Title: Antidepressant drugs in the treatment of Generalised Anxiety Disorder
1Antidepressant drugs in the treatment of
Generalised Anxiety Disorder what is known and
what remains uncertain?
- David S. Baldwin
- Reader in Psychiatry 1 and Honorary Consultant
Psychiatrist 2 - 1. Clinical Neuroscience Division, School of
Medicine, University of Southampton - 2. Mood and Anxiety Disorders Service, Hampshire
Partnership Trust
2(No Transcript)
3Declaration of interests
- I do not have shares in any pharmaceutical
company, nor do family members - I do not accept any personal retainer from any
pharmaceutical company - I have acted as a consultant to a number of
companies with an interest in anxiety disorders
(Asahi, AstraZeneca, Cephalon, Eli Lilly, GSK,
Lundbeck, Organon, Pharmacia, Pierre Fabre,
Pfizer, Roche, Servier, Sumitomo, Wyeth) - I hold or have held research grants (on behalf of
my employer) from a number of companies with an
interest in anxiety and depressive disorders
(Cephalon, Eli Lilly, GSK, Lundbeck, Organon,
Pfizer, Pharmacia, Roche, Wyeth) - I have accepted paid speaking engagements in
industry supported satellite symposia at
international and national meetings - I do not accept hospitality or travel not related
to a speaking engagement - I am co-author of the British Association for
Psychopharmacology evidence-based guidelines on
the treatment of anxiety disorders - I am a medical patron of the National Phobics
Society
4Outline of talk
- recent evidence-based guidelines for GAD
- response rates to initial treatment approaches
- predictors of response to treatment
- duration of treatment after response
- management approaches after initial non-response
- future treatments and guidelines
51. Recent guidelines for treatment of GAD
6www.bap.org.uk
7Summary of BAP guidelines for GAD
- choose an evidence-based treatment
- take account of clinical features, patient
preference, and availability - consider an SSRI for first-line treatment
- treatment periods of up to 12 weeks may be needed
to fully assess efficacy - routine combination with psychological treatments
is not recommended - continue treatment for at least 6 months in
patients who have responded - use a treatment known to be efficacious in
preventing relapse - monitor efficacy and tolerability during
long-term treatment - after non-response to initial treatment,
consider - switch to other evidence-based treatment
- combining drug treatment with cognitive-behaviour
therapy - combining treatment only when there are no
contraindications - referral to specialist services in refractory
patients
Baldwin DS et al. J Psychopharmacol 2005 19
567-596
8Canadian guidelines for drug treatment in GAD
First line Paroxetine, escitalopram, sertraline, venlafaxine
Second line Alprazolam, bromazepam, lorazepam, diazepam Buspirone, imipramine, pregabalin, bupropion
Third line Mirtazapine, citalopram, trazodone, hydroxyzine
Not recommended Beta-blockers (propranolol)
Canadian Psychiatric Association. Can J Psychiatr
2006 51 9S-91S.
92. Response rates to initial treatment
10Response rates with SSRIs in acute treatment
double-blind randomised placebo-controlled
studies
Study Treatment Dose (mg/day) Length (Wks) Active response () Placebo response ()
Pollack et al 2001 Paroxetine 20-50 8 62 56
Rickels et al 2003 Paroxetine 20, 40 8 61.7, 68 45.6
Rynn et al 2001 (children) Sertraline lt 50 9 90 10
Allgulander et al 2004 Sertraline 50-150 12 63 37
Brawman-Mintzer et al 2006 Sertraline 50-200 10 59.2 48.2
Davidson et al 2004 Escitalopram 10-20 8 58 38
Goodman et al 2005 Pooled analysis, 3 studies Escitalopram 10-20 8 52 37
Baldwin et al 2006 Escitalopram Paroxetine 5, 10, 20 20 12 70.9, 78.4, 74.2 66.2 63
p lt 0.05 p lt 0.01 p lt 0.001,
advantage for active treatment over placebo
11Response rates with SNRIs in acute treatment
double-blind randomised placebo-controlled
studies
Study Treatment Dose (mg/day) Length (Wks) Active response () Placebo response ()
Koponen et al 2007 Duloxetine 60, 120 9 63, 65 34
Rynn et al 2007 Duloxetine 60-120 10 52 41
