88 Cancer of the Breast N. Lynn Henry, Payal D. Shah, Irfanullah Haider, Phoebe E. Freer, Reshma Jagsi, and Michael S. Sabel

SUMMARY OF KEY POINTS Incidence and Epidemiology cancer should be referred for genetic starting at age 40. The 2015 • Breast cancer is the most evaluation. American Cancer Society (ACS) frequently diagnosed cancer in Biology and Estimation of Risk guidelines recommend a yearly women in the United States, • The expression of nuclear estrogen mammogram starting at age 45, with accounting for an estimated and progesterone receptors plays an the option to transition to biennial 268,670 new cases of invasive important role in the differentiation screening at age 55, and the option breast carcinoma, 63,960 new cases and growth of normal breast to begin screening between ages 40 of in situ carcinoma, and 41,400 epithelium and the response of and 44. Similarly, the American deaths in 2018. breast cancer cells to hormonal College of Radiology recommends • In the United States, the age-specific therapeutics. yearly mammography starting at age incidence of breast cancer increases • ERBB2 (human epidermal growth 40. The 2016 US Preventive Services with age, to a lifetime risk of breast factor receptor 2 [HER2]) is a Task Force (USPSTF) breast cancer cancer of 1 in 8; by age 40 years, growth-signaling molecule on the screening guideline recommends approximately 1 in 203 women will surface of normal breast cells that is biennial screening mammography for have been diagnosed with breast overexpressed in approximately 20% women ages 50 to 74 years and an cancer; at 60 years of age, the figure of breast cancer tumors, contributing individualized approach for women is 1 in 28 women. to growth autonomy and genomic ages 40 to 49 years. The ACS also • Incidence rates have been rising instability. recommends that breast magnetic about 1% per year; mortality rates • Molecular analyses have defined at resonance imaging (MRI) be used in stayed relatively constant until least four biologic subtypes of breast addition to mammographic screening recently, when annual decreases cancer, including basal type, two in women with a lifetime risk of were seen. luminal types (luminal A and luminal breast cancer greater than 20%. • A sharp decline in the incidence of B), and the HER2-enriched type. • Microcalcification and soft tissue breast cancer in the early 2000s • BRCA1 and BRCA2 are tumor density are the major indications for followed a decrease in the use of suppressor genes that play a critical biopsy after mammographic postmenopausal hormone role in homologous recombination screening; the mammographic replacement therapy; the incidence repair (HRR) of DNA damage; abnormality with the highest rate of rate has since remained stable. inherited mutations in these and malignancy is a mass with • Age, family history, and both other HRR genes are associated with associated calcification. endogenous and exogenous ovarian an increased risk of breast cancer. • For patients with breast symptoms hormone exposure have an • Regardless of the criteria used by an or palpable abnormalities, important effect on risk and have individual physician and patient to mammography, in conjunction with been incorporated into models that define high risk, four possible actions other imaging modalities such as predict individual risk of breast may be taken, some of which can be ultrasonography and breast MRI as cancer; diet, alcohol use, and other used simultaneously: (1) enhanced indicated, is used to characterize the factors play a smaller role. surveillance, (2) behavioral suspicious area, evaluate the • African Americans with breast cancer modification, (3) chemopreventive remainder of the breast for occult are diagnosed at more advanced strategies, and (4) prophylactic lesions, and assess the contralateral stages and have worse outcomes, mastectomy or oophorectomy. breast. including survival. • Malignant breast masses classically • Inherited mutations in BRCA1, Screening and Diagnosis are nontender and firm, with irregular BRCA2, PTEN, TP53, STK11, ATM, • Although exact guideline borders. CHEK2, PALB2, and other genes recommendations for mammography • Diagnostic methods include can increase the risk of breast screening remain controversial and needle-core biopsy with ultrasound cancer; individuals with a personal or are debated, all guidelines agree that or stereotactic guidance, fine-needle family history of hereditary breast most lives are saved with screening aspiration cytology, and excisional

1560 Cancer of the Breast • CHAPTER 88 1561

biopsy, with or without wire chemotherapy in addition to • The results of the serial localization. endocrine therapy in node-negative, meta-analyses performed by the Management of Noninvasive hormone receptor–positive breast Early Breast Cancer Trialists’ Disease cancer, and is undergoing Collaborative Group (EBCTCG) show evaluation for use in node-positive that the survival benefits of adjuvant • Lobular carcinoma in situ (LCIS) is a disease. chemotherapy and endocrine nonpalpable lesion that usually is • The seventh edition of the American therapy persist after 15 years of discovered with another indicator for Joint Committee on Cancer TNM follow-up. biopsy; it is an indicator of risk of staging system takes into account • Challenges that require resolution subsequent invasive breast cancer. the increasing use of advanced include identification of biologic The rare pleomorphic variant of LCIS imaging and pathology techniques, parameters that more precisely seems to have a more aggressive such as sentinel node biopsy and predict the natural history of disease behavior and is considered more as immunochemistry; it also considers and the response to systemic a precursor of invasive lobular the number of involved nodes as a therapy; more effective treatments cancer rather than a marker of strong prognostic factor. are needed. breast cancer risk. • Sentinel lymph node mapping is now • Management of LCIS has shifted Management of Locally the standard of care for many toward observation after biopsy Recurrent Disease women with early breast cancer rather than mastectomy; increasing • Local recurrence is an indicator of because it is more accurate and less evidence shows that tamoxifen systemic relapse in most cases; an morbid than axillary dissection. should be considered as a exception may be local relapse in a • For patients with stages I and II preventive approach. breast that has undergone disease, breast conservation • Unlike LCIS, ductal carcinoma in situ conservation therapy. (lumpectomy followed by radiation (DCIS) is almost always first • Surgical removal alone may be therapy) and simple mastectomy are identified with mammography; it sufficient in some cases, but it often the major therapeutic options; for accounts for most of the increasing is combined with locoregional most patients, breast conservation is number of carcinoma in situ lesions radiotherapy and/or systemic the preferred approach. diagnosed. therapy. • Adjuvant therapy with cytotoxic • DCIS is more likely to be localized to drugs, endocrine therapy, and/or Management of Metastatic one area of the breast; therefore anti-HER2 therapy is recommended Disease most patients are candidates for depending on stage of disease and • Although a small percentage of breast conservation. Tamoxifen or, in pathologic characteristics. patients with metastatic breast postmenopausal women, aromatase • For patients with locally advanced cancer achieve long-term inhibitors should also be considered disease, multimodality therapy is disease-free survival, this stage of after lumpectomy and radiation recommended; the sequence of disease is generally not curable, and therapy to reduce the risk of another systemic therapy, surgery, and therapy is largely palliative. ipsilateral or a new contralateral radiation therapy depends largely on • A wide range of systemic, local, and breast cancer event, which can be the operability of the primary supportive therapies are available for invasive or noninvasive. disease. the palliation of metastatic breast Management of Early-Stage • For women with inoperable or cancer. Primary breast surgery may Breast Cancer inflammatory breast cancer, be indicated in selected patients with • Patients should undergo a complete preoperative chemotherapy is stage IV disease. history and physical examination. recommended, followed by surgery, • The selection of endocrine, • Bilateral mammography is indicated radiation therapy, and endocrine cytotoxic, or biologic therapy usually for all patients, and other breast therapy, if appropriate. Patients with is based on disease-free interval, imaging (ultrasound and MRI) should HER2-positive disease should also receptor status, HER2 status, be used as needed for each receive anti-HER2–targeted therapy. presence or absence of visceral individual patient; advanced imaging Preoperative endocrine therapy for involvement, performance status, studies are recommended only to patients with hormone receptor– age, and previous exposure to evaluate specific signs or symptoms positive disease may also be a systemic therapy. or in patients with locally advanced reasonable option. • The treatment approach for disease. • HER2 is a strong predictive marker, metastatic breast cancer generally • Prognostic factors include pathologic and accurate determination of HER2 involves sequential therapies, each tumor size, hormone receptor status identifies patients with T1bN0 continued until disease progression expression, HER2 status, axillary tumors and above who should be or unacceptable toxicity. nodal status, histologic subtype, and considered candidates for adjuvant • Patients with hormone receptor– tumor grade. therapy with a trastuzumab-based positive, HER2-nonamplified • Multiple gene expression profiling regimen. metastatic breast cancer should assays have been developed as • Expression of estrogen receptor and/ be considered initially for prognostic tests. or progesterone receptor is a endocrine-based therapy, with • The 21-gene recurrence score assay predictive marker for response to systemic chemotherapy reserved for predicts benefit from systemic endocrine therapy. patients who are significantly

Continued 1562 Part III: Specific Malignancies

symptomatic or have compromised and ado-trastuzumab emtansine • Treatment with bisphosphonates or organ function. have changed the natural history of denosumab is indicated to minimize • Anti-HER2 agents, such as patients with HER2-positive bone complications in patients with trastuzumab, lapatinib, pertuzumab, metastatic breast cancer. bone metastases.

Between 2008 and 2012, breast cancer was the most common EPIDEMIOLOGY cancer diagnosed in women in the United States (overall incidence Incidence 123.1 per 100,000 women), followed by lung cancer (overall inci- dence 54.1 per 100,000 women).2 However, the leading cause of Worldwide, breast cancer is the most common type of cancer among cancer death was lung cancer (37.8 per 100,000 women), followed women.1 In the United States, approximately 268,670 new cases of by breast cancer (21.9 per 100,000 women).2 Mortality rates from invasive breast carcinoma, 63,960 new cases of in situ carcinoma, and breast cancer have been decreasing over time in the United States 41,400 deaths were expected in 2018.2 In women, breast cancer accounts (see Fig. 88.1). Around the world, there are large variations in for 29% of new cases of cancer and 14% of cancer deaths, second incidence, mortality, and survival, which may be a result of several only to lung cancer as a cause of cancer-specific death. Approximately underlying complex factors, including age, ethnicity, diet, and 1% of breast cancers occur in males, and 90% of these are estrogen lifestyles, including reproductive issues.1 Breast cancer is increasing receptor (ER) positive.3 Incidence rates of breast cancer decreased by in less-developed countries, likely related to changes in lifestyle 2% per year from 1999 to 2005, and have remained relatively stable factors. Most of our knowledge regarding specific risk factors has since then (Fig. 88.1). The decline appears to be partly the result of been derived from large observational studies in developed countries, a decrease in the use of postmenopausal hormone replacement therapy such as the Nurses’ Health Study (NHS) and NHS II5 in the United (HRT).4 Although incidence rates (all races combined) are substantially States, and the Million Women Study in the United Kingdom.6 higher for women older than age 50 years, approximately 23% of Some of these risk factors are briefly summarized in the following breast cancers are diagnosed in women younger than age 45 years sections. (Table 88.1). Diet The best-established dietary factor associated with increased risk for breast cancer is higher consumption of alcohol.7 Initial studies correlated higher fat intake and decreased intake of fruit and vegetables with an US breast cancer incidence increased risk for breast cancer, but more recent prospective studies 160 have failed to confirm those observations.8–11 One randomized study (Women’s Healthy Eating and Living [WHEL] Study) in breast cancer 120 patients who consumed a low-fat and high–fruit-and-vegetable diet did not demonstrate a decreased risk of breast cancer recurrence.11 80 However, a second randomized study (Women’s Intervention Nutrition Study [WINS]) demonstrated a lower risk of recurrence with a decreased intake of dietary fat associated with modest weight loss in breast 40 cancer survivors.8 Dietary phytoestrogens, such as those found in soybeans, have a 0 chemical structure that is similar to that of 17 -estradiol and can

Incidence per 100.000 women β 1991 1995 1999 2003 2007 2011 bind to the ER to compete with estrogens. Their consumption may have a weak protective effect against breast cancer. More recent studies have suggested an inverse association of soy intake and breast cancer US breast cancer mortality 12,13 40 in Asian but not non-Asian populations. Although there is significant interest in the association of vitamin D intake and breast cancer risk, there are insufficient data to draw conclusions regarding potential 30 benefit at this time; some data suggest that there may be an association in postmenopausal women.14,15 20

10 Table 88.1 Risk of Breast Cancer in US Women

Mortality per 100.000 women 0 1991 1995 1999 2003 2007 2011 Age Range (years) Breast Cancer Risk 30–40 1 : 227 Whites African Americans Hispanics* 40–50 1 : 68 Asians/Pacific Islanders* American Indians/Alaska Natives* 50–60 1 : 44 *Incidence and mortality data not available before 1992. 60–70 1 : 29 Lifetime (to age 110 years) 1 : 8 Figure 88.1 • Incidence and mortality of breast cancer in women in the United States between 1991 and 2011, by race. (From National Cancer Institute. Data from National Cancer Institute. Breast cancer risk in American women. www. Breast cancer—patient version. https://www.cancer.gov/research/progress/ cancer.gov/cancertopics/factsheet/Detection/probability-breast-cancer. Accessed snapshots/breast. Accessed January 9, 2017.) December 1, 2017. Cancer of the Breast • CHAPTER 88 1563

Ionizing Radiation Table 88.2 American Board of Pathology Histologic The accumulated knowledge about radiation-related breast cancer Classification of Benign Disease risk in women derives mainly from epidemiologic studies of patients Histopathologic Findings Approximate Relative Risk exposed to diagnostic or therapeutic radiation and of Japanese atomic bomb survivors. Therapeutic radiation, particularly extended-field Nonproliferative No added risk radiation therapy for Hodgkin lymphoma, has been associated with Cysts an increased risk of breast cancer. The age of exposure is important, Duct ectasia with the highest breast cancer risk seen in women undergoing mantle Calcification irradiation around puberty, and is minimal for women exposed after menopause.16,17 Fibroadenoma Milk ductal epithelial hyperplasia Exogenous Hormones Sclerosing adenosis No added risk The NHS originally was designed to determine whether oral contracep- Papillomatosis Slight added risk tive use was associated with an increase in breast cancer. Analysis of Radial scars data from that study and many additional studies has confirmed that current users of exogenous hormones have a slightly increased risk of Complex sclerosing lesions ? developing breast cancer, although this risk resolves within 5 years of Moderate florid hyperplasia 1.5 : 1–2 : 1 discontinuation of therapy.18 In addition, use of exogenous hormones Atypical hyperplasia (ductal and 4 : 1 is associated with a decreased risk of colon and ovarian cancer; there lobular) 5 is no increase in overall cancer mortality associated with their use. Extensive ductal involvement of 7 : 1 The Women’s Health Initiative (WHI) randomized study of HRT in atypical hyperplasia postmenopausal women demonstrated an increased risk of invasive Lobular carcinoma in situ 10 : 1 breast cancer with the combination of estrogen plus progesterone therapy,19 but not with estrogen alone.20 Other studies have also Ductal carcinoma in situ 10 : 1 demonstrated that breast cancer risk is affected by type of HRT and time of initiation of HRT, with an increased risk seen with therapy started soon after menopause.21 After initial publication of the WHI results22 and widespread discontinuation of HRT, age-adjusted inci- dences of invasive breast cancer decreased in the United States between Familial History and Predictive Models of 2002 and 2003 by 7%, primarily in ER-positive tumors in women Breast Cancer Risk 50 years of age or older. Breast cancer is approximately twice as common among first- Reproductive Factors and Endogenous Hormones degree relatives of breast cancer patients as in those women with Breast cancer risk consistently has been correlated with an earlier age no family history of the disease. The two most prevalent breast of menarche, later age of menopause, nulliparity, and late age of first cancer susceptibility genes, BRCA1 and BRCA2, were identified by birth, all of which determine the cumulative number of ovarian cycles. linkage analyses in the mid-1990s but account for less than 20% This finding is consistent with the correlation between estrogen levels of familial clustering of breast cancer.30 Since then, rare germline and breast cancer risk. In postmenopausal women, an increased risk mutations in several other high-penetrance hereditary breast and of breast cancer has been demonstrated in those with estradiol levels ovarian cancer genes, such as TP53, PTEN, CDH1, and SKT11, in the highest quartile.23 have been shown to confer a substantially increased risk of breast cancer. Additional details about BRCA- and TP53-mutated cancers are Obesity and Body Habitus provided later. There is a clear association between obesity and breast cancer risk. Given that linkage analyses have failed to yield additional susceptibil- Women with a body mass index (BMI) in the overweight (25–30) or ity loci, it is likely that most of the remaining genetic susceptibility obese (>30) range have been shown to have an elevated risk of breast is a result of a number of moderate- and low-penetrance alleles that cancer. In addition, increased BMI is associated with an increased risk coexist in highly penetrant combinations in a polygenic model, although of breast cancer death.24,25 The increased risk of breast cancer with some familial risk may still be a result of rare, undiscovered, high- obesity in postmenopausal women appears to be mediated, at least penetrance gene mutations.31 Multiple moderate-penetrance hereditary in part, by elevated endogenous estrogen and, possibly, insulin levels.26 breast cancer genes including ATM, CHEK2, PALB2, and NBN have An increased risk of breast cancer has also been associated with increas- been identified.32 Pathogenic mutations in these genes are associated ing height, with a relative risk (RR) of 1.17 (95% confidence interval with an RR of breast cancer of 2 to 5, and risk management strategies [CI], 1.15–1.19) for every 10 cm of additional height.27 have been delineated.32,33 Apart from specific cancer susceptibility genes, an increasing Prior Breast Biopsy understanding of relevant predisposing factors has led to integrated Finding a nonproliferative lesion such as a cyst or mild hyperplasia efforts at evaluating a women’s individual risk.34 Several individualized of the usual type with a breast biopsy is generally not associated with risk assessment tools are available, including the Gail, Claus, BRCAPRO, an increased risk of breast cancer (Table 88.2). Proliferative lesions Tyrer Cuzick, and Breast and Ovarian Analysis of Disease Incidence without evidence of atypia, such as usual ductal hyperplasia, sclerosing and Carrier Estimation Algorithm (BOADICEA) models.35 Many of adenosis, and intraductal papillomas, are associated with a 1.5- to these tools are now available on the Internet—for example, http:// 2-fold increased risk of developing breast cancer. In contrast, the www.cancer.gov/bcrisktool/ can be accessed on the National Cancer risk of developing breast cancer is substantially increased in patients Institute (NCI) website. These tools are being used for clinical counsel- with a prior breast biopsy with evidence of atypical hyperplasia, ing for healthy women perceived to be at risk and to determine with an RR of breast cancer of 3.7 to 5.3.28,29 In contrast, the risk candidates for enhanced surveillance or chemoprevention.36,37 Given of developing breast cancer after a finding of flat epithelial atypia that for women living in the Western hemisphere the lifetime risk of (FEA) was only marginally increased compared with the risk from a breast cancer is approximately 11%, even a small RR reduction could nonproliferative lesion. translate into many thousands of lives saved. 1564 Part III: Specific Malignancies

BRCA1, BRCA2, TP53, and Hereditary Susceptibility to BIOLOGIC CHARACTERISTICS Breast Cancer AND PATHOLOGY Germline mutations in BRCA1 and BRCA2 represent the most common For years, the traditional breast cancer biomarkers (ER; progesterone identifiable cause of hereditary breast cancer. BRCA1 (chromosome receptor [PR]; and Her2/neu [HER2]) have been a principal deter- 17q) and BRCA2 (chromosome 13q) are tumor suppressor genes; minant of adjuvant and metastatic breast cancer therapy. Although it inherited mutations in BRCA1/2 confer impaired homologous was originally recognized in the late 1800s that antiestrogen treatments recombination repair of DNA and are highly penetrant for hereditary could be effective for breast cancer when Sir George Beatson first breast and ovarian cancer syndrome. Mutations are particularly prevalent reported the benefits of oophorectomy as palliative therapy for several in the Ashkenazi Jewish population. Depending on multiple factors young women with metastatic breast cancer,49 traditional cytotoxic including the specific gene mutation, modifiers, and the population chemotherapy was the mainstay of treatment for decades although it studied, the cumulative cancer risks by age 70 are estimated for a rarely led to the cure of metastatic cancer. More recently, the discovery female BRCA1 mutation carrier at approximately 60% to 70% for of cancer-specific therapeutic targets such as ER and HER2 has led breast cancer and 20% to 50% for ovarian cancer, and for a female to the successful development of therapies that are effective only in BRCA2 mutation carrier, 50% to 70% for breast cancer and 10% to the subset of patients whose tumors harbor these targets. Antiestrogen 30% for ovarian cancer,38,39 with higher estimates in earlier studies therapy in tumors that are ER and/or PR positive and anti-HER2 due to ascertainment bias and the early examination of families with therapy in tumors that demonstrate amplification of HER2 are strategies multiple breast and ovarian cancer cases.30,40 Individuals with pathogenic that have demonstrated a survival benefit beyond that provided by BRCA mutations may also be at increased risk for male breast cancer, cytotoxic therapy alone. prostate cancer, pancreatic cancer, and melanoma. Multiple other The modern diagnosis and therapy of breast cancer have evolved genes interact with BRCA1 and/or BRCA2 and play a role in homolo- from focusing almost entirely on these immunophenotypic features gous recombination repair, such as PALB2, ATM, and CHEK2; to considering other biologic targets and pathways that might promote mutations in several of these genes also confer risks of breast and tumor development and growth. Standard biologic therapies beyond other cancers, although the magnitude of these risks are lower than those targeting hormone receptors and HER2 include the mammalian the those conferred by BRCA.32,41 Why these genes predispose primarily target of rapamycin (mTOR) inhibitor everolimus and the cell cycle to breast and ovarian cancers remains unclear. Important to note, inhibitor palbociclib. prognosis of BRCA mutation carriers with breast cancer appears similar Perhaps the most significant paradigm shift has been in the setting to prognosis for sporadic breast cancer patients.42 of metastatic disease, with widespread active efforts to therapeutically Inheritance of a germline pathogenic p53 mutant allele causes the exploit the genomic profile of a tumor. This concept, which has been rare high-penetrance familial Li-Fraumeni syndrome, characterized successful in other cancers in which the underlying genetic aberration by multiple early-onset cancers, including sarcoma, leukemia, adre- can be targeted, such as imatinib mesylate in chronic myelogenous nocortical tumors, and breast cancer in women who survive childhood leukemia and gastrointestinal stromal tumor, has provided the impetus cancers. Thirty percent to 50% of sporadic cases of breast cancers for large-scale sequencing efforts of cancer genomes to identify addi- harbor somatic mutations in p53, based primarily on sequencing of tional targets. With the incorporation into the clinic of both germline exons 5 through 8 (the DNA-binding core), where approximately genetic and somatic molecular testing, clinicians are increasingly aware 90% of mutations are found. Currently, p53 mutations that have of factors such as deficient DNA repair and activation of molecular been observed in breast cancers are listed in the p53 database maintained signaling pathways that affect both standard and investigational breast by the International Agency for Research on Cancer. cancer treatment options. Myriad personal and family history factors should prompt In addition, microarray technologies that enable the simultaneous a clinician to refer a breast cancer patient for genetic evaluation, quantitation of all RNA species and genome-wide DNA copy number including diagnosis before the age of 45, Ashkenazi Jewish ancestry, changes have begun to provide diagnostic classifications that may be or diagnosis of triple-negative disease before age 60.33 Genetic evalu- superior to traditional histologic criteria for determining prognosis and ation begins with a comprehensive assessment of risk factors. Genetic therapy. Finally, the identification of both normal mammary gland and testing, if indicated, should be preceded by a thorough informed breast cancer stem cells has led to novel therapeutic approaches that consent that includes potential issues of uninformative results or are being tested in clinical trials. These advances and their implications variants of uncertain significance, and general implications of a for therapy are summarized in the following sections. positive result. Genetic testing for hereditary breast and ovarian cancer syndrome Histology mutations can occur in a targeted fashion or with full gene sequencing and evaluation of large genomic rearrangements. In addition, testing Invasive Breast Carcinoma can be conducted with multigene panels using massively parallel The majority of invasive breast cancers are epithelial in origin and sequencing that evaluates BRCA1/2 and multiple other genes simultane- are histologically heterogeneous. Most are adenocarcinomas arising ously.43 The clinical usefulness of testing for mutations in moderate- and from the terminal ducts. Invasive ductal carcinoma accounts for low-penetrance genes,44–46 particularly in the setting of phenotypic approximately 85% of breast cancers and can form ductal structures discordance, is often unclear. When testing with a multiplex panel, (Fig. 88.2). In contrast, invasive lobular carcinoma, which is character- pretest counseling should also include discussion of the possibilities ized by small, regular epithelial cells that tend to align in single file of unanticipated or phenotypically discordant findings in high- and grow around ducts and lobules, accounts for 5% to 15% of breast penetrance genes and low- or moderate-penetrance mutations that cancers (Fig. 88.3). The overall prognosis of classic invasive lobular may have undefined clinical validity.47,48 The most recent Genetic/ cancer is thought to be similar to that of the invasive ductal subtype, Familial High-Risk Assessment: Breast and Ovarian National Com- although there may be some differences in likelihood of response to prehensive Cancer Network (NCCN) Guidelines (Version 2.2017) chemotherapy and endocrine therapy compared with ER-positive, provides recommendation for management of breast cancer risk in HER2-negative invasive ductal carcinomas.50,51 the context of pathogenic mutations in multiple genes, including There are a number of less common types of breast carcinoma, ATM, BRCA1/2, CDH1, CHEK2, NBN, PALB2, PTEN, STK11, and including mucinous, tubular, and papillary, which are generally hormone TP53.33 Testing should be followed by posttest counseling including receptor positive.52 Tubular cancers often resemble normal mammary a discussion of the known and unknown data regarding cancer risks ducts (Fig. 88.4), whereas mucinous (colloid) carcinoma is characterized and management strategies. microscopically by abundant accumulation of extracellular mucin Cancer of the Breast • CHAPTER 88 1565

Figure 88.2 • Typical infiltrating ductal carcinoma. Irregularly dispersed Figure 88.5 • Invasive mucinous (colloid) carcinoma. Bland nests of tumor glands and cords of tumor cells are set in a desmoplastic stroma. cells float in abundant extracellular mucin.

around tumor cells (Fig. 88.5). In contrast, a frond-forming growth pattern characterizes papillary carcinoma (Fig. 88.6). All three subtypes usually have a more favorable prognosis compared with hormone receptor–positive invasive ductal carcinomas of similar size and nodal status. In fact, according to the NCCN guidelines, adjuvant endocrine therapy is not strongly recommended for node-negative favorable- histology tumors measuring up to 3 centimeters in size, whereas treatment is recommended for invasive ductal or lobular carcinomas that are larger than 1 cm. Although medullary carcinoma is typically ER, PR, and HER2 negative, which generally portends a poor prognosis, when it meets all the histologic criteria defined hereafter it too has a more favorable prognosis. The required histopathologic features include a well- circumscribed border, intense reaction with lymphocytes and plasma cells, poorly differentiated nuclei, a syncytial growth pattern, and little or no intraductal carcinoma (Fig. 88.7). On the other hand, atypical medullary tumors, which do not meet all of these criteria, do not have the same excellent outcome, and care must be taken to avoid an Figure 88.3 • Invasive lobular carcinoma. The tumor cells are small and incorrect diagnosis. form linear, single files. Adenoid cystic carcinoma and apocrine carcinoma are other rare subtypes that are typically ER, PR, and HER2 negative. In addition, apocrine carcinomas often express androgen receptor. Despite these characteristics, these tumors also typically have more indolent behavior. In contrast, some uncommon histologic types are associated with a poor prognosis and poor response to standard treatment. For example, metaplastic carcinomas generally have differentiation of neoplastic epithelial cells into other phenotypes, including squamous, spindle cell, or mesenchymal. Although they are less likely to have nodal involvement at the time of diagnosis, relapse rates are high. BRCA-Associated Breast Cancers BRCA1- and BRCA2-associated tumors have typical pathologic features. BRCA1-associated breast cancers are usually of the basal-like subtype and are more likely to be negative for hormone receptors and HER2 overexpression. Conversely, BRCA2-associated breast cancers have more variable pathologic features and are often phenotypically similar to sporadic cancers, with a predominance of hormone receptor–positive tumors.53,54 Once cancer has been diagnosed, standard therapy is dictated by similar prognostic and predictive features to sporadic breast cancer, and no specific regimens are indicated on the basis of genetic predisposition at this time. The involvement of BRCA proteins in Figure 88.4 • Infiltrating tubular carcinoma. The invasive glands are irregularly dispersed, lack a second myoepithelial cell layer, and are set in a DNA repair by homologous recombination suggests that these tumors desmoplastic stroma. Tubular carcinoma resembles normal breast ducts. Two would be particularly sensitive to chemotherapies that induce DNA normal ducts, visible in the center, appear darker than the tubular carcinoma interstrand cross-links such as platinum salts, and there are clinical and have two cell layers. data supportive of this hypothesis.55–58 Furthermore, data suggest that 1566 Part III: Specific Malignancies

AB

Figure 88.6 • (A) Invasive carcinoma arising in papillary ductal carcinoma in situ (invasive papillary carcinoma). The in situ component shown on the right has a frondlike appearance, and the invasive carcinoma component on the left consists of irregularly shaped small glands. (B) Invasive micropapillary carcinoma of the breast. The tumor resembles ovarian serous carcinoma, and small clusters of invasive tumor cells are separated from the stroma by spaces. A psammoma body is visible.

distended by proliferating epithelial cells (Fig. 88.8). LCIS cells have a fairly uniform pattern of clear cytoplasm containing rounded, bland nuclei. Intercellular spaces are preserved, and clear vacuoles within the cytoplasm may displace the nucleus. This often is referred to as classic LCIS. In contrast, pleomorphic LCIS is a morphologic variant that is more aggressive. Pleomorphic LCIS elicits a similar pattern of infiltrative growth as invasive lobular carcinoma, yet neoplastic cells have marked nuclear atypia and pleomorphism. The terminal ductal lobular unit has been proposed as the site of origin of most breast cancer, including DCIS (Fig. 88.9). In contrast to LCIS, which tends to be multicentric, DCIS is generally confined to one branching ductal system in the breast. DCIS, by definition, has intact basement membrane on light microscopy. Unlike pure DCIS, when microinvasion is present, type IV collagen and lamina are lost from the basement membrane in association with loss of membrane continuity. Such observations lend weight to the argument that DCIS is a precursor of invasive ductal cancer. Not all cases of DCIS progress Figure 88.7 • Medullary carcinoma characterized by a well-defined border, to invasive cancer. Identifying the predictors of this process is a major an intense lymphoplasmacytic reaction, and pleomorphic tumor cells with challenge in understanding the biology of this disease. vesicular chromatin. The pathologic classification of DCIS previously was based primarily on the architectural patterns of DCIS as seen microscopically, including solid, cribriform (Fig. 88.10), comedo, micropapillary, and papillary BRCA-associated breast cancers may have higher 21-gene recurrence variants. The features of DCIS that are currently routinely used to scores (RSs), potentially consistent with chemosensitivity of these classify lesions as low or high risk for clinical care include nuclear tumors.59 Finally, poly (ADP-ribose) polymerase 1 (PARP1) is involved grade, size, presence of comedo or necrosis, and, to a lesser extent, in the base excision repair pathway; when this repair pathway is the architectural patterns.61 pharmacologically inhibited in the setting of BRCA deficiency, synthetic lethality may result in cytotoxicity. The sensitivity of BRCA-associated Estrogen and Progesterone Receptors ovarian cancer to the PARP inhibitor olaparib is demonstrative of Estrogen plays a key role in the development of both normal breast this principle60; studies in BRCA-associated breast cancer have not yet epithelium and breast cancer, and the modulation of estrogen concentra- demonstrated the same magnitude or consistency of response to date. tions and ER signaling are key therapeutic modalities for the majority of breast cancers. Estrogens interact with mammary epithelial cells Noninvasive Breast Carcinomas via specific ERs that function as nuclear transcription factors.62 The Much evidence supports the view that the development of malignancy two known receptors, ERα and ERβ, are encoded by different genes is a multistep process and that invasive breast cancer has a preinvasive and share an overall sequence homology of approximately 30%, phase. During the carcinoma in situ (CIS) phase, normal epithelial although the homology is higher in the DNA- and hormone-binding cells undergo enough genetic alterations to result in malignant regions. ERα is the receptor most closely associated with breast cancer; transformation. Transformed epithelial cells proliferate and pile up it is expressed mainly in the breast, uterus, and ovary. ERβ is more within lobules or ducts but lack the additional required genetic alteration widely expressed, and its relationship to breast cancer is less clear.63 that enables cells to penetrate the investing basement membrane. Two Binding of estrogen to the receptor results in a conformational change, types of noninvasive breast cancers have been described: lobular displacement of heat shock proteins, and homodimerization. This carcinoma in situ (LCIS) and ductal carcinoma in situ (DCIS). homodimer then binds to the estrogen-responsive elements of the The original description of LCIS characterized the lesion as a target genes and recruits coactivators or corepressors that further lobular unit with a cluster of ductules or acini filled, distorted, and influence transcription.64 This participation in a multimeric signaling Cancer of the Breast • CHAPTER 88 1567

AB

Figure 88.8 • (A) Lobular carcinoma in situ. The lobular unit (see Fig. 88.9) is distended and distorted by proliferating cells. The cells are uniform and round, with bland nuclei. (B) The same specimen at a higher magnification.

Normal anatomy Subcutaneous Normal anatomy adipose tissue Terminal duct lobular unit Subsegmental duct Ductule Lobule Lobule: groupings composed Segmental of a cluster duct Extra- Intra- of acini lobular lobular (terminal Lactiferous ductules) sinus

Lactiferous duct

Nipple surface

Paget disease

Nipple adenoma Fibroadenomas and Duct ectasia pure cystic diseases

Most single Epitheliosis and solitary most cancers papillomas Traumatic Associated pathology fat necrosis Associated pathology

Figure 88.9 • Anatomy of the breast, showing the organization of the elements of the terminal duct lobular unit and their relationship to specific pathologic abnormalities. (From Hayes D. Breast cancer. In: Skarin AT, ed. Atlas of Diagnostic Oncology. Philadelphia: JB Lippincott; 1991; l64.) process likely contributes to the differential biologic effects that estrogen between ER activity and receptor tyrosine kinase (RTK) signal transduc- and estrogen-related compounds such as the selective estrogen receptor tion pathways. Approximately 70% of breast cancers express ERα modulator (SERM) tamoxifen can have in different target organs. ER (i.e., they are ER positive); these tumors tend to grow more slowly also has a minor ligand-independent transcriptional activating function, and appear better differentiated than ER-negative tumors. Moreover, which is modified by phosphorylation and can facilitate “cross talk” antiestrogen therapy is effective in preventing development of breast 1568 Part III: Specific Malignancies

Clinically, breast cancers marked by HER2 amplification have a more aggressive clinical course and decreased survival as compared with HER2-nonamplified tumors.72 HER2-amplified tumors can be associated with intracranial metastases.73 The identification ofHER2 as an oncogene in breast cancer facilitated the development of therapies directed against it. The development and incorporation of HER2- targeted therapies into the adjuvant and metastatic settings have altered the natural history of HER2-amplified breast cancer dramatically. Trastuzumab is a HER2-targeted humanized monoclonal antibody that forms the cornerstone of HER2 targeted treatment, with proposed mechanisms of action including downregulation of HER2 dimerization and growth factor signaling cascades and induction of antibody- dependent cytotoxicity.74 Trastuzumab and other anti-HER2–directed therapies for treatment of both early-stage and metastatic breast cancer are described in more detail later. Although the advent of HER2-targeted therapy has dramatically improved survival for patients with HER2-amplified breast cancer, de novo and acquired resistance to these treatments contributes to mortality. Studies aimed at elucidating mechanisms of resistance are providing further insight into the pathophysiology of HER2-positive breast cancer; two proteins, HER3 and PI3K, have emerged as key Figure 88.10 • Cribriform ductal carcinoma in situ, which accounts for most of the ductal carcinoma in situ detected. It is characterized by intercon- mediators of this resistance, including mutational activation of the 75,76 necting strands of hyperchromatic cells. Few necrotic cells are seen, and arches latter. HER3 can promote oncogenic PI3K signaling, even in of connecting strands appear rigid. the presence of HER2 inhibition with trastuzumab. Furthermore, the CLEOPATRA phase III trial demonstrated that patients whose tumors harbored somatic activating mutations in the PIK3CA gene had shorter survival than those with wild-type disease.77 Data in the preoperative systemic therapy setting similarly identified an association cancer, reducing the likelihood of recurrence in the adjuvant setting, between PIK3CA mutations and decreased pathologic complete response and prolonging survival once metastases have developed. (pCR).78 However, these findings, while intriguing, do not yet have PR, itself encoded by an estrogen-regulated gene, gives rise to two sufficient clinical usefulness to affect selection of therapy for individual distinct isoforms, PRA and PRB, by alternative splicing. PRB appears patients.79 Increased levels of insulin-like growth factor–1 receptor to be more specific to breast cancer. Interesting to note, a polymorphism (IGF1R) may bypass HER2 blockade by trastuzumab and allow growth in the PR promoter (+331 G/A) that increases transcription of PRB factor activation of AKT,80 and a similar outcome may result from is associated with an elevated risk of endometrial and breast cancer.65 PTEN loss. PR is variably expressed in ER-positive tumors, and this variability In addition to alterations that may circumvent HER2 signaling, has prognostic relevance. ER-positive/PR-negative tumors occur more changes in HER2 itself or effects of genes located nearby on chromo- commonly in women older than 50 years, tend to be more aneuploid, some 17 may also influence response to therapy. A truncated form of and present as larger tumors with more frequent nodal involvement the HER2 protein itself that lacks the extracellular domain but retains than ER- and PR-positive tumors. Furthermore, ER- and PR-positive the kinase activity has been described to correlate with trastuzumab tumors may be more likely to respond to antiestrogen therapy than resistance.81 Activating mutations in the HER2 gene in non-HER2- ER-positive/PR-negative tumors, although more recent data suggest amplified breast cancer have been identified that may confer sensitivity this likely reflects the better prognosis of the dual positive tumors as to anti-HER2 directed therapy.82,83 Finally, the amplicon containing opposed to increased responsiveness to tamoxifen.66,67 HER2 frequently contains additional genes that influence therapeutic efficacy. Although initially it was thought that HER2-positive disease ERBB2 (HER2) is more sensitive to anthracyclines, recent data suggest that this sensitiv- Human epidermal growth factor receptor 2 (HER2) is overexpressed ity is conferred by the topoisomerase II gene (TOP2A) when it is in approximately 20% of breast cancers as a result of amplification present in the amplicon.84 Other cellular mechanisms of anti-HER2 of the HER2/neu gene on chromosome 17q. HER2 functions as therapy resistance are under investigation.85,86 a transmembrane RTK, activated on dimerization with another member of the epidermal growth factor family of receptors, includ- PI3K and PTEN ing EGFR (ERBB1), ERBB3 (HER3), and ERBB4 (HER4). The The PI3K pathway includes the PI3K holoenzyme and the downstream dimerization domain of a partner receptor, such as EGFR or HER3, effector kinases AKT and mTOR; this pathway is critical for cancer is exposed on binding of a ligand such as heregulin68 or, alterna- cell growth, proliferation, and survival.87 This pathway is aberrantly tively, through ligand-independent mechanisms.69 Of note, HER2 activated through various mechanisms including activation of upstream itself does not have ligand-binding capacity. The HER2-HER3 RTKs including HER2, as described earlier, and activating mutations heterodimer is the most potent of these dimer pairs, based on strength in pathway components such as PIK3CA kinase domain mutations of interaction and downstream signaling.70 The oncogenic effects and loss of functional regulatory components such as PTEN. Activating of this dimer pair are predominantly exerted through activation of mutations in the PI3K catalytic subunit are common in breast cancer, intracellular signaling through the phosphatidylinositol 3-kinase particularly in ER-positive, HER2-nonamplified tumors, with an (PI3K)-Akt-mTOR pathway, which in turn promotes proliferation incidence in these tumors of approximately 30% to 40%.88 and cell survival. In contrast to EGFR, in which amplifications in PI3K pathway signaling is initiated by membrane-bound RTKs lung cancers frequently harbor activating mutations in the kinase that undergo activation, for example, by dimerization. These activated domain, such activating mutations have not been found frequently RTKs bind to the regulatory subunit of PI3K, p85. This binding in HER2 in breast cancer according to the Sanger Centre’s COSMIC relieves inhibitory interactions and allows activation of the kinase database,71 and HER2 oncogenic activity is exerted primarily through subunit of PI3K, called p110, which is encoded for by the PIK3CA amplification. gene. Activated PI3K phosphorylates PIP2 to the second messenger Cancer of the Breast • CHAPTER 88 1569

PIP3, which then ultimately results in activation of AKT and mTOR. Breast Cancer Genome Ultimately, cell survival and proliferation result. The tumor suppressor phosphatase and tensin homolog (PTEN) counteracts PI3K activity The elucidation of genetic aberrations in cancer genomes, including by dephosphorylating PIP3 to PIP2. mutations, deletions, and amplifications, has led to a better understand- Germline PTEN mutations cause a hereditary cancer predisposi- ing of cancer pathophysiology. In addition, this knowledge is assisting tion syndrome known as Cowden syndrome, characterized by a high with rational drug development by identifying potential targets. With incidence of breast, uterine, thyroid, and skin neoplasms.89 PTEN is the determination of the DNA sequence of the normal human genome, inactivated in a wide variety of human tumors as a result of either efforts have shifted toward a more comprehensive analysis of cancer mutation or, more commonly, epigenetic silencing through meth- genomes. The first report of a comprehensive sequence analysis of ylation, including in breast cancer.90 PTEN loss is associated with 21,000 of the best studied and annotated human genes was published genetic instability, and primary breast tumors that lack PTEN have that included data from 11 breast cancer samples.110 This screen increased aneuploidy.91 In addition, PTEN loss has been associated rediscovered cancer genes known to be associated with breast cancer with a decreased likelihood of response to anti-HER2 therapy with (e.g., p53 and BRCA1) in addition to genes not previously characterized trastuzumab.92 Data elucidating the relationship between presence as associated with cancer. An individual breast cancer was shown to of PIK3CA mutations and PTEN status show that they are inversely harbor an average of about 90 mutant genes, although only a minority correlated. are likely to be driver mutations. In addition, although the functionally The observed frequency of PI3K activation in breast cancer has led important mutations often differed among tumors, when the genes to clinical investigation of inhibitors of PI3K in the context of hormone were analyzed in functional groups, mutations in signal transduction receptor–positive93,94 and HER2-amplified disease (NCT02167854). pathway and transcription factor genes were found in nearly all breast These agents have demonstrated some efficacy but are also associated tumor samples. As detailed subsequently, rapid advancements in with substantial toxicity including rash, diarrhea, hyperglycemia, and sequencing technology and bioinformatics have since greatly expanded depression. Their role in breast cancer remains to be elucidated. the information available. TP53 Molecular Profiling in Breast Cancer The tumor suppressor TP53 (p53), also termed the “guardian of the Genome-wide RNA transcriptional profiling in combination with genome,” is the most frequently mutated gene in human cancer.88,95 novel bioinformatic approaches has led to the development of molecular It plays a central role in sensing genotoxic and nongenotoxic stresses classifications based on gene signatures comprising the quantitative and transducing an antiproliferative effect (cell cycle arrest or apoptosis) expression of thousands of genes. Tumors can be classified either in response. It is activated and regulated by posttranslational modifica- according to transcripts that tend to segregate together according to tions to the N-terminal region, including phosphorylation and underlying biologic differences (unsupervised clustering) or sorted ubiquitination. In addition, by binding as a tetramer to specific DNA according to a given end point such as prognosis or response to therapy sequences via a central DNA-binding core region, it exerts its primary (supervised clustering). biologic function by modulating the transcription of dozens of genes.96 Initial studies with array-based expression profiling showed the In transgenic mice, loss of p53 is associated with multiple spontaneous ability of the technology to classify breast cancer according to five tumors, although not of the mammary glands.97 However, p53 loss gene clusters: luminal subtype A, luminal subtype B, HER2-positive, accelerates the appearance of mammary tumors in murine mammary basal, and normal breast–like subtypes (Fig. 88.11).111 The ER-positive tissue that also overexpresses MYC, HER2, IGF1, and/or WNT1. group was characterized by high expression of many genes expressed These genetic studies are consistent with a role for p53 loss late in by breast luminal cells, and the ER-negative group showed gene tumor development.98 Indeed, the transgenic restoration of p53 function expression characteristic of basal epithelial cells. However, a third leads to tumor regressions.99 group showed genes related to HER2 overexpression, suggesting that Mutant p53 accumulates in the nucleus of neoplastic cells. Thus this molecular characteristic may have equal or greater weight than initial studies that described only a weak association between aberrant ER expression in subclassifying breast cancers. Finally, the normal p53 and adverse prognosis in breast cancer relied on immunohisto- breast-like group of breast cancers clustered with normal breast epi- chemistry (IHC) detection; more recent analyses based on mutation thelium. These distinct subtypes of breast tumors show distinctive detection have confirmed a strong association.100 Mutations in exons molecular signatures and appear to represent diverse biologic entities 5 through 8 are more common in ductal and medullary tumors, associated with distinct clinical outcomes, and the comparison of tumors with an aggressive phenotype (high grade, large size, node- several independently developed gene signatures appears to show similar positive cases, and low hormone receptor content), and in women prognostic information, suggesting the existence of a common set of younger than 60 years.101 Furthermore, the presence of a mutation biologic phenotypes.111–113 Luminal A–type cancers have the most conferred an overall 2.27-fold increased RR of breast cancer–specific favorable long-term survival, whereas HER2-positive and basal-like mortality, independent of other known prognostic markers (e.g., cancers may be more sensitive to chemotherapy but have the worst tumor size, node status, and estrogen receptor and PR expression). overall prognosis. These patterns of gene expression appear to provide Finally, although not all mutations confer the same biologic proper- more specific information than identification of a single gene with a ties (e.g., missense versus nonmissense), all have similar prognostic specific effect. Thus breast cancer is not a single disease with hetero- usefulness. geneous ER and HER2 expression but appears to comprise three to Breast cancers in individuals with germline p53 mutations are five molecularly and clinically distinct subtypes. frequently ER and HER2 positive.102,103 Adjuvant radiation therapy A similar approach can be taken by examining DNA copy number is generally avoided in breast cancer patients with Li-Fraumeni syndrome abnormalities, including amplifications and deletions, across the owing to the risk of radiation-induced sarcomas104; mastectomy, often genome.114 Mapping these abnormalities against both the biologic bilateral mastectomy, is more common. As noted earlier, many tumors subtypes described earlier and clinical outcome data demonstrates the harbor sporadic p53 mutations; p53 is an obvious target for cancer potential to identify high-level DNA amplification that, similar to therapy, despite the apparent intractability of a loss-of-function geno- HER2, may be useful in identifying therapeutic targets.115 Emerging type.105 Multiple approaches are being studied to overcome this limita- knowledge now aims to integrate patterns of gene expression, methyla- tion, including small molecules to restore p53 function106 or inhibit tion, copy number variation, and mutation. This approach, using The its interaction with MDM2,107 and adenoviral-mediated gene deliv- Cancer Genome Atlas (TCGA) Network, holds significant potential95 ery. 108,109 There are now multiple compounds in clinical trials, although but will require independent validation using data sets that include no results are yet available in breast cancer. treatment and outcome. 1570 Part III: Specific Malignancies

A

Basal-like ERBB2+ Normal Luminal Luminal Luminal breast-like subtype C subtype B subtype A

B

C

D

E

F

G

Figure 88.11 • Hierarchical clustering of 115 tumor tissues and 7 nonmalignant tissues using the “intrinsic” gene set. (A) Scaled-down representation of the entire cluster of 534 genes and 122 tissue samples based on similarities in gene expression. (B) Experimental dendrogram showing the clustering of the tumors into five subgroups. Branches corresponding to tumors with low correlation to any subtype are shown in gray. (C) Gene cluster showing the ERBB2 oncogene and other coexpressed genes. (D) Gene cluster associated with luminal subtype B. (E) Gene cluster associated with the basal subtype. (F) A gene cluster relevant for the normal breastlike group. (G) Cluster of genes including the estrogen receptor (ESR1) highly expressed in luminal subtype A tumors. Scale bar represents fold change for any given gene relative to the median level of expression across all samples. (From Sorlie TR, Tibshirani R, Parker J, et al. Repeated observation of breast tumor subtypes in independent gene expression data sets. Proc Natl Acad Sci U S A. 2003;100:8418–8423.)

