Presentation on theme: "Director of Application Support Services"— Presentation transcript:
1 Director of Application Support Services Raman Spectroscopy for Rapid Material Identification in the Pharmaceutical Industry April 12, 2013Katherine Bakeev, PhDDirector of Application Support ServicesB&W TekNewark, DEUSA
2 What is the need?Pharmaceutical companies must meet worldwide PIC/S initiatives, local Pharmacopeia and the US Food and Drug Administration’s regulations on Good Manufacturing Practice (GMP)There has been a trend of increasing product contamination incidents, product recalls and facility shutdownsCompanies must develop and implement improved analytics to check the identity, integrity and quality of ingredients and productsImplementation of 100% testing of raw materials is coming!to the US (and has been adopted in Europe)PIC/S Pharmaceutical Inspection Cooperation Scheme
3 2012: Japan and Korea applied for membership TaiwanPIC/S is the abbreviation and logo used to describe both the Pharmaceutical Inspection Convention (PIC) and the Pharmaceutical Inspection Co-operation Scheme (PIC Scheme) operating together in parallel.The PIC Scheme commenced operating on 2 November 1995 in conjunction with PIC which had already been operating since 1970 (see "Background on PIC"). Thus PIC/S became operational in November 1995.The need to form the PIC Scheme became necessary when it was realised that an incompatibility between PIC and European law did not permit individual EU countries that were members of PIC to sign agreements with other countries seeking to join PIC. Only the European Commission was permitted to sign agreements with countries outside Europe, and the Commission itself was not a member of PIC.Therefore, a less formal and more flexible cooperation scheme was developed to continue and enhance the work of PIC. Instead of being a legal treaty between countries (ie. like PIC), the PIC Scheme is a cooperative arrangement between Health authorities.PIC and the PIC Scheme, operating together as PIC/S, provide an active and constructive co-operation in the field of GMP (Good Manufacturing Practice). The purpose of PIC/S is to facilitate the networking between participating authorities and the maintenance of mutual confidence, the exchange of information and experience in the field of GMP and related areas, and the mutual training of GMP inspectors.New Zealand2012: Japan and Korea applied for membership
4 Industry Trends and Requirements Incremental Regulatory Control for Incoming Material IdentificationPIC/S recommendations on complete traceability of incoming materialsGMP is expanding to new productsGlobal Supply ChainCompanies are moving toward a more delocalized supply chainProblems associated with supplier quality assurance and transportationQuality Assurance and Cost ReductionIncrease analytical capabilitiesReduce operational costOptimize operational efficiencies
5 100% Raw Material Inspection Why do we need to perform 100% container Identity Testing?
6 GMP for Raw Material Identification EU GMP and PIC/S GMP GuideChapter 5. Production5.30 There should be appropriate procedures or measures to assure the identity of the contents of each container of starting material.Annex 8 -Sampling of starting and packaging materials2. Starting MaterialsThe identity of a complete batch of starting materials can normally only be ensured if individual samples are taken from all the containers and an identity test performed on each sample.This means: Europe requires 100% container testing
7 GMP for Raw Material Identification US FDA 21 CFR requires testing of raw materials:21 CFR (d)- “At least one test shall be conducted to verify the identity of each component of a drug product. Specific identity tests, if they exist, shall be used.”Raw materials are quarantined until identity verifiedRaw materials must meet predetermined specificationsUS FDA Compliance Program Guidance Manual Program (Drug Manufacturing Inspections-FDA CP )Material systems: at least one specific identity test is conducted on each lot of each componentDietary Supplement GMPs US 21 CFR 11121 CFR (a)(1)(i)Conduct at least one appropriate test or examination to verify the identity of any component that is a dietary ingredient
8 US FDA Warning Letter Review- Identity Tests Xian Libang Pharmaceutical Co., Ltd., China. (Jan 28, 2010)1. Failure of your quality unit to ensure that materials are appropriately tested and the results are reported.Your firm used the IR spectra for one lot to approve and release two subsequent incoming lots. This practice is unacceptable and raises serious concerns regarding the integrity and reliability of the laboratory analyses conducted by your firm.It is essential that at least one test be conducted to verify the identity of each lot of incoming material.In addition, the laboratory control records should include complete documentation of all raw data generated during each test, including graphs, charts and spectra from laboratory instrumentation.Requirement-development of methods/libraries with well-characterized materials
9 Regulatory Expectations for Spectroscopic-based Methods The Pharmaceutical Inspection Convention and Pharmaceutical Inspection Co-operation Scheme (jointly referred to as PIC/S) are two international instruments between countries and pharmaceutical inspection authorities, which provide together an active and constructive co-operation in the field of GMP.
