What Really Happened in the Original Fluoxetine (Prozac) Trials for Children and Adolescents?

Antidepressant use soared in the 1990s due to marketing, direct to physician and direct to consumer advertising producing a controversial era in which drug makers directly targeted consumers for their products.  As a result of the antidepressant marketing campaign, depression diagnoses also surged during this period.  Within a decade, the diagnosis of depression increased from 1 in 10,000 people to 1 in 10 people – simply a coincidence or the anticipated end goal?

The momentum in marketing extended into 2020 when 75% of TV advertisements were for pharmaceutical drugs, a statistic said to have increased since.  Pharmaceutical companies pay the media an estimated $6 billion a year on advertisements directly targeting consumers; this statistic from 2016 has likely increased since.  This is a 360% increase since 1997 when the FDA first allowed pharmaceutical advertising to consumers.

Scaling the walls of ethics and medicine, repeatedly seen in the media and medical literature is the lockstep effort of clinicians, marketers and manufacturers, also known as the “psychopharmaceutical complex”, publishing, pushing and marketing pharmaceuticals based on misrepresented, misinformed and even outright false data.

On the coat-tails of this ‘Anti-Depressant Era’, the next segment of the population to be marketed were children and adolescents.  What did the original Eli Lilly trials show for the use of Prozac for depression in children and adolescents?

Prozac, Beginning of the Anti-Depressant Era

The two most common classes of antidepressants are the selective serotonin reuptake inhibitors (SSRIs) and selective norepinephrine reuptake inhibitors (SNRIs).  Fluoxetine, commonly known by its brand name Prozac, was the first SSRI to be marketed in the US and is one of the most commonly prescribed SSRIs, as described in Prozac, Beginning of the Anti-Depressant Era.  It became FDA-approved in 1987 for the treatment of depression in adults.

In 2002, Prozac gained FDA approval for the treatment of depression in children and adolescents based on data from two peer-reviewed clinical trials (1997 trial, 2002 trial).  The trials concluded no major safety concerns and a small benefit over placebo for treatment of depression in children and adolescents.

Re-Analysis and Restoration of the Original Prozac Trials for Children and Adolescents

Researchers can “restore” original trial publications by analyzing data and documents originally submitted to drug regulators by drug companies.  Restoring Invisible and Abandoned Trials (RIAT) corrects misreporting and invisibility by publishing an unpublished trial, republishing a trial or just the parts initially misreported, providing people accurate information.

The two clinical trials used as the basis for Prozac’s drug approval in 2002 for children and adolescents have been re-analyzed by RIAT, deeming it both unsafe and ineffective.

Behind the re-analysis of the original trials are two researchers, Peter Gotzsche and David Healy.  Healy was an early advocate of Prozac.  They obtained clinical study reports and protocols from the Medicines and Healthcare products Regulatory Agency (MHRA), a UK agency responsible for ensuring the safety of medicines.  These reports were compared to the original trials in which major differences were found in what was reported versus what was published in the medical journals.  Their findings are published in the systematic review: “Restoring the two pivotal fluoxetine trials in children and adolescents with depression.”

Healy and Gotzsche Stance on Antidepressants in the General Population

Healy concluded, “SSRIs might benefit some people in the short term, but he finds little reason to think they help things to turn out better in the long run.  Most worryingly, he presents persuasive evidence the SSRIs make a significant number of people suicidal and a larger number addicted.”

Gotzsche wrote in the Lancet Psychiatry in 2014 about antidepressants: “One patient might benefit for every ten patients treated.  I believe those results were exaggerated…With respect to harms of antidepressants, most will experience side effects, of which the most frequent is sexual problems.  Sexual problems developed in 59% of patients who all reported no problems with sexual function before they started using an antidepressant.  Even when tapering off them slowly, half the patients have difficulty stopping the drugs because of withdrawal effects, which can be severe and long-lasting.  We noted that withdrawal symptoms were described in similar terms for benzodiazepines and SSRIs…However, they were not described as dependence for SSRIs.  To define similar problems as “dependence” in the case of benzodiazepines and as ‘withdrawal reactions’ in the case of SSRIs is irrational.  For patients, the symptoms are just the same; it can be very hard for them to stop either type of drug.”

With these conclusions of antidepressant research analyses available, why would we think it safe to give to children and adolescents still in the stages of development?

What Led to Re-Analysis of the Original Trials for Children and Adolescents?

As mentioned, fluoxetine was initially FDA-approved in 1987 for adults and later in 2002 for children and adolescents even though the FDA noted the drug did not have a statistically significant benefit in either children and adolescent trial.

Just two years after the FDA approved Prozac for children and adolescents, the FDA issued a black box warning for SSRIs and SNRIs for increased risk of suicidal thinking and behavior in children and adolescents.  The initial study of paroxetine (Paxil), another common SSRI, doubled the risk of suicidal events in restoration of the trial.

At the same time, the risk of harms related to suicidality and violence was also found to be doubled by SSRI and SNRI use in a meta-analysis of placebo-controlled trials of healthy adult volunteers.

