Efficacy of MRI and Mammography for Breast-Cancer Screening in Women with a Familial or Genetic Predisposition
http://www.100md.com
《新英格兰医药杂志》
ABSTRACT
Background The value of regular surveillance for breast cancer in women with a genetic or familial predisposition to breast cancer is currently unproven. We compared the efficacy of magnetic resonance imaging (MRI) with that of mammography for screening in this group of high-risk women.
Methods Women who had a cumulative lifetime risk of breast cancer of 15 percent or more were screened every six months with a clinical breast examination and once a year by mammography and MRI, with independent readings. The characteristics of the cancers that were detected were compared with the characteristics of those in two different age-matched control groups.
Results We screened 1909 eligible women, including 358 carriers of germ-line mutations. Within a median follow-up period of 2.9 years, 51 tumors (44 invasive cancers, 6 ductal carcinomas in situ, and 1 lymphoma) and 1 lobular carcinoma in situ were detected. The sensitivity of clinical breast examination, mammography, and MRI for detecting invasive breast cancer was 17.9 percent, 33.3 percent, and 79.5 percent, respectively, and the specificity was 98.1 percent, 95.0 percent, and 89.8 percent, respectively. The overall discriminating capacity of MRI was significantly better than that of mammography (P<0.05). The proportion of invasive tumors that were 10 mm or less in diameter was significantly greater in our surveillance group (43.2 percent) than in either control group (14.0 percent and 12.5 percent , respectively). The combined incidence of positive axillary nodes and micrometastases in invasive cancers in our study was 21.4 percent, as compared with 52.4 percent (P<0.001) and 56.4 percent (P=0.001) in the two control groups.
Conclusions MRI appears to be more sensitive than mammography in detecting tumors in women with an inherited susceptibility to breast cancer.
The cumulative lifetime risk of breast cancer among Dutch women is approximately 11 percent.1 A family history of breast cancer or the presence of a germ-line mutation of the BRCA1 or BRCA2 gene increases this risk considerably and is often associated with a diagnosis at a young age.2,3 Among high-risk women, the risk of breast cancer can be reduced by prophylactic mastectomy,4,5 prophylactic oophorectomy,6,7 or chemoprevention.8 Early diagnosis as a result of intensive surveillance may also decrease the rate of death from breast cancer.
Randomized trials have shown that mammographic screening of all women who are between 50 and 70 years of age can reduce mortality from breast cancer by about 25 percent.9 Although these findings were recently disputed,10 there is a consensus among clinicians that breast-cancer screening of women in this age group is effective. Screening is one of the main factors contributing to the decrease in mortality associated with breast cancer in the Netherlands.11 However, there is no consensus about the value of breast-cancer screening among women who are 40 to 49 years old.12,13,14 One of the reasons for the lack of agreement is the difficulty in detecting tumors by mammographic screening in younger women, who have denser breasts than postmenopausal women.15,16 Although screening is frequently offered to women with a genetic predisposition to breast cancer who are under the age of 50 years, the efficacy of this approach is unproven. Preliminary results of surveillance by mammography and clinical breast examination in such women showed that mammographic screening has a low sensitivity for detecting tumors, especially in carriers of a BRCA mutation.17,18,19,20,21 Possible reasons, apart from the high rate of growth of tumors in women with such mutations, include the atypical changes seen on screening mammograms and specific histopathological characteristics in carriers of BRCA mutations, as compared with noncarriers of the same age.22,23,24
In a diagnostic setting, magnetic resonance imaging (MRI) is a sensitive method of breast imaging, and it is virtually uninfluenced by breast density, but the specificity is variable and the costs are high.25,26,27 Because MRI may improve the sensitivity of screening in women with a familial or genetic predisposition to breast cancer, we prospectively compared MRI with mammography for screening women with such a predisposition in order to determine whether screening with MRI facilitated the early diagnosis of hereditary breast cancer.
Methods
Study Population
The design of our MRI screening study, in which six subcommittees in different disciplines were involved, has been described previously.28 Between November 1, 1999, and October 1, 2003, 1952 women with a genetic risk of breast cancer were recruited for the study by six familial-cancer clinics in the Netherlands. The six centers were Erasmus Medical Center–Daniel den Hoed Cancer Center, Rotterdam; the Netherlands Cancer Institute, Amsterdam; University Medical Center Nijmegen, Nijmegen; Leiden University Medical Center, Leiden; University Hospital Groningen, Groningen; and Free University Medical Center, Amsterdam. The study was approved by the ethics committees of all the centers. All the women who participated gave written informed consent.
The inclusion criteria for participation were a cumulative lifetime risk of breast cancer of 15 percent or more owing to a familial or genetic predisposition, according to the modified tables of Claus et al.,29 and an age of 25 to 70 years. Women could be tested at an age younger than 25 if they had a family history of breast cancer diagnosed before the age of 30 years, since testing began at an age 5 years younger than that at which the youngest family member was found to have breast cancer. Women with symptoms that were suggestive of breast cancer or women who had a personal history of breast cancer were excluded.
Surveillance
Surveillance consisted of a clinical breast examination performed by an experienced physician every six months and imaging studies performed annually by experienced radiologists. The imaging included a mammographic study (oblique and craniocaudal views and, if necessary, compression views or magnifications) and a dynamic breast MRI with gadolinium-containing contrast medium according to a standard protocol.25 Whenever possible, both imaging investigations were performed on the same day or in the same time period, between day 5 and day 15 of the menstrual cycle. The results of mammography and MRI were scored in a standardized way, according to the Breast Imaging Reporting and Data System (BI-RADS) classification,30,31 and the results were blinded so that the two examinations were not linked. When one of the examinations was scored as either BI-RADS category 3 ("probably benign finding") or category 0 ("need additional imaging evaluation"), further investigation by ultrasonography with or without fine-needle aspiration was advised, or mammography or MRI was repeated. When one of the two examinations was scored as BI-RADS category 4 ("suspicious abnormality") or category 5 ("highly suggestive of malignancy"), a cytologic or histologic evaluation of a biopsy specimen was performed. When the results of mammography and MRI were negative but the findings on clinical breast examination were rated as uncertain or suspicious, additional investigation was also performed. The diagnosis of malignant tumors was based on the results of a histologic examination. One of the investigators, an expert pathologist, reviewed all the biopsy specimens that formed the basis for the diagnosis of breast cancer.
Statistical Analysis
The women were divided into three categories according to the cumulative lifetime risk of breast cancer, as follows: carriers of the BRCA1 or BRCA2 or other mutations (cumulative lifetime risk, 50 to 85 percent), a high-risk group (risk, 30 to 49 percent), and a moderate-risk group (risk, 15 to 29 percent).28,29 The characteristics of the women in each risk group were compared by analysis of variance or Pearson's chi-square test.
The rates of detection of breast cancer for the group as a whole and for each of the three risk groups were calculated, and a Poisson distribution was assumed in order to calculate the 95 percent confidence intervals. Person-years at risk were calculated from the date of the first examination, irrespective of the type of examination, to the date of detection of breast cancer, bilateral prophylactic mastectomy, or death; the date that a patient stopped surveillance; or the cutoff date for this analysis (October 1, 2003). An "interval cancer" was defined as a carcinoma detected between two rounds of screening after initially negative findings on screening. In our analysis, we defined as positive a mammographic or MRI study with a BI-RADS score of 0, 3, 4, or 5 and a clinical breast examination that was classified as "uncertain" or "suspicious," because those were the results that triggered an additional examination.
