Pregabalin vs Venlafaxine

7/25/2019 Pregabalin vs Venlafaxine

1/10

Original article 87

Efficacy of pregabalin and venlafaxine-XR ingeneralized anxiety disorder: results of a double-blind,placebo-controlled 8-week trialSiegfried Kaspera, Barry Hermanb, Giancarlo Nivolic, Michael Van Ameringeng,

Antonino Petraliad

, Francine S. Mandelb

, Francesca Baldinettie

and Borwin Bandelowf

The objective of this study was to evaluate the anxiolytic

efficacy, and speed of onset of efficacy, of pregabalin

(PGB) and venlafaxine-XR (VXR) in patients with

generalized anxiety disorder (GAD). In this double-blind

trial, outpatients, ages 1865 years, who met Diagnostic

and Statistical Manual of Mental Disorders, 4th edition,

criteria for GAD were randomized to 8 weeks offlexible-dose treatment with PGB (300600 mg/day), VXR

(75225 mg/day), or placebo (PBO). The intent-to-treatsample consisted of 121 patients on PGB [least square

(LS) mean SE baseline Hamilton Anxiety Rating Scale(HAM-A), 27.6 0.4], 125 patients on VXR (baseline HAM-A,

27.4 0.4), and 128 patients on PBO (baseline HAM-A,26.8 0.4). Treatment with PGB was associated with a

significantly greater LS mean change in the HAM-A totalscore at last observation carried forward endpoint versus

PBO ( 14.5 0.9 vs. 11.7 0.9;P= 0.028). Treatmentwith VXR was not significant versus PBO at endpoint

(12.00.9; 11.70.9; P= 0.968). Treatment with PGB

showed an early onset of improvement, with significantly

greater LS mean change in the HAM-A by day 4 versusboth PBO ( 5.3 0.5 vs. 3.4 0.5;P= 0.008) and VXR

(2.90.5; P= 0.0012). The proportion of patients

reporting a severe adverse event was similar for PGB

(9.1%) and PBO (7.8%), but higher for VXR (20.0%;P< 0.05). In conclusion, PGB was a safe and effective

treatment of GAD, with a significantly earlier onset of

anxiolytic activity than VXR. Int Clin Psychopharmacol

24:8796 c 2009 Wolters Kluwer Health | Lippincott

Williams & Wilkins.

International Clinical Psychopharmacology 2009, 24:8796

Keywords: generalized anxiety disorder, pregabalin, venlafaxine-XR

aDepartment of General Psychiatry and Psychotherapy, Medical University ofVienna, Vienna, Austria, bPfizer Global Pharmaceuticals, Pfizer Inc., New York,USA, cUniversita di Sassari, Sassari, dClinica Psichiatrica, Universita degli Studidi Catania, Catania, eIRCCS S. Lucia Foundation-Neurology Department,University of Rome, Rome, Italy, fDepartment of Psychiatry and Psychotherapy,University of Gottingen, Gottingen, Germany and gMcMaster University MedicalCentre, Hamilton Health Sciences, Hamilton, Ontario, Canada

Correspondence to Professor Borwin Bandelow, Department of Psychiatry andPsychotherapy, University of Gottingen, von-Siebold-Str 5, D-37085 Gottingen,GermanyTel: + 49 551 39 6607; fax: + 49 551 39 8592;

e-mail: [email protected]

Received26 September 2008 Accepted13 October 2008

IntroductionGeneralized anxiety disorder (GAD) is a chronic illness

with an estimated 1 year prevalence of approximately 3%,

and a lifetime prevalence of approximately 6% (Kessler

et al., 2005; Lieb et al., 2005). GAD (without depression

comorbidity) is associated with significant impairment inquality of life (QoL) and functioning, which has been

found to be comparable with major depressive disorder(MDD), and chronic medical illnesses such as diabetes

and arthritis (Kessler et al., 1999; Wittchen, 2002).

At least half a dozen classes of drugs are available for the

pharmacologic management of GAD (Bandelow et al.,

2008), each acting through different mechanisms: benzo-

diazepines (diazepam, lorazepam, alprazolam etc), which

augment inhibitory g-amino butyric acid (GABA)-ergic

activity; monoaminergic reuptake inhibitors, consisting

of selective serotonin reuptake inhibitors drugs with

serotonin selectivity (paroxetine, escitalopram, sertra-

line), serotoninnorepinephrine reuptake inhibitor drugs

with dual serotonin/norepinephrine activity [venlafaxine-

XR (VXR) duloxetine], as well as some first-generation

tricyclics (imipramine); azapirones (buspirone), which

modulate monoaminergic transmission; and pregabalin(PGB) that acts presynaptically to inhibit excitatory

neurotransmission. Two other classes of medication(antihistamines such as hydroxyzine; antipsychotics such

as quetiapine) have also shown efficacy in the treatmentof GAD. Cross-study comparisons suggest that each class

of drug has a different benefitrisk profile. Relatively fewdouble-blind, placebo-controlled head-to-head trials,

however, have been published, which provide direct

comparisons of the efficacy and safety profiles of drugs

in each class. The primary objective of this study was

to evaluate the efficacy of PGB and VXR compared with

placebo in the treatment of GAD.

The efficacy of VXR in GAD has been confirmed based

on the results of eight large, placebo-controlled trials

(Davidsonet al., 1999; Gelenberget al., 2000; Rickelset al.,

0268-1315 c 2009 Wolters Kluwer Health | Lippincott Williams & Wilkins DOI: 10.1097/YIC.0b013e32831d7980


2/10

2000; Allgulander et al., 2001; Hackett et al., 2003;

Lenox-Smith et al., 2003; Hartford et al., 2007; Nicolini

et al., 2008). Four of these trials were flexible-dose studies

(Gelenberget al., 2000; Lenox-Smithet al., 2003; Hartfordet al., 2007; Nicoliniet al., 2008), and four were fixed-dose

studies (Davidson et al., 1999; Rickels et al., 2000;Allgulanderet al., 2001; Hackett et al., 2003), with sample

sizes ranging from 81 to 185 per arm. Four of the placebo-controlled trials also included an active comparator. In

one fixed-dose study (Davidson et al., 1999), endpoint

improvement in the Hamilton Anxiety Rating Scale

(HAM-A) total score was not significantly different from

placebo on the 75 and 150mg doses of VXR, or on the

buspirone comparator. In a second fixed-dose study with

a high placebo response rate, diazepam was significantly

superior to placebo, but neither dose of VXR was

different from placebo on HAM-A total score (Hackett

et al., 2003). In two recent three-arm, flexible-dose,

duloxetine comparator studies (Hartford et al., 2007;

Nicolini et al., 2008), VXR showed significant improve-ment compared with placebo. Finally, VXR and

duloxetine showed noninferiority in an a priori pooled

analysis of data from two placebo-controlled trials

(Allgulander et al., 2008). In a secondary analysis, VXR

showed significant endpoint improvement in the HAM-A

total score versus placebo.

