[251] Are Columnar Cell Alteration and Sclerosing Adenosis, Independent Risk Markers for Breast Cancer?

A Nassar, DW Visscher, CA Reynolds, RA Vierkant, SS Anderson, MH Frost, LC Hartmann. Mayo Clinic, Rochester, MN; University of Michigan, Ann Arbor, MI

Background: Previous studies have found that SA carries a 1.5 to 3.7 times increased risk of developing invasive cancer compared to those without SA. This relative risk (RR) may increase to 5.5, translating to a 1.2% risk per year of cancer, if atypia is associated with the SA. On the other hand, columnar cell alteration (CCA) is noted to be associated with more worrisome lesions (ADH, DCIS or invasive cancer) in 25-30% of cases, but the risk association with breast cancer is still unknown.
Design: SA and CCA were assessed in 9344 women in the Mayo Benign Breast Disease Cohort who underwent excisional breast biopsy between 1967-1991. Sclerosing adenosis is defined as a combination of adenosis (epithelial and myoepithelial proliferation of lobules) and stromal sclerosis. Columnar cell alteration is characterized by the presence of columnar-shaped epithelial cells lining enlarged terminal-duct lobular units (columnar cell change and hyperplasia without cytologic atypia). Heterogeneity of breast cancer risk across levels of SA and CCA was assessed, standardized to a control population (the Iowa Surveillance, Epidemiology, and End Results registry). Each factor (SA and CCA) was evaluated separately initially and also as a combined entity for association. Analyses adjusted for age at biopsy, histology (non-proliferative disease [NP], proliferative disease without atypia [PDWA] and atypical hyperplasia [AH], both ductal and lobular) and lobular involution (none, partial, complete).
Results: SA and CCA were identified in 2350 (25%) and 2025 (22%), respectively. SA and CCA were both associated with presence of proliferative disease, lower levels of involution and later age at BBD diagnosis (p<0.001 for each). After adjustment for these variables, no association was found between CCA and risk of breast cancer (p=0.54). In contrast, women with SA had increased risk compared to those without (RR=1.41, 95% CI 1.15-1.71, p=0.003). No significant heterogeneity of risk was observed when jointly examining the two factors (presence of either CCA and SA vs. neither, p=0.17; presence of both CCA and SA vs. not, p=0.28).
Conclusions: Sclerosing adenosis is an independent breast cancer risk factor, conferring a RR of 1.41; whereas CCA does not independently alter risk.
Category: Breast

Wednesday, March 11, 2009 9:30 AM

Poster Session V # 18, Wednesday Morning