Quantitative evaluation of mammographic densities: a comparison of methods of assessment

Examination of fully automated mammographic density measures using LIBRA and breast cancer risk in a cohort of 21,000 non-Hispanic white women

BACKGROUND

Breast density is strongly associated with breast cancer risk. Fully automated quantitative density assessment methods have recently been developed that could facilitate large-scale studies, although data on associations with long-term breast cancer risk are limited. We examined LIBRA assessments and breast cancer risk and compared results to prior assessments using Cumulus, an established computer-assisted method requiring manual thresholding.

METHODS

We conducted a cohort study among 21,150 non-Hispanic white female participants of the Research Program in Genes, Environment and Health of Kaiser Permanente Northern California who were 40-74 years at enrollment, followed for up to 10 years, and had archived processed screening mammograms acquired on Hologic or General Electric full-field digital mammography (FFDM) machines and prior Cumulus density assessments available for analysis. Dense area (DA), non-dense area (NDA), and percent density (PD) were assessed using LIBRA software. Cox regression was used to estimate hazard ratios (HRs) for breast cancer associated with DA, NDA and PD modeled continuously in standard deviation (SD) increments, adjusting for age, mammogram year, body mass index, parity, first-degree family history of breast cancer, and menopausal hormone use. We also examined differences by machine type and breast view.

RESULTS

The adjusted HRs for breast cancer associated with each SD increment of DA, NDA and PD were 1.36 (95% confidence interval, 1.18-1.57), 0.85 (0.77-0.93) and 1.44 (1.26-1.66) for LIBRA and 1.44 (1.33-1.55), 0.81 (0.74-0.89) and 1.54 (1.34-1.77) for Cumulus, respectively. LIBRA results were generally similar by machine type and breast view, although associations were strongest for Hologic machines and mediolateral oblique views. Results were also similar during the first 2 years, 2-5 years and 5-10 years after the baseline mammogram.

CONCLUSION

Associations with breast cancer risk were generally similar for LIBRA and Cumulus density measures and were sustained for up to 10 years. These findings support the suitability of fully automated LIBRA assessments on processed FFDM images for large-scale research on breast density and cancer risk.

Quantitative Assessment of Percent Breast Density: Analog versus Digital Acquisition

Breast density is a moderate risk factor for breast cancer based on quantitative measurement of percent breast density from film-screen mammograms. In this study, percent breast density was determined using computer-assisted interactive thresholding software from sixty consecutive mammograms of women undergoing digital screening mammography with a prior film-screen mammogram obtained within the last two years. Observations were made regarding discrepancies in density readings. Percent breast density was significantly lower for digital mammograms (mean 32.2%) compared to analog mammograms (mean 40.3%) (p<0.0001). This was not significant for women with less than 20% breast density (range +0.3 to −2.7%), but larger differences were seen with increasing density (12.5–14.9% lower for >50% density). Differences in density readings between analog and digital mammography were largely observed to be due to better recognition of the skin line on digital mammograms resulting in inclusion of more subcutaneous fat. Difficulties with appropriate recognition of subcutaneous breast tissue and fatty tissue near the chest wall were present for both analog and digital mammography. In conclusion, percent breast density is significantly lower when the mammogram is acquired in digital format compared to film-screen, largely due to better recognition of the skin line with resultant inclusion of more subcutaneous fat. Breast cancer risk predictions based on computerized assessment of breast density may be underestimated when applied to digital mammography.

Methods for assessing and representing mammographic density: an analysis of 4 case-control studies

To maximize statistical power in studies of mammographic density and breast cancer, it is advantageous to combine data from several studies, but standardization of the density assessment is desirable. Using data from 4 case-control studies, we describe the process of reassessment and the resulting correlation between values, identify predictors of differences in density readings, and evaluate the strength of the association between mammographic density and breast cancer risk using different representations of density values. The pooled analysis included 1,699 cases and 2,422 controls from California (1990-1998), Hawaii (1996-2003), Minnesota (1992-2001), and Japan (1999-2003). In 2010, a single reader reassessed all images for mammographic density using Cumulus software (Sunnybrook Health Sciences Centre, Toronto, Ontario, Canada). The mean difference between original and reassessed percent density values was -0.7% (95% confidence interval: -1.1, -0.3), with a correlation of 0.82 that varied by location (r = 0.80-0.89). Case status, weight status, age, parity, density assessment method, mammogram view, and race/ethnicity were significant determinants of the difference between original and reassessed values; in combination, these factors explained 9.2% of the variation. The associations of mammographic density with breast cancer and the model fits were similar using the original values and the reassessed values but were slightly strengthened when a calibrated value based on 100 reassessed radiographs was used.

