Tutorial- Monday at 12:45 PM EDT
Title: Leveraging Historical Controls Using Multisource Adaptive Design
Presenters: Brian Hobbs, University of Texas

Abstract:
Beneficial therapeutic strategies are established through a gradual process devised to define the safety and efficacy profiles of new strategies over a sequence of clinical trials. This system produces redundancies, whereby similar treatment strategies are replicated, either as experimental or comparator standard-of-care therapies, across development phases and multiple studies. This article describes a collection of web-based statistical tools hosted by MD Anderson Cancer Center that enable investigators to incorporate historical control data into analysis of randomized clinical trials using Bayesian hierarchical modeling as well as implement adaptive designs using the method described in Hobbs et al. (2013). By balancing posterior effective sample sizes among the study arms, the adaptive design attempts to maximize power on the basis of interim posterior estimates of bias. With balanced allocation guided by hierarchical modeling, the design offers the potential to assign more patients to experimental therapies and thereby enhance efficiency while limiting bias and controlling average type I error.


References:

Bretz, Pinheiro, Branson (2005) Combining multiple comparisons and modelling techniques in dose-response studies, Biometrics, vol. 61, p. 738-748.
Pinheiro, Bornkamp, Bretz (2006) Design and analysis of dose finding studies combining multiple comparisons and modeling procedures. Journal of Biopharmaceutical Statistics, 16(5), 639-656.
Pinheiro, Bornkamp, Glimm, Bretz (2014) Model-based dose finding under model uncertainty using general parametric models. Statistics in Medicine 33(10): 1646-661


Bio:
Brian Hobbs completed a doctoral degree in biostatistics at the University of Minnesota and then joined The University of Texas MD Anderson as an Assistant Professor of biostatistics. He was promoted to Associate in 2017, and then recruited to Cleveland Clinic to found a Section of Cancer Biostatistics. He joined The University of Texas Dell Medical School in August 2020 as a tenured Associate Professor. The Eastern North American Region of International Biometric Society selected his thesis paper for the John Van Ryzin Award in 2010.

In 2016, Dr. Hobbs was selected by The University of Minnesota for the Emerging Leader Award, an honor bestowed on alumni on the basis of impactful contributions within 10 years of graduating from one of The School of Public Health’s 20 programs. Recognized as an expert in clinical oncology research methodology, in 2017 Dr. Hobbs was invited to lead the publication of National Cancer Institute’s Clinical Trials Design Task Force with the goal of providing national, consensus recommendations for first-in-human cancer drug trials that use seamless designs.

In 2019, Dr. Hobbs was invited to describe recent advances and current issues with master protocol designs in the Journal of Clinical Oncology Precision Oncology. In 2020, he was invited to contribute to an article for Nature Reviews Clinical Oncology describing the current state of tumor agnostic trials. In 2021, Dr. Hobbs was invited to review the landscape of basket trials in the Journal of Clinical Oncology.