LAUSR

Unlike other addictions, opioid-use disorder has several highly effective medication treatments available, in particular methadone, buprenorphine, and naltrexone. Methadone is the least accessible or acceptable of these in many settings, so providers and patients are often faced with a choice between buprenorphine-naloxone (BUP-NX) and extended-release naltrexone injection (XR-NTX). Since XR-NTX became available, many providers and researchers expected it to have similar problems to oral naltrexone—difficulty starting the treatment, poor adherence, and concerns about the possibility of higher relapse and overdose compared with agonist medication. Until recently, no direct comparison trial was available to guide this treatment selection. Two concurrent trials comparing BUP-NX with XR-NTX have now been completed, one in Norway and one in the USA. The Norwegian randomised clinical non-inferiority study showed similar retention and effectiveness of both agents. In The Lancet, Joshua D Lee and colleagues report on the US study, a multicentre, open-label, randomised trial. Patients (aged 18 years and older) at inpatient detoxification units were randomly assigned to receive monthly XR-NTX (n=283) intramuscular injections or daily sublingual BUP-NX (n=287) for 24 weeks in typical community outpatient settings. The study addresses several outstanding questions about use of XR-NTX in real-world settings regarding feasibility of induction, and the safety and effectiveness of the treatment. To balance the success potential for the two treatments, randomisation and inductions occurred at different stages in the detoxification process. The main outcome of the study was higher relapse for XR-NTX than with BUP-NX in the intention-to-treat population (185 [65%] vs 163 [57%]; hazard ratio [HR] 1·36, 95% CI 1·10–1·68). However, this outcome was primarily due to fewer XR-NTX inductions and high relapse among induction failures. Therefore, once treatment was successfully initiated, both medications were similar in effectiveness and safety: among the 474 participants inducted to treatment, the proportion of opioid-relapse events was 52% for the XR-NTX group and 56% for the BUP-NX group (odds ratio 0·87, 95% CI 0·60–1·25; p=0·44), with no difference in the relative hazard of relapse over time (HR 0·92, 95% CI 0·71–1·18; p=0·49). There are two main implications of these results. First, these findings add to growing scientific literature supporting the effectiveness and safety of XR-NTX. The similar performance of the two medications after successful induction is consistent with the results of the Norwegian study. These are the only published direct comparison trials of these two drugs, and both studies show that XR-NTX is a valid alternative to agonist therapy that should be considered among effective treatment options for opioid use disorder. Second, the substantial induction hurdles for XR-NTX continue to present challenges, which are especially important because induction failure might lead to relapse and overdose. In fact, in this US study most XR-NTX induction failures did relapse (70 [89%] of 79), confirming the scope of this problem. Such treatment failures are even more worrisome because of the high risk of overdose with opioids and the continuing rise in overdose deaths in the USA. In this study, deaths were equivalent in the two groups (two for XR-NTX and three for BUP-NX). Differences in total overdoses (fatal and non-fatal) did not differ between groups, but the numbers were noteworthy—18 for XR-NTX versus ten for BUP-NX. Considering that the study was not powered to detect overdose differences, there should be continued evaluation of how failure Published Online November 14, 2017 http://dx.doi.org/10.1016/ S0140-6736(17)32872-6