The Journal of Alternative and Complementary Medicine
Published Online: 7 May 2020
https://doi.org/10.1089/acm.2020.0144
Editor’s Note: For those whose response to COVID-19 includes exploring beyond vaccines, conventional pharmaceuticals, and the watchful or healthy waiting until such tools might arrive, interest in cannabinoids has been high – and controversial. It has already stimulated one journal, the Liebert Cannabis and Cannabinoid Research, to issue a call for papers on COVID-19. The unique place of cannabis in the culture seems to always mark the herb with an exponential asterisk whenever basketed with the other natural health strategies that are both widely used, and as broadly derided. In this invited commentary, JACM Editorial Board member Michelle Sexton, ND starts by describing the multiple immune modulating effects associated with the herb. The University of California San Diego Assistant Adjunct Professor in Anesthesiology then asks: “Given these effects, can phytocannabinoids be either helpful, or harmful for immune competency, in the context of the current COVID-19 pandemic?” A skilled edge-walker, Sexton lets the research fall where it may in wending a path through this evidentiary maze. —John Weeks, Editor-in-Chief, JACM
The endogenous cannabinoid signaling system (ECS) is a highly conserved, ubiquitous, pleiotropic biochemical system known as a gatekeeper in immune homeostasis.1 A multitude of ECS-mediated immunosuppressive effects have been demonstrated to date, including inhibition of immune cell proliferation, migration and antibody production, induction of apoptosis, and cytokine suppression (via downregulation of immunoregulatory genes). Given these effects, are phytocannabinoids helpful or harmful for immune competency in the context of the current coronavirus disease 2019 (COVID-19) pandemic?
The plant cannabinoid delta9-tetrahydrocannabinol (THC) mimics the actions of endogenous cannabinoids (ECB) as a nonselective partial agonist (higher affinity than ECBs) at cannabinoid receptors 1 and 2 (CB1 and CB2). Both receptors are expressed on immune cells, with CB2 exclusively expressed in human immune cells and tissues. Agonism of these receptors on immune cells has been shown to reduce the production and secretion of inflammatory mediators.2 Cannabidiol (CBD) acts as a negative allosteric modulator with very low affinity at both cannabinoid receptors. CBD has also been shown to be immunosuppressive through diverse (non-ECS-mediated) mechanisms.3 The dietary sesquiterpenoid beta-caryophyllene, found in Cannabis and other plants, and alkylamides in Echinacea purpurea activate the CB2 receptor (Ki: 100 and 60 nM affinity, respectively).4,5
Generally speaking, the role of CB2 is underexplored relative to the role of CB1, particularly with regard to acute, innate immune responses. Activation of CB2 is associated with intracellular pathways that tone down immune responses. Because of this, CB2 agonists may hold promise as therapeutic agents in autoimmune diseases by suppression of antibody production through T-cell mechanisms.6
The innate immune response is nonspecific, intended to prevent the spread of infection through chemical and cellular mechanisms. Toll-like receptors (TLR), expressed on the surface of macrophage and dendritic cells, recognize commonly conserved pathogen-associated molecular patterns. Upon recognition, a pro-inflammatory signaling cascade (via cytokine release) is triggered that eventually dictates lymphocyte involvement (adaptive immunity). Cytokine release is mediated by several pathways known to be affected by the ECS: NF-κB, MAPK, and JAK-STAT.7
Cannabinoids have been shown to inhibit cytokine production in monocyte cell cultures and in animal models of acute infection, primarily through inhibition of TLR4-induced activation. This has also been demonstrated in human subjects who smoke cannabis, indicating that cannabinoids may impair TLR-induced immune activation.8–10Further, THC has been shown via CB2 to inhibit the macrophage co-stimulatory signaling required for T-cell activation, thus impairing the adaptive immune response (antibody production and immune memory).6 Early investigations suggested a TH1 to TH2 shift by cannabinoids, but this has not been replicated in humans with multiple sclerosis—a condition that would benefit from this shift.10
A systems-based analysis of the ECS biologic network revealed that tumor necrosis factor alpha (TNF-α) is one of the major nodes, or units connecting to other signaling units, in the network. Using data published from 2003 to 2013, investigators built a database of elements (including protein receptors, ECB, and other ligands) with connections to the two primary ECBs: AEA and 2AG. Calculations of node connectivity revealed TNF-α to be one of the eight most highly connected nodes in the ECS system topology. TNF-α is a pleiotropic cytokine that has a pivotal role as the master regulator in the pro-inflammatory cytokine cascade (primarily via the NF-κB pathway) by activating macrophage cells. Thus, TNF-α is crucial for promoting the acute phase reaction in immune cells and has been shown to have a central role in ECS signaling.11
Overall, these immunomodulatory effects warrant further exploration of the ECS either for treating chronic inflammation/autoimmune diseases or for potential impacts on healthy host immunocompetence.
