KRAS gene mutations in lung cancer: Particulars established and issues unresolved


Koji Okudela, MD, PhD, Department of Pathology, Yokohama City University Graduate School of Medicine, 3-9, Fukuura, Kanazawa-ku, Yokohama 236-0004, Japan. Email:


Lung cancer, like other cancers, is considered to develop through the accumulation of genetic alterations. Mutation of the KRAS gene is one of the most important events in carcinogenesis of the lung. The KRAS gene, belonging to the RAS gene family, encodes a membrane-bound 21-kd guanosine triphosphate (GTP)-binding protein. Single point mutations in this protein result in continuous activation to transmit excessive signals, promoting a variety of biological events. In lung cancers, the mutations concentrate at codon 12 and mostly affect adenocarcinomas (ADCs). They also affect atypical adenomatous hyperplasia, the precursor of ADCs. Therefore, mutation of the KRAS gene is suggested to confer a growth advantage to airway epithelial cells enabling them to expand clonally early in the development of ADCs. The mutation is also a reliable marker of an unfavorable response to certain molecular-targeting therapies. Furthermore, patients with ADCs affected by mutations have been reported to exhibit a significantly higher risk of postoperative disease recurrence. Thus, the significance of KRAS gene mutations has been investigated extensively. However, not all the details emerged. In this review, particulars that have been established are introduced, and important issues remaining to be resolved are discussed, with special reference to carcinogenesis of the lung.

Cancer of the lung, as well as other organs, is considered to develop through the accumulation of genetic alterations.1–3 There is no doubt that KRAS gene mutations are among the most important events in carcinogenesis of the lung.4–12

RAS genes, like many of the oncogenes, were originally discovered through the study of cancer-causing retroviruses in animals. Two rat-sarcoma (RAS)-inducing viruses, Harvey's RAS (hras) and Kirsten's ras (kras), each named after their respective discoverers, were identified in the 1960s.13,14 Thereafter, transforming genes were isolated from human cancer cell lines in the early 1980s.15,18 These genes turned out to be the human homologues of rat hras and kras 15–18 Subsequently, another ras family member, neuroblastoma RAS (NRAS), was identified.19,20 Further, additional ras homologues, the related RAS-1 (RRAS1), RRAS2 and RRAS3 (= muscle RAS (MRAS)) genes, have been identified.21–23

The RAS family genes encode a membrane-bound 21-kd guanosine triphosphate (GTP)-binding protein that transmit extracellular stimuli through the receptor tyrosine kinases such as epidermal growth factor (EGFR) family proteins to regulate cell growth, motility, differentiation, and death, by interacting with multiple effectors, including mitogen-activated protein kinase (MAPK), Ral guanin nucleotide dissociation stimulator (RAL GDS), signal transducer and activator of transcription (STAT), and phosphoinositide 3-kinase (PI3K).24–26 Single point mutations, most of which occur at codon 12, 13 or 61 in the RAS genes, result in the constitutive activation of the proteins.4,5,27–29 Among the RAS genes, KRAS is most commonly mutated in human cancers, although the prevalence of mutations and codons affected differ with type of cancer.7,30

Since the initial identification of the KRAS gene, the significance of its mutations in various cancers has been investigated extensively. Studies have uncovered the causative factors for the mutations, mechanism of occurrence, biological functions, roles in carcinogenesis, and clinical utility. However, there are still issues to be resolved. In this review, particulars that now have been established are introduced and issues remaining to be resolved are discussed, with special reference to its significance in carcinogenesis of the lung.


Lung cancers comprise four histological types; adenocarcinomas (ADCs), squamous cell carcinomas (SQCs), large cell carcinomas (LCCs) and small cell carcinomas (SCCs). Among these types, ADCs are most frequently affected by KRAS gene mutations.1,11,12,31–33 Adenocarcinomas can be divided into several subtypes.2 According to the standard classification proposed by the World Health Organization (WHO), these subtypes include mucinous bronchioloalveolar carcinoma (BAC), non-mucinous BAC, acinar carcinoma, papillary carcinoma, solid carcinoma, mixed carcinoma consisting of different subtypes, and other specific types.2 It was initially reported that mucinous BAC was almost exclusively affected by KRAS gene mutations (100% (10/10) in mucinous BAC, 23% (9/40) in non-mucinous BAC).34 Thereafter, several studies supported this finding (73% (8/11) in mucinous BAC, 10% (1/21) in non-mucinous ADC, 0% (0/15) in mixed ADC)35 (67% (6/9) in ADC with mucinous BAC component, 2% (2/82) in ADC with non-mucinous component, 7% (2/27) in ADC without BAC component).36 On the other hand, an extra-analysis in our recent publication11 revealed that mixed ADCs, mainly consisting of acinar and solid components, were frequently affected by the mutations (16.4% (14/85)), and the prevalence did not differ from that in mucinous BAC (20% (2/10)) (unpublished observation). Thus, the exclusive predisposition of KRAS gene mutations in mucinous BAC is difficult to accept, because these studies analyzed unsatisfactory numbers of cases with biases in sample selection and criteria of categorization. Association between occurrence of KRAS gene mutations and histological subtypes of ADCs has been still actually controversial.

