Organized Approach to Screening Mammography

Overview

This chapter will review the science behind screening mammography including a summary of different screening guidelines and discussion of controversies surrounding screening recommendations. The second part of this chapter will review how to systematically approach screening mammography interpretation.

Why Perform Screening Mammography? (Screening Evidence)

Historical Perspective

Before there was mammography or any breast imaging, breast cancers presented when they became clinically apparent, for example as a large palpable mass. Diagnosis was based on surgical excision, and many benign lesions were excised. With screening mammography, breast cancers are detected at an earlier stage before they are clinically apparent resulting in improved survival, less invasive therapy, and lower morbidity.

Screening Mammography Reduces Breast Cancer Mortality

Screening mammography has been proven to reduce breast cancer mortality in multiple randomized controlled trials. These trials included women aged 40 to 69, with an overall mortality benefit of approximately 20% across all trials as shown in Fig. 12.1 . Randomized controlled trial data is considered the strongest screening evidence to prove benefit of screening mammography but will necessarily underestimate the magnitude of benefit of screening mammography due to noncompliance and contamination. First, it is important to understand that the randomized controlled trials randomized invitation to screening, not the screening examination itself. In other words, half the study participants were sent a letter inviting the woman to attend screening while the other half did not receive an invitation letter.

Fig. 12.1, Summary of eight randomized controlled trials in women aged 40–69. A relative-risk of 1 means there is no benefit from screening mammography. The combined results of these randomized controlled trials show a relative risk of 0.8 indicating an overall 20% mortality benefit of screening mammography. The NBSS (Canadian) screening trials were outliers as the only trials to not show a benefit from screening mammography, and many critics have called for these trials to be discounted citing problems with randomization (palpable lumps and disproportionate number of advanced cancers in the screening arm) and substandard mammography image quality (direct x-ray instead of screen-film mammography) for 4 of 5 years of the trial. HIP, Health Insurance Plan; NBSS, National Breast Screening Study; RR, relative risk.

Noncompliance occurs when a woman ignores the invitation letter and does not get screened. If this woman unfortunately later dies of breast cancer, she is still counted in the screening arm.

Contamination occurs when a woman who did not receive an invitation letter seeks screening mammography on her own. If the woman’s life is subsequently saved through early detection of a breast cancer, she is still counted in the control or no screening arm.

Thus while randomized controlled trials rigorously prove the life-saving benefit of screening mammography, they necessarily underestimate the magnitude of mortality benefit due to noncompliance and contamination ( Box 12.1 ).

Box 12.1

Randomized Controlled Trials Underestimate Screening Benefit Due to Contamination and Noncompliance

■
Trials randomized invitation to screen versus no invitation to screen (not the same as randomizing who gets mammography)
■
Contamination: Person in “no invitation” group gets screened and life is saved by early detection of breast cancer.
■
Noncompliance: Person in “invitation to screen” group does not get screened and dies of breast cancer.

Service screening experience has demonstrated an approximately 40% breast cancer mortality reduction with screening mammography. As an example, the Canadian service screening experience is summarized in Fig. 12.2 . Service screening experience provides a more accurate estimate of the magnitude of benefit of screening mammography because it reflects real-world experience, includes larger patient populations, and includes modern screening technology such as digital mammography ( Box 12.2 ). Note that the randomized controlled trials did not include any digital mammography, only (now obsolete) analog mammography. Fig. 12.3 illustrates the evolution of mammography technology.

Fig. 12.2, Summary of service screening experience in Canada showing overall 40% mortality reduction among women screened with mammography. Service screening reflects real-world experience using screening mammography, includes a larger number of patients, and includes current technology such as digital mammography. CI, confidence interval; SMR, standardized mortality ratio

Box 12.2

Service Screening Experience (Observational Studies)

■
Reflect actual clinical practice and more current practice
■
Include larger populations
■
More accurately reflect the magnitude of benefit of screening mammography

Fig. 12.3, Craniocaudal views of extremely dense breasts illustrating the evolution of two-dimensional mammography technology. Note that the randomized controlled screening trials only included analog technology ( circled ).

Screening Guidelines

While organizations agree that screening mammography saves lives, they disagree on what ages to screen and frequency of screening. Hence, there are currently three major guidelines for screening average risk women. Please refer to Chapter 10: Breast Cancer Risk Assessment and Chapter 11: Breast Cancer Screening Guidelines and Supplemental Screening for discussion of screening for patients at higher than average risk and additional discussion of screening guidelines. The American College of Radiology (ACR) recommends annual screening starting at age 40 in order to save the most lives. These recommendations are supported by the National Comprehensive Cancer Network, the Society of Breast Imaging, and other organizations placing emphasis on early detection of breast cancer. The U.S. Preventive Services Task Force (USPSTF) guidelines recommend biennial screening starting at age 50 and ending at age 74, placing emphasis on reducing benign-outcome recalls and biopsies and with significantly lower mortality benefit (23% for biennial screening compared with nearly 40% mortality benefit for annual screening starting at age 40). The American Cancer Society guidelines are a hybrid of annual and biennial screening recommending annual screening between the ages of 45 and 54 and then biennial screening from 55 to 79. These guidelines result in a mortality benefit in between that of the ACR and the USPSTF.

