Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Dietary carcinogens and genetic variation in their metabolism: epidemiological studies on the risk of selected cancers
(USC Thesis Other)
Dietary carcinogens and genetic variation in their metabolism: epidemiological studies on the risk of selected cancers
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
DIETARY CARCINOGENS AND GENETIC VARIATION IN THEIR
METABOLISM:
EPIDEMIOLOGICAL STUDIES ON THE RISK OF SELECTED CANCERS
by
Chelsea Catsburg
A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(MOLECULAR EPIDEMIOLOGY)
May 2013
Copyright: 2013 Chelsea Catsburg
ii
Acknowledgements
First and foremost, I would like to thank and recognize the support of my mentor and
advisor, Dr. Mariana Stern. She has always gone above and beyond to help me in every step
of this process, and has been a constant source of inspiration and guidance. Dr. Stern has
made herself available to me at all times, and her passion and belief in this research has
consistently helped me to keep focused and on track. Beyond the academic, Dr. Stern has
also been there on a personal level, and her office door has always been open to talk about
any issue. Without her kindness and understanding, it is hard for me to imagine how
completing this step, while also raising a young family, would have been possible. Dr. Stern
has been my mentor in every sense of the word, and I cannot express enough gratitude.
I would also like to thank other faculty members here at USC, primarily my
dissertation committee: Dr. Sue Ingles, Dr. Juan Pablo Lewinger and Dr. Rayudu
Gopalakrishna. Thank you all for taking the time to support me through my qualification
and dissertation steps and providing insightful comments and input into my work. I would
also like to thank Dr. Anna Wu who served on my qualification committee, Dr. Ann
Hamilton who helped me with various projects at the beginning of my time here, Dr. Jean
Richardson, and Dr. Stanley Azen for academic advice here and beyond my PhD.
Past and present members of my lab have made coming in to USC every day a very
enjoyable experience. Dr. Amit Joshi was always (and is still always) available for analytical
and statistical advice, Roman Corral has helped me since day one, particular with lab
iii
techniques and anything of that nature. I have worked with Ahva Shahabi and Andre Kim
on various projects, and they have always made this enjoyable and I value their friendship.
Finally, I would like to thank my mother, Sherril Harries, for always believing in me.
My loving husband, Joseph Verbeck, for always supporting and listening to me. And my
daughter, Matilda Love, who has made it impossible to be unhappy, or lose focus, since the
day she was born.
iv
Table of Contents
Acknowledgements .............................................................................................................................. ii
List of Tables ....................................................................................................................................... vii
List of Figures ...................................................................................................................................... ix
Abstract ................................................................................................................................................. xi
CHAPTER 1: INTRODUCTION .............................................................................................. 1
1.1 Diet & Cancer ............................................................................................................................ 1
1.1.1 Dietary Carcinogens .......................................................................................................... 2
1.1.1.1 Heterocyclic Amines (HCAs) .................................................................................. 3
1.1.1.2 Polycyclic Aromatic Hydrocarbons (PAHs) ......................................................... 5
1.1.1.3 N-nitroso Compounds (NOCs) ............................................................................ 6
1.1.2 Dietary Antioxidants and Preventative Factors ............................................................ 7
1.2 Genetic Variation and Cancer Susceptibility Pathways ....................................................... 8
1.2.1 Carcinogen Metabolism Pathways and prostate cancer risk ..................................... 11
1.2.1.1 HCA Metabolism .................................................................................................... 11
1.2.1.2 PAH Metabolism ..................................................................................................... 12
1.2.1.3 NOC Metabolism .................................................................................................... 14
1.2.2 DNA Repair Pathways, NOCs, and bladder cancer risk ........................................... 16
1.2.2.1 Base Excision Repair Pathway (BER) .................................................................. 18
1.2.2.2 Non Homologous End Joining Pathway (NHEJ) ............................................. 18
1.2.2.3 Homologous Recombinational Repair Pathway (HRR) ................................... 19
1.2.2.4 Nucleotide Excision Repair Pathway (NER) ...................................................... 20
1.2.3 GWAS Studies and colorectal cancer susceptibility genes ........................................ 20
1.3 Selected Cancers ...................................................................................................................... 21
1.3.1 Prostate Cancer: Overview and Risk Factors .............................................................. 22
1.3.2 Bladder Cancer: Overview and Risk Factors ............................................................... 25
1.3.3 Colorectal Cancer: Overview and Risk Factors .......................................................... 28
CHAPTER 2: HYPOTHESIS & SPECIFIC AIMS ............................................................ 31
v
CHAPTER 3: METHODS .......................................................................................................... 33
3.1: Study Design and Study Population .................................................................................... 33
3.1.1 California Collaborative Case-Control Study of Advanced Prostate Cancer ......... 33
3.1.2 Singapore Chinese Health Study ................................................................................... 34
3.1.3 Los Angeles Bladder Cancer Study ............................................................................... 35
3.1.4 Colorectal Cancer Family Registry ................................................................................ 36
3.2: Data Collection ....................................................................................................................... 39
3.2.1 California Collaborative Case-Control Study of Advanced Prostate Cancer ......... 39
3.2.2 Singapore Chinese Health Study ................................................................................... 40
3.2.3 Los Angeles Bladder Cancer Study ............................................................................... 41
3.2.4 Colorectal Cancer Family Registry ................................................................................ 43
3.3: SNP Selection and Genotyping ........................................................................................... 44
3.3.1 California Collaborative Case-Control Study of Advanced Prostate Cancer ......... 44
3.3.2 Los Angeles Bladder Cancer Study ............................................................................... 45
3.3.3 Colorectal Cancer Family Registry ................................................................................ 46
3.3: Statistical Analysis .................................................................................................................. 47
3.4.1 California Collaborative Case-Control Study of Advanced Prostate Cancer ......... 47
3.4.2 Singapore Chinese Health Study ................................................................................... 50
3.4.3 Los Angeles Bladder Cancer Study ............................................................................... 52
3.4.4 Colorectal Cancer Family Registry ................................................................................ 54
CHAPTER 4: RESULTS ............................................................................................................. 57
4.1: Fish Intake, Metabolic Enzymes and Risk of Prostate Cancer ....................................... 57
4.1.1 Main Effects of Metabolic Enzymes on Prostate Cancer Risk ................................ 57
4.1.2 Fish intake, cooking methods, genetic variation in carcinogen metabolism
enzymes and advanced Prostate Cancer Risk .............................................................. 59
4.2: Fish, Red Meat and Poultry Intake and Risk of Prostate Cancer ................................... 68
4.3: Processed meats, NOCs, DNA Repair Enzymes and Risk of Bladder Cancer ........... 75
4.3.1 Main Effects of Processed meats and dietary NOCs on Bladder Cancer Risk ..... 77
4.3.2 Precursors of endogenous NOC formation and Bladder Cancer Risk ................... 81
4.3.3 Processed meats, modification by DNA Repair Enzymes and Bladder
Cancer Risk ....................................................................................................................... 85
4.4: Fish and Meat Intake, GWAS Studies, and Risk of Colorectal Cancer ......................... 92
4.4.1 Main Effects of Fish Intake on Colorectal Cancer Risk ............................................ 93
4.4.2 Meat and fish intake, modification by GWAS identified cancer
susceptibility genes, and risk of Colorectal Cancer ..................................................... 96
vi
CHAPTER 5: DISCUSSION ...................................................................................................... 98
5.1: Genetic variation in metabolic enzymes and the risk of prostate cancer ...................... 99
5.2: Fish intake, variation in metabolic enzymes, and the risk of prostate cancer ............ 102
5.3: Fresh and preserved red meat, poultry, and fish intake and the risk of
prostate cancer in Singapore ............................................................................................... 105
5.4: Processed meats, variation in metabolic enzymes, and risk of prostate cancer ......... 108
5.5: Variation in DNA Repair Enzymes, processed meat and NOC intake,
and the risk of bladder cancer .......................................................................................... 113
5.6: Fish intake and the risk of colorectal cancer ................................................................... 115
5.7: Meat and fish intake, variation in GWAS top hits, and risk of CRC ........................... 118
5.8: Strengths and weaknesses of epidemiological studies .................................................... 119
5.9: Summary of studies ............................................................................................................. 122
5.10: Overall Summary and Final Conclusions ....................................................................... 124
Bibliography ....................................................................................................................................... 127
vii
List of Tables
Table 4.1: Socio-demographic and lifestyle characteristics of cases and
controls with available DNA from blood, by stage of disease
and study site 58
Table 4.2: Polymorphisms in metabolic genes and prostate cancer risk,
by stage of disease 60
Table 4.3: Fish intake and risk of prostate cancer among men with
DNA available, by stage of disease 61
Table 4.4: Fish intake, cooking temperature and risk of prostate
cancer among men with available DNA, by stage of disease 63
Table 4.5: Fish intake, level of doneness and risk of PCA among men
with available DNA, by stage of disease 64
Table 4.6: Fish intake and prostate cancer risk stratified by PTGS2
genotype, by stage of disease 65
Table 4.7: Well-done white fish intake and risk of advanced prostate
cancer, stratified on PTGS2 genotype 66
Table 4.8: Demographic and dietary characteristics of study population
by quartiles of meat and fish intake 70
Table 4.9: Meat, poultry and fish intake and risk of prostate cancer
among Singapore Chinese, by stage of disease 71
Table 4.10: Preserved meat intake and risk of prostate cancer by BMI
and fruit and vegetable intake 73
Table 4.11: Socio-demographic and life-style characteristics of cases
and controls 76
Table 4.12: Associations between intake of processed meats and liver and
risk of bladder cancer 79
Table 4.13: Associations between nitrate, nitrite, NOCs and heme iron,
and risk of bladder cancer 82
viii
Table 4.14: Associations between intake of processed meats, liver and
heme, and risk of bladder cancer stratified by nitrate intake 84
Table 4.15: Significant SNP by Exposure interactions after Bonferonni
adjustment 90
Table 4.16: Distribution of cases and controls in the Colon CFR 92
Table 4.17: Total Fish Intake and Risk of Colorectal Cancer 93
Table 4.18: Intake of Shellfish and Non-Shellfish and risk of CRC 95
Table 4.19: Intake of Shellfish and risk of CRC stratified by familial
history of cases 95
Table 4.20: Meat intake variables and SNP interactions that survived
Bonferroni correction 97
ix
List of Figures
Figure 1.1: Exposures leading to selected carcinogen formation 3
Figure 1.2: Carcinogenic pathways are modified by genetic variation in
metabolism and repair 9
Figure 1.3: HCA metabolism 12
Figure 1.4: PAH metabolism 14
Figure 1.5: NOC metabolism 16
Figure 1.6: DNA Repair Pathways 17
Figure 1.7: WCRF panel judgment on dietary factors and PCA risk 25
Figure 1.8: WCRF panel judgment on dietary factors and bladder
cancer risk 27
Figure 1.9: WCRF panel judgment on dietary factors and CRC risk 30
Figure 3.1: Colon CFR Participating Centers 38
Figure 4.1: Salami/pastrami/corned beef intake and SNPs in the
BER pathway 87
Figure 4.2: Salami/pastrami/corned beef intake and SNPs in the
HRR pathway 87
Figure 4.3: Salami/pastrami/corned beef intake and SNPs in the
NER pathway 88
Figure 4.4: Salami/pastrami/corned beef intake and SNPs in the
NHEJ pathway 88
Figure 4.5: Effect modification of salami/pastrami/corned beef intake
by rs1077478 89
x
Figure 4.6: Salami/pastrami/corned beef intake and SNPs in the HRR
pathway among males only 91
Figure 4.7: Salami/pastrami/corned beef intake and SNPs in the HRR
pathway among females only 91
xi
Abstract
The central topic of this thesis involves investigating the potential role in cancer risk of
carcinogens present in the everyday human diet. In addition, I report on other relevant
dietary compounds, such as antioxidants, that can reduce the detrimental effect of these
carcinogens and can thus be preventative against cancer risk. Finally, I present results from
investigations on the underlying genetic variation involved in an individual’s metabolism of
these specific carcinogens and antioxidants.
Two of the key carcinogens considered in this thesis are polycyclic aromatic
hydrocarbons (PAHs) and heterocyclic amines (HCAs), which are known to form in meats
and fish, cooked at high temperatures, and are potent carcinogens. Upon entering the body,
PAHs and HCAs can form adducts with DNA, and can also generate oxidative DNA
damage, thus contributing to the mutational load that can lead to tumor formation. In
contrast, diets high in fish have been associated with decreased risk of several cancers,
including prostate, breast and colorectal, albeit results have been inconsistent. It has been
proposed that omega-3 polyunsaturated fatty acids (PUFA) found in some fish, particularly
dark and oily species, might explain this association.
A third group of dietary carcinogens investigated in this thesis are N-nitroso
compounds (NOCs), which have been demonstrated to cause bladder tumors in mice and
other experimental organisms. The main sources of dietary exposure to NOCs are processed
meats, such as cold cuts, sausages and bacon. Moreover, NOCs can be formed endogenously
in the body when amines undergo nitrosation by nitrites at various sites, including the oral
xii
cavity, stomach, intestines and urinary bladder, as well as other sites of inflammation or
infection. Endogenous formation of NOCs has been estimated to contribute between 45-
75% of total NOC exposure. Conversely, plant sources in diet contain high amounts of
compounds such as vitamin C and specific polyphenols that directly inhibit these
endogenous nitrosation reactions.
Investigation of dietary exposures to carcinogens such as those discussed above is of
public health relevance in order to identify dietary items, range of exposure, and cooking
practices that might be most critical to reduce cancer risk. Given the complex inter-
relationship between these various carcinogenic and anti-carcinogenic exposures, studying
the underlying genetic variation of the enzymes relevant for their metabolism provides
insight into which of these mutagens is more relevant for specific cancers.
1
CHAPTER 1: INTRODUCTION
1.1 Diet & Cancer
Many of the well-established risk factors for cancer are un-modifiable, such as age, family
history and certain genetic variants. Thus, from a public health and prevention aspect,
research focused on modifiable risk factors is highly relevant. One interesting area of
exploration is the role of diet in carcinogenesis, as dietary factors play a significant role in
cancer risk. An estimated one-third of annual cancer deaths in the United States are due to
dietary factors (ACS 2012). Moreover, a review on diet and cancer estimates that dietary
factors are responsible for up to 80 percent of cancers of the large bowel and prostate (ACS
2012). This link between diet and cancer is not new. In January 1892, Scientific
American printed the observation that "cancer is most frequent among those branches of the
human race where carnivorous habits prevail". Indeed, research has shown that cancer is
much more common in populations consuming diets rich in fatty foods, particularly meat,
and much less common in countries eating diets rich in grains, vegetables, and fruits
(Tantamango-Bartley, Jaceldo-Siegl et al. 2013). These associations are complex. According
to the second expert report of the World Cancer Research Fund (WCRF) and the American
Institute of Cancer Research (AICR), increasing body mass index (BMI) is associated with
the risk of at least seven cancers (WCRF/AICR 2011). Vegetables and fruits are low in
calories, and whole grains are rich in fiber and moderate in calories, thus intake may help to
maintain a healthy weight. In contrast, meats tend to be high in calories and saturated fat,
2
potentially contributing to weight gain and increased risk of cancer (Nahleh, Bhatti et al.
2011). Additionally, diet affects the action of hormones, metabolism, the immune system
and numerous other important processes in the body (Ferguson and Philpott 2007). Animal
products have been implicated as containing many potentially carcinogenic compounds,
whereas in contrast, fruits and vegetables contain a variety of vitamins, minerals,
antioxidants, and phytochemicals to protect the body. This is made even more complex as
evidence suggests it is not just one, but the synergy of compounds working together in the
overall diet that are ultimately responsible for cancer protection or cancer risk.
The following sections describe the actions of three key carcinogens that occur
naturally in the human diet, along with the characteristics of anti-carcinogenic dietary
components, such as antioxidants and certain polyunsaturated fatty acids.
1.1.1 Dietary Carcinogens
There are three well-known carcinogens that are known to occur naturally in the everyday
human diet: Polycyclic Aromatic Hydrocarbons (PAHs), Heterocyclic Amines (HCAs) and
N-Nitroso Compounds (NOCs). All three groups are associated with intake of meats,
including fish, and their formation or accumulation is largely dependent on the methods
used to prepare meats and fish before consumption. HCAs and PAHs are mostly associated
with the cooking of meats, especially at high temperatures; whereas NOCs are formed upon
preservation and curing of meats (Figure 1.1). Figure 1.1 also indicates sources outside of
3
diet that may contribute to PAH and NOC exposure, such as tobacco use and occupation.
Each of the three carcinogenic groups is explored in detail below.
CARCINOGENS
Environmental
Sources:
e.g.
Occupation/
Tobacco
Use
DIET
Fried
and/or
Oven
Broiled
Meats
&
Fish
Grilled
and/or
Barbequed
Meats
&
Fish
Cured
Meats
&
Fish
Polycyclic
Aromatic
Hydrocarbons
(PAHs)
Heterocyclic
Amines
(HCAs)
N-‐Nitroso
Compounds
(NOCs)
endogenous
formation
Figure 1.1: Exposures leading to selected carcinogen formation
1.1.1.1 Heterocyclic Amines (HCAs)
Heterocyclic amines (HCAs) are the most potent mutagens found in cooked meats, as
determined by the Ames test, and have been determined to be both mutagenic and
carcinogenic in animals (Sinha 2002, Sugimura, Wakabayashi et al. 2004). Three of the most
4
active HCAs that have been detected in meat are MeIQx, DiMeIQx and PhiP, with PhiP
being the most abundant (Kulp, Knize et al. 2000). Meat and fish contain abundant
precursors of these compounds, such as creatine, amino acids and reducing sugars (Overvik,
Kleman et al. 1989). Upon cooking, these precursors come to the surface of the meat, where
conditions such as decreased moisture and increased temperature are optimal for HCA
formation (Felton and Knize 1991). The amount of HCA formation is dependent on a
number of factors (Sinha and Rothman 1997). Among them, increased temperatures such as
those used in pan-frying, oven-broiling, grilling and barbequing are associated with high
levels of HCA formation (Knize, Salmon et al. 1999). In addition, increased time of cooking
can lead to an increase in HCA formation, as has been shown with the slow process of
smoking fish (Knize, Kulp et al. 2002). The oils/fats used in pan-frying act as an efficient
medium for heat transfer, increasing HCA levels (Johansson, Fredholm et al. 1995).
However, the type of oil/fat used can lead to further variation, with vegetable oils high in
unsaturated fatty acids resulting in less HCA formation than saturated fat mediums (Randel,
Balzer et al. 2007). Marinades high in anti-oxidants have been shown to decrease HCA
formation, as have certain pre-treatments such as microwaving (Knize, Kulp et al. 2002).
Because of this variability in amount of HCA, it is clear that epidemiological studies should
take into consideration the type of meat, cooking method and level of doneness to avoid
misclassification of HCA content (Sinha 2002). To this extent, in 1997, Sinha et al.
developed a standardized cooking module for use in food frequency questionnaires (Sinha
and Rothman 1997). When linked with the CHARRED database (Ziegler and Oei
2001)(Ziegler and Oei 2001), this module attempts to accurately estimate the HCA levels in
different meats, taking into account cooking methods and level of doneness.
5
1.1.1.2 Polycyclic Aromatic Hydrocarbons (PAHs)
A second class of carcinogens found in meat and fish are the polycyclic aromatic
hydrocarbons (PAHs). The most widely studied PAH in cooked meat is benzo(a)pyrene
(BaP) (Miller and Ramos 2001), found to be a potent animal carcinogen (Kazerouni, Sinha et
al. 2001). PAHs are formed from incomplete combustion of organic matter. In contrast to
HCA formation, which can occur upon any type of cooking, PAH formation can only occur
when there is direct content with the heat source (Cross and Sinha 2004). For example, in
grilling or barbequing, when meat or fish is cooked in direct flames or smoke, incomplete
combustion of the fat droppings forms PAHs which adhere to the surface of the food and
are subsequently consumed (Cross and Sinha 2004). The concentration of PAHs found in
meat is dependent on the fat content of meat, the type of fat present, the source of the heat
(e.g. wood vs. charcoal) and the position of the heat source (van Maanen, Moonen et al.
1994). Oven-broiling, where the heat source is above the meat, has been shown to result in
less PAH formation than grilling or barbequing (Kazerouni, Sinha et al. 2001). Cooking
methods which do not involve direct contact with the heat source, such as pan-frying,
stewing, microwaving, roasting and baking, do not contribute to the formation of PAHs
(Kazerouni, Sinha et al. 2001). The CHARRED database estimates levels of 25 PAHs
including BAP when paired with the complementing food frequency questionnaire capturing
cooking methods and doneness of meats (Ziegler and Oei 2001, Cross and Sinha 2004).
6
1.1.1.3 N-Nitroso Compounds (NOCs)
The main sources of exogenous dietary exposure to NOCs are processed meats such as cold
cuts, sausage and bacon, when nitrate or nitrite salts are added to meat as part of the
processing procedure (Stuff, Goh et al. 2009). However, NOCs can also form endogenously
when ingested amines undergo nitrosation by nitrites. Endogenously formed NOCs have
been estimated to account for 45-75% of total NOC exposure (Tricker 1997). Endogenous
nitrosation from these precursors may occur at various sites of the body, including the
urinary bladder, the stomach and the intestines, particularly under conditions of
inflammation and/or infection (Bartsch, Ohshima et al. 1989, Mirvish 1995). The
International Agency for Research on Cancer (IARC) concludes that ingested nitrate or
nitrite are probable carcinogens to humans under conditions that result in endogenous
nitrosation (IARC Monographs on the Evaluation of Carcinogenic Risks to Huams. Volume
94 (2010)). As well as contributing exogenous NOCs, processed meats are a major dietary
source of nitrite, amines and amides, all of the precursors that serve as substrates for
endogenous formation of NOCs (Mirvish 1995). Processing and storage of meats has been
shown to increase the amount of secondary amines available for nitrosation by nitrites (Ruiz-
Capillas and Jimenez-Colmenero 2004). Vegetables are the main dietary source of nitrates. A
number of bacterial strains can reduce nitrates found in vegetables to nitrites, and activated
macrophages also stimulate this nitrite production (Mirvish 1995). However, plant sources
contain high amounts of compounds such as vitamin C and specific polyphenols that
directly inhibit endogenous nitrosation reactions (Nabrzyski and Gajewska 1994). Heme
from red meat stimulates endogenous NOC formation (Lunn, Kuhnle et al. 2007). The
7
amount of available heme has been suggested as a limiting factor in these nitrosation
reactions, as heme intake is highly correlated with the amount of endogenous NOC
formation (Cross, Pollock et al. 2003, Jakszyn, Bingham et al. 2006).
1.1.2 Dietary Antioxidants and Preventative Factors
Epidemiological studies show that diets high in anti-oxidant rich food such as fruit and
vegetables are protective against many cancers (Borek 2004). Antioxidants may prevent
cancer through several mechanisms. Among them is the inhibition of endogenous formation
of NOCs (Ferguson, Philpott et al. 2004). The principal compounds that block in vivo
nitrosation are macronutrients such as vitamin C, vitamin E, and a variety of plant phenols
(Riboli and Norat 2003). These and other dietary antioxidants can also prevent
carcinogenesis by scavenging reactive oxygen species (ROS) and preventing free radical
formation (Ferguson, Philpott et al. 2004). Because of these properties, fruit and vegetable
consumption may be especially beneficial for people routinely exposed to dietary
carcinogens, such as those who often consume highly cooked and processed meats.
Diets high in fish have also previously been associated with decreased risk of many
cancers, including breast, prostate and colorectal, although the results have been inconsistent
(Reese, Fradet et al. 2009, Catsburg, Joshi et al. 2012, Joshi, John et al. 2012 ). It has been
proposed that omega-3 polyunsaturated fatty acids (PUFA) found in fish, particularly dark
and oily species, might explain this association (Wendel and Heller 2009). Omega-3 PUFAs
are essential fatty acids from which eicosanoids, important components of the cell
8
membrane, are derived. Eicosanoids produced from omega-6 fatty acids are pro-
inflammatory while eicosanoids produced from omega-3 fatty acids are anti-inflammatory,
thus a high “omega-3 to omega-6 ratio” has been linked to many health benefits (Reese,
Fradet et al. 2009, Hull 2013). In addition, there is evidence that omega-3 PUFAs are
selectively toxic to tumor cells (Johnson 2002). For these reasons, when considering fish
intake, the presence of omega-3 PUFAs may heavily confound the associations between
HCA and PAH exposure and carcinogenesis.
1.2 Genetic Variation and Cancer Susceptibility Pathways
Upon exposure to each of the aforementioned carcinogens, there is great variability in the
amount of effective toxicity from individual to individual. Once carcinogens enter the body
they undergo a series of metabolic reactions that can convert them into activated species,
which can then attack DNA. These activated species, or their precursors, can also serve as
substrates for detoxification enzymes that help excrete these carcinogenic substances from
the body. The ability of an individual to activate and/or detoxify each carcinogen will
directly affect the amount of toxicity to cells and thus the amount of DNA damage (Figure
1.2). Carcinogen metabolism has been shown to be variable in the human population, and
this variability is partially determined by genetic variants in genes that code for carcinogen
metabolism enzymes. Once DNA damage has occurred, genetic variation plays a further
role, by contributing to variability in the proficiency of various DNA repair pathways (Figure
9
1.2). An individual’s ability to repair damaged DNA will be a determinate factor as to
whether or not carcinogenesis will eventually develop.
Polycyclic
Aromatic
Hydrocarbons
(PAHs)
Heterocyclic
Amines
(HCAs)
N-‐Nitroso
Compounds
(NOCs)
DNA
Damage
Genetic
Instability
CANCER
Figure 1.2: Carcinogenic pathways are modified by genetic variation in metabolism
and repair
Since the turning point of the completion of the Human Genome Project in 2003 (Lander,
Linton et al. 2001, Venter, Adams et al. 2001), great advances have been made in identifying
and sequencing genetic variation. By gathering information on a study subject’s genetic
variation, we can gain insight into which variants in the population are likely to be associated
with disease, including cancer. From there, these associations may lead us to identifying
10
genes and pathways important for disease progression, which can in turn allow us to
pinpoint specific carcinogens that may be relevant risk factors. Moreover, when an exposure
is known to affect risk of disease, but may contain a number of potential carcinogens, gene x
exposure (GxE) analyses may allow us to identify which particular component of the
exposure is playing the more important role. Specific to this discussion, variation in the
genes involved in the metabolic pathways mentioned above, may translate into variability in
the ability of individuals to metabolize carcinogens, and this in turn may directly affect the
risk of cancer development.
The specific metabolic steps involved in activating and detoxifying each of the three
carcinogenic groups described before, are detailed in the following sections. These
descriptions are focused on the two organs, prostate and bladder, studied in this thesis as
part of the investigation of the role of genetic variation, dietary carcinogens and cancer risk.
Specifically, the sections below describe HCA and PAH metabolism pathways and prostate
cancer risk, NOC metabolism and prostate and bladder cancer risk, and DNA repair
pathways relevant for NOC-induced DNA damage and bladder cancer risk. Genetic
variation in genes that participate in these pathways has been postulated as putative cancer
susceptibility genes. Also discussed below, are cancer susceptibility genes identified via
Genome Wide Association Studies (GWAS).
11
1.2.1 Carcinogen Metabolism Pathways and prostate cancer risk
1.2.1.1 HCA Metabolism
HCAs present in cooked meats and fish are absorbed in the intestine and transported via the
hepatic portal system to the liver where they undergo metabolism (Aeschbacher and Ruch
1989) (Figure 1.3). In the liver, the majority of HCAs are activated by Phase I enzymes,
predominantly CYP1A2, into N-hydroxylamines (Boobis, Lynch et al. 1994). They can then
undergo further metabolism by Phase II enzymes such as SULT1A1, NAT1, NAT2, and
PTGS2 to generate reactive species that can damage DNA (Williams, Martin et al. 2000,
Moonen, Briede et al. 2002). Native HCAs and their metabolites can then be excreted via
bile back to the small intestine where they will eventually reach the large intestine via feces.
Alternatively, they can be transported out of the liver into the bloodstream and subsequently
reach other organs such as the prostate (Figure 1.3). Phase II enzymes in both the intestines
and prostate have similar functions as in the liver, generating reactive species from HCA
metabolites (Lawson and Kolar 2002, Sugimura, Wakabayashi et al. 2004). HCA-DNA
adducts have been found in both the prostate and colon (Nelson, Kidd et al. 2001,
Sugimura, Wakabayashi et al. 2004). The N-acetoxy-HCAs that result from NAT1 and
NAT2 metabolism are substrates for detoxification by the family of GST enzymes,
particularly GSTM1, GSTT1, and GSTP1 (Di Paolo, Teitel et al. 2005).
12
Absorption
N-‐oxidation
O-‐acetylation
Intestines
Liver
Excretion
via
urine
Circulation Prostate
Excretion
via
urine
CYP1A2
UGT1A1
UGT1A4
HCAs HCAs HCAs
HCAs
HCAs
SULT1A1
NAT2
GST1A1
NAT2
NAT1
SULT1A1
GSTP1
GSTM1
GSTT1
N-‐hydroxy-‐HCA
Sulfamyl-‐
HCA
HCA-‐N-‐
glucuronide
N-‐hydroxy-‐
HCA-‐N-‐
glucuronide
N-‐acetoxy-‐
HCA
Nitrenium
ion
DNA
adducts
N-‐hydroxy-‐
HCA
N-‐hydroxy-‐HCA
N-‐sulfonoloxy-‐HCA N-‐acetoxy-‐HCA
N-‐acetoxy-‐HCA
GSH-‐conjugate
DNA
adducts
Nitrenium
ion
N-‐sulfonoloxy-‐
HCA
Figure 1.3: HCA metabolism
1.2.1.2 PAH metabolism
Ingested PAHs are absorbed in the intestine and are transported to the liver, similarly to
HCAs. Once they reach the liver, PAHs are oxidized by cytochrome P450 enzymes,
predominantly CYP1A1 and CYP1B1, to generate PAH-epoxides (Shimada and Fujii-
Kuriyama 2004) (Figure 1.4). PAH-epoxides can be further activated by EPHX1 to form
ultimate carcinogens that can attack DNA, or they can be detoxified by either
glucuronidation (UGT1A1, UGT1A6, UGT1A9) or by glutathione conjugation (GSTA1,
13
GSTM1, GSTP1, GSTT1) (Bock, Gschaidmeier et al. 1999, Strange and Fryer 1999). PAH
glucuronides and PAH-GSH conjugates are either excreted out of the liver via bile, or via
the blood stream into the urine (Ramesh, Walker et al. 2004). In addition, intestinally
absorbed PCAs can enter the systemic circulation and reach other organs, such as the
prostate (Figure 1.4). The prostate contains similar enzymes to the liver, such as CYP1A1,
CYP1B1 and EPHX1, thus PAHs can also be transformed into their reactive states in this
organ (Di Paolo, Teitel et al. 2005). In addition, PTGS1 (COX-1) and PTGS2 (COX-2),
which are expressed during prostatic inflammation, can metabolize PAH-dihydrodiols to
reactive PAH-dihydrodiol-epoxides, capable of attacking DNA (Wiese, Thompson et al.
2001). It should be noted here that in addition to activating PAHs, the PTGS2 enzyme is
also responsible for producing omega-6 derived eicosanoids (Reese, Fradet et al. 2009).
Omega-3 and omega-6 fatty acids compete for the PTGS2 enzyme; therefore omega-3
PUFAs directly block the activity of PTGS2 and decrease production of pro-inflammatory
eicosanoids (Reese, Fradet et al. 2009). Reactive PAH metabolites can also be detoxified in
the prostate and intestines, again due to the presence of the GST and UGT family of
enzymes (Figure 1.4).
14
Absorption
Intestines
Liver
Excretion
via
urine
Circulation
Excretion
via
urine
CYP1A1
CYP1B1
UGT1A1
UGT1A6
UGT1A9
PAHs PAHs PAHs
PAHs
DD2
EPHX1 GSTM1
GSTP1
GSTT1
GSTA1
PAH-‐catechol
PAH-‐dihydrodiol
GSH-‐conjugate
PAH-‐glucuronide
Carbonium ion
DNA
damage
PAH-‐dihydrodiol-‐
epoxide
O-‐semiquinone
PAH
GSH-‐conjugate
DNA
damage
PAH
O-‐quinone
O-‐semiquinone
PAH-‐epoxide
PAH
O-‐quinone
CYP1A1
CYP1B1
O₂-‐
H₂O₂
O₂
O₂-‐
CYP1A1
CYP1B1
DD3
PAH-‐catechol
O₂-‐
H₂O₂
EPHX1
O₂
O₂-‐
COX-‐1
COX-‐2
PAH-‐dihydrodiol-‐
epoxide
CYP1A1
CYP1B1
Carbonium
ion
PAH-‐glucuronide
Prostate
PAH-‐epoxide
PAH-‐dihydrodiol
UGT1A6
UGT1A9
GSTM1
GSTP1
GSTT1
GSTA1
Figure 1.4: PAH metabolism
1.2.1.3 NOC metabolism
Ingested NOCs are mostly active in the esophagus, stomach and intestines. Although they
can reach farther organs by entering systemic circulation, the majority of preformed NOCs
are rapidly metabolized. This means that only a fraction of ingested NOCs reach organs such
as the bladder and prostate where they could directly exert a carcinogenic effect. However,
NOCs can also form endogenously at these sites when amines undergo nitrosation by
nitrites. A number of bacterial strains can reduce nitrates found in foods to nitrites, and
15
activated macrophages also stimulate nitrite production (Mirvish 1995). Endogenous
nitrosation occurs particularly under conditions of high bacterial presence, so is more
predominant in the stomach, intestines and bladder than the prostate (Bartsch, Ohshima et
al. 1989, Mirvish 1995). In addition, NOCs formed in the stomach or intestines often pass
through the bladder before excretion, providing a further route of exposure to these
carcinogens (Tricker 1997, Vermeer, Pachen et al. 1998, Ferrucci, Sinha et al. 2010). After
formation, NOCs are activated via α-oxidation (Guttenplan 1987) (Figure 1.5). NOCs are
mainly hydroxylated by CYP2E1 and CYP3A4, but can also be metabolized by CYP1A2
(Yamazaki, Inui et al. 1992). This generates α-hydroxy-NOCs that can either react directly
with DNA (Wong, Murphy et al. 2005), or can alternatively decompose into diazonium ions,
oxonium ions, and carbocations which are highly electrophilic and can immediately alkylate
DNA and induce DNA damage (Guttenplan 1987, Verna, Whysner et al. 1996, Wong,
Murphy et al. 2005). These highly reactive species act locally, therefore the expression of
their metabolic enzymes in the target tissue is extremely important to consider. Similarly to
HCA and PAH detoxification, NOC-derived electrophilic species can be detoxified by the
GST family of enzymes (Gichner and Veleminsky 1988) (Figure 1.5). Reactive oxidative
species (ROS) are produced as by-products of metabolic activation of NOCs in the target
tissues and may thus also contribute to the carcinogenic effects of NOCs. Studies in rats
shown an increase in ROS, as measured by various indicators of lipid peroxidation, after
exposure to NOCs (Ahotupa, Bussacchini-Griot et al. 1987).
