Reporting a clinical trial result on the European registry: My nightmare journey

In this guest blog, Jasper van Miert, a PhD student at the University of Groningen and an activist with Universities Allied for Essential Medicines, describes the time-consuming and frustrating process of uploading the results of his clinical trial onto the European clinical trial registry. He argues that the European Medicines Agency must simplify the process, and calls on universities to provide more support to individual researchers.

Finally, my clinical trial has been completed. Time to report its results.

My earlier experience with the cumbersome process of registering the trial on the European registry EudraCT sends shivers down my spine: will things go more smoothly at this stage?

With trepidation, I log into my EudraCT account – only to find I have no access to my trial. Strange, because I myself registered it there in the first place. It turns out that the Dutch regulator, the CCMO, is now listed as the registrant. (Apparently, this happens every time somebody in the Netherlands registers a trial on EudraCT.)

In order to regain access, the European Medicines Agency requires me to email it a letter signed by the authorised person in my hospital (whoever that may be). My first attempt was rejected (letter not in the right format), as was the second (letter not printed on the correct stationary), but the third letter is a charm. I'm in.

Now I can re-use the information that I entered into EudraCT a couple of years ago, right? Unfortunately, no. Only the title, description and contact information is automatically copied.

This is going to take a lot of time.

And a lot of time it took, and a lot of frustration it caused.

For example, my abstract contained the following sentence:

“241 patients were randomised: 120 to drug A, 121 to drug B.”

To people who sometimes read medical studies, this is clear. But EudraCT requires me to rephrase the first line by filling in many forms, to produce this:

Allocation method: randomised controlled Arms: two (mutually exclusive). Arm 1 name: A; Arm 1 type: active comparator; Arm 1 number started: 120; Arm 1 drug 1: A Arm 1 drug 1 different name: A' Arm 1 drug 1: mode of administration: "orally" Arm 1 drug 1: pharmaceutical form "tablet" Arm 1 drug 1: dosing information: ***.

To select a route of administration and a pharmaceutical form, I have to use impractical select forms that squeeze too many options into a small space, and which do not allow using the keyboard to jump to the right option. And the options are many.

There are over 500 (!) methods of administration, many of which I never heard of during nine years of medical study.

There are 11 types of inhalation vapour, and 36 types of powder and solvent. Given this confusing array of micro-choices, different academic researchers on the same study team would probably select completely different routes of administration.

Here another line from my abstract:

“After one year of follow-up, event X had occurred in 3 patients randomised to A, and 4 randomised to B (hazard ratio 0.75 (95% CI 0.17–3.35)).”

This is how this single sentence has to be entered into the EudraCT interface:

Name of period: one year (full trial) Subject analysis set: intention to treat (give definition) Endpoint 1: efficacy. Endpoint 1 countable or measurable: countable Endpoint 1 countable unit: patients (?) Endpoint 1 type: primary Endpoint 1 timeframe: 1 year Endpoint 1 period 1 arms included: A and B Endpoint 1 period 1 subject analysis sets: intention to treat Endpoint 1 period 1 reporting group 1 count: 3 Endpoint 1 period 1 reporting group 2 count: 4 Endpoint 1 period 1 subject analysis set 1 (ITT): count 7 [yep, surprisingly 3+4=7, better confirm] Endpoint 1 period 1 statistical analysis title: Hazard Ratio Endpoint 1 period 1 statistical analysis description: *** Endpoint 1 period 1 statistical analysis comparison groups: A and B, ITT Endpoint 1 period 1 statistical analysis specification: pre-specified Endpoint 1 period 1 statistical analysis type: other Endpoint 1 period 1 statistical analysis type comment: *** Endpoint 1 period 1 statistical analysis parameter estimate type: Hazard ratio (HR) Endpoint 1 period 1 statistical analysis parameter estimate point estimate: 0.75 Endpoint 1 period 1 statistical analysis parameter estimate confidence interval sides: 2-sided Endpoint 1 period 1 statistical analysis parameter estimate confidence interval level: 95% Endpoint 1 period 1 statistical analysis parameter estimate confidence interval lower limit: 0.17 Endpoint 1 period 1 statistical analysis parameter estimate confidence interval upper limit: 3.35

Argh! And I'm still not done, because my small trial has 1 primary and 2 secondary composite endpoints.

Next, I have to start inputting information on adverse events, even though my trial did not systematically collect adverse events.

EudraCT requires me to classify events using a data dictionary I have never heard of. Does the European Medicines Agency really think it is a good idea to get a junior doctor to classify badly collected events using a library he has never used before?

I close my computer in frustration. Let's try again tomorrow, after a good night's sleep.

If even I, who campaigns for greater clinical trial transparency, cannot bear the process of uploading trial results onto EudraCT, who else could?

Two things are required to improve the situation. First, the European Medicines Agency must make its trial registry more user friendly, including and especially for first-time users like myself. Second, universities must provide more assistance to researchers in registering trials and uploading their results onto trial registries, as industry is already doing.

Note: In TranspariMED’s experience, universities that excel at clinical trial transparency assume responsibility for, and oversight of, their trial registry portfolios, and actively support individual researchers in the reporting process. Universities can use TranspariMED's collection of transparency tools to strengthen their policies, processes and performance. The tool collection includes two case studies of universities that have successfully tackled the problem.

Under European Union rules adopted to protect patients’ and taxpayers' interests, every trial registered on the EudraCT registry has to post its summary results there within 12 months of trial completion, but many universities are still violating these rules.