Usage Examples
- The M7 assessment classified three process impurities as Class 2, so each needs a specification at or below the acceptable intake.
- Because dosing runs under one month, the M7 acceptable intake is 120 µg/day, not the 1.5 µg/day lifetime figure.
- Their M7 package ran a single (Q)SAR model, so the Class 5 calls will not survive review.
What is ICH M7?
ICH M7 is the harmonised guideline governing assessment and control of DNA reactive (mutagenic) impurities in pharmaceuticals, setting acceptable intake limits that ICH Q3A and Q3B leave undefined for genotoxic impurities.
ICH M7 exists because ICH Q3A(R2) and Q3B(R2) qualify impurities on general toxicity grounds and give limited guidance for those that are DNA reactive. A mutagen can cause cancer at levels far below any normal qualification threshold, so ICH M7 replaces percentage-of-dose logic with a cancer-risk calculation: how much of this impurity can a patient take, and for how long.
ICH M7 covers new drug substances and new drug products through clinical development and marketing applications, plus post-approval changes that create new impurities, create new degradation products, or shift the acceptable cancer risk level. ICH M7 does not cover biologics, peptides, oligonucleotides, radiopharmaceuticals, fermentation products, herbals, or advanced cancer indications inside the scope of ICH S9, and it is not applied retrospectively to products marketed before adoption.
ICH M7 is applied as a three-stage pipeline: classify every actual and potential impurity into Class 1 through 5, derive an acceptable intake for Classes 1, 2, and 3, then select one of four control options. Option 1 puts a test on the drug substance specification; Option 4 relies on fate-and-purge understanding with no analytical test at all. ICH M7 review questions almost always land on which option the data actually supports.
Not to be confused with
- ICH Q3A(R2) and Q3B(R2)
- Q3A and Q3B qualify and control the majority of impurities using thresholds expressed as a percentage of daily dose. ICH M7 governs the subset that is DNA reactive, where the limit is a fixed microgram-per-day intake derived from cancer risk, not a percentage.
- The ICH M7 Addendum
- the Addendum is a separate document holding compound-specific acceptable intakes for named mutagens. The M7(R2) revision split them apart in 2021, so citing "M7" without saying whether you mean the Guideline or the Addendum is ambiguous.
- Cohort of concern
- this is a carve-out inside ICH M7, not a parallel regime. Aflatoxin-like, N-nitroso, and alkyl-azoxy structures stay in M7 scope, but the 1.5 µg/day TTC does not apply to them and acceptable intakes are justified case by case at substantially lower levels.
- ICH S9
- S9 covers nonclinical evaluation for advanced cancer indications, and drug substances and products within its scope are explicitly excluded from ICH M7. Their impurities are controlled at levels acceptable for non-mutagenic impurities instead.
The obligations run in sequence; each one presumes the previous is documented.
What you must do
- 1Determine applicability before assessing anything, applying ICH M7 to new drug substances and products in development and marketing applications, and to post-approval submissions only where the change creates new impurities, new degradation products, or a materially different acceptable cancer risk levelICH M7(R2) Section 2
- 2Classify each actual and potential impurity as Class 1 through 5 from carcinogenicity and bacterial mutagenicity data, and where no such data exist, from two complementary (Q)SAR methodologies, one expert rule-based and one statisticalICH M7(R2) Section 6
- 3Control Class 1, 2, and 3 impurities at or below the acceptable intake, defaulting to the 1.5 µg/day TTC for treatment longer than 10 years where no carcinogenicity data existICH M7(R2) Section 7.1
- 4Select and justify one of the four control options, from a test on the drug substance specification through to process understanding with no analytical testing for that impurityICH M7(R2) Section 8.1
- 5File the classification and its rationale, the in silico (Q)SAR systems used, bacterial mutagenicity study reports, and the justification for the proposed specification and control approach in the marketing applicationICH M7(R2) Section 9.2
Common mistakes
Running one (Q)SAR model and calling the impurity clean
M7 requires two complementary methodologies, one expert rule-based and one statistical. A single-model negative does not support a Class 5 conclusion. The classification gets rejected and the fix is bacterial mutagenicity testing on isolated material, which is the slowest and most expensive point to discover the gap.
Treating 1.5 µg/day as the universal limit
1.5 µg/day is the lifetime figure. Table 2 permits 120 µg/day for treatment of one month or less, 20 µg/day for more than one to 12 months, and 10 µg/day for more than one to 10 years. Defaulting to the lifetime number forces analytical methods far more sensitive than the risk requires, and creates specification failures on a limit that was never applicable.
Claiming Option 4 without the process data to defend it
Option 4 removes the impurity from every specification, so it carries the heaviest justification burden. Where scientific principles alone are not sufficient, analytical data supporting the fate and purge argument are expected, including pilot or initial commercial scale confirmation when the small scale model may not represent commercial scale.
When This Matters
- The M7 assessment classified three process impurities as Class 2, so each needs a specification at or below the acceptable intake.
- Because dosing runs under one month, the M7 acceptable intake is 120 µg/day, not the 1.5 µg/day lifetime figure.
- Their M7 package ran a single (Q)SAR model, so the Class 5 calls will not survive review.
Frequently Asked Questions
The TTC-based acceptable intake in ICH M7 is 1.5 µg per person per day, associated with a theoretical excess cancer risk of less than 1 in 100,000 over a lifetime. It applies as the default limit for Class 2 and Class 3 impurities in pharmaceuticals taken for more than 10 years.
Related Use Cases
- Pharma Use Cases
Cut NDA and sNDA prep time by 60% with AI-assisted drafting and automated readiness checks
- Biotech Use Cases
Compress IND prep from 8-12 weeks to under 3 weeks with AI-assisted drafting and validation
- CMC Workflows
Track ICH quality guidelines automatically and get alerts when changes impact your products
- Quality/QA Workflows
Track GxP regulation changes and enforcement trends
Related Regulatory Intelligence
Related Actions
Sources & References
- ICH Harmonised Guideline M7(R2) - Assessment and Control of DNA Reactive (Mutagenic) Impurities in Pharmaceuticals to Limit Potential Carcinogenic Risk (Step 4, 3 April 2023)
- EMA - ICH M7 Assessment and control of DNA reactive (mutagenic) impurities in pharmaceuticals to limit potential carcinogenic risk
- ICH Multidisciplinary Guidelines (M series)
