Breast Cancer Screening: Handling the Trade-Offs

1
Breast Cancer Screening Handling the Trade-Offs

• Lisle Nabell MD
• Associate Professor of Medicine
• UAB Comprehensive Cancer Center

2
Ten Leading Cancer Types for Estimated New Cancer
Cases and Deaths, by Sex, United States, 2009
From Jemal, A. et al. CA Cancer J Clin
200959225-249.
3
Age-specific Breast Cancer Incidence Among
Females by Age Group and Race/Ethnicity, United
States, 2002 to 2004
From Mahoney, M. C. et al. CA Cancer J Clin
200858347-371.
4
Breast Cancer Disease Progression
0 5 10
Years of growth
1012 1010 108 106 104 102
10 cm
Very early breast cancer (undetectable)
Clinicalbreast cancer
1 cm
Number of cells
DCIS
1 mm
0 5 10 15 20 25 30 35 40
Number of cell doublings
DCIS Ductal carcinoma in situ. Note 90-day
doubling ? 40 doublings 3,600 days
(approximately 10 years). Harris JR, et al, eds.
Breast Diseases, 2nd ed. Philadelphia JB
Lippincott 1991165-189.
5
Mammography Screening for Breast Cancer

• Screening mammogram entails 2 x-ray images of
  each breast one taken from the top (craniocaudal
  or CC) and the other from the side (mediolateral
  oblique MLO)
• Randomized trials suggest an overall sensitivity
  of 75, lower in women with dense breast tissue
  and in women harboring BRCA mutations
• Digital mammography allows for digital storage
  and manipulation of images may be somewhat
  better with regard to detections of lesions in
  younger women

6
Mammography Screening for Breast Cancer

• Single-view mammography is less sensitive than
  2-view mammography. In a UK NHS comparison,
  sites using single-view films were 19 less
  successful in detecting breast cancer compared to
  2-view (76 versus 95) especially among
  smaller cancers

Young et al Br J Raadiol 199770482-488 Blanks
et al Clin Radiol 200560674-680
7
Breast density patterns
Cummings, S. R. et al. J. Natl. Cancer Inst. 2009
101384-398
8
Illustration of the quantitative estimation of
breast density from a digitized image of a
mammogram
Cummings, S. R. et al. J. Natl. Cancer Inst. 2009
101384-398
9
Magnetic Resonance Imaging (MRI)

• MRI utilizes magnetic fields to produce
  cross-sectional images of tissue structures
  requires dedicated breast MRI coils and
  experience
• Overall, the sensitivity of MRI is superior to
  mammography but specificity is lower, resulting
  in a higher false-positive rate
• Strongest data for use is in high-risk women

Leach et al Lancet. 20053651769-1778 Port et al
Ann Surg Oncol. 2007141051-1057. Kuhl et al JCO
2010281450-1457.
10
Prospective Study of Women with Elevated Breast
Cancer Risk The EVA Trial

• Study of 687 women at elevated risk of breast
  cancer development
• Annual screening with CBE, mammography, US, and
  MRI over a 3 year time period
• Mean age of 44 72 were premenopausal
• 63 had a family history of breast cancer
• 27 of these women (3.9) developed breast cancer
  during the study period with 21/27 having Tis or
  invasive cancer lt 10 mm

11
Cancer yield of the different imaging methods,
used alone or in combination
Kuhl, C. et al. J Clin Oncol 281450-1457 2010
12
ACS Guidelines for Screening

• Average Risk Women
• Clinical breast exam every 1-3 years starting at
  age 20 and annually after age 40
• Annual mammography beginning at age 40 no upper
  age limit
• Breast awareness

Consensus Statement from ACS and NCI in 1997 with
updates
13
ACS Guidelines for Screening

• High Risk Women
• Definition Life-time risk of Breast Cancer of
  20
• Known or suspected inherited susceptibility for
  breast/ovarian cancer
• Women over age 35 with increased risk of breast
  cancer development
• Having undergone mantle irradiation at a young
  age for Hodgkins (10-30 years of age) maybe
  less with current RT techniques

14
Cumulative incidence (middle line) of breast
cancer as a function of age of the cohort of
female survivors of Hodgkin's disease, with 95
CIs (upper and lower curves)
Bhatia, S. et al. J Clin Oncol 214386-4394 2003
15
Estimating Risk of Cancer Development or BRCA1/2
Status