To compare the three different screening methods, we calculated the sensitivity, specificity, and positive predictive value of each. The sensitivity used is that of one screening method relative to the others, meaning that a test result is a false negative when a proven cancer (diagnosed on the basis of a histologic examination) is detected in the interval or by one of the other methods. Receiver-operating-characteristic (ROC) curves for the two imaging methods were generated. The area under the curve was used as an index in evaluating the inherent capacity of a screening method to discriminate between "positive" and "negative" cases. We used a z-test to compare the area under the curve for the results of mammography and MRI. For the analysis of the screening variables, we used only the screening data that included the results of both mammography and MRI.
To determine whether breast cancer was diagnosed by screening at a stage more favorable to treatment, the characteristics of breast tumors detected in the study group were compared with those in two control groups. The first control group was derived from all women who had breast cancers diagnosed in 1998 in the Netherlands. These data were obtained from the National Cancer Registry. The second control group consisted of unselected patients who had received a diagnosis of primary breast cancer in Leiden or Rotterdam between 1996 and 2002 and who were participating in a prospective study of the prevalence of gene mutations.32 Subjects in both control groups were matched for age with the patients in the study group (in five-year categories). From this series of consecutive patients in the second control group, we chose all the unscreened patients who were between 25 and 60 years old and whose cumulative lifetime risk of breast cancer was more than 15 percent because of a family history of the disease — information that was routinely recorded in this database. The differences in tumor characteristics between the study group and the control groups were tested with the use of Pearson's chi-square test or the chi-square test for trend. A two-sided P value of less than 0.05 was considered to indicate statistical significance. All statistical analyses were performed with the use of SPSS software (version 9.0).
Results
Study Population
Of the women who were invited to participate in the study, 90 percent agreed. Initially, 1952 women were included; 8 withdrew from the study before their first screening visit and another 35 were excluded because they ultimately proved not to be carriers in a family with a proven mutation and therefore had less than a 15 percent cumulative lifetime risk of breast cancer. Of the 1909 remaining women, 88 (4.6 percent) left the study or were lost to surveillance before October 1, 2003; 65 of these 88 women underwent prophylactic mastectomy. Another 89 women (4.7 percent) remained under surveillance but later refused screening by MRI, because of claustrophobia or for other reasons.
Table 1 lists the characteristics of the 1909 women according to risk category. The mean age at entry was 40 years (range, 19 to 72). Within the group of 358 carriers of pathogenic mutations, 276 had a BRCA1 mutation, 77 had a BRCA2 mutation, 1 had both a BRCA1 and a BRCA2 mutation, 2 had a PTEN mutation, and 2 had a TP53 mutation.
Table 1. Characteristics of Participating Women at Study Entry, According to Risk Group.
Breast Cancers
From November 1, 1999, to October 1, 2003, 51 malignant tumors (44 invasive breast cancers, 6 ductal carcinomas in situ, and 1 non-Hodgkin's lymphoma) were detected (Figure 1), during a median follow-up period of 2.9 years (mean 2.7, range, 0.1 to 3.9 years); 1 lobular carcinoma in situ was also found. Table 2 shows the detection rate for the whole group and separately for the different risk groups. The overall rate of detection for all breast cancers (invasive plus in situ) was 9.5 per 1000 woman-years at risk (95 percent confidence interval, 7.1 to 12.3), with the highest rate (26.5 per 1000) in the group of women who were carriers of the BRCA1, BRCA2, PTEN, and TP53 mutations.
Figure 1. Women at Increased Risk for Breast Cancer Enrolled and Tumors Detected.
A total of 1952 women were enrolled, of whom 1909 were eligible for this analysis.
Table 2. Detection of Cases of Breast Cancer (Including Ductal Carcinoma in Situ) According to Risk Group.
Performance of the Screening Methods
Table 3 shows the results with the three screening methods. Of the 50 breast cancers that were detected, 5 were excluded from the analysis (Table 3). The 45 cancers that were evaluated in the comparison of the methods included 4 interval cancers (i.e., cancers detected between two episodes of screening). The first was symptomatic (30 mm in diameter, node-negative), detected seven months after screening by imaging and clinical breast examination and one month after screening by clinical breast examination only. The second (4 mm, node-negative) was detected in a specimen from a prophylactic mastectomy. The third was symptomatic (45 mm, node-negative) and was detected seven months after screening by imaging; the fourth, also symptomatic (13 mm, with isolated tumor cells in a lymph node), was detected three months after screening by imaging.
Table 3. Sensitivity, Specificity, and Positive Predictive Value (PPV) of the Three Screening Methods.
Overall, 32 breast cancers were found by MRI (22 of these were not visible on mammography), whereas 13 were missed by MRI (8 of the 13 were visible on mammography, including 5 ductal carcinomas in situ; 4 were interval cancers; and 1 tumor was detected only by clinical breast examination). In this group of 45 breast cancers, mammographic screening detected 18 tumors (10 of these were visible by MRI) and missed 27 tumors (including the 22 that were visible on MRI, the 4 interval cancers, and the 1 that was detected only by clinical breast examination).
With respect to all breast cancers (invasive and ductal carcinoma in situ), the sensitivity of clinical breast examination, mammography, and MRI was 17.8 percent, 40.0 percent, and 71.1 percent, respectively, when the BI-RADS score was 3 or higher (Table 3). For invasive cancers only, the respective percentages were 17.9 percent, 33.3 percent, and 79.5 percent. The specificity was 98.1 percent for clinical breast examination, 95.0 percent for mammography, and 89.8 percent for MRI.
Of the 41 cancers found by screening, 22 were detected at the first imaging screening in the study; of the women in whom cancer was detected, 16 had undergone mammographic screening before the start of the study. Two of the interval cancers were detected after the first imaging screening, and two others after a subsequent imaging screening. The sensitivity of mammography was 37.5 percent for the first screening and 42.9 percent for subsequent screening (P=0.71). The sensitivity of MRI was 79.2 percent for the first screening and 61.9 percent for subsequent screening (P=0.20).
Among the 83 clinical breast examinations with findings that were judged as probably benign or suspicious, or highly suggestive of cancer, 8 cases of malignant disease were confirmed, for a positive predictive value of 9.6 percent (Table 3). Among the 225 mammograms with findings categorized as BI-RADS 3 or higher, 18 cases of malignant disease were confirmed, for a positive predictive value of 8.0 percent. A total of 32 cancers were confirmed among 452 MRI screenings with such findings, for a positive predictive value of 7.1 percent (Table 3). With a cutoff level of BI-RADS 4, the sensitivity for both imaging methods decreased, whereas the specificity increased.
To evaluate the discriminating capacity of the imaging methods, we generated ROC curves (Figure 2). The area under the curve was 0.686 for mammography and 0.827 for MRI; the difference between the areas was 0.141 (95 percent confidence interval, 0.020 to 0.262; P<0.05).
