Transcript Open presentation - CTN Dissemination Library

NIDA
NATIONAL INSTITUTE
ON DRUG ABUSE
National Drug Abuse Treatment
Clinical Trials Network
A Randomized Controlled Trial of OROS-MPH + CBT in
Adolescents with ADHD and Substance Use Disorders:
Implications for Clinical Practice and Study Design
Paula Riggs MD
Professor of Psychiatry
University Colorado School of Medicine
NCDEU Boca Raton Florida, June 14, 2010
DISCLOSURES
MC NEIL PROVIDED ACTIVE MEDICATION
AND MATCHING PLACEBO FOR
MULTISITE TRIAL
GRANT SUPPORT PROVIDED BY THE
NATIONAL INSTITUTE ON DRUG ABUSE
2
ACKNOWLEDGEMENTS
NIDA CTN
11 participating community-based adolescent substance
treatment programs
CTN 0028 Lead Team
David Lui MD, NIDA
Co-Investigators:
Theresa Winhusen PhD
Robert Davies MD
Connie Klein LCSW
Marilyn Macdonald BA
Michelle Lohman RN
Suzell Klein MA
3
Study Overview
Background and Significance
 ADHD 3-5x more common in adolescents
with SUD (30-50%) compared to those
without SUD (7-10%)
 Associated with poorer treatment outcomes
 The safety and efficacy of psychostimulant
medication for ADHD in youths with cooccurring SUD is not known
Study Aims
AIM 1: To evaluate the efficacy of OROSMPH vs. placebo in adolescents with
ADHD
and SUD
concurrent
DSM IV ADHD
Checklist
(adolescentCBT
weekly,
parent 8 and 16 wks); CGI- I
AIM 2: Impact of OROS-MPH + CBT vs
placebo + CBT on substance treatment
outcomes TLFB baseline, monthly; UDS, weekly
AIM 3: Safety, tolerability, abuse
AES, SAEs, weekly; abuse /diversion questionnaire
Study Design
16-week randomized controlled trial
• OROS-MPH+ CBT* vs placebo + CBT
(target dose = 72 mg/day)
•
•
300 adolescents (13-18) with co-occurring
ADHD and SUD
11 participating sites
(community treatment programs)
Cognitive Behavioral Therapy
•
Weekly, individual, manual –standardized across
sites
Analytic Approach
Intent-to-treat (ITT)
 All randomized participants
 5% Type I error; two tailed
Power
 N=300 across 11 sites


> .80 power OROS-MPH vs placebo on ADHD
.80 power to detect mediated effect size
= or > 0.4 on primary substance treatment outcome
(days of past 28 day drug use)
Inclusion and Exclusion Criteria
Inclusion Criteria
 Adolescents 13-18
Exclusion Criteria
 Serious medical/cardiac
illness
years
 ADHD
 Bipolar, psychosis,
 DSM IV criteria (KSADS)
suicide risk, tic disorder
 ADHD-RS DSM IV symptom

checklist > = 22
 SUD
 DSM IV criteria at least 1
non-tobacco SUD
 past month use
Opiate dependence;
methamphetamine
abuse/dependence
 MH/SUD treatment
 Psychotropic medication
 Medically healthy
9
Rationale for Trial Design
Advantages
• Internal validity
• Decrease variability of
Disadvantages
• ? impact of new
intervention if
added to TAU
alone
TAU across sites
• Hybrid efficacy
/effectiveness trials
necessary if adequate
• ? Impact if
power in single site
different
trial can’t be achieved
behavioral tx
platform selected
Main Study Findings
1334 Individuals Screened by Telephone
Study Flow Diagram
450 Assessed for Eligibility and Consented
• Medication compliance = 80%
• Tolerability
147 Excluded (32.7%)
139 Not eligible (94.6%)
• 96% achieved 72 mg daily dose
8 Other (5.4%)
• 86% sustained across study
• Dose reduction
303 Participants Randomized
• OROS-MPH 16/149 (10.7%)
• Placebo
10/148 (6.8%)
151 Assigned to Receive OROS-MPH + CBT
118 Completed Trial (78.1%)
33 Non-Completers (21.9%)
11 Withdrew Consent
3 Moved from Area
2 Practical Problems
4 Incarceration
1 Pressure/Advice from Outsiders
9 Failed to Return to Clinic and Lost
3 Other
That’s weird
151 Included in Analysis
152 Assigned to Receive Placebo + CBT
109 Completed Trial (71.7%)
43 Non-Completers (28.3%)
11 Withdrew Consent
1 Moved from Area
3 Practical Problems
5 Incarceration
