Literature review of advances and challenges in KRAS G12C mutant non-small cell lung cancer

Abstract

Background and Objective

Kirsten rat sarcoma viral oncogene homolog (KRAS) mutations are one of the most common oncogenic driver mutations in non-small cell lung cancer (NSCLC), with the KRAS G12C mutation being the most common prevalent subvariant. The review aims to explore optimal diagnostic and therapeutic strategies for KRAS G12C mutant NSCLC, and to provide guidance for the development of precise treatment approaches for affected.

Methods

A comprehensive search was conducted in PubMed, Embase, Web of Science, MEDLINE, the Cochrane Library, and major international conferences proceedings for all English-language publications available up to December 31, 2024. Relevant studies were systematically reviewed, analyzed, and synthesized to inform this review.

Key Content and Findings

In this review, we explore the KRAS G12C mutation and its associated signaling pathways, detection techniques, recent advancements in drug development and mechanisms of therapeutic resistance. The KRAS G12C mutation was once considered “undruggable” until the breakthrough approval of two targeted inhibitors: AMG510 (sotorasib) and MRTX849 (adagrasib). In China, IBI351 and D-1553 have also been approved for the treatment of adult patients with advanced NSCLC harboring the KRAS G12C mutation. Although currently approved only as second-line therapies for metastatic disease, these inhibitors—along with ongoing development of additional KRAS-targeted agents—are significantly advancing our understanding of KRAS-driven tumor biology. Notably, recent findings indicate that combining dual immune checkpoint inhibitors (ICIs; durvalumab and tremelimumab) with chemotherapy (CT) in patients with advanced NSCLC, including those with KRAS mutations, can result in durable survival benefits. This approach is emerging as a promising first-line treatment strategy.

Conclusions

The landscape of KRAS-mutant NSCLC has undergone substantial progress, marked by the successive approval of multiple KRAS G12C inhibitors and the development of novel targeted therapies. Moreover, the POSEIDON trial has highlighted the potential of dual ICI therapy combined with CT to achieve sustained clinical benefits. Despite these advances, the heterogeneity of tumor responses underscores the need for further investigation into intrinsic resistance mechanisms and the strategic optimization of combination therapies to enhance treatment outcomes.

Introduction

Non-small cell lung cancer (NSCLC) is a malignant tumor primarily composed of adenocarcinoma and squamous cell carcinoma. Kirsten rat sarcoma viral oncogene homolog (KRAS) is one of the most common mutated genes in NSCLC, accounting for approximately 25% of lung adenocarcinomas in Western countries. In Asian populations, KRAS mutations are present in approximately 10–15% of patients (1). Among these, the KRAS G12C mutation is the most common subvariant, accounting for about 30–40% of all KRAS mutations (2). KRAS mutations in NSCLC are often associated with smoking history, affecting approximately 6% of never smokers and 34% of smokers (3). Notably, KRAS G12C mutations were more prevalent in female patients (43.4%, P=0.007), who are also typically younger than male patients with the same mutation (median age 65 vs. 69 years; P<0.001) (3). Additionally, NSCLC patients with KRAS mutations often harbor mutations in other genes, including tumor protein p53 (TP53, 39.4%), serine/threonine kinase 11 gene (STK11, 19.8%), and Kelch-like ECH-associated protein 1 gene (KEAP1, 12.9%) (4). While STK11 deletion does not significantly impact clinical outcomes, it can lead to poor responses to programmed cell death (ligand) 1 [PD-(L)1] inhibitors (5), in contrast to patients with combined TP53 mutations who had a better survival prognosis with immune-combination chemotherapy (CT) (6). Furthermore, co-mutations in KEAP1, SWI/SNF-related, matrix-associated, actin-dependent regulator of chromatin, subfamily A, member 4 (SMARCA4), and cyclin-dependent kinase inhibitor 2A (CDKN2A) are associated with significantly shorter progression-free survival (PFS) and overall survival (OS) in patients with KRAS G12C-mutant NSCLC (7). This article reviews the signaling pathways of G12C mutations, diagnostic standards and their latest drug advances, explores different mechanisms of drug resistance and strategies to deal with them, with a view to providing the best diagnostic and therapeutic options for patients with KRAS G12C mutant NSCLC. We present this article in accordance with the Narrative Review reporting checklist (available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2025-164/rc).

Methods

A comprehensive search was conducted in PubMed, Embase, Web of Science, MEDLINE, the Cochrane Library, and major international conferences proceedings for all English-language publications available up to December 31, 2024. The search terms included “KRAS G12C mutation”, “Non-small cell lung cancer”, “Advance”, “signaling pathways”, “Sotorasib”, “Adagrasib”, etc. Relevant articles were selected to provide detailed information about KRAS-mutant NSCLC. All the articles included are reported in the reference list. The search strategy is detailed and shown in Table 1.

Table 1. The search strategy summary.

Items	Specification
Date of search	1/11/2024–31/12/2024
Databases and other sources searched	PubMed, Embase, Web of Science, MEDLINE, the Cochrane Library, and major international conferences
Search terms used	MeSH terms: “KRAS G12C mutation”, “Non-small cell lung cancer”, “Progress”, “Advance”, “signaling pathways”, “Challenge”, “resistance”, “Treatment”, “Targeted therapy”, “Immunotherapy”
	Free text search terms: “KRAS G12C-mutated NSCLC”, “Sotorasib”, “Adagrasib”, “biomarkers in KRAS G12C-mutant NSCLC”, etc.
	Filters: the following filters are applied in the PubMed database: select ‘Review’ and ‘Clinical Trial’ for document type; set the publication time to ‘2010–2024’; and select ‘English’ for language. In other databases, similar criteria were used to obtain literature that met the requirements
Timeframe	1/1/2010–31/12/2024
Inclusion and exclusion criteria	Inclusion criteria:
	(I) The study subjects are non-small cell lung cancer patients with KRAS G12C mutation
	(II) The study involved the signaling pathway, detection method, treatment, drug resistance mechanism and other related fields of KRAS G12C mutation
	(III) Study type is based on clinical research, basic research and review literature, etc.
	(IV) The language is mainly English
	Exclusion criteria:
	(I) Duplicate publications
	(II) Incomplete and unavailable data
	(III) Conference abstracts without full text
	(IV) Exclude low-quality literature such as inappropriate trial methods and high rate of lost visits
Selection process	Literature screening was conducted independently by J.X.Y. and Y.H., following a two-stage process of initial screening and rescreening. In the primary screening stage, the investigators read the titles and abstracts of the literature to exclude literature that was clearly not relevant to the topic of KRAS G12C-mutant NSCLC, such as literature that examined other mutation types or other cancer types
	In the rescreening stage, the literature retained from the initial screening was read in full and screened based on predetermined inclusion and exclusion criteria
	In case of disagreement between the two researchers on the results of the literature screening, consensus will be reached through discussion; if still uncertain, a third senior expert will be consulted for adjudication
Any additional considerations, if applicable	Although multiple databases and comprehensive search terms were used, the search terms might not be set up well enough, resulting in some relevant literature not being retrieved and the risk of omission. Secondly, due to the rapid progress of research in this field, the search was conducted up to December 2024, which may not be able to cover the latest research results in a timely manner. These limitations may affect the comprehensiveness and accuracy of the review findings to some extent and need to be taken into account when interpreting the review results

NSCLC, non-small cell lung cancer.

