Program

Keynote Speeches

Short Courses (13 October 2026)

Draft Program (to be posted) 13-16 October 2026

Keynote Speeches:

Keynote Speaker: Armin Koch, Hannover Medical School, Germany

Title: A Sad Farewell to the Study-Wise Type I Error? Considerations about Innovation and Conservativism in Regulatory Statistics
Date: Wednesday, October 13, 2026

Abstract: Confirmatory clinical trials are the backbone of decision making about efficacy and a positive benefit/risk-ratio for new medications or licensed drugs in new indications. European Scientific Advice is the platform, where stakeholders meet to agree about what should constitute a sound basis for decision making in this context and agree on the future development plan. This is, as a consequence, the place to also directly discuss new approaches for planning, conducting and analyzing these late-stage clinical trials. Guidance should follow once experience with the interpretation of trials incorporating new approaches is available. Many “new” concepts like approaches to multiple testing, adaptive designs, synthetic co-variates found their place in regular study planning this way. This presentation makes a plea that scientific advice has to be somewhat conservative in many instances, provides recommendations, how a safe way forwards can be found, and tries to explain the difference between innovation and lowering the hurdle for regulatory approval. In this context we argue for maintaining a strict position regarding the control of the study-wise type-1-error in confirmatory clinical trials to be precise about the amount of information that will be available for assessment.

Dr. Armin Koch studied mathematics and chemistry at Heidelberg University and has worked as research assistant at the German Centre for the Research on Cancer (DKFZ) between 1984 and 1991. Thereafter he has been an employee at the Institute of Medical Biometry at Heidelberg University. In 1999 he joined the Federal Institute for Drugs and Medical Devices (BfArM) in Germany. From 2001 to 2008 he was head of the unit „Biostatistics and Experimental Design“. Since 2008 he is Director of the Institute for Biostatistics at Hannover Medical School. Prof. Koch is a member of the Scientific Advice Working Party (SAWP) and has been a member of the drafting group for ICH-E17 (multi regional clinical trials) and for ICH-E20 (adaptive designs, until step 2). His research interests are centered around experimental design for clinical trials and the aspects of regulatory statistics with applications in rare and frequent diseases.    
Doctorate (1997):

• Thesis (Dr. sc. hum. for Medical Biometry, University of Heidelberg)

Academic Education (09/1980 – 07/1988):

Study of Mathematics and Chemistry, first for becoming a teacher, then for a Diploma in Mathematics at the University of Heidelberg

Academic & Professional Experience:

Industry Keynote Speaker: Dr. Weili He, AbbVie

Winner, the Heyse Memorial Lecture Award

Title: From Evidence to Decisions: How RWE, Causal Inference, and AI Are Reshaping Biopharmaceutical Statistics

Date: Thursday, October 14, 2026

As real-world data become more widely available and decision-makers increasingly seek timely, relevant evidence across the product lifecycle, statisticians are being asked to expand their toolkit beyond traditional trial-based paradigms. This lecture will highlight how rigorous statistical thinking can help ensure that real-world evidence is fit for purpose, scientifically credible, and decision-relevant for both regulatory and health technology assessment contexts. I will discuss key methodological considerations in formulating the research question, assessing data fitness, selecting appropriate study designs, defining estimands, and conducting sensitivity analyses, emphasizing how these elements work together to transform heterogeneous data into reliable evidence.

The lecture will also examine the growing role of causal inference and AI-enabled methods in evidence generation. Machine learning, natural language processing, and other AI-based approaches offer powerful opportunities to improve efficiency, scale, and the use of unstructured data, but they also introduce important challenges related to transparency, reproducibility, and interpretability. I will argue that these methods should be viewed as complements to, not substitutes for, sound study design and causal reasoning. Looking ahead, the future of biopharmaceutical statistics lies in integrating RWE, causal inference, innovative designs, and AI within a coherent framework that supports timely, robust, and trustworthy decisions across the product lifecycle.

Dr. Weili He is a Distinguished Research Fellow and head of Medical Affairs and Health Technology Assessment (HTA) Statistics at AbbVie, where she leads a team of over 40 statisticians supporting major business activities in Medical Affairs and HTA. Her extensive career spans clinical development at Merck and post-market life-cycle management at AbbVie. Renowned as a leading biostatistician in pharmaceutical statistics, her areas of expertise include early-phase oncology study designs, late-stage clinical trial design and analysis, adaptive trial methodologies, benefit-risk assessment, real-world data and evidence generation, causal inference, and statistical approaches for HTA.

Dr. He has authored over 60 peer-reviewed statistical publications and served as lead editor for three influential books: “Practical Considerations for Adaptive Trials Design and Implementation” (Springer, 2014), “Benefit-Risk Assessment Methods in Medical Product Development” (CRC Press, 2016), and “Real-World Evidence in Medical Product Development” (Springer Nature, 2023). In January 2026, she assumes the role of Editor-in-Chief for Statistics in Biopharmaceutical Research, following her tenure as Associate Editor from 2014 to 2025.

