Abstract: Kirsten rat sarcoma viral oncogene homolog (KRAS) mutations are prevalent oncogenic drivers in colorectal cancer (CRC), historically considered "undruggable" due to the protein's high affinity for GTP and lack of deep binding pockets. The discovery of the KRAS G12C allosteric switch-II pocket revolutionized targeted therapy, leading to the development of sotorasib (AMG 510), a first-in-class, irreversible, covalent small-molecule inhibitor. While sotorasib has demonstrated groundbreaking efficacy and received regulatory approval for KRAS G12C-mutated non-small cell lung cancer (NSCLC), its monotherapy efficacy in metastatic colorectal cancer (mCRC) has been modest, characterized by lower objective response rates and shorter progression-free survival. This limited efficacy is primarily attributed to the rapid reactivation of upstream epidermal growth factor receptor (EGFR) signaling and the emergence of acquired resistance mutations. Consequently, current research directions in CRC heavily emphasize combinatorial strategies, particularly dual blockade of KRAS G12C and EGFR, to overcome intrinsic and acquired resistance. This review comprehensively analyzes the pharmacological activity, molecular mechanism, structure-activity relationship, current limitations, and future therapeutic perspectives of sotorasib in the context of KRAS G12C-mutated colorectal cancer.
1. Introduction
Colorectal cancer (CRC) is a leading cause of cancer-related mortality worldwide, with metastatic disease (mCRC) presenting a significant clinical challenge. Mutations in the RAS oncogene family, particularly KRAS, occur in approximately 40% of CRC cases and are associated with aggressive tumor biology, poor prognosis, and intrinsic resistance to standard anti-epidermal growth factor receptor (EGFR) therapies such as cetuximab and panitumumab [1] [2]. Among these, the KRAS G12C mutation—characterized by a glycine-to-cysteine substitution at codon 12—is found in approximately 3% to 4% of CRC patients [1] [4]. For decades, KRAS was deemed an "undruggable" target due to its picomolar affinity for intracellular GTP and the absence of accessible pharmacological binding pockets [6] [7].
The therapeutic landscape shifted dramatically with the discovery of a cryptic allosteric pocket (the switch-II pocket) in the KRAS G12C mutant protein, which preserves a near-wild-type intrinsic GTPase activity and cycles between active (GTP-bound) and inactive (GDP-bound) states [1] [4]. This vulnerability led to the development of sotorasib (AMG 510), a first-in-class small molecule that covalently binds to the mutant cysteine residue, locking the protein in its inactive state [4]. While sotorasib achieved accelerated FDA approval for KRAS G12C-mutated non-small cell lung cancer (NSCLC) [4] [7], its application in CRC has revealed unique biological challenges, prompting extensive research into its pharmacological profile and synergistic combination strategies [2].
2. Pharmacological Activity
Sotorasib is an orally bioavailable inhibitor that has been extensively evaluated in the phase I/II CodeBreaK 100 clinical trial. In the phase I dose-escalation and expansion cohorts, sotorasib demonstrated a manageable safety profile with no dose-limiting toxicities, establishing 960 mg once daily as the recommended phase II dose [1] [8]. Pharmacokinetically, sotorasib exhibits non-linear, time-dependent exposure, reaching steady-state concentrations within 22 days with a mean terminal elimination half-life of approximately 5 hours. It is primarily metabolized via oxidative metabolism by CYP3A and non-enzymatic conjugation [4].
In the context of mCRC, the clinical activity of sotorasib monotherapy has been modest compared to NSCLC. In the phase I portion of CodeBreaK 100 involving 42 heavily pretreated mCRC patients, the objective response rate (ORR) was 7.1%, with a disease control rate (DCR) of 73.8% and a median progression-free survival (mPFS) of 4.0 months [1] [3] [4]. The phase II expansion cohort, which included 62 refractory KRAS G12C-mutant mCRC patients, confirmed these findings, reporting an ORR of 9.7% to 10%, a DCR of 82.3%, an mPFS of 4.0 months, and a median overall survival (mOS) of 10.6 months [2] [8]. The most common treatment-related adverse events (TRAEs) across solid tumors include diarrhea, musculoskeletal pain, nausea, fatigue, and hepatotoxicity (elevated AST/ALT), which are generally manageable with dose modifications [4] [8].
3. Molecular Mechanism of Action
Sotorasib functions as a highly selective, irreversible covalent inhibitor of the KRAS G12C mutant protein. Unlike wild-type KRAS, the G12C mutation introduces a reactive thiol group at the cysteine 12 position. Sotorasib exploits the intrinsic GTP-to-GDP cycling ability of KRAS G12C by selectively binding to the switch-II pocket (S-IIP) only when the protein is in its inactive, GDP-bound conformation [1] [6].
