Breast Cancer

Synonyms and related keywords: carcinoma of the breast, cancer of the breast, adenocarcinoma of the breast, infiltrating ductal carcinoma, invasive ductal carcinoma, infiltrating lobular carcinoma, invasive lobular carcinoma, ductal carcinoma in situ, DCIS, lobular carcinoma in situ, LCIS, axillary lymph node dissection, ALND, BRCA1, BRCA2, Li-Fraumeni syndrome, Cowden disease, sentinel lymph node biopsy, SLNB, breast-conserving surgery, BCS, lumpectomy, radiation therapy, RT, atypical ductal hyperplasia, ADH, atypical lobular hyperplasia, ALH, radiotherapy, radiation oncology, chemotherapy, hormonal therapy, HT, radical mastectomy, RM, total mastectomy, TM, skin-sparing mastectomy, SSM, quadrantectomy, breast reconstruction, bone marrow micrometastasis, BMM, selective estrogen receptor modulators, SERMS, selective estrogen receptor down-regulators, SERDs, screening mammography, screening mammogram, lymphedema, estrogen-positive tumors, SBLLA syndrome, Peutz-Jeghers syndrome, Muir-Torre syndrome, nipple discharge, peau d'orange sign, skin dimpling, nipple retraction, lupus, Paget disease of the breast, axillary dissection, phantom breast syndrome


INTRODUCTION

Breast cancer is the most common cancer and the second most common cause of death from cancer in women. Because of the high frequency of the disease and the esthetic and symbolic value invested in the breast, breast cancer has always been a source of severe distress to patients and their families. For the same reasons, breast cancer research has increased dramatically during the last 2 decades, resulting in extraordinary progress in our understanding of the disease and in new, more efficient and less toxic treatments. Furthermore, the diffusion of knowledge, the medical advancements, and the increased public awareness have led to earlier diagnosis at stages usually amenable to complete resection and potential cure of the disease.

This article addresses the etiology, pathophysiology, clinical presentation, diagnosis, surgical and medical treatment, prognosis, and future directions of breast cancer.

For excellent patient education resources, visit eMedicine's Cancer and Tumors Center, Women's Health Center, and Imaging Center. Also, see eMedicine's patient education articles Breast Cancer, Mastectomy, Breast Lumps and Pain, Breast Self-Exam, Mammogram, and Ovarian Cancer.

Problem: Over the past few decades, breast cancer management has undergone significant changes characterized by less aggressive approaches to diagnosis and treatment. Mammogram and ultrasound or stereotactic biopsies have supplanted clinical diagnosis and surgical biopsy for the diagnosis; breast-conserving surgery (BCS) and sentinel lymph node biopsy (SLNB) have successfully replaced the more aggressive radical mastectomy (RM) and axillary lymph node dissection (ALND).

These changes are the results of a century-long experience whereby different models for the disease have been proposed and tested. RM, introduced by Halsted at the end of the last century, was based on the centrifugal model of tumor spread, according to which cancer spread starts locally then moves to the lymphatics and only then invades distant organs. Despite increasingly radical procedures, most patients relapsed with systemic disease.

New paradigms appeared to palliate the deficiency of this model. Both the systemic and the spectrum models acknowledge the role of the blood stream in tumor dissemination independent of lymphatic invasion, but they differ in their explication of the relationship between tumor size and distant metastases.

The systemic model considers breast cancer a systemic disease from its inception, while the spectrum model views breast cancer as a progressive disease in which invasion and metastases are a function of tumor growth and biological transformation. In addition, this model acknowledges that the disease may manifest over a spectrum of biological behaviors, with tumors that are metastatic from the beginning and others that may reach large sizes without dissemination.

Modern trials comparing different forms of locoregional control for patients at the same stage of disease show that variations in local treatment do not result in differences in long-term survival, validating the first premise of these models regarding the inadequacy of local approaches to control the disease in the absence of systemic therapy. In addition, screening mammography results in early detection of breast cancer (average size 1.4 cm vs 2.2 cm for tumors clinically detected) and is associated with a 25% decrease in the mortality rate for breast cancer, thus lending credibility to the spectrum model. This model stresses the importance of both local and systemic treatment.

ALND remains one of the mainstays of breast cancer management because clinical, imaging, or biological methods are insufficient to reliably define nodal status, the most reliable predictor of final outcome. Furthermore, ALND allows local control of the disease and may also improve survival. However, the extent of axillary dissection is still debated. Knowing that the procedure has risks of chronic morbidity in terms of arm mobility and lymphedema, the question is not a trivial one. Results from several prospective studies show that 10 nodes or more should be removed and found negative before declaring the axilla stage N0. This involves a level I dissection and, usually, a level II dissection.

Early detection of breast cancer in the screening mammogram era has raised the question of whether extensive axillary dissection could be replaced by more limited procedures or SLNB. Less invasive techniques, such as positron emission tomography scan and endoscopic axillary clearance, are being evaluated for staging and local treatment.

The indications for postoperative radiation therapy (RT) are being redefined for small estrogen-positive tumors, particularly in postmenopausal women.

Immediate reconstruction after mastectomy became an accepted practice that proved to be safe and compatible with early detection of recurrence. Because breast cancer is being detected at increasingly smaller sizes, plastic surgery will likely play a larger role in the initial treatment of breast cancer.

Frequency: The American Cancer Society estimated that 211,240 new cases of breast cancer (32.2% of all cancers in women) would be diagnosed in 2005 in the United States, making breast cancer the most-diagnosed cancer in women. Male breast cancer is a rare disease and 1690 cases were expected for 2005 in the US. The true incidence rates of breast cancer have been stable from 1987-1996 after a constant increase since 1979 (increase of 1% per year from 1979-1982; 4% per y from 1982-1987). The lack of decline of breast cancer incidence in the 1990s contrasts with a slight decline (decline of 1.3% per y from 1992-1997) of the incidence rate of cancer for all sites. Up to 40,870 cases of breast cancer-related deaths were expected for 2005 in the US.

Based on cancer cases diagnosed from 1995-1997, the probability of developing invasive cancer is 0.44% (1 in 225) for women younger than 39 years, 4.15% (1 in 24) for women aged 40-59 years, and 7.02% (1 in 14) for women aged 60-79 years. The estimated lifetime probability of developing breast cancer is 12.83% (~1 in 8). The likelihood of developing breast cancer is higher in white women than in women of any other racial or ethnic group.

Although the death rate from breast cancer has decreased an average of 2.2% per year from 1990-1997, the recorded number of deaths from breast cancer has remained stable, at approximately 43,000 per year. Deaths dropped to 41,737 in 1998 after reaching the highest number, 43,844, in 1995. Among women aged 20-59 years, breast cancer is the leading cause of death from cancer. However, lung cancer remains the leading cause of death from cancer in women aged 60 years or older.

Etiology: Breast cancer is a heterogeneous disease with no single characterized cause.

Epidemiological studies have identified many risk factors that increase the chance for a woman to develop breast cancer:

• Factors with relative risk greater than 4
  • Advanced age
  • Being born in North America or northern Europe
  • High premenopausal blood insulinlike growth factor (IGF)–1 level
  • High postmenopausal blood estrogen level
  • History of mother and a sister with breast cancer
• Factors associated with a relative risk of 2-4
  • High socioeconomic status
  • Age at first full-term pregnancy older than 30 years
  • History of cancer in one breast
  • Any first-degree relative with a history of breast cancer
  • History of a benign proliferative lesion, dysplastic mammographic changes, and a high dose of ionizing radiation to the chest
• Factors associated with a relative risk of 1.1-1.9
  • Nulliparity
  • Early menarche (age less than 11 y)
  • Late menopause (age >55 y)
  • Postmenopausal obesity
  • High-fat diet/saturated fat–rich diet
  • Residence in urban areas and northern United States
  • White race - Older than 45 years
  • Black race - Younger than 45 years
  • History of endometrial or ovarian cancer
• Identified factors with a protective role against breast cancer
  • Age at first period older than 15 years
  • Breastfeeding for longer than 1 year
  • Monounsaturated fat–rich diet
  • Physical activity
  • Premenopausal obesity

Genetic factors

As with other cancers, breast cancer is the result of multiple genetic changes or mutations. Early mutations may be inherited (eg, mutations of breast stem cells) or acquired (eg, somatic mutations due to ionizing radiation, chemical carcinogens, or oxidative damage).

Estrogens, by their proliferation-promoting effect on the breast epithelium later, increase the chance of DNA replication errors leading to carcinogenic mutations. Indeed, the common denominator to many of these risk factors is their effect on the level and duration of exposure to endogenous estrogenic stimulation.

Early menarche, regular ovulation, and late menopause increase lifetime exposure to estrogens in premenopausal women, while obesity and hormone replacement therapy increase estrogen levels in postmenopausal women. Conversely, late menarche, anovulation, and early menopause (spontaneous or induced) are protective, owing to their effect on lowering the level or shortening the duration of estrogenic exposure.

Lactation and premenopausal obesity are associated with lower estrogen levels as a result of anovulation. For unknown reasons, pregnancy decreases breast tissue susceptibility to somatic mutations; thus, the earlier the first pregnancy, the shorter the susceptibility period.

Hereditary breast cancers have been thought to represent a small proportion (5-10%) of all breast cancers. However, based on new data derived from the comparison of identical and nonidentical twins, up to 27% of breast cancers may be attributed to inherited factors. The mutated genes BRCA1 and BRCA2 are responsible for approximately 30-40% of inherited breast cancers.

The prevalence of BRCA1 in the general population is 0.1%, compared with 20% in the Ashkenazi Jewish population. The gene is encountered in 3% of the unselected breast cancer population and in 70% of women with inherited early-onset breast cancer. Up to 50-87% of women carrying a mutated BRCA1 gene develop breast cancer during their lifetime.

Risks for ovarian and prostate cancers are also increased in carriers of this mutation. BRCA2 mutations are identified in 10-20% of families at high risk for breast and ovarian cancers and in only 2.7% of women with early-onset breast cancer. The lifetime risk of developing breast cancer in female carriers is 25-30%. BRCA2 is also a risk factor for male breast cancer; carriers have a lifetime risk of 6% for developing the cancer. BRCA2 mutations are associated with other types of cancers, such as prostate, pancreatic, fallopian tube, bladder, non-Hodgkin lymphoma, and basal cell carcinoma.

Li-Fraumeni syndrome, characterized by a mutation of TP53, is associated with multiple cancers, including the SBLLA syndrome (sarcoma, breast and brain tumors, leukemia, laryngeal and lung cancer). Cancer susceptibility is transmitted by an autosomal dominant pattern, with penetrance approximating 90% by age 70 years. Li-Fraumeni syndrome is identified in 1% of women with early-onset breast cancer. Bilateral breast cancer is noted in up to 25% of patients.

Cowden disease is a rare genetic syndrome associated with papillomatosis of the lips and oral mucosa, multiple facial trichilemmomas, and acral keratosis. The prevalence rate of breast cancer in women with this disease is 29%. Benign mammary abnormalities (eg, fibroadenomas, fibrocystic lesions, ductal epithelial hyperplasia, nipple malformations) are also more common. Other rare genetic disorders, such as Peutz-Jeghers and Muir-Torre syndromes, are associated with an increased risk of breast cancer.

Pathophysiology: In a normal state, cells proliferate in response to external proliferation-promoting signals to fulfill a function such as replacing lost cells or repairing injured tissues. Once the goal has been reached, a set of proliferation-repressing signals is activated. These signals allow the cells to exit the proliferation cycle (cell cycle) by returning to the dormant state (G0), by differentiating, or by dying (apoptosis). Each of these functions is carried out by a complex system of interacting proteins. Constitutive expression by mutation or another genetic change of any component of the proliferation-promoting system may result in uncontrolled proliferation. The constitutively expressed component is called an oncogene.

Conversely, the loss by mutation or deletion of a proliferation-repressing gene results in an inability to stop the cell cycle and, thereby, continuous proliferation, possibly leading to cancer. The lost gene is called a tumor suppressor gene. Likewise, constitutive expression of antiapoptotic genes may result in immortalization of the cell, paving the way for further genetic changes and eventually cancer formation. Loss of proapoptotic genes may lead to similar results. Thus, autonomous proliferation and immortality shared by all cancers are the final result of successive genetic changes, which may be different from one cancer to another.

Breast cancer is not an exception in that regard. It is the result of multiple genetic changes that are different from those of other malignancies and that confer to this cancer its characteristic phenotype.

