Title: Tobacco Carcinogen and Toxicant Biomarkers
1Tobacco Carcinogen and Toxicant Biomarkers
- Stephen S. Hecht, Ph.D.
- Masonic Cancer Center
- University of Minnesota
2Outline of Presentation
- Tobacco and cancer
- Background on tobacco carcinogen biomarkers
- Examples of tobacco carcinogen biomarkers
- Total NNAL
- Formaldehyde-DNA adducts
3Cancers Caused by SmokingIARC Monograph Series
- Volume 38, 1986
- Lung
- Oral cavity
- Pharynx
- Larynx
- Esophagus
- Pancreas
- Bladder
- Volume 83, 2004, added
- Nasal Cavity
- Stomach
- Liver
- Kidney
- Ureter
- Cervix
- Myeloid leukemia
- Volume 100E, 2009, added
- Colorectum
- Ovary (mucinous)
4Cancer Deaths Due to Smoking
- Worldwide 21 (1,420,000 per year)
- United States 33 (185,000 per year)
IARC World Cancer Report, 2008
5Tobacco Use Prevalence, 2008
- Adult smokers, U.S. 46,000,000 (20.6)
- Ex-smokers, U.S. 48,100,000 (21.5)
- Smokers, worldwide 1,300,000,000
- Smokeless, worldwide xxx,000,000
CDC, MMWR, November 13, 2009
6World Smoking Prevalence Males
O. Shafey, M. Ericksen, H. Ross, J. Mackay (2009)
The Tobacco Atlas, 3rd Ed.
7World Smoking Prevalence Females
O. Shafey, M. Ericksen, H. Ross, J. Mackay (2009)
The Tobacco Atlas, 3rd Ed.
8Overall Goal
- Elucidate mechanisms of tobacco-induced
cancer and apply this knowledge to cancer
prevention.
9Conceptual Framework for Understanding Tobacco
Carcinogenesis
S.S. Hecht, JNCI, 911194-1210 (1999), Nature
Rev. Cancer 3733-744 (2003) Cancer Principles
and Practice of Oncology, 8th Edition, 147-155
(2008)
10Significantly Mutated Genes in Lung
Adenocarcinoma Based on Sequencing of 623 Genes
in 188 Tumors
Ding et al, Nature, 4551069-1075, 2008
11Other Factors Contributing to Tobacco-Induced
Cancer
- Receptor mediated effectsnicotine, nitrosamines
- Direct activation of EGFR and COX-2
- Down-regulation of FHIT
- Hyper-methylation of tumor suppressors
- Tumor promotion and co-carcinogenesis
- Oxidative damage and inflammation
- Cilia-toxicity
H. Takahashi et al, Cancer Cell 17 89 (2010) H.
Schuller, Nature Rev. Cancer 9 195-205 (2009)
K.A. West et al, J. Clin. Invest. 111 81-90
(2003) S.A. Belinsky, Carcinogenesis 26 1481
(2005) DAgostini et al, Cancer Res 66
3936-3941 (2006) Jin et al, Carcinogenesis 29
1614-1622 (2008) Bhutani et al, Cancer Prev.
Res. 1 39-44 (2008)
12Conceptual Framework for Understanding Tobacco
Carcinogenesis
S.S. Hecht, JNCI, 911194-1210 (1999), Nature
Rev. Cancer 3733-744 (2003) Cancer Principles
and Practice of Oncology, 8th Edition, 147-155
(2008)
13IARC Carcinogens in Tobacco Smoke
S.S. Hecht, In DeVita et al, Cancer (2010) IARC
Monographs No. 83 (2004) D. Hoffmann and S.S.
Hecht, Handbook Exp. Pharmacol. 9463-102 (1990)
14Goal
- Develop and validate tobacco carcinogen and
toxicant biomarkers - Urinary metabolites
- DNA and protein adducts
- Metabolites in blood, saliva, breath, nails, hair
- Use these biomarkers to identify those smokers
susceptible to cancer.
15Outline of Presentation
- Tobacco and cancer
- Background on tobacco carcinogen biomarkers
- Examples of tobacco carcinogen biomarkers
- Total NNAL
- Formaldehyde-DNA adducts
16Definitions
- Biomarker A distinctive biological or
biologically derived indicator (as a metabolite)
of a process, event, or condition
(Merriam-Websters Collegiate Dictionary) - Tobacco carcinogen biomarker Any quantifiable
substance, such as a metabolite, that can be
specifically related to human exposure to a given
tobacco carcinogen.
