ISPE Belgium Affiliate Technical Seminar "AI, shaping the future of pharma"

On November 20th, ISPE Belgium Affiliate vzw/asbl is organizing its annual technical seminar (full day seminar, including site visit) at Imec in Leuven about "AI, shaping the future of pharma".

Program:

08:00 - 09:00 : Registration and coffee

08:45 - 09:00 : Welcome (Wim Steenackers, President ISPE Belgium)

09:00 - 09:15 : Introduction (imec)

09:15 - 09:55 : Using new digital and AI capabilities in manufacturing operations: Pfizer Case Studies (Stefaan Vanderhaegen, Pfizer)

09:55 - 10:30 : Sensor Data Quality – The Lifeblood of Pharma 4.0 (Bert Baeck and Jonathan Boel, Timeseer and QBD Group)

10:30 - 10:45 : Coffee break with table top and networking

10:45 - 11:25 : AI as an enabler for fully automated cryo-em image processing for QC. (Carsten Jasper, Charles River Laboratories)

11:25 - 12:00 : The Future of Quality (Simon Cajot & Salman Saif , Argenx)

12:00 - 13:30 : Lunch with table top and networking

13:30 - 14:05 : Leveraging GenAI to accelerate deviation and investigation management (Quentin Hamouch, UCB)

14:05 - 14:40 : Accelerating drug development through microphysiological systems and AI (Peter Peumans, imec)

14:40 - 15:00 : imec site visit introduction

15:00 - 17:00 : imec cleanroom visit (3x20 persons - limited availability!)
In parallel: additional AI se cases by Elegent, Endexia and CoCEPTIO.

17:00 - 18:30 : Closing the day and network reception

Abstract details of the keynote speakers can be found in the Agenda section below.

Disclaimer: Photos and videos will be taken during this event and may be used for promotional purposes.

Info: Using new digital and AI capabilities in manufacturing operations: Pfizer has assessed and implemented several Case Studies where AI and new digital tools are used to optimise People, Process and Product aspects in pharmaceutical manufacturing.

Info: Sensor data is an indispensable factor in creating a digital business transformation in pharma. In fact, it is necessary to unlock the value that resides in many pharma 4.0 data-driven use cases (e.g. predictive maintenance, digital twins, predictive quality) that impact revenue, costs, risk, and customer satisfaction. However, it is the quality and reliability of this sensor data that will determine the success of these use cases and that can make or break many of your data-driven initiatives. Like the well-known saying goes: garbage in is garbage out. Problematic data can cause significant problems and hinder your digital transformation. Therefore, this presentation will cover why sensor data quality in pharma 4.0 is a crucial enabler for digital transformation at scale and will show its considerable impact through concrete examples. We will therefore cover how you can make sensor data a strategic asset within your organization and move from a reactive to a pro-active mindset.

Info: Adeno-associated virus (AAV) are a widely used vector within the field of ATMPs. Full/empty-ratio is a critical quality attribute that can be determined by various different analysis methods. In the past cryo-em was not a suitable method in a GMP-environment due to various reasons even though the method has significant potential for automation and high-throughput testing. This presentation showcases (using a real case study) how AI is utilised to automate this QC method including a discussion of some challenges regading compliance and how these are managed using synthetic data. The presentation will also touch some areas to consider from a validation perspective based on the new ISPE AI guide where presenter is one of the co-authors.

Info: The Digital Quality Strategy is a multi-year transformation designed to elevate quality operations through digital innovation, data integration, and AI enablement. Anchored to support the future of Quality to enable argenx’ 2030 vision, it aims to enhance visibility, predictability, and efficiency across quality systems and processes.

Key pillars include:
• Data-Driven Quality: Establishment of a validated Quality Data Warehouse and near real-time dashboards

• AI & Automation: Deployment of GenAI use cases for Q&A with procedural documents, procedural writing, Annual Product Quality Review

• Risk-Based Approaches: Automated risk assessments for audits and validations

• Learning Transformation: A shift to capability-based learning via the new Learning Management platform

• Inspection Readiness at all times: Centralized, inspection-ready data architecture and documentation

Info: The pharmaceutical industry faces a persistent challenge: the high cost, long timelines, and low translational success of drug development. Traditional preclinical models—whether animal-based or simpler in vitro systems—often fail to predict human responses, contributing to an attrition rate of up to 90% and average development costs exceeding $2.3 billion per drug candidate.

This presentation explores the use of advanced microphysiological systems (MPSs) combined with artificial intelligence (AI) to accelerate drug development and reduce risk of failure in clinical trials. MPS platforms that replicate human physiology using patient-derived cells have the potential to offer dynamic, multi-cellular, and longitudinally stable environments for testing drug efficacy and toxicity. These advanced MPS model will ultimately generate datasets used to train foundation AI models for predictive modeling of the effects of a drug in vivo.

Info: Visit of one of imec's semiconductor cleanroom facilities. The visit will be conducted in 3 waves, each consisting of a small group of 20 visitors. Tickets for this site visit are thus limited and distributed on a "first come, first served" basis. In parallel of the site visit, additional presentations are given in the main auditorium.

Info: Machine learning in direct compression: supercharging process and formulation design with quantitative tools - Elegent /// AI for Energy Sustainability in the Pharmaceutical Industry - Case GSK - CoCEPTIO /// Engineering Tool Asset Management - Endexia

Host or Publisher ISPE Belgium