Efficacy of extended-release naltrexone in patients with relatively higher severity of alcohol dependence.
ABSTRACT Because some literature reviews have suggested that naltrexone's benefit may be limited to less-severe alcohol dependence, and exclusively to reduction in heavy drinking rather than abstinence, we examined the efficacy of once per month, injectable extended-release naltrexone (XR-NTX 380 mg) in patients with relatively higher severity alcohol dependence.
Post hoc analyses examined data from a multicenter, placebo-controlled, 24-week randomized trial of XR-NTX for alcohol dependence (N = 624). We analyzed treatment effects in alcohol-dependent patients who had higher baseline severity, as measured by: (i) the Alcohol Dependence Scale (ADS) or (ii) having been medically detoxified in the week before randomization. Efficacy was also examined via the relationship between pretreatment severity indices and reporting at least 4 days of lead-in abstinence prior to treatment-a major predictor of good outcome in the original study.
Higher severity alcohol-dependent patients, defined by the ADS, when receiving XR-NTX 380 mg (n = 50) compared with placebo (n = 47), had significantly fewer heavy-drinking days in-trial (hazard ratio=0.583; p = 0.0049) and showed an average reduction of 37.3% in heavy-drinking days compared with 27.4% for placebo-treated patients (p = 0.039). Among those who had a detoxification just prior to randomization, these reductions were 48.9% (XR-NTX 380 mg; n = 11) and 30.9% (placebo; n = 15) (p = 0.004). Subjects with at least 4 days of pretreatment abstinence (n = 82) versus those without (n = 542) had significantly higher pretreatment ADS scores (p = 0.002) and were more likely to require detoxification prior to randomization (p < 0.001). Patients with lead-in abstinence experienced significantly better maintenance of initial and 6-month abstinence.
These secondary analyses support the efficacy of XR-NTX 380 mg in relatively higher severity alcohol dependence for both reduction in heavy drinking and maintenance of abstinence, with implications for the role of adherence pharmacotherapy.
- SourceAvailable from: James C Garbutt[Show abstract] [Hide abstract]
ABSTRACT: The goal of this systematic review was to identify moderators of naltrexone efficacy in the treatment of alcohol dependence. We searched Pubmed, CINHAL, Embase, Psycinfo and the Cochrane Library from 1990 to April 2012 and reference lists of pertinent review articles, which yielded 622 trial, pooled analysis, and review articles. Using pre-established eligibility criteria, two reviewers independently determined whether abstracts contained evidence of demographic or biological characteristics, i.e. moderators, influencing naltrexone response in alcohol dependence. We assessed each publication for risk of bias and evaluated the strength of the body of evidence for each moderator. Twenty-eight publications (on 20 studies) met criteria for data synthesis. These included 26 publications from 12 randomized, placebo-controlled trials, 3 non-randomized, non-placebo studies, and 1 randomized, non-placebo study. In addition, there were 2 publications from pooled analyses of 4 randomized, placebo-controlled trials. Family history of alcohol problems and the Asn40Asp polymorphism of the μ-opioid receptor gene showed a positive association with efficacy in four out of five and three out of five studies respectively. Other moderators reported to be associated with efficacy included male sex (2/5 studies), pretreatment drinking (2/2 studies), and high craving (2/5 studies). However, the overall risk of bias in the published literature is high. The identification of naltrexone-responsive alcohol dependent patients is still in development. Studies to date point to two potential moderators-family history and presence of the OPRM1 Asn40Asp polymorphism-as having the strongest evidence. However, the data to date are still insufficient to recommend that any moderator be used in determining clinical treatment.Addiction 03/2014; · 4.58 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: Alcoholic hepatitis is an acute manifestation of alcoholic liver disease with mortality as high as 40-50 % in severe cases. Patients usually have a history of prolonged alcohol abuse with or without a known history of liver disease. Although there is significant range in severity at presentation, patients with severe alcoholic hepatitis typically present with anorexia, fatigue, fever, jaundice, and ascites. The use of either pentoxifylline or corticosteroids in those with severe disease (Maddrey's discriminate function >32) has significant mortality benefit. The addition of N-acetylcysteine to corticosteroids decreases the incidences of hepatorenal syndrome, infection, and short-term mortality, but does not appear to significantly affect 6-month mortality. Nutritional support with high-calorie, high-protein diet is recommended in all patients screening positive for malnutrition. Liver transplantation for a highly selected group of patients with severe alcoholic hepatitis may be an option in the future, but is not currently recommended or available at most transplant institutions.Digestive Diseases and Sciences 05/2014; · 2.26 Impact Factor
Efficacy of Extended-Release Naltrexone in Patients with
Relatively Higher Severity of Alcohol Dependence
Helen M. Pettinati, Bernard L. Silverman, John J. Battisti, Robert Forman,
Edward Schweizer, and David R. Gastfriend
Background: Because some literature reviews have suggested that naltrexone’s benefit may be
limited to less-severe alcohol dependence, and exclusively to reduction in heavy drinking rather
than abstinence, we examined the efficacy of once per month, injectable extended-release naltrex-
one (XR-NTX 380 mg) in patients with relatively higher severity alcohol dependence.
Methods: Post hoc analyses examined data from a multicenter, placebo-controlled, 24-week
randomized trial of XR-NTX for alcohol dependence (N = 624). We analyzed treatment effects
in alcohol-dependent patients who had higher baseline severity, as measured by: (i) the Alcohol
Dependence Scale (ADS) or (ii) having been medically detoxified in the week before randomiza-
tion. Efficacy was also examined via the relationship between pretreatment severity indices and
reporting at least 4 days of lead-in abstinence prior to treatment—a major predictor of good out-
come in the original study.