17Tobacco Carcinogen Biomarkers
- DNA Adducts
- Protein Adducts
- Hemoglobin
- Albumin
- Metabolites
- Breath
- Saliva
- Nails and Hair
- Urine
- Blood
Reviewed in Carcinogenesis 23 907 and 1979
(2002) Nature Rev. Cancer 3 733 (2003)
18Applications of Tobacco Carcinogen Biomarkers
- Assessing exposure in smokers, smokeless tobacco
users, and non-smokers exposed to secondhand
smoke - Regulation of tobacco products
- Understanding mechanisms of human carcinogenesis
and identifying susceptible individuals - Not specifically designed for early detection of
cancer, but could have applications in screening
19A Panel of Tobacco Carcinogen and Toxicant
Biomarkers
Range of recent mean values (nmol/24h unless noted otherwise) Range of recent mean values (nmol/24h unless noted otherwise)
Urinary biomarkers Source Smokers Non-smokers
Nicotine equivalents Nicotine 70.4-154 µmol/24 h Not Detected
Total NNAL NNK 1.1 - 2.9 Not Detected
Total NNN NNN 0.049 - 0.24 Not Detected
1-HOP or PheT PAH 0.50 - 1.45 0.18 - 0.50
MHBMA 1,3-Butadiene 15.5 - 322 0.65 - 7.5
SPMA Benzene 3.2 - 32.1 0.17 - 3.14
HPMA Acrolein 5,869 - 11,190 1,131 - 1,847
HBMA Crotonaldehyde 1,965 - 26,000 242 - 3,200
HEMA Ethylene oxide 19.1 - 102 6.51 - 38.8
Cd Cadmium 2.3 - 12.8 1.34 - 8.04
8-epi-PGF2a Oxidative damage 1.48 - 2.80 0.62 - 1.13
PGE-M Inflammation 54 - 60 31.6 - 45.3
Based on 1.3g creatinine per 24h in smokers and
1.5g creatinine per 24h in non-smokers, or 1.5 l
urine per 24h.
S.S. Hecht, J-M Yuan, and D. Hatsukami, Chem.
Res. Toxicol., 2010
20Structures of the Urinary Biomarkers
21A Panel of Biomarkers
Recent data (pmol/g globin mean S.D.) Recent data (pmol/g globin mean S.D.)
Hemoglobin adducts Source Smokers Non-smokers
Cyanoethylvaline Acrylonitrile 112 81 6.5 6.4
Carbamoylethylvaline Acrylamide 84.1 41.8 27.8 7.1
Hydroxyethylvaline Ethylene oxide 132 92 21.1 12.7
4-Aminobiphenyl-globin 4-Aminobiphenyl 0.26 0.006a 0.067 0.009a
(fmol/µmol dN mean S.D.) (fmol/µmol dN mean S.D.)
Leukocyte DNA adducts Source Smokers Non-smokers
N6-hydroxymethyl-dAdo Formaldehyde 179 205 15.5 33.8
N2-ethylidene-dGuo Acetaldehyde 1,310 1,720 705 438
Mean concentrations Mean concentrations
Other Source Smokers Non-smokers
Exhaled CO Carbon monoxide 17.4 - 34.4 ppm 2.6 - 6.5 ppm
Carboxyhemoglobin Carbon monoxide 3.4 - 7.1 0.35 - 1.45
22Biomarker Validation
- Analytical
- Specificity, sensitivity, accuracy, precision
- With respect to tobacco
- Decreases upon cessation
- Dose-response
- With respect to cancer risk
23Persistence of Biomarkers Study
- Smokers provide baseline 24h urine samples.
- Eight days later, they quit smoking and receive
nicotine replacement therapy. - They provide 24h urine samples on days 3, 7, 14,
21, 28, 42, and 56 after quitting. - Urine samples are analyzed for mercapturic acids
(by LC-MS/MS) and other biomarkers.
S.G. Carmella, M. Chen, S. Han, A. Briggs, J.
Jensen, D. K. Hatsukami, and S. S. Hecht Chem.
