Won’t Get Fooled Again? GlaxoSmithKline and Access to Data

On November 22nd the European Medicines’ Agency (EMA) is holding a workshop on access to the data from clinical trials.

While there have been many efforts by many people over the years to make the clinical trial process more transparent, the EMA workshop has come about primarily following the efforts of Peter Gøtzsche of the Danish Cochrane Group and Peter Doshi and Tom Jefferson from another Cochrane group.

Peter Gøtzsche

In 2007, Gøtzsche began the ball rolling by asking EMA to provide access to study reports on the weight loss drugs rimonabant and sibutramine – both since withdrawn because of suicide risks among other issues. EMA refused citing commercial confidentiality. Gøtzsche argued EMA were in breach of EU treaties. He appealed to Nikiforous Diamandouros, the European Ombudsman, and also applied to the Danish Medicine’s Agency for comparable data. In August 2009, he got some of the data from the Danish Agency. Diamandouros then required the EMA to provide access to all the data, and in February 2011 the data finally came. Gøtzsche wrote up the saga and has made the entire correspondence available online.

Peter Doshi & Tom Jefferson

The other driving force came from Peter Doshi and Tom Jefferson’s efforts to conduct a Cochrane review on Roche’s antiviral for influenza – Tamiflu. With recent scares about Avian and later Swine Flu, and other potential pandemics, governments around the world stockpiled billions of dollars worth of this under the impression it saved lives by reducing transmission, kept people out of hospital and got them back to work faster. The first publications from Doshi and Jefferson’s efforts to review the evidence began to appear in 2009 in the BMJ. Doshi and Jefferson asked Roche for access to the data – they agreed – but never delivered. They have since told the story in many places (Op-Ed New York Times), outlining the billions of dollars lack of access has cost us. The BMJ have picked the issue up as the most clear cut case on which to campaign for data access. Roche’s resistance have been a gift that keeps giving for those in favor of data access.

Andrew Witty?

Meanwhile GlaxoSmithKline have been winning kudos after their CEO Andrew Witty promised to make the data from their clinical trials available. This offer has been welcomed glowingly by the BMJ, who civilly steered clear of mentioning the fraud action for withholding trial data that led to a $3 Billion fine for GSK.

Over a decade ago before merging with SmithKline, Glaxo-Wellcome made a comparable offer of transparency. Perhaps good Glaxo has got the upperhand on the more worldly SmithKline – lawyers sometimes put GSK’s aberrations down to legacy issues. But there are wrinkles – this is not unfettered access. A panel set up by the company will vet applications to access the data. The BMJ slid over this.

Maybe good Glaxo has got the upper hand but perhaps the BMJ should have resisted the temptation to gush. Even if they were just being civil, the gush factor will lead many who don’t understand the way the “English” read between lines to think that industry have undergone a Damascene conversion. GSK’s “gift” is a far more effective form of resistance than Roche’s.

Robert Gibbons?

The best way to explain what GSK may be up to is perhaps to repeat the wonderful story of Robert Gibbons – modest hero.

On May 1 this year, The Scientist ran an article, Data Diving, on the benefits of data access. It said what every article on this subject says – in the presence of full and unfettered access to scientific data the truth will out, while without access, companies can claim their trials show whatever the company wants them to show, get to charge whatever they want for their products and can hide what they know about any harms.

Data Diving opens with the struggles of Doshi, Jefferson and colleagues to get access to the data on Roche’s Tamiflu. It then portrays Robert Gibbons as a member of the same club as Gøtzsche, Doshi and Jefferson, a pioneer in favor of unfettered access to data, who with access to patient level data denied to others has been able to turn myths on their head.

The truth is infinitely more intriguing. Gibbons, a professor at the University of Illinois and regular expert witness for pharmaceutical companies, published two articles in Archives of General Psychiatry in early 2012 claiming that patient level data from trials of Lilly’s Prozac and Wyeth’s Effexor (see Coincidence a fine thing & May Fool’s Day) showed that in contrast to media hype antidepressants in fact work very well and don’t cause suicide. It was, it turns out, lack of access to the data that led us to these mistaken beliefs. Companies don’t engage in conspiracies, they are masters of the cock-up, and if given a choice of feet to shoot themselves in will opt for both feet. It needs independent academics like Robert Gibbons to wade in and put a stop to their self-injurious behavior, and show the world how good these drugs truly are.

