ISO 10993 Biocompatibility Testing for Medical Devices: The Complete Guide

What Biocompatibility Means — and Why It Matters

Biocompatibility is the ability of a medical device material to perform with an appropriate host response when applied as intended. That definition — drawn from David Williams' foundational work and adopted across the ISO 10993 series — is deceptively simple. It does not mean "inert." It does not mean "non-toxic." It means the material, in the context of its specific application and duration of contact with the body, does not produce an unacceptable biological response.

This distinction matters. A material that is perfectly biocompatible as a surface-contacting wound dressing might be entirely unacceptable as a permanent cardiovascular implant. Biocompatibility is not an intrinsic property of a material — it is a relationship between a material, a biological system, and a context of use.

Every medical device that contacts the patient — whether it touches intact skin for five minutes or resides permanently within the central nervous system — must undergo a biological evaluation. This evaluation determines whether the device poses unacceptable biological risks to the patient. The international standard governing this evaluation is ISO 10993, a multi-part series that provides a systematic framework for planning, executing, and interpreting biocompatibility assessments.

Getting biocompatibility wrong has consequences. Inadequate biocompatibility data is one of the most common reasons the FDA issues Additional Information (AI) requests during 510(k) review. In the EU, Notified Bodies scrutinize biological evaluation reports as a core element of technical documentation under the MDR. Globally, regulators expect manufacturers to demonstrate — with evidence — that their devices do not cause undue harm through their biological interactions.

This guide covers the entire landscape: the ISO 10993 framework, device categorization, the biological evaluation matrix, specific test methods, chemical characterization, risk-based approaches, FDA and EU MDR expectations, strategies for reducing testing, cost and timeline considerations, and the most common deficiencies that delay market clearance.

The ISO 10993 Series: Structure and Scope

ISO 10993 is not a single standard. It is a series of over 20 parts, each addressing a different aspect of biological evaluation. The series is maintained by ISO Technical Committee 194 (TC 194), which focuses on biological and clinical evaluation of medical devices.

Key Parts of ISO 10993

Part	Title	What It Covers
10993-1	Evaluation and testing within a risk management process	The master framework — categorization, endpoint selection, risk-based approach
10993-2	Animal welfare requirements	Ethical requirements for in vivo testing; principles of replacement, reduction, refinement (3Rs)
10993-3	Tests for genotoxicity, carcinogenicity and reproductive toxicity	Test methods and evaluation criteria for genetic and reproductive risks
10993-4	Selection of tests for interactions with blood	Hemocompatibility testing framework — thrombosis, coagulation, hematology, complement, platelets
10993-5	Tests for in vitro cytotoxicity	Cell culture-based toxicity screening methods
10993-6	Tests for local effects after implantation	Implantation study design and evaluation for tissue response
10993-7	Ethylene oxide sterilization residuals	Allowable limits for EtO and ethylene chlorohydrin residues
10993-9	Framework for identification and quantification of potential degradation products	Systematic degradation analysis methodology
10993-10	Tests for skin sensitization	Delayed-type hypersensitivity testing (guinea pig and murine local lymph node assay)
10993-11	Tests for systemic toxicity	Acute, subacute, subchronic, and chronic systemic toxicity evaluation
10993-12	Sample preparation and reference materials	Extraction conditions, extraction media, and sample preparation procedures
10993-13	Identification and quantification of degradation products from polymeric medical devices	Polymer-specific degradation analysis
10993-14	Identification and quantification of degradation products from ceramics	Ceramic-specific degradation analysis
10993-15	Identification and quantification of degradation products from metals and alloys	Metal and alloy degradation analysis
10993-16	Toxicokinetic study design for degradation products and leachables	Pharmacokinetic behavior of substances released from devices
10993-17	Establishment of allowable limits for leachable substances	Tolerable intake and margin of safety calculations
10993-18	Chemical characterization of medical device materials within a risk management process	Extractables, leachables, and material composition analysis
10993-23	Tests for irritation	Irritation testing methods (previously part of 10993-10)

The most critical part for any manufacturer is ISO 10993-1, which serves as the entry point to the entire series. It defines how to categorize your device, which biological endpoints to evaluate, and how to integrate the evaluation into your risk management process. Every other part in the series is called upon — or not — based on the decisions made under Part 1.

Edition History and Current Status

The 2018 edition of ISO 10993-1 (ISO 10993-1:2018) represented a major shift from earlier editions. The key changes included:

• Emphasis on a risk-based approach — testing is no longer automatic; chemical characterization and existing data can reduce or eliminate the need for biological testing
• Chemical characterization elevated — ISO 10993-18 (material and chemical characterization) moved from an optional consideration to a central pillar of the evaluation
• Biological evaluation plan (BEP) required — manufacturers must document a formal plan before conducting any testing
• Biological evaluation report (BER) required — a comprehensive report integrating all evidence must accompany the device to market
• Greater flexibility to use existing data — literature, clinical history, material equivalence arguments, and toxicological risk assessments can substitute for new testing when properly justified

These changes aligned ISO 10993-1 with the FDA's 2016 and subsequent guidance documents, which had already moved toward a risk-based, chemical-characterization-first approach. Understanding this philosophy is essential: the modern biocompatibility framework is not "run all the tests in the matrix." It is "identify the risks, characterize the chemistry, and test only what cannot be resolved by other means."

ISO 10993-1:2025 — The Sixth Edition

In November 2025, ISO published the sixth edition of ISO 10993-1 (ISO 10993-1:2025), titled "Requirements and general principles for the evaluation of biological safety within a risk management process." This revision introduces significant structural and conceptual changes from the 2018 edition. Manufacturers should understand these changes now, even though transition timelines vary by jurisdiction.

Key Changes in the 2025 Revision

Restructured device contact categories. The three-tier classification system (surface-contacting, external communicating, implant) has been simplified to four contact-based categories:

1. Intact skin
2. Intact mucosal membranes
3. Breached or compromised surface / internal tissues (excluding blood)
4. Blood

This eliminates the previous subcategories (e.g., "blood path, indirect" vs. "circulating blood") and categorizes devices solely by what tissue they contact.

Biological evaluation matrix replaced by four tables. The single prescriptive Table A.1 from the 2018 edition — the cornerstone of device evaluation planning — has been replaced by four separate tables, one for each contact category. Each table lists the biological effects to evaluate for that contact type and duration.

Terminology changes. Several key terms have been updated to align with ISO 14971 risk management vocabulary:

2018 Term	2025 Term
Biological endpoint	Biological effect
Implantation effects	Local effects after tissue contact
(Multiple systemic toxicity columns)	Consolidated systemic toxicity column

The rename of "implantation effects" to "local effects after tissue contact" reflects that local tissue response evaluation may also be relevant for non-implantable devices.

"Contact Day" concept introduced. The 2025 edition introduces the term "Contact Day" — defined as any day on which the device contacts the user, regardless of the duration of contact on that day. This concept, combined with clarified cumulative exposure calculations, may trigger recategorization of devices that were previously classified as limited or prolonged contact. Reusable devices are now explicitly categorized based on total cumulative exposure time rather than single-use duration.

Reasonably foreseeable misuse. The 2025 edition requires manufacturers to evaluate reasonably foreseeable misuse scenarios — defined as intentional or accidental but readily predictable human behavior not intended by the designer. This includes scenarios where:

• Users extend the device beyond its intended use duration
• The device contacts tissue sites other than those intended
• The device is used in different patient populations than intended

These misuse scenarios must be factored into the biological evaluation, potentially expanding the scope of endpoints to evaluate.

Life cycle evaluation. The revision emphasizes evaluating biological safety across the entire device life cycle — from initial use through end-of-life. For reusable devices, this means assessing biological safety after maximum reprocessing cycles, not just in the as-manufactured state.

Expanded genotoxicity and carcinogenicity requirements. Genotoxicity evaluation is now mandatory for all devices with prolonged (>24 hours to 30 days) or long-term (>30 days) tissue contact, except intact skin. Carcinogenicity assessment has been extended to include long-term intact mucosal membrane contact — a new requirement not present in the 2018 edition.

Strengthened 3Rs principles. The 2025 edition reinforces the Replace, Reduce, Refine (3Rs) principles for in vivo testing, encouraging greater use of in vitro alternatives and chemical characterization.

FDA Recognition Status of ISO 10993-1:2025

As of early 2026, the FDA has not recognized ISO 10993-1:2025. The U.S. delegation voted "No" through each stage of the FDIS (Final Draft International Standard) process. Key areas of disagreement include the restructured contact categories, changes to the biological evaluation matrix, and certain new requirements that the FDA considers insufficiently supported by scientific evidence.

For U.S. submissions, manufacturers should continue using ISO 10993-1:2018 as the recognized consensus standard. However, for EU MDR submissions and other international markets, the 2025 edition will become the applicable standard as Notified Bodies and regulatory authorities adopt it. Manufacturers pursuing simultaneous global submissions face the challenge of navigating two different editions of the same standard.

The FDA maintains its Recognized Consensus Standards database and may choose to recognize all, part, or none of the 2025 edition in the future. Monitor the FDA's consensus standards recognition list for updates.

ISO 10993-1: The Biological Evaluation Framework

ISO 10993-1 provides the master decision framework. Every biological evaluation starts here.

Step 1: Categorize the Device

The first task is to categorize the device by two dimensions: nature of body contact and contact duration. These two factors determine which biological endpoints must be evaluated.

Nature of Body Contact

ISO 10993-1 defines three primary categories, each with subcategories:

Category	Subcategory	Examples
Surface-contacting devices	Skin	Electrodes, external prostheses, compression bandages, tape, monitors
	Mucosal membrane	Contact lenses, urinary catheters, intravaginal devices, endotracheal tubes
	Breached or compromised surface	Wound dressings, surgical drapes contacting open wounds, occlusive patches on compromised skin
External communicating devices	Blood path, indirect	Blood administration sets, extension sets, IV tubing that does not directly contact circulating blood
	Tissue/bone/dentin communicating	Laparoscopic instruments, arthroscopic instruments, cement spacers, drain tubes
	Circulating blood	Intravascular catheters, temporary pacemaker electrodes, oxygenator tubing, dialysis components
Implant devices	Tissue/bone	Orthopedic pins, plates, screws, artificial ligaments, breast implants, pacemaker casings
	Blood	Heart valves, vascular grafts, stents, ventricular assist devices, total artificial hearts

A critical nuance: if a device has multiple components that contact different tissues, each component must be categorized separately. A hip replacement system, for example, includes components contacting bone (acetabular cup, femoral stem) and components contacting blood (surfaces exposed during surgery). Each is evaluated according to its own category.

