Child Onset Child Onset Depression: Depression: Is It a Different Is It a Different Disorder? Disorder? Neal Ryan Neal Ryan
Dec 20, 2015
Child Onset Depression:Child Onset Depression:Is It a Different Disorder?Is It a Different Disorder?
Neal RyanNeal Ryan
Conflict of Interest StatementConflict of Interest Statement
No industry-funds in prior two yearsNo industry-funds in prior two years PI on the Pittsburgh site of the Keller et. al study of PI on the Pittsburgh site of the Keller et. al study of
paroxetine funded by GSKparoxetine funded by GSK PI on Pittsburgh site of Wyeth study of child depressionPI on Pittsburgh site of Wyeth study of child depression Paid consultant (lifetime) to:Paid consultant (lifetime) to:
AbbottAbbott BMSBMS GSKGSK Johnson and JohnsonJohnson and Johnson PfizerPfizer WyethWyeth
Joaquim Puig-Antich, 1944-1989Joaquim Puig-Antich, 1944-1989
Joaquim Puig-AntichJoaquim Puig-Antich
Born: September 22, 1944, Barcelona SpainBorn: September 22, 1944, Barcelona Spain Died: December 2, 1989Died: December 2, 1989 Undergraduate: 1953-1960, La Salle Bonanova, Undergraduate: 1953-1960, La Salle Bonanova,
BarcelonaBarcelona Graduate: 1961-1967, Facultad de Medicine, Graduate: 1961-1967, Facultad de Medicine,
Universidad de BarcelonaUniversidad de Barcelona Post GraduatePost Graduate
1967: Clinique Rech., Montpellier, France, Resident in 1967: Clinique Rech., Montpellier, France, Resident in NeurosurgeryNeurosurgery
1970-1971: Sinai Hospital, Baltimore, Intern in Medicine1970-1971: Sinai Hospital, Baltimore, Intern in Medicine 1971-1973, Beth Israel, New York, Resident in Psychiatry1971-1973, Beth Israel, New York, Resident in Psychiatry 1973-1975, Albert Einstein, New York, Child Psychiatry Fellow1973-1975, Albert Einstein, New York, Child Psychiatry Fellow
Joaquim Puig-AntichJoaquim Puig-Antich
Appointments:Appointments: 1975-1977: Albert Einstein, Assistant Professor of 1975-1977: Albert Einstein, Assistant Professor of
PsychiatryPsychiatry 1977-1983, Columbia, Assistant Professor of Clinical 1977-1983, Columbia, Assistant Professor of Clinical
PsychiatryPsychiatry 1983-1984, Columbia, Associate Professor of Clinical 1983-1984, Columbia, Associate Professor of Clinical
PsychiatryPsychiatry 1984-1989, Pittsburgh, Professor of Psychiatry1984-1989, Pittsburgh, Professor of Psychiatry 1988-1989, Pittsburgh, Professor of Pediatrics1988-1989, Pittsburgh, Professor of Pediatrics
Additional AcknowledgementsAdditional Acknowledgements
Ron DahlRon Dahl David AxelsonDavid Axelson Boris BirmaherBoris Birmaher David BrentDavid Brent BJ CaseyBJ Casey Cam CarterCam Carter Mike De BellisMike De Bellis Erika ForbesErika Forbes
Ahmad HaririAhmad Hariri Joan KaufmanJoan Kaufman Chris KyeChris Kye Cecile Ladouceur Cecile Ladouceur Chris MayChris May Jim PerelJim Perel Scott WatermanScott Waterman Doug WilliamsonDoug Williamson
Department of PsychiatryDepartment of PsychiatryUniversity of Pittsburgh Medical CenterUniversity of Pittsburgh Medical CenterWestern Psychiatric Institute and ClinicWestern Psychiatric Institute and Clinic
Child DepressionChild Depression Duration and CourseDuration and Course
Duration 3-9 monthsDuration 3-9 months 6-10% last more than 2 years6-10% last more than 2 years 70% recurrence in 5 years70% recurrence in 5 years 20-40% become bipolar20-40% become bipolar
• But 25% total adolescent prevalence of unipolar in But 25% total adolescent prevalence of unipolar in epidemiologic studies versus epidemiologic studies versus 1-2% for bipolar disorders so numbers 1-2% for bipolar disorders so numbers don’t “add up”don’t “add up”
Clinical PictureClinical Picture
Clinical picture in child, adolescent and adult Clinical picture in child, adolescent and adult depression very similardepression very similar
Endogenicity/melancholic, suicide attempts, Endogenicity/melancholic, suicide attempts, lethality of suicide attempts, lethality of suicide attempts, and impairment of functioning increase with and impairment of functioning increase with ageage
Separation anxiety, phobias, somatic Separation anxiety, phobias, somatic complains and comorbid behavioral problems complains and comorbid behavioral problems decrease with agedecrease with age
Psychosocial OutcomesPsychosocial Outcomes
