found ofAtidy Disorders. Vol. 4. pp. 239-246. 1990 0887-6185!90 53.00 + .

cQ
Primd in the USA. All righu racrve.3. Cqyight 0 1990 Pergamcm Press plc

CLINICAL REPORTS

Adinazolam Sustained Release

Formulation in the Treatment of
Generalized Anxiety Disorder

DAVJD V. SHEEHAN, M.D., ASHOK B. RAJ, M.D., K. HARNETT-SHEEHAN, PH.D.,

AND SONIASOTO M.D.

Department of Psychiatry and Behavioral Medicine. University of South Florida College of

Medicine Tampa. Florida

CARL P. LEWIS, M.D., PH.D.

Clinical Development The Upjohn Company, Kalamazoo. Michigan

Abstract - A 6 week open label trial, followed by a l-6 week taper/discontinuation

phase, was conducted to assess the safety and efficacy of of the sustained release for-
mulation of adinazolam in 7 outpatients with generalized anxiety disorder.
Adinazolam-SR was well tolerated and produced a rapid and significant improve-
ment in the severity of anxiety, phobias, disability, and depressive symptoms. The
duration of therapeutic action of this formulation of adinazolam was 12.2 f 2.6 hours.
Relapse was common, using the studys speed of withdrawal. However, rebound
occurred in only one patient and withdrawal symptoms were few.

INTRODUCTION
Several benzodiazepines widely used in the treatment of generalized anxi-
ety disorder have a short duration of therapeutic action although their half

This study was supported in part by a grant from the Upjohn Pharmaceutical Company. The results
are based on the authors analysis of the data.
Correspondence should be addressed to D.V. Sheehan M.D., Professor of Psychiatry and Director of
Research, Department of Psychiatry, University of South Florida College of Medicine, 3515 E.
Fletcher, Ste 321, Tampa, Florida 33612,
This work was supported, in part, by a grant-in-aid provided by the Upjohn Company CNS
Research. We would like to thank Mrs. Roxy Baker, Robert Ftdford, Jeffrey Freeman for their
assistance in the organization and preparation of this study.

239
240 D.V.SHEEHANETAI_.

lives are intermediate or long. Alprazolam, for example, has a duration of

therapeutic action of only 4 to 6 hours although its half life is approximately
12 hours. The half life of a benzodiazepine is not as useful a guide to its
duration of therapeutic action as its rate of distribution or lipophilicity
(Goldberg & Finnerty,l979; Rickels, Weisman, Norstad et al. 1982).
Diazepam, for example, has a longer half life than lorazepam, but lorazepam
has a longer duration of therapeutic action than diazepam. Lorazepam is
more lipophilic and has a slower rate of distribution than diazepam.
Some benzodiazepines need to be taken several times a day to maintain ade-
quate control of anxiety symptoms. Benzodiazepines with a short duration of
action are associated with a subjective roller coaster experience (peaks of mild
sedation followed by troughs of mild anxiety symptoms). Frequent recurrence
of symptoms, every few hours, prompts excessive concern about timing of
each next dose, a phenomenon referred to as symptom driven clock watch-
ing. This phenomenon was already apparent in 1982 following our first
alprazolam panic disorder study (Sheehan et al., 1984). To correct this prob-
lem work was begun in 1983 to develop sustained release formulations of
alprazolam and an alprazolam analog, adinazolam.
A triazolo benzodiazepine, adinazolam is a recent dimethylamino derivative
of alprazolam. In its original compressed tablet formulation (adinazolam-
CT), it is reported to have anxiolytic (Pyke & Greenberg,l989) and antide-
pressant effects (Pyke, Cohn, Feighner & Smith,1983; Feighner, 1985;
Lahti, Sethy, Barsuhn & Hester, 1983; Lahti, Pyke, Sethy, 1983; Amsterdam,
Kaplan, Potter, Bloom & Rickels, 1986; DuMer, Myers, Khan, Avery, Ishiki,
& Pyke, 1987). However, adinazolam-CT has a very short half life and a very
short duration of therapeutic action (approximately 3 hours). Since patients
require frequent dosing, often several times a day, adinazolam-CT is often
impractical for the average case of generalized anxiety disorder.
The study reported here is the first published report using the new sustained
release formulation of adinazolam (adinazolam-SR) in the treatment of gener-
alized anxiety disorder. The study was designed to collect information which
would help in planning and implementing double blind placebo controlled
studies with this compound.

