Breast Cancer Res Treat
DOI 10.1007/s10549-014-2931-9
CLINICAL TRIAL
MRI breast screening in high-risk women: cancer detection
and survival analysis
Evans D. Gareth • Kesavan Nisha • Lim Yit • Gadde Soujanye • Hurley Emma
Nathalie J. Massat • Anthony J. Maxwell • Ingham Sarah • Eeles Rosalind •
Martin O. Leach • MARIBS Group • Howell Anthony • Duffy Stephen
Received: 13 March 2014 / Accepted: 18 March 2014
Ó Springer Science+Business Media New York 2014
Abstract Women with a genetic predisposition to breast
cancer tend to develop the disease at a younger age with
denser breasts making mammography screening less
effective. The introduction of magnetic resonance imaging
(MRI) for familial breast cancer screening programs in
recent years was intended to improve outcomes in these
women. We aimed to assess whether introduction of MRI
surveillance improves 5- and 10-year survival of high-risk
women and determine the accuracy of MRI breast cancer
detection compared with mammography-only or no
enhanced surveillance and compare size and pathology of
The members of the MARBIS Group are listed in Appendix.
Electronic supplementary material The online version of this
article (doi:10.1007/s10549-014-2931-9) contains supplementary
material, which is available to authorized users.
E. D. Gareth
K. Nisha
I. Sarah
H. Anthony
Genesis Breast Cancer Prevention Centre, University Hospital of
South Manchester NHS Foundation Trust,
Wythenshawe, Manchester M23 9LT, UK
E. D. Gareth (&)
Manchester University Department of Genomic Medicine,
Manchester Academic Health Science Centre, Central
Manchester Foundation Trust, St. Mary’s Hospital, Oxford Road,
Manchester M13 9WL, UK
e-mail: gareth.evans@cmft.nhs.uk
E. D. Gareth
H. Anthony
Manchester Breast Centre, Manchester Cancer Research Centre,
Christie Hospital, University of Manchester,
Withington, Manchester M20 4BX, UK
K. Nisha
L. Yit
G. Soujanye
H. Emma
A. J. Maxwell
Department of Breast Imaging, Nightingale Centre, University
Hospital of South Manchester NHS Foundation Trust,
Wythenshawe, Manchester M23 9LT, UK
•
cancers detected in women screened with MRI ? mam-
mography and mammography only. We used data from two
prospective studies where asymptomatic women with a
very high breast cancer risk were screened by either
mammography alone or with MRI also compared with
BRCA1/2 carriers with no intensive surveillance. 63 can-
cers were detected in women receiving MRI ? mammog-
raphy and 76 in women receiving mammography only.
Sensitivity of MRI ? mammography was 93 % with 63 %
specificity. Fewer cancers detected on MRI were lymph
node positive compared to mammography/no additional
screening. There were no differences in 10-year survival
between the MRI ? mammography and mammography-
only groups, but survival was significantly higher in the
MRI-screened group (95.3 %) compared to no intensive
screening (73.7 %; p = 0.002). There were no deaths
among the 21 BRCA2 carriers receiving MRI. There
N. J. Massat
D. Stephen
Centre for Cancer Prevention, Wolfson Institute of Preventive
Medicine, Queen Mary University of London, Charterhouse
Square, London EC1 6BQ, UK
E. Rosalind
Oncogenetics Team, Division of Genetics and Epidemiology,
The Institute of Cancer Research and Royal Marsden NHS
Foundation Trust, Downs Road, Sutton, Surrey SM2 5PT, UK
M. O. Leach
CRUK Cancer Imaging Centre, The Institute of Cancer Research
and Royal Marsden NHS Foundation Trust, Downs Road,
Sutton, Surrey SM2 5PT, UK
123

