KRAS^G12C is the most common mutation in KRAS-mutant NSCLC. Two covalent KRAS^G12C inhibitors, sotorasib and adagrasib, are currently highly anticipated.

Sotorasib was developed by Amgen. It received accelerated approval from the United States FDA on 28 May 2021 for treating patients with KRAS^G12C-mutant NSCLC (Blair, 2021). This irreversible inhibitor locks KRAS^G12C in the inactive GDP-bound state and binds to a special groove created by histidine 95 (Gentile et al., 2017). The recommended dosage is 960 mg orally once a day and its half-life is 5.5 h (Nagasaka et al., 2020). In a multicenter CodeBreaK 100 clinical trial (NCT03600883), the results showed that among the 124 patients with NSCLC who had previously received chemotherapy and/or immunotherapy, the objective response rate (ORR) reached 36% and 81% of patients achieved tumor control. The median progression-free survival (PFS) was 6.8 months and the median overall survival (OS) was 12.5 months (Li B. et al., 2021). Thus, sotorasib successfully fills the gap in the clinical treatment of KRAS^G12C NSCLC. The most common adverse effects are diarrhea, musculoskeletal pain, nausea, fatigue, and cough (Li B. et al., 2021). In addition, some randomized, open-label, multicenter trials on sotorasib are underway or recruiting patients, including NCT04303780, NCT04933695, NCT04625647, NCT05054725, NCT04185883, NCT04380753, and NCT04667234 (Blair, 2021).

Adagrasib was developed by Mirati Therapeutics. It binds to cysteine 12 within the Switch II pocket and also irreversibly locks KRAS^G12C in the inactive GDP-bound state. The recommended dosage is 600 mg orally twice a day and its half-life is 24.7 h (Hellmann et al., 2019). In an ongoing, multicenter, phase I/II trial (NCT03785249), 3/4 responders had NSCLC. Following adagrasib treatment, no patient exhibited brain metastasis. Adagrasib had better safety and the most common adverse events are grade 1 or 2 diarrhea, and nausea (Janne et al., 2019). In addition, certain randomized, open-label, multicenter trials are underway or recruiting patients, including NCT04613596, NCT04685135, NCT04330664, and NCT04975256. The basic features of sotorasib and adagrasib are shown in Table 1. Due to the small number of participants in clinical trials, the advantages of these two inhibitors are not comparable at present.

In addition, a number of new KRAS^G12C inhibitors are under development. Patients with advanced KRAS ^G12C-positive solid tumors were selected for the phase I trial for inhibitor JNJ-74699157 (NCT04006301). That trial was completed with promising results. Furthermore, LY3499446 and LY3537982, developed by Eli Lilly, have been included in NCT04165031 and NCT04956640, respectively (Nagasaka et al., 2020). Clinical trials on GDC-6036 and D-1553 are at the recruitment stage (NCT04449874 and NCT04585035) (Addeo et al., 2021).

However, acquired resistance to KRAS^G12C inhibitors still occurs. Resistance mechanisms have been previously described (Zhang et al., 2022). First and foremost, KRAS on-target mutations, including R68S, H95D/Q/R, G12D/R/V/W, G13D, Q61H and Y96C, disrupt the drug-binding interface (Awad et al., 2021). Next, KRAS amplification along with secondary mutations of KRAS^G12C (G12D/R/V/W) induce drug resistance (Skoulidis et al., 2021). Moreover, reactivating upstream and downstream pathways has been confirmed to be the reason for that drug resistance (Hellmann et al., 2019; Xue et al., 2020). Meanwhile, histological transformation, such as from adenocarcinoma to squamous cell carcinoma, is a non-genetic mechanism (Adachi et al., 2020). Finally, activating the dysregulation of cell cycle and focal adhesion kinase (FAK) signaling is a potential resistance mechanism (Tang et al., 2016; Hellmann et al., 2019). We therefore hope that an increasing number of promising drugs that can battle acquired resistance will be identified in the future.

