Breast Density and Recurrent Disease

Abstract & Commentary

Synopsis: Breast tissue density, as determined by mammography, is known to be a risk factor for the development of invasive breast cancer. In the current analysis of a subgroup of patients participating in the NSABP B-17 trial, those women with highly dense breast tissue were found to have significantly more subsequent breast cancer. Thus, a quantitative assessment of breast density may prove useful in assessment of additional breast cancer risk for patients having breast-conserving surgery for DCIS.

Source: Habel LA, et al. J Natl Cancer Inst. 2004;96: 1467-1472.

Among risk factors, mammographic density, or the extent of the breast occupied by radiologically-dense tissue remains a powerful predictor of subsequent disease. The appearance of breast tissue on mammogram is determined by the relative components of fat, connective tissue and epithelial tissue. Connective and epithelial tissue are more highly dense, whereas fat is translucent by this technique. Women with extensive areas of mammographic density have been found to be at increased risk of developing breast cancer.1

Women diagnosed with ductal carcinoma in situ (DCIS) and treated with breast conserving surgery are at an increased risk of developing invasive breast cancer in the ipsilateral breast, and to a lesser extent in the contralateral breast as well.2-3

Although numerous studies have examined the association between mammographic density and risk of primary breast cancer, this feature has not been previously investigated with regard to the prediction of subsequent breast cancer after either invasive breast cancer or DCIS. Habel and colleagues examined the data from 504 women participants of the National Surgical Adjuvant Breast and Bowel Project B-17. In this subgroup of patients approximately 6.6% had breasts categorized as highly dense (ie, > 75% of the breast occupied by dense tissue). After adjusting for treatment with radiotherapy, age, and body mass index, women with highly dense breasts had 2.8 (95% confidence interval [CI] = 1.3-6.1) times the risk of subsequent breast cancer (DCIS or invasive), 3.2 (95% CI = 1.2-8.5) times the risk of invasive breast cancer, and 3.0 (95% CI = 1.2-7.5) times the risk of any ipsilateral breast cancer, compared with women with less than 25% of the breast occupied by dense tissue.

These findings suggest that the risk of second breast cancers may be increased among DCIS patients with highly dense breasts. An assessment of mammographic density, and a reporting of such findings in quantitative terms may prove to be useful in assessing patient risk for subsequent invasive cancer and in investigating prevention strategies.

Comment by William B. Ershler, MD

Dense breast tissue is relatively more glandular (less fat) and therefore more likely to harbor malignant disease. Women with DCIS treated with breast conserving surgery are at substantially increased risk of subsequent cancer in the involved (ipsilateral) breast, and to a lesser extent, in the unaffected (contralateral) breast.2,3 This may be because of residual microscopic disease after surgery. To the extent that dense breast tissue is a reflection of a hormonal environment supportive of active breast proliferation, recurrence rates might be expected to be higher with increasing breast density. Furthermore, because mammograms are more difficult to assess in women with dense breasts, the extent of a DCIS tumor may be more difficult to determine allowing for the chance for incomplete excision. Furthermore, it is possible that disease surveillance (eg, mammography, clinical breast examination) after treatment for the primary DCIS tumor is less accurate among women with highly dense breasts.

In the current series, most of the patients were 50 years of age or older and approximately 6.6% had breasts categorized as highly dense. As expected, percent density decreased with increasing age and with body mass index (BMI). Density was not significantly associated with treatment-related factors (ie, radiotherapy, status of surgical margins). Of the 504 eligible women on this trial, mammograms of sufficient quality were available on 392, and these patients served as the analysis cohort. Of these, 130 had a subsequent breast cancer event; 91 cases in the ipsilateral breast, 28 in the contralateral breast and 11 with regional (nodal) or distant metastases. As was true for the entire B-17 study population,4,5 risk of subsequent ipsilateral breast cancer was independently associated with radiotherapy, comedo necrosis, and appearance of micro calcifications. In addition, risk of subsequent breast cancer was independently associated with age, menopausal status and BMI. Although qualitative measures of breast density was not associated with increased risk, quantitative assessment as performed in this cohort, was highly predictive of subsequent breast cancer event. Of note, however, the risk of subsequent breast cancer was not markedly different for women in the lower three categories of percent density and there was no statistical evidence of a linear increasing trend across ordinal categories of density. Of note, relative risks for developing breast cancer remained even when the follow-up period was initiated 2 or more years after DCIS diagnosis.

Thus, there is now reasonable evidence that the risk of subsequent breast cancer is increased among DCIS patients with highly dense breasts and that this association did not appear to be limited to certain time periods after DCIS diagnosis. Mammographic density assessment at diagnosis may aid in risk assessment for women with DCIS, although larger studies will be needed to confirm this finding. Furthermore, a standardized method of assessing and reporting mammography determined breast density will need to be developed before clinicians will be able to incorporate this finding into the prescription of a prevention strategy for individual patients.


1. Boyd NF, et al. Cancer Epidemiol Biomarkers Prev. 1998;7:1133-1144.

2. Habel LA, et al. Ann Surg. 19978;225:69-75.

3. Frykberg ER, et al. Surg Gynecol Obstet. 1993;177:425-440.

4. Fisher ER, et al. Cancer. 1999;86:429-438.

5. Fisher B, et al. J Clin Oncol. 1998;16:441-452.

William B. Ershler, MD, INOVA Fairfax Hospital Cancer Center, Fairfax, VA, Director, Institute for Advanced Studies in Aging, Washington, DC, is Editor of Clinical Oncology Alert.