KRAS-Mutant Non-Small-Cell Lung Cancer: From Past Efforts to Future Challenges

KRAS is the most frequently mutated oncogene identified in human cancers. Despite the numerous efforts to develop effective specific inhibitors against KRAS, this molecule has remained “undruggable” for decades. The development of direct KRAS inhibitors, such as sotorasib, the first FDA-approved drug targeting KRAS G12C, or adagrasib, was made possible with the discovery of a small pocket in the binding switch II region of KRAS G12C. However, a new challenge is represented by the necessity to overcome resistance mechanisms to KRAS inhibitors. Another area to be explored is the potential role of co-mutations in the selection of the treatment strategy, particularly in the setting of immune checkpoint inhibitors. The aim of this review was to analyze the state-of-the-art of KRAS mutations in non-small-cell lung cancer by describing the biological structure of KRAS and exploring the clinical relevance of KRAS as a prognostic and predictive biomarker. We reviewed the different treatment approaches, focusing on the novel therapeutic strategies for the treatment of KRAS-mutant lung cancers.


Introduction
In non-small-cell lung cancers (NSCLCs), identifying molecular mechanisms underlying tumor growth and progression is crucial to defining the best therapeutic approach for each patient. In recent years, great strides have been made to improve the individualized therapeutic approach for each patient. Among the molecular alterations predicting response to targeted treatment in lung cancer, epidermal growth factor receptor (EGFR) inhibition was the first to succeed [1]. Studies have shown that in patients with molecular rearrangements and specific mutations (e.g., EGFR, ALK and ROS1), targeted therapy with tyrosine kinase inhibitors (TKIs) leads to a significant improvement in survival and quality of life [2]. Some new molecules have shown efficacy in inhibiting rare oncogenic drivers, such as RET, BRAF, MET, NTRK and HER2 [3]. KRAS is the most common alteration identified in solid tumors and especially in pancreatic (88%), colorectal (45-50%) and lung cancers (31-35%) [4]. G12D, G12V, G12C, G13D and Q61R are five different mutations that account for 70% of all RAS-mutant patients. KRAS G12C mutations are frequently found in lung cancer due to G:C>T:A transversions associated with carcinogen-DNA adduct generated by the mutagens in tobacco smoke [5].
Given the high frequency of KRAS mutations in different types of aggressive tumors, numerous efforts have been made to block the function of this oncoprotein.
KRAS mutations may be associated with poor overall prognosis and response in advanced/metastatic NSCLC [6]. Traditionally, the standard treatment for patients with KRAS mutations in NSCLC is chemotherapy, and the average overall survival (OS) rate is less than 2 years [7]. Several studies showed that KRAS mutations in patients with NSCLC

RAS-Mutant Biological Heterogeneity
Intra-tumor heterogeneity and different treatment responses are a current challenge in developing targeted therapies in patients with lung cancer and KRAS mutation. Cooccurring molecular alterations in lung cancer have been revealed as a major concept on which the molecular diversity of tumor biology is based [17]. Concomitant mutations, including the TP53 tumor suppressor gene, serine/threonine kinase 11 (STK11) and kelchlike ECH-associated protein 1 (KEAP1) have been reported in about half of KRAS-mutant NSCLC [18]. The downstream signaling derived from different co-mutation partners leads to variances in genetic events, suggesting a difference in treatment response. Skoulidis et al. [18] explored data from an early stage and chemorefractory disease and examined the various heterogeneity of KRAS-mutant NSCLC. The following three KRAS subsets were defined based on the presence of co-mutations: STK11/LKB1 ('KL'), TP53 ('KP') and CDKN2A/B inactivation ('KC'). Another critical co-mutation described is KEAP1/NFIE2L2, which was included in the KL subgroup. Tumors with TP53 alterations demonstrated an improved overall response rate ([ORR] 7.4% KL vs. 35.7% KP vs. 28.6% KRAS mutations alone) [19]. Patients with co-mutations in KEAP1/NFE2L2 have a significantly shorter survival. Co-mutations of KRAS with STK11/LKB1 were associated with a lower response to immune checkpoint blockade, poor immune micro-environment, and low PD-L1 expression proved to be immune-inert tumors [19,20]. Instead, KRAS-TP53 co-mutant tumors expressed high levels of cytotoxic CD-8 Th1 and tumor-infiltrating lymphocytes [18]. These data highlight the heterogeneity of KRAS-mutant NSCLC, proposing the predictive value of concomitant genetic events.