Contact Duration

Duration Category	Time Period	Examples
Limited (A)	Less than or equal to 24 hours (single or cumulative)	Surgical instruments used intraoperatively, examination gloves, tongue depressors
Prolonged (B)	Greater than 24 hours to 30 days	Urinary catheters, wound drains, temporary bone fixation pins, central venous catheters
Permanent (C)	Greater than 30 days	Orthopedic joint replacements, dental implants, pacemakers, intraocular lenses, stents

Cumulative exposure matters. A device used for 20 minutes per session but used daily for months is not "limited contact." You must consider the cumulative exposure. If the same device is used repeatedly and the total contact time exceeds 24 hours, it is prolonged. If cumulative exposure exceeds 30 days, it is permanent. This catches many manufacturers off guard — reusable surgical instruments, patient monitoring electrodes, and dental appliances are common examples where cumulative use pushes the device into a higher duration category than a single-use analysis would suggest.

Step 2: Identify Required Biological Endpoints

Once you have the device category (nature of contact) and duration, ISO 10993-1 Table A.1 (the biological evaluation matrix) tells you which biological endpoints to evaluate. This is the core of the framework.

The Biological Evaluation Matrix

The matrix below is based on ISO 10993-1:2018, Table A.1. An "X" indicates the endpoint should be evaluated for that device category. "Evaluate" does not automatically mean "test" — it means the endpoint must be addressed, whether through testing, chemical characterization, literature review, clinical history, or material equivalence.

Biological Endpoint	Skin (A)	Skin (B)	Skin (C)	Mucosal (A)	Mucosal (B)	Mucosal (C)	Breached Surface (A)	Breached Surface (B)	Breached Surface (C)	Blood Path Indirect (A)	Blood Path Indirect (B)	Blood Path Indirect (C)	Tissue/Bone/Dentin (A)	Tissue/Bone/Dentin (B)	Tissue/Bone/Dentin (C)	Circulating Blood (A)	Circulating Blood (B)	Circulating Blood (C)	Tissue/Bone Implant (B)	Tissue/Bone Implant (C)	Blood Implant (B)	Blood Implant (C)
Cytotoxicity	X	X	X	X	X	X	X	X	X	X	X	X	X	X	X	X	X	X	X	X	X	X
Sensitization	X	X	X	X	X	X	X	X	X	X	X	X	X	X	X	X	X	X	X	X	X	X
Irritation or intracutaneous reactivity	X	X	X	X	X	X	X	X	X	X	X	X	X	X	X	X	X	X	X	X	X	X
Acute systemic toxicity	—	—	—	X	X	X	X	X	X	X	X	X	X	X	X	X	X	X	X	X	X	X
Material-mediated pyrogenicity	—	—	—	—	—	—	—	—	—	X	X	X	X	X	X	X	X	X	X	X	X	X
Subacute/subchronic toxicity	—	—	X	—	X	X	—	X	X	—	X	X	—	X	X	—	X	X	X	X	X	X
Chronic toxicity	—	—	X	—	—	X	—	—	X	—	—	X	—	—	X	—	—	X	—	X	—	X
Genotoxicity	—	—	X	—	—	X	—	—	X	X	X	X	X	X	X	X	X	X	X	X	X	X
Implantation	—	—	—	—	—	—	—	—	—	—	—	—	—	X	X	—	—	—	X	X	X	X
Hemocompatibility	—	—	—	—	—	—	—	—	—	X	X	X	—	—	—	X	X	X	—	—	X	X

Important clarification: This matrix is a starting point, not a checklist. ISO 10993-1:2018 explicitly states that the endpoints marked in the matrix should be evaluated — but the standard gives you latitude in how you address each one. For a well-characterized material with decades of clinical use, a literature-based justification may suffice for most endpoints. For a novel polymer with no clinical history, testing will likely be required for every marked endpoint and possibly additional endpoints not in the matrix.

Step 3: Develop the Biological Evaluation Plan (BEP)

The BEP is a document that must exist before any testing is conducted. It records:

• Device description, materials of construction, manufacturing processes, and sterilization method
• Device categorization (body contact and duration) with rationale
• Identification of all biological endpoints to be evaluated, with justification for inclusion or exclusion
• The approach for each endpoint: testing, chemical characterization with toxicological risk assessment, literature review, clinical data, material equivalence argument, or a combination
• Identification of applicable standards and test methods
• Extraction conditions per ISO 10993-12
• Rationale for any deviations from the standard matrix

The BEP is not a pro forma document. FDA reviewers and Notified Body auditors examine it critically. A BEP that simply lists all tests without justification — or that omits endpoints without explanation — will draw questions.

Step 4: Execute the Evaluation

With the BEP in place, you execute the evaluation. This may include chemical characterization, biological testing, literature review, or all of the above. The specific test methods are discussed in detail in the following sections.

Step 5: Write the Biological Evaluation Report (BER)

The BER integrates all evidence — test results, chemical characterization data, toxicological risk assessments, literature reviews, and clinical data — into a comprehensive document that concludes whether the device is biocompatible for its intended use.

A well-written BER includes:

• A summary of the device description, categorization, and intended use
• The BEP (or a reference to it)
• All test reports and their interpretation
• Chemical characterization results with toxicological risk assessment
• Literature review findings
• Gap analysis — does the evidence fully address all required endpoints?
• An overall conclusion on biocompatibility, authored or reviewed by a qualified toxicologist
• Identification of any residual biological risks and how they are managed

The BER must be signed by a person with appropriate qualifications — in practice, this means a board-certified toxicologist (DABT) or equivalent. Regulators will question a BER that lacks a qualified toxicological assessment.

Specific Test Methods: What Each Test Evaluates

Cytotoxicity (ISO 10993-5)

Cytotoxicity testing is the most fundamental biocompatibility test. It evaluates whether extracts from the device material are toxic to cells in culture. It is required for virtually every device category and duration.

Methods:

• Direct contact — device material placed directly on a cell monolayer
• Agar diffusion — device material placed on an agar overlay covering the cells
• Elution (extract) — device material extracted in culture medium, then the extract is applied to cells (most common for regulatory submissions)

The standard cell line is L-929 mouse fibroblasts, though other cell lines are used. The test measures cell viability (typically via MTT, XTT, or neutral red uptake assays) after 24-48 hours of exposure. A reduction in cell viability below 70% of the negative control is generally considered cytotoxic.

Qualitative vs. quantitative assessment:

ISO 10993-5 supports both qualitative (morphological grading) and quantitative (viability assay) evaluation. The qualitative MEM elution test uses a 0-4 grading scale:

Grade	Reactivity	Description
0	None	No detectable zone around or under sample; discrete intracytoplasmic granules, no cell lysis
1	Slight	Some malformed or degenerate cells; zone limited to area under sample
2	Mild	Zone limited to area under sample; greater than 50% of cells are viable
3	Moderate	Zone may extend up to 1.0 cm beyond sample; greater than 25% but less than 50% of cells are viable
4	Severe	Zone extends more than 1.0 cm beyond sample; nearly complete destruction of cell layer

A qualitative grade of 0, 1, or 2 is generally considered a pass. A grade of 3 or 4 indicates cytotoxicity. Quantitative assays (MTT, XTT, NRU, WST-1) use the 70% cell viability threshold — results at or above 70% of the negative control pass.

Extraction ratios for cytotoxicity testing: The standard extraction condition for cytotoxicity is 6 cm²/mL (surface area to volume) or 0.2 g/mL (mass to volume) at 37 +/- 1 degrees C for 24 +/- 2 hours in serum-supplemented culture medium. For irregularly shaped devices, the mass-to-volume ratio is used. The extraction temperature for cytotoxicity (37 degrees C) is lower than the exaggerated conditions used for other biological tests — this is intentional, as cytotoxicity extraction uses cell culture medium that would denature at higher temperatures.

Why it matters: Cytotoxicity is a screening test. It is fast (results in days), inexpensive (typically $1,500-3,000), and catches materials that are overtly toxic to cells. A cytotoxicity failure is a red flag that usually stops further testing until the root cause is identified.

Sensitization (ISO 10993-10)

Sensitization testing evaluates whether the device or its extracts can cause delayed-type hypersensitivity (allergic contact dermatitis) upon repeated exposure. This is relevant because many medical devices involve repeated skin or mucosal contact.

Methods:

• Guinea Pig Maximization Test (GPMT) — Magnusson-Kligman method. Involves intradermal injection and topical application of test extract to guinea pigs, followed by a challenge application after a sensitization period. The gold standard for decades.
• Guinea Pig Closed Patch Test (Buehler test) — Topical application only, less aggressive than GPMT. Used when the device only contacts intact skin.
• Murine Local Lymph Node Assay (LLNA) — A mouse-based alternative that measures lymph node cell proliferation as an indicator of sensitization potential. Increasingly preferred due to animal welfare considerations (3Rs principle) and quantitative results.

In vitro sensitization alternatives (non-animal methods):

The regulatory landscape for sensitization testing is evolving rapidly. Several fully in vitro methods targeting specific key events in the Adverse Outcome Pathway (AOP) for skin sensitization are gaining acceptance:

Method	OECD Test Guideline	Key Event Addressed	What It Measures
DPRA (Direct Peptide Reactivity Assay)	TG 442C	KE1 — Molecular initiating event	Reactivity of extractables with cysteine and lysine peptides (protein binding)
KeratinoSens	TG 442D	KE2 — Keratinocyte activation	Activation of the Nrf2-keap1 signaling pathway via luciferase reporter gene
h-CLAT (human Cell Line Activation Test)	TG 442E	KE3 — Dendritic cell activation	Upregulation of CD86 and CD54 surface markers on THP-1 cells
GARD (Genomic Allergen Rapid Detection)	Referenced in ISO 10993-10 Annex C	KE2/KE3 — Genomic response	Machine learning analysis of gene expression signatures in a dendritic cell-like cell line

These methods are referenced in Annex C of ISO 10993-10:2021 (the current edition). When combined into an Integrated Approach to Testing and Assessment (IATA) — using two or three complementary methods covering different key events — the predictive accuracy can reach 91% or higher (96.3% sensitivity, 87.1% specificity when combining DPRA, KeratinoSens, and h-CLAT). Defined approaches using combinations of these assays are increasingly accepted for regulatory submissions, particularly in the EU. FDA acceptance of fully in vitro sensitization approaches for medical devices is still evolving — the agency currently prefers the LLNA or GPMT for definitive sensitization assessment but has shown willingness to consider in vitro data as part of a weight-of-evidence approach.

Typical cost: $5,000-15,000 depending on method. LLNA is generally less expensive than GPMT. In vitro methods (DPRA, KeratinoSens, h-CLAT) are typically $3,000-8,000 each, but two or three are needed for a defined approach.

Timeline: 4-6 weeks for GPMT; 1-2 weeks for LLNA. In vitro methods typically complete in 1-3 weeks.

Irritation (ISO 10993-23)

Irritation testing evaluates whether the device causes local inflammatory responses at the site of contact. Previously covered under ISO 10993-10, irritation now has its own dedicated standard (ISO 10993-23, published in 2021).

