How should we screen for breast cancer Mammography by mikeholy



         CAROLYN F. NEMEC, MD                        JAY LISTINSKY, MD, PhD                  ALICE RIM, MD
         Women’s Health Center, Cleveland Clinic,
         Willoughby Hills, OH
                                                     Breast Imaging, University Hospitals,
                                                     Cleveland, OH
                                                                                             Head, Section of Breast Imaging,
                                                                                             Cleveland Clinic

How should we screen for breast cancer?
Mammography, ultrasonography, MRI
■ A B S T R AC T                                                                                                                    for
                                                                                             T breast cancer is, “Veryshould screen No
                                                                                                     HE ANSWER TO HOW WE
  Of the imaging techniques currently available to evaluate                                  screening procedure is perfect, women vary great-
  women for breast disease, mammography remains the                                          ly in their breast cancer risk, and screening may
  mainstay of breast cancer screening, but recent guidelines                                 lead to unnecessary procedures and alarm.
  have included magnetic resonance imaging (MRI) for the                                     Therefore, physicians must carefully consider
  screening of some women at high risk. Whole-breast                                         which screening regimen is right for each patient.
  ultrasonography for screening has not been established                                          While many issues surrounding breast can-
  as useful and so should not be offered routinely to                                        cer screening are still unresolved, general
  patients.                                                                                  guidelines have now been implemented on the
                                                                                             basis of data accrued over many years.
■ KEY POINTS                                                                                      In this article, we summarize the most cur-
                                                                                             rent guidelines and also comment briefly on
  Most major medical organizations recommend starting                                        screening examinations that hold promise but
  routine screening mammography for women at age 40.                                         have not yet earned a place in routine breast
                                                                                             cancer screening.
                                                                                                  Breast cancer survival has improved over
  If a screening mammogram is abnormal or has findings                                       the past few decades. In the 1940s, the 5-year
  of unclear significance, the patient should be referred for                                survival rate for early-stage localized disease
  diagnostic mammography.                                                                    (no lymph node involvement or metastasis)
                                                                                             was 72%, which has improved to 97% today.1
  If a palpable breast mass is discovered, the patient should                                This improvement in survival is in large mea-
  be referred for diagnostic mammography and                                                 sure attributable to the increased and effective
  ultrasonography.                                                                           use of screening mammography in asympto-
                                                                                             matic patients, with improved treatment pro-
  Breast MRI can be considered in addition to                                                tocols also playing a role. Early detection of
  mammography for screening in high-risk patients, such as                                   breast cancer, ie, before it is clinically appar-
  women with a BRCA gene mutation, a strong family                                           ent, is important both to patients and to their
  history of breast cancer or a personal history of ovarian                                  physicians. In this article, we also examine
                                                                                             how the careful use of breast imaging tech-
  cancer, or women who have received high-dose chest                                         niques—mammography, ultrasonography, and
  radiation, such as mantle radiotherapy for Hodgkin                                         magnetic resonance imaging (MRI)—can
  disease. Other high-risk groups in which breast MRI is                                     improve breast cancer detection in women.
  currently being studied are women with a personal
  history of breast cancer, women with a history of atypical                                 ■ WHO SHOULD UNDERGO
  duct hyperplasia or lobular carcinoma in situ, and women                                     MAMMOGRAPHIC SCREENING?
  with an elevated Gail breast cancer risk assessment score.
                                                                                             Experts have long agreed that screening
                                                                                             mammography reduces the rate of death

                                                    CLEVELAND CLINIC JOURNAL OF MEDICINE             VOLUME 74 • NUMBER 12      DECEMBER 2007   897
                         BREAST CANCER SCREENING                         NEMEC AND COLLEAGUES