Figure 2. Receiver-Operating-Characteristic Curves for Mammography and MRI.
The difference between the area under the curve (AUC) for mammography and the AUC for MRI was 0.141 (95 percent confidence interval, 0.020 to 0.262; P<0.05).
Additional Investigations
Ultrasonography was performed 889 times in 627 different women according to the protocol. Fine-needle aspiration was carried out 312 times: 267 times in combination with ultrasonography and 45 times with palpation. Biopsy was performed 85 times in 82 women and showed malignant disease in 50 cases and 1 lobular carcinoma in situ, making the rate of positive histologic findings 60.0 percent. Sixty-seven of these 85 biopsies were performed after a screening visit at which both MRI and mammography were performed. Of the 25 biopsies in women who had mammographic findings with a score of 3 or higher, 7 (28.0 percent) showed no cancer. Of the 56 biopsies in women who had MRI findings with a score of 3 or higher, 24 (42.9 percent) showed no cancer (Table 3). One of the 51 tumors was found in a specimen from a prophylactic mastectomy.
Tumor Characteristics
Table 4 compares the characteristics of tumors found in the study group with those of tumors in the two age-matched control groups. In the study group, 19 of the 44 women with an invasive breast cancer (43.2 percent) had a small tumor (10 mm in diameter) — a proportion that was significantly higher than that in the first control group (14.0 percent, P<0.001) or the second control group (12.5 percent, P=0.04). Six of 42 invasive tumors (14.3 percent) with known axillary status in the study group were node-positive and 3 (7.1 percent) had micrometastases (combined total, 21.4 percent). This rate was significantly lower than those in both control groups, in which the rates of node-positive cancer were 52.4 percent (P<0.001) and 56.4 percent (P=0.001), respectively. There were no major differences between the study and control groups with respect to histologic features, with the exception of a relatively high incidence of the medullary type in the study group (11.3 percent, vs. 1.8 percent in the first control group). In the study group, a high proportion of grade 1 tumors were in women at high risk (68.8 percent) or moderate risk (75.0 percent); however, the group of women with BRCA1, BRCA2, or other mutations had a high percentage of grade 3 tumors (63.2 percent), in addition to a high percentage of tumors that were negative for steroid receptors (Table 4).
Table 4. Characteristics of Women with Breast Cancer and Breast Cancers Detected in the Three Risk Groups and in the Two Control Groups.
Disease-free and Overall Survival
In the study group, none of the 50 patients with breast cancer (44 with invasive cancer and 6 with ductal carcinoma in situ) died before the end of the study period; the total follow-up after diagnosis was 87.6 woman-years for these 50 patients (median, 1.5 years). Contralateral breast cancer occurred in one patient. The patient with non-Hodgkin's lymphoma died.
Discussion
In this prospective study, we compared the efficacy of mammographic and MRI screening for breast cancer in women with a family history of the disease or a genetic predisposition to breast cancer. Among the women examined by both methods at the same screening visit, we detected 45 breast cancers (including 6 ductal carcinomas in situ): 32 by MRI (sensitivity, 71.1 percent) and 18 by mammography (40.0 percent); five other patients were excluded from this comparison for various reasons (Table 3). Thus, the sensitivity of MRI was higher than that of mammography, but both the specificity and positive predictive value of MRI were lower.
In our sensitivity and specificity calculations, we defined lesions that were in BI-RADS category 3 and higher as positive, but most other authors have included in their calculations only lesions in BI-RADS categories 4 and 5 as positive.21,33,34 If we had followed that policy, the sensitivity would have been 24.4 percent for mammography and 46.6 percent for MRI, in accord with the higher sensitivity previously reported for MRI.21,33,35,36 However, the previous studies enrolled small groups of women, included some retrospective data,35 evaluated heterogeneous groups that included women with previous breast cancers,21,33,36 or had a plan for follow-up after a suspicious finding on MRI that differed from the follow-up plan for a suspicious mammographic finding.33 All these factors might have artificially increased the sensitivity of MRI. We also investigated sensitivity in relation to specificity as determined by ROC curves, showing that the area under the curve was significantly higher for MRI than for mammography; this means that MRI screening could better discriminate between malignant and benign cases.
When we included only invasive breast cancers, the difference between the sensitivity of the MRI and mammography (79.5 percent vs. 33.3 percent) was even greater than the difference overall (71.1 percent vs. 40.0 percent). MRI detected 20 cancers (including 1 ductal carcinoma in situ) that were not found by mammography or clinical breast examination. The stage of these 20 cancers was favorable; 11 of the 19 invasive tumors were smaller than 10 mm, and only 1 was associated with a positive node.
Another important matter that we addressed was the best method for detecting carcinoma in situ. Our study showed that mammography had a higher sensitivity than MRI for detecting ductal carcinoma in situ: 83 percent (five out of six cancers detected), as compared with 17 percent (one out of six) for MRI (P=0.22).
To investigate whether screening improves the chance of diagnosing breast cancer at an early stage, we compared the distribution of tumor stages in our study with the distribution in two external control groups. The first group consisted of age-matched women in a database of all breast cancers diagnosed in 1998 in the Netherlands. A drawback of this group is that we had no information about whether or not they had been screened or the family history. Therefore, we added a second control group from a prospective population-based study of the prevalence of mutations in patients with breast cancer. From this group, we selected all patients with an age and a family history of breast cancer that were similar to the women in our surveillance study. The tumors in our study group were significantly smaller and were less likely to be node-positive than those in the two control groups. Most screening studies (without MRI) in high-risk women have shown a higher incidence of positive nodes (30 to 45 percent) than we found (21 percent).17,18,37 Moreover, Kollias et al.38 found no significant differences in the size or grade of invasive tumors or in lymph-node status between women who had symptoms of cancer and women whose cancers had been found on screening by mammography. So we may conclude that MRI screening did indeed contribute to the early detection of hereditary breast cancer.
However, larger tumors (>2 cm in diameter) were found more often in the women with BRCA1, BRCA2, PTEN, and TP53 mutations than in the other two risk groups in our study, suggesting that more frequent screening is needed for women with these mutations. A drawback of MRI screening is that it has a lower specificity than mammography, and as a result, MRI will generate more findings judged as uncertain, which require short-term follow-up or additional investigations.39 In our study, screening by MRI led to twice as many unneeded additional examinations as did mammography (420 vs. 207) and three times as many unneeded biopsies (24 vs. 7).
In conclusion, our study shows that the screening program we used, especially MRI screening, can detect breast cancer at an early stage in women at risk for breast cancer.
Supported by a grant (OG 98-03) from the Dutch Health Insurance Council.
We are indebted to Petra Bos, Titia van Echten, Irene Groot, Marijke Hogenkamp, Arjan Nieborg, Angelique Schlief, and Manita Verhoeven for data collection; to Leon Aronson for computer assistance; and to Truuske de Bock and Ronald Damhuis for help in selecting the control groups.
* Other investigators in the Magnetic Resonance Imaging Screening (MRISC) study are listed in the Appendix.
