For years, the conversation about cannabis has centred on mental health, dependency, and the gateway drug debate. Those are serious concerns. But scientific research is now pointing to something far less discussed: the relationship between cannabis exposure and cancer, including acute myeloid leukaemia (AML), one of the most aggressive blood cancers in existence.
The risk does not stop with the person using cannabis. Evidence shows the damage to DNA can be passed on to children who have never used cannabis at all.
Cannabis and AML: What the Data Shows
Multiple large-scale epidemiological studies, drawing on data from both the United States and Europe, have linked cannabis use to a range of cancers including AML, acute lymphoid leukaemia (ALL), breast, thyroid, liver, and pancreatic cancers.
A 2021 study published in Scientific Reports by Reece and Hulse examined US cancer and drug use data from 2003 to 2017. The researchers found that cannabis, THC, cannabigerol, and cannabichromene fulfilled causal criteria for AML, along with thyroid, liver, breast, and pancreatic cancers. The findings held up after controlling for ethnicity, income, and other drug use, with effect sizes ranging from medium to large.
The European data points the same way. A 2023 study published in the Journal of Xenobiotics by Reece, Bennett, and Hulse analysed cancer incidence data from the European Cancer Information System covering 2000 to 2020. Countries with high cannabis use had significantly higher cancer rates. Of 41 cancers examined, 25 were causally linked to cannabis exposure in multivariate models. Compared directly with tobacco and alcohol, cannabis was the stronger correlate of cancer incidence.
When the Damage Is Inherited: Childhood Cancer and the Transgenerational Risk
One of the most serious dimensions of cannabis and AML research involves children. Some cases of AML develop in early childhood, before the age of ten. These children have had no opportunity to use cannabis themselves. Their disease appears to reflect damage inherited from a parent who did.
Research published in 2023 examining sperm DNA methylation in cannabis-dependent individuals found that withdrawal from cannabis triggered 6.5 times more AML-related gene activations than cannabis dependence itself. The hypothesis is that the withdrawal involved in birth, when a newborn is no longer receiving cannabis compounds through the placenta, may activate leukaemogenic gene expression in children born to cannabis-using parents.
This is not a speculative claim. Earlier work from the Children’s Cancer Group in the United States and Canada had already identified a relationship between parental cannabis use and childhood non-lymphoblastic leukaemia. More recent genomic and epigenomic research has provided a mechanism: cannabinoids disrupt DNA methylation patterns in sperm and eggs in ways that carry through to the next generation.
How Cannabinoids Damage DNA
Cannabis genotoxicity, its capacity to damage genetic material, has been documented for more than fifty years. What has changed is our understanding of just how extensive that damage is.
Cannabinoids are toxic to actin and tubulin, the structural proteins cells rely on when they divide. This disrupts the mitotic spindle and can cause chromosomes to separate incorrectly. They also generate reactive oxygen species that damage cell membranes, including those surrounding micronuclei, small packets of fragmented chromosomal material. When those membranes rupture, the result can be chromothripsis, a catastrophic shattering of whole chromosome segments and a known driver of aggressive malignancies.
The genotoxic impact is not limited to THC. Cannabigerol (CBG) and cannabichromene (CBC) have both been implicated in cancer-related chromosomal damage. Even cannabidiol (CBD), widely marketed as safe, has appeared in research as a carcinogenic risk factor for liver and other cancers. The cannabinoid cancer risk appears to run across the entire class, not just the psychoactive compounds.
Dose-response relationships follow exponential and supra-linear curves, meaning risk escalates sharply with increasing exposure rather than rising at a steady rate.
What Legalisation Is Doing to the Data
The Reece and Hulse studies in the US found that jurisdictions that had liberalised cannabis laws had higher cancer rates, rising faster than in non-liberal states, after adjusting for income, ethnicity, and other drug use. Legalisation correlated with worse cancer outcomes, including those associated with cannabis and AML.
Since 1975, the rate of acute lymphoid leukaemia in children under 20 in the US has risen by over 93 per cent. Total paediatric cancer rates rose 42 per cent between 1975 and 2017. Research published in BMC Cancer in 2021 identified cannabis exposure as a major driver of paediatric ALL rates across the United States, with the signal confirmed through spatiotemporal and causal inferential modelling.
Cannabis and AML in Context: A Broader Carcinogenic Picture
AML is not the only blood cancer linked to cannabinoid exposure. Both Hodgkin and non-Hodgkin lymphomas, chronic lymphoid leukaemia, and chronic myeloid leukaemia have appeared in the European datasets. Combined with reproductive cancers of the testis, ovary, prostate, and breast, the pattern strongly implicates damage to germline DNA, the genetic material in sperm and eggs.
Testicular cancer has received particular attention. Meta-analyses have shown that cannabis exposure elevates testicular cancer risk by a factor of approximately 2.6, with development appearing to be significantly accelerated. Researchers estimated an oncogenic rate-incubation index around six times greater than baseline.
The chapter by Reece and Hulse in the 2023 Elsevier volume Cannabis, Cannabinoids, and Endocannabinoids draws on the convergent US and European data and concludes that cannabinoid genotoxicity operates at the scale of hundreds of megabases of the human genome. This is not a targeted mutation. It is broad-spectrum chromosomal disruption.
The Case for Prevention
The research linking cannabis and AML, including acute myeloid and childhood ALL, has moved well beyond early-stage correlation. Multiple independent datasets across different nations, using inverse probability weighting, causal inference models, and spatiotemporal analysis, reach the same conclusions.
Cannabis is a genotoxin. Its carcinogenic risk rivals and in some analyses exceeds that of tobacco. Unlike tobacco, the genetic damage it causes can be inherited. A parent who uses cannabis today may be placing a child not yet born at risk of leukaemia in their first decade of life.
Prevention matters here. Not as a footnote, but as a direct response to what the evidence shows. Every informed choice not to use cannabis is one that may protect more than just the person making it.
(Source: WRD News)