Naltrexone for Gambling Disorder
A patient making real progress in behavioral treatment for gambling disorder plateaus on one specific symptom: an intense urge he describes as feeling just like the alcohol cravings his late father once described to him.
W.P.'s father died two years ago, not long before W.P.'s own gambling started escalating in earnest — a timing he has thought about more than he has said out loud to anyone besides his wife. Both his father and an uncle struggled with alcohol use disorder, something he has always known and, he says, part of why he never drank heavily himself; he had assumed that vigilance made him safe from something like this. He is forty-seven, works as a claims adjuster, a detail-oriented job he's held for over fifteen years, and has always described himself, half-joking, as the most cautious person in any room — the one who reads every contract twice and keeps a spreadsheet for household expenses. He has coached his son's youth hockey team every winter weekend for six seasons, without exception, even through the two years his gambling problem was at its worst.
His gambling, primarily sports betting through mobile apps, escalated over roughly three years to the point of significant, self-disclosed financial harm that he and his wife are still working through in couples counseling — a process he says has been harder, in some ways, than stopping the gambling itself. He has no history of illicit substance use. He has been in weekly cognitive behavioral therapy targeting the gambling for four months, has made real progress reducing frequency, but describes one specific pattern he can't talk himself out of: an intense physical urge that builds over the course of a day, which he describes, in his own words, as feeling almost identical to how he imagines his father's drinking urges must have felt, based on things his father told him before he died.
A symptom that sounds like his father's, in a disorder his father never had
I'd offer a trial of naltrexone specifically for this. Kim and colleagues' original placebo- controlled trial established real efficacy for opioid antagonists in pathological gambling, and a later pooled predictor analysis across the nalmefene and naltrexone trial data found family history of alcoholism was the single strongest predictor of a positive response, with gambling-urge intensity itself also predicting response at the higher doses studied — which is close to exactly how he's describing his own experience.
I want to be honest about the evidence's actual size here: naltrexone has no FDA approval for gambling disorder, and while the trial data is real, it's a considerably smaller body of evidence than what supports its approved indications. He's already responding well to CBT on frequency — I'd want to consider intensifying or extending that first before adding a medication with a comparatively thin evidence base for this specific use.
His CBT response on frequency is real, but it's specifically the urge-intensity symptom that's plateaued despite continued therapy — that's a different axis than frequency, and more of the same treatment isn't obviously the right lever for a symptom it hasn't moved in four months.
I don't think this needs to be either-or. Continue his CBT exactly as it is — it's working on what it's working on. Add naltrexone specifically targeted at the urge-intensity symptom, and frame it to him plainly as addressing the piece therapy hasn't resolved yet, not as a sign therapy has failed or as a replacement for the real progress he's already made.
Agreed: naltrexone started as a targeted addition alongside unchanged CBT, framed to W.P. plainly as addressing the specific urge-intensity symptom therapy hasn't resolved rather than as a sign of treatment failure.
Not agreed: how much a patient's family history of a related but distinct substance use disorder should influence medication choice for gambling disorder specifically, given the trials showing that subgroup benefiting most are still relatively small. The addiction psychiatrist sees his family history as a real, clinically useful signal; the clinical pharmacologist cautioned against treating a subgroup finding from limited trials as though it were a validated, individualized predictor.