Source Information
From the Rotterdam Family Cancer Clinic, Department of Medical Oncology (M.K., C.T.M.B., J.G.M.K.), and the Departments of Radiology (I.M.O.) and Surgery (M.M.A.T.-L.), Erasmus Medical Center–Daniel den Hoed Cancer Center, Rotterdam; the Department of Radiology (C.B.) and the Department of Medical Oncology and Family Cancer Clinic (L.V.A.M.B.), University Medical Center Nijmegen, Nijmegen; the Departments of Radiology (P.E.B., S.H.M.), Pathology (H.P.), and Surgery (E.J.T.R.), Netherlands Cancer Institute, Amsterdam; the Departments of Radiology (H.M.Z.) and Surgery (R.A.E.M.T.), Leiden University Medical Center, Leiden; the Departments of Radiology (R.A.M.) and Surgery (S.M.), Free University Medical Center, Amsterdam; the Departments of Radiology (T.K.) and Clinical Genetics (J.C.O.), University Hospital Groningen, Groningen; and the Department of Public Health, Erasmus Medical Center, Rotterdam (H.J.K.) — all in the Netherlands.
Address reprint requests to Dr. Klijn at Erasmus Medical Center–Daniel den Hoed Cancer Center, Groene Hilledijk 301 3075 EA, Rotterdam, the Netherlands, or at j.g.m.klijn@erasmusmc.nl.
References
Visser O, Coebergh JWW, van Dijck JAAM, et al., eds. Incidence of cancer in the Netherlands 1998. Utrecht, the Netherlands: Netherlands Cancer Registry, 2002.
Ford D, Easton DF, Stratton M, et al. Genetic heterogeneity and penetrance analysis of the BRCA1 and BRCA2 genes in breast cancer families. Am J Hum Genet 1998;62:676-689.
Klijn JGM, Meijers-Heijboer H. Gene screening and prevention of hereditary breast cancer: a clinical view. Eur J Cancer Suppl 2003;1:13-23.
Hartmann LC, Sellers TA, Schaid DJ, et al. Efficacy of bilateral prophylactic mastectomy in BRCA1 and BRCA2 gene mutation carriers. J Natl Cancer Inst 2001;93:1633-1637.
Meijers-Heijboer H, van Geel B, van Putten WLJ, et al. Breast cancer after prophylactic bilateral mastectomy in women with a BRCA1 or BRCA2 mutation. N Engl J Med 2001;345:159-164.
Rebbeck TR, Lynch HT, Neuhausen SL, et al. Prophylactic oophorectomy in carriers of BRCA1 or BRCA2 mutations. N Engl J Med 2002;346:1616-1622.
Kauff ND, Satagopan JM, Robson ME, et al. Risk-reduction salpingo-oophorectomy in women with a BRCA1 or BRCA2 mutation. N Engl J Med 2002;346:1609-1615.
Cuzick J, Powels T, Veronesi U, et al. Overview of the main outcomes in breast-cancer prevention trials. Lancet 2003;361:296-300.
Nystr?m L, Andersson I, Bjurstam N, Frisell J, Nordenskjold B, Rutqvist LE. Long-term effects of mammography screening: updated overview of the Swedish randomised trials. Lancet 2002;359:909-919.
Olsen O, G?tzsche PC. Cochrane review on screening for breast cancer with mammography. Lancet 2001;358:1340-1342.
Otto SJ, Fracheboud J, Looman CWN, et al. Initiating of population-based mammography screening in Dutch municipalities and effect on breast-cancer mortality: a systematic review. Lancet 2003;361:1411-1417.
de Koning HJ, Boer R, Warmerdam PG, Beemsterboer PM, van der Maas PJ. Quantitative interpretation of age-specific mortality reductions from the Swedish breast cancer-screening trials. J Natl Cancer Inst 1995;87:1217-1223.
Breast-cancer screening with mammography in women aged 40-49 years. Int J Cancer 1996;68:693-699.
Elmore JG, Barton MB, Moceri VM, Polk S, Arena PJ, Fletcher SW. Ten-year risk of false positive screening mammograms and clinical breast examinations. N Engl J Med 1998;338:1089-1096.
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.
Mandelson MT, Oestreicher N, Porter PL, et al. Breast density as a predictor of mammographic detection: comparison of interval- and screen-detected cancers. J Natl Cancer Inst 2000;92:1081-1087.
Brekelmans CTM, Seynaeve C, Bartels CCM, et al. Effectiveness of breast cancer surveillance in BRCA1/2 gene mutation carriers and women with high family risk. J Clin Oncol 2001;19:924-930.
Chart PL, Franssen E. Management of women at increased risk for breast cancer: preliminary results from a new program. CMAJ 1997;157:1235-1242.
Macmillan RD. Screening women with a familial history of breast cancer -- results from the British Familial Breast Cancer Group. Eur J Surg Oncol 2000;26:149-152.
Scheuer L, Kauff N, 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.
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.
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.
Huo Z, Giger ML, Olopade OI, et al. Computerized analysis of digitized mammograms of BRCA1 and BRCA2 gene mutation carriers. Radiology 2002;225:519-526.
Adem C, Reynolds C, Soderberg CL, et al. Pathologic characteristics of breast parenchyma in patients with hereditary breast carcinoma, including BRCA1 and BRCA2 mutation carriers. Cancer 2003;97:1-11.
Boetes C, Stoutjesdijk M. MR imaging in screening women at increased risk for breast cancer. Magn Reson Imaging Clin N Am 2001;9:357-372.
Morris EA. Review of breast MRI: indications and limitations. Semin Roentgenol 2001;36:226-237.
Orel SG, Schnall MD. MR imaging of the breast for the detection, diagnosis, and staging of breast cancer. Radiology 2001;220:13-30.
Kriege M, Brekelmans CTM, Boetes C, et al. MRI screening for breast cancer in women with familial or genetic predisposition: design of the Dutch National Study (MRISC). Fam Cancer 2001;1:163-8.
Claus EB, Risch N, Thompson WD. Autosomal dominant inheritance of early-onset breast cancer: implications for risk prediction. Cancer 1994;73:643-651.
Illustrated Breast Imaging Reporting and Data System (BI-RADS). 3rd ed. Reston, Va.: American College of Radiology, 1998.
Liberman L, Menell JH. Breast Imaging Reporting and Data System (BI-RADS). Radiol Clin North Am 2002;40:409-430.
van Asperen CJ, Tollenaar RAEM, Krol-Warmerdam EMM, et al. Possible consequences of applying guidelines to healthy women with a family history of breast cancer. Eur J Hum Genet 2003;11:633-636.
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 2000;215:267-279.
Tilanus-Linthorst MMA, Obdeijn IMM, Bartels KCM, de Koning HJ, Oudkerk M. First experiences in screening women at high risk for breast cancer with MR imaging. Breast Cancer Res Treat 2000;63:53-60.
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.
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:Suppl:115-124.
Lalloo F, Boggis CRM, Evans DGR, Shenton A, Threlfall AG, Howell A. Screening by mammography, women with a family history of breast cancer. Eur J Cancer 1998;34:937-940.
Kollias J, Sibbering DM, Blamey RW, et al. Screening women aged less than 50 years with a familial history of breast cancer. Eur J Cancer 1998;34:878-883.
