Skin sensitization risk assessment model using artificial neural network analysis of data from multiple in vitro assays

The sensitizing potential of chemicals is usually identified and characterized using in vivo methods such as the murine local lymph node assay (LLNA). Due to regulatory constraints and ethical concerns, alternatives to animal testing are needed to predict skin sensitization potential of chemicals. For this purpose, combined evaluation using multiple in vitro and in silico parameters that reflect different aspects of the sensitization process seems promising.We previously reported that LLNA thresholds could be well predicted by using an artificial neural network (ANN) model, designated iSENS ver.1 (integrating in vitro sensitization tests version 1), to analyze data obtained from two in vitro tests: the human Cell Line Activation Test (h-CLAT) and the SH test. Here, we present a more advanced ANN model, iSENS ver.2, which additionally utilizes the results of antioxidant response element (ARE) assay and the octanol–water partition coefficient (Log P, reflecting lipid solubility and skin absorption). We found a good correlation between predicted LLNA thresholds calculated by iSENS ver.2 and reported values. The predictive performance of iSENS ver.2 was superior to that of iSENS ver.1. We conclude that ANN analysis of data from multiple in vitro assays is a useful approach for risk assessment of chemicals for skin sensitization.     

Integrated decision strategies for skin sensitization hazard

One of the top priorities of the Interagency Coordinating Committee for the Validation of Alternative Methods (ICCVAM) is the identification and evaluation of non‐animal alternatives for skin sensitization testing. Although skin sensitization is a complex process, the key biological events of the process have been well characterized in an adverse outcome pathway (AOP) proposed by the Organisation for Economic Co‐operation and Development (OECD). Accordingly, ICCVAM is working to develop integrated decision strategies based on the AOP using in vitro, in chemico and in silico information. Data were compiled for 120 substances tested in the murine local lymph node assay (LLNA), direct peptide reactivity assay (DPRA), human cell line activation test (h‐CLAT) and KeratinoSens assay. Data for six physicochemical properties, which may affect skin penetration, were also collected, and skin sensitization read‐across predictions were performed using OECD QSAR Toolbox. All data were combined into a variety of potential integrated decision strategies to predict LLNA outcomes using a training set of 94 substances and an external test set of 26 substances. Fifty‐four models were built using multiple combinations of machine learning approaches and predictor variables. The seven models with the highest accuracy (89–96% for the test set and 96–99% for the training set) for predicting LLNA outcomes used a support vector machine (SVM) approach with different combinations of predictor variables. The performance statistics of the SVM models were higher than any of the non‐animal tests alone and higher than simple test battery approaches using these methods. These data suggest that computational approaches are promising tools to effectively integrate data sources to identify potential skin sensitizers without animal testing. Published 2016. This article has been contributed to by US Government employees and their work is in the public domain in the USA.     

Multivariate models for prediction of human skin sensitization hazard

One of the Interagency Coordinating Committee on the Validation of Alternative Method's (ICCVAM) top priorities is the development and evaluation of non‐animal approaches to identify potential skin sensitizers. The complexity of biological events necessary to produce skin sensitization suggests that no single alternative method will replace the currently accepted animal tests. ICCVAM is evaluating an integrated approach to testing and assessment based on the adverse outcome pathway for skin sensitization that uses machine learning approaches to predict human skin sensitization hazard. We combined data from three in chemico or in vitro assays – the direct peptide reactivity assay (DPRA), human cell line activation test (h‐CLAT) and KeratinoSens™ assay – six physicochemical properties and an in silico read‐across prediction of skin sensitization hazard into 12 variable groups. The variable groups were evaluated using two machine learning approaches, logistic regression and support vector machine, to predict human skin sensitization hazard. Models were trained on 72 substances and tested on an external set of 24 substances. The six models (three logistic regression and three support vector machine) with the highest accuracy (92%) used: (1) DPRA, h‐CLAT and read‐across; (2) DPRA, h‐CLAT, read‐across and KeratinoSens; or (3) DPRA, h‐CLAT, read‐across, KeratinoSens and log P. The models performed better at predicting human skin sensitization hazard than the murine local lymph node assay (accuracy 88%), any of the alternative methods alone (accuracy 63–79%) or test batteries combining data from the individual methods (accuracy 75%). These results suggest that computational methods are promising tools to identify effectively the potential human skin sensitizers without animal testing. Published 2016. This article has been contributed to by US Government employees and their work is in the public domain in the USA.     