16
CYP2E1
CYP3A4
CYP1A2
α-‐hydroxy-‐NOCs
diazonium
ions
oxonium
ions
carbocations
DNA
DAMAGE
GSTs
Detoxification
NOCs
Figure 1.5: NOC metabolism
1.2.2 DNA Repair Pathways, NOCs, and bladder cancer risk
The types of DNA damage induced by NOCs are shown in Figure 1.6. Firstly, the
metabolism of all these types of carcinogens generates ROS which can cause single-stranded
breaks (SSBs), double-stranded breaks (DSBs), abasic sites, and modified bases (Joenje 1989,
Floyd 1990). Repair of base damage and abasic sites can also cause indirect SSBs, which in
turn can lead to DSBs. Base damage, abasic sites, and SSBs are primarily repaired through
the base excision repair pathway (BER). DSBs are repaired by either the non-homologous
17
end joining (NHEJ) pathway, or the homologous recombination repair (HRR) pathway
(Lieber, Ma et al. 2003). In addition to generating ROS, the electrophilic metabolites formed
from NOC metabolism can result in alkylated bases (Figure 1.6). Both bulky lesions and
alkylated bases are repaired via the nucleotide excision repair (NER) pathway. Due to high
genetic variation in DNA repair genes, DNA repair proficiency is very variable in the
population. Thus genetic variants in the mentioned DNA repair pathways are highly likely to
modify the effects of NOC exposure. The roles of specific genes involved in each of these
pathways are briefly described below.
ROS
Electrophilic
metabolites
Base
damage
SSB DNA
bulky
adducts
alkylated
bases
DSB
DNA
sequence
restored
Base
changes
Translocations
DNA
loss
Genetic
Instability
Base
changes
BER
XRCC1
OGG1
PARP
APE1
NEIL1
POLB
LIG3
HRR
RAD51
RAD52
XRCC2
XRCC3
NBN
MRE11
RAD50
NHEJ
XRCC4
XRCC5
XRCC6
PRLDC
DCLRE1C
LIG4
NER
XPA
XPC
ERCC2
ERCC4
ERCC5
ERCC1
POLD1
LIG1
pathways
that
repair
DNA
damage
pathways
that
contribute
to
DNA
damage
Processed
meats
Exogenous
NOCs
Amines
from
meat Nitrates/Nitrites
From
fruits,
vegetables,
processed
meats
Endogenous
NOCs
Figure 1.6: DNA Repair Pathways
18
1.2.2.1 Base Excision Repair Pathway (BER)
The first step in the repair of base damage is the recognition and removal of the specific
base by a DNA glycosylase. There are at least 10 distinct DNA glycosylases that recognize
different types of base damage, however two that are relevant for the type of oxidative
damage caused by ROS are OGG1 and NEIL2 (Wood, Mitchell et al. 2005). Removal of
these damaged bases by glycosylases forms apurinic/apyrimidic (AP) sites, which are acted
upon by the major human AP endonuclease, APE1 (Wilson and Barsky 2001). APE1
recruits XRCC1 which in turn recruits POLB and the AP site is cleaved, causing an indirect
SSB (Caldecott 2003). Ligation of these short-repair patches are conducted by DNA ligase
III (LIG3) (Caldecott 2003). Spontaneous SSBs resulting from exposure to ROS are
recognized by PARP, which in turn recruits XRCC1 (de Murcia, Schreiber et al. 1994,
Ziegler and Oei 2001). Again, this results in an XRCC1-POLB complex, and finally ligation
by LIG3 (Caldecott 2003).
1.2.2.2 Non Homologous End Joining pathway (NHEJ)
NHEJ is the principal pathway for DSB repair in mammalian cells, and is typically imprecise,
which can greatly contribute to the genetic changes needed for tumor promotion (Lieber,
Ma et al. 2003). The protein Ku, which consists of two subunits (Ku70 and Ku86, coded by
XRCC5 and XRCC6 respectively), recognizes DSBs (Karanjawala, Adachi et al. 2002, Li,
Yang et al. 2011). Ku is the regulatory subunit of DNA-PK, and can bind DNA ends. The
19
catalytic subunit of DNA-PK is coded by PRKDC and is recruited by Ku. This in turn loads
Artemis, coded by DCLRE1C to the DNA complex, and allows opening of hairpins and
synapsis of broken DNA ends (Lieber, Ma et al. 2003). Finally, rejoining of the DNA break
is done by DNA ligase IV (LIG4), which is stabilized and enhanced by its interaction with
XRCC4 (Lieber, Ma et al. 2003).
1.2.2.3 Homologous Recombinational Repair pathway (HRR)
HRR also repairs DSBs and serves as a backup for damage not repaired by NHEJ (Honma,
Izumi et al. 2003, Lieber, Ma et al. 2003). This pathway is more conservative, and less
prominent, than NHEJ and thus it may not be as important in cancer progression (Honma,
Izumi et al. 2003). One of the first steps in HRR is the binding of the MRN complex,
formed by the MRE11, RAD50 and NBN proteins (Valerie and Povirk 2003). This MRN
complex is responsible for resecting the 5’ ends of DSBs, creating 3’ single stranded DNA
tails (D'Amours and Jackson 2002). RAD51, an essential recombinase, polymerizes on these
ssDNAs, leading to the formation of nucleoprotein filaments (Qing, Yamazoe et al. 2011).
The polymerization of RAD51 at damage sites is strictly regulated by a number of accessory
factors including RAD52 (Qing, Yamazoe et al. 2011). The recruitment of RAD52 to these
overhanging tails forms a DNA-protein complex that is thought to be stabilized by both
XRCC2 and XRCC3 (Liu, Lamerdin et al. 1998). It is not known as to which DNA
polymerase and ligase are involved in the polymerization and ligation steps of HRR (Valerie
and Povirk 2003).
20
1.2.2.4 Nucleotide Excision Repair pathway (NER)
The first step involved in NER is recognition of the DNA distortion caused by bulky
adducts or alkylated bases. The XPC-TFIIH protein complex facilitates this recognition.
TFIIH is a ten-subunit complex involving proteins such as XPB and XPD, coded for by
ERCC5 (Gillet and Scharer 2006). The XPC-TFIIH complex in turn allows assembly of the
“pre-incision” complex, which results in opening of the DNA helix. The following step is
dual excision around the damage site, carried out by the concerted action of XPG and
ERCC1-XPF (Gillet and Scharer 2006). XPF is coded for by the ERCC4 gene. Finally, the
DNA polymerase responsible for filling the resulting gap is POLD1, and the nick is sealed
by the DNA ligase LIG1 (Gillet and Scharer 2006).
1.2.3 GWAS Studies and colorectal cancer susceptibility genes
Genome-wide association studies (GWAS) are an examination of many common genetic
variants, single-nucleotide polymorphisms (SNPs), in different individuals to see if any
particular variant is associated with a trait. In contrast to methods that specifically test one or
a few genetic regions, GWAS investigate the entire genome. This approach is therefore said
to be agnostic, as opposed to gene-specific candidate-driven studies (Manolio 2010). To date,
over 1200 human GWAS have been performed, examining over 200 diseases and traits. Of
these, 14 GWAS have been undertaken to look at variants associated with the risk of
colorectal cancer (CRC), and 16 unique SNPs have shown a positive association in one or
21
more of these studies (Hindorff 2013). Investigation of the location and function of these
identified variants may now lead to identifying genes or pathways important for the
promotion and/or progression of CRC. Specific to this discussion, we wish to consider the
interaction of these identified variants with common exposures known to influence the risk
of CRC, i.e. in this case, meat related exposures. Exploratory analyses such as this, may lead
to identification of new pathways that may help to explain the risk conferred by certain meat
exposures on CRC risk, and/or help identify key elements of the exposures that may be
playing a predominant role in CRC carcinogenesis.
1.3 Selected Cancers
The research that will be presented in this thesis is focused on three main cancers: prostate,
bladder and colorectal cancer. This section will describe the key characteristics and
established risk factors for each of these three cancers. In addition, this review will be
centered on the role of dietary sources of the carcinogens discussed previously and the
current knowledge of their associations with these three cancers.
22
1.3.1 Prostate Cancer: Overview and Risk Factors
Prostate cancer (PCA) is the most common cancer among men in the US, and the second
most common cancer worldwide, after lung cancer (WCRF 2007). Based on rates from
2007-2009, 1 in 6 men born in the US today will be diagnosed with cancer of the prostate at
some time during their lifetime. Black males have the highest incidence of PCA, with 236
cases per 100,000 men and a mortality rate of 53 deaths per 100,000 men; this is nearly twice
the rate of white males who have an incidence of 147 cases per 100,000 men, and a mortality
rate of 22 deaths per 100,000 men. Asian and American Indian males are least at risk with an
incidence rate of 80 cases per 100,000 men. The reasons for these racial disparities are not
clear.
Few well-established risk factors for PCA have been identified, other than family
history of cancer, age, and selected genetic variants identified from genome-wide association
studies. However, the disparities in incidence between high- and low-risk regions and in
migrant populations point to putative lifestyle and/or environmental risk factors, with
adoption of Western dietary habits and lifestyle as a likely potential factor (Hsing, Tsao et al.
2000). According to the World Cancer Research Fund (WCRF 2007), diets high in calcium
are probable risk factors for PCA, and diets high in selenium and lycopene are most likely
protective (Figure 1.7). Diets high in meat have also previously been identified as plausible
risk factors for PCA. However, in their most recent report the WCRF classifies the evidence
for red meat, poultry and fish intake in risk of PCA as too limited to make any conclusions,
with consumption of processed meat classified as a suggestive risk factor for PCA, and a
recommendation to limiting intake (Figure 1.7). PAH, HCA and NOC exposure, could all
23
play a role in the association between meat and PCA risk, as the prostate gland is able to
metabolize each of these chemicals into activated carcinogens (Shimada and Fujii-Kuriyama
2004, Di Paolo, Teitel et al. 2005). In addition, several further mechanisms have been
proposed to contribute to the association. Meats are a rich source of dietary fat, which has
long been proposed as a PCA risk factor (Huang, Narita et al. 2012), whereas fruit and
vegetables contain antioxidants, which have anti-carcinogenic properties, and may be
displaced by diets high in meat (Steinmetz and Potter 1991). Red meat also contains a high
amount of iron in the form of heme, which stimulates the endogenous formation of N-
nitroso compounds (NOCs) as well as catalyzes a number of additional carcinogenic
oxidative reactions (Venkateswaran and Klotz 2010). Despite this biological plausibility for
an association between PCA risk and meat intake, results from previous studies have been
inconclusive. Whereas several cohort studies have reported positive associations (Rodriguez,
McCullough et al. 2006, Sinha, Park et al. 2009, Alexander, Mink et al. 2010), recent meta-
analyses and other cohort studies (Park, Murphy et al. 2007, Allen, Key et al. 2008,
Alexander, Mink et al. 2010) provide limited evidence for an association between total red
meat intake and PCA risk.. Similarly, the epidemiological evidence for associations between
fish and PCA risk is inconclusive with some cohort studies reporting inverse associations
(Augustsson, Michaud et al. 2003, Chavarro, Stampfer et al. 2008, Pham, Fujino et al. 2009,
Szymanski, Wheeler et al. 2010), others reporting positive associations (Mills, Beeson et al.
1989, Allen, Sauvaget et al. 2004), and some reporting no associations (Astorg 2010). When
considering preserved or processed meat, three previous cohort studies have reported an
increase in risk with increased intake of preserved meats and risk of PCA, two within US
cohorts, and one in a cohort from the Netherlands (Schuurman, van den Brandt et al. 1999,
24
Rodriguez, McCullough et al. 2006, Sinha, Park et al. 2009). Both of the US cohorts found
this increase in risk to be stronger among metastatic disease (Rodriguez, McCullough et al.
2006, Sinha, Park et al. 2009). Three further cohort studies, all US-based, reported suggestive
but non-statistically significant associations with preserved meat intake and PCA risk
(Michaud, Augustsson et al. 2001, Cross, Peters et al. 2005, Rohrmann, Platz et al. 2007), and
two of these studies found the association to be stronger among metastatic disease
(Michaud, Augustsson et al. 2001, Cross, Peters et al. 2005). Two additional cohorts, in the
US and Europe (EPIC), reported no association (Park, Murphy et al. 2007, Allen, Key et al.
2008). It should be noted that most of these previous studies were conducted in Western
populations where the proportion of diet from meat and fish intake is vastly different than in
Asian and other populations. These conflicting results emphasize the need for more research
and clarification in this area, and perhaps reflect the impreciseness and inherent inaccuracy
involved with using standard food frequency questionnaires (FFQs). Using FFQs that are
tailored to estimate HCA and PAH formation, and thus take into account cooking method,
time and other cooking variables such as use of marinades, may help refine future estimates
and give us a better insight into the association between dietary carcinogens and the risk of
PCA.
25
Figure 1.7: WCRF panel judgment on dietary factors and PCA risk
1.3.2 Bladder Cancer: Overview and Risk Factors
Bladder cancer is the 6
th
most common cancer worldwide and is predominantly a disease of
high-income countries such as the US, where it is now the 3
rd
most common malignancy
(GLOBOCAN 2008, Cancer Incidence and Mortality Worldwide). Identified risk factors
include white race, male gender, some occupational exposures, and a history of smoking
26
tobacco – with smoking accounting for nearly 50% of all bladder cancer cases in the US
(IARC Monographs on the Evaluation of Carcinogenic Risks to Humans, 2007)(Brennan,
Bogillot et al. 2000). The role of diet in bladder cancer is only partially understood.
Epidemiological evidence in the past has supported a protective role for diets high in
calcium, yellow-orange vegetables, citrus fruits, and cruciferous vegetables, although, the
WCRF recently downgraded this protective role from probable to inconclusive
(WCRF/AICR (2007).) (Figure 1.8). Diets high in meats, especially processed meats, diets
high in fat, and intake of artificial sweeteners have been implicated as risk factors (Brinkman
and Zeegers 2008, Silberstein and Parsons 2010). Previous studies that have investigated the
relationship between meat intake and risk of bladder cancer have arrived at inconclusive
results. A positive association between diets high in red meat intake and bladder cancer risk
was reported in three cohort studies (Aune, De Stefani et al. 2009, Key, Appleby et al. 2009,
Ferrucci, Sinha et al. 2010), whereas other cohort studies have shown no association
(Michaud, Holick et al. 2006, Cross, Leitzmann et al. 2007, Larsson, Johansson et al. 2009,
Jakszyn, Gonzalez et al. 2011). In the study by Ferruci et al. (Ferrucci, Sinha et al. 2010), the
observed positive association between red meat intake and bladder cancer appeared to be
attributable to consumption of processed red meat. However, other studies did not find
evidence of an association between intake of processed meat and bladder cancer (Lumbreras,
Garte et al. 2008, Larsson, Johansson et al. 2009). Only one study has investigated the role of
heme intake with bladder cancer and found no association (Jakszyn, Gonzalez et al. 2011).
Two cohort studies that considered estimated levels of NOC intake in relation to bladder
cancer risk have reported null results (Zeegers, Selen et al. 2006, Ferrucci, Sinha et al. 2010),
while one case-control study conducted in Japan reported risk to be associated with
27
increasing levels of nitrite intake (Wilkens, Kadir et al. 1996). In contrast, several studies
have investigated exposure to endogenous nitrosation as risk factors for gastric and colon
cancers. Due to the fact that NOCs formed in the stomach or intestines often pass through
the bladder before excretion, this provides a route of exposure to these carcinogens that
deserves consideration (Tricker 1997, Vermeer, Pachen et al. 1998, Ferrucci, Sinha et al.
2010).
Figure 1.8: WCRF panel judgment on dietary factors and bladder cancer risk
28
1.3.3 Colorectal Cancer: Overview and Risk Factors
Colorectal cancer (CRC) is the 3
rd
most diagnosed cancer worldwide, and leads to greater
than 600,000 deaths per year (Center, Jemal et al. 2009). Similarly, to PCA, incidence rates of
CRC are stabilized in Western countries, while they continue to increase in developing
countries, indicating a significant role for environmental and lifestyle factors in colorectal
carcinogenesis (Center, Jemal et al. 2009). The WCRF finds the most convincing evidence of
an association between meat intake and risk of cancer when considering cancers of the colon
and rectum (Figure 1.9), especially with red and processed meats. A recent meta-analysis
focusing solely on prospective studies supports the findings of the WCRF (Chan, Lau et al.
2011). The accumulated evidence from prospective studies indicates that for every 100 g/day
increase of total red and processed meats, there is a 14% increase in risk of CRC, a 25%
increase in risk of colon cancer, and a 31% increase in risk of rectal cancer (Chan, Lau et al.
2011). When looking at processed meats separately, three previously published meta-analyses
have examined processed meat consumption and CRC risk and all have found significant
associations indicating an approximately 10-20% increase in CRC risk with high processed
meat consumption (Sandhu, White et al. 2001, Norat, Bingham et al. 2005, Larsson and
Wolk 2006). Interestingly, these associations were much stronger among men than women.
Since the latest meta-analysis published in 2006, several important prospective studies have
been published in this area. Two of these find very strong associations between processed
meat intake and risk of CRC (Oba, Shimizu et al. 2006, Cross, Leitzmann et al. 2007), while
two further studies find no association (Sato, Nakaya et al. 2006, Nothlings, Yamamoto et al.
2009). Regarding fish intake, Wu et al. recently conducted a meta-analysis involving 22
29
prospective cohort and 19 case-control studies, and found evidence of a significant 12%
reduction in CRC with high intake of fish (Wu, Feng et al. 2012). However, very few of the
included studies took into account cooking methods of the fish variables, which may partly
explain the significant heterogeneity observed by Wu et al. between studies. Given the
consistent findings with meat intake, it makes sense to study HCAs, PAHs and NOCs as
potential CRC risk factors. Since the introduction of the CHARRED database has made
estimating HCAs and PAHs from food frequency questionnaires feasible, a number of
epidemiological studies have considered estimated levels of HCAs (Le Marchand, Hankin et
al. 2002, Nowell, Coles et al. 2002, Butler, Sinha et al. 2003, Cross, Ferrucci et al. 2010), and
support a role for HCAs in colorectal cancer risk, although results are not conclusive. A
recent large prospective study found an association of MeIQx, DiMeIQx with the risk of
colon cancer, but not rectal cancer (Cross, Ferrucci et al. 2010). This study also found
associations between heme iron and rectal cancer risk and between nitrates from processed
meats and colorectal cancer, indicating a potential role for NOCs in CRC carcinogenesis.
30
Figure 1.9: WCRF panel judgment on dietary factors and CRC risk
31
Chapter 2: Hypothesis & Specific Aims
The underlying hypothesis of this thesis is that three groups of dietary carcinogens, HCAs,
PAHs and NOCs, found in cooked and processed meats and fish are risk factors for cancer.
Secondarily, I hypothesized that the inherent genetic variation in the genes that code for
enzymes responsible for metabolizing these compounds, and/or repairing the damage
induced by them, modifies their carcinogenic effect on cancer risk.
To test these hypotheses I proposed the following specific aims:
Specific Aim 1: To determine the role of dietary carcinogens present in fish and meat on
the risk of prostate cancer taking into account genetic variation in carcinogen metabolism
pathways (Section 4.1)
Aim 1A: To investigate the main effects of genetic polymorphisms in genes
involved in the HCA and PAH metabolism pathways on the risk of advanced and
localized prostate cancers in California (Section 4.1.1).
Aim 1B: To investigate the modifying role of the above polymorphisms on fish
intake and prostate cancer risk, while taking into account cooking methods (Section
4.1.2).
Aim 1C: To investigate the main effects of fish, red meat and poultry intake on
prostate cancer in Singapore, where fish intake is very high (Section 4.1.3).
32
Specific Aim 2: To determine the role of dietary sources of N-nitroso compounds (NOCs)
found in meats, particularly processed meats, on the risk of bladder cancer taking into
account genetic variation in DNA repair pathways (Section 4.2).
Aim 2A: To investigate the associations between intakes of specific processed and
organ meats and estimated levels of exogenous NOCs with the risk of bladder cancer
(Section 4.2.1).
Aim 2B: To investigate the role of precursors of endogenous nitrosation found in
the diet and the risk of bladder cancer (Section 4.2.2).
Aim 2C: To investigate the modifying role of polymorphisms in metabolism
enzymes and DNA repair pathways on the above exposures and their association
with bladder cancer risk (Section 4.2.3).
Specific Aim 3: To determine the role of dietary carcinogens found in meat and fish on the
risk of colorectal cancer taking into account genetic risk factors (Section 4.3).
Aim 3A: To investigate the main effects of fish intake and cooking methods on the
risk of colorectal cancer, while taking into account cooking methods (Section 4.3.1).
Aim 3B: To investigate the modifying role of genetic risk factors identified through
genome-wide association studies on the association between meat and fish intake
and colorectal cancer risk, taking into account cooking methods (Section 4.3.2).
33
Chapter 3: Methods
3.1 Study Design and Study Population
3.1.1 California Collaborative Case-Control Study of Advanced Prostate
Cancer
The California Collaborative Case-Control Study of Prostate Cancer is a multiethnic,
population-based case-control study that identified incident cases of prostate cancer through
two regional cancer registries in Los Angeles County and the San Francisco Bay Area (John,
Schwartz et al. 2005, Schwartz, John et al. 2010). In both study sites, PCA was classified as
advanced disease if the tumor: extended beyond the prostatic capsule or into the adjacent
tissue, involved regional lymph nodes, or metastasized to distant locations (SEER 1995
clinical and pathologic extent of disease codes 41-85). Intra-capsular prostatic cancers were
categorized as localized disease. In Los Angeles county (LAC), controls were identified
through a standard neighborhood walk algorithm (Pike, Peters et al. 1997) and frequency-
matched based on age (±5 years) and race/ethnicity. In the San Francisco Bay Area (SFBA),
controls were primarily identified through random-digit dialing. In addition, a subset of
controls aged ≥65 years were also identified through random selections from the rosters of
beneficiaries of the Health Care Financing Administration (HCFA). Controls were
frequency-matched to cases with advanced disease based on 5-year age group and
34
race/ethnicity. In LAC, study eligibility was restricted to men willing to both complete an in-
person interview and provide a blood sample; 1,232 cases (376 AA, 355 Hispanics and 501
NHW), of 1,870 eligible cases diagnosed from 1999-2003, and 594 controls (163 AA, 122
Hispanics and 309 NHW) participated in the study. In SFBA, in-person interviews were
completed with 568 (118 AA and 450 NHW) advanced cases (of 788 eligible cases diagnosed
from 1997-2000), and 545 (90 AA and 455 NHW) control men of 868 eligible controls.
Blood or mouthwash samples were obtained for 533 advanced cases (107 AA and 426
NHW) and 525 controls (85 AA and 440 NHW). No biospecimen samples were collected
for localized cases(John, Schwartz et al. 2005). At the time we performed the genotyping for
this study, DNA from blood and data on risk factors were available for a total of 800
controls (545 from LAC and 255 from SFBA), 535 localized cases (LAC) and 988 advanced
cases (597 from LAC and 391 from SFBA). Written informed consent was obtained from all
study participants at the time of in-person interview.
3.1.2 Singapore Chinese Health Study
The Singapore Chinese Health Study is a residential cohort of 63,257 middle-aged and older
(45-74 years) Singapore Chinese men and women assembled between 1993 and 1998
(Hankin, Stram et al. 2001). Enrollment in the cohort was restricted to individuals belonging
to the two major dialect groups of Chinese in Singapore, the Hokkiens and the Cantonese.
The Institutional Review Boards at the National University of Singapore and the University
of Pittsburgh have approved this study. These analyses were performed using data from the
35
27,293 men who did not have a history of cancer diagnosis at baseline. PCA cases were
identified through passive following of the cohort for death and cancer occurrence via
regular linkage with the population-based Singapore Cancer Registry and the Singapore
Registry of Births and Deaths, as we have previously described (Butler, Wong et al. 2010). As
of December 31
st
, 2007, 298 study subjects had developed PCA. For analyses in this study
we grouped patients into those with early disease and those with advanced disease. Early
(localized) disease included cases with pathologically staged organ-confined disease (T1–T2),
or clinically staged organ-confined disease and with either D’Amico good or intermediate
risk. Advanced cases included patients with pathologically staged extra-capsular (T3–T4) or
nodal involvement (N1), or clinically staged extra-prostatic extension or regional nodal
disease, or clinically organ-confined disease and with D’Amico high risk, or patients with
metastatic disease, which was defined as those with evidence of distant metastases, based on
clinical information. To summarize, 124 cases were determined to be early stage disease, and
162 cases were determined to be locally advanced or metastatic disease at diagnosis. 12 cases
did not have sufficient information to determine stage of disease.
3.1.3 Los Angeles Bladder Cancer Study
Incident cases of bladder cancer, specifically transitional cell carcinoma, were identified
through the Los Angeles County Cancer Surveillance Program, the population-based
Surveillance, Epidemiology and End Results (SEER) cancer registry of Los Angeles
County(Castelao, Yuan et al. 2001). Eligibility criteria included histologically confirmed first
36
urothelial carcinoma diagnosis between January 1, 1987 and April 30, 1996 among non-Asian
individuals between the ages of 25 and 64 years. Of 2098 eligible cases, 1671 (80%) were
enrolled in this study. For each enrolled case a control individual was recruited from the
neighborhood where the index case resided at the time of diagnosis. Controls were matched
by age (within 5 years), gender and race/ethnicity (non-Hispanic white, Hispanic, African
American). If a matched control was not found after surveying 150 households, the race
criterion was dropped. A race/ethnicity-matched control could not be found for 86 cases. If
a control could not be found after surveying an additional 150 households under this relaxed
criterion, the search was halted. A matched control was found for 1586 (95%) of cases. 11
cases were later determined not to be transitional cell carcinoma and so were excluded. Our
study included 1,660 cases (1,307 men and 353 women) and 1,586 controls (1,237 men and
349 women). All subjects signed informed consent forms approved by the University of
Southern California Human Subjects Committee.
3.1.4 Colorectal Cancer Family Registry
The Colorectal Cancer Family Registry (Colon CFR) is an international consortium of six
centers, formed to support studies on the etiology, prevention, and clinical management of
colorectal cancer (CRC) (Newcomb, Baron et al. 2007). The participating centers are Cancer
Care Ontario (CCO), Fred Hutchinson Cancer Research Center (FHCRC), Mayo Clinic,
University of Hawaii (UHI), University of Southern California (USC) and University of
Melbourne (Australasia) (Figure 2.1). The six centers have used varying designs and sampling
37
schemes to recruit families. Phase I population based recruitment began in 1997 and was
completed in July 2002. All six centers recruited families with incident CRC identified
through population-based cancer registries in defined geographical areas. Sampling
probabilities differed from center to center according to age at diagnosis, race, and/or family
history of CRC (Newcomb, Baron et al. 2007). During Phase I, controls were either
randomly sampled from the general population (CCO, FHCRC, Australasia), or ascertained
through the affected proband, either as the spouse (Mayo, UHI, Australasia) or an
unaffected family member (USC). Three centers also used clinic-based recruitment during
Phase I (Mayo, USC, and Australasia). Clinic based probands had one or more of the
following criteria: two or more relatives with a personal history of CRC or Lynch syndrome,
proband diagnosed with CRC at a young age, or proband presenting with Lynch syndrome
or Lynch-like syndrome. Controls were ascertained through the proband, either as the
spouse (Mayo, Australasia) or an unaffected family member (USC).
38
USC consortium
USC, ACC, UC, UM, CC,
DC, UNC (N = 2,789)
Hawaii
UH
(N = 1,047)
Mayo
Clinic
(N = 1,658)
Seattle
FHCRC
(N= 5,937)
Australasian CRC
Family Study
AU, NZ (N = 2,841)
Ontario (N = 4,864)
Colon Cancer Family
Registry
Participating Sites
IC
GU
Figure 3.1: Colon CFR Participating Centers
Phase II CFR recruitment began in 2002 and was focused on increasing the proportion of
study individuals with genetic factors and family history. Recruitment was expanded to
include more recently diagnosed population-based probands under the age of 50, and their
first-degree relatives, and more clinic-based families presenting at cancer family clinics.
FHCRC continued to recruit population-based controls throughout Phase II. Confirmation
of cancer diagnoses was sought for all reported CRCs. Pathology reports for all affected
probands’ qualifying colorectal cancers were reviewed, and data, including site, stage,
histology, grade, distant spread, and nodular involvement were abstracted onto a
39
standardized review form developed by the Pathology Working Group. Our study was
restricted to those individuals with detailed dietary information. In summary this study
included 3350 CRC affected probands (1992 colon, 907 rectum, 451 unspecified), 1607
population-based controls, 138 spouse controls and 1761 unaffected family members
(siblings).
3.2 Data Collection
3.2.1 California Collaborative Case-Control Study of Advanced Prostate
Cancer
A common structured questionnaire was used at both study sites, including a 74-item food
frequency questionnaire (FFQ) that was adapted from Block’s Health History and Habits
Questionnaire(John, Schwartz et al. 2005). The FFQ also included questions on cooking
methods and degree of doneness and browning (John, Stern et al. 2011) that were adapted
from a commonly used cooking module developed by Sinha et. al. (Sinha, Cross et al. 2005).
An aggregate level socio-economic status (SES) variable was derived from 2000 census data
(Schwartz, John et al. 2010). Body mass index (BMI) was calculated using the reported
weight in the reference year (defined as the calendar year before diagnosis for cases and the
calendar year before selection into the study for controls) and measured height at the time of
the interview. BMI was calculated as weight (in kilograms) divided by height (in meters)
40
squared and categorized as normal weight (BMI <25), overweight (BMI 25-29.9) and obese
(BMI ≥30). Underweight men (BMI <18.5, n=15) were grouped with normal-weight men.
The FFQ assessed all food intake and cooking methods during the reference year. Subjects
reported the usual portion size and frequency of consumption of tuna fish (including tuna
salad, tuna casserole and tuna sandwiches), deep fried fish (including fish sticks and fish
sandwiches), dark fish (salmon, mackerel, catfish, trout, herring, sardines) and white fish
(flounder, halibut, snapper, bass, cod or sole). Information was obtained on usual method of
preparation of dark fish and white fish (pan-frying, oven-broiling, grilling, baking/roasting,
microwaving and other methods). The participants also reported their usual preference for
the level of doneness of dark fish and white fish by choosing from a series of color
photographs showing three levels of doneness and browning: (1) just until done (not
browned on the outside), (2) well done (browned on the outside) and (3) very well done
(charred on the outside). There were only 74 men who usually consumed fish “very well
done - charred on the outside”; therefore, men in this category were combined with men
who usually ate fish well done. Cooking by pan-frying, oven-broiling or grilling, were
considered high temperature cooking methods, and cooking by baking or other methods
were considered to be low temperature cooking methods.
3.2.2 Singapore Chinese Health Study
At recruitment, all participants in the cohort were interviewed in their homes by trained
interviewers using a structured questionnaire. Information was collected on diet, smoking,
41
physical activity, medical history, occupation and other demographics. Current diet was
assessed via a 165-item food frequency questionnaire that assessed frequency and usual
serving size and was developed for and validated in this population (Hankin, Stram et al.
2001). Summary variables were created from the individual food items commonly consumed
by Singapore Chinese: total fresh red meat (including minced pork or beef patty, pork
spareribs, pork satay, pork slices, belly pork, mutton curry and hamburger), total preserved
red meat (including Chinese sausage, ham, pork hot dogs, luncheon meat, meat floss and
sweet barbeque meat), total organ meat (including pork liver and pork intestine), total
poultry (including deep fried chicken, pan or stir fried chicken, soy sauce chicken, chicken
satay, boiled or stewed chicken, chicken curry and roasted or stewed duck or goose), total
fresh fish and shellfish (including fish ball or cake, deep fried fish, pan or stir fried fish,
boiled or steamed fish, shrimp or prawn, squid or cuttlefish), and total preserved fish and
shellfish (including salted fish, ikan bilis, dried fish, other dried seafood such as dried shrimp,
dried oyster, dried cuttlefish, canned tuna, canned sardine).
3.2.3 Los Angeles Bladder Cancer Study
In-person structured interviews were conducted in participants’ homes. The questionnaire
requested information up to 2 years prior to the diagnosis of cancer for cases and 2 years
prior to diagnosis of cancer of the index case for matched controls. The questionnaire
included information on demographic characteristics, height, weight, lifetime use of tobacco
and alcohol, usual adult dietary habits, lifetime occupational history, prior medical conditions
42
and prior use of medications. Forty food groups were included in the dietary section of the
structured questionnaire. These foods were chosen to allow assessment of dietary vitamins A
and C, and pre-formed nitrosamines. Detailed dietary questions regarding intake of
processed meats were asked including assessing intake of fried bacon, ham, salami, pastrami,
corned beef, bologna, other lunch meats, hot dogs and Polish sausage. For each food group
the subject was asked to indicate his/her usual intake frequency during the reference year.
These values were then converted to frequency of intake per week. For any given food that
the individual ingested with uneven frequency during the year, his/her usual intake
frequency was obtained for each of the distinct seasons and then averaged. The
questionnaire also asked about current use of multiple vitamin supplements; frequency of
use and brand names of supplements were recorded. To compute intake of nutrients
including nitrates, nitrites, nitrosamines and heme, a standard portion size was assumed for
each of the individual food items that comprised the 40 food groups listed in the study
questionnaire. Weights (which summed to 100%) were assigned to individual food items in
each food group consisting of multiple items. Gram weights of food were linked to food
composition values reported by the US Department of Agriculture (USDA nutrient
database, 1994.). Heme iron was estimated from processed meats and liver only. The amount
of heme iron in meat ranges from 22-79% of total iron depending on the type of meat. As
the processed meats studied here can often be comprised of more than one type of meat, we
estimated a heme-iron content by multiplying total iron by a factor of 0.4, a method
proposed by Monsen and used in previous studies (Monsen 1988).
