In silico prediction of the full United Nations Globally Harmonized System eye irritation categories of liquid chemicals by IATA-like bottom-up approach of random forest method

As an alternative to in vivo Draize rabbit eye irritation test, this study aimed to construct an in silico model to predict the complete United Nations (UN) Globally Harmonized System (GHS) for classification and labeling of chemicals for eye irritation category [eye damage (Category 1), irritating to eye (Category 2) and nonirritating (No category)] of liquid chemicals with Integrated approaches to testing and assessment (IATA)-like two-stage random forest approach. Liquid chemicals (n = 219) with 34 physicochemical descriptors and quality in vivo data were collected with no missing values. Seven machine learning algorithms (Naive Bayes, Logistic Regression, First Large Margin, Neural Net, Random Forest (RF), Gradient Boosted Tree, and Support Vector Machine) were examined for the ternary categorization of eye irritation potential at a single run through 10-fold cross-validation. RF, which performed best, was further improved by applying the 'Bottom-up approach' concept of IATA, namely, separating No category first, and discriminating Category 1 from 2, thereafter. The best performing training dataset achieved an overall accuracy of 73% and the correct prediction for Category 1, 2, and No category was 80%, 50%, and 77%, respectively for the test dataset. This prediction model was further validated with an external dataset of 28 chemicals, for which an overall accuracy of 71% was achieved.

Differences in classification for skin corrosion/irritation in EU and Ukraine: Case study of alternative (in vitro and in silico) methods application for classification of pesticide active ingredient imazamox

In Ukraine Globally Harmonized System of classification of chemicals has not been implemented yet. In this article we analyze differences between GHS/CLP classification systems and Hygienic Classification of Pesticides by the Degree of Hazard currently in force in Ukraine in respect of approach and criteria for classification of effects on skin. As a case study, we conducted in silico modelling of herbicide imazamox using skin irritation/corrosion modules of ToxTree. The prediction of ToxTree was "Not Corrosive to skin". Then skin irritation and skin corrosion in vitro tests (OECD TGs 439, 431) were conducted. Classification of this substance based on in vitro and in vivo results according to GHS/CLP was the same, while it was not possible based on in vitro results to assign certain hazard class of Ukrainian classification due to difference in its and GHS/CLP criteria. However, ongoing process of harmonization of Ukrainian legislation with EU will give opportunity not only use alternative methods, but also adopt most recent advances and incorporate data from non-animal methods directly into classification criteria.

Designing Validation Studies More Efficiently According to the Modular Approach: Retrospective Analysis of the EPISKIN Test for Skin Corrosion

It is claimed that the modular approach to validation, which involves seven independent modules, will make the assessment of test validity more flexible and more efficient. In particular, the aspects of between-laboratory variability and predictive capacity are formally separated. Here, the main advantage of the approach is to offer the opportunity for reduced labour, and thus to allow study designs to be more time efficient and cost effective. The impact of this separation was analysed by taking the ECVAM validation study on in vitro methods for skin corrosivity as an example of a successful validation study — two of its methods triggered new OECD test guidelines. Lean study designs, which reduced the number of tests required by up to 60%, were simulated with the original validation data for the EPISKIN™ model. By using resampling techniques, we were able to demonstrate the effects of the lean designs on three between-laboratory variability measures and on the predictive capacity in terms of sensitivity and specificity, in comparison with the original study. Overall, the study results, especially the levels of confidence, were only slightly affected by the lean designs that were modelled. It is concluded that the separation of the two modules is a promising way to speed-up prospective validation studies and to substantially reduce costs, without compromising study quality.

Assessment of the Eye Irritating Properties of Chemicals by Applying Alternatives to the Draize Rabbit Eye Test: The Use of QSARs and In Vitro Tests for the Classification of Eye Irritation

Huggins has reported on the current situation relating to the development of alternatives to the Draize eye irritation test with rabbits, and an ECVAM Working Group have reviewed the efforts needed in order to replace this animal test within the next 10 years by using the results of non-animal assessment methods. Our report reviews regulatory experience gained over the last 20 years with the EU chemicals notification procedure with respect to the assessment of eye lesions observed in Draize tests. The nature of eye lesions and their importance for classification and labelling of possible hazards to human eyes are evaluated and discussed, with a view to promoting the development of specific in vitro assays which are able to discriminate between eye damage, moderate eye irritation, and minor irritation effects which are completely reversible within a few days. Structural alerts for the prediction of eye irritation/corrosion hazards to be classified and labelled according to international classification criteria, are presented, which should be validated in accordance with internationally agreed (OECD) principles for (Q)SAR system validation. Physicochemical limit values for prediction of the absence of any eye irritation potential relevant for human health can make available a definition of the applicability domains of alternative methods developed for the replacement of the Draize eye irritation test.

In Vitro Tests within the REACH Information Strategies

Tonnage-based information requirements are specified in the proposal on the regulation on the Registration, Evaluation and Authorisation of Chemicals (REACH) in the European Union. The hazard assessment for toxic endpoints should be performed by using a tiered approach, i.e. as an information strategy (IS), starting with an evaluation of all of the data already available, including animal in vivo and in vitro data, and human evidence and case reports, as well as data from (Quantitative)-Structure Activity Relationships ([Q]SARs) or read-across, before any further testing is suggested. To contribute to the implementation of the REACH system, the Nordic countries launched two projects: 1) a review of currently used testing strategies, including a comparison with the REACH requirements; and 2) the development of detailed ISs for skin and eye irritation/corrosion. The review showed that the ISs and classification criteria for the selected endpoints are inconsistent in many cases. In the classification criteria, human data and in vivo test results are usually the prerequisites. Other types of information, such as data from in vitro studies, can sometimes be used, but usually as supportive evidence only. This differs from the REACH ISs, where QSARs, read-across and in vitro testing are important elements. In the other part of the project, an IS for skin and eye irritation/corrosion was proposed. The strategy was “tested” by using four high production volume (HPV) chemicals: hydrogen peroxide, methyl tertiary-butyl ether (MTBE), trivalent chromium, and diantimony trioxide, but only MTBE and trivalent chromium are dealt with in this paper. The “test” revealed that in vivo data, human case reports and physical-chemical data were available and could be used in the evaluation. Classification could be based on the proposed IS and the existing data in all cases, except for the eye irritation/corrosion of trivalent chromium. Weight-of-evidence analysis appeared to be a useful step in the ISs proposed, and including it in the REACH strategies should be considered. For these chemicals, few in vitro and (Q)SAR data were available — more of these data would be generated, if the relevant guidance and legislation on classification were updated.