Her accomplishments are recognized by the American Statistical Association (ASA), where she was elected as an ASA Fellow in 2018. At AbbVie, she holds the distinction of AbbVie Distinguished Research Fellow—the company’s highest honor for scientific excellence. Professionally, Dr. He co-founded and now co-chairs the ASA Biopharmaceutical Section (BIOP) HTA Scientific Working Group in 2023, and previously, the BIOP RWE Scientific Working Group (2018–2022). She has held leadership roles on the ASA BIOP Executive Committee, including Chair-Elect, Chair, and Past-Chair (2020–2022), served as Industry Co-chair of the Regulatory-Industry Statistics Workshop (2017), and co-chaired the QSPI Benefit-Risk Working Group (2012–2018). In addition to her professional impact, Dr. He has mentored numerous junior statisticians, fostering the next generation of leaders in the field.

Regulatory Keynote Speaker: Peter Arlett, Head of the Data Analytics and Methods Taskforce, European Medicines Agency (EMA)

Title: Europe in Action: Regulatory Considerations on Clinical Evidence and Data Science

Date: Friday, October 16 (Morning), 2026

Abstract: To be posted.

Peter Arlett is Head of the Data Analytics and Methods Taskforce at the European Medicines Agency. In this role he leads on operations and transformation on clinical evidence at the EMA including clinical trials, real world evidence, safety reporting and data science including AI. He is Chair of the EMA Data Board, Co-Chair of the HMA-EMA Network Data Steering Group, Co-chair of the EMA AI Coordination Group, Co-chair of the Vaccine Monitoring Platform Steering Group, and Member of the ACT EU Steering Group. Prior to taking up this role in 2020, he held leadership roles within the EMA in the areas of pharmacovigilance, epidemiology, and risk management. Prior to starting at EMA in 2008, Peter worked on new legislation and international collaboration for the European Commission, was the UK delegate to the European Committee for Human Medicinal Products, and was an assessor and manager at the UK’s MHRA. He has a medical degree from University College London, and began his career as a hospital physician in Oxford and London. In addition to his role at EMA, Peter is Honorary Professor at the London School of Hygiene and Tropical Medicine. He is also a Fellow of the Royal College of Physicians of Edinburgh and of the Faculty of Pharmaceutical Medicines of London.

Academic Keynote Speaker: Professor Mark van der Laan, University of California, Berkeley

Winner, the Lai Memorial Lecture Award

Title: Targeted Learning for Generating Real World Evidence

Date: Friday, October 16 (afternoon), 2026

Abstract: Targeted Learning follows a general roadmap for 1) accurately translating the real world into a formal statistical estimation problem in terms  of causal estimand, a corresponding statistical estimand, and statistical model; 2) a corresponding template for construction of a targeted maximum likelihood estimator (TMLE) of the statistical estimand; and finally 3) a sensitivity analysis addressing the possible causal gap. The TMLE represents an optimal plug-in machine learning based estimator of the estimand combined with formal statistical inference. The three pillars of TMLE are super-learning, Highly Adaptive Lasso (HAL), and the TMLE-update step, where the latter has various regularizations such  cross-fitted TMLE, collaborative TMLE, and adaptive TMLE (Lars van der Laan et al., 2023).  Through super-learning it can  incorporate high dimensional and diverse data sources such as images, NLP features, and state of art algorithms tailored for such data sources. To optimize finite sample performance, the precise specification of TMLE can be tailored towards the precise experiment and statistical estimation problem in question, while being theoretically grounded, optimal, and benchmarked.  TMLE applies to any estimation problem including causal inference on survival outcomes allowing for time-dependent confounding of treatment and drop-out. We provide  a motivation, explanation, and overview of targeted learning; the key role of super-learning and HAL; discuss some of the key choices and considerations in specifying the TMLE-step; and discuss (a priori specified) SAP construction based on targeted learning, incorporating outcome-blind simulations to choose a best specification of the SAP. We also discuss various case studies including a Sentinel and FDA RWE demonstration project of targeted learning demonstrating SAP specification on real data.

