After decades of research, the treatment efficacy of advanced lung cancer has remarkably improved, by incorporating novel therapeutic strategies including targeted therapies inhibiting specific genetically activated proteins, or immunotherapies such as immune-checkpoint inhibitors (ICI) (1). Mutations affecting members of the RAS family genes (KRAS, HRAS, NRAS) are the most frequent genetic alterations in human cancer, affecting about 27% of all tumors, including lung, colorectal and pancreatic ductal adenocarcinoma, among others (2, 3).

Lung adenocarcinoma (LuAD) is a type of cancer with the largest number of oncogenic alterations that are therapeutically approachable (4). Approximately 20-25% of the LuADs harbor KRAS mutations, most of them affecting codons 12 (~85%), 13 (~10%) or 61 (~5%). In LuADs from smokers, the vast majority of KRAS mutations consist on guanine to thymine transversions, an effect that is associated with the tobacco carcinogens (5). At the aminoacid level, most of these mutations replace the glycine (G) in codon 12 by a cysteine (C) (G12C) and occurs in almost 50% of KRAS mutant tumors. On the other hand, KRAS mutations in never-smokers are less frequent, and the most prevalent changes involve nucleotide transitions that replace the glycine in codon 12 by an aspartic acid (G12D) or a valine (G12V) (6, 7) ( Figure 1 ).

In contrast to other oncogenic proteins activated in cancer and despite multiple attempts to harness it, KRAS was considered undruggable for a long time. The KRAS mutated proteins has a reduced capability to hydrolyze GTP or to interact with the GTPase-activating proteins, maintaining the oncogene and the downstream pathways constitutively activated. The lack of specific inhibitors targeting the KRAS hydrophobic pocket and the complexity of downstream pathways have contributed to the challenge of developing effective therapeutic strategies (8). After years of study, novel KRAS selective inhibitors became available for the KRAS G12C mutant protein, enabling a covalent binding that hinders downstream signaling, which led to promising results in the clinical setting. Two specific KRAS G12C inhibitors, first sotorasib (AMG510) and later adagrasib (MRTX849), earned the breakthrough designation by the US Food and Drugs Administration (FDA), to treat metastatic lung cancer patients harboring that particular KRAS mutation who have progressed to at least one prior systemic therapy. In the case of sotorasib, this was based on the efficacy results of the phase I/II CodeBreak-100 trial (NCT03600883) that reported an objective response rate (ORR) of 32% and disease control rate (DCR) of 88% among lung cancer patients. Thus, sotorasib has the FDA-approval for this clinical indication (9). On the other hand, data from the phase I/II KRYSTAL-1 trial (NCT03785249) for adagrasib showed significant benefit with an ORR of 45%, although the study is still ongoing and definitive conclusions cannot be inferred (10). Many other drugs targeting KRAS activation and other parallel and downstream pathways, as well as immunotherapeutic strategies, are currently under investigation in early-phase clinical trials (11).

Here, we aim to describe the tumor heterogeneity of KRAS mutant lung cancers and its immune-regulatory role, to report the efficacy with current immunotherapies, and to overview the therapeutic approaches targeting KRAS mutant tumors, other than KRAS G12C inhibitors.

Mutations on KRAS and on other actionable oncogenic drivers, such as EGFR or ALK, are often mutually exclusive. However, they co-occur with mutations in important tumor suppressor genes, such as STK11 (also known as LKB1), KEAP1, TP53, or CDKN2A, whose inactivation cooperates with KRAS in the oncogenic process and, thus, characterize the heterogeneous nature of KRAS mutant tumors (12, 13). In addition, the different KRAS mutated proteins differ on their biologic properties to hydrolyze GTP and to activate downstream signaling pathways, which determines the differences in their therapeutic vulnerabilities (14, 15).

Skoulidis et al. described a molecular classification of KRAS mutant LuAD according to the presence of co-mutations in tumor suppressor genes TP53 (~30%) and/or LKB1 (~30% each) (16). These co-mutation partners lead to variances in gene expression and distinct patterns of inflammatory and immune-checkpoint molecules release, which models the tumor microenvironment and promotes different responses to therapies (12, 17). Because of this tumor heterogeneity, the prognostic role of KRAS mutant cancers remains uncertain, although most studies report a major aggressive behavior of this type of cancer (18–20).