A 2014 study entitled “Differences in reporting serious adverse events in industry sponsored clinical trial registries and journal articles on antidepressant and antipsychotic drugs: a cross-sectional study,” examined serious adverse events from antipsychotic and antidepressant drug trials reported in journal articles and clinical trial summaries for both children and adults.  This study found significant discrepancies in data reporting:

• 43% of serious adverse events (SAEs) reported in trial summaries did not appear in journal articles
• 60% of SAEs in articles and 41% in trial summaries contained no description
• 62% of deaths and 53% of suicides were not reported in articles
• 50% of suicides are not reported and suicidal events “are often called something else” such as depression, emotional lability and hospital admission.

A 2016 systematic review and meta-analyses found Eli Lilly’s website to be misleading as 90% of suicidal events, suicidal ideation events, most cases of aggression and akathisia (inability to remain still, a precursor to aggression and violence) among people on fluoxetine and duloxetine (Cymbalta) were labeled incorrectly, excluded without explanation, or missing in the published trial reports.

In published psychiatric drug trials for both children and adults, about half of suicides are missing and suicidal events are often mislabeled as something else.

In light of above information, the original trials for the use of fluoxetine in children and adolescents were re-analyzed.

Major Discrepancies Found in the Original Trials of the Use of Prozac in Children and Adolescents Leading to Re-Analysis of Data  

Trial 1 consisted of 96 patients aged 7-18 with major depressive disorder and were treated for 8 weeks.

Trial 2 consisted of 219 patients aged 8-17 with major depressive disorder treated over the course of 19 weeks.

In addition to the small number of patients enrolled, when restoring the original trials, the following major discrepancies were found:

1. There were missing data and dropouts resulting in outcomes biased towards fluoxetine (demonstrated by missing data and dropouts).
2. 11 patients in the fluoxetine group and 3 in the placebo group used drugs with sedative properties (11 vs 3). This could obscure harms of fluoxetine such as agitation.
3. The trial was stated to be blinded but was not, meaning the trial investigators were aware of which patients received placebo or fluoxetine creating a bias to the outcome.
4. Patients recruited in Trial 1 already on antidepressants were only given a one week wash out period of their current medication before starting the trial randomization meaning some patients were experiencing withdrawal symptoms. In Trial 2, there was a two-week no drug period but it was not explained if new drugs were not allowed or if previous antidepressant treatment was discontinued.
5. Numbers were either inconsistent or not reported without explanation.
   1. More data were missing for patients in the placebo group than patients in the fluoxetine group.
   2. In the non-solicited adverse events table, Eli Lilly wrote there were no statistically significant differences but simultaneously did not note more patients on fluoxetine than on placebo experienced one or more events.
   3. In Trial 2, an additional 5 patients in the placebo group were excluded with no explanation why while no patients in the fluoxetine group were excluded.
6. Information was indexed in the clinical trial report but missing from the published report.
7. Patients were not routinely asked what concomitant medication they were on and more patients on fluoxetine than placebo used concomitant medication.
8. Patients were more severely affected than described and experienced irritability, insomnia, fatigue, decreased concentration, anger, hearing voices, mood swings, racing thoughts and temper tantrums, all suggesting drug induced psychosis.

Also, in the restoration of data it was found:

• Treating 6 patients with Prozac led to the harm of 1 person.
• Treating 10 patients with Prozac led to severe harm in 1 person.
• Fluoxetine reduced weight and height over 19 weeks by 1.1 kg and 1 cm.
  • There was no comment on these harms.
  • The FDA requested Eli Lilly conduct a one-year study evaluating fluoxetine’s effect on growth as this is a concern in developing young people. The company declined to do this study.  To date, this study has not been done.
• Patient rating scores did not find fluoxetine effective and no meaningful benefit on depression was found between fluoxetine and placebo.

“The company violated its trial protocols; essential information was scattered around in the reports; a lot was missing despite being indexed; there were many errors, inexplicable numerical inconsistencies, and unexplained exclusions of patients from analyses; and results that were inconsistent with the conclusion that fluoxetine is safe and effective were side-lined or explained away in a disturbing manner.”

Despite the significant flaws and discrepancies, Eli Lilly concluded fluoxetine safe and effective for children and adolescents with major depression and the FDA approved it.  It is also interesting to note, six of the eight authors of the study were Eli Lilly employees and the other two were paid consultants for Eli Lilly.

Conclusion

• Eli Lilly left out critical information without explanation including but not limited to attempted suicides and severe adverse events predisposing patients to violence and suicidality.
• Requests for corrections to the publications for both paroxetine (Paxil) and fluoxetine (Prozac) have been submitted with neither being corrected.
• The editor of JAMA Psychology rejected the concerns from the restoration of the trials and corrections have not been made. Time will tell if the record will be corrected as Prozac continues to be the preferred SSRI for pediatric patients and one of the most prescribed medications in the world.
• Re-analysis of the original fluoxetine (Prozac) trials for children and adolescents shows “fluoxetine is unsafe and ineffective.”

Have an awesome day!  Dr D and Drew Chernisky, PA