Res. Toxicol., 22 734-741 (2009)
24Metabolism of 1,3-Butadiene to Mercapturic Acids
a. GSH, GSTs b. g-glutamyltranspeptidase c.
cysteinylglycine dipeptidase d. cysteine
S-conjugate N-acetyltransferase
C.L. Sprague and A.A. Elfarra, Chem. Res.
Toxicol., 17 819-826 (2004)
25Mean Urinary MHBMA Reduction Upon Smoking
Cessation, N17
26Metabolism of Acrolein, Crotonaldehyde, Benzene,
and Ethylene Oxide to Mercapturic Acids
a. GSH, GSTs b. g-glutamyltranspeptidase c.
cysteinylglycine dipeptidase d. cysteine
S-conjugate N-acetyltransferase
27Mean Urinary HPMA Reduction Upon Smoking
Cessation, N17
28Mean Urinary HBMA Reduction Upon Smoking
Cessation, N17
29Mean Urinary SPMA Reduction Upon Smoking
Cessation, N17
30Mean Urinary HEMA Reduction Upon Smoking
Cessation, N17
31Structures of Urinary Biomarkers
S.G. Carmella, et al, Chem. Res. Toxicol., 22
734-741 (2009)
32Mean Urinary 1-HOP Reduction Upon Smoking
Cessation, N15
33Mean Urinary Total NNAL Reduction Upon Smoking
Cessation, N17
34Outline of Presentation
- Tobacco and cancer
- Background on tobacco carcinogen biomarkers
- Examples of tobacco carcinogen biomarkers
- Total NNAL
- Formaldehyde-DNA adducts
35A Panel of Tobacco Carcinogen and Toxicant
Biomarkers
Range of recent mean values (nmol/24h unless noted otherwise) Range of recent mean values (nmol/24h unless noted otherwise)
Urinary biomarkers Source Smokers Non-smokers
Nicotine equivalents Nicotine 70.4-154 µmol/24 h Not Detected
Total NNAL NNK 1.1 - 2.9 Not Detected
Total NNN NNN 0.049 - 0.24 Not Detected
1-HOP or PheT PAH 0.50 - 1.45 0.18 - 0.50
MHBMA 1,3-Butadiene 15.5 - 322 0.65 - 7.5
SPMA Benzene 3.2 - 32.1 0.17 - 3.14
HPMA Acrolein 5,869 - 11,190 1,131 - 1,847
HBMA Crotonaldehyde 1,965 - 26,000 242 - 3,200
HEMA Ethylene oxide 19.1 - 102 6.51 - 38.8
Cd Cadmium 2.3 - 12.8 1.34 - 8.04
8-epi-PGF2a Oxidative damage 1.48 - 2.80 0.62 - 1.13
PGE-M Inflammation 54 - 60 31.6 - 45.3
Based on 1.3g creatinine per 24h in smokers and
1.5g creatinine per 24h in non-smokers, or 1.5 l
urine per 24h.
S.S. Hecht, J-M Yuan, and D. Hatsukami, Chem.
Res. Toxicol., 2010
36Essential Facts About NNK,A Tobacco-Specific
Lung Carcinogen
- Present in tobacco and tobacco smoke specific to
tobacco products - Systemic lung carcinogen in rats, mice, hamsters,
and ferrets. - Also induces tumors of the pancreas, nasal
cavity, and liver in rats - Considered to be a cause of lung, oral cavity and
pancreatic cancer in people exposed to tobacco
products - NNK and NNN- Carcinogenic to humans Group 1
(IARC Volume 89, 2007) reaffirmed (Vol 100E,
2009)
S.S. Hecht, Chem. Res. Toxicol. 11559 (1998)
Nature Rev. Cancer 3733 (2003)
37Metabolism of NNK by Carbonyl Reduction
38NNAL Plus NNAL-Glucs (Total NNAL) A Biomarker
of NNK Exposure
- Quantified by GC-TEA or LC-MS/MS
- High analytical specificity and sensitivity
- Specific to tobacco product exposure
- Responsive to dose
- Measures uptake of a lung carcinogen
S.S. Hecht, Carcinogenesis 23 907 (2002)
S.G.Carmella et al, CEBP 4 635 (1995) 12 1257
(2003) D. Hatsukami et al, Nic. Tob. Res. 8
169 (2006)
39GC-TEA Chromatogram of NNAL in a Smoker's Urine
40Applications of the Total NNAL Biomarker
- Cessation of smoking or smokeless tobacco
- Reduction of smoking
- Carcinogen uptake from new and old tobacco
products - Omni, light and ultra-light cigarettes
- Snus and other smokeless products
- Ultra low nicotine cigarettes
- Evaluation of carcinogen dose in various groups
- Reducers vs. light smokers
- Smokers of differing numbers of cigarettes
- Ethnic groups, gender, and teen-age smokers
- Smokeless vs. smokers
- Duration of smokeless use
- Carcinogen uptake from secondhand cigarette smoke
- Relationship to lung cancer
D. Hatsukami, J. Jensen, A. Joseph, S. E. Murphy,
S.G. Carmella, S.S. Hecht, and co-workers.