Liberman L, Morris EA, Benton CL, Abramson AF, Dershaw DD. Probably benign lesions at breast magnetic resonance imaging: preliminary experience in high-risk women. Cancer 2003;98:377-388.(Mieke Kriege, M.Sc., Ceci)
Background The value of regular surveillance for breast cancer in women with a genetic or familial predisposition to breast cancer is currently unproven. We compared the efficacy of magnetic resonance imaging (MRI) with that of mammography for screening in this group of high-risk women.
Methods Women who had a cumulative lifetime risk of breast cancer of 15 percent or more were screened every six months with a clinical breast examination and once a year by mammography and MRI, with independent readings. The characteristics of the cancers that were detected were compared with the characteristics of those in two different age-matched control groups.
Results We screened 1909 eligible women, including 358 carriers of germ-line mutations. Within a median follow-up period of 2.9 years, 51 tumors (44 invasive cancers, 6 ductal carcinomas in situ, and 1 lymphoma) and 1 lobular carcinoma in situ were detected. The sensitivity of clinical breast examination, mammography, and MRI for detecting invasive breast cancer was 17.9 percent, 33.3 percent, and 79.5 percent, respectively, and the specificity was 98.1 percent, 95.0 percent, and 89.8 percent, respectively. The overall discriminating capacity of MRI was significantly better than that of mammography (P<0.05). The proportion of invasive tumors that were 10 mm or less in diameter was significantly greater in our surveillance group (43.2 percent) than in either control group (14.0 percent and 12.5 percent , respectively). The combined incidence of positive axillary nodes and micrometastases in invasive cancers in our study was 21.4 percent, as compared with 52.4 percent (P<0.001) and 56.4 percent (P=0.001) in the two control groups.
Conclusions MRI appears to be more sensitive than mammography in detecting tumors in women with an inherited susceptibility to breast cancer.
The cumulative lifetime risk of breast cancer among Dutch women is approximately 11 percent.1 A family history of breast cancer or the presence of a germ-line mutation of the BRCA1 or BRCA2 gene increases this risk considerably and is often associated with a diagnosis at a young age.2,3 Among high-risk women, the risk of breast cancer can be reduced by prophylactic mastectomy,4,5 prophylactic oophorectomy,6,7 or chemoprevention.8 Early diagnosis as a result of intensive surveillance may also decrease the rate of death from breast cancer.
Randomized trials have shown that mammographic screening of all women who are between 50 and 70 years of age can reduce mortality from breast cancer by about 25 percent.9 Although these findings were recently disputed,10 there is a consensus among clinicians that breast-cancer screening of women in this age group is effective. Screening is one of the main factors contributing to the decrease in mortality associated with breast cancer in the Netherlands.11 However, there is no consensus about the value of breast-cancer screening among women who are 40 to 49 years old.12,13,14 One of the reasons for the lack of agreement is the difficulty in detecting tumors by mammographic screening in younger women, who have denser breasts than postmenopausal women.15,16 Although screening is frequently offered to women with a genetic predisposition to breast cancer who are under the age of 50 years, the efficacy of this approach is unproven. Preliminary results of surveillance by mammography and clinical breast examination in such women showed that mammographic screening has a low sensitivity for detecting tumors, especially in carriers of a BRCA mutation.17,18,19,20,21 Possible reasons, apart from the high rate of growth of tumors in women with such mutations, include the atypical changes seen on screening mammograms and specific histopathological characteristics in carriers of BRCA mutations, as compared with noncarriers of the same age.22,23,24
In a diagnostic setting, magnetic resonance imaging (MRI) is a sensitive method of breast imaging, and it is virtually uninfluenced by breast density, but the specificity is variable and the costs are high.25,26,27 Because MRI may improve the sensitivity of screening in women with a familial or genetic predisposition to breast cancer, we prospectively compared MRI with mammography for screening women with such a predisposition in order to determine whether screening with MRI facilitated the early diagnosis of hereditary breast cancer.
Methods
Study Population
The design of our MRI screening study, in which six subcommittees in different disciplines were involved, has been described previously.28 Between November 1, 1999, and October 1, 2003, 1952 women with a genetic risk of breast cancer were recruited for the study by six familial-cancer clinics in the Netherlands. The six centers were Erasmus Medical Center–Daniel den Hoed Cancer Center, Rotterdam; the Netherlands Cancer Institute, Amsterdam; University Medical Center Nijmegen, Nijmegen; Leiden University Medical Center, Leiden; University Hospital Groningen, Groningen; and Free University Medical Center, Amsterdam. The study was approved by the ethics committees of all the centers. All the women who participated gave written informed consent.
The inclusion criteria for participation were a cumulative lifetime risk of breast cancer of 15 percent or more owing to a familial or genetic predisposition, according to the modified tables of Claus et al.,29 and an age of 25 to 70 years. Women could be tested at an age younger than 25 if they had a family history of breast cancer diagnosed before the age of 30 years, since testing began at an age 5 years younger than that at which the youngest family member was found to have breast cancer. Women with symptoms that were suggestive of breast cancer or women who had a personal history of breast cancer were excluded.
Surveillance
Surveillance consisted of a clinical breast examination performed by an experienced physician every six months and imaging studies performed annually by experienced radiologists. The imaging included a mammographic study (oblique and craniocaudal views and, if necessary, compression views or magnifications) and a dynamic breast MRI with gadolinium-containing contrast medium according to a standard protocol.25 Whenever possible, both imaging investigations were performed on the same day or in the same time period, between day 5 and day 15 of the menstrual cycle. The results of mammography and MRI were scored in a standardized way, according to the Breast Imaging Reporting and Data System (BI-RADS) classification,30,31 and the results were blinded so that the two examinations were not linked. When one of the examinations was scored as either BI-RADS category 3 ("probably benign finding") or category 0 ("need additional imaging evaluation"), further investigation by ultrasonography with or without fine-needle aspiration was advised, or mammography or MRI was repeated. When one of the two examinations was scored as BI-RADS category 4 ("suspicious abnormality") or category 5 ("highly suggestive of malignancy"), a cytologic or histologic evaluation of a biopsy specimen was performed. When the results of mammography and MRI were negative but the findings on clinical breast examination were rated as uncertain or suspicious, additional investigation was also performed. The diagnosis of malignant tumors was based on the results of a histologic examination. One of the investigators, an expert pathologist, reviewed all the biopsy specimens that formed the basis for the diagnosis of breast cancer.
Statistical Analysis
The women were divided into three categories according to the cumulative lifetime risk of breast cancer, as follows: carriers of the BRCA1 or BRCA2 or other mutations (cumulative lifetime risk, 50 to 85 percent), a high-risk group (risk, 30 to 49 percent), and a moderate-risk group (risk, 15 to 29 percent).28,29 The characteristics of the women in each risk group were compared by analysis of variance or Pearson's chi-square test.
The rates of detection of breast cancer for the group as a whole and for each of the three risk groups were calculated, and a Poisson distribution was assumed in order to calculate the 95 percent confidence intervals. Person-years at risk were calculated from the date of the first examination, irrespective of the type of examination, to the date of detection of breast cancer, bilateral prophylactic mastectomy, or death; the date that a patient stopped surveillance; or the cutoff date for this analysis (October 1, 2003). An "interval cancer" was defined as a carcinoma detected between two rounds of screening after initially negative findings on screening. In our analysis, we defined as positive a mammographic or MRI study with a BI-RADS score of 0, 3, 4, or 5 and a clinical breast examination that was classified as "uncertain" or "suspicious," because those were the results that triggered an additional examination.