---

Short Courses: 13 October 2026

SC1:	Improving Treatment Evaluation by Integrating Randomized Clinical Trials and Real-World Data: Causal Inference and Machine Learning Approaches (Full-Day)
Room	Hahia
Time	9:00 – 18:00
Instructor	Shu Yang, Professor of Statistics, North Carolina State University, USA
Moderator	TBD
Abstract	Motivation and Course Description: The 21st Century Cures Act, enacted in 2016, highlights the importance of precision medicine and the utility of real-world data (RWD) to accelerate the development and evaluation of new treatments. It encourages the FDA and other regulatory bodies to consider real-world evidence (RWE) alongside traditional randomized controlled trials (RCTs). RCTs are widely considered the gold standard for causal inference due to their internal validity. However, they often suffer from practical limitations such as restrictive eligibility criteria and limited sample sizes. In contrast, RWD offers broader population coverage and reflects clinical practice more realistically but is prone to confounding and other biases. Integrating RCTs with RWD offers the potential to combine the strengths of both data sources, achieving internal validity from RCTs and external validity from RWD, thereby enabling more generalizable, efficient, and timely treatment evaluations. This short course will introduce statistical frameworks and methodologies that facilitate the integration of RCTs and RWD to: Improve the generalizability of RCT findings to broader patient populations, Enhance the estimation of treatment effect heterogeneity for precision medicine, and Address challenges such as covariate shift, unmeasured confounding, and model misspecification through robust statistical and machine learning techniques. Simulated case studies and hands-on demonstrations using publicly available R packages will support the conceptual and methodological content. A foundational understanding of clinical trials and causal inference is recommended. Course Roadmap I. Introduction to Integrated Evidence Generation • Overview of RCTs vs. RWD and their complementary strengths • Motivation for data integration in regulatory and real-world settings • FDA guidance on the use of RWD for treatment evaluation • Research questions and causal inference roadmaps II. Generating Real-World Evidence through RCT-RWD Integration • Motivating example: non-small cell lung cancer • genRCT framework for integrating RCT and RWD • Empirical evaluation: simulation and case studies • R package: genRCT III. Assessing Treatment Effect Heterogeneity Using RWD • Key sources of bias in RWD • Classical and machine learning methods for treatment effect heterogeneity estimation • Integrative methods to improve robustness: o Test-then-pool strategies o Bias function modeling o Selective borrowing o Conformal inference o Randomization-based inference • Empirical evaluation: simulation and case studies • R packages: elasticIntegrative, inFRT Target Audience and Learning Objectives: Students, researchers, and professionals in academia, industry, or regulatory agencies who are interested in causal inference, health data science, and RCT/RWD integration. By the end of the course, participants will: • Understand the respective roles of RCTs and RWD in treatment evaluation and the benefits of integration • Be equipped to perform generalizable analyses of RCTs by leveraging RWD • Apply advanced statistical methods for estimating treatment effect heterogeneity using RWD • Gain hands-on experience with R packages developed for causal inference and data integration
Bios	Dr. Shu Yang is Professor of Statistics, Goodnight Early Career Innovator, and University Faculty Scholar at North Carolina State University. She earned her Ph.D. in Applied Mathematics and Statistics from Iowa State University and completed her postdoctoral training at the Harvard T.H. Chan School of Public Health. Her research focuses on causal inference and data integration with applications to comparative effectiveness research. She has made significant contributions to methods for missing data, spatial statistics, and real-world evidence, and has served as PI on multiple NIH, NSF, and FDA-funded projects. She is a recipient of the 2024 COPSS Emerging Leader Award. Website: https://shuyang.wordpress.ncsu.edu Dr. Yang has served as lead or co-lead instructor for numerous short courses at prominent professional conferences and academic events, including the Joint Statistical Meetings (JSM), New England Statistical Society (NESS), Duke Industry Statistics Symposium (DISS), Eastern North American Region (ENAR), International Chinese Statistical Association (ICSA), ASA Biopharmaceutical Section Regulatory-Industry Statistics Workshop (RISW), Society for Clinical Trials (SCT), and the Lifetime Data Science Conference (LiDS), as well as summer schools hosted by various universities. A detailed list of these engagements is provided below. Statistical methods for Randomized Clinical Trials with Information Borrowing from External Controls, Full-Day Short Course, Joint Statistical Meeting (JSM), Nashville, TN, USA. August 2–5, 2025 Statistical methods for time-to-event data from multiple sources: a causal inference framework, Full-Day Short Course, New England Statistical Society (NESS), New Haven, CT, USA. May 31, 2025 Statistical methods for time-to-event data from multiple sources: a causal inference perspective, Half-Day Short Course, Duke Industry Statistical Symposium (DISS), Duke University, Durham, NC, USA. April 9–11, 2025 Statistical methods for time to event data from multiple sources: a causal inference perspective, Half-Day Short Course, Eastern North American Region (ENAR) Spring Meetings, New Orleans, LA, USA. March 23–26, 2025 Statistical methods for time-to-event data from multiple sources: a causal inference framework, Full-Day Short Course, International Chinese Statistical Association (ICSA) Symposium, Nashville, TN, USA. June 16, 2023 Design and analysis of