Những Nguyên Tắc Thiết Yếu Trong Phẫu Thuật Tuyến Vú (Essentials of Breast Surgery)

7
Gynecomastia
ETIOLOGY
Genetic Disorders
Malignancy
Thyroid Disorders
Liver Disease
Renal Failure
Drugs
HIV-Positive Men
EVALUATION
History and Physical
TREATMENT
Medical Therapy
Surgery
Gynecomastia: Key Points
Understand the pathophysiology and risk factors for gynecomastia.
Be familiar with the medications that can cause gynecomastia.
Describe a thorough history, physical, and laboratory evaluation of the patient
with gynecomastia.
Describe the risk factors for male breast cancer and the indications for biopsy
of a male with a breast mass.
Know the treatment options for gynecomastia, including medical and surgical.
Gynecomastia is a benign proliferation of the
glandular tissue of the male breast. While it is a
common condition, it can not only be painful
for the patient, but can also be an obvious source
of psychological distress. Before adulthood,
there are two stages of physiologic gynecomas-
tia. Between 60% and 70% of infants have a
transient development of breast tissue due to
estrogenic stimulation from the mother. Once
the estrogens are cleared from the neonatal cir-
culation after delivery, the breast tissue gradually
regresses over 2 to 3 weeks. The second stage
comes at puberty, when 30% to 60% of boys
may exhibit gynecomastia. This usually resolves
by age 16 to 17.
In adulthood, the incidence of gynecomastia
increases with age, with a sharp increase after
age 50. The prevalence can be as high as 50%
to 60% depending on the criteria for defining
gynecomastia and the population studied. Most
of these patients were unaware that they had
any degree of gynecomastia, denying pain or
the presence of any palpable breast tissue.
It is important to differentiate gynecomastia
from lipomastia, which is excessive breast adi-
pose tissue but not enlargement of the ductal
91

elements. This is usually easily differentiated
on physical examination. Most patients with
gynecomastia present with complaints of a
unilateral
swelling
underneath
the
areola,
which is often tender. In many cases, this
might be a firm disc of retroareolar tissue, eas-
ily demarcated from the surrounding fat. In
other cases, there might be a dramatic increase
in the size of one breast over the other, or of
both breasts. The goal of the clinician is to rule
out a serious underlying endocrine or systemic
disease as the cause of the gynecomastia, and
if none is present, to guide treatment choices
for idiopathic gynecomastia.
Etiology
Examination of the subareolar tissue in nor-
mal men reveals the presence of major ductal
systems with rare secondary ductal branching.
This male breast tissue has estrogen receptors
and androgen receptors. With proliferation
of this secondary branching and the sur-
rounding stroma, gynecomastia occurs. Estro-
gens exert a trophic and stimulatory effect on
mammary
epithelial
tissue
growth
while
androgens inhibit this effect. The major andro-
gens include testosterone (synthesized by the
testes) and dihydrotestosterone, which is con-
verted from testosterone by several tissues
including the skin, adipose tissue, the prostate
and others. Estrogen is primarily present in
men as estradiol, which is synthesized from
peripheral aromatization of testosterone or
androstenedione (a weak androgen secreted
by the adrenals). Increased levels of estrogen
or decreased levels of androgens can tilt the
equation toward the growth of ductal breast
epithelium.
There are many underlying processes that
can lead to gynecomastia. As stated, gyneco-
mastia is not unusual in neonates because of
placental estrogens. Around puberty, there
can also be a point where the synthesis of
plasma estradiol increases before the maximal
synthesis of testosterone. This shift in the ratio
can lead to temporary gynecomastia. This may
occur in 30% to 60% of boys and may be more
common in boys with greater body mass. The
gynecomastia may be associated with asym-
metry, cosmetic concerns, pain, and tender-
ness, however this almost always resolves.
Finally, older men also have an increased inci-
dence of gynecomastia secondary to declining
plasma
testosterone
concentrations.
Again,
this appears to be more common in heavier
individuals.
There are, however, other causes of changes
in the estrogen/androgen ratio that not only
lead to the development of gynecomastia but
may be significant of a more serious health
condition. These include disease processes that
result in excess estrogen (either by increased
production or decreased clearance), those that
lead to decreased androgens, or those caused
by exogenous agents (medications, drugs).
Genetic Disorders
Several genetic disorders can be associated with
gynecomastia (Table 7–1). The familial aroma-
tase excess syndrome is caused by overexpres-
sion of the gene coding for aromatase, leading
to increased aromatization of androgens to
estrogens. Affected male patients present with
precocious puberty and gynecomastia. Absent
TABLE 7–1
Conditions Associated
with Gynecomastia
Physiologic (pubertal, aging)
Obesity
Exogenous estrogen (topical creams or ingested)
Malignancy
Increased estrogen
Sertoli cell (sex cord) tumors
Testicular germ cell tumors (choriocarcinoma,
seminoma, teratoma, embryonal carcinoma)
Leydig’s cell tumors
Adrenocortical carcinoma
Increased hCG production
Lung carcinoma
Liver carcinoma
Gastric carcinoma
Kidney carcinoma
Choriocarcinoma
Decreased testosterone synthesis
Anorchia, trauma, castration, viral orchitis,
irradiation
Hydrocele, varicocele, spermatocele
Testicular failure from hypothalamic or pituitary
disease
Klinefelter’s syndrome, hermaphroditism
Liver failure
Chronic renal failure
Hyperthyroidism
Refeeding syndrome
92
SURGICAL FOUNDATIONS: ESSENTIALS OF BREAST SURGERY

or defective androgen receptors may lead to
gynecomastia. Klinefelter syndrome (KS) is the
most common chromosomal deficiency asso-
ciated with hypogonadism and thus gyneco-
mastia. KS has a 47, XXY karyotype, which is
not only a risk for gynecomastia but also for
male breast cancer.
Malignancy
One of the most concerning sources of excess
estrogen comes from testicular tumors (Table
7–2). These can include gonadal stromal (non-
germinal) neoplasms and germinal cell tumors.
Nongerminal neoplasms include Leydig’s cell
(interstitial) tumors, Sertoli’s cell tumors, and
granulosa-theca cell tumors. Germ cell tumors
include choriocarcinoma, seminoma, teratoma,
and embryonal cell carcinoma. For this reason,
a testicular examination is a mandatory part of
the workup for gynecomastia. Any question of
a mass should prompt an ultrasound and referral
for urology consultation. Some of these tumors,
however, may be nonpalpable, small focal
lesions within the parenchyma of the testes.
Any young patient with gynecomastia that
cannot be explained, is associated with other
symptoms such as loss of libido or sexual dys-
function, or is associated with abnormal endo-
crine laboratory studies should also undergo a
testicular ultrasound and be referred for urology
consultation.
Other tumors can also lead to an excess estro-
gen state. Tumors of the adrenal cortex, both
adenomas and adrenocortical carcinoma, have
been associated with gynecomastia, but this is
rare. Adrenocortical carcinoma is a concern in
the patient with gynecomastia and symptoms
of Cushing’s disease or other mixed syndromes,
or with other signs of feminization.
Lung carcinoma can lead to an increase in
plasma chorionic gonadotropin and a simulta-
neous escalation inestrogen secretion. Rare cases
of malignant mesothelioma, kidney cancer, or
gastric cancers producing human chorionic
gonadotropin (hCG) have also been reported,
but again are rare. Finally, hepatocellular carci-
noma can be associated with gynecomastia, the
TABLE 7–2
Malignancies Associated with Gynecomastia
Tumor
Characteristics
Association with Gynecomastia
Leydig cell tumors
Rare tumors of testis, mostly
benign
May occur at any age, but is most
common in young and middle-
aged men
Secrete estradiol, raising estrogen and
suppressing LH, leading to reduction
in testosterone
Sertoli cell tumors
Rare tumors of testis, mostly
benign
Overexpression of aromatase, increasing
conversion to estrogens
hCG-secreting
tumors
Germ cell tumors of testes,
nontesticular tumors such as
lung, liver, gastric, or kidney
Tumors produce hCG, which stimulates
normal Leydig cells to secrete estradiol
preferentially
Take up steroid precursors from
circulation and aromatize them into
estrogens
Feminizing adrenal
tumors
Malignant, poorly differentiated
tumors with a poor prognosis
Peak incidence among young,
middle-aged men
Direct secretion of estrogen
Pituitary tumors
Secretion of LH or interference with
secretion of LH leading to a decrease in
androgen levels
Secretion of prolactin
Prostate cancer
Androgen-dependent tumor often
treated by androgen
deprivation or blockade
Treatment with estrogen, LHRH
analogues, nonsteroidal antiandrogens
or bilateral orchiectomy
hCG, Human chorionic gonadotropin; LH, luteinizing hormone; LHRH, LH-releasing homone.
93
7—GYNECOMASTIA

consequence of increased aromatase activity in
the hepatic neoplasm.
As androgen deprivation or androgen blockade
is a primary treatment of prostate cancer, gyneco-
mastia is often a result of therapy. Treatments
such as systemic estrogens, luteinizing hor-
mone-releasing hormone (LHRH) analogues,
nonsteroidal antiandrogens (flutamide, bicaluta-
mide), and bilateral orchiectomy can all lead to
the development of gynecomastia. Prophylactic
radiation therapy has been effective in decreas-
ing the risk of developing gynecomastia, but
there is concern regarding the long-term risks.
Tamoxifen has also been shown to be effective
at preventing gynecomastia in men about to
begin antiandrogen therapy.
Thyroid Disorders
Approximately 10% to 40% of men with thyro-
toxicosis may develop gynecomastia, although
the reasons are not clear. In hyperthyroid indi-
viduals, it may be direct mammotrophic effects
of thyroid hormone on ductal epithelium,
although many people believe it may be mediated
through alterations in estrogen metabolism,
specifically the increased peripheral conver-
sion of androgens to estrogens and increased
levels
of
sex
hormone-binding
globulin
(SHBG). Regardless, the gynecomastia almost
always resolves once the hyperthyroidism is
treated.
Liver Disease
Several chronic diseases of the liver can result in
gynecomastia, although the incidence may be
less than originally suspected. Gynecomastia is
present in up to 40% of cirrhotic men, but
may also be present in age-matched controls.
However, there are some physiologic reasons
for gynecomastia among cirrhotic patients.
These patients have not only a decreased pro-
duction of testosterone (alcohol has a direct
toxic effect on gonadal function), but also a
decreased clearance of androstenedione, leav-
ing more substrate available for aromatization.
Patients
with
liver
disease
also
have
an
increased concentration of SHBG, resulting in
an increase in protein bound testosterone and
a corresponding decrease in the free, biologi-
cally active fraction of plasma testosterone.
Gynecomastia has also been documented in
other chronic liver diseases such as idiopathic
hemochromatosis and fatty metamorphosis of
the liver.
Renal Failure
Gynecomastia can be a common result of ure-
mia and can be seen in up to half of patients
undergoing dialysis. This may be secondary
to histologic damage to the testes. In addition,
decreased metabolic clearance by the kidneys
may result in increased levels of luteinizing hor-
mone (LH) and subsequent increased estradiol
secretion. Many of these men may be nauseated
and anorexic, which improves after initiating
dialysis, leading to a refeeding syndrome (see
later text).
Drugs
Several drugs can lead to the development of
gynecomastia. Some of these do so because
they have direct estrogenic activity. These
include not only drugs such as estrogens, oral
contraceptives, and tamoxifen, but also ana-
bolic steroids, heroin, digitalis, and tetrahydro-
cannabinol, which is found in marijuana.
Other drugs inhibit testosterone or decrease its
synthesis. These include drugs such as cimeti-
dine, diazepam, flutamide, phenytoin, or spiro-
nolactone.
Several
other
drugs
have
been
linked to gynecomastia, although the mechan-
isms are unclear. Table 7–3 includes a list of
medications associated with gynecomastia. It
is estimated that approximately one in four
cases of gynecomastia is related to drugs.
Estrogens can clearly lead to gynecomastia,
but estrogen is rarely taken intentionally by
men (except in the case of prostate cancer
treatment or among male-to-female transsex-
uals). However, some men may be getting
unintentional exposures to estrogens. There
may be estrogen in certain skin creams or anti-
balding lotions. It can sometimes be absorbed
by men after intercourse with a woman using
a vaginal estrogen cream. Estrogen exposure
may also occur at work or through the diet,
although large amounts are typically needed
to develop gynecomastia (except in boys and
young men, in whom smaller exposures may
lead to gynecomastia) (Box 7–1).
HIV-Positive Men
Gynecomastia is being increasingly seen in men
with human immunodeficiency virus (HIV) dis-
ease, although the reasons are unclear and likely
multifactorial. Many of the antiretroviral medi-
cations used to treat HIV are associated with
gynecomastia. Some HIV-positive men use illicit
drugs (marijuana, heroin) or have concomitant
94
SURGICAL FOUNDATIONS: ESSENTIALS OF BREAST SURGERY

liver disease. Some HIV-positive patients have
hypogonadism. There may also be a refeeding
mechanism at play. This phenomenon was first
seen after World War II when men liberated from
prison camps developed gynecomastia after
resuming a normal diet. During starving, men
may develop hypogonadotropic hypogonadism,
and when these men eat normally and begin to
gain weight, there is a transient imbalance of
estrogens to androgens. This can also be seen in
refugees, in impoverished individuals, or after
dieting.
Evaluation
History and Physical
The patient should be queried as to the dura-
tion and timing of the breast enlargement
(Box 7–2). A thorough family history should
be obtained. A review of symptoms focusing
on signs and symptoms associated with hypo-
gonadism should be obtained, including erec-
tile dysfunction or decreased libido. Signs and
symptoms of systemic diseases associated with
gynecomastia should also be reviewed. Finally,
an
extremely
thorough
medication
history
and social history should be obtained. It is usu-
ally best to have family members wait outside
the examining room during the history and
physical to diminish the patient’s reluctance
to divulge potentially important information.
On physical examination, the breast tissue
can be elevated up off of the chest wall by
pinching, allowing for accurate measurement.
Tenderness should be noted and an attempt
should be made to elicit nipple discharge. Bilat-
eral breast examination should be performed to
compare symmetry, and a thorough examina-
tion for adenopathy should be performed. In
addition, percussion and palpation of the liver
should be performed, as should be a testicular
examination.
TABLE 7–3
Hormones or Regulators
of Hormone Synthesis

Anabolic steroids (nandrolone, testosterone
cypionate)

Estrogens and estrogen agonists (oral
contraceptives, hormone replacement, tamoxifen)

Human growth hormone

Goserelin acetate or leuprolide acetate
(gonadotropin hormone-releasing hormones)

Flutamide

Nilutamide

Antibiotics
○Metronidazole
○Ketoconazole
○Isoniazid
○Ethionamide
○Minocycline

Antiulcer medications
○Cimetidine
○Omeprazole
○Ranitidine

Cardiovascular drugs
○Spironolactone
○Bumetanide
○Digitoxin
○Digoxin
○Amiodarone
○Diltiazem
○Enalapril
○Methyldopa
○Nifedipine
○Reserpine
○Verapamil

Psychoactive agents
○Diazepam
○Haloperidol
○Phenothiazine
○Tricyclic antidepressants

Chemotheraputic agents
○Alkylating agents
○Methotrexate
○Vinca alkaloids
○Busulfan

Drugs of abuse
○Alcohol
○Amphetamines
○Heroin
○Marijuana
○Methadone

Auranofin

Calcitonin

Diethylpropion

Domperidone

Etretinate

Metoclopramide

Penicillamine

Phenytoin

Sulindac

Theophylline
BOX 7–1 UNINTENTIONAL
EXPOSURE TO ESTROGEN
Medications
Skin creams
Antibalding lotions
Vaginal estrogen cream (through
intercourse)
Occupations
Barbers (who apply estrogen-containing
creams into the scalp)
Morticians (who apply estrogen-
containing creams onto corpses)
Factory workers involved in
manufacturing estrogens
Diet
Milk or meat from estrogen-treated cows
Phytoestrogens in plants or animals
95
7—GYNECOMASTIA

The thickening should be concentric in the
breast.
Asymmetric
thickening
or
masses
should raise the possibility of malignancy.
Other worrisome features include ulceration,
immobility, bloody nipple discharge, and pal-
pable adenopathy.
Mammography
Mammography is extremely useful in imaging
gynecomastia.
Gynecomastia
appears
as
a
flame-shaped opacity extending into the sur-
rounding fat. Mammography can effectively
distinguish between malignant and benign
breast disease (Fig. 7–1). If the history, physi-
cal, and mammography all suggest simple
gynecomastia, biopsy may be avoided.
Biopsy
If at this point in the evaluation there is any
concern that an underlying malignancy might
be present, then a biopsy is indicated. Biopsy is
A
B
Figure 7–1. A and B, Craniocaudal and mediolateral oblique views of right breast demonstrate gynecomastia.
(Continued)
BOX 7–2 DETAILED HISTORY IN
PATIENT WITH GYNECOMASTIA


Onset and duration of breast
enlargement


Associated symptoms (breast pain,
tenderness, skin involvement, nipple
discharge)


Systemic symptoms (weight loss,
fatigue)


Family history of cancer


Symptoms of hyperthyroidism (heat
intolerance, nervousness, tachycardia,
weight loss, increased appetite)


Symptoms of liver disease


Changes in libido, impotence


Complete past medical history


Full list of medications (including over-
the-counter and “natural” medications)


Possible occupational or accidental
exposures to estrogens
96
SURGICAL FOUNDATIONS: ESSENTIALS OF BREAST SURGERY

not necessary in every patient with gyneco-
mastia, but patients with worrisome features
such as an asymmetric or hard mass, fixed
mass, skin ulceration, bloody nipple discharge,
or an abnormality on imaging should have a
biopsy
performed.
Core-needle
biopsy
has
been the biopsy method of choice in this
situation, although now that the cytologic
features
of
gynecomastia
have
been
well
described, fine-needle aspiration biopsy is an
option if an experienced cytopathologist is
available (Fig. 7–2).
Laboratory Evaluation
If the history does not suggest a drug as the
inciting agent, then several laboratory tests
may help find the cause (Table 7–4). Biochem-
ical assessment of the liver (liver function
tests), kidney (BUN/Cr), and thyroid (T3, T4,
thyroid-stimulating hormone) should be per-
formed because gynecomastia may be the
result of underlying liver disease, chronic renal
failure, or thyroid disease. A chest x-ray study,
if one has not been done recently, should be
obtained. This is particularly true in older indi-
viduals. In addition to these tests, measure-
ments of serum concentrations of several
hormones may be obtained. These include
serum estradiol (or estrone), testosterone, LH,
SHBG, hCG, and prolactin. These tests do not
need to be obtained routinely in every patient
with
gynecomastia,
but
their
use
should
be guided by the clinical situation. If any
abnormalities are discovered, further workup
and treatment should proceed accordingly.
However, the majority of cases will have a nor-
mal laboratory evaluation and will be labeled
as idiopathic gynecomastia.
Treatment
The treatment of gynecomastia depends on
the severity (Table 7–5) and whether an identi-
fiable cause is present. If a possible cause has
C
D
Figure 7–1—cont’d
C and D, Craniocaudal and mediolateral oblique views of left breast demonstrate a spi-
culated retroareolar mass with associated nipple retraction. Patient underwent mastectomy for treatment of
invasive breast cancer. (Images courtesy of Dr. Alexis Nees, Department of Radiology, University of Michigan.)
97
7—GYNECOMASTIA

Adult male patient with palpable mass
Biopsy
Surgery
See Chapter 22
Subcutaneous
mastectomy
Chest x-ray
Lab: hepatic/renal
Subcutaneous
masectomy
vs
Medical therapy
Observation weight
reduction
Labs: testosterone, LH
estradiol, sex hormone-
binding globulin, hCG,
prolactin, thyroid function
tests.
Work-up and
treatment as
Indicated
Thorough review of systems,
medication list, social history
History and physical, MGM
Physical with simple gynecomastia
Gynecomastia
Worrisome for cancer
Malignant
Abnormal
Normal
Painful or
distressing
to patient
Minimal
discomfort
or concern
Failure to
improve
Abnormal
See Table 7−4
No identifiable
cause
Identifiable cause
Normal
Figure 7–2. Approach to the male patient with a breast mass. Any suspicious mass should be biopsied.
A mass consistent with gynecomastia on history, physical, and mammogram may be observed, with a thor-
ough search for possible causes. How extensive of a biochemical evaluation should be undertaken in the
asymptomatic patient is controversial because there is a very low yield for identifying any abnormality.
TABLE 7–4
Laboratory Workup of Gynecomastia
Lab Results
Consider
Further Workup
"hCG
Germ cell tumor or hCG-secreting
nontrophoblastic neoplasm
Testicular ultrasound (testicular germ
cell tumor)
CT of chest and abdomen
(bronchogenic, renal, gastric, or
hepatic carcinoma)
" LH, # T
Primary hypogonadism
# LH, # T
" P
Prolactin-secreting pituitary
tumor
MRI of brain
# LH, # T
normal P
Secondary hypogonadism
" LH, " T
"T4, # TSH
Hyperthyroidism
" LH, " T
normal T4, TSH
Androgen resistance
# LH, " E2
Nongerminal testicular tumor,
adrenal tumor, increased
aromatase activity
Testicular ultrasound (rule out Leydig
or Sertoli cell tumor)
Abdominal CT or MRI (rule out
adrenal neoplasm)
CT, Computed tomography; E2, estradiol; hCG, human chorionic gonadotropin; LH, serum luteinizing hormone;
MRI, magnetic resonance imaging; P, prolactin; T, testosterone; T4, thyroxine; TSH, thyroid-stimulating hormone.

