Biostatistics Program

Scott Evans, PhD, MS, is a professor and founding chair of the Department of Biostatistics and Bioinformatics and the director of the Biostatistics Center at Milken Institute School of Public Health at the George Washington University. He is the director of the Statistical and Data Management Center for the Antibacterial Resistance Leadership Group (ARLG) funded by National Institute of Allergy and Infectious Diseases (NIAID)/ National Institutes of Health (NIH); principal investigator of the Coordinating Center for the Exercise and Nutrition Interventions to Improve Cancer Treatment-Related Outcomes (ENICTO) in Cancer Survivors Consortium, funded by the National Cancer Institute (NCI)/NIH, and the co-PI of the Data Coordinating Center of the Clamp OR Delay among neonates with Congenital Heart Disease (CORD-CHD) clinical trial, funded by the National Heart, Lung, and Blood Institute (NHLBI)/NIH. He is the co-chair of the Benefit-Risk Balance for Medicinal Products Working Group of the Council for International Organizations of Medical Sciences (CIOMS); editor of a mini-series on data and safety monitoring boards for NEJM Evidence; and the president-elect of the Society for Clinical Trials (SCT). He is a recipient of the Mosteller Statistician Award, the Zackin Distinguished Collaborative Statistician Award, the Founders Award from the American Statistical Association (ASA), an elected member of the International Statistical Institute (ISI), and is a fellow of the ASA, SCT, and the Infectious Disease Society of America (IDSA).

Abstract

Randomized clinical trials are the gold standard for evaluating the benefits and harms of interventions, though often fail to provide the necessary evidence to inform medical decision-making. Primary reasons are failure to recognize: (1) the most important questions for treating patients in clinical practice, and (2) that traditional approaches do not directly address these most important questions, and subsequently not using these recognitions as the motivation for the design, monitoring, analysis, and reporting of clinical trials. The standard approach of analyzing one outcome at a time, fails to incorporate associations between or the cumulative nature of multiple outcomes in individual patients, suffers from competing risk complexities during interpretation of individual outcomes, fails to recognize important gradations of patient responses, and since efficacy and safety analyses are often conducted on different populations, benefit:risk generalizability is unclear. Treatment effect heterogeneity is typically sub-optimally evaluated based on a single efficacy or safety endpoint, and rarely evaluated based on benefit:risk. The desirability of outcome ranking (DOOR) is a paradigm for the design, analysis, and interpretation of clinical trials based on comprehensive patient-centric benefit-risk evaluation, developed to address these issues and advance clinical trial science.

Darren McGuire, MD, is a general cardiologist in the division of cardiology at the University of Texas Southwestern Medical Center, and a clinical trialist with expertise in large-scale CV outcomes trial design and execution, and drug registration/regulation, with a focus on diabetes and cardiovascular disease. He is a distinguished teaching professor of medicine with tenure, is the Jere H. Mitchell MD Distinguished Chair in Cardiovascular Science and is the lead physician of the Parkland Health Cardiology Clinics. McGuire has participated in numerous international cardiovascular clinical outcomes trials, including T2DM, obesity, and lipid trials. He is a previous member of the FDA Cardiovascular and Renal Drugs Advisory Committee. Additionally, he is deputy editor of Circulation, senior editor of Diabetes and Vascular Disease Research, and co-editor of the textbook: “Diabetes in Cardiovascular Disease: A Companion to Braunwald’s Heart Disease.” McGuire has authored/co-authored more than 480 peer-reviewed manuscripts, reviews, editorials, and book chapters, and is recognized as a “Clarivate Highly Cited Researcher.”

Abstract 

This talk will review some of the history and examples of composite outcomes used in cardiovascular clinical outcomes trials. It will also review lessons learned from some trials missing the mark due to choices made about the components of such outcomes or their methods of analysis, and some of the variations in choosing composite outcomes with regard to current and future CV outcomes trials that provide a more clinically relevant and informative evaluation and synthesis of treatment benefit, safety and/or patient quality-of-life.