KRAS targets and signaling pathways

KRAS G12C mutations in NSCLC

RAS family encodes a membrane-bound regulatory protein (G protein) with a guanine triphosphatase (GTPase), originally described by Harvey and Kirsten in the 1960s as a retrotransposon involved in cellular proliferation, differentiation, and survival (8). RAS proteins include KRAS, the Harvey rat sarcoma viral oncogene homolog (HRAS) and the neuroblastoma rat sarcoma viral oncogene homolog (NRAS). Of these, KRAS mutations are the most prevalent, accounting for 75% of RAS-mutated tumors (9). KRAS mutations are common in pancreatic adenocarcinomas, colon and rectal adenocarcinomas, and lung adenocarcinomas. In lung adenocarcinoma, KRAS mutations account for 90% of RAS mutations, and about 97% of KRAS mutations in NSCLC occur at codons G12, G13, and Q61 (10). Analysis of KRAS-mutant lung adenocarcinoma patients from the Lung Cancer Mutation Consortium (LCMC) by El Osta et al. revealed that G12C was the most common subvariant, accounting for approximately 39% of cases, followed by G12V and G12D, each present in 18% of cases (11). Different KRAS mutations activate distinct downstream signaling pathways, for example, the KRAS G12A mutation preferentially activate the phosphatidylinositol-4,5-bisphosphate 3-kinase (PI3K) and mitogen-activated protein kinase (MAPK) pathways, while the G12C and G12V mutations preferentially activate the Ras-like protein (RAL) guanine nucleotide dissociation stimulator (Ral-GDS) signaling pathway and reduce protein kinase B (AKT) activation (12). The G12D mutation may enhance glutathione-mediated detoxification (13).

Cyclic regulation of KRAS protein activity

KRAS functions as a membrane-bound regulatory protein, cycling between an active guanosine triphosphate (GTP)-bound state and an inactive guanosine diphosphate (GDP)-bound state (specific mechanisms are presented in Figure 1). In the resting state, KRAS maintains the inactive state bound to GDP. When growth factors or other stimuli activate the corresponding receptors on the cell membrane, KRAS is switched to its active state by binding to GTP. This transition activates downstream signaling proteins, and the intrinsic GTPase activity of KRAS hydrolyzes GTP to GDP, deactivating the protein and halting signaling (14). This cycle ensures that KRAS-mediated signaling is brief and responsive to external stimuli. Normally, the GTP hydrolysis rate of KRAS is low, and it requires interaction with GTPase-activating proteins (GAPs) to accelerate GTP hydrolysis. Guanine nucleotide exchange factor (GEF) facilitate the conversion of KRAS protein to its active state. The coordinated function of GAPs and GEFs ensure efficient signaling. When mutations in the KRAS gene, such as the G12C mutation, result in structural changes to the protein, leading to a loss of intrinsic GTPase activity (15). As a result, KRAS remains in an activated state, constantly driving downstream signaling pathways. This persistent activation disrupts normal cellular processes, including regulation of proliferation and survival, contributing to uncontrolled cell growth and tumorigenesis.

Figure 1.

KRAS-related signaling pathways and mechanisms of action. AKT, protein kinase B; BADs, Bcl-XL/Bcl-2-associated death promoters; BCL-X, B-cell lymphoma-extra large; CaM, calmodulin; cAMP, cyclic adenosine monophosphate; CDC42, cell division cycle 42; DAG, diacylglycerol; ERK, extracellular-regulated protein kinase; FOXO, forkhead box O; GAP, GTPase-activating protein; GDP, guanosine diphosphate; GEF, guanine nucleotide exchange factor; GRB2, growth factor receptor-bound protein 2; GTP, guanosine triphosphate; IP3, inositol trisphosphate; KRAS, Kirsten rat sarcoma viral oncogene homolog; MEK, mitogen-activated protein kinase; mTOR, mammalian target of rapamycin; NF-κB, nuclear factor kappa-B; PI3K, phosphatidylinositol-4,5-bisphosphate 3-kinase; PIP2, phosphatidylinositol 4,5-bisphosphate; PIP3, phosphatidylinositol 3,4,5-trisphosphate; PLCε, phospholipase C ε; PLD, phospholipase D; PKC, proteins kinase C; RAC, Ras-related C3 botulinum toxin substrate; RAF, fibrosarcoma; RAL, Ras-like protein; RALGDS, RAL guanine nucleotide dissociation stimulator; RAS-GRF1, Ras protein specific guanylate releasing factor 1; RTK, receptor tyrosine kinase; SHP2, Src homology phosphatase 2; SOS, son of sevenless; TIAM1, T-cell lymphoma invasion and metastasis inducing protein 1; TBK1, TANK-binding kinase 1.

Activation of KRAS upstream pathway

The activation of KRAS protein requires the regulation of a variety of upstream regulators such as growth factors, chemokines, Ca²⁺ or receptor tyrosine kinase (RTK). The more common upstream signaling molecules are many growth factors such as epidermal growth factor (EGF). When EGF binds to its receptor, the EGF receptor (EGFR), it undergoes dimerization and phosphorylation. The phosphorylated EGFR then binds to growth factor receptor-bound protein 2 (Grb2) through its SH2 domain. The SH3 domain of Grb2 binds to son of sevenless 1 (SOS1, a kind of GEF), which activates SOS. SOS promotes the exchange of GDP for GTP on KRAS, converting it from an inactive to an active state. This activation allows KRAS to control downstream effectors and initiate cell proliferation and growth. In addition, stimulation of RTK, Ca²⁺, and chemokines can also activate KRAS protein (16), triggering further downstream signaling.

Downstream signaling pathways of KRAS

RAF-MEK-ERK pathway

KRAS acts as a binary molecular switch, cycling between GDP- and GTP-bond states. The downstream signaling pathways it mediates are complex and diverse. One of the most well-defined pathways is the fibrosarcoma (RAF)-mitogen-activated protein kinase (MEK)-extracellular-regulated protein kinase (ERK) signaling cascade. When activated, KRAS proteins promote the localization of RAF, a serine/threonine-specific protein kinase, to the cell membrane. This triggers conformational changes in RAF, activating it through dimerization. Activated RAF binds to and phosphorylates MEK1/2, which in turn activates ERK1/2. The RAF-MEK-ERK pathway plays a critical role in regulating cell proliferation, differentiation, migration, and survival (16).

PI3K-AKT-mTOR pathway

KRAS also plays a key role in the PI3K-AKT-mammalian target of rapamycin (mTOR) signaling pathway, which is involved in the regulation of cell proliferation, differentiation, apoptosis, glucose transport and other cellular life activities. When activated KRAS proteins activate PI3K, phosphorylated PI3K converts phosphatidylinositol 4,5-bisphosphate (PIP2) to phosphatidylinositol 3,4,5-trisphosphate (PIP3) (17), and PIP3 promotes the activation of phosphatidylinositol-dependent kinase 1 (PDK1), which in turn phosphorylates with mTOR complex 2 (mTORC2) different sites of AKT, resulting in the full activation of AKT (18). AKT activation activates or inhibits a number of downstream pathways, such as mTOR, Bcl-XL/Bcl-2-associated death promoters (BADs), the transcription factor Forkhead box O (FOXO), and nuclear factor kappa-B (NF-κB), thereby regulating the life activities of cells, such as proliferation, survival, metabolism, and apoptosis (19).

RalGDS-RAL-PLD pathway

Ral-GDS is another key effector of KRAS (20), which promotes the transition of RAL from GDP-bound to GTP-bound state, and downstream signaling to promote cell cycle progression. Downstream effectors of RAL proteins include phospholipase D (PLD) associated with endocytosis, TANK-binding kinase 1 (TBK1) associated with viral immunity, and cell division cycle 42 (CDC42) associated with cell migration (16). The Ral-GDS-RAL-PLD pathway plays an important role in KRAS-driven tumorigenesis, especially in the regulation of vesicle transport and cytoskeletal organization (8). In lung adenocarcinoma, KRAS G12C mutations preferentially activate this pathway (21).

