Mammography for Breast Cancer Screening:2003

1
Mammography for Breast Cancer Screening2003

• What clinicians and patients should know about
  mammography
• The major controversies
• Mammography as screening example
• Information to share with patients

2
(No Transcript)
3
Clinical Information Needed for Screening

• Risk and severity of condition being sought
• - Breast cancer incidence mortality
• Effectiveness of screening procedure and
  follow-up treatment in preventing untoward
  outcome
• - Effectiveness of mammography early treatment
  in preventing breast cancer mortality
• Ill effects due to screening
• - False-positive mammograms
• - Possible overdiagnosis with DCIS

4
Chances of Developing and Dying of Breast Cancer
in 10 YearsAmong 1000 Women
5
Risk Factors for Breast Cancer

• Major (RR gt 3.0)
• Increasing age
• Genetic mutation
• Increased breast density
• Atypical hyperplasia on biopsy

6
Risk Factors for Breast Cancer

• Moderate (RR 1.0 3.0)
• Mother or sister with BC
• Increased bone density
• Older age at first birth
• Older age at menopause
• Younger age at menarche
• Benign breast biopsy
• Alcohol
• HRT/Contraceptive pills

7
Protective Factors against Breast Cancer

• Oophorectomy before age 35
• Breast feeding
• Increased parity
• Exercise
• Lean post-menopausal body mass

8
(No Transcript)
9
Effectiveness of Mammography Screening for Breast
Cancer

• 8 RCTs with 500,000 women
• Reduction in BC mortality
• Age 50-69 20 to 35
• Age 40-49 20

10
Controversies

• DDoes mammography work in younger women (under
  age 50)?
• DDoes mammography work at all?

11
Lancet Cochrane Review by Olsen and Gotzsche in
2000 and 2001

• 5 of 8 studies (and part of a 6th) flawed
• Remaining 2 ½ studies showed no effect of
  mammography

12
What were the Fatal Flaws?

• Unequal distribution of characteristics
• E.g., breast lumps in HIP, age in Swedish
  studies, SES in Edinburgh
• Varying numbers of women reported
• Combined Swedish studies showed no overall
  mortality reduction
• Cause of death not always masked (HIP)

13
Answers by Investigators

• Varying numbers
• Age versus dates of birth
• Late exclusion of some ineligible women
• Unequal distribution of characteristics
• Cluster randomization in some studies
• Small absolute differences
• Some differences biased against screening
• Latest update of Swedish studies found decrease
  in overall mortality

14
Does Mammography Work for Women in their 40s ?

• All 8 RCTs showed effect in women 50-69
• Consensus not seen for younger women

15
(No Transcript)
16
Why Would Breast Cancer Screening Work Less Well
in Younger Women?

• Sensitivity of test lower because of denser
  breast tissue
• Faster growth rate of breast cancer in younger
  women
• ? Menopausal status, not age per se

17
Breast Cancer ScreeningImportant Possible
Adverse Effects

• False-positive tests
• Overdiagnosis
• Radiation
• Pain

18
Abnormal Screening Mammograms
NSMF Study Northern CA
Study All ages 40-49
years, 1st screen Abnormal 11
6.3 f-p reading 10.7 5.9
true positive 0.3 0.4 Follow-Up
Proced/ Abnormal Mamm 1.2 1.9
19
Estimated Risk of at Least One False Positive
Screening Mammogram
of Women with gt 1 False Positive Test
(2227)
(2073)
(1843)
(1491)
(1062)
(636)
(300)
(107)
(23)
(0)
Number of Mammograms
20
Predicting Cumulative Risk of False-Positive
Mammograms

• Highest risk woman 98 after 1 mammogram
• Young age (40), estrogen user, 3 previous
  biopsies, family hx of BC, no comparison with
  previous mammogram, 3 yrs between screens,
  radiologist tends to call positive mammograms
• Lowest risk woman 5 after 9 mammograms
• Old age (70), no estrogen, no breast biopsies,
  no fm hx of BC, mammogram compared to previous
  one, 1 yr between screens, radiologist does not
  tend to call positive mammograms

21
Consequences of F-P Mammograms

• Financial adds 33 to cost of screening program
• Personal Causes anxiety among women
• Health-care utilization Increases patient
  visits for non breast-related reasons
• Bottom line Patients do not react well to
  hearing, Your screening test was not quite
  normal.

