CHAPTER 4 

Aberrations of normal development and

involution (ANDI): a concept of benign breast
disorders based on pathogenesis

Key points and new developments

1. Terminology in benign breast disease has been confused by a multiplicity of terms which do not relate accurately to clinical or
histological patterns, and which are not based on sound concepts of pathogenesis.
2. Most benign breast disorders derive from minor aberrations of the normal processes of development, cyclical activity and
involution.
3. The ANDI classification allows precise definition of an individual patient problem in terms of pathogenesis, histology and
clinical significance.
4. ANDI replaces the conventional view of ‘normal’ and ‘disease’ with a spectrum ranging from normal, through slight
abnormality (aberration), to disease.
5. Recent developments in molecular biology give support to, and provide possible mechanisms for, the concept.

Introduction framework, with a management policy appropriate to her

individual status.
The situation with benign conditions has been dif-
The aberrations of normal development and involution ferent until recently. A large number of clinical and
(ANDI) classification of benign breast disorders1 provides histological conditions, such as fibroadenoma, duct
an overall framework for benign conditions of the breast, papilloma, subareolar abscess and nipple discharge
encompassing both pathogenesis and degree of abnor- right through to the ubiquitous fibrocystic disease have
mality. It was developed because the concepts used in been seen as individual and unrelated entities in terms
teaching about, and managing, benign conditions of the of pathogenesis and management. Alternatively, some
breast (particularly ‘fibrocystic disease’) have been bedev- workers equated the whole range of benign breast
illed by obscurities and inaccuracy. This is in marked disease with fibrocystic disease and tried to push all its
contrast with breast cancer, where there is a clearly defined manifestations into this one ill-defined complex, with
framework. A patient with cancer is investigated along even greater confusion. There was no overall framework
two directions – a ‘longitudinal’ direction which will for benign disorders into which an individual condition
assess the tumour in temporal terms, in situ or invasive, could be placed for assessment and management. Thus,
size and extent of primary tumour, presence or absence historically, patients with breast cancer have been
of lymph node involvement and systemic spread; and a managed more logically, consistently, comprehensively
‘horizontal’ direction assessing biology, histological type and with more conviction than those with benign
(e.g. lobular, ductal or other), grading, hormone recep- breast disease.
tor/growth factor status, etc. This will allow a patient to The ANDI classification, also a bidirectional frame-
be placed within this well-recognized and well-defined work, provides a means to reverse this disparity (Table

41

Descargado de ClinicalKey.es desde Univ Antioquia agosto 17, 2016.

Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2016. Elsevier Inc. Todos los derechos reservados.
 Benign disorders and diseases of the breast

4.1), and is based on the fact that most benign breast The basic principles underlying the ANDI classification
conditions arise from normal physiological processes. are set out in Table 4.3.
The horizontal assessment defines the position along Since the ANDI concept was first proposed in 1979,
a spectrum from normality, through mild abnormality and published in 1987, a great deal of new information,
(‘aberration’) through to severe abnormality (‘disease’). relating both to physiology and to pathology, has come
The vertical component defines the pathogenesis of the forward, providing a surprising degree of support to those
condition, for almost all conditions are related to the elements of the concept which were speculative at that
three different phases of activity in the breast during time. The ANDI classification was accepted and recom-
reproductive life. Together, the two provide a comprehen- mended by an international multidisciplinary working
sive framework, into which can be fitted most aspects of party in 1992.2
benign breast disorders, in terms of concept, pathogene- The ANDI framework is consonant with and builds on
sis and severity (Table 4.2). work of many earlier workers, extending back as far as
1922.3 Yet its postulates are in total contrast to the widely
accepted concepts of fibrocystic disease, and hence it is
necessary to present the background, rationale and sup-
Table 4.1  The basic bidirectional framework of the ANDI
porting material in some detail.
classification

Horizontal – Spectrum of
severity
Table 4.3  The principles underlying the ANDI concept
Normal Aberration Disease
1 Most benign disorders are related to normal
Vertical – Pathogenesis processes of reproductive life.
based on reproductive 2 There is a spectrum that ranges from normal to
period aberration, and occasionally to disease.
Development (15–25 years) 3 The definition of normal and abnormal is pragmatic.
Cyclical activity (25–45 years) 4 The ANDI concept embraces all aspects: symptoms,
Involution (35–55 years) signs, histology and physiology.

Table 4.2  Classification of the more important conditions of BBD into ANDI and non-ANDI

Stage Main clinical presentations

Normal process Aberration Disease

Early reproductive Lobular development Fibroadenoma Giant fibroadenoma

(15–25 years) Stromal development Adolescent hypertrophy Gigantomastia
Nipple eversion Nipple inversion Subareolar abscess/mammary duct fistula
Mature reproductive Cyclical changes of menstruation Cyclical mastalgia Incapacitating mastalgia
(25–40 years) Nodularity
Epithelial hyperplasia of pregnancy Bloody nipple discharge
Involution Lobular involution Macrocysts
(35–55 years) Sclerosing lesions
Duct involution Duct ectasia Periductal mastitis/abscess
  – dilatation Nipple retraction
  – sclerosis
Epithelial turnover Simple epithelial With atypia
hyperplasia
Non-ANDI Conditions of well-defined aetiology, such as fat necrosis, lactational abscess, etc., together with extrinsic
precipitating factors such as smoking and oro-nipple contact in severe nonpuerperal abscess.

42

Descargado de ClinicalKey.es desde Univ Antioquia agosto 17, 2016.

Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2016. Elsevier Inc. Todos los derechos reservados.
 Aberrations of normal development and involution 4

