If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Lung cancer is the leading cause of cancer death and oxidative stress secondary to carcinogens such as cigarette smoke has been implicated in its pathogenesis. Therefore, lung cancer patients were hypothesized to have higher levels of oxidative stress markers in their exhaled breath compared with controls.
Exhaled breath condensate (EBC) was collected from newly diagnosed subjects with non-small cell lung cancer (NSCLC) and control subjects in a cross-sectional observational study. The samples were then analyzed for hydrogen peroxide (H2O2), pH, 8-isoprostane, and antioxidant capacity.
A total of 71 subjects (21 NSCLC patients, 21 nonsmokers, 13 exsmokers, and 16 smokers) were recruited. NSCLC patients had significantly higher EBC H2O2 concentration (NSCLC subjects versus smokers, 10.28 μM, 95% confidence interval [CI]: 4.74–22.30 and 2.29 μM, 95% CI: 1.23–4.25, respectively, p = 0.003) and lower antioxidant capacity (NSCLC versus smokers, 0.051 mM, 95% CI: 0.042–0.063 and 0.110 mM, 95% CI: 0.059–0.206, p = 0.023; NSCLC versus all controls as a group, 0.051 mM, 95% CI: 0.042–0.063 and 0.087 mM, 95% CI: 0.067–0.112, p = 0.001). They also had significantly lower pH (5.9, 3.3–7.3) compared with exsmokers (6.7, 5.8–7, p = 0.009).
The significant increase of H2O2 and reduction in antioxidant capacity in the EBC of lung cancer patients further support the concept of the disequilibrium between levels of oxidants and antioxidants in lung cancer, which leads to increased oxidative stress. These findings suggest oxidative stress is implicated in the development of lung cancer and may be an early marker of the disease.
Oxidative stress has been implicated in the pathogenesis of many respiratory conditions including lung cancer (Figure 1). This is based on the hypothesis that the lungs are directly exposed to higher oxygen concentrations compared with other organs and hence more susceptible to increased oxidative stress, either through constant exposure to oxidants derived internally from normal metabolic processes or exposure to ambient air containing environmental irritants or pollutants such as cigarette smoke and free-radical generating environmental carcinogens.
The disequilibrium between levels of oxidants and antioxidants results in oxidative stress which has been implicated in the multistep process of carcinogenesis, because oxidants influence apoptosis, growth, and senescence as regulated by oncogenes and tumor suppressor genes such as CKIp21 and PTEN, and even to increase angiogenesis, which is a critical pathway in the proliferation of cancer cells.
Lung cancer subjects were hypothesized to have higher levels of oxidative stress markers in their exhaled breath compared with controls. The objective of this research was to measure specific oxidative stress markers in the exhaled breath condensate (EBC) of subjects, including hydrogen peroxide (H2O2), a surrogate measure of airway oxidant burden as it is soluble and equilibrates with air; 8-isoprostane, a prostaglandin analogue produced by free-radical-initiated peroxidation of arachidonic acid; pH that measures the levels of airway acidification; and antioxidant capacity that gives a general measure of the antioxidant status in the airway.
The study was approved by the South Eastern Sydney Area Health Service Research Ethics Committee. Subjects were recruited from Prince of Wales Hospital. Patients with newly diagnosed lung cancer before treatment were recruited from the oncology clinic. The control group included subjects without lung cancer, and no history of chronic obstructive pulmonary disease (COPD) or other respiratory conditions matched for socioeconomic and age group, and comprised nonsmokers, smokers, and exsmokers, defined as not having smoked for at least 1 year. Informed consent was obtained from subjects before sample collection and a questionnaire administered to obtain demographic details including medical history, medication, smoking history, and also staging and histologic typing of lung cancer as appropriate. The time since last meal, drinks (e.g., tea, coffee), or cigarette was also recorded and EBC was obtained from subjects only after at least 1 hour had elapsed since the consumption of food, beverages, or cigarettes. Individuals with recent or current infection/sepsis were not recruited in this study to prevent confounding effect of a local or systemic inflammation.
