Biomarkers associated with the treatment response heterogeneity hold potential for treatment selection. In practice, the decision regarding whether to adopt a treatment selection marker depends on the effect of treatment selection on the rate of targeted disease as well as additional cost associated with the treatment. We propose an expected benefit measure that incorporates both aspects to quantify a biomarker's treatment selection capacity. This measure extends an existing decision-theoretic framework, to account for the fact that optimal treatment absent marker information varies with the cost of treatment. In addition, we establish upper and lower bounds for the performance of a perfect treatment selection model, which provides a basis for standardizing markers' expected benefit. We develop model-based estimators for those measures in a randomized trial setting and evaluate their asymptotic properties. An adaptive bootstrap confidence interval is proposed for inference in presence of non-regularity. Alternative robust estimators based on a working model are also investigated. The application of our methods is illustrated using an example in the Diabetes Control and Complications Trial where we evaluate the expected benefit of the baseline hemoglobin A1C in selecting diabetes treatment.



Included in

Biostatistics Commons