In the meantime, although these classifications are robust across non–protein-coding DNA of transcription factor association, transcrip- a population, they have not yet been shown to be clinically applicable tion, chromatin structure, and histone modification with the goal of for individual treatment decision making. Furthermore, because much determining biochemical functions for many areas of the genome, of the molecular heterogeneity overlaps with conventional histopathol- including those mutated in cancer.117 As technology advances and ogy and is captured by determination of ER, PR, and HER2 status becomes more inexpensive, even more comprehensive analysis of breast and tumor grade, use of these molecular tools adds little to current cancer genomes will likely continue to increase our understanding of standard pathologic information.116 both breast cancer and how to use the information for both treatment and prevention. Comprehensive Genomic Analysis More recently, more comprehensive approaches have been undertaken Breast Cancer Stem Cells to examine the molecular profiles of breast cancer. A comprehensive and integrated analysis of 825 primary breast cancers using genomic Cancer cells have two somewhat contradictory traits. Although they DNA copy number arrays, DNA methylation, exome sequencing, originate from a single clone, they often display marked genetic, messenger RNA arrays, microRNA sequencing, and reverse-phase biologic, and morphologic heterogeneity. Historically, the prevailing protein arrays was conducted by TCGA and confirmed the existence hypothesis was that genomic instability contributed to continuing of four main subtypes of breast cancer.95 In addition, it provided a tumor evolution and emergence of different and ultimately more wealth of data regarding genetic and epigenetic information for potential virulent subclones, leading to disease progression and resistance therapeutic targeting.95 Complementary to TCGA, the ENCODE to therapy. More recently, an alternative explanation was provided project (Encyclopedia of DNA Elements) mapped regions in by groundbreaking work elucidating the “cancer stem cell” model Cancer of the Breast • CHAPTER 88 1571

division (such as Wnt and Notch pathways) might provide novel targets for therapy.120 A phase I trial of a Notch-targeting γ-secretase inhibitor in combination with cytotoxic chemotherapy in breast cancer Apoptosis identified preliminary evidence of efficacy.121 Additional trials of drugs targeting the Notch pathway and others involved in cancer stem cells are underway. Breast Cancer Detection in the Circulation Primary tumor Disseminated Tumor Cells

A Metastasis More than 90% of patients diagnosed with breast cancer have no evidence of disease outside of the breast and ipsilateral axillary lymph nodes. However, approximately 30% of patients with primary operable breast cancer have disseminated tumor cells (DTCs) that can be detected in bone marrow122 at the time of diagnosis.123 In a follow-up study, Apoptosis 15.5% of patients had detectable DTCs 3 years after diagnosis, a finding that was associated with increased risk of relapse and death.124 Despite these data, however, assessment of DTCs has not been incorporated into routine practice, in part because of the need for a relatively invasive procedure. Primary tumor Liquid Biopsies Tumor markers are proteins, such as those derived from MUC1 (e.g., CA15-3, CA27.29) and carcinoembryonic antigen, that are shed from tumors and identified in the circulation.125 However, their usefulness B Metastasis is limited to monitoring of response to therapy or progression of

Figure 88.12 • Impact of the cancer stem cell (CSC) model on the origin disease. More recently there have been considerable methodologic and biology of metastases. (A) According to standard cancer models, tumors advances in the assessment of circulating cancer biomarkers, including are composed of heterogeneous mixtures of independent subclones, originated circulating tumor cells (CTC) and circulating nucleic acids, which by divergent genetic mutations; different subclones are endowed with different may play a role in diagnosis, monitoring, and treatment of disease. functional properties, and only selected clones (red cells) can migrate and form It is possible to measure CTCs using an automated immunobead metastases. The metastasis is predicted to be a homogeneous monoclonal enrichment followed by pancytokeratin staining.126,127 With this expansion of an individual subclone, which in turn can accumulate further approach, presence of five or more CTCs per 7.5 mL of blood is mutations (red, green, and orange cells) and diverge even further from the prognostic, with significantly worse overall survival (OS; 8.2 months primary tumor. Overall, the model predicts that primary tumors and correspond- 128 ing metastases are substantially different. (B) The CSC model assumes that versus 18 months) compared with those patients with fewer CTCs. intratumor heterogeneity is caused mainly by cell differentiation, and that The clinical usefulness of CTCs for treatment decision making in only CSCs (red cells) can migrate and form overt metastases, while differentiated metastatic breast cancer was prospectively tested in a randomized cells (blue and yellow cells) undergo apoptosis. In the CSC model, metastatic phase III trial; however, use of the assay was not shown to result in cancer tissues undergo differentiation programs that closely resemble those improved disease outcomes and therefore it is unclear how best to observed in the corresponding primary tissues. Experimental evidence based incorporate it into routine clinical care.129 The immunobead method on gene expression microarrays tends to support the CSC model for human lacks sensitivity owing to the use of epithelial cell adhesion molecule epithelial tumors, such as breast and colon cancer. The two hypotheses are (EpCAM)–based detection, because less differentiated cells often lack not mutually exclusive, and elements of both are probably true. (From Dalerba EpCAM expression and will avoid detection, so newer methodologies P, Cho RW, Clarke MF. Cancer stem cells: models and concepts. Annu Rev Med. 2007;58:267–284.) including isolation of CTCs based on size or other properties are being developed.127,130 Furthermore, the ability to isolate CTCs permits single-cell analysis, including assessment of molecular markers such as ER expression and HER2 amplification in real time. (Fig. 88.12).118 Initially demonstrated in hematologic reconstitution Technologic advances have also resulted in the ability to detect and malignancies, cancer stem cells, although constituting only a small nucleic acids, including circulating tumor DNA (ctDNA) and minority of the tumor mass, retain the capacity for self-renewal and microRNA (miRNA), that are shed from dying tumor cells into the differentiation. Differentiated cells, although forming the bulk of the blood and may be more sensitive and specific than standard biomark- tumor, are themselves incapable of endless replication. Given that stem ers.130,131 Because analysis of circulating nucleic acids has the potential cells are long-lived and slow to replicate, this would explain both the to identify emerging mutations that could lead to altered treatment accumulation of genetic mutations over time and the resistance to options, their use has potential to improve disease outcomes. ctDNA therapy targeting dividing cells. It would also explain the frequently and can be specifically identified by the presence of tumor-specific observed concordance between primary and metastatic cells. mutations, including in more than 90% of patients with metastatic For a long time this model for epithelial cancers has been theoretical, breast cancer.132,133 Serial assessment of ctDNA via detection of these because it was technically difficult to identify these cells. However, mutations may be useful for monitoring of patients for disease recur- such cells have been isolated from both normal mammary glands and rence or progression134; relative levels have been shown to correlate breast cancers, based on the presence and absence of specific surface with tumor burden.133,135,136 Similar findings have been demonstrated markers. The ability to reconstitute human breast cancers in with miRNAs.137 immunocompromised mice seems to reside within a minority CD44+, More recently, it has become apparent that in addition to detec- CD24−/low subclone. Notably, tumors formed by these cells gave rise tion of cancer, assessment of circulating biomarkers may be useful to mixed populations of epithelial cells recapitulating the morphologic for monitoring changes in tumors that could affect treatment. For characteristics of the parent tumor.119 These findings have several example, multiple groups have identified activating mutations in the important implications for breast cancer therapy. As a first step, the ligand-binding domain of ESR1, the gene that encodes ERα, which molecular and biochemical pathways that control cancer stem cell are present in up to 30% of endocrine therapy–resistant patients 1572 Part III: Specific Malignancies

but which are only rarely detected in primary tumor specimens.138,139 modification of lifestyle factors, such as diet, alcohol consumption, Studies have also demonstrated the use of massively parallel sequencing and exercise, may be discussed. At this time, behavioral modifications of ctDNA for examining intratumor heterogeneity, which could also do not have demonstrated efficacy as primary prevention for breast potentially be used to identify targetable mutations.140 Companies cancer in high-risk individuals. are now capitalizing on the new technology and providing assays 141 for monitoring patients, although the clinical usefulness of these Chemoprevention commercial assays is currently still in question. As the sensitivity of detection of these circulating biomarkers improves with advances in A significant 49% reduction in risk of invasive breast cancer with technology, it is likely that liquid biopsies will become routinely used tamoxifen compared with placebo was seen in the National Surgical for monitoring patients with metastatic breast cancer. However, use Adjuvant Breast and Bowel Project (NSABP) P-1 study145; similar of this technology for diagnosis or detection of disease recurrence findings were noted in the International Breast Cancer Intervention remains questionable. Study (IBIS-I).146 Subsequently, two SERMs, tamoxifen and raloxifene, were directly compared in the NSABP STAR trial; long-term follow-up PREVENTION AND EARLY DETECTION demonstrated that although raloxifene was less toxic than tamoxifen, it was slightly less effective in decreasing invasive breast cancer.147 An individual can be deemed at high risk for breast cancer based on Based on these findings, multiple groups have recommended that a calculation from a breast cancer risk assessment model, a genetic SERMs be considered to reduce the risk of breast cancer in women test result, or other family or personal history considerations that with a 5-year projected risk of developing breast cancer of 1.66% or result in an estimation of an individual’s lifetime risk of breast cancer greater.148 More recently, the aromatase inhibitors (AIs) exemestane that is above population risk. There is no universally accepted definition and anastrozole have been shown to decrease the risk of invasive breast of “high risk.” Whatever criteria an individual physician and patient cancer in the randomized placebo-controlled, double-blind MAP.3 use to define high risk, four possible actions may be taken, some of and IBIS-II clinical trials, respectively.36,37 Longer follow-up is required which can be used simultaneously: (1) enhanced surveillance; (2) to confirm these data, including possible long-term side effects of the behavioral modification; (3) chemopreventive strategies; and (4) medications. prophylactic mastectomy and/or oophorectomy. Tamoxifen has been shown to reduce the risk of contralateral breast cancer by 50% to 60% in BRCA1 and BRCA2 carriers, respectively, 149 Increased Surveillance even though BRCA1-associated cancers are generally ER negative. However, two European chemoprevention studies performed at the As discussed in the section on mammographic screening, there is Royal Marsden Hospital150 and by the Italian Tamoxifen Prevention significant debate about the age at which to begin mammographic Study Group,151 respectively, did not show a decrease in the incidence screening in average-risk women. Nonetheless, ample evidence supports of breast cancer in women from high-risk families when tamoxifen increased surveillance in women at high risk for breast cancer, and was used. strategies are being developed.142 Carriers of germline BRCA mutations, For all of these agents, it is not known whether they reduce the for example, may begin mammographic surveillance at age 30. The incidence of breast cancer by preventing the formation of cancer or American Cancer Society (ACS) and the American College of Radiology by treating small, clinically occult cancers. In practice, the potential (ACR) appropriateness criteria recommend consideration of annual chemopreventive benefits of these agents, along with consideration breast magnetic resonance imaging (MRI) in addition to mammography of an individual’s baseline risk, must be balanced in the context of for individuals who have a 20% or greater lifetime risk of breast their known toxicity profiles.152 Indeed, patients and providers alike cancer, including those with germline BRCA mutations, those who are often cautious about chemoprevention in light of potential have a family history suggestive of a familial risk of breast cancer but adverse events. in whom genetic testing has not been performed, those with a history of atypia, and those with a personal history of breast cancer or chest Prophylactic Mastectomy and/or Oophorectomy irradiation between ages 10 and 30. The timing of such imaging is not evidence based. Some providers recommend alternating mam- Prophylactic mastectomy is associated with a reduction in breast cancer mography and MRI every 6 months, whereas others recommend risk of approximately 90%, which is a substantial absolute reduction annual simultaneous imaging for convenience. Recommendations in patients with either a strong family history of breast and/or ovarian regarding age at which breast surveillance should be initiated may be cancer153 or a genetic predisposition to breast cancer.154 Additional individualized based on earliest age of onset of breast cancer in the data show that contralateral prophylactic mastectomy (CPM) can family. In addition, other genetic mutations and syndromes also increase reduce the risk of breast cancer in patients with a previous diagnosis the risk for breast cancer and require initiation of screening at an of unilateral disease.155 A high prevalence of premalignant lesions, earlier age according to both the ACS and NCCN guidelines, although including atypical hyperplasia, is seen in prophylactically removed the recommended age of initiation of screening MRI varies by gene.33,143 breasts from women who are at hereditary risk for breast cancer.156 Finally, although controversy remains about the role of breast Statistical analysis has shown that, depending on the assumed self-examination in the general population, experts in the surveillance penetrance of the BRCA mutation, compared with surveillance alone, of high-risk women still recommend breast awareness for women with 30-year-old patients with early-stage breast cancer who have BRCA BRCA mutations144 in addition to clinical breast examinations. For mutations may gain 2.9 to 5.3 years from prophylactic bilateral males with pathogenic germline mutations, it is important to recognize mastectomy and 0.6 to 2.1 years from CPM.157 A survival analysis that the absolute risk of breast cancer remains relatively low (≤10%) suggested that prophylactic mastectomy at age 25 years and oopho- and that evidence for screening with use of any modality is limited. rectomy at age 40 years maximized survival at 70 years of age; it is important to note, however, that substituting enhanced breast cancer Behavior Modification screening (with MRI and mammography) for mastectomy resulted in similar survival.158 The gain in quality of life in these high-risk Multidisciplinary centers that provide counseling for women who are women who chose to undergo prophylactic mastectomy is controver- at substantial risk for breast cancer are well established throughout sial.159 Given the impact on quality of life, the potential medical the United States. These multidisciplinary consultations provide recom- repercussions of an operation, including the reconstruction that mendations for a range of surveillance and interventional approaches. sometimes follows, and the presence of an effective, alternate breast In addition to the available surgical and medical preventive strategies, cancer risk mitigation strategy (enhanced surveillance), the choice of Cancer of the Breast • CHAPTER 88 1573 prophylactic mastectomy is personal and certainly not mandated by did undergo mammography.164 There continues to be active and lively providers counseling high-risk women. However, if a patient opts for debate about the usefulness of breast cancer screening, which is beyond enhanced surveillance, the possibility of diagnosis of an early-stage the scope of this chapter. breast cancer that may require systemic chemotherapy must be clearly Mammography has evolved from film screen mammography to discussed. digital mammography, which is now used in more than 97% of all Although physicians caring for women at increased risk of mammography units because it allows for a lower average dose of breast cancer generally do not mandate prophylactic mastectomy, radiation, improvement in storage and retrieval, easier image sharing women at increased risk for ovarian cancer, for example because of across institutions, and computer-assisted diagnosis.165 Digital breast germline mutations in BRCA1/2, are strongly encouraged to undergo tomosynthesis (DBT), described in detail later, is a new method of risk-reducing salpingo-oophorectomy. Unlike breast MRI and mam- acquiring multiple mammographic projections that mitigates some mography, ovarian cancer screening with transvaginal ultrasound and of the limitations of standard plane mammography. Important to CA125 measurement is not effective at diagnosing ovarian cancer note, mammography and DBT are anatomic approaches to screening early, and ovarian cancer is generally diagnosed at a stage when based on calcifications, architectural distortion, and radiographically curative treatment options are limited. Furthermore, the benefits of apparent masses, in contrast to functional assessment methodologies risk-reducing oophorectomy in BRCA carriers are significant and such as breast MRI. include a reduction in breast and ovarian cancer risk, and reduced breast and ovarian cancer–specific mortality and all-cause mortal- Screening and Early Detection ity.154 Despite the limitations of ovarian cancer screening, for some women with BRCA mutations whose ovaries and fallopian tubes The efficacy of screening for occult cancer depends heavily on the are intact, transvaginal ultrasonography and CA125 screening are following factors: (1) tumor growth rate, (2) the sensitivity of the test recommended. related to tumor volume, and (3) the interval between screens. Growth The age at which risk-reducing bilateral salpingo-oophorectomy rates of breast cancer vary. In determining the efficacy of screening, is performed is not trivial, because the induction of early surgical four biases must be considered: lead-time, length, selection, and menopause may have long-term systemic effects including those on overdiagnosis bias. Lead-time bias is the interval that the diagnosis bone health and cardiac function, in addition to significant effects has been advanced by screening. Length bias concerns the timing of on quality of life. The age at which surgery is recommended is based detection. When screening is infrequent, fast-growing tumors are not on the age-adjusted risks of ovarian cancer and is generally slightly detected as early in their natural history as more slowly growing tumors. earlier for BRCA1 carriers than BRCA2 carriers.160 A limited duration Thus the outcome of cancers detected by screening is better than that of HRT is safe in terms of breast cancer risk for premenopausal women for interval cancers. Women who participate in breast cancer screening with no personal history of breast cancer and with intact breasts, and have been shown to be more health conscious; they also are more may provide substantial benefits in terms of quality of life and overall likely to obtain Pap smears, use seat belts, and not smoke.142 Thus it health. It is generally recommended that these women stop HRT is likely that their outcomes would be better even in the absence of around the age of natural menopause. Important to note, BRCA screening, contributing to selection bias. Overdiagnosis is the detection carriers who undergo risk-reducing salpingo-oophorectomy remain of cancers that would not be fatal for the patient in her lifetime and at risk for developing primary peritoneal cancer, for which no screening that would not be diagnosed without screening. The contributions or risk reduction strategies are delineated. of all of these biases make assessment of benefits of breast cancer screening programs more complicated. CLINICAL MANIFESTATIONS AND Screening guidelines PATIENT EVALUATION Table 88.3 lists the current screening guidelines for mammography Detection of Breast Cancer for the ACS, ACR, NCCN, and the US Preventive Services Task Force (USPSTF).161,166,167 All of the major screening guidelines agree Despite a lifetime probability of developing breast cancer that is that the most lives are saved from breast cancer when screening begins currently estimated at 1 in 8, only 30% of all women have one or at 40 and continues annually. However, some of the guidelines suggest more identifiable risk factors. At present, there is no preventive treatment routine screening starting at a later age (45 or 50) or biennially, based for breast cancer that is widely available to the average population. on different perceptions of the trade-offs between the benefit (mortality The goal of breast cancer screening is early detection that will lead reduction) and the risks of screening (false-positive recalls with psy- to a reduction in mortality and morbidity, with a minimum of false- chologic anxiety and overdiagnosis are the main cited risks in the positive findings and minimal risks. In addition, the screening program current guidelines). should be feasible to provide to the public, acceptable to both patients and physicians, and cost-effective. Because 70% to 75% of women Risks of screening diagnosed with breast cancer have no identifiable risk factors other There are several risks of screening noted in the breast cancer screening than being female, current screening and education programs include guidelines issued by the ACS166 and USPSTF.161,168 One major concern all women aged 40 and older regardless of risk, although there is some about breast cancer screening is the risk of overdiagnosis, which reflects variability in the recommendations from different medical societies. breast cancers that would not have led to symptomatic disease during In addition, as noted earlier, there is supplemental screening available a patient’s lifetime. False-positive findings are an inherent risk of a for women who are at high risk of developing breast cancer. screening program. They have been shown to result in increased Multiple international randomized controlled trials have repeatedly short-term anxiety, although this has not been shown to translate to demonstrated that mammography decreases the mortality from breast long-term anxiety or decrement in health utility.169 Furthermore, a cancer by 15% to 30% in women invited to be screened.161 A number cross-sectional study in the United States demonstrated that women of case-control and cohort observational trials have shown that the are aware of the possibility of a false-positive result and view it as an mortality reduction is even higher, up to 48% in women who actually acceptable consequence, rather than a harm.170 undergo mammographic screening relative to women who do not.162,163 Radiation exposure associated with breast cancer screening is a Furthermore, a failure analysis study looking at women diagnosed concern that some patients express. All of the major societies that with breast cancer in Massachusetts in the 1990s found that 71% of issue breast screening guidelines acknowledge that the current doses breast cancer deaths were in women who did not undergo screening of radiation used in screening mammograms are of negligible risk to mammography, whereas only 29% of the deaths were in women who the screening age patient.161,166,167 The average effective radiation dose 1574 Part III: Specific Malignancies

Several states in the United States have enacted breast density Table 88.3 Suggested Screening Guidelines for notification laws that require patient notification acknowledging and Mammography From Some US offering options for supplemental screening for those with dense breasts. Organizations Mammographic technologies have evolved to address the decreased AMERICAN CANCER SOCIETY limitations of screen-film mammography. A large multicenter trial comparing conventional and digital mammography showed an improve- For women with normal risk, yearly mammograms are recommended ment over conventional film mammography in the detection of breast starting at age 45 and every 2 years starting at age 55 as long as the cancer in young, premenopausal, and perimenopausal women, and patient is in good health and expected to live at least 10 years. in women with dense breasts.172 However, there was no significant AMERICAN COLLEGE OF RADIOLOGY difference in diagnostic accuracy between digital and film mam- mography in the population as a whole or in the other predefined For women with normal risk, yearly mammograms are recommended starting at age 40 for as long as the patient is in good health. subgroups. For women with an average risk of developing breast cancer, supplemental screening with other modalities such as ultrasonography NATIONAL COMPREHENSIVE CANCER NETWORK and breast MRI has not been shown to be cost-effective and has a Between the ages of 25 and 40 years, women at average risk are high rate of false positives, and accordingly may not be offered by recommended to undergo a clinical breast examination every 1–3 many practices because it is not currently endorsed by any major 161,173,174 years; breast awareness is encouraged. Beginning at age 40 years, societal recommendation for routine use. an annual clinical breast examination and mammogram are recommended with the consideration of tomosynthesis; breast Digital breast tomosynthesis awareness is encouraged. To mitigate some of the limitations of mammography, DBT screening US PREVENTIVE SERVICES TASK FORCE is a new method of anatomic imaging that has been introduced. DBT acquires several angular projections that may be evaluated spatially Women who are 50–74 years of age should receive screening across the acquired slices. DBT provides improved discrimination of mammography every 2 years. For women younger than the age of masses and architectural distortion, which may be otherwise obscured 50 years, regular biennial screening mammography should be an by overlapping breast tissue within the imaged slices. Conversely, an individual decision that takes into account patient context, including apparent mass on a single-projection two-dimensional (2D) image values regarding benefits and harms. may simply represent a summation artifact of breast tissue, and use of DBT helps avoid this false-positive finding. Overall, DBT results Modified from Oeffinger KC, Fontham ET, Etzioni R, et al. Breast cancer screening in lower false-positive recall of patients from screening and maintains for women at average risk: 2015 guideline update from the American Cancer Society. 175 JAMA. 2015;314:1599–1614; American College of Radiology practice parameter for the or increases the cancer detection rate. In particular, DBT further performance of screening and diagnostic mammography. https://www.acr.org/~/ increases the sensitivity and decreases the false-positive results from media/ACR/Documents/PGTS/guidelines/Screening_Mammography.pdf?la=en. dense breasts relative to digital mammography, especially in women Accessed January 12, 2017; Siu AL et al. Screening for breast cancer: US Preventive with heterogeneously dense breasts.176 Services Task Force recommendation statement. Ann Intern Med. 2016;164:279–296; and Bevers TB, Ward JH, Arun BK, et al. Breast cancer risk reduction, version 2.2015. J Standard DBT is nearly equivalent in radiation dose to digital Natl Compr Canc Netw. 2015;13:880–915. mammography. However, most practices that have adopted DBT for screening currently acquire both 2D mammograms and DBT scans, which results in increased radiation exposure. It is possible that the increased radiation dose from the additional projections may be offset from annual screening digital mammography is 0.44 mSv (0.56 mSv by use of the newer technology of synthesized mammography, which for screen-film mammography); in contrast, annual background natural is created from the DBT acquired images, thereby eliminating the radiation constitutes 3 mSv.171 Therefore radiation from digital mam- need to acquire a traditional 2D mammogram in addition. mography is equivalent to less than one-fifth of the natural background radiation to which patients are normally exposed, without considering Other methods of screening any radiation from medical sources. As noted earlier, mammography and DBT provide an anatomic approach to screening. To overcome some of the variables that affect the predictive Mammography value of mammography, a number of new technologies have been Screening mammography comprises craniocaudal and mediolateral developed. Extending beyond just anatomic abnormalities, techniques oblique views of each breast. The anatomic abnormalities detected such as MRI, scintimammography and contrast-enhanced spectral with mammograms include radiographically apparent masses, archi- mammography provide a functional approach to screening based on tectural distortion, and calcifications. Adequate compression is essential injected contrast material or radiotracer that is taken up by biologically to even out the breast tissue over the detector to provide more uniform active breast cancer. The aforementioned additional functional exposure. Compression decreases the possibility that glandular areas techniques, in addition to ultrasonography, are available for select will obscure masses or calcification. It also prevents motion, which women who have abnormalities on screening mammograms or who causes loss of fine resolution. Additional views may be required to are at higher risk of developing breast cancer.177 image all of the breast tissue, especially tissue adjacent to the chest Breast magnetic resonance imaging. The combination of angio- wall. Magnification films are valuable for evaluating areas of architectural genesis and associated increased vascular permeability creates leaky distortion and fine calcification. vessels, which, when combined with proteolysis, results in increased Although mammography is the best screening tool for breast cancer, intravenous contrast uptake in areas of biologically active malignancy it detects only 85% to 90% of biopsy-proven cancers in the best case that can be detected with breast MRI. Breast MRI provides excellent scenario. Mammography is not a substitute for tissue sampling and sensitivity (ranging from 75% to 97% for breast MRI alone),178,179 histologic evaluation of a palpable abnormality, nor is it a substitute although its ability to detect microcalcifications is limited. There is for careful physical examination. Different factors affect mammographic mildly decreased specificity with breast MRI, resulting in false-positive sensitivity. Having mammographically dense breasts, which have a findings at biopsy. Breast tissue can normally enhance, which is termed higher ratio of fibroglandular and stromal elements relative to fatty background parenchymal enhancement (BPE). BPE is highest during tissue, decreases the sensitivity of mammography. As patients age and the luteal phase of the menstrual cycle, and thus patients are typically breast tissue is replaced with fat, mammography is more effective for scheduled for imaging outside of that phase. Infrequently, BPE can detecting occult malignancy. be focal or asymmetric rather than diffuse, which can limit evaluation Cancer of the Breast • CHAPTER 88 1575

Figure 88.14 • Ultrasound image of the breast showing the palpable lump to be cystic. Cystic lesions have a characteristic hypoechoic pattern, Figure 88.13 • Ultrasound image confirms the presence of a solid mass. with prominent acoustic shadowing.

or lead to false-positive biopsy results. Unlike with conventional may appear normal on sonographic examination and still contain mammography, dense breasts do not limit the sensitivity. In addition, metastasis. there is no radiation involved with MRI. Breast MRI is not recom- Sonographic evaluation can also be used to image the subareolar mended for routine screening of women at average risk of breast area to evaluate for an intraductal mass, especially in the setting of cancer who are asymptomatic. Recommendations for screening patients nipple discharge. Calcifications are poorly characterized by ultrasound, at high risk of breast cancer with breast MRI are described earlier in although they can appear as an echogenic area with acoustic shadowing this chapter. posteriorly, particularly when large in size. As ultrasound contrast Ultrasonography. Although commonly used in the diagnostic examination is not indicated for use in the United States; there is workflow, whole-breast screening ultrasound examination is not widely limited indication for ultrasound for functional evaluation of masses. endorsed by guidelines and may not be cost-effective, in part because Doppler flow analysis provides limited information on blood flow of an increased false-positive rate. For asymptomatic women with within a mass. However, because malignant masses may demonstrate dense breasts and a negative screening mammogram, some sites supple- absence of detectable color Doppler flow, this is not used as a dif- ment with whole-breast screening ultrasonography because it results ferentiating factor. in a slightly increased incremental cancer detection rate. The addition of ultrasound to mammography enables detection of a median of 3.2 Screening the elderly patient additional cancers (range, 0.4–14.2) per 1000 patients screened, with The life expectancy for women in the United States increased from a biopsy positive predictive value median of 6.8 per 1000 patients 77.4 years in 1980 to 80.4 years in 2002, although it decreased to (range, 3.2–18.4).180 78.8 in 2014.181 Biologic parameters suggest a higher prevalence of On the other hand, ultrasonography is a very effective imaging higher-risk tumors in younger women, with better differentiated tool in the diagnostic setting and is indicated for evaluation of women and hormone receptor–rich, lower-risk tumors generally occurring with either palpable or mammographically detected lesions (Fig. 88.13). in older women. In addition, for a given tumor size, the likelihood Ultrasonography is very accurate (>95%) for diagnosing breast cysts, of nodal involvement is lower in older than in younger women.182 which have well-demarcated, smooth margins with an echo-free center These data in aggregate have led some, including the USPSTF, to (Fig. 88.14) and are usually rounded and thin-walled and produce suggest that it may be appropriate to increase the screening interval distal shadowing. Simple cysts require no further evaluation. Com- to 2 years in older women and to stop screening mammography at plicated cysts contain echogenic debris and are benign, often undergoing age 75 years, although there are insufficient data to suggest an age at inflammatory change. A cystic mass containing a solid component which routine screening can be abandoned. At 75 years of age, the requires biopsy, because the histologic findings can be papilloma, average additional life expectancy for a woman in the United States papillary carcinoma, or invasive ductal carcinoma. is 12 years or more. Multiple studies have demonstrated the benefits The intrinsic architecture of axillary lymph node anatomy may of screening detection of breast cancer in elderly women, because also be evaluated with ultrasound in the setting of breast cancer persistence with screening is associated with diagnosis of earlier-stage staging. Identification of cortical thickening, defined as greater disease and lower breast cancer–related mortality.183–186 Although the than 3 mm, at ultrasound examination is suggestive of lymph node benefits of screening appear to be seen in all age groups, the magnitude involvement in a woman with a known or suspected underlying of the benefit appears to diminish with age and with the severity of breast malignancy. However, the sensitivity of axillary lymph node comorbid illnesses. Indeed, it seems reasonable to forego screening sonography for evaluation of nodal metastasis remains limited. Therefore mammography for women whose life expectancy is less than 5 to sentinel node biopsy remains the gold standard because lymph nodes 10 years.187 1576 Part III: Specific Malignancies

Table 88.4 Breast Imaging Reporting and Data System (BI-RADS) Assessment Categories Category Assessment Likelihood of Malignancy Management 0 Incomplete N/A Recall for additional imaging 1 Negative Essentially 0% Routine screening 2 Benign Essentially 0% Routine screening 3 Probably benign ≤2% Short-term follow-up 4 Suspicious >2% but <95% Biopsy 4A Low suspicion >2% to ≤10% 4B Moderate suspicion >10% to ≤50% 4C High suspicion >50% to <95% 5 Highly suggestive of malignancy ≥95% Biopsy 6 Known biopsy-proven malignancy N/A Treatment

N/A, Not applicable. Modified from Breast Imaging Reporting and Data System. Reston, VA: American College of Radiology; 2013; 135. https://www.acr.org/~/media/ACR/Documents/PDF/ QualitySafety/Resources/BIRADS/01-Mammography/02--BIRADS-Mammography-Reporting.pdf?la=en. Accessed December 22, 2016.

Figure 88.16 • Figure 88.15 • Magnified view showing that a palpable lump is uniform Mammogram showing arterial calcifications in an otherwise in density, lacks microcalcifications, and has sharp, clear borders. Biopsy normal examination. established the lump to be fibroadenoma.

Mammographic Abnormalities The ACR Breast Imaging Reporting and Data System (BI-RADS) standardizes reporting nomenclature and defines specific terms to describe breast imaging findings.188 Based on the imaging find- ings, BI-RADS provides an assessment category (0 through 6) as shown in Table 88.4.189 Features with the highest positive predictive value for carcinoma are spiculated margins, irregular shape, linear calcification, and segmented or linear calcification distribution. In general, the positive predictive value of a BI-RADS category 5 lesion is greater than 80%, whereas that of a category 4 lesion approaches 30% to 40%. Figure 88.17 • Pleomorphic calcifications in an area of extensive ductal carcinoma in situ. Masses BI-RADS defines a mass as three dimensional, convex, space occupying, and seen in two projections. If a probable mass is seen in only one the breast background tissue with a stellate appearance. Usually there view, it is called an asymmetry. If seen on both views but lacks convex is some distortion of adjacent breast stroma. The mammographic borders, it is called a focal asymmetry. Benign masses typically are abnormality that has the highest rate of malignancy is a mass with well defined, with sharp margins, and have little effect on the sur- associated calcification (Figs. 88.16 and 88.17). rounding breast architecture. Fibroadenomas, papillomas, intramam- mary lymph nodes, and cysts are the most common causes of benign Calcifications masses (Fig. 88.15). Malignant masses classically have irregular borders Calcifications are highly radiodense and therefore best visualized with that blend into the surrounding tissue and often appear to infiltrate mammography compared with other modalities such as MRI and Cancer of the Breast • CHAPTER 88 1577 ultrasound. The type of calcification and the distribution of calcification For patients younger than 30 years, it may be reasonable to provide risk stratification. Benign calcifications usually are larger and perform an ultrasound scan only. This will be partly based on the coarser, and are often round with smooth margins. Benign causes of characteristics of the palpable abnormality. In this age group, masses microcalcification include involuting fibroadenoma, arteriosclerosis, that are round, smooth, and mobile are more likely to represent sclerosing adenosis, fat necrosis, and previous mastitis with ductal fibroadenomas or cysts. Ultrasonography is simple to perform and calcium deposits. A diffuse distribution pattern of calcifications and may distinguish solid from cystic masses. Cystic lesions can be aspi- skin calcifications are typically benign. Malignant calcifications typically rated. Cytologic examination of the aspirated fluid is not indicated. are linear, or small (<1 mm) in diameter and nonuniform in size. Chronically recurrent cysts, or bloody fluid on aspiration should Segmental or linearly distributed calcifications have the highest associa- prompt excision. In addition, complex breast cysts, defined as cysts tion with malignancy (60%–62% positive for cancer on biopsy), with with thick walls, thick septa, intracystic masses, or other discrete solid grouped distribution comprising a smaller proportion (20%–25%). components, may require either ultrasound-guided core biopsy of the solid components or surgical excision. For select younger patients Architectural Distortion with solid masses that have the classic clinical and radiologic features Distortion of the breast parenchyma is called architectural distortion of a fibroadenoma, observation with short-term clinical follow-up and may occur by itself or may be a feature associated with a mass, and repeat ultrasound examination to ensure stability may be calcifications, or asymmetry. The presence of architectural distortion reasonable. should prompt the search for an underlying abnormality and should For all other patients with solid masses on breast imaging, a biopsy be attributed either to a benign cause such as prior surgery, radial is strongly recommended. Fine-needle aspiration (FNA) biopsy is easy scar, complex sclerosing lesion, or fat necrosis, or to a malignant to perform in the office with minimal complications, although this condition such as invasive lobular or ductal carcinoma. does require an experienced cytopathologist for accurate interpretation. However, core biopsy has several advantages with respect to FNA in Approach to the Patient evaluating a solid breast mass, and is generally preferred. As opposed to the cytologic assessment enabled by an FNA, core biopsy provides Management of the Palpable Mass adequate tissue to differentiate between fibroadenoma and phyllodes At presentation, many patients will have a mass in the breast detected tumors, or in situ and invasive cancer. In the case of malignancy, it during self-examination or incidentally. Determination of whether a allows for immunohistochemical staining for ER, PR, and Her-2/neu palpable abnormality is malignant can be challenging. The first step expression, enabling definitive surgical or neoadjuvant planning. FNA should be a detailed history and physical examination. Patients with biopsy has a small risk of false-positive findings, so a positive FNA a palpable mass should undergo diagnostic bilateral mammography, biopsy result may still necessitate a core biopsy before one proceeds and often will undergo breast ultrasonography (Fig. 88.18). As opposed with surgery. It is important to remember that both FNA biopsy and to a screening mammogram, a diagnostic mammogram is typically core needle biopsy can have false-negative findings, so there must be reviewed immediately and additional images, including spot views, concordance between the biopsy results and the imaging or physical are obtained to characterize the specific area of concern. Bilateral examination findings. In cases in which a biopsy of a suspicious lesion mammography is important to evaluate the rest of the breast for results in a discordant benign finding, excisional or incisional biopsy additional occult lesions, in addition to the contralateral breast. should be performed.

Breast lump

History and clinical examination

Imaging: Age <30—ultrasonography (US) Age ≥30—diagnostic mammogram and US

Cyst Solid

Aspirate Biopsy (core needle or FNA)

Follow-up in 3-6 mo Benign Atypia DCIS Invasive

No recurrence Recurrence Continued Excise Adjuvant Local and screening therapy systemic therapy Follow-up in 6 mo Aspirate ADH/ (follow-up 1 mo) ALH

Recurrence

Excise

Figure 88.18 • Algorithm for breast lump management. ADH, Atypical ductal hyperplasia; ALH, atypical lobular hyperplasia; DCIS, ductal carcinoma in situ; FNA, fine-needle aspiration. 1578 Part III: Specific Malignancies

with or without an associated mass (Table 88.5). When a biopsy Management of the Nonpalpable Mammographic Abnormality is recommended, an image-guided core needle biopsy is preferred. As screening mammography is increasingly used, many patients will have This can be performed using ultrasound guidance (if the lesion is screening-detected mammographic abnormalities in the face of normal clearly visualized with ultrasound), or a stereotactic core biopsy may breast examination findings. If any abnormality is seen on screening be preferred. Some patients may not be candidates for stereotactic core mammograms, diagnostic mammograms including magnification views biopsy. These include patients with lesions close to the skin or chest will be required. These lesions are often then categorized according to wall, those with small or thin breasts, or those unable to lie prone BI-RADS as shown in Table 88.4. Short-term follow-up is typically and immobile for 30 to 45 minutes. In these cases, localization and recommended for BI-RADS category 3, and biopsy is recommended guided excisional biopsy with specimen radiography is recommended; for BI-RADS categories 4 and 5. Indications for biopsy include new localization has traditionally been performed with a wire placement. or changing densities, densities with spiculations, ill-defined borders There are newer techniques, including localization using radioac- or stellate distortion of the stroma, or suspicious microcalcifications, tive seeds or nonradioactive reflector chips, that may hold several

Table 88.5 Types and Distributions of Calcifications Type of Calcification Description TYPICAL BENIGN Skin (dermal) Typical lucent-centered deposits that are pathognomonic. Atypical forms may be confirmed by tangential views to be in the skin. Vascular Parallel track or linear tubular calcifications that are clearly associated with blood vessels. Coarse or popcornlike The classic calcifications produced by an involuting fibroadenoma. Large rodlike Benign calcifications forming continuous rods that occasionally may be branching. They usually are >1 mm in diameter, and may have lucent centers, if calcium surrounds rather than fills an ectatic duct. These are the kinds of calcifications found in secretory disease, “plasma cell mastitis,” and duct ectasia. Round When multiple, they may vary in size. They usually are considered benign, and when small (<1 mm), they often are formed in the acini of lobules. When <0.5 mm, the term “punctate” can be used. Rim Very thin, benign calcifications that appear as calcium deposited on the surface of a sphere. These deposits usually are <1 mm thick when viewed on edge. Although fat necrosis can produce these thin deposits, calcifications in the walls of cysts are the most common “rim” calcifications. Milk of calcium Consistent with sedimented calcifications in cysts. On the craniocaudal image, they often are less evident and appear as fuzzy, round, amorphous deposits; on the 90-degree lateral view, they are sharply defined, semilunar, crescent-shaped, curvilinear, or linear, defining the dependent portions of cysts. Suture Calcium deposited on suture material. They are relatively common in the postirradiation breast. They typically are linear or tubular in appearance, and knots often are visible. Dystrophic Calcifications that usually form in the irradiated breast or in the breast after trauma. Although irregular in shape, they usually are >0.5 mm in size. They often have lucent centers. INTERMEDIATE CONCERN Amorphous Often round or flake-shaped calcifications that are so small or hazy that a more specific morphologic classification cannot be determined. Heterogeneous Smaller than dystrophic calcification, these irregular calcifications are typically between 0.5 mm and 1 mm. HIGHER PROBABILITY OF MALIGNANCY Fine pleomorphic These have discrete sharply defined varying shapes and sizes, although they are usually <0.5 mm in diameter. Fine linear, or fine linear Thin, irregular calcifications that appear linear, but are discontinuous and <0.5 mm wide. Their appearance suggests branching (casting) filling of the lumen of a duct involved irregularly by breast cancer. DISTRIBUTION MODIFIERS Used as modifiers of the basic morphologic description. These terms describe the arrangement of the calcification. Multiple similar groups may be indicated when there is more than one group of calcifications that are similar in morphology and distribution. Grouped Although historically the term “clustered” has connoted suspicion, the term is now used as a neutral distribution modifier and may reflect benign or malignant processes. It is used when multiple calcifications occupy a small volume (<2 mL) of tissue. Linear Calcifications are arrayed in a line that may have branch points. Segmental Worrisome in that their distribution suggests deposits in a duct and its branches, raising the possibility of multifocal breast cancer in a lobe or segment of the breast. Although benign causes of segmental calcifications exist (e.g., secretory disease), this distribution is of greater concern when the morphology of the calcifications is not specifically benign. Regional Calcifications scattered in a large volume of breast tissue and not necessarily conforming to a duct distribution. They are likely benign, but are not everywhere in the breast, and do not fit the other, more suspicious categories. Diffuse Calcifications that are distributed randomly throughout the breast.