EBC was obtained by breathing into a nonsiliconized glass collection device, cooled by ice to 4°C as previously validated.
This method has been shown not to allow salivary contamination as judged by an assay for salivary amylase (unpublished data). Subjects breathed tidally for 10 to 15 minutes through a mouthpiece with a unidirectional valve leading into the cooling device to collect 1 ml of EBC. The samples were then aliquoted into Eppendorf tubes (120 μl), deaerated by argon degassing for 1 minute and agitated before storage under argon at −80°C before analyses for pH, H2O2, antioxidant capacity, and 8-isoprostane. The antioxidant butylated hydroxytoluene (5 μl of 0.13% BHT) was added into the 8-isoprostane aliquots to act as a free-radical scavenger before storage at −80°C.
Reagents were obtained from Sigma-Aldrich (Sydney, Australia) unless otherwise specified.
A pH meter (Model IQ125 Professional, IQ Scientific Instruments, San Diego, CA) was used to measure pH of the EBC samples. Three point calibrations, using pH 4, 7, and 14 buffers, were carried out before pH measurement. EBC samples were deaerated by argon and 10 μl of the samples was added to the probe tip before reading the values directly off from the display.
H2O2 was measured by a spectrophotometric assay method based on oxidation of 3,3′,5,5′-tetramethybenzidine by horseradish peroxidase (HRP) and H2O2. EBC samples (100 μl) were added to 420 μM 3,3′,5,5′-tetramethybenzidine (100 μl) in 0.42 M potassium citrate buffer and 52.5 U/ml horseradish peroxidase (10 μl). After 20 minutes at room temperature, the mixture was acidified to pH 1 with 2 M sulfuric acid (15 μl). The product, 3,3′,5,5′-tetramethyl-1,1′diphenoquinone 4,4′diamine, was then measured spectrophotometrically at 450 nm. Concentration of H2O2 in EBC samples was derived directly from a standard curve (absorbance against concentration of H2O2 in standards) with a range of 0 to 100 μM of H2O2. In this laboratory, the intraassay coefficient of variation is 8.2% and the limit of detection 0.2 μM.
8-isoprostane was measured using an enzyme immunoassay (Cayman, Ann Arbor, MI) according to the manufacturer's instructions. The intraassay coefficient of variation was 22.1% and the limit of detection 5 pg/ml.
Antioxidant capacity was measured by a spectrophotometric assay method (Cayman, Ann Arbor, MI) according to the manufacturer's instructions. The antioxidant capacity was measured by the inhibiton of the reaction of metmyoglobin with 2,2′-azino-di-(3-ethylbenzthiazoline sulfonate) measured by a change in absorbance at 750 nm. The antioxidant capacity was derived by comparing samples with Trolox standards (a tocopherol analogue). The intraassay coefficient of variation was 3.4% and the limit of detection 0.044 mM of Trolox equivalents.
H2O2, 8-isoprostane, and antioxidant capacity data were log-transformed to the Normal distribution and data were analyzed by parametric statistics (one-way analysis of variance [ANOVA], Bonferroni posthoc test, and independent unpaired two-tailed t test where appropriate). All values were expressed as geometric mean with 95% confidence intervals (CI) unless otherwise stated. pH data did not conform to the Normal distribution and were analyzed using nonparametric statistics (Kruskal-Wallis test and Mann-Whitney U test where appropriate). pH values were expressed as median, range. Statistical analyses were performed using SPSS 14.0.
A total of 71 patients were recruited: 21 non-small cell lung cancer (NSCLC) patients, 21 nonsmokers, 13 exsmokers, and 16 smokers. Some patients could not complete the entire collection process, resulting in limited number of samples collected for analysis of all markers for these patients.
Three of the lung cancer patients, two nonsmokers and one exsmoker were using inhaled glucocorticosteroids. Subject characteristics are summarized in Table 1.
