B re a s t D e n s i t y

Where Are We Now?

Eric Kim, MDa, Alana A. Lewin, MDa,b,*

KEYWORDS
 Breast density  Breast cancer risk  Dense breasts  Supplemental breast cancer screening

KEY POINTS
 Breast density is a heritable and dynamic trait associated with age, race/ethnicity, body mass index
and hormonal factors.
 Increased breast density may have a masking effect on mammography and is an independent risk
factor for breast cancer.
 The American College of Radiology has published appropriate use criteria for supplemental
screening based on breast density due to the negative impact of breast density on mammographic
sensitivity.

INTRODUCTION area1 of fibroglandular tissue within the breast

(Fig. 2).
Breast density refers to the amount of fibrogland- As a result, breast density remains an important
ular tissue relative to fat and is determined either factor in the interpretation of screening mammog-
visually or quantitatively on mammography. The raphy. Breast density not only has the potential to
assessment of breast density is important as it in- mask a malignancy but is an independent risk fac-
dicates the possibility that a lesion could be tor for breast cancer. Dense breast tissue also is
obscured by normal tissue. Fibroglandular tissue associated with an increased risk of interval breast
absorbs relatively more ionizing radiation cancer3 and an increased incidence of high-risk le-
compared with fatty tissue and projects as grada- sions and nonproliferative benign breast disease.4
tions of white on mammography. In an attempt to Owing to the ongoing attention toward personal-
standardize the description of breast density, the ized care in medicine and in particular breast can-
Breast Imaging Reporting and Data System (BI- cer screening, there is increasing interest in breast
RADS) fifth edition described four categories of density as a modifiable risk factor for breast can-
breast density: category A (almost entirely fatty), cer and a tool for risk stratification. In addition,
B (scattered fibroglandular tissue), C (heteroge- there is enhanced focus on methods to optimize
neously dense), and D (extremely dense).1 breast cancer screening and offer supplemental
(Fig. 1, Table 1). Dense breasts are common as tools to improve cancer detection in women with
approximately 36% of women more than 40 year dense breasts.
old have heterogeneously dense breasts and 7%
have extremely dense breasts.2 Previously, the
ASSESSMENT OF BREAST DENSITY
breast density categories were defined by the total
Visual Assessment
amount of fibroglandular tissue within the breast.
However, in order to reflect the relationship be- The most commonly used method to categorize
tween breast density and the risk of masking breast density is subjective visual assessment.
breast cancer, the BI-RADS fifth edition changed The fourth edition of BI-RADS which was released
the definition of density to reflect the densest in 2003 used quartile ranges of the percentage of
radiologic.theclinics.com

a
Department of Radiology, New York University Grossman School of Medicine, New York, NY, USA; b New
York University Grossman School of Medicine, New York University Langone Health, Laura and Isaac Perlmutter
Cancer Center, 160 East 34th Street 3rd Floor, New York, NY 10016, USA
* Corresponding author. 160 East 34th Street, 3rd Floor, New York, NY 10016.
E-mail address: [email protected]

Radiol Clin N Am 62 (2024) 593–605

https://doi.org/10.1016/j.rcl.2023.12.007
0033-8389/24/Ó 2023 Elsevier Inc. All rights reserved.
Descargado para Luis Hurtado ([email protected]) en Cayetano Heredia Pervuvian University de ClinicalKey.es por Elsevier en agosto 23, 2024.
Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2024. Elsevier Inc. Todos los derechos reservados.
594 Kim & Lewin

Fig. 1. Left mediolateral oblique (MLO) views demonstrate the BI-RADS fifth edition breast density categories: (A)
Almost entirely fatty. (B) Scattered fibroglandular density. (C) Heterogeneously dense. (D) Extremely dense.

dense fibroglandular tissue (<25%, 25%–50%, have areas of primarily dense tissue with other
50%–75%, and >75%) for the four density cate- areas that are primarily fatty, BI-RADS suggests
gories. However, the fifth edition of BI-RADS that it may be helpful to the referring clinician to
which was released in 2013 has removed these describe the location of the denser tissue. In
percentage ranges. Instead, BI-RADS now em- breasts that are not of equal density, the denser
phasizes the text descriptions of breast density breast should be used to categorize breast density.
to focus on the masking effect of dense fibrogland- A well-known limitation of visual assessment is
ular tissue to detect breast cancer. that there is considerable intraobserver and inter-
According to BI-RADS, radiologists should make observer variation in the categorization of breast
an overall assessment of breast density based on density.5–8 This may be due to multiple individual
the relative possibility of normal fibroglandular tis- factors including reader fatigue, the predisposition
sue to obscure an underlying breast lesion. This to overestimate or underestimate density, and var-
should be based on the densest area of tissue iable reading room conditions. A study which
within the breast; therefore, a breast may be cate- involved 30 radiology facilities in the United States
gorized as dense even when only a focal area of from the Population-based Research Optimizing
dense tissue is present in an otherwise predomi- Screening through Personalized Regimens con-
nantly fatty breast (see Fig. 2). In breasts that sortium demonstrated a wide range (6.3%–
84.5%) of screening mammograms that were
interpreted as dense by radiologists.8 In a recent
Table 1 study, statistically significant interobserver vari-
Breast density categories based on the breast ability also was seen at the international level
imaging reporting and data system (BI-RADS) when comparing breast imaging radiologists prac-
fifth edition ticing in Indonesia, the Netherlands, South Africa,
and the United States.9 Because of this variability,
Category Terms
quantitative and deep learning methods have
A The breasts are almost entirely fatty emerged in an effort to standardize the assess-
B There are scattered areas of ment of breast density.
fibroglandular density
C The breasts are heterogeneously Quantitative Assessment
dense, which may obscure small
Fully automated quantitative methods have been
masses
developed which calculate either the volume or
D The breasts are extremely dense,
area-based density percentage on mammog-
which lowers the sensitivity of
mammography
raphy. The two most well-known commercially
available and Food and Drug Administration

Descargado para Luis Hurtado ([email protected]) en Cayetano Heredia Pervuvian University de ClinicalKey.es por Elsevier en agosto 23, 2024.
Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2024. Elsevier Inc. Todos los derechos reservados.
 Breast Density 595

Fig. 2. Left mediolateral oblique

(MLO) (A) and craniocaudal (CC) (B)
views of a 46-year-old woman who
presented for screening mammogram.
Dense fibroglandular tissue is visual-
ized in the upper breast. Even though
the percentage density in the breast is
less than 50%, the mammogram was
classified as heterogeneously dense ac-
cording to the BI-RADS fifth edition
which states that density should be
based on the ability of the densest
area of tissue within the breast to
mask a lesion.

