Abstract: Kirsten rat sarcoma viral oncogene homolog (KRAS) is the most frequently mutated oncogene in human cancers, particularly in non-small cell lung cancer (NSCLC). For decades, KRAS was considered an "undruggable" target due to its picomolar affinity for GTP and the lack of deep binding pockets. The discovery of the Switch II allosteric pocket revolutionized this landscape, leading to the development of AMG 510 (Sotorasib), a first-in-class, small-molecule covalent inhibitor specifically targeting the KRAS G12C mutation. Sotorasib irreversibly binds to the mutant cysteine residue, locking the protein in an inactive GDP-bound state and halting downstream oncogenic signaling. Clinical trials, notably CodeBreaK 100, demonstrated significant objective response rates and progression-free survival, culminating in the FDA's accelerated approval of sotorasib for pretreated KRAS G12C-mutated NSCLC. Despite this breakthrough, the clinical utility of sotorasib is challenged by the rapid emergence of acquired resistance mechanisms, including secondary KRAS mutations (e.g., Y96D) and bypass signaling pathway activation, as well as limited efficacy in patients with specific co-mutations like KEAP1 and STK11. This review synthesizes the pharmacological activity, molecular mechanisms, structure-activity relationships, current limitations, and future perspectives of sotorasib in the treatment of NSCLC, highlighting the shift towards combination therapies and next-generation inhibitors to overcome resistance.
1. Introduction
Kirsten rat sarcoma viral oncogene homolog (KRAS) is a GTPase protein that serves as a critical signaling node in cell proliferation and survival. It is the most frequently mutated oncogene in human cancers, driving highly fatal malignancies including non-small cell lung cancer (NSCLC), colorectal cancer (CRC), and pancreatic ductal adenocarcinoma (PDAC) [3][6]. In NSCLC, the KRAS G12C mutation—a single amino acid substitution of glycine to cysteine at codon 12—is present in approximately 11% to 16% of patients [2]. Historically, patients harboring KRAS mutations have exhibited poor responses to standard therapies, representing a significant unmet clinical need [3].
For over 40 years, KRAS was deemed "undruggable." The protein's smooth surface lacks deep hydrophobic pockets for small-molecule binding, and its picomolar affinity for intracellular GTP makes competitive inhibition nearly impossible [2][3]. However, the identification of a cryptic allosteric pocket (the Switch II pocket) beneath the effector binding switch region provided a novel vulnerability [4][6]. This discovery paved the way for AMG 510 (Sotorasib), developed by Amgen. On May 28, 2021, sotorasib received accelerated approval from the US Food and Drug Administration (FDA) as the first targeted treatment for adult patients with KRAS G12C-mutated locally advanced or metastatic NSCLC who have received at least one prior systemic therapy [1][3][6].
2. Pharmacological Activity
Sotorasib has demonstrated robust pharmacological activity and clinical efficacy in NSCLC. In the pivotal Phase I/II CodeBreaK 100 trial, patients with heavily pretreated KRAS G12C-mutated NSCLC received oral sotorasib at a recommended dose of 960 mg once daily. The trial reported an objective response rate (ORR) of 37.1% and a disease control rate (DCR) of 80.6% [1][2][8]. The median progression-free survival (PFS) was 6.8 months, and the median overall survival (OS) reached 12.5 months, which compares favorably to historical benchmarks for salvage chemotherapy [2][8].
The Phase III CodeBreaK 200 trial subsequently compared sotorasib to docetaxel in previously treated metastatic NSCLC. Sotorasib significantly improved PFS (5.6 months vs. 4.5 months) and ORR (28.1% vs. 13.2%) compared to docetaxel. However, the study did not meet its secondary endpoint for OS, showing no statistically significant difference between the two arms (10.6 months for sotorasib vs. 11.3 months for docetaxel) [8][10].
Pharmacokinetically, sotorasib exhibits non-linear, time-dependent properties. Steady-state concentrations are achieved within 22 days with no appreciable accumulation, and the drug has a mean terminal elimination half-life of approximately 5 hours [1]. The safety profile is generally manageable. The most common treatment-related adverse events (TRAEs) include diarrhea, nausea, fatigue, and elevations in alanine aminotransferase (ALT) and aspartate aminotransferase (AST) [1][8]. Severe (Grade 3 or 4) hepatotoxicity and rare but potentially fatal interstitial lung disease (ILD)/pneumonitis have been noted, requiring dose modifications or discontinuation in a subset of patients [1].
3. Molecular Mechanism of Action
The KRAS protein functions as a molecular switch, oscillating between an active GTP-bound state and an inactive GDP-bound state. In its active form, KRAS engages downstream effectors to promote cellular proliferation and survival. The G12C mutation impairs intrinsic GTPase activity, causing the protein to accumulate in the active GTP-bound state [2][6].
