It is estimated that 85% of the human proteome is undruggable.¹ If true, we may never be able to access and modify a multitude of proteins associated with pathology nor develop medicines for a range of complex and challenging diseases such as cancer, autoimmune conditions and neurodegenerative disorders. Fortunately, scientists and investigators are rising to this challenge with technological innovation.

The term undruggable was coined to describe clinically relevant targets – usually proteins – that are ‘impossible’ to target pharmacologically.

Undruggable targets typically have limited interactions that may be exploited pharmacologically. Examples include:^2–5

• Absence of deep or defined ligand-binding pockets to accommodate a drug
• Large, flat or non-catalytic protein–protein interactions, where preventing such interaction is the target
• Binding site occupation resulting in complete loss of protein function, leading to disease
• Unclarified 3D structures, leading to a roadblock in drug development

In some cases, the reasons why a target is undruggable are unknown. In others, the use of high concentration(s) of a drug to outcompete naturally occurring ligands may cause off-target effects, unacceptable toxicity or a poor benefit-risk profile.

Undruggable or yet to be drugged?

Traditional drug design relies on phenotypic screening, establishing a clinical correlation between disease phenotypes and drug treatment. It identifies the binding site of the target and then designs molecules to bind to this pocket and achieve a therapeutic response. Despite increasing efforts to decipher the molecular and pathological pathways involved in the genesis of disease and a tremendous increase in the number of targets discovered, this approach cannot target undruggable proteins.^2–5

If undruggable is a surrogate for a difficult-to-treat disease – beyond the reach of small molecule compounds – then what other strategies can be employed? Recombinant protein-based therapeutics emerged in the 80s, and their development and use continues today.⁶ Equally, ribonucleic acid (RNA) therapeutics are gathering momentum, finding application in conditions as diverse as hypercholesterolaemia, hereditary transthyretin amyloidosis, and spinal muscular atrophy.⁷

However, even RNA therapeutics may still suffer from low tissue-specific delivery (except to the liver and kidney, where they are typically targeted), inefficient cellular uptake, poor endosomal escape and potential platform-specific toxicities.⁸

Induced protein proximity combines two factors to act like a molecular matchmaker, recruiting or hijacking endogenous factors to trigger therapeutic biological processes such as protein degradation, stabilisation or folding.⁹ Several induced protein proximity technologies have emerged that may hold the key to next-level therapeutics, key examples include:^10–12

• Proteolysis-targeting chimaeras (PROTACs), bivalent molecules that degrade a target protein in close proximity by cellular proteasomal machinery
• Ribonuclease-targeting chimaeras (RIBOTACs), bivalent molecules containing an RNA-binding module and a ribonuclease recruitment module that have the potential to degrade diverse types of RNA
• Molecular glues, small chemicals linked to natural digestive enzymes (for example, the E3 ubiquitin ligases of the ubiquitin-proteasome system) that stick to and destroy target pathogenic proteins
• Multispecific or bifunctional small molecules

Multispecific or bifunctional drugs don’t follow the principle of predecessors where one drug hits one target¹¹ – they are more ambitious than that. Multispecific drugs form two or more drug–target binding interfaces sequentially or concurrently to express their therapeutic effect. They are consequently more complex in both chemical structure and mechanism of action, conferring advantages over classical drugs, such as:

• High specificity (e.g. by concentrating the agent only where needed at the site of action), delivering the potential for increased therapeutic effect with reduced off-target toxicity to cells and tissues
• Achieving therapeutic efficacy independent of binding affinity (as they are ‘event-driven’ rather than ‘occupancy-driven’, like classical drugs)

What it could mean for patients

Targeting the undruggable is a drug discovery frontier, with new technologies expanding the scope of therapeutically reachable disease-causing proteins. Novel drug discovery approaches look promising for challenging and complex diseases.

References

1. Neklesa TK, et al. Pharmacol Ther. 2017;174:138–144.
2. Serapian SA, et al. ChemMedChem. 2021;16(10):1593–1599.
3. Zhang G, et al. Expert Opin Drug Discov. 2022;17(1):55–69.
4. He H, et al. Stroke Vasc Neurol. 2020;5(4):381–387.
5. The Cambridge Crystallographic Data Centre. www.ccdc.cam.ac.uk/Community/blog/undruggable-targets-strategy-drug-discovery. Accessed 13 Sep 2022.
6. Brooks SM, Alper HS. Nat Commun. 2021;12(1):1390.
7. Crooke S, et al. Cell Metab. 2018;27(4):714–739.
8. Yin W, Rogge M. Clin Transl Sci. 2019;12(2):98–112.
9. Mullard A. Nat Rev Drug Discov. 2022;21(3):172–173.
10. Dey SK, Jaffrey SR. Cell Chem Biol. 2019;26(8):1047–1049.
11. Wang Y, Yang S. Signal Transduct Target Ther. 2020;5(1):86.
12. Sasso, J. CAS. https://www.cas.org/resources/blog/molecular-glues. Accessed 13 Sep 2022.