Fooling the journals

The writers of The Scientist, the Lancet and BMJ were fooled by superficial similarities into classifying the Fool’s Gold of Gibbons with the real gold of Gøtzsche, Doshi and Jefferson. They managed this despite the fact that very few articles in recent years in the psychiatric or any other literature have received such withering critique as Gibbons’ articles – see boringoldman, and madinamerica by Bob Whitaker. Medscape, National Public Radio, the Los Angeles Times and other outlets from Canada to Australia gave Gibbons airtime and print space in what at one point looked to me like a campaign to roll back the Black Box warnings of suicidality on antidepressants.

Gibbons in fact didn’t have access to anything the original authors of the Prozac papers hadn’t already had access to. The original authors were all Lilly personnel with presumably even fuller access than Gibbons was later given. Gibbons also managed to steer clear of other data in the public domain that could be readily accessed that show incontrovertibly that Prozac and Effexor can not only trigger suicide but that the clinical trial data shows that on balance there are more lives lost on these drugs than saved.

Perverting science

By ignoring relevant accessible data in favor of data no-one else can access, Gibbons has turned the standard access to data argument on its head and made access to data that comes through companies a potential liability for all of us. It’s hard to call the Gibbons version of data access anything less than a perversion.

Last May (May Fool’s Day) I joked that the only explanation for The Scientist getting things so badly wrong seemed to be that they were trying to perpetrate a hoax. The best response of course was to carry on in joke mode and in this vein I suggested Dr. Gibbons should be invited to chair an interview panel to recruit academics to whom companies would be prepared to make data available in the manner Lilly and Wyeth had done.

The joke has turned extremely black. This is almost exactly the mechanism Andrew Witty has just proposed for GSK’s data – to warm applause from the BMJ.

Perverting journals

What happened next in the Gibbons story is a cautionary tale. Several sets of academics – Glen Spielman and colleagues, Barney Carroll, Mickey Nardo, and Matthew Miller and colleagues all wrote to Archives pointing out the lethal flaws in the Gibbons papers that ranged from inappropriate data inclusion to incompetent mathematical calculations to biased interpretations. Letters like this would once have been published in a good journal – science depends on robust debate about the data shows. Without debate of this kind error is propagated and the self-correcting function of science is disabled.

But Archives refused to print. They offered to consign the letters to an online limbo where they would be untraceable for anyone interested to chase the issues.

Why publish when on the face of it, Gibbons was claiming access to data that his critics didn’t have? Science does not depend on publishing the rants of people who are unhappy about something but have not analyzed the data. From an editor’s point of view allowing letters that were not based on the data seems to be tackling the man rather than the issue.

But in a bizarre twist six months later Glen Spielmans got a voice mail from someone named Debbie at the behest of the Archives editor (J Coyle) stating “Dr. Coyle has sent you an accept letter for your letter regarding the Gibbons papers…”. Why?

The plot to capture data access?

So what’s going on? Are GSK making an offer aimed at frustrating the EMA process? Offering to give access to their data through an academic body is the kind of move that would appeal to politicians and might lead them to step back from granting the kind of unfettered access to the data that EMA have been offering in recent months. For politicians, what’s not to like about the GSK offer? It would come wrapped in company commitments to ensure good research governance, panels of experts, and boxes to tick. And joy of joys it would produce a succession of Robert Gibbons rather than the Tom Jeffersons, who embarrass governments as well as companies.

The long march

Before asking the tantalizing question as to why Archives suddenly backed down, there’s a twist to the Gibbons’ Legacy to explore.

In June 2009, Gibbons had chaired an Institute of Medicine public workshop that brought together experts from industry, academia, government, and advocacy groups to discuss the issue of managing perceptions of suicidality on psychotropic drugs. The participants were asked to examine and discuss currently available data, data analysis, and the future of potential partnerships in the area of clinical trials involving the nervous system.

The Planning committee included

WILLIAM POTTER (Co-chair), Merck Research Laboratories, an NIMH based defender of Prozac in 1991 who soon after joined Lilly
ROBERT GIBBONS (Co-chair), University of Illinois at Chicago
CHARLES BEASLEY, JR., Eli Lilly and Company
DAVID BRENT, University of Pittsburgh School of Medicine – an outspoken critic of the Black Box.
YEATES CONWELL, U of Rochester School of Medicine – colleague of Eric Caine below, networked to Tom Laughren
WALTER KOROSHETZ, Nat Inst of Neurological Disorders & Stroke
THOMAS LAUGHREN, Food and Drug Administration – the FDA official who likely has hidden evidence of harm over a longer period than any other
HUSSEINI MANJI, Johnson & Johnson Pharmaceutical
DAVID MICHELSON, Merck & Co.
ATUL PANDE, GlaxoSmithKline, Inc.
PHILIP WANG, National Institutes of Health – expert witness for GSK and voter against the Black Box Warning

This was billed as a consensus conference. Consensus conferences are supposed to bring proponents of differing viewpoints together in an attempt to find common ground but it would be close to impossible to have found differences between these participants before the meeting began. Calling this a consensus conference is a wonderful irony.

