Presents modern Bayesian statistical models and methods for identifying treatments tailored to individual patients using their prognostic variables and predictive biomarkers.

Bayesian Precision Medicine presents modern Bayesian statistical models and methods for identifying treatments tailored to individual patients using their prognostic variables and predictive biomarkers. The process of evaluating and comparing treatments is explained and illustrated by practical examples, followed by a discussion of causal analysis and its relationship to statistical inference. A wide array of modern Bayesian clinical trial designs are presented, including applications to many oncology trials. The later chapters describe Bayesian nonparametric regression analyses of datasets arising from multistage chemotherapy for acute leukemia, allogeneic stem cell transplantation, and targeted agents for treating advanced breast cancer.

Features:

• Describes the connection between causal analysis and statistical inference
• Reviews modern personalized Bayesian clinical trial designs for dose-finding, treatment screening, basket trials, enrichment, incorporating historical data, and confirmatory treatment comparison, illustrated by real-world applications
• Presents adaptive methods for clustering similar patient subgroups to improve efficiency
• Describes Bayesian nonparametric regression analyses of real-world datasets from oncology
• Provides pointers to software for implementation

Bayesian Precision Medicine is primarily aimed at biostatisticians and medical researchers who desire to apply modern Bayesian methods to their own clinical trials and data analyses. It also might be used to teach a special topics course on precision medicine using a Bayesian approach to postgraduate biostatistics students. The main goal of the book is to show how Bayesian thinking can provide a practical scientific basis for tailoring treatments to individual patients.