Studies suggest that nitric oxide (NO) may have a possible role in lung carcinogenesis. This study is aimed to evaluate the association of the NO metabolites, namely, nitrite and nitrate, with lung cancer incidence. We conducted a matched case-control study (
Lung cancer is one of the most common causes of cancer death worldwide with a poor prognosis [
Endogenous nitric oxide (NO) is a multifunctional inflammatory molecule and promotes inflammation under physiological condition [
NO plays a pivotal role in cancer development. On the one hand, excessive NO is toxic and can prevent tumor growth by increasing the apoptosis rate of cells. While on the other hand, NO is a mediator of signaling pathways, which promote cancer progression and metastasis [
However, it is difficult to detect NO in tissues and biological fluids directly due to its highly reactive nature and low concentration. The end-products of NO metabolism, nitrite and nitrate, are much more stable and can be used to reflect the systemic NO production [
This investigation was based on the ESTHER cohort study (Epidemiologische Studie zu Chancen der Verhütung, Früherkennung und optimierten Therapie chronischer Erkrankungen in der älteren Bevölkerung [German]), which is a population-based, longitudinal study, with repeated investigation of inhabitants in the German federal state Saarland. The details of the ESTHER cohort study have been reported elsewhere [
Information on lung cancer was provided by the Saarland Cancer Registry up to the end of the year 2014. Linkage of ESTHER participants with data of the Saarland Cancer Registry was possible for 99.7% of the cohort’s participants. Lung cancer cases were ascertained according to the 10th revision of the International Statistical Classification of Diseases (ICD-10) code C34.
From 252 incident lung cancer cases, during a mean follow-up time of 13.4 years, 245 could be included in the present analysis because they donated a blood sample and a urine sample at baseline of the ESTHER cohort. Each of the lung cancer cases was matched with three controls from the same cohort on sex, age (±5 years), smoking status (never/former/current smoker), and pack-years of smoking (±10 years).
At baseline, a blood sample and a spontaneous spot urine sample were collected by general practitioners (GPs) during the health checkup and then shipped to the study center and maintained at -80°C until further processing. Urinary concentrations of nitrite/nitrate were determined using the nitrite/nitrate Colorimetric Assay Kit of Cayman Chemical (Ann Arbor, Michigan, USA). This method detects the sum of nitrate and nitrite. Urine samples were used directly after dilution to a proper concentration (1 : 5, 1 : 25, or 1 : 50 depending on the levels of nitrite/nitrate in the urine sample). For renal function adjustment of the spot urine samples, urinary creatinine was determined by the kinetic Jaffe method. In addition, the acute-phase inflammatory protein C-reactive protein (CRP) was measured in serum samples by immunoturbidimetry with the wrCRP antibody (Bayer, Leverkusen, Germany) on the ADVIA 2400. Furthermore, the levels of an established oxidative stress marker, urinary 8-isoprostane, were determined by the 8iso1 ELISA kit from Detroit R&D (Detroit, Michigan, USA).
Information on sociodemographic characteristics, including age, sex, smoking status, pack-years of smoking, education, physical activity, vegetable consumption, meat consumption, family history of lung cancer, and individual history of asthma, was collected by a standardized self-administered questionnaire. To calculate body mass index (BMI), height and weight were measured by the GPs during the health checkup and documented on a standardized form.
Baseline characteristics of cases and controls were expressed as medians (interquartile ranges) or proportions. Differences between the two groups were determined by the Wilcoxon tests for continuous variables and by the chi-square tests for categorical variables. To assess the determinants of nitrite/nitrate concentrations, distributions of nitrite/nitrate concentrations across categories of baseline characteristics were compared using the Wilcoxon-Mann-Whitney tests.
To evaluate the association of nitrite/nitrate levels and lung cancer incidence, conditional logistic regression was used to compute odds ratios (ORs) and 95% confidence intervals (CIs). The main model was adjusted for body mass index (BMI), education, family history of lung cancer, asthma prevalence, physical activity, and vegetable and meat consumption. In sensitivity models, the inflammatory marker CRP and the oxidative stress marker 8-isoprostane were additionally added to the main model. In order to address possible reverse causality, a sensitivity analysis was conducted, in which cancer cases that occurred in the first 5 years of follow-up were excluded. A dose-response analysis was conducted using restricted cubic spline (RCS) functions with five knots at the 10th, 30th, 50th, 70th, and 90th percentiles of the nitrite/nitrate distribution.
