CBD and acne: from sebum in the lab to a 368-patient trial
In the laboratory, CBD curbs excess lipid production by human sebocytes, the cells that make sebum. In humans, the largest registered controlled trial gave 368 participants a 2.5% or 5% CBD solution for 12 weeks, and it did no better than the same solution without CBD. A CBD cosmetic can promise nothing against acne, and the law forbids it to.
Key takeaways
- A 2014 study found that CBD returned the lipid output of stimulated sebocytes to normal, in cell culture and in donated human skin.
- The effect does not go through the CB1 and CB2 cannabinoid receptors.
- In the BTX 1503 phase 2 trial, inflammatory lesions fell by 8.5 to 12.0 with CBD and by 11.3 with the CBD-free solution.
- The trial used 13 to 26 times the 0.19% that the European Scientific Committee on Consumer Safety considers safe in a cosmetic.
- In law, a product presented as treating acne is a medicine.
What the 2014 sebocyte study shows
Oláh and colleagues published the reference study in 2014 in the Journal of Clinical Investigation. The team used a human sebocyte line (SZ95) and whole human skin kept in culture. After stimulating lipid production with “pro-acne” agents, they added CBD. CBD brought this abnormally high lipogenesis back to normal, slowed cell multiplication without killing the cells and lowered the expression of several inflammatory mediators.
The result comes from a laboratory. Cultured cells and donated skin cannot reproduce the length of a treatment, the way a product crosses the skin barrier or the other factors behind acne. Such a finding is a reason to run human trials, and it tells you nothing about what a cream does.
The receptors involved: CB2 present, CB1 absent
A common claim is that sebocytes “carry CB1 and CB2 receptors”, which is inaccurate. A 2008 study in the FASEB Journal showed that SZ95 sebocytes express the CB2 receptor but not CB1. In these cells, the endocannabinoids the body makes raise lipid production through CB2.
CBD takes another route. In the 2014 study, blocking CB1 and CB2 left its effect intact. The authors point to two other targets: the TRPV4 ion channel for the effect on lipids, and the adenosine A2a receptor for the anti-inflammatory effect they observed. Both mechanisms were described in cells, and they tell you nothing about what happens on a face.
The BTX 1503 trial: 368 participants, 12 weeks
A controlled trial is the most direct test. The BTX 1503 phase 2 trial, registered on ClinicalTrials.gov as NCT03573518, enrolled 368 people aged 12 to 40 with moderate to severe facial acne. Each of its four groups of 92 received one regimen: 5% CBD solution twice a day, 5% CBD once a day, 2.5% CBD once a day, or the vehicle, meaning the same solution without CBD. The trial ended in September 2019, and its results were posted on 18 April 2022.
| Group | Change in inflammatory lesions at day 84 |
|---|---|
| CBD 5%, twice a day | −8.5 |
| CBD 5%, once a day | −12.0 |
| CBD 2.5%, once a day | −11.7 |
| Vehicle without CBD | −11.3 |
All groups improved, the vehicle group too. A trial includes a vehicle group for this very reason: the figure that counts is the gap between the product and the same formula without CBD. Here that gap is at most 0.7 lesion in favour of CBD, against a standard deviation of about 12.6 lesions in each group. The highest-dose group even did slightly worse than the vehicle. On this measure, CBD brought no benefit.
0.19%: the safety ceiling for CBD cosmetics
A cosmetic is not a medicine, and its CBD content has a safety limit. On 24 April 2026, the European Scientific Committee on Consumer Safety (SCCS) published opinion SCCS/1685/25. It considers CBD safe up to 0.19% in cosmetics applied to the skin or used in the mouth, and THC present as an impurity up to 0.00025%. Products that can be inhaled, such as sprays, fall outside the opinion, and the committee notes that the data submitted were limited.
Set the two figures side by side. The BTX 1503 trial used 2.5% and 5% CBD, 13 to 26 times that ceiling. A cosmetic cream sold in France therefore holds at most a fraction of a dose that did no better than the vehicle. On the label, CBD appears in the INCI list as “cannabidiol”, and its concentration need not be stated. Our guide to assessing a CBD cream, balm or gel shows how to read the rest of the label.
What a CBD cosmetic may say, and what it may not
Three texts draw the line. In its Article 20, Regulation (EC) No 1223/2009 forbids using text, names, trade marks or pictures to attribute to a cosmetic characteristics or functions it does not have. Regulation (EU) No 655/2013 sets six criteria for every claim: legal compliance, truthfulness, evidential support, honesty, fairness and informed decision-making.
The third text shifts the boundary. Under Article L5111-1 of the French Public Health Code, any substance “presented as having curative or preventive properties with regard to human diseases” is a medicine. Acne vulgaris is a skin disease, so a product presented as “anti-acne” or “clears spots” leaves the cosmetic field. CBD also has a rule of its own: the MILDECA points out that products containing CBD may not make any therapeutic claim, on pain of criminal penalties.
| Wording | Status |
|---|---|
| “Light texture, matte finish” | Cosmetic claim, to be backed by tests |
| “For oily-prone skin” | Cosmetic claim, to be backed by tests |
| “Regulates sebum” | Effect claim, which needs solid evidence |
| “Anti-acne”, “clears spots” | Forbidden: presentation of a medicine |
The six criteria are taken one by one in our article on permitted claims for CBD cosmetics.
When acne is a matter for a doctor
If acne scars, spreads or lasts, see a doctor or dermatologist; prescription treatments exist, both topical and oral. No CBD medicine is authorised for acne. Epidyolex, the only cannabidiol-based medicine authorised in France, is reserved for certain rare forms of epilepsy.
A CBD cosmetic is a care and comfort product. Choose one for its texture, how well your skin tolerates it and its ingredient list, and expect no effect on acne. The same caution applies to other reactive skin, as our article on CBD, eczema and psoriasis explains.
Frequently asked questions
In cultured sebocytes and donated human skin, yes: CBD returns artificially stimulated lipid production to normal. In people with acne, the largest phase 2 trial found no difference in inflammatory lesions compared with the same solution without CBD.
No. Under Article L5111-1 of the French Public Health Code, a product presented as treating or preventing a disease is a medicine, and CBD products may make no therapeutic claim.
The European scientific committee considers CBD safe up to 0.19% in cosmetics applied to the skin or used in the mouth. Labels need not show the concentration, so ask the seller when it is missing.
The 2014 study points to the TRPV4 channel for lipid production and the A2a receptor for inflammation; the CB1 and CB2 receptors play no part. The sebocytes studied express CB2 and lack CB1.
No. Epidyolex, the only authorised cannabidiol-based medicine, is indicated for certain forms of epilepsy.
Written by the Phytogrammes team
Each journal article draws on cited primary sources: ANSM, EFSA, EUR-Lex and peer-reviewed publications. Product profiles are distinguished from documented laboratory results. Our approach.
Article published on . CBD is not a medicine.