been
identified,
then
therapy
should
be
directed at correction of that cause. Gyneco-
mastia will almost always resolve with the
treatment of the underlying disorder or with-
drawal of the causative medication. Resolution
can be seen within a few weeks.
If no cause is identified, if the underlying
disease cannot be treated, or if the causative
medication cannot be discontinued, then the
goal of treatment depends on the severity of
the condition. Because most cases of gyneco-
mastia are associated with hormonal imbal-
ances or drugs and because gynecomastia
itself carries no serious significance (assuming
serious causes have been ruled out), reassur-
ance is the first line of therapy, and in many
cases patients desire no further intervention.
The risk of male breast cancer does not appear
to be increased in patients with gynecomastia,
except in patients with KS. For patients who
have symptomatic gynecomastia, treatment is
primarily
surgical,
although
some
medical
therapies have been proposed. Unfortunately
there are few well-designed prospective trials
of medical therapies for gynecomastia.
Medical Therapy
Clomiphene citrate is an antiestrogen that
acts at the level of the hypothalamic-pituitary
axis
to
increase
gonadotropin
secretion.
Although it may be successful, particularly
in the adolescent population, the side effects
may
be
severe
(gastrointestinal
reactions,
rashes, visual disturbances), and in this popu-
lation spontaneous resolution often occurs. It
is therefore not indicated in the treatment of
gynecomastia. Likewise, percutaneous dihy-
drotestosterone heptanoate has been shown
to reduce gynecomastia in pubertal patients
but has never been tested in prospective
randomized trials and again, spontaneous res-
olution is common in this population, so its
true efficacy is unknown.
Danazol
(Danocrine)
is
the
2,3,isoxazol
derivative of 17a-ethyl testosterone and has
been tested in patients with gynecomastia of
varying causes with response rates of 77% to
100%. Treatment lasts for 3 to 16 weeks with
dose schedules of 100 to 400 mg/day for adults
or 200 to 300 mg/day for adolescents. Side
effects include acne, weight gain, fluid reten-
tion, muscle weakness, and cramps. Finally,
tamoxifen can be effective in treating gyneco-
mastia, but the long-term effects of tamoxifen
in men are not well studied. Doses of 10 to
20 mg/d for 3 to 9 months have been used,
with resolution in up to 90% of men. It can
be considered in patients who have severe idi-
opathic
gynecomastia
after
a
thorough,
exhaustive workup has failed to identify any
underlying causes. If the gynecomastia recurs
upon stopping the tamoxifen, a second course
of therapy may be attempted.
Medical therapies are most effective in men
with
new-onset
gynecomastia.
With
long-
standing gynecomastia, the stroma is more
fibrotic and less likely to resolve. Surgery is
the preferred treatment in these patients, as
well as in patients who decline or fail medical
therapy.
Surgery
Most patients with gynecomastia prefer surgical
therapy given the rapid cosmetic improvement
and
the
avoidance
of
medications
(many
patients are taking several other medications
already for their underlying conditions). The
primary treatment of gynecomastia is surgical,
consisting of a subcutaneous mastectomy. Sim-
ple mastectomy is not indicated. Patients must
be aware that they may be trading one cosmetic
deformity for another.
For most patients with a small amount of
gynecomastia, a periareolar incision can be
used to excise the breast tissue. A small disk
of breast tissue should be left underneath the
nipple to prevent the sunken nipple defor-
mity. For patients with more extensive gyneco-
mastia, excess skin may need to be removed to
restore the contour of the breast. Several tech-
niques have been described to accomplish
this. However, in these more extreme cases,
or in cases in which there is both an excess
of breast tissue and fat, liposuction may be a
useful alternative to surgery.
TABLE 7–5
Clinical Classification
of Gynecomastia
Grade
Definition
I
Mild breast enlargement without skin
redundancy
IIa
Moderate breast enlargement without
skin redundancy
IIb
Moderate breast enlargement with
skin redundancy
III
Marked breast enlargement with skin
redundancy and ptosis
99
7—GYNECOMASTIA

Suggested Readings
1. Braunstein GD. Clinical practice. Gynecomastia.
N Engl J Med 2007;357(12):1229–1237.
2. Gikas P, Mokbel K. Management of gynaecomastia:
An update. Int J Clin Pract 2007;61:1209–1215.
3. Narula HS, Carlson HE. Gynecomastia. Endocrinol
Metab Clin N Am 2007;36:497–519.
4. Patterson SK, Helvie MA, Aziz K, Nees AV. Outcome
of men presenting with clinical breast problems:
The role of mammography and ultrasound. Breast
J 2006;12(5):418–423.
5. Wise GJ, Roorda AK, Kalter R. Male breast disease.
J Am Coll Surg 2005;200:255–269.
100
SURGICAL FOUNDATIONS: ESSENTIALS OF BREAST SURGERY

8
Identifying and Managing
the High-Risk Patient
RISK FACTORS FOR BREAST
CANCER
Hereditary Risk Factors
Menstrual and Reproductive
Factors
Dietary Factors
Factors Related to the Breast
Other Factors
STATISTICAL MODELS TO
ESTIMATE THE RISK OF BREAST
CANCER
Gail and Claus Models
WHAT CAN I DO TO DECREASE
MY RISK?
Lifestyle Changes
Chemoprevention
Surgery
MANAGEMENT OF THE PATIENT
WITH LOBULAR CARCINOMA IN
SITU
MANAGEMENT OF THE PATIENT
WITH A BRCA MUTATION
Increased Surveillance
Risk Reduction Strategies
Identifying and Managing the High Risk Patient: Key Points
Know the genetic syndromes associated with an increased risk of
breast cancers, as well as the other malignancies also at elevated risk.
Understand the impact of family history and menstrual and reproductive
factors on the individual woman’s risk of breast cancer and the strengths and
limitations of predictive models.
Describe dietary and environmental factors that may increase or decrease the
risk of breast cancer.
Understand the implications and management of atypical hyperplasia.
Describe the management options for the woman diagnosed with lobular
carcinoma in situ.
Know the risks and benefits of tamoxifen and raloxifene for the
chemoprevention of breast cancer.
Recognize the options for the woman with a known or suspected BRCA1
or BRCA2 mutation.
101

All women are at risk of breast cancer, with
one in eight developing the disease within
their lifetime. The breast surgeon is going to
see a wide array of women, each one at a dif-
ferent risk of developing cancer, or in some
cases a second cancer. Knowing that risk will
impact several decisions, including when and
how to begin screening, whether an abnormal
finding should be biopsied or observed, the
best course of treatment for a woman diag-
nosed with cancer, implications for family
members, and when a woman without breast
cancer may benefit from chemoprevention or
prophylactic
surgery.
In
addition,
many
patients concerned about cancer will have
many questions regarding things they can do
to decrease their risk.
It is also up to the surgeon to allay patient’s
concerns and misconceptions regarding their
risk of breast cancer. Most women have been
inundated with media reports regarding the
importance of family history. Women with a
single, second-, or third-degree relative who
had breast cancer in her 70s may present
overly worried about her risk of breast cancer.
The emphasis of family history and breast can-
cer also has the effect of convincing women
that if they do not have a family history, they
cannot get breast cancer. A large majority of
breast cancer cases are not associated with a
positive family history. In fact, nearly half of
breast cancers patients have no identifiable
risk factor.
Discussing risk with a patient is not easy.
Telling the patient that she has a 1 in 8 risk
over her lifetime or that her risk of developing
breast cancer is 4% over the next 5 years is a
difficult concept for patients to interpret. It is
also important not to overly frighten patients.
Risk Factors for Breast Cancer
The biggest risk factor for developing breast
cancer is gender. Breast cancer is 100 times
more common in women than in men. The
second strongest risk factor is age. The inci-
dence of breast cancer rises sharply with age,
although the rise is less steep after age 50,
reflecting the impact of hormonal changes.
At 75 to 80 years of age, the risk starts to
decrease slightly. There are multiple other risk
factors for breast cancer, with varying preva-
lence in the population and varying relative
risks.
Certain risk factors may be surrogates for
true risk factors. For example, several studies
have demonstrated that women of higher
socioeconomic status are at greater risk for
breast cancer, whether one categorizes this as
economic status, education, occupation, or
other measures of status. This may be as great
as a two-fold difference between the highest
and lowest classes. The reasons for this are
not entirely clear, but it is hard to believe that
higher education directly leads to the develop-
ment of breast cancer. A more likely explana-
tion is that women of varying socioeconomic
factors have differences in their menstrual his-
tory. For example, women who spend more
time in school or who are dedicated to their
career may be delaying their first pregnancy
until later in life or have fewer pregnancies.
Another possibility is that female infants who
have better nutrition may develop menarche
at an earlier age.
Another risk factor for breast cancer is the
area of the world in which the woman lives.
For example, breast cancer incidence and mor-
tality is much higher in the United States and
Northern Europe than it is in Asia. The inci-
dence and mortality rates also vary throughout
the United States. The incidence is highest in
Hawaii and the lowest in Utah. Again, this
increased risk may be related to factors such
as genetics or differences in the number of
children, age at first live birth, or age at menar-
che. They may also represent differences in
environment or diet. For example, studies of
first-generation daughters of Japanese Ameri-
can women suggest that risk increases in Japa-
nese women when they move to the United
States, suggesting a lifestyle factor.
Hereditary Risk Factors
Race
There are considerable interracial differences
in the risk of developing breast cancer. The
rates
in
African-American
women,
Latina
women, and Asian-Americans are all lower
than those in Caucasians. The reasons for this
are not clear. Some of this might be genetic,
but some may be attributable to lifestyle, diet,
or environment. As an example, some of the
lowest rates of breast cancer were seen in Asia,
particularly in the 1970s and 1980s. How-
ever, the incidence of breast cancer in Asia
has been increasing, and Asian women living
in the United States have an increased risk of
breast
cancer,
suggesting
environmental
causes. This is discussed in more detail in
Chapter 22.
102
SURGICAL FOUNDATIONS: ESSENTIALS OF BREAST SURGERY

Family History
Family history is clearly an important risk fac-
tor. Familial clustering of breast cancer has
long been recognized, with the greatest risk
in women with a first-degree relative with
early-onset breast cancer. However, only 10%
of women diagnosed with breast cancer have
a positive family history. It should be empha-
sized with women that the absence of a family
history does not mean that breast cancer is not
a concern. Likewise, not all women with a rel-
ative who had breast cancer are at the same
increased risk. The risk increases with the age
of the patient, the age of the relative, the num-
ber of relatives with cancer, and the relation
between the subjects (e.g., first degree, second
degree). For example, in a metaanalysis using
data from 150,000 women, the risk of breast
cancer was increased three times for a woman
who had a relative with breast cancer before
age 30, but only 1.5 times if the relative was
diagnosed after age 60. As another example,
women with one affected first-degree relative
had a risk 1.8 times that of their counterparts,
whereas women with two affected first-degree
relatives saw risk increase threefold.
Genetic Mutations
One of the strongest risk factors for the develop-
ment of breast cancer is the family history. How-
ever, the number of breast cancers that are due to
the inheritance of a specific genetic mutation is
relatively low. It is believed that only 5% to
10% of all breast cancers are associated with the
inheritance of a high penetrance autosomal
dominant cancer predisposing gene such as
BRCA1, BRCA2, p53, ATM, PTEN, MLH1, or
MSH2.
Breast cancers associated with germ line muta-
tions tend to occur at younger ages. Genetic
mutations may be responsible for more than
25% of breast cancers in women younger than
30, compared to less than 1% for women older
than 70. There may also be multiple cancers
(e.g., bilateral breast cancers) or associations
with other malignancies such as ovarian cancer
(in the case of BRCA1/2 mutations) or sarcomas
(as with p53 mutations). The likelihood that a
breast cancer may be secondary to a genetic
mutation is more likely when multiple family
members have early-onset or bilateral breast can-
cers, or other malignancies (Table 8–1).
BRCA1 and BRCA2
With the discovery of BRCA1 and BRCA2 in
1994 and 1995, respectively, the management
of breast cancer changed dramatically. Physi-
cians caring for all women now need to
understand the benefits and pitfalls of genetic
testing and which populations should undergo
testing. This is particularly true for surgeons see-
ing patients with signs or symptoms suggestive
of cancer, or women already diagnosed with
breast cancer. If these patients harbor a BRCA1
or BRCA2 mutation, they have a risk of develop-
ing breast cancer as high as 70% to 80% in their
lifetime. How mutations in BRCA1/2 predispose
to cancer is not well understood, although the
evidence suggests that BRCA genes act as tumor
suppressor genes.
It should be obvious how the identification of
such a mutation may alter the surgical manage-
ment of the breast cancer patient, increase suspi-
cion in a patient with a breast complaint, as well
as impact family members. However, genetic
testing should not be obtained indiscriminately.
BRCA mutations are rare, occurring in approxi-
mately 0.1% of the general population and at a
TABLE 8–1
Genes Associated with an Increased Risk of Breast Cancer
Gene (Syndrome)
Breast Cancer Risk
by Age 70 (%)
Other Associated Cancers
P53 (Li-Fraumeni)
>90%
Sarcoma (soft tissue and osteo), brain tumors,
adrenocortical carcinoma, leukemia, colon
PTEN (Cowden’s)
25% to 50%
Thyroid, enodmetrial, genitourinary
STK11/LKB1 (Peutz-
Jeghers)
45% to 50%
Small intestine, colorectal, uterine, testicular, ovarian
sex cord
CDH1 (hereditary diffuse
gastric carcinoma)
39%
Diffuse gastric cancer
BRCA1 (HBOC)
39% to 87%
Ovarian, pancreatic
BRCA2 (HBOC)
26% to 91%
Ovarian, prostate, pancreatic, male breast cancer
Other genes associated with an increased risk: ATM (ataxia-telangiectasia), BRIP1 (Fanconi’s anemia), PALB2
103
8—IDENTIFYING AND MANAGING THE HIGH-RISK PATIENT

slightly higher rate in certain other populations.
For example, Ashkenazi Jewish women (meaning
of Eastern European descent) have a higher inci-
dence of BRCA mutations, about 2%. There are
substantial psychological and social risks asso-
ciated with genetic testing. In addition, one must
be cognizant of the fact that genetic testing is not
perfect. No single technique for genetic testing is
able to detect all mutations. Even when the
entire gene is sequenced, the detection rate is
not 100%. A negative test may give a woman
false reassurance about her subsequent risk of
breast cancer. Genetic testing can also produce
indeterminate or false-positive findings.
When considering a patient for genetic
counseling and testing, it is important to assess
the likelihood of finding a mutation. The great
majority of women do not have a family or per-
sonal history suggestive of a BRCA mutation.
Overall about 5% of women with breast cancer
will be found to have a BRCA mutation. This
risk increases with a decrease in the age of diag-
nosis. As mutations in these genes are inherited
in an autosomal dominant fashion, several
generations of women in affected families
may have breast and possibly ovarian cancer.
The chance of finding a mutation in BRCA1
or BRCA2 increases with the number of family
members who have had breast or ovarian can-
cer, the number of relatives who have had mul-
tiple cancers, and the younger age when family
members developed cancer (Table 8–2). A
detailed family history on both the paternal
and maternal sides should be obtained, record-
ing any malignancies (not just breast or ovar-
ian cancer), the age when the malignancy
occurred, and in the case of breast cancer,
whether the cancer was bilateral. A useful tool
to predict the likelihood of an individual
having a deleterious BRCA mutation is a com-
puterized program BRCAPRO, which incor-
porates six predictive models for inherited
breast cancer. This program is available free of
charge at
edu/BayesMendel/.
In women with breast cancer, other clues
that a BRCA mutation may be responsible
may be in the histology, because cancer sec-
ondary to BRCA1 mutations is more likely to
be
poorly
differentiated,
estrogen
receptor
(ER)/progesterone receptor (PR)–negative, and
may have a medullary-like histology.
TABLE 8–2
The Likelihood of Finding a BRCA Mutation Based on Family History
Family History
Chance of BRCA Mutation
1 Breast cancer younger than 30 years of age
12%
1 Breast cancer younger than 40 years of age
6%
1 Breast cancer between 40 and 49 years of age
3%
2 Breast cancers younger than 50 years of age
15%
2 Sisters with breast cancer younger than 40 years of age
37%
3 Breast cancers younger than 50 years of age
28%
4 Breast cancers younger than 50 years of age
47%
5 Breast cancers younger than 50 years of age
60%
Ovarian cancer younger than 50, not Ashkenazi Jewish
10%
Ovarian cancer older than 50, not Ashkenazi Jewish
3%
Ovarian cancer younger than 50, Ashkenazi Jewish
60%
Ovarian cancer older than 50, Ashkenazi Jewish
30%
1 Ovarian cancer and 1 breast cancer younger than 50 years of age
33%
1 Ovarian cancer and 2 breast cancers younger than 60 years of age
60%
1 Ovarian cancer and 3 breast cancers younger than 60 years of age
90%
2 Ovarian cancers, average age younger than 60 years of age
40%
2 Sisters with ovarian cancer younger than 50 years of age
61%
2 Ovarian cancers, average age older than 60 years of age
20%
2 Ovarian cancers and 1 breast cancer younger than 60 years of age
80%
2 Ovarian cancers and 2 breast cancers less younger than 60 years of age
95%-100%%
3 Ovarian cancers
90%
More than 3 ovarian cancers
100%
104
SURGICAL FOUNDATIONS: ESSENTIALS OF BREAST SURGERY

If after a careful history a BRCA mutation is
suspected (at least a 10% chance), the patient
should be referred to a genetic counselor. Genetic
testing should not be initiated by the clinician
without pretest counseling. Women need to con-
sider the implications of either a positive or nega-
tive result. The implications of both are complex
enough, and the potential psychosocial, medical,
and legal ramifications severe enough, that
counseling is necessary to help women make an
informed decision and plan how they and their
family are going to use the information. There
are
implications
regarding
insurance
and
employment discrimination. Although federal
legislation (the Health Insurance Portability and
Accountability Act of 1996) and state laws pro-
hibit insurers from excluding an individual from
health coverage based on genetic testing, there
are still some areas where the law fails to fully
extend this protection. Furthermore, genetic test-
ing is not inexpensive and not always fully cov-
ered by insurance.
Interpreting the results of genetic testing is not
straightforward. When dealing with a woman
with a strong family history, the ideal situation
is one in which there are family members alive
who had or have breast cancer. In this situation,
if a deleterious mutation has been identified in
the family member, genetic testing would be
extremely helpful for the patient. In this case, a
true positive result would occur when the in-
dividual being tested has the same mutation as
the affected family member. In the case of a
positive result, the patient would know her
increased risk and this would allow her to con-
sider prophylactic measures. A negative result
would inform the woman that, despite her fam-
ily history, she does not carry this elevated risk.
However, if a specific cancer-associated mu-
tation has not or cannot be identified in an
affected family member, genetic testing for
the patient may not be as helpful. A negative
test does not rule out an increased risk second-
ary to genes other than BRCA1 and BRCA2.
Or, there could be a mutation in BRCA1/2 that
was not detected by the available methods.
A positive result may also be uninformative.
Although it may represent a newly identified
mutation, it could also simply be a benign
polymorphism. It is also unclear what the
exact range of cancer risk might be for differ-
ent mutations.
p53 (Li-Fraumeni Syndrome)
Li-Fraumeni syndrome is an autosomal domi-
nant condition characterized by the devel-
opment of multiple tumors. Families have
identifiable mutations in the p53 gene. In
addition to early-onset breast cancer, carriers
have an increased risk of soft tissue sarcomas,
osteosarcomas, brain tumors, leukemias, and
adrenocortical malignancies. Half of Li-Frau-
meni patients develop some type of cancer by
age 30.
ATM (Ataxia Telangiectasia)
An autosomal recessive disorder, ataxia telan-
giectasia (AT) results from a mutation in the
ATM gene, mapped to chromosome 11q22.3.
It is quite rare (between 1 in 20,000 and 1 in
100,000 live births) and is characterized by
progressive cerebellar ataxia and other neuro-
logic abnormalities, oculocutaneous telangiec-
tasias, immune deficiencies, diabetes mellitus,
and a predisposition to malignancy, including
breast cancer Although the disease is rare, it
is estimated that as many as 1% to 2% of Cau-
casians in the United States may carry one
defective gene. Several studies support an
increased risk of breast cancer in AT carriers
(as high as 4 times that of noncarriers).
PTEN (Cowden Syndrome)
Mutations in the PTEN gene (phosphatase and
tensin homologue), also referred to as MMAC1,
can lead to Cowden syndrome, characterized by
hamartomas in the skin, oral mucosa, breast,
and intestine. PTEN is a tumor-suppressor gene,
and Cowden syndrome is inherited in an auto-
somal dominant fashion. In addition to the
hamartomas, carriers are at an increased risk of
both breast and thyroid cancers. Up to 75% of
women with Cowden syndrome will manifest
benign
breast
conditions
(fibroadenomas,
papillomas), whereas cancer will develop in
25% to 50%. Benign thyroid conditions (adeno-
mas, goiter) will develop in half of those
affected, whereas thyroid cancer develops in
about 10% of individuals. Endometrial and
renal cancers may also be associated with Cow-
den syndrome.
STK11 (Peutz-Jeghers Syndrome)
Peutz-Jeghers syndrome (PJS) is a result of muta-
tions in STK11, a serine threonine kinase, and is
characterized by hamartomatous polyps in the
gastrointestinal tract. Affected individuals are
at increased risk of many malignancies includ-
ing gastrointestinal (gastric, pancreatic, colon)
and cancers outside the gastrointestinal tract
(breast, ovarian, endometrial). PJS is also asso-
ciated with mucocutaneous pigmentation in
the buccal mucosa, lips, fingers, and toes.
105
8—IDENTIFYING AND MANAGING THE HIGH-RISK PATIENT