Multiple imputation was applied to impute covariates with missing values [
All statistical tests were two-sided using a significant level of 0.05. All analyses were performed with the Statistical Analysis System (SAS) version 9.4 (SAS Institute Inc., Cary, NC).
Table
Baseline characteristics of the incident lung cancer cases and matched controls, the ESTHER study (2000-2014).
Characteristics | Incident lung cancer cases | Controls | |||||
---|---|---|---|---|---|---|---|
% | Median (IQR) | % | Median (IQR) | ||||
Age (years) | 245 | — | 62 (59-68) | 735 | — | 63 (59-68) | 0.885 |
Sex | |||||||
Female | 75 | 30.6 | 226 | 30.7 | — | ||
Male | 170 | 69.4 | 509 | 69.3 | — | ||
Smoking status | 0.937 | ||||||
Never smoker | 29 | 12.1 | 93 | 13.0 | — | ||
Former smoker | 87 | 36.2 | 260 | 36.2 | — | ||
Current smoker | 124 | 51.7 | 365 | 50.8 | — | ||
Pack-years of smoking | 221 | — | 34.8 (19.3-48.0) | 624 | — | 33.8 (14.0-47.4) | 0.423 |
School education (years) | 0.077 | ||||||
≤9 | 198 | 83.2 | — | 551 | 77.0 | — | |
10-11 | 19 | 8.0 | — | 82 | 11.4 | — | |
≥12 | 21 | 8.8 | — | 83 | 11.6 | — | |
Physical activity | 0.182 | ||||||
Inactive | 67 | 27.6 | — | 165 | 22.5 | — | |
Sedentary | 110 | 45.3 | — | 351 | 48.0 | — | |
Vigorously active | 66 | 27.1 | — | 216 | 29.5 | — | |
BMI (kg/m2) | |||||||
<25 | 83 | 33.9 | — | 202 | 27.0 | — | |
25 - <30 | 103 | 42.0 | — | 330 | 44.2 | — | |
≥30 | 59 | 24.1 | — | 215 | 28.8 | — | |
Meat consumption | 0.711 | ||||||
<once/week | 81 | 36.0 | — | 259 | 37.4 | — | |
Once/week | 133 | 54.2 | — | 369 | 53.3 | — | |
>once/week | 22 | 9.8 | — | 65 | 9.4 | — | |
Vegetable consumption | 0.338 | ||||||
<once/week | 25 | 10.6 | — | 110 | 15.6 | — | |
Once/week | 160 | 68.1 | — | 442 | 62.5 | — | |
>once/week | 50 | 21.3 | — | 155 | 21.9 | — | |
Asthma | 0.292 | ||||||
Yes | 21 | 8.9 | — | 49 | 6.8 | — | |
No | 216 | 91.1 | — | 671 | 93.2 | — | |
Family history of lung cancer | 0.709 | ||||||
Yes | 27 | 11.4 | — | 76 | 10.5 | — | |
No | 210 | 88.6 | — | 646 | 89.5 | — | |
CRP (mg/L) | 243 | — | 2.6 (1.3-5.8) | 729 | — | 2.2 (1.1-5.1) | 0.164 |
8-Isoprostane (nmol/mmol creatinine) | 240 | 0.25 (0.18-0.33) | 729 | 0.23 (0.17-0.31) | 0.059 | ||
Nitrite/nitrate ( |
245 | — | 122 (80-206) | 733 | — | 114 (75-170) | 0.081 |
Abbreviation: BMI: body mass index; CRP: C-reactive protein. Note: lung cancer cases and controls were 1 : 3 matched for age, sex, smoking status, and pack-years of smoking.
Table
Median (IQR) of a nitrite/nitrate concentration according to population characteristics in cases and controls, the ESTHER study (2000-2014).
