The Continuous Update Project
Breast cancer survivors
and prostate cancer
World Cancer Research Fund International, London UK
Professor Martin Wiseman FRCP FRCPath FAfN
NIHR BRC Southampton and University of Southampton
Who we are What we do
AICR (1982)
WCRF UK (1990)
WCRF Netherlands (1994)
WCRF Hong Kong (1997)
WCRF International (1999)
Fund research on the relationship of
nutrition, physical activity and body
weight to cancer risk
Interpret the accumulated scientific
literature to derive
Recommendations for Cancer
Prevention
Educate people through our national
Health Information programmes
Advocate effective policies to help
people and populations to reduce
their chances of developing cancer
Breast
Global variation in cancer
incidence
Colorectum
Migration data
0
5
10
15
20
25
30
1960 1965 1970 -
'71
1973 -
'77
1979 -
'82
1983 -
'87
1988 -
'92
1995 1997
Colon
Breast
Cancer Incidence in Japan*
* Per 100,000, world population standard
CancerIncidence
Hanahan & Weinberg (2011) Cell; Hanahan & Coussens (2012) Cancer Cell
Hallmarks of cancer
Two enabling characteristics for acquiring hallmarks
WCRF/AICR EXPERT REPORT
The most authoritative
• New method
• Systematic reviews
• Review of evidence separate from
judgement
• Panel of international experts
• Predetermined criteria for judgements
– Epidemiology
– Mechanisms
• Flexibility
• Continuous update of evidence
NUTRITION AND CANCERS
• ADIPOSITY
– BREAST (PM), COLORECTUM, ENDOMETRIUM,
OESOPHAGUS, PANCREAS, GALLBLADDER, KIDNEY, OVARY,
PROSTATE (ADVANCED)
• PHYSICAL (IN)ACTIVITY
– COLON, BREAST
• MEAT – RED AND PROCESSED
– COLORECTAL
• ALCOHOL
– MPL, BREAST, COLORECTUM, LIVER
• PLANT FOODS (F&V, PULSES, WHOLEGRAINS)
– MPL, OESOPHAGUS, STOMACH, COLORECTAL (DF), LUNG
• BREASTFEEDING
– BREAST (MOTHER), OBESITY (CHILD)
The Panel emphasises the
importance of not smoking and of
avoiding exposure to tobacco smoke
USA UK BRAZIL CHINA
Mouth, pharynx,
larynx
63 67 63 44
Oesophagus 63 71 50 33
Lung 36 33 36 38
Stomach 47 45 41 33
Pancreas 19 15 11 8
Gallbladder 21 16 10 6
Liver 15 17 6 6
Colorectum 50 47 41 22
Breast 33 38 22 11
Ovary 5 4 3 1
Endometrium 59 44 37 21
Prostate (advanced) 11 10 5 4
Kidney 24 19 13 8
Total for these
cancers
30 32 25 24
Total for all cancers 21 24 17 20
Estimates of
cancer
preventability by
appropriate diet,
nutrition,
physical activity
and body fatness
CANCER FREQUENCY AND
AGE
IMPACT OF OFFSPRING SIZE AND GROWTH
ON CANCER RISK
Height and risk
of CVD and cancer
CVD Cancer
Continuous Update Project: CUP
 Same process for systematically
reviewing evidence for 2007 Report
 Researchers at Imperial College London
– CUP database of epi research
– Systematically review the evidence
 Expert Panel
– Draw conclusions
– Make recommendations
CUP database
Research Team
Second Expert Report databases
Grading criteria
Predefined requirements for
 Number and types of studies
 Quality of exposure and outcome assessment
 Heterogeneity within and between study types
 Exclusion of chance, bias or confounding
 Biological gradient
 Evidence of mechanisms
 Size of effect
Grading the evidence
Decreases
risk
Increases
risk
Strong
evidence
Convincing
Probable
Limited
evidence
Limited -
suggestive
Limited – no
conclusion
Strong
evidence
Substantial
effect on risk
unlikely
Basis for recommendations
Year Publication
2010 Breast
2011 Colorectum
2012 Pancreas
2013-14 Endometrium, ovary, breast cancer survivors, prostate
2015 Liver, bladder, kidney, gallbladder
2015-16 Stomach, oesophagus, lung
2016-17 Mouth, pharynx, larynx, nasopharynx, breast, colorectum
2017 Review Recommendations for Cancer Prevention
Publications timetable
Prostate cancer
Prostate cancer – sub-types
 Non-advanced
 Advanced
 Fatal
Prostate cancer – advanced
 Advanced
