Breast Anatomy and Routes of Spread

The mammary gland is composed of glandular tissue, subcutaneous fat, and dense fibrous stroma containing an intricate network of lymphatics, nerves, and blood vessels. The gland is supported between the superficial fascia attached to the dermis and the deep fascia overlying the chest wall muscles interconnected by the Cooper ligament. The breast is mainly situated on the pectoralis muscle and extends craniocaudally between the second and sixth anterior ribs and mediolaterally from the sternum to the axillary midline with a portion of the breast reaching into the low axilla referred to as the tail of Spence. For staging purposes only, the chest wall includes the ribs, intercostal muscles, and serratus anterior but not the pectoralis muscles.

The glandular tissue of the breast is made up of between 4 and 18 milk ducts emanating, not always radially, from the nipple-areola complex. The ducts branch early on after leaving the nipple and form a highway of pathways, which terminate in ductal-lobular complexes. The lobules consist of specialized cells, which secrete milk products that travel down the ducts to the nipple predominantly during lactation. Most breast cancers originate from the interface of the ductal-lobular complex.

The lymphatic channels are present in the subareolar skin and follow the duct and lobular complexes and most frequently drain into the lymph node chains located in the axillary basin. Breast lymphatics can also directly communicate with the infraclavicular/supraclavicular or internal mammary lymph node chains. Intramammary nodes are located within the breast parenchyma and can contain the metastatic tumor from the primary site. The axillary nodes are divided into three levels (I-III) based on their relationship with the pectoralis minor muscle (Fig. 58-1). Level I is located caudal and lateral, level II nodes are beneath the muscle, and level III (infraclavicular) nodes are located cranial and medial to the pectoralis minor. Rotter nodes or intrapectoral nodes are sandwiched between the pectoralis major and minor muscles. Orderly spread of tumor cells from the primary breast tumor most commonly starts at the level I axillary nodes and then proceed to level II and level III lymph nodes. “Skip” metastases that defy this order can occur but are less likely and a standard axillary dissection for sentinel positive disease requires pathologic analysis of level I and level II axillary nodes.

FIGURE 58-1 • Schematic representation of the regional nodes for breast cancer.

(From Donegan WL, Spratt JS: Cancer of the breast, ed 3, Philadelphia, 1988, Saunders, p 19.)

The supraclavicular lymph nodes are located within the space defined by the omohyoid muscle and tendon (lateral and superior borders), the internal jugular vein (medial border), and the clavicle and subclavian vein (inferior border).² The internal mammary lymph node chain is encased within the endothoracic fascia in the parasternal space and runs alongside the corresponding artery and vein. The nodes located in the first through third intercostal spaces are the most common lymphoscintigraphic sites of drainage. Table 58-1 presents the location of the sentinel nodes as identified on lymphoscintigrams following peritumoral or intratumoral injection of technetium-99 sulfur colloid in women with early stage breast cancer. In the NSABP B-32 trial and the multicenter validation study, lymphoscintigrams were not required and the location of the sentinel nodes was determined by a hand-held gamma detector.^3,4 This may account for the lower reported incidence of internal mammary nodes in these two studies. Overall, level I-II axillary nodes were the most common location for sentinel nodes regardless of tumor location. Internal mammary nodes were identified in less than 5% of women when lymphoscintigrams were not performed. Internal mammary nodes are not identified if the injection is intradermal.⁶ Estourgie et al.⁵ reported localization to the internal mammary node (IMN) node in 21.8% of all patients with the use of lymphoscintigrams. Inner quadrant tumors had a higher incidence of hot spots in the IMN nodes with the highest frequency in tumors in the lower inner quadrant. Upper quadrant lesions are more commonly localized to the second, third, and fourth interspaces while central and lower quadrant lesions are more commonly localized to the second, third, fourth, and fifth interspaces. Drainage to the supraclavicular nodes occurred in less than 1% of patients except for those with a central primary tumor. It should be noted that localization to a nodal region does not mean nodal positively. Biopsy of hot internal mammary nodes has demonstrated nodal positively in 8% to 27% of women.⁷