Modified from Breast Imaging Reporting and Data System. Reston, VA: American College of Radiology; 1998; 27. Cancer of the Breast • CHAPTER 88 1579 advantages over wire localization.190 After excision, the specimen is if untreated. In breast cancer, the standard biomarkers are ER, PR, sent to the radiology department to ensure that the suspicious lesion and HER2, which are all critical for both patient assessment and or microcalcifications, and the wire or localization seed or chip, have treatment decision making. However, reliability of marker assessment been removed. was lacking.194 Guidelines have now been developed by the College of Wire localization biopsy is recommended when the core biopsy American Pathologists and the American Society of Clinical Oncol- results are either nondiagnostic or equivocal, or when certain benign ogy (ASCO) in order to optimize standardization and reliability of diagnoses are encountered. If atypia, including atypical ductal hyper- the assays.193,195,196 plasia (ADH), atypical lobular hyperplasia (ALH), FEA, or LCIS, is Multiple additional prognostic markers have been identified to seen at core biopsy, it is generally recommended that excisional biopsy predict the likelihood of recurrence of invasive carcinoma, including be performed to rule out concomitant DCIS or invasive cancer, which number of involved axillary lymph nodes, tumor size, and tumor may be present in 10% to 15% of patients, although there may be grade.197 Additional pathologic characteristics of the primary tumor, some bias in patient selection.191 Surgical biopsy is also recommended including histologic subtype, Ki67, and lymphovascular invasion, have when the core biopsy reveals a papillary lesion or a radial scar, as these also been evaluated as additional prognostic factors. Ki67 in particular may also be upstaged after excision. has been studied extensively, and although it appears to be useful for With the increasing use of MRI screening in women at high risk estimating tumor proliferation, the assay lacks standardization across for breast cancer, radiologists and breast surgeons are often faced with laboratories and therefore is of questionable clinical usefulness at this the need to perform biopsy on a lesion not visible with mammography time and is not recommended for routine clinical use by the ASCO or ultrasonography. When an abnormality requiring biopsy is detected guidelines.198,199 with MRI but is not visible on mammograms, a focused ultrasound Newer prognostic and predictive biomarkers, such as gene expression examination should be performed. If the lesion is visible, ultrasound- profiles200–207 and assessment of tumor-infiltrating lymphocytes, have guided core biopsy may be performed. MRI may be necessary after also been developed or are currently in development to complement biopsy to ensure that the ultrasound-detected lesion and the MRI- traditional pathologic markers, as described elsewhere in this chapter. detected lesion are one and the same. If no lesion can be detected with ultrasound examination, MRI-guided biopsy should be performed. Open surgical biopsy with MRI needle localization is rarely needed, PRIMARY THERAPY but if it is, it is important to note that the nonmagnetic wire used Management of Noninvasive Breast Cancer for MRI localization is thinner and more easily transected during surgery. In addition, specimen mammography may not be useful, so Because of the increasing acceptance of breast cancer screening in the postexcisional MRI may be necessary to confirm removal of the lesion. United States, CIS accounts for an increasing proportion of all new breast cancers. Between 1975 and 1978, surveys suggested an incidence STAGING of pure in situ lesions of 1.4% to 5.1%. By 2001, nearly 20% of all new breast cancers in the United States were DCIS, and this denomina- Once a tissue diagnosis of breast cancer has been made, additional tor did not include cases of LCIS.160 Important research questions evaluation will determine if the patient has operable disease that is attempt to address which CIS is preinvasive cancer, which indicates potentially curable with multimodality therapy. Treatment options an unstable epithelium that represents an increased risk of subsequent include one or more of the following: surgery to the breast and axilla, invasive cancer, and how to intervene optimally (with surgery or other radiation therapy, and systemic therapy. Selection of treatment is preventive measures). Until better predictive markers are identified, dependent on both stage of disease and predictive and prognostic traditional methods of estimating risk are still in use. factors, in addition to individual patient characteristics. Lobular Carcinoma in Situ Seventh Edition of the TNM Staging System LCIS is a microscopic diagnosis and is usually not associated with a mammographic abnormality. When LCIS is found after needle- The American Joint Committee on Cancer (AJCC) staging system is localization biopsy of suspicious microcalcifications, the calcifications used to stage breast cancer according to the TNM.192 Changes have usually are outside the LCIS and the LCIS itself is usually an inci- been made over time to reflect advances in surgical treatment and dental finding. LCIS often is multicentric and is associated with an imaging and pathologic assessment of breast cancer. For example, in increased risk of subsequent invasive cancer in both breasts.208 The risk the current version of the AJCC staging system for breast cancer, the appears to be higher in the ipsilateral breast, suggesting that it may be a seventh edition, stage IB was added to reflect the diagnosis of tumors precursor lesion.209 2.0 cm or smaller (T1) with associated micrometastatic (N1mi) nodal Classic LCIS itself does not require treatment. Evaluation for disease. Another update was the use of clinical measurement to risk-reducing surgery in LCIS must take into account other risk determine the tumor size before neoadjuvant therapy and the use of factors, including family history and inherited genetic predisposition. gross and microscopic histologic features to describe the size of the Chemoprevention can also be considered to reduce the risk of subse- residual disease after neoadjuvant chemotherapy. quent breast cancer diagnosis, as discussed earlier. Pleomorphic LCIS is a morphologic variant with an aggressive behavior. Accumulating Prognostic and Predictive Factors for clinical data demonstrate that it may therefore require more aggressive Invasive Carcinoma treatment, including free margins at the excision, similar to treatment for DCIS.210,211 Pure prognostic markers predict the likelihood of patient outcome in the absence of therapy, whereas pure predictive markers predict Ductal Carcinoma in Situ response to a specific therapy.193 Many markers serve dual roles. For Before the widespread use of mammography, DCIS was not commonly example, ER is predictive for response to treatment with adjuvant diagnosed and usually manifested as a palpable mass or bloody nipple endocrine therapy, and it is also a moderate positive prognostic factor discharge. The increasing use of screening mammography has resulted because patients with tumors with high levels of ER expression generally in a significant increase in the number of patients diagnosed with have a lower likelihood of disease recurrence even without treatment. DCIS. Most of these cases are clinically occult. Even when palpable Similarly, HER2 is predictive of response to anti-HER2 therapy, but DCIS is present, the mammographic findings are quite characteristic, it is a moderate negative prognostic factor because patients with with a diffuse, often linear, and extensive pattern of pleomorphic high levels of HER2 have a higher likelihood of disease recurrence calcifications. However, the screening mammogram usually is the first 1580 Part III: Specific Malignancies

Table 88.6 Guidelines for Evaluation and Table 88.7 Multivariate Analysis of Risk Factors Treatment of Nonpalpable Ductal Related to Local Recurrence Carcinoma in Situ Hazard 95% Confidence 1. Careful multiview mammography with or without ultrasound and Variable Ratio Interval P Value including magnification views AGE (YEARS) • Document extent of disease • Identify other areas of microcalcification >40 1 2. Suspicious microcalcifications and densities cleared with needle ≤40 1.89 1.12–3.19 .026 localization biopsy METHOD OF DETECTION 3. Specimen radiography with magnification techniques 4. Radiograph-directed histopathologic evaluation with orientation of Radiographic finding only 1 specimen by surgeon using multicolored inked margins Clinical symptoms 1.55 1.11–2.16 .012 5. Complete pathologic description to include: • Type of DCIS and size of tumor HISTOLOGIC TYPE • Relation to microcalcifications Well differentiated 1 • Distance of lesion from inked margins Intermediately 1.85 1.18–2.90 .024 • Presence of multifocality differentiated • Presence or risk of microinvasion 6. Repeat mammography with magnification to confirm successful Poorly differentiated 1.61 0.93–2.79 clearing of suspicious areas ARCHITECTURE 7. Repeat breast excision if: Clinging or micropapillary 1 • Residual microcalcifications are found • Margins are unacceptable Cribriform 2.39 1.41–4.03 .002 Solid or comedo 2.25 1.21–4.18 DCIS, Ductal carcinoma in situ. MARGINS Free 1 indication of DCIS. Less common mammographic findings include Not free 1.84 1.32–2.56 .0005 architectural distortion or a mass. Table 88.6 summarizes the evaluation TREATMENT and treatment of DCIS identified with imaging. LE + RT 1 Treatment of ductal carcinoma in situ LE 1.82 1.33–2.49 .0002 The overwhelming majority of cases of DCIS are diagnosed with LE, Local excision; RT, radiation therapy. mammography; almost three-quarters of those are diagnosed based on Data from Bijker N, Meijnen P, Peterse JL, et al. Breast-conserving treatment with the presence of microcalcifications without a concomitant mass. When or without radiotherapy in ductal carcinoma-in-situ: ten-year results of European offering the patient a breast-conserving approach, the surgeon must be Organisation for Research and Treatment of Cancer randomized phase III trial confident that the remaining breast is free of suspicious mammographic 10853—a study by the EORTC Breast Cancer Cooperative Group and EORTC abnormalities. Other clusters of microcalcifications within the breast Radiotherapy Group. J Clin Oncol. 2006;24:3381–3387. take on greater significance after a diagnosis of DCIS is established, and biopsy is often necessary to exclude multicentricity. In cases where multicentricity is documented, or the extent of calcifications requiring excision makes lumpectomy impossible because of the proportions the NSABP B-17 study, 818 women with DCIS were randomized to of the resection specimen and volume of the breast, simple mastec- lumpectomy with or without radiotherapy. At 12 years, radiation therapy tomy with or without reconstruction is appropriate local treatment. significantly reduced the cumulative incidence of invasive ipsilateral Otherwise, lumpectomy is a reasonable approach for treatment of breast tumor recurrence (from 16.8% to 7.7%) and noninvasive DCIS. Current guidelines recommend that surgical margins for DCIS ipsilateral breast tumor recurrence (from 14.6% to 8.0%).214 Similarly, be at least 2 mm in order to reduce the risk of in-breast recurrence.212 the European Organisation for the Research and Treatment of Cancer In general, because DCIS is noninvasive, regional staging via sentinel (EORTC) 10853 trial, in which 1010 women were randomized, revealed lymph node (SLN) biopsy is not necessary. However, the diagnosis a 48% reduction in the risk of DCIS recurrence and 42% reduction of DCIS is subject to sampling errors, and the cancer is often upstaged in invasive local recurrence. Although the effect of radiation therapy to invasive cancer at the time of lumpectomy or mastectomy. Most was homogeneous across risk factors, it is noteworthy that a number patients with DCIS undergoing lumpectomy may return to the of identifiable clinicopathologic features were independently associated operating room for SLN biopsy if upstaging occurs. However, it may with local recurrence risk in this study (Table 88.7).215 The patients from be reasonable to perform SLN biopsy at the time of lumpectomy in these two trials were combined with those enrolled in trials conducted patients at high risk for invasion, including patients with DCIS with in Sweden and in the United Kingdom, Australia, and New Zealand microinvasion (or areas suspicious for microinvasion), or patients with in a meta-analysis of individual patient data from 3925 women by DCIS manifesting with a palpable or mammographic mass. For patients the Early Breast Cancer Trialists’ Collaborative Group (EBCTCG).216 undergoing mastectomy for DCIS, SLN biopsy at the time is recom- This meta-analysis established that radiation therapy halves the risk mended. There is relatively low morbidity in performing SLN biopsy of local recurrence after lumpectomy in patients with DCIS, with an at the time of a mastectomy, and it would not be possible to return absolute 15% reduction in the 10-year risk of ipsilateral breast events for regional staging afterward if invasive cancer is detected. If SLN (both invasive and noninvasive) and little effect on contralateral or biopsy is not performed and the patient is diagnosed with invasive distant events and no effect on breast cancer mortality. cancer, a formal axillary lymph node dissection (ALND) would be Given the nontrivial burden and toxicity of radiation therapy, necessary to accurately stage the cancer. research has sought to identify subgroups of women with DCIS who The role of adjuvant radiotherapy after lumpectomy for DCIS might have a sufficiently low risk of local recurrence after lumpectomy was established by the consistent results of randomized trials.213 In alone to omit radiotherapy. Although clinicopathologic features do Cancer of the Breast • CHAPTER 88 1581 appear to predict risk of local recurrence, and retrospective studies Adjuvant radiation therapy decisions are made based on the final have suggested that tumor size, grade, age, and margin status can help surgery performed, pathologic characteristics of the tumor, and clinical to identify patients with DCIS who are at low risk of local recurrence factors such as age and comorbidities. Decisions about adjuvant systemic after lumpectomy alone,217 prospective studies enrolling such patients therapy are made using a combination of predictive and prognostic have not yet clearly identified patients in whom the long-term risk of factors. Additional details about each of these treatment modalities local recurrence is extremely low after excision alone. In a single-arm are provided in the following sections. trial of wide local excision alone at Harvard, even patients with limited amounts of DCIS excised with widely negative margins had a 5-year risk Surgery for Early-Stage Breast Cancer of local recurrence of 12%, leading the authors to conclude that even Mastectomy was the initial surgical approach to treating breast cancer. patients with favorable clinicopathologic features have a sufficient risk of Pioneering investigations in the 1920s and 1930s began to treat groups local failure to justify adjuvant radiotherapy.218 Similarly, in the ECOG of women with breast-conserving partial mastectomy, followed by 5194 single-arm trial, the 561 patients with low- or intermediate-grade irradiation to the intact breast, challenging the need for total mastec- DCIS smaller than 2.5 cm in extent, treated with excision with at least tomy. Results from those early studies were promising. Beginning in 3-mm negative margins, had a 12-year risk of local failure of 14%, and Europe and Canada, and later in the United States, several institutions the risk was 25% among 104 patients with high-grade disease 1 cm or initiated programs, including excision plus radiation therapy for select smaller.219 A genomic assay has been developed using these data to help breast cancer patients. In the early 1970s, several European reports provide more refined predictions of the risk of local recurrence in such created worldwide interest in nonmastectomy treatments based on patients, but given that the low-risk group identifiable with that assay individual series that suggested local recurrence rates of 5% to 10%, still appears to have a 10-year risk exceeding 10%, many patients and and similar survival. physicians continue to pursue adjuvant radiotherapy in this setting.220 These studies caused considerable controversy between breast Concerns about possible late recurrences have also limited the impact surgeons who embraced breast conservation and those who believed of the early results of the randomized Radiation Therapy Oncology mastectomy should remain standard of care. This led to six randomized Group (RTOG) 9804 trial, which showed a promisingly low risk of prospective trials across Europe and the United States comparing local recurrence (6.7% at 7 years) in patients who did not receive breast conservation and mastectomy, which demonstrated survival radiotherapy, and which was reduced to 0.9% with the addition of equivalency between the two approaches. Investigators of both the radiation therapy.221 Of note, 62% of the patients in the RTOG trial NSABP B-06 trial and the Milan Cancer Institute Trial reported received tamoxifen, which has been established in randomized trials 20-year results showing equivalent outcomes.225,226 The breast-conserving from the United Kingdom, Australia, and New Zealand and in NSABP therapy data were so convincing that in 1990, the NCI held a consensus B-24 to reduce the risk of both ipsilateral breast tumor recurrence and development conference on the treatment of early breast cancer and contralateral breast cancer in patients with DCIS, as described in detail declared that breast-sparing therapy not only was equivalent to subsequently.222 mastectomy, but was actually the “preferable” treatment.227 Approximately 50% to 75% of DCIS lesions express ER and/or In the NSABP B-06 trial, women were randomly assigned to three PR. Tamoxifen has been shown to reduce the risk of local recurrence treatment arms: mastectomy, breast conservation with radiation therapy, in DCIS. In the NSABP B-24 study, 1804 women were randomized and breast conservation without radiation therapy. Although all three to tamoxifen or placebo after lumpectomy and radiation therapy.214 arms had a similar survival, the breast conservation–alone arm With a median follow-up of 163 months, the risk of ipsilateral invasive experienced a nearly 40% in-breast local failure rate. The absence of recurrence was reduced by approximately 32%, although the absolute a relationship between local failure and distant metastases suggested risk reduction was only 3%. In the United Kingdom, Australia, and that this was an acceptable approach. However, the EBCTCG performed New Zealand DCIS trial, patients who underwent lumpectomy for a meta-analysis that included 78 randomized studies involving over DCIS were randomized to radiation therapy, tamoxifen, both, or 42,000 patients. These researchers found that avoidance of a local neither.222 In those patients who underwent lumpectomy, after a median recurrence affected 15-year breast cancer mortality. Improved local follow-up of 12.7 years tamoxifen reduced recurrence of DCIS with control translates to avoiding one breast cancer death for every four a hazard ratio (HR) of 0.70 (95% CI, 0.51–0.86), but not invasive local recurrences avoided. Therefore lumpectomy alone is generally disease. In the same trial, radiotherapy reduced the incidence of both discouraged, although there are some groups in which this may be ipsilateral invasive disease (HR, 0.32 [0.19–0.56]) and ipsilateral DCIS acceptable, as discussed further later.228 (HR, 0.38 [95% CI, 0.22–0.63]). Meta-analysis of the results of these two randomized trials showed that the addition of tamoxifen to Resection of the Primary Tumor breast-conserving therapy for DCIS reduced the risk of both ipsilateral Initially there was considerable variability in the techniques used for and contralateral DCIS, and revealed a trend toward a reduced risk resecting the primary tumor. In Europe, this often meant a segmen- of ipsilateral and contralateral invasive carcinoma, although there was tectomy or quadrantectomy, procedures that were characterized by a no survival benefit.223 For postmenopausal women, the AI anastrozole significantly larger resection volume compared with the lumpectomies is a reasonable alternative based on findings from the NRG Oncology/ or wide excisions performed in the United States.229 The concept of NSABP B-35 trial, in which anastrozole was associated with a longer large excisions has been progressively replaced by the concept of free breast cancer–free interval compared with tamoxifen.224 margins. There has been considerable debate as to the definition of the free margin distance, with multiple retrospective studies examin- Management of Early-Stage Invasive Breast Cancer ing the relationship between local recurrence and the microscopic margin distance. The Society of Surgical Oncology and the American The management of stages I to III breast cancer is multidisciplinary Society for Radiation Oncology convened a multidisciplinary panel and involves specialists in breast imaging, pathology, surgical oncology, and performed a meta-analysis of 33 studies including over 28,000 radiation oncology, medical oncology, and reconstructive surgery. patients.230,231 They concluded that although a positive margin was Depending on the extent of disease at diagnosis and other factors associated with a greater than twofold increase in local recurrence such as family history, patients may be candidates either for breast- (odds ratio [OR], 2.44 [95% CI, 1.97–3.03]), margins wider than conserving therapy with lumpectomy plus radiation therapy or for “no ink on tumor” were not associated with a lower incidence of mastectomy. Preoperative systemic therapy may also be a consideration, local recurrence. However, the meta-analysis was based on retrospec- with the goal of treating subclinical micrometastatic disease, reducing tive studies with significant selection bias, so there should be some local and regional tumor bulk, and increasing the likelihood of successful flexibility in application, with re-excision recommended for select surgical resection. patients. 1582 Part III: Specific Malignancies

Orientation of the specimen is mandatory to allow accurate Skin flaps are carefully developed—thin enough to remove all apparent pathologic assessment of the margins and guide re-excision. Pathologists breast tissue, but without removing the subcutaneous tissue of the ink the surface of the specimen by using six different colored inks. flaps that carry the blood vessels of the skin. The inferior flap extends The specimen is then cut and histologic sections are prepared. Extra inferiorly to the inframammary crease. The superior flap is extended attention is paid to areas near the surgical margin whenever tumor is to just beneath the clavicle. The medial extension of the flaps reaches seen to encroach grossly. the lateral edge of the sternum; the lateral extension should include Several intraoperative techniques have been described to reduce the axillary tail of Spence but, unless a modified radical mastectomy the re-excision rate, which in some studies can be as high as 40% to is being performed, should not include the axillary contents. During 50%. Obtaining routine cavity shave margins in all anatomic directions dissection, it is important to spare the medial pectoral nerves, which at the time of the lumpectomy can reduce the rate of re-excision.232 wrap around the lateral border of the pectoralis minor muscle and Intraoperative margin analysis has been shown to significantly decrease insert into the posterior aspect of the pectoralis major. Injury to re-excision rates, but implementation can be challenging.233–235 Cytologic these nerves leads to atrophy of the central portion of the pectoralis touch preparation analysis and intraoperative margin assessment devices major muscle. When bilateral mastectomies are being performed, it can also be used. There is no prevailing standard of care, and practices is critical not to cross the sternum, creating a tunnel between the vary widely.236 At the completion of the lumpectomy, small titanium two medial flaps. Once the flaps are developed, the breast is dissected clips should be placed at the margins to guide the radiation oncologist from the chest wall by dissecting the pectoralis major fascia off of for precise placement of a boost, or for partial breast irradiation. After the muscle. lumpectomy, the skin should be closed, but reapproximation of the Reconstruction of the breast is a valid option that can be done breast tissue should be avoided because this leads to distortion of the immediately or after the procedure (i.e., “delayed”). With attempts breast shape.237 For patients with a significant defect, oncoplastic to improve overall cosmesis after mastectomy and breast reconstruction, techniques can be used to rearrange the breast parenchyma to avoid more “conservative’ mastectomy procedures have evolved. Skin-sparing excess concavity.238 These are best planned preoperatively. mastectomy, in which removal of the skin is limited to the nipple-areolar Although most women today are candidates for breast conservation, complex (sometimes with a small portion of skin), is an oncologically there are several contraindications: safe approach that minimizes deformity and improves cosmesis. Retrospective series of skin-sparing mastectomy and breast reconstruc- • Scleroderma, cutaneous lupus erythematosus, and other collagen tion show similar rates of local control with those of conventional vascular diseases are considered relative contraindications to breast mastectomy. The nipple-sparing mastectomy, in which only the breast conservation, particularly if active, because evidence suggests an parenchyma is removed, has also emerged as an acceptable choice for increased severity and frequency of radiation-related complications select early-stage breast cancers and high-risk women pursuing mas- in such patients. tectomy, although the ductal tissue below the nipple may be an • Diffuse microcalcifications throughout the breast. oncologic concern when this is being performed for cancer. Nipple- • Multicentric invasive cancer or DCIS. sparing mastectomy should be considered for cancer patients only • Inability to achieve negative margins despite repeated attempts at when the tumor is at least 2 cm from the nipple-areolar complex and re-excision. in the absence of skin or nipple involvement, widespread microcalcifica- • Patients with previous chest wall irradiation require evaluation by tions, or microcalcifications extending to the nipple. Complete removal a radiation oncologist to determine the feasibility of breast conserva- of this tissue may compromise the vascularity of the nipple and lead tion; innovative approaches such as partial breast reirradiation are to necrosis. Intraoperative analysis of the subareolar tissue is often under investigation,239 but a history of prior therapeutic doses of performed, and the nipple-areolar complex is sacrificed if findings in radiotherapy to the whole breast is still generally considered a frozen sections are positive. contraindication to breast conservation. Contralateral Prophylactic Mastectomy There has been a recent trend toward women with unilateral breast Mastectomy cancer opting for removal of both breasts—the affected ipsilateral For patients who have a contraindication to breast conservation, breast along with a CPM. Even while the indications for mastectomy mastectomy is indicated. When patients are unable to achieve complete have decreased, more women deemed eligible for breast-conserving removal of the cancer and an acceptable cosmetic result, mastectomy therapy have opted for bilateral mastectomies.240 This started with and reconstruction may be a better option. Some patients may strongly the availability of BRCA testing in the preoperative period in 2002, desire conservation despite a less than satisfactory cosmetic outcome, but the rates of bilateral mastectomies have markedly increased since and several techniques exist for addressing defects in the breast including that time.241,242 Today, most patients who undergo CPM do not have local tissue flaps or fat grafting. strong genetic or familial risk factors for developing contralateral breast For many decades, the Halsted radical mastectomy was the treatment cancer.243 This trend is primarily being driven by an overestimation of choice for all stages of breast cancer. This involved the resection of the risk of contralateral breast cancer and potentially unrealistic of the breast parenchyma, a large portion of the skin overlying the outcomes from CPM.244,245 Although women who have had one breast breast, and the major and minor pectoral muscles en bloc with levels cancer do have an increased risk of a second cancer, this risk is not I to III of the axillary nodes. This eventually gave way to a less-aggressive as high as expected, estimated to be approximately 0.4% per year operation—the modified radical mastectomy—without compromising among BRCA1/2-negative patients with hormone receptor–positive survival. The modified radical mastectomy consists of resecting the tumors taking antiestrogen therapy, and 0.5% per year among hormone breast parenchyma en bloc with levels I and II axillary lymph nodes, receptor–negative patients.246 CPM does nothing to decrease the risk but preserving adequate skin and the pectoralis muscles. of distant recurrence, which is significantly higher than the risk of a With the evolution of SLN biopsy for axillary staging, the efficacy second primary cancer, and thus CPM has no impact on survival. of systemic therapies, and the use of postmastectomy radiation therapy, The misconceptions regarding the benefits of CPM come from multiple the modified radical mastectomy has become less common. Total or sources, including physicians and the media.247–249 For these reasons, simple mastectomy, with removal of enough skin to allow closure and groups such as the Society of Surgical Oncology and the American the nipple-areolar complex and all the breast tissue without the axillary Society of Breast Surgeons have advocated for having detailed conversa- lymph nodes, has supplanted modified radical mastectomy for most tions with patients regarding their expectations, educating them on patients. When elliptical or transverse incisions are used, it is ideal to the true benefits and risks of CPM, and discouraging CPM among excise any previous biopsy scar along with the nipple-areolar complex. women without a high risk of future breast cancer.250,251 Cancer of the Breast • CHAPTER 88 1583

Ninety percent of the patients in the IBCSG 23-01 trial and a Management of the Axilla similarly high proportion of the women in the Z0011 trial received In addition to resection of the primary lesion, the surgical treatment whole-breast irradiation, which covers much of the axilla. In the of invasive breast cancer involves management of the axilla. This is Z0011 trial, although the protocol called for no radiation therapy not only important in maximizing locoregional control, but significant to the axilla, several patients had either adjusted tangents or direct prognostic information is gained by determining the pathologic nodal axillary radiation therapy.260 In addition, the EORTC conducted status, which helps guide adjuvant systemic therapy decisions and the AMAROS trial, in which 4806 patients with T1 or T2 tumors radiation planning. with positive SLNs were randomized to either ALND or axillary For patients with no evidence of regional metastases at either radiotherapy.261 There were no significant differences in the axillary physical examination or imaging (clinically node-negative patients), recurrence rate (0.43% versus 1.19%), DFS, or OS. Lymphedema SLN biopsy is the standard surgical approach to staging the axilla. was significantly more common after ALND than after radiation More than 60 observational studies (in more than 6000 patients) in therapy. Looking at all three trials, it seems reasonable that patients which SLN biopsy was followed by ALND have established a greater with a positive SLN may avoid completion ALND if irradiation is than 95% success rate for SLN identification and a false-negative rate planned (either for breast-conserving therapy or postmastectomy below 10%. This false-negative rate, although potentially affecting radiation therapy). Whether to adjust the tangents or perform direct adjuvant therapy decisions, does not pose an increased risk for the axillary irradiation should be based on individual clinicopathologic development of axillary recurrence among patients staged with SLN features, including tumor burden within the SLNs. ALND should be biopsy alone. Data from several observational studies252–254 and one performed for a positive SLN when no radiation therapy or partial breast randomized trial255 demonstrate that axillary recurrence after a negative irradiation is planned. ALND should also still be performed in SLN- SLN biopsy result is a rare event, occurring in 0.2% of cases. Possible positive patients with three or more positive nodes, when extranodal explanations may include the local effect of systemic therapy and the use extension is present, or when the SLN is positive after neoadjuvant of breast radiotherapy, which inevitably treats a portion of the axillary chemotherapy. lymph nodes. Clinically node-positive patients—those who have adenopathy at NSABP B-32 was a prospective randomized phase III clinical trial physical examination or suspicious nodes at imaging—require a different designed to compare SLN biopsy with conventional axillary dissection approach. Any suspicious nodes at examination or imaging should in clinically node-negative patients.256 A total of 5611 patients with be biopsy proven, because these may often be reactive. For biopsy-proven clinically negative axillae were randomly assigned to SLN biopsy regional metastases, ALND is recommended. Axillary clearance for and ALND versus SLN biopsy and ALND only if the SLN was breast cancer typically involves levels I and II, those lymph nodes positive. There was significantly less morbidity with SLN biopsy, but lateral and inferior to the pectoralis minor muscle. If there is significant no difference in disease-free survival (DFS) or OS between the two axillary disease, the level III nodes should also be included. The level groups. As expected, patients with hematoxylin-eosin (H&E)–positive III nodes lie medial to the pectoralis minor and are bordered medially nodes had worse OS with respect to node-negative patients. After by the Halsted ligament. If possible, the pectoralis minor muscle is pathologic examination at the treating institution, paraffin blocks of preserved. all SLN-negative specimens from B-32 were sent to a central labora- If the axilla is to be dissected, either in continuity with the breast tory for a clinically blinded search for occult metastases. Additional in a modified radical mastectomy or as a separate axillary dissection, sections were evaluated with H&E and cytokeratin-IHC stains, with the clavipectoral fascia is divided. Fat from the axilla then pops through additional micrometastases and IHC-positive cells detected.257 Although this division, and differs in appearance from the subcutaneous fat. these were associated with a worse prognosis than truly negative SLN The inferior border of the axillary vein is roughly two fingerbreadths patients, the difference was minimal (1.2% decrease in OS). For this below the highest extent of the pectoralis major fascia. Staying below reason, the presence of isolated tumor cells (ITCs) is considered the axillary vein is critical in order to avoid brachial plexus injury and node-negative disease in the AJCC staging system, with an “i+” spare the brachial lymphatics, minimizing the risk of lymphedema. designation (N0i+). The thoracodorsal vein is identified inferiorly and posteriorly to the The management of the positive SLN has changed dramatically. axillary vein. The thoracodorsal nerve is seen to emerge from behind Initially, all patients with a positive SLN were recommended to undergo the axillary vein, just medial to the thoracodorsal vessels. It joins these completion ALND. However, over half of patients had no additional vessels, continuing onto the anterior surface of the thoracodorsal vein. disease identified, and SLN-positive patients go on to receive systemic Between 1 and 2 cm below the axillary vein, the highest branch of therapy and adjuvant radiation therapy, which may help control any the intercostobrachial nerve is seen coming from the chest wall and residual micrometastatic disease in the axilla, calling into question going to the arm. If there is no concern regarding its proximity to the benefit of the completion dissection. The American College of cancer, it is possible to spare this nerve branch to avoid the dysesthesias Surgeons Oncology Group (ACOSOG) Z0011 study was designed and numbness of the inner posterior arm. When the thoracodorsal to address this question among patients with a positive SLN undergoing bundle has been identified along its length (to its insertion into the breast-conserving surgery followed by radiotherapy.258 Patients with latissimus dorsi), elevating the pectoralis minor muscle and sweeping T1 or T2 tumors and fewer than three positive SLNs were randomized the axillary fat downward off the chest wall moves the level II axillary to completion ALND or observation. There were similar 5-year OS nodes out from behind the pectoralis minor muscle. The long thoracic rates (92.5% versus 91.9%) and DFS rates (83.9% versus 82.2%). nerve is found in the same anteroposterior plane as the thoracodorsal The trial was underpowered and was heavily biased toward micro- nerve. At this point, the three major midaxillary nerves have been metastatic disease in the SLNs. Nevertheless, the very low ipsilateral identified. The axillary contents can now be cleared inferiorly, with regional recurrence rate (0.5% in the ALND arm and 1.5% in the all of the important nerves and vessels in view. SLN-alone arm) was convincing that for patients similar to those When suspicious palpable lymph nodes are present high within enrolled in Z0011, ALND is not necessary for a positive SLN, par- level II or level III, removal of the level III nodes is advised. If possible, ticularly in patients with micrometastatic disease. This was confirmed the pectoralis minor muscle is preserved. The lateral borders of the by the International Breast Cancer Study Group (IBCSG) 23-01 trial, pectoralis major and minor muscles are dissected, thereby preserving in which patients with tumors smaller than 5 cm and nodal metastases the medial pectoral nerve. In this way it is possible to dissect the apex in the SLN smaller than 2 mm were randomized to completion ALND without division or removal of the pectoralis minor. Both pectoral or no further surgery.259 Although this study also had low accrual, it muscles are retracted medially to maintain the exposure of the level found no statistically significant differences in DFS (84% versus 88%) III region. This may require the use of a Thompson retractor and or OS (98% versus 97.6%). bringing the arm across the chest to relax the muscles. This is why 1584 Part III: Specific Malignancies

the entire arm is usually prepared when an ALND is planned. For halves the risk of any breast cancer recurrence and yields a modest patients with very bulky disease, it may be necessary to divide the survival benefit, it is widely considered to be the standard of care after pectoralis minor to ensure adequate clearance of level III. lumpectomy for invasive breast cancer. Still, the absolute benefits of treatment vary considerably across subgroups. Therefore just as in the Irradiation of the Intact Breast case of DCIS, research has focused on trying to identify women with The numerous randomized trials comparing outcomes after lumpectomy sufficiently low risk of recurrence in the absence of radiotherapy to with or without adjuvant radiotherapy have been combined by the justify its omission, particularly among patients with endocrine-sensitive EBCTCG (Fig. 88.19).262 Because adjuvant whole-breast radiotherapy disease who receive effective systemic therapies.

Any first recurrence Breast cancer death

Women with pNO disease (n = 7287)

60 60 10-year gain 15.4% (SE, 1.1) 15-year gain 3.3% (SE, 1.3) RR, 0.49 (95% CI, 0.45−0.55) RR, 0.83 (95% CI, 0.73−0.95) 50 Logrank 2P < .00001 50 Logrank 2P− .005

40 40

BCS 30 22.5% 31.0% 30 BCS 20.5% 20 20 12.7% Any first recurrence (%) 15.6% Breast cancer death (%) 17.2% BCS+RT BCS+RT 10 10 5.5% 10.6% 10.9% 4.6% 0 0

Women with pN+ disease (n = 1050)

BCS 63.7% 60 60 53.7% BCS 50 50 51.3% 42.6% 42.5% 42.8% 40 BCS+RT 40 BCS+RT

30 30 34.2% 31.1% 22.4% Any first recurrence (%) 20 Breast cancer death (%) 20 19.8%

10 10-year gain 21.2% (SE, 3.4) 10 15-year gain 8.5% (SE, 3.4) RR, 0.53 (95% CI, 0.44−0.64) RR, 0.79 (95% CI, 0.65−0.95) Logrank 2P < .00001 Logrank 2P < .01 0 0 0 5 10 15 0 5 10 15

Years Years

Figure 88.19 • Effect of radiation therapy after breast-conserving surgery on risk of first recurrence (locoregional or distant) and on breast cancer mortality. Data from 17 trials. Vertical lines indicate 1 standard error (SE) above or below the 5-, 10-, and 15-year percentages. CI, Confidence interval;RR, relative risk. (From Early Breast Cancer Trialists’ Collaborative Group et al. Effect of radiotherapy after breast-conserving surgery on 10-year recurrence and 15-year breast cancer death: meta-analysis of individual patient data for 10,801 women in 17 randomised trials. Lancet. 2011;378[9804]:1707–1716.) Cancer of the Breast • CHAPTER 88 1585

In the Cancer and Leukemia Group B (CALGB) 9343 trial, patients patients.267 Judicious selection of patients expected to benefit meaning- aged 70 and older with clinical stage I, ER-positive invasive breast fully from boost treatment includes consideration of factors known cancer treated with lumpectomy and tamoxifen were randomized to to correlate with the risk of local recurrence, such as age, grade, margin adjuvant whole-breast radiotherapy or not. Even in this highly select status, and systemic therapy. group, adjuvant radiotherapy significantly reduced the risk of local Given the observation that most local failures occur in the vicinity of recurrence (from 10% to 2% at 10 years).263 Still, because this reduction the original lesion, partial breast irradiation, which can be accomplished in local recurrence was not associated with an improvement in survival, with a variety of approaches ranging from external beam delivery to freedom from mastectomy, or other end points, this trial has been brachytherapy implants or intraoperative treatment, has also been the interpreted as establishing radiotherapy omission as a reasonable option subject of active investigation. Although concerns have been raised for consideration in this small subgroup of patients. In all other patients by the observation of a small increase in the rate of mastectomy in a who undergo lumpectomy outside of the context of a clinical investiga- claims-based study of patients who received balloon implants in the tion, radiation therapy remains the standard of care. Ongoing studies early days of development of that technique (a finding that might may identify additional populations at sufficiently low risk of local reflect toxicity before experience established parameters for safe spacing recurrence after lumpectomy alone to warrant avoidance of radiation of the balloon with respect to skin surface, for example, or perhaps therapy. an indication that early patient selection was suboptimal)268 and the Attempts to reduce the burden of adjuvant radiation therapy in observation of impaired cosmetic outcomes among patients treated this setting have included investigation of whole-breast hypofraction- with external beam approaches in a large Canadian trial,269 promis- ation, in which the whole breast is treated in fewer fractions (e.g., ing early results from other studies270 motivate eager anticipation of 40 Gy in 15 fractions or 42.5 Gy in 16 fractions) rather than the mature results from the large RTOG 0413/NSABP B-39 trial. In the conventional fractionation schedule of 50 Gy in 25 fractions. Strong meantime, consensus guidelines exist to select patients appropriate for evidence from randomized trials in the United Kingdom264 and treatment with partial breast approaches outside of the context of an Canada265 have established this to be equivalent in terms of both ongoing trial.271 efficacy and safety for women with early-stage invasive cancer, and For patients who have no positive axillary nodes after ALND or observational data suggest good outcomes also in patients with DCIS. have negative SLN biopsy findings, the axilla and other regional nodal Whole-breast irradiation is typically delivered with tangentially basins (supraclavicular and internal mammary) are not targets of oriented beams that treat the entire breast while minimizing dose to treatment. For those in whom at least one axillary node is positive, the underlying heart and lungs. Particularly given recent studies consideration is given to irradiating these regions in conjunction with emphasizing the dose-response relationship for cardiac toxicity of treatment to the whole breast. Treatment of the supraclavicular, radiotherapy,266 care is taken to shield the heart or to alter the anatomic infraclavicular, and internal mammary nodal regions is well accepted relationships between the breast tissue and the heart with use of for patients with four or more lymph nodes involved, for the same techniques such as breath holding or prone positioning (Fig. 88.20). reasons that postmastectomy radiotherapy is widely accepted in these This is often followed by boost treatment of an additional 10 to 16 Gy patients, as discussed later. For patients with fewer than four lymph to the tumor bed. A randomized trial from the EORTC established nodes involved, there is greater controversy, given that some of these the benefit of this boost treatment in improving local control in all patients were eligible for trials that have demonstrated benefits from patients, but with considerably lower absolute benefits in older regional nodal radiotherapy after lumpectomy and ALND (NCIC MA.20272 and EORTC 22922273) and for the ACOSOG Z0011 trial, which, as more fully described earlier, demonstrated excellent outcomes in patients who underwent lumpectomy with SLN biopsy alone.274 The low (levels I and II) axilla itself may be treated with minor modifications of the standard tangential beams used to treat the breast, an approach often called “high tangent” radiation therapy. This is often an approach favored in patients whose node-positive disease was identified at SLN biopsy and who did not undergo ALND; for patients with favorable features similar to those in the ACOSOG Z0011 and IBSCG 23-01 trials, many radiation oncologists will use this minor modification of treatment fields, which is not believed to cause meaningful changes in toxicity.260 Alternatively, in patients who have more concerning features, such as numerous lymph nodes involved or ER-negative disease, the additional lymph nodes in the infraclavicular (level III axilla) and supraclavicular regions may then also be targets of treatment; a third anterior field may be added to include just those regions or also to include the axilla if undissected or heavily involved. The internal mammary nodes may also be targeted in high-risk patients.

Complications of Treatment Advanced treatment planning and delivery approaches allow for improved dose homogeneity and visualization and exclusion of critical normal tissues, reducing the risks of radiation-related toxicity. Even with treatment to the whole breast and regional nodes after lumpectomy and ALND, the MA20 trial showed only a 1.2% incidence of radiation pneumonitis (and that rate was only 0.2% with breast irradiation alone, without regional nodal irradiation) and an 8.4% risk of lymphedema (a risk that was 4.5% with breast irradiation alone). Pneumonitis typically resolves spontaneously or after a short course Figure 88.20 • Left breast irradiation using prone breast technique can of corticosteroids and is not expected to result in clinically significant spare lung, left ventricle, and coronary arteries. long-term changes in lung function; lymphedema can be more 1586 Part III: Specific Malignancies

problematic but is reduced in the modern era with less extensive support the use of postmastectomy radiotherapy for patients with surgery. Rib fractures occur infrequently and heal spontaneously, but more extensive disease, including T3 or T4 primary tumors, four or the possibility merits care to ensure that when such a finding is dis- more lymph nodes involved, or residual axillary involvement after covered, appropriate consideration is given to treatment-related toxicity preoperative systemic therapy. rather than disease recurrence if the fracture is within the radiation treatment field. Dose-dependent cardiac toxicity has been shown in Adjuvant Systemic Therapy historical trials in which mean heart doses were substantially higher than those delivered in modern US practice; observational studies Along with mammography, adjuvant systemic therapy changed the suggest very low risk of cardiac sequelae from regimens administered course of early-stage breast cancer.280 Since the 1970s, randomized in contemporary practice with appropriate care to reduce cardiac trials have shown a survival benefit from adjuvant chemotherapy in exposure. Brachial plexus injury has been reported from treatment to operable, node-positive disease. Starting in 1985, the systematic reviews the supraclavicular region, and because this risk appears related to and meta-analyses by the EBCTCG led to the rapid adoption of daily fraction size, US practitioners frequently use conventional adjuvant systemic therapy, with updates published on chemotherapy281,282 fractionation rather than hypofractionation in patients in whom regional and endocrine therapy.67,283 Although adjuvant chemotherapy was irradiation is intended. Perhaps the most dreaded consequence of initially used primarily to treat patients with node-positive disease, therapeutic radiotherapy is secondary malignancy; the risk of soft by the mid-1990s it was routinely used to treat node-negative disease tissue sarcoma is approximately 0.2% at 15 years, and the risk of lung as well, regardless of hormone receptor status. In the same time period, cancer is also increased by radiotherapy, with epidemiologic analyses tamoxifen started being used for treatment of those with ER-positive suggesting that patients who smoke are at highest risk. disease. Soon, data began to suggest that not all patients with ER-positive Adjuvant Postmastectomy Irradiation disease obtained benefit from treatment with adjuvant chemotherapy. For the average patient, decision algorithms based on routine clini- Even after mastectomy, some patients have sufficient risk of harboring copathologic factors (e.g., tumor size, nodal status, grade, ER, and residual disease in the chest wall or regional nodes that adjuvant HER2) had proved quite useful for decision making. However, by radiotherapy may be beneficial. Historical trials initially indicated a the mid-2000s new genomic tools for prognosis and prediction became detriment from the addition of radiotherapy in this setting because available to better estimate recurrence risk and likelihood of benefit the benefits in local control were offset by increases in treatment-related from chemotherapy, respectively, to further individualize clinical toxicity, particularly cardiac mortality. However, landmark trials from decisions. Data from prospective and retrospective studies demonstrated Denmark275,276 and British Columbia,277 conducted in samples that that gene expression profiles such as Oncotype DX, described later, primarily included node-positive patients, revealed that postmastectomy and routine standard clinicopathologic parameters offered independent radiotherapy reduced local recurrence by two-thirds and improved prognostic usefulness in ER-positive disease.284–286 Therefore it was survival. The generalizability of these trials, particularly with respect important to develop optimal ways to integrate the complementary to the findings in patients with only one to three lymph nodes involved, data provided by standard clinicopathologic tumor assessment and was questioned in light of the limited extent of ALND in the Danish molecular markers. trials. Therefore prior consensus guidelines recommended postmas- Although screening mammography and consequent earlier diagnosis tectomy radiotherapy in patients with four or more lymph nodes of breast cancer was responsible for at least half of the breast mortal- involved, but failed to recommend for or against treatment in patients ity reduction observed between 1990 and 2003, the introduction of with one to three positive nodes and tumors 5 cm or smaller in size. adjuvant systemic therapy, including both cytotoxic chemotherapy However, a recent EBCTCG meta-analysis that focused on 3786 and endocrine therapy, also significantly reduced the odds of disease patients who were randomized to radiotherapy or not but who had recurrence and death.280 Analysis of large databases also indicates all received complete ALND also revealed substantial benefits, including that the 5-year survival rate in women with small hormone-receptor improved breast cancer mortality, from postmastectomy radiation positive tumors is not likely to be affected by their disease, and that therapy in node-positive patients (Fig. 88.21).278 Even in the 1314 chemotherapy offers minimal potential benefit.287 However, data from patients with axillary dissection and only one to three nodes involved, the NSABP suggest improvements in both recurrence-free survival and the improvement in breast cancer mortality was substantial (RR, 0.80; OS in women with ER-positive and ER-negative tumors no more than P = .01). Therefore the most recent consensus guidelines note that 1 cm in size who are treated with adjuvant chemotherapy.288 These postmastectomy radiotherapy reduces the risks of locoregional failure, discrepancies likely result from the modest treatment benefits obtained any recurrence, and breast cancer mortality for patients with T1 or with available systemic therapies at a population level, combined with T2 breast cancer with one to three positive axillary notes, but because the considerable heterogeneity observed in breast cancer even when some subsets are likely to have such a low risk of locoregional failure considering tumors with a similar profile based on standard pathologic that the absolute benefit of radiation therapy is outweighed by risks, characteristics (e.g., tumor size, nodal status, ER and PR expression). they call for individualized decision making in this context.279 Physicians Although systematic reviews and computerized nomograms have been are encouraged to consider in their recommendations a constellation quite useful to demonstrate the average benefit for specific patient of factors that may affect the risk of locoregional failure, attenuate subgroups, especially when the absolute benefit is otherwise small, the benefit of reduced breast cancer–specific mortality, and/or increase these efforts fail to recognize the variability among individuals. complication risks. These factors include patient characteristics such It is now understood that the small to modest therapeutic effects as age, life expectancy, coexisting conditions that might increase risk noted in individual clinical studies are of great value if applied to the of complications; pathologic features such as tumor size, lymphovascular large population of women with breast cancer. The 2005 EBCTCG invasion, number and size of nodal metastases, response to neoadjuvant meta-analysis demonstrated a significant survival advantage after systemic therapy if any administered; and biologic characteristics such polychemotherapy for all adequately studied age categories, although as grade, hormonal sensitivity, and subtype. The panel further encour- the number of patients older than 70 who were included was ages multidisciplinary decision making in this complex context, shared limited.67,281,282 There is now Level I evidence that chemotherapy is with patients to incorporate values regarding the weighting of benefits of benefit for older women with ER-negative disease.282 Likewise, and risks in the light of the best estimates that physicians can provide. chemotherapy has been shown to be effective in patients with either Although the most recent consensus guidelines primarily focus on node-negative or node-positive disease.281 Polychemotherapy was the groups for whom there has been greatest controversy over the past demonstrated to be superior to monochemotherapy, and more prolonged decade, it is important to stress that there is strong consensus to administration of chemotherapy (for 12 months or longer) has not Cancer of the Breast • CHAPTER 88 1587

700 pNO women with Mast+AD Locoregional recurrence first Any first recurrence Breast cancer mortality 100 100 100 90 90 90 80 Logrank 2P > 0.1 NS 80 10-year loss 1.3% (SE, 3.3) 80 20-year loss 2.2% (SE, 3.6) RR 1.06 (95% CI, 0.76−1.48) RR 1.18 (95% CI, 0.89−1.55) 70 70 logrank 2P > 0.1 NS 70 logrank 2P > 0.1 NS 60 60 60 50 50 50 40 40 40 RT RT 25.9 28.8% 30 30 30 22.4% 18.4

Any first recurrence (%) No RT 20 20 15.5 No RT 20 RT Breast cancer mortality (%) 23.9 26.6% 21.1% 11.0 Locoregional recurrence first (%) 18.3 10 3.0% No RT 10 10 1.9 13.3 1.6% 10.6 AB0 0 C 0 1.2 3131 pN+ women with Mast+AD Locoregional recurrence first Any first recurrence Breast cancer mortality 100 100 100 90 90 90 Logrank 2P < .00001 80 80 80 No RT 70 70 No RT 70 62.6 66.4% 62.5% 60 60 60 54.1 52.4 RT RT 50 50 51.9% 50 55.4 58.3% 37.2 48.0 40 40 43.0 40 No RT 30 21.3 30 30 33.9 26.0% 20 Any first recurrence (%) 20 20 10-year gain 10.6% (SE, 2.0) Breast cancer mortality (%) 20-year gain 8.1% (SE, 2.0)

Locoregional recurrence first (%) RT RR, 0.75 (95% CI, 0.67−0.83) RR, 0.84 (95% CI, 0.76−0.94) 10 10 10 8.1% logrank 2P < 0.00001 logrank 2P = 0.001 6.6 D 0 E 0 F 0 0510 15 20 0 510 15 20 0510 15 20 Years Years Years

