Secular stability and reliability of measurements of the percentage of dense tissue on mammograms

Determinants of the reliability of ultrasound tomography sound speed estimates as a surrogate for volumetric breast density

PURPOSE

High breast density, as measured by mammography, is associated with increased breast cancer risk, but standard methods of assessment have limitations including 2D representation of breast tissue, distortion due to breast compression, and use of ionizing radiation. Ultrasound tomography (UST) is a novel imaging method that averts these limitations and uses sound speed measures rather than x-ray imaging to estimate breast density. The authors evaluated the reproducibility of measures of speed of sound and changes in this parameter using UST.

METHODS

One experienced and five newly trained raters measured sound speed in serial UST scans for 22 women (two scans per person) to assess inter-rater reliability. Intrarater reliability was assessed for four raters. A random effects model was used to calculate the percent variation in sound speed and change in sound speed attributable to subject, scan, rater, and repeat reads. The authors estimated the intraclass correlation coefficients (ICCs) for these measures based on data from the authors' experienced rater.

RESULTS

Median (range) time between baseline and follow-up UST scans was five (1-13) months. Contributions of factors to sound speed variance were differences between subjects (86.0%), baseline versus follow-up scans (7.5%), inter-rater evaluations (1.1%), and intrarater reproducibility (∼0%). When evaluating change in sound speed between scans, 2.7% and ∼0% of variation were attributed to inter- and intrarater variation, respectively. For the experienced rater's repeat reads, agreement for sound speed was excellent (ICC = 93.4%) and for change in sound speed substantial (ICC = 70.4%), indicating very good reproducibility of these measures.

CONCLUSIONS

UST provided highly reproducible sound speed measurements, which reflect breast density, suggesting that UST has utility in sensitively assessing change in density.

Mammographic density and risk of breast cancer by age and tumor characteristics

Introduction

Understanding whether mammographic density (MD) is associated with all breast tumor subtypes and whether the strength of association varies by age is important for utilizing MD in risk models.

Methods

Data were pooled from six studies including 3414 women with breast cancer and 7199 without who underwent screening mammography. Percent MD was assessed from digitized film-screen mammograms using a computer-assisted threshold technique. We used polytomous logistic regression to calculate breast cancer odds according to tumor type, histopathological characteristics, and receptor (estrogen receptor (ER), progesterone receptor (PR), human epidermal growth factor receptor (HER2)) status by age (<55, 55–64, and ≥65 years).

Results

MD was positively associated with risk of invasive tumors across all ages, with a two-fold increased risk for high (>51%) versus average density (11-25%). Women ages <55 years with high MD had stronger increased risk of ductal carcinoma in situ (DCIS) compared to women ages 55–64 and ≥65 years (P_{age-interaction} = 0.02). Among all ages, MD had a stronger association with large (>2.1 cm) versus small tumors and positive versus negative lymph node status (P’s < 0.01). For women ages <55 years, there was a stronger association of MD with ER-negative breast cancer than ER-positive tumors compared to women ages 55–64 and ≥65 years (P_{age-interaction} = 0.04). MD was positively associated with both HER2-negative and HER2-positive tumors within each age group.

Conclusion

MD is strongly associated with all breast cancer subtypes, but particularly tumors of large size and positive lymph nodes across all ages, and ER-negative status among women ages <55 years, suggesting high MD may play an important role in tumor aggressiveness, especially in younger women.

Beyond breast cancer: mammographic features and mortality risk in a population of healthy women

BACKGROUND

Breast fibroglandular (dense) tissue is a risk factor for breast cancer. Beyond breast cancer, little is known regarding the prognostic significance of mammographic features.

METHODS

We evaluated relationships between nondense (fatty) breast area and dense area with all-cause mortality in 4,245 initially healthy women from the Breast Cancer Detection Demonstration Project; 1,361 died during a mean follow-up of 28.2 years. Dense area and total breast area were assessed using planimeter measurements from screening mammograms. Percent density reflects dense area relative to breast area and nondense area was calculated as the difference between total breast area and dense area. Hazard ratios (HRs) and 95% confidence intervals (CIs) were estimated by Cox proportional hazards regression.

