18/07/2024 MicroNIR Golden Batch Functionality Calibration on 3 batches Validation 2024 45 45 Summary PAT is about process understanding and providing predictive control. PAT is therefore Proactive. PAT allows the investigation into a process to find the causes of variability so that Quality by Design can be enabled. PAT requires a mindset shift and a new set of analytical tools to fully investigate the multivariate data generated. There are a number of PAT techniques in use today, however, Near Infrared (NIR) spectroscopy is by far the main technology used. To ensure acceptable and reproducible product and manufacturing process, we should consider both chemical attributes of incoming material and their process-ability. PAT offers many advantages over the traditional approach. PAT reduces regulatory burden, enhance quality assurance, improve production efficiency and increase product competitiveness in the market If you are not thinking PAT moving forward, you may be left behind very soon 2024 46 46 23  18/07/2024 Our PAT Solutions SPECTROSCOPY VISCOSITY TALK TO US NEAR INFRARED VISCOMETER SPECTROMETER In-line In-line, On-line, At-line At-line RAMAN SPECTROMETER In-Line MODEL DEVELOPMENT SERVICES Pharma PAT Solutions PARTICLE SIZE ANALYZER LASER HEADSPACE GAS At-line ANALYZER In-line At-line In-line I Kadek Sumerta Ilham A. Ibrahim PARTICLE SIZE HEADSPACE GAS @ Exhibition Area 2024 47 47 PT. VIRTUS ANALITIKA MITRATAMA www.virtus-analitika.com E: [email protected] T: +62 21 3886 9000 C: +62 856 9363 5138 The material in this presentation is the proprietary property of PT. Virtus Analitika Mitratama and its partner and may not be copied or reproduced in any form without the express permission of the speaker 48 48 24