(FDA)-approved methods are Volpara (Volpara semiautomated method that requires human input
Health, Wellington, New Zealand) (Fig. 3) and including segmentation of the breast. These
Quantra (Hologic, Danbury, Connecticut). Both methods have limited clinical utility because they
calculate the volumetric breast density percentage are time-consuming and less reliable than volume-
by dividing the volume of fibroglandular tissue by based methods.10,16 Fully automated area-based
the total breast volume and convert this into an methods also were developed but never achieved
appropriate BI-RADS breast density category. widespread use.
These automated methods have been shown to
be both consistent and reproducible.10,11 Howev-
Artificial Intelligence
er, studies have shown mixed results in regard to
correlation between different automated methods Recently, deep learning-based methods have been
as well as between automated methods and developed to categorize breast density on mam-
radiologists.12–15 mograms with encouraging results.17–19 Lehman
Previously, area-based density percentage and colleagues trained a deep convolutional neural
methods also were used, primarily in the research network to assess breast density on more than
setting. The most well-known is Cumulus (University 41,000 screening mammograms. They were one
of Toronto, Toronto, Ontario, Canada), which is a of the first to implement their model in the clinical

Fig. 3. Bilateral mediolateral oblique (MLO) and craniocaudal (CC) views from screening mammogram demon-
strate dense breast tissue (A). The volumetric breast density percentage was calculated using fully-automated Vol-
para software and results in BI-RADS breast density category D or extremely dense breasts (B). (Courtesy of Dr.
Stamatia DeStounis.)

Descargado para Luis Hurtado ([email protected]) en Cayetano Heredia Pervuvian University de ClinicalKey.es por Elsevier en agosto 23, 2024.
Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2024. Elsevier Inc. Todos los derechos reservados.
596 Kim & Lewin

setting and demonstrated very good (94%) agree- relationship with mammographic density and age
ment between their model and radiologists in the bi- continues postmenopausally and is most pro-
nary categorization of dense versus non-dense nounced in the menopausal transition. These
breasts.17 In a more recent prospective study with age-related density changes are seen across
more than 85,000 consecutive screening mammo- ethnic groups of women worldwide, suggesting a
grams, Dontchos and colleagues evaluated clinical common intrinsic biological mechanism.26 Parity
implementation of their deep learning model at is inversely associated with breast density.27,28 In
three sites. Interestingly, they demonstrated a sig- addition, two recent studies27,28 found that older
nificant reduction in the radiologist classification age at first birth and older age at menarche are
of mammograms as dense after exposure to their associated with higher density. A retrospective
deep learning model. They also reported reduced study of BRCA1/2 mutation carriers who under-
reader variability in density assessment at the sites went bilateral oophorectomy showed that breast
exposed to the model.20 density significantly decreased after oophorec-
There are now multiple FDA-approved AI and tomy.29 Younger patients demonstrated greater
machine learning tools to assess breast density absolute decreases compared with older
including commercial options from Volpara and patients.29
Quantra.21 The limitations of deep learning models Menopausal hormone therapy (MHT) is used to
include the need to verify the models in diverse pa- relieve the common symptoms of menopause,
tient populations, across different mammographic including hot flashes, sleep disturbances, mood
vendors, and across multiple institutions. Further alteration, muscle, and joint pain. Studies have
studies need to be performed to compare the re- shown an association between systemic MHT
sults in breast density categorization between and increased density. Specifically, estrogen plus
deep learning methods and quantitative volume- progestin MHT is associated with higher density
based methods. than estrogen alone.27 Mammographic density at
least partially accounts for the association be-
FACTORS THAT INFLUENCE BREAST DENSITY tween MHT and breast cancer.27,30
Tamoxifen, a selective estrogen receptor
Breast density is a heritable and dynamic trait. modulator, can reduce breast density and breast
Studies show that the heritability of breast density cancer risk. High-risk women receiving tamoxifen
is high, with genetic factors accounting for 58% to chemoprevention who experienced  10%
67% of breast density variation.22–24 Breast den- reduction in breast density had a 63% reduction
sity is associated with age, hormonal factors, in breast cancer risk compared with placebo
race/ethnicity, and body mass index (BMI).25 group.31 Women who had less than 10% reduc-
tion in breast density had no breast cancer risk
Hormonal
reduction.31 Further research has shown a large
Exogenous and endogenous hormonal factors decrease in breast density over the first year of
impact breast density. Generally, mammographic tamoxifen therapy with continued decline there-
density gradually decreases as age increases in after.32 Studies on aromatase inhibitors and
the premenopausal period, with the exception of breast density reduction demonstrate mixed
pregnancy and breastfeeding (Fig. 4). This inverse results.33–35

Fig. 4. Bilateral mediolateral oblique (MLO) views from a 35-year-old woman who presents for mammogram
while breastfeeding demonstrate heterogeneously dense breasts (A). Bilateral mediolateral oblique views
when the same patient presents for screening mammogram 5 years later demonstrate interval decreased breast
density, now characterized as scattered areas of fibroglandular density (B).

Descargado para Luis Hurtado ([email protected]) en Cayetano Heredia Pervuvian University de ClinicalKey.es por Elsevier en agosto 23, 2024.
Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2024. Elsevier Inc. Todos los derechos reservados.
 Breast Density 597

