Breast Cancer Overview

Risk Factors, Screening, Genetic Testing, and Prevention

Epidemiology
Etiology and risk factors
Genetic cancer risk assessment
Signs and symptoms
Screening and diagnosis
Prevention
Staging and prognosis
Suggested reading

Breast cancer is the most common malignancy in women, accounting for 32% of all female cancers. Breast cancer also is responsible for 18% of cancer deaths in women, making it the number two cause of cancer death. An estimated 193,700 new breast cancer cases will be diagnosed in the United States in the year 2001, and 40,600 women will die from this cancer. With improved awareness on the part of both women and health-care providers, however, more breast cancers are being diagnosed while still in situ.

This chapter provides an overview of breast cancer, with discussions of epidemiology, etiology and risk factors, genetic cancer risk assessment, signs and symptoms, screening and diagnosis, prevention (including lifestyle changes and chemoprevention), staging, and prognosis. The three chapters to follow focus on the management of stages 0 and I, stage II, and stages III and IV breast cancer, respectively.

Epidemiology

Gender Breast cancer is relatively uncommon in men; the female to male ratio is approximately 100:1.

Age The risk of developing breast cancer increases with age. The disease is uncommon in women under the age of 40 years; only about 0.8% of breast cancers occur in women < 30 years old and approximately 6.5% develop in women between 30 and 40 years old.

Race White women have a higher overall rate of breast cancer than African-American women; however, this difference is not apparent until age 40 and is marked only after menopause. The incidence of breast cancer in US Asian and Hispanic women is approximately half that in US Caucasian women. Breast cancer risk is extremely low in Native American women.

Geography There is at least a fivefold variation in the incidence of breast cancer among different countries, although this difference appears to be narrowing. The incidence of breast cancer is significantly lower in Japan, Thailand, Nigeria, and India than in Denmark, the Netherlands, New Zealand, Switzerland, the United Kingdom, and the United States. It has been suggested that these trends in breast cancer incidence may be related, in some way, to dietary influences, particularly dietary fat consumption (see “Etiology and risk factors”).

Socioeconomic status The incidence of breast cancer is greater in women of higher socioeconomic background. This relationship is most likely related to lifestyle differences, such as age at first birth.

Disease site The left breast is involved more frequently than the right, and the most common locations of the disease are the upper outer quadrant and retroareolar region.

Survival Survival rates for patients with nonmetastatic breast cancer (all stages) have improved in recent years (Table 1). These improvements may be secondary to advances in adjuvant chemotherapy and radiation therapy. In addition, early detection of recurrent disease after breast-conservation therapy allows for salvage surgery.

Etiology and risk factors

Numerous risk factors have been associated with the development of breast cancer, including genetic, environmental, hormonal, and nutritional influences. Despite all of the available data on breast cancer risk factors, 75% of women with this cancer have no risk factors.

Genetic factors Hereditary forms of breast cancer constitute only 5%-7% of breast cancer cases overall. However, the magnitude of the probability that a woman will develop cancer if she inherits a highly penetrant cancer gene mutation justifies the intense interest in predictive testing. Commercial testing is available for several genes (BRCA1, BRCA2, and p53) that are associated with a high risk for breast cancer development (see “Genetic cancer risk assessment”).

Elevated risk for breast cancer is also associated with mutations in the PTEN gene in Cowden’s syndrome (described below), and modest increased risk (relative risk of 3.9-6.4) may be seen in women who are heterozygous for a mutation in the ATM gene, which is associated with the recessive disease ataxia-telangiectasia in the homozygous state. It has been suggested that other BRCA genes will be discovered as well.

BRCA1 gene The BRCA1 gene is located on chromosome 17. This gene is extremely large and complex, and more than 500 different mutations have been discovered, distributed along the entire gene. BRCA1 mutations are inherited in an autosomal-dominant fashion and are associated with an increased risk for breast, ovarian, and to a lesser degree, prostate cancers. A BRCA1 mutation carrier has a lifetime risk of developing breast cancer on the order of 56%-85%, and a 15%-45% lifetime risk of developing ovarian cancer.

