CHAPTER 12: Breast Disease

Breast disease in women encompasses a spectrum of benign and malignant disorders, which present most commonly as breast pain, nipple discharge, or palpable mass. The specific causes of these symptoms vary with patient age. Benign disorders predominate in young premenopausal women, whereas malignancy rates increase with advancing age. Evaluation of breast disorders usually requires the combination of a careful history, physical examination, imaging, and, when indicated, biopsy.

Ductal System

The glandular portion of the breast is composed of 12 to 15 independent ductal systems that each drain approximately 40 lobules (Fig. 12-1). Each lobule consists of 10 to 100 milk-producing acini that empty into small terminal ducts (Parks, 1959). Terminal ducts drain into larger collecting ducts that merge into even larger ducts, which exhibit a saccular dilation just below the nipple called a lactiferous sinus (Fig. 12-2).

In general, only six to eight openings are visible on the nipple surface. These drain the dominant ductal systems, which account for approximately 80 percent of the breast’s glandular volume (Going, 2004). Minor ducts either terminate just below the nipple surface or open on the areola near the base of the nipple. The areola itself contains numerous lubricating sebaceous glands, called Montgomery glands, which are often visible as punctate prominences.

In addition to epithelial structures, the breast is composed of varying proportions of collagenous stroma and fat. The distribution and abundance of these stromal components accounts for a breast’s consistency when palpated and for its imaging characteristics.

Lymphatic Drainage

Afferent lymphatic drainage of the breast is provided by dermal, subdermal, interlobar, and prepectoral systems (Fig. 12-3)(Grant, 1953). Each of these may be viewed as a lattice of valveless channels that interconnect with every other system and that ultimately drain into one or two axillary lymph nodes (the sentinel nodes). Because all of these systems are interconnected, the breast drains as a unit, and injection of colloidal dyes in any part of the breast at any level will result in accumulation of dye in the same one or two axillary sentinel lymph nodes. The axillary lymph nodes receive most of the lymphatic drainage of the breast and consequently are the nodes most frequently involved with breast cancer metastases (Hultborn, 1955). However, there are also alternate drainage pathways that do not appear to interconnect with other networks and that drain directly into internal mammary, supraclavicular, contralateral axillary, or abdominal lymph node basins.

During fetal development, the primordial breast arises from the basal layer of the epidermis. Before puberty, the breast is a rudimentary bud composed of a few branching ducts capped with alveolar buds, end buds, or small lobules (Osin, 1998). At puberty, usually between the ages of 10 and 13 years, ovarian estrogen and progesterone cooperate to direct organized communication between breast epithelial cells and mesenchymal cells, resulting in extensive branching of the ductal system and development of lobules (Ismail, 2003). Specific disorders of this development are discussed in Chapter 14. Final differentiation of the breast is mediated by progesterone and prolactin and is not completed until the first full-term pregnancy (Grimm, 2002; Ismail, 2003).

During the reproductive years, terminal ducts near the acini and the acini themselves are most sensitive to ovarian hormones and prolactin. Most forms of benign and malignant breast disease arise in these terminal duct-acinar structures. Breast epithelial cells proliferate during the luteal phase of the menstrual cycle when estrogen and progesterone levels are increased, and then undergo programmed cell death at the end of the luteal phase, when levels of these hormones decline (Anderson, 1982; Soderqvist, 1997). This effect is mediated by paracrine signaling induced by estrogen-receptor activation and is associated with an increase in the water content of the extracellular matrix (Stoeckelhuber, 2002). This is often recognized as breast fullness and tenderness in the week preceding menses.

At menopause, when ovarian estrogen production ceases, breast lobules involute, and the collagenous stroma is replaced by fat. Because estrogen receptor expression is negatively regulated by estrogen, there is an increase in estrogen receptor expression after menopause (Khan, 1997). Despite a decline in ovarian estrogen production, postmenopausal women continue to produce estrogen through the action of the enzyme aromatase, which converts adrenal androgens to estrogen (Bulun, 1994). Aromatase is found in fat, muscle, and breast tissue.

It is not possible to distinguish benign from malignant or cystic from solid breast masses by clinical examination. However, findings from clinical examination, interpreted in conjunction with imaging and pathology (the triple test), contribute significantly to management decisions (Hermansen, 1987).

Physical Examination

The breast is comma shaped, and the comma’s tail corresponds to the axillary tail of Spence. This extension can be large, especially during pregnancy and lactation, and is frequently mistaken for an axillary mass.

Clinical examination of the breast begins with inspection of the breast to determine whether there is dimpling, nipple retraction, or skin changes. This examination is described in Chapter 1. The presence and character of expressible nipple discharge is recorded. In addition, the location of a mass is specifically documented according to its clock position and then measured along the long axis using a ruler or caliper (Fig. 12-4). The distance from the center of the nipple to the center of the mass is specified. Since numerous health care providers are typically involved in the evaluation and management of the same breast mass, the most useful entry in the clinical record will define the location and size of the mass (e.g., right breast, 2-cm mass, 3:00, 4 cm from the nipple). Although clinical examination alone can never exclude malignancy, noting that a mass has benign features such as smoothness, roundness, and mobility will factor into the ultimate decision to excise or observe a lesion. Evaluation also includes careful examination of the axillae, infraclavicular fossa, and supraclavicular fossa to identify lymphadenopathy.

Diagnostic Imaging

Imaging of a suspected mass may begin with mammography that includes magnification, extra compression, or extra views beyond the usual medial lateral oblique and cranial caudal views that are typically used for screening. Unlike screening mammography, diagnostic mammography may be appropriate for women of any age. In addition, sonography is invaluable for determining whether a mass is cystic or solid and is a component of most diagnostic imaging algorithms. Certain features of solid masses, such as irregular margins, internal echoes, or a width-to-height ratio <1.7 may suggest malignancy (Stavros, 1995).

Diagnostic imaging results should be summarized according to the Breast Imaging Reporting and Data System (BI-RADS) classification (Table 12-1) (D’Orsi, 2013). Lesions that are graded BI-RADS 5 are highly suggestive of malignancy, and ≥95 percent of these are ultimately proven to be cancerous. Decreasing numerical grades are associated with diminishing probability of malignancy.

Breast Biopsy

Evaluation of a solid breast mass is completed by needle biopsy. These biopsies should be performed after an imaging test or a minimum of 2 weeks prior to an imaging test. This is because resulting tissue trauma can produce image artifacts that simulate malignancy (Sickles, 1983). Options include fine-needle aspiration (FNA) biopsy or core-needle biopsy. The trend in recent years favors core-needle biopsy (Tabbara, 2000). Although FNA takes less time to perform and is less expensive than core-needle biopsy, it is less likely to provide a specific diagnosis and has a higher insufficient sample rate (Shannon, 2001). FNA retrieves clusters of epithelial cells that may be interpreted as benign or malignant, but it cannot reliably differentiate between benign proliferative lesions and fibroepithelial neoplasms or between ductal carcinoma in situ and invasive cancer (Boerner, 1999; Ringberg, 2001).

