PUPSIT & the Proposed Annex 1 Revision

Since its publication in December 2017, the proposed Annex 1 revision has been much discussed. As coleader of the team that prepared PDA’s comments on the revision, I am intimately familiar with the intricacies of the document. As such, I want to share some thoughts on the revision, culminating in four pieces of advice concerning one of the most debated points of contention within Annex 1.

The draft revision to the EMA Annex 1 (Manufacturing of Sterile Medicinal Products) strongly encourages the use of quality risk management (QRM) principles in the design and implementation of process control measures. The draft, however, continues to include prescriptive methods for process control that may dissuade companies from using alternative risk-based approaches.

Some European regulators have asserted that these prescriptive methods are guidance and should not be considered requirements. Yet there remains concern that these methods will be viewed by both industry and inspectors as requirements. Among these prescriptive methods, one in particular is drawing considerable scrutiny: the requirement that the sterilized filter and assembly be integrity tested prior to use, more commonly known by the acronym “PUPSIT” for “pre-use, post-sterilization integrity testing.”

For industry, the use of PUPSIT presents a dilemma. There has been a significant increase in regulatory enforcement of PUPSIT by some regulatory bodies, thus posing a compliance risk to companies not using PUPSIT. But as companies use or contemplate the use of PUPSIT, they recognize that the complexity of the PUPSIT assembly and procedure may add significant risk to the aseptic process, exceeding the relatively low risk of undetected filter flaws. As a result, PUPSIT has become the one most talked about issue regarding the draft Annex 1 revision.

First, Some Regulatory Background

Although not specifically designated, the PUPSIT requirement has appeared in Annex 1 for decades. It is important to note that while the PUPSIT language has appeared in previous versions of Annex 1, enforcement as a requirement, has increased since the last publication in 2008, most notably within the last 2–3 years.

The current Annex 1 version states in Section 113 that … “The integrity of the sterilized filter should be verified before use and should be confirmed immediately after use by an appropriate method such as a bubble point, diffusive flow or pressure hold test.”

The draft 2018 revision language restates this requirement in Section 8.84, “The integrity of the sterilized filter assembly should be verified by testing before use, in case of damage and loss of integrity caused by processing and should be verified by on line testing immediately after use by an appropriate method such as a bubble point, diffusive flow, water intrusion or pressure hold test.”

The revised Annex then adds the following limited exclusion: … “It is recognised that for small batch sizes, this may not be possible; in these cases, an alternative approach may be taken as long as a formal risk assessment has been performed and compliance is achieved.”

At this point, PUPSIT is primarily a European concern but keep in mind that the current draft revision was prepared by an international PIC/S working group that included non-European representatives, such as the U.S. FDA. Currently there is no requirement, written guidance, or apparent expectation from FDA on the need for or use of PUPSIT. Risk assessments of PUPSIT assemblies may indicate an additional level of risk to aseptic process and product sterility posed by the inclusion of the PUPSIT procedure itself. To date, there does not appear to be much FDA official opinion one way or the other.

What is the PUPSIT Procedure?

PUPSIT is a test to mitigate situations where a sterilizing grade filter passes the supplier’s initial integrity testing but may become defective during sterilization, and that defect is no longer detectable during end of use or post-use integrity testing. In this scenario, it is surmised that material in the solution being filtered can clog the hole or flaw in the filter, thus blinding or masking the defect during the end of use filter integrity test.

PUPSIT requires complex designs of the filtration system and filters downstream side. Since the filter must be wetted and the filtrate side set at atmospheric pressure conditions, additional vent filters, aseptic connections, valving and tubing is added to the typical filtration design. This means that test system leakage, sterilization process difficulties, condensation collection, improper value sequencing, back pressure, increase of user installation faults due to the higher complexity and other complications can pose additional risk to the performance of the product filtration system, including microbiological contamination. These additional risks would very likely not be detected downstream by the sterility test.

Is the benefit of uncovering a yet-to-be established masking effect worth these risks? Some European regulators say “yes,” many companies say “no.” The draft revision allows the use of risk assessments to determine the answer to that question. Although not specifically mentioned in the PUPSIT sections of the draft revision, it has been suggested that those companies that do not think PUPSIT is appropriate, can perform a comprehensive risk assessment to support that position.

The draft revision of Annex 1 states in the Principles section that “Risk assessments should be used to justify alternative approaches to those specified in this Annex only, if these alternative approaches meet or surpass the intent of this Annex.” Some European regulators have expressed concern that these PUPSIT related risk assessments appear biased with a predetermined outcome of avoiding PUPSIT. Therefore, the assessment is being found to be non-persuasive.