Epidemiology

Breast cancer is a major public health problem throughout the world. In 2007,⁸ it was estimated that 1.3 million women were diagnosed with new cases of breast cancer and 465,000 deaths from breast cancer were expected. Breast cancer remains the most frequently diagnosed cause of cancer death in women worldwide, only slightly surpassed by cervical cancer in economically underdeveloped countries. In the last 25 years, overall breast cancer rates have risen, including in Africa and Asia, continents with very low rates of breast cancer. The highest incidence of breast cancer occurs in North America, Northern and Western Europe, and Australia.^8–10 The San Francisco Bay area has the highest rate of breast cancer in the United States.¹¹ Currently a woman living in the United States has a 12% or a 1 in 8 lifetime risk of developing breast cancer. In 2009 in the United States, an estimated 192,370 new cases of invasive breast cancer will be diagnosed among women along with an estimated 62,280 additional cases of in situ breast cancer.¹ Approximately 40,930 women are expected to die from breast cancer. About 1910 cases of breast cancer are expected to occur among men with 450 deaths.

From 1980 to 2001, breast cancer incidence rates initially increased rapidly followed by a slow increase after 1987. From 2001 through 2004, the incidence rates dropped 3.5% per year. The increase in the 1980s is largely attributed to the introduction and routine use of screening mammography and is reflected by the shift in the increase in detection of smaller previously nonpalpable breast tumors (2 cm) and a decrease in tumors of 3 cm by nearly a third.¹² The increase has also been attributed to the popular use of hormone replacement therapy (HRT) and lifestyle choices that changed reproductive patterns. After 2000, the incidence rates began to drop in the United States, due largely in part to the declining use of HRT following the publication of the Women’s Health Initiative randomized trial in 2002^13–16 and a measurable reduction in the number of annual screening mammograms.¹⁷ Among women younger than 50, incidence rates have remained unchanged since 1986. Early detection by screening mammography also brought an increased incidence of in situ breast tumors in the 1980s and 1990s, particularly in women 50 years and older, but has since leveled off in this population. However, the incidence of in situ breast cancer continues to rise in young women.¹⁸

Epidemiologic studies of breast cancer focusing on age, race, ethnicity, and geographical location have shown disparities.¹⁹ For example, although whites have a higher incidence of breast cancer than African American women, African American women have a higher incidence before the age of 40 years and a higher mortality rate at any age.^9,20

The rate of breast cancer mortality has decreased since 1990-2004 by 2.2% annually with most benefit witnessed in women under age 50 years and in white women rather than African Americans or other racial and ethnic groups. Improved survival has been attributed to early detection and improvements in treatment.

Pathology

The American Joint Committee on Cancer classifies breast cancers into the following histopathologic categories^73,74 in situ cancers and invasive cancers. The in situ cancers include those that are not otherwise specified (NOS), ductal carcinoma in situ (DCIS, intraductal cancer), and Paget disease with DCIS. The invasive cancers include those that are not otherwise specified, ductal, lobular, medullary NOS, or medullary with lymphoid stroma, mucinous, tubular, papillary (predominantly micropapillary pattern), Paget with invasive cancer, inflammatory, undifferentiated, squamous cell, adenoid cystic, secretory, and cribriform.

The term lobular carcinoma in situ (LCIS) is a misnomer in that LCIS is considered a marker for an increased risk for the development of breast cancer and not a cancer in itself. LCIS was first described by Foote and Stewart in 1941.⁷⁵ It is characterized by the presence of large, nonuniform in size, discohesive epithelial cells that fill the acinar spaces to varying degrees.^76,77 Mastectomy specimens have demonstrated LCIS to be multicentric in 90% of cases and bilateral in 35% to 59%.^76,77 LCIS is usually estrogen receptor-positive. C-erbB-2 and p53 are infrequently overexpressed.^77,78 Loss of E-cadherin (CDH1), an adhesion molecule, is found in 95% of LCIS.⁷⁹ LCIS, when not associated with an invasive cancer, is usually detected as an incidental finding at the time of a benign biopsy. It is found more frequently in premenopausal women with an average age of 45 years.⁷⁷ The invasive cancer, which develops subsequent to a diagnosis of LCIS, is most often an invasive ductal cancer, which can occur in either breast.^76,80 In the National Surgical Adjuvant Breast and Bowel Project (NSABP) randomized trial for DCIS, 182 patients were inadvertently entered with LCIS. Five percent of these women who had excision alone developed a subsequent invasive cancer in the index breast compared with 5.6% in the contralateral breast.⁸¹ The respective rates for DCIS were 9% for the index breast and 1.2% for the contralateral breast. The median follow-up was 12 years. It was of interest that all of the cancers that developed in the index breast occurred in the vicinity of the LCIS. Treatment options for LCIS in the absence of an invasive cancer range from bilateral mastectomy to excision alone. Because of the propensity for the bilateral risk, unilateral mastectomy is not appropriate. The most commonly accepted approach is excision alone with close follow-up. The role of negative resection margins for LCIS in diminishing the subsequent risk is unknown. Tamoxifen has been shown to decrease (but not eliminate) the risk of cancer following a diagnosis of LCIS and has been used as a prevention strategy. In the NSABP P-1 prevention trial, tamoxifen resulted in a 56% decrease in the risk of a cancer in women with LCIS.^82,83