Why Has Targeting KRAS Been Difficult?
One of the most challenging and, at the same time, attractive therapeutic targets in cancer is represented by mutant KRAS. RAS proteins lack a deep pocket for small molecules to bind with high affinity; high intracellular GTP concentrations and its strong affinity to KRAS brought poor results in the various efforts to find small drugs that bind tightly to RAS proteins. Furthermore, the indiscriminate inhibition and the high blood levels of proteins needed to block KRAS function could lead to great potential toxicity [21][22][23].
Therefore, alternative approaches have been adopted, focusing on the downstream inhibition of KRAS activity. These indirect strategies demonstrated low efficacy explainable by extensive post-transcriptional modifications and the activation of different KRAS signaling pathways, leading to the need to target the RAS protein with different simultaneous strategies [21]. The oncogenic transforming activity of RAS requires farnesylation; interfering with the farnesylation of Ras proteins prevents membrane localization and significantly reduces transformation activity. Several farnesyltransferase inhibitors (FTIs) have been tested to target the post-translational modification of KRAS and disrupt the farnesylation of oncogenic RAS (Table 1).

How to Target
FTIs showed in vitro activity in KRAS-mutant lung cancer in mice [24,25], but its clinical activity has not been confirmed in NSCLC clinical trials [26][27][28].
The unsuccessful strategy to target KRAS for cancer therapy through FTIs might be explained by the existence of different RAS-mutant isoforms [29]. Preclinical studies showed FTIs failed to inhibit KRAS and NRAS function due to alternative membranebinding mechanisms; therefore, only HRAS-mutant cancer cells were sensitive to FTIs.
Salirasib inhibits all isoforms of Ras in contrast to FTIs, and its efficacy as a single agent was tested in a phase II trial [30]. Salirasib did not show significant antitumor activity in this study, and no partial responses were observed.
Recently, several clinical trials investigated the efficacy of vaccination. mRNA-5671 was examined as monotherapy, and in combination with pembrolizumab in an ongoing phase I trial (ClinicalTrials.gov Identifier: NCT03948763).  [31]. Table 2 summarises the clinical trials examining strategies to prevent KRASmutant activity in a large number of patients with metastatic NCLSC by interfering with downstream pathways.
Since KRAS was found to be involved in lipogenesis inducing fatty acid synthesis [32], different phase I trials demonstrated promising preliminary results of fatty acid synthase (FASN) inhibitors in KRAS-mutant NSCLC [33][34][35]. Falchook et al. studied TVB-2640 in a phase I trial, the first-in-class FASNs, aiming to investigate the safety of this small molecule as a monotherapy and in combination with paclitaxel or docetaxel [33]. When TVB-2640 was administered as a monotherapy, the DCR was 42%, and no patient had a complete (CR) or partial response (PR). In combination with paclitaxel, the PR rate was 11%, and the DCR was 70%. A phase II trial is ongoing to determine the response rate of TVB-2640 in KRAS-mutant NSCLC patients by examining the RECIST and toxicity profile (ClinicalTrials.gov Identifier: NCT03808558).
Selumetinib, a MEK pathway inhibitor, was tested in the phase II Selected-1 trial. The study included more than 500 patients previously treated, and the association of MEK inhibition and chemotherapy did not improve PFS compared with docetaxel alone in KRASmutated NSCLC (3.9 vs. 2.8 months; HR 0.93, p = 0.44); the median OS was 8.7 months with selumetinib + docetaxel and 7.9 months with placebo + docetaxel (HR, 1.05; p = 0.64) [36].
A preclinical trial revealed different activity of the MEK inhibitor selumetinib using murine models with co-occurring KRAS mutations. The study proved that the concomitant loss of either TP53 or Lkb1 (also known as STK11) reduced the response of KRAS-mutant cancers to docetaxel monotherapy. The addition of selumetinib improved results in mice with KRAS-mutant cancers and TP53 co-mutations, while those with KRAS and STK11 co-mutations developed primary resistance to this combination therapy [37].
