Characteristics of Pediatric Antidepressant Trials

1
Characteristics of PediatricAntidepressant
Trials

• Gregory M. Dubitsky, MD
• Division of Neuropharmacological Drug Products
• FDA

2
Objective

• Descriptive only.
• Will not address effect of study characteristics
  on suicidality risk
• See Dr. Hammads Presentation

3
Study Pool

• 23 placebo-controlled studies conducted
  between 1984 and 2001.
• 9 drugs.
• 5 indications (major depression, OCD, GAD,
  SAD, ADD).

4
Common Design Features

• Randomized.
• Double-blind.
• Placebo-controlled.
• Parallel group.
• Flexible-dose.

5
Basic Study Design

• See Handout Table 1
• Drug, study, indication.
• Age range.
• Number of patients in analysis by TX.
• Duration of double-blind TX.
• Protocol-specified dose range.

6
Screening and Key Exclusionary Criteria

• See Handout Table 2
• Extensive diagnostic screening.
• Placebo lead-in preceding DB TX (and exclusion
  of placebo responders).
• Exclusionary criteria H/O treatment
  resistance, current suicide risk, H/O suicide
  attempt, bipolar disorder, family H/O bipolar
  disorder.

7
Other Study Features (see Appendix to my review)

• study dates/location/number of centers.
• stratified randomization by age group.
• exclusionary criteria
• -homicidal risk.
• -psychotic symptoms.
• -alcohol/drug abuse.
• -borderline personality disorder.
• -eating disorder.

8
Notable Study Differences

• Prozac HCCJ (MDD) small terminated early.
• One active-control study
• - Paxil 329 (MDD) (imipramine).
• Two studies included inpatients
• - Celexa 94404 (MDD) Wellbutrin 75 (ADD).
• Three studies used extensive DX screening
• -Prozac X065 HCJE Paxil 329 (all MDD).

9
Post-DB Phase TX Options

• Variable across trials
• Taper of acute TX (8 trials).
• Abrupt D/C (7 trials).
• Open-label TX (5 trials).
• Continued DB TX (3 trials).
• Also variable within trials
• e.g., Paxil 329 (MDD) - responders continued
  DB TX, non-responders tapered.
• Variable follow-up hindered W/D effect analysis.

10
No study was specifically designed to assess
suicidality.

• suicide attempts and ideation detected only by
  routine safety monitoring
• -treatment-emergent adverse events.
• -suicide-related items on depression scales
  (CDRS-R, K-SADS, HAM-D, and MADRS).
• AE description often incomplete or vague.

11
Conclusions

• Potential influence of study design
  characteristics on suicide risk
• See Dr. Hammads analysis.