Screening for Breast Cancer

Breast cancer ranks as the most common cancer among women, accounting for more than one in four cancer diagnoses, 250,000 new cases, and 40,000 deaths annually in the United States. The lifetime probability that an American woman will develop breast cancer is approximately 12% (one in eight). The incidence of male breast cancer is substantially lower (˜11 new cases per million annually in the United States).

The prevalence, morbidity, and mortality associated with breast cancer make it a topic of intense public and personal interest. As the majority of breast cancers in American women are detected by screening tests, an understanding of the advantages and the limitations of these screening methods is crucial. Breast cancer is one of only a few conditions for which benefits of screening have been well-documented in randomized, controlled trials. Advances in molecular biology, including identification of breast cancer susceptibility genes, have improved our ability to estimate the risk and predict the behavior of breast cancer for some women. However, the clinical introduction of these techniques poses new challenges as chemoprevention and prophylactic mastectomy become considerations. Moreover, controversy has emerged about which women should be screened and how often, pitting recommendations by advocacy and specialty groups against those of the U.S. Preventive Services Task Force (USPSTF).

Primary care professionals deal with breast cancer screening, diagnosis, and associated fears on a daily basis. Sometimescontradictory recommendations and broad-based epidemiologic studies do not easily inform an individual woman’s screening decision. One must understand the benefits and harms of breast cancer screening and prevention in order to communicate these effectively and help each woman make an informed decision on screening that makes best sense for her.

EPIDEMIOLOGY AND RISK FACTORS (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28 and 29)

The epidemiology of breast cancer has been studied extensively. Furthermore, analysis of familial patterns and identification of breast cancer susceptibility genes have refined our understanding of risk associated with age and family history.

Age

Risk of breast cancer increases with age. The median age at the time of diagnosis is about 55 years; 45% of cases occur after age 65 years. However, breast cancer is not uncommon in women younger than 40 years, in whom approximately 20% of cases occur. When breast cancer does occur earlier in life, it is more likely to be associated with a susceptibility gene. For example, in women diagnosed with breast cancer before age 40 years, about 5% of non-Jewish women and about 25% of Jewish women can be expected to have BRCA1 mutations.

Family History

Overall, a family history of breast cancer, whether in maternal or paternal relatives, increases the risk approximately two- to threefold. For women who have a family history, specific risk depends heavily on the age at onset of cancer for combinations of first-degree and second-degree relatives. For example, if one first-degree or second-degree relative had cancer diagnosed at or after age 50 years, the cumulative breast cancer risk by age 80 years is approximately 10%, not significantly different from risk for the general population. If a first-degree relative had cancer before age 50 years, cumulative risk may be as high as 20%. With a history of early-onset breast cancer in two firstdegree relatives, the probability that the family is affected by a dominant breast cancer susceptibility gene is substantial, and cumulative cancer risk by age 80 years approaches 50%.

Susceptibility genes account for about 5% to 10% of all breast cancers. In addition to family clustering and early age at onset of disease, family history of bilateral or multifocal breast cancer and ovarian cancer suggests inherited cancer predisposition. BRCA1, which accounts for about half of all inherited breast cancers, confers a cumulative breast cancer risk of approximately 85% and an ovarian cancer risk of 50%. Women with BRCA2 mutations face approximately the same risk of breast cancer but a significantly lower (10% to 20%) risk of ovarian cancer. Many members of families affected by breast cancer susceptibility genes have indicated a preference not to be tested. In addition, the response to learning about susceptibility is varied. Some women choose to have prophylactic surgery (bilateral mastectomies and/or oophorectomies); others do not. Bilateral mastectomy reduces risk of breast cancer by about 90% in such persons.

Because of the variability in what women want to know about their risk and what they might choose to do with that knowledge, recommendations for women with a significant family history of breast cancer generally focus on the need for thorough genetic counseling before BRCA testing is performed.

Reproductive History

Generally, there is an inverse relationship between breast cancer risk and parity, but maternal age at the time of first full-term pregnancy may be the most important factor related to reproductive history. In a woman with high parity whose first birth occurs before the age of 20 years, the risk of breast cancer is one half of that of a nulliparous woman and one third of that of a woman with one or two births after the age of 30 years. It has been suggested that childbirth transiently increases the risk of breast
cancer but reduces the risk in later years. Either spontaneous or induced abortion may increase risk slightly, but the association is weak and could be due to reporting bias. Lactation appears to reduce slightly the risk of premenopausal cancer.