Similarly, trametinib, another MEK inhibitor, showed a similar PFS and response rate in patients with previously treated KRAS-mutant NSCLC, randomized to receive the MEK inhibitor or chemotherapy with docetaxel in a phase II trial [38].
The MAPK paradox explained that anti-BRAF agents in KRAS-mutated cancers are contraindicated because of the activation of tumorigenesis by the binding to BRAF [39]. Heidorn et al. demonstrated that the inhibition of BRAF in the presence of oncogenic or growth factor-activated RAS induces BRAF binding to CRAF (a type of RAF protein), leading to CRAF hyperactivation of this signaling pathway.
Therefore, other targets were explored. Sorafenib is an oral RAF inhibitor tested in the MISSION, a phase III, placebo-controlled trial conducted in molecularly unselected relapsed or refractory non-squamous NSCLC who failed at least two previous lines [40]. PFS was significantly longer in the KRAS-mutant subpopulation than in the KRAS wildtype group (2.6 versus 1.7 months; HR, 0.46; p = 0.007). However, the PFS advantage did not translate into a survival gain. Indeed, median OS was similar in the sorafenib and placebo groups with KRAS mutations (6.4 versus 5.1 months; HR, 0.76; p = 0.279) and wild-type KRAS (11.0 versus 9.1 months; HR, 0.79; p = 0.078).
Research on the PI3K/AKT/mTOR pathway showed that inhibiting those alterations with monotherapy inhibitors might not be sufficient. In a phase II trial, 79 patients received ridaforolimus, an oral inhibitor of mTOR [41]. The ORR, expressed by the complete response and partial response, at 8 weeks was 1%, and no significant benefit in OS was shown. Post-translational modification of KRAS promotes the membrane localization of this protein and permits KRAS signaling.

Immunotherapy in KRAS-Mutant NSCLC
The treatment landscape of metastatic NSCLC has been changed with the advent of monoclonal antibodies targeting PD-1 and its main ligand, PD-L1 [43][44][45][46]. Based on a subgroup analysis, in the CheckMate 057 trial, patients with KRAS-mutant lung cancers achieved the greatest OS benefit when comparing immune checkpoint inhibitors (ICIs) with chemotherapy (HR 0.52, 95% CI: 0.29-0.95) [43]. In the OAK study, a phase III clinical trial, the subgroup of patients with KRAS mutation also benefitted from immunotherapy with atezolizumab in terms of OS (HR = 0.71; 95% CI: 0.38-1.35) [45].
KRAS-mutant NSCLC is the target of interest for PD-1/PD-L1 inhibition for the following reasons: KRAS-mutant lung cancers are typically smoking-associated tumors, and are therefore often associated with a high mutational burden [47,48]; these cancers frequently show abundant T-cell infiltration; PD-L1 expression is present in approximately 24-55% of KRAS-mutant lung adenocarcinomas [49,50].
Nevertheless, KRAS-mutant lung cancers respond differently to immunotherapy and show differences in terms of the immunogenic profile [18].
As previously underlined, KRAS co-occurring mutations are fundamental factors determining distinct immune phenotypes within KRAS-mutant cancers. KRAS/STK11 co-mutations represent 25% of KRAS-mutant NSCLC, characterized by the low presence of tumor-infiltrating lymphocytes and reduced immune markers and PD-L1 levels. STK11 mutations constitute a major genomic driver of primary resistance to immune checkpoint inhibitors [19]. In a subgroup of patients with KRAS-mutant NSCLC in the phase III CheckMate 057 trial, patients with KRAS/STK11 co-mutations demonstrated significantly lower response rates than patients with KRAS/TP53 adenocarcinomas [19]. KRAS-mutant lung adenocarcinomas associated with KEAP1 mutational inactivation demonstrated lower expression rates of PD-L1 and other immune markers, proving refractory to anti-PD-1 antibody therapy [51,52]. Co-mutation in KEAP1 was also associated with shorter OS from the start of immune therapy in KRAS-mutant metastatic NSCLC [53]. KRAS/p53 co-mutations are associated with significant treatment responses and major susceptibility to immunotherapies, showing improvement in progression-free and overall survival [19]. These observations are based on the KRAS/p53 immune profile, which is characterized by an inflammatory response [21,53].