RESULTS

In age-adjusted models, greater nondense and total breast area were associated with increased risk of death (HR 1.17, 95% CI 1.10-1.24 and HR 1.13, 95% CI 1.06-1.19, per SD difference) while greater dense area and percent density were associated with lower risk of death (HR 0.91, 95% CI 0.86-0.95 and HR 0.87, 95% CI 0.83-0.92, per SD difference). Associations were not attenuated with adjustment for race, education, mammogram type (x-ray or xerogram), smoking status, diabetes and heart disease. With additional adjustment for body mass index, associations were diminished for all features but remained statistically significant for dense area (HR 0.94, 95% CI 0.89-0.99, per SD difference) and percent density (HR 0.93, 95% CI 0.87-0.98, per SD difference).

CONCLUSIONS

These data indicate that dense area and percent density may relate to survival in healthy women and suggest the potential utility of mammograms beyond prediction of breast cancer risk.

Reproducibility of BI-RADS breast density measures among community radiologists: a prospective cohort study

Using data from the Vermont Breast Cancer Surveillance System (VBCSS), we studied the reproducibility of Breast Imaging Reporting and Data System (BI-RADS) breast density among community radiologists interpreting mammograms in a cohort of 11,755 postmenopausal women. Radiologists interpreting two or more film-screen screening or bilateral diagnostic mammograms for the same woman within a 3- to 24-month period during 1996-2006 were eligible. We observed moderate-to-substantial overall intra-rater agreement for use of BI-RADS breast density in clinical practice, with an overall intra-radiologist percent agreement of 77.2% (95% confidence interval (CI), 74.5-79.5%), an overall simple kappa of 0.58 (95% CI, 0.55-0.61), and an overall weighted kappa of 0.70 (95% CI, 0.68-0.73). Agreement exhibited by individual radiologists varied widely, with intra-radiologist percent agreement ranging from 62.1% to 87.4% and simple kappa ranging from 0.19 to 0.69 across individual radiologists. Our findings underscore the need for additional evaluation of the BI-RADS breast density categorization system in clinical practice.

Use of clinical history affects accuracy of interpretive performance of screening mammography

OBJECTIVE

To examine how use of clinical history affects radiologist's interpretation of screening mammography.

STUDY DESIGN AND SETTING

Using a self-administered survey and actual interpretive performance, we examined associations between use of clinical history and sensitivity, false-positive rate, recall rate, and positive predictive value, after adjusting for relevant covariates using conditional logistic regression.

RESULTS

Of the 216 radiologists surveyed (63.4%), most radiologists reported usually or always using clinical history when interpreting screening mammography. Compared with radiologists who rarely use clinical history, radiologists who usually or always use it had a higher false-positive rate with younger women (10.7 vs. 9.7), denser breast tissue (10.1 for heterogeneously dense to 10.9 for extremely dense vs. 8.9 for fatty tissue), or longer screening intervals (> prior 5 years) (12.5 vs. 10.5). Effect of current hormone therapy (HT) use on false-positive rate was weaker among radiologists who use clinical history compared with those who did not (P=0.01), resulting in fewer false-positive examinations and a nonsignificant lower sensitivity (79.2 vs. 85.2) among HT users.

CONCLUSION

Interpretive performance appears to be influenced by patient age, breast density, screening interval, and HT use. This influence does not always result in improved interpretive performance.

Women's features and inter-/intra-rater agreement on mammographic density assessment in full-field digital mammograms (DDM-SPAIN)

