Accutane-Psychiatric Disorders

1
Accutane-Psychiatric Disorders

• A Pharmacoepidemiological Safety Assessment
• Robert C. Nelson, PhD
• RCN Associates, Inc. Annapolis, MD

2
Presentation Overview

• Objectives
• Literature review and model generation
• Spontaneous report reviews
• Epidemiology
• Conclusions

3
Objectives

• To determine the nature and extent of any
  relationship between Accutane therapy and
  psychiatric morbidity
• 1) Describe the types of reported psychiatric
  disorders
• 2) Identify all associated risk factors
• 3) Assess the magnitude of the identified risk
  factors
• 4) Evaluate causality within the
  pharmacoepidemiologic framework

4
Methodological Overview

• Literature Review
• determine scope of work and related disciplines
• comprehensively review and evaluate literature
  (250 citations)
• review etiology and epidemiology of psychiatric
  conditions (including suicidal behavior)
• conceptualize proposed relationships
• Review Spontaneous Reports
• evaluate spontaneous reports for category,
  quality and content
• determine the value of spontaneous reports in
  explaining proposed relationships
• Review Epidemiology - Conduct Epidemiologic
  Analyses
• evaluate relative likelihood of all risk factors
  identified
• derive relevant conclusions

5
Presentation Overview

• Objectives
• Literature review (Dr. Jacobs)
• Spontaneous report reviews
• Epidemiology
• Conclusions

6
Presentation Overview

• Objectives
• Literature review and model generation
• Spontaneous report reviews
• Epidemiology
• Conclusions

7
Spontaneous Report Case Retrieval

• WHOART SOC. 500 - All Psychiatric Cases by
  Reporter Term
• 8 functional diagnostic categories (DSM-IV)
• Reviewed reports received between 1982 and April
  30, 1999
• Worldwide

8
ResultsDistribution of Cases By Category
Mood disorders 1247 (53.1) Anxiety disorders
249 (10.6) Psychotic disorders
120 (5.1) Cognitive disturbances 174 (7.4) Sle
ep disorders 82 (3.5) Personality disorders
41 (1.7) Suicidal behavior 168 (7.1) Excluded
terms 265 (11.3) Total 2346 (100)
9
Reasons For Suspecting A Possible Relationship
Between A Drug And An Adverse Event

• Temporal association
• Dose-response
• Dechallenge
• Rechallenge
• Mechanism
• Class Effect
• Absence of alternatives

10
Methodology For Evaluation of Individual
Spontaneous Reports

• Review individual case
• Assess coded/reported term
• Assess data quality
• Assess consistency of data, including
  onset/offset
• Above relative to the 7 reasons on prior slide
• Then, review as case series

11
Interpretation of Spontaneous Reports

• When spontaneous reports are well documented and
  for rare ADRs that have background rates that are
  low, the spontaneous reports yield the most
  defensible data
• Spontaneous reports are of very diminished
  value when the outcome has a common background
  rate

12
Mood Disorder Results

• Of 1247 mood disorder reports
• 367 dechallenge reports
• 23 positive dechallenge and rechallenge reports
• 37 reports had mood disorder diagnosis subsequent
  to exposure

Dechallenge/Rechallenge Reports (34) Mood
Symptoms or Disorders
Health Professional
Consumer
Other
Total
Dx
No Dx
Dx
No Dx
Dx
No Dx
Dx
No Dx
3
14
1
8
0
8
4
30
13
Mood Disorder Results (contd)

• Even the dechallenge reports had a high level of
  diversity/inconsistency
• In summary At the individual case level, a
  small number of cases imply a causal association
  between depressive symptoms and/or mood
  disorders, and Accutane

14
Psychotic Disorder Results

• Of 120 psychotic disorder reports
• 20 dechallenge reports
• 5 positive dechallenge and rechallenge reports
• 3 reports had a diagnosis of psychotic disorder
• 100 other reports
• 9 additional reports with diagnosis
• No pattern of consistency in the available data
• In summary At the individual case level, at
  least 3 reports imply a causal association
  between the described psychotic disorder and
  Accutane administration

15
Suicidal Behavior Reports

• Total of 168 reports of suicidal behavior (1982
  to April 30, 1999)
• 104 reported suicide attempts
• 64 reports of completed suicides
• Overall suicide reports were poorly documented
  none had psychological autopsy
• No apparent dose relationship
• Ratio of M/F is 5 to 1 for completed suicide
  reports
• In summary None of the 168 reports implies
  direct causality between suicidal behavior and
  Accutane administration, at the individual case
  level

16
Presentation Overview

• Objectives
• Literature review and model generation
• Spontaneous report reviews
• Epidemiology Epidemiological Analyses
• Conclusions

17
Demographics of Accutane Patients

• Accutane use
• 70 in 15-24 age group
• 85 of males
• 55 of females

18
Prevalence of Major Depressive Disorder (MDD)
Age 15-24

• 30-day 6.1 (4.3 M8.2 F)
• 12-month 10.3 (9.5 M16.3 F)
• Lifetime 17.1 (11.0 M20.8 F)

