Friday, September 18, 2015

Transparency 2

A re-analysis of data from the heavily criticized Study 329 involving the antidepressant paroxetine (Paxil) turned up more cases of suicidality in young people than previously reported -- a finding that suggests greater need for reporting of patient-level data by pharmaceutical companies, researchers said.

While the original 2001 publication of the study reported five cases of suicidality -- and subsequent analyses by drugmaker GlaxoSmithKline (GSK) and FDA during litigation found nine and 10 cases, respectively -- Jon Jureidini, PhD, MBBS, of the University of Adelaide in Australia, and colleagues found a total of 12 cases when they went through the 93 patient-level case reports made available to them by GSK.

"There were several of, what looks like to the impartial observer, deliberate attempts to play down the adverse effect profile," Jureidini said during a press call with editors from the BMJ, where the study was published. "We can't be definite about that, but that's what it looks like."

Study 329 has long been a subject of controversy in child and adolescent psychiatry. Published in 2001 in the Journal of the American Academy of Child and Adolescent Psychiatry (JAACAP) after being rejected by JAMA, it was criticized in letters to the editor, and the FDA never approved Paxil in adolescents based on the data.

Only fluoxetine (Prozac) and escitalopram (Lexapro) are officially indicated for use in adolescents, but antidepressants are often used off-label in this population. All antidepressants carry a black box warning about increased suicidality in people under age 25.

Data manipulation in the study also played a key role in federal regulators' legal action against GSK, which ultimately ended up paying a $3 billion fine for illegal off-label marketing of drugs, including Paxil being promoted for young people.

Despite all of these problems, the paper remains part of the literature. JAACAP editors declined to retract it after an internal investigation found no evidence of wrongdoing, Peter Doshi, PhD, of the University of Maryland in Baltimore and an associate editor at BMJ, noted in an accompanying feature article.
Brown University of Providence, R.I., where lead author of the JAACAP paper Martin Keller, MD, is an emeritus professor, never confirmed having made an internal investigation at all, Doshi said.

As part of a settlement with the New York Attorney General's office, GSK publicly disclosed all Clinical Study Reports (CSRs) from Study 329, but they left out Appendix H, which contained individual patient-level data.

After negotiations with GSK, Jureidini and colleagues were able to get access to 77,000 pages worth of individual patient reports, but they weren't allowed to print or download those records and could only make manual notations.

They scrutinized 93 patient case reports, out of the original 275 patients (34%).
Overall they found paroxetine wasn't effective at treating depression in adolescents (neither was the comparator imipramine; neither drug was better than placebo)....

And for the specific adverse event of suicidality, the researchers found a total of 12 cases among 93 children that had case reports -- far higher than the five reported in the Keller study, they said.

"It's hard to think there wasn't some mischief being done when a severe suicide attempt requiring hospitalization was coded as 'emotional liability,'" Jureidini said during the press call about one of the specific cases they reviewed and re-classified....

"Our re-analysis of Study 329 showed considerable variations in the way adverse events can be reported, demonstrating several ways in which the analysis and presentation of safety data can influence the apparent safety of a drug," they wrote in the paper.

In an emailed statement, GSK emphasized its participation in giving the researchers access to patient-level data.

"This reflects our commitment to data transparency," the statement said. "We publish the results of all our studies regardless of whether they are positive or negative and we are the only pharmaceutical company to be part of the AllTrials campaign. We have also led the way in giving external researchers access to the very detailed patient-level data behind our studies, granting access to more than 50 research teams around the world so they can independently use our data in their research."...

Doshi noted that GSK has had access to their paper since March and has not disputed the findings.


  1. Le Noury J, Nardo JM, Healy D, Jureidini J, Raven M, Tufanaru C, Abi-Jaoude E.
    Restoring Study 329: efficacy and harms of paroxetine and imipramine in treatment
    of major depression in adolescence. BMJ. 2015 Sep 16;351:h4320.



