Abstract
Background
We previously reported that higher levels of mitochondrial DNA copy number (mtDNA CN) were associated with lung cancer risk among male heavy smokers (i.e., ≥20 cigarettes per day) in the Alpha-Tocopherol Beta-Carotene (ATBC) study. Here, we present two additional prospective investigations nested in the Prostate, Lung, Colorectal, and Ovarian (PLCO) cancer screening trial and the Shanghai Women’s Health Study (SWHS), and pooled with previously published data from ATBC.
Materials
All DNA was extracted from peripheral whole blood samples using the phenol–chloroform method, and mtDNA CN was assayed by fluorescence-based quantitative polymerase chain reaction. Multivariate unconditional logistic regression models were used to estimate odds ratios (OR) and 95% confidence intervals (95% CI) for the association of mtDNA CN and lung cancer risk.
Results
Overall, mtDNA CN was not associated with lung cancer risk in the PLCO, SWHS, or pooled populations (all P-trends > 0.42, P-heterogeneity = 0.0001), and mtDNA CN was inversely associated with lung cancer risk among male smokers in PLCO, the opposite direction observed in ATBC. Additionally, the mtDNA CN association observed among male heavy smokers in ATBC was the opposite direction in PLCO.
Conclusions
mtDNA CN was not consistently associated with lung cancer risk across three prospective study populations from Europe, Asia, and the US.
Impact
This pooled study suggests no consistent association between pre-diagnostic mtDNA CN levels and lung cancer risk across several populations.
Keywords: Mitochondrial DNA copy number, lung cancer, cohorts, pooled, nested case-control
Introduction
Mitochondria are primarily responsible for energy production in eukaryotic cells (1). Mitochondria have a singular circular mitochondrial DNA molecule. Mitochondria copy number (mtDNA CN) varies to meet energy needs and cope with oxidative stress (2). Oxidative stress disrupts homeostasis and forms reactive oxygen species leading to DNA damage and genomic instability (3). In a prospective study among Finnish male smokers, we observed a positive association between mtDNA CN in peripheral blood leukocytes and lung cancer risk (4).
To replicate the mtDNA CN-lung cancer risk association, nested case-control studies were conducted in two prospective studies, the Prostate, Lung, Colorectal, and Ovarian (PLCO) cancer screening trial, a population of men and women across the US, and the Shanghai Women’s Health Study (SWHS), a population of mostly non-smoking Chinese women. These data were then pooled with published data from the Alpha-Tocopherol Beta-Carotene (ATBC) study (4).
Materials and Methods
Study Subjects
The ATBC (5), PLCO (6), and SWHS (7) studies have been described in detail. From ATBC, a total of 229 incident cases with an available blood specimen for analysis were previously identified (4). Controls were matched to cases 1:1 based on date of birth. From PLCO, a total of 442 cases of first primary incident lung cancer with serum at baseline from the screening arm were identified. Controls were matched by age, sex, race, blood collection date, and diagnosing center. From SWHS, 226 incident lung cancer cases with adequate blood for mtDNA analysis were identified, with one control matched to case by age, smoking status, and date of blood collection.
Laboratory analysis
In all three studies, DNA was extracted from peripheral whole blood by the phenol–chloroform method. mtDNA CN was assayed in the same lab by quantitative polymerase chain reaction using the ratio of the estimated threshold cycle number of ND1 mitochondrial gene, and the B-globin nuclear gene, HBB (8). The coefficient of variation for PLCO, SWHS, and ATBC was 14%, 7%, and 13%, respectively. After excluding bad runs, poor quality DNA, or subjects with missing demographic variables, 227 cases and 227 controls in ATBC, 426 cases and 436 controls in PLCO, and 221 cases and 222 controls in SWHS were included in the final analysis.