4 Pieces of Advice to Industry

1. Employ Unbiased QRM principles

In evaluating PUPSIT, consider the objective of the assessment less an indictment of the PUPSIT process and more of a QRM assessment for preventing filtration failures. PUPSIT is not a stand-alone objective, instead, the objective is a robust process to manufacture sterile product.The outcome of the QRM effort should be more than a rejection or defense of PUPSIT. It should recognize the steps to prevent and mitigate filtration-related risks in a holistic view of the overall contamination control strategy.

Employing QRM and risk-based thinking approaches to assessing the value and need for PUPSIT combines several efforts. Information from studies and data analysis would be useful in determining the conditions, product types and probability of masking. Risk assessments of the filter manufacturing, transport, sterilization and use processes help determine where potential process weaknesses that might result in filer defects, and that can be masked, might occur. Best practice guidance for filter integrity testing, including PUPSIT if warranted, helps determine effective process control strategies that minimize risk to product sterility.

2. Assess the Relative Risk of a Masking Effect

PUPSIT detractors claim there is no evidence of masking phenomenon occurring. PUPSIT supporters, including some European regulators, claim there is anecdotal evidence and a supposition of risk. Published data, however, is scarce at best, with little, if any, conclusive evidence of masking phenomenon cited in literature. The hypothesis that defects can form in integral filters during sterilization, and then be masked during filtration, to the extent that the flaws are not uncovered during end of use integrity tests, should be further investigated.

Analysis of existing filtration performance and historic study data can be useful in providing the missing data needed to show if, and under what conditions, filter defect masking can occur. This may be done at the end user level, showing that the post-use integrity test data are consistent without any unusual outliers, a sign that the membrane matrix and blocking rate has been as consistent. Furthermore, laboratory masking studies may also need to be performed to uncover under what, if any conditions, and with which products and product types masking may occur. It is believed that if this effect can occur, it would depend on a select combination of product, process conditions and filter defect characteristics.

3. Assess the Relative Risk of Filter Manufacturing and Use Related Defects and Failures

European regulators have expressed concern over the robustness of control measures designed to mitigate filter defects, the level of control by filter manufacturers, the robustness of the filter manufacturing process, user handling of filters and assemblies, and the effects of sterilization as possible failure modes for filter defects. The concerns were addressed in a joint statement by four major filter suppliers published in the April 2018 PDA Letter (1). To further address these concerns, QRM principles and assessments can be performed on the overall filter manufacturing, transfer, sterilization and use processes, conditions, and procedures. These QRM activities should include the assessment of potential risks to sterile filtration, the development of control strategies to prevent filter failure, and appropriate use of detection controls (e.g., PUPSIT). The focus of these assessments should be prevention of those conditions that might result in flaws and masking, rather than reliance on testing. Risk assessment guides for filter manufacturing, handling, shipping, sterilization, usage, and integrity testing can also be helpful as a guide for audit of the effectiveness of filter manufacturer control systems.

4. Determine Best Practices

If the data analysis and process assessments suggest that the filtration conditions, product composition and mitigation strategies should include PUPSIT, then it is important that PUPSIT be performed correctly. PUPSIT is a complex procedure, employing complicated valving, connections, interventions and sequencing procedures. Failure to properly perform PUPSIT can place product quality at risk. Therefore, best practice guidance, procedures, and training are needed for the selection, design, installation, qualification,and operation of PUPSIT.

Some Final Thoughts...For Now

The question is not whether to perform PUPSIT, rather, are there conditions that pose a risk of filter flawing and masking, and if so, what steps can be taken to reduce or eliminate those risks? PUPSIT is not a process objective. The objective should be using QRM throughout the sterile filtration process to ensure reliable product sterilization and maintenance of sterility throughout the process.

More information is needed to determine the true risk of filter failure and the masking effect, as well as the relative risk of mitigating actions, including PUPSIT. Industry associations such as PDA are providing a valuable platform for conducting studies, analyzing data, drawing conclusions from results, providing QRM and risk assessment guidance, and helping develop best practices.

Regulators should avoid prescribing or requiring one control or test method. Instead, encourage companies to develop comprehensive well-designed control strategies. Sound, well-performed risk assessments of the overall filtration preparation and process should be designed to uncover the conditions that present risk of filter defects, and the integrity test failure to uncover those defects. PUPSIT may or may not be part of that depending on process specific conditions.

Once risks are understood, then control strategies and actions can be developed. Any actions or steps considered for inclusion in the aseptic process, including PUPSIT, should be first assessed for relative risk to product quality, and any resulting procedures should be carefully planned and performed with those assessed risks considered and adequately addressed.

Reference

1. “Sterilizing Grade Filters and PUPSIT.” PDA Letter 54 (April 2018) 19 www.pda.org/pda-letter-portal/archives/full-article/sterilizing-grade-filters-and-pupsit (accessed Jan. 23, 2019).