Pleomorphic LCIS is a variant of LCIS. The majority of reported cases of pleomorphic LCIS have occurred in association with pleomorphic invasive lobular cancer.⁸⁴ Pleomorphic LCIS has been found to have a higher proliferation rate and greater percentage of p53 positivity when compared with classic LCIS and is estrogen-receptor-positive, HER-2/neu negative, and E-cadherin negative.⁸⁴ Necrosis may be seen in 40%.⁸⁴ Recent studies have suggested that pleomorphic LCIS is more closely related to LCIS than DCIS.^84,85 The biologic course of pleomorphic LCIS is unknown. In the past many of these lesions were classified as DCIS due to the presence of necrosis and calcifications. Sneige et al.⁸⁴ reported outcome in 10 cases. None of the seven cases who underwent adequate excision or simple mastectomy recurred.

Ductal carcinoma in situ by definition is carcinoma confined to the pre-existing duct system of the breast without penetration of the basement membrane by light microscopy. Pathologically, DCIS may be characterized by its architectural pattern (comedo, solid, cribriform, papillary, and micropapillary), nuclear grade (low, intermediate, high), and the presence or absence of comedonecrosis. Less common architectural patterns include apocrine, clinging, signet-cell cystic, hypersecretory, and neuroendocrine.⁸⁶ It is not uncommon for more than one architectural pattern to be present in a single lesion. DCIS currently represents approximately 21% of all breast cancers and 20% to 30% of those that are mammographically detected.

The natural history of DCIS is characterized by a low risk for axillary metastases (<5%) and an extremely low breast cancer mortality (<5%). DCIS is detected as a mammographic abnormality (usually calcifications) in 80% to 90% of women with 10% to 20% having a physical finding (palpable mass, bloody nipple discharge). The extent of DCIS may be underestimated by the area of calcification on the mammogram, especially if magnification views are not obtained.^87–89 In addition, pathologic examinations of mastectomy specimens in women with DCIS not infrequently reveal discontinuous extension in the duct system.⁹⁰ Contiguous extension has been described in two directions: one centrally towards the nipple and one peripherally and laterally.⁹¹

This somewhat unpredictable pattern of spread contributes to the difficulty in assessing the pathologic extent of DCIS. Micropapillary and low-grade lesions tend to have a more diffuse and discontinuous pattern of spread.⁹² In contrast to high-grade DCIS, two thirds of low-to intermediate grade DCIS have been reported to have a multifocal disease with distances between foci of up to 1 cm.^87,90 More recently, MRI has been used to evaluate the extent of DCIS.^93,94 In a study of 45 patients from the University of California, San Francisco, MRI overestimated the pathologic extent of disease more than twofold in 23% and underestimated it in 9%.⁹³ A discrepancy between the pathologic size and MRI extent was more frequently observed with the MRI patterns of regional or multiregional enhancement and in estrogen receptor-positive DCIS.