Menstrual and Reproductive
Factors
Prolonged exposure to both exogenous and
endogenous estrogen will increase the risk of
breast
cancer.
For
premenopausal
women,
estrogen subtypes (estradiol, estriol, estrone)
come from the ovaries. After menopause, the
main source of estrogen is dehydroepiandros-
terone (DHEA). Produced in the adrenal gland,
DHEA is metabolized in peripheral fat tissue to
estradiol and estrone. The exposure to estro-
gen, and thus the risk of breast cancer, can
therefore be estimated by assessing key repro-
ductive factors.
Age at Menarche
The later the onset of regular menstrual cycles,
the later the exposure to estrogen and less life-
time exposure. Therefore the older a young
woman is when she begins the menstrual
cycle, the lower the risk of developing breast
cancer. There is an approximately 10% reduc-
tion in breast cancer risk for every 2-year delay
in the onset of menarche.
Age at Menopause
In contrast to the age of menarche, the later a
woman undergoes menopause, the higher her
risk for breast cancer, which is thought to
reflect longer exposure to endogenous hor-
mones. Women who undergo bilateral oopho-
rectomy before the age of 40 can reduce
their risk by 50%. However, this risk reduc-
tion disappears if the woman takes hormone
replacement.
Pregnancy
The number of pregnancies a woman has and
the age at which she has her first full-term preg-
nancy can influence breast cancer risk as well.
The protective effect comes from the full cellular
differentiation in the breast that occurs during
and after pregnancy (Chapter 1). Nulliparous
women have a relative riskof about 1.5.Whether
multiple
pregnancies
decrease
that
risk
is
unclear. Also somewhat confusing is the impact
of the age of a woman at her first full-term preg-
nancy. As a general rule, women giving birth
at older ages have a higher risk than their
younger counterparts. However, it is not so sim-
ple. For example, women who give birth to their
first child after the age of 30 appear to have a
higher
risk
than
nulliparous
women.
And
women are at a higher risk of breast cancer in
the years immediately after pregnancy, with the
decrease in risk not coming until later in life
(about 10 years after delivery).
This protection is conferred by full-term preg-
nancy, so this does not apply to disrupted
pregnancies.
Because
abortion
disrupts
the
pregnancy and the cellular differentiation of
the breast, opponents to abortion have pro-
moted the idea that abortion increases the risk
of breast cancer. Population-based cohort stud-
ies, however, fail to support an association
between abortion and breast cancer risk, and
the National Cancer Institute accepted the con-
clusions of a workshop examining reproductive
events and breast cancer that there is no evi-
dence of a link between induced abortion and
breast cancer risk.
Hormone Levels
The association between reproductive factors
and breast cancer risk centers on exposure to
hormones, primarily estrogen. This is sup-
ported by observational data, epidemiologic
studies, animal models, and the clear evidence
of risk reduction by decreasing estrogen levels.
So it is reasonable to believe that measuring
serum estrogen levels would identify high-risk
individuals. However, the correlation between
breast cancer risk and the levels of hormones
in the blood have not been consistent. Part
of this is due to difficulties with the assays.
More
responsible
is
the
large
variability
between patients, and within individuals, of
hormone concentrations. Estrogen levels fluc-
tuate during the menstrual cycle, so studies
of premenopausal women, particularly when
only one blood sample is used, are difficult
to interpret. The data are clearer for postmeno-
pausal women, in whom multiple prospective
epidemiologic studies have found a positive
relationship between serum estradiol concen-
tration and breast cancer risk. The association
is strongest when looking at ER- and PR-
positive tumors.
A surrogate for circulating estrogen levels is
bone mineral density, in that bone contains
estrogen receptors and is highly sensitive to
estrogen levels. Women with high bone den-
sity appear to have a higher risk of breast can-
cer than those with low bone density.
Hormone Replacement Therapy
There is
clear evidence
of an association
between the use of hormone replacement ther-
apy (HRT) and breast cancer. Although this
106
SURGICAL FOUNDATIONS: ESSENTIALS OF BREAST SURGERY

risk has received extensive media attention
and is consistent across studies, it is a relatively
modest risk, estimated to be about a 1.24-fold
increase. However, the link between
HRT
and breast cancer is not so straightforward.
Although the risk has been demonstrated for
women taking combined estrogen/progester-
one compared to placebo, the same risk has
not been demonstrated for women taking
unopposed
estrogen,
which
may
even
be
associated with a lower breast cancer risk. The
length of time a woman uses HRT is also
important. Long-term use of HRT is associated
with the highest risk. Using HRT for a limited
amount of time does not appear to increase
the risk of breast cancer significantly.
Oral Contraceptives
The evidence is also fuzzy about whether oral
contraceptives are linked to breast cancer. Mul-
tiple studies were unable to demonstrate an
increased risk, although a large metaanalysis
calculated a small but significant increase in
the relative risk (RR 1.24). The risk appears to
decrease after stopping use and by 10 years after
stopping is back to normal. This metaanalysis
has been criticized, however, for limited fol-
low-up, and two subsequent studies found no
evidence of increased risk, including among
women with a family history of breast cancer.
Depot medroxyprogesterone acetate (DMPA)
is a long-lasting, injectable contraceptive that
suppresses ovulation for 90 days. It is approved
for contraceptive use in the United States. It
was thought that the serum estradiol-lowering
effects of this drug might decrease breast cancer
risk but there is no evidence for this in epidemi-
ologic studies. In contrast, a case control study
by the World Health Organization showed a
slight increased breast cancer among current
users, but not in past users.
Dietary Factors
Many of the factors discussed are beyond the
control of the individual. Family history, age,
race, genetic factors, and age of menarche or
menopause are all factors that cannot be
altered. Other factors, such as age at first preg-
nancy or the number of pregnancies, can be
altered, but more important considerations
tend to influence family planning decisions.
The decision to breast-feed and for how long,
and the decision to take hormone replacement
therapy are controlled by the patient. Other
areas within the control of the individual are
factors related to diet and exercise.
Height and Weight
The height and weight of an individual has
been associated with a higher risk of breast can-
cer in a number of studies. The reasons for this
are complex. A woman’s height and weight var-
ies with her nutritional status in her develop-
mental years, which is also related to the age
of menarche. Young women who have signifi-
cant nutritional deficiencies, such as those with
anorexia nervosa, have a decreased incidence of
breast cancer. In addition, a high body mass
index (BMI) is associated with increased levels
of estrogens. Obesity is strongly associated with
an increased risk of breast cancer.
Physical Activity
Exercise has been linked to a decreased risk of
breast cancer, but again, the reasons are com-
plex. This does not need to be overly strenuous
exercise, because a decreased risk of breast can-
cer can be seen in women who do brisk walking
for 10 hours or more per week. The protective
effect is seen in both premenopausal and post-
menopausal
women.
Several
mechanisms
might explain this. The most obvious would
be the link between exercise and reducing obe-
sity, although a direct link between weight loss
and risk reduction has not been demonstrated.
Physical activity also appears to reduce serum
estrogens,
particularly
in
postmenopausal
women, and this may also play a role in the
decreased risk among women with increased
physical activity.
Specific Foods
Epidemiologic studies of diet and cancer are
extremely difficult for obvious reasons. They
are often heavily dependent upon patient
recall, which can often be inaccurate. The
great variability of foods within a particular
group makes categorization difficult. There
are often confounding factors associated with
diet that may bias the results. It is therefore
difficult to make any solid recommendations
on risk reduction based on dietary factors.
However,
many
of
the
recommendations
(decreased dietary fat, increased fruits and
vegetables) have other health benefits.
Ecologic studies comparing breast cancer
incidence and dietary fat intake have shown a
strong positive correlation between the two,
and for many years this was suspected to be
the primary reason for the increased breast
cancer risk seen in many Western countries.
However,
the
results
of
prospective
case
107
8—IDENTIFYING AND MANAGING THE HIGH-RISK PATIENT

control and cohort studies have been mixed.
In the Nurses’ Health Study, women who had
the lowest fat intake had a decreased risk of
breast cancer compared to those women with
the highest fat intake. This was among preme-
nopausal women. A pooled analysis of more
than 300,000 women, mostly postmenopau-
sal, failed to demonstrate any connection
between dietary fat and breast cancer. It may
be that the influence of dietary fat is more sig-
nificant in younger women, particularly in the
prepubertal years. This may have more impact
on age of menarche and subsequent BMI.
A prospective study, The Women’s Health Study,
which randomly assigns women to a low-fat or
average American diet, is presently under way
and may provide more answers on this topic.
There is presently no convincing evidence
that taking vitamins or nutrients, such as vita-
mins A, C, E, or selenium, will alter the risk of
breast cancer. Although caffeine is linked to
fibrocystic disease of the breast, a number of
studies have failed to show any increased risk
of breast cancer with caffeine intake.
Alcohol is associated with an increased risk
of breast cancer. This effect may be additive
with HRT.
There has been considerable interest in the
effects of phytoestrogens on breast cancer risk.
Phytoestrogens are natural plant substances
that have potential antioxidant activity. There
are several types of phytoestrogens: isofla-
vones, coumestans, and lignans, which are
found in a variety of fruits and vegetables.
Overall, the data suggesting a protective effect
of phytoestrogens in general have been weak.
However, two phytoestrogens found in soy,
the isoflavones genistein and daidzein, may
function as weak estrogens in the body. In
effect,
these
isoflavones
may
behave
like
tamoxifen, functioning as relative antiestro-
gens by displacing estradiol. Given the high
soy intake in Asia and the low rates of breast
cancer, many people believe that soy con-
sumption decreases breast cancer risk.
Factors Related to the Breast
Previous History of Breast Cancer
One of the strongest risk factors for the devel-
opment of breast cancer is a personal history
of either in situ or invasive breast cancer. The
risk of a second breast cancer in women with
a history of breast cancer is approximately 1%
per year among premenopausal women and
0.5% per year for postmenopausal women.
Breast Density
Dense breast tissue often hampers both physi-
cal examination of the breast as well as mam-
mographic detection of breast cancers. There
is some evidence now that the presence of
dense breast tissue is also associated with an
increased risk of developing breast cancer,
with a relative risk between 1.8 and 6 times
that of women with less dense tissue.
Breast-Feeding
Epidemiologic studies correlating breast-feeding
with breast cancer have been quite variable,
with no clear results. Studies in the United
States have failed to demonstrate a connection,
although studies from Asian countries have. It
should be pointed out that women in these
countries tend to breast-feed longer than Amer-
ican women. An analysis of 47 studies investi-
gating this question concluded that the risk
for breast cancer was decreased with breast-
feeding, but this effect is quite small, perhaps
a 4.3% decrease in risk for every 12 months of
breast-feeding. This is in addition to a decrease
of 7% for each birth. The exact mechanisms
for this protective effect is unknown, although
it is most likely due to a delay in the return of
the ovulatory cycles and a decrease in estrogen
production.
Proliferative Lesions without Atypia
Many women will present with complaints of
breast masses or mammographic abnormalities
that will ultimately turn out to have benign
disease. While these lesions may not require
immediate treatment, some may indicate an
increased risk of breast cancer for the individ-
ual. Benign lesions can be categorized as prolif-
erative or nonproliferative. Nonproliferative
lesions are not associated with an increased
risk for breast cancer, whereas proliferative
lesions may be associated with an increased
risk of either noninvasive or invasive disease.
The risk associated with proliferative lesions
depends upon the degree of atypia associated
with the lesion.
Proliferative lesions without atypia include
fibroadenoma, moderate or florid hyperplasia,
sclerosing adenosis, radial scar, and intraductal
papillomas. In these women, the relative risk
of breast cancer is approximately 1.3 to 2 times
that of other women. This varies with the spe-
cific lesion and the features of that lesion. For
example,
the
presence
of
a
fibroadenoma
appears to be associated with an overall risk of
108
SURGICAL FOUNDATIONS: ESSENTIALS OF BREAST SURGERY

breast cancer of 1.4 to 1.7 that of the general
population. However, the risk is not uniform
among all fibroadenomas. The majority of the
elevated risk is among women with fibroadeno-
mas exhibiting a complex histology or hyper-
plasia or those that occur in women with a
first-degree relative with breast cancer (a risk
above that of the family history alone). The
remainder of women with a fibroadenoma
(about 70%) are left without an apparent
increased risk.
Proliferative Lesions with Atypia
(Atypical Hyperplasia and Lobular
Carcinoma in Situ)
When atypia is present in a proliferative
lesion, the risk of breast cancer is significantly
increased.
The
relative
risk
of
developing
breast cancer in a patient with atypical hyper-
plasia is increased 4.5 to 5 times. Atypical
hyperplasia includes patients with atypical
ductal hyperplasia and atypical lobular hyper-
plasia. Lobular carcinoma in situ is sometimes
categorized as a proliferative atypical lesion
and sometimes categorized as in situ disease.
It is most appropriately discussed as a risk fac-
tor, associated with a 7 to 9 times increased
risk and an absolute lifetime risk of breast can-
cer of 20%.
The atypical hyperplasias are proliferative
lesions that possess some, but not all, of the
features of carcinoma in situ. Atypical ductal
hyperplasia (ADH) has some of the architec-
tural and cytologic features of low-grade ductal
carcinoma in situ (DCIS), and a distinction
between the two is sometimes difficult. ADH
is typically confined to a single lobular unit,
seldom larger than about 3 mm overall. The
hyperplastic cells may form tufts, micropapil-
lae, or bridges within the involved space. The
cells are relatively uniform with monomorphic
round nuclei (Fig. 8–1). Atypical lobular hyper-
plasia (ALH) usually involves multiple lobular
units with or without involvement of inter-
vening ducts. The cells are monomorphic and
evenly spaced with round or oval eccentric
nuclei. The cytoplasm is pale with intracyto-
plasmic vacuoles. (Fig. 8–2).
Lobular carcinoma in situ (LCIS) is a nonin-
vasive lesion that arises from the lobules and
terminal ducts of the breast. LCIS is almost
always an incidental finding in a breast biopsy
in that it does not form a mass, does not have
calcifications or other mammographic find-
ings, and does not lead to nipple discharge
or other breast symptoms. When LCIS is
discovered on a biopsy prompted by calcifica-
tions, the calcifications are often located in
normal epithelial cells adjacent to the LCIS,
rather than the involved lobules.
Histologically, LCIS is characterized by acini
filled with a monomorphic population of small,
round, polygonal, or cuboidal cells. The cells are
often pale and slightly eosinophilic. These pos-
sess a thin rim of clear cytoplasm with a high
nuclear-to-cytoplasm
ratio.
The
nuclei
are
uniform. Many cells contain clear vacuoles.
Figure 8–1. Histology of atypical ductal hyperpla-
sia. There is a proliferation of relatively uniform cells
with monomorphic round nuclei, in some but not all
areas. In other areas the cells maintain their orienta-
tion to each other. Thus there are some of the fea-
tures of low-grade DCIS. (Image courtesy of Maria
Braman, MD, Department of Pathology, University
of Michigan.)
Figure 8–2. Histology of atypical lobular hyperpla-
sia (ALH). Within the acini are small, uniform, evenly
spaced cells. When less than half the acini are filled,
distorted, or distended, a diagnosis of ALH is favored
over lobular carcinoma in situ. (Image courtesy of
Maria Braman, MD, Department of Pathology, Univer-
sity of Michigan.)
109
8—IDENTIFYING AND MANAGING THE HIGH-RISK PATIENT

The cells are loosely cohesive and regularly
spaced. Without disturbing the overall lobular
architecture, the cells fill and distend the acini
(Fig. 8–3). A characteristic of LCIS is pagetoid
spread, wherein the neoplastic cells extend
along adjacent ducts between intact overlying
epithelium and underlying basement mem-
brane (Fig. 8–4).
When the LCIS cells demonstrate marked
pleomorphism and distinctly larger, eccentri-
cally placed nuclei, this is a distinct entity
known as pleomorphic LCIS (PLCIS). Sometimes
these cells have a signet ring cell appearance
(Fig. 8–5). These cells are more discohesive
than in classic LCIS, and central necrosis and
calcifications, which are rarely seen with LCIS,
are more common with PLCIS. Recognition of
the pleomorphic subtype is important for sev-
eral reasons. The cellular features, necrosis,
and calcifications can often make the differen-
tiation from DCIS difficult. Frequently, PLCIS
is associated with an infiltrating pleomorphic
lobular carcinoma, which has a similar cyto-
logic appearance. Whereas the treatment para-
digm for typical LCIS is one of prevention (see
later), the treatment of PLCIS is more akin to
that of DCIS: complete excision of the lesion
with tumor-free margins and no residual calci-
fications on mammography, either via lump-
ectomy
or
mastectomy.
If
a
pleomorphic
invasive component is present, axillary staging
should also be performed. The exact role of
radiation after lumpectomy is unclear because
the natural history of PLCIS is unknown.
Overall, the diagnosis of atypical hyperplasia
implies an increased risk of 3.5 to 5.0 times
that
of
the
general
population.
The
risk
appears to be greater for ALH (RR 5.0 to 6.0)
than that of atypical ductal hyperplasia (RR
3.5 to 5.0). ALH involving both lobules and
ducts appear to have a higher risk than ALH
involving only the lobules. The risk is much
higher when atypical hyperplasia is diagnosed
in a patient with a family history of breast
cancer.
The risk of subsequent cancer is significantly
higher when LCIS is diagnosed. The risk of
developing an invasive cancer ranges from
5% to more than 30%, depending on the pop-
ulation being studied and the length of time
patients are followed. Overall, the risk of
developing subsequent invasive breast cancer
Figure 8–3. Histology of lobular carcinoma in situ.
There is a solid proliferation of small, uniform
round/cuboidal cells within the acini. (Image cour-
tesy of Maria Braman, MD, Department of Pathology,
University of Michigan.)
Figure 8–4. Pagetoid spread. The atypical cells are
extending beneath the luminal epithelium of the
extralobar ducts. This is characteristic of lobular car-
cinoma in situ. (Image courtesy of Maria Braman, MD,
Department of Pathology, University of Michigan.)
Figure 8–5. Pleomorphic lobular carcinoma in situ
(PLCIS). The cells are larger, more pleomorphic, and
eccentric. Some may have necrosis and dystrophic
microcalcifications, making differentiation from DCIS
difficult. (Image courtesy of Maria Braman, MD,
Department of Pathology, University of Michigan.)
110
SURGICAL FOUNDATIONS: ESSENTIALS OF BREAST SURGERY

