Studies Provide New Data on Mammography and Breast Cancer Mortality
Mammography can reduce mortality from breast cancer, but the magnitude of benefit may differ across age groups and as treatment improves.
Two Scandinavian studies present new information about the extent to which routine mammography in different age groups can reduce the risk of death from breast cancer. However, several researchers cautioned that, because both studies were observational, they have limitations that can affect how the findings are interpreted and are unlikely to resolve the controversy over the extent of mammography’s benefit in reducing breast cancer mortality.
The studies examined the effect of routine mammography in regions of Sweden and Norway where breast cancer screening programs were rolled out over time, county by county. Whereas the Swedish study focused strictly on routine mammography’s impact on breast cancer deaths among women age 40 to 49, the Norwegian study analyzed mammography’s mortality benefit in an older age group and took into account the broader context of changes in breast cancer awareness and improvements in treatment over time. (See the box at the bottom of the page.) The studies came to different conclusions, with the Swedish study finding that routine mammography screening greatly reduced breast cancer mortality in young women and the Norwegian study finding a much smaller reduction in women age 50 to 69.
Although the two studies looked at different age groups and had disparate results, “they both show that fewer women are dying of breast cancer in the presence of screening,” said Dr. Stephen Taplin of NCI’s Division of Cancer Control and Population Sciences. And although it’s been known that mammography can reduce mortality, he continued, “what has been changing is the recognition that the magnitude of that benefit may differ across age groups and that treatment has been making gains.”
Looking at a Younger Population
The Swedish study, dubbed SCRY, came to the conclusion that there was a 29-percent reduction in breast cancer deaths in women age 40 to 49 who underwent screening compared with women who were not invited to undergo screening. Approximately 1,250 women had to undergo screening to prevent or delay one death from breast cancer, reported the study’s senior author, Dr. Hakan Jonsson, at the American Society of Clinical Oncology’s 2010 Breast Cancer Symposium in Washington, DC. The study results also appeared online September 29 in Cancer.
“This report clearly shows a benefit to screening mammograms in this age group,” wrote American Cancer Society Deputy Chief Medical Officer Dr. Len Lichtenfeld, on his blog. But he also acknowledged that there would “be considerable discussion among the experts about what is right with this study and what may be questionable, what questions it answers, and what questions remain open.”
According to Dr. Barry Kramer, associate director for disease prevention at NIH, the way in which the data were analyzed could have led to a substantial overestimation of mammography’s mortality benefit. To calculate the excess number of breast cancer deaths among women in counties without routine screening versus those in counties with it, he noted, the researchers compared only those breast cancer deaths that occurred in women who had been diagnosed with breast cancer (typically via mammography) rather than comparing the overall breast cancer mortality rates in each county. Because mammography can detect nonlethal cancers, “their analysis would not take into account the overdiagnosis associated with screening,” Dr. Kramer explained. “That’s not a subtle issue. It’s an important bias that would favor screening.”
Other methodological issues, Dr. Kramer continued, such as not adjusting the analysis for differences in breast cancer deaths in the counties before routine screening was implemented and socioeconomic and county-level resource differences that likely influenced whether screening was offered, could also have inflated the mortality benefit associated with mammography.
Whether women between the age of 40 and 49 should undergo routine mammography has been a matter of much debate and discussion. In November 2009, the U.S. Preventive Services Task Force (USPSTF) issued updated guidelines on breast cancer screening that recommended that women ages 50 to 74 get a mammogram every 2 years. But the task force did not endorse routine mammography for women in the 40 to 49 age group who have average breast cancer risk, noting that the absolute risk reduction is less in younger than older women and that false-positive results are a greater concern in younger women. Instead, the decision about mammography for these women “should be an individual one and take patient context into account,” the task force recommended. Some cancer organizations, including the American Cancer Society and the National Comprehensive Cancer Network, continue to recommend annual mammography for women age 40 to 49.
Many women in this age group “want unambiguous recommendations regarding whether to undergo mammography,” said Dr. Jennifer Obel from NorthShore University HealthSystem in Illinois during a press briefing on the study results. “I think the critical message is that all women, beginning at age 40, should speak with their doctors about mammography to try to understand the potential benefits and risks of the test and to determine what is best for them as individuals.”
Closer Scrutiny of an Older Population
Meanwhile, Dr. Mette Kalager and her colleagues from Oslo University Hospital in Norway found that routine mammography in women ages 50 to 69 reduced breast cancer deaths, but only modestly. The findings appeared online September 23 in the New England Journal of Medicine.
