Blood test identified a DNA methylation marker linked to lifetime cannabis use | Research

The takeaway

An epigenome-wide study of over 2,500 women identified and replicated a blood-based DNA methylation site associated with lifetime cannabis use, though its predictive accuracy was modest.

Genetics researchers, epidemiologists, and anyone interested in biological markers of cannabis exposure.

First replicated blood-based epigenetic marker of lifetime cannabis use

What the researchers found

Researchers identified a replicated association between lifetime cannabis use and DNA methylation at site cg15973234 in the CEMIP gene (combined P = 3.3 x 10^-8). The methylation-based classifier had modest predictive accuracy with AUC of 0.60 in discovery and 0.54 in replication samples.

Why it matters

Objective biomarkers of cannabis use are needed for epidemiological research. While urine and blood tests detect recent use, DNA methylation could potentially serve as a marker of cumulative lifetime exposure, capturing long-term use patterns that self-report may miss.

The numbers in context

Discovery N = 1,730 (855 ever-users); replication N = 853 (392 ever-users); cg15973234 in CEMIP gene P = 3.3 x 10^-8; AUC 0.60 discovery, 0.54 replication

How the study worked

Epigenome-wide association study using blood-based DNA methylation data from a case-cohort within the Sister Study (discovery N = 1,730; replication N = 853). Researchers tested associations between methylation at over 450,000 sites and self-reported lifetime cannabis use.

What this study cannot tell us

All-female cohort (Sister Study) limits generalizability. Modest predictive accuracy. Cannot distinguish frequency, recency, or duration of use. Self-reported cannabis use as the comparator.

How to read the evidence

Well-powered EWAS with independent replication, though the effect size is small and the cohort is all-female.

When this study was published

Published in 2021.

The bigger picture

This is the first blood-based epigenome-wide study of cannabis use. While the predictive accuracy is too modest for clinical use, the replicated finding at CEMIP suggests cannabis leaves a detectable molecular footprint that could improve epidemiological research.

Questions still open

Does this methylation change have functional consequences? Would it perform better in populations with heavier use patterns? Could combining multiple methylation sites improve prediction? Does the CEMIP methylation change persist after cessation?

Common questions

Can a blood test tell if someone has ever used cannabis?

This study found a DNA methylation signal associated with lifetime cannabis use, but its accuracy was too modest (AUC 0.54-0.60) for reliable individual-level detection.

What is DNA methylation?

DNA methylation is a chemical modification to DNA that can change how genes are expressed without altering the genetic code itself. Environmental exposures, including substance use, can alter methylation patterns.

Read the original research

Epigenome-wide analysis uncovers a blood-based DNA methylation biomarker of lifetime cannabis use.

American journal of medical genetics. Part B, Neuropsychiatric genetics : the official publication of the International Society of Psychiatric Genetics, 186(3), 173-182

View on PubMed ↗Publisher’s paper ↗

Citation

Markunas, Christina A; Hancock, Dana B; Xu, Zongli; Quach, Bryan C; Fang, Fang; Sandler, Dale P; Johnson, Eric O; Taylor, Jack A. (2021). Epigenome-wide analysis uncovers a blood-based DNA methylation biomarker of lifetime cannabis use.. American journal of medical genetics. Part B, Neuropsychiatric genetics : the official publication of the International Society of Psychiatric Genetics, 186(3), 173-182. https://doi.org/10.1002/ajmg.b.32813