Screening Breast MRI: Current Status

1
Screening Breast MRI Current Status

• Brandi Nicholson, M.D.
• Assistant Professor, Mammography
• Department of Radiology

2
Goals

• To understand the risks and benefits of screening
  breast MRI
• To define who is high risk for developing breast
  cancer
• To outline the current suggestions for
  appropriate use of screening breast MRI at the
  University of Virginia
• American Cancer Society

3

• What Makes a Good Screening Test?

4
World Health Organization

• Disease-Specific Criteria for screening tests
• Disease is an important cause of morbidity,
  disability, or mortality
• The natural history of the disease is
  sufficiently well known
• Treatments are sufficiently effective
• There is a recognizable latent stage
• The efficacy of interventions for reducing
  morbidity, disability, or mortality is
  sufficiently high to conclude that screening will
  result in a net benefit

5
Breast Cancer Specifics

• Most common cancer affecting women
• 1 in 8 women will be diagnosed in a lifetime
• Globally 1.1 million new diagnoses and gt370,000
  deaths in 2000 alone
• Highest incidence in US and Northern Europe
• 40,000 US women die each year
• Natural history of disease is very well
  understood and treatment works
• The smaller the tumor the better the prognosis
• gt96 survival with early detection
• If found at Stage IV, 5-yr survival is lt25

6
World Health Organization

• Test-Specific Criteria for screening tests
• Requirements for acceptability of a test for
  patients and their physicians
• Requirements for sensitivity and specificity of
  the test

7
MRI Acceptable Test

• Risks posed by screening MRI are extremely low
• Very rare contrast reactions
• Category C drug
• Nephrogenic Systemic Fibrosis (NSF)
• Pacemakers, aneurysm clips or severe
  claustrophobia make patients ineligible
• Overall studies in high risk women have shown the
  test is acceptable to the target population
• lt1 of patients unable to complete the exam due
  to claustrophobia in two high-risk studies

8

• A Word About NSF

9
Nephrogenic Systemic Fibrosis

• First described in U.S. in 1997
• New, rare, acquired, multisystem fibrosing
  disease that occurs in patients with renal
  failure that affects skin
• Presents with firm, erythematous, indurated
  plaques of the skin associated with subcutaneous
  edema
• Flexion contractures, limited range of motion,
  pain, paresthesias, severe pruritis
• Up to 50 of patients become severely disabled
• Link to gadolinium exposure suggested in January
  2006, warnings June 06-March 07
• Currently no effective treatment
• Diagnosed with skin biopsy

J Am Soc Nephrol 2006172359 AJR
2007188586 Radiology 2007243148
10
NSF Perspective

• 200-500 cases NSF reported worldwide
• gt 200 million patients exposed to gadolinium
  contrast agents since 1980s
• Rare disease that does not occur in patients
  without renal disease
• Advanced renal disease markedly delays excretion
  of gadolinium agents prolonging exposure to
  potential toxic effects

11
NSF Necessary Factors
AJR 2007188586 Radiology 2007243148 CJASN
20072264 CJASN 20072200

• Renal impairment
• Exposure to IV gadolinium agents
• Concurrent proinflammatory condition
• Major surgery
• Infection
• Vascular event - thrombosis, ischemia
• Report of 2 kidney transplant patients who
  developed NSF without Gd exposure

Am J Transplant 2007102425
12
NSF Prevent by Screening

• Age over 70 years
• History of renal disease
• History of diabetes
• History of paraproteinemia syndromes such as
  multiple myeloma
• History of collagen vascular disease
• History of cardiovascular disease

Serum Creatinine Calculated GFR within 90 days
Ann Pharmacother 2007411481
13
NSF ACR Recommendations

• GFR gt 90 No special treatment
• GFR gt 60 No Omniscan
• Lab screening not warranted
• Query kidney disease or dialysis
• GFR 30-60 assess risk vs. benefit
• No Omniscan
• Use lowest dose
• GFR lt 30 or on dialysis at high risk for NSF

Kanal et al. ACR guidance document for safe MR
practices 2007. AJR 20071881447-1474
14
NSF UVa Policy

• Obtain labs on Pts for Gd administration using
  same criteria as for Iodinated Contrast
• Serum Cr, Calculated GFR
• GFR gt 60 Give Gd as indicated
• GFR 30-60
• Recheck need for Gd
• Patient completes a self consent form
• GFR lt 30 or Pt on dialysis
• Do not use Gd unless dire urgency
• Faculty to Faculty discussion

