Non-animal methods to predict skin sensitization (I): the Cosmetics Europe database*

Cosmetics Europe, the European Trade Association for the cosmetics and personal care industry, is conducting a multi-phase program to develop regulatory accepted, animal-free testing strategies enabling the cosmetics industry to conduct safety assessments. Based on a systematic evaluation of test methods for skin sensitization, five non-animal test methods (DPRA (Direct Peptide Reactivity Assay), KeratinoSens^TM, h-CLAT (human cell line activation test), U-SENS^TM, SENS-IS) were selected for inclusion in a comprehensive database of 128 substances. Existing data were compiled and completed with newly generated data, the latter amounting to one-third of all data. The database was complemented with human and local lymph node assay (LLNA) reference data, physicochemical properties and use categories, and thoroughly curated. Focused on the availability of human data, the substance selection resulted nevertheless resulted in a high diversity of chemistries in terms of physico-chemical property ranges and use categories. Predictivities of skin sensitization potential and potency, where applicable, were calculated for the LLNA as compared to human data and for the individual test methods compared to both human and LLNA reference data. In addition, various aspects of applicability of the test methods were analyzed. Due to its high level of curation, comprehensiveness, and completeness, we propose our database as a point of reference for the evaluation and development of testing strategies, as done for example in the associated work of Kleinstreuer et al. We encourage the community to use it to meet the challenge of conducting skin sensitization safety assessment without generating new animal data.     

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.     

Guiding principles for the implementation of non-animal safety assessment approaches for cosmetics: Skin sensitisation

Characterisation of skin sensitisation potential is a key endpoint for the safety assessment of cosmetic ingredients especially when significant dermal exposure to an ingredient is expected. At present the mouse local lymph node assay (LLNA) remains the ‘gold standard’ test method for this purpose however non-animal test methods are under development that aim to replace the need for new animal test data. COLIPA (the European Cosmetics Association) funds an extensive programme of skin sensitisation research, method development and method evaluation and helped coordinate the early evaluation of the three test methods currently undergoing pre-validation. In May 2010, a COLIPA scientific meeting was held to analyse to what extent skin sensitisation safety assessments for cosmetic ingredients can be made in the absence of animal data. In order to propose guiding principles for the application and further development of non-animal safety assessment strategies it was evaluated how and when non-animal test methods, predictions based on physico-chemical properties (including in silico tools), threshold concepts and weight-of-evidence based hazard characterisation could be used to enable safety decisions. Generation and assessment of potency information from alternative tools which at present is predominantly derived from the LLNA is considered the future key research area.     

Computational approaches for skin sensitization prediction

Abstract

Drugs, cosmetics, preservatives, fragrances, pesticides, metals, and other chemicals can cause skin sensitization. The ability to predict the skin sensitization potential and potency of substances is therefore of enormous importance to a host of different industries, to customers’ and workers’ safety. Animal experiments have been the preferred testing method for most risk assessment and regulatory purposes but considerable efforts to replace them with non-animal models and in silico models are ongoing. This review provides a comprehensive overview of the computational approaches and models that have been developed for skin sensitization prediction over the last 10 years. The scope and limitations of rule-based approaches, read-across, linear and nonlinear (quantitative) structure–activity relationship ((Q)SAR) modeling, hybrid or combined approaches, and models integrating computational methods with experimental results are discussed followed by examples of relevant models. Emphasis is placed on models that are accessible to the scientific community, and on model validation. A dedicated section reports on comparative performance assessments of various approaches and models. The review also provides a concise overview of relevant data sources on skin sensitization.     

Estimating skin sensitization potency from a single dose LLNA

Skin sensitization is an important aspect of safety assessment. The mouse local lymph node assay (LLNA) developed in the 1990s is an in vivo test used for skin sensitization hazard identification and characterization. More recently a reduced version of the LLNA (rLLNA) has been developed as a means of identifying, but not quantifying, sensitization hazard. The work presented here is aimed at enabling rLLNA data to be used to give quantitative potency information that can be used, inter alia, in modeling and read-across approaches to non-animal based potency estimation. A probit function has been derived enabling estimation of EC3 from a single dose. This has led to development of a modified version of the rLLNA, whereby as a general principle the SI value at 10%, or at a lower concentration if 10% is not testable, is used to calculate the EC3. This version of the rLLNA has been evaluated against a selection of chemicals for which full LLNA data are available, and has been shown to give EC3 values in good agreement with those derived from the full LLNA.     

