An analysis of data from nearly 6 million screening mammograms found no evidence for a clear cutoff age to stop breast cancer screening. Screening mammography among women aged 75 years was associated with higher cancer detection and lower recall rates than among younger women in the study. These findings add support for guidelines that encourage breast cancer screening decisions based on individual patient values, comorbidities, and overall health status, according to Cindy S. Lee, MD, the lead author of the study and Assistant Professor of Radiology at the University of California, San Francisco.
Dr. Lee presented the findings at the 2016 Radiological Society of North America (RSNA) Annual Meeting in Chicago.1 The study received strong positive feedback from her professional colleagues, Dr. Lee said in an interview with The ASCO Post. “We just haven’t had large current data like this,” she noted, and physicians’ advice to older women about screening mammography has been based on expert opinion and small studies, not entirely evidence. “My breast cancer colleagues in medical oncology and also surgery and radiology were excited to finally see some data, so that we can look at whether it is reasonable to screen older people.”
Evidence had been lacking because “all prior randomized, controlled trials excluded women older than 75, limiting data to small observational studies,” Dr. Lee said. When previous mammography trials were conducted years ago, “women were dying earlier,” she noted, and their limited life expectancy and overall health status may have precluded them from clinical trials. “But now women’s life expectancy is pushing into the 80s, and some people live until they are 100. Moreover, there have been many technologic advances in screening mammography and breast cancer treatment in the past decade, with survival and morbidity benefits. So there are good reasons for new studies to look at older people.”
Higher Cancer Detection Rates
Dr. Lee and her research team analyzed data for over 5.6 million screening mammograms performed in over 2.5 million women aged 40 and older between January 2008 and December 2014 at 150 facilities in the National Mammogram Database. Data were sorted by 5-year age intervals, starting with ages 40 to 44 and continuing to past 90. The percentage of patients in each age group was substantially lower for women over 70 years old.
Four standard performance metrics were calculated to evaluate screening mammography for each age group: cancer detection rate, percentage of cancers found among cases recommended for biopsy (PPV2), percentage of cancers found among biopsied cases (PPV3), and recall rate. For each of these four metrics, screening mammography among women aged 75 and older outperformed screening mammography among younger women.
The overall mean cancer detection rate was 3.74 cancers per 1,000 exams, but there was a significant upward trend. There were also significant upward trends with age for the positive predictive values reflecting the percentage of cancers found among cases recommended for biopsy and among cases where biopsy was performed.
For these positive predictive values, “ideally you want a number as high as possible, so you only perform biopsies on women who need them, not everyone,” Dr. Lee said. “If you look at the trends, the positive predictive values are also going up with increasing age, meaning we have fewer false-positives, and we are biopsying more people who actually need it.”
Lower Recall Rates
The only downward trend with age was for recall rate, where lower was better. Looking at the proportion of women who were called back for additional imaging or testing, your chance of getting called back is half as much if you are older than age 75. The overall recall rate was 10% in the United States.
Dr. Lee attributed the lower recall rate among older women to two factors—(1) having more prior mammograms for comparison is known to decrease the recall rate; and (2) older women tend to have more fatty and less dense breasts, making cancers more obvious to see. With this reduced breast density and a greater number of prior exams, “the radiologists do better and call back fewer false-positives,” Dr. Lee noted.
“We want lower recall rates and higher cancer detection in screening, and this seems to work out perfectly with older women, because breast cancers tend to happen more in older people. Most breast cancers (8 out of 10) occur in women who are older than age 50. As you get older, your risk for breast cancer is much higher, so it makes sense that we are finding more cancer in the older age groups,” Dr. Lee said. “The ideal screening test should have the higher cancer detection rate, the higher positive predictive values, and the lower recall rate, which is what we see in older people.”
‘Are We Saving Lives?’Error loading Partial View script (file: ~/Views/MacroPartials/TAP Article Portrait Widget.cshtml)
A CNN news report on the study included comments from American Cancer Society Chief Medical Officer Otis Brawley, MD. While he believes early detection of breast cancer among older women can prevent deaths, Dr. Brawley said the study does not address that issue, and “more research is needed to provide evidence to support that conclusion.”2
“That is obviously the key—are we saving lives?” Dr. Lee agreed. “That is the next part of this study. We are working on linking our data with mortality data.”