Sotorasib is a highly selective, covalent inhibitor that exploits the fact that mutant KRAS G12C still cycles, albeit abnormally, between GTP- and GDP-bound states. Sotorasib specifically binds to the inactive GDP-bound conformation of the KRAS G12C protein [1][8]. It forms an irreversible covalent bond with the mutant cysteine residue at position 12. By occupying the Switch II pocket (S-IIP), sotorasib locks the KRAS protein in its inactive GDP-bound state [6]. This steric blockade prevents guanine nucleotide exchange factors (like SOS1) from catalyzing the exchange of GDP for GTP, thereby shutting down downstream mitogen-activated protein kinase (MAPK) signaling (e.g., MEK/ERK pathways) and inducing tumor regression [1][8].
4. Structure-Activity Relationship (SAR)
The development of sotorasib represents a triumph of structure-based drug design. Early prototype inhibitors, such as ARS-853 and ARS-1620, validated the concept of targeting the Switch II pocket and covalently binding to Cys12. However, these early molecules were limited by suboptimal potency and pharmacokinetic properties due to the small volume of the S-IIP, which restricted protein-ligand interactions [2][6][8].
The critical SAR breakthrough for AMG 510 (sotorasib) was the identification of an adjacent surface groove formed by an alternative orientation of the Histidine 95 (His95) residue on the KRAS G12C protein [2][3]. Sotorasib features a novel quinazolinone scaffold with an acrylamide moiety that acts as the warhead to form the covalent bond with Cys12 [6]. Crucially, the aromatic ring of sotorasib is optimized to occupy and bind to the His95 groove. This enhanced interaction provides a more rigid and favorable conformation, resulting in a nearly 10-fold improvement in inhibitory potency compared to its predecessor, ARS-1620 [2][3][11]. The precise spatial arrangement allows sotorasib to maintain high selectivity for the G12C mutant over wild-type KRAS [1].
5. Current Limitations
Despite its clinical success, the efficacy of sotorasib is hindered by several significant limitations:
Acquired Resistance: Resistance to sotorasib develops rapidly and is highly heterogeneous. On-target resistance mechanisms involve secondary mutations within the Switch II binding pocket (e.g., Y96D, R68S, H95D/Q/R, and G13D) that sterically hinder sotorasib binding [2][3][4]. The Y96D mutation is particularly notable as it confers resistance to all current KRAS G12C inhibitors [3]. Off-target bypass mechanisms are also prevalent, including MET amplification, activating mutations in NRAS, BRAF, MAP2K1 (MEK), and RET, as well as oncogenic fusions (ALK, RET, FGFR3) and loss-of-function mutations in NF1 and PTEN [3][4]. Histological transformation from adenocarcinoma to squamous cell carcinoma has also been documented as an escape mechanism [2][4].
Impact of Co-mutations: The genomic landscape of KRAS-mutant NSCLC heavily influences sotorasib efficacy. Patients harboring concurrent mutations in KEAP1, SMARCA4, or CDKN2A experience significantly inferior clinical outcomes, lower ORRs, and early disease progression (PFS < 3 months) [5][10]. STK11 co-mutations also serve as a negative prognostic biomarker, attenuating the durability of the response [5].
CNS Efficacy and Survival Benefit: Patients with active brain metastases were excluded from initial trials, leaving the central nervous system (CNS) penetrance and efficacy of sotorasib not fully established, though some intracranial activity has been observed [2][5]. Furthermore, the lack of a statistically significant overall survival (OS) benefit in the Phase III CodeBreaK 200 trial compared to docetaxel highlights the need for improved therapeutic strategies [10].
6. Future Perspectives
To address the limitations of monotherapy, the future of KRAS G12C targeting lies in rational combination strategies and the development of next-generation inhibitors.
Combination Therapies: Vertical inhibition of the RTK-RAS-MAPK pathway is being extensively explored to prevent or overcome bypass resistance. Clinical trials (e.g., CodeBreaK 101) are evaluating sotorasib in combination with SHP2 inhibitors (e.g., TNO155), SOS1 inhibitors, MEK inhibitors, and EGFR inhibitors (e.g., cetuximab, panitumumab) [2][4][7]. Additionally, because sotorasib has been shown to induce a pro-inflammatory tumor microenvironment and drive anti-tumor immunity, combinations with PD-1/PD-L1 immune checkpoint inhibitors (e.g., pembrolizumab, atezolizumab) hold significant promise and are currently under clinical investigation [3][6][9].
Next-Generation Inhibitors: To combat secondary mutations like Y96D, researchers are developing novel compounds. These include RAS(ON) inhibitors (e.g., RM-018, RM-007) that act as molecular glues targeting the active GTP-bound state of KRAS, and pan-RAS inhibitors capable of targeting multiple KRAS variants [3][9]. Proteolysis-targeting chimeras (PROTACs) designed to degrade the KRAS protein entirely are also in early development [9].
Biomarker-Driven Approaches: The integration of liquid biopsies (ctDNA) and comprehensive molecular profiling will be essential to identify co-mutations (e.g., KEAP1, STK11) and emerging resistance mechanisms in real-time. This will enable clinicians to stratify patients effectively, personalizing treatment sequences and selecting the most appropriate combination therapies for individual patients [2][5].
