Articles
Medical device genotoxicity: what ISO 10993-3 and ISO 10993-33 require
A medical device rarely acts through a single molecule the way a drug does. Instead, the biological question is about everything the device might release into the body over its contact time — monomers, additives, processing residues, degradation products. Among the endpoints regulators care about most is genotoxicity: could any of those released substances damage DNA?
For medical devices, that question is answered under the ISO 10993 family — specifically ISO 10993-3 and the more detailed ISO 10993-33. This article explains what they require, how the testing differs from pharmaceutical genotoxicity, and how it fits into the wider biological evaluation demanded by the EU MDR and the FDA.
Key points
Part of a bigger evaluation
Two standards
You test the extract
A two-assay battery
Genotoxicity within the ISO 10993 evaluation
ISO 10993-1 sets the overall logic: biological evaluation is risk-based and driven by the nature and duration of body contact. A surface device in brief contact carries a different endpoint set from an implant in permanent contact. Genotoxicity becomes a required consideration for many devices — particularly those with prolonged or permanent contact, or contact with blood or internal tissues.
Within that framework, ISO 10993-3 (“Tests for genotoxicity, carcinogenicity and reproductive toxicity”) defines the principles, while ISO 10993-33 provides the up-to-date, detailed methodology — including how the assays map onto the corresponding OECD Test Guidelines. In practice, -33 is the operational reference for designing a compliant genotoxicity programme.
At a glance
ISO 10993-33
The detailed, OECD-aligned genotoxicity test guidance for medical devices
2 extracts
Polar and non-polar, to capture substances of differing solubility
≥2 assays
A gene-mutation test plus a chromosomal-damage test, covering both mechanisms
MDR + FDA
Both frameworks expect biological evaluation to ISO 10993 principles
Why device testing is different: it starts with the extract
The defining feature of device genotoxicity is that you rarely test the device material directly. Instead, following ISO 10993-12, you prepare extracts designed to simulate what the body would be exposed to — usually a polar extract (e.g. saline) and a non-polar extract (e.g. vegetable oil), to draw out both water- and lipid-soluble substances. Extraction conditions (temperature, time, surface-area-to-volume ratio) are chosen to be exaggerated but relevant.
Those extracts — not the bulk polymer or metal — are what the genotoxicity assays actually see. Getting extraction right is therefore as important as the assays themselves: an under-designed extraction can miss the very leachables that carry the hazard.
The standard battery
Device genotoxicity mirrors the OECD-based logic used for chemicals and drugs, typically combining:
- a bacterial reverse mutation test (Ames, OECD 471) for gene mutations; and
- an in vitro mammalian assay for chromosomal damage — the in vitro micronucleus test (OECD 487) or the in vitro chromosomal aberration test (OECD 473), covering clastogenic and (for the micronucleus test) aneugenic events.
This two-assay design covers the two principal mechanisms of genetic damage — point mutations and structural/numerical chromosome changes — with in vitro methods, in line with the 3Rs.
Fitting into the regulatory picture
Under the EU Medical Device Regulation (MDR) and for FDA submissions, genotoxicity data feed the overall biological safety section of the evaluation. Increasingly, that assessment is combined with chemical characterisation and extractables & leachables (E&L) data: identifying and quantifying released substances first, then using toxicological assessment (and targeted assays) to close any genotoxic-risk gaps. The two approaches are complementary — chemistry defines what is present, genotoxicity testing confirms whether it matters.
Getting it right the first time
- Classify the device by contact type and duration to confirm genotoxicity is required.
- Design the extraction (polar + non-polar) under ISO 10993-12 to match real exposure.
- Run the two-assay battery to ISO 10993-33 / OECD, under GLP.
- Integrate with chemical characterisation and E&L to build one coherent biological-safety argument.
Frequently asked questions
What does ISO 10993-3 cover?
ISO 10993-3 defines the principles for testing medical devices for genotoxicity, carcinogenicity and reproductive toxicity. The detailed, OECD-aligned genotoxicity methodology is given in ISO 10993-33.
Which genotoxicity tests are required for medical devices?
Typically a battery of two in vitro assays: a bacterial reverse mutation test (Ames, OECD 471) for gene mutations, plus an in vitro micronucleus test (OECD 487) or chromosomal aberration test (OECD 473) for chromosomal damage.
Do you test the device itself or an extract?
You test extracts. Under ISO 10993-12, polar and non-polar extracts are prepared to simulate what the body would be exposed to, and the genotoxicity assays are run on those extracts rather than on the raw material.
How does genotoxicity relate to extractables and leachables?
Chemical characterisation and E&L identify and quantify the substances a device releases; genotoxicity testing confirms whether those substances pose a DNA-damage risk. The two are combined into a single biological-safety assessment for the MDR or FDA.
Partner with GenEvolutioN
GenEvolutioN brings GLP-compliant, OECD-aligned genetic toxicology to medical-device developers — from extraction strategy to the full ISO 10993-33 battery, integrated with chemical characterisation where needed. We help you turn regulatory constraints into a clear, defensible biological-safety file.
See how we turn GLP and ISO standards into allies, or discuss your device programme with our team.
Have a regulatory toxicology project?
Our team supports you across all your in vitro studies — OECD methods, REACH dossiers, biocides, cosmetics.
Discuss your project
