Abstract
Global statistics estimate that approximately 25% of patients with lung cancer are never smokers. We suggest that genes related to susceptibility to metabolic syndrome were present among those related to susceptibility to lung adenocarcinoma (AC) in never smokers. There are many questions concerning lung AC in never smokers, which is increasing in incidence, with female predominance, good prognosis, unique genes related to susceptibility and good response to treatment with specific agents. The purpose of this review was to investigate the carcinogenesis of lung AC in never smokers focusing on genes related to susceptibility to lung AC and carcinogens, including environmental factors. In order to clarify the carcinogenesis of lung AC in never smokers, the definition of never smokers, survey of environmental tobacco smoke, the presence of the physical characteristics of metabolic syndrome, and other carcinogens should be investigated for primary prevention of lung AC.
- Never smoker
- lung adenocarcinoma
- environmental tobacco smoke (ETS)
- susceptible gene
- metabolic syndrome
- review
Lung cancer was the leading cause of death in developed countries in 2012 in both males and females, and the leading cause in less developed countries in males and the second cause in females (1). Lung cancer is classified into two morphological groups, small cell lung cancer (SCLC) and non-small cell lung cancer (NSCLC). NSCLC includes squamous cell lung cancer (SC) and adenocarcinoma (AC). Lung AC is on the rise in both sexes, especially women, although the number is not as high as among men, in both developed and developing countries (2). Lung AC is associated with life-style throughout the world (1, 3).
Islami et al. reported that global incidence and mortality rates of lung cancer are still closely associated with smoking prevalence (4). However, high-income countries have shown remarkable decrease in smoking in both males and females (4). Parkin et al. reported that 15% of lung cancer cases in men and 53% in women are not attributable to smoking, and approximately 25% of patients with lung cancer were never smokers, according to global statistics in 2002 (5). This was a very shocking report. A review article by Sun, et al. (6) and others (7-10) reported that never-smoker patients with lung AC were predominantly female (6, 10), were significantly younger (7, 8) and had better prognosis (6, 7, 10-12), with especially good response to treatment with epidermal growth factor receptor-tyrosine kinase inhibitors (EGFR-TKIs) (7, 13-15). Such patients were frequently Asian females (4, 16).
We previously wrote a review on genes related to susceptibility to lung AC in never smokers and demonstrated that genes associated with metabolic syndrome were present among these (17). Many questions arose, e.g. whether lung AC may be a different disease (6, 7), whether it may involve different genes (8), why female predisposition is present, whether the number of patients continues to increase (16), and why there are so many patients with AC among never smokers with NSCLC (6, 16, 18). For primary prevention, it is most important to clarify the mechanism of carcinogenesis, especially to identify carcinogens related to genes susceptibility to lung cancer. Moreover, it should also be elucidated whether and how metabolic syndrome plays a role in these patients (17).
Concerning lung carcinogenesis in never smokers, environmental tobacco smoke (ETS) at home, work and other places have been supposed to be the cause of lung cancer in never smokers (4, 19-24). If never smokers have been exposed to ETS, it may act as a trigger, converting stem cells into cancer cells in the lung (25) with involvement of metabolic syndrome (26). Besides ETS, reported carcinogens include radon (27-29), cooking oil vapor (9, 30), indoor coal burning, hormonal factors (after hormonal replacement therapy) (6, 8, 9), occupational chemical exposure including asbestos/heavy metals (8), infectious factors (31, 32), and air pollution (8).
The purpose of the present study was to clarify the carcinogenesis of lung AC in never smokers from the viewpoints of both genes related to susceptibility, environmental factors and lifestyle. It is necessary to clarify why lung AC is increasing in Asian women never smokers (15, 29). Finally this review shows the interrelation between genes related to susceptibililty and suspected carcinogens from original articles on lung AC in never smokers, and discusses a primary prevention strategy for lung cancer in never smokers.
