Docstoc

Nutrition and breast cancer

Document Sample
Nutrition and breast cancer Powered By Docstoc
					Preventing cancer: the role of
diet and nutrition

 Tim Key
 Cancer Epidemiology Unit
 University of Oxford
Diet and cancer

•   How important is diet?
•   Where do we get the hypotheses?
•   How do we test the hypotheses?
•   What do we know for certain?
Proportion of cancer deaths attributable to environmental and
behavioural factors in the UK

         Tobacco
               Diet
        Hormonal
           Alcohol
        Infections
Ionizing radiation
       Occupation
         Pollution
  Ultraviolet light
                      0   5   10   15    20    25     30    35
                                   Percent
Adapted from Doll and Peto, 2003
           Colorectal cancer incidence among men in
           Britain and Japan, 1960 to 1995
 Age-standardised incidence per 100,000




                                          60                                                    Japan
                                          50

                                          40
                                                                                                Britain
                                          30

                                          20

                                          10

                                          0
                                               1960   1965   1970   1975   1980   1985   1990   1995
                                                                       Year
Data from Cancer Incidence in Five Continents, IARC
Meat consumption in Britain and Japan, 1950s to 1990s
                             180

                             160                 Britain
  Meat intake g/person/day




                             140

                             120

                             100

                              80                                    Japan
                              60

                              40

                              20

                               0
                                   1952   1957   1962   1967   1972 1977   1982   1987   1992
                                                                Year
Data from National Nutrition Survey in Japan and National Food Survey in Britain
Testing hypotheses

 • Observational studies (epidemiology)
   – Limitations include “confounding”


 • Randomized controlled trials
   – Limitations include feasibility (pills/diet)
What do we know for certain?

 • Obesity
   – Cancers of the oesophagus, bowel, kidney,
     breast, uterus
 • Alcohol
   – Cancers of the mouth, throat and oesophagus,
     bowel, liver, breast
 Body mass index and breast cancer risk among postmenopausal
 women: results from the Million Women Study
       Risk relative to BMI 22.5-24.9 (95% CI)
 1.6
 1.4
 1.2
   1
                                         Trend: risk increases by 40% per 10 units
 0.8
 0.6
 0.4
 0.2
   0
           <22.5             22.5-               25-       27.5-           30+
                               Body mass index (kg/sq m)
Data for 5629 incident cancers. Reeves et al, BMJ 2007
  Body mass index and free oestradiol




1,100 women in EPIC study: Key et al, Proc Nutr Soc, 2001
Data for 59,000 cases and 95,000 controls from 53 studies worldwide.
Beral and Collaborative Group on Hormonal Factors in Breast Cancer 2002
Diet composition and cancer risk?

        Fibre
        Meat
        Fruit and vegetables
        Anti-oxidant vitamins
        Folic acid
        Vitamin D
        Calcium
        Selenium
        Fat
        N-3 fatty acids
        Soya
        Green tea
        Broccoli
        etc…….
EPIC: 500,000 men and women in 10 countries
      Tromsø                                               TROMSØ


      Umeå
      Malmö
      Aarhus
      Copenhagen                                                         UMEÅ

      Oxford
      Cambridge
      Potsdam
      Heidelberg
      Utrecht                                       AARHUS
      Bilthoven                                                      MALMÖ
                                                                    COPENHAGEN
      Paris (nationwide)                    CAMBRIDGE    UTRECHT
                                                                      POTSDAM
                                                    BILTHOVEN
                                      OXFORD
      Turin
      Milan
      Florence                                                  HEIDELBERG
      Naples                                            PARIS
      Ragusa
      Oviedo                                    IARC             MILAN
                                                           LYONTURIN
      San Sebastian         OVIEDO
      Pamplona                                                       FLORENCE
      Murcia                SAN SEBASTIAN
      Granada                        PAMPLONA
                                                    BARCELONA             NAPLES
      Athens (nationwide)
                                                                                      ATHENS
                                                MURCIA
                                                                             RAGUSA
                                        GRANADA
      Relative risks, adjusted for age, sex and other relevant factors
1.2

1.0

0.8

0.6

0.4

0.2

0.0
                   Very low Medium low     Medium Medium High Very high

                                Intake of dietary fibre in food

Dietary fibre and risk of colorectal cancer: 1065 cases among
500,000 men and women in EPIC. Lancet 2003;361:1496-501
Meat and colorectal cancer: results from EPIC
       Incidence rate ratio relative to low meat-eaters (95% CI)
   2

1.5

   1
                                                             P for trend=0.03
0.5

   0
           <10        20-40        40-80       80-160        160+
                      Red and processed meat intake, g/day

Data for 1329 incident cancers. Norat, Bingham, Ferrari et al JNCI 2005
 Fruit and vegetables and colon cancer risk
      Relative risk and 95% CI
1.2

1.1

1.0

0.9
                                                                   P trend=0.06
0.8

0.7

0.6
                   <200          200-     400-        600-       800+
                             Total fruit and vegetables, g/day

5838 cases among 756,217 men and women. The Pooling Project of Prospective
Studies, Koushik et al 2007
Fruit and vegetables and prostate cancer:
Fruit and vegetables: data for 1,104 cases among 130,000
data for 1,104
men in EPIC cases among 130,000 men in EPIC
      Relative risk and 95% CI
1.4

1.2
                                              5-a-day=400 g
1.0

0.8

0.6

0.4

0.2

0.0
                     170         244        321        414     634
                                 Fruit and vegetables, g/day
Key et al 2004
Insulin-like growth factor-I (IGF-I) and prostate cancer:
3,700 cases and 5,200 controls in 12 prospective studies
              Risk relative to low est IGF-I concentration (95% CI)
        1.6


        1.4
                                                                        P trend <0.001
        1.2


          1


        0.8
                  Low            2             3             4        High
                                     Fifth of plasma IGF-I

Roddam, Allen, Appleby, Key and the Endogenous Hormones and Prostate Cancer
Collaborative Group, Ann Intern Med October 2008
Randomized controlled trials

•   Beta-carotene and vitamin E do not prevent lung cancer
•   Selenium and vitamin E do not prevent prostate cancer
•   Calcium and vitamin D do not prevent bowel cancer
•   Low fat diet does not prevent breast cancer (?)
•   High fruit and vegetable and low fat intake does not
    improve prognosis in breast cancer

• But:
    – Anti-oestrogenic drugs can prevent breast cancer
    – Anti-androgenic drug can prevent prostate cancer
Future research – how to resolve the uncertainties?
• Better measurements of diet
   Repeated records of food intake
   Biomarkers


• Existing studies
   More collaborative re-analyses to standardize approach and avoid reporting
    biases


• Identification of intermediate factors and mechanisms
   e.g. hormones for breast and endometrial cancer


• Randomized trials
  Previous trials have not identified protective effects
  But trials provide very important evidence
Diet and cancer: current recommendations

 • Avoid overweight and obesity
 • Limit alcohol consumption
 • “Healthy diet”
    – Adequate fibre from cereals, fruit and vegetables
    – Not too much meat
    – [Also limit saturated fat, salt, sugar]


 • Future research may produce further
   recommendations

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:5
posted:8/14/2011
language:English
pages:21