Hartford et al 2007 Duloxetine Venlafaxine 60-120 75-225 10 10 56 60 42
Davidson et al 1999 Venlafaxine Buspirone 75, 150 30 8 62, 49 55 39
Gelenberg et al 2000 Venlafaxine 75-225 28 67 33
Rickels et al 2000 Venlafaxine 75, 150, 225 8 NS, NS. NR
Allgulander et al 2001 Venlafaxine 37.5, 75, 225 24 63, 73, 81 47
Lenox-Smith et al 2003 Venlafaxine 75-225 24 65 46
Rynn et al 2007 Venlafaxine Flexible dose in children 6-17 yrs 8 69 48
p lt 0.05 p lt 0.01 p lt 0.001, advantage
for active treatment over placebo , mean CGI-I
score, estimates from published figure NR not
reported, NS not significantly different from
placebo
12Fixed-dose escitalopram and paroxetine in
GADreduction in mean HAMA score
p lt 0.05, p lt 0.01, p lt 0.001, versus
placebo p lt 0.05 versus paroxetine mean HAMA
scores at baseline placebo 27.1, ESC 5 27.1, ESC
10 26.0, ESC 20 27.7, PAR 27.3
Baldwin et al. Br J Psychiatry 2006 189 262-272
13Pregabalin in acute treatment of GAD Endpoint
(LOCF) CGI-I responder rates
70
60
50
40
CGI-I responder rates ()
30
20
10
0
PBOn85
PBOn83
PGB 300 n89
PGB450n88
PGB600n85
PGB200n75
PGB400n85
PBOn100
PGB400n94
PGB600n104
PGB450n85
ALP 1.5 n88
VEN75n110
Total daily dose
Rickels et al 2005 (4 wks)
Pohl et al 2005 (6 wks)
Montgomery et al 2006 (6 wks)
Responder rate defined as CGI-I of 1 or 2
plt0.05 vs placebo plt0.01 vs placebo
14Efficacy vs. effectiveness of benzodiazepinessyst
ematic review and meta-analysis of RCTs
Withdrawals due to lack of efficacy
Withdrawals for any reason
Martin JLR et al. J Psychopharmacol 2007 21
774-782
15RCTs of antidepressant treatment in elderly GAD
Study Medication Dose (mg/d) N Mean age (yrs) Length (wks) Baseline HAMA score Placebo response Active response
Katz et al 2002 Venlafaxine (pooled) 37.5-225 195 66.1 8 25.9 41.0 66.0
Lenze et al 2005 Citalopram 10-30 34 69.4 8 22.3 24.0 65.0
Davidson et al 2008 Duloxetine 60-120 73 70.6 9-10 20.9 48.0 29.0
Montgomery et al In press Pregabalin 150-600 273 72.4 8 26.0 48.4 58.4
p lt 0.05 p lt 0.01, vs placebo 30/34
patients had GAD HAMA-response, not significant
16Pregabalin in acute treatment of older
patientsresponse (reduction in HAMA by 50 or
more) rates
Montgomery SA et al. Br J Psychiatry. In press
17What equates to symptom remission in GAD?
Bandelow et al. J Clin Psychiatry 2006 67
1428-1434
18Fixed-dose escitalopram and paroxetine in GAD
symptomatic remission (HAMA score lt 7)
p lt 0.05, p lt 0.01, p lt 0.001, versus
placebo p lt 0.05 versus paroxetine
Baldwin et al. Br J Psychiatry 2006 189 262-272
19Symptomatic remission with paroxetine in GAD
p lt 0.05, p lt 0.01 vs placebo 20-mg and
40-mg groups combined Rickels et al. J Clin
Psychiatry 2006 67 41-47
203. Prediction of response to treatment
21Predictors of good and poor response in GAD
- shorter duration of symptoms (venlafaxine,
fluoxetine)1, 2 - co-morbid dysthymia (venlafaxine)1, psychiatric
co-morbidity3 - history of depression, panic disorder
(venlafaxine)4 - severity of psychosocial impairment (TAU) 3
- lower symptom severity (escitalopram) 5
- history of benzodiazepine use (venlafaxine) 4
- longer duration of untreated illness (SSRI,
venlafaxine) 6
1. Perugi G et al. Neuropsychobiol 2002 46
145-149 2. Simon NM et al. Depress Anx 2006 23
373-376 3. Rodriguez BF et al. J Nerv Ment Dis
2006 194 91-97 4. Pollack M et al. J Clin
Psychopharmacol 2003 23 250-259 5. Stein DJ et
al. J Clin Psychiatry 2006 67 1741-1746 6.
Altamura CA et al. CNS Spectrums 2008 13 415-422
22Early onset of effect and subsequent response
Study Treatment Signpost Destination
1 Diazepam Week 1 Reduction in HAMA Week 6. Status as high, medium or low improvers strongly predictive of response.
2 Buspirone Lorazepam Week 2 Reduction in HAMA by less than 25 Week 6. Lack of early response strongly predictive of non-response (less than 50 reduction in HAMA) at endpoint
3 BZDs Azapirones Placebo Weeks 1 and 2 Reduction in HAMA Week 8. Response levels (robust, partial, poor) significantly predictive of end response (reduction in HAMA of gt50 from baseline).