43
3.2.4 Colorectal Cancer Family Registry
All Colon CFR centers used a standardized core questionnaire collecting details on family
history and personal exposures (Risk Factor Questionnaire). This questionnaire was
completed either in person (USC), by telephone (FHCRC, USC, Australasia), or by mail
(CCO, UHI, Mayo). Information on established and suspected risk factors for CRC was
captured, including medical history, reproductive history (females), physical activity,
demographics, alcohol and tobacco use, race and ethnicity and limited dietary data. Four
Colon CFR centers administered a detailed dietary food frequency questionnaire (UHI,
CCO, USC, Australasia). The dietary questionnaire developed at UHI for use by the
Multiethnic Cohort study in Hawaii and California was used at UHI, CCO, and USC (Stram,
Hankin et al. 2000). Australasian centers used a locally validated dietary questionnaire. This
study was restricted to only those participants who completed the UHI developed
questionnaire, and thus only included participants from UHI, CCO, and USC. The UHI
dietary questionnaire was comprised of a food frequency portion in addition to a cooking
module. The individual fish variables asked about in the food frequency portion were as
follows: fried shrimp or other shellfish; cooked, canned or raw shellfish; fried fish; baked,
broiled, boiled or raw fish; canned tunafish; other canned fish; salted and dried fish. In the
cooking module, information was collected on intake frequency of pan-fried fish, oven-
broiled fish and grilled or barbequed fish in addition to level of doneness (light brown,
medium brown, dark brown). Combining the above variables created a summary variable of
total fish, and then this was separated into two further summary variables: total fish
(excluding shellfish) and total shellfish.
44
3.3 SNP selection and Genotyping
3.3.1 California Collaborative Case-Control Study of Advanced Prostate
Cancer
We genotyped individuals for 12 SNPs in 9 genes: GSTP1 Ile105Val (rs1695), PTGS2 -765
G/C (rs20417), CYP1A2 -154 A/C (rs762551), EPHX1 Tyr113His (rs1051740), CYP1B1
Leu432Val (rs1056836), UGT1A6 Thr181Ala and Arg184Ser (rs1105879, rs2070959) and
NAT2 Arg197Gln, Gly286Glu, Arg64Gln and Ile114Thr (rs1799930, rs1799931, rs1801279,
rs1801280), in addition to two genes that had copy number variants, GSTM1 and GSTT1.
All genotypes were obtained using Taqman assays, available on request from ABI (Applied
Biosystems, Foster City, CA), following manufacturer's instructions. No differences were
found between observed genotypic frequencies and those expected under the Hardy-
Weinberg Equilibrium. Call rates were >97%. The four NAT2 SNPs genotyped infer the
NAT2 phenotype with estimated 98.4% accuracy in populations with similar SNP
frequencies as in this study (Hein and Doll 2012). In agreement with the existing
classification, carriers of at least one copy of the wild type haplotype were classified as “fast”
and carriers of all other haplotypes as “slow” phenotype (Hein, Grant et al. 2000).
45
3.3.2 Los Angeles Bladder Cancer Study
From the Los Angeles Bladder Cancer Study, we had 355 cases and 409 controls with
available DNA. These study participants were genotyped for 627 tag SNPs (tSNPs) across
the four investigated DNA repair pathways. To summarize, we genotyped 127 tSNPs across
7 genes from the BER pathway (XRCC1, OGG1, PARP, APE1, NEIL1, POLB and LIG3),
156 tSNPs across 7 genes from the HRR pathway (RAD51, RAD52, XRCC2, XRCC3,
NBN, MRE11 and RAD50), 187 tSNPs from the NHEJ pathway (XRCC4, XRCC5,
XRCC6, PRKDC, DCLRE1C and LIG4) and 156 tSNPs from the NER pathway (XPA,
XPC, ERCC2, ERCC4, ERCC5, ERCC1, POLD1 and LIG1). tSNPs were selected using the
program Snagger (Edlund, Lee et al. 2008) based on data from the White CEPH population
(Utah residents with ancestry from northern and western Europe) found in the International
HapMap Project (HapMap, release 21, July 2006). tSNPs were selected to cover all SNPs
with an MAF of 0.05 or greater, with a pairwise r
2
of ≥0.80 in the region covering each gene
of interest as well as an added 20 kb upstream and 10 kb downstream of the gene.
Genotyping of the tSNPs was completed in the USC Norris Comprehensive Cancer Center’s
genomics core facility, using the Illumina Golden Gate Assay
TM
(Illumina, Inc., San Diego,
CA, USA). Samples were excluded from this analysis if the genotyping success rate was
<90% after eliminating SNPs that failed genotyping completely. Samples were also excluded
if they had a minor allele frequency (MAF) of less than 5% in this study population. After
these exclusions we were left with 549 tSNPs ready for analysis. Deviation from HWE in
non-Hispanic whites was evaluated for each SNP and quality control measures also included
using blinded duplicate samples and mixing cases and controls on genotyping plates.
46
3.3.3 Colorectal Cancer Family Registry
All participants provided a blood sample at the time of recruitment. DNA samples were
genotyped with the Illumina Human1M (n individuals=1,973; m=1,072,820 SNPs) or
Human1M-Duo (n individuals=374; m=1,199,187 SNPs) BeadChip platforms. Samples with
GenCall scores <0.15 at any locus were considered ‘no calls’. Each 96-well plate included
one inter-plate positive quality control sample (NA06990 - Coriell Cell Repositories). In
addition, 27 blinded and 22 un-blinded quality control replicates from the study sample were
genotyped. SNP data obtained from both the Coriell and study sample replicates showed a
very high concordance rate of called genotypes: 99.95% and >99.94%, respectively (for
samples with call rates >90%). The Human1M and Human1M-Duo contain 415 and 436
SNPs, respectively, that were genotyped as part of a candidate gene study on the Illumina
GoldenGate platform on a subset of the individuals genotyped in this study (N=444). A
high concordance rate (>98%) was observed for >99% of the samples with a call rate >90%.
47
3.4 Statistical Analysis
3.4.1 California Collaborative Case-Control Study of Advanced Prostate
Cancer
In our study population, total caloric intake was associated with prostate cancer risk
(p<0.001). Residuals obtained from regression of fish intake variables on calories showed
non-normality and heteroscedasticity (even after log transformation) because our fish
exposure variables had few unique values. Hence, we used the multivariate nutrient density
approach to adjust for energy intake (Willett, Howe et al. 1997). Nutrient densities of white
fish and dark fish were created multiplying their reported daily intake with the reciprocal of
the total calories consumed per day. Nutrient density variables were categorized into three
levels of consumption: never/rarely (quintile 1), low (quintiles 2-4), and high (quintile 5).
These categories were chosen to circumvent spurious negative associations that are possibly
driven by the influence of large variation in the denominator of the density variable (inverse
calorie intake) as compared to the variance of the nutrient of interest in the numerator.
When tabulating the quintiles of nutrient density intake against the quintiles of total calorie
intake, we observed that the middle three quintile categories were strongly affected by this
phenomenon and the division among them could almost exclusively be explained by energy
intake, thus the middle three quintiles were collapsed. Given differences in the distribution
of race/ethnicity and SES between the two study sites, we created a variable that classified
subjects according to study site (LAC or SFBA), SES (5 level variable from low to high, as
48
previously described (Schwartz, John et al. 2010)), as well as race/ethnicity (AA, Hispanic,
NHW) to group individuals in conditional logistic regression models. SES was collapsed into
3 categories (quintiles 1-2, 3, 4-5) at the SFBA site and 4 categories (quintiles 1, 2, 3, 4-5) at
the LAC site, leaving 6 SES/race groups from SFBA and 12 from LAC. Odds ratios (ORs)
and 95% confidence intervals (CIs) were estimated from conditional logistic regression, with
separate analyses performed for localized and advanced cases. We estimated ORs and CIs
for each genotype using dummy variables and per variant allele assuming a log-additive
model. All OR estimates were adjusted for age (years, continuous) and family history of PCA
in first-degree relatives (yes, no). Minor allele frequencies (MAF), ranging from 0.16 to 0.49,
did not differ significantly from those reported for the general reference population
(www.hapmap.org). Our study design and sample size can detect ORs as small as 1.25
(MAF=0.49) - 1.34 (MAF=0.16) among localized cases and ORs as small as 1.21
(MAF=0.49) - 1.29 (MAF=0.16) among advanced cases with 80% power when testing at the
5% significance level (all power calculations performed with Quanto (Gauderman 2002)).
When investigating main effects of fish we also adjusted for potentially confounding lifestyle
and dietary factors. These variables were total fat intake (g/day), dietary vitamin D intake
(g/day), alcohol consumption (g/day), total dairy intake (g/day), cigarette smoking (pack-
years), total fruit consumption (g/day), total vegetable consumption (g/day), red meat
consumption (g/day), white meat consumption (g/day) and processed meat consumption
(g/day) (Table 4.6). We did not see any appreciable differences in estimates and overall
trends; therefore further gene-environment analyses were not adjusted for these additional
variables. Analyses of gene-by-diet interactions involved conducting one degree of freedom
likelihood ratio tests, with fish consumption (never, low, high) coded as an ordinal variable
49
and SNPs coded according to a log-additive model. Interaction models were adjusted for age
(years, continuous), BMI (<25, 25.0-29.9, ≥30), total calorie intake (kcal/day, continuous),
and family history of PCA in first-degree relatives (yes, no). Analyses of white fish were
adjusted for dark fish intake and vice versa. To analyze gene-by-fish interactions while taking
into account cooking practices, we created variables that captured frequency of intake of
either white or dark fish when cooked using high temperature methods (pan-frying, grilling,
oven-broiling and barbequing) or when cooked until well done. Models that tested for
interaction using variables that captured fish intake and cooking practices were also adjusted
for total well-done meat consumption (g/day) and consumption of total meat cooked at high
temperature (g/day). Total meat variables included red meat, poultry and processed meats.
Tests of trend across categories of fish consumption were done by coding each category by
its median value and modeling the category variable as continuous. To account for multiple
testing we used a p-value cutoff of 0.0056=0.05/9 for each of the exposures tested, which is
the Bonferroni corrected p-value = 0.05 for testing interactions between each given
exposure and 9 different genotypes or phenotypes. Further correction accounting for the
two stages of disease set the p-value cutoff at 0.0028=0.05/18. Our study design and sample
size can detect interaction ORs as small as 1.62 (MAF=0.49) - 1.84 (MAF=0.16) among
localized cases and interaction ORs as small as 1.49 (MAF=0.49) - 1.69 (MAF=0.16) among
advanced cases with 80% power when testing at the 5% significance level (all power
calculations performed with Quanto (Gauderman 2002)). All hypothesis tests were two-sided
and all analyses were done using the statistical software Stata S/E 11.0 for Windows
(STATA Corporation, College Station, TX). After excluding 124 individuals (87 cases, 37
50
controls) with dietary data considered unreliable (i.e., daily energy intake <600kcal or
>6000kcal), all dietary analyses were based on 1,420 cases and 746 controls.
3.4.2 Singapore Chinese Health Study
Cox proportional hazards models were used to calculate Hazard Ratios (HR) and 95%
confidence intervals (CI) that were used to estimate the association between dietary variables
and PCA risk. Person-years of follow-up were assessed starting from the recruitment date
and ending at the date of PCA diagnosis, death, loss to follow-up, or December 31
st
, 2007,
whichever occurred first. Dietary variables were adjusted for energy intake using the nutrient
density approach, i.e. multiplying their reported daily intake in grams with the reciprocal of
the total calories consumed per day. Quartiles of food variables were investigated for their
association with PCA risk. Linear trend tests were based on median values of the quartile
levels of intake. Associations were also considered separately for early and advanced disease.
All analyses were adjusted for age at baseline interview (years), dialect group (Cantonese,
Hokkien), year of interview (1993-94, 1995-96, 1997-98), family history of cancer, education
(no formal education/primary school/secondary school or higher), nutrient density adjusted
fat intake and caloric intake based on literature and previous reports from this study. Models
that estimated associations for fish variables were additionally adjusted for energy-adjusted
red meat intake, energy-adjusted poultry intake and energy-adjusted vegetable intake based
on a 10% change in the β estimate for trend of one or more of the fish variables. Meat
variables were additionally adjusted for smoking status (never/ex/current), number of
51
alcoholic drinks per week, energy-adjusted poultry intake, energy-adjusted dairy intake and
energy-adjusted vegetable intake based on a 10% change in the β estimate for trend of one
or more of the meat variables. Poultry variables were additionally adjusted for energy-
adjusted red meat intake based on a 10% change in the β estimate for trend of one or more
of the poultry variables. Additional inclusion of BMI, fruit intake, green tea consumption,
weekly use of a vitamin supplement, and weekly physical activity did not significantly change
any of the estimates. Based on previous reports from this study and the literature we
conducted stratified analyses using interaction terms between each potential modifier and
each meat variable and used likelihood ratio tests to examine whether the associations
between red meat, fish and poultry variables and PCA risk differed by the following: stage
(early/advanced PCA), age at baseline (below/above median), dialect group
(Cantonese/Hokkien), and BMI (below/above median). In addition, we also tested whether
fruit and vegetable intake and green tea intake modified the effect of these exposures, under
the hypothesis that both fruit and vegetables and green tea contribute a large amount of
dietary antioxidants and therefore individuals with low intake may be more susceptible to
dietary carcinogens. Moreover, we also tested whether smoking status modified the effect of
these exposures, given that smoking contains many of the same carcinogens that can
accumulate in cooked meats and can induce some of the enzymes that are required for their
activation and/or detoxification. All hypothesis tests were two-sided and all analyses were
done using the statistical software Stata S/E 11.0 for Windows (STATA Corporation,
College Station, TX).
52
3.4.3 Los Angeles Bladder Cancer Study
Study subjects were grouped into bins according to reference age (<38, 38-42, 43-47, 48-52,
53-57, and ≥58 years) and gender. Conditional logistic regression models, matching on these
bins, were used to estimate odds ratios (ORs) and 95% confidence intervals (95% CIs) for
the associations between dietary factors and/or nutrients and bladder cancer. There were
427 cases and 487 controls that were current users of vitamin supplements. There were no
appreciable differences in the associations between macronutrients and bladder cancer risk
whether or not nutritional data from supplements was included in the analysis. Thus this
study is restricted to nutrients derived from food sources only. Dietary variables were
categorized into quintiles based on the distribution among controls. Tests for trend were
conducted using the median level of exposure among controls at each level. We estimated
ORs using models that included intensity and duration of tobacco smoking, race/ethnicity,
level of education, history of diabetes, total vegetable intake in servings per day, vitamin A
intake in IU per week, vitamin C intake in mg per week, carotenoid intake in mcg per week
and reported number of servings of food per day. The covariates were selected based on
previous reports of association with bladder cancer in other studies and associations
estimated in this study (p<0.05, Table 4.11) (Castelao, Yuan et al. 2004). In addition, we also
included categorical BMI (<25, 25-<30 and ≥30), as a surrogate for total caloric intake,
which we could not compute as our questionnaire was not a full food frequency
questionnaire. Moreover, we created a variable that summed daily number of servings of
each reported food (total food servings), which we used as a further surrogate for caloric
intake under the assumption that intake of reported foods is highly correlated with intake of
53
all food. Interactions between intake of specific meats and dietary factors, smoking status
(never/ever) and gender were investigated using conditional logistic regression models that
included product terms (e.g. specific meat intake x estimated NOCs or gender) and
likelihood-ratio tests. Dietary factors that we investigated for interactions included NOCs,
fruit intake, vegetable intake, Vitamin A intake, Vitamin C intake, calcium intake and
carotenoid intake. We conducted both 4-degrees of freedom (df) interaction tests by treating
the intake exposure variable as categorical and 1-df interaction tests by treating this variable
as ordinal. Additional covariates included in interaction models were intensity and duration
of tobacco smoking, current smoking status, race/ethnicity, categorical BMI
(underweight/normal <25, overweight 25-30 and obese ≥30), level of education, history of
diabetes, total vegetable intake in servings per day, total servings of food per day, vitamin C
intake in mg per week, vitamin A intake in IU per week and total carotenoids in mcg per
week. We sub-typed tumors by risk of progression (low or high) following the classification
used by Kiemeney (low risk = Ta (non-invasive papillary carcinoma) tumors with WHO
1973 differentiation grade 1 or 2; high risk = T1-T4 stage tumors, carcinoma in situ (CIS),
and all tumors with WHO 1973 differentiation grades 3-4) (Kiemeney, Thorlacius et al.
2008). All Ta tumors were papillary. We tested for heterogeneity of the genotype effect
across different histological subtypes using case-only analyses.
Each of the individual meat variables, in addition to nitrite, nitrate and nitrosamine
intake, was further tested for interactions with each of the 627 SNPs from the four DNA
repair pathways. Again, these tests were conducted using conditional logistic regression
models that included product terms and likelihood-ratio tests. For these SNP by exposure
interactions, dietary variables were re-categorized into tertiles based on the distribution
54
among controls, due to the reduced sample size with DNA available. To account for
multiple testing, we used Bonferroni correction at three levels. We corrected for the number
of SNPs in each gene, the number of SNPs in each pathway, and then finally, across all
SNPs combined. We further stratified these analyses by gender, to see if there were
appreciable differences in associations between male and female. All statistical tests were
two-sided, and all analyses were conducted using the statistical software package STATA
version 11 (STATA Corporation, College Station, TX).
3.4.4 Colorectal Cancer Family Registry
In order to account for the three different types of controls in this dataset (population-
based, sibling and spouse controls), analyses were performed using a number of methods.
Predominantly, generalized estimating equation analysis was used to obtain Odds Ratios
(OR) and 95% confidence intervals (CI) for associations between fish variables and CRC risk
in models that included all three types of controls. Parameter estimates from GEE analyses
are less sensitive to variance in structure specification, and thus these types of analyses are
often used in multi-site cohort studies because they can handle many types of unmeasured
dependence between outcomes. We also performed analyses using sub-groups of the
controls for comparison with the GEE models. These included using 1:1 conditional logistic
regression models when restricting to spouse controls alone, N:M conditional logistic
regression models when restricting to sibling controls, and unconditional logistic regression
models when restriction to population-based controls. Dietary variables were categorized
55
into quintiles based on the distribution among controls. Tests for trend were conducted
using the median level of exposure among controls at each level. Dietary variables that were
further specified according to cooking method were categorized into quartiles based on the
distribution among controls, due to the decrease in sample size arising from this
specification. We estimated ORs using models that included age (years), BMI (<25, 25.0-
29.9, ≥30), total calorie intake (kcal/day, continuous), gender (male, female), recent physical
activity (yes, no), center (Hawaii, USC, Ontario), saturated fat intake (g/kcal, continuous),
fiber intake (g/kcal, continuous), vegetable intake (g/kcal, continuous) and race (NHW,
Black, AA, Other). We conducted stratified analyses to examine whether the associations
between fish variables and CRC risk differed by the following: site (colon, cecum and
appendix, ascending, transcending and descending colon, sigmoid colon, and rectum), center
(Hawaii, USC, Ontario), MMR status (proficient, deficient), and familial history of cases (yes,
no; based on Amsterdam criteria). We also used case only analysis to determine if there were
significant differences between the following sub-groups: site (colon vs. rectum), MMR
status (proficient vs. deficient) and familial history of cases (familial vs. non-familial; based
on Amsterdam criteria).
Individual meat and fish variables, in addition to indexes for HCAs, were further
tested for interactions with each of the 16 SNPs identified as top candidates from GWA
studies. Overall, we tested 16 different meat-related exposures with each of the 16 SNPs.
These variables were as follows: total meat, red meat, white meat, red processed meat, white
processed meat, beef, pork, organ meat, sausage, all fish, shellfish, non-shellfish, HAA, PhIP,
DiMeiQx and MeiQx. GxE tests were conducted using logistic regression models that
included product terms and likelihood-ratio tests. For these SNP by exposure interactions,
56
dietary variables were re-categorized into tertiles based on the distribution among controls,
due to the reduced sample size with DNA available. To account for multiple testing, we used
Bonferroni correction. Since each exposure was considered an independent a priori
hypothesis we corrected across SNPs for each exposure, but not across exposures. All
statistical tests were two-sided, and all analyses were conducted using the statistical software
package STATA version 11 (STATA Corporation, College Station, TX).
57
Chapter 4: Results
4.1 Fish Intake, Metabolic Enzymes and Risk of Prostate
Cancer: California Collaborative Case-Control Study of
Advanced Prostate Cancer Study
4.1.1 Main Effects of Metabolic Enzymes on Prostate Cancer Risk
In Table 4.1 we summarize the demographic and other relevant characteristics of cases and
controls included in the present analysis. Age was normally distributed among controls.
Mean age was comparable between study sites and between advanced cases and controls;
however, localized cases had a higher mean age than both controls and advanced cases.
Mean caloric intake was higher among LAC relative to SFBA as well as cases in relation to
controls, but was similar among advanced and localized cases.
Allelic frequencies did not differ between racial/ethnic groups or study sites for any
of the SNPs or GST null polymorphisms genotyped.
58
Table 4.1 Socio-demographic and lifestyle characteristics of cases and controls with available DNA from blood, by stage of disease and study site.
Socio-demographic characteristics
Controls Localized
PCA cases
Advanced
PCA cases
LAC SFBA LAC LAC SFBA
N=545 (%) N=255 (%) N=535 (%) N=597 (%) N=391 (%)
Race/ethnicity
African-American
Hispanic
Non-Hispanic White
149 (27)
101 (19)
295 (54)
37 (15)
0 (0)
218 (85)
206 (39)
127 (24)
202 (38)
137 (23)
179 (30)
281 (47)
86 (22)
0 (0)
305 (78)
p
het
<0.001
Family History of PCa
No
Yes
Missing
475 (87)
67 (12)
3 (1)
227 (89)
28 (11)
0 (0)
423 (79)
111 (21)
1 (0)
480 (80)
114 (19)
3 (1)
312 (80)
79 (20)
0 (0)
p
het=
0.579
Socio-economic Status
1 (Low)
2
3
4
5 (High)
97 (18)
104 (19)
108 (20)
109 (20)
127 (23)
8 (3)
14 (5)
50 (20)
75 (29)
108 (42)
151 (28)
116 (22)
91 (17)
96 (18)
81 (15)
137 (23)
109 (18)
131 (22)
103 (17)
117 (20)
11 (3)
26 (7)
56 (14)
89 (23)
209 (53)
p
het
<0.001
Body Mass Index (kg/m
2
)
<25
25-29.9
≥30
Missing
139 (26)
243 (45)
153 (28)
10 (2)
68 (27)
117 (46)
69 (27)
1 (0)
137 (26)
261 (49)
133 (25)
4 (1)
135 (23)
281 (47)
176 (29)
5 (1)
113 (29)
187 (48)
91 (23)
0 (0)
p
het
=0.671
Age (years)
Mean (SD)
63 (10)
Mean (SD)
63 (8)
Mean (SD)
68 (9)
Mean (SD)
64 (9)
Mean (SD)
64 (8)
p
het
<0.001
Calorie Intake (kcal/day)¹
2704 (1153) 2523 (918) 2947 (1174) 3015 (1165) 2728 (1096)
p
het
=0.452
¹124 individuals had unreliable dietary data; p
het
is the p-value for the difference between localized and advanced disease
59
We observed a 1.2-1.7 fold increased risk of localized prostate cancer associated with three
polymorphisms in three different genes: EPHX1 Tyr113His (per His allele OR = 1.27; 95%
CI = 1.04-1.56), CYP1B1 Leu432Val (per Val allele OR = 1.32; 95% CI = 1.09-1.61) and
GSTT1 null/present (null OR = 1.68; 95% CI = 1.19-2.38) (Table 4.2). We observed no
comparable associations for associations for advanced PCA. A test of heterogeneity for the
difference of these associations between advanced and localized disease did not reach
statistical significance for any gene. Moreover, inheritance of the variant alleles did not yield
differences in the risk estimates for localized/advanced PCA across the three racial/ethnic
groups and any of the analyzed genes.
4.1.2 Fish intake, cooking methods, and genetic variation in carcinogen
metabolism enzymes and advanced Prostate Cancer risk
Among the cases and controls included in this study, we observed a 30% reduction of both
localized and advanced PCA risk in men with a low or high intake of dark fish compared
to those who never/rarely consumed dark fish (OR localized disease = 0.72; 95% CI =
0.53-0.96; OR advanced disease = 0.73; 95% CI = 0.55-0.85) (Table 4.3).
60
Table 4.2 Polymorphisms in metabolic genes and prostate cancer risk, by stage of disease
Gene Controls Localized cases Advanced cases p
heterogeneity
n n OR
1
(95% CI) n OR
1
(95% CI)
GSTP1 (Ile105Val)
ile/ile
ile/val
val/val
per val allele
300
339
110
183
235
75
1.0
Ref
1.09 (0.82-1.45)
0.94 (0.64-1.38)
0.99 (0.83-1.20)
386
402
131
1.0
Ref
0.90 (0.73-1.12)
0.88 (0.65-1.19)
0.93 (0.81-1.07)
0.409
0.355
PTGS2 (-765 G/C)
G/G
G/C
C/C
per C allele
481
237
38
297
165
34
1.0
Ref
0.98 (0.74-1.30)
1.05 (0.61-1.85)
1.01 (0.81-1.24)
595
304
36
1.0
Ref
1.07 (0.87-1.35)
0.85 (0.50-1.35)
0.99 (0.85-1.18)
0.839
0.919
CYP1A2 (-154 A/C)
A/A
A/C
C/C
per C allele
368
306
76
223
212
51
1.0
Ref
0.95 (0.71-1.25)
0.96 (0.61-1.51)
0.96 (0.79-1.18)
479
358
82
1.0
Ref
0.88 (0.72-1.08)
0.80 (0.56-1.13)
0.89 (0.77-1.03)
0.607
0.320
EPHX1 (Tyr113His)
tyr/tyr
tyr/his
his/his
per his allele
400
295
52
254
180
48
1.0
Ref
1.13 (0.84-1.49)
1.91 (1.14-3.15)
1.27 (1.04-1.56)
462
361
77
1.0
Ref
1.04 (0.84-1.28)
1.14 (0.82-1.74)
1.04 (0.91-1.24)
0.155
0.087
CYP1B1 (Leu432Val)
leu/leu
leu/val
val/val
per val allele
231
335
190
107
223
166
1.0
Ref
1.51 (1.09-2.15)
1.75 (1.19-2.65)
1.32 (1.09-1.61)
291
396
236
1.0
Ref
1.00 (0.79-1.24)
1.07 (0.80-1.43)
1.03 (0.89-1.19)
0.142
0.120
NAT2
2
fast phenotype
slow phenotype
356
402
242
255
1.0
Ref
1.04 (0.84-1.40)
469
464
1.0
Ref
0.92 (0.76-1.12)
0.352
UGT1A6
3
*1
*2
*3/*4
490
223
25
326
132
21
1.0
Ref
1.04 (0.84-1.29)
0.94 (0.58-1.54)
616
261
29
1.0
Ref
0.91 (0.78-1.07)
0.91 (0.61-1.36)
0.744
GSTT1
present
null
583
153
411
80
1.0
Ref
1.68 (1.19-2.38)
747
162
1.0
Ref
1.18 (0.92-1.52)
0.061
GSTM1
present
null
417
321
299
188
1.0
Ref
0.94 (0.74-1.16)
475
418
1.0
Ref
1.15 (0.95-1.45)
0.192
OR¹ adjusted for age (years) and family history of PCA (yes/no)
NAT2
2
fast phenotype is defined as having at least one copy of the wildtype haplotype (i.e. Arg, Gly, Arg, Ile) for the NAT2 SNPs: Arg197Gln, Gly286Glu, Arg64Gln and Ile114Thr;
slow phenotype is as all other combination of haplotypes
UGT1A6
3
*1 defines the wiltype haplotype (i.e. Thr, Arg) for the UGT1A6 SNPs: Thr181Ala and Arg184Ser; *2 defines the haplotype Ala, Ser; *3 defines the haplotype Thr, Ser and
*4 defines the haplotype Ala, Arg
61
Table 4.3 Fish intake and risk of prostate cancer among men with DNA available, by stage of disease
Controls Localized Cases Advanced Cases
Fish Intake n n OR
1
95% CI OR
2
95% CI n OR
1
95% CI OR
2
95% CI
White fish intake
Never/Rarely
Low Intake
High Intake
P
trend
263
370
126
176
236
83
1.0
Ref
0.98
0.88
0.72-1.32
0.59-1.32
0.576
1.0
Ref
1.00
0.85
0.73-1.36
0.59-1.29
0.516
359
393
183
1.0
Ref
0.79
1.14
0.63-0.99
0.85-1.52
0.823
1.0
Ref
0.82
1.27
0.66-1.03
0.94-1.71
0.385
Dark fish intake
Never/Rarely
Low Intake
High Intake
P
trend
272
362
126
194
213
90
1.0
Ref
0.72
0.78
0.53-0.97
0.53-1.16
0.111
1.0
Ref
0.72
0.68
0.53-0.99
0.44-1.06
0.043
417
377
142
1.0
Ref
0.68
0.79
0.54-0.84
0.59-1.06
0.017
1.0
Ref
0.70
0.87
0.56-0.88
0.63-1.20
0.081
Dark fish intake
Never/Rarely
Low/High Intake
P
trend
272
488
194
303
1.0
Ref
0.74
0.56-0.98
0.034
1.0
Ref
0.72
0.53-0.96
0.030
477
519
1.0
Ref
0.70
0.55-0.85
<0.001
1.0
Ref
0.73
0.55-0.85
<0.001
OR
1
Adjusted for age (years), BMI (<25, 25-29, ≥30), total calorie intake (kcal/ day) and family history of PCA (yes/no)
OR
2
Adjusted for age (years, continuous), BMI (\25, 25–29, C30), total calorie intake(kcal/day), family history of PCA (yes or no), total fat intake
(g/day), dietary vitamin D intake (g/day), alcohol consumption (g/day), total dairy intake (g/day), cigarette smoking (pack-years),total fruit
consumption (g/day), total vegetable consumption (g/day), red meat consumption(g/day), white meat consumption (g/day), processed meat
consumption (g/day)
62
Furthermore, PCA risk was reduced by 50% among men who usually used low
temperature cooking methods, although the interaction between cooking methods did not
reach statistical significance (OR localized disease/high intake = 0.52; 95% CI = 0.26-1.02;
OR advanced disease/high intake = 0.53; 95% CI = 0.31-0.89) (Table 4.4). High white fish
consumption was associated with a 50-60% increase in risk of both localized and advanced
PCA among men who consumed well-done fish; however, the risk estimate only reached
significance among advanced cases (OR localized disease = 1.53; 95% CI = 0.81-2.86; OR
advanced disease =1.65; 95% CI = 1.02-2.69) (Table 4.5). These findings are similar to those
previously reported for the full study population regardless of DNA availability (Joshi, John
et al. 2012 ).
We considered interactions between each of the eight SNP genotypes and NAT2
predicted phenotype and intake of white or dark fish. We observed evidence that the PTGS2
765 G/C polymorphism modified the association between high white fish intake and
advanced PCA risk (crude p for interaction=0.002, Bonferroni corrected p for interaction =
0.018) (Table 4.6). Specifically, the association between high white fish intake and advanced
PCA risk was restricted to carriers of the C allele, reaching a more than three-fold increased
risk (for high vs. no/rare white fish intake) among carriers of two C alleles (OR = 3.56; 95%
CI = 1.61-7.88). There was no evidence of interaction with white fish intake for any of the
other polymorphisms investigated for either localized or advanced disease. When
considering dark fish intake there was an inverse association between dark fish intake and
advanced PCA risk among men with the GG genotype, but not among carriers of the C
allele (Table 4.7).
63
Table 4.4 Fish intake, cooking temperature and risk of prostate cancer among men with available DNA, by stage of disease
Controls Localized Cases Advanced Cases
Fish Intake Low
Temp
High
Temp
Low Temperature High Temperature Low Temperature High Temperature
White fish intake n
a
n
a
n
b
OR
1
95% CI n
b
OR
1
95% CI n
c
OR
1
95% CI n
c
OR
1
95% CI
Never/Rarely
Low Intake
High Intake
p
interaction
263
113
45
263
245
80
176
56
18
1.0
Ref
0.70
0.50
0.44-1.05
0.24-1.02
176
171
65
1.0
Ref
1.18
1.23
0.83-1.67
0.74-2.03
0.542
359
100
42
1.0
Ref
0.74
0.86
0.53-1.04
0.52-1.41
359
272
128
1.0
Ref
0.84
1.26
0.65-1.10
0.87-1.65
0.360
Dark fish intake
Never/Rarely
Low Intake
High Intake
p
interaction
272
119
42
272
236
82
194
59
24
1.0
Ref
0.50
0.52
0.32-0.80
0.26-1.02
194
148
62
1.0
Ref
0.72
0.67
0.50-1.03
0.41-1.11
0.847
417
101
34
1.0
Ref
0.49
0.53
0.35-0.69
0.31-0.89
417
267
104
1.0
Ref
0.66
0.71
0.51-0.85
0.49-1.04
0.896
a,b,c
reference groups the same; OR
1
Adjusted for age (years), BMI (<25, 25-29, ≥30), total calorie intake (kcal/ day) and family history
of PCA (yes/ no) ), total well-done meat consumption (g/day), total meat consumption with high-temperature methods (g/day); dark
fish and white fish intakes were mutually adjusted for each other
64
Table 4.5 Fish intake, level of doneness and risk of PCA among men with available DNA, by stage of disease
Controls Localized Cases Advanced Cases
Fish Intake JUD WD Just Until Done Well Done Just Until Done Well Done
n
a
n
a
n
b
OR
1
95% CI n
b
OR
1
95% CI n
c
OR
1
95% CI n
c
OR
1
95% CI
White fish intake
Never/Rarely
Low Intake
High Intake
P
interaction
263
188
60
263
153
34
176
93
29
1.0
Ref
0.85
0.86
0.57-1.27
0.46-1.61
176
128
46
1.0
Ref
1.12
1.53
0.76-1.64
0.81-2.86
0.531
359
199
81
1.0
Ref
0.83
1.18
0.63-1.11
0.78-1.78
359
165
78
1.0
Ref
0.78
1.65
0.58-1.06
1.02-2.69
0.139
Dark fish intake
Never/Rarely
Low Intake
High Intake
P
interaction
272
127
40
272
213
52
194
51
26
1.0
Ref
0.79
1.19
0.48-1.29
0.60-2.37
194
154
56
1.0
Ref
0.59
0.67
0.41-0.85
0.38-1.16
0.530
417
160
50
1.0
Ref
0.77
0.79
0.56-1.06
0.48-1.29
417
205
80
1.0
Ref
0.56
0.85
0.42-0.73
0.55-1.31
0.183
a,b,c
reference groups the same; OR
1
Adjusted for age (years), BMI (<25, 25-29, ≥30), total calorie intake (kcal/ day) and family
history of PCA (yes/ no), total well-done meat consumption (g/day), total meat consumption with high-temperature methods (g/day); dark fish and
white fish intakes were mutually adjusted for each other; JUD = Just until done; WD = well-done
65
Table 4.6 Fish intake and prostate cancer risk stratified by PTGS2 genotype, by stage of disease
Controls Localized Cases Advanced Cases
n n OR
1,2
(95% CI) n OR
1,2
(95% CI)
PTGS2
White Fish
G/G G/C C/C G/G G/C C/C G/G G/C C/C G/G G/C C/C G/G G/C C/C
Never/Rarely
Low
High
p
trend
p
3
heterogeneity of trend
p
heterogeneity by stage
149
241
91
91
114
32
21
13
3
108
142
46
54
77
34
14
16
3
1.0
Ref
0.87 (0.59-1.28)
0.82 (0.48-1.40)
0.422
0.78 (0.55-1.11)
0.98 (0.65-1.48)
1.10 (0.58-2.09)
0.233
0.61 (0.30-1.24)
1.05 (0.58-1.91)
1.36 (0.47-3.89)
0.105
0.089
227
262
105
117
120
67
15
10
11
1.0
Ref
0.80 (0.60-1.06)
0.96 (0.66-1.41)
0.619
0.81 (0.62-1.07)
0.79 (0.58-1.09)
1.85 (1.19-2.89)
0.002
0.66 (0.38-1.15)
0.79 (0.47-1.33)
3.56 (1.61-7.88)
0.001
0.002
0.991
Dark Fish
Never/Rarely
Low
High
p
trend
p
3
heterogeneity of trend
p
heterogeneity by stage
160
237
84
91
109
37
16
16
5
123
119
54
58
80
27
12
14
7
1.0
Ref
0.60 (0.40-0.88)
0.64 (0.38-1.07)
0.042
0.79 (0.55-1.14)
0.68 (0.45-1.02)
0.72 (0.42-1.24)
0.681
0.63 (0.30-1.31)
0.77 (0.41-1.46)
0.82(0.32-2.07)
0.578
0.191
269
251
74
128
115
61
20
10
6
1.0
Ref
0.62 (047-0.82)
0.53 (0.35-0.80)
<0.001
0.92 (0.71-1.21)
0.59 (0.43-0.80)
0.89 (0.58-1.36)
0.346
0.85 (0.50-1.46)
0.56 (0.33-0.94)
1.47 (0.68-3.18)
0.485
0.046
0.733
OR¹ odds ratios are from models with genotype coded as log-additive (i.e. number of C alleles, treated as continuous)
OR
2
adjusted for age (years), BMI (<25, 25-29, ≥30), total calorie intake (kcal/ day) and family history of PCA (yes/no); dark fish and white fish
intakes were mutually adjusted for each other
p
3
values unadjusted for Bonferroni; Bonferroni adjusted p-values can be found in text
66
Table 4.7 Well-done white fish intake and risk of advanced prostate cancer, stratified on PTGS2 genotype
Controls Advanced Cases
n n OR
1,2
(95% CI)
PTGS2 G/G G/C C/C G/G G/C C/C G/G G/C C/C
Fish Intake
Never/Rarely
Low Intake
High Intake
p
trend
p
3
heterogeneity of trend
149
121
45
91
56
20
21
4
1
227
124
58
117
64
39
15
6
5
1.0
Ref.
