Julio Cesar Nunes, MD. Psychiatry resident, Yale University. New Haven, CT.
Noah Capurso, MD, MHS. Assistant professor of psychiatry, Yale University. New Haven, CT. Editor-in-chief of The Carlat Addiction Treatment Report.
Dr. Nunes and Dr. Capurso have no financial relationships with companies related to this material.
Naltrexone is best known for its FDA-approved indications: treating alcohol use disorder (AUD) and opioid use disorder (OUD). However, as a mu-opioid receptor antagonist, it modulates dopaminergic circuits that play critical roles in the neurobiology of addiction—potentially giving it much wider applications in addiction treatment (Capurso NA and Ross DA, Biol Psychiatry 2017;81(11):e79–e81). Some have even wondered if it might be a “pan-addiction treatment” given its potential widespread application and relatively benign side effect profile (Aboujaoude E and Salame WO, CNS Drugs 2016;30(8):719–733). But even though naltrexone is effective in some areas, it falls flat in others. Here, we review the evidence for its uses and discuss how best to apply it to clinical practice.
Alcohol use disorder
Naltrexone was first approved for AUD in 1994. But it was the COMBINE trial over a decade later that established it as one of the best first-line agents for AUD (Anton RF et al, JAMA 2006;295(17):2003–2017). This large, randomized, multicenter trial found that naltrexone was superior to both acamprosate and psychotherapy on a number of drinking outcomes. Interestingly, combining naltrexone with other interventions did not enhance its effects. More recently, a large epidemiologic study found that a naltrexone prescription was associated with a reduced risk of hospitalization and death due to alcohol-related causes (Heikkinen M et al, Addiction 2021;116(8):1990–1998).
Despite this compelling evidence, naltrexone is vastly underutilized for AUD. The 2019 National Survey on Drug Use and Health showed that only 7% of people with AUD receive any treatment at all, and less than 4% receive medication (www.tinyurl.com/ca4kuwan). To put this into context, 35%–45% of patients with high cholesterol receive the recommended statin treatment; far from ideal, but much better than AUD (Bucholz EM et al, Circulation 2018;137(21):2218–2230). The take-home message: Naltrexone is effective for AUD, it reduces alcohol-related morbidity and mortality, and we should be using it more often.
Opioid use disorder
Given naltrexone’s mechanism of action, it is not surprising that it has shown promise in treating OUD. Initially approved in 1984, naltrexone was seen by many as largely ineffective and only to be used when buprenorphine or methadone could not be employed. But evidence in naltrexone’s favor was published in the form of two clinical trials in 2017 and 2018, which found that long-acting injectable naltrexone (XR-NTX, Vivitrol) was non-inferior to flexibly dosed buprenorphine (Tanum L et al, JAMA Psychiatry 2017;74(12):1197–1205; Lee JD et al, Lancet 2018;391(10118):309–318).
But there was a caveat: XR-NTX is more difficult to initiate than buprenorphine. As an opioid blocker, naltrexone can precipitate opioid withdrawal. This means anybody receiving naltrexone must be completely opioid free for at least seven days and sometimes longer if the receiver has been using long-acting opioids. This presented a substantial barrier in one of the two studies, with only 72% of participants able to start XR-NTX compared to 94% able to start buprenorphine (Lee et al, 2018). Those who received XR-NTX also had an increased risk of overdose relative to those who received buprenorphine (Ajazi EM et al, J Addict Med 2022;16(4):440–446). While many still consider buprenorphine first line for OUD, all is not lost for naltrexone; a recent analysis found that XR-NTX worked better than buprenorphine for people experiencing homelessness (Nunes EV Jr et al, Am J Psychiatry 2021;178(7):660–671).
Stimulant use disorder
Stimulant use disorder has no approved medications. Naltrexone monotherapy does not seem to be particularly effective, though combining XR-NTX with bupropion showed promise in a recent high-profile study (Trivedi MH et al, N Engl J Med 2021;384(2):140–153). It should be noted that the doses of both medications used in the trial were on the high side: 380 mg IM every three weeks for XR-NTX and 450 mg daily for extended-release bupropion. The treatment’s overall effect was modest, with 13.6% of participants responding, but it did do significantly better than placebo, which had a treatment response of only 2.5%. Though certainly not a cure-all, XR-NTX plus bupropion remains one of the most promising new treatments for this notoriously difficult-to-treat condition.
