Breast Cancer Surveillance in Women with Hereditary Risk Due to BRCA1 or BRCA2 Mutations

comprehensive

review

Breast Cancer Surveillance in Women with Hereditary Risk Due to BRCA1 or BRCA2 Mutations Mark Robson Abstract Women with germline mutations in BRCA1 or BRCA2 are known to be at substantially elevated risk for breast cancer. With increasing acceptance of genetic testing, significant numbers of mutation carriers are being identified, but evidence-based guidelines for the management of women at hereditary risk are lacking. This article reviews the most commonly recommended modalities employed in breast cancer surveillance for women at increased risk. It is apparent that the standard techniques of breast self-examination, clinical breast examination, and mammography are suboptimal for the identification of hereditary breast cancer. At least half of the cancers in this population appear to be detected by physical examination in the intervals between routine radiographic surveillance. Host factors (eg, breast density) and tumor features (rapid proliferative rates) likely contribute to the relative insensitivity of mammography. These factors may be mitigated by the deployment of screening techniques for breast cancer such as ultrasound and magnetic resonance imaging. However, the effect of incremental screening on either stage at diagnosis or breast cancer mortality has not been defined. In addition, the impact of the relatively limited specificity of these techniques on the quality of life (QOL) of women at risk has not been studied. Further research is needed to evaluate the effect of incremental radiographic screening on outcomes, to delineate the best way to integrate the different modalities in terms of sequencing and frequency, and to identify interventions that will minimize the impact of intensive surveillance programs on the QOL of the women engaged in them. Clinical Breast Cancer, Vol. 5, No. 4, 260-268, 2004 Key words: Breast magnetic resonance imaging, Mammography, Ultrasound imaging

Introduction Since BRCA1 and BRCA2 were identified in 1994 and 1995,1,2 genetic testing to define a hereditary breast-ovarian cancer predisposition has become a commonplace. A recent paper described 10,000 individuals who had undergone mutation analysis through a major commercial laboratory,3 and significant numbers of additional women have undergone testing in the context of various research studies around the world. Individuals have a number of reasons for seeking testing, whether clinically or in the context of a research study. In either setting, 2 prominent motivations for testing are to acMemorial Sloan-Kettering Cancer Center, New York, NY Submitted: Sep 17, 2003; Revised: Dec 1, 2003; Accepted: Dec 15, 2003 Address for correspondence: Mark Robson, MD, Clinical Genetics and Breast Cancer Medicine Services, Department of Medicine, Memorial Sloan-Kettering Cancer Center, 1275 York Ave, New York, NY 10021 Fax: 212-434-5166; e-mail:[email protected]

quire information for the benefit of family members and to define personal risk to guide the decision-making process regarding cancer surveillance and prevention.4,5 These motivations presume that, if an elevated risk were to be demonstrated through genetic analysis, then management strategies that are different from those offered to women at average risk would be both appropriate and effective. Several groups have published recommendations for specialized surveillance regimens for women at risk for hereditary breast cancer (Table 1).6-9 Unfortunately, these programs are based almost entirely upon expert opinion, which has led to considerable variation from center to center in terms of the regimen recommended.10 In addition, these recommendations were developed before the availability of data regarding the potential effectiveness of new technologies; particularly breast magnetic resonance imaging (MRI). The outcomes of women engaged in these programs are only now being defined. The purpose of the current review is to discuss the available literature re-

Electronic forwarding or copying is a violation of US and International Copyright Laws. Authorization to photocopy items for internal or personal use, or the internal or personal use of specific clients, is granted by Cancer Information Group, ISSN #1526-8209, provided the appropriate fee is paid directly to Copyright Clearance Center, 222 Rosewood Drive, Danvers, MA 01923 USA 978-750-8400.

260 • Clinical Breast Cancer October 2004

Table 1 Recommendations for Breast Cancer Surveillance in Women at Risk for Hereditary Breast Cancer6-9 Group

Breast Self-Examination

Clinical Breast Examination

Monthly

1-2 per year, begin at age 25-35

1 per year, begin at age 25-35

CGSC6 FNAHC7

Mammogram

Other –

Not mandatory

2-3 per year, begin at age 20

1 per year, begin at age 30

Research MRI

SIACR8

Monthly

2 per year

1 per year, begin at age 30

Annual ultrasound

NCCN9

Monthly, begin at 18

2 per year, begin at age 25

1 per year, begin at age 25

–

Abbreviations: CGSC = Cancer Genetics Studies Consortium; FNAHC = French National Ad Hoc Committee; SIACR = Swiss Institute for Applied Cancer Research; NCCN = National Comprehensive Cancer Network

garding those outcomes, and to describe investigational screening modalities that may be employed in this limited group of women. This approach may help identify breast cancer, if it arises, at the earliest possible stage.

Prevalence of BRCA Mutations and Risk of Hereditary Breast Cancer The prevalence of germline BRCA mutations in series of patients with breast cancer varies significantly depending on the characteristics of the population studied. Mutations are generally more common in clinic-based, as opposed to population-based, series, because of an obvious bias towards ascertainment in familial cancer clinics of women with significant family histories of breast or ovarian cancer. Mutations are also more likely to be identified in women with an earlier age at breast cancer diagnosis, but even in this group, the prevalence of mutations is only approximately 10% (Table 2).11-16 Mutations are more common if the woman is a member of a population in which founder mutations are relatively prevalent, such as Ashkenazi or Icelandic populations.17-20 The level of breast cancer risk associated with a deleterious BRCA mutation was originally reported to be as high as 87% by the age of 70.21,22 A recent systematic review of 22 case series unselected for family history has proposed that the risk may in fact be somewhat lower.23 In this analysis, the risk of breast cancer by age 70 was 65% (95% CI, 44%-78%) for BRCA1 mutation carriers and 45% (95% CI, 31%-56%) for BRCA2 mutation carriers. In the aggregate analysis, there