Statistical analysis
Differences between cases and controls for demographics characteristics were tested with Wilcoxon-signed rank sum test for continuous variables and Pearson chi-square test for categorical variables. mtDNA CN was categorized by quartiles among controls in each study. Unconditional logistic regression models generated odds ratios (OR) and 95% confidence intervals (95% CI) to estimate the association of mtDNA CN and risk of lung cancer. Models were adjusted for age, body mass index (BMI), race, pack-years smoking, and date of blood collection. PLCO was additionally adjusted for sex and study center. Pooled analyses were additionally adjusted for study. Additional analyses were stratified by sex, smoking status, and smoking pack-years. P-trend modeled mtDNA CN quartiles continuously. Between-study heterogeneity was tested by random-effects with the rmeta package (Thomas Lumley (2012). rmeta: Meta-analysis. R package version 2.16. http://CRAN.R-project.org/package=rmeta). Analyses were performed in SAS 9.3 (SAS Institute, Cary, NC).
Results
Selected demographics characteristics of the study population are described in Table 1. Statistically significant differences between cases and controls in BMI and years of smoking were observed in the ATBC and PLCO populations but not in SWHS. Age, sex, race (all Caucasian in ATBC, Asian in SWHS), and mtDNA CN levels were not different between cases and controls across the three populations.
Table 1.
Demographic and mtDNA CN characteristics of ATBC, PLCO, and SWHS populations
|
|
|
Cases |
Controls |
|
||
|---|---|---|---|---|---|---|
| Study (Ca/Co) | Characteristic | Mean/N | SD/% | Mean/N | SD/% | P-value* |
| ATBC (227/227) | BMI (kg/m2) | 25.59 | 3.49 | 26.35 | 3.87 | 0.045 |
| Age | 56.67 | 5.00 | 58.41 | 4.79 | 0.68 | |
| Years Smoking | 38.47 | 7.11 | 35.83 | 9.22 | 0.005 | |
| Ever Smoker | 227 | 100% | 227 | 100% | 1 | |
| mtDNA CN | 134.23 | 45.27 | 127.85 | 36.07 | 0.26 | |
| median, IQR | 125.86 | (102.04–160.18) | 124.79 | (100.63–151.51) | ||
| PLCO (426/436) | BMI (kg/m2) | 26.84 | 4.39 | 27.43 | 4.36 | 0.028 |
| Age | 64.07 | 4.95 | 63.67 | 4.71 | 0.22 | |
| Years Smoking | 34.89 | 15.15 | 15.57 | 17.47 | <0.0001 | |
| Ever Smoker | 380 | 89.20% | 238 | 54.59% | <0.0001 | |
| mtDNA CN | 128.37 | 128.26 | 127.39 | 59.96 | 0.079 | |
| median, IQR | 110 | (87–140) | 114 | (90–145) | ||
| Sex | ||||||
| Male | 259 | 60.80% | 267 | 61.24% | 0.89 | |
| Female | 167 | 39.20% | 169 | 38.76% | ||
| Race | ||||||
| White | 399 | 93.66% | 404 | 92.66% | 0.81 | |
| Black | 15 | 3.52% | 18 | 4.13% | ||
| Other | 12 | 2.82% | 14 | 3.21% | ||
| Center | ||||||
| University of Colorado | 23 | 5.40% | 23 | 5.28% | 1 | |
| Georgetown University | 19 | 4.46% | 19 | 4.36% | ||
| Pacific Health Research and Education Institute | 9 | 2.11% | 9 | 2.06% | ||
| Henry Ford Health System | 66 | 15.49% | 63 | 14.45% | ||
| University of Minnesota | 126 | 29.58% | 131 | 30.05% | ||
