A low-THC cannabis extract just cleared its biggest autism trial yet

Researchers at a pediatric neurology clinic in Victoria, Australia enrolled 61 children ages 8-17 with autism spectrum disorder in a Phase II/III randomized, double-blind, placebo-controlled trial of NTI164 — a full-spectrum cannabis extract with less than 0.3% THC. Fifty-four completed the eight-week double-blind phase, published in Neurotherapeutics as "The Harmony Study."

Kids on NTI164, dosed at 10-20 mg/kg of body weight daily, showed significantly greater improvement than the placebo group on the Clinical Global Impression scale (average improvement score 2.62 versus 4.04, p<0.001) — plus gains in communication, daily living skills and social responsiveness. Forty-one adverse events were logged across the trial, and researchers characterized all of them as mild, non-serious and temporary.

It's one of the larger controlled trials to date testing a cannabis product specifically for pediatric autism, and the low-THC formulation matters for the same reason it does in dementia and epilepsy trials: it's aimed at avoiding intoxication while still using the full cannabinoid profile — CBDA, CBD, CBGA and CBDV alongside trace THC — rather than isolating a single compound.

One 8-week trial in one clinic isn't a green light for pediatricians to start recommending cannabis extracts broadly, and NTI164 remains an investigational product, not something available over a dispensary counter. But it adds real controlled-trial weight to a body of autism research that's mostly been small, open-label or observational until now — the kind of data that makes a case for larger, multi-site follow-up trials.