BRCA2 gene The BRCA2 gene has been localized to chromosome 13. BRCA2 is approximately twice as large as BRCA1, and similarly complex.

Alterations in BRCA2 have been associated with an increased incidence of breast cancer in both women (similar to BRCA1) and men (6% lifetime risk). Increased risk for ovarian cancer, pancreatic cancer, and melanoma has also been reported. Together, BRCA1 and BRCA2 account for most hereditary breast and ovarian cancer families, and approximately half of hereditary breast cancer families.

The incidence of BRCA gene mutations in the general breast cancer population is unknown since most of the data have come from studies of high-risk populations. In one population-based study of women with breast cancer, only 9.4% of women < 35 years of age at the time of diagnosis and 12.0% of women < 45 years old who also had a first-degree relative with breast cancer had germline BRCA1 or BRCA2 mutations. However, a 40-year-old woman of Ashkenazi Jewish ancestry who has breast cancer has a 20%-30% probability of bearing one of three founder BRCA gene mutations based on data from high-risk clinics, testing vendors, and Israeli series probability of bearing a BRCA gene mutation.

Li-Fraumeni syndrome This rare syndrome is characterized by premenopausal breast cancer in combination with childhood sarcoma, brain tumors, leukemia and lymphoma, and adrenocortical carcinoma. Tumors frequently occur in childhood and early adulthood, and often present as multiple primaries in the same individual. Germline mutations in the p53 gene on chromosome 17p have been documented in persons with this syndrome. Inheritance is autosomal dominant with a penetrance of at least 50% by age 50.

Cowden’s syndrome is inherited as an autosomal-dominant trait and is manifested primarily by mucocutaneous lesions. Patients with this uncommon syndrome have a higher incidence of GI polyps and thyroid disorders; lifetime estimates for breast cancer among women with this syndrome range from 25% to 50%. Germline mutations in the PTEN gene, located on chromosome 10q23 are responsible for this syndrome.

Family history The overall relative risk of breast cancer in a woman with a positive family history in a first-degree relative (mother, daughter, or sister) is 1.7. Premenopausal onset of the disease in a first-degree relative is associated with a three-fold increase in breast cancer risk, whereas postmenopausal diagnosis increases relative risk by only 1.5. When the first-degree relative has bilateral disease, there is a fivefold increase in risk. The relative risk for a woman whose first-degree relative developed bilateral breast cancer prior to menopause is nearly nine.

Proliferative breast disease The diagnosis of certain conditions after breast biopsy is also associated with an increased risk for the subsequent development of invasive breast cancer. These include moderate or florid ductal hyperplasia and sclerosing adenosis, which pose only a slightly increased risk of breast cancer (1.5-2 times); atypical ductal or lobular hyperplasia, which moderately increases risk (4-5 times); and lobular carcinoma in situ (LCIS), which markedly increases risk (8-11 times). Patients who have a family history of breast cancer along with a personal history of atypical epithelial hyperplasia have an eight-fold increase in breast cancer risk when compared to patients with a positive family history alone and an 11-fold increase in breast cancer risk when compared to patients who do not have atypical hyperplasia and have a negative family history.

Personal cancer history A personal history of breast cancer is a significant risk factor for the subsequent development of a second, new breast cancer. This risk has been estimated to be as high as 1% per year from the time of diagnosis of an initial sporadic breast cancer. The risk for development of a second primary breast cancer is significantly higher for women with hereditary breast cancer, approximately 5% per year (50%-60% lifetime risk). Women with a history of endometrial, ovarian, or colon cancer also have a higher likelihood of developing breast cancer than those with no history of these malignancies.

Menstrual and reproductive factors Early onset of menarche (< 12 years old) has been associated with a modest increase in breast cancer risk (two-fold or less). Women who undergo menopause before age 30 have a two-fold reduction in breast cancer risk when compared to women who undergo menopause after age 55. A first full-term pregnancy before age 30 appears to have a protective effect against breast cancer, whereas a late first full-term pregnancy or nulliparity may be associated with higher risk. There is also a suggestion that lactation protects against breast cancer development.