Race/Ethnicity Although malignant calcifications remain well visu-

alized, noncalcified cancers (approximately 45%
Studies on breast density and its association with
of invasive cancers49) can be masked by dense
race and ethnicity have shown mixed results.36
tissue. The sensitivity of mammography to detect
Breast density is well studied among Asian women
breast cancer diminishes with increasing breast
who have a high prevalence of heterogeneously
density. The sensitivity range of mammography
dense and extremely dense breasts.14 According
is 81% to 93% for fatty breasts, 84% to 90% for
to the Breast Cancer Surveillance Consortium
breasts with scattered fibroglandular density,
(BCSC), in the United States, Asian women have
69% to 81% for heterogeneously dense breasts,
the highest proportion of dense breasts, followed
and 57% to 71% for extremely dense breasts.46
in descending order by non-Hispanic White
In addition, women with dense breasts have
women, Hispanic/Latina women, and non-
more interval cancers, which are cancers diag-
Hispanic Black/African American women.25 How-
nosed in between successive screening mammo-
ever, the effect of race and ethnicity on breast
grams, than women with nondense breasts.50,51
density are often limited by confounding factors
Even with the use of tomosynthesis,52 approxi-
such as use of hormone replacement therapy,
mately 25% of breast cancers are missed on
BMI, and reproductive factors.37 When correcting
mammography in women with heterogeneously
for these factors, there is still variability in breast
dense breasts and about 40% in women with
density by race/ethnicity.38
extremely dense breasts46,48,50
Nonetheless, higher breast density is associ-
ated with an increased risk of breast cancer within
certain racial/ethnic groups,38 Although a smaller Independent Risk Factor for Breast Cancer
proportion of Black/African American women
Multiple studies have demonstrated that breast
have dense breasts, Black/African American
density is an independent risk factor for the devel-
women with dense breasts are at an increased
opment of breast cancer, although there is
risk of breast cancer compared with Black/African
continued debate about the degree to which it in-
American women with nondense breasts.39
creases lifetime risk. This was first described by
Wolfe when he demonstrated an increased risk
Lifestyle and Environmental Factors for developing breast cancer based solely on a
There is growing interest in the impact of lifestyle classification system for the appearance of breast
and environmental factors on breast density. BMI parenchyma on mammography.53,54 Since then,
is inversely related to breast density in adult- numerous additional studies have been published
hood.40 Interestingly, high BMI increases the risk confirming that dense breast tissue increases the
of developing postmenopausal breast cancer, risk of breast cancer.
but not premenopausal breast cancer.38,41 In a meta-analysis which included 42 studies,
Reduced breast cancer risk following weight loss McCormack and colleagues demonstrated a
or bariatric surgery may be due in part to reduction 4.64-fold (95% CI 3.65–5.91) increased risk of
in total volume of dense breast tissue.42,43 In light breast cancer in patients with the most dense
of this information, breast density must be cor- breasts compared with those with the least dense
rected for BMI and age in the Tyrer-Cuzick (TC) breasts.55 A more recent meta-analysis by Bod-
model to add the independent contribution of ewes and colleagues published in 2022 analyzed
breast density to breast cancer risk.44 studies which used the breast density classifica-
The role of alcohol and tobacco use on breast tion system from the more recent BI-RADS fifth
density has also been explored. Studies show edition.56 The analysis included nine studies which
that there is a relationship between higher breast together demonstrated a 3.89-fold (95% CI 2.47–
density and alcohol intake and lower breast den- 6.13) increased breast cancer risk in patients
sity and tobacco use.27,45 with extremely dense breast tissue in comparison
with those with extremely fatty breast tissue. How-
ever, less than 10% of women have either
CLINICAL IMPLICATIONS
extremely dense or extremely fatty breasts. There-
Masking Effect on Mammography: Decreased
fore, the investigators also investigated the relative
Sensitivity
risk of patients with dense breasts to those with
Mammographic performance and resultant scattered density, which they described as the
screening outcomes are significantly impacted average breast density. In this comparison, they
by breast density. Both increased false-positive found a 1.28-fold increase (95% CI 1.19–1.37) in
findings46,47 and reduced cancer detection rates women with heterogeneously dense breasts and
are seen in women with dense breasts48 (Fig. 5). a 2.11-fold (95% CI 1.84–2.42) in women with

Descargado para Luis Hurtado ([email protected]) en Cayetano Heredia Pervuvian University de ClinicalKey.es por Elsevier en agosto 23, 2024.
Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2024. Elsevier Inc. Todos los derechos reservados.
598 Kim & Lewin

Fig. 5. A 67-year-old woman who presented for annual screening mammogram and screening ultrasound. (A)
Left mediolateral oblique (MLO) and craniocaudal (CC) views demonstrate breast density category D: extremely
dense. There is no mammographic evidence of malignancy. However, screening ultrasound (B) demonstrates a
0.6 cm irregular, hypoechoic mass at left 5:00, 3 cm from the nipple. This underwent ultrasound core biopsy
yielding invasive ductal carcinoma, well-differentiated, estrogen receptor positive, progesterone receptor posi-
tive, HER2-neu receptor negative.

extremely dense breasts compared with those multiple limitations in the use of current risk
with scattered density. models. Most of the currently used risk models
In a literature review including 102 studies, provide general risk estimates at the population
increased breast density has been associated level rather than at the individual level, and
with increased tumor size, although there are con- different models are known to underestimate
flicting data regarding lymph node status and no breast cancer risk in different subgroups of pa-
evidence to support an increased risk of distant tients.63 A recently published study also showed
metastasis.57 that there is substantial variability between
Increased mammographic density has also different models in estimating breast cancer risk.64
been shown to be associated with an increased The most common risk models currently used
rate of interval cancers.58,59 This may be in part are the Gail, TC, and BCSC models. Each model
due to the masking effect of breast density, uses a different combination of factors to calculate
although some suggest that underlying biological individual risk (Table 2). For instance, the Gail
changes may also play a role. The findings are model, also known as the Breast Cancer Risk
important because interval cancers are known to Assessment Tool, includes criteria such as repro-
be more aggressive and have a poorer prognosis ductive history, history of breast biopsies, and
than screen-detected cancers.60 A pooled anal- first-degree relatives with breast cancer. The TC
ysis which included six studies and 3492 invasive model, also known as the International Breast
breast cancers demonstrated a positive correla- Intervention Study model, is more comprehensive,
tion between percent mammographic density including additional criteria such as greater family
with an increased risk of breast cancer across all history (first- and second-degree relatives with his-
molecular subtypes of breast cancer.61 Interest- tory of breast cancer) and BMI.
ingly, a recently published longitudinal study found Breast density is not included as a factor in
that the decrease in breast density over time was many of the risk models, such as the widely used
slower in patients who developed breast cancer, Gail model. However, at least three models,
suggesting that breast density is a dynamic risk including the latest version of the TC model, the
factor which can change over time.62 BSCS model, and the Breast and Ovarian Analysis
of Disease Incidence and Carrier Estimation Algo-
Overview of Risk Models rithm model, include breast density information.
There are several models used to assess the risk Studies have shown that the addition of breast
for developing breast cancer, although there are density improves the performance of the Gail

Descargado para Luis Hurtado ([email protected]) en Cayetano Heredia Pervuvian University de ClinicalKey.es por Elsevier en agosto 23, 2024.
Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2024. Elsevier Inc. Todos los derechos reservados.
 Breast Density 599