Cell-cycle deregulation in breast cancer

Estrogen and progesterone induce cyclin D1 and c-myc expression. Although both sex hormones provide directionality by shifting the CDKI p21 from CDK2 to CDK4, progesterone promotes maturation by inducing p27, while only estrogen allows multiple cycles. Recent studies have reported common amplification of cyclin D1 (a third of breast cancers), inactivation of p16, and mutation of TP53 in breast cancer.

c-myc overexpression is one of the most common genetic alterations encountered in persons with breast cancer (a third of patients). Depending on the availability of its different partners, it may result in proliferation and chromosomal instability (Myc-Max) or differentiation (Myc-Mad), probably by sequestering Myc and reducing its availability. Amplification of the c-myc gene is associated with a poor prognosis and a high S-phase.

Estrogen receptor (ER)–positive breast cancer cells undergo apoptosis after withdrawal of estrogen, suggesting that this hormone functions not only as a mitogen but also as a survival factor. The antiapoptotic factor Bcl-2 is commonly overexpressed in ER-positive breast cancers.

ER negativity is observed in a third of primary breast cancers and a third of recurrences of ER-positive primaries. The ER gene is usually intact with no identifiable deletions or mutations. Although the exact mechanism of this lack of expression is not known, hypermethylation used normally by the genome to silence certain genes is a possible explanation. Methylation of cytosine-rich areas (called CpG islands) of the ER-gene promoter region has been described in the majority of ER-negative breast cancers and in a small fraction of ER-positive breast cancers. Demethylation of these areas with specific agents (eg, 5-azacytidine) restores ER expression and its function in vitro.

A progesterone receptor (PR) is present in approximately 50% of all ER-positive tumors. Its presence depends on the expression of functional ER, which explains its absence in almost all ER-negative breast cancers. The mitogenic effect of progesterone in breast cancer may depend on the induction of local growth hormone production in the hyperplastic mammary epithelium. However, high doses of progestins have proven inhibitory effects on breast cancer growth mediated by the down-regulation of G1-phase CDKs and cyclin D1 leading to cell differentiation.

Regarding adhesion-dependent cell regulation, the transmembrane glycoproteins, ie, epithelial cadherins (E-cadherins), mediate with their extracellular domain cell-to-cell interactions, thus stabilizing the cell in the epithelial tissue. Their intracellular domain interacts with and controls the transcription factors B-catenins. A mutation or the absence of E-cadherins results in cell detachment, increased motility and invasiveness, and release of B-catenins, which up-regulates c-myc expression.

Expression of E-cadherins is down-regulated in breast cancer. Another family of adhesion molecules, the integrins, is involved in cell-to-matrix interactions. Integrins signal through the Fak-Src pathway, which activates PI3K and AKT, resulting in enhanced survival, proliferation, and motility. The main components of this pathway (Fak, PI3K, and AKT) are inhibited by PTEN (ie, phosphatase on chromosome 10 gene product, mutated in Cowden disease), which results in the suppression of survival and apoptosis.

The epidermal growth factor (EGF) receptor family plays a critical role in mammary tumorigenesis. Other than the EGF receptor itself, 3 other members of this family have been described, including c-erb-B2 (HER2, HER2/neu), c-erb-B3, and c-erb-B4; the latter is called a kinase-dead receptor because it does not carry a kinase function on the cytoplasmic domain of the receptor, which is in contrast to the other members of the family.

These receptors interact with many ligands, including EGF, transforming growth factor (TGF)–alpha, hergulin (or Neu differentiation factor), heparin-binding EGF-like growth factor, beta-cellulin, and epiregulin. Upon binding of a ligand with its cognate receptor, a homodimerization or heterodimerization process occurs, followed by autophosphorylation of the intracellular domain and activation of the intrinsic catalytic domain.

Transmission of the signal is operated via phosphorylation of adaptor proteins (GRB2-SOS, Shc, IRS-1/2, STAT) docked or recruited to the cytoplasmic domain of the receptor, followed by activation of the RAS-GTP protein, then 1 of 3 pathways to the nucleus (ie, Raf/MEK/ERK-1/2, MEKK1/MEK4/7/JNK, PI3K/AKT/GSK).

Transmission of the signal from adaptor proteins to the nucleus without mediation by RAS is possible through Fak/Src or Rho/Rac/CDC42 pathways. In addition, phospholipase C-gamma is activated by direct interaction with the phosphorylated c-erb-B-2; however, its intracellular pathway is not fully known. The final result is the induction of transcription factors (ie, Myc, NF-kB, ATF, Ets, AP-1, EIK, SRF) that drive the cell cycle by up-regulating cyclins and inhibiting CDKIs and proapoptotic signals.

Once the biological message has been executed, the complex ligand receptor is internalized and destroyed in the lysosomes. The pattern of heterodimerization, the intracellular pathway used, and the rate of internalization and destruction of the receptor depend on the specific ligand bound to the receptor.

Although c-erb-B2 does not have a specific ligand, its role in the signal transmission from the epidermal growth factor receptor is crucial. Cell lines lacking c-erb-B2 are resistant to the tumorigenic effect of EGF, while those with a kinase-deficient or carboxyl terminal–truncated EGF receptor with intact c-erb-B2 can still execute all the functions of the wild type.

The discovery of the role of HER2 in breast cancer was one of the landmarks in breast cancer research in the last 2 decades. HER2 is overexpressed in 20-30% of breast cancers. Tumor cells overexpressing HER2/neu may have up to 2 million copies of the receptor on their surface compared with 20,000-50,000 copies in normal breast epithelial cells. Because of this abundance of HER2/neu, many heterodimers contain HER2/neu, resulting in potent intracellular signaling and malignant growth.

Despite persistent controversy regarding certain aspects of its biology, prognostic value, and methods of evaluation, HER2 overexpression or amplification is generally accepted to be correlated with a high histologic grade, the absence of hormone receptor expression, aneuploidy, a high proliferation index, tumor size, and a poor clinical outcome. Its role as a predictor of response to chemotherapy and hormonal therapy (HT) is not clearly defined.

Certain clinical studies using tamoxifen showed not only a lack of response, but also a detrimental effect in the group of patients overexpressing HER2. Retrospectively reviewed chemotherapy data showed longer disease-free survival and overall survival in HER2 overexpressors who received high doses of doxorubicin-containing chemotherapy compared with HER2-negative patients. Available data concerning the interaction between cyclophosphamide, methotrexate, and 5-fluorouracil (CMF) chemotherapy and HER2 overexpression are inconclusive and do not allow the formation of final conclusions. Taxanes seem to have high efficacy in patients overexpressing HER2 (relative risk, 65%) compared with the HER2-negative group (relative risk, 35%).

The IGF family consists of IGF-1, IGF-2, IGF receptor-1, IGF receptor-2, and IGF-binding proteins. The members of this family play an important role in normal mammary development and tumorigenesis. Both IGF-1 and IGF-2 bind to IGF receptor-1, whose strong mutagenic effect is synergistic with estrogen. In breast cancer, IGF receptor-1 and IGF-1 and IGF-2 are overexpressed in epithelial cells and stromal cells, respectively. Paradoxically, this overexpression correlates with a good prognosis, perhaps reflecting simple hormone dependence or association. IGF receptor-2 plays a tumor-suppressing role by down-regulating IGF-2.

The TGF-beta family consists of 3 TGF-beta and 2 interdependent serine-threonine kinase receptors. In normal mammary epithelial cells, TGF-beta blocks the expression of cyclin A, an S-phase–promoting protein and (to a lesser degree) cyclins D and E involved in the G1 phase and induces the expression of the CDKI p15, which results in cell-cycle arrest and, potentially, apoptosis. This explains its role in postlactational mammary gland regression.

Expression of TGF-beta in breast cancer is increased and seems to correlate with disease progression rather than tumor suppression. Mutation of TGF-beta receptors (type I and II) or any of the downstream molecules (Smad4) involved in intracellular signal transduction renders breast cancer cells resistant to its suppressive effects. However, its role in promoting angiogenesis and invasion and in suppressing the immune system becomes advantageous for the cancer cells, which acquire a proliferative advantage by losing sensitivity to TGF-beta and developing a way to escape the host immunosurveillance.

Mutated BRCA1 and BRCA2, breast cancer susceptibility genes, are proven risk factors for breast cancer. More than 500 mutations have been described in the BRCA1 gene (band 17q21), and 250 have been described in the BRCA2 gene (band 13q12-13). The mutations occurring at either end of the BRCA1 gene are associated with more aggressive tumors; those occurring at the 5` extremity are associated with breast and ovarian cancers, while those closer to the 3` end are associated with only breast cancer.

The biological function of the BRCA1 gene product is not well known. Accumulated evidence suggests that BRCA1 is a nuclear protein involved in other genes' expression, in cell cycle progression, and in the response to DNA damage. DNA damage results in activation and interaction of BRCA1, BRCA2, RAD51, and TP53 with subsequent expression of p21, which leads to a cell cycle pause until the damage is repaired. The mutation or absence of BRCA1 results in failure to repair the damaged DNA, and the cell cycle continues to accumulate further mutations, eventually leading to tumorigenesis. BRCA2 seems to play a role similar to that of BRCA1 in the cell cycle, other genes' expression, and DNA damage repair.

Clinical: In the past, the great majority of patients presented with a painless palpable mass. Although more than 80% of palpable masses are benign, the decision to observe such lesions should be made only after careful clinical, mammographic, and pathologic workup. Cystic lesions identified clinically or on ultrasound images should be explored using fine-needle aspiration (FNA) biopsy. Nonbloody fluid and complete resolution of the cyst confirm its benign nature. If the fluid is bloody or the cyst does not resolve after aspiration or has a complex appearance on the ultrasound image, a biopsy is indicated.

Other symptoms, such as breast pain or deformity, nipple discharge, and erythema or skin ulceration, occasionally occur. Patients with Paget disease present with a long-standing eczematoid rash of the nipple-areola complex, itching, tenderness, burning, and occasional bloody discharge from the nipple. Skin dimpling, the result of shortening or retraction of the Cooper ligaments induced by the tumor, does not have prognostic value, while the ominous peau d'orange sign reflects the invasion of the subdermal lymphatic plexus and portends a shortened survival.

Symptoms related to distant metastases, such as bone pain, dyspnea, or meningitic syndrome, are encountered in some cases.

In current practice, increasing numbers of breast cancers are mammographically diagnosed in the preclinical stage. Screening mammography has resulted in earlier diagnosis of breast cancer, which has translated in recent years into a 25% improvement in the mortality rate related to breast cancer. Mammographic signs suggestive of cancer include architectural distortions, microcalcifications, or masses. These changes require further evaluation using diagnostic mammograms with or without ultrasound. Biopsies are indicated if these changes are confirmed. Features helpful in the evaluation of palpable breast masses are as follows:

• Malignant masses
  • Hard
  • Painless: Malignant masses are painful in only 10-15% of patients.
  • Irregular
  • Possibly fixed to the skin or chest wall
  • Skin dimpling
  • Nipple retraction
  • Bloody discharge
• Benign masses
  • Firm, rubbery mass
  • Frequently painful
  • Regular margins
  • Not fixed to skin or chest wall, mobile
  • No skin dimpling
  • No nipple retraction
  • No bloody discharge
• Cysts: No reliable features distinguish cysts from solid masses based on clinical data.

Nipple discharge may be spontaneous or induced, unilateral or bilateral, and have different colors and textures. If the discharge is associated with one or more of the suggestive features, further investigation is necessary. Clinical characteristics of nipple discharges are as follows:

• Malignant discharge
  • Unilateral
  • Spontaneous
  • One duct orifice
  • Bloody, serosanguineous, or serous
• Benign discharge
  • Bilateral
  • Spontaneous or induced
  • Multiple duct orifices
  • Thick green or yellow, induced and bilateral (duct ectasia)

INDICATIONS

Ductal carcinoma in situ

Mastectomy has been the criterion standard for many years, with significantly low local recurrence and mortality rates (0.75% and 1.7%, respectively). The idea of using BCS has gained acceptance after good success in women with invasive disease.

When BCS is used, RT is often a part of the treatment plan because the prevalence of ipsilateral noninvasive breast cancer was reduced from 13.4% to 8.2% and, for invasive cancer, from 13.4% to 3.9%, after using RT (National Surgical Adjuvant Breast Project [NSABP] B-17). Because most of the tumors in this study were smaller than 1 cm, a wise plan is to continue offering mastectomy to (1) patients with extensive multicentricity, (2) those with multifocality, and (3) those in whom negative margins cannot be obtained with wide excision. Axillary node dissection or adjuvant chemotherapy has no role in the treatment of this disease.

