While mammography screening has been proven effective in reducing breast cancer deaths, treatment advances cannot overcome the disadvantage of being diagnosed with an advanced-stage tumor. New screening guidance from the American College of Radiology and Society of Breast Imaging recommends annual mammography screening beginning at age 40, which provides the greatest mortality reduction, diagnosis at earlier stage, better surgical options, and more effective chemotherapy. Screening should continue past age 74 years, without an upper age limit unless severe comorbidities limit life expectancy. Updated guidelines also cover screening for transgender patients.
Breast cancer remains the most common nonskin cancer, the second leading cause of cancer death, and the leading cause of premature death, measured by the average and total years of life lost, for women in the United States.1 Breast cancer currently accounts for 30% of all new invasive cancer diagnoses in women. For 2021, 333 490 new invasive and in situ cases and 44 130 breast cancer deaths are expected nationwide.2
Breast imaging experts from the ACR and Society of Breast Imaging (SBI) have reviewed extensive data from randomized controlled trials (RCTs), observational trials, and peer-reviewed literature on digital mammography, digital breast tomosynthesis (DBT), ultrasound, and MRI. Our analysis incorporates the ACR Appropriateness Criteria, which use accepted robust strength-of-evidence methodology.3
Since our most recent guideline4, new data are available to evaluate the impact of screening using advanced-stage and fatal breast cancers, to assess the utilization and benefits of DBT, and to evaluate disparities in breast cancer screening and treatment among minority women. With this guideline, we offer recommendations more inclusive of all women of average risk for breast cancer.
Mammography remains the principal modality of early detection for women of average risk. Early detection allows for diagnosis of tumors of smaller sizes with fewer nodal metastases and less histologic grade progression, making treatment more effective.5- 8 The decline in advanced-stage disease afforded by mammography screening is directly related to substantial declines in breast cancer mortality.5,9- 12 The ACR and SBI recommend annual mammography screening starting at age 40 to maximize these benefits. Both benefits and risks of mammography screening should be considered to assist women in making informed choices.
Benefits of Mammography Screening
Mammography screening has been proven effective in reducing breast cancer death in women age 40 years and over.4,9,13- 17
The 11 mammography RCTs have demonstrated a collective 22% reduction in breast cancer mortality9,18, despite the limitations of RCT design. RCTs test only the invitation to screening, not actual screening. RCTs that achieved a 20% or greater reduction in advanced-stage disease (a measure of trial efficacy) had average mortality reduction of 28% for invited women—corresponding to an approximately 40% mortality reduction for women actually screened.9
Numerous observational trials have demonstrated mortality reductions of 40% or greater with organized screening.13,14,19- 23 The largest service screening trial to date screened 2.8 million women and achieved a 40% mortality reduction in every province in Canada in every age group, including those 40 to 49 years.14
Cancer Intervention and Surveillance Modeling Network (CISNET) models have been used to evaluate US screening. The US Preventive Service Task Force has commissioned CISNET to examine outcomes of particular interest to them.24- 26 For example, 2016 CISNET models considered screening only to age 74, even though data exist beyond that age, so that recommendations by organizations other than the US Preventive Service Task Force could not be evaluated.26,27 Regardless, independent researchers studying CISNET models have confirmed that annual screening starting at age 40 would result in a 40% reduction in breast cancer mortality24,27, consistent with observational studies worldwide.
There is risk in not screening. Treatment advances are important but cannot overcome the disadvantage of being diagnosed with an advanced-stage tumor. Using the incidence rates of fatal breast cancers, Tabar et al have shown that, with equivalent state-of-the-art treatment, screened women had 60% lower breast cancer mortality at 10-year follow-up and 47% lower mortality at 20-year follow-up than unscreened women over a 58-year study period.6 Using similar methodologies, Duffy et al found that among 559 091 women who participated in screening, there was a statistically significant 41% reduction in breast cancer deaths within 10 years of diagnosis and a 25% reduction in the rate of advanced cancers, compared with unscreened women, independent of treatment regimen.5 Participation in screening saved lives that would otherwise be lost with current therapies.
