KRAS mutations drive roughly a quarter of all human cancers, and for nearly four decades the protein was considered impossible to target with drugs. That changed in 2021 when sotorasib became the first approved KRAS inhibitor, followed by adagrasib. Both lock onto a specific mutant form of KRAS called G12C, and they have reshaped treatment for a subset of lung cancers. But G12C is only one of many KRAS mutations, and even patients who respond to these drugs typically develop resistance within months. The field is now racing to expand the playbook with combination regimens, inhibitors that hit other KRAS mutations, drugs that attack KRAS in its active state, and strategies that destroy the protein entirely.
Why KRAS Was Called Undruggable
KRAS is a small signaling protein that acts as a molecular switch. When it is turned on, it tells cells to grow and divide. Mutations that lock it in the “on” position are among the most common genetic drivers of cancer, appearing frequently in lung, colorectal, and pancreatic tumors. Despite knowing this for decades, researchers could not find a way to block the protein with a drug. The surface of KRAS is smooth and lacks the deep pockets that most drugs latch onto, and the mutant versions are nearly identical to the normal protein the body needs, making it hard to hit one without disabling the other.1Annual Review of Cancer Biology. Targeting KRAS Directly
How the First KRAS Inhibitors Broke Through
The breakthrough came from an unexpected angle. The G12C mutation swaps one amino acid for a cysteine, a residue with unusual chemistry that can form a permanent bond with the right kind of molecule. Researchers discovered a hidden groove on the protein’s surface, called the Switch II pocket, that opens only when KRAS is in its inactive, GDP-bound state. By designing small molecules that slip into this pocket and form a covalent bond with the mutant cysteine, they could trap KRAS G12C in its “off” position.2PubMed Central. Discovery of Tetrahydropyridopyrimidines as Irreversible Covalent Inhibitors of KRAS-G12C with In Vivo Activity Sotorasib and adagrasib, the two drugs that emerged from this work, both exploit the same core strategy: they wait for KRAS G12C to cycle back to its inactive form and then lock it there permanently.3PubMed. Beyond KRAS(G12C): Biochemical and Computational Characterization of Sotorasib and Adagrasib Binding Specificity and the Critical Role of H95 and Y96
The Lung Cancer and Colorectal Cancer Gap
Early clinical trials showed that G12C inhibitors could shrink tumors in lung cancer, colorectal cancer, and other solid tumors carrying the mutation.4PubMed. Targeting KRAS in Colorectal Cancer But the responses were not equal across cancer types. In non-small cell lung cancer, the drugs produced meaningful and sometimes lasting responses. In colorectal cancer, the results on their own were disappointing. The reason comes down to biology: colorectal tumors have built-in backup signaling routes, particularly through receptors on the cell surface like EGFR, that can reactivate growth even when KRAS is blocked. Because of these alternative resistance pathways, using a G12C inhibitor alone in colorectal cancer is not considered an effective strategy.5PubMed. Activity and resistance to KRAS(G12C) inhibitors in non-small cell lung cancer and colorectal cancer
Combination Strategies in Colorectal Cancer
The recognition that colorectal tumors rewire around KRAS blockade led to pairing G12C inhibitors with drugs that shut down those escape routes. The most extensively tested combinations add an anti-EGFR antibody to the mix, blocking the receptor that colorectal cancer cells lean on when KRAS is silenced. A trial combining sotorasib with panitumumab in patients with refractory colorectal cancer showed that the combination outperformed KRAS inhibition alone.6PubMed. Sotorasib plus Panitumumab in Refractory Colorectal Cancer with Mutated KRAS G12C Similarly, adagrasib combined with cetuximab produced a response rate of about 46% in evaluable patients, with a median time before the disease progressed of roughly seven months.7PubMed Central. Adagrasib with or without Cetuximab in Colorectal Cancer with Mutated KRAS G12C A newer G12C inhibitor called divarasib, paired with cetuximab, has also shown promising activity with a manageable safety profile.8Nature Medicine. Divarasib plus cetuximab in KRAS G12C-positive colorectal cancer: a phase 1b trial These combination results have shifted the treatment paradigm: in colorectal cancer, KRAS inhibitors are now being developed as partners rather than solo agents.