Improved performance of the SH test as an in vitro skin sensitization test with a new predictive model and decision tree

Demands for the elimination and replacement of animal experiments for cosmetic safety assessment have increased in recent years. Evaluation of skin sensitization, however, is a critical issue in cosmetic safety assessment. The SH test is an in vitro skin sensitization test method that evaluates protein binding of chemical substances, which is an important event in skin sensitization. We previously verified the technical transferability and between-laboratory reproducibility of the SH test, a domestic test method for which no scientific research has been conducted, and improved the protocol, but also noted some unresolved issues. Therefore, in the present study, we successfully improved the operational efficiency and clarity of the final judgment of the SH test by (i) developing a new decision-making system that can make a final judgment without statistical processing, (ii) changing the statistical method, and (iii) evaluating and determining the maximum number of repetitions necessary for optimal efficiency. The improved SH test was verified by comparing it with existing test methods already adopted by the Organization for Economic Cooperation and Development. The results of this study demonstrated excellent performance of the improved SH test, with high reproducibility, reliable predictability, and good operational efficiency. The predictive performance of the improved method does not differ significantly from that of the conventional method, although it is clearer and more efficient. Therefore, the results of the present improved method are consistent with those obtained using the conventional method, with higher efficiency.     

Test battery with the human cell line activation test,direct peptide reactivity assay and DEREK based on a 139 chemical data set for predicting skin sensitizing potential and potency of chemicals

To develop a testing strategy incorporating the human cell line activation test (h‐CLAT), direct peptide reactivity assay (DPRA) and DEREK, we created an expanded data set of 139 chemicals (102 sensitizers and 37 non‐sensitizers) by combining the existing data set of 101 chemicals through the collaborative projects of Japan Cosmetic Industry Association. Of the additional 38 chemicals, 15 chemicals with relatively low water solubility (log Kow > 3.5) were selected to clarify the limitation of testing strategies regarding the lipophilic chemicals. Predictivities of the h‐CLAT, DPRA and DEREK, and the combinations thereof were evaluated by comparison to results of the local lymph node assay. When evaluating 139 chemicals using combinations of three methods based on integrated testing strategy (ITS) concept (ITS‐based test battery) and a sequential testing strategy (STS) weighing the predictive performance of the h‐CLAT and DPRA, overall similar predictivities were found as before on the 101 chemical data set. An analysis of false negative chemicals suggested a major limitation of our strategies was the testing of low water‐soluble chemicals. When excluded the negative results for chemicals with log Kow > 3.5, the sensitivity and accuracy of ITS improved to 97% (91 of 94 chemicals) and 89% (114 of 128). Likewise, the sensitivity and accuracy of STS to 98% (92 of 94) and 85% (111 of 129). Moreover, the ITS and STS also showed good correlation with local lymph node assay on three potency classifications, yielding accuracies of 74% (ITS) and 73% (STS). Thus, the inclusion of log Kow in analysis could give both strategies a higher predictive performance. Copyright © 2015 John Wiley & Sons, Ltd.     

Non‐animal assessment of skin sensitization hazard: Is an integrated testing strategy needed,and if so what should be integrated?

There is an expectation that to meet regulatory requirements, and avoid or minimize animal testing, integrated approaches to testing and assessment will be needed that rely on assays representing key events (KEs) in the skin sensitization adverse outcome pathway. Three non‐animal assays have been formally validated and regulatory adopted: the direct peptide reactivity assay (DPRA), the KeratinoSens? assay and the human cell line activation test (h‐CLAT). There have been many efforts to develop integrated approaches to testing and assessment with the “two out of three” approach attracting much attention. Here a set of 271 chemicals with mouse, human and non‐animal sensitization test data was evaluated to compare the predictive performances of the three individual non‐animal assays, their binary combinations and the “two out of three” approach in predicting skin sensitization potential. The most predictive approach was to use both the DPRA and h‐CLAT as follows: (1) perform DPRA – if positive, classify as sensitizing, and (2) if negative, perform h‐CLAT – a positive outcome denotes a sensitizer, a negative, a non‐sensitizer. With this approach, 85% (local lymph node assay) and 93% (human) of non‐sensitizer predictions were correct, whereas the “two out of three” approach had 69% (local lymph node assay) and 79% (human) of non‐sensitizer predictions correct. The findings are consistent with the argument, supported by published quantitative mechanistic models that only the first KE needs to be modeled. All three assays model this KE to an extent. The value of using more than one assay depends on how the different assays compensate for each other's technical limitations. Copyright © 2017 John Wiley & Sons, Ltd.     

