The Data Transparency Autumn Event 2025 brought together 519 attendees spanning 140 companies and 25 countries.

With discussions on evolving policies, the ethical responsibilities of data sharing, the challenges of privacy and risk management, and AI and innovation, the event offered a rich mix of perspectives and expertise, with Q&A sessions fueling thought-provoking discussions.

All presentation slides and recordings are now available on the PHUSE website.

Continue reading for reflections from our event hosts.

Day 1: Clinical Trial Transparency – Policy, Practice & Ethics

– Devaki Thavarajah, Independent

Day one of the Data Transparency Event covered the evolving landscape of clinical trial transparency, focusing on new regulatory demands, compliance challenges, and strategic approaches to data sharing and disclosure. Speakers emphasised the need to balance regulatory expectations with practical implementation, while ensuring participant trust and scientific value.

Francine Lane – Beyond Compliance: The Transparency Shift in the GCP Guidance You May Have Missed

Francine Lane highlighted a significant shift in clinical trial transparency driven by ICH E6(R3) principles, which has expanded guidance to all human studies and requires plain language summaries (PLS) to enhance trust with participants. She also noted the ongoing challenges of disseminating results globally in a consistent and accessible manner.

Vidhi Vashisht – From Regulation to Practice: Assessment of Scientific and Plain Language Clinical Trial Results Summaries Under EU CTR 536/2014

Vidhi Vashisht presented an assessment of compliance with EU CTIS reporting requirements, revealing that out of 9,400 records, only a small proportion had results available. Many Phase II to IV trial summaries ending before September 2024 were found to be non-compliant and often failed to meet Annex IV standards, with missing trial information and incomplete result analyses.

Stephanie Fayolle & Toni Spears – Sanofi and TransCelerate: 10 Years of Advancing Clinical Data Sharing Using the DataCelerate® Solution

Stephanie Fayolle and Toni Spears described TransCelerate’s Clinical Data Sharing (CDS) initiative, which for the past decade has facilitated the responsible reuse of patient-level data through the DataCelerate® platform. This member-driven approach aims to accelerate scientific insights and improve drug development efficiency across participating organisations.

Abby McDonell – Transparency-by-Design: Best Practices in Writing Clinical Documents for Downstream Public Disclosure

Abby McDonell introduced the concept of transparency-by-design as a proactive framework for preparing clinical documents with disclosure in mind. She emphasised the importance of minimising non-essential data, normalising variables into predictable formats, and standardising structures in line with guidance such as ICH E3 to simplify review and reduce redaction.

Closing Discussion

The day finished with a Q&A session involving the event hosts and speakers, offering attendees the opportunity to explore the practical and regulatory challenges of clinical trial transparency in more depth.

Day 2: Data Privacy, Anonymisation & Risk Management in an Evolving Regulatory Landscape

– Abby McDonell, Privacy Analytics

The second day of the Data Transparency Event explored the evolving regulatory, operational and technical challenges of managing sensitive data in clinical research. Speakers highlighted new compliance requirements, advances in privacy-enhancing technologies, and the practical complexities of handling confidential and commercially sensitive information.

Rebecca Li & David Peloquin – Navigating Data Sharing Compliance Under the US DOJ Final Rule on Transfers of Sensitive Personal Data to Countries of Concern – Operational Impacts

Rebecca Li and David Peloquin examined the implications of the U.S. Department of Justice’s Final Rule on transfers of sensitive personal data to countries of concern. They explained how prohibitions and restrictions on specific transactions, particularly those involving bulk health and genomic data, would affect clinical research collaborations. The session outlined exemptions, compliance obligations, and how platforms such as Vivli had adapted processes to manage data access requests in line with the new regulations.

Parisa Osivand – Policy in Practice: Operationalising Transparency and Privacy Through Privacy-Enhancing Technologies (PETs)

Parisa Osivand presented research into the adoption of privacy-enhancing technologies (PETs) in industry in Canada. She highlighted barriers such as regulatory uncertainty, skills gaps, and limited return on investment, contrasting these with the potential benefits of approaches such as federated learning and synthetic data. Parisa’s talk emphasised the role of governance frameworks, including the NIST Risk Management Framework and ISO 42001, in bridging the gap between policy intent and operational practice.

Nirpal Virdee – “You Structured the Anonymisation – But What About the CI?”

Nirpal Virdee discussed the risks of focusing solely on anonymisation of patient data while overlooking confidential information (CI) such as manufacturing methods, vendor details and strategic narratives. He described how current ad hoc practices have led to inefficiencies, regulatory pushback and potential loss of intellectual property. Nirpal proposed a structured CI workflow built on governance, classification, monitoring and defensibility to improve trust, compliance and operational efficiency.

