PUPSIT filter tests are a regulatory requirement in pharmaceutical production for European markets. Which options and strategies there are for this in systems engineering, what needs special attention in the design, and further background information will be presented below.
PUPSIT describes a process for filter tests, specifically and exclusively for product filters. “Pre-use post sterilize integrity testing” – indicates that the disposable filters will be tested for integrity after installation and sterilization, yet prior to use. This will ensure that the filter unit tested by the filter manufacturer (and possibly pre-sterilized) was not damaged during handling or installation, but especially not by the sterilization process in the system. A second filter test must be carried out at the end of a batch to recon- firm the correct condition.
PUPSIT has been and is still a controversial issue. The fact is that the European Medicines Agency (EMA) makes the implementation of PUPSIT filter tests for sterile medicinal products for the European market compulsory, while the American FDA does not stipulate this for its home market. Actually, even among experts of the PDA (Parenteral Drug Association) for example, further data evidence is requested to be able to better weigh the advantages and potential risks, which cannot be ruled out when conducting PUPSIT tests. One thing is certain: Anyone planning a system today must adhere to the regulations, which – as outlined – are clear.
In the CIP/SIP applications (Clean-in-Place/Sterilize-in- Place) the sterilization process is compulsory, since the product filter is manually built into a non-sterile system at the designated place at this point in time. The SIP process ensures that the microbiological contamination (load) in the complete system, in filters and pipework, is reduced to the regulatorily intended sterile level. The steam sterilization applied at temperatures over 121 °C stresses the filter material and the pressure difference, which a filter with- stands during steam sterilization, is clearly lower than at room temperature. Consequently, the test seems to be entirely appropriate at this point – just like a test that is carried out after production.
The filter integrity test works and proceeds as follows: It is crucial that a built-in filter is initially completely moistened. Only in this state will the filter material develop enough airtightness to prevent almost uninhibited air flow through the filter. This condition is used by increasing air pressure on the filter with a special test device. As a criterion for integrity, the airflow or the pressure drop is measured via the filter membrane.
The integrity of the filter can be verified on the basis of these two principles. During the bubble point test, the pressure limit is determined by the point at which an air bubble is formed through the gas (pressure drop) on the sterile side of the filter. The slightly gentler forward flow test measures the flow rate of the airflow exiting from the back. In both cases the specific measuring devices record this data, which is used in turn to clearly determine the filter integrity.
A key question for executing the PUPSIT is whether the filter should be moistened with Water for Injection (WFI) or with the product itself? Both variants demonstrate various features, which will be treated below.
First, the process with WFI: This comes with certain challenges for sterile handling. Because the WFI must be fed through the previously steam-sterilized product pipe, a previously closed sterile system. This system is “interrupted” again for the connection between WFI container and product pipe, to dock a container, which must be considered as a potential source of contamination.
There is also the issue of the aseptic level of the WFI used. Its quality is defined, but the low aseptic level of a steam sterilization is not reached via distillation or membrane technology, according to critics. Also, during the PUPSIT test method the WFI must have a temperature of 20 °C. If the WFI was generated using distillation, this must first cool off again, which involves a certain waiting period and thus a certain risk of germination. At this connection point, another filter is therefore often installed for the WFI or also for the air introduced, to ensure pharmaceutical safety. In addition, it must be noted that WFI can still be found in the system after the test. A drying process ensures that during the filling process with pharmaceuticals, the undiluted quality of the medicinal product required is achieved as quickly as possible and only a minimum of product will be discarded.
The PUPSIT process with product creates its own challenges. Here the complexity increases in proportion, depending on how many different products are processed on one machine. Every product type has specific flow proper- ties, meaning that the parameters which are recorded and provide information regarding the leak tightness of the filter, are dependent on the product and must be specifically defined for it. Particularly with the ever more specific medicinal products or machines, which are designed for a wide range of products, this can entail considerable effort. Increasingly more sensitive medicinal products are being processed that require cooling. This can make a filter test with product difficult, if not impossible, if this needs to be processed cooled, although the test, as mentioned, requires 20 °C. Another argument that can be used against tests with product is that in the case of a damaged filter the machine is already practically in production mode, this needs to be interrupted and the tests and preparations must be restarted. This also represents additional effort compared to tests with WFI.
Testing no type in particular can be viewed per se as the better option. The individual conditions under which a system is designed will decide on this. In fact, Optima is currently building predominantly systems which handle both test types. This is often advisable, if various application scenarios are conceivable over the time horizon of system use. Because one thing is certain: Subsequent conversion of an existing certified and validated piping system with SIP entails a very great effort, which should be avoided wherever possible. The same applies to the decision of companies considering a market entry into Europe. The decision should be made for or against the implementation of PUPSIT tests during the design phase. Any subsequent change to the system is hardly realistic.
The question regarding what the options are for companies that manufacture on their machines for European markets as well as the US market is also interesting. It is certain that the technical requirements for carrying out PUPSIT filter tests on such systems must be met, and that these must be verifiably performed during production for European markets. Which strategies these companies apply, if producing on the same systems for the US market, is open and remains an individual decision for each company.
On the system side, there is ultimately another design option which produces no difference in terms of function. The control of the filter tester, which measures and evaluates parameters, can be integrated into the system and operated via the HMI. This dispenses with individual manual work steps, which are required for connecting the tester. The control, however, also can remain attached to the testing device itself and brought with it to the point of use. This decision has no effect on the actual test and the test procedure.
Finally, it is worth noting that the system manufacturer should ensure the blow-drying of filter cartridges after a test with WFI is as gentle as possible. There is also the risk of damaging the filters, which in this case, however – unlike with steam sterilization – would only be detected after completion of a batch. Optima thus equips the air inlet with special screens, to optimize pharmaceutical safety.