Cannabis use disorder
We don’t know much about naltrexone for people who use cannabis, though there have been some promising preliminary findings. One double-blinded, placebo-controlled study found that participants treated with naltrexone smoked less cannabis than those not receiving naltrexone (Haney M et al, Neuropsychopharmacology 2015;40(11):2489–2498). A small open-label trial found that XR-NTX decreased the number of days per week people with cannabis use disorder (CUD) smoked cannabis (Notzon DP et al, Am J Drug Alcohol Abuse 2018;44(6):619–627). At this point, behavioral-based therapies remain the standard for CUD, though you could consider naltrexone as an adjunct or for patients who are not ready to engage in therapy.
Nicotine use disorder
The evidence for naltrexone in nicotine use disorder is disappointing. There is some evidence that it can mitigate post-quit weight gain, but we recommend sticking to approved treatments with a solid evidence base: varenicline, nicotine replacement, and bupropion (King AC et al, J Clin Psychopharmacol 2012;32(5):630–636).
Investigations of naltrexone for the treatment of behavioral addictions are still in their infancy, with studies mostly limited to small uncontrolled trials and case series. Nonetheless, there is some preliminary evidence that naltrexone might be helpful for pathologic gambling, problematic pornography use, compulsive sexual behaviors, compulsive buying disorder, kleptomania, and even trichotillomania (Mouaffak F et al, Eur Addict Res 2017;23(4):201–210). Though the data are far from conclusive, naltrexone is generally well tolerated and could be considered if better-studied behavioral interventions are not effective.
Naltrexone is often used to treat non-suicidal self-injurious behaviors (SIB), such as cutting. Evidence for its utility in SIB stretches back to the 1980s (Herman BH et al, Ann Neurol 1987;22(4):550–552), though studies have been small, open label, or at the level of case reports. Numerous small trials have found an effect specifically for SIB in people with intellectual disability, though a large meta-analysis found that there were not enough data to make a definitive conclusion about its efficacy one way or another (Rana F et al, Cochrane Database Syst Rev 2013;(4):CD009084). Psychotherapy remains the first-line treatment for SIB, but given its low side-effect profile, naltrexone could be considered for patients unable to engage in therapy or if therapy is not effective.
Dosing and administration
Naltrexone is available as 50 mg oral tablets and a 380 mg gluteal injection given every four weeks. For indications other than OUD, patients can start at the full dose of 50 mg daily. It is generally well tolerated, with the most common side effects being mild gastrointestinal (GI) distress in about a third of patients and transient transaminase increases in about a fifth of patients. GI distress can be severe in a small subset of patients, limiting medication adherence. For these patients, it can be helpful to start at 25 mg for a few days and increase to 50 mg (O’Malley SS et al, J Clin Psychopharmacology 2000;20(1):69–76). If side effects are an issue, decrease to 25 mg for a few days and then increase back to 50 mg. Tenderness or a small lump at the injection site is common after receiving XR-NTX.
Patients with OUD must have all opioids out of their system when they receive their first dose of naltrexone, otherwise they will experience withdrawal. Patients taking short-acting opioids, including tramadol, should be abstinent for seven to 10 days before receiving their first dose. Patients taking methadone or buprenorphine should be off all opioids for 14 days. If there is any doubt, consider a naloxone challenge test. Details on how to start naltrexone in patients with OUD, including the naloxone challenge, can be found here: www.tinyurl.com/bdeuwcw5.
XR-NTX is the formulation shown to be effective in treating OUD (Sullivan MA et al, Am J Psychiatry 2019;176(2):129–137). It’s less clear whether the PO or IM formation is better in AUD, so the choice can be left up to patient preference. We don’t have enough data to recommend one formulation over another for any of the other indications.
The oral dose can be increased to 100 mg daily if the 50 mg dose is not effective. Likewise, you can increase the XR-NTX dose by giving it as frequently as every three weeks. Finally, because naltrexone is hepatically metabolized, it is not recommended to start either formulation if transaminases are more than three to five times the upper limit of normal. If that is the case, LFTs can be trended and the medication started when levels normalize. Naltrexone also should be avoided in patients with severe liver impairment. But as long as your patient does not have known liver disease, LFTs don’t need to be regularly monitored (www.tinyurl.com/mryy5yc3). For more information on hepatic dosing, see our interview with Dr. Bataller in CATR March/April 2022.
Though not quite a "jack of all trades," naltrexone is very safe and does have its uses. The best evidence is for AUD, where it is a first-line treatment and remains vastly underutilized. The IM formulation is also effective for OUD, though the requirement of an opioid-free induction is a barrier for many. Data for CUD and behavioral addictions are preliminary, though promising.
PO Box 626, Newburyport MA 01950