19
MDD Morbidity During Any Six Month Exposure
Period
No Change or 80 Abated then35 Recur
152,500Current Disease
30,500 42,700
2.5 MillionUS IndividualsAged 15-24 Years
84,000
240,000By History Only
35 Recur orNo Change
182,500
25,290
1. 2 incident cases or No change
2,107,000 No Disease
???
t1
? of Accutane-Induced Cases
t2
Major DSM IV Depressive Disorder(With acne
estimates not available)
20
Observed vs Expected
All Other New Total Depressive
Incidence Incidence Symptoms MDD MDD
Disorders Observed 102 336 911 (102234) Exp
ected 25,000 152,000 500,000 -1,000,000
Major Depressive Disorder
21
Dose Time to Onset of Background Occurrences
Dose10 - 120mg
Time to Onset 0 - 180 days
22
Estimated Prevalence of Substance Abuse In
Accutane-Treated Cohort (15-24 yrs)

• 7.2 (180,000) heavy alcohol users
• Approximately same number of illicit drug abusers
  
• Assuming extensive co-morbidity, combined alcohol
  and illicit drug abuse includes approximately
  250,000 (10 of exposed) young Accutane patients

23
Alternate Risk Factors

• Nearly one-quarter million alcohol/substance
  abusers are amongst the 2.5 million of
  Accutane-exposed cohort
• Many of these abusers are also comorbid with the
  152,000 incident mood disorder cases or with
  people with depressive symptoms or different
  disease
• The total numbers in the cohort with some form
  DSM-IV disorder may be up to 16-20 (up to 1/2
  million) of the total cohort

24
Psychotic Disorders - Alternate Risk Factors

• Prevalence in ages 15-24 is 1 (25,000 expected
  amongst exposed)
• Prevalence in ages 25-34 is 2 (14,000 expected
  amongst exposed)
• Add alcohol and drug abuse covariates (gt3,900
  expected amongst exposed)

25
Suicidal Behavior
26
Risk of a Serious Suicide Attempt Adjusted for
Confounding Mental Disorders, by Gender and Age,
for 302 Persons Who Made Serious Suicide Attempts
and 1,028 Comparison Subjects
Age lt 30 Years
Age gt 30 Years
95 Population 95 Population DSM-III-R
Category Odds Confidence Attributable
Odds Confidence Attributable By Gender
Ratio Interval P Risk Ratio Interval P
Risk
WOMEN Any mood disorder 21.3 9.4-48.3 lt
0.0001 79.2 58.3 26.3-129.4 lt 0.0001 80.8 Any
substance use 2.7 1.0-7.3 gt 0.05 22.4 3.8 1.2-11.
5 lt 0.03 26.2 disorder Any antisocial
disorder 2.3 0.9-6.2 gt 0.05 21.5 1.2 0.2-9.1 gt
0.05 6.3 Any anxiety disorder 1.5 0.5-4.7 gt
0.05 7.0 1.3 0.5-3.3 gt 0.05 5.0 Any eating
disorder 8.6 0.8-91.7 gt 0.05 14.8 1.1 0.5-6.5 gt
0.05 1.6
SOURCE Beautrais, A.L, et al. (1996)
Prevalence and Comorbidity of Mental Disorders in
Persons Making Serious Suicide Attempts A
Case-Control Study. American Journal of
Psychiatry, 153(8) 1012
27
Observed vs Expected Suicidesin Accutane-Exposed
Cohort (United States Data and Estimates)
Observed Suicide by Gender Age
Male (33) Female (5) Under 25 Over
25 Unk Under 25 Over 25 Unk Total 27 4 2 5 0 0
38 Male (319) Female
(79) Under 25 Over 25 Under 25 Over
25 Total 262 57 48 31 398
Expected Suicide by Gender Age
28
Presentation Overview

• Objectives
• Literature review and model generation
• Spontaneous report reviews
• Epidemiology
• Conclusions

29
ConclusionMood Disorders

• There are a small number of reported cases that
  imply causality between depressive symptoms or
  mood disorders and Accutane administration, at
  the individual case level.
• However, an assessment in the context of natural
  history and alternative risk factors provides
  strong evidence that the described symptomatology
  and disorders are much more likely to be
  associated with factors other than Accutane.
• Unfortunately the analysis of these kinds of data
  do not allow any potential risk factor to be
  completely ruled out, no matter how unlikely it
  may appear.

30
ConclusionPsychotic Disorders

• There are a very small number (3) of reported
  cases that imply causality between a described
  psychotic disorder and Accutane administration,
  at the individual case level.
• However, an assessment in the context of natural
  history and alternative risk factors provides
  strong evidence that the described symptomatology
  and disorders are much more likely to be
  associated with factors other than Accutane.
• Unfortunately the analysis of these kinds of data
  do not allow any potential risk factor to be
  completely ruled out, no matter how unlikely it
  may appear.

31
ConclusionSuicidal Behavior

• There are no reports amongst the 168 reviewed
  that imply direct causality between suicidal
  behavior and Accutane.
• An assessment in the context of natural history
  and alternative risk factors provides strong
  supporting evidence that the reported cases are
  much more likely to be due to factors other than
  Accutane.

32
Overall Conclusion

• Given
• no biological plausibility
• no consistent patterns in the data
• complex environment of background symptoms
• very high background rates of disease
• very high background rates of alternative risk
  factors
• There is no evidence in these data to support a
  causal relationship between Accutane
  administration and psychiatric disorders