Patients with early-stage invasive breast cancer (TNM stage I and II) may benefit from BCS. BCS should be offered to patients with small tumors and adequate breast size. Family history, tumor location, and the presence of pathologically or clinically involved axillary lymph nodes are not contraindications to such a surgery. Absolute contraindications are 2 or more primary tumors in separate quadrants, previous RT to the breast, persistent positive margins, and pregnancy.

An extensive intraductal component within the index lesion and the surrounding tissue is not associated with a high risk of recurrence as long as clear margins can be achieved.

For Paget disease, mastectomy should be performed in patients with a tumor located beyond the central portion of the breast, while BCS may be considered in patients with disease limited to the retroareolar area, with excision of the nipple-areolar complex and complete excision of the mammographic abnormalities with a 2-cm cone of the retroareolar tissue. If surgical margins are negative, RT should complete the treatment; otherwise, mastectomy is the treatment of choice.

Indications for mastectomy include patient preference, an inability to achieve clean margins without unacceptable deformation of the remaining breast tissue, multiple primary tumors, previous chest wall irradiation, pregnancy, and severe collagen vascular diseases (eg, lupus), which are considered absolute contraindications to BCS.

Simple mastectomy is used in the treatment of ductal carcinoma in situ (DCIS), after lumpectomy or axillary dissection with or without radiation, if the specimen shows positive margins, in frail patients in whom an axillary dissection is contraindicated, and as a prophylactic measure.

Modified RM is performed in the presence of contraindications to BCS or as a patient preference.

ALND remains the standard of care in the management of invasive carcinoma of the breast. ALND is not indicated in carcinoma in situ, unless the presence of an invasive component is strongly suggested.

SLNB has recently emerged as a credible alternative to ALND, with comparable, if not better, staging results and much less morbidity. In expert hands, SLNB identifies the sentinel lymph node (SLN) in 85-98% of patients and correctly stages the axilla in more than 95% of patients, with a false-negative rate of less than 5%. The combined use of technetium colloid and blue dye techniques increases the detection rate of the SLN.

SLNB is usually offered to patients with early-stage breast cancer (stage I or II) who have no gross axillary lymph node involvement. SLNB is contraindicated in patients with clinically palpable axillary lymphadenopathy, multifocal disease, or locally advanced lesions. Patients with positive SLNB findings should undergo level I and II ALND. Because of the slow learning curve for this new procedure, the American Society of Breast Surgeons considers that an individual surgeon should perform at least 20 SLNB procedures with ALND to minimize the risk of false-negative results.

RELEVANT ANATOMY AND CONTRAINDICATIONS

Relevant Anatomy: The breasts are between the second and sixth ribs and are composed of breast tissue, skin, and subcutaneous tissue. The breast tissue is composed of parenchyma and stroma. The parenchyma is composed of 15-25 lobes, and each lobe contains 20-40 lobules. Each lobule consists of 10-100 alveoli. Fifteen to 25 lactiferous ducts provide separate drainage to the corresponding lobes. Before opening at the nipple, these ducts become dilated, forming the lactiferous sinuses. The breast tissue is enveloped superficially by the superficial pectoral fascia and deeply by the deep pectoral fascia, with the 2 layers connected by fibrous bands called Cooper suspensory ligaments.

The lymphatic drainage of the breast is unidirectional, from the superficial to the deep lymphatic plexus. The lymph then flows centrifugally to the regional lymph nodes after traveling through the lymphatic vessels of the lactiferous ducts. Ninety-seven percent of this flow is collected in the axillary lymph nodes, while only 3% goes to the internal mammary nodes.

Axillary lymph nodes are divided into apical lymph nodes, interpectoral (Rotter) lymph nodes, axillary vein lymph nodes, central lymph nodes, scapular lymph nodes, and external mammary lymph nodes. An arbitrary method divides these lymph nodes into 3 levels relative to their relationship with the pectoralis minor muscle. Lateral to the lateral border of this muscle lie the level I lymph nodes, and medial to it lie the level III lymph nodes. Level II lymph nodes are located between and behind the muscle.

Several structures, including vessels and muscles with their nerve supply, are related to the breasts and should be preserved during mastectomy or axillary node dissection. The pectoralis minor muscle and the lateral portion of the pectoralis major muscle are innervated by the medial pectoral nerve. Preservation of this nerve is particularly important to prevent atrophy of the pectoral muscles if a submuscular implant reconstruction is planned.

Additionally, the serratus anterior muscle is innervated by the long thoracic nerve of Bell, whose preservation is crucial to prevent winging of the scapula. Resection of the thoracodorsal nerve supplying the latissimus dorsi muscle should be avoided whenever possible, although resection does not result in any cosmetic or functional sequelae. Exposure of the axillary artery and brachial plexus should be avoided.

Also, injury to certain sensory branches of the brachial plexus that occasionally pass superficially to the axillary vein may result in arm numbness extending to the wrist. Injury to the intercostobrachial nerve results in numbness over the triceps area. It can be identified by its large size (2 mm) and its location near the axillary vein. Occasionally, this nerve is composed of multiple, thin branches that cannot be preserved. In this case, the nerve should be sectioned with the knife to prevent postoperative causalgia related to the use of electrocautery.

Contraindications:

• Absolute contraindications to lumpectomy/RT
  • Pregnancy
  • Prior irradiation to the breast
  • Two or more gross foci of cancer in separate quadrants of the same breast
  • Mammographic findings suggestive of diffuse areas suggestive of malignancy
  • Failure to obtain negative margins despite several surgical attempts
  • Collagen vascular diseases
• Relative contraindications to lumpectomy/RT
  • Tumors larger than 5 cm
  • Very large or very small breasts

WORKUP

Lab Studies:

• CBC count with differential and platelet count

• Chemistry and renal function studies

• Liver function tests

• Calcium and phosphorus evaluations

Imaging Studies:

• Mammography: Bilateral study is necessary for screening, diagnosis, and follow-up care. Malignant and benign breast lesions have the following mammographic characteristics:

• Malignant breast lesions
  • Irregular speculated mass
  • Clustered calcifications
  • Calcifications - Smaller than 0.5 mm in diameter
  • Architectural distortion
  • Focal asymmetric density

• Benign breast lesions
  • Solid- or lucent-centered spheres
  • Smooth and round calcifications
  • Calcifications - Larger than 1 mm in diameter
  • Architectural distortion - Usually not present

• Chest radiograph

• CT scan of the brain, chest, abdomen, and pelvis: Obtain CT scans if the patient has neurologic symptoms, abnormal chest radiograph results, supraclavicular lymphadenopathy and hepatosplenomegaly, or abnormal liver function test results.
• Skeletal radiograph: Use this for symptomatic areas only.
• Bone scan: Perform a bone scan if any of the following conditions are present:

• Advanced local disease

• Lymph node metastases

• Distant metastases

• Bony symptoms

Other Tests:

• Pathologic study of tumor specimens: Three features are evaluated and given scores from 1-3, ie, tubule formation, nuclear pleomorphism, and mitotic activity.

• Tubular grade is defined based on the degree of development of tubular formations.
  • Well differentiated if tubular structures occupy more than 75% of the tumor - 1 point
  • Moderately differentiated if tubular structures represent 10-75% of the tumor - 2 points
  • Poorly differentiated if the tubular structures represent less than 10% of the tumor - 3 points

• The nuclear grade is defined based on the nucleus size, stain density, and shape variations.
  • Small and uniformly staining nucleus, good prognosis - 1 point
  • Moderate variation in nuclear size and shape, intermediate prognosis - 2 points
  • Marked nuclear polymorphism with dark staining, poor prognosis - 3 points

• Determine the cycling fraction (mitotic index and S-phase). While the determination of the S-phase requires the use of flow cytometry, the mitotic index is the easiest and fastest way of assessing proliferation. Score the mitotic index as follows:
  • Low (0-3.3/mm²) - 1 point
  • Medium (3.3-7/mm²) - 2 points
  • High (>7/mm²) - 3 points

• The histologic grade is a composite index obtained by totaling the tubular, nuclear, and mitotic scores. Invasive breast cancer is graded as follows:
  • Well differentiated - 3-5 total points
  • Moderately differentiated - 6-7 total points
  • Poorly differentiated - 8-9 total points

• ER and PR evaluation: Two types of assays are used to quantitate ERs and PRs.

• For ligand-binding methods (eg, dextran-coated charcoal assay), results are expressed in femtomoles of receptor protein per milligram of cytosol protein (fmol/mg). Cutoffs vary from 3-20 fmol/mg, depending on the laboratory. To ensure accuracy and reproducibility, specimens should be large and must be immediately fixed in liquid nitrogen. Results of this method may be affected by the presence of estrogens or tamoxifen in the specimen.

• With regard to monoclonal antibody–based methods (eg, immunohistochemistry [IHC], enzyme immunoassay), IHC has 2 advantages. First, it can be performed on any type or size of specimen, including cell blocks from body fluids or those fixed or imbedded in paraffin. Second, it measures total protein; therefore, it is not affected by the presence of estrogens or tamoxifen. IHC is a semiquantitative technique that depends on the observer and the type of antibodies used. Enzyme immunoassay results are more objective because of the use of a spectrophotometer to quantitate the receptor protein; however, the technique is limited by the need for a sufficiently large fresh, frozen specimen.

• HER2/neu status: Several methods have been used to detect and quantitate HER2/neu.

• IHC methods have been extensively used. This method is a semiquantitative assay using a monoclonal antibody.

• Scoring of HER2/neu overexpression using the DAKO HercepTest (DAKO Cytomation; Carpinteria, Calif) follows. The cell membrane staining pattern, the interpretation, and the score are listed.
  • Strong complete membrane staining in more than 10% of tumor cells - Interpreted as strongly positive; score of 3+
  • Weak-to-moderate complete membrane staining in more than 10% of cells - Interpreted as weakly positive; score of 2+
  • Faintly perceptible membrane staining in more than 10% of tumor cells - Interpreted as negative; score of 1+
  • No staining or staining in 10% of tumor cells - Interpreted as negative; score of 0
  • Cytoplasmic staining of any intensity - Interpreted as negative; score of any

• Detection of gene amplification with fluorescence in situ hybridization (FISH) is highly specific and has an 82% overall concordance rate with IHC. However, when FISH results are expressed in the function of the IHC scores, they are positive in 92% of 3+ and 39% of 2+ IHC specimens. Only 7% of 1+ IHC specimen results are FISH-positive.

• Certain authorities consider FISH to be the criterion standard for HER2/neu evaluation; however, because this test is not readily available in many laboratories, they recommend use of IHC as a first-line test; 3+ and 1+ IHC results correspond to positive and negative expression of HER2/neu, respectively, and a 2+ IHC result is considered borderline. In these cases, only FISH is performed because a significant number of true HER2/neu–positive patients may be identified in the 2+ IHC group.

Diagnostic Procedures:

• Surgical procedures for nonpalpable lesions

• Image-guided core-needle biopsy
  • This is the preferred method for needle biopsy of a nonpalpable lesion. Note that because of sampling error, it carries a higher risk of false-negative findings than open biopsy. Negative or equivocal results in the face of suggestive mammogram findings or residual calcifications should be followed by an open biopsy. False-negative results are encountered in 1-10% of biopsies, with the highest rates occurring with the least-experienced operators.
  • For the technique, ultrasound is the method of choice to guide the core-needle biopsy. Stereotactic mammographic guidance is used in lesions not visualized on ultrasound images. Stereotactic core-needle techniques have the advantages of lower complication rates and lower costs, although they cannot be used when the lesion is very close to the chest wall or areola, where open biopsy is the best approach. The radiograph should be compared with the mammogram to ensure that all calcifications are included within the core biopsy specimen.

• Open biopsy with needle localization
  • Invasive localization techniques with small radiopaque needles to guide a surgical biopsy are used more commonly than noninvasive techniques.
  • For the technique, local anesthesia with or without intravenous sedation is sufficient in most cases. A thin needle and a fine wire with a thickened distal segment are used for immediate preoperative localization of the lesion. The incision may include the wire entry site if the lesion is superficial. A core of tissue along and around the wire is excised (including the lesion easily identified by the previous placement of the thickened segment inside it) and sent en bloc for radiographic evaluation. However, when the wire entry site is located far from the lesion, the incision should be placed directly over the lesion; dissection is then performed to find the wire.
  • Once identified, the free end of the wire is pulled up through the incision. A core of tissue around the wire is excised (including the lesion) and sent for radiographic evaluation. Closure should not proceed until radiographic confirmation that the entire lesion was excised.

