Long-Term Effect of Breast Symptoms Reported at Mammography Screening Visit on Cancer Incidence and Mortality

Abstract

Background: Efforts to reduce mortality through early detection and diagnosis has intensified in the recent decade. An important risk factor, 'breast symptoms' reported by women during screening visit, is still overlooked. Aim: To study the association between breast symptoms reported at screening visit and the risk of cancer incidence and mortality in a prospective manner over a period of 24-years. Methods: This population based matched cohort study was based on the follow-up of the ongoing Finnish National Breast Cancer Screening Program (FNBCSP) that began in 1987. Symptomatic subjects who attended screening with symptoms (lump, 39,965 visits; retraction, 24,190 visits; nipple discharge, 7882 visits) were identified from the Finnish Cancer Registry database. For each visit with symptoms, nonsymptomatic controls were matched (1:1 for lump and retraction; 1:2 for nipple discharge) based on age at screening visit (within 2 years), year of invitation (2 years band), number of invited visits, and municipality of invitation. The primary outcomes were incidence of breast cancer and incidence-based mortality, including all-cause mortality. Results: Women who reported lump or retraction had about twofold risk of breast cancer incidence, threefold risk of breast cancer mortality and all-cause mortality respectively as compared with women without respective symptoms. We found a substantial difference in mortality rates throughout the follow-up period between symptomatic and asymptomatic group. In absolute terms, for lump, in every 1000 screening visits, 20 women died of breast cancer as compared with 7 women without lump, and 30 vs 11 all-cause deaths in women with and without lump, respectively after 24 years of follow-up. We also found difference in the number of deaths in women who reported retraction or nipple discharge. Conclusion: This study provides comprehensive evidence that women with breast symptoms remain in a higher risk of dying over a very long period. Guidelines to reduce these inequalities needs to be developed.