The activation of KRAS upstream pathway signals may lead to the activation of KRAS mutations, which may cause continuous stimulation of its downstream signaling pathways and promote tumorigenesis (Patricelli et al., 2016). Thus, suppressing the upstream and downstream effectors of the KRAS pathways, including EGFR, rapidly accelerated fibrosarcoma (RAF), MAPK/ERK kinase (MEK), extracellular regulated protein kinase (ERK), phosphatidylinositol 3 kinase (PI3K), mammalian target of rapamycin (mTOR) and FAK, may be an effective therapeutic method. Certain single-agent inhibitors which have been studied in clinical trials are listed as follows.

Anti-programmed cell death protein 1 (anti-PD-1) inhibitor, anti-programmed cell death-ligand protein 1 (anti-PD-L1) inhibitor and anti-cytotoxic T-lymphocyte associated protein 4 (anti-CTLA-4) inhibitor are common immune checkpoint inhibitors. In a randomized, open-label, phase III trial CheckMate057 (NCT01673867), patients were administered either an anti-PD-1 antibody (nivolumab; 3 mg/kg every 2 weeks) or docetaxel monotherapy (75 mg/m² every 3 weeks). The results indicated that patients receiving nivolumab monotherapy had a better OS (12.2 vs. 9.4 months), while they experienced minor adverse events, including fatigue, nausea, decreased appetite, and asthenia (Borghaei et al., 2015). Moreover, in a randomized, open-label, multi-center, phase III trial OAK (NCT02008227), patients were administered either an anti-PD-L1 antibody (atezolizumab, 1,200 mg every 3 weeks intravenously) or docetaxel (75 mg/m² every 3 weeks intravenously), respectively. Atezolizumab monotherapy also led to a longer median OS (17.2 vs. 10.5 months). Adverse events following atezolizumab monotherapy were fewer than following docetaxel monotherapy (Rittmeyer et al., 2017). Thus, anti-PD-1/L1 antibodies are effective in treating KRAS-mutant NSCLC. In addition, there are ongoing studies on anti-CTLA-4 inhibitors (NCT02477826 and NCT03215706) (Hellmann et al., 2019; Paz-Ares et al., 2021). Studies have also suggested that combination treatment between covalent KRAS inhibitors and immune checkpoint inhibitors is worthy of investigation (Canon et al., 2019; Hallin et al., 2020). Of note, a retrospective study reported that a KRAS^G12D mutation was a negative prognostic factor while a KRAS^G12C mutation was a positive prognostic factor following immunotherapy (Aredo et al., 2019). Besides, since high expressions of PD-L1, CD8⁺ T cells, and TMB are associated with KRAS/P53 mutations, patients with such simultaneous mutations can benefit the most from PD-1 inhibitors (Dong et al., 2017). By contrast, LKB1-loss is associated with a low PD-L1 expression, as well as a lack of infiltrating CD8⁺ TILs, resulting in innate resistance to PD-1 inhibitors of patients with KRAS-mutant NSCLC (Skoulidis et al., 2015; Listì et al., 2018). Therefore, individualized immunotherapy will be needed in the future, due to the heterogeneity of KRAS mutations.

In addition to common immunotherapy, certain immune-related emerging therapeutics, such as cancer vaccines and adoptive cell therapy, are being investigated in clinical studies.