It is pivotal to elucidate how the diverse histological types of lung cancer develop. Differences in cell type, carcinogens, and genetic alterations, have been proposed. Confirmation that specific subtypes of ADC are predisposed to KRAS gene mutations is an important step in resolving this issue.


KRAS gene mutations affect almost exclusively ADCs with a frequency of about 5% to 40%.1,7,11,12,31–33,36,37 Some 80% to 90% are single point mutations at codon 12, and the rest are relatively rare (codon 13, less than 10%; codon 61, less than 5%).7,11,36–38 There are ethnic differences in occurrence, with rates of 5% to 15% in Asians and 25% to 50% in Westerners.7,11,36–38 Interestingly, mutations of the KRAS gene and the EGFR gene, another important oncogene of ADCs, occur mutually exclusively.7,11,36–38 Epidermal growth factor mutations, in contrast to KRAS mutations, affect Asians more than Westerners, with a frequency of 25% to 50% and 5% to 15%, respectively.7,11,36–38 The finding suggests distinct differences in the causes and mechanisms responsible for ADCs between Asians and Westerners. Studies have as yet found no contribution of genetic background and lifestyle to the ethnic difference in occurrence of mutations of the two oncogenes in ADC.39,40 It will be very interesting to solve this issue in future investigations to discover the novel causes and mechanisms of developing lung cancers.

In addition to the ethnic difference, there are distinct differences in sex predisposition and smoking habits between patients with KRAS- and EGFR-mutated ADCS.11,12,37,38,41–45 KRAS gene mutations mostly affect ADCs in male and smoker patients,11,12,31–33,38 whereas EGFR mutations affect ADCs in females and non-smokers.11,12,38,41–45 Importantly, a recent large-scale study with ADCs from about 550 Western patients demonstrated that KRAS mutations occurred in 22% of the overall population and 15% of ADCs form subjects who had never smoked.7,46 Transition mutations (G to A) were more common in the non-smokers. In contrast, transversion mutations (G to T or G to C) were more common in former or current smokers.7,46 This finding suggests that tobacco smoke causes certain but not all types of mutation of the KRAS gene, promoting the development of some forms of ADC. On the other hand, potential carcinogens targeting the EGFR gene are unclear.41–45 The identification of potential carcinogens causing the KRAS transition mutations and EGFR mutations will contribute to the prevention of lung cancers.


KRAS gene mutations are a reliable marker of unfavorable response to treatment with of EGFR-targeting tyrosine kinase inhibitors, such Gefitinib and Erlotinib.7,38–45,47–49 Studies reported that ADCs with KRAS gene mutations also tended to be resistant against the conventional adjuvant chemotherapy with cisplatin/vinorelbine.50,51 Furthermore, the mutations were reported to be a predictor of a higher risk of postoperative disease recurrence and death of it in patients with ADCs.12,52,53 Thus, the KRAS gene mutation is likely to be a negative factor in survival of patients with ADCs. However, contrarily, there have been studies not confirmative of the above findings.7,39,40,54 Biases in analytical subjects (disease stages, histological types and personal character) and detection methods among studies could be account for this discrepancy.

The recent success with EGFR-targeting therapies has dramatically improved the survival of patients with ADCs affected by EGFR mutations.7,39–48 However, the clinical benefits of KRAS-targeting reagents, namely chemical inhibitors for farnesytransferase, geranylgeranyltransferase, carboxymethyltrasferase, and so on, have not been defined yet, although these reagents have been demonstrated to suppress the growth of cancer cells in in vitro experiments.55,56

The confirmation of prognostic value of KRAS gene mutations and establishment of clinically beneficial reagents targeting oncogenic KRAS is a pressing need and expected to further improve the survival of patients with lung cancers, especially ADCs.