Radiation exposure An increased rate of breast cancer has been observed in survivors of the atomic bomb explosions in Japan, with a peak latency period of 15-20 years. More recently, it has been noted that patients with Hodgkin’s disease who are treated with mantle irradiation, particularly women who are under age 20 at the time of radiation therapy, have an increased incidence of breast cancer.

Exogenous hormone use The data on the possible association between use of oral contraceptives or hormone replacement therapy and breast cancer are controversial. Some data suggest that prolonged use of oral contraceptives by nulliparous women or the use of oral contraceptives before a first full-term pregnancy heightens breast cancer risk. In addition, the literature concerning the possible breast cancer risk posed by prolonged use of estrogens in perimenopausal and postmenopausal women is inconclusive.

There are clear benefits of exogenous hormone replacement for menopausal women, which include reductions in cardiovascular disease and osteoporosis, as well as alleviation of menopausal symptoms. Patients considering hormone replacement therapy should carefully weigh the risks and benefits.

Alcohol Moderate alcohol intake (two or more drinks per day) appears to modestly increase breast cancer risk.

High-fat diet Diets that are high in fat have been associated with an increased risk for breast cancer. As mentioned previously, it has been suggested that differences in dietary fat content may account for the variations in breast cancer incidence observed among different countries.

Obesity Alterations in endogenous estrogen levels secondary to obesity may enhance breast cancer risk.

Genetic cancer risk assessment

Dramatic advances in our understanding of the genetic bases for cancer have led to the development of new technologies and tools for genetic cancer risk assessment. Tests for BRCA1 and BRCA2 mutations, responsible for the majority of hereditary breast and ovarian cancer (HBOC) families, are now available commercially.

Genetic testing clearly has the potential to benefit carefully selected and counseled families. Education and adequately trained health care professionals are key elements in the successful integration of genetic cancer risk assessment into clinical practice.

The genetic risk assessment process begins with an assessment of perceived risk and the impact of cancer on the patient and her family. This information forms the framework for counseling (Table 2).

Comprehensive personal and family history Detailed information regarding personal, reproductive, and hormonal risk factors is noted. Family history, including age at disease onset, types of cancer, and current age or age at death, is obtained for all family members in going back at least three generations.

Documentation of cancer cases is crucial to accurate risk estimation. Pathology reports, medical record notes, and death certificates may all be used in determining the exact diagnosis.

Pedigree construction and evaluation The family pedigree is then constructed and analyzed to determine whether a pattern of cancer in the family is consistent with genetic disease. Sometimes, small family structure or lack of information about the family limits assessment of a hereditary trait; other times, clues, such as ancestry or early age at diagnosis, influence risk assessment and the usefulness of genetic testing.

Individual risk assessment Empiric cancer risk estimates are derived from the information gathered, as well as an estimate of the likelihood that a detectable BRCA1 or BRCA2 mutation is responsible for the disease in the family. The BRCAPRO computer program is a cancer risk assessment tool that uses a family history of breast or ovarian cancer in first- and second-degree relatives, and includes a Bayesian calculation to account for age-specific penetrance differences, to calculate the probabilities that either a BRCA1 or BRCA2 mutation is responsible for the disease.

Education about the principles of genetics and hereditary cancer patterns is provided. Information on the application of genetic testing (appropriateness, limitations, advantages, and disadvantages) is also given.

Genetic counseling and testing Informed consent is obtained before genetic testing is performed. For individuals who decide to undergo testing, a post-test counseling session is scheduled to disclose and explain the results in person.

Customized screening and prevention recommendations Regardless of whether or not the woman undergoes genetic testing, a customized management plan is delineated, with the goal of prevention or early detection of malignancy, within the context of her personal preferences and degree of risk.

Follow-up care and support The genetic cancer risk assessment service also provides follow-up care and support. This may include cancer surveillance measures, as well as assistance with family dynamics and advising patients about sharing information with at-risk relatives.

Models for predicting the likelihood of a BRCA mutation

Several studies have assessed the frequency of BRCA1 or BRCA2 mutations in women with breast or ovarian cancer from clinical referral centers. These data are likely to be subject to some selection biases. Personal and family characteristics that are associated with an increased likelihood of a BRCA1 or BRCA2 mutation are summarized in Table 3.

