The Skin Has Its Own Cannabinoid Receptors. That Is Not the Same as Topicals Working.

A 2005 immunohistochemical study of human skin found CB1 and CB2 immunoreactivity in cutaneous nerve fiber bundles, mast cells, macrophages, epidermal keratinocytes, and the epithelial cells of hair follicles, sebaceous glands, and eccrine sweat glands. In keratinocytes, hair follicles, and sebaceous glands, the two receptors were distributed in a complementary pattern rather than an overlapping one.

Receptor presence establishes that a tissue can respond to a cannabinoid if one reaches it. It does not establish that any commercially available topical delivers enough cannabinoid to that tissue, or that doing so improves a skin condition.

The cannabis skincare category is built on a single scientific claim, repeated everywhere: the skin has an endocannabinoid system. The claim is accurate. What gets built on top of it usually is not.

Knowing which cells carry which receptor, what activating them does in laboratory models, and what happens when the same idea is tested in a person lets a clinician answer the question a patient is actually asking, which is whether the product on their counter is doing anything.

Sonja Ständer and colleagues at University Hospital Münster mapped cannabinoid receptor distribution in sections of human skin using immunohistochemistry, publishing in the Journal of Dermatological Science in 2005. CB1 and CB2 immunoreactivity appeared in cutaneous nerve fiber bundles, mast cells, macrophages, epidermal keratinocytes, and the epithelial cells of hair follicles, sebocytes, and eccrine sweat glands.

Two details from that work carry real interpretive weight. In epidermal keratinocytes, hair follicles, and sebaceous glands, CB1 and CB2 were distributed in a complementary rather than identical fashion, which suggests distinct roles rather than redundancy. And double-staining with an antibody against calcitonin gene-related peptide indicated cannabinoid receptors on small peptidergic afferent nerves, the fibers most relevant to itch and to pain.

The receptor picture in cutaneous immune cells has an older and quantitatively stronger foundation. Sylvaine Galiègue and colleagues, publishing in the European Journal of Biochemistry in 1995, measured CB1 and CB2 transcripts across human tissues and blood cell subpopulations. CB2 expression in immune tissue ran 10 to 100 times higher than CB1, and among blood cell subsets the rank order was B cells, then natural killer cells, then monocytes, then neutrophils, then CD8 and CD4 T cells.

That gradient is the anatomical basis for the claim that cannabinoids are immunomodulatory. It is a measurement of receptor message abundance, not a measurement of a clinical effect, and the distinction holds throughout this topic.

Tamás Bíró and colleagues, writing in Trends in Pharmacological Sciences in 2009, proposed the organizing idea that has framed this field since: the main physiological function of the cutaneous endocannabinoid system is to constitutively control balanced proliferation, differentiation, and survival of skin cells, together with immune competence and tolerance.

On that model, the system is a set point rather than a switch. Disruption of the balance is proposed to contribute to acne, seborrhea, allergic dermatitis, itch and pain, psoriasis, hair growth disorders, systemic sclerosis, and cutaneous cancer. That is a hypothesis about disease mechanism, offered as such, and it is frequently repeated as though it were a list of conditions cannabinoids treat.

The most concrete laboratory demonstration involves the sebaceous gland. Attila Oláh and colleagues, publishing in the Journal of Clinical Investigation in 2014, applied cannabidiol to cultured human sebocytes and human skin organ culture and found it suppressed lipid production and sebocyte proliferation. The mechanism they identified is worth noting, because it is not a cannabinoid receptor: the antiproliferative effect ran through TRPV4 ion channels, and the anti-inflammatory effect through A2a adenosine receptor dependent upregulation of TRIB3 with inhibition of NF-kappaB signaling.

That finding is genuinely interesting, and it is a cell culture and organ culture result. The authors concluded that cannabidiol has potential as a therapeutic agent for acne. Potential is the operative word, and more than a decade later the confirmatory human trial has not appeared.

Andrea Telek and colleagues, publishing in the FASEB Journal in 2007, used human scalp hair follicle organ culture to test cannabinoid effects on hair growth. Human hair follicles turned out to be both sources and targets of endocannabinoids.

The direction of the effect is the part that matters. Anandamide and delta-9-THC dose-dependently inhibited hair shaft elongation and the proliferation of hair matrix keratinocytes, induced intraepithelial apoptosis, and drove premature follicle regression into catagen. A selective CB1 antagonist blocked those effects, and CB1 was expressed in the follicular epithelium in a hair-cycle-dependent manner while CB2 was not.

The authors drew the logical clinical inference in both directions: CB1 agonists might help manage unwanted hair growth, while CB1 antagonists might counteract hair loss. Read that carefully. In this model, cannabinoid receptor activation suppressed hair growth.

Cannabis-based hair growth products are a substantial retail category. The most direct experimental evidence on cannabinoids and human hair follicles points the other way, and any product claiming otherwise should be asked which study it is relying on.

A 2022 review in JID Innovations by Torunn E. Sivesind and colleagues screened the dermatologic literature for clinical use of cannabinoids. From 248 nonduplicated studies, 26 articles met inclusion criteria, with 13 concerning systemic cannabinoids and 14 concerning topical cannabinoids. Their conclusion was that available evidence suggests cannabinoids may be effective for various inflammatory skin disorders and that additional research is needed to evaluate efficacy and determine dosing, safety, and treatment guidelines.

The specific positive findings they identified are mostly systemic rather than topical: selective CB2 agonists in diffuse cutaneous systemic sclerosis and dermatomyositis, dronabinol for trichotillomania, and sublingual cannabidiol and THC for pain associated with epidermolysis bullosa. Those are systemic exposures with a route that reliably reaches the target.