KRAS gene mutations are thought to occur early in the carcinogenesis of ADCs, as they are found in atypical adenomatous hyperplasia (AAH), the precursor of certain types of ADC.1,2,57–60 However, the mutations alone are unlikely to be sufficient to generate AAHs, as discussed below.


Our previous studies8–10 demonstrated that the forced expression of oncogenic KRAS induced severe growth suppression in primary human airway epithelial cells, and also in Simian virus 40 large T antigen (SV40LT)—immortalized human airway epithelial cell lines.8–10 Similarly, others reported that the forced expression of oncogenic HRAS and/or KRAS markedly suppressed the growth of primary human fibroblastic cells61–63 and/or some human neuroblastoma cell lines,64,65 supporting that RAS mutations alone are insufficient to transform human cells and also suggesting that RAS could function as a tumor suppressor in certain instances.

Two types of oncogenic RAS-induced growth suppression and/or cell death, namely premature senescence66–69 and autophagy (type II programmed cell death),70,71 have been recognized. In addition, apoptosis induced by certain stimuli such as treatment of chemotherapeutic agents,72,73 cytokines,66,74–76 and depletion of growth factors,66 is also promoted through the RAS-mediated signaling pathways. Premature senescence is a complete growth arrest characterized by a marked flattening and enlargement of cells.61–63,77 Autophagy is cell death with cytoplasmic vacuolization.66,70,71 Apoptosis is cell death with nuclear shrinkage and fragmentation.66,78 The three could be mediated through different, but partly common, signaling pathways.66,77 Transcriptional activation of p16INK4A by ETS2 through the RAS-MAPK pathway was found to be essential to oncogenic RAS-induced premature senescence in primary fibroblastic cells.61–63,69,77,79,80 p53's activation by ataxia telangiectasia gene product (ATM) through the RAS-MAPK pathway was also reported to be essential to the induction of premature senescence, autophagy and apoptosis.62,63,66,75–77,81–83 Moreover, disruption of the machinery for controlling the cell cycle by oncogenic RAS, caused DNA-damage that triggered activation of the p53 pathway, resulting in apoptosis.77,84 On the other hand, the RAS-PI3K-mediated signal cascade was found to be crucial to a specific type of cell death similar to autophagy in some neuroblastoma cell lines.64,65 As described above, oncogenic KRAS induced severe growth suppression in primary and also SV40LT-immortalized human airway epithelial cells.9 The cells were markedly enlarged and had many vacuoles of various sizes in their cytoplasm (Fig. 1a).9 This morphologic change was consistent with autophagy.8 The SV40LT protein is known to inactivate both p53 and retinoblastoma protein (RB).85 Oncogenic KRAS induced autophagy-like growth suppression also in H1299 lung cancer cells whose p53 was inactivated through homozygous deletion and p16INK4A gene was inactivated through the promoter's hypermethylation.9,86 Also, oncogenic KRAS induced the same type of growth suppression in some other lung cancer cell lines whose p53 and p16INK4A was inactivated.9 KRAS mutations have been found in 25% to 35% of AAHs, but the inactivation of both p53 and p16INK4A merely accompanies.1,3,6,11,57 Moreover, the forced expression of a constitutively active phosphoinositide 3-kinase catalytic α (PIK3CA), a catalytic subunit of PI3K, did not induce growth suppression in immortalized airway cells and some lung cancer cells.87 Thus, p53, p16INK4A/RB, or PI3K is not indispensable to the oncogenic KRAS-induced growth suppression, and other unidentified effectors could be involved.

Figure 1.

(a) Morphological change on oncogenic KRAS-induced growth suppression in a Simian virus 40 large T antigen-immortalized human airway epithelial cell line.9,95 Empty vector (MOCK), wild-type KRAS (G12), or mutated oncogenic KRAS (V12) was transduced. After an appropriate period of selection during which cells not gene-transduced died, surviving cells were fixed and stained with the Papanicolaou method. The cells transduced with oncogenic KRAS (V12) were markedly enlarged and flattened, and had vacuoles in their cytoplasm. Magnification is ×400, each. (b) The selected cells were grown and passed several times. The means and standard deviations (error bars) of cumulated population doublings (PDLs) from triplicate experiments are presented.


