Use of the ‘purge tool’ in assessing mutagenic impurities

Historically, industry has responded to regulatory concerns by generating significant amounts of supporting analytical data. This approach exemplifies a quality by testing (QbT) paradigm – it is very resource-intensive (especially when applied to every PMI/MI that could occur in a synthetic process), it can be technically challenging, and it runs contrary to the underlying principles of quality by design (QbD). Tellingly, this QbT approach fails to acknowledge that these reactive MIs will likely be purged by prevailing downstream chemistry conditions. As a consequence, impurity fate mapping (or purging capability), analytical testing and control strategies for MIs/PMIs in drug substances and products have received significant focus during the evolution of this guidance.

One of the areas of particular interest to Industry is purging of MIs, because of the huge analytical resource burden required to support the current GRA process. Significant reduction in the analytical resource burden can be achieved by analysing the purge factor , i.e., the ability of a synthetic process to purge or remove a particular MI. The underlying principles supporting the purge factor are straightforward and involve identifying those intrinsic physicochemical factors affecting the processes’ capability to remove a specified MI. The most important parameters are reactivity, solubility, volatility, ionisability, as well as any other physical processes that remove impurities, e.g., preparative chromatography. These parameters are then allocated a relative weight (based on a pre-defined scoring system) in the overall purge factor.

The chemical reactivity (R) of the MIs/PMIs is the reactivity towards other reagents that will be encountered during the downstream chemistry. The reactivity is classified as high (R=100), moderate (R=10) or low (R=1). The solubility (S) of the specific MI/PMI is determined in the designated process solvents and can be classified as freely (S=10), moderately (S=3) or sparingly soluble (S=1). The volatility (V) of the MI/PMI is defined relative to the temperature of the reaction process and is allocated values of V=10, 3 or 1. Ionisability (I) is based on the potential to reduce levels of MI/PMIs based on manipulating the pH of the medium and liquid/liquid extraction. Finally, chromatography is allocated values of 10-100 based on the efficiency of the final separation. For each stage, the individual purge factors are multiplied to attain a stage purge factor. Then each stage purge factor is multiplied to produce the overall purge factor^2,3.

Lhasa recently initiated a cross-industry collaboration (Mirabilis) to further develop the purge tool to generate a semi-quantitative and reproducible approach to predicting the probable purge factor for these impurities at each stage within a synthetic process. One of the key goals of this consortium is to make the process consistent, open and transparent, so that regulators can easily view and interrogate the findings, increasing the chances of regulatory success and providing regulators with increased assurance of the output. The initial version of the Mirabilis software was released to the existing consortium (comprising Lhasa and eight pharmaceutical companies) in late 2014⁴.

Conclusion

The implementation of the purge factor and the activities of the Mirabilis consortium have the potential to significantly decrease analytical testing of MI/PMIs without any additional impact on patient safety. This approach is clearly aligned with the guidance provided in ICH M7, i.e., as part of the control section (option 4). Additionally, there is an emerging view that the purge tool may also be applicable to non-mutagenic impurities, i.e., ICH Q3A⁵ impurities.

References

1. ICH M7. 2014. Assessment and control of DNA reactive (mutagenic) impurities in pharmaceuticals to limit potential carcinogenic risk. Step 4. June 2014
2. Teasdale A, Elder DP, Chang S-J, Wang S, Thompson R, Benz N, Sanchez Flores, IH. 2013. Risk assessment of genotoxic impurities in new chemical entities: Strategies to demonstrate control. Org Proc. Res. Dev. 17, 221-230
3. Teasdale A, Elder, DP, Fenner, S. 2011. Chapter 9. Strategies for the evaluation of Genotoxic Impurity Risk. In Genotoxic Impurities: Strategies for identification and control, Teasdale, A. Ed. Willey, New York
4. Lhasa Limited Mirabilis. 2015. http://www.lhasalimited.org/products/mirabilis.htm. Accessed on 15th August 2015
5. ICH Q3A(R2). 2006. Impurities in new drug substances. Step 4, October 2006