A Research Update on the Neurobiology and Treatment of Eating Disorders Katherine A' Halmi, M'D' Wei

1
A Research Update on the Neurobiology and
Treatment of Eating DisordersKatherine A.
Halmi, M.D.Weill Cornell Medical College

• World Psychiatric Association
• Athens, Greece
• March 12, 2005

2
Multi-Conditional Model of Pathogenesis of
Anorexia Nervosa and Bulimia Nervosa
Biologic Vulnerability
Starvation Effects
Genetic, physiologic predisposition
Weight Loss Mental Changes
Psychologic Predisposition
Early experience Family influences
Intrapsychic conflict
Bulimia
Personality Change
Dieting Malnutrition
Physiologic Changes
Social Climate
Societal influence and expectations
(modified from Pike and Ploog, 1984, adopted
from Lucas, 1991)
Sustaining
3
Genetic Studies in Eating Disorders

• Family and twin studies suggest eating disorders
  are highly familial with a substantial portion of
  the observed familiality due to additive genetic
  factors

4
Identifying Susceptibility Genes for Complex
Traits

• Candidate gene approach
• Requires an hypothesis for a specific gene
• The hypothesis is tested by determining if
  certain forms of the gene are increased in an
  affected group compared to a control group
• Genome wide linkage analysis
• Requires no hypothesis about the nature of genes
  that are involved
• Novel, unknown genes can be identified

5
Genetic Studies in Eating Disorders

• Case-control association designs have yielded
  sporadically significant and typically
  non-replicated findings. The three main targets
  of inquiry have been indolamines, catecholamines
  and neuropeptides.

6
Association Studies in Eating Disorders
7
Association Studies in Eating Disorders
8
Association Studies in Eating Disorders
9
Association Studies in Eating Disorders
10
Association Studies in Eating DisordersNeuropepti
des
11
Association Studies in Eating Disorders
12
Genetic Studies in Eating Disorders

• Linkage studies give evidence for the presence of
  susceptibility loci on chromosome 1 for anorexia
  nervosa and chromosome 10 for bulimia nervosa.

13
Behavioral Covariates

• Drive for Thinness
• Y-BOCS Obsessionality
• A group of ASP who are extreme and highly
  concordant assumption is this group is typical
  for the linked (disease-marker) group.

14
Incorporation of Behavioral Covariates into
Linkage Analysis

• Drive for Thinness and Obesssionality
• Areas of Significant Disequalibrium on
  Chromosomes 1, 2 and 13

15
Future of Genetic Research in Eating Disorders

• Develop methodologies to study how variations in
  genotype can influence individuals to be
  differentially sensitive to environment events.
• The majority of women diet at some point in their
  life, only a small fraction develop frank eating
  disorders.

16
Salient Features of Neuropeptidesin Eating
Disorders

• Signals from the periphery, including gut-related
  peptides and adipokines, interact with
  hypothalamic peptides in the regulation of
  feeding behavior and body weight.
• State-related changes in nutrition and body
  weight influence cerebrospinal fluid and plasma
  levels and receptor activity of neuropeptides.

17
Hypothalamic Neuropeptides
18
Hypothalamic Neuropeptides

• Melanocortins interact with leptin and
  serotonin in the suppression of
• eating behavior.
• Severity of binge eating in obese patients is
  associated with mutations in the melanocortin 4
  receptor gene.
• Some patients with AN BN have circulating
  autoantibodies that bind to alpha-melanocyte-stimu
  lating hormone or ACTH.
• Aguoti related protein, an antagonist of
  central melanocortin receptors, has gene
  polymorphisms in AN.
• Farooqi et al, 2003
• Branson et al, 2003
• Fetissov et al, 2002
• Lu et al, 1994
• Vink et al, 2001

19
Adipokines
20
Gut-Related Peptides
21
Future Neuropeptide Research

• Development of centrally active agonists and
  antagonists for treatment interventions.

22
Functional Brain Imaging Studiesin Anorexia
Nervosa

• AN Resting Studies (9)
• 1 PET FDG, 6 SPECT, 1 SPECT 5HT2a, 1 PET 5-HT2A
• 8 ill state, 2 Recovered
• Decreased frontal cortex, occipital cortex,
  cingulate cortex, temporal/amygdala
• AN Activation Studies (9)
• 4 SPECT, 1 PET rCBF, 4 FMRI
• 9 ill, 1 Recovered
• 2 Eating food, 5 Food images, 2 Body images
• Frontal cortex normal, Temporal/Amygdala
• normal to increased, Cingulate cortex-
• normal to increased, Occipital - normal