• Surgical procedures for palpable lesions

• Fine-needle aspiration biopsy
  • In experienced hands, FNA biopsy may provide a high accuracy rate when combined with physical examination and mammography (sensitivity ~80-98%, specificity ~100%). However, negative results from a palpable lesion cannot exclude carcinoma. Lesions most suited for this procedure are T3 and T4 tumors and axillary or chest wall relapses.
  • Owing to the high false-negative rate with FNA in lesions smaller than 1 cm in diameter, another diagnostic procedure should be used. Because false-positive rates are extremely low (less than 2%), positive results are sufficient to plan surgery, without the need for further investigation. However, perioperative frozen sections are necessary to distinguish between invasive and in situ carcinoma and to determine the need for axillary dissection because FNA results cannot be used to make this distinction.
  • According to some experts, a negative FNA biopsy finding, a physical examination suggestive of a benign lesion, and a normal mammography result (ie, the triple-negative criteria) are sufficient to stop the workup without further investigation.
  • The equipment required is simple, ie, a small needle (22- to 25-gauge) for solid lesions or a larger one (20- to 21-gauge) mounted on a 10-mL syringe with 1-2 mL of air in it for easy visualization of the collected sample, which allows aspiration of a cystic lesion.
  • After sterile preparation of the skin, the needle is advanced toward the lesion while stabilized by the fingers of the nondominant hand. Once the needle is in place, a back-and-forth movement of its tip along a 5- to 10-mm track is applied simultaneously with strong suction on the syringe. The suction is released when the collected sample reaches the syringe or the hub of the needle and before withdrawing the needle to prevent contamination with normal tissue.
  • If the collected sample is fluid, it should be sent for analysis if (1) it is bloody, (2) a residual mass remains after complete aspiration, or (3) a recurrent cyst is present. If the sample is made of cellular material, it should be expelled on glass slides to make thin smears for cytological analysis.

• Cutting-needle (core-needle) biopsy
  • Although the cutting-needle biopsy technique is associated with a higher true-positive rate than the FNA technique, a negative result may reflect sampling error. This biopsy provides a cylinder of tissue for pathological rather than cytological analysis. Determination of ERs and PRs is possible. Its best indications are large tumors and chest wall relapses. Small lesions or those surrounded by fibrocystic tissue are better studied using FNA.
  • After applying local anesthesia to the skin, a small nick is made to allow the entry of the large biopsy needle (14 gauge), which can be placed by biopsy gun or by hand while the nondominant hand is stabilizing the lesion. A core sample is harvested and sent for pathological analysis.

• Excision (open) biopsy
  • The entire lesion is removed with excisional biopsy, along with a margin of normal breast tissue.
  • Local anesthesia, occasionally with intravenous sedation, is sufficient to perform most of these procedures. A curvilinear incision should be placed directly on the tumor mass and oriented in such a way that it could be included within a future mastectomy incision. In extremely lateral or medial lesions, a radial incision placed over the lesion is preferable.
  • Once the tumor is removed, the margins should be inked. After hemostasis is achieved, deep breast tissue approximation is performed only if this does not result in deformity of the breast contour. Lastly, the skin is closed with a subcuticular closure.

• Incisional biopsy: This is indicated for lesions 4 cm or larger and whenever neoadjuvant chemotherapy or RT is contemplated.

Histologic Findings: The various histopathologic types are as follows:

• I - Ductal
  • Intraductal (in situ)
  • Invasive with predominant intraductal component: Infiltrating or invasive ductal cancer is the most common breast cancer histologic type, comprising 70-80% of all cases.
  • Invasive, not otherwise specified
  • Scirrhous
  • Tubular
  • Medullary with lymphocytic infiltrate
  • Mucinous (colloid)
  • Papillary
  • Inflammatory
  • Comedo
  • Other
• II - Lobular
  • In situ
  • Invasive with predominant in situ component
  • Invasive
• III - Nipple
  • Paget disease, not otherwise specified
  • Paget disease with intraductal carcinoma
  • Paget disease with invasive ductal carcinoma
• IV - Undifferentiated carcinoma
• V - Rare tumor subtypes: The following are not considered typical breast cancers.
  • Cystosarcoma phyllodes
  • Angiosarcoma
  • Primary lymphoma

In situ carcinoma

In situ carcinoma is characteristically contained within the epithelium, with the basement membrane intact, and no signs of invasion. Many features distinguish lobular carcinoma in situ (LCIS) from DCIS.

LCIS arises from the terminal duct lobular apparatus and shows a rather diffuse distribution throughout the breast, without evidence of a palpable mass. DCIS originates from the major lactiferous ducts and tends to be a localized disease frequently associated with a palpable mass.

This explains the high incidence of synchronous involvement in the contralateral breast (bilaterality) or in other quadrants of the same breast (multicentricity) in LCIS (90-100%) compared with DCIS (10-15%), suggesting that complete surgical resection of DCIS is not only possible but also desirable, particularly if the index lesion manifests as a palpable mass. Conversely, locoregional modalities, such as surgery or RT, have no role in the treatment of LCIS except for prophylactic bilateral mastectomy.

DCIS originates by proliferation of the ductal luminal cells, which form protrusions into the lumen (papillary DCIS). These become more coalescent, leaving a few empty, rounded spaces (cribriform DCIS). When the lumen is filled with proliferating cells, it becomes completely obliterated (solid DCIS). Central areas of these ducts undergo necrosis because of the ischemic microenvironment (comedo DCIS), with secondary deposition of calcium responsible for the appearance of microcalcifications, a typical radiographic feature of this disease.

DCIS is probably a continuum of successive steps of the same process, with increasing malignant potential as the disease progresses from papillary to comedo forms. Indeed, cells in the earliest stages are well differentiated with no atypia or mitoses, while cells in advanced stages become more anaplastic with frequent mitotic figures.

From a practical standpoint, dividing DCIS into 2 categories, comedo-type and non–comedo-type, is helpful. Both DCIS and LCIS occasionally may be difficult to differentiate from atypical hyperplasia (eg, atypical ductal hyperplasia, atypical lobular hyperplasia), which is a benign change of the mammary gland preceding the in situ disease.

The characteristics of comedo and noncomedo (ie, cribriform, micropapillary, papillary, solid) DCIS are as follows:

• Other terms used
  • Comedo - Poorly differentiated, high-grade
  • Noncomedo - Well differentiated, low-grade
• Ploidy
  • Comedo - Aneuploid (80% of lesions)
  • Noncomedo - Diploid
• Nuclear grade
  • Comedo - High
  • Noncomedo - Low
• Mitoses
  • Comedo - Numerous
  • Noncomedo - Infrequent
• Estrogen receptors
  • Comedo - Negative
  • Noncomedo - Positive
• HER2/neu overexpression
  • Comedo - Present
  • Noncomedo - Absent
• Distribution
  • Comedo - Continuous
  • Noncomedo - Multifocal (gap areas)
• Necrosis
  • Comedo - Present
  • Noncomedo - Absent
• Local recurrence
  • Comedo - High
  • Noncomedo - Low
• Prognosis
  • Comedo - Poor
  • Noncomedo - Good

Invasive ductal carcinoma develops in 30-50% of patients with DCIS over a 10-year period, usually in the same quadrant where the DCIS was found. In contrast, only 10-37% of patients with LCIS develop invasive carcinoma, mostly of the ductal variety and with equal frequency in both breasts. Therefore, DCIS may be considered a true precancerous process, while LCIS is only a marker of increased risk for cancer.

Infiltrating ductal carcinoma with productive fibrosis (scirrhous, simplex, not otherwise specified) represents approximately 80% of ductal carcinoma of the breast cases, with the highest prevalence associated with perimenopause or early postmenopause. Typically, the mass is solitary, firm, and nontender with poorly defined borders. Heterogeneity is a characteristic feature of the malignant cells, which are arranged in single rows, producing the so-called Indian filing. Typical sites of metastases are bone, lung, and liver.

Lobular carcinoma

Invasive lobular carcinoma is the second most common histologic type after ductal carcinoma, accounting for 5-10% of all breast cancers. It is associated with a high rate of multifocality and bilaterality. LCIS is identified in 70-80% of cases. It may manifest as a palpable mass clinically and mammographically indistinguishable from ductal carcinoma, except that the extent of the tumor is often underestimated in lobular carcinoma.

Typical lobular carcinoma is composed of small, homogenous cells that invade the stroma in a single-file pattern. Signet-ring cells may be observed. Their stromal desmoplastic reaction is mild or absent. While ERs and PRs are expressed in the majority of tumors, HER2/neu overexpression and TP53 mutation are rare.

Tumor behavior is characterized by more common bone metastases and less frequent lung, liver, and brain metastases than ductal carcinoma. Metastases to the leptomeninges, peritoneum, retroperitoneum, GI tract, and reproductive system seem to be more common in lobular carcinoma compared with ductal carcinoma. Patients with the classic form of lobular carcinoma share the same prognosis as those who have ductal carcinoma, patients with the tubulolobular variant fare slightly better, and those with the signet-cell variant fare significantly worse than average.

Medullary carcinoma is relatively uncommon (5-7%) and occurs in younger persons. It manifests as a bulky palpable mass, with axillary lymphadenopathy in 40% of patients. Microscopically, the tumor has a syncytial growth pattern, without tubuloglandular differentiation in 75% of the tumor, mixed with intense lymphoplasmacytic infiltrate and, in most cases, associated with reactive lymphadenopathy. Nuclei are large and pleomorphic (grade 2 or 3). DCIS may be observed in the neighboring normal tissues, although there is no increased risk of bilaterality or multicentricity. ER, PR, and HER2/neu are usually negative and TP53 is commonly mutated. The prognosis of patients with this type is usually very good.

Mucinous carcinoma is another uncommon histologic type of invasive breast cancer (less than 5%). It is more common in the seventh decade of life and manifests clinically as a palpable mass or mammographically (with increasing frequency) as a poorly defined tumor with rare calcification. Its histologic hallmark is the presence of mucin production occupying more than 90% of the tumor in pure mucinous forms and variable percentages in the mixed forms. The cells are clustered in small islets dispersed in a pool of mucin. DCIS is often present in the vicinity of the tumor. ERs and PRs are positive in 90% and 68% of cases, respectively. HER2/neu overexpression is extremely rare. Patients with pure mucinous carcinoma have a better prognosis than those with mixed forms or other breast cancers of no special type.

Inflammatory breast cancer is diagnosed clinically based on the association of edema, erythema, and skin ridging (peau d'orange). Although subdermal lymphatic and vascular invasion is nearly constant, it is not a mandatory criterion for diagnosis. The mass is not palpable in most cases. Molecularly, ERs and PRs are negative, HER2/neu is overexpressed, TP53 is commonly mutated, and the thymidine-labeling index is frequently high. Inflammatory carcinoma is an aggressive but fortunately rare disease, with a sudden onset and a rapidly progressive course. It is uniformly fatal if not treated with multimodality therapy. It should be differentiated from benign cellulitis by the characteristic absence of polymorphonuclear leukocytes in the involved area and from locally advanced breast cancer with a secondary inflammatory component, which has a more indolent course and is often responsive to HT.

Nipple

Paget disease of the breast is a relatively rare entity comprising approximately 1% of all breast cancer cases, with the highest incidence in the seventh decade of life. However, pathologic evidence of the disease can be observed in 2-5% of mastectomy specimens. Approximately half the patients present with an underlying mass, which is an invasive cancer in 93% and a DCIS in 7%. In patients with no mass upon presentation, invasive cancer is present in approximately 40% and DCIS in approximately 60%. Histologically, the disease is localized to the epithelium of the nipple-areola complex and the characteristic cells, or Paget cells, are contained within the basement membrane. Paget cells are large, pale cells with prominent nuclei and large nucleoli, dispersed between the keratinocytes as single or clusters of cells.

Other

Tubular carcinoma is an uncommon type with limited metastatic potential and a very good prognosis. The average size of pure tubular carcinomas is smaller than 1 cm, and they are associated with axillary metastases in approximately 15% of cases. Mammographically detected tumors (60-70%) tend to be smaller and less frequently associated with nodal metastases than clinically detected tumors. Characteristic features of this type include a single layer of epithelial cells with low-grade nuclei and apical cytoplasmic "snoutings" arranged in well-formed tubules and glands. Tubular elements comprise more than 90% of pure tubular carcinomas and different proportions of mixed tubular tumors. DCIS is associated with most of these tumors. ERs and PRs are positive in 70-100% and 60-83%, respectively. HER2/neu overexpression and TP53 mutation are very uncommon in this type.