Peptide reactivity associated with skin sensitization: The QSAR Toolbox and TIMES compared to the DPRA

The molecular initiating event (MIE) of skin sensitization is the binding of a hapten to dermal proteins. This can be assessed using the in chemico direct peptide reactivity assay (DPRA) or in silico tools such as the QSAR Toolbox and TIMES SS. In this study, the suitability of these methods was analyzed by comparing their results to in vivo sensitization data of LLNA and human studies.Compared to human data, 84% of non-sensitizers and sensitizers yielded consistent results in the DPRA. In silico tools resulted in ‘no alert’ for 83%–100% of the non-sensitizers, but alerted only 55%–61% of the sensitizers. The inclusion of biotic and abiotic transformation simulations yielded more alerts for sensitizers, but simultaneously dropped the number of non-alerted non-sensitizers. In contrast to the DPRA, in silico tools were more consistent with results of the LLNA than human data. Interestingly, the new “DPRA profilers” (QSAR Toolbox) provided unsatisfactory results.Additionally, the results were combined in the ‘2 out of 3’ prediction model with in vitro data derived from LuSens and h-CLAT. Using DPRA results, the model identified 90% of human sensitizers and non-sensitizers; using in silico results (including abiotic and biotic activations) instead of DPRA results led to a comparable high predictivity.     

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.     

Binary test battery with KeratinoSens™ and h-CLAT as part of a bottom-up approach for skin sensitization hazard prediction

Skin sensitization is one of the key safety endpoints for chemicals applied directly to the skin. Several integrated testing strategies (ITS) using multiple non-animal test methods have been developed to accurately evaluate the sensitizing potential of chemicals, but there is no regulatory-accepted ITS to classify a chemical as a non-sensitizer. In this study, the predictive performance of a binary test battery with KeratinoSens™ and h-CLAT compared to the local lymph node assay (LLNA) and human data was examined using comprehensive dataset of 203 chemicals. When two negative results indicate a non-sensitizer, the binary test battery provided sensitivity of 93.4% or 94.4% compared with the LLNA or human data. Taking into account the predictive limitations (i.e. high log Kow, pre-/pro-haptens and acyl transfer agents (or amine-reactive)), the binary test battery had extremely high sensitivity comparable to that of the 3 out of 3 ITS where three negative results of the DPRA, KeratinoSens™ and h-CLAT indicate a non-sensitizer. Therefore, the data from KeratinoSens™ or h-CLAT may provide partly redundant information on the molecular initiating event derived from DPRA. Taken together, the binary test battery of KeratinoSens™ and h-CLAT could be used as part of a bottom-up approach for skin sensitization hazard prediction.     

Putting the parts together: Combining in vitro methods to test for skin sensitizing potentials

Allergic contact dermatitis is a common skin disease and is elicited by repeated skin contact with an allergen. In the regulatory context, currently only data from animal experiments are acceptable to assess the skin sensitizing potential of substances. Animal welfare and EU Cosmetic Directive/Regulation call for the implementation of animal-free alternatives for safety assessments. The mechanisms that trigger skin sensitization are complex and various steps are involved. Therefore, a single in vitro method may not be able to accurately assess this endpoint. Non-animal methods are being developed and validated and can be used for testing strategies that ensure a reliable prediction of skin sensitization potentials. In this study, the predictivities of four in vitro assays, one in chemico and one in silico method addressing three different steps in the development of skin sensitization were assessed using 54 test substances of known sensitizing potential. The predictivity of single tests and combinations of these assays were compared. These data were used to develop an in vitro testing scheme and prediction model for the detection of skin sensitizers based on protein reactivity, activation of the Keap-1/Nrf2 signaling pathway and dendritic cell activation.     