Other signaling pathways

KRAS is also linked to additional signaling pathways. It has been shown that recombinant T-cell lymphoma invasion and metastasis inducing protein 1 (TIAM1) is a downstream effector of KRAS proteins (22), which acts as a ras-related C3 botulinum toxin substrate 1 (RAC1)-specific GEF to promote RAC binding to GTP to activate the RAC1 signaling pathway that affects cell migration, adhesion, actin protein cytoskeleton formation, endocytosis, and membrane transport. In addition, KRAS proteins can also regulate the phospho-acylinositol signaling pathway through activation of phosphatidylinositol-specific phospholipase C ε (PLCε). PLCε is a key signaling enzyme that hydrolyzes PIP2 to generate inositol trisphosphate (IP₃), which stimulates an increase in intracellular Ca²⁺ and activation of proteins kinase C (PKC) (23). Stephen et al. point out that potential downstream effectors of RAS include the novel RAS effector 1A (NORE1A), Af6, RAS and Rab interactor 1 (RIN1), and growth factor receptor 14 (Grb14) (21,24). In conclusion, the signaling pathways regulated by KRAS are intricate and multifaceted, influencing various cell processes such as cell proliferation, differentiation, migration, and apoptosis. The diverse downstream effects of KRAS mutations contribute significantly to tumorigenesis, making KRAS an important target for therapeutic intervention in cancers such as lung adenocarcinoma.

Detection methods

With the successful approval of several targeted drugs, it is crucial to choose appropriate assays for the accurate detection of KRAS G12C. Currently, commonly used methods for KRAS mutation detection include Sanger sequencing, reverse transcription polymerase chain reaction (RT-PCR), next-generation sequencing (NGS), and liquid biopsy, among others. Each detection method has its own set of advantages and limitations.

Sanger sequencing amplifies the KRAS gene fragment by PCR, then sequences the amplified product and determines the mutation site based on the base peak map. As the gold standard for KRAS mutation detection, it provides unquestionable accuracy and can detect all KRAS mutations. However, its sensitivity is relatively low and generally only detects mutations with a frequency greater than 15% (25). Therefore, more sensitive genetic tests are required to detect complex or low-frequency mutations.

RT-PCR detects the abundance of specific PCR targets by monitoring the production of PCR products in real time. Compared to Sanger sequencing, RT-PCR has higher sensitivity for detecting KRAS mutations, reliably identifying mutations as low as 1%. It is also simple and technically proficient, making it the preferred method for KRAS mutation detection. However, it has low throughput and is only suitable for detecting known loci.

NGS involves breaking DNA or RNA into small fragments, constructing libraries, and performing high-throughput sequencing. It identifies KRAS mutations through bioinformatics analysis of large datasets. Compared to the previous two sequencing methods, NGS offers both high sensitivity and high throughput, capable of detecting various mutation types, such as short sequence variants, gene copy number variants, and gene rearrangements (26,27). However, NGS is complex, time-consuming, and costly, highlighting the need for the development of an affordable and comprehensive NGS panel.

NGS is also indicated for liquid biopsies. In recent years, liquid biopsy has gained considerable attention due to its minimally invasive nature and its ability to consistently track disease progression and treatment response (28). However, early-stage tumors often yield false-negative results, likely due to the limited presence of circulating tumor DNA (ctDNA) in the blood. This underscores the importance of exploring emerging technologies such as multimodal liquid biopsies, which combine ctDNA, exosomes, and circulating tumor cells (29). Liquid biopsy is not routinely used for RAS mutations but can be performed when tissue samples are unavailable, especially in patients with advanced primary diagnoses or those who have developed resistance to targeted therapy. cfDNA liquid biopsy can then be used to detect a broad range of molecular markers, including KRAS mutations, providing a basis for targeted therapy.

Treatment strategies

AMG510

The KRAS protein consists of two major structural domains, the G domain and the hypervariable region (HVR). The G domain includes the switch I, II and P loop regions, which undergo conformational changes between GDP/GTP-bound states. The HVR contains the CAAX motif, where KRAS is isoprenylated at a cysteine residue (14), anchoring the protein to cellular membranes. AMG510 is the first Food and Drug Administration (FDA)-approved small molecule inhibitor that specifically targets KRAS G12C (the evolution of the chosen treatment is presented in Figure S1), blocking oncogenic signaling by specifically binding to a pocket in the KRAS switch II region that is present only in the inactivated GDP-binding conformation, thereby rendering the KRAS G12C-mutant protein inactive. AMG510 has demonstrated efficient and highly selective inhibition of cell viability in both pancreatic and lung adenocarcinoma cell lines. It has been shown to inhibit ERK phosphorylation in KRAS G12C mutants and enhance the antitumor efficacy of chemotherapeutic and targeted drugs in xenografts and homozygous mouse models, as well as in patient-derived xenografts (30). Additionally, AMG510 has been reported to restore effective immune response against tumors (31).

In the phase I first-in-human clinical trial study (CodeBreak 100: NCT03600883), 129 patients with advanced solid tumors were evaluated, including 59 patients with NSCLC. Among these, 19 patients [32.2%, 95% confidence interval (CI): 20.62% to 45.64%] showed objective remission, while 52 patients had controlled disease (88.1%, 95% CI: 77.07% to 95.09%). The median PFS (mPFS) was 6.3 months (all specific research data are presented in Table 2) (32). These promising results demonstrated the significant efficacy of AMG510 monotherapy, prompting the development of further single-arm phase II trials in patients with KRAS G12C-mutated NSCLC. In a phase II trial, 126 patients were enrolled, 81% of whom had previously received platinum-based CT and PD-1 or PD-L1 immunotherapy. The study results found an objective remission rate (ORR) of 37.1%, a mPFS of 6.8 months, and a median OS (mOS) of 12.5 months, consistent with the results of the phase I results. In terms of safety, 25 patients (19.8%) experienced grade 3 adverse events (AEs) and 1 patient (0.8%) experienced a grade 4 AE (33).

Table 2. Clinical trial results of drugs for KRAS-mutant NSCLC (accessible at clinicaltrials.gov).

