Potential Genotoxic Impurity Consulting and Evaluation Services

Limits for organic impurities (process and API-related) which do not have mutagenic potential are covered under ICH guideline Q3A(R2).  For the majority of APIs (those with a daily dose less than or equal to 2 grams), impurities are permitted, but must be qualified in toxicology safety studies if present at 0.15% relative to the API, or 1.0 mg per daily intake (whichever is lower).

In contrast, limits set for substances possessing mutagenic potential under ICH M7(R1) are based on permissible daily intake limits to the patient, expressed in mcg/day.  Since cancer risk of a continuous low dose over a lifetime would be equivalent to the cancer risk associated with an identical cumulative exposure averaged over a shorter duration, acceptable intake limits are staged according to the projected duration of treatment for a given pharmaceutical:

 

To express the acceptable intake as a percent of the API, the maximum intended daily dose must be factored in:

Hence, the ICH M7(R1) guideline has its greatest impact on pharmaceuticals that are higher in dose, and those projected to be taken over a longer timeframe.

No.

Guidances that predated ICH M7(R1) (e.g. EMEA 2006, FDA 2008) instructed manufacturers to allow chemicals with mutagenic potential to be used only if unavoidable, and otherwise to control the impurity levels to “as low as reasonably practicable” (ALARP principle) while also not exceeding an agreed upon “threshold of toxicological concern” (TTC) patient exposure limit of 1.5 mcg/day.  It has been estimated, however, that 20–25% of all intermediates used in API synthetic manufacturing processes would prove to be mutagenic as determined by bacterial reverse mutation (Ames) testing.¹  The reality is that the reactivity that makes chemical intermediates synthetically useful also makes a high proportion of those substances DNA-reactive.²

Appropriately, the more recently issued ICH guideline M7(R1) does not require that applicants justify the process chemicals being employed in making APIs, but it does require applicants to develop suitably sensitive analytical test methods, and it incentivizes them to apply Quality by Design (QbD) principles to be aggressive in their process workups and scientifically demonstrate the virtually complete elimination of those substances prior to API isolation.

 

¹ Delaney, E . Reg. Toxicol. Pharmacol. 2008, 49, 107−124 – Request a reprint from author

²Elder, D. P. and Teasdale, A. Org. Process Res. Dev. 2007, 11, 985−995

Yes.

The ICH M7(R1) guideline states that “A computational toxicology assessment should be performed using Quantitative Structure-Activity Relationship ((Q)SAR) methodologies that predict the outcome of a bacterial mutagenicity assay.  The absence of structural alerts from two complementary (Q)SAR methodologies (expert rule-based and statistical) is sufficient to conclude that the impurity is of no mutagenic concern, and no further testing is recommended.”

Of course, not all companies can rationalize the expense of purchasing in-silico software programs designed to predict mutagenicity (e.g. DEREK, MultiCASE, Leadscope), or to hire an in-house toxicologist to interpret the data.  Third-party consultants who have purchased in silico software can assist in running the queries but the output data is frequently conflicting and ambiguous, and regulators’ interpretation of the predictive data tends to be highly conservative.

No.

In many instances it is worthwhile to conduct Ames testing, particularly if in-silico results aren’t strongly positive and if reduction of the impurity to extremely trace levels in your process steps could prove problematic.  If a given process impurity is flagged by in-silico testing and virtually certain to be a mutagen, however, a practical option is to skip the Ames test and just accept that the substance is a potential genotoxin.  This is particularly true if it is not difficult to prove that the substance is consistently removed under existing or modified process conditions.

ICH M7(R1) outlines four approaches for control of process-related, potentially mutagenic impurities (PGIs):

1. Include a suitable method among the API release tests with an acceptance criterion at or below the acceptable limit specified under M7(R1).
2. Include a suitable method for known PGIs among the release tests for each starting material, intermediate, or reagent with an acceptance criterion at or below the acceptable limit specified under M7(R1).
3. Include a suitable method for known PGIs among the release tests for each starting material, intermediate, or reagent with an acceptance criterion above the acceptable limit specified under M7(R1), coupled with a demonstrated understanding of downstream fate and purging and associated process controls in lab studies assure that no further testing is needed to confirm that the impurity is consistently below the M7(R1) limit by at least 30%.
4. Understand process parameters and impact on residual impurity levels (including fate and purge knowledge) with sufficient confidence that the level of the impurity in the drug substance will be below the acceptable limit such that no analytical testing is recommended for this impurity. (i.e., the impurity does not need to be listed on any specification).

Pharmaceutical companies tend to favor approaches 3 and 4 since:

• They are required under draft ICH Q11 guideline requirements to identify Critical Quality Attributes (CQAs) and apply Quality by Design (QbD) principles during API development anyway
• Sufficiently sensitive analytical methods may require the purchase of new instrumentation (and training) that may not have other applications in a typical QC laboratory.
• Highly sensitive analytical methods, if not sufficiently specific or thoroughly developed, may be more prone to error, thereby delaying batch releases, investigations, and completion of associated regulatory documentation.

Contact us to learn how we can help you meet both your ICH Q11 and ICH M7(R1) needs.

Yes.

If specific carcinogenicity bioassay data exists on a given substance, a toxicology expert may be able to create a compelling argument for specification limits higher than the default limits listed in ICH M7(R1).  Several examples, in fact, are described in the addendum recently added to ICH M7(R1).  However, for most API-related impurities or degradants this approach is not practical because:

1. published carcinogenicity bioassay data does not exist,
2. the costs involved in generating rodent carcinogenicity bioassay data are high,
3. a rodent carcinogenicity bioassay takes over two years to complete,
4. generation of an adequate supply of the impurity for the study adds further cost and time

For the vast majority of cases, it makes more sense to develop controls and demonstrate purging of the impurity to a level well below the specific threshold applicable under ICH M7(R1).

There is enormous pressure placed on process development scientists and engineers to deliver process technology on a “just-in-time” basis to prepare needed Phase II and Phase III clinical supplies.  Large pharmaceutical companies frequently have sufficient capacity of trained chemists and analysts to allow potential genotoxic impurity issues to be addressed concurrent with process development, while companies with more limited resources generally try to address them after processes have already been defined and demonstrated.  The problem in the latter case is that  modifications made to the process are severely limited once large scale trials are completed to supply Phase III trials and registrational stability studies.  Having an experienced third party evaluate your evolving API process for PGI issues concurrent with process development offers some major advantages:

1. Timely identification of opportunities to rectify future PGI issues before final scaled demonstration, after which time the API process becomes locked into place prior to CMC submission,
2. Independent studies designed and carried out by an impartial, experienced third party provide added credibility for the client when presenting its PGI control strategies and data in regulatory submissions.
3. The evaluation of a process for PGI process can be coupled with a somewhat broader evaluation of critical quality attributes and quality by design aspects, which helps to bolster the process.