Transcript Nutrition and Cancer Prevention

Nutrition and Cancer
Prevention
Jackilen Shannon, PhD, RD
Overview

Historical perspective
◦ Population data to molecular mechanisms

Diet and Breast Cancer
◦ What is the evidence?
◦ Overview of findings from Shanghai
◦ Future research directions

Challenges and future directions in Nutrition and
Cancer Prevention Research
Columbus
“discovers” America
Yong-He Yan
Poor nutrition a
cause for
esophageal cancer
1492
960-1279 AD
Roger Williams ‘probably no
single factor is more potent in
determining the outbreak of
cancer in the predisposed,
then excessive feeding’ citing
specifically, ‘deficient eating
and probably lack of
sufficient vegetable food’.
(The Natural History of
Cancer)
1855
1676
1914-1918
1908
Wiseman - cancer may
arise from “ an errour
in Diet, a great
acrimony in the meats
and drinks meeting
John Snow tracks
with a fault in the first source of Cholera
Concoction (digestion)” outbreak
– advised abstention
from ‘salt, sharp and
gross meats’
Great
Depression
Watson & Crick
publish the
structure of DNA
1929-1939 1939-1945
1937
World War I
World War II
Frederick Hoffman
(founder of ACS) concluded from a
systematic literature
review ‘excessive nutrition
if not the chief cause is at
least a contributory factor
of the first importance’
1953
Population studies:
Correlational results

1975 – correlational study of incidence of 27 cancers in 23
countries and dietary intake.
Armstrong B, Doll R. Int J Cancer 1975; 15: 617-31
Further Evidence:
Case-Control and Cohort Studies
Doll
and Peto (1981) 35% of cancer deaths may be
attributed to dietary factors
Doll R, Peto R. The causes of cancer. JNCI 1981; 66:1191-1308
World
Cancer Research Fund (1997) Cancer incidence can
be reduced by 30%-40% with diet, physical activity and
appropriate body size.
World Cancer Research Fund,. Food, Nutrition and the Prevention of Cancer: a Global Perspective. Washington DC
(USA): American Institute for Cancer Research; 1997: 310-323.
Role of Diet in the Cancer Process Metabolism & Excretion of Carcinogens
•Block metabolic activation
• Increase metabolic detoxification
Bioactive dietary constituents
B-vitamins, glutathione,
flavonoids
Alcohol
Dietary Carcinogens
Aflotoxin, heterocyclic amines
N-nitroso compounds
Smoking,
chewing tobacco,
betel
Smoking
Workplace
Genes
Procarcinogen
Phase I metabolizing enzymes
P450’s etc.
EXCRETION
Bioactive dietary constituents
Isothiocyanate, selenium,
other phytochemicals
Ultimate carciongen
Genes
Phase II metabolizing enzymes
glutahione, glutathione transferase,
N-acetyl transferases
EXCRETION
Role of Diet in the Cancer Process –
Initiation
Folate Deficiency
Heterocyclic Amines
Inadequate Methyl
groups
PhiP
DNA adducts
Genes
Physical Activity
Energy Intake
DNA Repair
Hypomethylation
of P53
Somatic alteration of oncogenes,
Tumour-suppressor genes and
DNA repair genes
NORMAL DNA
Role of Diet in the Cancer Process –
Promotion
 Physical Activity
Phytoestrogens
 Energy
 Fat
Hormones
Obesity
Abnormal DNA & cell replication
Dietary factors
 Protein
 Methionine
 Cholesterol
Growth factors
Specific nutrients e.g.
carotenoids, retinol
Colonic Bacteria
Fibre
REDIFFERENTIATION
APOPTOSIS
Volatile fatty acids
-3 fatty acids
Precancerous lesions & dysplasia
Genes
Role of Diet in the Cancer Process –
Progression
•Less work in this area
•Some of the same factors that function in initiation
DNA Repair Genes
DNA Damage
Dietary factors
Precancerous lesions & dysplasia
Immune System
Growth factors
Cancer
Hormones
Smoking &
other exposures
DNA Damage
Metastasis
DNA Repair Genes
Diet in the Cancer Process
Dietary Carcinogens,
heterocyclic amines,
amines, PAHs
Procarcinogen
B-vitamins, glutathione,
flavonoids
Folate Deficiency
Phase I metabolizing enzymes
Ultimate carcinogen
Isothiocyanate, selenium,
other phytochemicals
Phase II metabolizing enzymes
Hypomethylation
of P53
DNA adducts
Obesity
Physical
Activity
 Protein
 Methionine
 Cholesterol
Specific nutrients e.g.
carotenoids, retinol
Somatic alteration of oncogenes,Tumoursuppressor genes and DNA repair genes
Hormones & Growth Factors
Abnormal DNA & cell replication
Redifferentiation
Apoptosis
Precancerous lesions & dysplasia
Cancer
Metastasis
-3 fatty acids
Specific Dietary Factors Associated with
Breast Cancer Risk
Fruits and Vegetables.
 Total Fat
 Red Meat
 Phytoestrogens.