Results: Higher severity alcohol-dependent patients, defined by the ADS, when receiving XR-
NTX 380 mg (n = 50) compared with placebo (n = 47), had significantly fewer heavy-drinking
days in-trial (hazard ratio=0.583; p = 0.0049) and showed an average reduction of 37.3% in
heavy-drinking days compared with 27.4% for placebo-treated patients (p = 0.039). Among those
who had a detoxification just prior to randomization, these reductions were 48.9% (XR-NTX
380 mg; n = 11) and 30.9% (placebo; n = 15) (p = 0.004). Subjects with at least 4 days of pre-
treatment abstinence (n = 82) versus those without (n = 542) had significantly higher pretreat-
ment ADS scores (p = 0.002) and were more likely to require detoxification prior to
randomization (p < 0.001). Patients with lead-in abstinence experienced significantly better main-
tenance of initial and 6-month abstinence.
Conclusions: These secondary analyses support the efficacy of XR-NTX 380 mg in relatively
higher severity alcohol dependence for both reduction in heavy drinking and maintenance of
abstinence, with implications for the role of adherence pharmacotherapy.
Key Words: Extended-Release Naltrexone, Alcohol Dependence, Severity, Heavy Drinking,
population (Grant et al., 2004). While alcohol dependence at
any level of severity is a public health concern, individuals
with high-severity alcohol dependence are particularly at risk
for negative consequences from their alcohol consumption.
Compared with alcohol-dependent individuals at low severity
LCOHOL DEPENDENCE IS a common and debili-
tating disorder afflicting almost 4% of the adult U.S.
levels, those with relatively higher severity have worse symp-
toms of tolerance and withdrawal, abnormal values on
biomarkers of alcohol dependence, higher rates of crav-
ing, greater tobacco use, more psychiatric symptoms, and
impaired quality of life. Notably, they also are more likely to
endorse abstinence as their treatment goal (Berggren et al.,
2007; Donovan et al., 2006; Lima et al., 2005; Morgan et al.,
There are multiple aspects of the disorder of alcohol depen-
severity can be measured in terms of the quantity and fre-
quency of patterns of alcohol consumption (Goldstein et al.,
2006; Li et al., 2007). In the widely used Addiction Severity
Index (McLellan et al., 1992), severity is defined as the degree
of related social, legal, employment, medical, and psychologi-
cal problems that co-occur with alcohol use, as well as the fre-
quency and quantity of alcohol use per se. Symptom counts
have also been used to define severity of alcohol dependence
(e.g., Hicks et al., 2004; Rose et al., 2004; Saha et al., 2006;
Young et al., 2000). Instruments that have been used to mea-
sure severity of alcohol dependence, such as the Alcohol
Dependence Scale (ADS; Skinner and Allen, 1982) and the
From the Department of Psychiatry (HMP), University of Pennsyl-
vania School of Medicine, Philadelphia, Pennsylvania; Alkermes, Inc.
(BLS, RF, DRG), Waltham, Massachusetts; Neuroscience Program
(JJB), University of Colorado School of Medicine, Aurora, Colorado;
Department of Clinical Sciences (JJB), California Northstate College
of Pharmacy, Rancho Cordova, California; and Paladin Consulting
Group (ES), Hoboken, New Jersey.
Received for publication November 2, 2010; accepted February 4, 2011.
Reprint requests: Helen M. Pettinati, PhD, Department of Psychi-
atry, Treatment Research Division, Center for the Studies of Addic-
tion, University of Pennsylvania School of Medicine, Philadelphia,
PA; Tel.: 215-222-3200, Ext. 139; Fax: 215-386-5106; E-mail:
Copyright ? 2011 by the Research Society on Alcoholism.
Alcoholism: Clinical and Experimental Research
Vol. 35, No. 10
1804Alcohol Clin Exp Res, Vol 35, No 10, 2011: pp 1804–1811
Alcohol Use Disorders Identification Test (Saunders et al.,
1993), rely on both alcohol consumption and alcohol-related
In evaluating treatment options for individuals with alcohol
dependence, severity of the disorder is likely to be one of the
key considerations. Medical detoxification is often a clinical
recommendation for those with more severe dependence. One
aspect of severity, the presence of alcohol withdrawal symp-
toms, is a central element of the American Society of Addic-
tion Medicine (ASAM) Patient Placement Criteria that are
used to match patients with levels of care (Mee-Lee et al.,
The role of pharmacological treatments of alcohol depen-
dence within the subgroup of more severely dependent
patients is less clear. Available pharmacological treatment
studies have rarely reported the impact of severity of alcohol
dependence on outcome. There is some evidence, however,
that questions the efficacy of certain medications, particu-
larly oral naltrexone, among those with more severe alcohol
dependence. Oral naltrexone was not found to be signifi-
cantly superior to placebo in a large-scale randomized clini-
cal trial conducted with chronic, severe alcohol-dependent
veterans (Krystal et al., 2001). In another study comparing
oral naltrexone, acamprosate, and placebo, naltrexone was
found to be superior to acamprosate only within the sub-
sample of alcohol-dependent patients who had lower sever-
ity. There were no differences among these 3 treatments in
the high-severity patients (Morley et al., 2009). Furthermore,
a link between severity and medication adherence was
recently suggested by a study that reported that high-
severity alcohol dependence was associated with drinking dis-
tilled spirits (rather than wine or beer), which, in turn, was
related to lower adherence to combined psychosocial and
medication (oral naltrexone or topiramate) treatments for
alcohol dependence (Baltieri et al., 2009). Across 10 studies of
oral naltrexone for alcohol dependence, a meta-analytic
review found no significant efficacy in regard to abstinence
rates and stated that ‘‘naltrexone seems more…geared to con-
trolled consumption,’’ a conclusion, which suggests that nal-
trexone is more helpful to less severely dependent individuals
(Bouza et al., 2004, p. 825). The perspective that oral naltrex-
one is potentially ineffective in severely dependent patients
has been cited as one of the reasons that this medication is
not more widely used to treat alcohol dependence (Thomas
et al., 2003).
