Program

Day 1 - 23 September 2025

Day 1 – 23 September 2026

8:30 - 9:00

Registration

9:00 - 9:10

Welcome and opening by the Scientific Committee

9:10 - 9:55

Leonhard Held

University of Zurich (Switzerland)

The Experimental Unit Information Index: Balancing Evidentiary Value and Sample Size of Adaptive Designs

9:55 - 10:15

Ondrej Libiger

Johnson & Johnson (USA)

Assurance-based sample size determination for replication studies under publication bias

10:15 - 10:35

Lisa-Maria Winter

Boehringer Ingelheim Pharma and Ulm University (Germany)

"ONITH: An Ontology-based Non-clinical Integration and Term Harmonization Pipeline facilitating Data Reuse in 3R"

10:35 - 11:00

Coffee break

11:00 - 11:20

Jonathan Rathjens

Evidenze Germany GmbH (Germany)

Bayesian prediction intervals for genetic toxicology assays using historical control data

11:20 - 11:40

Andreas Schulz

Sanofi (Germany)

Dynamic Borrowing for Virtual Control Groups: Adapting Robust Meta-Analytic Priors to Reduce Control Size and Enhance Precision

11:40 - 12:00

Guillemette Duchateau-Nguyen Paolo Piraino, Stefan Anca

Organon SRL (Romania)
Roche (Switzerland)
Universitat Oberta de Catalunya

Data-driven classification of endpoints patterns in toxicology studies: toward innovative study design using virtual control groups

12:00 - 12:20
MY TALK IN 5 MINUTES

Ludwig A. Hothorn

Leibniz University Hannover (Germany)

Is the Dunnett procedure, the standard method for analyzing chronic toxicity studies, truly appropriate for the purpose?

Odile Coudert Berthion

Ipsen Innovation (France)

Flexible Modeling of Repeated Tumor Volume Data Using Generalized Additive Mixed Models

Arno Strouwen

PumasAI (USA) and KU Leuven (Belgium)

InVitroInVivoCorrelation.jl: A Modular Julia Package for Level A IVIVC

12:20 - 13:35

Lunch

13:35 - 14:20

Kalliopi Mylona

King’s College London (UK)

Efficient experimental designs for pharmaceutical production

14:20 - 14:40

Ying Chen

KU Leuven (Belgium)

Optimal Design of Experiments for Powerful Equivalence Testing

14:40 - 15:00

Joff Jones

AstraZeneca (UK)

Optimal Design of High Dimensional Experiments

15:00 - 15:20

Beate Ehrhardt

University of Bath (UK)

Spatially Aware Plate Layouts (SAPL): A Web-Based Tool Integrating Experimental Constraints into Randomization for Optimal Plate Designs

15:20 - 15:40

Caroline Lévéder

Sanofi (France)

Process Characterization and associated DoEs considerations in R&D/CMC

15:40 - 16:10

Coffee break

16:10 - 16:30

Thomas de Marchin

Cencora - PharmaLex (Belgium)

Production: Optimizing Cell Culture Processes with Digital Twins and Hybrid Modeling

16:30 - 16:50

Tobias Eilert

Boehringer Ingelheim Pharma (Germany)

Building an E2E Digital Twin for Your Entire Production Process—As Simple as Stacking Lego Blocks

16:50 - 17:10

Thomas Zahel

Körber Pharma (Austria)

Physics Informed AI for CMC Applications

17:10 - 17:30

Birgit Niederhaus

Sanofi (Germany)

Statistical approaches for comparability of stability data

17:30 - 17:50

Marco Galliani

Politecnico di Milano (Italy)

Physics-Informed Statistical Learning for Accelerated Stability Studies

17:50 - 18:10

Martin Otava

Johnson & Johnson (Czechia)

Calibrating dissolution surrogate model via conditioning for real time release testing of drug product

18:10 - 18:35
MY TALK IN 5 MINUTES

Francisca Galindo Garre

Johnson & Johnson (The Netherland)

Predicting and Ranking Formulations in a Sparse Categorical DOE for Potency

Kshitij Singh

Leukocare AG (Germany)

A Two-Stage Bootstrap framework for long term stability modelling of biologics

Hugo Zuin, Marion Berger

Sanofi (France)

Determination of Acceptance Criteria and Designs for Analytical Method Validation to support ICH Q2(R2) implementation

Aniek Sies

Johnson & Johnson (The Netherland)

A statistical risk-based framework for sunsetting one-sided release tests: from limited data to regulatory approval

18:35 - 18:40

Conclusion of the day by the Scientific Committee

Day 2 – 24 September 2026

8:30 - 8:35

Introduction of the day by the Scientific Committee

8:35 - 9:20

Francesco Denti

University of Padova (Italy)

Bayesian Nonparametrics Methods for Complex Biomedical Data

9:20 - 9:40

Jean-François Michiels

Cencora - PharmaLex (Belgium)

Robust Bayesian model for the determination of the confirmatory cut point during the immunogenicity assessment

9:40 - 10:00

Maryam Kazemi Naeini

AstraZeneca (UK)