is approximately 1% per year, a similar risk
that is quoted for patients with a previous his-
tory of invasive breast cancer.
Surprisingly, the subsequent cancer is more
likely to be ductal rather than lobular. Even
so, the incidence of infiltrating lobular cancer
is higher than that seen in the general popula-
tion without LCIS (25% to 37% of cancers
compared to 5% to 10% in the general popula-
tion). The subsequent cancer is also possible in
either breast. More than 50% of patients with
LCIS show multiple foci, and about 30% of
patients have further LCIS in the contralateral
breast. This is one reason why LCIS has been
considered to be only an indicator of risk
rather than a direct precursor of invasive can-
cer, and management is based on risk reduc-
tion and not surgical excision of the LCIS.
Other Factors
Cigarette
smoking
has
been
implicated
as
increasing breast cancer risk, but the relation-
ship is controversial and weak at best. Silicone
breast implants, electromagnetic fields, electric
blankets, antiperspirants, and hair dyes have
all been implicated as causes of breast cancer,
but have shown no increased risk in most
studies.
An association between antibiotic use and
breast cancer was suggested by a case control
study and received significant media atten-
tion. When cases and controls were compared,
increasing cumulative days of antibiotic use
for any condition was associated with a signif-
icantly greater risk of breast cancer. However,
there were several confounding factors includ-
ing the fact that women with greater antibiotic
use were older, had a younger age at menar-
che, were more likely to have a family history
of breast cancer, and had a higher use of hor-
mone replacement. There may also have been
a detection bias at play. It is unclear biologi-
cally how antibiotic use (or infections) may
be associated with breast cancer and the clini-
cal significance of these data is uncertain.
Clearly, antibiotics should not be withheld
among
high-risk
women
when
indicated,
although
careful
consideration
should
be
given to prescribing antibiotics in the absence
of a clear indication.
Viruses have been linked to other solid
tumors (nasopharyngeal cancer, cervical can-
cer, hepatocellular carcinoma) and have been
suggested as a possible cause of breast cancer.
Fragments of the Epstein-Barr virus genome
and sequences similar to the mouse mammary
tumor virus (MMTV) have been identified in
invasive breast cancers, although the signifi-
cance of this finding is unclear. Further research
is needed to determine whether viruses play a
role in breast cancer etiology.
Exposure to ionizing radiation at a young age is
associated with an increased risk of breast
cancer. Women may have been exposed for
therapeutic purposes (Hodgkin’s lymphoma,
thymic enlargement) or accidental exposure
(nuclear fallout). Women are most vulnerable
during the prepubertal years (10–14) but even
women in their 40s who are exposed to radia-
tion have an increased risk. The low radiation
associated with diagnostic imaging and the
older age at which imaging studies are usually
obtained make it highly unlikely that diag-
nostic x-rays increase breast cancer risk. Electro-
magnetic radiation exposure has been suggested as
a risk factor for breast cancer. Most women are
exposed to electromagnetic radiation from power
lines, transformer substations, or use of electrical
appliances. Residential magnetic field exposure
does not appear to increase breast cancer risk.
The link between melatonin and estrogen has
been proposed as a risk factor for breast cancer,
specifically women who are awake at night.
Melatonin is produced by the pineal gland
and exposure to light at night suppresses
this nocturnal process. This might result in
increased estrogen exposure. Three studies
have shown an increased risk of breast cancer
among women who work at night or who
did not typically sleep between 1 AM and 2 AM
(when melatonin levels are highest). Again,
the exact association between nocturnal light
exposure and breast cancer has not been eli-
cited and the strength of the association is
variable.
Statistical Models to Estimate
the Risk of Breast Cancer
It is apparent that with so many different fac-
tors
potentially
related
to
an
individual
woman’s risk of developing breast cancer, it
may be difficult to counsel patients. Models
of risk prediction have been created using the
strongest
risk
factors.
These
were
created
before genetic testing for BRCA1 and BRCA2,
but may still be useful for women who have
not undergone genetic testing. Other models
are available that may predict a woman’s risk
of harboring a BRCA mutation, such as the
BRCAPRO model. Each model is based on
111
8—IDENTIFYING AND MANAGING THE HIGH-RISK PATIENT

slightly different data and study designs, so
they
may
result
in
slightly
different
risk
estimations.
Gail and Claus Models
Gail and colleagues, using data from more
than 2800 patients and more than 3000
matched controls, created the Gail model for
unaffected individuals with a limited family
history. This model boils down to five vari-
ables: current age, age at first live birth, age at
menarche,
number of first-degree
relatives
with breast cancer, and number of prior breast
biopsies.
Tables
and
computer
programs
(including several on the Internet) are avail-
able to estimate the 5-year and lifetime risks
of developing breast cancer. The Claus model
is based on data from more than 4300 breast
cancer cases and 4600 control cases and is
based primarily on age and family history.
Even though these models are widely used,
particularly the Gail model, they are not with-
out limitations. They are not relevant for
women with AH or LCIS. The Gail model does
not include information on race, which can
impact breast cancer risk (see Chapter 22)
and so may overestimate the risk among Afri-
can-Americans or Asian-Americans. It does
not take into account the age at diagnosis of
the relative’s breast cancer, so the woman
whose mother had breast cancer at age 70
has the same reported risk as the woman
whose mother developed the disease at age
30. By not including second-degree relatives,
it overlooks the possible contribution from
the paternal side of the family. The Claus
model does include second-degree relatives
and the age of onset, but does not include
nonfamilial risk factors such as hormonal fac-
tors. Neither model takes into account bilateral
breast cancer or ovarian cancer.
The newest model is the Tyler-Cuzick model.
This model not only takes into account many
of the relevant details of family history and
hormonal factors, but also includes BMI and
the presence of LCIS. A computerized version
of this model is not yet available.
What Can I Do to Decrease
My Risk?
Many of the factors related to an increased
risk of breast cancer cannot be altered. A
woman cannot change her parents or genes.
While it is possible to delay the onset of men-
arche (through increasing physical activity,
altering the diet, or using LH-releasing ago-
nists), this seems a rather extreme approach
to decreasing risk and offers little to most
women who explore risk reduction strategies
in their adult years. Likewise, having a child
before the age of 25 years will decrease breast
cancer risk, but the decision whether and
when to have a child is influenced by many
factors.
There are some changes that women can
make to decrease their risk of developing
breast cancer (Table 8–3). Their relative impact
varies, and some are only theoretical.
Lifestyle Changes
Breast-feeding may decrease the risk of devel-
oping
breast
cancer,
although
supporting
data are not definitive. Given the benefits
of
breast-feeding
for
the
infant,
however,
there is little reason not to encourage breast-
feeding.
TABLE 8–3
Factors That May
Be Modified to Decrease the Risk
of Breast Cancer
Factor
Comments
Pregnancy before age
25 years or multiple
pregnancies
20% to 30% decreased
risk but family
planning should be
based on other
factors
Breast-feeding
Small decrease in risk
but multiple benefits
to the baby
Stopping or avoiding oral
contraceptives
Impact unclear
Stopping or avoiding
hormone replacement
therapy
Balanced against
symptoms
Stopping or minimizing
alcohol intake
Exercise
Maintaining a healthy
body weight
Maintaining a low-fat
diet
Consuming low-fat dairy
products, calcium, and
vitamin D
Impact unclear
Consuming fruits and
vegetables
Impact unclear
Soy products
Impact unclear
112
SURGICAL FOUNDATIONS: ESSENTIALS OF BREAST SURGERY

Alcohol
consumption
has
been
clearly
shown to increase the risk of breast cancer,
and limiting alcohol intake is an effective
way to reduce risk (in addition to other health
benefits). For those who do drink, adding folic
acid to the diet may counteract the effects of
alcohol on breast cancer and mitigate this risk.
Even though a woman may not have con-
trol over when she goes into menopause, she
does have a choice as to whether she takes
postmenopausal estrogens. HRT is effective at
reducing the symptoms associated with meno-
pause, but it does increase the risk of breast
cancer. Menopause symptoms can be managed
using nonhormonal therapies, such as citalo-
pram, fluoxetine, venlafaxine, or gabapentin.
Dietary changes, although difficult to imple-
ment for many women, can decrease breast can-
cer risk. A healthy body weight should be
maintained through both diet and moderate
exercise. These have been clearly associated with
a decreased risk of breast cancer, although the
mechanisms still need further investigation.
Data are less clear when one looks at specific
foods or diets. Some epidemiologic studies
have noted a lower risk of breast cancer among
patients who eat a relatively large amount of
fruits and vegetables, although a direct con-
nection is difficult to show. Soy consumption
has also been linked to a decreased risk of
breast cancer, but this likely needs to be imple-
mented at a young age. There is no strong evi-
dence that supplementing the typical Western
diet with soy-based products will impact risk.
Low-fat diets have been suggested to not only
decrease the risk of breast cancer but also
decrease the risk of recurrence after treatment
for
breast
cancer
patients.
However,
this
involves lowering fat levels far below that of
the average American woman. Intake of low-
fat dairy products may exert a protective effect,
especially in premenopausal women. In the
Nurses’ Health Study, an inverse association
between breast cancer risk and intake of low-
fat dairy products, calcium, and vitamin D
was
suggested
for
premenopausal
women.
The exact impact that this may have on breast
cancer risk is unclear, but again, it has other
health benefits and should be encouraged.
Chemoprevention
Aspirin
Because nonsteroidal antiinflammatory drugs
(NSAIDs), through their inhibition of COX-2,
have been shown to inhibit polyp formation
and the development of colon cancer, there was
hope that a similar effect would be seen in breast
cancer. Whereas a modest benefit has been sug-
gested by population-based studies, the results
are not clear and are difficult tointerpret. In addi-
tion, data from the Nurses’ Health study showed
no protective effect of NSAIDs’ use on breast can-
cer risk. At this time, there is no evidence to sug-
gest that aspirin use will decrease breast cancer
risk, although there are other health benefits
associated with daily low-dose aspirin.
Tamoxifen
Tamoxifen was originally synthesized in the
mid-1960s during a search for oral contracep-
tives and fertility agents. It failed at both, but
it was found to bind to the estrogen receptor
and halted the growth of malignant breast
cancer cells in vitro. Further studies revealed
that in both normal breast tissue and in breast
cancer, tamoxifen could act as an estrogen
antagonist, occupying the estrogen receptor
and blocking the effects of estrogen. This
results in a decrease in epithelial cellular prolif-
eration
(Box
8–1).
Ultimately,
randomized
clinical trials established a role for tamoxifen
in the treatment of metastatic disease and as
an adjuvant treatment for breast cancer. (See
Chapter 17, Hormonal Therapy). It was noted
in these trials that the risk of a contralateral
breast cancer was decreased by 40% to 50%,
leading to the idea that tamoxifen might be
used for breast cancer prevention. This led to
four large trials of breast cancer chemopreven-
tion using tamoxifen.
The Royal Marsden Hospital Prevention Trial ran-
domized 2471 healthy women with a family his-
tory of breast cancer to receive tamoxifen or a
placebo. After a median of 70 months, there
was no difference in the incidence of breast can-
cer between the two groups (34 for tamoxifen
and 36 for placebo).
BOX 8–1 MECHANISMS BY WHICH
TAMOXIFEN DECREASES EPITHELIAL
CELLULAR PROLIFERATION


Reduces in transforming growth
factor-alpha (TGF-a)


Reduces insulin-like growth factor-I
(ILGF-1)


Stimulates transforming growth
factor-beta (TGF-b)


Binds to calmodulin


Inhibits protein kinase activity
113
8—IDENTIFYING AND MANAGING THE HIGH-RISK PATIENT

The Italian Tamoxifen Prevention Study rando-
mized 5408 women to receive tamoxifen or
placebo. These patients were at low to normal
risk of breast cancer. Overall, there was no dif-
ference in the development of breast cancer
between the two groups (45 and 34, respec-
tively). One subset that did appear to benefit
was the group of women who were taking hor-
mone replacement therapy. There was a signif-
icant reduction in breast cancer cases among
women taking HRT and tamoxifen compared
with women taking HRT and placebo (6 vs.
17, P ¼ .02).
The International Breast Cancer Intervention
study (IBIS-I) randomized 7152 women with a
family history of breast cancer (either one first-
degree relative with bilateral breast cancer or
onset before age 50, or two or more relatives with
breast cancer) to receive tamoxifen or placebo.
With a median follow-up of 50 months, there
were significantly less breast cancers among
tamoxifen users than women receiving placebo
(69 vs. 101, P ¼ .013). Although this translated
to a risk reduction of 32%, there was an excess
of deaths in the tamoxifen group (25 vs. 11).
The thromboembolic risk was increased 2.5-fold
with tamoxifen.
The largest study of tamoxifen for chemo-
prevention is the Breast Cancer Prevention Trial
(NSABP P-01) (Table 8–4.) Between June 1,
1992, and September 30, 1997, a total of
13,388 women were randomized to receive
tamoxifen (20 mg/day) or placebo. High risk
was defined as (1) age older than 60 years, (2)
a Gail model 5-year estimated risk of at least
1.66%, or (3) 35 years of age or older with a
history of LCIS. After a median follow-up of
48 months (the study was stopped early when
TABLE 8–4
Average Annual Rates for Outcomes in the NSABP Breast Cancer
Prevention Trial by Patient Characteristic
Invasive Breast CA per 1000 Women
Tamoxifen
Placebo
Risk Ratio (95% CI)
Age
<49
3.8
6.7
0.56 (0.37-0.85)
50 to 59
3.1
6.3
0.49 (0.29-0.81)
>60
3.3
7.3
0.45 (0.27-0.74)
LCIS
5.7
13.0
0.44 (0.16-1.06)
Atypical hyperplasia
1.4
10.1
0.14 (0.03-0.47)
# Relatives with breast CA
0
3.0
6.4
0.46 (0.24-0.84)
1
3.0
6.0
0.51 (0.35-0.73)
2
4.8
8.7
0.55 (0.30-0.97)
3 or more
7.0
13.7
0.51 (0.15-1.55)
Gail model 5 year risk
<2
2.1
5.5
0.37 (0.18-0.72)
2.01-3.0
3.5
5.2
0.68 (0.41-1.11)
3.01-5.0
3.9
5.9
0.66 (0.39-1.09)
>5.0
4.5
13.3
0.34 (0.19-0.58)
Complications per 1000 Women
Invasive endometrial CA <49
1.3
1.1
1.21 (0.41-3.6)
Invasive endometrial CA >50
3.0
0.8
4.01 (1.7-10.90)
Fractures
4.3
5.3
0.81 (0.63-1.05)
Stroke
1.4
0.9
1.59 (0.93-2.77)
Transient ischemic attack
0.7
1.0
0.76 (0.40-1.44)
Pulmonary embolism
0.7
0.2
3.01 (1.15-9.27)
Deep vein thrombosis
1.3
0.8
1.60 (0.91-2.86)
Data from Fisher B, Constantino JP, Wickerham DL, et al. Tamoxifen for prevention of breast cancer: Report of the
NSABP P-1 Study. J Natl Cancer Inst 1998;90:1371–1388.
CA, Cancer; LCIS, lobular carcinoma in situ.
114
SURGICAL FOUNDATIONS: ESSENTIALS OF BREAST SURGERY

an interim analysis found statistical signifi-
cance), it was found that tamoxifen decreased
the relative risk of breast cancer by 49%. The
reduction was entirely in ER-positive tumors;
there was no change in the occurrence of ER-
negative tumors. The reduced risk was seen
among all age groups in the trial (0.56 for
women 49 years or younger; 0.49 for women
50 to 59; 0.45 for women 60 years or older).
Although not as dramatic as with the IBIS-I
study,
there
were
significant
increases
in
adverse effects with tamoxifen. Most were in
women older than 50 years of age. These
included
increases
in
endometrial
cancer,
stroke, pulmonary embolus (PE), and deep
vein thrombosis (DVT).
When the data from all four trials are com-
bined, the overall risk reduction associated
with
tamoxifen
is
approximately
32%
to
38%. However, the differences in the studies
make it difficult to compare them. The Royal
Marsden and IBIS trials were limited to women
with a strong family history, the P-01 trial had
a broad range of what was defined as high-risk,
and the Italian study was limited to low to
normal risk. There are also variations in sam-
ple size and compliance. The P-01 trial was
the largest and had more power to detect a
small difference; however, it has been criti-
cized for including patients who developed
tumors shortly after initiation. Thus some of
the benefit demonstrated may have been in
treating occult primary tumors rather than
preventing cancers.
Who Should Be Considered for
Tamoxifen Chemoprevention?
Tamoxifen has been approved for the preven-
tion of breast cancer for women at high risk.
However, the potential benefits of tamoxifen
must be weighed against the risks on an indi-
vidual basis (Box 8–2). The two most worri-
some adverse events are endometrial cancer
and thromboembolic events. These occurred
predominantly in women older than age 50.
The relative risk of endometrial cancer in the
P-01 trial was 2.53. All were localized (stage I)
and no deaths from the disease occurred. In
addition, several cases of uterine sarcoma have
been seen with long-term follow-up among
the tamoxifen uses (compared to no cases in
the control group). The relative risk of throm-
bolic
events
was
also
increased
in
older
women, with risk ratios of 1.59, 3.01, and
1.60 for stroke, pulmonary embolism, and
DVT,
respectively.
Therefore
the
risks
of
tamoxifen are going to be higher in older
women, particularly those who have not had
a hysterectomy. Women with a history of
stroke, transient ischemic attack (TIA), DVT,
or PE should not use tamoxifen. Tamoxifen
also has other side effects to consider. There
was an increase of approximately 14% in the
rate of cataract development compared to pla-
cebo, so women with a history of cataracts or
cataract surgery should also avoid tamoxifen
use. Hot flashes and vaginal discharge also
increased significantly with tamoxifen use.
Women with LCIS, ALH, or ADH
Women with LCIS have an annual risk of
invasive breast cancer of more than 1% per
year. In the P-01 trial, 6% of the participants
had LCIS and 9% had atypical hyperplasia.
A benefit was seen in both women with LCIS
(RR 0.44, 95% CI 0.16 to 1.06) and atypical
hyperplasia (RR
ratio
0.14, 95% CI 0.03
to 0.47).
Women with a Family History
of Breast Cancer
Women with a strong family history should be
counseled by a medical geneticist or genetic
counselor about the appropriateness of genetic
testing. If they are found to harbor a BRCA1 or
BRCA2 mutation, tamoxifen is one option for
preventing breast cancer (see later). In women
who do not choose to be tested or who do not
possess a mutation, tamoxifen is an appropri-
ate consideration for reducing risk. Data from
the P-01 trial did demonstrate a risk reduction
even in women with 3 or more immediate
relatives with breast cancer.
BOX 8–2 WOMEN FOR WHOM
THE BENEFIT OF TAMOXIFEN APPEARS
TO OUTWEIGH THE RISK


Women younger than age 50 who
have a projected 5-year risk of invasive
breast cancer of between 1.5% and 7%


Women ages 50 to 59 years who have a
uterus and a projected 5-year risk of
invasive breast cancer of 6.0%


Women age 50 to 59 years without a
uterus who have a projected 5-year risk
of invasive breast cancer of 3.0%


Women ages 60 to 69 years without a
uterus who have a projected 5-year risk
of invasive breast cancer of 5.5%
115
8—IDENTIFYING AND MANAGING THE HIGH-RISK PATIENT

