Grand Rounds

How Effective Is Mammography in Reducing the Number of Advanced Breast Cancer Diagnoses?

BACKGROUND AND PURPOSE:

The goal of cancer screening programs is to reduce cancer mortality by decreasing the number of individuals who present for cancer care with advanced disease
In 1988, the Netherlands adopted a population based breast cancer screening program
Follow-up studies indicated deaths from breast cancer declined following 1995
- From 1989–1997, there was decline in the incidence of tumors ≥ 20 mm with positive nodes and/or metastases in women between 50-69 years of age
- Tumors <20 mm with metastases or tumors ≥ 20 mm but no metastases were unchanged
Autier et al. (BMJ, 2017) updated the above data and analyzed the change in rates of stage specific incidence of breast cancer and breast cancer mortality from 1989-2012

Population based study
- 1989-1996: Women age 50-69 offered biennial mammography
- 1997: Women age 70-75 were also offered biennial mammography
- 2006: Digital mammography replaces film based mammography
- Approximately 80% of women opted for screening, mostly between ages 50-75
Threshold diameter between stage 1 and stage 2 is 20 mm
- Early cancer: Stage 1 and in-situ
- Advanced cancer: Stage 2-4
- Survival in stage 1 is closer to women without breast cancer, while stages 2-4 are associated with reduced survival
Overdiagnosis was defined as the incidence of in situ and stage 1 breast cancer in excess to the incidence that would be expected in the absence of screening, adjusting for clinical lead time
Basic assumptions from the literature
- Out of 3 stage 1 tumors detected by screening that would have been diagnosed at stage 2-4 without, two deaths from breast cancer would be prevented (3:2 ratio)
Breast cancer mortality reductions during 2010-12 and overdiagnosis during 2009-2011 were computed without (scenario 1) and with (scenario 2) a cohort effect on mortality
- Cohort effect is independent of screening
- Investigators have identified a gradual decrease in lifetime risk of breast cancer death based on birth cohort and adjusted for this in scenario 2

In 1989, the incidence of stage 2-4 breast cancer in women was 168/100,000 and decreased negligibly to 166/100,000 by 2012
The incidence of in situ and stage 1 cancers increased
- For ages 50-69, there was sharp increases in incidence in 1990-1996, with further increases to 2007, and then another sharp increase with the introduction of digital mammography
Screening in women ≥50 years was associated with
- 5% mortality reduction in scenario 1
- No change in scenario 2 (adjusted for cohort effects)
  - Cohort effect independently reduced mortality by 5%
- Regardless of cohort effect, improved treatments would result in a 28% breast cancer mortality reduction
- Timing and magnitude of decreases in mortality were similar among women <50 years (not in the screening program) and those age 50-69

Authors acknowledge limitations
- Due to the high rate of screening uptake, they do not have a direct, contemporaneous cohort without screening to serve as controls for comparison
The results of this study demonstrates that the Dutch mammography screening program has had little impact on the burden of advanced breast cancers
- Therefore, one can assume only a marginal effect on breast cancer mortality
Overdiagnosis has steadily increased over time and increasing with opening up the program to older women and using advanced digital technologies
As of 2012, about half of screen detected breast cancers represent overdiagnosis
The authors calculate that for one woman who would not die from breast cancer, 14 women would be overdiagnosed with an in situ or stage 1 cancer
- Using scenario 2 and correcting for cohort effect, there would be no additional early detection of cancers that would be stage 2-4 without the screening program

< Previous

Next >