In my article on the 3 main limitations to public clinical trial registries, I mentioned that public registries were not designed for the purposes of feasibility. While this is true, the scientific community was intended as one of the beneficiaries of the initiative right from the start.
To understand better why public registries hold back research (and I strongly recommend Maya Zlatanova’s article on how data sometimes gets in the way of research) we need to take a closer look at the legislation that gave start to public registries and their history. It is a surprisingly short journey, yet packed with legislative initiatives and reforms.
Public Registries Then
The history of clinical trial registries goes barely 23 years back when the FDA issued the Food and Drug Administration Modernization Act of 1997 and gave start to the first-ever public trial registry – ClinicalTrials.gov.
For decades the public and the scientific community had been concerned about two main issues of clinical research. The first one, the reporting bias: negative or null results of clinical trials were less likely to be published, in turn, skewing the public perception of certain medical interventions. The second issue was more strictly ethical – study sponsors benefited from clinical trial participants but were not required to give back to the same community, i.e. informing them about the results of the trial.
ClinicalTrials.gov was created out of this need for transparency and objective reporting of results and it went public in 2000 – just 20 years ago. Yet, the 1997 federal law did not provide any mechanism for penalizing non-compliance. It also limited its scope to trials of serious and life-threatening diseases.
Two events in 2004 necessitated a change: the then-New York State Attorney General Eliot Spitzer sued GSK on grounds that the company had failed to publish results from clinical trials revealing the harmful side effects of certain antidepressants. Then, the International Committee of Medical Journal Editors refused to consider for publication reports of clinical trials that had not been registered.
This was a strong incentive for the FDA Amendments of 2007 that included more trials to be registered and summary results to be submitted (including adverse events) for some trials; it also set penalties for those who failed to comply. In 2008 the Declaration of Helsinki was revised to include the statement that every trial must be publicly registered and accessible before the first subject enrolled.
In 2007 the WHO launched an international registry which main purpose is full transparency and access to information for all healthcare stakeholders.
Public Registries Now
ClinicalTrials.gov remains the biggest public clinical trial registry with data that includes trials in all 50 states and 216 countries. According to a 2013 study, it is followed by the EU registry and the Japan registries network.
Despite the abundance of information you can find on CT.gov, the data there is in no way presented in a feasibility-friendly way or even a patient-friendly way. There are a lot of missing pieces like incomplete or outdated sites & investigators contacts and even in the best-case scenario things like RRs (recruitment rates) are not readily available and require manual calculations for the purposes of feasibility.
In a recent poll we did on the TrialHub LinkedIn page, we asked feasibility experts what their main data challenge is and 44% answered: “To analyze the data”. This is a bigger issue than trusting public sources (22%), manual calculations (11%) or too many open tabs (11%), i.e. gathering information from a lot of different places.
And indeed, the data you can find on CT.gov or any other registry does not speak for itself. Not to mention that the WHO registry only recently opened its search again for the public (the data there was only available to developers which is how our TrialHub team got access to it in order to update the platform… but more on that in another post).
Even so, many registries like the WHO and national registries do not provide the site-level data you need for competition analysis. Their database of sites & investigators is at best incomplete, only containing local-level information with the rest of the trial left out.
And the search is not very sophisticated either – even CT.gov’s advanced search only allows you to search for either one or all countries, so you cannot compare Germany to Italy to Austria for example.
This being said, are public registries at least helpful for patients and their doctors looking for treatments in development? Not as much as they could be. The many cases of incomplete or missing contacts make it hard for the patient to get in touch with the study team.
The information there is also not presented in a patient-friendly way and requires some basic knowledge of the clinical trial process to even begin to comprehend it – and many patients just lack this knowledge. They’ve never had to learn what PI means or what the Phases of a trial mean in terms of drug development or even what a protocol is. And in many cases, they are scared, exhausted, and need someone to guide them through the process.
In short, public registries are much needed but could be optimized to serve researchers and patients better.
The Future of Public Registries
The 2017-2027 strategic plan by the National Library of Medicine (NLM) describes the NLM’s commitment to modernize ClinicalTrials.gov:
“to deliver a modern user experience on a flexible, extensible, scalable, and sustainable platform that will accommodate growth and enhance efficiency.”
The NLM wants to ensure that CT.gov will continue to build public trust in clinical research and serve its mission of transparency and mitigating reporting bias. The first main aim of these Modernization efforts is to collect complete data about clinical studies by introducing changes to the process of registration. By first undergoing automatic validation and then manual quality control, errors and inconsistencies will be detected in advance.
The second aim of the Modernization initiative is to facilitate the use of information by helping both the public and researchers find trials of interest.
The NLM welcomes your input on:
- Website functionality
- Information submission
- Data standards
Are you curious to know how we address the data issues through TrialHub? Stay tuned for an upcoming interview with our developers – the people behind our algorithm.