0.65 (0.45-0.93)
1.24 (0.70-2.22)
0.582
0.84 (0.64-1.12)
0.90 (0.60-1.37)
2.17 (1.05-4.48)
0.037
0.71 (0.41-1.24)
1.27 (0.59-2.72)
3.79 (0.92-15.70)
0.020
0.021
OR¹ odds ratios are from models with genotype coded as log-additive (i.e. number of C alleles, treated as continuous)
OR
2
adjusted for age (years), BMI (<25, 25-29, ≥30), total calorie intake (kcal/ day) and family history of PCA (yes/no),
total well-done meat consumption (g/day), total meat consumption with high-temperature methods (g/day) and dark fish
p
3
values unadjusted for Bonferroni
67
However, this interaction was not statistically significant after correcting for Bonferroni. We
found no evidence of interactions between dark fish intake and the other polymorphisms.
For both white and dark fish intake similar trends were observed for localized and advanced
disease; however, the tests for interaction were not statistically significant for localized
disease. Interactions between the PTGS2 variant and low/high intake of white/dark fish in
relation to localized/advanced PCA did not vary across the three racial/ethnic groups.
We further assessed whether polymorphisms in metabolic enzymes might modify the
association between localized and/or advanced prostate cancer risk and fish intake when
taking into account cooking methods and level of doneness. For this purpose we
investigated fish intake when cooked using high temperature cooking methods only (pan-
frying, grilling, oven-broiling and barbequing), and intake of well-done fish only. We
observed that PTGS2 genotype modified the association between high intake of well-done
white fish and advanced PCA similarly to all white fish, although the interaction term was no
longer significant after adjusting for Bonferroni (crude p for interaction = 0.021) (Table 4.7).
As before, this association was restricted to carriers of the C allele, with estimated two- to
four-fold increased risks (for high vs. no/rare white fish intake) for carriers of one or two C
alleles (OR CG genotype = 2.17; 95% CI = 1.05-4.48; OR CC genotype = 3.79; 95% CI =
0.92-15.70). These estimates were slightly stronger than what we observed for white fish
without considering level of doneness. We observed no evidence of interactions when
considering intake of white or dark fish cooked using high temperature methods.
68
4.2 Fish, Red Meat, and Poultry Intake and Risk of Prostate
Cancer: Singapore Chinese Health Study
Table 4.8 summarizes demographic and dietary characteristics for men in this cohort
stratified by level of intake of red meat or fish, the two main meat types in this population.
We observed that increased intake of both red meat and fish was associated with increased
BMI and decreased physical activity. Men with diets high in red meat intake were also more
likely to be younger, current smokers, regular drinkers of alcohol and green tea, and less
likely to take a vitamin supplement. Men with diets high in fish intake were more likely to be
Hokkein vs. Cantonese, less likely to drink alcohol and were more likely to drink green tea
(Table 4.8). Intake of red meat was positively associated with caloric intake and inversely
associated with vegetable, fruit and Vitamin C intake. In contrast, increased fish intake was
associated with a decrease in caloric intake and increased vegetable, fruit and Vitamin C
intake. Red meat and fish intake were both positively associated with each other as well as
poultry intake and total fat intake (Table 4.8).
Diets high in total red meat, fresh red meat, preserved red meat and organ meat
showed no association with PCA risk (Table 4.9). Total poultry was not associated with PCA
risk, however when we considered whether poultry was eaten with or without the skin, we
observed a positive association between diets high in poultry eaten with all its skin and PCA
risk when considering risk of advanced disease (OR(Q4 vs. Q1) = 3.5; 95% CI = 1.1-11.2;
p
trend
= 0.013), although the number of cases in this restricted subgroup was relatively small
69
and a test for heterogeneity of disease was not statistically significant (p
heterogeneity
= 0.378)
(Table 4.9).
Total fish and shellfish intake and fresh fish and shellfish intake were not associated
with PCA risk. However, when we considered preserved fish and shellfish intake we
observed heterogeneity of the association with PCA risk by stage of disease (p
heterogeneity
=
0.041) (Table 4.9). There was a borderline statistically significant increasing trend (p
trend
=
0.070), among advanced cases only, with the third and fourth quartiles showing a 30%
increase in risk compared to the first quartile of intake. No evidence of this trend was seen
among early PCA cases (Table 4.9). Diets high in all fresh meats combined (red meat,
poultry, and fish) were not associated with PCA risk. Instead, diets high in all preserved
meats combined were positively associated with PCA risk (p
trend
= 0.060). This association
differed by stage with the observed positive association being restricted to advanced cases
and not evident among early cases (p
heterogeneity
= 0.027). Among advanced cases, the highest
quartile of intake of all preserved meats was associated with a 40% increase in risk of PCA
compared to the lowest quartile (95% CI = 0.9-2.3) with a significant increasing trend (p
trend
= 0.018) (Table 4.9).
The observed positive association between diets high in all preserved meats and PCA
risk was modified by both BMI (p
interaction
= 0.010) and by fruit and vegetable intake (p
interaction
= 0.020) (Table 4.10). Among those with a BMI below the median (<22.9kg/m
2
), there was
no evidence of an association between preserved meat intake and PCA risk.
70
Table 4.8 Demographic and dietary characteristics of study population by quartiles of meat and fish intake
Red meat (g/1000kcal) Fish (g/1000kcal)
<12.0 12.0-18.5 18.5-26.5 >26.5 p
trend
<22.6 22.6-32.4 32.4-43.8 >43.8 p
trend
N
Demographic
6824 6823 6823 6823 6824 6823 6823 6823
Mean age 57.7 56.6 56.3 55.8 <0.001 56.9 56.4 56.4 56.7 0.127
Dialect (% Hokkien) 55.7 55.3 56.4 56.2 0.372 53.3 54.0 55.5 60.8 <0.001
Education (%)
No formal education 12.1 10.8 10.3 10.2 11.8 9.9 10.0 11.7
Primary school 50.3 50.5 52.1 51.5 50.4 50.4 51.7 51.8
Secondary school or higher 37.7 38.7 37.7 38.2 0.053 37.7 39.7 38.4 36.5 0.194
Mean BMI 22.9 23.0 23.1 23.1 <0.001 22.7 23.0 23.1 23.2 <0.001
Any physical activity (%) 45.2 44.0 42.9 41.0 <0.001 44.1 45.2 43.1 40.6 <0.001
Smoking status (%)
Never 45.2 43.2 41.8 38.6 42.1 43.3 42.2 41.2
Former 23.4 21.2 20.8 20.1 20.9 21.9 21.1 21.7
Current 31.4 35.6 37.4 41.2 <0.001 37.0 34.8 36.6 37.1 0.212
Regular Alcohol Use (%) 24.9 31.9 33.9 35.0 <0.001 31.0 33.2 32.5 29.0 <0.001
Regular Green Tea Use (%) 41.6 44.0 47.3 50.6 <0.001 44.4 46.4 46.6 46.2 0.044
Weekly vitamin intake (%) 5.4 5.2 4.5 4.1 <0.001 5.1 4.7 4.9 4.5 0.176
Dietary (Mean Values)
Calories (kcal) 1619.6 1725.9 1786.6 1880.7 <0.001 1758.4 1840.9 1795.9 1617.6 <0.001
Red meat (gpd) - - - - - 16.6 10.1 21.7 22.2 <0.001
Fish (gpd) 30.3 33.5 35.9 37.7 <0.001 - - - - -
Poultry (gpd) 8.5 12.4 14.5 16.5 <0.001 11.4 13.3 13.9 13.4 <0.001
Vegetables (gpd) 65.9 63.5 64.5 63.0 <0.001 56.2 61.7 65.7 73.4 <0.001
Fruit (gpd) 139.6 125.7 117.3 103.6 <0.001 118.0 121.9 121.4 125.1 <0.001
Total Fat (ND) 21.0 23.2 25.3 28.3 <0.001 21.9 24.1 25.3 26.5 <0.001
71
Table 4.9 Meat, poultry & fish intake and risk of prostate cancer among Singapore Chinese by stage of disease
All Early Advanced
Dietary Variable (g/1000kcal) Cases HR
1
95% CI Cases HR
1
95% CI Cases HR
1
95% CI p
heterogeneity
Total red meat
Q1: <12.0 89 1.0
Ref.
48 1.0
Ref.
40 1.0
Ref.
Q2: 12.0-18.5 74 0.8 0.6-1.2 24 0.5 0.3-0.8 46 1.2 0.8-1.8
Q3: 18.5-26.5 65 0.7 0.5-1.0 25 0.5 0.3-0.9 36 0.9 0.6-1.5
Q4: >26.5 70 0.8 0.5-1.2 27 0.6 0.3-1.0 40 1.0 0.6-1.7
Trend 0.223 0.080 0.743 0.847
Fresh red meat
Q1: <10.3 88 1.0
Ref.
46 1.0
Ref.
41 1.0
Ref.
Q2: 10.3-16.1 77 0.9 0.6-1.2 25 0.5 0.3-0.9 50 1.2 0.8-1.8
Q3: 16.1-23.3 67 0.8 0.5-1.1 28 0.6 0.4-1.0 34 0.8 0.5-1.3
Q4: >23.3 66 0.7 0.5-1.1 25 0.6 0.3-1.0 37 0.8 0.5-1.4
Trend 0.098 0.076 0.261 0.848
Preserved red meat
Q1: <0.17 78 1.0
Ref.
32 1.0
Ref.
44 1.0
Ref.
Q2: 0.17-0.90 73 1.0 0.7-1.4 32 1.1 0.7-1.8 39 0.9 0.6-1.5
Q3: 0.90-2.12 61 0.9 0.6-1.2 26 0.9 0.5-1.6 31 0.8 0.5-1.2
Q4: >2.12 86 1.3 0.9-1.8 34 1.3 0.7-2.1 48 1.2 0.8-1.9
Trend 0.113 0.379 0.287 0.610
Organ meat
Q1: <0.12 79 1.0
Ref.
40 1.0
Ref.
36 1.0
Ref.
Q2: 0.12-0.30 84 1.1 0.8-1.5 35 0.9 0.6-1.5 46 1.3 0.9-2.1
Q3: 0.30-0.86 69 1.0 0.7-1.3 24 0.7 0.4-1.1 40 1.2 0.8-1.9
Q4: >0.86 66 1.0 0.7-1.4 25 0.8 0.4-1.3 40 1.3 0.8-2.1
Trend 0.807 0.408 0.523 0.871
Total poultry
Q1: <6.14 75 1.0
Ref.
30 1.0
Ref.
41 1.0
Ref.
Q2: 6.14-11.1 78 1.1 0.8-1.5 36 1.3 0.8-2.1 40 1.0 0.6-1.6
Q3: 11.1-17.8 61 0.9 0.6-1.3 24 0.9 0.5-1.6 36 0.9 0.6-1.5
Q4: >17.8 84 1.3 0.9-1.8 34 1.4 0.8-2.4 45 1.2 0.7-1.9
Trend 0.239 0.376 0.462 0.378
Poultry with all skin
Q1: <7.67 10 1.0
Ref.
4 1.0
Ref.
4 1.0
Ref.
Q2: 7.67-12.2 21 1.5 0.7-3.2 10 1.8 0.6-5.8 10 1.8 0.5-5.7
Q3: 12.2-18.5 15 1.0 0.4-2.2 9 1.5 0.4-5.0 5 0.8 0.2-3.1
Q4: >18.5 28 2.0 0.9-4.5 8 1.4 0.4-5.2 19 3.5 1.1-11.2
Trend 0.084 0.888 0.013 0.378
Poultry with no/some skin
Q1: <5.54 65 1.0
Ref.
26 1.0
Ref.
37 1.0
Ref.
Q2 : 5.54-10.6 57 1.0 0.7-1.4 26 1.2 0.7-2.1 30 0.9 0.6-1.5
Q3: 10.6-17.4 46 0.9 0.6-1.3 15 0.8 0.4-1.5 31 1.0 0.6-1.6
Q4: >17.4 56 1.1 0.7-1.6 26 1.5 0.8-2.7 26 0.8 0.5-1.5
Trend 0.701 0.320 0.591 0.364
72
Table 4.9 (cont.) Meat, poultry & fish intake and risk of prostate cancer among Singapore Chinese by stage of disease
All Early Advanced
Dietary Variable (g/1000kcal) Cases HR
1
95% CI Cases HR
1
95% CI Cases HR
1
95% CI p
heterogeneity
Total fish & shellfish
Q1: <22.6 68 1.0
Ref.
30 1.0
Ref.
36 1.0
Ref.
Q2 : 22.6-32.4 83 1.2 0.9-1.7 35 1.4 0.8-2.2 43 1.1 0.7-1.8
Q3: 32.4-43.8 63 0.9 0.7-1.3 25 0.8 0.5-1.5 37 1.0 0.6-1.6
Q4: >43.8 84 1.3 0.9-1.8 34 1.4 0.8-2.4 46 1.3 0.8-2.1
Trend 0.265 0.373 0.339 0.857
Fresh fish & shellfish
Q1: <20.8 69 1.0
Ref.
29 1.0
Ref.
36 1.0
Ref.
Q2 : 20.8-30.3 76 1.1 0.8-1.5 37 1.2 0.7-2.0 39 1.0 0.7-1.6
Q3: 30.3-41.7 72 1.0 0.7-1.5 22 0.9 0.5-1.6 42 1.1 0.7-1.8
Q4: >41.7 81 1.2 0.9-1.7 36 1.3 0.7-2.1 45 1.3 0.8-2.0
Trend 0.300 0.565 0.285 0.840
Preserved fish & shellfish
Q1: <0.48 75 1.0
Ref.
34 1.0
Ref.
38 1.0
Ref.
Q2 : 0.48-1.33 68 0.9 0.7-1.3 34 1.1 0.7-1.8 31 0.8 0.5-1.3
Q3: 1.33-2.59 79 1.2 0.8-1.6 28 0.9 0.6-1.6 46 1.3 0.8-2.0
Q4: >2.59 76 1.1 0.8-1.6 28 1.0 0.6-1.6 47 1.3 0.9-2.1
Trend 0.321 0.800 0.070 0.041
Fresh all meats
2
Q1: <35.5 79 1.0
Ref.
37 1.0
Ref.
41 1.0
Ref.
Q2 : 35.5-48.5 78 1.2 0.9-1.6 34 1.1 0.7-1.8 41 1.2 0.8-1.8
Q3: 48.5-62.7 66 1.0 0.7-1.4 26 0.9 0.5-1.6 34 1.0 0.6-1.6
Q4: >62.7 75 1.1 0.7-1.6 27 0.9 0.5-1.7 46 1.2 0.7-2.0
Trend 0.896 0.629 0.637 0.988
Preserved all meats
2
Q1: <1.23 76 1.0
Ref.
36 1.0
Ref.
38 1.0
Ref.
Q2 : 1.23-2.61 59 0.8 0.6-1.2 29 0.9 0.5-1.4 26 0.7 0.4-1.2
Q3: 2.61-4.73 82 1.2 0.9-1.7 30 1.0 0.6-1.6 49 1.4 0.9-2.2
Q4: >4.73 81 1.2 0.9-1.7 29 1.0 0.6-1.7 49 1.4 0.9-2.3
Trend 0.060 0.911 0.018 0.027
1
All
analyses
were
adjusted
for
age
at
baseline
interview
(years),
dialect
group
(Cantonese,
Hokkien),
year
of
interview
(1993-‐94,
1995-‐96,
1997-‐98),
education
(no
formal
education/primary
school/secondary
school
or
higher,
family
history
of
cancer,
nutrient
density
adjusted
fat
intake
and
caloric
intake;
meat
variables
were
additionally
adjusted
for
smoking
status
(never/ex/current),
number
of
alcoholic
drinks
per
week,
energy
adjusted
poultry
intake,
energy
adjusted
dairy
intake
and
energy
adjusted
vegetable
intake;
poultry
variables
were
additionally
adjusted
for
energy
adjusted
red
meat
intake;
fish
variables
were
additionally
adjusted
for
energy
adjusted
red
meat
intake,
energy
adjusted
poultry
intake
and
energy
adjusted
vegetable
intake.
2
Combines
red
meat,
poultry
and
fish.
73
Table 4.10 Preserved meat intake and risk of prostate cancer by BMI and fruit and vegetable intake
BMI
Preserved meat
2
<22.9 >22.9 p
interaction
All Cases Controls Cases HR
1
95% CI Controls Cases HR
1
95% CI
Q1: <1.23 3423 44 1.0
Ref.
3325 32 1.0
Ref.
Q2 : 1.23-2.61 3332 34 0.9 0.6-1.4 3432 25 0.8 0.5-1.3
Q3: 2.61-4.73 3394 39 1.0 0.7-1.6 3347 43 1.5 1.0-2.4
Q4: >4.73 3393 33 0.9 0.5-1.4 3349 48 1.8 1.1-2.8
Trend 0.707 0.002 0.010
Advanced Cases Controls Cases HR
1
95% CI Controls Cases HR
1
95% CI
Q1: <1.23 3423 25 1.0
Ref.
3325 13 1.0
Ref.
Q2 : 1.23-2.61 3332 16 0.7 0.4-1.3 3432 10 0.8 0.3-1.7
Q3: 2.61-4.73 3394 18 0.8 0.4-1.5 3347 31 2.7 1.4-5.1
Q4: >4.73 3393 18 0.8 0.4-1.5 3349 31 2.7 1.4-5.3
Trend 0.715 <0.001 0.002
Fruit and Vegetable Intake
Preserved meat
2
<168.7 >168.7
All Cases Controls Cases HR
1
95% CI Controls Cases HR
1
95% CI
Q1: <1.23 3295 30 1.0
Ref.
3453 46 1.0
Ref.
Q2 : 1.23-2.61 3497 28 1.0 0.6-1.6 3267 31 0.8 0.5-1.2
Q3: 2.61-4.73 3459 36 1.3 0.8-2.2 3282 46 1.2 0.8-1.8
Q4: >4.73 3256 46 1.9 1.2-3.0 3486 35 0.9 0.6-1.4
Trend 0.002 0.946 0.020
Advanced Cases Controls Cases HR
1
95% CI Controls Cases HR
1
95% CI
Q1: <1.23 3295 18 1.0
Ref.
3453 20 1.0
Ref.
Q2 : 1.23-2.61 3497 13 0.7 0.4-1.5 3267 13 0.7 0.4-1.4
Q3: 2.61-4.73 3459 21 1.3 0.7-2.4 3282 28 1.7 0.9-3.0
Q4: >4.73 3256 31 2.0 1.1-3.7 3486 18 1.0 0.5-1.9
Trend 0.003 0.632 0.072
1
Analyses
adjusted
for
age
at
baseline
interview
(years),
dialect
group
(Cantonese,
Hokkien),
caloric
intake,
year
of
interview
(1993-‐94,
1995-‐96,
1997-‐98),
education
(no
formal
education/primary
school/secondary
school
or
higher),
smoking
status
(never/ex/current),
family
history
of
cancer,
number
of
alcoholic
drinks
per
week,
energy
adjusted
dairy
intake
and
nutrient
density
adjusted
fat
intake;
BMI
interaction
additionally
adjusted
for
energy
adjusted
vegetable
intake.
2
Combines
red
meat,
poultry
and
fish.
74
However, among men with a BMI above the median value, there was a statistically
significant trend (p
trend
= 0.002), with those in the highest quartile of intake of all preserved
meats showing an 80% increase in risk of PCA compared to the lowest quartile (95% CI =
1.1-2.8). This association seemed stronger when restricting analyses to advanced cases only,
with those in the two highest quartiles of intake of all preserved meats showing 2.7 times the
risk of PCA compared to the lowest quartile (p
trend
< 0.001). When repeating analyses
restricting preserved meats to preserved red meats we observed a similar effect (p
interaction
=
0.048), but an interaction was not seen when restricting to preserved fish only (p
interaction
=
0.322), suggesting that this effect modification may be driven by preserved red meat. When
considering fruit and vegetable intake as potential modifiers, we observed that among men
with diets below the median level of daily fruit and vegetable intake (<168.7g/1000kcal),
there was a positive association with preserved meat intake and PCA risk (Table 4.10). Those
in the highest quartile of intake were 90% more likely to develop PCA than those in the
lowest quartile of intake (95% CI = 1.2-3.0), with a significant trend (p
trend
= 0.002). This
positive association was not seen among those with high fruit and vegetable intake. Results
were very similar when restricting analyses to advanced cases. Again, similar results were
seen when restricting analyses to preserved red meats only (p
interaction
= 0.020), but not
preserved fish (p
interaction
= 0.419), suggesting this modification may also be driven by
preserved red meat.
75
4.3 Processed meats, NOCs, DNA Repair Enzymes and Risk of
Bladder Cancer
Table 4.11 shows demographic information and other key characteristics of cases and
controls. The mean age at diagnosis of cases was 54.4 years and the mean age of controls at
recruitment was 54.4 years. Our study included 2,992 non-Hispanic whites (1,523 cases and
1,469 controls), 152 Hispanics (80 cases and 72 controls) and 102 African Americans (57
cases and 45 controls). Males made up 78.7% of cases and 78.0% of controls. Level of
education differed between cases and controls (p<0.001), with controls more likely to have
achieved a higher level of education. Cases were more likely to have a history of diabetes
(p=0.016). As expected, more cases were ever smokers, and among ever smokers, mean
duration and intensity of cigarette smoking were significantly higher in cases than in controls
(p<0.001). Controls reported greater numbers of daily servings of fruit and vegetables
(p<0.001), higher Vitamin A and C intake (p<0.001), higher total carotenoid intake
(p=0.004), and more servings of all food combined (p=0.001).
76
Table 4.11. Socio-demographic and life-style characteristics of cases and controls
Socio-demographic characteristics Controls Cases
Race N (%) N (%) p-value for difference
NHW 1469 (93) 1523 (92)
Hispanic 72 (5) 80 (5)
AA 45 (3) 57 (3) 0.571
Gender
Male 1237 (78) 1307 (79)
Female 349 (22) 353 (21) 0.609
BMI
Normal 786 (50) 827 (50)
Overweight 626 (39) 651 (39)
Obese 173 (11) 180 (11) 0.982
Missing 1 (0) 2 (0)
Education
High School or less 491 (31) 689 (42)
1-4 years college 742 (47) 712 (43)
Post/Grad school 353 (22) 259 (16) <0.001
History of Diabetes
No 1523 (96) 1565 (94)
Yes 62 (4) 95 (6) 0.016
Missing 1 (0) 0 (0)
Cigarette Use
Never 574 (36) 299 (18)
Former 373 (24) 756 (46)
Current 639 (40) 605 (36) <0.001
Tumor Stage
T
a
847 (51)
T
1
442 (27)
T
2
190 (11)
T
3
52 (3)
T
4
32 (2)
CIS
85 (5)
Unknown
12 (1)
Tumor Grade
Grade 1
352 (21)
Grade 2
709 (43)
Grade 3
478 (29)
Grade 4
93 (6)
Unknown 28 (2)
77
Table 4.11 cont. Socio-demographic and life-style characteristics of cases and controls
Socio-demographic characteristics Controls Cases
Tumor Stage/Grade
N (%)
T
a
/grade 1-2
1
757 (46)
All other stage/grade
2
885 (53)
Unknown
18 (1)
Mean (SD) Mean (SD)
Age 54 (8) 54 (8) 0.991
Years Smoked among ever smokers 26 (13) 30 (12) <0.001
Cigarettes per day among ever smokers 23 (14) 27 (15) <0.001
Servings fruit per day 1.6 (1.0) 1.4 (1.0) <0.001
Servings vegetables per day 2.3 (1.1) 2.1 (1.1) <0.001
Servings dairy per day 2.4 (1.6) 2.4 (1.6) 0.881
Total servings food per day 8.2 (3.1) 7.8 (3.1) 0.001
Vitamin A per day (IU) 10397 (5792) 9619 (5979) <0.001
Vitamin C per day (mg) 120 (70) 108 (71) <0.001
Vitamin D per day (IU) 144 (121) 140 (125) 0.418
Calcium per day (mg) 573 (368) 551 (381) 0.098
Total Carotenoids per day (mcg) 9869 (6375) 9158 (7421) 0.004
Lifetime use of NSAID (g) 610 (2593) 569 (2501) 0.647
1
Low probability of progression as defined by Kiemeney et al. (23)
2
High probability of progression as defined by Kiemeney et al. (23)
4.3.1 Main Effects of Processed Meats and dietary NOCs on the risk of
Bladder Cancer
We investigated associations between dietary sources of exogenous NOCs, which can also
serve as precursors of endogenous NOC formation. We observed positive associations
between bladder cancer risk and intake of salami/pastrami/corned beef (p
trend
=0.008), with
each of the two highest categories of intake associated with a 30% increased risk of bladder
cancer risk compared with the lowest category (Table 4.12). Evidence of effect modification
78
by smoking status (never/ever) was evident for a number of the variables examined;
therefore, we also present results stratified by smoking status. The association of bladder
cancer with intake of salami/pastrami/corned beef was strongest among never smokers
(p
trend
=0.006), with the highest categories at twice the risk of bladder cancer compared with
the lowest category. There was also a statistically significant association between bladder
cancer and highest category of liver intake when compared with the lowest category
(OR=1.26; 95% CI=1.00-1.60; p
trend
=0.039). This association was also strongest among
never smokers (OR=1.76; 95% CI=1.09-2.85; p
trend
=0.016), although a test of heterogeneity
of trends between never and ever smokers was not statistically significant (p=0.118). When
considering intake of total processed meats and bladder cancer risk we also observed
heterogeneity between never and ever smokers (p=0.017), although none of the risk
estimates were significant. These dietary variables were not modified further by gender or by
risk of progression based on stage and grade information.
79
Table 4.12 Associations between intake of processed meats and liver and risk of bladder cancer
All Subjects Never Smokers Ever Smokers
Co/Ca OR
1
95% CI Co/Ca OR 95% CI Co/Ca OR
95% CI
Fried Bacon/Ham
< once a month 457/412 1.0
Ref
213/10
1
1.0
Ref
244/311 1.0
Ref
1-2 times a
month
170/172 1.04 0.80-
1.35
68/39 1.23 0.77-
1.96
102/133 0.97 0.71-1.34
1.35 3-4 times a
month
427/490 1.12 0.91-
1.36
154/79 1.06 0.73-
1.53
273/411 1.14 0.90-145
weekly 245/276 1.00 0.79-
1.27
66/43 1.25 0.78-
2.01
179/233 0.94 0.71-1.24
2+ times a week 274/297 0.94 0.73-
1.21
70/33 0.84 0.49-
1.45
204/264 0.99 0.75-1.32
p-trend
0.453
0.683
0.707 p
het
=0.153
Salami/Pastrami/Corned Beef
< twice a year 405/369 1.0
Ref
194/68 1.0
Ref
211/301 1.0
Ref
2-11 times a year 501/495 1.07 0.87-
1.30
177/96 1.51 1.03-
2.21
324/399 0.90 0.71-1.14
monthly 204/202 1.12 0.86-
1.45
68/37 1.61 0.98-
2.66
136/165 0.92 0.68-1.25
twice monthly 284/349 1.34 1.07-
1.69
86/62 2.19 1.40-
3.43
198/287 1.10 0.84-1.44
weekly 179/232 1.33 1.02-
1.74
46/32 1.95 1.10-
3.46
133/200 1.14 0.84-1.55
p-trend
0.008
0.006
0.112 p
het
=0.003
Bologna/Other Lunch Meats
never 320/343 1.0
Ref
134/76 1.0
Ref
186/267 1.0
Ref
< once a month 390/363 0.83 0.66-
1.03
163/70 0.74 0.49-
1.12
227/293 0.87 0.67-1.14
monthly 435/458 0.87 0.70-
1.08
157/73 0.77 0.51-
1.16
278/385 0.90 0.70-1.17
weekly 162/198 0.96 0.73-
1.28
48/35 1.14 0.65-
1.97
114/163 0.93 0.67-1.29
twice weekly 266/285 0.81 0.63-
1.04
69/41 0.94 0.56-
1.59
197/244 0.80 0.59-1.07
p-trend
0.406
0.558
0.249 p
het
=0.097
80
Table 4.12 (cont.) Associations between intake of processed meats and liver and risk of bladder cancer
All Subjects Never Smokers Ever Smokers
Co/Ca OR
1
95% CI Co/Ca OR 95% CI Co/Ca OR
95% CI
Hot Dogs/Polish Sausage
< 4 times a year 339/319 1.0
Ref
133/70 1.0
Ref
206/249 1.0
Ref
4-11 times a year 444/450 1.01 0.81-1.25 171/84 0.88 0.59-1.31 273/366 1.05 0.82-1.36
monthly 262/263 1.00 0.78-1.27 104/42 0.75 0.46-1.19 158/221 1.14 0.85-1.52
twice monthly 357/446 1.18 0.94-1.48 119/72 1.10 0.71-1.70 238/374 1.25 0.96-1.62
weekly 171/169 0.88 0.66-1.18 44/27 1.08 0.58-2.00 127/142 0.87 0.62-1.21
p-trend
0.926
0.433
0.705 p
het
=0.331
Total Processed Meat
< once a week 328/281 1.0
Ref
157/76 1.0
Ref
171/205 1.0
Ref
1-2 times a week 304/275 0.96 0.76-1.23 135/53 0.79 0.51-1.21 169/222 1.06 0.78-1.42
3 times a week 322/365 1.11 0.87-1.41 115/61 1.03 0.67-1.59 207/304 1.13 0.85-1.50
4-6 times a week 304/381 1.23 0.96-1.58 92/59 1.26 0.80-1.98 212/322 1.25 0.93-1.68
1+ times a day 315/345 0.97 0.74-1.27 72/46 1.20 0.69-2.07 243/299 0.95 0.69-1.30
p-trend
0.846
0.167
0.682 p
het
=0.017
Liver
never 537/523 1.0
Ref
210/100 1.0
Ref
327/423 1.0
Ref
< once a year 152/152 1.08 0.82-1.41 72/31 0.94 0.57-1.55 80/121 1.18 0.85-1.64
1-3 times a year 255/240 0.97 0.77-1.22 91/46 1.12 0.72-1.74 164/194 0.95 0.73-1.23
4-11 times a year 371/423 1.10 0.90-1.34 116/61 1.18 0.79-1.78 255/362 1.08 0.86-1.35
monthly 258/309 1.26 1.00-1.60 82/57 1.76 1.09-2.85 176/252 1.15 0.88-1.52
p-trend
0.039
0.016
0.291 p
het
=0.118
OR
ref
= reference category; ORs
adjusted for BMI (underweight/normal <25, overweight 25-30, obese >30), race/ethnicity (non-Hispanic
white/Hispanic/black or other), education (high school/1-4 years college/grad school), history of diabetes (yes/no), total vegetable intake per day, vitamin
A intake (IU per day), vitamin C intake (mg per week), carotenoid intake (mcg per day) & total servings of food per day; smoking duration (years
smoked), smoking intensity (cigarettes per day); OR
1
further adjusted for smoking status (never/ex/current); p
het
= term for heterogeneity of trend
between never and ever smokers
81
When considering estimated total dietary intake of nitrates, nitrites, nitrosamines, and heme
iron, we did not observe associations with bladder cancer risk, with the exception of a non-
statistically significant positive association with high levels of heme intake (Q5 vs. Q1
OR=1.32; 95% CI=1.00-1.73, p
trend
= 0.191) (Table 4.13). This association was stronger
among never smokers (heterogeneity p=0.002). Among never smokers, high intake heme
was associated with almost twice the risk of bladder cancer when compared to the lowest
quintile (OR=1.97; 95% CI=1.16-3.33; p
trend
=0.010). This trend was not seen in ever
smokers. Similar effect modification by smoking was observed for total nitrite intake
(heterogeneity p=0.002) and total nitrosamine intake (heterogeneity p=0.045), with non-
significant associations between the highest levels of intake and bladder cancer risk only
among never smokers (Table 4.13). There was no effect modification of these variables by
gender or risk of progression based on stage and grade information.