Figure 88.21 • Effect of radiation therapy (RT) after mastectomy and axillary dissection (AD) on locoregional recurrence and on breast cancer mortality. Data from 22 trials. Vertical lines indicate 1 standard error (SE) above or below the 5-, 10-, 15-, and 20-year percentages. CI, confidence interval;RR, relative risk. (From Early Breast Cancer Trialists’ Collaborative Group et al. Effect of radiotherapy after mastectomy and axillary surgery on 10-year recurrence and 20-year breast cancer mortality: meta-analysis of individual patient data for 8135 women in 22 randomised trials. Lancet. 2014:383[9935]:2127–2135.) been associated with greater benefit compared with shorter duration predictive markers of response (e.g., ER, PR, and HER2), and patient of treatment (e.g., 6 months). factors such as comorbidities (Fig. 88.22).294 By including all three For patients with hormone receptor–positive disease, the EBCTCG categories of data, one can accurately estimate both the potential also showed that adjuvant tamoxifen improves survival, irrespective benefits and risks of treatment for an individual patient. Breast cancer of age or menopausal status.67 In contrast to chemotherapy, longer patients traditionally have overestimated the absolute value of systemic duration of treatment with tamoxifen (i.e., 5 years) provides greater therapy, and a common misinterpretation is that the treatment benefit benefit than a single year of administration. Although initial studies is similar across all patients, with most having some benefit. It is revealed no additional benefit when tamoxifen was continued beyond important to consider the estimated individual risk of recurrence, 5 years, more recent large randomized clinical trials including Adjuvant comorbidity, and personal patient preferences when discussing the Tamoxifen Long Versus Short (ATLAS) and Adjuvant Tamoxifen potential benefits of adjuvant systemic therapy. Treatment, Offer More? (aTTom) demonstrated superiority of 10 years over 5 years.289,290 Ovarian function suppression (OFS) reduces Adjuvant Chemotherapy the mortality rate in women younger than 50 years when compared with no therapy and is similar to the benefit offered by first-generation Who Should Receive Chemotherapy? chemotherapy.291 In addition, recent data demonstrated improvement What tools are available to guide chemotherapy decision making? in DFS when added to adjuvant endocrine therapy, especially in specific Using standard pathologic data, multiple tools are available to clinicians. patient subsets.292,293 Clinical practice guidelines often are used, such as those from NCCN295 However, it is now generally acknowledged that breast cancer is a or ASCO296 in the United States or the St. Gallen International Expert heterogeneous disease. Decisions about whether to consider adjuvant Consensus Panel Meeting in Europe.297 Quantitative tools such as systemic therapy must take into account a combination of prognostic Adjuvant! Online (www.adjuvantonline.com) were developed to help markers of risk (e.g., nodal status, tumor grade, and tumor size), patients and health care providers estimate the potential actual benefit 1588 Part III: Specific Malignancies

Adjuvant Treatment Algorithm

Early-Stage Breast Cancer

HR-positive HR-positive HR-negative HER2-negative HER2-positive HER2-negative

Treatment options: Treatment options: Treatment options: - Endocrine therapy - Endocrine therapy - Chemotherapy3 - ± chemotherapy1 - Chemotherapy - Anti-HER2 therapy2

1Consider evaluation with multiparameter gene expression assay for decision making about chemotherapy 2If >10 mm or node positive; if ≤10 mm, consider chemotherapy + anti-HER2 therapy 3If >10 mm or node positive; if 6−10 mm, consider chemotherapy

Figure 88.22 • Treatment algorithm for systemic therapy for early-stage breast cancer. HR, Hormone receptor. (Modified from Gradishar WJ et al. Breast cancer, version 1. J Natl Compr Canc Netw. 2016;13[12]:1475–1485; and Denduluri N et al. Selection of optimal adjuvant chemotherapy regimens for human epidermal growth factor receptor 2 (HER2)–negative and adjuvant targeted therapy for HER2-positive breast cancers: an American Society of Clinical Oncology guideline adaptation of the Cancer Care Ontario Clinical Practice Guideline. J Clin Oncol. 2016;34[20]:2416–2427.)

from adjuvant systemic therapy.298,299 However, estimates of benefit randomized clinical trials are prospectively testing use of the Oncotype from adjuvant trastuzumab for patients with HER2-positive disease DX assay to guide decision making regarding the addition of chemo- have not yet been incorporated into this model. More recently, the therapy to endocrine therapy. Prospective observational data from PREDICT (http://www.predict.nhs.uk/) tool was developed; it also patients with node-negative disease enrolled on TAILORx who had includes HER2 status and Ki67 expression levels, although most studies RS 0 to 10 demonstrated very low rates of invasive DFS (93.8%) and about the tool published to date have focused on its prognostic use freedom from recurrence of breast cancer at a distant or locoregional rather than its role in prediction of response to therapy.300,301 site (98.7%) at 5 years.311 Multiple gene expression profiles were included in the 2016 ASCO MammaPrint (Agendia), a gene expression assay with prognostic guidelines for use for guiding decisions on chemotherapy in select usefulness based on a 70-gene prognostic signature, became the first populations.199 These include the 21-gene RS Oncotype DX assay,200,201 US Food and Drug Administration (FDA)–cleared in vitro diagnostic the 12-gene EndoPredict assay,302,303 the PAM50 risk of recurrence multivariate assay based on prospective and retrospective analyses.202,203 Prosigna assay,304–306 the Breast Cancer Index,307–309 and the combination It was not recommended in the aforementioned ASCO guidelines of urokinase plasminogen activator and plasminogen activator inhibitor that were published in early 2016,199 although prospective data were type 1 (uPA/PAI-1).310 The strength of each recommendation varies subsequently published from the MINDACT trial examining its role by assay; additional details about a subset of these assays are provided in chemotherapy decision making.207 Patient tumors were characterized subsequently. with both MammaPrint, which dichotomizes tumors into low and For patients with hormone receptor–positive, HER2-negative, high risk based on the genomic signature, and Adjuvant! Online, node-negative breast cancer, gene expression profiling with Oncotype which uses standard clinicopathologic characteristics. Patients whose DX was validated with a prospective and retrospective approach as a tumor classification was discordant between the two approaches were tool to identify the predictive benefit from adding adjuvant chemo- randomized to receive chemotherapy or not. In the 23% of enrolled therapy to endocrine therapy.200,201 Patients with a low recurrence patients who were at high clinical risk but low genomic risk and who score (0–17) have a less than 10% risk of distant recurrence over 10 did not receive chemotherapy, the 5-year rate of survival without years, assuming treatment with 5 years of tamoxifen, and minimal or distant metastases was 94.7% (95% CI, 92.5–96.2), which met the no benefit from chemotherapy.311 On the other hand, patients with criteria for success in the trial.207 These findings suggested that a subset a high RS (31–100) have a greater than 20% risk of distant recurrence of patients at high risk of recurrence based on clinical characteristics over 10 years despite 5 years of tamoxifen therapy, and obtain substantial may not obtain substantial benefit from treatment with chemotherapy. benefit from the addition of chemotherapy. The most beneficial Studies have explored the combined value of routine markers such treatment approach for patients with intermediate RS (18–30) has as ER, PR, Ki67, and HER2 as a prognostic score (so-called IHC4) not yet been defined. More recently, prospective and retrospective when compared with gene expression profiling,313 but this approach data from patients with lymph node involvement demonstrated similar requires optimal standardization of all these individual parameters.198 trends, although the magnitude of risk of recurrence was greater than Until then, use of commercially available assays will continue to provide for patients with node-negative disease.284,312 additional information about prediction and prognosis to complement The main use of predictive assays such as the Oncotype DX appears standard clinicopathologic assessment. In addition, access to accurate to be in clinical situations in which chemotherapy is being considered, and reliable routine clinicopathologic markers for all patients worldwide as opposed to situations in which it is clear that chemotherapy is or remains critical to ensure the best outcome for all patients. is not indicated. There are no data to support use of the assay results At present, no molecular tools have been developed to guide for selection of individual chemotherapy regimens. Many questions treatment selection for patients with either triple-negative disease or remain to be addressed, including how best to treat patients with HER2-positive disease, which tend to be more aggressive tumors with node-negative disease and intermediate RS, and how best to use results higher risk of recurrence. Consequently, decision algorithms for these for patients with node-positive disease. To address these questions, breast cancer subtypes are exclusively based on clinicopathologic factors the TAILORx (NCT00310180) and RxPONDER (NCT01272037) such as tumor size, nodal involvement, and tumor grade. Studies are Cancer of the Breast • CHAPTER 88 1589 ongoing to test new molecular assays based on DNA methylation, anthracycline and often a taxane, are expected to offer higher benefit immune markers, and other gene expression signatures. At present, in patients with more chemoresponsive disease (e.g., hormone receptor adjuvant chemotherapy is routinely offered to all triple-negative negative) but also are associated with greater toxicity.317–319 It is worth node-negative breast cancer patients with at least T1b tumors, despite noting that the administration of taxane-containing regimens in lymph the fact that most patients with triple-negative, node-negative disease node–positive, early-stage breast cancer appears to be particularly useful remain disease free in the long term when treated with locoregional in those with hormone receptor–negative320 and HER2-positive321 therapy alone.314 Similarly, chemotherapy plus trastuzumab is offered disease. However, it is important to note that these analyses were to most patients with at least T1c node negative tumors, and to many retrospective in nature. with T1b tumors, because of the inability to accurately assess the Patients with lower-risk disease often are offered potentially more likelihood of disease recurrence without treatment for individual tolerable first- or second-generation regimens such as TC (docetaxel patients.315,316 plus cyclophosphamide),322 classic CMF (cyclophosphamide, methotrex- ate, and 5-fluorouracil), AC (doxorubicin plus cyclophosphamide), Chemotherapy Regimens or sometimes just endocrine therapy if they have no nodal involvement For patients who are candidates for adjuvant chemotherapy, treatment and their tumors are strongly ER positive. regimens for early-stage breast cancer have evolved over time because There is debate about which patients obtain benefit from anthracycline- of the conduct of multiple large randomized controlled trials. Poly- based chemotherapy. The ABC trials compared anthracycline-based chemotherapy regimens are now considered standard practice over doxorubicin, cyclophosphamide and taxane-containing regimens with monotherapy regimens, and the EBCTCG has confirmed the improved the non–anthracycline-containing docetaxel-cyclophosphamide recurrence and survival outcome observed with anthracycline-based regimen.323 These investigators were unable to demonstrate noninferior- regimens (Table 88.8).281,282,296 However, the magnitude of this survival ity of the nonanthracycline regimen, so the trial was reported early benefit is modest, especially for patients with lower-risk disease. For for futility; longer-term follow-up is planned. In summary, the reason an individual patient, it remains challenging to determine the benefit why some individual patients may derive more benefit from from treatment with different chemotherapeutic regimens. anthracycline-containing regimens remains uncertain, and no clear In view of the potential toxicities and costs, the actual absolute predictors of benefit from anthracyclines have yet been identified to benefit offered by individual therapies must be considered. Predic- guide treatment selection.296 Table 88.9 lists examples of commonly tive factors of benefit (e.g., hormone receptor and HER2 status) and used adjuvant chemotherapy regimens. prognostic factors of risk (e.g., size, nodal status, and tumor grade) must be discussed with each patient, along with potential short- and Adjuvant Therapy for Triple-Negative Breast Cancer long-term toxicities, preferences, and comorbidities, as part of a shared In triple-negative breast cancer (TNBC), neither antiendocrine therapy decision model. More intensive combination regimens, usually with an nor anti-HER2 therapy is effective. Therefore at present, cytotoxic

Table 88.8 Overall Survival Improvements With Adjuvant Systemic Therapy

PROPORTIONAL EFFECT ON ANNUAL BREAST 15-YEAR BREAST CANCER MORTALITY WITH TREATMENT CANCER MORTALITY RATE (RISK [%] AND ABSOLUTE GAIN) VERSUS CORRESPONDING RISK (TREATMENT VS CONTROL) WITHOUT TREATMENT (M) M = 12.5 (E.G., LOW RISK, M = 25 (E.G., M = 50 (E.G., NODE NEGATIVE) NODE NEGATIVE) NODE POSITIVE) Systemic Adjuvant Treatment Ratio of Proportional and Age at Diagnosis (Years) Rates (R) Reduction Risk Gain Risk Gain Risk Gain CHEMOTHERAPY ONLY IN ER-POOR OR ER-POSITIVE DISEASEa None (any age) 1.0 — 12.5 — 25.0 — 50.0 — Anthracycline (age <50) 0.62 38% 7.9 4.6 16.3 8.7 34.9 15.1 Anthracycline (50–69) 0.80 20% 10.1 2.4 20.6 4.4 42.6 7.4 Anthracycline (≥70) ? ? ? ? ? ? ? ? ENDOCRINE, OR CHEMOENDOCRINE, THERAPY IN ER-POSITIVE DISEASEa None (any age) 1.0 — 12.5 — 25.0 — 50.0 — Tamoxifen (any age) 0.69 31% 8.8 3.7 18.0 7.0 38.0 12.0 Anthracycline + tamoxifen (age <50) 0.62 × 0.69 57% 5.6 6.9 11.6 13.4 25.7 24.3 Anthracycline + tamoxifen (50–69) 0.80 × 0.69 45% 7.1 5.4 14.7 10.3 31.8 18.2 Anthracycline + tamoxifen (≥70) ? × 0.69 ? ? ? ? ? ? ?

Anthracycline: About 6 months of anthracycline-based adjuvant chemotherapy with regimens such as FAC or FEC, as in the reviewed trials. Tamoxifen: About 5 years of adjuvant tamoxifen. The 15-year survival probability with treatment is calculated as (1 − M/100) to the power R. aFor women of given nodal status the 5-year mortality is greater for ER-poor than for ER-positive disease, but the 15-year risks may be similar, as may the 15-year benefits of anthracycline-based chemotherapy (because the age-specific breast cancer mortality ratios for anthracycline-based versus no chemotherapy do not depend significantly on ER status). Combination of the direct and indirect randomized evidence yields breast cancer death rate ratios (treatment versus control) of 0.62 (standard error [SE] 0.05) at younger than age 50 years and 0.80 (SE, 0.04) at age 50–69 years for allocation to anthracycline and 0.69 (SE, 0.03) for allocation to tamoxifen. (Allowance for any inappropriate noncompliance with the treatment allocations in these trials would, in expectation, further reduce breast cancer mortality.) ER, Estrogen receptor; FAC, fluorouracil, doxorubicin (Adriamycin), and cyclophosphamide; FEC, 5-fluorouracil, epirubicin, and cyclophosphamide. From Early Breast Cancer Trialists’ Collaborative Group (EBCTCG). Effects of chemotherapy and hormonal therapy for early breast cancer on recurrence and 15-year survival: an overview of the randomised trials. Lancet. 2005;365:1687–1717. 1590 Part III: Specific Malignancies

HER2-positive tumors compared with HER2-negative tumors of the Table 88.9 Examples of Commonly Used Adjuvant same size and nodal status.326 Chemotherapy Regimens for Early- Multiple HER2-targeted therapies have been developed or are in Stage Breast Cancer development. The first therapeutic agent was trastuzumab, a humanized NON–TRASTUZUMAB-BASED REGIMENS monoclonal antibody directed toward the extracellular domain of HER2 that acts to block HER2 activity through several mechanisms.74 AC × 4 (doxorubicin + cyclophosphamide) every 2a or 3 wk followed by A few decades later a second monoclonal antibody, pertuzumab, was paclitaxel × 4 every 2 wka or weekly × 12 a developed; it also binds the extracellular domain of HER2, but at a AC × 4 (doxorubicin + cyclophosphamide) every 2 or 3 wk followed by separate site.327 Pertuzumab prevents homodimerization and heterodi- docetaxel × 4 every 3 wk merization of HER2 with other members of the EGFR family and TAC × 6 (docetaxel + doxorubicin + cyclophosphamide)a b inhibits signaling downstream of HER2. In addition to these antibody- TC × 4 (docetaxel + cyclophosphamide) × 4 based therapeutics, the small-molecule tyrosine kinase inhibitors FAC/CAF (fluorouracil doxorubicin cyclophosphamide) + + lapatinib and neratinib have been developed; they target HER2 and FEC/CEF (cyclophosphamide + epirubicin + fluorouracil) 328 FEC (or FAC) × 4 followed by paclitaxel weekly × 12 other members of the HER family. CMF (cyclophosphamide + methotrexate + fluorouracil) Several studies now show that anti-HER2 therapy is effective in the adjuvant setting for patients with HER2-positive disease, defined TRASTUZUMAB-BASED REGIMENS as 3+ based on immunohistochemical staining or fluorescence in situ 296 AC × 4 followed by weekly paclitaxel plus concurrent trastuzumabc × hybridization (FISH) ratio of 2.0 or greater. Five international, 12 wk followed by trastuzumabc to complete 1 year of anti-HER2 prospective randomized clinical trials have demonstrated that treatment therapy with adjuvant trastuzumab for 1 year reduces the risk of recurrence TCH (docetaxel + carboplatin + trastuzumab)c × 6 followed by and mortality by one-half and one-third, respectively, in patients with trastuzumabc to complete 1 year of anti-HER2 therapy early-stage breast cancer compared with the same chemotherapy regimen TH (paclitaxel + trastuzumab) weekly × 12 followed by trastuzumab to without anti-HER2 therapy.329–334 Additional studies investigating the complete 1 year of anti-HER2 therapy optimal duration of trastuzumab therapy demonstrated that 6 months of therapy was inferior to 1 year,335 and that 2 years of therapy was aWith colony-stimulating factor support. not more effective than 1 year.336 Therefore the standard-of-care bConsider colony-stimulating factor support. cConsider adding pertuzumab. treatment in the adjuvant setting is chemotherapy in combination Modified from NCCN Clinical Practice Guidelines in Oncology. Breast Cancer with 1 year of trastuzumab therapy. No data currently exist to support Version 2.2016. www.nccn.org/professionals/physician_gls/PDF/breast.pdf; and giving trastuzumab without chemotherapy for treatment of lower risk Denduluri N et al. Selection of optimal adjuvant chemotherapy regimens for human disease. Because of concerns about overtreatment of patients, a single- epidermal growth factor receptor 2 (HER2)–negative and adjuvant targeted therapy arm trial of paclitaxel and trastuzumab weekly for 12 weeks followed for HER2-positive breast cancers: an American Society of Clinical Oncology guideline by 9 months of trastuzumab monotherapy was conducted in patients adaptation of the Cancer Care Ontario clinical practice guideline. J Clin Oncol. 2016;34:2416–2427. with tumor size of 3.0 cm or smaller and node-negative, HER2-positive disease.337 With 4 years of follow-up, 3-year rate of survival free from invasive disease was demonstrated to be 98.7% (95% CI, 97.6–99.8), which is substantially better than the rate of recurrence in historical controls. Based on its approval in the metastatic setting,338 lapatinib was chemotherapy regimens are the mainstay of treatment. However, tested with trastuzumab in the adjuvant setting in patients with HER2- although TNBC is generally chemotherapy sensitive, recurrence rates positive, lymph node–positive disease. Initial results in the preoperative are high, so the addition of other agents has been tested to try to systemic therapy setting were promising, with increased rates of pCR make treatment more effective. with dual targeting of HER2.339–341 However, results of the combination In particular, data from randomized phase II trials suggest that in the adjuvant ALTTO trial demonstrated no improvement in DFS the addition of platinums to standard regimens may increase response or OS for dual targeting of HER2 plus chemotherapy compared to therapy, especially for patients with tumors that are deficient in with trastuzumab plus chemotherapy without lapatinib.342 Therefore DNA repair.576,577 For example, CALGB C40603 investigated doxo- lapatinib is not indicated for treatment of early-stage breast cancer. rubicin and cyclophosphamide followed by paclitaxel with or without More recently, 12 months of adjuvant neratinib after completion of carboplatin in the preoperative setting.576 The addition of carboplatin adjuvant trastuzumab was shown to increase 2-year invasive DFS from resulted in a 13% improvement in pCR in the breast and axilla, from 91.6% for placebo to 93.9% for neratinib (stratified HR, 0.67 [95% 41% to 54%. Similar findings were noted in the GeparSixto trial.324 CI, 0.50–0.91]; P = .0091), and is FDA approved in this setting.343 However, no improvement in OS with the addition of a platinum More recently, pertuzumab was studied in combination with has been identified. trastuzumab in the neoadjuvant setting. In the Neosphere study, The CREATE-X randomized clinical trial investigated the use of the addition of pertuzumab to trastuzumab plus docetaxel led to an capecitabine in patients who were treated with primary systemic therapy increase in pCR rate compared with trastuzumab plus docetaxel, from using standard chemotherapy and who had residual disease at the 29.0%. to 45.8%.344 These promising results led to provisional FDA time of surgery.325 Two-year DFS was 87.3% for the capecitabine-treated approval of pertuzumab in the neoadjuvant setting. The confirmatory patients, compared with 80.5% for those who did not receive additional adjuvant trial (APHINITY, NCT01358877) is currently underway, therapy (HR, 0.69 [98.7% CI 0.48–0.99]; logrank P = .001). Additional although results are not yet available. Interesting to note, the pCR research is underway to investigate benefit from the use of other rate for trastuzumab and pertuzumab alone without chemotherapy chemotherapy and targeted agents on disease outcomes in breast cancer, was 16.8%. and TNBC in particular. Trastuzumab therapy is associated with a risk of cardiac toxicity, especially among older patients and those previously exposed to an Adjuvant Anti-HER2 Therapy anthracycline.345,346 In the adjuvant setting, the risk of severe congestive HER2 is amplified or overexpressed in approximately 18% to 20% heart failure may reach 4%.347 Lower rates of cardiotoxicity were of breast cancers. HER2 overexpression is a negative prognostic factor, identified in the non–anthracycline-containing arm in the BCIRG-006 but is also predictive of response to anti-HER2 therapies. Even for trial.333 Although the majority of patients have partial recovery of T1 tumors, rates of recurrence are substantially greater for heart function after discontinuation of trastuzumab, concerns have Cancer of the Breast • CHAPTER 88 1591 been raised about the likelihood of full recovery based on long-term of the bulk of the tumor could lead to increased risk of recurrence. data from BCIRG-006.348 Late onset of cardiac dysfunction from However, initial data did not show a survival advantage for either trastuzumab after discontinuation of treatment appears uncommon, approach.355,356 however.345,348 More recent data have demonstrated that pathologic response These results confirm that HER2 is a useful marker for decision (especially if complete) after preoperative chemotherapy correlates making for patients with breast cancer. Treatment with an anti-HER2 with improved DFS or OS, especially for triple-negative and HER2- agent is recommended for all patients with HER2-positive breast positive breast cancer. Therefore the initial response to preoperative cancer that is node positive or has a tumor size at least 1.0 cm, and systemic therapy offers the potential of in vivo assessment of sensitivity should be considered for treatment of tumors smaller than 1 cm.349 or resistance to treatment.357–360 Of great interest, the identification These findings also emphasize the importance of having an accurate of intermediate (surrogate) markers, including imaging-, tissue-, and and reliable HER2 assay, and confirm the critical need to assess HER2 blood-based markers, that correlate with disease outcome, such as tumor expression in all patients with a new diagnosis of invasive breast pathologic response or survival, in carefully selected patient groups cancer. A false-positive finding could result in exposure of patients to may have clinical usefulness, especially those that can identify early the toxicity of anti-HER2 therapy without the potential for benefit, response or resistance to therapy.361 whereas a false-negative result would deny a patient treatment with Also as a result of these findings, there has been increasing interest a potentially useful drug. Because of prior excessive testing inaccuracy, in leveraging the preoperative setting to conduct clinical trials of new joint ASCO and College of American Pathologists guidelines were therapies with surrogate end points that are faster to complete and developed to guide appropriate testing and interpretation of the results require smaller sample sizes compared with adjuvant trials. Administra- of HER2 assays.195 tion of systemic therapy after surgery has been the gold standard for Interesting to note, retrospective data from some of the original examining the effect of new treatment strategies on disease end points adjuvant trastuzumab trials suggest that a DFS and OS benefit from in the adjuvant setting. Traditionally, trials have required large sample trastuzumab is also observed in patients tested locally as having HER2- sizes and long-term follow-up to achieve the number of events required positive disease but centrally as having HER2-negative disease.350 This to assess adjuvant end points such as DFS and OS. This problem is raises questions about heterogeneity of HER2 expression within compounded by the increasing number of patients diagnosed with tumors.351 Of note, measures of ER and HER2 expression were only earlier stages of disease. minimally discordant when test results between core biopsy and There are potential downsides to preoperative therapy, however, definitive surgical specimens in otherwise unselected patients were because it could lead to the loss of baseline parameters (e.g., tumor compared.352 The potential benefit of anti-HER2 therapy for patients size and lymph node involvement) that generally are used to guide with tumors defined as being HER2 IHC 1+ or IHC 2+/FISH-negative recommendations for adjuvant therapy. Results from NSABP B-27355 is now being prospectively tested in the NSABP B-47 randomized confirm the feasibility of performing SLN biopsy after preoperative trial (NCT01275677). At this time, currently available Level I evidence chemotherapy. In contrast, in the ACOSOG Z1071 clinical study from metastatic trials does not support a role for anti-HER2 therapy of patients with clinically positive nodes before preoperative chemo- in HER2-negative breast cancer.353 therapy, the false-negative rate of SLN biopsy after chemotherapy was unacceptably high (12.5%).362 Similarly, in the SENTINA trial Preoperative Systemic Therapy the false-negative rate was 14.2% in all patients, although for those Although chemotherapy was traditionally administered after surgery, with three or more lymph nodes removed it was 7.3% or lower.363 preoperative systemic therapy, also referred to as neoadjuvant therapy, Therefore this issue remains controversial. In addition, many radiation with either chemotherapy or endocrine therapy is increasingly also oncologists have wanted to know the axillary status at presentation to an option. Preoperative systemic therapy was traditionally reserved guide use of postmastectomy radiation therapy and/or regional nodal for women with locally advanced breast cancer (LABC) to enhance irradiation, although as described earlier, there is increasing interest in the likelihood of negative surgical margins. It is also used for down- potentially using preoperative systemic therapy to reduce the extent staging, to increase the likelihood of breast conservation in select of necessary locoregional therapy and tailor adjuvant radiotherapy women who are not initially lumpectomy candidates. However, the decisions. absolute increase in the number of women ultimately treated with In terms of systemic therapies, patients with hormone receptor– breast conservation after preoperative systemic therapy is small. In a positive disease will all receive 5 to 10 years of endocrine therapy after large clinical trial, preoperative chemotherapy therapy increased the completion of local therapy, regardless of pathologic findings. Similarly, proportion of patients eligible for breast-conserving therapy from 41% those with HER2-positive disease will receive a total of 1 year of to 64%, although only 77% of eligible patients chose to undergo anti-HER2 therapy. However, there is uncertainty about what to do lumpectomy.354 for patients with ER-negative, HER2-negative disease with residual There is growing interest in preoperative systemic therapy to reduce disease at surgery; as described later, ongoing clinical trials are investigat- the extent of needed locoregional treatments, not only by allowing ing the use of additional systemic therapy after surgery in this setting. for certain patients who would have required mastectomy to conserve Interesting to note, there is now the potential for different treatment their breasts but also by potentially downstaging axillary disease to regimens to be administered in the preoperative versus the postoperative allow for omission of ALND or postmastectomy radiotherapy. The setting, at least for short periods of time. One example is for treatment optimal radiotherapeutic management of patients who have pCR to of HER2-positive breast cancer, wherein the FDA provided conditional preoperative systemic therapy in the axilla is an area of ongoing approval for pertuzumab specifically in the preoperative setting based investigation; for patients with residual disease, studies are exploring on data from multiple primary systemic therapy trials.344,364,365 Final whether radiation therapy to the axilla can obviate the need for complete approval of the drug will be based on the results of an adjuvant trial ALND. However, it is important to note that the standard of care that is still underway. Although initially after the FDA’s action per- remains postmastectomy radiation therapy for patients with residual tuzumab could be given only in the preoperative setting, the NCCN axillary involvement after neoadjuvant therapy. subsequently added a footnote to its guidelines recommending con- Initially there were concerns about the use of preoperative systemic sideration of the addition of pertuzumab to adjuvant chemotherapy therapy. On one hand, preclinical data suggested that the administra- in the setting in which a patient would have met criteria for treatment tion of systemic therapy before surgery may be associated with early in the preoperative setting. eradication of micrometastases and improved long-term outcomes by Preoperative systemic therapy might allow trials with smaller sample decreasing the risk of drug resistance and leading to more favorable sizes and adaptive designs to test new strategies that can then quickly growth kinetics. Conversely, there was concern that delaying removal evolve into larger trials with survival as the primary end point. One 1592 Part III: Specific Malignancies

example is the Investigation of Serial Studies to Predict Your Therapeutic Tamoxifen Response Through Imaging and Molecular Analysis 2 (I-SPY 2) SERMs, such as tamoxifen and raloxifene, may function as ER agonists, platform trial, which uses adaptive randomization to examine the antagonists, or mixed agonist-antagonists, depending on the target efficacy, based on pCR, of multiple drugs compared with a single tissue. Tamoxifen has been in use for decades and is approved in the control backbone regimen in patients with high-risk breast cancer.366 United States for risk reduction in high-risk women, for reduction in Each patient is assigned to one of multiple competing regimens based the risk of invasive breast cancer after breast conservation in women on her tumor’s biomarker subtype. with DCIS, in the adjuvant treatment of hormone receptor–positive The findings for two drugs in I-SPY 2 have been reported to disease, and in the management of advanced disease. Important to date. In patients with TNBC, the pCR rate for the PARP inhibitor note, tamoxifen can be used to treat both premenopausal and post- veliparib plus carboplatin and paclitaxel in patients with TNBC was menopausal women. greater than for paclitaxel alone for 12 weeks.366 Similarly, in patients Data from the 2005 EBCTCG meta-analysis showed that 5 years with HER2-positive disease, the pCR rate for paclitaxel plus neratinib of adjuvant tamoxifen for ER-positive disease reduced the annual was greater than for paclitaxel plus trastuzumab for 12 weeks; all breast cancer death rate by 31% regardless of age or chemotherapy patients had also received doxorubicin plus cyclophosphamide for four use.376 Duration of tamoxifen therapy is important. Assuming treatment cycles prior to surgery.367 Both drugs are considered to have “gradu- for a total of 5 years, the annual breast cancer mortality rates are ated” to phase III testing in those particular subsets of patients with similar during years 0 to 4 and 5 to 14, with a cumulative reduction breast cancer. in mortality twice as large at 15 years as after 5 years since diagnosis, As mentioned earlier, in the NeoALTTO trial preoperative trastu- demonstrating a carryover effect following completion of therapy. zumab combined with lapatinib was shown to result in a higher rate No predictive markers of response to tamoxifen have been validated. of pCR compared with trastuzumab alone.339 However, although pCR Some data suggest that host factors, such as polymorphisms of the has been shown to correlate with improved disease outcomes in patients cytochrome P450 CYP2D6 gene, may identify patients who are poor with HER2-positive breast cancer, improvement in this intermediate metabolizers (*4/*4 genotype variant) and have reduced levels of the end point in the NeoALTTO trial failed to translate to a DFS or OS active tamoxifen metabolite endoxifen; the clinical usefulness of this benefit in the definitive large ALTTO clinical trial.342 Therefore the information is uncertain.578–580 usefulness of neoadjuvant trials is now in question. Originally, data supported the use of adjuvant tamoxifen for 5 Optimal patient selection for preoperative chemotherapy is key, years.377,378 However, two large international trials, ATLAS and aTTom, especially if the goal is tumor shrinkage (to allow breast conservation) demonstrated superiority of 10 years of tamoxifen compared with 5. and pCR (for prognostic purposes). Hormone receptor status and In the multinational ATLAS trial, women with early-stage hormone tumor histology should be considered carefully, because patients with receptor–positive breast cancer who had been treated with 5 years of ER-positive disease and lobular histologic type are less likely to respond adjuvant tamoxifen were randomized to take either tamoxifen or to preoperative chemotherapy, even though their long-term outcome placebo for an additional 5 years.290 Compared with placebo, extended (especially among the small minority of patients who achieve a pCR) tamoxifen therapy reduced the risk of recurrence (18% versus 21%; is improved overall.368 P = .002), breast cancer mortality (9.7% versus 11.6%; P = .01), and overall mortality (19% versus 21%; P = .01), although the risk of Adjuvant Endocrine Therapy toxicity, including venous thromboembolic disease and uterine cancer, Adjuvant endocrine therapy arguably is the most effective targeted persisted during the subsequent 5 years. Most of the improvement in therapy in women with early-stage, ER-positive breast cancer, regardless disease outcomes from extended tamoxifen was noted once 10 years of age or nodal status. Surgical hormonal manipulation (e.g., oopho- of treatment was complete. No differences in benefit were noted based rectomy, adrenalectomy, and hypophysectomy) in metastatic disease on stage, age at diagnosis, or menopausal status. Overall, these data was considered the first example of targeted antitumor therapy, and support the use of extended adjuvant tamoxifen. has largely been supplanted by pharmacologic approaches with SERMs such as tamoxifen and toremifene, AIs, the selective estrogen receptor Ovarian function suppression degrader (SERD) fulvestrant, and ovary-suppressing luteinizing The ovary is the primary site of estrogen production in premenopausal hormone–releasing hormone (LHRH) agonists, alone or in combina- women. OFS is effective when compared with no therapy in women tion. For postmenopausal women, both tamoxifen and the AIs have younger than age 50 years.372,376 Similar outcomes are seen with OFS been shown to be active agents in the adjuvant setting.369 versus CMF chemotherapy379–381 and with OFS plus tamoxifen versus Although breast cancer is primarily a disease of older women, up chemotherapy.382,383 At this time, though, data comparing OFS with to 25% of all patients newly diagnosed with invasive disease are younger contemporary chemotherapy regimens containing anthracyclines and than age 50 years, half of whom have hormone receptor–positive taxanes are lacking. Clinical benefit was seen in young premenopausal disease. Either tamoxifen alone or ovarian suppression/ablation plus women who received OFS following chemotherapy,323,384,385 but it endocrine therapy is standard of care for premenopausal women in was unclear if it was of added benefit in this group of women when the adjuvant setting.67,369–371 For those who decline or cannot tolerate they were also treated with endocrine therapy. Indirect evidence from endocrine therapy, OFS is a reasonable alternative.372 IBCSG trial 13-93 suggested that some degree of benefit from The survival benefit of tamoxifen in premenopausal women with chemotherapy-induced amenorrhea might exist even when tamoxifen hormone receptor–positive breast cancer was not fully recognized was also given.386 A 2007 meta-analysis of OFS with LHRH agonists until the mid-1990s. Factors that confounded therapeutic decisions in premenopausal women with ER-positive disease showed that for younger women included age-related chemotherapy effects on chemically induced ovarian suppression offered a significant reduction ovarian function and its indirect endocrine effect and the belief that in the relative risk for recurrence and death when added to tamoxifen, chemotherapy was more active than endocrine therapy in this popula- chemotherapy, or both.387 tion. Retrospective data show that very young premenopausal women The question of combined endocrine therapy for premenopausal (age <35 years) have a higher likelihood of worse disease outcomes women with OFS plus tamoxifen or OFS plus an AI was further when treated with chemotherapy alone,373 especially those with investigated in a number of clinical trials. Initially, the Austrian Breast hormone-sensitive cancer, a finding that may be due to a higher and Colorectal Cancer Study Group [ABCSG]-12 trial, which used likelihood of retaining ovarian function after chemotherapy.374,375 a 2 × 2 factorial design comparing 3 years of treatment with the AI Additional data from prospective randomized trials are now available anastrozole versus tamoxifen and treatment with zoledronic acid versus to guide treatment decision making for premenopausal women, as placebo, was conducted in 1803 premenopausal women who received outlined subsequently. concomitant OFS.388 Important to note, the majority of patients were Cancer of the Breast • CHAPTER 88 1593 not treated with chemotherapy. After a median follow-up of 62 months, exemestane,398 and after a median follow-up of 4.1 years, no differences there was no difference in DFS between those treated with tamoxifen were noted for DFS or OS. alone versus anastrozole alone (HR, 1.08 [95% CI, 0.81–1.44]; P = AIs are contraindicated for treatment in premenopausal women .59), although OS was worse in the anastrozole-treated patients (HR, without concomitant ovarian suppression or ablation. In addition, 1.75 [95% CI, 1.08–2.83]; P = .02). they must be used with caution in women who develop amenorrhea More recently, results from two large trials, the Suppression of after chemotherapy, because those women may have unrecognized Ovarian Function Trial (SOFT; IBCSG 24-02) and the Tamoxifen residual ovarian function or could regain ovarian function, which and Exemestane Trial (TEXT), were published.292,293 In SOFT, would render the AI therapy ineffective.399 In the presence of an 3066 premenopausal women who had regained biochemically AI, the reduced feedback of estrogen to the hypothalamus and confirmed ovarian function after chemotherapy or who did not pituitary may lead to an increase in the secretion of gonadotrophins receive chemotherapy were randomized to tamoxifen alone, tamoxi- and resumption of ovarian function, including the possibility of fen plus OFS, or exemestane plus OFS. In the TEXT trial, 2672 pregnancy.400 There is evidence from the MA.17 trial, however, that premenopausal women were treated with OFS and randomized to premenopausal women who become postmenopausal while on adjuvant tamoxifen or to exemestane at the time of initiation of adjuvant tamoxifen are good candidates for late administration of adjuvant therapy. letrozole.401 In the initial analysis of the SOFT trial, the difference in 5-year DFS between those patients treated with tamoxifen versus those treated Schedule and duration of adjuvant endocrine therapy with tamoxifen plus OFS was not statistically significant (84.7% versus The optimal schedule and duration of endocrine therapy remain 86.6%; HR, 0.83 [95% CI, 0.66–1.04]; P = .10), although there was undefined. ER-positive breast cancer has a long natural history, and a greater benefit in 5-year freedom from breast cancer in those who half of the residual risk of recurrence occurs after the first 5 years.290,402 received chemotherapy (78% versus 82.5%; HR, 0.78 [95% CI, Risks are higher in patients with higher-stage disease at diagnosis. 0.60–1.02]).293 However, in the joint analysis of the two trials, after Potential strategies for which there are data include the use of different 68 months of follow-up, patients treated with exemestane plus OFS sequences of tamoxifen and AIs, and the use of longer durations of had superior DFS (91.1%) compared with those treated with tamoxifen therapy, as described earlier. plus OFS (87.3%; HR, 0.72 [95% CI, 0.60–0.85]; P < .001).292 In a large meta-analysis from the EBCTCG that included patients Important to note, most of the benefit was identified in very young treated with up to 5 years of adjuvant endocrine therapy, AIs were women (<35 years of age) and those with higher-risk disease who had demonstrated to reduce risk of recurrence of breast cancer by about received chemotherapy. 30% compared with tamoxifen, and to reduce risk of breast cancer mortality by about 15%.283 When comparing patients receiving 5 years Aromatase inhibitors of tamoxifen versus 2 to 3 years of tamoxifen followed by AI therapy In postmenopausal women, estrogens are primarily generated through for 2 to 3 years, recurrence rates were statistically significantly lower for the conversion of androgens in peripheral tissues. AIs can block the those patients who switched therapy (RR, 0.56 [95% CI, 0.46–0.67]). action of the aromatase enzyme and reduce circulating estrogen In contrast, less difference in recurrence rates was noted for those concentrations up to 10-fold.389 First-generation inhibitors (e.g., treated with 5 years of an AI versus 2 to 3 years of tamoxifen followed aminoglutethimide) were associated with significant toxicity in the by AI therapy for 2 to 3 years (RR, 0.90 [95% CI, 0.81–0.99]; P = adjuvant setting because of nonspecific effects. Currently used third- .045). Subgroup analysis revealed that the benefit from AI therapy generation drugs include steroidal nonreversible inhibitors (e.g., was independent of age, stage, tumor grade, and PR and HER2 exemestane) and nonsteroidal reversible inhibitors (e.g., letrozole and expression. anastrozole). The more specific, and less toxic, third-generation AIs In sum, evidence-based guidelines from ASCO support the were assessed in several adjuvant trials in postmenopausal women administration of an AI at some point during the adjuvant treatment using different treatment strategies. The Arimidex, Tamoxifen, Alone of postmenopausal women with early-stage hormone-responsive breast or in Combination [ATAC]390 and Breast International Group [BIG] cancer.369 For patients with higher-risk disease, including premenopausal 1-98391 trials tested up-front therapy with an AI versus tamoxifen. women receiving concomitant OFS, the absolute difference in benefit Multiple trials, including BIG 1-98, the Intergroup Exemestane Study from treatment with an AI versus tamoxifen is substantial. There are (IES),392 and TEAM,393 evaluated a sequential approach with a switch no clinically important differences among the three commercially to an AI after 2 to 3 years of tamoxifen. Finally, trials including available AIs, and no predictors of benefit or toxicity have been identified MA.17394 and NSABP B-33395 tested extended therapy with an AI that can be used to select one AI over another for an individual patient. following 5 years of tamoxifen. In all cases, treatment with the AI Safety data support the use of an AI for up to 10 years as monotherapy medication either up-front or after 2 to 3 years of tamoxifen resulted or used sequentially after tamoxifen. in superior DFS compared with tamoxifen monotherapy.396 In addition, research is underway to identify molecular predictors More recently, data for treatment with an AI for 10 years versus of delayed recurrence of breast cancer. Evidence is available for a 5 were published (MA.17R).397 All patients had received a total of number of markers that may be both prognostic for late recurrence 4.5 to 6 years of letrozole, completed within 2 years before enrollment, and predictive of benefit from extended adjuvant therapy, including and more than 70% had received 4.5 to 6 years of tamoxifen before the Breast Cancer Index, although use of these markers for clinical initiating AI therapy. The 5-year DFS for letrozole-treated patients care is not currently recommended.199,309 For now, the decision regarding was 95%, compared with 91% for those who received placebo (P = which treatments to use, in which order, and for which duration, .01). Important to note, the annual incidence rate of contralateral should be made by the physician and patient after careful consideration breast cancer was lower in those treated with letrozole compared with of benefits and risks. placebo (0.21% versus 0.49%; P = 0.007). There was no difference in OS after a median follow-up of 6.3 years. Combined chemoendocrine therapy Are there differences in efficacy among the individual AI medica- Adjuvant endocrine therapy has a more favorable therapeutic index tions? Data demonstrate increased suppression of estrogens by letrozole than chemotherapy in patients with endocrine-responsive breast cancer, compared with the other two AI medications. However, few head- and endocrine therapy is considered the primary component of any to-head comparisons between AIs have been performed examining adjuvant systemic regimen in patients with hormone receptor–positive disease outcomes, and cross-trial comparisons suggest similar benefits disease, regardless of age. However, as described earlier, patients who and toxicity profiles for all AI medications. In the MA.27 clinical also have prognostic factors associated with an increased risk of recur- trial, postmenopausal women were randomized to anastrozole or rence and death may benefit from the addition of chemotherapy to 1594 Part III: Specific Malignancies

