What is the most successful preclinical strategy for drug discovery? Finding the answer to this question is important if pharma is to keep its pipeline full. With more drugs coming off patents, it may be time to consider what the numbers have to tell us.
There are two approaches to drug discovery. One is a target-based approach, popular since the 1990s. This develops drugs that target specific proteins related to disease pathogenesis. High-throughput screening facilitates identification of compounds that may interact with whole target families, such as receptor tyrosine kinases. Computers can help, with computational modeling of target-lead interactions.
Before such technology was available, the main approach to drug discovery was phenotypic assays, which address the question, “does it work?” Not, “how does it work?” The phenotypic approach screens candidates for their desired effect, without knowing the molecular mechanism of action (MMOA).
Though the hypothesis-driven target-based approach sounds like it should be more successful, it is not necessarily a sure-fire winner for drug discovery. Put simply, hypotheses can be wrong. Phenotypic screening, on the other hand, presents challenges to drug optimization. Not knowing the MMOA makes it more difficult to chemically tweak leads for better results. Phenotypic screens are also lower-throughput, which inherently limits drug discovery. So which is better?
An analysis of the drug discovery methods that led to FDA approval between 1998 and 2008 found that phenotypic screening should not be cast aside in favor of glitzy new technology. Of the small-molecule first-in-class drugs that were discovered during this ten-year span, 28 came from phenotypic screening, while only 17 came from target-based approaches.
This is not to say that target-based approaches are bad. To develop best-in-class drugs, knowing the MMOA is important. When it came to follower drugs, the target-based approach was more successful, accounting for 51% of approved drugs. Phenotypic screening accounted for 18%. Sometimes, it’s about knowing the right tool for the job.
- Swinney D.C. and Anthony J. How were new medicines discovered? Nature Reviews Drug Discovery 10, 507-519 (July 2011). doi:10.1038/nrd3480