Selection of Bibliography
The articles cited in this review dated before 2000 were selected from review articles. Original articles after 2000 sought from PubMed were screened by us. The key words used in the search were as follows: lung cancer, lung AC, never smoker, NSCLC, incidence rate of global lung cancer, mortality rate of lung cancer, smoking prevalence, genes susceptible to lung cancer, and genes susceptible to lung AC. Of the articles published after 2000 we selected those we considered to be most important and appropriate for this study. The purpose of this review was to investigate the carcinogenesis of lung AC in never smokers focusing on genes related to susceptibility and suspected carcinogens, including environmental factors, in order to improve primary prevention of lung AC.
Original Articles on Genes Related to Susceptibility to Lung AC in Never Smokers
Thirty-one original articles listed in our previous review (17) were examined once again from the following standpoints: the hypothesis or purpose stated by the authors, the genes they had investigated, the study style used, the definition of never smokers, and the etiology of lung AC they analyzed such as ETS. Among these, the study hypothesis was considered the most important. If the authors had found genes related to susceptibility to lung AC and these genes were analyzed by the grade of heavy smoking but did not include discussion on never smokers, we deleted these articles from the list of originals in the present study. We focused more precisely on genes related to susceptibility to lung AC in never smokers (Table I).
1. Classical Genes
The hypothesis stated in the original articles on classical genes (20, 33-43) is as follows: exposure of never smokers to ETS resulted in lung AC, hence the genes affected in these cases are the same as those seen in the tobacco-related cases (33-36). The relative risk of the genes suspected in patients with lung AC exposed to ETS was higher than the frequency of suspected genes in patients with lung AC without exposure to ETS (Table I). The slow and the fast genotypes among the polymorphisms of N-acetyltransferase 2 (NAT2) were shown to be related to the risk of lung cancer among women never smokers (37). NAT2 participates in the detoxification of aromatic amines. The authors took into consideration exposure to cooking oil fumes among never smoker women who developed lung cancer.
The tumor suppressor genes, TP53 and, TP63 may be involved in carcinogenesis of lung AC in female never smokers (38, 39), although TP53 mutation has been reported in smokers with lung SC (17). These two reports (38, 39) did not describe the survey methods regarding ETS. Of course the authors understood the possibility of ETS, or other environmental risk factors, such as exposure to combustion products of indoor heating and cooking solid fuel and cooking oil fumes, as possible etiologies (38). Zhang et al. (39) introduced studies by Cianchi et al. (68) and Yao and Rahman (69), reporting that exposure to external carcinogens, including cigarette smoking, infectious agents, and dietary carcinogens, can result in inflammation and play a role in tumor development, but ETS in their cases was not analyzed because it was not their purpose. The purpose of their studies was to discover novel genes. DNA-repair genes were investigated in never smokers (40-43) and the authors strongly suggested the possibility of ETS (41, 42) or cooking oil fumes (40) as suspected carcinogens.
2. Susceptible Genes Found by Genome-wide Association Study (GWAS)
Nicotine acetylcholine esterase [cholinergic receptor nicotinic alpha (CHRNA)] genes found by GWAS for association with lung cancer reported by Amos et al. (70), Hung et al. (71) and Thorgeirsson et al. (72) indicated the association of chromosome 15q variants for CHRNA (rs16969968 and rs8034191) and risk of lung cancer with odds ratios (ORs) between 1.30 and 1.32. This was not found in patients with lung AC in Asia by Wu et al. (44), who identified four novel single-nucleotide polymorphisms (SNPs) for CHRNA3 (rs2036534c>T, rs667282C>T, rs12916984G>A, and rs6495309T>C) associated with significantly increased lung cancer risk and smoking behavior. Shiraishi et al. reported CHRNA SNPs to be associated with lung cancer susceptibility in a small subset of the Japanese population in a smoking-independent manner (45). These two reports from Asia did not focus on never smokers, and the results regarding smoking habits were very different between the reports, stressing the vital importance of the survey method used when analyzing smoking habits in patients as well as in controls (Table I). Wang et al. found no evidence of association between 6p21.33 or 15q25.1 variation and risk of lung AC in never smokers (47), and this finding was widely accepted in an article written by authors of 69 Institutions (73).