Mammographic breast density patterns among a group of women in sub Saharan Africa

INTRODUCTION

Mammographic breast density is a measure of parenchymal breast patterns on film and in part a marker of cumulative exposure to oestrogen. The risk of breast cancer for women with increased density is up to six fold more than in women with less dense tissues. The pattern of mammographic breast density among Ugandan women is not known.

OBJECTIVE

To establish these as a contribution to baseline data.

METHODS

A cross sectional descriptive study that enrolled women presenting for mammography at the national referral hospital radiology department. Breast densities were scored using the BI-RADS categories. IRB approval was obtained.

RESULTS

Of the 190 women enrolled, 178 were scored, of those scored 10 (5.3%) had extremely dense breasts (grade IV) and 39 (20.5%) had heteregenous ones (grade III). The rest 129 (67.9%) had scattered fibroglandular or fat densities (Grades I & II). Most of the women were young 45.8 ± 12.5 years The majority had normal or benign mammographic findings and all were non pregnant.

CONCLUSION

Mammographic densities in this Ugandan population appear to be of low grade. The pattern established here is markedly different from findings in other studies that indicated much higher proportions for high dense tissues in other races. Mammographic interpretation of films could therefore be easier.

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.

Common breast cancer susceptibility variants in LSP1 and RAD51L1 are associated with mammographic density measures that predict breast cancer risk

BACKGROUND

Mammographic density adjusted for age and body mass index (BMI) is a heritable marker of breast cancer susceptibility. Little is known about the biologic mechanisms underlying the association between mammographic density and breast cancer risk. We examined whether common low-penetrance breast cancer susceptibility variants contribute to interindividual differences in mammographic density measures.

METHODS

We established an international consortium (DENSNP) of 19 studies from 10 countries, comprising 16,895 Caucasian women, to conduct a pooled cross-sectional analysis of common breast cancer susceptibility variants in 14 independent loci and mammographic density measures. Dense and nondense areas, and percent density, were measured using interactive-thresholding techniques. Mixed linear models were used to assess the association between genetic variants and the square roots of mammographic density measures adjusted for study, age, case status, BMI, and menopausal status.

RESULTS

Consistent with their breast cancer associations, the C-allele of rs3817198 in LSP1 was positively associated with both adjusted dense area (P = 0.00005) and adjusted percent density (P = 0.001), whereas the A-allele of rs10483813 in RAD51L1 was inversely associated with adjusted percent density (P = 0.003), but not with adjusted dense area (P = 0.07).

CONCLUSION

We identified two common breast cancer susceptibility variants associated with mammographic measures of radiodense tissue in the breast gland.

IMPACT

We examined the association of 14 established breast cancer susceptibility loci with mammographic density phenotypes within a large genetic consortium and identified two breast cancer susceptibility variants, LSP1-rs3817198 and RAD51L1-rs10483813, associated with mammographic measures and in the same direction as the breast cancer association.

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.

Vitamin D and mammographic breast density: a systematic review

Studies suggest a protective relationship between Vitamin D and breast cancer risk. Several studies assessed the association of Vitamin D with mammographic breast density, a known and strong breast cancer risk factor. Understanding the potential role of Vitamin D in the modification of breast density might open new avenues in breast cancer prevention. This systematic review summarizes published studies that investigated the association between Vitamin D and mammographic breast density and offers suggestions for strategies to advance our scientific knowledge.

Evaluation of mammographic density patterns: reproducibility and concordance among scales

BACKGROUND

Increased mammographic breast density is a moderate risk factor for breast cancer. Different scales have been proposed for classifying mammographic density. This study sought to assess intra-rater agreement for the most widely used scales (Wolfe, Tabár, BI-RADS and Boyd) and compare them in terms of classifying mammograms as high- or low-density.