In contrast, core-needle biopsy is performed using an automated device that takes one core at a time or is completed using a vacuum-assisted device that, once initially positioned, delivers multiple cores. Needle biopsy of solid masses should be done prior to excision, as the biopsy results contribute significantly to surgical planning (Cox, 1995).

Triple Test

The combination of clinical examination, imaging, and needle biopsy is called the triple test (Wai, 2013). When all three assessments suggest a benign lesion or all three suggest a breast cancer, the triple test is said to be concordant. A concordant benign triple test is >99-percent accurate, and breast lumps in this category can be followed by clinical examination alone at 6-month intervals. If any of the three assessments suggests malignancy, the lump should be excised regardless of results from the other two. It is always appropriate to offer excision of a fully evaluated breast lump, even after a benign concordant triple test result, as breast lumps can be a source of significant anxiety.

Cysts

Most breast cysts arise from apocrine metaplasia of lobular acini. They are generally lined by a single layer of epithelium that ranges from flattened to columnar. One autopsy series that included 725 women reported microcysts in 58 percent and cysts >1 cm in 21 percent (Davies, 1964). The incidence of breast cysts peaks between 40 and 50 years, and the lifetime incidence of palpable breast cysts is estimated to be 7 percent (Haagensen, 1986).

Breast cysts are diagnosed and classified by sonography. There are three types of cysts: simple, complicated, and complex (Berg, 2003). Simple cysts are sonolucent, have a smooth margin, and show enhanced through-transmission (Fig. 12-5). These lesions do not require special management or monitoring, but they may be aspirated if painful. Recurrent cysts can be reimaged and reaspirated, but recurrent symptomatic cysts are best managed by excision.

Complicated cysts show internal echoes during sonography and can sometimes be indistinguishable from solid masses. Internal echoes are usually caused by proteinaceous debris. Consideration is given to aspirating complicated cysts. The aspirated material may be submitted for culture, if it is purulent, or for cytology, if there are worrisome clinical or imaging features. If the sonographic abnormality does not resolve completely with aspiration, a core-needle biopsy is usually performed.

Complex cysts show septa or intracystic masses during sonographic evaluation. An intracystic mass usually represents a papilloma, but medullary carcinoma, papillary carcinoma, and some infiltrating ductal carcinomas can present as complex cysts. Although some advocate core-needle biopsy for the evaluation of complex cysts, this procedure can decompress a cyst, making it difficult to localize at the time of surgery. Additionally, papillary lesions diagnosed by needle biopsy will require excision. Thus, it seems reasonable to recommend excision of all complex cysts.

Fibroadenoma

This represents a focal developmental abnormality of a breast lobule and as such is not a true neoplasm. Histologically, fibroadenomas are composed of glandular and cystic epithelial structures surrounded by a cellular stroma. Fibroadenomas account for 7 to 13 percent of breast clinic visits and had a prevalence of 9 percent in one autopsy series (Dent, 1988; Franyz, 1951). They often present in adolescence, are recognized most frequently in premenopausal women, and usually spontaneously involute at menopause.

Fibroadenomas classified as benign concordant by the triple test can be safely followed without excision. Because some fibroadenomas may grow large, and because benign phyllodes tumors are often indistinguishable from fibroadenomas by imaging and needle biopsy, a fibroadenoma that is growing should be excised.

Phyllodes Tumors

These are true biphasic neoplasms characterized by epithelial-lined spaces surrounded by cellular stroma. Both the epithelial and stromal components can be monoclonal and clonally related (Karim, 2013). Phyllodes tumors are classified as benign, intermediate, or malignant, based on the degree of stromal cell atypia, number of mitoses, tumor margin characteristics, and abundance of stromal cells (Oberman, 1965). Phyllodes tumors account for less than 1 percent of breast neoplasms, and the median age at diagnosis is 40 years (Kim, 2013; Reinfuss, 1996).

Malignant phyllodes tumors can metastasize to distant organs, with lung being the primary site. Chest radiographs or chest computed-tomography (CT) scanning are appropriate staging tests for malignant cases. Phyllodes tumors rarely metastasize to lymph nodes, thus axillary staging is not required unless nodes clinically appear involved (Chaney, 2000).

Treatment consists of wide local excision with a minimum 1-cm margin. Mastectomy may be required to achieve this margin, as the median tumor size at presentation is 5 cm. Local recurrence rates for completely excised tumors range from 8 percent for benign lesions to 36 percent for malignant ones (Barth, 1999). Fibroproliferation in the surrounding breast tissue and necrosis are the strongest predictors of recurrence (Barrio, 2007). Postoperative adjuvant radiation therapy may be indicated for high-risk cases (Barth, 2009).

Fluid can be expressed from the nipple ducts of at least 40 percent of premenopausal women, 55 percent of parous women, and 74 percent of women who have lactated within 2 years (Wrensch, 1990). The fluid generally issues from more than one duct and may range from milky white to dark green or brown. Green coloration is related to the content of cholesterol diepoxides and does not suggest underlying infection or malignancy (Petrakis, 1988).

Multiduct discharges that are elicited only following manual expression are considered physiologic and do not require additional evaluation. However, spontaneous discharges merit evaluation (Fig. 12-6). Spontaneous milky nipple discharge, also called galactorrhea, results from various causes (Table 12-2) (Chap. 15). Of these, pregnancy is a frequent cause of new-onset spontaneous discharge, and a bloody multiduct discharge during pregnancy is not uncommon.

Pathologic nipple discharge is defined as a spontaneous single-duct discharge that is serous or bloody. The rate of underlying malignancy ranges from approximately 2 percent for young women with no associated findings on imaging or physical examination to 20 percent for older women with associated findings (Cabioglu, 2003; Lau, 2005). Most pathologic nipple discharges are caused by benign intraductal papillomas, which are simple milk duct polyps (Urban, 1978). They arise in the major milk ducts, generally within 2 cm of the nipple, and contain a velvety papillary epithelium on a central fibrovascular stalk.

Evaluation of a pathologic nipple discharge begins with breast examination. Careful evaluation can frequently locate a trigger point on the areolar edge that elicits the discharge when pressed. Occult-blood testing and microscopic examination of the discharge can provide additional information. A glass slide that has been touched to the discharge and immediately fixed in 95-percent alcohol may be used for cytologic assessment. Nipple fluid samples are acellular in 25 percent of cases and thus cannot exclude an underlying malignancy (Papanicolaou, 1958). However, malignant cells, if found, are highly correlated with an underlying cancer (Gupta, 2004).

