Title Page

© 2016 The Authors. Published by the British Institute of Radiology

Microcalcification on mammography: – approaches to interpretation and biopsy

Dr Louise S Wilkinson BA, BM, BCh, FRCR, Consultant Radiologist, Department of

Breast Imaging, St Georges Hospital, Tooting, London SW17 0BZ

FS
Dr Val Thomas MBBS (SGUL), PhD (UL) FRCPath, Consultant Pathologist,
Department of Pathology, St Georges Hospital, Tooting, London SW17 0BZ

O
Dr Nisha Sharma, Director Breast Screening. MBChB, BSc(Hons), MRCP, FRCR,
Leeds Teaching Hospital NHS Trust

O
R
P
D
E
T
C
E
R
R
O
C
N
U
JR
B
 Microcalcification on Mammography

Microcalcification on mammography: – approaches to interpretation and biopsy

This paper discusses the significance of microcalcifications on mammography

and the changes in technology that have influenced management; it also
describes a pragmatic approach to investigation of microcalcification in a UK
screening programme.

FS
1. Background and prevalence of microcalcifications

Microcalcifications result from the deposition of calcium oxalate and calcium

O
phosphate within breast tissue. The mechanism by which calcium deposition
occurs is not clearly understood; it may be an active cellular process, or an effect

O
of cellular degeneration. Calcification deposits are found within the ductal
system, the breast acini, stroma and vessels, mainly as calcium oxalate and

R
calcium phosphate.

P
Calcium oxalate is produced by apocrine cells in the breast. The crystals are
usually colourless and may be difficult to see on routine histopathology without
polarisation. They are mainly related to benign cystic change, but can also be

D
seen in association with breast cancer. Calcium oxalate cannot be metabolised
E
by mammalian cells and there is emerging evidence that exposure to high levels
of oxalate may affect epithelial cells by triggering cellular and genetic changes.1
T
Calcium phosphate, usually in the form of calcium hydroxyapatite (similar to the
C

form of calcium laid down in bone during skeletal growth2), is more easily
recognised in histopathology as it stains purple with haematoxylin and eosin. It
E

is more commonly associated with malignant lesions than calcium oxalate.3

Magnesium substituted hydroxyapatite has also been reported.4 There is
R

evidence that a change in levels and calcium carbonate content of hydroxyapatite

may influence breast cancer cell growth.5
R

Radiographic microcalcification was first described in 1913 by Albert Salomon, a

O

surgeon in Berlin. He imaged over 3000 surgical specimens describing the

association of microcalcifications with breast cancer; demonstrated tumour
C

spread to the lymph nodes, and postulated that there were different types of
breast cancer.6
N

Mammography developed as a specialty through the late 1950s and 1960s, with
U

the first screening equipment introduced in the late 1960s.7 Improvements in

technology with low kV, high definition screen/film combinations and
JR

magnification views allowed the diagnosis of pre-clinical breast cancer. In 1986,

Sickles proposed an interpretation scheme for microcalcifications utilizing a
structured approach of classification into ‘benign’ (requiring no further
intervention), ‘probably benign’ (managed by periodic mammography) and
B

‘suggestive of malignancy’ (requiring biopsy).8

The advent of organised screening during the late 1980s led to an increase in the
detection of microcalcifications and, as a result, an increase in the detection of
ductal carcinoma in situ (DCIS). The age-standardised incidence of DCIS in the

2
 Microcalcification on Mammography

UK has increased from around 3 per 100 000 before the advent of the NHS
Breast Screening Programme, to 23 per 100 000 in 2013. It continues to
increase with the introduction of digital mammography and the UK national trial
assessing the effect of increasing the age range of women invited for screening.9

There is no routinely published data from the UK national screening

programmes describing the radiological features prompting further assessment,

FS
but data from other national screening programmes is available. This indicates
that the recall rate for calcifications ranges from 0.4% to 2% of women screened.
Investigation of mammographic microcalcification results in a diagnosis of

O
malignancy in up to 0.3% of women screened (see Table 1).

O
2. a) Changes due to evolving technology - imaging

R
Triple assessment by clinical examination, imaging and biopsy remains the
fundamental approach to breast diagnosis. The conversion to digital

P
mammography has increased the conspicuity of microcalcification on
mammography and the introduction of increasingly sophisticated biopsy
techniques has facilitated tissue diagnosis.

D
E
Analogue mammography used high-resolution film/screen combinations which
were designed for optimal spatial and contrast resolution at low dose.
T
Computed radiography has been used as an interim step for cost reasons, but
digital mammography is now widely used. Digital mammography employs post-
C

processing of the image to enhance the appearance of microcalcifications: the

comparative data from the Netherlands in Table 1 demonstrates an increase in
E

calcium detection with the change to digital imaging. Computer aided diagnosis
algorithms (CAD) can further increase the detection of microcalcifications, but
R

do not improve cancer detection in a screening setting when mammograms are

double read.14 Magnification views can be used to enhance the morphology of
R

calcifications. Digital breast tomosynthesis does not substantially improve the

interpretation of microcalcifications.15
O

b) Changes due to evolving technology – localisation techniques

C

Microcalcification on mammography is relatively non-specific, and non-operative

N

diagnosis by image guided needle sample is essential. In the early days

localisation was performed using cranio-caudal and lateral mammograms with a
U

localisation compression grid. Early stereotactic approaches using two angled

views to give a three-dimensional coordinate for needle placement, were
JR

hampered by the delays of analogue film processing and patient movement.16

The breakthrough into small field digital technology was a spin-off of the Hubble
Space Telescope in the mid-1990s, when a joint project between NASA and
Scientific Imaging Technologies developed a new charged-couple device.17 The
B

technology allowed high resolution, wide dynamic range and low light
sensitivity, shortening exposure time while preserving image quality, resulting in
the LORAD Stereo Guided Breast Biopsy System. This has been incorporated into
two approaches to stereotactic guided biopsy; it can be performed on dedicated
equipment in the prone position that may be more comfortable for the patient

3
 Microcalcification on Mammography

and reduces the risk of fainting, but does not provide a conventional
mammography facility. Alternatively biopsy may be performed with the patient
seated or recumbent using an add-on device to an upright mammography
machine.