Foci of occult invasion are more often observed with high-grade DCIS, those with comedonecrosis, or with a size greater than 2.5 cm^88,92,95 The estrogen receptor is positive in 90% of low-grade DCIS and 25% of high-grade lesions. Overexpression of c-erBB-2 and P53 is present in less than 20% of low-grade lesions but in two thirds of high-grade lesions.⁹⁶ DCIS is considered a precursor for the development of an invasive ductal cancer in the index breast. However, it is not an obligate precursor in that not all DCIS progresses to an invasive cancer. In a study of women who had undergone a biopsy for what was interpreted at the time as benign disease and on subsequent review was found to be low-grade DCIS, 39% 15 to 42 years later developed an invasive cancer in the index breast.^97,98

Pathologic classification systems have been proposed to predict the clinical behavior of DCIS in terms of ipsilateral breast tumor recurrence (IBTR), either invasive or noninvasive after breast-conserving surgery and breast cancer specific survival. A system which stratifies risk for recurrence is useful in defining the roles of radiation and mastectomy. A 1997 international consensus conference⁹⁹ recommended that a number of features should be described in the pathology report including nuclear grade, necrosis, polarization, architectural pattern, margin status, size, presence of absence of calcifications, and relationship of calcifications to the DCIS. Silverstein¹⁰⁰ proposed a classification system (the University of Southern California/Van Nuys Prognostic Index-USC-VNPI) based on clinical and pathologic features to predict biologic behavior. The scoring system has four categories (age, size, margins, and nuclear grade combined with necrosis) and within each category there are three scores (1, 2, 3). The age scores are 1 greater than or equal to 60 years, 2 = 40 to 60 years, and 3 ≤ 40 years. The size scores are 1 less than or equal to 1.5 cm, 2 = 1.6-4 cm, and 3 greater than or equal to 4 cm. The margin scores are 1 greater than or equal to 1 cm, 2 = 1 to 9 mm, and 3 less than or equal to 1 mm. The pathologic scores are 1 = low grade and no necrosis, 2 = low grade with necrosis, and 3 = high grade. Summing each of the category scores results in total scores ranging from 4 to 12. These four factors were culled from a number of factors evaluated in a multivariate analysis of prognostic factors for IBTR in patients with DCIS treated in their practice. It should be noted that their work was based on detailed clinical (mammographic and operative findings) and pathologic correlation and required processing of the entire surgical specimen with a uniform thickness of 2 to 3 mm and sequential analysis, thereby providing the most accurate assessment of size and margin status. The reproducibility of this system has been questioned by a number of investigators in retrospective and prospective studies and their detailed pathologic evaluation is not commonly used in clinical practice.^101–104 At the present time, there is no one system which best predicts outcome and multiple factors should be considered.

Paget disease of the breast consists of 1% to 3% of all breast cancers.^105–107 It is characterized by the presence of neoplastic cells in the epidermis of the nipple. There is no penetration of the dermal basement membrane and therefore Paget is a form of in situ cancer. HER-2/neu overexpression is common in Paget disease.^108,109 Paget is frequently associated with underlying ductal carcinoma in-situ or an invasive cancer or both,^105,107,110 although the process may be isolated to the nipple. Chen et al.¹¹¹ reported a decrease in the overall incidence of Paget disease in the Surveillance, Epidemiology and End Results (SEER) registry from 1988-2002. However, no underlying breast cancer was identified in only 13% of the 1704 women with Paget disease. Paget disease has erythema and an eczematous scaling of the nipple skin. It may progress to crusting and frank ulceration. The prognosis of Paget disease is related to that of the underlying invasive cancer, if present.

Invasive ductal cancer is the most common invasive cancer.¹¹² Approximately 50% of invasive ductal cancers will have an associated DCIS component. The term extensive intraductal component (EIC) was first described by Schnitt et al. from Harvard.¹¹³ By definition, the entity consists of the simultaneous presence of DCIS comprising 25% or more of the primary tumor and its presence in the normal surrounding breast tissue. The definition also includes DCIS with focal areas of invasion (i.e., microinvasive cancer). While there have been various definitions of microinvasive cancer, it is currently defined by the AJCC staging manual as foci of invasion no greater than 0.1 cm.^73,74 In instances where there are multiple foci of invasion, only the largest focus of invasion should be considered. The clinical presentation of EIC positive tumors is usually the presence of mammographic calcifications without an associated mass.¹¹⁴ The clinical significance of an extensive intraductal component is its association with an increased risk of ipsilateral breast tumor recurrence after breast-conserving surgery with or without radiation.^115–122 This observation has been explained in part by the pathologic studies of Holland et al.¹²³ In a serial subgross and correlated mammographic examination of 217 mastectomy specimens, they correlated the presence of an extensive intraductal component in an invasive ductal cancer with the incidence of residual tumor following an excisional biopsy. EIC positive tumors were significantly more likely to have residual tumor (predominantly DCIS) and at greater distances from the primary than EIC negative tumors. At a distance of 6 cm from the edge of the invasive cancer, 21% of EIC positive tumors had residual cancer compared with 8% of EIC negative tumors. The presence of EIC has not been associated with an increased risk of chest wall recurrence after mastectomy.¹²⁴ EIC positive tumors have also not been found to have a higher risk of distant metastases¹¹³ and in some series have been reported to have a better prognosis.^124,125