Papillary carcinoma is rare (less than 1% of breast cancers), occurring mainly in postmenopausal women. Histologically, the tumor is circumscribed with cells arranged in delicate or blunt papillae. Nuclei are of intermediate grade. Occasionally, extracellular mucin production can be observed. ERs and PRs are positive in approximately 100% and 80% of cases, respectively. Lymphatic vessel invasion occurs in a third of the cases; however, axillary lymph node enlargement can be related to benign reactive changes in a significant number of cases. This type has a good prognosis. The papillae in the micropapillary type lack the lymphvascular core. This type is associated with lower ER and PR positivity and a higher percentage of HER2/neu overexpression, which explain its worse prognosis. Staging: The American Joint Committee on Cancer staging system groups patients based on the tumor size (T), lymph node status (N), and distant metastases (M) into 4 stages, thus allowing clinicians to derive prognostic information necessary for therapeutic decisions. ER and PR status in the tumor tissue, menopausal status, and the general health of the patient are the other factors required for the final therapeutic plan.

TNM definitions

• Primary tumor

• TX - Cannot be assessed
• T0 - No evidence of primary tumor
• Tis - Carcinoma in situ, intraductal carcinoma, LCIS, or Paget disease of the nipple with no associated tumor (Note: Paget disease associated with a tumor is classified according to the size of the tumor.)
• T1 - Tumor 2 cm or smaller in greatest dimension
  • T1mic - Microinvasion 0.1 cm or less in greatest dimension
  • T1a - Tumor larger than 0.1 cm but not larger than 0.5 cm in greatest dimension
  • T1b - Tumor larger than 0.5 cm but not larger than 1 cm in greatest dimension
  • T1c - Tumor larger than 1 cm but not larger than 2 cm in greatest dimension
• T2 - Tumor larger than 2 cm but not larger than 5 cm in greatest dimension
• T3 - Tumor larger than 5 cm in greatest dimension
• T4 - Tumor of any size with direct extension to (a) chest wall or (b) skin, only as described below (Note: Chest wall includes ribs, intercostal muscles, and serratus anterior muscle, but not pectoral muscle.)
  • T4a - Extension to chest wall
  • T4b - Edema (including peau d'orange) or ulceration of the skin of the breast or satellite skin nodules confined to the same breast
  • T4c - Both of the above (T4a and T4b)
  • T4d - Inflammatory carcinoma (Note: Inflammatory carcinoma is a clinicopathologic entity characterized by diffuse brawny induration of the skin of the breast with an erysipeloid edge, usually without an underlying palpable mass. Radiologically, a detectable mass and characteristic thickening of the skin may be present over the breast. This clinical presentation is due to tumor embolization of dermal lymphatics with engorgement of superficial capillaries.)

• Regional lymph nodes

• NX - Cannot be assessed (eg, previously removed)
• N0 - No regional lymph node metastasis
• N1 - Metastasis to movable ipsilateral axillary lymph node(s)
• N2 - Metastasis to ipsilateral axillary lymph node(s) fixed to each other or to other structures
• N3 - Metastasis to ipsilateral internal mammary lymph node(s)

• Pathologic classification

• pNX - Regional lymph nodes cannot be assessed (eg, not removed for pathologic study or removed previously)
• pN0 - No regional lymph node metastasis
• pN1 - Metastasis to movable ipsilateral axillary lymph node(s)
  • pN1a - Only micrometastasis (none >0.2 cm)
  • pN1b - Metastasis to lymph node(s), any larger than 0.2 cm
    • pN1bi - Metastasis in 1-3 lymph nodes, any larger than 0.2 cm and all smaller than 2 cm in greatest dimension
    • pN1bii - Metastasis to 4 or more lymph nodes, any larger than 0.2 cm and all smaller than 2 cm in greatest dimension
    • pN1biii - Extension of tumor beyond the capsule of a lymph node metastasis, smaller than 2 cm in greatest dimension
    • pN1biv - Metastasis to a lymph node 2 cm or larger in greatest dimension
• pN2 - Metastasis to ipsilateral axillary lymph node(s) fixed to each other or to other structures
• pN3 - Metastasis to ipsilateral internal mammary lymph node(s)

• Distant metastasis

• MX - Cannot be assessed
• M0 - No distant metastasis
• M1 - Distant metastasis present (includes metastasis to ipsilateral supraclavicular lymph nodes)

• American Joint Committee on Cancer stage groupings

• Stage 0 - Tis, N0, M0
• Stage I - T1 (includes T1mic), N0, M0
• Stage IIA
  • T0, N1, M0
  • T1 (includes T1mic), N1 (The prognosis of patients with pN1a disease is similar to that of patients with pN0 disease.), M0
  • T2, N0, M0
• Stage IIB
  • T2, N1, M0
  • T3, N0, M0
• Stage IIIA
  • T0, N2, M0
  • T1 (includes T1mic), N2, M0
  • T2, N2, M0
  • T3, N1, M0
  • T3, N2, M0
• Stage IIIB
  • T4, Any N, M0
  • Any T, N3, M0
• Stage IV - Any T, Any N, M1

TREATMENT

Medical therapy:

Treatment of in situ disease

Ductal carcinoma in situ

Approximately 85% of DCIS is detected mammographically, and this represents 20-30% of mammographically detected breast cancer. The risk of developing invasive cancer is approximately 40% in the ipsilateral breast and 5% in the contralateral breast. Regardless of the initial treatment modality, 50% of recurrences are invasive carcinoma.

Mastectomy cures 98-99% of all types of DCIS, with a recurrence rate of only 1-2%. Most recently, lumpectomy with RT was shown to yield local recurrence rates of 7-13.4%, compared with 26.8-43% for local excision alone. Furthermore, the addition of tamoxifen resulted in a 44% decrease of invasive breast cancer in the ipsilateral breast and a 52% decrease of invasive breast cancer in the contralateral breast.

Predictors of recurrence in patients undergoing BCS for DCIS are as follows:

• Residual microcalcifications: Risk of relapse is 100%.
• Comedo necrosis
• Positive margins: For margins less than 1 mm, the recurrence rate is 25%. For margins of 1-9 mm, the recurrence rate is 15%. For margins greater than 1 cm, the recurrence rate is 3%.
• Age younger than 50 years
• Bloody discharge

Lobular carcinoma in situ

This lesion is usually an incidental finding in breast biopsy specimens. LCIS is not a cancer; it is an indicator for increased risk for breast cancer. This risk is estimated at 1-1.5% per year and 20-30% over a lifetime. Of invasive carcinoma developing in this setting, 50% is ductal carcinoma; the other 50% is lobular carcinoma.

Patients may be observed or offered participation in a chemoprevention trial. Bilateral simple mastectomy with immediate reconstruction is the recommended surgery should the patient elect a radical treatment. Chemotherapy and RT have no role in the treatment of this lesion.

Treatment of invasive disease

Modern treatment of breast cancer is based on a multimodality approach combining surgery, chemotherapy, HT, and RT. Treatment is tailored for an individual patient based on tumor size, axillary lymph node involvement, ER and PR status (the most important variables identified by many historical studies), histologic tumor type, standardized pathologic grade, and menopausal status. The 2000 US National Institutes of Health Consensus Conference updated adjuvant therapy guidelines for breast cancer.

Adjuvant hormonal therapy

Adjuvant HT is indicated only in the presence of hormone receptors (ER and/or PR) on cancer tissue assessed using IHC. Adjuvant tamoxifen has shown a 50% decrease in the risk of breast cancer recurrence and a 28% decrease in breast cancer mortality, while ovarian ablation produced benefits similar to certain chemotherapies (20-25%) in this population.

In 2 cooperative trials (NSABP 23, Intergroup trial 0102), the addition of tamoxifen to chemotherapy was not associated with improvement in disease-free survival and overall survival in patients with ER-negative tumors and was even detrimental in premenopausal patients with ER-negative tumors. Therefore, HT is indicated only in case of ER and/or PR positivity, regardless of age, menopausal status, lymph node status, or tumor size.

The goal of HT in breast cancer is to induce an estrogen deprivation state at the tumor level. This may be achieved by (1) receptor blockade using one of the selective estrogen receptor modulators, such as tamoxifen or toremifene; (2) suppression of estrogen synthesis by aromatase inhibitors (eg, anastrozole, letrozole, exemestane) in postmenopausal women or by luteinizing hormone-releasing hormone analogues (eg, goserelin) in premenopausal women; or (3) ovarian ablation by surgical oophorectomy or external beam radiation therapy in premenopausal women.

Tamoxifen has been the most common form of adjuvant HT used to date. It can be used both in pre and post menopausal women. However, the recent publication of the results of many large aromatase inhibitor (AI) trials (ATAC, BIG 1-98, examestene trial) has shown that AIs are superior to tamoxifen as adjuvant HT in postmenopausal women. Anastrozole and letrozol are approved for use in first line hormonal therapy for HR-positive postmenopausal women and examestene is approved for use sequentially after 2-3 years of tamoxifen. When tamoxifen is chosen, administer 20 mg/d for 5 years. In asymptomatic women, no special screening procedures (eg, transvaginal ultrasound, endometrial biopsies) for endometrial cancer are recommended. The dose of anastrozole is 1 mg a day, of letrozol 2.5 mg and of examestene 25 mg a day.

The Early Breast Cancer Trialists' Collaborative Group's overview analysis suggests that ovarian ablation is effective as adjuvant HT for premenopausal receptor-positive breast cancer patients regardless of nodal status. Premenopausal patients who receive chemotherapy and maintain their ovarian function may benefit from ovarian ablation. If the patient is younger than 50 years and there is a question about her ovarian function, FSH/LH and estradiol should be checked to document her menopausal status.

Adjuvant chemotherapy

Combination chemotherapy is superior to single agents in the adjuvant setting. The body of knowledge about adjuvant chemotherapy for breast cancer has benefited from the serial updates of the Oxford Overview analysis and from other large randomized trials, which have shown slight but statistically significant superiority of anthracycline-containing regimens over traditional CMF. Adjuvant chemotherapy results in an approximately 25% decrease of breast cancer mortality. However, the determination of the anthracycline-containing regimen of choice is still under investigation. For the doses and schedules of the most common regimens used in breast cancer see Table 2.

Doxorubicin (Adriamycin) and cyclophosphamide (AC) has a threshold effect; thus, doses greater than 60 mg/m² and 600 mg/m², respectively, are of no additional benefit. The results of 3 large US trials (NSABP B22 and B25, Cancer and Leukemia Group B 9344) did not support any role for dose intensification of the AC combination. However, 2 other studies, one French (French Adjuvant Study Group -05) and the other Canadian, showed that when epirubicin was escalated in the fluorouracil-epirubicin-cyclophosphamide (FEC) combination, disease-free survival and overall survival were significantly improved in operable breast cancer with positive axillary lymph nodes.

High-dose chemotherapy with stem cell or bone marrow support did not prove superior to standard chemotherapy and is best reserved for clinical trials.

Results from two trials (Cancer and Leukemia Group B 9344 and NSABP-B28) exploring the role of taxanes in the adjuvant setting were encouraging. CALGB-9344 had a 3x2 factorial design with patients randomized to receive doxorubicin at 60, 75, or 90 mg/m² followed by paclitaxel at 175 mg/m² or no additional therapy. NSABP-B28 had similar design except that the dose of doxorubicin was not escalated and the dose of paclitaxel was 225 mg/m². From the first trial it was established unequivocally that doxorubicin dose escalation does not improve outcome while the addition of paclitaxel results in small but statistically significant improvement of both the risk of relapse and the risk of death. In NSABP-B28, relapse free survival was significantly improved by the addition of paclitaxel while the data about survival are not available yet.

In patients with node-negative, early-stage disease (stages I, IIA, and IIB), chemotherapy is not indicated if the tumor is smaller than 0.5 cm, regardless of the histologic subtype. In persons with invasive ductal and lobular carcinomas, tumors measuring 0.6-1 cm require chemotherapy only if they were associated with unfavorable features (eg, angiolymphatic invasion, high S-phase, high nuclear grade, high histologic grade), while all tumors larger than 1 cm require adjuvant chemotherapy alone (hormone-receptor negative) or in combination with tamoxifen (hormone-receptor positive). In patients with other histologic types (ie, tubular, colloid, medullary, adenoid), chemotherapy is indicated only for tumors larger than 3 cm. It may be considered in those from 1-2.9 cm, and it is not indicated for tumors smaller than 1 cm.

In patients with node-positive, early-stage disease, adjuvant chemotherapy with or without adjuvant HT is the mainstay of treatment. Anthracycline-containing chemotherapy, such as doxorubicin (Adriamycin) (60 mg/m²) and cyclophosphamide (600 mg/m²), administered for 4 cycles is the treatment of choice. Many oncologists consider this treatment insufficient and opt for 2 more cycles of the same treatment or 2-4 cycles of paclitaxel (175 mg/m²).