More recently tomosynthesis-guided biopsy technology has become available,

this is reported to be quicker and more effective for sampling low-contrast soft

FS
tissue lesions because it requires less repositioning. However a recent
technology evaluation on behalf of the NHSBSP indicated that a stereotactic
approach was preferred over tomosynthesis guided biopsy for soft

O
microcalcifications.18

O
c) Changes due to evolving technology – biopsy devices

R
At the advent of the NHSBSP in 1988, needle sampling was performed by fine
needle aspiration cytology (FNAC) to achieve pre-operative diagnosis.

P
Cytological analysis of fine needle specimens is a specialised technique requiring
particular expertise on the part of the operator and the cytologist. It is difficult
to assess sample adequacy at the time of procedure, and cytology cannot

D
distinguish between non-invasive and invasive malignancy. NHSBSP guidance
E
(2001) indicates a median absolute sensitivity of cytology of 57% - just over half
the carcinomas identified had pre-operative malignant cytology.19
T
In 1994, Parker published data on the outcomes of 6 152 core biopsies from 20
C

institutions, concluding that 14 gauge core breast biopsy is a reproducible and

reliable alternative to surgical biopsy.20 This became the percutaneous biopsy
E

method of choice, used as a reference for subsequent developments.

R

The shortcomings of 14G biopsies led to the introduction of larger cores assisted
by vacuum to ensure retrieval, which also allowed multiple samples to be
R

collected with a single percutaneous introduction.21,22 Such devices range from

7-12G and can be used to remove tissue volumes equivalent to the weight of a
O

surgical specimen. This allows the pathologist considerably more tissue for
analysis improving diagnostic accuracy, but requires additional processing and
C

reporting time to ensure the sample has been sufficiently scrutinised.

N

3. Radiological identification and interpretation of microcalcification

U

Microcalcifications are seen on many mammograms and there are well-

described patterns that help to distinguish benign from potentially malignant
JR

changes. The BI-RADS lexicon supports consistency in nomenclature and

provides descriptions to discriminate between benign and malignant.23
Approaches to interpretation include appreciation of the extent, morphology and
distribution of the calcifications. The Royal College of Radiologists Breast Group
B

has described a five-point scale to communicate the level of suspicion (see table
2).24 Review of prior mammograms to assess interval change is critical,
although malignant calcifications may occasionally show minimal change in
appearance over several years.25

4
 Microcalcification on Mammography

The primary feature of calcifications that prompts further analysis is clustering

(>5 calcifications in a square cm). Figure 1

Features that suggest benign change include:

 multiple similar clusters in more than one quadrant in one or both
breasts,

FS
 uniformity of the individual flecks
 lack of interval change.

Features that indicate further evaluation is required include:

O
 pleomorphism (variability in shape, size and density),
 linear and branching forms,

O
 segmental distribution within a lobe of the breast
 interval change.

R
The characteristic morphological features of calcification are less reliable in

P
small clusters and just under 50% of DCIS calcification clusters contain punctate
calcifications.25

D
Microcalcifications associated with a mass lesion should be reviewed carefully.
E
Some patterns are clearly benign (such as popcorn calcification in a
fibroadenoma), but malignant change may arise in any area of breast tissue and
T
it is possible, for example, to find ductal carcinoma in situ colonising a
fibroadenoma.
C

If microcalcifications cannot reasonably be assumed to be benign, the

E

appearance is classified as indeterminate to malignant (M3, M4, M5) and further

R

evaluation is required. Magnification views may be used to demonstrate the

morphology more clearly and display very fine calcifications not visible on
R

routine mammography. Lateral mammograms are useful to display the layering

of calcifications in the dependent aspect of microcysts, eliminating the need for
O

biopsy. MRI may be used for further evaluation of calcifications26 and has
potential to improve specificity by reliably identifying benign change, reducing
C

the number of cases requiring biopsy.

N

Any calcification that is not clearly benign should be considered for biopsy.
U

4. Localisation technique and sampling device

Biopsy is recommended when further imaging of calcification has not shown that
JR

changes are clearly benign. In general calcification is biopsied using a

stereotactic approach, or increasingly, tomosynthesis, for localisation. This
requires a team approach to enable accurate positioning and recognition of the
mammographic lesion. It is important to understand the geometry of the
B

localisation device to ensure precise needle placement.

When appropriate, biopsy can be performed using ultrasound. Successful

identification of calcification on ultrasound relies on accurate localisation of the
cluster in the correct quadrant, distance from the nipple, and depth below the

5
 Microcalcification on Mammography

skin surface. Calcification tends to be more conspicuous if there is any change in

the adjacent soft tissue, and ultrasound guided biopsy of calcification therefore
may have a higher yield of malignancy than mammographic imaging.27, 28 The
size of the biopsy needle varies with local protocols, some practitioners favour a
large vacuum assisted sample, and others prefer a 14G sample, with vacuum
assisted biopsy (VAB) for selected cases.