PGB is a new anxiolytic with potent antinociceptive

(Sonnett et al., 2006) and antiepileptic (Hamandi andSander, 2006) efficacy, as well as efficacy in treating

fibromyalgia (Croffordet al., 2005). Structurally, PGB is analkylated analogue of GABA, however, unlike GABA it has

no direct or indirect GABAergic activity. In contrast to thebenzodiazepines, which bind to a modulatory site on the

GABAreceptor complex to augment the inhibitory

effects of GABA, available data suggest that PGB acts

presynaptically to inhibit excitatory neurotransmission

(Dooleyet al., 2002; Finket al., 2002; Taylor et al., 2007).

PGB binds to the a2d subunit of N-type and P/Q-type

calcium channels, resulting in a reduction of calcium

influx in response to an action potential. As a result of

the inhibition of the inward calcium current, there is

a significant reduction in the ability of synaptic vesicles to

fuse with the presynaptic membrane, and release

neurotransmitter, thus propagating the action potential.

The efficacy of PGB in GAD has been confirmed based

on the results of seven large, placebo-controlled trials

(Feltner et al., 2003; Pande et al., 2003; Pohl et al., 2005;

Rickelset al., 2005; Montgomery, 2006; Montgomeryet al.,

2006; Montgomery et al., 2008). Six of these trials were

fixed-dose, and one trial (in the elderly) was flexible

dose, representing a combined total of 15 PGB treatment

arms, with sample sizes ranging from 66 to 177 per arm.

Overall, PGB showed significant efficacy when compared

with placebo. The 150-mg dose level, however, only

achieved significance in one of three studies, and thus

PGB seems to have a doseresponse curve between 150

and 300 mg, whereas no doseresponse relationship isevident between 300 and 600 mg (Bech, 2007). In

contrast, a clear doseresponse relationship has not beenshown for VXR across its approved dosage range of

75225 mg/day (Wyeth, 2007).

The primary objective of this study was to evaluate the

efficacy of PGB and VXR compared with placebo in the

treatment of GAD. The secondary objectives were to

evaluate the onset of anxiolytic activity in PGB compared

with both placebo and VXR; to evaluate the ability of

PGB to improve the QoL and functioning in patients

with GAD; to evaluate the ability of PGB to improve

insomnia, pain, and depressive symptoms commonly seen

in patients with GAD; and finally, to evaluate the

tolerability and safety of PGB and VXR.

MethodsStudy design

This was an 8-week, double-blind, placebo-controlledstudy of PGB and VXR for the treatment of GAD patients

who met Diagnostic and Statistical Manual of MentalDisorders, 4th edition (DSM-IV) criteria. After a 1-week,

open-label, lead-in period, study patients wererandomized to 8 weeks of double-blind, parallel-group

treatment with flexible doses of either PGB, venlafaxine,or placebo. PGB treatment was started at a dose of

150mg twice daily for the first week; thereafter, PGBdosing was flexible, based on clinical response and

tolerability, in the range of 300600 mg/day, administered

twice daily. VXR treatment was started at a dose of 75 mg/day (administered in the morning, with matching placebo

in the evening) for the first week. Thereafter, VXR dosing

was flexible, in the range of 75225 mg/day, administered

in the morning (with matching placebo in the evening).

Patients were randomized to one of the three treatment

groups based on a computer-generated randomization list.

The study was conducted at 47 investigational sites in

Belgium, Canada, France, Ireland, Italy, The Netherlands,

Spain, and Sweden. The protocol was approved at each

site by the appropriate institutional review board, and

written informed consent was obtained from each patientbefore enrolment.

Patient selection

Patients were recruited from clinic referrals and from

advertisements in local media. Male and female

outpatients ages 1865 years were eligible for study

entry if they met DSM-IV-Text Revision criteria for a

primary diagnosis of GAD based on a structured Mini

International Neuropsychiatric Interview (M.I.N.I. Plus;

version 5.0.0; Sheehan et al., 1997), and if their HAM-A

total score was Z 20 at both the screening and baseline

visits. The HAM-A psychic and somatic anxiety factors

88 International Clinical Psychopharmacology 2009, Vol 24 No 2


3/10

also were required to be Z 10 at both the screening and

baseline visits. Women of childbearing potential wererequired to have a negative serum b-human chorionic

gonadotropin pregnancy test and be practicing amedically accepted form of birth control.

Patients were excluded if they presented with any of

the following: (i) a current or past DSM-IV diagnosis ofbipolar disorder, schizophrenia, or any other psychotic

disorder; (ii) a DSM-IV diagnosis in the past 6 months

of MDD, dysthymic disorder, obsessivecompulsive

disorder, posttraumatic stress disorder, an eating disorder,

or alcohol or substance dependence and/or abuse; (iii) a

17-item Hamilton Depression Rating scale (HAM-D)

total score Z 15; (iv) a history of seizure disorder (except

febrile seizures in childhood); (v) any clinically signifi-

cant acute or unstable medical condition; (vi) a positive

urine drug screen (for benzodiazepines, ethanol, amphetamines, barbiturates, cocaine, opiates, cannabinoids, phen-

cyclidine); (vii) creatinine clearance rates r 60 ml/min;(viii) concurrent psychotherapy for GAD (psycho-

therapy not targeting GAD symptoms was permitted ifinitiated > 3 months before study enrolment); (ix) use

of concomitant psychotropic medications within 2 weeksof the baseline visit (5 weeks for fluoxetine; except,

zopiclone or zolpidem were permitted on two nights, asneeded, during the 1-week washout period); (x) current

suicide risk based on the clinical judgment of theinvestigator; (xi) previous treatment with either PGB or

VXR; and (xii) lactating women.