Maugeri-Saccà and colleagues explored intratumor heterogeneity and immunologic features among KEAP1/TP53-based subtypes in both blood-based NGS and tissue-based NGS cohorts. KEAP1 single mutant tumors had the shortest survival; the subgroup represented by KEAP1/TP53 double mutant had an intermediate prognosis similarly to the pure TP53-mutant subgroup; and the double wild-type group showed the longest survival [54].
A retrospective exploratory analysis of the IMpower150 phase III study showed that survival benefits, regardless of the treatment combination, were greater in the KRAS mutant and KEAP1/STK11 wild-type population compared to mKRAS and co-mutations in STK11 and/or KEAP1, suggesting both a prognostic and predictive value. Moreover, tumors presenting co-mutations in KRAS and TP53 had elevated PD-L1 expression. In contrast, mKRAS and co-occurring STK11and KEAP1-mutant tumors had reduced PD-L1 expression, suggesting that the addition of bevacizumab to atezolizumab may represent the preferred option for KRAS and TP53 co-mutated NSCLC [55].
An interesting treatment direction is to explore combination therapy using KRAS G12C inhibitors and immunotherapy. In a preclinical study, a durable response was produced in immune-competent mice with the administration of AMG510, resulting in the infiltration of tumors by CD8+ T cells and the development of a pro-inflammatory tumor microenvironment [56]. Several studies are underway to evaluate the synergistic activity between KRAS G12C inhibitors and anti PD-1/PD-L1 therapy (ClinicalTrials.gov Identifier: NCT03785249, NCT04185883, NCT03600883, NCT04613596).

CDK4/6 and SHP2 Inhibition
The description of a biological connection between KRAS and cyclin-dependent kinase (CDK)4/6 was observed in preclinical studies, suggesting that the robust and selective inhibition of CDK4/6 could respond to NSCLC patients carrying KRAS oncogenes [57,58].
Nevertheless, the JUNIPER trial [59] and the SWOG S1400C study [60], which tested the efficacy of abemaciclib and palbociclib as single agents, did not show great activity of CDK 4/6 inhibitors in mutant KRAS NSCLC tumors (Table 3).

Direct KRAS G12C Inhibition
The complex challenge of directly targeting the KRAS protein is highly connected to the unicity of this protein, characterized by a shallow surface, high affinity for GTP molecules and lack of active binding sites [63][64][65].
Developments in the field of crystallography facilitated the study of new molecules able to interact with the KRAS protein and its specific conformation [66]; with the aim of finding potent Switch II pocket inhibitors, several molecules were tested in vitro and in vivo (Table 4). Compound 12 had suboptimal pharmacologic properties, proved by the fact that the compound did not engage covalently KRAS G12C mutant cells [66,67]. Other molecules proved more potent and efficient in inhibiting KRAS mutant cells in vitro. ARS-1620 exhibited in vitro and in vivo potency and high selectivity for KRAS G12C [68].
Ostrem et al. demonstrated that small compounds could covalently bind to the switch II pocket of KRAS G12C; therefore, the numerous efforts with the aim of discovering this new pocket facilitated the direct targeting of KRAS. Recently, two molecules directly targeting KRAS G12C reached phase II/III trials. Their mechanism of action is based on the conversion of the preference of KRAS from GTP to GDP, holding a KRAS state from an active to inactive GDP-bound form and interrupting intracellular signaling and tumor cell growth [66].
Sotorasib (AMG 510) is an irreversible KRAS G12C inhibitor, which provided encouraging results in terms of response rate (RR) and duration of response (DOR) [69]. CodeBreaK 100 is a multicentre, single-arm, open-label clinical trial (ClinicalTrials.gov Identifier: NCT03600883) evaluating sotorasib alone and in combination with anti-PD-1/PD-L1, including patients with locally advanced or metastatic NSCLC with KRAS G12C mutations. Patients showed progression of the disease after three or fewer lines of therapy, including ICIs targeting PD-1/PD-L1, platinum-based combination chemotherapy or targeted therapy if they harbored EGFR, ALK, or ROS1 alterations.