23
Functional Brain Imaging Studiesin Bulimia
Nervosa

• BN Resting Studies
• 3 PET FDG, 1 SPECT, 1 PET, 1SPECT 5HTT, 1PET
  5HT2A, 1 PET 5HT1A
• 6 ill, 2 Recovered
• Frontal cortex normal to decreased,
  temporal/amygdala
• normal to increased, occipital cortex normal
• BN Activation Studies
• 1 FMRI, 1 Recovered, sweet taste
• R. Cingulate cortex decreased, L. Occipital
  cortex - decreased

24
Reviews of New Treatment Research in Eating
Disorders With Critical Analyses

• Cochrane Reviews The Cochrane Library, Issue 3,
  2004.
• Australian and New Zealand Clinical Practice
  Guidelines for Treatment of AN, Australian and
  New Zealand J. Psychiatry 2004 38659-670.
• NICE (National Institute for Clinical Excellence)
  Clinical Guideline 9, January 2004, Eating
  Disorders, UK www.nice.org.uk/CG009publicinfoengli
  sh

25
Few Controlled Treatment Trials forAnorexia
Nervosa

• Disorder relatively uncommon adequate sample
  size
• Resistance to treatment
• Medical complications withdrawal from protocols

26
Anorexia Nervosa Resistance toTreatment
Compliance

• Positive function in patients life
• Escape from aversive developmental issues.
• Escape from distressing life events
• Highly reinforcing terrifying to relinquish
• Egosyntonic nature patient denial

27
Psychosocial Intervention - RCTs
28
Family Therapy in AN
29
Pharmacotherapy for ANNew Randomized Controlled
Trials
30
Pharmacotherapy ANConclusion Recent Studies

• Promising pilot studies of antipsychotic
  medications during weight restoration phase.
  Example, olanzapine, quetiapine.
• Citalopram may reduce depression and anxiety
  during weight restoration.
• Fluoxetine is not beneficial in weight
  restoration but may decrease relapse rate in
  weight restored AN patients.
• Nutritional supplements with L-tryptophan do not
  increase effectiveness of fluoxetine.
  (Double-blind placebo controlled trial.)

31
Pharmacotherapy of Anorexia Nervosa

• Medications are only useful adjuncts in treating
  AN
• 1. Cyproheptadine (Periactin) facilitates weight
  gain in AN-restrictors and has an antidepressant
  effect.
• 2. Major tranquilizers - chlorpromazine,
  olanzepine may reduce severely obsessional,
  compulsive and agitated behavior, side effect -
  weight gain.
• 3. Fluoxetine (and other SSRIs) may reduce
  relapse of weight and eating disorder behaviors.

32
Signal Detection Methodology Is Useful To
Determine Optimal Algorithms For Prediction

• It is nonparametric and distribution free.
• It leads to an and/or rule that identifies
  which patients require attention and which do
  not.
• It requires an evaluation of the relative
  importance of false positive and false negative.
• It is highly sensitive to interactive effects of
  predictors.
• It can identify different subgroups of subjects
  who have similar probabilities of outcome but for
  different reasons.
• It can take the costs of evaluations into
  consideration.
• It is a hypothesis generating, not testing
  method.
• Kramer HC Assessment of 2x2 association
  generalization of signal detection methods.
• Am. Statistician 1988 4237-49.

33
Predictors of Treatment Acceptance Signal
Detection Analysis

• Medication CBT and CBT Fluoxetine
• 23 of 41 Patients (56) 66 of 81 Patients (81.5)
• YBC-EDS YBC-EDS
• Low lt8 Preoccupations High gt8 Preoccupations
• 13 of 25 Patients (52) 51 of 56 Patients (91)

34
Predictors of Treatment Completion

• 122 participants were randomly assigned
• 45 of the 89 treatment acceptors (50.6)
  completed treatment
• 45 of 122 (37) completed treatment
• Low Self Esteem High Self Esteem
• (RSE gt23) (RSE lt23)
• n 68 Acceptors n 21
• n 27 Completers n 18
• 39.7 86

35
Conclusions

• Anorexia Nervosa patients will not accept
  medications as the only treatment. (Drug refusal
  rate of 59 in Kaye et al, 2001).
• Behavioral characteristics (obsessive
  preoccupations and self esteem) predict
  differential dropout rates.
• Problems of treatment acceptance and dropout need
  to be solved before large comparison treatment
  trials in AN are conducted.
• Prevention of chronicity in AN focus on early
  diagnosis and adequate treatment of the younger
  anorectics.