Laboratory methods

Several techniques/strategies for detecting mutations in cancer genes have been adopted by different researchers and commercial vendors. Current technology misses 8%-10% of pathologic alterations in BRCA1 and BRCA2.

Directed assays are available for specific founder or ancestral mutations that are common in a gene population. Among Ashkenazi Jews, 1 in 40 individuals bear one of three founder mutations (185delAG and 5382insC in BRCA1, and 6174delT in BRCA2), and these mutations account for 25% of early-onset breast cancer in this population.

Limitations All of the approaches to detecting mutations have limitations. In general, discovery of an inactivating or “deleterious” mutation of either BRCA1 or BRCA2 indicates a high probability that the person will develop breast and/or ovarian cancer.

One of the greatest challenges is the interpretation of missense mutations. These mutations are more likely to be significant if located in an evolutionarily conserved or functionally critical region of the protein. In the absence of a clear disease association, it is often difficult to exclude the possibility that a given missense alteration simply represents a rare polymorphism. A testing service may designate such changes as “genetic variants of uncertain significance.”

Testing strategies

In general, testing should be initiated with the youngest affected individual in a given family. Even if one is convinced that a family has HBOC based on clinical criteria, there is only a 50% chance that an offspring or sibling of an affected patient will have inherited the deleterious allele. Therefore, only a positive test (detection of a known or likely deleterious mutation) is truly informative.

Until the “familial mutation” is known, a negative test result could mean either that the unaffected person being tested did not inherit the cancer susceptibility mutation, or that the person inherited the disease-associated gene but the mutation was not detectable by the methods used.

In many cases, no affected family members are available for testing. In that case, one may proceed with genetic testing of an unaffected person, but only after she has been thoroughly counseled regarding its risks, benefits, and limitations.

Unless there is suggestive family history, cancer susceptibility testing is not considered appropriate for screening unaffected individuals in the general population. However, it may be reasonable to test unaffected persons who are members of an ethnic group in which specific ancestral mutations are prevalent and whose family structure is limited (ie, the family is small, with few female relatives or no information due to premature death from noncancerous causes).

Impact of genetic cancer risk status on initial management

Data from the Breast Cancer Linkage Consortium (BCLC) suggest that the cumulative risk of developing a second primary breast cancer is approximately 5% per year (up to 65% by age 70) among BRCA gene mutation carriers who have already had a breast cancer. Thus, knowledge of the genetic status of a woman affected with breast cancer might influence the initial surgical approach (eg, bilateral mastectomy might be recommended for a mutation carrier instead of a more conservative procedure). Moreover, since ovarian cancer risk may be markedly increased in women with BRCA1 mutations (and to a lesser degree with BRCA2 mutations), additional measures, such as surveillance for presymptomatic detection of early-stage tumors or consideration of oophorectomy, may be warranted. According to data from BRCA1 carriers who underwent oophorectomy, breast cancer risk may also be decreased by the reduction in ovarian hormone exposure.

Recent retrospective data suggest that prophylactic or preventive bilateral mastectomy significantly decreases (by approximately 90%), but does not eliminate, the risk of developing breast cancer in women with a family history of the disease. The preferred surgical technique for breast cancer risk reduction is total complete mastectomy; skin-sparing techniques should involve the removal of all mammary tissue, including the nipple and areola.

Potential benefits and risks of genetic testing

The ability to identify individuals at highest risk for cancer holds the promise of improved prevention and early detection of cancers. Patients who are not at high risk can be spared anxiety and the need for increased surveillance. Recent studies suggest a better emotional state among at-risk relatives who undergo testing than among those who choose not to know their status. The patient’s perception of risk is often much higher than risk estimated by current models.

Potential risks Potential medical, psychological, and socioeconomic risks must be addressed in the context of obtaining informed consent for genetic testing.

Concerns about insurance Fear about adverse effects of testing on insurability remains the premier concern among patients. Close behind that is concern about the costs of analyzing large complex genes ($2,400 for BRCA1 and BRCA2) in an uncertain insurance coverage and reimbursement environment.

