Health Canada Guidance: Nitrosamine Impurities in Medications
Health Canada has updated its guidance on evaluating and managing N-nitrosamine impurities in human pharmaceutical, biological and radiopharmaceutical products. Originally published on 4 April 2022 and updated on 5 November 2026, the guidance outlines expectations for market authorization holders (MAHs), applicants and manufacturers regarding risk assessment, confirmatory testing, acceptable intake (AI) limits and lifecycle risk management.
1. Background and Scope
The guidance applies to human pharmaceutical products containing chemically synthesized and semi-synthetic active pharmaceutical ingredients (APIs), including prescription and non-prescription medicines. It also covers biological and radiopharmaceutical products, including vaccines, plasma proteins and cell-based fermentation products, as well as approved products not yet marketed or with dormant DIN status.
Cosmetics without a DIN, veterinary products, natural health products, antimicrobial agents and hard-surface disinfectants are currently excluded.
2. General Requirements and Responsibilities
Health Canada follows a three-step approach:
Step 1 – Risk Assessment: Identify potential sources and causes of nitrosamine formation or contamination.
Step 2 – Confirmatory Testing: Conduct testing where a potential risk is identified.
Step 3 – Market Authorization Changes: Implement required changes to specifications, manufacturing controls and risk mitigation measures.
The original deadlines have elapsed. MAHs are permitted up to three years for CAPA implementation for established AI limits, calculated from 1 August 2025 or the relevant AI publication date, as applicable. However, Health Canada expects impurity levels to be reduced below applicable limits as soon as possible.
MAHs remain responsible for product safety, efficacy and quality. They should collaborate with API, excipient and drug product manufacturers, maintain documented risk assessments and use qualified personnel. Supplier declarations alone cannot replace a comprehensive risk assessment.
Risk prioritisation should consider maximum daily dose, route and duration of administration, patient population, API toxicity, manufacturing conditions and emerging regulatory information. Foreign assessment results may be used when the Canadian and foreign products are identical in relevant composition, strength, manufacturing process and sites, supported by written justification.
The guidance also addresses Special Access Program products, third-party support for risk assessments, confirmatory testing exemptions where no risk is scientifically demonstrated, and expectations for new applications and products under development.
3. Regulatory Submissions and Market Actions
Nitrosamine-related changes should be submitted through the applicable Supplement, Notifiable Change or Post-DIN Change pathway using eCTD or non-eCTD format through the Common Electronic Submission Gateway.
Submissions should identify their risk-mitigation purpose and include relevant root cause investigation summaries and control strategy changes in the appropriate CTD sections.
Where nitrosamine levels exceed applicable AI limits, MAHs should assess patient risk and determine whether market action is necessary. Health Canada encourages prior engagement before initiating recalls to minimise potential medicine supply disruptions. Appropriate CAPA and control strategies should ensure impurity levels remain within limits throughout the API retest period or drug product shelf life.
4. Communications and International Collaboration
Health Canada shares information through stakeholder engagement and its dedicated nitrosamine webpage, including affected or recalled products and analytical testing results.
It also collaborates with international regulators, including authorities in Europe, the US, Japan, Switzerland, Singapore and Australia, and the WHO. The Nitrosamines International Strategic Group (NISG) and Nitrosamines International Technical Working Group (NITWG) support scientific discussions, information sharing and regulatory alignment.
5. Safety: Acceptable Intake Limits
Health Canada has established AI limits for several nitrosamine impurities, applicable to the maximum daily dose and all routes of administration.
AI limits may be established using compound-specific carcinogenicity data, enhanced Ames assay results, in vivo mutagenicity data, structure-activity relationship assessments and read-across approaches.
The Carcinogenic Potency Categorization Approach (CPCA) classifies nitrosamines into five potency categories based on molecular structural features, with corresponding AI limits ranging from 18 ng/day to 1,500 ng/day.
Where multiple nitrosamines are present, cumulative daily exposure should generally follow the most conservative applicable AI limit. The guidance also addresses products intended for advanced cancer indications and products containing APIs that are genotoxic at therapeutic concentrations.
An updated provision allows less-than-lifetime (LTL) AI adjustments for eligible products based on treatment duration, following ICH M7 principles:
Up to 1 month: 80 × AI
More than 1 to 12 months: 13.3 × AI
More than 1 to 10 years: 6.7 × AI
More than 10 years to lifetime: No adjustment
6. Quality: Risk Factors and Potential Root Causes
MAHs should assess potential nitrosamine formation and contamination throughout manufacturing, packaging and storage. Key risk factors include:
Reaction of secondary or tertiary amines with nitrosating agents.
Nitrite contamination in raw materials, reagents, excipients or water.
Inadequate removal of nitrosamines used as intermediates.
Contaminated recycled solvents, reagents and catalysts.
Cross-contamination in shared manufacturing facilities.
Packaging components that may introduce nitrosating agents or amines.
Manufacturing and storage conditions that promote nitrosamine formation.
All drug product components should be considered during risk assessments. The nitrosamines listed by Health Canada are not exhaustive, and product-specific risks should guide confirmatory testing.
7. Analytical Testing and Control Strategy
Confirmatory testing should use sensitive, quantitative and fully validated analytical methods at GMP-compliant facilities. The limit of quantitation (LOQ) should generally be equal to or below the applicable AI limit or relevant reporting threshold.
Testing should cover representative batches based on the identified risk. For new applications with an identified risk, at least six pilot-scale or three commercial-scale batches should generally be tested. For marketed products, testing frequency may be reduced when the root cause is understood and batch consistency is demonstrated.
Specifications should include appropriate individual and cumulative nitrosamine limits where required. Control strategies may involve API testing, process controls, intermediate specifications and drug product release and shelf-life testing. Stability testing should be considered where impurity levels may increase during storage.
The guidance also clarifies GMP expectations for analytical laboratories and when laboratories must be listed on the applicable Drug Establishment Licence (DEL) annex.
8. Appendices: Technical and Regulatory Support
The guidance includes five appendices supporting implementation:
Appendix 1 – Established AI Limits: Lists established acceptable intake limits for individual N-nitrosamine impurities.
Appendix 2 – Post-NOC Changes: Provides a decision-making approach for assessing nitrosamine risks associated with post-approval changes and submission requirements.
Appendix 3 – Enhanced Ames Assay: Describes enhanced bacterial reverse mutation test conditions for evaluating N-nitrosamines, including NDSRIs.
Appendix 4 – CPCA: Explains structural feature assessment and potency categorisation for determining AI limits, with practical examples.
Appendix 5 – Bioequivalence Studies: Covers reformulated products using nitrosamine-suppressing agents, including antioxidants and pH modifiers, and circumstances where BCS-based biowaivers or comparative bioavailability studies may be appropriate.
Reference
Comments