One of the great ironies that gives me chuckle in my day-to-day work is the seeming paradox that one of the purposes of the Trial Master File (TMF) is to REDUCE documentation, while in reality, the amount of documentation that goes into a TMF can honestly feel overwhelming.
What is a Trial Master File (TMF)?
Conceptually, a TMF is extremely simple—a TMF is nothing more than a collection of essential documents for a clinical trial. Operationally, however, TMFs are quite complex and require a significant amount of time to establish and maintain.
One factor that makes TMFs challenging to manage is simply the sheer volume of documentation. In the DIA Reference Model1, there are 250 artifacts. Hypothetically, if each artifact had an average of two versions, and each version had an average of ten pages, the TMF would consist of 5,000 pages—that’s more than 4 copies of War and Peace! And that’s a very conservative estimate. Most TMFs are significantly larger.
Why do TMFs exist?
The United States Food and Drug Administration requires clinical trials to be in compliance with the International Conference on Harmonization (ICH) Good Clinical Practice (GCP) guidelines.2 Section 8 of these guidelines provide a list of the minimum requirements for a TMF and clearly lays out the rationale for why TMFs exist. TMFs:
- Permit evaluation of the conduct of a trial
- Confirm the validity of the trial conduct and the integrity of data collected
- Demonstrate compliance with all applicable regulatory requirements
What does a TMF look like?
A TMF is just a set of organized documents and files. TMFs can be hardcopy, softcopy, or a combination thereof. For a hardcopy TMF, picture a binder [actually, multiple binders] divided into three hierarchical levels using labeled dividers. For an electronic TMF (eTMF) system, picture the files structure on your computer. In a TMF using the DIA Reference Model1, level 1 contains the zones, level 2 contains the sections, and level 3 contains the artifacts. There are 11 zones, 48 sections, and 250 artifacts.
An artifact is a document or a file. For example, an artifact can be anything from a PDF document to a SAS program file or a video recording mp4 file. This does not mean that every TMF should contain exactly 250 files. There is not a 1:1 relationship between an artifact and a file. Each artifact can correspond to multiple files. For example, a clinical trial may have two files that correspond to artifact 11.01.01, Statistical Analysis Plan: SAP version 1 and SAP version 2. Furthermore, not all artifacts are applicable to every trial. For example, artifact 11.02.01, Randomization Plan, will not apply to an open-label, non-randomized study.
Who is Drug Information Association (DIA)?
The Drug Information Association (DIA) Document and Records Management Community manages the TMF Reference Model. “The TMF Reference Model provides standardized taxonomy and metadata and outlines a reference definition of TMF content using standard nomenclature”1, and is the minimal list of essential documents.
The TMF Reference Model and other materials found at http://tmfrefmodel.com have been developed through a collaborative industry effort and published in the public domain for use by anyone for any purpose without restriction.
While it is mandatory to maintain a TMF for a clinical trial per ICH CGP Section 8, it is not mandatory that every study use the DIA TMF Reference Model. Be that as it may, I would personally encourage use of the DIA TMF Reference Model. As a general principle, I am very much in favor of standardization. Just look at the value standardization adds to commercial transport. The average truck height is 13 feet 6 inches while the average overpass height is 14-16 feet. This is not by accident; the Federal Highway Administration has standardized road constructions so that trucks aren’t constantly ramming into overpasses. (For those of you who live in Boston, we’ve seen enough tall trucks run into the bridges on Storrow Drive to truly appreciate the value of this standardization.)
In much the same way, DIA has taken great strides toward standardizing TMF content, structure, and naming conventions. This standardization provides a myriad of benefits, including making it easier to maintain a TMF, locate documents, audit a TMF, and engage external entities.
What’s the Golden Rule?
“The golden rule of clinical trial research is that if it isn’t written down, it didn’t happen.”3
My first job out of graduate school was in human resources. A recurring mantra that guided our daily work was “if it isn’t documented, it didn’t happen.” Whether I was assisting an employee with something as simple as providing them with a parking pass request form or something as serious as discussing inappropriate workplace behavior, the rule was the same—if I didn’t document the exchange in our employee management database, it didn’t happen. In the most extreme cases, this documentation (which may have seemed tedious at the time) could become crucial evidence should a former employee allege unlawful termination in court.
Clinical trials follow the same golden rule: if it isn’t written down, it didn’t happen. And where is this crucial documentation kept? The TMF, of course! PROMETRIKA’s Head of QA explains it in a way that really resonates with me. He likens auditors to archeologists. If an archeologist finds a perfectly preserved setting from the past, s/he can recreate the events and retell their story. Now, pretend you are an auditor looking at your clinical trials TMF 50 years from now. You should be able to recreate the clinical trial exactly as it happened by reading the TMF. So in your present day work, if you are ever wondering if you should take the time to write something down or if something should be filed in the TMF or not, let the auditor 50 years in the future be your guide.
Food for Thought
As you think about your TMF for your clinical trial, here are some key considerations:
- The three pillars of the TMF are “Completeness, Timeliness, and Quality.”4
- TMFs must be maintained in real-time, not as an afterthought.
- If outsourcing your TMF, select your vendor carefully…ultimately, the Sponsor is accountable.
- Have a Standard Operating Procedure (SOP) to dictate how TMFs are to be handled.
- Ensure your staff are properly trained to maintain your TMF.
- If keeping an electronic TMF, you must use a validated system.
- Test yourself. Conduct a mock inspection of your TMF to identify any issues before the FDA does.
- Consider hiring a TMF Operations Specialist, or if it interests you, make a career out of it!
As you look at your shelves full of TMF binders and your terabytes worth of TMF files, just remember the TMF only collects final, essential documents, and nothing else.
1Trial Master File Reference Model (a DIA Document & Records Management Community project. https://tmfrefmodel.com/. Accessed September 15, 2021.
2 U.S. HHS, FDA, CDER, CBER. E6(R2) Good Clinical Practice: Integrated Addendum to ICH E6(R1)
Guidance for Industry. March 2018. https://www.fda.gov/files/drugs/published/E6%28R2%29-Good-Clinical-Practice–Integrated-Addendum-to-ICH-E6%28R1%29.pdf. Accessed September 15, 2021.
3Ehdaivand S., Morrill J. Best Practices for Maintaining a Trial Master File. March 30, 2020. https://www.sharevault.com/blog/life-sciences/best-practices-for-maintaining-a-trial-master-file. Accessed September 15, 2021.
4Mitchell, T. 20 Trial Master File Resolutions for 2020. https://blog.montrium.com/blog/20-trial-master-file-best-practice-resolutions-for-2020. Accessed September 15, 2021.