Following these examinations, diagnostic mammography and an assessment of the subareolar ducts by sonography or ductography is indicated. Diagnostic mammography is usually negative, but it may occasionally identify an underlying ductal carcinoma in situ (DCIS). Mammary ductography, also known as galactography, requires cannulating the affected duct, injecting radiocontrast, and then performing mammography (Fig. 12-7).

An evaluation of the subareolar ducts, as described above, is required to localize an intraductal lesion for subsequent excision. However, pathologic nipple discharge is definitively diagnosed and treated by subareolar duct excision, which is also known as microductectomy (Locker, 1988). Subareolar duct excision can also be used to treat bothersome multiduct discharges not associated with pituitary prolactinoma.

Puerperal Infections

Breast infections are generally divided into puerperal, which develop during pregnancy and lactation, and nonpuerperal. Of these, pregnancy-related breast infection is characterized by warm, tender, diffuse breast erythema, associated with systemic signs of infection such as fever, malaise, myalgias, and leukocytosis. The most common organism is staphylococcus, and it is successfully treated with oral or intravenous antibiotics, depending on the severity. However, infection may also progress to form deep parenchymal abscesses that require surgical drainage (Branch-Elliman, 2012). Sonographic examination is highly sensitive for identifying underlying abscesses if mastitis does not improve rapidly with antibiotics or if an abscess is suggested clinically. Women with puerperal mastitis should continue to breast feed or breast pump during treatment to prevent milk stasis, which may contribute to infection progression (Thomsen, 1983). Cracked or excoriated nipples may provide entry for bacteria and are treated with lanolin-based lotions or ointments.

Appropriate antibiotics for puerperal mastitis include those covering staphylococcal species. Group A and B Streptococcus, Corynebacterium, and Bacteroides species and Escherichia coli are less frequently isolated. Commonly, cephalexin (Keflex) or dicloxacillin (Dynapen), each given at dosages of 500 mg orally four times daily, or the combination of amoxicillin and clavulanate (Augmentin), 500 mg orally three times daily, may be prescribed for 7 days. Erythromycin, 500 mg orally four times daily, will provide adequate coverage for those with a penicillin allergy. Methicillin-resistant Staphylococcus aureus (MRSA) is becoming a more prevalent community-acquired pathogen causing mastitis in pregnancy and the puerperium (Laibl, 2005; Stafford, 2008). If MRSA is suspected or if a patient fails to improve on an initial regimen, then trimethoprim-sulfamethoxazole double strength (Bactrim DS, Septra DS), one or two tablets orally twice daily, or clindamycin, 300 mg orally three times daily, is a suitable choice. In ill patients with extensive infection, hospitalization and intravenous (IV) antibiotics are typically required. In these complicated cases, MRSA coverage may be prudent, and clindamycin, 600 mg IV every 8 hours, or vancomycin, 1 g IV every 12 hours, can be administered. Intravenous antibiotics are typically given until the woman is afebrile for 24 to 48 hours. Oral antibiotics are then continued to complete a 7- to 10-day course.

Focal mastitis may result from an infected galactocele. A tender mass will usually be palpable at the site of skin erythema. Needle aspiration of the galactocele and antibiotics are frequently all that is required, but recurrence or progression may mandate surgical drainage.

Nonpuerperal Infections

Uncomplicated cellulitis in a nonirradiated breast and in a nonpuerperal setting is uncommon. Accordingly, its presence prompts imaging and biopsies to exclude inflammatory breast cancer, described on page 291.

Nonpuerperal breast abscesses are generally classified as peripheral or subareolar. Peripheral abscesses usually are skin infections such as folliculitis or infection of epidermal inclusion cysts or Montgomery glands. These abscesses are all adequately treated by drainage and antibiotics discussed in the previous section. In contrast, subareolar abscesses arise from keratin-plugged milk ducts directly behind the nipple. The abscess itself usually presents under the areola, and fistulous communications between multiple abscesses are common (Kasales, 2014). Simple drainage is associated with a recurrence rate of nearly 40 percent, thus effective treatment requires subareolar duct excision and complete removal of sinus tracts. In general, surgical drainage of nonpuerperal breast abscesses is usually always accompanied by biopsy of the abscess wall, as breast cancer occasionally presents as an abscess (Benson, 1989; Watt-Boolsen, 1987).

Idiopathic granulomatous mastitis (IGM) is not a true infection. It is included in this section because the painful masses, fluid collections, skin erythema, ulceration, and draining sinus tracts are often confused with infection. Core biopsy will show noncaseating granulomas, and fluid aspirated from apparent “abscesses” is nearly always sterile. Tissue stains can be used to exclude tuberculosis or mycotic infection. Wegener granulomatosis and sarcoidosis are considered in the initial differential diagnosis. This is a self-limiting condition that may take years to resolve. Procedures should be minimized as they will often result in painful draining sinuses. High-dose corticosteroids or methotrexate have been used for treatment, but it is not clear whether they are effective (Mohammed, 2013; Pandey, 2014).

The prevalence of breast pain is 66 percent and is higher for women nearing menopause than for younger women (Euhus, 1997; Maddox, 1989). The precise etiology of mastalgia is unknown, but it is likely related to estrogen- and progesterone-mediated changes in interstitial water content and therefore in interstitial pressure.

Mastalgia is generally classified as cyclic or noncyclic. Noncyclic mastalgia is often focal and shows no relationship to the menstrual cycle. Although focal mastalgia is frequently caused by a simple cyst, breast cancer occasionally presents as focal breast pain. Therefore, this complaint is evaluated by careful clinical examination, targeted imaging, and needle biopsy of any palpable or imaging abnormalities.

In contrast, cyclic mastalgia is usually bilateral, diffuse, and most severe during the late luteal phase of the menstrual cycle (Gateley, 1990). It remits with the onset of menstruation. Cyclic mastalgia requires no specific evaluation and is generally managed symptomatically with nonsteroidal antiinflammatory agents (Fig. 12-8). Various other treatments have been proposed including bromocriptine, vitamin E, or oil of evening primrose. However, outcomes are no better than placebo in the best randomized clinical trials, except for bromocriptine in the subset of women with elevated prolactin levels (Kumar, 1989; Mansel, 1990). For the most severe cases, several agents are effective when administered during the last 2 weeks of the menstrual cycle. These include: (1) danazol, 200 mg orally daily; (2) the selective estrogen-receptor modifier toremifene (Fareston), 20 mg orally daily; or (3) tamoxifen (Nolvadex), 20 mg orally daily. Pregnancy must first be excluded and then avoided if these medications are used.