randomized clinical trials with real-world data using causal inference framework and Bayesian methods, Half-Day Short course, ASA Biopharmaceutical Section Regulatory-Industry Statistics Workshop (RISW), Rockville, MD, USA. September 27–29, 2023 Unveiling the Power of Real-World Data: A Causal Inference Framework for Designing and Analyzing Randomized Clinical Trials, 2nd CANSSI-NISS sponsored Health Data Science Workshop, Half-Day Short Course, Waterloo, Ontario, Canada. August 2, 2023 Statistical methods for time-to-event data from multiple sources: a causal inference framework, Full-Day Short Course, Lifttime Data Science (LiDS) Conference, Raleigh, NC, USA. May 31, 2023 Design and analysis of randomized clinical trials with real-world data, Half-Day Preconference Workshop on the Society of Clinical Trial (SCT), Baltimore, MD, USA. May 21–24, 2023 Improved causal inference combining randomized clinical trials and observational studies, Half-day lectures, Canadian Statistical Sciences Institute (Collaborative Research Teams) Summer School, University of Ottawa, Ontario, Canada. (Invited). July 5–8, 2022
SC2:	Applied Longitudinal Data Analysis with AI for Pharmaceutical Science (Full-Day)
Room	La Paz
Time	9:00 – 18:00
Instructor	Depeng Jiang, Professor of Biostatistics, University of Manitoba, Canada
Moderator	TBD
Abstract	This one day, AI-informed workshop bridges longitudinal data analysis with modern data-science approaches to advance biopharmaceutical statistics. The course offers an applied, hands-on introduction to modeling repeated-measures data in pharmaceutical contexts, with practical guidance on interpretability, reproducibility, and responsible AI in regulated environments. Through real-world public-health and pharmaceutical datasets, participants will translate theory into actionable analysis in R, from data preparation and model fitting to evaluation, reporting, and communication with non-technical stakeholders. Emphasis is placed on transparent workflows, robust handling of missing data and time-varying covariates, and the integration of AI-inspired techniques while maintaining interpretability and regulatory relevance. Learning objectives • Select appropriate longitudinal analytic approaches (e.g., linear and generalized linear mixed-effects models) for intervention evaluation in pharmaceutical settings. • Fit, compare, and interpret mixed-effects models in R (lme4/nlme, with emphasis on random intercepts/slopes and model diagnostics). • Manage time-varying covariates and missing data using principled strategies (e.g., multiple imputation, maximum likelihood approaches) within longitudinal analyses. • Evaluate model performance using suitable criteria (AIC/BIC, likelihood ratio tests, predictive accuracy, calibration) and perform model validation. • Communicate results to non-technical stakeholders with clear narratives, tables, figures, and reproducible code-based reports. • Develop reproducible workflows in R (RStudio, version control, and R Markdown) to ensure auditability and transparency. • Explore AI-informed enhancements to longitudinal analyses while preserving interpretability and regulatory suitability. Target audience • Biostatisticians, data scientists, and researchers in biopharmaceutical science • Clinical trial statisticians, pharmacometricians, epidemiologists, and health outcomes researchers • Regulators, policy researchers, and graduate students in biostatistics or pharmaceutical sciences • Professionals seeking practical, reproducible skills for analyzing longitudinal data in pharmaceutical contexts Notes for participants • The course explicitly ties longitudinal data analysis to AI-for-health science goals, with emphasis on reproducible, interpretable results suitable for biopharmaceutical practice and regulatory contexts. If you have questions about fit or prerequisites, please reach out prior to registration. Prerequisites and recommendations • Basic proficiency in R and RStudio (data frames, tidyverse familiarity recommended) • Familiarity with regression or introductory mixed-effects concepts is helpful • Prior exposure to simple longitudinal analyses is advantageous but not required Proposed benefits for ISBS and attendees • A competitively designed short course that aligns with ISBS goals of standardizing statistical practice in biopharmaceutical settings. • A practical, hands-on experience focusing on reproducibility, interpretability, and responsible AI in pharmaceutical data analysis. • Access to a high-quality, code-based teaching approach that participants can reuse in their own organizations and collaborations.
Bios	Dr. Depeng Jiang, PhD, is a full professor of Biostatistics in the College of Community and Global Health, Rady Faculty of Health Sciences, University of Manitoba. He also leads the Biostatistics Group within the George and Fay Yee Centre for Healthcare Innovation (CHI). Dr. Jiang has extensive experience providing statistical consulting to diverse clients, training researchers and students in statistical methodology, and applying advanced statistical learning methods to health services research and program evaluation. His work spans local, national, and international venues, with recognized contributions to methodological development and practical application in biostatistics. He is renowned for bridging theory and practice, fostering multidisciplinary collaborations with researchers and policymakers, and delivering statistically rigorous, accessible training. His expertise includes modern statistical learning methods, mixed-effects modeling, AI-informed approaches, and the translation of complex analyses into actionable health insights. He has delivered short courses across both academic conference and university campus both locally, national and internationally.   What participants will gain (outcomes and benefits) • A solid, implementable understanding of when and how to apply mixed-effects models to longitudinal pharmaceutical data. • Practical, hands-on experience building, comparing, and validating mixed-effects models in R using real-world data. • Competence in handling time-varying features and missing data in longitudinal analyses, with appropriate methodological choices. • Skills to translate statistical results into actionable public health and pharmaceutical insights for stakeholders. • Ability to present clear, reproducible results using well-documented R code, tables, and figures. • A reproducible analysis workflow in R and RStudio, including code-enabled reporting for auditability and collaboration. • Insight into integrating AI-inspired approaches with longitudinal modeling in a way that maintains interpretability and regulatory alignment.
