Breast cancer screening in women at elevated risk
Phí, Xuân Anh
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date: 2018
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
Phí, X. A. (2018). Breast cancer screening in women at elevated risk: Comparative evaluation of screening modalities to inform practice. University of Groningen.
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.
Accuracy of screening women at familial risk
of breast cancer without a known gene mutation
- An Individual patient data meta-analysis
Xuan-Anh Phi Nehmat Houssami Maartje J. Hooning Christopher C. Riedl Martin O. Leach Francesco Sardanelli Isabelle Trop Sepideh Saadatmand Madeleine M.A Tilanus-Linthorst Thomas H. Helbich Edwin R. Van den Heuvel Harry J. de Koning Inge-Marie Obdeijn Geertruida H. De Bock European Journal of Cancer 2017 Nov;85:31-38ABSTRACT
Introdu on: Women with a strong family history of breast cancer (BC) and without a known gene muta on have an increased risk of developing BC. We aimed to inves gate the accuracy of screening using annual mammography with or without magne c resonance imaging (MRI) for these women outside the general popula on screening program.
1
Methods: An individual pa ent data (IPD) meta-analysis was conducted using IPD from six prospec ve screening trials that had included women at increased risk for BC: only women with a strong familial risk for BC and without a known gene muta on were included in this analysis. A generalised linear mixed model was applied to es mate and compare screening accuracy (sensi vity, specificity and predic ve values) for annual mammography with or without MRI.
1
Results: There were 2,226 women (median age: 41 years, interquar le range 35.3-47) with 7,478 woman-years of follow-up, with a BC rate of 11.6 (95% confidence interval 9.3-14.4) in 1,000 woman-years. Mammography screening had a sensi vity of 55.1% (standard error of mean [SE] 7.0) and a specificity of 93.7% (SE 1.3). Screening with MRI alone had a sensi vity of 89.3% (SE 4.6) and a specificity of 82.9% (SE 2.8). Adding MRI to mammography increased sensi vity to 98.4% (SE 1.8, P<0.01 compared to mammography alone) but lowered specificity to 78.9% (SE 2.7, P<0.01 compared with mammography alone).
1
Conclusion: In this popula on of women with strong familial BC risk but without a known gene muta on, in whom BC incidence was high both before and a er age 50, adding MRI to mammography substan ally increased screening sensi vity but also decreased its specificity.
1 Highlights
In women at familial risk and no known muta ons, we es mate a breast cancer (BC) rate of 12 per 1000 woman-years.
1
BC rate increased from 5.1 per 1000 in women <40 years to 21 per 1000 in women >50 years.
1
Mammography sensi vity ranged from 51% in women <40 to 67% in women ≥50 years. 1
Adjunct magne c resonance imaging increased sensi vity by 43% but decreased specificity by 15%.
Introdu on
1
About 15-20% of breast cancer (BC) cases are associated with a family history of BC (1) . Women without a known muta on in a hereditary BC gene, but with a family history of breast with/without ovarian cancer, are at a higher risk of developing BC, the extent of the increased risk depends on the number of affected rela ves and the age at cancer diagnosis in the rela ve(s) (2,3) . These women at familial risk, who have a cumula ve life me risk of developing BC over 15% to 20%, are usually offered a BC screening regimen outside of the general popula on screening program, star ng at an earlier age and including more frequent (annual) mammography (4,5) .
Results of many prospec ve trials evalua ng the accuracy of adding annual MRI to mammography for screening these women have been published (6-15). Although these studies emphasised the significantly greater sensi vity of annual magne c resonance imaging (MRI) and mammography in combina on for screening this high-risk popula on, several issues remain unclear. First, inclusion criteria were heterogeneous and all the studies also included women with known gene muta ons. Furthermore, the defini on of familial risk for BC varied across countries and centres depending on referral criteria and risk assessment tools. Also, few studies reported results separately for women at familial risk without a known gene muta on (8)(11,12) and none of the studies reported results stra fied by age for this popula on.
In this meta-analysis, pooling individual pa ent data (IPD) from prospec ve trials, we aimed to assess the accuracy of screening women at familial risk of BC without a known gene muta on, adding MRI to mammography and stra fying outcomes by age.