– AJCC 1992 stage 3-4
– Advanced
– Advanced or metastatic
– Metastatic
– Whitmore/Jewett 3-4
– Fatal (prostate specific)
– High stage or grade
– Gleason 7 or more
Issues for prostate cancer
Heterogeneity of disease
Screen detected
Advanced or aggressive
Fatal
Characterisation of exposure
Lycopene
Calcium/dairy
Breast Cancer Survivors
Breast Cancer Survivors Included
 Pre-menopausal women
 Post-menopausal women
 Incident in-situ breast cancer
 Incident invasive breast cancer
Criteria for Study Inclusion
 Randomised controlled trials
≥ 50 women
≥ 6 months follow-up
Only 2 identified (both low-fat diet trials)
 Prospective cohort (follow-up) studies
 Primary analysis, secondary analysis or
ancillary analysis of randomised controlled
trials, or follow-up studies in breast cancer
survivors
Exposures Included
 Specific foods
 Micro- and macro-nutrients
 Dietary patterns
 Alcohol
 Overweight, obesity, underweight, weight
change, BMI
 Body composition
 Dietary supplements
 Physical activity
Timing of Exposures
Pre-diagnosis
 Within 12 months following diagnosis
Period of intensive primary therapy (surgery,
radiation, chemotherapy)
 From 12 months after diagnosis
Patients may be receiving targeted therapy
during this period (e.g. hormonal therapy,
HER2neu, bone metastases prevention)
Data may extend 20 years or more
Outcomes Included
 Total mortality
 Breast cancer specific mortality
 Second primary breast cancer
Outcomes Included
 Total mortality
 Breast cancer specific mortality
 Second primary breast cancer
 Not:
 recurrence, QoL, lymphoedema etc
Literature search
(New search for CUP)
19831 unique records identified in Pubmed
and Embase until 30th June 2012 and 18
articles found in handsearch
897 full-text articles retrieved and
assessed for inclusion
319 articles on survival and health events
in women with breast cancer
18952 records excluded on the basis of
title and abstract
578 articles excluded for not fulfilling the
inclusion criteria
85 no original data
278 did not report on the associations
of interest
30 abstract/commentary
9 meta-analyses
94 irrelevant study design
33 follow-up less than 6 months
49 study smaller than 50 women
213 articles have mortality or any second
primary cancer as study endpoints
106 articles excluded on health events
other than death or second primary
cancer in women with breast cancer
Issues Relevant to
Survivor Research
Confounding effects of
 Treatment types, efficacies, adverse effects
 Stage of disease
 Comorbidities
 Type of cancer
Methodological
 Determining cause of death
 Screening for second primary breast cancer
 Increasing length of survival
• Various associations also found for:
– Physical activity
– Foods containing fibre
– Foods containing soy
– Total fat
– Saturated fatty acids
• Evidence not judged strong enough to
conclude causal effects
Need for studies that control for
confounding by clinical variables
Randomised controlled trials
Cohort studies with accurate diagnostic &
treatment variables
Future Considerations
Summary
 Although there were significant
associations between some exposures
and outcomes, incomplete adjustment for
potential confounders restricted the ability
to ascribe causality.
 CUP Panel concluded that evidence is
limited.
CONCLUSIONS
• Nutrition (diet, body composition and physical
activity) is key determinant of global cancer patterns
• Evidence from epidemiology and mechanistic data
• CUP is refining knowledge of risk factors
– Heterogeneity of prostate cancer
– Body fatness and advanced prostate cancer
– Lycopene? Calcium/dairy?
– Body fatness and early life events are important for many
cancers
– Poor nutrition is an adverse diagnostic factor in breast
cancer – better studies needed to determine causality
Thank you!
