|
Abstract:
|
In precision medicine, biomarkers are often used to identify subgroups of patients who are likely to benefit from investigative therapies. Continuously valued biomarker such as baseline HbA1c for T2DM and MMSE for Alzheimer's disease may be applicable. To guard against over optimism from searching through all possible cut-point values, one should adjust for the double-multiplicity of inference on marker-positive, marker-negative, and all-comes patient groups at each cut-point, and searching through all cut-point values. This presentation gives a principled approach, called Subgroup Mixable Estimation (SME), which provides, at each candidate cut-point, simultaneous confidence intervals for efficacy of the drug on marker-positive patients, marker-negative patients, and their mixture, which are then further adjusted for the multiplicity of searching through all possible cut-point values. These confidence intervals respect logical relationships among the subgroups and their mixture at each cut-point, and across the cut-point values.
|
