An investigation of new toxicity test method performance in validation studies: 1. Toxicity test methods that have predictive capacity no greater than chance.

An approach commonly used to measure new toxicity test method (NTM) performance in validation studies is to divide toxicity results into positive and negative classifications, and the identify true positive (TP), true negative (TN), false positive (FP) and false negative (FN) results. After this step is completed, the contingent probability statistics (CPS), sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) are calculated. Although these statistics are widely used and often the only statistics used to assess the performance of toxicity test methods, there is little specific guidance in the validation literature on what values for these statistics indicate adequate performance. The purpose of this study was to begin developing data-based answers to this question by characterizing the CPS obtained from an NTM whose data have a completely random association with a reference test method (RTM). Determining the CPS of this worst-case scenario is useful because it provides a lower baseline from which the performance of an NTM can be judged in future validation studies. It also provides an indication of relationships in the CPS that help identify random or near-random relationships in the data. The results from this study of randomly associated tests show that the values obtained for the statistics vary significantly depending on the cut-offs chosen, that high values can be obtained for individual statistics, and that the different measures cannot be considered independently when evaluating the performance of an NTM. When the association between results of an NTM and RTM is random the sum of the complementary pairs of statistics (sensitivity + specificity, NPV + PPV) is approximately 1, and the prevalence (i.e., the proportion of toxic chemicals in the population of chemicals) and PPV are equal. Given that combinations of high sensitivity-low specificity or low specificity-high sensitivity (i.e., the sum of the sensitivity and specificity equal to approximately 1) indicate lack of predictive capacity, an NTM having these performance characteristics should be considered no better for predicting toxicity than by chance alone.

An investigation of new toxicity test method performance in validation studies: 2. Comparison of three measures of toxicity test performance.

An area that requires further research is how best to measure test method performance in validation studies and how to set criteria that should be used to judge the adequacy of this performance. The studies reported here were designed to begin an investigation of these questions. Computer simulations were used to generate data sets similar to those that might be obtained from a large validation study. These data were then analysed using three procedures including determination of the 95% prediction interval (PI), calculation of Pearson's correlation coefficient and calculation of the contingent probability statistics (CPS), sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). The results of this work suggest that of the three approaches examined, quantitative measurements with calculation of the 95% PI provide the most information to allow discrimination between the performance of several different NTMs. The results also suggest that dividing data sets into positive and negative toxicity classifications followed by the calculation of CPS leads to considerable information loss. This loss of information may be so significant that it is not possible in certain circumstances to distinguish between NTMs that are adequate and those that are not.

An investigation of new toxicity test method performance in validation studies: 3. Sensitivity and specificity are not independent of prevalence or distribution of toxicity.

Often, the only measures of toxicity test performance provided in validation studies are the contingent probability statistics (CPS) sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). Sensitivity and specificity are generally used in preference to NPV and PPV since NPV and PPV are assumed to vary with changes in prevalence while sensitivity and specificity are assumed to be independent of changes in prevalence. The purpose of the studies reported here was to test whether or not sensitivity and specificity are actually independent of changes in prevalence. Results derived from these studies indicate that sensitivity and specificity vary significantly depending on the prevalence of toxic substances in the set of chemicals being tested. This means sensitivity and specificity should not always be considered constant indicators of toxicity test performance.

In vitro human tissue models in risk assessment: report of a consensus-building workshop.

Advances in the technology of human cell and tissue culture and the increasing availability of human tissue for laboratory studies have led to the increased use of in vitro human tissue models in toxicology and pharmacodynamics studies and in quantitative modeling of metabolism, pharmacokinetic behavior, and transport. In recognition of the potential importance of such models in toxicological risk assessment, the Society of Toxicology sponsored a workshop to evaluate the current status of human cell and tissue models and to develop consensus recommendations on the use of such models to improve the scientific basis of risk assessment. This report summarizes the evaluation by invited experts and workshop attendees of the current status of such models for prediction of human metabolism and identification of drug-drug interactions, prediction of human toxicities, and quantitative modeling of pharmacokinetic and pharmaco-toxicodynamic behavior. Consensus recommendations for the application and improvement of current models are presented.

Eye Irritation Testing: The Way Forward. The Report and Recommendations of ECVAM Workshop 34.