Mammography-detected tumors are more effectively treated.28- 33 Women in their 40s29 and women 75 and older30 with mammographically detected cancers have lower-stage disease, less treatment, lower recurrence rates, and better disease-specific survival. The method of detection is an independent prognostic factor for breast cancer mortality.28,31
Risks of Mammography Screening
The primary, quantifiable risks of screening mammography are recall for additional imaging and, after diagnostic evaluation, recommendation for biopsy that yields benign results. Recall occurs when an abnormality identified at screening requires further evaluation; if the ultimate result is not malignant, this is referred to by some as a false-positive mammogram. Lee et al reviewed 5 680 743 screening examinations in the National Mammography Database, the largest mammography registry in the United States. On average, 9.6% of screened women had to return for additional imaging.34 In other studies, risk of recall after a screening examination for women of all ages ranged from 9.6% to 11.6%.34- 37 Most abnormal screening examinations are resolved by imaging alone. Recommendations for biopsy, most performed with minimally invasive percutaneous techniques, occur in fewer than 2% of screened women.38
A woman screened annually starting at age 40 could expect recall for benign diagnostic workup once every 13 years and to undergo a benign biopsy once every 187 years, based on CISNET modeling.27 Modeling the risks of screening based on different screening regimens demonstrates more risks (primarily recalls) with annual screening beginning at age 40 compared with other strategies, but often these risks are overstated.26,27 Although recall and biopsy recommendations are higher with more frequent screening, so are life-years gained and deaths averted.27 Comparing biopsy risk to benefit shows that a woman screened annually starting at age 40 could expect 1 life-year gained for every benign biopsy.27
Efforts to reduce risks have been made through improvements in mammographic technology. Specifically, DBT decreases false-positive mammograms while simultaneously increasing cancer detection. In the largest study to date, involving 454 850 patients, Friedewald et al showed the recall rate decreased from 10.7% to 9.1% (relative decrease 15%) using DBT, with a concomitant increase in cancer detection rate (CDR) from 4.2 in 1000 to 5.4 in 1000 (relative increase 29%).39
These improvements are sustainable. The digital mammography recall rate decreased from 10.4% to 8.8% after the prevalent screen with DBT. After 5 years of DBT screening, the recall rate decreased to 6.1% while maintaining the improvement in CDR.40
Anxiety has been given as a reason to delay the onset and decrease the frequency of screening mammography.18,41 This fails to account for the highly variable experience of anxiety among women who attend screening42- 45 and also the added anxiety and regret women may endure when diagnosed with advanced breast cancer that might have been detected earlier with screening. Women have repeatedly indicated their willingness to undergo diagnostic mammograms and false-positive results in exchange for increased likelihood of early detection.46,47 Efforts to diminish anxiety, such as individual and public education of patients, rapid delivery of results, and other strategies, are encouraged.48- 52
Overdiagnosis occurs when a cancer is detected that would not have otherwise become clinically apparent in a woman's lifetime.53,54 Overdiagnosis cannot be measured directly. Unless specificity and sensitivity are 100%, the results of any test will either overdiagnose or underdiagnose disease. A meta-analysis by the EUROSCREEN group showed overdiagnosis ranged from 0% to 10% (including ductal carcinoma in situ) among adequately adjusted studies [55]. The higher end of the range occurs in older women, who are more likely to die of other causes than younger women. Current benchmarks produce less than 1% overdiagnosis for women in their 40s and less than 2% for women in their 50s (Table 1).54,56,57
Table 1. Overdiagnosis rates as a function of age and assuming an average lead time of 24 months
| Age (y) | DCIS | Invasive | All Cancer |
|---|
| 40 | 0.5 | 0.3 | 0.3 |
| 50 | 1.2 | 0.7 | 0.8 |
| 60 | 2.8 | 1.8 | 2.1 |
| 70 | 8.4 | 6.0 | 6.5 |
| 80 | 19.6 | 13.9 | 14.8 |
Delaying screening until age 45 or 50, or lengthening screening interval, will not decrease the already low levels of overdiagnosis.58 Overdiagnosed tumors persist on imaging just as lethal tumors do and will be diagnosed at the next screening, which, if delayed, could result in later-stage diagnosis.58 Delay in discovery of lethal tumors would result in underdiagnosis, with concomitant increases in anxiety, cost, treatment, morbidity, and mortality.24,33,59