10 Issues and Customer Concerns Can Raman help my project?What are the differences between Raman and NIR?Should I really perform 100% container ID test? What are my options?Do I need to validate the Raman method? If yes, what are the requirements for method validation?What is next after the Raman ID becomes a routine method?Do I need to know the filing expectations from regulatory agencies (JP/EMA/FDA..)?Can vendors provide GMP support & related documents?
11 Advantages of Raman Spectroscopy These materials cannot be identified by Raman, FTIR or NIR
12 Raman can measure through packaging: broad class BottlesManufacturerThicknessAmber GlassVWR3 mmClear Glass (A)I-Chem2 mmClear Glass (B)High Density Polyethylene (HDPE)1 mmTeflon FEPNalgenePolystyreneUlineVialsKimballClear GlassBagsPolypropylene (PP)2 milLow-Density Polyethylene (LDPE)
13 Raman Applications in Pharma IDIdentification of raw materials (building up an extensive library of Raman spectra); This is the application for NanoRamQuantitative analysisQuantitative determination of active substances in different formulations;PolymorphismSupporting polymorphic screenings (polymorphs have different solubility rates, thereby impacting the effective dosing);Process (PAT)Supporting chemical development process scale-up (as process steps are modified and refined to ascertain whether the desired chemical is being produced or not, and the rate at which it is formed).
14 Raman vs non Raman Feature FTIR Raman NIR Selectivity Interference Very high; fundamental vibrationsVery high; fundamental vibrations Low; broad, overlapping bands requiring chemometric modelsInterferenceStrong absorption of glass, plastics, waterNo issues with aqueous solutions; not sensitive to physical form of samples Influence from water; signal impacted by particle size, hardness,SamplingSample prep may be needed; direct compressed sample contact for ATRNo sample prep; measure through transparent or translucent containers; effective in the presence of solutions Can measure through glass, quartz; large sample area neededMethod DevelopmentCan be by visual examination or library matchingSharp, distinctive spectral peaks can be used for accurate identification Analysis typically requires chemometrics; low selectivityDevice PackagingCan be bulky and weigh > 2 kg; low battery lifeHandheld. Light weight (1 kg); > 4 hour battery life Bulky. Over 3 kg; low battery life4/14/2017
15 Benefits of Handheld Raman for ID Testing Reduced material movement – directly implement in warehouse, loading dock, fewer accidents/reduced manpowerEasy and simple operations- simple training & improves manpower usageReduce chemical exposure – scan through packaging materials & maximize personnel safetyFewer lab delays reduce cycle time – on-time production & higher throughputFree up laboratory instruments – increases lab capacityReduced transcription – fewer errors; lower personnel and equipment overhead; control costs
16 Method Development- Special Considerations for Raman Technique Fluorescence interference/presence of impuritiesBaseline shifts and background signalLaser powerConfirm that the sample is not being altered. Adjustable laser power from mW.Sample-position sensitivityUse of appropriate sampling accessory.
17 The Measurement Challenges- Measure Through Plastic Packaging Implementation advantages of spectroscopic methodsNo cross-contamination (the material inside the bag)Safe and easy to implementCan be implemented in warehouse, docking area or labBut…challenges exist if using NIR techniqueSpectral profiles include packaging absorptionsPackaging interferences may reduce the specificity of the method when identify similar materialsThe library requires assessment or change if the bag changes, i.e., not suitable for multiple suppliers of same materialsPackaging could have multiple layers, with different colors or materials
18 NIR Spectra of API (Plastic bag contribution) .4.6.8188.8.131.52.8850080007500700065006000550050004500Wavenumber (cm-1)NIR Spectra of material –with plastic bagTwo layer PE bagOne layer PE bagTwo Layers PE bag has higher NIR intensity as compared to One-Layer PE bag.No PE bag (pure sample)Required to evaluate the method specificity in the presence of bag!