Invasive cancers can be further classified by their histologic grade. A modification of the Scharf Richardson Bloom system is commonly used. The total score for grade is based on adding the individual scores for differentiation, mitoses, and nuclear pleomorphism with each having a score of 1 to 3 and total scores of 3 to 9. Low-grade tumors have scores of 3 to 5, intermediate grade scores of 6 to 7, and high grade of 8 to 9.¹²⁶ The Nottingham grading system¹²⁷ assigns scores of one to three for the percentage of tubule formation, the degree of nuclear pleomorphism and mitotic count in a defined field area. The sum of the three scores determines the final grade with three grade divisions. Histologic grade is a prognostic factor for local-regional recurrence following mastectomy^128–130 and breast-conserving surgery with or without radiation^131–138 and has been correlated with breast cancer specific survival.^134,139

The presence of lymphovascular invasion (LVI) has been associated with an increased risk of local-regional recurrence following breast-conserving surgery with or without radiation and mastectomy, and with an overall worse prognosis, especially in axillary node-negative women.^{115,140–150} Colleoni et al.¹⁵⁰ found extensive lymphovascular invasion to be associated with a significant decrease in disease-free and overall survival in axillary node-negative women.

Invasive lobular cancers represent 5% to 10% of all breast cancers and represent the second most common type of breast cancer. They arise from the lobules and are often characterized by multifocality and a higher incidence of bilaterality in some series.^151,152 There are five subtypes of invasive lobular cancers (classic, tubulolobular, solid, alveolar, and pleomorphic). Classic invasive lobular cancers are characterized by the presence of small relatively uniform cells that invade the stroma in a single-file pattern with little or no desmoplastic reaction. The pleomorphic variant has a similar pattern of invasion but the cells are larger and have more nuclear variation. The different subtypes of invasive lobular cancer have been associated with different prognoses.¹⁵³ The pleomorphic variant has been associated with a larger primary tumor size, a greater frequency of positive nodes and a worse prognosis.^154,155 Invasive lobular cancers are usually estrogen and progesterone receptor-positive and HER-2/neu negative.^156,157

Several of the more uncommon histologic subtypes of breast cancer have been associated with a very favorable prognosis. These include tubular and mucinous or colloid cancers. Axillary nodal metastases are less common and they tend to occur in older women.^158–161 In contrast, invasive micropapillary carcinoma is associated with a worse prognosis due to its frequent involvement of axillary nodes and larger primary tumor size.¹⁶² Medullary carcinomas are associated with a prominent lymphocytic infiltrate and are well circumscribed grossly and microscopically. While more often high grade, they tend to be associated with a better prognosis. They occur more frequently in younger women and have been associated with BRCA1 mutation carriers.¹⁶³

The more unusual types of breast cancers include adenoid cystic cancer, metaplastic cancer, and its variant spindle cell carcinoma, primary neuroendocrine small cell carcinoma, primary lymphoma, and the sarcomas (malignant cystosarcoma phylloides, stromal sarcoma, angiosarcoma, fibrosarcoma, carcinosarcoma, and liposarcoma).