Methods:

• Intracutaneous (intradermal) reactivity — Injection of device extracts intradermally in rabbits, with observation of erythema and edema over 72 hours. Applicable when the device contacts tissue below the skin surface.
• Primary skin irritation — Topical application to rabbit skin (or in vitro reconstructed human epidermis models). Applicable for skin-contacting devices.
• Mucosal irritation — Application to relevant mucosal tissue (oral, vaginal, rectal, penile) in appropriate animal models.
• Ocular irritation — For devices contacting the eye.
• In vitro irritation — Reconstructed human tissue models (e.g., EpiDerm, SkinEthic) are increasingly accepted as alternatives to animal testing for skin and ocular irritation.

In vitro reconstructed human tissue models for irritation:

Model	Manufacturer	Application	Standard Method
EpiDerm (EPI-200)	MatTek	Skin irritation	OECD TG 439
SkinEthic RHE	EPISKIN (L'Oreal)	Skin irritation	OECD TG 439
EpiSkin	EPISKIN (L'Oreal)	Skin irritation	OECD TG 439
EpiOcular (OCL-200)	MatTek	Ocular irritation	OECD TG 492
SkinEthic HCE	EPISKIN (L'Oreal)	Ocular irritation	OECD TG 492

These models use human-derived keratinocytes grown at the air-liquid interface to form a stratified, differentiated epithelium that mimics the structure of human skin or corneal epithelium. The device extract is applied topically for a defined period (typically 15-60 minutes for skin models), followed by a post-exposure incubation period. Tissue viability is measured using the MTT assay. A viability reduction below 50% of the negative control is generally considered irritant.

In vitro reconstructed human epidermis (RhE) models are now well-accepted for skin irritation assessment in the EU and are referenced in ISO 10993-23. For devices that only contact intact skin, in vitro irritation results are generally sufficient for EU regulatory submission without animal irritation data. FDA acceptance of RhE models is also improving, though the agency may still request in vivo irritation data for higher-risk devices or devices contacting mucosal membranes.

Typical cost: $3,000-8,000 depending on the specific model and route.

Acute Systemic Toxicity (ISO 10993-11)

Acute systemic toxicity testing determines whether device extracts cause systemic toxic effects after a single exposure. The test article (device extract) is administered to mice or rabbits via relevant routes (intravenous and/or intraperitoneal), and animals are observed for 24-72 hours for signs of toxicity, morbidity, or mortality.

Methods:

• Mouse systemic injection test — Intravenous and/or intraperitoneal injection of extracts in multiple vehicles (saline, alcohol/saline, PEG, cottonseed oil). Animals observed at 4, 24, 48, and 72 hours.
• Material-mediated pyrogenicity (fever response) — Rabbit pyrogen test (RPT) or Monocyte Activation Test (MAT, an in vitro alternative). Evaluates whether the device or its extracts cause a febrile (fever) response.

Route of administration matters. ISO 10993-11 specifies that the route of extract administration should simulate the clinical exposure route as closely as possible:

Route	When to Use	Extraction Vehicles
Intravenous (IV)	Blood-contacting devices, intravenous drug delivery components	Saline, alcohol/saline (5% ethanol)
Intraperitoneal (IP)	General default for internal-contact devices when IV is impractical	Saline, PEG 400, cottonseed oil
Subcutaneous/intramuscular	Tissue-contacting implant devices	Saline, PEG 400, cottonseed oil
Intracutaneous/intradermal	For intracutaneous reactivity testing specifically (irritation endpoint under ISO 10993-23)	Saline, cottonseed oil or sesame oil

For each route, extracts in at least a polar (saline) and a non-polar (oil-based) vehicle are tested. The mouse is the standard species (minimum 5 per group), with observation at 4, 24, 48, and 72 hours post-injection. Body weight loss exceeding 2 grams or clinical signs of systemic toxicity (piloerection, lethargy, labored breathing, convulsions, mortality) in the test group beyond what is observed in the vehicle control group constitutes a failure.

Note on pyrogen testing: Bacterial endotoxin testing (LAL/rFC) addresses contamination-based pyrogens. Material-mediated pyrogenicity testing addresses pyrogens intrinsic to the device materials themselves. Both may be required — they address different risk factors.

Typical cost: $5,000-12,000 for acute systemic toxicity; $2,000-5,000 for pyrogenicity.

Subacute, Subchronic, and Chronic Systemic Toxicity (ISO 10993-11)

For devices with prolonged or permanent contact, longer-duration systemic toxicity studies may be required:

Study Type	Duration	When Required
Subacute toxicity	Up to 28 days	Prolonged contact with certain tissues; permanent skin contact
Subchronic toxicity	90 days (typically)	Permanent contact with internal tissues, blood path, or circulating blood
Chronic toxicity	6-12 months	Permanent implants and blood-contacting devices when chemical characterization reveals substances of concern

These studies are significantly more expensive ($50,000-200,000+) and time-consuming (3-15 months) than acute studies. The 2018 edition of ISO 10993-1, combined with FDA guidance, has substantially reduced the frequency with which these studies are required — chemical characterization and toxicological risk assessment can often address systemic toxicity endpoints without animal testing.

Practical tip: Before committing to a subchronic or chronic toxicity study, exhaust chemical characterization first. If you can identify and quantify all extractable substances and demonstrate through toxicological risk assessment that their levels are below established tolerable intake limits, you may be able to justify not performing these expensive animal studies. This approach is explicitly encouraged by both ISO 10993-1:2018 and the FDA's biocompatibility guidance.

Genotoxicity (ISO 10993-3)

Genotoxicity testing evaluates whether device extracts can damage DNA or cause gene mutations. This is required for devices with permanent body contact and for devices with direct or indirect blood contact regardless of duration.

A standard genotoxicity battery typically includes three tests covering different genetic endpoints:

Test	Endpoint	System	Typical Cost
Bacterial reverse mutation (Ames test)	Gene mutation	Salmonella typhimurium and E. coli strains	$5,000-8,000
In vitro mammalian chromosomal aberration or mouse lymphoma assay	Chromosomal damage	CHO cells, CHL cells, or mouse lymphoma L5178Y cells	$8,000-12,000
In vitro micronucleus assay	Chromosomal damage (alternative to above)	Human lymphocytes or other mammalian cells	$6,000-10,000

The complete genotoxicity battery runs $15,000-30,000 and takes 6-10 weeks. A positive result in any single test does not automatically mean the device is genotoxic — but it triggers follow-up testing (typically in vivo studies) to confirm or refute the finding.

Test protocol details for the Ames test (ISO 10993-3 / OECD TG 471): The Ames test uses five Salmonella typhimurium strains (TA98, TA100, TA1535, TA1537, and TA97a or TA102) and optionally one E. coli strain (WP2 uvrA). The test is performed both with and without metabolic activation (S9 liver microsomal fraction from Aroclor 1254-induced rats) to detect both direct-acting mutagens and promutagens requiring metabolic activation. Device extracts in saline and DMSO (or other appropriate vehicles) are tested at multiple concentrations. A reproducible, dose-related increase in revertant colony counts (typically greater than or equal to 2-fold over the concurrent solvent control for strains TA98, TA100, and TA102, or greater than or equal to 3-fold for strains TA1535 and TA1537) constitutes a positive result.

In vivo follow-up for positive in vitro results: If one or more in vitro genotoxicity tests yield positive results, an in vivo genotoxicity study is typically required to determine whether the finding is relevant in a whole-organism context. The most common follow-up is the in vivo mammalian micronucleus test (OECD TG 474) — a mouse bone marrow micronucleus assay that evaluates chromosomal damage in vivo. This study costs $20,000-40,000 and takes 6-8 weeks. A negative in vivo result can override a positive in vitro finding, provided the target tissue was adequately exposed to the test substance.

Implantation (ISO 10993-6)

Implantation testing evaluates the local tissue response to a device material when placed in an appropriate tissue site. This is required for implant devices and some external communicating devices with prolonged or permanent tissue contact.

Study design:

• Test material and control material are implanted in appropriate tissue (subcutaneous, muscle, bone) in rabbits, rats, or other suitable species
• Implantation periods are based on contact duration: 1-4 weeks for short-term, 12 weeks for prolonged, 26-52+ weeks for permanent
• At the end of the implantation period, tissue surrounding the implant is harvested and evaluated histopathologically
• A semi-quantitative scoring system grades inflammatory cell infiltration, neovascularization, fibrosis, fatty infiltrate, and other tissue responses
• The test material response is compared to the control material response

Typical cost: $30,000-100,000+ depending on species, implantation duration, and number of time points.

Timeline: 3-15 months depending on implantation duration.

Hemocompatibility (ISO 10993-4)

Hemocompatibility testing evaluates the device's interaction with blood. This is required for any device that directly or indirectly contacts circulating blood.

ISO 10993-4 identifies five categories of blood interaction, each with specific test methods:

Category	What It Measures	Common Tests
Thrombosis	Blood clot formation on device surfaces	In vivo thrombosis models, flow loop studies
Coagulation	Activation of the coagulation cascade	Partial thromboplastin time (PTT), prothrombin time (PT), thrombin time (TT), fibrinogen levels
Platelets	Platelet activation, adhesion, and aggregation	Platelet count, platelet factor 4 (PF4), beta-thromboglobulin
Hematology	Effects on blood cells	Complete blood count (CBC), hemolysis (free hemoglobin), reticulocyte count
Complement system	Complement activation	C3a, C5a, SC5b-9 (terminal complement complex)

Hemolysis deserves special mention. It is the most commonly performed hemocompatibility test and evaluates whether the device causes red blood cell lysis (destruction). The hemolysis test (ASTM E2524 or equivalent) is relatively simple, fast, and inexpensive ($3,000-5,000). Most regulatory submissions for blood-contacting devices include hemolysis testing at minimum.

Hemolysis test protocol detail (ASTM E2524 / NIH method): The hemolysis test exposes the device material or extract to fresh human or rabbit blood (diluted in saline or PBS) at 37 degrees C for a defined contact period (typically 1-4 hours). After contact, the blood is centrifuged and the supernatant is analyzed spectrophotometrically at 540 nm for free hemoglobin. Results are reported as a hemolytic index — a value of 5% or greater is generally considered hemolytic, 2-5% is mildly hemolytic, and less than 2% is non-hemolytic. Positive (water) and negative (saline) controls are required. The test uses either direct contact (device material placed in blood) or extract contact (device extract added to blood), with direct contact being the preferred method when device geometry permits.

Partial thromboplastin time (PTT) and other coagulation tests are performed using citrated human plasma exposed to the device or its extract. Clinically significant shortening of PTT (indicating procoagulant effects) or prolongation (indicating anticoagulant or consumptive effects) relative to the negative control triggers further evaluation.