Efficacy measures

The primary efficacy measure was the 14-item, investi-gator-rated HAM-A total score (Hamilton, 1959), which

was completed at screen baseline, day 4 (a telephone

assessment), and at each assessment visit during

double-blind treatment. Secondary investigator-rated

measures consisted of: (i) the seven-item HAM-A psychic

anxiety factor score; (ii) the seven-item HAM-A somatic

anxiety factor score; (iii) the Clinical Global Impression

(CGI) Severity scale (completed at baseline and all

post-baseline assessment visits; Guy, 1996); (iv) the CGI

Improvement scale (completed at every post-baseline

visit; Guy, 1996); and (v) the 17-item HAM-D

(completed at screen, baseline, and weeks 1, 4, and 8;Hamilton, 1960).

Secondary patient-rated scales consisted of: (i) the

14-item Hospital Anxiety and Depression Scale

[(HADS); Zigmond and Snaith, 1983] total score, and

the seven-item anxiety (HADS-A) and seven-item

depression (HADS-depression) subscale scores (com-

pleted at baseline, and weeks 4 and 8); (ii) the 100 mm

Global Anxiety Visual Analogue Scale [GA-VAS;

completed at screen and baseline, day 4 (telephone

assessment), and at all assessment visits]; (iii) Daily Pain

Rating Scale [Dworkinet al., 2005; a single-item that rates

pain severity on an 11-point numerical scale; completed

at screen and baseline, day 4 (telephone assessment), andat all assessment visits]; (iv) QoL, Enjoyment, and

Satisfaction Scale (Endicott et al., 1993; completed atbaseline, and weeks 4 and 8); (v) the EuroQoL health

status profile questionnaire (EQ-5D; Brooks, 1996;completed at baseline, and weeks 1,3,4, 6, and 8); (vi)

the EuroQoL visual analogue scale (completed at base-line, and weeks 4 and 8); (vii) the Sheehan Disability

Scale (SDS; Sheehan et al., 1996) total score and Work

(SDS-work), Family (SDS-family), and Social (SDS-

social) subscale scores (completed at baseline, and weeks

2, 4 and8); and (viii) the 12-item Medical Outcomes

Study Sleep Scale (MOS-Sleep; Hayset al., 2005), and its

seven subscale scores and nine-item sleep problems index

(completed at baseline, and weeks 4 and 8).

Safety and tolerability measures

Blood chemistry, hematology, urinalysis, and physical

examination were carried out at the screen and week 8visits (or at the last visit, if the patient was discontinuing

prematurely). Vital signs and weight were obtained at

every study visit. Spontaneously reported or observed

adverse events were recorded, regardless of causality,

in terms of time of onset, severity, and outcome. Use of

concomitant medications was recorded in terms of daily

dose, stop and start dates, and reason for use.

Treatment-related sexual dysfunction was evaluated

using the male and female versions of the Changes

in Sexual Functioning Questionnaire (CSFQ-male;

CSFQ-female; Clayton et al., 1997), which were com-pleted at baseline, and weeks 4 and 8.

Statistical analyses

On the basis of data from previous PGB GAD studies,

it was estimated that a sample size of 130 patients

per treatment group would have 90% power to detect a

3-point difference in HAM-A total score as significant on

a two-sided test with a type I error rate of 0.05.

All statistical analyses were performed using SAS

statistical package (version 8) (SAS Institute, Cary, North

Carolina, USA, 2000) on the intent-to-treat population

consisting of all randomized patients who received at

least one dose of study medication.

The primary analysis was a comparison of endpoint

change in HAM-A total score between PGB and placebo

based on an analysis of covariance model with center

and baseline as covariates, and utilizing tests for center

by treatment and treatment by baseline interactions. In

addition, the two active treatment groups were compared

with placebo using Dunnetts test. A mixed-effects model

repeated-measures (MMRM) analysis was also carried

out post hoc on the HAM-A total score, and psychic and

Efficacy of pregabalin and venlafaxine Kasper et al. 89


4/10

somatic anxiety factor scores. All other secondary efficacy

outcomes were analyzed with an analysis of covariancemodel as described above.

All efficacy analyses were two sided with a = 0.05.

Hierarchical testing was used to control overall type I

error at a = 0.05. The hierarchy of hypotheses tests used

in the primary tests and key secondary analyses wereapplied at the a = 0.05 level for each hypothesis test.

ResultsFour-hundred-sixty-six patients provided informed con-

sent and were recruited into the study, of which 374

(80.3%) met eligibility criteria at baseline and agreed to

be randomized to 8 weeks of study treatment (Fig. 1).

The intent-to-treat sample, used in both safety and

efficacy analyses, consisted of 374 patients. No significantbaseline differences were observed in the demographic or

clinical characteristics of patients assigned to each

treatment group (Table 1).

Anxiolytic efficacy

Treatment with PGB resulted in significant improvement

compared with placebo on the primary endpoint, least

squares (LS) mean change in HAM-A total score at last

observation carried forward endpoint (Table 2). The

endpoint difference in HAM-A change score in the VXR

treatment group was not significantly different fromplacebo. A post-hoc, MMRM analysis confirmed that

there was a significant improvement on PGB, but notVXR, compared with placebo across the 8 weeks of study

treatment (Fig. 2). Post-hoc, MMRM analyses also foundsignificant improvement on PGB compared with placebo

on both the HAM-A psychic anxiety factor (Fig. 3), andthe HAM-A somatic anxiety factor (Fig. 4). Treatment

Fig. 1

Screen visitN=466

1-week drug-free

Baseline visitITT sample, N=374

Discontinued beforereceiving study drug, N=92

Did not meet entry criteria, N=73Withdrew consent, N=13

Other reasons, N=6

Pregabalin 150600 mgN=121

Venlafaxine-XR 75225 mgN=125

PlaceboN=128

8-week DB treatment 8-week DB treatment 8-week DB treatment

Discontinued during DB, N=33 (27.3%) Discontinued during DB, N=41 (32.8%) Discontinued during DB, N=35 (27.3%)Lack of efficacy, N=4 (3.3%)

Adverse events, N=15 (12.4%)Other reasons, N=14 (11.6%)

Completed DB, N=88 (72.7%) Completed DB, N=84 (67.2%) Completed DB, N=93 (72.7%)

Lack of efficacy, N=4 (3.2%) Lack of efficacy, N=12 (9.4%)Adverse events, N=22 (17.6%) Adverse events, N=7 (5.5%)Other reasons, N=15 (12.0%) Other reasons, N=16 (12.5%)

PGB 600 mg: 2 week taperPGB 450 mg: 1 week taper

PGB 300 mg: abrupt discontinuation

Completed taper, N=47 Completed taper, N=58 Completed taper, N=66

Venla-XR 225 mg: 2 week taper

Venla-XR 150 mg: 1 week taperVenla-XR 75 mg: abrupt discontinuation

Flow diagram. DB, double-blind; ITT, intent-to-treat.

