Every time I explain that our eISF and eTMF are connected, people look at me like I just said something revolutionary.
I didn’t. I said something obvious.
And that gap, between what is obvious and what the industry has normalised, is exactly what this post is about.
A quick reminder of what these things actually are
The Trial Master File is the sponsor’s responsibility. It holds all the essential documents that prove a trial was conducted in accordance with ICH GCP and applicable regulations. Protocol versions, informed consent templates, CRA visit reports, regulatory approvals, and hundreds of other records.
The Investigator Site File is the site’s responsibility. It holds the same essential documents from the site’s perspective. The signed agreements, the delegation logs, the training records, the local IRB approvals, the drug accountability logs.
They are not the same file. They never were. The regulations are clear on that. ICH E6(R3), which came into full effect in July 2025 across EMA territories and September 2025 under FDA, confirms it: the TMF belongs to the sponsor, and the site’s portion of the TMF, commonly managed as an ISF, belongs to the investigator.
But here is the part people forget: they are about the same study. They are documentation of the same events, the same patients, the same protocol, the same investigational product.
They are two halves of the same truth..
How we got here: history, not regulation
The separation between TMF and ISF was never a regulatory requirement. It was a practical reality of paper.
When clinical trials were run on paper binders, a sponsor could not physically hold the site’s documents. The site kept their binder. The sponsor kept theirs. CRAs travelled between them, manually verifying that the two sets of records were consistent.
Then the industry digitised, but it digitised in silos. Sponsors implemented eTMF systems. Sites got eRegulatory binders. And the two systems, much like the paper binders that preceded them, did not talk to each other. They lived in different platforms, maintained by different teams, accessed through different logins, reconciled through email chains, shared by attachment, and audited by CRAs who printed documents at one location and compared them to documents at another.
We replaced paper binders with digital binders. We did not replace the problem.
As Doug Bain wrote in a widely-read LinkedIn analysis in late 2025: “The separation we live with today isn’t a matter of compliance; it’s a matter of history, habits, and technology choices.”
He is right. And the industry knows it..
The real cost of keeping them apart
Let me be specific about what “keeping them apart” actually costs in practice.
It costs CRA time. A meaningful portion of every monitoring visit, estimates from industry surveys consistently put administrative overhead at 30-40% of total CRA time, is spent on tasks that exist purely because the two document sets are not connected: chasing the latest version of a document, reconciling what the sponsor holds against what the site holds, sending emails to request access to files that should already be visible.
It costs inspection readiness. Record-keeping and essential document management is, year after year, the single most cited finding category in GCP inspections. The MHRA has reported sponsor inspection cohorts where more than half of all inspected organisations received at least one critical TMF-related finding. The FDA’s ALCOA+ compliance gaps appear in approximately 35% of clinical investigator inspections. These are not obscure edge cases. This is the norm.
It costs visibility. When sponsor documents live in one system and site documents live in another, nobody has a complete picture of the study at any given moment. Sponsors estimate what sites have. Sites guess what sponsors need. CRAs act as the manual bridge between two incomplete datasets.
It costs time-to-market. Every day a trial runs at risk of an inspection finding is a day the sponsor is exposed to potential delays, additional queries, or remediation requirements. At a Phase III daily cost of roughly $55,000 per day, the administrative inefficiency created by disconnected document systems is not a minor inconvenience. It is a material business risk.
What ICH E6(R3) actually demands
The finalisation of ICH E6(R3) in 2025 is not just a compliance update. It is a signal.
The new guideline specifically emphasises risk-based approaches to trial management, calls for version-controlled and traceable documentation, and places greater weight on **the integrity of records across the entire trial ecosystem,** sponsor and site alike. It does not mandate integrated systems. But it makes the case for them implicitly: if you are expected to demonstrate contemporaneous, traceable, complete records across all stakeholders, managing those records in disconnected silos is not a risk-based approach. It is a risk-creating one.
Regulators are not going to tell you to buy a particular system. They are going to inspect your trial and ask you to demonstrate that your records are complete, current, and consistent. If your answer involves explaining how your TMF team emails the site coordinator every time they need a document, that answer is not going to age well.
The objection I hear most often
“But site documents cannot be visible to the sponsor. And sponsor documents cannot go to sites.”
Correct. And this is not an argument against integration. It is an argument for smart integration.
Proper permission architecture means site users see site documents. Sponsor users see sponsor documents. Shared documents, things like the current protocol version, the investigator brochure, the delegation log and training log, are visible to both parties because they are supposed to be.
This is not technically complicated. It is a matter of building a system where the data model reflects the actual regulatory structure of the study: each party owns their documents, both parties contribute to the same body of evidence, and nobody sees what they should not.
In Menken Trials, this is exactly how it works. The eISF and eTMF are connected. Not merged, connected. Site documents stay at the site. Sponsor documents stay with the sponsor. But both exist within the same compliance framework, with the same automated checks running across both, and with visibility that reflects the actual relationships in the study.
No emails requesting the latest delegation log. No CRA manually comparing two document sets at the end of a visit. No sponsor guessing whether the site’s records match what they hold centrally..
Why this still surprises people
I have thought about this a lot.
Part of it is habit. The industry has managed TMFs and ISFs separately for so long that the separation itself has become invisible: assumed, unquestioned, treated as a regulatory requirement when it is not one.
Part of it is vendor inertia. The eTMF market grew up around sponsor needs. The eISF market grew up around site needs. The companies serving each market had little incentive to build bridges between them, the integration was technically harder, commercially ambiguous, and required convincing two separate buyer personas simultaneously.
Part of it is risk aversion. Clinical trial professionals are trained to be cautious. When something has worked, even imperfectly, for decades, proposing a change to it feels riskier than it actually is.
And part of it is simply that nobody had built it properly yet.
The question worth asking
If you ran your trial today and a regulator asked to see a complete, current, consistent picture of all essential documents, sponsor-held and site-held, how quickly could you produce it?
If the answer involves exporting files, sending emails, waiting for responses, reconciling two separate systems, and hoping nothing has changed since the last monitoring visit, then the gap between what you have and what you need is not a technology gap.
It is a decision gap.
The technology exists. The regulatory framework supports it. The case for it is not subtle.
The only thing keeping your TMF and your ISF in separate rooms is the assumption that this is how it has always been done.
It does not have to be.
See What Menka AI Would Find in Your Study
We offer a 15-minute Menka AI session, no preparation required on your side, no commitment. We run a targeted check against your study documentation and show you exactly what Menka AI surfaces.
Most teams find it clarifying. Some find it uncomfortable. Either way, it is far better to know now than during an inspection.
Anna-Liisa Parts is co-founder of Menken Trials, an eTMF, eISF, and AI-powered compliance platform built for inspection-ready clinical studies. She spent her career at IQVIA as a CRA and clinical lead before building the tool she always wished existed.
If this resonated, follow for more on clinical trial compliance, TMF best practices, and what it actually takes to run an inspection-ready study.