                               Current breast biopsy techniques
        CORE NEEDLE BIOPSY                                              core-biopsy instruments are more widely available.
        A large-bore automated cutting needle is used to                Fine-needle aspiration can often make the diagno-
        remove three to five solid cylindrical tissue sam-              sis of malignancy but does not provide sufficient
        ples (“cores”). For adequate samples, a 14-gauge                tissue for more detailed studies, and thus generally
        or larger needle is used. These procedures are per-             necessitates obtaining a second (core) biopsy spec-
        formed with guidance by ultrasonography, stereo-                imen for study before definitive treatment can be
        tactic imaging, or MRI. In most cases this is the               planned. In addition, the satisfactory interpreta-
        preferred method of biopsy, since it usually pro-               tion of cytologic specimens requires pathologists
        vides adequate tissue for tumor grading and per-                with special expertise in cytopathology. Fine-nee-
        formance of receptor studies, both of which are                 dle aspiration is not recommended for the evalua-
        important in formulating the patient’s treatment                tion of suspected ductal carcinoma in situ.
                                                                        EXCISIONAL BIOPSY
        FINE-NEEDLE ASPIRATION                                          This procedure is performed by a surgeon in the
        A smaller-bore (usually 18- or 20-gauge) needle is              operating room, usually to remove the entire mass
        used to obtain cytologic samples from a suspicious              or suspicious area. Excisional biopsy requires pre-
        breast mass. This is technically easy to perform but            operative wire localization if the lesion is not pal-
        is less used by radiologists now that automated                 pable.

                     from breast cancer in women who begin                          A recent study6 considered the separate
                     screening in their 50s and 60s. These con-                effects of screening mammography and of
                     clusions are supported by results from eight              adjuvant therapy on the breast cancer death
                     randomized clinical studies of the efficacy of            rate, drawing on the experience of multiple
Screening            screening mammography. For women ages 50                  institutions. From 1975 to 2000, the overall
mammography          to 69, screening mammography decreased                    reduction in breast cancer deaths was 24%.
                     the death rate from breast cancer by 20% to               The study estimated that the portion of the
lowers the           35%.2,3                                                   reduction attributed to screening mammog-
mortality rate            The value of screening mammography for               raphy ranged from 28% to 65% (median
                     women in their 40s has been more recently                 46%), with the rest attributed to the use of
by about 20%         addressed; meta-analyses now reveal that                  adjuvant therapy.6 The variability in the
                     screening mammography decreases breast can-               reduction of the death rate was attributed to
                     cer death rates by about 20%.2,4                          variations in the inclusion criteria of the par-
                          For women over age 70 there are fewer                ticipating groups: eg, some groups included
                     studies. One study in the Netherlands found               patients with ductal carcinoma in situ, and
                     that mammographic screening in women over                 other studies included only patients with
                     age 65 led to a 55% decrease in the breast can-           invasive carcinoma.
                     cer death rate.5
                          Annual screening mammography for                     General recommendations
                     women age 40 and older is covered by major                for mammographic screening
                     insurance carriers in the United States. In                    Screening mammography should begin at
                     2007, the Medicare reimbursement for bilater-             age 40. This recommendation is supported by
                     al screening mammography was $93.03.                      major medical organizations, including the
                          The principal aim of screening mammogra-             American Cancer Society, the American
                     phy is the same for all age groups: to detect             College of Radiology, the National Cancer
                     breast cancer at an early stage, before it becomes        Institute, the American College of Obstetricians
                     clinically apparent, and thereby to avoid the ill-        and Gynecologists, and the American Medical
                     ness and death that accompany locally                     Association.7 In our practice, we follow the
                     advanced or widespread breast cancer.                     American College of Radiology recommenda-