19 Raman Spectra of API (Plastic bag effect) 20040060080010001200140016001800200022002400900270036004500540063007200810090009900Raman shift [cm-1]IntensityNo significant spectral absorption from PE bagTwo layer bagOne layer bagNo bag
20 Raman has greater specificity than NIR Same finished tablet measured by NIR and Raman- note sharp, distinct peaks in Raman4/14/2017
21 Regulatory Standards Compliance 21 CFR Part 11 Electronic Records; Electronic Signatures21 CFR Part Laser and Laser SystemsUS Pharmacopeia <1120> Raman SpectroscopyEuropean Pharmacopeia Ch guidelines for Raman SpectroscopyASTM (2007) Standard Guide for Raman Shift Standards for Wavelength Calibration
22 Validation Qualification Process of providing documented evidence that something does what it is intended to doProcess, system, methodQualificationInspection, testing and documentation reviewIs a part of the validation process which verifies module and system performance prior to being placed in routine useEquipment/instrument
23 Analytical Method Development Prerequisites for analytical method validation-Six “M”sManMachineMethodsqualifiedcalibratedcharacterizedrobustdocumentedskilledqualifiedsuitableQuality of the analytical methodQualified MaterialVibrationsTimeIrradi- ationAnalysts´ supportTempe- ratureQualityHumiditySuppliesMaterialMilieuManagement
27 Calibration Validation: part of the PQ Self check: by performing calibration validation using ASTM 1840 ref material: PolystyreneCalibration Validation: part of the PQStandard spectra of the following materialsReference spectra on :1. naphthalene (Mallinckodt #94848) 2. 1,4 bis (2-methylstyryl) benzene (BMB) (Aldrich #25,740-0) 3. sulfur (Aldrich #41,498-0) /50 (v/v) toluene/acetonitrile (Mallinckrodt Analytical Reagents) acetamidophenol (active ingredients of Tylenol) (Aldrich A730-2) 6. benzonitrile (Baker B883-07) 7. cyclohexane (Mallinckrodt Analytical Reagents) 8. polystyrene (Aldrich #18,243-5)
28 Near Infrared already established for ID Provides guidance on qualitative and quantitative analysis.Samples used in ID methods should be authenticated by appropriate means (certificate of analysis or relevant testing).
29 Spectroscopic-based Identity Method The methods/libraries must be developed and can then be run on routine basis2. Threshold p-value 0.05 based on PCA of 20 spectraMethod developmentMeasure 3-10 Batches;NanoRam uses 20 spectra3. Build & Test MethodTrained analysts
31 ID Method Development- Flowchart Define intended purpose & scopeChoose sampling accessory that will be usedAuthentic (or released) samplesFor library one spectrum of each sample is scannedCollect samples for methodsPowder in vial, in real package…Any anomalies or outliers ?The sources of variations (if any)Scan barcode and samples; visual evaluation of scanTablet: Two faces scans/per tabletSample spectrum added to method being created; 20 spectra neededAdd sample to methodCapsule: better to remove capsule shell (if possible)Method created after data collectedFor library method development see NanoRam users manual section 6.4The method on the NanoRam is automatically created after the 20 scans are collected. A user does not need to follow the steps of rigorous chemometric model building (as required with NIR). The data are normalized, and a second derivative applied. A PCA model of the 20 scans is developed, and model rank based on 90% explained variance of the data. In the Identification Mode samples are then projected onto the Method model, and if they fall within the 95% confidence of the PCA model (0.05 p value), they PASS the method ID.For method the p value is used as pass/fail criteria. Default setting p>0.05 is defined under SettingsThreshold can be modified during method development to decrease overlap of materials.ID Method validation
32 Raman Method Development Raman bands are distinct and can be easily related to chemical structure-very good for fingerprinting.Conduct a feasibility test:To ensure sufficient sensitivity for compounds to be included in the reference library (very weak scatterers)Minimize fluorescence interference/impuritiesLaser powers and exposure timesconfirm that the sample is not being alteredSamplingAppropriate sample accessoryCollect valid representative samplesUnderstand the sources of variationsGlass vial and packaging materials’ variabilityMaterial lot-to-lot variabilitySupplier variationsNeed to evaluate the performance on testing numerous lots of material
33 Raman Method Development Raman bands are distinct and can be easily related to chemical structure-very good for fingerprinting.Raman spectral data collectionMeasure through containers and plastic bags with appropriate accessory; contribution from the packaging components are usually very minorSystem automatically determines scan time to give sufficient Raman signal for each sampleCollect 20 scans for each method – best to use different lotsIdentification MethodThe method is based on the 20 scans of the materialA multivariate model of the spectra (PCA) with Hotelling’sT2 limit determined (95% confidence; p = 0.05)p-value used as Pass/Fail criterion
34 Tools for different verification modes of materials. Identification vs. defined method with p-value (significance level)Investigation for unknown samples: HQIIn general, we recommend the use of p-value for ID (identification mode).The HQI (Investigation mode) may be considered an additional tool for validation.