Women with BRCA1 and BRCA2
Mutation
Women with a mutation in BRCA1 or BRCA2
genes are at the highest risk of developing
breast cancer. Unfortunately, the data are not
clear as to the benefit of tamoxifen in this pop-
ulation. A case control study looking at tamox-
ifen use in 593 breast cancer patients with
either a BRCA1 or BRCA2 mutation showed a
decreased risk of contralateral breast cancer
with
tamoxifen.
However,
data
from
the
NSABP P-01 trial suggest that a preventive
benefit might only be in women with a BRCA2
mutation. Whereas there was a 62% reduction
in the incidence of breast cancer in women
with BRCA2 mutations, there was no reduc-
tion
in
women
with
BRCA1
mutations.
Although these are based on very small num-
bers, in that genetic testing was performed on
only a small fraction of the patients, these
findings may be due to the fact that only a
small percentage of tumors in BRCA-1 carriers
are ER positive (17% to 24%) compared to
BRCA-2 carriers (76% to 78%).
Women with a High Risk of Breast
Cancer Based on Their Gail Model
As described earlier, the Gail model takes into
account
several
factors
when
assessing
a
woman’s risk of breast cancer, including age,
age at menarche, age at first live birth, number
of first-degree relatives with breast cancer, and
the number of breast biopsies. High risk is typ-
ically defined as at least a 1.66% chance of
developing breast cancer in the next 5 years.
However, the benefits of tamoxifen in this
population
must
be
weighed
against
the
potential risks.
Raloxifene
Raloxifene (Evista) resulted from efforts to
design selective estrogen receptor modulators
(SERMs) that have estrogen activity in bone
and cardiovascular tissue but not in the repro-
ductive tissues, for the purpose of lowering the
risk of osteoporotic fractures. The Multiple Out-
comes of Raloxifene Evaluation (MORE) trial
randomized
7705
postmenopausal
women
with osteoporosis (and no history of breast or
endometrial cancer) to either receive one of
two doses of raloxifene or placebo. Although
the trial was designed to look at osteoporosis,
it was found that with 4 years of follow-up,
there was an 84% relative reduction in the inci-
dence of ER-positive breast cancer. There was no
effect on ER-negative breast cancer. As with
tamoxifen, raloxifene was associated with an
increased risk of venous thromboembolic dis-
ease, but it was not associated with an increased
risk of endometrial cancer.
These data prompted the Study of Tamoxifen
and Raloxifene (STAR) trial, which rando-
mized postmenopausal women who are high-
risk for the development of breast cancer to
receive either tamoxifen or raloxifene. One of
the largest breast cancer prevention studies
ever, STAR took place at more than 500 centers
across the United States, Canada, and Puerto
Rico. Initial results of STAR show that the drug
raloxifene is as effective as tamoxifen in reduc-
ing the breast cancer risk of the women in the
trial. In STAR, both drugs reduced the risk of
developing invasive breast cancer by about
50%. In addition, within the study, women
who were assigned to take raloxifene daily
had 36% fewer uterine cancers and 29% fewer
blood
clots
than
the
women
who
were
assigned to take tamoxifen.
Aromatase Inhibitors
Aromatase inhibitors reduce circulating estro-
gen by approximately 90% in postmenopausal
women and have been shown to be beneficial
in the adjuvant therapy of breast cancer (see
Chapter
17).
The ATAC
trial
(anastrazole,
tamoxifen, alone and in combination) rando-
mized postmenopausal women with early-
stage breast cancer to tamoxifen, anastrazole,
or the combination. Anastrozole was superior
to tamoxifen in prolonging disease-free sur-
vival. It also reduced the incidence of contra-
lateral
primary
invasive
breast
cancers.
However, additional data from trials looking
specifically at aromatase inhibitors as chemo-
prevention are needed before this class of
drugs can be used in this manner.
Surgery
Prophylactic Mastectomy
Bilateral prophylactic mastectomy is the most
logical preventive measure, although it is also
the most extreme. Prophylactic mastectomy
should be a simple mastectomy, incorporating
the nipple-areola complex and the breast tissue
extending into the axilla, but not the axillary
lymph nodes. Subcutaneous mastectomy is gene-
rally not thought to be appropriate as cancer
can form in the nipple-areolar complex, but
studies are examining the potential of nipple
and areolar sparing mastectomies. Immediate
reconstruction is appropriate and may reduce
116
SURGICAL FOUNDATIONS: ESSENTIALS OF BREAST SURGERY

the psychological stress associated with the
procedure. It must be stressed that mastec-
tomy does not remove all mammary tissue.
Residual mammary tissue can be found in
the axilla and pectoralis fascia. Therefore it
does not provide 100% protection against
breast
cancer.
Studies
of
both
high-risk
women
and
breast
cancer
patients
have
shown that bilateral mastectomy reduces the
risk by 90% to 95%.
While there is a clear reduction in the risk of
developing breast cancer (possibly for a second
time), the fact that even among high-risk
women, most will not develop cancer and
the cancers they do develop are usually cur-
able, bilateral prophylactic mastectomy may
not improve overall survival. This varies with
the age and health of the patient, the risk of
cancer, and the risk of systemic disease among
patients already diagnosed with breast cancer.
Thus a careful discussion needs to be held with
the patient regarding what she can expect, and
not expect, from bilateral mastectomy. The
benefits of bilateral prophylactic mastectomy
must be balanced against the physical and psy-
chological consequences.
Prophylactic Oophorectomy
Observational
studies
have
suggested
that
bilateral oophorectomy among premenopau-
sal women reduces their risk of breast cancer.
Even though this is probably not a reasonable
intervention for the typical high-risk woman,
it is a reasonable consideration for women
with a heritable predisposition to breast can-
cer. Women with a BRCA1 and BRCA2 muta-
tion are at an increased risk of not only
breast cancer but also ovarian cancer. One
study suggests that bilateral oophorectomy in
women with a BRCA mutation will signifi-
cantly reduce the risk of developing a breast
or gynecologic cancer. This should be strongly
considered for premenopausal women known
to have a BRCA mutation and who have com-
pleted child bearing. It should also be consid-
ered in postmenopausal women with a BRCA
mutation for the ovarian cancer benefit, but
it is unclear what impact this might have on
their subsequent breast cancer risk.
Management of the Patient
with Lobular Carcinoma
In Situ
The true incidence of LCIS in the population is
unknown but it is discovered incidentally on
approximately 0.5% to 3.8% of benign breast
biopsies. Management is based on the fact that
LCIS is not considered a malignant or prema-
lignant lesion, but rather an indicator lesion
for the risk of subsequent invasive carcinoma.
The presence of LCIS in a breast biopsy speci-
men is one of the strongest risks currently
known for the subsequent development of
invasive carcinoma. Efforts to identify features
of LCIS that are associated with a higher likeli-
hood of invasive disease have been unsuccess-
ful. Therefore all women with LCIS must be
considered high risk.
The management of atypical hyperplasia or
LCIS is twofold. The first step is to rule out the
presence of an invasive cancer. When ADH,
ALH, or LCIS is diagnosed by a core-needle
biopsy, a wire-localized excisional biopsy is
recommended to establish definitive histologic
results. Often the diagnosis is upgraded to DCIS
or invasive cancer (7% to 17% of cases). If an
excisional biopsy demonstrates LCIS extending
to the margins, however, reexcision is not nec-
essary to obtain negative margins. Bilateral
mammograms
should
be
reviewed
with
increased suspicion, and biopsy of any addi-
tional suspicious lesions should be undertaken.
Once it is confirmed that the patient does
not have DCIS or invasive cancer, a discussion
must be held with the patient regarding her
individual risk and prevention strategies. It
often takes some time to explain the nuances
of LCIS. It is important to explain why unilat-
eral mastectomy or wide surgical excision with
histologically negative margins and radiation
therapy are not needed. Patients often come
to the office having been told they have can-
cer, and they are confused when informed oth-
erwise. Some surgeons and pathologists prefer
the term “lobular neoplasia” to “LCIS” to
decrease anxiety of patients and help avoid
confusion, but LCIS remains the most com-
monly used name.
Patients diagnosed with LCIS have the fol-
lowing choices (Box 8–3):
1. Observation. As one might do for any
woman known to have an increased risk
of developing breast cancer, observation
is a reasonable option. Women should
be counseled that they have an approxi-
mately 1% per year risk of developing
cancer. Attempts to further stratify this
risk, such as by the extent of LCIS in the
biopsy specimen, have been unsuccess-
ful, although women with LCIS and a
first-degree relative with breast cancer
may have an even higher risk. In the
117
8—IDENTIFYING AND MANAGING THE HIGH-RISK PATIENT

absence of intervention, it is important
for these women to undergo appropriate
screening for breast cancer. At a mini-
mum, this should include monthly self
breast examinations, clinical examina-
tions at 6-month intervals, and annual
mammograms. There is no evidence that
more
frequent
mammography
is
of
benefit. Are there other more intensive
options for screening for the woman with
LCIS? MRI has recently been shown to
increase
the
detection
of
cancer
in
women at high risk based on known
genetic defects or strong family history
(see Chapter 2). However, it has not been
studied in the setting of LCIS, and it is
important to note that MRI will detect
many benign lesions. MRI-guided biopsy
is still not available at many institutions,
so a woman who has LCIS and an MRI-
detected abnormality could find herself
in a difficult situation. For now, MRI as
a screening examination in women with
LCIS cannot be routinely recommended.
Another surveillance examination being
touted for high-risk women is ductal
lavage. Ductal lavage detects epithelial
atypia among ductal cells collected from
microcatheter lavage. Although this may
someday be a promising tool for women
at risk for invasive breast cancer, the role
of ductal lavage in the surveillance of
women with LCIS is uncertain.
2. Bilateral prophylactic mastectomy. A sec-
ond and more drastic option for women
with LCIS is bilateral simple mastectomy,
usually with immediate reconstruction.
Treatment
strategies
addressing
one
breast are not recommended because the
risk of cancer in the contralateral breast
is roughly the same as the risk in the
biopsied
breast.
With
the
significant
reduction in invasive breast cancer with
tamoxifen treatment, recent trends of
fewer women undergoing mastectomy
to treat LCIS seem appropriate. However,
some women may opt for prophylactic
surgery.
3. Tamoxifen
or
Raloxifene.
Before
the
advent of tamoxifen as chemopreven-
tion, women with LCIS had only two
options: continued observation or bilat-
eral
mastectomy.
Today,
the
use
of
tamoxifen in women with LCIS is worthy
of serious consideration. In the NSABP P1
Breast Cancer Prevention Trial, the rate of
development of invasive breast cancer in
women with LCIS was significantly less
in the tamoxifen-treated group compared
to the controls (5.69 versus 12.99 per
1000 women). The 56% risk reduction
was slightly higher than that seen in
women at risk for breast cancer based on
other factors. Raloxifene is also an option
for postmenopausal women. This is not
to say that tamoxifen or raloxifene is
ideal for all women with LCIS. Many
women choose not to take tamoxifen
because of the adverse effects, which can
include hot flashes (an estrogen antagonist
effect), endometrial hyperplasia, fibroids,
and
an
increased
risk
of
endometrial
cancer
and
venous
thromboembolism.
Furthermore, women who receive tam-
oxifen
need
to
be
seen
for
ongoing
follow-up every 3 months, with regular
screening blood work to evaluate for the
possibility of adverse effects; this is not
necessary in women who undergo obser-
vation alone.
Management of the Patient
with a BRCA Mutation
Recent estimates of breast cancer risk would
suggest that a woman with a BRCA1 mutation
has a 90% risk of breast cancer and a 24%
risk of ovarian cancer by age 80. A woman
with a BRCA2 mutation has a 40% risk of
breast cancer and an 8% risk of ovarian cancer.
BOX 8–3 OPTIONS FOR WOMEN
DIAGNOSED WITH ATYPICAL
HYPERPLASIA OR LOBULAR
CARCINOMA IN SITU


Careful clinical follow-up is
appropriate for most women with LCIS;
the risk of invasive breast cancer is
approximately 1% per year.


Tamoxifen reduces the risk of invasive
breast cancer by 55% and should be
considered. Raloxifene may also be
considered for postmenopausal women.


Bilateral prophylactic mastectomy,
usually with reconstruction, is an
alternative approach for women
unwilling to undergo careful observation
or tamoxifen therapy.


Excision to negative margins, radiation
therapy, and chemotherapy have no role
in treatment of women with LCIS.
118
SURGICAL FOUNDATIONS: ESSENTIALS OF BREAST SURGERY

The rates of survival for these cancers are
roughly similar to those of cancers in non-
BRCA mutation carriers.
Management of these patients is difficult
because several issues must be considered and
the exact risk of breast cancer or ovarian can-
cer for any individual BRCA mutation carrier
is difficult to assess. Counseling is of the
utmost importance, taking into account the
family history, personal medical history of
the patient, ethnic group, and specific muta-
tion. Family members must also be counseled
regarding recommendations for testing and
screening. A thorough discussion of the risks
and benefits of increased surveillance, chemo-
prevention, and prophylactic surgery is neces-
sary. It is quite unlikely that a breast surgeon
has the time or expertise to discuss all of these
issues with the patient, which is why genetic
counseling is so important at this juncture.
The surgeon must work with not only the
genetic counselor, but also the medical oncol-
ogist (if chemoprevention is an option), the
plastic surgeon (if bilateral mastectomy is
considered), and the gynecologist (if bilateral
salpingo-oophorectomy
is
considered)
to
determine the best strategy for each individual
patient.
Increased Surveillance
Breast Examination
There are minimal data either supporting or
refuting the importance of self breast exami-
nation for women with a hereditary risk of
breast cancer. For women at average risk, self
breast examinations do not appear to be asso-
ciated with a decrease in breast cancer mortal-
ity but are associated with an increase in
breast biopsies. Clinical breast examination
will detect additional cancer beyond mammo-
graphic screening, although the numbers are
small; roughly 3% to 8% of cancers were
detected solely by clinical examination. Even
among women with a hereditary risk, clinical
examination detects only a small number of
cancers above and beyond imaging. However,
there are benefits to both self breast examina-
tion and clinical breast examination beyond
cancer detection and most cancer societies
continue to recommend their use for all
women, not just women with a hereditary
predisposition. There is minimal evidence to
support high-risk women undergoing a clini-
cal breast examination more often than every
6 months.
Mammogram
As opposed to breast examination, screening
mammograms have been clearly shown to
increase breast cancer detection and decrease
breast cancer mortality (see Chapter 2). How-
ever, mammographic screening among BRCA
carriers has several limitations compared to
the
screening
in
the
general
population.
Patients are generally younger than 40 years
of age, in whom mammographic screening is
less sensitive due to breast density, and it is
common for these patients to have interval
cancers (cancers that become clinically appar-
ent between yearly mammograms). Specialized
mammogram
screening
programs
among
women with a familial risk have failed to dem-
onstrate either an improved survival or a
decreased stage at presentation, and the sensi-
tivity
of mammograms specifically
among
BRCA mutation carriers has been questioned.
Even so, yearly mammography is recom-
mended for any woman with a known BRCA
mutation who has not opted for bilateral mastec-
tomy, beginning between the ages of 25 and
30. Digital mammography is recommended
because studies have suggested a higher accuracy
among women with dense breast tissue. At this
time, there is no evidence to suggest that mam-
mography more frequently than once a year is
of added benefit.
Ultrasonography
Ultrasound is typically thought of as a comple-
ment to mammography; a tool to further scru-
tinize a palpable mass or abnormality seen on
screening mammogram. Whole-breast screen-
ing ultrasonography does not yield a high
enough rate of breast cancer detection to rec-
ommend its routine use. Even among high-
risk patients, it is of questionable worth. It will
detect some mammographically occult can-
cers, and for women who do not have access
to MRI, may be worth adding to yearly mam-
mography. However, in women undergoing
screening MRI (see later), the addition of ultra-
sonography is of minimal value.
Magnetic Resonance Imaging
The use of MRI in the management of breast
disease is controversial and still evolving, but
one area where MRI has documented superior-
ity is in the screening of the high-risk woman.
Prospective studies of high-risk patients, includ-
ing many women with BRCA mutations, have
119
8—IDENTIFYING AND MANAGING THE HIGH-RISK PATIENT

shown that MRI more than doubles the num-
ber of cancers detected using routine physical
examination, mammography, and ultrasonog-
raphy. Rates of interval cancers are less than
10% when MRI is included in the screening.
Despite the high sensitivity of MRI, it is not
perfect, and both in situ and invasive cancers
may be missed. Thus MRI should be performed
with mammograms and should not be consid-
ered a substitute. Another problem with MRI
has been the specificity. MRI will lead to the
further evaluation and biopsy of a number of
benign lesions that would not have been
picked up on mammography and ultrasound.
Specificity has varied in these studies, but has
ranged from 81% to 97%. This variable speci-
ficity will affect the positive predictive value
(PPV). However, PPV is dependent upon the
frequency of the disease in question among
the population being studied. Therefore the
PPV is acceptable for high-risk women but sig-
nificantly lower and less likely to be cost-
effective among average risk women. Thus
MRI screening should be limited to patients
at high risk, especially those with BRCA muta-
tions. At this time, yearly MRI as an adjunct
to physical examination and mammography
is recommended, although further research is
necessary
to
define
the
most
appropriate
intervals, and the ideal population for MRI
screening including what age to begin and
what age to stop MRI screening.
The National Comprehensive Cancer Net-
work (NCCN) has recommended that women
with a genetic high risk of breast cancer be
counseled to perform self breast examination
on a monthly basis starting at age 18. Clinical
breast examination should be performed every
6 months starting at age 25. Mammography
and MRI should be performed yearly starting
at age 25 (Box 8–4).
Other
BRCA mutations predispose patients to non–
breast malignancies as well. Screening for
other BRCA-associated cancers (ovarian, pros-
tate,
pancreas)
is
recommended,
although
there are little data to suggest a survival benefit
or the ideal screening protocols (beyond those
recommended for the general population).
Although popular, screening BRCA mutation
carriers for ovarian cancer through serum CA-
125
measurements
and
transvaginal
ultra-
sounds has not been shown to downstage
ovarian tumors or improve survival. Likewise
there are no data to support serum CA19-9
measurements
or
abdominal
CT scans
to
screen for pancreatic cancer outside of a clini-
cal trial. The benefit of mammography for
male BRCA2 carriers is also unknown and not
recommended, although these patients should
be examined regularly by their internists for
signs of breast cancer. Male carriers should also
be monitored for signs of prostate cancer.
Risk Reduction Strategies
Tamoxifen
Five years of tamoxifen clearly decreases the
risk of breast cancer, decreasing the incidence
by 43% in the Breast Cancer Prevention Trial.
Raloxifene will provide a similar risk reduc-
tion, with a lower risk of uterine cancer or
venous thromboembolism among postmeno-
pausal women. It is unclear, however, if this
same degree of benefit can be expected among
BRCA mutation carriers. A subgroup analysis
of the Breast Cancer Prevention Trial failed to
demonstrate a benefit, but the numbers were
too small to exclude this possibility. Tamoxi-
fen does appear to decrease the risk of contra-
lateral cancers among BRCA patients treated
for breast cancer.
Bilateral Prophylactic Mastectomy
Bilateral
prophylactic
mastectomy
(BPM)
reduces the risk of breast cancer by at least
90% among carriers of a BRCA mutation, and
it should be presented and discussed with
any affected patient. The exact benefits of
bilateral mastectomy on survival are unclear.
Even though mathematical models would sug-
gest a survival benefit of bilateral mastectomy
BOX 8–4 INCREASED SCREENING
STRATEGY FOR WOMEN WITH A
KNOWN BRCA1/BRCA2 MUTATION
WHO DO NOT DESIRE RISK-
REDUCING SURGERY


Monthly breast self-examination
beginning at age 18.


Clinical breast examination every 3 to
6 months beginning at age 25.


Annual mammography and magnetic
resonance imaging study beginning at
age 25.


Twice yearly ovarian cancer screening
with ultrasound and serum CA-125 levels
beginning at age 35.
120
SURGICAL FOUNDATIONS: ESSENTIALS OF BREAST SURGERY

to routine surveillance, the degree of that
benefit compared with oophorectomy and/or
surveillance with MRI is not clear. Immediate
reconstruction is strongly recommended to
minimize the psychological trauma associated
with bilateral mastectomy. Patients still need
close follow-up because this prophylactic mea-
sure does not completely eradicate the risk of
breast cancer.
Risk-Reducing Salpingo-Oophorectomy
Bilateral salpingo-oophorectomy (BSO) has not
been evaluated in randomized trials, but both
retrospective and prospective data strongly sup-
port its use. BSO can result in an 80% to 96%
reduction in the risk of a BRCA-associated gyne-
cologic cancer. In addition, it can reduce the
risk of breast cancer by almost 50%. Certainly
there are drawbacks; it will prematurely put
young women into menopause with the asso-
ciated symptoms and diminished quality of life.
In addition, it should be delayed until after the
woman has completed child bearing, although
this can be a difficult compromise in a younger
woman, particularly one who is not married.
Women who undergo BSO are still at a small
risk of primary peritoneal carcinoma, estimated
at about 0.2% annually.
Suggested Readings
1. Ahlgren M, Melbye M, Wohlfahrt J, Sorensen TIA.
Growth patterns and the risk of breast cancer in
women. N Engl J Med 2004;351:1619–1626.
2. American Society of Clinical Oncology Policy State-
ment Update: Genetic Testing for Cancer Suscepti-
bility. J Clin Oncol 2003;21:2397.
3. Antoniou A, Pharoah PDP, Narod S, et al. Average
risks of breast and ovarian cancer associated with
mutations in BRCA1 or BRCA2 detected in case
series unselected for family history: A combined
analysis of 22 studies. Am J Hum Genet 2003;
72(5):1117–1130.
4. Beral V, Bull D, Doll R, et al. Breast cancer and abor-
tion: Collaborative re-analysis of data from 53
epidemiological studies, including 83,000 women
with
breast
cancer from
16
countries.
Lancet
2004;363:1007.
5. Berry DA, Iversen ES Jr, Gudbjartsson DF, et al.
BRCAPRO validation, sensitivity of genetic testing
of BRCA1/BRCA2, and prevalence of other breast
cancer susceptibility genes. J Clin Oncol 2002;
20(11):2701–2712.
6. Bodian CA, Perzin KH, Lattes R. Lobular neoplasia:
Long term risk of breast cancer and relation to
other factors. Cancer 1996;78:1024.
7. Breast
cancer
and
breastfeeding:
Collaborative
reanalysis of individual data from 47 epidemiologi-
cal studies in 30 countries, including 50302 women
with breast cancer and 96973 women without the
disease. Lancet 2002;360:187.
8. Breast cancer and hormonal contraceptives: Collab-
orative reanalysis of individual data on 53, 297
women with breast cancer and 100,239 women
without breast cancer from 54 epidemiological
studies. Collaborative Group on Hormonal Factors
in Breast Cancer. Lancet 1996;347:1713.
9. Calle EE, Rodriguez C, Walker-Thurmond K, Thun,
MJ. Overweight, obesity, and mortality from cancer
in a prospectively studied cohort of U.S. adults.
N Engl J Med 2003;348:1625.
10. Cauley JA, Norton L, Lippman ME, et al. Continued
breast cancer risk reduction in postmenopausal
women treated with raloxifene: 4-year results from
the MORE trial. Multiple Outcomes of Raloxifene
Evaluation. Breast Cancer Res Treat 2001;65:125.
11. Chemoprevention of breast cancer: Recommenda-
tions and rationale. Ann Intern Med 2002;137:56.
12. Chlebowski RT, Collyar DE, Somerfield MR, Pfister
DG. American Society of Clinical Oncology tech-
nology assessment on breast cancer risk reduction
strategies:
Tamoxifen
and
raloxifene.
JCO
1999;17:1939.
13. Chlebowski RT, Col N, Winer EP, et al. American
Society of Clinical Oncology technology assessment
of pharmacologic interventions for breast cancer risk
reduction including tamoxifen, raloxifene, and
aromatase inhibition. J Clin Oncol 2002;20:3328.
14. Chlebowski RT, Hendrix SL, Langer RD, et al.
Influence of Estrogen Plus Progestin on Breast Can-
cer and Mammography in Healthy Postmenopausal
Women: The Women’s Health Initiative Rando-
mized Trial. JAMA 2003;289:3243.
15. Clemons M, Goss P. Estrogen and the risk of breast
cancer. N Engl J Med 2001;344:276.
16. Cuzick J, Powles T, Veronesi U, et al. Overview of
the main outcomes in breast-cancer prevention
trials. Lancet 2003;361:296.
17. Fisher B, Constantino JP, Wickerham DL, et al.
Tamoxifen for prevention of breast cancer: Report
of the National Surgical Adjuvant Breast and Bowel
Project P-1 Study. J Natl Cancer Inst 1998;90:1371.
18. Fisher B, Costantino JP, Wickerhan L, et al. Tamox-
ifen for prevention of breast cancer: Report of the
National Surgical Adjuvant Breast and Bowel Proj-
ect P-1 study. J Natl Cancer Inst 1998;90:1371.
19. Fisher B, Costantino JP, Wickerham DL, et al. Tamox-
ifen for the prevention of breast cancer: Report of the
National Surgical Adjuvant Breast and Bowel Project
P-1 study. J Natl Cancer Inst 1998;90:1371.
20. Foote FW, Steward FW. Lobular carcinoma in situ: A
rare form of mammary carcinoma. Am J Surg
Pathol 1941;19:74.
21. Ford D, Easton DF, Bishop DT, Narod SA, Goldgar
DE, and the Breast Cancer Linkage Consortium.
Risk of cancer in BRCA-1 mutation carriers. Lancet
1994;343:692–695.
22. Ford D, Easton DF, Stratton M, et al. The Breast
Cancer Linkage Consortium: Genetic heterogeneity
and penetrance analysis of the BRCA1 and BRCA2
genes in breast cancer families. Am J Hum Genet
1998;62:676–689.
23. Foster MC, Helvie MA, Gregory NE, et al. Lobular
carcinoma in situ or atypical lobular hyperplasia
at core-needle biopsy: Is excisional biopsy neces-
sary? Radiology 2004;231:813–819.
24. Frank TS, Deffenbaugh AM, Reid JE, et al. Clinical
characteristics of individuals with germline muta-
tions in BRCA1 and BRCA2: Analysis of 10,000
individuals. JCO 2002;20:1480–1490.
121
8—IDENTIFYING AND MANAGING THE HIGH-RISK PATIENT