appeared to not only be differences in risk between BRCA1 and BRCA2 carriers, but also variation by birth cohort and possibly also by mutation position within the gene. Confirmation that risk may vary over the lifetime of a woman with a mutation is of particular importance, as this information may be used to time the introduction of specific surveillance interventions. In addition, information regarding the risk over defined time periods may be more meaningful to women at risk than lifetime estimates. Table 317,24 presents the estimated annual breast cancer incidence per 1000 BRCA mutation carriers derived from the 22 unselected case series described earlier, with the general population incidence in the United States derived from the Surveillance, Epidemiology, and End Results registry data as a comparator.24 As can be seen, breast cancer incidence in BRCA1 carriers begins to rise at age 25, peaks at 4.3% per year in the 45-49 year age group, and then declines. For BRCA2 carriers, the breast cancer risk also begins to rise at age 25, but does not exceed 1% per year until after age 45, and does not decline with age. These data are consistent with 2 small prospective trials of BRCA mutation carriers. In a report from the Netherlands of 113 BRCA1 and 15 BRCA2 carriers with a mean age at first screen of 37 years, the annual breast cancer incidence was 33 in 1000 (95% CI, 17-63).25 In another series of 156 BRCA1 and 77 BRCA2 carriers followed from a mean age of 46.8 years, the annual incidence of breast cancer was 41 in 1000 (95% CI, 20-62).26 Similar incidence figures can be derived from analysis of

Table 2 Prevalence of BRCA Mutations in Population-Based Series of Breast Cancer Cases11-16 Ascertainment

Number of Patients

BRCA1 Mutations

BRCA2 Mutations

Breast cancer, age 20-74 years

211

3 (2.6%; 95% CI, 0-5.5%)*

–

Breast cancer > 18 years

662

11 (1.6%; 95% CI, 0.8%-2.9%)

–

Breast cancer < 35 years Breast cancer < 45 years, FDR with BC

203 225

12 (5.9%; 95% CI, 3.1%-10.1%) 16 (7.1%; 95% CI, 4.1%-11.3%)

7 (3.4%; 95% CI, 1.4%-7.0%) 11 (4.9%; 95% CI, 2.5%-8.6%)

Peto et al14

Breast cancer < 36 years Breast cancer 36-45 years

254 363

9 (3.5%) 7 (1.9%)

6 (2.4%) 8 (2.2%)

Anglian Group et al15

Breast cancer < 55 years

1220

8 (0.7%; 95% CI, 0.3%-1.3%)

16 (1.3%; 95% CI, 0.8%-2.1%)

Loman et al16

Breast cancer < 41 years

234

16 (6.8%; 95% CI, 4.0%-11.0%)

5 (2.1%; 95% CI, 0.7%-4.9%)

Study Newman et al11 Anton-Culver et al

12

Malone et al13

*Adjusted for sampling probabilities.

Clinical Breast Cancer October 2004 • 261

Screening for Hereditary Breast Cancer Table 3 Breast Cancer Incidence Rates per 1000 Women with BRCA Mutations17,24 Age Group

BRCA117

BRCA217

US Population24

20-24 Years

0.2

0.2

0.0

25-29 Years

1.1

1.2

0.1

30-34 Years

7.4

3.6

0.3

35-39 Years

15.9

7.8

0.6

40-44 Years

29.2

9.1

1.2

45-49 Years

42.8

13.4

1.9

50-54 Years

26.5

17.6

2.3

55-59 Years

30.1

20.0

2.7

60-64 Years

27.0

21.7

3.3

65-69 Years

29.6

23.8

3.8

contralateral breast cancer risk in the Breast Cancer Linkage Consortium data set. The average annual incidence of metachronous cancer was 2.6% in affected BRCA1 carriers,27 and 1.6% in affected BRCA2 mutation carriers.28 It was suggested that the risk of BRCA-associated breast cancer begins to increase significantly at approximately age 25 and, by age 40, reaches approximately 3.0% per year in BRCA1 carriers and 1% per year in BRCA2 carriers. Modifying factors have not been clearly defined; it is likely that some women may experience risks that are higher or lower than these averages.

Effectiveness of Recommended Surveillance Programs Women at hereditary risk who do not choose to undergo risk-reducing mastectomy generally choose to pursue intensified breast cancer surveillance following a program similar to one of those presented in Table 1. The effectiveness of these programs is not established, and, indeed, available data strongly suggest that conventional approaches are suboptimal. In the combined early experience of 2 centers, 21 breast cancers (18 invasive, 3 ductal carcinoma in situ [DCIS]) were diagnosed among 293 mutation carriers undergoing intensified surveillance based on physical examination and mammography.25,26 Of the 18 invasive cancers, 10 (55.6%) were identified clinically during the intervals between radiographic screening, 12 (66.7%) were > 1 cm in maximum diameter, and 8 (44.4%) were associated with axillary lymph node metastases. Thus, it appears that, despite careful screening with self-examination, clinical examination, and mammography, a substantial proportion of BRCA mutation carriers who develop breast cancer will have relatively advanced disease. This poses a significant threat to their survival and will require systemic adjuvant therapy. This sobering finding prompts a more in-depth evaluation of the components of the program, and a consideration of incremental modalities that may improve the results.