35 Results from NanoRam Identification: Pass/Fail sample identity verified by comparison to defined method; p-value used as pass criterionInvestigation: Match/No Matchsample matched by correlation (HQI) to materials that are in the libraries available on the system
36 When running IDENTIFICATION and sample fails, there may show some potential hits. If an identification fails, the system will automatically run an investigation to suggest potential matches for reference purpose.
37 Analytical Method Validation Do we need to perform method validation of spectroscopic-based techniques ?
40 Spectroscopic Method Validation Method BuildingIdentification Method and operation preset settingsNotes should include ref to samplesSamplesAt least 3 lots (best to use more lots)Authentic (or released) samplesIf only 1 set is available is ok, method can be revisited at a later stage.(+) Positive Controls (one per product) – not being used in development(-) Negative Controls-similar materialsMethod Validationspecificity InternalExternal (+, -)Robustness*One Positive Control per product, uses materials used for method developmentOne to three Positive Controls One Negative Control* Robustness is not required from USP <1225> & ICH Q2(R1). In practice it is better to assess the variations of sampling techniques during early method development
41 Proximity Matrix www.bwtek.com Proximity Matrix: Testing of all the materials in Identification Mode using specific methods generated by each of the reference materialsMethodMaterialsMethod AMethod BMethod CMethod DMethod EMaterial APASSFailMaterial BMaterial CMaterial DMaterial EProvides an overview to illustrate which chemicals may produce:False PositiveFalse Negativegood signalbad signal…Green …no problemYellow… carefull, potential issuered… detail studied is requiredp-value> 0.05< 0.05< 10-3< 10-6zero
42 Method Validation for Qualitative analysis (ID) It is important to test methods to ensure correct PASS of known good materials and correct FAIL of materials known to not be what the method was developed forOur OQ and PQ procedures include tests for this – ensuring that there are not false positives or false negatives from the methodsValidation requires risk assessment and should be done at a level commensurate with the risk that an incorrect result carriesFocus on minimizing possible harm to patient.
43 Method ValidationDepending on the criticality of the Raman method, steps for checking the method validity should be done on a periodic basisA method performance test can include running a check sample by the Raman method and testing the results versus the compendial method used as reference for the method developmentPQ is done to test instrument performance at regular intervals
44 Spectroscopic Method:Iterative Nature Intended useBased on scienceConfiguration/SamplingSpectral-pretreatmentSample selectionCalibration/validation setsMainte-nanceCollectionMethodValidationDevelop-mentData storageContinuous performance verificationMethod updateRe-validationRaman SpectroscopyMethod developed on NanoRam with defined settingsChemometric/statistical methods/ algorithmsProtocol & reportValidation elements
45 Raman Method- Beyond Routine Analysis: Change Control & OOS Out of Specifications- ID FailurePeriodic Performance ReviewMethod and Library Update
47 Raman Routine Analysis- More Details Similar to QC methods, but…ID results - Out of SpecificationFollow internal SOPs, investigate root causesInstrument or operating errors?Variations of material sources or grades?Change ControlRisk assessment of incidents (e.g. laser, accessory, change of packaging material)Periodic Performance Verification (PQ)Should be run at regular intervals (or N batches)Event trigger: changes of hardware/software, change of material supplier or results trends
48 Standard Operating Procedures (SOPs) Guidance for Raman ID Method Development: B&W Tek doc availableSOPs for Raman General Operation: NanoRam User ManualProcedures that describe how to use the system: NanoRam User ManualSOPs for Raman Calibration and Maintenance, including relocation performance check : annual system recertificationSOPs for Raman System Disaster RecoveryUser’s internal SOPs for Change control plan-Hardware/software upgradeChange of sampling module (e.g., point and shoot to vial holder).SOPs for Raman Method Performance VerificationPeriodic performance verification PQSOPs for Method and Library updateConditions and decision to launch a method updateRequirements for electronic data archival
49 Regulatory Expectations for Spectroscopic-based Methods The Pharmaceutical Inspection Convention and Pharmaceutical Inspection Co-operation Scheme (jointly referred to as PIC/S) are two international instruments between countries and pharmaceutical inspection authorities, which provide together an active and constructive co-operation in the field of GMP.