In persons with advanced-stage disease (stage IIIA and IIIB), the same regimen of AC for 4 cycles followed by at least 2 more cycles of the same treatment or 4 cycles of a taxane is recommended after surgical treatment. Neoadjuvant chemotherapy may be offered to patients with this stage.

The role of neoadjuvant chemotherapy is being intensively investigated. The NSABP B-18 trial compared the effect of preoperative AC for 4 cycles to the same regimen postoperatively in subjects with early-stage breast cancer. Pathologic node-negativity of 60% and BCS of 68% could be achieved in the preoperative AC group compared with 42% and 60%, respectively, in the postoperative AC group. Five-year disease-free survival and overall survival rates for those who achieved pathologic complete response were 84% and 87% compared with 72% and 78%, respectively, in patients who had residual disease.

Based on these encouraging results, NSABP started its B-27 trial to address the question related to the role of 4 cycles of preoperative or postoperative docetaxel added to 4 AC cycles. The use of preoperative docetaxel almost doubled the pathologically confirmed complete remission rate compared with the AC arm (25.6% vs 13.7%).

Because the role of neoadjuvant therapy is not fully established, participation in clinical trials should be encouraged. In ER- and/or PR-positive tumors, neoadjuvant tamoxifen or letrozole may achieve the same magnitude of response as chemotherapy but with longer time to response. Although the neoadjuvant approach does not seem to prolong survival, its theoretical advantages include downstaging the tumor, in vivo testing of the chemosensitivity of the tumor, and allowing BCS.

The treatment of metastatic disease is mainly medical. In postmenopausal patients with ER- and/or PR-positive tumors, the use of tamoxifen or an aromatase inhibitor (eg, anastrozole, letrozole, exemestane) is the standard of care for bone disease and limited visceral disease. In many cases, at least a partial response can be achieved. Patients who progress on first-line HT may still respond to second- or third-line HT, aromatase inhibitors, and megestrol acetate, respectively.

ER down-regulators are a promising new class of HT agents that may have some efficacy if the previous treatments fail. In premenopausal women with ER- and/or PR-positive breast cancer, ovarian ablation and tamoxifen are the mainstays of treatment. Ovarian ablation can be achieved by medical (eg, luteinizing hormone-releasing hormone analogues such as goserelin, leuprolide, buserelin, and triptorelin), surgical (ie, bilateral oophorectomy), or RT methods.

Chemotherapy is indicated for patients with advanced visceral disease (visceral crisis) and those with hormone-refractory or hormone-insensitive tumors. The goals of chemotherapy in this setting are palliative and include control of symptoms, control of disease progression, and prolongation of life. The best response rates are obtained with first-line therapy and combination regimens. However, regardless of the response rates achieved with these therapies (including high-dose chemotherapy with autologous stem cell support), survival is not affected in most cases.

In 2001, Slamon and Pegram reported a survival advantage for the combination of chemotherapy and trastuzumab over chemotherapy alone in women with HER2-positive metastatic breast cancer.

Table 1. Single-Agent Chemotherapy for Metastatic Breast Cancer

Drug	Class	Dose/Schedule	Overall Response Rate	Toxicity	Comments
Paclitaxel	Microtubule-stabilizing	175 mg/m2 as 3-h infusion q3wk	21-32%	Myelosuppression, alopecia, neuropathy, myalgias, arthralgias, and allergic reactions	Corticosteroids are used to prevent allergic reactions, especially after first doses. They may be omitted later if no adverse effects occur.
		80-100 mg/m2/wk	23-53%	Mild and noncumulative myelosuppression; with prolonged use, neuropathy, fatigue, and fluid retention are limiting toxicities
Docetaxel	Microtubule-stabilizing	75-100 mg/m2 q3wk	30-68%	Myelosuppression, alopecia, skin reaction, mucositis, and fluid retention	Corticosteroids are used to prevent fluid retention. Initiate diuretics with the first signs of fluid retention.
		35-45 mg/m2/wk for 6 cycles, with 2 wk off	29-50%
Doxorubicin	Anthracycline	45-60 mg/m2 q3wk; not to exceed cumulative dose of 450-500 mg/m2	35-50%	Myelosuppression, mucositis, nausea, vomiting, and myocardial dysfunction	Dexrazoxane may be used as a cardioprotectant.
Epirubicin	Anthracycline	90 mg/m2 q3wk; not to exceed cumulative dose of 900 mg/m2	35-50%	Myelosuppression, mucositis, nausea, vomiting, and myocardial dysfunction
Doxil (liposomal encapsulated doxorubicin)	Liposomal anthracycline	20 mg/m2 IV q3wk		Less cardiotoxicity, neutropenia, and alopecia; stomatitis and hand-foot syndrome
Capecitabine	Oral fluoro-pyrimidine	2500 mg/m2/d for 2 wk, with 1 wk off	25-30%	Rash, hand-foot syndrome, mucositis, diarrhea, and mild neutropenia
Vinorelbine	Microtubule-assembly inhibitor	25-30 mg/m2/wk	35-45%	Myelosuppression, neuropathy, alopecia, constipation, fatigue, and phlebitis
Carboplatin	Same as cisplatin	Doses calculated for an area under the curve of 5 or 6.		Neuropathy, nausea, vomiting, kidney dysfunction, and mild myelosuppression

Predictors of poor response to chemotherapy in patients with metastatic breast cancer are poor performance status, multiple and/or visceral sites of disease, short disease-free intervals, and failure to respond to prior chemotherapy. Because the goals of chemotherapy in patients with metastatic breast cancer are palliative, many authorities recommend the sequential use of single chemotherapeutic agents rather than combinations in order to limit toxicity. Taxanes, anthracyclines, oral fluoropyrimidines, vinorelbine, and gemcitabine are the most effective drugs used in this setting.

Trastuzumab, a monoclonal antibody directed against the extracellular domain of the HER2/neu receptor, has shown significant antitumor activity in patients with metastatic breast cancer overexpressing HER2/neu. Response rates of 30-35% have been observed in patients with metastatic breast cancer who are receiving single-agent trastuzumab as a first-line therapy. The relative risk of death was decreased by 20% with a median follow-up of 30 months when trastuzumab was used in combination with chemotherapy. However, when combined with doxorubicin, a significant increase in cardiotoxicity was noted. For this reason, the current recommendation is to avoid trastuzumab in combination with or after doxorubicin.

The mechanism of action of trastuzumab is still debated. Trastuzumab induces down-regulation of HER2/neu and prevents its heterodimerization, reestablishing breast cancer cell sensitivity to HT and chemotherapy. As a result of this down-regulation, p27 is induced, resulting in cell-cycle arrest in the G1 phase. Furthermore, trastuzumab binds to the receptor at a site where the extracytoplasmic domain (ECD) is usually cleaved by metalloproteinases. The ECD cleavage results in a constitutively active truncated receptor and a more aggressive phenotype; this would be prevented by binding to trastuzumab. Finally, trastuzumab may induce antibody-dependent cytotoxicity, resulting in cell death.

Four large multicenter trials were conducted with trastuzumab in the adjuvant setting in patients with HER2/neu-positive breast cancer (NSABP B31, NCCTG N9831, The HERA trial and BCIRG 06). A combined analysis of the first two trials showed dramatic improvement of DFS in the group randomized to receive trastuzumab with chemotherapy consisting of AC x 4 followed by T x 4.

Radiotherapy

RT reduces the risk of local recurrence and has the potential to decrease long-term mortality from breast cancer. Although certain studies have shown that RT following chemotherapy results in better long-term survival rates than the opposite, recent updates from Bellon et al and the Joint Center randomized trial showed no significant differences between the CT-first and RT-first arms in any endpoint studied. A reduction of approximately 20% in local recurrence correlates with an absolute reduction of approximately 5% in long-term mortality from breast cancer 10-15 years later.

RT to the breast (with or without the supraclavicular area) is indicated after lumpectomy in persons with early-stage breast cancer as an integral part of the treatment plan, and it is indicated after mastectomy in the presence of a large tumor mass (>5 cm), positive margins, and 4 or more lymph nodes positive for disease.

The role of axillary RT after radical ALND is debated. Many experts believe that RT is best avoided after complete dissection of the axilla for level I, II, and III nodes (RM and modified RM). A 6- to 8-fold increase in the incidence of lymphedema in the ipsilateral arm was reported with the combined modality.

Table 2. Adjuvant Chemotherapy Regimens for Breast Cancer

Regimen	Dose and Schedule	Cycle Interval, d	Cycles
CMF (standard) Cyclophosphamide Methotrexate 5-Fluorouracil	100 mg/m2/d PO for 14 d 40 mg/m2/d IV days 1 and 8 600 mg/m2/d IV days 1 and 8	28 28 28	6 6 6
CMF (IV; in node-negative patients) Cyclophosphamide Methotrexate 5-Fluorouracil	600 mg/m2 IV 40 mg/m2 IV 600 mg/m2 IV	21 21 21	12 12 12
CAF Cyclophosphamide Doxorubicin (Adriamycin) 5-Fluorouracil	100 mg/m2/d PO for 14 d 30 mg/m2/d IV days 1 and 8 500 mg/m2/d IV days 1 and 8	28 28 28	6 6 6
CAF Cyclophosphamide Doxorubicin 5-Fluorouracil	600 mg/m2 IV day 1 60 mg/m2 IV day 1 600 mg/m2/d IV days 1 and 8	21-28 21-28 21-28	4-6 4-6 4-6
AC Doxorubicin Cyclophosphamide	60 mg/m2 IV day 1 600 mg/m2 IV day 1	21 21	4 4
AC followed by paclitaxel (Taxol) Doxorubicin Cyclophosphamide Paclitaxel	60 mg/m2 IV day 1 600 mg/m2 IV day 1 175 mg/m2 IV day 1	21 21 21	4 4 4 (after AC)
AC followed by CMF Doxorubicin Cyclophosphamide Methotrexate 5-Fluorouracil	75 mg/m2 IV day 1 600 mg/m2 IV day 1 40 mg/m2 IV day 1 600 mg/m2 IV day 1	21 21 21 21	4 8 (cycles 5-12) 8 (cycles 5-12) 8 (cycles 5-12)

Surgical therapy:

Procedures

RM is the en bloc resection of the breast, the overlying skin, the pectoralis muscles, and all of the axillary contents (level I, II, and III dissection). Extended RM is RM with removal of the internal mammary nodes. Both procedures are rarely performed in current practice. In addition to their mutilating effects, they result in functional impairment as a result of neurologic and lymphatic vessel damage. Resection of the involved portion of the muscle is satisfactory even when the tumor abuts or invades the pectoral fascia or when the muscle fascia was violated at the time of biopsy.

Modified RM implies the removal of breast tissue, the underlying fascia of the pectoralis major muscle, and some of the axillary lymph nodes. The nipple-areola complex and the area around the biopsy incision must be removed, but the remainder of the skin of the breast can be preserved.

Total mastectomy removes the entire breast. Both pectoralis muscles and the axillary nodes are preserved.

Skin-sparing mastectomy removes the breast parenchyma, the previous biopsy site(s), and the skin overlying superficial tumors through a periareolar incision. Thus, the breast envelope is preserved, allowing immediate reconstruction with improved cosmetic results. Skin-sparing mastectomy can be coupled with SLNB.

BCS (lumpectomy) removes the tumor with 0.5-1 cm of normal tissue. It is often coupled with ipsilateral axillary lymph node sampling and, increasingly, with SLNB. Quadrantectomy is the removal of the quadrant containing the cancer with the overlying skin and underlying pectoral fascia.

ALND, when performed in a 2-cm primary breast cancer to levels I and II, reduces the probability for regional recurrence from 20% to approximately 3%.

Regarding SLNB, the combined use of a radionuclide and a blue dye increases the detection rate of the SLN. Isosulfan blue (Lymphazurin) is the most commonly used blue dye. Five mL of isosulfan blue are injected into the parenchyma, and the injection site is gently massaged for 3-5 minutes before the SLNB procedure is performed. Technetium sulfur colloid, the radionuclide of choice, is used at a dose of 0.1-1 mCi. The isotope is injected into the parenchyma 1-2 hours before the procedure. The SLN is preoperatively localized using a handheld gamma probe.

Preoperative details: Conduct a thorough preoperative assessment, and discuss with the patient and family the risks related to the cancer itself or to any comorbid conditions that may impact the intended surgery. Focus this discussion on prognostic factors, therapeutic options for the stage of the disease (with their expected benefits and adverse effects), and alternatives to surgery if available.