4 Duloxetine Placebo Weeks 2 and 4 Reduction in HAMA 20, 40, 60, 80 Week 9 or 10. Degree of response strongly predictive of symptomatic remission
5 Escitalopram Week 2 Reduction in HAMA by 20 or more Endpoint. Onset strongly predictive of response (reduction of gt50 on HAMA).
1. Downing and Rickels. Acta Psychiatr Scand
1985 72 522-528 2. Laakmann et al.
Psychopharmacol 1998 136 357-366 3. Rynn et
al. Depress Anx 2006 23 461-465 4. Pollack et
al. J Psych Res 2008. In press 5. Baldwin et al.
Unpublished findings
23Onset of action and remission with duloxetine
proportion achieving remission (HAMA lt7) at
endpoint
Degree of reduction in HAMA score at week 2 and
week 4
Pollack MH et al. J Psych Res 2008. In press.
24Early onset of effect and later overall
responseanalysis of the escitalopram clinical
trial database
Baldwin DS et al, submitted for publication.
Poster presentation at 2nd IADC, Cape Town, March
2008
254. Duration of treatment after response
26Clinical outcomes in anxiety disordersprospective
, naturalistic, longitudinal (12 year) study
(HARP)
recovery at least 8 consecutive weeks with at
most residual symptoms
recurrence full diagnostic criteria for minimum
period (2 weeks, other than GAD)
Bruce et al. Am J Psychiatry 2005 162 1179-1187
27Outcome of DSM-III defined GAD in the Zurich
Studyprospective epidemiological study, 22-year
follow-up (n105)
These results are clearly at odds with the
common affirmation that GAD has a generally
chronic course, but are compatible with the
description of a waxing and waning course
however our data show that waning is more common
than waxing.
Angst J, Gamma A, Baldwin DS et al. Eur Arch
Psychiatry Clin Neurosci. In press.
28Placebo-controlled relapse prevention studies
p lt 0.001 vs placebo
n168
n278
n288
n187
n188
n170
pregabalin, 24 wks
paroxetine, 24 wks
escitalopram, 24-72 wks
duloxetine, 24 wks
Feltner et al. Int Clin Psychopharmacol 2008 23
18-28 Stocchi et al. J Clin Psychiatry 2003 64
250-258 Allgulander et al. Int J
Neuropsychopharmacol 2006 9 495505 Korb et
al. Presented at IADS Cape Town, March 2008
295. Management after initial non-response
30Possible treatment options after non-response
- increase in dose
- switch to alternative pharmacological treatment
- SNRI
- pregabalin
- buspirone
- agomelatine
- quetiapine
- augmentation with olanzapine or risperidone
- combination drug and psychological treatment
- use of complementary approaches
31SSRIs versus psychic and somatic
symptomsdouble-blind placebo-controlled
fixed-dose studies
Rickels et al 2003 Paroxetine Rickels et al 2003 Paroxetine Baldwin et al 2006 Escitalopram Baldwin et al 2006 Escitalopram Baldwin et al 2006 Escitalopram
Daily dose 20 mg 40 mg 5 mg 10 mg 20 mg
HAMA Yes Yes No Yes Yes
Psychic Factor Yes Yes No Yes No
Somatic Factor No No No Yes Yes
Boxes show significant differences from placebo (
lt 0.05, or less)
32SNRIs versus psychic and somatic
symptomsdouble-blind placebo-controlled
fixed-dose studies
Rickels et al 2000 Venlafaxine Rickels et al 2000 Venlafaxine Rickels et al 2000 Venlafaxine Allgulander et al 2001 Venlafaxine Allgulander et al 2001 Venlafaxine Allgulander et al 2001 Venlafaxine
Daily dose 75 mg 150 mg 225 mg 37.5 mg 75 mg 150 mg
HAMA Total No No Yes Yes Yes Yes
Psychic Factor No Yes Yes Yes Yes Yes
Somatic Factor No No No No Yes Yes
Boxes show significant differences from placebo
(plt 0.05, or less)
33Azapirones in acute treatment of GAD
- efficacious in acute treatment especially if
benzodiazepine-naïve - less well tolerated than benzodiazepines
Chessick CA et al. Cochrane Database Syst Rev
2006 3 CD006115
34Agomelatine in acute treatment of GADrandomised
double-blind flexible-dose placebo-controlled
12-week study
p 0.026
p 0.027
Mean baseline HAMA scores placebo 28.6,
agomelatine 29.0 Stein DJ et al. Eur
Neuropsychopharmacol 2007 17 (suppl. 4) S509-510
35Antipsychotic augmentation after
non-responsechange in HAMA score and response on
CGI-I
NS
p lt 0.05
Pollack MH et al. Biol Psychiatry 2006 59
211215 (mean olanzapine daily dose 8.7 7.1 mg,
following fluoxetine 20mg/day) Brawman-Mintzer
et al. J Clin Psychiatry 2005 66 13211325
(mean risperidone daily dose 1.1 mg, following
range of drugs)
36Quetiapine in acute treatment of GADrandomised
double-blind fixed-dose placebo-controlled 8-wk
study (LOCF, modified ITT)
Mean baseline HAMA scores placebo 27.3, QUET 50
mg 26.9, QUET 150 mg 26.6, PAR 20 mg 27.1
Bandelow B et al. Int J Psychiatry Clin Pract
2007 11 314-315 (abstract)
37Complementary medicine in GAD
Study and Agent Design Findings
Connor and Davidson 2006 Kava kava Randomised double-blind placebo- and comparator-controlled (venlafaxine) studies. Pooled analysis of 3 small studies (kava n28, placebo n30, VEN n6). No evidence for efficacy of kava kava. Placebo superior to kava and venlafaxine in more severely ill patients.
