French Trial Tragedy: Who Says What Data Should Be Published?

While French government inspectors pore over the results of their recent visit to the premises of Biotrial, the CRO that ran the ill-fated Phase I study with the FAAH inhibitor BIA 10-2474, the regulatory agency ANSM has revealed that the product's developer, Bial, has opposed the publication of the preclinical documentation on the drug.

The use of BIA 10-2474 in the study led to the death of one volunteer and suspected neurological damage in four others. ANSM originally gave the go-ahead to the dose-ranging trial of BIA 10-2474 in mid-2015, taking account of the European Medicines Agency's "Guideline on strategies to identify and mitigate risks for first-in-man clinical trials with investigational medicinal products," which took effect in September 2007.

The agency has published the clinical trial protocol – which was amended in November to change the name of the investigator – on its website. It also wanted to publish two other documents – the investigator's brochure (IB) and the investigational medicinal product dossier (IMDP) – but Bial refused "on the pretext of trade secrets protection," ANSM said.

In refusing publication of these documents, which contain a great deal of clinical and non-clinical data, Bial cited Article L311-6 of the French code on relations between the public and the government. This code includes a section on the kinds of documents that the government can make available to a member of the public on demand.

Within this section, Article L311-6 lists the kinds of documents that should not be made publicly available, including those that would "impair the protection of private life, medical secrets and commercial and industrial secrets."

This decision raises a number of questions, not least whether serious adverse reactions in a Phase I trial should justify the release of such data in the interests of public health – a proposal that has just been made by the British Pharmacological Society. In a statement, the BPS calls for much earlier publication of IMPDs and protocols in the case of studies that lead to "devastating outcomes," such as those seen with the Bial drug.

It says that current EU regulations state that data from early human experiments are "commercially sensitive and, therefore, may be kept outside the public domain. However, in this circumstance, we believe that patient safety should outrank commercial sensitivity. A better understanding amongst the scientific community of the nature of the drug, and the way in which it was dosed, would be helpful, even at this early stage, in preventing further harm to volunteers in other clinical trials, which may require adaptation or even discontinuation based on this knowledge. An early release of data from the manufacturer is crucial."

The statement said that the BPS recommended the release of the study design and protocol, the full IMPD and the batch release data for such studies "at the earliest possible stage." This level of transparency, it said, was "critical to maximise patient safety in the future and should outweigh considerations of commercial confidentiality."

Regulatory Position

So what is the regulatory position on the public disclosure of data relating to Phase I studies? Bial's position against publication would on the face of it seem to be supported not only by French law but also by the EU legislation. The European Medicines Agency, which provides support to the EU regulatory network in cases of severe adverse reactions in clinical trials and is now liaising with the French authorities and the US Food and Drug Administration on the Bial case, told Scrip there was "no specific EU rule that currently requires preclinical results to be made public."

Under existing regulations, the IMPD and the IB are submitted to the relevant national competent authorities as part of the clinical trial application dossier, but they are not made public through the EudraCT database, the EMA said. Some information from this database is published through the public EU Clinical Trial Register, such as summary data and results, but not preclinical information, the agency added.

This will change to some extent when the provisions of the 2014 Clinical Trial Regulation (CTR) take effect, although exceptions will still be made for some personal and commercially confidential information (CCI), including that relating to Phase I studies.

New EU Database

The new portal and database to be established under the CTR will contain summary information and results for all clinical trials conducted in the EU, as well as the full application dossier including the protocol, the IB and the IMPD. It will also contain information on various types of notifications (start/end of trial, temporary halt, early termination etc.), final assessment reports and their conclusions, reports on serious breaches, inspection reports, EU control reports, lay summary of results, and the final clinical study report (CSR).

But while the new database will generally be public, there will be certain exceptions to protect personal data and CCI, taking account of the product's marketing authorization status.

Generally speaking, the EMA noted, the default position is that information and documents will be published at the time that approval is given for the trial to go ahead. In the case of Phase I studies, this information will include the main characteristics of the trial (public title, key inclusion and exclusion criteria, study type, primary outcomes, and information on the investigator and sponsor).

However, for some of this information, as well as for other trial related documents such as the protocol, the subject information sheet, the IB and the IMPD, the sponsor may ask to defer publication.

According to an EMA document on transparency of information in the new database, which was signed off by the EMA last October, in the case of category 1 trials (which include Phase I studies) the sponsor can defer publication of product-specific documents, including the IB and the safety and efficacy sections of the IMPD, until the relevant marketing authorization is granted "or up to seven years after the end of the trial, whichever is earlier."

The section of the IMPD on IMP quality and the related requests for additional information, responses and assessment report sections, however, "should be considered to be commercially confidential and not be made public for any trial at any time, as these deal with the manufacturing and related pharmaceutical development information which continues to be CCI, post marketing authorisation," the document says.

In some cases, the EMA noted, the "overriding public interest in disclosure" may prevail, whereby documents and data can be made public earlier than normal, and a decision-making process will be established in order to make use of this possibility.

"This decision should be made by the member states concerned by the trial or trials in question, supported for technical, regulatory and consistency purposes by the EMA and the EU Commission," it said. Such decisions will only be made in exceptional circumstances, for example where there are "very serious safety incidents" such as those that occurred in the 2006 UK clinical trial of TeGenero's TGN1412, it declared.

But the provisions of the CTR will not take effect until 2018, and it is to be expected that more voices will be raised in support of wider publication of documents relating to Phase I trials. Ben Goldacre of transparency campaigning organization AllTrials said recently: "We do not yet know what went wrong in this Phase I trial in France. But for as long as we continue to allow dangerous secrecy to persist around such trials, we make these disasters more likely."