Measurement of mammographic density (MD), one of the leading risk factors for breast cancer, still relies on subjective assessment. However, the consistency of MD measurement in full-digital mammograms has yet to be evaluated. We studied inter- and intra-rater agreement with respect to estimation of breast density in full-digital mammograms, and tested whether any of the women's characteristics might have some influence on them. After an initial training period, three experienced radiologists estimated MD using Boyd scale in a left breast cranio-caudal mammogram of 1,431 women, recruited at three Spanish screening centres. A subgroup of 50 randomly selected images was read twice to estimate short-term intra-rater agreement. In addition, a reading of 1,428 of the images, performed 2 years before by one rater, was used to estimate long-term intra-rater agreement. Pair-wise weighted kappas with 95% bootstrap confidence intervals were calculated. Dichotomous variables were defined to identify mammograms in which any rater disagreed with other raters or with his/her own assessment, respectively. The association between disagreement and women's characteristics was tested using multivariate mixed logistic models, including centre as a random-effects term, and taking into account repeated measures when required. All quadratic-weighted kappa values for inter- and intra-rater agreement were excellent (higher than 0.80). None of the studied women's features, i.e. body mass index, brassiere size, menopause, nulliparity, lactation or current hormonal therapy, was associated with higher risk of inter- or intra-rater disagreement. However, raters differed significantly more in images that were classified in the higher-density MD categories, and disagreement in intra-rater assessment was also lower in low-density mammograms. The reliability of MD assessment in full-field digital mammograms is comparable to that for original or digitised images. The reassuring lack of association between subjects' MD-related characteristics and agreement suggests that bias from this source is unlikely.

Breast cancer risk assessment in women aged 70 and older

Although the benefit of screening mammography for women over 69 has not been established, it is generally agreed that screening recommendations for older women should be individualized based on health status and breast cancer risk. However, statistical models to assess breast cancer risk have not been previously evaluated in this age group. In this study, the original Gail model and three more recent models that include mammographic breast density as a risk factor were applied to a cohort of 19,779 Vermont women aged 70 and older. Women were followed for an average of 7.1 years and 821 developed breast cancer. The predictive accuracy of each risk model was measured by its c-statistic and associations between individual risk factors and breast cancer risk were assessed by Cox regression. C-statistics were 0.54 (95% CI = 0.52-0.56) for the Gail model, 0.54 (95% CI = 0.51-0.56) for the Tice modification of the Gail model, 0.55 (95% CI = 0.53-0.58) for a model developed by Barlow and 0.55 (95% CI = 0.53-0.58) for a Vermont model. These results indicate that the models are not useful for assessing risk in women aged 70 and older. Several risk factors in the models were not significantly associated with outcome in the cohort, while others were significantly related to outcome but had smaller relative risks than estimated by the models. Age-related attenuation of the effects of some risk factors makes the prediction of breast cancer in older women particularly difficult.

Physical activity and mammographic parenchymal patterns among Greek postmenopausal women

OBJECTIVE

To examine whether physical activity during the last five years is related to later breast mammographic density in postmenopausal Greek women.

METHODS

We designed a cross-sectional study in 724 women, of ages 45-67 years. An interview-administered questionnaire was used to obtain information on duration and intensity of recreational physical activity during five years preceding study recruitment. Mammograms were evaluated according to BIRADS classification and BIRADS score was also estimated. Multivariate ordinal logistic regression analysis was used to assess associations between physical activity index and breast density according to the BIRADS classification methods.

RESULTS

We observed a statistically significant inverse association of mammographic breast density measured by the BIRADS classification method and recreational exercise (OR=-0.10; 95% CI -0.018, -0.001; p=0.022). For one unit increase in physical activity as expressed by the MET-h/week score, the odds of lower versus higher breast density categories are 1.105 greater, given that all of the other variables in the model are held constant. A modifying effect by age at recruitment was evident among participants, with a stronger inverse association between recreational activity and mammographic breast density among older women (OR=-0.036; 95% CI -0.063, -0.009; p=0.009). An inverse association between physical activity and BIRADS score was evident, not reaching statistical significance (OR=0.00; 95% CI -0.009, 0.008; p=0.887).

CONCLUSIONS

Mammographic breast area was lower in postmenopausal women who participated in sports/recreational physical activity compared to inactive controls. Increasing physical activity levels among postmenopausal women might be a reasonable approach to reduce mammographic density. However, until more physical activity and mammographic breast density studies are conducted that confirm our findings, they have to be interpreted with caution, due to the retrospective nature of our data and the possibility of memory bias.

