Essential Medicines Policies, WHO, Geneva.

1. Essential Medicines Policies, WHO, Geneva. Technical Briefing Seminar: 2011 Prequalification Programme: Priority Essential Medicines WHO-PQ INSPECTIONS Presented by Deus K Mubangizi Technical Officer mubangizid@who.int

2. WHO-PQP Inspections In this presentation: • Procedures and standards used for WHO-PQP inspections • WHO-PQ Inspections: avenues for capacity building and collaboration • Observed deficiencies during Inspection of: • FPP and API manufacturers + QC Labs • Contract Research Organizations (CROs) • Summary and conclusion

3. WHO Prequalification: Inspection activities APIs, FPPs, BE/CROs, QCLs

4. Prequalification Programme: Use of Inspection reports from other NMRAs • Inspectorates whose reports are recognized: • PICS member inspectorates • EU (EDQM + EMA) • USFDA – new member of PICS • What GMP evidence to submit: • SMF – Up-to-date • Inspection report - conducted NMT 2 years • + CAPAs to deficiencies + final conclusion • Product Quality Review – not more than 1 year old • Review of the report: • scope covered the specific API • Is comprehensive and supports the final outcome. • PQP reserves the right to inspect the API manufacturer – as long as product is active in WHO-PQP. • on-going GMP compliance will be confirmed by WHO • Desk review may only be use once in every 5 years.

5. Prequalification: Inspection Processes • By a team of qualified and experienced inspectors • WHO representative (qualified inspector) • Inspector from well-established inspectorate (Pharmaceutical Inspection Cooperation Scheme countries – PIC/S) • National inspector/s invited to be part and observe the inspection • Observer from recipient/developing countries (nominated by DRA of the country) • Scope: • Compliance with guidelines: • GMP for API and FPP sites, • GCP for CROs, • GLP for FPP/API factory QCL, CRO-BAL, NQCL, IQCL • Data verification – data manipulation, falsification, (validation, stability, clinical, bioanalytical)

6. Risk-based approach in:definition and classification of deficiencies • Deficiencies are descriptions of non-compliance with GMP requirements. • A distinction is made between deficiencies as a result of: - • a defective system or, • failure to comply with the system. • Deficiencies may be classified as: • Critical Observation – potential risk harm to the user • Major Observation – major deviation from GMP/GCP • Minor or Other Observation – departure from good practice

7. Risk-based approach in:Conclusion following an inspection • When there are "other" observations only: • considered to be operating at anacceptable level of compliancewith WHO GMP. • The manufacturer is expected to provide CAPAs. • CAPAs are evaluation and followed up during the next routine inspection. • When the are "other" and a few "major" observations: • compliance with WHO GMP/ICHQ7 is madeafter the CAPAs have been assessed. • CAPAs for majors to include documented evidence of completion. • CAPAs paper evaluated ± an on-site follow up inspection. • When there are "critical" or several "major" observations: • considered to be operating at an unacceptable level of compliance with WHO GMP/ICHQ7 guidelines. • Another inspection will be required

8. Information put in public domain - available for use by NMRAs: WHOPIRs and NOCs • These are published in response to the WHA Resolution WHA57.14 of 22 May 2004, which requested WHO, among other actions: • "3. (4) to ensure that the prequalification review process and the results of inspection and assessment reports of the listed products, aside from proprietary and confidential information, are made publicly available;" • A WHO Public Inspection Report (WHOPIR) reflects a positive outcome after an inspection • A Notice of Concern (NOC) is a letter reflecting areas of concern where the non-compliances require urgent attention and corrective action by the manufacturer or research organization.

9. Prequalification Programme: International norms, standards and guidelines used in inspection activities to ensure wide applicability USP BP Ph. Eur. Ph. Int. Other guidelines e.g. ICH, ISO http://apps.who.int/prequal/assessment_inspect/info_inspection.htm#2

10. WHO-PQ offers new avenues for collaboration in inspection • WHO-PQ Collaborative Procedure in Inspections • nominated inspectors from NMRAs of selected member states are invited to participate in WHO-PQ organized inspections and in turn, the NMRAs is given appropriate access to outcomes of these inspections. • Capacity building of NMRAs inspectors. • Facilitating use of WHO-PQ inspection results in national regulatory environment for information and decision making. • Facilitation of harmonization through joint inspections and sharing of outcomes. • Share the workload and promote avoiding duplicative inspections. http://apps.who.int/prequal/info_general/documents/inspection/NMRAs/GUIDANCE_WHO-PQM_NMRAs_CollaborativeProcedure.pdf