The role of cancer vaccines is to motivate the immune system to recognize different antigen markers mainly expressed on tumor cells and to lyse these tumor cells (Even-Desrumeaux et al., 2011). Nowadays, various types of cancer vaccines exist, including in microparticle, nanoparticle and liposome formulations (Neek et al., 2019). By the end of 2017, Moderna had launched the candidate vaccine mRNA-5671/V941, encoding the G12C, G12D, G12V, and G13D KRAS-mutant antigens. Following an intramuscular injection, the mRNA nanoparticles are taken in by antigen-presenting cells, translated and presented on the cell surface, thereby causing T cells to respond to the new epitope of KRAS (Moore et al., 2020). A phase I, open-label, non-randomized, multi-center clinical trial on mRNA-5671/V941 as a monotherapy or combined with pembrolizumab is ongoing (NCT03948763). This study is due to be completed in 2025. However, Merck recently announced that it has abandoned its collaboration with Moderna to develop a cancer mRNA vaccine targeting KRAS mutation. The phase I clinical trial about mRNA-5671/V941 has also been discontinued. Merck noted that future partnerships with Moderna remain focused on oncology, with the two companies continuing to work together to develop personalized cancer vaccines such as mRNA-4157 as well as cancer vaccines that encode the four most common KRAS mutations. In addition to mRNA vaccines, cancer vaccines also include polypeptide vaccines and dendritic cell–based vaccines. The polypeptide vaccine is a kind of vaccine prepared by chemical synthesis technology, according to the amino acid sequence of some known or predicted epitope of the pathogen antigen gene. If the specific antigen in tumor cells can be identified and the vaccine can be developed based on the amino acid sequence, the relevant immune cells can be activated to kill tumor cells with the same specific antigen. At present, the polypeptide vaccine against KRAS-mutant NSCLC is still in the preclinical stage. The dendritic cell–based vaccine induces the generation of dendritic cells (DCs) by using the patient’s autologous monocytes in vitro, and then loads the corresponding tumor antigens to make DCs loaded with tumor antigens. Next, these DCs are injected into the body to stimulate the proliferation of tumor-killing lymphocytes in the body. These DCs play important roles in long-term tumor surveillance and tumor-killing to achieve the purpose of eliminating tumors. Dendritic cell-based vaccines have actually made a number of major breakthroughs in animal studies and early clinical trials, including treating KRAS-mutant pancreatic cancer, colorectal cancer and triple negative breast cancer (NCT03329248, NCT03387098, NCT03586869, NCT03136406, NCT03563157, NCT03387085). However, few clinical trials about dendritic cell-based vaccines against KRAS-mutant NSCLC has been conducted. Thus, only by finding out specific antigens in different patients with different tumors and developing relevant vaccines, can we activate the immune system of patients with specific targets and inhibit the growth of tumor cells.

Another strategy being studied is the injection of autologous T cells, which transduce the mutated KRAS-specific T cell receptors into the patients. This method is known as adoptive cell therapy (Salgia et al., 2021). Changhai Hospital is conducting a clinical trial on the activity and safety of mutant KRAS^G12V-specific T cell receptor transduced T cell therapy for patients with pancreatic cancer. The mechanism of adoptive cell therapy is the targeting of human leukocyte antigen-matched mutant KRAS tumor cells instead of normal cells. The primary outcome measures are side effects and ORR. This study is estimated to be finished in 2023 (NCT04146298). This novel strategy has also been tested for the treatment of gastrointestinal, gastric, colon, and rectal cancers (NCT03745326). Treating KRAS-mutant NSCLC using this method should be explored in the future.

The MAPK signaling pathway can transduce extracellular signals into cells, and transmit cell signals through a tertiary kinase cascade, thereby regulating cell proliferation, differentiation, apoptosis, inflammation and exerting many other important pathological effects. The MAPK pathway has four main routes, with the RAS/RAF/MEK/ERK pathway being the most studied (Wei and Hui, 2002).