Legal and privacy issues The legal and privacy issues surrounding genetic testing are as complex as the testing technologies. Although several state laws regarding the privacy of medical information, genetic testing, and insurance and employment discrimination have been passed, they vary widely.

The 1996 Health Insurance Portability and Accountability Act (US public law 104-191) stipulates that genetic information may not be treated as a preexisting condition in the absence of a diagnosis of the condition related to such information. It further prohibits group medical plans from basing rules for eligibility or costs for coverage on genetic information. However, the law does not address genetic privacy issues and doses not cover individual policies. Many states have laws addressing genetic discrimination, but gaps remain.

ASCO recommendations for genetic testing

The American Society of Clinical Oncology (ASCO) recommends that cancer predisposition testing be offered only when: (1) the person has a strong family history of cancer or very early onset of disease; (2) the test can be adequately interpreted; and (3) the results will influence the medical management of the patient or family member. The National Comprehensive Cancer Network (NCCN) recently published practice guidelines for genetics/familial high-risk cancer screening.

Signs and symptoms

Mammographic findings Increasing numbers of breast malignancies are being discovered in asymptomatic patients through the use of screening mammography. Mammographic features suggestive of malignancy include asymmetry, microcalcifications, a mass, or an architectural distortion (see Figure 1, Figure 2, Figure 3, Figure 4, and Figure 5).

When these features are identified on a screening mammogram, they should, in most cases, be further evaluated with a diagnostic mammogram (and, in some cases, with a breast ultrasound), prior to determining the need for a tissue diagnosis. Often, pseudolesions, such as those caused by summation artifact, dust on the mammographic cassettes, and dermal calcifications, are correctly identified in this manner. All mammographic lesions (and the examinations themselves) must be unambiguously categorized according to one of the six Breast Imaging Reporting and Data System (BI-RAD) classifications developed by the American College of Radiology (Table 4).

Breast lump When signs or symptoms are present, the most common presenting complaint is a lump within the breast. The incidence of this complaint can range from 65% to 76%, depending on the study.

Paget’s disease has been associated with intraductal carcinoma involving the terminal ducts of the breast and may have an associated invasive component. It presents as an eczematoid change in the nipple, a breast mass, or bloody nipple discharge. Cytology may be helpful in establishing the diagnosis; however, negative cytologic results should not preclude a biopsy.

Other local symptoms Breast pain is the presenting symptom in ~5% of patients, breast enlargement in 1%, skin or nipple retraction in ~5%, nipple discharge in ~2%, and nipple crusting or erosion in 1%.

Screening and diagnosis

Screening

Breast self-examination The American Cancer Society (ACS) recommends that women begin monthly breast self-examination at the age of 20. A recent meta-analysis of 12 studies involving a total of 8,118 patients with breast cancer correlated the performance of breast self-examination with tumor size and regional lymph node status. Women who performed breast self-examination were more likely to have smaller tumors and less likely to have axillary node metastases than those who did not.

A major problem with breast self-examination as a screening technique is that it is rarely performed well. Only 2%-3% of women do an ideal examination a year after instruction has been provided.

Clinical breast examination The American Cancer Society recommends that women begin clinical breast examination at the age of 20, and have an exam every 3 years between ages 20 and 39, and annually beginning at age 40. Beginning at age 40, the clinical breast examination should be timed to occur near to/prior to screening mammography. If the clinician detects an abnormality, the patient should then undergo diagnostic imaging rather than screening. Clinical breast examination should be performed and a complete breast history obtained when a woman presents for routine health care. The clinical examination should include inspection and palpation of the breast and regional lymph nodes. Between 14% and 21% of breast cancers are detected by clinical breast examination.

Mammography The American Cancer Society, the American College of Radiology, and the American Medical Association each have updated their guidelines since 1997 and recommend annual mammography beginning at age 40. The National Cancer Institute (NCI) also updated their guidelines in 1997, recommending that women undergo screening mammography every 1-2 years beginning in their 40s.

Screening mammography is performed in the asymptomatic patient to detect an occult breast cancer. This contrasts with diagnostic mammography, which is performed in a patient with a breast abnormality (palpable mass, bloody nipple discharge, or some other clinical finding) to further identify the etiology of the problem.

