When a woman presents with a breast complaint, initial management nearly always includes a clinical breast examination and an imaging study. Frequently, risk factors for breast cancer are assessed. Individualized risk predictions employing more formalized tools, such as the Gail model1 or the Claus model,2 are being used increasingly in screening populations. However, an important question remains: in symptomatic women, are risk factors for breast cancer still clinically important? Unfortunately, the answer is unclear.
The Steering Committee on Clinical Practice Guidelines for the Care and Treatment of Breast Cancer3 states that when a woman presents with a breast lump or a suspicious change in breast texture, her risk factors for breast cancer should be noted, but the presence or absence of risk factors should not influence decisions about further work-up. Similarly, recommendations for evaluation and follow-up of mammographic abnormalities generally are made without regard to individual breast cancer risk. However, improved use of breast cancer risk factors has the potential to reduce the number of biopsies performed in women who do not have cancer and to increase the percentage of positive biopsies.
In response to a topic nomination by Kaiser Permanente Northern California, the Agency for Healthcare Research and Quality (AHRQ) funded a systematic review of the literature.4 The objective was to assess published evidence on the relationship between risk factors, breast abnormalities (clinical symptoms or mammographic findings), and breast cancer, and to provide practical recommendations for applying this information.
The systematic review4 found that although many studies reported breast cancer incidence in association with risk factors (menstrual status, hormone therapy, pregnancy history, family history, age) or abnormal breast findings, relatively few studies reported the incidence in association with both. In addition, the literature suffers from a lack of standardization of terms for reporting information about breast disease. Hence, reported results vary, depending on whether breast cancer incidence is derived from the number of lesions or the number of affected patients.
The literature on mammography also is problematic. Although mammographic results almost always are given, variations in reporting formats make it impossible to combine data in a useful way. The Breast Imaging Reporting and Data System (BI-RADS) terminology was developed for the purpose of standardizing mammogram reports.5,6 Widespread use of the BI-RADS nomenclature (e.g., in studies that relate cancer incidence by age to BI-RADS scores) could make data integration possible.7–11
Thus, although risk factors for breast cancer are well established and commonly used to direct evaluation in other clinical scenarios, current evidence does not permit assessment of the impact of individual risk factors on the likelihood that a breast abnormality represents cancer. Family history,12–19 pregnancy and menstrual history,13,14 and hormone therapy20 lacked a consistent evidence base for inferring any conclusions about the risk of cancer when these factors were associated with a clinical or mammographic abnormality. The only exception is patient age. In this instance, studies show that age over 50 years greatly increases the risk of breast cancer in women with a clinical or mammographic abnormality.
At this time, no published evidence supports modifying the work-up of breast symptoms or mammographic abnormalities based on risk factors other than age.