New Toxicity Screening Tool Could Save Pharma Millions

The tool makes screening novel drugs for potential toxicities much easier.

A newly-created system that determines the risk of chemical products could potentially save pharmaceutical manufacturers millions on research and development costs.

Current methods used by regulators to screen chemical products for toxicities are only accurate approximately 50% of the time. This inaccuracy can become very costly to companies, who must re-evaluate their product due to suspected problems.

Scientists at the University of North Carolina at Chapel Hill have created a new approach that could improve the current manufacturing process, and increase accuracy up to 85%. The new tool could potentially save time and millions of dollars for manufacturers, and improve safety.

Agencies like the FDA and the EPA are required to evaluate the safety of new drugs and chemical products before they are marketed. The initial screening is heavily reliant on early analysis of the product’s molecular structure.

Groups of atoms associated with chemical toxicity trigger a structural alert, which then requires the company to conduct additional testing on the product, according to the study published by Green Chemistry.

Since these structural alerts are accurate only half of the time, there is a clear need to develop a more accurate system. The scientists created a novel computational approach that uses statistical analysis to establish whether the alert is accurate or if it should be dismissed.

“A lot of chemicals are incorrectly identified as potentially toxic even though in the end they are not toxic and that could have been predicted,” said lead scientist Alex Tropsha, PhD. “Companies are forced to run a lot of unnecessary and costly experiments, and because companies run these checks themselves before submitting their products to regulators, there are products that never see the light of day because they are flagged as toxic when they are not.”

However, drugs that are flagged do not always get rejected by the FDA in the final approval process. Lipitor, the commonly used drug that treats high cholesterol, was originally flagged because it has 5 elements in its molecular structure that are structural alerts, but the drug itself is non-toxic, according to the study.

The scientists used quantitative structure-activity relationship modeling over the current alert system to create a more accurate approach. The new model is able to view the structure of the entire chemical molecule, and then assigns a numerical value relative to the accuracy of the alert.

“Structural alerts are a convenient system, but there are few consequences for being wrong even though the stakes are potentially very high,” Dr Tropsha said. “If the alert is right, then it’s ‘we told you so.’ If it’s wrong, ‘well, it was just a warning anyway.’ But unfounded alerts unnecessarily add years and millions of dollars to the cost of bringing a new drug or product to market without improving safety. That is unacceptable, we think.”

The scientists are planning to make their new system available to regulatory agencies and scientists through a web-based computer software, according to the study.

“We want to alarm regulators that structural alerts over-predict toxicity while missing truly toxic substances, and offer them much more accurate tools to support regulatory decisions,” Dr Tropsha concluded.