Benign Breast Disease without Atypia

The primary tissue components of the breast are fat, fibrous stroma, and epithelial structures. The hormonally responsive component is the epithelium, but considerable paracrine communication exists between the epithelium and stroma. The natural hormonal changes of puberty, pregnancy, lactation, and menopause drive considerable physiologic remodeling of breast tissue during a woman’s lifetime, but pathologic remodeling is observed in some. This is initially characterized by acinar dilation and fibrosis, termed nonproliferative benign breast disease. Depending on the extent and pattern of these changes, a breast may appear mammographically dense, feel nodular to palpation, or both. The term “fibrocystic change” is often used to refer to palpably nodular breast tissue or to the histologic pattern of dilated ducts and acini invested with dense collagenous stroma. This is not a significant breast cancer risk factor and does not require any special management.

When this change is accompanied by accumulation of luminal epithelial cells (e.g., epithelial hyperplasia), it is called benign proliferative disease (Fig. 12-9). This change has been linked to higher levels of estrogen, insulin, and certain inflammatory cytokines, as well as reduced levels of the beneficial adipokine adiponectin (Catsburg, 2014). Benign proliferative breast disease without atypia is a modest breast cancer risk factor with a relative risk of 1.5 to 1.9 (Dupont, 1993; Hartmann, 2005; Sneige, 2002).

Benign Proliferative Disease with Atypia

Atypia refers to specific alterations in the size, shape, or nuclear features of individual epithelial cells in combination with the way groups of cells are organized. Atypical proliferation of ductal cells is termed atypical ductal hyperplasia (ADH), whereas similar changes in acinar cells are termed atypical lobular hyperplasia (ALH). As more and more terminal ducts or acini become involved, the condition is recognized as ductal carcinoma in situ or lobular carcinoma in situ, respectively, which are discussed in later sections (Ringberg, 2001).

Benign proliferative disease with atypia historically accounts for 4 percent of benign breast lesions (Hartmann, 2005). However, the incidence has recently decreased coincident to reductions in hormone replacement therapy use (Menes, 2009).

It is important to recognize that the difference between ADH and low-grade ductal carcinoma in situ is based on the area occupied by the proliferative epithelial cells (Vandenbussche, 2013). Accordingly, surgical excision is usually recommended when ADH is diagnosed by core biopsy, as 4 to 38 percent of cases will be upgraded to in situ or invasive cancer.

Chemoprevention is an excellent option for high-risk women with atypical hyperplasia. These lesions are estrogen-driven, and tamoxifen has been shown to reduce breast cancer risk by 52 to 86 percent for these women (Coopey, 2012; Fisher, 1999).

Benign proliferative disease with atypia is a marker of increased breast cancer risk. Relative risks are 4.5 to 5.0, and absolute risks approximate 1 percent per year for 20 to 30 years (Degnim, 2007; Dupont, 1993). This risk is higher for more extensive lesions. Risk does not appear to increase further with hormone replacement use.

Similar to proliferative ductal lesion, lobular carcinoma in situ (LCIS) differs from ALH by the greater extent of lobular cell proliferation with LCIS and increased acini distention. LCIS is not associated with any specific mammographic or palpable features and thus is only diagnosed incidentally. Classic LCIS has not traditionally been viewed as a direct precursor of breast cancer, but this view is changing. For example, although LCIS is associated with increased risk for both breasts, women with LCIS most commonly develop carcinoma in the ipsilateral breast (Fisher, 2004b; Ottesen, 1993; Salvadori, 1991). Moreover, infiltrating lobular cancers frequently show associated LCIS, and a clonal relationship between LCIS and subsequent invasive cancer has been demonstrated (Abner, 2000; Andrade, 2012; Sasson, 2001).

As noted, LCIS is also a marker of increased breast cancer risk. The risk of subsequent breast cancer approximates 1 percent per year but is modified upward by early age at diagnosis, family history of breast cancer, and extensive disease (Bodian, 1996).

Surgical excision is recommended if LCIS is diagnosed by needle biopsy, as an associated cancer will be identified in 2 to 25 percent of cases (Buckley, 2014). Excising to clear margins is not required (Sadek, 2013). These lesions are strongly estrogen receptor positive, and tamoxifen has been shown to reduce breast cancer risk by 56 percent in this setting (Fisher, 1999).

DCIS can be understood as a condition in which cancer cells fill portions of a mammary ductal system without invading beyond the duct’s basement membrane (Ringberg, 2001). Although DCIS cells have accumulated many of the DNA changes common to invasive breast cancer, they lack certain critical changes that would permit them to persist outside of the duct (Aubele, 2002). Ductal carcinoma in situ is classified as stage 0 breast cancer.

The U.S. incidence of DCIS has increased in parallel with that of invasive breast cancer during the past two decades. But, as with invasive breast cancer, the incidence has plateaued during the past several years (Virnig, 2010). DCIS currently accounts for 25 to 30 percent of all breast cancers in the United States. It is most commonly diagnosed by screening mammography as it is frequently associated with pleomorphic, linear, or branching calcifications (Fig. 12-10).

DCIS is classified by morphologic type, the presence or absence of comedonecrosis, and nuclear grade. The common morphologic types include cribriform, solid, micropapillary, and comedo (Fig. 12-11). Comedonecrosis appears as a necrotic eosinophilic core down the center of a duct packed with cancer cells. Of all of the classifying variables, nuclear grade is the most predictive for associated invasive cancer, extent of disease, and recurrence after treatment (Ringberg, 2001).

Incompletely treated DCIS may recur locally, and 50 percent of recurrences are associated with fully developed invasive breast cancer. The principal treatment of DCIS is wide excision with a negative margin. This may require mastectomy if DCIS is extensive or if there are other contraindications to breast conservation. When breast conservation is possible, postoperative breast irradiation will reduce the local recurrence rate from 18 percent to 9 percent and is considered standard adjuvant treatment (Fisher, 1993). For those treated with breast conservation and radiation, the breast cancer-specific survival rate is 96 percent (Fig. 12-12) (Solin, 1996).

Axillary staging is generally not included in the management of DCIS, although some have advocated sentinel node biopsy for large, high-grade DCIS diagnosed by needle biopsy and treated by lumpectomy, as occult invasive cancer is diagnosed in 10 percent (Wilkie, 2005). Sentinel lymph node (SLN) biopsy in conjunction with mastectomy is less controversial, as it is not possible to go back and perform SLN biopsy if an occult invasive cancer is diagnosed in this setting.

