Swissmedic’s technical interpretation of Annex 1

, , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , ,

by Giuliana Miglierini

New insights on the interpretation of the new Annex 1 to Good manufacturing practices (GMPs) comes from the Swiss regulatory authority Swissmedic, that at the end of October 2023 published the first revision of its Q&As document (you can find it on the Swissmedicines Inspectorate webpage)

The technical interpretation refers to the revised Annex 1 to the PIC/S GMP Guide (PE 009), adopted on 9 September 2022 and entered into force on 25 August 2023 (with the exception of point 8.123 on lyophilisation, which will enter into force on 25 August 2024). The Q&As follow the same scheme and chapters of Annex 1.

Scope and Premises

According to Swissmedic, for certain types of advanced medicinal products (e.g. ATMPs or allogenic and autologous cell therapy products) specific considerations are required with respect to the fact they cannot be terminally sterilised or filtered. The unsterile patient material should also be considered. Requirements of Annex 2A, paragraph 5.29(b) should be followed for aseptic processing, that should be maintained from the time of procurement of cells through manufacturing and administration back into the patient.

Exceptions to the application of Annex 1 need to be always justified: the Contamination Control Strategy (CCS) is the appropriate tool to detail all risk analysis performed on the basis of the specific manufacturing processes under consideration.

As for the Premises, segregated unidirectional flow airlocks for material and personnel for grade A and B cleanrooms are expected in the case of new facilities. Temporary separation of the flows in the airlocks is the minimum requirement for existing facilities, together with a detailed risk analysis to assess the need for additional technical or organisational measures.

The transfer of materials in and out of a critical grade A cleanroom should be based on the careful definition of the technical and procedural measures associated with it. For example, prior introduction of materials in an isolator followed by decontamination is considered possible only for small batches and for materials resistant to VHP treatment. In all other cases, materials have to be sterilised before entering the already sterile isolator. The transfer process is also subject to a risk analysis to be included in the CCS, as well as to measures to control the maintenance of the integrity and functionality of the systems (also with respect to aseptic process simulation, APS).

Swissmedic specifies that the cleanroom sequence for the transfer of materials via airlocks or passthrough hatches is expected to be fulfilled for zones A and B. In the case of the passage from grade A to C, qualification is needed to prove adequacy of the established systems and procedures. The corresponding risk analysis has to be included in the CCS.

Updating equipment to reach full compliance with the new Annex 1 may require high investments. According to the Q&As, older barrier technologies should be subject to an in-depth internal evaluation to assess the need for new technical measures. The document underlines that starting from 25 August 2023 all barrier technologies not compliant with the new Annex 1 are considered deficient, thus companies should start projects to evaluate the upgrading of background cleanrooms and to define CAPA plans and interim measures to reduce risks.

The risk assessment should also include the evaluation of all automated functionalities and processes associated with the use of the isolator and the activities taking place in it. To this instance, Swissmedic highlights that robotic systems may help improving the reproducibility of operations and minimising both errors and manual interventions. Automatic processes are also expected for the decontamination of isolators, while for RABS manual processes might be used, provided they are designed to ensure reproducibility and are subject to validation and regular monitoring. The absence of negative effects on the medicinal product associated to the cleaning or biodecontamination substances used should also be validated.

As for barrier technology systems with unidirectional air flow, air velocity must be defined so that uniform airflow conditions prevail at the working positions where high-risk operations take place. Alternative air speed ranges or measurements at different heights in the system have to be scientifically justified in the CCS.

Utilities and Personnel

The section on Utilities offers additional guidance on systems used for water generation, that should be designed to allow for routine sanitisation and/or disinfection. Procedures are needed to define regular preventive maintenance of the reverse osmosis system, including the regular change of membranes. A suitable sampling schedule should be in place to regularly check water quality. More stringent controls are needed for the sampling of water-for-injection distribution systems, including daily microbial and bacterial endotoxin testing. The monitoring of the process gas should be performed as close as possible before the sterilisation filter.