Reaching Beyond G12C
G12C accounts for only a fraction of all KRAS mutations. The G12D mutation, for instance, is especially common in pancreatic cancer, where KRAS mutations appear in the vast majority of tumors. Because G12D does not introduce a reactive cysteine, the covalent bonding trick that works for G12C does not apply. Researchers had to find a completely different way in. MRTX1133, the first potent and selective inhibitor of KRAS G12D, was identified through extensive structure-guided drug design. It is a noncovalent inhibitor, meaning it binds tightly to the protein through shape and charge complementarity rather than forming a permanent chemical bond.9Journal of Medicinal Chemistry. Identification of MRTX1133, a Noncovalent, Potent, and Selective KRASG12D Inhibitor Preclinical work showed it could shrink tumors in animal models carrying the G12D mutation.10PubMed Central. A Small Molecule with Big Impact: MRTX1133 Targets the KRASG12D Mutation in Pancreatic Cancer This matters enormously for pancreatic cancer patients, who until recently had no KRAS-targeted options at all.
The ambition goes even broader. A dual-state pan-KRAS inhibitor called MCB-294 can bind KRAS in both its active and inactive forms, potentially covering many different KRAS mutations with a single drug. Researchers have also built on MCB-294 to create MCB-36, a molecule designed not just to block KRAS but to tag it for destruction by the cell’s own protein-disposal machinery.11Cancer Cell. Dual-state pan-KRAS inhibition and targeted degradation suppress tumorigenesis and antitumor immunity
Active-State Inhibitors and the Tri-Complex Approach
The first wave of KRAS drugs can only grab the protein when it is in its “off” (GDP-bound) state. But in many cancers, mutant KRAS spends most of its time switched on and bound to GTP. To address this, a fundamentally different class of drugs has emerged. Daraxonrasib (RMC-6236) is a tri-complex inhibitor: instead of binding KRAS alone, it recruits a natural chaperone protein called cyclophilin A, creating a three-way complex that jams KRAS in a configuration where it can no longer send growth signals. This approach works on the active, GTP-bound form of KRAS and can hit multiple KRAS mutant variants as well as wild-type KRAS.12PubMed. Discovery of Daraxonrasib (RMC-6236), a Potent and Orally Bioavailable RAS(ON) Multi-selective, Noncovalent Tri-complex Inhibitor for the Treatment of Patients with Multiple RAS-Addicted Cancers Early clinical data have generated excitement, particularly in pancreatic cancer. However, resistance mutations that disrupt the interaction between the inhibitor and cyclophilin A have already been identified in laboratory studies, with certain mutations at the Y64 position on KRAS reducing the drug’s potency by roughly 20-fold.13Cell. Mechanisms of Resistance to RAS(ON) Tri-Complex Inhibitors
Attacking KRAS Indirectly
Rather than targeting the KRAS protein itself, another strategy aims at the proteins KRAS depends on to get activated. SOS1 is a key helper protein that loads KRAS with GTP and switches it on. Small-molecule inhibitors that wedge into the KRAS-SOS1 interface can prevent this loading step, keeping KRAS stuck in its inactive state.14PubMed Central. Discovery of potent SOS1 inhibitors that block RAS activation via disruption of the RAS-SOS1 interaction MRTX0902, one such SOS1 inhibitor, has shown the ability to block KRAS activation and slow proliferation in cancer cell lines carrying KRAS pathway alterations.15Molecular Cancer Therapeutics. The SOS1 Inhibitor MRTX0902 Blocks KRAS Activation and Demonstrates Antitumor Activity in Cancers Dependent on KRAS Nucleotide Loading Earlier work using stabilized peptides that mimic a portion of the SOS1 protein confirmed the concept, demonstrating that blocking the KRAS-SOS1 interaction could inhibit both normal and mutant KRAS.16PubMed Central. Direct inhibition of oncogenic KRAS by hydrocarbon-stapled SOS1 helices
SHP2 is another upstream node being exploited. This enzyme sits at the junction where growth factor receptor signals feed into KRAS activation. Blocking SHP2 increases the fraction of KRAS G12C stuck in its drug-targetable, GDP-bound state, which amplifies the effectiveness of G12C inhibitors. In preclinical models of both lung and pancreatic cancer, combining an SHP2 inhibitor with a G12C inhibitor overcame adaptive resistance that the G12C inhibitor alone could not.17PubMed Central. SHP2 inhibition diminishes KRASG12C cycling and promotes tumor microenvironment remodeling
How Tumors Fight Back
Resistance to KRAS inhibitors is the central challenge. It comes in two flavors. The first is on-target: the tumor acquires new mutations in KRAS itself that prevent the drug from binding. A comprehensive study of patients whose cancers progressed on G12C inhibitors identified a wide array of secondary KRAS mutations, including changes at positions G12, G13, Q61, R68, H95, and Y96, along with amplification of the KRAS G12C gene itself.18PubMed Central. Acquired Resistance to KRAS(G12C) Inhibition in Cancer Laboratory experiments confirmed this breadth, generating over 140 resistant cell clones and finding that about 87% harbored secondary KRAS mutations. Mutations at Y96 were particularly troublesome because they conferred resistance to both sotorasib and adagrasib.19PubMed. KRAS Secondary Mutations That Confer Acquired Resistance to KRAS G12C Inhibitors, Sotorasib and Adagrasib, and Overcoming Strategies: Insights From In Vitro Experiments
The second flavor is off-target: the tumor reactivates growth signaling through pathways that bypass KRAS entirely. In the majority of KRAS G12C models studied, rapid feedback reactivation of the RAS pathway was observed after treatment, driven by receptor tyrosine kinases activating wild-type RAS proteins that the G12C-specific drug cannot touch.20Clinical Cancer Research. Vertical Pathway Inhibition Overcomes Adaptive Feedback Resistance to KRASG12C Inhibition This is a major reason why combination strategies, whether adding SOS1 inhibitors, SHP2 inhibitors, or EGFR antibodies, are being pursued so aggressively: shutting down the escape routes is as important as hitting KRAS itself.