Table 1 Baseline clinical and demographic characteristics ofpatient sample

PregabalinN=121

Venlafaxine-XRN=125

PlaceboN=128

Female (%) 64 58 61Age, mean SD (years) 39.5 1 1. 9 42.6 11.8 40.2 1 2.1White (%) 71 70 72Weight, mean S D (kg) 72.5 1 7. 9 72. 6 14.6 74.4 1 4.6Time since first diagnosis of

GAD, mean (years)

3.1 4.0 4.6

HAM-A, LS mean SETotal score 27.6 0.4 27.4 0.4 26.8 0.8Psychic factor score 14.4 0 .3 14.0 0 .3 13.8 0 .3Somatic factor score 13.3 0 .3 13.4 0 .3 12.9 0 .3

HAM-D-17, LS mean S E 11.5 0 .2 11.5 0 .2 11.3 0 .2

GAD, generalized anxiety disorder; HAM-A, Hamilton Anxiety Rating Scale;HAM-D, Hamilton Depression Rating scale; LS, least square.



5/10

with VXR was not associated with significant improve-

ment compared with placebo in either HAM-A factor.

Treatment with PGB was associated with significant

endpoint improvement on one of the two secondary

patient-rated measures of anxiety (HADS-A; Table 2),

but not on the other, the GA-VAS. VXR showed trend

levels of significance on the HADS-A (P> 0.07), but was

nonsignificant on the GA-VAS. In addition, PGB-treated

patients showed significantly greater improvementcompared with placebo-treated patients on both the

CGI Improvement and CGI Severity measures (Table 2).

In contrast, patients treated with VXR did not show

efficacy on these two global measures. The proportion of

patients who met HAM-A responder criteria (Z 50%

reduction from baseline) was significantly higher on

PGB compared with both placebo and VXR (59 vs.

46%, Cochran-Mantel-Haenszel; P= 0.05, and 44%,

Cochran-Mantel-Haenszel;P= 0.05).

Early onset of anxiolytic efficacy

An a priori study objective was to evaluate the time

of onset of anxiolytic efficacy. Treatment with PGB

was associated with significantly greater improvement

Table 2 Primary and secondary efficacy variables at baseline and LOCF-endpoint in patients with generalized anxiety disorder treated withpregabalin, venlafaxine-XR, or placebo

Pregabalin N= 121 Venlafaxine-XR N= 125 Placebo N=128

Investigator-rated scales LS mean S E Pvaluea LS meanSE Pvaluea LS mean SE

HAM-A totalBaseline 27.6 0.4 27.4 0.4 26.8 0.4Endpoint change 14.5 0.9 0.028 12.0 0.9 0.97 11.7 09

HAM-A psychic anxietyBaseline 14.4 0.3 14.0 0.3 13.8 0.3Endpoint change 7.3 0.5 0.017 5.9 0.5 0.90 5.6 0.5

HAM-A somatic anxietyBaseline 13.3 0.3 13.4 0.3 12.9 0.3Endpoint change 7.3 0.4 0.11 6.1 0.5 0.996 6.2 0.5

CGI-SeverityBaseline 4.7 0.1 4.6 0.1 4.5 0.1Endpoint change 2.0 0.2 0.014 1.7 0.2 0.55 1.5 0.2

CGI-ImprovementEndpoint change 2.3 0.1 0.050 2.5 0.1 0.53 2.7 0.1

HAM-D17totalBaseline 11.5 0.2 11.5 0.2 11.3 0.2Endpoint change 4.4 0.5 0.018 3.6 0.5 0.36 2.8 0.5

Patient-rated symptom scalesHADS-anxiety scale

Baseline 14.2 0.3 13.8 0.3 13.9 0.3

Endpoint change 5.0 0.4 0.027 4.8 0.4 0.07 3.7 0.4Global Anxiety VAS

Baseline 54.1 1.8 48.5 1.9 49.4 1.8Endpoint change 18.5 2.4 0.10 16.9 2.4 0.26 12.8 2.3

MOS-Sleep problems indexBaseline 52.0 1.8 51.0 1.8 49.3 1.8Endpoint change 18.1 1.8 0.002 10.1 1.9 0.98 10.5 1.8

MOS-Sleep disturbanceBaseline 55.9 2.4 53.5 2.4 51.5 2.4Endpoint change 22.2 2.2 < 0.001 11.6 2.2 0.98 12.0 2.2

Daily pain rating scaleBaseline 3.8 0.2 3.1 0.2 3.1 0.2Endpoint change 1.0 0.2 0.99 1.1 0.2 0.53 0.9 0.2

Patient-rated QoL and functioning scalesQ-LES-Q total score

Baseline 47.1 1.4 47.3 1.4 50.1 1.4Endpoint change + 11.5 1.7 0.13 + 12.7 1.7 0.038 + 7.7 1.7

Sheehan Disability ScaleBaseline 16.2 0.6 16.0 0.6 15.5 0.6Endpoint change 6.7 0.8 < 0.005 4.7 0.8 0.50 3.7 0.8

EuroQoL EQ-5DBaseline 0.46 0.03 0.48 0.03 0.52 0.03Endpoint change + 0.22 0.03 0.40 + 0.21 0.03 0.48 + 0.18 0.03

EuroQoL VASBaseline 50.6 1.9 51.5 1.9 53.3 1.9Endpoint change + 14.4 2.0 0.25 + 12.7 2.1 0.63 + 10.6 2.1

Data shown are from an ANCOVA model with baseline and site as covariates.ANCOVA, analysis of covariance; CGI, Clinical Global Impression; HADS, Hospital Anxiety and Depression Scale; HAM-A, Hamilton Anxiety Rating Scale; HAM-D,Hamilton Depression Rating Scale; LOCF, last observation carried forward; LS mean, least square mean data; MOS, Medical Outcomes Study; Q-LES-Q, Quality of Life;QoL, quality of life, Enjoyment, and Satisfaction Scale; SE, standard error; VAS, visual analogue scale.aData shown are from an ANCOVA model comparing endpoint change scores with baseline and site as covariates; significance testing for pregabalin and venlafaxine-XRversus placebo was performed using Dunnetts test.