Recognition of the normality of much recognized as part of a spectrum extending from normal-
ity and probably arising from minor hormonal abnor-
benign breast ‘disease’ malities, no specific aetiological factors have been
identified, except for those of altered prolactin secretion
In the 1950s, a number of seminal studies demonstrated and possibly changes in fatty acid intake. Nor is there any
that the histological changes of fibrocystic disease are good evidence as to whether the condition has always
widely distributed in patients who had not claimed to occurred with the same frequency. One study of women
be symptomatic or demonstrated overt disease. Parks4 of all age groups showed fewer histological changes of
showed, from studies of both surgical and autopsy speci- ANDI than were reported in the 1950s and 1960s.6 While
mens, a complete gradation between normal histology microcysts were seen in most breasts, and duct ectasia
and disease; for example, between developing lobules in one-third, both conditions showed a uniform age
and fibroadenomas, and between involuting lobules and spread. In contrast, epithelial hyperplasia and sclerosed
macroscopic cysts. He also showed that epithelial hyper- lobules were uncommon, and no case of premalignant
plasia is so common around the menopause as to be hyperplasia was seen. Only further work will show
normal and that these lesions, so often in the past regarded whether this is a genuine epidemiological trend, since
as indicative of cancer risk, could regress without population (autopsy) studies common 40 years ago have
treatment. in general been replaced by more detailed study and
Extensive autopsy studies over the past 50 years have follow-up of biopsy material. There is some evidence of
shown that most of the benign changes previously con- changing incidence patterns in non-Western populations
sidered as disease are so common that they must be as social and dietary conditions change, as discussed in
regarded as lying within the spectrum of normality. For Chapter 7.
example, a study of 200 breasts at autopsy from 100
postmenopausal patients (mean age 62 years) without
clinical breast disease, confirmed earlier studies and Problems with the conventional view
showed changes of ‘fibrocystic condition’ (normal to of benign breast disease
aberration) in 54% of cases, with only 46% histologically
normal. Hyperplasia with severe atypia (disease) was seen
When problems are defined, solutions are more easily
in only 3% of patients.5
found. Even just recognizing the problems goes a long
Mastalgia provides an example of horizontal assess-
way towards resolving them. We see four main problems
ment. Studies in Cardiff, UK, over the past 20 years have
in the classical approach to benign breast disease:
shown the high incidence of painful nodularity: two-
thirds of a population of working women will experience • Nomenclature
mastalgia. The clinical significance of mastalgia was quan- • The borderline between normal and abnormal
tified in the Cardiff Mastalgia Clinic through accurate • Correlation of clinical symptoms and signs with
classification and assessment of impact on patients’ lives histological changes
by visual linear analogue scales and breast pain charts. • The assessment of premalignant potential.
This has demonstrated that much breast pain can be con-
sidered as ‘normal’, some is sufficiently troublesome to
Nomenclature
warrant attention and can be regarded as a ‘disorder’, but
in a minority pain is sufficiently severe as to be a major As discussed in Chapter 1, the situation has been beset
interference with quality of life and to be regarded as by multiple loose terminology. Each worker in the past
disease. has tended to introduce his or her own terminology,
Individual conditions can also be put into a longitudi- reflecting personal ideas of disease and its underlying
nal grouping related to the three main phases of breast pathology, without linking it to earlier studies. At first,
activity during reproductive life, since these conditions the terms were mainly clinical, e.g. ‘chronic mastitis’, but
can be recognized as arising from aberrations of normal later reflected biopsy appearance, e.g. ‘fibrocystic disease’,
processes within the breast. of symptomatic lesions without appreciating the range of
While the changes of ANDI, both clinical (painful histological appearance in patients without symptoms.
nodularity) and histological (‘fibroadenosis’), are now This leads on to the second problem.

43

Descargado de ClinicalKey.es desde Univ Antioquia agosto 17, 2016.

Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2016. Elsevier Inc. Todos los derechos reservados.
 Benign disorders and diseases of the breast

The borderline between normal and criteria because there is usually a spectrum of severity.
abnormal – what determines normality However, it is possible, using considerations such as
these, to assess where an individual patient lies along this
or abnormality?
spectrum, so that an appropriate management policy can
Many organs under endocrine control show a wide range be determined.
of appearances associated with cyclical or pulsed hormo-
nal secretion. This is especially true in females, where
Correlation of clinical symptoms and signs
cyclical changes are set against a background of the
broader changes of development and involution at
with histological changes
the extremes of reproductive life, and particularly so in In the past, a patient with a local clinical abnormality
the breast. Most of these cyclical changes show a spectrum such as an area of painful nodularity has been subjected
which on occasions may extend outside the normal range. to biopsy and the histological changes correlated with
Clearly, it is important to try to define the point, even that clinical nodule. This ignores the fact that the classic
if blurred, at which normality crosses the line to changes of ‘fibrocystic disease’, including fibrosis, adeno-
abnormality. sis, cyst formation and lymphocytic infiltration, may be
A number of factors must be assessed in deciding where seen in the asymptomatic breast, and ignores the dynamic
the boundary of normality lies, including incidence, clini- changes within the breast from month to month. Thus, a
cal impact and histology. Macrocysts are an example of second biopsy from a clinically identical area a few cen-
the importance of incidence; they are common and com- timetres away might show different histological changes,
monly multiple, while microcysts are found in almost all as might a biopsy a few months later. One study provides
breasts if looked for carefully enough. Clearly, microcysts interesting confirmation of this heterogeneity of response
must be regarded as normal, and macrocysts – with an within different areas of the breast. Normal volunteers
incidence of perhaps 10% of all women – are at most an given defined oral contraceptives showed diverse histo-
aberration of normality and cannot be considered as logical changes on biopsy, often with secretory and invo-
disease. lutional changes coexisting in different parts of the same
The importance of clinical impact can be illustrated by breast.8
mastalgia. A large proportion of women experience pre- The histology of a clinical lesion must be assessed
menstrual discomfort for a few days; this causes little against the broad spectrum of histological change which
interference with quality of life, and can be considered as might be seen within a ‘normal’ breast, symptomatic or
normal. However, when pain persists for 3 out of 4 weeks not. Failure to do so leads to confusion of significance;
of the cycle and is of great severity, it must be looked at painful nodularity, a clinical condition which has no dis-
in a different light. This is quite uncommon, and the tinct radiological or pathological counterpart, has been
clinical impact can be quite severe; hence, it can be called fibroadenosis, and correlated with cancer risk.
regarded as an aberration of normality and, perhaps in Only lesions which are both definite clinical and path-
the most extreme cases, as disease. ological entities, e.g. a macrocyst, can be assessed from
Histological appearances have given particular diffi- both points of view. Terms such as fibroadenosis or fibro-
culty in this area; even minimal degrees of hyperplasia cystic disease are misleading, because they imply that
have been regarded as carrying cancer risk in the past, but these histological patterns are abnormal and that they
recently the high incidence of simple hyperplasias in correlate with clinical conditions. Histological changes
normal women has caused a reassessment of significance. are best described by a general term such as ANDI, which
A number of pathologists, and Page and co-workers in correctly indicates pathogenesis without using mislead-
particular,7 have defined and categorized histological pat- ing specific terms, or by the specific individual histologi-
terns and their significance with considerable precision. cal elements present in the biopsy material.
This has allowed hyperplasias to be placed in three broad
groups: normal (no increased cancer risk), slightly
increased risk (equivalent to aberration) and high cancer
The assessment of premalignant potential
risk. This aspect has understandably dominated the efforts of
Benign breast conditions cannot always be categorized breast surgeon and pathologist alike. Both have been frus-
as definitively normal or abnormal using these three trated by the three factors discussed above and it is not

44

Descargado de ClinicalKey.es desde Univ Antioquia agosto 17, 2016.

Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2016. Elsevier Inc. Todos los derechos reservados.
 Aberrations of normal development and involution 4

surprising that attempts to assess premalignant potential book. We hope readers will persevere with the concepts
of fibrocystic disease have given almost infinitely variable of this chapter; we believe it makes detailed assessment
results from one study to another. Confusion of nomen- and management of individual clinical problems so much
clature is particularly great when discussing the epithelial easier.
hyperplasias. Because of their precancerous association,
the use of different terms in different countries has led to
Hormone-controlled processes of the breast
serious misunderstanding, compounded by failure to
define the borderline between normal and abnormal. Breast development
Only recently has this problem been addressed in a
The premenarchal breast consists of a few ducts only. The
uniform and structured manner, as with the work of Page
striking feature of the perimenarchal development of the
discussed above.9
breast is the addition of lobular structures to the already
Where do the answers to these problems lie? Firstly, in
developing duct system. The lobules develop particularly
defining the range of normality, both in terms of clinical
during early reproductive life at 15–25 years of age. At
symptoms and signs, and of histological appearance. Sec-
first, ‘primitive’ type 1 lobules of Russo and Russo (Ch.
ondly, recognizing that breast problems may be clinical,
3), they are gradually replaced by more mature and less
physiological or histological, and each problem may
active lobules during the cyclical period, and especially
sometimes reflect only one of these three: sometimes
with pregnancy. This explains the frequency of fibroade-
more than one. Thirdly, by providing a comprehensive
noma during early and midreproductive life, for it is a
framework within which individual clinical or histologi-
condition analogous to gross hypertrophy of a lobule.
cal situations can be placed so that they are seen within
Until the age of about 35 years, the luteal phase is also
an overall context. This must allow precise placement of
associated with enhanced acinar sprouting from the
a problem within the overall clinicopathological frame-
ductules.
work of benign breast disease and also encompasses the
A distinctive element of the lobule is its highly special-
decision whether an individual clinical situation lies
ized connective tissue, and the close interaction between
towards the normal or abnormal end of the spectrum of
epithelium and connective tissue separated only by a
a particular process.
basement membrane. This lobular connective tissue is
pale and loose (Fig. 4.1) with mononuclear infiltrate and
differs notably from the much less interesting and urbane
The physiological processes underlying interlobular fibrous stroma (see Ch. 3).
the andi concept

The physiology of the breast has been described in detail

in Chapter 3, but the broad outline is important in under-
standing ANDI. The main processes are related to hormo-
nal effects on the breast during reproductive life, in the
three phases of development, cyclical change and involu-
tion. As these three stages are of increasing complexity, it
is not surprising that the processes go wrong more often
during the long period of cyclical activity than during
development, and even more commonly in the long and
complex process of involution. Consideration of the
impact of these hormonal events on lobules, ducts and
stroma sets the scene for an understanding of the clinical
conditions which arise from them.
Pathogenesis, clinical presentation and management
of benign breast disorders are set out in this chapter only
Fig. 4.1  The perimenarchal breast showing early lobular
in conceptual terms. The details of clinical features, development. The pale, loose lobular connective tissue contrasts
pathology and management are given in the rest of the with the denser interlobular fibrous stroma.

45

Descargado de ClinicalKey.es desde Univ Antioquia agosto 17, 2016.

Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2016. Elsevier Inc. Todos los derechos reservados.
 Benign disorders and diseases of the breast

Cyclical change

Both epithelial and stromal elements of the lobule are

under hormonal control and there is evidence that the two
work in tandem. In fact, normality seems to be very much
dependent on a normal, balanced relationship between
both elements. The details of the interaction of hormones
and growth factors on epithelium, myoepithelium, base-
ment membrane and stroma are now being elucidated (as
discussed in Ch. 3) and give an inkling of the mechanisms
underlying this relationship. It is likely that interference
with these close relationships is responsible for many of
the conditions that are often included under the term Fig. 4.2  Section from a normal asymptomatic breast. It shows a
‘benign breast disease’. The changes occurring with each wide variety of histological appearances, the changes commonly
menstrual cycle have been summarized by Vorherr,10 and ascribed to ‘fibrocystic disease’.
expanded with more recent studies which demonstrate a
peak of mitosis in the late cycle followed by apoptosis (see
Ch. 3). They provide a potent opportunity for minor
upsets to occur with repetitive cyclical changes.
These cyclical changes are associated with clinical
symptoms of heaviness and fullness that are not associ-
ated with consistent histological change, but for which a
hormonal basis is being elucidated through recent studies.
A correlation with prolactin secretion in response to pitu-
itary stimulation, and particularly with bioactive forms of
prolactin,11 brings new supportive evidence not available
until recently. This illustrates the need to consider physi-
ological aspects as equal to, or more important than,
structural changes in assessing benign disorders. Super-
imposed on the cyclical changes are the much more Fig. 4.3  A section of normal pregnant breast, showing extreme,
uniform lobular development.
radical effects of pregnancy and lactation. With the
repeated development and involutional changes of men-
struation and pregnancy occurring throughout 40 years The variability of appearance within the normal breast
of reproductive life, there is abundant opportunity for is illustrated following pregnancy. Under the intense hor-
minor aberrations to occur. monal stimulation associated with pregnancy, a uniform
When one studies a section of normal breast from a pattern of lobular development and maturation is seen
patient who has no breast complaint or overt clinical (Fig. 4.3), but postlactational involution is patchy (Fig.
disease on examination, the striking feature is the wide 4.4).
spectrum of histological appearance. Figure 4.2 is from With such variable involution following the total stim-
an asymptomatic patient in her thirties, and within a ulation of pregnancy, it is not surprising that the more
small area may be seen well-developed lobules, poorly minor cyclical changes with menstruation can, com-
developed lobules, dilated ducts and normal ducts, pounded over a long period, produce marked differences
lobule-deficient fibrous tissue and fatty tissue. in the structure and appearance of various areas of the
These normal appearances provide the elements breast tissue on a purely random basis.
‘fibrosis’ and ‘adenosis’ that have been documented as the
histological appearance of biopsies taken from patients
Breast involution
with nodular breasts. It has not always been appreciated
that the same changes may be evident elsewhere in the Involution starts quite early and changes are obvious by
same breast, where there is no clinical complaint. 35 years of age and often earlier. Thus, cyclical change

46

Descargado de ClinicalKey.es desde Univ Antioquia agosto 17, 2016.

Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2016. Elsevier Inc. Todos los derechos reservados.
 Aberrations of normal development and involution 4

Fig. 4.4  Histological section from a postlactational breast. Fig. 4.6  A section of postmenopausal breast, showing advanced
Involution is patchy, varying from marked to negligible. lobular involution.