262 • Clinical Breast Cancer October 2004

Breast Self-Examination and Clinical Breast Examination Routine breast self-examination (BSE) is commonly recommended by familial cancer clinics for women at hereditary risk (Table 1).10 However, the effectiveness of this technique is not clear. The reported studies of BRCA mutation carriers undergoing screening do not generally specify whether interval breast cancers were identified incidentally through a regularly performed BSE or during a clinical breast examination (CBE). There are no data to support or refute the hypothesis that cancers identified by formal, planned BSE in women at hereditary risk are smaller or more likely to be node-negative than cancers identified incidentally or by clinical exam. In studies of women at average risk, BSE has not been shown to reduce breast cancer mortality, and is associated with an increased likelihood of undergoing a breast biopsy.29-31 Despite these discouraging data, clinical experience suggests that at least some affected women at risk discover an interval breast cancer in the course of self-examination. In one study, 5 of 6 interval breast cancers in BRCA mutation carriers were discovered by the patients themselves.26 Given the relatively high degree of breast cancer risk and relatively low sensitivity of mammographic screening, it seems prudent to educate mutation carriers in BSE. Women at potential risk are made aware of the limitations of the technique and supporting those who feel unable to perform it. There is a similar paucity of data regarding the effectiveness of periodic CBE in women at hereditary risk. Studies of combined CBE and mammography in average risk women report that 3%-45% of breast cancers diagnosed in the course of screening are identified by CBE alone.32,33 Clinical breast examinations may also improve the sensitivity of screening in women at hereditary risk. For example, in 3 studies of women engaged in special surveillance programs, 8%-45% of identified breast cancers were palpable, but mammographically occult.34-36 However, many of these palpable tumors were detected as interval cancers. The sensitivity of CBE as a screening modality in asymptomatic women was low (6.7%-25%) in 3 large recent studies of mutation carriers undergoing multimodality screening.37-39 Published literature does not clearly demonstrate an independent effect of CBE on breast cancer stage at diagnosis or on mortality in women at either average or increased risk. Nonetheless, CBE does appear to marginally improve the breast cancer detection rate compared to mammography alone, even in women at average risk. Because interval breast cancer is a significant problem in mutation carriers, regular CBE is an appropriate adjunct to radiographic screening in women at hereditary risk. No studies have examined the optimal frequency of CBE, but most programs suggest that it be performed at the time of screening mammography and on at least one other occasion during the year, preferably at the 6month interval. Whether quarterly examinations provide additional benefit is not known, but more frequent CBE may be useful for women who are unable or unwilling to perform BSE.

Mark Robson Table 4 Performance of Mammography in Surveillance Studies of Women at Increased Risk25,34,36,40-42 Study

Eligibility

Number of Number Breast Cancers of Patients (Prevalent/Incident)

Mammogram Detected

Invasive/ DCIS

Stage

Brekelmans et al25

BRCA mutation 30%-50% risk 15%-30% risk

449

3/32

21/32 (66%) incident

31/4

4 DCIS 10 ≤ 1.0 cm 20 LN negative

Tilanus-Linthorst et al34

> 25% risk 15%-25% risk

1078 128

26

16 (62%) incident

21/5

5 DCIS 19 ≤ 2.0 cm 16 LN negative

≥ 1 in 6 lifetime

621

19

9 (47%)

17/2

4 ≤ 1.0 cm 9 LN negative

Chart et al40

RR > 4 RR 2-4 RR 1-2

381 204 401

13/10

4/10 (40%) incident

7/3*

3 DCIS 3 ≤ 1.0 cm 7 LN negative

Lalloo et al41

< 50 years ≥ 1 in 6 lifetime

1259 1371

6/8†

12 (86%)

11/3

3 DCIS 7 ≤ 1.0 cm 4 LN negative

Kollias et al42

< 50 years Family history

384 295

11/18

17/29 (59%) overall 8/18 (44%) incident‡

23/6

6 DCIS 15 ≤ 2.0 cm 15 LN negative

Gui et al36

*Incident cancers only.

†Two LCIS excluded. ‡Mammogram every 2 years.

Abbreviations: LN = lymph node; RR = relative risk

Mammography Annual mammography is the cornerstone of all recommended surveillance programs for women at hereditary risk for breast cancer. A number of reports have described the outcomes of screening mammography in women with a family history of breast cancer (Table 4).25,34,36,40-42 The interpretation of these studies is complicated by differing eligibility criteria, varying methods of reporting outcomes, and a frequent lack of distinction between cancers detected at screening by CBE or mammogram. Nevertheless, the results indicate that, in women at increased risk by virtue of a family history, only 40%-66% of incident breast cancers are detected by screening mammography. This is consistent with, but slightly inferior to, the 63%-70% sensitivity reported by the National Cancer Institute Breast Cancer Surveillance Consortium for screening mammograms in younger women (< 50 years of age) with a family history of breast cancer.43 There are few studies specifically describing the outcome of mammography in women with BRCA mutations, but the available data suggest that sensitivity may be even worse than in the overall group of women at familial risk. In one series of 194 mutation carriers, only 5 of 13 cancers (38%) were diagnosed by screening mammogram, with the remainder being identified by interval BSE, MRI, or incidentally at prophylactic surgery.26 In another study of 128 carriers, 5 of 9 invasive cancers (55%) were identified by mammogrambased screening.25 The limited sensitivity of mammography among mutation carriers may result from 1 of 2 factors, which are not mutually exclusive. First, BRCA-associated tumors may develop rapidly, and thus progress from a radiographically undetectable state to a palpable mass in a rela-

tively short period of time. In this regard, BRCA-associated tumors commonly manifest features associated with interval cancers in the more general population of women undergoing screening, such as high proliferative rates and young age at diagnosis.44 Second, host characteristics such as breast density or tumor radiographic features such as smooth margins may reduce the sensitivity of mammography by obscuring the cancer or reducing the radiologist’s level of suspicion. Two small case-control studies have reported that only 38%65% of BRCA-associated breast cancers were identifiable by mammography at the time of diagnosis, which was significantly lower than for noncarriers.45,46 BRCA-associated tumors commonly manifested a smooth, “pushing” margin (as opposed to a spiculated, infiltrative edge), which was a greater contributor to mammographic insensitivity than breast density.46 In summary, the available studies indicate that the results of mammography are suboptimal in women at familial risk, including women with BRCA mutations. Therefore, investigation of supplemental modalities appears warranted. Two that have received considerable attention have been ultrasound (US) imaging and breast MRI.