15

• Back to MRI
• Test Acceptability

16
MRI Sensitivity and Specificity

• Results of 8 trials in Germany, Canada, Italy,
  The Netherlands, the United Kingdom, and the US
  have been published in the last decade
• In the largest 4 prospective studies
• 71-100 sensitivity (vs. 25-40 mammo in same
  population)
• 81-97 specificity
• Additional cancer yield 8-67/1000 women screened
  with MRI over mammo alone (average 31/1000)
• Typically screening average risk mammo detected
  cancers are 3-6/1000
• Sensitivity not affected by density

17
MRI Mortality Impact?

• No trials have addressed the impact of this
  increased detection/sensitivity on mortality
• But extrapolate from nodal status at diagnosis
• Nodal ds is a surrogate marker for breast cancer
  mortality
• 4x more likely to die with nodal ds than without
• The Netherlands study found women outside
  screening less likely to have T1N0 cancers (46
  vs. 81) and twice as likely to have
  node-positive ds (42 vs. 19)
• Combining 4 studies, 7 of invasive cancers seen
  on MRI alone were node-positive vs. 28 visible
  on mammography

18

• History of Screening

19
Screening Mammography Impact

• 1990 Death rate fell for first time
• RR 0.85 death reduction compared with unscreened
  population
• 20 decline in death rate over next 10 years
• Is this from screening?
• 50 likely is

20
Screening Mammography Studies

• First study done (1960s)
• Health Insurance Plan of NY
• Death rate reduced with earlier detection
• Breast Cancer Detection Demonstration Project
  (1970s)
• No control group, demonstration project after
  HIP
• Screening detected smaller and earlier stage
  cancers
• Swedish Two-County Trial published in 1985
• 30 decrease in breast cancer deaths

21
Screening Age Issues

• Most feel screening is beneficial
• Sudden increase in breast cancer incidence
  1983-1984 b/c increased numbers of women being
  screened
• DCIS increased abruptly at the same time
• 50 years was picked as cut off in studies to
  represent menopausal status
• No data on menopausal status included however
• 50 years, benefit in 2-3 years
• 40-49 years, benefit delayed to 5-7 years
• 1989 recommendations
• 40-49 yrs q1-2 years
• 50 years q1 year
• 1993 NCI pulled support for 40-49 yrs
• However most felt screening in this population
  could save lives

22
Screening Data for Young Women

• Malmo, Sweden trial
• 35 reduction in younger women
• Gothenburg trial
• 44 reduction in 40-49 years
• Five Swedish trials together
• 29 decrease for younger women
• Overall26 statistically significant
  benefitwhen add HIP and Edinburgh
• 16 benefit when NBSS data included

23
Screening Recommendations for Mammography

• 1997 NCI again supports 40-49 years
• q1-2 years 40
• 2002 US Preventive Services Task Force one of
  the last major advisory groups to support
  screening in younger women
• q1-2 years 40
• American Cancer Society, American College of
  Radiology, and the American Medical Association
• q1 year 40

24
Screening Modalities
US
Clinical Breast Exam does poor job screening for
small, less than 1 cm, tumors Kriege NEJM 2004
CBE 17.8 sensitivity in screening
setting Mammography in 40-50 years old has
sensitivity of 50-80 Mammography in over 50
years has sensitivity of 70-90 (ILC lower than
IDC) Data on US is new RSNA 2007 abstracts and
shows poor specificity
Mammography
MRI
25
MRI Nuts and Bolts

• Uses magnetic fields
• Detailed cross-sectional images
• Good soft tissue contrast
• Brightness of tissue depends on measured signal
  of mobility and magnetic environment of the H
  atoms in H2O and fat
• Need breast coil and parallel imaging to do
  bilateral breasts
• Need contrast (paramagnetic molecule, gadolinium
  based contrast agent)
• Cancers enhance differently than normal breast
  tissue

26
MRI Why Breast Coil?
Body coil
Breast coil
27
MRI Why Fat Suppression?

• Unlike mammography, fat interferes with MRI
• Both fat and cancer are white on MRI
• Two options
• Subtraction (lowers resolution)
• Fat Suppression (takes longer)

No Fat Saturation
With Fat Saturation
28
Axial T2 w/ Fat Sat
T1 Pre w/ Fat Sat
Current UVa studies average 2000 images
T1 Post x5 _at_ 1 min intervals
Sagittal
Subtraction
MIP
29
Screening Why Breast MRI?