Proposal for a risk assessment methodology for skin sensitization based on sensitization potency data

In this paper, we propose a quantitative risk assessment methodology for skin sensitization aiming at the derivation of 'safe' exposure levels for sensitizing chemicals, used e.g., as ingredients in consumer products. Given the limited number of sensitizers tested in human sensitization tests, such as the human repeat-insult patch test (HRIPT) or the human maximization test (HMT), we used EC3 values from the local lymph node assay (LLNA) in mice because they provide the best quantitative measure of the skin sensitizing potency of a chemical. A comparison of LLNA EC3 values with HRIPT and HMT LOEL, and NOEL values was carried out and revealed that the EC3, expressed as area dose, can be used as a surrogate value for the human NOEL in risk assessment. The uncertainty/extrapolation factor approach was used to derive (a) an 'acceptable non-sensitizing area dose' (ANSAD) to protect non-allergic individuals against skin sensitization and (b) an 'acceptable non-eliciting area dose' (ANEAD) to protect allergic individuals against elicitation of allergic contact dermatitis. For ANSAD derivation, interspecies, intraspecies and time extrapolation factors are applied to the LLNA EC3. For ANEAD derivation, additional application of a variable sensitization-elicitation extrapolation factor is proposed. Values for extrapolation factors are derived and discussed, the proposed methodology is applied to the sensitizers methylchloroisothiazolinone/methylisothiazolinone, cinnamic aldehyde and nickel and results are compared to published risk assessments.     

IL-1α and IL-1β as alternative biomarkers for risk assessment and the prediction of skin sensitization potency

Potential biomarkers of skin sensitization in RAW264.7 mouse macrophages were investigated as alternatives to animal experiments and risk assessment. The concentrations that resulted in a cell viability of 90% (CV90) and 75% (CV75) were calculated by using a water-soluble tetrazolium salt (WST)-1 assay and used to analyze the skin sensitization potency of 23 experimental materials under equivalent treatment conditions. In addition, the expression of interleukin (IL)-1α, IL-1β, IL-31, tumor necrosis factor (TNF)-α, inducible nitric oxide synthase (iNOS), prostaglandin E2 (PGE2), and cyclooxygenase-2 (COX-2) was analyzed utilizing Western blotting. In the cell viability analysis, skin sensitizers were generally more cytotoxic and exhibited increased skin sensitization potency. However, nonsensitizers did not show any marked cytotoxic tendency. Biomarker analysis demonstrated that IL-1α, IL-1β, and the combination of IL-1α and IL-1β (IL-1α + IL-1β) predicted reliably skin sensitization potential (1) sensitivities of 94.4%, 83.3%, and 83.3%, specificities of 100%, 100%, and 100%, and (2) accuracies of 95.7%, 87%, and 87%, respectively. These observations correlated most reliably as indicators for skin sensitization potency. Data suggest that IL-1α and IL-1β may serve as potential biomarkers for skin sensitization and provide an alternative method to animal experiments for prediction of skin sensitization potency and risk assessment.     