appears to be benefit from screening with MRI, particularly
in BRCA2 carriers. Extended follow-up of larger numbers
of high-risk women is required to assess long-term
survival.
Keywords MRI
Breast cancer
BRCA1
BRCA2,
survival
Introduction
Breast cancer is the most commonly diagnosed cancer
amongst women in the UK and ranks as the second most
common cause of cancer-related female mortality. In 2011,
49,936 women were diagnosed with the disease in the UK
(population 60 million) and as a result, 11,684 died [1].
Although a majority of breast cancers occur spontaneously,
*27 % are heritable as judged by twin studies [2]. Women
with a proven or suspected genetic predisposition are more
likely to develop cancer at a younger age. A number of
studies have shown that young women who develop breast
cancer have poorer survival rates than older women [3–8].
Some poor prognosis tumors are related to germline
mutations in the BRCA1, BRCA2, and TP53 genes [8, 9],
and detection of these mutations predicates more intensive
screening at a younger age.
Screening higher risk women for breast cancer in the UK
were predominantly based on the use of mammography [10,
11]. However, the higher breast density seen in younger
women reduces the sensitivity of mammography for cancer
detection [12]. Further, some studies suggest that cancers that
develop in women with BRCA1 or BRCA2 mutations are more
likely to have benign features on mammography [13, 14].
There is good evidence that mammographic screening of
women aged 50–70 years reduces breast cancer mortality
[11], and there is evidence for a probable mortality advantage
in young women predominantly from screening young
women with a moderate family history (FH01) [15]. A num-
ber of studies have suggested that women at high risk of
developing breast cancer would benefit from screening by
magnetic resonance imaging (MRI) in addition to mammog-
raphy [12, 16–22]. MRI is being used more commonly in this
high-risk group of women worldwide, and although the cancer
detection rates are higher with MRI ? mammography, there
is no evidence that survival is improved by the use of MRI.
This report aims to determine the accuracy of MRI in
detecting breast cancer in high-risk women, to compare the
size and pathology of cancer at detection between MRI-
screened patients and mammography-screened patients and
to determine whether the introduction of MRI ? mam-
mography surveillance improves the 5- and 10-year sur-
vival compared with mammography and no enhanced
screening.
123
Breast Cancer Res Treat
Material and method
Subjects
MRI group
From August 1997 to May 2004, a prospective cohort
study, magnetic resonance imaging breast screening
(MARIBS) [18], was conducted in the UK in which 649
asymptomatic women aged 35–55 years were selected to
receive annual MRI screening based on the presence of a
proven or likely BRCA1, BRCA2, or TP53 mutation. Fol-
lowing the MARIBS study, after endorsement of annual
MRI in the same high-risk group by NICE [23], a pro-
spective surveillance program was designed where 338
asymptomatic women accepted the offer to receive annual
MRI screening in addition to mammography at the
Nightingale Centre at the University Hospital of South
Manchester. Women received annual mammography and
MRI 6 months apart rather than simultaneously. NICE
criteria from 2006 to 2013 are shown in supplementary
Table 1. 70 % of these women had undergone mammog-
raphy screening prior to initiation of the combined
screening approach.
Data on patients from the Nightingale Centre were
obtained from the Family History Clinic database and
hospital notes. Information on the number of women who
had false positive MRI scans and women who were diag-
nosed with cancer and the number of scans prior to
detection were obtained. The method of detection, cancer
size, pathological type and grade, and lymph node status
were recorded.
Invasive cancer size was grouped into four categories, as
follows:
1a [ 1 mm B 10 mm; 1b [ 10 mm B 20 mm; 2 [
20 mm B 50 mm; and 3 [ 50 mm.
For the survival analysis, women were censored on
30/05/2013 or at date of death. Survival was confirmed
from the NHS tracing system and by flagging of MARIBS
patients on the national ONS (Office of National Statistics)
database [19].
Mammogram group
Women who were BRCA1/2 mutation carriers (usually
identified after the breast cancer diagnosis) and/or were at
equivalent risk (i.e., 40 % lifetime risk) of developing
breast cancer to MARIBS and NICE recommendations and
who had received mammography screening only were
selected as controls. Breast cancers were diagnosed
between 1990 and 2013 and were limited to those aged
B55 years.
Breast Cancer Res Treat
In women who were diagnosed with cancer, the tumor
size at detection, pathology, and lymph node status were
compared between the women who received MRI screen-
ing ? mammography and those who received only
mammography.
Unscreened group
This comprises 557 women with BRCA1/2 mutations who
were identified from the Manchester genetic database as
having been diagnosed with breast cancer after 1990 aged
B55 years and who had not undergone intensive surveil-
lance (a subset aged 50–55 years had undergone population
3 yearly mammography).
Statistics
All statistical analyses were performed in R version 3.0.2
(The R Foundation for Statistical Computing).
The difference between the MRI ? mammography
group and the mammogram-only group with regard to
tumor size at detection, pathology, and lymph node status
was analyzed by cross-tabulation, and significance was
tested using Pearson’s v2 tests. A survival analysis was
conducted by comparing the date of diagnosis of cancer to
the last follow-up date or the date of death. Kaplan–Meier
survival curves (proportion surviving) of time-to-diagnosis
were produced comparing the two screened groups to the
unscreened cohort among BRCA1/2 women. We deter-
mined statistical significance using log-rank tests. Cox
(proportional hazards) regression was also performed,
adjusting for age at diagnosis (fitted using a natural spline).
p values of \0.05 were considered statistically significant.
Results
Sixty-three breast cancers in 959 women were detected in
the MRI ? mammography group—37/647 in the MARIBS
cohort and 26/312 in the Nightingale Centre group
screened after completion of entry to the MARIBS study
(70 % of the latter had undergone previous mammogra-
phy). The age at diagnosis ranged from 29 to 51 years
(mean age 40.2 years). Of the 24 Nightingale Centre can-
cers, 3 were prevalent cancers, 21 were incident cancers,
and two were interval cancers (both identified at risk
reducing mastectomy without symptoms—12 mm DCIS in
a BRCA1/2 negative woman 12-months after last MRI scan
and 2 mm DCIS in a BRCA1 carrier 3-months after last
MRI scan, respectively). A total of 750 screening MRI
scans were performed. Of these, 24 (3 %) detected cancers,
and a further 14 were false positive for malignancy on
biopsy. Sensitivity and specificity of MRI scans in this
study group were 93 and 63 %, respectively. If the two
DCIS cases found at risk reducing mastectomy are exclu-
ded from the analysis on the basis that they are not true
interval cancers, the sensitivity in the Manchester cohort is
100 % for the remaining 24 cancers. 20 cancers were
identified on the alternating program: five cancers (25 %)
were detected on the alternating mammogram and 15
(75 %) on the MRI scan. A 6 mm DCIS (BRCA2), 4 mm
DCIS (BRCA1-later identified with microscopic invasion at
risk reducing mastectomy), and 7 and 10 mm invasive
ductal cancers (BRCA2) were found on mammography
6 months after normal MRI screens. A 29 mm invasive
cancer was found 6 months after an MRI showing a sus-
picious 13 mm area invisible on ultrasound. A 6-month
follow-up MRI also demonstrated the now enlarged lesion
in this BRCA2 carrier.
Seventy-six of 1,223 women had cancers detected in the
mammography-only group (Tables 1, 2). The age at time
of cancer diagnosis ranged from 25 to 55 years
(mean = 43.4 years).
BRCA carriers in study groups
Of the women who developed breast cancer in the com-
bined MRI screening groups, 24 (38 %) were known
BRCA1 carriers, and 21 (33 %) were known BRCA2 car-
riers. In the mammography-only group, there were 27
BRCA1 and 30 BRCA2 carriers (Table 1). There were only
two TP53 carriers both in the MRI group.
Invasive tumor size in the combined MRI-screened
group ranged from 2 to 31 mm (mean 13.2 mm; median
11 mm). Tumor size in the mammography-screened group
ranged from 1 to 50 mm (mean 14.6 mm; median 12 mm
p = NS). In the MRI-screened group, nearly half of tumor
sizes were 1–10 mm. 85 % of tumors detected on MRI
were \20 mm compared to 81 % of invasive tumors on
mammography (Table 2). There were 557 BRCA1/2
mutation carriers identified in the unscreened group. There
were missing data on over 50 % of the invasive tumors
(Table 2). Invasive tumor size ranged from 2 to 120 mm
(mean 23 mm; median 20 mm). MRI screen detected
tumors were significantly more likely to be smaller, non-
invasive, and lymph node negative than unscreened tumors
(p \ 0.0001) but not the mammography-only detected
tumors. (Table 2).
Sixteen (24 %) of the cancers detected in the
MRI ? mammography-screened group were DCIS, and 47
(76 %) were invasive carcinomas. Twelve (19 %) cancers
detected in the mammogram-screened group were DCIS,
and 64 (81 %) were invasive. The proportions of grade 3
cancers were 61, 61, and 74 % in the MRI ? mammog-
raphy, mammography-only, and unscreened groups
respectively.
123