Characteristics | Cases | Controls | ||
---|---|---|---|---|
Nitrite/nitrate ( |
Nitrite/nitrate ( | |||
Age (year) | ||||
50-60 | 79 | 122.1 (85.2-202.1) | 245 | 122.0 (88.0-182.9) |
60-64 | 62 | 130.3 (79.4-256.5) | 184 | 120.9 (80.4-187.2) |
65-69 | 60 | 103.2 (69.7-160.6) | 176 | 98.9 (64.9-158.8) |
70-75 | 44 | 136.2 (88.7-237.5) | 130 | 97.1 (62.7-147.0) |
| 0.323 | |||
Sex | ||||
Female | 75 | 121.5 (67.8-227.0) | 226 | 117.6 (78.8-184.0) |
Male | 170 | 120.8 (81.4-181.7) | 509 | 111.2 (71.9-164.3) |
| 0.595 | 0.199 | ||
Smoking status | ||||
Never smoker | 29 | 104.7 (60.4-125.5) | 93 | 108.9 (74.7-146.3) |
Former smoker | 87 | 112.6 (70.0-172.7) | 260 | 102.2 (64.8-151.4) |
Current smoker | 124 | 139.3 (92.8-231.1) | 365 | 127.2 (87.0-188.7) |
| ||||
Pack-years of smoking | ||||
≤15.0 | 73 | 107.1 (60.4-167.1) | 274 | 106.9 (71.8-158.1) |
15.0 - ≤34.0 | 59 | 117.7 (69.4-204.8) | 160 | 114.5 (74.6-162.5) |
34.0 - ≤47.5 | 56 | 141.7 (89.5-198.6) | 148 | 114.6 (75.9-171.7) |
>47.5 | 57 | 136.5 (94.5-248.1) | 153 | 122.0 (79.4-197.1) |
| 0.071 | 0.105 | ||
Education levels (years) | ||||
≤9 | 198 | 122.0 (78.7-195.8) | 551 | 113.7 (74.1-168.6) |
10-11 | 19 | 112.6 (79.9-235.3) | 82 | 112.7 (67.7-163.5) |
≥12 | 21 | 145.7 (93.4-250.4) | 83 | 122.7 (92.6-211.9) |
| 0.788 | 0.103 | ||
Physical activity | ||||
Inactive | 67 | 121.6 (77.9-193.6) | 165 | 118.6 (76.7-183.5) |
Sedentary | 110 | 123.7 (78.7-227.2) | 351 | 110.8 (72.5-170.6) |
Vigorously active | 66 | 121.2 (94.1-216.0) | 216 | 118.3 (74.6-167.7) |
| 0.846 | 0.510 | ||
BMI (kg/m2) | ||||
<25 | 83 | 135.8 (85.9-222.0) | 199 | 118.4 (83.2-168.8) |
25 - <30 | 103 | 134.4 (86.4-229.2) | 325 | 114.3 (72.1-173.8) |
≥30 | 59 | 98.7 (58.6-170.7) | 211 | 105.0 (70.4-162.5) |
| 0.337 | |||
Meat intake frequency | ||||
<once/week | 81 | 125.5 (85.0-191.6) | 259 | 114.3 (75.2-177.2) |
Once/week | 122 | 112.9 (79.4-207.5) | 369 | 113.8 (74.3-165.7) |
>once/week | 22 | 121.8 (64.6-195.8) | 65 | 120.5 (76.2-170.0) |
| 0.933 | 0.877 | ||
Vegetable consumption frequency | ||||
<once/week | 25 | 108.5 (71.3-181.7) | 108 | 111.6 (75.7-152.3) |
Once/week | 160 | 125.4 (86.1-202.0) | 435 | 113.8 (72.5-163.8) |
>once/week | 50 | 105.8 (67.0-188.3) | 152 | 134.6 (80.8-227.3) |
| 0.240 | 0.134 | ||
Asthma prevalence | ||||
No | 216 | 121.5 (79.4-200.6) | 671 | 114.0 (74.7-170.0) |
Yes | 21 | 107.2 (80.7-229.7) | 49 | 112.7 (78.9-147.0) |
| 0.520 | 0.921 | ||
Family history of lung cancer | ||||
No | 210 | 121.2 (79.9-217.7) | 646 | 113.9 (74.7-168.7) |
Yes | 27 | 132.3 (71.3-191.6) | 76 | 111.3 (72.0-186.0) |
| 0.870 | 0.907 | ||
CRP (mg/L) | ||||
≤1.175 | 52 | 114.3 (63.5-236.5) | 191 | 118.0 (78.1-178.3) |
1.175 - ≤2.325 | 59 | 148.9 (85.9-229.7) | 184 | 108.9 (70.8-162.2) |
2.325 - ≤5.140 | 67 | 113.0 (89.4-207.5) | 177 | 111.2 (78.2-165.7) |
>5.140 | 65 | 118.1 (76.0-171.5) | 177 | 114.8 (70.3-174.6) |
| 0.520 | 0.739 | ||
8-Isoprostane (nmol/mmol creatinine) | ||||
≤0.175 | 54 | 108.3 (69.4-158.0) | 189 | 105.0 (70.5-155.7) |
0.174 - ≤0.231 | 54 | 112.6 (79.5-193.6) | 188 | 108.4 (73.4-162.5) |
0.231 - ≤0.308 | 59 | 107.9 (67.1-176.9) | 182 | 114.1 (75.7-175.5) |
>0.308 | 73 | 167.0 (112.0-267.3) | 170 | 138.9 (88.3-195.9) |
|
Abbreviations: BMI: body mass index; CRP: C-reactive protein.
