EU Annex 1 draft: GIC A3P Proposals Appendix 1

L20 December 2017, after several months of waiting, EMA, PIC / s and WHO finally published the draft of the new version of Annex 1 relating to the production of sterile drugs, long awaited by manufacturers. This draft follows the publication of 2 February 2015's "Concept paper on the revision of Annex 1 of the guidelines on good manufacturing practice - manufacture of sterile medicinal product". This concept paper introduced the reasons and objectives of the 1 review.

1. Background Annex 1 and Review Process by Authorities (EMA / PICs / WHO)

The current version of the 1 Appendix in effect since March 2009 (but published the November 25 2008) is the latest updated version of a series of updates to 1996, 2003, 2005, and 2007 since the release of the initial release. However, since the document has never been revised in its entirety, the German authorities (ZLG) have requested in 2012 a complete revision to the Inspections Working Group (IWG) of the EMA. A working group of 16 representatives (EMA / PICS / WHO) has therefore been set up in 2014 whose leader is Andrew Hopkins, MHRA inspector, GDMP expert. The draft published the 20 December 2017 proposes a document of 50 pages (instead of 16 pages of the 2008 version) composed of 11 chapters some of which introduce new concepts or new technologies (Pharmaceutical Quality System, Contamination Control Strategy, RABS, closed systems , Single Use Systems) and others a complete restructuring such as chapter 9 combining viable / nonviable monitoring and Aseptic Process Simulation. A glossary has been added (chapter 11).

2. Recap of the chapters of the draft Appendix 1 and points developed

The draft was opened for public consultation until March 20 2018.

3. Contributions from the ICG A3P Appendix 1

The 1 GIC Appendix 2014 created early and aggregating industry, consultants, and vendors had already submitted its proposed changes in response to the February 2 2015 concept paper on the revision of the 1 Appendix by producing 15 general comments and about 120 specific comments. Following the publication of the draft 20 December 2017 the GIC was reactivated early January 2018 for the complete review of the document.

The ICG Annex 1 having reviewed the draft was composed of the following members:
Sophie Amadio (Lilly), Patrick Balériaux (Aseptictech) Marc Besson (formerly SANOFI and GIC leader Annex 1), Lydia Breteau (Boiron) Olivier Chancel (Merial BI), Pierre Devaux (UPS Consultants), Philippe Duhem (Intertek) Jean-François Dulière (Technip) John Hargreaves (JHAC) Eric Hurtubise (Theapharma) Benoit Ramon (Sanofi), Philippe Robin (NOX group), Jonnathan Tafforin (Recipharm) Julien Triquet (GSK), Patrick Turlier (ex GSK), Dominique Weill (Delegate France AFNOR, CEN ISO).

The 11 chapters of the draft have been split between 5 under expert groups. The subgroups worked for 2 months by teleconference. The restitution and the consolidation of the proposals were validated in plenary meeting at the beginning of March.

During the review and in its proposals, the GIC wished to have a pragmatic and innovative industrial vision while retaining the GMP spirit necessary for the sterility assurance of sterile drugs.

A total of 9 general comments and 120 specific comments have been proposed in response to the draft 1 Annex. The specific comments are distributed as follows according to the 11 chapters:

3.1 General Comments

1 / The draft of the 1 Appendix uses the word "should" in almost all chapters and "shall" in very few cases (eg § 5.16). A3P suggests giving a clear definition of the meaning of "should" in this document, because generally (and especially in EN standards) "should" is considered a recommendation whereas "shall" should be used to indicate a requirement that is mandatory, which must be implemented and its implementation verified.

2 / A3P fully appreciates the introduction of a glossary but A3P thinks that some important definitions are still missing or should be reassigned such as (non-exhaustive list):

  • Inherent / Corrective Interventions (9.36 a / b): A3P suggests to add definitions according to PDA TR 22.
  • The terms LIMITS of alert / action are widely used in the document. However, only the alert / action LEVELS are defined in the glossary.
  • Pure / clean steam: "pure" steam and "clean" steam are still used by manufacturers. A3P suggests adding a clarification in the document and in the glossary to confirm the term to be used for the steam needed to sterilize the parts in contact with the product and to differentiate the vapor used for the non-product contact parts.
  • RABS: The glossary provides only an "open" RABS definition. The glossary should also provide a definition of "closed" RABS as this type of RABS is also used in industry. A3P suggests that definitions are also provided to differentiate RABS closed / open by design and RABS closed / open in operation.
  • Classification of controlled areas: A3P suggests to have all the definitions of classes A, B, C and D in the glossary and not in 5.3 so as to be consistent with the structure of the document. For example, these classes are widely mentioned in the 4 (personal) section whereas the definitions are only available in 5 (Locals). In addition, the term "A / B grade" is widely used in the document (eg 196 lines, 202, 204, 263, 281, 356, 381 ...) without a clear definition.
  • Disqualification of staff: A3P suggests to differentiate disqualification and suspension of staff when the suspension is due to prolonged absence and disqualification due to repeated discrepancies when monitoring personnel or the environment or to participation in a Media Non-compliant fill or inadequate aseptic techniques / practices.

3 / The draft of Annex 1 should clarify the scope of the document and confirm whether the approved version will apply to human and veterinary drugs as some requirements may not apply to both types of products. For example, 8.26 requires that all containers filled with parenteral products be inspected individually for external contamination or other defects. In the context of veterinary products, samples (not all individual units) should be verified for external contamination or other defects using validated methods and in accordance with Quality risk management principles. A3P suggests that a clear statement be included in the scope of the document.