This is the report of the thirty-fourth of a series of workshops organised by the European Centre for the Validation of Alternative Methods (ECVAM). ECVAM's main goal, as defined in 1993 by its Scientific Advisory Committee, is to promote the scientific and regulatory acceptance of alternative methods which are of importance to the biosciences and which reduce, refine or replace the use of laboratory animals. One of the first priorities set by ECVAM was the implementation of procedures which would enable it to become well-informed about the state-of-the-art of non-animal test development and validation, and the potential for the possible incorporation of alternative tests into regulatory procedures. It was decided that this would be best achieved by the organisation of ECVAM workshops on specific topics, at which small groups of invited experts would review the current status of various types of in vitro tests and their potential uses, and make recommendations about the best ways forward (1). The workshop on Eye Irritation Testing: The Way Forward was held in Egham, UK, on 15-17 June 1998, under the chairmanship of Michael Balls (ECVAM, Italy). The workshop had two aims, the first of which was to review some of the previous multi-laboratory validation studies on alternatives to the Draize eye test and assess why many promising alternative methods were not successful in these studies. The second aim was to discuss strategies for making progress toward the short-term reduction, refinement, and eventual replacement, of the Draize test, including: a new approach to the validation of in vitro tests for eye irritancy, based on the use of reference standards, which promises to overcome some of the problems encountered in previous studies; the use of stepwise testing strategies which reduce and refine the use of animals in eye irritation testing; the use of multivariate and other statistical techniques for the further analysis of data generated in previous validation studies; and a programme of research aimed at understanding the underlying mechanisms of eye irritation.

Validation of alternative methods for toxicity testing.

Before nonanimal toxicity tests may be officially accepted by regulatory agencies, it is generally agreed that the validity of the new methods must be demonstrated in an independent, scientifically sound validation program. Validation has been defined as the demonstration of the reliability and relevance of a test method for a particular purpose. This paper provides a brief review of the development of the theoretical aspects of the validation process and updates current thinking about objectively testing the performance of an alternative method in a validation study. Validation of alternative methods for eye irritation testing is a specific example illustrating important concepts. Although discussion focuses on the validation of alternative methods intended to replace current in vivo toxicity tests, the procedures can be used to assess the performance of alternative methods intended for other uses.

In vitro alternatives for ocular irritation.

The necessity of using animals to test whether new chemicals and products are eye irritants has been questioned with increasing frequency and fervor over the last 20 years. During this time many new nonanimal methods have been proposed as reliable alternatives to the traditional rabbit (Draize) test. To date, however, none of these nonanimal (in vitro) tests have become universally accepted as a complete replacement for the Draize test. To understand why a complete replacement has not been found, one has to first understand the reasonably complex structure of the eye, the standard Draize scoring scale--which is based on a qualitative evaluation of three different tissues--the differences between human and rabbit eyes, the intrinsic variability of the animal test, and the details of the different in vitro tests that have been proposed as replacements. The in vitro tests vary from relatively simple assays using single cells to more sophisticated assays that use discarded animal tissue or artificially constructed human tissue. It is clear that appropriately designed in vitro tests will eventually give more useful mechanistic information about ocular injury from which we can more comfortably predict the risk of human eye irritation from new products and ingredients.

The ECVAM International Validation Study on In Vitro Tests for Skin Corrosivity. 2. Results and Evaluation by the Management Team.

As a follow-up to a prevalidation study on in vitro tests for replacing the in vivo rabbit test for skin corrosivity, an international validation study was conducted during 1996 and 1997 under the auspices of ECVAM. The main objectives of the study were to: (a) identify tests capable of discriminating corrosives from non-corrosives for selected types of chemicals and/or all chemicals; and (b) determine whether these tests could identify correctly known R35 (UN packing group I) and R34 (UN packing groups II & III) chemicals. The tests evaluated were the rat skin transcutaneous electrical resistance (TER) assay, CORROSITEX(TM), the Skin(2TM) ZK1350 corrosivity test and EPISKIN(TM). Each test was conducted in three independent laboratories. 60 coded chemicals were tested. All of the tests evaluated showed acceptable intralaboratory and interlaboratory reproducibilities, and the TER, Skin(2) and EPISKIN tests proved applicable to testing a diverse group of chemicals of different physical forms, including organic acids, organic bases, neutral organics, inorganic acids, inorganic bases, inorganic salts, electrophiles, phenols and soaps/surfactants. Two of the four tests evaluated, the TER assay and EPISKIN, met the criteria agreed by the Management Team concerning acceptable underprediction and overprediction rates for them to be considered scientifically validated for use as replacements for the animal test for distinguishing between corrosive and non-corrosive chemicals for all of the chemical types studied [objective (a)]. EPISKIN was the only test able to distinguish between known R35 (UN packing group I) and R34 (UN packing groups II & III) chemicals, for all of the chemical types included, on an acceptable number of occasions [objective (b)]. The corrosive potentials of about 40% of the test chemicals could not be assessed with CORROSITEX, and the assay did not meet all of the criteria for it to be considered acceptable as a replacement test. However, CORROSITEX may be valid for testing specific classes of chemicals, such as organic bases and inorganic acids. The Skin(2) assay did not meet the criteria for it to be considered scientifically validated. Thus, the validities of (i) the TER and EPISKIN assays for discriminating corrosives from non-corrosives, and (ii) the EPISKIN assay for identifying correctly known R35/I and R34/II & III chemicals, have been demonstrated in this study. CORROSITEX appears to be valid when used only with certain types of chemicals.