Table 2
Factors included in commonly used breast cancer risk models

Gail Tyrer-Cuzick (Version 8) Breast Cancer Surveillance Consortium

 Age  Age  Age
 Age at menarche  Age at menarche  Race/ethnicity
 Age at first live birth  Age at first live birth  Family history of breast cancer
 Race/ethnicity  Age at menopause  First-degree relatives
 Family history of  Body mass index  History of prior breast biopsies
breast cancer  Breast density  BI-RADS breast density
 First-degree  Hormone replacement therapy
relatives use
 History of prior  BRCA gene
breast biopsies  Family history of breast cancer
 First-degree relatives
 Second-degree relatives
 History of prior breast biopsies
 History of ovarian cancer

and TC risk models.65–67 In a systematic review patients are notified about their breast density. In
with 11 studies including mammographic breast 2011, the FDA Mammography Quality Assurance
density improved risk prediction in all of the tested Advisory Committee advised that the FDA should
breast cancer risk models.68 McCarthy and col- require breast density reporting both in mammog-
leagues demonstrated that the BCSC model had raphy reports to health care providers and in lay
higher accuracy than other models that they language summaries to patients. On March 10,
tested, including the Gail and TC models, in pre- 2023, the FDA issued a rule (effective September
dicting breast cancer risk among women who 10, 2024) that will require the BI-RADS density
had breast density information available in their category (using the 2013 BI-RADS language1) to
cohort of more than 35,000 women.69 Volume- be included in mammography reports sent to
based breast density percentage has been shown health care providers.74 This also stipulates spe-
to be more strongly associated with breast cancer cific language in mammography letters sent to pa-
risk than area-based breast density percentage.70 tients, which cannot be altered (Fig. 6).
Recently, deep learning models trained on In addition, facilities will have to comply with all
mammographic images have been developed to applicable federal, state, and local reporting re-
estimate breast cancer risk. Early results demon- quirements. The FDA estimates that there will be
strate improvement over breast density-based reduced mortality and reduced breast cancer treat-
scores and models,71,72 suggesting that there is ment costs as a result of density notification.74
more image-based information than density alone However, in order to observe these benefits,
which plays an important role in risk prediction. women and their providers must be educated
Newer risk models are being developed to include about supplemental screening and have access to
additional information such as genetics to further supplemental screening modalities.
improve breast cancer risk prediction. Improving
the performance of risk models is critical in identi-
fying patients who may benefit from supplemental Supplemental Screening Based on Breast
imaging or chemoprevention. Density
Research shows that there are socioeconomic,
SUPPLEMENTAL SCREENING BASED ON geographic, racial, and ethnic disparities in both
BREAST DENSITY the awareness of breast density and its implica-
tions for supplemental screening.75 The modalities
Although mammography reduces breast cancer
available to patients with dense breasts are
mortality,73 the impact of breast density on
impacted by practice type (academic vs private
mammographic performance underscores the
practice) and geography.76 The most common
need for more effective screening strategies.
supplemental modalities available are whole-
breast ultrasound (WBUS), digital breast tomosyn-
Changes to Dense Breast Notification Laws
thesis (DBT), and breast MR imaging.
In 2009, Connecticut was the first state to enact In 2021, the American College of Radiology
dense breast notification laws, mandating that (ACR) published appropriate use criteria for

Descargado para Luis Hurtado ([email protected]) en Cayetano Heredia Pervuvian University de ClinicalKey.es por Elsevier en agosto 23, 2024.
Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2024. Elsevier Inc. Todos los derechos reservados.
600 Kim & Lewin

Fig. 6. Food and Drug Administration (FDA) language required in mammogram results letters to patients as of
September 10, 2024. Breasts assessed by the radiologist as “not dense” include BI-RADS density category A
(almost entirely fatty) and B (scattered areas of fibroglandular density). “Dense” breasts include those assessed
as BI-RADS category C (heterogeneously dense, which may obscure small masses) and D (extremely dense,
which lowers the sensitivity of mammography).

supplemental breast cancer screening based on cancer detection rate (CDR) of 2 to 2.7/1000.81 How-
breast density.77 The document reviewed six clin- ever, as demonstrated in the multi-institutional
ical scenarios based on breast cancer risk and ACRIN 6666 trial and other studies, the supple-
breast density and presented evidence for the mental cancer detection with ultrasound is associ-
appropriate use of various imaging studies in the ated with high screening recall rates, high rate of
specific contexts. In 2023, the ACR released short-term follow-up, and lower positive predictive
updated breast cancer screening guidelines78 (Ta- value for biopsy compared with mammography,
ble 3), some of which address scenarios of pa- and MR imaging.82–84 Therefore, the ACR recom-
tients with varying risk factors and different mends that women with elevated risk due to breast
breast density. density may consider adjunctive screening with ul-
Multiple studies confirm the incremental cancer trasound after weighing the risks and benefits.78
detection of WBUS. Whole-breast screening ultra- Multiple prospective studies in women at
sound (WBUS) increases the detection of early inva- higher-than-average risk of breast cancer demon-
sive node-negative breast cancers in women with strate that breast MR imaging has higher sensi-
mammographically dense breast tissue and varying tivity than mammography, ultrasound, and
patient risk factors73,79,80 with an incremental mammography plus ultrasound combined. The

Table 3
Comparison of the American College of Radiology recommendations for annual imaging (with
consideration of breast density)

ACR Recommendations 201876 Updated ACR Recommendations 202377

Lifetime risk 20% Annual DM  DBT Annual DM  DBT
(regardless of Annual MR imaging Annual MR imaging (starting at age 30 y)
breast density) (starting at age 30 y)
Personal history of Annual DM  DBT (from the Annual DM  DBT (from the age of diagnosis)
breast cancer age of diagnosis) Annual MR imaging if dense breasts or if
Annual MR imaging if dense diagnosed before age 50 y
breasts or if diagnosed
before age 50
Dense breast tissue Annual DM  DBT consider Annual DM  DBT annual MR imaging
annual MR imaging or Consider CEM or ultrasound as alternative to
ultrasound (age 40 y or MR imaging (age 40 y or earlier if other
earlier if other risk factors) risk factors)

Abbreviations: DBT, digital breast tomosynthesis; DM, digital mammography; MR imaging, contrast-enhanced breast MR
imaging.
Table adapted and modified from: Monticciolo DL et al. Breast Cancer Screening for Women at Higher-Than-Average
Risk: Updated Recommendations From the ACR. J Am Coll Radiol 2023 May 5;S1546-1440(23)00334-4.

Descargado para Luis Hurtado ([email protected]) en Cayetano Heredia Pervuvian University de ClinicalKey.es por Elsevier en agosto 23, 2024.
Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2024. Elsevier Inc. Todos los derechos reservados.
 Breast Density 601

