Introduction
Lung cancer, especially non-small cell lung cancer (NSCLC), presents an epidemic on the rise, accounting for more deaths per year than the next three leading cancers combined [
1]. Although smoking cessation is fundamental for lung cancer prevention, currently most lung cancers develop in ex-smokers [
2,
3]. More importantly, a significant proportion of lung cancers occur in non-smokers and women [
4] and there is evidence to support that these cases are governed by a different pathobiology [
5]. Hence additional strategies for lung cancer prevention are needed to complement smoking bans, prevention, and cessation [
6]. For this to be achieved, better understanding of the molecular pathways that promote airway epithelial carcinogenesis is essential.
Previous work has linked inflammation and carcinogenesis in the gastrointestinal epithelium, and has identified the transcription factor nuclear factor (NF)-
κΒ as an important tumor promoter [
7,
8]. We and others have proposed that, in the lungs, carcinogen-induced inflammation and airway epithelial neoplasia are connected via activation of pro-inflammatory NF-
κΒ [
9‐
11]. However, experimental studies addressing the association of inflammation with lung carcinogenesis have so far focused on innate immune responses, such as those observed in the lungs of heavy smokers and patients with chronic obstructive pulmonary disease [
12‐
15].
Several epidemiologic studies have detected increased incidence of lung cancer in non-smoking patients with asthma [
16‐
20]. The increased risk has been estimated to be 1.5-3.0-fold compared to healthy non-smokers without asthma, while some studies have reported synergy of asthma with female gender, atopy, or polymorphisms in the interleukin (IL)-6 gene towards increasing lung cancer risk [
16‐
20]. One study also found increased risk of dying from lung cancer among patients with asthma [
18]. Although observational evidence supports an association of lung cancer with asthma, and although both disease processes have been extensively modeled in mice [
9‐
11,
21‐
26], no study to date has functionally evaluated the effects of the allergic adaptive immune response that characterizes asthma on lung carcinogenesis.
In the present studies, we aimed at determining the impact of experimental-induced allergic airway inflammation on chemical-induced lung carcinogenesis in mice. We hypothesized that either acute or chronic allergic airway inflammation promotes lung carcinogenesis. We chose the most widely used models to emulate the two conditions, the ovalbumin mouse model of allergic respiratory inflammation and the urethane mouse model of lung adenocarcinoma. We used the Balb/c strain of inbred mice, which uniquely displays susceptibility to both compounds. Studies were designed to dissect the effects of allergic airway inflammation on distinct time-periods of tumor initiation and promotion in the respiratory tract. Surprisingly, we found that ovalbumin-induced asthma does not functionally impact urethane-induced lung carcinogenesis.
Discussion
In the present studies we experimentally tested the hypothesis that allergic airway inflammation, such as that observed in asthma, promotes lung carcinogenesis. For this, we generated both ovalbumin-induced allergic inflammation and urethane-induced carcinogenesis in the lungs of Balb/c mice, sensitive to both compounds. Allergic inflammation was induced in both an acute and a chronic fashion and studies were designed for induction of respiratory allergy during distinct time-periods of multi-stage lung carcinogenesis. In stark contrast to what we anticipated based on reports of increased lung cancer incidence in asthmatic humans, we found no evidence that allergic inflammation influences chemical carcinogenesis in the murine lung. This was the case during both tumor initiation and tumor progression in the respiratory tract. Even animals with long-standing allergic airway inflammation resulting in marked structural alterations of pulmonary airways and parenchyma did not display evidence of enhanced tumor formation or progression. These results were furthered by studies on mice genetically engineered to lack IL-5, a mediator of asthma consistently up-regulated in our mice with ovalbumin-induced allergic respiratory inflammation. These mice did not exhibit any difference in lung tumor induction by urethane compared to wt littermates. Collectively, these studies indicate that allergic airway inflammation does not functionally affect chemical-induced lung carcinogenesis; that not all types of airway inflammation influence lung carcinogenesis; and that a mechanistic link between asthma and lung cancer may not exist.
Most cases of lung cancer are caused by smoking [
31]. In addition to genetic damage, smoking provokes chronic inflammation in the lungs, represented by the spectrum of illness coined chronic obstructive pulmonary disease (COPD) [
32]. Multiple lines of evidence from humans, cell and mouse models support that, in addition to the mutational stress imposed by tobacco carcinogens, chronic inflammation caused by smoking and/or in the context of COPD [
12] can induce or promote lung cancer formation and progression [
33]. In this regard, observations of increased lung cancer incidence in smokers with COPD compared with smokers without COPD after correction for smoking intensity and duration [
34,
35] have been coupled with functional studies in animal models that have identified and validated candidate molecular culprits for this link, including NF-κΒ, tumor-related protein 53(TRP53, P53), and Janus kinase (JNK) [
9‐
11]. These lines of evidence have established an association between innate immune responses in the lungs and lung carcinogenesis.