endocrine therapy as part of the adjuvant regimen. For most treatment Secondary acute myeloid leukemia and myelodysplastic syndrome regimens that include treatment with both chemotherapy and endocrine have been linked to exposure to alkylating agents, topoisomerase II therapy, it is recommended that chemotherapy be completed before inhibitors, and antimetabolites.417,418 The risk of acute myeloid leukemia initiation of endocrine therapy. This recommendation is primarily or myelodysplastic syndrome after four cycles with conventional doses based on the results of SWOG 8814 (Intergroup 0100), which of doxorubicin and cyclophosphamide (60 and 600 mg/m2, respectively) demonstrated a survival advantage associated with sequential versus at 5 years is at least 0.21%, and cumulative incidence is 0.48% by concurrent administration of chemotherapy and tamoxifen.403 A second 10 years. Age has been associated with increased risk of developing trial (GEICAM 9401) compared epirubicin and cyclophosphamide myeloproliferative neoplasms. with sequential versus concurrent tamoxifen, and showed a trend Other potential toxicities are not life-threatening but can negatively favoring a sequential approach.404 Based on these results, chemotherapy affect health-related quality of life. CIPN is a long-term toxicity caused and endocrine therapy are generally not given simultaneously for by taxane and platinum therapy that can persist in a subset of patients treatment of endocrine-sensitive breast cancer. and affect function. Despite considerable research, no agents to prevent CIPN or to allow patients to regain sensation have yet been identified.419 Preoperative endocrine therapy However, data from randomized phase III trials are now available to Some patients may be appropriate candidates for preoperative therapy support the use of medications, including duloxetine, for treatment yet are unlikely to respond well to cytotoxic chemotherapy. In this of painful neuropathy.581 setting, preoperative endocrine therapy is an attractive alternative to Symptoms of fatigue, insomnia, weight gain, and cognitive dysfunc- chemotherapy, especially in older patients with strong ER-positive tion have been reported with some frequency after the completion of tumors who may not be good candidates for chemotherapy or patients chemotherapy.420,421 Both short- and long-term cognitive dysfunction with strongly hormone receptor–positive, low-grade tumors. However, has been reported, including poor memory, impaired concentration, clinical responses are slower to occur and are rarely complete. Optimal and language deficits.422 Confirmation of an association with chemo- markers to predict long-term outcome after preoperative endocrine therapy has been challenging, because of both limitations with objective therapy are lacking, but expression of tumor Ki67 after short-term assessment of the symptom and confounding by factors such as onset endocrine treatment may help predict response to therapy by combining of menopause and use of adjuvant endocrine therapy. Fatigue may the prognostic value of Ki67 level at baseline with observed changes also affect cognitive function. Although data to support the use of in levels at 2 weeks.405–407 pharmacologic interventions for fatigue and insomnia are lacking, nonpharmacologic approaches such as exercise and cognitive behavioral Secondary Effects of Adjuvant Systemic Therapy therapy have been shown to be beneficial.423 Long-term survivors Secondary effects of chemotherapy report more frequent physical and menopausal symptoms than do As more patients with earlier stages of disease are offered adjuvant healthy women, but their health-related quality of life and sexual systemic therapy, its benefits must be carefully examined, taking into functioning are comparable to those reported by healthy, age-matched consideration the potential short- and long-term toxicity. The che- control subjects.424 Still, the various long-term needs of breast cancer motherapy regimens commonly administered to patients with breast survivors and the crucial role to be played by both oncologists in the cancer are generally well tolerated. Acute life-threatening toxicity in short run and gynecologists and internal medicine specialists in the patients receiving therapy is quite rare. Common short-term effects long run cannot be overemphasized.424,425 include alopecia, gastrointestinal symptoms, myelosuppression, and Young women are at risk for premature menopause after adjuvant febrile neutropenia or neutropenic infection. Chemotherapy-induced chemotherapy, and the risk correlates with both age and adjuvant peripheral neuropathy (CIPN), arthralgia, myalgia, capillary leak chemotherapy regimen (classic CMF for six cycles > AC followed by syndrome, and skin and nail changes also are seen with taxanes. Use paclitaxel = TAC with docetaxel > AC for four cycles).375,426 Temporary of growth factors to stimulate white blood cell production has reduced cessation of menses during adjuvant therapy also correlates with earlier the risk of febrile neutropenia and neutropenic infection.408 However, occurrence of menopause.427 Although ovarian suppression as a result because of concerns about a potential worse survival outcome with of chemotherapy may further decrease the odds of recurrence and the use of erythropoietin-stimulating agents, these agents are generally death in premenopausal women with endocrine-responsive disease, no longer used in the adjuvant setting, where the intent is cure.409 premature menopause can have a significant effect on quality of life Although many women accept these short-term toxicities, greater because of severe hot flashes and vaginal dryness. Hot flashes resulting attention is being paid to long-term and late effects from systemic from discontinuation of menopausal hormonal therapy or loss of therapy because the number of patients who will become long-term ovarian function are a common symptom after chemotherapy, although survivors is increasing. Long-term effects are those that occur during nonhormonal options are now available, as described later.428 For treatment and do not resolve, whereas late effects are those that develop women who retain fertility, pregnancy does not appear to increase many years after completion of therapy. For some, the actual absolute the risk of another breast cancer event.429 Data support the use of reduction in risk of recurrence and death from treatment is relatively LHRH agonists during chemotherapy to preserve fertility in women small compared with the potential risk of late toxicity. with hormone receptor–negative breast cancer.430 A few rare long-term or late effects of therapy are serious or life- Side effects from LHRH agonists occur earlier but are short-lasting threatening. One example is anthracycline-related cardiac toxicity. when compared with chemotherapy, and may include hot flashes, Approximately 8% of women may have asymptomatic systolic dysfunc- decreased libido, mood lability, and potential cardiovascular and skeletal- tion 10 years after receiving doses of doxorubicin around 300 mg/m2, related sequelae such as bone loss. In the SOFT trial, the addition of and cardioprotective drugs such as dexrazoxane are not approved for OFS to endocrine therapy increased the rate of side effects compared use in the adjuvant setting.410 Although the risk of clinical congestive with tamoxifen alone.431 However, in the groups treated with OFS heart failure associated with conventional doses of anthracyclines is small plus either tamoxifen or AI therapy, although the side effect profiles (≤1%),411 this risk is increased in patients subsequently treated with were different and reflected the specific adjuvant endocrine therapy, adjuvant trastuzumab.345,412 Older age, lower left ventricular ejection overall patient-reported health-related quality of life was similar.292 fraction at baseline, and presence of multiple cardiac comorbidities One potentially long-lasting effect of chemotherapy-induced ovarian appear to increase this risk, but good predictive models and biomarkers failure or OFS is the negative impact on bone mineral density (BMD), are not yet sufficiently refined for use in clinical practice.413 In addition, which can increase the risk of osteoporosis and fracture.432 NCCN other than a single nucleotide polymorphism in an intergenic region guidelines derived from the Bone Health Task Force recommend that was recently identified in a genome-wide association study,414 baseline and on-treatment monitoring of BMD, with advice on lifestyle genetic predictors of increased risk have not yet been validated.415,416 interventions, calcium and vitamin D supplementation, and Cancer of the Breast • CHAPTER 88 1595 antiosteoclast therapy initiation depending on the dual-energy x-ray reported side effects. However, vaginal dryness is challenging to manage absorptiometry (DXA)–derived T-score or Fracture Risk Assessment because of the concern about potential systemic absorption of intra- Tool (FRAX) 10-year risk of fracture.433 vaginal estrogen, which could negate the benefits from AI therapy.400 In AI-treated women, recently presented data suggest that intravaginal Secondary effects of endocrine therapy dehydroepiandrosterone (DHEA) may be effective for sexual function,449 Tamoxifen is associated with an increase in BMD in the axial skeleton and topical lidocaine may improve penetrative dyspareunia.450 Unfavor- and with stabilization in the appendicular skeleton in postmenopausal able changes in lipid profile including decreases in high-density women.434 Although tamoxifen in premenopausal women leads to lipoprotein also have been noted,451,452 although a large meta-analysis bone mineral loss in the lumbar spine and hip, the NSABP P-1 study did not demonstrate an increase in cardiovascular events.453 showed a 19% reduction in fractures of the hip, radius, and spine One provocative study reported that patients who experienced in all age groups with tamoxifen compared with placebo, especially treatment-emergent symptoms, either AIMSS or hot flashes, within among those aged 50 years and older.145 Tamoxifen is also associated the first 3 months of AI therapy had improved disease outcomes.454 with a reduction in low-density-lipoprotein cholesterol, and indi- Subsequent analyses of other large trials were mixed, with some reporting vidual studies have suggested that tamoxifen might reduce the risk of similar findings for some or all symptoms,455–458 and others reporting coronary heart disease. However, the EBCTCG meta-analysis did not no association.459,460 These analyses were limited by the lack of informa- show any statistically significant effect of cardiovascular events from tion about either baseline symptoms or symptom severity, and therefore tamoxifen on OS.67 it remains unknown how to apply these findings clinically. Similar to its antiestrogen effects in breast tissue, the toxicity profile of tamoxifen is also related to its tissue-specific effects on ER. There Long-Term Follow-Up are a few key serious but rare tamoxifen-related toxicities: uterine malignancies, venous thromboembolic disease, and ocular toxicity. A number of evidence-based guidelines have been published describing Both the NSABP P-1 prevention trial and the 1995 EBCTCG meta- appropriate follow-up for breast cancer survivors.425,461 Breast cancer analysis showed a higher incidence of uterine malignancies with survivors should undergo evaluation with history and physical examina- tamoxifen treatment.145,290 Although uterine cancer is likely to be tion every 3 to 12 months, depending on time since diagnosis, and diagnosed at an early stage and cured with surgery alone, longer annual mammography. Available evidence does not support use of any duration of tamoxifen therapy is associated with worse histology and surveillance blood tests (e.g., complete blood counts, liver function higher grade,435 and 2% to 5% of all cases represent uterine sarcoma.436 tests, tumor markers) or imaging studies other than mammography Important to note, tamoxifen is expected to cause endometrial thickness, in patients with no signs or symptoms concerning for recurrence.425,461 although neither transvaginal ultrasound nor endometrial biopsy is a Depending on the presence of long-term or late toxicities, patients useful screening tool in an asymptomatic patient.437,438 Instead, women should be referred to mental health providers, rehabilitation specialists, who are taking tamoxifen should undergo routine gynecologic evaluation cardiologists, and endocrinologists, as indicated. Cancer specialists and notify their physicians of abnormal vaginal spotting or bleeding should ensure adequate coordination of care in a shared care model or pelvic pain. Thromboembolic complications caused by tamoxifen, with gynecologists and primary care physicians, who will have an including pulmonary embolus and deep vein thrombosis, occur in increasingly important role in the long-term care of these breast cancer fewer than 1% of patients. Risk factors are poorly defined but include survivors (Table 88.10). In addition to surveillance for possible late obesity, smoking, and possibly factor V Leiden mutation, although complications resulting from local and systemic therapies given as part results have been mixed.439,440 Ocular toxicity, including retinopathy of their breast cancer treatment,462 breast cancer survivors also require and cataract formation, is also uncommon. age-appropriate routine health maintenance care. In many settings, Tamoxifen is more commonly associated with bothersome side noncancer specialists may assume the primary long-term follow-up effects, including hot flashes, night sweats, leg cramps, decreased libido, care of these patients,463 but it is critical that good communication and vaginal discharge; a statistically significant negative impact on be maintained with cancer specialists, especially for patients with quality of life has not been noted.441 The NSABP prevention P-1 hormone receptor–positive disease who receive endocrine therapy for trial showed no detrimental effects on quality of life, mood, or sexual at least 10 years. Accreditation bodies now require survivorship care function and no increase in weight gain in the tamoxifen-treated plans to facilitate communication,464 although minimal data exist to group.442 Selective serotonin reuptake inhibitors and serotonin norepi- demonstrate the effectiveness of these plans for improving patient care.465 nephrine reuptake inhibitors such as citalopram and venlafaxine, and antiepileptics such as gabapentin and pregabalin, have been shown to New Strategies in Adjuvant Treatment be effective for managing hot flashes.582–585 Data to support the use of clonidine are mixed, however,443 and popular herbal supplements such Greater understanding of the biologic subtypes of breast cancer is as black cohosh and evening primrose oil are ineffective.444 helping with the selection of treatment strategies targeting specific In contrast, AIs act by lowering circulating estrogen concentrations, tumor subtypes. ER, PR, and HER2 remain the most useful markers which is a different mechanism compared with tamoxifen, and therefore at present, although, as noted previously, gene expression profiles are have a different toxicity profile. The primary serious toxicity is loss being tested prospectively as prognostic and predictive markers to ensure of bone density and concomitant increased risk of fracture. Women the optimal identification of the patients most likely to benefit from a considered for AI therapy should have a baseline bone density and specific strategy, including treatment with adjuvant chemotherapy and/ vitamin D assessment, and ensure adequate daily intake of calcium or the use of extended endocrine therapy. For patients with HER2- and vitamin D. Concurrent therapy with an antiosteoclast inhibitor positive disease, adjuvant trials are in progress to examine the use of can ameliorate this bone loss.445,446 Raloxifene should not be used in additional anti-HER2 agents, including pertuzumab, ado-trastuzumab AI-treated women to increase bone density based on the data from emtansine, and neratinib, in conjunction with trastuzumab. the ATAC trial, which noted an increased risk of recurrence when As noted earlier, whether or not a patient achieves a pCR with anastrozole and tamoxifen were given concurrently compared with primary systemic therapy is predictive of long-term disease outcomes, anastrozole alone.447 especially in those patients with TNBC or HER2-positive disease. Multiple bothersome toxicities have been reported. The most Therefore a number of clinical trials are ongoing to evaluate the use common toxicities are aromatase inhibitor–associated musculoskeletal of additional cytotoxic agents or novel targeted drugs in patients with symptoms (AIMSS), including arthralgias, myalgias, joint stiffness, residual disease at the time of surgery. One trial that was recently carpal tunnel syndrome, and tendinopathies.448 Other menopausal reported was the randomized phase III CREATE-X trial, which symptoms, including hot flashes and vaginal dryness, are also commonly demonstrated a benefit from the addition of capecitabine in patients 1596 Part III: Specific Malignancies

distant recurrence. In the absence of distant disease, local in-breast Table 88.10 Guidelines for Surveillance of failure after lumpectomy and radiotherapy is typically managed with Asymptomatic Early Breast Cancer mastectomy. For select patients (e.g., women older than 70 with Survivors From the American Cancer hormone receptor–positive disease) repeat lumpectomy may be reason- Society/American Society of Clinical able. For the rare woman treated with lumpectomy alone who experi- Oncology and the National ences an in-breast recurrence, repeat lumpectomy followed by Comprehensive Cancer Network radiotherapy is an option. For the more common case of a woman treated with lumpectomy and adjuvant radiotherapy, the safety and Recommended Patient education about signs and symptoms efficacy of repeat breast conservation with partial breast re-irradiation of recurrence is under investigation. For patients who experience a chest wall recur- History and physical examination every 3–12 rence after mastectomy, wide excision with negative margins followed months for first 5 years as clinically by irradiation (if postmastectomy radiation therapy was not part of appropriate, and annually thereafter the primary treatment) should be performed. For those who experience Evaluation for genetic counseling chest wall recurrence after prior radiation therapy, consideration may Annual mammography (if lumpectomy and be given to reirradiation in combination with hyperthermia. In pooled adjuvant radiation therapy) data from two randomized studies—EORTC 10801 and the Danish Annual gynecologic follow-up if on tamoxifen Counseling regarding adherence to Breast Cancer Cooperative Group (DBCG) 82TM trial—that compared endocrine therapy, if applicable breast conservation therapy and radiation therapy versus mastectomy, both therapies were associated with a similar risk of local failure and Assessment and management of long-term 473 and late effects of cancer and treatment OS following salvage treatment. Coordination of care among providers The decision to use additional multidrug chemotherapy or hormone Not recommended for Routine laboratory testing (complete blood therapy is based on the nature of the breast recurrence, including the detection of disease count, chemistry panel) ER, PR, and HER2 expression status, and on whether the patient recurrence (in the Tumor markers (e.g., CEA, CA27.29, CA15-3) previously received adjuvant chemotherapy. In the CALOR trial, absence of suggestive Bone or positron emission tomography scans patients with completely excised isolated locoregional recurrence were signs or symptoms) Chest radiograph shown to have improved 5-year DFS when they received additional Computed tomography scan of chest, chemotherapy compared with no chemotherapy (69% versus 57%; 474 abdomen, pelvis, and brain P = .046). These findings were stronger for patients with hormone Liver ultrasound receptor–negative breast cancer. Breast magnetic resonance imaging (unless meets high-risk criteria for screening) MANAGEMENT OF METASTATIC DISEASE Modified from NCCN Clinical Practice Guidelines in Oncology: Breast Cancer Because most patients with metastatic breast cancer ultimately die of Version 2.2016. www.nccn.org/professionals/physician_gls/PDF/breast.pdf; and their disease, the primary goal of therapy is palliation of symptoms Runowicz CD et al. American Cancer Society/American Society of Clinical Oncology and prolongation of life. At the same time, there has been an improve- breast cancer survivorship care guideline. J Clin Oncol. 2016;34:611–635. ment in the survival of patients with metastatic disease over the last few decades as a result of more effective therapies.475 Historically, with TNBC with residual disease after primary systemic therapy, as median survival of patients after the diagnosis of metastatic breast described earlier.325 Other trials using this strategy are currently in cancer was 2 to 3 years. However, some patients survive long term, progress, including trials to evaluate targeted therapies such as PARP and a very small number of patients with “oligometastatic” disease inhibitors and mTOR inhibitors, and immunotherapies such as PD-L1 may even benefit from multimodality therapy that includes surgical inhibitors (NCT02954874, NCT02032823). resection of an isolated visceral metastasis with curative intent.476,477 Much attention has also been directed toward the study of Also, specific therapies such as trastuzumab have dramatically changed bisphosphonates as adjuvant therapy to reduce the risk of breast the natural history of HER2-positive metastatic breast cancer for a cancer recurrence, distinct from their use to support bone health. subset of patients.478 Although initial studies of adjuvant clodronate demonstrated a survival Approximately 75% of metastases occur within the first 5 years benefit and reduction in bony recurrences,466–468 a larger randomized after the diagnosis of early-stage disease, especially among patients NSABP trial of adjuvant clodronate versus placebo failed to show any with hormone receptor–negative disease. Unfortunately, patients overall benefit.469 Mixed results have also been observed with adjuvant with hormone receptor–positive disease have a significant residual zoledronic acid.388,470 In the ABCSG-12 trial, in which premenopausal risk of recurrence beyond the first 5 years, and metastases have been women treated with goserelin were randomized to zoledronic acid or no documented as late as 20 to 30 years after the initial diagnosis.402 therapy for 3 years, treatment with the bisphosphonate improved DFS Although most patients with metastatic disease are expected to experi- (HR, 0.68 [95% CI, 0.51–0.91]; P = .009) but had no impact on OS.388 ence progression of disease at some point, certain clinical and tumor More recently, a large meta-analysis demonstrated an improvement in characteristics are useful in predicting prognosis. Patients with a long breast cancer recurrence in bone with bisphosphonate therapy in women interval since initial diagnosis, excellent performance status, hormone who had been postmenopausal for at least 5 years.471 SWOG S0307, receptor–positive disease that primarily involves bone or soft tissue, which compared three different bisphosphonates (oral clodronate, oral and only a few sites of visceral involvement are likely to have a better ibandronate, and intravenous zoledronic acid) in the adjuvant setting, long-term prognosis. In addition, some patients with HER2-positive revealed no difference among the three treatment regimens.472 Therefore disease can also have prolonged survival, even after the diagnosis and although the data support consideration of adjuvant bisphosphonate treatment of brain metastases.478,479 Available locoregional, systemic, therapy for postmenopausal women at risk of breast cancer recurrence, and supportive care treatments can result in significant regression the optimal drug and schedule remain uncertain. of disease, relief of symptoms, and, in some cases, prolongation of survival. Although the goal of treatment of metastatic breast cancer Recurrence After Breast Conservation Therapy seldom is cure, palliation with improved quality of life can be achieved in many patients. Patients who experience a local recurrence after breast-conserving Preliminary evidence indicates a role for combined multimodality therapy or mastectomy should undergo imaging to rule out a concurrent therapy in patients with small-volume metastatic disease.480,481 If Cancer of the Breast • CHAPTER 88 1597 confirmed, this could have significant implications for the current for disease monitoring in a subset of patients, and a baseline value recommendations for no surveillance in the absence of specific can be useful for assessing trends.199 Imaging studies, including bone symptoms.425 Previous exposure to adjuvant therapy predicts a lower scintigraphy and contrast computed tomography (CT) scan, fluorine-18 response to first-line chemotherapy in patients with metastatic breast fluorodeoxyglucose–positron emission tomography (FDG-PET)/CT, cancer. However, retrospective data suggest that patients who have a and/or plain radiographs, provide a baseline for the evaluation of recurrence long after completing adjuvant therapy may respond to response to the planned treatment modality. Skeletal scintigraphy similar regimens.482 IBCSG data also suggest that quality-of-life scores (bone scan) remains a reasonable option to screen for bone metastasis, may correlate with outcome in metastatic breast cancer,483 and improve- although scintigraphic flare response can make the interpretation of ment in symptoms such as pain and shortness of breath may correlate changes with treatment challenging.489,490 Other metabolic studies with greater response to therapy.484 such as FDG-PET may have a role, especially when integrated with Prompt initiation of supportive measures and specific anticancer conventional CT imaging for anatomic information, although it remains therapy in patients with significant symptoms or life-threatening unclear whether FDG-PET can replace bone scintigraphy scan,491 and complications (e.g., spinal cord compression, destructive bone lesions cost remains a barrier. Recent reports also suggest an apparent increase in weight-bearing areas, hypercalcemia, and symptomatic pleural or in the prevalence of central nervous system metastases in breast cancer, pericardial effusions and ascites) can offer significant palliation of especially in patients with HER2-positive disease.492 This increase is symptoms. Management of oncologic problems common to cancer in great part a result of improved control of systemic disease and lack and its therapy is described in detail elsewhere in this book. of penetration of antibody therapy in sanctuary sites.

Evaluation of Suspected Metastases Endocrine Therapy At the time of recurrence, many patients have nonspecific symptoms, Endocrine therapy is often very effective with minimal associated such as new pain, weight loss, or dyspnea. Guidelines now recom- toxicity, and therefore should be considered the primary option over mend tissue acquisition for diagnostic confirmation of recurrent cytotoxic chemotherapy for frontline therapy in patients with bone-only breast cancer and reassessment of biomarker status (ER, PR, and or asymptomatic visceral metastatic hormone receptor–positive HER2) to guide treatment decision making.485 Biopsy is particularly disease.493,494 Patients in visceral crisis, however, should be considered important in the setting of solitary lesions, which may be due to a for frontline multiagent cytotoxic chemotherapy because of its faster disease process other than breast cancer. In addition, recent studies have onset of action. The majority of patients whose tumors coexpress demonstrated considerable biomarker discordance between primary HER2 benefit from HER2-based therapies in the frontline setting, and metastatic breast tumors,486,487 and this can also have substantial but single-agent oral therapy with an endocrine agent remains an implications for treatment. One caveat is patients with primary hormone option for a selected subset of patients.495 Fig. 88.23 lists a treatment receptor–positive disease who are found to have ER-negative disease algorithm for palliative endocrine regimens. at bone biopsy. It is possible that the lack of ER expression is due to a laboratory artifact related to bone decalcification, and therefore Selective Estrogen Receptor Modulators frontline endocrine therapy could still be considered in this context.488 Tamoxifen, 20 mg daily, is a SERM that has been in clinical use for Circulating tumor markers such as CA15-3, CA27.29, and CEA, decades, and was the standard-of-care endocrine therapy for treatment although not diagnostic for breast cancer, can provide useful information of metastatic disease until the advent of AI therapy. Toremifene, which

Treatment Algorithm Metastatic HR+, HER2− Breast Cancer—Postmenopausal1,2

No prior endocrine Prior endocrine therapy therapy >1 year

Prior endocrine therapy <1 year Treatment option: Treatment option: - AI ± palbociclib - AI ± palbociclib - Tamoxifen - AI - AI + fulvestrant - Tamoxifen Treatment option: - AI - Fulvestrant + palbociclib - Tamoxifen Progression (if no palbociclib 1st line) - Megestrol acetate - Exemestane + everolimus - Fluoxymesterone (if progressed on NSAI) - Ethinyl estradiol - Fulvestrant

1Treatment options are similar for premenopausal women with the addition of ovarian suppression or ablation 2Treatment options are similar for men, although when AI is given should add LHRH agonist therapy

Figure 88.23 • Treatment algorithm for treatment of hormone receptor–positive, HER2-negative metastatic breast cancer with palliative endocrine therapy. AI, Aromatase inhibitor; HR, hormone receptor; LHRH, luteinizing hormone–releasing hormone. (Modified from Gradishar WJ et al. Breast cancer, version 1. J Natl Compr Canc Netw. 2016;13[12]:1475–1485; and Rugo HS et al. Endocrine therapy for hormone receptor–positive metastatic breast cancer: American Society of Clinical Oncology guideline. J Clin Oncol. 2016;34[25]:3069–3103.) 1598 Part III: Specific Malignancies

is also available for use, has a similar profile and is cross-resistant with supporting combinations of antiestrogens with other targeted agents tamoxifen. In general, tamoxifen is no longer used in the first-line in the first and later lines of therapy. Examples include CDK4/6 setting, especially for postmenopausal women, and instead is used in inhibitors, inhibitors of the PI3K-AKT-mTOR pathway, and anti- the second-line or later setting. Minimal data are available regarding HER2 agents. Most patients will be treated with all three classes of tamoxifen in the second-line setting, in which approximately 10% of antiestrogen therapy at some point, in addition to at least a subset patients had an objective response and the clinical benefit rate was of the targeted therapies. However, there are few data regarding the almost 50%.496 optimal sequence of therapy, either monotherapy or combination Acquired resistance to tamoxifen is multifactorial and can be due regimens. to a variety of mechanisms, including activation of alternate growth Two studies have evaluated anastrozole alone versus in combination pathways, mutations in ER, alterations in tamoxifen metabolism, with fulvestrant in previously untreated metastatic disease. One trial inadequate concentrations of intracellular tamoxifen, and differen- showed similar outcomes,512 whereas the second showed an improved tial expression of steroid-receptor transcriptional coactivators and PFS favoring the combination, possibly related to prior tamoxifen corepressors.497 As an example, recently there has been a recognition exposure.513 In the latter trial there was no required crossover to that mutations in the ligand-binding domain of ER develop in as fulvestrant following anastrozole monotherapy, so it is unknown whether many as 30% to 40% of patients treated with endocrine therapy in the combination is superior to sequential therapy. the metastatic setting and result in resistance to endocrine therapy, CDK4/6 inhibitors regulate cell cycle progression and arrest 498 514 including tamoxifen and AIs. Interesting to note, very few ER cells in the G1 phase. In patients with ER-positive and HER2- mutations have been identified in primary tumors or after treatment negative disease, data have demonstrated a PFS advantage for adding with adjuvant endocrine therapy. the CDK4/6 inhibitor palbociclib to letrozole for treatment of metastatic disease. In the PALOMA-2 phase III trial, the addition Aromatase Inhibitors of palbociclib to letrozole resulted in a 10-month improvement in In postmenopausal women with no or distant previous exposure to DFS compared with letrozole alone (24.8 versus 14.5 months; P = antiestrogen agents, the AIs show similar or modestly superior efficacy .0004) in postmenopausal women with previously untreated metastatic compared with tamoxifen.499 Therefore these drugs are often used in disease, with an HR for disease progression or death of 0.58 (95% the frontline setting for postmenopausal women, and for premenopausal CI, 0.46–0.72; P < .001).515 No OS benefit has been reported. In women in combination with ovarian suppression or ablation. Although addition, similar results were seen with the combination of a dif- letrozole appears to more potently suppress both total-body aromatiza- ferent CDK4/6 inhibitor, ribociclib, and letrozole compared with tion and plasma estrogen levels than anastrozole,500,501 direct comparison letrozole alone, with an improvement in 18-month PFS from 42.2% as second-line therapy in metastatic breast cancer showed no convincing to 63%.516 In the second-line setting in patients with progression on a clinical advantage of one AI over the other.502 AIs should be avoided nonsteroidal AI, there was an improvement in PFS with the addition as a single agent in men with metastatic disease, because their chronic of palbociclib to fulvestrant, from 4.6 to 9.5 months (P < .0001).517,518 administration may lead to a significant increase in levels of follicle- No predictors of response to therapy have yet been identified. Based stimulating hormone and testosterone without any change in levels on these data, palbociclib received accelerated approval from the FDA of estradiol, but they appear to be effective when combined with an in 2015, and other CDK4/6 inhibitors are also being considered LHRH agonist as chemical castration.503 for approval. Activation of the mTOR pathway may represent a mechanism of Ovarian Ablation resistance to antiestrogens in ER-positive disease. A PFS benefit has The preferred endocrine therapy in premenopausal women with been shown with the addition of the mTOR inhibitor everolimus to endocrine-responsive disease and recent exposure to tamoxifen is AI exemestane in patients with ER-positive and HER2-negative disease therapy plus OFS with surgical or chemical techniques (LHRH that is refractory to a nonsteroidal AI, from 4.1 months with exemestane agonist).504 Radiation ablation is less reliable and technically more alone to 10.6 months with the combination.519 However, no PFS or challenging, and the results are not as immediate.505 Available data OS benefit was observed when the mTOR inhibitor temsirolimus show both an OS advantage and a progression-free survival (PFS) was added to letrozole as first-line therapy in otherwise AI-naïve patients, advantage with the addition of tamoxifen to an LHRH agonist,506 even though approximately half of these patients had prior exposure but there are limited data on OFS and an AI in the metastatic setting.507 to adjuvant tamoxifen.520 As described later, chemotherapy plus anti-HER2–directed therapy Other Antiestrogens is recommended as frontline therapy for most patients with ER-positive The pure antiestrogen fulvestrant downregulates ER and lacks the and HER2-positive metastatic breast cancer.493 However, there are agonistic activity of tamoxifen. The standard dose of fulvestrant is data to guide the use of anti-HER2 therapy plus AI therapy in the 500 mg intramuscularly every 2 weeks for three doses then monthly frontline setting for metastatic disease from at least two large trials.521,522 based on data that demonstrated improved PFS when compared with Guidelines recommend consideration of an endocrine therapy–based the previously recommended dose of 250 mg intramuscularly regimen for select patients, including those with comorbidities that monthly.508 It has also been shown to have a comparable clinical preclude use of HER2-directed therapy, those with minimal disease benefit rate compared with the steroidal AI exemestane in women burden, and those with a prolonged disease-free interval.493 whose ER-positive disease progressed on prior nonsteroidal AI.509 More recently, data from the phase II FIRST trial demonstrated improved Chemotherapy PFS and OS with frontline fulvestrant versus anastrozole (OS, 54.1 versus 48.4 months; P = .04).510 Data reported from the phase III Palliative chemotherapy should be considered for patients with FALCON trial similarly demonstrated a PFS benefit of fulvestrant symptomatic visceral disease, ER- and PR-negative disease, or hormone compared with anastrozole; OS data are not yet mature.511 receptor–positive disease that is resistant to endocrine therapy. Given the palliative goal and need to appropriately balance the benefits and Combination Regimens With Antiestrogens for toxicities of cytotoxic therapy, the challenge for the oncologist is in deciding when to initiate chemotherapy and with which regimen. Postmenopausal Patients Many appropriate chemotherapy regimens are available, including In addition to the option of using antiestrogens as monotherapy, both single-agent and multiagent regimens, and there is little evidence since 2012 new data have been reported regarding combinations of to support use of combination therapy over sequential single-agent antiestrogens. In addition, more recent data have been published chemotherapy (Table 88.11). Cancer of the Breast • CHAPTER 88 1599

The nontaxane microtubule inhibitor eribulin was shown to offer Table 88.11 Commonly Used Cytotoxic an OS benefit when compared with best clinician choice for patients Chemotherapy and Anti-HER2 Therapy with prior exposure to at least two chemotherapy regimens for metastatic Drugs in Metastatic Breast Cancer disease (13.1 versus 10.6 months; P = .041).531 However, in a random- CYTOTOXIC CHEMOTHERAPY ized phase III trial of eribulin versus capecitabine in patients with metastatic disease, 75% of whom had received no or one prior che- Albumin-bound paclitaxel motherapy regimen, there was no difference in the coprimary end Capecitabine points of DFS and OS,532 which leaves uncertainty regarding the Cisplatin optimal use of this drug. Carboplatin As described earlier, anthracyclines are commonly used in the Docetaxel adjuvant setting, but they can also be used for treatment of metastatic Doxorubicin disease. Because of concern for heart failure with cumulative exposure, Epirubicin Eribulin the amount that can be given in the metastatic setting is sometimes Gemcitabine limited. However, there is the option of giving liposomal doxorubicin, which has been shown to have equivalent efficacy to standard doxo- Ixabepilone 533,534 Vinorelbine rubicin in the first-line setting. Capecitabine is an oral prodrug of 5-fluorouracil that is active in ANTI-HER2 THERAPY patients who showed disease progression after previous taxane regi- Ado-trastuzumab emtansine mens.535 In addition to having benefit similar to eribulin in the frontline Lapatinib setting,532 it has also been shown to have similar activity to liposomal Pertuzumab doxorubicin.536 Because it is oral and usually associated with minimal Trastuzumab alopecia and peripheral neuropathy, it is generally a convenient option for patients. Platinum analogues such as cisplatin and carboplatin, which act via cross-linking DNA, interfere with DNA replication. Although it is unclear which subtypes of breast cancer are most sensitive to treatment with platinums, a subset of TNBCs including those associated with There often is a fine line between premature use of chemotherapy BRCA1 mutations appears to be susceptible to the action of platinum in the asymptomatic patient without disease-related complications salts because of defects in the DNA double-strand break repair versus delaying therapy until deterioration of performance status mechanism.537 Although previously published data are mixed, a phase significantly decreases the likelihood of response to or tolerance of II trial of cisplatin for metastatic breast cancer demonstrated a relatively therapy. There is considerable interest in identifying biologic parameters high response rate, especially in patients who harbored BRCA1 or that may predict the success of specific chemotherapy regimens. The BRCA2 germline mutations or whose tumors had a BRCA-like genomic organ distribution of metastases and the patient’s symptoms, history instability signature.537 of exposure to chemotherapy, and general medical condition are helpful considerations in determining the time of initiation of chemotherapy. Combination Chemotherapy Age alone should never be a contraindication to treatment. Similarly, Combination regimens that are administered in the adjuvant setting lack of response or clinical benefit after a few sequential chemotherapy can also be considered for use in the metastatic setting. In addition, (as few as three) regimens and/or a poor performance status of 3 or gemcitabine plus paclitaxel was shown to significantly improve OS worse may identify patients who are unlikely to benefit from further when compared with paclitaxel alone (18.6 versus 15.8 months; P = standard chemotherapy and in whom the primary focus should be .0489).538 Subsequently, a similar study design demonstrated that the on other palliative measures for symptom management.523 combination of ixabepilone plus capecitabine resulted in superior PFS compared with capecitabine alone (5.8 versus 4.2 months), which led Single-Agent Chemotherapy to FDA approval in 2007 of ixabepilone in combination with Commonly used single-agent chemotherapy drugs in patients with capecitabine.527 No OS benefit was identified. However, although advanced disease include microtubule inhibitors, anthracyclines, and these combination regimens were superior to treatment with a single capecitabine. Because no data suggest a markedly strong benefit of agent, it is unknown whether the use of two drugs in combination one class of drugs over another, decisions should be made based on would lead to improved PFS or OS compared with use of the same patient convenience and toxicity profile. Other active agents include two drugs in sequence. In addition, combination regimens are often gemcitabine, 5-fluorouracil, platinum compounds, and etoposide. associated with increased toxicity compared with sequential single Multiple agents act via inhibition of microtubules, including taxanes, agents, which may not be appropriate given the palliative intent of epothilones, vinorelbine, and eribulin. The activity of taxanes in patients the treatment. with anthracycline-resistant disease is well documented524 and may High-dose chemotherapy with autologous or allogeneic stem cell extend to patients previously treated with another taxane. Taxanes support remains investigational. Available data from multiple random- appear to have a better toxicity profile compared with doxorubicin.525 ized trials do not support its use as a standard approach.539 The epothilone B analogue ixabepilone also has clinical activity in patients with disease resistant to paclitaxel, doxorubicin, and HER2-Targeted Therapy capecitabine.526,527 However, although the nanoparticle albumin formula- Anti-HER2 therapy is another example of targeted therapy. Available tion of paclitaxel (nab-paclitaxel) had shown a PFS benefit versus data demonstrate that the potential benefits of trastuzumab are restricted paclitaxel528 and docetaxel529 in the metastatic setting, a trial comparing to patients whose tumors overexpress HER2; current evidence does these agents in the frontline setting in combination with bevacizumab not support its use against HER2-negative disease.540 Single-agent demonstrated that weekly ixabepilone was inferior to paclitaxel, and therapy with trastuzumab results in an objective response rate of 26% there was a trend toward inferiority for weekly nab-paclitaxel as well.530 in the first-line setting for HER2-positive disease.541 Although the Such data suggest that nab-paclitaxel might be of primary interest in initial phase II trial of combination regimens demonstrated a DFS patients with metastatic disease with allergy to paclitaxel or poor toler- and OS benefit when the anti-HER2 monoclonal antibody trastuzumab ance to dexamethasone premedication that is often routinely used was added to anthracycline and cyclophosphamide or paclitaxel, with paclitaxel. excessive cardiac toxicity limits the ability to combine anthracyclines 1600 Part III: Specific Malignancies

with trastuzumab.542 Additional small studies have been conducted therapy, switching from every 4 weeks to every 12 weeks was noninferior to support the use of trastuzumab in combination with other cytotoxic to continuing infusion every 4 weeks (NCT00320710). Similarly, chemotherapies. In addition, there are limited data to support continu- preliminary data from CALGB 70604 demonstrated that administration ation of trastuzumab during subsequent regimens.543 every 12 weeks starting at the time of diagnosis of bone metastases The dual tyrosine kinase inhibitor lapatinib is approved in combina- was noninferior to administration every 4 week.557 These more potent tion with capecitabine for the treatment of patients who previously third-generation bisphosphonates and denosumab are associated with were treated with trastuzumab,338 and improves OS when combined an increased risk of osteonecrosis of the jaw, and long-term safety with paclitaxel versus paclitaxel alone in the frontline setting.544 In data beyond 1 or 2 years are not available. Use of oral supplementation addition, the combination of two HER2-directed agents, trastu- with calcium and vitamin D is recommended, along with a baseline zumab and lapatinib, is superior to lapatinib alone in patients whose preventive dental evaluation before therapy is started. cancer progressed on trastuzumab-based therapy for HER2-positive 545 metastatic breast cancer. In addition to the effect of lapatinib on UNUSUAL PROBLEMS ENCOUNTERED IN extracranial disease, data also suggest that lapatinib in combina- tion with chemotherapy has activity against HER2-positive brain BREAST CANCER 546 metastases. Inflammatory Disease Pertuzumab, a second anti-HER2 monoclonal antibody, has no meaningful clinical activity as a single agent or against HER2-negative Inflammatory breast cancer is the most aggressive form of nonmetastatic disease. It was granted FDA approval based on the results of a phase breast cancer.558 Although it accounts for only 1% to 4% of breast III randomized trial (CLEOPATRA) comparing docetaxel and cancers in the United States, its presentation is striking. Characteristics trastuzumab with and without pertuzumab in the first-line setting, essential to the clinical diagnosis include rapid enlargement and which demonstrated a PFS and OS benefit.327,547 Of note, enrolled generalized induration of the breast, often without an associated mass. patients had received no prior trastuzumab therapy or had received Diffuse skin erythema affecting more than one-third of the breast is it at least 6 months prior to enrollment. OS increased by 15.7 months, the most distinctive clinical feature of the disease. Retraction of the from 40.8 months to 56.5 months (HR, 0.68 [95% CI, 0.56–0.84]; nipple, diffuse breast warmth, and peau d’orange skin changes are P < .001).547 A number of clinical trials are currently underway to commonly observed. Not infrequently, patients are thought to have investigate the use of pertuzumab in later lines of therapy. mastitis and are treated with antibiotics. Ado-trastuzumab emtansine, originally known as T-DM1, is an Pathologically, inflammatory cancer is not a distinct histologic antibody-drug conjugate that links the microtubule inhibitor DM1 entity, although studies are ongoing to try to identify a signature to trastuzumab. It binds HER2 at the same site as trastuzumab and specific for inflammatory breast cancer.558 Inflammatory breast cancer delivers the cytotoxic agent to the HER2-overexpressing cancer cells.548 can be hormone receptor–positive or –negative, and HER2 positive or T-DM1 has been shown to have activity in trastuzumab-refractory negative. At present, dermal lymphatic involvement is the pathologic HER2-positive disease, and was shown to provide both PFS and hallmark of the disease. Since the first description by Bryant in 1887, OS benefit when compared with the combination of capecitabine many studies have associated the clinical findings with carcinoma in and lapatinib.548 Similarly, T-DM1 demonstrated an improvement the lymphatics of the skin. Thorough examination of mastectomy in PFS compared with physician’s choice for those with prior specimens confirms dermal lymphatic involvement in as many as 70% trastuzumab and lapatinib therapy.549 However, initial results of a of women with clinical signs of inflammatory carcinoma. phase III trial of first-line treatment with a taxane plus T-DM1 with Dispute exists in the literature about the criteria necessary to pertuzumab or T-DM1 with placebo demonstrated noninferiority of diagnose inflammatory breast cancer. Although some believe that PFS between the T-DM1–containing arms and controls; T-DM1 was clinical findings alone are adequate to make the diagnosis, others associated with less toxicity.550 Therefore for now, the use of docetaxel, argue that a skin biopsy confirming dermal lymphatic involvement trastuzumab, and pertuzumab remains the standard of care for first- is needed. Both presentations share a similar poor outcome. line therapy. Combined modality therapy with primary systemic chemotherapy, modified radical mastectomy, and postmastectomy radiation therapy Therapies Targeting Angiogenesis has become the standard approach to the treatment of inflammatory Although the humanized monoclonal antibody bevacizumab, which breast cancer, similar to the evolution of treatment for other types of targets the vascular endothelial growth factor–A ligand, was originally LABC. Retrospective data are convincing for a survival benefit and granted accelerated approved by the FDA for the treatment of metastatic support the routine use of systemic therapy in this disease. Depending breast cancer, the approval was revoked when confirmatory studies on the choice of chemotherapeutic and locoregional treatment, DFS failed to demonstrate a sufficient PFS benefit given the toxicities, and rates at 5 years generally exceeded 25% to 30%, with 5-year survival also demonstrated no OS benefit. First-line data showed a significant rates approaching 40%.559 increase in PFS when the drug was added to paclitaxel in a randomized 551 first-line trial in women with metastatic disease (ECOG 2100). Male Breast Cancer However, other confirmatory studies failed to show a similar PFS or any OS benefit.552,553 Therefore although bevacizumab is still available Male breast cancer is rare, accounting for 0.2% of male cancers and for treatment of other malignancies, is it not approved for treatment less than 1% of new breast cancers.3 Mutations in the BRCA2 gene of breast cancer in the United States. predispose men to breast cancer and may account for up to 40% of all cases. Most present as infiltrating ductal carcinoma with unilateral, Bisphosphonates firm, painless masses. Nipple discharge should be taken seriously and Bone is the most common site of metastasis in breast cancer.554 Monthly is an indication for FNA or core or excisional biopsy. Mammography injections of antiosteoclast inhibitors, including bisphosphonates and and ultrasound may help to differentiate breast cancer from gyneco- the receptor activator of nuclear factor–κB ligand (RANKL) inhibitor mastia. A negative finding on FNA or core biopsy necessitates an denosumab, for up to 2 years can reduce the risk of skeletal events excision procedure, and cytologic findings that show gynecomastia in patients who have lytic bone metastases and are receiving systemic mandate close follow-up. The tumor phenotype appears similar to therapy.555,556 Zoledronate is an effective alternative to pamidronate that observed in women. because of its shorter infusion time, and denosumab is convenient Treatment is similar to that for female breast cancer. There is because it is a subcutaneous injection. Preliminary data from the evidence that SLN biopsy can be safely applied in male breast cancer.560 OPTIMIZE-2 trial suggest that after 1 year of monthly bisphosphonate Because most patients have endocrine-responsive disease, orchiectomy Cancer of the Breast • CHAPTER 88 1601 or chemical castration with LHRH agonists often is used in patients Paget Disease of the Breast with advanced disease. Tamoxifen is the most common endocrine therapy used, although AIs can also be used in combination with Paget disease of the breast is a less common presentation of breast LHRH agonist therapy.561 cancer, occurring in approximately 1% to 3% of new female breast Male breast cancer accounts for only approximately 1% of all cancer cases. Patients typically have a scaly, raw or excoriated lesion cases of breast cancer, and men with DCIS only are estimated to be of the nipple and areola. The differential diagnosis can include benign 7% of that already small group. Little information exists on the lesions (eczema, dermatitis) and malignant lesions (basal cell carcinoma, management of DCIS in men. Total mastectomy or wide excision Bowen disease, melanoma). A short course of steroids is often attempted, with free margins may be considered a reasonable treatment. but biopsy should be performed on any suspicious or persistent lesion to avoid any delay in diagnosis. This can be done with a punch biopsy; Breast Cancer and Pregnancy often, malignant, intraepithelial adenocarcinoma cells (Paget cells) will be present. Breast Cancer During Pregnancy Paget disease is associated with an underlying in situ or invasive Carcinoma of the breast, although rare in pregnant women, occurs cancer of the breast in 85% to 88% of cases. A thorough breast in about 1 to 3 patients per 10,000 deliveries and is the most common examination and breast imaging are mandatory. Because of the incidence malignancy associated with pregnancy. European data suggest that of occult cancer with Paget disease, MRI is often recommended in the prognosis of breast cancer during pregnancy is similar to that of addition to mammography, although the data are limited. Biopsy breast cancer diagnosed in women who are not pregnant.562 Diagnosis should be performed on any associated palpable or image-detected and staging are more difficult in pregnant women because of physiologic abnormalities. changes in the mother and radiation risk to the fetus. Mammograms Treatment will be guided by the presence or absence of any are not routinely performed because little information can be gained additional masses or mammographic abnormalities. If an in situ or because of pregnancy-related increased breast density. Ultrasound and invasive cancer is detected, both the nipple-areolar complex and the MRI can be used, although based on currently available data, gado- underlying cancer will need to be excised. Although this often requires linium should be avoided. mastectomy, breast-conserving therapy is an option if resection of In terms of local therapy, methylene blue dye should not be used both the cancer and the nipple-areolar complex can be accomplished for SLN biopsy because of risks to the fetus, although radiolabeled (central lumpectomy), followed by definitive breast radiation therapy. sulfur colloid appears to be safe.295,563 Similarly, radiotherapy to the This will often require reduction on the contralateral side along with breast is not performed while the patient is pregnant. Therefore systemic nipple-areolar reconstruction. For patients who have no additional chemotherapy is often given before surgery because a subset of the findings on imaging or physical examination, treatment options include chemotherapeutic agents (doxorubicin and cyclophosphamide) appear mastectomy or resection of the nipple-areolar complex followed by to be able to be given safely after the first trimester, although there whole-breast irradiation. is some debate about the safety of taxanes.563,564 Tamoxifen and trastuzumab are not recommended during pregnancy because of Phyllodes Tumors of the Breast concerns about teratogenesis. Phyllodes tumors are uncommon breast tumors, accounting for less Pregnancy After Breast Cancer than 0.5% of all breast malignancies, that have a diverse range of Many women now maintain fertility after breast cancer treatment, biologic behaviors.571 These can range from benign tumors with a and some will become pregnant.565 There is increasing evidence that propensity to recur locally to a malignant sarcoma that can metastasize. in contrast to women diagnosed with breast cancer during pregnancy, The term phyllodes means leaf-like, and these tumors characteristically those diagnosed with breast cancer during the first few years after have leaf-like papillary projections at pathologic assessment. Although pregnancy have an approximately threefold increased risk of developing in the past these were called cystosarcoma phyllodes, this is no longer metastatic disease and dying from breast cancer.566,567 The biologic an appropriate term because these are not true sarcomas and rarely reasons for this distinction remain unclear. have a cystic component. Phyllodes tumors have characteristics very similar to those of Axillary Metastases With Occult Breast Cancer fibroadenomas; most patients have smooth, well-defined mobile masses. One-fifth of patients will have nonpalpable mammographic abnormali- A woman with clinically suspicious axillary lymph nodes despite ties.572 What typically distinguishes these from fibroadenomas clinically negative breast examination findings and mammogram requires careful is their rapid growth pattern, and phyllodes should be suspected in evaluation for breast cancer. FNA often allows the diagnosis of adeno- any patient with a large or rapidly growing mass. Core needle biopsy carcinoma versus other tumor types. The breast is the most common is required in order to make the diagnosis because FNA is not accurate primary source when dealing with adenocarcinoma, although for differentiating fibroadenomas and phyllodes tumors. gastrointestinal, pulmonary, or thyroid sources should also be con- Histologically, phyllodes tumors are categorized as benign, bor- sidered. Breast imaging with ultrasound and MRI should be performed, derline, or malignant. Benign lesions have circumscribed tumor margins, although if no breast primary is identified, then additional imaging mild to moderate cellular atypia, a low mitotic rate, and a lack of with CT scan of the chest and abdomen should be considered. stromal overgrowth. Phyllodes tumors are described as borderline when The definition of “occult breast carcinoma” is a breast cancer they have a greater degree of atypia and a higher mitotic rate (4–9 manifesting with metastatic axillary nodes without evidence of the mitoses per 10 high-power fields) but still a lack of stromal overgrowth. primary tumor. Occult breast carcinoma manifesting as axillary Malignant tumors are characterized by marked atypia, high mitotic metastases is rare, accounting for less than 0.4% of primary operable rate (>10), and the presence of stromal overgrowth. breast cancer. A large single-institution study of 69 patients with an Complete surgical excision with negative margins is the standard occult primary showed that breast MRI was able to depict a primary of care for phyllodes tumors. Positive margins are associated with high carcinoma in 63% of the patients, and no carcinoma was found in local recurrence rates. Margins of 1 cm are generally recommended the mastectomy specimen when the MRI finding was negative.568 for borderline or malignant tumors, because margins less than 1 cm Such patients should be managed with axillary dissection and may have been associated with a higher recurrence rate.571,573,574 As long as receive either mastectomy of the breast or whole-breast radiation appropriate margins can be obtained with a satisfactory cosmetic therapy for breast conservation; breast conservation should be considered outcome, mastectomy is not necessary. Axillary metastases are extremely because local recurrence rates are very low.569,570 rare, so neither SLN biopsy nor ALND is indicated. 1602 Part III: Specific Malignancies

Radiation therapy is not necessary for benign phyllodes tumors but made, new treatment strategies will be required in both the curative is recommended after excision for borderline or malignant phyllodes and metastatic settings. Research is ongoing to identify effective tumors. In a meta-analysis of eight studies, adjuvant radiation therapy treatments for patients with residual disease after preoperative systemic was associated with a significantly decreased risk of local recurrence (HR, therapy, and those with endocrine-sensitive tumors who are at risk of 0.31 [95% CI, −0.10 to 0.72]) but no impact on OS.575 Hormonal late recurrence of disease. In the advanced-disease setting, there are therapy is not effective, and there are few data regarding the efficacy active areas of research examining both new cytotoxic agents and of chemotherapy for malignant phyllodes tumors. Patients with benign combinations of chemotherapy, targeted therapy, and/or immuno- or borderline phyllodes are cured with local therapy alone, whereas therapies to treat refractory cancers. Finally, with the increasing number patients with malignant phyllodes tumors have a 5-year survival rate of cancer survivors across the globe, it will be important to evaluate between 60% and 80%.572 Patients with large (>5 cm), high-risk symptom management and survivorship strategies to reduce long-term or recurrent malignant phyllodes tumors may be considered for and late effects of therapy and to improve coordination of care for chemotherapy. patients and providers. Overall there has been tremendous progress in both the understanding and management of breast cancer in the FUTURE STRATEGIES past decade, and there is considerable promise for increased knowledge and new treatment approaches in the future. As noted earlier, a greater understanding of the biologic subtypes of breast cancer is helping with the selection of treatment strategies target- The complete reference list is available online at ing specific tumor subtypes. Although considerable progress has been ExpertConsult.com.