Physical examination and mammography are complementary. Mammography has a sensitivity of 85%-90% and, thus, would miss 10%-15% of clinically evident tumors, while detecting the majority of cases an average of 2 years prior to any perceptible clinical signs or symptoms.

Screening recommendations for average-risk patients No upper age limit has been suggested, and the previous recommendation for a "baseline" mammogram between the ages of 35 and 40 has been withdrawn.

Screening recommendations for high-risk patients Based on epidemiologic evidence that premenopausal familial breast cancer often presents at similar ages among affected family members, many breast imaging centers recommend that yearly screening for such high-risk individuals begin 5-10 years prior to the youngest age at which their first-degree relative was diagnosed with breast cancer. For example, according to this algorithm, a woman whose mother developed breast cancer at age 45 could begin yearly screening at age 35, in addition to biannual clinical breast examinations. Screening for women at genetic risk may begin at age 25.

Evaluation of a cystic mass

Fine-needle aspiration (FNA) When a dominant breast mass is present and the history and physical examination suggest that it is a cyst, the mass can simply be aspirated with a fine needle. Aspiration of a simple benign breast cyst should yield nonbloody fluid and result in complete resolution of the lesion.

Ultrasonography can also be used to determine whether a lesion is solid or cystic, and also whether a cyst is simple or complex. A complex cyst does not meet the strict criteria of a simple cyst. For example, a complex cyst may demonstrate low level echoes within the cyst fluid or a thickened cyst wall. These features may also be caused by cyst aspiration (presumably due to postaspiration bleeding).

Biopsy A biopsy is indicated if the cyst fluid is bloody, the lesion does not resolve completely after aspiration, or the cyst recurs after repeated aspirations. Cytologic examination of the fluid is not routinely indicated, as the yield for positive cytology is so low. Cystic carcinoma accounts for < 1% of all breast cancers. However, an intraluminal solid mass is a worrisome sign suggesting (intra) cystic carcinoma, and should be biopsied.

Evaluation of a solid mass

A solid mass can be evaluated in a variety of ways. The decision to observe a patient with a breast mass that appears to be benign should be made only after careful clinical, radiologic, and cytologic examinations.

Mammography is used to assess the radiologic characteristics of the mass and is important for the evaluation of the remainder of the ipsilateral breast as well as the contralateral breast.

FNA is a simple, easy-to-perform method for obtaining material for cytologic examination. The overall incidence of false-positive results ranges from 0%-2.5% (0.7% when done by experienced technicians) and the incidence of false-negatives varies from 3% to 27% (3%-9% in experienced hands). Reasons for false-negative readings include less-than-optimal techniques in preparing the cytologic material, missing the lesion on aspiration, tumor necrosis, and incorrect cytologic interpretation.

Biopsy A core biopsy (18 gauge or larger needle biopsy) can be advantageous since architectural as well as cellular characteristics can be evaluated. An excisional biopsy, in which the entire breast mass is removed, definitively establishes the diagnosis. When the mass is extremely large, an incisional biopsy (which entails removal of only a portion of the mass) may be more appropriate.

Evaluation of nonpalpable mammographic abnormalities

Excisional biopsy Prior to 1991, almost all nonpalpable mammographic lesions were diagnosed by surgical excision, and this remains a major diagnostic tool today. Prior to surgery, a breast imager places a hook-wire at the lesion in order to guide the surgeon to it accurately. After the target lesion has been excised, a specimen film is then obtained to ensure that it was successfully removed and, in some cases, to assess the gross adequacy of the margins around the lesion.

Stereotactic- and ultrasound-guided core biopsies have revolutionized the management of nonpalpable mammographic lesions, and currently the majority of these lesions can be diagnosed with these percutaneous techniques. At various facilities around the United States, the percentage of benign (false-positive) breast biopsies with these techniques ranges from 60% to 93%.

Stereotactic-guided core biopsy Several different biopsy devices are available, from a 14-gauge core biopsy needle, to an 11-gauge vacuum-assisted biopsy gun (Mammotome and MIB), to a 20-mm wide percutaneous excisional cannula (ABBI). With each device, the lesion is accurately localized in three dimensions by the use of a stereotactic table, which takes a pair of mammographic images at a fixed angle to each other for lesion triangulation.