Table 1 Mutation status of women with cancers detected on screening with mammography alone or with additional MRI
Treatment Negative BRCA1
Mammogram 18 (24 %) 27 (35.5 %)
Mammogram ? MRI 15 (25 %) 24 (38 %)
Total 25 (18 %) 51 (36.5 %)
In the MRI-screened group, the axillary lymph node
status was negative in 83 % of cases compared to 70 % in
the mammography-screened group (p = 0.2) and 65 % in
the unscreened group (p \ 0.0001).
Survival analysis
In the MRI ? mammography group, five women died–
three due to breast cancer (both BRCA1 mutation carriers
one with family history of Li Fraumeni Syndrome without
genetic testing), one due to acute myeloid leukemia (a
TP53 mutation carrier), and one due to a pulmonary
embolus (Table 3).
In the mammography-only group, nine deaths occurred.
Six women died of breast cancer (two BRCA1 mutation
carriers, two BRCA2 mutation carriers, and two who tested
negative for BRCA1/2), two of ovarian cancer (both
BRCA1 mutation carriers), and one from lung cancer (a
BRCA2 mutation carrier). Six of the nine women had
axillary lymph node metastases at diagnosis, including five
of six women dying from breast cancer.
In the MRI group, the median follow-up time from the
time of diagnosis of primary tumors in the 43 BRCA1/2
surviving patients was 11.75 years (range 0.34–18.61
years). In the mammogram-only group, the median follow-
up from diagnosis of primary tumors in the 75 surviving
patients was 6.60 years (range 0.30–17.89 years).
In the unscreened cohort, there were 128 deaths of 557
women (23 %). 109 were from breast cancer, 12 from
ovarian cancer, and 7 from other causes (such as myocar-
dial infarction, stomach cancer, pulmonary embolism, lung
cancer, and endometrial cancer.)
Of 128 deaths that occurred in the unscreened group, 64
(50 %) were BRCA1, and 64 (50 %) were BRCA2 mutation
carriers. In the unscreened sample, there were a total of
64/287 (22 %) deaths in women who were known BRCA1
carriers and 64/270 (24 %) BRCA2 carriers. Median time
from breast cancer diagnosis to identification as a mutation
carrier was 4.27 years with a mean of 5.37 years.
A survival analysis was conducted between
MRI ? mammography-screened carriers of BRCA1/2
mutations and unscreened mutation carriers. Overall
10-year survival in the MRI-screened cohort was 95.3 %
compared to 73.7 % in the unscreened cohort (Table 4,
log-rank test for overall survival p = 0.002; Fig. 1). After
123
Breast Cancer Res Treat
BRCA2 TP53 Unknown Total
30 (40.0 %) 0 (39.5 %) 1 (1 %) 76
21 (33 %) 2 (3 %) 1 (2 %) 63
51 (36.5 %) 2 (1.5 %) 2 (1 %) 139
adjusting for age at diagnosis, the hazard rate ratio (HR)
was 0.13 (95 % CI 0.032–0.53, p = 0.004).
In the mammography-only group, the overall 10-year
survival was 87.7 %; the hazard rate ratio was not signif-
icantly different from 1 (HR = 0.52, 95 % CI 0.24–1.11,
p = 0.09) when compared to the unscreened group.
The difference in survival between the MRI ? mam-
mography and the mammography-only groups was not
significant at the 5 % level (log-rank test for overall sur-
vival p = 0.075).
BRCA1 carriers who received MRI screening were
compared to BRCA1 carriers who were not in an intensive
screening program. We observed significantly better sur-
vival in BRCA1 carriers in the MRI group (HR = 0.21,
95 % CI 0.051–0.89, p = 0.03). As there were no BRCA2
deaths in the MRI-screened cohort, a comparison between
BRCA2 carriers receiving screening by MRI versus mam-
mography could not be carried out. Instead, a comparison
was made between women in an intensive screening pro-
gram (MRI or mammography) versus women who were not
in an intensive program. BRCA2 mutation carriers who
were in an intensive screening program had an overall
improved survival (HR = 0.21, 95 % CI 0.066–0.67,
p = 0.008) compared to unscreened women.
Finally, we assessed whether carriers of BRCA1 and
BRCA2 mutations in the unscreened group differ from each
other: we could not detect any difference in survival
[HR = 1.00, 95 % CI 0.70–1.43, p = 0.99 (Table 5)].
Discussion
Since the publication of a number of cohort studies of MRI
screening in women at high risk of breast cancer [12, 16–
22], there has been hope that such screening would trans-
late into a survival advantage [23–26]. However, formal
evidence for a survival advantage has not so far been
published. Indeed a recent publication from Norway
showed only 69 % survival at 10-years in BRCA1 mutation
carriers undergoing annual MRI and mammography [27].
In a previous report by Moller et al. [28], the 5-year sur-
vival for women with Stage 1 breast cancer was only 82 %
for BRCA1 mutation carriers compared to 98 % in the
general population [28]. This suggests that simply
 Breast Cancer Res Treat