• Risk Factor Models
• Gail model using age, race, personal history of
  atypical hyperplasia, reproductive history,
  number of biopsies components are in
  www.cancer.gov/bcrisktool
• Claus model includes information on up to 2
  first/second degree relatives with breast cancer,
  age of onset most helpful for women with a
  family history
• Cuzick-Tryer model assesses FH, hormonal factors,
  low-penetrantgenes and benign breast disease J
  Med Genetics 200340807-814.
• BRCAPRO incorporates data on BRCA1/2 mutation
  prevalence http//astor.som.jhmi.edu/BayesMendel/
  brcapro.html
• BOADICEA model of genetic susceptibility in Br J
  Cancer 2004911580-1590

16
ACS Guidelines for Screening High Risk Women

• Known or Suspected Mutation Carrier or Radiation
  Exposure or Elevated Lifetime Risk gt 20
• Annual screening mammography and MRI beginning
  at age 30
• Clinical breast exam every 6-12 months
• Breast awareness

17
Guidelines from the US Preventative Services Task
Force (USPSTF)

• Independent body established in 1984 comprised of
  nonfederal experts in primary care and preventive
  medicine
• The USPSTF examined the effectiveness of 5
  screening modalities with regard to benefit and
  harm of screening and used population modeling to
  compare expected health outcomes and resource
  requirements

18
Guidelines from the US Preventative Services Task
Force (USPSTF)

• Rating system
• A, recommendation for the service, with
  expectation of high benefit
• B, recommendation for the service, with
  expectation that the net benefit is moderate
• C, recommendation against routinely providing the
  service
• D, recommendation against the service
• I, insufficient evidence to recommend for or
  against the service

19
Nelson H D et al. Ann Intern Med 2009151727-737
20
(No Transcript)
21
Mammography Screening Trials For Women Age 40-49
Included in Meta-analysis.
Nelson H D et al. Ann Intern Med 2009151727-737
22
Pooled relative risk for breast cancer mortality
from mammography screening trials compared with
control for women aged 39 to 49 years.CNBSS-1
Canadian National Breast Screening Study-1 CrI
credible interval HIP Health Insurance Plan of
Greater New York. Swedish Two-County trial.
Nelson H D et al. Ann Intern Med 2009151727-737
23
Pooled RRs for Breast Cancer Mortality From
Mammography Screening Trials for All Ages.
Nelson H D et al. Ann Intern Med 2009151727-737
24
Age-Specific Screening Results From the BCSC.
Nelson H D et al. Ann Intern Med 2009151727-737
25
Screening for Breast Cancer SummaryUSPSTF Ann
Int Med, 2009151716-726
Population Women aged 40-49 Women aged 50-74 Women aged gt 75 year
Recommendation Do not screen routinely. Individualize decision to begin biennial screening according to patients context and values Grade C Screen every 2 years Grade B No recommendations Grade I
Risk Assessment This recommendation applies to women gt 40 year who are not at increased risk by virtue of a known genetic mutation or history of chest radiation. Increasing age is the most important risk factor for most women. This recommendation applies to women gt 40 year who are not at increased risk by virtue of a known genetic mutation or history of chest radiation. Increasing age is the most important risk factor for most women. This recommendation applies to women gt 40 year who are not at increased risk by virtue of a known genetic mutation or history of chest radiation. Increasing age is the most important risk factor for most women.
Screening tests Standardization of film mammography has led to improved quality. Refer patients to facilities certified under the MQSA Standardization of film mammography has led to improved quality. Refer patients to facilities certified under the MQSA Standardization of film mammography has led to improved quality. Refer patients to facilities certified under the MQSA
Timing of screening Evidence indicates that biennial screening is optimal. A biennial schedule preserves most of the benefit of annual screening and cuts the harms nearly in half. A longer interval may reduce the benefit. Evidence indicates that biennial screening is optimal. A biennial schedule preserves most of the benefit of annual screening and cuts the harms nearly in half. A longer interval may reduce the benefit. Evidence indicates that biennial screening is optimal. A biennial schedule preserves most of the benefit of annual screening and cuts the harms nearly in half. A longer interval may reduce the benefit.
Balance of harms and benefits There is convincing evidence that screening with film mammography reduces breast cancer mortality, with a greater absolute reduction for women aged 50-74 years than for younger women. Harms of screening include psychological harms, additional medical visits, imaging, and biopsies in women without cancer, inconvenience due to false-positive screening results, harms of unnecessary treatment, and radiation exposure. Harms seem moderate for each age group. There is convincing evidence that screening with film mammography reduces breast cancer mortality, with a greater absolute reduction for women aged 50-74 years than for younger women. Harms of screening include psychological harms, additional medical visits, imaging, and biopsies in women without cancer, inconvenience due to false-positive screening results, harms of unnecessary treatment, and radiation exposure. Harms seem moderate for each age group. There is convincing evidence that screening with film mammography reduces breast cancer mortality, with a greater absolute reduction for women aged 50-74 years than for younger women. Harms of screening include psychological harms, additional medical visits, imaging, and biopsies in women without cancer, inconvenience due to false-positive screening results, harms of unnecessary treatment, and radiation exposure. Harms seem moderate for each age group.
26
The Age Trial