An extended-release formulation of naltrexone (XR-NTX,
VIVITROL?; Alkermes, Inc., Waltham, MA), approved by
the U.S. Food and Drug Administration (FDA) in April
2006, has been developed to address adherence issues with
oral naltrexone. Administered as an intramuscular monthly
injection, XR-NTX 380 mg has been shown to maintain con-
tinuous plasma levels of naltrexone for more than 30 days fol-
lowing a single injection (Dunbar et al., 2006). A 6-month,
randomized, multicenter, double-blind, placebo-controlled
study of XR-NTX 380 and 190 mg in combination with psy-
chosocial management found that, among the subgroup of
alcohol-dependent patients who achieved at least 4 days of
abstinence prior to initiating treatment, there were about 3
times as many patients treated with XR-NTX 380 mg as com-
pared with placebo who achieved 24 weeks of consecutive
abstinence (32 vs. 11%; p = 0.02) and had a 3-fold longer
median time to their first drink after starting treatment (41 vs.
12 days, p = 0.02) (Garbutt et al., 2005; O’Malley et al.,
2007). XR-NTX 380 mg was approved by the FDA with an
indication for alcohol-dependent patients who are able to
abstain from alcohol prior to treatment initiation.
While the Garbutt and colleagues (2005) report documents
the efficacy of XR-NTX 380 mg for most alcohol-dependent
patients, regardless of their severity of dependence, there are
persistent questions about the efficacy of naltrexone for more
severely dependent individuals. This report describes a sec-
ondary analysis of data from the Garbutt and colleagues
(2005) trial that examined the efficacy of XR-NTX 380 mg
for both heavy drinking and abstinence outcomes in patients
with relatively severe alcohol dependence with respect to the
impact of lead-in abstinence prior to treatment.
MATERIALS AND METHODS
Study Design and Patients
The current study is a post hoc analysis of alcohol consumption
data from a 6-month, double-blind, randomized controlled trial of
XR-NTX versus placebo injection, conducted at 24 U.S. public hos-
pitals, private and Veterans Administration clinics, and tertiary care
medical centers (details are given in the primary manuscript; Garbutt
et al., 2005).
The study recruited adult patients (men and women aged ‡ 18
years) with a diagnosis of alcohol dependence, who had at least 2 epi-
sodes per week of heavy drinking (‡5 drinks per day for men, ‡4 for
women) in the 30 days prior to enrollment. Excluded from the trial
were individuals who had any clinically significant medical condition
that might adversely affect safety or study participation; major
depression with suicidal ideation, psychosis, or bipolar disorder; or
dependence on benzodiazepines, opioids, or cocaine within the past
year. Patients with more than 7 days of inpatient treatment for sub-
stance abuse during the 30 days prior to screening were also
Patients were randomized to receive treatment with a single intra-
muscular injection of XR-NTX 380 mg, XR-NTX 190 mg, or pla-
cebo every 4 weeks for the 24-week treatment period. All the patients
also received 12 sessions of concurrent low-intensity psychosocial
counseling (Volpicelli et al., 2001).
Alcohol consumption was measured using the timeline follow-back
(TLFB) method (Sobell and Sobell, 1992) for the 30 days prior to
treatment, during treatment at weekly intervals for the first 4 weeks,
and then every 2 weeks for the next 20 weeks or at the final visit. The
TLFB assessment used calendars and recall of drinking on specific
days to record the frequency and amount of drinking since the previ-
ous assessment. This yielded a continuous record of daily drinking,
heavy-drinking days, and abstinence rates for the 24-week treatment
Pretreatment (baseline) TLFB data were used to identify patients
with ‡4 days of voluntary abstinence prior to treatment initiation.
Patients in the XR-NTX efficacy trial with at least 4 days of absti-
nence prior to treatment initiation showed a clinically meaningful
and robust treatment effect with XR-NTX 380 mg compared with
EFFICACY OF EXTENDED-RELEASE NALTREXONE
placebo (O’Malley et al., 2007). A 4-day lead-in abstinent period has
been a requirement in many other studies of pharmacotherapies for
alcohol dependence, including the large-scale COMBINE Study
(Anton et al., 2006). Four days is also the median length in the
United States for complete detoxification episodes (Drug and
Alcohol Services Information System, 2004).
Severity of alcohol dependence was measured in 2 ways: (i) the
ADS (Skinner and Horn, 1984) and (ii) the occurrence of detoxifica-
tion just prior to randomization to medication treatment.
The ADS is a widely used 25-item self-report scale that quantita-
tively measures alcohol dependence severity by focusing on alcohol
withdrawal symptoms, impaired control over drinking, awareness of
a compulsion to drink, increased tolerance to alcohol, and salience of
drink seeking behavior. A score ‡9 is highly predictive of DSM-IV
alcohol dependence (Ross et al., 1990). The ADS is used within the
ASAM Patient Placement Criteria for level of care planning for alco-
hol treatment (American Society of Addiction Medicine, 1996).
Because the ADS scale was added to the protocol after the trial had
begun, only about half of the patients in the trial received the ADS.
A median split on the ADS was used to identify patients with rela-
tively higher severity (ADS > 16). There were 97 patients (380 mg:
50, placebo: 47) classified as severely dependent on the ADS.