• Of all imaging modalities for breast, MRI is the
  most sensitive
• Reproducible across centers as long as contrast
  is used
• 80-100 sensitive
• Better than mammography in each study
• The improvement of MRI over with mammography is
  dependant on density and tumor type
• Misses are typically from obscuration by early
  enhancing normal tissue around the mass
• Sensitivity not affected by density, scar tissue,
  radiation therapy, implants or reconstruction
• IDC is higher sensitivity than DCIS
• But up to 40 of DCIS are seen only with MRI

DCIS
30
Screening Why Breast MRI?

• High negative predictive value
• As high as 98 in single-center trials
• Highest of all imaging modalities
• More recent multicenter trial 85.4
• Finds early cancers
• 1 cm and less, invasive, high grade, node
  negative breast cancers
• Kriege NEJM, 2004

IDC
31

• The Data
• Very High-Risk Patients
• Clinical Risk Factors
• Psychological Effects

32
Screening MRI StudiesVery High Risk Patients

• Mid-late 90s 6 prospective, nonrandomized studies
  in very high-risk women
• MRI sensitivity gt MG
• Interval cancer rate lt10
• Kuhl J Clin Oncol 2005
• 529 women 30 yrs with 20 LTR (5 yrs screening)
• MRI 91 sens vs. MG 33 (spec 97 both together)
• 43 cancers (34 inv) with 1 interval cancer
• 19 node ()
• Sardanelli Eur Radiol 2007
• 278 women 25 yrs 27 were BRCA mutation
  carriers or 1st degree with BRCA (median 1.4
  rounds screening)
• MRI 94 sens vs. MG 59 (US 65, CBE 50) (spec
  MRI 99)
• 18 cancers (14 inv)

33
Screening MRI StudiesVery High Risk Patients

• Warner JAMA 2004
• 236 women 25-65 yrs with BRCA mutation (3 yrs
  screening)
• MRI 77 sens vs. MG 36 (spec MRI 95 vs. MG
  99.8)
• 22 cancers (16 inv) with 1 interval cancer
• 50 1 cm
• 13 node ()
• Lehman Cancer 2005 International Breast
  Consortium
• 390 women 25 yrs with 25 LTR (one time imaging)
• FN unknown (spec MRI 95 vs. MG 98)
• 4 cancers on MRI (only 1 on MG)

34
Screening MRI StudiesVery High Risk Patients

• Kriege NEJM 2004
• 1909 women 25-70 yrs (mean 3 rounds screening)
• MRI 71.5 sens vs. MG 40 (spec MRI 90 vs. MG
  95)
• 50 cancers (44 inv)
• 19 BRCA1 or 2 carriers with 26.5/1000 cancers
• LTR 15-30 7.8/1000 cancers
• LTR 30-50 5.4/1000 cancers
• Invasive cancers, MRI 79 vs. MG 33
• 43 1 cm
• 67 node (-)

35
Screening MRI StudiesVery High Risk Patients

• Leach MARIBS, Lancet 2005
• 649 women 35-49 yrs 25 LTR (19 BRCA) (median 3
  rounds screening)
• MRI 77 sens vs. MG 40, combined 94 (spec MRI
  77 vs. MG 40)
• 35 cancers (29 inv) with 2 interval cancers
• Differences most notable in gene carriers
• 45 1 cm
• 14 node (-)

36
Screening MRI StudiesClinical Risk Factors

• Studies in women with clinical factors placing
  them at increased risk less numerous
• Port Ann Surg Oncol 2007
• 252 women LCIS, 126 ADH or ALH
• 50 MRI, biennial CBE, MG (younger and ?FHx)
• 50 CBE and MG
• 6 cancers in 5 women (LCIS)
• 4 of women getting MRI
• Small advantage (no cancers in atypia patients)
  but more biopsies
• MRI 75 sens, 92 spec, 13 PPV
• MRI cancers stage 0/I vs. MG cancers stage I/II