4.3.2 Precursors of endogenous NOC formation and Bladder Cancer risk
Dietary nitrates can contribute to endogenous formation of NOCs when combined with
amines, including those present in various meats. We therefore hypothesized that diets high
in both sources may contribute to high levels of endogenous NOC formation, thereby
increasing bladder cancer risk.
82
Table 4.13 Associations between nitrate, nitrite, NOCs and heme iron, and risk of bladder cancer
All Subjects Never Smokers Ever Smokers
Daily Intake Co/Ca OR
1
95% CI Co/Ca OR 95% CI Co/Ca OR
2
95% CI
Nitrate (mg)
Q1: ≤64.3 315/467 1.0
Ref
116/68 1.0
Ref
199/399 1.0
Ref
Q2: 64.4-91.4 314/329 0.79 0.63-1.01 111/60 0.95 0.59-1.53 203/269 0.75 0.57-0.98
Q3: 91.5-117.3 315/293 0.74 0.57-0.97 110/54 0.91 0.54-1.53 205/239 0.71 0.52-0.96
Q4: 117.4-148.3 315/274 0.78 0.58-1.06 107/62 1.17 0.65-2.11 208/212 0.70 0.49-1.00
Q5: ≥148.4 314/284 0.90 0.60-1.35 127/51 0.81 0.36-1.83 187/233 0.96 0.60-1.54
p-trend
0.598
0.844 0.759 p
het
=0.333
Nitrite (mcg)
Q1: ≤234 314/400 1.0
Ref
142/76 1.0
Ref
172/324 1.0
Ref
Q2: 235-311 316/287 0.75 0.59-0.94 131/54 0.81 0.52-1.27 185/233 0.72 0.54-0.95
Q3: 312-400 313/302 0.81 0.63-1.03 120/59 1.03 0.65-1.62 193/243 0.74 0.55-0.99
Q4: 401-532 315/314 0.82 0.64-1.07 94/52 1.19 0.71-1.99 221/262 0.73 0.54-0.98
Q5: ≥533 315/344 0.89 0.66-1.20 84/54 1.56 0.85-2.87 231/290 0.77 0.54-1.08
p-trend
0.921
0.063 0.341 p
het
=0.002
Nitrosamine (ng)
Q1: ≤14.6 316/290 1.0
Ref
142/68 1.0
Ref
174/222 1.0
Ref
Q2: 14.7-24.8 313/318 1.08 0.85-1.37 143/75 1.01 0.67-1.54 170/243 1.09 0.81-1.46
Q3: 24.9-36.2 314/333 1.01 0.79-1.29 109/54 0.99 0.63-1.56 205/279 1.03 0.77-1.38
Q4: 36.3-54.4 316/341 1.03 0.80-1.32 103/42 0.77 0.47-1.26 213/299 1.12 0.83-1.50
Q5: ≥54.5 314/365 1.03 0.78-1.36 74/56 1.52 0.86-2.66 240/309 0.96 0.69-1.33
p-trend
0.984
0.281 0.701 p
het
=0.045
Heme
3
(mg)
Q1: ≤1.0 315/253 1.0
Ref
160/65 1.0
Ref
155/188 1.0
Ref
Q2: 1.0-2.1 314/350 1.24 0.98-1.57 134/63 1.15 0.75-1.77 180/287 1.22 0.90-1.64
Q3: 2.2-3.4 313/334 1.14 0.90-1.46 109/58 1.35 0.86-2.12 204/276 1.06 0.78-1.43
Q4: 3.4-5.1 315/312 1.04 0.80-1.34 82/47 1.41 0.86-2.31 233/265 0.90 0.66-1.21
Q5: ≥5.2 316/398 1.32 1.00-1.73 86/62 1.97 1.16-3.33 230/336 1.15 0.83-1.59
p-trend
0.191
0.010 0.851 p
het
=0.002
OR
ref
= reference category; ORs adjusted for BMI (underweight/normal <25, overweight 25-30, obese >30), race/ethnicity (non-Hispanic
white/Hispanic/black or other), education (high school/1-4 years college/grad school), history of diabetes (yes/no), total vegetable intake per
day, vitamin A intake (IU per day), vitamin C intake (mg per week), carotenoid intake (mcg per day) & total servings of food per day; OR
1
further adjusted for smoking duration (years smoked), smoking intensity (cigarettes per day) & smoking status (never/ex/current); OR
2
further
adjusted for smoking duration (years smoked) & smoking intensity (cigarettes per day); p het = term for heterogeneity of trend between never
and ever smokers;
3
Heme from processed meats and liver only
83
To address this hypothesis, we conducted analyses treating nitrate intake as a potential
modifier of associations between bladder cancer risk and consumption of each of the
processed meats, liver and heme intake (Table 4.14). Estimates of associations between
intake of each meat type and bladder cancer risk were consistently higher among groups
reporting high nitrate intake compared to those reporting low nitrate intake (Table 4.14). In
particular, nitrate intake appeared to modify the associations between bladder cancer risk and
intake of liver (p=0.010), hot dogs/Polish sausage (p=0.005) and heme (p=0.010).
Specifically, our findings suggest that among individuals who had diets high in nitrate intake
(i.e. greater than the median intake among controls, or 103 mg per day), diets high in liver
were positively associated with bladder cancer risk (highest category versus lowest category
OR=1.48; 95% CI=1.09-2.01; p
trend
=0.001). In contrast, no such association was observed
among individuals with low nitrate intake (highest category versus lowest category OR=0.96;
95% CI=0.68-1.35;
p
trend
=0.646). Similarly, among individuals with high nitrate intake, diets
high in hot dogs/Polish sausage were positively associated with bladder cancer risk (highest
category versus lowest category OR = 1.36; 95% CI = 0.91-2.04; p
trend
=0.062); whereas
among individuals with low nitrate intake, an association with opposite direction (i.e., an
inverse association) was observed (highest category versus lowest category OR = 0.58; 95%
CI = 0.39-0.87; p
trend
=0.052). Finally, among individuals with high nitrate intake we observed
a positive association between high heme intake and risk of bladder cancer (highest category
versus lowest category OR = 1.76; 95% CI = 1.21-2.55; p
trend
=0.007). Modification of the
total processed meat – bladder cancer association by nitrate intake also approached
significance (p
interaction
=0.055). We found no evidence that intake of nitrites, nitrosamines, nor
various antioxidants and/or dietary factors known to inhibit endogenous nitrosation (i.e.
84
fruits and vegetables, Vitamin A, Vitamin C, and carotenoids), modified effects of sources of
amines.
Table 4.14 Associations between intake of processed meats, liver and heme, and risk of bladder cancer stratified
by nitrate intake
Exposure
Low Nitrate Intake
(<103 mgs per day)
High Nitrate Intake
(≥103 mgs per day)
Interaction p-
value
Co/Ca OR
1
95% CI Co/Ca OR
1
95% CI
Fried Bacon/Ham
< once a month
220|232 1.0
Ref.
237|180 1.0
Ref.
1-2 times a month
97|105 0.98 0.69-1.39 73|67 1.13 0.75-1.69
3-4 times a month
225|293 1.08 0.82-1.41 202|197 1.18 0.88-1.57
weekly
123|155 0.90 0.65-1.24 122|121 1.14 0.81-1.60
2+ times a week
122|155 0.87 0.62-1.21 152|142 1.05 0.75-1.47
p-trend
0.267
0.980 0.396 (1df)
Salami/Pastrami/Corned Beef
< twice a year
210|230 1.0
Ref.
191|137 1.0
Ref.
2-11 times a year
269|307 1.02 0.79-1.33 219|184 1.14 0.84-1.55
monthly
106|113 1.01 0.71-1.42 106|79 1.29 0.88-1.89
twice monthly
126|165 1.14 0.83-1.56 159|194 1.60 1.16-2.22
weekly
76|125 1.34 0.94-1.93 112|128 1.37 0.94-2.00
p-trend
0.078
0.035 0.763 (1df)
Bologna/Other Lunch Meats
never
122|173 1.0
Ref.
198|170 1.0
Ref.
< once a month
212|212 0.72 0.53-0.99 178|151 0.94 0.68-1.28
monthly
238|276 0.75 0.55-1.01 197|182 1.02 0.76-1.39
weekly
81|124 0.92 0.63-1.36 81|74 0.98 0.66-1.47
twice weekly
134|155 0.67 0.47-0.95 132|130 0.98 0.69-1.39
p-trend
0.239
0.992 0.390 (1df)
Hot Dogs/Polish Sausage
< 4 times a year
139|181 1.0
Ref.
177|140 1.0
Ref.
4-11 times a year
243|267 0.82 0.61-1.10 210|186 1.26 0.92-1.71
monthly
145|177 0.94 0.68-1.31 117|96 0.99 0.69-1.44
twice monthly
172|235 0.94 0.69-1.29 188|200 1.49 1.09-2.03
weekly
88|80 0.58 0.39-0.87 95|100 1.36 0.91-2.04
p-trend
0.052
0.062 0.005 (1df)
85
4.3.3 Processed Meats, modification by DNA Repair Enzymes and risk of
Bladder Cancer
We explored the potential modifying role of genetic variation in 28 DNA repair genes (549
tSNPs) that participate in four key DNA repair pathways, on the effect of dietary sources of
NOCs. Figures 4.1 through 4.4 show plots of the significance of each SNP by exposure
Table 4.14 (cont.) Associations between intake of processed meats, liver and heme, and risk of bladder cancer
stratified by nitrate intake
Exposure
Low Nitrate Intake
(<103 mgs per day)
High Nitrate Intake
(≥103 mgs per day)
Interaction p-
value
Co/Ca OR
1
95% CI Co/Ca OR
1
95% CI
Total Processed Meat
< once a week
151|162 1.0
Ref.
177|119 1.0
Ref.
1-2 times a week
156|162 0.95 0.69-1.33 148|113 0.97 0.68-1.39
3 times a week
184|231 0.98 0.72-1.35 138|134 1.30 0.91-1.85
4-6 times a week
146|206 1.06 0.76-1.48 158|175 1.47 1.05-2.08
1+ times a day
150|179 0.81 0.57-1.14 165|166 1.20 0.83-1.74
p-trend
0.305
0.154 0.055 (1df)
Liver
never
291|316 1.0
Ref.
246|207 1.0
Ref.
< once a year
80|105 1.30 0.92-1.85 72|47 0.82 0.53-1.25
1-3 times a year
138|162 1.08 0.81-1.45 117|78 0.83 0.58-1.19
4-11 times a year
179|244 1.15 0.88-1.49 192|179 1.06 0.79-1.41
monthly
99|113 0.96 0.68-1.35 159|196 1.48 1.09-2.01
p-trend
0.646
0.001 0.010
Heme Intake
2
Q1 163/159 1.0
Ref.
152/94 1.0
Ref.
Q2 174/219 1.13 0.83-1.55 141/131 1.40 0.97-2.03
Q3 157/211 1.09 0.79-1.51 157/124 1.24 0.86-1.79
Q4 161/174 0.87 0.62-1.21 154/138 1.33 0.92-1.92
Q5 133/178 1.01 0.71-1.43 183/221 1.76 1.21-2.55
p-trend
0.532
0.007 0.010
OR
ref
= reference category; OR = adjusted for smoking duration (years smoked), smoking intensity (cigarettes per day),
smoking status (never/ex/current), BMI (underweight/normal <25, overweight 25-30, obese >30), and race/ethnicity (non-
Hispanic white/Hispanic/black or other), education (high school/1-4 years college/grad school), history of diabetes (yes/no),
total vegetable intake per day, vitamin A intake (IU per day), vitamin C intake (mg per week), carotenoid intake (mcg per
day) and total servings of food per day;
2
Heme from processed meats and liver only
86
interaction in each of the four pathways; these plots use the salami/pastrami and corned
beef intake exposure as an illustration, although these tests were run for all exposures. Table
4.15 shows interactions between dietary variables and SNPs that remained significant after
Bonferroni adjustment for multiple testing. Only one exposure by SNP interaction was
statistically significant after applying multiple testing corrections across all SNPs, genes, and
pathways tested. This SNP (rs10774478) is in the RAD52 gene, active in the HRR pathway
(Table 4.15, Figure 4.2, Figure 4.5), and modified the association between
salami/pastrami/corned beef intake and bladder cancer risk. The corrected p-value for this
effect modification was 0.042. Among individuals with two copies of the rs10774478 minor
allele (AA), high salami intake was associated with five times the risk of bladder cancer
compared to those with low salami intake (Figure 4.5). This effect was not seen when
considering those with two copies of the rs1077478 major allele (GG). These associations
were most evident when restricted to male subjects only (Figure 4.6, Figure 4.7). Among
males only, a second SNP in the HRR pathway, rs6464262 in the XRCC2 gene was also
significant when adjusted for multiple testing across all SNPs (Figure 4.6). Another SNP of
note in this analysis was rs1571069 in the ERCC5 gene, which was significant at the gene
level for two exposures, salami/pastrami/corned beef intake and nitrite intake (Table 4.15).
This SNP is found in the NER pathway.
87
*In this, and the following figures, each dot plots the significance of SNP by exposure interaction (y-axis) by the SNP
position (x-axis)
**The three lines represent gene-wide, pathway-wide and all SNP significance cut-offs respectively
Figure 4.1: Salami/pastrami/corned beef intake and SNPs in the BER pathway
Figure 4.2: Salami/pastrami/corned beef intake and SNPs in the HRR pathway
88
Figure 4.3: Salami/pastrami/corned beef intake and SNPs in the NER pathway
Figure 4.4: Salami/pastrami/corned beef intake and SNPs in the NHEJ pathway
89
Figure 4.5: Effect modification of salami/pastrami/corned beef intake by rs1077478
90
Table
4.15
Significant
SNP
by
Exposure
interactions
after
Bonferonni
adjustment
Exposure
Pathway
Gene
SNP
Level
of
Correction
gene
pathway
overall
Fried
bacon/Ham
BER
HR
NEIL1
MRE11A
XRCC3
rs11635996
rs1805363
rs4900589
rs8008451
x
x
x
x
Salami/Pastrami/Corned
Beef
HR
NER
RAD52
XRCC2
ERCC5
rs10774478
rs6464262
rs876430
rs1571069
rs9557946
x
x
x
x
x
x
x
Bologna/Other
Lunch
Meats
NER
ERCC5
rs4150355
x
Hot
dogs/Polish
Sausage
NHEJ
LIG4
rs1555902
x
Total
processed
meat
none
Liver
none
Nitrate
none
Nitrite
NER
ERCC5
rs1571069
rs1886087
x
x
Nitrosamine
none
91
Figure 4.6: Salami/pastrami/corned beef intake and SNPs in the HRR pathway
among males only
Figure 4.7: Salami/pastrami/corned beef intake and SNPs in the HRR pathway
among females only
92
4.3 Fish and Meat Intake, Metabolic Enzymes and Risk of
Colorectal Cancer: Colorectal Family Registry
Table 4.16 outlines the distribution of cases and types of controls in the CFR, according to
gender, race and center. All three centers included in this analysis had sibling controls;
whereas Hawaii only had spouse controls, and population based controls came solely from
Ontario. The majority of study participants were Non-Hispanic White making up 59%,
African-Americans made up 21%, Asians made up 13% and 7% were other ethnicities.
Ontario contributed the most subjects to this analysis (49%), followed by the USC
consortium (40%) and Hawaii the least (11%).
Table 4.16. Distribution of cases and controls in the Colon CFR
Cases All controls
Sibling
Controls
Spouse
Controls
Population-
Based
Controls
Characteristics N N N N N
Total 4503 4197 2410 157 1630
Sex
Male 2124 1922 1005 64 853
Female 2379 2275 1405 93 777
Race
Non-Hispanic White 2677 3133 1604 31 1498
African-American 924 307 303 2 2
Asian 573 502 342 99 61
Other 329 255 161 25 69
Center
Ontario 2192 2672 1042 0 1630
Los Angeles 1808 981 981 0 0
Hawaii 503 544 387 157 0
93
4.3.1 Main Effects of Fish intake on risk of Colorectal Cancer
There was no association seen between total fish intake and risk of colorectal cancer in this
dataset. Table 4.17 illustrates results from four different case-control comparisons we
performed for each variable. We found no significant differences among the different
control groups, therefore the following tables will be displaying the most comprehensive
comparison, which compares all probands (cases) to all available controls (siblings, spouses,
population-based).
Table 4.17: Total Fish Intake and Risk of Colorectal Cancer
Proband – All Controls Proband – Sibling Controls
GEE Regression N:M Conditional Logistic Regression
Total Fish Co/Ca OR
1
95% CI Co/Ca OR
1
95% CI
Q1
701/626
1.0
Ref.
317/196
1.0
Ref.
Q2 701/630 1.0 0.8-1.1 333/217 1.0 0.8-1.3
Q3 701/667 1.0 0.9-1.2 302/230 1.1 0.8-1.5
Q4 700/734 1.1 0.9-1.3 337/246 1.2 0.9-1.5
Q5 701/693 1.0 0.8-1.2 321/191 0.9 0.7-1.2
p
trend
0.881
0.644
Proband – Spouse Controls Proband – Population Based Controls
1:1 Conditional Logistic Regression Unconditional Logistic Regression
Total Fish Co/Ca OR
1
95% CI Co/Ca OR
1
95% CI
Q1 10/11 1.0
Ref.
345/626 1.0
Ref.
Q2 20/15 0.7 0.2-2.5 318/630 1.0 0.8-1.2
Q3 28/23 0.6 0.2-2.0 340/667 1.0 0.8-1.2
Q4 25/26 1.1 0.3-3.9 312/734 1.1 0.9-1.3
Q5 45/53 1.1 0.3-3.4 292/693 0.9 0.7-1.1
p
trend
0.358 0.389
ORs
1
adjusted for age (years), BMI (<25, 25.0-29.9, ≥30), total calorie intake (kcal/day, continuous), gender (male,
female), recent physical activity (yes, no), center (Hawaii, USC, Ontario), saturated fat intake (g/kcal, continuous),
fiber intake (g/kcal, continuous), vegetable intake (g/kcal, continuous) and race (NHW, Black, AA, Other).
94
Shellfish intake was a risk factor for colorectal cancer in this dataset (Table 4.18). Those in
the highest quintile of shellfish intake were 1.3 times as likely to develop colorectal cancer as
those who never ate shellfish. There was also a significant trend (p=0.028). Intake of non-
shellfish (all fish excluding shellfish) showed no association with risk of CRC in this dataset
(Table 4.18). When stratifying the shellfish results by familial history of cases (based on
Amsterdam criteria), although we found evidence of the association in both familial and
non-familial cases, the effect was stronger in cases with familial history (p
heterogeneity
= 0.049)
(Table 4.19). When comparing familial cases to all controls, those in the highest quintile of
shellfish intake now had twice the risk of CRC compared to those in the lowest quintile.
Among non-familial cases, the association was similar to all cases combined (Table 4.19). We
saw no differences in association with fish intake and CRC risk when comparing fish that
was cooked using different methods, i.e., pan-frying, grilling or barbequing. We also saw no
difference in associations when stratifying by MSI status. The association between shellfish
intake and CRC risk appeared stronger in the distal part of the colon, in particular, the
sigmoid colon and the rectum. However, case-only analysis did not show these differences to
be significant, and numbers in sub-groups were small.
95
Table 4.18: Intake of Shellfish and Non-Shellfish and risk of CRC
Shellfish (g/kcal) Co/Ca OR
1
95% CI
Q1: 0-0.1 702/555 1.0
Ref.
Q2: 0.1-0.3 700/647 1.3 1.1-1.5
Q3: 0.3-0.8 700/632 1.2 1.0-1.4
Q4: 0.8-1.9 701/685 1.2 1.0-1.4
Q5: 1.9-34.7 701/831 1.3 1.1-1.5
p-for trend 0.028
Non-Shellfish (g/kcal)
Q1: 0-2.7 701/601 1.0
Ref.
Q2: 2.7-5.0 701/728 1.2 1.0-1.4
Q3: 5.0-7.7 701/659 1.0 0.9-1.2
Q4: 7.7-11.6 700/684 1.1 0.9-1.3
Q5: 11.6-95.0 701/678 1.0 0.9-1.2
p-for trend 0.99 0.672
ORs
1
adjusted for age (years), BMI (<25, 25.0-29.9, ≥30), total calorie intake (kcal/day, continuous), gender (male,
female), recent physical activity (yes, no), center (Hawaii, USC, Ontario), saturated fat intake (g/kcal, continuous),
fiber intake (g/kcal, continuous), vegetable intake (g/kcal, continuous) and race (NHW, Black, AA, Other).
Table 4.19: Intake of Shellfish and risk of CRC stratified by familial history of cases
1
Familial cases
1
vs. All controls
Co/Ca OR
2
95% CI
Q1 702/22 1.0
Ref.
Q2 700/25 1.1 0.6-2.0
Q3 700/22 0.9 0.5-1.7
Q4 701/35 1.5 0.8-2.6
Q5 701/38 2.0 1.1-3.6
p-for trend 0.001
Non-familial cases vs. non-familial controls
Q1 679/519 1.0
Ref.
Q2 670/591 1.2 1.1-1.5
Q3 673/589 1.2 1.0-1.4
Q4 680/616 1.1 1.0-1.3
Q5 678/763 1.3 1.1-1.5
p-for trend 0.073
p-for heterogeneity 0.049
1
cases defined as familial based on Amsterdam criteria
ORs
2
adjusted for age (years), BMI (<25, 25.0-29.9, ≥30), total calorie intake (kcal/day, continuous), gender (male,
female), recent physical activity (yes, no), center (Hawaii, USC, Ontario), saturated fat intake (g/kcal, continuous),
fiber intake (g/kcal, continuous), vegetable intake (g/kcal, continuous) and race (NHW, Black, AA, Other)
96
4.3.2 Meat and fish intake and modification by GWAS identified cancer
susceptibility genes
We explored the potential modifying role on meat intake variables and the risk of CRC of 16
SNPs that had previously been identified as important in CRC risk via GWAS. Overall, we
tested 16 different meat-related exposures with each of the 16 SNPs. These variables were as
follows: total meat, red meat, white meat, red processed meat, white processed meat, beef,
pork, organ meat, sausage, all fish, shellfish, non-shellfish, HAA, PhIP, DiMeiQx and
MeiQx. Of these multiple comparisons, two survived Bonferroni correction for the number
of SNPs tested (Table 4.20). A SNP in the SMAD7 gene, rs4939827, modified the effect of
processed white meat intake, with a corrected interaction p-value of 0.0064. Among
individuals homozygous for the wild-type allele (T/T), high intake of processed white meat
was associated with twice the risk of CRC (OR = 2.2; 95% CI = 1.4-3.5; p
trend
< 0.001). This
association was not seen among T/C heterozygotes, and among C/C homozygotes, intake
of processed white meat was associated with a decrease in risk (Table 4.20). In addition, an
intergenic SNP, rs6691170, modified the effect of sausage intake with a corrected interaction
p-value of 0.0432. Among individuals homozygous for the wild-type allele (G/G), high
intake of sausage showed no association with the CRC risk (Table 4.20). However, among
G/T heterozygotes and even more so T/T homozygotes, increasing intake of sausage
increased CRC risk. Among individuals homozogous for the minor allele (T/T), high
sausage intake was associated with twice the risk of CRC (OR = 2.0; 95% CI = 1.1-3.8; p
trend
< 0.001) (Table 4.20). Although these interactions remained signification after correction
97
across all SNPs tested, it should be noted that neither interaction would remain significant if
we had corrected for all GxE tests performed (256).
Table 4.20: Meat intake variables and SNP interactions that survived Bonferroni correction
Co/Ca OR
1
(95% CI)
rs4939827
Processed White Meat
Intake
T/T T/C C/C T/T T/C C/C
Never/Rarely
Low
High
p
trend
p
2
heterogeneity of trend
70/47
62/70
42/84
91/106
94/103
104/115
35/43
40/35
50/46
1.0
Ref
1.5 (0.9-2.3)
2.2 (1.4-3.5)
<0.001
1.4 (1.0-1.8)
1.3 (0.9-1.9)
1.5 (1.0-2.3)
0.658
1.9 (1.1-3.4)
1.2 (0.7-2.0)
1.0 (0.6-1.6
0.014
0.0004
rs6691170
Sausage Intake
G/G G/T T/T G/G G/T T/T
Never/Rarely
Low
High
p
trend
p
2
heterogeneity of trend
59/50
59/51
53/48
65/62
62/55
59/84
20/5
12/13
14/31
1.0
Ref
0.8 (0.5-1.4)
0.8 (0.4-1.3)
0.331
0.7 (0.5-1.1)
0.9 (0.5-1.4)
1.2 (0.8-2.0)
0.006
0.5 (0.2-1.1)
0.9 (0.4-1.9)
2.0 (1.1-3.8)
<0.001
0.0027
OR¹ odds ratios are from models with genotype coded as log-additive (i.e. number of C alleles, treated as continuous)
p
2
values unadjusted for Bonferroni; Bonferroni adjusted p-values can be found in text
98
CHAPTER 5: DISCUSSION
It is estimated that environmental factors are responsible for upwards of 80% of all cancers.
Of these environmental factors, diet accounts for approximately 30% of all cancers in
Western countries, and approximately 20% of cancers in the developing world (Nahleh,
Bhatti et al. 2011). This is good news from a preventative point of view, as dietary patterns
are highly modifiable. If we can identify dietary factors that put us at risk for cancer, in
addition to those that protect us against cancer, we are well on our way towards preventing a
large number of cancer cases. In fact, diet is considered second only to tobacco as a
preventable cancer cause (Nahleh, Bhatti et al. 2011). Research has long pointed to “healthy”
dietary patterns, i.e. diets high in fresh fruit and vegetables and whole grains, as protective
when compared to “unhealthy” dietary patterns, i.e. diets high in animal products, saturated
fat and overall calories. Exactly which aspect of the “unhealthy” dietary pattern is
responsible for an increase in cancer risk is not entirely clear. Moreover, it is hypothesized
that there exists a synergy between different dietary and inherent genetic factors.
Recent years have seen great advances in molecular technology, allowing us to now
quickly and inexpensively gather large amounts of genetic information on an individual.
Incorporating genetic information into epidemiological studies has become a valuable tool in
gaining insight into the potential mechanisms of carcinogens and risk factors. By gathering
information on a study subject’s genetic variation, and combining this with information on
dietary patterns, we may be able to understand the mechanisms by which specific dietary
factors confer risk or protection against cancer. The aim of this is to translate this knowledge
99
into dietary recommendations and development of cooking and preparation practices that
avoid the harmful effects of certain foods and preparation methods, while taking advantage
of any protective effects. In the sections below I summarize our findings from the specific
studies conducted and discuss the overall role of dietary carcinogens and associated genetic
variants in the risk of carcinogenesis.
5.1 Genetic variation in metabolic enzymes and the risk of
prostate cancer
In the California Collaborative Case-Control Study of Prostate Cancer we found that genetic
variants in several metabolic enzymes, CYP1B1 Leu432Val, EPHX1 Tyr113His and GSTT1
present/null, are associated with risk of localized PCA (Catsburg, Joshi et al. 2012). We also
found evidence that the PTGS2 765 G/C polymorphism modifies the association between
diets high in white or dark fish and advanced PCA risk. The modifying effect appears
stronger for consumption of well-done white (lean) fish.
There exists substantial biological plausibility for a role of CYP1B1 in the
development of PCA. PAHs that enter the prostate induce CYP1B1 expression. CYP1B1 is
over-expressed in prostate carcinomas (Carnell, Smith et al. 2004) and can activate PAHs
into mutagenic metabolites that are capable of forming DNA adducts (Martin, Patel et al.
2010). The CYP1B1 protein coded by the codon 432 Valine allele is more active than the
one coded by the Leucine allele (Tang, Green et al. 2000); therefore, our finding of an
100
association between the Valine (Val) allele and higher PCA risk is compatible with the
known functional impact of this variant allele on the CYP1B1 protein. We distinguished
between advanced and localized PCA and we found the CYP1B1 codon 432 Val allele to be
associated with increased risk of localized disease only. Our findings are supported by two
studies that found the CYP1B1 Leu432Val Val allele to be associated with an increased risk
of PCA among Caucasian (Tang, Green et al. 2000) and Japanese (Fukatsu, Hirokawa et al.
2004) men. These studies did not distinguish between localized or advanced cases; therefore
it is difficult to fully compare those findings with ours. Another study reported the Val allele
being associated with a decreased risk of advanced disease in Hispanic men and with
decreased disease aggressiveness (localized vs. advanced) in non-Hispanic white men
(Beuten, Gelfond et al. 2008) which is consistent with our study finding the increased risk in
localized disease only.
EPHX1 codes for the epoxide hydrolase protein mEH, a Phase 1 enzyme that
converts PAH-epoxides to PAH-dihydrodiol, which can undergo further metabolic
activation by CYP1A1 and CYP1B1, or detoxification by enzymes in the GST or UGT
family (Pastorelli, Guanci et al. 1998). Therefore, this enzyme plays an important role in the
generation of species that are at the crossroads between carcinogenic activation and
detoxification. The EPHX1 codon 113 Histidine (His) allele codes for a protein with
reduced activity compared to the one coded by the Tyrosine allele (Hassett, Lin et al. 1997).
Our finding of a positive association between the His allele and localized PCA risk suggests
that the slow EPHX1 His allele may contribute toward increased metabolic activation,
perhaps by reducing the bioavailability of substrates for PAH-epoxide detoxification.
Alternatively, reduced levels of EPHX1 may result in an increased bioavailability of pro-
101
carcinogenic PAH epoxides. PAH-epoxides can serve as direct substrates for AKR1C3 in the
prostate where it can undergo conversion to catechols, which may lead to the accumulation
of DNA-damaging PAH-O-quinones (Penning, Burczynski et al. 1999). One study reported
a positive association between the codon 113 His allele and advanced PCA risk in an Indian
population, and another reported an association between this allele and higher-grade tumors
in an Israeli population (Figer, Friedman et al. 2003, Mittal and Srivastava 2007).
GSTT1 belongs to the glutathione-S-transferase (GST) superfamily and is
responsible for detoxifying many carcinogens, including HCAs and PCAs. Approximately
20% of Caucasians have a complete deletion of this gene, which may predispose to cancer
due to a decreased ability to detoxify these substrates (Strange and Fryer 1999). In recent
years, GSTT1 status has been extensively studied as a PCA risk factor; however, the results
are inconsistent. A 2009 meta-analysis concluded that there was no significant evidence that
the GSTT1 deletion increased risk for PCA, regardless of racial group (Mo, Gao et al. 2009).
Since then, six additional studies have investigated the association with mixed results. Three
reported an approximately two-fold increase in risk with GSTT1 deletion in Iranian, Tunisian
and Danish populations (Souiden, Mahdouani et al. 2010, Norskov, Frikke-Schmidt et al.
2011, Safarinejad, Shafiei et al. 2011), whereas Tailoi et al reported an inverse association
with GSTT1 deletion in populations of African descent (Taioli, Flores-Obando et al. 2011).
Two additional studies reported no associations in the German EPIC cohort (Steinbrecher,
Rohrmann et al. 2010) and a population of African descent (Lavender, Benford et al. 2009).
Few of these studies distinguished between localized and advanced cases; however,
Safarinejad et al. (Safarinejad, Shafiei et al. 2011) found the increase to be even more
substantial in those with advanced PCA. In our study population we found the GSTT1
102
deletion variant to be associated with increased risk of localized PCA, with a similar trend,
albeit not statistically significant, for advanced PCA. Our results are compatible with the
hypothesis that lack of GSTT1 detoxification may contribute to accumulation of activated
HCA and PAH metabolites, which could contribute to prostate carcinogenesis.
It is interesting to note that for these three polymorphisms statistically significant
associations were only found for localized disease, but not advanced disease. As we reviewed
above, there are previous studies that have reported on associations between the EPHX1
Tyr113His and the GSTT1 deletion and risk of advanced disease. Therefore, further studies
with larger numbers of both localized and advanced cases are needed to clarify whether
these enzymes play a differential role for early or late disease.
5.2 Fish intake, variation in metabolic enzymes, and the risk of
prostate cancer
In the California Collaborative Case-Control Study of Prostate Cancer, we observed
evidence that the PTGS2 765 G/C polymorphism modified the association between high
white fish intake and advanced PCA risk (Catsburg, Joshi et al. 2012). We had slightly less
statistical power for detecting significant gene-diet interactions for localized disease than
advanced disease. However, the fact that we found no strong evidence of effect modification
of the association between consumption of white fish cooked at high temperature and
103
localized or advanced PCA risk and any of the three polymorphisms that showed a main
effect, may suggest that there are other relevant substrates for these enzymes. Larger studies
are needed to confirm this. Our observed association between high intake of white fish
cooked at high temperature (i.e., mostly pan-frying) and PCA risk could be driven by other
changes that occur in the fish during the cooking process (Joshi, John et al. 2012 ).
Specifically, it is plausible that changes in white fish fatty acid composition may occur when
white fish is cooked at high temperature, such as pan-frying, and that these changes are the
ones driving the observed associations (Joshi, John et al. 2012 ). If this hypothesis were true,
it would explain why we failed to find gene-by-environment interactions for these three
metabolism enzyme polymorphisms, in spite of their overall association with localized PCA.
Interestingly, as we discuss below, we only observed an effect modification for PTGS2,
which participates in both carcinogen metabolism and polyunsaturated fatty acid
metabolism.