Medical detoxification in the week before randomization was uti-
lized as a reasonable measure of severity because medical detoxifica-
tion is typically instituted when there are signs of severe dependence,
e.g., symptoms of tolerance and withdrawal. Therefore, a clinician’s
decision to have a patient undergo medical detoxification would
patients (11 in the XR-NTX 380 mg group; 15 in the placebo group)
who wereclassifiedwithseverealcoholdependencebased onreceiving
a medical detoxification immediately prior to randomization.
The primary statistical analysis paralleled that conducted for the
main efficacy report (Garbutt et al., 2005) except the focus in this
investigation is on the XR-NTX 380-mg dose versus placebo because
the 380-mg dose is the clinically approved and available dose of XR-
NTX. Among the subgroup of more severely dependent patients,
XR-NTX 380 mg and placebo were compared on the occurrence of
heavy-drinking events during the 24-week treatment period. This
comparison was performed using a stratified generalized Andersen–
Gill recurrent-event Cox model (Andersen and Gill, 1982). To adjust
for the impact of patient discontinuation during the study on the
occurrence of heavy-drinking events, a pattern mixture-model
approach was implemented in the generalized Andersen–Gill recur-
rent-event Cox model (Hogan et al., 2004). Site was also included in
The analysis included all heavy-drinking events captured during
the treatment period, up to 30 days following the last injection (dose)
of study medication. All randomized subjects within the defined
severely dependent subgroup who received at least 1 dose of study
drug were included in this analysis. The primary descriptive statistic
from the Andersen–Gill analysis is a hazard ratio that compares the
failure rate, where failure here is the occurrence of heavy-drinking
events, for the XR-NTX 380 mg and placebo group. A hazard ratio
below 1 indicates a slower failure rate (i.e., a longer time to the occur-
rence of heavy-drinking events for XR-NTX 380 mg compared with
The Andersen–Gill analysis was conducted for those patients
defined as having high severity on the ADS. The sample of patients
who had undergone a detoxification prior to randomization was too
small to conduct the Andersen–Gill analysis with the pattern mixture
addition and site as a covariate. For the detoxification sample, and
for the severely dependent sample defined by the ADS, the percent-
age of heavy-drinking days in the XR-NTX 380 mg and placebo
groups were compared with the baseline. For this comparison, an
analysis of covariance was conducted specifying the contrast of the
XR-NTX 380 mg group versus placebo and using baseline percen-
tage heavy-drinking days as a covariate. The percentage of respond-
ers to treatment was compared for XR-NTX 380 mg versus placebo
using Fisher’s exact test. A clinical response was defined as £2 heavy-
drinking days in any 28-day period during 6-month treatment phase
(O’Malley et al., 2007). This definition of clinical response was
selected as representing substantial improvement from the minimum
of 8 days of heavy drinking in the month prior to randomization that
was required for enrollment. To present a complete understanding of
the impact of severity, responder rates in the low severity subgroups
(no detox; ADS £ 16) were also examined.
Finally, we used logistic regression to examine all 3 binary vari-
ables of higher versus lower severity as predictors of clinical response
to XR-NTX 380 mg: (i) ADS score >16 versus £16; (ii) detoxifica-
tion versus no detoxification; and (iii) lead-in abstinence versus no
Baseline Demographic and Clinical Characteristics of
Table 1 provides demographic and clinical information
on 3 patient subgroups derived from the total patient sam-
ple reported in the Garbutt and colleagues (2005) trial. Each
of the subgroups is further subdivided into patients who
received either XR-NTX 380 or placebo. The 3 patient sub-
groups listed in Table 1 are (i) those who were above versus
below the median ADS score of 16 for the sample, (ii) those
who received versus did not receive a medical detoxification
immediately prior to randomization, and (iii) patients who
had versus did not have at least 4 days of lead-in abstinence
prior to treatment initiation. Within each of the 3 subgroups
displayed in Table 1, there were no significant pretreatment
(baseline) differences on demographic and clinical character-
istics for the patients who had received XR-NTX 380 versus
For each definition of severity, there were no significant
demographic differences between the high and low severity
subgroups, but there were some notable clinical characteristic
differences (Table 1). Patients with higher (vs. lower) severe
alcohol dependence on the ADS were significantly more likely
to (i) have had a medical detoxification just prior to rando-
mization (p = 0.042), (ii) lead-in abstinence of ‡4 days (p =
0.038), and (iii) endorsed abstinence as their treatment goal
(p = 0.035). Those who had been detoxified just prior to
treatment (vs. those who were not) were significantly more
likely to endorse abstinence as their treatment goal (p =
0.004) and have ‡4 days of lead-in abstinence (p < 0.001).
Patients with lead-in abstinence of ‡4 days, compared with
those who did not have lead-in abstinence of ‡4 days, were
significantly more likely to have severe alcohol dependence, as
evidenced by significantly higher scores on the ADS (p =
0.002), and require detoxification prior to randomization
(p < 0.001).
A significant overlap between a medical detoxification
just prior to randomization and lead-in abstinence of ‡4 days
would be anticipated. However, the overlap was limited:
PETTINATI ET AL.