CON4EI: CONsortium for in vitro Eye Irritation testing strategy - EpiOcular™ time-to-toxicity (EpiOcular ET-50) protocols for hazard identification and labelling of eye irritating chemicals

Assessment of acute eye irritation potential is part of the international regulatory requirements for testing of chemicals. The objective of the CON4EI (CONsortium for in vitro Eye Irritation testing strategy) project was to develop tiered testing strategies for eye irritation assessment for all drivers of classification. A set of 80 reference chemicals (38 liquids and 42 solids) was tested with eight different alternative methods. Here, the results obtained with reconstructed human cornea-like epithelium (RhCE) EpiOcular™ in the EpiOcular time-to-toxicity Tests (Neat and Dilution ET-50 protocols) are presented. The primary aim of this study was to evaluate whether test methods can discriminate chemicals not requiring classification for serious eye damage/eye irritancy (No Category) from chemicals requiring classification and labelling for Category 1 and Category 2. In addition, the predictive capacity in terms of in vivo drivers of classification was investigated. The chemicals were tested in two independent runs by MatTek In Vitro Life Science Laboratories. Results of this study demonstrate very high specificity of both test protocols. With the existing prediction models described in the SOPs, the specificity of the Neat and Dilution method was 87% and 100%, respectively. The Dilution method was able to correctly predicting 66% of GHS Cat 2 chemicals, however, prediction of GHS Cat 1 chemicals was only 47%-55% using the current protocols. In order to achieve optimal prediction for all three classes, a testing strategy was developed which combines the most predictive time-points of both protocols and for tests liquids and solids separately. Using this new testing strategy, the sensitivity for predicting GHS Cat 1 and GHS Cat 2 chemicals was 73% and 64%, respectively and the very high specificity of 97% was maintained. None of the Cat 1 chemicals was underpredicted as GHS No Category. Further combination of the EpiOcular time-to-toxicity protocols with other validated in vitro systems evaluated in this project, should enable significant reduction and even possible replacement of the animal tests for the final assessment of the irritation potential in all of the GHS classes.

The local lymph node assay in practice: a current regulatory perspective

Following the formal acceptance of the local lymph node assay (LLNA) as an Organization for Economic Cooperation and Development (OECD) guideline in April 2002, the UK Health and Safety Executive (HSE) informed notifiers that this was now the method of choice for the assessment of skin sensitization potential under the EU notification scheme for new industrial chemicals (NONS). This paper summarizes the experience of the HSE for the 2-year period immediately following the issuing of this statement, during which 48 LLNA study reports were assessed for notification purposes. The issues discussed here include adherence to the OECD guideline, interpretation of results, and classification outcomes. Generally, notifying laboratories followed the OECD guideline successfully, with regard to the sex/ strain/numbers of mice used, the precise process used for measurement of cell proliferation, and the use of recommended vehicles and positive controls. Initially, use of the individual animal approach (measuring the cell proliferation in each animal rather than for a pooled dose group) highlighted problems caused by technical inexperience, but these were overcome by practice. Toxicity or irritation were found to be minor factors in dose selection; more important was the choice of vehicle to correctly maximize the test substance concentration, while maintaining appropriate application properties. Contrary to concerns that the LLNA would prove to be less sensitive or more sensitive than the traditionally used Guinea Pig Maximization Test (GPMT), the proportion of new substances classified as skin sensitizers was within the range observed in previous years. Although the sample size is relatively small, the experience of the HSE indicates that the LLNA is satisfactory for routine regulatory use.

A Study of the Potential of the Pig as a Model for the Vaginal Irritancy of Benzalkonium Chloride in Comparison to the Nonirritant Microbicide PHI-443 and the Spermicide Vanadocene Dithiocarbamate

A porcine model was established to test the mucosal toxicity potential of a thiophene thiourea (PHI-443)-based anti-HIV microbicide and a vanadocene-based spermicide, vanadocene dithiocarbamate (VDDTC) in comparison to benzalkonium chloride (BZK). Nine domestic pigs (Duroc) in nonestrus stage received a single intravaginal application of 2% BZK, 2% PHI-443, or 0.1% VDDTC-containing gel. At various times after gel application, cell differentials and levels of inflammatory cytokines (IL-1 β, IL-4, IL-6, IL-8, IL-10, IL-18, IFN- γ, and TNF- α) in cervicovaginal lavage (CVL) fluid were monitored by flow cytometry and ELISA, respectively. Eight pigs were exposed intravaginally to a gel with and without BZK or VDDTC for 4 consecutive days and vaginal tissues were scored histologically for inflammation using a new scoring system. Only CVL fluid from pigs exposed to BZK showed a significant increase of IL-1 β, IL-8, and also IL-18 production when compared to the controls, PHI-443 or VDDTC-treated groups. Maximum levels of BZK-induced IL-1 β (100-fold), IL-8 (2,500-fold), IL-18 (80-fold), and IFN- γ(10-fold) were found at 24 hours. In the in vivo porcine vaginal irritation model, increased levels of vaginal IL-1 β, IL-8, and IL-18 were associated with histological changes consistent with vaginal inflammation. These results demonstrate that key cervicovaginal inflammatory cytokines are useful in vivo biomarkers for predicting the mucosal toxicity potential of vaginal products in the physiologically relevant and sensitive porcine model.

Assessment of human epidermal model LabCyte EPI-MODEL for in vitro skin irritation testing according to European Centre for the Validation of Alternative Methods (ECVAM)-validated protocol

A validation study of an in vitro skin irritation testing method using a reconstructed human skin model has been conducted by the European Centre for the Validation of Alternative Methods (ECVAM), and a protocol using EpiSkin (SkinEthic, France) has been approved. The structural and performance criteria of skin models for testing are defined in the ECVAM Performance Standards announced along with the approval. We have performed several evaluations of the new reconstructed human epidermal model LabCyte EPI-MODEL, and confirmed that it is applicable to skin irritation testing as defined in the ECVAM Performance Standards. We selected 19 materials (nine irritants and ten non-irritants) available in Japan as test chemicals among the 20 reference chemicals described in the ECVAM Performance Standard. A test chemical was applied to the surface of the LabCyte EPI-MODEL for 15 min, after which it was completely removed and the model then post-incubated for 42 hr. Cell v iability was measured by MTT assay and skin irritancy of the test chemical evaluated. In addition, interleukin-1 alpha (IL-1alpha) concentration in the culture supernatant after post-incubation was measured to provide a complementary evaluation of skin irritation. Evaluation of the 19 test chemicals resulted in 79% accuracy, 78% sensitivity and 80% specificity, confirming that the in vitro skin irritancy of the LabCyte EPI-MODEL correlates highly with in vivo skin irritation. These results suggest that LabCyte EPI-MODEL is applicable to the skin irritation testing protocol set out in the ECVAM Performance Standards.

[Establishment of reconstituted human engineered skin model for evaluating cosmetic skin corrosion/irritation]

OBJECTIVE

Constructing reconstituted human engineered skin model, to establish an alternative approach for evaluating cosmetics skin corrosion and irritation.

METHODS

A cut of human dermal tissue was taken, and the epidermis and dermis were separated respectively for primary cell culture. The reconstituted skin models were onstructed with rat tail collagen. Then, the models under air-liquid face were cultured. Skin models were affected by Chemicals for 3 min or 1 h in corrosivity test, and the models were affected for 15 min in irritation test following with 42 h culture. Then, the cytoactive were detected by MTT essay and IL-1alpha releases level were detected by ELISA.

RESULTS

In corrosion test, the cytoactive was declined < 50% (3 min) and 15% (1 h) after giving three corrosive chemicals. However, the cytoactive wasn't change for two un-corrosive chemicals. In irritation test, three of four irritation chemicals caused cytoactive less than 50%, and another one made the IL-1alpha release more than 60 pg/ml in ELISA test. It seemed that all of four chemicals were irritations.

CONCLUSION

It was suggested that constructing reconstituted human engineered skin model could be established as an alternative approach for predicting skin corrosion and irritation as possible.