Woelk et al 2007 Ginkgo biloba EGb761 Randomised double-blind placebo-controlled fixed-dose (240 mg/day or 480 mg/day) study. Mixed diagnostic group GAD (n82) and adjustment disorder with anxious mood (n25). EGb761 was significantly superior to placebo on primary (HAMA) and secondary outcome measures (inc. CGI-I).
Herrera-Arettano et al 2007 Galphimia glauca Randomised comparator-controlled (lorazepam 1mg/day) study (n152) of capsules of aqueous extract. Similar reduction in anxiety symptoms severity observable from first week of treatment.
Connor KM, Davidson JRT. Int Clin Psychopharmacol
2006 21 249-253. Woelk H et al. J Psychiatr
Res 2007 41 472-480. Herrera-Arettano et al.
Planta Med 2007 73 713-717.
38Combining CBT and drug treatment in GAD
- due to lack of data, not currently possible to
draw conclusions for GAD - continuing need for large RCT of CBT vs SSRI vs
CBTSSRI
Bandelow B et al. World J Biol Psychiatr 2007 8
175-187
396. Future developments in drug treatment
40Emotion recognition and anxiety disorders
- recent cross-sectional studies indicate
attentional bias in social phobia - high social anxiety associated with selective
attention towards threatening images
- social phobia associated with deficits of
discriminating fear from other emotions - accuracy of recognition in depression can be
enhanced by tryptophan supplementation
41Correlates of disorder-specific contextual threat
processing Virtual Reality
42Acute SSRI treatment and cognitive bias in GAD
p lt .05
Mogg, Baldwin et al. Psychopharmacol 2004 176
466-470
43Genetic polymorphisms, efficacy and tolerability
- genotyping for cytochrome P450 enzymes might be
useful in deciding which SSRI to use and at what
dose 1 - selection of drug treatment on the presence of
genetic polymorphisms known to be predictive of
response or tolerability 2 - 5-HTTLPR polymorphisms predicts response to SSRI
in patients with generalised social phobia 3
1. Perlis RH. Br Med J 2007 334 259 2. Wong ML,
Licinio J. Nature Rev Drug Discovery 2004 3
136-151 3. Stein MB et al. Psychopharmacol 2006
187 68-72
44New targets for anxiolytic drugs
- SSRI plus 5-HT1A and 5-HT1B autoreceptor
antagonists - SSRI plus 5-HT2C antagonists
- melatonin receptor agonists
- metabotropic glutamate receptor antagonists
- cholecystokinin antagonists
- neuropeptide Y agonists
- adenosine A1 and A2A receptor agonists
- refined enhancement of GABA activity
45What might be in future BAP guidelines?
- recognition of evidence of efficacy in acute
treatment - agomelatine
- duloxetine
- pregabalin
- quetiapine
- reduction in length of initial treatment before
assessment - recognition of efficacy of pregabalin in
preventing relapse - recommendation for 12 months of treatment after
response
46Expanding the evidence base in GAD
- neuropsychobiological measures in response
prediction - studies in patients with comorbid depressive
disorders - dosage escalation studies after initial
non-response - placebo-controlled augmentation studies
- pregabalin, buspirone, agomelatine,
antipsychotics - large randomised controlled trials of combination
treatment versus CBT or drug treatment, when
given alone
47Thank you very much!