tions for routine screening at yearly intervals.    but it can also be caused by hormone thera-
     Screening mammography is most effec-           py. Women on long-term hormone therapy
tive between ages 50 and 59 and should be           are called back more often for further evalu-
routinely recommended.                              ation and have a higher rate of benign breast
     There is no established upper age limit        biopsies carried out to evaluate suspicious
to the beneficial use of screening mammog-          findings.15
raphy. According to the 2004 revised                     The addition of breast ultrasonography or
American College of Radiology guidelines,8          MRI in selected cases of mammogram density
“It is unclear at what age, if any, women cease     may be useful. Digital mammography may also
to benefit from screening mammography.              play a role in patients with mammogram den-
Because this age is likely to vary depending on     sity and is currently undergoing evaluation in
the individual’s overall health, the decision as    large-scale trials.16 Its clinical usefulness has
to when to stop routine mammography                 yet to be fully elucidated.
screening should be made on an individual
basis by each woman and her physician.”8 The        ■ WHAT HAPPENS IF THE MAMMOGRAM
American Cancer Society further recom-                SHOWS AN ABNORMALITY?
mends that “as long as a woman is in reason-
ably good health and would be a candidate for       Screening mammography may detect a mass
treatment, she should continue to be screened       lesion, suspicious microcalcifications, focal
with mammography.”9                                 asymmetry, or architectural distortion. It may
                                                    detect lymph nodes of abnormal size, contour,
Special recommendations                             or density. It may also detect more subtle
     If a patient has a first-degree relative       changes such as skin thickening. If any of
who has had breast cancer, screening mam-           these findings is noted, diagnostic mammogra-
mography should commence 10 years earlier           phy is recommended.
than the age at which that relative was diag-            About 10% of women who undergo
nosed, or at age 25, whichever is older. For        screening mammography are called back for
example, if the patient’s mother was diag-          further evaluation. Of these patients, around       There is as yet
nosed with breast cancer at age 39, screening       10% will have a breast biopsy, of which 25% to      no upper age
mammography for her daughters should begin          40% will be positive for breast cancer.17,18
at age 29.                                          Using estimates from the Mammography                limit to the
     Patients with a personal history of atyp-      Quality Standards Act guidelines, of 1,000          beneficial use
ical duct hyperplasia or lobular carcinoma in       asymptomatic patients screened by mammogra-
situ are candidates for increased surveillance,     phy, 2 to 10 patients will be shown to have         of screening
usually including a clinical breast examina-        breast cancer. This range allows for variations     mammography
tion and mammography every 6 months.                encountered in different screening populations.
(High-risk patients may also be candidates for
breast MRI, as discussed later in this article).    ■ WHEN SHOULD WE ORDER
     Patients who have received high-dose             DIAGNOSTIC MAMMOGRAPHY?
chest radiation (mantle radiotherapy) are at
increased risk of developing radiation-induced      If a patient presents with a breast complaint,
breast cancer. For this reason, a woman with a      or if an abnormality is noted on the clinical
diagnosis of Hodgkin lymphoma for example,          breast examination, then diagnostic mam-
who received mantle radiotherapy, should            mography should be ordered. In contrast to
begin mammographic screening 8 years after          screening mammography, which consists of
her radiation treatment.                            two standard views of each breast, diagnostic
     ‘Mammogram density’ is now consid-             mammography includes extra views or studies
ered a risk factor for breast cancer. In fact,      tailored to evaluate the finding in question.
the risk associated with mammogram density               Reasons to refer a patient for diagnostic
may be greater than the risk from many other        mammography include an abnormal screening
risk factors.10–13 Mammogram density is             mammogram, a breast mass or thickening on
thought to be an inherited phenomenon,14            palpation, focal breast pain, clear or bloody

                                      CLEVELAND CLINIC JOURNAL OF MEDICINE   VOLUME 74 • NUMBER 12   DECEMBER 2007   899
                       BREAST CANCER SCREENING                         NEMEC AND COLLEAGUES

                                                                           FIGURE 3. Ultrasonography further defines
                                                                           the mass (arrow) as a simple cyst.