    To reanalyse SmithKline Beecham's Study 329 (published by Keller and colleagues in 2001), the primary objective of which was to compare the efficacy and safety of paroxetine and imipramine with placebo in the treatment of adolescents with unipolar major depression. The reanalysis under the restoring invisible and abandoned trials (RIAT) initiative was done to see whether access to and reanalysis of a full dataset from a randomised controlled trial would have clinically relevant implications for evidence based medicine.


    Double blind randomised placebo controlled trial.


    12 North American academic psychiatry centres, from 20 April 1994 to 15 February 1998.


    275 adolescents with major depression of at least eight weeks in duration. Exclusion criteria included a range of comorbid psychiatric and medical disorders and suicidality.


    Participants were randomised to eight weeks double blind treatment with paroxetine (20-40 mg), imipramine (200-300 mg), or placebo.


    The prespecified primary efficacy variables were change from baseline to the end of the eight week acute treatment phase in total Hamilton depression scale (HAM-D) score and the proportion of responders (HAM-D score ≤8 or ≥50% reduction in baseline HAM-D) at acute endpoint. Prespecified secondary outcomes were changes from baseline to endpoint in depression items in K-SADS-L, clinical global impression, autonomous functioning checklist, self-perception profile, and sickness impact scale; predictors of response; and number of patients who relapse during the maintenance phase. Adverse experiences were to be compared primarily by using descriptive statistics. No coding dictionary was prespecified.


    The efficacy of paroxetine and imipramine was not statistically or clinically significantly different from placebo for any prespecified primary or secondary efficacy outcome. HAM-D scores decreased by 10.7 (least squares mean) (95% confidence interval 9.1 to 12.3), 9.0 (7.4 to 10.5), and 9.1 (7.5 to 10.7) points, respectively, for the paroxetine, imipramine and placebo groups (P=0.20). There were clinically significant increases in harms, including suicidal ideation and behaviour and other serious adverse events in the paroxetine group and cardiovascular problems in the imipramine group.


    Neither paroxetine nor high dose imipramine showed efficacy for major depression in adolescents, and there was an increase in harms with both drugs. Access to primary data from trials has important implications for both clinical practice and research, including that published conclusions about efficacy and safety should not be read as authoritative. The reanalysis of Study 329 illustrates the necessity of making primary trial data and protocols available to increase the rigour of the evidence base.

  2. Doshi P. No correction, no retraction, no apology, no comment: paroxetine
    trial reanalysis raises questions about institutional responsibility. BMJ. 2015
    Sep 16;351:h4629.

    A major reanalysis just published in The BMJ of tens of thousands of pages of original trial documents from GlaxoSmithKline’s infamous Study 329, has concluded that the antidepressant paroxetine is neither safe nor effective in adolescents with depression. This conclusion, drawn by independent researchers, is in direct contrast to that of the trial’s original journal publication in 2001, which had proclaimed paroxetine “generally well tolerated and effective.” The new paper, published under the restoring invisible and abandoned trials (RIAT) initiative, has reignited calls for retraction of the original study, putting additional pressure on academic and professional institutions to publicly address the many allegations of wrongdoing...

    Yet this same year, according to the New York State Attorney General’s office, which sued GSK, over two million prescriptions were written for children and adolescents in the United States, all off-label, after a marketing campaign that characterized Study 329 as demonstrating “REMARKABLE Efficacy and Safety.”...

    Then there are the matters of “editorial assistance” and undisclosed financial conflicts of interests of one of the paper’s authors. The first draft of the manuscript ultimately published in the Journal of the American Academy of Child and Adolescent Psychiatry (JAACAP) was not written by any of the 22 named authors but by an outside medical writer hired by GSK. And the paper’s lead author—Brown University’s chief of psychiatry, Martin Keller—had been the focus of a front page investigation in the Boston Globe in 1999 that documented his under-reporting of financial ties to drug companies. Senator Charles Grassley, who led a congressional investigation and published a report on ghostwriting in the medical literature, reportedly wrote to Brown University about Keller...

    None of the paper’s 22 mostly academic university authors, nor the journal’s editors, nor the academic and professional institutions they belong to, have intervened to correct the record. The paper remains without so much as an erratum, and none of its authors—many of whom are educators and prominent members of their respective professional societies—have been disciplined...