Telomerase reverse transpeptidase (TERT)–Cleft lip and palate transmembrane 1-like protein (CLPTM1L) variants may be involved in lung AC as a gene related to lung AC in never smokers (46-51) (Table I). The survey method regarding ETS was not described in this excellent and important article (Table I). The 5p15.33 region was associated significantly with lung AC in Asian female never smokers (48). Landi et al. revealed that 5p15.33 rs2736100 (TERT) was associated with risk of AC [odds ratio (OR)=1.23] (49). Jin et al. confirmed that 5p15.33, especially in the TERT gene, may also predispose to susceptibility to lung AC in Chinese female never smokers (50). Fine-mapping analysis of genetic variants in the 5p15.33 region conducted by Pande et al. revealed four SNPs associated with lung cancer risk (46): rs4975538, which is an intronic SNP in TERT, rs451360 and rs370348, which are intronic SNPs in CLPTM1L, and rs4975615, which is in the intergenic region between the two genes. However, these excellent articles did not discuss any association with ETS among never smokers.
Lan et al. made the extremely important finding that the strongest association signal, rs7086803 at 10q25.2, located at intron 7 of the vesicle transport through interaction with t-SNAREs homologue 1A gene (VTl1A) gene was implicated in lung carcinogenesis (52). VTl1A is involved in ACRP30-containing vesicles in adipocytes, and lower amounts of VTl1A in cultured adipocytes can inhibit adiponectin secretion (74). However, it is not clear how this international team defined the term never smokers or how several thousand cases and controls were surveyed for smoking habits.
Li et al. demonstrated a strong correlation between the transcription level of the gene glypican-5 (GPC5) and genotypes of the replicated SNP (rs2352028 at 13q31.3) in 77 non-tumor lung tissue samples, and the expression levels of GPC5 in the matched lung AC tissue were lower by half than in normal tissues (54). The main function of membrane-attached glypican is to regulate the signaling pathway of wingless transformation, hedgehog, fibroblast growth factors, and bone morphogenetic proteins (75). The authors showed that down-regulation of GPC5 contributes to the development of lung cancer in never smokers. Their results were discussed from the viewpoints of former smokers and never smokers who had quit smoking (54) in relation to ETS exposure in never smokers (76). Landi et al. did not agree with this observation (55). However, the study by Li et al. (54) opened the way to a new epidemiological approach to clarify the interrelationships between the susceptible genes found by GWAS and carcinogens in ETS.
Tessema et al. reported that never smokers with primary AC had a significantly higher prevalence of methylation of tumor necrosis factor receptor superfamily member 10C (TNFRSF10C), basic helix-loop-helix (bHLH) transcription factor 5 (BHLHB5) and boule-like RNA-binding protein (BOLL) (regulating meiotic G2/M transition) than current and former smokers (56). Genotypes of C3ORF21, which plays an important role in the formation of NOTCH EGF repeats, were nominally associated with a reduced risk of lung AC among never smokers (39). Another locus, 18p11.22, near the adenomatous polyposis coli down-regulated 1 (APCDD1), N-ethylmaleimide-sensitive factor attachment protein gamma (NAPG) and family with sequence similarity 38, member B (FAM38B) genes was reported in Korean never smokers with NSCLC by Ahn et al. (53). These findings for never smokers are very important, but the survey method of smoking was not sufficiently described.
3. Driver Genes
Driver genes are of interest as genes related to susceptibility to lung AC in Asian women never smokers from the standpoints of signal transduction for cell proliferation, survival migration and angiogenesis, as well as good treatment response to EGFR-TKIs if EGFR mutation was present (57-64) (Table I). Driver genes include mutations of EGFR, Kirsten rat sarcoma viral oncogene (KRAS), protooncogene B-Raf (BRAF), phosphatidylinositol-4,5-bisphosphate 3-kinase, catalytic subunit alpha (PIK3CA), and fusion of echinoderm microtubule-associated protein-like 4 (EML4) and the intracellular signaling portion of the anaplastic lymphoma kinase (ALK) protein (EML4–ALK).
