Naltrexone-bupropion shows promise for meth withdrawal

In individuals with moderate or severe methamphetamine use disorder (MUD), a combination of naltrexone and bupropion (NB) appeared to be more effective than placebo in facilitating methamphetamine (meth) withdrawal, results of the ADAPT-2* trial suggest.

“[Our] primary outcome was a response, defined as at least three out of four meth-negative urine samples obtained at the end of [both study stages],” the researchers explained. “[Our findings show a higher] response … among participants who received [NB] than those who received placebo.”

MUD is a serious disorder that entails functional and mental impairment and frequent relapses. [Drug Alcohol Depend 2014;143:11-21; Drug Alcohol Depend 2013;129:167-179] The surge of MUD in the US has propelled meth as one of the leading causes of death by drug overdose. [Natl Vital Stat Rep 2019;68:1-16; Drug Alcohol Depend 2018;193:14-20] “[With] no medication approved by the FDA for MUD treatment … effective treatment has been identified as an essential public health goal,” they stressed.

Naltrexone and bupropion have demonstrated efficacy for MUD treatment – both individually and as a combination regimen. [Neuropsychopharmacology 2015;40:2347-2356, Addiction 2014;109:1878-1886, CNS Drugs 2020;34:337-365; J Addict Med 2016;10:236-243] As such, the investigators aimed to investigate its efficacy and safety using a sequential parallel comparison design comprising two 6-week study stages.

Stage 1 (S1) saw 403 participants (mean age 41 years, 68.7 percent male) who wanted to quit from or reduce their almost-daily meth use. Participants were randomized in a 0.26:0.74 ratio to receive injectable naltrexone 380 mg extended-release (ER) Q3W plus oral bupropion ER 450 mg QD or matching placebo. Placebo recipients who did not respond in S1 (n=225) were rerandomized 1:1 in S2 to receive NB or placebo. [N Engl J Med 2021;384:140-153]

There was a greater fraction of participants who had a response with NB vs placebo during both stages (16 percent vs 3 percent [S1] and 11 percent vs 2 percent [S2]). Weighted average response across both stages was also higher with NB vs placebo (14 percent vs 2 percent), generating a significant between-group difference of 11.1 percentage points (Wald z-test statistic, 4.53; p<0.001).

Sensitivity analyses also saw higher response rates with NB vs placebo among participants who provided all four urine samples in the last 2 weeks of each study stage (29 percent vs 5 percent [S1] and 16 percent vs 1 percent [S2]) and under the assumption of equal weight for each stage (16 percent vs 3 percent and 11 percent vs 2 percent, respectively). Between-group differences in these respective subanalyses were 18.7 and 11.4 percentage points.

AEs that occurred more frequently with NB vs placebo in both stages were nausea (38 percent vs 15 percent [S1] and 28 percent vs 7 percent [S2]) and vomiting (12 percent vs 2 percent and 10 percent vs 3 percent, respectively). Other common AEs with NB during S1 were headache, dizziness, and constipation. AEs were mostly mild to moderate.

The objective primary outcome, as well as the high adherence (>75 percent in the NB arm during both stages) and low attrition rates, were considered some of the study’s strengths. However, the latter two may have also limited generalizability of the findings, the researchers noted. Adherence to the oral drug also cannot be confirmed, as these were only self-reported and were not monitored by clinicians.

“The results … may be difficult to explain to patients and practitioners because of the sequential parallel comparison design … This method was intended to enhance the likelihood of detecting efficacy of the combination treatment,” explained the researchers.

“Replication of our results in a more naturalistic effectiveness design could be a next step,” they added, noting a possible continuation of the study. “The 12-week duration of a trial of a substance use disorder requires consideration of how the treatment can be adapted to practice.”