| Washington University | 50 | 11.74% | 50 | 11.47% | ||
| University of Pittsburgh | 43 | 10.09% | 46 | 10.55% | ||
| University of Utah | 29 | 6.81% | 31 | 7.11% | ||
| Marshfield Clinic Research Foundation | 45 | 10.56% | 47 | 10.78% | ||
| University of Alabama Birmingham | 16 | 3.76% | 17 | 3.90% | ||
| SWHS (221/222) | BMI (kg/m2) | 24.58 | 3.45 | 25.01 | 3.53 | 0.16 |
| Age | 59.22 | 8.20 | 59.18 | 8.35 | 0.93 | |
| Years Smoking | 2.97 | 10.67 | 1.74 | 8.19 | 0.23 | |
| Ever Smoker | 17 | 7.69% | 11 | 4.95% | 0.24 | |
| mtDNA CN | 191.84 | 315.45 | 185.88 | 268.6 | 0.82 | |
| median, IQR | 118 | (94–158) | 124 | (93–162) | ||
Abbreviations: Alpha-Tocopherol, Beta-Carotene study (ATBC), Prostate Lung Colorectal Ovarian study (PLCO), Shanghai Women’s Health Study (SWHS), body mass index (BMI), mitochondrial DNA copy number (mtDNA CN), interquartile range (IQR, standard deviation (SD) cases (Ca), controls (Co)
Wilcoxon rank sum test for continuous variables/Pearson chi-square for categorical variables
mtDNA CN was not associated with lung cancer risk in the PLCO or SWHS data; pooled analysis was also null with evidence of heterogeneity across studies (P-heterogeneity = 0.0001) (Table 2). mtDNA CN and lung cancer risk associations were inverse in male smokers in PLCO, the opposite direction of ATBC. Similarly, comparing heavy smokers to non/light smokers, associations in PLCO were the opposite direction of ATBC (data not shown). No differential associations were observed by histology or follow-up time (data not shown).
Table 2.
mtDNA copy number and risk of lung cancer across ATBC, PLCO, and SWHS studies by sex and smoking status
| ATBC (Ca 227/ Co 227)
|
PLCO (Ca 426/Co 436)
|
SWHS (Ca 221/Co 222)
|
Pooled |
|||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| mtDNA% | Ca/Co | OR* | 95% CI | Ca/Co | OR* | 95% CI | Ca/Co | OR* | 95% CI | Ca/Co | OR* | 95% CI |
| Overall | ||||||||||||
| Q1 | 52/57 | 1.00 | (ref) | 117/107 | 1 | (ref) | 54/56 | 1.00 | (ref) | 223/220 | 1.00 | (ref) |
| Q2 | 59/56 | 1.34 | (0.71–2.37) | 119/110 | 0.98 | (0.64–1.50) | 72/58 | 1.27 | (0.48–1.47) | 250/224 | 1.08 | (0.82–1.42) |
| Q3 | 43/57 | 1.15 | (0.61–2.17) | 96/108 | 0.84 | (0.55–1.29) | 44/53 | 0.84 | (0.48–1.47) | 183/218 | 0.83 | (0.62–1.11) |
| Q4 | 73/57 | 2.38 | (1.12–5.06) | 94/111 | 0.98 | (0.63–1.52) | 51/55 | 0.91 | (0.53–1.57) | 218/223 | 1.04 | (0.79–1.39) |
| P-trend | 0.049 | P-trend | 0.74 | P-trend | 0.42 | P-trend | 0.78 | |||||
| Female | ||||||||||||
| Q1 | - | - | 32/40 | 1 | (ref) | 54/56 | 1.00 | (ref) | 86/96 | 1.00 | (ref) | |
| Q2 | - | - | 38/41 | 1.51 | (0.74–3.11) | 72/58 | 1.27 | (0.48–1.47) | 110/99 | 1.34 | (0.89–2.02) | |
| Q3 | - | - | 41/44 | 1.47 | (0.73–2.98) | 44/53 | 0.84 | (0.48–1.47) | 85/97 | 1.03 | (0.67–1.58) | |
| Q4 | - | - | 56/44 | 2.12 | (1.07–4.22) | 51/55 | 0.91 | (0.53–1.57) | 107/99 | 1.24 | (0.82–1.89) | |