KEY REFERENCES

19. Chlebowski RT, Anderson GL, Gass M, et al. 199. Harris LN, Ismaila N, McShane LM, et al. Use open-label, phase 3 non-inferiority trial. Lancet Estrogen plus progestin and breast cancer incidence of biomarkers to guide decisions on adjuvant Oncol. 2014;15(12):1303–1310. and mortality in postmenopausal women. JAMA. systemic therapy for women with early-stage 262. Early Breast Cancer Trialists’ Collaborative Group, 2010;304(15):1684–1692. invasive breast cancer: American society of clinical Darby S, McGale P, et al. Effect of radiotherapy 30. King MC, Marks JH, Mandell JB, New York Breast oncology clinical practice guideline. J Clin Oncol. after breast-conserving surgery on 10-year recurrence Cancer Study G. Breast and ovarian cancer risks 2016;34(10):1134–1150. and 15-year breast cancer death: meta-analysis of due to inherited mutations in BRCA1 and BRCA2. 200. Paik S, Shak S, Tang G, et al. A multigene assay individual patient data for 10,801 women in 17 ran- Science. 2003;302(5645):643–646. to predict recurrence of tamoxifen-treated, node- domised trials. Lancet. 2011;378(9804):1707–1716. 33. National Comprehensive Cancer Network. Genetic/ negative breast cancer. N Engl J Med. 2004;351(27): 263. Hughes KS, Schnaper LA, Bellon JR, et al. Lumpec- Familial High-Risk Assessment: Breast and Ovarian, 2817–2826. tomy plus tamoxifen with or without irradiation version 2.2017. https://www.nccn.org/professionals/ 201. Paik S, Tang G, Shak S, et al. Gene expression and in women age 70 years or older with early breast physician_gls/pdf/genetics_screening.pdf. Accessed benefit of chemotherapy in women with node- cancer: long-term follow-up of CALGB 9343. December 22, 2016. negative, estrogen receptor-positive breast cancer. J Clin Oncol. 2013;31(19):2382–2387. 54. Turner N, Tutt A, Ashworth A. Hallmarks of ‘BRCA- J Clin Oncol. 2006;24(23):3726–3734. 264. Haviland JS, Owen JR, Dewar JA, et al. The ness’ in sporadic cancers. Nat Rev Cancer. 2004; 202. van de Vijver MJ, He YD, van’t Veer LJ, et al. A gene- UK standardisation of breast radiotherapy (START) 4(10):814–819. expression signature as a predictor of survival in breast trials of radiotherapy hypofractionation for treatment 67. Davies C, Godwin J, Gray R, et al. Relevance of cancer. N Engl J Med. 2002;347(25):1999–2009. of early breast cancer: 10-year follow-up results of breast cancer hormone receptors and other factors to 207. Cardoso F, van’t Veer LJ, Bogaerts J, et al. 70-gene sig- two randomised controlled trials. Lancet Oncol. the efficacy of adjuvant tamoxifen: patient-level meta- nature as an aid to treatment decisions in early-stage 2013;14(11):1086–1094. analysis of randomised trials. Lancet. 2011;378(9793): breast cancer. N Engl J Med. 2016;375(8):717–729. 271. Correa C, Harris EE, Leonardi MC, et al. Accelerated 771–784. 214. Wapnir IL, Dignam JJ, Fisher B, et al. Long-term partial breast irradiation: executive summary for the 72. Slamon DJ, Clark GM, Wong SG, Levin WJ, Ullrich outcomes of invasive ipsilateral breast tumor recur- update of an ASTRO evidence-based consensus A, McGuire WL. Human breast cancer: correlation rences after lumpectomy in NSABP B-17 and B-24 statement. Pract Radiat Oncol. 2016;epub Sept 17. of relapse and survival with amplification of the randomized clinical trials for DCIS. J Natl Cancer 272. Whelan TJ, Olivotto IA, Parulekar WR, et al. HER-2/neu oncogene. Science. 1987;235(4785):177– Inst. 2011;103(6):478–488. Regional nodal irradiation in early-stage breast 182. 216. Early Breast Cancer Trialists’ Collaborative Group, cancer. N Engl J Med. 2015;373(4):307–316. 95. Cancer Genome Atlas Network. Comprehensive Correa C, McGale P, et al. Overview of the ran- 278. Early Breast Cancer Trialists’ Collaborative Group, molecular portraits of human breast tumours. Nature. domized trials of radiotherapy in ductal carcinoma McGale P, Taylor C, et al. Effect of radiotherapy 2012;490(7418):61–70. in situ of the breast. J Natl Cancer Inst Monogr. after mastectomy and axillary surgery on 10-year 130. Ignatiadis M, Lee M, Jeffrey SS. Circulating tumor 2010;2010(41):162–177. recurrence and 20-year breast cancer mortality: meta- cells and circulating tumor DNA: challenges and 224. Margolese RG, Cecchini RS, Julian TB, et al. analysis of individual patient data for 8135 women opportunities on the path to clinical utility. Clin Anastrozole versus tamoxifen in postmenopausal in 22 randomised trials. Lancet. 2014;383(9935): Cancer Res. 2015;21(21):4786–4800. women with ductal carcinoma in situ undergoing 2127–2135. 134. Garcia-Murillas I, Schiavon G, Weigelt B, et al. lumpectomy plus radiotherapy (NSABP B-35): 281. Peto R, Davies C, Godwin J, et al. Comparisons Mutation tracking in circulating tumor DNA a randomised, double-blind, phase 3 clinical between different polychemotherapy regimens for predicts relapse in early breast cancer. Sci Transl trial. Lancet. 2016;387(10021):849–856. early breast cancer: meta-analyses of long-term Med. 2015;7(302):302ra133. 250. Boughey JC, Attai DJ, Chen SL, et al. Contralateral outcome among 100,000 women in 123 randomised 145. Fisher B, Costantino JP, Wickerham DL, et al. prophylactic mastectomy consensus statement from trials. Lancet. 2012;379(9814):432–444. Tamoxifen for the prevention of breast cancer: the American society of breast surgeons: additional 283. Early Breast Cancer Trialists’ Collaborative Group, current status of the national surgical adjuvant considerations and a framework for shared decision Dowsett M, Forbes JF, et al. Aromatase inhibitors breast and bowel project P-1 study. J Natl Cancer making. Ann Surg Oncol. 2016;23(10):3106–3111. versus tamoxifen in early breast cancer: patient-level Inst. 2005;97(22):1652–1662. 258. Giuliano AE, McCall L, Beitsch P, et al. Locoregional meta-analysis of the randomised trials. Lancet. 154. Domchek SM, Friebel TM, Singer CF, et al. Associa- recurrence after sentinel lymph node dissection with 2015;386(10001):1341–1352. tion of risk-reducing surgery in BRCA1 or BRCA2 or without axillary dissection in patients with sentinel 290. Davies C, Pan H, Godwin J, et al. Long-term mutation carriers with cancer risk and mortality. lymph node metastases: the American college of effects of continuing adjuvant tamoxifen to 10 JAMA. 2010;304(9):967–975. surgeons oncology group z0011 randomized trial. years versus stopping at 5 years after diagnosis of 166. Oeffinger KC, Fontham ET, Etzioni R, et al. Ann Surg. 2010;252(3):426–432. oestrogen receptor-positive breast cancer: ATLAS, Breast cancer screening for women at average 261. Donker M, van Tienhoven G, Straver ME, et al. a randomised trial. Lancet. 2013;381(9869):805– risk: 2015 guideline update from the American Radiotherapy or surgery of the axilla after a positive 816. cancer society. JAMA. 2015;314(15):1599– sentinel node in breast cancer (EORTC 10981- 292. Pagani O, Regan MM, Walley BA, et al. Adju- 1614. 22023 AMAROS): a randomised, multicentre, vant exemestane with ovarian suppression in Cancer of the Breast • CHAPTER 88 1603

premenopausal breast cancer. N Engl J Med. 2014;371: 2-06; NCCTG [alliance] n063d) comparing one hormone receptor–positive breast cancer: the BIG 107–118. year of anti-HER2 therapy with lapatinib alone (L), 1-98 randomised clinical trial at 8.1 years median 293. Francis PA, Regan MM, Fleming GF, et al. Adjuvant trastuzumab alone (T), their sequence (T→L), or follow-up. Lancet Oncol. 2011;12(12):1101–1108. ovarian suppression in premenopausal breast cancer. their combination (T+L) in the adjuvant treatment 397. Goss PE, Ingle JN, Pritchard KI, et al. Extending N Engl J Med. 2015;372(5):436–446. of Her2-positive early breast cancer (EBC). J Clin aromatase-inhibitor adjuvant therapy to 10 years. 311. Sparano JA, Gray RJ, Makower DF, et al. Prospective Oncol. 2014;32(5S suppl):abstr LBA4. N Engl J Med. 2016;375(3):209–219. validation of a 21-gene expression assay in breast 343. Chan A, Delaloge S, Holmes FA, et al. Neratinib after 425. Runowicz CD, Leach CR, Henry NL, et al. American cancer. N Engl J Med. 2015;373(21):2005–2014. trastuzumab-based adjuvant therapy in patients with cancer Society/American society of clinical oncology 329. Piccart-Gebhart MJ, Procter M, Leyland-Jones B, et al. HER2-positive breast cancer (ExteNET): a multicen- breast cancer survivorship care guideline. J Clin Trastuzumab after adjuvant chemotherapy in HER2- tre, randomised, double-blind, placebo-controlled, Oncol. 2016;34(6):611–635. positive breast cancer. N Engl J Med. 2005;353(16): phase 3 trial. Lancet Oncol. 2016;17(3):367–377. 510. Ellis MJ, Llombart-Cussac A, Feltl D, et al. 1659–1672. 359. Cortazar P, Zhang L, Untch M, et al. Pathological Fulvestrant 500 mg versus anastrozole 1 mg for 332. Perez EA, Romond EH, Suman VJ, et al. Four- complete response and long-term clinical benefit the first-line treatment of advanced breast cancer: year follow-up of trastuzumab plus adjuvant in breast cancer: the CTNeoBC pooled analysis. overall survival analysis from the phase II FIRST chemotherapy for operable human epidermal growth Lancet. 2014;384(9938):164–172. study. J Clin Oncol. 2015;33(32):3781–3787. factor receptor 2-positive breast cancer: joint analysis 362. Boughey JC, Suman VJ, Mittendorf EA, et al. 515. Finn RS, Martin M, Rugo HS, et al. Palbociclib of data from NCCTG n9831 and NSABP B-31. Sentinel lymph node surgery after neoadjuvant and letrozole in advanced breast cancer. N Engl J J Clin Oncol. 2011;29(25):3366–3373. chemotherapy in patients with node-positive breast Med. 2016;375:1925–1936. 333. Slamon D, Eiermann W, Robert N, et al. Adjuvant cancer: the ACOSOG z1071 (Alliance) clinical trial. 547. Swain SM, Baselga J, Kim SB, et al. Pertuzumab, trastuzumab in HER2-positive breast cancer. N Engl JAMA. 2013;310(14):1455–1461. trastuzumab, and docetaxel in HER2-positive J Med. 2011;365(14):1273–1283. 390. Cuzick J, Sestak I, Baum M, et al. Effect of metastatic breast cancer. N Engl J Med. 2015;372(8): 337. Tolaney SM, Barry WT, Dang CT, et al. Adjuvant anastrozole and tamoxifen as adjuvant treatment 724–734. paclitaxel and trastuzumab for node-negative, for early-stage breast cancer: 10-year analysis of the HER2-positive breast cancer. N Engl J Med. ATAC trial. Lancet Oncol. 2010;11(12):1135–1141. 2015;372(2):134–141. 391. Regan MM, Neven P, Giobbie-Hurder A, et al. 342. Piccart-Gebhart MJ, Holmes AP, Baselga J, et al. Assessment of letrozole and tamoxifen alone and First results from the phase III ALTTO trial (BIG in sequence for postmenopausal women with steroid Cancer of the Breast • CHAPTER 88 1603.e1

REFERENCES

1. Bray F, Jemal A, Grey N, Ferlay J, Forman D. estrogens among postmenopausal women with prior unselected for family history: a combined analysis of Global cancer transitions according to the human hysterectomy: a randomized controlled trial. JAMA. 22 studies. Am J Hum Genet. 2003;72(5):1117–1130. development index (2008–2030): a population-based 2011;305(13):1305–1314. 39. Mavaddat N, Peock S, Frost D, et al. Cancer risks study. Lancet Oncol. 2012;13(8):790–801. 21. Beral V, Reeves G, Bull D, Green J, Million Women for BRCA1 and BRCA2 mutation carriers: results 2. Siegel RL, Miller KD, Jemal A. Cancer statistics, Study C. Breast cancer risk in relation to the interval from prospective analysis of EMBRACE. J Natl 2018. CA Cancer J Clin. 2018;68:7–30. between menopause and starting hormone therapy. Cancer Inst. 2013;105(11):812–822. 3. Korde LA, Zujewski JA, Kamin L, et al. Multidis- J Natl Cancer Inst. 2011;103(4):296–305. 40. Narod SA. Modifiers of risk of hereditary breast and ciplinary meeting on male breast cancer: summary 22. Rossouw JE, Anderson GL, Prentice RL, et al. Risks ovarian cancer. Nat Rev Cancer. 2002;2(2):113–123. and research recommendations. J Clin Oncol. and benefits of estrogen plus progestin in healthy 41. Pennington KP, Swisher EM. Hereditary ovarian 2010;28(12):2114–2122. postmenopausal women: principal results from the cancer: beyond the usual suspects. Gynecol Oncol. 4. Ravdin PM, Cronin KA, Howlader N, et al. The Women’s health initiative randomized controlled 2012;124(2):347–353. decrease in breast-cancer incidence in 2003 in trial. JAMA. 2002;288(3):321–333. 42. Goodwin PJ, Phillips KA, West DW, et al. Breast the United States. N Engl J Med. 2007;356(16): 23. Farhat GN, Cummings SR, Chlebowski RT, et al. Sex cancer prognosis in BRCA1 and BRCA2 mutation 1670–1674. hormone levels and risks of estrogen receptor–nega- carriers: an international prospective breast cancer 5. Colditz GA, Hankinson SE. The Nurses’ health tive and estrogen receptor–positive breast cancers. family registry population-based cohort study. J Clin study: lifestyle and health among women. Nat Rev J Natl Cancer Inst. 2011;103(7):562–570. Oncol. 2012;30(1):19–26. Cancer. 2005;5(5):388–396. 24. Calle EE, Kaaks R. Overweight, obesity and cancer: 43. Easton DF, Pharoah PD, Antoniou AC, et al. Gene- 6. Travis RC, Reeves GK, Green J, et al. Gene- epidemiological evidence and proposed mechanisms. panel sequencing and the prediction of breast-cancer environment interactions in 7610 women with breast Nat Rev Cancer. 2004;4(8):579–591. risk. N Engl J Med. 2015;372(23):2243–2257. cancer: prospective evidence from the million women 25. Calle EE, Rodriguez C, Walker-Thurmond K, Thun 44. Desmond A, Kurian AW, Gabree M, et al. Clinical study. Lancet. 2010;375(9732):2143–2151. MJ. Overweight, obesity, and mortality from cancer actionability of multigene panel testing for hereditary 7. Chen WY, Rosner B, Hankinson SE, Colditz GA, in a prospectively studied cohort of U.S. adults. N breast and ovarian cancer risk assessment. JAMA Willett WC. Moderate alcohol consumption during Engl J Med. 2003;348(17):1625–1638. Oncol. 2015;1(7):943–951. adult life, drinking patterns, and breast cancer risk. 26. Hvidtfeldt UA, Gunter MJ, Lange T, et al. Quantify- 45. Kurian AW, Hare EE, Mills MA, et al. Clinical JAMA. 2011;306(17):1884–1890. ing mediating effects of endogenous estrogen and evaluation of a multiple-gene sequencing panel 8. Chlebowski RT, Blackburn GL, Thomson CA, insulin in the relation between obesity, alcohol for hereditary cancer risk assessment. J Clin Oncol. et al. Dietary fat reduction and breast cancer consumption, and breast cancer. Cancer Epidemiol 2014;32(19):2001–2009. outcome: interim efficacy results from the Women’s Biomarkers Prev. 2012;21(7):1203–1212. 46. Ricker C, Culver JO, Lowstuter K, et al. Increased intervention nutrition study. J Natl Cancer Inst. 27. Green J, Cairns BJ, Casabonne D, et al. Height yield of actionable mutations using multi-gene 2006;98(24):1767–1776. and cancer incidence in the million women study: panels to assess hereditary cancer susceptibility in 9. Key TJ. Fruit and vegetables and cancer risk. Br J prospective cohort, and meta-analysis of prospective an ethnically diverse clinical cohort. Cancer Genet. Cancer. 2011;104(1):6–11. studies of height and total cancer risk. Lancet Oncol. 2016;209(4):130–137. 10. Linos E, Willett WC. Diet and breast cancer risk 2011;12(8):785–794. 47. Domchek SM, Bradbury A, Garber JE, Offit K, reduction. J Natl Compr Canc Netw. 2007;5(8): 28. Collins LC, Baer HJ, Tamimi RM, Connolly JL, Robson ME. Multiplex genetic testing for cancer 711–718. Colditz GA, Schnitt SJ. Magnitude and laterality susceptibility: out on the high wire without a net? 11. Pierce JP, Natarajan L, Caan BJ, et al. Influence of breast cancer risk according to histologic type of J Clin Oncol. 2013;31(10):1267–1270. of a diet very high in vegetables, fruit, and fiber atypical hyperplasia: results from the Nurses’ health 48. Robson ME, Bradbury AR, Arun B, et al. American and low in fat on prognosis following treatment for study. Cancer. 2007;109(2):180–187. society of clinical oncology policy statement update: breast cancer: the Women’s healthy eating and living 29. Marshall LM, Hunter DJ, Connolly JL, et al. Risk of genetic and genomic testing for cancer susceptibility. (WHEL) randomized trial. JAMA. 2007;298(3): breast cancer associated with atypical hyperplasia of J Clin Oncol. 2015;33(31):3660–3667. 289–298. lobular and ductal types. Cancer Epidemiol Biomark- 49. Beatson GT. On the treatment of inoperable cases 12. Dong JY, Qin LQ. Soy isoflavones consumption ers Prev. 1997;6(5):297–301. of carcinoma of the mamma: suggestions for a new and risk of breast cancer incidence or recurrence: 30. King MC, Marks JH, Mandell JB, New York Breast method of treatment with illustrative cases. Lancet. a meta-analysis of prospective studies. Breast Cancer Cancer Study G. Breast and ovarian cancer risks 1896;2:104. Res Treat. 2011;125(2):315–323. due to inherited mutations in BRCA1 and BRCA2. 50. Metzger Filho O, Giobbie-Hurder A, Mallon 13. Wu AH, Yu MC, Tseng CC, Pike MC. Epidemiology Science. 2003;302(5645):643–646. E, et al. Relative effectiveness of letrozole com- of soy exposures and breast cancer risk. Br J Cancer. 31. Antoniou AC, Easton DF. Models of genetic sus- pared with tamoxifen for patients with lobular 2008;98(1):9–14. ceptibility to breast cancer. Oncogene. 2006;25 (43): carcinoma in the BIG 1-98 trial. J Clin Oncol. 14. Chung M, Lee J, Terasawa T, Lau J, Trikalinos TA. 5898–5905. 2015;33(25):2772–2779. Vitamin D with or without calcium supplementation 32. Tung N, Domchek SM, Stadler Z, et al. 51. Cristofanilli M, Gonzalez-Angulo A, Sneige N, et al. for prevention of cancer and fractures: an updated Counselling framework for moderate-penetrance Invasive lobular carcinoma classic type: response to meta-analysis for the US preventive services task cancer-susceptibility mutations. Nat Rev Clin Oncol. primary chemotherapy and survival outcomes. J Clin force. Ann Intern Med. 2011;155(12):827–838. 2016;13(9):581–588. Oncol. 2005;23(1):41–48. 15. Bauer SR, Hankinson SE, Bertone-Johnson ER, 33. National Comprehensive Cancer Network. Genetic/ 52. Dieci MV, Orvieto E, Dominici M, Conte P, Ding EL. Plasma vitamin D levels, menopause, Familial High-Risk Assessment: Breast and Ovarian, Guarneri V. Rare breast cancer subtypes: histologi- and risk of breast cancer: dose-response meta- version 2. 2017. https://www.nccn.org/professionals/ cal, molecular, and clinical peculiarities. Oncologist. analysis of prospective studies. Medicine (Baltimore). physician_gls/pdf/genetics_screening.pdf. Accessed 2014;19(8):805–813. 2013;92(3):123–131. December 22, 2016. 53. Mavaddat N, Barrowdale D, Andrulis IL, et al. 16. Ronckers CM, Erdmann CA, Land CE. Radiation 34. Hankinson SE, Colditz GA, Willett WC. Towards Pathology of breast and ovarian cancers among and breast cancer: a review of current evidence. an integrated model for breast cancer etiology: the BRCA1 and BRCA2 mutation carriers: results Breast Cancer Res. 2005;7(1):21–32. lifelong interplay of genes, lifestyle, and hormones. from the consortium of investigators of modifiers 17. Swerdlow AJ, Cooke R, Bates A, et al. Breast cancer Breast Cancer Res. 2004;6(5):213–218. of BRCA1/2 (CIMBA). Cancer Epidemiol Biomarkers risk after supradiaphragmatic radiotherapy for Hodg- 35. Amir E, Freedman OC, Seruga B, Evans DG. Prev. 2012;21(1):134–147. kin’s lymphoma in England and Wales: a national Assessing women at high risk of breast cancer: a 54. Turner N, Tutt A, Ashworth A. Hallmarks of cohort study. J Clin Oncol. 2012;30(22):2745–2752. review of risk assessment models. J Natl Cancer Inst. ‘BRCAness’ in sporadic cancers. Nat Rev Cancer. 18. Breast cancer and hormone replacement therapy: col- 2010;102(10):680–691. 2004;4(10):814–819. laborative reanalysis of data from 51 epidemiological 36. Cuzick J, Sestak I, Forbes JF, et al. Anastrozole 55. Arun B, Bayraktar S, Liu DD, et al. Response to neo- studies of 52,705 women with breast cancer and for prevention of breast cancer in high-risk post- adjuvant systemic therapy for breast cancer in BRCA 108,411 women without breast cancer. Collaborative menopausal women (IBIS-II): an international, mutation carriers and noncarriers: a single-institution group on hormonal factors in breast cancer. Lancet. double-blind, randomised placebo-controlled trial. experience. J Clin Oncol. 2011;29(28):3739– 1997;350(9084):1047–1059. Lancet. 2014;383(9922):1041–1048. 3746. 19. Chlebowski RT, Anderson GL, Gass M, et al. 37. Goss PE, Ingle JN, Ales-Martinez JE, et al. Exemes- 56. Byrski T, Gronwald J, Huzarski T, et al. Pathologic Estrogen plus progestin and breast cancer incidence tane for breast-cancer prevention in postmenopausal complete response rates in young women with and mortality in postmenopausal women. JAMA. women. N Engl J Med. 2011;364(25):2381–2391. BRCA1-positive breast cancers after neoadjuvant 2010;304(15):1684–1692. 38. Antoniou A, Pharoah PD, Narod S, et al. Average chemotherapy. J Clin Oncol. 2010;28(3):375–379. 20. LaCroix AZ, Chlebowski RT, Manson JE, et al. risks of breast and ovarian cancer associated with 57. Kriege M, Jager A, Hooning MJ, et al. The efficacy Health outcomes after stopping conjugated equine BRCA1 or BRCA2 mutations detected in case series of taxane chemotherapy for metastatic breast cancer 1603.e2 Part III: Specific Malignancies

in BRCA1 and BRCA2 mutation carriers. Cancer. 77. Baselga J, Cortes J, Im SA, et al. Biomarker analyses 97. Attardi LD, Donehower LA. Probing p53 biological 2012;118(4):899–907. in CLEOPATRA: a phase III, placebo-controlled functions through the use of genetically engineered 58. Robson ME, Chappuis PO, Satagopan J, et al. A study of pertuzumab in human epidermal growth mouse models. Mutat Res. 2005;576(1–2):4–21. combined analysis of outcome following breast factor receptor 2-positive, first-line metastatic breast 98. Gunther EJ, Moody SE, Belka GK, et al. Impact of cancer: differences in survival based on BRCA1/ cancer. J Clin Oncol. 2014;32(33):3753–3761. p53 loss on reversal and recurrence of conditional BRCA2 mutation status and administration of adju- 78. Loibl S, von Minckwitz G, Schneeweiss A, et al. Wnt-induced tumorigenesis. Genes Dev. 2003;17(4): vant treatment. Breast Cancer Res. 2004;6(1):R8–R17. PIK3CA mutations are associated with lower rates of 488–501. 59. Shah PD, Patil S, Dickler MN, Offit K, Hudis pathological complete response (pCR) to anti-HER2 99. Ventura A, Kirsch DG, McLaughlin ME, et al. CA, Robson ME. Twenty-one-gene recurrence score therapy in primary HER2-overexpressing breast Restoration of p53 function leads to tumour assay in BRCA-associated versus sporadic breast cancer. J Clin Oncol. 2014;32:3212–3220. regression in vivo. Nature. 2007;445(7128):661–665. cancers: differences based on germline mutation 79. Henry NL, Schott AF, Hayes DF. Assessment of 100. Borresen-Dale AL. TP53 and breast cancer. Hum status. Cancer. 2016;122(8):1178–1184. PIK3CA mutations in human epidermal growth Mutat. 2003;21(3):292–300. 60. Kaufman B, Shapira-Frommer R, Schmutzler factor receptor 2-positive breast cancer: clinical 101. Olivier M, Langerod A, Carrieri P, et al. The RK, et al. Olaparib monotherapy in patients with validity but not utility. J Clin Oncol. 2014;32(29): clinical value of somatic TP53 gene mutations in advanced cancer and a germline BRCA1/2 mutation. 3207–3209. 1,794 patients with breast cancer. Clin Cancer Res. J Clin Oncol. 2015;33(3):244–250. 80. Nahta R, Yuan LX, Zhang B, Kobayashi R, Esteva 2006;12(4):1157–1167. 61. Consensus conference on the classification of ductal FJ. Insulin-like growth factor-I receptor/human 102. Masciari S, Dillon DA, Rath M, et al. Breast cancer carcinoma in situ. The consensus conference com- epidermal growth factor receptor 2 heterodimeriza- phenotype in women with TP53 germline mutations: mittee. Cancer. 1997;80(9):1798–1802. tion contributes to trastuzumab resistance of breast a Li-fraumeni syndrome consortium effort. Breast 62. Cordera F, Jordan VC. Steroid receptors and their cancer cells. Cancer Res. 2005;65(23):11118–11128. Cancer Res Treat. 2012;133(3):1125–1130. role in the biology and control of breast cancer 81. Arribas J, Baselga J, Pedersen K, Parra-Palau 103. Melhem-Bertrandt A, Bojadzieva J, Ready KJ, et al. growth. Semin Oncol. 2006;33(6):631–641. JL. P95HER2 and breast cancer. Cancer Res. Early onset HER2-positive breast cancer is associated 63. Saji S, Hirose M, Toi M. Clinical significance of 2011;71(5):1515–1519. with germline TP53 mutations. Cancer. 2012;118(4): estrogen receptor beta in breast cancer. Cancer 82. Ben-Baruch NE, Bose R, Kavuri SM, Ma CX, 908–913. Chemother Pharmacol. 2005;56(suppl 1):21–26. Ellis MJ. HER2-mutated breast cancer responds 104. Heymann S, Delaloge S, Rahal A, et al. Radio- 64. Bjornstrom L, Sjoberg M. Mechanisms of estrogen to treatment with single-agent neratinib, a induced malignancies after breast cancer postop- receptor signaling: convergence of genomic and second-generation HER2/EGFR tyrosine kinase erative radiotherapy in patients with Li-fraumeni nongenomic actions on target genes. Mol Endocrinol. inhibitor. J Natl Compr Canc Netw. 2015;13(9):1061– syndrome. Radiat Oncol. 2010;5:104. 2005;19(4):833–842. 1064. 105. Turner N, Moretti E, Siclari O, et al. Targeting triple 65. De Vivo I, Hankinson SE, Colditz GA, Hunter 83. Bose R, Kavuri SM, Searleman AC, et al. Activating negative breast cancer: is p53 the answer? Cancer DJ. A functional polymorphism in the progesterone HER2 mutations in HER2 gene amplification nega- Treat Rev. 2013;39(5):541–550. receptor gene is associated with an increase in breast tive breast cancer. Cancer Discov. 2013;3(2):224–237. 106. Bykov VJ, Issaeva N, Zache N, et al. Reactivation of cancer risk. Cancer Res. 2003;63(17):5236–5238. 84. Press MF, Sauter G, Buyse M, et al. Alteration of mutant p53 and induction of apoptosis in human 66. Arpino G, Weiss H, Lee AV, et al. Estrogen recep- topoisomerase II-alpha gene in human breast cancer: tumor cells by maleimide analogs. J Biol Chem. tor–positive, progesterone receptor–negative breast association with responsiveness to anthracycline-based 2005;280(34):30384–30391. cancer: association with growth factor receptor chemotherapy. J Clin Oncol. 2011;29(7):859–867. 107. Andreeff M, Kelly KR, Yee K, et al. Results of expression and tamoxifen resistance. J Natl Cancer 85. Lavaud P, Andre F. Strategies to overcome trastu- the phase I trial of RG7112, a small-molecule Inst. 2005;97(17):1254–1261. zumab resistance in HER2-overexpressing breast MDM2 antagonist in leukemia. Clin Cancer Res. 67. Davies C, Godwin J, Gray R, et al. Relevance of cancers: focus on new data from clinical trials. 2016;22(4):868–876. breast cancer hormone receptors and other factors to BMC Med. 2014;12:132. 108. Tazawa H, Kagawa S, Fujiwara T. Advances in the efficacy of adjuvant tamoxifen: patient-level meta- 86. Rexer BN, Arteaga CL. Intrinsic and acquired adenovirus-mediated p53 cancer gene therapy. analysis of randomised trials. Lancet. 2011;378(9793): resistance to HER2-targeted therapies in HER2 gene- Expert Opin Biol Ther. 2013;13(11):1569– 771–784. amplified breast cancer: mechanisms and clinical 1583. 68. Baselga J, Swain SM. Novel anticancer targets: implications. Crit Rev Oncog. 2012;17(1):1–16. 109. Cristofanilli M, Krishnamurthy S, Guerra L, et al. revisiting ERBB2 and discovering ERBB3. Nat 87. Vivanco I, Sawyers CL. The phosphatidylinositol A nonreplicating adenoviral vector that contains Rev Cancer. 2009;9(7):463–475. 3-kinase AKT pathway in human cancer. Nat Rev the wild-type p53 transgene combined with chemo- 69. Junttila TT, Akita RW, Parsons K, et al. Ligand- Cancer. 2002;2(7):489–501. therapy for primary breast cancer: safety, efficacy, and independent HER2/HER3/PI3K complex is 88. Ellis MJ, Perou CM. The genomic landscape of breast biologic activity of a novel gene-therapy approach. disrupted by trastuzumab and is effectively inhibited cancer as a therapeutic roadmap. Cancer Discov. Cancer. 2006;107(5):935–944. by the PI3K inhibitor GDC-0941. Cancer Cell. 2012;3(1):27–34. 110. Sjoblom T, Jones S, Wood LD, et al. The consensus 2009;15(5):429–440. 89. Eng C, Peacocke M. PTEN and inherited hamar- coding sequences of human breast and colorectal 70. Gala K, Chandarlapaty S. Molecular pathways: toma-cancer syndromes. Nat Genet. 1998;19(3):223. cancers. Science. 2006;314(5797):268–274. HER3 targeted therapy. Clin Cancer Res. 2014;20(6): 90. Garcia JM, Silva J, Pena C, et al. Promoter methyla- 111. Sorlie T, Tibshirani R, Parker J, et al. Repeated 1410–1416. tion of the PTEN gene is a common molecular observation of breast tumor subtypes in independent 71. Catalogue of Somatic Mutations in Cancer. http:// change in breast cancer. Genes Chromosomes Cancer. gene expression data sets. Proc Natl Acad Sci USA. www.sanger.ac.uk/genetics/CGP/cosmic/, Accessed 2004;41(2):117–124. 2003;100(14):8418–8423. December 27, 2016. 91. Puc J, Keniry M, Li HS, et al. Lack of PTEN 112. Hu Z, Fan C, Oh DS, et al. The molecular portraits 72. Slamon DJ, Clark GM, Wong SG, Levin WJ, Ullrich sequesters CHK1 and initiates genetic instability. of breast tumors are conserved across microarray A, McGuire WL. Human breast cancer: correlation Cancer Cell. 2005;7(2):193–204. platforms. BMC Genomics. 2006;7:96. of relapse and survival with amplification of the 92. Nagata Y, Lan KH, Zhou X, et al. PTEN activation 113. Fan C, Oh DS, Wessels L, et al. Concordance among HER-2/neu oncogene. Science. 1987;235(4785):177– contributes to tumor inhibition by trastuzumab, gene-expression-based predictors for breast cancer. 182. and loss of PTEN predicts trastuzumab resistance N Engl J Med. 2006;355(6):560–569. 73. Ramakrishna N, Temin S, Chandarlapaty S, in patients. Cancer Cell. 2004;6(2):117–127. 114. Naylor TL, Greshock J, Wang Y, et al. High resolu- et al. Recommendations on disease management 93. Krop IE, Mayer IA, Ganju V, et al. Pictilisib for tion genomic analysis of sporadic breast cancer using for patients with advanced human epidermal oestrogen receptor–positive, aromatase inhibitor- array-based comparative genomic hybridization. growth factor receptor 2-positive breast cancer resistant, advanced or metastatic breast cancer Breast Cancer Res. 2005;7(6):R1186–R1198. and brain metastases: American society of clinical (FERGI): a randomised, double-blind, placebo- 115. Chin K, DeVries S, Fridlyand J, et al. Genomic and oncology clinical practice guideline. J Clin Oncol. controlled, phase 2 trial. Lancet Oncol. 2016;17(6): transcriptional aberrations linked to breast cancer 2014;32(19):2100–2108. 811–821. pathophysiologies. Cancer Cell. 2006;10(6):529– 74. Hudis CA. Trastuzumab—mechanism of action and 94. Mayer IA, Abramson VG, Formisano L, et al. A phase 541. use in clinical practice. N Engl J Med. 2007;357(1): ib study of alpelisib (BYL719), a PI3Kalpha-specific 116. Pusztai L, Ayers M, Stec J, et al. Gene expression 39–51. inhibitor, with letrozole in ER+/HER2− metastatic profiles obtained from fine-needle aspirations of 75. Chandarlapaty S. Negative feedback and adaptive breast cancer. Clin Cancer Res. 2016;23(1):26–34. breast cancer reliably identify routine prognostic resistance to the targeted therapy of cancer. Cancer 95. Cancer Genome Atlas Network. Comprehensive markers and reveal large-scale molecular differences Discov. 2012;2(4):311–319. molecular portraits of human breast tumours. Nature. between estrogen-negative and estrogen-positive 76. Chandarlapaty S, Sakr RA, Giri D, et al. Frequent 2012;490(7418):61–70. tumors. Clin Cancer Res. 2003;9(7):2406–2415. mutational activation of the PI3K-AKT pathway 96. Lacroix M, Toillon RA, Leclercq G. P53 and 117. ENCODE Project Consortium. An integrated in trastuzumab-resistant breast cancer. Clin Cancer breast cancer, an update. Endocr Relat Cancer. encyclopedia of DNA elements in the human Res. 2012;18(24):6784–6791. 2006;13(2):293–325. genome. Nature. 2012;489(7414):57–74. Cancer of the Breast • CHAPTER 88 1603.e3

118. Visvader JE, Lindeman GJ. Cancer stem cells in metastatic breast cancer. Sci Transl Med. 2015;7(313): women with breast cancer and BRCA1 or BRCA2 solid tumours: accumulating evidence and unresolved 313ra182. mutations. JAMA. 2000;283(5):617–624. questions. Nat Rev Cancer. 2008;8(10):755–768. 140. De Mattos-Arruda L, Weigelt B, Cortes J, et al. 158. Kurian AW, Sigal BM, Plevritis SK. Survival analysis 119. Al-Hajj M, Wicha MS, Benito-Hernandez A, Mor- Capturing intra-tumor genetic heterogeneity by of cancer risk reduction strategies for BRCA1/2 rison SJ, Clarke MF. Prospective identification of de novo mutation profiling of circulating cell-free mutation carriers. J Clin Oncol. 2010;28(2):222–231. tumorigenic breast cancer cells. Proc Natl Acad Sci tumor DNA: a proof-of-principle. Ann Oncol. 159. Tercyak KP, Peshkin BN, Brogan BM, et al. Quality USA. 2003;100(7):3983–3988. 2014;25(9):1729–1735. of life after contralateral prophylactic mastectomy 120. Reya T, Morrison SJ, Clarke MF, Weissman IL. 141. Lanman RB, Mortimer SA, Zill OA, et al. Analyti- in newly diagnosed high-risk breast cancer patients Stem cells, cancer, and cancer stem cells. Nature. cal and clinical validation of a digital sequencing who underwent BRCA1/2 gene testing. J Clin Oncol. 2001;414(6859):105–111. panel for quantitative, highly accurate evaluation 2007;25(3):285–291. 121. Schott AF, Landis MD, Dontu G, et al. Preclinical of cell-free circulating tumor DNA. PLoS ONE. 160. Finch AP, Lubinski J, Moller P, et al. Impact of and clinical studies of gamma secretase inhibitors 2015;10(10):e0140712. oophorectomy on cancer incidence and mortality with docetaxel on human breast tumors. Clin Cancer 142. Domchek SM, Eisen A, Calzone K, Stopfer J, in women with a BRCA1 or BRCA2 mutation. Res. 2013;19(6):1512–1524. Blackwood A, Weber BL. Application of breast J Clin Oncol. 2014;32(15):1547–1553. 122. Braun S, Vogl FD, Naume B, et al. A pooled analysis cancer risk prediction models in clinical practice. 161. Siu AL, US Preventive Services Task Force. Screen- of bone marrow micrometastasis in breast cancer. J Clin Oncol. 2003;21(4):593–601. ing for breast cancer: US preventive services task N Engl J Med. 2005;353(8):793–802. 143. Saslow D, Boetes C, Burke W, et al. American cancer force recommendation statement. Ann Intern Med. 123. Weaver DL, Ashikaga T, Krag DN, et al. Effect of society guidelines for breast screening with MRI as an 2016;164(4):279–296. occult metastases on survival in node-negative breast adjunct to mammography. CA Cancer J Clin. 2007; 162. Broeders M, Moss S, Nystrom L, et al. The impact of cancer. N Engl J Med. 2011;364(5):412–421. 57(2):75–89. mammographic screening on breast cancer mortality 124. Janni W, Vogl FD, Wiedswang G, et al. Persistence 144. DeMichele A, Weber BL. Risk management in in Europe: a review of observational studies. J Med of disseminated tumor cells in the bone marrow BRCA1 and BRCA2 mutation carriers: lessons Screen. 2012;19(suppl 1):14–25. of breast cancer patients predicts increased risk for learned, challenges posed. J Clin Oncol. 2002;20(5): 163. Schopper D, de Wolf C. How effective are breast relapse—a European pooled analysis. Clin Cancer 1164–1166. cancer screening programmes by mammography? Res. 2011;17(9):2967–2976. 145. Fisher B, Costantino JP, Wickerham DL, et al. Review of the current evidence. Eur J Cancer. 125. Hayes DF, Zurawski VR Jr, Kufe DW. Comparison Tamoxifen for the prevention of breast cancer: 2009;45(11):1916–1923. of circulating CA15-3 and carcinoembryonic antigen current status of the national surgical adjuvant 164. Webb ML, Cady B, Michaelson JS, et al. A failure levels in patients with breast cancer. J Clin Oncol. breast and bowel project P-1 study. J Natl Cancer analysis of invasive breast cancer: most deaths from 1986;4(10):1542–1550. Inst. 2005;97(22):1652–1662. disease occur in women not regularly screened. 126. Pantel K, Brakenhoff RH, Brandt B. Detection, clini- 146. Cuzick J, Forbes JF, Sestak I, et al. Long-term Cancer. 2014;120(18):2839–2846. cal relevance and specific biological properties of dis- results of tamoxifen prophylaxis for breast cancer— 165. http://www.fda.gov/Radiation-EmittingProducts/ seminating tumour cells. Nat Rev Cancer. 2008;8(5): 96-month follow-up of the randomized IBIS-I trial. MammographyQualityStandardsActandProgram/ 329–340. J Natl Cancer Inst. 2007;99(4):272–282. FacilityScorecard/ucm465058.htm. Accessed December 127. Alix-Panabieres C, Pantel K. Challenges in circulating 147. Vogel VG, Costantino JP, Wickerham DL, et al. 10, 2016. tumour cell research. Nat Rev Cancer. 2014;14(9): Update of the national surgical adjuvant breast and 166. Oeffinger KC, Fontham ET, Etzioni R, et al. Breast 623–631. bowel project study of tamoxifen and raloxifene cancer screening for women at average risk: 2015 128. Cristofanilli M, Budd GT, Ellis MJ, et al. Circulating (STAR) P-2 trial: preventing breast cancer. Cancer guideline update from the American cancer society. tumor cells, disease progression, and survival in Prev Res (Phila). 2010;3(6):696–706. JAMA. 2015;314(15):1599–1614. metastatic breast cancer. N Engl J Med. 2004;351(8): 148. Visvanathan K, Hurley P, Bantug E, et al. Use of 167. Bevers TB, Ward JH, Arun BK, et al. Breast cancer 781–791. pharmacologic interventions for breast cancer risk risk reduction, version 2.2015. J Natl Compr Canc 129. Smerage JB, Barlow WE, Hortobagyi GN, et al. Cir- reduction: American society of clinical oncology Netw. 2015;13(7):880–915. culating tumor cells and response to chemotherapy clinical practice guideline. J Clin Oncol. 2013;31(23): 168. Nelson HD, Pappas M, Cantor A, Griffin J, Daeges in metastatic breast cancer: SWOG s0500. J Clin 2942–2962. M, Humphrey L. Harms of breast cancer screening: Oncol. 2014;32(31):3483–3489. 149. Phillips KA, Milne RL, Rookus MA, et al. Tamoxifen systematic review to update the 2009 US preventive 130. Ignatiadis M, Lee M, Jeffrey SS. Circulating tumor and risk of contralateral breast cancer for BRCA1 and services task force recommendation. Ann Intern Med. cells and circulating tumor DNA: challenges and BRCA2 mutation carriers. J Clin Oncol. 2013;31(25): 2016;164(4):256–267. opportunities on the path to clinical utility. Clin 3091–3099. 169. Tosteson AN, Fryback DG, Hammond CS, et al. Cancer Res. 2015;21(21):4786–4800. 150. Powles TJ, Ashley S, Tidy A, Smith IE, Dowsett M. Consequences of false-positive screening mam- 131. Calin GA, Croce CM. MicroRNA signatures in Twenty-year follow-up of the royal marsden random- mograms. JAMA Intern Med. 2014;174(6):954–961. human cancers. Nat Rev Cancer. 2006;6(11):857–866. ized, double-blinded tamoxifen breast cancer preven- 170. Schwartz LM, Woloshin S, Sox HC, Fischhoff B, 132. Schwarzenbach H, Pantel K. Circulating DNA as bio- tion trial. J Natl Cancer Inst. 2007;99(4):283–290. Welch HG. US women’s attitudes to false-positive marker in breast cancer. Breast Cancer Res. 2015;17(1): 151. Veronesi U, Maisonneuve P, Rotmensz N, et al. mammography results and detection of ductal 136. Tamoxifen for the prevention of breast cancer: carcinoma in situ: cross-sectional survey. West J 133. Dawson SJ, Tsui DW, Murtaza M, et al. Analysis late results of the Italian randomized tamoxifen Med. 2000;173(5):307–312. of circulating tumor DNA to monitor metastatic prevention trial among women with hysterectomy. 171. Hendrick RE. Radiation doses and cancer breast cancer. N Engl J Med. 2013;368(13):1199– J Natl Cancer Inst. 2007;99(9):727–737. risks from breast imaging studies. Radiology. 1209. 152. Freedman AN, Yu B, Gail MH, et al. Benefit/risk 2010;257(1):246–253. 134. Garcia-Murillas I, Schiavon G, Weigelt B, et al. assessment for breast cancer chemoprevention with 172. Pisano ED, Gatsonis C, Hendrick E, et al. Diagnostic Mutation tracking in circulating tumor DNA raloxifene or tamoxifen for women age 50 years or performance of digital versus film mammography predicts relapse in early breast cancer. Sci Transl older. J Clin Oncol. 2011;29(17):2327–2333. for breast-cancer screening. N Engl J Med. 2005; Med. 2015;7(302):302ra133. 153. Hartmann LC, Schaid DJ, Woods JE, et al. Efficacy 353(17):1773–1783. 135. Forshew T, Murtaza M, Parkinson C, et al. Non- of bilateral prophylactic mastectomy in women with 173. Melnikow J, Fenton JJ, Whitlock EP, et al. invasive identification and monitoring of cancer a family history of breast cancer. N Engl J Med. Supplemental screening for breast cancer in women mutations by targeted deep sequencing of plasma 1999;340(2):77–84. with dense breasts: a systematic review for the US DNA. Sci Transl Med. 2012;4(136):136ra168. 154. Domchek SM, Friebel TM, Singer CF, et al. Associa- preventive services task force. Ann Intern Med. 136. Murtaza M, Dawson SJ, Tsui DW, et al. Non- tion of risk-reducing surgery in BRCA1 or BRCA2 2016;164(4):268–278. invasive analysis of acquired resistance to cancer mutation carriers with cancer risk and mortality. 174. Sprague BL, Stout NK, Schechter C, et al. Benefits, therapy by sequencing of plasma DNA. Nature. JAMA. 2010;304(9):967–975. harms, and cost-effectiveness of supplemental 2013;497(7447):108–112. 155. McDonnell SK, Schaid DJ, Myers JL, et al. Efficacy ultrasonography screening for women with dense 137. Heneghan HM, Miller N, Lowery AJ, Sweeney KJ, of contralateral prophylactic mastectomy in women breasts. Ann Intern Med. 2015;162(3):157–166. Newell J, Kerin MJ. Circulating microRNAs as novel with a personal and family history of breast cancer. 175. Skaane P, Bandos AI, Gullien R, et al. Comparison minimally invasive biomarkers for breast cancer. Ann J Clin Oncol. 2001;19(19):3938–3943. of digital mammography alone and digital mam- Surg. 2010;251(3):505–511. 156. Hoogerbrugge N, Bult P, de Widt-Levert LM, mography plus tomosynthesis in a population-based 138. Segal CV, Dowsett M. Estrogen receptor mutations et al. High prevalence of premalignant lesions in screening program. Radiology. 2013;267(1):47– in breast cancer—new focus on an old target. Clin prophylactically removed breasts from women at 56. Cancer Res. 2014;20(7):1724–1726. hereditary risk for breast cancer. J Clin Oncol. 2003; 176. Rafferty EA, Durand MA, Conant EF, et al. Breast 139. Schiavon G, Hrebien S, Garcia-Murillas I, et al. 21(1):41–45. cancer screening using tomosynthesis and digital Analysis of ESR1 mutation in circulating tumor 157. Schrag D, Kuntz KM, Garber JE, Weeks JC. Life mammography in dense and nondense breasts. JAMA. DNA demonstrates evolution during therapy for expectancy gains from cancer prevention strategies for 2016;315(16):1784–1786. 1603.e4 Part III: Specific Malignancies

