A new systematic review published in brain, behavior, immunity This suggests that cannabinoid use does not have a single direct effect on the human immune system. Rather, regular use tends to increase both pro-inflammatory and anti-inflammatory biological markers simultaneously. These findings provide evidence that cannabis products may modulate the immune system and require more detailed nuances to be fully understood.
Cannabis use is increasing worldwide due to changes in legal policy and public perception of its safety. At the same time, medicine views cannabis through a different lens. Certain ingredients are being studied as treatments for pain and epilepsy, but regular use is also associated with cognitive and psychiatric risks.
Preclinical studies using animals and isolated cells often show that cannabinoids can reduce inflammation. In these laboratory settings, cannabinoids typically suppress immune cell activation. This early evidence led many to view cannabis as a broad-spectrum anti-inflammatory.
However, translating these experimental results to human biology has been incredibly inconsistent. Some previous human studies have reported increases in certain immune markers, while others have shown no effect or only reduced inflammation alone. Previous statistical reviews on this topic have often been limited as they have focused on a very narrow set of biological markers.
To address this confusion, researchers began a comprehensive analysis of the existing literature. They aimed to determine the association between regular cannabinoid use and peripheral inflammatory biomarkers. These biomarkers are measurable proteins and cells present in the blood that indicate how actively the immune system is responding to a threat.
Elucidating this relationship is of great public health relevance. Long-term, low-grade inflammation tends to contribute to heart disease, metabolic disorders, and mental illness. Scientists wanted to know whether cannabinoids move the immune system toward or away from this dangerous condition.
“We were motivated by the fact that cannabinoids are often described as having anti-inflammatory properties, but human evidence is inconsistent and dispersed across different biomarkers and populations. Importantly, we wanted to examine this more systematically by excluding samples with physical illnesses or other conditions that could directly alter inflammatory status. This allowed us to see the association more clearly in non-medical and psychiatric populations.” Study author Martino Belvederri Murri is an associate professor at the University of Ferrara.
Scientists performed a systematic review and meta-analysis. This is a statistical method that combines data from multiple independent studies to identify overall trends. The research team searched major scientific databases for studies published through late 2025 that compared inflammatory biomarkers between people who used and did not use cannabinoids.
These included data from 46 studies with a total of 54,382 participants. Participants were diverse, including physically healthy people, people with mental disorders such as schizophrenia, and people with substance use disorders. The researchers extracted data on various immune markers from blood, serum, and plasma samples.
These markers include pro-inflammatory proteins that signal the body to increase immune activation and cause inflammation. Examples include C-reactive protein and various interleukins, which surge during infection or tissue damage. Researchers also looked at anti-inflammatory markers that work to calm the immune response and restore balance.
The included studies included 34 cross-sectional and case-control studies. This type of study examines data from a single time point and compares different groups. The scientists also included data from two prospective studies and four randomized controlled trials that followed people over long periods of time.
They utilized an advanced statistical approach known as a Bayesian multilevel model to synthesize 190 different effect sizes. This advanced mathematical approach allows us to account for the complex nature of our data. For example, it helped reconcile the fact that multiple different biomarkers are often measured within the same single research group.
This analysis revealed that regular use of cannabinoids in observational studies was associated with higher levels of both pro-inflammatory and anti-inflammatory biomarkers. For example, cannabinoid users tended to have increased numbers of white blood cells and neutrophils, cells that rush to the site of infection. At the same time, users showed increased levels of certain proteins specifically designed to suppress the immune system.
Rather than completely suppressing the immune system or driving it into a state of chronic inflammation, cannabinoids appear to exert complex regulatory effects. The researchers noted that the exact changes in these biological markers varied systematically depending on the study design and the specific demographics of the participants. Case-control studies that intentionally select patients with specific characteristics have yielded slightly different patterns of biomarker change than broad cross-sectional studies.
For example, recent cannabis use has higher levels of these biomarkers compared to past use. This suggests a direct and temporal relationship between exposure to active substances and fluctuations in the immune system. Researchers also found that synthetic cannabinoids were associated with much greater inflammatory effects than natural cannabis products.
Demographic factors also played a role, with the proportion of male participants and user age influencing the magnitude of immune changes. The research team emphasized that the user’s context is very important. They noted that anti-inflammatory effects are likely to be more predominant in the presence of overt medical disease, but that in otherwise healthy populations, the biological response is more complex.
In randomized controlled trials, researchers found a slightly different pattern. These experimental studies primarily tested cannabidiol, commonly known as CBD, in healthy people over short periods of time. Combined data from these studies suggested a slight increase in pro-inflammatory markers after CBD administration.
This finding provides evidence that contrasts with the common notion that CBD has strictly anti-inflammatory effects in humans without medical conditions. Most of the preclinical studies demonstrating anti-inflammatory effects have relied on animal models facing acute and severe inflammatory challenges. In healthy humans without an active infection or crisis, CBD appears to induce a mild rather than calming immune response.
“What was striking is that the human evidence does not clearly support an overall anti-inflammatory signal,” Belvedeli-Mulli told SciPost. “Instead, the pattern was mixed. One possible interpretation is that in the presence of overt medical disease, anti-inflammatory effects are likely to predominate, whereas in non-medical populations where this is not the case, the picture may be more mixed, with pro- and anti-inflammatory changes occurring simultaneously. That said, the evidence for anti-inflammatory biomarkers in our review is based on a small number of studies, so some of the picture remains more uncertain.”
Despite the large data set, the scientists urge readers to be cautious when translating these biological changes into everyday clinical meaning. The main limitation is that most of the included studies were observational in nature. This means that this study can only show an association and cannot prove that cannabinoid use directly causes the observed changes in immune markers. People who use cannabinoids often differ from non-users in several other ways that also affect the immune system.
“These findings should not be simplified to ‘cannabis causes inflammation’ or ‘cannabis reduces inflammation,'” Belvedeli-Mulli explained. “Most of the included studies were observational, so causality remains uncertain, and cannabinoid users may differ from non-users in many ways that also affect inflammation, such as smoking, alcohol or other drug use, body mass index, medications, and psychiatric comorbidities. Also, ‘cannabinoid use’ is a broad category that includes very different compounds and exposure patterns. ”
“An important next step is to better characterize exposure by better distinguishing between THC, CBD, synthetic cannabinoids, mixed products, dose, potency, and recency of use. Longer-term experimental studies, especially in well-characterized samples, are also needed to understand when cannabinoid-related immune changes are adaptive, neutral, or potentially harmful.”
These future studies will help determine whether cannabinoid-related immune changes are neutral, adaptive, or potentially detrimental to long-term health. As cannabinoid use continues to become more prevalent in society, understanding its true immunological footprint will be critical to predicting downstream clinical outcomes.
The study, “Regular use of cannabinoids and inflammatory biomarkers: a systematic review and hierarchical meta-analysis,” was authored by Martino Belvederi Murri, Riccardo Guglielmo, Alessio Zizzi, Angela Muscettola, Maria Giulia Nanni, Manuela Dall’Oro, Gianluca Serafini, Alberto Inuggi, Andrea Escelsior, Mario Amore, and Luigi. Grassi.