Among never smokers with AC, the frequency of EGFR mutation was reported to be 42 ~50% in Asia (57, 58, 60), and 55% in those of European descent (63), while the frequencies of those with fusion of EML4–ALK, PTEN, PIK3CA, c-MET, KRAS, STK11, and BRAF were 9.3%, 9.1%, 5.2%, 4.8%, 4.5%, 2.7%, and 1.9%, respectively (57). Among smoker patients with AC, the frequency of EGFR mutation was reported to be 22.0% (57) or 10% to 20% (60), while the mutation frequency of STK11 was 19.0%, and of KRAS was 12.0% (57). That is, never smokers with lung AC showed increased EGFR mutation and EML4–ALK fusion protein. These mutations were at low frequency in smokers, but other mutations were seen in smokers with lung cancer (57). Although some cases showed overlap of EGFR and KRAS mutations, these were generally few in AC (57). The EML4–ALK fusion gene has been identified in a small subset of patients with lung AC (58). Ren et al. reported that 82.7% of lung AC in Asian female non-smokers showed well-known oncogenic mutations in EGFR, KRAS, HER2, BRAF and PIK3CA, and a majority of the mutations were mutually exclusive as noted above, except two with EGFR mutation and BRAF mutation, one with EML4–ALK fusion and PIK3CA mutation (59). Li et al. identified seven (among 208 cases; 3.37%) patients with the EML4–ALK fusion gene, of whom four had variant 3, two had variant 1, and one case had variant 1. Of these cases, six were non-smokers, and five were found among 33 cases of female non-smokers. EML4–ALK translocation was predominant in never smokers with AC (58).
Lung AC with EGFR mutation in never smokers showed somewhat good prognosis (57). The effect of ETS on EGFR mutation was reported in lung cancer in never smokers, and the rate of EGFR mutation among never smokers exposed to ETS was significantly low, as discussed below (60). Moreover, EGFR-TKIs, e.g. gefitinib and erlotinib, have been reported to be associated with high sensitivity to AC in Asian women non-smokers (57-61, 77, 78). Activating EGFR mutation, including 19 in-frame deletions and exon 21 L858R substitutions, have been shown to be the most potent biological predictors of sensitivity to EGFR-TKIs (60). Jou et al. found significant frequency of EGFR intron 1 SNP in Asian female never smoker patients with lung AC, as shown in Table I (61).
Other driver genes, human epidermal growth factor receptor type 2 (HER2) and BRAF, have been reported in female never smokers with AC (62, 63). Jo et al. found that three SNPs of HER2 (-3444C>T, -1985G>T and P1170A C>G) were significantly frequent in Asian female non-smokers and nondrinker patients with lung AC (62). Suzuki et al. examined HER2 gene mutation in 1,275 patients (1,055 ACs, 146 SCs, and 74 others) and detected HER2 gene mutation in 46 (3.6%) of all cases. Mutation-positive cases were all ACs, comprising 4.3% of ACs (79). These positive cases were younger never smokers, with smaller tumor size (79).
Jang et al. found novel fusion transcript formed between exon 1-9 of staphylococcal nuclease domain-containing protein 1 (SND1) and exons 2 to 3’ end of BRAF (63). This was observed in 3/89 tested tumors and 2/64 never-smoker lung ACs (63). BRAF mutations and their fusion transcripts lead to constitutive activation of Ser/Thr kinase activity and are located downstream of the RAF/MEK/ERK pathway in lung, melanoma, thyroid and colon cancer (63). SND1 is a component of the RNA-induced silencing complex and plays a role as a regulator for transcription of specific mRNAs by mediating RNA interference reported in colon, prostate and liver cancer (63).