In the phase I monotherapy arm of CodeBreaK 100, 59 patients with NSCLC achieved an ORR of 32.2%, median duration of response of 10.9 months and a median progressionfree survival of 6.3 months [69][70][71][72]. The dose escalation of 960 mg twice daily proved to be active and safe. The drug was well tolerated, with a grade 3/4 TRAE rate of 11.6%.
The phase II trial evaluated the efficacy and safety of sotorasib in a patient with locally advanced or metastatic KRAS G12C NSCLC who was previously treated with standard therapies. Among the 126 enrolled patients, Skoulidis et al. demonstrated an RR of 37.2%, a median PFS of 6.3 months and a median DOR of 10 months [73].
In May 2021, the US FDA granted accelerated approval to sotorasib based on Code-BreaK 100 for patients with NSCLC who received at least one prior line of anti-cancer therapy (immunotherapy and/or chemotherapy). This small molecule became the first drug targeting KRAS to be approved for treating locally advanced or metastatic NSCLC with KRAS G12C mutation [74].
The CodeBreak 101 study (ClinicalTrials.gov Identifier: NCT04185883) investigated sotorasib monotherapy and in combination with other anti-cancer treatments in patients with advanced solid tumors with KRAS G12C mutation. The CodeBreaK 200 study is a phase III trial comparing docetaxel to sotorasib in patients with KRAS G12C mutation in the second-line setting (ClinicalTrials.gov Identifier: NCT04303780). Skoulidis and colleagues conducted an exploratory analysis of the phase II trial that evaluated the activity of sotorasib and investigated the association between mutations in STK11, KEAP1 and TP53 and their response to sotorasib treatment. In the subgroup with mutated STK11 and wild-type KEAP1, a benefit in terms of response was seen in 50% of the patients. Patients with both mutated STK11 and KEAP1 showed a response of 23%; otherwise, wild-type STK11 and mutated KEAP1 demonstrated a response of 14%. Future prospective studies could validate the identification of subgroups of patients who may benefit differently from sotorasib treatment [75].
Recently, Jänne et al. published the results from cohort A of the phase I/II KRYSTAL-1 trial with single agent adagrasib in previously treated patients with KRAS G12C-mutated NSCLC [78]. After a median follow-up of 12.9 months, the objective response was 43%, with a disease control rate of 80% among the 112 patients with measurable disease at baseline. The median duration of response was 8.5 months, with 50% of NSCLC patients remaining on treatment. The median PFS was 6.5 months, and the median OS was 12.6 months, with 6-and 12-month OS rates of 71% and 51%, respectively. The authors evaluated the efficacy among 33 patients with treated, stable central nervous system (CNS) metastases. The intracranial ORR was 33%, with a DCR of 85%; the median intracranial PFS was 5.4 months. The treatment was well-tolerated, with 43% of patients experiencing grade 3 or 4 adverse events.
Subsequent clinical trials are currently investigating adagrasib as monotherapy or in association with other compounds in patients with advanced or metastatic solid tumors. The phase III KRYSTAL-12 study is recruiting pre-treated patients with NSCLC to MRTX 849 versus docetaxel [79]. In the KRYSTAL-1 trial, a subgroup analysis was performed with the aim of exploring co-mutations in STK11, KEAP1, P53 and CDKN2A. There were no differences in these subgroups except for patients with STK11 wild-type, KEAP1-mutant disease, which presented inferior response rates (only one of seven patients had a response) [80]. The phase II trial Lung-MAP S1900E with sotorasib (NCT04625647) will further clarify the impact of co-mutations on the activity of KRAS G12C inhibitors.

Resistance to KRAS G12C Inhibitor
Despite the evident clinical benefit, underlying mechanisms lead to the development of resistance in most patients treated with KRAS G12C inhibitors ( Figure 2).
As evidenced by clinical studies, KRAS G12C inhibitors did not determine significant tumor shrinkage in about half of the patients included in clinical trials. Furthermore, disease progression was evidenced in about 10% of patients. During the treatment with target therapy, the emergence of resistance mechanisms in cancer cells can determine the progression of the disease after an initial response or stability of the disease. therapy, the emergence of resistance mechanisms in cancer cells can determine the progression of the disease after an initial response or stability of the disease. Intercellular variability and intratumoral heterogeneity are recognized as the most prominent causes of resistance to KRAS G12C inhibitors; nevertheless, multiple resistance mechanisms are described, leading to the idea that combination strategies and a more personalized therapy are necessary to overcome resistances.
