The Trial Master File (TMF) & TMF Reference Model
The Trial Master File (TMF) is the collection of essential documents that allows the conduct of a clinical trial to be reconstructed and evaluated, and the sponsor’s and investigator’s compliance with Good Clinical Practice to be demonstrated. It is the trial’s evidentiary backbone: if it is not in the TMF, for regulatory purposes it did not happen. This page covers what the TMF is, how it differs from the site-side file and the operational systems around it, and the industry Reference Model that structures it — including the custodianship change that current training content still gets wrong. The oversight duty the TMF evidences is the [sponsor oversight](/topics/sponsor-oversight) explainer.
What an explainer is not
A topic explainer is SPEQ’s synthesis of what a practice involves, cited to the standards that govern it. It does not reproduce their text, and it does not determine which of them apply to your product or process.
[ POSITION IN THE FRAMEWORK ]
7 DIMENSIONS · 20 LINKSThe Trial Master File lets a trial be reconstructed and its GCP compliance proven; what counts is being complete, timely, and quality-controlled throughout, not 'complete at close-out', structured by the CDISC-stewarded TMF Reference Model.
06 · QUALITY MATURITY — THE TRIAL MASTER FILE (TMF) & TMF REFERENCE MODEL, REACTIVE TO ADAPTIVE
The TMF is scrambled together before an inspection; documents are back-filled and gaps are found by the inspector, not the sponsor.
An eTMF exists mapped to the Reference Model, but filing lags and completeness is checked only near milestones.
Essential documents are filed contemporaneously with QC (right version, signed, legible); completeness is checked continuously against the model.
Completeness, timeliness, and quality are trended as leading metrics; the TMF/CTMS/ISF seams are reconciled to defined rules of record.
Inspection-readiness is a continuous state by design; the TMF is the demonstrable evidence of oversight, current at any moment.
SPEQ’s shared five-stage progression, labelled synthesis — not the FDA QMM rating scale. Where does your organization sit? Score your quality system →
07 · REGULATORY & EVIDENCE
GOVERNING STANDARDS · 3
Derived from the 3 standards SPEQ maps to this subject, across 3 regulatory bodies: ICH, FDA, EMA.
RECORDS & OBJECTIVE EVIDENCE
- An index/plan mapping the TMF to the Reference Model zones, sections, and artifacts
- Essential documents filed contemporaneously as they are generated
- TMF quality-control records (correct version, signed, legible)
- Completeness checks against the expected document set during the trial
- A defined system-of-record rule reconciling TMF, CTMS, and EDC divergences
COMMON INSPECTION FINDINGS
- A TMF back-filled before inspection rather than maintained throughout
- Essential documents missing or filed months late
- TMF and investigator site file treated as one file
- Monitoring visit reports generated in the CTMS but never filed in the eTMF
- Enrolment counts diverging between CTMS and EDC with no reconciliation rule
What the TMF is for — reconstruction and proof
The TMF exists to do two things: allow the trial to be **reconstructed** — to show what was done, when, by whom, and why — and to **demonstrate GCP compliance**. Essential documents are those that individually and collectively permit evaluation of the conduct of the trial and the quality of the data produced. That purpose sets the standard: the TMF is not a filing formality but the evidence base an inspector, an auditor, or a future investigator uses to judge whether the trial was run properly. The operative principle is stark — for regulatory purposes, if an activity is not documented in the TMF, it did not happen.
This is why the defining quality of a TMF is not that it is *complete at the end* but that it is complete, timely, and quality-controlled *throughout*. A TMF assembled in a scramble before an inspection may contain all the documents and still fail the test, because contemporaneous filing is itself part of the evidence: a document filed months late raises the question of whether it existed when it should have. Inspection-readiness is a continuous state, not a pre-inspection project — the three failure axes an inspection probes are completeness (is anything missing?), timeliness (was it filed contemporaneously?), and quality (are the documents correct, signed, and current?).
TMF vs the site file vs the operational systems
Several records are routinely confused with the TMF. The **TMF is the sponsor’s** essential-document set for the trial; the **investigator site file (ISF)** — or its electronic form, the eISF — is the *site-side* counterpart, held by the investigator, with distinct obligations and overlapping but not identical content. Treating them as one file is a common error; each has documents the other does not, and inspection looks at both.