METHODS

Design
An open label, nonparallel controlled, flexible dose, pilot study was carried
out to investigate the efficacy, safety, duration of therapeutic action, and
dosage regimen of adinazolam-SR in seven outpatients with a diagnosis of
Generalized Anxiety Disorder. Patients were free of all psychotropic medica-
tion for one week prior to baseline and were then treated with adinazolam-SR
for 6 weeks.
15 mg tablets of adinazolam-SR were dispensed at each visit to be taken
once or twice daily by each patient. Medication was initiated at one 15 mg.
tablet per day and gradually increased on a flexible dose schedule. The maxi-
 TREATMENT OF GENERALIZED ANXIETY DISORDER 241

mum dose permitted was 8 tablets (120 mg adinazolam) per day. After 6
weeks of treatment, adinazolam-SR was tapered slowly over a 1 to 6 week
period. The withdrawal schedule varied from a maximum decrement of 50%
per week when the dose was above 2 tablets per day followed by a decrease of
a l/2 tablet every 4 days at lower doses, to a minimum of a l/2 tablet per
week.
All patients gave informed consent to this protocol which was approved by
the institutional review board of the study site.

Entry Criteria
Male and female patients aged 18 to 65 who met DSM-III-R criteria for
generalized anxiety disorder, using a revision of SCID-UP-R, the Upjohn ver-
sion of the Structured Clinical Interview for DSM-III-R (Spitzer & Williams,
1987), were eligible. Patients were excluded if they were pregnant or lactat-
ing, had a history of seizures, had acute suicidal ideation or an unstable medi-
cal condition as measured by physical examination, vital signs and clinical
laboratory determinations. Patients were also excluded if they had concurrent
evidence of a DSM-III-R diagnosis of major depressive disorder, concurrent
evidence or a recent (within 6 months) diagnosis of alcohol or other substance
abuse/dependence, obsessive/compulsive disorder, hypomania or paranoia, or
if they had concurrent evidence or a recent (within 2 years) DSMIII-R diagno-
sis of psychosis, mania, dementia or cyclothymia.

Clinical Measures and Evaluations

Patients were assessed weekly by the research psychiatrists during treat-
ment, taper, and for two post discontinuation visits. The following instruments
were used: the Sheehan Clinician Rated Anxiety Scale (Sheehan, 1986); the
Hamilton Anxiety Scale (Hamilton, 1959); the Hopkins Symptom Checklist
(SCI-90-R) devbeloped by Derogatis, Lipman, Rickels, Uhlenhuth & Covi
(1974); the Hamilton Depression Scale (Hamilton, 1967); the Beck
Depression Inventory, 21 item, (Beck, 1961); an 11 point Disability Scale,
Phobia Scale, 21 point Clinician Rated Global Improvement Scale (rated
from -10 to +lO), and 21 point Patient Rated Global Improvement Scale (rated
from -10to +lO, all developed by Sheehan (1986). Safety assessments (physi-
cal examinations, vital signs, blood chemistries [SMA 251, complete blood
count, urinalysis, electrocardiogram) were performed as indicated during the
study. Potential side effects were monitored weekly using the Medical Event
Checklist developed by Lewis (1985).