Recent studies pointed out the level of expression to be an important factor in determining whether oncogenic RAS accelerates or suppresses cell growth.77,88,89 Extremely high level of oncogenic KRAS was reported to induce growth suppression.77,88,89 In contrast, physiological level of oncogenic KRAS promoted transformation.77,88,89 The expression of transduced genes via viral promoters in vectors is generally too strong. However, in our investigations, oncogenic KRAS severely suppressed growth, but some subpopulations survived and grew continuously (Fig. 1b).9 The level of KRAS protein did not differ between the cells at the dying cells and growing cells, suggesting that the level of KRAS is not necessarily sufficient to determine whether airway cells die or are transformed and also implying that the oncogenic KRAS-induced growth suppression is not simply an artifact produced by in vitro experiments.


Aside from the complexity of the signaling pathway(s) promoting the KRAS-induced growth suppression, its potential significance to carcinogenesis is also an important issue. We found that the SV40LT-immortalized human airway epithelial cell lines and lung cancer cell lines, which were temporarily affected by the oncogenic KRAS-induced growth suppression, had subpopulations resistant to the suppression with an incidence of 1/1000 to 1/100 of the overall population.9 Similarly, our preliminary study using a highly sensitive detection method of KRAS gene mutation (restriction enzyme-mediated selective polymerase chain reaction)90 found that a tobacco-specific nitrosamine (4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone [NNK])91 induced KRAS gene mutations in murine lung with a frequency of at least more than 1/1000 early (2 weeks) after a single intraperitoneal injection, where considerable number of airway cells in whole lung were estimated to be affected by the mutations, but only a few, less than ten, tumors developed (32 weeks) (data not shown). Also, conditional expression of oncogenic KRAS in whole murine airway epithelia in vivo was reported to result in severe pneumonia (alveolar damage) at an early stage92 and in turn adenomatous neoplasms later on.93 Moreover, in our in vitro experiment, forced expression of oncogenic KRAS led to severe growth suppression in lung cancer cell lines without KRAS gene mutations, but had only a modest effect on cell lines with the mutations,9 suggesting that lung cancer cells with KRAS gene mutations had overcome the oncogenic KRAS-induced growth suppression during their development. It is speculated that airway epithelial cells affected by specific genetic and/or epigenetic alterations could acquire resistance (Fig. 2a), or airway epithelia could physiologically pool such resistant subpopulations (Fig. 2b). Alternatively, specific genetic polymorphism(s) might be involved in the resistance and a molecular basis to the development of multiple lung tumors consisting of ADCs and AAHs (Fig. 2c). Thus, it is of great interest to investigate the potential cause and mechanism of the resistance to oncogenic KRAS-induced growth suppression.

Figure 2.

Hypothetical schema of KRAS gene mutation-associated carcinogenesis of lung adenocarcinoma (ADC). (a) Airway epithelial cells (green columns) affected by specific genetic and/or epigenetic alteration(s) (yellow stars) acquired the potential resistance to oncogenic KRAS-induced growth suppression at the initiation stage. The KRAS gene mutation (red bursts) promoted the growth of resistant cells to form atypical adenomatous hyperplasia (AAH), but suppressed the growth of others (light blue columns), during the promotion stage. Additional alterations (purple bursts) caused AAH to develop into overt ADC during the progression stage. (b) Airway epithelia physiologically pool specific cells potentially resistant to the oncogenic KRAS-induced growth suppression (yellow columns). The resistant cells (yellow columns) affected by KRAS gene mutations (red bursts) instantly form AAH, but conventional cells (yellow columns) affected by the mutation (red bursts) are growth-suppressed and removed (light blue columns). Additional alterations (purple bursts) caused AAH to develop into overt ADC. (c) Airway epithelial cells in certain individuals with specific genetic polymorphism(s) (orange stars) making cells resistant to the oncogenic KRAS-induced growth suppression, instantly form AAH by KRAS gene mutation (red bursts). In such instances, multiple AAHs easily develop. Additional alterations (purple bursts) caused AAH to develop into overt ADC.


The multistep theory of carcinogenesis proposes that initiation, promotion, and progression are essential stages in the development of overt malignant neoplasms.3 Initiation is when cells are ready to transform, but yet to form tumors.3 Promotion is the stage when the initiated cells form tumors.3 Progression is where tumors become more aggressive and highly malignant. These three stage progressing, multiple genetic and/or epigenetic alterations have been suggested to accumulate in neoplastic cells.3 Indeed, oncogenes (KRAS and EGFR, etc.) and tumor suppressors (p53 and p16INK4A, etc.) potentially involved in the promotion and/or progression stages have been identified.3 Meanwhile, the initiation step is conceptual and hypothetical. The pre-neoplastic cells involved in this step are not recognizable by histopathological examination. Thus, one cannot selectively isolate initiated cells. Therefore, potential genetic and/or epigenetic alterations specifically involved in initiation have not been identified. The in vitro experimental system producing the oncogenic KRAS-induced growth suppression is expected to make it possible to uncover the potential molecular basis of this step. We have speculated that the inactivation of factor(s) triggering, or activation of factor(s) preventing, the oncogenic KRAS-induced growth suppression could be involved in the initiation step (Fig. 2).