| P-trend | 0.043 | P-trend | 0.42 | P-trend | 0.59 | |||||||
| Male | ||||||||||||
| Q1 | 52/57 | 1.00 | (ref) | 85/67 | 1 | (ref) | - | - | 137/125 | 1.00 | (ref) | |
| Q2 | 59/56 | 1.34 | (0.71–2.37) | 81/69 | 0.76 | (0.44–1.28) | - | - | 140/125 | 0.89 | (0.61–1.29) | |
| Q3 | 43/57 | 1.15 | (0.61–2.17) | 55/64 | 0.59 | (0.34–1.05) | - | - | 98/121 | 0.69 | (0.46–1.02) | |
| Q4 | 73/57 | 2.38 | (1.12–5.06) | 38/67 | 0.55 | (0.29–0.99) | - | - | 111/124 | 0.92 | (0.62–1.35) | |
| P-trend | 0.049 | P-trend | 0.028 | P-trend | 0.39 | |||||||
| Never Smoker | ||||||||||||
| Q1 | - | - | 9/51 | 1 | (ref) | 52/53 | 1.00 | (ref) | 61/104 | 1.00 | (ref) | |
| Q2 | - | - | 11/51 | 1.35 | (0.49–3.72) | 66/58 | 1.15 | (0.68–1.94) | 77/109 | 1.17 | (0.74–1.85) | |
| Q3 | - | - | 12/42 | 1.23 | (0.43–3.57) | 40/51 | 0.79 | (0.44–1.40) | 51/93 | 0.86 | (0.52–1.41) | |
| Q4 | - | - | 14/54 | 1.49 | (0.55–4.08) | 46/49 | 0.94 | (0.54–1.65) | 60/103 | 1.03 | (0.64–1.68) | |
| P-trend | 0.49 | P-trend | 0.54 | P-trend | 0.79 | |||||||
| Ever Smoker | ||||||||||||
| Q1 | 52/57 | 1.00 | (ref) | 108/56 | 1 | (ref) | 2/3 | 1.00 | (ref) | 162/116 | 1.00 | (ref) |
| Q2 | 59/56 | 1.34 | (0.71–2.37) | 108/59 | 0.91 | (0.56–1.47) | 6/0 | - | - | 173/115 | 1.03 | (0.73–1.47) |
| Q3 | 43/57 | 1.15 | (0.61–2.17) | 84/66 | 0.72 | (0.44–1.17) | 4/2 | 4.33 | (0.24–79.49) | 132/125 | 0.78 | (0.54–1.12) |
| Q4 | 73/57 | 2.38 | (1.12–5.06) | 80/57 | 0.84 | (0.50–1.39) | 5/6 | 1.05 | (0.087–12.81) | 158/120 | 1.05 | (0.73–1.49) |
| P-trend | 0.049 | P-trend | 0.33 | P-trend | 0.46 | P-trend | 0.81 | |||||
| Male Smoker | ||||||||||||
| Q1 | 52/57 | 1.00 | (ref) | 78/38 | 1 | (ref) | - | - | 130/95 | 1 | (ref) | |
| Q2 | 59/56 | 1.34 | (0.71–2.37) | 78/40 | 0.8 | (0.45–1.45) | - | - | 137/96 | 0.94 | (0.63–1.39) | |
| Q3 | 43/57 | 1.15 | (0.61–2.17) | 53/45 | 0.58 | (0.32–1.06) | - | - | 96/102 | 0.69 | (0.46–1.03) | |
| Q4 | 73/57 | 2.38 | (1.12–5.06) | 33/37 | 0.50 | (0.25–0.98) | - | - | 106/94 | 0.95 | (0.63–1.43) | |
| P-trend | 0.049 | P-trend | 0.021 | P-trend | 0.45 | |||||||
Alpha-Tocopherol Beta-Carotene (ATBC), Prostate, Lung, Colorectal, Ovarian (PLCO), Shanghai Women’s Health Study (SWHS), Cases (Ca), Controls (Co), Odds Ratio (OR), 95% Confidence Interval (95% CI), Quartile (Q1–Q4), Median (M1, M2)
%Quartiles: ATBC: <100.6, ≥100.6–124.8, ≥124.8–151.5, ≥151.5; PLCO: <90, ≥90–114, ≥114–145, ≥145; SWHS: <93, ≥93–124, ≥124–162, ≥162; Median: ATBC: ≤124.8; PLCO: ≤114, SWHS: <124
Adjusted for age, BMI, pack-years, race, sex, date of enrollment, center (if applicable), study
Discussion
The mtDNA CN and lung cancer risk association observed in ATBC did not replicate in PLCO, SWHS, and pooled study populations. There was no consistent evidence of an association across populations by sex or smoking status/intensity. This pooled study suggests no consistent association between pre-diagnostic mtDNA CN levels and lung cancer risk across several populations.