FS
The accuracy of 14G biopsy depends on the size of the microcalcification cluster,
the volume of representative tissue obtained, and the nature of the pathology.
Today there is a range of automated needles available commercially. Most

O
studies describe the use of 14 gauge needles with a long throw (around 2cm); a
smaller gauge or shorter throw provides less tissue for analysis, which reduces

O
the accuracy of the biopsy. The needles are single use, either for use with a
reusable biopsy device, or a fully disposable needle. The re-usable device usually

R
has a more rigorous spring, but can become contaminated with blood tracking
back up the needle and therefore should be sterilized between procedures. The

P
fully disposable devices are marginally more costly and are available in a range
of gauge and throw. Some needles have a single action of advancing the inner
stylet with sample trough, followed by the outer cutting cannula. Others which

D
allow initial advancement of the stylet followed by advancement of the cutting
E
cannula, aid more precise needle placement in the case of small lesions or
minimal tissue depth. All require multiple insertions to retrieve multiple
T
samples, and may only obtain scanty samples in dense or fibrous breast tissue.
C

Immediate specimen radiography is invaluable to assess the adequacy of the

specimen – multiple calcifications are essential, preferably in more than one
E

core, depending on the extent of calcification on the mammogram. At least 5

flecks of calcification should be seen, or flecks in 3 separate cores to ensure that
R

the sample is representative.29 It is important to ensure that calcification seen

on specimen radiography correlates to the size and morphology of the
R

calcification on mammography. A 14G needle sample may be confidently used to

establish a benign diagnosis such as microcystic or fibroadenomatoid change, or
O

a malignant diagnosis of invasive cancer, but underestimates the nature of

disease in approximately 27% of cases when ductal carcinoma in situ and
C

indeterminate lesions such as atypia are present.30

N

Much of the current literature on percutaneous biopsy of the breast describes

outcomes of large sample volumes obtained through a vacuum assisted needle.
U

This has the advantage of reaching a definitive diagnosis with a single procedure,
the duration of the procedure is reduced as the needle is only introduced once
JR

and the samples are large volume. Haemostasis may take longer, but there is no
significant difference in complication rates and the procedure is well tolerated
by patients.31
B

Although vacuum assisted large core biopsy has many benefits, the cost of
consumables is substantially greater than 14G biopsy. In most instances a 14G
biopsy will be sufficient to make the diagnosis, and even small clusters may be
successfully sampled if sufficient care is taken over localisation. It is not
essential to leave a marker clip in situ if calcium remains visible after biopsy,

6
 Microcalcification on Mammography

although an ultrasound visible marker makes it easier to identify the biopsy area
when surgery is anticipated. Exceptions include very small or scattered clusters,
when vacuum assisted biopsy is preferred in the first instance. In 2016, the cost
of a 14 G needle and needle guides is approximately £20. A vacuum needle
requires a dedicated vacuum system and marker clip placement is considered
essential. The cost of a vacuum needle, guides and marker clip is around £300.
In addition, the increase in work for the pathologist is considerable as the

FS
greater volume of tissue may mean that it is difficult to identify small clusters of
calcification and that more levels of multiple blocks need to be examined.

O
5. The role of the pathologist in interpretation of biopsies

O
Accurate diagnosis of microcalcifications depends on effective collaboration
between the radiologist and pathologist. It is important that the radiologist

R
understands the process of sample preparation. The specimen and request
should be fully labeled, including adequate information regarding the nature of

P
the lesion sampled. The radiologist should comment on the presence of
calcification and give their opinion of the likely pathology: the pathologist should
have access to the specimen radiograph. Segregating the samples containing

D
calcification is useful after vacuum assisted biopsy so that subsequent leveling
E
can be concentrated on the relevant material. Adequate fixation is necessary
and larger volume samples take longer to fix. At embedding, despite the use of
T
heated forceps, it is possible for tiny fragments of biopsy to be conveyed into
subsequent samples. In this event, the pathologist may see fragments of
C

irrelevant tissue separate from the main sample, which rarely present a
diagnostic dilemma. This effect can be minimised by ensuring that breast
E

biopsies are interspersed with non-breast samples during embedding. The

samples are embedded in wax and the block is then rough cut until the sample is
R

apparent. Occasionally tissue discarded during this process may include the
relevant calcification.
R

Successive levels are then cut for staining. Current guidance indicates a
O

minimum of 3 levels, a 0.004mm level is cut, then 10 are discarded, the next level
is preserved and 10 discarded. This means that approximately 10% of the first
C

0.13mm of the block is available for review. A 14G core biopsy is approximately
1.6mm thick, so the first 3 levels represent less than 10% of the specimen.
N

Further levels are necessary if the calcification is not visible. In practice, it is

more efficient to cut 6 levels in the first instance and review, before cutting
U

further levels if required. A 9G vacuum sample is approximately 3.6mm thick

and 3 levels constitute less than 4% of the tissue volume. For reference,
JR

microcalcifications are demonstrated on mammography at 0.1mm or larger and

cancer cells are approximately 0.03 mm.

The sections are routinely stained with haematoxylin and eosin. Additional
B

staining, including immunohistochemistry, may be used to assist in diagnosis.32

The pathological entities that are associated with microcalcification have been
well described in the NHSBSP guidance.33 Some commonly encountered entities
are included in Table 3.

7
 Microcalcification on Mammography

There is a spectrum of benign changes described which may be associated with

epithelial proliferation with or without atypia. A variety of lesions are classified
as ‘indeterminate’, some because they show atypical morphology and others
such as radial scar, papilloma and mucinous lesions because they may be
associated with malignancy and are deemed inadequately sampled until
completely removed. The diagnosis and management of indeterminate lesions

FS
will be discussed in a subsequent review. The distinction between atypia and
low grade in situ carcinoma depends on the extent of changes. If the abnormality
measures more than 2mm, or more than one duct system is involved, the lesion

O
is best described as low grade DCIS rather than atypia. In these circumstances, a
larger volume of tissue at pre-operative diagnosis supports more accurate

O
assessment by the pathologist.