                                                                           tic mammography (FIGURES 1–3).
                                                                                If diagnostic mammography or ultra-
                                                                           sonography reveals suspicious findings, biop-
                                                                           sy is recommended (see “Current breast biop-
                                                                           sy techniques,” page 898). In a small number
                    FIGURE 1. Screening mammography reveals
                    a mass (arrow).                                        of cases in which suspicion of cancer is very
                                                                           low, a conclusion of “probably benign” may
                                                                           be given, with recommendations for short-
Screening for                                                              term follow-up.
breast cancer
                                                                           ■ A PALPABLE MASS
should begin                                                                 WITH NEGATIVE MAMMOGRAPHIC
with                                                                         AND ULTRASONOGRAPHIC STUDIES
mammography                                                                What should be done when a mass is palpable,
                                                                           but mammographic and ultrasonographic
                                                                           studies are negative? This will depend on the
                                                                           situation. Often, ultrasonography will show
                                                                           the palpable lump to be a normal fibrous ridge
                                                                           or a region of benign asymmetry that corre-
                                                                           sponds definitely to the palpable findings. In
                                                                           these cases, no further workup is needed. But
                                                                           if clinical suspicion remains high despite the
                                                                           negative imaging results, biopsy should be
                    FIGURE 2. Diagnostic mammography mag-                       Fine-needle aspiration is an easy, mini-
                    nifies the mass (arrow).                               mally invasive way to obtain a sample of a pal-
                                                                           pable mass. It can be performed in the office.
                    nipple discharge, nipple retraction, a concern-        Core biopsy can be carried out on a palpable
                    ing palpable lymph node, or abnormal skin              mass using imaging guidance in order to assure
                    changes such as erythema or peau d’orange. If          a safe trajectory for the cutting needle.
                    a breast mass is palpated, an ultrasonographic              If the palpable mass is not easily accessible
                    examination is ordered in addition to diagnos-         (eg, if it is close to the chest wall or in the axil-

la), or if the patient is very anxious, an exci-    ing programs… At the present time, it is not
sional biopsy can be performed to remove the        the standard of care to offer or perform this
entire mass.                                        examination.” In our practice, we do not rec-
     No imaging technique can exclude breast        ommend or offer screening with breast ultra-
cancer with 100% accuracy. This is true for         sonography.
mammography, ultrasonography, and MRI.
                                                    ■ MAGNETIC RESONANCE IMAGING
                                                    What is breast MRI?
Should we use ultrasonography to screen for         MRI uses magnetic fields along with radiofre-
breast cancer? In a word, no. The rationale         quency transmitters and receivers to produce
and limitations summarized below—from the           cross-sectional images of the human body. To
Position Statement on Screening Breast              image the breast, specialized imaging receivers
Sonography in Dense Breasts, promulgated by         (“coils”) that encompass each breast are used.
the Society of Breast Imaging19—reflect the              For cancer detection protocols, a contrast
expert consensus on this issue.                     agent that contains gadolinium is injected
     Screening mammography is an important          intravenously to help identify tissues that
tool, but it can miss some breast cancers, espe-    “handle” the agent in an abnormal way, which
cially in women with dense breast tissue. For       is a possible sign of the neovascularity seen in
this reason, it was thought that ultrasono-         many breast cancers. MRI involves no ioniz-
graphic screening might improve breast can-         ing radiation, and most patients tolerate the
cer detection rates.20 Indeed, breast ultra-        contrast agent well.
sonography can detect some invasive cancers              The procedure. The patient lies prone on
that mammography and physical examination           a padded table that contains the two breast
miss, but the number of cancers found with          coils. The breasts are positioned within the
ultrasonography alone remains small. Also,          coils, sometimes with mild compression to
ultrasonography does not detect most micro-         maintain constant positioning. The table is         No imaging
calcifications, which are the typical findings      advanced into the magnet and a preselected          study can
in ductal carcinoma in situ. In fact, 75% of        series of scans is carried out, both before and
cancers missed by ultrasonography were duc-         after the injection of contrast material. The       exclude
tal carcinoma in situ and 25% were invasive         entire study takes about 40 to 50 minutes,          breast cancer
carcinomas.                                         with about half of that time devoted to
     In addition, the results of ultrasonography    patient preparation, placement of a small-          with 100%
can vary widely, depending on the expertise of      gauge intravenous catheter, and patient posi-       accuracy
the technician. Indeterminate findings can          tioning. The scans are acquired rapidly, but
lead to the increased use of costly and perhaps     image reconstruction and post-processing
unnecessary interventions. There are at pres-       continue after the patient has left the scanner.
ent no data to support whole-breast ultrasono-           Contraindications to MRI include
graphic screening to decrease breast cancer         implant expanders, cardiac pacemakers, neu-
death rates. A study by the American College        rostimulator devices, extreme claustrophobia,
of Radiology Imaging Network (called                and morbid obesity.
ACRIN Study 6666) is under way to investi-
gate this issue.                                    What are the disadvantages of breast MRI?
     Screening for breast cancer should begin       An important disadvantage of breast MRI is
with mammography. Ultrasonography can be            the rate of false-positive results. The sensitiv-
added to evaluate a mass or to clarify focal        ity of breast MRI is high but the specificity is
mammographic findings. According to a posi-         low. Breast cysts, fibroadenomas, papillomas,
tion statement of the Society of Breast             and fibrocystic changes may all appear as
Imaging,19 it “has not been established that        abnormalities on contrast images, resulting in
women will benefit from the incorporation of        unnecessary biopsies. In a study of breast MRI
sonography into routine breast cancer screen-       in high-risk women, many of whom were