4. Genes Related to Inflammation and Natural Immunity
Kang et al. used an Affymetrix custom-made GeneChip for finding novel genes related to susceptibility to lung cancer in female never smokers, and found three SNPs involved: colony stimulating factor 1 receptor (CSF1R) rs10079250 A>G, TP63 rs7631358G> A and core-pressor interacting with RBPJ 1 (C1R1) rs130090791 (65). CSF1R rs10079250 A>G exhibited an increased level of phosphorylated c-Jun NH(2)-terminal kinase (JNK), a downstream molecule of the CSF1-CSF1R signaling pathway. They did not analyze genes from the viewpoint of ETS. Kiyohara et al. reported that an association of IL1B (rs1143634, 3954 C>T) polymorphism with lung cancer was seen in never smokers (OR=1.11); higher risk was of course observed for smokers (OR=2.48) (66). Olivo-Marston, et al. reported that ETS exposure during childhood is associated with increased lung cancer risk among never smokers, especially never smokers with the haplotype mannose-binding lectin-2 (MBL2) (OR=2.52), suggesting alteration in inherited genetic variants of innate immunity genes (67). In these reports, the authors described in detail their surveys of smoking habits.
Definition of Never Smoker and Survey Methods for Smoking Habits
We sent e-mails with a questionnaire regarding the ETS exposure of never smokers to authors who had published original articles on genes and lung AC among never smokers in December of 2015. Very few responses were received. The reason for this is easily understandable because the hypothesis or purpose of the original articles was not to clarify carcinogenesis but focus on novel genes. Then, we examined the original articles listed in Table I again.
It is globally accepted that the definition of never smoker is as follows: a never smoker is an individual who has had a lifetime exposure of fewer than 100 cigarettes (University of California at Los Angeles, Harvard, Mayo, and International Agency for Research on Cancer studies) (24). This definition is the same as that of World Health Organization nomenclature (80).
Other definitions such as the Hawaii study (those who smoked fewer than 180 cigarettes in their lifetime), the Seoul study (those who smoked fewer than 200 cigarettes in their lifetime), the Liverpool study (those who never smoked more than 10 cigarettes per week regularly), the CREST study (those who either smoked less than 400 cigarettes in their lifetime or less than one cigarette per day for one year), the Aichi and GenAir studies (those who reported they had never smoked) have been reported (24).
We checked the definitions of never smoker described in original articles on novel susceptibility genes for lung AC in never smokers as shown in Table I, and found that most described never smokers as individuals with a lifetime exposure of fewer than 100 cigarettes.
Regarding the method used to survey the smoking habits of the patients, most authors described that a trained interviewer administered the questions regarding smoking. ETS at home, work or other places has been reported to be the cause of lung cancer in never smokers (19-24; Figure 1). De Andrade et al. showed that from 1997 to 2001, 810 women with lung cancer were interviewed to obtain data including the source, intensity, and duration of ETS exposure (76). In this descriptive study, relationships between smoking history, ETS exposure, and lung cancer histological subtypes were analyzed. Among the 810 patients, 773 (95.4%) reported personal smoking or ETS exposure, including 170 out of 207 (82%) never smokers. Among the never smokers with a history of ETS exposure, the mean years of exposure were 27 from smoking spouses, 19 from parents, and 15 from co-workers. For each major subtype of lung cancer (AC, SC, unclassified NSCLC, small cell, or carcinoids) among never smokers, 75-100% of patients had ETS exposure. Trends for AC, SC and small cell carcinoma were found to be statistically significant using the Cochran-Armitage Test for Trend I (p<0.001) among never smokers without ETS exposure, never smokers with ETS exposure, former smokers, and current smokers.