37
Screening MRI StudiesPsychological Impacts

• Facts about psychological health impact from MRI
  are taken from subsets of other studies
• MARIBS Lancet 2005
• 611 women, 89 intended to return for further MRI
  screening (1 intended not to return)
• 4 found MRI extremely distressing
• 47 continued to have intrusive thoughts about
  MRI exam 6 weeks after
• Kriege NEJM 2004
• 357 women had psychological stress within normal
  limits throughout the screening as a whole
• ? breast cancer-specific distress related to
  screening was found in excessive breast
  self-examiners (1/week), risk overestimators,
  and women closely involved in the breast cancer
  case of a sister

38
Screening MRI StudiesPsychological Impacts

• Warner Ann Oncol 2004
• Small sample followed for 2 years
• No evidence of any effect on global anxiety,
  depression, or breast-cancer-related anxiety
• Hill Abstract 4036, San Antonio 2006
• 50 elevated baseline general and/or breast
  cancer-specific anxiety (57 women)
• 77 of these attributed to life event, including
  relatives with cancer
• Nonsignificant increase seen in general anxiety
  and breast-cancer-related anxiety in subset of
  women recalled for further imaging or biopsy

39
MRI Limitations

• Relatively poor specificity
• As low as 30 specificity in early data
• In studies that treated a finding on MRI
  equivalent to doing a biopsy for any and all
  findings seen on a mammogram
• Relatively low positive predictive value
• Numbers likely much better, newer data
• Similar specificity to mammography
• Higher specificity than US
• 35-64 positive predictive values in high risk
  women

Fibrocystic Changes
40
MRI Other Aspects to Consider

• Call back rate 8-17
• Limited other tests to do (second-look US)
• Rate decreases after 1st round (learning curve)
• Warner Radiology 2005 steady decrease from 26
  to 10 in year three
• Kuhl Radiology 2000 decrease from 15 to 9 in
  second round
• Biopsy rates range from 3-15
• PPV 20-40
• Overall a high-risk woman has about a 10 chance
  of an abnormal MRI and a 5 chance of a benign
  biopsy
• FN secondary to technical factors, patient
  characteristics, QA issues, human errors,
  medical-legal issues, patient request for biopsy

Sclerosing adenosis
41
Comparison Mammography Numbers

• PPV1 based on abnormal screening exam 5 - 10
• PPV2 when biopsy recommended 25 - 40
• Tumors found - Stage 0 or 1 gt 50
• Tumors found Minimal cancer gt 30
• Node positivity lt 25
• Cancers per 1000 cases 2 - 10
• Prevalent cancers per 1000 first-time exams 6 -
  10
• Incident cancers per 1000 F/U exams 2 - 4
• Recall Rate lt 10
• Sensitivity gt 85
• Specificity gt 90

42
MRI Mortality Endpoint

• Not likely to happen
• There are no RCTs with mortality as an endpoint
  for women at high risk for breast cancer
• If we waited for this we would not be able to
  make recommendations about
• Genetic testing, prophylactic mastectomy,
  chemoprevention, screening mammography, or any
  form of early detection or intervention
• Would take and 15 years for data
• Current studies realize there is a lead-time bias

43
MRI Cost Issues

• Adding MRI to mammography compared to mammography
  alone in BRCA1 and BRCA2 results in QALY gained
  of 45,000 to 700,000
• Depends on age and BRCA
• MRI is 10x more expensive than mammography
• Imaging Mammo 100 vs. MRI 1200
• Biopsy Mammo 800 vs. US 500 vs. MRI 1000
• National Comprehensive Cancer Network Guidelines
  for Genetic/Familial High-Risk Assessment
• Recommends starting mammography screening at 25
  yrs

44
MRI Cost Effectiveness

• Cost per QALY saved for annual MRI with
  mammography versus mammography alone depends on
• Age, density, breast cancer risk factors,
  mammography sensitivity, MRI sensitivity level,
  MRI cost, and QOL gains from MRI
• 55,420 (BRCA1) vs. 130,695 (BRCA2) for 35-54
  years of age
• 41,183 (BRCA1) vs. 98,454 (BRCA2) in dense
  breasts

Pleveritis JAMA 2006
45
MRI Cost Effectiveness

• Incremental cost per cancer detected
• 50,911 50 risk of having BRCA mutation
• 27,544 known mutation carriers
• UK analysis supported introducing screening MRI
  in high risk women for 30-39 yrs at 10-yr-risk
  gt8, familial risk 40-49 yrs at 10-yr-risk gt20
  (or 12 with dense breasts)
• 14,005 in 40-49 years with 31 10-yr-risk group
  (BRCA1)
• 53,320 in 12 10-yr-risk 96,379 in 6
  10-yr-risk
• 24,275 in 30-39 years with 11 10-yr-risk group
  (BRCA1)
• 70,054 in 5 10-yr-risk

