Content area
Full Text
In silico ADME predictive models promise faster cycle times, lower costs, and early indication of drug failure
The first part of this three-part series on ADME/Tox (absorption, distribution, metabolism, elimination, toxicology), which appeared in the April issue of Drug Discovery & Development, discussed the importance and reliance on in vitro assays for ADME/Tox testing. It also featured some of the technological improvements that were enabling these in vitro assays to be performed in an automated, high-throughput fashion. This, the second of a three-part article series, highlights the importance of in silica predictive tools for ADME/Tox testing, which are now gaining popularity. In silico testing is typically a high-throughput process, and has relatively low associated costs because it is neither labor- nor material-intensive. These tools are used to predict physicochemical and pharmacokinetic (PK) properties, which in turn helps prioritize compounds for screening and enables early decision making in drug discovery.
Making early decisions on whether to advance a compound through the drug pipeline largely depends on knowing how the compound is going to behave once it enters the clinic. As shown in the graphic on this page, the more a test system resembles the physiological environment in humans, the more accurate and relevant the information. Hence, many companies rely heavily on in vivo data and perform animal testing while the compound is still in discovery.
"For each lead molecule or series, there is some early animal testing done along with the in vitro assays for metabolic stability, bioavailability, and cell transport," says Gerald Fisher, PhD, senior vice president, drug safety and metabolism at Wyeth Pharmaceuticals, Collegeville, Pa. But the animal testing done at this stage is not very detailed, and 80% of the time it's in a rodent, he says. While most in vitro assays mimic only two major cellular pathways-the oxidative and the conjugation pathways-there are several other pathways that are involved in vivo in drug metabolism, says Fisher. "It's the artificial nature of the biochemical assay that doesn't reflect the complexity of the human environment," he says.
HOW ACCURATE IS THE PREDICTION?
Given that even in vitro testing done in a biochemical or cellular environment has its limitations when compared to animal testing, how then is an in silica model expected to...