This week, the TMF community came together at the CDISC US Interchange, and one topic dominated the TMF sessions: the TMF Standard Model V1.
After almost three years of work, and contributions from more than 300 people across sponsors, CROs, sites and technology providers, the draft standard is now just days away from community review. The sessions gave the clearest picture yet of what V1 looks like, how it was built, and what happens next.
Our headline takeaway is simple. The work is real, the direction is clear, and the next six months are where the industry gets to shape the final result.
From a reference model to a standard
The TMF Reference Model gave our industry a shared and flexible way to organise trial documents. Organisations took it, adapted it and made it their own. That flexibility served us well, but it also meant that no two TMFs looked quite the same.
V1 is different by design. It is a standard: a common structure and a clearly defined language for the evidence that shows a clinical trial was conducted in line with regulatory requirements.
Donna Dorozinsky, who leads the V1 effort, described how the project began as a set of updates to version 3.3.1 and grew into something much bigger. The team realised they were not building another reference model but a standard built on two foundations: the evidence expectations of ICH E6(R3), and a digital future where structured, machine-readable information supports automation and AI.
That alignment is not a loose principle. Every record type in V1 is linked back to the ICH E6(R3) requirement it provides evidence for.
What V1 actually looks like
For those who work in the TMF every day, the draft will feel familiar and new at the same time. It is still a structure you can export to a spreadsheet, but the thinking behind it has changed.
Twelve zones. The existing zones remain, joined by a new Zone 12 for eClinical systems. Study-specific system records, such as validation and risk assessment records, finally have a proper home.
Record groups. Sections and artifacts give way to record groups, which bring together record types around a business process or purpose and can span multiple zones. Gill Gittens gave the example of qualifications and participant information, where the records you need for a review currently sit across several zones. Record groups pull them together.
Record types. Each record type represents one distinct piece of evidence. The aim is to remove the familiar question of “it could go here, or it could go there” that leads to misfiling. Through consolidation and smarter use of metadata, the number of record types is expected to be significantly lower than the figures discussed earlier in the project. For example, different types of informed consent form and different clinical committees will be described through metadata rather than separate record types.
Context instead of levels. Trial, country and site are being reframed as context: study, country or region, and site. Because record groups cut across the structure, V1 is no longer a simple hierarchy.
No more core and recommended. ICH E6(R3) talks about the essentiality of records, not core records. If a record is produced and needed as evidence, it belongs in the TMF.
Consistent definitions. Every record type has a purpose and description written in a consistent style, along with draft dating conventions so it is clear which date matters for each record. Unique identifiers are moving from three digits to five, with a reserved range for organisations that need their own record types.
Evidence over format. There is no catch-all record type for communications or notes to file. These are filed alongside the evidence they relate to. Meetings, on the other hand, are called out specifically, because meeting records are often critical evidence of oversight.
Terminology has also been aligned with ICH E6(R3) throughout, so you will see “study”, “participant” and “service provider” used consistently.
One standard, five subsets
One of the most significant achievements presented was the harmonisation of several separate models into a single standard. Paul Carter explained how V1 brings together five subsets:
- Drug and biologic
- Medical device
- In vitro diagnostics (IVDR)
- Real world evidence and non-interventional studies
- Investigator Site File (ISF)
Medical device and IVDR started as one subset and were later separated, as the teams recognised the evidence requirements are genuinely different. Rather than maintaining parallel models, each record type now carries attributes that show which subsets it applies to.
For sponsors and CROs running studies across multiple domains, this means one consistent structure. For sites, it means a much closer relationship between the ISF and the sponsor TMF. For technology providers and inspectors, it means one way of defining and retrieving evidence.
The timeline
The dates shared at the conference give the industry a clear path to release:
- 19 October 2026: the V1 draft is released for community review, including zones, record groups, record types, purposes and descriptions, draft dating conventions, subsets, ICH E6(R3) alignment and provisional metadata for informed consent, committees and investigational products
- 20 November 2026: community review closes
- Q1 2027: an updated draft, including metadata and controlled terminology, is released for public review
- April 2027: planned release of the TMF Standard Model V1
Once V1 is released, a change control board will manage its evolution, and an implementation guide will accompany the standard to set out how it works and the rules behind it. Training on the standard is also being developed to align with how CDISC supports its other data standards.
