Rima Izem, Director in Biostatistics, Novartis

Dr Rima Izem is Real World Evidence Scientific Lead in the Advanced Quantitative Sciences group at Novartis in Basel, Switzerland. She has nearly 20 years of experience in pharmaceutical research, including ten years at the US Food and Drug Administration and prior academic appointments. Her work has supported benefit-risk evaluation and regulatory decision-making across multiple therapeutic areas and throughout the product lifecycle, from clinical development to the post-authorization setting. Her recent interests include the application of causal inference methods and hybrid study designs that integrate evidence from clinical trials and real-world data sources, particularly in rare events and rare diseases.

This presentation will provide an introduction of safety evaluation in clinical trials. This will begin by outlining the key scientific objectives that underpin these evaluations, including the protection of trial participants and the evaluation of a treatment’s benefit-risk profile. Then, the presentation will introduce some safety- specific terminology, concepts, and/or principles and illustrate these concepts on a toy example. The presentation will put recent methodological development, existing methods and challenges, in a historical context of evolving legislation, guidelines, and advances in therapeutic development.

Matthias Trampisch, Expert TAM Statistician, Boehringer Ingelheim

Dr. Matthias Trampisch is an Expert TAM Statistician and independent statistician (iSTAT) within the Independent Statistical Analysis Team (iSAT) at Boehringer Ingelheim, where he has been supporting independent statistical activities for almost 10 years now. He provides statistical support to Data Monitoring Committees (DMCs) and conducts independent unblinded analyses during ongoing clinical trials, including assessments of anticipated serious adverse events (SAEs) and emerging safety signals. His work focuses on supporting FDA IND safety reporting requirements while maintaining trial integrity through robust independent review processes. He played a key role in establishing Boehringer Ingelheim's statistical framework, standards, and processes for independent analyses and serves as an R Governance Process Owner (R GPO), supporting the compliant and sustainable use of R across the organization.

Beyond Individual Cases: Operationalizing FDA IND Safety Reporting for Anticipated Serious Adverse Events

The FDA requires expedited reporting of suspected unexpected serious adverse reactions (SUSARs) during clinical trials, including the assessment of anticipated serious adverse events (SAEs). Implementing these requirements – specifically for the assessment of anticipated events - can be challenging, particularly when unblinded analyses are needed to evaluate treatment group imbalances while preserving trial integrity.

This presentation discusses practical approaches for conducting independent unblinded safety analyses, comparing external CRO models with internal Independent Statistical Analysis Teams (iSATs). Key topics include governance, Safety Surveillance Plans, trigger-based and periodic review strategies, output validation, and maintaining operational independence.

Matteo Pedone, GSK

Matteo Pedone is a Principal Statistician in GSK Vaccines, where he has been part of the Quantitative Sciences and Innovation methodological team since February 2023. His current interests center on mediation models, predictive models, and futility rules for clinical trials. He previously supported protocol and statistical analysis plan development for early-phase vaccine and HIV trials within GSK's Bacterial Vaccines biostatistics group, and contributed to work on safety signal detection, Bayesian causal mediation models, and sample size re-estimation. He holds a Ph.D. in Statistics from the University of Florence, where his research focused on covariate-dependent Bayesian models for heterogeneous populations. His broader work spans Bayesian nonparametrics, variable selection, and causal inference, with applications to microbiome data, adaptive clinical trials, and precision medicine.

From Theory to Practice: Implementing BDRIBS for Blinded Safety Surveillance

This presentation uses a realistic toy example to demonstrate how sponsor project teams can design and implement Bayesian Detection of potential Risk using Inference on Blinded Safety Data (BDRIBS) as a blinded safety surveillance strategy. Attendees will gain an understanding of the statistical principles underpinning BDRIBS, explore key implementation considerations, and discuss practical challenges and opportunities when deploying the approach in clinical development programs.

We hope to spark a discussion on barriers to implementation, and would welcome attendees sharing their own experiences with blinded safety surveillance methods.

Alex Spiers, GSK

Alex is a statistician at GSK's Quantitative Sciences Innovation group, working as part of a team that develops novel methods and creates tools to support the implementation of high-impact statistics methodology in clinical trials. His methodological work spans multiplicity strategies, trial design, and he maintains the open-source {multigrain} R package for optimising graphical multiple testing. Alex began his career teaching science in London comprehensive schools through the Teach First programme. After completing his MSc in Applied Statistics from Birkbeck, University of London, he worked briefly as a Data Science Consultant at DecisionLab, developing machine learning and simulation models for clients in government and the water industry. In 2019, he began his PhD in Epidemiology and Biostatistics at Imperial College London, focusing on adolescent mental health. His career focus in clinical trials began in 2022 when he joined the clinical trials unit at King's College London's Institute of Psychiatry, Psychology and Neuroscience as a research associate statistician, before joining GSK in 2023. In addition to his work at GSK, Alex remains dedicated to STEM outreach, both as a volunteer giving talks and workshops to schools and as a member of the PSI Careers Schools committee.

From Theory to Practice: Implementing BDRIBS for Blinded Safety Surveillance

This presentation uses a realistic toy example to demonstrate how sponsor project teams can design and implement Bayesian Detection of potential Risk using Inference on Blinded Safety Data (BDRIBS) as a blinded safety surveillance strategy. Attendees will gain an understanding of the statistical principles underpinning BDRIBS, explore key implementation considerations, and discuss practical challenges and opportunities when deploying the approach in clinical development programs.

We hope to spark a discussion on barriers to implementation, and would welcome attendees sharing their own experiences with blinded safety surveillance methods.