Numerous studies comparing the sensitivity and specificity of stereotactic biopsy vs biopsy have consistently found the two procedures to be statistically equivalent. A recent large series demonstrated a false-negative rate of 1.4% for stereotactic core biopsy after long-term follow-up, which equals best published results with surgical biopsy.

Up to 80% of nonpalpable mammographic lesions are candidates for stereotactic core biopsy. Lesions near the chest wall or immediately behind the nipple often cannot be reached on the stereotactic table. Diffuse lesions, such as scattered calcifications or a large asymmetric density, are subject to undersampling with the percutaneous approaches. Some patients are unable to lie prone on the stereotactic table for the duration of the examination. Finally, stereotactic units and trained personnel are not universally available.

Ultrasound-guided core biopsies are another accurate percutaneous technique, useful for lesions best imaged by ultrasound. Since the biopsy gun is hand-held and guided in real time by the ultrasound imager, there is more variability in performance, depending on the experience and skill of the practitioner. The overall reported accuracy rate of ultrasound-guided biopsy is comparable to rates achieved with stereotactic and surgical biopsies.

Ultrasound- or stereotactic-guided FNA is another biopsy option. Although somewhat less invasive than core biopsy, FNA provides only cytologic (not histologic) pathology results. This technique can result in both false-positive and false-negative results, whereas a false-positive has not been reported to date for core breast biopsies. FNA is most successful in centers that have an experienced cytopathologist, who, ideally, is available on site to review smears for adequacy during FNA procedures.

Breast MRI is a sensitive tool for detecting occult breast cancer foci. Due to its limited specificity and high cost, however, MRI is not likely to become a screening tool.

MRI is currently used primarily to search for a subtle primary breast carcinoma in a patients with metastatic disease, to evaluate the extent of disease in a biopsy-proven breast carcinoma (useful if breast conservation is being considered), or to screen very high-risk women with dense mammograms. If an occult lesion is discovered by MRI, localization of the lesion may be problematic, unless a MRI localization device is available.

Ultrasound Stavros et al have described ultrasound features of solid masses that suggest benign or malignant disease, such as sharp margins (benign) and taller-than-wide lesions (malignant). Although these features are useful for clinical decision-making, their utility in increasing the specificity of the breast lesion work-up has not been verified.

Sestamibi nuclear medicine scanning can help differentiate benign from malignant mammographic asymmetries, and may play a role in evaluating palpable masses. Due to its limited spatial resolution and scatter, this technique is not reliable for lesions 1 cm or smaller.

Prevention

LIFESTYLE CHANGES ASSOCIATED WITH BREAST CANCER RISK REDUCTION

There is increasing evidence that lifestyle changes may alter an individual’s breast cancer risk.

Physical activity has been associated with a reduction in breast cancer risk. The benefit was greatest in premenopausal women, as compared with postmenopausal women, and was larger in younger as opposed to older women. The activity can be related to leisure or work time activities.

It has been suggested that women who exercise 3½-4 times per week have a reduced incidence of breast cancer, as compared with women who do not exercise. The protective effect of exercise may be associated with a reduction in the frequency of ovulatory cycles and circulating estrogen and progesterone levels.

Alcohol consumption Numerous studies that have evaluated the effects of alcohol consumption on breast cancer risk and the results of a cohort study addressing this issue were recently published. When compared to nondrinkers, women who consumed 2.3-4.5 bottles of beer per day, 2.5-5.6 glasses of wine per day or 2-4 shots of liquor per day had a 41% higher risk of developing invasive breast cancer. Therefore, a reduction in alcohol consumption is likely to reduce breast cancer risk.

The biological basis for the association between alcohol consumption and an increased risk of breast cancer is unclear. It has been proposed that there is a positive correlation between alcohol and estrogen levels.

Alterations in diet and tobacco use A reduced incidence of breast cancer has been observed in countries where the populations’ diet is typically low in fat. However, no reduction in breast cancer risk has been observed in the United States when women followed low fat diets. An association between red meat consumption or tobacco use and breast cancer risk has not been demonstrated.