Kurahashi et al. reported a population-based prospective study (Cohort I in 1990 and Cohort II in 1993) including 28,414 nonsmoking women aged 40-69 years old, 28,414 nonsmoking women with exposure to ETS from their husband, at the workplace and during childhood (21). Over 13 years of follow-up, 109 cases with lung cancer were diagnosed. Among them 82 had lung AC. The hazard ratio (HR) in women with smoker husbands compared to those with never-smoker husbands was 1.34, and an association with lung AC had an HR of 2.03, with a dose–response relationship with the husbands' smoking. Tse et al. demonstrated an association between ETS on lung cancer in nonsmokers using a population-based, case-referent study in Hong Kong during 2004-2006, including 132 Chinese male nonsmoker cases with lung cancer and 536 nonsmoking community referents (22). They found a weak association between lung cancer and ETS exposure from household/workplace with OR of 1.11, and an increased risk of lung AC with OR of 1.68.
A recent report by the International Lung Cancer Consortium (ILCCO) in 2014 (24) is valuable in understanding the situation of ETS exposure among never smokers; it includes eight studies in North America, four studies in Europe and six studies in Asia/Oceania. The data comprise 12,688 lung cancer cases and 14,452 controls of which 2,504 cases and 7,276 controls were never smokers. This study showed that exposure to ETS increased risk of lung cancer among both ever smokers and never smokers. The association between ETS exposure and lung cancer development among never smokers was follows: OR=1.35 in males, OR= 1.27 in females; OR=1.56 in those ≥65 years old, OR=1.10 in those<65 years old. The adjusted OR comparing SCLC with NSCLC was 1.28 in the overall population and 2.11 in never smokers.
Clinical Features of Lung AC in Never Smokers
As noted above, global statistics estimate that 15% of lung cancer in men and 53% in women are not attributable to smoking, indicating that worldwide approximately 25% of patients with lung cancer are never smokers (5). Moreover, never smoker patients with lung cancer with AC were female, younger and had a better prognosis (6, 7, 10-12), although Pallis et al. noted that a correlation between younger age in never smoker lung cancer and the severity of cases was doubtful (9). NSCLC in never smokers is currently on the rise (16, 18, 22) and there is a question as to whether it may not be a different disease altogether (23, 24). Moreover, lung AC harboring an EGFR mutation responds to EGFR-TKIs, while that without EGFR mutation does not (85, 86). HER2 amplification was an unfavorable prognostic factor, but HER2 phosphorylation was a favorable prognostic factor (79).
Histopathology of Lung Cancer in Never Smokers
Incidence of lung AC in men surpassed the incidence of lung SC in the 1960s and 1970s in the USA and Europe, and in the 1980s and 1990s in Japan (3). The report of the International Association for the Study of Lung Cancer (IASLC) noted imbalance with respect to gender and histology: among 2,341 female patients, 55% had AC, 25% SC, while among 6,796 male patients, 30% had AC and 57% SC (1990~2000) (87).
Yang et al. showed a strong association between cigarette smoking history and lung AC in a prospective cohort of 41,836 Iowa women aged 55-69 years with 13 years of follow-up (84). Two-thirds of the enrolled population were never smokers; lung AC was seen in 25%, SC in 8% and SCLC in 4% of all lung cancer. Moreover, they found that women who developed SC consumed more alcohol but less fruit than women with other cancer types, and women who developed SCLC had higher waist circumferences than women with other types of cancer. Women who developed lung cancer were compared with smoker cancer cases, and the differential factors seen in never-smoker cases were higher education, consumption of more fruit, less alcohol and less physical activity.
Seki et al. reported the effects of ETS of spouses on lung cancer in 1,670 cases and 5,855 controls and noted a marginal association of ETS with female lung cancer risk (OR=1.31), whereas no significant association was observed for lung cancer in men (23). Moreover, a recent report by ILCCO noted that ETS exposure in never smokers or former and current smokers showed a strong association with SCLC (24).