50 Raman Related Documents (1) PharmacopeiaUnited States Pharmacopeia (USP34/NF29) General Chapter <1120> Raman SpectroscopyEuropean Pharmacopeia (Ph Eur 7.0) Raman SpectroscopyPharmacopeia of the People‘s Republic of China (2010)- Raman Spectroscopy 拉曼光谱法指导原则EMA Guidance – effective Oct 1, 2012EMA/CHMP/QWP/811210/2009-Final Guideline on Real Time Release Testing (formerly Guideline on Parametric Release) …may utilize process analytical technology (PAT) tools e.g. near infrared spectroscopy (NIR) and Raman spectroscopy
51 Raman Related Documents (2) ASTM E (2007) Standard Guide for Raman Shift Standards for Spectrometer CalibrationASTM E (2007) Standard Practice for Testing the Performance of Scanning Raman SpectrometersASTM E Standard Guide for Testing the Resolution of a Raman SpectrometerUS FDA Advancing Regulatory Science at FDA: A Strategic Plan (August 2011): Section 2. Develop new analytical methods:a) Investigate feasibility and value of using emerging and improved analytical technologies like Nuclear Magnetic Resonance (NMR), mass spectrometry, or near infrared or Raman spectroscopy for evaluating product quality of pharmaceutical agents, and evaluate whether these technologies should replace existing methods;
52 Raman Method–How do you make a regulatory filing? ChangesUS FDAEU (new)NIR/Raman based analytical method (new or alternative method)API: starting material, intermediateFP: excipient, productAnnual ReportPAS (if high risk to quality)*Type IB as “other changes to a test procedure (including replacement or addition).Changes requiring prior approvalChange in test procedure for the immediate packaging or reagents (low quality risk) of the API & FPType IAMinor variation- Not requiring immediate notification1.Real-Time Release Test2.Introduction of new or extension of existing Design SpacePAS (Post Approval Supplement)Type IIMajor var. Significantimpact on Q/S/EReference only- Need to consult with specific Regulatory Affairs contactRef: US FDA 21CFR314.70Rev. 2012, and EMA "Guidelines on the details of the various categories of variations for medical products for human use and veterinary medicinal products" (2010/C 17/01).
53 Challenges for using Raman Methods Within Pharma organizationIntroducing new technology can be a challengeQA or regulatory staff adverse to changeConfusion about the validation elements & procedures:B&W Tek can helpLocal regulatory agenciesLack of existing USP, EP or EMA guidance specific to Raman IDDifferent outcomes and comments from reviewersInstrument vendors/sales representativeLimited knowledge of the regulatory processNot always staffed for applications support
54 Strategy and Approach for Raman Implementation Define objectives- intended useProject Team: Requires complete preparation and coordinationIdentify project members and leader (coordinator)Treat Raman technique as a general GMP methodUnderstand advantages and limitations of techniqueReview GMP regulation (EU, US FDA, PIC/S) and pharmacopeia (USP, EP, EMA) requirementsContact instrument supplier/vendor and evaluate their technical/GMP/supporting capability; partner with supplierFull participation of internal departments (QC, QA, Regulatory affairs, Manufacturing, R&D) with open discussions to reduce communication/technical barriers
56 Q1: What is the difference between ID and purity tests “Identity” and “Purity” are two separate specifications.“Purity” implies some type of quantitative method; typically HPLC“Purity” is a test for specific impurities such as low level contaminants/adulterants.An identity test says nothing about purity, strength, or composition.Usually gives “binary result”: Positive/Negative; Pass/Fail; Yes/NoIdentity verifies it is similar to what has been used in the pastEvery proposed method should be validated if it is used for quality decision!
