MATRICS: an example of government industry academic collaboration

1
MATRICS an example of government / industry /
academic collaboration

• Michael F. Green, Ph.D.,
• Department of Psychiatry and Biobehavioral
  Sciences,
• Geffen School of Medicine at UCLA
• Department of Veterans Affairs, VISN 22 Mental
  Illness, Research, Education and Clinical Center
  (MIRECC)
• ASENT Washington D.C.
• March 2007

2
Cognition in Schizophrenia Core Feature of the
Illness

• Present before onset of clinical symptoms
• Seen in unaffected first-degree relatives
• Relatively stable across clinical state life
  span
• Low cross sectional correlations w/ psychotic
  symptoms
• Discrepancy between clinical and cognitive
  effects of antipsychotic meds
• Schizophrenia profile of deficits (w/ variation)

3
Separable Cognitive Domains in Schizophrenia

• Speed of Processing
• Attention/Vigilance
• Working Memory
• Verbal Learning

• Visual Learning
• Reasoning and Problem Solving
• Social Cognition

Nuechterlein et al. (2004), Schizophrenia
Research.
4
Magnitude of Cognitive Deficits in Schizophrenia
Meta-Analysis 204 studies, 7420 patients and
5865 controls
Characteristic profile in schizophrenia maximal
impairment in memory, attention, and executive
function relative preservation of old learning
and visual perceptual skills.
SD Units
Heinrichs Zakzanis Neuropsychology 1998
Healthy Comparison Mean
5
Attentional Deficits and Symptoms
Neuroleptic Naive
Neuroleptic Withdrawn
Mean of global scores for SANS and SAPS
Finkelstein et al. (1997), J Abnormal Psychology
6
Profiles of Neurocognitive Deficits Are Very
Similar for Schizoaffective and Schizophrenic
Disorders
Schizophrenia (n115)
Schizoaffective (n24)
Mood Disorder (n22)
50
45
Mean T-Score
40
35
30
ATTN
ABST
PMOT
LEARN
MOTOR
MEM
RETENT
VERBAL
From Evans et al. (1999)
7
Cognition and Functional Outcome in
Schizophrenia

• Cognitive deficits are reliable correlates and
  predictors of functional outcome (disability)
• Functional outcome includes work outcome, social
  outcome, independent living, skills acquisition
• Magnitude of associations medium for specific
  domains large for summary scores
• Relationships are stronger than between psychotic
  symptoms functional outcome
• Cognitive deficits are linked to success in
  psychiatric rehabilitation

8
Cognition and Functional Outcome in
Schizophrenia Strengths of Relationships
Large -
Medium -
Small -
Problem Solving
Immediate Memory
Verbal Learning
Sustained Attention
Summary Scores
Effect size based on Cohens r
Green et al. Schizophrenia Bulletin 2000
9

• Identify determinants
• Build models
• Test models

10
Model Building Neurocognition, Interventions,
Mediators, and Functional Outcome
Green, Meltzer, Newcomer Bellack, APA 2000
11
Model Testing Neurocognition, Social Cognition,
Functional Outcome (Path Analysis)
.56
.30
Adjusted Goodness of Fit Index .99
Brekke, Kay, Kee Green (2005)
plt .05 p lt .01, one-tailed
12
MATRICS Measurement and Treatment Research to
Improve Cognition in Schizophrenia

• Steve Marder, M.D., P.I.
• Michael Green, Ph.D., Co-P.I.
• NIMH Project Officer Wayne Fenton, M.D.
• NIMH Division Director, DMDBA Ellen Stover,
  Ph.D.
• www.matrics.ucla.edu

13
Wayne Fenton 1953-2006 Science That Matters
14
MATRICS Background and Rationale

• Increasing evidence that cognitive deficits are
  core features of schizophrenia
• Increasing support for relationships between
  cognition and functional outcome in schizophrenia
  
• Increasing research focus on the basic studies
  of neuropharmacology of cognition

15
Targeting Cognition in Schizophrenia Why the
Bottleneck?

• Lack of consensus regarding cognitive targets
• No widely accepted endpoint
• Ambiguity regarding optimal clinical trial design
• Lack of path to FDA approval and labeling
• Barriers to compound acquisition for testing

16
NIMH MATRICS Goals and Products

• Create Standardized Measure for use in Clinical
  Trials
• Define Optimal Experimental Designs
• Establish path to FDA Approval
• Attract large pharmaceutical companies to focus
  efforts on this important clinical target
• Success requires involvement of NIMH, FDA,
  pharmaceutical industry, and academia

www.matrics.ucla.edu
17
The First Challenge of MATRICS Pseudospecificity

• A claim of a drug effect that is considered to
  be artificially narrow
• Serves only promotional purpose
• Implied advantage over other drugs in class
  regarding subgroup/symptom
• Misleading (since no evidence to support)
• FDA Position is that a claim is pseudospecific
  until proven otherwise.

