|
Abstract:
|
Treatment selection biomarkers are used to identify patients who will benefit from a specific therapy. Typically, treatment selection biomarkers are used to refine treatment protocols, so that treatment is determined by the measured biomarker values. An imperfect example is OncoTypeDX, which is used to refine treatment for node negative, hormone receptor positive breast cancer patients treated with endocrine therapy, dividing them into a group that receives adjuvant chemotherapy and a group that does not. This is called ``marker-guided therapy." This and other examples will be discussed. We present novel sample size methods for inference about the change in the the population proportion of patients who survive a specified time (e.g., 5 years) if all receive marker-guided therapy, a metric previously developed by others. The methods are appropriate for studies that produce right-censored survival data.
|
