Discord Remains On EU Clinical Trials Transparency, Even Within Industry

EU regulators, industry and researchers remain at odds on the level of transparency needed in clinical trials, despite a fast approaching deadline and efforts by EFPIA and PhRMA to break the diplomatic logjam.

Although a tight race is on to get the new EU Clinical Trials Regulation through the European Parliament ahead of elections in May 2014, the drug industry, its regulators, researchers and clinicians remain at odds over what data should be made public, to whom and how.

The joint proposal to enhance responsible clinical trial data sharing, published on July 24 by the European Federation of Pharmaceutical Industries and Associations (EFPIA) and the Pharmaceutical Research and Manufacturers of America (PhRMA) was supposed to have bridged the gap between the conflicting parties, but has perhaps raised more questions than given answers. This is not necessarily a bad thing, as debate would bring these parties closer to a deal on data sharing. Still, time for debate may soon be over (Also see "Time Is Running Out For Revised EU Clinical Trials Legislation" - Pink Sheet, 20 Aug, 2013.).

The EFPIA/PhRMA deal as presented by EFPIA Director General Richard Bergström at the “A Roadmap for Sharing Clinical Trial Data” in Brussels, Belgium on Aug. 28 represents a major commitment to transparency by the pharmaceutical industry. It would entail biopharmaceutical companies making available, at the very least, the synopses of clinical study reports after the FDA or European Medicines Agency (EMA), or an EU Member State medicines agency had approved a medicine or a new indication for an already approved drug.

Key Transparency Campaigner Wants More

For Ben Goldacre, author of “Bad Science” and “Bad Pharma” and co-founder of the AllTrials campaign, which calls for all past and present clinical trials to be registered and their results reported, the commitment by EFPIA does not go far enough.

Speaking at the event organized by Vital Transformation, Goldacre said that there are a number of glaring omissions in the EFPIA document.

“It does not go as far as the EMA draft in which almost all the CSRs are to be published, which is very good and that also has data on rejected drugs, which is proposed to be shared and is extremely important, particularly for orphan diseases,” he said.

Goldacre cited an array of loopholes and exemptions in the EFPIA draft. “It misses off-label uses, despite the fact that we know off label use is highly prevalent, and it covers no retrospective trials,” Goldacre said. This was a fundamental omission, he suggested, because the prescriptions of today are based on the trials of yesterday. “Because this promise gets us nothing from the past, it will do nothing to improve the evidence base for the prescribing decisions that we are making today,” Goldacre explained.

He went on to outline a number of complexities in the EFPIA draft that he suggested simply rendered access to the industry's offering confusing and problematic. “It bundles up protocols in with individual patient data in the sense that people have to apply to get access to the protocol by going through the same elaborate process that you have to go through to get individual line-by-line patient data. This is clearly not proportionate: the protocol should be publicly available before a trial even begins,” Goldacre said.

He also fired off a salvo at the proposed individual drug manufacturers' scientific review boards, saying: “The panels, described as independent are not independent; they are only required to exclude employees, they can still have people with close contacts with the company.”

Goldacre also called for a routine audit of requests and correspondence so that interested parties could follow the paper trail. “We need to know: who asked, who rejected them and what were their reasons for rejecting them,” he said.

EFPIA had also proposed that researchers requesting data should submit a research proposal to document the legitimacy of the research questions and the qualifications of the requester. “I fully agree with that, but I think we have to apply that same high standard to people doing analysis on their own data,” Goldacre said.

The industry appears to be calling for advanced research proposal submissions from third parties to avoid the pitfalls of so-called “data dredging” in the negative sense, which equates to snooping.

Goldacre wants drug companies also to share their statistical analysis plans in advance, “because the risk of people doing data dredging on their own trial data is exactly the same as the risk of people going data dredging on other people's trial data.” Presumably this would avoid cases of distorted reporting with the aim of producing a false positive in favor of a drug.

Industry Conciliatory

But if Goldacre had come to the debate armed for a fight, he failed to get one. Industry's approach was more studied, more open and more conciliatory than ever before.

Acknowledging that both regulators and the industry had come slow to the debating table, Bergström stressed that EFPIA's new deal was about sharing more data on a large scale. “We don't have to be defensive in any way. Maybe we could have done this a year ago or even two years ago, but nevertheless, it's not too late, I think,” he said.

Bergström said that he was ready to work towards removing the hurdles that made access to the study protocol and plan difficult. And he suggested the industry be held to account for its efforts or lack thereof in this direction, adding; “We are ready to listen and to make sure that the processes are manageable.”

The sticking point for Goldacre and his colleagues on the AllTrials campaign is the need for retrospective data from trials going back up to two decades.

“I would be interested to know why companies think it is right to withhold the results” Goldacre said. He cited non-available data from drugs used today such as Pfizer Inc.'s antidepressant Effexor (venlafaxine) and Eli Lilly & Co.'s antipsychotic Zyprexa (olanzapine).“Why can't we have access to those results from those trials for the drugs that I myself am prescribing right now?”