There has been some media hype about Cannabis for “treating” the COVID-19-associated cytokine storm, a fatal immune dysregulation during the course of disease. While there is a paucity of human data on acute viral infections and cannabinoids, in vitro and in vivo studies shed some light on their role in immune suppression in viral influenza illness.
For example, mice infected with attenuated influenza A were administered intraperitoneal (IP) THC (75 mg/kg; wild type vs. CB1/2 knockout). The percentage of CD4+ (but not CD8+) cells increased, while the percentage of natural killer (NK) cells decreased, in bronchoalveolar lavage fluid of wild-type mice but not knockout mice. Also, THC significantly suppressed interleukin (IL)-17-producing NK cells (needed to target and kill infected lung epithelial cells) in the wild-type mice. Further, THC suppressed the ability of macrophages and dendritic cells to migrate into the lungs. This effect was only observed in wild-type mice. Functionally, THC also attenuated interferon gamma production, which is a critical lymphocytic cytokine in immune responses to viral infections. Cannabinoids may be a therapeutic strategy in certain chronic viral infections (that invade the central nervous system), but other viral studies have shown increased viral replication and disease pathology.12
These data illustrate how cell populations necessary for a healthy immune response are affected by THC in a CB1/2-dependent manner, impairing migration of antigen-presenting cells to the lungs, subsequent cytokine production, and antigen presentation needed for T-cell responses for healthy adaptive immunocompetence. Collectively, suppression of host immunity against influenza was demonstrated in this mouse model, although it should be noted that the dose of THC was administered IP at a supraphysiologic dose.
Similar to tobacco smoking, chronic cannabis smoking can lead to long-term effects of increased cough, sputum production, and wheeze, along with airway disease such as chronic bronchitis and decline in lung function. Additionally, data from healthy, adult-use cannabis smokers demonstrated a global reduction in cytokine production.8–10 Further, the use of vape pens (vaporization devices with concentrated cannabis) may pose an even greater risk through concentration, adulteration, or contamination of the extracts. The concentration of cannabinoids and terpenes can be increased by 3.2- to 4-fold and 2.7- to 8.9-fold, respectively (depending on extraction process and terpene structure), and this concentrated form may also contribute to respiratory symptoms and dysfunction.13 Avoidance of smoking cannabis and vaping of concentrates is particularly relevant for pulmonary health in light of COVID-19. These administration methods may diminish the respiratory system’s efficacy in responding to infection and thereby increase the risk of rapid progression to hypoxemia.
Cannabis has a long history of relatively safe use as a botanical medicine, with therapeutic benefit achievable at doses below the threshold for intoxication. THC dosing for recreational use versus therapeutic use can be widely divergent. Cannabinoids are known to have bi-phasic effects, and higher doses are commonly associated with adverse events.8,14
A recent survey of health care professionals specializing in cannabinoid medicine reported 44.7% recommend 6–10 mg/day of THC for chronic pain, fibromyalgia, arthritis, sleep disorders, anorexia, and other conditions—an average dose of 2–3.3 mg per dose (t.i.d.).15 This dose is significantly less than recreational dosing and miniscule compared to 75 mg/kg (used in the aforementioned mouse study).16 This translates to a human equivalent dose of 360 mg for a 60 kg individual. Clearly, THC dose is a critical factor in assessing clinical impact.