An evaluation of the proposed OECD testing strategy for skin corrosion.

The use of testing strategies which incorporate a range of alternative methods and which use animals only as a last resort is widely considered to provide a reliable way of predicting chemical toxicity while minimising animal testing. The widespread concern over the severity of the Draize rabbit test for assessing skin irritation and corrosion led to the proposal of a stepwise testing strategy at an OECD workshop in January 1996. Subsequently, the proposed testing strategy was adopted, with minor modifications, by the OECD Advisory Group on Harmonization of Classification and Labelling. This article reports an evaluation of the proposed OECD testing strategy as it relates to the classification of skin corrosives. By using a set of 60 chemicals, an assessment was made of the effect of applying three steps in the strategy, taken both individually and in sequence. The results indicate that chemicals can be classified as corrosive (C) or non-corrosive (NC) with sufficient reliability by the sequential application of three alternative methods, i.e., structure-activity relationships (where available), pH measurements, and a single in vitro method (either the rat skin transcutaneous electrical resistance (TER) assay or the EPISKIN™ assay). It is concluded that the proposed OECD strategy for skin corrosion can be simplified without compromising its predictivity. For example, it does not appear necessary to measure acid/alkali reserve (buffering capacity) in addition to pH for the classification of pure chemicals.

IRAG working group 5. Other assays. Interagency Regulatory Alternatives Group.

As part of the Interagency Regulatory Alternatives Group (IRAG) program to evaluate the state of the art in the development of alternative (non-whole animal) eye irritation tests, academic and industrial organizations were invited to submit in vitro eye irritation data generated in their laboratories to one of several working groups for review. The assays reviewed in this report (from Working Group 5. "Other Assays") were the EYTEX assay, tissue equivalent assay, a cytotoxicity assay using three-dimensional human fibroblast constructs, the Microtox assay, and other miscellaneous assays. Each submission consisted of raw data for chemicals and products tested, a description of the methodology, and an analysis (generally by regression analysis and Pearson's correlation coefficient) for the performance of the in vitro test relative to its ability to predict individual ocular tissue scores or total ocular score. In vivo data were generated according to the scoring methods proposed by Draize. Working Group 5 evaluated the submissions and commented on the utility of the assays. The variability of the in vivo data made conclusions difficult in many situations. Most of these assays were deemed useful (within limited chemical classes) for screening purposes or for use in conjunction with other toxicological information.

The CTFA Evaluation of Alternatives Program: an evaluation of in vitro alternatives to the Draize primary eye irritation test. (Phase III) surfactant-based formulations.

The CTFA Evaluation of Alternatives Program is an evaluation of the relationship between data from the Draize primary eye irritation test and comparable data from a selection of promising in vitro eye irritation tests. In Phase III, data from the Draize test and 41 in vitro endpoints on 25 representative surfactant-based personal care formulations were compared. As in Phase I and Phase II, regression modelling of the relationship between maximum average Draize score (MAS) and in vitro endpoint was the primary approach adopted for evaluating in vitro assay performance. The degree of confidence in prediction of MAS for a given in vitro endpoint is quantified in terms of the relative widths of prediction intervals constructed about the fitted regression curve. Prediction intervals reflect not only the error attributed to the model but also the material-specific components of variation in both the Draize and the in vitro assays. Among the in vitro assays selected for regression modeling in Phase III, the relationship between MAS and in vitro score was relatively well defined. The prediction bounds on MAS were most narrow for materials at the lower or upper end of the effective irritation range (MAS = 0-45), where variability in MAS was smallest. This, the confidence with which the MAS of surfactant-based formulations is predicted is greatest when MAS approaches zero or when MAS approaches 45 (no comment is made on prediction of MAS > 45 since extrapolation beyond the range of observed data is not possible). No single in vitro endpoint was found to exhibit relative superiority with regard to prediction of MAS. Variability associated with Draize test outcome (e.g. in MAS values) must be considered in any future comparisons of in vivo and in vitro test results if the purpose is to predict in vivo response using in vitro data.

