FDA’s Waiting Game Exposes Children to Preventable Risks

Comments by Vera Hassner Sharav, President, The Alliance for Human Research Protection

Submitted to
FDA Advisory Committee Psychopharmacological Drugs Advisory Committee and Pediatric Subcommittee Drugs Advisory Committee
September 13-14, 2004 Meeting
RE: Suicidality Associated with Antidepressant Drug Treatment

The Alliance for Human Research Protection (AHRP) wishes to alert the committee that the two analyses by FDA medical experts – demonstrating a twofold risk of suicide-related behaviors in children prescribed an antidepressant compared to placebo – are confirmed by a much larger body of analysis of adverse drug effects linked to antidepressants. The finding of a twofold risk for children taking an antidepressant has been consistently validated and corroborated by every independent review of the SSRI clinical trials before and since the Feb 2, 2004 advisory committee meeting. In particular, two independent teams in the UK and Australia published their meta-analyses in prestigious peer-reviewed journals, The Lancet and the British Medical Journal,[1] confirming both a lack of efficacy of the drugs and an increased risk of suicide-related behaviors. But these reports have not been introduced as relevant evidence to the committees and not one of the authors has been invited to address the panel. On the contrary, FDA officials have provided the panel with selective information, much as drug companies have withheld data that contradicted their own claims about the efficacy and safety of antidepressants.

In fact, at the Feb 2, 2004 meeting, FDA officials prevented the agency’s own expert medical officer, Dr. Andrew Mosholder, from presenting his report and his resulting recommendations to restrict the use of antidepressants in children – which is what the British Committee on Safety in Medicines had done. When questioned by the British Medical Journal[2], Dr. Temple of the FDA defended the embargo on Dr. Mosholder’s report, maintaining that both the raw data and Mosholder’s interpretation were “imperfect,” and suggesting that some behaviors labeled suicidal were “highly suspect and could have been accidents.” Dr. Mosholder identified 78 cases of serious, suicide-related behavior. The Columbia panel reviewed the same data and identified an additional 17 events (95 cases total). Dr. Temple, how certain do you need to be before you take action?

The real problem confounding the integrity of clinical trial findings and the significance of the adverse incidence reports would largely be resolved if drug manufacturers did not have quasi-total control over trial design and subject selection, and total proprietary possession of the data and total control over its release. Manipulation and concealment of adverse event data has corrupted the entire enterprise of testing drugs and identifying their benefits and dangers. Industry’s disregard for regulation that interferes with marketing is best demonstrated by a 1998 GlaxoSmithKline memo detailing how failed trial data would be withheld.[3] Concealment of clinical data is now at the center of a public debate about drug safety standards, the credibility of unvalidated company-controlled data, the integrity of the scientific literature, and the erosion of public trust in medicine.

FDA’s inclusion/exclusion criteria for invited presentations at its public advisory meetings were designed to prevent a fair and open debate inasmuch as independent scientists, whose published reports triggered the current debate, have been excluded. Ironically, those invited to make presentations – Eli Lilly and Company and Pfizer[4] – continue to conceal evidence of adverse drug effects, apparently with FDA’s blessing. The conclusion seems inescapable: FDA’s standards for proving drug efficacy and safety are designed to help industry manufacture favorable results. To prove efficacy requires merely eliciting an effect greater than placebo on symptom rating scales. Dr. Thomas Laughren [5] acknowledged the particularly low bar for showing efficacy in antidepressant trials. Dr. Bernard Carroll, a past chairman of the FDA Advisory Committee for Psychotropic Drugs, also noted “a dumbing down of expectations for antidepressant efficacy in recent years.” [6] A miniscule difference between the drug and placebo is often termed “statistically significant.” But what is the value of a “statistically significant” effect if it is clinically trivial?

At the Feb 2 meeting, Dr. Temple acknowledged that the small pediatric studies were essentially designed to test efficacy, not safety. Thus, the studies were not designed to detect relatively uncommon adverse effects such as suicidal ideation, suicide attempt, or suicide. Indeed, participants in clinical trials are excluded if they are considered at risk of such behavior during initial screening. The emergence of such serious drug safety problems in these small and mostly failed efficacy trials underscores that the risk of increased suicidality is likely to be genuine. FDA’s own analysis found a 2.52 increased suicide-related risk in sertraline (Zoloft) trials; a 2.19 increased risk in paroxetine (Paxil) trials; a 2.54 increased risk in citalopram (Celexa) trials; and an overall 1.89 increased risk in all pediatric trials. Can you say with certainty that these are random occurrences? Can you say this even merely with “confidence”? Dr. Temple, how certain do you need to be before you take action to protect children?