Screening Breast Ultrasound Increased risk for breast cancer may be a partial explanation for the reduced sensitivity of mammographic screening in these women. An alternative technology, which did not depend on X-ray through-transmission to define a mass, would, in theory, circumvent this limitation and improve the detection of breast cancer. To this end, a number of investigators have evaluated breast US as a supplement to mammogra-

Clinical Breast Cancer October 2004 • 263

Screening for Hereditary Breast Cancer Table 5 Performance of Screening Ultrasound in Women with Mammographically Dense Breasts47,50,51,53,54 Number of Patients

Number of Patients (Biopsy)*

Number of Cancers (PPV)

Cancer Stage

Number Recommended Interval Follow-up

Crystal et al47

1517

21 (1.4%)

7 (33.3%)

4/7 ≤ 1.0 cm 6/7 LN negative

62 (4.0%)

Kaplan et al50

1862

56 (3.0%)

6 (11.8%)

70% ≤ 1.0 cm 89% "stage 0/I"

–

Series

O'Driscoll et al51 Buchberger et Kolb et al54

al53

149*

10 (6.7%)

1 (10%)

Adenoid cystic

72 (3.9%)

8103

330 (4.1%)

32/362 lesions (10.3%)

–

–

13,547 studies (5418 women)

358 (2.6%)

37 (10.3%)

4/6 ≤ 1.0 cm 5/5 LN negative

400 (2.9%)

*“Moderate-risk” patients. Abbreviation: PPV = positive predictive value

phy.47-54 A further potential advantage of breast US is that it could be employed more than once a year without fear of increasing the risk of radiation-induced malignancy. Used in this way, US could theoretically identify interval cancers that arise because of rapid growth rates. The results of the reported US screening studies are presented in Table 5,47,50,51,53,54 and are the basis for a proposed trial to be conducted under the auspices of the American College of Radiology Imaging Network.55 These studies largely describe the results of screening US performed in women with dense breasts Breast Imaging Reporting and Data System [BIRADS] category 3-5. These women had a putatively normal mammogram and CBE, although a number of women with findings on other modalities are also included in some trials. The studies are not restricted to women at increased risk for breast cancer. Overall, the results of these reports are quite consistent. A finding (usually solid mass) requiring core biopsy was identified in 1.4%-6.7% of cases, with a significant number of additional aspirations performed for complex cysts. Breast cancer was identified in 9.7%-33.3% of core biopsy samples, with a positive predictive value of approximately 10% in most studies. The majority of cancers identified were invasive, but were subcentimeter, node-negative lesions. In the largest experience reported, nearly 40% of prevalent nonpalpable cancers were detected with US alone.54 It is important to note that, in addition to the relatively frequent need for aspiration of complex cysts, findings leading to a recommendation for an interval follow-up US were identified in 3%4% of cases. Although US clearly identifies mammographically occult cancers in a small number of women, the positive predictive value of the procedure is modest, and there is a significant likelihood that additional procedures will be required after the screening examination. In addition, there are no studies that rigorously evaluated the issue of sensitivity on the incidence of interval cancer after a normal US. Nonetheless, in some studies of women at high risk undergoing screening, the sensitivity of US has actually been reported to be slightly superior to mammogram. It appears that the sets of breast cancers identified by US and mam-

264 • Clinical Breast Cancer October 2004

mography are overlapping, but not completely congruent. Therefore, it may be advantageous to screen women at risk with dense breasts by mammography and US. There are no randomized trial data demonstrating downstaging or a mortality reduction when an US is added to mammography. However, the studies discussed earlier suggest that the procedure is an attractive incremental option to be discussed with high-risk women undergoing surveillance. In particular, US as an interval screening technique between annual mammograms is theoretically attractive, as it is relatively inexpensive and technologically straight forward. However, the sensitivity of the procedure is likely to be somewhat operator dependent. Women electing to undergo this incremental surveillance may require additional testing to either evaluate or follow up a sonographic abnormality. Because of these limitations, other techniques continue to be of interest, particularly MRI.

Magnetic Resonance Imaging Breast MRI has been under development for a number of years, and advances in both technology and interpretation have made the technique increasingly important in the evaluation and management of women with breast cancer.56,57 As MRI can clearly detect breast cancers that are not identified with other techniques, particularly mammography, it is not surprising that a number of investigators have studied MRI as a potential screening tool.33,35,36,57-65 It is important to note that these studies have been performed in cohorts of women at increased risk, and there are no studies reported that assess MRI in women at average risk. Entry criteria for the reported studies were somewhat variable, with some women having simultaneous screening with MRI and other modalities, and others undergoing MRI only if mammography was unrevealing. Nevertheless, the results of these studies are quite consistent (Table 6).35,37,38,58-63,65 Screening MRI detected a high proportion of the diagnosed breast cancers, including DCIS. Most encouragingly, MRI was able to detect significant numbers of malignancies in women who had putatively normal mammograms, and the sensitivity of MRI, when reported, was substantially better than that of mammography (Table 7).35,37,38,58,60,62,63 These data can