Fruits and Vegetables
and Breast Cancer Risk
Procarcinogen
B-vitamins, glutathione, flavonoids
Isothiocyanate, selenium,
other phytochemicals
Phase I metabolizing
enzymes
Ultimate carcinogen
Phase II
metabolizing
enzymes
DNA adducts
Folate
Deficiency
Hypomethylation
of P53
Somatic alteration
Abnormal DNA & cell replication
Specific nutrients e.g.
carotenoids, retinol
Redifferentiation
Precancerous lesions & dysplasia

Biologic Mechanism –
Cancer
Metastasis
◦ Unclear – but thought to be primarily due to antioxidants
(phytoestrogens (lignans), other biochemical substance)
◦ Phytochemicals have been shown to induce detoxifying (Phase II)
enzymes.
◦ May also function later in cancer process – redifferentiation
(promotion)

Epidemiologic Evidence:
◦ Strong correlational evidence
◦ Case-control studies – Probable
◦ Cohort studies – inconsistent
Epidemiologic Evidence

Recent cohort studies have cast
doubt◦ Pooling project
 8 cohort studies
 No evidence of a protective effect
Meta-analysis of 15 Case-control and 10 Cohort studies.
Change in Breast Cancer risk with each
additional 100g intake. OR (95% CI)
1.1
1.05
1
Smith-Warner SA, Spiegelman D, Yaun SS, et al. Intake of fruits and
vegetables and risk of breast cancer: a pooled analysis of cohort studies.
JAMA. 2001; 285:769-776.
0.95
0.9
0.85
◦ EPIC (8 European countries)
 285,526 women, 5.4 years followup
 Vegetables, OR = 0.98 (95% CI,
0.84-1.14)
 Fruit, OR=1.09 (95% CI, 0.94-1.25)
Van Gils C, Peeters PHM et al. Consumption of Vegetables and Fruits
and Risk of Breast Cancer. JAMA. 2005;293:183-193
0.8
0.75
0.7
Fruits
Vegetables
Fruits
Vegetables
Riboli E, Norat T. Epidemiologic evidence of the protective effect of fruit and
vegetables on cancer risk. Am J Clin Nutr 2003;78(suppl):559S–69S.
Evidence for the Role of Fat in Breast
Carcinogenesis
 Animal evidence:

High Fat –
High Meat
1940’s Tannenbaum:
◦ Kcal restricted = ↓mammary tumors
◦ ↑ fat diet = ↑ mammary tumors.
Procarcinogen
◦ Indirectly impacts breast cancer risk
through altering hormonal pathways

1997 Fay & Freedman – Metaanalyses of animal studies.
◦ Similar findings, still unclear if effect
is due to fat, calories, type of fat
High Fat
etc…
◦ Indirect impacts of fat on –
 Calories
 Protein
 Meat
 F&V
 Body size
Dietary Carcinogens,
heterocyclic amines,
amines, PAHs
Ultimate carcinogen
Lipid peroxidation
-6 fatty
acids
DNA adducts
Somatic alteration
Obesity
 Protein
 Methionine
 Cholesterol
Hormones &
Growth Factors
Abnormal DNA &
cell replication
Apoptosis
Precancerous lesions
& dysplasia
Cancer
Metastasis
-3 fatty
acids
Dietary Fat and Breast Cancer Risk
 Epidemiologic Evidence:
Since 1996 nearly 500 articles have been
published on fat intake and breast cancer
risk in humans.
◦ Strongest evidence of an association from
correlational studies
◦ Inconsistent evidence from case-control studies
◦ No association found in cohort studies
Red Meat and Breast Cancer

Biologic Mechanism:
◦ Directly- May contribute procarcinogens –
 heterocyclic amines- through overcooking
 N-nitroso compounds (proteins)
◦ Indirectly – Associated with high fat and energy intake.