An in vitro human skin test for assessing sensitization potential

Sensitization to chemicals resulting in an allergy is an important health issue. The current gold‐standard method for identification and characterization of skin‐sensitizing chemicals was the mouse local lymph node assay (LLNA). However, for a number of reasons there has been an increasing imperative to develop alternative approaches to hazard identification that do not require the use of animals. Here we describe a human in‐vitro skin explant test for identification of sensitization hazards and the assessment of relative skin sensitizing potency. This method measures histological damage in human skin as a readout of the immune response induced by the test material. Using this approach we have measured responses to 44 chemicals including skin sensitizers, pre/pro‐haptens, respiratory sensitizers, non‐sensitizing chemicals (including skin‐irritants) and previously misclassified compounds. Based on comparisons with the LLNA, the skin explant test gave 95% specificity, 95% sensitivity, 95% concordance with a correlation coefficient of 0.9. The same specificity and sensitivity were achieved for comparison of results with published human sensitization data with a correlation coefficient of 0.91. The test also successfully identified nickel sulphate as a human skin sensitizer, which was misclassified as negative in the LLNA. In addition, sensitizers and non‐sensitizers identified as positive or negative by the skin explant test have induced high/low T cell proliferation and IFNγ production, respectively. Collectively, the data suggests the human in‐vitro skin explant test could provide the basis for a novel approach for characterization of the sensitizing activity as a first step in the risk assessment process. Copyright © 2015 John Wiley & Sons, Ltd.     

Probabilistic derivation of the interspecies assessment factor for skin sensitization

An interspecies sensitization assessment factor (SAF) is used in the quantitative risk assessment (QRA) for skin sensitization when a murine-based NESIL (No Expected Sensitization Induction Level) is taken as point of departure. Several studies showed that, on average, the murine sensitization threshold is in good correspondence with that determined in humans. However, on an individual level, the murine and human sensitization thresholds may differ considerably. In this study, the interspecies SAF was quantified by using a probabilistic approach, to be able to take these cases into account. As expected, the geometric means of the probability distributions of murine and human sensitization threshold ratios were close to one, but taking the 95 ^th percentile of these distributions resulted in an interspecies SAF of 15. By using this value, one is sure that with 95% probability, the sensitization threshold determined in mice does not underestimate the human threshold. It can be concluded that a murine-based NESIL requires the use of an interspecies SAF (of 15) in the QRA for skin sensitization, to correct for the differences between mice and humans. This empirically derived interspecies SAF contributes to refinement of the risk assessment methodology.     

Prediction of skin sensitization potency using machine learning approaches

The replacement of animal use in testing for regulatory classification of skin sensitizers is a priority for US federal agencies that use data from such testing. Machine learning models that classify substances as sensitizers or non‐sensitizers without using animal data have been developed and evaluated. Because some regulatory agencies require that sensitizers be further classified into potency categories, we developed statistical models to predict skin sensitization potency for murine local lymph node assay (LLNA) and human outcomes. Input variables for our models included six physicochemical properties and data from three non‐animal test methods: direct peptide reactivity assay; human cell line activation test; and KeratinoSens™ assay. Models were built to predict three potency categories using four machine learning approaches and were validated using external test sets and leave‐one‐out cross‐validation. A one‐tiered strategy modeled all three categories of response together while a two‐tiered strategy modeled sensitizer/non‐sensitizer responses and then classified the sensitizers as strong or weak sensitizers. The two‐tiered model using the support vector machine with all assay and physicochemical data inputs provided the best performance, yielding accuracy of 88% for prediction of LLNA outcomes (120 substances) and 81% for prediction of human test outcomes (87 substances). The best one‐tiered model predicted LLNA outcomes with 78% accuracy and human outcomes with 75% accuracy. By comparison, the LLNA predicts human potency categories with 69% accuracy (60 of 87 substances correctly categorized). These results suggest that computational models using non‐animal methods may provide valuable information for assessing skin sensitization potency. Copyright © 2017 John Wiley & Sons, Ltd.     

Evaluation of in vitro assays for the assessment of the skin sensitization hazard of functional polysiloxanes and silanes

The skin sensitization potential of chemicals has traditionally been evaluated in vivo according to OECD testing guidelines in guinea pigs or the mouse local lymph node assay. There has lately been a great emphasis on establishing in vitro test methods reflecting the key biological events in the adverse outcome pathway (AOP) for skin sensitization as published by the OECD. Against this background, a group of 8 polysiloxanes and silanes, seven of them aminofunctionalised, for which in vivo data were already available, has been tested in vitro in the direct peptide reactivity assay (DPRA), the KeratinoSens™ and the human cell line activation test (h-CLAT) and in the modified myeloid U937 skin sensitization test (mMUSST) as far as technically feasible. The main objective of the programme was to determine the utility of these systems for this heterogeneous group of silicone-based substances, recognizing that some substances are outside the assays applicability domains. The presented data provided some interesting mechanistical insights into the performance of these assays for functionalised siloxanes and silanes. The data also allow for a preliminary evaluation of proposed integrated testing strategies (ITS) to determine the skin sensitization potential of chemicals which were not considered in the training sets of the respective ITS.     