Mark Robson Table 6 Performance of Magnetic Resonance Imaging in Studies of Women at Increased Risk35,37,38,58-63,65 Entry

Number (Studies)

Warner et al35

Family history or BRCA mutation

196

7

Kriege et al37

> 15% risk

1848

Kuhl et al38

“Definite/probable mutation carrier”

Kuhl et al58

“High risk”

Study

Cancer Identified (PPV)

Stage

23 (11.7%)‡

6 (26%)

1 DCIS, 6/6 < 1.0 cm, All node-negative

30

NS

NS

6 DCIS 46% ≤ 1.0 cm 77% node-negative

462

51

86 (18.6%)

49 (57%)

–

192 (363)

9

14 (3.9%)

9 (64%)

2 DCIS, All < 2.0 cm, All node-negative

“High risk,” normal MMG

157

–

28 (17.8%)

5 (18%)

–

BRCA mutation

54 (129)

3

15 (11.6%)

3 (20%)

2 DCIS

Family history or BRCA mutation

105

8

9 (8.6%)

8 (89%)

3 DCIS

> 15% Risk

139 (258)

9

30 (11.6%)†

13 (43%)

3 DCIS 2/10 ≤ 2.0 cm 5/10 node-negative

> 25% Risk > 50% Density

109

3

9 (8.2%)

3 (33%)

2/3 ≤ 1.0 cm All node-negative

“High-risk,” normal MMG

367

14

59 (14.8%)

14 (24%)

8 DCIS, 4/6 node-negative

Lo et al59 Robson et al60 Podo et al61

Number Abnormal (Cancers) MRI*

Stoutjesdijk et al62 Tilanus-Linthorst et al63 Morris et al65

*Defined as screening study leading to biopsy. †Defined as BIRADS equivalent ≥ 3. ‡Defined as BIRADS equivalent ≥ 4.

Abbreviations: MMG = mammography; PPV = positive predictive value

certainly be used to justify offering MRI screening to selected women, but a number of caveats should be noted. The early studies, describing nearly 100% sensitivity for MRI, were from single institutions or limited numbers of centers, presumably with significant experience in the interpretation of breast MRI. The high level of sensitivity is not likely to be maintained when the technique is more widely used by centers with less experience. This is clearly demonstrated by the largest MRI screening series yet reported, involving numerous centers across the Netherlands, in which the sensitivity of MRI was only 71% overall, and 79.5% for invasive disease.37 In addition, in this very large series, 21% of tumors were node positive

at diagnosis and 25% were > 2 cm in maximum diameter. In the subset of mutation carriers, only 82.6% of the cancers were detected by screening, 35% were > 2 cm at diagnosis, and only 63% were completely node negative. Although these results appear to be superior to those achieved without MRI screening, it is clear from this experience that even intensive screening with the best technology available does not guarantee that cancer will be diagnosed at a minimal stage. A second cautionary note that is repeatedly sounded regarding MRI screening is the comparatively low specificity described in most of the reported studies. In these trials, the proportion of MRI studies resulting in a recommendation of

Table 7 Sensitivity and Specificity of MRI and Other Screening Modalities in Studies of Women at High Risk35,37,38,58,60,62,63 Mammography

MRI

Study

Ultrasound

Sensitivity

Specificity

Sensitivity

Specificity

Sensitivity

Specificity

Warner et al35

86% (100% invasive)

91%

43% (33% invasive)

99.5%

43% (60% invasive)

93%

Kriege et al37

71% (83% invasive)

88%

36% (26% invasive)

95%

–

–

Kuhl et al38

96.1%

95%

43%

94.3%

47%

88.4%

al58

100%

95%

33%

93%

33%

80%

100%

83.3%

–

–

–

–

100%*

93%

–

–

–

–

100%

94%

42%

96%

–

–

Kuhl et

Robson et al60 Stoutjesdijk et

al62

Tilanus-Linthorst et al63

*Threshold of BIRADS equivalent category 3.

Clinical Breast Cancer October 2004 • 265

Screening for Hereditary Breast Cancer biopsy ranges from 3.9% to 18.6%, with most reports clustering in the 8%-11% range. The positive predictive values of a biopsy recommendation ranged from 18% to 89% (Table 6), and the calculated specificity of MRI ranges from 83.3% to 95% (Table 7) but is generally lower than that of mammography. There is some evidence that specificity improves after the first screening round39 and also with reader experience.60 Fortunately, because of the high disease prevalence in the high-risk populations that have been studied, the positive predictive value of an abnormal MRI remains quite high despite suboptimal specificity. In groups of women at lesser degrees of risk, the positive predictive value will likely be considerably lower, which raises concerns about the rapid deployment of screening MRI into lower-risk groups before ways have been found to reduce false-positive readings. A number of features that may be useful in the prediction of malignancy have been proposed, such as enhancement pattern and kinetics, T2 signal intensity, character of the margins of a mass, shape of the mass, and level of clinical suspicion graded on a BIRADS-like scale. Although multiple different parameters may be useful in adjusting the level of suspicion, no one feature or combination of features can be used as an absolute discriminator between benign and malignant lesions.65 It should be noted that the calculations of specificity reported in the literature generally define a false-positive examination as one in which a biopsy is recommended which does not result in a diagnosis of malignancy. However, a “probably benign” finding that leads to a recommendation for further studies or a repeat interval MRI has been described in 11%-24% of studies.60,64 Although most of these findings remain stable on follow-up examination, and do not represent cancer, interval examination is required as up to 10% of these women develop malignant disease in the area of initial concern.66 Although, “probably benign” findings requiring interval follow-up or further evaluation are not included in the calculation of specificity, the additional investigations required might generate considerable anxiety on the part of women who already perceive themselves to be at very high risk. Studies have not yet elucidated the impact of this anxiety on QOL, adherence to future screening, and decisions regarding prophylactic surgery. Of course, the impact of the anxiety of screening must be balanced by each prospective participant against the negative impact of a diagnosis of advanced breast cancer. In summary, breast MRI offers significant promise as an incremental screening modality for women at hereditary risk for developing breast cancer. Sensitivity appears to be very high overall, and it is clear that this technique can detect a significant proportion of cancers that are occult to mammography and physical examination. It has not yet been formally demonstrated that the incorporation of MRI into a surveillance program results in downstaging of the cancers that are diagnosed, but it seems likely that this will be the case. An impact on survival has not, of course been demonstrated, and no randomized trials have been performed. To continue the