177. Berg WA, Zhang Z, Lehrer D, et al. Detection of pathologists clinical practice guideline update. J Clin 214. Wapnir IL, Dignam JJ, Fisher B, et al. Long-term breast cancer with addition of annual screening Oncol. 2013;31(31):3997–4013. outcomes of invasive ipsilateral breast tumor recur- ultrasound or a single screening MRI to mam- 196. Hammond ME, Hayes DF, Dowsett M, Allred DC, rences after lumpectomy in NSABP B-17 and B-24 mography in women with elevated breast cancer Hagerty KL, Badve S. American society of clinical randomized clinical trials for DCIS. J Natl Cancer risk. JAMA. 2012;307(13):1394–1404. Oncology/college of American pathologists guideline Inst. 2011;103(6):478–488. 178. Granader EJ, Dwamena B, Carlos RC. MRI recommendations for immunohistochemical testing 215. Early Breast Cancer Trialists’ Collaborative Group, and mammography surveillance of women at of estrogen and progesterone receptors in breast Bijker N, et al. Breast-conserving treatment with increased risk for breast cancer: recommendations cancer. J Clin Oncol. 2010;28(16):2784–2795. or without radiotherapy in ductal carcinoma-in- using an evidence-based approach. Acad Radiol. 197. Ries LAG, Eisner MP. Cancer of the female breast. situ: ten-year results of European organisation for 2008;15(12):1590–1595. In: Ries LAG, Young JL, Keel GE, Eisner MP, Lin research and treatment of cancer randomized phase 179. Warner E, Messersmith H, Causer P, Eisen A, YD, Horner M-JD, eds. SEER Survival Monograph: III trial 10853–a study by the EORTC breast cancer Shumak R, Plewes D. Systematic review: using Cancer Survival Among Adults: U.S. SEER Program, cooperative group and EORTC radiotherapy group. magnetic resonance imaging to screen women 1988-2001, Patient and Tumor Characteristics. J Clin Oncol. 2006;24(21):3381–3387. at high risk for breast cancer. Ann Intern Med. Bethesda, MD: National Cancer Institute; SEER 216. Early Breast Cancer Trialists’ Collaborative Group, 2008;148(9):671–679. Program, NIH Pub. No. 07-6215. 2007. Correa C, McGale P, et al. Overview of the ran- 180. Institute for Clinical and Economic Review. The 198. Dowsett M, Nielsen TO, A’Hern R, et al. Assessment domized trials of radiotherapy in ductal carcinoma Comparative Clinical Effectiveness and Value of of ki67 in breast cancer: recommendations from the in situ of the breast. J Natl Cancer Inst Monogr. Supplemental Screening Tests Following Negative international ki67 in breast cancer working group. 2010;2010(41):162–177. Mammography in Women with Dense Breast J Natl Cancer Inst. 2011;103(22):1656–1664. 217. Silverstein MJ, Lagios MD. Choosing treatment Tissue. http://cepac.icer-review.org/wp-content/ 199. Harris LN, Ismaila N, McShane LM, et al. Use of for patients with ductal carcinoma in situ: fine uploads/2013/11/Action-Guide-Dense-Breast-Final. biomarkers to guide decisions on adjuvant systemic tuning the university of southern California/van pdf. Accessed December 17, 2016. therapy for women with early-stage invasive breast nuys prognostic index. J Natl Cancer Inst Monogr. 181. Centers for Disease Control and Prevention. https:// cancer: American society of clinical oncology clini- 2010;2010(41):193–196. www.cdc.gov/nchs/fastats/life-expectancy.htm. cal practice guideline. J Clin Oncol. 2016;34(10): 218. Wong JS, Kaelin CM, Troyan SL, et al. Prospective Accessed December 22, 2016. 1134–1150. study of wide excision alone for ductal carcinoma 182. Tabar L, Vitak B, Chen HH, et al. The Swedish Two- 200. Paik S, Shak S, Tang G, et al. A multigene assay in situ of the breast. J Clin Oncol. 2006;24(7): county trial twenty years later. Updated mortality to predict recurrence of tamoxifen-treated, node- 1031–1036. results and new insights from long-term follow-up. negative breast cancer. N Engl J Med. 2004;351(27): 219. Solin LJ, Gray R, Hughes LL, et al. Surgical excision Radiol Clin North Am. 2000;38(4):625–651. 2817–2826. without radiation for ductal carcinoma in situ of 183. Galit W, Green MS, Lital KB. Routine screen- 201. Paik S, Tang G, Shak S, et al. Gene expression and the breast: 12-year results from the ECOG-ACRIN ing mammography in women older than 74 benefit of chemotherapy in women with node- e5194 study. J Clin Oncol. 2015;33(33):3938– years: a review of the available data. Maturitas. negative, estrogen receptor–positive breast cancer. 3944. 2007;57(2):109–119. J Clin Oncol. 2006;24(23):3726–3734. 220. Solin LJ, Gray R, Baehner FL, et al. A multigene 184. Malmgren JA, Parikh J, Atwood MK, Kaplan 202. van de Vijver MJ, He YD, van’t Veer LJ, et al. A gene- expression assay to predict local recurrence risk for HG. Improved prognosis of women aged 75 and expression signature as a predictor of survival in breast ductal carcinoma in situ of the breast. J Natl Cancer older with mammography-detected breast cancer. cancer. N Engl J Med. 2002;347(25):1999–2009. Inst. 2013;105(10):701–710. Radiology. 2014;273(3):686–694. 203. van ‘t Veer LJ, Dai H, van de Vijver MJ, et al. 221. McCormick B, Winter K, Hudis C, et al. RTOG 185. Smith-Bindman R, Kerlikowske K, Gebretsadik T, Gene expression profiling predicts clinical outcome 9804: a prospective randomized trial for good-risk Newman J. Is screening mammography effective in of breast cancer. Nature. 2002;415(6871):530–536. ductal carcinoma in situ comparing radiotherapy elderly women? Am J Med. 2000;108(2):112–119. 204. Ma XJ, Wang Z, Ryan PD, et al. A two-gene with observation. J Clin Oncol. 2015;33(7):709–715. 186. Vyas A, Madhavan S, Sambamoorthi U. Association expression ratio predicts clinical outcome in breast 222. Cuzick J, Sestak I, Pinder SE, et al. Effect of tamoxi- between persistence with mammography screening cancer patients treated with tamoxifen. Cancer Cell. fen and radiotherapy in women with locally excised and stage at diagnosis among elderly women diag- 2004;5(6):607–616. ductal carcinoma in situ: long-term results from the nosed with breast cancer. Breast Cancer Res Treat. 205. Parker JS, Mullins M, Cheang MC, et al. Supervised UK/ANZ DCIS trial. Lancet Oncol. 2011;12(1): 2014;148(3):645–654. risk predictor of breast cancer based on intrinsic 21–29. 187. Walter LC, Schonberg MA. Screening mam- subtypes. J Clin Oncol. 2009;27(8):1160–1167. 223. Staley H, McCallum I, Bruce J. Postoperative mography in older women: a review. JAMA. 206. Duffy MJ, McGowan PM, Harbeck N, Thomssen C, tamoxifen for ductal carcinoma in situ. Cochrane 2014;311(13):1336–1347. Schmitt M. uPA and PAI-1 as biomarkers in breast Database Syst Rev. 2012;(10):CD007847. 188. Sickles E, Bassett L, D’Orsi C. Breast Imaging cancer: validated for clinical use in level-of-evidence-1 224. Margolese RG, Cecchini RS, Julian TB, et al. Reporting and Data System, BI-RADS: Mammography. studies. Breast Cancer Res. 2014;16(4):428. Anastrozole versus tamoxifen in postmenopausal Reston, VA: American College of Radiology; 2013. 207. Cardoso F, van’t Veer LJ, Bogaerts J, et al. 70-gene sig- women with ductal carcinoma in situ undergoing 189. American College of Radiology. ACR BI-RADS nature as an aid to treatment decisions in early-stage lumpectomy plus radiotherapy (NSABP B-35): Atlas—Mammography. 2013. https://www.acr. breast cancer. N Engl J Med. 2016;375(8):717–729. a randomised, double-blind, phase 3 clinical org/~/media/ACR/Documents/PDF/QualitySafety/ 208. Schnitt SJ, Morrow M. Lobular carcinoma in situ: trial. Lancet. 2016;387(10021):849–856. Resources/BIRADS/01-Mammography/02- current concepts and controversies. Semin Diagn 225. Fisher B, Anderson S, Bryant J, et al. Twenty-year -BIRADS-Mammography-Reporting.pdf?la=en. Pathol. 1999;16(3):209–223. follow-up of a randomized trial comparing total Accessed December 22, 2016. 209. Li CI, Malone KE, Saltzman BS, Daling JR. mastectomy, lumpectomy, and lumpectomy plus 190. Gray RJ, Salud C, Nguyen K, et al. Randomized Risk of invasive breast carcinoma among women irradiation for the treatment of invasive breast prospective evaluation of a novel technique for diagnosed with ductal carcinoma in situ and lobular cancer. N Engl J Med. 2002;347(16):1233–1241. biopsy or lumpectomy of nonpalpable breast lesions: carcinoma in situ, 1988-2001. Cancer. 2006;106(10): 226. Veronesi U, Cascinelli N, Mariani L, et al. radioactive seed versus wire localization. Ann Surg 2104–2112. Twenty-year follow-up of a randomized study Oncol. 2001;8(9):711–715. 210. Middleton LP, Palacios DM, Bryant BR, Krebs P, Otis comparing breast-conserving surgery with radical 191. Calhoun BC, Collins LC. Recommendations CN, Merino MJ. Pleomorphic lobular carcinoma: mastectomy for early breast cancer. N Engl J Med. for excision following core needle biopsy of the morphology, immunohistochemistry, and molecular 2002;347(16):1227–1232. breast: a contemporary evaluation of the literature. analysis. Am J Surg Pathol. 2000;24(12):1650–1656. 227. NIH consensus conference. Treatment of early-stage Histopathology. 2016;68(1):138–151. 211. Vargas AC, Lakhani SR, Simpson PT. Pleomorphic breast cancer. JAMA. 1991;265(3):391–395. 192. Breast. In: Edge SB, Byrd DR, Compton CC, lobular carcinoma of the breast: molecular pathol- 228. Clarke M, Collins R, Darby S, et al. Effects of Fritz AG, Greene FL, Trotti A, eds. AJCC Cancer ogy and clinical impact. Future Oncol. 2009;5(2): radiotherapy and of differences in the extent of Staging Manual. 7th ed. New York, NY: Springer; 233–243. surgery for early breast cancer on local recurrence 2010:347–376. 212. Morrow M, Van Zee KJ, Solin LJ, et al. Society of and 15-year survival: an overview of the randomised 193. Henry NL, Hayes DF. Cancer biomarkers. Mol surgical Oncology–American society for radiation trials. Lancet. 2005;366(9503):2087–2106. Oncol. 2012;6(2):140–146. Oncology–American society of clinical oncology 229. Zurrida S, Costa A, Luini A, Galimberti V, Sac- 194. Allred DC. Commentary: hormone receptor testing consensus guideline on margins for breast-conserving chini V, Intra M. The veronesi quadrantectomy: in breast cancer: a distress signal from Canada. surgery with whole-breast irradiation in ductal an established procedure for the conservative Oncologist. 2008;13(11):1134–1136. carcinoma in situ. J Clin Oncol. 2016;34:4040– treatment of early breast cancer. Int J Surg Investig. 195. Wolff AC, Hammond ME, Hicks DG, et al. 4046. 2001;2(6):423–431. Recommendations for human epidermal growth 213. Mamounas EP. NSABP breast cancer clinical trials: 230. Houssami N, Macaskill P, Marinovich ML, Morrow factor receptor 2 testing in breast cancer: American recent results and future directions. Clin Med Res. M. The association of surgical margins and local society of clinical Oncology/college of American 2003;1(4):309–326. recurrence in women with early-stage invasive breast Cancer of the Breast • CHAPTER 88 1603.e5

cancer treated with breast-conserving therapy: a 248. Jagsi R, Hawley ST, Griffith KA, et al. Contralateral of early breast cancer: 10-year follow-up results meta-analysis. Ann Surg Oncol. 2014;21(3):717–730. prophylactic mastectomy decisions in a population- of two randomised controlled trials. Lancet Oncol. 231. Moran MS, Schnitt SJ, Giuliano AE, et al. Society based sample of patients with early-stage breast 2013;14(11):1086–1094. of surgical Oncology–American society for radiation cancer. JAMA Surg. 2016;epub December 21, 2016. 265. Whelan TJ, Pignol JP, Levine MN, et al. Long-term oncology consensus guideline on margins for breast- 249. Sabel MS, Dal Cin S. Trends in media reports of results of hypofractionated radiation therapy for conserving surgery with whole-breast irradiation in celebrities’ breast cancer treatment decisions. Ann breast cancer. N Engl J Med. 2010;362(6):513–520. stages I and II invasive breast cancer. J Clin Oncol. Surg Oncol. 2016;23(9):2795–2801. 266. Darby SC, Ewertz M, McGale P, et al. Risk of 2014;32(14):1507–1515. 250. Boughey JC, Attai DJ, Chen SL, et al. Contralateral ischemic heart disease in women after radiotherapy 232. Chagpar AB, Killelea BK, Tsangaris TN, et al. A ran- prophylactic mastectomy consensus statement from for breast cancer. N Engl J Med. 2013;368(11):987– domized, controlled trial of cavity shave margins in the American society of breast surgeons: additional 998. breast cancer. N Engl J Med. 2015;373(6):503–510. considerations and a framework for shared decision 267. Bartelink H, Horiot JC, Poortmans PM, et al. 233. Sabel MS, Jorns JM, Wu A, Myers J, Newman making. Ann Surg Oncol. 2016;23(10):3106–3111. Impact of a higher radiation dose on local control LA, Breslin TM. Development of an intraoperative 251. Society of Surgical Oncology. Position Statement on and survival in breast-conserving therapy of early pathology consultation service at a free-standing Prophylactic Mastectomy. http://www.surgonc.org/ breast cancer: 10-year results of the randomized ambulatory surgical center: clinical and economic resources/consensus-statements/position-statement- boost versus no boost EORTC 22881-10882 trial. impact for patients undergoing breast cancer surgery. on-prophylactic-mastectomy. Accessed on January J Clin Oncol. 2007;25(22):3259–3265. Am J Surg. 2012;204(1):66–77. 9, 2017. 268. Smith GL, Xu Y, Buchholz TA, et al. Association 234. Fukamachi K, Ishida T, Usami S, et al. Total- 252. Chung MA, Steinhoff MM, Cady B. Clinical axillary between treatment with brachytherapy vs whole- circumference intraoperative frozen section analysis recurrence in breast cancer patients after a negative breast irradiation and subsequent mastectomy, reduces margin-positive rate in breast-conservation sentinel node biopsy. Am J Surg. 2002;184(4): complications, and survival among older women surgery. Jpn J Clin Oncol. 2010;40(6):513–520. 310–314. with invasive breast cancer. JAMA. 2012;307(17): 235. Olson TP, Harter J, Munoz A, Mahvi DM, Breslin 253. Jeruss JS, Winchester DJ, Sener SF, et al. Axillary 1827–1837. T. Frozen section analysis for intraoperative margin recurrence after sentinel node biopsy. Ann Surg 269. Olivotto IA, Whelan TJ, Parpia S, et al. Interim assessment during breast-conserving surgery results Oncol. 2005;12(1):34–40. cosmetic and toxicity results from RAPID: a random- in low rates of re-excision and local recurrence. Ann 254. Smidt ML, Janssen CM, Kuster DM, Bruggink ized trial of accelerated partial breast irradiation using Surg Oncol. 2007;14(10):2953–2960. ED, Strobbe LJ. Axillary recurrence after a negative three-dimensional conformal external beam radiation 236. Blair SL, Thompson K, Rococco J, Malcarne sentinel node biopsy for breast cancer: incidence and therapy. J Clin Oncol. 2013;31(32):4038–4045. V, Beitsch PD, Ollila DW. Attaining negative clinical significance.Ann Surg Oncol. 2005;12(1): 270. Livi L, Meattini I, Marrazzo L, et al. Accelerated margins in breast-conservation operations: is there 29–33. partial breast irradiation using intensity-modulated a consensus among breast surgeons? J Am Coll Surg. 255. Veronesi U, Paganelli G, Viale G, et al. A randomized radiotherapy versus whole breast irradiation: 5-year 2009;209(5):608–613. comparison of sentinel-node biopsy with routine survival analysis of a phase 3 randomised controlled 237. Morrow M, Strom EA, Bassett LW, et al. Standard for axillary dissection in breast cancer. N Engl J Med. trial. Eur J Cancer. 2015;51(4):451–463. breast conservation therapy in the management of inva- 2003;349(6):546–553. 271. Correa C, Harris EE, Leonardi MC, et al. Accelerated sive breast carcinoma. CA Cancer J Clin. 2002;52(5): 256. Krag DN, Anderson SJ, Julian TB, et al. Sentinel- partial breast irradiation: executive summary for the 277–300. lymph-node resection compared with conventional update of an ASTRO evidence-based consensus 238. De La Cruz L, Blankenship SA, Chatterjee A, axillary-lymph-node dissection in clinically node- statement. Pract Radiat Oncol. 2016;epub Sept 17. et al. Outcomes after oncoplastic breast-conserving negative patients with breast cancer: overall survival 272. Whelan TJ, Olivotto IA, Parulekar WR, et al. surgery in breast cancer patients: a systematic findings from the NSABP B-32 randomised phase Regional nodal irradiation in early-stage breast literature review. Ann Surg Oncol. 2016;23(10):3247– 3 trial. Lancet Oncol. 2010;11(10):927–933. cancer. N Engl J Med. 2015;373(4):307–316. 3258. 257. Weaver DL, Le UP, Dupuis SL, et al. Metastasis 273. Poortmans PM, Collette S, Kirkove C, et al. Internal 239. Arthur DW, Winter K, Kuerer HM, et al. NRG detection in sentinel lymph nodes: comparison of mammary and medial supraclavicular irradiation in Oncology/RTOG 1014: 1-year toxicity report from a limited widely spaced (NSABP protocol B-32) breast cancer. N Engl J Med. 2015;373(4):317–327. a phase II study of repeat breast preserving surgery and a comprehensive narrowly spaced paraffin block 274. Giuliano AE, Hunt KK, Ballman KV, et al. Axillary and 3d conformal partial-breast reirradiation (PBrI) sectioning strategy. Am J Surg Pathol. 2009;33(11): dissection vs no axillary dissection in women with for in-breast recurrence. Int J Radiation Oncol Biol 1583–1589. invasive breast cancer and sentinel node metastasis: Phys. 2015;93(3 (suppl)):S58–S59. (abstract 130). 258. Giuliano AE, McCall L, Beitsch P, et al. Locoregional a randomized clinical trial. JAMA. 2011;305(6): 240. Sabel MS, Kraft CT, Griffith KA, et al. Differ- recurrence after sentinel lymph node dissection with 569–575. ences between breast conservation-eligible patients or without axillary dissection in patients with sentinel 275. Overgaard M, Hansen PS, Overgaard J, et al. Post- and unilateral mastectomy patients in choosing lymph node metastases: the American college of operative radiotherapy in high-risk premenopausal contralateral prophylactic mastectomies. Breast J. surgeons oncology group z0011 randomized trial. women with breast cancer who receive adjuvant che- 2016;22(6):607–615. Ann Surg. 2010;252(3):426–432. motherapy. Danish breast cancer cooperative group 241. Tuttle TM, Habermann EB, Grund EH, Morris 259. Galimberti V, Cole BF, Zurrida S, et al. Axillary 82b trial. N Engl J Med. 1997;337(14):949–955. TJ, Virnig BA. Increasing use of contralateral dissection versus no axillary dissection in patients 276. Overgaard M, Jensen MB, Overgaard J, et al. Post- prophylactic mastectomy for breast cancer patients: with sentinel-node micrometastases (IBCSG 23-01): operative radiotherapy in high-risk postmenopausal a trend toward more aggressive surgical treatment. a phase 3 randomised controlled trial. Lancet Oncol. breast-cancer patients given adjuvant tamoxifen: J Clin Oncol. 2007;25(33):5203–5209. 2013;14(4):297–305. Danish breast cancer cooperative group DBCG 242. Yao K, Stewart AK, Winchester DJ, Winchester DP. 260. Jagsi R, Chadha M, Moni J, et al. Radiation field 82c randomised trial. Lancet. 1999;353(9165): Trends in contralateral prophylactic mastectomy for design in the ACOSOG z0011 (Alliance) trial. J 1641–1648. unilateral cancer: a report from the national cancer Clin Oncol. 2014;32(32):3600–3606. 277. Ragaz J, Olivotto IA, Spinelli JJ, et al. Locoregional data base, 1998-2007. Ann Surg Oncol. 2010;17(10): 261. Donker M, van Tienhoven G, Straver ME, et al. radiation therapy in patients with high-risk breast 2554–2562. Radiotherapy or surgery of the axilla after a positive cancer receiving adjuvant chemotherapy: 20-year 243. Hawley ST, Jagsi R, Morrow M, et al. Social and sentinel node in breast cancer (EORTC 10981- results of the British columbia randomized trial. clinical determinants of contralateral prophylactic 22023 AMAROS): a randomised, multicentre, J Natl Cancer Inst. 2005;97(2):116–126. mastectomy. JAMA Surg. 2014;149(6):582–589. open-label, phase 3 non-inferiority trial. Lancet 278. Early Breast Cancer Trialists’ Collaborative Group, 244. Abbott A, Rueth N, Pappas-Varco S, Kuntz K, Oncol. 2014;15(12):1303–1310. McGale P, Taylor C, et al. Effect of radiotherapy Kerr E, Tuttle T. Perceptions of contralateral breast 262. Early Breast Cancer Trialists’ Collaborative Group, after mastectomy and axillary surgery on 10-year cancer: an overestimation of risk. Ann Surg Oncol. Darby S, McGale P, et al. Effect of radiotherapy recurrence and 20-year breast cancer mortality: meta- 2011;18(11):3129–3136. after breast-conserving surgery on 10-year recurrence analysis of individual patient data for 8135 women 245. Rosenberg SM, Tracy MS, Meyer ME, et al. Percep- and 15-year breast cancer death: meta-analysis of in 22 randomised trials. Lancet. 2014;383(9935): tions, knowledge, and satisfaction with contralateral individual patient data for 10,801 women in 17 ran- 2127–2135. prophylactic mastectomy among young women with domised trials. Lancet. 2011;378(9804):1707–1716. 279. Recht A, Comen EA, Fine RE, et al. Postmastectomy breast cancer: a cross-sectional survey. Ann Intern 263. Hughes KS, Schnaper LA, Bellon JR, et al. Lumpec- radiotherapy: an American society of clinical oncol- Med. 2013;159(6):373–381. tomy plus tamoxifen with or without irradiation ogy, American society for radiation oncology, and 246. Portschy PR, Tuttle TM. Contralateral prophylactic in women age 70 years or older with early breast society of surgical oncology focused guideline update. mastectomy: perceptions versus reality. Ann Intern cancer: long-term follow-up of CALGB 9343. J Clin Oncol. 2016. Med. 2013;159(6):428–429. J Clin Oncol. 2013;31(19):2382–2387. 280. Berry DA, Cronin KA, Plevritis SK, et al. Effect 247. Yao K, Belkora J, Sisco M, et al. Survey of the deficits 264. Haviland JS, Owen JR, Dewar JA, et al. The UK of screening and adjuvant therapy on mortality in surgeons’ knowledge of contralateral prophylactic standardisation of breast radiotherapy (START) trials from breast cancer. N Engl J Med. 2005;353(17): mastectomy. JAMA Surg. 2016;151(4):391–393. of radiotherapy hypofractionation for treatment 1784–1792. 1603.e6 Part III: Specific Malignancies

281. Peto R, Davies C, Godwin J, et al. Comparisons of the cancer care Ontario clinical practice guideline. oestrogen-receptor-positive breast cancer on chemo- between different polychemotherapy regimens for J Clin Oncol. 2016;34(20):2416–2427. therapy: a retrospective analysis of a randomised early breast cancer: meta-analyses of long-term 297. Coates AS, Winer EP, Goldhirsch A, et al. Tailoring trial. Lancet Oncol. 2010;11(1):55–65. outcome among 100,000 women in 123 randomised therapies—improving the management of early breast 313. Cuzick J, Dowsett M, Pineda S, et al. Prognostic trials. Lancet. 2012;379(9814):432–444. cancer: st gallen international expert consensus on value of a combined estrogen receptor, progesterone 282. Early Breast Cancer Trialists’ Collaborative Group, the primary therapy of early breast cancer 2015. receptor, Ki-67, and human epidermal growth factor Clarke M, Coates AS, et al. Adjuvant chemotherapy Ann Oncol. 2015;26(8):1533–1546. receptor 2 immunohistochemical score and compari- in oestrogen-receptor-poor breast cancer: patient- 298. Olivotto IA, Bajdik CD, Ravdin PM, et al. son with the genomic health recurrence score in early level meta-analysis of randomised trials. Lancet. Population-based validation of the prognostic model breast cancer. J Clin Oncol. 2011;29(32):4273–4278. 2008;371(9606):29–40. ADJUVANT! for early breast cancer. J Clin Oncol. 314. Dignam JJ, Dukic V, Anderson SJ, Mamounas EP, 283. Early Breast Cancer Trialists’ Collaborative Group, 2005;23(12):2716–2725. Wickerham DL, Wolmark N. Hazard of recurrence Dowsett M, Forbes JF, et al. Aromatase inhibitors 299. Ravdin PM, Siminoff LA, Davis GJ, et al. Computer and adjuvant treatment effects over time in lymph versus tamoxifen in early breast cancer: patient-level program to assist in making decisions about adjuvant node-negative breast cancer. Breast Cancer Res Treat. meta-analysis of the randomised trials. Lancet. therapy for women with early breast cancer. J Clin 2009;116(3):595–602. 2015;386(10001):1341–1352. Oncol. 2001;19(4):980–991. 315. O’Sullivan CC, Bradbury I, Campbell C, et al. 284. Dowsett M, Cuzick J, Wale C, et al. Prediction 300. Wishart GC, Bajdik CD, Dicks E, et al. PREDICT Efficacy of adjuvant trastuzumab for patients with of risk of distant recurrence using the 21-gene plus: development and validation of a prognostic human epidermal growth factor receptor 2-positive recurrence score in node-negative and node-positive model for early breast cancer that includes HER2. early breast cancer and tumors

328. Rimawi MF, Schiff R, Osborne CK. Targeting HER2 344. Gianni L, Pienkowski T, Im YH, et al. Efficacy and 360. von Minckwitz G, Untch M, Blohmer JU, et al. Defi- for the treatment of breast cancer. Annu Rev Med. safety of neoadjuvant pertuzumab and trastuzumab nition and impact of pathologic complete response 2015;66:111–128. in women with locally advanced, inflammatory, or on prognosis after neoadjuvant chemotherapy in 329. Piccart-Gebhart MJ, Procter M, Leyland-Jones B, early HER2-positive breast cancer (NeoSphere): a various intrinsic breast cancer subtypes. J Clin Oncol. et al. Trastuzumab after adjuvant chemotherapy randomised multicentre, open-label, phase 2 trial. 2012;30(15):1796–1804. in HER2-positive breast cancer. N Engl J Med. Lancet Oncol. 2012;13(1):25–32. 361. Hylton NM, Blume JD, Bernreuter WK, et al. 2005;353(16):1659–1672. 345. Romond EH, Jeong JH, Rastogi P, et al. Seven-year Locally advanced breast cancer: MR imaging for 330. Romond EH, Perez EA, Bryant J, et al. Trastuzumab follow-up assessment of cardiac function in NSABP prediction of response to neoadjuvant chemo- plus adjuvant chemotherapy for operable HER2- B-31, a randomized trial comparing doxorubicin therapy—results from ACRIN 6657/I-SPY TRIAL. positive breast cancer. N Engl J Med. 2005;353(16): and cyclophosphamide followed by paclitaxel (ACP) Radiology. 2012;263(3):663–672. 1673–1684. with ACP plus trastuzumab as adjuvant therapy 362. Boughey JC, Suman VJ, Mittendorf EA, et al. 331. Gianni L, Dafni U, Gelber RD, et al. Treatment with for patients with node-positive, human epidermal Sentinel lymph node surgery after neoadjuvant trastuzumab for 1 year after adjuvant chemotherapy growth factor receptor 2-positive breast cancer. J chemotherapy in patients with node-positive breast in patients with HER2-positive early breast cancer: Clin Oncol. 2012;30(31):3792–3799. cancer: the ACOSOG z1071 (Alliance) clinical trial. a 4-year follow-up of a randomised controlled trial. 346. Perez EA, Rodeheffer R. Clinical cardiac tolerability JAMA. 2013;310(14):1455–1461. Lancet Oncol. 2011;12(3):236–244. of trastuzumab. J Clin Oncol. 2004;22(2):322–329. 363. Kuehn T, Bauerfeind I, Fehm T, et al. Sentinel- 332. Perez EA, Romond EH, Suman VJ, et al. Four- 347. Tan-Chiu E, Yothers G, Romond E, et al. Assessment lymph-node biopsy in patients with breast cancer year follow-up of trastuzumab plus adjuvant of cardiac dysfunction in a randomized trial compar- before and after neoadjuvant chemotherapy chemotherapy for operable human epidermal growth ing doxorubicin and cyclophosphamide followed by (SENTINA): a prospective, multicentre cohort factor receptor 2-positive breast cancer: joint analysis paclitaxel, with or without trastuzumab as adjuvant study. Lancet Oncol. 2013;14(7):609–618. of data from NCCTG n9831 and NSABP B-31. therapy in node-positive, human epidermal growth 364. O’Sullivan CC, Connolly RM. Pertuzumab and its J Clin Oncol. 2011;29(25):3366–3373. factor receptor 2-overexpressing breast cancer: accelerated approval: evolving treatment paradigms 333. Slamon D, Eiermann W, Robert N, et al. Adjuvant NSABP B-31. J Clin Oncol. 2005;23(31):7811–7819. and new challenges in the management of HER2- trastuzumab in HER2-positive breast cancer. N Engl 348. Slamon DJ, Eiermann W, Robert NJ, et al. Ten year positive breast cancer. Oncology (Williston Park). 2014; J Med. 2011;365(14):1273–1283. follow-up of BCIRG-006 comparing doxorubicin 28(3):186–194, 196. 334. Joensuu H, Bono P, Kataja V, et al. Fluorouracil, plus cyclophosphamide followed by docetaxel 365. Schneeweiss A, Chia S, Hickish T, et al. Pertu- epirubicin, and cyclophosphamide with either (AC→T) with doxorubicin plus cyclophosphamide zumab plus trastuzumab in combination with docetaxel or vinorelbine, with or without trastu- followed by docetaxel and trastuzumab (AC→TH) standard neoadjuvant anthracycline-containing and zumab, as adjuvant treatments of breast cancer: final with docetaxel, carboplatin and trastuzumab (TCH) anthracycline-free chemotherapy regimens in patients results of the FinHer trial. J Clin Oncol. 2009;27(34): in HER2+ early breast cancer. Cancer Res. 2016;76(4 with HER2-positive early breast cancer: a random- 5685–5692. (suppl)):Abstract S5-04. ized phase II cardiac safety study (TRYPHAENA). 335. Pivot X, Romieu G, Debled M, et al. 6 months 349. Gradishar W, Salerno KE. NCCN guidelines update: Ann Oncol. 2013;24(9):2278–2284. versus 12 months of adjuvant trastuzumab for breast cancer. J Natl Compr Canc Netw. 2016;14(5 366. Rugo HS, Olopade OI, DeMichele A, et al. Adaptive patients with HER2-positive early breast cancer suppl):641–644. randomization of veliparib-carboplatin treatment in (PHARE): a randomised phase 3 trial. Lancet Oncol. 350. Paik S, Kim C, Wolmark N. HER2 status and benefit breast cancer. N Engl J Med. 2016;375(1):23–34. 2013;14(8):741–748. from adjuvant trastuzumab in breast cancer. N Engl 367. Park JW, Liu MC, Yee D, et al. Adaptive randomiza- 336. Goldhirsch A, Gelber RD, Piccart-Gebhart MJ, J Med. 2008;358(13):1409–1411. tion of neratinib in early breast cancer. N Engl J et al. 2 years versus 1 year of adjuvant trastuzumab 351. Perez EA, Reinholz MM, Hillman DW, et al. HER2 Med. 2016;375(1):11–22. for HER2-positive breast cancer (HERA): an and chromosome 17 effect on patient outcome in 368. Guarneri V, Broglio K, Kau SW, et al. Prognostic open-label, randomised controlled trial. Lancet. the n9831 adjuvant trastuzumab trial. J Clin Oncol. value of pathologic complete response after primary 2013;382(9897):1021–1028. 2010;28(28):4307–4315. chemotherapy in relation to hormone receptor 337. Tolaney SM, Barry WT, Dang CT, et al. Adjuvant 352. Arnedos M, Nerurkar A, Osin P, A’Hern R, Smith status and other factors. J Clin Oncol. 2006;24(7): paclitaxel and trastuzumab for node-negative, HER2- IE, Dowsett M. Discordance between core needle 1037–1044. positive breast cancer. N Engl J Med. 2015;372(2): biopsy (CNB) and excisional biopsy (EB) for 369. Burstein HJ, Prestrud AA, Seidenfeld J, et al. 134–141. estrogen receptor (ER), progesterone receptor (PgR) American society of clinical oncology clinical practice 338. Geyer CE, Forster J, Lindquist D, et al. Lapatinib and HER2 status in early breast cancer (EBC). Ann guideline: update on adjuvant endocrine therapy plus capecitabine for HER2-positive advanced breast Oncol. 2009;20(12):1948–1952. for women with hormone receptor–positive breast cancer. N Engl J Med. 2006;355(26):2733–2743. 353. Di Leo A, Gomez HL, Aziz Z, et al. Phase III, cancer. J Clin Oncol. 2010;28(23):3784–3796. 339. Baselga J, Bradbury I, Eidtmann H, et al. Lapatinib double-blind, randomized study comparing lapatinib 370. Burstein HJ, Lacchetti C, Anderson H, et al. Adju- with trastuzumab for HER2-positive early breast plus paclitaxel with placebo plus paclitaxel as first-line vant endocrine therapy for women with hormone cancer (NeoALTTO): a randomised, open-label, treatment for metastatic breast cancer. J Clin Oncol. receptor–positive breast cancer: American society multicentre, phase 3 trial. Lancet. 2012;379(9816): 2008;26(34):5544–5552. of clinical oncology clinical practice guideline 633–640. 354. Golshan M, Cirrincione CT, Sikov WM, et al. update on ovarian suppression. J Clin Oncol. 340. Gianni L, Eiermann W, Semiglazov V, et al. Impact of neoadjuvant therapy on eligibility for 2016;34(14):1689–1701. Neoadjuvant chemotherapy with trastuzumab fol- and frequency of breast conservation in stage II-III 371. Burstein HJ, Temin S, Anderson H, et al. Adjuvant lowed by adjuvant trastuzumab versus neoadjuvant HER2-positive breast cancer: surgical results of endocrine therapy for women with hormone chemotherapy alone, in patients with HER2-positive CALGB 40601 (Alliance). Breast Cancer Res Treat. receptor–positive breast cancer: American society locally advanced breast cancer (the NOAH trial): a 2016;160(2):297–304. of clinical oncology clinical practice guideline focused randomized controlled superiority trial with a parallel 355. Rastogi P, Anderson SJ, Bear HD, et al. Preoperative update. J Clin Oncol. 2014;32(21):2255–2269. HER2-negative cohort. Lancet. 2010;375:377–384. chemotherapy: updates of national surgical adjuvant 372. Griggs JJ, Somerfield MR, Anderson H, et al. Ameri- 341. Carey LA, Berry DA, Cirrincione CT, et al. breast and bowel project protocols B-18 and B-27. can society of clinical oncology endorsement of the Molecular heterogeneity and response to neoadjuvant J Clin Oncol. 2008;26(5):778–785. cancer care Ontario practice guideline on adjuvant human epidermal growth factor receptor 2 targeting 356. Kaufmann M, von Minckwitz G, Mamounas EP, ovarian ablation in the treatment of premenopausal in CALGB 40601, a randomized phase III trial of et al. Recommendations from an international women with early-stage invasive breast cancer. J Clin paclitaxel plus trastuzumab with or without lapatinib. consensus conference on the current status and future Oncol. 2011;29(29):3939–3942. J Clin Oncol. 2016;34(6):542–549. of neoadjuvant systemic therapy in primary breast 373. Aebi S, Gelber S, Castiglione-Gertsch M, et al. Is 342. Piccart-Gebhart MJ, Holmes AP, Baselga J, et al. cancer. Ann Surg Oncol. 2012;19(5):1508–1516. chemotherapy alone adequate for young women with First results from the phase III ALTTO trial (BIG 357. Symmans WF, Peintinger F, Hatzis C, et al. Measure- oestrogen-receptor–positive breast cancer? Lancet. 2-06; NCCTG [alliance] n063d) comparing one ment of residual breast cancer burden to predict 2000;355(9218):1869–1874. year of anti-HER2 therapy with lapatinib alone (L), survival after neoadjuvant chemotherapy. J Clin 374. Goldhirsch A, Gelber RD, Yothers G, et al. Adjuvant trastuzumab alone (T), their sequence (T->L), or Oncol. 2007;25(28):4414–4422. therapy for very young women with breast cancer: their combination (T+L) in the adjuvant treatment 358. Carey LA, Metzger R, Dees EC, et al. American joint need for tailored treatments. J Natl Cancer Inst of Her2-positive early breast cancer (EBC). J Clin committee on cancer tumor-node-metastasis stage Monogr. 2001;30:44–51. Oncol. 2014;32(5S suppl)):abstr LBA4. after neoadjuvant chemotherapy and breast cancer 375. Petrek JA, Naughton MJ, Case LD, et al. Incidence, 343. Chan A, Delaloge S, Holmes FA, et al. Neratinib after outcome. J Natl Cancer Inst. 2005;97(15):1137–1142. time course, and determinants of menstrual bleeding trastuzumab-based adjuvant therapy in patients with 359. Cortazar P, Zhang L, Untch M, et al. Pathological after breast cancer treatment: a prospective study. HER2-positive breast cancer (ExteNET): a multicen- complete response and long-term clinical benefit J Clin Oncol. 2006;24:1045–1051. tre, randomised, double-blind, placebo-controlled, in breast cancer: the CTNeoBC pooled analysis. 376. Early Breast Cancer Trialists’ Collaborative Group. phase 3 trial. Lancet Oncol. 2016;17(3):367–377. Lancet. 2014;384(9938):164–172. Effects of chemotherapy and hormonal therapy 1603.e8 Part III: Specific Malignancies

