Development of an Endpoint-Specific Fingerprint to Identify Structurally Similar Chemicals for Skin Sensitization...

Area of Science:

• Toxicology
• Computational Chemistry
• Risk Assessment

Background:

• Human health risk assessment for cosmetics requires evaluating skin sensitization.
• Next Generation Risk Assessment (NGRA) promotes animal-free testing methods.
• Read-across (RAx) is a key NGRA approach, using data from similar chemicals to fill knowledge gaps.

Purpose of the Study:

• To develop and evaluate a novel chemical fingerprint optimized for skin sensitization RAx.
• To address the challenge of selecting appropriate analogues for RAx.
• To establish guidelines for fingerprint selection in skin sensitization assessments.

Main Methods:

• Utilized a large dataset of murine local lymph node assay (LLNA)-positive chemicals.
• Evaluated the performance of various chemical fingerprints for RAx.
• Developed a novel fingerprint by integrating Protein Binding Alert-based Fingerprint (PBAF) and Klekota-Roth fingerprint (KRFP) features into ToxPrint.

Main Results:

• The novel fingerprint, incorporating PBAF and KRFP features within ToxPrint, demonstrated superior performance.
• This optimized fingerprint effectively distinguished suitable from unsuitable chemical pairs for RAx.
• Fingerprint choice significantly impacts Tanimoto coefficient (Tc) calculations and analogue selection.

Conclusions:

• Endpoint-specific fingerprint development is crucial for enhancing analogue selection in RAx.
• The developed fingerprint improves the reliability of skin sensitization risk assessments.
• This work supports the advancement of animal-free safety testing methodologies.