or ovarian cancer.157 Once an individual tests positive for a mutation, she should inform her first- and

second-degree relatives of their risk status and the potential for DNA testing.

Figure 74-16. Pedigree of genetic breast cancer. The family illustrated in the pedigree carries a mutation of the BRCA1 gene;

multiple relatives are affected with breast and ovarian cancer.

Other breast cancer susceptibility genes include p53, ATM, CDH1, and PTEN. The Li–Fraumeni or

SBLA syndrome corresponds with p53 mutations and includes a broad spectrum of cancers including

sarcoma, breast cancer and brain tumors, lung and laryngeal cancer, leukemia, lymphoma, and an

adrenal cortical carcinoma.158 The ATM gene mutation results in a childhood neurologic disorder

referred to as ataxia—telangiectasia, which is characterized by progressive cerebellar and neuromotor

deterioration, telangiectasia of the conjunctivae and facial skin, and growth retardation. Females with

the ATM mutation have a five times higher risk for developing breast cancers than the general

population.159 Females with the CDH1 or E-cadherin gene mutation have a 70% lifetime risk for diffuse

gastric cancer and a 40% risk for lobular breast cancer.160 Distinctive mucocutaneous lesions referred to

as trichilemmomas characterize Cowden disease, which is associated with an increased risk of breast,

thyroid, and genitourinary tract cancers in women, resulting from a PTEN gene mutation.161

Clinical Risk Assessment

Through large population studies, numerous factors that can increase breast cancer risk have been

identified as summarized in Table 74-8. In an effort to quantitate these potential risks such that

preventive strategies can be better tailored, multiple risk assessment tools have been developed

including the BCRAT or the Gail model 2,162 BRCAPRO,163 BOADICEA,164 Tyrer-Cuzick,165 and the

Breast Cancer Surveillance Consortium.166 The Gail model calculates risk over a defined time interval

and predicts risk accurately in groups of women 35 years of age and older who undergo annual

mammography. The model is not for women who have a strong family history of breast cancer (such as

genetic mutation carriers) or those with LCIS. The components of the Gail model include current age,

age at menarche, age at time of first live birth, number of first-degree relatives with breast cancer,

number of prior breast biopsies, prior breast biopsy with atypia, and race. A sample risk calculation is

shown in Figure 74-17. Women are considered to be at high risk if their 5-year risk is >1.7%, or if their

lifetime risk of breast cancer is at least 20%.

Table 74-7 Estimated Lifetime Cancer Risks for BRCA1 and BRCA2 Mutations (To

Age 70)

2058

CLASSIFICATION

Table 74-8 Magnitude of Known Breast Cancer Risk Factors

Management

All women at high risk for breast cancer should undergo genetic counseling, with education provided

about the risk factors, genetics, and natural history of breast cancer. These women should be followed

with enhanced surveillance, and chemoprevention or prophylactic surgery should be considered. For

women who have a genetic predisposition, such as those with a BRCA1 or BRCA2 mutation, many

experts support aggressive surveillance including CBE biannually beginning at age 20 years and annual

mammography and MRI screening performed in a staggered fashion such that imaging occurs every 6

months starting at age 25. The use of screening MRI for women at high risk is supported by numerous

studies (see the section on Screening MRI). Prophylactic mastectomies can be considered at any age;

many women prefer to defer this until after childbearing as they would like to breast-feed. Ovarian

cancer screening for BRCA1 and BRCA2 carriers includes transvaginal ovarian ultrasound, serum CA125, and annual pelvic examinations starting at age 30. When childbearing is complete, prophylactic

oophorectomies should be considered. In women with a family history but no known genetic

abnormality, screening mammography should begin 10 years before the age of onset in the youngest

diagnosed first-degree relative or at age 40. Furthermore, in the absence of a genetic abnormality,

women who have a lifetime risk of breast cancer >20% should also be monitored with annual screening

MRI.

2059

Figure 74-17. Breast cancer risk assessment according to the Gail model. This woman’s 5-year risk for the development of breast

cancer is 1.6%, and her lifetime risk is 18%.

In addition to its role in the treatment of breast cancer, endocrine therapy (chemoprevention) can

reduce the risk of invasive and/or in situ breast cancers in women at high risk for breast cancer. Women

who are known to benefit from chemoprevention include those older than 35 years with a history of

LCIS, DCIS, or atypical proliferative lesions of the breast, women between 35 and 59 years of age with

a Gail model risk of breast cancer >1.7% over 5 years, and women with known BRCA1 or BRCA2

mutations who do not undergo bilateral prophylactic mastectomies. Currently, there are no data

regarding the risks and benefits of chemoprevention in male patients with breast cancer and as such, the

use of these agents in men is not recommended.