25. Gail MH, Brinton LA, Byer DP, et al. Projecting indi-
vidualized probabilities of developing breast cancer
for white females who are being examined annu-
ally. J Natl Cancer Inst 1989;81:1879.
26. Goldberg JI, Borgen PI. Breast cancer susceptibility
testing: Past, present and future. Exp Rev Antican-
cer Ther 2006;6(8):1205–1214.
27. Gump FE. Lobular carcinoma in situ: Pathology
and treatment. Surg Clin N Am 1990;70: 873.
28. Hartmann LC, Schaid DJ, Woods JE, et al. Efficacy
of bilateral prophylactic mastectomy in women
with a family history of breast cancer. N Engl J
Med 1999;340:77–84.
29. Hartmann LC, Sellers TA, Schaid DJ, et al. Efficacy
of bilateral prophylactic mastectomy in BRCA1
and BRCA2 gene mutation carriers. J Natl Cancer
Inst 2001;93:1633–1637.
30. IBIS investigators. First results from the Interna-
tional Breast Cancer Intervention Study (IBIS-I): A
randomised
prevention
trial.
Lancet
2002;360:
817–824.
31. Lippman ME, Krueger KA, Eckert S, et al. Indicators
of lifetime estrogen exposure: Effect on breast cancer
incidence and interaction with raloxifene therapy
in the multiple outcomes of raloxifene evaluation
study participants. J Clin Oncol 2001;19:3111.
32. Martino S, Cauley JA, Barrett-Connor E, et al.
Continuing outcomes relevant to Evista: Breast
cancer incidence in postmenopausal osteoporotic
women in a randomized trial of raloxifene. J Natl
Cancer Inst 2004;96:1751.
33. Meijers-Heijboer EJ, van Geel B, van Putten WLJ,
et al. Breast cancer after prophylactic bilateral mas-
tectomy in women with a BRCA1 or BRCA2 muta-
tion. N Engl J Med 2001;345:159–164.
34. Metcalfe K, Lynch HT, Ghadirian P, et al. Contralat-
eral breast cancer in BRCA1 and BRCA2 mutation
carriers. J Clin Oncol 2004;22:2328.
35. Narod SA, Brunet JS, Ghadirian P, and the Heredi-
tary Breast Cancer Clinical Study Group. Tamoxifen
and risk of contralateral breast cancer in BRCA1
and BRCA2 carriers: A case control study. Lancet
2000;356:1876–1881.
36. Narod SA, Brunet JS, Ghadirian P, et al. Tamoxifen
and risk of contralateral breast cancer in BRCA1
and BRCA2 mutation carriers: A case-control study.
Hereditary Breast Cancer Clinical Study Group.
Lancet 2000;356:1876.
37. NCCN Guidelines on Genetic/Familial High-Risk
Assessment: Breast and ovarian. Available at www.
nccn.com/physician_gls/f_guidelines.html.
38. Page DL, Kidd TE, Depont WD, et al. Lobular neo-
plasia of the breast: Higher risk for subsequent inva-
sive cancer predicted by more extensive disease.
Hum Pathol 1991;22:1232.
39. Page DL, Schuyler PA, Dupont WD, et al. Atypical
lobular hyperplasia as a unilateral predictor of
breast cancer risk: A retrospective cohort study.
Lancet 2003;361:125.
40. Rebbeck TR, Lynch HT, Neuhausen SL, et al. Reduc-
tion in cancer risk after bilateral prophylactic
oophorectomy in BRCA1 and BRCA2 mutation
carriers. N Engl J Med 2002;346:1616–1622.
41. Risks and benefits of estrogen and progestin in
healthy postmenopausal women: Principal results
from the Women’s Health Initiative randomized
controlled trial. JAMA 2002;288:321.
42. Rivers A, Newman LA. Ductal lavage for breast can-
cer risk assessment. Surg Oncol Clin North Am
2005;14:45–68.
43. Robson M, Offit K. Management of an inherited
predisposition to breast cancer. N Engl J Med
2007;357:154.
44. Rosen PP, Lieberman PH, Braun DW, et al. Lobular
carcinoma in situ of the breast: Detailed analysis
of 99 patients with average follow-up of 24 years.
Am J Surg Pathol 1978;2:225.
45. Russo J, Moral R, Balogh GA, et al. The protective
role of pregnancy in breast cancer. Br Canc Res
2005;7(3): 131–142.
46. Sasco AJ, Kaaks R, Little RE. Breast cancer: Occur-
rence, risk factors and hormone metabolism. Exp
Rev Anticancer The 2003;3(4):546–562.
47. Singletary KW, Gapstur SM. Alcohol and breast can-
cer: Review of epidemiologic and experimental
evidence and potential mechanisms. JAMA 2001;
286:2143.
48. Struewing J, et al. The risk of breast cancer associated
with specific mutations of BRCA1 and BRCA2
among Ashkenazi
Jews.
N
Engl J
Med 1997;
336:1401–1407.
49. Wysowski DK, Honig SF, Beitz J. Uterine sarcoma
associated with tamoxifen use. N Engl J Med 2002;
346:1832.
122
SURGICAL FOUNDATIONS: ESSENTIALS OF BREAST SURGERY

9
Reading the Pathology Report
HISTOLOGY
Invasive Ductal Carcinoma
Invasive Lobular Carcinoma
Tubular Carcinoma
Cribriform Carcinoma
Medullary Carcinoma
Mucinous Carcinoma
Papillary Carcinoma
Secretory Carcinoma
Metaplastic Carcinoma
Other Forms of Breast
Cancer
TUMOR SIZE
MARGIN STATUS
GRADE
HORMONE RECEPTOR AND
HER-2/NEU EXPRESSION
LYMPHOVASCULAR INVASION
EXTENSIVE INTRADUCTAL
COMPONENT
Reading the Pathology Report: Key Points
Be familiar with the different subtypes of invasive breast cancer and
the implications on surgical management.
Know how margins are assessed and the need for reexcision for either a close
or positive margin.
Understand how tumor size, grade and other histologic features may impact
surgical or adjuvant therapy.
Although there are many important relation-
ships among the various members of the
breast cancer multidisciplinary team, the rela-
tionship between the surgeon and the pathol-
ogist is a unique one. The two are closely tied
because the success of surgical therapy is
dependent upon the accuracy of the pathology
report, and the accuracy of the pathology
report depends upon a firm grasp of the
surgical procedure performed. To succeed, this
relationship requires excellent communica-
tion, including the accurate labeling of speci-
mens, the proper orientation of specimens,
and possibly the presence of the surgeon in
the pathology laboratory to help facilitate the
process.
123

Histology
The first question when one receives a pathol-
ogy report back after a biopsy is what type of
cancer is this? Specifically is this in situ or
invasive disease and what type of invasive
cancer is it? The pathology of in situ disease
is discussed in Chapter 11. For invasive can-
cers, the two most common histologic types
are invasive ductal and invasive lobular. How-
ever, there are many other subtypes that may
have different implications in terms of surgical
or adjuvant therapy (Box 9–1).
Invasive Ductal Carcinoma
Invasive ductal carcinoma is the most com-
mon type of invasive breast cancer. Tumors
can be either pure invasive ductal carcinoma,
or mixed, which consists of combinations of
ductal carcinoma and other types. Grossly,
invasive ductal tumors are hard in consistency,
with a gray-white color. This is secondary to
an often present stromal desmoplasia. The
tumors can either be well circumscribed or be
spiculated with irregular margins. Histologi-
cally, the tumor cells are arranged as glandular
structures, either as nests, cords, trabeculae,
or solid sheets (Fig. 9–1). Depending on the
differentiation, the cells can look similar to
normal breast epithelial cells or show marked
cellular and nuclear pleomorphism.
Overall, invasive ductal carcinoma has the
worst prognosis of the invasive histologies; how-
ever, ductal cancers can be quite heterogeneous,
and within this group, subsets can be identified
with more favorable prognoses.
Invasive Lobular Carcinoma
The second most frequent type of invasive
breast cancer, invasive lobular carcinomas,
account for approximately 10% to 15% of
breast cancers. Even though invasive lobular
tumors can be hard and discrete like ductal
carcinomas, they are often much more subtle.
There may not be a hard palpable mass, but
rather a vague thickening
with indiscrete
margins. The same goes for imaging studies,
wherein invasive lobular carcinomas may be
mammographically occult or may be poorly
defined areas of asymmetry. Often, the tumor
is larger than expected based on physical
examination or imaging studies.
Histologically, classic invasive lobular carci-
noma is composed of a uniform population
of cells that infiltrate the breast parenchyma
in a so-called “single file” (Fig. 9–2). The cells
follow each other in a linear fashion to sur-
round both benign ductal or lobular struc-
tures. Cytologically, the cells are relatively
small with minimal cytoplasm, large non-
pleomorphic nuclei, and few if any mitoses.
There are several variants of lobular carcinoma
based on cytology (pleomorphic, histiocytoid,
signet-ring cell) or pattern (alveolar, solid,
tubulolobular). Sometimes the diagnosis is
made not only by pathology, but also by
taking into account the clinical and radiologic
presentation.
BOX 9–1 TYPES OF BREAST
CANCER
In situ cancer
Ductal carcinoma in situ
Lobular carcinoma in situ
Paget disease
Invasive cancer
Invasive lobular
Invasive ductal
Tubular
Cribriform
Papillary
Secretory
Cystic hypersecretory
Mucinous
Medullary
Apocrine
Acinic cell
Adenoid cystic
Sarcomatoid
Figure 9–1. Histology of invasive ductal carcinoma.
The cancer cells form glandular structures with
cellular and nuclear pleomorphism. These nests infil-
trate the surrounding breast tissue in a haphazard
fashion. (Image courtesy of Maria Braman, MD,
Department of Pathology, University of Michigan.)
124
SURGICAL FOUNDATIONS: ESSENTIALS OF BREAST SURGERY

Overall, lobular carcinoma is prognostically
similar to ductal carcinoma, although some
studies have shown better survival rates for lob-
ular carcinoma. However, studies that include
both classic lobular carcinoma and its variants
are difficult to interpret (Box 9–2). Pure, classic
invasive lobular carcinoma appears to have a
slightly better prognosis compared with solid
or alveolar patterns. Pleomorphic lobular carci-
noma and the signet ring cell variant appear to
have a worse prognosis.
There are two misconceptions regarding
invasive lobular carcinoma. The first is that
breast conservation therapy is inappropriate
for invasive lobular carcinoma, because the
tumors are often larger than what is found
on physical examination or on mammo-
graphy. Even though this is true and reexci-
sion may be necessary to ultimately achieve
negative margins, if an adequate lumpectomy
can
be
performed,
breast
conservation
is
perfectly appropriate. The second assumption
is that because lobular carcinoma in situ
(LCIS) is associated with an increased risk of
bilateral breast cancer, patients with invasive
lobular carcinoma are at an increased risk of
contralateral breast cancer compared to other
histologies. Thus some clinicians recommend
bilateral mastectomies more often for lobu-
lar carcinomas. However, long-term follow-up
of patients with invasive lobular carcinoma
shows a similar rate of contralateral breast
cancer to invasive ductal carcinoma.
Tubular Carcinoma
Tubular
carcinoma
is
an
extremely
well-
differentiated form of breast cancer. Tubular
carcinoma is rare, representing only 2% of
breast cancers. Histologically, tubular carci-
noma is composed of small, single-layered,
angulated tubules that are surrounded by a
dense cellular stroma, rich in hyaline and
elastotic tissue (Fig. 9–3). In this way, it is sim-
ilar to benign conditions such as sclerosing
adenosis and radial scar.
Compared with invasive ductal carcinoma,
tubular carcinoma tends to be more common
in older patients. It is more commonly multi-
centric and bilateral. They appear to be very
slow growing, with most (80% to 90%) being
less than 1 cm. Because they rarely get large
enough to cause a palpable mass and they are
often
associated
with
microcalcifications
(50% of cases), they are almost always detected
by mammography. Thus the incidence of
tubular carcinoma is rising with increased
mammographic screening.
The prognosis is quite good. The tumors
are usually small, rarely larger than 2 cm with
BOX 9–2 VARIANTS OF LOBULAR
CARCINOMA
Classic
Solid variant
Loose alveolar variant
Tubuloalveolar
Pleomorphic lobular
Figure 9–3. Histology of tubular carcinoma. Tubu-
lar carcinoma is characterized by small, single-
layered angulated tubules. They are surrounded by
a dense cellular stroma. (Image courtesy of Maria
Braman, MD, Department of Pathology, University
of Michigan.)
Figure 9–2. Histology of invasive lobular carci-
noma. The cancer cells infiltrate the parenchyma in
a linear, “single-file” pattern. The cells themselves
are small with minimal cytoplasm and large nuclei.
(Image courtesy of Maria Braman, MD, Department
of Pathology, University of Michigan.)
125
9—READING THE PATHOLOGY REPORT

a median size of 0.8 cm. Axillary node metas-
tases are less common than other invasive
subtypes, but not unheard of, so lymphatic
mapping is still recommended. Axillary node
involvement is more common when multi-
centric disease is present, but even when
nodes are involved, the prognosis is favorable.
Cribriform Carcinoma
Another well-differentiated form of invasive
cancer is cribriform carcinoma. It is also rare,
with an incidence of 0.3% to 4%. A cribriform
pattern of growth is characterized by nests
of tumor cells containing round to oval arch-
ed glandular spaces. This gives it a fenestra-
ted pattern with sharp “cookie-cutter” spaces
within a desmoplastic stroma. The cells in
these nests have low-grade nuclei and low
mitotic rates.
There
are
very
few
studies
specifically
addressing cribriform carcinoma. Axillary node
metastases are again less common than other
subtypes, but can occur, so lymphatic map-
ping is worthwhile. Cribriform carcinoma ap-
pears to be a low-grade malignancy, with a
better prognosis than other forms of breast
cancer.
Medullary Carcinoma
Medullary carcinomas are nicely circumscribed
tumors with a good prognosis despite a poorly
differentiated
appearance.
The
pathologist
must be certain that they are dealing with a
pure medullary carcinoma, and not an inva-
sive ductal carcinoma with medullary features,
because the latter has a prognosis similar to
invasive ductal cancer (Fig. 9–4). Medullary
carcinoma is often overdiagnosed. There are
strict criteria that must be met to call a cancer
medullary (Box 9–3).
Medullary carcinomas are more common in
younger women, as well as in both Japanese
and African-American women in the United
States.
Because
they
are
so
well
circum-
scribed, they are often confused for fibroade-
nomas. Prognosis is good, even when lymph
node involvement is present. Not uncom-
monly, medullary carcinomas are associated
with a reactive lymphadenopathy that may
be confused for clinical nodal involvement.
fine-needle aspiration should be performed
to confirm metastatic disease, and if negative,
sentinel node biopsy should be performed in
the presence of palpable adenopathy to avoid
unnecessary node dissection.
Mucinous Carcinoma
These tumors comprise less than 2% of inva-
sive breast cancers and, as their name implies,
are composed of large amounts of extracellular
mucin. Mucinous carcinomas are also known
as colloid or gelatinous carcinoma. They are
more common in older women and present
as a softer mass. Sometimes crepitance, known
as a “swish sign,” can be heard on palpation.
Most are mammographically occult; others
appear benign on imaging. Even follow-up
mammograms may
not increase
suspicion
Figure 9–4. Histology of medullary carcinoma. Med-
ullary carcinoma is characterized by broad irregular
sheets of cells devoid of glandular differentiation.
The borders between the cells are indiscriminate and
the cells are poorly differentiated, with high-grade
nuclei and frequent mitoses. (Image courtesy of Maria
Braman, MD, Department of Pathology, University of
Michigan.)
BOX 9–3 CRITERIA FOR
DIAGNOSING MEDULLARY
CARCINOMA


Microscopically circumscribed,
characterized by a smooth pushing
border


A lymphoplastic reaction present within
and at the periphery of the tumor


Germinal centers may be present,
mimicking an intramammary lymph node


A syncytial growth pattern is present,
characterized by broad irregular sheets
or islands of cells devoid of glandular
differentiation