    Such stark differences between the original paper and the rewrite are bound to put particular pressure on Andrés Martin, Yale University professor and current editor in chief of JAACAP. Martin has been under pressure to retract the paper for years, including from within his own society.

    Last October, Martin was compelled to address the academy’s assembly about Study 329. According to the minutes, members heard how Martin had investigated the matter thoroughly by consultation with the authors, the Committee on Publication Ethics (COPE), clinical experts, “a whole range of attorneys, and more.” Martin’s assessment, completed in July 2010, concluded that no further action was necessary. A follow-up inquiry, again by Martin, in 2012, after GSK was fined $3bn, similarly concluded “no basis found for editorial action against the article.”...

    Ivan Oransky, cofounder of the Retraction Watch blog, says that transparency is vital. “GSK agreed to pay a $3bn fine and you’re [Martin] saying you had completely different results? Great. Show me.”

  3. A reanalysis of the now infamous Study 329 of the antidepressant paroxetine (multiple brands) shows that the drug is ineffective and unsafe for adolescents, prompting experts to call for a retraction of the original study...

    The reanalysis used previously confidential trial documents and found that the efficacy of paroxetine and imipramine (multiple brands) was not statistically or clinically significantly different from placebo for any prespecified primary or secondary efficacy outcome, the researchers report.

    Further, they observed "clinically significant increases in harms, including suicidal ideation and behavior and other serious adverse events in the paroxetine arm and cardiovascular problems in the imipramine arm."

    "Hundreds of thousands" of adolescents took this drug, and "we can't know how many were harmed," Jon Jureidini, PhD, Critical and Ethical Medical Research Group, University of Adelaide, South Australia, told Medscape Medical News.

    "This is the first reanalysis" of a randomized controlled trial under the Restoring Invisible and Abandoned Trials (RIAT) initiative, he noted.

    In 2013, in the face of "selective reporting of outcomes of randomized controlled trials," an international group of researchers formed RIAT and called on funders and investigators of abandoned (unpublished) or misreported trials to publish undisclosed outcomes or correct misleading publications, Dr Jureidini and colleagues explain in the article...

    Study 329 was criticized by the US Food and Drug Administration in 2002. In 2004, the agency added a black box warning advising against use of paroxetine in adolescents. In 2012, GSK was fined $3 billion, in part for fraudulently promoting paroxetine.

    The reanalysis of Study 329 "illustrates the necessity of making primary trial data and protocols available to increase the rigour of the evidence base," the investigators write...

    "Rationally, I can't see how they cannot retract it, but my guess is they won't," Dr Jureidini told Medscape Medical News.

    In a statement, BMJ Editor-in-Chief Fiona Godlee, MD, said publication of the reanalyzed data from Study 329 "sets the record straight" and "shows the extent to which drug regulation is failing us." It also shows that the public and clinicians do not have the unbiased information they need to make informed decisions...

    Medscape Medical News also approached the American Academy of Child and Adolescent Psychiatry (AACAP) for comment on the BMJ article and the calls for retraction of the study.

    "AACAP has the utmost respect for the BMJ, and we thank them for their continued efforts to bring science to the forefront. AACAP supports transparency in clinical trial reporting and welcomes the RIAT initiative. The statements and opinions expressed in JAACAP articles are those of the authors and not necessarily those of AACAP, the editors, or the publisher. Inquiries about the articles and study in question should be addressed to their respective authors."

  4. In medicine, trust is paramount. We rightly hold doctors to the highest standards of integrity and expect them to be honest and open with us about our treatment, its side-effects and its likely outcome. If we expect such standards of the people who prescribe the drugs – why not the people who make them?

    The findings of a long overdue reanalysis of a 14-year-old study of the antidepressant paroxetine and its effects on adolescent patients, published in the BMJ, are astonishing. Not because of the overall conclusion that the drug is ineffective and unsafe in this age group – thankfully that has already been well established by further research and clinical experience in the intervening years.