SC3:	Implementing Bayesian Methodology in Clinical Trials: Navigating New FDA and EMA Guidance (Half-Day)
Room	Caracas
Time	9:00 – 12:30
Instructors	Yong Zhang, Associate Professor of Biostatistics, Indiana University, USA Ying Yuan, Bettyann Asche Murray Distinguished Professor, Department of Biostatistics, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA
Moderator	TBD
Abstract	In January 2026, the U.S. Food and Drug Administration (FDA) issued landmark draft guidance on the use of Bayesian methodology in clinical trials for drugs and biologics. This paradigm-shifting guidance signals a new era in which the Agency is increasingly open to Bayesian methods for primary analysis and as pivotal evidence for drug approval. Similarly, the EMA has issued a position paper echoing these sentiments while highlighting the key technical challenges involved. This half-day short course provides a comprehensive overview of Bayesian methodology in clinical trials, closely aligned with the latest FDA and EMA regulatory frameworks. In the first part of this short course, we will explore the strategic opportunities and practical challenges of integrating Bayesian approaches into drug development, emphasizing real-world applications. Through clinical case studies and hands-on demonstrations of open-source software, attendees will gain the technical skills necessary to implement these designs and the communication strategies required to engage effectively with regulatory authorities. In the second part of this short course, we will focus on Bayesian information borrowing methods for clinical trials, a key methodological area highlighted in FDA guidance. We will review several widely used approaches, including the power prior, commensurate prior, and robust meta-analytic prior. We will then introduce the dynamic information borrowing strategy, which is recommended in FDA guidance and adapts the extent of information borrowing based on the degree of prior–data conflict between current and historical data. Finally, we will present recent methodological extensions in this area, many of which have been developed through collaborations with industry partners. Key Topics Include: 1. Fundamental concepts of Bayesian statistics within a regulatory context. 2. Defining and calibrating Bayesian success criteria to ensure robust operating characteristics. 3. Strategic considerations for prior distribution selection and expert elicitation. 4. Advanced methods for information borrowing from historical and real-world data (RWD). 5. Best practices and common pitfalls in Bayesian regulatory submissions. 6. Bayesian information borrowing methods. 7. Dynamic information borrowing and its extensions. Learning Objectives:  By the conclusion of this course, participants will be able to: • Navigate the evolving regulatory landscape, including the features and challenges of contemporary Bayesian guidance. • Demystify the intricacies of Bayesian decision rules, calibration, and information borrowing, while critically evaluating their respective advantages and limitations. • Master the implementation of innovative Bayesian designs through hands-on training with freely available software, facilitating data-driven decision-making in drug development.
Bios	Yong Zang, PhD, is a Showalter Scholar Associate Professor in the Department of Biostatistics and Health Data Science at the Indiana University School of Medicine, where he also serves as Co-Director of Clinical Research for the Biostatistics and Data Management Core at the IU Simon Comprehensive Cancer Center. He received his Ph.D. in Statistics from the University of Hong Kong and completed postdoctoral training at The University of Texas MD Anderson Cancer Center. His research focuses on clinical trial design and health data informatics, with over 80 peer-reviewed publications supported by the National Institutes of Health, the Showalter Trust, and Eli Lilly. An expert in Bayesian adaptive clinical trial design, Dr. Zang has taught a dedicated course on the topic annually since 2016 and has delivered numerous short courses at major international conferences including AACR, ICSA, New England Statistics Symposium and DahShu Data Science Symposium. Ying Yuan, PhD, is a Bettyann Asche Murray Distinguished Professor and Deputy Chair in the Department of Biostatistics at the University of Texas MD Anderson Cancer Center. Dr. Yuan is an internationally renowned researcher in innovative Bayesian adaptive designs with over 300 publications. The designs and software developed by Dr. Yuan’s lab (www.trialdesign.org) have been widely used in medical research institutes and pharmaceutical companies. The BOIN design, developed by Dr. Yuan’s team, is a groundbreaking oncology dose-finding design that has been recognized by the FDA as a fit-for-purpose drug development tool. Dr. Yuan is a member of the FDA Advisory Committee for Genetic Metabolic Diseases, an elected Fellow of the American Statistical Association, and the lead author of two books: Bayesian Designs for Phase I-II Clinical Trials and Model-Assisted Bayesian Designs for Dose Finding and Optimization, both published by Chapman & Hall/CRC.
SC4:	Graphical multiple comparison procedures: Combining flexibility with optimality (Half-Day)
Room	Landres