36
Medical Management of Anorexia NervosaRandomized
Controlled Studies

• Hospitalization use of liquid formula in early
  stage of weight gain with gradual exposure to
  food and increased activity is most effective for
  weight gain (Okomoto et al, 2002).
• Hospitalization warming therapy with a heating
  vest for 3 hours/day over 21 days did not
  increase rate of weight gain (Birmingham et al,
  2004).
• Recombinant human growth hormone (rhGH) 0.05
  mg/kg subocutaneously for 28 days did not
  increase rate of weight gain (hill et al, 2000).

37
Future Directions for AN Treatment Research

• Continue to develop therapies that address
  motivational stage (Touyz et al, 2003 Geller,
  2002).
• Devise interventions to effect the anorectics
  interpersonal attachment style (Ward et al,
  2000).
• Develop interventions to effect anorectics
  overall sense of attachment to life (Bachar et
  al, 2002).
• Deal with Resistance to Treatment.

38
Practice Recommendations

• CBT is the treatment of choice for BN
• Combining CBT with an SSRI is indicated in cases
  of severe comorbid depression
• IPT is a viable alternative, particularly when
  specialists are available in IPT and not CBT
• Individuals who fail to respond to CBT by session
  10 should be considered for alternative treatment
  (eg, IPT or SSRI)

SSRIselective serotonin reuptake
inhibitor. Wilson GT, Fairburn CG. Eating
disorders. In Nathan PE, Gorman JM, eds.
Treatments That Work. 2nd ed. New York Oxford
University Press 2002559-592.
39
Predictors of Outcome in CBT

• Early response (70 decrease in frequency of
  purging at 4 weeks 6 sessions) in CBT is a
  robust predictor of outcome
• Dropouts from CBT have higher bulimic cognitions,
  greater concern about shape, and greater
  impulsivity
• Current depression, low BMI, and poor social
  adjustment predict poor CBT outcome in
  individuals with BN
• Higher levels of preoccupation and ritualization
  of eating, and low motivation for change were
  predictive of relapse following abstinence at
  posttreatment

Agras WS et al. Am J Psychiatry.
20001571302-1308. Halmi KA et al. Arch Gen
Psychiatry. 2002591105-1109.
40
Medication Management of Bulimia Patients

• Fluoxetine, 60mg/d is the drug of first
  choiceevidence of beneficial effect, favorable
  side-effect profile relatively
• If first trial is unsuccessful, there is evidence
  another antidepressant trial may be effective
• Minimum duration of successful treatment should
  be 6 months
• Baseline labCBC, serum electrolytes,
• liver function, BUN/Cr, thyroid function,
  EEG, r,

41
Topiramate Pharmacotherapy for Bulimia Nervosa
69 BN patients randomized 10 week
trial Median dose 100mg/d, range 25-400
42
Topiramate Pharmacotherapy for Bulimia Nervosa

• Discontinuation and Side Effect Profile
  
• Randomized n69
• Topiramate 13 discontinued
• adverse event1, lost to
  follow-up9, other3
• Placebo 16 discontinued
• Side Effects
• fatigue 32, influenza-like symptoms 29,
• paresthesia 24, hypoesthesia 21,
• nausea 18, constipation 15,
• difficulty concentrating /attention 15

43
Summary of BED Treatment Research

• CBT and IPT produce substantial and long-lasting
  changes in the specific and general
  psychopathology of BED
• Cessation of binge eating is associated with both
  weight loss and maintenance of those losses over
  a 1-year period
• Questions about the optimal treatment regimen for
  BED need further evaluation

Wilfley DE. Psychological treatment of binge
eating disorder. In Fairburn CG, Brownell D,
eds. Obesity and Eating Disorders. 2nd ed. New
York Guilford Press 2002350-353. Wilson GT,
Fairburn CG. Eating disorders. In Nathan PE,
Gorman JM, eds. Treatments That Work. 2nd ed. New
York Oxford University Press 2002559-592.
44
Pharmacotherapy of BEDSummary of Results
1. McCann UD, Agras WS. Am J Psychiatry.
19901471509-1513. 2. Alger SA et al. Am J Clin
Nutr. 199153865-871. 3. Hudson JI et al. Am J
Psychiatry. 19981551756-1762. 4. Arnold LM et
al. J Clin Psychiatry. 2002631028-1033. 5.
McElroy SL et al. Am J Psychiatry.
20001571004-1006. 6. McElroy SL et al. 2003,
submitted.
45
Topiramate in BED With ObesityMean Binges/Week
MeanBinges/Wk
? 46
? 94
Weeks
Adapted with permission from McElroy SL et al. Am
J Psychiatry. 2003160255-261.
46
Topiramate in BED With Obesity Mean Weight
Change (ITT)
Weeks
MeanWeightChange(kg)
ITTintent-to-treat P.005 Adapted with
permission from McElroy SL et al. Am J
Psychiatry. 2003160255-261.