RESULTS

Short Term Treatment

Seven patients (five women and two men), who met DSM-III-R criteria for
generalized anxiety disorder but not panic disorder (using the SCID-UP), were
242 D.V.SHEEHANETAL.

given medication and completed 6 weeks of treatment. Their mean fS.D. age
was 37.7 f 9.5 years. Mean duration of illness was 12.3 f 8.3 years. Marital
status included 3 married, 3 divorced and 1 separated. Six were employed.
Four described their current condition as a recurrence, 2 as a worsening and 1
as a continuation of a longstanding event.

Dosage and Duration of Therapeutic Action

At the end of the acute phase (week 6) patients were taking an average of
6.4 +1.8 tablets (95.8 + 28.0 mgs adinazolam) per day, range, 2-8 tablets (30
to 120 mgs adinazolam) per day. Most required twice daily dosing, with 2/3
of the dose taken in the morning and l/3 taken in the evening.
The mean + S.D. duration of therapeutic action reported by these patients
was 12.2 * 2.6 hours, range 9-16 hours.

Outcome

Effectiveness of treatment was evaluated using repeated measures analysis

of variance of the outcome variables at baseline and treatment weeks 1
through 6. Post hoc comparisons of the weekly scores were performed using
Fishers PLSD test with the significance set at p=<.Ol to determine the point at
which significant improvement over baseline occurred and to determine when
maximum therapeutic effectiveness was reached.
In the interests of brevity, Table 1 gives only the mean + S.D. scores on the
outcome variables for baseline, week 1, midpoint (week 3), endpoint (week
6). This table demonstrates significant improvement over baseline on all of
the anxiety scales beginning in the first week and persisting through the acute
phase of the study. Significant improvement was also early and persistent on
the clinician and patient rated global scales, the Hamilton Depression Scale,
the SCl-90 depression subscale, and the fear component of four target phobias.
Maximum effectiveness was usually reached in the first week for anxiety, pho-
bic, and depressive symptoms, in the second week for disability, and in the
third week for global improvement ratings. Improvement was not statistically
significant using repeated measures anova on the SCL-90 phobia subscale, the
avoidance component of the four target phobias, overall phobia severity rat-
ing, the Beck Depression Scale, or the work component of the disability scale.
Four patients experienced a 50% improvement on anxiety symptom mea-
sures (Hamilton Anxiety Scale, Sheehan Clinician Rated Anxiety Scale and
anxiety subscale of the SCL-90-R) beginning in week 1 or 2 and persisting for
the most part through the 6 week trial. Three had a 50% improvement in the
phobic anxiety subscale of the SCL-90-R by week I and five had a similar