11. Inspection observers have been from:⃟ AFRO Region - External Observers⃟ WPRO Region – Host country Observers

12. Medicines Prequalification Process Expression of Interest Product dossier SMF Inspections Assessment Additional information and data Corrective actions Compliance Compliance Prequalification Dossier maintenance (variations) Handling of complaints Monitoring

13. Zidolam-NZidovudine 300mg, Lamivudine 150mg and Nevirapine 200mg tablets • Prequalified on 23 May 2006 with reference number HA275 • Manufactured by Hetero at its Unit III, Andhra Pradesh, India

14. Source of Information: MSF12th September 2011 PHOTO by MSF Alert from CHMP to MSF

15. Immediate Investigation13th September 2011 • Alerted Kenya Pharmacy and Poisons Board • Had not been notified by CHMP (Kenya) or MSF (Kenya) • Special Inspection of Hetero Unit 3: • Retention samples • Manufacturing records • Analysis records • Distribution records • Observed re-analysis of retention samples

16. Findings (1) Genuine Zidolam-N Batch No. E10076 Falsified Zidolam-N Batch No. E10076

17. Findings (2)ZIDOLAM-N Batch No. E100766

18. Findings (3) • Affected batches; • E100766, E110467, A9351, A9357, A9366 • the genuine batches E100766 and E110467 were never supplied to the Kenyan market • the quantities of Zidolam-N with a reference to “batch number A9351, A9357 or A9366”, found in Kenya, exceed the quantities manufactured, packed and dispatched by Hetero as batches A9351, A9357 and A9366 to Kenya.

19. Findings (4) • Results of analysis by Kenya’s NQCL and Hetero: • Intact samples comply with the manufacturers’ and with international pharmacopoeia specifications. • Open samples discoloured with high friability, low assay, fungal growth: Possible poor handling by patient. • Nature of falsification: • Relabeling and repackaging of donated batches (whose expiry dates are unknown) in order to divert them to the commercial market. • Extent and conditions under which the falsification (re-labelling) was undertaken is unknown and thus the quality of the products cannot be fully ascertained

20. Actions taken • Collaborated with Kenya authorities in conducting investigation and taking necessary actions. • Special inspection of the manufacturing site. • Analysis of complaint and retention samples. • KPPB ordered a recall of the batches. • Issued public notices which were updated as more information became available – careful not to cause anxiety and fear.

22. Lessons for the future • Inadequate market controls and procurement procedures: • Diversion of donations, relabeling and repackaging. • Purchase from middlemen at very low prices without question. • NMRA not informed promptly. • PQ prompt reaction and coordination facilitated quick investigation. • Collaboration between PQ, MSF, KPPB, KNQCL and Hetero was crucial for the success of investigations.

23. Recommendation • Supply chain should be fully known and actions should be taken to shorten it wherever possible • A supply chain is no stronger than its weakest link

24. Inspection of FPP manufacturers

25. Deficiencies observed during PQ InspectionsPREMISES • Poor design and construction of premises: • Inadequate segregation. • Illogical process flow. • Inadequate provision for Utilities: HVAC, water, compressed gases • Poor design and management of the HVAC system: • Multipurpose plant used re-circulated air but had no HEPA filters. • Adequate pressure differentials: reversal of air flow. • No sequence of switching on and off of AHUs of adjacent areas. Mix-ups Contamination Cross contamination

26. Deficiencies observed during PQ InspectionsMATERIALS • Inadequate goods and materials management • Starting materials: sourcing and sampling – ID per container. • Packaging materials: inadequate sampling – ISO2859 or BS6001. • Intermediate and bulk products – holding time not set, or justified, or respected. • Finished products: Release procedures – no adequate review by QA or QP. • Rejected materials and products: not adequate segregation or disposal. • Reagents and culture media: no GPT, positive and negative control • Reference Standards: inadequate standardisation, storage and use.

27. Deficiencies observed during PQ InspectionsQC LaboratoriesIR: What not to do!

28. Deficiencies observed during PQ InspectionsQUALITY CONTROL: Microbiology • Media Preparation: • No positive and negative control • No separate room for media preparation • Equipment: • No separate autoclave for sterilization of media and decontamination of used media. • Environmental Monitoring: • Inadequate exposure of plate method, adequate air sampling or swabbing. • Validation of Sanitising Agents: no challenge tests using the standard stock cultures (103-104/0.1ml) in order to ascertain the minimum dilutions to effect a kill.