                                      CLEVELAND CLINIC JOURNAL OF MEDICINE   VOLUME 74 • NUMBER 12   DECEMBER 2007   901
                       BREAST CANCER SCREENING                         NEMEC AND COLLEAGUES

                                                                           by age 70. The lifetime risk for carriers of
                                                                           BRCA1 or BRCA2 mutations is 50% to
                                                                           85%.24 Breast cancers in women with a muta-
                                                                           tion often occur at a young age, are “aggres-
                                                                           sive” with a high nuclear grade, and lack estro-
                                                                           gen receptors.25 At the time of diagnosis, half
                                                                           of these breast cancers have already spread to
                                                                           axillary lymph nodes.22
                                                                                Up to this point, strategies to follow these
                                                                           patients have included bilateral prophylactic
                                                                           mastectomy, prophylactic chemotherapy with
                                                                           tamoxifen, and early surveillance, ie, begin-
                                                                           ning clinical breast examinations and mam-
                                                                           mographic screening at age 25 to 30.
                                                                                The usefulness of breast MRI in these high-
                                                                           risk patients is now being studied. Investigators
  FIGURE 4. This mammogram shows mild asymmetry of the                     in the United Kingdom looked at mammogra-
  left breast.                                                             phy vs contrast-enhanced MRI in 649 asymp-
                                                                           tomatic women with a known BRCA1,
                young and had very dense breast tissue,                    BRCA2, or TP53 mutation or a strong family
                screening MRI led to three times as many                   history of breast cancer. MRI was significantly
                benign breast biopsies as mammography.21                   more sensitive (77% vs 40%) but less specific
                However, one could argue that in high-risk                 (81% vs 93%) than mammography. In particu-
                populations the sensitivity of mammography is              lar, MRI was significantly more sensitive than
                quite low and may be of limited value, making              mammography in patients with a BRCA1
                MRI an imperfect but better tool.22                        mutation and their first-degree relatives (92%
                    Another disadvantage of breast MRI is                  vs 23%).21,26
Ultrasonography that it does not detect microcalcifications,                    Screening breast MRI is proving to be a
can clarify a   which are often associated with ductal carci-              useful adjunct to mammography in patients
                noma in situ. These calcifications are readily             with a BRCA mutation. In addition to these
focal finding   apparent on mammography. Breast MRI is also                patients, the American Cancer Society has
on the clinical not sensitive in detecting lobular cancers.                recently recommended that other high-risk
                    A recent study in Germany in high-risk                 patients pursue breast MRI. These include
exam or         women found breast MRI to have a sensitivity               women with a 20% to 25% or greater lifetime
mammogram       of 92% in detecting ductal carcinoma in situ               risk of breast cancer: ie, women with a strong
                (DCIS),23 but in general DCIS remains a                    family history of breast cancer or ovarian can-
                mammographic diagnosis.                                    cer and women with a history of mantle radio-
                    Other major drawbacks to using breast                  therapy for Hodgkin disease.27
                MRI include cost and limited access. Fees for                   Insurance often covers the cost of breast
                breast MRI range from $3,000 to $4,000,                    MRI for patients in these high-risk groups.
                which is 10 times that of mammography.                     High-risk patients can also obtain breast MRI
                Unlike mammography, breast MRI is not rou-                 under research protocols in which there is no
                tinely covered by insurance. Screening MRI                 charge to the patient.
                protocols have yet to be developed, and access                  In our practice, candidates for annual
                to breast MRI remains limited.                             screening breast MRI include:
                                                                           • Women who carry a BRCA mutation
                    Who is a candidate for screening with MRI?             • Women with a strong family history of
                    The average lifetime risk of breast cancer for              breast cancer
                    an American woman is now one in seven.17               • First-degree relatives of a BRCA carrier,
                    However, the risk of breast cancer in women                 but untested
                    with a BRCA1 gene mutation is 3.2% by age              • Women with a strong family history or a
                    30, 19% by age 40, 50% by age 50, and 85%                   personal history of ovarian cancer