Five years of tamoxifen is recommended for estrogen-receptor-positive DCIS treated by breast conservation (Fisher, 1999). Although tamoxifen is not associated with a statistically significant improvement in overall survival rates, it does significantly reduce the incidence of ipsilateral invasive cancer and also reduces the risk of contralateral breast cancer.

Paget Disease of the Nipple

This type of DCIS presents as a focal eczematous rash of the nipple (Fig. 12-13). Ductal carcinoma cells, responding to chemoattractants secreted by cells in the dermis, migrate to the surface of the nipple, inducing skin breakdown (Schelfhout, 2000). The condition is easily diagnosed histologically by punch biopsy or excision of the affected nipple tip after nipple-areolar blockade using local anesthetic. Evaluation also includes careful clinical examination, as an associated mass is identified in approximately 60 percent of cases (Ashikari, 1970). Among those with no palpable abnormalities, mammography will show suspicious densities or calcifications in 21 percent (Ikeda, 1993). An underlying DCIS is identified in about two thirds of cases, and an invasive cancer in approximately one third (Ashikari, 1970).

Treatment includes wide excision with negative margins. Breast conservation, which requires central breast resection including the nipple-areolar complex and all identifiable underlying disease, is followed by postoperative breast irradiation (Bijker, 2001). Axillary staging by sentinel node biopsy is not required unless an invasive component is identified or total mastectomy is performed.

The most profound breast cancer risk factor is female gender. In addition, the incidence of breast cancer, as for most other cancers, increases with advancing age. Only 12 to 30 percent of breast cancer has a significant familial component, and apart from radiation exposure early in life, convincing environmental causes have not been elucidated (Baker, 2005; Locatelli, 2004). Breast cancer risk factors are listed in Table 12-3, but the strong link between ovarian function and breast tissue remodeling warrants more detailed description. In general, all of these risk factors are more prevalent in developed countries than in those that are less developed. Consequently, breast cancer is more common in industrialized cultures (Parkin, 2001).

Reproductive Factors

Ovulatory Cycles

Ovulatory menstrual cycles exert stress on the breast epithelium by inducing proliferation in the late luteal phase. If conception does not occur, proliferation is followed by programmed cell death (Anderson, 1982; Soderqvist, 1997). Early age at menarche is associated with earlier onset of ovulatory cycles and increased breast cancer risk (den Tonkelaar, 1996; Vihko, 1986). Conversely, early menopause, whether it is natural or surgical, is associated with a reduced breast cancer risk (Kvale, 1988). Pregnancy generates very high levels of circulating estradiol, which is associated with a transient increase in short-term risk. But pregnancy also induces terminal differentiation of breast epithelium and provides relief from ovarian cycling. Consequently, increasing parity is associated with reduced lifetime risk.

Pregnancy

The breast is unique among all human organs in that it exists as a primordium for a decade or more before entering a highly proliferative state at menarche, and then does not fully mature until the first live birth. Immature breast epithelium is more susceptible to carcinogens than postlactational epithelium (Russo, 1996). Therefore, the longer a first live birth is delayed, the greater the breast cancer risks. Relative to nulliparity, a first live birth before the age of 28 years is associated with reduced breast cancer risk, whereas one occurring later is associated with increased risk (Gail, 1989). Both early age at first live birth and greater numbers of live births are associated with reduced breast cancer risk (Layde, 1989; Pike, 1983).

Hormone Replacement Therapy

The postmenopausal use of combined estrogen and progestin hormone replacement therapy is a modest breast cancer risk factor, and relative risks range from 1.26 to 1.76 (Beral, 2011; Hulley, 2002; Rossouw, 2002). The risk is higher with longer durations of use and with a shorter interval between the onset of menopause and the start of the medication (Beral, 2011). Estrogen-only replacement is not convincingly associated with increased breast cancer risk, but there is a relationship with body mass index that yields a lower risk for obese women and higher risk for thin women (Anderson, 2004).

Approaches for managing breast cancer risk include: (1) lifestyle modification to achieve and sustain ideal body weight, (2) enhanced surveillance that includes screening magnetic resonance (MR) imaging, (3) chemoprevention with a selective estrogen-receptor modifier or an aromatase inhibitor, and (4) prophylactic surgery including oophorectomy or mastectomy for those at highest risk (Cuzick, 2014; Domchek, 2010; Goss, 2011; Heemskerk-Gerritsen, 2007; Vogel, 2010). Beyond beneficial lifestyle modification, each intervention introduces new risks, and thus breast cancer risk quantification is essential for making prevention decisions.

The American Cancer Society has endorsed screening MR imaging for women with a lifetime breast cancer risk that exceeds 20 percent (Saslow, 2010). The Food and Drug Administration (FDA) has approved tamoxifen chemoprevention for women older than 35 years with >1.7 percent breast cancer risk over 5 years. Similarly, raloxifene (Evista) is approved for increased-risk postmenopausal women. To maximize benefit and minimize harm, breast cancer risk, age, race, and prior hysterectomy all factor into chemoprevention decisions (Freedman, 2011).

The foregoing highlights the importance of quantitative breast cancer risk stratification. Several computer models are available for this. The most thoroughly validated is the Gail model, available at http://www.cancer.gov/bcrisktool/ (Costantino, 1999; Gail, 1989; Rockhill, 2001). Although the most generally applicable model, the Gail model is insufficient when there is a strong family history of breast cancer, male breast cancer, or ovarian cancer (Euhus, 2002). Genetic models such as BRCAPRO, Tyrer-Cuzick, or BOADICEA are more appropriate in these settings (Berry, 1997; Lee, 2014; Tyrer, 2004).

Breast Cancer Genetics

Twin studies suggest that only 12 to 30 percent of breast cancer is primarily genetic in origin, and modeling studies implicate autosomal dominant inheritance of single genes as the most important mechanism (Lichtenstein, 2000; Locatelli, 2004; Risch, 2001). As such, genetic testing is one of the most powerful risk stratification tools available. It can identify women at very high risk for cancer who could reasonably consider risk-reducing surgery. In breast cancer patients, it can also contribute directly to decisions regarding surgery, radiation, and systemic therapies (Euhus, 2013). Mutations in BRCA1 or BRCA2 genes are the most frequently identified germline alterations in familial breast cancer, but the list of predisposition genes is growing (Table 12-4). Commercialized massive parallel sequencing, namely, next-generation sequencing, now allows testing for mutations in a few to dozens of genes simultaneously (Euhus, 2015).

Obtaining a reasonably detailed cancer family history is essential for identifying individuals who may benefit from genetic counseling and testing. At a minimum, the relationship and age at diagnosis is recorded for every cancer in the family. Family histories that may suggest inherited susceptibility include early-onset breast cancer (<50 years), bilateral breast cancer, male breast cancer, multiple affected relatives in one generation, breast cancer in multiple generations, development of cancers that are known to be associated with a particular syndrome, and two or more cancers in one relative, especially if they develop at an early age.