Methods 1
Study design
1
An IPD meta-analysis was conducted including individual data from 6 of 12 prospec ve trials in which women at high risk of BC due to an inherited BRCA gene muta on or a strong family history of BC were screened with annual mammography and MRI, and the accuracy of each screening modality was reported separately (16,17). All studies were performed in developed countries. More details about the study inclusion criteria, data acquisi on and assembly and quality assessment were reported in our previous publica on which focused on BRCA1/2 gene muta on carriers (17). In the present study, we focus only on women with a strong family history of BC (defined as a cumula ve life me BC risk of at least 15%) and without a
known gene muta on. Specific inclusion criteria for the original studies contribu ng to this IPD meta-analysis, outlining family history criteria and whether women with a personal history of BC were included are summarised in Supplementary appendix 1.
Study popula on
1
Women aged 25 or older who had a strong family history of BC and no known gene muta on and had completed at least one screening round, were included in this analysis. A completed screening round was defined as a screening round in which both MRI and mammography were performed within a me interval of less than 3 months, with results of the two tests interpreted separately using blinded methods. Screens were included if there was either a pathology test or at least 1 year follow-up to confirm the presence or absence of BC. Women who were proven to be non-muta on carriers from a BRCA family were excluded, as their risk is generally considered to be comparable to that of the general popula on. Screen-detected or interval cancers were counted in this analysis (BCs found during preven ve mastectomy were not considered).
Primary outcome and defini ons
1
Primary outcomes were screening accuracy including sensi vity, specificity and posi ve/nega ve predic ve value (PPV/NPV). To adjust for mul ple screenings of the same women and differences between studies, the es mates for each modality were model-based with the following: (1) sensi vity defined as the number of BCs detected over the total number of BC diagnosed; (2) specificity defined as the number of true-nega ve tests over the total number of screens without BC; (3) PPV defined as the number of true-posi ve over the total number of posi ve tests; (4) NPV as the number of true-nega ves over the total number of nega ve tests.
Imaging scores of BI-RADS 0, 3, 4 or 5 (Breast imaging-repor ng and data system) were considered to be a posi ve screening result. Using this threshold allowed harmonisa on of the outcomes across studies. The combina on of MRI and mammography was classified as a posi ve result if either one of these tests was posi ve. BI-RADS 1 or 2 was considered a nega ve test, and a nega ve outcome of the combina on was based on both tests having nega ve results. For posi ve test results, the presence of BC was based on the results of histologic examina on. The absence of BC was ascertained by histologic examina on or 1 year follow-up with nega ve screening or stable imaging. Where more than one tumour was
diagnosed in a woman in the same screening round, the largest BC was included. Where more than one BC was diagnosed in a woman at different screening years, the first BC was included. For analy c purposes, a BC was considered an interval cancer when it was not detected by a posi ve screening test (mammography or MRI) and was diagnosed between two annual rounds of screening. The above defini ons were applied to all the studies when assembling the individual data to obtain consistency and comparable data from all studies, although there were slight differences between these defini ons and the defini ons that may have been applied in the original studies.
Sta s cal analysis
1
Characteris cs of the women (follow-up me, cancer incidence, median age at entry with interquar le range [IQR]) and cancer characteris cs were reported for the total study popula on and for each age-group. Overall cancer incidence in total and stra fying by screening round, were calculated as the number of BCs per 10,000 woman-years and the 95% confidence interval (CI) was computed assuming the incidence follows a Poisson distribu on.
A generalised linear mixed model (Procedure Glimmix using the QUAD op on for the likelihood method of es ma on the binomial distribu on with logit link func on, SAS version 9.4) was applied to es mate sensi vity, specificity and the predic ve values of each screening modality and then compare these measures for the two screening modali es and the combina on using Wald tests. One analysis is done for sensi vity and specificity simultaneously and another for predic ve values. For sensi vity and specificity, repeated measurements were summarised for each woman to the total number of screens with proven BC, total number of screens without BC, the number of posi ves and the number of true-nega ves. Each woman had six records: two ascertained outcomes (with or without proven BC) for each of three screening modali es (mammography, MRI and the combina on).In the model, the numbers of true-posi ve/nega ve tests followed a binomial distribu on with the total number of screens with/ without proven BCs and a propor on that was modelled as a func on of screening modality. To address heterogeneity between studies, a bivariate random variable with an unstructured correla on matrix was added to model the study effect for each screening modality and each ascertained outcome. Analysis was performed for each age-group separately. Sensi vity and specificity were modelled simultaneously to take into account their nega ve correla on. The same approach was used for posi ve and NPVs by replacing the ascertained outcomes by modality outcomes (test posi ve or nega ve) and
posi ve or nega ve) and number of screens with/without BCs by the number of screens with posi ve/nega ve test. Screening accuracy for the three modali es was compared within age-groups defined by age at screening, as follows: younger than 40 years, 40-49 years and 50 years and older. In two sensi vity analyses, the year of screening was added to the model to explore its impact on the results, and screening accuracy was es mated for first and subsequent rounds to allow for prevalent cases.
