Читать книгу Mutagenic Impurities - Группа авторов - Страница 117

The purge ratio is determined by simply dividing the predicted purge by the required purge. For example, if the predicted purge was 1 × 10⁶ and the required purge was 100, then the purge ratio is 10 000, indicating that the PMI in question is anticipated to be removed by the process to levels at least 10 000‐fold lower than that required to reduce to the acceptable limit established for the individual PMI.

As illustrated in Figure 3.2, the purge ratio can then be utilized to determine the most appropriate ICH M7 control strategy. Based on this ratio, Barber et al. [19] defined a series of action limits incorporating both the purge ratio and the phase of development, see Table 3.4. The action limits recommend the extent of data required to specifically support an Option 4 based control strategy. Clearly, the ratio reflects the extent of risk and hence the data requirements are directly proportionate to this, i.e. that increasing amounts of supporting data are required as the purge ratio decreases. The aim is ultimately to support the voracity of the predicted purge through experimental data, ensuring the robustness of the proposed control strategy.

Barber et al. [19] also set out a series of recommendations on reporting expectations for purge ratio justifications within regulatory submissions. In accord with the phase‐dependent data requirements defined within section 9 of the ICH M7 [8] guideline, this covers both clinical development and post‐approval, marketing phase.

Experience has shown that in reporting purge factors and the proposed control option, particularly Option 4, transparency is key. It is thus recommended to include predicted purge factors for each key purging step and within each step, each unit operation in the process. That should be augmented by inclusion of any supporting experimental physicochemical data that strengthens and increases confidence in the final prediction.