To our knowledge, the only metabolic enzyme reported to date as a potential
modifier of the association of PCA with high fish intake is the PTGS2 enzyme (Hedelin,
Chang et al. 2007, Fradet, Cheng et al. 2009). In our study, we investigated eight other
biologically plausible enzymes jointly with fish intake and found that the PTGS2 765 G/C
polymorphism was the only one that showed evidence of interaction with white and dark
fish intake. PTGS2 is over-expressed in prostate cancer cells and is induced by HCAs and
PAHs, which it can activate into carcinogens (Kirschenbaum, Klausner et al. 2000, Moonen,
Briede et al. 2002, Wang, Bergh et al. 2004). PTGS2 is also involved in the metabolism of
omega-6 PUFAs, such as arachidonic acid, into prostaglandins that can contribute to
carcinogenesis. Moreover, omega-3 PUFAs found in fish, which are particularly abundant in
104
dark fish, have been shown to block the activity of PTGS2 in hepatic and colon cells (Lim,
Han et al. 2009). Therefore, it is biologically plausible that genetic variation in PTGS2 might
modify the association between fish intake and risk of PCA. The PTGS2 765 G/C
polymorphism is located in a binding site upstream of the translation initiation site, its effect
on PTGS2 gene expression in the prostate is still unclear. In vitro, the C allele had a lower
activity than the G allele when transfected into human cervical epithelium cancer cells, but
opposite results were obtained when transfection was conducted on human neural cells
(Papafili, Hill et al. 2002, Cui, Salehi-Rad et al. 2005). Sanak and co-workers (2010) recently
reported an increase in prostaglandin production associated with the C allele in heart disease
patients; however, they simultaneously reported reduced PTGS2 activity with the C allele in
both human endothelial and human leukemia cell lines (Rybicki, Nock et al. 2006, Sanak,
Plutecka et al. 2010). We found that diets high in white fish are associated with PCA among
carriers of the C allele, and that diets high in dark fish are inversely associated with PCA risk
only among carriers of the G allele. These data suggest that the C allele might be the more
active form of PTGS2 in the prostate, and that under conditions of exposure to high levels
of pro-carcinogens such as HCAs, PAHs and/or omega-6 PUFAs, all found in diets high in
fish intake, it might be associated with risk of PCA. We found the PTGS2 interaction with
white fish to be stronger when we restrict the analysis to men who consumed well-done
white fish. This could lend support to the hypothesis that the interaction is due, at least in
part, to the formation of HCAs and PAHs, as we would expect that fish cooked longer (i.e.,
until well done) would accumulate more of these carcinogens and therefore intake of well-
done fish would increase risk to a greater extent than intake of lightly browned fish.
However, we cannot disregard the hypothesis that when white fish is cooked for a long time,
105
more fatty acid changes are accumulated that may also contribute to PCA risk, as we
previously discussed (Joshi, John et al. 2012 ). The fact that we failed to find interactions for
any of the other metabolism enzymes investigated seems to offer more support for the latter
rather than the former hypothesis. However, given that we did not include in our study all
potential enzymes that participate in HCA and PAH metabolism, we cannot dismiss a role
for these carcinogens in the observed association between fish and PCA risk.
5.3 Fresh and preserved red meat, poultry, and fish intake and
the risk of prostate cancer in Singapore
In the Singapore Chinese Health Study we found an association between diets high in
preserved meats (red meat and fish) and risk of PCA, particularly among advanced or
metastatic PCA, in a prospective cohort of Singapore Chinese. Preserved fish alone showed
a borderline statistically significant association among advanced PCA only. We also observed
that both BMI and fruit and vegetable intake may modify this association, which seems
restricted to men with above average BMI and/or below average fruit and vegetable intake.
In addition, we report an increased risk of metastatic PCA with intake of poultry with skin.
Processed meat refers to any meat that is preserved by salting, curing, smoking or
adding chemical preservatives (WCRF 2007), thus the terms processed and preserved are
used interchangeably in the literature. Existing studies point to preserved meats as a
suggestive risk factor for PCA (WCRF 2007). One mechanism proposed to explain this
106
association is the presence in preserved meats of NOCs, which are known chemical
carcinogens (Bartsch, Ohshima et al. 1988). NOC formation can be exogenous, when
nitrates and nitrates are added to meats for preservation (Mirvish 1995); or endogenous,
when biogenic amines from ingested meat react with ingested nitrates and nitrites, from
various sources, and form NOCs at various sites of the body (Hecht and Hoffman 1998).
The processing and subsequent increased storage of meats has been shown to greatly
increase biogenic amine formation, thus contributing to endogenously formed NOCs (Ruiz-
Capillas and Jimenez-Colmenero 2004). In addition, heme iron, found in high levels in red
meat and lower levels in other meats, directly stimulates this endogenous NOC formation
from amines present in non-preserved meats and nitrates and nitrites ingested from various
sources (Lunn, Kuhnle et al. 2007). Three previous cohort studies have reported an increase
in risk with increased intake of preserved meats, two were conducted within US cohorts, and
one was conducted within a cohort from the Netherlands (Schuurman, van den Brandt et al.
1999, Rodriguez, McCullough et al. 2006, Sinha, Park et al. 2009). Consistent with our
findings, both of the US cohorts found this increase in risk to be stronger among metastatic
disease (Rodriguez, McCullough et al. 2006, Sinha, Park et al. 2009). Three further cohort
studies, all US-based, reported suggestive but non-statistically significant associations with
preserved meat intake and PCA risk (Michaud, Augustsson et al. 2001, Cross, Peters et al.
2005, Rohrmann, Platz et al. 2007), two of these studies found the association to be stronger
among metastatic disease (Michaud, Augustsson et al. 2001, Cross, Peters et al. 2005). Two
additional cohorts, in the US and Europe (EPIC), reported no association (Park, Murphy et
al. 2007, Allen, Key et al. 2008). A key difference between these previous cohorts and our
study is the range of preserved meat intake investigated. The US and the Netherlands studies
107
all had very comparable preserved meat intakes with median daily intake ranging from 16.8-
26.9 grams, most of them closer to 18 grams. The EPIC study had an even higher intake
with a median intake of 53 grams per day (gpd). In contrast, the men in our study had a
median of 4.3 gpd, much lower than the 20
th
percentile of all the other studies (range was
from 0-14.6 gpd). Interestingly, our findings support the suggested association between
preserved meats and PCA risk in spite of the lower levels of intake compared to other
cohorts. This finding may suggest that there might be genetic predisposition factors in the
Singapore Chinese population that may predispose men to PCA in conjunction with lower
levels of NOC exposure compared to other populations.
We found that BMI modified the effect of preserved meat intake on PCA risk, with
the risk being much stronger among those with above median BMI. Increasing body size is
associated with hormonal and metabolic changes, including lower free testosterone levels in
men (Discacciati, Orsini et al. 2012). Lower testosterone has previously been associated with
risk of more advanced or fatal PCA (Cao and Ma 2011). In addition, being overweight can
promote disease by impairing the ability of cells to detoxify radicals and repair DNA damage
(Mattson 2009). This could explain the synergistic effect we observed in our data i.e.
overweight individuals could be less able to repair DNA damage caused by NOC
compounds or other meat carcinogens.
Fruit and vegetables contribute high levels of dietary antioxidants and other
chemicals that block carcinogenic reactions, including NOC formation (Bartsch, Ohshima et
al. 1988). Because of this, we hypothesized that individuals with low intake may be more
susceptible to damage from dietary carcinogens and thus we tested whether fruit and
vegetable intake modified the effect of our exposures. Consistent with this hypothesis, the
108
association of PCA risk and preserved meat intake was much stronger in individuals with
lower fruit and vegetable intake in this dataset.
Cooking poultry is associated with HCA formation, in particular the HCA PhIP,
which is known to form DNA adducts in human tissues (Sinha, Rothman et al. 1995). Pan-
fried, oven broiled and grilled poultry have been reported to accumulate high amounts of
PhIP. In particular, chicken with skin has higher mutagenicity measured by the Ames test
compared to skinless chicken or chicken cooked with the skin but the skin discarded
(Jakszyn, Gonzalez et al. 2011). Consistent with this, we report an association between
metastatic PCA risk and intake of poultry when the skin is not removed. The majority of
previous studies did not differentiate between poultry with and without skin; therefore, we
cannot compare our findings. To our knowledge, only one other study has looked at poultry
intake with and without skin and has found that intake of poultry with skin is associated with
increased progression of disease (Richman, Kenfield et al. 2011). Our findings may warrant
future studies to consider this factor when investigating poultry intake.
5.4 Processed meats, NOCs, and the risk of bladder cancer
In the Los Angeles Bladder Cancer study we investigated the role of dietary sources of
NOCs as potential bladder cancer risk factors. We considered processed meats as direct
sources of exogenous NOCs, but also examined possible sources of precursors for
endogenous formation of NOCs, such as liver and processed meats, which are abundant
109
sources of amines, nitrates, nitrites and heme, as well as fruits and vegetables, which are
sources of nitrates. Analyses addressed dietary factors that inhibit endogenous nitrosation
(e.g. vitamin C) as well as those that promote it (e.g. heme). Our findings suggest that diets
high in salami/pastrami/corned beef or liver may increase the risk of bladder cancer,
particularly among non-smokers (Table 4.12). In addition to contributing NOC precursors,
the liver stores many compounds that could be carcinogenic and contribute to this increased
risk. We observed an association between bladder cancer risk and heme intake among non-
smokers (Table 4.13). Moreover, our findings suggest that these associations may be
restricted to individuals with diets high in nitrate, suggesting a synergistic effect between
NOCs precursors (nitrates, nitrites, amines) present in processed meats and liver, and total
nitrate levels (Table 4.14). Fruits and vegetables are main contributors to total nitrate intake.
Therefore, our findings suggest that even though, overall, fruits and vegetables are protective
factors against bladder cancer, as previously demonstrated by our group and others, the
combination of high fruit and vegetable intake and high processed meats and/or liver have a
synergistic positive association with bladder cancer risk. Importantly, our findings highlight
the seemingly important role of dietary amines and heme iron from meats as required
precursors for endogenous nitrosation.
Most nitrosamines are rapidly metabolized in vivo and are thus not readily transported
in the blood, therefore we speculate that a very small fraction of ingested exogenous
nitrosamines reach the bladder where they could influence malignant potential of urothelial
cells (Mirvish 1995). Also, in this study population, the estimated dietary contribution of
NOCs was low, ranging from 0 to 2.1 mcg per day, whereas endogenous nitrosation has
been estimated to contribute upwards of 90 mcg per day (Jakszyn, Bingham et al. 2006,
110
Cross, Leitzmann et al. 2007). As expected, we saw no evidence of an association with
preformed NOCs and bladder cancer risk in this study.
High intake of fruit and vegetables, specifically dark-green and yellow-orange
vegetables and citrus fruits, has previously been reported as inversely associated with bladder
cancer risk in this study (Castelao, Yuan et al. 2004). However, vegetables, especially green
leafy vegetables such as spinach, lettuce and parsley, also have a very high content of nitrates
(Nabrzyski and Gajewska 1994). Nitrates are inherently present in all plant materials, and
they accumulate when the plant matures in a nitrate-rich environment; therefore, the nitrate
content of vegetables depends greatly on the nitrate level of the soil in which they were
cultivated (Wang, Zong et al. 2002). The higher level of nitrates found in some vegetables is
usually not considered to be harmful as these foods contain high levels of antioxidants and
other chemicals that block carcinogenic reactions, including nitrosation (Bartsch, Ohshima et
al. 1988, Tang and Zhang 2004, Tang and Zhang 2004, Sugie, Vinh et al. 2005). In our study,
we observed evidence of a synergistic effect between high nitrate intake and consumption of
large amounts of processed meats and liver (Table 4.14). We hypothesized that under
conditions of high amounts of nitrosation precursors, such as amines from liver and
processed meats and nitrates from fruits and vegetables, the antioxidants and other
protective compounds (e.g. polyphenols, isothiocyanates and vitamin C) present in fruits and
vegetables may no longer be sufficient to block the nitrosation reactions leading to
formation of carcinogenic NOCs in the intestines and bladder. In addition, heme – a
constituent of red meats including processed meat and liver – promotes endogenous
nitrosation. In feeding studies, the amount of NOCs found in feces is linearly correlated with
intake of heme from red meat (Cross, Pollock et al. 2003). Our data are consistent with this
111
hypothesis that heme intake from processed meats and liver was also a risk factor for
bladder cancer in our study, under conditions of concurrent high nitrate intake. In addition,
salami and liver typically contain the highest levels of secondary amines, with amounts on
the order of ten times those of bacon, ham, hot dogs and bologna (HernandezJover,
IzquierdoPulido et al. 1997). However, the amount of amines present in processed meats,
such as salami, is highly variable and dependent on a number of factors involved in the
fermenting process, including age of meat, temperature and strain of bacteria (Suzzi and
Gardini 2003). We see an interaction between high nitrate and liver intake, but although the
estimates are higher for salami among the high nitrate group, the interaction does not reach
significance. We speculate this could be due to varying amounts of amines present in
different salami meats. In addition, in our study salami intake was ascertained jointly with
pastrami and corned beef.
Previous studies have investigated the relationship between meat intake and risk of
bladder cancer, with inconclusive results. A positive association between diets high in red
meat intake and bladder cancer risk was reported in three cohort studies (Aune, De Stefani
et al. 2009, Key, Appleby et al. 2009, Ferrucci, Sinha et al. 2010), whereas other cohort
studies have shown no association (Michaud, Holick et al. 2006, Cross, Leitzmann et al.
2007, Larsson, Johansson et al. 2009, Jakszyn, Gonzalez et al. 2011). In the study by Ferruci
et al. (Ferrucci, Sinha et al. 2010), the observed positive association between red meat intake
and bladder cancer appeared to be attributable to consumption of processed red meat.
However, other studies did not find evidence of an association between intake of processed
meat and bladder cancer (Lumbreras, Garte et al. 2008, Larsson, Johansson et al. 2009). To
our knowledge, only one study has investigated the role of heme intake with bladder cancer
112
and found no association (Jakszyn, Gonzalez et al. 2011). Two cohort studies that
considered estimated levels of NOC intake in relation to bladder cancer risk have reported
null results (Zeegers, Selen et al. 2006, Ferrucci, Sinha et al. 2010), while one case-control
study conducted in Japan reported risk to be associated with increasing levels of nitrite
intake (Wilkens, Kadir et al. 1996). As far as we know, this present study is the only one that
has considered the role of processed meat in bladder cancer risk while taking into account
potential modifying effects of other sources of nitrosamine precursors, such as nitrate, nitrite
and heme intake. In contrast, several studies have investigated exposure to endogenous
nitrosation as risk factors for gastric and colon cancers, and NOCs formed in the stomach or
intestines often pass through the bladder before excretion, providing a route of exposure to
these carcinogens that deserves consideration (Tricker 1997, Vermeer, Pachen et al. 1998,
Ferrucci, Sinha et al. 2010). An interesting aspect of our findings was the increase of risk
found in non-smokers as compared to smokers. Smoking has been shown to induce
CYP450 enzymes, as well as a host of detoxifying enzymes that also play a role in the
metabolism of meat mutagens such as NOCs. Up regulation of these detoxifying enzymes
may lead to more efficient detoxification of NOCs in smokers when compared to non-
smokers. Alternatively, we speculate that those individuals most susceptible to DNA damage
and carcinogenesis at this site may have already developed bladder cancer in response to
exposure to the multitude of tobacco carcinogens, thus, among smokers, carcinogenic
exposures from NOCs had no further effect. No previous study has stratified results on
smoking status for comparison.
113
5.5 Variation in DNA Repair Enzymes, processed meat and
NOC intake, and the risk of bladder cancer
In the Los Angeles Bladder Cancer study we investigated the effect of genetic variation in
DNA repair enzymes on the association between processed meat intake, NOC intake, and
the risk of bladder cancer. We found that there was significant modification of the
association between salami, pastrami and corned beef intake and bladder cancer, by a SNP in
the RAD52 gene that participates in the HRR pathway, rs10774478 (Table 4.15). This
association remained significant even after Bonferroni adjustment for multiple testing. Also
of note, was a SNP in the ERCC5 gene that participates in the NER pathway, rs1571069.
This SNP modified the association of both salami/pastrami and corned beef intake and
nitrite intake and the risk of bladder cancer. Although, these associations did not remain
significant after correcting all SNPs tested, they were significant at the gene correction level,
i.e. correcting for the number of SNPs in the specific gene (Table 4.15).
rs10774478 is a polymorphism in RAD52, the gene coding for the RAD52 protein,
in the HRR DNA repair pathway. HRR repairs DSBs in DNA, such as those caused by ROS
that can arise as a result of exposure to both exogenous and endogenously formed NOCs
(Figure 1.6). RAD52 interacts with RAD51, an essential recombinase in HRR, and
participates in the regulation of its polymerization (Qing, Yamazoe et al. 2011). Thus,
variation in RAD52 may disrupt the DSB repair function of RAD51. In addition, RAD52
may also participate in the BER pathway as it has recently been shown to cooperate with
OGG1 to repair base damage and SSBs, again the type of DNA damage typical of oxidative
114
stress (de Souza-Pinto, Maynard et al. 2009). rs10774478 is an A/G SNP that has not
previously been reported in the literature as having any association with susceptibility to
carcinogenesis. This polymorphism is found in the promoter region of RAD52, thus we
hypothesize that this SNP, or a SNP in high linkage disequilibrium with this SNP, may play a
role in the regulation of RAD52 expression. Many SNPs in this region have been previously
reported as potential cancer susceptibility variants. The closest SNP to rs10774478 that has
been reported as having a clinical association is rs10849605, which is associated with the
antibody deficiency syndrome IgAG, a syndrome highly regulated by HRR (Offer, Pan-
Hammarstrom et al. 2010).
rs1571069 is an A/G SNP found in an intronic region of the ERCC5 gene. The
ERCC5 protein is part of the XPC-TFIIH protein complex, responsible for recognizing
DNA distortion caused by bulky adducts or alkylated bases, and targeting them for NER
(Gillet and Scharer 2006). This type of DNA damage can arise from exposure to
electrophilic metabolites of NOCs (Figure 1.6). Variation in genes coding for proteins
involved in the XPC-TFIIH complex may affect its ability to recognize this bulky damage.
rs1571069 has not previously been reported in the literature as being associated with any
type of carcinogenesis, nor has any SNP in the region of this polymorphism. We mention it
here as it was associated with modification of two sources of NOCs, salami, pastrami and
corned beef intake, and nitrite intake. Thus we hypothesize that variance in rs1571069, or in
the genetic region tagged by rs1571069, could be involved in susceptibility to damage caused
by endogenous NOC formation.
115
5.6 Fish intake and the risk of colorectal cancer
In the Colorectal Family Registry Study, we found an association between diets high in
shellfish and risk of CRC, particularly among those with identified familial cases of CRC.
Here familial cases are defined as those meeting Amsterdam criteria for familial history. We
did not find any evidence of association with any other type of fish and the risk of CRC;
moreover cooking methods of the fish did not seem to affect these associations.
This study found that those subjects in the highest quintile of shellfish consumption
had a 30% increased risk of developing CRC than those with no or very little shellfish
consumption (Table 4.18). Among familial cases, this increased risk was significantly higher;
subjects in the highest quintile of shellfish consumption were at 2 times the risk of
developing familial CRC than those in the lowest quintile. This risk was also higher when
restricted to cases of rectal cancer alone and when restricted to cases of the sigmoid colon,
suggesting that it is the distal portion of the colon most affected. It should be noted here
that shellfish consumption was very low in this population, thus we see this effect even at
these very low levels of intake. The average person in the highest quintile ate approximately
42g of shellfish per week, which is the equivalent of 2 small oysters or 15 medium shrimp
per week.
Shellfish intake has previously been identified as a plausible risk factor for CRC
(Manerio, Rodas et al. 2008). Shellfish naturally accumulate diarrhetic shellfish poisoning
(DSP) toxins when they feed on toxic microalgae. The most common DSP toxin found in
shellfish is okadaic acid (OA). OA has been shown to be carcinogenic in both rodent and
116
human cell lines (Jahan, Iijima et al. 1996). In animal models, exposure to OA has been
shown to alter the permeability of the intestinal wall, damage which can lead to tumor
promotion after repeat exposure (Tripuraneni, Koutsouris et al. 1997). This altered
permeability of the intestine is also responsible for DSP syndrome, a food-borne illness that
is associated with exposure to large amounts of DSP toxin, often found in contaminated
mussels, oysters, clams and scallops. For this reason, DSP toxin levels are monitored and
must not exceed more than 0.16µg OA per gram of shellfish (Manerio, Rodas et al. 2008).
Despite these regulations, outbreaks of shellfish poisoning do still occur, and furthermore,
the tumor promoting potential of OA may be occurring at much lower levels needed than
the threshold for gastrointestinal illness. In addition to OA, shellfish have also been reported
as containing a wide range of other chemicals, such as methyl mercury, polychlorinated
dibenzo-p-dioxins and organochlorine residues, some of which have been shown to be
mutagens or animal carcinogens (Lee, Shu et al. 2009). Not many large studies have
considered shellfish as a separate entity from total fish when investigating the associations
with colorectal cancer. Our results are consistent with a recent report from the Shanghai
Women’s Health Study that also found a significant association with shellfish intake and
CRC in women in Shanghai (Lee, Shu et al. 2009). An ecological study investigating DSP
levels in shellfish and cancer rates in men that ate high quantities of shellfish, found a
positive association with DSP toxin level and colon cancer in men only (Cordier, Monfort et
al. 2000).
The Second Export Report from the WCRF was updated in 2011 and concluded
that there was limited, but suggestive evidence that dietary fish intake protects against
colorectal cancer risk (Figure 1.9). This conclusion was based on the review of 35 cohort
117
studies and over 50 case-control studies. Approximately half of published cohort studies
have shown a decreased risk of CRC risk with high fish intake (WCRF/AICR 2011).Fish has
long been looked at as a potential preventative factor for CRC as it contains high levels of
beneficial nutrients such as vitamin D, selenium and omega-3 PUFAs. In particular, omega-3
PUFA supplementation has consistently been shown to have anti-neoplastic properties in
rodent models (Hull 2011). Despite these previous findings, we did not find any evidence
that fish intake decreased CRC in this study. Although fish contains many beneficial
compounds, fish is also susceptible to HCA and PAH accumulation when cooked at high
temperatures, and to NOC formation upon preservation, similar to red meat (IARC 1997).
When we looked at fish that was pan-fried, the method known to cause the most carcinogen
accumulation, we saw a slight increase in risk compared to grilled fish, however this was not
statistically significant. We hypothesize that the opposing effects of the previously
mentioned beneficial compounds, and carcinogen formation may confound the association
between fish intake and risk of CRC. Heterogeneity of both amount of omega-3 PUFA in
fish, and of cooking methods that lead to carcinogen formation, may account for the
inability of epidemiological studies to find a consistent decrease of CRC risk with high fish
intake.
118
5.7 Meat and fish intake, variation in GWAS top hits, and risk
of CRC
In the Colon CFR study, we found evidence that a polymorphism in the SMAD7 gene,
rs4939827, modified the association between processed white meat intake and the risk of
CRC. Among individuals homozygous for the wild-type allele (T/T) of this SNP, high intake
of processed white meat was associated with twice the risk of CRC, an association not seen
in carriers of the minor allele (C). We also found that an intergenic SNP, rs6691170,
modified the association of sausage intake and risk of CRC. Among individuals homozogous
for the minor allele (T/T) of this SNP, high sausage intake was associated with twice the risk
of CRC, an association not seen in carriers of the wild-type allele (G).
rs4939827 is one of three SNPs in the SMAD7 gene associated with risk
of colorectal cancer. The minor (C) allele is associated with a decreased risk of CRC; with
estimated ORs of 0.73 for (C/C) homozygotes and 0.86 for (C/T) heterozygotes (Broderick,
Carvajal-Carmona et al. 2007). SMAD7 is involved in the TGF-β signaling pathway, and
codes for an inhibitory SMAD protein that functions as a negative feedback regulator of
TGF-β. There is evidence that the over-expression of SMAD7 could promote tumorigenesis
via disturbing TGF-β-induced growth inhibition and apoptosis (Song, Zhu et al. 2012). It
has been demonstrated in animal models that TGF-β is up regulated upon the feeding of
nitrosamines, similar to those found in processed meats (Tessitore and Bollito 2006). Under
conditions of over-expression of SMAD7 and concurrent up regulation of TGF-β, such as
with high intake of processed meat, it is possible that the normal signaling pathway of TGF-
119
β could be disrupted and lead to tumor promotion. However, if this were the case, we would
expect to see evidence of this interaction with intake of red processed meat also, and we
found no such association in this study.
rs6691170 is also a marker for risk, specifically for rectal cancer (Lubbe, Whiffin et
al. 2012). However, it is difficult to hypothesize how the association seen here with sausage
intake might be interpreted biologically as this is an intergenic SNP and the functional basis
behind the observed cancer risk has yet to be elucidated (Lubbe, Whiffin et al. 2012).
5.8 Strengths and weaknesses of epidemiological studies
The four studies presented here all involved the use of administered food frequency
questionnaires to estimate dietary exposures. Unfortunately, there are inherent accuracy
problems in epidemiological studies whenever a food frequency questionnaire is used to
collect data. Misclassification may be present, particularly in the case-control studies, since
dietary exposure recalled at diagnosis may not accurately represent diet at time of cancer
formation, decades before. Moreover, recall bias might be present if cases tend to remember
certain exposures differently than controls, for example, a tendency might be observed for
cases to remember eating potentially harmful food items more often than controls in an
attempt to find explanations for their disease. However, given that at the time these studies
were conducted there was no widespread knowledge of the role of meat or fish intake and
120
the risk of cancer, any such recall bias is likely non-differential, and thus likely to bias results
towards the null.
Particular strengths of the CCPC study include the use of population-based cancer
registries for the ascertainment of cases, the oversampling of advanced cases, the inclusion
of three racial/ethnic groups, and the consideration of different types of fish (dark and
white) and various cooking methods and doneness levels. Weaknesses of this study are the
inclusion of a limited number of metabolic enzymes that do not comprehensively cover the
HCA and PAH metabolism pathways and the inclusion of selected polymorphisms instead
of a comprehensive tagSNP or sequencing approach. Moreover, we genotyped only four of
the 30 SNPs that accurately classify NAT2 haplotype (Hein, Grant et al. 2000). These four
SNPs have been shown to infer NAT2 phenotype with an estimated 98.4% accuracy (Hein
and Doll 2012); however, not genotyping every SNP may have misclassified some “slow”
phenotypes as “fast” (wild type), thereby biasing the results towards the null. For these
reasons, we cannot exclude the possibility that some of the genes we studied, for which we
found no association, may indeed play a role in PCA risk. Although the study had a large
sample size, small numbers in some subgroup analyses may explain our failure to find
interactions with the genes investigated. Also, dietary questionnaires have been reported to
result in an underestimation of HCA intake (Sinha, Cross et al. 2005), which could attenuate
gene-diet interactions towards the null. Additionally, we did not have information on the use
of marinades or use of cooking pre-treatments (microwaving, boiling), which can reduce the
amount of HCA formation (Salmon, Knize et al. 1997) and may further contribute to
exposure misclassification. Our analyses included advanced cases sampled from both LAC
121
and SFBA, but localized cases were from LAC only, as the SFBA center did not collect
blood for localized cases.
Strengths of the Singapore Chinese Health study include the use of a well-
characterized prospective cohort specially developed to investigate the role of diet, which
limits the effects of recall bias and differential misclassification. However, due to the age of
the cohort and the prevalence of PCA in Singapore our sample size was limited, particularly
when investigating subgroups and effect modification in our data. However, using a fully
validated questionnaire that was specifically developed for use in this population
strengthened our study.
Specific strengths of the Los Angeles Bladder Cancer study include a large
population based-sample, and availability of extensive data on multiple sources of
nitrosamine exposures, together with major established bladder cancer risk factors and
several additional potential confounders and effect modifiers. A potential weakness is that
we did not use a comprehensive food frequency questionnaire and were therefore unable to
estimate total red meat intake, which is highly correlated with endogenous nitrosation, and
heme iron intake. Instead, our questionnaire focused on processed meat intake and
estimating nitrate, nitrite and NOCs from dietary sources as well as Vitamin C and Vitamin
A. A further limitation is our inability to consider total caloric intake as a potential
confounder of the dietary factors investigated. However, we were able to use other surrogate
measures of total caloric intake, such as BMI and total food portions consumed. Another
limitation is that our data for the bladder study was restricted to the standard portion sizes
assumed in each questionnaire, which may reduce accuracy. Further limitation arose when
estimating dietary components such as heme, nitrate and nitrite levels. Heme data was
122
limited and estimated from processed meats and liver only. Heme content varies depending
on the type of meat. The processed meat categories investigated here could be comprised of
different meats, a standard heme-iron content of 40% iron was assumed. This
standardization potentially reduces our accuracy. In addition, nitrate and nitrite estimations
were limited to items asked about in the questionnaire, which disregards potentially
important exposures from other sources such as nitrate in drinking water.
Particular strengths of the Colon CFR are a large sample size of cases and controls,
enriched for familial cases and thus those with strong possible genetic factors. The Colon
CFR FFQ was very comprehensive and included a detailed cooking module; however, this
FFQ was not used across all centers, so our analyses were restricted to only those three
centers that completed this questionnaire. We were also restricted, as information on type of
fish was not asked about in the cooking module, so we could not make inferences about the
levels of nutrients that could be interacting with the cooking methods.
5.9 Summary of Studies
Overall the findings of these studies support an important role for dietary factors in the risk
of three common cancers; prostate, bladder and colorectal cancer. These studies also
support a role for common variants in metabolic and DNA repair pathways in modifying the
way that these dietary factors will lead to carcinogenesis. Moreover, the findings of these
gene-by-diet interactions strengthen the hypothesis that the carcinogens present in these
123
dietary factors might be responsible for the observed associations. There is evidence to
suggest that HCA and PAH metabolism is involved in prostate tumor formation. The CPCC
Study reports associations between functional polymorphisms in three enzymes involved in
HCA and PAH metabolism and risk of localized PCA. Furthermore, we also report that the
association between diets high in fish and advanced PCA may be modified by a
polymorphism that may affect activity of the PTGS2 enzyme, which is known to be involved
in HCA and PAH metabolism as well as metabolism of polyunsaturated fatty acids found in
fish. Overall, our finding of an interaction with PTGS2 supports a role for HCAs and PAHs,
which accumulate in well-done fish, and/or fatty acid changes that occur when fish is
cooked at high temperature in PCA carcinogenesis.
Overall, our findings in the Singapore Chinese Health Study support an association
between preserved meat intake and risk of advanced stage PCA. This association seems
more relevant among men less able to detoxify carcinogens, such as those with an increased
BMI or those with low levels of fruit and vegetable intake. Our findings come from a
population where intake of preserved meats has traditionally been very low and suggest that
increase in preserved meat intake may account for some of the increase in PCA risk seen
with adoption of a Westernized diet. We also report an increase in advanced stage PCA risk
with intake of poultry with skin. Although numbers are small, these findings may warrant
future studies to consider the role of cooked poultry, taking into account preparation
methods, in PCA risk.
We found intake of liver and of salami/pastrami/corned beef to be associated with
risk of bladder cancer. In particular, the effects of consuming these meats may be greater
when accompanied by overall high nitrate intake. Results from the Los Angeles Bladder
124
Cancer study are consistent with a role of dietary sources of NOC precursors from
processed meats in bladder cancer risk, suggesting consumption of meats with high amine
and heme content such as salami and liver as a risk factor for bladder cancer. Our findings
support a role for endogenous nitrosation as a potential risk factor for bladder cancer,
particularly among non-smokers.
In the Colon CFR study we found that high intake of shellfish was associated with
increased risk of CRC. In particular, these effects were stronger among familial cases of
CRC. These findings suggest that consumption of shellfish with high levels of DSP toxins
may contribute to CRC progression, and supports previous evidence that regulation on DSP
toxin levels in shellfish should be re-evaluated. In addition, in the Colon CFR, we found that
a SMAD7 SNP involved in the TGF-β signaling pathway may modify the association of
dietary carcinogens with risk of CRC.
5.10 Overall Summary and Final Conclusions
This thesis has investigated the role of three key carcinogens that occur naturally in the
human diet, HCAs, PAHs and NOCs, and the risk of carcinogenesis. HCAs and PAHs are
known to form when meats and fish are cooked at high temperatures, and their effects have
been much studied in the risk of gastrointestinal cancers. Consistent with the hypothesis of
this dissertation, we find here evidence of a role for HCA and PAH metabolism genes in the
role of prostate cancer in a California based study. Moreover, we find that a gene important
125
for metabolism of these carcinogens modifies the effect of fish intake on the risk of prostate
cancer, adding further support to the hypothesis. HCAs and PAHs can enter the
bloodstream from the gastrointestinal tract and thus provide a route of exposure for other
organs. These findings suggest that the role of HCA and PAHs may be important in
tumorigenesis of the prostate, and perhaps a variety of cancers other than those in the
gastrointestinal tract, as HCAs and PAHs can move readily through the bloodstream.
However, contrary to these findings in California, and to the stated hypothesis, we did not
find evidence of HCA or PAH involvement when investigating the same exposures in a
Singapore based cohort.
NOCs have also been extensively studied in cancers of the stomach and intestine.
Processed meats are known to be sources of both exogenous and endogenous NOC
formation. Studies presented here, support the proposed hypothesis of a role of preserved
meat intake and NOC exposure in the risk of two further cancers, bladder and prostate.
These findings may indicate a role of endogenous NOC formation in the risk of these
cancers. Overall this thesis provides evidence and lends support to recommendations from
the WCRF to limit red meat intake, especially cooked at high temperature methods, and to
limit processed meat intake.
An unforeseen finding of this research, that was additional to the hypotheses
proposed, was the findings here of an association with high shellfish intake and risk of
colorectal cancer. Shellfish intake has not been extensively looked at as a cancer risk factor,
perhaps due to relatively low levels of consumption in the Western diet. These findings
require replication, but may support a re-evaluation of the level of shellfish toxins such as
DSP toxins that are deemed safe for human consumption.
126
Studies presented here have combined dietary information with genetic factors to
gain insight into metabolic and DNA repair pathways that may be important in
carcinogenesis. These gene-by-diet interactions are valuable. However, when we have
information on a number of polymorphisms, there still exist limitations in how we can
extract meaningful data after performing such a large number of tests. Future work in this
area involves elucidating how to best make use of the wealth of information now at our
fingertips, and in developing new and better analyses to deal with the large amounts of data
available. Furthermore gene-by-diet analysis and pathway-by-diet analysis have the potential
to identify reliable biological markers for exposure to dietary carcinogens. This would reduce
our current dependence on food frequency questionnaires and increase accuracy of
epidemiological dietary studies. Moving forward it is clear that investigating the role of
genetic variation in the metabolism of dietary carcinogens can provide valuable insight into
the biological mechanisms of diet related carcinogenesis.
127
Bibliography
ACS. (2012). "American Cancer Society." Cancer Facts and Figures.
Aeschbacher, H. U. and E. Ruch (1989). "Effect of heterocyclic amines and beef extract on
chromosome aberrations and sister chromatid exchanges in cultured human lymphocytes."
Carcinogenesis 10(3): 429-433.
Ahotupa, M., V. Bussacchini-Griot, J. C. Bereziat, A. M. Camus and H. Bartsch (1987). "Rapid
oxidative stress induced by N-nitrosamines." Biochem Biophys Res Commun 146(3): 1047-1054.
Alexander, D. D., P. J. Mink, C. A. Cushing and B. Sceurman (2010). "A review and meta-analysis of
prospective studies of red and processed meat intake and prostate cancer." Nutr J 9: 50.