Fisetin is a natural flavonoid found in many fruits and vegetables, which has antioxidant, anti-cancer and neuroprotective effects. Dr. Liao’s team first showed that fisetin could inhibit KRAS^G12S-mutant cancer cell A549 invasion, migration and metastasis by suppressing ERK1/2 phosphorylation, as well as the binding abilities of nuclear factor kappa B (NF-κB) and activator protein-1 (AP-1). Fisetin has also been shown to reduce the expression levels of NF-κB, c-Fos, c-Jun, matrix metalloproteinase-2 (MMP-2) and urokinase-type plasminogen activator (u-PA) (Liao et al., 2009). PJ-1 and PJ-9 are flavonoid glycosides purified from Pulicaria jaubertii, which inhibit A549 proliferation and metastasis by reducing the expression of mutant K-Ras/B-Raf proteins, transforming growth factor-beta (TGF-β) and interleukin 8 (IL-8). Meanwhile, the gene expression of tumor suppressor p53 was rescued (Abd El Maksoud et al., 2019). Additionally, curcumin is a natural phenolic compound extracted from Curcuma longa with anti-inflammatory, antioxidant, anti-proliferation and anti-angiogenesis effects. It has been demonstrated that curcumin can inhibit the invasion and migration of A549 by blocking the MEKK and ERK signaling pathways, finally inducing cell apoptosis by decreasing MMP-2 and MMP-9 expression (Lin et al., 2009). Moreover, thymoquinone, which is extracted from Nigella sativa Linn seed oil, has been shown to exert anti-oxidant, anti-inflammatory, anti-cancer and hepatoprotective effects. In a previous study, it blocked the ERK1/2 signaling pathway to downregulate MMP-2 and MMP-9, inhibiting the proliferation, migration and invasion of A549 cells (Yang et al., 2015). Another natural compound, krukovine, a bisbenzylisoquinoline alkaloid isolated from Abuta grandifolia, exhibited potent cytotoxicity to A549 and KRAS^Q12H-mutant H460 cancer cells. It suppressed cell proliferation by inactivating the RAF-ERK and AKT pathways. It also significantly promoted cell cycle arrest at the G1 phase and induced apoptosis (Lai et al., 2018). Besides, nootkatone, derived from grapefruit and Alpiniae Oxyphyllae fructus, could activate the AMPK pathway, induce G1 cell arrest as well as inhibit the activation of AKT and ERK proteins. Studies have shown that either treatment with nootkatone alone or combination treatment with nootkatone and adriamycin (ADR) suppressed the progression of ADR-resistant A549/ADR lung cancer in vivo and in vitro (Le et al., 2019). Meanwhile, homoharringtonine is a cytotoxic alkaloid extracted from Cephalotaxus harringtonia. In vivo, Dr. Weng’s team used a Kras^G12D-expressing LL2 tumor-bearing mouse model and a Kras^G12C-driven spontaneously transgenic mouse model to demonstrate that homoharringtonine exhibited an obvious anti-cancer activity by downregulating IL-12, and upregulating CD80, CD86, and CD69 in B220 + B cells. In vitro, homoharringtonine suppressed A549 cell proliferation by decreasing the oncogenic KRAS, ERK, AKT, and STAT3 protein expression (Weng et al., 2018). Finally, phloretin, a polyphenolic compound found in apples and other Rosaceae plants, has been shown to induce apoptosis in A549 cells mainly by decreasing the expression of Bcl-2 and increasing that of the degraded form of PARP, cleaved caspase-3, cleaved caspase-9, and BAX. At the same time, phloretin upregulated ERK1/2, JNK1/2, and P38 MAPK phosphorylation. In vivo, phloretin played a significant inhibitory role in a nude mouse model (Min et al., 2015).

The PI3K/AKT/mTOR pathway is also regulated by KRAS mutations. The hyperactive PI3K/AKT/mTOR pathway reduces apoptosis and promotes the proliferation of tumor cells. Thus, certain natural compounds may exert inhibitory effects on KRAS-mutant cancer cells by blocking the PI3K/AKT/mTOR pathway.

Krukovine, nootkatone, and homoharringtonine inhibited KRAS-mutant cancer cells not only by blocking the MAPK pathway, but also by suppressing the PI3K/AKT/mTOR pathway, as discussed in the above section. In addition, fisetin could block PI3K/AKT/mTOR signaling in A549 and KRAS^G12C-mutant NSCLC H1792 cells in a dose-dependent manner (Khan et al., 2012). Betulinic acid is a natural pentacyclic triterpenoid compound extracted from Zizyphus mauritiana, which has anti-inflammatory, anti-malaria, anti-AIDS and anti-tumor activities. The results showed that combination treatment with betulinic acid and sorafenib inhibited the proliferation of A549, H358 (KRAS^G12Cmutation) and A427 (KRAS^G12Dmutation) cells by downregulating AKT and mTOR phosphorylation, decreasing Bcl-2 and Bcl-X_L expression, increasing Bak and Bax expression, inducing CHOP overexpression and activating the caspases and cleavage of PARP (Kutkowska et al., 2017). Besides, gallic acid is a type of natural polyhydroxyphenol compound found in plant derivatives, fruits and green tea, which has antitumor, antioxidant, antibacterial, anti-inflammatory and antibacterial activities. Gallic acid has been found to suppress EGFR phosphorylation to inhibit PI3K and AKT phosphorylation. It has also been shown to activate tumor suppressor p53 and decrease the expression of PD-L1 (Kang et al., 2020). In addition, honokiol is a biphenolic phytochemical derived from the magnolia tree, which readily crosses the blood brain barrier. It obviously induced apoptosis, G1 arrest and autophagy in H460, A549, and H358 cells by interrupting AMPK/mTOR and Sirt3/Hif-1α signaling (Luo et al., 2017). Finally, allicin exhibits a high antibacterial activity and inhibits the growth of multiple microorganisms. It was considered as an anti-invasive drug through changing the tissue inhibitor of metalloproteinase (TIMP)/matrix metalloproteinase (MMP) balance and inhibiting PI3K/AKT pathway (Huang et al., 2017).

