N-Acetyl Cysteine, a Glutamate-Modulating Agent, in the Treatment of Pathological Gambling: A Pilot Study

N-Acetyl Cysteine, a Glutamate-Modulating Agent, in the Treatment of Pathological Gambling: A Pilot Study Jon E. Grant, Suck Won Kim, and Brian L. Odlaug Background: Although pathological gambling (PG) is relatively common, pharmacotherapy research for PG is limited. N-acetyl cysteine (NAC), an amino acid, seems to restore extracellular glutamate concentration in the nucleus accumbens and therefore offers promise in reducing addictive behavior. Methods: Twenty-seven subjects (12 women) with DSM-IV PG were treated in an 8-week open-label trial of NAC with responders (defined as a ⱖ 30% reduction in Yale Brown Obsessive Compulsive Scale Modified for Pathological Gambling [PG-YBOCS] total score at end point) randomized to 6 weeks of double-blind NAC or placebo. Results: The PG-YBOCS scores decreased from a mean of 20.3 ⫾ 4.1 at baseline to 11.8 ⫾ 9.8 at the end of the open-label phase (p ⬍ .001). Sixteen of 27 subjects (59.3%) met responder criteria. The mean effective dose of NAC was 1476.9 ⫾ 311.3 mg/day. Of 16 responders, 13 entered the double-blind phase. Of those assigned to NAC, 83.3% still met responder criteria at the end of the double-blind phase, compared with only 28.6% of those assigned to placebo. Conclusions: The efficacy of NAC lends support to the hypothesis that pharmacological manipulation of the glutamate system might target core symptoms of reward-seeking addictive behaviors such as gambling. Larger, longer, placebo-controlled double-blind studies are warranted.

Key Words: Addiction, amino acid, gambling, glutamate, pharmacology, treatment

P

athological gambling (PG), a significant public health problem characterized by persistent and recurrent maladaptive patterns of gambling, is associated with impaired functioning, reduced quality of life, and high rates of bankruptcy and divorce (Argo and Black 2004; Grant and Kim 2005). Past-year adult prevalence rates for PG are estimated at 1% (Cunningham-Williams et al. 1998; Shaffer et al. 1999). Although PG occurs frequently in primary care (6.2% of outpatients), it often goes unrecognized and untreated (Pasternak and Fleming 1999). Because untreated PG can impair function in multiple domains (Potenza et al. 2002), validated treatments are needed to optimize mental health care. At the present time, there is no treatment for PG that has been approved by the Food and Drug Administration, despite almost a decade of intense research. Controlled clinical trials evaluating medication treatments for PG have demonstrated that opioid antagonists (naltrexone and nalmefene) (Grant et al. 2006; Kim et al. 2001) as well as lithium (Hollander et al. 2005a) show promise in reducing gambling urges, thoughts, and behaviors. Similarly, psychosocial interventions, particularly cognitive behavioral therapy, might also be beneficial for some individuals with PG (Hodgins and Petry 2004). These treatments, however, do not seem effective for all individuals with PG, and so additional treatment options are needed. Preclinical studies have suggested that levels of glutamate within the nucleus accumbens mediate reward-seeking behavior (Kalivas and Volkow 2005; McFarland et al. 2003). Studies examining cocaine addiction in rats demonstrate that N-acetyl From the Department of Psychiatry (JEG, SWK, BLO), University of Minnesota School of Medicine, Minneapolis, Minnesota. Address reprint requests to Jon E. Grant, J.D., M.D., M.P.H., Department of Psychiatry, University of Minnesota School of Medicine, 2450 Riverside Avenue, Minneapolis, MN 55454; E-mail: [email protected]. Received October 12, 2006; revised November 7, 2006; accepted November 23, 2006.

0006-3223/07/$32.00 doi:10.1016/j.biopsych.2006.11.021

cysteine (NAC), an amino acid and cysteine pro-drug, seems to increase extracellular levels of glutamate, stimulate inhibitory metabotropic glutamate receptors, and thereby reduce synaptic release of glutamate. Restoring extracellular glutamate concentration in the nucleus accumbens seems to block reinstitution of compulsive behaviors and decrease cravings. Although not previously studied in PG or other impulse control disorders, NAC has shown benefit in reducing the reward-seeking behavior in rats with cocaine dependence (Baker et al. 2003). In the only study involving humans, LaRowe et al. (2006) established the safety of NAC in individuals with cocaine addiction and reported a modest reduction in cue-induced craving in a double-blind cross-over design. On the basis of these encouraging preliminary reports of NAC in cocaine dependence in both animals and humans, we conducted a pilot study to examine the tolerability and efficacy of NAC in the treatment of PG. We hypothesized that on the basis of NAC’s potential to reduce drug cravings, it would reduce gambling cravings as well as the gambling behavior that results from cravings in individuals with PG.