Accounting for data variability,a key factor in in vivo/in vitro relationships: application to the skin sensitization potency (in vivo LLNA versus in vitro DPRA) example

When searching for alternative methods to animal testing, confidently rescaling an in vitro result to the corresponding in vivo classification is still a challenging problem. Although one of the most important factors affecting good correlation is sample characteristics, they are very rarely integrated into correlation studies. Usually, in these studies, it is implicitly assumed that both compared values are error‐free numbers, which they are not. In this work, we propose a general methodology to analyze and integrate data variability and thus confidence estimation when rescaling from one test to another. The methodology is demonstrated through the case study of rescaling the in vitro Direct Peptide Reactivity Assay (DPRA) reactivity to the in vivo Local Lymph Node Assay (LLNA) skin sensitization potency classifications. In a first step, a comprehensive statistical analysis evaluating the reliability and variability of LLNA and DPRA as such was done. These results allowed us to link the concept of gray zones and confidence probability, which in turn represents a new perspective for a more precise knowledge of the classification of chemicals within their in vivo OR in vitro test. Next, the novelty and practical value of our methodology introducing variability into the threshold optimization between the in vitro AND in vivo test resides in the fact that it attributes a confidence probability to the predicted classification. The methodology, classification and screening approach presented in this study are not restricted to skin sensitization only. They could be helpful also for fate, toxicity and health hazard assessment where plenty of in vitro and in chemico assays and/or QSARs models are available. Copyright © 2016 John Wiley & Sons, Ltd.     

Transferability and within‐ and between‐laboratory reproducibilities of EpiSensA for predicting skin sensitization potential in vitro: A ring study in three laboratories

The epidermal sensitization assay (EpiSensA) is an in vitro skin sensitization test method based on gene expression of four markers related to the induction of skin sensitization; the assay uses commercially available reconstructed human epidermis. EpiSensA has exhibited an accuracy of 90% for 72 chemicals, including lipophilic chemicals and pre?/pro‐haptens, when compared with the results of the murine local lymph node assay. In this work, a ring study was performed by one lead and two naive laboratories to evaluate the transferability, as well as within‐ and between‐laboratory reproducibilities, of EpiSensA. Three non‐coded chemicals (two lipophilic sensitizers and one non‐sensitizer) were tested for the assessment of transferability and 10 coded chemicals (seven sensitizers and three non‐sensitizers, including four lipophilic chemicals) were tested for the assessment of reproducibility. In the transferability phase, the non‐coded chemicals (two sensitizers and one non‐sensitizer) were correctly classified at the two naive laboratories, indicating that the EpiSensA protocol was transferred successfully. For the within‐laboratory reproducibility, the data generated with three coded chemicals tested in three independent experiments in each laboratory gave consistent predictions within laboratories. For the between‐laboratory reproducibility, 9 of the 10 coded chemicals tested once in each laboratory provided consistent predictions among the three laboratories. These results suggested that EpiSensA has good transferability, as well as within‐ and between‐laboratory reproducibility.     

Evaluation for skin sensitization based on published literatures (existing information) of major PRTR designated chemical substances in Japan

Of the 354 substances designated as class I under the Pollutant Release and Transfer Register (PRTR) law in Japan, we reviewed the sensitization data of the selected 144 substances and analyzed it from various aspects comparing human and animal data, determining the relationship between skin sensitization and chemical structure and comparing the various international organizations.Although most of them were expected to be hazardous substances, 49 out of the 144 substances lacked both human and animal sensitization data. Positive substances accounted for 69% and 42% of the substances for which sensitization data were available in the case of humans and animals, respectively.A correlation was observed between the chemical structures of the substances and sensitization, despite the relatively few substances examined in this study and the limited homogeneity of the collected data. In particular, epoxides clearly had sensitizing potentials and more than half of carboxylic esters or dicarboxyl anhydrides, aliphatic aldehydes, and aromatic compounds with at least two hydroxyl groups also had sensitizing potentials.Also, this study clearly demonstrated the lack of consistency across the sensitization assessment criteria adopted by different countries or among those adopted by the same country on the basis of different laws or administrative measures.     