Epidemiologic Evidence:
◦ Strong correlational evidence
◦ Case-control studies – inconsistent
◦ Cohort studies – inconsistent –
 pooling project found no association
Soy and Breast Cancer Risk

Soy products are high in phytoestrogens
◦ Competitive binding of the estrogen receptor
◦ Increase production of SHBG and thus reduce free estrogen
◦ Decrease cell proliferation and induce apoptosis

Majority of evidence comes for ecologic and in vitro
studies.

Epidemiologic studies◦ Inconsistent findings, but few large long-term studies.
◦ Two recent case-control studies suggest assoc. with intake in adolescence
◦ Little evidence of an increased risk
Why the conflicting findings?

What is the important time of intake?
◦ BC and height (growth)

What is the correct nutrient/ food to
measure?
◦ Oils? Monounsaturated vs. Poly
◦ -3 vs. -6
◦ All phytoestrogens v. Isoflavones v. Lignan

Foods vs. Nutrients
◦ Fruits and Vegetables v. Carotenoids
Randomized Trial of
Breast Self Examination (BSE)
Cell Proliferation
Study (CPS)
The Nutrition Study (1995-2000)
Primary Aim: To determine if
Reproductive
Health Study
increased risk of breast cancer is
associated with high consumption
of fat and red meat, and low
consumption of soy, and fruits
and vegetables.
What can be learned from the Shanghai analysis?

Potential for greater variation in total fat and soy intake.
Soy exposure in Western studies
Distribution of % Calories from Fat
Soy exposure in proposed
study
China 1989*
China 1993*
U.S. 1993**
high
Cancer
Incidenc
e
Rate
0%
4%
8%
12% 16% 20% 24% 28% 32% 36% 40% 44% 48% 52%
% kcals from fat
low
low
high
Soy Foods Exposure
Proposed dose response association between dietary exposures
and cancer risk. McMichael A, Potter J. JNCI 1985
What can be learned from the Shanghai
analysis?

Variation in meat consumption (heme iron) with greater
intake of organ meats.

Higher consumption of particular types of vegetables
(e.g leguminosae, cruciferea) than seen in Western
populations.
The Shanghai Nutrition Study
◦ Primary Aim:
 To determine if increased risk of breast cancer is
associated with high consumption of fat and red meat,
and low consumption of soy, and fruits and vegetables
in a population of women in Shanghai, China.
Nutrition Study Design
Women Recruited into the Breast Self-Examination Study (1989-1991)
(n= 266,064)
Recruit and Interview Women with Breast Biopsies into
The Cell Proliferation Study (1992-1997)
Recruit and Interview Women for
The Diet Study (1995-2000)
Complete FFQ & blood draw
Biopsy
Select Controls
from unaffected cohort
Complete FFQ &
blood draw
(n=367)
Benign
(n=949+)
Malignant
(n=378/ 436)
Fibroadenoma
(n=327)
Fibrocysitic
(n=551/ 622)
Select Controls
from unaffected cohort
Complete FFQ &
blood draw
(n=703/ 862)
Food Frequency Questionnaire

Modified from a validated NCI questionnaire used
previously in Shanghai.

24- hour recall portion was added.

Reviewed by colleagues in Shanghai

Portion size section was dropped

Added items regarding dietary change, supplement and
herbal remedy use.
The Food Frequency Questionnaire
Dietary Assessment
Interviewers were
trained by J. Shannon
 Pilot dietary data were
collected from 100
retrospective cases.
 Reviewed for face
validity

Blood specimens
Pre-biopsy blood
specimens were
collected.
 Processed for
assessment of
antioxidants, fatty acids
& DNA
 Stored in Shanghai

Analyses

Daily intake was determined using the reported
frequency and average portion sizes reported on
the Chinese Health and Nutrition Survey (1993).

Individual food intake converted to intake per
month.

Groupings created based on traditional food
groups and botanical groups.
Analyses cont.

Food group intake converted from
continuous variable to categories of
intake- based on control group
consumption.

Why group foods and create categorical
variables for intake??
Analyses cont.

Association between food groupings and breast cancer risk modeled
using conditional logistic regression stratified by year of interview.
◦ ODDS RATIOS (OR) and 95% Confidence Interval

All foods models adjusted for age and total energy, botanical models
adjusted for age and total fruits and vegetables.

Covariates considered for inclusion (maintained in model if change
OR >10%) Only duration of breast feeding maintained.

