Sarah Dolan was diagnosed with young onset Parkinson’s Disease 7 years ago. After leaving a 30-year career in the biopharma industry she has participated in multiple clinical trials, is a current member of Critical Path to Parkinson’s Endpoints Team, a current Consumer Representative for the FDA PCNS Advisory Committee, and is an active Ambassador for the Davis Phinney Foundation. Sarah lives in a cabin on a river in Cody, Wyoming and loves cycling, her horses, and spending time with her husband and three children.

There was no active community of TMF’ers in Japan, and each company has been struggling to manage challenges in TMF operation. To address these aspects, CDISC Japan User Group (CJUG) launched a TMF team in January 2025, bringing together 20+ TMF’ers from pharma, CRO, tech service provider, consultant and academia. We focus on 4 key challenges including ICH-E6(R3) vs TMF, Sponsor-CRO collaboration, TMF in academia and TMF RM penetration in Japan.
One of the practical challenges for Japanese pharma/CRO is managing Japan-specific regulatory requirements while complying with global standards from EMA, MHRA, and FDA. This presentation highlights a case study on safety information documentation in Japan, which uniquely requires IRB review for study continuation rather than just PI’s review as in EU/US. This process creates additional documentation and potential duplication, so CJUG TMF team aims to streamline TMF management practices while fostering greater integration with the global TMF community.

• Matt Lowery, MGH
• Colleen Butler, Syneos
• Steph Viscomi, Apellis

In this session we will introduce a proposal developed under the TMF RM V4 initiative to define the core record types for computerized systems used in clinical trials. These records are those that sponsors and their service providers should consider essential for inclusion in the Trial Master File (TMF), supporting both oversight activities and regulatory submissions.

The proposal reflects inputs from a multidisciplinary consultation and is aligned with the guiding principles of the TMF RM V4 initiative and CDISC standards. It is designed to meet known regulatory expectations while offering a practical and adaptable approach for diverse sponsor organizations and trial needs.

Additionally, we will present a template for documenting the trial system inventory, aimed at enhancing inspection readiness and facilitating efficient retrieval of both trial-specific and enterprise-level records.

• Jamie Toth, BeOne Medicines USA, Inc.
• Bryan Souder, Merck
• Deb Wells, Novartis

Is it possible to interact with the CDISC Library API using natural language in plain English? The answer is yes. This paper introduces a web application enabling users to query ADaM/SDTM variable information(metadata and codelist) through natural language by leveraging AI-driven Natural Language Processing (NLP).

The underlying logic involves initially creating a SAS macro to extract ADaM/SDTM variable and codelist information from the CDISC Library API, followed by converting this SAS code into Python while preserving its original functionality. This Python script serves as the application's core logic, interfaced with AI to manage both input and output. Code is open-sourced via GitHub. During this conference, the step by step implementation and practical applications of CDISC Genius will be demonstrated.

This poster addresses the increasingly important requirement for anonymized clinical trial data for research by academia, regulatory evaluation, and training of AI models. The FDA, EMA, and Health Canada, among others, require traceability of data, creating a conflict between protecting patient privacy and regulatory requirements. We propose an approach utilizing CDISC SDTM, ADaM, and value harmonization techniques. The regulatory framework created through laws like GDPR and HIPAA disallows the reuse of data but enables its reuse in proper anonymization. Our suggested approach categorizes variables according to EMA Policy 0070 as direct identifiers, quasi-identifiers, and non-identifying variables. Generalization, suppression, and perturbation are approaches used to handle these variables. Quasi-identifiers are assessed based on how easy it is to replicate, differentiate, and know them. The poster as a whole contributes practical guidance on anonymizing data to satisfy privacy and regulatory needs, enabling responsible worldwide data sharing.

Digital Health Technologies (DHTs), such as wearable sensors, offer real-time, quantitative insights into patient activity, enhancing the evaluation of therapeutic efficacy in clinical trials. However, their integration presents operational challenges, including device misassignment, incomplete data uploads, and unfamiliarity among sites and participants. These issues can lead to data loss and reduced data quality. To address this, it is key to implement automated quality control checks and leverages cloud-based APIs for secure data access. CDISC-compliant data entry (e.g., DXSTDTC, DXENDTC, DXTRT, SPDEVID) anchors DHT data to participant timelines, enabling cross-validation and query generation. Standardized data models allow for reusable programmatic checks, improving efficiency and consistency across studies. This approach ensures robust data monitoring, preserves statistical power, and supports the generation of novel endpoints. By addressing DHT-specific complexities early in study design, sponsors can enhance data integrity and optimize the value of digital endpoints in modern clinical research.

Behind every high-quality, CDISC-compliant SDTM dataset lies a team of data-driven heroes: Data Managers, Standards Managers, Clinical Programmers, and Biostatisticians, each bringing their own unique skillsets and expertise to the fight for clean, conformant data.