During depression and after recovery:During depression and after recovery: Worse functioning with friends and familyWorse functioning with friends and family Impaired performance in schoolImpaired performance in school Higher rate of pregnancyHigher rate of pregnancy More smoking – “gateway”More smoking – “gateway” Clear long-term persistence after “successful” Clear long-term persistence after “successful”
treatment of depressiontreatment of depression
Increase in rate of depression, Increase in rate of depression, particularly great in girls, correlated particularly great in girls, correlated
with puberty and not age with puberty and not age per seper se
Point PrevalencePoint Prevalence 0.4 to 2.5% in children0.4 to 2.5% in children
1:1 sex ratio1:1 sex ratio 0.4 to 8.3% in 0.4 to 8.3% in
adolescentsadolescents2:1 female excess2:1 female excess
Lifetime prevalence in Lifetime prevalence in adolescenceadolescence 15% - 25%15% - 25% boys girls
What other disorders are like Major What other disorders are like Major Depressive DisorderDepressive Disorder
SimilarSimilar Complex genetic disorders, large environmental Complex genetic disorders, large environmental
contribution, exacerbated by stress, treatment but no contribution, exacerbated by stress, treatment but no “cure”, the group that is responsible for most medical “cure”, the group that is responsible for most medical morbiditymorbidity
• HypertensionHypertension• Obesity, adult onset diabetesObesity, adult onset diabetes• Alcohol AbuseAlcohol Abuse
Not similarNot similar• Communicable diseases (avian flu)Communicable diseases (avian flu)• Single-gene disorders (ALS, sickle cell disease)Single-gene disorders (ALS, sickle cell disease)• Being struck by lightningBeing struck by lightning
A Model for Genesis and Maintenance of Child Depression
Child to Adult DepressionChild to Adult Depression
ContinuitiesContinuities Clinical pictureClinical picture Clinical courseClinical course Responds to CBT Responds to CBT
and IPT in and IPT in adolescentsadolescents
Responds to (at Responds to (at least some) SSRIsleast some) SSRIs
DiscontinuitiesDiscontinuities Probably Probably
unresponsive or unresponsive or minimally responsive minimally responsive to TCAsto TCAs
Some biological Some biological correlates of correlates of depression show depression show maturational effectsmaturational effects
Child Depression vs. Adolescent Child Depression vs. Adolescent Depression vs. Adult DepressionDepression vs. Adult Depression
Adolescent Depression is continuous with Adolescent Depression is continuous with adult depression; child depression shows adult depression; child depression shows less continuity but studies are very limitedless continuity but studies are very limited
Perinatal insults, motor skill deficits, care Perinatal insults, motor skill deficits, care taking instability and family-of-origin taking instability and family-of-origin psychopathology increases hazard for psychopathology increases hazard for child depression but not adult depression child depression but not adult depression (Jaffee 2002)(Jaffee 2002)
Outcome of Child/Adolescent Outcome of Child/Adolescent DepressionDepression
More depression and anxietyMore depression and anxiety probably a direct result of prior depression episodeprobably a direct result of prior depression episode
More nicotine dependence, alcohol abuse, More nicotine dependence, alcohol abuse, suicide attempts, educational underachievement, suicide attempts, educational underachievement, unemployment and unemployment and early parenthoodearly parenthood
possibly as a result of shared risk factors for depression and possibly as a result of shared risk factors for depression and other adverse outcomes (Fergusson other adverse outcomes (Fergusson et al, 2002) though not all data supports this conclusionet al, 2002) though not all data supports this conclusion
Initiation of smokingInitiation of smoking
Age in Years
4 6 8 10 12 14 16 18 20
Cum
ulat
ive
Pro
port
ion
Sm
oke
Fre
e
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
MDD High-RiskLow-Risk
Substance abuse / dependenceSubstance abuse / dependence
Age in Years
0 5 10 15 20 25 30
Cum
ulat
ive
Pro
port
ion
Sub
stan
ce F
ree
0.0
0.2
0.4
0.6
0.8
1.0
1.2
MDDHigh RiskLow-Risk
PubertyPuberty
Puberty is starting earlierPuberty is starting earlier
Puberty and Brain DevelopmentPuberty and Brain Development
