Garbutt 2014: who responds best to naltrexone for alcohol

What they tested

Two reviewers screened 622 articles published from 1990 to April 2012 for evidence that personal or biological traits, called moderators, changed how well naltrexone worked for alcohol dependence. They rated each publication for risk of bias and judged the strength of the evidence for each trait.

28 publications covering 20 studies qualified. Most came from 12 randomized, placebo-controlled trials. The rest came from three non-randomized studies, one randomized study without a placebo, and pooled analyses of four trials.

What they found

A family history of alcohol problems was linked to better results on naltrexone in four of five studies that looked at it. The Asn40Asp variant of OPRM1, the gene for the mu-opioid receptor, was linked to better results in three of five.

Other traits reported to go with better results were being male (two of five studies), heavier drinking before treatment (two of two) and high craving (two of five).

The overall risk of bias in the published studies was high. The authors concluded that identifying who will respond to naltrexone is still in development, and that the data are not yet enough to recommend using any trait to decide treatment.

Limits

• The authors rated the overall risk of bias in the studies as high.
• Each trait was examined in only two to five studies.
• The studies show links between traits and response, not proof that a trait causes a better response.

What it means if you're considering naltrexone

No test can tell you in advance whether naltrexone will work for you. This review found hints, mainly family history and one gene variant, but not enough to act on. The site's guides add that a 2020 meta-analysis of seven trials found no clear effect of that gene variant once it allowed for multiple comparisons.