Validation of alternative methods for toxicity testing.

Many studies have been conducted in order to assess the validity of alternative methods as replacements forin vivo toxicity tests. The purpose of this review is to build on what has been learned in the course of this work by presenting a practical process that can be used to conduct future validation programmes. The important role of a clearly stated prediction model, which defines how to use the results from an alternative method to predict anin vivo toxicity endpoint, has been introduced. Computer simulations have been used to demonstrate that data-based guidance can be developed to assist in judging the performance of alternative methods assessed in a validation study. Additionally, statistical procedures have been used in order to provide guidance on choosing the appropriate number of reference test substances and number of participating laboratories to include in a validation study. The validation of alternative methods for eye irritation testing is used as a specific example to illustrate important concepts. Although the focus of the discussion is on the validation of alternative methods intended to replace currentin vivo tests, the procedures can be used to assess the performance of alternative methods intended for other uses. This review will be particularly useful to those who require a practical guide for conducting a validation study and to those who must assess the results of such programmes.

A collaborative evaluation of seven alternatives to the Draize eye irritation test using pharmaceutical intermediates.

Much of the data which have been generated on in vitro alternatives to the Draize eye irritation test have dealt with compounds within a specific chemical class or product category. However, in the pharmaceutical industry, it is often necessary to evaluate materials which are not related in structure or properties. It was thus decided to evaluate a diverse series of chemicals in seven in vitro methods for estimating ocular irritation. Thirty-seven test materials were chosen to represent a broad range of pH, solubility, and in vivo irritation potential. Assays were chosen to include as many different types of end points as practical. The group of assays was composed of TOPKAT (assessing structure-activity relationships), bovine corneal opacity-permeability (BCO-P; corneal opacity/toxicity), Eytex (protein coagulation), neutral red uptake (cytotoxicity), MTT in living dermal equivalent (cytotoxicity), Microtox (cytotoxicity in bacteria), and CAMVA (inflammation/toxicity). The results of the study indicated that, in general, the cytotoxicity end points did not correlate well with the in vivo data. The BCO-P, CAMVA, and Eytex assays had the best overall concordance (88.9, 75.8, and 75.0%, respectively) with this set of compounds. Estimation of irritation potential based on structure-activity (TOPKAT) was possible for only approximately 50% of the compounds; however, the assay showed 100% sensitivity (i.e., no false negatives), but low specificity (i.e., negatives correctly identified only 54.5% of the time). These data suggest that for screening of chemicals of diverse structure and properties, the more mechanism-based assays, as opposed to general cytotoxicity assays, hold more promise and should be further evaluated.

The CTFA Evaluation of Alternatives Program: an evaluation of in vitro alternatives to the Draize primary eye irritation test. (Phase II) oil/water emulsions.

The Cosmetic, Toiletry and Fragrance Association (CTFA) Evaluation of Alternatives Program is an evaluation of the relationship between Draize ocular safety test data and comparable data from a selection of in vitro tests. In Phase II, 18 representative oil/water-based personal-care formulations were subjected to the Draize primary eye safety test and 30 in vitro assay protocols (14 different types of in vitro endpoints were evaluated; the remainder were protocol variations). Correlation of in vitro with in vivo data was evaluated using analysis of sensitivity/specificity and statistical analysis of the relationship between maximum average Draize score (MAS) and in vitro endpoint. Regression modelling is the primary approach adopted in the CTFA Program for evaluating in vitro assay performance. The objective of regression analysis is to predict MAS for a given test material (and to place upper and lower prediction interval bounds on the range in which the MAS is anticipated to fall with high probability) conditional on observing an in vitro assay score for that material. The degree of confidence in prediction is quantified in terms of the relative widths of prediction intervals constructed about the fitted regression curves: the narrower the prediction interval, the more predictive of the Draize score is the in vitro test result. 16 assays were shown to have the greatest agreement with the Draize procedure and were therefore selected for regression analysis. Based on the magnitude of the 95% prediction bounds of each of the 16 selected assays over the range of test data, it may be inferred that prediction of MAS values from experimentally determined in vitro scores is more accurate for oil/water-based formulations with lower rather than higher irritancy potential. The assays selected for modelling in Phase II generally exhibited weaker relationships with MAS than those selected in Phase I (evaluated using hydroalcoholic formulations), even though several assays were common to both Phases.