Resistance to KRAS G12C inhibitors can be either intrinsic or acquired. It has been suggested that putative resistance mechanisms could lead to intrinsic resistance to KRAS G12C targeting; the presence of non-KRAS G12C mutations may be evident at baseline, leading to significant clinical implications for patients eligible for KRAS G12 inhibitors [81].
Two main types of biological mechanisms of adaptative resistance are described, namely on-target resistance, determined by mutations or amplification in KRAS cells, and off-target resistance, with the activation of another oncogenic signaling pathway.
Ryan et al. examined the adaptive feedback response to KRAS G12C inhibition and observed evidence of rapid RAS pathway reactivation in most KRAS G12C models. These data showed that vertical pathway inhibition strategies, such as combinations of KRAS G12C inhibitors with SHP2 inhibitors, may be effective in interrupting feedback reactivation of the RAS pathway following KRAS G12C inhibition and may represent a possible therapeutic approach for KRAS G12C cancers [82].
As already mentioned, a phase I/II clinical trial (NCT04330664) tested the combination of adagrasib and TNO-155 (investigational SHP-2 inhibitor) and several other trials of SHP2 inhibitors alone or in combination with other drugs are currently ongoing.
The nuclear factor NRF2 (erythroid 2-like 2, NFE2L2), regulated by the KEAP1 protein, is a key transcription factor in the cellular antioxidant response. In addition to predicting poor response to checkpoint inhibitor immunotherapy [53], KEAP1 or NRF2 mutations may also be related to resistance to KRAS inhibitors, such as adragasib [83].
A recent clinical study evaluated 38 patients, of which 27 with NSCLC with KRAS G12C-mutant cancers were treated with adagrasib monotherapy [84]. The authors analyzed genetic variations in circulating tumor DNA or tissue samples from patients resistant to adagrasib through next-generation sequencing. In total, 42% of those patients showed a putative resistance mechanism to adagrasib, including secondary mutations or Intercellular variability and intratumoral heterogeneity are recognized as the most prominent causes of resistance to KRAS G12C inhibitors; nevertheless, multiple resistance mechanisms are described, leading to the idea that combination strategies and a more personalized therapy are necessary to overcome resistances.
Resistance to KRAS G12C inhibitors can be either intrinsic or acquired. It has been suggested that putative resistance mechanisms could lead to intrinsic resistance to KRAS G12C targeting; the presence of non-KRAS G12C mutations may be evident at baseline, leading to significant clinical implications for patients eligible for KRAS G12 inhibitors [81].
Two main types of biological mechanisms of adaptative resistance are described, namely on-target resistance, determined by mutations or amplification in KRAS cells, and off-target resistance, with the activation of another oncogenic signaling pathway.
Ryan et al. examined the adaptive feedback response to KRAS G12C inhibition and observed evidence of rapid RAS pathway reactivation in most KRAS G12C models. These data showed that vertical pathway inhibition strategies, such as combinations of KRAS G12C inhibitors with SHP2 inhibitors, may be effective in interrupting feedback reactivation of the RAS pathway following KRAS G12C inhibition and may represent a possible therapeutic approach for KRAS G12C cancers [82].
As already mentioned, a phase I/II clinical trial (NCT04330664) tested the combination of adagrasib and TNO-155 (investigational SHP-2 inhibitor) and several other trials of SHP2 inhibitors alone or in combination with other drugs are currently ongoing.
The nuclear factor NRF2 (erythroid 2-like 2, NFE2L2), regulated by the KEAP1 protein, is a key transcription factor in the cellular antioxidant response. In addition to predicting poor response to checkpoint inhibitor immunotherapy [53], KEAP1 or NRF2 mutations may also be related to resistance to KRAS inhibitors, such as adragasib [83].