Results
Study popula on and breast cancers during the study
1
Data on 2,226 women at familial risk with at least one completed screening round were included in this analysis, represen ng 7,478 woman-years of follow-up (median 3 years, IQR 2-5) with ascertained outcomes (Table 1). There were 106 (4.8%) women with a personal history of BC and 193 (8.7%) women with a nega ve gene c test result. Amongst these 2,226 women with a median age of 41 years (IQR 35.3-47) at study entry, 87 BCs were diagnosed at a median age of 48.4 (IQR 42.6-54) years. BC rate was es mated as 11.6 (95%CI 9.3-14.4) per 1,000 woman-years. BC incidence increased with increasing age: 5.1 (95%CI 2.8-8.6) per 1,000 woman-years in women who underwent screening before age 40, 11.9 (95%CI 8.3-16.4) per 1,000 woman-years in women aged 40-49, and 21.6 (95%CI 15.2-29.8) per 1,000 woman-years in women aged 50 and older. Of all BCs, 37 were prevalent cancers (detected at the first screening round); excluding those prevalent cancers the rate (per 1,000 woman-years) was 9.5 (95%CI 7.1-12.6) in all women, 5.7 (95%CI 2.7-10.5) in women aged <40, 8.3 (95%CI 4.9-13) in women aged 40-50 and 16.6 (95%CI 10.4 – 25.1) in women aged >50. Amongst 87 BCs observed in the study popula on, three were interval cancers: all were invasive, smaller than 2cm, and diagnosed in women aged 37, 46 and 57 years. The youngest group had more invasive BCs and more large invasive BCs than the older age group (Table 1).
Table 1: Overview of the women at familial risk and their breast cancers characteris cs, stra fying by age at screening (N=2,226; BCs=87)
Table 2: Screening accuracy in women at familial risk of breast cancer (N=2,226, BCs=87, median screening rounds: 3 [IQR 2-5])
Screening accuracy in all women
In the total study popula on, mammography sensi vity was 55.1% (standard error of mean [SE] 7.0) (Table 2). The combina on of MRI and mammography had the highest sensi vity but the lowest specificity. The combined sensi vity was 98.4% (SE 1.8) versus 89.3% (SE 4.6) for MRI alone (P<0.001) and 55.1% (SE 7.0) for mammography alone (P<0.001). The combina on had the lowest specificity of 78.9% (SE 2.7) versus 82.9% (SE 2.8) for MRI alone (P<0.01) and 93.7% (SE 1.3) for mammography alone (P<0.01). The PPVs of the three screening modali es were generally comparable around 9%, whereas differences were observed for NPVs, as summarised in Table 2. When adjus ng for year at screening (results not shown) or excluding the first screening round (Table 4), the es mates for sensi vity and specificity did not change. PPV: posi ve predic ve value; NPV: nega ve predic ve value; SE: Standard error of mean
*Compared to the combina on sensi vity: P=0.0008; **Compared to the combina on sensi vity: P<0.0001; µ Compared to the combina on specificity: P=0.002; ¥ Compare to the combina on specificity: P=0.002; α Compared to the combina on NPV: P= 0.001; β Compared to the combina on NPV: P= 0.0002;
Table 2: Screening accuracy in women at familial risk of breast cancer (N=2,226, BCs=87, median screening rounds: 3 [IQR 2-5])
PPV: posi ve predic ve value; NPV: nega ve predic ve value; SR: screening round, median and interquar le range; SE: Standard error of mean
*Compared to the combina on sensi vity: P=0.003; **Compared to the combina on sensi vity: P=0.002; α Compared to the combina on specificity; P=0.04
¥Compared to the combina on sensi vity: P=0.003;π Compared to the combina on sensi vity: P<0.0001; ᵟ Compared to the combina on specificity; P=0.006; β Compared to the combina on NPV: P=0.02 £Compared to the combina on sensi vity: P=0.0003; µ Compared to the combina on specificity: P=0.03; & Compared to the combina on NPV: P=0.02;
Cancer detec on: contribu on of screening modali es in different age groups
5