57 Q2: How long does it take to get a Raman Method in place Bottleneck:Document preparation & internal reviewAssembling samplesExample: Develop an ID method for 10 materials:Training < 1 weekMethod development 1-2 weeksSpecificity & robustness tests < 1 weekDocumentation (prepare, review & approve) 2-4 weeksThe first ID method validation takes weeksThe next ID validation project takes only 1-2 weeksB&W Tek can provide Support Services
58 Q3: How do traditional vs. spectroscopic-based methods compare? In traditional methods often have a univariate calibration modelSpectroscopic method based on chemometric modeling or statistical calculationsIdentification/QualitativeCalculate spectral similarity between method and unknown samplesSpecificity test- interval validation evaluation focus on the model correctness of spectral libraryNote: Different terminology:Calibration in chemometrics- method/model calibrationCalibration in GMP environment - e.g., instrument calibration
59 Q4: Can I use a commercial spectral library Motivation: reduce development time & manpower if the spectral library or database can be supplied by Vendor or Third Party.Challenges: under Pharma GMP: qualified raw material library should be evaluated:Authentic materials?- identify sources, lot number, expiration date etc. and assess the authenticity (with test record)Qualified instrument? – assess instrumentation/configurations (resolution, range, probe etc),experimental and GMP qualification documents.Suitability: Commercial library/database could be used for general screening purpose (e.g., contract lab, food/chemical industry),Understand intended use- e.g., “…This library is designed for those laboratories studying pharmaceutical formulations using Raman spectroscopy”Question: Can a Pharma Lab use commercial library for an ID test ?
60 Issues on using Commercial Spectral Library or Database US FDA Warning LettersJilin Shulan Synthetic Pharmaceutical Co 5/13/2010#5. You routinely failed to utilize a suitable standard for the spectral comparison during the FTIR testing of XXX. Instead, you compared sample spectra with the spectrum of XXX in the Chinese Pharmacopeia. The USP requires that you compare the spectrum of your test sample with that of your reference standard.Yunnan Hande Bio-Tech. Co. Ltd. 10/15/10#4. a. Your firm does not concurrently analyze the FTIR samples with a XXX reference standard. Instead, you analyze the sample against a spectrum stored in the memory of the FTIR. During our inspection, you were unable to provide data to show when and how you prepared and qualified the FTIR spectrum standard.EP Raman Spectroscopy…This database is then valid for use only with the originating instrument, or with a similar instrument, provided the transferred database has been demonstrated to remain valid.
61 Summary Increase analysis RamanIncrease analysisCost reductionImprove operationsCompetitivenessSummaryRaman has proven to be a promising tool to increase operational capabilities and reduce cost while providing rapid material identification.61
62 Thank you for your attention Dr. Katherine BakeevDirector Application Support ServicesB&W Tek, Inc.x141662
65 Why do Raman ID methods fail? (1) Failure to include lot-to-lot variabilityuse numerous non-consecutive lotsrealize material potential variability (particle size, moisture, mechanics etc.)Failure to understand the limitation of Ramanfluorescence interferencelow Raman scattering signal (e.g., cellulose, salts, metals)low concentration (e.g., tablet API)Potentially close materials give ambiguous results (pass more than one method)Adjust p-value to narrow acceptance range for material, and limit overlap of methods
66 Why do Raman ID methods fail? (2) Occasional operational errorsSampling/scanning problemQuality of sample vials (sealed, glass etc.), plastic bags or containersMethod specific to particular sampling accessory and/or packagingLimited knowledge or misuse of identity algorithmsPerformance/sensitivity of identification method (p-value is based on a PCA model on the 20 scans in the method)
67 Deficiencies of Filed Spectroscopic Methods (1) Absence of instrument descriptionLight source, design, detector, sampling module, software packageMinimal description of chemometric techniquesAlgorithms for pass/fail resultsMethod validationLimited results for method validationWithout acceptance criteria (ID thresholds)Detailed results (with predicted values, not just pass or fail)Inappropriate selection of negative control samples (material structurally similar or closely related to the analyte)
68 Deficiencies of Filed Spectroscopic Methods (2) Inadequate SOPs and plan for change controlSpectroscopic method performance verification?Spectroscopic method maintenance plan?Lack of information for library/method update planSOPs for method/library update?