METHODS

The study covered 3572 mammograms drawn from women included in the DDM-Spain study, carried-out in seven Spanish Autonomous Regions. Each mammogram was read by an expert radiologist and classified using the Wolfe, Tabár, BI-RADS and Boyd scales. In addition, 375 mammograms randomly selected were read a second time to estimate intra-rater agreement for each scale using the kappa statistic. Owing to the ordinal nature of the scales, weighted kappa was computed. The entire set of mammograms (3572) was used to calculate agreement among the different scales in classifying high/low-density patterns, with the kappa statistic being computed on a pair-wise basis. High density was defined as follows: percentage of dense tissue greater than 50% for the Boyd, "heterogeneously dense and extremely dense" categories for the BI-RADS, categories P2 and DY for the Wolfe, and categories IV and V for the Tabár scales.

RESULTS

There was good agreement between the first and second reading, with weighted kappa values of 0.84 for Wolfe, 0.71 for Tabár, 0.90 for BI-RADS, and 0.92 for Boyd scale. Furthermore, there was substantial agreement among the different scales in classifying high- versus low-density patterns. Agreement was almost perfect between the quantitative scales, Boyd and BI-RADS, and good for those based on the observed pattern, i.e., Tabár and Wolfe (kappa 0.81). Agreement was lower when comparing a pattern-based (Wolfe or Tabár) versus a quantitative-based (BI-RADS or Boyd) scale. Moreover, the Wolfe and Tabár scales classified more mammograms in the high-risk group, 46.61 and 37.32% respectively, while this percentage was lower for the quantitative scales (21.89% for BI-RADS and 21.86% for Boyd).

CONCLUSIONS

Visual scales of mammographic density show a high reproducibility when appropriate training is provided. Their ability to distinguish between high and low risk render them useful for routine use by breast cancer screening programs. Quantitative-based scales are more specific than pattern-based scales in classifying populations in the high-risk group.

Association between mammographic density and age-related lobular involution of the breast

PURPOSE

Mammographic density and lobular involution are both significant risk factors for breast cancer, but whether these reflect the same biology is unknown. We examined the involution and density association in a large benign breast disease (BBD) cohort.

PATIENTS AND METHODS

Women in the Mayo Clinic BBD cohort who had a mammogram within 6 months of BBD diagnosis were eligible. The proportion of normal lobules that were involuted was categorized by an expert pathologist as no (0%), partial (1% to 74%), or complete involution (>or= 75%). Mammographic density was estimated as the four-category parenchymal pattern. Statistical analyses adjusted for potential confounders and evaluated modification by parity and age. We corroborated findings in a sample of women with BBD from the Mayo Mammography Health Study (MMHS) with quantitative percent density (PD) and absolute dense and nondense area estimates.

RESULTS

Women in the Mayo BBD cohort (n = 2,667) with no (odds ratio, 1.7; 95% CI, 1.2 to 2.3) or partial (odds ratio, 1.3; 95% CI, 1.0 to 1.6) involution had greater odds of high density (DY pattern) than those with complete involution (P trend < .01). There was no evidence for effect modification by age or parity. Among 317 women with BBD in the MMHS study, there was an inverse association between involution and PD (mean PD, 22.4%, 21.6%, 17.2%, for no, partial, and complete, respectively; P trend = .04) and a strong positive association of involution with nondense area (P trend < .01). No association was seen between involution and dense area (P trend = .56).

CONCLUSION

We present evidence of an inverse association between involution and mammographic density.

Reproducibility of visual assessment on mammographic density

BACKGROUND

High mammographic density was an independent risk factor for breast cancer and has a higher associated risk than most other known risk factors. The reproducibility remains a major issue in assessment of breast parenchymal patterns. Misclassification of mammographic pattern can lead to significant underestimation of risk estimates. The purpose of this study was to assess the inter-rater and intra-rater reliability based on visual subjective mammographic density measurements.

