Participation in Mammography Screening Accomplishes an Enduring and Significant Reduction in Breast Cancer Mortality in the Era of Modern Therapy: An Observational Study Spanning 52 Years

Abstract

Background: A significant decrease in breast cancer mortality has been demonstrated in populations invited to mammography screening. There have been questions regarding the value of mammography in the era of modern therapies, and the relative contribution of mammography screening and modern therapy on mortality reduction. Aim: We have sought to clarify the extent to which early detection through mammography screening contributes to the reduction of breast cancer death in the era of modern adjuvant therapies. We address this question by studying five decades of continuous data from a stable population spanning the pre- and postscreening and adjuvant therapy periods. Methods: We analyzed changes in breast cancer mortality in a stable population in Dalarna County, Sweden, among women aged 40-69 years during the 52 years from 1964 through 2015. Mortality data were obtained from the National Cause of Death Register in Sweden, and incidence data from the Swedish National Cancer Registry. Crude and incidence-based mortality rates were compared among four successive 13-year periods: the prescreening period from 1964 through 1976, the Swedish Two-County randomized screening trial period from 1977 through 1989, and two service screening periods from 1990-2002 and from 2003-2015. Furthermore, we measured the effect of early detection upon breast cancer mortality in women exposed to mammography screening by comparison with breast cancer death in women not exposed to mammography screening, within these three screening periods. Data were analyzed by Poisson regression, with corrections for lead time and self-selection bias. Results: There were 5844 incident cases and 1425 breast cancer deaths during the 52 years of observation. The relative breast cancer mortality rates associated with exposure to screening, adjusted for self-selection bias, were 0.46 (95% CI 0.30-0.69) in the trial period (1977-1989), 0.44 (95% CI 0.30-0.65) in the 1990-2002 period, and 0.37 (95% CI 0.24-0.56) in the 2003-2015 period. The significant reductions in incidence-based breast cancer mortality associated with exposure to screening were independent of contemporaneous changes in therapy. Conclusion: The combination of early detection of breast cancer through mammography screening and the resultant earlier treatment has significantly reduced breast cancer mortality in Dalarna County in the women exposed to screening, compared with the women not participating in screening, by a factor of 2.2 in the screening trial period, increasing to a factor of 2.7 in the most recent service screening period. These mortality benefits can be attributed to the far greater effectiveness of modern therapeutic methods upon cancers detected at screening compared with the poorer effectiveness of the same therapeutic methods in women not participating in screening.