Dr. Lee explained that while the National Mammogram Database is a large nationwide database with 218 active facilities in 39 states, “we only collect information about diagnosis of breast cancer. We don’t know how the patients are treated, and we don’t know whether the treatment worked,” Dr. Lee said. “To understand the effect of screening on mortality and morbidity, we need more time, and the National Mammogram Database started in 2008, so in the future, we hope to collect data on mortality from a different source and then see if we can merge the data sets to figure out what happened to all the breast cancer patients.” That postdiagnosis follow-up information, she said, might come from Surveillance, Epidemiology, and End Results (SEER) data and/or from state tumor registries.
Impact on Guidelines
Dr. Lee said that the study adds support for guidelines that encourage screening decisions based on individual patients and their health status. The American Cancer Society guidelines say, “Screening should continue as long as a woman is in good health and is expected to live 10 more years or longer.”3 The Society of Breast Imaging and American College of Radiology recommendations for imaging screening for breast cancer say annual screening with mammography should stop “when life expectancy is < 5 to 7 years on the basis of age or comorbid conditions” and “when abnormal results of screening would not be acted on because of age or comorbid conditions.”4
Dr. Lee stressed, however, that she and her coauthors are not developing or proposing new guidelines. “This is a health services research study. This is to provide evidence and facilitate the discussion between patients and doctors about screening mamography,” she noted.
The study has not yet been published, Dr. Lee said, but “we hope to publish this work very soon, so that people can actually reference it” and look at the methodology and results. She hoped that the study would be included by the U.S. Preventive Services Task Force (USPSTF) in the next update of its breast cancer screening guidelines, because it addresses a knowledge gap. In the current guidelines, “the USPSTF concludes that the current evidence is insufficient to assess the balance of benefits and harms of screening mammography in women 75 years or older.”5
About the Database
Part of the American College of Radiology (ACR)’s National Radiology Data Registry, the National Mammography Database contains more than 11 million mammograms. Most of the exams are digital, reflecting the current screening practice in the United States. The database “leverages data that radiology practices are already collecting under federal mandate by providing them with comparative information for national and regional benchmarking,” according to the National Radiology Data Registry website. “Participants receive semiannual feedback reports that include important benchmark data such as true-positive rates, positive predictive value rates, and recall rates.”
“We wanted to provide a quality improvement tool to empower radiologists to audit their performance at interpreting screening mammograms,” Dr. Lee explained. “The feedback report compares the facility and physician to others in the same geographic region, practice type, practice setting, and annual exam volume. It is completely anonymous and de-identified.6 So you can see if you are outperforming others or among the outliers.” The quarterly reports allow comparisons of facilities across the nation and within specific geographic regions, as well as by type of facility, such as community or metropolitan facilities, or facilities with similar patient volumes. “This is about the continuation of improving practice and patient care nationwide.”
Additional information about the National Mammography Database and how to join is available at nrdr.acr.org. There is a fee to participate, but currently, the fee is waived for any ACR-accredited Breast Imaging Center of Excellence. ■
Disclosure: Dr. Lee reported no potential conflicts of interest.
1. Lee C, Sengupta D, Chatfield M, et al: Current era screening mammography outcomes from the National Mammography Database, involving nearly 6 million examinations. 2016 Radiological Society of North America Annual Meeting, Presented November 28, 2016.
4. Lee CH, Dershaw DD, Kopans D, et al: Breast cancer screening with imaging: Recommendations from the Society of Breast Imaging and the ACR on the use of mammography, breast MRI, breast ultrasound, and other technologies for the detection of clinically occult breast cancer. J Am Coll Radiol 7:18-27, 2010.
In the discussion between patient and doctor, the focus here is patient-centered care. It shouldn’t be about this ‘magic number’ of 75, after which you are no longer allowed to participate in screening.— Cindy S. Lee, MD
Tweet this quote
A study finding that there is no clear cutoff ...