involution shown in Figure 4.5 is seen. Eventually, by the

time the menopause has been reached and passed, invo-
lution is extensive (Fig. 4.6) with only a few ducts remain-
ing, and few if any lobular structures.
But it does not always happen in that way, and minor
aberrations of this process are very common during a
period of fluctuating involution extending over 20 years.
It appears that normal epithelial involution of the lobule
is dependent on the continuing presence of the special-
ized stroma around it. Should the stroma disappear too
early, the epithelial acini remain and may form micro-
cysts. The exact mechanism is not known, but one finding
is that the structural protein fodrin (expressed in normal
Fig. 4.5  Section from a normal involuting breast. The orderly breast epithelium) is not expressed in micro- or macro-
regression of both epithelial and stromal elements of the lobule is cysts,12 although it is not certain whether this is a primary
obvious. The lobular stroma has been replaced by fibrous tissue, or secondary phenomenon. Microcysts are obviously a
and there is little residual epithelial tissue.
prime target for macroscopic cyst formation if pressure
disparity occurs between secretion and drainage, as might
and involution run in tandem for 20 years or more, occur with obstruction of the draining ductule. Microcyst
increasing the chance of aberration of normality, as formation is very common in normal breasts (Fig. 4.7),
reflected in the high frequency of presentation to a breast as demonstrated by Parks,4 and, in this process of cystic
clinic during this period. The involution affects the lobules lobular involution, microcysts may appear even though
particularly and is much dependent on the relationship there is still specialized stroma present.
between the epithelium and specialized stroma of the Presumably, this arises from minor obstruction to the
lobule. In Figure 4.5, which shows an involuting breast, duct by kinking or compression from fibrous tissue, or
an orderly regression of lobules and surrounding fibrous perhaps by vigorous secretion from still active epithelial
tissue can be clearly seen. tissue. (It is interesting that the lobular vein exits from
During this process of lobular involution, the loose the lobule alongside the ductule, so venous compression
hormone-responsive intralobular connective tissue is readily occurs.)
replaced by the more standard interlobular type of fibrous Recent work has shown that there is continual turnover
tissue. If this replacement is well coordinated with the of fluid within cysts, but, paradoxically, that particular
regression of epithelial tissue, the uniform picture of molecules (e.g. hormones) may persist in a cyst for

47

Descargado de ClinicalKey.es desde Univ Antioquia agosto 17, 2016.

Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2016. Elsevier Inc. Todos los derechos reservados.
 Benign disorders and diseases of the breast

Fig. 4.7  Involuting breast showing microcystic lobular change Fig. 4.9  In a fully developed macrocyst, the bands in the wall
(thick section technique). reflect the origin from gross distension of a number of acini
(ductules) within a single lobule.

known that the number of ovarian follicles decreases

progressively from the age of 35 years, so there is a pro-
gressive withdrawal of hormonal stimulation over the full
period in which these involutional changes and aberra-
tions occur.
Thus, three periods occur and overlap: lobule develop-
ment at 15–25 years, cyclical change at 15–50 years and
involution at 35–55 years. Each period has its own clini-
cal presentations, but overlapping and interacting
processes also lead to complex clinical situations. The
introduction of hormone replacement therapy has com-
Fig. 4.8  A further stage in the evolution from microcystic plicated the situation by extending these changes beyond
involution to macrocyst formation. the menopause. Thus, benign disorders expected to
resolve at the menopause may persist, or even arise de
months or even years (see Ch. 10). As long as some of novo.
the specialized stroma remains, the lobule can still invo-
lute normally in spite of these microcysts, but should the
A framework based on pathogenesis
specialized stroma disappear early then further cystic
change is likely (Fig. 4.8).
Mechanical duct obstruction leading to macrocysts Table 4.2 sets out a classification of the more important
(Fig. 4.9) is almost bound to occur in a proportion of clinical benign breast disorders that make up the main
lobules, because there are many possible mechanisms: constituents of ANDI. An important point of the classifi-
from internal blockage by epithelial cells or debris, cation is the replacement of the term disease by disorders
through simple kinking and angulation, to strangulation in the interpretation of benign breast disorders (BBDs).
by the surrounding maturing fibrous tissue. This does not mean that there is no benign breast disease,
Recent work on the neonatal breast is of great interest, but recognizes that most breast complaints are due to
since it shows that the involution which occurs in the first disorders based on the normal processes of development,
year of life following withdrawal of maternal hormones cyclical change and involution, and lie towards the
follows an identical pattern to that seen over 20 years of normal or aberration end of the spectrum, with only a
adult involution, even to microcyst formation as the few severe enough to be placed at the disease end. The
stroma reverts to the banal interlobular form. It is now concept that conditions such as fibroadenoma and duct

48

Descargado de ClinicalKey.es desde Univ Antioquia agosto 17, 2016.

Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2016. Elsevier Inc. Todos los derechos reservados.
 Aberrations of normal development and involution 4

ectasia lie within the normal or minor aberration range 4.10) and will involute, to be replaced by hyaline con-
is foreign to conventional teaching in pathology and nective tissue in concert with the rest of the breast in the
surgery. Hence, we give the reason in some detail. perimenopausal period.
These hormonal responses are much more complete
than those usually seen in benign tumours. Rarely, a
Reasons for including various benign fibroadenoma will continue to grow to a size of 3 cm,
breast disorders as part of andi although this is sufficiently common to be regarded as
within the normal spectrum. Growth beyond 5 cm is suf-
Disorders of development ficiently uncommon in Western populations as to justify
being regarded as a disease, known as giant fibroadenoma.
Fibroadenoma Similarly, a multiple fibroadenoma (five or more lesions
Since it can be shown that fibroadenomas arise from in one breast) is so uncommon in Western populations,
lobules, it is not surprising that these are seen predomi- and its implications so uncertain, as also to justify being
nantly in women in the 15–25 age group, even though considered a disease. Thus, fibroadenoma fits well into the
they may not be diagnosed until later, when postpreg- ANDI classification: small fibroadenomas are normal,
nancy or involutional changes facilitate clinical recogni- clinical fibroadenomas are a mild aberration of the normal
tion in the softer, drooping, involutional breast, or processes, and giant and multiple fibroadenomas are
ultrasound demonstrates impalpable lesions. placed to the disease end of the spectrum.
What is the evidence to support the contention that Good evidence can be put forward to support this view
fibroadenoma should be placed in the benign breast dis- that fibroadenoma should not be regarded as a neoplasm,
order side of ANDI rather than regarded as a neoplasm? as discussed in Chapter 7. All the cellular elements of
Parks4 showed that hyperplastic lobules, histologically fibroadenoma are normal on conventional and electron
identical to clinical fibroadenomas, are present so com- microscopy, and epithelium and myoepithelium maintain
monly as to be regarded as normal; they can probably be a normal relationship,13 while molecular biology studies
found in all breasts if they are sought sufficiently care- have shown that fibroadenomas are polyclonal in keeping
fully. A full spectrum can be found between these hyper- with hyperplasia, in contrast to phyllodes tumours which
plastic lobules and clinical fibroadenomas, which do not are monoclonal in keeping with a neoplastic condition.
show the inexorable growth typical of true neoplasms.
They usually grow to 1 or 2 cm in diameter and then stay Adolescent hypertrophy
constant in size. They show hormonal dependence similar
This condition is associated with gross stromal hyperpla-
to normal lobules by lactating during pregnancy (Fig.
sia at the time of breast development. The aetiology is
unknown, and this is not surprising because so little is
known about the control of breast stroma, important
though this is. Nevertheless, it is likely that there is a
hormonal basis to the condition, a view supported by
recent reports that danazol (an antigonadotrophin) may
have a beneficial effect. The continuous spectrum from a
small breast through to massive hyperplasia fits the hori-
zontal element of the ANDI concept, with excessively
large breasts an aberration, and the extreme hyperplasia
of gigantomastia placed at the disease end.