•   Women who were treated for Hodgkin
    disease with radiation to the chest.
• Women known to have a hereditary
    breast cancer syndrome and their first-
    degree relatives.
    Screening MRI is currently under investi-
gation in patients with a history of atypical
duct hyperplasia or lobular carcinoma in situ
or with an elevated Gail score (ie, a 5-year risk
of developing breast cancer ≥ 1.7%).
    In carefully selected cases, MRI may be
helpful in the evaluation of equivocal or sus-
picious mammographic findings; however,
detailed mammographic evaluation and ultra-
sonography should be done first.
    Important note: Breast MRI may not
detect some in situ carcinomas and other low-
grade benign or malignant lesions and is only
an adjunct to mammography. Breast MRI
should never be offered as a substitute for con-     FIGURE 5. Breast MRI shows abnormal enhancement of a
ventional screening mammography.                     single lesion corresponding in size and location to the left
                                                     mammographic asymmetry.
  GUIDELINES AS APPLIED                              MRI detected contralateral breast cancer in
  IN DAILY PRACTICE                                  up to 10% of women initially diagnosed with
                                                     unilateral breast cancer. The contralateral
A 56-year-old white woman is seen at the             breast cancers in these women were missed by
breast center for her annual examination and         both clinical and mammographic evaluation.          MRI is not a
mammographic evaluation. She is considered                These results suggest that breast MRI may      substitute for
at high risk because of a family history of          be a valuable tool in evaluating women at the
breast cancer—her mother and sister—and              time of the initial breast cancer diagnosis.        conventional
her elevated 5-year Gail score (5.0%).               However, more evidence is needed from large-        screening
     Her breast examination is negative and her      scale clinical trials before clinicians can rec-
mammogram shows mild asymmetry on the left           ommend routine screening MRI for long-term          mammography
(FIGURE 4). Because she is at high risk, breast      follow-up of patients with a personal history of
MRI is ordered and reveals a small mass on the       breast cancer.
left near the area of asymmetry (FIGURE 5).
     She is taken to surgery and is found to         ■ MUCH WORK IS YET TO BE DONE
have a 5-mm breast cancer with negative axil-
lary lymph nodes. She undergoes radiation            In a report just released by the American
therapy and begins anastrozole (Arimidex)            Cancer Society, the breast cancer mortality
therapy. Cytotoxic chemotherapy was not              rate decreased by 2.2% per year between 1990
necessary.                                           and 2004.29 However, in African American
                                                     women the breast cancer death rate declined
■ PATIENTS WITH BREAST CANCER:                       by only 1.6% per year and remained
  ANOTHER HIGH-RISK GROUP                            unchanged among Asian Americans/Pacific
                                                     Islanders and Alaskans/Native Americans.
No discussion of breast MRI would be com-            Reasons for the differences in mortality rates
plete without mentioning another high-risk           remain unclear, and further research is needed.
group: patients with breast cancer.                  In the same report, it appears that breast cancer
    A study of breast MRI in women recently          incidence is more complex than previously
diagnosed with breast cancer28 found that            thought. Access to mammography and the

