Assessing Association Between the Risk of Clinical Outcome Event and Patient's Disease Status That Vary by Time

• Activity Number: 368
• Type: Contributed
• Date/Time: Tuesday, August 11, 2015 : 10:30 AM to 12:20 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Yiran Hu* and Bo Fu and Yuanyuan Tang
• Companies: and AbbVie and AbbVie
• Keywords: Clinial Outcome ; Clinical Trial ; Hepatitis C ; SVR

Abstract

In clinical studies with advancing diseases, the effectiveness of a treatment, to both the doctors and the patients, not only means the one-time "cure" of the disease but also means delayed or prevented progression of related diseases. For example, in the Hepatitis C therapeutic area, the cure of the disease is generally indicated by subject's SVR12 (sustained virologic response 12 weeks follow treatment) status, but doctors and patients are as well interested in whether the cure of the hepatitis disease will reduce or delay the occurrence of advanced liver-related diseases or even death. In many literatures, the occurrence of liver-related outcomes is modeled by Cox proportional hazards where SVR12 status is treated as a constant. Our research interest lies in the challenge when patient's disease status can vary by time and we would like to investigate whether the time-dependent SVR12 status are associated with the risk for a liver-related advanced clinical outcome. Simulated data is constructed based on the historical information from literature to examine alternative ways to incorporate time-varying covariates in the model.