Table 2 Invasive status of all cancers, and size, grade, TNM stage and lymph node status of invasive breast cancers (all women included)
Cohort Treatment CIS (%) Invasive Invasive Invasive Invasive Invasive size Invasive Invasive Invasive Invasive Invasive LN Invasive LN Stage 1 Stage Stage Stage
(%) grade grade grade 3 Missing (%) size 1a size 1b size 2 (%) size 3 Missing (%) negative (%) (%) 2a (%) 2b 3 (%)
missing (%) 1–2(%) (%) (%) (%) (%) (%)

G1 No screening 17 (3.1) 540 (96.9) 165 (30.6) 96 (17.8) 279 (51.7) 290 (53.7) 27 (10.8) 114 (45.6) 103 (41.2) 6 (2.4) 288 (53.3) 163 (64.7) 117 (42) 98 (35) 53 11 (4)
(19)
G2 Mammogram 12 (15.8) 64 (84.2) 0 25 (39.1) 39 (60.9) 2 (3.1) 25 (40.3) 25 (40.3) 12 (19.4) 0 (0.0) 1 (1.6) 44 (69.8) 40 (62.5) 15 7 (11) 2 (3)
(23.5)
G3 Mammogram ? MRI 16 (25.4) 47 (74.6) 0 18 (39.1) 29 (61.7) 0 (0.0) 24 (51.0) 16 (34.0) 7 (15.0) 0 (0.0) 0 (0.0) 39 (83) 36 (77) 8 (17) 3 (7) 0
Total 45 (6.3) 651 (93.7) 165 (25.4) 139 (21.4) 345 (53.2) 292 (14.1) 76 (20.9) 155 (43.3) 122 (34.1) 6 (1.7) 289 (44.5 246 (67.9) 193 121 63 13
2
Pearson’s v test with Yates’ continuity correction for independence
No screening vs Mammo ? MRI
Invasiveness: p value \0.0001
Invasive size: p value \0.0001
Invasive LN status: p value \0.0001
Stage: p value \0.0001
Mammo vs Mammo ? MRI
Invasiveness: p value = 0.3
Invasive size: p value = 0.6
Invasive LN status: p value = 0.2
Stage: p value = 0.4

123
 Breast Cancer Res Treat

Table 3 Deaths occurring in women undergoing MRI screening

Patient no. Genetic status Tumor size (mm) Pathology Grade Lymph node status Cause of death

1 BRCA1 31 IDC ?DCIS 3 Negative Breast Cancer

2 BRCA1 30 IDC 3 Positive Breast cancer
3 TP53 25 DCIS High grade Negative Acute myeloid leukaemia
4 LFS family 17 Invasive ? DCIS 2 Negative Breast cancer
5 Br/ov family 20 IDC 3 Negative Pulmonary embolus

Table 4 Five- and 10-year overall survival in BRCA women

Cohort Treatment Follow-up Number at risk Number of events % Overall survival (95% CI)

G1 No screening 5-year 320 59 86.7 (83.6–90.0)

10-year 172 101 73.7 (69.3–78.4)
G2 Mammogram 5-year 35 4 90.7 (82.4–99.8)
10-year 18 5 87.7 (78.0–98.5)
G3 Mammogram ? MRI 5-year 35 2 95.3 (89.3–100.0)
10-year 23 2 95.3 (89.3–100.0)
Log-rank test results for overall survival
No screening vs Mammo ? MRI: p = 0.002
Mammo vs Mammo ? MRI: p = 0.075
No screening vs Mammo: p = 0.1

studies [12, 16–22]. A prospective study by Kuhl et al. [12]

Fig. 1 Overall survival in BRCA women
identifying early stage disease may not translate to a sur-
vival advantage in BRCA1 carriers in particular.
In the present study of younger high-risk women, the
sensitivity of MRI was 93–100 %. This compares to the
reported sensitivity of MRI of 88-100 % in various other
reported MRI sensitivity of 100 % (31/31) for asymptom-
atic women at moderate to high risk. The 63 % specificity
of MRI in this study is comparable to specificities of
37–70 % in other studies (Table 6).
There appears to be no difference in the proportion of
DCIS to invasive carcinoma in the groups screened by
either mammography or MRI. Likewise there was no dif-
ference in the grades of cancer detected by these two
screening modalities. However, in the clinically important
parameter of lymph node involvement, there were signifi-
cantly fewer patients with positive lymph nodes at the time
of surgery in the MRI-screened group compared to the
mammogram-screened group, but only significantly less
than in the non-enhanced screening group. This is similar
to the relatively high proportion of node-negative women
in MRI-based surveillance studies of 14–29 % [12, 16–22;
Table 6].
Thus, overall we observed a substantial and significant
advantage in survival for those under surveillance com-
pared to those not and a borderline significant advantage in
survival between those screened with mammogra-
phy ? MRI compared to those screened with mammog-
raphy alone. Due to small numbers of deaths, multivariable
survival analysis was not feasible, but the survival differ-
ences are consistent with differences in size, node status,
and grade of the tumors diagnosed.

