Skip to main content
EU Science Hub

Eye irritation is the production of changes in the eye following the application of a test substance to the anterior surface of the eye, which are fully reversible within 21 days of application (OECD Test Guideline 405, in vivo).

The EU has implemented the United Nations Globally Harmonized System (UN GHS) for the classification and labelling of hazardous chemicals with Directive 1272/2008 (EU CLP). The guidelines are provided in UN GHS and adopted in the EU CLP.

In brief, four representative in vivo endpoint(s) for final classification (corneal opacity, iritis, conjunctiva chemosis, conjunctiva redness) receive mean scores calculated following grading at 24, 48 and 72 hours after installation of the test material to the eye sack of three different animals.

The test material is classified as "Category 1" when at least in one animal the effects caused by the application of the test material are not expected to reverse, or have not fully reversed within an observation period of normally 21 days.

Otherwise, if all effects have fully reversed within an observation period of normally 21 days, the test material is classified as "Category 2". UN GHS provides the option to distinguish "Category 2" test materials into two optional sub-categories (not implemented in EU CLP):

  • ‘Category 2A’ (irritating to eyes) when the eye effects listed above are not fully reversible within 7 days of observation;
  • ‘Category 2B’ (mildly irritating to eyes) when the eye effects listed above are fully reversible within 7 days of observation.

Test materials that fall in none of the above two categories are not classified as eye irritants.

A number of OECD Test Guidelines based on in vitro methods related to the endpoint "Eye Irritation" already exist or are currently close to their final endorsement/update: 437, 438, 460. In addition, in Test Guideline 405, a stepwise testing strategy is described for the determination of the eye irritation/corrosion properties of substances.

Using this strategy, as well as a weight-of-evidence analysis (where all available information on the eye irritation potential is considered prior to proceeding to in vivo testing) is important to avoid the unnecessary use of animals. Reducing substance testing resulting in severe animal responses promotes both animal welfare and sound science.

Although TG 405 is an in vivo test method, it also supports the 3Rs principle in reduction and refinement of animal testing.

EURL ECVAM validated test methods

In the endpoint of eye irritation it is generally accepted that, in the foreseeable future, no single in vitro eye irritation test will be able to replace the in vivo Draize eye test to predict across the full range of irritation for different chemical classes. However, strategic combinations of several alternative test methods within a (tiered) testing strategy may be able to replace the Draize eye test.

A possible conceptual framework for such a (tiered) testing strategy was developed within an ECVAM workshop (Scott et al., 2010). The framework is based on alternative eye irritation methods that vary in their capacity to detect either severe irritant substances (EU R41; GHS 'Category 1') or substances considered non-irritant (EU 'Non-Classified'; GHS 'No Category').

According to this framework, the entire range of irritancy may be resolved by arranging tests in a tiered strategy that may be operated from either end: to detect first severe irritants and resolve absence of irritancy ("Top-Down Approach") or to proceed inversely, starting with the identification of non-irritants first ("Bottom-Up Approach"). Mild irritancy will be resolved in a last tier in both approaches.

To evaluate the scientific validity of possible building blocks of such a test strategy and to assess their possible placement within a Bottom-Up and Top-Down Approach, several validation studies of different in vitro methods were undertaken or are ongoing:

  1. Organotypic in vitro assays: Bovine Corneal Opacity & Permeability test (BCOP), Isolated Chicken Eye test (ICE), Isolated Rabbit Eye test (IRE), Hens Egg Test on the Chorio-Allantoic Membrane (HET-CAM).
  2. Cytotoxicity and cell function-based assays: Cytosensor Microphysiometer (CM), Fluorescein Leakage (FL), Neutral Red Release (NRR), Red Blood Cell test (RBC).