6/10

in HAM-A total score than both placebo and VXR,

respectively, on day 4 (by phone assessment: 5.3 0.5 vs. 3.4 0.5;P= 0.008; and vs. 2.94 0.6;P= 0.001); and

on day 7 ( 7.9 0.6 vs. 5.4 0.6; P= 0.002; and vs. 5.6 0.6; P= 0.005). Twenty percent or greater

improvement from baseline in HAM-A total score wasachieved on day 4 by significantly more patients treated

with PGB (36.3%) compared with VXR (18.3%; P= 0.008)or placebo (20.3%; P= 0.002).

On an MMRM analysis, treatment with PGB resulted

in greater early improvement, on day 4, compared

with VXR and placebo on the HAM-A psychic factor

score ( 3.51 0.48 vs. 2.57 0.50; P= 0.02; and vs. 2.13 0.49; P= 0.0003), and on the HAM-A somatic

anxiety factor score ( 3.69 0.52 vs. 3.02 0.54;P= 0.12; and vs. 2.54 0.54; P= 0.008).

On the patient-rated GA-VAS, treatment with PGB

was associated with significantly greater improvement

than both placebo and VXR, respectively, on day 4:

1 1.3 1.6 vs. 4 .3 1.6; P=0 .002; and vs.

5.2 1.6; P= 0.01; and on day 7: 15.3 1.8 vs.

5.2 1.8;P< 0.0001; and vs. 8.6 1.8;P= 0.01.

Fig. 2

5

0

Week1

Week2

Week3

Week4

Week6

Week8

LOCF

Placebo comparison

P


7/10

Secondary symptom measures: insomnia, depression, pain

PGB-treated patients showed significantly greater

endpoint improvement than placebo-treated patients on

both the four-item MOS-Sleep Disturbance Scale, and

the nine-item MOS-Sleep problems index (Table 2). The

effect of VXR on sleep measures was similar to placebo.

In this study sample, HAM-D and Daily Pain Ratingscores were relatively low at pretreatment baseline. On

the 17-item HAM-D total score, the PGB group showed

significantly greater endpoint improvement than placebo,whereas VXR did not (Table 2). No difference was foundbetween PGB or VXR and placebo on the Daily Pain

Rating Scale (Table 2).

Quality of life and functioning

The treatment effect of PGB and VXR on patient-rated

QoL and functioning was modest and only intermittently

significant. At endpoint, PGB-treated patients showed

significant improvement compared with placebo on the

SDS, but not on any of the three QoL measures (Table 2).

In contrast, patients treated with VXR showed significant

improvement compared with placebo on the Quality ofLife, Enjoyment, and Satisfaction Scale, but not on the

SDS, or the other two QoL scales (Table 2).

Tolerability and safety

The mean maximal daily dose of PGB was 424 118 mg;

the proportion of patients using a daily dose of 450 mg

(37.2%) was somewhat higher than the proportion of

patients using 300 mg (33.1%) and 600 mg (28.9%). The

mean maximal daily dose of VXR was 155 60 mg; the

proportion of patients using a daily dose of 75 mg (37.6%)

was somewhat higher than the proportion of patients

using 150 mg (32.0%), 225 mg (29.6%), or 300 mg (0.8%).

Both PGB and VXR were well tolerated in the dosage

range used in this study. The most frequent adverse

events on PGB were dizziness, headache, and nausea;

whereas the most frequent adverse events on VXR were

nausea, headache, and fatigue (Table 3). The proportion

of adverse events reported as severe was significantly

higher on VXR (20.0%) compared with placebo (7.8%;

Fishers exact test,P= 0.002), whereas the rate of severe

adverse events was intermediate on PGB (9.1%).Consistent with this, the rate of attrition because of

adverse events was numerically higher on VXR comparedwith placebo (17.6 vs. 5.5%), and again was intermediate

(12.4%) on PGB.

Treatment with PGB and VXR had only modest effects on

the CSFQ. In males, there was no significant endpoint

difference between PGB or VXR and placebo in the CSFQ.

In females, there was an isolated significant worsening in

LS mean CSFQ change scores at week 4 on VXR compared

with placebo ( 0.34 0.14 vs. + 0.14 0.12; P= 0.006).

Females treated with PGB showed a modest but significant

improvement as compared with VXR in LS mean CSFQ

change scores at last observation carried forward endpoint

(+0.500.14 vs. +0.060.14; P= 0.015).

The proportion of patients showing a clinically significant

(Z 7%) increase in body weight at endpoint was similar

on PGB (N= 2), VXR (N= 1), and placebo (N=1). No

clinically significant, treatment-emergent changes were

observed in vital signs or in laboratory parameters.

DiscussionThis randomized, placebo-controlled trial evaluated the

efficacy of flexible doses of PGB and VXR in patients

with GAD. The doses achieved in the trial were within

Fig. 4

4

2

0

Week1

Week2

Week3

Week4

Week6

Week8

LOCF

Placebo comparison:

P< 0.05; P< 0.01

Day4

10

8

6

Pregabalin (N=121)

Venla-XR (N=125)

Placebo (N=128)

LSmeansomaticanxietycha

ngescore

Mean change in Hamilton Anxiety Rating Scale somatic anxiety factor score by treatment week and at endpoint [last observation carried forward(LOCF)]. Day to week 8 results are based on a mixed-effects model repeated measures (MMRM) analysis LOCF-endpoint results based on analysisof convariance, LS, least square.



8/10

the recommended therapeutic dosing range of both PGB

and VXR. Treatment with PGB resulted in significantly

greater improvement compared with placebo in the

primary outcome, the HAM-A total score. Onset of

significant efficacy versus placebo occurred at the first

post-baseline assessment, by day 4, and was maintainedthroughout the 8 weeks of study treatment. Similarly,

significant levels of improvement on PGB occurred in

both psychic and somatic symptoms of anxiety, with the

onset of improvement observed by day 4 on both HAM-A

subscales.

