Small molecules accounted for approximately 65% of FDA new drug approvals in 2025, continuing a resurgence that has challenged predictions of their gradual displacement by biologics and other therapeutic modalities. Their enduring advantages are familiar: oral bioavailability, scalable manufacturing, cellular permeability, and broad therapeutic flexibility. What has changed is the nature of the molecules themselves.

The current generation of small molecule therapeutics are increasingly sophisticated in terms of their design, mechanism of action, and ability to interact with the process that drive, or may correct, disease states. This diversity in small molecule therapeutics now includes targeted protein degraders/induced-proximity therapeutics, covalent agents, and next-generation kinase inhibitors. These modalities, by virtue of their complex mechanism of action, require new approaches to the workflows that shaped classical medicinal chemistry workflows over the past several decades. We now need assays for such complex processes as ternary complex formation, selectivity of protein degradation, models for the relationship between kinetics of binding reversibility and therapeutic outcome, and many more. Furthermore molecular weight, polarity, conformational flexibility, complexity of synthetic routes, and the need for new analytical and bioanalytical methods all require new strategies to measure and optimize these parameters.

Small molecules clearly remain relevant. The harder question is whether drug discovery infrastructure, organizational models, and translational strategies can evolve quickly enough to support this growing molecular complexity.

Chemically induced proximity and the transition to event-driven pharmacology

One of the most consequential conceptual shifts in modern drug discovery is the emergence of strategies to use small molecules to create, or stabilize, interactions between two biomolecules. The first, and perhaps most mature, example of this field is the application of the concept to initiate targeted protein degradation (TPD). Traditional small molecules generally rely on sustained occupancy of a binding site to inhibit protein function. Degraders instead exploit the cell’s endogenous protein disposal machinery, converting transient binding events into catalytic elimination of the target protein itself.

This distinction has expanded the accessible target landscape. Proteins lacking well-defined inhibitor or activator binding sites, structurally dynamic proteins, and scaffolding proteins previously considered poorly tractable are now entering discovery pipelines.

The final validation of the field arrived in 2026, when Arvinas and Pfizer received FDA approval for vepdegestrant, the first proteolysis targeting chimera (PROTAC) therapeutic approved for patients with advanced or metastatic breast cancer. The approval represents more than a milestone for a single modality. It establishes induced proximity as a clinically viable pharmacological principle.

Importantly, PROTACs are only one branch of a broader induced-proximity landscape. Proteasome-directed degradation remains largely restricted to intracellular proteins, prompting increasing interest in lysosomal pathways capable of targeting extracellular and membrane-associated proteins. Emerging approaches such as LYTACs, AUTACs, ATTECs, and KineTACs reflect an expanding effort to manipulate cellular trafficking and degradation pathways beyond the ubiquitin–proteasome system.

Despite their promise, degraders continue to expose the limits of conventional small molecule optimization paradigms. Oral bioavailability remains challenging because many degraders occupy physicochemical space beyond traditional drug-likeness heuristics. Linker composition, conformational dynamics, and ternary complex formation introduce additional layers of design complexity. Resistance mechanisms, including mutations in the E3 ligases that ‘tag’ target proteins for degradation, such as CRBN or VHL, further illustrate that induced-proximity pharmacology introduces new evolutionary pressures distinct from those associated with occupancy-driven inhibitors.

These challenges are reshaping medicinal chemistry priorities. Efforts increasingly focus on identifying alternative E3 ligases (sometimes with selective expression profiles), reducing molecular size, optimizing linker topology, and developing prodrug strategies capable of improving permeability, oral exposure and even CNS permeability. More broadly, degraders are forcing the field to reconsider long-standing assumptions regarding acceptable physicochemical boundaries for orally active therapeutics.

Covalent chemistry moves from exception to rational strategy

Covalent drugs have existed for more than a century, yet for much of modern medicinal chemistry they were either discovered serendipitously, or approached cautiously because of concerns surrounding selectivity and off-target reactivity. That perception has changed substantially over the past decade.