Full hemocompatibility evaluation (all five categories) is expensive ($50,000-150,000+) and complex. For many devices, hemolysis plus a subset of the other categories — guided by risk assessment — is acceptable. ISO 10993-4 provides a risk-based framework for selecting which hemocompatibility categories to evaluate based on the device's blood contact characteristics (direct vs. indirect, high vs. low flow, arterial vs. venous).

Additional Endpoints

Depending on the device, additional biological endpoints may need evaluation:

Endpoint	When Required	What It Evaluates
Carcinogenicity	Permanent implants when materials are known or suspected carcinogens, or when chemical characterization identifies carcinogenic substances	Long-term tumor-forming potential
Reproductive/developmental toxicity	Devices that may expose reproductive tissues or the developing fetus to leachable substances	Effects on fertility, embryonic development, and offspring
Degradation (ISO 10993-9, -13, -14, -15)	Absorbable/biodegradable devices, devices that may degrade in vivo	Identification and quantification of degradation products
Toxicokinetics (ISO 10993-16)	When leachable substances are identified and their systemic distribution needs characterization	Absorption, distribution, metabolism, and excretion of device-derived substances

Carcinogenicity studies are extremely rare for medical devices — they cost $500,000-2,000,000+, take 2+ years, and require hundreds of animals. Chemical characterization with toxicological risk assessment has effectively replaced carcinogenicity testing for nearly all medical device applications. The same trend applies to reproductive toxicity testing.

Chemical Characterization: ISO 10993-18

The 2018 revision of ISO 10993-1 elevated chemical characterization from a supplementary activity to a central pillar of biological evaluation. ISO 10993-18 provides the framework.

Why Chemical Characterization Matters

The logic is straightforward: if you know exactly what chemicals a device releases and you can demonstrate through toxicological risk assessment that those chemicals are at safe levels, you may not need biological testing to address the corresponding endpoints. This is faster, cheaper, and reduces animal use.

Chemical characterization answers three questions:

1. What is the device made of? (Material composition)
2. What chemicals can be extracted from the device under exaggerated conditions? (Extractables)
3. What chemicals actually leach from the device under clinical use conditions? (Leachables)

Material Composition

Identify every material in the device, including:

• Base polymers, metals, ceramics, and composites
• Additives: plasticizers, stabilizers, colorants, antioxidants, processing aids
• Coatings, adhesives, lubricants, printing inks
• Sterilization residuals
• Manufacturing residuals (mold release agents, machine oils, cleaning agents)

Material composition is typically documented through supplier certificates of analysis, material safety data sheets, master formulation records, and analytical testing (FTIR, XRF, EDS).

Extractables vs. Leachables

Characteristic	Extractables	Leachables
Definition	Substances that can be released from the device under exaggerated (aggressive) extraction conditions	Substances that actually migrate from the device under normal or simulated clinical use conditions
Extraction conditions	Aggressive solvents (hexane, isopropanol, water), elevated temperatures, extended times per ISO 10993-12	Simulated use conditions — clinically relevant media, physiological temperature, actual contact duration
Purpose	Identify the universe of what could be released — worst-case scenario	Determine what is released under real-world conditions
Typical order	Performed first	Performed if extractables study identifies substances of concern
Regulatory role	Required as baseline chemical characterization	May be requested if extractables data raises questions, or if the device has prolonged/permanent contact

Extraction Conditions: ISO 10993-12 in Detail

ISO 10993-12:2021 governs sample preparation and extraction conditions. Choosing the correct extraction parameters is critical — incorrect conditions are a frequent cause of FDA Additional Information requests.

Standard Extraction Temperatures and Durations

ISO 10993-12 specifies four standardized exaggerated extraction conditions:

Temperature	Duration	When to Use
(37 +/- 1) degrees C	(72 +/- 2) hours	Devices used at body temperature where material stability at higher temperatures is uncertain; also the standard condition for cytotoxicity extraction (but at 24 hours in culture medium)
(50 +/- 2) degrees C	(72 +/- 2) hours	The most commonly used and FDA-accepted exaggerated condition for most polymeric devices; provides sufficient extraction aggression without causing material degradation for most polymers
(70 +/- 2) degrees C	(24 +/- 2) hours	An alternative exaggerated condition for materials stable at this temperature; useful for some elastomers and thermoplastics
(121 +/- 2) degrees C	(1 +/- 0.1) hours	The most aggressive condition; appropriate only for materials confirmed stable at autoclave temperatures (metals, ceramics, some high-temperature polymers like PEEK). Can cause degradation of many polymers and should not be used unless material stability is verified

The FDA's 2023 biocompatibility guidance and the September 2024 draft guidance on chemical analysis both emphasize using exaggerated conditions. The 50 degrees C / 72-hour condition is the most widely accepted default for polymeric devices. Using only 37 degrees C extraction (except for cytotoxicity) may draw an AI request.

Extraction Vehicles (Solvents)

ISO 10993-12 requires extraction in both polar and non-polar vehicles to capture the full range of potential extractables:

Vehicle Category	Examples	What It Captures
Polar	Purified water, physiological saline (0.9% NaCl), culture medium without serum	Water-soluble organic compounds, salts, ionic species, hydrophilic additives
Non-polar	Freshly refined vegetable oil (sesame oil, cottonseed oil) of pharmacopoeial quality	Lipid-soluble organic compounds, plasticizers, antioxidants, hydrophobic additives
Semi-polar (supplementary)	Ethanol/water mixtures, ethanol/saline, DMSO, PEG 400 (diluted to physiological osmotic pressure)	Compounds of intermediate polarity; useful when polar and non-polar vehicles alone provide incomplete coverage

For chemical characterization (ISO 10993-18), more aggressive solvents are typically used — hexane and isopropanol are common non-polar and semi-polar solvents for extractables studies, as they extract more aggressively than vegetable oil. The FDA's September 2024 draft guidance recommends "harsh non-polar and semi-polar solvents (e.g., hexane and isopropanol)" for extractables studies but acknowledges these can be incompatible with certain polymers such as silicones.

Surface Area-to-Volume Ratios

Device Form	Extraction Ratio	When to Use
Flat or sheet-like devices	6 cm2/mL (surface area to volume) for biological testing; 3 cm2/mL for chemical characterization	Use surface area ratio whenever the device surface area can be measured or reasonably estimated
Irregular shapes, porous devices, powders	0.2 g/mL (mass to volume)	When surface area cannot be determined; mass ratio provides a conservative alternative
Tubing, fluid path devices	Fill-and-soak using the internal volume, or external surface area ratio	For devices that form a fluid path, extraction through the lumen simulates clinical use

The surface area ratio should be used whenever feasible. The mass-to-volume ratio is a fallback for irregularly shaped devices where surface area measurement is impractical.

Solvent compatibility warning: The selected extraction solvent must not compromise the device material through severe swelling, particulate generation, or degradation. If a solvent causes visible material changes, it is not appropriate for that device — but this must be documented, and an alternative solvent must be justified. The FDA's 2024 draft guidance states that when compatible solvents cannot be identified after feasibility studies, chemical characterization may be inappropriate and biological testing should be performed instead.

Analytical Methods for Chemical Characterization

Technique	Abbreviation	What It Detects
Gas chromatography-mass spectrometry	GC-MS	Volatile and semi-volatile organic compounds
Liquid chromatography-mass spectrometry	LC-MS	Non-volatile organic compounds, polar compounds
Inductively coupled plasma-mass spectrometry	ICP-MS	Metals and trace elements
Inductively coupled plasma-optical emission spectrometry	ICP-OES	Metals at higher concentrations
Headspace GC-MS	HS-GC-MS	Volatile organic compounds and residual solvents
Fourier-transform infrared spectroscopy	FTIR	Polymer identification, surface chemistry
Total organic carbon	TOC	Non-specific organic content screen
Non-volatile residue	NVR	Total extractable non-volatile mass

A robust extractables study typically employs multiple complementary techniques — GC-MS for volatile and semi-volatile organics, LC-MS for non-volatile organics, and ICP-MS/ICP-OES for metals — to achieve comprehensive coverage.

Non-Targeted vs. Targeted Analysis

Chemical characterization studies use two complementary analytical strategies:

Non-targeted analysis (screening) surveys the extract for all detectable compounds without prior knowledge of what is present. This is the primary approach for initial extractables profiling. GC-MS, LC-MS, and ICP-MS are run in full-scan mode to capture the broadest possible range of compounds. Chromatographic peaks above the Analytical Evaluation Threshold (AET) are identified through spectral library matching (e.g., NIST mass spectral library for GC-MS), accurate mass determination (for high-resolution LC-MS/QTOF), and structural elucidation. Non-targeted analysis answers the question: "What is in this extract?"

Targeted analysis (confirmation) focuses on specific compounds — either those identified in non-targeted screening that require quantitative confirmation, or known compounds of concern (e.g., phthalates, bisphenol A, PAHs, specific monomers or catalysts). Targeted methods use reference standards for confirmed identification and validated quantitation. Targeted analysis answers the question: "Exactly how much of this specific substance is present?"

The FDA's September 2024 draft guidance on chemical analysis emphasizes that compound identifications must reach "confident" or "confirmed" levels to be used in toxicological risk assessment. "Tentative" or "unknown" identifications — which are common for low-level extractables — require additional analytical work (orthogonal detection methods, reference standard confirmation) before they can support regulatory conclusions.

Method Selection Guidance

Compound Class	Primary Method	Complementary Method	Detection Limit Range
Volatile organics (bp <200 degrees C)	HS-GC-MS (headspace)	Purge-and-trap GC-MS	0.1-1 micrograms/mL
Semi-volatile organics (bp 200-400 degrees C)	GC-MS (liquid injection)	GC-MS/MS for confirmatory work	0.1-1 micrograms/mL
Non-volatile organics, polar compounds	LC-MS (or LC-MS/MS)	LC-UV/PDA as orthogonal detection	0.1-5 micrograms/mL
High molecular weight compounds	GPC/SEC (gel permeation chromatography)	LC-MS with ESI or APCI	Variable
Metals and trace elements	ICP-MS	ICP-OES for higher concentrations	0.01-0.1 micrograms/mL (ICP-MS); 0.1-10 micrograms/mL (ICP-OES)
Residual solvents	HS-GC-MS or HS-GC-FID	GC-MS with direct injection	0.1-1 micrograms/mL
Non-specific organic content	TOC (total organic carbon)	NVR (non-volatile residue, gravimetric)	0.5-5 micrograms/mL (TOC)

High-resolution mass spectrometry (HRMS): Quadrupole time-of-flight (QTOF) instruments, now widely available, provide accurate mass data (less than 5 ppm mass error) that can determine molecular formulas even for unknown compounds. QTOF-based LC-MS and GC-MS methods are increasingly used in non-targeted extractables screening because they generate richer structural information than unit-resolution instruments, improving identification confidence for unknowns.