Fortunato (Fred) Senatore, MD, PhD, FACC, has been a medical officer in the division of cardiology and nephrology at the U.S. Food and Drug Administration (FDA) since 2012, and has served as team leader/lead physician since January 2020. He has published on shock, medication adherence/representative populations, benefit/risk assessment, heart failure, and clinical trial challenges during the COVID-19 pandemic. He co-edited a recently released textbook on basic and clinical research, published by Springer. Additionally, Senatore teaches a variety of courses on trial design and data analysis. Prior to his tenure at the FDA, he served in the pharmaceutical industry for 17 years and was a professor of chemical engineering at Texas Tech University, where he specialized in artificial organ technology, biocompatibility, hemodynamics, and modeling/simulation of biological processes. He received his BA in biochemistry and MS in bioengineering from Columbia University, his PhD in chemical engineering from Rutgers University, his MD degree from the Texas Tech University Health Sciences Center School of Medicine, his internal medicine training at the Mayo Clinic, and his cardiology training at Harvard Medical School/Massachusetts General Hospital.

Abstract

This history and definitions of “Major Adverse Cardiac Events” (MACE) as a composite endpoint in randomized clinical trials over the past decade are reviewed. Increasing the number of components in the composite endpoint introduces subjectivity and consequently reduces prognostic relevance of the composite endpoint. Alternative endpoints deemed acceptable include the 6-minute walking distance and the Kansas City Cardiomyopathy Questionnaire based on the feel-function-survive paradigm and on pre-specified criteria for clinically meaningful results. Considerable effort has focused on validating biomarkers as surrogates for clinical outcome. This requires a strong correlation between the biomarker and clinical outcome void of confounding covariates and mechanistic pathways that may disengage the effect of an intervention on the biomarker and on the clinical outcome, respectively. To further streamline a clinical trial, it is no longer necessary to engage an external endpoint adjudication committee in cardiovascular clinical trials with cardiologists as the principal investigators.

Shanthi Sethuraman, PhD, has more than 25 years of experience in the pharmaceutical industry, spanning various cross-functional roles across global statistical sciences. These include regulatory, chemistry, control and manufacturing, and project management. Currently, she is senior vice president and chief analytics officer of global statistical sciences, research and development, at Eli Lilly and Company. There, she leads statisticians who are integral to driving innovation and supporting discovery research for pre-clinical; chemistry, manufacturing, and controls (CMC); clinical drug development; post- launch patient access; advanced analytics; and the statistical innovation center. Sethuraman has been active in driving leadership development internally and externally. She is an active member of the Leadership in Practice Committee and Biopharmaceutical Statistics Leadership Consortium.

Abstract

With advances in science, many breakthroughs have been made in the treatment of metabolic related diseases including treatment of obesity with incretins, the reduction of Lp(a) for cardiovascular risks, and flourishment of experimental treatments for liver and kidney diseases. There is an unprecedented need for conducting more outcome studies to confirm the clinical benefits of these treatments in a more efficient manner. The presenters will share thoughts regarding ideas and initial work in using analytics to achieve efficiency in outcome studies. The idea is to use concepts of artificial intelligence (AI) and machine learning (ML) to improve the adjudication process. This talk will review the current adjudication process and presenters will share thoughts on how to improve this process, which could potentially speed up the adjudication process and reduce adjudication cost. Secondly, potential opportunities to optimize the enrollment inclusion criteria to balance the event rate, patient population and enrollment speed, and the relative risk reduction for a treatment will be discussed. Specifically, this presentation will introduce the concepts of unmodifiable and modifiable risk factors, using the risk model to optimize the relative risk reduction, absolute risk reduction, and patient population size. Following the presentation, the presenters will seek input from the audience on additional opportunities to enable efficient and effective outcome studies for the most appropriate patient population.