Toxmatch--a chemical classification and activity prediction tool based on similarity measures

Chemical similarity forms the underlying basis for the development of (Quantitative) Structure-Activity Relationships ((Q)SARs), expert systems and chemical groupings. Recently a new software tool to facilitate chemical similarity calculations named Toxmatch was developed. Toxmatch encodes a number of similarity indices to help in the systematic development of chemical groupings, including endpoint specific groupings and read-across, and the comparison of model training and test sets. Two rule-based classification schemes were additionally implemented, namely: the Verhaar scheme for assigning mode of action for aquatic toxicants and the BfR rulebase for skin irritation and corrosion. In this study, a variety of different descriptor-based similarity indices were used to evaluate and compare the BfR training set with respect to its test set. The descriptors utilised in this comparison were the same as those used to derive the original BfR rules i.e. the descriptors selected were relevant for skin irritation/corrosion. The Euclidean distance index was found to be the most predictive of the indices in assessing the performance of the rules.

Development of the short time exposure (STE) test: an in vitro eye irritation test using SIRC cells

Using SIRC (rabbit corneal cell line) cells, we developed an alternative eye irritation test: the short time exposure (STE) test. This STE test is a cytotoxicity test using physiological saline or mineral oil as the test solvent. Evaluation exposure time is short (5 min), which is similar to actual exposure situations, and uses the cell viability (CV) at a constant concentration as the endpoint for irritation potential. First, in order to confirm the usefulness of this STE test in assessing eye irritation potential of chemicals, 51 raw materials were tested and the correlation between CV in the STE test and the eye irritation score in the Draize test was examined. For the undiluted raw materials tested in the Draize test, the 5% test concentration in the STE test gave irritation classes that correlated well with the irritation classes from the Draize test (accuracy: 89.6%). For those materials tested as a 10% solution in the Draize test, STE irritation classes with 0.05% test concentration corresponded well with the Draize irritation classes (accuracy: 80.0%). Next, using the cell viabilities at these two concentrations, the STE prediction model (PM) was developed. A score of 1 or 2 was given for the results from each tested concentration in the STE test and Draize test. The scores from each test were then summed to yield a 3-level (Rank 1: minimally irritant, Rank 2: moderate irritant, Rank 3: severe irritant) eye irritation potential classification. Rank classification in the STE test showed a good correlation mostly to that in the Draize test (irritation class correspondence rate: 70.2%, but after exclusion of data of alcoholic materials, the rate was 91.7%). In most cytotoxicity test, the cytotoxicity of acids and amines is generally underestimated due the use of medium as the solvent. This is the result of the buffering capacity of the media. On the other hand, the STE test could predict the eye irritation potential by evaluating the chemical with a 5% test concentration. Eleven water insoluble materials such as toluene, octanol, and hexanol could be evaluated by using mineral oil as test solvent in the STE test. The STE test demonstrated itself to be simple, promising, have great potential, be of value, and to be an easily standardized alternative eye irritation test.

Fragrance material review on trans-2-hexenyl salicylate

A toxicologic and dermatologic review of trans-2-hexenyl salicylate when used as a fragrance ingredient is presented.

Fragrance material review on 4-(2,4,6-trimethyl-3-cyclohexen-1-yl)-3-buten-2-one

A toxicologic and dermatologic review of 4-(2,4,6-trimethyl-3-cyclohexen-1-yl)-3-buten-2-one when used as a fragrance ingredient is presented.

Fragrance material review on ethyl salicylate

A toxicological and dermatological review of ethyl salicylate when used as a fragrance ingredient is presented.

Evaluation and application of the RD50 for determining acceptable exposure levels of airborne sensory irritants for the general public

BACKGROUND

The RD(50) (exposure concentration producing a 50% respiratory rate decrease) test evaluates airborne chemicals for sensory irritation and has become an American Society for Testing and Materials (ASTM) standard method. Past studies reported good correlations (R(2)) between RD(50)s and the occupational exposure limits, particularly threshold limit values (TLVs).

OBJECTIVE

The main purpose of this study was to examine the relationship between RD(50)s and human sensory irritation responses in a quantitative manner, particularly for chemicals that produce burning sensation of the eyes, nose, or throat, based on lowest observed adverse effect levels (LOAELs) reported for human subjects.

METHODS

We compared RD(50)s with LOAELs and acute reference exposure levels (RELs). RELs, developed by the California Environmental Protection Agency's Office of Environmental Health Hazard Assessment, represent a level at which no adverse effects are anticipated after exposure. We collected RD(50)s from the published literature and evaluated them for consistency with ASTM procedures. We identified LOAELs for human irritation and found 25 chemicals with a corresponding RD(50) in mice.

DISCUSSION

We found the relationship between RD(50)s and LOAELs as log RD(50) = 1.16 (log LOAEL) + 0.77 with an R(2) value of 0.80. This strong correlation supports the use of the RD(50) in establishing exposure limits for the public. We further identified 16 chemical irritants with both RD(50)s and corresponding acute RELs, and calculated the relationship as log RD(50) = 0.71 (log REL) + 2.55 with an R(2) value of 0.71. This relationship could be used to identify health protective values for the public to prevent respiratory or sensory irritation.

CONCLUSION

Consequently, we believe that the RD(50) has benefits for use in setting protective levels for the health of both workers and the general population.

Fragrance material review on cinnamyl butyrate

A toxicologic and dermatologic review of cinnamyl butyrate when used as a fragrance ingredient is presented.

Surfactant-induced TRPV1 activity--a novel mechanism for eye irritation?

The pain receptor transient receptor potential vanilloid type 1 (TRPV1) has been reported as one of the key components in the pain pathway. Activation of the receptor causes a Ca2+ influx with secondary effects leading to neurogenic inflammation. Here we report specific activation of TRPV1 by detergent-containing hygiene products measured as intracellular Ca2+ influxes in stably TRPV1-expressing neuroblastoma SH-SY5Y cells. Children products marketed as "painless" (containing lower concentration of detergents), and conditioners (without detergents) did not induce specific TRPV1 activation. Furthermore, low concentrations of the detergent sodium lauryl sulfate dose-dependently induced Ca2+ influxes that could be addressed to TRPV1. These results reveal a novel mechanistic pathway for surfactant-induced nociception, which may be an important endpoint in in vitro test batteries as alternatives to Draize's rabbit eye test for classification of eye irritating products.

Evaluation of SARs for the prediction of eye irritation/corrosion potential: structural inclusion rules in the BfR decision support system

The proposed REACH regulation within the European Union (EU) aims to minimise the number of laboratory animals used for human hazard and risk assessment while ensuring adequate protection of human health and the environment. One way to achieve this goal is to develop non-testing methods, such as (quantitative) structure-activity relationships ([Q]SARs), suitable for identifying toxicological hazard from chemical structure and physicochemical properties alone. A database containing data submitted within the EU New Chemicals Notification procedure was compiled by the German Bundesinstitut für Risikobewertung (BfR). On the basis of these data, the BfR built a decision support system (DSS) for the prediction of several toxicological endpoints. For the prediction of eye irritation and corrosion potential, the DSS contains 31 physicochemical exclusion rules evaluated previously by the European Chemicals Bureau (ECB), and 27 inclusion rules that define structural alerts potentially responsible for eye irritation and/or corrosion. This work summarises the results of a study carried out by the ECB to assess the performance of the BfR structural rulebase. The assessment included: (a) evaluation of the structural alerts by using the training set of 1341 substances with experimental data for eye irritation and corrosion; and (b) external validation by using an independent test set of 199 chemicals. Recommendations are made for the further development of the structural rules in order to increase the overall predictivity of the DSS.