TABLE 1Subject Characteristics
Subjects with Lung Cancer Stage 1 (1); Stage 2 (2); Stage 3 (9); Stage 4 (9)
Spirometry results were available for 14 NSCLC patients and approximately two thirds of the patients showed a degree of airway obstruction. The mean spirometry results were: forced vital capacity = 3.16 ± 1.03 l, forced expiratory volume in 1 second = 2.10 ± 0.78 l and forced expiratory volume in 1 second/forced vital capacity = 65 ± 13%.
There were significant differences in mean H2O2 concentrations between groups when patients with NSCLC (10.28 μM, 95% CI: 4.74–22.30) were compared with the control groups (nonsmokers, 4.87 μM, 95% CI: 3.48–6.81; exsmokers, 5.72 μM, 95% CI: 3.20–10.23; smokers, 2.29 μM, 95% CI: 3.20–10.23, p = 0.003, ANOVA, Figure 2). The mean H2O2 concentration of the NSCLC patients was significantly higher than the smokers (p = 0.003, unpaired t test, Figure 2); whereas current healthy smokers also had significantly lower H2O2 concentration compared with nonsmokers (p = 0.020, unpaired t test); and exsmokers (p = 0.028, unpaired t test).
There were significant differences in mean 8-isoprostane concentrations between the groups (p = 0.043, ANOVA, Figure 3). All samples had detectable levels of 8-isoprostane. The mean 8-isoprostane levels of each group are as follow: NSCLC, 37.9 pg/ml (95% CIL: 17.8–80.8); 28.9 pg/ml (95% CI: 13.9–60.1); exsmokers, 116.9 pg/ml (95% CI: 39.5–345.5); and smokers, 20.7 pg/ml (95% CI: 8.9–48.4). There was a significant difference between smokers and exsmokers (p = 0.011, unpaired t test). There was, however, no significant difference between NSCLC and the smokers (p = 0.26, unpaired t test); the exsmokers (p = 0.064, unpaired t test); and the nonsmokers (p = 0.59, unpaired t test).
Subjects with an EBC antioxidant capacity below the limit of detection of the assay kit were assigned a value equivalent to the lower limit of detection (i.e., 0.044 mM). NSCLC subjects had a lower mean EBC antioxidant capacity compared with smokers (0.051 mM, 95% CI: 0.042–0.063 versus 0.110 mM, 95% CI: 0.059–0.206, p = 0.023, unpaired t test). The mean EBC antioxidant capacity of exsmokers and nonsmokers were 0.067 mM (95% CI: 0.034–0.135) and 0.066 mM (95% CI: 0.040–0.107), respectively. However, there was no significant difference between groups (p = 0.132, ANOVA).
Lung cancer subjects had significantly lower antioxidant capacity when compared with the amalgamated control groups (0.080 mM, 95% CI: 0.058–0.110, p = 0.012, t test, Figure 4).
Significant differences in median EBC pH were observed between groups (p = 0.011, Kruskal-Wallis test, Figure 5). The median EBC pH of lung cancer patients (5.9, 3.3–7.3) was significantly lower compared with the exsmokers (6.7, 5.8–7, p = 0.009, Mann-Whitney). However, the median EBC pH of nonsmokers (6.0, 4.7–7.0, p = 0.003, Mann-Whitney); and smokers (6.0, 4.1–7.2, p = 0.004, Mann-Whitney) were also significantly lower compared with exsmokers.
This study showed that increased levels of oxidative stress in lung cancer were detected in the EBC of lung cancer subjects compared with that of controls. Four oxidative stress markers, pH, 8-isoprostane, H2O2, and antioxidant capacity, were tested with differing results.
H2O2 is formed from the conversion of superoxide anions by superoxide dismutase and is a marker of oxidative stress, which has been shown to be elevated in various respiratory conditions such as asthma, COPD, bronchiectasis, acute respiratory distress syndrome, common cold, and allergic rhinitis.