By the end of 2024, more than 50 covalent drugs had been approved globally, including a growing number designed intentionally around covalent mechanisms rather than discovered retrospectively. Ibrutinib (BTK) and afatinib (EGFR) are widely cited as the first rationally designed covalent inhibitors, validating the strategy of introducing covalent warheads into reversible scaffolds to enhance potency and selectivity. The field’s maturation reflects improvements in structural biology, computational chemistry, and kinetic characterization that now allow covalent engagement to be engineered with far greater precision.

The development of KRAS G12C inhibitors marked a defining moment. Sotorasib and adagrasib, the two approved KRAS G12C inhibitors that broke the field’s decades-long stalemate on this target, are now widely referenced as historical inflection points for covalent design. KRAS had long been considered undruggable because of its smooth surface topology and limited pocket accessibility. Electrophile-guided discovery strategies transformed that assumption by exploiting a mutant cysteine residue unique to the oncogenic allele. Rather than optimizing reversible affinity alone, medicinal chemists designed molecules capable of selectively forming covalent bonds with the target itself.

The implications extend well beyond KRAS. Covalent chemistry is now being applied to residues beyond cysteine, including lysine, tyrosine, serine, and threonine. Softer electrophiles such as acrylamides and nitriles have enabled more tunable reactivity profiles, improving selectivity while reducing concerns regarding indiscriminate protein modification.

At the same time, covalent drugs continue to challenge traditional pharmacokinetic modeling strategies. Exposure metrics alone may poorly predict biological activity because pharmacodynamic duration can substantially outlast plasma residence time of the free drug molecule. As a result, parameters such as target engagement kinetics, residence time, kinact/KI relationships, and time-dependent inhibition profiles are becoming increasingly important during optimization of efficacy, as well as prediction of toxicology profiles.

Covalent chemistry is therefore evolving from a niche strategy into a broader design discipline — one that integrates structural biology, reaction kinetics, and systems pharmacology into an alternative medicinal chemistry workflow and decision-making strategy.

The kinome remains far from exhausted

The landmark approval by the FDA of the 100th kinase inhibitor in 2025 underscored the extraordinary durability of kinase biology as a therapeutic engine. Yet despite more than two decades of clinical success since imatinib, much of the kinome remains largely unexplored as a strategy for therapeutic benefit.
The human cell expresses more than 500 kinases. Approved kinase inhibitors act on roughly 10% of them. The broader small molecule literature reports activity against about 68% of the kinome. The remaining third, often called the dark kinome, are  without any pharmacological tools to investigate function, and consequently in many cases, poorly understood.

Historically, these kinases were deprioritized for understandable reasons. Many lacked strong genetic validation, and selective inhibitors frequently failed to produce sufficiently clear phenotypes in reductionist systems. Increasingly, however, researchers recognize that kinase function is deeply network-dependent. Biological outcomes often emerge not from single-node inhibition, but from perturbations across interconnected signaling systems.

This shift is contributing to renewed interest in rational polypharmacology. Early kinase inhibitors were frequently described as “dirty” because of their multi-target profiles. That characterization increasingly appears over-simplified. Multi-kinase activity may in some contexts be therapeutically advantageous, provided that the relevant signaling liabilities can be understood and controlled.
The next generation of kinase therapeutics may therefore rely less on maximal selectivity and more on deliberate network modulation — designing compounds that engage beneficial signaling nodes while avoiding pathways associated with toxicity or resistance.

Discovery infrastructure is a determinant of scientific success

“The field of small molecule drugs is now moving toward a more deliberate discovery paradigm, where advances in screening, structural biology, and systems-level mechanistic and functional analysis are beginning to make the process increasingly predictable and engineerable.” Said Dr. Dave Madge, VP, Discovery Services from WuXi AppTec.

For much of the past two decades, innovation focused heavily on improving throughput and reducing cycle times. Increasingly, however, the emphasis is shifting toward improving translational success rates. This transition is reflected in the evolution of discovery and development technologies themselves.