Table
Associations of nitrite/nitrate concentration quintiles with lung cancer incidence, the ESTHER study (2000-2014).
Nitrite/nitrate levels ( |
Main modela |
Sensitivity model 1b |
Sensitivity model 2c | ||
---|---|---|---|---|---|
Quintile 1 | <66.9 | 48/147 | Ref. | Ref. | Ref. |
Quintile 2 | 66.9 - <97.2 | 39/147 | 0.81 (0.59-1.12) | 0.81 (0.59-1.12) | 0.82 (0.60-1.12) |
Quintile 3 | 97.2 - <134.1 | 43/147 | 0.88 (0.65-1.20) | 0.89 (0.65-1.21) | 0.88 (0.65-1.20) |
Quintile 4 | 134.1 - <192.8 | 48/147 | 1.00 (0.74-1.36) | 1.01 (0.75-1.37) | 1.01 (0.74-1.36) |
Quintile 5 | ≥192.8 | 67/147 |
aAdjusted for body mass index (BMI), education, family history of lung cancer, asthma, physical activity, and vegetable and meat consumption frequency. In addition, potential confounding by the following factors was controlled by matching age, sex, smoking status, and pack-years of smoking. bAdjusted for variables of the main model+C-reactive protein. In addition, potential confounding by the following factors was controlled by matching age, sex, smoking status, and pack-years of smoking. cAdjusted for variables of the main model+8-isoprostane. In addition, potential confounding by the following factors was controlled by matching age, sex, smoking status, and pack-years of smoking. Note: numbers in bold: statistically significant estimates compared to the quintile 1 (
Associations of a nitrite/nitrate concentration with lung cancer incidence in a sensitivity analyses excluding lung cancer cases which occurred in the first 5 years of follow-up, the ESTHER study (2000-2014).
Nitrite/nitrate levels ( |
Main modela |
Sensitivity model 1b |
Sensitivity model 2c | ||
---|---|---|---|---|---|
Quintile 1 | <67.7 | 37/101 | Ref. | Ref. | Ref. |
Quintile 2 | 67.7 - <97.4 | 29/101 | 0.88 (0.60-1.29) | 0.88 (0.602-1.29) | 0.88 (0.60-1.29) |
Quintile 3 | 97.4 - <133.7 | 31/101 | 0.95 (0.65-1.38) | 0.95 (0.66-1.39) | 0.95 (0.66-1.39) |
Quintile 4 | 133.7 - <188.8 | 26/101 | 0.81 (0.55-1.18) | 0.80 (0.55-1.18) | 0.81 (0.55-1.19) |
Quintile 5 | ≥188.77 | 45/100 | 1.36 (0.97-1.92) | 1.36 (0.97-1.92) | 1.34 (0.95-1.90) |
aAdjusted for body mass index (BMI), education, family history of lung cancer, asthma, physical activity, and vegetable and meat consumption frequency. In addition, potential confounding by the following factors was controlled by matching age, sex, smoking status, and pack-years of smoking. bAdjusted for variables of the main model+C-reactive protein (CRP). In addition, potential confounding by the following factors was controlled by matching age, sex, smoking status, and pack-years of smoking. cAdjusted for variables of the main model+8-isoprostane. In addition, potential confounding by the following factors was controlled by matching age, sex, smoking status, and pack-years of smoking.
