Abstract: Kirsten rat sarcoma viral oncogene homolog (KRAS) is one of the most frequently mutated oncogenes in human cancers, historically considered an "undruggable" target. The development and subsequent FDA approval of AMG 510 (Sotorasib), a first-in-class covalent inhibitor specifically targeting the KRAS G12C mutation, marked a paradigm shift in precision oncology. Sotorasib exerts its pharmacological activity by irreversibly binding to the mutant cysteine residue (Cys12) within the Switch-II pocket, locking the KRAS protein in an inactive GDP-bound state and halting downstream MAPK signaling. Despite its impressive clinical efficacy in non-small cell lung cancer (NSCLC), the therapeutic durability of sotorasib is significantly limited by the rapid emergence of acquired resistance. This literature review comprehensively examines the pharmacological and structural profile of AMG 510, with a primary focus on the complex, multifaceted mechanisms of acquired resistance. These mechanisms include on-target secondary KRAS mutations (such as Y96D and H95 alterations), off-target bypass signaling through upstream receptor tyrosine kinases (RTKs) and alternative pathways (e.g., PI3K-AKT, MET, BRAF), phenotypic transformations like epithelial-to-mesenchymal transition (EMT), and histological shifts. Finally, the review explores future perspectives, including rational combination therapies and next-generation inhibitors designed to overcome these resistance barriers.
1. Introduction
The RAS gene family, including KRAS, HRAS, and NRAS, encodes GTPase proteins that act as critical molecular switches regulating cell proliferation, survival, and differentiation. KRAS is the most frequently mutated oncogene in human cancers, particularly prevalent in pancreatic ductal adenocarcinoma (PDAC), non-small cell lung cancer (NSCLC), and colorectal cancer (CRC) [1]. Oncogenic KRAS mutations disrupt GTPase-activating protein (GAP)-mediated GTP hydrolysis, trapping the protein in a continuously active, GTP-bound state [5]. For over four decades, KRAS was deemed "undruggable" due to its picomolar affinity for intracellular GTP and the lack of deep, traditional small-molecule binding pockets on its surface [4][6].
This historical barrier was broken with the discovery of a cryptic allosteric pocket (Switch-II pocket) in the KRAS G12C mutant, leading to the development of covalent inhibitors [6]. AMG 510 (Sotorasib) emerged as the first-in-class, highly selective, and irreversible small-molecule inhibitor of KRAS G12C. In May 2021, based on the promising results of the CodeBreaK 100 clinical trial, the U.S. Food and Drug Administration (FDA) granted accelerated approval to sotorasib for the treatment of adult patients with locally advanced or metastatic KRAS G12C-mutated NSCLC who have received at least one prior systemic therapy [1][9]. While this approval represents a milestone in targeted therapy, the clinical benefit of sotorasib is increasingly challenged by the rapid onset of acquired resistance, necessitating a deeper understanding of tumor adaptation mechanisms [3].
2. Pharmacological Activity
The clinical efficacy of sotorasib was robustly demonstrated in the phase I/II CodeBreaK 100 trial (NCT03600883). In the phase II NSCLC cohort (n=126), sotorasib administered orally at 960 mg once daily yielded an objective response rate (ORR) of 37.1% and a disease control rate (DCR) of 80.6% [2][9]. The median duration of response was 11.1 months, with a median progression-free survival (PFS) of 6.8 months and a median overall survival (OS) of 12.5 months [2][4]. Sotorasib demonstrated a favorable toxicity profile, with the most common treatment-related adverse events being diarrhea, nausea, fatigue, and reversible elevations in liver aminotransferases (ALT/AST) [1][9].
Interestingly, the pharmacological activity of sotorasib varies significantly by tumor histology. While highly active in NSCLC, its efficacy in colorectal cancer (CRC) is more modest. In the CRC cohort of the CodeBreaK 100 trial, the ORR was only 7.1% to 10%, with a DCR of 73.8% and a median PFS of 4.0 months [2][8]. This discrepancy is largely attributed to higher basal receptor tyrosine kinase (RTK) activation, specifically EGFR signaling, in CRC, which drives rapid feedback reactivation of the MAPK pathway upon KRAS inhibition [1][6].
3. Molecular Mechanism of Action
Unlike wild-type KRAS, the KRAS G12C mutant retains intrinsic GTPase activity and cycles between active (GTP-bound) and inactive (GDP-bound) states [9]. Sotorasib exploits this cycling by selectively and irreversibly binding to the mutant cysteine residue (Cys12) exclusively when the protein is in its inactive, GDP-bound conformation [3][6]. By occupying the Switch-II pocket (S-IIP), sotorasib sterically blocks the interaction between KRAS and guanine nucleotide exchange factors (GEFs), such as SOS1 [4]. This interaction traps the KRAS G12C protein in the inactive state, subverting its native nucleotide preference and profoundly impairing its ability to bind to downstream effector proteins like RAF [6]. Consequently, sotorasib durably suppresses the downstream RAF-MEK-ERK (MAPK) signaling cascade, leading to the inhibition of tumor cell proliferation and the induction of apoptosis [9].