Pathobiological effects of acetaldehyde in cultured human epithelial cells and fibroblasts.

The ability of acetaldehyde, a respiratory carcinogen present in tobacco smoke and automotive emissions, to affect cell viability, thiol status and intracellular Ca2+ levels and to cause DNA damage and mutations has been studied using cultured human cells. Within a concentration range of 3-100 mM, a 1 h exposure to acetaldehyde decreases colony survival and inhibits uptake of the vital dye neutral red in bronchial epithelial cells. Acetaldehyde also causes both DNA interstrand cross-links and DNA protein cross-links whereas no DNA single strand breaks are detected. The cellular content of glutathione is also decreased by acetaldehyde, albeit, without concomitant changes in the glutathione redox status or in the content of protein thiols. Transient or sustained increases in cytosolic Ca2+ occur within minutes following exposure of cells to acetaldehyde. Moreover, acetaldehyde significantly decreases the activity of the DNA repair enzyme O6-methylguanine-DNA methyltransferase. Finally, a 5 h exposure to acetaldehyde causes significant levels of 6-thioguanine resistance mutations in an established mutagenesis model involving skin fibroblasts. The results indicate that mM concentrations of acetaldehyde cause a wide range of cytopathic effects associated with multistep carcinogenesis. The fact that acetaldehyde, in relation to its cytotoxicity, causes comparatively higher genotoxicity and inhibits DNA repair more readily than other major aldehydes in tobacco smoke and automotive emissions is discussed.

Cytotoxicity of tacrine and velnacrine metabolites in cultured rat, dog and human hepatocytes.

Clinical trials with tacrine (THA) and its principal (1-OH) metabolite (velnacrine) for the treatment of Alzheimer's disease have been hampered by adverse hepatic events that were undetected in preclinical studies. As part of integrated in vivo/in vitro efforts to characterize the role of metabolites in these events, cultured cells were evaluated for their suitability for further mechanistic studies. The relative cytotoxic potentials of THA, three monohydroxy metabolites of THA (including velnacrine, a racemate), the two velnacrine enantiomers, and several known and suspected dihydroxy velnacrine metabolites were determined. Cytotoxicity was evaluated in 24-hour cultures by morphology and by the Neutral Red Uptake Assay. All test articles were evaluated in primary rat hepatocytes and in a human hepatoma cell line (HepG2). THA and velnacrine were also tested in a rat hepatoma cell line (H4) and in primary dog hepatocytes. The metabolic competency of each cell type was determined. Sensitivity to THA and velnacrine was greatest in H4 cells, followed by primary rat and HepG2 cells; dog cells were least sensitive. In HepG2 cells, THA was clearly more cytotoxic (LC50:54 micrograms/ml) than its monohydroxy metabolites (LC50 values: 84 to 190 micrograms/ml); dihydroxy velnacrine metabolites were the least cytotoxic (LC50 values:251 to 434 micrograms/ml); the relative order was comparable in primary rat hepatocytes. Roles for reactive metabolites and/or altered metabolic capabilities of Alzheimer's patients are suggested.

Genetic toxicology of acrylic acid.

Acrylic acid was tested for gene mutations in the in vitro CHO/HGPRT assay, for chromosome aberrations in CHO cells in culture, and for potential to induce unscheduled DNA synthesis in rat hepatocytes in culture. In vivo assays performed included the Drosophila sex-linked recessive lethal assay by both the feeding and injection routes, the in vivo cytogenetic assay in rat bone marrow cells after both a 1-day and 5-day oral dosing regimen, and a dominant lethal assay in mice by both an acute and 5-day dosing regimen. All results were negative (non-mutagenic) except for the in vitro chromosome aberration assay. This latter result is consistent with the previously reported possible clastogenic activity suggested by the results of the mouse lymphoma L5178Y TK locus assay in which a predominance of small-colony mutants was observed (Moore et al., Environmental and Molecular Mutagenesis 1988, 11, 49-63). The rapid clearance of acrylic acid in animals and the weight of evidence of genetic toxicity testing, including negative in vivo data in both somatic and germ cells, indicate a lack of genetic toxicity of acrylic acid in vivo.