Screening with the combina on of MRI and mammography had higher sensi vity compared with mammography alone in all age groups (Table 2). In women younger than 40 years, the sensi vity of the combina on was 95.4% (SE 5.6) versus 51% (SE 13.4) for mammography (P<0.01) and MRI alone detected seven invasive cancers (of 14 cancers, 50%, four invasive tumours 1-2cm), which were not detected at mammography (Table 3). In women aged 40-49, the sensi vity of the combina on was 98.4% (SE 2) compared to mammography sensi vity 57.2% (SE 8.2, P<0.01) (Table 3) and MRI detected 17 cancers (of 36 cancers, 47.2%) which were not detected at mammography, among which 12 were invasive cancers (six tumours were ≤1cm), four were ductal carcinoma in situ (DCIS) and one was unspecified. In women ≥ 50, sensi vity of the combina on was 97.4% (SE 3.0) (compared to 66.6%, SE 7.8 P>0.05, for mammography), and MRI alone detected 12 cancers (of 37 cancers, 32.4%, including two cases of DCIS and one invasive tumour ≤1cm) that were not detected by mammography (Table 3).
1
The sensi vity of mammography improved to some extent with increasing age. The sensi vity of mammography was low in women <40 years (51% SE 13.4) (Table 3). In the women < 40 years, mammography visualised six of 14 cancers and only for one case, the tumour (DCIS) would have been missed if mammography would not have been performed. In women age 40-49, mammography sensi vity was 57.2% (SE 8.2). In this age-group, 18 of 35 cancers could be visualised with mammography, and four cancers (three DCIS and one invasive ≤1cm) were missed by MRI (Table 3). In women aged >=50, mammography had a sensi vity of 66.6% (SE 7.8) (Table 3). Mammography visualised 24 of 37 cancers and four of those (one DCIS and three invasive) were not detected by MRI.
Predic ve value of screening modali es in different age groups
1
The PPV of each modality increased by age, whereas the NPV remained at about 99% or higher: Details are shown in Table 3.
Discussion
1
Our IPD meta-analysis examined the accuracy of screening mammography with or without MRI in women with a strong family history of BC and without a known gene muta on: Based on data for 2,226 women, the observed BC incidence rate was high (11.6 per 1,000 woman-years) and this was evident in both younger and older women, highligh ng the BC burden in this popula on. The sensi vity of
Table 4: Screening accuracy in women at familial risk of breast cancer, stra fied by screening rounds
TP: true posi ve; TN: true nega ve; FP: false posi ve; FN: false nega ve; SE: standard error of mean *compare to the combina on: p<0.01
mammography was only 55.1% (SE 7.0). However, mammography was the most specific modality compared to MRI alone or the combina on of MRI and mammography. Combining MRI and mammography detected the great majority of cancers with a sensi vity of 98%, significantly higher than the 55.1% sensi vity of mammography or 89% of MRI alone (P<0.001). The higher sensi vity of the combina on of mammography and MRI was evident in all age groups. The combina on, however, had the lowest specificity due to a rela vely high number of false-posi ves from MRI.
1
The accuracy of screening mammography for women at elevated risk due to family history and without a proven muta on was examined in some of the original primary studies contribu ng data for this IPD meta-analysis, and the results were generally comparable to this IPD meta-analysis, though based on fewer cases. Of those eligible studies that did not par cipate in this IPD meta-analysis (6,7,9,10,18) , one study reported a low mammography sensi vity (25%) in women with an es mated life me BC risk of 21-40% . Another study reported that mammography detected two of four BCs in 142 women with > 25% life me risk of developing BC (18). Although outside the scope of this IPD, a prospec ve screening study including women (mean age 55, range 25-91) at familial risk and who had heterogeneously dense or extremely dense parenchyma reported a mammography sensi vity of 50% (95%CI 33.8-66.2) (19). Another retrospec ve study showed that annual mammography did not contribute to cancer detec on over annual MRI in a retrospec ve cohort of women younger than 40 years with a life me risk of more than 20%, in whom four BCs were diagnosed (20). Similarly, our IPD meta-analysis showed that mammography sensi vity was rela vely modest (51% [SE 7.0]) whereas adding MRI increased the sensi vity up to 95.4% (SE 1.8).