The dose-response relationship between nitrite/nitrate concentrations and lung cancer incidence is presented in Figure
Dose-response relationship of a nitrite/nitrate concentration and lung cancer incidence, the ESTHER study (2000-2014). Notes: results of a logistic regression model adjusted for BMI, education, family history of lung cancer, asthma, physical activity, and vegetable and meat consumption frequency. In addition, potential confounding by the following factors was controlled by matching age, sex, smoking status, and pack-years of smoking. Knots: 10th, 30th, 50th, 70th, and 90th percentile. Solid red line: estimation for the odds ratio. Dashed grey lines: 95% confidence interval bands. Dashed green line:
In this prospective matched case-control study from Germany, we investigated the determinants of urinary nitrite/nitrate levels and the association between nitrite/nitrate levels and lung cancer incidence. We observed that high nitrite/nitrate levels were associated with lower age in controls and a lower BMI among cases. Current smoking and high 8-isoprostane levels were associated with high nitrite/nitrate levels in both groups. Furthermore, the association of nitrite/nitrate levels was not linear, and only participants with nitrite/nitrate levels greater than approximately 200
Previous studies have demonstrated that NO is a mediator and regulator in inflammatory responses. Whereas low NO is anti-inflammatory, excessively elevated NO promotes inflammation and oxidative stress under pathological conditions [
Figure
Schematic illustration of observed associations of nitrite/nitrate, 8-isoprostane, and CRP concentrations with lung cancer incidence. Abbreviation: CRP: C-reactive protein; iNOS: inducible nitric oxide synthase; ROS: reactive oxygen species.
The following mechanisms could explain the observed association of nitrite/nitrate concentrations in urine with lung cancer incidence. NO can react with other reactive oxygen species (ROS) which is converted to reactive nitrogen species (RNS), which are subsequently metabolized to nitrite and nitrate [
To the best of our knowledge, the associations of NO metabolites and the risk of lung or any other cancer site have not been investigated by prospective, population-based studies before. However, a case-control study nested in a prostate cancer cohort from Sweden observed that high compared to low/negative iNOS immunoreactivity in prostate tumor epithelial cells was associated with a strongly increased 10-year prostate cancer mortality (OR (95% CI), 3.80 (1.45-9.97)) [
In addition to 8-isoprostane levels, nitrite/nitrate levels were associated with current smoking and a low BMI among cases. This might be explained by the long latency period of lung cancer, and some lung cancer cases diagnosed during follow-up may have already been subclinical at the baseline examination. Lung cancer patients often experience loss of appetite and weight for a long period of time before they are diagnosed [
Furthermore, we observed that nitrite/nitrate levels were inversely associated with age in the control group, which confirms previous observations in healthy individuals [
It has been shown that NO has pathophysiological effects on asthma and COPD. Elevated NO can lead to nitrosative stress in the airway epithelium, which may be responsible for steroid resistance or ineffectiveness in inflammatory pulmonary diseases [
Our study has several strengths. First, a prospective matched case-control study design was used, and subjects with a history of lung cancer before baseline were excluded from the analyses to avoid reverse causality. In addition, cancer cases diagnosed within the first 5 years of follow-up were excluded in a sensitivity analysis to address the lag time of lung cancer development, and the results were consistent with the main results. Second, linkage to cancer registries ensured high certainty regarding lung cancer diagnoses and minimized attrition bias, which often affects cohort studies with long-term follow-up. Third, in order to control for confounding, cases were matched to controls for age, sex, smoking status, and pack-years of smoking, and models were comprehensively adjusted for other potential confounders, including 8-isoprostane levels and CRP. Nevertheless, several limitations in our study need to be considered when interpreting the results. First, the sampling of controls reduces precision and power compared to a cohort design. Second, residual confounding can generally not be excluded in an observational study. Third, the number of cases in each quintile was relatively small and led to a rather low statistical power for quintile comparisons. Fourth, urinary nitrite/nitrate concentrations may also be influenced by a nitrate-rich diet [
In conclusion, the current study observed that high urinary nitrite/nitrate levels were associated with high lung cancer incidence although the comparison between cases and controls was matched for age, sex, and smoking and controlled for other biomarkers of oxidative stress and inflammation. This suggests an independent mechanism that links pathologically high levels of NO to lung cancer development.
The total ESTHER study data cannot be made freely available due to data security regulations. Requests for access to the data used for this publication can be made to the corresponding author, Dr. Ben Schöttker (
The authors have no conflicts of interest to disclose.
This study was funded by a grant from the German Research Foundation (grant No. SCHO 1545/3-1) and by the China Scholarship Council (grant No. 201506010268 to Xīn Gào). The ESTHER study was funded by grants from the Ministry of Science, Research and the Arts Baden-Württemberg (Stuttgart, Germany), the German Federal Ministry of Education and Research (Berlin, Germany), the Federal Ministry of Family Affairs, Senior Citizens, Women and Youth (Berlin, Germany), and the Saarland State Ministry of Social Affairs, Health, Women and Family. The authors gratefully thank the study participants and their general practitioners as well as the laboratory and administrative staff of the ESTHER study team. The authors also gratefully acknowledge Prudence R. Carr for her effort to improve the English of the manuscript as well as her comments on the manuscript.