123
Breast Cancer Res Treat

Table 5 Cox regression in BRCA1/2 women

Subset Cohort Treatment Number at risk Number of events Hazard rate ratio 95 % CI p value

BRCA1/2 657 137

G1 No screening 557 128 – – –
G2 Mammogram 57 7 0.52 0.24–1.11 0.09
G3 Mammogram ? MRI 43 2 0.13 0.032–0.53 0.004
BRCA1 337 70
G1 No screening 287 64 – – –
G2 Mammogram 27 4 0.63 0.23–1.74 0.4
G3 Mammogram ? MRI 24 2 0.21 0.051–0.89 0.03
BRCA2 320 67
G1 No screening 270 64 – – –
G2 & G3 Mammogram & Mammogram 51 3 0.21 0.066–0.67 0.008
? MRI
BRCA1/2 G1 No screening 557 128
BRCA1 287 64 – – –
BRCA2 270 64 1.00 0.70–1.43 0.99

Table 6 Comparison of current study to similar studies of breast MRI surveillance in high-risk women
Study Study Group MRI MRI MRI invasive Invasive CIS Invasive LN ?ve
Type sensitivity specificity cancer size cancer grade 3
(%) (%) (\10 mm)

Kuhl [16] P H 100 98 NR 34/43 (79 %) 9/43 (21 %) 11/24 (46 %) 5/31 (16 %)
Warner [17] P M 95.5 NR 5/16 (31 %) 16/22 (73 %) 6/22 (27 %) NR 2/14 (14.2 %)
Leach [18] P H 94 77 11/29 (38 %) 29/35 (83 %) 5/35 (14 %) 19/29 (66 %) 5/24 (21 %)
Sardenelli P H 100 NR 3/14 (21 %) 14/18 (78 %) 4/18 (22 %) 8/14 (57 %) 3/13 (23 %)
[19]
Hagen [20] P M 86 NR 6/21 (29 %) 21/25 (84 %) 4/25 (16 %) 13/21 (62 %) 6/21 (29 %)
Rijnsberger P H 77.4 89.7 30/74 (40 %) 78/98 (80 %) 20/98 (20 %) 28/72 (39 %) 22/72 (31 %)
[22]
Current P H 93-100 63 24/47 (51 %) 47/63 (75 %) 16/63 (25 %) 29/47 (61 %) 8/47 (17 %)
study
P prospective study, H high-risk women including mutation carriers, M mutation carriers, NR not recorded

The mean size of the MRI-detected invasive cancers in
the German study [14] was 12.4 mm (median 11.0 mm),
which is similar to our MRI study group’s mean of
13.3 mm (median 11.0 mm). Our study, however, includes
a higher percentage of BRCA mutation carriers. One might
expect the cancers that developed in this group to be larger
and more aggressive, and it is reassuring that histologically
the tumors were similar to those occurring in women who
were at lower risk.
Among the BRCA1/2 carriers, women who were
screened by MRI had an overall 10-year survival of 95.3 %
compared to 87.7 % in women who only had mammog-
raphy screening. The most striking differences were
observed when women who had been screened with MRI
were compared with women who had not been screened,
who had a 10-year survival of 73.7 % (p = 0.002). Among
BRCA2 carriers, there were no deaths in the MRI group,
and the combined group of screened versus non-screened
had a substantial survival advantage compared with the
non-screened group (HR = 0.21; p = 0.008). Although
numbers are small, there was still a survival advantage
among the 24 BRCA1 carriers undergoing MRI versus no
screening (HR-0.21; p = 0.03). This survival advantage is
encouraging in view of the poor survival in Norwegian
BRCA1 carriers [26].
In our prospective series, after the MARIBS study was
completed, we employed a strategy of 6-monthly alter-
nating mammography and MRI as reported in one previous