American Cancer Society advocates for screening a state insurance law is in effect, some insurance
with breast MR imaging in high-risk women plans are exempt from state laws, including na-
regardless of breast density. In addition, the ACR tional plans like Medicare and the Veterans
recommends that women with genetics-based Administration.
increased risk (including BRCA1 carriers), those
with a calculated lifetime risk of  20%, and those
FUTURE DIRECTIONS
exposed to chest radiation at a young age are rec-
Textural Analysis and Radiomics
ommended to have MR imaging surveillance start-
ing at age 25 to 30 years, regardless of breast Breast density provides a global assessment of
density.78 the relative amount of fibroglandular tissue in the
Breast MR imaging also is recommended for breast. However, there is increasing interest in
women with dense breasts who desire supple- analyzing the more complex parenchymal pattern
mental screening. For those who qualify for but of the breast. These more granular textural fea-
cannot undergo breast MR imaging, contrast- tures have been proposed as possible imaging
enhanced mammography (CEM) or ultrasound biomarkers for breast cancer risk. The idea behind
may be considered.78 A recent meta-analysis by this is that medical images contain information
Hussein and colleagues found that for women at reflecting underlying biological processes which
average or intermediate risk for breast cancer with can be analyzed quantitatively.93 Radiomics refers
dense breasts, MR imaging was statistically supe- to the practice of high-throughput extraction of
rior to other supplemental modalities, with better mineable quantitative data from digital medical im-
cancer detection rate than ultrasound or DBT.85 ages. Advantages to this include reproducibility
Given the recent evidence that women with dense and scalability over subjective or qualitative
breast tissue alone may benefit from supplemental methods. There is growing evidence that textural
screening with MR imaging,86,87 the European Soci- features have the ability to provide more informa-
ety of Breast Imaging has updated their recommen- tion than breast density alone in breast cancer
dations to reflect this change.88 In the DENSE risk assessment.
(Breast Cancer Screening With MR imaging in One of the first studies to look at textural anal-
Women Aged 50–75 Years With Extremely Dense ysis was performed by Byng and colleagues.94
Breast Tissue) randomized controlled trial in the The investigators conducted a nested case-
Netherlands, the prevalence-round MR imaging control study using automated analysis of two
screening in women with extremely dense breasts textural mammographic features, one describing
yielded 16.5/1000 cancers and a 50% reduction in the distribution of breast density and the other
interval cancer rate, suggesting a mortality benefit89; characterizing texture. The study demonstrated a
this was shown to be cost-effective in modeling.90 In moderately increased relative risk of breast cancer
the second MR imaging screening round, the CDR with these two textural features after adjusting for
was 5.8/1000 and the false-positive rate decreased the effect of other risk factors. Several additional
from 79.8% to 26.3%.91 It should be noted that studies since then have also demonstrated
women were invited for biennial screening and increased breast cancer risk after analyzing
therefore that the impact of MR imaging on interval textural features from mammography.95–97
cancer rates might vary compared with a study eval- The investigators of a recent systematic review
uating annual screening. which included 28 articles published between
2016 and 2021 suggest that risk prediction model
performance increases when textural features are
Insurance Coverage
added to breast density.98 For example, Kontos
Issues with insurance coverage/denials and out- and colleagues demonstrated a significant associ-
of-pocket costs remain a concern for patients ation of their radiomics-derived textural features
recommended for supplemental screening. State from mammography with breast cancer. They
insurance laws for supplemental screening are also demonstrated improved performance of their
inconsistent. Currently, 21 states and the District model after including textural features compared
of Columbia have enacted insurance laws for with a baseline model which only included per-
either supplemental US and/or MR imaging. Of centage density and BMI (area under the curve
note, for most of the state insurance laws, out- [AUC] of 0.84 vs 0.80).99
of-pocket costs such as copays and deductibles Most of the studies thus far have been retro-
still apply. As of June 19, 2023, 14 states either spective case-control studies. As a result, addi-
have no-cost sharing legislation in effect or soon tional and larger prospective studies are needed
to be in effect (ie, no expense to the patient) for before textural features are included in breast can-
supplemental screening.92 However, even when cer risk prediction models. Of note, it is also

Descargado para Luis Hurtado ([email protected]) en Cayetano Heredia Pervuvian University de ClinicalKey.es por Elsevier en agosto 23, 2024.
Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2024. Elsevier Inc. Todos los derechos reservados.
602 Kim & Lewin

difficult to perform a meta-analysis or assess 2. Prevalence of mammographically dense breasts in

reproducibility in this area given the heterogeneity the United States. J Natl Cancer Inst 2014;106.
in the approach and number of features that are 3. Mandelson MT, Oestreicher N, White PL. Breast
included. Despite academic interest, clinical appli- density as a predictor of mammographic detection:
cation or evaluation of radiomic approaches are comparison of interval- and screen-detected can-
still lacking. However, radiomics remains a prom- cers. J Natl Cancer Inst 2000;92:1081–7.
ising field with the potential to improve our knowl- 4. Boyd NF, Jensen HM, Cooke G, et al. Relationship be-
edge of breast cancer risk, diagnosis, and tween mammographic and histological risk factors for
treatment. breast cancer. J Natl Cancer Inst 1992;84:1170–9.
5. Ciatto S, Houssami N, Apruzzese A, et al. Categoriz-
SUMMARY ing breast mammographic density: intra- and inter-
observer reproducibility of BI-RADS density
Breast density is a heritable trait that is influenced categories. Breast 2005;14:269–75.
by age, hormonal factors, race/ethnicity, and BMI. 6. Ooms EA, Zonderland HM, Eijkemans MJ, et al.
There are important clinical implications of having Mammography: interobserver variability in breast
increased breast density including decreased density assessment. Breast 2007;16:568–76.
mammographic sensitivity and increased risk of 7. Redondo A, Comas M, Macià F, et al. Inter- and in-
developing breast cancer. Standardized methods traradiologist variability in the BI-RADS assessment
of density assessment that are reproducible and and breast density categories for screening mam-
reliable are becoming increasingly important. This mograms. Br J Radiol 2012;85:1465–70.
will allow for the addition of breast density informa- 8. Sprague BL, Conant EF, Onega T, et al. Variation in
tion to improve the accuracy of risk assessment Mammographic Breast Density Assessments
models and assess the need for supplemental Among Radiologists in Clinical Practice: A Multi-
screening on an individual basis. Breast density re- center Observational Study. Ann Intern Med 2016;
mains a significant and complex topic in breast im- 165:457–64.
aging that will continue to evolve as we learn more 9. Portnow LH, Malek L, Leung JWT, et al. International
about its associations and clinical implications. Interobserver Variability of Breast Density Assess-
ment. J Am Coll Radiol 2023. https://doi.org/10.
CLINICS CARE POINTS 1016/j.jacr.2023.03.010.
10. Alonzo-Proulx O, Mawdsley GE, Patrie JT, et al. Reli-
ability of automated breast density measurements.
Radiology 2015;275:366–76.
 Mammographic breast density is traditionally 11. Engelken F, Jasmin-Maya S, Eva-Maria F, et al. Volu-
assessed visually, although there are now metric breast composition analysis: reproducibility
Food and Drug Administration-approved, of breast percent density and fibroglandular tissue
fully automated quantitative, and artificial volume measurements in serial mammograms.
intelligence methods available. Acta radiol 2014;55:32–8.
 Reproducible and reliable breast density 12. Lee HN, Sohn Y-M, Han KH. Comparison of mammo-
assessment is important to improve the accu- graphic density estimation by Volpara software with
racy of breast cancer risk assessment models radiologists’ visual assessment: analysis of clinical-
and to optimize personalized breast cancer radiologic factors affecting discrepancy between
screening. them. Acta radiol 2015;56:1061–8.
 Supplemental screening with whole-breast 13. Morrish OWE, Tucker L, Black R, et al. Mammo-
ultrasound, digital breast tomosynthesis, graphic breast density: comparison of methods
and breast MR imaging should be considered for quantitative evaluation. Radiology 2015;275:
in patients with dense breasts. 356–65.
14. Brandt KR, Scott CG, Ma L, et al. Comparison of
Clinical and Automated Breast Density Measure-
DISCLOSURE ments: Implications for Risk Prediction and Supple-
No relevant relationships. mental Screening. Radiology 2016;279:710–9.
15. Gweon HM, Youk JH, Kim J-A, et al. Radiologist
assessment of breast density by BI-RADS cate-
REFERENCES
gories versus fully automated volumetric assess-
1. Acr. 2013 ACR BI-RADS Atlas: Breast Imaging Re- ment. AJR Am J Roentgenol 2013;201:692–7.
porting and Data System. (2014). Available at: 16. Yaffe MJ. Mammographic density. Measurement of
https://www.acr.org/Clinical-Resources/Reporting- mammographic density. Breast Cancer Res 2008;
and-Data-Systems/Bi-Rads. 10:209.