Assessment of the skin irritation potential of chemicals by using the SkinEthic reconstructed human epidermal model and the common skin irritation protocol evaluated in the ECVAM skin irritation validation study

Currently, two reconstructed human skin models, EpiDerm and EPISKIN are being evaluated in an ECVAM skin irritation validation study. A common skin irritation protocol has been developed, differing only in minor technical details for the two models. A small-scale study, applying this common skin irritation protocol to the SkinEthic reconstructed human epidermis (RHE), was performed at ZEBET at the BfR, Berlin, Germany, to consider whether this protocol could be successfully transferred to another epidermal model. Twenty substances from Phase III of the ECVAM prevalidation study on skin irritation were tested with the SkinEthic RHE. After minor, model-specific adaptations for the SkinEthic RHE, almost identical results to those obtained with the EpiDerm and EPISKIN models were achieved. The overall accuracy of the method was more than 80%, indicating a reliable prediction of the skin irritation potential of the tested chemicals when compared to in vivo rabbit data. As a next step, inter laboratory reproducibility was assessed in a study conducted between ZEBET and the Department of Experimental Toxicology, Schering AG, Berlin, Germany. Six coded substances were tested in both laboratories, with three different batches of the SkinEthic model. The assay results showed good reproducibility and correct predictions of the skin irritation potential for all six test chemicals. The results obtained with the SkinEthic RHE and the common protocol were reproducible in both phases, and the overall outcome is very similar to that of earlier studies with the EPISKIN and EpiDerm models. Therefore, the SkinEthic skin irritation assay test protocol can now be evaluated in a formal "catch-up" validation study.

Simple stochastic fingerprints towards mathematical modeling in biology and medicine. 3. ocular irritability classification model

MARCH-INSIDE methodology and a statistical classification method--linear discriminant analysis (LDA)--is proposed as an alternative method to the Draize eye irritation test. This methodology has been successfully applied to a set of 46 neutral organic chemicals, which have been defined as ocular irritant or nonirritant. The model allow to categorize correctly 37 out of 46 compounds, showing an accuracy of 80.46%. Specifically, this model demonstrates the existence of a good categorization average of 91.67 and 76.47% for irritant and nonirritant compounds, respectively. Validation of the model was carried out using two cross-validation tools: Leave-one-out (LOO) and leave-group-out (LGO), showing a global predictability of the model of 71.7 and 70%, respectively. The average of coincidence of the predictions between leave-one-out/leave-group-out studies and train set were 91.3% (42 out of 46 cases)/89.1% (41 out of 46 cases) proving the robustness of the model obtained. Ocular irritancy distribution diagram is carried out in order to determine the intervals of the property where the probability of finding an irritant compound is maximal relating to the choice of find a false nonirritant one. It seems that, until today, the present model may be the first predictive linear discriminant equation able to discriminate between eye irritant and nonirritant chemicals.

The minipig in dermatotoxicology: Methods and challenges

In view of more morphological and physiological similarities between human and porcine skin than for other laboratory animal species, the minipig is a preferred model to evaluate the safety profile of dermally applied xenobiotics. Different methods of dermal administration and examples of non-invasive and invasive investigations during the in-life phase of the studies are described. Routine and special post-mortem examinations in dermal studies are presented to assess responses to the topical treatment of minipig skin. Challenges in dermal minipig studies are discussed with respect to animal welfare and husbandry, test formulations, application methods and different types of investigations. One of the most significant issues for dermal minipig studies is the extensive measures required to prevent cross-contamination of blood and tissue samples taken to monitor local and systemic exposure to the test item.

In vitro skin irritation testing on reconstituted human epidermis: Reproducibility for 50 chemicals tested with two protocols

Since it is of high importance to establish the skin irritation potential of industrial chemicals, toxicologists developed tests on various skin models. Most test data come from the rabbit Draize test, but its reproducibility is questionable. Some human in vivo test data exist, but they concern only few compounds. The emergence of new tools such as reconstituted human skin tissues offers a promising future to alternative methods. We describe here two in vitro skin irritation test protocols performed on reconstituted human epidermis. One is a direct topical application test and the other an in vitro patch test. Both protocols were performed using multiple endpoint analysis including cell viability (MTT reduction), histology, and IL-1alpha release. Fifty chemicals were tested: 20 compounds were used in the ECVAM pre-validation study and 30 products were previously tested in a human in vivo patch test. These in vitro skin irritation tests have not only the advantages of enhanced convenience and of reduced costs, but a good reproducibility is observed by endpoint, and by compound. A prediction model is proposed to classify the chemicals as irritant or non-irritant, and the results are compared to available rabbit and human data. We do not wish to overgeneralize from these 50 compounds; but, instead suggest that this data set be extensively extended to include chemicals of varying physico-chemical properties.

Comparative study of red blood cell method in rat and calves blood as alternatives of Draize eye irritation test

INTRODUCTION

Red blood cell assay (RBC) is used to estimate potential irritation of tensioactive agents and detergents. Cell membrane lysis and cell protein denaturation are measured photometrically. This study was aimed to determine if rat blood cells can be used to predict eye potential irritation in the same way of calves blood cells in RBC assay.

METHODS

We evaluated 20 cosmetic formulations using rat and calves blood according to INVITOX protocol No 37. Data of media hemolysis concentration, denaturation index and the ratio of both parameters were compared with in vivo data of eye irritancy.

RESULTS

There was a significant difference (p<0.01) between H50 value when evaluated the standard SDS with red blood cell method in rat and calves blood. According to the exact probability of Fisher taking as approach the acceptance or rejection of the substance there are no significant differences between in vitro assay with calves blood and in vivo results. Not happening the same way for the RBC assay with rat blood where significant differences were obtained (p<0.01) among the classification of in vitro and in vivo test.

DISCUSSION

The RBC assay using calves blood showed better results. Several test substances were false negatives with rat blood. This high false negative rate would be correctly identified by the animal test but it may also lead to increased animal consumption. For that RBC assay with calf blood cells is preferable to the employment of rat blood as screening method with a reduction and refinement strategy.

Reconstituted human corneal epithelium: a new alternative to the Draize eye test for the assessment of the eye irritation potential of chemicals and cosmetic products

The aim of this study was to evaluate the interest of a new three-dimensional epithelial model cultivated from human corneal cells to replace animal testing in the assessment of eye tolerance. To this end, 65 formulated cosmetic products and 36 chemicals were tested by means of this in vitro model using a simplified toxicokinetic approach. The chemicals were selected from the ECETOC data bank and the EC/HO International validation study list. Very satisfactory results were obtained in terms of concordance with the Draize test data for the formulated cosmetic products. Moreover, the response of the corneal model appeared predictive of human ocular response clinically observed by ophthalmologists. The in vitro scores for the chemicals tested strongly correlated with their respective scores in vivo. For all the compounds tested, the response of the corneal model to irritants was similar regardless of their chemical structure, suggesting a good robustness of the prediction model proposed. We concluded that this new three-dimensional epithelial model, developed from human corneal cells, could be promising for the prediction of eye irritation induced by chemicals and complex formulated products, and that these two types of materials should be tested using a similar protocol. A simple shortening of the exposure period was required for the chemicals assumed to be more aggressively irritant to the epithelial tissues than the cosmetic formulae.