Probabilistic hazard assessment for skin sensitization potency by dose–response modeling using feature elimination instead of quantitative structure–activity relationships

Supervised learning methods promise to improve integrated testing strategies (ITS), but must be adjusted to handle high dimensionality and dose–response data. ITS approaches are currently fueled by the increasing mechanistic understanding of adverse outcome pathways (AOP) and the development of tests reflecting these mechanisms. Simple approaches to combine skin sensitization data sets, such as weight of evidence, fail due to problems in information redundancy and high dimensionality. The problem is further amplified when potency information (dose/response) of hazards would be estimated. Skin sensitization currently serves as the foster child for AOP and ITS development, as legislative pressures combined with a very good mechanistic understanding of contact dermatitis have led to test development and relatively large high‐quality data sets. We curated such a data set and combined a recursive variable selection algorithm to evaluate the information available through in silico, in chemico and in vitro assays. Chemical similarity alone could not cluster chemicals' potency, and in vitro models consistently ranked high in recursive feature elimination. This allows reducing the number of tests included in an ITS. Next, we analyzed with a hidden Markov model that takes advantage of an intrinsic inter‐relationship among the local lymph node assay classes, i.e. the monotonous connection between local lymph node assay and dose. The dose‐informed random forest/hidden Markov model was superior to the dose‐naive random forest model on all data sets. Although balanced accuracy improvement may seem small, this obscures the actual improvement in misclassifications as the dose‐informed hidden Markov model strongly reduced " false‐negatives" (i.e. extreme sensitizers as non‐sensitizer) on all data sets. Copyright © 2015 John Wiley & Sons, Ltd.     

Mechanistic applicability domain classification of a local lymph node assay dataset for skin sensitization

The goal of eliminating animal testing in the predictive identification of chemicals with the intrinsic ability to cause skin sensitization is an important target, the attainment of which has recently been brought into even sharper relief by the EU Cosmetics Directive and the requirements of the REACH legislation. Development of alternative methods requires that the chemicals used to evaluate and validate novel approaches comprise not only confirmed skin sensitizers and non-sensitizers but also substances that span the full chemical mechanistic spectrum associated with skin sensitization. To this end, a recently published database of more than 200 chemicals tested in the mouse local lymph node assay (LLNA) has been examined in relation to various chemical reaction mechanistic domains known to be associated with sensitization. It is demonstrated here that the dataset does cover the main reaction mechanistic domains. In addition, it is shown that assignment to a reaction mechanistic domain is a critical first step in a strategic approach to understanding, ultimately on a quantitative basis, how chemical properties influence the potency of skin sensitizing chemicals. This understanding is necessary if reliable non-animal approaches, including (quantitative) structure-activity relationships (Q)SARs, read-across, and experimental chemistry based models, are to be developed.     

Anchoring molecular mechanisms to the adverse outcome pathway for skin sensitization: Analysis of existing data

Allergic contact dermatitis (ACD) is a hypersensitivity immune response induced by small protein-reactive chemicals. Currently, the murine local lymph node assay (LLNA) provides hazard identification and quantitative estimation of sensitizing potency. Given the complexity of ACD, a single alternative method cannot replace the LLNA, but it is necessary to combine methods through an integrated testing strategy (ITS). In the development of an ITS, information regarding mechanisms and molecular processes involved in skin sensitization is crucial. The recently published adverse outcome pathway (AOP) for skin sensitization captures mechanistic knowledge into key events that lead to ACD. To understand the molecular processes in ACD, a systematic review of murine in vivo studies was performed and an ACD molecular map was constructed. In addition, comparing the molecular map to the limited human in vivo toxicogenomic data available suggests that certain processes are similarly triggered in mice and humans, but additional human data will be needed to confirm these findings and identify differences. To gain insight in the molecular mechanisms represented by various human in vitro systems, the map was compared to in vitro toxicogenomic data. This analysis allows for comparison of emerging in vitro methods on a molecular basis, in addition to mathematical predictive value. Finally, a survey of the current in silico, in chemico, and in vitro methods was used to indicate which AOP key event is modeled by each method. By anchoring emerging classification methods to the AOP and the ACD molecular map, complementing methods can be identified, which provides a cornerstone for the development of a testing strategy that accurately reflects the key events in skin sensitization.     

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.     

LuSens: A keratinocyte based ARE reporter gene assay for use in integrated testing strategies for skin sensitization hazard identification

Allergic contact dermatitis can develop following repeated exposure to allergenic substances. To date, hazard identification is still based on animal studies as non-animal alternatives have not yet gained global regulatory acceptance. Several non-animal methods addressing key-steps of the adverse outcome pathway (OECD, 2012) will most likely be needed to fully address this effect. Among the initial cellular events is the activation of keratinocytes and currently only one method, the KeratinoSens™, has been formally validated to address this event. In this study, a further method, the LuSens assay, that uses a human keratinocyte cell line harbouring a reporter gene construct composed of the antioxidant response element (ARE) of the rat NADPH:quinone oxidoreductase 1 gene and the luciferase gene. The assay was validated in house using a selection of 74 substances which included the LLNA performance standards. The predictivity of the LuSens assay for skin sensitization hazard identification was comparable to other non-animal methods, in particular to the KeratinoSens™. When used as part of a testing battery based on the OECD adverse outcome pathway for skin sensitization, a combination of the LuSens assay, the DPRA and a dendritic cell line activation test attained predictivities similar to that of the LLNA.     