Longitudinal association of anthropometry with mammographic breast density in the Study of Women's Health Across the Nation

High percent mammographic breast density is strongly associated with increased breast cancer risk. Though body mass index (BMI) is positively associated with risk of postmenopausal breast cancer, BMI is negatively associated with percent breast density in cross-sectional studies. Few longitudinal studies have evaluated associations between BMI and weight and mammographic breast density. We studied the longitudinal relationships between anthropometry and breast density in a prospective cohort of 834 pre- and perimenopausal women enrolled in an ancillary study to the Study of Women's Health Across the Nation (SWAN). Routine screening mammograms were collected and read for breast density. Random intercept regression models were used to evaluate whether annual BMI change was associated with changes over time in dense breast area and percent density. The study population was 7.4% African-American, 48.8% Caucasian, 21.8% Chinese, and 21.9% Japanese. Mean follow-up was 4.8 years. Mean annual weight change was +0.32 kg/year, mean change in dense area was -0.77 cm(2)/year, and mean change in percent density was -1.14%/year. In fully adjusted models, annual change in BMI was not significantly associated with changes in dense breast area (-0.17 cm(2), 95% CI -0.64, 0.29). Borderline significant negative associations were observed between annual BMI change and annual percent density change, with percent density decreasing 0.36% (95% CI -0.74, 0.02) for a one unit increase in BMI over a year. This longitudinal study provides modest evidence that changes in BMI are not associated with changes in dense area, yet may be negatively associated with percent density.

Recreational physical activity and mammographic breast density characteristics

Increased mammographic breast density is considered an intermediate marker of breast cancer risk. Physical activity is believed to reduce breast cancer risk; however, its effect on breast density is not well understood. We studied the association between recreational physical activity and mammographic characteristics of the breast among a population of premenopausal and postmenopausal women enrolled as controls (n = 728) in a case-control study of mammographic breast density and breast cancer. Women were enrolled shortly after obtaining their regular screening mammograms, and participants reported their current and lifetime recreational physical activity history using a self-administered, reliable questionnaire at study enrollment. Linear regression was used to determine associations between physical activity variables and the dense breast area, non-dense area, total breast area, and percent density. Age-adjusted analyses revealed significant inverse associations between physical activity variables and the non-dense area and total area and positive associations with percent breast density. These associations were attenuated and nonsignificant after adjustment for body mass index (BMI). Adjustment for additional factors did not substantially change the results. Physical activity was not associated with the dense breast area before or after adjustment for BMI. Self-reported recreational physical activity was not significantly associated with the mammographic characteristics of the breast after adjustment for BMI in this population. These results suggest that the mechanism by which physical activity reduces breast cancer risk may not involve breast density.

Mammographic density in a multiethnic cohort

OBJECTIVES

To compare mammographic density among premenopausal and early perimenopausal women from four racial/ethnic groups and to examine density and acculturation among Japanese and Chinese women.

DESIGN

The study included 391 white, 60 African American, 171 Japanese, and 179 Chinese participants in the Study of Women's Health Across the Nation, a multisite study of US women transitioning through menopause. Mammograms done when women were premenopausal or early perimenopausal were assessed for area of dense breast tissue and the percent of the breast occupied by dense tissue (percent density). Information on race/ethnicity, acculturation, and other factors was obtained from standardized instruments. Multiple linear regression modeling was used to examine the association between race/ethnicity or acculturation and density measures.

RESULTS

Age-adjusted mean percent density was highest for Chinese (52%) and lowest for African American (34%) women. After additional adjustment for body mass index, menopause status, age at first birth, breast-feeding duration, waist circumference, and smoking, African Americans had the highest mean percent density (51%) and Japanese women had the lowest (39%). In contrast, the area of dense tissue was highest for African Americans and similar for white, Japanese, and Chinese women. Less acculturated Chinese and Japanese women tended to have a larger area of density and a higher percent density.

CONCLUSIONS

Neither the age-adjusted nor fully adjusted results for percent density or area of dense tissue reflected current differences in breast cancer incidence rates among similarly aged African American, Japanese, Chinese, and white women. In addition, mammographic density was higher in less acculturated Asian women.

Projecting absolute invasive breast cancer risk in white women with a model that includes mammographic density

BACKGROUND

To improve the discriminatory power of the Gail model for predicting absolute risk of invasive breast cancer, we previously developed a relative risk model that incorporated mammographic density (DENSITY) from data on white women in the Breast Cancer Detection Demonstration Project (BCDDP). That model also included the variables age at birth of first live child (AGEFLB), number of affected mother or sisters (NUMREL), number of previous benign breast biopsy examinations (NBIOPS), and weight (WEIGHT). In this study, we developed the corresponding model for absolute risk.