Summary:
Before diagnosis- BMI
Total mortality
Breast cancer
mortality
Second primary
breast cancer
N
deaths
RR (95%CI)
N
deaths
RR (95%CI)
N
events
RR (95%CI)
High vs. Low 8318 1.41 (1.29-1.54) 9854 1.34 (1.23-1.46) 701 1.43 (0.87-2.34)
Underweight vs.
normal weight
4944 1.10 (0.92-1.31) 4479 1.02 (0.85-1.21) -
-
Per 5 kg/m2 6261 1.17 (1.13-1.21) 6600 1.17 (1.11-1.24) 701 1.21 (1.04-1.40)
Premenopause 644 1.25 (1.10-1.43) 1350 1.06 (0.85-1.32) -
-
Postmenopause 1103 1.16 (1.01-1.34) 2866 1.15 (1.05-1.25) -
-
Summary:
Around diagnosis- BMI
Total mortality
Breast cancer
mortality
Second primary
breast cancer
N
deaths
RR (95%CI)
N
deaths
RR (95%CI)
N
events
RR (95%CI)
High vs. Low
BMI
16925 1.27 (1.16-1.38) 10063 1.36 (1.23-1.50) 3478 1.30 (1.14-1.48)
Premenopause 4604 1.28 (1.16-1.42) 586 0.96 (0.45-2.06) -
-
Postmenopause 4614 1.13 (1.03-1.23) 1067 1.57 (1.31-1.89) -
-
Underweight vs.
normal weight
2598 1.23 (0.93-1.63) 1455 1.52 (1.26-1.84) -
-
Per 5 kg/m2 5875 1.11 (1.06-1.17) 1918 1.18 (1.11-1.25) 3186 1.13 (1.06-1.21)
Summary:
After diagnosis – BMI
Total mortality
Breast cancer
mortality
Second primary
breast cancer
N
deaths
RR (95%CI)
N
deaths
RR (95%CI)
N
events
RR (95%CI)
High vs. Low 2289 1.21 (1.06-1.38)
2 studies
Both  risk, 1 significant
No studies
Underweight vs.
normal weight
1361 1.29 (1.02-1.63)
Per 5 kg/m2 1703
1.08 (1.01-1.15)
4, 0%, p=0.52
Summary: physical activity
Total mortality
Breast cancer
mortality
Events
RR (95%CI)
N, I2, P het
Events
RR (95%CI)
N, I2, P het
Before diagnosis
Total
H vs. L 505
0.83 (0.62-1.12)
2, 23%, p=0.25
338
0.80 (0.59-1.10)
2, 0%, p=0.88
Before diagnosis
Recreational
H vs. L 2892
0.74 (0.67-0.83)
8, 5%, p=0.39
1750
0.76 (0.61-0.95)
7, 49%, p=0.06
After diagnosis
Total H vs. L 514
0.63 (0.41-0.97)
3, 44%, p=0.16
217
0.81 (0.48-1.36)
2, 0%, p=0.63
Per 10 MET-
h/week
514
0.90 (0.79-1.03)
3, 79%, p=0.009
- -
After diagnosis
Recreational
H vs. L 2337
0.61 (0.50-0.74)
5, 46% ,p=12
392
0.71 (0.45-1.12)
2, 33%, p=0.22
After diagnosis
Recreational
Per 10 MET-
h/week
2337
0.81 (0.73-0.90)
5, 64%, p=0.03
- -
Summary: dietary fibre
Total mortality
Breast cancer
mortality
Events
RR (95%CI)
N, I2, P het
Events
RR (95%CI)
N, I2, P het
Before diagnosis H vs. L 417
0.50 (0.35-0.73)
2, 0%, p=0.83
2 studies
Non-sig  risk
Per 10g/d 443
0.68 (0.55-0.84)
3, 0%, p=0.41
After diagnosis H vs. L 1092
0.76 (0.58-0.98)
3, 0%, p=0.99
332
0.82 (0.57-1.20)
3, 0%, p=0.95
Per 10g/d 1092
0.88 (0.78-0.99)
3, 0%, p=0.97
332
0.93 (0.80-1.07)
3, 0%, p=0.64
Summary: isoflavones
Total mortality
Events
RR (95%CI)
N, I2, P het
Before diagnosis H vs. L 624
0.87 (0.65-1.17)
3, 59%, p=0.06
After diagnosis H vs. L 794
0.70 (0.56-0.88)
3, 8%, p=0.33
Per 10mg/d 794
0.91 (0.83-1.00)
3, 68%, p=0.05
Summary: total fat
Total mortality
Breast cancer
mortality
Events
RR (95%CI)
N, I2, P het
Events
RR (95%CI)
N, I2, P het
Before diagnosis H vs. L (g/d) 655
1.87 (0.76-4.57)
3, 90%, p<0.001
521
1.13 (0.71-1.77)
4, 54%, p=0.09
Per 10g/d 178
1.19 (1.01-1.41)
4, 82%, p=0.001
- -
Per 10% energy 178
1.82 (1.41-2.36)
3, 0%, p=0.38
- -
After diagnosis H vs. L (g/d) 1436
1.08 (0.90-1.30)
3, 24%, p=0.27