CBD, primarily from hemp-based products, is also being touted as a potential treatment for COVID-19. CBD has been shown to have anti-inflammatory effects, as demonstrated in an animal model of collagen-induced arthritis, by suppressing lymphocyte and macrophage functions.17 Without high-quality evidence in humans, however, effective anti-inflammatory doses for CBD are unknown. Based on effective doses of CBD in the existing clinical literature, an anti-inflammatory dose is likely to be quite high, and not approximated in most of the hemp-based products currently marketed. Table 1 presents a sample of completed and recruiting clinical trials of CBD for various therapeutic purposes (referenced from a search in clinicaltrials.gov), showing the range of what is considered to be therapeutic dosing. (Only two trials for anti-inflammatory effects were identified.) Because CBD is principally available as an unregulated dietary supplement ingredient, it is difficult for patients and their doctors to know exactly what is contained in the products they are purchasing.15
| Sample of completed trials with CBD (from search in PubMed) | ||||
|---|---|---|---|---|
| CBD dose | Condition | Author, year | Size of study | Type of study |
| 5–50 mg/kga | Treatment-resistant epilepsy | Szaflarski et al., 201818 | N = 72 (child) | Open label |
| N = 60 (adult) | ||||
| 600 mg/day, fixed dose | Schizophrenia | Boggs et al., 201819 | N = 36 | Randomized, placebo controlled, parallel group, fixed dose |
| 600 mg, single dose | Public-speaking anxiety model | Bergamaschi et al., 201120 | N = 12 | Double blind, placebo controlled |
| 200 mg, q.i.d. | Schizophrenia | Leweke et al., 201221 | N = 33 | Double blind, randomized, parallel group |
| Up to 20 mg/kg/day | Dravet syndrome | Devinsky et al., 201722 | N = 34 | Double blind, placebo controlled |
| 1000 mg/day | Schizophrenia | McGuire et al., 201823 | N = 43 (T) | Double blind, randomized, placebo controlled, parallel group |
| N = 45 (C) | ||||
| Up to 50 mg/kg/day | Severe, intractable, childhood-onset, treatment-resistant epilepsy | Devinsky et al., 201624 | N = 162 (safety and tolerability) | Open label |
| N = 137 (efficacy) | ||||
| 20 mg/kg/day | Lennox–Gastaut syndrome | Thiele et al., 201825 | N = 86 (T) | Randomized, double blind, placebo controlled (Phase III) |
| N = 85 (C) | ||||
| 10 or 20 mg/kg/day | Drop seizure in Lennox–Gastaut syndrome | Devinsky et al., 201826 | N = 225 | Double blind, placebo controlled |
| 20 mg/kg | Drug-resistant seizure in Dravet syndrome | Devinsky et al., 201722 | N = 120 | Double blind, placebo controlled |
| 75 or 300 mg/day | Parkinson’s disease | Chagas et al., 201427 | N = 119 | Double blind, placebo controlled |
| Single ascending dose from 1500 to 6000 mg; multiple dose 750 or 1500 mg; 1500 mg single dose | Healthy subjects | Taylor et al., 201928 | N = 6–12 per arm | Phase I open label |
| 50–150 mg b.i.d. | Chronic pain in kidney transplant patients | Cunetti et al., 201829 | N = 7 | Open-label |
| 750, 1500, or 4500 mg dose compared to alprazolam and dronabinal | Healthy recreational polydrug users | Schoedel, 201830 | N = 43 | Single dose, randomized, double blind, double dummy, placebo and active controlled crossover design |
| 600 mg | High risk of clinic psychosis | Battacharyya et al., 201831 | N = 33 | Single dose, parallel group, double blind placebo controlled |
| Sample of trials listed on clinicaltrials.gov (currently recruiting), N = 43 | ||||