Identifying and characterizing chemical skin sensitizers without animal testing: Colipa’s research and method development program

The sensitizing potential of chemicals is usually identified and characterized using one of the available animal test methods, such as the mouse local lymph node assay. Due to the increasing public and political concerns regarding the use of animals for the screening of new chemicals, the Colipa Skin Tolerance Task Force collaborates with and/or funds research groups to increase and apply our understanding of the events occurring during the acquisition of skin sensitization. Knowledge gained from this research is used to support the development and evaluation of novel alternative approaches for the identification and characterization of skin sensitizing chemicals. At present one in chemico (direct peptide reactivity assay (DPRA)) and two in vitro test methods (cell based assays (MUSST and h-CLAT)) have been evaluated within Colipa inter-laboratory ring trials and accepted by the European Centre for the Validation of Alternative Methods (ECVAM) for pre-validation. Data from all three test methods will be used to support the development of testing strategy approaches for skin sensitizer potency prediction. The replacement of the need for animal testing for skin sensitization risk assessment is viewed as ultimately achievable and the next couple of years should set the timeline for this milestone.     

LuSens: A keratinocyte based ARE reporter gene assay for use in integrated testing strategies for skin sensitization hazard identification

Allergic contact dermatitis can develop following repeated exposure to allergenic substances. To date, hazard identification is still based on animal studies as non-animal alternatives have not yet gained global regulatory acceptance. Several non-animal methods addressing key-steps of the adverse outcome pathway (OECD, 2012) will most likely be needed to fully address this effect. Among the initial cellular events is the activation of keratinocytes and currently only one method, the KeratinoSens™, has been formally validated to address this event. In this study, a further method, the LuSens assay, that uses a human keratinocyte cell line harbouring a reporter gene construct composed of the antioxidant response element (ARE) of the rat NADPH:quinone oxidoreductase 1 gene and the luciferase gene. The assay was validated in house using a selection of 74 substances which included the LLNA performance standards. The predictivity of the LuSens assay for skin sensitization hazard identification was comparable to other non-animal methods, in particular to the KeratinoSens™. When used as part of a testing battery based on the OECD adverse outcome pathway for skin sensitization, a combination of the LuSens assay, the DPRA and a dendritic cell line activation test attained predictivities similar to that of the LLNA.     

Predicting skin sensitization potential and inter-laboratory reproducibility of a human Cell Line Activation Test (h-CLAT) in the European Cosmetics Association (COLIPA) ring trials

Regulatory policies in Europe prohibited the testing of cosmetic ingredients in animals for a number of toxicological endpoints. Currently no validated non-animal test methods exist for skin sensitization. Evaluation of changes in cell surface marker expression in dendritic cell (DC)-surrogate cell lines represents one non-animal approach. The human Cell Line Activation Test (h-CLAT) examines the level of CD86 and CD54 expression on the surface of THP-1 cells, a human monocytic leukemia cell line, following 24 h of chemical exposure. To examine protocol transferability, between-lab reproducibility, and predictive capacity, the h-CLAT has been evaluated by five independent laboratories in several ring trials (RTs) coordinated by the European Cosmetics Association (COLIPA). The results of the first and second RTs demonstrated that the protocol was transferable and basically had good between-lab reproducibility and predictivity, but there were some false negative data. To improve performance, protocol and prediction model were modified. Using the modified prediction model in the first and second RT, accuracy was improved. However, about 15% of the outcomes were not correctly identified, which exposes some of the limitations of the assay. For the chemicals evaluated, the limitation may due to chemical being a weak allergen or having low solubility (ex. α-hexylcinnamaldehyde). The third RT evaluated the modified prediction model and satisfactory results were obtained. From the RT data, the feasibility of utilizing cell lines as surrogate DC in development of in vitro skin sensitization methods shows promise. The data also support initiating formal pre-validation of the h-CLAT in order to fully understand the capabilities and limitations of the assay.