To our knowledge, this is the first study to show

significant efficacy within the first 4 days of treatment

for a nonbenzodiazepine in patients with GAD. The early

onset of improvement was clinically significant, with

36% of patients reporting a 20% or greater reduction in

HAM-A severity on day 4. Earlier (before day 4) efficacy

assessment was not carried out. However, PGB hasshowed rapid onset of anxiolytic effect (within 34 h) in

a dental model of anxiety symptoms (Nutt et al., 2008).

As GAD is often comorbid with symptoms of depression,it is important to evaluate the influence of depressive

symptoms on the efficacy of PGB. Treatment with PGB

was associated with significant improvement in secondary

depressive symptoms as measured by the HAM-D. PGB

also showed significant efficacy in treating symptoms

of insomnia as measured by the MOS-Sleep factors. In

contrast, venlafaxine did not show significant efficacy

when compared with placebo on either the HAM-Dor MOS-Sleep factors. Improvement on PGB in subsyn-

dromic depressive symptoms, using the Bech Melancholia

Scale, has been reported in a pooled analysis of earlier

GAD studies (Pollacket al., 2003). However, it should be

noted that baseline HAM-D scores were relatively low

(11.5) in this study, and a comorbid diagnosis of MDD or

dysthymic disorder was reason for exclusion. No clinical

trial data are available on the efficacy of PGB as

a monotherapy in patients with GAD that is comorbid

with MDD or dysthymic disorder. The only published

data we are aware of that have evaluated the efficacy

selective serotonin reuptake inhibitors or serotonin

norepinephrine reuptake inhibitor therapy in GAD

comorbid with MDD or dysthymia is a post-hoc subgroupanalysis that found significant anxiolytic and antidepres-

sant efficacy for VXR, but not fluoxetine (Silverstone andSalinas, 2001).

A battery of QoL and functioning scales was used in this

study to determine whether improvement in anxiety wasassociated with parallel improvement in patient-centered

outcomes. The results were disappointing, perhaps

because of several reasons. First, the measures used in

this study may not have been sensitive to change in

outpatients diagnosed with GAD. For example, the

EuroQoL EQ-5D has not been validated for use in

psychiatric disorders such as GAD and, in fact, contains

items of questionable relevance in this population (e.g.

I have some problems walking about; I am confined to

bed; and I have some problems washing and dressing

myself). Second, the apparent lack of treatment effect

on QoL and functional outcomes may be attributable tothe fact that at least half of the patients seem to have had

minimal anxiety-related impairment in QoL or function-

ing at baseline. Study entry criteria did not require

impairment in QoL or functioning as inclusion criteria,

nor was the study powered to detect treatment effects

on QoL or functional measures. Finally, it is possible

that improvement in QoL and functioning lags behindimprovement in anxiety symptomatology, and that

a longer treatment period would be required to showefficacy in these outcome domains.

The interpretation of the results of this study is

complicated by the relatively high placebo response.High placebo response, and inability to show significance,

has been reported to occur, even among drugs of provenefficacy in GAD, in approximately 50% of clinical trials

(Khan et al., 2002, 2005). In 8-week trials evaluating

the efficacy of VXR in GAD, the endpoint HAM-A

change score on placebo ranged from 8.1 to 11.7, with

three out of seven trials having a high placebo response

( > 10-point improvement in HAM-A; [45] Thase, 2006).

In this study, the endpoint improvement in the HAM-A

total score on placebo was 11.7, identical to the HAM-A

change score in the largest previous VXR trial with an

active comparator control (diazepam; Hackett et al.,2003). In the previous study, [16]Hackett and colleagues

(2003) reported an endpoint improvement on the 150 mg

dose of VXR that was similar to what was observed in the

current trial ( 12.8 vs. 12.0). In both studies, the VXR

treatment groups were not significant versus placebo,

whereas diazepam (in the study by Hackett and

colleagues, 2003) and PGB (in this study) did show

significant efficacy versus placebo. In contrast, an earlier

placebo-controlled trial of PGB, conducted in European

sites similar to this study, did show significant efficacy

versus placebo for a lower (75 mg) fixed dose of

venlafaxine (Montgomeryet al., 2006).

Table 3 Incidence (%) of treatment-emergent adverse events(all-causality, with incidence 5%)

PregabalinN=121

Venlafaxine-XRN=125

PlaceboN=128

Nausea 12.4 25.6 8.6Dizziness 20.7 9.6 6.3Headache 17.4 16.0 11.7Dry mouth 10.7 12.0 3.9

Fatigue 9.9 12.8 3.9Somnolence 9.1 4.8 2.3Insomnia 4.1 9.6 4.7Vertigo 13.2 8.0 3.1Constipation 4.1 5.6 3.1Hyperhidrosis 2.5 8.0 5.5Any severe adverse event 9.1 20.0 7.8



9/10

Various factors have been suggested (Benedetti, 2008;

Price et al., 2008) that may increase the likelihood ofa high placebo response during a clinical trial, including

illness characteristics (shorter duration; low severity;low comorbidity); study design characteristics (higher

number of active treatment arms; too many outcomemeasures, especially as the latter may increase the time

of doctorpatient contact resulting in an increase innonspecific psychotherapy effects); various longitudinal

effects (spontaneous remission; regression to the mean);

and patient variables (expectancy effects). The specific

reasons for the high placebo response observed in this

study are uncertain, but we suspect that expectancy

effects may have played a role. In particular, the use of a

day 4 assessment may have created the impression that

the treatment under study might be powerful and rapid

in onset. Clinical trials in central nervous system

disorders are especially susceptible to this type of subject

and observer bias (Benedetti, 2008; Price et al., 2008).

Treatment with both PGB and VXR were generally

well tolerated. Treatment with PGB was associated with

higher rates of dizziness, whereas treatment with VXR

was associated with higher rates of nausea. Overall, the

incidence of adverse events rated as severe was

significantly higher on VXR than with PGB. No clinically

significant, treatment-emergent changes were observed

on either drug in vital signs or in laboratory parameters.

In conclusion, this study provides further confirmation of

the efficacy of PGB in the treatment of GAD. The onset

of significant antianxiety effect for PGB occurred by day4, with rapid improvement occurring in both psychic and

somatic symptoms of anxiety.

