CBD Gel in Fragile X Syndrome: A Subgroup Finding Does Not Replace the Primary Result
The CONNECT-FX trial did not establish a significant benefit on its primary social-avoidance outcome in the full analysis group. Favorable results in selected methylation subgroups were exploratory and should be presented alongside that primary result.
The full trial came first
Researchers randomized 212 children and adolescents with fragile X syndrome to 12 weeks of transdermal CBD gel or placebo alongside standard care. The study used weight-based daily doses of 250 or 500 mg. The efficacy analysis included 210 participants with the required baseline and postbaseline measurements.
The main endpoint was change in social avoidance on a fragile-X-specific version of the Aberrant Behavior Checklist. This was a human behavioral-treatment trial, not an assay of cannabis impairment and not evidence about other childhood conditions.
Improvement within a group was not superiority
Mean improvements in social avoidance, irritability and social unresponsiveness/lethargy were numerically greater with CBD, but the full-group between-treatment differences were not statistically significant. The key clinician-rated global-improvement comparison was also nonsignificant.
Some caregiver measures were more favorable. Those observations should not be merged into a single claim that the primary outcome was positive. Different scales, responder thresholds and analyses answer different questions.
The methylation analyses generated a narrower hypothesis
The authors then examined participants with at least 90% methylation of the FMR1 promoter, a group comprising about 80% of those randomized. In that ad hoc analysis, the social-avoidance treatment difference favored CBD by one point, with a nominal p value of 0.020. Additional analyses in the fully methylated group also produced favorable signals.
“Ad hoc” and “nominal” matter here. Examining selected subgroups and multiple endpoints after the main analysis increases the need for independent confirmation. A plausible biological explanation for a subgroup response does not turn that analysis into the original confirmatory test.
What the safety findings and limits mean
No serious or severe treatment-emergent adverse events were reported, and measured laboratory changes were comparable between groups. Application-site pain was more frequent with CBD. These are observations over 12 weeks, rather than a guarantee about rare reactions or prolonged treatment.
The study involved children and adolescents with a defined genetic syndrome and a particular gel. Results cannot automatically be extended to adults, all FMR1 methylation profiles or oral cannabis products. The public-facing conclusion should preserve both parts of the evidence: a negative full-group primary comparison and a promising subgroup hypothesis requiring a trial designed to test it.
Related analysis: CBD-Rich Extract in Autism: Interview Findings and Standardized Scales Diverged.
Primary research
Elizabeth Berry-Kravis and colleagues. A randomized, controlled trial of ZYN002 cannabidiol transdermal gel in children and adolescents with fragile X syndrome (CONNECT-FX). Journal of Neurodevelopmental Disorders (2022). DOI: 10.1186/s11689-022-09466-6.