A recent clinical study evaluated 38 patients, of which 27 with NSCLC with KRAS G12C-mutant cancers were treated with adagrasib monotherapy [84]. The authors analyzed genetic variations in circulating tumor DNA or tissue samples from patients resistant to adagrasib through next-generation sequencing. In total, 42% of those patients showed a putative resistance mechanism to adagrasib, including secondary mutations or amplifications to KRAS; the activation of the RTK-RAS signaling pathway; and the histological transformation of adenocarcinoma to squamous cell carcinoma. Mutations identified in the switch II pocket of KRAS may impede the binding of KRAS inhibitors. In addition, Awad et al. identified acquired resistance mechanisms, including MET amplification; activating mutations in NRAS, BRAF, MAP2K1, and RET; oncogenic fusions involving ALK, RET, BRAF, RAF1, and FGFR3; and mutations with loss-of-function in NF1 and PTEN.
Recently, Tanaka et al. described acquired resistance to adagrasib in an NSCLC patient with 10 secondary genetic mutations on the RAS-RAF-MEK-ERK pathway [85]. Among them, a new KRAS Y96D mutation was observed directly affecting the switch-II pocket, resulting in resistance to sotorasib, adagrasib or ARS-1620 across multiple models. Thereby, the authors suggested that the previously undescribed Y96D mutation may have a novel and specific role in driving resistance to KRAS G12C inhibitors [85]. RM-018 is a novel KRAS G12C inhibitor that binds specifically to the GTP-bound, active state of KRAS G12C. Analyzing the different mechanisms of action of this class of inhibitor, they hypothesized that RM-018 could have the ability to bind and inhibit KRAS G12C/Y96D and may represent a potential therapeutic strategy to overcome this acquired resistance mechanism [85].
Patients treated with KRAS G12C-targeted therapies can be affected by intrinsic and acquired resistance co-occurring in the same patient [86]. Combining KRAS G12C inhibitors with other treatments could improve the clinical response and overcome resistances.
Several mechanisms can contribute to the development of resistance to KRAS G12C inhibitors. The results of various studies highlight the need to develop KRAS inhibitors capable of overcoming resistances, determine the appropriate sequence of treatments and develop effective combination therapy regimens.

Open Questions and Future Challenges
The efficacy of KRAS G12C inhibitors could be enhanced by the combination of treatments with targeted agents, immune checkpoint inhibitors, and downstream KRAS inhibitors; several studies are ongoing to evaluate the clinical efficacy and safety of combination therapy.
In recent years, numerous efforts have been made to implement therapeutic possibilities for patients with KRAS-mutant NSCLC. Several specific drugs have been introduced into preclinical and current clinical practice.
The position of these inhibitors in the KRAS G12C lung cancer treatment strategy algorithm is still evolving; primarily, the activity of KRAS G12C may be limited due to the genetic heterogeneity and complex biology of KRAS.
A multiplex assay for mutations and fusion genes with next-generation sequencing (NGS) is essential for the appropriate selection of patients, as recommended by ESMO Clinical Practice Guidelines [3]. A possible approach is to treat patients unsuitable for standard first-line therapy with direct KRAS G12C inhibitors. Co-mutations of KRAS G12C and STK11 provided a poorer response to ICI therapy. Therefore, treatment with KRAS G12C inhibitors could represent a valid option in this first-line setting. Another strategy could be the association of KRAS G12C inhibitors with ICI and/or chemotherapy.
Overcoming the mechanisms of innate/acquired resistance that reduce the clinical efficacy of KRAS G12C inhibition constitutes a future challenge to obtaining a durable response. Combining KRAS G12C inhibitors with other therapy strategies, such as immunotherapy or KRAS pathway inhibitors, as previously described, could improve clinical benefits and overcome resistances.
Co-occurring genetic alterations and the mutant KRAS allele copy number gains define a variety of tumor microenvironments, which lead similar histological tumors of different drug sensitivities. Individualized treatments are needed to overcome the biological heterogeneity of KRAS-mutant NSCLC and ensure the development of new effective treatment strategies.
In addition, further developments are warranted to identify selective agents against KRAS G12D and G12V mutations, representing 38% of all KRAS-mutant lung adenocarcinomas. It is necessary to develop selective inhibitors against single mutations or pan-KRAS inhibitors able to block all mutant proteins.
Furthermore, many questions remain, and additional studies are needed to optimize these therapeutic strategies in the clinical setting; however, recent results are encouraging and demonstrate the potential to adequately treat patients with KRAS-mutant NSCLC.
A major concern is identifying patients who can potentially benefit from KRAS-targeted monotherapies or combination therapy strategies.