Organotypic in vitro assays

A retrospective validation study on four organotypic assays, i.e. the Bovine Corneal Opacity & Permeability test (BCOP), the Isolated Chicken Eye test (ICE), the Isolated Rabbit Eye test (IRE) and the Hens Egg Test on the Chorio-Allantoic Membrane (HET-CAM), was conducted between 2003 and 2006 by ICCVAM-NICEATM, with ECVAM collaboration, to assess the ability of these assays for detecting severe eye irritants and ocular corrosives. After peer review, two assays, the BCOP and ICE tests, were endorsed as scientifically valid to identify ocular corrosives and severe irritants in the US and EU.

Test guidelines for these two test methods were adopted by OECD Council on 7 September, 2009.

With regard to the evaluation of the four organotypic assays for identifying mild or non irritants, a retrospective analysis of the collected data was conducted by ICCVAM. In May 2009, the ICCVAM Ocular Peer Review Panel recommended BCOP for the identification of non-classified materials .

Much more information is available on these methods:

Cytotoxicity and cell function-based assays

The retrospective validation study of four cytotoxicity and cell function-based assays, i.e. the Neutral Red Release (NRR), the Red Blood Cell test (RBC), the Fluorescein Leakage assay (FL) and the Cytosensor Microphysiometer (CM), took place between May 2006 and October 2008. The study was based on the retrospective collection of existing data compiled according to the ECVAM modular approach to validation and weight-of-evidence principles.

After peer review, two assays, the CM and the FL, were scientifically validated for being used as an initial step within a Top-Down approach to identify ocular corrosives and severe irritants from all other classes for the chemical applicability domain of water soluble chemicals.

In addition, the CM was scientifically validated for being used as an initial step within a Bottom-Up approach useful to identify non-irritants from all other classes for water-soluble surfactants, and water-soluble surfactant-containing mixtures.

Much more information is available on these methods:

Reconstructed human tissue assays

Two reconstructed human tissue-based test methods, the SkinEthic Human Corneal Epithelium (HCE) assay and the EpiOcular Eye IrritationTest (EIT) were subjected to validation studies.

More information on the EpiOcular Eye Irritation Test (EIT), which was considered validated following these studies is available here.

Otherwise, you can find more information on the validation studies below.

SkinEthic is based on 3D constructs prepared from immortalized human corneal epithelial cells while EpiOcular uses a 3D construct prepared from non-transformed, human-derived epidermal keratinocytes. These construct tissues are used to test, under different conditions, chemicals for the assessment of ocular toxicity.

They are both advanced in vitro methods currently undergoing validation by EURL ECVAM (Eskes et al., 2005). Notably, EURL ECVAM and COLIPA are jointly sponsoring an EURL ECVAM‑co‑ordinated full prospective validation study on two reconstructed tissue models.

Although these assays appear to be promising for specific purposes and applicability domains, they may not fully address the relevant mechanisms of ocular toxicity. Some examples include the reversibility of effects due to tissue remodelling following inflammation.

To achieve full replacement of the animal test, the use of tiered test strategies has been recommended, which exploit the strengths of particular in vitro assays, in order to address the required ranges of irritation potential, chemical classes and/or mechanisms of action.

In 2010 ECVAM has, together with external experts, published a simple approach of aligning test methods in a strategic manner. This so‑called bottom‑up/top‑down approach describes how to use specific eye irritation test methods for either identifying serious eye irritants from the rest of the chemicals or for identifying non‑irritants (i.e., chemicals that do not require classification) from the rest of the population (Scott et al., 2010).

ECVAM is promoting this approach as means of a testing strategy using various in vitro methods which may in their totality and if used in this strategic manner allow replacing the Draize animal test. It was moreover recommended to advance the development of mechanistically‑based models, in order to address the currently existing mechanistic gaps.

The ICCVAM/ECVAM symposium on Mechanisms of Chemically-Induced Ocular Injury and Recovery, held in May 2005, identified the following aspects of ocular irritation where further investigation and development would be helpful:

  • the assessment of depth of injury as potential indicator of irreversibility/reversibility of effects (Maurer et al., 2002);
  • the use of complex human models (e.g., 3D human corneal tissue reconstructs); and
  • the use of inflammatory mediators (e.g., cytokines, adhesion proteins) as potential markers and predictors of ocular irritation.