Slug species- and population-specific effects on the end points of the Slug Mucosal Irritation test

The Slug Mucosal Irritation test seems to be a promising method for the evaluation of the local tolerance of products applied to the mucosa. Furthermore, the Slug Mucosal Irritation test is a reliable method to classify chemicals accurately into three eye irritation categories based on the mucus production and the score for tissue damage. Until now the slug Arion lusitanicus collected in Belgium was always used as test organism. The present study investigated the effects of the slug population and species on the end points of the test and on the eye irritation classification. For this purpose, the eye irritation/damage test procedure and eye reference chemicals were used. Comparison of the results of one Belgian and two Swiss A. lusitanicus populations indicated that the geographic and ecological origins of slug populations did neither influence the mucus production, nor the score for tissue damage. Slug species-specific effects on the test end points were investigated by comparing the data of Belgian Limax flavus with corresponding data of Belgian A. lusitanicus. L. flavus produced more mucus than A. lusitanicus, so that the mucus production cut-off values had to be increased. Therefore, the results indicated that the test procedure and prediction model have to be optimised and validated, if other slug species are used instead of A. lusitanicus.

Epidermal-skin-test 1000 (EST-1000)—A new reconstructed epidermis for in vitro skin corrosivity testing

The determination of a possible corrosive or irritative potential of certain products and ingredients is necessary for their classification and labeling requirements. Reconstructed skin as a model system provides fundamental advantages to single cell culture testing and leads to promising results as shown by different validation studies (for review: Fentem, J.H., Botham, P.A., 2002. ECVAM's activities in validating alternative tests for skin corrosion and irritation. ATLA 30(Suppl. 2), 61-67). In this study we introduce our new reconstructed epidermis "Epidermal-Skin-Test" (EST-1,000). This fully grown epidermis consists of proliferating as well as differentiating keratinocytes. EST-1,000 shows a high comparability to normal human skin as shown by histological and immunohistochemical data. Characteristic markers (KI-67, CK 1/10/5/14, transglutaminase, collagen IV, involucrin, beta 1 integrin) can be identified easily. The main focus of this work was to characterize EST-1,000 especially with respect to its barrier function by testing several substances of known corrosive potential. Skin corrosion was detected by the cytotoxic effect of the substances on a reconstructed epidermis after short-term application to the stratum corneum. The effect was determined by standard MTT assay and accompanying histological analysis. Hence EST-1,000 shows a very high predictive potential and closes the gap between animal testing and the established full-thickness model Advanced-Skin-Test 2,000 (AST-2,000) (Noll, M., Merkle, M.-L., Kandsberger, M., Matthes, T., Fuchs, H., Graeve, T., 1999. Reconstructed human skin (AST-2,000) as a tool for pharmaco-toxicology. ATLA 27, 302).

A study of the relationship between cornea permeability and eye irritation using membrane-interaction QSAR analysis

A methodology termed membrane-interaction QSAR (MI-QSAR) analysis has been used to develop QSAR models to predict drug permeability coefficients across cornea and its component layers (epithelium, stroma, and endothelium). From a training set of 25 structurally diverse drugs, significant QSAR models are constructed and compared for the permeability of the cornea, epithelium, and stroma plus endothelium. Cornea permeability is found to depend on the measured distribution coefficient of the drug, the cohesive energy of the drug, the total potential energy of the drug-membrane "complex," and three other energy refinement descriptor terms. The endothelium may be a more important barrier in cornea permeation than the stroma. Moreover, an investigation of the correlation between cornea permeation and eye irritation is presented as an example of a cross study on different ADMET properties using MI-QSAR analysis. Thirteen structurally diverse drugs, whose molar-adjusted eye irritation scores (MES) have been measured using the Draize rabbit-eye test, were chosen as an eye irritation comparison set. A poor correlation (R(2) = 0.0232) between the MES measures and the predicted cornea permeability coefficients for the drugs in the eye irritation set suggests there is no significant relationship between eye irritation potency and the cornea permeability.

Assessment of contact allergens by dissociation of irritant and sensitizing properties

The human organotypic skin explant culture (hOSEC) model is a promising alternative in vitro model for screening contact allergens. In this model, the chemical-induced migration of Langerhans cells (LCs) out of the epidermis, evaluated after a 24-h exposure period, is used as a measure of sensitizer potential. As skin irritants can also induce LC migration it is essential that concentrations of test chemicals are used that are not even weakly irritant. Using the hOSEC irritation model chemicals are classified as weak irritants if they are toxic after a 48-h exposure period. Toxicity is determined by methyl green-pyronine (MGP) staining of hOSEC. We studied three frequently used non-sensitizing skin irritants and six potent or frequent human sensitizers in a dose-response. A complete discrimination between non-sensitizers and contact sensitizers was obtained for the chemicals tested when the concentrations used were lower than the weak irritant concentrations. Frequency of positive allergen reactions in patch test of human populations correlated with the difference between weak irritant concentrations and the lowest concentration inducing significant LC migration. Sensitizer potency correlated with chemical irritancy as determined by keratinocyte death. For the compounds tested, the hOSEC model predicted allergenicity in humans better than the guinea pig maximization test and the mouse local lymph node assay.

Refinement of the Slug Mucosal Irritation test as an alternative screening test for eye irritation

The objective of this study was to limit the test procedure time of the Slug Mucosal Irritation test to one day and to determine whether it is a relevant and reliable method to predict the eye irritation potential of chemicals. The irritation potential of several eye reference chemicals can be estimated by the amount of mucus produced when tested at a 1% concentration (60 min). Since some in vivo irritating chemicals did not influence this endpoint the effect of increasing concentrations on membrane damage was investigated. This study revealed that when tested at a 3.5% concentration (60 min) the underestimated chemicals induced an increased protein and/or enzyme release. A two-step classification prediction model was developed that classified the chemicals first by the amount of mucus produced (1%, 60 min). Chemicals that did not affect this endpoint were classified based on the membrane damage induced by a second treatment (3.5%, 60 min). The results were compared with the corresponding EU classification (NI, R36 and R41) and 71% of the chemicals were correctly classified with a specificity and sensitivity of 75% and 94%, respectively. Repeated testing of the chemicals revealed a good intra-laboratory reproducibility. The test seems to be a promising method for screening the eye irritating potential of chemicals.

A two-centre evaluation of the human organotypic skin explant culture model for screening contact allergens

Animal models are considered to be the "gold standard" for determining the potential contact allergenicity of low molecular weight chemicals. However, governmental regulations and ethical considerations limit the use of animals for such purposes. There is therefore a need for in vitro alternative models. The human organotypic skin explant culture (HOSEC) model is reported to be a promising alternative method for the predictive testing of contact allergens. The accelerated migration of Langerhans cells from the epidermis upon exposure to contact allergens is used to identify chemicals that are potentially capable of inducing a delayed-type hypersensitivity. In the study described in this paper, the model was further refined, and used, in two independent laboratories, to screen 23 low molecular weight compounds of known classification for their allergenicity. Each laboratory was able to accurately detect the contact allergens, despite small variations in the protocols used. However, the classification of dermal irritants, which have often been falsely classified as allergens, varied between the two laboratories. Despite the current limitations of the HOSEC model, the accuracy of the predictions made (sensitiser or non-sensitiser) compare favourably with classifications obtained with commonly used animal models. The HOSEC model has the potential to be developed further as an in vitro alternative to animal models for screening for contact allergens.