KRAS mutant tumors are characterized by enabling tumor cells to escape immunosurveillance as one of the hallmarks of cancer (21). NF-κB, STAT3, and certain suppressive inflammatory cytokines such as IL-6, IL-1β and GM-CSF, are key transducers of the immunosuppressive properties of KRAS driven tumors (22–24). Other mechanisms in this tumor type consist in increasing the expression of immune-checkpoints (e.g., PD-L1) to prevent T-cell effector functions, eliciting the release of myeloid-derived suppressor cells, regulatory T cells, and M2-differentiated tumor-associated macrophages, all of which impair antitumor immunity and facilitates tumor growth (25, 26) ( Figure 1 ).

In preclinical mice models, Stk11/Kras mutant tumors produced abundant IL-6 and were associated with neutrophil accumulation and inflammatory cytokines with immunosuppressive properties in the tumor microenvironment, along with increased levels of T-cell exhaustion markers, compared with normal lungs (27). Accordingly, KRAS/STK11-mutant tumors are associated to a “cold immunophenotype” with lower T-cell infiltration and lower rates of PD-L1 immunostaining, among other immunosuppressive features. Lung cancer patients with this tumor genetic profile showed a worse response to ICI than did patients with KRAS-mutant tumors without LKB1 co-mutations (27–30).

More recently, STK11 inactivation in KRAS-mutant tumors, has been shown to enhance the silencing of the STimulator of INterferon Genes (STING) protein, in part through epigenetic mechanisms (31). STING is a key component of the innate immunity and acts as a sensor for double-strand DNA (dsDNA) in the cytosol from virus and pathogens, which mediates the type I interferon production (32). By silencing STING expression, KRAS/STK11-mutant tumors become insensitive to cytoplasmatic dsDNA sensing, avoiding T-cell inflammation and promoting the recruitment of exhausted T-cell lymphocytes. Thus, restoring STING expression or activation of the cyclic GMP–AMP synthase (cGAS)-STING pathway with STING agonists, could induce T cell infiltration and turn this subset of tumors into more immunogenic, increasing the synergistic effects in combination with ICI (33).

Besides the immune-related nature of KRAS driven tumors, the smoking habit of patients with KRAS mutations has been associated with higher tumor mutational burden (TMB), which might predict better responses to ICI (34, 35).

Mazieres et al. retrospectively evaluated a cohort of non-small cell lung cancer (NSCLC) harboring different oncogenic driver mutations. Among them, the subset of KRAS mutant tumors expressed higher rates of PD-L1 and responded better to ICI than tumors with other oncogenic driver alterations (36). On the other hand, most phase 3 clinical trials evaluating all-comers with NSCLC treated with ICI did not stratify by KRAS status, and only post-hoc analyses have been performed on that subset. The following section compiles the evidence derived from exploratory analyses of the most relevant phase 3 clinical trials evaluating ICI and real-world data from retrospective cohorts of NSCLC patients treated with ICI.

The efficacy of nivolumab has been reported in two international studies. In the Italian Expanded Access Program out of the 530 patients who received 2^nd or 3^rd line nivolumab, 206 (39%) were positive for KRAS mutation. KRAS status did not influence nivolumab efficacy in terms of ORR (20% vs 17%, p = 0.39) nor DCR (47% vs 41%, p = 0.23) in patients with KRAS-mutant tumors when compared to KRAS-wild type (wt). No statistically significant differences were found in the median progression-free survival (mPFS) nor in the median overall survival (mOS) between both groups, although the 3-months PFS was significantly higher in KRAS-mutant patients (53% vs 42%, p = 0.01) (37). On the other hand, in the phase 3 study CA209-057, patients who had progressed to previous platinum-based chemotherapy (ChT), were randomized to receive either nivolumab or docetaxel. Among the 582 patients studied, KRAS was tested in 185 patients and 33% showed a KRAS mutation. When comparing the latter with KRAS-wt patients in the groups who received ICI, the hazard ratio (HR) for OS and PFS were 0.52 (95% confidence interval [CI]: 0.29-0.95) and 0.82 (95% CI: 0.47-1.43), respectively, in favor to KRAS mutant patients (38).