Guidelines for the basic elements of a pathology report for breast cancer have been established by the College of American Pathologists.¹⁶⁴ The following is a list of information that is pertinent for the radiation oncologist to consider in the decision-making process:

Staging

The current staging system for breast cancer is presented in Table 58-2, A and can be found in the seventh edition of the AJCC Cancer Staging Manual.⁷⁴ The prior staging system is presented in Table 58-2, B.⁷³ The staging system relies on primary tumor size, the presence and extent of regional node involvement, and distant metastases to define prognostic categories. The clinical classification (cTNM) is based on findings from the physical examination, imaging studies including mammography, ultrasound, MRI, CT scans, but not lymphoscintigraphy and pathologic examination of breast and/or regional node biopsies. Dimpling of the skin or nipple retraction does not imply T₄ disease. Extensive imaging is not required to designate a case as M₀ provided the clinical history and examination do not suggest the presence of distant metastases. The term M₀ (i+) has been introduced to denote patients with circulating tumor cells in the blood or bone marrow.

Table 58-2B TNM Classification for Breast Cancer from the AJCC Staging Manual, 6th Edition

Rights were not granted to include this table in electronic media. Please refer to the printed book.

The pathologic classification (pTNM) includes information from a detailed pathologic examination of the primary tumor, regional nodes, and distant metastases when present. For invasive cancers, the pathologic size is that of the invasive component only. The microscopic measurement is considered the most accurate for small invasive cancers and for larger tumors the gross measurement may be used. Adding the size of the invasive cancer obtained from the core biopsy to the residual invasive cancer at the time of primary surgery is not recommended. For multiple synchronous primary tumors, the size of the largest invasive focus determines the pathologic tumor size. The size of each individual focus should not be added together. Focal dermal involvement with an intact epidermis results in a classification based on the size of the primary tumor. The combination of the T, N, and M categories results in a clinical and pathologic stage.

Lymph node metastases are staged by the size of the largest focus regardless of the method of detection. cN₁ designates clinically positive axillary nodes that are nonfixed while nodes that are fixed to one another or matted are cN_2a. Internal mammary nodes detected by CT or ultrasound but not lymphoscintigraphy are cN_2a in the absence of clinically positive axillary nodes. Clinically positive axillary and IMN nodes are cN_3b. Clinically positive supraclavicular nodes are cN_3c. If the clinically positive node(s) is confirmed by fine-needle aspiration (FNA) or core biopsy, the (f) modifier is added (i.e., cN₁[f]).

Pathologically positive nodes are classified as macrometastases, if the metastatic deposit is more than 2 mm. Micrometastases are defined as deposits more than 0.2 mm and less than 2 mm. T₁N₁miM₀ tumors are now classified as stage IB. Isolated tumor cells are defined as clusters of 0.2 mm or not exceeding 200 cells in a single cross section of a node (pN_0i+). They are usually detected by immunohistochemistry or molecular methods (pN_0mol+) but can be verified by hematoxylin and eosin stains. Intramammary nodes are considered axillary nodes for the staging system and metastatic deposits in the axillary fat are classified as positive nodes. The number of positive nodes and their location further defines the staging groups (Table 58-2, B). The modifier (sn) is used for patients who have sentinel node biopsies only or when the total number of nodes removed is less than 6 nodes.

Following neoadjuvant chemotherapy or hormonal therapy, the pathologic classification is designated at yp TNM. The size of the residual primary tumor may be difficult to assess after neoadjuvant systemic therapy, and in the presence of multiple foci, the size of the largest focus determines the yp T stage with the modifier “m” for multiple foci. The extent of the residual tumor can be assessed by determining the distance over which it extends, the number of slides/blocks in which it is seen, and its cellularity. Patients with ypN₀(i⁺) disease in the axilla in the absence of residual disease in the breast are not considered to be pathologic complete responders.

The new staging system reflects changes in the evaluation and treatment of breast cancer. At present it does not incorporate assessment of the biology of the cancer as measured by genetic profiles, molecular subtypes, and biochemical markers. Future modifications may include this information to enhance the prediction of outcome for the individual patient, resulting in improvements in individualized therapy.

Surgical Considerations

The goal of breast-conserving surgery (BCS) is to eradicate both invasive and in situ disease and to achieve a desirable cosmetic result, which can be appreciated by both the surgeon and patient. Breast-conserving surgery has been shown to be an effective treatment when compared with mastectomy, yet BCS lacks the psychological impact and impaired body image that may accompany mastectomy.^165,166 The decision to undergo BCS should be a carefully thought out plan that incorporates information from pathology and radiology and patient preference in an attempt to individualize care.