    What is astonishing is that the original report of this study, funded by what was then SmithKline Beecham, reached exactly the opposite conclusion. Only now, years later, thanks to the publication of previously confidential trial documents by what is now GlaxoSmithKline (GSK) – and the hard work of independent researchers – can the totality of the evidence be assessed. It confirms what we already knew, and what the original report should have made clear.

    The original research, known as Study 329, has become totemic of all that is wrong in the clinical trials process. Key documents never saw the light of day and the final report was drafted by a medical writer paid for their trouble by the drug company itself. GSK was able to claim the study had shown “remarkable” efficacy and safety and in one year profited from more than two million prescriptions of the drug for children and adolescents in the US....

    But Study 329 was never retracted or disowned by the 22 professionals whose names are still on it: a symbol of the academic and pharmaceutical community’s continued reluctance to confront these issues. Dr Fiona Godlee, editor of the BMJ, says that the saga shows “the extent to which drug regulation is failing us”. But it also has lessons for the way ahead...

    All Trials, co-founded by the BMJ and supported by 615 medical and research organisations, is calling for every clinical trial to be registered and all of their results published, ensuring comprehensive assessment can be made of all findings, and for trials to be subject to the scrutiny of the entire scientific community. Dr Godlee also calls for an end to clinical trials funded and managed by industry.

    These are ambitious goals and legitimate concerns around patient confidentiality in an era of open data will have to be addressed. But there can be no argument that such a model would be supremely preferable to the current situation, in which the very people who stand to profit from positive findings can control how the data – and even which data – is made public.

    Unfortunately GSK remains the only pharmaceutical company signed up to All Trials. If this final judgement of the infamous and outrageous Study 329 carries lessons for anyone, it is for its Big Pharma competitors. The truth will out eventually. You owe it to medicine, to science and most of all to patients never to stand in its way.

  5. Study leader Jon Jureidini said it raised wider implications about the need for medical researchers to publish their underlying data — standard practice in genomics and astronomy, but rare in medicine.

    “Unless you’ve got access to patient data, you can’t be sure about the integrity of the way a study is reported,” said Professor Jureidini, of Adelaide University’s Critical and Ethical Mental Health Research Group.''

    The new study re-analysed records extracted from the company during legal proceedings and subsequent negotiations. It found that more than 10 per cent of the teenagers who took paroxetine during the trial had harmed themselves or attempted suicide, compared with about 1 per cent of patients who took a placebo.

    The reinterpreted data also showed that the drug did little to relieve depression, proving no more effective than the placebo.

    Professor Jureidini said the paper had been drafted by a ghostwriter employed by GSK. He said the trial had “fudged” its outcomes by downplaying the suicide attempts, overlooking other adverse events and using excessive doses of another drug for comparison.

    All this led to the “inescapable” conclusion that the 2001 study had deliberately misrepresented the data, he said....

    A GSK spokeswoman said the side-effects of the drug were now well understood, and the company no longer owned the product in Australia. She said the company was committed to open disclosure of trial data and had allowed the Adelaide researchers access to the Study 329 records. “We disagree with any suggestion that we didn’t help the team with its analysis,” she said.

    Professor Jureidini said GSK had granted access to the data via a server that allowed the researchers to look at the 80,000-odd records only one at a time, and repeatedly logged them out. One researcher had spent more than 1000 hours painstakingly extracting information.

  6. Keller MB, Ryan ND, Strober M, Klein RG, Kutcher SP, Birmaher B, Hagino OR,
    Koplewicz H, Carlson GA, Clarke GN, Emslie GJ, Feinberg D, Geller B, Kusumakar V,
    Papatheodorou G, Sack WH, Sweeney M, Wagner KD, Weller EB, Winters NC, Oakes R, McCafferty JP. Efficacy of paroxetine in the treatment of adolescent major
    depression: a randomized, controlled trial. J Am Acad Child Adolesc Psychiatry.
    2001 Jul;40(7):762-72.



    To compare paroxetine with placebo and imipramine with placebo for the treatment of adolescent depression.