• The new trial added to the group (Age Trial)
  studied women aged 39 to 41 were invited to
  annual mammography or usual care through age 48
  (Lancet 20063682053-2060)
• A study of breast cancer mortality in 23 NHS
  centers in UK, Scotland, and Wales
• First mammogram performed utilized two-view film,
  was not digital
• Subsequent annual mammogram was performed in MLO
  only
• Compliance was relatively poor with 41 of the
  invited group attending all screening rounds

27
The Age Trial

• 10 years of follow-up demonstrated a RR of 0.83
  or 17 reduction in mortality in women invited to
  screen (CI of 0.66-1.04)
• Those who attended screening the RR was 0.76 with
  CI of 0.51-1.01

28
Randomized Controlled Trial Assessment

• Mortality reduction from breast cancer is
  assessed in the intervention group as compared to
  controls is this the optimal goal?
• Should other data outside of randomized
  controlled trials that assess efficacy be
  employed to avoid noncompliance or contamination?
  
• Is mammography the best screening tool? Should
  there be a standard for usage?
• What is the chance of being diagnosed with breast
  cancer if you are a 40 year old woman?

29
Breast Cancer Detection Rate from a Single
Mammogram/1000 Breast Cancer Surveillance
Consortium (BCSC) 1996-2007
Age (years) Cancer Detection Rate
40-44 1.69
45-49 2.60
50-54 3.23
55-59 4.20
60-64 4.70
65-69 5.25
70-74 5.95
75-89 6.96
Any age 4.00
http//breastsceening.cancer/gov/data/performance
30
Mammography as the Standard

• Mammography alone will miss approximately 20 of
  cancers rates that are further decreased in
  women with dense breast tissue
• Digital mammography results in better tissue
  contrast, improving cancer detection
• MRI may prove to be superior but is unproven in
  women of average risk and while more sensitive,
  has a higher recall rate

Pisano et al NEJM 20053531773-1783
31
Mammography as the Standard
Event Number per 1000 screened
Recall 80-100
False positive 45-65
6 month review 20
Biopsy 15
Breast Cancer Diagnosis 2-5
Rosenburg et al Radiology 200624155-66
32
Lifetime Risk of Breast Cancer Following Mammogram
Age at Exposure (yrs) Lifetime Incidence of Excess Breast Cancer per 100,000 Excess Breast Cancer Mortality per 100,000
20 17 4
30 10 2
40 5.6 1.4
50 2.8 0.8
60 1.2 0.4
70 0.5 0.2
80 0.2 0.1
BEIR VII, Phase 2 Washington DC National
Academies Press 2006
33
Observational Studies that Support Mammography

• Observational Studies of women who undergo
  screening mammography have been performed in
  Sweden and British Columbia
• In Sweden, screening may occur at age 40 or age
  50 once started it continues every 18 months
  through age 54
• Across all age groups, with 20 years of
  follow-up, there was a 44 decrease in risk of
  death from breast cancer

Coldman et al Int J Cancer 2007120(5)1076-1080 T
abar et al Lancet 20033611405-1410
34
Cumulative 20-year Mortality from Incident Tumors
in Women aged 40-69
Tabar et al Lancet 20033611405-1410
35
Observational Studies

• These observational studies are biased by
  self-selection for screening, and the possibility
  that smaller tumors are detected (equating longer
  survival)
• While neither the Sweden nor British Columbia
  trials separate mammography from CBE effects,
  both suggest a reduction of 30-40 in women aged
  40-49

