"The risk of overdose with naltrexone results when patients try to overcome μ-opioid receptor antagonism by taking increasingly high doses of unregulated drugs., Concerns about overdose misclassification in naltrexone trials have also been raised.Large epidemiological studies have found naltrexone to be less protective against overdose than buprenorphine, consistent with known pharmacology.Post-marketing data suggest waning protection with the monthly extended-release injectable naltrexone formulation used in X:BOT,although this is contested by the manufacturer.

"The original report found that induction failures affected the naltrexone group differentially: 28% for naltrexone versus 5·9% for buprenorphine. In a commentary from 2022, Lee and colleagues correctly acknowledge that the corresponding per-protocol analysis might, therefore, not represent a sample with treatment assigned at random.Statistical methods accounting for induction failures substantiate differential overdose risk., Lee and colleagues also suggest focusing on the maintenance period (ie, study days 22–168, after induction and before therapy discontinuation), when most overdoses (61%) occurred. Doing so, we found HR 3·77 (95% CI 1·23–11·57; Wald p=0·02; appendix) for intention-to-treat analysis, and 3·81 (95% CI 1·01–14·36; p=0·05) for per-protocol analysis. X:BOT shows increased risk of overdose with naltrexone than buprenorphine even during therapy.In X:BOT, Lee and colleaguesconcluded that “once initiated, both medications were equally safe”. We believe this statement is not supported by the public data. Although we think it is important to acknowledge that the number of participants experiencing an overdose in X:BOT could appear small from a statistical perspective (28 [5%] of 570), the other major trial comparing these two medications had only one overdose among 143 participants in Norway.

"Over the past decade drug overdoses in the USA have increased substantially. Given the demonstrated safety risk, naltrexone should not be presented as the only option for opioid dependence treatment, as is the case in many prisons and drug courts. With more than 100 000 overdose deaths annually, clinicians must make treatment decisions based on accurate information. We believe the original X:BOT reportshould be retracted or corrected quickly."


Dasgupta N, Ajazi EM, Schwartz TA, Marshall SW. Misclassification of overdose events in the X:BOT study. Lancet. 2023;402(10401):526-527. doi:10.1016/S0140-6736(23)00113-7