| ext">Breast cancer is the most common form of | | | | years later, another large study compared film |
| cancer in women and the second leading cause of | | | | mammography with digital mammograms read |
| cancer deaths in American women. In 2009, | | | | with computer-aided detection software. The |
| approximately 194,280 patients are estimated to | | | | authors found that diagnostic specificity |
| be diagnosed with invasive breast cancer, and | | | | significantly decreases from 90.2% to 87.2% with |
| 62,280 with carcinoma in situ. An estimated | | | | computer-aided detection, whereas sensitivity |
| 40,610 will die of this disease. For a woman of | | | | does not change. The rate of biopsy increases by |
| average risk, the lifetime incidence of breast | | | | 19.7%. Thus, more expensive technology does |
| cancer is one in eight. | | | | not necessarily translate into better outcome. |
| Screening of asymptomatic women has been | | | | In women at high risk for developing breast |
| accredited for the decline in mortality of breast | | | | cancer, screening may also involve breast |
| cancer. The current recommendations from the | | | | ultrasound and/or MRI (magnetic resonance |
| American Cancer Society for normal-risk women | | | | imaging). High risk factors include BRCA gene |
| are as follows: 1) Yearly mammogram starting at | | | | carriers, personal or strong family history of |
| age 40. The age at which screening should be | | | | breast cancer, prior atypia such as ADH (atypical |
| stopped should be individualized by considering the | | | | ductal hyperplasia) or LCIS (lobular carcinoma in |
| potential risks and benefits of screening in the | | | | situ), and prior chest irradiation. Sonography may |
| context of overall health status and longevity; and | | | | be useful in dense breasts. Screening ultrasound |
| 2) Clinical breast exam every 3 years for women | | | | can lead to biopsy in 2%-4% of women, of which |
| in their 20’s and 30’s, and every | | | | carcinoma was found in 10%-16% of these |
| year for women 40 and older. The evidence to | | | | biopsies. The ACR Imaging Network trial 6666 is |
| justify mammography for population-based | | | | evaluating screening sonography in high-risk |
| screening is derived from both randomized and | | | | women. Breast MRI has also recently been used |
| several non-randomized clinical trials. Eight | | | | by clinicians in many high-risk patients. In previous |
| randomized trials totalling hundreds of thousands | | | | reports, MRI resulted in biopsy in 7%-18% of |
| of patients include the Health Insurance Plan of | | | | women, of which breast cancer was detected in |
| New York, four studies from Sweden, one from | | | | 24%-88% of these biopsies. We do not know |
| the United Kingdom, and two from Canada. The | | | | yet whether ultrasound or MRI decreases breast |
| US BCDDP (Breast Cancer Detection | | | | cancer mortality in high-risk population, beyond |
| Demonstration Project), the largest study of | | | | that achieved by screening mammography. |
| mammography and clinical breast exam, also | | | | Currently, breast MRI is also indicated in women |
| demonstrated that screening decreases breast | | | | with newly diagnosed unilateral breast cancer. A |
| cancer mortality. | | | | recent publication showed that MRI can detect |
| During the past decade, advances in | | | | cancer in the contralateral breast that is missed |
| mammography include digital techniques and | | | | by mammography and clinical examination in 3.1% |
| computer-aided detection. Film (non-digital) | | | | of cases. The sensitivity of contralateral breast |
| mammography has been estimated to have | | | | MRI is 91%, the specificity 88%, and the negative |
| approximately 65-80% sensitivity at the desired | | | | predictive value 99%. |
| specificity of 90%. The investigators from the | | | | Despite the proven success of breast cancer |
| DMIST (Digital Mammographic Imaging Screening | | | | screening with mammography, many investigators |
| Trial) reported that the overall diagnostic accuracy | | | | in the US have noted a decline in its use. The |
| of digital and film mammography is similar. | | | | NHIS (National Health Interview Survey) estimates |
| However, digital mammograms are more | | | | from 2005 demonstrated a decline in |
| accurate in women under the age of 50 years, | | | | mammogram screening to 66%, from 70% in |
| pre-menopausal or peri-menopausal women, and | | | | 2000. This may unfortunately explain the reported |
| those with radiographically dense breasts. Two | | | | decrease in the incidence of breast cancer, i.e. |