Methods and Materials Subjects Men and women ages 18 –75 with a primary diagnosis of PG on the basis of DSM-IV criteria were recruited by newspaper advertisements. Subjects were recruited from January 2006 through June 2006. All subjects met the DSM-IV criteria for PG with the clinician-administered Structured Clinical Interview for Pathological Gambling (SCI-PG) (Grant et al. 2004). All subjects were required to have gambled within 1 week before enrollment. Eligible subjects were required to have a score of 15 or greater on the Yale Brown Obsessive Compulsive Scale Modified for Pathological Gambling (PG-YBOCS), a clinician-administered scale with a range from 0 to 40 (Pallanti et al. 2005). Women’s participation required negative results on a ␤-human chorionic gonadotropin pregnancy test and stable use of a medically accepted form of contraception. Exclusion criteria included: 1) infrequent gambling (i.e., ⬍ one time/week) that did not meet DSM-IV criteria for PG; 2) unstable medical illness or clinically significant abnormalities on BIOL PSYCHIATRY 2007;62:652– 657 © 2007 Society of Biological Psychiatry

J.E. Grant et al. physical examination at screen; 3) history of seizures; 4) myocardial infarction within 6 months; 5) current pregnancy or lactation or inadequate contraception in women of childbearing potential; 6) lifetime history of bipolar disorder type I or II, dementia, schizophrenia, or any psychotic disorder determined by Structured Clinical Interview for DSM-IV (SCID); 7) current or recent (past 3 months) DSM-IV substance abuse or dependence; 8) positive urine drug screen at screening; 9) initiation of psychotherapy or behavior therapy within 3 months before study baseline; and 10) previous treatment with NAC. Subjects who were currently taking psychotropic medications were allowed into the study as long as the dose of medication had been stable for 3 months before study inclusion and there were no plans to modify the dose during the study duration. Similarly, subjects attending Gamblers Anonymous were allowed to participate if attendance had been ongoing for at least 6 months before study entry. Subjects who changed doses of medication or started therapy or Gamblers Anonymous, on the basis of their self-report, were discontinued from the study. The institutional review board for the University of Minnesota approved the study and the informed consent. One investigator discussed potential risks of the study as well as alternative treatments with subjects. After complete description of the study, subjects provided written informed consent. This study was carried out in accordance with the Declaration of Helsinki. Study Design The study consisted of 8 weeks of open-label NAC. All eligible study subjects were started on unblinded NAC 600 mg/day for 2 weeks. The dose was then increased to 1200 mg/day for 2 weeks and increased again to 1800 mg/day for the remainder of the study. Doses were not increased or the scheduling of increases was modified if clinical improvement was attained at a lower dose (clinical improvement was assessed by the investigator with respect to gambling thoughts, urges, and behavior). Subjects who were not compliant with their use of study medication (i.e., failing to take medication for 3 or more consecutive days) were discontinued from the study. Screening Assessments Subjects were evaluated at entry into the study by the Structured Clinical Interview for Pathological Gambling (SCIPG), a reliable and valid diagnostic instrument using criteria for DSM-IV PG (Grant et al. 2004). Psychiatric comorbidity was assessed with the SCID (First et al. 1995) and SCID-compatible modules for impulse control disorders (Grant et al. 2005). Medical history, physical examination, and urine toxicology were obtained. The PG symptom severity was assessed with the clinician-rated Yale Brown Obsessive Compulsive Scale Modified for Pathological Gambling (PG-YBOCS) (Pallanti et al. 2005) and the self-rated Gambling Symptom Assessment Scale (G-SAS) (Kim et al. 2001). Efficacy and Safety Assessments Subjects were seen weekly for 4 weeks and then every 2 weeks for the remaining 4 weeks of the open-label phase of the study. The primary outcome measure was the PG-YBOCS (Pallanti et al. 2005). Response was defined a priori as a 30% or greater reduction in PG-YBOCS total score at end point compared with baseline (a measure used in previous PG studies) (Hollander et al. 2000). The PG-YBOCS is a reliable and valid, 10-item, clinician-administered scale that rates gambling symptoms within the last 7 days, on a severity scale from 0 to 4 for