Rui (Sammi) Tang, PhD, is a leading expert of biostatistics in the biotech/pharmaceutical industry. She is currently the vice president, global head of biometrics, at Servier Pharmaceuticals. Tang is also co-founder of DahShu, a 501(c)(3) non-profit organization, founded to promote research and education of its 5000 members. She is currently leading teams in the innovative design scientific working group of oncology drug development and the small population working group for rare disease statistical methodology development. Prior to joining Servier, Tang was the biostatistics therapeutic area head for multiple teams at Shire Pharmaceuticals including oncology, transplants, ophthalmology, and prematurity neonates. She has worked at several other organizations, including Vertex, Amgen, Mayo Clinical, and Merck. Her research interests are primarily in the area of adaptive clinical trial design and statistical issues in precision medicine. She has authored more than 50 articles in peer-reviewed scientific journals on methodology, study design, data analysis and reporting and is a co-inventor of several patents. Tang is also an active member in ASA (American Statistics Association) and ICSA (International Chinese Statistics Association).

Abstract

In the landscape of clinical drug development, the strategic selection of endpoints stands as a cornerstone, pivotal in evaluating the effectiveness of interventions or drugs on enhancing patient survival and quality of life. This selection process is intricately influenced by a comprehensive understanding of both clinical and regulatory frameworks, which collectively shape the research question at hand. From a clinical perspective, this encompasses a deep understanding of the disease, available treatment options, therapeutic strategies, and the viewpoints of both patients and healthcare providers. Regulatory considerations further extend to the specific objectives of the endpoint, including its role in product labeling. The meticulous selection of endpoints in clinical trials is essential, as it underpins the ability of trial results to inform healthcare decision-making, spanning policy formulation to clinical practice. This presentation will embark on a journey through the realm of endpoint development, beginning with a high-level introduction to the estimand framework. This framework sets the stage for a deeper exploration of therapeutic trial endpoint considerations across different case studies, each illustrating unique challenges and strategies outside the direct scope of the estimand framework. Concluding the presentation, we delve into the evolving landscape of endpoint development within real-world evidence in clinical trials, highlighting how real-world data is increasingly influencing the selection and validation of endpoints in a way that resonates with both clinical and regulatory imperatives.

Lee-Jen (LJ) Wei, PhD, is a professor of biostatistics at Harvard University. His research focus is on clinical trial methodology, specializing in design, monitoring, and analysis of studies. His recent research is concentrated on translational statistics and the personalization of medicine under the risk-benefit paradigm via biomarkers and the revitalizing of clinical trial methodology. He has extensive working experience in regulatory science for developing and evaluating new drugs/devices. Wei has developed numerous novel statistical methods which are often utilized in practice. He received the prestigious Wald Medal in 2009 from the American Statistical Association for his contribution to clinical trial methodology. He is a fellow of American Statistical Associating and Institute of Mathematical Statistics. In 2014, to honor his mentorship, Harvard T.H. Chan School of Public Health established a Wei-Family scholarship to support students who are studying biostatistics. He has more than 260 publications and served on numerous editorial and scientific advisory boards including data monitoring for governments and industry.

Abstract 

The goal of conducting a clinical study is to obtain robust, clinically interpretable treatment effect estimate with respect to harm-benefit perspectives at the patient’s level via efficient and reliable quantitative procedures. To execute a valid and successful comparative trial, a crucial step is to choose appropriate study endpoints with which we can quantify the clinically interpretable treatment effects/differences. Generally, for a clinical study, multiple outcomes from each study participant would be collected, which reflect the disease burden/progression observed from different angles and perspectives. The conventional wisdom is to identify a primary endpoint and various secondary endpoints for designing and analyzing the study. This approach may not be ideal or efficient for decision making clinically and statistically, especially for rare disease studies. This talk will discuss various issues and concerns about the current practice via real-world examples. It will present simple remedies, which would be appreciated by the patients, clinicians, and other stakeholders for treatment selection decisions.