In addition, investigation of the potential effects on the ocular nervous system could be of value. Such an endpoint is not always identifiable in the animal test, although it is important to avoid pain and discomfort during clinical trials and to the end consumer. In addition, it might be possible to gain an insight into the neurogenic components of inflammation, which could contribute to the general chemically-induced inflammatory response of the cornea (Garle and Fray, 2003).

Currently, the COLIPA eye irritation research programme addresses some of the above mentioned mechanistic features, by supporting the development of mechanistically based assays, such as 3D human reconstructs and the assessment of inflammatory mediators and toxicogenomics as biomarkers to predict eye irritation. In addition, various isolated efforts exist within the scientific community addressing the same endpoints.

EURL ECVAM is following such efforts, to ensure that the most mechanistically relevant assays proceed toward validation according to internationally agreed principles, and eventually lead, most likely in combination with other assays, to the full replacement of specific animal tests, in particular the Draize-test (Bouvier d’Yvoire et al., 2012).

In July 2009, ESAC endorsed the CM (INVITTOX Protocol 102 modified) as scientifically valid for use as an initial step within a top-down approach to identify ocular corrosives and severe irritants (GHS Cat 1, EU R41, EPA Cat I) from all other classes, for water-soluble chemicals (substances and mixtures).

Furthermore, the CM (INVITTOX Protocol 102 modified) was considered to have been scientifically validated for use as an initial step within a bottom-up approach to identify chemicals not classified as irritants (GHS no category, EU no category, 19 EPA Cat IV) from all irritant classes, only for water-soluble surfactants, and water-soluble surfactant containing mixtures.

Regarding the remaining two of the four organotypic assays, the Hen’s Egg Test on the Chorio-allantoic Membrane (HET-CAM) assay and the Isolated Rabbit Eye (IRE) test, they were evaluated by ESAC who requested that further work was to be performed before a statement on their scientific validity to identify ocular corrosives and severe irritants could be made.

In vivo refinement test method

The retrospective validation study of the refinement/reduction Low Volume Eye Test (LVET) method for the use domain of household detergents and cleaning products as well as their main ingredient classes took place between 2006 and 2009.

After peer review, the LVET was not recommended for prospective use, i.e. to generate new data but it was recommended that existing LVET data of the limited use domain of household detergents and cleaning products as well as their main ingredient classes may be used for purposes of classification and labeling decisions.

Moreover, it was recommended that existing LVET data of this limited use domain may be used as supplementary data for future validation studies. No additional testing should be however performed to further develop or validate the LVET test.

More information is available on LVET here.

EURL ECVAM template for in vivo eye irritation classification of chemicals with existing Draize data to assist the development, optimisation and validation of alternative test methods or strategies

The Excel template and accompanying explanatory document available for download below were developed at EURL ECVAM. The template includes built-in algorithms for calculating eye irritation classifications from Draize in vivo eye irritation data, according to the rules of the classification systems of the European Union Dangerous Substances Directive (EU DSD), the European Union Regulation on Classification, Labelling and Packaging (EU CLP), the United Nations Globally Harmonized System of Classification and Labelling of Chemicals (UN GHS) and of the United States Environmental Protection Agency (US EPA).

The explanatory document provides detailed instructions on how to use the template. EURL ECVAM strongly recommends the use of this template to generate eye irritation classifications from existing reference Draize in vivo data for chemicals that shall be used in the development, optimisation and validation of alternative test methods or strategies for eye irritation.

For more information on the template and to download the files, go here.

Development and optimisation of alternative methods

Having been occasionally the lead organisation for the drafting and regulatory acceptance of several Test Guidelines (405 (in vivo), 437, 438, 460, draft TG on Cytosensor Microphysiometer) on in vitro eye irritation, EURL ECVAM is continuing activities in collaboration with the OECD test guidelines programme in view of improving the scientific use of in vitro eye irritation test methods and their standardisation.