For pre- and postmenopausal women at increased risk of breast cancer, both tamoxifen and raloxifene

can be used. As selective estrogen receptor modulators (SERMs), both agents block the effects of

endogenous estrogen on normal breast tissue and breast cancer. Using data from four trials in the 2013

USPSTF meta-analysis, tamoxifen has been shown to reduce the risk of estrogen receptor (ER)-positive

breast cancer (RR, 0.70; 95% CI, 0.59 to 0.82). Similarly, using data from two trials in the 2013 USPSTF

meta-analysis, raloxifene reduced the risk of ER-positive breast cancers (RR, 0.44; 95% CI, 0.27 to

0.71).167,168 In terms of the choice between tamoxifen and raloxifene, the STAR trial directly compared

tamoxifen with raloxifene and showed that the former was slightly more effective at preventing

invasive breast cancer (Fig. 74-18). However, neither reduced breast cancer-specific or all-cause

mortality rates.169

Both tamoxifen and raloxifene have estrogenic activity on the bone resulting in a decreased incidence

of fractures, while only tamoxifen has estrogenic activity on the uterus and coagulation pathways.

Therefore, tamoxifen use is associated with an increased risk of endometrial cancer (4 cases in 1,000

women, RR 2.13; 95% CI, 1.36 to 3.32) and thromboembolic events (4 cases in 1,000 women; RR 1.93;

95% CI, 1.41 to 2.64),167 with a trend toward higher stroke rates and pulmonary embolism.170 Despite

these risks, quality-of-life measures (depression, anxiety, major psychosocial outcomes) show no

differences in women receiving the drug versus placebo. However, an increase in vasomotor symptoms,

gynecologic symptoms, and problems of sexual functioning was noted.171,172 As congenital anomalies

have been associated with tamoxifen, women with childbearing potential must implement an effective

contraceptive method. For women desiring a pregnancy, tamoxifen should be discontinued for at least 2

months prior to attempt at pregnancy.

2060

Figure 74-18. Both tamoxifen and raloxifene when taken for 5 years in the National Surgical Adjuvant Breast and Bowel Project P2 (STAR trial) were equally effective at reducing the risk of invasive breast cancer.

In postmenopausal women, studies show that aromatase inhibitors (AIs) are a reasonable alternative

to tamoxifen and raloxifene. AIs suppress plasma estrogen levels by inhibition of aromatase, an enzyme

that converts androgens to estrogens. Although AIs are commonly used for breast cancer treatment,

they are not approved in the United States for breast cancer prevention at this time. Also, while both

the USPSTF and the American Society of Clinical Oncology (ASCO) support the use of SERMs for both

premenopausal and postmenopausal women, only ASCO considers the use of exemestane as a reasonable

alternative to SERMs for postmenopausal women.173 More recently, the International Breast Cancer

Intervention Study (IBIS-II trial) showed that high-risk postmenopausal women who took the AI

anastrozole had a 50% reduction in the number of invasive breast cancers and DCIS compared with

placebo. The side effects of anastrozole include musculoskeletal problems, hypertension, vaginal

dryness, and vasomotor symptoms.174 AIs may also be associated with a loss of bone density.

The decision to proceed with endocrine therapy chemoprevention must be individualized. In general,

the data support the use of endocrine therapy versus observation. For premenopausal and

postmenopausal women, tamoxifen and raloxifene are both reasonable options. If a woman’s primary

concern is breast cancer prevention, then tamoxifen should be chosen. However, if there are risk factors

for uterine cancer and or thromboembolic complications, raloxifene is a reasonable alternative. If a

postmenopausal woman does not want to take a SERM, then an AI can be prescribed.

DUCTAL CARCINOMA IN SITU

DCIS is commonly defined as the proliferation of malignant epithelial cells within the mammary ductal

system with no evidence of invasion into the surrounding stroma on routine light microscopic

examination (Fig. 74-19). It is further categorized by its size, nuclear grade, presence and extent of

comedo necrosis, and architectural pattern. Common histologic subtypes of DCIS include comedo,

cribriform, micropapillary, papillary, and solid.175 DCIS is designated as Tis (DCIS) by the TNM staging

system and considered a stage 0 breast cancer.176 The incidence of DCIS increases with age and is

uncommon in women younger than 30 years. In women aged 50 to 64 years, the risk is as high as 88

per 100,000 women.177 Among the breast cancers diagnosed in the United States, 25% are DCIS.178 The

risk of development of metastases and/or death in women with DCIS is rare (<1%).179

Figure 74-19. Photomicrograph of ductal carcinoma in situ. The abnormal cells do not cross the ductal basement membrane. The

necrotic debris in the center of the duct is responsible for the calcification visible on mammography.