| Dose | Condition | Preliminary investigator | Size of study | Type of study |
| Starting at 25 mg/b.i.d. and increasing to 150 mg/b.i.d. | Steroid-sparing effects in stable autoimmune hepatitisb | Stero Biotechs Ltd. | N = 15 | Open label, Phase II |
| Starting at 25 mg/b.i.d. and increasing to 150 mg/b.i.d. | Steroid-sparing effects in Crohn’s diseaseb | Stero Biotechs Ltd. | N = 28 | Open label, Phase IIa, randomized, crossover, placebo controlled |
| Sublingual, 10 mg/t.i.d. (30 mg/day) | Anxiety | Stacey Gruber, MD | N = 16 | Phase I: open label; Phase II: placebo controlled |
| 150–300 mg/day | Bipolar | Márcia Kauer-Sant’Anna, MD, PhD | N = 100 | Double blind, randomized, placebo controlled |
| 600 mg/day | Alcohol use disorder in patients with PTSD | Charles Marmar, MD | N = 48 | Double blind, randomized, parallel |
| 2.5–5 mg/kg up to 20 mg/kg/day (in divided doses) | Gastropareisis and functional Dyspepsia | Michael Camilleri, MD | N = 96 | Randomized, double blind, parallel design |
| 20–40 mg/kg/day | Prader–Willi syndrome | Ahmed Elkashef, MD | N = 66 | Open label (safety trial) |
| 20 mg/kg/day | Chronic back pain | Jodi Gilman | N = 20 | Open label |
| 400 or 800 mg/day | Cocaine addiction | N = 110 | Randomized, parallel design, placebo controlled | |
| Up to 2.5 mg/kg/day | Motor symptoms in Parkinson’s disease | Maureen Leehey | N = 60 | Parallel design, double blind, randomized controlled trial |
| 3, 6, or 9 mg/kg/day | Autism spectrum disorder | Francisco Castellanos, MD | N = 30 | Open label, single group, Phase II study |
| 75, 150, or 300 mg (p.o.) b.i.d. | Graft-versus-host disease | Ram Ron, MD | N = 36 | Open label |
| 300 mg/day | Post-traumatic stress disorder | Michael J. Telch, PhD | N = 120 | Randomized |
In summary, there are dichotomous effects of phtyocannabinoids on immunocompetence. On the one hand, targeting the ECS for anti-inflammatory benefits in chronic inflammatory/autoimmune disease in humans may be a possibility, although effective dosing has not yet been documented. On the other hand, detrimental effects in the setting of acute infection may also be a possibility, although dosing guidelines to avoid immunosuppression in humans has not been documented.
When interacting with patients on the topic of cannabis use, respiratory and immune health education on mode of administration and dosing is critical. A harm-reduction approach for individuals smoking cannabis would be substituting orally administered products at low doses (<5 mg THC suggested)15 or using vaporized dried flower material (to avoid byproducts of combustion). For patients using low-dose oral products, it appears likely that clinically significant immunosuppression is not a risk. However, this should still be on health care professionals’ radar. An individualized approach to assessing the patient, including respiratory and cardiovascular health or existing immunocompromise (e.g., cancer patients, use of biologic drugs), should guide the health care provider. Reduced innate defense is being considered as a driving feature of COVID-19, and to be clear, no data suggest that THC or CBD is a proven therapeutic intervention for treating COVID-19.
Acknowledgments
Thanks to Dr. Jamie Corroon for his help with preparation of the manuscript.
Author Disclosure Statement
Dr. Sexton is a member of the scientific advisory board for Versea LLC.
Funding Information
No funding was received for this article.