1
Adding MRI screening has been shown to improve screening accuracy compared to mammography alone in other popula ons at increased BC risk, specifically in women with BRCA1/2 muta on (16). However, adding MRI to mammography gave a significantly higher number of false-posi ve results. In addi on, adding MRI is costly. It should be noted that in these studies, MRI and mammography were performed and interpreted independently, and a posi ve result was referred to whenever either test was posi ve joint interpreta on may poten ally help reduce the number of false posi ves. Yet there s ll remains a lack of evidence regarding the long term effects of annual MRI plus mammography compared with annual mammography.
1
The strength of this IPD meta-analysis is that it collected individual data from six prospec ve screening studies, crea ng the first pooled analysis of women with strong family history and
without a known gene muta on. Further, it allowed implementa on of common defini ons and thresholds as well as subgroup analyses, through the collec ve datasets and through the use of IPD methodology. Nonetheless, there are several limita ons to this meta-analysis. First, the number of women and cancers in some subgroup analyses remained too small to obtain either a clear trend or sta s cal significance. Second, we might have underes mated the accuracy of screening because the data collected in these studies are based on rela vely older imaging technology (17). Although we explored this issue by adding year of screening to our model and found that it did not substan ally change the es mates for sensi vity and specificity. We acknowledge that higher accuracy may be expected for current mammography and MRI technology, due to improved MRI technology (21,22) , be er defined MRI BI-RADS descriptors and diagnos c categories (23) and the increasing use of breast digital tomosynthesis for screening (24,25). Third, we included IPD from six prospec ve screening trials, yet each original study had its own recruitment me frame, age at recruitment, inclusion criteria and risk assessment tool for women without a proven muta on (Supplementary 1); thus, there was unavoidable heterogeneity in the format for family history data and probably risk level, reflec ng real-life prac ces. It was therefore not possible to es mate BC risk for all the study popula on using one common criterion or assessment model without making some assump ons. In addi on, despite some small differences between countries, the overall life me risk to develop BC in countries where the included studies were performed is 1 in 8 (26,27), which is not expected to have any impact on the here presented results on compara ve accuracy. Furthermore, there was no informa on about CHEK2 prevalence in these cohorts, and the effect of this gene muta on in screening accuracy was not examined. Future research focussing on the burden of family history and other gene muta ons may poten ally help refine and personalise screening strategies in these women. Finally, as the included studies inves gated screening accuracy and did not report data on long-term outcomes such as survival, we could not include such outcomes in our analysis.
Our IPD meta-analysis highlights that adding MRI to mammography significantly improves BC detec on in women with a strong family history of BC and without known gene muta ons; however this should be considered against the higher false-posi ve rate (lower specificity) caused by adding MRI screening. Our findings might lead to the conclusion that MRI screening alone may be appropriate for these women; however, compara ve accuracy studies, such as reported in this work, need to be complemented by health-economic evalua on to determine whether such an approach could be cost-effec ve. Also, future research in this
research in this popula on of women is cri cally needed to examine alternate screening approaches, inves ga ng new, faster (and less costly) MRI techniques and tomosynthesis as a replacement for 2-D mammography to develop screening strategies that op mise BC detec on without significantly increasing the false-posi ve recall burden.
Acknowledgements: The authors acknowledge collaborators from HIBCRIT study (Filippo Santoro and
Franca Podo) and from MARIBS study (Gek Kwan-Lim) for contribu ng in data transferring and crosschecking.
Funding source: There was no specific project funding. N. Houssami received support through a
Na onal Breast Cancer Founda on (NBCF Australia) Prac oner Fellowship.
Conflict of interest statement
Three authors reported consul ng or advisory roles: Thomas H. Helbich (Siemens; Philips); Francesco Sardanelli (Bayer Healthcare including honoraria; Bracco Imaging including funding; IMS-Gio o including funding); Edwin R. van den Heuvel (MSD). All other authors have no conflict of interest to report.
