Over the last few days I’ve been seeing a few posts on LinkedIn that have been getting me a little frustrated. The reason? Poor use of terminology! Let me explain.
The Trial Master File – or TMF – is defined by the International Conference on Harmonisation (ICH) Good Clinical Practice (GCP) guideline as being “those documents which individually and collectively permit evaluation of the conduct of a trial and the quality of the data produced. These documents serve to demonstrate the compliance of the investigator, sponsor and monitor with the standards of Good Clinical Practice and with all applicable regulatory requirements.” There is something really important to note here. The TMF includes not only the records generated by and managed by the trial sponsor. It includes the records generated by and managed by the investigator. They are held separately from the sponsor documents but they’re still part of the TMF.
We sometimes use different terms for the investigator-managed records…. typically the Investigator Site File or Regulatory Binder. But listen up: they are part of the Trial Master File!! So, when you hear people talking about the TMF or when read guidance about the TMF, it is not just referring to the sponsor essential documents, it includes the ISF too.
Why is this important at the moment?
An industry group has developed a reference model for trial essential documents over the last 9 years called the TMF Reference Model. It was created with the full scope of the TMF in mind…. in other words, it includes not just sponsor essential documents but also investigator essential documents. Yes, that’s right! The TMF Reference Model already includes the ISF. It doesn’t just contain 2 site documents (the subject log and source data)….. it includes them all! It is not called the “Sponsor TMF Reference Model” or “ISF Reference Model” because that’s not the terminology that ICH GCP uses. ICH GCP refers to both of these components as “the TMF” so that’s what the reference model is called! I’m going to make this absolutely clear here so there can be no ambiguity. If someone tells you the TMF Reference Model does not include ISF documents, they’re wrong. Just take a look for yourself. Download the Reference Model here: https://tmfrefmodel.com/resources/. Open up the spreadsheet and take a look at column O. Where you see an “X” in the cell, these are ISF documents for trials on medicinal products. Look at column Q. Where you see an “X” in the cell, these are ISF documents for trials on devices.
So, here’s the next question. Is the TMF Reference Model perfect for the ISF essential documents? No it is not. This is recognised by the project team and one of the things we’ve committed to do is to get it more closely aligned with the requirements of the site. For example, for each artifact in the model we list individual document types that correlate with the definition and purpose of the artifact. These are currently mostly focused on the sponsor’s needs but it would not take a great amount of effort to also include specific site documents. A separate initiative (NOT part of the TMF Reference Model!) is current embarking on a project to create a completely separate ISF Reference Model. An initial draft has just been published and I can say with confidence that EVERY document is already included in the TMF Reference Model with full classification. So I ask myself, why is a completely separate reference model being created for Investigator Site Files when we already have a Reference Model that includes this…. it just needs improving. I also ask myself, which scenario would it be better to have: a single TMF Reference Model that meets the needs of both the sponsor and the site, or sponsor and site using two different models, with different terminology, different filing structure, and different metadata? Surely a single reference model would support efficient information exchange between sponsor and site and ensure we’re aligned with the way we name and file our documents?
Recent surveys indicate the TMF Reference Model is used on over 80% of clinical trials. The TMF Reference Model has been downloaded over >10,000 times from the TMF Reference Model website. Let’s work together to make this work as effectively for sites as it does for sponsors. How? Join the team and make it work!!
Need help developing your TMF processes and systems? I’ve worked on over 30 TMF projects. Give me a call! I’m here to help.
nb. For the avoidance of doubt, whilst I am a member of the TMF Reference Model Steering Committee, this is posted as a personal opinion and not as an official representative of the TMF Reference Model project team.