The Netherlands' terrible clinical trial reporting can be fixed. Here's how.
A recent report by TranspariMED and Health Action International documenting widespread violations of European transparency rules by major Dutch institutions running clinical trials drew a lot of public attention in the country, and led to questions being asked by parliament.
Yesterday, the responsible Dutch government agency, the Centrale Commissie Mensgebonden Onderzoek (CCMO), published a statement setting out its perspective.
Below, we reproduce some passages of the CCMO statement as rendered by Google Translate, together with our comments.
“[T]he CCMO also asked for attention for the obligation of the client to register his results in the appropriate portals, so that the research results per study are made public. For example, in 2019 the CCMO issued a joint call for this with, among others, the European Medicines Agency. The CCMO regrets that this has not led to an increase in the publication of research results in the public trial registers."
"However, there are currently no options for enforcing this legally. For drug research this will change when the European regulation for drug research comes into force, expected at the end of next year. This makes the registration of research results for drug research in the European research portal and thus publication in the public register a legal obligation against which no objection is possible. The CCMO is working with all parties involved to implement the implementation of this new law in practice.”
TranspariMED welcomes CCMO’s efforts to inform Dutch companies and universities of their obligation to upload clinical trial results onto the European registry. However, to date, Dutch sponsors have only uploaded an estimated 3% of their due trial results.
In contrast, Irish sponsors have uploaded nearly 100% of their due results, and UK sponsors around 80%; major sponsors in Denmark, Germany and Austria are currently rapidly improving their performance.
TranspariMED does not share the assumption that the problem of missing trial results will be automatically resolved when the European Clinical Trial Regulation fully comes into force in late 2021. While trial reporting will then become a clear legal requirement, Dutch universities in particular will only be able to achieve compliance by 2021 if they start putting into place appropriate policies and systems now.
CCMO should immediately adopt successful approaches taken by public bodies in other countries:
Sending warning letters directly to trial sponsors has rapidly improved reporting performance in the UK and Denmark.
Bimonthly round tables with trial sponsors hosted by regulator BASG have helped Austrian trial sponsors to rapidly improve their regulatory compliance.
The Danish regulator DKMA has announced that it will begin issuing fines to trial sponsors.
UK authorities will in future make data on trial sponsors’ reporting performance public and refuse to grant ethics approvals for new studies to those who break the rules.
In addition, CCMO should clarify exactly how they intend to enforce the law from 2021 onwards, including a detailed mechanism for automatically imposing sanctions on companies and institutions that break the law.
“The CCMO believes it is very important that the results of research are made public as much as possible. However, there is an important nuance to be added to the NRC articles. It is not the case that research results are only made public in 3% of drug research, as the first article indicates. The results of research are primarily made public by means of publication in scientific journals and open access platforms, after a thorough peer review."
"In this way, the healthcare professionals who have to put the findings into practice and implement them in treatment guidelines are reached, so that patients can benefit from the research results. This does not detract from the fact that publication in public registers is important to inform society more widely about the results of research.”
Our study found that an estimated 97% of completed and due Dutch clinical trials violate European Union transparency rules, which is far above the European average. The study did not assess, or claim to assess, academic publication rates. TranspariMED stands by its original estimate (see here for more details).
While publication in academic journals is also important, there are good reasons why EU rules (and WHO best practices) require institutions running clinical trial to upload their results specifically onto the European trial registry.
In short, registry reporting accelerates medical progress, reduces the risk of research waste, and typically gives a more comprehensive and accurate picture of the safety and efficacy of the drugs being tested. See this blog for more details.
Dutch companies and universities almost certainly do not consistently publish all their trial results in journals, let alone within the 12 month timeframe set out by EU rules.
A recent study showed that German universities failed to make hundreds of trial results public in academic journals, causing over a quarter of all trials to become research waste; the results of other trials were only made public after long delays (examples listed on pages 6-10 here).
Currently, there is a strong public expectation for all Covid-19 clinical trials to make their results public as rapidly as possible. TranspariMED believes that cancer patients also have a right to expect that all cancer trial results are made public as rapidly as possible. Only trial registries can ensure such consistent and rapid reporting of all trial results.
For these reasons, TranspariMED welcomes CCMO’s confirmation that that “publication in public registers is important to inform society more widely about the results of research”.
“In 2012, the Medical Research with Human Subjects Act (WMO) was amended in the Netherlands, giving the CCMO the authority to publish the research results submitted by the sponsor of the research on the Dutch submission portal for clinical research (ToetsingOnline). in the CCMO register. The client was then given the opportunity to object to publication in the CCMO register. The number of objections to disclosure of research results is minimal.”
Uploading trial results onto ToetsingOnline is not an acceptable substitute for posting results onto the European trial registry.
Medical research is a global endeavour. ToetsingOnline is not part of the WHO primary registry network, and researchers outside the Netherlands are highly unlikely to be aware of the database and locate results posted there, severely limiting its contribution to medical progress.
Note also that recent research on similar national databases in Spain and in Germany [study forthcoming] has shown that many trial results are missing on these platforms, despite national legal requirements to make results available there.
Till Bruckner, founder of TranspariMED, said:
“Currently, if a Dutch patient parks her car in the wrong place outside the hospital where she volunteers to participate in a clinical trial, she has to pay a fine, but if the hospital fails to make the results of that clinical trial public, it does not have to pay a fine. This is deeply unfair to Dutch patients, endangers public health and leads to costly research waste."
"As the responsible public body, CCMO should end this impunity and ensure that in future, all interventional clinical trials in the Netherlands, including non-drug trials, make their results public on a WHO primary registry within 12 months of trial completion, as per WHO best practices.”
Note: The comments above exclusively reflect the position of TranspariMED. Health Action International, which co-published the initial report, was not involved in formulating this response to the CCMO statement.