Triplicate Extraction Requirement

The FDA's September 2024 draft guidance recommends that three separate batches of device materials be extracted and analyzed separately using all analytical techniques. The highest concentrations detected across all three batches should be used for toxicological risk evaluation. This triplicate requirement reflects manufacturing variability — a single extraction may not capture the worst-case extractables profile. Individual extraction of single materials (rather than the complete multi-material assembly) is also recommended when feasible, to correlate each extractable substance to its source material.

Toxicological Risk Assessment (ISO 10993-17)

Once chemical characterization identifies the extractable or leachable substances, each substance is evaluated for toxicological risk using ISO 10993-17 principles:

1. Identify the substance — by chemical name, CAS number, and structure
2. Determine the tolerable intake (TI) or tolerable contact level (TCL) — derived from published toxicological data (NOAELs, LOAELs, benchmark doses) divided by appropriate uncertainty factors
3. Calculate the estimated exposure — based on the amount released, device contact area or volume, and frequency of use
4. Calculate the margin of safety (MoS) — TI divided by estimated exposure. An MoS greater than 1 indicates acceptable risk for the evaluated endpoint

This approach allows you to address multiple biological endpoints (systemic toxicity, genotoxicity, carcinogenicity, reproductive toxicity) through chemical data and published toxicological information rather than new animal testing.

Example: An extractables study on a polyurethane catheter identifies 2-ethylhexanol at 0.5 micrograms per device. Published toxicological data establishes a tolerable intake of 340 micrograms per day. Estimated patient exposure from one catheter per day is 0.5 micrograms. The margin of safety is 680 (340/0.5) — far above 1. Systemic toxicity, genotoxicity, and carcinogenicity endpoints for this specific substance are adequately addressed without animal testing.

Analytical Evaluation Threshold (AET)

The AET is the concentration below which an extracted substance does not need individual identification and toxicological evaluation. It is calculated based on the Threshold of Toxicological Concern (TTC) concept — if a substance is present below the TTC-derived threshold, it is considered to pose negligible toxicological risk regardless of its identity.

For most non-genotoxic substances, the TTC is 1.5 micrograms per day (derived from Kroes et al., 2004). For substances with structural alerts for genotoxicity, lower thresholds apply (typically 0.15 micrograms per day, aligned with ICH M7 principles for mutagenic impurities).

The AET allows labs to focus identification and risk assessment efforts on the substances that matter, rather than chasing every trace-level peak in a chromatogram.

The Risk-Based Approach to Biological Evaluation

The 2018 edition of ISO 10993-1 fundamentally reframed biocompatibility evaluation as a risk management activity — not a testing checklist. This aligns with the overarching principles of ISO 14971 (risk management for medical devices).

The Decision Framework

For each biological endpoint identified in the matrix, the manufacturer should follow this decision hierarchy:

1. Can the endpoint be addressed through existing data? This includes published literature, clinical history of the material, data from substantially equivalent devices, or supplier data.
2. Can chemical characterization with toxicological risk assessment address the endpoint? If extractables and leachables are identified and quantified, and toxicological risk assessment demonstrates adequate margins of safety, biological testing may not be needed.
3. Can in vitro testing address the endpoint? In vitro tests (cytotoxicity, genotoxicity, irritation models) are preferred over in vivo tests when scientifically valid alternatives exist.
4. Is in vivo testing required? Only when the above approaches cannot adequately address the endpoint.

This hierarchy explicitly prioritizes non-testing approaches and in vitro methods over animal testing. It is not just a philosophical preference — it reflects the scientific reality that chemical characterization with expert toxicological interpretation often provides more relevant safety information than a biological test on an extract.

When Testing Can Be Reduced or Waived

Several scenarios support reducing or eliminating biological testing:

Well-characterized materials with extensive clinical history. Surgical-grade stainless steel (316L/316LVM), commercially pure titanium (Grade 1-4), Ti-6Al-4V alloy, UHMWPE, PMMA bone cement, medical-grade silicone, and many other materials have decades of clinical use data. For devices made entirely of these materials and manufactured using established processes, literature-based justification can address most or all biological endpoints.

Material equivalence arguments. If your device uses the same material (same grade, same supplier, same processing) as a legally marketed device with an established biocompatibility profile, you can argue equivalence. This requires demonstrating that the material is truly identical — not just similar — and that manufacturing, sterilization, and packaging processes do not introduce new biological risks.

Chemical characterization demonstrating negligible extractables. If exhaustive extraction studies reveal no detectable substances above the AET, the toxicological risk from the device is demonstrably negligible, and many biological endpoints can be addressed without testing.

Predicate device data (FDA pathway). For 510(k) submissions, if your device uses the same materials and manufacturing processes as a predicate device that has established biocompatibility, you can reference the predicate's biocompatibility data. However, you must demonstrate material equivalence — not just claim it.

Clinical data. If the device (or a substantially equivalent device) has been used clinically without adverse biological effects, this clinical evidence supports biocompatibility. Post-market surveillance data, published clinical studies, and complaint history all contribute.

Warning: "Reduced testing" does not mean "no documentation." Every decision to reduce or omit testing must be thoroughly justified in the BEP and BER. A one-sentence claim that "the material is well known" without supporting evidence will not survive regulatory review. You must cite specific literature, provide material certificates, present clinical history data, and articulate a clear toxicological rationale.

FDA Guidance on Biocompatibility

The FDA recognizes ISO 10993-1 as a consensus standard and has published guidance documents that describe its expectations for biocompatibility evaluation. The key document is "Use of International Standard ISO 10993-1, Biological evaluation of medical devices — Part 1: Evaluation and testing within a risk management process" (most recently updated in 2023).

FDA September 2024 Draft Guidance: Chemical Analysis for Biocompatibility Assessment

In September 2024, the FDA published a significant draft guidance titled "Chemical Analysis for Biocompatibility Assessment of Medical Devices." This guidance represents the most detailed FDA articulation to date of expectations for chemical characterization under ISO 10993-18 and has substantial implications for how manufacturers plan extractables and leachables studies.

Scope and purpose. The draft guidance describes recommended methodological approaches for chemical analysis as an alternative to biological testing for evaluating specific biocompatibility endpoints, including acute, subacute, subchronic, and chronic systemic toxicity, genotoxicity, carcinogenicity, and reproductive/developmental toxicity. It covers information gathering, test article extraction, chemical analysis, and data reporting.

Key requirements and recommendations:

Area	FDA Recommendation	Impact
Triplicate extraction	Three batches of device materials must be extracted and analyzed separately; worst-case (highest) concentrations used for risk evaluation	Increases testing cost by approximately 2-3x compared to single-batch extraction
Solvent selection	Harsh non-polar and semi-polar solvents (hexane, isopropanol) recommended; deviations must be justified	May be incompatible with certain polymers (e.g., silicones), potentially requiring biological testing as fallback
Exhaustive extraction	Extraction must be demonstrated to be exhaustive — serial extractions until no additional substances are detected above the AET	Adds extraction cycles and analytical cost
Chemical identification confidence	Identifications must reach "confident" or "confirmed" levels for use in toxicological risk assessment; "tentative" or "unknown" identifications require additional work	Increases analytical burden for trace-level extractables
Non-targeted + targeted analysis	Non-targeted screening required as baseline; targeted analysis required to confirm identity and quantify specific substances of concern	Two-phase analytical approach is mandatory, not optional
Worst-case reprocessing	Reusable devices must be tested after maximum reprocessing cycles, including all combinations of cleaning and sterilization	Can multiply test conditions substantially for multi-cycle reusable devices
Material and supplier transparency	FDA increasingly expects detailed material composition, supplier identity, and manufacturing information	Goes beyond what 510(k) submissions have traditionally required
Cohort of concern	Special attention to genotoxic, mutagenic, and carcinogenic substances; lower identification thresholds apply	Aligns with ICH M7 principles for mutagenic impurities

Industry response. The draft guidance has been described by industry observers as "detailed and burdensome." Key concerns include:

• The requirement for confirmed chemical identities may be unattainable for many low-level extractables despite best analytical efforts
• Worst-case reprocessing testing for reusable devices could require hundreds or thousands of reprocessing cycles
• The guidance's contact duration classification for repeated-use devices may conflict with ISO 10993-1 categorization rules
• The overall cost and timeline implications may negate the theoretical time savings of chemical characterization over biological testing for certain device types

The comment period closed on December 19, 2024. Manufacturers should monitor the final guidance publication and prepare for these more detailed requirements to become the de facto standard for chemical characterization submitted to the FDA.

Strategic note: Despite the increased burden, chemical characterization remains the preferred pathway for addressing systemic toxicity, genotoxicity, carcinogenicity, and reproductive toxicity endpoints. The draft guidance raises the bar for how chemical characterization must be performed — but it does not change the fundamental advantage: a well-executed chemical characterization program addresses multiple biological endpoints simultaneously and avoids long-duration, expensive animal studies.

Key FDA Positions

Chemical characterization is expected. The FDA has been clear since 2016: chemical characterization is not optional for most devices. The agency expects manufacturers to characterize the chemical composition of their devices and perform extractables/leachables studies as a foundation for biological evaluation. Testing alone — without understanding what the device releases — is no longer considered adequate.

Risk-based approach is endorsed. The FDA supports the risk-based framework of ISO 10993-1:2018, including the use of toxicological risk assessment to reduce or eliminate biological testing. However, the agency expects the risk assessment to be thorough and performed by qualified toxicologists.

The biocompatibility matrix is a starting point. The FDA uses the ISO 10993-1 matrix as the baseline for determining which endpoints to evaluate, but reserves the right to request additional testing based on the specific device, materials, or patient population.

Extraction conditions matter. The FDA's guidance specifies exaggerated extraction conditions (per ISO 10993-12) and expects polar and non-polar extraction vehicles to be used. The agency has been known to issue AI requests when manufacturers use only a single extraction vehicle or insufficiently aggressive conditions.

Device-specific guidance may apply. For certain device types, the FDA has published device-specific biocompatibility guidance. For example, the contact lens guidance has specific biocompatibility requirements beyond the general ISO 10993 framework.

FDA Expectations for 510(k) Submissions

For a 510(k) submission, the FDA expects:

Element	What FDA Expects
BEP	Documentation of the evaluation plan, including device categorization, endpoint identification, and rationale for the chosen approach
BER	Comprehensive report integrating all evidence, with conclusions signed by a qualified toxicologist
Device categorization	Clear identification of body contact type and duration, with rationale
Chemical characterization	Extractables/leachables data for most devices, especially those with prolonged or permanent contact
Toxicological risk assessment	For any identified extractables/leachables — with MoS calculations per ISO 10993-17
Test reports	Full GLP-compliant test reports for any biological testing performed (not just summaries)
Material equivalence (if claimed)	Detailed comparison demonstrating identical material composition, grade, supplier, and processing
Literature review	Systematic review of relevant published data supporting biocompatibility conclusions

Common FDA 510(k) Deficiencies Related to Biocompatibility

Biocompatibility deficiencies are among the most frequently cited reasons for FDA Additional Information (AI) requests. Understanding these patterns can save months of delay.