Our study included diverse study populations. Additional strengths included the combined large sample size and the standardized specimen processing and mtDNA CN assay across the three studies. The primary weakness of this study was the single measurement of mtDNA CN. Copy number could change over time, and this study was unable to determine intrapersonal variation over time.
Acknowledgments
Financial support: NIH intramural research program
Footnotes
Conflicts of interest: None
References
- 1.Hatefi Y. The mitochondrial electron transport and oxidative phosphorylation system. Annual review of biochemistry. 1985;54:1015–69. doi: 10.1146/annurev.bi.54.070185.005055. [DOI] [PubMed] [Google Scholar]
- 2.Lee HC, Yin PH, Lu CY, Chi CW, Wei YH. Increase of mitochondria and mitochondrial DNA in response to oxidative stress in human cells. The Biochemical journal. 2000;348(Pt 2):425–32. [PMC free article] [PubMed] [Google Scholar]
- 3.Lawless MW, O’Byrne KJ, Gray SG. Oxidative stress induced lung cancer and COPD: opportunities for epigenetic therapy. Journal of cellular and molecular medicine. 2009;13:2800–21. doi: 10.1111/j.1582-4934.2009.00845.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Hosgood HD, 3rd, Liu CS, Rothman N, Weinstein SJ, Bonner MR, Shen M, et al. Mitochondrial DNA copy number and lung cancer risk in a prospective cohort study. Carcinogenesis. 2010;31:847–9. doi: 10.1093/carcin/bgq045. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.The alpha-tocopherol, beta-carotene lung cancer prevention study: design, methods participant characteristics, and compliance. The ATBC Cancer Prevention Study Group. Annals of epidemiology. 1994;4:1–10. doi: 10.1016/1047-2797(94)90036-1. [DOI] [PubMed] [Google Scholar]
- 6.Hayes RB, Sigurdson A, Moore L, Peters U, Huang WY, Pinsky P, et al. Methods for etiologic and early marker investigations in the PLCO trial. Mutat Res-Fund Mol M. 2005;592:147–54. doi: 10.1016/j.mrfmmm.2005.06.013. [DOI] [PubMed] [Google Scholar]
- 7.Zheng W, Chow WH, Yang G, Jin F, Rothman N, Blair A, et al. The Shanghai Women’s Health Study: rationale, study design, and baseline characteristics. Am J Epidemiol. 2005;162:1123–31. doi: 10.1093/aje/kwi322. [DOI] [PubMed] [Google Scholar]
- 8.Liu CS, Tsai CS, Kuo CL, Chen HW, Lii CK, Ma YS, et al. Oxidative stress-related alteration of the copy number of mitochondrial DNA in human leukocytes. Free radical research. 2003;37:1307–17. doi: 10.1080/10715760310001621342. [DOI] [PubMed] [Google Scholar]