The number of women with a personal history of breast cancer is increasing, and these women have a long-term risk of developing a new or recurrent cancer in the conserved breast or a new cancer in the contralateral breast. Consensus-based recommendations for surveillance of breast cancer survivors include annual mammography; however, high-quality evidence is lacking on outcomes of mammography screening in these women. Evidence of screening mammography’s benefits and harms in women who have had breast cancer comes from nonrandomized studies and from extrapolating benefit (reduced breast cancer mortality) from population-based mammography screening trials. It is important to minimize disease burden by detecting second breast cancers early, when they are most easily treated.
We recently examined the accuracy and outcomes of mammography screening performed on women with a personal history of early-stage breast cancer (ductal carcinoma in situ or stage I or II invasive breast cancer) compared with mammography performed on similar women who had no previous breast cancer. Using data on mammograms performed from 1996 to 2007 at Breast Cancer Surveillance Consortium–affiliated mammography facilities, we studied 58,870 screening mammograms in 19,078 women who had had breast cancer and matched their mammograms on breast density, age-group, mammography year, and registry to 58,870 screening mammograms in 55,315 women who had not.
We found that within 1 year following screening, 655 breast cancers occurred in women with a personal history of breast cancer, compared with 342 cancers in the other women—almost a doubling of the cancer rate in breast cancer survivors versus other women (10.5/1000 vs 5.8/1000 screens). Screening mammography detected 6.8 cancers/1000 screens in women who had a history of breast cancer compared with 4.4/1000 screens in women who had not. Most of the detected cancers were early-stage in both groups, but the rate of “missed” (interval) cancers was more than double in breast cancer survivors compared with other women (3.6/1000 vs 1.4/1000 screens).
The sensitivity of mammography screening (ie, proportion of breast cancers detected by mammography) was significantly lower in breast cancer survivors than in the other women, at 65.4% (95% confidence interval [CI], 61.5%- 69.0%) and 76.5% (95% CI, 71.7%-80.7%), respectively. This decreased sensitivity occurred despite more diagnostic workups performed on women with a personal history of breast cancer: 18.1% (95% CI, 17.6%-18.6%) of screening mammograms in women with a history had additional imaging performed on the same day or led to a recommendation for additional imaging, compared with only 8.3% (95% CI, 8.1%- 8.5%) of mammograms on women with no such history. And 2.2% (95% CI, 2.1%-2.3%) of mammograms performed on breast cancer survivors led to a recommendation for biopsy or surgical consultation, compared with only 1.4% (95% CI, 1.2%-1.5%) of mammograms on the other women.
The accuracy of mammography and the rates of second breast cancers varied by characteristics of the women and the treatment received for the first cancer. For example, mammography was less likely to detect a second breast cancer in women who were younger or had dense breast tissue—women who were also at increased risk of a second breast cancer— relative to women who were older or had fatty or scattered fibroglandular breast tissue. Mammography was less likely to detect second breast cancers in women who received adjuvant systemic therapy (chemotherapy or endocrine therapy), although these women were at lower risk of a second breast cancer than those who did not receive adjuvant therapy.
Women with a personal history of breast cancer should be encouraged to have annual screening mammography. Despite relatively modest screening sensitivity, we found that mammography detects second breast cancers early in previously affected women. Breast cancer survivors should also be advised to have breast symptoms (such as a lump or nipple discharge) promptly investigated, even if they had a recent mammogram, because they are at increased risk of an interval (missed) cancer. In addition, women who have had breast cancer should be warned that they will likely need to undergo additional imaging (including diagnostic mammography, ultrasound, or breast magnetic resonance imaging) and possibly even a breast biopsy to rule out a second cancer. Most of these exams will not result in a cancer diagnosis, so they should not be anxious if they are called back for diagnostic workup following their screening mammogram.
Houssami N, Abraham LA, Miglioretti DL, et al. Accuracy and outcomes of screening mammography in women with a personal history of early-stage breast cancer. JAMA. 2011;305:790-799.