Hitesh Manchanda – Optimising CCI Identification To Reduce Rejection Rates and Establish Best Practices for Health Canada PRCI and EMA Policy 0070

Hitesh Manchanda shared an analysis of rejection rates for commercially confidential information (CCI) under Health Canada PRCI and EMA Policy 0070. Reviewing submissions from 2023 to 2024, he demonstrated how inconsistent justifications and reliance on public information had contributed to high rejection rates. By implementing structured decision trees, stronger justification frameworks and stakeholder engagement, rejection rates were significantly reduced. The session concluded that robust, proactive identification and justification of CCI supports regulatory acceptance.

Closing Discussion

The day concluded with an engaging Q&A session involving the event hosts and speakers. Attendees raised questions on the practical application of new regulations, the future role of privacy-enhancing technologies, and how organisations could best balance transparency with the protection of sensitive information. The panel provided clarifications, shared differing perspectives, and highlighted areas where further collaboration and guidance would be needed. This interactive session offered a valuable opportunity to consolidate the day’s discussions and connect regulatory insights with real-world operational challenges.

Day 3: AI & Innovation in Transparency and Communication

– Christa Polidori, Regeneron

Day three of the Data Transparency Event focused on advancing patient literacy, enhancing operational efficiency, and responsibly integrating artificial intelligence into clinical trial transparency. Presenters emphasised the importance of striking a balance between innovation and regulatory standards, human oversight and public trust.

Lauren Haggerty – PLS AI Automation: Challenges and Progress

Lauren Haggerty described Bristol Myers Squibb’s experience of integrating AI into the production of plain language summaries (PLS). She explained how AI reduced authoring time by drafting most of the content and improving consistency, though limitations remained with data tables and variable formats across studies. The human role shifted from writer to editor, ensuring patient-friendly language and regulatory accuracy. After a year of use, refinements included enhanced prompts, development of a glossary, and ongoing process improvements to expand patient access to understandable trial results.

Woo Song – Using AI in Production: Beyond Pilots

Woo Song examined the challenges of moving generative AI from proof-of-concept to production in regulated environments. He highlighted how early demos created excitement but masked the complexity of engineering robust, scalable and compliant systems. Key issues included non-deterministic outputs, extensive document processing, token limits, API throttling, cost, and speed constraints. Proposed solutions involved multi-agent frameworks, reflection agents, structured pre-processing, caching, and supervisor models, combined with strong security, audit trails and compliance safeguards. Woo’s talk emphasised the necessity of human-in-the-loop oversight to ensure scientific validity, regulatory adherence, and effective handling of ambiguity. It concluded that successful adoption requires rigorous architecture, expert-driven workflows and gradual scaling, balancing technological capability with regulatory and operational realities.

Devaki Thavarajah – Data Privacy and Data Sharing in Clinical Trials: A PHUSE Data Transparency Project Update – How To Enhance Health Literacy Surrounding Complex Data-Related Concepts

Devaki Thavarajah presented the PHUSE Data Transparency Working Group’s efforts to improve public health literacy on clinical trials, data sharing and privacy. She discussed common misconceptions and misinformation, particularly during the COVID-19 pandemic, and outlined how the team created animated videos and infographics to communicate complex concepts in accessible language. Feedback from patients and advocates informed the development of terminology, design choices, and character development, fostering trust and engagement. The project demonstrated the power of collaboration across stakeholders in promoting accurate, empathetic and inclusive education.

Zach Weingarden – How To Balance Risk and Utility in Data and Image Anonymisation for Clinical Trial Transparency

Zach Weingarden addressed how clinical trial data and images could be anonymised while balancing privacy protection with research value. He showed that risk scores measure re-identification potential and ensure compliance, while utility scores preserve analytical usefulness. The challenge was described as meeting privacy thresholds without overly degrading data quality. Traditional manual methods of assigning utility scores were contrasted with AI-driven approaches, which provided more objective and efficient weighting of variables. Finally, Zach demonstrated how interactive decision-making tools and an AI wizard help users explore anonymisation options, compare trade-offs, and adopt strategies that harmonise privacy with data value.

Closing Discussion

The day concluded with a Q&A session featuring the event hosts and speakers, which allowed attendees to delve deeper into the themes of AI, regulatory challenges and health literacy.

Looking Ahead

Data Transparency Pavillion: 22–26 March 2026

The Data Transparency Pavilion will debut at the US Connect 2026, 22–26 March, as a dedicated space for bold content, real-time discussion and collaboration with peers.

What is a PHUSE Pavilion™? Pavilion events are immersive, self-curated “conference within a conference” spaces that give participants the chance to explore a focused theme through interactive formats and expo-style setups, rather than traditional presentations.

Call for Speakers is open until 6 October 2025. Submit your abstract to contribute to these important conversations and share your expertise with the global transparency community.

Thank You to Our Sponsors

To join our below sponsors in supporting this new initiative, contact events@phuse.global.

Data Transparency Autumn Event: 15–17 September 2026

Save the date: the next Data Transparency Event will take place 15–17 September 2026.

Sponsorship opportunities are available, offering you a unique platform to showcase your commitment to advancing data transparency.

View the Return on Investment Summary and the prospectus for more information. The earlier you sign up, the more exposure you receive!