Once a patient is determined to be a candidate for BCS, attention turns to the technique for surgical excision. Localization of the tumor is performed either by palpation or placement of a wire under mammography, ultrasound, or MRI guidance. New techniques for localization include placement of a radioactive seed at the tumor, which can be identified intraoperatively using a hand-held gamma probe. However, only a few institutions have incorporated this into surgical practice.^167–169 The size of excised breast tissue depends on multiple factors, which include tumor size and location, breast size, and radiographic findings. Complete excision of the tumor and any suspicious radiologic findings should be done in the most cosmetically appreciated way. Some surgeons advocate removing additional margins at the time of surgery to decrease the incidence of positive or close margins at final pathology. It is not necessary to remove the skin overlying the tumor except in cases where the skin is tethered or in intimate contact with the cancer.

After removal of the specimen, it should be adequately marked by the surgeon to ensure proper orientation for pathologic assessment. The surgeon should place sutures to signify distinct margins or the specimen can be painted in the operating theater with different colors of ink to each margin. Specimen orientation is an important step that is essential if the patient requires re-excision for a compromised margin. Proper orientation allows the surgeon to return to the operating theater and direct the re-excision to the identified margin, limiting the extent of tissue removal and allowing for continued care of cosmesis. Typically, nonpalpable lesions are then evaluated under mammography to identify complete excision of the known malignancy and suspicious radiographic findings. Once the surgeon has been notified of these results, the lumpectomy cavity is then marked with titanium clips (anterior, posterior, medial, lateral, superior, and inferior), which aid in delivery of radiation and subsequent mammographic surveillance.

Standard wound closure is a two-layer closure at the dermal/epidermal level. Generally, the cavity is left alone to fill with seroma fluid so as to maintain the shape of the breast. Recently, new techniques of wound closure have incorporated breast-flap advancement for volume replacement following large partial mastectomy resections.^170–173 In an attempt to offer breast-conserving surgery, patients who had previously been destined for mastectomy are now assessed for “oncoplastic” surgery. The approach to oncoplastic surgery requires the surgical team to rethink the traditional curvilinear incisions in the upper pole and radial incisions in the lower pole, and focus on an approach that allows incorporation of fibroglandular rearrangement. These newer techniques have broadened the surgical choices for many patients who have concerns regarding mastectomy; however, there are inherent risks that cannot be overlooked.¹⁷⁴ Further excision for compromised margins could be very difficult with this volume-replacement technique, and disruption of lymphatic channels may have consequences as it pertains to future lymphatic mapping in cases of local recurrence. Planning for radiation therapy must also be considered in terms of the identification of the primary tumor site for delineation of the boost volume. Placement of surgical clips in the tumor bed is especially important in these women. There have been no randomized controlled trials involving oncoplastic surgery and the risk of a local recurrence. Small cohorts have shown promising results for both aesthetics and local control, but careful consideration must be taken into account when planning oncoplastic techniques until better, long-term data are available to support this technique for routine use.^175–177

If a patient desires BCS but has a tumor/breast ratio that initially would not be amendable to lumpectomy, there is the possibility of tumor “downstaging” with neoadjuvant therapy. Chemotherapy has been shown to be equally effective whether it is given presurgery (neoadjuvant) or postsurgery (adjuvant).^178,179 The discussion of neoadjuvant therapy should be a multidisciplinary approach to ensure appropriate recommendations from medical, radiation, and surgical oncology. Magnetic resonance imaging (MRI) has become a valuable tool for predicting BCS success and is rapidly gaining wide use in this setting.^180–182 The MRI images can direct the surgeon to the volume of breast tissue that should be removed in relation to response to therapy. However, mammography is equally important in assessing the extent of residual calcifications that need to be excised at the time of lumpectomy. At the start of neoadjuvant therapy, the patient should have a titanium clip placed under the core biopsy of the tumor to mark the location. The titanium clip can be used for localization in cases where a complete radiographic response is achieved.¹⁸³ Clip placement has been associated with improved local control in patients receiving neoadjuvant chemotherapy and undergoing breast-conserving surgery.¹⁸⁴ The volume of breast tissue excised after neoadjuvant chemotherapy is generally the residual nidus and not the original volume. However, all malignant appearing calcifications must be excised before radiation. Margin assessment may be more difficult in tumors that regress in a multifocal pattern.