Cancer Res., JNCI, CEBP, Nic. Tob. Res.,
1999-2008
41Tobacco Harm Reduction Continuum of Risk
Conventional cigarettes Modified tobacco
cigarettes Cigarette reduction Cigarette-like
delivery devices Extra-low nicotine
cigarettes Smokeless tobacco products
Nicotine delivery devices Smoking Cessation
Most toxic
Least toxic
D. Hatsukami et al., Nicotine Tob. Res.
9S537-S553 2007
42Applications of the Total NNAL Biomarker
- Cessation of smoking or smokeless tobacco
- Reduction of smoking
- Carcinogen uptake from new and old tobacco
products - Omni, light and ultra-light cigarettes
- Snus and other smokeless products
- Ultra low nicotine cigarettes
- Evaluation of carcinogen dose in various groups
- Reducers vs. light smokers
- Smokers of differing numbers of cigarettes
- Ethnic groups, gender, and teen-age smokers
- Smokeless vs. smokers
- Duration of smokeless use
- Carcinogen uptake from secondhand cigarette smoke
- Relationship to lung cancer
D. Hatsukami, J. Jensen, A. Joseph, S. E. Murphy,
S.G. Carmella, S.S. Hecht, and co-workers.
Cancer Res., JNCI, CEBP, Nic. Tob. Res.,
1999-2008
43Non-Smokers Exposure to NNK Throughout Life by
Measurement of Urinary Total NNAL
Exposed Group Type of Exposure Total NNAL (fmol/ml urine) of Amount in Smokers' Urinea
Fetus Transplacental 25 29 (amniotic fluid) 1.3
Newborns Transplacental 130 150 6.5
Infants (lt1 year old) Air 83 20 4.2
Elementary School Children Elementary School Children Elementary School Children Elementary School Children
Minneapolis Air 56 76 2.8
Moldova Air 90 77 4.5
Women Living with Smokers Air 50 68 2.5
Hospital Workers Air 59 28 3.0
Casino Patrons Air 18 15 0.9
Restaurant and Bar Workers Air 33 34 1.7
a based on 2 pmol/ml total NNAL in smokers S.S.
Hecht, Carcinogenesis 23907 (2002) S.S. Hecht
et al, CEBP 15988 (2006)
44Total NNAL measurements in secondhand
smoke-exposed non-smokers have impact
- It can only come from secondhand smoke.
- It represents uptake of a lung carcinogen.
- It is found in the urine of non-smokers.
- It is the only lung carcinogen biomarker
consistently elevated in exposed non-smokers. - These studies should spur clean air legislation
in the remaining countries, states and localities.
45Median Serum Cotinine Levels in Non-Smokers, by
Age Group 1988-2002
CDC NHANES Study MMWR 55 1130 (2006)
46Regulation of Indoor Smoking and Tobacco Control
- Regulation of indoor smoking
- Reduces cues for smoking
- Reduces amount smoked
- Can change social norms
- Regulation of indoor smoking, along with
counter-advertising and taxation, are the most
effective methods in tobacco control.
47Applications of the Total NNAL Biomarker
- Cessation of smoking or smokeless tobacco
- Reduction of smoking
- Carcinogen uptake from new and old tobacco
products - Omni, light and ultra-light cigarettes
- Snus and other smokeless products
- Ultra low nicotine cigarettes
- Evaluation of carcinogen dose in various groups
- Reducers vs. light smokers
- Smokers of differing numbers of cigarettes
- Ethnic groups, gender, and teen-age smokers
- Smokeless vs. smokers
- Duration of smokeless use
- Carcinogen uptake from secondhand cigarette smoke
- Relationship to lung cancer
D. Hatsukami, J. Jensen, A. Joseph, S. E. Murphy,
S.G. Carmella, S.S. Hecht, and co-workers.
