"Difficult questions about difficult endpoints: Carryover effects on incident hypertension"
Thomas Lumley, PhD Professor of Biostatistics Department of Statistics University of Auckland
Abstract: A small but interesting class of trials delivers a treatment that is known to be effective, then stops treatment and tries to see how much longer the benefit lasts. A concrete example is the TROPHY trial using an angiotensin receptor blocker to postpone incident hypertension, but the same issue is relevant for intensive lifestyle interventions. Assessing carryover effects is especially hard when the endpoint is hard to pin down in time, as for incident hypertension or diabetes, where multiple noisy measurements are needed to determine whether a disease threshold has been crossed. Simulations show it is infeasible to choose a measurement schedule that allows a simple comparison of cumulative incidence. I describe how to fix the problem by analysis, taking advantage of an unusual situation where the MAR assumption is actually true. This is joint work with Gwynn Sturdevant.
"Teaching the Statistical Investigation Process with Simulation-Based Inference to Improve Statistical Thinking about Biomedical Research"
Nathan Tintle, PhD
Associate Professor, Dordt College, Sioux Center, Iowa
"Sample Size Re-Estimation: Is Bigger Always Better?"
Leslie Ain McClure, PhD, Professor and Chair
Department of Epidemiology and Biostatistics
“Adaptive Modeling: Methods Overview and Example Analyses”
George J. Knafl, PhD, Center Investigator
School of Nursing, University of North Carolina at Chapel Hill
"The Utility of Conditional Autoregressive (CAR) Models for Modeling Efficacy of Molecularly Targeted Agents in Early-Phase Trials"
Thomas M. Braun, PhD Professor of Biostatistics University of Michigan School of Public Health
Abstract: Phase I trials in oncology have advanced from studies of the safety of a single agent to studies of the simultaneous toxicity and efficacy of two agents. Furthermore, the new agents under study are often molecularly targeted agents (MTAs) rather than chemotherapeutic agents. The nature of MTAs brings into question the adage from the era of chemotherapy that “more is better”, requiring deeper thought into how we model efficacy as a function of dose changes of either or both of the MTAs. We demonstrate that several published modeling choices implicitly place strong correlation constraints among the dose combinations that lead to over smoothing of the data. As an alternative, we propose the use of a conditional autoregressive (CAR) model, which allows us to model the correlation directly and leads to a direct control on the amount of smoothing that is used. We describe the general structure of CAR models and then present simulation results comparing the operating characteristics of a CAR model-based design with other existing designs. We then conclude with a discussion of several nuances of CAR models that require further study.
"Orthogonality and Constraints: Powerful Tests for Gene-treatment and Gene-environment Interactions"
James Y. Dai, PhD Associate Member and Affiliate Associate Professor Fred Hutchinson Cancer Research Center Seattle, WA
Abstract: Gene-treatment and gene-environment interactions are frequently of interest in cancer research. The power for detecting interactions is typically low in genome-wide studies, and few significant interactions have been discovered. In this talk, I will present statistical methods that exploit prior knowledge, for example the gene-treatment independence dictated by randomization in cancer trials, and biological plausibility for a more focused search for gene-treatment interactions. Generalization of case-only estimators will be evaluated. For multiple gene-environment interactions in cancer epidemiological studies, I will discuss testing strategies such as restricting the multivariate alternative hypothesis to certain plausible direction, thereby gaining power.
Mi-Ok Kim, PhD Associate Professor Biostatistics and Epidemiology Deptartment of Pediatrics University of Cincinnati
“From Treatment Targets to Disparities:Improving Outcomes with Patient-Centered Research in Chronic Inflammatory Skin Diseases”