Questions around the need for a pre-use post-sterilization integrity testing (PUPSIT) strategy have challenged operations and their filtration experts since the inclusion of PUPSIT in Annex 1. Finding the correct answers became increasingly significant with the ongoing debate about the wording and interpretation in the Annex 1 redraft.

To respond to specific needs from regulators and end-users, and determine when and why a risk assessment would require the use of PUPSIT for sterilizing-grade filters, the PDA and BioPhorum came together to form the Sterile Filtration Quality Risk Management (SFQRM) consortium in 2017.

After completing a significant body of work, the consortium has fulfilled its mission and deliverables and so, from December 2020, the team will transition into a BioPhorum SFQRM Community of Practice (COP) for the implementation of a risk-based approach to PUPSIT.

“The team has given the industry the best possible answers to when and how to use PUPSIT,” said Tony White, Director at BioPhorum. “Its three years of multi-company data mining, filter studies and method comparisons have given industry the best chance yet to break the cycle of confusion and uncertainty with science and data. For this, we should give each person a massive vote of thanks.

“It has been a privilege to witness from a distance the commitment, effort and intellect directed at the PUPSIT conundrum by so many outstanding industry filtration specialists,” White added. “It is a measure of their commitment to the cause that they are not happy with their outstanding publications. This group is determined to continue to help the whole industry absorb and master the decisions around PUPSIT.”

Strategy roadmap

The COP will use a strategy roadmap with the same name to guide work in 2021 and beyond. The roadmap is a tool for companies wanting to understand how to use the following SFQRM publications in their PUPSIT decision-making process:

1. Points to consider for the risks associated with sterilizing grade filters and sterilizing filtration (June 2020)
2. Datamining to determine the influence of fluid properties on the integrity test values (July 2020)
3. Test process and results of potential masking of sterilizing grade filters (July 2020)
4. Points to consider for implementation of pre-use post-sterilization integrity testing (PUPSIT) (July 2020)
5. The use of scientific data to assess and control risks associated with sterilizing filtration (July 2020)
6. Impact of fouling propensities of filtered drug product on the need for conducting PUPSIT – submitted to PDA for inclusion as an appendix in PUPSIT Points to Consider (expected Q1 2021).

All of these SFQRM consortium documents can be found on www.biophorum.com or www.pda.org. The sites also contain other useful PUPSIT educational, implementation and communication tools, such as:

• Article in American Pharmaceutical Review by team member Will Peterson, Merck MSD 
• Three-part series of educational video vignettes
• BPI and Aseptic Processing Conference presentations
• PDA and BioPhorum webinar presentation recordings and files
• PUPSIT FAQ page.

The PUPSIT conundrum

The original industry concern was the potential of filter flaw masking, where a flaw is not detected with the post-use integrity test. This started the sporadic enforcement of PUPSIT implementation and the urgent need to determine whether there is such a masking propensity. Also, the risk of masking, filter handling, transportation, etc., had to be weighed against the risk of a PUPSIT implementation as it added significant complexity and therefore risk to the sterility assurance of the process. This meant science-based tests and evaluations had to be performed to create a substantial database for such an appropriate risk assessment.

Over three years, the consortium reviewed available industry data, performed risk assessments from filter production through to installation, and performed fouling studies on flawed filters to expand the available scientific data. During these investigations, the consortium kept regulatory authorities informed and collaborated closely with them.

The result of four workstreams demonstrated that filter masking will only happen during extreme conditions and blocking rates (>80%). They also found that quality control measurements within filter production are excellent and filter flaws are rare. The teams recognized that the implementation of PUPSIT is a significant undertaking and requires careful assessment. In summary, a robust risk assessment of the terminal sterilizing filtration step is much more valuable than the generic implementation of PUPSIT.

After completing such a major body of work, the COP is now looking to the next phase of its development – helping industry implement its PUPSIT approach.