Developing and applying novel statistical design methods and computational tools to support clinical trials and research efforts
The design of a clinical trial is a key component in the trial’s viability. Adequate design methods and statistical analyses of data ensure scientific integrity within clinical trials and laboratory research. To assist in these efforts, our group supports physicians and researchers by developing novel design methods for clinical trials and research initiatives.
Outside our state-of-the-art biostatistical collaboration with clinicians and researchers, we develop computational tools for analyzing high-dimensional data. The goal of our work is to provide optimal design methods and biostatistical analyses to support clinical trials and research efforts in childhood cancer and catastrophic disease.
Novel design methods and computational tools ensure clinical trials and laboratory research efforts collect and interpret meaningful data. To contribute to these efforts, our research focuses on the development and application of novel Bayesian and Frequentist adaptive design methods.
Beyond our research and software-development interests, we collaborate with clinicians and researchers at St. Jude to design and analyze studies across the spectrum of laboratory research and early- and late-phase clinical trials for pediatric tumors.
Methodology development and clincial collaboration
Our group’s research interests concentrate on novel design methodologies for use in early-phase (phase I/II) and late-phase (phase III) clinical trials in pediatric tumors. In phase I trials, the goal is to evaluate the safety of the drug or intervention. A portion of our research efforts focus on this early phase, and our goal is to create algorithms that help decide how to identify optimal dosing—in terms of maximum tolerated dose (MTD) or optimal biological dose (OBD)—of patients in phase I trials. We have developed several efficient designs with software packages/apps that offer support in the implementation and simulation of study designs for single-drug and drug-combination phase I trials.
In phase II or III trials, the goal is to evaluate a new intervention without the concurrent control or to evaluate the effectiveness of a drug or intervention(s) compared to existing standard treatments. To meet these goals, we contribute novel design models to these trials. As an example of our work in this area, we developed a screened selection design software package to assist in the successful implementation of phase II trials.
While our focus remains on clinical trials design methods, we also investigate how to appropriately analyze and interpret the trials’ results. For instance, we try to align our collaborative clinical trials to the ICH E9(R1) Estimand guideline.
A defining aspect of our group is our in-depth collaboration with clinicians as we support their development and analyses of clinical trials. Our laboratory analyzes data from the clinic and applies existing and novel biostatistical methods. Much of our collaborative statistical services focus on solid tumors and other tumors through our partnership with the Department of Oncology (Solid Tumor Division) and the Department of Psychology and Biobehavioral Sciences.
Because of the collaborative partnerships with clinicians and researchers, our group has the notable opportunity to develop and apply new biostatistical methods and computational tools that provide an impact to exploratory clinical care and research at St. Jude. In all our work, we strive to develop and apply novel design methods and computational tools that push the boundaries of what is possible in the biostatistical design and analysis of clinical trials.
Phase I effective dose-finding design
Phase I/II design
Phase II/III design