R
6. The impact of investigating microcalcification

P
Calcification represents a challenge in both perception and interpretation. Small
clusters of calcification are easy to miss and difficult to interpret. An aggressive
approach to recall and investigation may result in high rates of benign biopsies

D
but reducing the number of women recalled is likely to mean some significant
E
changes are not investigated. The benefit of biopsy is early diagnosis, meaning
treatment can be easier and more effective, with a mortality benefit. The balance
T
of overdiagnosis and overtreatment are difficult to model but were described for
the NHS Breast Screening Programme in 2012.34 Some of the challenges of
C

choosing an approach that balances risk and benefit are discussed below.
E

The rate of microcalcification and DCIS identified at screening depends on the

age of the population and the frequency of screening. It is therefore difficult to
R

establish baseline expected levels for assessment and rate of cancer diagnosis
from calcifications.
R

Farshid published a series of 2545 cases investigated between 1992 and 2007,
O

where microcalcification without soft tissue change was biopsied. Almost half
(47.7%) were graded as indeterminate, 28.3% as suspicious and 24.0% as highly
C

suspicious. After assessment 47.9% of cases were malignant, 4.8%

indeterminate (including atypia) and 47.3% benign. Less than one third (30.9%)
N

of DCIS was low grade, and the features predicting higher grade included
radiological suspicion, extent, and presence of a palpable mass.35
U

National audit data for UK screening units in 2014/15 indicates the rate of
JR

diagnosis of DCIS ranges from 0.5 to 3.1 cases per 1000 women screened
(average 1.8).36 It is likely that this variation is due to different thresholds for
biopsy. Maintaining a high threshold for sampling microcalcifications will
reduce the number of women recalled and subjected to needle sampling. This
B

minimises unnecessary stress and discomfort in many cases, and reduces the
potential for overdiagnosis of low grade DCIS and indeterminate lesions that are
treated but may never affect a woman in her lifetime. However, this is at the cost
of missing some cases of both DCIS and invasive cancer, which may present at
the next screen or as an interval cancer.

8
 Microcalcification on Mammography

A recent analysis of data for over 5 million women screened between 2003 and
2007 investigated the relationship between the detection of DCIS and
subsequent diagnosis of interval cancer in the UK. This showed that the average
frequency of DCIS detected at screening was 1.6 per 1000 women screened (unit
range: 0.54–3.56 per 1000 women screened). There was a significant negative
association of screen-detected DCIS cases with the rate of invasive interval

FS
cancer; for every three cases of DCIS diagnosed there was one less interval
cancer.37

O
Microcalcification was seen more frequently in cancers that were only identified
by one of two readers than in cancers detected by both readers in a screening

O
environment.38 Reviews of imaging of women presenting with screen detected
and interval cancers show that approximately 30% were missed on the prior

R
mammogram. Of these, 18% were microcalcifications with digital
mammography and 32% were microcalcifications with screen/film

P
mammography.39 Warren et al reviewed 193 cases where cancer was diagnosed
after assessment, and found that microcalcifications were more likely to have
been inadequately assessed than other lesions.40 A review of the prior

D
mammograms of women with DCIS showed abnormality in 22% of cases.41 The
E
calcification morphology on the prior mammogram was more indeterminate,
indicating that a lower threshold for sampling indeterminate calcifications would
T
increase the diagnosis of early DCIS.
C

In light of the discussion regarding overdiagnosis and overtreatment,

alternatives to surgical excision for low-grade lesions are being considered. The
E

LORIS trial has been designed to test the efficacy of vacuum assisted excision and
regular surveillance for low-grade DCIS.42
R

The appearance and effect of treating screen-detected DCIS is being recorded by

R

the Sloane Project; a UK wide prospective audit of screen detected DCIS and
atypical hyperplasias of the breast.43 The Sloane project began collecting data in
O

2003/4, including information about pre-operative findings as well as the

management of DCIS. It has identified variation in the use of post-operative
C

radiotherapy, oestrogen receptor measurement, and surveillance protocols in

the UK. Of interest to radiologists, the Sloane Project has demonstrated that
N

typical calcifications in DCIS change with size of lesion. Casting calcifications are
typical of larger areas of DCIS, including low grade, but small clusters of punctate
U

or granular calcifications may represent high grade DCIS where an aggressive

clinical approach is recommended.44
JR

7. A pragmatic approach to investigation of microcalcifications

B

Assessment of microcalcification is described in UK national guidance.45

Calcifications that are not clearly benign at screening mammography are recalled
for assessment, including further views, ultrasound and clinical examination.
Biopsy is recommended in all cases where further imaging is not entirely normal

9
 Microcalcification on Mammography

or benign. A summary of assessment and microcalcification biopsy outcomes for

Southwest London Breast Screening Service is shown in Table 4.

If the microcalcification is confidently seen on ultrasound, biopsy may be

performed under ultrasound guidance. Ultrasound guidance allows real time
visualisation of the needle and is more comfortable for both the patient and the

FS
operator. Occasionally more calcification is seen on ultrasound than on
mammography and it is advisable to place a marker clip at the site of ultrasound-
guided biopsy for calcification to ensure that the site of biopsy may be

O
subsequently demonstrated on mammography.

O
If the microcalcification is not seen with confidence on ultrasound, then
stereotactic biopsy with in-room specimen radiography is necessary. In our

R
practice 14G biopsy is chosen as first line approach for most microcalcifications.
First line vacuum biopsy is used if the cluster is small (less than 5mm) or the

P
calcification is scanty.