Drugs	Clinical trial	Phase	Setting	No. of patients	ORR	DOR (months)	DCR	mPFS (months)	mOS (months)	TRAEs (grade ≥3)
Directly targeted inhibitors
AMG510	CodeBreak100 (NCT03600883) (32,33)	I	Previous radiotherapy and chemotherapy	59	32.2%	4	88.1%	6.3
		II		126	37.1%	11.1	80.6%	6.8	12.5	20.6%
	CodeBreak200 (NCT04303780) (34,35)	III	AMG510 vs. docetaxel	171 vs. 174	28.1% vs. 13.2%	8.6 vs. 6.8	82.5% vs. 60.3%	5.6 vs. 4.5	10.6 vs. 11.3	33% vs. 40%
	J.A. Stratmann et al., 2024 (36)	Retrospective cohort study		163	38.7%	7.9	78.7%	4.8	9.8	22.8%
MRTX849	KRYSTAL-1 (NCT03785249) (37)	I	Previous radiotherapy and chemotherapy	15	53.3%	16.4		11.1	NR
	KRYSTAL-1 (NCT03785249) (38)	II		116	42.9%	8.5	79.5%	6.5	12.6	44.8%
	KRYSTAL-12 (NCT04685135) (39)	III	MRTX849 vs. docetaxel	301 vs. 152	31.9% vs. 9.2%	8.3 vs. 5.4		5.49 vs. 3.84		47.0% vs. 45.7%
GDC-6036	NCT04449874 (40)	I	Dosages 50–400 mg	58	53.4%	14		13.1
			Dosages 400 mg	39	56.4%	11.9		13.7
D-1553	NCT05383898 (41)	I		79	40.5%	7.1	91.9%	8.2		38%
			Dosages 600 mg	62	38.7%	6.9	90.3%	7.6
	NCT05383898 (42)	II		123	50%	12.8	89%	7.6	Unachieved	50%
JDQ443	KontRASt-01 (NCT04699188) (43)	Ib/II	JDQ443 200 mg/bid	84	54.5%					7.1%
			JDQ443 + TNO155/tislelizumab	Active, not recruiting
JAB-21822	NCT05009329 (44)	IIb		119	47.9%					38.7%
IBI351	NCT05005234 (45,46)	Ia		166	45.5%	Unachieved	92.1%	9.6	Unachieved	36.7%
			Dosages 600 mg	126	46.8%		92.9%	9.6		33.3%
		II		116	49.1%	Unachieved		9.7	Immature	41.4%
Immunotherapy drugs
ICIs monotherapy	IMMUNOTARGET (47)			271	26%			3.2	13.5
pembrolizumab	KEYNOTE-042 (NCT02220894) (48)	III	Pembrolizumab vs. chemotherapy	TPS ≥50%	39% vs. 32%	10.8		7.1 vs. 6.4	20 vs. 12.2	18% vs. 41%
				TPS ≥20%	33% vs. 29%	8.3		6.2 vs. 6.6	17.7 vs. 13
				TPS ≥1%	27% vs. 27%	8.3		5.4 vs. 6.5	16.7 vs. 12.1
CT/D/T	POSEIDON trial NCT03164616 (49)	III	CT	337	33.4%	4.2		4.8	11.7	35.14%
			D + CT	338	48.5%	6.0		5.5	13.3	40.12%
			T + D + CT	338	46.3%	7.4		6.2	14	44.24%
Combined treatment
AMG510 and others	CodeBreak 101 NCT04185883) (50)	Ib	KRAS G12C-mutated mCRC	40	30%	5.3	92.5%	5.7	12.5	27%
JAB-21822 and AB-3312	NCT05288205 (51)	I/II		80	72.5%		96.3%			≥41.9%
MEK inhibitor and chemotherapy	NCT00890825 (52)	II	Selumetinib + docetaxel vs. placebo group	44 vs. 43	37% vs. 0%			5.3 vs. 2.1	9.4 vs. 5.2	82% vs. 67%
	NCT01933932 (53)	III		251 vs. 254	20.1% vs. 13.7%	2.9 vs. 4.5	42% vs. 37%	3.9 vs. 2.8	8.7 vs. 7.9	67% vs. 45%
MEK and PI3K inhibitors	NCT01363232 (54)	Ib		89	77.5%					60.7%

CT, chemotherapy; D, durvalumab; DCR, disease control rates; DOR, duration of remission; ICI, immune checkpoint inhibitor; KRAS, Kirsten rat sarcoma viral oncogene homolog; mCRC, metastatic colorectal cancer; mOS, median overall survival; mPFS, median progression-free survival; MEK, mitogen-activated protein kinase; ORR, objective remission rate; PI3K, phosphatidylinositol-4,5-bisphosphate 3-kinase; T, tremelimumab; TRAE, treatment-related adverse event.

A multinational, multicenter, randomized phase III trial (CodeBreak 200: NCT04303780) compared AMG510 to docetaxel in NSCLC patients with KRAS G12C mutations. AMG510 significantly improved mPFS (5.6 vs. 4.5 months) and ORR (28.1% vs. 13.2%) compared to docetaxel, with a better safety profile but no significant difference in mOS (35), The incidence of grade 3 or higher AEs was lower with AMG510 (33% vs. 40%), with diarrhea (12%), alanine aminotransferase (ALT) elevation (8%), and aspartate aminotransferase (AST) elevation (5%) being the most common side effects (34). A retrospective study reaffirmed AMG510’s significant efficacy in KRAS G12C-mutant NSCLC, though patients with KEAP1 co-mutations showed limited benefit (36). Ongoing phase II trials (NCT05398094, NCT04933695) continue to recruit patients with advanced KRAS G12C-mutant NSCLC (all clinical trials are presented in Table 3).

Table 3. Ongoing clinical studies (accessible at clinicaltrials.gov).