Adequate training and qualification of all people working in grade A and B areas, including aseptic gowning and aseptic behaviors, is essential. According to Annex 1, this should include an annual successful APS. Swissmedic adds that, even if not explicitly required, practical process simulations, including manual interventions, should be carried out under the supervision of qualified trainers/QA; the company can choose if to integrate these process simulations into the APS.

Production and specific technologies

As for lyophilisation, initial loading patterns must be always validated, and revalidated annually. The Q&As specify cases where revalidation can be skipped, adding that a theoretical reference load is not acceptable. Revalidation has also to include temperature mapping for moist heat sterilisation systems.

Should a closed system be opened, this should be followed by cleaning (if required) and a validated sterilisation process. Alternatively, the system can be opened in a decontaminated isolator; a class A cleanroom with a class B background might be considered only for exceptional cases.

Non-aseptic connections can be carried out for coupling closed systems, provided a validated sterilisation cycle (SIP) occurs prior to use. Sterile aseptic connectors can be used if the supplier was checked and validated; data from the supplier can be used to file the relevant documentation, but handling of these parts has to be included in the APS.

Swissmedic also underlines that piercing a septum with a needle is to be regarded as a breach of the sterile barrier, and thus avoided for ascetic steps. Should this not be possible, temporary measures should be undertaken to prevent contamination.

Tube welding has also to be qualified and validated, and included in the APS if it is part of the aseptic filling process. The advice is to use more reliable systems, to avoid risks of undetected integrity deficiencies.

Critical single use systems (SUS) should always be tested for integrity by the end user on site before they are used in production. In case of difficult to test, small single use systems, the decision not to test their integrity must be justified in the CCS, as well as the decision to make use of test results provided by suppliers. To this instance, Swissmedic underlines that the comprehensive assessment (including quality system, etc.) should cover the SUS manufacturer/ s, as well as any subcontractors involved in critical services or processes.

Furthermore, the intended use of a SUS in the specific manufacturing process represents the basis for setting the respective acceptance criteria. The Q&As also detail the modalities for the visual inspection of SUSs and the possible acceptance of validation data provided by their suppliers.

As for extractables, the end user is expected to assess the data provided by the suppliers in order to define the need for additional evaluation or leachable studies. A redundant filtration step through a sterile sterilising grade filter, to be included as close to the point of fill as possible, is also encouraged, and its absence has to be justified. A risk analysis is required to justify the choice not to include pre-use/post-sterilisation integrity testing (PUPSIT) of sterilising grade filters used in aseptically processes.

Environmental and process monitoring

According to ICH Q9 (R1), the frequency of the risk review should be based on the level of risk determined for the specific process under consideration, as well as on the level of uncertainty of previous assessments. The recommendation of Swissmedic for new plants is to review the risk assessment after the first year of operations, so to take into due consideration the acquired experience. The document also suggests cases where more stringent action limits may be needed, and the type of statistics to be used to establish alert levels.

The use of rapid microbiological methods (RMM) requires validation and demonstration of equivalence with more traditional approaches. Details on the frequency of the interventions and their inclusion in the APS are also discussed, as well as the container/closure configuration and the distinction between liquid filling and lyophilisation.

The APS of campaign manufacturing represents a complex case for Swissmedic, for which the start-of-campaign (including aseptic assemblies if the case) and end-of-campaign studies should be both conducted. The Q&As also confirm that any contaminated unit with a contamination > 0 CFU results in a failed APS and requires the activation of the consequent actions. Production should resume only after completion of a successful revalidation.

Quality control

A university degree or an equivalent diploma in the field of microbiology (or other natural sciences, or medicine) together with a good understanding of the manufacturing processes under consideration are required for the person in charge of supporting the design of manufacturing activities and environmental monitoring.

As for raw materials, the need for microbiological testing should be evaluated taking into consideration their nature and respective use in the process. All specifications should be discussed and justified in the CCS.

Swissmedic also confirms that the bioburden has to be tested on each batch of raw material as incoming control as well as on the compounding solution in which it is formulated before sterile filtration. In the case of products with short shelf life, should an out-of-specification (OOS) event appear after release of the batch, a procedure is needed to inform doctors, patients, and health authorities, and to assess the connected risks and define remediation actions.