improvement by week 6. Four showed marked and persistent improvement (a
score of 27) on the clinician rated global improvement scale at the end of
week 1 of treatment and persisting throughout the trial.
Hamilton, discussing his own long experience with his scale, has said that a
good anxiolytic drug should reduce the Hamilton Anxiety Scale score below
10. If the endpoint Hamilton Anxiety score is higher than 12 the clinical ben-
 TABLE 1
OUTCOME SCORISIOR UASI!I.INI!. WIXKS 1,3 AND 6 or: 7 PATIIVVISwrn~ GI!NI!RAI.VXII ANXIETY DISOHO~?R ADINM\ZO~,AM-SR
wl~o RE:(.~~vI:II
Weeks First Week Las1 Taper Weeks Firs1 Week Lssl Taper
overall Significantly of Maximum Mean SD
Bareline Week 1 Week 3 Week 6 Significance Different TVXUllClll
MCXUl SD MetXl SD Mean SD MCZUI SD F df P from BawlincA EffCClM
AllXlCly
Hamihon Anxiety 22.3 4.9 11.7 2.7 12.3 7.1 9.3 5.8 5.1 6 0.0007 1.2.3,4.5.6 1 13.7 5.3
SCRAS 31.6 11.7 18.3 4.5 15.9 10.5 13.4 8.7 5.7 6 0.@303 1 .2 .3 9
4 35 .6 1 22.5 13.4
SCL-90 Anxiety 17.4 10.4 11.0 9.6 8.3 8.5 4.6 3.4 4.3 6 o.cQ21 1.2.3.4.5.6 2 6.8 3.6
XL-90 Somatizstion 17.3 5.8 9.6 3.7 6.6 3.8 6.0 4.4 6.4 6 0.0001 1.2.3.4.5.6 1 9.1 2.4
Phobic Anxiety
SCL-90 Phobic Anxiety 7.0 6.6 5.3 6.34 4.1 5.1 3.7 4.7 1.2 6 0.3294 3.8 4.5
4 Targel Fhobias/Fear 20.7 17.9 13.7 14.8 15.7 14.3 10.9 13.3 2.6 6 0.0353 1 14.5 14.3
4 Target Phobias/Avoidance 6.7 6.1 5.7 5.6 6.4 5.6 4.9 5.5 I.5 6 0.1929 6.8 6.1
Overall Phobia Severity 5.8 4.5 4.0 4.3 3.5 4.4 3.5 4.1 1.93 6 0.1092 4.3 4.1
Depression 3
Hamilton Depression 17.9 7.0 8.9 4.4 9.6 6.0 6.1 5.4 4.3 6 0.0024 1.2.3.4.5.6 1 10.8 5.2 pI
Beck Depression 16.1 12.8 13.1 l3.0 10.7 7.5 7.3 7.3 1.3 6 0.2764 6 9.8 6.9
SCL-90 Depression 19.7 12.6 Il.1 7.9 10.1 9.2 7.6 6.6 2.4 6 0.0437 1.2.3.4.5.6 I 7.8 4.8
Dl.sablllty
Work Disability 5.4 2.9 5.1 3.5 3.7 3.1 4.0 3.7 2.0 6 0.0900 3.7 2.0
Social Disability 5.7 2.9 4.7 2.8 3.3 2.7 2.9 2.0 2.3 6 0.0556 23.6 4.5 1.4
Family Disability 5.6 2.9 5.1 3.3 4.9 2.0 2.6 2.0 3.0 6 0.0173 26 3.3 I.0
Clinician Rating 4.1 0.9 2.5 0.5 2.4 1.0 1.7 1.3 5.2 6 0.0007 I v2.3.4.5.6
of Severity
Global Improvement
Clinician Rated 0 0 4.0 IA 6.0 2.6 7.4 2.3 15.4 6 O.cQOl 1.2.3.4.5.6 3.7 3.8
Pa&mt Rated 0 0 4.0 2.2 5.3 2.5 7.3 2.4 15.6 6 O.ooOl 1 .2 *3 1
4 .5 .6 3.8 3.9
n Weekssignificdy different fran baseline using Fishers PSLD with the significance set PI P=<.Ol.
nn The first pin1 st which lhe scorn was significsnlly below the baseline scan but nd significantly different fran any subsequent score using
Firhen PSLD with he significance set 81p=<.Ol
AM Means derived fmm last taper visit for 6 parients who cmcrcd taper. No significant dilfenncu found between mum II last
taper visit and baseline for thhese6 patimrs.
244 D. V. SHEEHAN ET AL.