Allen, N. E., T. J. Key, P. N. Appleby, R. C. Travis, A. W. Roddam, A. Tjonneland, N. F. Johnsen,
K. Overvad, J. Linseisen, S. Rohrmann, H. Boeing, T. Pischon, H. B. Bueno-de-Mesquita, L.
Kiemeney, G. Tagliabue, D. Palli, P. Vineis, R. Tumino, A. Trichopoulou, C. Kassapa, D.
Trichopoulos, E. Ardanaz, N. Larranaga, M. J. Tormo, C. A. Gonzalez, J. R. Quiros, M. J. Sanchez,
S. Bingham, K. T. Khaw, J. Manjer, G. Berglund, P. Stattin, G. Hallmans, N. Slimani, P. Ferrari, S.
Rinaldi and E. Riboli (2008). "Animal foods, protein, calcium and prostate cancer risk: the European
Prospective Investigation into Cancer and Nutrition." Br J Cancer 98(9): 1574-1581.
Allen, N. E., C. Sauvaget, A. W. Roddam, P. Appleby, J. Nagano, G. Suzuki, T. J. Key and K.
Koyama (2004). "A prospective study of diet and prostate cancer in Japanese men." Cancer Causes
Control 15(9): 911-920.
Astorg, P. (2010). ω-3 PUFAs and Prostate Cancer: Epidemiological Studies. Dietary Omega-3
Polyunsaturated Fatty Acids and Cancer. G. Calviello and S. Serini, Springer Netherlands. 1: 109-148.
Augustsson, K., D. S. Michaud, E. B. Rimm, M. F. Leitzmann, M. J. Stampfer, W. C. Willett and E.
Giovannucci (2003). "A prospective study of intake of fish and marine fatty acids and prostate
cancer." Cancer Epidemiol Biomarkers Prev 12(1): 64-67.
Aune, D., E. De Stefani, A. Ronco, P. Boffetta, H. Deneo-Pellegrini, G. Acosta and M. Mendilaharsu
(2009). "Meat consumption and cancer risk: a case-control study in Uruguay." Asian Pac J Cancer
Prev 10(3): 429-436.
128
Bartsch, H., H. Ohshima and B. Pignatelli (1988). "Inhibitors of endogenous nitrosation.
Mechanisms and implications in human cancer prevention." Mutat Res 202(2): 307-324.
Bartsch, H., H. Ohshima, B. Pignatelli and S. Calmels (1989). "Human exposure to endogenous N-
nitroso compounds: quantitative estimates in subjects at high risk for cancer of the oral cavity,
oesophagus, stomach and urinary bladder." Cancer Surv 8(2): 335-362.
Beuten, J., J. A. Gelfond, J. J. Byrne, I. Balic, A. C. Crandall, T. L. Johnson-Pais, I. M. Thompson, D.
K. Price and R. J. Leach (2008). "CYP1B1 variants are associated with prostate cancer in non-
Hispanic and Hispanic Caucasians." Carcinogenesis 29(9): 1751-1757.
Bock, K. W., H. Gschaidmeier, H. Heel, T. Lehmkoster, P. A. Munzel and B. S. Bock-Hennig (1999).
"Functions and transcriptional regulation of PAH-inducible human UDP-glucuronosyltransferases."
Drug Metab Rev 31(2): 411-422.
Boobis, A. R., A. M. Lynch, S. Murray, R. de la Torre, A. Solans, M. Farre, J. Segura, N. J.
Gooderham and D. S. Davies (1994). "CYP1A2-catalyzed conversion of dietary heterocyclic amines
to their proximate carcinogens is their major route of metabolism in humans." Cancer Res 54(1): 89-
94.
Borek, C. (2004). "Dietary antioxidants and human cancer." Integr Cancer Ther 3(4): 333-341.
Brennan, P., O. Bogillot, S. Cordier, E. Greiser, W. Schill, P. Vineis, G. Lopez-Abente, A. Tzonou, J.
Chang-Claude, U. Bolm-Audorff, K. H. Jockel, F. Donato, C. Serra, J. Wahrendorf, M. Hours, A.
T'Mannetje, M. Kogevinas and P. Boffetta (2000). "Cigarette smoking and bladder cancer in men: a
pooled analysis of 11 case-control studies." Int J Cancer 86(2): 289-294.
Brinkman, M. and M. P. Zeegers (2008). "Nutrition, total fluid and bladder cancer." Scand J Urol
Nephrol Suppl(218): 25-36.
Broderick, P., L. Carvajal-Carmona, A. M. Pittman, E. Webb, K. Howarth, A. Rowan, S. Lubbe, S.
Spain, K. Sullivan, S. Fielding, E. Jaeger, J. Vijayakrishnan, Z. Kemp, M. Gorman, I. Chandler, E.
Papaemmanuil, S. Penegar, W. Wood, G. Sellick, M. Qureshi, A. Teixeira, E. Domingo, E. Barclay, L.
Martin, O. Sieber, D. Kerr, R. Gray, J. Peto, J. B. Cazier, I. Tomlinson and R. S. Houlston (2007). "A
genome-wide association study shows that common alleles of SMAD7 influence colorectal cancer
risk." Nat Genet 39(11): 1315-1317.
Butler, L. M., R. Sinha, R. C. Millikan, C. F. Martin, B. Newman, M. D. Gammon, A. S. Ammerman
and R. S. Sandler (2003). "Heterocyclic amines, meat intake, and association with colon cancer in a
population-based study." Am J Epidemiol 157(5): 434-445.
129
Butler, L. M., A. S. Wong, W. P. Koh, R. Wang, J. M. Yuan and M. C. Yu (2010). "Calcium intake
increases risk of prostate cancer among Singapore Chinese." Cancer Res 70(12): 4941-4948.
Caldecott, K. W. (2003). "Protein-protein interactions during mammalian DNA single-strand break
repair." Biochem Soc Trans 31(Pt 1): 247-251.
Cao, Y. and J. Ma (2011). "Body mass index, prostate cancer-specific mortality, and biochemical
recurrence: a systematic review and meta-analysis." Cancer Prev Res (Phila) 4(4): 486-501.
Carnell, D. M., R. E. Smith, F. M. Daley, P. R. Barber, P. J. Hoskin, G. D. Wilson, G. I. Murray and
S. A. Everett (2004). "Target validation of cytochrome P450 CYP1B1 in prostate carcinoma with
protein expression in associated hyperplastic and premalignant tissue." Int J Radiat Oncol Biol Phys
58(2): 500-509.
Castelao, J. E., J. M. Yuan, M. Gago-Dominguez, P. L. Skipper, S. R. Tannenbaum, K. K. Chan, M.
A. Watson, D. A. Bell, G. A. Coetzee, R. K. Ross and M. C. Yu (2004). "Carotenoids/vitamin C and
smoking-related bladder cancer." Int J Cancer 110(3): 417-423.
Castelao, J. E., J. M. Yuan, P. L. Skipper, S. R. Tannenbaum, M. Gago-Dominguez, J. S. Crowder, R.
K. Ross and M. C. Yu (2001). "Gender- and smoking-related bladder cancer risk." J Natl Cancer Inst
93(7): 538-545.
Catsburg, C., A. D. Joshi, R. Corral, J. P. Lewinger, J. Koo, E. M. John, S. A. Ingles and M. C. Stern
(2012). "Polymorphisms in carcinogen metabolism enzymes, fish intake, and risk of prostate cancer."
Carcinogenesis 33(7): 1352-1359.
Center, M. M., A. Jemal, R. A. Smith and E. Ward (2009). "Worldwide variations in colorectal
cancer." CA Cancer J Clin 59(6): 366-378.
Chan, D. S., R. Lau, D. Aune, R. Vieira, D. C. Greenwood, E. Kampman and T. Norat (2011). "Red
and processed meat and colorectal cancer incidence: meta-analysis of prospective studies." PLoS One
6(6): e20456.
Chavarro, J. E., M. J. Stampfer, M. N. Hall, H. D. Sesso and J. Ma (2008). "A 22-y prospective study
of fish intake in relation to prostate cancer incidence and mortality." Am J Clin Nutr 88(5): 1297-
1303.
Cordier, S., C. Monfort, L. Miossec, S. Richardson and C. Belin (2000). "Ecological analysis of
digestive cancer mortality related to contamination by diarrhetic shellfish poisoning toxins along the
coasts of France." Environ Res 84(2): 145-150.
130
Cross, A. J., L. M. Ferrucci, A. Risch, B. I. Graubard, M. H. Ward, Y. Park, A. R. Hollenbeck, A.
Schatzkin and R. Sinha (2010). "A large prospective study of meat consumption and colorectal cancer
risk: an investigation of potential mechanisms underlying this association." Cancer Res 70(6): 2406-
2414.
Cross, A. J., M. F. Leitzmann, M. H. Gail, A. R. Hollenbeck, A. Schatzkin and R. Sinha (2007). "A
prospective study of red and processed meat intake in relation to cancer risk." PLoS Med 4(12): e325.
Cross, A. J., U. Peters, V. A. Kirsh, G. L. Andriole, D. Reding, R. B. Hayes and R. Sinha (2005). "A
prospective study of meat and meat mutagens and prostate cancer risk." Cancer Res 65(24): 11779-
11784.
Cross, A. J., J. R. Pollock and S. A. Bingham (2003). "Haem, not protein or inorganic iron, is
responsible for endogenous intestinal N-nitrosation arising from red meat." Cancer Res 63(10): 2358-
2360.
Cross, A. J. and R. Sinha (2004). "Meat-related mutagens/carcinogens in the etiology of colorectal
cancer." Environ Mol Mutagen 44(1): 44-55.
Cui, J. G., S. Salehi-Rad, E. Rogaeva and W. J. Lukiw (2005). "Characterization of a cyclooxygenase-
2-765G-->C promoter polymorphism in human neural cells." Neuroreport 16(6): 575-579.
D'Amours, D. and S. P. Jackson (2002). "The Mre11 complex: at the crossroads of dna repair and
checkpoint signalling." Nat Rev Mol Cell Biol 3(5): 317-327.
de Murcia, G., V. Schreiber, M. Molinete, B. Saulier, O. Poch, M. Masson, C. Niedergang and J.
Menissier de Murcia (1994). "Structure and function of poly(ADP-ribose) polymerase." Mol Cell
Biochem 138(1-2): 15-24.
de Souza-Pinto, N. C., S. Maynard, K. Hashiguchi, J. Hu, M. Muftuoglu and V. A. Bohr (2009). "The
recombination protein RAD52 cooperates with the excision repair protein OGG1 for the repair of
oxidative lesions in mammalian cells." Mol Cell Biol 29(16): 4441-4454.
Di Paolo, O. A., C. H. Teitel, S. Nowell, B. F. Coles and F. F. Kadlubar (2005). "Expression of
cytochromes P450 and glutathione S-transferases in human prostate, and the potential for activation
of heterocyclic amine carcinogens via acetyl-coA-, PAPS- and ATP-dependent pathways." Int J
Cancer 117(1): 8-13.
131
Discacciati, A., N. Orsini and A. Wolk (2012). "Body mass index and incidence of localized and
advanced prostate cancer--a dose-response meta-analysis of prospective studies." Ann Oncol 23(7):
1665-1671.
Edlund, C. K., W. H. Lee, D. Li, D. J. Van Den Berg and D. V. Conti (2008). "Snagger: a user-
friendly program for incorporating additional information for tagSNP selection." BMC
Bioinformatics 9: 174.
Felton, J. S. and M. G. Knize (1991). "Occurrence, identification, and bacterial mutagenicity of
heterocyclic amines in cooked food." Mutat Res 259(3-4): 205-217.
Ferguson, L. R. and M. Philpott (2007). "Cancer prevention by dietary bioactive components that
target the immune response." Curr Cancer Drug Targets 7(5): 459-464.
Ferguson, L. R., M. Philpott and N. Karunasinghe (2004). "Dietary cancer and prevention using
antimutagens." Toxicology 198(1-3): 147-159.
Ferrucci, L. M., R. Sinha, M. H. Ward, B. I. Graubard, A. R. Hollenbeck, B. A. Kilfoy, A. Schatzkin,
D. S. Michaud and A. J. Cross (2010). "Meat and components of meat and the risk of bladder cancer
in the NIH-AARP Diet and Health Study." Cancer.
Figer, A., T. Friedman, A. E. Manguoglu, D. Flex, A. Vazina, I. Novikov, A. Shtrieker, A. A. Sidi, T.
Tichler, E. E. Sapir, J. Baniel and E. Friedman (2003). "Analysis of polymorphic patterns in candidate
genes in Israeli patients with prostate cancer." Isr Med Assoc J 5(10): 741-745.
Floyd, R. A. (1990). "The role of 8-hydroxyguanine in carcinogenesis." Carcinogenesis 11(9): 1447-
1450.
Fradet, V., I. Cheng, G. Casey and J. S. Witte (2009). "Dietary omega-3 fatty acids, cyclooxygenase-2
genetic variation, and aggressive prostate cancer risk." Clin Cancer Res 15(7): 2559-2566.
Fukatsu, T., Y. Hirokawa, T. Araki, T. Hioki, T. Murata, H. Suzuki, T. Ichikawa, H. Tsukino, D. Qiu,
T. Katoh, Y. Sugimura, R. Yatani, T. Shiraishi and M. Watanabe (2004). "Genetic polymorphisms of
hormone-related genes and prostate cancer risk in the Japanese population." Anticancer Res 24(4):
2431-2437.
Gauderman, W. J. (2002). "Sample size requirements for association studies of gene-gene
interaction." Am J Epidemiol 155(5): 478-484.
132
Gichner, T. and J. Veleminsky (1988). "Mechanisms of inhibition of N-nitroso compounds-induced
mutagenicity." Mutat Res 202(2): 325-334.
Gillet, L. C. and O. D. Scharer (2006). "Molecular mechanisms of mammalian global genome
nucleotide excision repair." Chem Rev 106(2): 253-276.
Guttenplan, J. B. (1987). "N-nitrosamines: bacterial mutagenesis and in vitro metabolism." Mutat Res
186(2): 81-134.
Hankin, J. H., D. O. Stram, K. Arakawa, S. Park, S. H. Low, H. P. Lee and M. C. Yu (2001).
"Singapore Chinese Health Study: development, validation, and calibration of the quantitative food
frequency questionnaire." Nutr Cancer 39(2): 187-195.
Hassett, C., J. Lin, C. L. Carty, E. M. Laurenzana and C. J. Omiecinski (1997). "Human hepatic
microsomal epoxide hydrolase: comparative analysis of polymorphic expression." Arch Biochem
Biophys 337(2): 275-283.
Hecht, S. S. and D. Hoffman (1998). "N-nitroso compounds and man: sources of exposure,
endogenous formation and occurrence in body fluids." Eur J Cancer Prev 7(3): 244-246.
Hedelin, M., E. T. Chang, F. Wiklund, R. Bellocco, A. Klint, J. Adolfsson, K. Shahedi, J. Xu, H. O.
Adami, H. Gronberg and K. A. Balter (2007). "Association of frequent consumption of fatty fish
with prostate cancer risk is modified by COX-2 polymorphism." Int J Cancer 120(2): 398-405.
Hein, D. W. and M. A. Doll (2012). "Accuracy of various human NAT2 SNP genotyping panels to
infer rapid, intermediate and slow acetylator phenotypes." Pharmacogenomics 13(1): 31-41.
Hein, D. W., D. M. Grant and E. Sim (2000). "Update on consensus arylamine N-acetyltransferase
gene nomenclature." Pharmacogenetics 10(4): 291-292.
HernandezJover, T., M. IzquierdoPulido, M. T. VecianaNogues, A. MarineFont and M. C.
VidalCarou (1997). "Biogenic amine and polyamine contents in meat and meat products." Journal of
Agricultural and Food Chemistry 45(6): 2098-2102.
Hindorff, M. J. E. B. I., Morales J (European Bioinformatics Institute), Junkins HA, Hall PN, Klemm
AK, and Manolio TA (2013). "A Catalog of Published Genome-Wide Association Studies. Available
at: http://www.genome.gov/gwastudies. Accessed 2013.".
133
Honma, M., M. Izumi, M. Sakuraba, S. Tadokoro, H. Sakamoto, W. Wang, F. Yatagai and M.
Hayashi (2003). "Deletion, rearrangement, and gene conversion; genetic consequences of
chromosomal double-strand breaks in human cells." Environ Mol Mutagen 42(4): 288-298.
Hsing, A. W., L. Tsao and S. S. Devesa (2000). "International trends and patterns of prostate cancer
incidence and mortality." Int J Cancer 85(1): 60-67.
Huang, M., S. Narita, K. Numakura, H. Tsuruta, M. Saito, T. Inoue, Y. Horikawa, N. Tsuchiya and
T. Habuchi (2012). "A high-fat diet enhances proliferation of prostate cancer cells and activates
MCP-1/CCR2 signaling." Prostate 72(16): 1779-1788.
Hull, M. A. (2011). "Omega-3 polyunsaturated fatty acids." Best Pract Res Clin Gastroenterol 25(4-
5): 547-554.
Hull, M. A. (2013). "Nutritional agents with anti-inflammatory properties in chemoprevention of
colorectal neoplasia." Recent Results Cancer Res 191: 143-156.
IARC (1997). International Agency for Research on Cancer. Some Naturally Occurring Substances:
Food Items and Constituents, Heterocyclic Aromatic Amines and Mycotoxins, vol56.
Jahan, I., M. Iijima, T. Kondo and M. Namba (1996). "Effects of okadaic acid on cell growth,
anchorage-independent growth, and co-cultures of normal (KMS-6), immortalized (KMST-6), and
neoplastically transformed (KMST-6T and KMST-6/RAS) human fibroblasts." J Cancer Res Clin
Oncol 122(4): 249-253.
Jakszyn, P., S. Bingham, G. Pera, A. Agudo, R. Luben, A. Welch, H. Boeing, G. Del Giudice, D.
Palli, C. Saieva, V. Krogh, C. Sacerdote, R. Tumino, S. Panico, G. Berglund, H. Siman, G. Hallmans,
M. J. Sanchez, N. Larranaga, A. Barricarte, M. D. Chirlaque, J. R. Quiros, T. J. Key, N. Allen, E.
Lund, F. Carneiro, J. Linseisen, G. Nagel, K. Overvad, A. Tjonneland, A. Olsen, H. B. Bueno-de-
Mesquita, M. O. Ocke, P. H. Peeters, M. E. Numans, F. Clavel-Chapelon, A. Trichopoulou, C.
Fenger, R. Stenling, P. Ferrari, M. Jenab, T. Norat, E. Riboli and C. A. Gonzalez (2006).
"Endogenous versus exogenous exposure to N-nitroso compounds and gastric cancer risk in the
European Prospective Investigation into Cancer and Nutrition (EPIC-EURGAST) study."
Carcinogenesis 27(7): 1497-1501.
Jakszyn, P., C. A. Gonzalez, L. Lujan-Barroso, M. M. Ros, H. B. Bueno-de-Mesquita, N. Roswall, A.
M. Tjonneland, F. L. Buchner, L. Egevad, K. Overvad, O. Raaschou-Nielsen, F. Clavel-Chapelon, M.
C. Boutron-Ruault, M. S. Touillaud, J. Chang-Claude, N. E. Allen, L. A. Kiemeney, T. J. Key, R.
Kaaks, H. Boeing, S. Weikert, A. Trichopoulou, E. Oikonomou, D. Zylis, D. Palli, F. Berrino, P.
Vineis, R. Tumino, A. Mattiello, P. H. Peeters, C. L. Parr, I. T. Gram, G. Skeie, M. J. Sanchez, N.
Larranaga, E. Ardanaz, C. Navarro, L. Rodriguez, D. Ulmert, R. Ehrnstrom, G. Hallmans, B.
134
Ljungberg, A. W. Roddam, S. A. Bingham, K. T. Khaw, N. Slimani, P. A. Boffetta, M. Jenab, T.
Mouw, D. S. Michaud and E. Riboli (2011). "Red meat, dietary nitrosamines, and heme iron and risk
of bladder cancer in the European Prospective Investigation into Cancer and Nutrition (EPIC)."
Cancer Epidemiol Biomarkers Prev 20(3): 555-559.
Joenje, H. (1989). "Genetic toxicology of oxygen." Mutat Res 219(4): 193-208.
Johansson, M. A., L. Fredholm, I. Bjerne and M. Jagerstad (1995). "Influence of frying fat on the
formation of heterocyclic amines in fried beefburgers and pan residues." Food Chem Toxicol 33(12):
993-1004.
John, E. M., G. G. Schwartz, J. Koo, D. Van Den Berg and S. A. Ingles (2005). "Sun exposure,
vitamin D receptor gene polymorphisms, and risk of advanced prostate cancer." Cancer Res 65(12):
5470-5479.
John, E. M., M. C. Stern, R. Sinha and J. Koo (2011). "Meat consumption, cooking practices, meat
mutagens, and risk of prostate cancer." Nutr Cancer 63(4): 525-537.
Johnson, I. T. (2002). "Anticarcinogenic effects of diet-related apoptosis in the colorectal mucosa."
Food Chem Toxicol 40(8): 1171-1178.
Joshi, A. D., E. M. John, J. Koo, S. A. Ingles and M. C. Stern (2012 ). "Fish intake, cooking practices,
and risk of prostate cancer: results from a multi-ethnic case-control study." Cancer causes & control :
CCC 23(4): 405-420.
Karanjawala, Z. E., N. Adachi, R. A. Irvine, E. K. Oh, D. Shibata, K. Schwarz, C. L. Hsieh and M. R.
Lieber (2002). "The embryonic lethality in DNA ligase IV-deficient mice is rescued by deletion of
Ku: implications for unifying the heterogeneous phenotypes of NHEJ mutants." DNA Repair
(Amst) 1(12): 1017-1026.
Kazerouni, N., R. Sinha, C. H. Hsu, A. Greenberg and N. Rothman (2001). "Analysis of 200 food
items for benzo[a]pyrene and estimation of its intake in an epidemiologic study." Food Chem
Toxicol 39(5): 423-436.
Key, T. J., P. N. Appleby, E. A. Spencer, R. C. Travis, N. E. Allen, M. Thorogood and J. I. Mann
(2009). "Cancer incidence in British vegetarians." Br J Cancer 101(1): 192-197.
Kiemeney, L. A., S. Thorlacius, P. Sulem, F. Geller, K. K. Aben, S. N. Stacey, J. Gudmundsson, M.
Jakobsdottir, J. T. Bergthorsson, A. Sigurdsson, T. Blondal, J. A. Witjes, S. H. Vermeulen, C. A.
Hulsbergen-van de Kaa, D. W. Swinkels, M. Ploeg, E. B. Cornel, H. Vergunst, T. E. Thorgeirsson,
135
D. Gudbjartsson, S. A. Gudjonsson, G. Thorleifsson, K. T. Kristinsson, M. Mouy, S. Snorradottir,
D. Placidi, M. Campagna, C. Arici, K. Koppova, E. Gurzau, P. Rudnai, E. Kellen, S. Polidoro, S.
Guarrera, C. Sacerdote, M. Sanchez, B. Saez, G. Valdivia, C. Ryk, P. de Verdier, A. Lindblom, K.
Golka, D. T. Bishop, M. A. Knowles, S. Nikulasson, V. Petursdottir, E. Jonsson, G. Geirsson, B.
Kristjansson, J. I. Mayordomo, G. Steineck, S. Porru, F. Buntinx, M. P. Zeegers, T. Fletcher, R.
Kumar, G. Matullo, P. Vineis, A. E. Kiltie, J. R. Gulcher, U. Thorsteinsdottir, A. Kong, T. Rafnar
and K. Stefansson (2008). "Sequence variant on 8q24 confers susceptibility to urinary bladder
cancer." Nat Genet 40(11): 1307-1312.
Kirschenbaum, A., A. P. Klausner, R. Lee, P. Unger, S. Yao, X. H. Liu and A. C. Levine (2000).
"Expression of cyclooxygenase-1 and cyclooxygenase-2 in the human prostate." Urology 56(4): 671-
676.
Knize, M. G., K. S. Kulp, C. P. Salmon, G. A. Keating and J. S. Felton (2002). "Factors affecting
human heterocyclic amine intake and the metabolism of PhIP." Mutat Res 506-507: 153-162.
Knize, M. G., C. P. Salmon, P. Pais and J. S. Felton (1999). "Food heating and the formation of
heterocyclic aromatic amine and polycyclic aromatic hydrocarbon mutagens/carcinogens." Adv Exp
Med Biol 459: 179-193.
Kulp, K. S., M. G. Knize, M. A. Malfatti, C. P. Salmon and J. S. Felton (2000). "Identification of
urine metabolites of 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine following consumption of a
single cooked chicken meal in humans." Carcinogenesis 21(11): 2065-2072.
Lander, E. S., L. M. Linton, B. Birren, C. Nusbaum, M. C. Zody, J. Baldwin, K. Devon, K. Dewar,
M. Doyle, W. FitzHugh, R. Funke, D. Gage, K. Harris, A. Heaford, J. Howland, L. Kann, J.
Lehoczky, R. LeVine, P. McEwan, K. McKernan, J. Meldrim, J. P. Mesirov, C. Miranda, W. Morris, J.
Naylor, C. Raymond, M. Rosetti, R. Santos, A. Sheridan, C. Sougnez, N. Stange-Thomann, N.
Stojanovic, A. Subramanian, D. Wyman, J. Rogers, J. Sulston, R. Ainscough, S. Beck, D. Bentley, J.
Burton, C. Clee, N. Carter, A. Coulson, R. Deadman, P. Deloukas, A. Dunham, I. Dunham, R.
Durbin, L. French, D. Grafham, S. Gregory, T. Hubbard, S. Humphray, A. Hunt, M. Jones, C. Lloyd,
A. McMurray, L. Matthews, S. Mercer, S. Milne, J. C. Mullikin, A. Mungall, R. Plumb, M. Ross, R.
Shownkeen, S. Sims, R. H. Waterston, R. K. Wilson, L. W. Hillier, J. D. McPherson, M. A. Marra, E.
R. Mardis, L. A. Fulton, A. T. Chinwalla, K. H. Pepin, W. R. Gish, S. L. Chissoe, M. C. Wendl, K. D.
Delehaunty, T. L. Miner, A. Delehaunty, J. B. Kramer, L. L. Cook, R. S. Fulton, D. L. Johnson, P. J.
Minx, S. W. Clifton, T. Hawkins, E. Branscomb, P. Predki, P. Richardson, S. Wenning, T. Slezak, N.
Doggett, J. F. Cheng, A. Olsen, S. Lucas, C. Elkin, E. Uberbacher, M. Frazier, R. A. Gibbs, D. M.
Muzny, S. E. Scherer, J. B. Bouck, E. J. Sodergren, K. C. Worley, C. M. Rives, J. H. Gorrell, M. L.
Metzker, S. L. Naylor, R. S. Kucherlapati, D. L. Nelson, G. M. Weinstock, Y. Sakaki, A. Fujiyama, M.
Hattori, T. Yada, A. Toyoda, T. Itoh, C. Kawagoe, H. Watanabe, Y. Totoki, T. Taylor, J.
Weissenbach, R. Heilig, W. Saurin, F. Artiguenave, P. Brottier, T. Bruls, E. Pelletier, C. Robert, P.
Wincker, D. R. Smith, L. Doucette-Stamm, M. Rubenfield, K. Weinstock, H. M. Lee, J. Dubois, A.
Rosenthal, M. Platzer, G. Nyakatura, S. Taudien, A. Rump, H. Yang, J. Yu, J. Wang, G. Huang, J.
Gu, L. Hood, L. Rowen, A. Madan, S. Qin, R. W. Davis, N. A. Federspiel, A. P. Abola, M. J. Proctor,
136
R. M. Myers, J. Schmutz, M. Dickson, J. Grimwood, D. R. Cox, M. V. Olson, R. Kaul, N. Shimizu,
K. Kawasaki, S. Minoshima, G. A. Evans, M. Athanasiou, R. Schultz, B. A. Roe, F. Chen, H. Pan, J.
Ramser, H. Lehrach, R. Reinhardt, W. R. McCombie, M. de la Bastide, N. Dedhia, H. Blocker, K.
Hornischer, G. Nordsiek, R. Agarwala, L. Aravind, J. A. Bailey, A. Bateman, S. Batzoglou, E. Birney,
P. Bork, D. G. Brown, C. B. Burge, L. Cerutti, H. C. Chen, D. Church, M. Clamp, R. R. Copley, T.
Doerks, S. R. Eddy, E. E. Eichler, T. S. Furey, J. Galagan, J. G. Gilbert, C. Harmon, Y. Hayashizaki,
D. Haussler, H. Hermjakob, K. Hokamp, W. Jang, L. S. Johnson, T. A. Jones, S. Kasif, A. Kaspryzk,
S. Kennedy, W. J. Kent, P. Kitts, E. V. Koonin, I. Korf, D. Kulp, D. Lancet, T. M. Lowe, A.
McLysaght, T. Mikkelsen, J. V. Moran, N. Mulder, V. J. Pollara, C. P. Ponting, G. Schuler, J. Schultz,
G. Slater, A. F. Smit, E. Stupka, J. Szustakowski, D. Thierry-Mieg, J. Thierry-Mieg, L. Wagner, J.
Wallis, R. Wheeler, A. Williams, Y. I. Wolf, K. H. Wolfe, S. P. Yang, R. F. Yeh, F. Collins, M. S.
Guyer, J. Peterson, A. Felsenfeld, K. A. Wetterstrand, A. Patrinos, M. J. Morgan, P. de Jong, J. J.
Catanese, K. Osoegawa, H. Shizuya, S. Choi and Y. J. Chen (2001). "Initial sequencing and analysis
of the human genome." Nature 409(6822): 860-921.
Larsson, S. C., J. E. Johansson, S. O. Andersson and A. Wolk (2009). "Meat intake and bladder
cancer risk in a Swedish prospective cohort." Cancer Causes Control 20(1): 35-40.
Larsson, S. C. and A. Wolk (2006). "Meat consumption and risk of colorectal cancer: a meta-analysis
of prospective studies." Int J Cancer 119(11): 2657-2664.
Lavender, N. A., M. L. Benford, T. T. VanCleave, G. N. Brock, R. A. Kittles, J. H. Moore, D. W.
Hein and L. C. Kidd (2009). "Examination of polymorphic glutathione S-transferase (GST) genes,
tobacco smoking and prostate cancer risk among men of African descent: a case-control study."
BMC Cancer 9: 397.
Lawson, T. and C. Kolar (2002). "Human prostate epithelial cells metabolize chemicals of dietary
origin to mutagens." Cancer Lett 175(2): 141-146.
Le Marchand, L., J. H. Hankin, L. M. Pierce, R. Sinha, P. V. Nerurkar, A. A. Franke, L. R. Wilkens,
L. N. Kolonel, T. Donlon, A. Seifried, L. J. Custer, A. Lum-Jones and W. Chang (2002). "Well-done
red meat, metabolic phenotypes and colorectal cancer in Hawaii." Mutat Res 506-507: 205-214.
Lee, S. A., X. O. Shu, G. Yang, H. Li, Y. T. Gao and W. Zheng (2009). "Animal origin foods and
colorectal cancer risk: a report from the Shanghai Women's Health Study." Nutr Cancer 61(2): 194-
205.
Li, R., Y. Yang, Y. An, Y. Zhou, Y. Liu, Q. Yu, D. Lu, H. Wang, L. Jin, W. Zhou, J. Qian and Y. Y.
Shugart (2011). "Genetic polymorphisms in DNA double-strand break repair genes XRCC5, XRCC6
and susceptibility to hepatocellular carcinoma." Carcinogenesis 32(4): 530-536.
137
Lieber, M. R., Y. Ma, U. Pannicke and K. Schwarz (2003). "Mechanism and regulation of human
non-homologous DNA end-joining." Nat Rev Mol Cell Biol 4(9): 712-720.
Lim, K., C. Han, Y. Dai, M. Shen and T. Wu (2009). "Omega-3 polyunsaturated fatty acids inhibit
hepatocellular carcinoma cell growth through blocking beta-catenin and cyclooxygenase-2." Mol
Cancer Ther 8(11): 3046-3055.
Liu, N., J. E. Lamerdin, R. S. Tebbs, D. Schild, J. D. Tucker, M. R. Shen, K. W. Brookman, M. J.
Siciliano, C. A. Walter, W. Fan, L. S. Narayana, Z. Q. Zhou, A. W. Adamson, K. J. Sorensen, D. J.
Chen, N. J. Jones and L. H. Thompson (1998). "XRCC2 and XRCC3, new human Rad51-family
members, promote chromosome stability and protect against DNA cross-links and other damages."
Mol Cell 1(6): 783-793.
Lubbe, S. J., N. Whiffin, I. Chandler, P. Broderick and R. S. Houlston (2012). "Relationship between
16 susceptibility loci and colorectal cancer phenotype in 3146 patients." Carcinogenesis 33(1): 108-
112.
Lumbreras, B., S. Garte, K. Overvad, A. Tjonneland, F. Clavel-Chapelon, J. P. Linseisen, H. Boeing,
A. Trichopoulou, D. Palli, M. Peluso, V. Krogh, R. Tumino, S. Panico, H. B. Bueno-De-Mesquita, P.
H. Peeters, E. Lund, C. Martinez, M. Dorronsoro, A. Barricarte, M. D. Chirlaque, J. R. Quiros, G.
Berglund, G. Hallmans, N. E. Day, T. J. Key, R. Saracci, R. Kaaks, C. Malaveille, P. Ferrari, P.
Boffetta, T. Norat, E. Riboli, C. A. Gonzalez and P. Vineis (2008). "Meat intake and bladder cancer
in a prospective study: a role for heterocyclic aromatic amines?" Cancer Causes Control 19(6): 649-
656.
Lunn, J. C., G. Kuhnle, V. Mai, C. Frankenfeld, D. E. Shuker, R. C. Glen, J. M. Goodman, J. R.
Pollock and S. A. Bingham (2007). "The effect of haem in red and processed meat on the
endogenous formation of N-nitroso compounds in the upper gastrointestinal tract." Carcinogenesis
28(3): 685-690.
Manerio, E., V. L. Rodas, E. Costas and J. M. Hernandez (2008). "Shellfish consumption: a major
risk factor for colorectal cancer." Med Hypotheses 70(2): 409-412.
Manolio, T. A. (2010). "Genomewide association studies and assessment of the risk of disease." N
Engl J Med 363(2): 166-176.
Martin, F. L., Patel, II, O. Sozeri, P. B. Singh, N. Ragavan, C. M. Nicholson, E. Frei, W. Meinl, H.
Glatt, D. H. Phillips and V. M. Arlt (2010). "Constitutive expression of bioactivating enzymes in
normal human prostate suggests a capability to activate pro-carcinogens to DNA-damaging
metabolites." Prostate 70(14): 1586-1599.
138
Mattson, M. P. (2009). "Roles of the lipid peroxidation product 4-hydroxynonenal in obesity, the
metabolic syndrome, and associated vascular and neurodegenerative disorders." Exp Gerontol
44(10): 625-633.