266 • Clinical Breast Cancer October 2004

development of breast MRI as a screening modality, eligible women should be enrolled on clinical trials of this promising procedure. In the absence of a suitable trial, it seems reasonable to offer women at the highest levels of risk screening offstudy. Prior to off-study screening, however, a frank discussion should be held, outlining the significant risk of a falsepositive examination and exploring the effect that such a result may have on the woman's QOL. Currently, unresolved issues surrounding the specificity of MRI are likely to limit its utility in the screening of women at lesser degrees of increased risk, such as women with a single affected first-degree relative. The positive predictive value of MRI is likely to be low in this population, making it difficult to justify the incremental expense and anxiety of routine MRI screening in this group. However, the specificity of MRI is likely to improve with greater experience among breast radiologists. Once the learning curve has been climbed, further evaluation and definition of the utility of MRI in women at lesser degrees of risk will be warranted.

Surveillance Program for Women at Hereditary Risk: 2004 Nearly 10 years after the discovery of BRCA1, few data comparing different surveillance regimens are available for women at the highest levels of breast cancer risk, and the optimum regimen remains a matter of expert opinion. It is clear that women with BRCA mutations are at significant risk, with an annual breast cancer incidence of 2%-4%. It is also clear that screening with annual mammogram and semiannual CBE is suboptimal, as a significant number of interval cancers are diagnosed in women following such a program. These are not infrequently unacceptably large tumors and/or metastatic to axillary nodes. For this reason, it seems prudent to recommend routine BSE, so that such interval cancers may be detected as quickly as possible. More frequent (eg, quarterly) clinical examinations may be an appropriate substitute for women who are unable to perform BSE for reasons of anxiety, or for women whose breasts are particularly difficult to examine.

Conclusion Given the limited sensitivity of mammography in BRCA mutation carriers, supplemental radiographic screening appears sensible, although it has not been clearly shown that such incremental screening results in cancer downstaging or an improvement in mortality. Preliminary studies suggest that both US and MRI can improve the sensitivity screening. In experienced hands, MRI is the more sensitive of the 2 incremental techniques and is preferred, if appropriate expertise is available. A woman undergoing MRI screening should be willing to assume the attendant psychological risk that she will have a false-positive examination, requiring further studies, biopsy, or interval follow-up. There are theoretical advantages to performing MRI (if available) or US imaging at 6-month intervals between mammograms, in order to identify rapidly growing

Mark Robson interval cancers at their earliest possible stage. Further research is clearly needed to address the optimal integration of different screening technologies, to improve the specificity of the newer techniques, and, perhaps most important, to determine how best to incorporate these technologies into the care of women at risk without producing a counterproductive disruption of their QOL.

