Nitrosamine Impurities Testing: Risk Assessment, AI Limits and Methods

Nitrosamine impurities testing is the confirmatory step in a three-part control strategy: assess the risk that N-nitrosamines can form in or contaminate your API or drug product, test at-risk batches with a sensitive, validated method (usually LC-MS/MS, LC-HRMS or GC-MS/MS), and change the process or set a specification if levels are significant. The limits are extremely low, typically tens to hundreds of nanograms per day, so the analytical method has to be designed around the acceptable intake (AI) and the maximum daily dose from the start. This article covers how risk assessment works, how AI limits are set (including the Carcinogenic Potency Categorisation Approach, CPCA), how to convert them into ppm specifications, and what to look for in a testing method.

Why nitrosamines are treated differently

ICH M7(R2) (Step 4, 3 April 2023) controls mutagenic impurities using a threshold of toxicological concern (TTC) of 1.5 µg/day, which corresponds to a theoretical 1 in 100,000 excess lifetime cancer risk. However, M7 places N-nitroso compounds in the "cohort of concern", alongside aflatoxin-like and alkyl-azoxy compounds: high-potency mutagenic carcinogens for which acceptable intakes would likely be significantly lower than the TTC. That is why nitrosamines have their own AI limits rather than the default 1.5 µg/day.

For context, FDA stated in 2019 that the acceptable daily intake limit for NDMA in the US is 96 ng/day, and FDA's current guidance recommends 26.5 ng/day as the AI where no other limit can be determined.

Small-molecule nitrosamines vs NDSRIs

FDA's guidance Control of Nitrosamine Impurities in Human Drugs (Revision 2, September 2024) describes two classes:

• Small-molecule nitrosamines do not share structural similarity with the API. Examples include NDMA, NDEA, NMBA, NDIPA, NEIPA and NDBA. They typically arise from reagents, solvents, recovered materials or cross-contamination.
• Nitrosamine drug substance-related impurities (NDSRIs) share structural similarity with the API, usually because a secondary or tertiary amine in the API (or an API fragment) is nitrosated. They are unique to each molecule, which is why reference standards and methods often have to be developed specifically.

Step 1: risk assessment and root causes

A credible risk assessment looks for the combination of a vulnerable amine and a nitrosating agent anywhere in the life cycle of the drug substance and drug product. FDA's guidance lists root causes that should be checked, including:

• Nitrite salts or other nitrosating agents reacting with secondary, tertiary or quaternary amines, often under acidic conditions.
• Amide solvents such as DMF that can degrade to amines, and amine impurities in reagents.
• Contaminated vendor raw materials, for example nitrite in sodium azide or nitrosamine-contaminated solvents.
• Recovered solvents, reagents and catalysts, especially where recovery involves a nitrous acid quench.
• Quenching nitrous acid directly in the main reaction mixture, and poorly controlled pH, temperature or reaction sequence.
• Nitrite in excipients, potable water or packaging, and nitrogen oxides from forced-air drying or milling.

For drug products, the API structure is the starting point. Any secondary amine, or tertiary amine that can dealkylate, is a potential NDSRI precursor, and excipient nitrite levels combined with wet granulation or high-temperature processing can drive formation. The assessment should conclude with a clear list of the specific nitrosamines considered plausible, which then defines the scope of testing.

Step 2: acceptable intake limits and the CPCA

Where compound-specific carcinogenicity data exist, regulators derive an AI directly or by read-across from a structurally similar surrogate. For most NDSRIs there are no such data, so FDA's guidance Recommended Acceptable Intake Limits for Nitrosamine Drug Substance-Related Impurities (August 2023) introduced the CPCA. The CPCA predicts potency from structure, based on the α-hydroxylation pathway of metabolic activation, and assigns each nitrosamine to one of five potency categories.

How the potency score works

The CPCA counts the α-hydrogens on each side of the N–N=O group, written as a pair with the lower number first. Combinations such as 0,0, 0,1 and 1,1, or a tertiary α-carbon, lack or strongly disfavour activation and place the compound in category 5. Otherwise the potency score is the α-hydrogen score plus scores for deactivating and activating features; the higher the total score, the lower the predicted potency and the higher the category number. FDA maintains a web page of recommended AI limits for specific NDSRIs, listed by the API that could form them, which is updated regularly and should be checked before you set a limit.

Worked example: converting AI to ppm

FDA's guidance gives the conversion as ppm = AI (ng/day) / maximum daily dose (mg).