BIOL PSYCHIATRY 2007;62:652– 657 653 each item (total scores range from 0 to 40 with higher scores reflecting greater illness severity). The first five items comprise the gambling urge/thought subscale (time occupied with urges/ thoughts; interference and distress due to urges/thoughts; resistance against and control over urges/thoughts), and items 6 –10 comprise the gambling behavior subscale (time spent gambling and amount of gambling; interference and distress due to gambling; ability to resist and control gambling behavior). The following secondary measures were used at each study visit: Gambling Symptom Assessment Scale. The Gambling Symptom Assessment Scale (G-SAS) (Kim et al. 2001) is a 12-item, reliable and valid self-rated scale assessing gambling urges, thoughts, and behaviors during the previous 7 days. Each item is rated 0 – 4 with a possible total score of 48. Higher scores reflect greater severity of PG symptoms. Clinical Global Impression-Improvement and Severity scales. The Clinical Global Impression-Improvement and Severity scales (CGI) (Guy 1976) consist of two reliable and valid 7-item Likert scales used to assess severity and change in clinical symptoms. The scale ranges from 1 ⫽ “very much improved” to 7 ⫽ “very much worse.” Both clinician- and subject-rated improvement scores were performed at each visit. The CGI severity scale was used at each visit and ranges from 1 ⫽ “not ill at all” to 7 ⫽ “among the most extremely ill.” Improvement in CGI was used to note change only in PG symptoms. Sheehan Disability Scale. The Sheehan Disability Scale (SDS) (Sheehan 1983) is a three-item, reliable and valid scale that assesses functioning in three areas of life: work, social or leisure activities, and home and family life. Total scores range from 0 to 30, with higher scores reflecting greater functional impairment. Hamilton Depression Rating Scale. The Hamilton Depression Rating Scale (HAM-D) (Hamilton 1960) is a valid and reliable, 17-item, clinician-administered rating scale assessing severity of depressive symptoms. Hamilton Anxiety Rating Scale. The Hamilton Anxiety Rating Scale (HAM-A) (Hamilton 1959) is a reliable and valid, clinician-administered, 14-item scale that provides an overall measure of global anxiety. A secondary measure used only at baseline and open-label study end point was the Quality of Life Inventory (QOLI) (Frisch et al. 1993). The QOLI is a 16-item self-administered rating scale that assesses life domains such as health, work, recreation, friendships, love relationships, home, self-esteem, and standard of living. The QOLI has demonstrated excellent reliability and validity in nationwide normative studies and in studies of other impulse control disorders (Grant and Kim 2005). Safety assessments at each visit included evaluations of sitting blood pressure, heart rate, and weight. Adverse effects were documented and included time of onset and resolution, severity, action taken, and outcome. Use of concomitant medications was recorded in terms of daily dosage, start and stop dates, and reason for use. Urine toxicology and urine pregnancy tests were performed only at screening. Compliance was monitored by pill count. Double-Blind Discontinuation Those subjects who “responded” to open-label NAC (response being defined a priori as a 30% or greater reduction in PG-YBOCS total score at end point compared with baseline) were randomized by the university’s investigational pharmacy (in blocks of four with computer-generated randomization with no clinical information) either to the NAC dose at which they www.sobp.org/journal

654 BIOL PSYCHIATRY 2007;62:652– 657 completed the open-label phase or to an equivalent number of identical-appearing placebo capsules containing lactose powder. No adjustments were made to dosage during the double-blind phase. The double-blind phase lasted 6 weeks, with subjects seen every 2 weeks. Efficacy and safety assessments were performed at each visit during the double-blind discontinuation phase, with one investigator responsible for noting side effects and another administering outcome measures. Data Analysis The primary efficacy measure was the PG-YBOCS. Data were analyzed with the last observation carried forward in the intentto-treat (ITT) population. Only subjects with at least one follow-up evaluation were included. In the open-label phase, baseline and subsequent scores on continuous study measures were compared with paired t tests, two-tailed. For the doubleblind phase of the study, the percentage of subjects who continued as “responders” (on the basis of the PG-YBOCS) was compared after 6 weeks with the Fisher exact test. In addition, changes in scores on outcome measures during the 6-week double-blind phase were assessed with a Mann–Whitney test. Because we performed multiple comparisons, we used an adjusted ␣ level of p ⬍ .01; we did not adjust the ␣ level to reflect all statistical comparisons, because this is the first study of this topic and is therefore exploratory.

Results Twenty-nine subjects with current DSM-IV PG were enrolled. Two subjects never returned for their first evaluation (both subjects reported that they decided against participating in the study). Twenty-seven subjects (mean age ⫽ 50.8 ⫾ 12.1 years [range 21– 65]; 12 women [44.4%]) with current DSM-IV PG were included in the ITT analysis. Demographics and clinical characteristics of the ITT sample are presented in Table 1. Eighteen subjects (66.7%) reported slots, 6 (22.2%) reported blackjack, and 3 (11.1%) reported Internet sports betting as their primary gambling activity. Over the previous 12 months, subjects reported losing a mean of 57.5% of their gross income to gambling. The mean gambling losses/ week were $456 (⫾ $341). Thirteen subjects (48.1%) met diagnostic criteria for at least one lifetime co-occurring Axis I disorder: major depressive disorder (n ⫽ 8; 29.6%) and alcohol use disorder (n ⫽ 5; 18.5%) were the most common (Table 1). Four subjects (14.8%) met criteria for past-year major depressive disorder but were currently asymptomatic. Eight (29.6%) met criteria for current nicotine dependence. Of the 27 subjects, 5 (18.5%) were currently taking psychotropic medications: 2 were taking sertraline 100 mg/day for at least 1 year; 2 were taking fluoxetine 20 mg/day for at least 1 year; and 1 had been taking venlafaxine 300 mg/day for the past 2 years. Only one subject was receiving ongoing supportive therapy for relationship issues. No one else was receiving any type of therapy. No subject was attending Gamblers Anonymous. 8-Week Open-Label Phase Of the 27 subjects, 23 (85.2%) completed the 8-week openlabel study. The 4 subjects who discontinued treatment reported being unable to keep the study schedule (they did not differ significantly from those who remained in the study on any baseline measure). Scores on the PG-YBOCS decreased from a mean of 20.3 ⫾ 4.08 at baseline to 11.8 ⫾ 9.81 at the end of the open-label phase [t (50) ⫽ 4.17, p ⬍ .001], a decrease of 41.9%. www.sobp.org/journal