The panel needs to know that these drugs pose a much higher cumulative risk for severe adverse effects whose incidence rates are much higher than suicidality: FDA’s medical review linked Prozac to stunted growth,[7] prolonged QT interval. Other studies link mania and an alarming rise in manic-depression in children to increased use of antidepressants, defining the phenomenon as “manic conversion.” A recent Yale analysis of 88,000 mental health users, aged 5 to 29, concluded: “Treatment with antidepressants is associated with highest conversion hazards among children aged 10 to 14 years.”[8] Given that these drugs repeatedly failed to demonstrate a drug benefit for children in controlled trials, how do the FDA and NIMH justify exposing children to reduced growth, mania, manic-depression, and suicide?

FDA’s failure to put children’s welfare ahead of corporate interests led New York State Attorney General, Eliot Spitzer, to file a lawsuit charging GlaxoSmithKline with engaging “in repeated and persistent fraud”[9] by concealing negative findings from physicians about Paxil. The charge can be leveled against all manufacturers of SSRIs. The consent agreement[10] (reached between GSK and the AG in less than three months) raises the standard of disclosure in all GSK clinical trials. Eli Lilly and others have indicated plans to post their unpublished trials as well – but “plans” and “intentions” are not enforceable. Whereas the British regulatory agency took precautionary action – all but banning antidepressants for children – as soon as its panel of experts analyzed the data, the FDA has stalled and thrown roadblocks to impede full public disclosure of the risks. The revised, FDA-approved labels use equivocating language suggesting that the risk for suicide is equal “whether or not [patients] are taking antidepressant medications.” [11]

The Best Pharmaceuticals Children’s Act requires manufacturers to disclose pediatric trial findings on drug labels – but the FDA has allowed SSRI drug manufacturers to conceal failed efficacy tests results in violation of the law. Experiments such as the Prozac study HCJE (Emslie, 2002)[12] raise serious concern about the children’s safety in these trials. The stated objective was to compare the efficacy of Prozac at 20 mg to placebo. But those who responded to placebo (phase 1) were discontinued, and those who couldn’t tolerate Prozac at 10mg (phase 2) were also excluded – raising doubts about the integrity of the results claimed. Furthermore, children as young as 8 years were pushed to 40 and 60 mg (phase 5), then randomized to drug or placebo for 32-weeks (phase 6). Inasmuch as most severe adverse events, such as suicidal acts, occur when doses are changed, why did FDA’s analysis[13] of suicidality exclude these clinically and scientifically crucial phases of the experiment?

FDA’s failure to warn about the risks and to disclose failed trial outcomes, coupled with its failure to stop drug companies from making false and misleading claims in advertisements and sponsor-manipulated published reports that hide the unfavorable data led to the promulgation of unsubstantiated pronouncements that mislead healthcare professionals and the public about the safety and efficacy of SSRIs. For example, Drs. Brent and Birmaher[14] (2002) asserted in The New England Journal of Medicine: “SSRIs are the most commonly used treatment for adolescent depression, because of the proven efficacy of fluoxetine, citalopram, and paroxetine in placebo-controlled trials, with a response rate of approximately 60 percent and a favorable side-effect profile” (p. 668). FDA’s own analysis refutes these preposterous assertions.[15] Yet, these assertions have undoubtedly helped fuel the spiraling increased use of psychotropic drugs in children and adolescents in the US.[16] [17] [18] [19] [20] The greatest percentage increase of psychotropic drug use for children (between 1995 and1999) was for SSRIs (62%), from 7.9 per 1000 children in 1995 to 12.8 per thousand in 1999.

If the FDA is truly concerned about the safety of patients, its safety standard needs to be upgraded to safeguard children and adults from hazardous drugs. The safety of patients should not be sacrificed for industry’s financial interests and market shares. If these widely prescribed drugs are not beneficial for most children, and if they pose life-threatening risks for some children, how can the FDA possibly justify its failure to warn?

The AHRP respectfully reminds the committee that the issue of the causation of suicide-related events in clinical trials of SSRIs is separate from the issue facing it today. Why and how psychotropic drugs increase the risk of suicide-related behaviors is a complex question, whose elucidation is urgently needed. This question, however, must not distract the committee from its responsibility to advise the FDA on necessary actions to undertake given the established association between exposure to SSRIs and increased risk of suicide-related events in children and adolescents.

A related concern is that the FDA and investigators funded by pharmaceutical corporations have studiously ignored the crucial issue of identifying predictors of response or nonresponse to antidepressant drugs. The scientific standard of antidepressant trials has degenerated to the lowest common generic denominator, a practice that may potentially expand the market for the drugs but that ensures that many patients with the nonspecific diagnosis “major depression” are exposed to all the risks but none of the benefits of these agents.

AHRP recommendations for immediate FDA action:

(1) information concerning the increase in suicide-related events associated with exposure to antidepressants in pediatric clinical trials be placed at the front of product labels of antidepressants of the SSRI and SNRI class;

(2) this information should be contained in a bold black box warning in the product label;

(3) manufacturers of SSRIs (selective serotonin reuptake inhibitors) and SNRIs (selective serotonin norepinephrine reuptake inhibitors) should be required to send “Dear Doctor” letters to the nation’s physicians detailing the risk of suicide-related events associated with their products, reminding physicians that these products have not demonstrated a benefit for children and have not been approved by the FDA for pediatric use.