Final report on the safety assessment of 6-Amino-m-Cresol, 6-Amino-o-Cresol, 4-Amino-m-Cresol, 5-Amino-4-Chloro-o-Cresol, 5-Amino-6-Chloro-o-Cresol, and 4-Chloro-2-Aminophenol

Each of these ingredients function as hair colorants. 5-Amino-4-Chloro-o-Cresol and 5-Amino-6-Chloro-o-Cresol are identified as oxidative hair dyes, that is, they are combined with an oxidizing agent before being applied to the hair. 6-Amino-m-Cresol, 6-Amino-o-Cresol, 4-Amino-m-Cresol, and 5-Amino-4-Chloro-o-Cresol are used in oxidative hair dyes, but it is not known if they are also used in nonoxidative (semipermanent) hair dyes. No toxicologically significant impurities are present with these two ingredients. To supplement the safety test data on these ingredients, available data on related ingredients (4-amino-2-hydroxytoluene and p-,m-, and o-aminophenol) previously found safe as used by the Cosmetic Ingredient Review (CIR) Expert Panel were summarized. 5-Amino-4-Chloro-o-Cresol and 5-Amino-6-Chloro-o-Cresol do not absorb significant ultraviolet radiation in the UVB region and none in the UVA region, although 4-Amino-m-Cresol had a symmetrical UV absorption peak at 300 nm. Percutaneous penetration of 5-Amino-4-Chloro-o-Cresol and 5-Amino-6-Chloro-o-Cresol alone was significant, but when combined with oxidative developer, skin absorption was extremely low. Both of these dyes are excreted rapidly via the urine. Repeated exposure of animal skin to 5-Amino-4-Chloro-o-Cresol and 5-Amino-6-Chloro-o-Cresol failed to produce any cumulative irritation and single exposures up to 10%were not irritating to animal skin. 5-Amino-4-Chloro-o-Cresol and 5-Amino-6-Chloro-o-Cresol combined with oxidizer were not sensitizers in guinea pig maximization tests. Ocular irritation resulted from exposure of animals to undiluted 5-Amino-4-Chloro-o-Cresol, but not to a 5%solution. Only minor irritation was observed with 5%5-Amino-6-Chloro-o-Cresol. Subchronic toxicity testing in animals using 5-Amino-4-Chloro-o-Cresol, 5-Amino-6-Chloro-o-Cresol, and 4-Amino-m-Cresol did not yield any adverse reactions. 6-Amino-m-Cresol and 4-Amino-m-Cresol were generally not mutagenic in in vitro and in vivo tests. Exposure to 5-Amino-4-Chloro-o-Cresol, 5-Amino-6-Chloro-o-Cresol, 6-Amino-m-Cresol and 4-Amino-m-Cresol from cosmetics were several orders of magnitude below developmental toxicity no-observed-adverse effect levels (NOAELs). Although irritation data on several ingredients are absent, products containing these ingredients must include a caution statement and patch test instructions for determining whether the product causes skin irritation. The Expert Panel expects that following this procedure would identify individuals who would have an adverse reaction and allow them to avoid significant exposures. These compounds, when tested alone, are moderate skin sensitizers, but when combined with the developer, these ingredients are not sensitizers in animal tests. This information, coupled with the available animal test data, supports the safety of these ingredients in oxidative hair dyes. In the absence of systemic toxicity data, however, the available data are insufficient to support the safety of 6-Amino-o-Cresol and 4-Chloro-2-Aminophenol in semipermanent hair dyes. The types of data required for these two ingredients for this use include (1) physical and chemical properties, including the octanol/water partition coefficient; (2) impurities data, especially regarding the presence of m-cresol, other organic molecules, and heavy metals; (3) data demonstrating that the metabolism is similar to that of 4-amino-2-hydroxytoluene and/or p-,m-, and o-aminophenol, or 28-day dermal toxicity with histopathology, dermal reproductive toxicity data, and an in vitro genotoxicity study for 6-Amino-o-Cresol and one genotoxicity study in a mammalian system; if positive, a 2-year dermal carcinogenicity study using National Toxicology Program methods may be needed.

The benefits of the 3T3 NRU test in the safety assessment of cosmetics: long-term experience from pre-marketing testing in the Czech Republic

We have introduced the 3T3 NRU cytotoxicity test for methodological, economical and ethical reasons as a regular part of tier pre-marketing testing to assess local tolerance of raw materials for cosmetics, household chemicals and final cosmetic products. Using the 3T3 cell line according to the standard INVITTOX protocol No.64 (NRU Assay) the borderline concentration, relevant to the highest tolerated dose, is determined for each material. The toxic effect is reached at different concentration levels specific for individual cosmetics categories, depending on their chemical characteristics. Typical ranges of cytotoxicity for specific categories of cosmetics were established after testing of hundreds of materials. The range lies between 1 microg/ml (anti-dandruff shampoos), up to 2000 microg/ml (toothpastes and mouthwashes). The 3T3 NRU cytotoxicity test is a sensitive tool able to identify more aggressive products, that are also more likely to evoke irritation in human skin. It was even possible to detect protective effects of one natural herbal ingredient. The comparative study of cytotoxicity test results and human patch test results from a group of essential oils is presented. Cytotoxicity tests represent a highly ethical approach for estimation of irritancy. On the basis of in vitro test results suggesting low risk we can proceed to confirmatory tests in human volunteers.

Tiered testing strategies--acute local toxicity

More work has been done to develop alternatives to animal use in the areas of eye and skin irritation than in any area other than carcinogenicity. There has long been a belief both in the scientific community and among the public that the development of nonanimal tests in these areas should be simple and straightforward. After more than 20 yr of research, we can identify materials corrosive to the skin without using animals, but the assessment of irritation using in vitro methods alone is still an illusive goal. This review of current recommendations and industry practices that reduce the number of animals needed for these two tests concludes that animal use for skin irritation testing is not necessary today, with currently available and accepted methodology, except for regulatory reasons. Scientifically sound improvements in current eye irritation methods are also available. Advances in the understanding of the mechanisms of eye irritation that have been made in the last 5 yr should lead to improved in vitro methods for this endpoint. In the meantime, changes should be made in the current animal protocol to reduce pain and distress. This paper provides an overview of the progress that has been made toward discontinuing the use of animals in tests to determine the potential of materials to cause skin or eye damage after a single acute exposure. It also discusses some additional changes that could be made now to reduce animal use further or to reduce pain and distress in the testing that must still be done until such time as we can meet the ultimate goal--validated and accepted nonanimal methods for these endpoints.

In-house assessment of a modified in vitro cytotoxicity assay for higher throughput estimation of acute toxicity

The main objective of this study was to assess an in-house 3T3 NRU cytotoxicity assay for compatibility with a prediction model for acute rodent oral toxicity endorsed by an NIEHS-ICCVAM workshop. The aim is to use the NRU assay as one test component of HTS strategies for both acute oral toxicity and acute skin irritation, enabling the rejection of the most toxic materials and prioritisation of other materials for further testing. Groups of model cytotoxins and irritants were tested using the NRU assay and their EC50 values obtained from dose-response curves. These values were compared with those estimated from a limited (three)-dose protocol, deemed more suitable for HTS. A good correlation was observed between the EC50 values from both dose-response curves (R2=0.94). The relationships between EC50 values and acute rodent oral toxicity were compared by application of the prediction model to the model cytotoxins. The results from both full and limited dose-responses fitted within the acceptance limits of the prediction model, with regression lines similar to that of the model. Results indicated that the performance of the currently used 3T3 NRU cytotoxicity assay was similar to that of the assays used to generate the data employed in developing the prediction model. This prediction model can be applied with both the standard and HT assays to estimate acute rodent oral toxicity.