Descargado para Luis Hurtado ([email protected]) en Cayetano Heredia Pervuvian University de ClinicalKey.es por Elsevier en agosto 23, 2024.
Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2024. Elsevier Inc. Todos los derechos reservados.
 Breast Density 603

17. Lehman CD, Yala A, Schuster T, et al. Mammographic 31. Cuzick J, Warwick J, Pinney E, et al. Tamoxifen-
Breast Density Assessment Using Deep Learning: induced reduction in mammographic density and
Clinical Implementation. Radiology 2019;290:52–8. breast cancer risk reduction: a nested case-control
18. Saffari N, Rashwan HA, Abdel-Nasser M, et al. Fully study. J Natl Cancer Inst 2011;103:744–52.
Automated Breast Density Segmentation and Classi- 32. Brentnall AR, Warren R, Harkness EF, et al. Mammo-
fication Using Deep Learning. Diagnostics 2020;10. graphic density change in a cohort of premeno-
19. Matthews TP, Singh S, Mombourquette B, et al. pausal women receiving tamoxifen for breast
A Multisite Study of a Breast Density Deep Learning cancer prevention over 5 years. Breast Cancer
Model for Full-Field Digital Mammography and Syn- Res 2020;22:101.
thetic Mammography. Radiol Artif Intell 2021;3: 33. Cigler T, Richardson H, Yaffe MJ, et al. A randomized,
e200015. placebo-controlled trial (NCIC CTG MAP.2) exam-
20. Dontchos BN, Cavallo-Hom K, Lamb LR, et al. ining the effects of exemestane on mammographic
Impact of a Deep Learning Model for Predicting breast density, bone density, markers of bone meta-
Mammographic Breast Density in Routine Clinical bolism and serum lipid levels in postmenopausal
Practice. J Am Coll Radiol 2022;19:1021–30. women. Breast Cancer Res Treat 2011;126:453–61.
21. Center for Devices & Radiological Health. Artificial 34. Vachon CM, Suman VJ, Brandt KR, et al. Mammo-
Intelligence and Machine Learning (AI/ML)- Enabled graphic breast density response to aromatase inhi-
Medical Devices. U.S. Food and Drug Administra- bition. Clin Cancer Res 2013;19:2144–53.
tion. 2022. Available at: https://www.fda.gov/ 35. Prowell TM, Blackford AL, Byrne C, et al. Changes in
medical-devices/software-medical-device-samd/ breast density and circulating estrogens in post-
artificial-intelligence-and-machine-learning-aiml- menopausal women receiving adjuvant anastrozole.
enabled-medical-devices. Accessed August 28, Cancer Prev Res 2011;4:1993–2001.
2023. 36. McCarthy AM, Keller BM, Pantalone LM, et al.
22. Mavaddat N, Michailidou K, Dennis J, et al. Poly- Racial Differences in Quantitative Measures of
genic Risk Scores for Prediction of Breast Cancer Area and Volumetric Breast Density. J Natl Cancer
and Breast Cancer Subtypes. Am J Hum Genet Inst 2016;108.
2019;104:21–34. 37. El-Bastawissi AY, White E, Mandelson MT, et al.
23. Boyd NF, Dite GS, Stone J, et al. Heritability of Reproductive and hormonal factors associated
mammographic density, a risk factor for breast can- with mammographic breast density by age (United
cer. N Engl J Med 2002;347:886–94. States). Cancer Causes Control 2000;11:955–63.
24. Holowko N, Eriksson M, Kuja-Halkola R, et al. Herita- 38. Bissell MCS, Kerlikowske K, Sprague BL, et al.
bility of Mammographic Breast Density, Density Breast Cancer Surveillance Consortium. Breast Can-
Change, Microcalcifications, and Masses. Cancer cer Population Attributable Risk Proportions Associ-
Res 2020;80:1590–600. ated with Body Mass Index and Breast Density by
25. Chalfant JS, Hoyt AC. Breast Density: Current Race/Ethnicity and Menopausal Status. Cancer Epi-
Knowledge, Assessment Methods, and Clinical Im- demiol Biomarkers Prev 2020;29(10):2048–56.
plications. J Breast Imaging 2022;4:357–70. 39. Friebel-Klingner TM, Ehsan S, Conant EF, et al. Risk
26. Burton A, Maskarinec G, Perez-Gomez B, et al. factors for breast cancer subtypes among Black
Mammographic density and ageing: A collaborative women undergoing screening mammography.
pooled analysis of cross- sectional data from 22 Breast Cancer Res Treat 2021;189(3):827–35.
countries worldwide. PLoS Med 2017;14:e1002335. 40. Lee KH, Chae SW, Yun JS, et al. Association be-
27. Azam S, Sjölander A, Eriksson M, et al. Determinants tween skeletal muscle mass and mammographic
of Mammographic Density Change. JNCI Cancer breast density. Sci Rep 2021;11:16785.
Spectr 2019;3:kz004. 41. Premenopausal Breast Cancer Collaborative Group
28. Alexeeff SE, Odo NU, McBride R, et al. Reproduc- et al. Association of Body Mass Index and Age
tive Factors and Mammographic Density: Associa- With Subsequent Breast Cancer Risk in Premeno-
tions Among 24,840 Women and Comparison of pausal Women. JAMA Oncol 4, e181771 (2018).
Studies Using Digitized Film-Screen Mammography 42. Hassinger TE, Mehaffey JH, Knisely AT, et al. The
and Full-Field Digital Mammography. Am J Epide- impact of bariatric surgery on qualitative and quan-
miol 2019;188:1144–54. titative breast density. Breast J 2019;25:1198–205.
29. Lecler A, Dunant A, Delaloge S, et al. Breast tissue 43. Williams AD, So A, Synnestvedt M, et al. Mammo-
density change after oophorectomy in BRCA muta- graphic breast density decreases after bariatric sur-
tion carrier patients using visual and volumetric anal- gery. Breast Cancer Res Treat 2017;165:565–72.
ysis. Br J Radiol 2018;91(1083):20170163. 44. Brentnall AR, Cuzick J. Risk Models for Breast Can-
30. Rice MS, Bertrand KA, VanderWeele TJ, et al. cer and Their Validation. Stat Sci 2020;35:14–30.
Mammographic density and breast cancer risk: a 45. McBride RB, Fei K, Rothstein JH, et al. Alcohol and
mediation analysis. Breast Cancer Res 2016;18:94. Tobacco Use in Relation to Mammographic Density