Based largely on clinical trials conducted in the late 1980s and early 1990s, the mortality reduction associated with routine mammography in this age group has been estimated to be as high as 25 percent. But the Norwegian study found that implementation of a screening program was associated with a 10-percent reduction in breast cancer mortality. Breast cancer mortality was reduced by another 18 percent over the same time period, even in counties that had not implemented the screening program. This component of the reduction was likely due to greater awareness of breast cancer and improvements in treatment, such as adjuvant therapy, the authors concluded.
Dr. Taplin agreed that treatment likely played a role in the mortality improvements. As part of the screening program in Norway, dedicated interdisciplinary breast cancer management teams were also established in each county. “The fact that organized support may have contributed to mortality reduction also suggests that it is needed, so we need to be testing these approaches here [in the United States],” Dr. Taplin said.
The Norwegian study avoided some of the methodological issues seen in the Swedish study, Dr. Kramer explained. Dr. Kalager’s team used population-based breast cancer death rates and compared mortality in each county before and after the implementation of routine mammography; these features “add to the strength and plausibility of the data,” Dr. Kramer said.
The Norwegian findings suggest that it “is quite plausible that screening mammography was more effective in the past than it is now,” wrote Dr. H. Gilbert Welch of Dartmouth Medical School in an accompanying editorial. “If women with new breast lumps now present earlier for evaluation, the benefit of screening will be less. If treatment of clinically detected breast cancer (i.e., tumors that are detected by means other than screening) has now improved, the benefit of screening will be less.”
The average follow-up after diagnosis was 2.2 years in the Norwegian study, which makes it difficult to draw definitive conclusions from it, Dr. Taplin cautioned. Longer follow-up is needed to get a clearer picture of screening’s impact on mortality, he noted.
As part of its ongoing efforts to evaluate mammography, Dr. Taplin explained, NCI is developing a new initiative called Population-based Research Optimizing Screening through Personalized Regimens (PROSPR) to evaluate how to improve the screening processes for breast cancer (as well as colon and cervical cancer) in the U.S. community practice setting. “While we are saving more lives, we are losing too many,” he said. “So we must continue to work harder to improve screening techniques, treatment approaches, and the coordination of care.”
Finer Details of the Studies
In 1996, Norway began to roll out, county by county, a breast cancer screening program for women age 50 to 69. Women could receive a mammogram every 2 years and multidisciplinary breast cancer management teams were established in each county before routine screenings were offered.
More than 40,000 women diagnosed with breast cancer between 1986 and 2005 were included in the Norwegian screening study. There were four groups of women: those in counties with established screening programs between 1996 and 2005, those in counties awaiting screening programs between 1996 and 2005, and those in two historical control groups living in the same counties between 1986 and 1995. The comparison of the two “current” groups allowed the researchers to avoid confounding by factors that can change over time, such as treatment and awareness that might be associated with a reduction in breast cancer mortality. The use of the historical groups allowed the researchers to adjust for differences between the counties in factors such as rates of death from breast cancer and to achieve equal follow-up time in each county. The average follow-up duration was slightly more than 2 years after diagnosis, and the longest was nearly 9 years.
Compared with women in the corresponding historical control group, women in the current screening group had a 28-percent relative reduction in breast cancer mortality (breast cancer mortality was 25.3 per 100,000 person-years in the historical control but 18.1 per 100,000 person-years in the screened group, a difference of 7.2 deaths per 100,000 person-years). For women in the current unscreened group, however, breast cancer mortality was also reduced substantially compared with that in the historical control group; in this case, there was an 18-percent relative reduction in mortality (26.0 versus 21.2 deaths per 100,000 person-years, a difference of 4.8 deaths per 100,000 person-years).
As a result, the authors concluded, only “the overall between-group difference [of 2.4 deaths per 100,000 person-years] can be attributed to the screening program alone, representing a third of the total estimated reduction in mortality (2.4 of 7.2).” In other words, the screening program resulted in an absolute 10-percent decline in breast cancer mortality.
The Swedish study was different in that it directly compared breast cancer deaths among women diagnosed with breast cancer at age 40 to 49 in counties that offered routine mammography (the study group) with those in counties that didn’t offer routine mammography (the control group). Beginning in 1986, invitations for routine mammography were mandated for women ages 50 to 69 but were voluntary for the younger age group.
Over the course of the Swedish study, 1986 to 2005, data from more than 1 million women were analyzed. Average follow-up was more than 14 years. There were 803 breast cancer deaths among diagnosed women in the study group (during 7.3 million person-years) and 1,238 such deaths in the control group (8.8 million person-years). There was a 29-percent reduction in breast cancer deaths among women who actually underwent screening and a 26-percent reduction in breast cancer mortality when the analysis was restricted to those women who were invited to receive screening.