Some brain changes Some brain changes precedeprecede pubertal increase pubertal increase in hormones and body changesin hormones and body changes
Some brain changes appear to be the Some brain changes appear to be the consequenceconsequence of some pubertal processes of some pubertal processes
Some adolescent brain maturation appears to be Some adolescent brain maturation appears to be independentindependent of pubertal processes of pubertal processes
Potential for creating internal dis-synchronyPotential for creating internal dis-synchrony
Slide courtesy R. Dahl
Puberty and Motivation/EmotionPuberty and Motivation/Emotion
Strongest links to pubertal changes per-se are in Strongest links to pubertal changes per-se are in the domains of romantic motivation, sexual the domains of romantic motivation, sexual interest, emotional intensity, sleep/arousal interest, emotional intensity, sleep/arousal regulation, appetite, and affective disorders regulation, appetite, and affective disorders
A general increase in risk-taking, novelty-A general increase in risk-taking, novelty-seeking, sensation-seeking (reward-seeking).seeking, sensation-seeking (reward-seeking).
Animal studies also show increase in novelty-Animal studies also show increase in novelty-taking (risk-taking?) in the peri-adolescent taking (risk-taking?) in the peri-adolescent period (Spear 2000)period (Spear 2000)
Slide courtesy R. Dahl
PubertyPuberty
A number of developmental hormonal changes occur A number of developmental hormonal changes occur during the pubertal transition (reproductive hormones, during the pubertal transition (reproductive hormones, adrenal androgens, growth hormones).adrenal androgens, growth hormones).
Hormone levels fluctuate across hours and days.Hormone levels fluctuate across hours and days. Increased stress exposure during adolescence also leads Increased stress exposure during adolescence also leads
to hormonal (cortisol) and brain changes.to hormonal (cortisol) and brain changes. There are complex interactions between reproductive There are complex interactions between reproductive
hormones, stress-related hormones, and neural systems hormones, stress-related hormones, and neural systems that regulate behavioral affect.that regulate behavioral affect.
There are profound individual differences in developmental There are profound individual differences in developmental trajectories in each of these systems.trajectories in each of these systems.
Slide courtesy R. Dahl
Brain DevelopmentBrain Development
Brain Development by Anatomic Region (145 Children & Brain Development by Anatomic Region (145 Children & Adolescents age 4-22 years of age who underwent 243 MRI Adolescents age 4-22 years of age who underwent 243 MRI
Scans) [Giedd et al]Scans) [Giedd et al]
90%
95%
100%
105%
110%
115%
4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22Age (in years)
Total Brain CerebellumFrontal TemporalParietal
**Parietal
Frontal
Temporal
Total Brain Cerebellum
**
*
Peak Cerebellum vs. Other Peaks: * <.002, ** <.0001
Overview of Psychotherapy Overview of Psychotherapy Studies in Child and Studies in Child and
Adolescent DepressionAdolescent Depression
PsychotherapyPsychotherapy
CBTCBT CBT works better than wait-list and better than CBT works better than wait-list and better than
some other psychotherapies in child and some other psychotherapies in child and adolescent major depression (Reinecke 1998, adolescent major depression (Reinecke 1998, Harrington 1998; Brent 1997; Clarke 1999)Harrington 1998; Brent 1997; Clarke 1999)
IPT IPT Works in depressed adolescents Works in depressed adolescents
(Mufson, 1999)(Mufson, 1999)
TADS ResultsTADS Results
SSRI+CBT and SSRI better then placebo SSRI+CBT and SSRI better then placebo and better than CBT alone aggregating and better than CBT alone aggregating across measuresacross measures
CBT seemed to protect against suicidality CBT seemed to protect against suicidality while SSRI may increase itwhile SSRI may increase it
Combination better than SSRI alone but Combination better than SSRI alone but by small marginby small margin
A RCT of CBT to Prevent A RCT of CBT to Prevent Adolescent DepressionAdolescent Depression