Occasionally, stereotactic biopsy is not possible because the individual is unable

D
to tolerate the procedure or the calcification cannot be targeted on the small field
E
biopsy device. When stereotactic biopsy is not possible, it can help to draw a
skin mark over the calcifications during attempted stereotactic localisation to aid
T
localisation on ultrasound.
C

When a firm diagnosis of DCIS or invasive cancer is made, the radiologist aims to
define the extent of disease such that the surgeon is able to remove all disease in
E

a single operation. If the lesion is focal and amenable for local excision then only
the area of most concern is biopsied. If the microcalcification is extensive and
R

heterogeneous, or multifocal, such that mastectomy might be considered, two (or

more) areas may need to be biopsied and marker clips deployed to determine
R

disease extent.
O

Examination of the ipsilateral breast and axilla with ultrasound may

demonstrate soft tissue change associated with invasion, and can give further
C

information on axillary node changes. Nodes in the lower axilla with a thickened
cortex are sampled by FNAC or core biopsy.
N
U

8. Documentation and Communication with the patient

JR

There should be thorough documentation of the procedure, including

identification of the clinician and radiographer, radiation dose, drugs
administered and confirmation of the correct site check in keeping with the
National Safety Standards for Invasive Procedures.46 Details of implanted marker
B

clips should be recorded. As with all procedures, it is important to have formal

training and update procedures in place, and to evaluate the service
continuously through audit and comparison with local and national standards
and targets.

10
 Microcalcification on Mammography

The cooperation of the patient is critical and is best gained by providing a calm
environment and avoiding delay during the intervention. The patient should be
fully informed regarding the nature of the procedure and the need to stay still. It
helps if she can be supported by a healthcare assistant throughout. Written
information should be given well before the procedure so that the patient has
sufficient time to digest the information and ask questions if necessary. There is
variation in approach to confirmation of consent depending on local protocols.

FS
As the patient is fully conscious, written consent is not essential.47

The two main risks associated with biopsy are firstly the harm of recall and

O
intervention in a normal woman who is not diagnosed with cancer, and secondly
the treatment of women who have an abnormal diagnosis which would not cause

O
harm during their lifetime. This is explained in the screening invitation leaflet.48
In addition, specific risks include haematoma which can occasionally be extreme,

R
and ongoing haemorrhage which may need surgical intervention. Infection is
rare. Post-biopsy pain is described but appears sporadic and unpredictable, it

P
may be related to the extent of anxiety prior to the procedure.49

Communication of biopsy results to women is important. When the biopsy is

D
benign, women often ask whether they need more frequent follow up but they
E
should be reassured that the area of the breast sampled is no more likely to
develop malignant change than surrounding tissue. Identification and
T
management of indeterminate lesions will be discussed separately. DCIS may be
a difficult diagnosis to communicate and it is often helpful to use diagrams to
C

demonstrate the difference between DCIS and invasive cancer. Clinicians vary in
the phrases they use to describe non-invasive disease; some refer to it as “early
E

cancer”, others as “pre-cancer”, and some feel strongly that it should not be
referred to as cancer at all, because DCIS is not an obligate precursor of invasive
R

disease. The BBC has an iWonder Interactive Guide that can be helpful.50
Women may wish to know whether the biopsy can cause seeding along the
R

biopsy tract. This may occur, but research has shown that the transplanted cells
are not viable.51
O

Summary
C

The identification and investigation of microcalcifications found on

N

mammography have become more common with improving technology and

there has been a parallel increase in the variety of associated lesions in
U

pathology. This has resulted in an increase in the diagnosis of DCIS. In some

cases women may not benefit (overdiagnosis), but in others early treatment may
JR

pre-empt the development of invasive cancer. This is likely to have contributed

to the reduction in mortality from breast cancer seen since the advent of
screening. Clinicians responsible for the investigation of mammographic
calcification should remain mindful of the need to balance harm and benefit.
B

11
 Microcalcification on Mammography

References

1. Castellaro AM, Tonda A, Cejas HH, et al. Oxalate induces breast

cancer. BMC Cancer. 2015;15:761. PMID: 26493452
2. Cox, Rachel F, Morgan MP. Microcalcifications in breast cancer: Lessons
from physiological mineralization. Bone. 2013 Apr;53(2):437 – 450.
PMID: 23334083

FS
3. Morgan MP, Cooke MM, McCarthy GM.
Microcalcifications associated with breast cancer:
an epiphenomenon or biologically significant feature of selected tumors? J

O
Mammary Gland Biol Neoplasia. 2005 Apr;10(2):181-7. PMID: 16025224
4. Scimeca M, Giannini E, Antonacci C, Pistolese CA, Spagnoli LG, Bonanno E.

O
Microcalcifications in breast cancer: an active phenomenon mediated by
epithelial cells with mesenchymal characteristics. BMC Cancer. 2014 Apr

R
23;14:286. PMID: 24758513
5. Choi S, Coonrod S, Estroff L, Fischbach C.Chemical and physical properties

P
of carbonated hydroxyapatite affect breast cancer cell behavior. Acta
Biomater. 2015 Sep;24:333-42. PMID: 26072364
6. Weerakkody Y, Kruger G et al. http://radiopaedia.org/articles/dr-albert-

D
salomon-1. June 2016
E
7. Gold RH, Bassett LW, Widoff BE. Highlights from the History of
Mammography. RadioGraphics 1990 Nov;10:1111-1131. PMID: 2259767
T
8. Sickles EA. Breast calcifications: mammographic evaluation.
Radiology. 1986 Aug;160(2):289-93. PMID: 3726103
C

9. Web content: Cancer Research UK,

http://www.cancerresearchuk.org/health-professional/cancer-
E

statistics/statistics-by-cancer-type/breast-cancer/incidence-in-
situ#heading-Twoge, Accessed June 2016.
R

10. Weigel S, Decker T, Korsching E, Hungermann D, Böcker W, Heindel W.

Calcifications in Digital Mammographic Screening: Improvement of Early
R

Detection of Invasive Breast Cancers? Radiology 2010 255:3, 738-745.

