EMA Nitrosamine Article 5(3)

In 2018, the discovery of N‑nitrosodimethylamine in valsartan triggered large-scale recalls and exposed systemic vulnerabilities in impurity control. To set uniform scientific expectations across the European Union, the European Medicines Agency initiated an Article 5(3) referral under Regulation (EC) No 726/2004. This mechanism allows the Committee for Medicinal Products for Human Use (CHMP) to issue a scientific opinion on complex questions affecting the evaluation of medicines.

Adopted on 26 September 2019, the CHMP opinion established a structured, Union‑wide program for marketing authorisation holders of chemical medicinal products. The program requires three sequenced actions: a risk evaluation of potential nitrosamine presence or formation, confirmatory analytical testing where risks are identified, and implementation of controls with corresponding dossier updates. The framework has been refined through rolling updates to EMA’s Questions and Answers, which capture evolving toxicology, structure–activity read‑across for drug‑substance‑related nitrosamines, and practical control expectations.

Operationally, the referral aligns with ICH quality principles and EU GMP, requiring evidence‑based decision making, proportionality, and lifecycle management. It is expressly designed to integrate with a product’s overarching control narrative, drawing on risk management discipline, process knowledge, and supplier intelligence. Embedding these expectations coherently within the product’s overall control concept is essential; documentation should clearly connect toxicological rationales, analytical capabilities, and process controls within the product’s established control strategy and impurity specifications.

Teams should design their programs so that scientific reasoning is traceable, auditable, and durable. A coherent dossier narrative, supported by contemporaneous records and justified testing choices, will stand up to regulatory scrutiny and inspections. Consistency with ICH and EU GMP tenets is not optional; it is the backbone of the approach envisioned by the referral.

The referral applies to human medicinal products with chemically synthesised active substances, regardless of authorisation route—centralised, national, mutual recognition, or decentralised. It obliges marketing authorisation holders to evaluate both the active substance and the finished product, and to consider all plausible pathways by which nitrosamines could form or be introduced. This includes assessing upstream precursors and nitrosating conditions that may occur within the supply chain, manufacturing, packaging, and storage.

Although the initial mandate targeted non‑biological products, the EMA has clarified that certain biologicals may warrant risk review when process conditions, raw materials, or excipients could plausibly enable nitrosamine formation. Because scientific understanding continues to evolve, companies must check the latest EMA Q&A before finalising scope decisions, and document rationale when choosing to include or exclude product classes, strengths, or presentations from testing.

Effective implementation demands cross‑functional orchestration. Chemistry and manufacturing groups should feed structured risk inputs to regulatory teams, while quality units maintain the governance of evidence and changes. Supplier data—especially nitrite levels in excipients and amine‑bearing reagents—are critical, and must be curated under robust document control with traceability to certificates, trends, and capability statements. Centralising these judgments in an auditable quality risk register supports consistent outcomes and defensible decisions.

The referral’s expectations also extend to contractors and critical suppliers. Marketing authorisation holders remain responsible for ensuring their network demonstrates suitable knowledge of nitrosating conditions, control of inbound variability, and prompt escalation of deviations relevant to nitrosamines. Where contractual controls are thin, strengthen oversight through qualified status, data transparency, and regular risk dialogues anchored in structured supplier risk management.

The program sequences effort to achieve speed without sacrificing science. Step 1 screens each product using a structured risk evaluation focused on reasonably foreseeable formation or contamination. Only where a plausible risk is identified does Step 2 proceed with confirmatory analytical testing targeted to the nitrosamines predicted by process chemistry or historically associated with the route, reagents, or excipient set. Step 3 then locks in enduring controls and dossier changes, tuned to patient exposure and the EMA’s Acceptable Intakes.

Timelines have been adjusted several times to accommodate method development complexity and the workload across portfolios. Authorities still expect diligent, documented progress and prioritisation commensurate with risk. Where testing reveals levels above the applicable Acceptable Intake, companies must promptly mitigate risk and engage regulators, including potential market actions consistent with EU pharmacovigilance and defect reporting obligations.

The success of the sequence turns on articulation. Step 1 must produce a reasoned hypothesis set. Step 2 must test those hypotheses with sensitivity and selectivity at or below health‑based limits. Step 3 must translate learnings into robust specifications, in‑process controls, and ongoing monitoring that reflect realistic worst‑case formation over shelf life and use.