METHODS

We combined the relative risk model with data on the distribution of the variables AGEFLB, NUMREL, NBIOPS, and WEIGHT from the 2000 National Health Interview Survey, with data on the conditional distribution of DENSITY given other risk factors in BCDDP, with breast cancer incidence rates from the Surveillance, Epidemiology, and End Results program of the National Cancer Institute, and with national mortality rates. Confidence intervals (CIs) accounted for variability of estimates of relative risks and of risk factor distributions. We compared the absolute 5-year risk projections from the new model with those from the Gail model on 1744 white women.

RESULTS

Attributable risks of breast cancer associated with DENSITY, AGEFLB, NUMREL, NBIOPS, and WEIGHT were 0.779 (95% CI = 0.733 to 0.819) and 0.747 (95% CI = 0.702 to 0.788) for women younger than 50 years and 50 years or older, respectively. The model predicted higher risks than the Gail model for women with a high percentage of dense breast area. However, the average risk projections from the new model in various age groups were similar to those from the Gail model, suggesting that the new model is well calibrated.

CONCLUSIONS

This new model for absolute invasive breast cancer risk in white women promises modest improvements in discriminatory power compared with the Gail model but needs to be validated with independent data.

Prospective breast cancer risk prediction model for women undergoing screening mammography

BACKGROUND

Risk prediction models for breast cancer can be improved by the addition of recently identified risk factors, including breast density and use of hormone therapy. We used prospective risk information to predict a diagnosis of breast cancer in a cohort of 1 million women undergoing screening mammography.

METHODS

There were 2,392,998 eligible screening mammograms from women without previously diagnosed breast cancer who had had a prior mammogram in the preceding 5 years. Within 1 year of the screening mammogram, 11,638 women were diagnosed with breast cancer. Separate logistic regression risk models were constructed for premenopausal and postmenopausal examinations by use of a stringent (P<.0001) criterion for the inclusion of risk factors. Risk models were constructed with 75% of the data and validated with the remaining 25%. Concordance of the predicted with the observed outcomes was assessed by a concordance (c) statistic after logistic regression model fit. All statistical tests were two-sided.

RESULTS

Statistically significant risk factors for breast cancer diagnosis among premenopausal women included age, breast density, family history of breast cancer, and a prior breast procedure. For postmenopausal women, the statistically significant factors included age, breast density, race, ethnicity, family history of breast cancer, a prior breast procedure, body mass index, natural menopause, hormone therapy, and a prior false-positive mammogram. The model may identify high-risk women better than the Gail model, although predictive accuracy was only moderate. The c statistics were 0.631 (95% confidence interval [CI] = 0.618 to 0.644) for premenopausal women and 0.624 (95% CI = 0.619 to 0.630) for postmenopausal women.

CONCLUSION

Breast density is a strong additional risk factor for breast cancer, although it is unknown whether reduction in breast density would reduce risk. Our risk model may be able to identify women at high risk for breast cancer for preventive interventions or more intensive surveillance.

A Longitudinal Investigation of Mammographic Density: The Multiethnic Cohort

Mammographic densities are hypothesized to reflect the cumulative exposure to risk factors that influence breast cancer incidence. This report analyzed percent densities over time and explored predictors of density change in relation to age. The study population consisted of 607 breast cancer cases and 667 frequency matched controls with 1,956 and 1,619 mammographic readings, respectively. Mammograms done over >20 years and before a diagnosis of breast cancer were assessed for densities using a computer-assisted method. Using multilevel modeling to allow for repeated measurements, we estimated the effect of ethnicity, case status, reproductive characteristics, hormonal therapy, body mass index, and soy intake on initial status and longitudinal change. After integrating the area under the percent density curve, cumulative percent density was compared with age-specific breast cancer rates in Hawaii. Percent densities decreased approximately 5.6% per 10 years but a nonlinear effect indicated a faster decline earlier in life. Cumulative percent densities and age-specific breast cancer rates increased at very similar rates; both standardized regression coefficients were >0.9. Japanese ancestry, overweight, estrogen/progestin treatment, and, to a lesser degree, estrogen-only therapy predicted a slower decline in densities with age. Case status and adult soy intake were related to higher densities whereas overweight and having any child were associated with lower densities at initial status. Risk factors that influence the decline in mammographic densities over time may be important for breast cancer prevention because cumulative percent densities may reflect the age-related increase in breast cancer risk.