Disorders of cyclical change

Mastalgia and nodularity
Premenstrual enlargement and postmenstrual involution
Fig. 4.10  A fibroadenoma removed in the postpartum period.
It shows lactation similar to the normal breast, indicating that of the breast occurring with each cycle is so commonly
fibroadenomas respond readily to normal physiological stimuli. associated with discomfort and nodularity as to lie firmly

49

Descargado de ClinicalKey.es desde Univ Antioquia agosto 17, 2016.

Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2016. Elsevier Inc. Todos los derechos reservados.
 Benign disorders and diseases of the breast

within the spectrum of normality. We have used the term Cyst formation
‘cyclical pronounced mastalgia’ or ‘severe painful nodu-
The desirable integrated involution of stroma and epithe-
larity’ to differentiate the clinical disorder from the more
lium outlined earlier in this chapter is not always seen,
common physiological discomfort and lumpiness. A
and minor aberrations of the process are understandably
duration of painful nodularity of more than 1 week of
common during a period of fluctuating involution extend-
the cycle is a useful definition for differentiation from
ing over 20 years. The exact mechanism of this involution
normal discomfort, and the severity of the pain can be
is not well understood,16 but it appears that the normal
quantified with a pain chart. This is a more pragmatic
epithelial involution of the lobule is dependent on the
approach than the inaccurate histological concepts of
continuing presence of the specialized stroma around it.
fibrocystic disease, or the concept of ‘non-disease’14 which
If the stroma disappears too early, the epithelial acini
is unhelpful to those unfortunate women who suffer
remain and may form microcysts, setting the pattern for
from its more severe manifestations. While no histologi-
macrocyst development by obstruction of the efferent
cal basis has been defined for these changes, the objection
ductule as discussed above. This concept of the macrocyst
of such women to the concept of non-disease is sup-
as being an involutional aberration (and hence part of
ported by hormone studies,11 which show an underlying
ANDI), rather than a disease, fits in with its common
physiological abnormality demonstrated by excess prol-
occurrence and the fact that it is so frequently multiple
actin release from the pituitary following stimulation of
and subclinical.
the hypothalamic–pituitary axis.
The fact that macrocysts appear to develop in two
These findings stress the importance of taking a broad
directions – apocrine and non-apocrine cysts17 – is some-
view of benign breast disorders, avoiding undue empha-
thing which is as yet poorly understood, but the evidence
sis on non-specific histological changes and giving due
is strong that both develop from a common origin of
attention to significant physiological changes. It is likely
microcystic involution.18
that subtle stromal and epithelial changes accompany
physiological variations in more severe cases of cyclical
nodularity and mastalgia, but more sensitive techniques Sclerosing adenosis
may be necessary to demonstrate them. Oedema, stromal
or lobular, can be demonstrated in the late cycle, but This condition may be considered as an aberration of
at present good evidence correlating this with clinical either the cyclical or the involutional phase of breast
symptoms is lacking. activity because it can show histological changes which
Painful nodularity of the predominantly cyclical phase are both proliferative and involutional. This illustrates
of reproductive life (20–35 years) merges into and over- the complexity on the one hand, but the simplicity of
laps those symptoms which are more typically part of the concept on the other, of regarding these as aberrations of
involutional phase, especially cyst formation and scleros- so many interacting normal processes. Considering the
ing adenosis. While all have been lumped together as complex interrelationship of stromal fibrosis and epithe-
fibrocystic disease (or fibroadenosis) in the past, the lial regression occurring during involution, superimposed
clinical problems and management differ. on cyclical changes of ductal sprouting, it is not surprising
that this complex picture, in which epithelial acini are
strangled and distorted by fibrous tissue, should arise. It
Disorders of involution is surprising that it does not occur more commonly.
Since the process of involution extends over 20 years
of monthly cycles of mitosis and apoptosis, it is not
Duct ectasia and periductal mastitis
surprising that a number of aberrations should arise
involving different elements of the normal breast. As The second major group of benign breast disorders con-
discussed below, the evidence favours an aberration of sists of those associated with duct ectasia and periductal
normal processes for most of these. It is interesting that mastitis. The pathogenesis of duct ectasia is obscure. The
the incidence of these changes is similar in a number classic theory, proposed by Haagensen,19 regards duct
of races with widely differing cancer incidences,15 sup- ectasia (dilated ducts) as being the primary event leading
porting the ‘normal’ view against the ‘precancerous’ to stagnation of secretion, epithelial ulceration and
view. leakage of duct secretions containing chemically irritant

50

Descargado de ClinicalKey.es desde Univ Antioquia agosto 17, 2016.

Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2016. Elsevier Inc. Todos los derechos reservados.
 Aberrations of normal development and involution 4

fatty acids into periductal tissue to give a chemical inflam- found in their autopsy study that 59% of women over the
matory process. This secondary inflammation is then seen age of 70 exhibited some degree of epithelial hyperplasia.
as leading to periductal fibrosis, with subsequent fibrous Sloss et al.22 concluded from their autopsy study that ‘the
contraction and nipple retraction. mere presence of blunt duct adenosis, apocrine epithe-
An alternative theory sees the primary process as peri- lium and intraductal epithelial hyperplasia in the breast
ductal mastitis, perhaps on an autoimmune basis, leading of women is insufficient to warrant such tissue being
to weakening of the muscle layer of the ducts and second- called disease’.
ary dilatation. It is likely that both processes may occur Hence the simple epithelial hyperplasias may be placed
separately or in conjunction, thus explaining the wide firmly within the concept of benign breast disorder.
spectrum of clinical behaviour in this condition. Both However, careful studies by Page et al.7 and Wellings
duct dilatation and duct sclerosis may represent an aber- et al.23 have shown that the other ends of the spectrum –
ration of involution. Periductal fibrosis can occur in the atypical lobular hyperplasia and atypical ductal hyperpla-
absence of duct ectasia or of inflammation20 and proba- sia – particularly as seen in the terminal ductal lobular
bly represents part of the normal involutional process. unit (TDLU), are sufficiently commonly associated with
Duct ectasia is so common in the postmenopausal breast malignancy as to be regarded as associated or premalig-
that it must be regarded as part of the normal ageing nant conditions. Hence, epithelial hyperplasias with
process. marked atypia belong firmly under the column of BBD.
The wide variety of clinical symptoms associated with There is at present insufficient evidence to allow a firm
this condition – nipple discharge, nipple retraction, opinion as to whether conditions placed under the benign
inflammatory masses and abscesses (sterile or bacterial) breast disorder column present a continuous spectrum,
– can best be explained and understood by accepting with the implication that the ‘disorders’ move to ‘dis-
more than one process in the pathogenesis. Duct ectasia eases’. They may be entirely separate processes and cer-
appears to be a simple involutional process in the elderly, tainly, at the present time, no such progression of
and congenital inversion of the nipple is an aberration of hyperplasias to cancer in situ should be assumed; it
nipple eversion during breast development. Both are should be left as an open question.
aberrations of ANDI, which may progress to disease, as
discussed fully in Chapter 11. Nonbacterial periductal
inflammation is multifactorial, due to smoking in some An extension of the concept of
cases, idiopathic and possibly involutional in others. The andi to include most benign
fact that some aspects of this complex condition are
clearly parts of ANDI, some parts probably are not, and
breast disorders?
others are of unknown cause at present, illustrates the
view that the ANDI concept should be utilized where it Since the conceptual thinking prior to ANDI was encom-
is appropriate, but that those aspects where an associa- passed in the chronic mastitis/fibroadenosis/fibrocystic
tion is not obvious can be left outside until the patho- disease attitude to BBD, it is not surprising that ANDI is
genesis has been clearly established. equated sometimes with these conditions and used as a
synonym for painful nodularity. This is restrictive, as it
has been shown that a majority of benign breast disorders
Epithelial hyperplasias
lie within the concept, some clearly do not and some
The third element of the benign breast disorder complex, remain inconclusive.24 Our current view of the classifica-
epithelial hyperplasias, has given rise to most confusion tion of all benign disorders is shown in Table 4.2. Most
and problems in management. Many people would see conditions are regarded as part of ANDI, and the old
this lying firmly on the side of disease rather than disor- term ‘fibroadenosis’ is designated ‘painful nodularity of
der, but a number of studies have shown that this is not ANDI’.
so with simple hyperplasias. Parks4 showed that lobular Some of the less common benign breast disorders also
and intraductal papillary hyperplasia is common in the fit well into the concept of aberrations of normality and
premenopausal period and tends to regress spontane- it is useful to consider the arguments for including them
ously after the menopause, and hence should be regarded within this framework. Briefly, these arguments are as
as an aberration of normal involution. Kramer and Rush21 follows:

51

Descargado de ClinicalKey.es desde Univ Antioquia agosto 17, 2016.

Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2016. Elsevier Inc. Todos los derechos reservados.
 Benign disorders and diseases of the breast

• Nipple inversion: This is an aberration of development Fibroadenoma

of the terminal ducts, preventing the normal
The concept of fibroadenoma being a part of ANDI and
protrusion of ducts and areola.
not a neoplasm has been one of the conceptual changes
• Mammary duct fistula: Nipple inversion predisposes
behind us moving gradually from active to conservative
to terminal duct obstruction, leading to recurrent
management over the past 20 years. The results from
subareolar abscess and mammary duct fistula, the
many centres now justify this approach, as discussed in
usual form of periductal mastitis seen in younger
Chapter 7.
women. Extraneous factors such as smoking and
oro-nipple contact interact with the processes of
ANDI. Inverted nipple
• Epithelial hyperplasia of pregnancy: Marked hyperplasia
This is the failure of development of major ducts, and
of the duct epithelium occurs in pregnancy, and the
treatment short of severing the ducts is unlikely to give a
papillary projections sometimes give rise to bilateral
long-term satisfactory result. Hence, treatment is related
bloody nipple discharge, a condition which is always
to the patient’s view of the cosmetic deformity, and
benign when occurring in pregnancy.
recommended only after recognizing the uncertain
• Benign duct papilloma: This is a common condition
long-term results of minor surgical procedures for this
during the period of cyclical activity and shows
condition, and the consequences of total duct division for
minimal if any malignant potential. It is reasonable
cosmesis and lactation if the defect is to undergo total
to regard it as an aberration of cyclical epithelial
correction.
activity.
Likewise, recurrent subareolar abscess associated with
• Adenosis: For similar reasons it might be considered
nipple inversion will require eradication of the sump-like
logical to extend the concept of aberration of
dilated duct underlying the recurrent infection.
normality to encompass adenosis as a manifestation
of involution.
Cyclical mastalgia and nodularity
Implications for the management It is generally accepted that these manifestations are phys-
of benign breast disorders iological, and that most cases can be treated by adequate
reassurance. But the view that the most severe cases, those
lasting perhaps 2 weeks or more of the menstrual cycle,
It follows from the foregoing that most of the conditions interfere so much with the quality of life as to merit con-
listed under benign breast disorders can be regarded as sideration as disease is confirmed by the high incidence
minor aberrations of normality and hence do not demand of hormonal abnormalities in these patients. It is unhelp-
active specific treatment. This being the case, any active ful to tell the patient that she has a non-disease if her
management of these conditions should be based on symptoms interfere with her quality of life. Rather, the
considerations such as accurate diagnosis, patient concern severity of her mastalgia should be assessed objectively
and interference with quality of life. No treatment is with pain charts so that her condition can be placed on
required based solely on inherent pathological signifi- the normal–aberration–disease scale and appropriate
cance until the disease end of the spectrum is reached. management instituted. There is equally strong evidence
This concept of management is outlined here and details that most cases are not psychologically based, and hence
of management are given in the appropriate chapters. endocrine-related treatment is appropriate in those cases
severe enough to warrant therapy.25

Adolescent hypertrophy
Cysts
In its more severe forms, treatment is indicated because
of psychological and physical morbidity from the size Macrocysts and microcysts are so common as to need no
and weight of the breast. It is the degree of this morbidity active treatment other than that required to allow diag-
that determines treatment with hormonal or surgical nosis and allay patient concern. It is now well established
therapy. by practice that simple cysts are satisfactorily treated by

52

Descargado de ClinicalKey.es desde Univ Antioquia agosto 17, 2016.

Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2016. Elsevier Inc. Todos los derechos reservados.
 Aberrations of normal development and involution 4