29. Poor aseptic techniques: Extensive movement of operators in Grade A close to the open vials. Vials that were not stoppered by the machine taken from the conveyor belt into class B area and manually placed back into class A to be manually stoppered. Inadequate Media Fill Tests Environmental monitoring: The viable particle continuous monitoring for Grade A zone does not cover the whole time period of setting up of the equipment and the vial filling-capping process. Personnel garments and gloves were not monitored after manufacturing operations in grade A/B areas Deficiencies observed during PQ InspectionsFPP manufacturers: – sterile products

30. Inspection of API manufacturers

31. Out of 126 API sites participating in PQ activities, 49 were accepted based on approval by PICS inspectorates and/or ICH countries while 31 were inspected.

32. The sites inspected were the ones producing APIs used in most FPPs (average each API site representing 21 FPPs). Thus maximizing use of available inspection resources.

33. Deficiencies observed during PQ InspectionsAPI manufacturers • The most frequently found deficiencies were: • Material management • SOPs • Cleaning • Others included: • Batch records • Labelling • Cross contamination

34. Inspections of Contract Research Organizations (CROs)

35. http://apps.who.int/prequal/

36. Prequalification Programme: International norms, standards and guidelines used in inspection activities to ensure wide applicability • HANDBOOK FOR GOOD CLINICAL RESEARCH PRACTICE (GCP) Guidance for implementation http://whqlibdoc.who.int/publications/2005/924159392X_eng.pdf • Guidelines for good clinical practice (GCP) for trials on pharmaceutical products. World Health Organization, 1995 (WHO Technical Report Series, No. 850), Annex 3. http://apps.who.int/prequal/info_general/documents/TRS850/WHO_TRS_850-Annex3.pdf • Additional guidance for organizations performing in vivo bioequivalence studies. WHO Technical Report Series, No. 937, 2006, Annex 9 http://apps.who.int/prequal/info_general/documents/TRS937/WHO_TRS_937__annex9_eng.pdf • Guidelines for the preparation of a contract research organization master file. World Health Organization, WHO Technical Report Series, No. 957, 2010 Annex 7, Page 271. http://www.who.int/medicines/publications/TRS957_2010.pdf Other guidelines e.g. ICH

37. CRO/BE Inspections: Problems with integrity, archiving and retrieval of documents

38. CRO/BE Inspections: Inadequate data integrity Source data either not available or authenticity questionable: • Source data could not be located to verify entries in VRFs • destroyed accidently by fire or rain • Sponsor claims the data were kept by the CRO, and the CRO claims the data were kept by the sponsor • Two of the ECGs shown to the inspectors, bearing different subject numbers and initials, were found to be identical. • Other ECGs bearing different subject numbers and initials appear to have been recorded from a single subject. Out of 95 ECGs copied by the inspectors, 43 appear to have been recorded from the same and single subject during a single session

39. The following images are lead III of the screening and follow-up ECGs of volunteers AND and KRK, respectively, as copied during the inspection. They show no difference in QRS complex morphology. • Lead III, screening ECG, subject AND • Lead III, follow-up ECG, subject AND • Lead III, screening ECG, subject KRK • Lead III, follow-up ECG, subject KRK

40. Manual reintegration of peak was done inappropriately and inconsistently for all peaks inclusive internal standard For some samples checked, especially QCs or standards close or outside the 15% of their nominal concentration, the baseline of the chromatograms were modified manually. This was not done appropriately and consistently for all peaks inclusive internal standard. For modified integration, initial integration was not available. No paper or electronic audit trail of manual integration available. Each analytical run did not include calibration and quality control samples. CRO/BE Inspections: Data manipulation – inappropriate manual integration of peaks

41. CRO/BE Inspections:Data manipulation - Identical chromatograms had different peak areas but the same area percent

42. Deficiencies observed during PQ InspectionsQC Laboratories: Conclusions • Data manipulation and misrepresentation is not acceptable to the WHO and according to the law of most national regulatory authorities (could result in the publication of NOC or NOS on behalf of the WHO). • The honest way is always the "right way". • Good ethics are key to reliable data. • Always prioritize data integrity – not quick batch release at any cost.

43. WHO-PQ InspectionsSummary and Conclusions ధన్యవాదాలు • API, FPP and CRO/BE Inspections are an important part of the WHO-PQP evaluation and continuous monitoring process • International norms, standards and guidelines are used in inspection activities to ensure wide applicability • Collaborative and Risk management principles are applied to ensure efficient use of available resources • Information put in public domain - available for use by NMRAs: WHOPIRs and NOCs • Inspection results show thatthere are still a lot of poor manufacturing practices out there. Collaborative effort and skills are needed to ensure access to medicines of assured quality. Results show that WHO-PQP has made tremendous contribution in this respect. • The support of NRAs in providing co-inspectors and observers is appreciated. This is good for: • Tapping into international skills • Ensuring transparency • Facilitating ownership • Contributing to capacity building