Do mammograms reduce breast cancer mortality? The debate continues
Do mammograms reduce breast cancer mortality? The debate continues
13/2/2014
In a 25-year follow-up report from a Canadian screening program, annual mammography did not reduce breast cancer mortality in women ages 40-59. This sparked further an ongoing debate on the common practice and its effects.
In particular, women screened annually by mammography for 5 years had had a breast cancer mortality hazard of 1.05 compared with the control group during the screening period. During follow-up for a mean of 22 years, the mammography group had a breast cancer mortality hazard of 0.99 versus the control group. Neither value was statistically significant.
After 15 years of follow-up, the mammography group had an excess of 106 breast cancers attributable to overdiagnosis.
"Although the difference in survival after a diagnosis of breast cancer was significant between those cancers diagnosed by mammography alone and those diagnosed by physical examination screening, this is due to lead time, length of time bias, and overdiagnosis," Anthony B. Miller, MD, of the University of Toronto School of Public Health, and colleagues said of their findings.
"At the end of the screening period, an excess of 142 breast cancers occurred in the mammography arm compared with the control arm, and at 15 years, the excess remained at 106 cancers. This implies that 22% (106 of 484) of the screen-detected cancers in the mammography arm were overdiagnosed."
The findings suggest a need to reassess the value of screening mammography, they added.
Quick Retort
The publication drew a quick and forceful response from the American College of Radiology (ACR) and the Society of Breast Imaging (SBI). In a joint statement, officials of the two organizations characterized the results as "an incredibly misleading analysis based on the deeply flawed and widely discredited Canadian National Breast Screening Study (CNBSS)."
Noting the 32% rate of cancer detection by mammography, the ACR and SBI said "this extremely low number is consistent with poor-quality mammography." Mammography alone should detect twice that many cancers, they added. The organizations noted that a prior outside review of the CNBSS confirmed the poor quality of mammography in the study.
The ACR/SBI statement also questioned the randomization process used by CNBSS, pointing out that all women had a clinical breast examination prior to allocation, providing investigators with advance knowledge about which patients had breast lumps or enlarged lymph nodes.
The findings should come as no surprise, said Richard C. Wender, MD, of the American Cancer Society (ACS). The CNBSS has been an "outlier" from the initial report, and the lack of benefit with mammography wouldn't be expected to change with additional follow-up.
Miller and colleagues reported 25-year follow-up data from the CNBSS, which began in 1980. All women, ages 50 to 59, had annual clinical breast examinations, as did women 40 to 49 in the mammography arm. Younger women in the control arm had a clinical breast exam at enrollment, followed by usual care.
The study included 89,835 women enrolled at 15 centers in six Canadian provinces. During the 5-year mammographic-screening period, 666 invasive cancers were diagnosed in the mammography arm and 524 in the control group. Additionally, 180 women randomized to mammography died of breast cancer, as did 171 in the control group.
The hazard ratio (HR) for breast cancer-specific mortality during the screening period was 1.05 for mammography versus control.
During the entire study, 3,250 women in the mammography group developed breast cancer compared with 3,133 in the control group, and 500 breast cancer deaths occurred in the mammographically screened patients versus 505 in the control group, resulting in an HR of 0.99.
The authors of an accompanying editorial noted that some evidence suggests that improved treatment, rather than breast cancer screening, has fueled the decline in breast cancer mortality in recent years. Regardless of the rates found in different studies, overdiagnosis represents a larger problem.
"The real amount of diagnosis in current screening programs might be even higher than that reported in the Canadian study, because ductal carcinoma in situ, which accounts for one in four breast cancers detected in screening programs, was not included in the analysis," said Mette Kalager, MD, of the University of Oslo in Norway, and colleagues.
"We agree with Miller and colleagues that 'the rationale for screening by mammography be urgently reassessed by policymakers,'" Kalager and co-authors added.
Retort to the Retort
In a reply to the ACR/SBI comments, Miller called the statement "deeply flawed and misleading, as it ignores all our previous clarifications on the issues raised."
To the assertion of poor mammographic quality, Miller said the study demonstrated an excellent cancer detection rate, participating centers used modern (for the time) imaging machines, and technologists were trained in accordance with prevailing standards for North America.
"The randomization in the trial was assessed by two internationally recognized epidemiologists and was deemed to comply fully with accepted standards," Miller said by email. "Other commentators have agreed and regarded our study as high quality."
In planning the trial, investigators overestimated the number of breast cancer deaths that would occur, a reflection of the care the patients received.
"We had to postpone our first definitive analysis for 2 years so that sufficient events would accumulate," Miller continued. "However, our long-term follow-up has resulted in sufficient events so that we can state with a high degree of confidence that mammography screening does not result in a reduction of breast cancer mortality."
Referring directly to the ACR and SBI, Miller said the study's outcome "has to be unwelcome to this highly financially conflicted group, but which will be of substantial interest to policy makers in considering the future of screening for breast cancer."
Not the Last Word
Acknowledging the strong emotions underlying viewpoints on mammography, Wender said the issues will continue to be debated. ACS plans to review its mammography recommendations this year, and the United States Preventive Services Task Force (USPSTF) has indicated that a panel may revisit the organization's recommendations later this year.
"My own sense is that we did not learn a lot from the Canadian study," Wender told MedPage Today. "These numbers have been out for a long time. Maybe the one new thing we learned is that if you go out 25 years, the number of cases in the screening and control groups were pretty similar."
A key issue often gets overlooked or lost in the debate over statistics and methodology.
"There is unanimity of opinion among the major guidelines groups that attempt to synthesize all of the data," Wender said. "That is, mammography reduces age-adjusted mortality rates. It does avert/prevent premature breast cancer death -- not just in women over 50 but in women in their 40s."
"Ultimately, the guideline groups are charged with balancing the benefits and risks of screening, and there is no perfect way to do that," he added.
As a final word on the matter (for now) Wender said the ACS and USPSTF recommendations (arguably the best known in the U.S.) are more similar than different. "I think we all look at the differences."
Source: MedPage Today: http://www.medpagetoday.com/HematologyOncology/BreastCancer/44263