Early detection of breast cancer is vital to reducing the morbidity and mortality associated with this disease. After a brief overview of breast cancer epidemiology and risk assessment, this article describes screening and diagnostic imaging techniques as they are currently practiced to promote early breast cancer detection. We conclude with a review of image-guided needle biopsy techniques and a recommended approach to breast cancer screening in the general population.
EPIDEMIOLOGY OF BREAST CANCER: DAUNTING BUT SLOWLY IMPROVING
After nonmelanoma skin cancers, breast cancer is the most common form of cancer in women today, accounting for more than 1 in 4 cancers diagnosed in US women.1 If the current incidence of breast cancer remains constant, US females born today have an average risk of 12.7% of being diagnosed with breast cancer during their lifetime (ie, 1-in-8 lifetime risk), based on National Cancer Institute statistics.2,3 The American Cancer Society estimated that 178,480 new cases of invasive breast cancer and 62,030 new cases of in situ breast cancer would be diagnosed in the United States in 2007, and that 40,460 US women would die from breast cancer that year.1 Only lung cancer accounts for more cancer deaths in women.
The role of race and ethnicity
Breast cancer risk varies by race and ethnicity in the United States. After age 40 years, white women have a higher incidence of breast cancer compared with African American women; conversely, before age 40, African American women have a higher incidence compared with white women. African American women are more likely than their white counterparts to die from their breast cancer at any age. Incidence and death rates from breast cancer are lower among Asian American, American Indian, and Hispanic women compared with both white and African American women.1
Recent hopeful trends
Despite the daunting incidence numbers reviewed above, recent years have seen encouraging trends in US breast cancer epidemiology.
For invasive breast cancer, the growth in incidence rates slowed during the 1990s, and rates actually declined by 3.5% per year during the period 2001–2004.1 These changes are likely attributable to multiple factors, including variations in rates of mammography screening and decreased use of hormone replacement therapy after the 2002 publication of results from the Women’s Health Initiative trial. Still, the trend is encouraging.
Incidence rates of in situ breast cancer rose rapidly during the 1980s and 1990s, largely due to increased diagnosis by mammography, but have plateaued since 2000 among women aged 50 years or older while continuing to rise modestly in younger women.1
Meanwhile, the overall death rate from breast cancer in women declined by 2.2% annually from 1990 to 2004.1
RISK FACTORS AND RISK MODELING
Risk factors for breast cancer have been well described and include the following:
- Age ( ≥ 65 years vs < 65 years, although risk increases across all ages up to 80 years)
- Family history of breast cancer
- Late age at first full-term pregnancy (> 30 years)
- Never having a full-term pregnancy
- Early menarche and/or late menopause
- Certain genetic mutations for breast cancer (eg, in the BRCA1, BRCA2, ATM, and CHEK2 genes)
- Certain breast disorders, such as atypical hyperplasia or lobular carcinoma in situ
- High breast tissue density
- High bone density (postmenopausal)
- High-dose radiation to the chest.
The above risk factors are, in general, fixed. More elusive risk factors, in that they are variable and modifiable, include obesity, use of exogenous hormones (recent and long-term hormone replacement therapy; recent oral contraceptive use), alcohol use, tobacco use, diet, and a low level of physical activity. Breast implants are not a risk factor for breast cancer, though their presence does obscure breast tissue on imaging, limiting the detectability of a tumor when it does develop (see “Screening the Surgically Altered Breast” below).
Women with a genetic predisposition to breast cancer merit special consideration. Hereditary breast cancers account for about 5% to 10% of breast cancer cases, and the BRCA1 and BRCA2 mutations are responsible for 80% to 90% of these cases, while other gene mutations (noted above) or genetic syndromes account for the rest. Clinical options for managing women with a genetic predisposition include surveillance, chemoprevention, and prophylactic surgery.4 Detailed discussion of the management of these women is beyond the scope of this article, but readers are referred to www.nccn.org/professionals/physician_gls/PDF/ genetics_screening.pdf for practice guidelines from the National Comprehensive Cancer Network.5
Tools for risk assessment
Several tools are available to predict a woman’s risk of developing breast cancer. Probably the most widely used is the Gail model,6 which was published in 1989 and forms the statistical basis for the National Cancer Institute’s Breast Cancer Risk Assessment Tool, which is available for downloading at www.cancer.gov/bcrisktool.7 The model uses a woman’s personal medical and reproductive histories and her family history of breast cancer to predict her 5-year and lifetime risk of developing invasive breast cancer. Factors included in the risk calculation are age, race, number of first-degree relatives with a history of breast cancer, age at first live birth (or nulliparity), age at menarche, number of breast biopsies, and presence or absence of a history of atypical hyperplasia. The relative risk for each of these factors is multiplied to generate a composite risk. The Gail model has been validated for white women but has been shown to underestimate breast cancer risk in African American women; it remains to be validated for Hispanic women, Asian women, and other subgroups of women.7
The commonly taught “triple test” for palpable breast lesions is another risk model that incorporates clinical findings. It consists of a physical examination, mammography, and fine-needle aspiration8 (in the “modified triple test,” ultrasonography replaces mammography9). When all three elements of the test are concordant (either all benign or all malignant), the triple test has been reported to have 100% diagnostic accuracy.8,9