123

study [21] and a health economic evaluation [29]. In con-
trast to the 13 screen detected cancers in a North American
study that were all MRI detected, 5/20 (25 %) were found
in the present study on the alternating mammogram which
suggests there may be a benefit from this approach rather
than carrying out simultaneous screening annually. The
alternating approach was shown to be cost effective,
especially for BRCA1 carriers [29].
The present study has several weaknesses. Although the
study is prospective based on two groups undergoing MRI
screening, the number of BRCA1 and BRCA2 carriers
diagnosed with cancer remains relatively small. The com-
parison group was not a randomized control group, and
there may be differences between the populations in terms
of breast awareness and lifestyle. Nonetheless, the control
group had to be alive to provide a blood sample following
the breast cancer diagnosis, whereas the two study groups
were all fully prospective. Median time to genetic testing
from breast cancer diagnosis in the unscreened group was
over 4 years. This creates a survival advantage for the
control group as those that died prior to considering genetic
testing would not have been identified. There is also likely
to be a short-lead time bias in any screening study [30, 31];
however, there was no difference in 5- and 10-year survival
in MRI-screened women suggesting that this would have
had no effect in the present study.
Conclusion
In conclusion, the present study has shown that MRI
screening with additional mammography detects the great
majority of cancers in high-risk women reliably and early.
A survival benefit is demonstrated compared to an
unscreened group for both BRCA1 and BRCA2.
Acknowledgments We acknowledge support from the NIHR to the
Biomedical Research Centre at The Institute of Cancer Research and
Royal Marsden NHS Foundation Trust.
Conflict of interest None.
Appendix
MARIBS study group Study Advisory Group (past and
present) MO Leach (Chairman and Principal Investigator,
Professor of Physics as Applied to Medicine) J Brown
(Health Economist) A Coulthard (Consultant Radiologist)
AK Dixon (Professor and Honorary Consultant Radiolo-
gist) JM Dixon (Consultant Surgeon and Senior Lecturer)
D Easton (Professor of Genetic Epidemiology) RA Eeles
(Reader in Clinical Cancer Genetics and Honorary Con-
sultant in Cancer Genetics and Clinical Oncology) DG
123
Breast Cancer Res Treat
Evans (Consultant Geneticist) FJ Gilbert (Professor of
Radiology) J Hawnaur (Consultant Radiologist) P Kessar
(Consultant Radiologist) SR Lakhani (Professor of Breast
Cancer Pathology) S Moss (Epidemiologist) A Nerurkar
(Consultant Pathologist) AR Padhani (Consultant Radiol-
ogist) AJ Potterton (Consultant Radiologist) BAJ Ponder
(Professor and Head of Department of Oncology) J Sloane
(Professor of Pathology; deceased) LW Turnbull (Professor
of Radiology and Honorary Consultant) LG Walker (Pro-
fessor of Cancer Rehabilitation) RML Warren (Consultant
Radiologist) Study Staff (past and present) LJ Pointon
(Study Coordinator) RJC Hoff (Assistant Study Coordina-
tor) K Chan (Data Manager) M Khazen (Image Analysis
Physicist) RML Warren (Study Radiologist) J Anderson
(Health Psychologist)C Levesley (Psychology Research
Assistant) I Griebsch (Health Economist) D Thompson
(Statistician) C Hayes (Study Physicist) R Gregory (Study
Physicist) M Sydenham (Acting Study Coordinator) K
Bletcher (Data Manager) GP Liney (Study Physicist) B
Browne (Data Manager) Data Monitoring and Ethics
Committee K McPherson (Chairman, Visiting Professor of
Public Health Epidemiology) R Blamey (Professor Emer-
itus and Consultant Breast Surgeon) SW Duffy (Professor
of Cancer Screening). Trial Steering Committee A Howell
(Chairman, Professor of Medical Oncology) D Easton
(Study Statistician, Genetic Epidemiologist) DG Evans
(Study Representative, Consultant Geneticist) JE Husband
(Host Institution Representative, Professor of Radiology) E
Maher(Independent Member, Professor of Medical Genet-
ics) MJ Michell (Independent Member, Consultant Radi-
ologist) RML Warren (Study Radiologist, Consultant
Radiologist) W Watson (Consumer Representative, Foun-
der of the Hereditary Breast Cancer Group). Recruiting
centres (Number of women recruited) Aberdeen: NE Ha-
ites, B Gibbons, H Gregory, M McJannett, L McLennan
(29); Belfast: PJ Morrison, L Jeffers (12); Birmingham: T
Cole, L Burgess, CmcKeown, JEV Morton (24); Bristol
Royal Infirmary: Z Rayter (3);Cambridge: J Mackay, J
Rankin, LG Bobrow, S Downing, S Everest, A Middleton,
B Newcombe (67);Dundee: D Goudie, D Young
(24);Edinburgh: M Steel, EDC Anderson, J Campbell, JM
Dixon, P Walsh (60); Frenchay HospitalBristol: SJ Caw-
thorn, M Shere, C Dawe (29); Glasgow: R Davidson, CM
Watt (20); Guy’s and St Thomas’ London: SV Hodgson, S
Watts (43); Leeds: C Chu, G Turner, E Hazell, L Rae (55);
Liverpool: I Ellis, J Birch, C Holcombe, S Holcombe, K
Makinson (16);Manchester Regional Genetics Service: DG
Evans, G Hall, A Shenton (157);Newcastle: F Douglas, G
Seymour (111);Northwick Park: J Paterson, C Cummings,
L Jackson (9); Sheffield: OWJ Quarrell, JA Cook, D
Kumar (14);Southampton: DM Eccles, G Crawford, S
Goodman (34);Sutton and St George’s (or collaborators
who referredto this centre): RA Eeles, S Allan, A Ardern-
Breast Cancer Res Treat
Jones, E Bancroft, C Brewer, R Carpenter, C Chapman, DL
Christensen, RC Coombes, S Ebbs, I Fentiman, S Furnell,
R Given-Wilson, S Goff, S Gray, M Greenall, G Gui, T
Homfray, R Houlston, MW Kissin, I Laidlaw, F Lennard, I