Descargado para Luis Hurtado ([email protected]) en Cayetano Heredia Pervuvian University de ClinicalKey.es por Elsevier en agosto 23, 2024.
Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2024. Elsevier Inc. Todos los derechos reservados.
604 Kim & Lewin

in 23,456 Women. Cancer Epidemiol Biomarkers 60. Mook S, Van ’t Veer LJ, Rutgers EJ, et al. Indepen-
Prev 2020;29:1039–48. dent prognostic value of screen detection in invasive
46. Kerlikowske K, Zhu W, Tosteson ANA, et al. Identi- breast cancer. J Natl Cancer Inst 2011;103:585–97.
fying women with dense breasts at high risk for inter- 61. Kleinstern G, Scott CG, Tamimi RM, et al. Associa-
val cancer: a cohort study. Ann Intern Med 2015; tion of mammographic density measures and breast
162:673–81. cancer “intrinsic” molecular subtypes. Breast Can-
47. Conant EF, Barlow WE, Herschorn SD, et al. Associ- cer Res Treat 2021;187(1):215–24.
ation of Digital Breast Tomosynthesis vs Digital 62. Jiang S, Bennett DL, Rosner BA, et al. Longitudinal
Mammography With Cancer Detection and Recall Analysis of Change in Mammographic Density in
Rates by Age and Breast Density. JAMA Oncol Each Breast and Its Association With Breast Cancer
2019;5:635–42. Risk. JAMA Oncol 2023;9:808–14.
48. Destounis S, Johnston L, Highnam R, et al. Using 63. Amir E, Freedman OC, Seruga B, et al. Assessing
Volumetric Breast Density to Quantify the Potential women at high risk of breast cancer: a review of
Masking Risk of Mammographic Density. AJR Am risk assessment models. J Natl Cancer Inst 2010;
J Roentgenol 2017;208:222–7. 102:680–91.
49. Gajdos C, Tartter PI, Bleiweiss IJ, et al. Mammo- 64. Paige JS, Lee CI, Pin-Chieh W, et al. Variability
graphic appearance of nonpalpable breast cancer Among Breast Cancer Risk Classification Models
reflects pathologic characteristics. Ann Surg 2002; When Applied at the Level of the Individual Woman.
235:246–51. J Gen Intern Med 2023. https://doi.org/10.1007/
50. Wanders JOP, Holland K, Karssemeijer N, et al. The s11606-023-08043-4.
effect of volumetric breast density on the risk of 65. Brentnall AR, Harkness EF, Astley SM, et al.
screen-detected and interval breast cancers: a Mammographic density adds accuracy to both the
cohort study. Breast Cancer Res 2017;19(1):67. Tyrer-Cuzick and Gail breast cancer risk models in
51. Holland K, van Gils CH, Mann RM, et al. Quantifica- a prospective UK screening cohort. Breast Cancer
tion of masking risk in screening mammography with Res 2015;17:147.
volumetric breast density maps. Breast Cancer Res 66. Brentnall AR, et al. A Case-Control Study to Add
Treat 2017;162:541–8. Volumetric or Clinical Mammographic Density into
52. Berg WA, Rafferty EA, Friedewald SM, et al. the Tyrer-Cuzick Breast Cancer Risk Model.
Screening Algorithms in Dense Breasts: Expert Panel J Breast Imaging 2019;1:99–106.
Narrative Review. AJR Am J Roentgenol 2021;216: 67. Tice JA, Cummings SR, Ziv E, et al. Mammographic
275–94. breast density and the Gail model for breast cancer
53. Wolfe JN. Risk for breast cancer development deter- risk prediction in a screening population. Breast
mined by mammographic parenchymal pattern. Cancer Res Treat 2005;94:115–22.
Cancer 1976;37:2486–92. 68. Vilmun BM, Vejborg I, Lynge E, et al. Impact of add-
54. Wolfe JN. Breast patterns as an index of risk for ing breast density to breast cancer risk models: A
developing breast cancer. AJR Am J Roentgenol systematic review. Eur J Radiol 2020;127:109019.
1976;126:1130–7. 69. McCarthy AM, Guan Z, Welch M, et al. Performance
55. McCormack VA, dos Santos Silva I. Breast density of Breast Cancer Risk-Assessment Models in a
and parenchymal patterns as markers of breast can- Large Mammography Cohort. J Natl Cancer Inst
cer risk: a meta-analysis. Cancer Epidemiol Bio- 2020;112(5):489–97.
markers Prev 2006;15:1159–69. 70. Gastounioti A, Pantalone L, Scott CG, et al. Fully
56. Bodewes FTH, van Asselt AA, Dorrius MD, et al. Automated Volumetric Breast Density Estimation
Mammographic breast density and the risk of breast from Digital Breast Tomosynthesis. Radiology 2021;
cancer: A systematic review and meta- analysis. 301(3):561–8.
Breast 2022;66:62–8. 71. Yala A, Lehman C, Schuster T, et al. A Deep
57. Shawky MS, Huo CW, Henderson MA, et al. A review Learning Mammography- based Model for
of the influence of mammographic density on breast Improved Breast Cancer Risk Prediction. Radiology
cancer clinical and pathological phenotype. Breast 2019;292:60–6.
Cancer Res Treat 2019;177:251–76. 72. Dembrower K, Liu Y, Azizpour H, et al. Comparison
58. Boyd NF, Guo H, Martin LJ, et al. Mammographic of a Deep Learning Risk Score and Standard
density and the risk and detection of breast cancer. Mammographic Density Score for Breast Cancer
N Engl J Med 2007;356:227–36. Risk Prediction. Radiology 2020;294(2):265–72.
59. Pollán M, Ascunce N, Ederra M, et al. Mammo- 73. Smith RA, Duffy SW, Gabe R, et al. The randomized
graphic density and risk of breast cancer according trials of breast cancer screening: what have we
to tumor characteristics and mode of detection: a learned? Radiol Clin North Am 2004;42:793–806, v.
Spanish population-based case-control study. 74. Website. Department of Health and Human Ser-
Breast Cancer Res 2013;15:R9. vices. Food and Drug Administration Mammography