Hereditary Breast-Ovarian Cancer Syndrome

This syndrome accounts for 5 to 7 percent of breast cancers in the United States and is most frequently caused by BRCA1 or BRCA2 mutation (Malone, 2000). Hallmarks of the form linked with BRCA1 include early age at breast cancer diagnosis (median 44 years); high-grade, estrogen- and progesterone-receptor negative breast cancers; and associated ovarian cancer (Foulkes, 2004). Other associated cancers are pancreatic cancer and melanoma.

For BRCA1 mutation carriers, the lifetime breast cancer risk ranges from 45 to 81 percent, and ovarian cancer risk from 16 to 54 percent (Antoniou, 2008; Brohet, 2014; Ford, 1998; King, 2003; Mavaddat, 2013). Individuals who have developed both breast and ovarian cancer have an 86-percent probability of carrying a BRCA mutation (Cvelbar, 2005).

Among BRCA2 carriers, lifetime risk for breast cancer ranges from 27 to 85 percent, and ovarian cancer risk from 6 to 27 percent. Women with BRCA2 mutations develop breast cancer later in life than BRCA1 carriers, thus age at diagnosis is not usually a good criterion for recognizing this syndrome (Panchal, 2010). Similar to sporadic breast cancer, most BRCA2-associated breast cancers are hormone receptor positive (Lakhani, 2002). Ovarian cancer is an associated cancer but develops less frequently than it does in BRCA1-affected families. Five to 13 percent of male breast cancers are associated with BRCA2 mutations. Lifetime breast cancer risk is estimated at 1.8 percent for men with BRCA1 mutations and 8.3 percent for BRCA2 (Tai, 2007).

For affected women, early premenopausal bilateral oophorectomy reduces breast cancer risk by 37 to 72 percent and also lowers breast cancer-specific and all-cause mortality rates (Domchek, 2010; Finch, 2014; Kauff, 2008). This is discussed further in Chapter 35. Bilateral prophylactic mastectomy reduces breast cancer risk by more than 90 percent but has not yet been shown to improve survival rates (Hartmann, 2001; Heemskerk-Gerritsen, 2007; Meijers-Heijboer, 2001).

With the introduction of next-generation sequencing panel tests, clinicians are increasingly confronted with rare syndromes for which there are scarce data to guide management (see Table 12-4) (Euhus, 2015). Involvement of professional genetic counselors and careful assessment of a three-generation family cancer history is essential for estimating and managing cancer risk.

Surgical options for breast cancers that arise in the context of an inherited predisposition syndrome are the same as for sporadic breast cancers (Pierce, 2010). However, patients are counseled that the risk of an ipsilateral second primary breast cancer in a preserved breast can be as high as 3 to 4 percent annually (Haffty, 2002; Seynaevea, 2004). Moreover, lifetime contralateral breast cancer risk is 83 percent for BRCA1 mutation carriers and 62 percent for BRCA2 carriers, and growing evidence supports bilateral mastectomy to improve survival rates (Evans, 2013; Mavaddat, 2013; Metcalfe, 2014).

In the United States, digital mammography has largely replaced film-screen mammography, and 3-dimensional (3-D) tomosynthesis is gradually replacing 2-D mammography. This technique generates hundreds of images as the x-ray source arcs over the top of the breast. Digital reconstruction allows a radiologist to visually scroll through breast images and significantly attenuates overlying breast densities at each level (Kopans, 2013). Compared with 2-D mammography, tomosynthesis reduces the false-positive rate (recall rate) by 15 to 30 percent and increases the cancer detection rate by 10 to 29 percent (Greenberg, 2014; Haas, 2013; Skaane, 2013). This is achieved with slightly higher radiation doses (Feng, 2012).

The Screening Mammography Controversy

In 2009, the U.S. Preventive Services Task Force recom­mended biennial screening mammography for women aged 50 to 74 years and individualized screening decisions for women aged 40 to 49. Several influential organizations including the American College of Obstetricians and Gynecologists (2014) and the American College of Radiology suggest that yearly screening mammography begin at age 40 (Lee, 2010). The American Cancer Society recommends yearly screening beginning at age 45, but with an opportunity to begin at age 40. They promote a transition to biennial screening at age 55, although yearly screening may be elected (Oeffinger, 2015). The controversy centers on: (1) the true mortality rate benefit, (2) the harm from false-positive results, and (3) the harm from diagnosing clinically irrelevant breast cancers.

However, most data available for addressing these issues are derived from eight large, but older, randomized prospective trials. The most recent trial was completed in the 1980s. Recent technological advances have significantly improved the sensitivity of mammography, but breast cancer treatment has also advanced, reducing the mortality rate improvement from early detection. Based on 30-year-old data, it is generally agreed that screening mammography starting at age 50 reduces breast cancer mortality rates by approximately 27 percent, and one metaanalysis reported an 18-percent reduction for women aged 40 to 49 (Hendrick, 1997; Kerlikowske, 1997). However, screen-detected breast cancer is a heterogeneous disease. Some cancers will eventually develop clinical metastases no matter how small they are when first detected, and some will never become lethal no matter how long diagnosis is delayed. This latter form is the one most likely to be detected by periodic screening (length time bias).

The practice of screening mammography is based on the assumption that early intervention in some subgroup of tumors will interrupt progression and save lives. Since the introduction of screening mammography more than three decades ago, there has been a large increase in the detection of early-stage breast cancer but only a small decrease in the diagnosis of node-positive or metastatic disease (Bleyer, 2012). This suggests that many breast cancers will never progress (overdiagnosed) and that the fraction of breast cancers whose progression can be interrupted by surgery may be modest.

For now, annual mammography beginning at age 40 as recommended by several professional societies is reasonable, but women are counseled of the risks and benefits. Among 1000 U.S. women aged 50 years who are screened annually for a decade, 0.3 to 3.2 are estimated to avoid a breast cancer death, 490 to 670 will have at least 1 false alarm, and 3 to 14 will be overdiagnosed and treated needlessly (Welch, 2014). There is no arbitrary age above which screening should cease. Women should have at least 10 years of remaining life to realize a mortality benefit from screening mammography (Lee, 2013).

Breast Magnetic Resonance Imaging

Breast MR imaging is commonly used as an adjunct to screen high-risk women and to establish disease extent in certain breast cancer patients. It is more sensitive for breast cancer detection than is mammography, but it is expensive and has a high false-positive rate. In addition, some evidence links its use with increased mastectomy rates but without reducing reexcision rates or improving breast cancer outcome (Houssami, 2013, 2014; Pilewskie, 2014; Turnbull, 2010).