Deficiency	Description	How to Avoid
Missing or incomplete BEP/BER	No formal evaluation plan or report; endpoints not systematically addressed	Prepare a complete BEP before testing and a BER that integrates all evidence
Incorrect device categorization	Wrong body contact category or duration; failure to account for cumulative exposure	Carefully analyze all contact scenarios; consider worst-case cumulative use
Missing chemical characterization	No extractables/leachables data, particularly for prolonged/permanent contact devices	Perform chemical characterization per ISO 10993-18 as the foundation of your evaluation
Incomplete extraction conditions	Using only one extraction vehicle; extraction conditions not per ISO 10993-12; insufficient extraction time or temperature	Use polar and non-polar vehicles; follow ISO 10993-12 exaggerated conditions
No toxicological risk assessment	Chemical characterization performed but no TRA linking extractables to biological endpoints	Engage a qualified toxicologist to perform MoS calculations per ISO 10993-17
Insufficient material equivalence justification	Claiming equivalence to a predicate without demonstrating identical material composition	Provide material certificates, supplier documentation, and analytical data
Missing endpoints	Not all matrix-required endpoints addressed; no justification for omissions	Systematically address every endpoint in the matrix — either through testing, TRA, or justified omission
Non-GLP testing	Biological tests not performed under GLP (Good Laboratory Practice) conditions	Ensure all biological testing is GLP-compliant per 21 CFR Part 58
Outdated test standards	Testing performed to superseded versions of ISO 10993 parts	Use current standard editions; verify before testing begins
Sterilization residual not addressed	EtO-sterilized devices without ISO 10993-7 residual analysis	If using EtO sterilization, include 10993-7 residual data in the BER

Practical tip: Before submitting, compare your biocompatibility section against the FDA's biocompatibility guidance document and the relevant device-specific guidance (if one exists). Walk through each requirement systematically. An hour spent on this self-audit can prevent a 90-day AI request cycle.

Relationship to EU MDR General Safety and Performance Requirements

Under the EU Medical Device Regulation (MDR 2017/745), biocompatibility is addressed through the General Safety and Performance Requirements (GSPRs) in Annex I.

Relevant GSPRs

GSPR	Requirement	Relationship to ISO 10993
GSPR 10.1	Devices shall be designed and manufactured in such a way as to ensure that the characteristics and performance requirements are fulfilled. Particular attention shall be paid to the choice of materials and substances, particularly as regards toxicity.	Requires biological evaluation of all patient-contacting materials
GSPR 10.4	Devices shall be designed and manufactured in such a way as to reduce as far as possible risks posed by substances or particles that may be released from the device, including wear debris, degradation products, and processing residuals.	Drives extractables/leachables and degradation studies per ISO 10993-18, -9, -13, -14, -15
GSPR 10.4.1 (Subclause on CMR/ED substances)	Devices or device parts that are invasive and contact the body, or (re)administer substances to/from the body, shall only contain CMR (carcinogenic, mutagenic, reprotoxic) or endocrine-disrupting substances above 0.1% w/w if adequately justified	Requires specific assessment of CMR and endocrine-disrupting substances — goes beyond standard ISO 10993 requirements
GSPR 11	Devices and their manufacturing processes shall be designed to eliminate or reduce as far as possible the risk of infection to patients, users, and third persons.	Intersects with biocompatibility via sterilization residuals and material contamination

EU MDR-Specific Considerations

The EU MDR imposes requirements that go beyond what ISO 10993 alone covers:

CMR and endocrine-disrupting substances. GSPR 10.4.1 requires manufacturers to identify and justify the presence of substances that are carcinogenic, mutagenic, reprotoxic (CMR Category 1A or 1B under CLP Regulation), or endocrine-disrupting above 0.1% w/w concentration in invasive devices. This requires explicit material composition analysis and comparison against the ECHA Candidate List and CLP classifications. ISO 10993-18 chemical characterization supports this but must be supplemented with specific CMR/ED substance analysis.

Phthalates. The MDR specifically calls out devices containing phthalates classified as CMR. If your device contains DEHP or other classified phthalates (common in PVC-based medical tubing and blood bags), you must label the device accordingly and justify the use in your technical documentation.

Nanomaterials. The MDR includes specific provisions for devices containing or releasing nanomaterials. Biological evaluation of nanomaterials may require additional endpoints and considerations beyond the standard ISO 10993 framework.

Notified Body expectations. Notified Bodies reviewing CE marking applications scrutinize the BER as a core element of the technical documentation. They expect to see a complete biological evaluation per ISO 10993-1:2018, chemical characterization per ISO 10993-18, and specific attention to CMR/ED substances per GSPR 10.4.1. Notified Bodies have become increasingly rigorous in their review of biocompatibility documentation since the MDR came into full effect.

Transition to ISO 10993-1:2025 in the EU. Notified Bodies in the EU are expected to transition to ISO 10993-1:2025 as the applicable harmonized standard for biocompatibility evaluation. The timeline for formal harmonization under the MDR (listing in the Official Journal of the European Union) is pending, but manufacturers with upcoming Notified Body audits should begin assessing the impact of the 2025 edition on their existing BERs. Key areas likely to require BER updates include: recategorization of devices under the new four-category system, reassessment of exposure duration using the "Contact Day" concept, evaluation of reasonably foreseeable misuse scenarios, expanded genotoxicity evaluation for prolonged mucosal contact devices, and lifecycle evaluation documentation.

Testing Labs and Costs

GLP Requirements: What Must Be GLP-Compliant and What Does Not

Understanding when Good Laboratory Practice (GLP) regulations apply — and when they do not — is essential for planning a biocompatibility testing program efficiently.

21 CFR Part 58 Requirements for Medical Device Biocompatibility

GLP regulations (21 CFR Part 58 in the US; OECD GLP principles internationally) apply to nonclinical laboratory studies conducted to support regulatory submissions. For medical devices, this means:

Study Type	GLP Required?	Notes
In vivo biological tests (sensitization, implantation, systemic toxicity, pyrogenicity, hemocompatibility)	Yes	FDA requires GLP for PMA and IDE submissions; strongly prefers GLP for 510(k)
In vitro biological tests (cytotoxicity, genotoxicity battery, hemolysis, MAT)	Yes	Even though these are bench tests, they are nonclinical safety studies under 21 CFR Part 58
Chemical characterization / E&L studies (ISO 10993-18)	Generally no — ISO 17025 accreditation is the applicable quality standard	However, if chemical characterization data is used to directly support biocompatibility conclusions (substituting for biological testing), some FDA reviewers expect GLP-like documentation rigor
Exploratory / feasibility studies	No	Basic studies to determine whether a device has potential utility or to characterize physical/chemical properties are explicitly exempt
Routine lot release testing	No	Governed by GMP (21 CFR 820), not GLP
Sterilization validation studies	No	Governed by relevant sterilization standards and GMP

Key GLP Compliance Elements

For studies that require GLP, 21 CFR Part 58 mandates:

• Named Study Director — A single individual responsible for the technical conduct of the study and the final report. The Study Director must be located at the testing facility.
• Quality Assurance Unit (QAU) — An independent function that inspects critical study phases and audits the final report. The QAU must be independent of the study conduct — the Study Director cannot serve as QA for their own study.
• Written protocols and amendments — The study protocol must be approved before study initiation. Any changes during the study must be documented as formal amendments with justification.
• Test article characterization — The test article (finished device) must be characterized by identity, composition, lot number, and stability. For medical devices, this means documenting materials, manufacturing lot, sterilization method, and shelf-life status.
• Raw data integrity (ALCOA+ principles) — All raw data must be Attributable, Legible, Contemporaneous, Original, and Accurate, plus complete, consistent, enduring, and available.
• Final report requirements — 21 CFR Part 58 specifies the minimum contents of a final study report, including identification of the Study Director and all contributing scientists, dates of study phases, description of methods, statistical analysis, and conclusions.
• Archive — All raw data, protocols, reports, and specimens must be archived and retrievable for required retention periods.

GLP vs. ISO 17025: Understanding the Distinction

Aspect	GLP (21 CFR Part 58 / OECD)	ISO/IEC 17025
Primary purpose	Ensure reliability of safety data submitted to regulatory agencies	Demonstrate technical competence of testing laboratories
Applicable to	Nonclinical safety studies (in vivo and in vitro biocompatibility, toxicology)	Chemical and analytical testing (E&L studies, material characterization, sterilization residuals)
FDA recognition	Required for biological safety studies	Not formally recognized by FDA for biological safety, but accepted for chemical characterization
Key requirement	Study Director, QAU, protocol-driven study conduct	Method validation, measurement uncertainty, proficiency testing
Accreditation	Monitored through FDA BIMO (Bioresearch Monitoring) inspections	Accredited by accreditation bodies (A2LA, NVLAP, UKAS, DAkkS)

Important: The FDA does not consider ISO 17025 accreditation sufficient for biological safety studies. A lab accredited to ISO 17025 but not GLP-compliant cannot conduct biocompatibility testing for FDA submissions. Conversely, GLP compliance alone does not demonstrate analytical competence for chemical characterization — ISO 17025 accreditation is the appropriate credential for E&L studies.

FDA ASCA Pilot Program

The FDA's Accreditation Scheme for Conformity Assessment (ASCA) pilot program includes biocompatibility testing laboratories. Under ASCA, accreditation bodies (such as A2LA) perform laboratory assessments that encompass both in vitro and in vivo biocompatibility testing. Labs assessed under ASCA may have their biocompatibility data accepted by the FDA with reduced review burden, potentially accelerating submission timelines. Manufacturers should inquire whether their testing lab participates in the ASCA program.

Choosing a Biocompatibility Testing Lab

Select a lab that meets these criteria:

• GLP compliance for biological testing — The lab must operate under Good Laboratory Practice regulations (21 CFR Part 58 in the US, OECD GLP principles internationally). Non-GLP biological test data will be rejected by the FDA for PMA submissions and questioned for 510(k) submissions.
• ISO 17025 accreditation for chemical characterization — For extractables/leachables studies, material characterization, and analytical testing. Accreditation should cover the specific test methods being performed (GC-MS, LC-MS, ICP-MS).
• Experience with medical devices — Labs that specialize in pharmaceutical testing may not understand the nuances of device extraction (ISO 10993-12) and device-specific endpoints. Ask for references from medical device clients and inquire about the lab's experience with your specific device type and materials.
• Regulatory familiarity — The lab should understand FDA and EU MDR expectations and flag potential issues proactively. Ask whether the lab's studies have been the subject of FDA AI requests or Notified Body findings.
• Toxicology support — Some labs offer in-house toxicology services for risk assessment and BER preparation. This integration can save time and reduce errors.
• ASCA participation — Labs participating in the FDA's ASCA pilot program for biocompatibility may provide a streamlined path for FDA acceptance of test data.
• Capacity and turnaround time — Biocompatibility testing labs frequently have backlogs of 4-12 weeks before study initiation. Inquire about current capacity and expected start dates during lab selection, not after contracts are signed.

