Barry Blackwell: The Baby and the Bathwater
Leemon B. McHenry and Jay D. Amsterdams’s reply to Glen Spielman’s comment on Jay Amsterdam’s comment on Edward Shorter’s comment
Evidence-Based Medicine and the RIAT Initiative
One of the most promising developments for restoring confidence in evidence-based medicine is Peter Doshi's Restoring Invisible and Abandoned Trials (RIAT) initiative (Doshi, Dickersin, Healy et al. 2013). The idea is to identify unpublished or misreported trials, gain access to the patient-level, raw data and perform an independent, objective reanalysis of the trial. This process was successfully achieved in the landmark reanalysis of GlaxoSmithKline’s(GSK) paroxetine study 329 by Le Noury et al. in 2015. Study 329, which was ghost written by Sally Laden of Scientific Therapeutics Information, Inc., in conjunction with GSK employees, and published in the Journal of the American Academy of Child & Adolescent Psychiatry under the so-called authorship of Keller et al. (Keller, Ryan, Strober et al. (2001). The published article appeared to show a significant and “positive” finding for paroxetine compared to imipramine and placebo. By contrast, the objective RIAT reanalysis of the study indicated that it was negative for efficacy and positive for harm.
Forest Laboratories CIT-MD-18 trial is easily compared to GlaxoSmithKline’s study 329 for a misreported trial ofSSRI antidepressant use in pediatric depression (Amsterdam, McHenry and Jureidini 2017). We have now issued a “Call to action: RIAT reanalysis of Forest Laboratories CIT-MD-18: A Randomized, Double-Blind, Placebo-Controlled Evaluation of the Safety and Efficacy of Citalopram in Children and Adolescents with Depression” in the British Medical Journal(McHenry 2018) to launch a RIAT reanalysis of Forest's CIT-MD-18 study, entitled “A Randomized, Double-Blind, Placebo-Controlled Evaluation of the Safety and Efficacy of Citalopram in Children and Adolescents with Depression.” In order to assemble a team of investigators and begin the RIAT reanalysis of the Forest CIT-MD-18 Trial, on July 18, 2018, we formally requested the patient level data from the current owner of the citalopram CIT-MD-18 trial, Allergan Inc., which purchased Forest Laboratories, Inc. (as part of a corporate merger deal). To date, Allergan has not yet responded to this written request.
Forest Laboratories CIT-MD-18 trial was one of two RCTs that Forest used in its original application to the FDA in order to gain approval in 2009 of escitalopram in the treatment of adolescent depression. However, in the case of the CIT-MD-18 trial, citalopram failed to outperform placebo on the protocol-specified primary and secondary outcomes measures. The study was conducted between 1999 and 2002 and was designated as a 9-week, 20-site, randomized, double-blind, comparison of the safety and efficacy of citalopram verses placebo in children aged 7-11 and adolescents aged 12-17 with major depressive disorder. It was designated a Phase III registration trial supporting an FDA indication for depression in pediatric patients. The study protocol specified that the primary efficacy measure was the change from baseline to week 8 on the Children’s Depression Rating Scale-Revised (CDRS-R) total score. Protocol-specified secondary efficacy measures included the Clinical Global Impression severity and improvement subscales, Kiddie Schedule for Affective Disorders and Schizophrenia―depression module and Children’s Global Assessment Scale.
In our view, the final study report and the published Wagner et al. CIT-MD-18 article contained potentially inaccurate and misleading claims regarding efficacy and safety that were inconsistent with the protocol-designated criteria (Wagner Robb, Findling et al. 2004). The study sample size was increased by adding eight of nine subjects who, per protocol, should have been excluded because they were inadvertently dispensed unblinded study drug (Jureidini, Amsterdam and McHenry 2016). When the eight unblinded subjects were added back into the analysis, a statistically insignificant efficacy outcome was transformed into a statistically significant outcome. Because both patients and investigators were unblinded, CIT-MD-18 was not, in fact, a completely double-blind study. Other issues that were evident in the published Wagner et al.article include: the presentation of favorable post hoc results as though they were protocol-designated secondary outcomes; failure to present negative and unfavorable outcomes; obscuring a statistically significant age-by-treatment interaction effect by presenting misleading effect size results; and inaccurately reporting safety results.(A more comprehensive description of these data analysis inaccuracies was previously published on the INHN web site). CIT-MD-18 is still misrepresented in some clinical prescribing guidelines for citalopram use in pediatric depression (Sussex Partnership NHS2014).According to Allergan’s policy of data sharing:
Allergan accepts requests to share information beyond what is available in public registries or in published literature with External Researchers.
Allergan will accept and review external scientific research proposals and provide applicable anonymized patient-level data to support external research involving our clinical trial data. This applies to protocols, clinical study reports, and de-identified patient-level and study-level data from clinical trials (Allergan, 2018).
The purpose of the RIAT reanalysis of is purely academic and scholarly, and has no commercial intention. In the case of CIT-MD-18, there is sufficient evidence that the trial was misreported in the Wagner et al. and subsequent scientific publications. The goal of the research team of scientists engaged in a RIAT reanalysis is to analyze and write a manuscript for publication, independently of the sponsor, that reports the trial with fidelity to the original protocol.
Allergan’s failure to provide the patient-level data for scientific investigation not only contradicts their own data sharing policy, but is also antithetical to the views and process of scientific investigation. If their data are not disseminated as part of their own stated data sharing policy, the scientific process is compromised. The integrity of evidence-based medicine depends on the transparency of all stakeholders.
References:
Allergan, Data Sharing Policy 2018.Available from: www.allerganclinicaltrials.com/PatientDataRequest.htm.
Amsterdam J, McHenry L, Jureidini J. Industry-corrupted psychiatric trials. Psychiatr Pol 2017; 51: 993-1008.
Doshi P, Dickersin K, Healy D, Vedula SS, Jefferson T. Restoring invisible and abandoned trials: a call for people to publish the findings. BMJ 2013;346:f2865. https://www.bmj.com/content/346/bmj.f2865.
Jureidini J, Amsterdam J, McHenry L. The citalopram CIT-MD-18 pediatric depression trial: A deconstruction of medical ghostwriting, data manipulation and academic malfeasance. Int J Risk Saf Med 2016, 28: 33-43.
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 Psych 2001; 40: 762-72.
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; 16; 35.
McHenry LB. Re: Restoring invisible and abandoned trials: a call for people to publish the findings. BMJ 2018;346:f2865. https://www.bmj.com/content/346/bmj.f2865/rr-5.
Sussex Partnerships NHS, Guidance on the use of antidepressants in children and adolescents, Version 2, January 2014.
Wagner KD, Robb AS, Findling RL, Jin J, Gutierrez MM, Heydorn WE. A randomized, placebo-controlled trial of citalopram for the treatment of major depression in children and adolescents. Am J Psych 2004, 161: 1079-83.
November 15, 2018