Start by mapping each unit operation for the active substance and finished product, from raw material receipt through packaging and storage. Catalogue all potential amine sources, including secondary and tertiary amines and quaternary ammonium precursors, as well as nitrosating species like nitrite under acidic conditions. Inventory reagents, catalysts, recycled solvents, and process aids that can introduce either amines or nitrite as impurities. Confirm actual usage conditions with process historians and batch records rather than relying on development summaries.

Quantify conditions that drive nitrosation: pH, time–temperature profiles, and concentration. Evaluate steps such as drying, coating, and hold times, and consider storage scenarios where humidity or headspace oxides could influence formation over shelf life. Examine excipients with variable nitrite content using supplier data and targeted testing. Integrate cleaning validation and shared‑equipment assessments to address cross‑contamination risk, reinforcing practices for spillage and cross‑contamination control.

Document the logic behind testing decisions. When foregoing testing, maintain a clear, auditable rationale supported by supplier confirmations, development studies, purge calculations, and trend data. Use controlled repositories and master data discipline so that reagent identities, grades, and impurity profiles remain traceable across changes, anchored by robust material master data and targeted raw material sampling plans. Where feasible, pilot preventive process analytical technology or in‑process checks to reduce formation potential early in the flow.

Confirmatory testing should reflect the specific chemistry outlined in Step 1. Select analytes based on likely formation pathways, known associations with your synthetic route, and realistic excipient contributions. Methods typically use LC–MS/MS or GC–MS with nitrosamine‑specific chromatographic selectivity and mass transitions capable of distinguishing close analogues. Ensure sampling plans capture within‑batch and across‑batch variability for both API and finished product.

Validation must demonstrate selectivity, sensitivity, and robustness in relevant matrices. Limits of quantification should meet or undercut the applicable Acceptable Intake when normalised to the maximum daily dose. Assess matrix effects, ion suppression, recovery, and carryover. Employ appropriate internal standards and system suitability checks consistent with a defensible stability‑indicating method, strengthened by rigorous risk‑based validation and laboratory governance.

Treat out‑of‑trend or unexpected positives as signals rather than anomalies. Investigate specificity (e.g., co‑eluting amines), check blanks, and evaluate potential in‑situ formation during sample preparation. Lock versioned methods, raw data, and review workflows under controlled document control and laboratory QA practices, with redundancy in data integrity safeguards and system suitability test criteria.

Once testing has characterised actual or potential nitrosamine profiles, convert findings into specifications, in‑process controls, and process modifications that reliably prevent formation or keep levels at or below the AI. Choose control locations that are closest to the source of risk, and avoid relying solely on end‑product testing where upstream prevention is demonstrably superior. Ensure analytical capability and sampling plans are commensurate with formation kinetics over shelf life.

Update the dossier with clear toxicological rationale, method summaries, control points, and stability commitments. Align impurity sections with ICH Q3A on impurities in drug substance and ICH Q3B on impurities in drug product, keeping the full control narrative coherent with the product’s control strategy. Where appropriate, position in‑process measures under in‑process controls (IPC) and anchor routine monitoring in a robust stability program.

If results exceed AI, act immediately. Implement risk‑reducing measures, assess patient exposure, and contact authorities to agree on mitigations and market actions if needed. Thereafter, manage lifecycle changes through your quality system and timely variations, ensuring consistency between manufacturing records, release decisions, and dossier content supported by a mature qms.

Acceptable Intakes (AIs) for nitrosamines are health‑based exposure limits that correspond to a theoretical excess lifetime cancer risk of 1 in 100,000 under established ICH principles. EMA maintains compound‑specific AIs and guidance on setting limits when only partial toxicology data exist. Where data are limited, default daily intakes derived from conservative carcinogenic potency assumptions may be applied until compound‑specific evaluations are available.

For nitrosamine drug‑substance‑related impurities, EMA applies a read‑across approach that estimates potency from closely related structural analogues and considers features relevant to metabolic activation. These decisions are periodically refined in the Q&A as new structure–activity information emerges. When multiple nitrosamines may co‑occur, the sum of their contributions is generally compared with an overall cap unless compound‑specific toxicology justifies an alternative approach.

Translating AIs into specifications requires normalising to the maximum daily dose across strengths, dosage forms, and likely use patterns. Limits should be paired with validated LOQs, sampling plans, and stability commitments that reflect realistic formation potential over shelf life. Maintain traceable calculations, rationales, and change histories under disciplined quality assurance process control, and use analytics to trend results against control limits and to detect early signals of drift.

The referral’s science dovetails with ICH quality guidelines and EU GMP. Quality risk management principles should guide prioritisation, decision making, and lifecycle updates, with records maintained for inspection. Align impurity narratives with ICH Q3A and Q3B, and ensure manufacturing and control choices are consistent with EU GMP expectations for knowledge management, change control, and deviation handling. Where APIs are made under contract, confirm that GMP for APIs is reflected in technical agreements and operational oversight.

Real‑world non‑conformances have often stemmed from gaps in excipient nitrite understanding, insufficient attention to equipment carryover, or testing the wrong nitrosamines relative to process chemistry. Others arise when dose changes or new strengths are introduced without recalculating AIs and reassessing method sensitivity. These missteps are avoidable with disciplined planning, change impact assessment, and strong supplier engagement.

Strengthen interfaces with your suppliers through qualified status, transparent data flows, and clear responsibilities for escalation and change notification. Build inspection‑readiness with version‑controlled dossiers, contemporaneous evidence, and accessible governance artifacts. Use digital tools for audit readiness, issue management via structured deviations, and up‑to‑date lists of qualified sources through an approved supplier list and supplier portal.

• Do not generalise across products without chemistry‑based justification; each route and formulation can change nitrosation risk.
• Avoid relying entirely on end‑product testing where upstream prevention can demonstrably reduce formation risk.
• Do not fix limits once and forget them; re‑check EMA’s Q&A when strengths, doses, or excipients change.
• Do not overlook cleaning validation and shared equipment risks; carryover can confound results.
• Avoid testing for the wrong nitrosamines; target analytes predicted by your actual process and excipient set.
• Do not under‑document negatives; record evidence and rationale when choosing not to test.

Inspectors will look for coherent, end‑to‑end governance of nitrosamine risks. Expect deep dives into how you framed Step 1 hypotheses, how Step 2 methods achieve sufficient selectivity and sensitivity, and how Step 3 controls are anchored in prevention. Evidence should connect supplier capability, process knowledge, and analytical performance to patient‑focused limits. Records must be contemporaneous, versioned, and logically cross‑referenced to batch decisions and submissions.

Maintain the nitrosamine program within normal quality operations rather than treating it as an exceptional project. Embed recalculation triggers into change control so that dose, strength, route, or excipient changes automatically initiate AI reviews and method verifications. Tie deviations, out‑of‑trends, and supplier notifications into a central signal‑management process that can escalate and initiate targeted studies as needed. Use notifications to route time‑critical events to accountable owners.

Lifecycle resilience depends on disciplined knowledge management. Keep your justification files and variation histories under robust document control, integrate with your stability program, and preserve traceability from risk registers to manufacturing instructions. Where sterile or highly controlled environments are used, ensure relevant GMP annexes and process‑specific requirements remain harmonised with nitrosamine controls and testing cadence.

V5 integrates nitrosamine obligations into routine operations. Author teams can build a single, queryable risk register mapped to unit operations, suppliers, and materials, then link it to analytical methods, specifications, and stability commitments. Evidence is versioned, time‑stamped, and cross‑referenced to batches, ensuring that Step 1 reasoning, Step 2 data, and Step 3 controls stay coherent throughout the lifecycle and during inspections.

Laboratory and quality modules maintain validated methods, LOQs, and results under controlled review. Automated routing captures out‑of‑trends and exceedances, driving immediate containment and communication. Supplier and material capabilities, including nitrite trends in excipients and amine impurity statements, are managed via an approved supplier list and a collaborative supplier portal, strengthening evidence for supply chain risk decisions.

Change control and dossier governance are synchronised so that recalculated AIs, revised specifications, and updated monitoring plans move together. Teams leverage qms, lab-qc, traceability, and audit readiness to embed nitrosamine controls into daily practice, while structured deviations, analytics, and document control provide transparency and rapid, defensible actions across markets.