The investigators plan to investigate the safety, tolerability, and efficacy of the opioid antagonist naltrexone in Pathological Gambling. We hypothesize that naltrexone will be superior to placebo in reducing gambling urges and behavior, when combined with adjuvant non-pharmacological treatment as usual.

Pathological gambling (PG) is a significant public health problem that can cause significant devastation for the individuals affected and their families. Those afflicted may experience unemployment, considerable debt, marital problems, family dysfunction, legal troubles, incarceration, and mental health issues including suicidality. Prevalence estimates vary but most estimates put it at between 1% and 2%, with annual costs of over $5 billion in the United States alone. Thus, PG is costly not only for the affected individuals and their families, but also for society in general.

The current treatment as usual for PG is limited to various types of counseling, psychotherapy (e.g. cognitive behavioral therapy), and self-help groups such as Gamblers Anonymous. However, such treatment modalities have not been shown to be particularly effective. High rates or relapse are common and treatment attrition is often a concern.

Currently, no drugs are approved by the U.S. Food and Drug Administration for the treatment of PG. Pharmacological treatment studies are still in their infancy but show considerable promise. Several placebo-controlled, randomized clinical trials have been conducted, but results have been limited by small sample sizes, short durations, exclusion of individuals with co-occurring disorders, and heterogeneity in treatment response measures and diagnostic criteria used for inclusion.

Selective serotonin reuptake inhibitors (SSRIs) have shown mixed results in PG. Some studies have suggested a benefit of active drug over placebo while other studies have not. Mood stabilizers have not been extensively studied but some reports suggest that certain patients, such as those with co-morbid bipolar spectrum disorders, may benefit from this type of medication. Atypical antipsychotics have also been tried with limited success and may be more appropriate for patients with co-occurring psychotic disorders.

Opioid antagonists such as naltrexone and nalmefene, possibly through their modulation of the mesolimbic dopamine system, have demonstrated preliminary efficacy superior to placebo in treating PG. As with substance use disorders (SUD), it has been suggested that opioid antagonists may exert their therapeutic benefit by helping to reduce the appetitive urges or cravings present in symptomatology of addiction.

From the neurochemical perspective, the pharmacological action of opioid antagonists is to block the effects of endogenous endorphins on mu-opioid receptors and may inhibit dopamine release in the nucleus accumbens. The mu-opioid system is generally thought to be involved in the mediation of hedonic, rewarding and reinforcing behaviors. The mu-opioid and mesolimbic pathways have been implicated in PG. For example, problem gamblers have been shown to have elevated levels of the endogenous opioid β-endorphin during gambling (Shinohara et al 1999).

Naltrexone, a pure opioid antagonist, is an FDA-approved medication with two labeled indications. Firstly, for the blockade of the effects of exogenously administered opioids. And secondly, for the treatment of alcohol dependence. However, it's labeling is clear that it has not been shown to provide any therapeutic benefit except as part of an appropriate plan of management for addiction. Naltrexone has been investigated in PG in part due to its proposed ability to modulate the mesolimbic dopamine pathway. In preliminary studies, it has shown efficacy superior to placebo. As with clinical trials of alcohol dependence, it appears that naltrexone targets craving and urge states. In fact, naltrexone has shown to be particularly effective in individuals with stronger urges to gamble at treatment onset.

A double-blind, placebo-controlled twelve-week investigation of naltrexone in 83 subjects (of which 45 were used for analysis) has been described(Kim et al 2001). Doses were initiated at 25 mg/day and titrated to a maximum dosage of 250 mg/day, with an average final dose of 187.50 mg/day (SD=96.45). Naltrexone was superior to placebo and was associated with statistically significant improvement in various measures of gambling severity, both self-reported and clinician-administered. A post hoc analysis showed that naltrexone was more effective in gamblers who reported more severe urges.

More recently, an eighteen-week double-blind, placebo-controlled study of naltrexone for PG was reported(Grant et al 2008). Following a single-blinded one-week placebo lead-in, seventy-seven PG subjects were randomized to daily doses of 50mg, 100mg, or 150mg naltrexone. Unlike the previous shorter naltrexone study, this group included subjects with a range of co-occurring disorders. Outcomes did not significantly differ among the three dosages. The three active arms of the study were combined and compared to placebo. Analyses showed that naltrexone was more effective than placebo in decreasing PG severity, gambling urges, gambling behavior, and psychological functioning.

Nalmefene hydrochloride is a similar-acting opioid antagonist. Contrary to naltrexone, nalmefene is not associated with possible liver toxicity. A sixteen-week multicenter, randomized, dose-ranging, double-blind, placebo-controlled investigation was conducted at 15 outpatient treatment centers in the U.S., including at Yale(Grant et al 2006). Two hundred and seven male and female subjects were randomized to either 25mg, 50 mg, 100mg per day or to an equivalent placebo. All three active arms began with a one-week course of 25mg per day. This study did not include individuals with co-occurring disorders. Subjects assigned to the active arms showed statistically significant reductions in gambling severity. The 50mg and 100mg doses resulted in intolerable side effects. It appears that the lower 25mg dose was the most efficacious. In fact, the 25mg dose was unique in terms of demonstrating superiority to placebo based on overall response to treatment as measures by the Clinical Global Impression (CGI) improvement scale.

Based on previous encouraging results both at Yale and elsewhere, this study will attempt to replicate and extend the safety, tolerability, and efficacy findings of opioid antagonists in the management of PG. In addition, this study will provide much needed information regarding pharmacotherapy in conjunction with treatment as usual for PG.

Inclusion Criteria:

1. Men or women over age 18
2. Current DSM-IV PG Diagnosis as determined by a score of ≥ 5A criteria and B criterion present on the SCI-PG and a score ≥ 5 on the SOGS
3. Gambling behavior within 2 weeks prior to enrollment
4. For women, stable use of a medically accepted form of contraception and negative results on urine pregnancy test at study onset
5. Currently entering, enrolled, or interested in treatment for PG

Exclusion Criteria:

1. Gambling that does not meet DSM-IV criteria for PG
2. Unstable medical illness or clinically significant abnormalities on laboratory tests, EKG, or physical examination at screen
3. Past or current acute hepatitis or liver failure
4. History of renal impairment
5. Current or recent (within one week) treatment with an opioid agonist/opioid analgesic or current opioid withdrawal
6. Opiate agonist maintenance therapy (e.g. methadone)
7. Known sensitivity to opioid antagonists
8. Current pregnancy or lactation, or inadequate contraception in women of childbearing potential
9. A need for medication with unfavorable interactions with naltrexone
10. Clinically significant suicidality
11. Lifetime history of dementia, schizophrenia, or any psychotic disorder determined by SCID
12. Clinically significant cognitive impairment
13. Previous treatment with naltrexone or nalmefene
14. Treatment with investigational medication or depot neuroleptics within 3 months
15. Lack of proficiency in written and spoken English
16. Unable to travel to study sites for appointments

3 years

Enrollment was lower than expected and due to the low numbers of subjects completing the study (7 subjects completed all interventions) the numbers were not powerful enough to conduct a full analysis or provide any meaningful statistical analyses.