• DNA-encoded libraries, fragment-based screening, and high-throughput computational methods are expanding access to difficult target classes. 

• Direct-to-biology platforms compress design–make–test cycles by integrating synthesis and biological evaluation earlier in the process. 

• High-resolution mass spectrometry (LC-HRMS) offers the resolution and precision to confidently identify and quantify drugs and biomarkers.

•  Cell type-specific and spatial selective analysis of the outcomes of target engagement, at the level of RNA and proteins, is becoming more widely available to inform the desired and undesired outcomes of drug binding. 

•  Enabling technologies such as flow chemistry, biocatalysis, and automated reaction optimization are reshaping how complex molecules are synthesized and scaled.

Importantly, these technologies are not operating independently. Modern discovery increasingly depends on the integration of chemical synthesis, structural biology, computational modeling, translational biology, and manufacturing sciences into a coordinated system capable of supporting highly complex molecular architectures.
In that sense, the next challenge for small molecule drug discovery may be less about identifying whether a target is druggable, and more about making an informed choice amongst different target engagement strategies, and creating an appropriate workflow to support that choice. In each case of course the later-stage development challenges may also be part of the selection criteria for any specific small-molecule type, but as each type populates more of the clinical development pipeline these concerns will dissipate. 

“Overall, drug discovery is shifting from a compound-centric exercise toward an outcome-oriented discipline, where success increasingly depends on the integration of scientific and development capabilities across the discovery continuum.” Said Dr. Tao Guo, Senior Vice President, Research Chemistry Services, Integrated Program Management at WuXi AppTec.

The trends that may re-shape the success of small molecule drug discovery in the coming decade

Will induced-proximity therapeutics expand access to new targets?

Targeted protein degraders, molecular glues, and other induced-proximity therapeutics are expanding the range of tractable biology, allowing us to engage with targets and pathways previously not drugged, as well as creating routes to ‘re-wire’ cellular biology to therapeutic benefit. In doing so, they are also challenging long-standing assumptions surrounding the molecular weight, polarity, and permeability of drug-like molecules. Rather than eliminating classical drug-likeness principles, these modalities are pushing the field toward a more nuanced understanding of how conformational dynamics, intramolecular interactions, and exposure profiles influence oral bioavailability.

Is covalent chemistry entering a golden age?

The success of KRAS G12C inhibitors demonstrated that covalent engagement can unlock targets long considered inaccessible to small molecules. The field is now rapidly expanding beyond cysteine targeting into broader amino acid space, including lysine, tyrosine, serine, and threonine residues. Increasingly, covalent chemistry is being approached not as a fallback strategy, but as a deliberate design principle integrating structural biology, kinetics, and systems pharmacology.

How will we explore new therapeutic opportunities across the kinome?

Despite more than two decades of clinical success, much of the human kinome remains poorly characterized biologically and chemically. These “dark kinases” may represent one of the largest remaining frontiers in targeted therapeutics. At the same time, the field is moving beyond the early paradigm of maximal selectivity. Rational polypharmacology — intentionally designing compounds to modulate signaling networks rather than single nodes — is increasingly emerging as a legitimate therapeutic strategy.

Is discovery infrastructure becoming a scientific differentiator?

As small molecule modalities become more complex, the developers need timely and efficient access to enabling technologies. Modern programs often require simultaneous advances in medicinal chemistry, computational design, analytical characterization, process chemistry, translational biology, and scalable manufacturing. The ability to integrate these functions into a coherent development architecture may increasingly determine whether sophisticated molecular concepts can ultimately become viable medicines.

What could define the next era of small molecule drug discovery?

Induced-proximity pharmacology, covalent design, and systems biology are collectively expanding what small molecules can achieve biologically. The next decade will likely test not only the creativity of medicinal chemistry, but also the industry’s ability to integrate increasingly complex science into reproducible clinical success.