The project was supported by the National Academy of Sciences, the National Science Foundation, the Alzheimer’s Association, Astra-Zeneca, CeNeRx Biopharma, the Department of Health and Human Services, National Institutes of Health (including the Nat Inst on Aging, Nat Inst on Alcohol Abuse and Alcoholism, Nat Inst on Drug Abuse, Nat Eye Inst, the NIH Blueprint for Neuroscience Research, NIMH, Nat Inst of Neurological Disorders & Stroke), Eli Lilly, GE Healthcare, GSK, Johnson & Johnson, Merck Research Laboratories, the Society for Neuroscience and Wyeth. So it was clearly very important to a lot of people.

At it Gibbons recommended examining the item-3 of the weekly Hamilton Depression Rating Scale, the suicide item, and the degree of suicidal ideation and planning according to four levels of severity— plus overall Hamilton weekly ratings to determine treatment responsiveness and suicide attempts. He and Charles Beasley of Eli Lilly, he said, were conducting a reanalysis of the RCT data to determine whether this approach improves suicidal ideation as a predictor of suicidality. This he suggested was better than the approach taken by FDA in 2004-2006.

This was the exact same approach taken by Beasley and Lilly in 1991. It had been rejected by FDA in 2004 and 2006. It was this “new” research that gave rise to the papers he published in 2012 – without Beasley’s name on them.

A report came out of the meeting. “This report has been reviewed in draft form by individuals chosen for their diverse perspectives and technical expertise in accordance with procedures approved by the NRC’s report review committee. The purpose of this independent review is to provide candid and critical comments that will assist the institution in making its published report as sound as possible and to ensure that the report meets institutional standards for objectivity, evidence and responsiveness to the study charge.”

The reviewers included Charles Beasley, Eric Caine (See Y Conwell above), J Greenhouse and R Temple (T Laughren’s boss at FDA). This was a very closed circle. What were they up to?

The 2009 watershed

There was another “consensus conference” in March 2009 that featured many of the same participants, notably Charles Beasley, along with Herb Meltzer and other academics in collaboration with FDA and industry on the topic of suicidality. This was written up in the Journal of Clinical Psychiatry.

These groups appeared to be all desperately grappling with a need to find some way to stop people – patients or doctors – believing the evidence of their own eyes. When someone becomes suicidal on an antidepressant and it clears up when the drug is stopped and reappears when the drug is started again, this is conclusive evidence the drug causes suicide. But over the last 20 years companies have deployed rating scale data, arcane classification systems for suicidal events, miscoding of events, statistical significance and lack of access to the clinical trial data as ways to obscure the problem.

There was a lot more happening in 2009. To me all this activity looked like a concerted effort to roll back the Black Box. The only reason to think not, was that it didn’t make any sense that people would want to roll back the Black Box – which had had almost no effect on clinical practice or the sales of drugs. But whatever this effort might have started out as, the real fruit of the exercise may well have been the Gibbons experiment. It’s hard to see any reason for rolling out a set of publications than repeated a 20 year old discredited article other than as an experiment to see what might happen if companies stood the data access argument on its head.

Why did Archives suddenly back down? Having tested the model and found it works beautifully, putting GSK in a wonderful position to make their offer, secure in the knowledge that this was a close to perfect way to control the scientific process – why draw attention to what’s going on?

Why would Archives play ball? For journals, there is the tempting prospect on the one hand of a series of apparently data-driven Gibbons-type papers that have all been through a strict research governance process that will win the journal huge amounts of money in reprint fees, rather than on the other hand a series of articles a la Doshi, Gøtzsche and Jefferson, that will get them nothing in reprint fees, will risk tying them up in legal knots and might even put them out of business.

Too paranoid? Not paranoid enough.

Too English? Far too English.

[Conflict of Interest: I have a serious conflict of interest with this piece that will be laid out in the next post.]