Annual screening breast MR imaging is frequently selected for genetically high-risk women and in women with a lifetime breast cancer risk exceeding 20 percent (Saslow, 2010). This increases the diagnosis of smaller, lymph node negative breast cancers but does not improve survival rates (Gareth, 2014; Moller, 2013).

Breast MR imaging is not routinely performed in women with newly diagnosed breast cancer. Its primary value is assessing response to neoadjuvant chemotherapy in women contemplating breast conservation and evaluating women with breast cancer metastatic to axillary lymph nodes from an unknown primary (Morrow, 2011). Additionally, it can aid establishing the extent of disease prior to breast conservation for a subset of patients in whom uncertainty persists after careful clinical examination, mammography, and sonography.

Other Breast Imaging Modalities

Adding almost any imaging modality to screening mammography will incrementally increase the cancer detection rate but at the cost of an increased false-positive rate and more biopsies. Modalities that are occasionally useful, but not recommended for routine use, include screening sonography, breast-specific gamma imaging, and breast positron emission tomography (PET) (Kalinyak, 2014; Merry, 2014; Rechtman, 2014). These latter two tests are associated with significantly higher radiation exposure. Evidence is accumulating that medical radiation exposure before age 30 can increase breast cancer risk, and thus caution is advised (Berrington de Gonzalez, 2009; ch012-bib018).

Screening Physical Examination

The value of a screening clinical breast examination (CBE) performed by health care providers should not be neglected (Jatoi, 2003). Four of the large, randomized mammography trials mentioned earlier collected information on CBE and found that 44 to 74 percent of the breast cancers were detected by this approach. Sensitivity and specificity were higher for CBE than mammography among young women.

In contrast, enthusiasm for patients to perform breast self-examination (BSE) has diminished after a very large randomized trial from Shanghai, China, found no improvement in mortality rates (Thomas, 2002). Although there is less interest in promoting regimented BSE, encouraging women to remain breast-aware is reasonable.

In the United States, breast cancer is the most common cancer in women and the second most frequent cause of cancer-related mortality (second to lung) (Siegel, 2014). Although the incidence of breast cancer increased steadily in this country through the 1980s and 1990s, it has leveled at approximately 125 cases per year per 100,000 postmenopausal women and is declining for some ethnicities. Concurrently, survival rates steadily improve (Fig. 12-14) (Howlader, 2013).

Tumor Characteristics

Primary cancers of the breast comprise 97 percent of malignancies affecting the breast, whereas 3 percent represent metastases from other sites. The most common of these, in descending order, are the contralateral breast, sarcoma, melanoma, serous epithelial ovarian cancer, and lung cancer (DeLair, 2013). Cancers of mammary epithelial structures account for most of primary breast cancer. Infiltrating ductal carcinoma is the most common form of invasive breast cancer (∼80 percent), and infiltrating lobular carcinoma is the second most frequent (∼15 percent). Other malignancies such as phyllodes tumors, sarcoma, and lymphoma form the remainder.

Apart from stage, the primary tumor characteristics that most influence prognosis and treatment decisions are hormone receptor status, nuclear grade, and Her-2/neu expression (Harris, 2007). Approximately two thirds of breast cancers are estrogen- and progesterone-receptor positive. This feature is generally associated with a better prognosis and more treatment options.

Her-2/neu is a membrane tyrosine kinase that cooperates with other Her-family receptors to generate proliferation and survival signals in breast cancer cells. Approximately 25 percent of breast cancers have increased expression of Her-2/neu (Masood, 2005). The list of medications that specifically target HER2 overexpressing breast cancer is growing and includes trastuzumab (Herceptin), trastuzumab emtansine (Kadcyla), pertuzumab (Perjeta), neratinib, and lapatinib (Tykerb)(Tolaney, 2014).

Gene expression profiling has identified several “intrinsic subtypes” of breast cancer with prognostic significance (Cadoo, 2013). Multigene assays are now available in the clinic for individualized prediction of prognosis and treatment response, especially for estrogen-receptor positive tumors (Rouzier, 2013).

Breast Cancer Staging

Careful breast cancer staging is essential for predicting outcome, planning treatment, and comparing treatment effects in clinical trials. Each patient is assigned both a clinical and a pathologic stage. The clinical stage is based on examination and radiographic findings, whereas the pathologic stage is based on actual tumor measurements and histologic assessments of lymph nodes after primary surgery. Surgical staging of breast cancer is based on the TNM system, which includes primary tumor size (T), regional lymph node involvement (N), and presence of distant metastases (M) (Table 12-5). For patients with a clinically negative axilla, sentinel lymph node biopsy has replaced complete axillary dissection for nodal staging (Lyman, 2014).

The most common distant metastatic site in breast cancer is bone, followed by lung, liver, and brain. Thus, for newly diagnosed breast cancer patients, a complete blood count and liver function tests including alkaline phosphatase are recommended. Whole body screening with CT of the chest, CT or MR imaging of the abdomen and pelvis, and bone scan or whole body PET/CT are only recommended for clinical suspicion of metastases or for patients with clinical stage III disease (National Comprehensive Cancer Network, 2014).

Breast Cancer Treatment

Breast cancer is best managed by a multidisciplinary team of breast surgeons, medical oncologists, and radiation oncologists. Goals of surgery and radiation therapy are elimination of all local or regional tumor in a way that maximizes cosmesis and minimizes the risk of local or regional recurrence. There is some evidence that these local modalities reduce the risk of subsequent metastases and therefore increase survival rates (Darby, 2011). However, a significant proportion of patients with apparently localized disease have tumor cells detectable in their blood or bone marrow at diagnosis (Braun, 2005; Giuliano, 2011a). For these women, systemic treatment with chemotherapy, hormone manipulation, or targeted therapies is the primary approach for reducing metastasis risk and death (Dowsett, 2010; Peto, 2012).

Surgery

Halstead (1894) revolutionized the treatment of breast cancer by demonstrating improved outcome for patients treated with radical mastectomy. However, results from recent randomized clinical trials have appropriately fostered a trend toward less aggressive surgery. Specifically, lumpectomy with postoperative radiation therapy results in the same breast cancer-specific survival rate as total mastectomy (Fisher, 2002).