However, not all cases of lung cancer are caused by smoking. An estimated 10-15% of lung cancers is attributed to other genetic and environmental factors, such as occupational or domestic exposure to gases, fumes, or irritants and inherited somatic mutations or genetic polymorphisms [
36,
37]. In this regard, development of adenocarcinomas in never-smoking women in south-east Asia has been the focus of debate [
4], and there is evidence to support that these cases are governed by a different pathobiology [
5]. In addition, not all inflammatory lung disorders are smoking-related. Importantly, miscellaneous inflammatory and fibrotic pulmonary conditions like pulmonary fibrosis, tuberculosis, or asthma have been reported to be associated with increased lung cancer incidence [
36,
37]. Several reports now have linked asthma with increased lung cancer incidence [
16‐
20], setting the question of whether allergic airway inflammation promotes carcinogenesis in the respiratory tract.
In an effort to address this issue, we functionally modeled both asthma and chemical-induced lung cancer in mice. We used the most widely available models for this and set power analysis-based criteria to design this work [
9‐
11,
21‐
26]. Evidence for effective induction of asthma-like allergic airway inflammation was sought: ovalbumin-treated mice developed marked airspace eosinophilia and IL-5 up-regulation, widely used biomarkers of asthma [
38]. In addition, mice chronically exposed to the allergen developed structural changes reminiscent of the chronic airway remodeling that occurs in humans with difficult-to-treat asthma [
23,
38]. Despite the above efforts to discover a possible impact of experimental allergic airway inflammation on chemical carcinogen-induced tumor initiation or progression in the lungs of mice, our results show the absence of such an effect. In addition, we found that genetic deficiency in IL-5, a central mediator of allergy and asthma, has no impact on urethane-induced adenocarcinoma formation. This stands in contrast to previous observations from our group on the role of the cytokine in adenocarcinoma progression, in the forms of malignant pleural effusion [
39] and metastasis (unpublished data). In fact, we have observed a marked role of IL-5 in promoting intravenous and intrapleural tumor progression via immunomodulatory effects on the host response to tumor. These different results collectively indicate that the effects of IL-5 on malignant effusion and metastasis are specific and do not apply to more early stages of tumor induction, and that different components of the host immune system are involved during the different phases of tumor formation and progression in the respiratory tract.
Inflammation has been linked with cancer formation and progression. However, in contrast to a generalized effect of any type of inflammation on cancer formation, it is more probable that specific cellular, humoral, and transcriptional components of inflammation are involved in lung cancer formation and progression. In the lungs, while tobacco smoke [
11] and bacterial product-induced [
40] inflammation promote carcinogenesis, our study shows that allergic inflammation characterized by specific induction of eosinophil and IL-5 accumulation does not enhance chemical carcinogenesis. In this regard, while macrophages and neutrophils can function as potent promoters of tumor progression [
41,
42], eosinophils are probably mere bystanders recruited to tumor sites of necrosis [
43]. In addition, while mediators of innate inflammation positioned within the NF-κΒ pathway, such as tumor necrosis factor, promote lung carcinogenesis [
7‐
11,
29], our studies provide evidence that inflammatory mediators involved in other inflammatory signaling pathways, such as IL-5, do not affect lung tumor formation and progression.
The shortcomings of our studies are not to be overlooked. We only modeled allergic airway inflammation and chemical lung carcinogenesis using Balb/c mice, a single allergen, and a single carcinogen. In addition, we used the resistant C57BL/6 strain to study the role of IL-5 in lung carcinogenesis. However, Balb/c mice developed both allergic inflammation in response to ovalbumin and lung tumors in response to urethane, and should thus be an appropriate model for the study of the interactions between the two conditions. Moreover, sufficient lung tumors were induced in C57BL/6 mice by multiple urethane doses, facilitating the study of the role of IL-5 in lung tumor formation.
Since the original induction of urethane-induced lung tumors in C57BL/6 mice [
28], another group [
44] and we have observed increased tumor numbers in the lungs of urethane-treated C57BL/6 mice. This phenomenon could be ascribed to background strain variation, urethane batch variation, or other unidentified reasons. There is no evidence that C57BL/6 mice are currently more sensitive to other commonly used carcinogens, such as 3'-methylcholanthrene, since the original report by Miller et al [
28]. Although we and others have observed higher tumor numbers than Miller et al., C57BL/6 mice are still highly resistant to urethane-induced lung tumorigenesis.
Although negative, our study holds value in streamlining future research [
45,
46]. Our negative findings may aid in focusing future basic investigations into the relationship of lung carcinogenesis with inflammation in pertinent directions. In addition, the proposed absence of a mechanistic impact of allergic inflammation on lung carcinogenesis may aid towards focusing on other possible explanations for increased lung cancer detection in patients with asthma [
16‐
20], such as increased medical surveillance of this patient population. Another possible explanation for the human epidemiologic studies showing increased cancer detection in asthmatics is non-reported (occult) smoking in self-reported non-smokers, since these studies did not employ tobacco exposure biomarker assessment, such as cotinine or carboxyhemoglobin.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
KD carried out mouse experiments and immunoassays, performed histologic analyses, and helped to draft the manuscript. SPK participated in mouse experiments and immunoassays, performed histology, and helped to draft the manuscript. CAK participated in mouse experiments and helped to draft the manuscript. DCMS participated in the design of the study and helped to draft the manuscript. CR, SGZ, IK, and TSB participated in the design of the study helped to draft the manuscript. GTS conceived and designed and coordinated the study, carried out mouse experiments, performed histologic analyses, analyzed the data, wrote the manuscript, and revised the paper after peer-review. All authors read and approved the final manuscript.