AcknowledgementsThe study was funded by Pfizer Inc. Paid editorial

support was provided by Edward Schweizer, MD, andfunded by Pfizer Inc. The authors acknowledge the

contribution of the many individual investigators for

their participation in this trial. Dr Kasper has received

grants/research support, consulting fees, and honoraria within

the last 3 years from AstraZeneca, Bristol-Myers Squibb,

CSC, Eli Lilly, GlaxoSmithKline, Janssen Pharmaceutical,Lundbeck, MSD, Novartis, Organon, Pierre Fabre, Pfizer,

Schwabe, Sepracor, Servier, Wyeth. Dr Herman is a full-time

employee of Pfizer Inc. Dr Nivoli has no disclosures to

declare. Dr Van Ameringen has received grant/research

support form AstraZeneca, Cephalon, GlaxoSmithKline,

Janssen-Ortho Inc., National Institute of Health, Novartis,

Pfizer, and Wyeth-Ayerst; has served as a consultant for

Biovail, Cephalon, GlaxoSmithKline, Janssen-Ortho Inc.,

Novartis, Pfizer, and Wyeth-Ayerst; and has served on the

speakers bureaus of GlaxoSmithKline, Janssen-Ortho Inc.,

Pfizer, and Wyeth-Ayerst. Dr Petralia has no disclosures to

declare Dr Mandel is a full-time employee of Pfizer Inc.

Dr Baldinetti was a full-time employee of Pfizer Inc. at the

time this study was conducted, and at the time the initialdraft manuscript was prepared. She currently is affiliated

with IRCCS S. Lucia Foundation-Neurology Department,University of Rome, Rome, Italy. Dr Bandelow has received

consulting fees and honoraria within the last 3 years fromAstraZeneca, Bristol-Myers-Squibb, Cephalon, Dainippon-

Sumitomo, Glaxo, Janssen, Jazz, Lilly, Lundbeck, Pfizer,Roche, Servier, Solvay, and Wyeth. ClinicalTrials.gov Identi-

fier: NCT00151450. Funded by Pfizer Inc.

ReferencesAllgulander C, Hackett D, Salinas E (2001). Venlafaxine extended release (ER)

in the treatment of generalised anxiety disorder: twenty-four-week placebo-controlled dose-ranging study. Br J Psychiatry179:1522.

Allgulander C, Nutt D, Detke M, Erickson J, Spann M, Walker D, et al. (2008).A non-inferiority comparison of duloxetine and venlafaxine in the treatment of adultpatients with generalized anxiety disorder. J Psychopharmacol22:417425.

Bandelow B, Zohar J, Hollander E, Kasper S, Moller HJ; World Federation ofSocieties of Biological Psychiatry (WFSB) (2008). Guidelines for thepharmacological treatment of anxiety, obsessive-compulsive and post-traumaticstress disordersFirst Revision. World J Biol Psychiatry9:248312.

Bech P (2007). Dose-response relationship of pregabalin in patients withgeneralized anxiety disorder. A pooled analysis of four placebo-controlledtrials.Pharmacopsychiatry40:163168.

Benedetti F (2008). Mechanisms of placebo and placebo-related effects acrossdiseases and treatments. Annu Rev Pharmacol Toxicol48:3360.

Brooks R (1996). EuroQol: the current state of play. Health Policy37:5372.Clayton AH, McGarvey EL, Clavet GJ (1997). The Changes in Sexual Function-

ing Questionnaire (CSFQ): development, reliability, and validity. Psycho-pharmacol Bull33:731745.

Crofford LJ, Rowbotham MC, Mease PJ, Russell IJ, Dworkin RH, Corbin AE, et al.(2005). Pregabalin for the treatment of fibromyalgia syndrome: results ofa randomized, double-blind, placebo-controlled trial. Arthritis Rheum52:12641273.

Davidson JR, DuPont RL, Hedges D, Haskins JT (1999). Efficacy, safety, andtolerability of venlafaxine extended release and buspirone in outpatients withgeneralized anxiety disorder. J Clin Psychiatry60:528535.

Dooley DJ, Donovan CM, Meder WP, Whetzel SZ (2002). Preferential action ofgabapentin and pregabalin at P/Q-type voltage-sensitive calcium channels:inhibition of K + -evoked [3H]-norepinephrine release from rat neocorticalslices.Synapse 45:171190.

Dworkin RH, Turk DC, Farrar JT, Haythornthwaite JA, Jensen MP, Katz NP, et al.(2005). Core outcome measures for chronic pain clinical trials: IMMPACTrecommendations.Pain 113:919.

Endicott J, Nee J, Harrison W, Blumenthal R (1993). Quality of Life Enjoymentand Satisfaction Questionnaire: a new measure. Psychopharmacol Bull29:321326.

Feltner DE, Crockatt JG, Dubovsky SJ, Cohn CK, Shrivastava RK, Targum SD,et al. (2003). A randomized, double-blind, placebo-controlled, fixed-dose,multicenter study of pregabalin in patients with generalized anxiety disorder.J Clin Psychopharmacol23:240249.

Fink K, Dooley DJ, Meder WP, Suman-Chauhan N, Duffy S, Clusmann H, et al.(2002). Inhibition of neuronal Ca(2 + ) influx by gabapentin and pregabalinin the human neocortex. Neuropharmacology42:229236.

Gelenberg AJ, Lydiard RB, Rudolph RL, Aguiar L, Haskins JT, Salinas E (2000).Efficacy of venlafaxine extended-release capsules in nondepressedoutpatients with generalized anxiety disorder: a 6-month randomizedcontrolled trial. JAMA283 :30823088.

Hackett D, Haudiquet V, Salinas E (2003). A method for controlling for a highplacebo response rate in a comparison of venlafaxine XR and diazepamin the short-term treatment of patients with generalized anxiety disorder.Eur Psychiatry18 :182187.

Hamandi K, Sander JW (2006). Pregabalin: a new antiepileptic drug for refractoryepilepsy.Seizure15 :7378.

Hamilton M (1959). The assessment of anxiety states by rating. Br J Med Psychol32:5055.

Hamilton M (1960). A rating scale for depression.J Neurol Neurosurg Psychiatry23:5662.