Appendix
Supplementary 1: Overview of inclusion criteria of studies included in the IPD
Study/
Number of women (N)/ Number of breast cancers (BCs)
Time frame
Age at recruitment (median, IQR) Gene c status (%) Inclusion criteria MRISC study (12) N=1,426 BCs=38 * Time frame: 11/1999-3/2006 Age: 40 [34-47] Nega ve tes ng: 0% Unknown muta on: 100%
Women at high risk (cumula ve life me risk 30-50%) according to modified Claus model, defined as women who have at least one of the following criteria:
+ A first-degree family member with a BRCA1/2 muta on
+ A first-degree-family member and two other first or second degree family members affected with breast or ovarian cancer
+ Two first-degree or one first and one second-degree family members with breast cancer, mean age at diagnosis 45 years or younger.
Or women at moderate risk (cumula ve life me risk 15-30%) according to modified Claus model, defined as women who have at least one of the following criteria:
+ A second-degree family member with a BRCA1/2 muta on.
+ One first-degree family member with breast cancer younger than 40 years
+ Two first-degree family members or one first- and one second degree family member with breast cancer together, mean age at diagnosis between 45 and 60 years.
+ Three second-degree family members with breast or ovarian cancer
+ Two first or one first and one second-degree family members, 1 affected with breast cancer younger than 55 and one with ovarian cancer
Austria study (11) N=457 BCs=22 Time frame: 1/2002-5/2011 Age: 43 [36-50] Nega ve tes ng: 19% Unknown muta on: 81%
Women with life me risk >20%, who have at least one of the following criteria:
+ Three or more rela ves on the same side of the family with breast cancer diagnosed before the age of 61 or one rela ve diagnosed with ovarian cancer at any age,
+ Two or more rela ves on the same side of the family with breast cancer diagnosed before the age of 51 or one rela ve diagnosed with ovarian cancer at any age, and
+ Rela ve with breast cancer diagnosed before the age of 36.
In any case, the women had to be a first-degree rela ve of one of the affected rela ves or be one of the affected herself.
Exclude women who were proved not to be carriers in a family with a proven muta on because their breast cancer risk was not assumed to be elevated; women with clinical signs of breast cancer at their first visit were also excluded but became eligible to par cipate 1 year a er treatment.
MARIB study (8) N=452 BCs=8 Time frame: 8/1997-5/2004 Age: 40 [37-44.5] Nega ve tes ng: 0% Unknown muta on: 100%
Women age 35-49 who have at least one of following criteria: + At 50% risk of being carrier because of known muta on in parent or sibling, or
+ At a 50% risk of carrying BRCA1/2 because of family history: having 4 or more cases of female breast cancer below the age of 60; or having 4 or more cases of female breast and ovarian cancer where the breast cancer is below the age of 60 and the ovarian cancer is at any age; or having 4 or more cases of female and male breast cancer where female cancer is below the age of 60 and the male cancer is at any age
Bilateral breast cancer or breast and ovarian cancer in the same person, can contribute two cases to the above provided that both are proven primary carcinomas.
Or women aged 25-49 who have at least one of the following criteria:
+ At 50% risk of carrying TP53 because of known muta on in parent or sibling, or
+ At 50% risk of carrying TP53 in families with a 75% prior probability of being due to TP53. These are family with classical Li-Fraumeni syndrome(sarcoma
aged <45 years with one of the following in first degree rela ves: sarcoma (any age), brain tumour, early onset of breast cancer, leukaemia,
adrenocor cal carcinoma, and another tumour in a close rela ve which is either cancer aged <45 or sarcoma at any age)
Exclusion: previous breast cancer (include DCIS) HIBCRIT 1 study (13) N=152 Bcs=16 Time frame: 6/2000-1/2007 Age: 44 [36-60] Nega ve tes ng: 30% Unknown muta on: 70%
Women from age 25 who have at least one of the following criteria:
+ Have personal unknown muta onal status but who were first-degree rela ves of BRCA1 or BRCA2 muta on carriers
+ Have a strong family history of breast or ovarian cancer, defined as: having three or more events of breast or ovarian cancer in first- or second-degree rela ves in either the maternal or the paternal line. These three or more events could have included female breast cancer in rela ves younger than 60 years, ovarian cancer in rela ves at any age, or male breast cancer in rela ves at any age.
