Can Metabolomics Distinguish Cannabis Use History From Acute Intoxication?

Blood metabolomic patterns may help distinguish use history from acute cannabis effects, but an exploratory classification result is not a validated impairment test. A controlled human study investigated those two questions separately.

Baseline differences came before intoxication

Madrid-Gambin and colleagues studied 35 occasional and chronic cannabis users under cannabis and placebo conditions. Blood was collected before treatment and repeatedly afterward. Sustained attention and subjective high were assessed during the acute observation period.

Fourteen metabolites distinguished the use-history groups at baseline. A discriminant model classified chronic versus occasional use with approximately 80% accuracy. That number concerns use-history classification; it is not an 80%-accurate diagnosis of impairment.

Acute patterns were not identical

Occasional users showed attentional impairment and greater subjective high after cannabis, accompanied by changes in organic acids and other metabolic signals. Chronic users exhibited a different and more limited acute metabolic pattern.

Those observations support investigating adaptation and individual response. They do not show that every chronic user is unaffected, or that a metabolite change uniquely identifies cannabis as its cause in a different population.

From discovery to a decision rule

Metabolic pathways respond to numerous influences. Before applying a panel in practice, an expert would need evidence about independent validation, reproducibility, other exposures, health conditions, and errors at the proposed decision threshold.

The observation window was short, and the sample was small. Blood metabolic signals also are not direct measurements of brain function. The study offers candidate markers and mechanistic questions, while leaving open whether they can reliably classify an individual’s functional state outside the controlled experiment.

Related reading: why cannabis effects can differ across memory tests.

Primary research

Francisco Madrid-Gambin; Noemí Haro; Natasha L. Mason; Pablo Mallaroni; Eef L. Theunissen; Stefan W. Toennes; Oscar J. Pozo; Johannes G. Ramaekers. Metabolomic profiling of cannabis use and cannabis intoxication in humans. Neuropsychopharmacology (2025). DOI: 10.1038/s41386-025-02082-7.