Many German universities struggle to navigate the complexities of clinical trial reporting. This blog post discusses five common misconceptions, and suggests how universities can improve their performance.
But first some context:
Current university reporting performance is weak because there has been a diffusion of responsibilities within institutions and no clear oversight. In addition, regulators in Germany and Brussels have in the past failed to clearly communicate regulatory requirements to the institutions they are supposed to regulate.
In addition, regulators failed to notify trial sponsors when they were breaking the rules, so lapses often went unnoiticed. This is changing - the European Medicines Agency started sending out reminders in 2019. German regulators have yet to begin monitoring trial registration and reporting on a national level.
Now to the misconceptions.
Misconception 1: “Registry reporting is optional”
German regulator BfArM has clearly and consistently stated that it expects trial sponsors to upload their results onto EudraCT.
The second national regulator, Paul Ehrlich Institut, has in the past sent more mixed messages. However, in a statement provided to TranspariMED in early January, Paul Ehrlich Institut clarified that:
"Transparency in clinical trials is essential to protect health and foster innovation in medical research. The Paul-Ehrlich-Institut, Federal Institute for Vaccines and Biomedicines, expressly supports the joint call by the European Commission (EC), the European Medicines Agency (EMA) and the Heads of Medicines Agencies (HMA) for those responsible for clinical trials to make the results of their authorized clinical trials available in EudraCT."
The Institute's statement adds that:
"There is currently no general legal obligation to publish clinical trials results in EudraCT. In contrast to a Directive [“Richtlinie”] or a Regulation [“Verordnung”], the European Commission Guideline (2012/C 302/03) [Leitlinie] on the entry and publication of results-oriented information on clinical trials does not provide a binding legal basis."
This may (or may not) currently be the case. However, there will be an unambiguously "binding legal basis" once the Clinical Trial regulation comes into force in late 2020 or early 2021. Any current disputes about legal technicalities will soon be a thing of the past anyway.
Misconception 2: “Reporting trial outcomes in journals is enough”
Reporting trial results in peer-reviewed journals is important, but European Union rules require trial sponsors to report the results of trials on the European database EudraCT, whether or not these have already been – or will in future be – reported in journals or at academic conferences.
There are numerous good reasons for this. The most important ones are that (1) it speeds up medical progress because getting results published in journals can take several years, as opposed to only one year for registry reporting, and (2) it ensures that no trial remains unreported.
Academics often assume that their career incentives alone will ensure that all university trial results get published in a journal sooner or later. This is not the case. A 2019 study found that 28% of all clinical trials run by German universities never made their results public – anywhere. The results of these trials, which aimed to recruit 56,730 patients, are in acute danger of being lost forever. Reporting their results on a registry within 12 months of completion would have avoided this. TranspariMED’s recent report discusses several such trials whose outcomes remain completely unknown.
Important: Making trial results public on registries will not prevent subsequent publication by a journal.
Misconception 3: “The results are available on Pharmnet.Bund”
Two German universities recently told the media that their trial results were available on other platforms, pointing to the German national Pharmnet.Bund database and the American trial registry Clinicaltrials.gov.
Sadly, this is not the case. For example, TranspariMED’s report discussed LMU Munich’s kidney transplant trial 2013-004956-39, whose results are missing from the European trial registry. Its results appear to never have been published in a journal; the university told us that “we could not figure out whether it had already been published.”
The trial itself can be located on Pharmnet.Bund, but its results are missing there too. Even if they were available on Pharmnet.Bund, they would de facto be invisible to the global medical research community. Pharmnet.Bund entries may (or may not) have value from the perspective of German regulators, but the databank is not part of the global WHO Primary Registry Network, so posting results there does not satisfy WHO transparency requirements.
As Pharmnet.Bund is completely unknown outside Germany, and rarely used by German researchers, posting trial outcomes onto that database does not qualify as making them “public” in any meaningful sense of the word.
