Enhancing Genotoxicity Assessment: New Insights on OECD QSAR Toolbox Profilers
- Elena Fioravanzo
- Jul 17
- 2 min read
In the evolving landscape of chemical safety and regulatory toxicology, Quantitative Structure-Activity Relationship (QSAR) models and in silico tools are becoming increasingly vital for efficient and ethical genotoxicity assessment. These tools offer a powerful first-tier screening approach, helping to prioritize compounds and reduce reliance on animal testing.
We recently came across a highly relevant and insightful paper that sheds light on the performance and application of a key tool in this domain: "Validation of OECD QSAR Toolbox profilers for genotoxicity assessment of pesticides using the MultiCase genotoxicity database," published in Computational Toxicology (Volume 34, 2025, 100356).
While we are not authors of this study, we believe its findings are crucial for anyone leveraging the OECD QSAR Toolbox or engaged in regulatory genotoxicity assessment.
Integrating these vital insights into our training:
At ToxNavigation, we are committed to providing the most current and practically relevant training in New Approach Methodologies (NAMs). The results and recommendations from this study are directly applicable to best practices in regulatory genotoxicity assessment using in silico tools.
Therefore, we are thrilled to announce that the key findings and practical implications of this validation study will be fully integrated into our new eLearning training course: "NAMs - application of QSAR and read-across with regulatory genotoxicity add-on".
This course is designed to equip you with the knowledge and skills to effectively apply QSAR and read-across for regulatory purposes, with a specific focus on robust genotoxicity assessment according to current guidelines.
Learn more and enroll in our comprehensive training program here: https://courses.toxnavigation.com/course/ichm7
The paper meticulously explores the accuracy of using the Toolbox's in silico profilers as a lower tier in genotoxicity assessment, providing valuable data to inform regulatory concerns.
The researchers investigated relevant profilers within the OECD QSAR Toolbox, validating them against an extensive external dataset derived from the MultiCASE Genotoxicity database, which includes both AMES mutagenicity and in vivo micronucleus (MNT) experimental results (even expanded with crucial pesticide data from regulatory documents). A particularly important aspect of their work was the incorporation of metabolism simulations by the OECD QSAR Toolbox and assessing their influence on profiler performance.
Key takeaways from this significant study include:
Reliability of Negative Predictions: The study found a strong correlation between the absence of profiler alerts and experimentally negative genotoxicity outcomes, reinforcing their utility for initial screening.
Varying Accuracy: While overall performance is good, the calculated accuracy for MNT-related and AMES-related profilers varies, emphasizing the need for nuanced interpretation (41%-78% for MNT-related and 62%-88% for AMES-related profilers).
Impact of Metabolism: Incorporating metabolism simulations significantly boosts accuracy, increasing it by 4–6% for the AMES dataset and 4–16% for the MNT dataset, underscoring the importance of considering metabolic activation.
Need for Expert Review: The paper strongly advises that while profilers are excellent for flagging potential concerns, any triggered alerts require critical evaluation and should be complemented by results from third-party QSAR models and expert review, rather than being used directly for definitive predictions.
We encourage all toxicologists, regulatory scientists, and chemists working with genotoxicity assessments to read this paper to enhance your understanding and improve your application of the OECD QSAR Toolbox.
#Genotoxicity #QSAR #OECDQSARToolbox #NAMs #ComputationalToxicology #RegulatoryToxicology #Pesticides #InSilico #Training #ChemicalSafety #ToxNavigation
You can access the full paper here
Comments