J.E. Grant et al. Table 1. Baseline Demographic and Clinical Characteristics of Individuals With Pathological Gambling Subjects (n ⫽ 27) Age Mean (⫾ SD), [range], yrs Gender, n (%) Female Race/Ethnicity, n (%) Caucasian African-American Native-American Marital Status, n (%) Single Married Widow/Separated/Divorced Education, n (%) High school grad or less Part college College grad or post-college education Unemployed, n (%) Age at PG Onset, Mean (⫾ SD), [range], yrs Primarily Strategic Gamblers, n (%)a Previous Gamblers Anonymous Attendance, n (%) Previous Treatment for Gambling, n (%) Percentage of Gross Income Lost to Gambling in The Previous Yr Committed Illegal Acts Due to Gambling, n (%)b Currently Taking Medication With Possible Psychotropic Effects, n (%) Comorbid Lifetime Disorders, n (%)c Major depressive disorder Dysthymia Social phobia Generalized anxiety disorder Panic disorder Agoraphobia Post-traumatic stress disorder Any anxiety disorder Any eating disorder Alcohol abuse/dependence Drug abuse/dependence Nicotine dependence Attention-deficit hyperactivity disorder Any other impulse control disorder Any somatoform disorder Any personality disorder

50.8 (12.1) [21–65] 12 (44.4) 26 (96.3) 0 (0) 1 (3.7) 12 (44.4) 8 (29.6) 7 (25.9) 2 (7.4) 13 (48.1) 12 (44.4) 3 (11.1) 37.1 (12.8), [18–58] 9 (33.3) 5 (18.5) 4 (14.8) 57.5% 11 (40.7) 5 (18.5) 8 (29.6) 0 (0) 0 (0) 1 (3.7) 1 (3.7) 0 (0) 0 (0) 2 (7.4) 1 (3.7) 5 (18.5) 1 (3.7) 10 (37.0) 0 (0) 2 (7.4) 0 (0) 2 (7.4)

PG, pathological gambling. a Strategic gambling ⫽ for example, cards, dice, sports (compared with non-strategic, which includes slots, pull tabs). b Illegal activities includes acts even if not known by authorities (e.g., forging checks, embezzlement, prostitution). c Comorbid lifetime disorders did not include bipolar disorder or psychotic disorders, because these were exclusion criteria.

Sixteen (59.3%) subjects met criteria as treatment responders (ⱖ 30% decrease in PG-YBOCS total score) at the end of the open-label phase. On the basis of CGI-Improvement scores, 14 of the 27 subjects (51.9%) were “very much improved” and 2 (7.4%) were “much improved” by the end of the open-label phase. Rates of subjects who reported themselves as “very much improved” and “much improved” were 37.0% (n ⫽ 10) and 22.2% (n ⫽ 6), respectively. When only study completers are considered (n ⫽ 23), 16 (69.6%) were responders on the PG-YBOCS. Of the 16 who responded, 10 (62.5%) reported abstinence from

BIOL PSYCHIATRY 2007;62:652– 657 655

J.E. Grant et al.

Table 2. Change in Study Measures in 27 Pathological Gambling Subjects Treated With Open-Label N-Acetyl Cysteine Measure PG-YBOCS Total PG-YBOCS Urge/Thought Subscale PG-YBOCS Behavior Subscale G-SAS Total Score CGI-Improvement, n (%)b Very much improved Much improved CGI-Severityc Hamilton Depression Rating Scale Sheehan Disability Scale Hamilton Anxiety Rating Scale Quality of Life Inventory Money Lost Gambling week

Baseline

Week 8 - End Point

20.3 ⫾ 4.08 9.62 ⫾ 2.62 10.4 ⫾ 4.39 32.3 ⫾ 5.54 NA

11.8 ⫾ 9.81 4.73 ⫾ 4.75 6.56 ⫾ 5.64 19.3 ⫾ 13.6

5.00 ⫾ .83 6.70 ⫾ 4.02 13.8 ⫾ 6.43 5.19 ⫾ 3.51 33.1 ⫾ 14.5 $456 ⫾ 341

10 (37.0) 6 (22.2) 3.00 ⫾ 2.18 3.93 ⫾ 4.38 7.67 ⫾ 8.88 3.26 ⫾ 3.74 39.7 ⫾ 16.1 $223 ⫾ 327