The TMF is also distinct from the operational systems that feed it. A **CTMS** (clinical trial management system) holds operational metadata — sites, enrolment, visit scheduling, monitoring-visit tracking, milestones — and answers *how is the trial running?* The **eTMF** holds the regulated document set and answers *can we prove it was run properly?* The seam between them is where findings appear: a monitoring visit report is generated in the CTMS but must be *filed* in the eTMF, and divergence between the two is a classic finding. Enrolment counts exist in both the CTMS and the EDC (the clinical-data system) and routinely disagree — neither is automatically authoritative, so the reconciliation rule has to be written down. Modern suites merge CTMS and eTMF, which makes "system of record" a **governance decision**, not a product feature.
The TMF Reference Model — and who stewards it now
Because "essential documents" is a principle rather than a fixed list, the industry converged on a shared structure: the **TMF Reference Model**, a standardised taxonomy that organises the thousands of possible essential documents into zones, sections, and artifacts so that TMFs are consistent, complete, and interoperable across sponsors and vendors. It is not a regulation — it is a voluntary industry standard — but it is the de facto framework most electronic TMFs are built on, and mapping a TMF to it is how completeness is demonstrated.
The custodianship detail matters because a lot of training content is out of date on it. The Reference Model originated as a volunteer initiative under the **Drug Information Association (DIA)**, which released version 1.0 in 2010. In April 2022 the TMF Reference Model group formally **affiliated with CDISC** (the Clinical Data Interchange Standards Consortium), which is now its steward; the current published version is **v3.3.1 (2023)**, and CDISC has set out a roadmap toward a data-driven version 4. Stating that the model is "a DIA model" is therefore a decaying error — being correct that CDISC now stewards it is a small but real marker of current knowledge.
Inspection-readiness as the operating standard
Everything about the TMF points back to one operating standard: it must be able to withstand an inspection at any time, not just at close-out. That means filing essential documents contemporaneously as they are generated, running quality control on what is filed (right document, right version, signed, legible), and continuously checking completeness against the expected document set — usually via the Reference Model — so gaps are found and closed as the trial runs rather than discovered by an inspector.
The reason this discipline is non-negotiable is that the TMF is the concrete form of the sponsor’s oversight: a sponsor that oversaw its trial well but cannot evidence it in the TMF has, for regulatory purposes, a gap; and a TMF that is complete but was clearly back-filled undermines confidence in the whole trial record. The TMF is where "we ran this trial properly" stops being a claim and becomes something that can be demonstrated — which is exactly why its completeness, timeliness, and quality are inspected as carefully as the data itself.
FREQUENTLY ASKED
What is the Trial Master File (TMF)?
The collection of essential documents that allows the conduct of a clinical trial to be reconstructed and evaluated, and the sponsor’s and investigator’s GCP compliance to be demonstrated. It is the trial’s evidentiary backbone: for regulatory purposes, if an activity is not documented in the TMF, it did not happen. Its defining quality is being complete, timely, and quality-controlled throughout the trial, not merely complete at close-out.
What is the difference between the TMF and the investigator site file?
The TMF is the sponsor’s essential-document set for the trial; the investigator site file (ISF), or electronic eISF, is the site-side counterpart held by the investigator. They have distinct obligations and overlapping but not identical content — each holds documents the other does not — and inspections look at both. Treating them as one file is a common error.
Who maintains the TMF Reference Model?
CDISC (the Clinical Data Interchange Standards Consortium). The Reference Model began as a DIA volunteer initiative that released version 1.0 in 2010, but the group affiliated with CDISC in April 2022, which now stewards it; the current version is v3.3.1 (2023), with a roadmap toward a version 4. Describing it as "a DIA model" is out of date. The model is a voluntary industry taxonomy of zones, sections, and artifacts, not a regulation.
Why is contemporaneous filing so important for the TMF?
Because timeliness is itself part of the evidence. A TMF assembled just before an inspection may contain every document and still fail, since a document filed months late raises the question of whether it existed when it should have. Inspection-readiness is a continuous state — completeness, timeliness, and quality maintained throughout — not a pre-inspection project. A back-filled TMF undermines confidence in the whole trial record.