Hartford J, Kornstein S, Liebowitz M, Pigott T, Russell J, Detke M, et al. (2007).Duloxetine as an SNRI treatment for generalized anxiety disorder: results froma placebo and active-controlled trial. Int Clin Psychopharmacol22:167174.



10/10

Hays RD, Martin SA, Sesti AM, Spritzer KL (2005). Psychometric properties ofthe medical outcomes study sleep measure. Sleep Med6:4144.

Guy W (1996). ECDEU Assessment Manual for Psychopharmacology.Washington, DC: National Institute of Mental Health-US Dept of Health,Education, and Welfare publication (ADM); pp. 276338.

Kessler RC, DuPont RL, Berglund P, Wittchen HU (1999). Impairment in pureand comorbid generalized anxiety disorder and major depression at 12months in two national surveys. Am J Psychiatry156 :19151923.

Kessler RC, Brandenburg N, Lane M, Roy-Byrne P, Stang PD, Stein DJ, et al.

(2005). Rethinking the duration requirement for generalized anxiety disorder:evidence from the National Comorbidity Survey Replication. Psychol Med35:10731082.

Khan A, Khan S, Brown WA (2002). Are placebo controls necessary to test newantidepressants and anxiolytics? Int J Neuropsychopharmacol5:193197.

Khan A, Kolts RL, Rapaport MH, Krishnan KR, Brodhead AE, Browns WA (2005).Magnitude of placebo response and drug-placebo differences acrosspsychiatric disorders. Psychol Med35:743749.

Lenox-Smith AJ, Reynolds A (2003). A double-blind, randomised, placebocontrolled study of venlafaxine XL in patients with generalised anxietydisorder in primary care. Br J Gen Pract53:772777.

Lieb R, Becker E, Altamura C (2005). The epidemiology of generalized anxietydisorder in Europe. Eur Neuropsychopharmacol15 :445452.

Montgomery SA (2006). Pregabalin for the treatment of generalised anxietydisorder. Expert Opin Pharmacother7:21392154.

Montgomery SA, Tobias K, Zornberg GL, Kasper S, Pande AC (2006). Efficacyand safety of pregabalin in the treatment of generalized anxiety disorder:

a 6-week, multicenter, randomized, double-blind, placebo-controlledcomparison of pregabalin and venlafaxine. J Clin Psychiatry67:771782.

Montgomery S, Chatamra K, Pauer L, Whalen E, Baldinetti F (2008). Efficacyand safety of pregabalin in elderly people with generalised anxiety disorder.Br J Psychiatry193 :389394.

Nicolini H, Bakish D, Duenas H, Spann M, Erickson J, Hallberg C, et al. (2008).Improvement of psychic and somatic symptoms in adult patientswith generalized anxiety disorder: examination from a duloxetine,venlafaxine extended-release and placebo-controlled trial. Psychol Medhttp://journals.cambridge.org/action/displayAbstract?fromPage=online&aid

= 1880804

Nutt D, Mandel F, Baldinetti F (2008). Early onset anxiolytic efficacy after asingle dose of pregabalin: double-blind, placebo- and active-comparatorcontrolled evaluation using a dental anxiety model. J Psychopharmacolhttp://jop.sagepub.com/cgi/rapidpdf/0269881108094722v1

Pande AC, Crockatt JG, Feltner DE, Janney CA, Smith WT, Weisler R, et al.(2003). Pregabalin in generalized anxiety disorder: a placebo-controlled trial.Am J Psychiatry160 :533540.

Pohl RB, Feltner DE, Fieve RR, Pande AC (2005). Efficacy of pregabalin in thetreatment of generalized anxiety disorder: double-blind, placebo-controlledcomparison of BID versus TID dosing.J Clin Psychopharmacol25:151158.

Pollack MH, Zimbroff DZ, Tobias K, Zornberg GL (2003), Pregabalin ingeneralized anxiety disorder: analyses of subsyndromic depression. posterpresentation, 156th annual meeting of the American psychiatric Association

May 1722, San Francisco.Price DD, Finniss DG, Benedetti F (2008). A comprehensive review of the

placebo effect: recent advances and current thought. Annu Rev Psychol59:565590.

Rickels K, Pollack MH, Sheehan DV, Haskins JT (2000). Efficacy of extended-release venlafaxine in nondepressed outpatients with generalized anxietydisorder.Am J Psychiatry157:968974.

Rickels K, Pollack MH, Feltner DE, Lydiard RB, Zimbroff DL, Bielski RJ, et al.(2005). Pregabalin for treatment of generalized anxiety disorder: a 4-week,multicenter, double-blind, placebo-controlled trial of pregabalin andalprazolam. Arch Gen Psychiatry62:10221030.

Sheehan DV, Harnett-Sheehan K, Raj BA (1996). The measurement of disability.Int Clin Psychopharmacol11 (Suppl 3):8995.

Sheehan DV, Lecrubier Y, Harnett-Sheehan K, Janavs J, Weiller E, Bonara I,et al.(1997). Reliability and validity of the MINI international neurospychiatricinterview (M.I.N.I): according to the SCID-P. Eur Psychiatry12:232241.

Silverstone PH, Salinas E (2001). Efficacy of venlafaxine extended release in

patients with major depressive disorder and comorbid generalized anxietydisorder.J Clin Psychiatry62:523529.

Sonnett TE, Setter SM, Campbell RK (2006). Pregabalin for the treatment ofpainful neuropathy. Expert Rev Neurother6:16291635.

Taylor CP, Angelotti T, Fauman E (2007). Pharmacology and mechanism of actionof pregabalin: the calcium channel alpha2-delta (alpha2-delta) subunit asa target for antiepileptic drug discovery. Epilepsy Res 73:137150.

Thase ME (2006). Treatment of anxiety disorders with venlafaxine XR. Expert RevNeurother6:269282.

Wittchen HU (2002). Generalized anxiety disorder: prevalence, burden, and costto society. Depress Anxiety16 :162171.

Wyeth Venlafaxine-XR US prescribing information. Accessed on the web on Nov 12,2007 at: www.wyeth.com/content/ShowLabeling.asp?id= 100

Zigmond A, Snaith RP (1983). The hospital anxiety and depression scale.Acta Psychiatr Scand67:361370.


Pregabalin vs Venlafaxine

Documents

Transcript of Pregabalin vs Venlafaxine