Optical response of the cultured bovine lens; testing opaque or partially transparent semi-solid/solid common consumer hygiene products

This study determines the relative ocular lens irritancy of 16 common partially transparent or non-transparent consumer hygiene products. The irritancy was found by measuring the changes in the sharpness of focus [referred to as the back vertex distance (BVD) variability] of the cultured bovine lens using a scanning laser In Vitro Assay System. This method consists of a laser beam that scans across the lens, and a computer, which then analyses the average focal length (mm), the BVD variability (mm), and the intensity of the beam transmitted. Lenses were exposed to the 16 hygiene products and the lens' focusing ability was monitored over 192 h. The products are semi-solids or solids (e.g. gels, lotions, shampoos). They are categorized into six groups: shampoos, body washes, lotions, toothpastes, deodorant, and anti-perspirant. Damage (measured by > 1 mm BVD variability) occurred slower for the shampoos, especially in the case of baby shampoo. The results indicate that shampoos exhibit the lowest level of ocular lens toxicity (irritability) while the deodorant is the most damaging.

Draize rabbit eye test compatibility with eye irritation thresholds in humans: a quantitative structure-activity relationship analysis

Draize rabbit eye test scores, as modified maximum average score (MMAS), for 68 pure bulk liquids were adjusted by the liquid-saturated vapor pressure P. These 68 adjusted scores, as log (MMAS/P), were shown to be completely equivalent to eye irritation thresholds (EIT), expressed as log (1/EIT), for 23 compounds in humans. Thus, for the first time the Draize eye test in rabbits for pure bulk liquids is shown to be perfectly compatible with eye irritation thresholds in humans. The total data set for 91 compounds was analyzed by the general solvation equation of Abraham. Values of log (MMAS/P) or log (1/EIT) could be fitted to a five-parameter equation with R2 = 0.936, SD = 0.433, AD = 0.000, and AAD = 0.340 over a range of 9.6 log units. When divided into a training set of 45 compounds, the corresponding equation could be used to predict the remaining 46 compounds in a test set with AD = -0.037 and AAD = 0.345 log units. Thus, the 91-compound equation can now be used to predict further EIT values to around 0.4 log units. It is suggested that the mechanism of action in the Draize test and in the human EIT involves passive transfer of the compound to a biophase that is quite polar, is a strong hydrogen bond base, a moderate hydrogen bond acid, and quite hydrophobic. The biophase does not resemble water or plasma, but resembles an organic solvent such as N-methylformamide.

Harmonization of thresholds for primary skin irritation from results of human repeated insult patch tests and laboratory animal skin irritation tests

Classification of a chemical or chemical product as a primary skin irritant using laboratory animal tests has been defined rigorously and codified in Consumer Products Safety Commission (CPSC) regulations. However, no regulatory agency, including the CPSC, has defined a primary skin irritant threshold for human repeat insult patch tests (RIPTs) that are typically conducted on products, such as cosmetics, anticipated to come into direct contact with humans. Further more, the protocols for animal and human tests are significantly different, as are the schemes for grading responses. Consequently, comparing the results of one type of test with those from the other type has proved to be difficult. In this short communication, we propose a procedure to harmonize the results from these two types of skin irritation tests and suggest that a score of "5" in animal tests, considering 24-h results on unabraded skin, is equivalent to a score of "3" in human RIPTs.

Comparison of in vivo (Draize method) and in vitro (Corrositex assay) dermal corrosion values for selected industrial chemicals

Skin irritation is a common occupational hazard for employees engaged in the manufacture, transport, and use of industrial chemicals. The most common method used to evaluate dermal irritation and/or corrosion has typically been in vivo tests using rabbits (Draize method). Several in vitro test methods have been developed, with Corrositex being the first to gain approval by a regulatory agency (U.S. Department of Transportation). The purpose of this study was to compare the results of in vitro (Corrositex) assays of dermal irritation/corrosion to in vivo test data for several industrial chemical formulations and to determine the predictability and usefulness of the Corrositex assay for these types of products. Twenty-four (24) formulations were qualified, categorized, and evaluated using the Corrositex method and the results compared to available animal data for each of the formulations. The Corrositex assay accurately predicted a corrosive end point in 8 (57.1%) of the 14 formulations identified as corrosive by the in vivo evaluations. Corrositex accurately predicted a noncorrosive end point for 1 (10%) of 10 formulations determined to be noncorrosive in animal studies. The Corrositex assay overpredicted the packing group for 12 (50%) of the 24 formulations, and underpredicted the packing group for 7 (29.2%) of the 24 formulations. Compared to the in vivo results, Corrositex correctly classified as corrosive or noncorrosive 37.5% of the formulations tested. A concordance of 20.8% for the packing group assignments of the evaluated formulations was calculated. The Corrositex assay did not accurately predict a corrosive end point or packing group assignment for all of the formulations used in this study. Manufacturers should assess the relevance of this method to their products prior to relying on it for compliance with hazardous material and worker safety regulations.

An analysis of ISO intracutaneous reactivity test results to justify a reduction in animal requirements

The ISO intracutaneous reactivity test is the standard protocol for determining a medical device's potential for causing irritation. The authors present data indicating that the number of animals required per test can be reduced from three rabbits to two.

Using In Vitro Prediction Models Instead of the Rabbit Eye Irritation Test to Classify and Label New Chemicals: A Post Hoc Data Analysis of the International EC/HO Validation Study

The international validation study on alternative methods to replace the Draize rabbit eye irritation test, funded by the European Commission (EC) and the British Home Office (HO), took place during 1992–1994, and the results were published in 1995. The results of this EC/HO study are analysed by employing discriminant analysis, taking into account the classification of the in vivo data into eye irritation classes A (risk of serious damage to eyes), B (irritating to eyes) and NI (non-irritant). A data set for 59 test items was analysed, together with three subsets: surfactants, water-soluble chemicals, and water-insoluble chemicals. The new statistical methods of feature selection and estimation of the discriminant function's classification error were used. Normal distributed random numbers were added to the mean values of each in vitro endpoint, depending on the observed standard deviations. Thereafter, the reclassification error of the random observations was estimated by applying the fixed function of the mean values. Moreover, the leaving-one-out cross-classification method was applied to this random data set. Subsequently, random data were generated r times (for example, r = 1000) for a feature combination. Eighteen features were investigated in nine in vitro test systems to predict the effects of a chemical in the rabbit eye. 72.5% of the chemicals in the undivided sample were correctly classified when applying the in vitro endpoints lgNRU of the neutral red uptake test and lgBCOPo5 of the bovine opacity and permeability test. The accuracy increased to 80.9% when six in vitro features were used, and the sample was subdivided. The subset of surfactants was correctly classified in more than 90% of cases, which is an excellent performance.