13-18 yo adolescents who were at high risk 13-18 yo adolescents who were at high risk for MDD because of family history (parental) for MDD because of family history (parental) for treated MDD current or in past year and for treated MDD current or in past year and who currently had subsyndromal who currently had subsyndromal depressive symptomsdepressive symptoms
Randomized to usual care (N=49) or 15 one-hour Randomized to usual care (N=49) or 15 one-hour sessions of group CBT (N=45)sessions of group CBT (N=45)
26 month f/u26 month f/u 9.3% MDD in CBT group versus 28.8 in usual care by 14 9.3% MDD in CBT group versus 28.8 in usual care by 14
monthsmonths Preventive effect persisted but somewhat diminished at 18 Preventive effect persisted but somewhat diminished at 18
and 24 monthsand 24 months
Long Term Course and Long Term Course and MaintenanceMaintenance
Little evidence for long-term effect of short-Little evidence for long-term effect of short-term treatmentterm treatment CBT better than other therapy acutely but no CBT better than other therapy acutely but no
difference in longitudinal course (Birmaher 2000)difference in longitudinal course (Birmaher 2000) Modest evidence for long-term maintenanceModest evidence for long-term maintenance
Fluoxetine better than placebo in preventing Fluoxetine better than placebo in preventing relapse over 1 year in fluoxetine responders, 34 vs relapse over 1 year in fluoxetine responders, 34 vs 60% relapse (Emslie 2001)60% relapse (Emslie 2001)
However, overall there is little dataHowever, overall there is little data
Pharmacotherapy of Child Pharmacotherapy of Child and Adolescent Depressionand Adolescent Depression
PharmacotherapyPharmacotherapy
In adultsIn adultsSSRIs = SNRIs = TCAsSSRIs = SNRIs = TCAs
In childrenIn childrenSSRIs > TCAsSSRIs > TCAs
Fluoxetine Published, Simeon et al 49
Industry-funded, single-site, N=40, adolescents
No statistical difference between fluoxetine and placebo but had very high placebo response rate
Published, Emslie et al. 50
NIH-funded, single-site N=96, children and adolescents
Fluoxetine superior to placebo on primary endpoints
Published, Emslie et al. 53
Industry-funded, multi-site, N=219, children and adolescents
Primary endpoint, CDRS-R, did not show significant difference (p < .10) but all other endpoints significantly better for fluoxetine
Published, March et al. (TADS group) 41
NIH-funded, multi-site, N= 439 in four cells (placebo, CBT, fluoxetine, fluoxetine+CBT), adolescents
Fluoxetine-only group superior to placebo on one of two primary endpoints. Considered across the four groups, most of the positive effect of treatment appears attributable to medication not psychotherapy.
Sertraline Published, Wagner et al. 52
Pair of identical-methodology independent industry funded multi-site studies combined into one omnibus test, N=365 across both studies, children and adolescents.
Each study separately did not reach significance. When combined (preplanned), the overall aggregate found sertraline significantly better than placebo on primary endpoint
Citalopram Published, Wagner et al. 54
industry funded, multi-site, N=174, children and adolescents
Citalopram significantly better than placebo on primary and secondary endpoints
Unpublished, some data available on FDA web site 55
Industry funded, multi-site, N=244, adolescents
No indication of efficacy
Paroxetine Published, Keller et al 48 with additional data on GSK website 56
Industry funded, multi-site, N= 180 in paroxetine and placebo cells (with another 95 in an imipramine cell not further considered here), adolescents
Both primary endpoint measures did not reach significance but paroxetine significantly better than placebo on four of six secondary endpoint measures
Unpublished, presented at scientific meeting, Emslie 57 and data available on GSK web site 56
Industry funded, multi-site, N=206, children and adolescents
No indication of efficacy
Unpublished, presented at scientific meeting, Milin 58 and data available on GSK web site 56
Industry funded, multi-site, N=286, adolescents
No indication of efficacy
Escitalopram Unpublished, presented at scientific meeting, Wagner 59
Industry funded, multi-site, N=264, children and adolescents
No indication of efficacy
Venlafaxine Unpublished, presented at scientific meeting, Emslie 60
Industry funded, multi-site, N=165, children and adolescents
No indication of efficacy
Unpublished, presented at scientific meeting, Emslie 60
Industry funded, multi-site, N=201, children and adolescents.