To compare the three different screening methods, we calculated the sensitivity, specificity, and positive predictive value of each. The sensitivity used is that of one screening method relative to the others, meaning that a test result is a false negative when a proven cancer (diagnosed on the basis of a histologic examination) is detected in the interval or by one of the other methods. Receiver-operating-characteristic (ROC) curves for the two imaging methods were generated. The area under the curve was used as an index in evaluating the inherent capacity of a screening method to discriminate between "positive" and "negative" cases. We used a z-test to compare the area under the curve for the results of mammography and MRI. For the analysis of the screening variables, we used only the screening data that included the results of both mammography and MRI.
To determine whether breast cancer was diagnosed by screening at a stage more favorable to treatment, the characteristics of breast tumors detected in the study group were compared with those in two control groups. The first control group was derived from all women who had breast cancers diagnosed in 1998 in the Netherlands. These data were obtained from the National Cancer Registry. The second control group consisted of unselected patients who had received a diagnosis of primary breast cancer in Leiden or Rotterdam between 1996 and 2002 and who were participating in a prospective study of the prevalence of gene mutations.32 Subjects in both control groups were matched for age with the patients in the study group (in five-year categories). From this series of consecutive patients in the second control group, we chose all the unscreened patients who were between 25 and 60 years old and whose cumulative lifetime risk of breast cancer was more than 15 percent because of a family history of the disease — information that was routinely recorded in this database. The differences in tumor characteristics between the study group and the control groups were tested with the use of Pearson's chi-square test or the chi-square test for trend. A two-sided P value of less than 0.05 was considered to indicate statistical significance. All statistical analyses were performed with the use of SPSS software (version 9.0).
Results
Study Population
Of the women who were invited to participate in the study, 90 percent agreed. Initially, 1952 women were included; 8 withdrew from the study before their first screening visit and another 35 were excluded because they ultimately proved not to be carriers in a family with a proven mutation and therefore had less than a 15 percent cumulative lifetime risk of breast cancer. Of the 1909 remaining women, 88 (4.6 percent) left the study or were lost to surveillance before October 1, 2003; 65 of these 88 women underwent prophylactic mastectomy. Another 89 women (4.7 percent) remained under surveillance but later refused screening by MRI, because of claustrophobia or for other reasons.
Table 1 lists the characteristics of the 1909 women according to risk category. The mean age at entry was 40 years (range, 19 to 72). Within the group of 358 carriers of pathogenic mutations, 276 had a BRCA1 mutation, 77 had a BRCA2 mutation, 1 had both a BRCA1 and a BRCA2 mutation, 2 had a PTEN mutation, and 2 had a TP53 mutation.
Table 1. Characteristics of Participating Women at Study Entry, According to Risk Group.
Breast Cancers
From November 1, 1999, to October 1, 2003, 51 malignant tumors (44 invasive breast cancers, 6 ductal carcinomas in situ, and 1 non-Hodgkin's lymphoma) were detected (Figure 1), during a median follow-up period of 2.9 years (mean 2.7, range, 0.1 to 3.9 years); 1 lobular carcinoma in situ was also found. Table 2 shows the detection rate for the whole group and separately for the different risk groups. The overall rate of detection for all breast cancers (invasive plus in situ) was 9.5 per 1000 woman-years at risk (95 percent confidence interval, 7.1 to 12.3), with the highest rate (26.5 per 1000) in the group of women who were carriers of the BRCA1, BRCA2, PTEN, and TP53 mutations.
Figure 1. Women at Increased Risk for Breast Cancer Enrolled and Tumors Detected.
A total of 1952 women were enrolled, of whom 1909 were eligible for this analysis.
Table 2. Detection of Cases of Breast Cancer (Including Ductal Carcinoma in Situ) According to Risk Group.
Performance of the Screening Methods
Table 3 shows the results with the three screening methods. Of the 50 breast cancers that were detected, 5 were excluded from the analysis (Table 3). The 45 cancers that were evaluated in the comparison of the methods included 4 interval cancers (i.e., cancers detected between two episodes of screening). The first was symptomatic (30 mm in diameter, node-negative), detected seven months after screening by imaging and clinical breast examination and one month after screening by clinical breast examination only. The second (4 mm, node-negative) was detected in a specimen from a prophylactic mastectomy. The third was symptomatic (45 mm, node-negative) and was detected seven months after screening by imaging; the fourth, also symptomatic (13 mm, with isolated tumor cells in a lymph node), was detected three months after screening by imaging.
Table 3. Sensitivity, Specificity, and Positive Predictive Value (PPV) of the Three Screening Methods.
Overall, 32 breast cancers were found by MRI (22 of these were not visible on mammography), whereas 13 were missed by MRI (8 of the 13 were visible on mammography, including 5 ductal carcinomas in situ; 4 were interval cancers; and 1 tumor was detected only by clinical breast examination). In this group of 45 breast cancers, mammographic screening detected 18 tumors (10 of these were visible by MRI) and missed 27 tumors (including the 22 that were visible on MRI, the 4 interval cancers, and the 1 that was detected only by clinical breast examination).
With respect to all breast cancers (invasive and ductal carcinoma in situ), the sensitivity of clinical breast examination, mammography, and MRI was 17.8 percent, 40.0 percent, and 71.1 percent, respectively, when the BI-RADS score was 3 or higher (Table 3). For invasive cancers only, the respective percentages were 17.9 percent, 33.3 percent, and 79.5 percent. The specificity was 98.1 percent for clinical breast examination, 95.0 percent for mammography, and 89.8 percent for MRI.
Of the 41 cancers found by screening, 22 were detected at the first imaging screening in the study; of the women in whom cancer was detected, 16 had undergone mammographic screening before the start of the study. Two of the interval cancers were detected after the first imaging screening, and two others after a subsequent imaging screening. The sensitivity of mammography was 37.5 percent for the first screening and 42.9 percent for subsequent screening (P=0.71). The sensitivity of MRI was 79.2 percent for the first screening and 61.9 percent for subsequent screening (P=0.20).
Among the 83 clinical breast examinations with findings that were judged as probably benign or suspicious, or highly suggestive of cancer, 8 cases of malignant disease were confirmed, for a positive predictive value of 9.6 percent (Table 3). Among the 225 mammograms with findings categorized as BI-RADS 3 or higher, 18 cases of malignant disease were confirmed, for a positive predictive value of 8.0 percent. A total of 32 cancers were confirmed among 452 MRI screenings with such findings, for a positive predictive value of 7.1 percent (Table 3). With a cutoff level of BI-RADS 4, the sensitivity for both imaging methods decreased, whereas the specificity increased.
To evaluate the discriminating capacity of the imaging methods, we generated ROC curves (Figure 2). The area under the curve was 0.686 for mammography and 0.827 for MRI; the difference between the areas was 0.141 (95 percent confidence interval, 0.020 to 0.262; P<0.05).
Figure 2. Receiver-Operating-Characteristic Curves for Mammography and MRI.
