Genetic models for the familial aggregation of mammographic breast density

Abstract

Background: Mammographic breast density (MBD) has a strong genetic component. Investigating the genetic models for mammographic density may provide further insights into the genetic factors affecting breast cancer risk.

Purpose: To evaluate the familial aggregation of MBD and investigate the genetic models of susceptibility.

Methods: We used data on 746 women from 305 families participating in the Sisters in Breast Screening study. Retrieved mammograms were digitized, and percent mammographic density was determined using the Cumulus software. Linear regression analysis was done to identify the factors that are associated with mammographic density and a multivariate regression model was constructed. Familial correlations between relative pairs were calculated using the residuals from these models. Genetic models of susceptibility were investigated using segregation analysis.

Results: After adjusting for covariates, the intraclass correlation coefficient among the residuals was 0.26 (95% confidence interval, 0.16-0.36) in sister-sister pairs and 0.67 (0.27-1.00) among the monozygotic twin pairs. The most parsimonious model was a Mendelian single major gene model in which an allele with population frequency 0.39 (95% confidence interval, 0.33-0.46) influenced mammographic density in an additive fashion. This model explained 66% of the residual variance.

Conclusion: These results confirm that MBD has a strong heritable basis, and suggest that major genes may explain some of the familial aggregation. These results may have implications for the search of genes that control mammographic density.