Time	9:00 – 12:30
Instructors	Frank Bretz, Distinguished Quantitative Research Scientist, Novartis                                                                                           Yao Chen, Statistical Consultant, Novartis Dong Xi, Gilead Sciences
Moderator	TBD
Abstract	Addressing multiplicity is essential in confirmatory clinical trials to ensure valid statistical inference. Various multiple comparison procedures (MCPs) have been developed, including fixed-sequence, fallback, and gatekeeping procedures, which allow trialists to reflect the relative importance and inter-relationships of study objectives in a tailored multiple test procedure. This course focuses on graphical approaches which enable the construction and exploration of tailored MCPs to meet specific study objectives, such as comparisons of multiple treatments against a common control and multiple endpoint analyses. In these approaches, MCPs are represented by directed, weighted graphs, where each node corresponds to an elementary hypothesis. A simple algorithm then facilitates the sequential testing of hypotheses. Optimizing MCPs to maximize the probability of success is often a key concern for clinical trial teams. We will discuss clinically relevant objective functions for optimization and introduce an efficient algorithm, based on constrained nonlinear optimization, to identify optimal graphs. Case studies will illustrate the flexibility and practicality of these approaches in clinical trial settings. We will also introduce the graphicalMCP R package, which implements weighted Bonferroni tests, weighted parametric tests (accounting for correlations between test statistics), and weighted Simes’ tests. We also briefly consider power and sample size calculation. Example code for optimizing graphs will be demonstrated and shared, providing participants with practical tools to implement these methods effectively.
Bios	Dr. Frank Bretz is a Distinguished Quantitative Research Scientist at Novartis. He has supported the methodological development in various areas of pharmaceutical statistics, including dose finding, estimands, multiple comparisons, and adaptive designs. Frank is an Adjunct Professor at the Hannover Medical School (Germany) and the Medical University Vienna (Austria). Frank is a Fellow of the American Statistical Association. Frank Bretz has taught short courses at JSM, ASA BIOP RISW, and other statistical conferences. Dr. Yao Chen is a Statistical Consultant in the Advanced Methodology and Data Science group at Novartis. He has supported development and implementation of innovative statistical methodologies in multiple comparisons and treatment effect heterogeneity. Yao Chen has taught short course at ASA BIOP RISW. Dr. Dong Xi is a Senior Director in the Biostatistics Innovation Group at Gilead Sciences. He has supported development and implementation of innovative statistical methodologies in multiple comparisons, dose finding, group sequential designs, estimands and causal inference. He is an associate editor of Statistics in Biopharmaceutical Research and a committee member of the International Conference of Multiple Comparison Procedures. Dong Xi has taught short courses at ICSA symposium, JSM, ASA BIOP RISW, and other statistical conferences.
SC5:	Data Visualization with Applications to the Life Sciences (Half-Day)
Room	Paris
Time	9:00 – 12:30
Instructors	Richard Zink, Principal Research Fellow, JMP Statistical Discovery
Moderator	TBD
Abstract	(Bio)statisticians and data scientists are effective at utilizing cutting-edge methodologies to address the complexities of data in order to produce meaningful results. Despite this technical prowess, quantitative scientists struggle to communicate the story contained within these data to their non-statistical colleagues. Given the volume of data to review, the variety of analyses to perform, and the need to turn findings into actionable outcomes, it should come as no surprise that clear insight is often out of reach to the research team. The traditional means of data summary – static tables and listings – are ineffective for understanding the story hiding in plain sight; data visualization is the key to effective communication for the modern quantitative scientist. The goal of this short course is to describe visualization methodologies to aid in the interpretation and communication of data from applications in clinical trials and other life science topics. Categorical, continuous, and time-to-event endpoints are discussed, and numerous practical illustrations are presented. Learning Objectives 1. Describe the transition from traditional methods of data analysis and summary to visual approaches 2. Explore and interpret life science data using one or more data visualizations 3. Assess the strengths and limitations of various graphical techniques 4. Communicate the “data story” through numerous examples  Topics covered in lecture 1. Features of study design 2. Distributional summaries 3. Signal detection 4. Data integrity 5. Patient journeys 6. Time trends, time-to-event, recurrence 7. Clustering, correlation and co-occurrence 8. Subgroups 9. Meta-analysis 10. Benefit risk
Bios	Richard C. Zink is Principal Research Fellow at JMP Statistical Discovery and has spent 20+ years in and around clinical trials and medical product development. Richard is author, editor, and contributor to 10 books on statistical topics in clinical trials and clinical research. He holds a Ph.D. in Biostatistics from the University of North Carolina at Chapel Hill, where he serves as Adjunct Professor of Biostatistics and Adjunct Assistant Professor of Public Health Leadership and Practice. Richard was awarded the distinction of Fellow of the American Statistical Association in 2020. My research and day-to-day work since 2011 has included data visualization as a necessary component for effective communication of statistical concepts. I previously offered this course at the 2017 ASA Biopharmaceutical Regulatory-Industry Statistics Workshop, multiple DIA Annual Meetings (2018-2021), and as part of the 2025-2026 ASA Council of Sections Traveling Courses. This material will serve as the foundation for a course data visualization that I will teach in the Department of Biostatistics at the University of North Carolina at Chapel Hill starting in Fall 2026, and I am currently writing a book for CRC Press based on this material, tentatively due in August 2027.
