“Water does not resist. Water flows…But water always goes where it wants to go, and nothing in the end can stand against it.” – Margaret Atwood
“The path of least resistance leads to crooked rivers and crooked men.” – Henry David Thoreau
What fraction of potential protein targets is accessible to conventional therapeutic modalities such as small molecules and protein biologics? The “druggable genome”, a term coined by Hopkins and Groom in 2002 (here), provides an estimate: approximately 10% of proteins in the human body are druggable by small molecule therapeutics. Greg Verdine and others estimate that an additional 10% of protein targets – those that are extracellular proteins – are druggable by biologics (here; excellent podcast by Janelle Anderson, humanPoC, here). Derek Lowe, however, has blogged that there is a lot of uncertainty in these estimates (here, here).
But just because a protein is druggable does not necessarily make it a potential drug target, “for that honour belongs only to proteins that are also linked to disease”. That is, proteins that are compelling targets based on causal human biology may not be druggable.
These two issues create a natural tension for drug hunters at the start of a drug discovery program: pursue those targets that are druggable or those targets with the most compelling human evidence. Ideally, drug hunters would pick targets with both. However, life is not so convenient, and a drug hunter is often forced to choose.
Which brings me to the title of the blog. Too often, drug hunters pick the “Water” option, the path of least resistance: targets that are druggable. I favor the “Thoreau” option. I favor targets with compelling human biology, even if that is the path of most resistance.
Let me back up a few steps and provide context.
[Disclaimer: I am a Merck/MSD employee. The opinions I am expressing are my own and do not necessarily represent the position of my employer.]
What makes a good drug? In brief: a drug that binds and modulates a molecular target in such a way that is safe and effective in the disease context for which it is administered. Ideally, at the very start of the drug R&D process, it would be helpful to know which targets, when perturbed, alter human physiology in such a way that the drug is safe and effective in humans. This is what I refer to as “causal human biology”.
Once a compelling target has been identified, the next step in a drug discovery program is to develop a therapeutic that modulates the target in a specific manner. There are two major challenges. First, the therapeutic molecule must gain access to the protein target of interest. As described above, it is estimated that only ~20% of human proteins are accessible by either small molecules (which target hydrophobic pockets) or biological therapeutics (which bind to extracellular targets), which leaves the majority of protein targets “undruggable” (see here). Second, once a therapeutic molecule engages its target, it must exert an effect consistent with the underlying therapeutic hypothesis. Both small molecule and biologics are restricted in the mechanism by which they perturb a target (e.g., orthosteric inhibition for small molecules, neutralization of extracellular proteins by biologics). As a consequence, only a small portion of potential drug targets is considered tractable for a new drug discovery program.
Key steps in the “Thoreau” model are to select targets with causal human biology and to develop a drug that recapitulates causal human biology. To enable this step, it is critical to understand directionality of desired therapeutic modulation (e.g., increase or decrease activity of protein target) as well as the mechanism by which therapeutic modulation is expected to act (e.g., enzymatic activity, signaling). For example, if the human immune system leads to autoimmune destruction of a specific cell type that secretes a specific protein ligand (e.g., orexin secreting neurons in patients with narcolepsy), then the therapeutic intervention should inhibit the ligand-receptor interaction (e.g., orexin-receptor antagonists to promote sleep).
In the absence of a disciplined approach, targets may be selected based on the “Water” option (i.e., druggability) rather than the “Thoreau” option (i.e., causal human biology). Indeed, there is a historical bias towards protein classes considered druggable, such as kinases, ion channels, G-protein coupled receptors, and extracellular cytokines and cytokine receptors (here).
To overcome the challenges of the druggable genome, new approaches are being developed to expand mechanisms by which small molecules and biologics exert therapeutic effect (e.g., positive allosteric modulators, conjugated nanobodies). Moreover, phenotypic screens can be used to uncover unexpected mechanisms by which small molecules modulate a target or pathway (e.g., binding to RNA).
Beyond small molecules and biologics, new therapeutic modalities are gaining traction: mRNA delivery (e.g., Moderna Therapeutics), siRNA (e.g., Alnylam Pharmaceuticals) and anti-sense oligonucleotides (e.g., Ionis Pharmaceuticals), and gene editing with CRISPR-cas9 (here), amongst others. These new modalities should expand the ability to recapitulate causal human biology in the form of a therapeutic.
Why do I believe in the “Thoreau” path? Think about the productivity crisis in drug R&D: the number new therapeutic drug approvals per dollar spent has decreased over the last several decades (here). While the number of FDA approvals has seen an uptick since 2012 (here), R&D costs, driven by failure in Phase II-III clinical trials, continue to rise. And most Phase II-III failures are due to lack of efficacy.
In other words, most drugs fail because the choice of target at the beginning of a drug discovery program is wrong. And drug hunters too often make the wrong choice because they chose the “Water” path rather than the “Thoreau” path. The path of least resistance leads to crooked rivers and failed drugs.