only 28% of the 82 patients with ‡4 days of lead-in absti-
nence had received a medical detoxification, and only about
half (54.8%) of those who had received a medical detoxifica-
tion also had ‡4 days of lead-in abstinence. Also, Table 1
depicts that those with (vs. without) lead-in abstinence of
‡4 days reported significantly (p < 0.001) fewer heavy-drink-
ing days in the past 30 days. However, we believe this is a
temporary artifact of detoxification and voluntary drinking
taper just prior to entering the study because when a
longer pretreatment time period is examined (90 days, as
seen in Fig. 1), the lead-in abstinence group displayed
higher levels of pretreatment drinking, with an average of
about 1.5 drinks per day more than the no lead-in abstinence
Table 1. Demographic and Clinical Characteristics: Baseline Data for XR-NTX 380 mg and Placebo by Potential Severity Indices
Alcohol Dependence Scale
(ADS) £ 16 ADS > 16No DetoxDetox
N = 53
N = 53
N = 50
N = 47
N = 194
N = 194
N = 11
N = 15
Age, years; Mean
Sex, N (%)
Race, N (%)
Treatment goal is
abstinence, N (%)
N (%) with ADS
N (%) with >16 ADS
Detox status, N (%)
‡4 days of abstinence
prior to randomization,
Percentage of days with
heavy drinking in
month prior to
baseline, Mean (SD)
46.0 (9.6) 47.4 (10.4) 43.5 (10.4)43.3 (10.7)44.8 (10.1)44.5 (11.0) 48.0 (9.8) 46.9 (8.1)
62.2 (27.1)60.7 (24.5) 60.9 (24.3)66.1 (23.0) 64.4 (26.3)66.3 (24.6)56.1 (15.6) 51.6 (23.3)
No lead-in abstinenceLead-in abstinence ‡ 4 days
XR-NTX 380 N = 177Placebo N = 181 XR-NTX 380 N = 28Placebo N = 28
Age, years; mean (±SD)
Sex, N (%)
Race, N (%)
Treatment goal is abstinence, N (%)
N (%) with ADS
Mean (SD) score
N(%) with >16 ADS
Detox status, N (%)
‡4 days of abstinence prior to
randomization, N (%)
Percentage of days with heavy drinking in
month prior to baseline, Mean (SD)
44.6 (9.9)44.7 (10.9)47.5 (11.4)44.5 (10.3)
66.6 (26.0)68.3 (24.3)47.3 (17.9)45.5 (18.6)
No significant differences between XR-NTX 380 and placebo. XR-NTX, extended-release naltrexone.
EFFICACY OF EXTENDED-RELEASE NALTREXONE
Efficacy of XR-NTX for Patients with Relatively High-
Severity Alcohol Dependence
Results of the Andersen–Gill recurrent-event Cox model
revealed that the more severely dependent group, as defined
by the ADS score, had significantly longer time for the occur-
rence of heavy-drinking events with XR-NTX 380 mg versus
placebo. The overall occurrence of heavy drinking per unit
time in the 380 mg group was 42% lower than the placebo
group over the course of the 24-week treatment period (haz-
ard ratio = 0.583; p = 0.0049 [95% CI: 0.400–0.849]).
Patients with higher ADS scores treated with XR-NTX
380 mg also showed a significantly greater reduction from
baseline in the percentage heavy-drinking days (37.3%) versus
placebo (27.4%; p = 0.039) (see Fig. 2). This effect was even
more pronounced when severity was defined by the presence
of a detoxification immediately prior to randomization: the
XR-NTX 380 mg group showed a 48.9% reduction in heavy-
drinking days compared with 30.9% for placebo (p = 0.004)
(see Fig. 2).
For those with an ADS score >16, responder rates (clinical
response = having £2 heavy-drinking days in any 28-day
period during 6-month treatment phase) tended to be greater,
by 86%, for XR-NTX 380 mg than for placebo (28.3%
[13⁄46] vs. 15.2% [7⁄46]; p = 0.13). Among those with ADS
£ 16, responder rates were 32.7% (17⁄52) for XR-NTX
380 mg and 6% (3⁄50) for placebo (p = 0.001). For those
who had a detoxification immediately prior to randomization,
responder rates tended to be greater, by 122%, for XR-NTX
380 mg (63.6% [7⁄11] vs. 28.6% [4⁄14]; p = 0.08). Among
those who did not have a detoxification prior to randomiza-
tion, responder rates were 28.6% (53⁄185) for XR-NTX
380 mg and 17.6% (33⁄188) for placebo (p = 0.014).
Prediction of Clinical Response to XR-NTX 380 mg
Logistic regression analyses indicated that detoxification
status and lead-in abstinence status were significant univariate
predictors of a clinical response to XR-NTX 380 mg. In a
multivariate logistic regression, only lead-in abstinence status
remained as a significant predictor of clinical response (see
These analyses for this paper were derived from one of
the largest (N = 624) and longest (6 months), multisite,
double-blind, randomized clinical trials of alcohol pharmaco-
therapies yet reported in the literature, and results are placebo
controlled. The main finding of this study was that XR-NTX
380 mg combined with a low-intensity psychosocial interven-
tion demonstrated efficacy compared with placebo in patients
Fig. 1. Drinks per day in the 90 days prior to treatment for patients with and without lead-in abstinence.
ADS > 16 Detox Occurred
Change in Percent Heavy Drinking Days
XR-NTX 380 mg
P = .039
P = .004
Definition of Severe Alcohol Dependence
Fig. 2. Change from baseline in % heavy-drinking days for patients with
relatively severe alcohol dependence. ADS, Alcohol Dependence Scale;
XR-NTX, extended-release naltrexone.
PETTINATI ET AL.
with relatively higher severity alcohol dependence. Efficacy
was measured as significant reductions in heavy drinking and
longer maintenance of abstinence over the course of 24 weeks
of treatment. These results were evident with either of 2
definitions of pretreatment severity—high ADS score or a
medical detoxification just prior to randomization. Clinical
response to XR-NTX 380 mg versus placebo among those
with higher severity was found to be similar to those with
lower severity, as measured on the ADS. Neither of the 2
severity measures significantly predicted clinical response to
XR-NTX 380 mg once the overlap with lead-in (‡4 days) of
abstinence was statistically considered in a multivariate analy-
sis. Nonetheless, the comparison with placebo among the
high-severity subgroup indicates that XR-NTX 380 mg is effi-
cacious for alcohol-dependent patients with relatively higher
An additional interesting and important finding of this
study was that lead-in abstinence of ‡4 days was associated
with a relatively higher severity alcohol dependence. Lead-in
abstinence was associated with a higher ADS score, greater
probability of receiving a medical detoxification, and more
drinks per day prior to randomization (particularly from 20
to 90 days prior to first dose of study medication, i.e., exclud-
ing the 3 weeks prior to study initiation during which detoxifi-
cation or drinking taper may have been under way).