1 Pressure/Advice from Outsiders
17 Failed to Return to Clinic and Lost
1 Feels Treatment not Working
4 Other
152 Included in Analysis
Demographics
All
non-significant
differences
Race, Ethnicity
All non-significant differences
14
Baseline Clinical Characteristics
ADHD Diagnostic Subtype
70
60
50
40
68.6%
30
28.1%
20
10
2.6 %
0
inattentive
8 INATTENTIVE ; 3.4 HI
HI
5.1 INATTENTIVE; 7.0
HI
ADHD NOS=0.7% (n=2)
combined
8.4 INATTENTIVE ; 7.8 HI
Baseline Clinical Characteristics
ADHD
DSM IV ADHD symptom
checklist score (adolescent)
OROS-MPH + CBT
N=151
PLACEBO + CBT
N=152
38.1 (9.0)
39.3 (8.7)
2.1 (1.2)
1.9 (1.3)
14.0/28 days
(9.6)
15.1/28 days
(9.4)
Substance Diagnoses
& Past 28 day use
No. of Abuse +
Dependence dx
Days of past 28 day
drug use
16
Baseline DSM-IV Substance Use Disorders (CIDI)
Baseline DSM-IV Substance Use Disorders (CIDI)
ADHD OUTCOMES
19
Adolescent DSM IV ADHD Checklist by Treatment Group
• Clinically and statistically significant reduction in
Clinically and statistically significant decrease in ADHD
ADHD
inbut
both
groups between
symptomssymptoms
in both groups
no difference
groups + CBT 45%; p<0.0001)
(OROS + CBT 46%; placebo
• No difference
between
• OROS-MPH
(-19.2, groups
50% reduction, p<0.001)
• Placebo ( -21.2, 54% reduction, p<0.001)
OROS-MPH: CGI-I ADHD treatment responders =23.6%
Placebo: CGI-I ADHD treatment responders =21.3%
NS
20
Secondary
ADHD
Outcome
Measures
Mixed results
Secondary outcomes suggest some
“added benefit” with OROS-MPH
P<0.02
compared to placebo
Mean diff = 7
P=0.002
+7
Mean diff = 4
P=0.023
+4.2
ARCQ
ARCQ
P<0.0015
ARCQ
Mean
diff=4.4
P=0.02
Mean
diff= 6.7
P<.0.001
SUBSTANCE OUTCOMES
22
Past 28 Day Substance Use
OROS > PLACEBO “ADDED BENEFIT
“
RESPONDERS > NON-RESPONDERS
Clinically and statistically significant reduction in
past 28 day drug use in both groups
REGARDLESS
OF MEDICATION
The trajectories of past 28 day drug use based on
adolescent self-reports
did not
GROUP
(CGI-I)
).
•More
UDSgroups (Chi-square = 3.04,
differ Negative
between treatment
3 df, pASSIGNMENT
= 0.3855 ; Proc Glimmix
18
•More Negative UDS
16
(p< 0.05)
14
12
Placebo= - 5.2 days; 34% p<0.001
Days of Use
• Trend for more subjects
NS
with > 75% reduction in
8
•More Days of Abstinence
days of drug use
OROS MPH= - 5.7 days; 41%; p<0.001
6
10
Placebo
4
• Mean difference = 15
(p<0.001)
OROS
(p=0.08)
2
0
0
4
Week of8Study
12
16
Significant Differences in Adverse Events By Treatment Group
Adverse Effect
OROS-MPH
Placebo
P-value
Statistical
• OROS-MPH
Significance
+ CBT > study-related AEs/subject
Excoriationcompared to placebo
14/151 (9.3%)
> (2.4
4/152 (2.6%)
0.016
+ CBT
v 1.6; p=0.022)
Abdominal Discomfort
• 70% reported
Nervousness
8/151 (5.3%)
>
2/152 (1.3%)
0.016
Palpitations (males only)
5/122 (4.1%)
>
0/117 (0.9%)
0.024
taking
medication
same
12/151
(7.9%) > on
3/152
(2.0%) day as
0.018
but
Heart Rate drugs/alcohol
Increased
8/151only
(5.3%)2.5%
> reported
1/152 (0.7%) adverse
0.019
Heart rate =interactions
> 100
19/149 (12.8%) >
8/148 (5.4%)
0.028
SAEs
• OROS-MPH
= 14/151
study
related
SAE/4
total
Migraine
Headaches
(2.6%
)
> 0/152
(0.0%)
• Placebo = 3 study6/151
related
Anorexia
(4.0%) SAEs/7
>
1/152total
(0.7%)
Statistical Trend
0.06
0.067
Mood Altered
4/151 (2.6%)
>
0/152 (0.0%)
0.06
Road Traffic Accidents
4/151 (2.6%)
>
0/152 (0.0%)
0.06
Limb Injury
1/151 (0.7%)
<
7/152 (4.6%)
0.067
Systolic BP = > 140
27/149 (18.1%)
> 16/148 (14.5%)
0.073
Summary of Main Study Findings
1. OROS-MPH safe, well-tolerated despite nonabstinence
2. ADHD outcomes as good or better than
in adolescents without SUD
3. Results suggest contribution of CBT to both SUD
and ADHD outcomes
4. Substance outcomes as good or better than in youth
with less severe psychopathology