Regarding pembrolizumab, the clinical outcomes of two phase 3 clinical trials including KRAS mutant population have been reported. The KEYNOTE-042 trial (NCT0222089) evaluated pembrolizumab compared to platinum-based ChT as first-line treatment for PD-L1 positive tumors. The exploratory analysis presented by Herbst et al. showed that, out of 301 patients, 22.9% harbored KRAS mutations (9.6% were KRAS G12C) and presented higher levels of PD-L1 and tissue TMB. Both mOS and mPFS favored KRAS mutant tumors when treated with pembrolizumab compared to KRAS-wt, with a HR of 0.42 (95% CI: 0.22-0.81) and 0.51 (95% CI: 0.29-0.87), respectively (39). On the other hand, the KEYNOTE-189 trial evaluated platinum-based ChT either alone or in combination with pembrolizumab as first-line setting in advanced disease. Among 289 patients, 30.8% harbored a KRAS mutation and 12.8% a KRAS G12C. These tumors also presented higher levels of PD-L1 and higher tissue TMB. Pembrolizumab-based therapy was associated with improved clinical outcomes in terms of OS, PFS, and ORR regardless of KRAS status (40). Finally, results from real-world data published by Frost et al. from a multicenter, retrospective study evaluated the efficacy of first-line pembrolizumab in 119 patients with KRAS mutant LuAD with high PD-L1 expression (≥50%). Co-mutations in TP53 were also evaluated, and patients with KRAS G12C/TP53 had significantly higher ORR (100% vs 27.3%; p = 0.003) and longer mPFS (33.3 vs 2.8 months; HR, 0.18; 95% CI: 0.06-0.53; p = 0.002) than tumors with KRAS nonG12C/TP53 mutations (41), suggesting that KRAS G12C present better outcomes to immune-based therapies. Furthermore, Noordhof et al. reported another retrospective study that evaluated the outcomes of first-line pembrolizumab in 595 patients with metastatic LuAD and high PD-L1 expression. KRAS mutations were found in 57% of the cases. Although not statistically significant, mOS with ICI was higher in KRAS mutant patients than in those with KRAS-wt (19.2 vs 16.8 months; p = 0.86) (42).

In relation to atezolizumab, efficacy results are derived from a single trial, IMpower 150 (NCT02366143), a phase 3 study with first-line ChT and bevacizumab in combination or not with atezolizumab in patients with metastatic NSCLC. An exploratory analysis evaluated atezolizumab efficacy in KRAS mutant population according to STK11/KEAP1 mutation status. Among 920 evaluable patients, 24.5% harbored a KRAS mutation, which in 45% of the cases were co-mutated with STK11 and/or KEAP1. Greater benefits in terms of OS and PFS were observed in patients harboring KRAS mutations in the immunotherapy-based arm, regardless of STK11 and KEAP1 status (43).

Finally, results on durvalumab in stage III NSCLC patients after ChT-radiotherapy are available from a retrospective study that was carried out in 134 patients from MD Anderson Hospital. Patients with driver oncogenic mutations, including KRAS mutations (n=26) and targetable driver mutations (n=24) in EGFR, ALK translocations, ROS1 fusions, MET exon 14 skipping, RET fusion, and/or BRAF, had significantly worse mPFS compared to those without driver mutations (n=84) (8.9 months vs 26.6 months; HR 2.62 p < 0.001), particularly in cases with KRAS mutant tumors (mPFS 7.9 months, HR 3.34, p < 0.001), with no impact on OS based on driver mutation status (44).

KRAS is a GTPase that, when mutated, loses the ability to turn back to the GDP-bound state and leads to a constitutively active GTP-bound state. This, in turn, activates downstream signaling pathways, including MAPK, PI3K/AKT/mTOR, and Ras-like GEF, among others, all of them responsible for cell proliferation and survival (45, 46) ( Figure 1 ). In addition, the heterogeneity of KRAS mutations results in a variety of different diseases, which hinders the finding of a unique common therapy to address all of them. Thus, while KRAS G12C specific inhibitors have proven efficacy against their target, many other therapeutic strategies are currently under development for those KRAS mutant tumors with no druggable genetic alteration (47). Clinical trials addressed to KRAS mutant NSCLC non specific for KRAS G12C are listed in Table 1 .