The fact that lead-in abstinence was associated with greater
severity of alcohol dependence provides a further understand-
ing of the results of the primary efficacy trial on XR-NTX
(Garbutt et al., 2005). In the Garbutt and colleagues (2005)
report, and elaborated upon in more depth in a subsequent
article (O’Malley et al., 2007), it was found that XR-NTX
380 mg was particularly efficacious among the subgroup of
patients with lead-in abstinence. For patients who had at least
4 days of lead-in abstinence, those who received XR-NTX
380 mg, compared with placebo, had a 3-fold longer duration
of initial abstinence (41 vs. 12 days, respectively) and almost 3
times as many achieved a full 24 weeks of consecutive absti-
nence (31% vs. 11%, respectively) (O’Malley et al., 2007).
The current report clarifies that these patients who achieve
lead-in abstinence (and for whom XR-NTX 380 mg is
approved for use by the FDA) are more likely to be the rela-
tively severely dependent individuals.
Some clinicians may perceive that patients who achieve
abstinence before treatment is administered are able to do so
primarily because their illness is less severe. However, the cur-
rent data contradict this finding. This clinical perception is
likely reinforced by the awareness of individuals who stop
alcohol consumption on their own without treatment. How-
ever, data on individuals who stop drinking without treat-
ment have focused on problem drinkers rather than those
who are dependent on alcohol (e.g., Sobell et al., 1996).
Within a sample of treatment-seeking alcohol-dependent
patients, the data presented here indicate that, despite con-
suming more alcohol on a daily basis in the 90 days prior to
treatment, the more severe alcohol-dependent patients are
more likely to endorse abstinence as a goal of treatment and
become initially abstinent (either on their own or through a
medical detoxification) prior to beginning treatment. Previous
research has also found that those more severely dependent
on alcohol are more likely to endorse abstinence as a treat-
ment goal (Donovan et al., 2006).
The finding in this study of efficacy of XR-NTX 380 mg
among the more severely alcohol-dependent patients chal-
lenges previous findings that have suggested that naltrexone
be confined to alcohol-dependent patients with lower severity.
Oral naltrexone was used in these previous studies that did
not show efficacy compared with placebo in high-severity
patient subgroups. The link between reduced efficacy of oral
naltrexone and high-severity alcohol dependence may be med-
iated by poor adherence among patients with more severe
dependence. High-severity alcohol dependence has been asso-
ciated with lower adherence to medication treatments for
alcohol dependence (Baltieri et al., 2009). Adherence to oral
naltrexone has repeatedly been found to be essential to the
efficacy of the medication (Baros et al., 2007; Chick et al.,
2000; Krystal et al., 2001; Pettinati et al., 2000). Psychosocial
means of managing adherence have been developed and vali-
dated, including behavioral couples therapy (O’Farrell and
Fals-Stewart, 2002) and laboratory monitoring (Cone et al.,
1974; Meyer et al., 1984). A pharmacological approach to
addressing adherence is the extended-release formulation of
naltrexone, because it maintains continuous plasma levels of
naltrexone for more than 30 days following a single injection
(Dunbar et al., 2006). By addressing adherence pharmacoki-
netically, XR-NTX 380 mg may be particularly well-suited
for the treatment of the severely alcohol-dependent indi-
vidual. This suggests that, whereas outcomes from medi-
cation-assisted treatment employing oral agents may be
compromised by nonadherence, an approach that might be
termed ‘‘adherence pharmacotherapy,’’ may hold promise,
even for patients with severe alcohol dependence.
Severity may occur in other dimensions besides withdrawal
or physical dependence. Two recent studies have examined
treatment with XR-NTX in another high-severity subpopula-
tion, that is, alcohol-dependent patients with repeat driving-
while-intoxicated (DWI) offenses. A 10-patient case series
Table 2. Logistic Regressions Predicting Responder Rates from Severity
Criteria Within XR-NTX 380 mg Group
(N = 98; 17
Odds ratiop-valueOdds ratiop-value
Alcohol Dependence Scale
>16 (n = 50) vs. £16
(n = 53)
Detox (n = 7) vs. no detox
(n = 189)
status ‡ 4 days (n = 19)
vs. not (n = 177)
4.36 0.0236.50 0.136
XR-NTX, extended-release naltrexone.
EFFICACY OF EXTENDED-RELEASE NALTREXONE
reported significant within-subject decreases from pre- to
posttreatment in drinks per day and drinks per drinking day,
and significant increases in abstinent days (Lapham and
McMillan, 2010). A study of 64 case-matched multiple off-
enders (averaging approximately 3 prior DWI offenses) found
that XR-NTX patients versus patients given treatment-as-
usual in the DUI⁄Drug Court system in Michigan and
Missouri had a 62% reduction in the annualized re-arrest rate
(8% vs. 26% for treatment-as-usual; p < 0.05) (Finigan
et al., 2010). Taken together, these findings suggest that XR-
NTX may be a useful treatment for high severity in a variety
of severity dimensions.