aspiration, a policy that might have been predicted from Epithelial hyperplasias
the ANDI concept. The recent differentiation of cysts into
apocrine and non-apocrine17 does not at present alter this Epithelial hyperplasias without atypia usually found as a
conservative therapeutic approach, because any breast chance histological finding fall somewhere between nor-
cancer risk is a general one, and not related to the indi- mality and an aberration in significance. With atypical
vidual cyst. Likewise, since the condition is a minor aber- hyperplasias, the emphasis moves towards the disease
ration of a normal process, excision is not required for state, and special consideration should be given to assess-
multiple or recurrent cysts, except in the presence of a ment of cancer risk, as defined particularly by Page and
bloody aspirate or a residual lump, both requiring exclu- Dupont.9
sion of coexisting cancer.
Aberration to disease?
Sclerosing adenosis
The ANDI concept is based on the progression from
This condition causes diagnostic problems to surgeons, normal to aberration to disease. Sound evidence for the
radiologists and histopathologists and sometimes prob- progression from normal to aberration exists, as discussed
lems to the patient in the form of a lump or pain. It in the section on fibroadenoma. However, no evidence
requires no treatment other than careful exclusion of exists for or against the supposition that giant fibroadeno-
cancer, although it sometimes requires symptomatic mas arise from the continued progression of small
treatment for pain. fibroadenomas, i.e. that aberration progresses to disease.
It is possible that giant fibroadenoma is a separate condi-
tion de novo, or an added factor may lead to progression
Postmenopausal nipple retraction from a ‘standard’ fibroadenoma. There is certainly evi-
The only importance of this is to recognize that it may be dence from molecular biology that a change from poly-
caused by the simple involutional process of periductal clonality occurs with phyllodes tumours, although it is
fibrosis. It requires no active management other than not clear whether this is a secondary change in a fibroad-
exclusion of cancer. In our experience, nipple retraction enoma, or is present in a phyllodes tumour de novo (see
is more commonly due to ductal fibrosis than to cancer, Ch. 7).
and fortunately the diagnosis is easily made in older Likewise, it is likely that normal epithelium and apo-
patients because the postmenopausal breast lends itself crine changes form a continuum with simple hyperpla-
to accurate mammography. sias, but there is less evidence to support the direct
progression from the latter to atypical hyperplasias or
carcinoma in situ. Thus, on the basis of present evidence
we do not know whether disease in this case is a progres-
Duct ectasia/periductal mastitis complex
sion from aberration or whether disease and aberration
The symptoms of this condition that can be categorized are two separate conditions.
under the aberration column cause little clinical upset This is one of the most interesting areas of benign
and require no therapy other than reassurance for opaque, breast disorders, and new evidence is starting to provide
non-bloody discharge. answers to some of these questions. For instance, simple
Rarely, the discharge may be sufficiently profuse to fibroadenoma is classified as an aberration, but multiple
cause social embarrassment. In this case, after a pituitary fibroadenomas are sufficiently rare and troublesome to
adenoma has been excluded, the patient may be treated be classified as disease. It has recently been found that
in mechanical fashion by a total duct excision. This transplant patients on ciclosporin have an increased inci-
approach is rarely necessary, but it again demonstrates dence of multiple fibroadenomas, giving a new insight
the value of assessing the impact of symptoms on quality into progression from aberration to disease. Cigarette
of life to place them appropriately along the spectrum of smoking has a similar significance for duct ectasia/peri-
severity. ductal mastitis, and it is likely that the various factors that
Management of blood-related discharge is directed lead to atypical hyperplasias will become elucidated in
towards excluding more serious pathology. the near future.

53

Descargado de ClinicalKey.es desde Univ Antioquia agosto 17, 2016.

Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2016. Elsevier Inc. Todos los derechos reservados.
 Benign disorders and diseases of the breast

The time is approaching when we are likely to have Cyst formation and duct ectasia
much greater insight into these questions. Meanwhile,
this does not compromise the utility of the ANDI classi- These processes have largely eluded research efforts into
fication in helping understand benign breast disorders, pathogenesis, although the absence of fodrin in the wall
and the concepts are sufficiently flexible to allow us to of cysts may be important.12 Animal experiments are noto-
incorporate new information as it becomes available. The riously irrelevant to human breast disease, although the
past 10 years have seen the concept strengthened and action of keratinocyte growth factor (KGF-7) commands
refined by new knowledge. some interest in such an otherwise sterile field. Yi et al.30
have shown that this stromal cell-derived paracrine medi-
ator for epithelial proliferation causes hyperplastic
changes in ducts in rats similar to ANDI. In mice it causes
Recent developments having a bearing dilatation of ducts along much of their length (duct
on the andi concept ectasia?), and when given together with exogenous oestro-
gen and progesterone, it produces numerous end-buds
Apocrine metaplasia with a picture resembling the histological changes of
Apocrine metaplasia has a particular interest in that it is ANDI.
a frequent finding in ANDI, suggesting normality, yet
Fibroadenoma and hyperplasias
carrying a slight but definite increase in cancer risk. For
that reason, there is much interest in possible mecha- It seems likely that an imbalance between cell prolifera-
nisms for indicating cancer risk, and which might throw tion and apoptosis may be involved in the development
light on the evolution from normal to aberration to of ANDI conditions, as breast tissues under hormonal
disease. control undergo continuing remodelling, involving a
Haagensen26 suggested three possible mechanisms by balance between quiescence, proliferation and apoptosis.
which apocrine metaplasia might relate to cancer: Evidence supporting this is becoming available: Ferrieres
et al.31 found that bcl-2 levels in normal ducts and lobules
• Apocrine metaplasia is a precursor to malignant varied with the menstrual cycle, being higher in the fol-
transformation.
licular than luteal phases, with high progesterone levels
• Apocrine metaplasia may result from a response to

apparently suppressing bcl-2 activity. Levels of bcl-2 were
the same stimulus as can cause cancer.
higher and the progesterone effect absent in fibroadeno-
• Apocrine metaplasia might itself have a higher mas, giving a possible mechanism for excessive lobular
propensity for malignant change.
growth in this condition. There is also evidence for a role
Kumar and colleagues27 demonstrated by an immuno- for matrix metalloproteinases in this hormone control of
histochemical technique that cells of apocrine metaplasia reproductive organs.32
have very high levels of prolactin not seen in normal Allan et al. studied apoptosis in normal epithelial cells
ductal cells, blunt duct adenosis, lobules or fibroade- adjacent to pathology (fibroadenoma, fibrocystic disease
noma. Tschugguel et al.28 found that apocrine metaplastic and cancer), and found reduced apoptosis in the case of
cells demonstrate endothelial calcium-dependent nitric the latter two.33 They suggest that this reduced apoptosis
oxide synthase, unlike other cells of benign breast disor- may be the cause of cellular build-up in ANDI, although
ders, suggesting that the vascular effects of nitric oxide an alternative is that the pathological tissue could affect
might play a part in progression to malignant change. A apoptosis by a (secondary) paracrine mechanism.
study of fetal breast tissue has made the situation even Loss of heterozygosity (LOH) is an indicator of a
more complex.29 Epithelial cells bearing a biochemical clonal, neoplastic condition rather than a simple hyper-
marker for apocrine cells (anti-GCDFP-15 monoclonal plasia. Lakhani et al. have shown that LOH is a feature of
antibody) were found in some duct cells of fetal breasts, atypical ductal hyperplasia (ADH) as well as ductal and
and in lobules of adult breasts, although no cells with lobular carcinoma in situ (DCIS and LCIS), suggesting
histological or ultrastructural apocrine features were that the essential step towards malignancy (aberration to
found. This suggests that apocrine cells appear when disease) has already taken place at the stage of ADH. They
some unknown stimulus causes apocrine precursor cells have recently shown34 evidence of LOH in some cases of
to take on the typical morphology. ductal hyperplasia without atypia, but not in apocrine

54

Descargado de ClinicalKey.es desde Univ Antioquia agosto 17, 2016.

Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2016. Elsevier Inc. Todos los derechos reservados.
 Aberrations of normal development and involution 4

cysts or benign papillomas. The cases showing LOH could differing results, emphasizing the preliminary nature of
not be distinguished morphologically from those without. these approaches.
This would suggest that progression from aberration to These findings as yet raise as many questions as they
disease may occur early in some cases of hyperplasia answer, but they do hold out hope that similar tech-
without atypia, presumably those that will progress to niques may soon give a much deeper understanding of
more severe pathology, or that these cases may be differ- the mechanisms of ANDI, and of the relationship between
ent ab initio. However, other workers35 have found aberration and disease.