Trend OR determined entering category score as a continuous
variable and using Wald test of significance.
Blood analyses

Isoflavones (Daidzein/ Genestein) -- liquid

RBC fatty acids analyzed by GCMS @ FHCRC.
Membrane Percent of individual Fatty Acids was
determined.

Levels of fatty acid groups (e.g. total omega-3) were
calculated.

Ratio groups were calculated –
chromatography-coularray method (LC-coularray) and
liquid chromatography-mass spectrometry (LC-MS)
◦ Omega-3:omega-6, Saturation index (Palmitic: Palmitoleic and
Stearic:Oleic)
Associations with Reported Dietary Intake
Shannon J, Ray RM, Wu C, Nelson ZC, Gao DL, Li GD, Wei HY, Lampe JW, Horner N, Abouta JS, Patterson R, Fitzgibbons ED,
Thomas DB. Food and botanical groupings and risk of breast cancer: Cancer Epidemiol Biomarker & Prev 2005;14:81-90.
2.25
Trend OR (95%CI)
2.00
1.75
1.50
1.25
1.00
0.75
0.50
0.25
Total Meat
Milk
Seafood
Eggs
Fruits
Soyfood
Rice
Fried foods
Cured
foods
Desserts
Vegetables
Trend OR and 95% CI for Each quartile Vs. next lowest quartiles of food groups
Conditional Logistic regression adjusted for age, total energy and breastfeeding
Summary of Questionnaire Findings
 A diet high in fruits and vegetables may be protective against
breast cancer.
◦ The association does not appear to be due entirely to any single
botanical group assessed.
 Egg intake may be protective against breast cancer but difficult to
determine what we are actually measuring.
 We found no association between soy food intake and breast
cancer risk in this population of high soy consumers.
Odds ratios (OR) and 95% confidence intervals (CI) of fibrocystic
breast conditions and breast cancer in relation to quartiles of plasma
daidzein and genistein concentrations.
Lampe JW, Nishino Y, Ray RM, Wu C, Li W, Lin MG, Gao DL, Hu Y, Shannon J, Stalsberg H, Porter PL, Frankenfeld CL, Wähälä K, Thomas
DB.Plasma isoflavones and fibrocystic breast conditions and breast cancer among women in Shanghai, China. Cancer Epidemiol Biomarkers Prev.
2007 Dec;16(12):2579-86.
N of women (%)*
Control
FBC
Cancer
FBCs vs.
controls
†
OR 95% CI
Cancer vs.
controls
OR† 95% CI
Cancer vs. FBCs
OR† 95% CI
Daidzein (ng/ml)
Q1
Q2
Q3
Q4
(<6.718)
(6.718-18.515)
(18.515-42.092)
(≥ 42.092)
239 (25.0)
239 (25.0)
239 (25.0)
239 (25.0)
956 (100)
115 (41.4)
82 (29.5)
50 (18.0)
31 (11.2)
278 (100)
57 (32.4)
53 (30.1)
43 (24.4)
23 (13.1)
176 (100)
p trend
1.00
0.61 0.36-1.02
0.28 0.16-0.50
0.24 0.13-0.45
1.00
0.85
0.48
0.23
< 0.0001
0.47-1.56
0.25-0.91
0.12-0.48
1.00‡
1.40 0.52-2.48
1.62 0.50-2.59
1.07 0.30-2.62
<0.0001
0.4388
0.31-0.92
0.33-0.97
0.13-0.50
1.00 a
1.23 0.76-2.00
1.71 0.998-2.93
0.73 0.38-1.41
0.0001
0.8152
Genistein (ng/ml)
Q1
Q2
Q3
Q4
(<9.418)
( 9.418-31.761)
(31.761-76.954)
( ≥ 76.954)
p trend

