After earning her PhD in pharmacology, Barbara Slusher developed small molecule therapeutics in the pharmaceutical industry for 18 years. Then, she brought her expertise to academia, leading the Johns Hopkins Drug Discovery program. In this interview with The Scientist, Slusher discussed her approach to drug discovery, noting challenges and emerging technologies in the field.
Barbara Slusher is an expert in drug discovery, leading multidisciplinary teams to identify small molecule therapeutics for development.
Johns Hopkins University
How do you begin the discovery phase of drug development?
The first stages are target identification and validation. This involves selecting a target, such as a protein, receptor, or enzyme, and demonstrating that its modulation is likely to have a meaningful benefit in a patient. You want to make sure that the target is valid, there is a path to success, and someone will care. Common techniques include large-scale omics experiments and knocking out targets to see the effect. We also look at human genetics and disease-associated pathways.
Once you have a target, what is the next step?
The next stage is hit identification to find a molecule that interacts with the target. Most big pharmaceutical companies have a large compound library that they developed over their history, and high-throughput screening of this internal library results in decent, proprietary, drug-like compounds. In academia, you can purchase libraries of nonproprietary compounds from vendors. Because everyone has access to these libraries, our center established a high-throughput screening agreement with industry. We identify targets and develop assays which they use to screen their library—it is a real win-win.
In addition to high-throughput screening, alternative techniques include fragment-based screening, virtual screening, and phenotypic screening. Fragment-based screening is when you start with really little molecules, and then you build the compound around that fragment. If you know the structure of your target, you can virtually screen drugs that might bind the target to bring you initial leads. Phenotypic screening is when you do not know what the target is, but you screen a library of compounds to see if they improve a phenotype. Once you get a promising compound, you have no idea how it works, so then you have to deconvolute what it might be doing.
How do you optimize a hit to become a clinical candidate?
The hit-to-lead optimization phase is really the meat of drug discovery. This is the integrated process where the medicinal chemistry, toxicology, and pharmacokinetic groups work together to try to refine the hits to become more potent, selective, and drug-like. If you think about it like a circle, medicinal chemistry makes a drug. Then, it goes to the assay group that says whether the drug is potent. If it is not potent, we abandon it. If it is potent, then the drug goes into a metabolic screening cascade, which is usually assessing the metabolism in plasma or liver microsomes across species. That information goes back to the chemist to improve the molecule, and then we do that cycle again until we have a compound that is potent and metabolically stable. Then, we test its efficacy in vivo. Hit-to-lead optimization is an iterative process of making, characterizing, and optimizing compounds, and then the last stage is the early assessment of developability, including stability, efficacy, and toxicity.
What are the biggest challenges in drug discovery right now?
I think the biggest challenges are our incomplete understanding of human biology and that there can be poor translation from preclinical to human models. Another difficulty is that there are a lot of redundancies or compensatory pathways in biology.
Thinking about the operations of drug discovery, we can come up with potent compounds pretty easily but getting ones that have the appropriate pharmacokinetic and stability profile is tough. The ability to optimize all those parameters at the same time is one of the most challenging parts of drug discovery.
Barbara Slusher uses her experience in the pharmaceutical industry to inform drug discovery research at Johns Hopkins University.
Johns Hopkins University
What new technologies have emerged that have changed drug discovery?
We still have improvements to make, but artificial intelligence for target identification and compound optimization has been impactful in prioritizing candidates. Also, human genetics and functional genomics have enhanced target identification. From a technological perspective, structural biology, formulation, and drug delivery have also improved. People use human disease models, such as induced pluripotent stem cells and organoids, that are better reflective of the human condition and make drug discovery more successful. There are new therapeutic modalities as well, which allow us to have targets that we would not have been able to investigate previously. These new technologies shift us from serendipity to more data-driven decision-making that will hopefully result in faster cycle times and better outcomes.
The interview has been condensed and edited for clarity.