No indication of efficacy
Nefazodone Published abstract only, Emslie et al. 61
Industry funded, multi-site, N=195, adolescents
Nefazodone superior on secondary outcome measure
Unpublished, some data available on FDA web site 55
Industry funded, multi-site, N=284, children and adolescents
No indication of efficacy
Mirtazapine Unpublished, data available on FDA web site 55
Industry funded, multi-site, N=126, children and adolescents
No indication of efficacy
Unpublished, some data available on FDA web site 55
Industry funded, multi-site, N=133, children and adolescents
No indication of efficacy
SummarySummary
Signal that SSRIs workSignal that SSRIs work But less than half of studies are “positive”But less than half of studies are “positive” This is like adult industry studiesThis is like adult industry studies
The more sites in the study, the smaller the The more sites in the study, the smaller the effect size found (Brent et al., in press)effect size found (Brent et al., in press)
Rushed studies probably decrease measured Rushed studies probably decrease measured effect sizeeffect size
However, even if this is true, you don’t have any However, even if this is true, you don’t have any way to say how much this decrease is!way to say how much this decrease is!
SummarySummary
Fluoxetine best replicatedFluoxetine best replicated FDA does not feel that data available for FDA does not feel that data available for
other agents sufficient for indication other agents sufficient for indication Data not bad for citalopram and sertralineData not bad for citalopram and sertraline Data quite mixed for paroxetineData quite mixed for paroxetine
Putting Child and Putting Child and Adolescent Depression in Adolescent Depression in
it’s place (compared to it’s place (compared to adult depression)adult depression)
No difference in rates of adult MDD between MDD, anxiety and control prepubs, but 59% of prepub MDD had recurrence of depression. In those, there was elevated rate of MDD in relatives.
ResultsResults More melancholic symptoms in depressed adolescents More melancholic symptoms in depressed adolescents
but otherwise children and adolescents have similar but otherwise children and adolescents have similar symptomatology, duration, and severity of the index symptomatology, duration, and severity of the index episodeepisode
Similar rates of recovery and recurrenceSimilar rates of recovery and recurrence Similar comorbid disordersSimilar comorbid disorders Similar parental history of psychiatric disordersSimilar parental history of psychiatric disorders Index episode of both groups lasted on average 17 Index episode of both groups lasted on average 17
months.months. 85% of children and and adolescent recovered85% of children and and adolescent recovered 40% had at least one recurrence40% had at least one recurrence Guilt and female sex predicted longer episodesGuilt and female sex predicted longer episodes Prior history of MDD and father MDD predicted lower Prior history of MDD and father MDD predicted lower
recovery and increased risk for recurrencerecovery and increased risk for recurrence
• trait-like marker for depression (or depression and anxiety
• stable through development and adulthood
• may be result of early life stressors (e.g. macaque variable foraging paradigm)
ResultsResults Increased activity in amygdala during Increased activity in amygdala during
presentation of fearful faces and a decrease in presentation of fearful faces and a decrease in activation with repeated exposure to fearful activation with repeated exposure to fearful facesfaces
Developmental differences in amygdala Developmental differences in amygdala response to fearful and neutral facesresponse to fearful and neutral faces Adults show increased amygdala activity for fearful Adults show increased amygdala activity for fearful
facesfaces Children show more amygdala activity in response to Children show more amygdala activity in response to
neutral facesneutral faces• Children may find neutral faces to be more ambiguous than Children may find neutral faces to be more ambiguous than
adults do or even more ambiguous than fearful faces.adults do or even more ambiguous than fearful faces.
14 MDD and 17 control children 9-17 years
Of 14 with MDD, 10 also had comorbid anxiety disorder
Reward Related Decision MakingReward Related Decision Making
Anxiety disordersAnxiety disorders Increased response in cingulate and left Increased response in cingulate and left
caudate (reward related areas) during caudate (reward related areas) during anticipation of reward and in caudate after anticipation of reward and in caudate after receiving large-magnitude rewardreceiving large-magnitude reward
MDDMDD Decreased response caudate after receiving a Decreased response caudate after receiving a
large-magnitude reward. large-magnitude reward.
ThanksThanks