efit is suspect (David Wheatley, personal communication, 1988). The mean

endpoint Hamilton Anxiety Scale score in this study was 9.9, reduced from
22.3 at baseline.

Side Effects
Treatment emergent symptoms and side effects, defined as symptoms which
were not present at baseline but emerged during treatment, or symptoms
which were present at baseline but became worse during treatment, were
mostly limited to sedation or drowsiness. Six patients experienced sedation to
a degree which was worse than baseline at some point during the trial.
However, sedation diminished as the trial progressed with only two patients
reporting it at week five and four at week 6. There were transient reports of an
emergence, or worsening compared to baseline, incoordination, paresthesia,
lightheadedness, panic attacks, depression. heightened sensory perception,
tremor, tachycardia, muscle cramps, nausea, diarrhea, constipation, nasal con-
gestion, appetite increase, and weight gain. Although each of these symptoms
was reported as worse than baseline by 3 or 4 patients at some point in the
study, none, with the exception of nasal congestion, was reported by more
than two patients at any week and most were confined to one or two patients a
week and to one or two weeks. Adinazolam SR was not found to produce any
abnormalities in CBC or SMA-25.

Taper1 Withdrawal
Since one patient was lost to follow-up after the acute phase, six were
entered into taper. Of these, one required the addition of another anxiolytic
after week 2, and one had to be switched to another anxiolytic at week 3 for
reasons unrelated to taper. The remaining four patients tapered to zero in 3
weeks or less.
Analysis of taper/withdrawal data included all six patients who entered
taper. However, only pure taper data, ie. data collected before the addition
or substitution of another anxiolytic, was used.

Relapse/Rebound
Relapse was defined as a loss of therapeutic gain, i.e. deterioration during
discontinuation period as compared with improvement during the therapeutic
trial (Pecknold, Swinson, Kuch, & Lewis, 1988). Individual patients were
judged to have relapsed if their scores at any point in the taper showed a dete-
rioration relative to their scores at week 6. Using these criteria, relapse was
common. Almost all patients who had taper data experienced relapse on some
of the study scales. Relapse was most frequent on the clinician and patient
rated global improvement scales with five of six patients showing relapse.
Relapse was also common on the Hamilton Anxiety Scale (4 patients) and
 TREATMENT OF GENERALIZED ANXlElY DISORDER 235

Sheehan Clinician Rated Anxiety Scale (four patients).

Rebound was defined as a worsening of the patients condition relative to
baseline. Individual patients were judged to have rebound if any score in the
taper phase showed a worsening relative to their scores at baseline. Using
these criteria, rebound was experienced by only one of the patients who
attempted taper. This patient had worse scores in taper on the Hamilton
Anxiety Scale, Sheehan Clinician Rated Anxiety Scale, and Beck Depression
Inventory. However, on none of the scales did the score worsen by 50% or
more. None of the other patients experienced rebound.
No significant differences were found on any of the scales when scores
from the last taper visit were compared to baseline scores, using paired t-tests.
However, Table 1 indicates that mean scores on almost all of the scales were
higher (better) at the last taper visit than at baseline.

Withdrawal Symptoms
Discontinuation emergent symptoms were defined as those symptoms
occurring at greater severity during the taper or post taper periods than previ-
ously at baseline or during treatment. In these patients, four events were each
reported as discontinuation emergent 4 times. They were paresthesias (2
patients), tinnitus (2 patients), tremor (1 patient) and dry mouth (1 patient).

DISCUSSION
The results of this study suggest that adinazolam-SR is highly effective in
controlling generalized anxiety disorder. It provided a rapid and statistically
significant improvement (time effect) in the severity of overall anxiety symp-
toms, in phobias, and in disability and depression scores in this group of
patients. Its duration of therapeutic action is considerably longer than that of
other benzodiazepines studied for DSM-III-R Generalized Anxiety Disorder.
It was well tolerated without any significant ill effects, clinical or biochemi-
cal. Relapse was common during the taper phase of the study, using this
studys speed of withdrawal. However, rebound occurred in only one patient
and symptoms of withdrawal were few.
The main advantage of adinazolam-SR over existing anxiolytics is its
longer duration of therapeutic action. Since adinazolam-SR requires only only
once or twice daily dosing, the typical roller-coaster phenomenon of other
benzodiazepines (peaks of side effects and valleys of symptom recurrence)
can be avoided or minimized.
While caution must be advised in interpreting these results until replicated
in double-blind placebo-controlled studies, the preliminary findings are
encouraging. Benzodiazepines with a longer duration of therapeutic action
(particularly sustained release formulations) have advantages clinically over
benzodiazepines with a shorter duration of action. They provide an effective
alternative for patients who have recurrent anxiety breakthroughs throughout
the day, those who forget to take their medications on schedule, and those
who dislike feeling dependent on frequent dosing.
246 D. V. SHEEHANET AL.