Menstrual History

Both late menarche and early natural menopause reduce breast cancer risk. Women who experienced menarche after age 16 years have half the risk of those who experienced it earlier. Women in whom menopause occurred before age 45 years have half the risk of those in whom menopause occurred after age 55 years. Women with early surgical menopause seem to be similarly protected.

History of Benign Breast Disease

The relative risk of breast cancer among women who have a history of benign breast disease compared to those who do not is roughly 1.5. However, there are complex interactions among a woman’s age, family history, and the histology of the benign breast disease. Histologic appearance is strongly associated with risk. In one study, the relative risks for nonproliferative changes, proliferative changes, and atypia were 1.27, 1.66, and 4.24, respectively. For women with nonproliferative changes and no family history, there was no increased risk; this subset accounted for more than half of all women with benign disease that prompted biopsies. However, among women with a strong family history, even nonproliferative changes were associated with a relative risk of 1.62. Age is also an important factor; for example, among women younger than 45 years and older than 55 years, the relative risks conferred by a finding of atypia were 6.99 and 3.37, respectively.

Breast Density

Breast density is also a risk factor for developing breast cancer and an impediment to screening. Increased breast density is associated with younger age, family history of dense breast tissue, and current use of hormone replacement therapy. Women who have dense tissue in 75% or more of the volume of the breast have been reported to face a risk of cancer four to six times greater than women with little or no dense tissue. However, women with dense breast tissue who develop breast cancer do not face an increased risk of dying compared to women with breast cancer who do not have dense breasts.

As fat is radiolucent and stroma and epithelium are radiographically dense, breast density reduces the sensitivity of screening mammography. In recent years, several states in the United States have passed legislation requiring radiologists to communicate to patients with dense breast tissue on mammography that they have an increased risk of breast cancer and that mammographic test sensitivity is reduced. Such legislation has been controversial, since the finding of dense breast tissue is quite common, and the best screening for breast cancer after a finding of increased breast density is uncertain (see later discussion). Short-term interruption of hormone therapy prior to mammogram does lead to a small reduction in breast density but does not reduce mammogram recall rates.

History of Previous Malignancy or Cancer Treatment

Approximately 10% of women who survive 10 years after the diagnosis of breast cancer will have a second primary malignancy, usually in the contralateral breast. Increased use of breast-sparing surgery for early-stage breast cancer (even if complemented by radiation) increases the incidence—over 10 years, about 10% to 20% of women choosing such treatment experience an ipsilateral in-breast recurrence.

Presence or treatment of other cancers also increases risk of breast cancer. Women with a history of endometrial carcinoma have slightly increased risk for breast cancer. Those who have undergone chest wall irradiation between ages 10 and 30 years (most commonly for Hodgkin lymphoma) have a substantially increased risk of breast cancer and are advised to have screening MRI for breast cancer per the guidelines of the National Comprehensive Cancer Network.

Diet, Drugs, and Other Factors

Diet, Alcohol, and Tobacco

Breast cancer incidence varies fivefold among different countries and is positively associated with national dietary intake of fat. Diets high in fat have produced increased rates of mammary tumors in laboratory animals. Data from case-control studies have suggested a positive association, but prospective studies, including a recent overview of studies involving more than 300,000 women and nearly 5,000 cases, have found no evidence of a positive association between dietary fat and breast cancer. The Women’s Health Initiative Dietary Modification Trial among postmenopausal women failed to demonstrate a statistically significant reduction in breast cancer among women randomized to a low-fat diet, but there was a trend in that direction, with a relative risk for the intervention group of 0.91 (95% confidence interval 0.83 to 1.01). Low levels of vitamin A have also been linked to increases in risk. Obesity has been associated with minor increases in risk.

Several studies have demonstrated an association between daily moderate alcohol consumption and modest increase in breast cancer risk (e.g., relative risk 1.15 for one drink/day in the Nurses’ Health Study). Tobacco smoking initiated at an early age and for prolonged periods has been associated with moderate increase in breast cancer risk.

