Virtualizing?chemical safety assessment: QSAR modelling to guide reduction of animal testing.

Author(s): Melchor Sanchez Martinez

Editor(s): Maryam Salehi, Andrew Whitehouse & Natalja Kurbatova

This article will discuss good old QSAR models and their role in modern drug discovery.

Summary

QSAR (Quantitative Structure-Activity Relationship) modelling is a computational technique used in drug discovery and chemical assessment. It reduces costs, time, and animal testing while improving sustainability. Regulatory agencies support its use, provided it is aligned with current best practices. Zifo has expertise in building high-quality QSAR models. Contact us at?[email protected] to explore how QSAR can benefit your needs.

What QSAR models are designed to do

QSAR is a computational technique widely used in drug discovery, environmental sciences and many other fields involving?the discovery and usage of chemical compounds.?It is?a?relatively old technique, coming from the 60s, but still valid as exemplified by the hundreds of related papers indexed in PubMed just this year.

QSAR refers to mathematical models based on statistical and AI /machine learning algorithms that learn patterns from a set of molecules with known activities and properties and then apply those patterns to make predictions for new molecules. These models can be used for different purposes.?One important application is chemical safety assessment, the prediction of the toxicity of chemicals. Getting this data prior to production can facilitate sustainable product development and green chemistry.

QSAR modelling can reduce the time and cost of drug and molecule discovery and production processes, for example,?by reducing the number of experiments required.?This reduction saves resources and has an important ethical dimension, especially regarding animal testing.?While animal testing is necessary to assess the safety and efficacy of new drugs and chemicals, there are growing concerns about this practice's ethical and practical implications.?Avoiding unnecessary animal testing has become a societal debate and a goal of regulatory agencies.

Regulations

Regulatory agencies such as the FDA (Food and Drug Administration) and the EMA (European Medicines Agency) have actively advocated for the?reduction?of animal testing.?They support developing and using alternative methods, such as QSAR models.?The FDA has established the Predictive Toxicology Roadmap (FDA's Predictive Toxicology Roadmap | FDA) to promote developing and using QSAR models and other alternative methods in drug development.?Similarly, the?Chemical Safety for the 21st Century Act of 2016 and the?FDA Modernization Act 2.0 of 2022 encourage and support the use of alternative methods to reduce animal testing.?The EU's REACH regulation (Registration, Evaluation, Authorisation and Restriction of Chemicals) also supports using these models in the regulatory assessment of new chemicals.

However, QSAR models cannot be applied indiscriminately. There are rules to follow to apply these models and for their results to be valid (how to use alternatives to animal testing, how to use and report QSAR models?). For example, valid models need a defined applicability domain or appropriate performance measures. The Zifo team has extensive experience in building QSAR models that meet the necessary quality standards, so you do not have to worry. We can help!

?Animal testing reduction

Can QSAR modelling and other in silico techniques help to reduce animal testing? The logical answer is yes. The technology is there, many companies already have it in their pipelines, and regulatory agencies are actively promoting its use. We are currently in the perfect scenario for this to happen. However, time is needed to answer the question adequately.?In the meantime, if you are interested in further discussing how QSAR modelling?can help in the regulatory assessment of new chemicals, don't hesitate to contact us at?[email protected].