A gene-chip microarray analysis revealed that oncogenic KRAS elevated the mRNA levels of many putative growth suppressors, not only growth activators, in a SV40LT-immortalized human airway epithelial cell line (Tables 1,2).9 The result suggested the KRAS-mediated oncogenic pathway to be tightly regulated through negative feedback via downstream factors and supported that KRAS gene mutations alone are not sufficient to produce neoplasms (Fig. 3). These downstream factors are considered as important clue to the underlying mechanism(s) mediating the oncogenic KRAS-induced growth suppression, and also to provide candidates for novel tumor suppressors as well as oncogenes playing important roles in carcinogenesis of the lung. Indeed, we previously identified several candidates for tumor suppressors (DUSP6, FXYD3, IGFBP-2 and IGFBP-4), which could be involved in the progression to a highly malignant neoplasm (Fig. 3).9,10,94 However, the potential effectors making cells resistant to oncogenic KRAS-induced growth suppression, which would play an important role in the initiation of carcinogenesis, have yet to be identified. Such effectors are expected to improve our understanding of the multiple steps of carcinogenesis and to help establish a novel paradigm.

Table 1.  Downstream targets of oncogenic KRAS (upregulated genes)
CommonGenbankMapMOCKKRAS/G12KRAS/V12Ratio V12/MOCKRatio V12/G12
Signal valueFlagsSignal valueFlagsSignal valueFlags
  1. MOCK, empty vector-transduced; KRAS/G12-transduced; KRAS/V12-transduced cells; Common, common gene name; Map, Chromosomal locus; GenBank, gene bank accession number; Flags indicate whether gene expression is present (P) or absent (A). This table is quoted and modified from reference.9

Table 2.  Downstream targets of oncogenic KRAS (downregulated genes)
CommonGenbankMapMOCKKRAS/G12KRAS/V12Ratio V12/MOCKRatio V12/G12
Signal valueFlagsSignal valueFlagsSignal valueFlags
  1. MOCK, empty vector-transduced; KRAS/G12-transduced; KRAS/V12-transduced cells; Common, common gene name; Map, Chromosomal locus; GenBank, gene bank accession number; Flags indicate whether gene expression is present (P) or absent (A). This table is quoted and modified from reference.9

 M78162 2.2374306P1A0.2505935A0.112000560.25059348
Figure 3.

Negative feedback regulation of the KRAS-mediated pathway and its disruption during progression stage of lung adenocarcinoma. Left panel: Oncogenic KRAS activates (standard arrows) and (T-shape arrows) suppresses multiple downstream factors to transform cells. Among the factors, DUSP6 and IGFBP2/4 control the KRAS-mediated pathway in a negative feedback manner. Right Panel: DUSP6 and IGFBP2/4 inactivated though the genetic or epigenetic alterations during progression,11,94 resulting in augmentation of the KRAS-mediated oncogenic signal to form highly malignant neoplasm. During this progression stage, atypical adenomatous hyperplasia (AAH) develops into bronchioloalveolar carcinoma (BAC) and it further into overt adenocarcinoma (ADC).


More than thirty years have passed since the initial identification of RAS genes. The significance of their oncogenic mutations has been investigated extensively. This review focused on KRAS gene mutations with special reference to lung cancers, and pointed out important issues remaining to be resolved. Notably, the biological role of the mutations in the development of lung cancers is still not entirely established, although KRAS has long been believed to be a transforming gene. Investigation of the potential role of mutated KRAS in carcinogenesis is as ever fundamental to cancer research.


This work was supported by the Japanese Ministry of Education, Culture, Sports, and Science (Tokyo Japan), Smoking Research Foundation (Tokyo, Japan), and by a grant from Yokohama Medical Facility (Yokohama, Japan). We especially thank Masaichi IKEDA and Hideaki MITSUI (Department of Pathology, Yokohama City University Graduate School of Medicine), Nobuo OGAWA (Division of General Thoracic Surgery, Kanagawa Prefectural Cardiovascular and Respiratory Center Hospital), and Emi HONDA, Shigeko IWANADE (Division of Pathology, Kanagawa Prefectural Cardiovascular and Respiratory Center Hospital) for assistance.