Currently, immunotherapy is a popular cancer treatment option. In a recent preclinical study, KRAS mutations were closely associated with PD-L1 (Liu et al., 2020).

Except for gallic acid, which has previously been mentioned, luteolin and its derivative, apigenin, were found by our team to clearly suppress H358 and H460 cell proliferation, induce cell apoptosis and downregulate the IFN-γ-induced PD-L1 expression by inhibiting STAT3 phosphorylation. In vivo, three mouse models (nude mice, xenograft mice, and genetically engineered KRASLA2 mice) were used to test the inhibitory effect of luteolin and apigenin (Jiang et al., 2021). Additionally, ginsenoside Rg3, one of the active compounds of ginseng root, inhibits vascular endothelial growth factor expression and exhibits an antitumor activity. Dr. Jiang demonstrated that it attenuated cisplatin resistance by decreasing PD-L1 expression, which was associated with the inhibition of the NF-κB pathway in the treatment of KRAS-mutant NSCLC (Jiang et al., 2017). In addition, our research group recently found that tricin (C₁₇H₁₄O₇), a flavonoid constituent isolated from Rhizoma Phragmitis, rice and wheat, had the ability to selectively target H358 and H2122 cells (Li JX. et al., 2021). Our preliminary results exhibited that tricin decreased PD-L1 expression in H358 and H2122 cells. The relationship between tricin and the PD-1/PD-L1 axis will be explored further by our team in the future.

In a word, naturally extracted compounds have been shown to play important roles in treating KRAS-mutant NSCLC in preclinical studies. We hope that an increasing number of compounds will enter clinical trials and benefit patients.

One interesting combination treatment is that of KRAS^G12C inhibitors and anti-PD-1/L1 inhibitors. Dr. Canon found that, in a CT-26 KRAS ^G12C mouse model, monotherapy with covalent KRAS^G12C inhibitor sotorasib only caused the regression of 1/10 tumors, and so did monotherapy with anti-PD-1 inhibitor. However, combining sotorasib with anti-PD-1 antibody promoted cancer regression in 9/10 mice and extended survival (Canon et al., 2019). Thus, clinical trials evaluating the efficacy and safety of this combination treatment are ongoing, and the results are highly anticipated by patients. Considerable research attention has been paid on evaluating the combined effect of KRAS^G12C inhibitors such as sotorasib, adagrasib, GDC-6036 or JDQ443, and anti-PD-1/L1 inhibitors such as pembrolizumab, atezolizumab or spartalizumab (NCT04185883, NCT03785249, NCT04613596, NCT03600883, NCT04449874, and NCT04699188).

There is also interest in the study of the combination of KRAS^G12C inhibitors with inhibitors targeting the KRAS-related pathways. Dr. Canon’s team conducted experiments on KRAS^G12C-mutant cell lines, the results of which showed that combining sotorasib with EGFR, MEK, PI3K or AKT inhibitors resulted in stronger synergy. In vivo, sotorasib combined with the MEK inhibitor also exhibited much more obvious antitumor effects. These preclinical data suggested that this type of combination treatment might eliminate bypass signaling, which induced drug resistance (Canon et al., 2019). The combination of sotorasib (KRAS^G12C inhibitor) and afatinib (EGFR inhibitor) for the treatment of advanced NSCLC is being investigated (NCT04185883) in a clinical study. Another phase Ib, non-randomized, open-label trial is investigating the safety, tolerability and activity of the combination of adagrasib (KRAS^G12C inhibitor) and afatinib (EGFR inhibitor) (NCT03785249). Furthermore, participants with KRAS^G12C-mutant NSCLC received GDC-6036 (KRAS^G12C inhibitor) in combination with erlotinib (EGFR inhibitor). This phase I dose-escalation and dose-expansion trial is still ongoing (NCT04449874; due to be completed in August 2023).

