Lenzner (1) recently reported that the US Food and Drug Administration
(FDA) had prevented an internal report on increased suicidality among
children on serotonin reuptake inhibitors from becoming public in February
2004. This is very alarming to the medical community because it is usually
only the regulatory agencies such as the FDA which have full access to
clinical trials data submitted from pharmaceutical companies.
The situation, however, is now different with regard to paroxetine.
Full reports of its clinical trials are publicly available on the
internet, and we can now critically examine on our own the three trials
(study 329, study 377 and study 701) that GSK has to date conducted in
paediatric and adolescent major depression. As I read their reports, I was
alarmed by three unmistakable facts that emerged.
1) What the company calls gemotional lability, explained as
crying, mood fluctuations, hostility, self-harm, suicidal thoughts and
attempted suicide is none other than suicidal tendencies. Crying
and mood fluctuations may be there but the case descriptions clearly
indicate that the reports did not label the symptoms emotional
lability unless there was suicidality.
2) Two cases of suicidal tendencies were not counted as those with emotional lability. In study 329, one patient was treated as a case of
aggression but not of emotional lability. According to the report,
however, this 14-year-old male patient had been on paroxetine for 14 days,
when he became very angry. He punched pictures, broke glass, and
sustained lacerations that required six sutures. His anger subsided, but
he expressed hopelessness and possible suicide thoughts. The patient was
hospitalized due to his severe anger outburst and a worsening of his
depression. In study 701, an 11-year-old male with the following
clinical course was counted as a case of acute exacerbation of
depression but not of emotional lability; He was randomized to
paroxetine and was titrated up to 30 mg per day. The patient stopped
taking study medication on Day 28; no reason has been provided. Two days
later, the patient threatened to harm himself and was hospitalized with an
acute exacerbation of major depressive disorder.
3) Taking all these cases of suicidal tendencies into account, the
pooled odds ratio for suicidality on paroxetine in comparison with placebo
is 2.77 (fixed effect model 95%CI: 1.03 to 7.41). The effect size is large
and no heterogeneity is present.
There is one clear lesson to be learned here. All clinical trials,
not only those conduced by drug companies but all the others included,
must be reported in details and made publicly available as reasonably soon
as possible. Without such policy internationally, neither healthcare
professionals nor consumers can ever make sufficiently informed decisions.
1 Lenzer J. Secret US report surfaces on antidepressants in children.
BMJ 2004; 329: 307
Competing interests:
I have received research grants and fees for speaking from some pharmaceutical companies, which market antidepressants (paroxetine, fluvoxamine, milnacipran, trazodone, mianserin), antipsychotics (risperidone, olanzapine, quetiapine), nootropics (donepezil) and anxiolytics (loflazepate).
Competing interests:
No competing interests
10 August 2004
Toshi A. Furukawa
Professor
Dept Psychiatry, Nagoya City University Med School, Mizuho-cho, Mizuho-ku, Nagoya 467-8601, JAPAN
Rapid Response:
No more secrets
Sir:
Lenzner (1) recently reported that the US Food and Drug Administration
(FDA) had prevented an internal report on increased suicidality among
children on serotonin reuptake inhibitors from becoming public in February
2004. This is very alarming to the medical community because it is usually
only the regulatory agencies such as the FDA which have full access to
clinical trials data submitted from pharmaceutical companies.
The situation, however, is now different with regard to paroxetine.
Full reports of its clinical trials are publicly available on the
internet, and we can now critically examine on our own the three trials
(study 329, study 377 and study 701) that GSK has to date conducted in
paediatric and adolescent major depression. As I read their reports, I was
alarmed by three unmistakable facts that emerged.
1) What the company calls gemotional lability, explained as
crying, mood fluctuations, hostility, self-harm, suicidal thoughts and
attempted suicide is none other than suicidal tendencies. Crying
and mood fluctuations may be there but the case descriptions clearly
indicate that the reports did not label the symptoms emotional
lability unless there was suicidality.
2) Two cases of suicidal tendencies were not counted as those with emotional lability. In study 329, one patient was treated as a case of
aggression but not of emotional lability. According to the report,
however, this 14-year-old male patient had been on paroxetine for 14 days,
when he became very angry. He punched pictures, broke glass, and
sustained lacerations that required six sutures. His anger subsided, but
he expressed hopelessness and possible suicide thoughts. The patient was
hospitalized due to his severe anger outburst and a worsening of his
depression. In study 701, an 11-year-old male with the following
clinical course was counted as a case of acute exacerbation of
depression but not of emotional lability; He was randomized to
paroxetine and was titrated up to 30 mg per day. The patient stopped
taking study medication on Day 28; no reason has been provided. Two days
later, the patient threatened to harm himself and was hospitalized with an
acute exacerbation of major depressive disorder.
3) Taking all these cases of suicidal tendencies into account, the
pooled odds ratio for suicidality on paroxetine in comparison with placebo
is 2.77 (fixed effect model 95%CI: 1.03 to 7.41). The effect size is large
and no heterogeneity is present.
There is one clear lesson to be learned here. All clinical trials,
not only those conduced by drug companies but all the others included,
must be reported in details and made publicly available as reasonably soon
as possible. Without such policy internationally, neither healthcare
professionals nor consumers can ever make sufficiently informed decisions.
1 Lenzer J. Secret US report surfaces on antidepressants in children.
BMJ 2004; 329: 307
2. GlaxoSmithKline Media Room http://www.gsk.com/media/paroxetine.htm
(accessed 17 July 2004)
Competing interests:
I have received research grants and fees for speaking from some pharmaceutical companies, which market antidepressants (paroxetine, fluvoxamine, milnacipran, trazodone, mianserin), antipsychotics (risperidone, olanzapine, quetiapine), nootropics (donepezil) and anxiolytics (loflazepate).
Competing interests: No competing interests