SC6:	Causal AI with Targeted Learning (Half-Day)
Room	Caracas
Time	14:30 – 18:00
Instructors	Susan Gruber, Co-Founder, Targeted ML Solutions Alan Hubbard, Professor of Biostatistics and co-director of the Center for Targeted Machine Learning, University of California at Berkeley, CA, USA
Moderator	TBD
Abstract	Sound decision-making is built on transparent, interpretable evidence.  However, practical challenges can make it difficult to obtain efficient unbiased estimates from randomized controlled trial (RCT),  observational, or real-world data (RWD) studies. Enrollment difficulties and smaller than expected effect sizes reduce power  for answering the question of interest. Lack of baseline randomization or adherence to treatment and informative loss to follow-up can bias estimates. Targeted Learning (TL) unifies causal inference, machine learning (ML) and statistical theory to provide a framework for evaluating causal effects from data. Targeted maximum likelihood estimation (TMLE) is an estimator that incorporates powerful machine learning to address these challenges. The process is guided by  This short course will first describe the Targeted Learning Causal Estimation Roadmap that guides study design, analysis, and interpretation. The focus then shifts to theory and practice underling Targeted maximum likelihood estimation (TMLE) and super learning (SL). TMLE is a maximally efficient double robust estimator that relies on fewer assumptions than propensity score-based methods that are often used to analyze health care data. Unlike most machine learning approaches used in isolation, TMLE+SL provides accurate control of the Type-I error rate, and good confidence interval coverage.    Case studies of actual and simulated data in rare disease and oncology will illustrate the application of the TL roadmap for (1) study design and planning, (2) efficient and flexible estimation using TMLE and SL, (3) diagnostics that increase transparency, and (4) sensitivity analysis and interpretation of findings. Practical guidance on how to pre-specify the TMLE + SL procedure tailored to characteristics of the data and the estimation problem will be emphasized. The course will conclude with an overview of recent advances.
Bios	Dr. Gruber is co-founder with Mark van der Laan of Targeted ML Solutions, a company working to deliver TMLE+SL based software to expert and novice users. A former Harvard Medical School faculty and Reagan-Udall Foundation leader, Susan is a pioneer in targeted learning and real-world evidence, She released the first open-source package for  data analysis using TMLE in 2010, which now has over 100k downloads world-wide. Dr. Hubbard is Professor of Biostatistics and co-director of the Center for Targeted Machine Learning at UC Berkeley. His research focuses on the application of statistics to population studies with particular expertise in semi-parametric models and the use of machine learning in causal inference, as well as applications in high dimensional biology.
SC7:	Baseline Covariate Adjustment in Randomized Clinical Trials (Half-Day)
Room	Landres
Time	14:30 – 18:00
Instructors	Kelly Van Lancker, Assistant Professor of biostatistics, Ghent University and Vrije Universiteit, Brussels Bohdana Ratitch, Bayer
Moderator	TBD
Abstract	Background and Motivation: Baseline covariate adjustment in randomized clinical trials (RCTs) has received renewed attention following recent regulatory guidance, notably the 2023 FDA and the 2015 EMA guidelines on adjusting for covariates in RCTs. These documents encourage sponsors to adjust for baseline covariates associated with the outcome to improve statistical efficiency, while carefully distinguishing between conditional and marginal treatment effects. Despite this regulatory momentum, practitioners face a complex landscape of methods, each with distinct assumptions, strengths, and limitations — particularly regarding model misspecification, non-collapsibility, variance estimation, and performance in small samples. This short course aims to bridge the gap between the evolving methodological literature and practical usage in clinical trial settings. Course Overview: This course provides a comprehensive, practically oriented overview of baseline covariate adjustment methods for estimating treatment effects in RCTs. The course will cover the following topics: • Introduction and foundational concepts: The estimand framework, with a focus on the distinction between conditional and marginal treatment effects, and a causal roadmap for covariate adjustment; the role of randomization in providing validity versus balance, and why covariate adjustment can yield meaningful increases in statistical power; stratification versus analysis with covariate adjustment. • Estimation: We briefly introduce the estimation of conditional estimands using direct covariate-adjusted regression, highlighting its role as a foundational component for estimating marginal estimands. Building on this, we focus on standardization (G-computation) and then turn to Augmented Inverse Probability Weighting (AIPW) and Targeted Maximum Likelihood Estimation (TMLE), which are particularly useful in more complex settings where machine learning methods are employed. We also consider prognostic scores that incorporate historical data. • The phenomenon of non-collapsibility whereby certain effect measures such as the odds ratio yield numerically different conditional and marginal estimates even in the absence of confounding. • Special topics and practical