18
Examples of Pseudospecificity

• Subgroup of recognized syndrome (e.g., major
  depression in women)
• Symptom of recognized syndrome (e.g.,
  hallucinations in schizophrenia)
• Claiming specific benefit in single disease model
  for recognized nonspecific symptom (e.g., dental
  pain)

• Cognition in schizophrenia would be
  pseudospecific if
• Cognitive impairment resulted from other illness
  features (e.g., psychosis)
• Cognitive impairment was general and not specific
  to schizophrenia

19
Principles for Developing Consensus

• Consensus should be as broad as possible
• Inclusion of academia, NIMH, industry, FDA,
  consumer representatives
• Transparency of process
• A priori development of a path to consensus
  (e.g., RAND Panel)
• Management of conflicts of interest

20
Steps to MATRICS-NIMH Consensus Battery
Subgroup of NCC survey of experts
Survey of experts
NCC, based on survey of experts
NCC
RAND Panelists
MATRICS Team
NCC, based on ratings of Panelists
NCC MATRICS Neurocognition Committee PASS
MATRICS Psychometric and Standardization Study
21
Essential Criteria for Consensus Cognitive
Battery for Clinical Trials in Schizophrenia

• Battery
• Inclusion of the seven cognitive domains
• Valid assessment of cognition at the level of all
  individual major cognitive domains
• Individual Tests
• High test-retest reliability
• High utility as a repeated measure
• Demonstrated relationship to functional outcome
• Demonstrated tolerability and practicality

22
The MATRICS Psychometric and Standardization
Study (PASS)

• 5 Sites UCLA / LA VA, Duke, MPRC, Harvard, U of
  Kansas
• PHASE 1
• To collect psychometric data for beta version of
  battery -- provide basis for selection of final
  consensus battery
• Test-retest reliability, repeatability,
  relationship to outcome, practicality /
  tolerability
• 176 schizophrenia patients at baseline 167 at 4
  weeks
• PHASE 2
• To provide co-norms for the final MATRICS battery
  so that results can be standardized
• 300 adults from general community representative
  demographics

23
MATRICS Consensus Cognitive Battery (MCCB)
Estimated administration time 63.5 min

• Speed of Processing
• Category Fluency
• BACS Symbol Coding
• Trial Making A
• Attention / Vigilance
• Continuous Performance Test - Identical Pairs
  version
• Working Memory
• Maryland Letter Number Span
• WMS-III Spatial Span

• Verbal Learning
• Hopkins Verbal Learning Test-R
• Visual Learning
• Brief Visuospatial Memory Test-R
• Reasoning and Problem Solving
• NAB Mazes
• Social Cognition
• MSCEIT Managing Emotions

24
MATRICS Consensus Cognitive Battery (MCCB)

• Distributed by
• Harcourt Assessment, Inc
• Multi-Health Systems (MHS)
• Psychological Assessment Resources (PAR)

25
Co-primary Measure in Addition to Cognitive
Performance

• Arguments for Community Status as Co-primary
• Increase face validity for consumers and
  clinicians
• Increase acceptance by consumers and clinicians
• The ultimate goal is better community functioning
• Arguments against Community Status Co-primary
• Measurement challenges
• Distance from biological systems (sensitivity to
  cognitive-enhancing drugs likely length of time
  to change)
• Intervening variables / level of psychosocial
  support
• Arguments by analogy (e.g., not needed for
  hallucinations, depression, Parkinsons tremor)

26
Cognition and Co-Primary Measures The
compromise

• Agreement that community outcome may set the bar
  too high
• Too dependent on factors that lie outside a
  clinical trial
• Search for other outcome measures that could be
  functionally meaningful
• Either related to community functioning, or
  something that a patient could notice, or
  something that a caregiver could notice

27
Approaches to Co-primary Measures

• Functional Capacity (Proxy) Measures
• Measures that simulate daily activities or social
  problem solving situations
• Demonstrate a person is capable of performing the
  functional task not whether they do so in the
  community
• Interview-based Assessments of Cognition
• Interview estimate of level of cognitive
  abilities, or how much cognitive deficits affect
  community functioning
• Can be self-ratings from the patients or
  estimates of a patients cognitive abilities from
  a caregiver

28
Key Linkages for Cognition and Functional
Outcome
Community Outcome
Community
Assessment Environment
Clinic
Cognitive Performance Measures
Laboratory
Distal
Proximal
Within individual
Interpersonal
Sensitivity to cognitive drug effects
Less sensitive
More sensitive
29
Next Steps for Assessment in Trials of Cognition
Enhancement in Schizophrenia

• 1) Translation of MCCB into other languages
• For use in multi-national trials
• Linguistic translation and co-norming
• 2) NIMH Initiative for cognitive neuroscience
  measures in clinical trials (CNTRICS)
• Enhanced construct validity / precision
• Psychometrics
• Practicality / Tolerability
• 3) Formation of Industry / Academic Consortium
• Uses NIH Foundation
• Will reach consensus co-primary measure
• Will fund translation of MCCB

30
NIMH-TURNS Treatment Units for Research on
Neurocognition and Schizophrenia

• Network of 7 research sites.
• Identify, obtain and test efficacy of new
  agents targeting cognition in schizophrenia.
• Develop and validate clinical trial
  methodology.
• Conduct two three NIMH supported trials.
• To eventually obtain federal, foundation or
  industry support for additional trial.

31
TURNS Research Sites

• Sites
• Columbia University
• Duke University
• Harvard / Mass General
• Nathan Kline Institute
• University of Maryland
• UCLA
• Washington University

32
TURNS Call for Nominations
33
6 Things I Learned in MATRICS

• Sometimes the process IS the product
• Sometimes it is good to work for a boss
• Sometimes it is important to have idealists at
  the table
• Sometimes it is important to ask idealists to
  leave the table
• Sometimes you need to start a business knowing
  you will never make a cent
• Sometimes you need to make it up as you go