Drugs	Clinical trial	Phase	Title	Study status
AMG510	NCT05398094	II	Clinical Trial of AMG510 in Stage III Unresectable NSCLC KRAS p.G12C Patients and Ineligible for Chemo-radiotherapy (MERIT-lung)	Recruiting
	NCT04933695	II	A Study of AMG 510 in Participants With Stage IV NSCLC Whose Tumors Harbor a KRAS p.G12C Mutation in Need of First-line Treatment (CodeBreaK201)	Active, not recruiting
MRTX849	NCT05673187	II	Adagrasib in Patients With KRASG12C-mutant NSCLC Who Are Elderly or Have Poor Performance Status (ADEPPT)	Recruiting
GDC-6036	NCT06497556	III	A Study Evaluating the Efficacy and Safety of Divarasib Versus Sotorasib or Adagrasib in Participants With Previously Treated KRAS G12C-positive Advanced or Metastatic Non-Small Cell Lung Cancer (Krascendo 1)	Recruiting
D-1553	NCT05492045	Ib/II	A Study to Evaluate D-1553 in Combination Therapy in Non-Small Cell Lung Cancer	Terminated
	NCT06300177	III	D-1553 Tablet Versus Docetaxel Injection for KRAS G12C Mutation-positive Locally Advanced or Metastatic Non-small Cell Lung Cancer After Prior Standard Therapy Failure	Recruiting
JDQ443	KontRASt-02 (NCT05132075)	III	Study of JDQ443 in Comparison With Docetaxel in Participants With Locally Advanced or Metastatic KRAS G12C Mutant Non-small Cell Lung Cancer	Active, not recruiting
LY3537982	SUNRAY-01 (NCT06119581)	I	A Study of First-Line LY3537982 and Pembrolizumab With or Without Chemotherapy in Patients With Advanced KRAS G12C-Mutant Non-small Cell Lung Cancer	Recruiting
	NCT04956640	I/II	Study of LY3537982 in Cancer Patients With a Specific Genetic Mutation (KRAS G12C)	Recruiting
MK-1084	NCT05067283	I	A Study of MK-1084 in KRAS Mutant Advanced Solid Tumors (MK-1084-001)	Recruiting
RMC-6236	NCT05379985	I	Study of RMC-6236 in Patients With Advanced Solid Tumors Harboring Specific Mutations in RAS	Recruiting
D3S 001	NCT05410145	I/II	A Study of D3S 001 Monotherapy or Combination Therapy in Subjects With Advanced Solid Tumors With a KRAS p.G12C Mutation	Recruiting
BBO-8520	NCT06343402	I	Open-label Study of BBO-8520 in Adult Subjects with KRASG12C Non-small Cell Lung Cancer	Recruiting
BEBT-607 tablets	NCT06117371	I	Study of BEBT-607 Tablets in The Treatment of Advanced or Metastatic Solid Tumors With KRAS G12C Mutation	Recruiting
SHP2 inhibitors
TNO155	NCT03114319	I	Dose Finding Study of TNO155 in Adult Patients With Advanced Solid Tumors	Active, not recruiting
BBP-398	NCT04528836	I	First-in-Human Study of the SHP2 Inhibitor BBP-398 in Patients With Advanced Solid Tumors	Terminated
JAB-3068	NCT03518554	I	A First in Human, Dose Escalation Study of JAB-3068 (SHP2 Inhibitor) in Adult Patients With Advanced Solid Tumors	Completed
SOS1 inhibitors
BI-1701963	NCT04111458	I	A Study to Test Different Doses of BI 1701963 Alone and Combined With Trametinib in Patients With Different Types of Advanced Cancer (Solid Tumours With KRAS Mutation)	Active, not recruiting
Combined treatment
AMG510 and pembrolizumab	CodeBreak202 (NCT05920356)	III	A Study Evaluating Sotorasib Platinum Doublet Combination Versus Pembrolizumab Platinum Doublet Combination as a Front-Line Therapy in Participants With Stage IV or Advanced Stage IIIB/C Nonsquamous Non-Small Cell Lung Cancers	Recruiting
AMG510 and SBRT	NCT06127940		K-SAB Trial - Sotorasib Followed by SBRT to 1-3 Lesions in Advanced NSCLC With KRASG12C Mutation	Recruiting
MRTX849 and TNO155	KRYSTAL 2 (NCT04330664)	I/II	Adagrasib in Combination With TNO155 in Patients With Cancer (KRYSTAL 2)	Active, not recruiting
MRTX849 and Pembrolizumab	KRYSTAL-7 (NCT04613596)	II/III	Phase 2 Trial of Adagrasib Monotherapy and in Combination With Pembrolizumab and a Phase 3 Trial of Adagrasib in Combination in Patients With a KRAS G12C Mutation	Recruiting
MRTX849 and BI-1701963	KRYSTAL-14 (NCT04975256)	I/I b	MRTX849 in Combination With BI-1701963 in Patients With Cancer (KRYSTAL 14)	Terminated
MRTX849 and Palbociclib	KRYSTAL-16 (NCT05178888)	I/I b	Adagrasib in Combination With Palbociclib in Patients With Advanced Solid Tumors (KRYSTAL-16)	Active, not recruiting
ERAS-007/ERAS-601 and others	NCT04959981	I b	A Study of Anti-Cancer Therapies Targeting the MAPK Pathway in Patients With Advanced NSCLC (HERKULES-2)	Completed
GDC-6036 and others	NCT05789082	I b/II	A Study Evaluating the Safety, Activity, and Pharmacokinetics of Divarasib in Combination With Other Anti-Cancer Therapies in Participants With Previously Untreated Advanced or Metastatic Non-Small Cell Lung Cancer With a KRAS G12C Mutation	Recruiting
Durvalumab + tremelimumab + chemotherapy	NCT06008093	IIIb	A Study to Investigate the Efficacy of Durvalumab Plus Tremelimumab in Combination With Chemotherapy Compared With Pembrolizumab in Combination With Chemotherapy in Metastatic NSCLC Patients With Non-squamous Histology Who Have Mutations and/or Co-mutations in STK11, KEAP1, or KRAS (TRITON)	Recruiting

NSCLC, non-small cell lung cancer; SBRT, stereotactic body radiation therapy.

MRTX849

MRTX849 is the second FDA-approved KRAS G12C inhibitor, known for its irreversible and highly selective properties. Like AMG510, MRTX849 impairs cell viability and inhibits ERK phosphorylation, although it does not effect on AKT activation (55). MRTX849 also boasts a significant advantage in its half-life (23 vs. 5 hours) (38). Briere et al. found that MRTX849 reduced intratumoral myeloid-derived suppressor cells (MDSCs) and increased M1-polarized macrophages, dendritic cells, and CD4⁺ and CD8⁺ T cells (56).

In the phase I KRYSTAL-1 trial (NCT03785249), MRTX849 demonstrated clinical activity in KRAS G12C-mutant NSCLC patients previously treated with platinum-based CT and PD-1/PD-L1 inhibitors (37). A phase II single-arm trial (NCT03785249) enrolled 116 patients, with 112 having measurable disease at baseline. Of these, 48 patients (42.9%) showed objective responses, with a mPFS of 6.5 months (95% CI: 4.7–8.4) and a mOS of 12.6 months (95% CI: 9.2–19.2) (38). Among them, the ORRs confirmed in patients with STK11, KEAP1, TP53 and CDKN2A co-mutations were 40.5%, 28.6%, 51.4% and 58.3%, respectively. In terms of safety, 52 patients (44.8%) experienced grade 3 or higher AEs, with the most commonly diarrhea (70.7%), nausea (69.8%), fatigue (59.5%), and vomiting (56.9%). Two patients experienced fatal AEs (heart failure and pulmonary hemorrhage) (38).

The phase III validated KRYSTAL-12 trial (NCT04685135) compared MRTX849 to conventional CT (docetaxel) and showed significant improvement in mPFS (5.49 vs. 3.84 months, HR 0.58). The ORR was also higher in the MRTX849 group (31.9%, 95% CI: 26.7–37.5%) compared to docetaxel (9.2%, 95% CI: 5.1–15.0%) (39). Although grade 3 or higher AEs occurred at similar rates in both groups (47.0% vs. 45.7%), fewer patients in the MRTX849 group discontinued treatment due to AEs (7.7% vs. 14.3%) (39). Of note, Gadgeel et al. presented the findings at the World Conference on Lung Cancer (WCLC) suggesting that dose adjustments of MRTX849 did not significantly impact treatment efficacy, with a 2-year OS rate of 32.1% in the dose-adjusted group compared to 31.3% in the overall population (57).

Clinical studies of other directly targeted inhibitors

Although the successive approvals of AMG510 and MRTX849 have dispelled the notion that KRAS G12C mutations are ‘undruggable’, these agents are currently approved only for second-line treatment of metastatic disease. Furthermore, studies have shown that multiple mechanisms may contribute to resistance against KRAS G12C inhibitors (see ‘Resistance mechanisms’ for details). This underscores the urgent need for the development of additional novel therapies. Targeted inhibitors currently in clinical development include GDC-6036, D-1553, JDQ443 and LY3537982, MK-1084, JAB-21822, IBI351 etc. These inhibitors aim to irreversibly lock KRAS in an inactive state, offering potential for more potent and selective targeting of KRAS G12C-mutant tumors.

GDC-6036 (divarasib)

GDC-6036 binds covalently to cysteine residues in the KRAS protein, irreversibly locking it in an inactive state, effectively shutting down its oncogenic signaling. In vitro studies have shown that GDC-6036 is 5 to 20 times more potent and 50 times more selective than AMG510 and MRTX849 (40). A phase I study (NCT04449874) evaluating GDC-6036 in 137 patients with advanced or metastatic solid tumors, including 60 with KRAS G12C-mutant NSCLC, demonstrated a confirmed response rate of 53.4%, a mPFS of 13.1 months, and a median duration of remission (DOR) of 14.0 months (40).