A limitation of the present report is that the analyses were
post hoc and were conducted on a subsample of the original
randomized sample and therefore had reduced statistical
power relative to the parent study because of the small sample
sizes. It would therefore be important to confirm the current
results in a larger fully powered prospective study that targets
high-severity patients. In addition, as with the parent study,
the results reported here may not generalize to all patients
with alcohol dependence (the trial excluded patients with
unstable major depressive disorder, bipolar disorder, psycho-
sis, or dependence within the past year on benzodiazepines,
opiates, or cocaine). Another limitation of this research is that
the lack of occurrence of a medical detoxification is not
always an indicator of low severity of alcohol dependence.
Some individuals with high severity of alcohol dependence
might not receive a detoxification because of logistical barriers
(i.e., availability of clinicians who can provide this type of
treatment and financial constraints). Additional research is
also needed to determine the optimal duration of treatment
and the causal relationships between severity of alcohol
dependence, initial abstinence, attrition from treatment, and
In summary, the data from these analyses demonstrated
that patients with lead-in abstinence of at least 4 days had
more severe alcohol dependence. More severely dependent
patients receiving XR-NTX 380 mg over the course of
24 weeks of treatment were found to have a significantly
reduced rate of heavy-drinking days, compared with placebo.
XR-NTX 380 mg also demonstrated efficacy for maintaining
initial and total abstinence in the subset of more severely
dependent patients who had lead-in abstinence. The results
support the use of XR-NTX 380 mg for maintaining
abstinence as well as reduction in heavy drinking among those
with relatively more severe alcohol dependence. The impli-
cation of these data, in contrast to the conclusions drawn in
literature reviews of oral naltrexone trials, is that adher-
ence pharmacotherapy with XR-NTX 380 mg may have an
impact in treating alcohol-dependent patients with higher
Robert Gallop, PhD, provided assistance in the analysis of
the data under a contract from Alkermes, Inc.
American Society of Addiction Medicine (1996) Patient Placement Criteria for
the Treatment of Substance-related Disorders. 2nd edn. American Society
of Addiction Medicine, Inc., Chevy Chase, MD.
Andersen PK, Gill RD (1982) Cox’s regression model for counting processes:
a large sample study. Ann Stat 10:1100–1120.
Anton RF, O’Malley SS, Ciraulo DA, Cisler RA, Couper D, Donovan DM,
Gastfriend DR, Hosking JD, Johnson BA, LoCastro JS, Longabaugh R,
Mason BJ, Mattson ME, Miller WR, Pettinati HM, Randall CL, Swift R,
Weiss RD, Williams LD, Zweben A, COMBINE Study Research Group
(2006) Combined pharmacotherapies and behavioral interventions for alco-
hol dependence. The COMBINE Study: a randomized controlled trial.
Baltieri DA, Daro FR, Ribeiro PF, De Andrade AG (2009) The role of alco-
holic beverage preference in the severity of alcohol dependence and adher-
ence to the treatment. Alcohol 43:185–195.
Baros AM, Latham PK, Moak DH, Voronin K, Anton RF (2007) What role
does measuring medication compliance play in evaluating the efficacy of
naltrexone? Alcohol Clin Exp Res 31:596–603.
Berggren U, Berglund K, Fahlke C, Aronsson E, Eriksson M, Balldin J (2007)
Tobacco use is associated with more severe alcohol dependence, as assessed
by the number of DSM-IV criteria, in Swedish male type 1 alcoholics. Alco-
hol Alcohol 42:247–251.
Bouza C, Magro A, Mun ˜ oz A, Amate JM (2004) Efficacy and safety of nal-
trexone and acamprosate in the treatment of alcohol dependence: a system-
atic review. Addiction 99:811–828.
Chick J, Anton R, Checinski K, Croop R, Drummond DC, Farmer R,
Marshall J, Moncrieff J, Morgan MY, Peters T, Ritson B (2000) A multi-
centre, randomized, double-blind, placebo-controlled trial of naltrexone
in the treatment of alcohol dependence or abuse. Alcohol Alcohol 35:587–
Cone EJ, Gorodetzky CW, Yeh SY (1974) The urinary excretion profile of
naltrexone and metabolites in man. Drug Metab Dispos 2:506–512.
Donovan DM, Kivlahan DR, Doyle SR, Longabaugh R, Greenfield SF
(2006) Concurrent validity of the Alcohol Use Disorders Identification Test
(AUDIT) and AUDIT zones in defining levels of severity among out-
patients with alcohol dependence in the COMBINE study. Addiction
Drug and Alcohol Services Information System (2004) The DASIS Report:
Discharges from Detoxification: 2000. Office of Applied Studies, Substance
Abuse and Mental Health Services Administration (SAMHSA). Available
at: http://www.oas.samhsa.gov. Accessed August 18, 2010.
Dunbar JD, Turncliff RZ, Dong Q, Silverman BL, Ehrich EW, Lasseter KC
(2006) Single and multiple dose pharmacokinetics of long-acting naltrexone.
Alcohol Clin Exp Res 30:480–490.
Finigan M, Perkins T, Zold-Kilbourn P (2010) Preliminary evaluation of
extended-release naltrexone (XR-NTX) in Michigan and Missouri drug
courts. Presented at the American Psychiatric Association, Boston, MA,
Garbutt JC, Kranzler HR, O’Malley SS, Gastfriend DR, Pettinati
HM, Silverman BL, Loewy JW, Ehrich EW, Vivitrex Study Group
(2005) Efficacy and tolerability of long-acting injectable naltrexone for
alcohol dependence: a randomized controlled trial. JAMA 293:1617–
Goldstein BI, Velyvis VP, Parikh SV (2006) The association between moderate
alcohol use and illness severity in bipolar disorder: a preliminary report.
J Clin Psychiatry 67:102–106.