With young patients or those who desire a reconstruction, also discuss immediate versus delayed plastic surgery. For all patients with locally advanced disease, seek consultation with a medical oncologist to consider the role of neoadjuvant chemotherapy.

Intraoperative details: Whenever BCS is performed, the surgeon has the responsibility to ensure that all involved tissues are removed. Frozen-section studies are performed perioperatively, followed by permanent-section studies. Perform total mastectomy if the margins are not clear or if the specimen contains multicentric disease.

Lumpectomy

Lumpectomy or wide local excision may be performed with the patient under local anesthesia if no axillary node dissection is planned. General anesthesia is preferred for large excisions or if axillary dissection is intended. A long, curvilinear incision is placed directly over the lesion in such a way as to allow its inclusion in a future mastectomy incision if the margins are positive.

A radial incision is better suited for lesions located far laterally or in the areolar area. The dissection is performed around the palpable tumor, which is removed with 1-1.5 cm of grossly normal tissue. In cases of a nonpalpable lesion, remove 2-3 cm of tissue around the localization wire.

Exercise care for proper orientation and margin inking of the specimen to ensure appropriate pathological study of the margins. Reexcision can be performed immediately if instantaneous study results are available, or it can be performed later. Subcuticular closure is performed after appropriate hemostasis and approximation of deep tissues if the latter does not result in major deformity of the breast contour.

Axillary dissection

In axillary dissection, include level I and II lymph nodes and axillary fibrofatty tissue. Level III lymph nodes are preserved unless gross disease is present. The borders of the dissection are the axillary vein and the medial aspect of the pectoralis minor muscle superiorly, the latissimus dorsi muscle laterally, the pectoral muscles and the anterior serratus muscle medially, and the tail of the breast inferiorly.

Make a transverse incision in the lower third of the hair-bearing skin of the axilla; the incision can be extended posteriorly to the latissimus dorsi muscle for larger exposure. The dissection proceeds through the true axillary fat pad and its fascia to allow exposure of major pectoralis and latissimus dorsi muscles. Recognition of anatomical landmarks such as the axillary vein, the thoracodorsal bundle, and the long thoracic nerve help to maintain proper orientation and to preserve important structures.

Low-lying branches of the brachial plexus are preserved after identification of the axillary vein under the pectoralis major muscle medially. The thoracodorsal bundle is identified at its junction with either the axillary vein or the latissimus dorsi muscle. Medial to this bundle, the long thoracic vein is recognized lying on the chest wall.

Retraction is applied on the pectoralis major and latissimus dorsi muscles to ensure good exposure. The fat and the lymph nodes located medial to the thoracodorsal bundle are removed up to the level of the axillary vein, and then the dissection proceeds medially to level II fat and nodes beneath the pectoralis major muscle, where the medial pectoral nerve may be encountered.

Dissection of the axillary fat continues inferiorly along the anterior serratus muscle, where the long thoracic nerve is identified and preserved. Lastly, the axillary fat is released from its attachment to the tail of the breast by knife or electrocautery. Most surgeons drain the axilla after axillary dissection.

Sentinel lymph node biopsy

This procedure can be performed with the patient under local or general anesthesia. The isotope and the blue dye are injected 2 hours and 5 minutes before the procedure, respectively. Through an incision made below the "hot spot" identified by the gamma probe, the surgeon can identify a blue-green lymphatic representing the afferent lymphatic leading to the SLN.

The removed SLN is examined ex vivo using the gamma probe to determine its count, which serves as a reference for a reexamination of the axilla. Remove hot spots with counts exceeding 10% of the hottest ex vivo node. In patients undergoing lumpectomy, perform the SLNB before the lumpectomy; for those undergoing modified RM, perform the procedure during ALND before or after the mastectomy.

Mastectomy

Mastectomy (modified radical or simple) implies removing all the breast tissue and the overlying skin, including the nipple-areola complex, and leaving viable skin flaps. Whether a transverse or an upward-angled incision is chosen, include the nipple-areola complex and previous biopsy incisions in the excised area.

After determining upper and lower endpoints and while the breast is pulled firmly downward, a straight line is drawn to define the upper incision. The lower incision is similarly drawn while the breast is pulled upward.

Skin flaps are started by dissecting through the avascular plane between the subcutaneous fat and the breast tissue and are then followed down by dividing the Cooper ligaments. Perform the dissection medially to the sternum, inferiorly to the inframammary fold, laterally to the latissimus dorsi muscle, and superiorly to the clavicle. Include in the specimen the fascia of the pectoralis major muscle, which constitutes the deep margins of the dissection.

In a simple mastectomy, the dissection is stopped before the axillary fat pad is entered. In a modified RM, and according to the surgeon's preference, the dissection may start with the breast and proceed through the axilla or it may start in the axilla and finish in the breast. Avoid excessive hemostasis to preserve the viability of the flaps. Closure is performed without tension or redundant skin after placement of 2 closed suction drains.

Postoperative details: Offer breast reconstruction to patients with breast cancer as an integral part of the multidisciplinary therapeutic approach. It may be immediate or delayed.

Immediate reconstruction is offered to patients with stage 0, I, or II disease. Its psychological, esthetic, and practical advantages outweigh its disadvantages, which mainly include the need for a multidisciplinary team collaborating during the same operative time and the absence of final histology results during surgery. Complications are similar to those of ablative surgery alone. Immediate reconstruction can be combined with any type of ablative surgery, including modified RM. Delayed reconstruction can be performed in all breast cancer stages. It usually follows chemotherapy and RT.

Two reconstructive techniques are used: implant insertion and autologous graft. Implants are made of a silicone shell filled with silicone gel or saline. Although the surgical procedure itself is short and easy, long-term complications are common and include capsular contracture/wrinkling, skin necrosis, and leakage (33%, 8.5%, and 6%, respectively). Because of this high rate of problems, multiple revisions are required, which can add significantly to the total cost. Implants are indicated for small- to medium-sized breasts, in women with poor general health, and in those with short life expectancy. They are contraindicated in patients who require chest wall RT, owing to the high rate of complications.

Although breast reconstruction using autologous tissue is more complex, its advantages far outweigh those of the implant. Esthetically, breast reconstruction results in a warm, soft texture of the breasts with good imitation of natural ptosis. Results are long-lasting, with a low rate of complications and subsequent procedures. Breast reconstruction is probably the only valid option for patients with partial mastectomy and those who require RT.

The most commonly used flaps are the pedicled latissimus dorsi myocutaneous flap, the thoracodorsal artery perforator skin-flat flap, the pedicled or free transverse rectus abdominis myocutaneous flap, the free deep inferior epigastric perforator skin-flat flap, the superior and inferior gluteus myocutaneous flap, and the free superior gluteal artery perforator skin-flat flap. The choice of the donor site and type of procedure depend not only on local conditions at the donor and recipient sites but also on the surgeon's experience.

COMPLICATIONS

• Mastectomy
  • Wound infection and abscess
  • Necrosis of skin flap
  • Paresthesia of chest wall
  • Phantom breast syndrome
  • Postsurgical pain syndrome
  • Seroma
  • Lymphedema
• Axillary lymph node dissection
  • Injury to or thrombosis of the axillary vein
  • Seroma formation
  • Lymphedema: The reported prevalence rate of lymphedema is approximately 11%, with extremes ranging from 5-30%. Extensive surgery, RT, and advanced age are recognized risk factors for arm edema. Although the risk may decrease with time, it does not completely disappear.
  • Impairment of shoulder movements: Depending on whether the patient has received RT to the axilla, the incidence rate varies from 12-15% (RT) and 7-8% (no RT). Symptoms include decreased range of motion of the shoulder, a problem that may be improved with early participation in a physical therapy program.
  • Damage to the brachial plexus, with chronic pain and varying degrees of decreased grip strength occurring in up to 15% of patients and lasting for more than a year after surgery
  • Chest wall pain
• Chemotherapy
  • Cyclophosphamide: Adverse effects may include hemorrhagic cystitis, and amenorrhea.
  • Methotrexate: Adverse effects may include liver toxicity, increased toxicity in the presence of pleural effusion, and ascites.
  • Fluorouracil: Adverse effects may include mucositis, hand-foot syndrome, and cerebellar ataxia.
  • Doxorubicin: Adverse effects may include myocardial dysfunction, alopecia, nausea, vomiting, mucositis, and neutropenia.
  • Paclitaxel: Adverse effects may include myelosuppression, peripheral neuropathy (less common if less than 170 mg/m² is used), hypersensitivity reaction (premedication with steroids, H1- and H2-receptor blockers), cardiac toxicity, alopecia, mucositis, nausea, vomiting, and typhlitis.
  • Docetaxel: Adverse effects may include myelosuppression, mucositis, conjunctivitis, edema due to capillary leak syndrome (>80% of patients if not medicated; less than 10% if premedicated with steroids), hypersensitivity reactions, neurotoxicity (less frequent than with paclitaxel), nausea, vomiting, and alopecia.
• Radiation therapy
  • Necrosis of the breast soft tissue, prolonged breast edema, rib fracture (low rates, 1-3%)
  • Decreased shoulder mobility (low rates, 1-3%)
  • Brachial plexopathy with paresthesia and arm pain (low rates, 1-3%)
  • Lymphedema
  • Angiosarcoma: The 30-year cumulative risk is lower than 1%, with a peak incidence at 6 years.
  • Lung cancer: Ipsilateral lung cancer may occur, with an increased risk in persons who smoke.
  • Coronary artery disease: The risk has significantly decreased with newer RT techniques.
  • Symptomatic pneumonitis: It is relatively infrequent, affecting 3-6% of women treated with RT for breast cancer. Patients present 3-12 months after competition of RT with dry cough, dyspnea, and low-grade fever. Chest radiographs may reveal intersitial infiltrate in the ipsilateral lung, which can evolve to fibrosis.
• Adverse effects of tamoxifen
  • Endometrial cancer: This rare complication occurs in 2 of every 1000 women receiving tamoxifen. Most of these cases are detected at an early stage and are easily cured by surgery. Other than yearly gynecologic examinations in asymptomatic women, the National Cancer Institute and the American Society of Obstetricians and Gynecologists recommend prompt evaluation with endometrial biopsy in women on tamoxifen who experience vaginal bleeding.
  • Perimenopausal symptoms: Hot flashes and mood changes may occur in women on tamoxifen and occasionally are very severe, thus posing a serious threat to the woman's quality of life. Selective serotonin uptake inhibitors have been used with some success to treat these symptoms.
  • Cataracts: These have also been reported in women receiving tamoxifen, justifying a yearly eye examination.
• Adverse effects of trastuzumab
  • Cardiac toxicity: In the phase III trial, trastuzumab alone resulted in cardiac dysfunction in 7% (class III/IV, 1.5%) of cases. The prevalence rate increased to 11% when trastuzumab was administered with paclitaxel (class III/IV, 0%). With the combination of anthracycline and cyclophosphamide, the prevalence rate increased to 28% (class III/IV, 6%). Apparently, the combination of an anthracycline and trastuzumab is particularly cardiotoxic. The mechanism of this dysfunction is yet to be defined.

    Although HER2/neu is minimally expressed on the cardiomyocytes, trastuzumab-induced phosphorylation of HER2 has been documented in vitro and is associated with cytoskeletal infrastructure dysfunction, which seems to be the underlying mechanism. These abnormalities may be reversed by adding neuregulin to the culture milieu. Clinically, trastuzumab-induced cardiomyopathy is partly reversible by medical therapy.

  • Fever, chills, nausea, vomiting, and pain with first infusion: These are relatively common but become infrequent with subsequent infusions.

OUTCOME AND PROGNOSIS

Prognostic factors help predict the clinical outcome of the disease, while predictive factors are those that help predict the response to therapy. Certain factors are both prognostic and predictive.

Prognostic factors

Lymph node status is a significant prognostic factor. Axillary lymph node involvement and the number of lymph nodes involved remain the most important prognostic factors for invasive breast cancer. Although 75% of the lymphatic drainage from the breast goes to the axilla and 25% to the internal mammary lymph nodes, isolated metastasis to the internal mammary lymph nodes is extremely rare (~5%).