Locke, AM Lucassen, F McDuff, K McReynolds, G
Mitchell, MWE Morgan, V Murday, U Querci della Ro-
vere, N Rahman, N Sacks, A Salmon, S Shanley, S Shro-
tria, N Sodha, A Stacey-Clear, C Webster (130).Magnetic
resonance image readers (number of cases read)Aberdeen:
FJ Gilbert (132), G Needham (75); Barnet: GR Kaplan
(19); Belfast: JG Crothers (13); Birmingham: CP Walker
(48); Bristol Royal Infirmary: A Jones (10); Cambridge:
PD Britton (161), AK Dixon (104), R Sinnatamby (25),
RML Warren (759); Dundee: JM Rehman (14), D Shepp-
ard (20); Edinburgh: J Walsh (426); Frenchay Hospital
Bristol: ID Lyburn (23), NF Slack (50); Glasgow: LM
Wilkinson (24); Guy’s and St Thomas’ London: S Rankin
(222); Hillingdon Hospital Middlesex: K Raza (100); Hull:
G Hall (81), P Balan (47), LW Turnbull (221); Liverpool:
GH Whitehouse (47); Manchester—Christie Hospital/
Nightingale Centre: CRM Boggis (80), E Hurley (16), A
Jain (4), S Reaney (49), M Wilson (63); Manchester
Medical School: JM Hawnaur (183), J Jenkins (4); New-
castle: A Coulthard (234), AJ Potterton (321); Northwick
Park: B Shah (57), W Teh (92);Paul Strickland Scanner
Centre London: AR Padhani (269); Royal Hospital Haslar
Gosport: PJ Buxton (2), JM Domjan (2), PAL Gordon (6);
Southampton: M Briley (55), C Rubin (72); Sutton and St
George’s: P Kessar (now at Bromley Hospitals NHS Trust;
256); University College Hospital: MA Hall-Craggs (23).
XRM Readers (number of films read) Aberdeen: HE Deans
(42), K Duncan (47), L Gomersall (30), G Iyengar (3), G
Needham (8); Barnet: GR Kaplan (4); Belfast: JG Crothers
(12), J McAllister (12), JM Kirby (1);Birmingham: S
Bradley (47), MG Wallis (45); Bristol Royal In.rmary: JE
Basten (56), E Kutt (52); Cambridge: PD Britton (185), R
Davies (5), CDR Flower (9), AH Freeman (240), D
O’Driscoll (4), R Sinnatamby (310), RML Warren (426);
Dundee: AM Cook (25), CM Walker (25); Edinburgh: A
Buttimer (55), A Gilchrist (35), BB Muir (106), J Murray
(126), L Smart (10), M Smith (8); Glasgow: C Cordiner
(18), J Litherland (16); Guy’s and St Thomas’ London: A
Jones (51), S McWilliams (76); Hull: AE Hubbard (146);
Liverpool: A Ap-Thomas (1), DA Ritchie (33), F White
(32); Manchester—Christie Hospital/Nightingale Centre:
DL Asbury (46), U Beetles (14), CRM Boggis (212), R
Dobrashian (3), MDJ Harake (15), E Hurley (34), A Jain
(20), S Reaney (74), M Wilson (117); Newcastle: B Kaye
(55), M McElroy (180), L McLean (145), W Wotherspoon
(230);Northwick Park: G Markham (8); Southampton: A
Bisset (2), S Hegarty (57), G Michaels (59), N Robson (2);
Sutton and St George’s: G Brown (41), J Husband (6), KT
Khaw (1), D MacVicar(10), E Moskovic (7), J Murfitt (23).
Other radiology/magnetic resonance staff Aberdeen: ML
Muirhead, TW Redpath, S Semple; Barnet: M Cunning-
ham, S Turnell; Belfast: Creynolds, R Bridcut, J Winder;
Birmingham: P Fergusson, Z Vegnuti; Bristol Royal
Infirmary: S Cowley, K Isaacs, P Richardson; Cambridge: J
Green, I Joubert, J Pinney, C Pittock, E van Rooyen;
Dundee: SJ Gandy, P Martin, T McLeay; Edinburgh: T
Lawton, I Marshall, L Thomson; Frenchay Hospital Bris-
tol: H Albarran, V Blake, J Robson; Glasgow: M Cock-
burn; Guy’s and St Thomas’ London: J Goodey, K
McBride; Hull: D Fagge, S Hunter, G Liney; Liverpool: J
Chance, J Davies, Z Hussain; Manchester—Christie Hos-
pital/Nightingale Centre: Chammond, W Johnson; Man-
chester Medical School: JE Adams, Y Watson; Newcastle:
L Lewis, M Myers; Northwick Park: D Fox, J Johnson, J
Shah; Paul Strickland Scanner Centre, London: L Culver,
R Sale, JJ Stirling, NJ Taylor; Royal Hospital Haslar,
Gosport: E Boyd, J Evans, W Johnston, S Lindsay, R
MacKenzie, H Stansby, B Tailor, L Watts, L WomackS-
outhampton: A Darekar, S King, N Shepherd; Sutton and St
George’s: G Charles-Edwards, E Charles-Edwards, E Scurr
(on behalf of all the MRI radiographers Sutton)
References
1. Cancer Research UK (2010) CancerStats incidence—UK.
[online] Cancer Research UK. Available at: www.cancerre
searchuk.org Accessed Jan 2014
2. Lichtenstein P et al (2000) Environmental and heritable factors in
the causation of cancer—analyses of cohorts of twins from
Sweden, Denmark, and Finland. N Engl J Med 343:78–85
3. Chung M, Chang HR, Bland KL, Wanebo HJ (1996) Younger
women with breast carcinoma have a poorer prognosis than older
women. Cancer 77:97–103
4. Nixon AJ, Neuburg D, Hayes DF et al (1994) Relationship of
patient age to pathologic features of the tumor and prognosis for
patients with stage I and II breast cancer. J Clin Oncol
12:888–894
5. El Saghir NS, Seoud M, Khalil MK et al (2006) Effects of young
age at presentation on survival in breast cancer. BMC Cancer
6:194
6. Anders CK, Hsu DS, Broadwater G et al (2008) Young age at
diagnosis correlates with worse prognosis and defines a subset of
breast cancers with shared patterns of gene expression. J Clin
Oncol 26:3324–3330
7. Copson E, Eccles B, Maishman T, Gerty S, Stanton L, Cutress RI,
Altman DG, Durcan L, Simmonds P, Lawrence G, Jones L, Bliss
J, Eccles D, POSH Study Steering Group (2013) Prospective
observational study of breast cancer treatment outcomes for UK
women aged 18–40 years at diagnosis: the POSH study. J Natl
Cancer Inst 105(13):978–988
8. Evans DG, Moran A, Hartley R, Dawson J, Bulman B, Knox F,
Howell A, Lalloo F (2010) Long term outcomes of breast cancer
in women aged 30 years or younger, based on family history,
pathology and BRCA1/BRCA2/TP53 status Brit J. Cancer
102(7):1091–1098
9. Lakhani SR, Van De Vijver MJ, Jacquemier J, Anderson TJ, Osin
PP, McGuffog L, Easton DF (2002) The pathology of familial
123