PMID: 20501713
O

11. Glynn CG, Farria DM, Monsees BS, Salcman JT, Wiele KN, Hildebolt CF.
Radiology. Effect of transition to digital mammography on clinical
C

outcomes. 2011 Sep;260(3):664-70. PMID: 21788529

12. Bluekens AM, Holland R, Karssemeijer N, Broeders MJ, den Heeten GJ.
N

Comparison of Digital Screening Mammography and Screen-Film

Mammography in the Early Detection of Clinically Relevant Cancers: A
U

Multicenter Study. Radiology 2012 265:3, 707-714. PMID: 23033499

13. Farshid G, Sullivan T, Jones S, Roder D. Performance indices of
JR

needle biopsy procedures for the assessment of screen detected

abnormalities in services accredited by BreastScreen Australia. Asian Pac
J Cancer Prev. 2014;15(24):10665-73. PMID: 25605157
14. Gilbert FJ, Astley SM, McGee MA, Gillan MG, Boggis CR, Griffiths PM, Duffy
B

SW.Single reading with computer-aided detection and double reading

of screening mammograms in the United Kingdom National Breast
Screening Program. Radiology. 2006 Oct;241(1):47-53. PMID: 16990670
15. Spangler ML, Zuley ML, Sumkin JH, Abrams G, Ganott MA, Hakim C, Perrin
R, Chough DM, Shah R, Gur D. Detection and Classification of Calcifications

12
 Microcalcification on Mammography

on Digital Breast Tomosynthesis and 2D Digital Mammography: A

Comparison. American Journal of Roentgenology 2011 196:2, 320-324.
PMID: 21257882
16. Bolmgren J, Jacobson B, Nordenström B.
Stereotaxic instrument for needle biopsy of the mamma. AJR Am J
Roentgenol. 1977 Jul;129(1):121-5. PMID: 409122

FS
17. https://www.spacefoundation.org/programs/space-technology-hall-
fame/inducted-technologies/stereotactic-breast-biopsy-technology.
Accessed June 2016

O
18. Practical evaluation of Hologic Affirm digital breast tomosynthesis biopsy
system NHS Breast Screening Programme Equipment Report 1501.

O
https://www.gov.uk/government/uploads/system/uploads/attachment_
data/file/488948/Practical_evaluation_of_Hologic_tomosynthesis_biopsy

R
_system_FINAL_291215.pdf
19. Non-operative Diagnosis Subgroup of the National Coordinating Group for

P
Breast Screening Pathology. Guidelines for non-operative diagnostic
procedures and reporting in breast cancer screening. NHSBSP Publication
No 50 .June 2001. https://www.gov.uk/government/publications/nhs-

D
breast-screening-non-operative-diagnostic-procedures
20. Parker SH, Burbank F, Jackman RJ, Aucreman CJ, Cardenosa G, Cink
E
TM, Coscia JL Jr, Eklund GW, Evans WP 3rd,Garver PR, et al. Percutaneous
large-core breast biopsy: a multi-institutional study. Radiology. 1994
T
Nov;193(2):359-64. PMID: 7972743
C

21. Park HL, Hong J. Vacuum-assisted breast biopsy for breast cancer. Gland
Surg. 2014 May;3(2):120-7. PMID: 25083505
E

22. Schrading S, Distelmaier M, Dirrichs T, Detering S, Brolund L, Strobel

K, Kuhl CK. Digital breast tomosynthesis-guided vacuum-assisted breast
R

biopsy: initial experiences and comparison with prone stereotactic

vacuum-assisted biopsy. Radiology. 2015 Mar;274(3):654-62. PMID:
R

25386875
23. Sickles, EA, D’Orsi CJ, Bassett LW, et al. ACR BI-RADS® Mammography. In:
O

ACR BI-RADS® Atlas, Breast Imaging Reporting and Data System. Reston,
VA, American College of Radiology; 2013.
C

24. Maxwell AJ, Ridley NT, Rubin G, Wallis MG, Gilbert FJ, Michell MJ; Royal
College of Radiologists Breast Group. The Royal College of Radiologists
N

Breast Group breast imaging classification. Clin Radiol. 2009

Jun;64(6):624-7. PMID:19414086
U

25. Evans A. The diagnosis and management of pre-invasive breast disease:

radiological diagnosis Review Article. Breast Cancer Res. 2003. PMCID:
PMC314425.
JR

26. Stehouwer BL, Merckel LG, Verkooijen HM, Peters NH, Mann RM, Duvivier
KM, et al. 3-T breast magnetic resonance imaging in patients with
suspicious microcalcifications on mammography. Eur Radiol. 2014
B

March; 24(3): 603–609. PMID: 24078055

27. Bae S, Yoon JH, Moon HJ, Kim MJ, Kim EK.Breast Microcalcifications:
Diagnostic Outcomes According to Image-Guided Biopsy Method. Korean J
Radiol. 2015 Sep-Oct;16(5):996-1005. PMID: 26357494

13
 Microcalcification on Mammography

28. Kim TE, Kim DB, Jung JH, Lee EK. Sonographic Visibility and Feasibility of
Biopsy under Ultrasound Guidance of Suspicious Microcalcification-only
Breast Lesions: a Single-centre Study TE Kim, DB Kim, JH Jung, EK Lee.
Hong Kong J Radiol. 2015;18:125-33
29. Bagnall MJ, Evans AJ, Wilson AR, Burrell H, Pinder SE, Ellis IO. When have
mammographic calcifications been adequately sampled at needle core
biopsy? Clin Radiol. 2000 Jul;55(7):548-53. PMID: 10924380

FS
30. Houssami N, Ciatto S, Ellis I, Ambrogetti D. Underestimation of
Malignancy of Breast Core-Needle Biopsy Concepts and Precise Overall
and Category-Specific Estimates Cancer. 2007 Feb 1;109(3):487-95.