245 (25.0)
246 (25.0)
245 (25.0)
246 (25.0)
982 (100)
128 (44.0)
74 (25.4)
48 (16.5)
41 (14.1)
291 (100)
73 (38.8)
51 (27.1)
44 (23.4)
20 (10.6)
188 (100)
1.00
0.49 0.30-0.80
0.30 0.18-0.51
0.40 0.23-0.70
< 0.0001
1.00
0.54
0.57
0.26
*missing data were excluded in the analysis†adjusted for age and isoflavone analysis method and stratified by year of blood
draw‡ futher adjusted for the status of proliferative changes
Shannon J, King IB, Lampe JW, Gao DL, Ray RM, Lin M-G, Stalsberg H, Thomas DB.
Erythrocyte fatty acids and risk of proliferative and non-proliferative fibrocystic disease
in women in Shanghai, China. Am J Clin Nutr. 12/2008; e-pub, ahead of print.
Odds Ratio (95% Confidence Interval)
3
2.5
=non-proliferative FCD v. control
=proliferative FCD v. control
=proliferative FCD v. cancer
=cancer v. control
2
1.5
1
0.5
0
Total Omega-3
Fatty Acids
Eicosapentaenoic
Acid
Docosahexaenoic
Acid
OR and 95% CI for Highest Vs. Lowest Quartiles of RBC Fatty Acid.
Conditional logistic regression, adjusting for age, stratified by year of interview.
Summary of RBC Fatty Acid Findings
Normal
1. ↓ risk with total n-3 PUFA, specifically EPA
2. No association with n-3 or n-6
3. ↓ risk with ↑ total n-3 PUFA,
specifically EPA
1
Fibroadenoma
Non-Proliferative
FCD
2
3
4. ↓ risk with total n-3 PUFA, specifically
EPA, DHA, n3:n6 ratio
4
Proliferative FCD
with & without
Atypia
DCIS
Invasive
Cancer
OHSU Cancer
Institute –
pilot funding
OHSU
Comprehensive
Breast Cancer
Clinic
OHSU CoInvestigators
• John Vetto
• Philippe Thuillier
• Shannon
McWeeney
• Rosalie Sears
OCTRI / CTRC
Resources
EPIC Imaging
Collaborators
• Amy Thurmond
• Judith Richmond
• Maureen Filipek

Aim 1: Determine the effect of purified fish oil
supplementation (75% EPA / DHA) on markers of
cancer progression in women newly diagnosed with
DCIS.

Aim2a: Determine the effect of n-3 fatty acids on the
targets identified in Aim 1 in breast cancer cells.


Treatment/ Intervention:
◦ ~2 week supplementation with 2.0 gram EPA/
DHA (ROPUFA 75) or placebo.

Analyses for Primary Endpoints:
◦ Pre- post changes in erythrocyte and NAF
fatty acid levels
◦ Pre- post changes in gene expression using
genome wide array
◦ Pre- post changes in c-myc phosphorylation
and stem cell markers.
Table 3.0. Schedule of Events (w-3 FA and DCIS/ADH)
STUDY VISITS
PreTrial
Test/ Procedure
Initial Biopsy (request tissue
sample)
Screening
Eligibility (Incl./ Excl.
Criteria)
Informed Consent
Randomization
Height, weight, blood
pressure
Baseline/
Registration
Visit
Week
1
Week
2
Week
3
Week
4
Weeks
5-8
Surgery1
or PostIntervention
Appt
X
X
X
X
X
X
X
X
Research Specimens:
Plasma, Serum, Urine, Nipple
Aspirate
X
X
Confirm eligibility:
Urine HcG test
X
Diet & Family History
Questionnaire
X
Questionnaire: Adverse
Events
X
Questionnaire: Changes to
Diet and Medications
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Questionnaire:
Placebo/Supplement
Study Supplement Dispensed
Study
Followup
X
X
* Supplements to be provided only if surgery is delayed due to non-study related concerns
1 Request tissue sample if surgery is completed
X*
Inclusion criteria:

Biopsy confirmed diagnosis of any of the
following:
◦ DCIS or ADH or both
◦ DCIS with a component of invasive
carcinoma
◦ ADH with a component of invasive
carcinoma
◦ DCIS and ADH with a component of
invasive carcinoma

Age over 21 years (no upper age limit)

English or Spanish speaking

Female patients
Exclusion criteria:

Using therapeutic anticoagulation

Pure invasive breast cancer on biopsy without
a component of DCIS or ADH

Pregnancy (as determined by urine hCG test)

Male patients

Patient reported allergy to fish oil or olive oil

Patient reported current use of fish oil greater
than 1 gram per day

Any condition which, in the opinion of the
study clinician, would make participation in the
study harmful to the subject
Recruit and randomize 40 women
Challenges in Diet and Cancer
Prevention Research
The role of genetic variation.
 Incomplete knowledge of compounds in
foods.
 Difficulty in capturing “true” dietary
intake.
 A constantly changing food supply.

Bridging the disciplines
Case-control
Population
Studies Cohort
Correlational
Human Trials
Randomized
Clinical Trials
Intervention Studies
Laboratory Studies
In Vitro
Animal Studies
What Next?
Green Tea
Nutrient Supplements
Soy