Evaluation of the human epidermal keratinocyte neutral red release and neutral red uptake assay using the first 10 MEIC test materials.

Two methodologies used in vitro to estimate cytotoxicity in cell culture systems were compared: these were the neutral red uptake assay (NRU), which is used to measure toxicity caused by an extended (48-hr) exposure to the test material, and the neutral red release assay (NRR), which is used to measure toxicity caused by a short-term (1-min) exposure to the test material. Both methodologies used the normal human epidermal keratinocyte (NHEK)-based NeutralRed Bioassay supplied by Clonetics Corporation (San Diego, CA, USA). 10 materials (paracetamol, acetylsalicylic acid, ferrous sulphate, diazepam, amitriptyline, digoxin, ethylene glycol, methanol, ethanol and isopropanol), which are part of the Multicenter Evaluation of In Vitro Cytotoxicity (MEIC) panel, were tested. NRU(50) values for the 10 compounds covered more than an eight-log range from 0.004 mum (digoxin) to 1.0 x 10(6) mum (methanol). Because of solubility limits, NRR(50) values for diazepam, digoxin, ferrous sulphate and paracetamol could not be determined. NRR(50) values for the remaining six compounds covered approximately a three-log range from 3.2 x 10(3) to 7.1 x 10(6) mum. When compared with documented values for either the human acute oral lethal dose or the human acute lethal blood concentration, the NRU assay was found to be much more useful in predicting human acute toxicity than was the NRR assay.

Genotoxicity evaluation of lithium hypochlorite.

Lithium hypochlorite (LiOCl), the pool and spa sanitizer/algicide, was evaluated for genotoxicity in a battery of studies designed to evaluate potential mutagenicity, DNA damage and chromosome aberrations. LiOCl was not mutagenic in the Ames test when tested in Salmonella typhimurium strains TA98, TA100, TA1535, TA1537, TA1538 or in the hypoxanthine-guanine phosphoribosyl transferase (HGPRT) mutation assay in Chinese hamster ovary (CHO) cells without metabolic activation. LiOCl did not induce DNA damage in the unscheduled DNA synthesis assay using rat primary hepatocytes. Effects on metaphase chromosomes were evaluated in vitro in CHO cells at 12 and 18 h exposure without S9 and at 12 and 22 h following a 2 h exposure with S9. LiOCl induced a statistically significant increase in chromosome aberrations at the high dose only at both harvest times without S9 and at the late harvest time with S9. There were significant increases in chromosome aberrations at the low dose, low-mid and high doses, but not at the high mid-dose at the early harvest time with S9. However, LiOCl did not increase chromosome aberrations when tested orally in rats at maximally tolerated doses. Bone marrow cells, collected 6, 24 and 48 h after a single oral dose of LiOCl to rats (100, 500, 1000 mg/kg in males; 50, 250, 500 mg/kg in females) showed no increase in the incidence of aberrations. In general, the weight of the evidence indicates that LiOCl is not genotoxic.

Evaluation of two in vitro assays to screen for potential developmental toxicants.

To evaluate two in vitro assays for their ability to detect known developmental toxicants and nontoxicants, a series of 44 coded compounds were assayed by two independent laboratories using standardized protocols. The two test systems were (1) the human embryonic palatal mesenchymal cell growth inhibition assay and (2) the mouse ovarian tumor cell attachment inhibition assay. After all compounds were tested, they were decoded and ranked according to the minimum IC50 value (the millimolar concentration of compound which inhibits growth or attachment by 50% compared to the solvent control) from either test. The in vitro test result concordance with established in vivo animal and human test results was examined over a wide range of concentration levels (above which the in vitro results were called positive and below which they were considered negative). A positive response from either test was defined as a positive in vitro response. Concordance was defined as the number of correct responses divided by the number of chemicals tested. At the 1 mM level, the concordance of data from the combined in vitro assays with the in vivo data was 66% in one laboratory and 58% in the other. The maximum agreement between the combined in vitro and in vivo data was reached at the 20 mM level, where there was a 73 and 74% concordance of results in the two laboratories. At that level there was a 16 and 10% incidence of false negative results, and a 54 and 77% incidence of false positive results. A portion of these false negative compounds may require metabolic activation. The use of either assay alone was not as accurate as using a positive result from either test. Agreement of the in vitro data at the 10 mM level with available human data was 71 and 75% for each laboratory. The data indicate that the two assays are complimentary and as such the combination of these assays could be useful as a preliminary screen to establish priorities for in vivo developmental toxicity testing.