Catching Resistance Early With Liquid Biopsy
One of the more practical advances is using blood-based tests, or liquid biopsies, to detect resistance mutations before they show up on imaging scans. In one documented case, a patient on adagrasib developed three new KRAS resistance mutations (G12D, G13D, and Q61H) that were detectable in circulating tumor DNA while CT and MRI scans still showed stable disease. Clinical deterioration did not occur until roughly eight weeks after the blood test first flagged the emerging resistance.21PubMed Central. ctDNA Detection of Polyclonal KRAS Resistance in Adagrasib-Treated NSCLC Similar discordances have been reported between traditional tissue biopsies and liquid biopsies, with the blood-based approach capturing resistance clones that a single tissue sample might miss due to the spatial diversity of a tumor.22PubMed Central. First report of dual KRAS Y96C/Y96S resistance mutations detected by liquid biopsy in KRAS G12C-mutant metastatic colorectal cancer If validated in larger studies, routine liquid biopsy monitoring could give oncologists a window to switch therapies before resistance becomes clinically obvious.
Destroying KRAS Instead of Just Blocking It
A conceptually different approach skips inhibition altogether and tries to eliminate the KRAS protein from the cell. PROTACs (proteolysis-targeting chimeras) are bifunctional molecules: one end grabs KRAS, the other end recruits the cell’s protein-disposal system, tagging KRAS for destruction. Because a single PROTAC molecule can destroy multiple copies of the target protein through repeated cycles, these degraders can in theory achieve deeper suppression than a traditional inhibitor that must occupy every copy of KRAS simultaneously. Several KRAS-directed PROTACs have shown increased activity compared to their parent inhibitors in cell and animal models, though questions about toxicity in normal tissues remain unanswered.23PubMed Central. Development of PROTACS degrading KRAS and SOS1 The pan-KRAS degrader MCB-36, mentioned earlier, represents a convergence of these ideas: broad mutation coverage paired with a destruction mechanism.
Safety and Side Effects of Current Drugs
The two approved G12C inhibitors share some side effects but differ in their risk profiles in important ways. A real-world pharmacovigilance study drawing on thousands of adverse event reports found that sotorasib’s most prominent safety signal was liver toxicity, with elevated liver enzymes reported at rates far above background. Adagrasib, by contrast, showed a lower risk of liver problems but a higher rate of gastrointestinal side effects like nausea, vomiting, and diarrhea, along with a higher rate of serious adverse events overall.24PubMed. Safety assessment of KRAS (G12C) inhibitors based on the FDA Adverse Event Reporting System (FAERS) database A dedicated analysis of sotorasib’s liver effects confirmed that liver cell damage and elevated liver enzymes were substantially overrepresented compared to what would be expected across all cancer drugs.25PubMed Central. Hepatobiliary Adverse Events Associated With the KRAS p.G12C Inhibitor Sotorasib
The practical reality for patients on adagrasib is that side effects are usually manageable. Most patients need anti-nausea medications, and about half use anti-diarrheal drugs, but the rate of patients stopping treatment entirely because of side effects has been low, around 7%.26The Oncologist. Practical Guidance for the Management of Adverse Events in Patients with KRASG12C-Mutated Non-Small Cell Lung Cancer Receiving Adagrasib These side-effect differences can matter when choosing between the two drugs, particularly for patients who already have compromised liver function or are prone to gastrointestinal problems.