Table 12.1 compares these different screening strategies in terms of number of mammograms, lives saved, and life years gained. This comparison of differing screening strategies comes from the Cancer Intervention and Surveillance Modeling Network (CISNET) models. The CISNET models used the original randomized controlled trial data to model outcomes for different screening strategies. CISNET models were considered by the USPSTF in the development of their 2009 and 2016 (current) screening recommendations.

Table 12.1

Comparison of Three Screening Strategies for Average-Risk Women

Arleo EK, Hendrick RE, Helvie MA, Sickles EA. Comparison of recommendations for screening mammography using CISNET models. Cancer Oct 1;123(19):3673–3680, 2017.

Screening Strategy	No. of Examinations ^a	Pecent Mortality Reduction	Breast Cancer Deaths Averted ^a	LYG ^a	NNS per Death Averted	NNS per LYG
A40–84	36,500	39.6%	11.9	189	84	5.3
H45–79	19,846	30.8%	9.25	149	108	6.7
B50–74	11,066	23.2%	6.95	110	144	9.1

The most lives are saved and the most life years are gained with annual screening from 40 to 84.

A40–84 , American College of Radiology recommendation for annual screening from 40 to 84; H45–79 , American Cancer Society hybrid strategy of annual screening from 45 to 54, then biennial screening from 55 to 79; B50–74 , U.S. Preventive Services Task Force recommendation for biennial screening from 50 to 74. All organizations agree the most lives are saved with annual screening. All screening guidelines allow screening starting at age 40; LYG , life years gained; NNS , number needed to screen.

a Per 1000 women screened. Based on mean values of six 2009 CISNET models.

All organizations agree the most lives are saved with annual screening starting at age 40. The American Cancer Society guidelines note that the majority of American women would choose to screen annually at age 40. The decision to screen and what guideline to follow is ultimately a woman’s choice based on shared decision making. All guidelines allow women to screen annually starting at age 40.

Screening Controversies: Commonly Cited Screening “Harms” (Risks)

Much of the controversy surrounding when to start screening and at what interval is due to cited “harms” or risks of screening. Although the word harm is often used, this term is pejorative, meaning that it implies more substantial adverse outcomes than most women would think. A more appropriate term to use is risk , analogous to the terminology for adverse outcomes used during the informed consent process. For example, when consenting for a breast biopsy we discuss the risks versus benefits of a procedure with the patient, not the harms versus benefits.

The four most commonly cited risks (“harms”) of mammography screening are radiation, benign-outcome recalls and biopsies, patient anxiety, and overdiagnosis. As will be discussed in the subsequent sections, none of these risks are substantial.

Radiation

We are all exposed to natural background radiation. Our annual background dose is approximately 3 mSv per year at sea level. People living at higher elevations are exposed to slightly more background radiation. For a standard two-view screening mammogram, the radiation dose is equivalent to approximately 2 months of natural background radiation at sea level. Hence, radiation risk from mammography is classified in the “minimal” risk range, and it is generally accepted that the potential benefits of screening mammography far outweigh any potential risk of dying of breast cancer due to radiation exposure from mammography ( Box 12.3 ).

Box 12.3

Mammography Radiation Risk

■
Equivalent to ~2 months natural background radiation for bilateral two-view mammogram
■
Benefits of screening far outweigh negligible radiation risks

Benign-Outcome Recalls and Biopsies (False Positives)

An abnormal screening mammogram (Breast Imaging Reporting and Data System [BI-RADS] 0: incomplete) that is ultimately resolved as negative (BI-RADS 1) or benign (BI-RADS 2) at diagnostic imaging is considered a false positive. A benign biopsy is another example of a false positive. In the United States, out of 100 women screened, approximately 10 women will be recalled from screening for additional diagnostic evaluation consisting of additional mammographic views and/or breast ultrasound. Often, the diagnostic evaluation provides reassurance to the patient, for example in the case of an asymmetry resolved as superimposition ( Fig. 12.4 ), and the patient’s anxiety is relieved (see Chapter 1 ).

Fig. 12.4, Example of false positive at screening. Asymmetry seen on screening mammography examination, visible only on the mediolateral oblique (MLO) view ( circle ) and not on the craniocaudal view (not shown). Examination was interpreted as BI-RADS 0: Incomplete with a recommendation for additional imaging evaluation. Repeat MLO view at time of diagnostic examination resolves the asymmetry as superimposition of normal breast structures ( dotted circle ). Final assessment of BI-RADS 1: Negative with recommendation to return to annual screening.

One to two patients out of the original 100 women screened may be recommended to undergo a breast biopsy. Breast biopsies are minimally invasive, well-tolerated outpatient procedures with few contraindications. See Chapter 7 for further discussion of breast interventions. Breast biopsy is recommended for BI-RADS 4 (suspicious) and BI-RADS 5 (highly suggestive of malignancy lesions).

The majority of breast biopsies are benign and considered false-positive biopsies. But it is important to recognize that about one-quarter to one-third of breast biopsies are true positives, meaning cancer is diagnosed.

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here