breast cancer: predictive value of immunohistochemical markers
estrogen receptor, progesterone receptor, HER-2, and p53 in
patients with mutations in BRCA1 and BRCA2. J Clin Oncol
20(9):2310–2318
10. Duffy SW, Mackay J, Thomas S, Anderson E, Chen TH, Ellis I,
Evans G, Fielder H, Fox R, Gui G, Macmillan D, Moss S, Rogers
C, Sibbering M, Wallis M, Warren R, Watson E, Whynes D,
Allgood P, Caunt J (2013) Evaluation of mammographic sur-
veillance services in women aged 40–49 years with a moderate
family history of breast cancer: a single-arm cohort study. Health
Technol Assess 17(11: vii–xiv):1–95
11. Independent UK (2012) Panel on breast cancer screening the
benefits and harms of breast cancer screening: an independent
review. Lancet 380(9855):1778–1786
12. Lehman CD, Blume JD, Weatherall P et al (2005) Screening
women at high risk for breast cancer with mammography and
magnetic resonance imaging. Cancer 103:1898–1905
13. Hamilton LJ, Evans AJ, Wilson AR et al (2004) Breast imaging
findings in women with BRCA1- and BRCA2-associated breast
carcinoma. Clin Radiol 59:895–902
14. Tilanus-Linthorst M, Verhoog L, Obdeijn IM et al (2002) 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 102:91–95
15. FH01 Collaborative Teams (2010) Mammographic surveillance
in women younger than 50 years who have a family history of
breast cancer: tumor characteristics and projected effect on
mortality in the prospective, single-arm, FH01 study. Lancet
Oncol 11:1127–1134
16. Kuhl CK, Schrading S, Leutner CC et al (2005) Mammography,
breast ultrasound, and magnetic resonance imaging for surveil-
lance of women at high familial risk for breast cancer. J Clin
Oncol 23:8469–8476
17. Warner E, Plewes DB, Hill KA et al (2004) Surveillance of
BRCA1 and BRCA2 mutation carriers with magnetic resonance
imaging, ultrasound, mammography, and clinical breast exami-
nation. J Am Med Assoc 292:1317–1325
18. MARIBS Study Group (2005) Screening with magnetic reso-
nance imaging and mammography of a UK population at high
familial risk of breast cancer: a prospective multicentre cohort
study (MARIBS). Lancet 365:1769–1778
19. Sardanelli F, Podo F, D’Agnolo G et al (2007) Multicenter
comparative multimodality surveillance of women at genetic-
familial high risk for breast cancer (HIBCRIT Study): interim
results. Radiology 242:698–715
20. Hagen AI et al (2007) Sensitivity of MRI versus conventional
screening in diagnosis of BRCA-associated breast cancer in a
national prospective series. Breast. doi:10.1016/j.breast.2007.01.
006
21. Le-Petross HT, Whitman GJ, Atchley DP, Yuan Y, Gutierrez-
Barrera A, Hortobagyi GN, Litton JK, Arun BK (2011) Effec-
tiveness of alternating mammography and magnetic resonance
imaging for screening women with deleterious BRCA mutations
at high risk of breast cancer. Cancer 117(17):3900–3907
22. Rijnsburger AJ, Obdeijn IM, Kaas R, Tilanus-Linthorst MM,
Boetes C, Loo CE, Wasser MN, Bergers E, Kok T, Muller SH,
123
Breast Cancer Res Treat
Peterse H, Tollenaar RA, Hoogerbrugge N, Meijer S, Bartels CC,
Seynaeve C, Hooning MJ, Kriege M, Schmitz PI, Oosterwijk JC,
de Koning HJ, Rutgers EJ, Klijn JG (2010) 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 28(36):5265–5273
23. McIntosh A, Shaw C, Evans G, Turnbull N, Bahar N, Barclay M,
Easton D, Emery J, Gray J, Halpin J, Hopwood P, McKay J,
Sheppard C, Sibbering M, Watson W, Wailoo A, Hutchinson A
(2004 updated 2006) Clinical Guidelines and Evidence Review
for The Classification and Care of Women at Risk of Familial
Breast Cancer, London: National Collaborating Centre for Pri-
mary Care/University of Sheffield. NICE guideline CG041.
www.nice.org.uk
24. Kurian AW, Munoz DF, Rust P et al (2012) An online tool to
guide decisions for BRCA1/2 mutation carriers. J Clin Oncol
30(5):497–506
25. Norman RP, Evans DG, Easton DF, Young KC (2007) The cost-
utility of magnetic resonance imaging for breast cancer in
BRCA1 mutation carriers aged 30–49. Eur J Health Econ
8(2):137–144
26. Griebsch I, Brown J, Boggis C, Dixon A, Dixon M, Easton D,
Eeles R, Evans DG, Gilbert FJ, Hawnaur J, Kessar P, Lakhani
SR, Moss M, Nerurkar A, Padhani AR, Pointon L, Potterton J,
Thompson D, Turnbull LW, Walker LG, Warren R, Leach MO
for the UK Magnetic Resonance Imaging in Breast Screening
(MARIBS) Study Group (2006) Cost-effectiveness of contrast
enhanced magnetic resonance imaging versus X-ray mammog-
raphy of women at a high familial risk of breast cancer. Brit J
Cancer 95(7):801-810
27. Møller P, Stormorken A, Jonsrud C, Holmen MM, Hagen AI,
Clark N, Vabø A, Sun P, Narod SA, Mæhle L (2013) Survival of
patients with BRCA1-associated breast cancer diagnosed in an
MRI-based surveillance program. Breast Cancer Res Treat
139(1):155–161
28. Moller P, Borg A, Evans DG, Haites N, Reis MM, Vasen H,
Anderson E, Steel CM, Apold J, Goudie D, Howell A, Lalloo F,
Maehle L, Gregory H, Heimdal K (2002) Survival in prospec-
tively ascertained familial breast cancer: analysis of a series
stratified by tumor characteristics, BRCA mutations and oopho-
rectomy. Int J Cancer 101(6):555–559
29. Cott Chubiz JE, Lee JM, Gilmore ME, Kong CY, Lowry KP,
Halpern EF, McMahon PM, Ryan PD, Gazelle GS (2013) Cost-
effectiveness of alternating magnetic resonance imaging and
digital mammography screening in BRCA1 and BRCA2 gene
mutation carriers. Cancer 119(6):1266–1276
30. Maurice A, Evans DGR, Shenton A, Boggis C, Wilson M, Duffy
S, Howell A (2006) The screening of women aged less than 50
years at increased risk of breast cancer by virtue of their family
history. Eur J Cancer 42(10):1385–1390
31. Maurice A, Evans DG, Affen J, Greenhalgh R, Duffy SW, Ho-
well A (2012) Surveillance of women at increased risk of breast
cancer using mammography and clinical breast examination:
further evidence of benefit. Int J Cancer 131(2):417–425