The prognosis depending on the number of axillary lymph nodes involved in patients who received adjuvant chemotherapy is as follows:

• With 0 positive nodes
  • Recurrence rate at 5 years - Approximately 20%
  • Survival rate at 10 years - 65-80%
• With 1-3 positive nodes
  • Recurrence rate at 5 years - 30-40%
  • Survival rate at 10 years - 35-65%
• With 4 positive nodes
  • Recurrence rate at 5 years - Approximately 44%
  • Survival rate at 10 years - Not available
• With more than 4 positive nodes
  • Recurrence rate at 5 years - 54-82%
  • Survival rate at 10 years - 13-24%

Additionally, tumor size is highly correlated with lymph node involvement and clinical outcome. Tumor size and the percentage of axillary node involved is as follows:

• Tumor smaller than 0.5 cm - Approximately 20%
• Tumor 0.5-0.9 cm - Approximately 20%
• Tumor 1-1.9 cm - 33%
• Tumor 2-2.9 cm - 45%
• Tumor 3-3.9 cm - 52%
• Tumor 4-4.9 cm - 60%
• Tumor larger than 5 cm - 70%

The 5-year survival rate based on tumor size and axillary lymph node status is as follows:

• Tumor smaller than 2 cm
  • Negative nodes - 96%
  • One to 3 positive nodes - 87%
  • Four or more positive nodes - 66%
• Tumor 2-5 cm
  • Negative nodes - 89%
  • One to 3 positive nodes - 79%
  • Four or more positive nodes - 58%
• Tumor larger than 5 cm
  • Negative nodes - 82%
  • One to 3 positive nodes - 73%
  • Four or more positive nodes - 45%

Hormone receptor status helps predict the prognosis. The expression of ERs and/or PRs portends a good prognosis. It also helps predict the response to HT.

Finally, the histopathologic grade helps predict the patient's prognosis. This is a composite index based on nuclear differentiation (nuclear grade I, II, III; the higher, the worse), histologic differentiation (I, II, III; the higher, the worse), and mitotic index (I, II, III; the higher, the worse).

New prognostic or predictive factors

Cancers overexpressing HER2/neu are frequently poorly differentiated and lymph node–positive. HER2/neu overexpression correlates with more aggressive behavior and shortened disease-free survival and overall survival rates. HER2/neu overexpression is a marker of response to chemotherapy and HT, ie, relative resistance to tamoxifen and CMF and sensitivity to anthracycline- and taxane-containing regimens. HER2/neu overexpression predicts the response to trastuzumab and to monoclonal anti-HER2/neu antibodies in the metastatic setting. A few recent reports suggest that high serum HER2/neu levels predict resistance to HT. HER2 shedding may result in constitutive activation of its cytoplasmic domain and a more aggressive phenotype, with possible estrogen-independent activation of ER receptors, hence the resistance to tamoxifen.

Other predictive or prognostic factors include the EGF receptor family, the S-phase, and DNA ploidy. Overexpression of the EGF receptor family is inversely correlated with ER positivity and is usually associated with a poor prognosis. A high S-phase indicates a rapid proliferation rate and is associated with a worse prognosis. Because most breast cancers are aneuploid, the significance of aneuploidy awaits definition. However, diploid tumors are usually associated with a good prognosis.

Bone marrow micrometastasis is an area of current active research. Occult bone marrow micrometastasis has been discovered to help predict disease-free survival and overall survival of breast cancer patients with both lymph node–positive and lymph node–negative disease. The presence of bone marrow micrometastasis is associated with larger tumor size, higher tumor grade, and lymph node involvement. Bone marrow micrometastasis is an independent prognostic indicator with a predictive value superior to axillary lymph node status, tumor stage, and tumor grade.

The last of the new predictive or prognostic factors includes angiogenesis, peritumoral lymphatic invasion and perineural invasion, cathepsin D, and obesity.

FUTURE AND CONTROVERSIES

Hormonal therapy

The selective estrogen receptor modulator family, headed by tamoxifen, has been enlarged by the addition of new members (ie, toremifene, raloxifene, arzoxifene, lasofoxifene, idoxifene, EM-652, GW5638). Some have received approval for the same indications as tamoxifen (toremifene), and others are still the subjects of intense investigation. The introduction of aromatase inhibitors to the armamentarium of HT, first in the metastatic setting and more recently in the adjuvant setting, has opened new horizons for this modality. The new class of selective estrogen receptor down-regulators is being evaluated in clinical trials, and results are promising. Lastly, the rehabilitation of ovarian ablation with the use of luteinizing hormone-releasing hormone analogue has added another layer of complexity to modern HT for breast cancer.

The publication of the results of many aromatase inhibitor clinical trials including the ATAC, BIG 1-98 and the examestene trials evaluating the role or anastrozole, letrozol and examestene, respectively has announced the end of the tamoxifen domination era in the adjuvant setting. Furthermore, aromatase inhibitors have shown the capability to significantly reduce the prevalence of new contralateral breast primaries, paving the way for their use in chemoprevention trials. However, the following questions must be addressed.

• Which selective estrogen receptor modulators will be most effective in adjuvant and prevention settings?
• What is the role of selective estrogen receptor down-regulators?
• Which aromatase inhibitor is the best adjuvant therapy for breast cancer and if these agents are equivalent in terms of efficacy are they different in terms of toxicity?
• Using modern biotechnology (ie, genomics, proteomics), will it be possible to predict the responsiveness to any of these hormonal therapies?
• Will prolonged treatment of aromatase inhibitors beyond 5 years be superior to 5 years of therapy (more than 5 years of tamoxifen is not superior to 5 year) ?
• What is the best sequence to provide these patients with the longest survival possible?
• What are the patterns of resistance to HT, and are there methods available to overcome them?
• Will long-term use of aromatase inhibitors result in an increased prevalence of skeletal or other complications?

The overview analysis from the Early Breast Cancer Trialists' Collaborative Group has shown that ovarian ablation results in significant improvement of disease-free survival and overall survival, similar to cytotoxic chemotherapy. Because chemotherapy itself results in ovarian failure in 30-40% of patients younger than 40 years and in 70-90% of patients older than 40 years, the relative contribution of chemotherapy and ovarian ablation is difficult to determine. Several ongoing adjuvant trials are examining the role of ovarian ablation versus chemotherapy, the role of combined chemotherapy and ovarian ablation, and, lastly, the role of total estrogen ablation using tamoxifen and ovarian ablation.

Chemotherapy

Several concepts are being explored in this area.

The precise role of taxanes in the adjuvant setting is still awaiting the results of completed (ie, Breast Cancer International Research Group 001, Eastern Cooperative Oncology Group 2197) or ongoing trials. Three ongoing adjuvant trials are examining the role of sequential administration of taxanes following AC chemotherapy in early-stage breast cancer (Cancer and Leukemia Group B 9344, NSABP B-28, Eastern Cooperative Oncology Group 1199). Four other adjuvant trials are exploring the potential of doxorubicin/paclitaxel combinations (Breast Cancer International Research Group 001, Eastern Cooperative Oncology Group 2197, NSABP B-30, Breast Cancer International Research Group 005). Three trials are testing the effect of sequential anthracycline-containing regimens and paclitaxel in the neoadjuvant setting (NSABP B-27, M.D. Anderson Cancer Center, Aberdeen).

Many trials are addressing the role of trastuzumab in the adjuvant setting. The development of trastuzumab was a major triumph of transnational research into finding targeted therapy for cancer. Consequently, once this new drug proved its efficacy in the metastatic setting, it followed that researchers tried to include it in adjuvant regimens. However, to overcome its unexpected cardiac toxicity, new chemotherapy combinations and trial designs had to be sought.

Preclinical and clinical data have indicated dramatic synergistic interactions, with platinum salts and docetaxel forming the basis for an ongoing trial comparing sequential AC and docetaxel with or without trastuzumab to cisplatin or carboplatin with docetaxel and trastuzumab (Breast Cancer International Research Group 006).

Two other trials are examining the efficacy of 4 cycles of AC followed by paclitaxel in different schedules with or without trastuzumab (NSABP B-31, North Central Cancer Treatment Group 9831). Vaccine trials are another type of trial targeting HER2/neu in the adjuvant setting. The theory that maximum efficacy of cancer vaccines is more likely to occur in minimal residual disease than in bulky disease is now accepted.

The success of bisphosphonates in reducing skeletal complications in persons with metastatic breast cancer has suggested a possible role for these noncytotoxic agents in the adjuvant setting. Previous trials have also shown a possible reduction of recurrences in sites other than bone and a survival benefit. To test the potential of this therapy in the adjuvant setting, 3 trials are underway (NSABP B-34, Southwest Oncology Group S9905, CLB-79809).

Postmastectomy radiotherapy

The 2000 National Institutes of Health Consensus Conference has recognized the benefit of postmastectomy RT in patients with breast cancer who have 4 or more positive lymph nodes, but participants could not make final recommendations for patients with 1-3 positive lymph nodes. Some benefit is probably derived in patients with 1-3 positive lymph nodes; however, because the absolute risk of failure is small, the demonstration of small absolute benefit requires a large clinical trial.

To lift the uncertainty surrounding this issue, the Southwest Oncology Group S9927 trial proposes to assess postmastectomy RT in women with stage II breast cancer with 1-3 positive lymph nodes. The impact of irradiating internal mammary and medial supraclavicular lymph nodes on survival, disease-free survival, metastasis-free survival, and cause of death in women with resected stage I/II/III breast cancer is being evaluated by European Organization for Research and Treatment of Cancer trial 10925. Another European study will compare complete ALND to axillary irradiation in SLN-positive women with operable invasive breast cancer (European Organization for Research and Treatment of Cancer 10981).

Therapy duration

Optimal duration of different breast cancer therapies is still the subject of discussion. The currently accepted theory is that in the adjuvant setting, 6 months of chemotherapy is equivalent to longer durations. However, controversy remains as to whether 4 months of chemotherapy (AC for 4 cycles) is equivalent to 6 months and, in the event that a longer regimen is used, whether a non–cross-resistant regimen should be implemented.

Adjuvant tamoxifen therapy for 5 years is superior to shorter durations, but whether longer durations are beneficial or detrimental remains to be determined. With the advent of aromatase inhibitors to the adjuvant arena, the question of their optimal duration will fuel the debate for the coming years.

Because of the absence of serious adverse effects, many oncologists tend to use trastuzumab in patients with metastatic breast cancer until progression. To date, whether the continuation of trastuzumab with a different cytotoxic agent is of any benefit to patients whose disease progresses while on combination therapy with trastuzumab and another cytotoxic agent remains unknown. The optimal duration of trastuzumab in the adjuvant setting has not yet been defined and will depend on knowledge gained regarding its mechanism of action. If trastuzumab is mainly a cytostatic drug, its prolonged use is justified; however, if it is found to be cytotoxic, a limited duration may be reasonable.

HER/neu as a predictor or prognostic factor

The role of tissue HER2/neu as a prognostic and predictive factor has not been resolved. Early reports on the prognostic significance of tissue HER2/neu levels based on human and animal research have suggested that HER2/neu overexpression is a poor prognostic factor, although later clinical studies could not confirm these results. Similarly, the role of HER2/neu as a predictive indicator of response to chemotherapy or HT was tempered by the retrospective nature of these studies.

A part of this discrepancy is also related to the lack of a validated measurement assay. Several large retrospective studies analyzing serum banks from HT trials have suggested that elevated serum ECD-HER2 levels are correlated with a poor prognosis and resistance to HT.

Other smaller retrospective studies found similar results regarding serum ECD-HER2, ie, a negative prognostic but not predictive role. Owing to its simplicity and the possibility to express the results quantitatively, serum ECD-HER2 measurements should be included in future prospective trials to determine their exact value.

Patient selection

Controversy remains regarding the test of choice to select patients who may benefit from trastuzumab. Because this is a targeted therapy, reliably identifying patients who carry the overexpressed target, HER2/neu, is important. Early trials have used IHC, which is a semiquantitative method with high rates of false-positive and false-negative results compared with the more precise method, ie, detection of gene amplification by FISH.

Retrospective analysis of the FISH status of patients with metastatic breast cancer treated with trastuzumab as first-, second-, or third-line therapy showed that FISH was by far superior to IHC in predicting response to this agent. Response to therapy in patients with FISH-negative tumors was close to zero, while all the patients who experienced a clinical response had FISH-positive tumors.

The problem with FISH testing is the need for expensive equipment and consequent limited availability in pathology laboratories. To mitigate these deficiencies, effort was directed toward quantitation of the shedding of HER2/neu in the serum. Although a good correlation exists between serum HER2 levels and HER2/neu expression in the tumor, this test, critics argue, does not accurately reflect the HER2/neu tumor status and cannot register single-cell expression, both required for clinical decision-making processes. The Breast Cancer International Research Group 006 trial is comparing peripheral levels of shed HER2 with FISH for predicting outcome, while the North Central Cancer Treatment Group N9831 trial is examining whether pretreatment levels of shed HER2 and HER1 and their autoantibodies are prognostic.