ECVAM's research and validation activities in the fields of topical toxicity and human studies

This paper outlines the research, prevalidation and validation activities that ECVAM has undertaken in collaboration with its partners in the field of topical toxicity testing and human volunteer studies, from its creation until now (1994-2002).

Validation successes: chemicals

The ECVAM validation concept, which was defined at two validation workshops held in Amden (Switzerland) in 1990 and 1994, and which takes into account the essential elements of prevalidation and biostatistically defined prediction models, has been officially accepted by European Union (EU) Member States and by the Federal regulatory agencies of the USA and the OECD. The ECVAM validation concept was introduced into the ongoing ECVAM/COLIPA validation study of in vitro phototoxicity tests, which ended successfully in 1998. The 3T3 neutral red uptake in vitro phototoxicity test was the first experimentally validated in vitro toxicity test recommended for regulatory purposes by the ECVAM Scientific Advisory Committee (ESAC). It was accepted by the EU into the legislation for chemicals in the year 2000. From 1996 to 1998, two in vitro skin corrosivity tests were successfully validated by ECVAM, and they were also officially accepted into the EU regulations for chemicals in the year 2000. Meanwhile, in 2002, the OECD Test Guidelines Programme is considering the worldwide acceptance of the validated in vitro phototoxicity and corrosivity tests. Finally, from 1997 to 2000, an ECVAM validation study on three in vitro embryotoxicity tests was successfully completed. Therefore, the three in vitro embryotoxicity tests, the whole embryo culture (WEC) test on rat embryos, the micromass (MM) test on limb bud cells of mouse embryos, and the embryonic stem cell test (EST) including a permanent embryonic mouse stem cell line, are considered for routine use in laboratories of the European pharmaceutical and chemicals industries.

Kinetic analysis on the in vitro cytotoxicity using Living Skin Equivalent for ranking the toxic potential of dermal irritants

The extent of cytotoxicity injured by several skin irritants was kinetically measured and analyzed in a three-dimensional cultured human skin model, Living Skin Equivalent-002 (LSE). Colorimetric thiazoyl blue (MTT) conversion assay was selected as a cytotoxicity assay, and olive oil (OO), lactic acid (LA), Triton X-100 (TX) and sodium lauryl sulfate (SLS) were evaluated as model irritants. OO had almost no effect on the viability of LSE. When the other irritants were applied on the full-thickness LSE, two first-ordered decreasing phases, initial slow and following rapid phases, were found in the viability of LSE. LA and TX showed a bigger difference between the slow and rapid rates than SLS to show an inflection. The inflection time point from the slow to rapid rate was dependent on the kind and concentration of irritants applied. The higher the concentration of irritants applied, the more rapid the inflection point was observed. When LA and SLS were applied on the stratum corneum-stripped LSE, on the other hand, viability was mono-exponentially decreased with time. LA, TX and SLS probably decrease the barrier function of the stratum corneum to increase the rate of cytotoxicity during the irritant application. Interestingly, the rate of cytotoxicity on the stripped skin was similar to the late rapid rate on the full-thickness skin in LA not in SLS. These results suggest that cytotoxicity of skin irritants on the full-thickness LSE can be represented by two first-order kinetics, and that the skin irritation rate is closely related by the barrier function of skin as well as the application concentration and intrinsic toxicity of irritants.

Ocular safety: a silent (in vitro) success story

Ocular irritation testing has been one of the animal test methods most criticised by animal welfare advocates. Additional criticism has arisen from within the scientific community, based on the variability of the animal test results and the questionable relevance of the extremely high dose levels employed. As a result, the Draize eye irritation test has been one of the main targets for in vitro replacement. Despite extensive efforts, however, there is still no in vitro method that is fully validated as a regulatory replacement. In spite of this, many individual companies are using diverse in vitro ocular irritation tests to gain important safety and efficacy information about their products and raw materials, eliminating the need for animal testing in the process. This is done in a safe fashion by applying intelligent testing paradigms. ECVAM has played a major role in this success, through its many programmes that have emphasised the importance of understanding the true toxicological need, and then using in vitro tests to provide that information. Thus, even in the absence of a successfully validated regulatory assay, the desired result of reducing animal testing is being met.

ECVAM's activities in validating alternative tests for skin corrosion and irritation

ECVAM has funded and managed validation studies on in vitro tests for skin corrosion, resulting in the validities of four in vitro tests being endorsed by the ECVAM Scientific Advisory Committee: the rat skin transcutaneous electrical resistance (TER) assay, two tests based on the use of commercial reconstituted human skin equivalents, EPISKIN and EpiDerm, and another commercially-produced test, CORROSITEX. In the European Union (EU), a new test method on skin corrosion (B.40), incorporating the rat skin TER and human skin model assays, was included in Annex V of Directive 67/548/EEC in mid-2000, thereby making the use of in vitro alternatives for skin corrosion testing of chemicals mandatory in the EU. At the recommendation of its Skin Irritation Task Force, ECVAM has funded prevalidation studies on five in vitro tests for acute skin irritation: EpiDerm, EPISKIN, PREDISKIN, the pig-ear test, and the mouse-skin integrity function test (SIFT). However, none of the tests met the criteria (set by the Management Team for the studies) for inclusion in a large-scale formal validation study. Thus, to date, there are no validated in vitro tests for predicting the dermal irritancy of chemicals. Following further work on the EPISKIN, EpiDerm and SIFT test protocols and/or prediction models after the completion of the prevalidation studies, it appears that the modified tests could meet the performance criteria defined for progression to a validation study. This will now be assessed independently by the ECVAM Skin Irritation Task Force, with the objective of taking a decision before the end of 2002 on whether to conduct a formal validation study.

An in vitro model for detecting skin irritants: methyl green-pyronine staining of human skin explant cultures

We evaluated the potential of human organotypic skin explant cultures (hOSECs) for screening skin irritants. Test chemicals were applied to the epidermis of the skin explants which were incubated for 4, 24 or 48 h in tissue culture medium. A decrease in epidermal RNA staining, visualised in frozen sections using a modified methyl-green pyronine (MGP) staining procedure, was used as a marker of irritancy. A decrease in epidermal RNA after a 4-, 24- or 48-h exposure to a certain concentration of a test chemical equated to a MGP score of 3, 2 or 1, respectively. The MGP score was 0 if there was no keratinocyte cytotoxicity after a 48-h exposure. A minimum of three donors were used per chemical and the average MGP score was used to classify the chemical as irritant or not. Chemicals with an average MGP score > or =1.5 were classified as irritants (R38), at that concentration. Chemicals with a MGP score <1.5 were not classified (NC), at that concentration. The results obtained using human skin in vitro were compared with published data obtained using cultured porcine skin, the cutaneous Draize test (from this point referred to as the "rabbit skin irritation test") and volunteer studies. There was an excellent correlation between the classification of a chemical, as R38 or NC, based on hOSEC and results of volunteer studies. The hOSEC model predicted perfectly the irritation hazard of the 22 chemicals for which volunteer data were available. The porcine OSEC correctly predicted the classification of 21 of 22 (95%) chemicals and the rabbit skin irritation test correctly predicted the classification of 14 of 15 chemicals (93%) for which data were available. In conclusion, MGP staining of human skin explant cultures can be used to predicted human skin irritancy in vivo. In addition, the data validate the use of porcine skin as an alternative to human skin for screening for dermal irritants in vitro.