4. Structure-Activity Relationship (SAR)
The structural design of sotorasib represents a significant optimization over early prototype inhibitors like ARS-853 and ARS-1620. Sotorasib features a novel quinazolinone scaffold that occupies the Switch-II pocket, coupled with an acrylamide warhead that forms a covalent bond with the nucleophilic thiol group of the mutant Cys12 residue [6]. A critical structural breakthrough in the development of sotorasib was the incorporation of an aromatic ring that binds to a cryptic surface groove formed by an alternative orientation of the Histidine 95 (His95) residue on the KRAS protein [1][4]. This enhanced protein-ligand interaction provides a more rigid and favorable conformation, increasing the inhibitory potency of sotorasib by nearly 10-fold compared to ARS-1620 [1][8].
Furthermore, the molecular architecture of sotorasib incorporates an axially chiral biaryl linkage, introducing the phenomenon of atropisomerism [12]. The configurationally stable axial chirality was a deliberate structural tool utilized during the drug's design to optimize its three-dimensional fit within the small volume of the Switch-II pocket, thereby maximizing target occupancy and metabolic stability [12].
5. Current Limitations
The primary limitation of sotorasib therapy is the rapid development of acquired resistance. Clinical and preclinical studies have revealed that resistance to KRAS G12C inhibitors is highly heterogeneous and often polyclonal, with multiple mechanisms occurring simultaneously within a single patient [1][8]. These mechanisms can be broadly categorized into on-target alterations, off-target bypass signaling, phenotypic transformations, and specific biomarker-driven adaptations.
5.1 On-Target Secondary KRAS Alterations
Acquired mutations within the KRAS gene itself are a major source of resistance. Secondary mutations frequently occur within the Switch-II binding pocket, directly impeding the binding of sotorasib. Identified pocket mutations include R68S, H95D, H95Q, H95R, Y96C, and Y96D [2][4]. Notably, structural differences between inhibitors dictate cross-resistance profiles: mutations at the H95 residue (e.g., H95D/Q/R) confer resistance to adagrasib but remain sensitive to sotorasib, whereas mutations at the Y96 residue (e.g., Y96D, Y96C) disrupt the binding of both drugs [3][5]. Additionally, patients develop activating non-G12C KRAS mutations (e.g., G12D, G12V, G12W, G13D, Q61H) often in a trans allelic configuration, which maintain the protein in an active GTP-bound state that sotorasib cannot inhibit [3][8]. High-level amplification of the KRAS G12C allele is also frequently observed [1].
5.2 Off-Target Bypass Signaling
Tumors frequently bypass KRAS G12C blockade by reactivating the MAPK pathway or parallel survival cascades. Vertical pathway reactivation often occurs via the upregulation or amplification of upstream RTKs, including EGFR, HER2, FGFR, and MET [3][6]. Downstream of KRAS, activating mutations in NRAS, BRAF, and MAP2K1 (MEK), as well as oncogenic fusions involving ALK, RET, BRAF, RAF1, and FGFR3, have been documented [1][8]. Furthermore, parallel signaling through the PI3K-AKT-mTOR pathway—often mediated by the IGFR-IRS1 axis or loss-of-function mutations in the tumor suppressor PTEN—provides an alternative survival mechanism independent of KRAS [5][6].
5.3 Phenotypic and Histological Transformations
Therapeutic pressure from sotorasib can induce epithelial-to-mesenchymal transition (EMT), driven by TGF-β signaling and SNAIL expression, rendering cells resistant to KRAS inhibition [3][5]. Additionally, a histological transformation from lung adenocarcinoma to squamous cell carcinoma (SCC) has been observed in patients progressing on KRAS G12C inhibitors, mirroring resistance patterns seen with EGFR inhibitors [4][8].
5.4 Biomarkers of Resistance
Co-occurring genomic alterations significantly influence the durability of sotorasib therapy. Mutations in KEAP1, STK11, SMARCA4, and CDKN2A are critical prognostic biomarkers associated with early disease progression (PFS < 3 months) and inferior clinical outcomes [10][11]. Specifically, KEAP1 mutations have been identified as a strong negative predictive biomarker for KRAS G12C inhibitor efficacy [10].
6. Future Perspectives
To overcome the complex, polyclonal resistance mechanisms associated with sotorasib, the future of KRAS-targeted therapy relies heavily on rational combination strategies. Clinical trials are currently evaluating sotorasib in combination with downstream MEK inhibitors, upstream EGFR inhibitors (e.g., cetuximab, particularly for CRC), and SHP2 or SOS1 inhibitors (e.g., TNO155, BAY-293) to prevent RTK-mediated feedback reactivation [4][5][7]. Combinations with PI3K/AKT inhibitors and immune checkpoint inhibitors (PD-1/PD-L1) are also under active investigation to address parallel pathway activation and leverage tumor microenvironment modulation [6][9].
Beyond combination therapies, the development of next-generation inhibitors is paramount. Novel pan-RAS inhibitors and RAS(ON) inhibitors (e.g., RM-018), which act as molecular glues to target the active GTP-bound state of KRAS, show promise in overcoming secondary pocket mutations like Y96D [1][5]. Additionally, targeted protein degradation via Proteolysis-Targeting Chimeras (PROTACs) represents an innovative frontier for eliminating oncogenic RAS entirely [5]. Finally, the integration of circulating tumor DNA (ctDNA) liquid biopsies into clinical practice will be essential for the real-time monitoring of resistance mutations, enabling dynamic and personalized sequencing of therapies [4].
