Time-Varying Effects of Prognostic Factors Associated with Disease-Free Survival in Breast Cancer

• Activity Number: 561
• Type: Contributed
• Date/Time: Thursday, August 6, 2009 : 8:30 AM to 10:20 AM
• Sponsor: Section on Statistics in Epidemiology
• Author(s): Karen Messer*+ and Minya Pu and Barbara Parker and Cynthia Thomson and Bette Caan and Shirley Flatt and Lisa Madlensky and Richard Hajek and Wael Al-Delaimy and Nazmus Saquib and Ellen Gold and John Pierce and Loki Natarajan
• Companies: University of California, San Diego and University of California, San Diego and University of California, San Diego and University of Arizona and Kaiser Permanente Northern California and University of California, San Diego and University of California, San Diego and The University of Texas M.D. Anderson Cancer Center and University of California, San Diego and University of California, San Diego and University of California, Davis and University of California, San Diego and University of California, San Diego
• Keywords: Time-varying hazards ; Cox proportional hazards models ; breast cancer survivorship

Abstract

Early detection and effective treatments have improved breast cancer survivorship. In long-term cancer studies it is unlikely that exposure-disease risks remain constant over the follow-up period, necessitating models that incorporate time-varying effects for covariates. Effects of tumor characteristics on breast-cancer free survival were investigated in a randomized dietary intervention trial with 3088 participants. A piecewise-constant penalized spline approach incorporating time-varying coefficients was adopted allowing for deviations from the proportional hazards assumption. Significant results include: compared to Stage I tumors, the HR of having Stage II/III decreased from 3.1during the first 2.5-years to 2.1 after 7.7 years but remained a significant risk factor; similar diminishing effects were found for poorly differentiated tumors but reverse effects were found for being ER+.