METHOD

Three density measures, Wolfe parenchymal pattern, Boyd classification scale, and a percentage of densities in total breast, were investigated. The study included 101 women who were participants of the International Breast Cancer Intervention Study I (IBIS I) for up to 7 years. Seven sets of mammograms were collected for each woman. Left breast mediolateral oblique films were digitized, and the scanned images were independently reviewed by two readers. These images were reassessed by one reader after a year. The agreements of measures were evaluated by Kappa statistics (Wolfe and Boyd scale) and intraclass correlation coefficient (percentage densities).

RESULTS

For the inter-rater agreement, Weighted Kappa for Wolfe scale was 0.89 (P < 0.0001) and for Boyd scale was 0.84 (P < 0.0001). The intraclass correlation coefficient was 0.94 for percentage densities. For the intra-rater agreement, Weighted Kappa for Wolfe scale was 0.87 (P < 0.0001) and for Boyd scale was 0.86 (P < 0.0001). The intraclass correlation coefficient was 0.96 for percentage densities.

CONCLUSION

The study concludes that both visual qualitative and quantitative measurements on mammographic density are highly reproducible in the breast cancer research studies if appropriate training is provided. The method is appropriate for risk assessment in a prevention trial.

Mammographic breast density, dense area, and breast area differences by phase in the menstrual cycle

BACKGROUND

Mammographic breast density may be greater in the luteal phase (days 15-30) than the follicular phase (days 1-14) of the menstrual cycle; this may have implications for when mammography screening should occur.

OBJECTIVE

Examine whether percent breast density, breast area, or dense area differ by menstrual phase.

METHODS

We identified 204 premenopausal women with regular periods who were <55 years (mean = 45.0 years) and had two screening mammograms within 9 to 18 months, with one screening between days 9 and 14, and one screening between days 22 and 35 of the menstrual cycle. We measured percent breast density, breast area, and dense area using the Cumulus software. We used linear regression to test for differences in breast density, breast area, and dense area from follicular to luteal phase, adjusting for change in weight and time between exams.

RESULTS

The mean (SD) percent breast density was 35.8% (21.3) in the follicular phase and 36.7% (21.3) in the luteal phase. Multivariable analyses showed small but not statistically significant increases in percent density [1.1%; 95% confidence interval (95% CI), -0.2% to 2.3%] and breast area (16.7 cm(2); 95% CI, -2.8 to 36.2) and a statistically significant increase in dense area (13.1 cm(2); 95% CI, 0.1-26.1) in the luteal compared with the follicular phase.

CONCLUSIONS

Breast density, breast area, and dense area have small, but probably not clinically meaningful, increases in the luteal phase of the menstrual cycle. However, there are other factors that may differ by menstrual cycle phase that we were unable to assess (e.g., breast compression), which may ultimately influence mammographic sensitivity by menstrual cycle phase.

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.

Mammographic Density Measured with Quantitative Computer-aided Method: Comparison with Radiologists' Estimates and BI-RADS Categories

PURPOSE

To retrospectively compare computer-aided mammographic density estimation (MDEST) with radiologist estimates of percentage density and Breast Imaging Reporting and Data System (BI-RADS) density classification.

MATERIALS AND METHODS

Institutional Review Board approval was obtained for this HIPAA-compliant study; patient informed consent requirements were waived. A fully automated MDEST computer program was used to measure breast density on digitized mammograms in 65 women (mean age, 53 years; range, 24-89 years). Pixel gray levels in detected breast borders were analyzed, and dense areas were segmented. Percentage density was calculated by dividing the number of dense pixels by the total number of pixels within the borders. Seven breast radiologists (five trained with MDEST, two not trained) prospectively assigned qualitative BI-RADS density categories and visually estimated percentage density on 260 mammograms. Qualitative BI-RADS assessments were compared with new quantitative BI-RADS standards. The reference standard density for this study was established by allowing the five trained radiologists to manipulate the MDEST gray-level thresholds, which segmented mammograms into dense and nondense areas. Statistical tests performed include Pearson correlation coefficients, Bland-Altman agreement method, kappa statistics, and unpaired t tests.

