Issue link: https://www.e-digitaleditions.com/i/1373953
8 Pharmaceutical Technology BIOLOGICS AND STERILE DRUG MANUFACTURING EBOOK 2021 P h a r mTe c h . c o m Manufacturing The results of the these tests established scien- tific data corroborating the fact that f lawed filter masking is a possibility, albeit at extreme condi- tions. Although these conditions are not normally seen in commercial processes, results suggests that manufacturers check the product and process conditions, since a f luid capable of high degrees of filter fouling and/or blocking could mask filter f laws and lead to false pass integrity test results. Either the sterilizing-grade filter must be protected against blocking or PUPSIT will be necessary. Data mining In this phase of the consortium's work, the goal was to mine the results of the integrity test data during the product-specific bacterial challenge test. As mentioned previously, sterilizing grade filters must undergo process validation activities. One of these activities is the product bacteria chal- lenge test, which determines whether the steriliz- ing grade filter creates a sterile filtrate, with the organism being subjected to the product and the filter challenged under process conditions. Typi- cally, using this method, 0.2-micron test filters and 0.45-micron controls are bubble point integrity tested pre-challenge, then challenged with 10 7 /cm 2 organism density, followed by bubble point test- ing after the challenge (11). The evaluation was needed to show that the number of cases in which the product and/or high-density challenge would shift the bubble point integrity test value of the pre-challenge test versus that of the post-challenge test. This shift would suggest the number of times one would find potential filter-masking conditions based on various fluids used for the challenge tests. In all, 2080 data sets, representing 531 bacteria challenge tests, were collected using different 0.2- and 0.45-micron filters with 518 different f luid streams. Results showed that a pre- and post-use bubble point value shift was extremely rare. Of 531 bacteria challenge tests performed, only five were outliers that ref lected a shift of the bubble point value pre- versus post-challenge. It was determined that the challenge f luid used within these five outliers were high foulant f luids, which very much supports the findings of the masking studies. These results also suggest that f luids with high levels of fouling potential could be de- termined during the process validation exercise, creating an indicator of whether there would be a high or low risk of fouling the filter. The study underscored the fact that the bac- terial challenge test analysis, which is typically performed during filter sizing and process vali- dation, could also be used to determine whether the f luid used has a high or low potential to foul filters, and whether the use and additional risk entailed by PUPSIT can be justified. Risk assessment Regulators expect the risk of critical process-control measures to be assessed, and this practice has also become standard procedure for the industry. Lan- guage in the proposed revision of EU Annex 1 ap- peared to allow for a risk-based assessment of the filtration process as criteria for offering an alterna- tive to the use of PUPSIT. Companies have used such risk assessments as justification for choosing not to use PUPSIT. These assessments usually take into account the risk of masking compared to the risk of sterile product pathway contamination posed by the PUPSIT process. Some regulators have found risk assessments to be non-persuasive and biased, with predetermined