D-1553

D-1553 is one of the earliest KRAS G12C inhibitors showing promising clinical efficacy. In a phase I dose-escalation and dose-expansion study (NCT05383898), 79 patients with KRAS G12C-mutant NSCLC were enrolled. The study reported an ORR rate of 40.5%, a disease control rate (DCR) of 91.9%, a mPFS of 8.2 months, and a DOR of 7.1 months (41). Among 62 patients who received the recommended phase II dose (600 mg), 24 patients (ORR, 38.7%) experienced partial remission and 32 patients (DCR, 90.3%) were stable with a PFS of 7.6 months (41). Grade 3 or higher AEs were reported in 30 patients (38.0%), the most notably hepatic abnormalities and gastrointestinal events. Encouragingly, early clinical efficacy of D-1553 monotherapy was also observed in NSCLC patients with brain metastases, with an intracranial ORR of 17% and an intracranial DCR of 100% (41). A subsequent phase II study of D-1553 in Asian NSCLC patients treated 123 individuals, with 50% achieving objective remission and an mPFS of 7.6 months, though half of the patients experienced grade 3 or higher AEs, again primarily hepatic and gastrointestinal (42). D-1553 was finally officially approved for marketing by the National Medical Products Administration (NMPA) on 8 November 2024 for the treatment of adult patients with advanced NSCLC with KRAS G12C mutation who have received at least one systemic therapy.

JDQ443

JDQ443 binds to KRAS G12C in a novel manner beneath the switch-II loop, potentially reducing the risk of resistance due to mutations in the RAS gene (58). The KontRASt-01 (NCT04699188) clinical trial has demonstrated the efficacy of JDQ443 as a single-agent treatment for advanced solid tumors with KRAS G12C mutations. This study reported an ORR of 54.5%, with only 7.1% of grade 3 or higher treatment-related AEs (43). Additionally, combination studies involving JDQ443 with TNO155 [a Src homology phosphatase 2 (SHP2) inhibitor] or tislelizumab (an anti-PD-1 monoclonal antibody) are underway, with enrollment now closed and results pending.

JAB-21822

JAB-21822 is another novel KRAS G12C inhibitor, which formally filed a New Drug Marketing Application with the NMPA for the treatment of patients with advanced or metastatic NSCLC bearing the KRAS G12C mutation in second-line and above therapy. A multicenter, single-arm phase IIb study (NCT05009329) is investigating the efficacy of JAB-21822 in 119 KRAS G12C-mutant NSCLC patients. Early results show an ORR of 47.9%, as assessed by Independent Review Committee (IRC) (44). Additionally, a clinical trial (NCT05288205) is exploring the combination of JAB-21822 with JAB-3312, an SHP2 inhibitor, for the treatment of KRAS G12C-mutant solid tumors (51).

IBI351

Preclinical studies suggest that IBI351 covalently and irreversibly modifies cysteine residues in KRAS G12C, locking the protein in its inactive GDP-bound state, which induces apoptosis and cell cycle arrest in tumor cells (59). In a phase I study (NCT05005234), IBI351 demonstrated promising efficacy with an ORR of 46.8%, a DCR of 92.9%, and a treatment-related AE rate of 33.3% in patients receiving the recommended phase II dose (45). A subsequent phase II study reported an updated ORR of 49.1% and mPFS of 9.7 months for IBI135 treatment, which is the longest PFS benefit of any currently marketed KRAS G12C inhibitor (46), and showed superior anti-tumor activity compared to AMG510 and MRTX849. In August 2024, NMPA formally approved IBI351 for adult patients with advanced NSCLC with the KRAS G12C mutation who have received at least one systemic therapy (60). This is the first KRAS G12C inhibitor approved in China, which fills the gap of KRAS G12C mutation targeted therapy in China, and provide a new treatment option for patients with KRAS G12C-mutant advanced NSCLC.

Programs of combined treatment

While KRAS G12C inhibitors such as AMG510 and MRTX849 have shown promising results in clinical trials and are already approved for clinical use, the emergence of acquired resistance poses a significant challenge to their long-term efficacy. Therefore, combination therapies are considered a critical strategy to enhance efficacy, reduce toxicity, and delay resistance.

Given the complexity of the KRAS signaling pathways, targeting protein-protein interactions between KRAS and its downstream effectors offers an attractive therapeutic approach. Inhibiting key upstream and downstream molecules involved in KRAS signaling could further disrupt its oncogenic functions. Clinical trials are currently exploring combinations of KRAS inhibitors with other targeted therapies, such as SHP2 inhibitors, immune checkpoint inhibitors (ICIs), and other signaling modulators, to improve patient outcomes.

Combined upstream molecular inhibitors

RTKs inhibitors

Reactivation of KRAS and its downstream signaling pathways via adaptive RTKs is one of the mechanisms of acquired resistance to KRAS G12C inhibitors. Therefore, dual combination of KRAS G12C inhibitors and RTKs inhibitors may be a promising therapeutic strategy (16). RTKs inhibitors work by increasing KRAS-GDP binding and inhibiting GDP/GTP exchange, thus increasing the therapeutic effect of KRAS G12C inhibitors. Lito et al. were the first time to demonstrate that co-targeting KRAS G12C with RTK inhibitors enhanced antiproliferative effects in various cell models (59). However, efficacy can vary depending on the specific RTK inhibitor used, underscoring the need for careful selection of RTK inhibitors based on the RTK type involved.

SHP2 inhibitor

SHP2 is a non-receptor protein tyrosine phosphatase that mediates the activation of the RAS-MAPK-ERK signaling pathway through interactions with multiple RTKs (61). SHP2 has been identified as a critical player in the oncogenic process of KRAS-mutant NSCLC (62). Preclinical studies have shown that SHP2 inhibitors can reduce KRAS-GTP activation and enhance KRAS-GDP binding, helping to overcome resistance to KRAS G12C inhibitors (63). For instance, SHP-099, the first reported SHP2 inhibitor, demonstrated efficacy in a mouse tumor xenograft model and enhanced MAPK signaling blockade when combined with KRAS G12C inhibitor (64). However, concerns regarding hepatotoxicity and cardiovascular risks have been raised (65). Several SHP2 inhibitors, including TNO155 (NCT03114319), JAB-3068 (NCT03518554), and BBP-398 (NCT04528836), are currently in clinical trials for various cancers, with combination studies exploring their use with KRAS G12C inhibitors (NCT04185883, NCT04330664).

SOS1 inhibitor

SOS1 plays a key role in KRAS activation. Phosphorylation of SOS1 by the MAPK pathway can lead to uncoupling of the Grb2-SOS1 complex, promoting further RAS pathway activation (8,66). BI-3406 is a potent and selective SOS1 inhibitor that reduces KRAS-GTP formation both in vitro and in vivo, and alleviates acquired resistance to MEK inhibitors (67). The combination of BI-3406 with KRAS G12C inhibitors and MEK inhibitors holds promise for enhancing treatment efficacy in KRAS-mutant cancers. A phase I clinical trial (NCT04111458), is evaluating the safety and efficacy of BI-1701963, a novel SOS1 inhibitor, in combination with the MEK inhibitor trametinib.

Combined downstream molecular inhibitors

KRAS signaling involves multiple downstream effectors, including RAF, MEK, PI3K, and mTOR, all of which are critical for cell survival and proliferation. Preclinical studies emphasize that simultaneous inhibition of multiple downstream effectors is necessary to effectively block KRAS-mediated signaling (68). Monotherapy with multikinase inhibitors like sorafenib, MEK inhibitors like selumetinib and trametinib, or mTOR inhibitors such as ridaforolimus, has shown limited efficacy, as indicated by lower ORRs and modest PFS improvements (69-72).