Our heroes may hail from different domains, but they share a common trait: a working knowledge of CDISC standards and familiarity with the tools of the trade. This foundational knowledge becomes their secret weapon to enabling smarter, more collaborative workflows. Understanding the strengths of each role unlocks opportunities to rethink traditional processes, distribute tasks more effectively, and innovate with workflows that leverage each team member’s unique superpowers.

This poster will present our SDTM Squad, outlining the signature skills each role possesses, supercharged by emerging contributions of AI. By highlighting the strengths of this dynamic team, we’ll demonstrate that with the right strategy and collaboration, regulatory compliance can be both faster and more heroic. 

As a Canadian academic research institute, Population Health Research Institute (PHRI) has historically utilized multiple clinical trial management systems (CDMS) across studies and clinical trials. The challenge is to create harmony among the multiple CDMS in study design, Case Report Forms creation (CRFs), variable names, data types, formats and codes to make optimized studies and clinical trials. Each CDMS system has unique preferences in study design and programming. The PHRI CDASH Working Group was established to manage the standardization process and to create a custom PHRI CDASH library.

The poster will explain the process of creating the PHRI CDASH-P library and provide examples of the challenges met and solutions found for applying CDASH to clinical trials across different studies and multiple CDMS in an academic organization. Currently the standardization applies to the CDMS DFnet DFdiscover and AnjuEDC TrialMaster.  Our goal is to expand the library for RedCap and other systems.

Processing Trial Design domains in clinical research is a complex and time-consuming task, primarily because much of the required information is embedded within unstructured documents such as clinical trial protocols, rather than being derived directly from subject-level data. This Poster explores how Artificial Intelligence (AI) and Machine Learning (ML) techniques can be leveraged to automate and streamline the generation of Trial Design datasets. By extracting relevant data from protocols, Case Report Forms (CRFs), and other Study Data Tabulation Model (SDTM) domains, AI/ML models can significantly reduce manual effort and improve accuracy. We propose an ensemble approach that combines multiple models to classify and interpret various sections of the protocol, enabling precise extraction of key elements necessary for constructing Trial Design domains. Furthermore, we highlight the integration of the Unified Study Definition Model (USDM) as a foundational framework to standardize and facilitate the creation of these domains.

Autumn is the perfect time to focus on the changing landscape of TMF culture.
 

Let go of outdated habits and embrace a new season of consistency and shared purpose for all.

Key Themes:
Harvesting Data
•    Conduct thorough TMF reviews to gather valuable insights.
•    Use this data to improve processes and ensure compliance.
Turning Over a New Leaf
•    Refresh and update your TMF Plan and workflows.
•    Leave behind inefficient habits and adopt best practices.
Storing for Winter
•    Focus on inspection readiness to prepare for future audits.
•    Ensure your TMF is complete, organized, and ready for any challenge.
“This Autumn, let’s embrace the season of change and prepare for a future of success!”

Effective management of the Trial Master File (TMF) is essential for ensuring regulatory compliance and maintaining inspection readiness. However, clinical research operational teams frequently encounter challenges related to clarity, confidence, and consistent engagement in fulfilling their TMF-related responsibilities.

This poster presents how the design and implementation of a TMF Reference Model tool, built in Smartsheet, has empowered multidisciplinary operational teams to gain a clearer understanding of TMF zones and artifacts, and clarify responsibilities and expectations.

 For a clinical pharmacology organization operating within short, high-intensity timelines, the development of a streamlined, practical resource to support TMF comprehension and process adherence was crucial. This targeted approach has driven stronger engagement, improved role clarity, and heightened accountability, ultimately enhancing operational efficiency in achieving inspection readiness, and ensuring the delivery of complete and accurate final TMFs.

Maintaining a state of Trial Master File (TMF) inspection readiness, particularly pertaining to completeness, timeliness, and document quality, often poses a challenge for clinical researchers. Through a collaborative effort, Gilead Sciences and Trialwise piloted a TMF inspection readiness program that assessed Clinical Data Science (CDS) TMF documents for quality and completeness in real time and utilized TMF quality metrics to monitor trends for process improvement. Over the course of the program, 21 studies were included. Elements of the program included data collection, training, document review, process development, workflow optimization, and leadership support. Following receipt of a study demographics questionnaire, a documentation specialist (DS) conducted a retrospective review of all previously filed documents for critical document issues. Next, the filers routed new documents to the DS for a “real time” quality control review using a specified workflow in the eTMF platform, where findings were resolved prior to eTMF filing. Outside of the reviews, the DS periodically provided guidance and support to document filers. Over the course of the program, quality metrics improved up to 6.40%. For previously filed documents, TMF completeness increased 19.51% across all artifact types. Review findings continue to decrease over time as process improvement measures are implemented to guide filers on avoiding common findings. In summary, this program has shown increasing improvement across all quality metrics (completeness, timeliness, and quality) since its inception in 2023.

• Rob Jones, Cencora
• Guillaume Gerard, Agatha
• Eleanor Hewes, Syneos