Prediction of skin sensitization potency using machine learning approaches

Abstract

The replacement of animal use in testing for regulatory classification of skin sensitizers is a priority for U.S. federal agencies that use data from such testing. Machine learning models that classify substances as sensitizers or nonsensitizers 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: the 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/nonsensitizer responses and then classified the sensitizers as strong or weak sensitizers. The two-tiered model using 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 (60/87 substances correctly categorized). These results suggest that computational models using non-animal methods may provide valuable information for assessing skin sensitization potency.

Alternative Integrated Testing for Skin Sensitization: Assuring Consumer Safety

“…Non-animal testing strategies orientated toward regulatory acceptance for chemicals testing (such as REACH) commonly aim to identify sensitizers, and so usually enhance sensitivity, at the expense of specificity. 29,67,68 By application of the most commonly used strategy, often referred to as the ''2 out of 3'' method, 29,31,50 this dataset delivers a sensitivity of 93%, but with a specificity of only 59% due to the relatively high proportion of false positives. However, in the work reported here, while retaining a high sensitivity in terms of the accurate identification of skin sensitizing chemicals, confidence that an ingredient is truly non-sensitizing (specificity) has been increased to 90%.…”

Artificial Intelligence Applications in Dermatology: Where Do We Stand?

“…Research is also accumulating on using AI to minimize exposure to skin-sensitizing substances (61)(62)(63)(64)(65). A representative example by Zang et al described an application capable of analyzing physiochemical properties of substances (e.g., melting point) and determining whether the substance could be a sensitizer or not (65). This application yielded an accuracy of 81% when the substances were studied in a human cohort.…”

Skin toxicology and 3Rs—Current challenges for public health protection

“…For all other substances, a combination of one to three studies should then lead to correct classification according to potency for sensitisation. There are several validated methods available for such an iterative approach which is even accessible to machine learning . The only limitation is the lack of distinction between sensitising categories 1A and 1B.…”