Drugs

The role of drugs, especially exogenous hormone use, has been the subject of considerable study and debate. Evidence for increased risk after oral contraceptive use is equivocal; risk may be slightly increased (relative risks in the range of 1.1 to 1.2) for younger women and for long-time users. Breast cancer risk from postmenopausal estrogen and combined estrogen-progestin hormone replacement therapy appears to be of greater concern. The 16-year cohort analysis of the Nurses’ Health Study found increased relative risk among women currently using hormones (1.32 for estrogen alone and 1.41 for estrogen plus progestin) as compared with postmenopausal women who had never used hormones. The Women’s Health Initiative randomized trial was stopped early after a mean of 5.2 years of follow-up because of a breast cancer hazard ratio of 1.26. After the report of this finding to the public in mid-2002, there was a sharp decline in the use of hormone replacement therapy, followed by a noticeable decline in breast cancer incidence a year later.

Use of aspirin and other nonsteroidal anti-inflammatory drugs (NSAIDs) has been associated with a 20% to 40% relative risk reduction in breast cancer; however, in recent reports from the Nurses’ Health Study, this association has not held out. Prostaglandin E2 has been shown to stimulate estrogen and progesterone in laboratory experiments. A selective reduction in risk of hormone receptor-positive tumors supports the hypothesis that this is mediated by inhibition of prostaglandin synthesis. A study suggesting that women who had undergone breast augmentation had lower cancer risk received substantial media attention. However, the analysis was found to be faulty, and a subsequent analysis showed no difference in risk after breast implants.

Physical Activity and Sleep Pattern

Moderate physical activity appears to be protective against breast cancer, but risk is increased in persons who perform night-shift work, hypothesized to suppress nocturnal melatonin production.

NATURAL HISTORY AND EFFECTIVENESS OF TREATMENT (30, 31, 32, 33, 34, 35, 36 and 37)

Natural History

Little can be said with certainty about the natural history of breast cancer. The few observational studies of untreated invasive breast cancer have shown widely variable tumor doubling times ranging from less than 1 week to more than 6 months.

Younger women tend to have faster-growing tumors than older women. The average time for development of an invasive tumor that can be detected by mammography for women aged 35 to 49 years is 15 months. The average time that a tumor is detectable by mammography before a clinical diagnosis is made is believed to be approximately 3 years. Again, however, progression is likely to be faster among younger women.

Ductal Carcinoma In Situ

There is even greater uncertainty about the natural history of noninvasive breast cancer. Dramatic increases in the incidence of ductal carcinoma in situ (DCIS) in recent years are attributable to mammographic screening. DCIS incidence rates increased approximately 4% annually during the decade ending in 1983 and approximately 18% annually for the subsequent decade. More than 25,000 new cases of DCIS are now diagnosed among US women each year. Many of these lesions appear to be slow growing; some undetected lesions likely regress without specific intervention, and evidence suggests that at least 50% of untreated women with DCIS would not develop invasive disease. DCIS does not affect survival during the first 1 to 9 years after diagnosis and treatment.

Treatment of Early-Stage Disease (see also Chapter 122)

Therapy for early-stage breast cancer is highly effective. Women with small invasive cancers (e.g., 1 cm) and negative axillary nodes can expect a disease-free survival of 90% to 95% during the 5 to 10 years after either mastectomy or breast-conserving surgery followed by radiation. Cumulative recurrence risk rises sharply with increase in tumor size and/or metastasis to axillary lymph nodes. Breast cancer screening has dramatically increased the proportion of breast cancers that are in earlier stages of development at the time of diagnosis.

Chemoprevention (see also Chapter 122)

Chemoprevention of breast cancer with selective estrogen receptor modulators (SERMs) or estrogenic activity regulators is an important consideration for women at high risk. Multiple randomized trials have addressed the impact of SERMs, specifically tamoxifen and raloxifene, and more recently the selective tissue estrogenic activity regulator tibolone when taken for primary prevention of breast cancer. Although not uniformly positive, most studies demonstrate reductions in relative risk in the range of 50% to 75%; the risks of thrombosis and stroke are also significantly increased, especially when used in older women. Use of tamoxifen also increases risks of endometrial cancer and cataracts. The positive effects have been limited to prevention of estrogen receptor-positive cancers. Based on this evidence, the USPSTF recommends against the routine use of chemoprevention for primary prevention in women not at increased risk of breast cancer and its application discussed with those who are.

Though supported only by observational and epidemiologic data, use of aspirin and other NSAIDs can be discussed as well. Epidemiologic studies indicate a relative risk reduction of 20% to 40%, with the effect evident for hormone receptor-positive tumors but not hormone receptor-negative tumors.