O
PMID: 17186530
31. Soo AE, Shelby RA, Miller LS, Balmadrid MH, Johnson KS, Wren AA, et al.

O
Predictors of Pain Experienced by Women During Percutaneous Imaging-
Guided Breast Biopsies. Journal of the American College of Radiology.

R
July; 11(7):709 – 716. PMID: 24993536
32. Members of the National Coordinating Committee for Breast Pathology,

P
Professor Ian Ellis, Nottingham City Hospital (Writing Group Lead).
Tissue pathways for breast pathology. September 2010. The Royal College
of Pathologists.

D
33. Pathology Reporting of Breast Disease: A Joint Document Incorporating
E
the Third Edition of the NHS Breast Screening Programme’s Guidelines
for Pathology Reporting in Breast Cancer Screening and the Second
T
Edition of The Royal College of Pathologists’ Minimum Dataset for Breast
Cancer Histopathology NHSBSP Publication No 58 January 2005.
C

https://www.gov.uk/government/uploads/system/uploads/attachment_
data/file/465530/nhsbsp58-high-resolution.pdf
E

34. Marmot, MG, Altman, DG, Cameron, DA, Dewar, JA, Thompson, SG, Wilcox,
M., & The Independent UK Panel on Breast Cancer Screening. (2013). The
R

benefits and harms of breast cancer screening: an independent review: A

report jointly commissioned by Cancer Research UK and the Department
R

of Health (England) October British Journal of Cancer, 108(11), 2205–

2240. PMCID: PMC36934502012.
O

35. Farshid G, Sullivan T, Downey P, Gill PG, Pieterse S. Independent

predictors of breast malignancy in screen-detected microcalcifications:
C

biopsy results in 2545 cases. Br J Cancer. 2011 Nov 22;105(11):1669-75.

PMID: 22052156
N

36. NHS breast screening programme and Association of Breast Surgery. An

audit of screen detected breast cancers for the year of screening April
U

2014 to March 2015. Public Health England. Published May 2016.

http://www.associationofbreastsurgery.org.uk/media/63035/nhsbsp_ab
JR

s_breast_screening_audit_201415_full_audit_v3.pdf
37. Duffy SW, Dibden A, Michalopoulos D, Offman J, Parmar D, et al. Screen
detection of ductal carcinoma in situ and subsequent incidence of invasive
interval breast cancers: a retrospective population-based study. Lancet
B

Oncol. 2016 Jan;17(1):109-14. PMID: 26655422

38. Hofvind S1, Geller BM, Rosenberg RD, Skaane P. Screening-
detected breast cancers: discordant independent double reading in a
population-based screening program. Radiology. 2009 Dec;253(3):652-
60. PMID: 19789229

14
 Microcalcification on Mammography

39. Hoff SR, Abrahamsen AL, Samset JH, Vigeland E, Klepp O, Hofvind S.
Breast Cancer: Missed Interval and Screening-detected Cancer at Full-
Field Digital Mammography and Screen-Film Mammography — Results
from a Retrospective Review. Radiology 2012 264:2, 378-386. PMID:
22700555.
40. Warren R, Allgood P, Hunnam G, Godward S, Duffy S; East Anglian Breast
Screening Programme. An audit of assessment procedures in women who

FS
develop breast cancer after a negative result. J Med Screen.
2004;11(4):180-6. PMID: 15624238
41. Evans AJ, Wilson ARM, Burrell HC, Ellis IO, Pinder SE. Mammographic

O
features of ductal carcinoma in situ (DCIS) present on previous
mammography. Clin Radiol. 1999;54:644–649. PMID: 10541387

O
42. LORIS: A Phase III Trial of Surgery versus Active Monitoring for Low Risk
Ductal Carcinoma in Situ (DCIS).

R
http://www.birmingham.ac.uk/research/activity/mds/trials/crctu/trials
/loris/index.aspx Accessed June 2016

P
43. The Sloane Project. http://www.sloaneproject.co.uk/ Accessed June 2016
44. Evans A, Clements K, Maxwell A, Bishop H, Hanby A, Lawrence G, Pinder
SE. Lesion size is a major determinant of the mammographic features of

D
ductal carcinoma in situ: findings from the Sloane Project. Clinical
E
Radiology 2010; 65: 181-184. PMID: 20152272
45. Liston J, Wilson R, Robin Wilson. Clinical Guidelines for Breast Cancer
T
Screening Assessment, Third edition NHSBSP Publication No 49 June
2010.
C

https://www.gov.uk/government/uploads/system/uploads/attachment_
data/file/465528/nhsbsp49_June2010.pdf
E

46. National Safety Standards for Invasive Procedures (NatSSIPs). Version

number: 1 First published: 7 September 2015. NHS England Patient Safety
R

Domain and the National Safety Standards for Invasive Procedures Group.
https://www.england.nhs.uk/patientsafety/wp-
R

content/uploads/sites/32/2015/09/natssips-safety-standards.pdf
47. The Royal College of Radiologists. Standards for patient consent
O

particular to radiology, Second Edition, The Royal College of Radiologists,

2012. https://www.rcr.ac.uk/standards-patient-consent-particular-
C

radiology-second-edition
48. NHS Breast Screening: Helping you decide. Published by NHS Cancer
N