Lab Selection Due Diligence Checklist

Before committing to a biocompatibility testing lab, verify the following:

• Current GLP compliance status — request the lab's most recent GLP inspection report or FDA Form 483 history
• ISO 17025 accreditation certificate and scope of accredited methods (for chemical characterization labs)
• Named Study Directors with relevant experience for your study types
• Independent QAU staffing and procedures
• Archive facilities and data retention policies
• Recent FDA BIMO inspection results (for GLP labs) — check the FDA's BIMO inspection database
• Experience with your specific device type, materials, and target markets
• Ability to perform all required analytical methods in-house (vs. subcontracting, which adds time and introduces quality oversight complexity)
• Standard turnaround times and current backlog
• Insurance coverage and indemnification terms
• Communication protocols — will you have a dedicated project manager or single point of contact?

Major Biocompatibility Testing Laboratories

Several contract research organizations (CROs) specialize in biocompatibility testing for medical devices:

• Nelson Laboratories (Salt Lake City, UT) — One of the largest medical device testing labs globally
• WuXi AppTec (formerly NAMSA) (Northwood, OH / global) — Full-service biocompatibility and chemical characterization
• SGS (global) — Multinational testing, inspection, and certification company
• Eurofins (global) — Extensive chemical characterization and biological testing capabilities
• Toxikon (Bedford, MA) — Specializes in biocompatibility and preclinical testing
• Pacific Biolabs (Hercules, CA) — West Coast option for biological testing
• RQMIS / Intertek (global) — Full-service testing and regulatory consulting

Cost Estimates for Biocompatibility Testing

The following cost ranges are approximate and based on typical US CRO pricing as of 2025-2026. Costs vary significantly by lab, region, number of samples, and study complexity.

Test/Activity	Typical Cost Range	Timeline
Cytotoxicity (ISO 10993-5)	$1,500-3,500	2-4 weeks
Sensitization — LLNA	$5,000-10,000	3-5 weeks
Sensitization — GPMT	$8,000-15,000	6-10 weeks
Irritation — intracutaneous	$3,000-7,000	3-5 weeks
Irritation — in vitro	$2,000-5,000	2-4 weeks
Acute systemic toxicity	$5,000-12,000	3-5 weeks
Pyrogenicity (rabbit)	$3,000-6,000	2-4 weeks
Pyrogenicity (MAT in vitro)	$2,000-5,000	2-3 weeks
Hemolysis	$2,500-5,000	2-4 weeks
Genotoxicity battery (3 tests)	$15,000-30,000	8-12 weeks
Subchronic systemic toxicity (90-day)	$80,000-200,000	5-7 months
Chronic systemic toxicity	$150,000-500,000+	9-18 months
Implantation (short-term, 2-4 weeks)	$30,000-50,000	3-4 months
Implantation (long-term, 26 weeks)	$50,000-100,000	8-10 months
Implantation (52 weeks)	$80,000-150,000	14-16 months
Hemocompatibility (comprehensive)	$50,000-150,000	3-6 months
Extractables/leachables (ISO 10993-18)	$15,000-50,000	6-12 weeks
Toxicological risk assessment	$5,000-20,000	4-8 weeks
BEP + BER preparation	$10,000-30,000	4-12 weeks

Total Program Cost Estimates by Device Category

Device Category	Typical Total Cost	Key Driver
Surface-contacting, limited (e.g., EKG electrode)	$10,000-25,000	Basic triad (cytotoxicity, sensitization, irritation) + chemical characterization
Surface-contacting, permanent (e.g., hearing aid shell)	$30,000-60,000	Extended endpoint coverage including genotoxicity and subchronic toxicity assessment
External communicating, prolonged (e.g., urinary catheter)	$40,000-80,000	Broad endpoint coverage + chemical characterization + possible systemic toxicity
Blood-contacting, prolonged (e.g., central venous catheter)	$60,000-120,000	Hemocompatibility + systemic toxicity + genotoxicity + chemical characterization
Permanent implant, tissue (e.g., orthopedic screw)	$80,000-200,000	Implantation studies + genotoxicity + subchronic/chronic assessment + full chemical characterization
Permanent implant, blood (e.g., heart valve)	$150,000-500,000+	Full hemocompatibility + implantation + all systemic endpoints + comprehensive chemical characterization

These are rough ranges. Costs decrease significantly when chemical characterization and risk assessment can replace biological testing, and when material equivalence to a predicate can be demonstrated.

Timeline for a Biocompatibility Testing Program

Planning is essential. Biocompatibility testing is often on the critical path for regulatory submission, and delays in testing directly push out market entry.

Typical Timeline: Surface-Contacting or Simple External Communicating Device

Phase	Duration	Activities
BEP development	2-4 weeks	Device categorization, endpoint identification, test planning, lab selection
Sample preparation	2-4 weeks	Producing finished, sterilized test samples in representative final form
Chemical characterization	6-12 weeks	Extractables study, analytical testing, data interpretation
Biological testing	6-12 weeks	Cytotoxicity, sensitization, irritation, acute systemic toxicity (if applicable) — most run in parallel
Toxicological risk assessment	4-8 weeks	Can begin as chemical characterization data becomes available
BER preparation	4-8 weeks	Integration of all data, conclusion writing, toxicologist review
Total	4-8 months

Typical Timeline: Permanent Implant Device

Phase	Duration	Activities
BEP development	4-6 weeks	Complex categorization, comprehensive endpoint analysis
Sample preparation	4-8 weeks	Finished, sterilized implant samples; may require custom shapes for implantation studies
Chemical characterization	8-16 weeks	Comprehensive extractables in multiple vehicles, possible leachables study
Biological testing (short-term)	8-12 weeks	Cytotoxicity, sensitization, irritation, genotoxicity battery, hemolysis
Biological testing (long-term)	6-15 months	Implantation studies (12-52 weeks), subchronic/chronic toxicity if required
Hemocompatibility (if blood-contacting)	3-6 months	Full hemocompatibility evaluation
Toxicological risk assessment	6-10 weeks	Comprehensive assessment of all identified substances
BER preparation	6-12 weeks	Complex integration, multiple data packages
Total	12-24 months	Long-term implantation and chronic studies drive the timeline

Planning advice: Start biocompatibility evaluation early in the development cycle — ideally during design verification, not after design freeze. Material selection decisions made during design development have the largest impact on biocompatibility testing requirements and timeline. Choosing a well-characterized material with clinical history can save 6-12 months compared to a novel material requiring full testing.

Parallelization Strategy

Smart scheduling can compress timelines significantly:

• Chemical characterization and short-term biological tests run in parallel. Start cytotoxicity, sensitization, and irritation testing as soon as finished sterilized samples are available, while simultaneously running extractables studies.
• Genotoxicity battery can start immediately. It does not depend on chemical characterization results.
• Begin TRA as chemical data becomes available. Do not wait for all analytical testing to complete — start the risk assessment on early results and update as additional data arrives.
• Start BER writing while final tests are in progress. The framework, background sections, and chemical characterization discussion can be written while long-term studies are still ongoing.

Material Equivalence Arguments

Material equivalence is one of the most powerful tools for reducing biocompatibility testing — and one of the most commonly misapplied.

What Material Equivalence Requires

To claim material equivalence, you must demonstrate that:

1. Same material specification — Identical material grade, formulation, and specification (e.g., ASTM F138 for 316LVM stainless steel, ASTM F136 for Ti-6Al-4V ELI)
2. Same or equivalent supplier — Ideally the same material supplier. If a different supplier, you must demonstrate equivalent material composition through analytical testing.
3. Same or equivalent manufacturing process — Processing affects biocompatibility. A material that is safe when injection molded may release different extractables when 3D printed or machined.
4. Same or equivalent sterilization — Different sterilization methods (EtO, gamma, e-beam, steam) can affect material chemistry and extractables profiles.
5. Same or equivalent geometry and surface area — A device with a larger surface area may release more extractables per unit time.
6. Same or less aggressive contact conditions — Contact duration and tissue type should be equal or less aggressive than the reference device.

Documentation Requirements

A material equivalence argument must include:

• Material certificates of analysis (CoA) for both the reference and new device
• Supplier documentation confirming material identity and grade
• Analytical comparison data (FTIR, DSC, or other appropriate characterization) if suppliers differ
• Manufacturing process comparison
• Sterilization method comparison
• Surface area and geometry comparison
• Reference to the biocompatibility data for the equivalent device — with evidence that this data exists and is adequate

Common Pitfalls

• "Same material" is not enough. Saying "both devices use PEEK" without specifying the exact grade, supplier, and any additives or colorants is insufficient. PEEK from different suppliers, or with different colorant packages, can have very different extractables profiles.
• Manufacturing process changes everything. Two devices made from the same raw material can have different biological risk profiles if one is injection molded and the other is machined, because machining introduces cutting fluids and surface characteristics that injection molding does not.
• Sterilization is not interchangeable. EtO sterilization leaves residuals (addressed by ISO 10993-7). Gamma irradiation can cause polymer chain scission and generate new extractables. Steam sterilization can hydrolyze certain materials. A material equivalence argument that ignores sterilization differences is incomplete.

Biocompatibility for Combination Products

Combination products — devices combined with drugs or biologics — present unique biocompatibility challenges.

Drug-Device Combinations

For drug-eluting devices (drug-eluting stents, antibiotic-loaded bone cement, hormonal IUDs), the biological evaluation must address both the device materials and the drug substance. The drug itself is evaluated through pharmaceutical toxicology standards, but the interaction between the drug and device materials, and any novel extractables created by the combination, falls under the biocompatibility evaluation.

Key Considerations

• Extraction studies must account for the drug substance. The drug may appear as an extractable, complicating analytical chemistry.
• Drug-material interactions may generate degradation products not present in either component alone.
• Regulatory pathway affects expectations. In the US, combination products are assigned a primary mode of action (PMOA) that determines the lead review center (CDER, CBER, or CDRH). The biocompatibility expectations may differ depending on which center leads the review.

Special Considerations

Reprocessed Single-Use Devices

Devices originally marketed as single-use but reprocessed for reuse must undergo biocompatibility re-evaluation. Reprocessing (cleaning, sterilization) can alter material properties, generate new extractables, and change surface characteristics. The biological evaluation must address the reprocessed device — not just the original.