In this study, we demonstrated that there were significant differences between groups, with the lung cancer group having significantly higher mean H2O2 concentration when compared with the smokers. This finding has not been described previously. It is further supported by the finding that there was a significantly lower EBC antioxidant capacity in lung cancer patients compared with smokers and the amalgamated control group. This pair of novel findings lends further support to the theory that the disequilibrium between levels of oxidants and antioxidants results in oxidative stress which is implicated in the multistep process of carcinogenesis. The next step would be to further identify the specific types of antioxidants. Several enzymes are associated with H2O2 degradation, such as catalase and gluthatione-associated enzymes,
and it will be of interest to determine which of these enzymes are actually down-regulated/up-regulated so as to elucidate the pathways that are altered during carcinogenesis.
A perhaps unexpected finding was that smokers had the lowest H2O2 concentration and highest antioxidant capacity compared with the other groups. It was hypothesized that smokers would have higher levels of H2O2 compared with nonsmokers and exsmokers as smokers are exposed to significant quantities of reactive oxygen species in both gas and tar phase of cigarette smoke.
Furthermore, smokers have been shown to have higher total white blood cells counts, specifically with increase in neutrophils and these neutrophils in turn have higher levels of myeloperoxidase, which would then cause an increase in H2O2 production.
where smoking subjects were found to have higher levels of H2O2 compared with nonsmoking subjects. Smoking subjects in this study, however, have an approximately 20 pack-years longer smoking history and the mean age of the subjects of this study is approximately 20 years older than those in the study by Novak et al., which may have contributed to the difference in results, and it might have been expected that these smokers should have even higher levels of H2O2 given the greater exposure to the oxidative insult from cigarette smoking. Current understanding is that free radicals in cigarette smoke activate inflammatory cells with recruitment of neutrophils and macrophages into the airway, which in turn generate high levels of reactive oxygen and nitrogen species and other toxic metabolites.
The recruitment of these inflammatory cells may nevertheless increase antioxidant capacity at the particular site by bringing along with them antioxidant enzymes contained within these inflammatory cells and with some of these (e.g., gluthatione) being released for the protection of the extracellular milieu.
Furthermore, superoxide dismutase and catalase activities have been shown to be increased in human alveolar macrophages in chronic smokers, and gluthatione-associated enzymes and manganese superoxide dismutase induced by cigarette smoke.
have also demonstrated that 16 of 44 antioxidant-related genes were increased in the epithelium of smokers compared with nonsmokers. The results suggest that smoking may increase the antioxidant capacity of the airway so as to counterbalance the influx of oxidants from the cigarette smoke, potentially contributing to the significant decrease in H2O2 concentration when compared with healthy controls and lung cancer patients. This suggestion that chronic exposure to oxidants resulting in increased antioxidant gene expression before the antioxidant system being overwhelmed by oxidative stress as disease progresses is supported by the finding that healthy smokers have the highest levels of glutathione in bronchoalveolar lavage, followed by COPD patients and then COPD patients during an acute exacerbation.
For example, in extracellular superoxide dismutase (ECSOD), a single nucleotide substitution resulted in a 10-fold increase in circulating ECSOD activity and a decrease in tissue binding of ECSOD, with reports suggesting that this particular polymorphism may be a risk factor for developing chronic lung disease.
and it may be worthwhile performing a comparative study of more antioxidant enzymes and their functional polymorphisms in the various populations (e.g., smokers, exsmokers, lung cancer patients) so as to determine whether they are risk factors for the development of lung cancer or whether they confer protection against the development of neoplasia.
The pH of EBC is a measure of airway acidification and can be affected in disease processes such as acute asthma, cystic fibrosis, COPD, bronchiectasis, and acute lung injury, where hydrogen ion concentrations were found to be elevated.