648
1.19 (0.94-1.50)
4, 0%, p=0.41
Per 10g/d - -
575
1.01 (0.95-1.08)
3, 24%, p=0.27
After diagnosis: saturated fat
Outcome Author
Year
Study design Number of
events
Contrast RR
(95% CI)
Total
mortality
Beasley
2011
Follow-up of
cases from case-
control study
525 13% vs. 7%
energy
1.40 (1.06-1.87)
Total
mortality
Holmes
1999
Cancer survivors
of a cohort study
(NHS)
378 Q5 vs. Q1
g/d
1.23 (0.89-1.69)
Breast
cancer
mortality
Beasley
2011
Follow-up of
cases from case-
control study
137 13% vs. 7%
energy
1.55 (0.88-2.75)
Breast
cancer
mortality
Rohan
1993
Follow up of
cases from case-
control study
112 ≥45 vs <20g/d 1.65 (0.73-3.75)
For each of total mortality and breast cancer
mortality there were only two studies
Summary: Alcohol
Total mortality
Breast cancer
mortality
Second primary/
contralateral
breast cancer
N deaths RR (95%CI) N deaths RR (95%CI)
N
events
RR (95%CI)
Before
diagnosis:
High vs. Low
2650 0.93 (0.82-1.06) 1329 1.18 (0.81-1.72) - -
Before :
Per 1
drink/week
2676 1.00 (0.99-1.00) 1296 1.00 (0.97-1.02) -
-
After
diagnosis:
High vs. Low
3827
0.89 (0.72-1.09)
403 1.22 (0.88-1.69) 2347 1.19 (0.96-1.47)
After
diagnosis:
Per 10g/d
3779 0.98 (0.93-1.03) 403 1.06 (0.79-1.42) 2347 1.01 (0.99-1.03)

The Continuous Update Project | Prof. Martin Wiseman

  • 1.
    The Continuous UpdateProject Breast cancer survivors and prostate cancer World Cancer Research Fund International, London UK Professor Martin Wiseman FRCP FRCPath FAfN NIHR BRC Southampton and University of Southampton
  • 2.
    Who we areWhat we do AICR (1982) WCRF UK (1990) WCRF Netherlands (1994) WCRF Hong Kong (1997) WCRF International (1999) Fund research on the relationship of nutrition, physical activity and body weight to cancer risk Interpret the accumulated scientific literature to derive Recommendations for Cancer Prevention Educate people through our national Health Information programmes Advocate effective policies to help people and populations to reduce their chances of developing cancer
  • 4.
    Breast Global variation incancer incidence Colorectum
  • 5.
  • 6.
    0 5 10 15 20 25 30 1960 1965 1970- '71 1973 - '77 1979 - '82 1983 - '87 1988 - '92 1995 1997 Colon Breast Cancer Incidence in Japan* * Per 100,000, world population standard CancerIncidence
  • 8.
    Hanahan & Weinberg(2011) Cell; Hanahan & Coussens (2012) Cancer Cell Hallmarks of cancer Two enabling characteristics for acquiring hallmarks
  • 10.
    WCRF/AICR EXPERT REPORT Themost authoritative • New method • Systematic reviews • Review of evidence separate from judgement • Panel of international experts • Predetermined criteria for judgements – Epidemiology – Mechanisms • Flexibility • Continuous update of evidence
  • 12.
    NUTRITION AND CANCERS •ADIPOSITY – BREAST (PM), COLORECTUM, ENDOMETRIUM, OESOPHAGUS, PANCREAS, GALLBLADDER, KIDNEY, OVARY, PROSTATE (ADVANCED) • PHYSICAL (IN)ACTIVITY – COLON, BREAST • MEAT – RED AND PROCESSED – COLORECTAL • ALCOHOL – MPL, BREAST, COLORECTUM, LIVER • PLANT FOODS (F&V, PULSES, WHOLEGRAINS) – MPL, OESOPHAGUS, STOMACH, COLORECTAL (DF), LUNG • BREASTFEEDING – BREAST (MOTHER), OBESITY (CHILD)
  • 13.