3D-Printed (Additively Manufactured) Devices

Additive manufacturing introduces unique biocompatibility considerations:

• Residual powders and support materials may not be fully removed by post-processing
• Layer-by-layer fabrication creates surface characteristics different from conventionally manufactured materials
• Post-processing (sintering, HIP, surface finishing) significantly affects extractables profiles
• Material equivalence arguments are difficult — the same alloy produced by casting versus laser powder bed fusion may have different microstructure, porosity, and surface chemistry

The FDA has published guidance on additively manufactured devices that addresses biocompatibility considerations specific to these manufacturing methods.

Devices with Nanomaterials

Nanomaterials (particles, fibers, coatings with features less than 100 nm in at least one dimension) may exhibit biological behavior different from their bulk counterparts due to their high surface area-to-volume ratio. Standard ISO 10993 testing may not be sufficient — additional characterization of particle size, shape, surface area, surface chemistry, and agglomeration behavior may be needed. The FDA and SCENIHR (EU) have published guidance on nanomaterial-containing medical devices.

Absorbable and Biodegradable Devices

Devices designed to degrade in the body (absorbable sutures, bioresorbable scaffolds, drug-delivery implants) require evaluation of degradation products in addition to the intact device. ISO 10993-9 provides the framework for identifying and quantifying degradation products. The biological evaluation must address the toxicological risk of degradation products at each stage of the degradation process — not just the initial device composition.

Putting It All Together: A Practical Workflow

Phase 1: Material Selection and Early Risk Assessment (During Design Development)

• Identify all patient-contacting materials
• Perform preliminary risk assessment: is the material well-characterized with clinical history?
• Check published literature and material databases for existing biocompatibility data
• Identify potential material equivalence to predicate or reference devices
• Flag novel materials or novel manufacturing processes that will require full testing
• Estimate timeline and budget for biocompatibility evaluation

Phase 2: BEP Development (During Design Verification)

• Finalize device categorization (body contact, duration)
• Document all materials, manufacturing processes, and sterilization method
• Identify all required biological endpoints per ISO 10993-1 matrix
• Determine the approach for each endpoint: testing, chemical characterization + TRA, literature, equivalence
• Select testing lab(s) and obtain quotes
• Document everything in the BEP

Phase 3: Chemical Characterization (After Design Freeze)

• Prepare finished, sterilized samples per ISO 10993-12
• Conduct extractables studies (polar and non-polar vehicles)
• Conduct leachables studies if needed
• Begin toxicological risk assessment as data becomes available

Phase 4: Biological Testing (In Parallel with Phase 3)

• Submit samples to testing lab(s)
• Execute GLP-compliant biological tests per the BEP
• Monitor test progress; address any anomalous results promptly

Phase 5: BER Preparation (As Data Completes)

• Integrate all data: chemical characterization, TRA, biological test reports, literature review
• Perform gap analysis: are all endpoints adequately addressed?
• Draft conclusions: is the device biocompatible for its intended use?
• Qualified toxicologist reviews and signs the BER
• Incorporate BER into regulatory submission (510(k), technical documentation, etc.)

Phase 6: Post-Market Maintenance

Biocompatibility evaluation is not a one-time activity. You must reassess when:

• Materials or suppliers change
• Manufacturing processes change
• Sterilization method changes
• New toxicological data emerges for materials or extractables
• Post-market surveillance reveals adverse biological events
• Regulatory requirements change (new standard editions, new guidance)

Document reassessments and update the BER accordingly.

Frequently Asked Questions

Do I need biocompatibility testing if my device only contacts intact skin?

You still need a biological evaluation, but the testing burden is typically low. Skin-contacting devices with limited duration often require only the basic triad: cytotoxicity, sensitization, and irritation. If the material is well-characterized (e.g., medical-grade stainless steel, established medical-grade polymers), even these tests may be addressed through literature and material equivalence rather than new testing. You still need a BEP and BER documenting this rationale.

Can I use biocompatibility data from my material supplier?

Possibly, but with significant caveats. Supplier data on the raw material may not reflect the finished device, which has been through manufacturing processes (molding, machining, welding, coating, cleaning) and sterilization that can alter the extractables profile. Supplier data is useful as supporting evidence but is rarely sufficient as the sole basis for a biocompatibility conclusion on a finished device. The evaluation must be performed on the finished, sterilized device in its final form.

My device is made of well-known materials (stainless steel, titanium, PEEK). Do I still need chemical characterization?

Generally, yes — but the scope may be limited. Even well-known materials can contain unexpected extractables from manufacturing processes (cutting fluids, cleaning residues, surface treatments, coatings, marking inks). A targeted chemical characterization study focused on process-related extractables may be sufficient, rather than a comprehensive multi-vehicle extraction. Document your rationale in the BEP.

How long are biocompatibility test results valid?

ISO 10993 does not specify an expiration date for test data. However, the data is valid only as long as the device design, materials, manufacturing process, and sterilization remain unchanged. Any change to these factors requires re-evaluation (though not necessarily re-testing — a risk assessment of the change may suffice). As a practical matter, test data older than 5-10 years may draw questions from reviewers, particularly if testing was performed to superseded standard editions.

What if a test fails?

A failed biocompatibility test does not necessarily mean the device cannot go to market, but it does require investigation. Common responses include:

• Root cause analysis — Identify why the test failed. Was it a material issue, a manufacturing contaminant, a test artifact, or a genuine biological hazard?
• Repeat testing — If the failure is suspected to be an artifact, repeating the test with proper controls may be justified. Document the rationale for repeating.
• Material or process change — If the failure is material-related, modify the material or manufacturing process and retest.
• Risk-benefit analysis — In rare cases (typically for high-benefit implants), a device with a marginally positive test result may still be acceptable if the clinical benefit outweighs the biological risk and the risk is disclosed in labeling. This requires strong justification and regulatory discussion.

Does the FDA accept in vitro alternatives to animal testing?

Increasingly, yes. The FDA supports the 3Rs principles and has accepted in vitro alternatives for several endpoints:

• Irritation — In vitro reconstructed human epidermis models (EpiDerm, SkinEthic) are accepted for skin irritation assessment
• Pyrogenicity — The Monocyte Activation Test (MAT) is accepted as an alternative to the rabbit pyrogen test
• Sensitization — The LLNA (while still using animals) uses fewer animals than the GPMT; fully in vitro sensitization assays (DPRA, KeratinoSens, h-CLAT) are gaining acceptance as part of defined approaches
• Cytotoxicity — Already an in vitro test by definition

For other endpoints (systemic toxicity, implantation, hemocompatibility), in vivo testing remains the standard when testing is required — though chemical characterization with TRA can often eliminate the need for these tests entirely.

Defined approaches for sensitization (2024-2025 developments): The most significant recent advancement is the development of "defined approaches" — standardized combinations of in vitro sensitization assays (e.g., DPRA + KeratinoSens + h-CLAT) with fixed data interpretation procedures. These defined approaches are described in OECD Guideline 497 and referenced in ISO 10993-10:2021 Annex C. The EU regulatory framework is increasingly accepting defined approaches for skin-contacting medical devices, and ISO TC 194/WG 8 is actively working to incorporate validated in vitro methods into the normative (mandatory) section of ISO 10993-10. GARD (Genomic Allergen Rapid Detection) was added to ISO 10993-10 Annex C in 2021 and is under evaluation for normative inclusion.

For manufacturers seeking to avoid animal sensitization testing, the most defensible current approach is to perform a defined approach combining two or three in vitro assays (covering at least two different key events in the AOP) and to include this data alongside a thorough chemical characterization and toxicological risk assessment. This combined evidence package may support a conclusion of negligible sensitization risk without GPMT or LLNA testing — but discuss this strategy with your testing lab and toxicologist before committing, as regulatory acceptance varies by jurisdiction and device risk class.

Navigating US-International Regulatory Divergence (2025-2026)

The publication of ISO 10993-1:2025 and the FDA's September 2024 draft guidance on chemical analysis have created a period of regulatory divergence that manufacturers pursuing global market access must navigate carefully.

The Current Landscape

Jurisdiction	Applicable Standard	Chemical Characterization Guidance	Key Implication
United States (FDA)	ISO 10993-1:2018 (recognized consensus standard)	FDA's 2023 final guidance + September 2024 draft guidance on chemical analysis	Continue using 2018 edition; prepare for enhanced chemical characterization requirements per the draft guidance
European Union (MDR)	ISO 10993-1:2018 (transitioning to 2025)	ISO 10993-18:2020	Begin gap assessment for 2025 edition; address CMR/ED substance requirements per GSPR 10.4.1
Japan (PMDA)	Typically follows ISO with national deviations	JIS T 0993-1 (harmonized with ISO 10993-1)	Monitor PMDA adoption timeline for 2025 edition
China (NMPA)	GB/T 16886.1 (equivalent to ISO 10993-1:2018)	National standards aligned with ISO 10993-18	Chinese adoption of the 2025 edition may lag; maintain 2018-based documentation for current submissions

Dual-Track Strategy for Global Submissions

Manufacturers targeting both the US and EU markets should consider a dual-track approach:

1. Maintain a BER compliant with ISO 10993-1:2018 for US FDA submissions. This remains the recognized standard, and the FDA has given no timeline for recognizing the 2025 edition.
2. Prepare a supplemental assessment addressing ISO 10993-1:2025 changes for EU submissions. Key additions include: recategorization under the new four-table system, Contact Day exposure calculation, reasonably foreseeable misuse evaluation, expanded genotoxicity requirements, and lifecycle documentation.
3. Structure chemical characterization studies to satisfy both the FDA's draft guidance and ISO 10993-18. In practice, this means performing triplicate extraction, using both polar and non-polar harsh solvents, pursuing confirmed-level chemical identifications, and documenting exhaustive extraction — the FDA's requirements are generally more demanding than ISO 10993-18 alone.

This dual-track approach adds documentation burden but avoids the risk of having a BER that satisfies one jurisdiction but not the other.

Summary: Key Takeaways

ISO 10993 biocompatibility evaluation is not about running every test in the matrix. It is a risk-based, evidence-driven process that integrates chemical characterization, toxicological risk assessment, biological testing, literature review, and clinical data to determine whether a device poses unacceptable biological risk.

The modern approach, endorsed by both ISO 10993-1:2018 and the FDA, follows a clear hierarchy: characterize the chemistry first, assess the risk toxicologically, test only what cannot be resolved by other means, and document everything in a BEP and BER reviewed by a qualified toxicologist.

Getting this right — early in the development process, with proper planning and qualified expertise — prevents the most common regulatory delays. Getting it wrong — treating biocompatibility as an afterthought, skipping chemical characterization, or submitting incomplete documentation — reliably results in AI requests, Notified Body findings, and months of lost time.

Start early. Plan thoroughly. Characterize before you test. Document everything. Engage a toxicologist. These five principles will carry you through any biocompatibility evaluation, for any device, in any market.