Cancer Res., JNCI, CEBP, Nic. Tob. Res.,
1999-2008
48Relationship of Urinary NNAL to Lung Cancer in
Two Prospective Cohorts of Cigarette Smokers
- Collaboration with Professors Mimi Yu and
Jian-Min Yuan - Two prospective cohorts of Chinese cigarette
smokers Shanghai and Singapore - Nested case control study of 246 cases of lung
cancer and 245 matched controls - Total NNAL and cotinine quantified in stored
urine samples collected prior to lung cancer
diagnosis
49Joint Effect of Urinary Total NNAL and Cotinine
on Lung Cancer Risk
NNAL in tertile Cotinine in tertile Cotinine in tertile Cotinine in tertile Cotinine in tertile Cotinine in tertile Cotinine in tertile
NNAL in tertile 1st (low) 1st (low) 2nd 2nd 3rd (high) 3rd (high)
NNAL in tertile Ca/Co1 OR (95 CI)2 Ca/Co1 OR (95 CI)2 Ca/Co1 OR (95 CI)2
1st (low) 9/47 1.00 23/25 3.93 (1.54, 10.05) 11/10 5.08 (1.63, 15.89)
2nd 14/24 3.01 (1.11, 8.10) 31/32 4.15 (1.70, 10.12) 22/26 4.48 (1.78, 11.31)
3rd (high) 8/10 3.41 (1.08, 11.25) 30/25 5.58 (2.25, 13.84) 93/46 8.47 (3.69, 19.46)
1 No. of cases/no. of controls 2 Odds ratios (OR)
were adjusted for age, year of interview, year of
sample collection, gender and dialect group,
study location (Shanghai versus Singapore),
number of cigarettes smoked per day, and number
of years of smoking CI, confidence interval.
J. Yuan, M. Yu, S.E. Murphy, S. Carmella, S.S.
Hecht et al., Cancer Res.,69 2990 (2009)
50Conclusions of the Shanghai and Singapore Study
- Total NNAL significantly associated with risk of
lung cancer in a dose-dependent manner, after
adjustment for smoking history and urinary
cotinine. - Cotinine was independently associated with lung
cancer, consistent with previous data. - Smokers in the highest tertiles of urinary total
NNAL and cotinine exhibited an 8.5 fold increased
risk for lung cancer, relative to those with
comparable smoking history, but in the lowest
tertiles.
J. Yuan et al, Cancer Res. 69 2990 (2009)
51Similar Results in the PLCO Study
- The Prostate, Lung, Colon, and Ovarian Cancer
Screening Trial started 1993 - 77,468 subjects (25,000 smokers) screened over
1,000 lung cancer cases diagnosed - Questionnaire data and blood samples collected
prospectively - 100 lung cancer cases and 100 controls without
lung cancer selected all were smokers of gt 10
CPD - Pre-diagnostic serum analyzed for total NNAL and
cotinine
T. Church, K. Anderson, M. Geisser, Y. Zhong, C.
Le, N. Caporaso, S. Carmella, A. Benoit, S. S.
Hecht, CEBP,18 260 (2009)
52Total NNAL and Lung Cancer
- Total NNAL is a risk biomarker.
- Results are consistent with all experimental and
previous clinical and epidemiologic data. - Results further implicate NNK as an independent
etiologic risk factor in lung cancer.
53Cigarette Smoke Constituents Targeted for
Regulation by WHO, and Their Biomarkers
- Benzoapyrene
- NNK, NNN
- Acrolein
- Benzene
- 1,3-Butadiene
- Carbon monoxide
- Acetaldehyde
- Formaldehyde
- Nicotine
- 1-HOP or PheT in urine
- NNAL and NNN in urine
- 3-HPMA in urine
- SPMA in urine
- MHBMA in urine
- Exhaled CO
- Leukocyte DNA adducts
- Leukocyte DNA adducts
- Nicotine metabolites
D.M. Burns et al. Tob. Control. 17 132-141
(2008) S.S. Hecht, Carcinogenesis 23 907-922
(2002) L. Chen et al. Chem. Res. Toxicol. 20
108-133 (2007) S.G. Carmella et al. Chem. Res.