4 / Promoting Quality Risk Management Principles (QRM)
The draft 1 Annex introduces the QRM principles in several chapters but the principle (46-48 lines) that internally determined measurements and verifications must meet or exceed the requirements of the Annex is not consistent with the QRM concepts and may be subject to interpretation. Annex 1 should state that the results of the risk assessment (based on scientific considerations) are acceptable when they meet at least the requirements listed in Annex 1.

5 / Promotion of barrier technologies
A3P considers that the use of barrier technologies should be promoted more specifically for new lines and processes when aseptic distribution is used for sterile products exposed in a Class A environment. When designing new RABS, the Annex 1 should emphasize the need to design RABS and associated processes / equipment for routine operation with closed doors as much as possible.

6 / Sterile Product Filters Integrity Test (PUPSIT)
The draft 1 Appendix 8.84 (XNUMX) still considers that the sterilizing filter product should be tested after sterilization and before use. This requirement may imply that the integrity of the filter is tested online. A3P considers that this test should be evaluated on a case by case basis by an objective risk assessment. For some installations and processes in place, the PUPSIT may add risk by requiring handling downstream of the sterilized filter. The RE must take into account the following aspects: risks of contamination downstream of the filter due to additional handling, risk of clogging of the filter media due to the formulation of the product, risk of deterioration of the filter media during sterilization, impact of the PUPSIT on the dilution of the product, risk of modification of the existing lines to implement the PUPSIT ... If the results of the RE show that the implementation of the PUPSIT does not entail additional risk for the sterility of the product, the modification of the Existing lines should be planned and realized. However, if the risk assessment establishes that the PUPSIT increases the risk of compromising the sterility of the product et it is shown that a loss of integrity of the filter (throughout its use) will always be detected during the filter integrity test after use, then the PUPSIT should not be implemented. However, A3P confirms that the implementation of the PUPSIT should be fully integrated during the design of new lines.

7 / Considerations on chemical agents used for decontamination and sterilization
Vaporized hydrogen peroxide (VH2O2) is widely used successfully for the decontamination of surfaces in insulators. It is recognized that this agent is used as a surface decontamination agent but is not claimed as a sterilizing agent. However, robust VH decontamination cycles2O2 can be developed to get more 6 log reduction for spores Geobacillus stearothermophilus. When effective and validated surface cleaning is achieved prior to decontamination, a validated and robust VH2O2 cycle eliminates microbial bioburden from product contact surfaces and primary components to meet the requirements of §6.6. A3P suggests that Annex 1 recognizes the effectiveness of commonly used chemical agents for the decontamination and sterilization of surfaces, specifies the requirements for their use and clarifies in particular what is expected for the contact surfaces of primary components in insulators such as that bowls stoppers or conveyors.

8 / LEVELS of Alert / Action vs LIMITS of Alert / Action
The draft of Annex 1 alternatively uses the terms limit et levels in many paragraphs. The A3P suggests that the meaning of these two terms be clarified in Annex 1. As stated in General Comments 2, the definitions of the Alert / Action LIMITS should be added in the Glossary. A3P considers that the recommended maximum LIMITS proposed in the draft of Annex 1 should not be considered as an extension of "product specifications" as there is no systematic relationship between environmental excursions and product contamination. Each manufacturer should establish its own ALERT / ACTION LEVELS based on qualification data and historical trends in order to continuously monitor its validated processes. However, A3P confirms that any microbiological contamination in the environment of a class A (or even an adjacent class B) must be investigated and impact tested for the product before release.

9 / Consideration of particles> 5 μm for the monitoring of zones in grade A, B, C and D: The draft of Annex 1 removed particle measurements> 5 μm at classification according to the new version of the ISO 14644-1 standard but still considers these particles for the monitoring non-viable particles in grades A, B, C and D. Some requirements are also contradictory, eg 9.13, 2 which states that alert limits should be defined on the basis of historical data and qualification. A3P suggests that monitoring of these particles be recommended (and not required as stated in 9.2). Each manufacturing site should establish its own QRM-based monitoring strategy and this strategy should be part of the overall site contamination control strategy. The alert / action levels should be defined according to each specific line / manufacturing process and, therefore, the recommended limits for particles> 5 μm in the 5 table should be deleted in Annex 1.

3.2 Specific Comments

The detailed review of the draft gave rise to 120 specific comments ranging from simple correction (typo, punctuation ...) to major changes in the proposed text. Each change was supported by a justification and a new text proposal was proposed (in red in the text).

The following tables show some of the major proposals of the GIC that may have significant consequences for the industrial sector in terms of changes to premises, processes, investments or organization. Specific comments referring to general comments (eg PUPSIT or Particles> 5μm) are not all included in these tables.

5. Next steps

The A3P responses were submitted to the EMA 19 March 2018. The European agency should receive many contributions from different associations such as PHSS, ISPE, LEEM ... To date, it is not possible to know the exact process of review within the IWG and the date of publication of the final version of the new Annex 1 (taking into account in particular the relocation constraints of the Agency following Brexit). Nevertheless a round table should be organized during the international congress A3P Biarritz 13 November 15 2018 during which a status of the revision and a selection of key points of the Annex 1 could be discussed between representatives of the authorities , GIC Annex 1 and industrial.

Share Article

2018 04 27 screenshot to 09.04.14


G Pharmacist, worked 30 years in the Pharmaceutical industry including 25 years at SANOFI where he held various positions including the Quality Department for several years on a site of production of injectable drugs under isolator for the USA, Japan and the EU . He also participated in Corporate Quality Operations in projects for the development of new sterile production units on SANOFI Group sites in China, Russia and India. In particular, he collaborates within the A3P in different working groups and leads the GIC dedicated to the new version of the 1 Annex.