Toronto study (15) N=100 BCs=3 Time frame: 11/1997-6/2009 Age: 42.5 [36-47] Nega ve tes ng: 30% Unknown muta on: 70%
Women who have at least one of the following criteria: + First degree rela ve with a BRCA muta on, but not tested
+ Three or more rela ves on the same side of the family with breast cancer diagnosed before age 50 or ovarian cancer.
+ Women with a past history of breast cancer were eligible if they had one unaffected breast.
Montreal study (14) N=39 BCs=0 Time frame: 8/2003-5/2007 Age: 40 [34-48] Nega ve tes ng: 30% Unknown muta on: 70%
Women who had a priori carrier probability of ≥30% es mated by BRCAPRO model.
*In the publication of original study, there were 47 breast cancers. Among those, 9 cancers were not included due to being screened with only one modality (MRI or mammography) (n=8) and unclear information on whether the breast cancer was related to the screening (n=1). There were two breast cancers reported as screen detected cancer in the original study. However, they were detected by physical examination and missed by MRI or mammography. According to our definition, those two tumours were considered as interval cancers in this IPD.
Reference 1
(1) Lynch HT, Silva E, Snyder C, Lynch JF. Hereditary breast cancer: part I. Diagnosing hereditary breast cancer syndromes. Breast J 2008 Jan-Feb;14(1):3-13.
1
(2) Claus EB, Risch N, Thompson WD. Autosomal dominant inheritance of early-onset breast cancer. Implica ons for risk predic on. Cancer 1994 Feb 1;73(3):643-651.
1
(3) van Asperen CJ, Jonker MA, Jacobi CE, van Diemen-Homan JE, Bakker E, Breuning MH, et al. Risk es ma on for healthy women from breast cancer families: new insights and new strategies. Cancer Epidemiol Biomarkers Prev 2004 Jan;13(1):87-93.
1
( 4 ) N I C E . C G 1 6 4 F a m i l i a l b r e a s t c a n c e r : N I C E G u i d a n c e . 2 0 1 3 ; Av a i l a b l e a t : h p://guidance.nice.org.uk/CG164/NICEGuidance/pdf/English. Accessed 09/23, 2013.
1
(5) Zonderland HM, van Vegchel T, van Asperen CJ, Benraadt J, de Bock GH, den Heeten GJ, et al. Richtlijn mammacarcinoom (Breast Cancer Na onal Guideline). 2012; Available at:
h p://www.heelkunde.nl/uploads/_a/Gk/_aGkhzjRsgiYhbhbFWFETQ/Mammacarcinoom2012.pdf. Accessed 02/02, 2014.
1
(6) Kuhl C, Weigel S, Schrading S, Arand B, Bieling H, Konig R, et al. Prospec ve mul center cohort study to refine management recommenda ons for women at elevated familial risk of breast cancer: the EVA trial. J Clin Oncol 2010 Mar 20;28(9):1450-1457.
1
(7) Kuhl CK, Schrading S, Leutner CC, Morakkaba -Spitz N, Wardelmann E, Fimmers R, et al. Mammography, breast ultrasound, and magne c resonance imaging for surveillance of women at high familial risk for breast cancer. J Clin Oncol 2005 Nov 20;23(33):8469-8476.
1
(8) Leach MO, Boggis CR, Dixon AK, Easton DF, Eeles RA, Evans DG, et al. Screening with magne c resonance imaging and mammography of a UK popula on at high familial risk of breast cancer: a prospec ve mul centre cohort study (MARIBS). Lancet 2005 May 21-27;365(9473):1769-1778. 1
(9) Lehman CD, Blume JD, Weatherall P, Thickman D, Hylton N, Warner E, et al. Screening women at high risk for breast cancer with mammography and magne c resonance imaging. Cancer 2005 May 1;103(9):1898-1905.
1
(10) Lehman CD, Isaacs C, Schnall MD, Pisano ED, Ascher SM, Weatherall PT, et al. Cancer yield of mammography, MR, and US in high-risk women: prospec ve mul -ins tu on breast cancer screening study. Radiology 2007 Aug;244(2):381-388.
1
(11) Riedl CC, Lu N, Bernhart C, Weber M, Bernathova M, Tea MK, et al. Triple-Modality Screening Trial for Familial Breast Cancer Underlines the Importance of Magne c Resonance Imaging and Ques ons the Role of Mammography and Ultrasound Regardless of Pa ent Muta on Status, Age, and Breast Density. J Clin Oncol 2015 Feb 23.