Application of in vitro skin penetration measurements to confirm and refine the quantitative skin sensitization risk assessment of methylisothiazolinone

Use of quantitative risk assessment (QRA) for assessing the skin sensitization potential of chemicals present in consumer products requires an understanding of hazard and product exposure. In the absence of data, consumer exposure is based on relevant habits and practices and assumes 100% skin uptake of the applied dose. To confirm and refine the exposure, a novel design for in vitro skin exposure measurements was conducted with the preservative, methylisothiazolinone (MI), in beauty care (BC) and household care (HHC) products using realistic consumer exposure conditions. A difference between measured exposure levels (MELs) for MI in leave-on versus rinse-off BC products, and lower MELs for MI in HHC rinse-off compared to BC products was demonstrated. For repeated product applications, the measured exposure was lower than estimations based on summation of applied amounts. Compared to rinse-off products, leave-on applications resulted in higher MELs, correlating with the higher incidences of allergic contact dermatitis associated with those product types. Lower MELs for MI in rinse-off products indicate a lower likelihood to induce skin sensitization, also after multiple daily applications. These in vitro skin exposure measurements indicate conservatism of default exposure estimates applied in skin sensitization QRA and might be helpful in future risk assessments.     

A quantitative in silico model for predicting skin sensitization using a nearest neighbours approach within expert‐derived structure–activity alert spaces

Dermal contact with chemicals may lead to an inflammatory reaction known as allergic contact dermatitis. Consequently, it is important to assess new and existing chemicals for their skin sensitizing potential and to mitigate exposure accordingly. There is an urgent need to develop quantitative non‐animal methods to better predict the potency of potential sensitizers, driven largely by European Union (EU) Regulation 1223/2009, which forbids the use of animal tests for cosmetic ingredients sold in the EU. A Nearest Neighbours in silico model was developed using an in‐house dataset of 1096 murine local lymph node (LLNA) studies. The EC3 value (the effective concentration of the test substance producing a threefold increase in the stimulation index compared to controls) of a given chemical was predicted using the weighted average of EC3 values of up to 10 most similar compounds within the same mechanistic space (as defined by activating the same Derek skin sensitization alert). The model was validated using previously unseen internal (n = 45) and external (n = 103) data and accuracy of predictions assessed using a threefold error, fivefold error, European Centre for Ecotoxicology and Toxicology of Chemicals (ECETOC) and Globally Harmonized System of Classification and Labelling of Chemicals (GHS) classifications. In particular, the model predicts the GHS skin sensitization category of compounds well, predicting 64% of chemicals in an external test set within the correct category. Of the remaining chemicals in the previously unseen dataset, 25% were over‐predicted (GHS 1A predicted: GHS 1B experimentally) and 11% were under‐predicted (GHS 1B predicted: GHS 1A experimentally). Copyright © 2017 John Wiley & Sons, Ltd.     

Evaluation of physiologically based pharmacokinetic models for use in risk assessment

Physiologically based pharmacokinetic (PBPK) models are sophisticated dosimetry models that offer great flexibility in modeling exposure scenarios for which there are limited data. This is particularly of relevance to assessing human exposure to environmental toxicants, which often requires a number of extrapolations across species, route, or dose levels. The continued development of PBPK models ensures that regulatory agencies will increasingly experience the need to evaluate available models for their application in risk assessment. To date, there are few published criteria or well-defined standards for evaluating these models. Herein, important considerations for evaluating such models are described. The evaluation of PBPK models intended for risk assessment applications should include a consideration of: model purpose, model structure, mathematical representation, parameter estimation, computer implementation, predictive capacity and statistical analyses. Model purpose and structure require qualitative checks on the biological plausibility of a model. Mathematical representation, parameter estimation, computer implementation involve an assessment of the coding of the model, as well as the selection and justification of the physical, physicochemical and biochemical parameters chosen to represent a biological organism. Finally, the predictive capacity and sensitivity, variability and uncertainty of the model are analysed so that the applicability of a model for risk assessment can be determined. Published in 2007 by John Wiley & Sons, Ltd.