Based on limited existing data, the international scientific consensus recognizes a linear relationship between radiation dose and risk of radiation-induced solid organ cancers, including breast cancer, for doses above 100 mGy.60,61 The mean glandular dose from a screening mammogram is approximately 3 mGy, or the equivalent of 6 weeks of natural background radiation.62 Based on models that consider the unverified possibility (there are no documented cases) of radiation-induced breast cancer at doses of 3 mGy63- 65, lives saved by screening mammography far outweigh the theoretical possibility of lives lost from radiation exposure by a ratio of 50 to 75:1.62,66,67
Interval and Age
Annual screening is recommended. This results in more screening-detected tumors, tumors of smaller size, and fewer interval cancers than biennial screening, the latter a key determinant of survival33,68- 75. Webb et al found that interval cancers presenting in the second year after a negative mammogram caused 34% of deaths from breast cancer in screened women and 47% of breast cancer deaths in women 40 to 49 years old.33 Modeling by Wu et al shows that biennial screening misses the diagnosis of breast cancer in preclinical stages in nearly 2 of 3 women.71
Annual mammography before a breast cancer diagnosis is predictive of increased overall survival; women who had missed any of the previous five annual mammograms had a 2.3-fold increase in all-cause mortality compared with those who had missed none.76 Although annual surveillance is thought to be more important in younger women, Sanderson et al used the National Cancer Institute's Surveillance, Epidemiology, and End Results (SEER) data to show that women 69 to 84 years of age screened annually had lower 10-year breast cancer mortality than women screened biennially or sporadically regardless of race.77 CISNET models confirm that annual screening affords the best mortality reduction for all age groups.24,27
Screening should begin at age 40. Screened women in their 40s are more likely to have early-stage disease, negative lymph nodes, and smaller tumors than unscreened women.78,79 A 10-year review showed statistically significant improvement in 5-year disease-free survival (94% versus 71%) and overall survival (97% versus 78%) for screened versus unscreened women ages 40 to 49 years with breast cancer, all of whom had similar treatments [79]. For a single cohort who turned 40 years old in 2000, annual screening from 40 to 84 years would avert 29 369 breast cancer deaths, 71% more deaths averted than biennial screening limited to ages 50 to 74.27
Recommendations to delay screening until age 50 adversely affect minority women as one-third of all breast cancers in Black, Asian, and Hispanic women are diagnosed under age 50.80 The SEER program describes the different ethnic groups as non-Hispanic White (NHW), non-Hispanic Black (NHB), Hispanic, American Indian or Alaska Native, and Asian or Pacific Islander. Analysis of SEER 21 data by Hendrick et al81 shows that, among all women with breast cancer, minority women are 72% more likely to be diagnosed with invasive breast cancer under age 50 years and 58% more likely to be diagnosed with advanced-stage disease under age 50 years than NHW women. Among women dying of breast cancer, minority women are 127% more likely to die under age 50 years than NHW women (personal communication, R.E. Hendrick). Comparison of younger (15-44 years) and older (45-64 years) women using SEER 18 data found a greater proportion of younger women were diagnosed at late stages compared with older women (48.1% versus 38.7%, P < .0001) and a significant 54.5% of younger NHB women and 52.9% of younger Hispanic women were diagnosed with late-stage disease, compared with 46.0% for younger NHW women (P < .0001 for both).82
Screening should continue past age 74 without an upper age limit, unless severe comorbidities limit life expectancy or ability to accept treatment. One in five breast cancers occurs in women 75 years or older.83 Performance metrics among women ages 75 to 90 years are increasingly favorable with higher CDRs, sensitivity, and specificity and fewer false-positives compared with younger women.34 Destounis et al found a substantial CDR of 8.4 in 1000 in 18 497 women aged 75 or older, with most cancers invasive but highly treatable.84 CISNET confirms that screening women after age 74 results in benefits in terms of breast cancer deaths averted and life-years gained with no upper age limit, although both metrics steadily decline with age.85 In addition, quality-adjust life-years remain positive for screening up to age 90.27,85
Mammography-detected invasive breast cancer in women aged 75 and older is associated with significantly better 5-year disease-specific survival (97% versus 87%, P < .001).30 Even with mild to moderate comorbidities, the relative risk of death for mammography-detected tumors is 53% less for women aged 75 to 79 years and 48% less for women 80 years and older.83 Using recent SEER data, Hendrick et al showed that, among women over 40, women aged 70 to 79 years have had the steepest increase in distant-stage breast cancer rates and slowest mortality declines86, highlighting the need to continue screening this cohort.