RESULTS

There was a close correlation between the reference standard and radiologist-estimated density (R = 0.90-0.95) and MDEST density (R = 0.89). Untrained radiologists overestimated percentage density by an average of 37%, versus 6% for trained radiologists (P < .001). MDEST showed better agreement with the reference standard (average overestimate, 1%; range, -15% to +18%). MDEST correlated better with percentage density than with qualitative BI-RADS categories. There were large overlaps and ranges of percentage density in qualitative BI-RADS categories 2-4. Qualitative BI-RADS categories correlated poorly with new quantitative BI-RADS categories, and 16 (6%) of 260 views were erroneously classified by MDEST.

CONCLUSION

MDEST compared favorably with radiologist estimates of percentage density and is more reproducible than radiologist estimates when qualitative BI-RADS density categories are used. Qualitative and quantitative BI-RADS density assessments differed markedly.

Factors Affecting Sensitivity and Specificity of Screening Mammography and MRI in Women with an Inherited Risk for Breast Cancer

BACKGROUND

The MRISC study is a screening study, in which women with an increased risk of hereditary breast cancer are screened by a yearly mammography and MRI, and half-yearly clinical breast examination. The sensitivity found in this study was 40% for mammography and 71% for MRI and the specificity was 95 and 90%, respectively. In the current subsequent study we investigated whether these results are influenced by age, a BRCA1/2 mutation, menopausal status and breast density.

PATIENTS AND METHODS

From November 1999 to October 2003, 1909 eligible women were screened and 50 breast cancers were detected. For the current analysis, data of 4134 screening rounds and 45 detected breast cancers were used. For both imaging modalities, screening parameters, receiver operating characteristic (ROC) curves and uni- and multivariate odds ratios (ORs) were calculated. All analyses were separately performed for age at entry (< 40, 40-49, > or =50), mutation status, menopausal status and breast density.

RESULTS

Sensitivity of MRI was decreased in women with high breast density (adjusted OR 0.08). False-positive rates of both mammography (OR(adj) 1.67) and MRI (OR(adj) 1.21) were increased by high breast density, that of MRI by pre-menopausal status (OR(adj) 1.70), young age (OR(adj) 1.58 for women 40-49 years versus women > or =50 years) and decreased in BRCA1/2 mutation carriers (OR(adj) 0.74). In all investigated subgroups the discriminating capacity (measured by the area under the ROC-curve) was higher for MRI than for mammography, with the largest differences for BRCA1/2 mutation carriers (0.237), for women between 40 and 49 years (0.227) and for women with a low breast density (0.237).

CONCLUSIONS

This report supports the earlier recommendation that MRI should be a standard screening method for breast cancer in BRCA1/2 mutation carriers.

Evaluating hormone therapy-associated increases in breast density comparison between reported and simultaneous assignment of BI-RADS categories, visual assessment, and quantitative analysis

RATIONALE AND OBJECTIVES

Changes in breast density, which are commonly associated with hormone replacement therapy (HRT) use, may imply changes in breast cancer risk. This study explores the ability of different methods to detect hormone replacement therapy (HRT)-associated increases in breast density.

MATERIALS AND METHODS

Between 1997 and 2001, 51 postmenopausal women were reported to have HRT-associated increases in breast density at our institution. Twenty postmenopausal women not reported to have an increase in density during the same period were selected as controls. Mammograms from date of report and earlier comparison were used. Breast Imaging Reporting and Data System (BI-RADS) density categories from both dates were obtained from the mammography report. Mammograms were reviewed at separate time points and density changes evaluated by assigning BI-RADS density categories, visual assessment, and computer-assisted quantitative analysis.

RESULTS

Mammogram reports were not available for two patients. The remaining 49 women with reported HRT increases in density were included. Reported BI-RADS categories resulted in detection of 57%, simultaneous BI-RADS assignment in 61%, visual assessment in 100%, and quantitative assessment in 94% of women with HRT-associated increases in density. Reported BI-RADS category change was the only method that resulted in false-positive increases in density for control patients. Minimal HRT associated increases in density were the most difficult to detect, with 90% of these 21 cases not detected by simultaneous BI-RADS category assignment and 3 cases not detected by quantitative methods when defined as an increase of at least 5%.