Brain Metastases and Getting Drugs Past the Blood-Brain Barrier
Lung cancer frequently spreads to the brain, and historically most targeted therapies have struggled to reach adequate concentrations there. Adagrasib appears to cross the blood-brain barrier more effectively than many oncology drugs. In preclinical brain metastasis models, it penetrated into cerebrospinal fluid, shrank brain tumors, and extended survival. In two patients with untreated brain metastases, measured drug concentrations in spinal fluid exceeded the level needed to inhibit the target, and both patients experienced regression of their brain lesions.27Clinical Cancer Research. Activity of Adagrasib (MRTX849) in Brain Metastases: Preclinical Models and Clinical Data from Patients with KRASG12C-Mutant Non–Small Cell Lung Cancer This is a meaningful clinical advantage. For patients whose KRAS G12C lung cancer has spread to the brain, the ability of adagrasib to reach that compartment could influence drug selection.
Combining KRAS Inhibitors With Immunotherapy
KRAS mutations do not just drive tumor growth; they also help tumors evade the immune system. Blocking KRAS can remodel the tumor’s local environment, making it more visible to immune cells. This creates a strong rationale for pairing KRAS inhibitors with immune checkpoint therapies like anti-PD-1 or anti-PD-L1 antibodies. In mouse models of lung and colorectal cancer, the combination of a KRAS G12C inhibitor with anti-PD-1 therapy produced strong synergy and in some cases led to long-term cures in the majority of treated animals.28Cancer Cell. KRAS Inhibitor: Emerging Strategies for Cancer Therapy An analysis of patients with KRAS-mutant non-small cell lung cancer found that combining immunotherapy with chemotherapy produced a response rate of about 46%, compared to roughly 35-37% with either approach alone.29CancerNetwork. Erica C. Nakajima, MD, Analyzes Frontline Immune Checkpoint Inhibitors in KRAS+, PD-L1+ Non-Small Cell Lung Cancer
Adding SHP2 inhibition to this mix may push the immune response further. In preclinical lung tumor models that were resistant to immunotherapy alone, combining a RAS inhibitor with an SHP2 inhibitor sensitized the tumors to checkpoint blockade, leading to efficient immune-mediated tumor rejection.30PubMed Central. Combining RAS(ON) G12C-selective inhibitor with SHP2 inhibition sensitises lung tumours to immune checkpoint blockade Triple combinations of KRAS inhibitor, SHP2 inhibitor, and immune checkpoint antibody are now being explored clinically.
Personalized Immune Approaches Targeting Mutant KRAS
Beyond checkpoint combinations, some researchers are treating the mutant KRAS protein itself as a target for the immune system. In a landmark case, a patient with metastatic colorectal cancer driven by KRAS G12D received an infusion of over 100 billion of their own immune cells that had been selected specifically because they recognized the mutant KRAS protein. All seven of the patient’s lung metastases regressed.31PubMed Central. T-Cell Transfer Therapy Targeting Mutant KRAS in Cancer This approach is limited by the fact that the immune recognition depends on specific tissue-compatibility molecules that vary between individuals, so it cannot be applied broadly without further engineering. But it offers proof of concept that the immune system can be directed specifically at mutant KRAS, opening a path toward vaccines and off-the-shelf cell therapies built on the same principle.
Pancreatic Cancer and the Stromal Barrier
Pancreatic ductal adenocarcinoma is the cancer type most thoroughly dominated by KRAS mutations, yet it has been among the hardest to treat with any targeted therapy. One major reason is the dense, fibrous tissue (desmoplasia) that surrounds pancreatic tumors and physically blocks drugs and immune cells from reaching the cancer. Early preclinical data suggest that combining KRAS pathway inhibition with agents that break down this stromal barrier can substantially reduce the deposition of structural proteins like collagen and fibronectin within the tumor. Immune profiling of treated tumors showed a robust increase in the infiltration of killer T cells, along with suppression of a survival protein that helps cancer cells resist death signals.32Cancer Research. Combined KRAS pathway inhibition and ONIOYID treatment enhances tumor regression, attenuates desmoplasia, and augments T cell infiltration in pancreatic ductal adenocarcinoma If these findings translate into patients, they could mark the beginning of meaningful KRAS-targeted therapy for a cancer that has resisted nearly every prior attempt at precision treatment.