REFERENCES

1. Hughes LE, Mansel RE & Webster DJTW. Aberrations of 13. Archer F & Omar N. The fine structure of fibroadenoma
normal development and involution (ANDI): a new of the human breast. Journal of Pathology 1969; 99: 113–
perspective on pathogenesis and nomenclature of benign 117.
breast disorders. Lancet 1987; 2: 1316–1319. 14. Love SM, Gelman RS & Silen W. Fibrocystic ‘disease’ of
2. Hughes LE, Smallwood J & Dixon JM. Nomenclature of the breast – a non disease. New England Journal of
benign breast disorders: report of a working party on the Medicine 1982; 307: 1010–1014.
rationalisation of concepts and terminology of benign 15. Bartow SA, Pathak DR, Black WC et al. Prevalence of
breast conditions. The Breast 1992; 1: 15–17. benign, atypical and malignant breast lesions in
3. McFarland J. Residual lactation acini in the female breast. populations at different risk of breast cancer. Cancer
Their relationship to chronic cystic mastitis and 1987; 60: 2751–2760.
malignant breasts. Archives of Surgery 1922; 5: 1–64. 16. Azzopardi JG. Problems in Breast Pathology. London: WB
4. Parks AG. The microanatomy of the breast. Annals of the Saunders; 1979.
Royal College of Surgeons of England 1959; 25: 295–311. 17. Miller WR, Dixon JM, Scott WN et al. Classification of
5. Sarnelli R & Squartini F. Fibrocystic condition and ‘at human breast cysts according to electrolyte and androgen
risk’ lesions in asymptomatic breasts: a morphological conjugate composition. Clinical Oncology 1983; 9: 227–
study of postmenopausal women. Clinical and 232.
Experimental Obstetrics and Gynecology 1991; 18: 271–279.
18. Dixon JM, Scott WN & Miller WR. An analysis of the
6. Hutson SW, Cowen PN & Bird CC. Morphometric studies content and morphology of human breast microcysts.
of age related changes in normal human breast and their European Journal of Surgical Oncology 1985; 11: 151–154.
significance for evolution of mammary cancer. Journal of
19. Haagensen CD. Mammary duct ectasia – a disease that
Clinical Pathology 1985; 38: 281–287.
may simulate carcinoma. Cancer 1951; 4: 749–761.
7. Page DL, Vander-Zwag R, Roger LW et al. Relationship
20. Davies JD. Inflammatory damage to ducts in mammary
between component parts of fibrocystic disease complex
dysplasia: a cause of duct obliteration. Journal of Pathology
and breast cancer. Journal of the National Cancer Institute
1975; 117: 47–54.
1978; 61: 1055–1063.
21. Kramer WM & Rush BF. Mammary duct proliferation in
8. DiLieto A, De Rosa G, Albano G et al. Desogestrone
the elderly – a histological study. Cancer 1973; 31: 130–
versus gestodene in oral contraceptives: influence on the
137.
clinical and histomorphological features of BBD. European
Journal of Obstetrics, Gynecology, and Reproductive Biology 22. Sloss PT, Bennett WA & Clagett OT. Incidence in normal
1994; 55: 71–83. breasts of features associated with chronic cystic mastitis.
9. Page DL & Dupont WD. Anatomic indicators (histologic American Journal of Pathology 1957; 33: 1181–1191.
and cytologic) of increased breast cancer risk. Breast 23. Wellings SR, Jensen HM & Marcum RG. An atlas of
Cancer Research and Treatment 1993; 28: 157–162. subgross pathology of the human breast with reference to
10. Vorherr H. The Breast. Morphology, Physiology and Lactation. possible pre-cancerous lesions. Journal of the National
New York: Academic Press; 1974. Cancer Institute 1975; 55: 231–273.
11. Kumar S, Mansel RE, Hughes LE et al. Prediction of 24. Hughes LE. Classification of benign breast disorders.
response to endocrine therapy in pronounced cyclical British Medical Bulletin 1991; 47: 251–257.
mastalgia, using dynamic tests of prolactin release. 25. Pye JK, Mansel RE & Hughes LE. Clinical experience
Clinical Endocrinology 1985; 23: 699–704. of drug treatments for mastalgia. Lancet 1985; ii: 373–
12. Simpson JF & Page DL. Loss of expression of fodrin (a 377.
structural protein) in cystic changes in the human breast. 26. Haagensen DE Jr. Is cystic disease related to cancer?
Laboratory Investigation 1993; 68: 537–540. American Journal of Surgical Pathology 1991; 15: 687–694.

55

Descargado de ClinicalKey.es desde Univ Antioquia agosto 17, 2016.

Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2016. Elsevier Inc. Todos los derechos reservados.
 Benign disorders and diseases of the breast

27. Kumar S, Mansel RE & Jasani B. Presence and possible absence of variation in fibroadenomas. Journal of
significance of immunohistochemically demonstrable Pathology 1997; 183: 204–211.
prolactin in breast apocrine metaplasia. British Journal of 32. Hulboy DL, Rudolph LA & Matrisian LM. Matrix
Cancer 1987; 55: 307–309. metalloproteinases as mediators of reproductive function.
28. Tschugguel W, Knogler W, Czerwenka K et al. Presence of Molecular Human Reproduction 1997; 3: 27–45.
endothelial calcium-dependent nitric oxide synthase in 33. Allan DJ, Howell A, Roberts SA et al. Reduction in
breast apocrine metaplasia. British Journal of Cancer 1996; apoptosis relative to mitosis in histologically normal
74: 1423–1426. epithelium accompanies fibrocystic change and
29. Viacava P, Naccarato AG & Bevilacqua G. Apocrine carcinoma in the premenopausal breast. Journal of
metaplasia of the breast: does it result from metaplasia? Pathology 1992; 167: 25–32.
Virchows Archiv. A, Pathological Anatomy and Histopathology 34. Lakhani SR, Slack DN, Hamoudi RA et al. Detection of
1997; 431: 205–209. loss of heterozygosity (LOH) indicates that mammary
30. Yi ES, Bedoya AA, Lee H et al. Keratinocyte growth factor hyperplasia of usual type is a clonal, neoplastic
causes cystic dilatation of the mammary glands in mice. proliferation. Journal of Pathology 1996; 178(Suppl): 5A.
American Journal of Pathology 1994; 145: 1015–1022. 35. Kasami M, Vnencak-Jones CL, Manning S et al. Loss of
31. Ferrieres G, Cuny M, Simony-Lafontaine J et al. Variation heterozygosity and microsatellite instability in breast
of bcl-2 expression in breast ducts and lobules in relation hyperplasia. American Journal of Pathology 1997; 150:
to plasma progesterone levels: overexpression and 1925–1932.

56

Descargado de ClinicalKey.es desde Univ Antioquia agosto 17, 2016.

Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2016. Elsevier Inc. Todos los derechos reservados.