[1] Jureidini JN, Doecke CJ, Mansfield PR, Haby MM, Menkes DB, Tonkin AL, Efficacy and safety of antidepressants for children and Adolescents, British Medical Journal, Apr 2004; 328: 879 – 883. online at: http://bmj.bmjjournals.com/cgi/content/full/328/7444/879?

[2] Lenzer J. Secret US report surfaces on antidepressants in children, British Medical Journal, Aug 7, 2004, 329: 307

[3] The memo was first revealed by Dr. David Healy at a press briefing convened by the Alliance for Human Research Protection [3] (AHRP) on Feb. 2. http://www.ahrp.org/risks/SSRI0204/GSKpaxil/index.php

[4] Armstrong D and MATHEWS AW. Pfizer Case Signals Tougher Action On Off-Label Drug Use, The Wall Street Journal. May 14, 2004; Page B1

[5] Dr. Laughren participated in a symposium at the University of Texas in Houston, April. 2000, at which he acknowledge the low threshold met by antidepressants.

[6] Carroll BJ. Sertraline and the Cheshire Cat in Geriatric Depression, American Journal of Psychiatry, 2004 Apr;161(4):759

[7] Mosholder A. Medical Review. Fluoxetine-Prozac Application 18-936/SE5-064, 2001

[8] Martin A, Young C, Leckman JF, Mukonoweshuro C, Rosenheck R, Leslie D. Age effects on antidepressant-induced manic conversion. Arch Pediatr Adolesc Med. 2004 Aug;158(8):773-80. See also: Martin A, Young C, Leckman JF, Mukonoweshuro C, Rosenheck R, Leslie D. Age effects on antidepressant-induced manic conversion. Arch Pediatr Adolesc Med. 2004 Aug;158(8):773-80.

[9] Office of New York State Attorney General. Press Release, June 2, 2004


[10] Harris G. Maker of Antidepressant Drug to Release All Trial Results, The New York Times, August 26, 2004, p. A-1 http://www.nytimes.com/2004/08/26/business/26CND-DRUG.html?ex=1094539557&ei=1&en=b5e0eacae077a2e7

[11] See FDA-approved (April 2004) label: http://www.celexa.com/prescribing_information/celexa_pi.pdf

[12] Emslie GJ, Heiligenstein JH, Wagner KD, Hoog SL, Ernest DE, Brown E, Nilsson M, Jacobson JG. Fluoxetine for acute treatment of depression in children and adolescents: a placebo-controlled, randomized clinical trial. J Am Acad Child Adolesc Psychiatry. 2002 Oct;41(10):1205-15.

[13] Dubitsky G. Review & evaluation of clinical data placebo-controlled antidepressant studies in pediatric patients, FDA, August 6, 2004. http://www.fda.gov/ohrms/dockets/ac/04/briefing/2004-4065b1-08-TAB06-Dubitsky-Review.pdf

[14] Brent, D.A., & Birmaher, B. (2002). Adolescent Depression. The New England Journal of Medicine, 347, 667-671.

[15] Llaughren P. U.S. Food and Drug Administration CDER. Memorandum to Members of PDAC and PEDS AC January 5, 2004.

[16] Safer DJ. Changing patterns of psychotropic medications prescribed by child psychiatrists in the 1990s. J Child Adolescent Psychopharmacoogyl. 1997. 7:267-274; Zito JM, Safer DJ, dosReis S, Gardner JF, Magder L, Soeken K, Boles M, Lynch F, Riddle MA. Psychotropic Practice Patterns for Youth A 10-Year Perspective, Archives of Pediatric and Adolescent Medicine, 2003. 157:17-25.

[17] Summary, Evidence Report/Technology Assessment Number 7: Treatment of Depression – Newer Pharmacotherapies. Rockville, Md: Agency for Health Care Policy and Research; March 1999. Available at: http://www.ahcpr.gov/clinic/deprsumm.htm

[18] Olfson M, Marcus SC, Weissman MM, and Jensen JS. National trends in the use of psychotropic medications by children. Journal of the American Academy of Child and Adolescent Psychiatry. 2002. 41:514-521

[19] Shatin D and Drinkard C. Ambulatory use of psychotropics by employer-insured children and adolescentsin a national Managed Care organization. Ambulatory Pediatrics, 2002, 2: 111-119.

[20] Zito JM, Safer DJ, DosReis S, Gardner JF, Magder L, Soeken K, Boles M, Lynch F, Riddle MA. Trends in the prescribing of psychotropic medications to preschoolers. Journal of the American Medical Association 2000, 283:1025-1030.

Alliance for Human Research Protection (AHRP)

Tel. 212-595-8974
Fax: 212-595-9086
142 West End Ave. Suite 28P, New York, NY 10023