Descargado para Luis Hurtado ([email protected]) en Cayetano Heredia Pervuvian University de ClinicalKey.es por Elsevier en agosto 23, 2024.
Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2024. Elsevier Inc. Todos los derechos reservados.
 Breast Density 605

Quality Standards Act final rule. Available at: https:// Imaging for Breast Cancer Screening in Dense
public-inspection.federalregister.gov/2023-04550. Breast Tissue. Acad Radiol 2017;24:316–20.
pdf. 87. Chen S-Q, Huang M, Shen Y-Y, et al. Abbreviated
75. Huang S, Houssami N, Brennan M, et al. The impact MRI Protocols for Detecting Breast Cancer in
of mandatory mammographic breast density notifi- Women with Dense Breasts. Korean J Radiol 2017;
cation on supplemental screening practice in the 18:470–5.
United States: a systematic review. Breast Cancer 88. Mann RM, Athanasiou A, Baltzer PA, et al. Breast
Res Treat 2021;187:11–30. cancer screening in women with extremely dense
76. Choudhery S, Patel BK, Johnson M, et al. Trends of breasts recommendations of the European Society
Supplemental Screening in Women With Dense of Breast Imaging (EUSOBI). Eur Radiol 2022;32:
Breasts. J Am Coll Radiol 2020;17:990–8. 4036–45.
77. Expert Panel on Breast Imaging et al. ACR Appropri- 89. Bakker MF, de Lange SV, Pijnappel RM, et al. Sup-
ateness CriteriaÒ Supplemental Breast Cancer plemental MRI Screening for Women with Extremely
Screening Based on Breast Density. J. Am. Coll. Ra- Dense Breast Tissue. N Engl J Med 2019;381:
diol. 18, S456–S473 (2021). 2091–102.
78. Monticciolo DL, Newell MS, Moy L, et al. Breast Can- 90. Geuzinge HA, Bakker MF, Heijnsdijk EAM, et al.
cer Screening for Women at Higher-Than-Average Cost-Effectiveness of Magnetic Resonance Imaging
Risk: Updated Recommendations From the ACR. Screening for Women With Extremely Dense Breast
J Am Coll Radiol 2023. https://doi.org/10.1016/j. Tissue. J Natl Cancer Inst 2021;113:1476–83.
jacr.2023.04.002. 91. Veenhuizen SGA, de Lange SV, Bakker MF, et al.
Supplemental Breast MRI for Women with Extremely
79. Kolb TM, Lichy J, Newhouse JH. Comparison of the
Dense Breasts: Results of the Second Screening
performance of screening mammography, physical
Round of the DENSE Trial. Radiology 2021;299:
examination, and breast US and evaluation of fac-
278–86.
tors that influence them: an analysis of 27,825 pa-
92. Berg WA, Seitzman RL, Pushkin J. Implementing the
tient evaluations. Radiology 2002;225:165–75.
National Dense Breast Reporting Standard, Expand-
80. Kim WH, Chang JM, Lee J, et al. Diagnostic perfor-
ing Supplemental Screening Using Current Guide-
mance of tomosynthesis and breast ultrasonogra-
lines, and the Proposed Find It Early Act. J Breast
phy in women with dense breasts: a prospective
Imaging 2023;wbad034.
comparison study. Breast Cancer Res Treat 2017;
93. Gillies RJ, Kinahan PE, Hricak H. Radiomics: Images
162:85–94.
Are More than Pictures, They Are Data. Radiology
81. Berg WA, Vourtsis A. Screening Breast Ultrasound
2016;278:563–77.
Using Handheld or Automated Technique in Women
94. Byng JW, Yaffe MJ, Lockwood GA, et al. Automated
with Dense Breasts. J Breast Imaging 2019;1:
analysis of mammographic densities and breast car-
283–96.
cinoma risk. Cancer 1997;80:66–74.
82. Corsetti V, Houssami N, Ferrari A, et al. Breast 95. Manduca A, Carston MJ, Heine JJ, et al. Texture fea-
screening with ultrasound in women with tures from mammographic images and risk of breast
mammography-negative dense breasts: evidence cancer. Cancer Epidemiol Biomarkers Prev 2009;18:
on incremental cancer detection and false positives, 837–45.
and associated cost. Eur J Cancer 2008;44:539–44. 96. Wei J, Ping Chan H, Yi-Ta W, et al. Association of
83. Berg WA, Zhang Z, Lehrer D, et al. Detection of computerized mammographic parenchymal pattern
breast cancer with addition of annual screening ul- measure with breast cancer risk: a pilot case-
trasound or a single screening MRI to mammog- control study. Radiology 2011;260:42–9.
raphy in women with elevated breast cancer risk. 97. Zheng Y, Keller BM, Ray S, et al. Parenchymal
JAMA 2012;307:1394–404. texture analysis in digital mammography: A fully
84. Brem RF, Tabár L, Duffy SW, et al. Assessing automated pipeline for breast cancer risk assess-
improvement in detection of breast cancer with ment. Med Phys 2015;42:4149–60.
three-dimensional automated breast US in women 98. Anandarajah A, Chen Y, Colditz GA, et al. Studies of
with dense breast tissue: the SomoInsight Study. parenchymal texture added to mammographic
Radiology 2015;274:663–73. breast density and risk of breast cancer: a system-
85. Hussein H, Abbas E, Keshavarzi S, et al. Supple- atic review of the methods used in the literature.
mental Breast Cancer Screening in Women with Breast Cancer Res 2022;24:101.
Dense Breasts and Negative Mammography: A Sys- 99. Kontos D, Winham SJ, Oustimov A, et al. Radiomic
tematic Review and Meta-Analysis. Radiology 2023; Phenotypes of Mammographic Parenchymal
306:e221785. Complexity: Toward Augmenting Breast Density in
86. Chen S-Q, Huang M, Shen Y-Y, et al. Application of Breast Cancer Risk Assessment. Radiology 2019;
Abbreviated Protocol of Magnetic Resonance 290:41–9.

Descargado para Luis Hurtado ([email protected]) en Cayetano Heredia Pervuvian University de ClinicalKey.es por Elsevier en agosto 23, 2024.
Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2024. Elsevier Inc. Todos los derechos reservados.