REFERENCES
American Psychiatric Association (1987). Diagnostic and StaWical Manual of Mental
Disorders. 3rd edition Revised (DSM-III-R). Washington, D.C.
Amsterdam, J.. Kaplan ,M.. Potter, L.. Bloom. L. & Rickels, K. (1986). Adinarolam. a new oi-
azolobenzodiarepine and imipramine in the treatment of major depressive disorder.
Springer Verlag.
Beck, A.T. Ward, C.H. Mendelson. M.. Mock, J.E., & Erbaugh, J.K. (1961). An inventory for
measuring depression. Archives of General Psychiatry. 4,561-571.
Derogatis, L.R.. Lipman, R.S., Rickels, K.. Uhlenhuth, E.H.. & Covi L: The Hopkins Symptom
Checklist (HSCL) (1974). A self report symptom inventory. Behavioral Science, 19. l-15.
Dunner, D., Myers, J., Khan, A., Avery, D., Ishiki, D., & Pyke. R. (1987). Adinazolam, a new
antidepressant: findings of a placebo-controlled double-blind studying outpatients with major
depression. Journal Clinical Psychopharmacology, 7. 170-172.
Feighner, J.P. (1985). Controlled studies of Deracyn@ (adinazolam mesylate) for major depressive
episode. Abstract presented at NCDEU Annual Meeting, Key Biscayne, Florida.
Goldberg, H.L. & Finnerty, R-l. (1979). The comparative efficacy of buspirone and diazepam in
the treatment of anxiety. American Journal of Psychiatry , 136, I184-1187.
Hamilton, M. (1959). The assessment of anxiety states by rating. British Journal of Medical
Psychology. 32.50.
Hamilton, M. (1967). Development of a rating scale for primary depressive illness. British
Journal of Social and Clinical Psychology, 6.278-296.
Lahti. R.A. Sethy, V.H., Barsuhn, C., & Hester, J.B. (1983). Pharmacological profile of the antide-
pressant adinazolam, a triazolobenzodiazepine. Neuropharmacology ,22,1277-1282.
Lahti, R.A., Pyke, R.E., & Sethy. V.H (1983). Adinazolam, a triazolobenzodiazepine antidepres-
sant: preclinical and clinical data. Abstract, International Union of Pharmacology, 9th
International Congress of Pharmacology, London.
Lewis, C. (1985). Medical Event Checklist. Kalamazoo: Upjohn Company.
Pecknold, J.C., Swinson, R.P., Kuch, K., & Lewis, C.P. (1988) Alprazolam in panic disorder and
agoraphobia: Results from a multicenter trial, III Discontinuation Effects. Archives of
General Psychiatry. 4.5,429-436.
Fyke. R.E., Greenberg, H.S. (1989) Double blind comparison of alprazolam and adinazolam for
panic and phobic disorders. Journal of Clinical Psychopharmacology ,9, 15-21.
Pkye, R.E., Cohn, J.B., Feighner, J.D., & Smith, W.T. (1983) Open label studies of adinazolam
in severe depression. Psychopharmacology Bullerin , 19,96-98.
Rickets, K., Weisman. K., Norstad, N., Singer, M ., Stoltz, D.. Brown, A., & Danton J. (1982).
Buspirone and diazepam in anxiety: a controlled study. Journal of Clinical Psychiatry ,
43.81-86.
Sheehan, D.V. (1986) The Anxiety Disease. New York: Bantam Books (revised paperback edition).
Sheehan, D.V., Greenblatt, DJ., Levine, PH., Schildraut, J.J., Watkins, D. (1984). Some biologi-
cal correlates of panic attacks with agoraphobia and their response to a new treatment.
Journal of Clinical Psychopharmacology, 4,66-75.
Spitzer, R.L. & Williams, J.B.W. (1987). Structured clinical interview for DSM-III-R, Upjohn
Version (SCID-UP-R). New York. Biometrics Research Department, New York State
Psychiatric Institute.