Before the advent of KRAS G12C inhibitors, several strategies were tested to target the broad spectrum of KRAS mutant tumors, but most of them failed. Historically, the majority of these strategies focused on downstream effectors. For instance, the MEK inhibitor selumetinib combined with docetaxel showed good responses in early trials but failed to improve survival in a randomized phase 3 trial (48, 49). The cyclin-dependent kinase (CDK4/6) inhibitor abemaciclib was also tested in a randomized phase 3 trial against erlotinib but could not achieve its OS primary endpoint (50). Combining downstream effector inhibitors targeting MEK/PI3K demonstrated moderate responses but unacceptable toxicity profiles (51). Another therapeutic strategy focusing on the RAS family was the interruption of its anchoring to the cell membrane by inhibiting the post-translational farnesylation (e.g. tipifarnib). Unfortunately, despite a seemingly effective strategy in early phase trials, phase 2 and 3 trials assessing the effectivity in KRAS mutant tumors could not meet the expected outcomes (52). Hence, ChT remained the main treatment for these tumors for a long time, albeit with limited success.

Since KRAS mutant tumors form a heterogenous disease, nowadays, investigational efforts are focused on a subset of targeted therapies that can be further classified in different families according to their mechanism of action. Furthermore, to achieve better outcomes, the different families can be combined with themselves, with KRAS G12C inhibitors, or with immunotherapy. We will subsequently develop further each one of these potential options:

The high frequency of KRAS mutations in cancer justifies the multiple efforts invested in developing novel therapeutic strategies targeting KRAS. A deeper understanding of the cancer biology and immune system interactions that fuel carcinogenesis in KRAS mutant tumors is essential for developing new drugs and improving disease prognosis. Besides KRAS G12C specific inhibitors, several other drugs targeting KRAS directly or indirectly are being investigated. In addition, the list of actionable KRAS mutations in lung cancer will likely increase in the upcoming years.

Current immunotherapies seem to be effective for subset of KRAS mutant tumors, due in part, by the influence of smoking related nature of KRAS G12C mutations. The presence of co-mutations such as STK11 or KEAP1 shape the tumor immune microenvironment and might has an impact on treatment efficacy. Incorporating these genetic alterations in diagnostic panels as predictive markers represent a useful strategy for therapeutic decisions, including immunotherapy-based regimens.

Finally, the genomic complexity of KRAS mutant tumors will ultimately require tailored application of therapeutic approaches and upcoming data from clinical trials will contribute to provide the most promising strategies.

Conceptualization, EC, MC, and MS. Writing—original draft preparation, MC, EC, and MS. Writing—review and editing, MS-C, CH, LN, and AE. Supervision, MS, MC, EC, and MS-C. All authors have read and agreed to the published version of the manuscript.

This research received no external funding. MS is supported by a Juan-Rodés contract from the Instituto de Salud Carlos III (JR20/00015).

MC reports advisory/consultancy: Roche, Bristol-Myers Squibb, AstraZeneca; Travel/expenses: Pfizer, Roche. MSC reports a sponsored research agreement with Merck Serono Pharmaceuticals. CH reports research grants from Merck, and speaker’s bureau from MSD, Lilly and Ipsen. AE reports honoraris from Roche, MSD, AstraZeneca and Pharmamar. Travel/expenses: Roche, MSD, AstraZeneca, Lilly, Pfizer, Pharmamar. EC reports advisory/consultancy: AstraZeneca, Boehringer Ingelheim, Bristol Myers Squibb, MSD, Novartis, Roche, Takeda. Speaker bureau: AstraZeneca, Boehringer Ingelheim, Bristol-Myers Squibb. MSD, Novartis, Pfizer, Roche, Takeda, Amgen. Travel/accommodation/expenses: Bristol Myers Squibb, Pfizer, Roche, Takeda. MS reports a sponsored research agreement with Merck Serono Pharmaceuticals. Advisory/consultancy: Roche, Bristol-Myers Squibb, AstraZeneca, Takeda; Travel/expenses: Pfizer, Roche.

The remaining author declares that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.