Most patients with DCIS will present with no breast-related symptoms or findings on physical

examination. More than 90% of all cases of DCIS are detected on imaging studies – commonly as

microcalcifications on mammography.180,181 Prior to routine screening mammography, DCIS presented

as a palpable mass, in association with nipple discharge, or as Paget disease.182,183 Although most

patients with DCIS have microcalcifications of indeterminate morphology, patterns that are highly

suggestive of DCIS include linear branching or segmental types of pleomorphic microcalcifications.181

Once an abnormal lesion is detected by diagnostic imaging, the diagnosis of DCIS is typically

confirmed by breast biopsy. Of note, when DCIS is diagnosed initially by core biopsy, 15% to 20% of

lesions will be upgraded to invasive cancer at the time of definitive resection. Large, high-grade DCIS

lesions are more likely to have an invasive component.184

2061

5 Surgical options for the treatment of DCIS include breast-conserving surgery as a component of

breast conservation therapy (BCT) (partial mastectomy and postoperative radiation therapy) or simple

mastectomy. Although there is no randomized trial that has ever compared the treatment of DCIS by

mastectomy with treatment by BCT, a large number of clinical studies have shown that the local

recurrence rate of BCT is 10% to 15% after 10 years and that approximately half of the local

recurrences are invasive carcinoma.185–189 Mastectomy is curative in 98% of patients with DCIS,

regardless of the tumor grade or the size of the lesion.185,186 However, both surgical treatment options

are associated with equivalent long-term survival as shown in both retrospective and prospective

analyses.190–192 Given that partial mastectomy is a much less invasive surgical procedure, BCT is

typically recommended over mastectomy. Mastectomy should be performed if BCT is contraindicated,

or based on patient preference. BCT is contraindicated with multicentric disease, extensive disease

where resection would result in poor cosmesis, and in patients who cannot receive radiation therapy.

For patients undergoing mastectomy, immediate breast reconstruction options should be considered.

BCT typically includes a partial mastectomy with negative surgical margins, followed by radiation

therapy to eradicate any residual disease. Radiation therapy is often recommended, except in rare

circumstances (see the section on Breast Cancer in the Elderly). Four prospective, randomized clinical

trials – NSABP B-17188,189 (1998, 2001), EORTC 10853193 (2006), United Kingdom194 (2003), and the

SweDCIS195 (2008) have focused on the role of radiation therapy in the treatment of DCIS (Table 74-9).

Despite differences in study design, the results are very similar, demonstrating a 50% to 60% reduction

in local recurrence with surgical excision and radiation therapy compared to surgical excision alone.

Furthermore, the results of NSABP B-17 confirm that the reduction in local recurrence persists through

12 years of follow-up. However, no survival benefit from radiation therapy was observed. Patients with

comedo necrosis, uncertain margins, and/or those at highest risk for recurrence appear to benefit most

from radiation therapy.196 However, even in the most favorable subgroup (absent or slight comedo

necrosis, clear margins), radiation therapy was associated with a reduction in the absolute incidence of

breast recurrence by 7% at 8 years.

Given the risks, inconvenience, and cost associated with radiation therapy, attempts have been made

to define a subgroup of patients with DCIS who may not require radiation therapy. The ECOG 5194

study investigated patients undergoing local excision alone in women with low-risk DCIS (low- to

intermediate-grade DCIS lesions were <2.5 cm, and high-grade DCIS lesions were <1.0 cm). Margins

were at least 3 mm and a postexcision mammogram was obtained for all patients to document the

adequacy of surgical resection. Tamoxifen following excision was allowed but not required. At a median

follow-up of 6.7 years, the local recurrence rate at 5 years was 6.1% (95% CI, 4.1% to 8.2%) for

patients with low- to intermediate-grade DCIS (n = 565), and 15.3% (95% CI, 8.2% to 22.5%) for

patients with high-grade DCIS (n = 105).197 These findings suggest that local excision alone may be

sufficient for select patients with low- to intermediate-grade DCIS. However, identifying these patients

remains a challenge.

RESULTS

Table 74-9 Results of Randomized Trials of Radiotherapy in Ductal Carcinoma in

Situ

Other models that have attempted to predict the likelihood of local recurrence following excision

alone in patients with DCIS include the Van Nuys Prognostic Index (VNPI)198 and its updated version,

the University of Southern California Van Nuys Prognostic Index (USC/VNPI).199 Patients are classified

into three groups on the basis of pathologic classification: (1) non–high-grade DCIS without comedotype necrosis, (2) non–high-grade DCIS with comedo-type necrosis, and (3) high-grade DCIS with or

2062

without comedo-type necrosis. The USC/VNPI score (range is 4 to 12) is then calculated on the basis of

the pathologic classification score and points given for margins, tumor size, and the patient’s age. The

study’s conclusion is that women with an overall low score (4, 5, 6) did not require radiation therapy.