Poorly differentiated cells


High nuclear grade


Mitotic activity
126
SURGICAL FOUNDATIONS: ESSENTIALS OF BREAST SURGERY

because
they
are
relatively
slow
growing.
Grossly, they are well-circumscribed with a
soft, currant-jelly gelatinous cut surface. Histo-
logically, they are defined by the presence of
well-differentiated cells floating in a sea of
extracellular mucin separated by fibrous septa
(Fig. 9–5). The mucin is composed of neutral
and
acidic
mucopolysaccharides
expressing
MUC-5 and MUC-2. It is possible that the
MUC-2 contributes to the favorable prognosis
of mucinous carcinomas, by not only confer-
ring tumor suppressor activity but also hinder-
ing the spread of cells. Mucinous carcinomas
have a lower rate of axillary node metastases
and an improved survival compared to other
subtypes. It is reasonable to treat small, node-
negative mucinous carcinomas without the
need for systemic adjuvant therapy. Recur-
rence can be quite late after initial treatment,
as long as 25 to 30 years after therapy.
Papillary Carcinoma
Another rare form of breast cancer that is more
common in elderly women is papillary carci-
noma. As with papillomas, papillary carcino-
mas are more common in the subareolar
region and can be associated with nipple dis-
charge.
However,
papillary
carcinomas
are
not limited to this area and can occur any-
where in the breast. Papillary carcinomas can
be difficult in that on mammography they
are well circumscribed, suggesting a benign
pathology. If part of the peripheral contour
lacks
circumscription,
invasion
should
be
suspected.
Under the microscope, papillary carcinoma
consists of a complex network of papillae: thin
fibrovascular cores covered by layers of cuboi-
dal or columnar epithelial cells (Fig. 9–6). The
cells
have
high
nuclear-to-cytoplasm
ratio
and hyperchromatic nuclei. It is not only diffi-
cult to differentiate benign from malignant
on mammography and ultrasound, but it is
also difficult on pathology. For these reasons,
core biopsies of a papillary lesion usually need
to be followed up by an excisional biopsy to
completely rule out a carcinoma. Papillary-like
lesions on mammography are therefore often
best removed by wire-localized biopsy to make
the diagnosis.
There are several variants of papillary carci-
noma. Intracystic papillary carcinoma describes
a papillary carcinoma that is entirely con-
tained within a large cystic duct. Alone, it is
an in situ papillary carcinoma with an excel-
lent prognosis and an extremely low rate of
axillary metastases. However, in many cases,
nonpapillary invasive ductal carcinoma is
seen in association with intracystic papillary
carcinoma. Papillary transitional cell carcinoma
is a variant that has both intraductal and
intracystic papillary growth and bears simi-
larities
to
papillary
urothelial
neoplasms.
Invasive micropapillary carcinoma is a much
more aggressive form of breast cancer, and
compared with other forms of papillary car-
cinoma, occurs in relatively younger women.
Often
mammographically
occult,
it
more
commonly presents with advanced disease
and axillary metastases. Chest wall recur-
rences are common and 50% of patients
Figure
9–6. Histology
of
papillary
carcinoma.
Within the network of thin fibrovascular cores are
layers of cuboidal or columnar epithelial cells. (Image
courtesy of Maria Braman, MD, Department of Pathol-
ogy, University of Michigan.)
Figure 9–5. Histology of mucinous carcinoma. Well-
defined cells floating in a sea of extracellular mucin
separated by fibrous septa. (Image courtesy of Maria
Braman, MD, Department of Pathology, University of
Michigan.)
127
9—READING THE PATHOLOGY REPORT

succumb to widespread disease within 2 years.
Histologically invasive micropapillary carci-
noma has a distinct pattern of neoplastic cells
in small micropapillae and ring-shaped nests.
These are immersed in a hyaline stroma giv-
ing a “spongy” appearance. There is often
extensive lymphatic invasion, and the disease
is often hormone receptor-positive and HER-
2/neu-positive.
Secretory Carcinoma
Secretory carcinoma, also known as juvenile
carcinoma, is a rare form of breast cancer that
can occur in children and teenagers as well
as in adults. It occurs over a wide age range
(3 to 73 years) but most patients are younger
than 30. A well-circumscribed mass, it has
several histologic growth patterns that can
include cribriform, microcystic, solid, and pap-
illary areas. The cells have abundant pale and
finely granular cytoplasm, round nuclei, and
prominent nucleoli. These can be confused
with apocrine or acinic carcinoma.
A
low-grade
neoplasm,
the
carcinoma
appears to have a good prognosis in children
and is slightly more aggressive in adults.
Tumor size less than 2 cm and an absence of
stromal invasion at the periphery also portend
a good prognosis. Wide excision is the treat-
ment of choice, with special attention being
paid to preserving the breast bud in prepuber-
tal girls. Recurrences do occur after lumpec-
tomy alone, as late as 20 years after therapy.
However, the benefit of radiation or chemo-
therapy is difficult to assess given the rarity
of secretory carcinoma. Sentinel node biopsy
should be done because there may be axillary
metastases in up to 30% of cases.
Metaplastic Carcinoma
Metaplastic breast cancers are rapidly growing
tumors, often presenting as a well-circumscribed,
firm mass larger than the average invasive
ductal carcinoma. The mean size is 3 to 4 cm
with more than half being more than 5 cm
and some are larger than 20 cm. The tumors
may be necrotic, and the largest tumors can
have associated skin and nipple changes. These
cancers are thought to be invasive ductal can-
cers transforming to nonglandular compo-
nents, and they consist of a combination of
mammary carcinoma with spindle, squamous,
chondroid, or osseous elements (metaplasia is
defined as a change from one cell type to
another). Despite the unusual appearance, the
prognosis is roughly equal to that of invasive
ductal carcinoma.
Metaplastic carcinomas are subdivided into
two categories: squamous and heterologous.
Squamous metaplasia has histologic features
similar to squamous cell carcinoma: intercellu-
lar bridges, dyskeratotic cells, and keratin pearl
formation. Focal squamous metaplasia can be
seen in a small percentage of cases of invasive
ductal carcinoma, but squamous metaplastic
carcinoma consists predominantly of squa-
mous-type changes. Heterologous metaplastic
carcinomas have two components, an epithe-
lial one and a heterologous one, which can
be quite bland or can contain spindle cells,
bone, cartilage, muscle, adipose tissue, and
vascular differentiation. When bone or carti-
lage is present, the carcinoma is known as a
matrix-producing carcinoma. When these com-
ponents are poorly differentiated, it is called
carcinosarcoma.
Other Forms of Breast Cancer
Apocrine carcinomas are characterized by apo-
crine differentiation of ductal cells. There are
two types of normal apocrine cells within the
mammary gland. Type A apocrine cells have
eosinophilic granular cytoplasm while type B
apocrine cells have abundant foamy cyto-
plasm filled by small vacuoles. These apocrine
cells can proliferate into atypical or malignant
lesions. Apocrine features can be seen with in
situ disease, invasive ductal carcinoma, and
invasive lobular carcinomas. Pure apocrine
carcinomas are rare, constituting 1% to 2%
of breast carcinomas. They possess similar
growth patterns to invasive ductal carcinoma
and similar clinical features. What makes apo-
crine carcinoma different is the appearance
of the cells, which have abundant granular
eosinophilia or cytoplasm with fine empty
vacuoles.
Adenoid
cystic
carcinomas
are
low-grade
tumors that usually arise in the salivary gland
but can occur in the breast, albeit quite rarely.
They often present as painful or tender masses
in the subperiareolar region. They have a good
prognosis with few cases of axillary or distant
metastases. Distant metastases occur in about
10% of patients, usually to the lungs.
Neuroendocrine carcinomas are defined by an
endocrine growth pattern and the expression
of neuroendocrine markers. They are more
common in older women and may occasionally
128
SURGICAL FOUNDATIONS: ESSENTIALS OF BREAST SURGERY

present with symptoms related to ectopic hor-
mone production such as adrenocorticotropic
hormone
(ACTH),
parathyroid
hormone
(PTH), calcitonin, and epinephrine.
Tumor Size
Besides the presence of regional metastases,
the size of the invasive component of the
tumor is, to date, the most important factor
used in therapy decisions. The size of a breast
cancer is generally determined by the measure-
ment of the largest contiguous area of invasive
cancer. Pathologic size does not necessarily
correspond with imaging size (although ultra-
sound is slightly more accurate than mammo-
gram) or clinical estimations. This may be due
to a lack of discrimination on the imaging
studies, the presence of benign components,
as well as decreased tumor size secondary to
multiple core-needle biopsies.
Multifocal tumors may extend over a larger
area, but the tumor size is reported as the larg-
est microscopic area of contiguous invasive
tumor. For staging and adjuvant therapy deci-
sions, this is most accurate, but it may not
reflect the true area needed to be resected sur-
gically. This is particularly true for lobular
carcinoma and after neoadjuvant chemother-
apy. Tumor size is also difficult to reconstruct
in situations in which a portion of the tumor
was removed by an excisional biopsy with
residual tissue on the reexcision lumpectomy.
For example, if 6 mm of tumor is reported in
the biopsy and 8 mm of tumor is reported
in the reexcision lumpectomy, is the tumor
8 mm (the size of the largest component) or
14 mm (the combined size), which would
upstage it from T1b to T1c? It is not reasonable
to assume the two portions add together
(Fig. 9–7), so this is generally discouraged, with
the tumor size being reported as the larger of
the two components.
Margin Status
One of the most important aspects of the
pathology report to the surgeon is the margin
status. In theory, tumors are generally uni-
centric and pathologic assessment of negative
margins should mean complete excision. Even
when ductal carcinoma in situ (DCIS) is pres-
ent, it is usually a contiguous spread through
the ductolobular tree rather than multiple foci.
However, theory does not match practice in
that there is tissue artifact and it is impossible
to completely examine the entire periphery
of a tumor. Negative margins do not preclude
the possibility of residual disease just as posi-
tive margins do not necessarily mean that
tumor was left behind. However, there is an
inverse relationship between the likelihood of
a complete excision and the proximity of the
tumor to the margins (as well as the amount
of tumor near the margin).
It is clear that positive margins are not
acceptable for breast conservation and require
reexcision. In an older study of lumpectomy
alone (no radiation), early local recurrence
rates were 45% for positive margins and 9%
for negative margins. In the modern era of
breast conservation (lumpectomy and radia-
tion), even though radiation improves local
recurrence, it does not compensate for positive
margins. Local recurrence rates range from 0%
to 9% for negative margin lumpectomies fol-
lowed by radiation therapy (XRT) compared
with 10% to 20% when margins are positive.
The more controversial question regarding
margin status is the “close margin.” This is
0.4 cm
0.4 cm
0.8 cm
1.4 cm
0.8 cm
0.8 cm
0.6 cm
0.4 cm
Specimen A
Specimen B
Tumor size = 1.4 cm, T1c
Tumor Size = 0.8 cm, T1b
Combining the two specimens:
OR
Figure 9–7. Difficulty in accurately determining the
true tumor size when there is residual tissue on the
reexcision. Depending on how the two sections com-
bine, the tumor may have a maximal dimension of
1.4 cm (and be staged as T1c) or 0.8 cm (and be
staged as T1b).
129
9—READING THE PATHOLOGY REPORT

variably defined as margins that are negative
but within 0.1 and 3 mm. Some authors say a
margin less than 1 mm is close, whereas other
authors say a margin within 3 mm is close.
Studies have shown inconsistent results as to
the local recurrence risk associated with close
margins. In general, however, studies show an
increasing risk of recurrence with the distance
between the tumor and the margin, and sev-
eral studies have shown a significant risk of
finding residual disease when reexcisions are
performed
for close
margins.
This
should
prompt an aggressive use of reexcision when
margins are within 2 to 3 mm although this
needs to be balanced against the size of the
close margin (broad vs. focal), the age of the
patient, the impact of re-excision and whether
a boost is planned by radiation oncology.
Grade
The grade of a tumor is primarily used for inva-
sive ductal carcinomas, recognizing the fact that
these can be quite heterogeneous. This allows
clinicians to stratify those invasive ductal can-
cers with a better prognosis from those with a
worse outcome. The grade of the tumor is based
on the degree of differentiation of the cells, com-
paring the nuclear and structural features with
those of normal mammary cells. This can be
based on the nuclear features, the architecture,
or a combination of features. The most com-
monly used grading system in breast cancer is
based on three features: nuclear grade, tubular
formation, and mitotic rate. Originally known
as the Bloom-Richardson grade, it was modified
by Elston and Ellis to remove some of the subjec-
tive qualities and make the grading of breast can-
cer more reproducible. Thus a semiquantitative
evaluation of the three variables is given (Box
9–4). Along with tumor size and lymph node sta-
tus, the grade of the tumor is the only other prog-
nostic factor that repeatedly and significantly
correlates
with
recurrence
and
survival.
Although grade is not included in the staging
system, it is used in adjuvant therapy decisions.
Hormone Receptor and Her-2/
neu Expression
Several factors have been identified that can
help select which patients are most likely to
benefit from adjuvant therapy. Most are con-
troversial or of limited benefit (Ki-67, p53,
S-phase, ploidy), although two are extremely
important and should be considered standard
of care in the histopathologic assessment of
breast cancer. These include staining for hor-
mone receptors (estrogen and progesterone)
and for Her-2/neu. These not only help select
appropriate therapies (hormonal therapy is
only effective in patients who express the hor-
mone receptors, and trastuzumab [Herceptin]
is only effective in patients who overexpress
Her-2/neu), but also provide prognostic infor-
mation because their expression correlates
with outcome. Both of these are most com-
monly evaluated by direct immunohistochem-
istry, which allows the pathologist to not only
visually evaluate the number of cells involved
and the intensity of the reaction, but also con-
firm that the expression is truly on the inva-
sive component. If the immunohistochemical
staining for Her-2/neu is equivocal, fluores-
cence in situ hybridization (FISH) is recom-
mended. These are discussed in more detail
in Chapters 16 and 17.
Lymphovascular Invasion
Angiolymphatic invasion, lymphovascular inva-
sion, lymphatic invasion, and vascular invasion
BOX 9–4 ELSTON AND ELLIS
MODIFICATION OF BLOOM-
RICHARDSON GRADE
Tubules
>75% of tumor composed
of tubules
1 point
10 to 75% of tumor
2 points
<10% of tumor
3 points
Nuclear Grade
Nuclei small and uniform
1 point
Moderate variation in
nuclear size and shape
2 points
Marked nuclear
proliferation
3 points
Mitotic Rate (Dependent on Microscopic
Field)
Low
1 point
Intermediate
2 points
High
3 points
Point Total
Histologic
Grade
3 to 5
1 (well-
differentiated)
6 to 7
2 (moderately
differentiated)
7 to 9
3 (poorly
differentiated)
130
SURGICAL FOUNDATIONS: ESSENTIALS OF BREAST SURGERY

are terms often used interchangeably among
pathologists to describe the histologic finding
of tumor cells within a vessel (Fig. 9–8). The term
“vascular invasion” should be restricted to cases
inwhich tumorcells are seen within a vesselcon-
taining a muscular wall (with or without red
blood cells) and the term “lymphatic invasion”
should be limited to evidence of intraluminal
groups of tumor cells conforming to the shape
of the endothelial-lined space. Whereas the for-
mer is relatively rare, the latter is much more
common. The term “lymphovascular invasion”
should be limited to cases in which both are
seen, but this is rarely the case in practice. In
addition, the reproducibility of this finding is
in question because it is often hard to tell the
difference between lymphatic invasion and arti-
fact. Immunohistochemical stains are emerging
that will help differentiate true lymphatic inva-
sion; this is important because this finding
implies a higher chance of finding nodal meta-
stases. Likewise, the finding of lymphatic or
vascular invasion correlates with prognosis in
node-negative
patients,
including
decreased
disease-free survival and increased local recur-
rence after breast conservation therapy (BCT).
Even though lymphovascular invasion is not a
contraindication to BCT, its presence may influ-
ence adjuvant chemotherapy and radiation ther-
apy decisions.
Extensive Intraductal
Component
Extensive intraductal component (EIC) refers
to cases in which at least 25% of the mass of
an invasive tumor is intraductal carcinoma,
and the intraductal component is present both
within and outside of the tumor mass. EIC is
important because EIC-positive patients have
a higher risk of residual DCIS in reexcision
lumpectomy specimens. In some series, EIC
has been associated with an increased local
recurrence rate, whereas other studies found
no difference, that is if negative margins
around both the invasive and intraductal com-
ponent are obtained. This emphasizes the
importance of attaining negative margins for
both
invasive
and
in
situ
disease
when
performing a lumpectomy, but does not pre-
clude
breast
conservation
in
EIC-positive
patients. However, EIC may be a contraindica-
tion to newer, experimental approaches to
breast cancer including in situ ablation and
partial breast irradiation.
Suggested Readings
1. Colleoni M, Rotmensz N, Maisonneuve P, et al.
Prognostic role of the extent of peritumoral vascular
invasion in operable breast cancer. Ann Oncol
2007;18(10):1632–1640.
2. Harris L, Fritsche, Mennel R, et al. American Society
of Clinical Oncology. American Society of Clinical
Oncology 2007 update of recommendations for the
use of tumor markers in breast cancer. J Clin Oncol
2007;25(33):5287–5312.
3. Hayes DF. Prognostic and predictive factors revisited.
Breast. 2005;14(6):493–499.
4. Kapoor A, Vogel VG. Prognostic factors for breast
cancer and their use in the clinical setting. Expert
Review in Anticancer Therapy 2005;5(2):269–281.
5. Mirza AN, Mirza NQ, Vlastos G, Singletary SE.
Prognostic factors in node-negative breast cancer:
A review of studies with sample size more than 200
and follow-up more than 5 years. Ann Surgery
2002;235(1):10–26.
6. Wong SL, Chao C, Edwards MJ, et al. University of
Louisville Breast Cancer Study Group. Frequency of
sentinel lymph node metastases in patients with
favorable breast cancer histologic subtypes. Am J
Surg 2002;184(6):492–498.
7. Yoder BJ, Wilkinson EJ, Massoll NA. Molecular and
morphologic distinctions between infiltrating ductal
and lobular carcinoma of the breast. Breast J 2007;
13(2):172–179.
Figure 9–8. Angiolymphatic invasion. Tumor cells
are seen within the dilated endothelial lined space.
The cells conform to the shape of the lymphatic
space, but appear to float in it. (Image courtesy of
Maria Braman, MD, Department of Pathology, Univer-
sity of Michigan.)
131
9—READING THE PATHOLOGY REPORT

This page intentionally left blank

10
Workup and Staging of the
Breast Cancer Patient
BREAST CANCER STAGING
T Stage
N Stage
M Stage
Other Information Not Included
in Staging
GENETIC COUNSELING
PRESENTATION AT A
MULTIDISCIPLINARY TUMOR
BOARD
Workup and Staging of the Breast Cancer
Patient: Key Points
Know the TNM staging system for breast cancer.
Describe the preoperative assessment of the woman diagnosed with
breast cancer.
Appreciate the potential benefits and drawbacks to obtaining an MRI to
stage breast cancer.
Develop an algorithm for evaluating the lymph nodes (N stage) in the clinically
node-negative and clinically node-positive patient.
Understand which patients need a more thorough staging (CT, bone scan)
before proceeding with surgery.
Appreciate which patients might benefit from meeting with a genetic counselor
before proceeding with therapy.
Understand the advantages of discussing the patient’s case and formulating
a treatment plan in a multidisciplinary setting.
Breast Cancer Staging
The purpose of staging breast cancer patients, or
all cancer patients for that matter, is to classify
groups of patients with similar outcomes. Even
though patients are often extremely concerned
with their stage, the information is of limited
use to them and to us, because even with our
current methods of stratifying patients there is
still considerable heterogeneity within each
133

stage. A primary reason for staging patients is to
provide consistency for both clinical and basic
science research. By staging patients appropri-
ately, outcomes across different populations
can be effectively compared. In this manner,
the efficacy of specific treatments can be better
assessed.
Staging patients is also important in deli-
neating a treatment plan, but in breast cancer
treatment, decisions are based on factors well
beyond simple TNM staging. Therefore it is
important not to focus only on garnering
information to provide the TNM staging, but
rather on information that will impact how
the patient is treated. Conversely, it is often a
waste of time and money to order a battery
of tests that are unlikely to influence treat-
ment, either because they are of low yield or
because treatment would not change pending
the results of the test. It is therefore necessary
to use clinical acumen when deciding what
staging tests to order before proceeding with
therapy.
Staging systems for breast cancer began in
1940 with the Manchester classification, a
clinical four-stage system based on the local
extent of the primary tumor and the involve-
ment of axillary lymph nodes. This gave way
to the Columbia staging system, which had
more
clinical
utility
in
predicting
which
patients were appropriate for surgery. The
Columbia classification served as a precursor
to the TNM classification system, introduced
in 1958 by the International Union against
Cancer (UICC). Whereas the Manchester and
Columbia systems were entirely clinical, the
TNM system was based on histopathology.
Independent of the UICC, the American Joint
Committee on Cancer (AJCC) developed a
series of site-specific staging systems in the late
1960s and early 1970s, including breast cancer.
Ultimately, in the late 1980s, the two bodies
agreed on a single staging system, which is in
use today.
It is important to remember that the staging
systems are fabricated, based on our knowl-
edge of important prognostic factors in breast
cancer. As our understanding of the biology
of breast cancer changes and as new prognos-
tic tests emerge, the staging system is updated.
The AJCC updates the staging system every
few years, incorporating new data, so that the
staging system better stratifies patients into
homogenous groups. This, however, makes
staging more complicated and more difficult
to commit to memory. In addition, as we
understand more about the molecular biology
of breast cancer, TNM staging may become
more archaic, replaced by a more sophisticated
stratification based on the presence, absence,
overexpression,
or
mutation
of
particular
genes within the tumor.
TNM staging of breast cancer patients is
typically done at two points. The first is the
clinical staging, which is based on physical
examination and imaging studies. The patient
is usually restaged after surgery using patho-
logic information. Even though the pathologic
TNM stage is generally more important and
better stratifies patients, clinical staging is
often used to base treatment decisions on,
particularly surgery and the use of neoadju-
vant chemotherapy. After neoadjuvant che-
motherapy, the pathologic stage may change
(hopefully the patient will be downstaged),
and the pretreatment pathologic stage may
never be truly known. It is important to clarify
whether one is referring to clinical or patho-
logic staging when discussing an individual
patient, as well as knowing when additional
clinical staging information may be necessary
to make treatment decisions.
T Stage
The first component of the TNM staging sys-
tem is the size of the tumor (T). Table 10–1
outlines the present T-staging system. The
T stage is based solely on the invasive compo-
nent of the tumor, thus a 3-cm area of ductal
carcinoma in situ (DCIS) with a 4-mm invasive
component is staged as T1a based on the
4 mm of invasive breast cancer. This can only
be determined after surgery and evaluation of
the resected specimen. The T stage is based
on the maximum diameter of the primary
tumor. Even after surgery, however, the T stage
is not always clear. For patients with more
than one tumor within the breast, the T stage
is based on the largest of the two lesions and
not the sum of the greatest diameter. Thus a
patient with a 1.3-cm tumor and a 1.0-cm
tumor is a T1b tumor (based on the 1.3 cm)
and not a T2 tumor based on the sum of 1.3
and 1.0. For patients with bilateral cancers,
each cancer is staged separately. However,
when a patient has multifocality within a
tumor, the T stage is less clear. Whether it
should be based on the size of the largest sin-
gle focus or based on the entire extent of the
multifocality is less clear and varies among
tumors and among pathologists. Either way,
it is important to document the presence of
multifocality.
134
SURGICAL FOUNDATIONS: ESSENTIALS OF BREAST SURGERY