Upon binding, sotorasib forms a covalent bond with the mutant cysteine, effectively trapping the KRAS protein in the inactive state and subverting its nucleotide preference from GTP to GDP [6]. This irreversible blockade prevents the interaction of KRAS with guanine nucleotide exchange factors (GEFs) such as SOS1, thereby halting SOS1-catalyzed nucleotide exchange [4]. Consequently, sotorasib disrupts the engagement of KRAS with downstream effector proteins like RAF, leading to a durable suppression of the mitogen-activated protein kinase (MAPK) signaling cascade (evidenced by reduced ERK phosphorylation) and selectively impairing the viability of KRAS G12C-mutant cancer cells without affecting wild-type KRAS [4] [6].
4. Structure-Activity Relationship (SAR)
The structural design of sotorasib represents a significant optimization over early prototype KRAS G12C inhibitors such as ARS-853 and ARS-1620. Sotorasib features a novel quinazolinone scaffold that precisely occupies the switch-II pocket of the KRAS protein [6]. A critical component of its structure is the acrylamide warhead, which acts as an electrophile to form a covalent bond with the nucleophilic thiol group of the mutated Cys12 residue [6].
A major breakthrough in the SAR of sotorasib was the substitution of the quinazoline nitrogen (N1) present in earlier molecules, allowing the drug to exploit a novel surface groove formed by an alternative orientation of the histidine 95 (His95) residue on the KRAS protein [2] [7]. The aromatic ring of sotorasib engages in a strong interaction with His95, which enhances the rigidity and favorable conformation of the drug-target complex. This specific interaction increases the binding affinity and overall inhibitory potency of sotorasib by approximately 10-fold compared to its predecessor, ARS-1620, enabling profound tumor regression in preclinical models [2] [6] [10].
5. Current Limitations
Despite its success in NSCLC, the clinical efficacy of sotorasib monotherapy in mCRC is severely limited by intrinsic and acquired resistance mechanisms. The primary intrinsic limitation is the high redundancy of the KRAS signaling network in CRC. Inhibition of KRAS G12C in CRC cells leads to a rapid, adaptive feedback reactivation of upstream receptor tyrosine kinases (RTKs), predominantly EGFR. Because CRC remains highly dependent on EGFR signaling, this feedback loop restores MAPK pathway activation, thereby blunting the anti-tumor effect of sotorasib [2] [10].
Furthermore, acquired resistance inevitably develops in patients who initially respond. Genomic and histologic analyses have identified multiple bypass mechanisms, including secondary mutations within the KRAS switch-II pocket (e.g., Y96D, Y96C, H95D/Q/R, R68S) that sterically hinder sotorasib binding [2] [3] [6]. Other resistance mechanisms involve polyclonal alterations converging on RAS-MAPK reactivation, such as activating mutations in NRAS, BRAF, MAP2K1, alternative KRAS mutations (G12D/V/W, G13D), high-level amplification of the KRAS G12C allele, and MET amplification [2] [3]. Additionally, phenotypic adaptations, such as epithelial-to-mesenchymal transition (EMT) induced by PI3K or TGF-β activation, have been implicated in rendering CRC cells resistant to KRAS G12C blockade [2] [6].
6. Future Perspectives
To overcome the limitations of sotorasib monotherapy in mCRC, the clinical paradigm has rapidly shifted toward rational combination therapies. Given the prominent role of EGFR feedback reactivation, dual blockade of KRAS G12C and EGFR is the most promising strategy. In the phase Ib CodeBreaK 101 trial, the combination of sotorasib and the anti-EGFR antibody panitumumab demonstrated a highly encouraging ORR of 30% and a DCR of 93% in refractory mCRC patients, significantly outperforming monotherapy [2]. This has led to the initiation of phase III randomized trials, such as CodeBreaK 300 and CodeBreaK 301, which are evaluating sotorasib plus panitumumab with or without FOLFIRI chemotherapy against standard-of-care regimens in mCRC [2] [11].
Beyond EGFR inhibitors, vertical pathway inhibition is being explored by combining sotorasib with downstream MEK inhibitors, PI3K/AKT/mTOR inhibitors, and upstream SHP2 allosteric inhibitors (e.g., TNO155) to prevent adaptive resistance [1] [3] [6]. Furthermore, preclinical evidence suggests that sotorasib can convert an immunosuppressive "cold" tumor microenvironment into a pro-inflammatory one by increasing the infiltration of CD8+ T cells, macrophages, and dendritic cells [2] [10]. Consequently, combinations of sotorasib with immune checkpoint inhibitors (anti-PD-1/PD-L1) are currently under investigation, although overlapping toxicities, particularly hepatotoxicity, require careful dose optimization [2] [6]. Ultimately, the integration of sotorasib into multi-targeted regimens holds the potential to transform KRAS G12C from an undruggable target into a manageable vulnerability in colorectal cancer.