Test Statistica

p

4.17 4.59 2.80 4.60 NA

⬍ .001 ⬍ .001 .007 ⬍ .001 NA

4.45 2.43 2.91 1.95 ⫺1.34 2.05

⬍ .001 .019 .005 .056 .188 .048

Last observation carried forward. All values shown as mean ⫾ SD, unless otherwise indicated. PG-YBOCS, Pathological Gambling Modification of the Yale Brown Obsessive Compulsive Scale; G-SAS, Gambling Symptom Assessment Scale. a Paired t tests 2-tailed; df ⫽ 50. b Clinical Global Impression Improvement. c A score of 1 ⫽ not ill at all; 2 ⫽ borderline mentally ill; 3 ⫽ mildly ill; 4 ⫽ moderately ill; 5 ⫽ markedly ill; 6 ⫽ extremely ill; and 7 ⫽ among the most extremely ill.

gambling. The mean effective dose of NAC was 1476.9 ⫾ 311.3 mg/day. None of the 5 subjects taking psychotropic medications responded to open-label NAC. During the open-label phase, secondary measures of gambling symptomatology also showed improvement between baseline and end point in the intent-to-treat population (Table 2). The PG-YBOCS behavior subscale demonstrated a 36.9% reduction, and there was a 50.8% reduction in the PG-YBOCS urge/thought subscale. The G-SAS demonstrated a 40.2% reduction in selfreported gambling symptoms. Level of functioning improved between baseline and end point as indicated by a 44.4% reduction on the SDS scale. Although overall quality of life improved by 16.6% as measured by the QOLI, depressive symptoms decreased by 41.3%, and anxiety symptoms improved by 37.2%, none of these changes were statistically significant at an adjusted ␣ level of p ⬍ .01. Adverse events were few. Two subjects reported mild flatulence. No other adverse events were reported. 6-Week Double-Blind Discontinuation Phase Of the 16 subjects who responded to NAC and completed the 12-week open-label phase, 13 agreed to continue with the

100 90 80 70 60 50 40 30 20 10 0

Active Placebo

Week 0 Week 2 Week 4 Week 6

Figure 1. Percentage of subjects meeting responder criteria each week of the double-blind discontinuation phase. Fisher exact test ⫽ .078 at week 6; ␸ coefficient ⫽ ⫺.55.

double-blind discontinuation. The 16 responders did not differ significantly on any baseline measure when compared with nonresponders. The 3 responders who did not agree to doubleblind discontinuation reported being unwilling to risk loss of improvement in gambling symptoms. Of the 13 subjects, 6 were randomized to NAC and 7 to placebo. Of those assigned to NAC, 83.3% still met responder criteria at the end of the double-blind phase, compared with only 28.6% of those assigned to placebo (Fisher exact test ⫽ .078; ␸ coefficient ⫺.55 considered “relatively strong” [Rea and Parker 1992]; Figure 1). There was also a statistical trend for NAC to maintain improvement on the primary outcome measure (PG-YBOCS total score) and two secondary outcome measures (PG-YBOCS urge subscale and G-SAS) (Table 3).

Table 3. Change in Study Measures in 13 Pathological Gambling Subjects Randomized to 6 Weeks of N-Acetyl Cysteine or Placebo Outcome Measure PG-YBOCS Total Score Placebo Active PG-YBOCS Urge Subscale Score Placebo Active PG-YBOCS Behavior Subscale Score Placebo Active G-SAS Total Score Placebo Active Sheehan Disability Scale Total Score Placebo Active

Mean Rank

Z Score

p

10.6 7.2

⫺1.67

.095

5.0 3.6

⫺1.88

.060

6.0 4.0

⫺1.46

.144

13.0 7.4

⫺1.88

.060

4.4 6.0

.42

.675

Differences between first visit and last visit were calculated. The difference scores were rank ordered from low to high and a Mann–Whitney was run. PG-YBOCS, Pathological Gambling Modification of the Yale Brown Obsessive Compulsive Scale; G-SAS, Gambling Symptom Assessment Scale.