The maximum dimension of the invasive
component strongly correlates with the risk
of regional and distant disease and hence prog-
nosis. Thus it is crucial information upon
which to base adjuvant therapy decisions.
However, for the surgeon, the most important
question in planning surgical therapy is the
size of the tumor relative to the size of the
breast and specifically whether a lumpectomy
is possible. In our patient with 3 cm of DCIS
and
4 mm
of
invasive
cancer,
the
3 cm
becomes the more important variable. Accu-
rate clinical T staging helps determine the
patient’s surgical options.
Clinical T staging is done by physical exami-
nation, mammography, ultrasonography, and
possibly magnetic resonance imaging (MRI).
Bilateral mammography is essential, not only
for estimating the size of the tumor but also
for looking for synchronous lesions or diffuse
microcalcifications. These may preclude breast
conservation or mandate biopsies of other
lesions before proceeding with surgery. Ultra-
sound is not mandatory, but can often provide
additional
information
about
the
primary
tumor. In addition, ultrasound of the regional
nodes can provide crucial staging information
and potentially alter surgical therapy (see later).
A more controversial topic is the routine role
of MRI in the staging of the primary tumor.
Several studies have suggested that MRI is the
preferred method for detecting the extent of
disease within the breast before surgical deci-
sion making. This appears particularly true
for
younger
women
with
dense
breasts,
women with an extensive intraductal compo-
nent, and women with invasive lobular carci-
noma. MRI also has been shown to increase
the detection of contralateral breast cancers
over mammography alone.
MRI is not without its drawbacks, however.
The primary concern is the specificity, in that
MRI has been associated with a high false-
positive rate. This partly depends upon the
technique and expertise of whoever is reading
the MRI, but even in the best hands, MRI can
be associated with false-positive findings. The
decision to perform a mastectomy based on an
MRI finding without tissue diagnosis will lead
to unnecessary mastectomies. Therefore any
abnormality discovered on MRI that might alter
the surgical management must be biopsied;
however, not all of the institutions performing
MRI have the capability of performing MRI-
guided biopsy. Patients should be guided to only
undergo breast MRI at an institution where
subsequent biopsy can be performed if neces-
sary, because it is very difficult and not always
possible (given the differences in machines, pro-
tocols, and interpretations between institutions)
to have a lesion seen at one hospital biopsied at
another.
A more important, and unanswered ques-
tion, is what impact MRI might have on recur-
rence rates. Proponents of MRI would point
out the frequency at which MRI detects lesions
that alter surgical management, leading to
either a wider excision or a mastectomy. How-
ever, this appears to be occurring at a fre-
quency higher than local recurrence rates,
suggesting that much of the disease seen on
MRI but missed on mammography is eradi-
cated by radiation or systemic therapy. This
has raised a significant concern that MRI in
this setting may lead to considerable surgical
overtreatment and a sharp increase in the
mastectomy rate.
Until more data are available, the routine
use of MRI as a preoperative staging test can-
not be recommended. It may be considered
in cases wherein mammography is limited by
breast density or in cases of invasive lobular
carcinoma, but even in these situations the
exact benefit of MRI has yet to be established.
It may also be used after lumpectomy if there
TABLE 10–1
Primary Tumor Staging
TX
Primary tumor cannot be assessed
T0
No evidence of primary tumor
Tis
Carcinoma in situ
Tis (DCIS)
Ductal carcinoma in situ
Tis (LCIS)
Lobular carcinoma in situ
Tis (Paget’s)
Paget disease of the nipple
with no tumor
T1
Tumor 2 cm or less in greatest dimension
T1mic
Microinvasion 0.1 cm or less
T1a
0.1 to 0.5 cm
T1b
>0.5 to 1.0 cm
T1c
>1.0 to 2.0 cm
T2
Tumor >2 to 5 cm in greatest dimension
T3
Tumor >5 cm in greatest dimension
T4
Tumor of any size with direct extension to
chest wall or skin
T4a
Extension to chest wall not including
pectoralis muscle
T4b
Skin involvement, including edema,
peau d’orange, ulceration, or
ipsilateral satellite skin nodules
T4c
T4a and T4b
T4d
Inflammatory carcinoma
135
10—WORKUP AND STAGING OF THE BREAST CANCER PATIENT

is a concern that bulky residual disease may be
present at the margins or elsewhere in the
breast, which may preclude an attempt at reex-
cision (although MRI cannot detect micro-
scopic disease and cannot be used to decide
whether a positive or close margin should
be reexcised). Patients should be counseled to
both the potential benefits of MRI and the
risks, and if an MRI is to be performed,
patients should be steered only to centers that
have
the
ability
to
perform
MRI-guided
biopsies.
N Stage
The absence or presence of disease in the
regional lymph nodes represents the N stage
and
has
undergone
dramatic
changes
in
recent years (Table 10–2). This is secondary
not only to changes in our understanding of
biology, but also to changes in how the
regional lymph nodes are approached surgi-
cally. Sentinel lymph node (SLN) biopsy has
essentially replaced axillary lymph node dis-
section (ALND) as the method for surgically
staging the regional lymph nodes. This has
not only reduced the morbidity of breast
cancer surgery, but also allowed for more
detailed histologic evaluation of the regional
nodes. Extremely small foci of disease within
the lymph nodes can now be detected that
would have been missed on routine ALND.
The clinical implications of these findings
are a source of confusion and controversy,
so care must be taken not to overestimate
or underestimate the importance of these
micrometastases.
As with Tstage, the N stage is first approached
clinically, with the pathologic N stage deter-
mined after surgery. In contrast to the T stage,
there are two different N-staging systems,
depending on whether the patient was assessed
clinically or pathologically. Table 10–2 demon-
strates the clinical N staging. The stage is based
on not only the absence or presence of ade-
nopathy, but also the location of disease (axil-
lary, internal mammary, supraclavicular) and
physical examination findings (movable ver-
sus fixed or matted). Thus a careful physical
examination of the axilla and neck is a crucial
part of the clinical N stage. It is important to
point out that not all palpable adenopathy
identified on clinical examination of the breast
cancer patient represents regional disease; in
some cases reactive lymph nodes may be pres-
ent. Fine-needle aspiration (FNA) biopsy of
palpable nodes in this setting is highly accurate
and should be routinely obtained to confirm
the presence of regional metastases.
Clinical staging of the regional nodes has
primarily been based on physical examina-
tion. Imaging studies were rarely indicated
to look at the regional nodes, and the iden-
tification
of
regional
disease
was
usually
an incidental finding on a computed tomog-
raphy (CT) scan obtained to look for distal
disease. However, the use of axillary ultra-
sound is expanding and should be consid-
ered either a selective (based on tumor size)
or routine part of the staging workup in any
patient
with
invasive
breast
cancer.
(See
Chapter 13.) The identification of axillary
metastases by ultrasound and ultrasound-
guided FNA biopsy can spare patients from
undergoing an unnecessary sentinel node
biopsy, instead proceeding directly to axil-
lary node dissection. This not only decreases
the number of operations performed but also
decreases the costs of breast cancer care.
In addition, the identification of disease in
the regional nodes before neoadjuvant che-
motherapy not only accurately stages the
patient, but also allows for the assessment of
response after chemotherapy.
Pathologic staging has become more com-
plex with the introduction of the SLN biopsy.
With
the
staging
ALND,
the
pathologist
bisected each identified node in search of
metastases. With only one to three nodes to
examine, SLN allows the pathologist to ana-
lyze multiple sections and allows more time
to identify metastases. The use of immuno-
histochemical (IHC) staining and molecular
detection of disease by polymerase chain re-
action
(PCR)
have
also
become
possible,
although the clinical significance is still in
question. The new staging guidelines for breast
cancer incorporate these changes into the
staging
system,
including
designations
for
the identification of small metastases of un-
known significance under N0. Disease not
visible on hematoxylin and eosin (H&E) but
visible on IHC and less than 0.2 mm is desig-
nated
N0(iþ)
and
usually
referred
to
as
“isolated tumor cells.” These are still categor-
ized as N0, and until more information is
available, should be treated as such. Likewise,
disease detected by reverse transcriptase PCR
(RT-PCR) but not by H&E or IHC is designated
as N0(molþ) and likewise should still be trea-
ted as N0 disease (RT-PCR of SLN should only
be done as research).
136
SURGICAL FOUNDATIONS: ESSENTIALS OF BREAST SURGERY

TABLE 10–2
Regional Nodal Staging: Clinical and Pathologic
Clinical
NX
Regional lymph nodes cannot be assessed
N0
No regional lymph node metastases
N1
Metastasis in movable ipsilateral axillary lymph nodes
N2
Metastasis in ipsilateral axillary lymph nodes fixed or matted, or in clinically apparent ipsilateral
internal mammary nodes in the absence of clinically evident axillary lymph node metastasis
N2a
Metastasis in ipsilateral axillary lymph nodes fixed to one another (matted) or to other structures
N2b
Metastasis only in clinically apparent ipsilateral internal mammary nodes in the absence of
clinically evident axillary lymph node metastasis
N3
Metastasis in ipsilateral infraclavicular lymph node(s) with or without axillary lymph node
involvement, or in clinically apparent ipsilateral internal mammary lymph node(s) and in
the presence of clinically evident axillary lymph node metastasis; or metastasis in ipsilateral
supraclavicular lymph node(s) with or without axillary or internal mammary lymph node
involvement
N3a
Metastasis in ipsilateral infraclavicular lymph node(s)
N3b
Metastasis in ipsilateral internal mammary lymph node(s) and axillary lymph node(s)
N3c
Metastasis in ipsilateral supraclavicular lymph node(s)
Pathologic (pN)
pNX
Regional lymph nodes cannot be assessed (e.g., previously removed or not removed for pathologic
study)
pN0
No regional lymph node metastasis histologically, no additional examination for isolated tumor
cells
pN0(i-)
No regional lymph node metastasis histologically, negative IHC
pN0(iþ)
No regional lymph node metastasis histologically, positive IHC but no IHC cluster greater than
0.2 mm
pN0(mol-)
No regional lymph node metastasis histologically, negative molecular findings (RT-PCR)
pN0(molþ)
No regional lymph node metastasis histologically, positive molecular findings (RT-PCR)
pN1
Metastasis in 1 to 3 axillary lymph nodes and/or in internal mammary nodes with microscopic
disease detected by SLN biopsy but not clinically apparent
pN1mic
Micrometastasis (greater than 0.2 mm, none greater than 2.0 mm)
pN1a
Metastasis in 1 to 3 axillary lymph nodes
pN1b
Metastasis in internal mammary nodes with microscopic disease detected by SLN biopsy but not
clinically apparent
pN1c
Metastasis in 1 to 3 axillary lymph nodes and in internal mammary nodes with microscopic
disease detected by SLN biopsy but not clinically apparent
pN2
Metastasis in 4 to 9 axillary lymph nodes, or in clinically apparent internal mammary lymph
nodes in the absence of axillary lymph node metastasis
pN2a
Metastasis in 4 to 9 axillary lymph nodes (at least one tumor deposit >2 mm)
pN2b
Metastasis in clinically apparent internal mammary lymph nodes in the absence of axillary lymph
node metastasis
pN3
Metastasis in 10 or more axillary lymph nodes, or in infraclavicular lymph nodes, or in clinically
apparent ipsilateral internal mammary lymph nodes in the presence of 1 or more positive
axillary lymph nodes; or in more than 3 axillary lymph nodes with clinically negative
microscopic metastasis in internal mammary lymph nodes; or in ipsilateral supraclavicular
lymph nodes
pN3a
Metastasis in 10 or more axillary lymph nodes or metastasis to the infraclavicular lymph nodes
pN3b
Metastasis in clinically apparent ipsilateral internal mammary lymph nodes in the presence of
1 or more positive axillary lymph nodes; or in more than 3 axillary lymph nodes with
microscopic metastasis in internal mammary lymph nodes detected by SLN biopsy
pN3c
Metastasis in ipsilateral supraclavicular lymph nodes
IHC, Immunohistochemistry; SLN sentinel lymph node.
137
10—WORKUP AND STAGING OF THE BREAST CANCER PATIENT

Metastases greater than 0.2 mm in size but
less than 2.0 mm in size are referred to as
micrometastases and are staged as N1mi. The
term micrometastases is often used inappropri-
ately,
both
to
describe
lesions
less
than
0.2 mm or greater than 2.0 mm. As N1 disease,
micrometastases should be treated as such,
prompting a completion node dissection (see
Chapter 13) and considered when contemplat-
ing adjuvant systemic therapy. The nodal
staging goes up with the number of nodes
involved, as well as with the location of metas-
tases, as outlined in Table 10–2.
Metastases to the internal mammary and
supraclavicular nodes must also be considered
in the N stage. Previously, supraclavicular
nodes were considered metastatic disease, and
patients
with
documented
disease
in
the
supraclavicular nodes were considered stage
IV. However, outcomes of patients with locally
advanced breast cancer with supraclavicular
involvement are similar to those of patients
with locally advanced breast cancer without
supraclavicular disease, and both are signifi-
cantly better than those with metastatic dis-
ease,
prompting
reclassification
of
these
metastases as pN3c disease.
The internal mammary nodes were not a sig-
nificant issue before sentinel node biopsy
came into use because they were only rarely
discovered incidentally on a staging CT study.
However, lymphatic mapping often demon-
strates drainage to the internal mammary
nodes and some surgeons routinely or selec-
tively excise these nodes (see Chapter 13).
Whether the disease is clinically discovered or
discovered on SLN biopsy and whether there
are internal mammary (IM) metastases in con-
junction with axillary metastases or not are
factored into the staging system.
M Stage
Finally, the M stage documents the presence of
distant
metastases
(see
Table
10–3).
The
staging system is simple: MX if the patient
cannot be clinically assessed, M0 if there are
no distant metastases, and M1 if there are.
The TNM staging can then be used to deter-
mine the AJCC stage (Table 10–4).
Whether a patient is M0 or M1 will in some
part depend on how hard one looks. All
patients
should
have
a
thorough
history
(including complete review of systems) and
physical as part of their staging workup (as
well as their diagnostic workup), and any
symptoms consistent with possible metastases
should be fully evaluated. Bone is the most
common site of distant spread (Box 10–1). Up
to 70% of women who die of breast cancer will
have bone metastases. However, very few
women will have identifiable bone metastases
at
presentation.
Patients
with
bone
pain,
recent fractures, lower extremity weakness, or
bowel or bladder dysfunction (signs of spinal
cord involvement) should have a thorough
workup including bone scan, and CT scans or
x-ray studies as indicated. Lung metastases
are the next most common site of distant
spread, representing about 25% of patients
with stage IV breast cancer, although again
only a small percentage have identifiable dis-
ease at presentation. Whereas lung metastases
often cause no symptoms initially, the pres-
ence of dyspnea, a dry cough, chest pain, or
heaviness or pain on deep inspiration should
prompt a chest CT and further studies as
indicated. Liver is the third most common site
TABLE 10–3
Distant Metastatic
Staging
MX
Distant metastasis cannot be assessed
M0
No distant metastasis
M1
Distant metastasis
TABLE 10–4
Stage Grouping
Stage
T
N
M
Stage 0
Tis
N0
M0
Stage 1
T1
N0
M0
Stage 2A
T0-1
N1
M0
T2
N0
M0
Stage 2B
T2
N1
M0
T3
N0
M0
Stage 3A
T0-3
N2
M0
T3
N1
M0
Stage 3B
T4
N0-2
M0
Stage 3C
Any T
N3
M0
Stage 4
Any T
Any N
M1
BOX 10–1 BONES MOST
COMMONLY AFFECTED BY
METASTASES
Spine
Ribs
Pelvis
Skull
Long bones of the extremities
138
SURGICAL FOUNDATIONS: ESSENTIALS OF BREAST SURGERY

of spread. It is exceedingly uncommon for
patients to have any symptoms suggestive of
liver disease. Finally, the central nervous sys-
tem (brain, cranial nerves, spinal cord, lepto-
meninges) is involved in 20% of patients
with stage IV disease (Box 10–2). Symptoms
of brain or spinal cord involvement should
prompt an MRI of the brain.
Even though the indications for imaging
studies in symptomatic patients are clear, the
majority of breast cancer patients are asymp-
tomatic and the question arises as to which,
if any, of these patients should have a workup
for distant metastases. Clearly, not all patients
need to undergo a complete workup. The prev-
alence of detectable metastases is extremely
low in the early stage of breast cancer, and
while their presence might alter therapy, a
complete workup in all patients is not only
exceedingly costly, but also leads to unneces-
sary
stress,
repeat
studies,
and
biopsies
prompted by false-positive results or indeter-
minate lesions often detected on these studies.
The evidence is clear that bone, CT, MRI , and
positron emitting tomography (PET) scans are
not indicated in asymptomatic patients with
early-stage disease. They have an extremely
low detection rate and poor sensitivity.
Chest x-ray studies are a common preopera-
tive staging examination. However, they have
a very low yield, detecting disease in only
approximately 0.3% of breast cancer patients.
Several studies have shown that their routine
use is neither cost-effective nor clinically use-
ful. However, many surgeons obtain a chest
x-ray study not so much to look for distant
disease but rather to look for cardiopulmonary
disease before administering general anesthe-
sia. Although the use of chest x-ray studies
in
asymptomatic
patients
without
known
underlying pulmonary disease is debatable, it
is a well-established practice. If a chest x-ray
study
was
obtained
and
demonstrates
an
abnormal finding, a follow-up CT scan is usu-
ally indicated.
Serum blood tests are often part of the rou-
tine preoperative search for distant disease
and include a complete blood count (CBC)
comprehensive
metabolic
panel
including
alkaline phosphatase and liver function tests.
Some physicians also order tumor markers
such as CA15-3 or CEA. A further metastatic
workup would then be initiated in patients
with anemia or abnormalities in any of these
laboratory values. As with radiographic stud-
ies, however, the yield of detecting metastases
based on serum laboratory abnormalities is
extremely low, and this approach does not
appear to be cost-effective.
For the symptomatic patient, the patient
with a more advanced cancer, and even the
asymptomatic patient with newly diagnosed
breast cancer, PET scanning is being examined
as an alternative to CT and bone scan. The
Center for Medicare and Medicaid Services
approved the use of PET scanning in breast
cancer patients in 2002, although how PET
scanning should be used is not yet clear. Even
though it is being used more often, the yield of
PET scanning in asymptomatic patients with
small primary tumors and clinically negative
nodes is very low (<1%) and not recom-
mended. For the symptomatic patient, or the
patient with more advanced disease, PET scan-
ning may be complementary to CT of the
chest/abdomen/pelvis and bone scan. As an
alternative, however, because infection and
inflammation will lead to increased uptake of
fluorodeoxyglucose, false-positive findings are
not
uncommon,
and
treatment
decisions
should not be based on PET scan results alone.
PET scan may play a bigger role in the workup
of the patient suspected to have recurrent dis-
ease (see Chapter 21).
Other Information Not Included
in Staging
Although the TNM staging system does a rea-
sonable job at stratifying the risk of recurrence,
and hence the benefit of therapy, there is addi-
tional information that goes into the decision-
making process, particularly with regard to
adjuvant systemic therapy (see Chapter 16).
As our knowledge of breast cancer genetics
advances and genetic analysis of breast cancer
specimens
become
commercially
available,
the TNM staging system becomes less relevant.
For years, factors beyond TNM have been
examined as methods to stratify risk.
BOX 10–2 SYMPTOMS OF CENTRAL
NERVOUS SYSTEM INVOLVEMENT BY
BREAST CANCER METASTASES
Motor deficit or gain disturbance
Seizures
Headaches
Cognitive dysfunction
Nausea and vomiting
Visual changes
Cranial nerve dysfunction
Cerebellar symptoms
Speech disturbances
139
10—WORKUP AND STAGING OF THE BREAST CANCER PATIENT