Coldman et al Int J Cancer 20071201076-1080 Taba
r et al Lancet 20033611405-1410
36
Harms Associated with Mammography Screening

• Radiation Exposure
• Screening 2-view films average 7 mGy of
  low-energy radiation.
• Overdiagnosis
• Pain During Procedure
• Anxiety, Distress
• False-Positive and False-negative Results

37
USPSTF Breast Cancer Screening Recommendations

• The USPSTF did not recommend against women having
  mammograms The USPSTF recommends against
  routine screening mammography in women aged 40-49
  years. The decision to startshould be an
  individual one and take patient context into
  account, including the patients values regarding
  specific benefits and harms.
• USPSTF changed the recommended screening interval
  for women 50-74 from 1-2 years to biennial
  screening and noted that there was insufficient
  evidence for or against screening women gt 75
  years (I level).

USPSTF Ann Int Med, 2009151716-726
38
USPSTF Breast Cancer Screening Recommendations

• The USPSTF did not oppose insurance coverage for
  mammography
• The USPSTF did not oppose breast self-examination
  but recommended against teaching standardized
  examination procedures (D rating)

39
Analysis of the Data Set

• Methodology of the randomized controlled trials
  used in the meta-analysis used only fair data
  troubled by noncompliance contamination with
  the Age trial adding a substandard method of
  screening
• CBE will add little and rightly was not
  recommended
• Majority of women have no clear identifiable risk
  factors for breast cancer development

40
Benefits and Harms of Mammography Screening

• Take the test, not the chance has become the
  mantra for advocacy groups, physician report
  cards, celebrities and others
• How do we balance the trade-offs?

41
Summary of Benefits and Harms
Risks by Age, years Risks by Age, years
40-49 50-59
Benefit Reduced 10-y chance of dying from breast cancer No screening Screening Avoid breast cancer death because of screening 3.5/1000 3.0/1000 0.5/1000 5.3/1000 4.6/1000 0.7/1000
Harms of screening False-positive test requiring a biopsy 60-200/1000 50-200/1000
Overdiagnosis unnecessary treatment 1-5/1000 1-7/1000

Mandelblatt et al Ann Int Med 2009151(10)738-747
Elmore JG et al N Engl J Med 1998338(16)1089-10
96 Woloshin et al JAMA 2010303(2)164-165
42
Assessing Breast Cancer Risk
Factors Relative Risk
BRCA1/2 deleterious mutation Lifetime risk of 50-80
Radiation to chest during adolescence 30-50 risk by age 50
Family history in first degree relative 1.5-3.3
LCIS or ADH on a biopsy 1.9-2.5
Estrogen Factors After age 30 at first live birth Early menses or late menarche Post-menopausal obesity (BMI 80th) Exogenous hormones (estrogenprogesterone) Physical activity after menopause Alcohol consumption of over 2 drinks daily 1.7-1.9 1.3-1.5 1.2-1.9 1.2 20 decrease in risk 1.2
Breast Density on a mammogram ?
25-Hydroxyvitamin D levels ?
43
Factors Influencing Breast Cancer Development
Breast Cancer Risk

• Modifiable
• Diet
• Body mass index
• Exercise
• Smoking
• Exogenous estrogen usage
• Alcohol consumption

• Not Modifiable
• Genetics/Family History
• Radiation as an adolescent
• Atypical Ductal Hyperplasia
• LCIS
• Age
• Race
• Ethnicity
• Age at menarche

• Potentially Modifiable
• Age at first birth
• Breast feeding

44
Prevention of Breast Cancer in Women Reducing
Risk in Modifiable Factors

• Surveillance
• Chemoprevention options with SERMs (tamoxifen or
  raloxifene)
• Prophylactic surgery
• Lifestyle modification

45
ASCO Tech Assessment 2009
Agent 2009 recommendations
Tamoxifen 20 mg/day for 5 years Offered to reduce the risk of ER-positive invasive breast cancer for premenopausal and postmenopausal women with a 5-year projected risk of BC gt 1.66 or women with LCIS Not recommended for women with a prior history of DVT, PE, stroke, or TIA F/U should include annual gynecologic examination and timely work-up of abnormal vaginal bleeding Use of concomitant HRT not recommended
Raloxifene 60 mg/day for 5 years Offered to reduce risk of ER-positive invasive breast cancer in postmenopausal women with a 5 year projected risk of BC gt 1.66, or with LCIS. Should not be used in premenopausal women. Not recommended in women with a history of DVT, PE, stroke, or TIA