The difference between the area under the curve (AUC) for mammography and the AUC for MRI was 0.141 (95 percent confidence interval, 0.020 to 0.262; P<0.05).
Additional Investigations
Ultrasonography was performed 889 times in 627 different women according to the protocol. Fine-needle aspiration was carried out 312 times: 267 times in combination with ultrasonography and 45 times with palpation. Biopsy was performed 85 times in 82 women and showed malignant disease in 50 cases and 1 lobular carcinoma in situ, making the rate of positive histologic findings 60.0 percent. Sixty-seven of these 85 biopsies were performed after a screening visit at which both MRI and mammography were performed. Of the 25 biopsies in women who had mammographic findings with a score of 3 or higher, 7 (28.0 percent) showed no cancer. Of the 56 biopsies in women who had MRI findings with a score of 3 or higher, 24 (42.9 percent) showed no cancer (Table 3). One of the 51 tumors was found in a specimen from a prophylactic mastectomy.
Tumor Characteristics
Table 4 compares the characteristics of tumors found in the study group with those of tumors in the two age-matched control groups. In the study group, 19 of the 44 women with an invasive breast cancer (43.2 percent) had a small tumor (10 mm in diameter) — a proportion that was significantly higher than that in the first control group (14.0 percent, P<0.001) or the second control group (12.5 percent, P=0.04). Six of 42 invasive tumors (14.3 percent) with known axillary status in the study group were node-positive and 3 (7.1 percent) had micrometastases (combined total, 21.4 percent). This rate was significantly lower than those in both control groups, in which the rates of node-positive cancer were 52.4 percent (P<0.001) and 56.4 percent (P=0.001), respectively. There were no major differences between the study and control groups with respect to histologic features, with the exception of a relatively high incidence of the medullary type in the study group (11.3 percent, vs. 1.8 percent in the first control group). In the study group, a high proportion of grade 1 tumors were in women at high risk (68.8 percent) or moderate risk (75.0 percent); however, the group of women with BRCA1, BRCA2, or other mutations had a high percentage of grade 3 tumors (63.2 percent), in addition to a high percentage of tumors that were negative for steroid receptors (Table 4).
Table 4. Characteristics of Women with Breast Cancer and Breast Cancers Detected in the Three Risk Groups and in the Two Control Groups.
Disease-free and Overall Survival
In the study group, none of the 50 patients with breast cancer (44 with invasive cancer and 6 with ductal carcinoma in situ) died before the end of the study period; the total follow-up after diagnosis was 87.6 woman-years for these 50 patients (median, 1.5 years). Contralateral breast cancer occurred in one patient. The patient with non-Hodgkin's lymphoma died.
Discussion
In this prospective study, we compared the efficacy of mammographic and MRI screening for breast cancer in women with a family history of the disease or a genetic predisposition to breast cancer. Among the women examined by both methods at the same screening visit, we detected 45 breast cancers (including 6 ductal carcinomas in situ): 32 by MRI (sensitivity, 71.1 percent) and 18 by mammography (40.0 percent); five other patients were excluded from this comparison for various reasons (Table 3). Thus, the sensitivity of MRI was higher than that of mammography, but both the specificity and positive predictive value of MRI were lower.
In our sensitivity and specificity calculations, we defined lesions that were in BI-RADS category 3 and higher as positive, but most other authors have included in their calculations only lesions in BI-RADS categories 4 and 5 as positive.21,33,34 If we had followed that policy, the sensitivity would have been 24.4 percent for mammography and 46.6 percent for MRI, in accord with the higher sensitivity previously reported for MRI.21,33,35,36 However, the previous studies enrolled small groups of women, included some retrospective data,35 evaluated heterogeneous groups that included women with previous breast cancers,21,33,36 or had a plan for follow-up after a suspicious finding on MRI that differed from the follow-up plan for a suspicious mammographic finding.33 All these factors might have artificially increased the sensitivity of MRI. We also investigated sensitivity in relation to specificity as determined by ROC curves, showing that the area under the curve was significantly higher for MRI than for mammography; this means that MRI screening could better discriminate between malignant and benign cases.
When we included only invasive breast cancers, the difference between the sensitivity of the MRI and mammography (79.5 percent vs. 33.3 percent) was even greater than the difference overall (71.1 percent vs. 40.0 percent). MRI detected 20 cancers (including 1 ductal carcinoma in situ) that were not found by mammography or clinical breast examination. The stage of these 20 cancers was favorable; 11 of the 19 invasive tumors were smaller than 10 mm, and only 1 was associated with a positive node.
Another important matter that we addressed was the best method for detecting carcinoma in situ. Our study showed that mammography had a higher sensitivity than MRI for detecting ductal carcinoma in situ: 83 percent (five out of six cancers detected), as compared with 17 percent (one out of six) for MRI (P=0.22).
To investigate whether screening improves the chance of diagnosing breast cancer at an early stage, we compared the distribution of tumor stages in our study with the distribution in two external control groups. The first group consisted of age-matched women in a database of all breast cancers diagnosed in 1998 in the Netherlands. A drawback of this group is that we had no information about whether or not they had been screened or the family history. Therefore, we added a second control group from a prospective population-based study of the prevalence of mutations in patients with breast cancer. From this group, we selected all patients with an age and a family history of breast cancer that were similar to the women in our surveillance study. The tumors in our study group were significantly smaller and were less likely to be node-positive than those in the two control groups. Most screening studies (without MRI) in high-risk women have shown a higher incidence of positive nodes (30 to 45 percent) than we found (21 percent).17,18,37 Moreover, Kollias et al.38 found no significant differences in the size or grade of invasive tumors or in lymph-node status between women who had symptoms of cancer and women whose cancers had been found on screening by mammography. So we may conclude that MRI screening did indeed contribute to the early detection of hereditary breast cancer.
However, larger tumors (>2 cm in diameter) were found more often in the women with BRCA1, BRCA2, PTEN, and TP53 mutations than in the other two risk groups in our study, suggesting that more frequent screening is needed for women with these mutations. A drawback of MRI screening is that it has a lower specificity than mammography, and as a result, MRI will generate more findings judged as uncertain, which require short-term follow-up or additional investigations.39 In our study, screening by MRI led to twice as many unneeded additional examinations as did mammography (420 vs. 207) and three times as many unneeded biopsies (24 vs. 7).
In conclusion, our study shows that the screening program we used, especially MRI screening, can detect breast cancer at an early stage in women at risk for breast cancer.
Supported by a grant (OG 98-03) from the Dutch Health Insurance Council.
We are indebted to Petra Bos, Titia van Echten, Irene Groot, Marijke Hogenkamp, Arjan Nieborg, Angelique Schlief, and Manita Verhoeven for data collection; to Leon Aronson for computer assistance; and to Truuske de Bock and Ronald Damhuis for help in selecting the control groups.
* Other investigators in the Magnetic Resonance Imaging Screening (MRISC) study are listed in the Appendix.