How to take part in the community review
Sydney Stamper, project manager for the initiative, walked through how the community review will work. The message was clear: this is not a formality, and the team wants practical, real-world feedback.
When the draft lands, the team’s recommendation is to:
- Familiarise yourself with the zone, record group and record type structure first
- Test it against the studies you actually work on
- Look beyond record names to context and subsets
- Submit one issue per feedback form, as many forms as you need
- Explain why something will not work, not just what
The feedback form will be shared with the TMF Reference Model general meeting mailing list, so now is the time to make sure your team is on it.
Community tools are also in development to help people explore the draft. A free, web-based navigator that visualises V1 across zones, record groups and record types was demonstrated during the session by our Head of Clinical Services, Rob Jones, who leads the community’s tools sub-team.
What this means for your organisation today
Some of the most useful moments came from the panel discussion, where the V1 management team answered questions directly. A few themes stood out.
Start by understanding, not mapping. There will be no official mapping guide from version 3.3.1, because organisations have customised the Reference Model in so many different ways. The panel’s advice was to digest V1 as the new way forward first. Where mapping is needed, the unique identifier is the anchor between the two.
You do not need to migrate everything. Moving ongoing studies is an investment and risk decision for each organisation. Ongoing studies can continue on the version documented in their TMF plan. A risk-based impact assessment is a sensible first step.
Use V1 as an evidence lens now. As Donna put it, inspectors are not going to cite you for not using the standard. They are going to ask to see your evidence. Reviewing V1 may highlight ICH E6(R3) evidence your studies are not yet collecting, and that is worth acting on whether you adopt V1 tomorrow or in two years.
Talk to your eTMF provider. Each technology provider will have its own timeline for supporting V1. The vendors have been aligned with the project throughout, and they are the best people to explain how and when their platforms will be ready.
Plan for change now. V1 will affect structures, SOPs, work instructions and training. Organisations that start change management planning early will find the transition far smoother.
What comes after V1
Paul Carter closed the conference by looking beyond release. V1 deliberately focuses on a stable, credible core that organisations can adopt and build on. The areas already identified for future development include:
- Expanding the metadata standard
- Regional subsets, developed with the growing regional communities in Japan, China, the UK, France, Italy, Denmark and beyond
- Connecting evidence to the processes that produced it, so the TMF can tell the story of a study rather than having it reconstructed
- A dedicated focus on what the standard needs to support AI
- Closer alignment with the wider CDISC standards landscape, including USDM
- Continued dialogue with regulators, including the EMA and ICH
The community itself is also evolving, with a federated model to support global governance, specialist teams and regional communities as the standard grows.
There was also news on leadership. Jamie Toth will take over as chair of the steering committee from the start of 2027, with Donna Dorozinsky as incoming chair. Our thanks go to Paul Carter for his leadership as chair, and to Karen Roy, one of the founders of the original Reference Model, for everything she has given the community.
Our view
The TMF Standard Model V1 is the most significant change to how our industry defines TMF evidence in over a decade. It is ambitious, but it is grounded in years of real-world use, built by the community, aligned with ICH E6(R3) and designed for the digital tools that are already changing how we work.
The draft opens for review on 19 October. We would encourage every sponsor, CRO, site and service provider to take part. The quality of the final standard depends on the quality of the feedback it receives.
April 2027 is closer than it sounds, and the organisations that start preparing now will be the ones best placed to benefit.
Arithmos supports sponsors and CROs with independent analytics and clinical services that provide objective insight into TMF quality, oversight and inspection readiness. If you would like support assessing the impact of the TMF Standard Model V1 on your organisation, get in touch with our Clinical Services team.
Talk to our Clinical Services team
Sources
Session content: TMF track sessions and panel discussion, CDISC US Interchange 2026 (October 2026)
ICH E6(R3) Good Clinical Practice Guideline (Step 4, January 2025)
https://database.ich.org/sites/default/files/ICH_E6%28R3%29_Step4_FinalGuideline_2025_0106.pdf
CDISC
https://www.cdisc.org
About Arithmos
With deep expertise in the life sciences industry, Arithmos supports companies in their digital transformation journey to achieve the best value from technology-enabled solutions and excellence in business operations.
The Arithmos team shares extensive knowledge of the full GxP regulated environment for both deliveries of services and technologies, and our domains range from Clinical Development, Quality & Compliance, Pharmacovigilance & Safety and Regulatory Affairs.