    The Panel emphasisesthe importance of not smoking and of avoiding exposure to tobacco smoke
  • 14.
    USA UK BRAZILCHINA Mouth, pharynx, larynx 63 67 63 44 Oesophagus 63 71 50 33 Lung 36 33 36 38 Stomach 47 45 41 33 Pancreas 19 15 11 8 Gallbladder 21 16 10 6 Liver 15 17 6 6 Colorectum 50 47 41 22 Breast 33 38 22 11 Ovary 5 4 3 1 Endometrium 59 44 37 21 Prostate (advanced) 11 10 5 4 Kidney 24 19 13 8 Total for these cancers 30 32 25 24 Total for all cancers 21 24 17 20 Estimates of cancer preventability by appropriate diet, nutrition, physical activity and body fatness
  • 15.
  • 16.
    IMPACT OF OFFSPRINGSIZE AND GROWTH ON CANCER RISK
  • 17.
    Height and risk ofCVD and cancer CVD Cancer
  • 18.
    Continuous Update Project:CUP  Same process for systematically reviewing evidence for 2007 Report  Researchers at Imperial College London – CUP database of epi research – Systematically review the evidence  Expert Panel – Draw conclusions – Make recommendations
  • 19.
    CUP database Research Team SecondExpert Report databases
  • 20.
    Grading criteria Predefined requirementsfor  Number and types of studies  Quality of exposure and outcome assessment  Heterogeneity within and between study types  Exclusion of chance, bias or confounding  Biological gradient  Evidence of mechanisms  Size of effect
  • 21.
    Grading the evidence Decreases risk Increases risk Strong evidence Convincing Probable Limited evidence Limited- suggestive Limited – no conclusion Strong evidence Substantial effect on risk unlikely Basis for recommendations
  • 22.
    Year Publication 2010 Breast 2011Colorectum 2012 Pancreas 2013-14 Endometrium, ovary, breast cancer survivors, prostate 2015 Liver, bladder, kidney, gallbladder 2015-16 Stomach, oesophagus, lung 2016-17 Mouth, pharynx, larynx, nasopharynx, breast, colorectum 2017 Review Recommendations for Cancer Prevention Publications timetable
  • 23.
  • 24.
    Prostate cancer –sub-types  Non-advanced  Advanced  Fatal
  • 25.
    Prostate cancer –advanced  Advanced – AJCC 1992 stage 3-4 – Advanced – Advanced or metastatic – Metastatic – Whitmore/Jewett 3-4 – Fatal (prostate specific) – High stage or grade – Gleason 7 or more
  • 33.
    Issues for prostatecancer Heterogeneity of disease Screen detected Advanced or aggressive Fatal Characterisation of exposure Lycopene Calcium/dairy
  • 34.
  • 35.
    Breast Cancer SurvivorsIncluded  Pre-menopausal women  Post-menopausal women  Incident in-situ breast cancer  Incident invasive breast cancer
  • 36.
    Criteria for StudyInclusion  Randomised controlled trials ≥ 50 women ≥ 6 months follow-up Only 2 identified (both low-fat diet trials)  Prospective cohort (follow-up) studies  Primary analysis, secondary analysis or ancillary analysis of randomised controlled trials, or follow-up studies in breast cancer survivors
  • 37.
    Exposures Included  Specificfoods  Micro- and macro-nutrients  Dietary patterns  Alcohol  Overweight, obesity, underweight, weight change, BMI  Body composition  Dietary supplements  Physical activity
  • 38.
    Timing of Exposures Pre-diagnosis Within 12 months following diagnosis Period of intensive primary therapy (surgery, radiation, chemotherapy)  From 12 months after diagnosis Patients may be receiving targeted therapy during this period (e.g. hormonal therapy, HER2neu, bone metastases prevention) Data may extend 20 years or more
  • 39.
    Outcomes Included  Totalmortality  Breast cancer specific mortality  Second primary breast cancer
  • 40.
    Outcomes Included  Totalmortality  Breast cancer specific mortality  Second primary breast cancer  Not:  recurrence, QoL, lymphoedema etc
  • 41.
