A Pragmatic Approach to Controlling Potential Genotoxic Impurities for Phase I

Qualifying and controlling impurities is a key activity during the development of new drugs. ICH Q3A and Q3B provide the appropriate guidance for the drug developers as to what needs to be met at the finish line, when the product is ready for the market. One very important consideration given in these guidance documents is that impurities at levels below 0.05 % “fly below the radar cover” and are not considered unless they are of special concern.

Genotoxic impurities (GTI’s) are one group that requires particular attention since such substances may cause genetic damage after a single exposure at low levels. Genotoxicity may be exerted either by direct interaction with DNA, or by interaction with proteins that regulate the structure or function of DNA. While DNA-reactive substances are generally considered to be harmful at any exposure level, those substances that affect DNA by indirect mechanisms may have a safe threshold of exposure.

In 2006 EMEA issued new guidelines on the limits of genotoxic impurities in new drugs.  The maximum allowed exposure of genotoxic impurities, the Threshold of Toxicological Concern (TTC), is set at such a level (1.5 μg/day) that the excess cancer risk over a lifetime is minimal (1 in 10⁵ to 1 in 10⁶). Lower limits are required for aflatoxins, N-nitroso- and azoxy- compounds.

The pharmaceutical industry published a response in 2006 to the EMEA’s guideline with a proposal on how to handle the threshold levels of Potential GTI’s (PGI) during the development phase of a new drug. To summarize, the industry suggested a staged approach taking into account the duration of the exposure (length of study) and the daily dose. In a short study, such as a phase I study, the daily accepted exposure to PGI’s could be higher than in a study that extends over a longer period such as a phase III study, or for a lifelong treatment.

In a 2008 Q&A response EMEA accepted the use of a staged approach during the drug development phase.  However, EMEA set the maximum daily exposure at half of that suggested by the industry for all stages of development (see Table 1). The same levels are being proposed in a draft guidance document by the FDA, with one notable exception: An acceptable daily intake (ADI) of 120µg is acceptable for administration up to 14 days.

Table 1. Acceptable daily intake (ADI) for genotoxic impurities 

	Duration of exposure
	Single dose	1 month	3 month	6 month	12 month
ADI	120 μg	60 μg	20 μg	10 μg	5 μg
Maximum concentration of PGI if daily dose < 100 mg	1200 ppm 0.12 %	600 ppm 0.06 %	200 ppm 0.02 %	100 ppm 0.01 %	50 ppm 0.005 %

Key recommendations for Phase I

1.Identify any GTI’s and appoint PGI’s. Focus on later parts of synthesis.

Chemicals that are added or generated during the later phases of API synthesis are more likely to end up in the API than chemicals added or generated during the earlier stages. A reasonable approach for phase I is to limit the search for PGI’s to the last 3-4 chemical steps and to consider only chemicals added and desired products (intermediates).However, impurities that may form due to side-reactions caused by the presence of sulfonic acids (such as mesylate reacting with alcohols) should be considered.

Identify GTI’s from literature data and appoint those impurities that contain alerting functional groups² as PGI’s. To identify PGI’s in the synthetic process search for alerting functional groups (see e.g. ref 2), use computer-based structure activity analysis (SAR) for genotoxicity such as DEREK, MCASE and MDL-QSAR.

2.Purify API until TTC levels are reached.

Most often, PGI’s are very reactive chemical species such as alkylators, and as such can be expected to be easily removed during extractive work-up and/or crystallization. This normally means that the contamination levels allowed in a batch for phase I can be attained with reasonable effort.

3.Control and demonstrate acceptable levels (ADI) in API.

EMEA has acknowledged that wherever PGI’s are controlled at the TTC level, no genotoxic testing is obligatory.³ At the 0.06 % level, analytical techniques such as HPLC-UV or HPLC-MS are normally sufficiently sensitive to monitor the PGI’s with reasonable effort. Limit tests may be considered in difficult cases, although actual quantification is preferred whenever possible.  

An example

In Scheme 1, the final four chemical steps in the synthetic sequence of an API are given. The intermediate, “FIM-1”, is an alkylator and as such contains a typical structural alert functionality with genotoxic potential. “FIM-1” is therefore appointed as a PGI and controlled at a level with a maximum daily intake of 60 μg for a typical phase I study (see Table 1). In the penultimate step a known GTI, thiourea, is used. This reagent should be controlled at the same level as the PGI, in this case an intake of 60 μg/day ADI.

Finally, in the acetylation of FIM-3 affording intermediate FIM-2, the class 1 solvent benzene is used. Benzene is a confirmed human carcinogen, the risk levels for which have been calculated based on substance-specific cancer potency data. The maximum ADI has been calculated as 20 µg/day for lifetime exposure. Based on this data, the ADI would be 800 µg per day for 1 month for a risk level of 1 in 10⁶. However, it would be prudent to reduce the level as much as reasonably possible considering that it is recommended in ICH Q3C that Class 1 solvents, including benzene, should be avoided as often as possible. We suggest an ADI for benzene of 60 μg/day for a short study with a duration of one month or less.

 

Figure 1. Four last chemical steps in the manufacture of an API. (FIM = final intermediate)

Conclusion: In this example with a study duration of 20 days and a daily dose of approximately 100 mg, the specification limit for all three of the above-mentioned process impurities is set at 0.06 % (w/w), a level attained with reasonable effort by the manufacturing process and controlled by HPLC-UV and GC-FID.  

What’s next? 

Experimental confirmation of genotoxicity is desirable at later stages of the drug development process. These studies will indicate whether or not there is reason for further concern. The recommended primary test to confirm that a PGI is genotoxic is Ames test. 

Furthermore, the TTC approach provides guidance for establishing limits for PGI’s where no cancer data is available. In cases where cancer data is available from animal or human studies, substance-specific ADI’s can be set. These may be higher than the ADI specified by the TTC. Higher levels may also be set when cancer data is available for closely related structures showing a lower potency than that on which the TTC is based, or when there is data supporting a threshold mechanism of action for the genotoxicity.  

FOOTNOTES:

1.  George Bolcsfoldi, PhD, consultant in genetic toxicology, is acknowledged for his input to this article (www.toxassist.se)

2.  CHMP: European Medicines Agency, June 2006.  Guideline on the Limits of Genotoxic Impurities 

3.  L. Muller et al Regulatory Toxicology and Pharmacology 2006, 44, 198-211

4.  CHMP: European Medicines Agency, June 2008.  Questions and answers on the CHMP Guideline on the Limits of Genotoxic Impurities

5.  US FDA Guidance for Industry, Genotoxic and Carcinogenic Impurities in Drug Substances and Products: Recommended Approaches, Draft, December 2008

6.  Pierson et al OPRD, 2009

7.  Reasonably anticipated to be a human carcinogen: Report on Carcinogens, Eleventh Edition

8.  ICH Q3C Impurities: Guideline for residual solvents, July 1997