Screening Programmes. June 2013

https://www.gov.uk/government/uploads/system/uploads/attachment_
U

data/file/440798/nhsbsp.pdf Accessed June 2016

49. Miller SJ, Sohl SJ, Schnur JB, Margolies L, Bolno J, Szabo J, Hermann
JR

G, Montgomery GH. Pre-biopsy psychological factors

predict patient biopsy experience. 2014 Feb;21(1):144-8. PMID:
23065421
50. BBC iWonder. Why isn't breast cancer screening totally reliable?
B

http://www.bbc.co.uk/guides/zcq7xnb Accessed June 2016

51. Loughran CF, Keeling CB, BA. Seeding of tumour cells following breast
biopsy: a literature review. Br J Radiol. 2011 Oct; 84(1006): 869–874.
PMCID: PMC3473763

15
 Microcalcification on Mammography

Table 1. Published data on investigation of microcalcifications in population

screening programmes

Germany United States Netherlends Australia

(Wiegel (Glynn 201111) (Bluekens 201212) (Farshid
201010) 201413)

FS
Screening Biennial not stated ? annual Biennial Biennial
interval
Modality Digital Analogue Digital Analogue Digital Not
(year 1 stated

O
and 2)
Number 24 067 32 600 19 282 1 045 152 1494 809

O
screened* 978 515
Recall rate 7.5% 6.0% 8.5% 1.5% 2.4% 4.6%

R
Cancer 1.0% 0.33% 0.55% 0.52% 0.6% 0.52%
detection rate

P
Assessment data for calcifications **:
Recall for 1.7% 0.79% 1.82% 0.20% 0.67% 0.42%
calcifications had

D
biopsy
E for calc
DCIS from 0.20% 0.04% 0.09%
calcifications
T
% of women 0.32% 0.12% 0.20% 0.15%
diagnosed
C

with
malignancy
E

from
calcifications
R

PPV of biopsy 36% 41.1% 22.6% 35.8%

of
R

calcification
* proportion of initial and subsequent attenders may differ between cohorts
O

**rates are estimated from numbers of lesions/cancers and numbers of women

screened
C
N
U
JR
B

16
 Microcalcification on Mammography

Table 2. The RCR Breast Group Classification for Breast Imaging.24

Mammographic Description
grade
M1 Normal

FS
M2 Benign
M3 Indeterminate/probably benign
M4 Suspicious of Malignancy
M5 Highly Suspicious of Malignancy

O
O
R
P
D
E
T
C
E
R
R
O
C
N
U
JR
B

17
 Microcalcification on Mammography

Table 3
Pathology identified on percutaneous breast biopsy
Benign Proliferative change (B2)
 Fibroadenoma
 Fibrocystic change

FS
 Sclerosing adenosis
 Columnar cell change

O
Indeterminate lesions (B3)
 Atypical epithelial ductal proliferation (AEDIP )

O
 In situ lobular neoplasia, including lobular carcinoma in situ and

R
atypical lobular hyperplasia (ILN)
 Papilloma

P
 Radial scar
 Mucinous lesions

D
Non-invasive cancer (B5a) E
 Ductal carcinoma in situ and intracystic carcinoma
T
 Pleomorphic lobular carcinoma in situ
C

Cancer which extends less than 1mm outside the duct wall is classified as
microinvasive (B5c)
E
R

Invasive cancer (B5b)

 Invasive ductal carcinoma
R

 Invasive lobular carcinoma

O

 Special type including papillary, tubular and mucinous carcinomas

C
N
U
JR
B

18
 Microcalcification on Mammography

Table 4. Data for women assessed in South West London Breast Screening
Service between April 2013 and March 2016 (from NBSS* assessment report)
*NBSS: National Breast Screening System

FS
 7443 women recalled for assessment
 4338 biopsies performed
 1824 (42% of biopsies) for microcalcification

O
 69 (3.8%) of biopsies were repeated for non-concordance (B1)


O
12 (0.65%) biopsies were repeated after B4 diagnosis
 0 cancers identified arising from an area previously assessed for calcification

R
Final B1 B2 B3 B4 B5a/c B5b No

P
pre- (no (benign and (indeterminate (suspicious (in situ/ (invasive biopsy
operative calcification) concordant) pathology) for microinvasive cancer)
diagnosis malignancy) cancer)

D
Number 45 1212 152 5 360 50 332
of lesions
sampled
% of 2.5% 66.5% 8.3%
E
0.3% 19.7% 2.7%
T
total
biopsies
C
E
R
R
O
C
N
U
JR
B

19
Figure Legends

Microcalcification on Mammography

1 Figure 1a. Patterns of calcification associated with benign change.

2
3 Figure 1b. Examples of calcification associated with benign change
4 Figure 1c. Examples of malignant calcification
5
6
7

FS
8
9
10
11

O
12
13

O
14
15
16

R
17
18

P
19
20
21
22

D
23
24
25
26
E
T
27
28
C

29
30
31
E

32
33
R

34
35
R

36
37
38
O

39
40
C

41
42
43
N

44
45
U

46
47
48
49
JR

50
51
52
53
B

54
55
56
57
58
59
60
61
62
63
64
65
Figure 1a Click here to download Figure fig 1a.tiff

FS
O
O
R
P
D
E
T
C
E
R
R
O
C
N
U
JR
B
Figure 1b Click here to download Figure fig1b.tiff

FS
O
O
R
P
D
E
T
C
E
R
R
O
C
N
U
JR
B
Figure 2 Click here to download Figure fig 2.tiff

FS
O
O
R
P
D
E
T
C
E
R
R
O
C
N
U
JR
B