Possibility of Metabolic Syndrome Participating in Lung AC in Never Smokers
Yang et al. surveyed body mass index (BMI) and waist circumferences in a prospective cohort of 41,836 Iowa women aged 55-69 years with 13 years of follow-up as described above (84). They showed patients with lung cancer had significantly higher waist circumference (p<0.15). Zhang et al. reported that a comprehensive analysis of adiponectin quantitative trait loci (QTLs) associated with gene expression correlation identified genes related to metabolic syndrome with a potential role in carcinogenesis (88). As discussed above, genes related to susceptibility to lung AC in never smokers include classical genes (participating in detoxifying or metabolizing carcinogenic agents derived mainly from tobacco smoke, tumor-suppressor genes such as TP53 and TP63, and DNA damage-repair genes), genes found by GWAS (nicotine acetylcholine esterase genes, TERT and CLPTM1L, VTL1A, GPC5, TNFRSF10C, C3ORF21, hypermethylation of TNFSF10C, BHLHB5, and BOLL), and driver genes (EGFR, KRAS, BRAF, PIK3CA, EML4-ALK) have been reported (17). Although these genes are not related to metabolic syndrome, EGFR, VTL1A, TNFRSF10C, C3ORF21 and hyper-methylation of TNFSF10C, BHLHB5, and BOLL are involved in the metabolic pathways of metabolic syndrome (Table II). Mazieres et al. reported a close relationship between lung AC in never smokers and metabolic syndrome (89). They examined 140 women with AC (63 never smokers and 77 former/current smokers) and found that never smokers were characterized by a higher frequency of lipidic features (60.3% vs. 37.7%) compared with smokers (89). Obesity, lack of physical activity, heavy alcohol consumption and a diet of food with high fat content all lead to metabolic syndrome and a high percentage of cancer (90, 91). Metabolic syndrome is frequently complicated with type-2 diabetes mellitus, which is associated with increased risk of lung cancer, especially among female diabetic patients with RR=1.14 (92).
Predominance of Women Among Never Smokers
Among the never-smoker patients with lung AC, there were fewer female patients than male patients. However, the percentage of never smokers among total females with lung AC was over 50%. It has been reported that these female patients were younger and had better prognoses, as mentioned above (6-12).
It remains unknown why lung AC is seen in Asian women never smokers (4, 16). Association of EGFR mutation and estrogen receptor (ER)-α and –β with lung carcinogenesis has been reported (93, 94). Mazieres et al. observed increased frequency of EGFR mutation and ERα expression in never smokers with AC with higher frequency of lipidic features (89). Li et al. also revealed that EGFR mutations were seen in 24.5% (51/208 cases) of patients with lung cancer. These mutations were identified with higher frequency in females (47.5% vs. 15.0% in males), never smokers (42.3% vs. 13.9% in smokers), and patients with AC (44.2% vs. 8.0% in patients with non-AC) (58). Moreover, human papillomavirus (HPV) infection may be involved in Asian non-smoking lung AC, which responds to EGFR-TKIs (95). Women with early menarche or late menopause showed significantly increased risk of lung cancer (96). Further epidemiological research should be conducted to clarify these points.
Interaction of Susceptibility Genes and Suspected Carcinogens
There is much information on genes related to susceptibility to lung AC in never smokers as detailed above. Comparing original articles of the 1990s with those published after 2000 by meta-analysis, Okazaki et al. reported that the role of classical genes associated with lung cancer is decreasing and novel genes are emerging (2010), a fact that may reflect changes in lifestyle in Japan (3). Although the carcinogenesis of lung AC in smokers is simple, that in never smokers is complicated. However, clarification is necessary for primary prevention of lung AC. Moreover, the numbers of patients with lung AC in never smokers is increasing for both sexes. Generally the frequency among female never smokers was predominant in lung AC. In developed countries, even though the frequency in men is less than in women, that in never smoker male patients increases after quitting smoking (4).
Different genes associated with susceptibility to cancer of course relate to different metabolic pathways to lung AC. For example, never smokers exposed to ETS exhibited the changes related to classical genes as Bennett et al. suggested (34). Members of the same group, Olivo-Marston et al. showed that ETS exposure during childhood might result in more susceptibility to ETS in never smokers with a haplotype of MBL2, suggesting alteration in inherited genetic variants of innate immunity genes (67). Li et al. found glypican-5 to be a novel gene related to susceptibility to lung AC in never smokers, and showed the possibility of a role for ETS in carcinogenesis in never smokers (54). Although the role of carcinogens in never smokers with most defined genes related to susceptibility remains unknown, the reported interrelations are shown in Table III.
