Designing a biocompatibility testing strategy in 2026 is not about running every ISO 10993 test you can think of. It is about selecting the minimum evidence necessary to convince a regulator that your device is safe for its intended clinical use. Since the 2018 FDA guidance and the 2020 ISO 10993-1 revision, chemical characterization has moved from optional to central, and the biocompatibility testing strategy you file today looks materially different from one filed a decade ago. This guide explains how to design a biocompatibility testing strategy that is scientifically sound, cost-effective, and aligned with the 3Rs principle of reducing animal use.
What a modern biocompatibility testing strategy looks like
The ISO 10993-1 biological evaluation framework lists the endpoints a device may need to address — cytotoxicity, sensitisation, irritation, acute systemic toxicity, subchronic toxicity, genotoxicity, implantation, haemocompatibility, and more — based on the device category, tissue contact, and contact duration. A defensible biocompatibility testing strategy is not the maximal test matrix; it is the strategy that chooses each test because it genuinely adds evidence that cannot be obtained by another means.
The 2020 revision of ISO 10993-1 codified the risk management approach: start with device characterisation, conduct chemical characterisation per ISO 10993-18, perform a toxicological risk assessment per ISO 10993-17, and then decide which biological endpoints still need in vitro or in vivo testing. This order is not optional — it is the expected flow.
The four pillars of an effective biocompatibility testing strategy
Pillar 1 — Device characterisation
Before any testing is scoped, the device must be fully characterised: materials of construction, manufacturing processing, residuals from sterilisation, additives and colourants, configuration, patient-contact surface, and clinical use duration. Gaps in characterisation are the single biggest risk to a biocompatibility testing strategy because they leave reviewers asking questions that prior testing cannot answer.
Pillar 2 — Chemical characterization
ISO 10993-18 extractables studies provide the chemical inventory. The extractables profile, interpreted against the Analytical Evaluation Threshold (AET), identifies substances that require toxicological evaluation. Our E&L studies design guide walks through the analytical principles.
Pillar 3 — Toxicological risk assessment
Every extractable above AET enters the TRA per ISO 10993-17. Tolerable Intake is derived, exposure is estimated, and the Margin of Safety is calculated. Where MoS is robust, chemistry alone can replace animal testing for that endpoint; where MoS is narrow, targeted biological testing is triggered. See our ISO 10993-17 TRA guide for the full workflow.
Pillar 4 — Residual biological testing
Endpoints that cannot be adequately addressed by chemistry — sensitisation and implantation, in particular — still need biological evidence. But the range of endpoints requiring in vivo testing has narrowed significantly under a chemistry-led biocompatibility testing strategy.
Matching tests to device category
Limited-contact external devices
A short-term skin-contact device (< 24 hours) typically needs cytotoxicity, sensitisation, and irritation endpoints addressed. Chemical characterisation can often cover cytotoxicity and irritation; sensitisation usually still requires a guinea pig maximisation or local lymph node assay unless justified otherwise.
External-communicating devices
Devices in contact with mucosa, wound surfaces, or compromised tissue need additional systemic-toxicity and haemocompatibility evidence. A well-constructed biocompatibility testing strategy here leans heavily on chemical characterisation to bound systemic exposure.
Implants
Long-term and permanent implants require the most extensive evidence: implantation, chronic systemic toxicity, carcinogenicity (via TRA), and sometimes reproductive toxicity. The biocompatibility testing strategy for an implant is built around the chemistry-driven TRA with targeted biological testing for endpoints where chemistry is insufficient.
Reducing animal use through a chemistry-led strategy
The 3Rs principle — replacement, reduction, refinement — now has regulatory force in the EU via MDR Article 5(5) and is strongly encouraged by FDA. A chemistry-led biocompatibility testing strategy is the most scalable way to comply. Every endpoint that can be adequately addressed by extractables data plus toxicological derivation saves animal studies and accelerates the programme.
Validated in vitro alternatives continue to expand. OECD test guidelines now cover in vitro skin sensitisation (TG 442D, 442E), skin irritation (TG 439), eye irritation (TG 492), and acute oral toxicity (TG 432) among others. Integrating validated in vitro tests into the biocompatibility testing strategy is expected by reviewers, particularly in the EU.
Regulatory alignment across jurisdictions
FDA, EU notified bodies under MDR, and PMDA all accept ISO 10993 as the biocompatibility framework, but each has jurisdiction-specific expectations. FDA’s 2023 biocompatibility guidance explicitly recognises chemical characterisation as primary evidence for many endpoints. EU MDR technical documentation under Annex II requires a clinical-evaluation-coherent biological risk evaluation. PMDA aligns with ISO 10993 but places particular emphasis on residual solvent and sterilant evaluation. The ISO 10993-1:2018 standard is the common reference.
Common mistakes to avoid
Mistake 1: Testing first, characterising later. A biocompatibility testing strategy built without a chemistry-first foundation frequently ends up with expensive animal data that cannot answer the right questions.
Mistake 2: Reusing legacy test data without justification. Data from before 2018 may not satisfy current reviewer expectations, particularly around identification confidence in extractables studies.
Mistake 3: Ignoring manufacturing changes. A change in sterilisation, colourant, or polymer grade can invalidate the biological evaluation. The biocompatibility testing strategy must include a change-triggered reassessment procedure.
Mistake 4: Treating the TRA as a spreadsheet. Regulators read the risk narrative. A TRA with calculations but no argument is weaker evidence than a shorter TRA with a clear safety conclusion.
Mistake 5: Under-using read-across. Structurally related compounds with published data can substitute for compound-specific evidence when the similarity is well-justified. Our Read-Across Assessment method documents the case.
How ToxLibrary supports your biocompatibility testing strategy
A modern biocompatibility testing strategy depends on high-quality chemical characterisation data, robust toxicological derivations, and a risk narrative that convinces reviewers. Our Toxicology Monographs cover more than 7,200 compounds with verified Tolerable Intake values. Our TRA service builds the risk narrative and documents every uncertainty factor. Reach out — a chemistry-led biocompatibility strategy is faster, cheaper, and more ethical than legacy animal-heavy approaches.