                                       CLEVELAND CLINIC JOURNAL OF MEDICINE   VOLUME 74 • NUMBER 12   DECEMBER 2007   903
                     BREAST CANCER SCREENING                                NEMEC AND COLLEAGUES

                  decline in hormone therapy use may be con-                          tine mammograms are not receiving a recom-
                  tributing factors.                                                  mendation for mammography from their
                       The decrease in screening mammogram                            physician. We must continue to communicate
                  rates is also of concern. Almost one-third of                       to our patients the importance of routine
                  American women are not undergoing mam-                              screening mammography and offer our high-
                  mographic screening at appropriate inter-                           risk patients additional breast imaging when
                  vals.30 Most women who are not having rou-                          appropriate.                              ■

                  ■ REFERENCES
                   1. Jemal A, Siegel R, Ward E, Murray T, Xu J, Thun MJ.             18. Smith-Bindman R, Chu PW, Miglioretti DL, et al.
                      Cancer statistics, 2007. CA Cancer J Clin 2007; 57:43–66.           Comparison of screening mammography in the United
                   2. Preventive Services Task Force. Screening for breast                States and the United Kingdom. JAMA 2003;
                      cancer: recommendations and rationale. Ann Intern                   290:2129–2137.
                      Med 2002; 137:344–346.                                          19. Screening Breast Sonography in Dense Breasts. SBI
                   3. Fletcher SW, Black W, Harris R, Rimer BK, Shapiro S.                Position Statements, Society of Breast Imaging.
                      Report of the international workshop on screening for               Available at Last accessed
                      breast cancer. J Natl Cancer Inst 1993; 85:1644–1656.               November 7, 2007.
                   4. Berry DA. Benefits and risks of screening mammogra-             20. Kolb TM, Lichy J, Newhouse JH. Comparison of the
                      phy for women in their forties: a statistical appraisal. J          performance of screening mammography, physical
                      Natl Cancer Inst 1998; 90:1431–1439.                                examination, and breast ultrasound and evaluation of
                   5. van Dijck JA, Verbeek AL, Beex L, et al.                            factors that influence them: an analysis of patient
                      Mammographic screening after the age of 65 years:                   evaluations. Radiology 2002; 225:165–175.
                      evidence for a reduction in breast cancer mortality. Int        21. Leach MO, Boggis CRM, Dixon AK, et al. Screening
                      J Cancer 1996; 66:727–731.                                          with magnetic resonance imaging and mammography
                   6. Berry DA, Cronin KA, Plevritis SK, et al. Effect of                 of a UK population at high familial risk of breast can-
                      screening and adjuvant therapy on mortality from                    cer: a prospective multicentre cohort study (MARIBS).
                      breast cancer. N Engl J Med 2005; 353:1784–1792.                    Lancet 2005; 365:1769–1778.
                   7. Fletcher SW, Elmore JG. Mammographic screening for              22. Liberman L. Breast cancer screening with MRI: what
                      breast cancer. N Engl J Med 2003; 348:1672–1680.                    are the data for patients at high risk? N Engl J Med
                   8. American College of Radiology. ACR Practice Guideline               2004; 351:497–500.
                      for the Performance of Screening Mammography.                   23. Kuhl CK, Schrading S, Bieling HB, et al. MRI for diag-
                      Revised 2004. Available at Last accessed               nosis of pure carcinoma in situ: a prospective observa-