Bayesian dose–response mixed modeling: application to perturbation screening data

10:00 - 10:20

Alice Raillard

Sanofi (France)

Bayesian vs. Frequentist Methods: Implications for Preclinical Vaccine Research

10:20 - 10:40

Kim Ji Young

Takeda Pharmaceuticals (USA)

Comparability Assessment for Changes to Manufacturing Processes: Industry Recommendations on Bayesian Practices

10:40 - 11:00

Luwis Diya

Johnson & Johnson (Belgium)

PPQ Sampling plan based on Bayesian dynamic borrowing

11:00 - 11:20

Coffee break

11:20 - 12:05

Andreas Mayr

Philipps University Marburg (Germany)

Machine learning meets statistical modelling: boosting, simplicity and predictions

12:05 - 12:25

Christian Schmid

Roche (Switzerland)

Statistical Evaluation of Probabilistic Generative AI Models

12:25 - 12:45

Silke Werz

Roche (Switzerland)

Opportunities and Risks: Navigating the AI Revolution in CMC Statistics

12:45 - 13:05

Gianmarco Barberi

University of Padova (Italy)

Digital AI-driven methodologies to support and accelerate mAbs development in the biopharmaceutical industry

13:05 - 14:05

Lunch

14:05 - 14:25

Bruno Boulanger

SANAITIO (Belgium)

A Unified Bayesian Framework for AI Uncertainty and Reliability in Pharmaceutical Development

14:25 - 14:45

Hannah Lennon

Johnson & Johnson (France)

The evolution of role of (nonclinical) statisticians in an AI-era

14:45 - 15:45

Moderator:
Thomas Zahel

Körber Pharma (Austria)

PANEL DISCUSSION
Benefits and Risks of Agentic AI in CMC and Non-clinical Statistics

Panelists:
Christian Schmid

Roche (Switzerland)

Beate Presser

Boehringer Ingelheim Pharma (Germany)

Thomas de Marchin

Cencora - PharmaLex (Belgium)

Hannah Lennon

Johnson & Johnson (France)

15:45 - 15:50

Conclusion of the day by the Scientific Committee

15:50 - 16:10

Final coffee

Day 3 – 25 September 2026

8:30 - 8:35

Introduction of the day by the Scientific Committee

8:35 - 9:20

Raphael Gottardo

Translational Data Science - CHUV | Lausanne University Hospital (Switzerland)

Computational Innovations in Single-Cell and Spatial Transcriptomics for Translational Research

9:20 - 9:40

David Willé

GSK (UK)

Analysing data where groups are completely censored

9:40 - 10:00

Simone Modolo

Evotec (Italy)

Modeling Longitudinal Behavioural Data in Preclinical Home Cage Monitoring Studies

10:00 - 10:30
MY TALK IN 5 MINUTES

Alexandre Segers

Johnson & Johnson (Belgium)

Dimensionality reduction of proteomics and transcriptomics data with omicsGMF

Koen Van den Berge

Johnson & Johnson (Belgium)

Identification of cell type aggregates in spatial transcriptomics data

Michael A. Debebe

University of Cologne (Germany)

ZONAR: A Probabilistic Framework for Liver Zonation Mapping Integrating Spatial Omics and Anatomical Regularization

Gabriele Dolfi

Università di Padova (Italy)

Effects of misclustering in multi-sample, multi-group scRNA-seq studies: a stability-based approach

Yuyue Liao

Bayer AG Pharma (Germany)

Statistical Methodologies for Pragmatic AUC Estimation in Longitudinal Imaging Data

10:30 - 11:00

Coffee break

11:00 - 11:20

Ezgi Tanriver-Ayder

GSK (UK)

Making the Most of Indirect Evidence: IPD Network Meta Analysis for Sparse Preclinical Networks

11:20 - 11:40

Beate Presser

Boehringer Ingelheim Pharma (Germany)

EFSPI CMC Statistical Network Europe: Recent Achievements, Future Directions, and Position Papers on dissolution and ICHQ6

11:40 - 12:00

Bernard G. Francq

GSK (Belgium)

From Antiquated Frequentist Methods to the Bayesian Revolution: Driving Innovation and Approval in Non-Clinical Statistics

12:00 - 12:30
MY TALK IN 5 MINUTES

Yiming Peng

Genentech, Roche Group (USA)

From Reactive to Proactive: Scaling AI for Product Health

Marco Brendolan

University of Padova (Italy)

Hybrid modelling for improved prediction of intestinal solubility

Edoardo Tamiazzo

University of Padova (Italy)

Improved metabolic understanding in biopharmaceutical product development through hybrid modelling

Swetlana Berger

UCB Biosciences GmbH (Germany)

Applying DoE to Bioassays: Faster Optimization, Relevant Performance, Better Decisions

Michele Firmo

Roche (Switzerland)

Statistically Derived MAD for TOST: A Combined Accuracy and Precision Approach for PSD Co-validation Studies

12:30 - 12:45

Closing remarks and farewell by the Scientific Committe