    Literature search (New searchfor CUP) 19831 unique records identified in Pubmed and Embase until 30th June 2012 and 18 articles found in handsearch 897 full-text articles retrieved and assessed for inclusion 319 articles on survival and health events in women with breast cancer 18952 records excluded on the basis of title and abstract 578 articles excluded for not fulfilling the inclusion criteria 85 no original data 278 did not report on the associations of interest 30 abstract/commentary 9 meta-analyses 94 irrelevant study design 33 follow-up less than 6 months 49 study smaller than 50 women 213 articles have mortality or any second primary cancer as study endpoints 106 articles excluded on health events other than death or second primary cancer in women with breast cancer
  • 45.
    Issues Relevant to SurvivorResearch Confounding effects of  Treatment types, efficacies, adverse effects  Stage of disease  Comorbidities  Type of cancer Methodological  Determining cause of death  Screening for second primary breast cancer  Increasing length of survival
  • 46.
    • Various associationsalso found for: – Physical activity – Foods containing fibre – Foods containing soy – Total fat – Saturated fatty acids • Evidence not judged strong enough to conclude causal effects
  • 49.
    Need for studiesthat control for confounding by clinical variables Randomised controlled trials Cohort studies with accurate diagnostic & treatment variables Future Considerations
  • 50.
    Summary  Although therewere significant associations between some exposures and outcomes, incomplete adjustment for potential confounders restricted the ability to ascribe causality.  CUP Panel concluded that evidence is limited.
  • 51.
    CONCLUSIONS • Nutrition (diet,body composition and physical activity) is key determinant of global cancer patterns • Evidence from epidemiology and mechanistic data • CUP is refining knowledge of risk factors – Heterogeneity of prostate cancer – Body fatness and advanced prostate cancer – Lycopene? Calcium/dairy? – Body fatness and early life events are important for many cancers – Poor nutrition is an adverse diagnostic factor in breast cancer – better studies needed to determine causality
  • 52.
  • 54.
    Summary: Before diagnosis- BMI Totalmortality Breast cancer mortality Second primary breast cancer N deaths RR (95%CI) N deaths RR (95%CI) N events RR (95%CI) High vs. Low 8318 1.41 (1.29-1.54) 9854 1.34 (1.23-1.46) 701 1.43 (0.87-2.34) Underweight vs. normal weight 4944 1.10 (0.92-1.31) 4479 1.02 (0.85-1.21) - - Per 5 kg/m2 6261 1.17 (1.13-1.21) 6600 1.17 (1.11-1.24) 701 1.21 (1.04-1.40) Premenopause 644 1.25 (1.10-1.43) 1350 1.06 (0.85-1.32) - - Postmenopause 1103 1.16 (1.01-1.34) 2866 1.15 (1.05-1.25) - -
  • 55.
    Summary: Around diagnosis- BMI Totalmortality Breast cancer mortality Second primary breast cancer N deaths RR (95%CI) N deaths RR (95%CI) N events RR (95%CI) High vs. Low BMI 16925 1.27 (1.16-1.38) 10063 1.36 (1.23-1.50) 3478 1.30 (1.14-1.48) Premenopause 4604 1.28 (1.16-1.42) 586 0.96 (0.45-2.06) - - Postmenopause 4614 1.13 (1.03-1.23) 1067 1.57 (1.31-1.89) - - Underweight vs. normal weight 2598 1.23 (0.93-1.63) 1455 1.52 (1.26-1.84) - - Per 5 kg/m2 5875 1.11 (1.06-1.17) 1918 1.18 (1.11-1.25) 3186 1.13 (1.06-1.21)
  • 56.
    Summary: After diagnosis –BMI Total mortality Breast cancer mortality Second primary breast cancer N deaths RR (95%CI) N deaths RR (95%CI) N events RR (95%CI) High vs. Low 2289 1.21 (1.06-1.38) 2 studies Both  risk, 1 significant No studies Underweight vs. normal weight 1361 1.29 (1.02-1.63) Per 5 kg/m2 1703 1.08 (1.01-1.15) 4, 0%, p=0.52
  • 57.