for early breast cancer on recurrence and 15-year 390. Cuzick J, Sestak I, Baum M, et al. Effect of 407. Dowsett M, Smith IE, Ebbs SR, et al. Prognostic survival: an overview of the randomised trials. Lancet. anastrozole and tamoxifen as adjuvant treatment value of ki67 expression after short-term presurgical 2005;365(9472):1687–1717. for early-stage breast cancer: 10-year analysis of the endocrine therapy for primary breast cancer. J Natl 377. Fisher B, Dignam J, Bryant J, Wolmark N. Five ATAC trial. Lancet Oncol. 2010;11(12):1135–1141. Cancer Inst. 2007;99(2):167–170. versus more than five years of tamoxifen for lymph 391. Regan MM, Neven P, Giobbie-Hurder A, et al. 408. Smith TJ, Bohlke K, Lyman GH, et al. Recommen- node-negative breast cancer: updated findings from Assessment of letrozole and tamoxifen alone and dations for the use of WBC growth factors: American the national surgical adjuvant breast and bowel in sequence for postmenopausal women with steroid society of clinical oncology clinical practice guideline project B-14 randomized trial. J Natl Cancer Inst. hormone receptor–positive breast cancer: the BIG update. J Clin Oncol. 2015;33(28):3199–3212. 2001;93(9):684–690. 1-98 randomised clinical trial at 8.1 years median 409. Rizzo JD, Brouwers M, Hurley P, et al. American 378. Stewart HJ, Prescott RJ, Forrest AP. Scottish adjuvant follow-up. Lancet Oncol. 2011;12(12):1101–1108. society of clinical Oncology/American society of tamoxifen trial: a randomized study updated to 15 392. Bliss JM, Kilburn LS, Coleman RE, et al. Disease- hematology clinical practice guideline update on years. J Natl Cancer Inst. 2001;93(6):456–462. related outcomes with long-term follow-up: an the use of epoetin and darbepoetin in adult patients 379. International Breast Cancer Study Group, updated analysis of the intergroup exemestane study. with cancer. J Clin Oncol. 2010;28(33):4996–5010. Castiglione-Gertsch M, O’Neill A, et al. Adjuvant J Clin Oncol. 2012;30(7):709–717. 410. Zambetti M, Moliterni A, Materazzo C, et al. chemotherapy followed by goserelin versus either 393. van de Velde CJ, Rea D, Seynaeve C, et al. Adjuvant Long-term cardiac sequelae in operable breast modality alone for premenopausal lymph node- tamoxifen and exemestane in early breast cancer cancer patients given adjuvant chemotherapy with negative breast cancer: a randomized trial. J Natl (TEAM): a randomised phase 3 trial. Lancet. or without doxorubicin and breast irradiation. J Cancer Inst. 2003;95(24):1833–1846. 2011;377(9762):321–331. Clin Oncol. 2001;19(1):37–43. 380. Jonat W, Kaufmann M, Sauerbrei W, et al. Gos- 394. Goss PE, Ingle JN, Pater JL, et al. Late extended 411. Perez EA, Suman VJ, Davidson NE, et al. Effect of erelin versus cyclophosphamide, methotrexate, and adjuvant treatment with letrozole improves doxorubicin plus cyclophosphamide on left ventricu- fluorouracil as adjuvant therapy in premenopausal outcome in women with early-stage breast cancer lar ejection fraction in patients with breast cancer patients with node-positive breast cancer: the zoladex who complete 5 years of tamoxifen. J Clin Oncol. in the north central cancer treatment group n9831 early breast cancer research association study. J Clin 2008;26(12):1948–1955. intergroup adjuvant trial. J Clin Oncol. 2004;22(18): Oncol. 2002;20(24):4628–4635. 395. Mamounas EP, Jeong JH, Wickerham DL, et al. 3700–3704. 381. Schmid P, Untch M, Wallwiener D, et al. Benefit from exemestane as extended adjuvant 412. Perez EA, Suman VJ, Davidson NE, et al. Cardiac Cyclophosphamide, methotrexate and fluorouracil therapy after 5 years of adjuvant tamoxifen: inten- safety analysis of doxorubicin and cyclophosphamide (CMF) versus hormonal ablation with leuprorelin tion-to-treat analysis of the national surgical adjuvant followed by paclitaxel with or without trastuzumab acetate as adjuvant treatment of node-positive, breast and bowel project B-33 trial. J Clin Oncol. in the north central cancer treatment group premenopausal breast cancer patients: preliminary 2008;26(12):1965–1971. n9831 adjuvant breast cancer trial. J Clin Oncol. results of the TABLE-study (takeda adjuvant breast 396. Dowsett M, Cuzick J, Ingle J, et al. Meta-analysis 2008;26(8):1231–1238. cancer study with leuprorelin acetate). Anticancer of breast cancer outcomes in adjuvant trials of 413. Armenian SH, Lacchetti C, Barac A, et al. Prevention Res. 2002;22(4):2325–2332. aromatase inhibitors versus tamoxifen. J Clin Oncol. and monitoring of cardiac dysfunction in survivors 382. Boccardo F, Rubagotti A, Amoroso D, et al. 2010;28(3):509–518. of adult cancers. J Clin Oncol. 2016;epub December Cyclophosphamide, methotrexate, and fluorouracil 397. Goss PE, Ingle JN, Pritchard KI, et al. Extending 5, 2016. versus tamoxifen plus ovarian suppression as adju- aromatase-inhibitor adjuvant therapy to 10 years. 414. Schneider BP, Shen F, Gardner L, et al. Genome-wide vant treatment of estrogen receptor–positive pre-/ N Engl J Med. 2016;375(3):209–219. association study for anthracycline-induced conges- perimenopausal breast cancer patients: results of the 398. Goss PE, Ingle JN, Pritchard KI, et al. Exemestane tive heart failure. Clin Cancer Res. 2017;23(1):43–51. Italian breast cancer adjuvant study group 02 ran- versus anastrozole in postmenopausal women 415. Blanco JG, Sun CL, Landier W, et al. Anthracycline- domized trial. J Clin Oncol. 2000;18(14):2718–2727. with early breast cancer: NCIC CTG MA.27—a related cardiomyopathy after childhood cancer: role 383. Jakesz R, Hausmaninger H, Kubista E, et al. randomized controlled phase III trial. J Clin Oncol. of polymorphisms in carbonyl reductase genes—a Randomized adjuvant trial of tamoxifen and gos- 2013;31(11):1398–1404. report from the Children’s oncology group. J Clin erelin versus cyclophosphamide, methotrexate, and 399. Henry NL, Xia R, Banerjee M, et al. Predictors of Oncol. 2012;30(13):1415–1421. fluorouracil: evidence for the superiority of treatment recovery of ovarian function during aromatase inhibi- 416. Hertz DL, Caram MV, Kidwell KM, et al. Evidence with endocrine blockade in premenopausal patients tor therapy. Ann Oncol. 2013;24(8):2011–2016. for association of SNPs in ABCB1 and CBR3, but with hormone-responsive breast cancer—austrian 400. Smith IE, Dowsett M, Yap YS, et al. Adjuvant not RAC2, NCF4, SLC28a3 or TOP2b, with breast and colorectal cancer study group trial 5. J aromatase inhibitors for early breast cancer chronic cardiotoxicity in a cohort of breast cancer Clin Oncol. 2002;20(24):4621–4627. after chemotherapy-induced amenorrhoea: patients treated with anthracyclines. Pharmacogenom- 384. Arriagada R, Le MG, Spielmann M, et al. Random- caution and suggested guidelines. J Clin Oncol. ics. 2016;17(3):231–240. ized trial of adjuvant ovarian suppression in 926 2006;24(16):2444–2447. 417. Wolff AC, Blackford AL, Visvanathan K, et al. Risk premenopausal patients with early breast cancer 401. Goss PE, Ingle JN, Martino S, et al. Impact of of marrow neoplasms after adjuvant breast cancer treated with adjuvant chemotherapy. Ann Oncol. premenopausal status at breast cancer diagnosis in therapy: the national comprehensive cancer network 2005;16(3):389–396. women entered on the placebo-controlled NCIC experience. J Clin Oncol. 2015;33(4):340–348. 385. Davidson NE, O’Neill AM, Vukov AM, et al. Che- CTG MA17 trial of extended adjuvant letrozole. 418. Smith RE, Bryant J, DeCillis A, Anderson S, moendocrine therapy for premenopausal women with Ann Oncol. 2013;24(2):355–361. National Surgical Adjuvant B, Bowel Project E. axillary lymph node-positive, steroid hormone recep- 402. Pan H, Gray RG, Davies C, et al. Predictors of Acute myeloid leukemia and myelodysplastic tor–positive breast cancer: results from INT 0101 recurrence during years 5-14 in 46,138 women with syndrome after doxorubicin-cyclophosphamide (E5188). J Clin Oncol. 2005;23(25):5973–5982. ER+ breast cancer allocated 5 years only of endocrine adjuvant therapy for operable breast cancer: the 386. International Breast Cancer Study Group, therapy (ET). J Clin Oncol. 2016;34(suppl):abstr national surgical adjuvant breast and bowel project Colleoni M, Gelber S, et al. Tamoxifen after adjuvant 505. experience. J Clin Oncol. 2003;21(7):1195–1204. chemotherapy for premenopausal women with lymph 403. Albain KS, Barlow WE, Ravdin PM, et al. Adjuvant 419. Hershman DL, Lacchetti C, Dworkin RH, et al. node-positive breast cancer: international breast cancer chemotherapy and timing of tamoxifen in post- Prevention and management of chemotherapy- study group trial 13-93. J Clin Oncol. 2006;24(9): menopausal patients with endocrine-responsive, induced peripheral neuropathy in survivors of adult 1332–1341. node-positive breast cancer: a phase 3, open-label, cancers: American society of clinical oncology clinical 387. Cuzick J, Ambroisine L, Davidson N, et al. Use of randomised controlled trial. Lancet. 2009;374(9707): practice guideline. J Clin Oncol. 2014;32(18):1941– luteinising-hormone-releasing hormone agonists as 2055–2063. 1967. adjuvant treatment in premenopausal patients with 404. Pico C, Martin M, Jara C, et al. Epirubicin- 420. Bower JE. Behavioral symptoms in patients hormone-receptor-positive breast cancer: a meta- cyclophosphamide adjuvant chemotherapy plus with breast cancer and survivors. J Clin Oncol. analysis of individual patient data from randomised tamoxifen administered concurrently versus sequen- 2008;26(5):768–777. adjuvant trials. Lancet. 2007;369(9574):1711– tially: randomized phase III trial in postmenopausal 421. Goodwin PJ, Ennis M, Pritchard KI, et al. Adjuvant 1723. node-positive breast cancer patients. A GEICAM treatment and onset of menopause predict weight 388. Gnant M, Mlineritsch B, Stoeger H, et al. 9401 study. Ann Oncol. 2004;15(1):79–87. gain after breast cancer diagnosis. J Clin Oncol. Adjuvant endocrine therapy plus zoledronic acid 405. Ellis MJ, Tao Y, Luo J, et al. Outcome prediction 1999;17(1):120–129. in premenopausal women with early-stage breast for estrogen receptor–positive breast cancer based on 422. Jim HS, Phillips KM, Chait S, et al. Meta-analysis cancer: 62-month follow-up from the ABCSG-12 postneoadjuvant endocrine therapy tumor character- of cognitive functioning in breast cancer survivors randomised trial. Lancet Oncol. 2011;12(7):631–641. istics. J Natl Cancer Inst. 2008;100(19):1380–1388. previously treated with standard-dose chemotherapy. 389. Geisler J, Helle H, Ekse D, et al. Letrozole is 406. Dowsett M, Smith IE, Ebbs SR, et al. Proliferation J Clin Oncol. 2012;30(29):3578–3587. superior to anastrozole in suppressing breast cancer and apoptosis as markers of benefit in neoadjuvant 423. Bower JE, Bak K, Berger A, et al. Screening, assess- tissue and plasma estrogen levels. Clin Cancer Res. endocrine therapy of breast cancer. Clin Cancer Res. ment, and management of fatigue in adult survivors 2008;14(19):6330–6335. 2006;12(3 Pt 2):1024s–1030s. of cancer: an American society of clinical oncology Cancer of the Breast • CHAPTER 88 1603.e9

clinical practice guideline adaptation. J Clin Oncol. 441. Ganz PA. Impact of tamoxifen adjuvant therapy on analysis of the monotherapy arms of the BIG 1-98 2014;32(17):1840–1850. symptoms, functioning, and quality of life. J Natl trial. J Clin Oncol. 2011;29(suppl):abstr 522. 424. Rowland JH, Hewitt M, Ganz PA. Cancer survivor- Cancer Inst Monogr. 2001;30:130–134. 459. Henry NL, Stearns V. Treatment-emergent effects ship: a new challenge in delivering quality cancer 442. Day R, Ganz PA, Costantino JP, Cronin WM, may predict benefit from endocrine therapy.J Clin care. J Clin Oncol. 2006;24(32):5101–5104. Wickerham DL, Fisher B. Health-related quality Oncol. 2013;31(18):2233–2235. 425. Runowicz CD, Leach CR, Henry NL, et al. American of life and tamoxifen in breast cancer prevention: a 460. Stearns V, Chapman JA, Ma CX, et al. Treatment- cancer Society/American society of clinical oncology report from the national surgical adjuvant breast and associated musculoskeletal and vasomotor symptoms breast cancer survivorship care guideline. J Clin bowel project P-1 study. J Clin Oncol. 1999;17(9): and relapse-free survival in the NCIC CTG MA.27 Oncol. 2016;34(6):611–635. 2659–2669. adjuvant breast cancer aromatase inhibitor trial. 426. Swain SM, Land SR, Ritter MW, et al. Amenorrhea 443. Pandya KJ, Raubertas RF, Flynn PJ, et al. Oral J Clin Oncol. 2015;33(3):265–271. in premenopausal women on the doxorubicin- clonidine in postmenopausal patients with breast 461. Khatcheressian JL, Hurley P, Bantug E, et al. Breast and-cyclophosphamide-followed-by-docetaxel arm cancer experiencing tamoxifen-induced hot flashes: cancer follow-up and management after primary of NSABP B-30 trial. Breast Cancer Res Treat. a university of rochester cancer center community treatment: American society of clinical oncology 2009;113(2):315–320. clinical oncology program study. Ann Intern Med. clinical practice guideline update. J Clin Oncol. 427. Partridge A, Gelber S, Gelber RD, Castiglione- 2000;132(10):788–793. 2013;31(7):961–965. Gertsch M, Goldhirsch A, Winer E. Age of meno- 444. Jacobson JS, Troxel AB, Evans J, et al. Randomized 462. Ganz PA. Monitoring the physical health of cancer pause among women who remain premenopausal trial of black cohosh for the treatment of hot flashes survivors: a survivorship-focused medical history. following treatment for early breast cancer: long-term among women with a history of breast cancer. J J Clin Oncol. 2006;24(32):5105–5111. results from international breast cancer study Clin Oncol. 2001;19(10):2739–2745. 463. Grunfeld E, Levine MN, Julian JA, et al. Randomized group trials V and VI. Eur J Cancer. 2007;43(11): 445. Gnant M, Pfeiler G, Dubsky PC, et al. Adjuvant trial of long-term follow-up for early-stage breast 1646–1653. denosumab in breast cancer (ABCSG-18): a multi- cancer: a comparison of family physician versus 428. Loprinzi CL, Sloan JA, Stearns V, et al. Newer anti- centre, randomised, double-blind, placebo-controlled specialist care. J Clin Oncol. 2006;24(6):848–855. depressants and gabapentin for hot flashes: an indi- trial. Lancet. 2015;386(9992):433–443. 464. Institutes of Medicine. From Cancer Patient to vidual subject pooled analysis. J Clin Oncol. 2008;26: 446. Van Poznak C, Hannon RA, Mackey JR, et al. Cancer Survivor: Lost in Transition. Washington, 9537a. Prevention of aromatase inhibitor-induced bone loss D.C.: National Academies Press; 2006. 429. Azim HA Jr, Kroman N, Paesmans M, et al. using risedronate: the SABRE trial. J Clin Oncol. 465. Grunfeld E, Julian JA, Pond G, et al. Evaluating Prognostic impact of pregnancy after breast cancer 2010;28(6):967–975. survivorship care plans: results of a randomized, clini- according to estrogen receptor status: a multicenter 447. Baum M, Budzar AU, Cuzick J, et al. Anastrozole cal trial of patients with breast cancer. J Clin Oncol. retrospective study. J Clin Oncol. 2013;31(1):73– alone or in combination with tamoxifen versus 2011;29(36):4755–4762. 79. tamoxifen alone for adjuvant treatment of postmeno- 466. Diel IJ, Solomayer EF, Costa SD, et al. Reduction in 430. Moore HC, Unger JM, Phillips KA, et al. Goserelin pausal women with early breast cancer: first results of new metastases in breast cancer with adjuvant clodro- for ovarian protection during breast-cancer adjuvant the ATAC randomised trial. Lancet. 2002;359(9324): nate treatment. N Engl J Med. 1998;339(6):357–363. chemotherapy. N Engl J Med. 2015;372(10): 2131–2139. 467. Powles T, Paterson A, McCloskey E, et al. Reduc- 923–932. 448. Henry NL, Azzouz F, Desta Z, et al. Predictors tion in bone relapse and improved survival with 431. Bernhard J, Luo W, Ribi K, et al. Patient-reported of aromatase inhibitor discontinuation due to oral clodronate for adjuvant treatment of operable outcomes with adjuvant exemestane versus tamoxifen treatment-emergent symptoms in early-stage breast breast cancer [ISRCTN83688026]. Breast Cancer in premenopausal women with early breast cancer cancer. J Clin Oncol. 2012;30(9):936–942. Res. 2006;8(2):R13. undergoing ovarian suppression (TEXT and SOFT): 449. Barton DL, Sloan JA, Shuster LT, et al. Impact of 468. Saarto T, Blomqvist C, Virkkunen P, Elomaa I. a combined analysis of two phase 3 randomised vaginal dehydroepiandrosterone (DHEA) on vaginal Adjuvant clodronate treatment does not reduce the trials. Lancet Oncol. 2015;16(7):848–858. symptoms in female cancer survivors: trial n10c1 frequency of skeletal metastases in node-positive 432. Shapiro CL, Manola J, Leboff M. Ovarian failure (Alliance). J Clin Oncol. 2014;32(5S):abst 9507. breast cancer patients: 5-year results of a randomized after adjuvant chemotherapy is associated with rapid 450. Goetsch MF, Lim JY, Caughey AB. A practical controlled trial. J Clin Oncol. 2001;19(1):10–17. bone loss in women with early-stage breast cancer. solution for dyspareunia in breast cancer survi- 469. Paterson AH, Anderson SJ, Lembersky BC, et al. J Clin Oncol. 2001;19(14):3306–3311. vors: a randomized controlled trial. J Clin Oncol. Oral clodronate for adjuvant treatment of operable 433. Gralow JR, Biermann JS, Farooki A, et al. NCCN 2015;33(30):3394–3400. breast cancer (national surgical adjuvant breast task force report: bone health in cancer care. J Natl 451. Bell LN, Nguyen AT, Li L, et al. Comparison of and bowel project protocol B-34): a multicentre, Compr Canc Netw. 2013;11(suppl 3):S1–S50. changes in the lipid profile of postmenopausal women placebo-controlled, randomised trial. Lancet Oncol. 434. Love RR, Mazess RB, Barden HS, et al. Effects of with early stage breast cancer treated with exemes- 2012;13(7):734–742. tamoxifen on bone mineral density in postmeno- tane or letrozole. J Clin Pharmacol. 2012;52(12): 470. Coleman RE, Marshall H, Cameron D, et al. Breast- pausal women with breast cancer. N Engl J Med. 1852–1860. cancer adjuvant therapy with zoledronic acid. N 1992;326(13):852–856. 452. Lonning PE, Geisler J, Krag LE, et al. Effects of Engl J Med. 2011;365(15):1396–1405. 435. Bergman L, Beelen ML, Gallee MP, Hollema H, exemestane administered for 2 years versus placebo 471. Early Breast Cancer Trialists’ Collaborative G, Benraadt J, van Leeuwen FE. Risk and prognosis of on bone mineral density, bone biomarkers, and Coleman R, Powles T, et al. Adjuvant bisphospho- endometrial cancer after tamoxifen for breast cancer. plasma lipids in patients with surgically resected early nate treatment in early breast cancer: meta-analyses Comprehensive cancer Centres’ ALERT group. breast cancer. J Clin Oncol. 2005;23:5126–5137. of individual patient data from randomised trials. Assessment of liver and endometrial cancer risk fol- 453. Amir E, Seruga B, Niraula S, Carlsson L, Ocana Lancet. 2015;386(10001):1353–1361. lowing tamoxifen. Lancet. 2000;356(9233):881–887. A. Toxicity of adjuvant endocrine therapy in 472. Gralow J, Barlow WE, Paterson AHG, et al. Phase 436. Wysowski DK, Honig SF, Beitz J. Uterine sarcoma postmenopausal breast cancer patients: a system- III trial of bisphosphonates as adjuvant therapy in associated with tamoxifen use. N Engl J Med. atic review and meta-analysis. J Natl Cancer Inst. primary breast cancer: SWOG/alliance/ECOG- 2002;346(23):1832–1833. 2011;103(17):1299–1309. ACRIN/NCIC clinical trials Group/NRG oncology 437. Barakat RR, Gilewski TA, Almadrones L, et al. 454. Cuzick J, Sestak I, Cella D, Fallowfield L. Treatment- study s0307. J Clin Oncol. 2015;33(suppl):abst 503. Effect of adjuvant tamoxifen on the endometrium emergent endocrine symptoms and the risk of breast 473. van Tienhoven G, Voogd AC, Peterse JL, et al. in women with breast cancer: a prospective study cancer recurrence: a retrospective analysis of the Prognosis after treatment for loco-regional recurrence using office endometrial biopsy. J Clin Oncol. ATAC trial. Lancet Oncol. 2008;9(12):1143–1148. after mastectomy or breast conserving therapy in 2000;18(20):3459–3463. 455. Fontein DB, Houtsma D, Hille ET, et al. Relation- two randomised trials (EORTC 10801 and DBCG- 438. American congress of obstetricians and gynecologists ship between specific adverse events and efficacy of 82TM). EORTC breast cancer cooperative group committee opinion no. 601: tamoxifen and uterine exemestane therapy in early postmenopausal breast and the Danish breast cancer cooperative group. cancer. Obstet Gynecol. 2014;123(6):1394–1397. cancer patients. Ann Oncol. 2012;23(12):3091–3097. Eur J Cancer. 1999;35(1):32–38. 439. Abramson N, Costantino JP, Garber JE, Berliner 456. Fontein DB, Seynaeve C, Hadji P, et al. Specific 474. Aebi S, Gelber S, Anderson SJ, et al. Chemotherapy N, Wickerham DL, Wolmark N. Effect of factor V adverse events predict survival benefit in patients for isolated locoregional recurrence of breast cancer leiden and prothrombin G20210→a mutations on treated with tamoxifen or aromatase inhibitors: an (CALOR): a randomised trial. Lancet Oncol. 2014; thromboembolic risk in the national surgical adjuvant international tamoxifen exemestane adjuvant mul- 15(2):156–163. breast and bowel project breast cancer prevention tinational trial analysis. J Clin Oncol. 2013;31(18): 475. Chia SK, Speers CH, D’Yachkova Y, et al. The impact trial. J Natl Cancer Inst. 2006;98(13):904–910. 2257–2264. of new chemotherapeutic and hormone agents on 440. Garber JE, Halabi S, Tolaney SM, et al. Factor 457. Huober J, Cole BF, Rabaglio M, et al. Symptoms of survival in a population-based cohort of women V leiden mutation and thromboembolism risk in endocrine treatment and outcome in the BIG 1-98 with metastatic breast cancer. Cancer. 2007;110(5): women receiving adjuvant tamoxifen for breast study. Breast Cancer Res Treat. 2014;143(1):159–169. 973–979. cancer. J Natl Cancer Inst. 2010;102(13):942– 458. Huober JB, Cole BF, Wu J, et al. Symptoms of 476. Hortobagyi G. Can we cure limited metastatic breast 949. endocrine treatment and outcome: a retrospective cancer? J Clin Oncol. 2002;20(3):620–623. 1603.e10 Part III: Specific Malignancies

477. Pagani O, Senkus E, Wood W, et al. International growth factor receptor 2–positive breast cancer: 500 mg vs. anastrozole for hormone receptor–positive guidelines for management of metastatic breast American society of clinical oncology clinical practice advanced breast cancer. Ann Oncol. 2016;27(suppl 1): cancer: can metastatic breast cancer be cured? J guideline. J Clin Oncol. 2014;32(19):2078–2099. abstract LBA 14_PR. Natl Cancer Inst. 2010;102(7):456–463. 496. Thurlimann B, Robertson JF, Nabholtz JM, Buzdar 512. Bergh J, Jonsson PE, Lidbrink EK, et al. FACT: an 478. Murthy P, Kidwell KM, Schott AF, et al. Clinical A, Bonneterre J. Efficacy of tamoxifen following open-label randomized phase III study of fulvestrant predictors of long-term survival in HER2-positive anastrozole (‘Arimidex’) compared with anastrozole and anastrozole in combination compared with metastatic breast cancer. Breast Cancer Res Treat. following tamoxifen as first-line treatment for anastrozole alone as first-line therapy for patients 2016;155(3):589–595. advanced breast cancer in postmenopausal women. with receptor-positive postmenopausal breast cancer. 479. Yardley DA, Tripathy D, Brufsky AM, et al. Long- Eur J Cancer. 2003;39(16):2310–2317. J Clin Oncol. 2012;30(16):1919–1925. term survivor characteristics in HER2-positive 497. Nardone A, De Angelis C, Trivedi MV, Osborne 513. Mehta RS, Barlow WE, Albain KS, et al. Combina- metastatic breast cancer from registher. Br J Cancer. CK, Schiff R. The changing role of ER in endocrine tion anastrozole and fulvestrant in metastatic breast 2014;110(11):2756–2764. resistance. Breast. 2015;24(suppl 2):S60–S66. cancer. N Engl J Med. 2012;367(5):435–444. 480. Ricardi U, Badellino S, Filippi AR. Clinical 498. Jeselsohn R, Buchwalter G, De Angelis C, Brown 514. Morikawa A, Henry NL. Palbociclib for the treatment applications of stereotactic radiation therapy for M, Schiff R. ESR1 mutations—a mechanism for of estrogen receptor–positive, HER2-negative meta- oligometastatic cancer patients: a disease-oriented acquired endocrine resistance in breast cancer. Nat static breast cancer. Clin Cancer Res. 2015;21(16): approach. J Radiat Res. 2016. Rev Clin Oncol. 2015;12(10):573–583. 3591–3596. 481. Hanrahan EO, Broglio KR, Buzdar AU, et al. 499. Mauri D, Pavlidis N, Polyzos NP, Ioannidis JP. 515. Finn RS, Martin M, Rugo HS, et al. Palbociclib Combined-modality treatment for isolated recur- Survival with aromatase inhibitors and inactivators and letrozole in advanced breast cancer. N Engl J rences of breast carcinoma: update on 30 years of versus standard hormonal therapy in advanced Med. 2016;375:1925–1936. experience at the university of texas m.D. anderson breast cancer: meta-analysis. J Natl Cancer Inst. 516. Hortobagyi GN, Stemmer SM, Burris HA, et al. cancer center and assessment of prognostic factors. 2006;98(18):1285–1291. Ribociclib as first-line therapy for HR-positive, Cancer. 2005;104(6):1158–1171. 500. Geisler J, Haynes B, Anker G, Dowsett M, Lonning advanced breast cancer. N Engl J Med. 2016;375(18): 482. Cara S, Tannock IF. Retreatment of patients PE. Influence of letrozole and anastrozole on total 1738–1748. with the same chemotherapy: implications for body aromatization and plasma estrogen levels in post- 517. Turner NC, Ro J, Andre F, et al. Palbociclib in clinical mechanisms of drug resistance. Ann Oncol. menopausal breast cancer patients evaluated in a ran- hormone-receptor-positive advanced breast cancer. 2001;12(1):23–27. domized, cross-over study. J Clin Oncol. 2002;20(3): N Engl J Med. 2015;373(3):209–219. 483. Coates AS, Hurny C, Peterson HF, et al. Quality- 751–757. 518. Cristofanilli M, Turner NC, Bondarenko I, et al. of-life scores predict outcome in metastatic but not 501. Geisler J, Ekse D, Helle H, Duong NK, Lonning Fulvestrant plus palbociclib versus fulvestrant early breast cancer. International breast cancer study PE. An optimised, highly sensitive radioimmunoas- plus placebo for treatment of hormone-receptor- group. J Clin Oncol. 2000;18(22):3768–3774. say for the simultaneous measurement of estrone, positive, HER2-negative metastatic breast cancer 484. Geels P, Eisenhauer E, Bezjak A, Zee B, Day A. estradiol and estrone sulfate in the ultra-low range that progressed on previous endocrine therapy Palliative effect of chemotherapy: objective tumor in human plasma samples. J Steroid Biochem Mol (PALOMA-3): final analysis of the multicentre, response is associated with symptom improvement in Biol. 2008;109(1–2):90–95. double-blind, phase 3 randomised controlled trial. patients with metastatic breast cancer. J Clin Oncol. 502. Rose C, Vtoraya O, Pluzanska A, et al. An open Lancet Oncol. 2016;17(4):425–439. 2000;18(12):2395–2405. randomised trial of second-line endocrine therapy in 519. Baselga J, Campone M, Piccart M, et al. Everolimus in 485. Van Poznak C, Somerfield MR, Bast RC, et al. Use advanced breast cancer. comparison of the aromatase postmenopausal hormone-receptor-positive advanced of biomarkers to guide decisions on systemic therapy inhibitors letrozole and anastrozole. Eur J Cancer. breast cancer. N Engl J Med. 2012;366(6):520–529. for women with metastatic breast cancer: American 2003;39(16):2318–2327. 520. Wolff AC, Lazar AA, Bondarenko I, et al. Random- society of clinical oncology clinical practice guideline. 503. Giordano SH, Hortobagyi GN. Leuprolide acetate ized phase III placebo-controlled trial of letrozole J Clin Oncol. 2015;33(24):2695–2704. plus aromatase inhibition for male breast cancer. J plus oral temsirolimus as first-line endocrine therapy 486. Karlsson E, Appelgren J, Solterbeck A, Bergenheim Clin Oncol. 2006;24(21):e42–e43. in postmenopausal women with locally advanced M, Alvariza V, Bergh J. Breast cancer during follow- 504. Tan SH, Wolff AC. Luteinizing hormone-releasing or metastatic breast cancer. J Clin Oncol. 2013; up and progression—a population based cohort on hormone agonists in premenopausal hormone recep- 31(2):195–202. new cancers and changed biology. Eur J Cancer. tor–positive breast cancer. Clin Breast Cancer. 2007; 521. Kaufman B, Mackey JR, Clemens MR, et al. 2014;50(17):2916–2924. 7(6):455–464. Trastuzumab plus anastrozole versus anastrozole alone 487. Amir E, Clemons M, Purdie CA, et al. Tissue 505. Hughes LL, Gray RJ, Solin LJ, et al. Efficacy of for the treatment of postmenopausal women with confirmation of disease recurrence in breast cancer radiotherapy for ovarian ablation: results of a breast human epidermal growth factor receptor 2-positive, patients: pooled analysis of multi-centre, multi- intergroup study. Cancer. 2004;101(5):969–972. hormone receptor–positive metastatic breast cancer: disciplinary prospective studies. Cancer Treat Rev. 2012; 506. Klijn JG, Blamey RW, Boccardo F, Tominaga T, results from the randomized phase III TAnDEM 38(6):708–714. Duchateau L, Sylvester R. Combined tamoxifen and study. J Clin Oncol. 2009;27(33):5529–5537. 488. Darvishian F, Singh B, Krauter S, Chiriboga L, luteinizing hormone–releasing hormone (LHRH) 522. Schwartzberg LS, Franco SX, Florance A, O’Rourke Gangi MD, Melamed J. Impact of decalcifica- agonist versus LHRH agonist alone in premeno- L, Maltzman J, Johnston S. Lapatinib plus letrozole tion on receptor status in breast cancer. Breast J. pausal advanced breast cancer: a meta-analysis of as first-line therapy for HER-2+ hormone recep- 2011;17(6):689–691. four randomized trials. J Clin Oncol. 2001;19(2): tor–positive metastatic breast cancer. Oncologist. 489. Hamaoka T, Madewell JE, Podoloff DA, Hortobagyi 343–353. 2010;15(2):122–129. GN, Ueno NT. Bone imaging in metastatic breast 507. Carlson RW, Theriault R, Schurman CM, et al. 523. Palumbo R, Sottotetti F, Riccardi A, et al. Which cancer. J Clin Oncol. 2004;22(14):2942–2953. Phase II trial of anastrozole plus goserelin in the patients with metastatic breast cancer benefit from 490. Vogel CL, Schoenfelder J, Shemano I, Hayes DF, treatment of hormone receptor–positive, metastatic subsequent lines of treatment? An update for Gams RA. Worsening bone scan in the evaluation carcinoma of the breast in premenopausal women. clinicians. Ther Adv Med Oncol. 2013;5(6):334– of antitumor response during hormonal therapy of J Clin Oncol. 2010;28(25):3917–3921. 350. breast cancer. J Clin Oncol. 1995;13(5):1123–1128. 508. Di Leo A, Jerusalem G, Petruzelka L, et al. Results of 524. Rivera E, Holmes FA, Frye D, et al. Phase II study 491. Manohar K, Mittal BR, Senthil R, Kashyap R, the CONFIRM phase III trial comparing fulvestrant of paclitaxel in patients with metastatic breast Bhattacharya A, Singh G. Clinical utility of f-18 250 mg with fulvestrant 500 mg in postmenopausal carcinoma refractory to standard chemotherapy. FDG PET/CT in recurrent breast carcinoma. Nucl women with estrogen receptor–positive advanced Cancer. 2000;89(11):2195–2201. Med Commun. 2012;33(6):591–596. breast cancer. J Clin Oncol. 2010;28(30):4594–4600. 525. Paridaens R, Biganzoli L, Bruning P, et al. Paclitaxel 492. Lin NU, Winer EP. Brain metastases: the HER2 509. Chia S, Gradishar W, Mauriac L, et al. Double-blind, versus doxorubicin as first-line single-agent chemo- paradigm. Clin Cancer Res. 2007;13(6):1648–1655. randomized placebo controlled trial of fulvestrant therapy for metastatic breast cancer: a European 493. Partridge AH, Rumble RB, Carey LA, et al. Che- compared with exemestane after prior nonsteroidal organization for research and treatment of cancer motherapy and targeted therapy for women with aromatase inhibitor therapy in postmenopausal randomized study with cross-over. J Clin Oncol. human epidermal growth factor receptor 2-negative women with hormone receptor–positive, advanced 2000;18(4):724–733. (or unknown) advanced breast cancer: American breast cancer: results from EFECT. J Clin Oncol. 526. Perez EA, Lerzo G, Pivot X, et al. Efficacy and safety society of clinical oncology clinical practice guideline. 2008;26(10):1664–1670. of ixabepilone (BMS-247550) in a phase II study J Clin Oncol. 2014;32(29):3307–3329. 510. Ellis MJ, Llombart-Cussac A, Feltl D, et al. of patients with advanced breast cancer resistant to 494. Rugo HS, Rumble RB, Macrae E, et al. Endocrine Fulvestrant 500 mg versus anastrozole 1 mg for an anthracycline, a taxane, and capecitabine. J Clin therapy for hormone receptor–positive metastatic the first-line treatment of advanced breast cancer: Oncol. 2007;25(23):3407–3414. breast cancer: American society of clinical oncology overall survival analysis from the phase II FIRST 527. Thomas ES, Gomez HL, Li RK, et al. Ixabepilone guideline. J Clin Oncol. 2016;34(25):3069–3103. study. J Clin Oncol. 2015;33(32):3781–3787. plus capecitabine for metastatic breast cancer pro- 495. Giordano SH, Temin S, Kirshner JJ, et al. Systemic 511. Elllis MJ, Bondarenko IM, Trishkina E, et al. gressing after anthracycline and taxane treatment. therapy for patients with advanced human epidermal FALCON: a phase III randomised trial of fulvestrant J Clin Oncol. 2007;25(33):5210–5217. Cancer of the Breast • CHAPTER 88 1603.e11

528. Gradishar WJ, Tjulandin S, Davidson N, et al. Phase breast cancer: a german breast group 26/breast 560. Port ER, Fey JV, Cody HS 3rd, Borgen PI. Sentinel III trial of nanoparticle albumin-bound paclitaxel international group 03-05 study. J Clin Oncol. lymph node biopsy in patients with male breast compared with polyethylated castor oil-based 2009;27(12):1999–2006. carcinoma. Cancer. 2001;91(2):319–323. paclitaxel in women with breast cancer. J Clin Oncol. 544. Guan Z, Xu B, DeSilvio ML, et al. Randomized 561. Zagouri F, Sergentanis TN, Azim HA Jr, Chrysikos 2005;23(31):7794–7803. trial of lapatinib versus placebo added to paclitaxel D, Dimopoulos MA, Psaltopoulou T. Aromatase 529. Gradishar WJ, Krasnojon D, Cheporov S, et al. in the treatment of human epidermal growth factor inhibitors in male breast cancer: a pooled analysis. Significantly longer progression-free survival with receptor 2-overexpressing metastatic breast cancer. Breast Cancer Res Treat. 2015;151(1):141–147. nab-paclitaxel compared with docetaxel as first-line J Clin Oncol. 2013;31(16):1947–1953. 562. Amant F, von Minckwitz G, Han SN, et al. Prognosis therapy for metastatic breast cancer. J Clin Oncol. 545. Blackwell KL, Burstein HJ, Storniolo AM, et al. Ran- of women with primary breast cancer diagnosed 2009;27(22):3611–3619. domized study of lapatinib alone or in combination during pregnancy: results from an international 530. Rugo HS, Barry WT, Moreno-Aspitia A, et al. with trastuzumab in women with ErbB2-positive, collaborative study. J Clin Oncol. 2013;31(20): Randomized phase III trial of paclitaxel once per trastuzumab-refractory metastatic breast cancer. 2532–2539. week compared with nanoparticle albumin-bound J Clin Oncol. 2010;28(7):1124–1130. 563. Amant F, Loibl S, Neven P, Van Calsteren K. Breast nab-paclitaxel once per week or ixabepilone with 546. Bachelot T, Romieu G, Campone M, et al. Lapatinib cancer in pregnancy. Lancet. 2012;379(9815): bevacizumab as first-line chemotherapy for locally plus capecitabine in patients with previously untreated 570–579. recurrent or metastatic breast cancer: CALGB brain metastases from HER2-positive metastatic 564. Loibl S, Han SN, von Minckwitz G, et al. Treatment 40502/NCCTG n063h (Alliance). J Clin Oncol. breast cancer (LANDSCAPE): a single-group phase of breast cancer during pregnancy: an observational 2015;33(21):2361–2369. 2 study. Lancet Oncol. 2013;14(1):64–71. study. Lancet Oncol. 2012;13(9):887–896. 531. Cortes J, O’Shaughnessy J, Loesch D, et al. Eribulin 547. Swain SM, Baselga J, Kim SB, et al. Pertuzumab, 565. Pagani O, Partridge A, Korde L, et al. Pregnancy monotherapy versus treatment of physician’s choice in trastuzumab, and docetaxel in HER2-positive after breast cancer: if you wish, ma’am. Breast Cancer patients with metastatic breast cancer (EMBRACE): metastatic breast cancer. N Engl J Med. 2015; Res Treat. 2011;129(2):309–317. a phase 3 open-label randomised study. Lancet. 372(8):724–734. 566. Borges VF. Management of the patient with 2011;377(9769):914–923. 548. Verma S, Miles D, Gianni L, et al. Trastuzumab postpartum breast cancer. Oncology (Williston Park). 532. Kaufman PA, Awada A, Twelves C, et al. Phase III emtansine for HER2-positive advanced breast cancer. 2014;28(9):768–770. open-label randomized study of eribulin mesylate N Engl J Med. 2012;367(19):1783–1791. 567. Callihan EB, Gao D, Jindal S, et al. Postpartum versus capecitabine in patients with locally advanced 549. Krop IE, Kim SB, Gonzalez-Martin A, et al. Trastu- diagnosis demonstrates a high risk for metastasis and or metastatic breast cancer previously treated with zumab emtansine versus treatment of physician’s merits an expanded definition of pregnancy-associated an anthracycline and a taxane. J Clin Oncol. 2015; choice for pretreated HER2-positive advanced breast breast cancer. Breast Cancer Res Treat. 2013;138(2): 33(6):594–601. cancer (TH3RESA): a randomised, open-label, phase 549–559. 533. O’Shaughnessy JA. Pegylated liposomal doxorubicin 3 trial. Lancet Oncol. 2014;15(7):689–699. 568. Buchanan CL, Morris EA, Dorn PL, Borgen PI, Van in the treatment of breast cancer. Clin Breast Cancer. 550. Ellis PA, Barrios CH, Eiermann W, et al. Phase Zee KJ. Utility of breast magnetic resonance imaging 2003;4(5):318–328. III, randomized study of trastuzumab emtansine in patients with occult primary breast cancer. Ann 534. O’Brien ME, Wigler N, Inbar M, et al. Reduced (T-DM1) ± pertuzumab (P) vs trastuzumab + taxane Surg Oncol. 2005;12(12):1045–1053. cardiotoxicity and comparable efficacy in a phase (HT) for first-line treatment of HER2-positive MBC: 569. Olson JA Jr, Morris EA, Van Zee KJ, Linehan DC, III trial of pegylated liposomal doxorubicin HCl primary results from the MARIANNE study. J Clin Borgen PI. Magnetic resonance imaging facilitates (CAELYX/Doxil) versus conventional doxorubicin Oncol. 2015;33(suppl):abstr 507. breast conservation for occult breast cancer. Ann for first-line treatment of metastatic breast cancer. 551. Miller K, Wang M, Gralow J, et al. Paclitaxel plus Surg Oncol. 2000;7(6):411–415. Ann Oncol. 2004;15(3):440–449. bevacizumab versus paclitaxel alone for metastatic 570. Varadarajan R, Edge SB, Yu J, Watroba N, 535. Blum JL, Dieras V, Lo Russo PM, et al. Multicenter, breast cancer. N Engl J Med. 2007;357(26): Janarthanan BR. Prognosis of occult breast carcinoma phase II study of capecitabine in taxane-pretreated 2666–2676. presenting as isolated axillary nodal metastasis. metastatic breast carcinoma patients. Cancer. 552. Robert NJ, Dieras V, Glaspy J, et al. RIBBON-1: Oncology. 2006;71(5–6):456–459. 2001;92(7):1759–1768. randomized, double-blind, placebo-controlled, 571. Reinfuss M, Mitus J, Duda K, Stelmach A, Rys J, 536. Harbeck N, Saupe S, Jager E, et al. A randomized phase III trial of chemotherapy with or without Smolak K. The treatment and prognosis of patients phase III study evaluating pegylated liposomal bevacizumab for first-line treatment of human with phyllodes tumor of the breast: an analysis of doxorubicin versus capecitabine as first-line therapy epidermal growth factor receptor 2-negative, locally 170 cases. Cancer. 1996;77(5):910–916. for metastatic breast cancer: results of the PELICAN recurrent or metastatic breast cancer. J Clin Oncol. 572. Macdonald OK, Lee CM, Tward JD, Chappel study. Breast Cancer Res Treat. 2016. 2011;29(10):1252–1260. CD, Gaffney DK. Malignant phyllodes tumor of 537. Isakoff SJ, Mayer EL, He L, et al. TBCRC009: 553. Miles DW, Chan A, Dirix LY, et al. Phase III the female breast: association of primary therapy a multicenter phase II clinical trial of platinum study of bevacizumab plus docetaxel compared with cause-specific survival from the surveillance, monotherapy with biomarker assessment in meta- with placebo plus docetaxel for the first-line treat- Epidemiology, and end results (SEER) program. static triple-negative breast cancer. J Clin Oncol. ment of human epidermal growth factor receptor Cancer. 2006;107(9):2127–2133. 2015;33(17):1902–1909. 2-negative metastatic breast cancer. J Clin Oncol. 573. Barth RJ Jr, Wells WA, Mitchell SE, Cole BF. 538. Albain KS, Nag SM, Calderillo-Ruiz G, et al. 2010;28(20):3239–3247. A prospective, multi-institutional study of Gemcitabine plus paclitaxel versus paclitaxel mono- 554. Solomayer EF, Diel IJ, Meyberg GC, Gollan C, adjuvant radiotherapy after resection of malignant therapy in patients with metastatic breast cancer and Bastert G. Metastatic breast cancer: clinical course, phyllodes tumors. Ann Surg Oncol. 2009;16(8):2288– prior anthracycline treatment. J Clin Oncol. 2008; prognosis and therapy related to the first site of 2294. 26(24):3950–3957. metastasis. Breast Cancer Res Treat. 2000;59(3): 574. Kapiris I, Nasiri N, A’Hern R, Healy V, Gui GP. 539. Berry DA, Ueno NT, Johnson MM, et al. High- 271–278. Outcome and predictive factors of local recurrence dose chemotherapy with autologous hematopoietic 555. Van Poznak CH, Temin S, Yee GC, et al. American and distant metastases following primary surgical stem-cell transplantation in metastatic breast cancer: society of clinical oncology executive summary of treatment of high-grade malignant phyllodes tumours overview of six randomized trials. J Clin Oncol. the clinical practice guideline update on the role of of the breast. Eur J Surg Oncol. 2001;27(8):723–730. 2011;29(24):3224–3231. bone-modifying agents in metastatic breast cancer. 575. Zeng S, Zhang X, Yang D, Wang X, Ren G. Effects of 540. Johnston S, Pippen J Jr, Pivot X, et al. Lapatinib J Clin Oncol. 2011;29(9):1221–1227. adjuvant radiotherapy on borderline and malignant combined with letrozole versus letrozole and placebo 556. Stopeck AT, Lipton A, Body JJ, et al. Denosumab phyllodes tumors: a systematic review and meta- as first-line therapy for postmenopausal hormone compared with zoledronic acid for the treatment of analysis. Mol Clin Oncol. 2015;3(3):663–671. receptor–positive metastatic breast cancer. J Clin bone metastases in patients with advanced breast 576. Sikov WM, et al. Impact of the addition of Oncol. 2009;27(33):5538–5546. cancer: a randomized, double-blind study. J Clin carboplatin and/or bevacizumab to neoadjuvant 541. Vogel CL, Cobleigh MA, Tripathy D, et al. Efficacy Oncol. 2010;28(35):5132–5139. once-per-week paclitaxel followed by dose-dense and safety of trastuzumab as a single agent in first- 557. Himelstein AL, Foster JC, Khatcheressian JL, doxorubicin and cyclophosphamide on pathologic line treatment of HER2-overexpressing metastatic et al. Effect of longer-interval vs standard dosing complete response rates in stage II to III triple- breast cancer. J Clin Oncol. 2002;20(3):719– of zoledronic acid on skeletal events in patients with negative breast cancer: CALGB 40603 (Alliance). 726. bone metastases: a randomized clinical trial. JAMA. J Clin Oncol. 2015;33(1):13–21. 542. Slamon DJ, Leyland-Jones B, Shak S, et al. Use of 2017;317(1):48–58. 577. von Minckwitz G, et al. Neoadjuvant carboplatin in chemotherapy plus a monoclonal antibody against 558. Woodward WA. Inflammatory breast cancer: unique patients with triple-negative and HER2-positive early HER2 for metastatic breast cancer that overexpresses biological and therapeutic considerations. Lancet breast cancer (GeparSixto; GBG 66): a randomised HER2. N Engl J Med. 2001;344(11):783–792. Oncol. 2015;16(15):e568–e576. phase 2 trial. Lancet Oncol. 2014;15(7):747–756. 543. von Minckwitz G, du Bois A, Schmidt M, et al. 559. Giordano SH, Buzdar AU, Smith TL, Kau SW, 578. Rae JM, et al. CYP2D6 and UGT2B7 genotype and Trastuzumab beyond progression in human epi- Yang Y, Hortobagyi GN. Is breast cancer survival risk of recurrence in tamoxifen-treated breast cancer dermal growth factor receptor 2-positive advanced improving? Cancer. 2004;100(1):44–52. patients. J Natl Cancer Inst. 2012;104(6):452–460. 1603.e12 Part III: Specific Malignancies

579. Regan MM, et al. CYP2D6 genotype and with chemotherapy-induced painful peripheral for alleviating hot flashes, N07C1. J Clin Oncol. tamoxifen response in postmenopausal women neuropathy: a randomized clinical trial. JAMA. 2010;28(4):641–647. with endocrine-responsive breast cancer: the breast 2013;309(13):1359–1367. 585. Loprinzi CL, et al. Newer antidepressants and gaba- international group 1-98 trial. J Natl Cancer Inst. 582. Barton DL, et al. Phase III, placebo-controlled trial pentin for hot flashes: an individual patient pooled 2012;104(6):441–451. of three doses of citalopram for the treatment of analysis. J Clin Oncol. 2009;27(17):2831–2837. 580. Harris LN, et al. Use of biomarkers to guide deci- hot flashes: NCCTG trial N05C9. J Clin Oncol. sions on adjuvant systemic therapy for women with 2010;28(20):3278–3283. early-stage invasive breast cancer: American Society 583. Loprinzi CL, et al. Newer antidepressants and of Clinical Oncology Clinical Practice Guideline. J gabapentin for hot flashes: a discussion of trial Clin Oncol. 2016;34(10):1134–1150. duration. Menopause. 2009;16(5):883–887. 581. Smith EM, et al. Effect of duloxetine on pain, 584. Loprinzi CL, et al. Phase III, randomized, double- function, and quality of life among patients blind, placebo-controlled evaluation of pregabalin