Lee et al. clearly demonstrated that EGFR mutation rate was 56.6% in lung cancer in never smokers without ETS exposure, 44.0% in lung cancer in never smokers exposed for <45 ETS smoker years and 25.7% in lung cancer in never smokers exposed for ≥45 ETS smoker years (60). The same trend was seen for childhood ETS exposure, and household vs. workplace ETS, i.e. EGFR mutation rate was higher in never smokers without ETS, but a high level of ETS was associated with a low EGFR mutation rate; the EGFR mutation rate was lowest in smokers with lung cancer. According to the review of Choi et al., radon is recognized as the second leading carcinogen of lung cancer, and they listed the genes related to susceptibility as shown in Table III (29).
Subramanian and Govindan reviewed pre-existing lung diseases, oncogenic viruses and human papillomavirus in carcinogenesis of lung cancer in never smokers (97), and they also referred to Schabath et al.'s study (98), in which estrogen replacement therapy was used to exert a protective effect on women against developing lung cancer. Hormone or hormone replacement therapy should be investigated from the viewpoint of carcinogenesis between carcinogen and susceptible genes; further research is necessary.
We reported the possibility of genes associated with metabolic syndrome being related to lung AC in never smokers (17). However, while ETS and the other agents shown in Table III are suspected carcinogens, it is not yet clear whether they actually are. The references shown in Table II did not reveal the physical characteristics of metabolic syndrome such as body weight, body length, BMI, waist circumferences, hypertension, hyperlipidemia, blood sugar levels and hemoglobin A1c (HbA1c) levels (99). Even if the study subjects of these reports had the physical characteristics of metabolic syndrome, ETS or other carcinogens, drugs, infectious agents might be responsible for causing lung cancer because metabolic syndrome, obesity and type 2 diabetes mellitus are known to be complicating factors in malignant disease. The carcinogen(s) involved in never-smoker patients with lung cancer and metabolic syndrome requires further elucidation.
Future Directions
For primary prevention of lung AC in never smokers, of course, it is necessary to educate the general population that they should quit smoking and reduce ETS. We should promote the conduct of epidemiological surveys of suspected carcinogens in individual districts in order to clarify the carcinogenesis of lung AC in each case with genes related to susceptibility to cancer.
The relationship between AC and metabolic syndrome is especially important, because in general Asians are increasingly suffering from increased obesity and type-2 diabetes mellitus (92) as their life-styles change to include more Western-type foods and they engage in less physical activity. A recent report by Zanetti et al. showed the importance of ethnic groups with respect to susceptibility genes (100). The interaction of these genes and carcinogens, even if they are only suspected, should be investigated. The well-known review by Sun et al. (6) pointed out that EGFR mutation in never smokers vs. KRAS mutation and TP53 mutation in never smokers are frequently seen, and these differences in never smokers support the idea that different carcinogens are involved for different groups of never smokers. Moreover, the study of different oncogenic mutation spectra of lung AC in never smokers can lead to better selection of effective treatments and improve prognosis. We could establish strategies for primary prevention of lung AC in never smokers if we start to conduct surveys in individual districts.
Acknowledgements
The Authors thank Ms. Yoshie Muroya for her systematic research of the literature and Ms. Cecilia Hamagami for her assistance in the English revision of the manuscript.
The Authors are grateful to the Smoking Research Foundation (Tokyo, Japan) for support of this research in 2015~2016 (IO. SI, YS).
Footnotes
This article is freely accessible online.
Conflicts of Interest
The Authors confirm that there are no conflicts of interest regarding the contents of this article. This study was approved by the Ethics Committee of the International University of Health and Welfare (13-B-130; October 10, 2015).
- Received October 8, 2016.
- Revision received October 30, 2016.
- Accepted November 3, 2016.
- Copyright© 2016 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved