Toxicological Risk Assessment (TRA) for Drug Impurities

Impurities in Drug Substances and Drug Products: Is Your Toxicological Risk Assessment Strategy Defensible?

By Jaiprakash Bhelonde

Finding an impurity is rarely the hardest part anymore. The harder question is what you do next.

Your analytical team may detect a new degradation product during stability testing. A manufacturing change may introduce an impurity that was not present in the original toxicology batches. An existing impurity may appear above the level previously qualified. Or an extractable or leachable may emerge from a container-closure system.

At that point, analytical data alone will not answer the regulatory question.

Is the impurity safe at the proposed level, and can you defend that conclusion scientifically?

That is where Toxicological Risk Assessment, or TRA, becomes critical.

A strong TRA connects the identity and level of an impurity with its toxicological profile, patient exposure and the applicable regulatory framework. It should give CMC, Quality, Toxicology and Regulatory Affairs teams a clear basis for deciding whether an impurity is adequately qualified, needs further control or requires additional investigation.

One impurity framework does not fit every situation

The first mistake is to treat every impurity as the same toxicological problem. It is not.

ICH Q3A(R2) addresses impurities in new drug substances, while ICH Q3B(R2) covers impurities and degradation products in new drug products. But several impurity classes require more specific frameworks.

Potential DNA-reactive mutagenic impurities fall under ICH M7(R2). Residual solvents require assessment under ICH Q3C. Elemental impurities fall under ICH Q3D. Extractables and leachables require their own risk-based assessment.

Nitrosamines add another layer.

So before asking, “What level is acceptable?”, you first need to ask: What type of impurity are we dealing with, and which scientific framework applies?

That decision influences the literature search, testing strategy, exposure calculation, control approach and final regulatory justification.

Non-mutagenic impurities are receiving renewed regulatory attention

One of the most relevant recent developments came from EMA in 2026.

In February 2026, EMA published its adopted Reflection Paper on the Qualification of Non-Mutagenic Impurities, with an updated version issued in March.

The paper addresses a common problem: what happens when your existing nonclinical or clinical data do not adequately qualify a newly identified non-mutagenic impurity, or a higher level of an impurity already known?

EMA specifically discusses situations where impurities arise after manufacturing-process changes, appear after safety studies have been completed, or need qualification at levels above those previously supported.  

The paper discusses approaches including read-across, metabolism information, Threshold of Toxicological Concern, in-vitro methods and other non-animal approaches where scientifically appropriate.

This raises the standard of scientific reasoning.

If you rely on read-across, why is the selected analogue appropriate? How closely do the structure, metabolism and toxicological characteristics match? What uncertainty remains?

A TRA must explain that logic clearly. A database search alone will not.

ICH M7(R2): computational toxicology still requires expert judgement

Potential mutagenic impurities require a different decision pathway.

ICH M7(R2), adopted by FDA as final guidance, sets out the framework for assessing and controlling DNA-reactive mutagenic impurities to limit potential carcinogenic risk. Where bacterial mutagenicity data are unavailable, the guideline supports structure-based assessment using two complementary (Q)SAR methodologies: one expert rule-based and other statistical-based. 
But software does not make the regulatory decision. You still need to interpret alerts, understand model applicability, review supporting information and document the scientific rationale behind the classification.

A prediction is an input. A defensible toxicological assessment is the output.

Nitrosamines continue to reshape impurity risk assessment

Nitrosamines remain one of the most closely watched impurity topics in pharmaceutical regulation.

FDA's revised Control of Nitrosamine Impurities in Human Drugs, published in September 2024, covers both small-molecule nitrosamines and nitrosamine drug substance-related impurities, or NDSRIs. It recommends that manufacturers assess the risk of nitrosamine formation, conduct confirmatory testing where appropriate and implement mitigation and control strategies when unacceptable levels occur.

EMA continues to update its own nitrosamine framework, including acceptable intake limits, the Carcinogenic Potency Categorisation Approach (CPCA), read-across and enhanced Ames testing. EMA's Appendix 1 on established acceptable intakes for N-nitrosamines was updated on 24 June 2026. 

The focus has also moved upstream.

EMA revised its Guideline on the Chemistry of Active Substances in March 2026 to add further recommendations on cohort-of-concern impurities, principally N-nitrosamines, alongside expanded guidance on starting materials, recovery and re-processing. For drug-substance manufacturers, that links toxicological risk directly to process knowledge.

Extractables and leachables are moving toward greater harmonisation

Another development worth watching is ICH Q3E.

The draft guideline for Extractables and Leachables proposes a risk-based framework for assessing and controlling leachable impurities arising from manufacturing systems, packaging and delivery-device components. It complements the existing ICH Q3 impurity guidelines and ICH M7.

The distinction matters: Q3E remains draft guidance and is not yet a final harmonised standard. Pharma teams should understand the direction of travel without treating a draft framework as an implemented requirement.

Where does a TRA become difficult?

Usually, not when the evidence is clear. The challenge comes when data are incomplete.

You may have limited compound-specific toxicology information. A read-across analogue may exist, but its relevance needs justification. Two computational models may need expert interpretation. An impurity may exceed an established qualification threshold. A new process may change the impurity profile after pivotal safety studies are complete.

Now chemistry, analytical science, toxicology and regulatory strategy need to work together.

Chemistry tells you where the impurity comes from.

Analytical science tells you how much is present.

Toxicology addresses the hazard and exposure.

Regulatory science determines how that evidence supports your specification, control strategy and submission.

So ask yourself: Are you producing a TRA to complete a document, or to support a regulatory decision?

What should a regulatory-ready TRA do?

A useful TRA should create a clear scientific trail.

It should define the impurity and exposure scenario, review relevant toxicological evidence, identify the applicable regulatory framework and explain the basis for the safety conclusion.

Where direct data are insufficient, the assessment may require scientifically justified read-across, (Q)SAR analysis or another appropriate methodology. The conclusion should answer the question that triggered the assessment:

Can the impurity be supported at the proposed level?

Can an acceptable daily intake be justified?

Is additional testing necessary?

Does the product or manufacturing control strategy need to change?

That is where TRA becomes useful to the wider regulatory program.

How Freyr supports Toxicological Risk Assessment

Freyr's current toxicological risk assessment (TRA services )cover multiple impurity categories across drug substances and drug products.

These include genotoxic impurity risk assessment in line with ICH M7; TRA of impurities in drug substances under ICH Q3A and drug products under ICH Q3B; residual solvents under ICH Q3C; elemental impurities under ICH Q3D; extractables and leachables from container-closure systems. 

Freyr's toxicologists use structured literature searches, read-across approaches and (Q)SAR assessment using validated, regulatory-accepted software in line with ICH M7 recommendations.

Freyr also supports TRA and derivation of Acceptable Daily Intake for out-of-specification impurities, including non-ICH Q3D elemental impurities, and can support the planning and design of nonclinical studies required for impurity qualification, including study placement, monitoring and review of study reports for regulatory submissions.

For extractables and leachables, Freyr provides risk-assessment support for pharmaceuticals and medical devices with reference to ISO 10993-17, ISO 10993-18 and ISO/TS 21726. Its published TRA offering includes review and approval by qualified toxicologists, with a team that includes professionals certified by the American Board of Toxicology and European Registered Toxicologists.

The objective is not simply to produce another toxicology report. It is to reach a conclusion your Regulatory teams can use- and one that can stand up to scientific scrutiny.

Next time, when you face a challenge dealing with impurities, the first question should not be:

“Who can perform the TRA?”

It should be:

“What scientific and regulatory decision does this TRA need to support?” 

Facing an impurity assessment challenge?

Talk to a Freyr Toxicologist to discuss your impurity profile, available toxicological evidence, and the most appropriate assessment approach for your drug substance or drug product.