In the case of selumetinib, a combination with the chemotherapeutic agent docetaxel showed an ORR of 37%, mPFS of 5.3 months, and mOS of 9.4 months in a randomized type II clinical trial (NCT00890825), significantly improving outcomes compared to the placebo group (ORR of 0%, an mOS of 5.2 months, mPFS of 2.1 months) (52). However, this benefit came with a higher incidence of AEs (82% vs. 67%), and the results were not confirmed in a subsequent phase III trial (NCT01933932) (53). Further, a mouse model of KRAS-mutant lung adenocarcinoma demonstrated that dual inhibition of PI3K and mTOR was more effective than single-agent therapy (73). A phase Ib/II trial (NCT01363232) is currently exploring the combination of BKM120 (a PI3K inhibitor) and MEK162 (a MEK1/2 inhibitor) in patients with solid tumors, including NSCLC (54).

Combined immunotherapy

KRAS mutations are associated with a highly immunogenic and inflammatory microenvironment in NSCLC tumors, and several studies have consistently shown that patients with advanced KRAS-mutated NSCLC benefit more from ICIs than from CT alone (74,75). An exploratory analysis (KEYNOTE-042) found that pembrolizumab monotherapy in patients with locally advanced or metastatic NSCLC had higher OS and a lower rate of AEs than CT (18% vs. 41%), without an advantage in ORR or mPFS (48). Therefore, ICIs-based treatment strategies are consistently recommended for patients with advanced KRAS-mutant NSCLC. And it has been shown that first-line immunotherapy is more effective in patients with KRAS G12C mutations than in patients with other KRAS mutations (76).

The mechanisms of KRAS mutations during the immune response are intricate. Upregulation of intercellular adhesion molecule 1 (ICAM1) by KRAS activation promotes pro-inflammatory M1 macrophage recruitment. However, co-activation with MYC can increase anti-inflammatory M2 macrophage recruitment by releasing CCL9 and IL-23 (77). Notably, KRAS-mutant tumors may also express higher levels of PD-L1 compared to non-KRAS-mutant tumors, suggesting a potential benefit from ICIs (78), which is the reason why KRAS mutant tumors respond better to immunotherapy. By contrast, Ricciuti et al. found that the large amount of native genomic data revealed that survival of NSCLC patients with STK11 or KEPA1 co-mutated were significantly shorter than KRAS single mutant (2.0/1.8 vs. 4.8/4.6 months, 6.2/4.8 vs. 17.3/18.4 months) (79,80), which may be due to the lack of PD-L1 expression caused by STK11 and KEAP1 mutations (81). Therefore, polyimmune combinations or immune-combined targeted inhibitors are highly warranted.

It is encouraging that dual immune checkpoint inhibition—using the PD-L1 inhibitor durvalumab and the CTLA-4 inhibitor tremelimumab—has shown efficacy in overcoming resistance to PD-(L)1 blockade, as demonstrated in the recently published phase III POSEIDON trial (NCT03164616) (49). In this trial, previously untreated patients with stage IV NSCLC were randomized into three groups: (I) standard CT; (II) durvalumab plus CT (DCT); and (III) durvalumab and tremelimumab plus CT (TDCT). The TDCT group showed a significant improvement in mOS [11.7 vs. 13.3 vs. 14 months; hazard ratio (HR) =0.77, P=0.003] and mPFS (4.8 vs. 5.5 vs. 6.2 months; HR =0.72, P<0.001), compared to the CT and DCT groups.

Moreover, after a median follow-up of 63.4 months, the 5-year OS rate in the TDCT group was more than double that of the CT group (15.7% vs. 6.8%; HR =0.76). This survival benefit was particularly pronounced in patients with non-squamous NSCLC. These findings suggest that TDCT has the potential to become a first-line treatment strategy for this subgroup. Additionally, the trial highlighted the poorer prognosis of patients with STK11 and KEAP1 mutations—a finding consistent with other studies (82). However, TDCT therapy was found to significantly extend long-term clinical benefit in patients with mutations in STK11, KEAP1, and KRAS, regardless of mutation subtype (83). The upcoming phase IIIB TRITON trial (NCT06008093), which plans to enroll approximately 280 patients with non-squamous NSCLC harboring STK11, KEAP1, or KRAS mutations/co-mutations, will further evaluate the efficacy of this combination therapy. We eagerly await its findings.

Canon et al. found stronger anti-tumor efficacy of AMG510 in combination with PD-1 inhibitors than monotherapy of them through the CT-26 KRAS-G12C series of models (30), and the synergistic effect of RAF/MEK inhibitors in combination with immune drugs has long been demonstrated in a mouse model of BRAF-mutant melanoma (84). Several clinical trials are also underway to evaluate the efficacy and safety of AMG510, MRTX849, SHP2 and SOS1 inhibitors, and downstream signaling pathway-related inhibitors in combination with anti-PD-1/PD-L1 drugs.

Nevertheless, the combination therapy may bring more side effects (85). For example, Chour et al. found that patients treated with PD-L1 inhibitors prior to AMG510 therapy showed a higher incidence of severe hepatotoxicity (33% vs. 11%, P=0.006) than patients not treated with PD-L1 inhibitors (86). Undeniably, there are limitations in current studies investigating combination therapies, largely due to the incomplete understanding of the interactions between KRAS G12C mutations and the tumor microenvironment. Additionally, the limited number of approved drugs further constrains treatment strategies. As a result, further research is needed to identify the optimal combination regimens and dosing strategies, as well as to thoroughly evaluate their safety and therapeutic efficacy. These areas represent critical opportunities for future breakthroughs.

Combined Chinese medicine ingredients

The use of traditional Chinese medicines (TCMs) is gaining attention as an adjunct to cancer treatment, particularly for KRAS-mutant tumors. Active compounds from TCM herbs have been shown to regulate critical signaling pathways such as Ras-Raf-MEK-ERK, and modulate processes like apoptosis, autophagy, and the cell cycle in tumor cells (87).

For example, andrographolide (AP), derived from the herb Andrographis paniculata, has been shown to inhibit tumorigenesis and metastasis by targeting EGFR, AKT, and mTOR pathways (88-90). In a KRAS-mutant colorectal cancer (CRC) cell model, AP was found to enhance the effect of cetuximab on KRAS-mutant CRC cells by targeting PDGFRβ and EGFR, decreasing PI3K and AKT expression, and reducing epithelial-mesenchymal transition (EMT). This combination significantly reduced tumor growth and lung metastases in KRAS-mutant CRC models. Another TCM-derived compound, β-elemene, induces ferroptosis and inhibits EMT, further suppressing KRAS-mutant tumor cell migration and metastasis when combined with cetuximab (91).

Additionally, TCM compounds such as Shengmai-water decoction (SM-WD) have demonstrated anti-tumor effects by modulating oxidative stress and inhibiting Ras/MAPK pathway overactivation, suggesting their potential as adjuvant therapies for KRAS-mutant cancers (92).

Overall, the integration of KRAS G12C inhibitors with TCM compounds or complexes offers a novel and promising strategy for the treatment of KRAS-mutant tumors.

Resistance mechanisms

Although promising preliminary clinical results have been obtained with KRAS G12C inhibitors, tumorigenesis is an intricate, multistep process involving various oncogenes and tumor suppressor genes. Resistance to these inhibitors can arise through several mechanisms, including primary resistance and acquired resistance.

Primary resistance

Primary resistance is the underlying cause of malignant tumors that do not respond to inhibitors such as AMG510, and researchers analysing the 2-year data from the trial CodeBreaK 100 found that early disease progression (PFS <3 months) occurred in approximately 36% of patients treated with AMG510 (93). Primary resistance may be associated with co-mutations in KEAP1, SMARCA4, and CDKN2A, and several large-sample analyses have shown KEAP1 mutations to be a marker of poorer prognosis in NSCLC patients treated with AMG510/MRTX849 (7,94). The specific mechanisms of resistance mediated by these gene co-mutations may need to be further explored by investigators across tumor types. Looking forward, integrating gene co-mutation types could help predict different types of resistance to KRAS inhibition and provide effective strategies for clinical treatment selection.