However, attempt to replicate and validate the USC/VNPI score in other data sets have had varying

success

200–202 and as such it is not commonly used.

More recently, gene expression analysis has been explored in an attempt to distinguish patients who

may omit postexcision radiation therapy. The DCIS Recurrence Score is a multigene assay that was

developed using tumors from 327 patients who participated in the ECOG5194 trial as described

previously.203 The DCIS Recurrence Score is designed to stratify patients into three groups

corresponding to the 10-year risk of developing ipsilateral DCIS or invasive breast cancer. Those with a

low score (<39) had a risk of ipsilateral recurrence of 12% and 5%, respectively; intermediate score

(39–54) had a risk of ipsilateral recurrence of 25% and 9%; and high score (>55) had a risk of

ipsilateral recurrence of 27% and 19%. It must be noted that for some patients, a 12% risk of an

ipsilateral recurrence at 10 years may not justify omission of radiation therapy. Furthermore, validation

of this multigene assay is required prior to its implementation into clinical practice.

Axillary lymph node involvement in DCIS is rare, and sentinel lymph node biopsy is not indicated

during breast-conserving surgery. In the NSABP B-17 and B-24 trials, the axillary recurrence rates,

irrespective of treatment, were 0.83 and 0.36 per 1,000 patient-years of follow-up.204 If a patient is

undergoing a mastectomy following core needle biopsy, SLN biopsy should be considered as it is

possible that an occult invasive breast cancer will be identified, and SLNB is often not feasible after

mastectomy. Of note, SLNB after a mastectomy has been explored through skin mapping. Radiocolloid

with or without blue dye is injected superior to the mastectomy incision into the skin flap, with

successful identification of a sentinel lymph node in 13 out of 20 patients (65% success rate).205 Our

institution has had slightly higher success rates with the injection of both radiocolloid and blue dye.

After breast-conserving surgery, a postexcision mammogram should be obtained prior to the initiation

of radiation therapy to ensure that there are no persistent residual suspicious calcifications. Treatment

for DCIS after breast conserving surgery is not considered complete without adjuvant radiation therapy

(see the section on Radiation Therapy).

There is no role for chemotherapy in DCIS. However, endocrine therapy is commonly recommended.

Fifty percent to seventy-five percent of DCIS specimens express estrogen and progesterone

receptors.206,207 In the United States, only tamoxifen has been approved to prevent breast cancer

recurrences in women with DCIS. Currently, no data exist to support the use of AIs in this population.

The benefit of tamoxifen therapy in decreasing local recurrence rates is supported by both a metaanalysis of two randomized trials and by NSABP B-24. The meta-analysis consists of 3,375 women and

shows that tamoxifen after surgery for DCIS reduces recurrence of both ipsilateral DCIS (HR, 0.75; 95%

CI, 0.61 to 0.92) and contralateral DCIS (HR, 0.50; 95% CI, 0.28 to 0.87) (Fig. 74-20). There is also a

trend toward decreased ipsilateral invasive cancer (HR, 0.79; 95% CI, 0.62 to 1.01) and reduced

contralateral invasive cancer (HR, 0.57; 95% CI, 0.39 to 0.83). The number needed to treat in order for

tamoxifen to have a protective effect against all breast events was 15. However, there was no difference

in all-cause mortality (RR, 1.11; 95% CI, 0.89 to 1.39).208 In a retrospective review of NSABP B-24, it

was found that at 10 years of follow-up, patients (n = 732) with ER-positive DCIS treated with

tamoxifen had significant decreases in any subsequent (invasive and/or noninvasive; ipsilateral and/or

contralateral) breast cancer events compared with those who received placebo (HR, 0.58; 95% CI, 0.42

to 0.81; p <0.0015).209

The NCCN Task Force guidelines for management of patients with DCIS who have not undergone

bilateral mastectomies include consideration of tamoxifen for 5 years rather than observation,

recognizing that although there is an improvement in recurrence rates, there is no proven survival

benefit.210 In women who have undergone bilateral mastectomies, the risks associated with tamoxifen

use likely outweigh the benefits. Furthermore, tamoxifen is not typically recommended for patients

with ER-negative DCIS.

EVALUATION OF THE PATIENT WITH BREAST CANCER

Breast cancer is a heterogenous disease that can be characterized further on the basis of clinical and

pathologic features. Further characterization and staging of breast cancer in patients with newly

diagnosed breast cancer can facilitate definitive treatment. Pathologic, biologic, and anatomic

characteristics are used in combination to predict prognosis and response to therapy.

2063