Misconception 4: “The results are on other trial registries”
What about Clinicaltrials.gov, the American trial registry?
The kidney transplant trial discussed above was never registered on Clinicaltrials.gov. This is a good thing – as it is already registered on the European trial registry, which is part of the global ICTRP network, double registering it on Clinicaltrials.gov would have been an unnecessary duplication.
However, in other cases, registration on Clinicaltrials.gov (or the German DRKS registry, which is part of the global registry network) is appropriate and even necessary because the European registry will only accept certain kinds of drug trials. That means that any trial of medical devices, surgery or non-drug treatments such as physiotherapy must be registered elsewhere.
German universities frequently do use the American database, but they usually do not upload trial results there either. A quick search for trials sponsored by LMU Munich shows numerous trials completed more than 12 months ago that are missing results on Clinicaltrials.gov:
In fact, a comprehensive study of German university trials registered on Clinicaltrials.gov and DRKS found that tabular results had been uploaded onto these registries for only 13 (!) out of the 1,509 trials listed on these registries – less than 0.1%. In other words, German universities’ reporting performance on other registries is even worse than for trials listed on EudraCT.
Misconception 5: “Regulatory compliance is good enough”
Trials on EudraCT
Many German universities – including LMU Munich – are now starting to clear their backlogs of unreported trials on the European trial registry EudraCT, and putting into place systems that ensure that in future, the results of drug trials will routinely be uploaded onto EudraCT within 12 months of trial completion. This will ensure that these universities will not face sanctions once the European Clinical Trial Regulation fully comes into force in late 2020 or early 2021.
That is a good start – but not good enough.
Trials on Clinicaltrials.gov
Some trials run by German universities are subject to reporting requirements under the U.S. Food and Drug Administration Amendments Act. Under that law, German universities risk being fined up to $11,000 for every day that a trial’s results are overdue. While the FDA has not yet begun collecting fines, that may change very soon. Since determining whether or not a trial is subject to FDAAA is really complex, it is far easier to ensure that going forward, the results of all new trials are routinely uploaded to Clinicaltrials.gov within 12 months of trial completion.
Trials on DRKS
There is currently no legal obligation to upload the results of clinical trials listed on the German registry DRKS. However, posting results onto DRKS is essential for meeting transparency standards set out by the WHO. Many medical research funders, including the Gates Foundation and Wellcome Trust, now expect their grantees to comply with WHO best practices. German universities that fail to do so risk becoming ineligible for future international research funding.
What should German universities do?
Focus on scientific excellence and patient welfare
Excellence in reporting research results is an integral part of overall research excellence. Unpublished clinical trials distort the scientific evidence base. They also betray the trust of trial participants, and do not benefit patients.
The only way to consistently and reliably excel scientifically and meet ethics requirements is to set up strong systems that ensure that results are routinely and rapidly shared via trial registries. Strong systems can also save researchers valuable time because they minimise the time required to register trials and manage registry entries.
Adopt global best practices
German universities can save themselves a lot of headaches by aiming beyond narrow regulatory compliance now and immediately leapfrogging to fully implementing WHO best practices (see checklist below).
This means setting up systems and processes that ensure that in future, the results of every single interventional clinical trial are uploaded onto every trial registry where that trial has been registered within 12 months of trial completion (6 months for paediatric trials).
Regarding already completed trials, universities should:
Use the EU Trials Tracker to identify trials listed on EudraCT missing results and then upload those results
Use the IntoValue data set to identify trials listed on Clinicaltrials.gov and DRKS that have never published their outcomes anywhere, and upload the results for those trials onto these registries before they are lost forever
Transparency tools and sources of support
TranspariMED has curated a collection of Transparency Tools to support universities in their efforts.
Universities can also contact the European Medicine Agency’s helpdesk for trial sponsors (this has recently become far more responsive), the EU Trials Tracker team, and the Clinicaltrials.gov PRS support team for assistance with reporting the results of individual trials.