As such, we postulated that pH of the EBC of lung cancer patients should be lower compared with the healthy controls. Although our results demonstrated significant differences between groups, the median pH of the lung cancer group was similar to the other groups except for the exsmokers. The exsmokers have significantly higher pH compared with the lung cancer patients, nonsmokers, and smokers. One explanation for the marked elevation of pH in exsmokers may be that the exsmokers were hyperventilating, resulting in the accumulation of bicarbonates in the EBC. However, this explanation may not be valid as all EBC samples were deaerated before storage and analysis. Therefore, carbon dioxide should have been driven out of the EBC samples and such a problem should not have occurred. Furthermore, subjects were told to breathe normally, and none of the subjects were observed to be hyperventilating. We observed that the pH values in this study were slightly lower when compared with other studies, which could be attributed to the differences in the degassing time. Nevertheless, it would be expected the class difference to remain and given that pH of lung cancer subjects is not significantly different from the other healthy controls, this may suggest that airway acidification is not implicated in the pathogenesis of lung cancer and its associated neoplastic processes.
8-isoprostane is a marker of lipid peroxidation, which has been shown to be elevated in the EBC of patients with chronic obstructive pulmonary disease but not in those with asthma.
Our study showed significant differences in the concentration of 8-isoprostane between groups, with higher levels of 8-isoprostane in the EBC of lung cancer patients compared with that of the smokers. Khyshiktyev et al.
have reported, without stating the actual marker, that levels of lipid peroxidation were lower in EBC of patients with lung cancer compared with controls, thereby making the association of lung cancer with increased lipid peroxidation measured in EBC a contentious subject. This is in contrast with several studies that have suggested that lipid peroxidation is implicated in pulmonary carcinogenesis with various products of lipid peroxidation (e.g., malondialdehyde, thiobarbituric acid-reactive substances, lipid hydroperoxide) shown to be elevated in the urine and tumors of lung cancer patients.
Further research is required to determine the relationship between the presence of various products of lipid peroxidation in the airways and the pathogenesis of lung cancer.
In this study, exsmokers had fewer pack years compared with smokers, but this is not surprising as age-matched current smokers would be expected to have a higher exposure. These variables could be adjusted for in a larger study by multivariate analysis, which could not be performed in this study. Although both control smokers and exsmokers had on average fewer pack-years of smoking compared with lung cancer patients, this difference was not significant and suggests that the observed differences in EBC markers are not related to differences in smoking history.
There are limitations in this study of EBC. Some of the samples were below the limit of detection of the antioxidant assay. Therefore, concentrating the EBC samples before analysis, or siliconising the glass collection device, may actually increase the yield of antioxidant capacity of some EBC markers as Liu et al.
have previously reported that collection devices significantly affect the levels of biomarkers obtained in EBC. This may have contributed to this study having higher levels of EBC H2O2 when compared with other studies which have collected EBC with the EcoScreen (Erich Jaeger GmbH, Hochberg, Germany), but the latter can contaminate EBC with exogenous NOx.
Although all subjects were recruited before treatment (e.g., radiation or chemotherapy), most of the lung cancer patients had more advanced disease at presentation (i.e., stage III–IV). Thus, our results may not be directly applicable to the use of EBC analysis as a screening tool for early detection of lung cancer in the high-risk population, and further research is necessary.
Future studies could include increasing the sample size to allow for analysis of lung cancer histopathology subtypes and effect of stage of disease. This will allow more specific biomarkers to be targeted for each type of lung cancer. The effect of treatment (i.e., chemotherapy or radiotherapy) on patients’ EBC could also be studied and used in tandem with other measures such as tumor size on computed tomography scan to assess response to treatment.
In conclusion, this study has shown significant differences in levels of H2O2 and antioxidant capacity between lung cancer subjects and other control groups. EBC may therefore be a novel method of sampling the lung to study markers involved in the pathogenesis of lung cancer and the role of oxidants and antioxidants.
The authors thank the subjects for donating their time to participate in this study, and the Australian Lung Foundation for awarding HPC with The Australian Lung Foundation/Lung Cancer Consultative Group Undergraduate Grant-in-Aid for Lung Cancer Research.
Superoxide dismutases in the lung and human lung diseases.