REVIEW OF: Nunes EV Jr et al, Am J Psychiatry 2021;178(7):660–671

For decades, methadone and buprenorphine (Bup) have been upheld as the gold standard of opioid use disorder (OUD) treatments, with naltrexone largely considered second line. However, a pair of landmark studies challenged that wisdom by showing the non-inferiority of long-acting injectable naltrexone (XR-NTX) as compared to sublingual Bup (Lee JD et al, Lancet 2018;391(10118):309–318; Tanum L et al, JAMA Psychiatry 2017;74(12):1197–1205). These papers showed that the treatments are equally effective for patients who manage to start treatment. However, patients are much more likely to tolerate starting treatment with Bup because they need to be abstinent from opioids for only about 12 hours before Bup induction. Conversely, patients must be opioid-free for at least seven days before they can start XR-NTX—a tall order for many people with OUD.

The authors of this recent study hoped to drill deeper into these data to better understand which patients might do better with Bup vs XR-NTX. They did this by reexamining data from one of those original papers: the X:BOT trial (Lee et al, 2018). In this trial, 287 patients with OUD were randomized to Bup (8–24 mg sublingual daily) and 283 patients to XR-NTX (380 mg IM every 28 days). Participants were assessed weekly with urine drug screens and self-reported substance use. Authors examined the data for demographic characteristics predictive of successful medication initiation and rate of returning to use.

Overall, it was easier for participants to successfully start Bup (5.9% failure rate) compared to XR-NTX (27.9% failure rate). The following characteristics were most predictive of successful initiation with Bup versus XR-NTX:

1. Presence of chronic pain. Participants with moderate to severe chronic pain had a much higher failure rate with XR-NTX compared to Bup (32.4% vs 2%, OR = 23.68). This makes sense because having chronic pain would make it difficult to stop opioid use for several days.


2. Recent use of opioids. Individuals randomized to XR-NTX soon after their last use (less than three days) had a higher rate of treatment failure compared to Bup (41.3% vs 1.5%, OR = 47.79). This is no surprise, since use of naltrexone soon after opioid use may trigger withdrawal symptoms.


3. Stated preference for Bup. Participants who stated a preference for Bup had a much lower failure rate if they received Bup than if they received XR-NTX (0.88% vs 33.0%, OR = 55.28). Those who did not state a preference for Bup had similar failure rates with both medications.

The following characteristics tended to predict success with XR-NTX:

1. Being on probation/parole. XR-NTX fared better in participants on probation/parole as compared to the study participants overall. In this group, XR-NTX had a similar failure rate to Bup (17.39% vs 14.39%, OR = 1.25, CI = 0.42–3.61).


2. Being homeless. XR-NTX had an edge among homeless individuals where the rate of returning to opioid use was lower in those receiving XR-NTX as compared to Bup (51.6% vs 70.4%, OR = 0.45). This is the only patient characteristic that was predictive of return to use, as opposed to failure to start the medication.

It is worth noting that only opioid-related outcomes were measured, and that other potentially relevant outcomes unrelated to opioid use were not reported. For example, a patient with comorbid alcohol use disorder (AUD) might fare better with XR-NTX since it is an effective medication for both AUD and OUD. XR-NTX might also be preferred for patients with comorbid benzodiazepine use disorder, since Bup and benzos can produce fatal respiratory suppression when combined.

CATR’S Take
This study largely confirms previous findings that Bup and XR-NTX are both effective treatments for OUD once started, but that Bup is easier to initiate than XR-NTX. Factors predicting better success with initiating Bup include chronic pain, recent opioid use, and preference for Bup. Initiation of Bup and XR-NTX fared similarly for patients on probation. Homeless patients had lower rates of return to use with XR-NTX. These data provide useful guidance, but other patient factors should still be considered when making a final medication choice.