Rehash of 2001 Study Finds Paroxetine, Imipramine Ineffective, Unsafe for Teens

Results Largely Contradict Original Findings, Raise Transparency Concerns

September 30, 2015 10:12 am Chris Crawford

The antidepressant paroxetine (Paxil, Pexeva) is ineffective and unsafe for treating adolescents with depression, according to a reanalysis( of a 2001 study that was published online in BMJ on Sept. 16.

[Female doctor giving teen female a pill in exam room]

In addition, the reanalysis found that paroxetine (also sold as Brisdelle for the treatment of moderate to severe vasomotor symptoms in menopausal women) increased harms, including suicidal ideation, among adolescents who used the drug. Finally, the study confirmed that another widely used antidepressant, imipramine (Tofranil), was ineffective in treating teens for depression, even at high doses, and increased the incidence of cardiovascular events among those studied.

These findings largely contradict conclusions from the original research,( known as Study 329, which was funded by pharmaceutical manufacturer SmithKline Beecham (now GlaxoSmithKline). That original study, which was published in the Journal of the American Academy of Child and Adolescent Psychiatry, compared the safety and effectiveness of paroxetine and imipramine with that of placebo for treating adolescents diagnosed with major depression and concluded that paroxetine was safe and effective for this indication. Study 329 did find imipramine to be ineffective for treating depression compared with placebo.

Story highlights
  • Paroxetine is ineffective and unsafe for treating adolescents with depression, according to a recent reanalysis of a 2001 study.
  • The reanalysis also confirmed that another popular antidepressant, imipramine, was ineffective in treating teens for depression and increased the incidence of cardiovascular events among those studied.
  • These findings largely contradict conclusions from the original research, known as Study 329, which found paroxetine to be safe and effective in treating major depression in teens.

The reanalysis was published by BMJ under an initiative known as RIAT (Restoring Invisible and Abandoned Trials),( which encourages further review and correction of abandoned and misreported studies. The RIAT proposal offers specific definitions of these terms: "Invisible trials are those that have never been published. Abandoned trials are unpublished trials that sponsors are no longer actively working to publish or published trials that, although documented as misreported, have not been corrected by the authors."

According to the reanalysis authors, "The article by Keller and colleagues, which was largely ghostwritten, claimed efficacy and safety for paroxetine that was at odds with the data. This is problematic because the article has been influential in the literature supporting the use of antidepressants in adolescents."

Specifically, they noted, Study 329 researchers failed to follow the study protocol, even as amended. "There were four outcome variables in the (clinical study report) and in the published paper that were not specified in the protocol. These were the only outcome measures reported as significant. They were not included in any version of the protocol as amendments (despite other amendments), nor were they submitted to the institutional review board."

Further complicating reanalysis authors' attempts to rehabilitate Study 329 was the difficulty they experienced in obtaining the original set of case report forms from GlaxoSmithKline. In the end, more than a thousand pages of those reports were not retrieved despite their efforts.

Reanalysis Details

The reanalysis of Study 329 reviewed data originally collected from 275 adolescents ages 12-18 diagnosed with major depression of at least eight weeks' duration. The original study was conducted at 12 North American academic psychiatric centers (10 U.S., two Canada) from April 20, 1994, to Feb. 15, 1998. Participants spent eight weeks in double-blind, randomized and parallel-designed trials using either paroxetine (20-40 mg), imipramine (titrated to 200-300 mg) or placebo. The acute phase was followed by a six-month continuation phase.

Study 329's prespecified primary efficacy variables were change from baseline to the end of the eight-week acute treatment phase in two areas:

  • total Hamilton depression scale (HAM-D) score and
  • the proportion of responders (i.e., a HAM-D score less than or equal to 8 or a reduction of 50 percent or more in baseline HAM-D) at acute endpoint.

Prespecified secondary outcomes were changes from baseline to endpoint in the following areas:

  • depression items in the Kiddie Schedule for Affective Disorders and Schizophrenia for School-aged Children (Lifetime Version), clinical global impression, autonomous functioning checklist, self-perception profile, and sickness impact scale;
  • predictors of response; and
  • number of patients who relapsed during the maintenance phase.

Adverse events were compared primarily by using descriptive statistics.

Reanalysis results showed 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, 9 and 9.1 points, respectively, for the paroxetine, imipramine and placebo groups.

However, as previously mentioned, there were clinically significant increases in harms among both intervention groups. For paroxetine, this included suicidal thoughts and behavior, as well as other serious adverse events. Participants in the imipramine group saw increases in cardiovascular events.

Family Physician Expert Offers Take

Ken Schellhase, M.D., M.P.H., who is professor and associate director of the Center for Healthy Communities and Research in the Department of Family and Community Medicine at the Medical College of Wisconsin in Milwaukee, told AAFP News that this reanalysis of Study 329 proves family physicians need to approach industry-funded research with great caution.

"No doubt many industry-sponsored trials have been conducted, and reported, with the utmost integrity," he said. "However, this example makes clear the corrosive effect commercialism may have on scientific integrity. This is especially egregious when it relates to a population of vulnerable children."

In response to the reanalysis authors' call for greater transparency in clinical research data-sharing, Schellhase said he believes all trials that receive any public funding should make all original data available to other investigators.

"Moreover, any drug which is paid for with public dollars should likewise have all clinical trial data made publicly available as a condition of acceptance into publicly funded insurance formularies," he said. "This is consistent with the principles of science in the first place: If I conduct a trial, you should be able to replicate the results using my methods."

In support of clinical trial transparency, the AAFP recently signed on to the AllTrials USA campaign, which calls for all past and present clinical trials in the United States to be registered and report their results.

As for a good medication alternative to paroxetine and imipramine, Schellhase said the best evidence at this point leans toward using fluoxetine (Prozac, Sarafem) in adolescents and children with moderate to severe depression.

"Even still, treatment with fluoxetine or other (selective serotonin reuptake inhibitors) should be undertaken with great caution, and close follow-up," he added.

The AAFP recommends screening adolescents (ages 12-18) for major depressive disorder when systems are in place to ensure accurate diagnosis, psychotherapy and followup. This recommendation will be updated after the U.S. Preventive Services Task Force's releases its final recommendation on screening teens for depression, which was issued as a draft recommendation statement Sept. 8.

Conclusions and Implications for the Future

The authors of the reanalysis concluded that researchers and clinicians should recognize the potential biases in published research, including barriers to accurate reporting of harms, and understand that a second (or more) look at findings actually strengthens the scientific process.

"When the data become accessible to others, it becomes clear that scientific authorship is provisional rather than authoritative," the report said.

Schellhase agreed, saying confidence must be restored in the integrity of the scientific process, and he strongly supports the efforts of the AAFP and others to require open and free access to all clinical trial data.

"Even this will not always prevent the problems highlighted by this publication, but it will be a critical addition to the scientific process to discourage sloppy science, or worse, conscious deceit that risks the health of our patients -- who we have an ethical duty to serve and protect from harm," he said.

Related AAFP News Coverage
USPSTF Proposes Screening All Adults for Depression
Clinical Support Systems Should Be in Place, Says Task Force


Fluoxetine, Paroxetine Linked to Increased Birth Defect Risks
Study Didn't Find Similar Risks for Other SSRIs


please wait Processing