CONCLUSION

Visual and quantitative assessment best identified women with HRT-associated increases in density, including those with minimal increases. Simultaneous assignment of BI-RADS categories was considerably better than use of reported BI-RADS categories. This information may be helpful in guiding research design of studies evaluating changes in density from the HRT use.

Likelihood of additional work-up among women undergoing routine screening mammography: the impact of age, breast density, and hormone therapy use

BACKGROUND

Mammography screening can involve subsequent work-up to determine a final screening outcome. Understanding the likelihood of different events that follow initial screening is important if women and their health care providers are to be accurately informed about the screening process.

METHODS

We conducted an analysis of additional work-up following screening mammography to characterize use of supplemental imaging and recommendations for biopsy and/or surgical consultation and the factors associated with their use. We included all events following screening mammography performed between 1/1/1998 and 12/31/1999 on a population-based sample of 37,632 New Hampshire women. We calculated adjusted odds ratios (OR) and 95% confidence intervals (CI) for supplemental imaging and recommended biopsy and/or surgical consultation as function of age, menopausal status and HRT use, breast density, and family history of breast cancer.

RESULTS

Ninety-one percent of women (n = 34,445) did not require supplemental imaging. Among those who did (n = 3187), 84% had additional views, 9% ultrasound, and 7% received both. Supplemental imaging was affected by age (OR 0.84; 95% CI = 0.76-0.94 for 50-59; OR = 0.66; 95% CI = 0.58-0.75 for > or = 60 versus < 50), menopausal status, and HRT use (OR = 1.33; 95% CI = 1.21-1.47 for peri- or post-menopausal HRT users; OR = 1.14; 95% CI = 1.01-1.29 for premenopausal versus peri- or post-menopausal non-HRT users), breast density (OR = 1.43; 95% CI = 1.33-1.55 for dense versus fatty breasts) and family history (OR = 1.15; 95% CI = 1.06-1.25 for any versus none). In women with supplemental imaging, age (OR = 1.80; 95% CI = 1.11-2.90 for > or = 60, relative to <50) and imaging type (OR = 3.23; 95% CI = 2.38-4.38 for ultrasound with or without additional views versus additional views only) were significantly associated with biopsy and/or surgical consultation recommendation. In those with no supplemental imaging, breast density was associated with recommended biopsy and/or surgical consultation (OR = 1.53; 95% CI = 1.13-2.07 for dense versus fatty breasts).

CONCLUSIONS

Breast density and HRT use are both independent predictors of use of supplemental imaging in women. With advancing age (age 60 and older), women were less likely to require follow-up imaging but more likely to receive a recommendation for biopsy and/or surgical consultation. This information should be used to inform women about the likelihood of services received as part of the screening work-up.

Quantitative Assessment of Mammographic Breast Density: Relationship with Breast Cancer Risk

Increased mammographic breast density is a moderate independent risk factor for breast cancer, with findings of published studies in which quantitative methods of assessment were used showing a positive association. Breast density may be quantified by using visual assessment or planimetry. Although the category definitions vary, the odds ratio for developing breast cancer for the most dense compared with the least dense breast tissue categories ranges from 1.8 to 6.0, with most studies yielding an odds ratio of 4.0 or greater. Plausible explanations for the association of breast density with increased breast cancer risk may be the development of premalignant lesions such as atypical ductal hyperplasia, elevated growth factors, or increased estrogen production within the breast due to overactive aromatase. The amount of breast density may be due in part to genetic heredity. However, unlike other risk factors, breast density may be influenced. Specifically, breast density is very hormonally responsive and potentially may be influenced by lifestyle factors such as alcohol intake and diet. Assessment of breast density may become useful in risk assessment and prevention decisions.

Mammography screening in African American women: evaluating the research

BACKGROUND

Notwithstanding some controversy regarding the benefits of screening mammography, it is generally assumed that the effects are the same for women of all race/ethnic groups. Yet evidence for its efficacy from clinical trial studies comes primarily from the study of white women. It is likely that mammography is equally efficacious in white and African American women when applied under relatively optimal clinical trial conditions, but in actual practice African Americans may not be receiving equal benefit, as reflected in their later stage at diagnosis and greater mortality.