5. Treatment compliance, completion superior to that
reported in studies of youths with less severe
SUD and psychopathology
25
Study Limitations
• Study was not designed to evaluate separate and
combined effects of OROS/CBT on treatment response
•
If CBT contributed to ADHD response, statistical power
compromised
• Heavy reliance on adolescent self-reports as primary
outcome measures
• No long-term follow-up to evaluate possible emergent
treatment effects
26
Interpretation of Results in the
Context of Previous Research
27
Results are inconsistent with Results are consistent with
• With most controlled
• With 3 controlled
psychostimulant vs
placebo (alone) trials
for ADHD (large effect
psychostimulant trials in
adults concurrently
receiving weekly
individual CBT for SUD
size)
• Questions
– medication
compliance?
– validity of self
report?
– contribution CBT?
(Levin et al 2006; 2007; Schubiner et al
2004)
•
Whereas, in the absence of
behavioral tx platform, active
medication separates from
placebo
•
Pemoline, adolescents (Riggs et
al 2004)
•
Atomoxetine, adults (Wilens et al
2008)
Interpretation of Results in the Context of Previous
Research
Cognitive Behavioral and Behavioral Tx effective for
ADHD in individuals without SUD
Adults
•
•
• CBT is most empirically-supported tx modality
Safren
et al, 2005: 56% treatment
responders
for child/adolescent
depression,
anxietymedication
disorders + CBT vs
13%(PTSD,
treatment
responders
medication
alone
OCD);
disruptive
behavior
disorders
Solanto
et al, 2010: Meta-cognitive
effective for adults
• Meta-analysis
of controlled therapy
trials adolescent
with ADHD Amer J Psychiatry, March, 2010
substance treatment interventions between 19602008 showed greater
effect size for individual
Children/Adolescents
(CBT) > family-based interventions (Tripoldi et al
•
Arch Ped
Med
2010)
Fabiano
et alAdol
2009:
Meta
analysis of 114 studies shows “strong
and consistent evidence that behavioral treatments are effective
for treating ADHD” with effect sizes ranging from .83 for between
group, .70 for pre-post, and 2.64 for within group studies”
If Replicated,
Results Have Important Clinical Implications
•
Results suggest that ADHD clinical
response may be important to substance
outcomes
•
OROS-MPH may have some added
benefit over placebo
•
In the context of individual CBT (for
SUD), significant reduction in ADHD
symptoms may occur with or without
pharmacotherapy.
30
If Replicated,
Results Have Important Clinical Implications
• If ADHD does not improve early in
treatment, OROS-MPH may be
considered
• Well-tolerated, good safety profile and
likely effective for ADHD even in nonabstinent adolescents
• Low abuse/diversion liability with
weekly monitoring
31
Implications for Study Design
Efficacy
Assumed fixed site
effects by
standardizing TAU
•
• standardize behavioral platform in efficacy
trials
CTN
study
demonstrated
the
• initial multi-site
efficacy
trials may
be
warranted if sufficient
enrollment
feasible at
feasibility
of not
recruitment
and vs “wimpy”
• robust
single site
behavioral
enrollment, high compliance
of platform
• randomize within site
difficult to recruit adolescents
• ? 3 arm trial incl
with co-occurring disorders
CBT alone
Effectiveness
• ? Manage variability of TAU in effectiveness
trials
• post-hoc control for
• standardize length
Feasibility
of medication
of trial
addressing
even if “placebo”
futureresponse
behavioral platform/TAU
is not
research gaps
• ? differential
• randomize within site
symptom response CBT
• analyses include site and treatment by site
(alone) vs CBT +
• consider population and setting results
medication
generalize to
32
Thank you for your attention
Comments?
Questions?
Discussion?
33