    After a 7- to 14-day screening period, 275 adolescents with major depression began 8 weeks of double-blind paroxetine (20-40 mg), imipramine (gradual upward titration to 200-300 mg), or placebo. The two primary outcome measures were endpoint response (Hamilton Rating Scale for Depression [HAM-D] score < or = 8 or > or = 50% reduction in baseline HAM-D) and change from baseline HAM-D score. Other depression-related variables were (1) HAM-D depressed mood item; (2) depression item of the Schedule for Affective Disorders and Schizophrenia for Adolescents-Lifetime version (K-SADS-L); (3) Clinical Global Impression (CGI) improvement scores of 1 or 2; (4) nine-item depression subscale of K-SADS-L; and (5) mean CGI improvement scores.


    Paroxetine demonstrated significantly greater improvement compared with placebo in HAM-D total score < or = 8, HAM-D depressed mood item, K-SADS-L depressed mood item, and CGI score of 1 or 2. The response to imipramine was not significantly different from placebo for any measure. Neither paroxetine nor imipramine differed significantly from placebo on parent- or self-rating measures. Withdrawal rates for adverse effects were 9.7% and 6.9% for paroxetine and placebo, respectively. Of 31.5% of subjects stopping imipramine therapy because of adverse effects, nearly one third did so because of adverse cardiovascular effects.


    Paroxetine is generally well tolerated and effective for major depression in adolescents.

  7. “That one can do better reanalyzing adverse-event data using refinements in approach that have accrued in the 15 years since a study’s publication is unsurprising and not a valid critique of our study as performed and presented,” Keller and his colleagues noted in a letter in response to the BMJ article. The authors concluded, “In summary, to describe our trial as 'misreported' is pejorative and wrong, both from consideration of best research practices at the time, and in terms of a retrospective from the standpoint of current best practices.

    “This new reanalysis has sparked a healthy discussion over the value of open access to clinical trial data, the importance of following protocol specified analyses, and the risks of confirmation bias, which is a tendency to search for information that confirms the investigator's preconceptions,” Mark Olfson, M.D., M.P.H., a professor of psychiatry at Columbia University Medical Center who was not involved with either study, told Psychiatric News. “However, the new reanalyses does not alter the totality of clinical trial evidence that continues to support the safety and efficacy of SSRIs for adolescent depression.”

    Daniel Pine, M.D., chief of NIMH's Section on Development and Affective Neuroscience, added, “We have known for some time that antidepressant medications have both significant benefits for some children as well as significant risks for other children. This new analysis really does nothing to change this knowledge, and provides no new insights into what we have known about these medications for the past few years.”

    “AACAP supports transparency in clinical trial reporting and welcomes the RIAT initiative, which enables publicly available primary data to be reanalyzed and published as new, potentially revised reports,” Paramjit T. Joshi, M.D., President of the American Academy of Child and Adolescent Psychiatry, told Psychiatric News. “JAACAP is a forum for scientific reporting and scholarly discussion. The scientific process builds on itself over time through a cycle of new research, analysis, and ongoing dialog. This process stimulates debate and moves the field forward toward a better understanding of critical issues.”

    See Keller and colleagues letter at:


  8. In addition, study 329 was ghostwritten with the help of ghostwriter Sally Laden. Internal emails housed in the Drug Industry Document Archive at the University of California, San Francisco show Dr. Keller saying, "You did a superb job with this. Thank you very much. It is excellent. Enclosed are some rather minor changes from me...” (

    Dear Dr Keller:

    We are pleased to enclose all of the necessary materials for you to submit your manuscript, "Efficacy of Paroxetine but Not Imipramine in the Treatment of Adolescent Major Depression: A Randomized, Controlled Trial," to the Journal of the American Academy of Child and Adolescent Psychiatry.

    Please find enclosed the following items:
    • Five copies of the manuscript (submit four to the journal; keep one for your files)
    • One set of glossy prints of the figures (submit to the journal)
    • A draft cover letter to Dr Dulcan, editor of JAACAP (please retype on your own letterhead and revise as you like)