(12) Rijnsburger AJ, Obdeijn IM, Kaas R, Tilanus-Linthorst MM, Boetes C, Loo CE, et al. BRCA1-associated breast cancers present differently from BRCA2-associated and familial cases: long-term follow-up of the Dutch MRISC Screening Study. J Clin Oncol 2010 Dec 20;28(36):5265-5273.
1
(13) Sardanelli F, Podo F, Santoro F, Manoukian S, Bergonzi S, Trecate G, et al. Mul center surveillance of women at high gene c breast cancer risk using mammography, ultrasonography, and contrast-enhanced magne c resonance imaging (the high breast cancer risk italian 1 study): final results. Invest Radiol 2011 Feb;46(2):94-105.
1
(14) Trop I, Lalonde L, Mayrand MH, David J, Larouche N, Provencher D. Mul modality breast cancer screening in women with a familial or gene c predisposi on. Curr Oncol 2010 Jun;17(3):28-36.
1
(15) Passaperuma K, Warner E, Causer PA, Hill KA, Messner S, Wong JW, et al. Long-term results of screening with magne c resonance imaging in women with BRCA muta ons. Br J Cancer 2012 Jun 26;107(1):24-30.
1
(16) Phi XA, Saadatmand S, De Bock GH, Warner E, Sardanelli F, Leach MO, et al. Contribu on of mammography to MRI screening in BRCA muta on carriers by BRCA status and age: individual pa ent data meta-analysis. Br J Cancer 2016 Feb 23.
1
(17) Phi X, Houssami N, Obdeijn I, Warner E, Sardanelli F, Leach MO, et al. Magne c Resonance Imaging Improves Breast Screening Sensi vity in BRCA Muta on Carriers Age >/= 50 Years: Evidence From an Individual Pa ent Data Meta-Analysis. J Clin Oncol 2014 Dec 22.
1
(18) Weinstein SP, Localio AR, Conant EF, Rosen M, Thomas KM, Schnall MD. Mul modality screening of high-risk women: a prospec ve cohort study. J Clin Oncol 2009 Dec 20;27(36):6124-6128.
1
(19) Berg WA, Blume JD, Cormack JB, Mendelson EB, Lehrer D, Bohm-Velez M, et al. Combined screening with ultrasound and mammography vs mammography alone in women at elevated risk of breast cancer. JAMA 2008 May 14;299(18):2151-2163.
1
(20) Narayan AK, Visvanathan K, Harvey SC. Compara ve effec veness of breast MRI and mammography in screening young women with elevated risk of developing breast cancer: a retrospec ve cohort study. Breast Cancer Res Treat 2016 Aug;158(3):583-589.
1
(21) Pinker K, Helbich TH, Morris EA. The poten al of mul parametric MRI of the breast. Br J Radiol 2017 Jan; 90(1069): 20160715.
1
(22) Bickelhaupt S, Laun FB, Tesdorff J, Lederer W, Daniel H, S eber A, et al. Fast and Noninvasive Characteriza on of Suspicious Lesions Detected at Breast Cancer X-Ray Screening: Capability of Diffusion-weighted MR Imaging with MIPs. Radiology 2016 Mar;278(3):689-697.
1
(24) Bernardi D, Macaskill P, Pellegrini M, Valen ni M, Fanto C, Os llio L, et al. Breast cancer screening with tomosynthesis (3D mammography) with acquired or synthe c 2D mammography compared with 2D mammography alone (STORM-2): a popula on-based prospec ve study. Lancet Oncol 2016 Aug;17(8):1105-1113.
1
(25) Cia o S, Houssami N, Bernardi D, Caumo F, Pellegrini M, Brunelli S, et al. Integra on of 3D digital mammography with tomosynthesis for popula on breast-cancer screening (STORM): a prospec ve comparison study. Lancet Oncol 2013 Jun;14(7):583-589.
1
(26) Arnold M, Karim-Kos HE, Coebergh JW, Byrnes G, An lla A, Ferlay J, et al. Recent trends in incidence of five common cancers in 26 European countries since 1988: Analysis of the European Cancer Observatory. Eur J Cancer 2015 Jun;51(9):1164-1187.
1
(27) Canadian Breast Cancer Founda on. Breast cancer in Canadian, 2016. a on. Breast cancer in Canadian, 2016.