References 1. Miki Y, Swensen J, Shattuck-Eidens D, et al. A strong candidate for the breast and ovarian cancer susceptibility gene BRCA1. Science 1994; 266:66-71. 2. Wooster R, Bignell G, Lancaster J, et al. Identification of the breast cancer susceptibility gene BRCA2. Nature 1995; 378:789-792. 3. Frank TS, Deffenbaugh AM, Reid JE, et al. Clinical characteristics of individuals with germline mutations in BRCA1 and BRCA2: analysis of 10,000 individuals. J Clin Oncol 2002; 20:1480-1490. 4. Bluman LG, Rimer BK, Berry DA, et al. Attitudes, knowledge, and risk perceptions of women with breast and/or ovarian cancer considering testing for BRCA1 and BRCA2. J Clin Oncol 1999; 17:1040-1046. 5. Chen WY, Garber JE, Higham S, et al. BRCA1/2 genetic testing in the community setting. J Clin Oncol 2002; 20:4485-4492. 6. Burke W, Daly M, Garber J, et al. Recommendations for follow-up care of individuals with an inherited predisposition to cancer. II. BRCA1 and BRCA2. Cancer Genetics Studies Consortium. JAMA 1997; 277:997-1003. 7. Eisinger F, Alby N, Bremond A, et al. Recommendations for medical management of hereditary breast and ovarian cancer: the French National Ad Hoc Committee. Ann Oncol 1998; 9:939-950. 8. Pichert G, Bolliger B, Buser K, et al. Evidence-based management options for women at increased breast/ovarian cancer risk. Ann Oncol 2003; 14:9-19. 9. NCCN Clinical Practice Guidelines in Oncology. http://www.nccn.org/ professionals/physician_gls/default.asp. Accessed September 13, 2003. 10. Vasen HF, Haites NE, Evans DG, et al. Current policies for surveillance and management in women at risk of breast and ovarian cancer: a survey among 16 European family cancer clinics. European Familial Breast Cancer Collaborative Group. Eur J Cancer 1998; 34:1922-1926. 11. Newman B, Mu H, Butler LM, et al. Frequency of breast cancer attributable to BRCA1 in a population-based series of American women. JAMA 1998; 279:915-921. 12. Anton-Culver H, Cohen PF, Gildea ME, et al. Characteristics of BRCA1 mutations in a population-based case series of breast and ovarian cancer. Eur J Cancer 2000; 36:1200-1208. 13. Malone KE, Daling JR, Neal C, et al. Frequency of BRCA1/BRCA2 mutations in a population-based sample of young breast carcinoma cases. Cancer 2000; 88:1393-1402. 14. Peto J, Collins N, Barfoot R, et al. Prevalence of BRCA1 and BRCA2 gene mutations in patients with early-onset breast cancer. J Natl Cancer Inst 1999; 91:943-949. 15. Anglian Breast Cancer Study Group. Prevalence and penetrance of BRCA1 and BRCA2 mutations in a population-based series of breast cancer cases. Br J Cancer 2000; 83:1301-1308. 16. Loman N, Johannsson O, Kristoffersson U, et al. Family history of breast and ovarian cancers and BRCA1 and BRCA2 mutations in a population-based series of early-onset breast cancer. J Natl Cancer Inst 2001; 93:1215-1223. 17. Thorlacius S, Sigurdsson S, Bjarnadottir H, et al. Study of a single BRCA2 mutation with high carrier frequency in a small population. Am J Hum Genet 1997; 60:1079-1084. 18. Thorlacius S, Struewing JP, Hartge P, et al. Population-based study of risk of breast cancer in carriers of BRCA2 mutation. Lancet 1998; 352:1337-1339. 19. Tonin P, Weber B, Offit K, et al. Frequency of recurrent BRCA1 and BRCA2 mutations in Ashkenazi Jewish breast cancer families. Nat Med 1996; 2:1179-1183. 20. Warner E, Foulkes W, Goodwin P, et al. Prevalence and penetrance of BRCA1 and BRCA2 gene mutations in unselected Ashkenazi Jewish women with breast cancer. J Natl Cancer Inst 1999; 91:1241-1247. 21. Ford D, Easton DF, Bishop DT, et al. Risks of cancer in BRCA1-mutation carriers. Breast Cancer Linkage Consortium. Lancet 1994; 343:692-695. 22. Ford D, Easton DF, Stratton M, et al. Genetic heterogeneity and penetrance analysis of the BRCA1 and BRCA2 genes in breast cancer families. The Breast Cancer Linkage Consortium. Am J Hum Genet 1998; 62:676-689. 23. Antoniou A, Pharoah PD, Narod S, et al. Average risks of breast and ovarian cancer associated with BRCA1 or BRCA2 mutations detect-

ed in case series unselected for family history: a combined analysis of 22 studies. Am J Hum Genet 2003; 72:1117-1130. 24. Surveillance, Epidemiology, and End Results (SEER) Program, April 2003. National Cancer Institute, DCCPS, Surveillance Research Program, Cancer Statistics Branch. Available at: www.seer.cancer.gov. Accessed September 13, 2003. 25. Brekelmans CT, Seynaeve C, Bartels CC, et al. Effectiveness of breast cancer surveillance in BRCA1/2 gene mutation carriers and women with high familial risk. J Clin Oncol 2001; 19:924-930. 26. Scheuer L, Kauff ND, Robson M, et al. Outcome of preventive surgery and screening for breast and ovarian cancer in BRCA mutation carriers. J Clin Oncol 2002; 20:1260-1268. 27. Easton DF, Ford D, Bishop DT. Breast and ovarian cancer incidence in BRCA1-mutation carriers. Breast Cancer Linkage Consortium. Am J Hum Genet 1995; 56:265-271. 28. Breast Cancer Linkage Consortium. Cancer risks in BRCA2 mutation carriers. J Natl Cancer Inst 1999; 91:1310-1316. 29. Kosters JP, Gotzsche PC. Regular self-examination or clinical examination for early detection of breast cancer. Cochrane Database Syst Rev 2003; CD003373. 30. Hackshaw AK, Paul EA. Breast self-examination and death from breast cancer: a meta-analysis. Br J Cancer 2003; 88:1047-1053. 31. Thomas DB, Gao DL, Ray RM, et al. Randomized trial of breast selfexamination in Shanghai: final results. J Natl Cancer Inst 2002; 94:1445-1457. 32. Jatoi I. Screening clinical breast examination. Surg Clin North Am 2003; 83:789-801. 33. Barton MB, Harris R, Fletcher SW. The rational clinical examination. Does this patient have breast cancer? The screening clinical breast examination: should it be done? How? JAMA 1999; 282:1270-1280. 34. Tilanus-Linthorst MM, Bartels CC, Obdeijn AI, et al. Earlier detection of breast cancer by surveillance of women at familial risk. Eur J Cancer 2000; 36:514-519. 35. Warner E, Plewes DB, Shumak RS, et al. Comparison of breast magnetic resonance imaging, mammography, and ultrasound for surveillance of women at high risk for hereditary breast cancer. J Clin Oncol 2001; 19:3524-3531. 36. Gui GP, Hogben RK, Walsh G, et al. The incidence of breast cancer from screening women according to predicted family history risk: Does annual clinical examination add to mammography? Eur J Cancer 2001; 37:1668-1673. 37. Kriege M, Brekelmans CTM, Boetes C, et al. Efficacy of MRI and mammography for breast-cancer screening in women with a familial or genetic predisposition. N Engl J Med 2004; 351:427-437. 38. Kuhl CK, Schrading S, Leutner CC, et al. Surveillance of "high risk" women with proven or suspected familial (hereditary) breast cancer: First mid-term results of a multi-modality clinical screening trial. Proc Am Soc Clin Oncol 2003; 22:2 (Abstract #4). 39. Warner E, Plewes DB, Shumak RS, et al. Surveillance of BRCA1 and BRCA2 mutation carriers with magnetic resonance imaging, ultrasound, mammography, and clinical breast examination. J Am Med Assoc 2004; 292:1317-1325. 40. Chart PL, Franssen E. Management of women at increased risk for breast cancer: preliminary results from a new program. CMAJ 1997; 157:1235-1242. 41. Lalloo F, Boggis CR, Evans DG, et al. Screening by mammography, women with a family history of breast cancer. Eur J Cancer 1998; 34:937-940. 42. Kollias J, Sibbering DM, Blamey RW, et al. Screening women aged less than 50 years with a family history of breast cancer. Eur J Cancer 1998; 34:878-883. 43. Kerlikowske K, Carney PA, Geller B, et al. Performance of screening mammography among women with and without a first-degree relative with breast cancer. Ann Intern Med 2000; 133:855-863. 44. Gilliland FD, Joste N, Stauber PM, et al. Biologic characteristics of interval and screen-detected breast cancers. J Natl Cancer Inst 2000; 92:743-749. 45. Goffin J, Chappuis PO, Wong N, et al. Re: Magnetic resonance imaging and mammography in women with a hereditary risk of breast cancer. J Natl Cancer Inst 2001; 93:1754-1755. 46. Tilanus-Linthorst M, Verhoog L, Obdeijn IM, et al. A BRCA1/2 mutation, high breast density and prominent pushing margins of a tumor independently contribute to a frequent false-negative mammography. Int J Cancer 2002; 102:91-95. 47. Crystal P, Strano SD, Shcharynski S, et al. Using sonography to screen women with mammographically dense breasts. AJR Am J Roentgenol 2003; 181:177-182. 48. Kolb TM, Lichy J, Newhouse JH. Occult cancer in women with dense breasts: detection with screening US—diagnostic yield and tumor characteristics. Radiology 1998; 207:191-199. 49. Buchberger W, DeKoekkoek-Doll P, Springer P, et al. Incidental findings on sonography of the breast: clinical significance and diagnostic workup. AJR Am J Roentgenol 1999; 173:921-927.