Lactation Although it has been suggested that lactation may protect against breast cancer, it is unclear whether lactation reduces breast cancer risk. A recent study failed to demonstrate any breast cancer risk reduction in women who breast-fed and showed no dose-response effect in women who breast-fed for longer time periods.

CHEMOPREVENTION

Breast Cancer Prevention Trial

The National Institutes of Health (NIH) and NCI have publicized the results of the National Surgical Adjuvant Breast and Bowel Project (NSABP) Breast Cancer Prevention Trial (BCPT). Women who had a risk of developing breast cancer equivalent to that of women 60 years of age qualified as participants in this double-blind, randomized trial. (For representative eligibility profiles, see Table 5.) A total of 13,388 women were randomized to tamoxifen (Nolvadex) or placebo.

Benefits of therapy The summary results indicate that tamoxifen prevented about half of both invasive and noninvasive breast cancers in all age groups (Table 6). In addition to this reduction in invasive and noninvasive breast cancer, a secondary benefit of tamoxifen appeared to be a reduction in the incidence of hip fracture (Table 7).

At present, no survival advantage has been shown for participants in this trial.

Side effects Tamoxifen-treated women under age 50 had no apparent increase in side effects. However, women over age 50 experienced serious side effects, including vascular events and endometrial cancer. Particularly worrisome was the increased incidence of endometrial cancer in the tamoxifen-treated patients (Table 8). In addition, a significant increase in pulmonary embolism and deep vein thrombosis was noted, especially in women over age 50 (Table 9).

Current recommendations

Based on results of the BCPT, the FDA recently approved tamoxifen for use in women at high risk of breast cancer.

The NCI and NSABP are in the process of developing risk profiles based on age, number of affected first-degree relatives with breast cancer, number of prior breast biopsies, presence or absence of atypical hyperplasia or LCIS, age at menarche, and age at first live birth. These risk profiles may help guide women in making the decision of whether or not to take tamoxifen.

An ASCO working group recently published an assessment of tamoxifen use in the setting of breast cancer risk reduction. All women older than 35 years with a Gail model risk of > 1.66% (or the risk equivalent to that of women 60 years of age) should be considered candidates for this treatment strategy. Comorbid conditions, such as a history of deep venous thrombosis, must be a part of the consent process and treatment decision.

Staging and prognosis

Staging system The most widely used system to stage breast cancer is the American Joint Committee on Cancer (AJCC) classification, which is based on tumor size, the status of regional lymph nodes, and the presence of distant metastasis (Table 10).

Clinical staging is done initially and is determined after the physical examination and appropriate radiologic studies have been performed.

Pathologic staging Pathologic stage is determined following surgery for operable breast cancer. Pathologic tumor size may differ from clinical tumor size. In addition, axillary nodal metastases that were not clinically evident may be detected after pathologic examination.

Prognostic factors Numerous prognostic factors for breast cancer have been identified.

Lymph node status Axillary nodal metastases are the most important prognostic factor. Axillary node involvement and survival were evaluated in patients with breast cancer. Survival was examined relative to the number of nodes involved and the location of nodes that contained metastatic deposits. For any given number of positive nodes, survival was independent of the level of involvement but was directly related to the number of involved nodes.

Overall, patients who are node-negative have a 10-year survival rate of 70% and a 5-year recurrence rate of 19%. As the number of positive nodes increases, so does the likelihood of relapse. Patients with > 10 positive lymph nodes have a recurrence rate of 72%-82%. The majority of patients who develop recurrence after initial curative treatment for early-stage breast cancer will have distant metastases.

Tumor size and hormone-receptor status also correlate with outcome.

Other factors that have been utilized to predict outcome are histologic grade, lymphovascular permeation, S-phase fraction, and ploidy.

More recently, molecular prognostic factors have been evaluated to determine their utility in predicting outcome. These include the growth factor receptors (epidermal growth factor receptor and c-erbB-2/neu), tumor-suppressor genes (p53), proteolytic enzymes that may be associated with invasion and metastasis (cathepsin D), and metastasis-suppressor genes (nm23).