Single Blog

Why KRAS Was Long Considered “Undruggable”

For decades, KRAS was the poster child of “undruggable” oncogenes. Mutated KRAS drives up to 30% of all cancers, including subsets of non–small cell lung cancer (NSCLC), colorectal cancer, and pancreatic cancer. Yet its smooth, compact surface and picomolar affinity for GTP/GDP made traditional small molecule binding seem impossible. Medicinal chemists repeatedly failed to find a pocket where a drug could meaningfully bind and switch off signaling.

This changed when structural biology and covalent fragment screening converged, revealing a cryptic allosteric pocket in a specific KRAS mutation: KRAS^G12C. That single insight triggered a wave of next‑generation small molecule inhibitors now reshaping precision oncology.

From Impossible to Drugged: The KRAS^G12C Breakthrough

Exploiting a Hidden Allosteric Pocket

KRAS^G12C replaces glycine with cysteine at position 12, introducing a nucleophilic handle. Structure‑based design uncovered an allosteric “switch II” pocket adjacent to this cysteine. Covalent fragments were engineered to form an irreversible bond with Cys12, locking KRAS in its inactive, GDP‑bound state and preventing downstream signaling through RAF–MEK–ERK and PI3K pathways (doi:10.1038/nature22307).

This strategy produced the first clinically successful KRAS inhibitors, including sotorasib and adagrasib, both FDA‑approved for KRAS^G12C NSCLC. They validated a core idea: even “undruggable” proteins may harbor transient or mutation‑created pockets accessible to small molecules.

Beyond G12C: Targeting Other KRAS Mutations

Noncovalent Inhibitors and State‑Selective Binding

The next frontier is KRAS mutants without a cysteine at position 12, such as G12D, G12V, and G13D, which dominate in pancreatic and colorectal cancers. Without a reactive cysteine, covalent chemistry is off the table. Researchers are now engineering noncovalent small molecules that bind selectively to inactive or active conformations of KRAS, stabilizing druggable pockets that appear only transiently (doi:10.1016/j.ccell.2020.11.006).

These state‑selective inhibitors often exploit subtle conformational differences between wild‑type and mutant KRAS, enabling mutant‑selective targeting while sparing normal cells. Early clinical candidates against KRAS^G12D and other alleles are already in trials, signaling a shift from “one mutation, one drug” to a broader KRAS mutation toolkit.

Overcoming Resistance: Combination Strategies and Next‑Gen Designs

Adaptive Signaling and On‑Target Resistance

As with other targeted therapies, resistance to first‑generation KRAS inhibitors is emerging. Tumors adapt via secondary KRAS mutations, bypass pathway activation (e.g., EGFR or MET upregulation), or phenotypic transformation. Medicinal chemists are responding with:

• Next‑generation KRAS binders that retain activity against resistance mutations in the switch II pocket.
• Upstream/downstream combinations, such as pairing KRAS inhibitors with SHP2, MEK, or EGFR inhibitors to blunt pathway reactivation (doi:10.1038/s41568-021-00362-1).
• Rational triplet regimens that integrate chemotherapy or immunotherapy to deepen and prolong responses.

AI‑Driven Design Meets Structure‑Guided Chemistry

While the first KRAS inhibitors emerged from classical fragment‑based and structure‑guided campaigns, the current wave increasingly leverages AI. Generative models help explore chemical space around the switch II pocket, optimize covalent warheads, and predict off‑target liabilities. Machine learning‑guided docking and free energy calculations accelerate lead optimization, reducing the time from hit to clinical candidate.

The synergy of high‑resolution cryo‑EM/X‑ray structures, covalent chemistry, and AI‑driven design is expanding what counts as a “druggable” surface on KRAS and other small GTPases.

What This Means for the Future of Small Molecule Oncology

The success of KRAS inhibitors has rewritten the rules of druggability. It shows that:

• Mutation‑specific pockets can turn “smooth” proteins into small molecule targets.
• Covalent and noncovalent designs can be tailored to individual KRAS alleles.
• Resistance is not a dead end but a roadmap for smarter, combination‑ready molecules.

As next‑generation KRAS inhibitors reach the clinic, they are likely to move from niche salvage options to backbone therapies in molecularly defined lung, colorectal, and pancreatic cancers. The once “undruggable” KRAS is rapidly becoming a model for how to crack other recalcitrant oncogenes with innovative small molecule chemistry.

Key References

• Ostrem JM et al. Inhibition of KRAS^G12C by a small molecule that targets a switch‑II pocket. Nature. 2013;503:548–551. doi:10.1038/nature12796
• Canon J et al. The clinical KRAS^G12C inhibitor AMG 510 drives anti‑tumour immunity. Nature. 2019;575:217–223. doi:10.1038/s41586-019-1694-1
• Ryan MB et al. Targeting RAS‑mutant cancers: is ERK the key? Cancer Cell. 2020;38(5):604–624. doi:10.1016/j.ccell.2020.11.006
• Moore AR et al. RAS‑targeted therapies: is the undruggable drugged? Nat Rev Cancer. 2020;20:671–688. doi:10.1038/s41568-020-00131-0