    Summary: physical activity Totalmortality Breast cancer mortality Events RR (95%CI) N, I2, P het Events RR (95%CI) N, I2, P het Before diagnosis Total H vs. L 505 0.83 (0.62-1.12) 2, 23%, p=0.25 338 0.80 (0.59-1.10) 2, 0%, p=0.88 Before diagnosis Recreational H vs. L 2892 0.74 (0.67-0.83) 8, 5%, p=0.39 1750 0.76 (0.61-0.95) 7, 49%, p=0.06 After diagnosis Total H vs. L 514 0.63 (0.41-0.97) 3, 44%, p=0.16 217 0.81 (0.48-1.36) 2, 0%, p=0.63 Per 10 MET- h/week 514 0.90 (0.79-1.03) 3, 79%, p=0.009 - - After diagnosis Recreational H vs. L 2337 0.61 (0.50-0.74) 5, 46% ,p=12 392 0.71 (0.45-1.12) 2, 33%, p=0.22 After diagnosis Recreational Per 10 MET- h/week 2337 0.81 (0.73-0.90) 5, 64%, p=0.03 - -
  • 58.
    Summary: dietary fibre Totalmortality Breast cancer mortality Events RR (95%CI) N, I2, P het Events RR (95%CI) N, I2, P het Before diagnosis H vs. L 417 0.50 (0.35-0.73) 2, 0%, p=0.83 2 studies Non-sig  risk Per 10g/d 443 0.68 (0.55-0.84) 3, 0%, p=0.41 After diagnosis H vs. L 1092 0.76 (0.58-0.98) 3, 0%, p=0.99 332 0.82 (0.57-1.20) 3, 0%, p=0.95 Per 10g/d 1092 0.88 (0.78-0.99) 3, 0%, p=0.97 332 0.93 (0.80-1.07) 3, 0%, p=0.64
  • 59.
    Summary: isoflavones Total mortality Events RR(95%CI) N, I2, P het Before diagnosis H vs. L 624 0.87 (0.65-1.17) 3, 59%, p=0.06 After diagnosis H vs. L 794 0.70 (0.56-0.88) 3, 8%, p=0.33 Per 10mg/d 794 0.91 (0.83-1.00) 3, 68%, p=0.05
  • 60.
    Summary: total fat Totalmortality Breast cancer mortality Events RR (95%CI) N, I2, P het Events RR (95%CI) N, I2, P het Before diagnosis H vs. L (g/d) 655 1.87 (0.76-4.57) 3, 90%, p<0.001 521 1.13 (0.71-1.77) 4, 54%, p=0.09 Per 10g/d 178 1.19 (1.01-1.41) 4, 82%, p=0.001 - - Per 10% energy 178 1.82 (1.41-2.36) 3, 0%, p=0.38 - - After diagnosis H vs. L (g/d) 1436 1.08 (0.90-1.30) 3, 24%, p=0.27 648 1.19 (0.94-1.50) 4, 0%, p=0.41 Per 10g/d - - 575 1.01 (0.95-1.08) 3, 24%, p=0.27
  • 61.
    After diagnosis: saturatedfat Outcome Author Year Study design Number of events Contrast RR (95% CI) Total mortality Beasley 2011 Follow-up of cases from case- control study 525 13% vs. 7% energy 1.40 (1.06-1.87) Total mortality Holmes 1999 Cancer survivors of a cohort study (NHS) 378 Q5 vs. Q1 g/d 1.23 (0.89-1.69) Breast cancer mortality Beasley 2011 Follow-up of cases from case- control study 137 13% vs. 7% energy 1.55 (0.88-2.75) Breast cancer mortality Rohan 1993 Follow up of cases from case- control study 112 ≥45 vs <20g/d 1.65 (0.73-3.75) For each of total mortality and breast cancer mortality there were only two studies
  • 62.
    Summary: Alcohol Total mortality Breastcancer mortality Second primary/ contralateral breast cancer N deaths RR (95%CI) N deaths RR (95%CI) N events RR (95%CI) Before diagnosis: High vs. Low 2650 0.93 (0.82-1.06) 1329 1.18 (0.81-1.72) - - Before : Per 1 drink/week 2676 1.00 (0.99-1.00) 1296 1.00 (0.97-1.02) - - After diagnosis: High vs. Low 3827 0.89 (0.72-1.09) 403 1.22 (0.88-1.69) 2347 1.19 (0.96-1.47) After diagnosis: Per 10g/d 3779 0.98 (0.93-1.03) 403 1.06 (0.79-1.42) 2347 1.01 (0.99-1.03)