Another potential approach is that of combining anti-PD-1/L1 inhibitors with inhibitors targeting the KRAS-related pathways. In a preclinical study, Dr. Liu demonstrated that MEK inhibitors combined with anti-PD-1 inhibitors increased the number of CD8⁺ tumor infiltrating lymphocytes (Liu et al., 2015). At the same time, in a model of BALB/c mice inoculated with CT26 cells, Dr. Ebert’s team found that monotherapy with anti-PD-L1 inhibitor or MEK inhibitor G-38963 only slightly inhibited tumor growth. However, combination treatment with these two inhibitors induced an obvious suppression, and in some cases led to complete regression and extended survival (Ebert et al., 2016). A phase Ib, non-randomized, open-label clinical trial is currently investigating the safety and efficacy of the combination of pembrolizumab (anti-PD-1 inhibitor) and trametinib (MEK inhibitor). The primary outcome measure is dose-limiting toxicity (NCT03299088). In addition, a phase I study on the combination of oral LXH254 (RAF inhibitor) with PDR001 (anti-PD-1 inhibitor) was recently completed, but its results have not been published as of yet (NCT02607813). Similarly, there is also a phase II clinical trial studying how effective the combination of atezolizumab (anti-PD-L1 inhibitor) and cobimetinib (MEK inhibitor) is for the treatment of patients with metastatic, recurrent and refractory NSCLC (NCT03600701; due to be completed in 2022).

Several preclinical studies have shown that KRAS-mutant cancer treatment required co-targeting the RAF/MEK/ERK and PI3K/AKT/mTOR pathways, due to multipoint crosstalk, negative feedback and redundancy (Carracedo and Pandolfi, 2008; She et al., 2010). The dual inhibition of these two pathways will eliminate the compensatory effect and produce an improved anti-tumor efficacy (Haagensen et al., 2012; Roberts et al., 2012). Clinically, in a phase I, non-randomized, open-label trial, patients were orally administered MEK inhibitor AZD6244 and AKT inhibitor MK2206. The results showed that the durable tumor shrinkage occurred in NSCLC and low-grade ovarian carcinoma with KRAS mutation following the combination treatment. A dose of 135 mg MK2206 weekly and 100 mg AZD6244 once a day was proposed as the recommended dose (NCT01021748) (Tolcher et al., 2015). In another phase Ib, open-label, dose-escalation, multi-center study, a lot of patients discontinued combination treatment with PI3K inhibitor BKM120 and MEK1/2 inhibitor MEK162, due to the intolerable toxicities associated with it. Despite the encouraging results in terms of efficacy, this combination strategy needs to be further explored to identify a better-tolerated dosage (NCT01363232) (Bardia et al., 2020). In addition, clinical trials on the combination of MEK and EGFR inhibitors, that of MEK and RAF inhibitors, and that of ERK and RAF inhibitors are being explored (NCT01229150, NCT02230553, NCT01337765, NCT01392521, NCT01390818, NCT03284502, NCT01859026, NCT02450656, NCT02974725, and NCT03905148).

All in all, compared with monotherapy, combination treatment may exhibit a much stronger antitumor effect. On the other hand, it may also lead to more serious adverse events. Potential approaches to overcoming toxicities include reducing the dose of both agents, evaluating the recommended dose and undertaking the pulsatile drug delivery. The advantages of pulsatile drug delivery include mitigating toxicities and delaying resistance (Bardia et al., 2020). Furthermore, natural compounds extracted from plants and herbs have fewer side effects. Therefore, combining natural compounds with western medicines may be promising.