Overcoming TMF misfiles and metadata errors to reduce inspection risk

Trial master file misfiles are a common occurrence, with as many as one in every 10 documents being misfiled.  

In some ways, misfiles are an even greater issue now with eTMF systems, where filing is driven by metadata and even a single keystroke slipup might mean filing a document in the wrong site or even wrong study.  

If these errors aren’t spotted during quality control reviews, the document might be irretrievable. Consider, for example, a large, multi-site phase 3 trial with tens of thousands of documents and a few documents with the wrong date or wrong site number. Tracking those down would be next to impossible.  

Sponsors recognize the risks misfiles pose, which is why they typically invest heavily in QC. This, however, adds both an administrative and cost burden, because not only does it mean having someone to index the document but also at least one other person to check the metadata and perhaps a third to do additional QC. 

Holdups and regulatory backlash 

Failure to adequately maintain the trial master file can have severe consequences from both a regulatory and process perspective. Regulatory authorities are taking management of the TMF seriously, viewing it as an indicator of study health and compliance with good clinical practice (GCP).  

The European Medicines Agency’s guidance on trial master files, introduced in June 2019, offers details on various areas of TMF compliance, including sponsor, CRO, and investigator site responsibilities, and the technology requirements for electronic trial master file (eTMF) systems. 

On March 9, 2023, the EMA published a guideline on computerized systems and electronic data in clinical trials, setting out principles that apply to all computerized systems used in clinical trials. Specifically, the guideline notes that to ensure data integrity, data should be “collected, accessed, and maintained in a secure manner, to fulfil the ALCOA++ principles of being attributable, legible, contemporaneous, original, accurate, complete, consistent, enduring, available when needed and traceable”. 

Regulators expect the TMF to be complete when they carry out inspections. If they can’t find the documents they require, they may not be able to complete their inspection. Indeed, the UK’s regulator, the MHRA, has been vocal about the fact that they will stop an inspection, deem the eTMF to be unusable and issue a critical finding if they cannot find the documents they are looking for.  

There are also potential data security breaches with eTMF document misfiles, since if a document is filed in the wrong study, patients’ medical history may be exposed to teams working on a different study and who would therefore not have permission to view those documents.  

Beyond the regulatory and legal ramifications, misfiles can potentially hold up a clinical trial, which may then delay a company’s submission and marketing strategies. For example, if an institutional review board (IRB) approval document is misfiled, the study team may not know approval has been granted, which would delay the start of a trial. 

Improving the eTMF process to improve TMF health 

Given the importance of good TMF health (defined as ongoing and high levels of quality, completeness, and timeliness), ensuring documents are properly handled from the outset is of paramount importance to the success of the clinical trial.  

Clear processes are an important first step to good TMF health. For example, if a company is centralizing its indexing and has a large enough team, it may be possible to have people focused on one study at a time. This minimizes the likelihood that those people will put a document into the wrong study. 

Having clearly defined document creation processes can also help to reduce errors. Templates that have some key identifiers on of all the documents, such as a study number in the page footer, would be more efficient than relying on people to scan through multiple pages to determine which study a document belongs to. 

Digital technology, such as artificial intelligence, is playing a key role in reducing errors and misfiles. Next-generation eTMF AI systems can now help prevent errors before they occur at the critical document upload step, by making initial suggestions as to where a document should be filed – with human oversight to provide confirmation. This serves to increase TMF quality as well as reduce the burden on busy study teams for QC and remediation. 

These more sophisticated eTMF AI engines also go a step further and automatically populate document fields – such as Zone, Section, Artifact, Subartifact, Country, Language, File Type, and more – based on the selection made by the user. This speeds up the process, frees up people to focus on more value-added activities, and reduces inspection risk. 

Getting it right for better outcomes 

With regulators taking a tougher stance on errors in the trial master file and recognition that good TMF health is also crucial for patient safety, companies must take steps to reduce misfiles, improve TMF processes, and ensure they are inspection ready. A more process-centric approach, supported by next-generation technologies such as AI, is the key to good TMF health and, as such, fewer clinical trial delays.