22
Effect of Education on Anxiety after
False-Positive Mammograms
23
Effect of Immediate Reading on Anxiety after
False-Positive Mammograms
24
Ductal Carcinoma in Situ
300 Incidence 1983-1995

• Criteria for Diagnosis
• Prevalence of DCIS
• Natural History
• Appropriate Treatment

25
DCIS - Prognosis

• Almost all women survive in first 9 years
  (Ernster et al, 1996)
• Recurrent cancers over 8 years (Fisher et al,
  1998)
• Invasive
• All Recurrences Recurrences
• Lumpectomy 26.8 13.4
• Lumpectomy Radiation 12.1 3.9

26
Modern Screening Quandry

• Technology can find lesions that look but dont
  act like cancer in large numbers of people
• We do not know which of these lesions will
  progress to act like cancer
• The quandry - what to do?

27
Approach for Clinicians

• At any age
• Ask about family history of breast and/or
  ovarian cancer.
• (http//www.isds.duke.edu/gp/brcapro.html)

28
Approach for Clinicians

• Women 40-70
• Discuss BC risk and benefits and hazards of
  screening.
• Recommend mammography every 1-2 years between
  ages 50 and 69.
• For all women, take into account individual
  values.
• Record screening decision.
• Women gt 70
• Consider screening if life expectancy at least 10
  years.

29
560
Will experience at least 1 false-positive
mammogram
470
360
Will experience at least 1 needle or open biopsy
190
190
190
See enlargement
Will develop breast cancer
37
28
15
30
37
Will develop breast cancer
28
Will be cured of breast cancers by treatment
regardless of screening
18
15
14
Will have DCIS diagnosed because of mammography
Will be saved by screening mammography
8
7
7
6
4
3
2
Alive 20 years after BC diagnosis Assuming RCTs
have valid results
31
Approach for Clinicians

• Refer patients to experienced mammographers with
  recall rates no more than 10
• Encourage patients to obtain previous studies
• ? Screen at least every 18 months

32
Malpractice Claims for Failure to Diagnose Breast
Cancer

• 1 reason for malpractice claims
• 68 - women lt 50 years old
• 33 - women lt 40 years old
• 59 - patient found a lesion
• 80 - negative or equivocal mammogram
  result
• 54 - MD exam negative
• Take-Home Message Patient complaint is not the
  same as screening
• Close follow-up is important

33
(No Transcript)
34
HIP Example

• OG Criticism
• Women with a history of BC were excluded after
  randomization in an unbalanced way favoring the
  screened group.
• Screened group - All women
• Control group - Only a few
• Because a history of BC confers added risk of
  subsequent BC and BC death, the screened group
  was favored.

35
HIP Example (Continued)

• Method in study
• 60,000 women were randomized to 2 groups.
• Screened group Asked at the first visit for hx
  of previous BC. All women reporting such history
  were excluded.
• Control group Not contacted.
• Both groups Followed for BC occurrence and BC
  death. For any woman diagnosed with breast
  cancer during the study, the medical record was
  reviewed to determine any previous hx of breast
  cancer, and all such women were excluded.

36
HIP Example (Continued)

• Result
• Among women who developed breast cancer during
  the study, none with a previous BC were included
  in either arm of the study. However, overall
  fewer women were excluded in the unscreened
  group. This means that the rates of BC incidence
  and BC deaths would have been (slightly)
  artificially higher in the screened group as
  compared to the control group. Therefore any
  bias would have been against, not for, a
  screening effect.