www.sobp.org/journal

656 BIOL PSYCHIATRY 2007;62:652– 657 Discussion This pilot study, the first to examine the efficacy of a glutamatergic agent in the treatment of PG, found that PG symptoms improved in a majority of subjects. Furthermore, the trend toward a significant difference between NAC and placebo in the double-blind discontinuation phase, suggests that there seems to be a drug effect underlying this improvement in PG symptoms. These findings suggest that pharmacological modulation of the glutamate system might prove useful in controlling the reward-driven behavior seen in PG, either as a primary agent or possibly as an augmenting agent. The efficacy of NAC lends support to the hypothesis that pharmacological manipulation of the glutamate system (within the nucleus accumbens) might target core symptoms of addictions (Hemby et al. 2005; Peters and Kalivas 2006), such as PG. It is hypothesized, on the basis of animal studies, that the key pathophysiologies of substance and behavioral addictions are mediated through glutamate and dopamine within the core of nucleus accumbens (Kalivas et al. 2005; McFarland et al. 2003). Repeated behaviors associated with reward produce persistent neuroplasticity in extracellular glutamate levels in the nucleus accumbens (Kalivas and Volkow 2005; Robinson and Kolb 1997). The cysteine-glutamate antiporters are the source of extracellular glutamate, and they modulate the release of glutamate and dopamine via stimulation of the glutamate group 2/3 metabotropic glutamate receptors (Kalivas and Volkow 2005; McFarland et al. 2003). Furthermore, in animal studies NAC seems to increase the activity of cysteine-glutamate antiporters in the nucleus accumbens and abolishes the reward-seeking behavior (Kalivas and Volkow 2005; McFarland et al. 2003). Behaviorally, NAC administration leads to diminished cocaine-induced reinstatement of drug-seeking behavior in animal studies (McFarland et al. 2003), and in humans it leads to reduced drug cravings (LaRowe et al. 2006). Pharmacotherapies, such as NAC, that target the prefrontal glutamatergic drive to the nucleus accumbens (Kalivas and Volkow 2005) might, therefore, also correct the underlying pathophysiology and symptoms of PG. This study’s findings that gambling cravings were reduced by treatment with NAC support this theory. Although NAC might be a promising treatment for PG, prior pharmacological studies in PG have shown that particular treatments have not been effective for all individuals with PG (Grant and Potenza, in press). These prior results might be due to a failure to recognize the heterogeneity of individuals with PG and how that heterogeneity might necessitate individually tailored treatment approaches (Grant and Potenza 2006). Recent research has suggested that the obsessive compulsive spectrum disorders might have various clusters that cohere on the basis of clinical and possibly genetic grounds (Lochner et al. 2005). Further research examining various abnormalities of the impulsivecompulsive neurocircuitry might lead us to a finding that there is a subtype of PG that is more like an addiction (Grant and Potenza 2006), whereas other PG subtypes have more in common with other disorders such as obsessive compulsive disorder (OCD) (Grant and Potenza, 2006) or bipolar disorder (Hollander et al. 2005a). Those with a subtype of PG that is similar to addictions might respond preferentially to a pharmacological treatment, such as NAC, targeting cravings. Although these notions remain speculative and require additional studies to examine their appropriateness, one future direction for the treatment of PG might be to better define subtypes of PG (with clinical symptomwww.sobp.org/journal

J.E. Grant et al. atology, neuroimaging, and pharmacogenetics) to guide pharmacological treatment selection. The hypothesized effects of glutamate in PG from this study are consistent with findings from neuroimaging studies in PG. Preclinical research in addictions have demonstrated alterations in the prefrontal cortex and corticofugal (glutamatergic) projections to the nucleus accumbens (Goldstein and Volkow 2002). In a study of unmedicated subjects with PG, investigators measured relative metabolic rate with 18F-deoxyglucose in positron emission tomography imaging to compare computer blackjack for monetary rewards versus points only (Hollander et al. 2005b). They noted significantly higher relative metabolic rates in the primary visual cortex, cingulate gyrus, putamen, and prefrontal cortex in the monetary condition compared with points only, suggesting heightened sensory and limbic activation with increased valence/risk (Hollander et al. 2005b). An interesting finding in this study was also that those few subjects taking psychotropic medications did not respond to NAC. Although the numbers were small, this finding seems inconsistent with promising reports of glutamatergic augmentation of selective serotonin reuptake inhibitors (SSRIs) in individuals with OCD (Coric et al. 2005; Lafleur et al. 2006). In individuals with OCD, glutamatergic agents might reduce regional cortico–striatal–thalamic hyperactivity, (Carlsson 2001; Coric et al. 2005). Although the exact nature of how glutamatergic agents in combination with SSRIs might reduce OCD symptoms is unknown (Coric et al. 2005), a similar neurochemical mechanism might be at work in some individuals with PG. One explanation for our discrepant findings might be that the doses of SSRI medication in four of the five subjects were lower than in OCD-resistant subjects (Coric et al. 2005; Lafleur et al. 2006) (sertraline 100 mg/day and fluoxetine 20 mg/day accounted for the concomitant psychotropic medications in four subjects), and our fifth subject was taking a serotonergic-noradrenergic agent (venlafaxine). Additionally, the five subjects on psychotropic medication failed to respond to either their antidepressant drugs before the study or NAC augmentation. On the basis of those results, these five subjects might also represent a treatmentresistant subtype of PG or yet another undefined subtype that merits further examination. N-acetyl cysteine has been extensively studied in a variety of medical problems (e.g., acetaminophen overdose, AIDS), and its lack of significant side effects might present an advantage over other pharmacologic agents. Adverse events reported in this study were consistent with oral NAC’s previously reported safety profile (LaRowe et al. 2006). This study represents, to our knowledge, the first trial of a glutamatergic agent in PG, but there exist several limitations. First, the double-blind findings in this study showed only a trend toward significance. It is possible, however, that a larger study, appropriately powered, might demonstrate a significant separation from placebo. Of course by including only responders in the double-blind phase we might have artificially increased the NAC treatment effect relative to placebo. Second, PG is a chronic disease that might require long-term therapy. By design, this study did not assess treatment effects beyond the 14-week treatment period, and longer-term effects thus require further evaluation. It is possible that a longer course of therapy could result in continued and even greater reductions in gambling symptoms. Alternatively, NAC’s therapeutic effects in PG might not endure beyond 14 weeks. Third, we enrolled subjects seeking pharmacological treatment and not psychotherapy. Therefore, these results might not generalize to the larger