Acquired resistance through genetic mechanisms

Awad et al. observed that among patients who initially respond to KRAS G12C inhibitors, most develop resistance within 4–6 months (95). Notably, nearly all resistance mutations identified involve mechanisms that reactivate RAS-MAPK signaling. Within the MAPK pathway, MRTX849 appears to be associated with a higher frequency of resistance mutations compared to AMG510. These include secondary mutations in the KRAS gene itself—such as G12X, G13X, and Q61H—as well as alterations in the Switch II pocket (e.g., R68, H95, or Y96) (see Table 4) (95-97).

Table 4. Frequency of acquired resistance mutations.

Clinical & resistance-related metrics	Koga T et al. (96)	Zhao Y et al. (97)	Awad MM et al. (95)
Drug	AMG510/ MRTX849	AMG510	MRTX849
Tumor type	Ba/F3 clones resistant to either AMG510/MRTX849	NSCLC dominant	NSCLC dominant
Number of patients evaluated	142 (68/74)	43	38
Number of patients with any resistance alteration	124 (52/72)	27	17
Secondary KRAS mutations	87.3% (76.5%/97.3%)	14.8%	58.8%
KRAS G12C gene amplification		11.1%	11.8%
Activating mutations in NRAS/HRAS		11.1%	5.9%
Upstream RTK (amplification/fusion/mutation)		25.9%	35.3%
Downstream effect genes (amplification/fusion/mutation)		22.2%	41.1%

HRAS, Harvey rat sarcoma viral oncogene homolog; KRAS, Kirsten rat sarcoma viral oncogene homolog; NRAS, neuroblastoma rat sarcoma viral oncogene homolog; NSCLC, non-small cell lung cancer; RTK, receptor tyrosine kinase.

Similar to other oncogenic drivers in NSCLC, Secondary mutations in KRAS are a well-established cause of acquired resistance to KRAS G12C inhibitors. Using the KRAS G12C series Ba/F3 cell model, Koga et al. found that approximately 76.5% of cell models developed secondary KRAS mutations that were resistant to AMG510, with the most common secondary mutation being KRAS G13D (accounting for 23% of the KRAS secondary mutations), followed by R68M and A59S (21.2% each) (96). In contrast, secondary mutations occurred in up to 97.3% of cell models treated with MRTX849, with the most common secondary mutation being KRAS Q99L (52.8% of KRAS secondary mutations), followed by Y96D and R68S (15.3% and 13.9%, respectively). These findings demonstrate significant differences in secondary mutations between AMG510 and MRTX849, which suggests that switching between the two inhibitors might help overcome resistance in many cases (96). However, mutations shared by both drugs, such as A59S and Y96D, will likely require additional strategies. Potential approaches include the use of SOS1 inhibitors like BI-3406 (67) or SHP2 inhibitors such as TNO155 (65).

Adaptive resistance

Notably, a large proportion of patients do not have identifiable genetic alterations to explain primary or acquired resistance, suggesting that non-genetic mechanisms are another cause of resistance. This mechanism of species resistance is associated with feedback reactivation of bypass signaling pathways.

When KRAS is mutated, the cellular feedback inhibitory mechanism usually restricts the activity of upstream RTKs. The inhibition of the MAPK pathway by KRAS inhibitor treatment leads to the disappearance of this feedback inhibition, which results in the up-regulation of RTKs and the shift of KRAS to an active state mediated by SHP2 and wild-type RAS (NRAS or HRAS), etc., thus decreasing the sensitivity of KRAS inhibitors. Studies have shown that the RAS-MAPK pathway can be reactivated following AMG510 treatment, driven by the activation of wild-type RAS, which is unaffected by KRAS G12C-specific inhibitors. This reactivation can be counteracted by combining KRAS G12C inhibitors with SHP2 inhibitors, effectively inhibiting this bypass signaling (64).

Furthermore, research by Atish Mohanty and colleagues found that acquired resistance to AMG510 in homozygous cells was associated with increased expression of integrin β4 (ITGB4) and β-catenin. This overexpression can activate AKT-mTOR signaling pathways, contributing to resistance. Interestingly, this bypass mechanism does not confer resistance to MRTX849. For tumors overexpressing ITGB4, combining AMG510 with a proteasome inhibitor like carfilzomib (CFZ) may help reverse resistance by downregulating ITGB4 and β-catenin expression (98).

Histological transformation

Histologic transformation, particularly to small cell lung cancer or squamous cell carcinoma, represents another mechanism of acquired resistance to KRAS G12C inhibitors in NSCLC (99). This phenomenon has been well-documented in various cancer therapies, including those targeting KRAS. Adachi et al. found that EMT was linked to resistance against AMG510, which can be counteracted by combining PI3K inhibitors and SHP2 inhibitors (100). Similarly, Suzuki et al. also found that MET amplification in AMG510-resistant cells to activation of both the RAF-MEK-ERK pathway and AKT-mTOR pathway contributing to resistance (101). It has also been shown that feedback reactivation of the RAS-MAPK pathway after MEK inhibition varies depending on the epithelial or mesenchymal status of the tumor, and the combination of FGFR1 (an RTK) and MAPK inhibitors may be an effective strategy for the future treatment of mesenchymal-like KRAS-mutant NSCLC (102). Thus, assessing the EMT status of the tumor could help to provide the best treatment options for patients. In addition, two NSCLC patients treated with MRTX849 showed histological transformation from adenocarcinoma to SCC (95), Tong et al. also found that MRTX849 promotes squamous cell transformation after long-term treatment (103), further confirming that such transformations can be a significant mechanism of resistance to specific KRAS-targeted therapies.

Conclusions

KRAS G12C mutation is the most common subtype in NSCLC, initially considered undruggable. With the successive approval of several KRAS G12C inhibitors, including AMG510, MRTX849, IBI351 and D-1553, the victory of KRAS mutation-targeted therapies has finally arrived. These inhibitors, along with ongoing development of other KRAS-targeted therapies, are expanding our understanding of KRAS-driven tumor biology. As larger clinical cohorts are evaluated, efforts to address KRAS G12C inhibitor resistance are intensifying. The recent good news from the POSEIDON trial is even more encouraging, marking the potential for durvalumab in combination with tremelimumab and CT to become a first-line treatment strategy for advanced NSCLC, which includes KRAS-mutant tumor types with a poor prognosis. If the TRITON study can further validate this finding, then we believe it will be another major breakthrough in the treatment of advanced NSCLC.

Nevertheless, due to the complexity and heterogeneity of tumor responses, significant limitations remain in the current understanding of KRAS G12C-mutant NSCLC. Chief among these is the development of drug resistance. The reported frequency of resistance mechanisms may be subject to bias, as most existing studies are based on small sample sizes. Moreover, emerging evidence suggests that multiple resistance pathways often coexist within the same tumor, further complicating therapeutic outcomes.

Currently, KRAS G12C inhibitors are approved only as second-line treatments. Prior therapies—such as immunotherapy or radiotherapy—may alter the internal tumor microenvironment, potentially influencing resistance patterns; however, there is a lack of robust evidence-based data to clarify these effects. Thus, further investigation into the intrinsic mechanisms of resistance and the strategic optimization of combination therapies remains an essential focus for future research.

Supplementary

The article’s supplementary files as

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.