One of the better-designed topical studies is a pilot. Erika McCormick and colleagues, publishing in the Journal of the American Academy of Dermatology in 2024, randomized blinded buttock sites in 19 participants to nanoencapsulated cannabidiol cream or vehicle twice daily for 14 days before ultraviolet A irradiation. The cannabidiol sites showed less ultraviolet-induced epidermal hyperplasia and reduced ultraviolet-induced mitochondrial DNA deletions. It is a mechanistic photoprotection endpoint in 19 people, described by its authors as limited by sample size.

The separate question of what actually crosses skin into the bloodstream has been measured, and we cover it in detail in our review of how much cannabinoid reaches the blood by method. The short version: in a controlled study of commercial hemp topicals, cannabidiol was measurably absorbed from some products after days of repeated use, delta-9-THC stayed below detection, and the participants showed no discernible subjective or physiological effects.

The anatomy and the efficacy question are separate, and a patient deserves both answers rather than whichever one supports the product in front of them.

Yes, the skin has cannabinoid receptors. They sit on keratinocytes, sebocytes, follicular epithelium, sweat gland epithelium, sensory nerve endings, mast cells, and macrophages. The endocannabinoids that act on them are produced locally. That is settled anatomy, not marketing.

No, that does not establish that a cream you can buy delivers a meaningful concentration to those cells, or that doing so improves acne, eczema, psoriasis, or hair growth. The dermatology literature contains a small number of studies, mostly small, mostly uncontrolled, with the stronger positive results coming from systemic routes.

For a patient with a diagnosed skin condition, the practical advice is unchanged by any of this. Treat the condition with what has evidence behind it. A topical cannabinoid used alongside established treatment is a low-risk experiment, and one used instead of established treatment for psoriasis, eczema, or acne is a poor trade.

The anatomical evidence is strong and old enough to be uncontroversial. Immunohistochemical localization in human tissue sections and quantitative transcript measurement across human blood cell subpopulations are appropriate methods for the question of where receptors are, and the findings have been reproduced by multiple groups over two decades.

The therapeutic evidence is weak. The sebocyte and hair follicle findings come from cell and organ culture, which can establish mechanism and cannot establish clinical effect. The dermatologic clinical literature is small, heterogeneous, and weighted toward case reports and small series, with its clearest positive results arising from systemic rather than topical administration.

Receptor presence is the weakest possible premise for a therapeutic claim, because receptors are nearly everywhere. Opioid receptors are present in skin as well, and nobody argues that this makes a topical opioid cream an established dermatologic treatment.

The cannabidiol sebocyte result is often cited as evidence that cannabidiol treats acne. The mechanism the investigators identified runs through TRPV4 and adenosine receptors rather than CB1 or CB2, which means the finding is not straightforwardly an endocannabinoid system result at all, and it was obtained in culture at concentrations that a topical product may never achieve in a living sebaceous gland.

The hair growth literature is the clearest case of marketing running against the primary data. In the best available human follicle model, cannabinoid receptor activation shortened the growth phase rather than extending it.

None of this shows that a commercially available cannabis topical treats acne, eczema, psoriasis, or hair loss. The cell and organ culture findings identify mechanisms worth testing, and the confirmatory human trials have not been run.

It does not show that topical cannabinoids reach the dermal or follicular targets at concentrations comparable to those used in laboratory models. Absorption studies of commercial topicals measured systemic exposure, not tissue concentration at the receptor.

It does not establish which cannabinoid, at what concentration, in what vehicle, applied how often, would be worth testing. Those formulation questions remain open, and the products on the market were not designed to answer them.

Dermatology has been through this before with other plant-derived actives, where a compelling in vitro mechanism drove a consumer category for years before controlled trials arrived. Some of those compounds survived the transition and most did not, and the ones that survived did so on the strength of randomized data rather than on receptor maps.

The cutaneous endocannabinoid system is likely to remain scientifically important regardless of what happens to the product category, because it sits at the intersection of barrier function, itch signaling, sebaceous biology, and cutaneous immunity. Those are central problems in dermatology, and a system that touches all four deserves the research attention it is getting.

When a patient tells me the skin has an endocannabinoid system, they are right, and I say so. The receptors are there, the endocannabinoids are made locally, and this is settled biology rather than a talking point someone invented.

Then I ask what they are treating. If it is dry irritated skin and a CBD balm feels good, I have no argument with that, and most of the benefit is probably the emollient base doing what emollient bases do. If it is psoriasis or moderate acne and they have replaced their prescription with a cannabis cream, that is a bad trade, and the literature does not support it.

The hair growth products are the ones I find hardest to be polite about. The clearest experimental evidence in human hair follicles shows cannabinoid receptor activation pushing follicles out of the growth phase. Somebody is selling the opposite of what the study found.

The skin endocannabinoid system is documented anatomy: CB1 and CB2 on keratinocytes, sebocytes, follicular and sweat gland epithelium, sensory nerves, mast cells, and macrophages, with CB2 heavily concentrated in immune cells. That anatomy makes cutaneous cannabinoid pharmacology worth studying. It does not make any commercial topical an evidence-based treatment, and for hair growth the primary data point the opposite way.

Carry forward the receptor map, which is solid, and the observation that the strongest clinical dermatology results came from systemic rather than topical routes. Do not carry forward the sebocyte culture finding as evidence that cannabidiol treats acne in people, and do not carry forward any hair growth claim, which the primary human follicle data contradict.