Source Information
From the Rotterdam Family Cancer Clinic, Department of Medical Oncology (M.K., C.T.M.B., J.G.M.K.), and the Departments of Radiology (I.M.O.) and Surgery (M.M.A.T.-L.), Erasmus Medical Center–Daniel den Hoed Cancer Center, Rotterdam; the Department of Radiology (C.B.) and the Department of Medical Oncology and Family Cancer Clinic (L.V.A.M.B.), University Medical Center Nijmegen, Nijmegen; the Departments of Radiology (P.E.B., S.H.M.), Pathology (H.P.), and Surgery (E.J.T.R.), Netherlands Cancer Institute, Amsterdam; the Departments of Radiology (H.M.Z.) and Surgery (R.A.E.M.T.), Leiden University Medical Center, Leiden; the Departments of Radiology (R.A.M.) and Surgery (S.M.), Free University Medical Center, Amsterdam; the Departments of Radiology (T.K.) and Clinical Genetics (J.C.O.), University Hospital Groningen, Groningen; and the Department of Public Health, Erasmus Medical Center, Rotterdam (H.J.K.) — all in the Netherlands.
Address reprint requests to Dr. Klijn at Erasmus Medical Center–Daniel den Hoed Cancer Center, Groene Hilledijk 301 3075 EA, Rotterdam, the Netherlands, or at j.g.m.klijn@erasmusmc.nl.
References
Visser O, Coebergh JWW, van Dijck JAAM, et al., eds. Incidence of cancer in the Netherlands 1998. Utrecht, the Netherlands: Netherlands Cancer Registry, 2002.
Ford D, Easton DF, Stratton M, et al. Genetic heterogeneity and penetrance analysis of the BRCA1 and BRCA2 genes in breast cancer families. Am J Hum Genet 1998;62:676-689.
Klijn JGM, Meijers-Heijboer H. Gene screening and prevention of hereditary breast cancer: a clinical view. Eur J Cancer Suppl 2003;1:13-23.
Hartmann LC, Sellers TA, Schaid DJ, et al. Efficacy of bilateral prophylactic mastectomy in BRCA1 and BRCA2 gene mutation carriers. J Natl Cancer Inst 2001;93:1633-1637.
Meijers-Heijboer H, van Geel B, van Putten WLJ, et al. Breast cancer after prophylactic bilateral mastectomy in women with a BRCA1 or BRCA2 mutation. N Engl J Med 2001;345:159-164.
Rebbeck TR, Lynch HT, Neuhausen SL, et al. Prophylactic oophorectomy in carriers of BRCA1 or BRCA2 mutations. N Engl J Med 2002;346:1616-1622.
Kauff ND, Satagopan JM, Robson ME, et al. Risk-reduction salpingo-oophorectomy in women with a BRCA1 or BRCA2 mutation. N Engl J Med 2002;346:1609-1615.
Cuzick J, Powels T, Veronesi U, et al. Overview of the main outcomes in breast-cancer prevention trials. Lancet 2003;361:296-300.
Nystr?m L, Andersson I, Bjurstam N, Frisell J, Nordenskjold B, Rutqvist LE. Long-term effects of mammography screening: updated overview of the Swedish randomised trials. Lancet 2002;359:909-919.
Olsen O, G?tzsche PC. Cochrane review on screening for breast cancer with mammography. Lancet 2001;358:1340-1342.
Otto SJ, Fracheboud J, Looman CWN, et al. Initiating of population-based mammography screening in Dutch municipalities and effect on breast-cancer mortality: a systematic review. Lancet 2003;361:1411-1417.
de Koning HJ, Boer R, Warmerdam PG, Beemsterboer PM, van der Maas PJ. Quantitative interpretation of age-specific mortality reductions from the Swedish breast cancer-screening trials. J Natl Cancer Inst 1995;87:1217-1223.
Breast-cancer screening with mammography in women aged 40-49 years. Int J Cancer 1996;68:693-699.
Elmore JG, Barton MB, Moceri VM, Polk S, Arena PJ, Fletcher SW. Ten-year risk of false positive screening mammograms and clinical breast examinations. N Engl J Med 1998;338:1089-1096.
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.
Mandelson MT, Oestreicher N, Porter PL, et al. Breast density as a predictor of mammographic detection: comparison of interval- and screen-detected cancers. J Natl Cancer Inst 2000;92:1081-1087.
Brekelmans CTM, Seynaeve C, Bartels CCM, et al. Effectiveness of breast cancer surveillance in BRCA1/2 gene mutation carriers and women with high family risk. J Clin Oncol 2001;19:924-930.
Chart PL, Franssen E. Management of women at increased risk for breast cancer: preliminary results from a new program. CMAJ 1997;157:1235-1242.
Macmillan RD. Screening women with a familial history of breast cancer -- results from the British Familial Breast Cancer Group. Eur J Surg Oncol 2000;26:149-152.
Scheuer L, Kauff N, 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.
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.
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.
Huo Z, Giger ML, Olopade OI, et al. Computerized analysis of digitized mammograms of BRCA1 and BRCA2 gene mutation carriers. Radiology 2002;225:519-526.
Adem C, Reynolds C, Soderberg CL, et al. Pathologic characteristics of breast parenchyma in patients with hereditary breast carcinoma, including BRCA1 and BRCA2 mutation carriers. Cancer 2003;97:1-11.
Boetes C, Stoutjesdijk M. MR imaging in screening women at increased risk for breast cancer. Magn Reson Imaging Clin N Am 2001;9:357-372.
Morris EA. Review of breast MRI: indications and limitations. Semin Roentgenol 2001;36:226-237.
Orel SG, Schnall MD. MR imaging of the breast for the detection, diagnosis, and staging of breast cancer. Radiology 2001;220:13-30.
Kriege M, Brekelmans CTM, Boetes C, et al. MRI screening for breast cancer in women with familial or genetic predisposition: design of the Dutch National Study (MRISC). Fam Cancer 2001;1:163-8.
Claus EB, Risch N, Thompson WD. Autosomal dominant inheritance of early-onset breast cancer: implications for risk prediction. Cancer 1994;73:643-651.
Illustrated Breast Imaging Reporting and Data System (BI-RADS). 3rd ed. Reston, Va.: American College of Radiology, 1998.
Liberman L, Menell JH. Breast Imaging Reporting and Data System (BI-RADS). Radiol Clin North Am 2002;40:409-430.
van Asperen CJ, Tollenaar RAEM, Krol-Warmerdam EMM, et al. Possible consequences of applying guidelines to healthy women with a family history of breast cancer. Eur J Hum Genet 2003;11:633-636.
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 2000;215:267-279.
Tilanus-Linthorst MMA, Obdeijn IMM, Bartels KCM, de Koning HJ, Oudkerk M. First experiences in screening women at high risk for breast cancer with MR imaging. Breast Cancer Res Treat 2000;63:53-60.
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.
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:Suppl:115-124.
Lalloo F, Boggis CRM, Evans DGR, Shenton A, Threlfall AG, Howell A. Screening by mammography, women with a family history of breast cancer. Eur J Cancer 1998;34:937-940.
Kollias J, Sibbering DM, Blamey RW, et al. Screening women aged less than 50 years with a familial history of breast cancer. Eur J Cancer 1998;34:878-883.
Liberman L, Morris EA, Benton CL, Abramson AF, Dershaw DD. Probably benign lesions at breast magnetic resonance imaging: preliminary experience in high-risk women. Cancer 2003;98:377-388.(Mieke Kriege, M.Sc., Ceci)