Toxicol., 22 734-741 (2009)
54Outline of Presentation
- Tobacco and cancer
- Background on tobacco carcinogen biomarkers
- Examples of tobacco carcinogen biomarkers
- Total NNAL
- Formaldehyde-DNA adducts
55Formaldehyde Genetic Toxicology
- Genotoxic
- Mutagenic
- DNA protein cross-links
- DNA strand breaks
- Sister chromatid exchanges
- Chromosomal aberrations
- These changes initiated by reactions with DNA to
form adducts - No previous reports of formaldehyde DNA adducts
in humans - International Agency for Research on Cancer
Monographs, Volume 88 (2006)
56Structures of Formaldehyde-DNA Adducts
R. Shapiro et al. (1980) F. Beland et al. (1984)
57Conversion of N6-HOMe-dAdo to N6-Me-dAdo
M. Wang et al. Chem. Res. Toxicol. 20 1141-1148
(2007)
58Outline of Analytical Method for N6-HOMe-dAdo in
DNA
59LC-ESI-MS/MS-SRM Chromatograms of N6-Me-dAdo in
Smokers' Leukocyte DNA
60Typical LC-ESI-MS/MS-SRM Chromatograms of
N6-Me-dAdo in Leukocyte DNA
61Levels of N6-HOMe-dAdo (as N6-Me-dAdo) in
Leukocyte DNA of Smokers and Non-smokers
62Previous Studies of Leukocyte DNA Adducts in
Smokers vs. Non-smokers
- Most used 32P-postlabelling and immunoassay
inconsistent results comparing smokers and
non-smokers - Mixed results in 8-OH-dG analyses
- Marginally higher levels of acetaldehyde-DNA
adducts in smokers before stopping - Significant difference in BPDE-DNA adducts (2
per 108 vs 1 per 108 nucleotides) - Our results 5 per 108 vs. 0.5 per 108 nucleotides
IARC Monographs, Vol 83 (2004) Pavanello et al.
Mutat. Res 611 54 (2006) Chen et al Chem. Res.
Toxicol. 20 108 (2007)
63Sources of Formaldehyde-DNA Adducts
- Formaldehyde itself in cigarette smoke, but blood
levels were not elevated in volunteers exposed to
similar amounts - Smoking effect on endogenous metabolism
- Released as a metabolite from nicotine, NNK, or
related compounds - Transfer from formaldehyde-histone adducts
- Secondary metabolite from lipid peroxidation or
inflammation caused by smoking
64Conclusions Formaldehyde DNA Adducts
- First study to detect formaldehyde-DNA adducts in
humans - Highly significant differences between smokers
and non-smokers - Results indicate a previously unrecognized and
potentially important role for formaldehyde in
smoking-induced cancer
M. Wang, G. Cheng, S.S. Hecht et al. Cancer Res.,
69 7170 (2009)
65Overall Goal
- Elucidate mechanisms of tobacco-induced
cancer and apply this knowledge to cancer
prevention.
66We are Making Progress in Tobacco Control
- Smoke-free legislation
- Increased taxation
- Aggressive anti-tobacco advertising
67Age-Adjusted Total U.S. Mortality Rates for Lung
and Bronchus Cancer
Source SEER data http//seer.cancer.gov/faststats
/
68(No Transcript)
69Dorothy Hatsukami
70Acknowledgements
- Hecht Lab
- Steven Carmella
- Mingyao Wang
- Irina Stepanov
- Pramod Upadhyaya
- Brad Hochalter
- Silvia Balbo
- Shaomei Han
- Menglan Chen
- Guang Cheng
- Lei Meng
- Yan Zhong
- Aleks Knezevich
- John Muzic
- Core Facilities
- Pete Villalta
- Chap Le
- Xianghua Luo
- Dorothy Hatsukami
- Joni Jensen
- Amanda Anderson
- Rachel Feuer
- Tim Church
- Kristin Anderson
- Mindy Geisser
- Jian-Min Yuan
- Mimi Yu
71Research Support
- National Cancer Institute
- National Institute of Environmental Health
Sciences - National Institute on Drug Abuse/NCI/NIAAA
(TTURC) - American Cancer Society
- Wallin Chair in Cancer Prevention