J.E. Grant et al. population of people with PG. Effective behavioral treatments for PG are emerging and should be considered in conjunction with pharmacotherapies (Hodgins and Petry 2004). Similarly, additional studies might consider a comparison of NAC with a more proven pharmacological agent (e.g., nalmefene) (Grant et al. 2006) to examine its efficacy. Fourth, subjects who met current criteria for depression and were taking stable doses of psychotropic medications were enrolled, and as such, they might have affected NAC’s efficacy. Given that the individuals taking medications did not respond to NAC, we believe that NAC’s effect is not simply due to augmenting another agent. Fewer exclusionary criteria in this study, however, suggest that this sample might generalize to a larger population of pathological gamblers. Finally, we have not yet examined the optimal NAC dose, and whether some subjects would have responded to higher doses needs to be examined. This investigation suggests that NAC might be effective in the acute treatment of PG. As effective treatments for PG emerge, it becomes increasingly important that physicians and mental health care providers screen for PG in order to provide timely treatment. Given the open-label design of the study and the small number of subjects participating (particularly in the discontinuation component), the interpretation of the efficacy results of this study is limited. This research was supported in part by a Career Development Award (K23 MH069754-01A1) to JEG. Dr. Grant has been a consultant to Somaxon Pharmaceuticals and has received research grants from Somaxon Pharmaceuticals, Forest Pharmaceuticals, and GlaxoSmithKline. Argo TA, Black DW (2004): Clinical characteristics. In: Grant JE, Potenza MN, editors. Pathological Gambling: A Clinical Guide to Treatment. Washington DC: American Psychiatric Publishing, 39 –53. Baker DA, McFarland K, Lake RW, Shen H, Toda S, Kalivas PW (2003): N-acetyl cysteine-induced blockade of cocaine-induced reinstatement. Ann N Y Acad Sci 1003:349 –351. Carlsson ML (2001): On the role of prefrontal cortex glutamate for the antithetical phenomenology of obsessive compulsive disorder and attention deficit hyperactivity disorder. Prog Neuropsychopharmacol Biol Psychiatry 25:5–26. Coric V, Taskiran S, Pittenger C, Wasylink S, Mathalon DH, Valentine G, et al. (2005): Riluzole augmentation in treatment-resistant obsessive-compulsive disorder: An open-label trial. Biol Psychiatry 58:424 – 428. Cunningham-Williams RM, Cottler LB, Compton WM III, Spitznagel EL (1998): Taking chances: Problem gamblers and mental health disorders—results from the St. Louis Epidemiologic Catchment Area study. Am J Public Health 88:1093–1096. First MB, Spitzer RL, Gibbon M, Williams JBW (1995): Structured Clinical Interview for DSM-IV-Patient Edition (SCID-I/P, Version 2.0). New York: Biometrics Research Department, New York State Psychiatric Institute. Frisch MB, Cornell J, Villaneuva M (1993): Clinical validation of the Quality of Life Inventory: A measure of life satisfaction for use in treatment planning and outcome assessment. Psychol Assess 4:92–101. Goldstein RA, Volkow ND (2002): Drug addiction and its underlying neurobiological basis: Neuroimaging evidence for the involvement of the frontal cortex. Am J Psychiatry 159:1642–1652. Grant JE, Kim SW (2005): Quality of life in kleptomania and pathological gambling. Compr Psychiatry 46:34 –37. Grant JE, Levine L, Kim D, Potenza MN (2005): Impulse control disorders in adult psychiatric inpatients. Am J Psychiatry 162:2184 –2188. Grant JE, Potenza MN (2006): Compulsive aspects of impulse control disorders. Psychiatr Clin North Am 29:539 –551. Grant JE, Potenza MN (in press): Treatments for Pathological Gambling and Other Impulse Control Disorders. In: Nathan PE, Gorman JM, editors. A Guide to Treatments that Work, 3rd ed. New York: Oxford University Press. Grant JE, Potenza MN, Hollander E, Cunningham-Williams R, Nurminen T, Smits G, Kallio A (2006): A multicenter investigation of the opioid antag-

