Monday, October 19, 2009
Grand Ballroom, Salons 1 & 2 (Renaissance Hollywood Hotel)
Rowan Iskandar, MA, Todd M. Tuttle, MD, MS and Karen M. Kuntz, ScD, University of Minnesota, Minneapolis, MN
Purpose: Several studies report the risk of developing a contralateral breast cancer (CBC) is 2 to 6 times the risk of unilateral breast cancer (UBC) in the general population. Because the group of women at risk for CBC is a selected high-risk group, this increase in risk may not be applicable to a particular woman diagnosed with breast cancer who is choosing from several possible CBC prevention strategies. We sought to estimate a more unbiased woman-specific relative risk of CBC risk vs. UBC risk (RRCBC) that is consistent with the population-based RRCBC estimates from the literature (i.e., selected group of UBC patients vs. general population). Method: We simulated UBC incidence among a population of breast-cancer-free women. We used an overall 5-year UBC risk of 2.7%, which we multiplied by a gamma random value to represent the variability of risk across women. We varied the standard deviation (SD) of this distribution between 1.1 (a published estimate of the SD of Gail scores among average-risk women) and 3 (to reflect variability not captured by the Gail model). Among women diagnosed with UBC we subsequently simulated her risk of CBC to estimate the population-based RRCBC, assuming that each woman’s risk for CBC is the same as her risk of UBC (i.e., woman-specific RRCBC=1). Result: When the distribution of risks is more variable (i.e., higher values for SD of UBC risk), RRCBC increases. RRCBC varied from 2.2 to 9.8 as the SD of UBC risk varied from 1.1 to 3, and SDs of 1.0 to 2.2 were consistent with the population-based RRCBC estimates of 2 to 6 from the literature. For a SD of 1.1, we predict that 95% of women with UBC would have a 5-year CBC risk between 1% and 14.6%, and 95% of breast-cancer-free women would have a 5-year UBC risk between 0.1% and 8.3%. Conclusion: Using a simple modeling exercise we found that the reported RRCBC in the literature, based on comparing risk among a selected group vs. an unselected group, is consistent with a scenario where an individual woman’s risk of CBC is the same as her risk was for UBC. These insights could play an important role in choosing the appropriate CBC prevention strategy for women diagnosed with breast cancer.
Candidate for the Lee B. Lusted Student Prize Competition