Clinical Breast Cancer October 2004 • 267

Screening for Hereditary Breast Cancer 50. Kaplan SS. Clinical utility of bilateral whole-breast US in the evaluation of women with dense breast tissue. Radiology 2001; 221:641-649. 51. O'Driscoll D, Warren R, Mackay J, et al. Screening with breast ultrasound in a population at moderate risk due to family history. J Med Screen 2001; 8:106-109. 52. Gordon PB, Goldenberg SL. Malignant breast masses detected only by ultrasound. A retrospective review. Cancer 1995; 76:626-630. 53. Buchberger W, Niehoff A, Obrist P, et al. Clinically and mammographically occult breast lesions: detection and classification with high-resolution sonography. Semin Ultrasound CT MR 2000; 21:325-336. 54. Kolb TM, Lichy J, Newhouse JH. Comparison of the performance of screening mammography, physical examination, and breast US and evaluation of factors that influence them: an analysis of 27,825 patient evaluations. Radiology 2002; 225:165-175. 55. Berg WA. Rationale for a trial of screening breast ultrasound: American College of Radiology Imaging Network (ACRIN) 6666. Am J Roentgenol 2003; 180:1225-1228. 56. Orel SG, Schnall MD. MR imaging of the breast for the detection, diagnosis, and staging of breast cancer. Radiology 2001; 220:13-30. 57. Kneeshaw PJ, Turnbull LW, Drew PJ. Current applications and future direction of MR mammography. Br J Cancer 2003; 88:4-10. 58. Kuhl CK, Schmutzler RK, Leutner CC, et al. Breast MR imaging screening in 192 women proved or suspected to be carriers of a breast cancer susceptibility gene: preliminary results. Radiology

268 • Clinical Breast Cancer October 2004

2000; 215:267-279. 59. Lo L, Rosen M, Schnall M, et al. Pilot study of breast MR screening of a high-risk cohort. Radiology 2001; 221:432. 60. Robson M, Morris EA, Kauff N, et al. Breast cancer screening utilizing magnetic resonance imaging (MRI) in carriers of BRCA mutations. Proc Am Soc Clin Oncol 2003; 22:91 (Abstract #362). 61. Podo F, Sardanelli F, Canese R, et al. The Italian multi-centre project on evaluation of MRI and other imaging modalities in early detection of breast cancer in subjects at high genetic risk. J Exp Clin Cancer Res 2002; 21(3 Suppl):115-124. 62. Stoutjesdijk MJ, Boetes C, Jager GJ, et al. Magnetic resonance imaging and mammography in women with a hereditary risk of breast cancer. J Natl Cancer Inst 2001; 93:1095-1102. 63. Tilanus-Linthorst MM, Obdeijn IM, Bartels KC, et al. First experiences in screening women at high risk for breast cancer with MR imaging. Breast Cancer Res Treat 2000; 63:53-60. 64. Warren R, Hayes C, Pointon LJ, et al. Is the specificity of breast MRI suitable for screening asymptomatic women at high risk? Radiology 2001; 221(suppl):432. 65. Morris EA, Liberman L, Ballon DJ, et al. MRI of occult breast carcinoma in a high-risk population. AJR Am J Roentgenol 2003; 181:619-626. 66. Liberman L, Morris EA, Benton CL, et al. Probably benign lesions at breast magnetic resonance imaging: preliminary experience in high-risk women. Cancer 2003; 98:377-388.