• Category 1 NDSRI, AI 26.5 ng/day, maximum daily dose 500 mg: 26.5 / 500 = 0.053 ppm (53 ppb). Ten per cent of that is about 0.005 ppm.
• Category 3 NDSRI, AI 400 ng/day, maximum daily dose 200 mg: 400 / 200 = 2.0 ppm. Ten per cent is 0.2 ppm.

The same structural class can therefore demand methods that differ in sensitivity by more than an order of magnitude, depending on dose and category.

More than one nitrosamine

FDA recommends that the total nitrosamine level should not exceed the AI of the most potent nitrosamine present. Where individual AIs differ greatly, FDA's guidance describes a more flexible approach that keeps the combined exposure within the 1 in 100,000 lifetime cancer risk outlined in ICH M7(R2).

Step 3: specifications and timelines

FDA's Revision 2 guidance links the control decision to confirmatory results:

• At or below 10% of the AI: no specification is needed, provided the root cause is understood and process controls are validated.
• Above 10% but within the AI: a control should be included in release and stability specifications.
• Above the AI: changes to formulation, process or packaging are warranted.

FDA recommended that NDSRI confirmatory testing and any resulting submissions be completed by 1 August 2025. In the EU, EMA expects marketing authorisation holders to implement corrective and preventive actions within three years of an acceptable intake being established and published.

Choosing and validating a test method

FDA's guidance asks for sensitive, validated methods with detection and quantitation limits commensurate with the level at which the impurity must be controlled, and notes that a method validated for one product or matrix must be validated again before use with another.

LC-MS/MS and LC-HRMS

Liquid chromatography with tandem or high-resolution mass spectrometry is the default for most NDSRIs and for thermally labile matrices. FDA's own laboratory developed an LC-HRMS platform covering NDMA, NDEA, NMBA, NMPA, NEIPA, NDIPA, NDBA and NDPA across ARBs, ranitidine and metformin, reporting quantitation limits as low as 0.005 ppm. ESI suits most polar NDSRIs, while APCI is often better for small, volatile nitrosamines. Read more on LC-MS analysis.

GC-MS and GC-MS/MS

Gas chromatography with mass spectrometry, often with headspace or direct liquid injection, works well for volatile small-molecule nitrosamines such as NDMA and NDEA in thermally stable matrices; see GC-MS testing. The key risk is artefact formation. In October 2019 FDA warned that a high-temperature method generated very high levels of NDMA from ranitidine because of the test procedure itself. Any GC method should be challenged for in-situ formation before results are relied on. Our comparison of LC-MS vs GC-MS covers the trade-offs in more detail.

Validation essentials

• Specificity: demonstrate no interference from the API, excipients or co-eluting isobaric species at the nitrosamine transition or accurate mass.
• LOQ: set at or below the level needed for the control decision, ideally no higher than 10% of the ppm limit if you intend to justify omitting routine testing.
• Accuracy: spike recovery in the actual product matrix at the LOQ and at the limit, using a characterised nitrosamine reference standard.
• Artefact control: run spiked and unspiked samples with and without a nitrosation inhibitor (for example sulfamic acid or ascorbic acid) to show the sample preparation does not create nitrosamines.
• Isotope-labelled internal standards: where available, they compensate for matrix suppression in MS and improve precision at trace levels.

For NDSRIs, the reference standard is often the bottleneck: it frequently has to be synthesised and characterised before validation can start. Our impurity reference standards guide explains what its CoA should contain.

How SDS Labs can help

SDS Labs supplies nitrosamine impurity reference standards from its catalogue and offers custom synthesis of impurities and labelled compounds where a specific NDSRI is not in stock. In-house LC-MS (ESI, APCI and APPI), GC-MS (EI and CI) and 400 MHz NMR support method development, validation and structural confirmation, with results generated in line with 21 CFR Part 11 and cGMP requirements under ISO/IEC 17025:2017.

To discuss a nitrosamine risk assessment outcome, a standard or a testing requirement, use the contact page, call +91 90295 20144 or email info@sdslabs.in.

Frequently asked questions

References

• FDA: Control of Nitrosamine Impurities in Human Drugs (Revision 2)
• FDA: Recommended Acceptable Intake Limits for Nitrosamine Drug Substance-Related Impurities
• FDA: CDER Nitrosamine Impurity Acceptable Intake Limits
• FDA: Determining Recommended Acceptable Intake Limits for N-nitrosamine Impurities in Pharmaceuticals
• ICH M7(R2): Assessment and Control of DNA Reactive (Mutagenic) Impurities
• EMA: Nitrosamine impurities guidance for marketing authorisation holders
• FDA: Updates and Press Announcements on NDMA in Zantac (ranitidine)