Research Seminars: Probability and Statistics

Spring 2024

Time & Location: All talks are on Wednesdays in Gibson 126 at 4:00 PM unless otherwise noted.
Organizer: Xiang jiMichelle Lacey and Yuwei Bao

Archives

 

February  2

Title: Stochastics in medicine: Delaying menopause and missing drug doses
Sean Lawley| University of Utah

Abstract:  Stochastic modeling and analysis can help answer pressing medical questions. In this talk, I will attempt to justify this claim by describing recent work on two problems in medicine. The first problem concerns ovarian tissue cryopreservation, which is a proven tool to preserve ovarian follicles prior to gonadotoxic treatments. Can this procedure be applied to healthy women to delay or eliminate menopause? How can it be optimized? The second problem concerns medication nonadherence. What should you do if you miss a dose of medication? How can physicians design dosing regimens that are robust to missed/late doses? I will describe (a) how stochastics theory offers insights into these questions and (b) the mathematical questions that emerge from this investigation. The first problem is based on joint work with Joshua Johnson (University of Colorado School of Medicine), John Emerson (Yale University), and Kutluk Oktay (Yale School of Medicine).

Time: 12:00 pm
Location: Stanley Thomas 316

 

February  28

Title: Heavy-tailed p-value combinations from the perspective of extreme value theory
Yeonwoo Rho - Michigan Technological University

Abstract:  Handling multiplicity without losing much power has been a long-standing problem in statistics. Recently, p-value combination methods based on heavy-tailed distributions have received much spotlight. In this talk, p-value combinations from the perspective of extreme value theory will be introduced. Distributions with regularly varying tails, a subclass of heavy tail distributions, are found to be useful in constructing such combined p-values. Three combined p-values are introduced, of which left tail probabilities are shown to be approximately uniform when the global null is true. The number of tests can be both finite and diverging. Connections to existing literature will also be discussed.

Time: 4:00 pm
Location: Gibson Hall 126

 

March  13

Title: Overview of Master Protocol Trials and Statistical Considerations
Xiaoyun(Nicole) Li – Senior Director at BeiGene

Abstract:  Master protocol is a trial structure that evaluates multiple diseases or multiple drugs (or drug combinations) within the same trial. There are three main types of master protocols, i.e., basket trials, umbrella trials and platform trials. Basket trials evaluate the same drug/drug combination in different diseases within the same trial, with the assumption that similar drug activities may seen and data may be borrowed. Various basket trial designs have been proposed over the years and I will give a flavor of it. Umberlla trials evaluate multiple drugs/drug combinations in the same disease within a trial. There is usually a shared control arm for all the different experimental arms to increase the efficiency. I will talk about the statistical consideration in terms of type I error control and other statistical errors in terms of umbrella trials. Platform trials refer to umbrella trials in a perpetual manner. Statistical considerations arise as to whether we could use the non-contemporaneous (non-concurrent) control and if so, how to use it. I will also talk about a phase 3 umbrella trial design as an illustration.

Time: 4:00 pm
Location: Zoom with meeting ID: 932 4354 5612

 

April  3

Title: Statistical methods used for clinical research
Hiya Banerjee – Director of Biostatistics at Eli Lilly

Abstract: 

In the technical presentation, I will showcase an innovative statistical method utilized to address a clinical question in the context of drug marketing. I will provide a comprehensive overview of how statisticians are involved in approaching and solving the problem, shedding light on the formulation of hypotheses and our collective endeavors to reach resolutions.

Besides that I will talk about  how our daily responsibilities influence the trajectory of drug development. Furthermore, I will touch upon the essential skills and behaviors that aspiring students can cultivate to successfully embark on a career in the industry. The conversation will be informal, allowing for ample time for interactions and questions, providing insights into potential careers.

Time: 4:00 pm
Location: Gibson 126

 

 

April 10

Title: The Proximal Distance Principle for Constrained Estimation
Alfonso Landeros – University of California, Riverside 

Abstract: 

Statistical methods often involve solving an optimization problem, such as in maximum likelihood estimation and regression. The addition of constraints, either to enforce a hard requirement in estimation or to regularize solutions, complicates matters. Fortunately, the rich theory of convex optimization provides ample tools for devising novel methods.

In this talk, I present applications of distance-to-set penalties to statistical learning problems. Specifically, I will focus on proximal distance algorithms, based on the MM principle, tailored to various applications such as regression and discriminant analysis. Special emphasis is given to sparsity set constraints as a compromise between exhaustive combinatorial searches and lasso penalization methods that induce shrinkage. 
 

Time: 4:00 pm
Location: Gibson 126

 

May: 1

Title: Extrapolation in pediatric clinical trials
Jingjing Ye – Executive director and global head at BeiGene

Abstract:  Pediatric population presents several barriers for clinical trial design and analysis, including ethical constraints on the sample size and slow accrual rate. Pediatric drug development lags adult development by about 8 years and children represent a large underserved population of "therapeutic orphans," as an estimated 80% of children are treated off-label (Mulugeta et al. in Pediatr Clin 64(6):1185-1196, 2017). It may be appropriate to extrapolate existing data when the course of the disease or condition and effects of the drugs are sufficiently similar between source and target populations. Pediatric extrapolation is an approach that the FDA and EMA are encouraging sponsors to adopt to minimize the number of children that are required to participate in clinical trials. With recent ICH harmonization on pediatric extrapolation framework, despite uncertainties, pediatric drug development programs should initially assume some degree of extrapolation. The degree to which extrapolation can be used lies along a continuum representing the uncertainties to be addressed through generation of new pediatric evidence (Gamalo et al. 2021). Represented by ASA Pediatric Working Group of Extrapolation Subgroup, this talk will help audience recognize when extrapolation is appropriate, understand the extrapolation landscape in pediatrics, how extrapolation was applied in a small pediatric trial through real examples, introduce recent statistical developments for extrapolation frameworks, and share our thoughts on the regulatory guideline of ICH E11.

Time: 4:00 pm
Location: Zoom with meeting ID: 975 7934 5167