Genotoxic impurities refer to compounds or substances that may cause damage to genetic material (DNA or RNA) in biological cells. These impurities may cause adverse effects such as genetic mutations, chromosomal aberrations, DNA breaks, thereby affecting normal cellular functions and growth, and even leading to diseases such as cancer. In recent years, the emphasis on genotoxic impurities has continued to increase, triggering more research and discussion on genotoxic impurities. In the process of new drug development, genotoxic impurities are an important issue that we must address. Thorough research and assessment of genotoxic impurities are crucial. It ensures patient safety, enhances drug quality, meets regulatory requirements, and is one of the key steps in new drug development.
01. Can the mini-Ames test be used instead of the Ames test?
Zeng Bi: In our experience, during the IND application stage, using the mini-Ames test for impurities can be declared as a substitute material for the formal Ames test. However, during the application for new drug and generic drug marketing, the standard Ames test should be employed. It’s important to note that the microscale testing referred to in Note 2 of M7(R2) does not refer to the mini-Ames test but rather to measures that can be taken when the quantity of the test compound is limited and cannot achieve test concentrations in line with regulatory guidelines.
Analytical Testing Center
02. In the development of generic drugs, does each impurity need to undergo software assessment?
Zeng Bi: Yes, it is clearly recommended in the M7(R2) Q&A that all impurities be simulated and assessed. Through the case studies shared in my livestream, it can be seen that some compounds may exhibit genotoxicity even without having structural alerts for genotoxicity. Therefore, even if certain compounds are not on our list of structural alerts, they still need to be assessed because they could potentially exhibit genotoxicity.
03. "If there is Ames test data available, is software evaluation still necessary?"
Zeng Bi: It depends on the situation. “If the Ames test data included in the database follows OECD guidelines or the standards of ICH S2 for the Ames test, meaning it uses 5 bacterial strains and conducts experiments with both metabolic activation and non-activation pathways, and the tests are conducted in compliance with GLP regulations, then if the Ames test results are negative, software evaluation may not be necessary. However, comprehensive database searches should still be conducted.”
04. At what stage of drug development is it evaluated?
Zeng Bi: From my understanding, each project needs to be analyzed on a case-by-case basis. My suggestion is to conduct the evaluation concurrently during the pilot production phase, after process stability has been achieved, such as when the process route is finalized and optimization work is completed. At this stage, significant adjustments to the previous process are typically not made, unless issues such as impurities exceeding standards or substantial decreases in yield are identified during subsequent scale-up processes. In general, it is recommended to conduct corresponding evaluations during the pilot testing phase. This allows for early method development and preliminary confirmation work, and as the scale-up process progresses step by step, data accumulation supports subsequent impurity control strategies.
05. Is the Case Ultra prediction compliant? Does CDE approve it as a replacement for the AMES test?
Zeng Bi: This is compliant, so you can rest assured. We have used Case Ultra for multiple generic drug projects, including new drug IND applications and impurity predictions commissioned by clients. There have been no issues.
Case Ultra has a Statistical Model (GT1_BMUT) which was developed and validated in collaboration with the US FDA through the RCA protocol. The training set includes 13,760 compounds. Additionally, there is an Expert Rule Model (GT_EXPERT), with a reference data set containing 17,309 compounds used for rule validation. There are 240 alert structures, including common positive alert structures, subtypes of reduced positive and enhanced positive alert structures, as well as globally reduced positive alert structures.
Case Ultra meets the ICH M7 requirement for using two models to predict compounds. All models have undergone rigorous validation, allowing users to view background data, alert structures, and related details. It has a clear algorithm and good prediction transparency, complying with OECD regulations .
06. In the case involving trifluoromethanesulfonic acid, the sample itself contains trifluoromethanesulfonic acid. Is this hydrolysis method still feasible?
Zeng Bi: In this case, during the initial method development phase, we tested the API for trifluoromethanesulfonic acid and did not detect this impurity. Based on this result, if I use this detection method and trifluoromethanesulfonic acid is detected in the sample itself, it will lead to an overestimation of the impurity. In such a case, it is possible to detect another fragment generated after hydrolysis, such as 3 -hydroxy-2-nitropyridine in this case.
07. What software can be used for impurity assessment?
Zeng Bi: In addition to Case Ultra, Derek and Sarah are also equally excellent commercial prediction software. Of course, besides commercial software, there are also some free software options, such as VEGA GUI. I have personally used this software, but the predicted results were not clear. This could be due to the relatively small database coverage and limited variety of compounds covered by the software, resulting in fewer compounds used to train the software algorithms. This impacts the reliability of the calculation results and presents significant challenges in terms of compliance. It’s advisable for users to exercise caution when using such software.
08. In clinical practice, if an impurity in a locally administered single-dose formulation exceeds the defined limit, what safety evaluations should be conducted?
Zeng Bi: For impurities exceeding defined limits in formulations, it is recommended to conduct studies following the decision tree for identification and qualification of degradation products as outlined in Attachment 3 of ICH Q3B (R2). Firstly, it must be determined whether the impurity can be reduced to or below the defined limit. If not, considerations should be made based on patient population and duration of administration for the following studies: mutagenicity studies (gene mutation, chromosomal aberration), general toxicity studies (single species, typically 14 to 90 days), or other toxicity endpoint studies as deemed appropriate.