Again, industry appears to be giving ground, with Bergström pointing out that the proposals that EFPIA put forward were a “baseline” from which further concessions, based on mutual understandings, could be made. “We really wanted to have one minimum standard for the entire industry,” he said.

Bergström added that a number of EFPIA's members had already pledged to open up their data banks and retroactively provide information to qualifying requesters. Still, he stressed that this type of data could only be made available on a case-by-case basis, due to the fact that some of the information could be commercially confidential in nature.

“There are increasingly products in development where we do not have patent protection,” Bergström explained. Indeed, Francis Collins, Director of the National Institutes of Health has laid out plans for drug repurposing or making older drugs effective in new indications, an obvious example being thalidomide.

EU data exclusivity laws currently offer up to 10 years of protection for a new drug, in order to prevent generic competition eroding the manufacturer's ability to recuperate costs. But Bergström argued that while the EMA was championing data exclusivity on the one hand, it was also making it far more difficult for companies to obtain this type of protection. “One of our member companies has been told that its new MS drug, just because it is a known compound, won't get any data protection,” he pointed out.

Bergström omitted the manufacturer's name, but it is interesting to note that in March this year, the EMA's Committee for Medicinal Products for Human Use refused to grant Sanofi's Aubagio (teriflunomide) new active substance (NAS) status – thereby denying it data exclusivity – on the grounds that it is in fact a metabolite of an existing, off-patent drug, Sanofi's own Arava (leflunomide) (Also see "Sanofi’s Aubagio And Biogen Idec’s Tecfidera Make It A Good Day For MS Sufferers At EMA" - Pink Sheet, 22 Mar, 2013.).

EFPIA is not looking to block access to researchers for information wholesale, but is offering to explore with interested parties a way in which to safeguard manufacturer's intellectual property in such cases.

The public deserves more, though, Goldacre suggested.

He said that it was not possible to hide the results from competitors in China while sharing them with researchers in Europe - the data nonetheless had to be shared. “I cannot prescribe a drug to patients, ethically, knowing that information about that drug is being withheld and that the best I have is a summary by somebody who I trust to be clue full and well-motivated, but I cannot expect regulators to be perfect,” Goldacre said.

This is clearly an impasse. And while Bergström said that he is sure that EFPIA member companies will be able to offer the kind of data that AllTrials is seeking, a far-reaching agreement may yet be a few months off.

Industry Not In Unison

EFPIA may, however, have more problems with ensuring a harmonized interpretation of its existing joint undertaking with PhRMA amongst its members.

Also speaking at the event in Brussels was Neal Parker, Section Head Legal - Biologics Strategic Development, AbbVie Inc. , the pharmaceutical manufacturer that is in fact taking the EMA to court for releasing data on its anti-inflammatory behemoth Humira (adalimumab) (Also see "InterMune, AbbVie Data Case Could Add Fuel To Wider European Disclosure Debates" - Pink Sheet, 6 May, 2013.).

“Each company must be permitted on its own to determine what information in which of its dossiers for which of its products it needs to protect in order to prevent competitive harm,” Parker said.

Presenting a three-stage slide on what AbbVie's view of what might or might not be disclosed, Parker said that the company considered that clinical study designs, subject-level data, study-level results and clinical study protocols “could”, depending on the circumstances, constitute confidential commercial material. AbbVie says that there should be reasonable access to the scientific community to such information to replicate results, prove or disprove the detail in the package insert or what the company has claimed is accurate, generate new ideas and move science forward.

But there is another category of information that AbbVie is totally unwilling to disclose. Contained in a white central sphere on the slide, these included: interpretive analyses and judgments, which include a sponsors characterization, theories and conclusions regarding subject-level and study-level results; internal, individual tactical decisions; how the company runs a study or engages with regulators; and how it solves challenges uncovered during clinical studies.

Referring to this as “subjective” information, Parker said; “the white represents the intellectual output of our scientists, which we consider among our company's most valuable assets.”

This type of information could give others a substantial competitive advantage. “Third party researchers do not need access to this sort of information to conduct robust independent analyses of our data and our products,” Parker said.

Bergström suggested that the biologics sector may be a bit more concerned about sharing data due to the fact that it was such a competitive area. “We need to find a model in which there is a possibility to blacken out more than is the normal case, but again, under these commitments that we have put forward today, this data, all of it, including that white spot, will be available for legitimate researchers,” Bergström said.

So it appears that not everyone in industry is reading from the same page as yet. And, with regulators continuing to express surprise and even consternation at what pharmaceutical manufacturers might seek to have classified as commercially confidential information, the hope for an all-encompassing settlement that can be agreed on by all parties, including the European Commission, the Council of Ministers and the Parliament, is dwindling fast.