METHODS

Initial searches of Medline using search terms related to screening mammography, race, and other selected topics were supplemented with national data that are routinely published for cancer surveillance. Factors that potentially compromise the benefits of mammography as it is delivered in the current health care system to African American women were examined.

RESULTS

While there have been significant improvements in mammography screening utilization, observational data suggest that African American women may still not be receiving the full benefit. Potential explanatory factors include low use of repeat screening, inadequate followup for abnormal exams, higher prevalence of obesity and, possibly, breast density, and other biologic factors that contribute to younger age at diagnosis.

CONCLUSIONS

Further study of biologic factors that may contribute to limited mammography efficacy and poorer breast cancer outcomes in African American women is needed. In addition, strategies to increase repeat mammography screening and to ensure that women obtain needed followup of abnormal mammograms may increase early detection and improve survival among African Americans. Notwithstanding earlier age at diagnosis for African American women, mammography screening before age 40 years is not recommended, but screening of women aged 40-49 years is particularly critical.

Mammographic tissue, breast cancer risk, serial image analysis, and digital mammography. Part 1. Tissue and related risk factors

This work is presented as a sequence of two parts. In this leading section, a review of the breast tissue-risk research is provided. Although controversy remains, there is substantial evidence indicating that dense mammographic tissue (a) is a breast cancer risk factor that is at least similar, if not greater, in magnitude with the other known breast cancer risk factors and (b) may be a partial biomarker for some of the other risk factors. Understanding these influences may provide a mechanism for measuring the dynamics of breast cancer risk. The totality of this work is to provide support for an automated serial mammography study under way at the authors' institution, where digital mammographic images are acquired with a full-field digital mammography system. This is a filmless imaging system, where the image is acquired in digital format. This electronic imaging acquisition system provides a prime opportunity to easily couple and manipulate the image data with patient information such as risk probability analysis or other pertinent personal history data for improved automated decision making. In this leading section, the main focus is on understanding elements that will assist in fusing risk probability analysis with automated computer-aided diagnosis. The evidence indicates that there are many factors that influence breast tissue at any given time and thus have the ability to alter the associated radiographic image appearance over time. At the initiation of the serial study it was clear that the authors did not fully understand the nature of the problem: automatically comparing similar mammographic scenes acquired at different times. In the second part of this sequence, the more time-related tissue influences are reviewed.

A breast density index for digital mammograms based on radiologists' ranking

The purpose of this study was to develop and evaluate a computerized method of calculating a breast density index (BDI) from digitized mammograms that was designed specifically to model radiologists' perception of breast density. A set of 153 pairs of digitized mammograms (cranio-caudal, CC, and mediolateral oblique, MLO, views) were acquired and preprocessed to reduce detector biases. The sets of mammograms were ordered on an ordinal scale (a scale based only on relative rank-ordering) by two radiologists, and a cardinal (an absolute numerical score) BDI value was calculated from the ordinal ranks. The images were also assigned cardinal BDI values by the radiologists in a subsequent session. Six mathematical features (including fractal dimension and others) were calculated from the digital mammograms, and were used in conjunction with single value decomposition and multiple linear regression to calculate a computerized BDI. The linear correlation coefficient between different ordinal ranking sessions were as follows: intraradiologist intraprojection (CC/CC): r = 0.978; intraradiologist interprojection (CC/MLO): r = 0.960; and interradiologist intraprojection (CC/CC): r = 0.968. A separate breast density index was derived from three separate ordinal rankings by one radiologist (two with CC views, one with the MLO view). The computer derived BDI had a correlation coefficient (r) of 0.907 with the radiologists' ordinal BDI. A comparison between radiologists using a cardinal scoring system (which is closest to how radiologists actually evaluate breast density) showed r = 0.914. A breast density index calculated by a computer but modeled after radiologist perception of breast density may be valuable in objectively measuring breast density. Such a metric may prove valuable in numerous areas, including breast cancer risk assessment and in evaluating screening techniques specifically designed to improve imaging of the dense breast.