Griebsch Br J Cancer 2006 NICE May 2006
46
MRI and -Summary

• Cost
• Anxiety
• F/U imaging
• Benign biopsies
• BUT
• Early detection and possible reduced mortality

47
MRI Who Should Do it?

• At institutions which perform many MRI exams
  (diagnostic and screening)
• Need the ability to also perform biopsy
  (MRI-guided)
• Have partnership with High Risk Clinic and/or
  Genetic Counseling
• Have ability to audit MRI like mammography

March/April 2007 edition of CA a Cancer Journal
for Clinicians
48
MRI Current Indications

• Implant Evaluation
• Axillary Metastasis with unknown primary
• Extent of Disease
• Assessing Response to Therapy
• Diagnostic Dilemmas
• High Risk Screening

49
MRI Current Indications

• American Cancer Society (2003) guideline for the
  early detection of breast cancer stated that
  women at increased risk of breast cancer might
  benefit from additional screening strategies
  beyond those offered to women at average risk,
  such as earlier initiation of screening, shorter
  screening intervals, or addition of screening
  modalities (such as breast US or MRI) other than
  mammography and physical exam.
• Decisions about screening options be based on
  shared decision making after a review of
  potential benefits, limitations, and harms of
  different screening strategies and the degree of
  uncertainty about each

Smith Ca Cancer J Clin 2003
50

• Determination of Risk
• Family history
• Genetic predisposition
• Clinical history

51
Family History

• 2 or more first- or second-degree relatives with
  breast or ovarian cancer
• Breast cancer before 50 years (premenopausal) in
  1st degree relative
• Family history of both breast and ovarian
• One or more relatives with 2 cancers (breast
  and/or ovarian)
• Male relatives with breast cancer

52
Family History

• Gail, Claus, Tyrer-Cusick
• Estimate of breast cancer risk is based on
• family history
• sometimes in combination with
• reproductive history
• prior biopsy
• Estimates are similar but can vary between models

March/April 2007 edition of CA a Cancer Journal
for Clinicians
53
Genetic Risk

• Two decision models for estimation of likelihood
  of being a BRCA mutation carrier
• BRCAPRO
• BOADICEA Breast and Ovarian Analysis of Disease
  Incidence and Cancer Estimation Alorithm
• Also estimates breast cancer risk

BRCAPRO family trees
BRCAPRO comparedwith genetics BRCA1BRCA2 results
54
Genetic Risk

• BRCA1
• Discovered 1990 Hall, Science 1990
• Prevalence 1/500-1/1000, 1/50 Jewish
• Autosomal Dominate
• Responsible for 3 of all cancers and up to 45
  of all hereditary cases
• LTR of approximately 80
• 65 risk by 70 years of age
• Up to 50 by 50 years of age
• BRCA2
• Discovered 1994 Wooster, Science 1994
• Responsible for 30 of all hereditary cases
• 45 risk by 70 years of age
• 6 LTF of male breast cancer
• 100x over general population

NEJM 1997 Am J Hum Genet 1997
55
Genetic Risk

• TP53 gene (Li-Fraumeni syndrome)
• 30 of cancers diagnosed before 15 yrs of age
• Adrenocortical carcinoma
• Soft-tissue sarcomas
• Acute leukemias
• Brain tumors
• Osteosarcomas
• Premenopausal breast cancer is common

TP53
56
Genetic Risk

• PTEN gene (Cowden and Bannayan-Riley Ruvalcaba
  syndromes)
• Lesions on the skin and mucous membranes,
  including facial tricholemmomas, acral
  keratoses, and papillomas
• Macrocephaly
• Intestinal polyps
• Breast, thyroid, and endometrial cancers
• Lipomas
• Developmental delay

Cowden
B-R-R
57
Genetic Risk Referral at UVA

• Early-onset breast cancer (40)
• Breast cancer 50 with
• 1 relative with breast cancer 50 and/or
• 1 relative with ovarian cancer
• Breast cancer at any age with
• 2 relative with breast cancer 50 and/or
• 2 relative with ovarian cancer
• Bilateral breast cancer (or 2 separate primaries
  in the same breast), if additional supporting
  personal/FHx
• Male breast cancer
• Personal history of ovarian cancer
• Ashkenazi Jewish ancestry, if additional
  supporting personal/FHx
• Suspicion of Cowden syndrome
• Suspicion of Li-Fraumeni syndrome