• NCI breast cancer risk assessment tool
  www.cancer.gov/bcrisktool
• ASCO 2009 Prevention Guidelines May 26, 2009

46
Current Randomized Trials for Br Ca Prevention
Study Entry Criteria Intervention Target Accrual Status
Hormone Replacement Therapy Opposed by Low-Dose Tam (HOT) Postmenopausal healthy women using HT at increased risk for breast cancer HRT plus tamoxifen versus placebo for 5 yrs 8,500 Started 12/06
International Breast Intervention Study-II (IBIS-II) Women at risk for breast cancer Anastrozole 1 mg/day versus placebo x 5 years 6,000 Started 12/08
Mammary Prevention Trial or MAP3 (ExCel) gt 35 years of age postmenopausal increased risk of breast cancer Exemestane 25 mg/d versus placebo x 5 years 4560 Started 01/09
Aromasin Prevention Study (ApreS) Postmenopausal, unaffected BRCA1/2 mutation carriers Exemestane 25 mg/d versus placebo x 5 years 666 Started
47
Strategies for Breast Cancer Risk Reduction

• SERM therapy ongoing trials of AI underway
• Prophylactic surgery
• Influence of postmenopausal hormonal therapy
• Diet/Nutrition
• Body size/Body mass index

48
Association of Food/Nutrients and Breast Cancer
Risk
Food Premenopausal Postmenopausal Level of Evidence
Alcohol 5-10 increase risk per 10 gram of ETOH/day 5-10 increase risk per 10 gram of ETOH/day Pooled analysis of 6 prospective studies
Total fat No association Weak positive association for saturated fat Nurses Health Study, pooled analysis of 8 studies
Fruits and vegetables No association No association Pooled analysis of 8 studies
Caffeine No association No association Observational cohort
Vitamin D Reduced risk with high vitamin D Possible reduced risk with high plasma vitamin D Nurses health study, obserational cohort
Soy/phytoestrogen 30 reduced risk with highest intake 20 reduced risk with highest intake Meta-analysis review
49
The Conundrum Obesity, Diet, Alcohol

• There is compelling evidence that a high BMI (BMI
  of 25 kg/m2)in postmenopausal women is
  associated with an increased risk of breast
  cancer
• Estimated 15 of breast cancer within the NHS
  attributable to weight gain after menopause
• Long term weight loss (10 kg or more) in
  menopause may be protective in breast cancer
  development weak links to physical activity
• Biological data suggests that exposure in early
  life may be important in predicting later breast
  cancer risk

50
Survival from Breast Cancer in Physically Active
Women

• Prospective study in 1500 women with early-stage
  breast cancer between 1991-2000 studied by
  dietary and activity recall
• Vegetables and fruits (V/F) were lumped into low
  (lt 5 servings daily) or high (gt5 servings)
• Physical activity (PA) was divided into low (lt540
  metabolic equivalent tasks per week) or high
  (gt540 MET) based on duration, intensity, and
  frequency
• Measured BMI

JCO 2007252345-52.
51
Kaplan-Meier survival after Women's Healthy
Eating and Living (WHEL) Study enrollment by four
diet and physical activity categories
Pierce, J. P. et al. J Clin Oncol 252345-2351
2007
52
What Do We Tell Our Patients?

• Majority of breast cancer is diagnosed in women
  without identifiable risk factors the USPSTF
  recommendation against routine screening is
  problematic
• Mammography is the only method proven to reduce
  breast cancer deaths benefit observed for women
  aged 40-74
• For women over age 75, screening appears
  reasonable as long as good health is maintained
• CBE adds little to mammography but women need to
  be counseled to notice changes in their breast(s)
• Biennial Screening may miss the more aggressive
  tumors in younger women and for optimal benefits,
  a woman in her 40s likely needs annual screening.

53
What Do We Tell Our Patients?

• Keep a healthy body weight through diet and
  exercise goal of not gaining more then 10 lb
  over weight at age 20
• Be moderately to vigorously active for at last 30
  minutes on most days stress this starting at a
  young age
• Little role for HRT after menopause when used,
  estrogen alone has only a modest increase in
  breast cancer risk
• Healthy eating no study demonstrates specific
  dietary component which can reduce breast cancer
  risk
• Moderate alcohol usage
• Chemoprevention drugs are an option but not a
  replacement for activity