Grant BF, Dawson DA, Stinson FS, Chou SP, Dufour MC, Pickering RP
(2004) The 12-month prevalence and trends in DSM-IV alcohol abuse and
dependence: United States, 1991–1992 and 2001–2002. Drug Alcohol
Hicks BM, Krueger RF, Iacono WG, McGue M, Patrick CJ (2004) Family
transmission and heritability of externalizing disorders: a twin-family study.
Arch Gen Psychiatry 61:922–928.
Hogan JW, Roy J, Korkontzelou C (2004) Handling dropout in longitudinal
studies. Stat Med 23:1455–1497.
PETTINATI ET AL.
Krystal JH, Cramer JA, Krol WF, Kirk GF, Rosenheck RA for the Veterans
Affairs Naltrexone Cooperative Study Group (2001) Naltrexone in the
treatment of alcohol dependence. N Engl J Med 345:1734–1739.
Lapham SC, McMillan GP (2010) Open-label pilot study of extended-release
naltrexone to reduce drinking and driving among repeat offenders. J Addict
Med Epub ahead of print, July 28.
Li TK, Hewitt BG, Grant BF (2007) Is there a future for quantifying drinking
in the diagnosis, treatment, and prevention of alcohol use disorders? Alco-
hol Alcohol 42:57–63.
Lima AF, Pechansky F, Fleck MP, de Boni R (2005) Association between psy-
chiatric symptoms and severity of alcohol dependence in a sample of Brazil-
ian men. J Nerv Ment Dis 193:126–130.
McLellan AT, Kushner H, Metzger D, Peters R, Smith I, Grissom G, Pettinati
H, Algeriou M (1992) The fifth edition of the Addiction Severity Index.
J Subst Abuse Treat 9:199–213.
Mee-Lee D, Shulman GD, Fishman M, Gastfriend DR, Griffith JH (2001)
ASAM Patient Placement Criteria for the Treatment of Substance-Related
Disorders, Second Edition-Revised (ASAM PPC-2R). American Society of
Addiction Medicine, Inc., Chevy Chase, MD.
Meyer MC, Straughn AB, Lo MW, Schary WL, Whitney CC (1984) Bio-
equivalence, dose-proportionality, and pharmacokinetics of naltrexone after
oral administration. J Clin Psychiatry 45:15–19.
Morgan MY, Landron F, Lehert P (2004) Improvement in quality of life after
treatment for alcohol dependence with acamprosate and psychosocial sup-
port. Alcohol Clin Exp Res 28:64–77.
Morley KD, Teesson M, Reid SC, Sannibale C, Thomson C, Phung N,
Weltman M, Bell JR, Richardson K, Haber PS (2009) Naltrexone
versus acamprosate in the treatment of alcohol dependence: a multi-
centre, randomized, double-blind, placebo-controlled trial. Addiction
O’Farrell TJ, Fals-Stewart W (2002) Behavioral couples and family therapy
for substance abusers. Curr Psychiatry Rep 4:371–376.
O’Malley S, Garbutt JC, Gastfriend DR, Dong Q, Kranzler HR (2007)
Efficacy of extended-release naltrexone (XR-NTX) in alcohol-dependent
patients who are abstinent before treatment. J Clin Psychopharmacol 27:
Pettinati HM, Volpicelli JR, Pierce JD Jr, O’Brien CP (2000) Improving nal-
trexone response: an intervention for medical practitioners to enhance
medication compliance in alcohol dependent patients. J Addict Dis 19:71–
Rose RJ, Dick DM, Viken RJ, Pulkkinen L, Kaprio J (2004) Genetic and
environmental effects on conduct disorder and alcohol dependence symp-
toms and their covariation at age 14. Alcohol Clin Exp Res 28:1541–1548.
Ross HE, Gavin DR, Skinner HA (1990) Diagnostic validity of the MAST
and the Alcohol Dependence Scale in the assessment of DSM-III alcohol
disorders. J Stud Alcohol 51:506–513.
Saha TD, Chou SP, Grant BF (2006) Toward an alcohol use disorder
continuum using item response theory: results from the National Epidemio-
logic Survey on Alcohol and Related Conditions. Psychol Med 36:931–941.
Saunders JB, Aasland OG, Babor TF, de la Fuente JR, Grant M (1993)
Development of the Alcohol Use Disorders Identification Test (AUDIT):
WHO collaborative project on early detection of persons with harmful alco-
hol consumption: II. Addiction 88:791–804.
Skinner HA, Allen BA (1982) Alcohol dependence syndrome: measurement
and validation. J Abnorm Psychol 91:199–209.
Skinner HA, Horn JL (1984) Alcohol Dependence Scale: Users Guide. Addic-
tion Research Foundation, Toronto, Canada.
Sobell L, Sobell M (1992) Timeline follow-back, a technique for assessing self-
reported alcohol consumption, in Measuring Alcohol Consumption (Litten
RZ, Allen JP, eds), pp 41–72. The Humana Press, Totowa, NJ.
Sobell LC, Cunningham JA, Sobell MB (1996) Recovery from alcohol prob-
lems with and without treatment: prevalence in two population surveys. Am
J Public Health 86:966–972.
Thomas CP, Wallack SS, Lee S, McCarty D, Swift R (2003) Research to prac-
tice: adoption of naltrexone in alcoholism treatment. J Subst Abuse Treat
Volpicelli JR, Pettinati HM, McLellan AT, O’Brien CP (2001) Combining
Medication and Psychosocial Treatments for Addictions: The BRENDA
Approach. The Guilford Press, New York, NY.
Young SE, Stallings MC, Corley RP, Krauter KS, Hewitt JK (2000) Genetic
and environmental influences on behavioral disinhibition. Am J Med Genet
EFFICACY OF EXTENDED-RELEASE NALTREXONE