challenges for marginal estimation such as small-sample performance and finite-sample corrections, multicenter trials, variable selection strategies, and considerations for group sequential designs. • Practical examples in R: Illustrations of key methods using R, demonstrating implementation and interpretation. • Review of available evidence on methods’ performance. The target audience includes statisticians and data scientists involved in the design and analysis of RCTs in pharmaceutical, biotech, and regulatory settings. Participants should have a working knowledge of generalized linear models, hypothesis testing, and the fundamentals of randomized clinical trial design. By attending this short course, the attendees will gain: (1) understanding of the motivation for use and benefits of covariate adjustment in randomized clinical trials (2) understanding of conditional versus marginal estimands and non-collapsibility; (3) practical knowledge of a range of covariate adjustment methods and their assumptions; (4) understanding of variance estimators aligned with regulatory expectations; and (5) awareness of methods’ performance trade-offs, including in small samples.
Bio	Prof. Kelly Van Lancker is an assistant professor in biostatistics at Ghent University and Vrije Universiteit Brussels. She obtained both her Master’s degree in Mathematics and her PhD in Statistical Data Analysis from Ghent University. Prior to her current position, she was a postdoctoral researcher at the Johns Hopkins Bloomberg School of Public Health. Her research focuses on causal inference in health sciences, particularly randomized clinical trials, with a strong emphasis on covariate adjustment to improve efficiency and power in treatment effect estimation. Kelly is widely recognized for her expertise in covariate adjustment. She has delivered over 15 workshops and 30 invited talks on the topic at international conferences, pharmaceutical companies, and academic institutions. Her work includes numerous peer-reviewed publications, and her PhD research specifically focused on causal inference for clinical trials, with covariate adjustment as a core component. Dr. Bohdana Ratitch works as Statistical Innovation Expert and is a Distinguished Science Fellow at Bayer. She has an advanced training in quantitative methods (applied mathematics and statistics and PhD with machine learning focus) and extensive applied clinical trials experience (over 20 years in pharmaceutical/biostatistics roles). Bohdana has a strong interest and expertise in causal inference and covariate adjustment (company-wide and cross-industry initiatives on covariate adjustment methods for RCTs; work spanning causal inference, estimands, missing data, and subgroup/heterogeneity evaluation). Bohdana regularly provides trainings on advanced statistical methods to colleagues at Bayer and presents at conferences and workshops. In December 2025, Bohdana was invited to provide a short course “Baseline Covariate Adjustment in RCTs” at the FDA Statistical Association 2025 Statistical Webinar series.
SC8:	Hands-on Introduction to Trial Design with East Horizon(TM) – Methodologies and Practical Applications (Half-Day)
Room	Paris
Time	14:30 – 18:00
Instructors	Martin Kappler, Director of Customer Success, Cytel Pantelis Vlachos, VP of Customer Success at Cytel
Moderator	TBD
Abstract	This short course provides a hands-on introduction to Cytel’s East Horizon platform for clinical trial design. Participants will work primarily with the Design and Explore modules to construct, simulate, compare, and evaluate candidate trial designs across a range of use cases. A shorter section will introduce enrolment and event prediction aas well as early development Go-NoGo decision making. The course will cover adaptive methods and their implementation in the software through practical examples from several case studies. Designs may include interim futility and efficacy stopping, sample size re-estimation, population enrichment, and multiplicity in designs with multiple endpoints and/or treatment arms, for event-driven trials as well as trials with continuous and binary endpoints. Emphasis will be placed on understanding how these features are specified, explored, and interpreted in practice. The course is intended for biostatisticians involved in clinical trial design who want practical experience with software-supported design evaluation. A certificate of participation will be provided.
Bios	Martin is Director of Customer Success at Cytel and worked for several years in Cytel’s Strategic Consulting unit as Expert Innovative Statistics Consultant. Martin is an expert in adaptive clinical trial design methodology and uses Cytel’s soſtware, including East Horizon, on a daily basis. Martin has over 25 years of experience and has been working as a trial, lead statistician and expert statistical consultant in different CROs, pharmaceutical companies and as independent consultant. Each participant will receive a 14-day East Horizon license to allow hands-on work during the course and independent reproduction of the examples afterward. Participants will gain practical experience in defining and parametrizing trial designs in East Horizon, exploring a broader design and scenario space through simulations, customizing design and analysis features through R-code integration, comparing candidate designs side by side, and interpreting of graphical and tabular outputs. Pantelis is VP of Customer Success at Cytel. Before joining Cytel in 2013, he was a Principal Biostatistician at Merck Serono and a Professor of Statistics at Carnegie Mellon University for 12 years. His research interests lie in adaptive designs, mainly from a Bayesian perspective, as well as hierarchical model testing and checking, although his secret passion is Text Mining. He has served as Managing Editor of the journal “Bayesian Analysis” as well as on editorial boards of several other journals and online statistical data and soſtware archives.