                      November 7, 2007.                                                   tional study. Lancet 2007; 370:485–492.
                   9. Smith RA, Saslow D, Sawyer KA, et al. American                  24. Burke W, Daly M, Garber J, et al. Recommendations
                      Cancer Society Guidelines for Breast Cancer Screening:              for follow-up care of individuals with an inherited pre-
                      Update 2003. CA Cancer J Clin 2003; 53:141–169.                     disposition to cancer. II. BRCA1 and BRCA2. Cancer
                  10. Tabar L, Dean PB. Mammographic parenchymal pat-                     Genetics Studies Consortium. JAMA 1997;
                      terns: risk indicator for breast cancer? JAMA 1982;                 277:997–1003.
                      247:185–189.                                                    25. Lakhani SR, Van De Vijver MJ, Jacquemier J, et al. The
                  11. Ciatto S, Zappa M. A prospective study of the value of              pathology of familial breast cancer: predictive value of
                      mammographic patterns as indicators of breast cancer                immunohistochemical markers, estrogen receptor,
                      risk in a screening experience. Eur J Radiol 1993;                  progesterone receptor, HER-2 and p53 in patients with
                      117:122–125.                                                        mutations in BRCA1 and BRCA2. J Clin Oncol 2002;
                  12. Byrne C, Schairer C, Wolfe J, et al. Mammographic fea-              20:2310–2318.
                      tures and breast cancer risk: effects with time, age,           26. Warner E, Causer PA. MRI surveillance for hereditary
                      and menopause status. J Natl Cancer Inst 1995;                      breast-cancer risk. Lancet 2005; 365:1747–1749.
                      87:1622–1629.                                                   27. Saslow D, Boetes C, Burke W, et al. American Cancer
                  13. Boyd NF, Guo H, Martin LJ, et al. Mammographic den-                 Society guidelines for breast screening with MRI as an
                      sity and the risk and detection of breast cancer. N Engl            adjunct to mammography. CA Cancer J Clin 2007;
                      J Med 2007; 356:227–236.                                            75–89.
                  14. Boyd NF, Dite GS, Stone J, et al. Heritability of mam-          28. Lehman CD, Gatsonis C, Kahl CK, et al. MRI evaluation
                      mographic density, a risk factor for breast cancer. N               of the contralateral breast in women with recently
                      Engl J Med 2002; 347:886–894.                                       diagnosed breast cancer. N Engl J Med 2007;
                  15. Banks E, Reeves G, Beral V, et al. Impact of use of hor-            356:1295–1303.
                      mone replacement therapy on false positive recall in            29. American Cancer Society. Breast Cancer Facts and
                      the NHS breast screening programme: results from the                Figures 2007-2008, Atlanta: American Cancer Society,
                      million women study. BMJ 2004; 328:1291–1292.                       Inc. Available at
                  16. Pisano ED, Gatsonis C, Hendrick E, et al. Diagnostic            30. Meissner HI, Breen N, Taubman ML, Vernon SW,
                      performance of digital versus film mammography for                  Graubard BI. Which women aren’t getting mammo-
                      breast cancer screening. N Engl J Med 2005;                         grams and why? (United States). Cancer Causes Control
                      353:1773–1783.                                                      2007; 18:61–70.
                  17. Sohlich RE, Sickles EA, Burnside ES, Dee KE.
                      Interpreting data from audits when screening and                ADDRESS: Carolyn F. Nemec, MD, Cleveland Clinic Willoughby
                      diagnostic mammography outcomes are combined.                   Hills, 2550 SOM Center Road, N Building, Suite 100,
                      AJR Am J Roentgenol 2002; 178:681–686.                          Willoughby Hills, OH 44094; e-mail


To top