Effect of Baseline Breast Density on Breast Cancer Incidence, Stage, Mortality, and Screening Parameters: 25-Year Follow-up of a Swedish Mammographic Screening

Abstract

We aimed to quantitatively assess the effect of baseline breast density on the incidence, stage, and mortality, and also the natural course of the disease, considering the sensitivity of mammography to clarify its causal or masking effect. In total, 15,658 women ages 45 to 59 years from the Kopparberg randomized controlled trial in Sweden were prospectively followed from 1977 until 2004 to ascertain breast cancer incidence and death. Dense breast tissue collected at the beginning of the study was defined as pattern IV or V by the Tabár classification. Conventional risk factors were also collected at baseline. The three-state Markov model was used to estimate the preclinical incidence rate and the mean sojourn time given the fixed sensitivity. Dense breast tissue was significantly associated with breast cancer incidence [relative risk (RR)=1.57 (1.18-1.67)] and with breast cancer mortality [RR=1.91 (1.26-2.91)] after adjusting for other risk factors. Cumulative incidence rates irrespective of nonadvanced and advanced breast cancer were higher in dense breast tissue compared with nondense tissue but no difference in survival was detected between dense and nondense breast tissue. Dense breast tissue had a higher preclinical incidence rate (causal effect) and shorter mean sojourn time (masking effect) compared with nondense breast tissue by controlling the sensitivity of mammography. We corroborated the effect of baseline breast density with a higher incidence and mortality and also showed its contribution to a masking effect with long-term follow-up data. Results suggest that the screening policy with a predominantly shorter screening interval and with alternative imaging techniques might be indicated in women with dense breast tissue.