BIOL PSYCHIATRY 2007;62:652– 657 657 onist nalmefene in the treatment of pathological gambling. Am J Psychiatry 163:303–312. Grant JE, Steinberg MA, Kim SW, Rounsaville BJ, Potenza MN (2004): Preliminary validity and reliability testing of a structured clinical interview for pathological gambling (SCI-PG). Psychiatry Res 128:79 – 88. Guy W (1976): ECDEU Assessment Manual for Psychopharmacology. US Dept Health, Education and Welfare publication (ADM) 76 –338. Rockville, Maryland: National Institute of Mental Health, 218 –222. Hamilton M (1959): The assessment of anxiety states by rating. Br J Med Psychiatry 32:50 –55. Hamilton M (1960): A rating scale for depression. J Neurol Neurosurg Psychiatry 23:56 – 62. Hemby SE, Tang W, Muly EC, Kuhar MJ, Howell L, Mash DC (2005): Cocaineinduced alterations in nucleus accumbens ionotropic glutamate receptor subunits in human and non-human primates. J Neurochem 95:1785– 1793. Hodgins DC, Petry NM (2004): Cognitive and behavioral treatments. In: Grant JE, Potenza MN, editors. Pathological Gambling: A Clinical Guide to Treatment. Washington DC: American Psychiatric Publishing, 169 –187. Hollander E, DeCaria CM, Finkell JN, Begaz T, Wong CM, Cartwright C (2000): A randomized double-blind fluvoxamine/placebo crossover trial in pathological gambling. Biol Psychiatry 47:813– 817. Hollander E, Pallanti S, Allen A, Sood E, Baldini Rossi N (2005a): Does sustained-release lithium reduce impulsive gambling and affective instability versus placebo in pathological gamblers with bipolar spectrum disorders? Am J Psychiatry 162:137–145. Hollander E, Pallanti S, Baldini Rossi N, Sood E, Baker BR, Buchsbaum MS (2005b): Imaging monetary reward in pathological gamblers. World J Biol Psychiatry 6:113–120. Kalivas PW, Volkow ND (2005): The neural basis of addiction: A pathology of motivation and choice. Am J Psychiatry 162:1403–1413. Kalivas PW, Volkow N, Seamans J (2005): Unmanageable motivation in addiction: A pathology in prefrontal–accumbens glutamate transmission. Neuron 45:647– 650. Kim SW, Grant JE, Adson DE, Shin YC (2001): Double-blind naltrexone and placebo comparison study in the treatment of pathological gambling. Biol Psychiatry 49:914 –921. Lafleur DL, Pittenger C, Kelmendi B, Gardner T, Wasylink S, Malison RT, et al. (2006): N-acetylcysteine augmentation in serotonin reuptake inhibitor refractory obsessive-compulsive disorder. Psychopharmacol 184:254 – 256. LaRowe SD, Mardikian P, Malcolm R, Myrick H, Kalivas P, McFarland K, et al. (2006): Safety and tolerability of N-acetylcysteine in cocaine-dependent individuals. Am J Addict 15:105–110. Lochner C, Hemmings SMJ, Kinnear CJ, Niehaus DJH, Nel DG, Corfield VA, et al. (2005): Cluster analysis of obsessive-compulsive spectrum disorders in patients with obsessive-compulsive disorder: Clinical and genetic correlates. Compr Psychiatry 46:14 –19. McFarland K, Lapish CC, Kalivas PW (2003): Prefrontal glutamate release into the core of the nucleus accumbens mediates cocaine-induced reinstatement of drug-seeking behavior. J Neurosci 23:3531–3537. Pallanti S, DeCaria CM, Grant JE, Urpe M, Hollander E (2005): Reliability and validity of the Pathological Gambling Modification of the Yale-Brown Obsessive-Compulsive Scale (PG-YBOCS). J Gambling Stud 21:431– 443. Pasternak AV, Fleming MF (1999): Prevalence of gambling disorders in a primary care setting. Arch Fam Med 8:515–520. Peters J, Kalivas PW (2006): The group II metabotropic glutamate receptor agonist, LY379268, inhibits both cocaine- and food-seeking behavior in rats. Psychopharmacology (Berl) 186:143–149. Potenza MN, Fiellin DA, Heninger GR, Rounsaville BJ, Mazure CM (2002): Gambling: an addictive behavior with health and primary care implications. J Gen Intern Med 17:721–732. Rea LM, Parker RA (1992): Designing and Conducting Survey Research. San Francisco: Jossey-Bass Publishers. Robinson TE, Kolb B (1997): Persistent structural modifications in nucleus accumbens and prefrontal cortex neurons produced by previous experience with amphetamine. J Neurosci 17:8491– 8497. Shaffer HJ, Hall MN, Vander Bilt J (1999): Estimating the prevalence of disordered gambling behavior in the United States and Canada: A research synthesis. Am J Pub Health 89:1369 –1376. Sheehan DV (1983): The Anxiety Disease. New York: Scribner’s.

www.sobp.org/journal