Diversity
In addition to being diagnosed at younger ages, NHB women have 40% higher breast cancer mortality than NHW women.87 Contributing factors include higher incidence of BRCA1 and BRCA2 mutations [88,89] and twice the incidence of aggressive, triple-negative breast cancers in NHB women.90- 92 Systems-based factors contribute to the mortality disparity and could be targeted to eliminate much of the mortality difference.93,94 For example, Black and NHW women had equal breast cancer mortality rates in the 1980s but rates declined only for NHW women after widespread utilization of screening mammography in the 1990s.94 Other minority groups face worse breast cancer outcomes as well. Examining the SEER-17 database, Banegas et al found that breast-cancer-specific mortality for Hispanic Black, Hispanic White and NHB women was 10% to 50% higher than for NHW women.95
Although the disadvantages of younger age at diagnosis and tumor biology cannot be eliminated, providing all women earlier risk assessment, equal access to high-quality screening mammography beginning at age 40, and equitable treatment could reduce disparity in the breast cancer mortality. Chen et al reported that screening mammography reduced the mortality disparity when Black and White women had equitable treatment of triple-negative breast cancer.96
Individuals who identify as lesbian, gay, bisexual, transgender, or queer are less likely to present for cancer screening than non–lesbian, gay, bisexual, transgender, or queer persons.97 Facilities must work to create a respectful environment that welcomes all people.98 For average-risk transgender patients, recommendations depend on sex assigned at birth, use and duration of hormones, and surgical history and are based on limited data and expert opinion. Annual screening at age 40 is recommended for transfeminine (male-to-female) patients who have used hormones for ≥5 years, as well as for transmasculine (female-to-male) patients who have not had mastectomy.99
Risk Assessment
In 2018, the ACR issued new recommendations that all women be evaluated for breast cancer risk by age 30, so that those at higher risk can be identified and begin screening before age 40.100 Similarly, the American College of Breast Surgeons advises all women over age 25 to undergo risk assessment. There are a variety of risk assessment models available, and new methods, including deep-learning-based systems, are being tested.100- 102
Other Modalities
Evidence continues to accumulate confirming the ability of supplemental screening technologies, including MRI, whole breast ultrasound, contrast-enhanced mammography, and molecular breast imaging, to detect incremental cancers after negative mammographic screening.103,104 Most studies involve women at elevated risk. Kuhl et al showed that screening MR found incremental cancers in women at average risk (Interval Cancer Detection Rate [ICDR] = 15.5 per 1000), 40% of which occurred in women with ACR category A or B breast density.105 However, at this time, there is insufficient evidence to support supplemental screening in average-risk women.
Other Considerations
The ACR Appropriateness Criteria are evidence-based recommendations using a process that entails extensive assessment of the literature and follows well-established methodologies to rate the appropriateness of imaging for specific clinical scenarios. The ACR Appropriateness Criteria for breast cancer screening in average-risk women assigned the highest rating (“usually appropriate”) to annual screening mammography or DBT beginning at age 40 with no upper age limit.106
The ACR Practice Parameters are expert-authored policy statements regarding the safe and effective use of imaging. The goal of the parameters is to narrow variability among radiology practices and provide guidance to optimize quality. The ACR Practice Parameter for the Performance of Screening and Diagnostic Mammography recommends annual screening mammography beginning at age 40 for average-risk women with no specific upper age for cessation of screening.107
The National Comprehensive Cancer Network (NCCN) is a not-for-profit organization of 30 leading cancer centers with multidisciplinary clinical experts.108 The NCCN states that the primary goal of screening is reduction of mortality and treatment-related morbidity and, therefore, recommends annual screening mammography starting at age 40.109 Per the NCCN, upper age of screening should be based on assessment of comorbid conditions that affect life expectancy and the individual's treatment desires. These recommendations are of Category 1 Evidence and Consensus.110
NCCN recognizes that all major organizations are in agreement that maximum mortality reduction is achieved when screening begins at age 40, but differences occur because of “subjective value judgments between the benefits versus the risks.” The ACR, SBI, and NCCN believe that each woman should have the opportunity to choose the regimen with the highest associated benefit and that there should be financial coverage for her choice.
Women have a choice of whether and how often to be screened for breast cancer. Annual mammography screening starting at age 40 provides the greatest breast cancer mortality reduction, diagnosis at earlier stage, better surgical options, and more effective chemotherapy. These benefits should be considered along with the possibility of recall for additional imaging and benign biopsy and the less tangible risks of anxiety and overdiagnosis. Assessing the relative importance of these factors is a value judgement, not a scientific calculation. As stated in our previous work, weighing benefits and risks should be done by women, not for women, once accurate information is given. Women who wish to maximize the benefits will choose annual screening starting at age 40 and will not stop screening prematurely. Delaying the age to start screening will result in unnecessary loss of life to breast cancer and will disadvantage minority women in particular. All women, especially minority women, should have a risk assessment by age 30, to ensure that they are not in a higher-risk category, for which supplemental screening might be considered.100
Mammography screening has been proven effective in reducing breast cancer deaths in women age 40 years and older, with a mortality reduction of 40% possible with regular screening.
Annual mammography screening starting at age 40 provides the greatest breast cancer mortality reduction by enabling diagnosis at smaller sizes and earlier stages, better surgical options, and more effective chemotherapy.
Delaying screening until age 45 or 50 results in unnecessary loss of life to breast cancer, adversely affecting minority women in particular.
Breast cancer screening in women ages 75 years and older has continued benefits in terms of deaths averted and life-years gained.
Annual screening results in more screening-detected tumors, tumors of smaller sizes, and fewer interval cancers than longer screening intervals.
There is risk in not screening; treatment advances are important but cannot overcome the disadvantage of being diagnosed with an advanced-stage tumor.