Michaud, D. S., K. Augustsson, E. B. Rimm, M. J. Stampfer, W. C. Willet and E. Giovannucci
(2001). "A prospective study on intake of animal products and risk of prostate cancer." Cancer
Causes Control 12(6): 557-567.
Michaud, D. S., C. N. Holick, E. Giovannucci and M. J. Stampfer (2006). "Meat intake and bladder
cancer risk in 2 prospective cohort studies." Am J Clin Nutr 84(5): 1177-1183.
Miller, K. P. and K. S. Ramos (2001). "Impact of cellular metabolism on the biological effects of
benzo[a]pyrene and related hydrocarbons." Drug Metab Rev 33(1): 1-35.
Mills, P. K., W. L. Beeson, R. L. Phillips and G. E. Fraser (1989). "Cohort study of diet, lifestyle, and
prostate cancer in Adventist men." Cancer 64(3): 598-604.
Mirvish, S. S. (1995). "Role of N-nitroso compounds (NOC) and N-nitrosation in etiology of gastric,
esophageal, nasopharyngeal and bladder cancer and contribution to cancer of known exposures to
NOC." Cancer Lett 93(1): 17-48.
Mittal, R. D. and D. L. Srivastava (2007). "Cytochrome P4501A1 and microsomal epoxide hydrolase
gene polymorphisms: gene-environment interaction and risk of prostate cancer." DNA Cell Biol
26(11): 791-798.
Mo, Z., Y. Gao, Y. Cao, F. Gao and L. Jian (2009). "An updating meta-analysis of the GSTM1,
GSTT1, and GSTP1 polymorphisms and prostate cancer: a HuGE review." Prostate 69(6): 662-688.
Monsen, E. R. (1988). "Iron nutrition and absorption: dietary factors which impact iron
bioavailability." J Am Diet Assoc 88(7): 786-790.
Moonen, H. J., J. J. Briede, J. M. van Maanen, J. C. Kleinjans and T. M. de Kok (2002). "Generation
of free radicals and induction of DNA adducts by activation of heterocyclic aromatic amines via
different metabolic pathways in vitro." Mol Carcinog 35(4): 196-203.
Nabrzyski, M. and R. Gajewska (1994). "[The content of nitrates and nitrites in fruits, vegetables and
other foodstuffs]." Rocz Panstw Zakl Hig 45(3): 167-180.
139
Nahleh, Z., N. S. Bhatti and M. Mal (2011). "How to reduce your cancer risk: mechanisms and
myths." Int J Gen Med 4: 277-287.
Nelson, C. P., L. C. Kidd, J. Sauvageot, W. B. Isaacs, A. M. De Marzo, J. D. Groopman, W. G.
Nelson and T. W. Kensler (2001). "Protection against 2-hydroxyamino-1-methyl-6-
phenylimidazo[4,5-b]pyridine cytotoxicity and DNA adduct formation in human prostate by
glutathione S-transferase P1." Cancer Res 61(1): 103-109.
Newcomb, P. A., J. Baron, M. Cotterchio, S. Gallinger, J. Grove, R. Haile, D. Hall, J. L. Hopper, J.
Jass, L. Le Marchand, P. Limburg, N. Lindor, J. D. Potter, A. S. Templeton, S. Thibodeau and D.
Seminara (2007). "Colon Cancer Family Registry: an international resource for studies of the genetic
epidemiology of colon cancer." Cancer Epidemiol Biomarkers Prev 16(11): 2331-2343.
Norat, T., S. Bingham, P. Ferrari, N. Slimani, M. Jenab, M. Mazuir, K. Overvad, A. Olsen, A.
Tjonneland, F. Clavel, M. C. Boutron-Ruault, E. Kesse, H. Boeing, M. M. Bergmann, A. Nieters, J.
Linseisen, A. Trichopoulou, D. Trichopoulos, Y. Tountas, F. Berrino, D. Palli, S. Panico, R. Tumino,
P. Vineis, H. B. Bueno-de-Mesquita, P. H. Peeters, D. Engeset, E. Lund, G. Skeie, E. Ardanaz, C.
Gonzalez, C. Navarro, J. R. Quiros, M. J. Sanchez, G. Berglund, I. Mattisson, G. Hallmans, R.
Palmqvist, N. E. Day, K. T. Khaw, T. J. Key, M. San Joaquin, B. Hemon, R. Saracci, R. Kaaks and E.
Riboli (2005). "Meat, fish, and colorectal cancer risk: the European Prospective Investigation into
cancer and nutrition." J Natl Cancer Inst 97(12): 906-916.
Norskov, M. S., R. Frikke-Schmidt, S. E. Bojesen, B. G. Nordestgaard, S. Loft and A. Tybjaerg-
Hansen (2011). "Copy number variation in glutathione-S-transferase T1 and M1 predicts incidence
and 5-year survival from prostate and bladder cancer, and incidence of corpus uteri cancer in the
general population." Pharmacogenomics J 11(4): 292-299.
Nothlings, U., J. F. Yamamoto, L. R. Wilkens, S. P. Murphy, S. Y. Park, B. E. Henderson, L. N.
Kolonel and L. Le Marchand (2009). "Meat and heterocyclic amine intake, smoking, NAT1 and
NAT2 polymorphisms, and colorectal cancer risk in the multiethnic cohort study." Cancer Epidemiol
Biomarkers Prev 18(7): 2098-2106.
Nowell, S., B. Coles, R. Sinha, S. MacLeod, D. Luke Ratnasinghe, C. Stotts, F. F. Kadlubar, C. B.
Ambrosone and N. P. Lang (2002). "Analysis of total meat intake and exposure to individual
heterocyclic amines in a case-control study of colorectal cancer: contribution of metabolic variation
to risk." Mutat Res 506-507: 175-185.
Oba, S., N. Shimizu, C. Nagata, H. Shimizu, M. Kametani, N. Takeyama, T. Ohnuma and S.
Matsushita (2006). "The relationship between the consumption of meat, fat, and coffee and the risk
of colon cancer: a prospective study in Japan." Cancer Lett 244(2): 260-267.
140
Offer, S. M., Q. Pan-Hammarstrom, L. Hammarstrom and R. S. Harris (2010). "Unique DNA repair
gene variations and potential associations with the primary antibody deficiency syndromes IgAD and
CVID." PLoS One 5(8): e12260.
Overvik, E., M. Kleman, I. Berg and J. A. Gustafsson (1989). "Influence of creatine, amino acids and
water on the formation of the mutagenic heterocyclic amines found in cooked meat." Carcinogenesis
10(12): 2293-2301.
Papafili, A., M. R. Hill, D. J. Brull, R. J. McAnulty, R. P. Marshall, S. E. Humphries and G. J. Laurent
(2002). "Common promoter variant in cyclooxygenase-2 represses gene expression: evidence of role
in acute-phase inflammatory response." Arterioscler Thromb Vasc Biol 22(10): 1631-1636.
Park, S. Y., S. P. Murphy, L. R. Wilkens, B. E. Henderson and L. N. Kolonel (2007). "Fat and meat
intake and prostate cancer risk: the multiethnic cohort study." Int J Cancer 121(6): 1339-1345.
Pastorelli, R., M. Guanci, A. Cerri, E. Negri, C. La Vecchia, F. Fumagalli, M. Mezzetti, R. Cappelli, T.
Panigalli, R. Fanelli and L. Airoldi (1998). "Impact of inherited polymorphisms in glutathione S-
transferase M1, microsomal epoxide hydrolase, cytochrome P450 enzymes on DNA, and blood
protein adducts of benzo(a)pyrene-diolepoxide." Cancer Epidemiol Biomarkers Prev 7(8): 703-709.
Penning, T. M., M. E. Burczynski, C. F. Hung, K. D. McCoull, N. T. Palackal and L. S. Tsuruda
(1999). "Dihydrodiol dehydrogenases and polycyclic aromatic hydrocarbon activation: generation of
reactive and redox active o-quinones." Chem Res Toxicol 12(1): 1-18.
Pham, T. M., Y. Fujino, T. Kubo, R. Ide, N. Tokui, T. Mizoue, I. Ogimoto, S. Matsuda and T.
Yoshimura (2009). "Fish intake and the risk of fatal prostate cancer: findings from a cohort study in
Japan." Public Health Nutr 12(5): 609-613.
Pike, M. C., R. K. Peters, W. Cozen, N. M. Probst-Hensch, J. C. Felix, P. C. Wan and T. M. Mack
(1997). "Estrogen-progestin replacement therapy and endometrial cancer." J Natl Cancer Inst 89(15):
1110-1116.
Qing, Y., M. Yamazoe, K. Hirota, D. Dejsuphong, W. Sakai, K. N. Yamamoto, D. K. Bishop, X. Wu
and S. Takeda (2011). "The epistatic relationship between BRCA2 and the other RAD51 mediators in
homologous recombination." PLoS Genet 7(7): e1002148.
Ramesh, A., S. A. Walker, D. B. Hood, M. D. Guillen, K. Schneider and E. H. Weyand (2004).
"Bioavailability and risk assessment of orally ingested polycyclic aromatic hydrocarbons." Int J
Toxicol 23(5): 301-333.
141
Randel, G., M. Balzer, S. Grupe, S. Drusch, B. Kaina, K. L. Platt and K. Schwarz (2007).
"Degradation of heterocyclic aromatic amines in oil under storage and frying conditions and
reduction of their mutagenic potential." Food Chem Toxicol 45(11): 2245-2253.
Reese, A. C., V. Fradet and J. S. Witte (2009). "Omega-3 fatty acids, genetic variants in COX-2 and
prostate cancer." J Nutrigenet Nutrigenomics 2(3): 149-158.
Riboli, E. and T. Norat (2003). "Epidemiologic evidence of the protective effect of fruit and
vegetables on cancer risk." Am J Clin Nutr 78(3 Suppl): 559S-569S.
Richman, E. L., S. A. Kenfield, M. J. Stampfer, E. L. Giovannucci and J. M. Chan (2011). "Egg, red
meat, and poultry intake and risk of lethal prostate cancer in the prostate-specific antigen-era:
incidence and survival." Cancer Prev Res (Phila) 4(12): 2110-2121.
Rodriguez, C., M. L. McCullough, A. M. Mondul, E. J. Jacobs, A. Chao, A. V. Patel, M. J. Thun and
E. E. Calle (2006). "Meat consumption among Black and White men and risk of prostate cancer in
the Cancer Prevention Study II Nutrition Cohort." Cancer Epidemiol Biomarkers Prev 15(2): 211-
216.
Rohrmann, S., E. A. Platz, C. J. Kavanaugh, L. Thuita, S. C. Hoffman and K. J. Helzlsouer (2007).
"Meat and dairy consumption and subsequent risk of prostate cancer in a US cohort study." Cancer
Causes Control 18(1): 41-50.
Ruiz-Capillas, C. and F. Jimenez-Colmenero (2004). "Biogenic amines in meat and meat products."
Crit Rev Food Sci Nutr 44(7-8): 489-499.
Rybicki, B. A., N. L. Nock, A. T. Savera, D. Tang and A. Rundle (2006). "Polycyclic aromatic
hydrocarbon-DNA adduct formation in prostate carcinogenesis." Cancer Lett 239(2): 157-167.
Safarinejad, M. R., N. Shafiei and S. H. Safarinejad (2011). "Glutathione S-transferase gene
polymorphisms (GSTM1, GSTT1, GSTP1) and prostate cancer: a case-control study in Tehran,
Iran." Prostate Cancer Prostatic Dis 14(2): 105-113.
Salmon, C. P., M. G. Knize and J. S. Felton (1997). "Effects of marinating on heterocyclic amine
carcinogen formation in grilled chicken." Food Chem Toxicol 35(5): 433-441.
Sanak, M., H. Plutecka, W. Szczeklik, W. Piwowarska, P. Rostoff and A. Szczeklik (2010).
"Functional promoter polymorphism of cyclooxygenase-2 modulates the inflammatory response in
stable coronary heart disease." Pol Arch Med Wewn 120(3): 82-88.
142
Sandhu, M. S., I. R. White and K. McPherson (2001). "Systematic review of the prospective cohort
studies on meat consumption and colorectal cancer risk: a meta-analytical approach." Cancer
Epidemiol Biomarkers Prev 10(5): 439-446.
Sato, Y., N. Nakaya, S. Kuriyama, Y. Nishino, Y. Tsubono and I. Tsuji (2006). "Meat consumption
and risk of colorectal cancer in Japan: the Miyagi Cohort Study." Eur J Cancer Prev 15(3): 211-218.
Schuurman, A. G., P. A. van den Brandt, E. Dorant and R. A. Goldbohm (1999). "Animal products,
calcium and protein and prostate cancer risk in The Netherlands Cohort Study." Br J Cancer 80(7):
1107-1113.
Schwartz, G. G., E. M. John, G. Rowland and S. A. Ingles (2010). "Prostate cancer in African-
American men and polymorphism in the calcium-sensing receptor." Cancer Biol Ther 9(12): 994-999.
Shimada, T. and Y. Fujii-Kuriyama (2004). "Metabolic activation of polycyclic aromatic hydrocarbons
to carcinogens by cytochromes P450 1A1 and 1B1." Cancer Sci 95(1): 1-6.
Silberstein, J. L. and J. K. Parsons (2010). "Evidence-based principles of bladder cancer and diet."
Urology 75(2): 340-346.
Sinha, R. (2002). "An epidemiologic approach to studying heterocyclic amines." Mutat Res 506-507:
197-204.
Sinha, R., A. Cross, J. Curtin, T. Zimmerman, S. McNutt, A. Risch and J. Holden (2005).
"Development of a food frequency questionnaire module and databases for compounds in cooked
and processed meats." Mol Nutr Food Res 49(7): 648-655.
Sinha, R., Y. Park, B. I. Graubard, M. F. Leitzmann, A. Hollenbeck, A. Schatzkin and A. J. Cross
(2009). "Meat and meat-related compounds and risk of prostate cancer in a large prospective cohort
study in the United States." Am J Epidemiol 170(9): 1165-1177.
Sinha, R. and N. Rothman (1997). "Exposure assessment of heterocyclic amines (HCAs) in
epidemiologic studies." Mutat Res 376(1-2): 195-202.
Sinha, R., N. Rothman, E. D. Brown, C. P. Salmon, M. G. Knize, C. A. Swanson, S. C. Rossi, S. D.
Mark, O. A. Levander and J. S. Felton (1995). "High concentrations of the carcinogen 2-amino-1-
methyl-6-phenylimidazo- [4,5-b]pyridine (PhIP) occur in chicken but are dependent on the cooking
method." Cancer Res 55(20): 4516-4519.
143
Song, Q., B. Zhu, W. Hu, L. Cheng, H. Gong, B. Xu, X. Zheng, L. Zou, R. Zhong, S. Duan, W.
Chen, R. Rui, J. Wu and X. Miao (2012). "A common SMAD7 variant is associated with risk of
colorectal cancer: evidence from a case-control study and a meta-analysis." PLoS One 7(3): e33318.
Souiden, Y., M. Mahdouani, K. Chaieb, R. Elkamel and K. Mahdouani (2010). "Polymorphisms of
glutathione-S-transferase M1 and T1 and prostate cancer risk in a Tunisian population." Cancer
Epidemiol 34(5): 598-603.
Steinbrecher, A., S. Rohrmann, M. Timofeeva, A. Risch, E. Jansen and J. Linseisen (2010). "Dietary
glucosinolate intake, polymorphisms in selected biotransformation enzymes, and risk of prostate
cancer." Cancer Epidemiol Biomarkers Prev 19(1): 135-143.
Steinmetz, K. A. and J. D. Potter (1991). "Vegetables, fruit, and cancer. II. Mechanisms." Cancer
Causes Control 2(6): 427-442.
Stram, D. O., J. H. Hankin, L. R. Wilkens, M. C. Pike, K. R. Monroe, S. Park, B. E. Henderson, A.
M. Nomura, M. E. Earle, F. S. Nagamine and L. N. Kolonel (2000). "Calibration of the dietary
questionnaire for a multiethnic cohort in Hawaii and Los Angeles." Am J Epidemiol 151(4): 358-370.
Strange, R. C. and A. A. Fryer (1999). "The glutathione S-transferases: influence of polymorphism on
cancer susceptibility." IARC Sci Publ(148): 231-249.
Strange, R. C. and A. A. Fryer (1999). "The glutathione S-transferases: influence of polymorphism on
cancer susceptibility." IARC Sci Publ 148: 231-249.
Stuff, J. E., E. T. Goh, S. L. Barrera, M. L. Bondy and M. R. Forman (2009). "Construction of an N-
nitroso database for assessing dietary intake." J Food Compost Anal 22(Suppl 1): S42-S47.
Sugie, S., P. Q. Vinh, K. M. Rahman, J. Ushida, H. Kohno, R. Suzuki, A. Hara, B. Quang le, T.
Tanaka and H. Mori (2005). "Suppressive effect of 1,4-phenylene diisothiocyanate on N-butyl-N-(4-
hydroxybutyl)nitrosamine-induced urinary bladder carcinogenesis in male ICR mice." Int J Cancer
117(4): 524-530.
Sugimura, T., K. Wakabayashi, H. Nakagama and M. Nagao (2004). "Heterocyclic amines:
Mutagens/carcinogens produced during cooking of meat and fish." Cancer Sci 95(4): 290-299.
Suzzi, G. and F. Gardini (2003). "Biogenic amines in dry fermented sausages: a review." Int J Food
Microbiol 88(1): 41-54.
144
Szymanski, K. M., D. C. Wheeler and L. A. Mucci (2010). "Fish consumption and prostate cancer
risk: a review and meta-analysis." Am J Clin Nutr 92(5): 1223-1233.
Taioli, E., R. E. Flores-Obando, I. Agalliu, P. Blanchet, C. H. Bunker, R. E. Ferrell, M. Jackson, L. C.
Kidd, S. Kolb, N. A. Lavender, N. McFarlane-Anderson, S. S. Morrison, L. Multigner, E. A.
Ostrande, J. Y. Park, A. L. Patrick, T. R. Rebbeck, M. Romana, J. L. Stanford, F. Ukoli, T. T.
Vancleave, C. M. Zeigler-Johnson, B. Mutetwa and C. Ragin (2011). "Multi-institutional prostate
cancer study of genetic susceptibility in populations of African descent." Carcinogenesis 32(9): 1361-
1365.
Tang, L. and Y. Zhang (2004). "Dietary isothiocyanates inhibit the growth of human bladder
carcinoma cells." J Nutr 134(8): 2004-2010.
Tang, L. and Y. Zhang (2004). "Isothiocyanates in the chemoprevention of bladder cancer." Curr
Drug Metab 5(2): 193-201.
Tang, Y. M., B. L. Green, G. F. Chen, P. A. Thompson, N. P. Lang, A. Shinde, D. X. Lin, W. Tan, B.
D. Lyn-Cook, G. J. Hammons and F. F. Kadlubar (2000). "Human CYP1B1 Leu432Val gene
polymorphism: ethnic distribution in African-Americans, Caucasians and Chinese; oestradiol
hydroxylase activity; and distribution in prostate cancer cases and controls." Pharmacogenetics 10(9):
761-766.
Tantamango-Bartley, Y., K. Jaceldo-Siegl, J. Fan and G. Fraser (2013). "Vegetarian Diets and the
Incidence of Cancer in a Low-risk Population." Cancer Epidemiol Biomarkers Prev 22(2): 286-294.
Tessitore, L. and E. Bollito (2006). "Early induction of TGF-beta1 through a fasting-re-feeding
regimen promotes liver carcinogenesis by a sub-initiating dose of diethylnitrosamine." Cell Prolif
39(2): 105-116.
Tricker, A. R. (1997). "N-nitroso compounds and man: sources of exposure, endogenous formation
and occurrence in body fluids." Eur J Cancer Prev 6(3): 226-268.
Tripuraneni, J., A. Koutsouris, L. Pestic, P. De Lanerolle and G. Hecht (1997). "The toxin of
diarrheic shellfish poisoning, okadaic acid, increases intestinal epithelial paracellular permeability."
Gastroenterology 112(1): 100-108.
Valerie, K. and L. F. Povirk (2003). "Regulation and mechanisms of mammalian double-strand break
repair." Oncogene 22(37): 5792-5812.
145
van Maanen, J. M., E. J. Moonen, L. M. Maas, J. C. Kleinjans and F. J. van Schooten (1994).
"Formation of aromatic DNA adducts in white blood cells in relation to urinary excretion of 1-
hydroxypyrene during consumption of grilled meat." Carcinogenesis 15(10): 2263-2268.
Venkateswaran, V. and L. H. Klotz (2010). "Diet and prostate cancer: mechanisms of action and
implications for chemoprevention." Nat Rev Urol 7(8): 442-453.
Venter, J. C., M. D. Adams, E. W. Myers, P. W. Li, R. J. Mural, G. G. Sutton, H. O. Smith, M.
Yandell, C. A. Evans, R. A. Holt, J. D. Gocayne, P. Amanatides, R. M. Ballew, D. H. Huson, J. R.
Wortman, Q. Zhang, C. D. Kodira, X. H. Zheng, L. Chen, M. Skupski, G. Subramanian, P. D.
Thomas, J. Zhang, G. L. Gabor Miklos, C. Nelson, S. Broder, A. G. Clark, J. Nadeau, V. A.
McKusick, N. Zinder, A. J. Levine, R. J. Roberts, M. Simon, C. Slayman, M. Hunkapiller, R. Bolanos,
A. Delcher, I. Dew, D. Fasulo, M. Flanigan, L. Florea, A. Halpern, S. Hannenhalli, S. Kravitz, S.
Levy, C. Mobarry, K. Reinert, K. Remington, J. Abu-Threideh, E. Beasley, K. Biddick, V. Bonazzi, R.
Brandon, M. Cargill, I. Chandramouliswaran, R. Charlab, K. Chaturvedi, Z. Deng, V. Di Francesco,
P. Dunn, K. Eilbeck, C. Evangelista, A. E. Gabrielian, W. Gan, W. Ge, F. Gong, Z. Gu, P. Guan, T.
J. Heiman, M. E. Higgins, R. R. Ji, Z. Ke, K. A. Ketchum, Z. Lai, Y. Lei, Z. Li, J. Li, Y. Liang, X. Lin,
F. Lu, G. V. Merkulov, N. Milshina, H. M. Moore, A. K. Naik, V. A. Narayan, B. Neelam, D.
Nusskern, D. B. Rusch, S. Salzberg, W. Shao, B. Shue, J. Sun, Z. Wang, A. Wang, X. Wang, J. Wang,
M. Wei, R. Wides, C. Xiao, C. Yan, A. Yao, J. Ye, M. Zhan, W. Zhang, H. Zhang, Q. Zhao, L.
Zheng, F. Zhong, W. Zhong, S. Zhu, S. Zhao, D. Gilbert, S. Baumhueter, G. Spier, C. Carter, A.
Cravchik, T. Woodage, F. Ali, H. An, A. Awe, D. Baldwin, H. Baden, M. Barnstead, I. Barrow, K.
Beeson, D. Busam, A. Carver, A. Center, M. L. Cheng, L. Curry, S. Danaher, L. Davenport, R.
Desilets, S. Dietz, K. Dodson, L. Doup, S. Ferriera, N. Garg, A. Gluecksmann, B. Hart, J. Haynes, C.
Haynes, C. Heiner, S. Hladun, D. Hostin, J. Houck, T. Howland, C. Ibegwam, J. Johnson, F. Kalush,
L. Kline, S. Koduru, A. Love, F. Mann, D. May, S. McCawley, T. McIntosh, I. McMullen, M. Moy, L.
Moy, B. Murphy, K. Nelson, C. Pfannkoch, E. Pratts, V. Puri, H. Qureshi, M. Reardon, R.
Rodriguez, Y. H. Rogers, D. Romblad, B. Ruhfel, R. Scott, C. Sitter, M. Smallwood, E. Stewart, R.
Strong, E. Suh, R. Thomas, N. N. Tint, S. Tse, C. Vech, G. Wang, J. Wetter, S. Williams, M.
Williams, S. Windsor, E. Winn-Deen, K. Wolfe, J. Zaveri, K. Zaveri, J. F. Abril, R. Guigo, M. J.
Campbell, K. V. Sjolander, B. Karlak, A. Kejariwal, H. Mi, B. Lazareva, T. Hatton, A. Narechania, K.
Diemer, A. Muruganujan, N. Guo, S. Sato, V. Bafna, S. Istrail, R. Lippert, R. Schwartz, B. Walenz, S.
Yooseph, D. Allen, A. Basu, J. Baxendale, L. Blick, M. Caminha, J. Carnes-Stine, P. Caulk, Y. H.
Chiang, M. Coyne, C. Dahlke, A. Mays, M. Dombroski, M. Donnelly, D. Ely, S. Esparham, C. Fosler,
H. Gire, S. Glanowski, K. Glasser, A. Glodek, M. Gorokhov, K. Graham, B. Gropman, M. Harris, J.
Heil, S. Henderson, J. Hoover, D. Jennings, C. Jordan, J. Jordan, J. Kasha, L. Kagan, C. Kraft, A.
Levitsky, M. Lewis, X. Liu, J. Lopez, D. Ma, W. Majoros, J. McDaniel, S. Murphy, M. Newman, T.
Nguyen, N. Nguyen, M. Nodell, S. Pan, J. Peck, M. Peterson, W. Rowe, R. Sanders, J. Scott, M.
Simpson, T. Smith, A. Sprague, T. Stockwell, R. Turner, E. Venter, M. Wang, M. Wen, D. Wu, M.
Wu, A. Xia, A. Zandieh and X. Zhu (2001). "The sequence of the human genome." Science
291(5507): 1304-1351.
Vermeer, I. T., D. M. Pachen, J. W. Dallinga, J. C. Kleinjans and J. M. van Maanen (1998). "Volatile
N-nitrosamine formation after intake of nitrate at the ADI level in combination with an amine-rich
diet." Environ Health Perspect 106(8): 459-463.
146
Verna, L., J. Whysner and G. M. Williams (1996). "N-nitrosodiethylamine mechanistic data and risk
assessment: bioactivation, DNA-adduct formation, mutagenicity, and tumor initiation." Pharmacol
Ther 71(1-2): 57-81.
Wang, W., A. Bergh and J. E. Damber (2004). "Chronic inflammation in benign prostate hyperplasia
is associated with focal upregulation of cyclooxygenase-2, Bcl-2, and cell proliferation in the
glandular epithelium." Prostate 61(1): 60-72.
Wang, Z., Z. Zong, S. Li and B. Chen (2002). "[Nitrate accumulation in vegetables and its residual in
vegetable fields]." Huan Jing Ke Xue 23(3): 79-83.
WCRF (2007). "Food, Nutrition, Physical Activity, and the Prevention of Cancer: A Global
Perspective World Cancer Research Fund and American Institute for Cancer Research Washington
DC."
WCRF/AICR (2011). World Cancer Research Fund/American Institute for Cancer Research.
continuous update project interim report summary. Food, nutrition, physical activity, and the
prevention of cancer. .
Wendel, M. and A. R. Heller (2009). "Anticancer actions of omega-3 fatty acids--current state and
future perspectives." Anticancer Agents Med Chem 9(4): 457-470.
Wiese, F. W., P. A. Thompson and F. F. Kadlubar (2001). "Carcinogen substrate specificity of
human COX-1 and COX-2." Carcinogenesis 22(1): 5-10.
Wilkens, L. R., M. M. Kadir, L. N. Kolonel, A. M. Nomura and J. H. Hankin (1996). "Risk factors for
lower urinary tract cancer: the role of total fluid consumption, nitrites and nitrosamines, and selected
foods." Cancer Epidemiol Biomarkers Prev 5(3): 161-166.
Willett, W. C., G. R. Howe and L. H. Kushi (1997). "Adjustment for total energy intake in
epidemiologic studies." Am J Clin Nutr 65(4 Suppl): 1220S-1228S; discussion 1229S-1231S.
Williams, J. A., F. L. Martin, G. H. Muir, A. Hewer, P. L. Grover and D. H. Phillips (2000).
"Metabolic activation of carcinogens and expression of various cytochromes P450 in human prostate
tissue." Carcinogenesis 21(9): 1683-1689.
Wilson, D. M., 3rd and D. Barsky (2001). "The major human abasic endonuclease: formation,
consequences and repair of abasic lesions in DNA." Mutation Research 485: 283-307.
147
Wong, H. L., S. E. Murphy and S. S. Hecht (2005). "Cytochrome P450 2A-catalyzed metabolic
activation of structurally similar carcinogenic nitrosamines: N'-nitrosonornicotine enantiomers, N-
nitrosopiperidine, and N-nitrosopyrrolidine." Chem Res Toxicol 18(1): 61-69.
Wood, R. D., M. Mitchell and T. Lindahl (2005). "Human DNA repair genes, 2005." Mutat Res
577(1-2): 275-283.
Wu, S., B. Feng, K. Li, X. Zhu, S. Liang, X. Liu, S. Han, B. Wang, K. Wu, D. Miao, J. Liang and D.
Fan (2012). "Fish consumption and colorectal cancer risk in humans: a systematic review and meta-
analysis." Am J Med 125(6): 551-559 e555.
Yamazaki, H., Y. Inui, C. H. Yun, F. P. Guengerich and T. Shimada (1992). "Cytochrome P450 2E1
and 2A6 enzymes as major catalysts for metabolic activation of N-nitrosodialkylamines and tobacco-
related nitrosamines in human liver microsomes." Carcinogenesis 13(10): 1789-1794.
Zeegers, M. P., R. F. Selen, J. C. Kleinjans, R. A. Goldbohm and P. A. van den Brandt (2006).
"Nitrate intake does not influence bladder cancer risk: the Netherlands cohort study." Environ
Health Perspect 114(10): 1527-1531.
Ziegler, M. and S. L. Oei (2001). "A cellular survival switch: poly(ADP-ribosyl)ation stimulates DNA
repair and silences transcription." Bioessays 23(6): 543-548.
Abstract (if available)
Abstract
The central topic of this thesis involves investigating the potential role in cancer risk of carcinogens present in the everyday human diet. In addition, I report on other relevant dietary compounds, such as antioxidants, that can reduce the detrimental effect of these carcinogens and can thus be preventative against cancer risk. Finally, I present results from investigations on the underlying genetic variation involved in an individual’s metabolism of these specific carcinogens and antioxidants. ❧ Two of the key carcinogens considered in this thesis are polycyclic aromatic hydrocarbons (PAHs) and heterocyclic amines (HCAs), which are known to form in meats and fish, cooked at high temperatures, and are potent carcinogens. Upon entering the body, PAHs and HCAs can form adducts with DNA, and can also generate oxidative DNA damage, thus contributing to the mutational load that can lead to tumor formation. In contrast, diets high in fish have been associated with decreased risk of several cancers, including prostate, breast and colorectal, albeit results have been inconsistent. It has been proposed that omega-3 polyunsaturated fatty acids (PUFA) found in some fish, particularly dark and oily species, might explain this association. ❧ A third group of dietary carcinogens investigated in this thesis are N-nitroso compounds (NOCs), which have been demonstrated to cause bladder tumors in mice and other experimental organisms. The main sources of dietary exposure to NOCs are processed meats, such as cold cuts, sausages and bacon. Moreover, NOCs can be formed endogenously in the body when amines undergo nitrosation by nitrites at various sites, including the oral cavity, stomach, intestines and urinary bladder, as well as other sites of inflammation or infection. Endogenous formation of NOCs has been estimated to contribute between 45-75% of total NOC exposure. Conversely, plant sources in diet contain high amounts of compounds such as vitamin C and specific polyphenols that directly inhibit these endogenous nitrosation reactions. ❧ Investigation of dietary exposures to carcinogens such as those discussed above is of public health relevance in order to identify dietary items, range of exposure, and cooking practices that might be most critical to reduce cancer risk. Given the complex interrelationship between these various carcinogenic and anti-carcinogenic exposures, studying the underlying genetic variation of the enzymes relevant for their metabolism provides insight into which of these mutagens is more relevant for specific cancers.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Genes and environment in prostate cancer risk and prognosis
PDF
Genetic variation in the base excision repair pathway, environmental risk factors and colorectal adenoma risk
PDF
Dietary and supplementary folate intake and prostate cancer risk
PDF
Fish consumption and risk of colorectal cancer
PDF
Genetic and environmental risk factors for childhood cancer
PDF
Meat intake, polymorphisms in the NER and MMR pathways and colorectal cancer risk
PDF
Red and processed meat consumption and colorectal cancer risk: meta-analysis of case-control studies
PDF
Robust feature selection with penalized regression in imbalanced high dimensional data
PDF
Identifying genetic, environmental, and lifestyle determinants of ethnic variation in risk of pancreatic cancer
PDF
Carcinogen metabolism genes, meat intake, and colorectal cancer risk
PDF
Molecular mechanisms of young-onset type 2 diabetes: integration of diet and multi-omics biomarkers
PDF
The effects of hormonal exposures on ovarian and breast cancer risk
PDF
Risk factors of pelvic floor disorders in the multiethnic cohort study
PDF
Dietary fats, fat metabolizing genes, and the risk of breast cancer
PDF
Disparities in gallbladder, intra-hepatic bile duct, and other biliary cancers among multi-ethnic populations: a California Cancer Registry study
PDF
Genetic studies of inflammation and cardiovascular disease
PDF
Identification of gene expression regulated by 1,25(OH)₂D3 in human endometriosis cell lines with next-generation sequencing
PDF
Prostate cancer: genetic susceptibility and lifestyle risk factors
PDF
Genetic risk factors in multiple myeloma
PDF
Carcinogenic exposures in racial/ethnic groups
Asset Metadata
Creator
Catsburg, Chelsea E.
(author)
Core Title
Dietary carcinogens and genetic variation in their metabolism: epidemiological studies on the risk of selected cancers
School
Keck School of Medicine
Degree
Doctor of Philosophy
Degree Program
Molecular Epidemiology
Publication Date
05/13/2013
Defense Date
02/19/2013
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
cancer,diet,Meat,metabolism,OAI-PMH Harvest
Format
application/pdf
(imt)
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Stern, Mariana C. (
committee chair
), Gopalakrishna, Rayudu (
committee member
), Ingles, Sue Ann (
committee member
), Lewinger, Juan Pablo (
committee member
)
Creator Email
catsburg@usc.edu,chelseacats@gmail.com
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c3-255192
Unique identifier
UC11293204
Identifier
etd-CatsburgCh-1687.pdf (filename),usctheses-c3-255192 (legacy record id)
Legacy Identifier
etd-CatsburgCh-1687.pdf
Dmrecord
255192
Document Type
Dissertation
Format
application/pdf (imt)
Rights
Catsburg, Chelsea E.
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA
Tags
diet
metabolism