58
Clinical Factors

• Hodgkins disease who had mantle radiation
• Risk ?15-30 years after XRT
• Prior biopsies demonstrating high risk lesions
• Lobular Neoplasia
• LTR 10-20, 0.5-1.0 per year
• LCIS (lobular carcinoma in-situ)
• 6-10x ?
• 50 ILC gt15 yrs after Dx LCIS (Bil)
• Incidence of 13/1000 inv ca (Bil)
• ALH (atypical lobular hyperplasia)
• ADH (atypical ductal hyperplasia)
• 4-5x ?at mean 12 years, more if FHx

ALH
ADH
59
Clinical Factors

• Dense breast tissue
• 4-6x greater than fatty
• Masses more often occur in dense areas of breast
• Personal history of breast cancer
• 0.5-1.0 per year
• 5-10 in first 10 years following Dx

60

• High-Risk Screening MRI
• Guidelines ACS
• Guidelines UVA

61
High-Risk Screening MRIACS Guidelines

• Released March/April 2007 edition of CA a Cancer
  Journal for Clinicians
• Based on literature published on screening MRI
  Sept 2002-July 2006
• Recommends screening MRI
• Women with 20-25 or greater lifetime risk
• Strong family history
• Women treated for Hodgkins disease (radiation
  b/n 10-30 years of age)
• Not enough evidence for other subsets

62
High-Risk Screening MRIACS Guidelines

• ACS suggests that women begin annual MRI
  screening at 30 years of age and continue as long
  as they are in good health
• Interval is based on data available (limited)

63
High-Risk Screening MRIACS Guidelines
Saslow, et al. American Cancer Society Guidelines
for Breast Screening with MRI as an Adjunct to
Mammography. CA Cancer J Clin 2007 57 75-89
64
ACR Recommendations Age

• BRCA 3 risk by 30 years and 19 by 40 years
• Expert panel suggested starting at 25-35 years
  old for BRCA1 or BRCA2
• Observational studies (group decision)
• Screening with MRI 5-10 years before earliest
  family history
• When stop? No data in women older than 69 yrs
• How often? Data is mostly annual
• Not much on incidence (follow up)

65
High-Risk Screening MRINBCC Response

• National Breast Cancer Coalition published
  response March 2007
• Randomized studies unlikely given that MRI so
  widely used in clinical practice
• ACS recommendations based primarily on
  observational studies and expert consensus
  opinion
• Concerns about demand increase, standards for
  equipment/acquisition/readers, high cost
• Guidelines needed.

66
ACR Guidelines MRI Performance

• Guidelines initially published in 2004
• Include recommendations for indications
• Plan to release accreditation process for
  performing breast MRI
• Imaging protocols
• MRI-guided procedures

http//www.acr.org/SecondaryMainMenuCategories/qua
lity_safety/guidelines/breast.aspx
67
ACR Guidelines Breast Imaging
68
High-Risk Screening MRIUVA Guidelines

• We feel a subset of women will benefit from MRI
  screening exams
• Need to continue mammographic screening
• Need up-to-date mammogram
• Either to coincide with mammogram or within 6
  months of study

69
High-Risk Screening MRIUVA Guidelines
risk factors
70
Future Direction and Discussion

• MRS (spectroscopy)
• DWI (diffusion-weighted imaging)
• Galactography
• New cancer patients
• Motivated by 3 risk of contralateral malignancy
  only seen by MRI (normal PE, MG and US)
• More to come on protocol/suggestions for patient
  management

Lehman NEJM 2007
71
Future Direction and Discussion
MRS
DWI
72
Future Direction and Discussion
Hirose, Otsuki et al. Magn Reson Med Sci. 2006.
73
Future Direction and Discussion
Hirose, Otsuki et al. Magn Reson Med Sci. 2006.
74
Summary

• MRI is highly sensitive but has limitations
• FP
• Biopsy
• Cost
• Stress
• High-risk patient is one at gt20-25 lifetime risk
• Familial, genetic, clinical
• Our protocol is for high risk annual MRI and
  mammography at 6 mn intervals
• Desire to improve specificity
• What about new cancer diagnosis patients !?