Axillary dissection, that is, near-complete axillary lymphadenectomy, was also once a standard part of breast cancer staging and treatment, but its role is diminishing (Rao, 2013). The procedure is still indicated for patients with clinically node positive disease at diagnosis. However, it is frequently omitted in selected patients with clinically negative nodes but positive sentinel nodes, because radiation therapy and systemic adjuvant therapies achieve the same low axillary recurrence rate and the same survival rate (Galimberti, 2013; Giuliano, 2010, 2011b). When required, axillary dissection results in lymphedema in 15 to 50 percent of women, depending on how it is measured (Morrell, 2005). Dissection is also associated with persistent shoulder or arm symptoms in up to 70 percent (Kuehn, 2000).

Radiation Therapy

After breast-conserving surgery, whole breast radiation reduces local recurrence from approximately 5 percent per year to 1 percent per year, although it may be omitted in elderly patients with favorable tumors (Fisher, 2002; Hughes, 2013). Shorter courses of partial breast irradiation may be appropriate in selected patients (Smith, 2009). Postmastectomy chest wall radiation improves survival in women with high-risk lymph node positive breast cancer (Overgaard, 1999; Ragaz, 2005). Recent clinical trial data are driving a marked increase use of extended-field radiation therapy (Early Breast Cancer Trialists’ Collaborative Group, 2014).

Chemotherapy

In the past, adjuvant chemotherapy was reserved for patients with nodal metastases and was always given after definitive surgery. However, randomized prospective trials have shown that adjuvant chemotherapy also improves survival rates for high-risk node-negative patients (Fisher, 2004a). Increasingly, however, the decision for chemotherapy is influenced by specific measures of tumor biology including results from multigene assays (Rouzier, 2013; Sparano, 2008).

If used, adjuvant chemotherapy is usually administered after primary surgery but before radiation therapy. Neoadjuvant chemotherapy is given prior to definitive surgery and is gaining popularity. Neoadjuvant chemotherapy permits assessment of a given tumor’s sensitivity to the selected agents, and tumor shrinkage permits less aggressive surgery (von Minckwitz, 2013).

Modern breast cancer chemotherapy often includes an anthracycline such as doxorubicin (Adriamycin), in conjunction with cyclophosphamide (Cytoxan) (Trudeau, 2005). The addition of a taxane has been shown to improve outcome (A’Hern, 2013). Platins such as cisplatin (Platinol) or carboplatin (Paraplatin) are increasingly used to replace doxorubicin when other cardiotoxic drugs such as trastuzumab are required and to treat certain tumor subtypes that have defects in homologous recombination. Chemotherapeutic agents are described more fully in Chapter 27.

Hormonal Therapy and Targeted Therapies

Adjuvant hormonal therapy is used for estrogen-receptor positive tumors. In pre- or postmenopausal women, one option is the selective estrogen-receptor modulator tamoxifen (Jaiyesimi, 1995). As discussed in Chapter 27, important side effects of tamoxifen include menopausal symptoms, increased risks of thromboembolic events, and higher rates of endometrial polyps and endometrial cancer. Although this cancer risk is increased, surveillance of the endometrium with routine transvaginal sonography or endometrial biopsy is not recommended. Endometrial evaluation is reserved for those with abnormal bleeding and follows that outlined in Chapter 8.

In postmenopausal women, aromatase inhibitors may be used. FDA-approved agents include anastrozole (Arimidex), letrozole (Femara), and exemestane (Aromasin) (Kudachadkar, 2005). In postmenopausal women, most circulating estradiol is derived from the peripheral conversion of androgens by the enzyme aromatase. Administration of aromatase inhibitors reduces circulating estradiol to nearly undetectable levels in these women. The addition of an aromatase inhibitor after tamoxifen is associated with a 23- to 39-percent improvement in the disease-free survival rate and a nearly 50-percent reduction in the contralateral breast cancer rate (Geisler, 2006). Although tamoxifen is commonly used as an initial antihormonal therapy in postmenopausal women, transition to an aromatase inhibitor and 10 years of treatment improves outcome (Johnston, 2014).

Unlike tamoxifen, aromatase inhibitors are associated with greater rates of bone loss and fractures. Accordingly, baseline bone mineral density testing and periodic monitoring is recommended. For women with mild or moderate bone loss, exercise and supplementation with vitamin D and calcium are encouraged. Various agents are available for managing severe loss, and a discussion of these drugs is found in Chapter 22.

Bisphosphonates such as zoledronic acid (Zometa) are often used to prevent cancer-treatment-induced bone loss (Hadji, 2011). In addition, the combination of aromatase inhibitors and zoledronic acid appears to improve outcome in hormone-receptor-positive breast cancer (Coleman, 2013).

Therapies that target specific biological pathways are becoming available. However, only HER2-targeted therapies are currently used routinely in early-stage breast cancer and only in HER2-amplified tumors. Targeting the mTOR pathway with everolimus (Afinitor) is now an FDA-approved strategy for advanced or metastatic hormone-receptor-positive breast cancer and is also being investigated for trastuzumab-resistant HER2-positive breast cancer (Andre, 2014; Dhillon, 2013). Comprehensive molecular profiling of tumors to identify targets for intervention is becoming more common (Frampton, 2013). Biologic agents for targeting cancer are described more fully in Chapter 27.

Surveillance

Long-term surveillance of breast cancer patients after treatment includes periodic history and physical examination. Women who elected breast conservation are counseled that the remaining breast tissue requires surveillance indefinitely. Ipsilateral, second primary breast cancers develop at a rate of approximately 1 percent per year and contralateral breast cancers at approximately 0.7 percent per year (Fatouros, 2005; Fisher, 1984; Gao, 2003). Laboratory and imaging tests are obtained to further evaluate specific signs or symptoms. Screening tests other than mammography to identify asymptomatic recurrences are not recommended (Khatcheressian, 2013).

Inflammatory Breast Cancer

Inflammatory breast cancer accounts for 1 to 5 percent of breast cancers (Chang, 1998; Dawood, 2010). This cancer presents with skin changes that can range from a faint red blush to a flaming-red rash associated with skin edema (peau d’orange change) (Fig. 12-15). It is distinguished from a neglected advanced primary breast cancer by its rapid onset and progression within just a few weeks. The cancer spreads rapidly throughout the entire breast and creates diffuse induration. As a result, the breast may enlarge to two to three times its original volume within weeks (Taylor, 1938).

Although mastitis or even congestive heart failure can produce a similar clinical appearance, inflammatory breast cancer must be definitively excluded. This always includes diagnostic mammography and punch biopsy of the skin. However, it also may require multiple biopsies and additional imaging such as MR imaging. Treatment begins with induction chemotherapy, followed by modified radical mastectomy (total mastectomy and axillary dissection), and then postoperative chest wall irradiation with or without additional chemotherapy (Cariati, 2005). The 5-year survival rate is 30 to 55 percent, which is significantly worse than for neglected advanced primary breast cancer (Brenner, 2002; Harris, 2003).