Breast cancer risk factors and mammographic breast density in women over age 70

BACKGROUND

Breast density is a strong risk factor for breast cancer, but little is known about factors associated with breast density in women over 70.

METHODS

Percent breast density, sex hormone levels and breast cancer risk factor data were obtained on 239 women ages 70-92 recruited from 1986 to 1988 in the United States. Multivariable linear regression was used to develop a model to describe factors associated with percent density.

RESULTS

Median (range) percent density among women was 23.7% (0-85%). Body mass index (beta=-0.345, p<0.001 adjusted for age and parity) and parity (beta=-0.277, p<0.001 adjusted for age and BMI) were significantly and inversely associated with percent breast density. After adjusting for parity and BMI, age was not associated with breast density (beta=0.05, p=0.45). Parous women had lower percent density than nulliparous women (23.7 versus 34.7%, p=0.005). Women who had undergone surgical menopause had greater breast density than those who had had a natural menopause (33.4 versus 24.8%, p=0.048), as did women who were not current smokers (26.0 versus 17.3% for smokers, p=0.02). Breast density was not associated with age at menarche, age at menopause, age at first birth, breastfeeding, estrogen levels or androgen levels. In a multivariable model, 24% of the variance in percent breast density was explained by BMI (beta=-0.35), parity (beta=-0.29), surgical menopause (beta=0.13) and current smoking (beta=-0.12).

CONCLUSION

Factors associated with breast density in older, post-menopausal women differ from traditional breast cancer risk factors and from factors associated with breast density in pre-menopausal and younger post-menopausal women.

A Prospective Study of Breast Cancer Risk Using Routine Mammographic Breast Density Measurements

Mammographic breast density is a major risk factor for breast cancer but estimates of the relative risk associated with differing density patterns have varied widely. It is also unclear how menopausal status influences this association and to what extent the effects of density are due to its correlation with other risk factors. Most recent investigations of breast density have been case-control studies, which provide indirect estimates of relative risk. We have prospectively followed 61,844 women for an average of 3.1 years to directly estimate risk among women in the four mammographic breast density categories defined by the American College of Radiology's Breast Imaging Reporting and Data System (BI-RADS). The study was population-based and used density assessments routinely made by community radiologists. Cox regression was used to obtain age-adjusted relative risk estimates and to control for other risk factors. Risk increased with density and the risk associated with extremely dense breasts, relative to entirely fatty breasts, was 4.6 (95% confidence interval, 1.7–12.6) for premenopausal women and 3.9 (95% confidence interval, 2.6–5.8) for postmenopausal women. Estimates for pre- and postmenopausal women did not differ significantly. Although breast density was significantly related to body mass index, age at first childbirth, and postmenopausal hormone use (P &lt; 0.001), adjustment for these variables only slightly altered the relative risk estimates. Our results correspond well to those from case-control studies using more quantitative measures of mammographic breast density and suggest that routine Breast Imaging Reporting and Data System density measurements may be useful in models for assessing breast cancer risk in individual women.

Polymorphism in the Androgen Receptor and Mammographic Density in Women Taking and Not Taking Estrogen and Progestin Therapy

There is some evidence that women with a higher number of CAG repeat lengths on the androgen receptor (AR) gene have increased breast cancer risk. We evaluated the association between AR-CAG repeat length and mammographic density, a strong breast cancer risk factor, in 404 African-American and Caucasian breast cancer patients. In postmenopausal estrogen progestin therapy users, carriers of the less active AR-CAG had statistically significantly higher mean percentage of density (41.4%) than carriers of the more active AR-CAG (25.7%; P = 0.04). Our results raise the question of whether the number of AR-CAG repeats predicts breast cancer risk in estrogen progestin therapy users.