Psoriasis, itching, herpes, sickle cell disease, leukaemia and MRSA all come up in claims that circulate online, each backed by a sentence or two. We put them in one article on purpose, because none of them has enough evidence for a full write-up. What they do have ranges from one small study to an unsourced assertion, and the point of this article is to tell those apart.
One safety warning belongs at the top. Claims circulating online suggest that external ulcers, warts and “abnormal moles” can be removed by applying cannabis oil under an occlusive dressing. Do not do this. An abnormal mole may be a melanoma. Destroying a skin lesion at home eliminates the tissue a pathologist needs to diagnose it, and it does nothing about cells that may already have spread. Any changing, irregular, bleeding or newly pigmented skin lesion needs a doctor, not a dressing. This is the most likely way anything in this article could cause serious harm.
Nothing else here is medical advice either. In Thailand, cannabis flower is a controlled herb (Notification of the Ministry of Public Health Re: Controlled Herbs (Cannabis) B.E. 2568) and may only be supplied against a prescription/certificate from an authorised practitioner (PT 33).
1. Chronic Itch (Pruritus)
The evidence. Szepietowski and colleagues at Wrocław Medical University in Poland (2005) studied 21 haemodialysis patients with itch caused by end-stage kidney disease. The patients applied a cream twice daily for three weeks, and itch disappeared completely in eight of them. The cream contained structured physiological lipids with endocannabinoid-like lipids, which are fatty-acid molecules of the kind the body makes itself, such as palmitoylethanolamide (PEA). It was not a cannabis-derived product and contained no plant cannabinoid.
How to read it. It was a small, uncontrolled preliminary study with no placebo arm, run over three weeks in one specific cause of itch. An emollient base alone relieves itch for many people, so the base cream and the active lipids cannot be separated. Topical itch is discussed further in community topic 1310.
Status: small, uncontrolled, not a cannabis product — indirect relevance to cannabis.
2. Psoriasis
The claim. Cannabis cream applied regularly helps psoriasis because cannabinoids are anti-inflammatory and modulate the immune system.
How to read it. Cannabinoid receptors are involved in immune regulation, and psoriasis is an immune-mediated skin disease, so the hypothesis makes sense. But no randomised controlled trial has shown that any cannabinoid works for psoriasis, and none is licensed or recommended in guidelines for it.
A caution about how this is often presented. Social-media posts sometimes show before-and-after photos that set a conventional drug such as methotrexate, with its side effects listed, against a few days of home-applied cannabis oil. That is a marketing device, not evidence. It is uncontrolled and unblinded, there is no way to know what else changed, and it is framed to make the established treatment look like the harmful one. Psoriasis also flares and settles on its own, which is exactly when before-and-after images mislead.
Modern psoriasis treatment includes topical therapies, phototherapy and biologics, all with substantial trial evidence behind them.
Status: plausible mechanism, no controlled evidence.
3. Herpes Simplex
The claim. THC treats herpes, speeds healing of blisters and reduces how often outbreaks happen.
How to read it. The research record points both ways, and no human trial has been done.
- Cell culture: Blevins and Dumic (Journal of General Virology, 1980) found that THC stopped herpes simplex virus types 1 and 2 from replicating in cultured cells.
- Animals: Cabral and colleagues (Proceedings of the Society for Experimental Biology and Medicine, 1986) found that THC-treated guinea pigs infected with HSV-2 developed more severe genital disease than untreated controls. A mouse study by the same group found that THC lowered the interferon response to the virus.
So a lab-dish result is being quoted while the animal evidence points the other way. That fits a real concern: herpes recurrence is held in check by cell-mediated immunity, and cannabinoids can suppress immune responses. Herpes also has effective, cheap, well-evidenced antiviral treatment: aciclovir and related drugs shorten outbreaks and, taken suppressively, make them less frequent.
Status: in vitro inhibition, animal evidence of harm, no human trial. Effective conventional treatment exists.
4. Sickle Cell Disease
The claim. Cannabis does not cure sickle cell anaemia but controls the pain, acts as an anti-inflammatory “without side effects”, and may reduce retinopathy and neuropathy.
How to read it. The first part is framed properly: it claims symptom control, not a cure. Sickle cell pain is a real and poorly served problem.
A randomised trial exists. Abrams and colleagues (JAMA Network Open, 2020) gave 23 adults with sickle cell disease and chronic pain vaporised cannabis with roughly equal THC and CBD, or a placebo. Each participant went through two five-day inpatient sessions and acted as their own control.
Results: pain was not significantly lower on cannabis than on placebo. Only one secondary outcome reached statistical significance: pain interfered less with mood. Pain interfering with walking and sleep also improved, but those were trends, not significant results. The authors called for larger studies.
“No side effects” is not accurate for any cannabinoid. The retinopathy and neuropathy claims have no human evidence.
A regional note. Sickle cell disease is rare in Thailand and Southeast Asia, where the common inherited blood disorders are thalassaemia and haemoglobin E disease. This trial tells us nothing about those conditions. Readers affected by them should talk to a haematologist.
Status: one small randomised trial; primary pain outcome not significant, only mood interference significant.
5. Leukaemia
CB2-mediated cell death in leukaemia cell lines is real preclinical work. However, such studies typically use concentrations around 10 μM, well above the levels reached in human blood after cannabis use, and no clinical evidence shows that “cannabis oil” treats leukaemia. The claim is especially hazardous for patients whose immune systems are severely weakened, because moulds and bacteria in unregulated cannabis are a documented risk for them. The dose gap and immune effects are discussed in community topics 1308 and 1314.
Status: preclinical only. The therapeutic claim is unsupported and hazardous.
6. MRSA
Appendino and colleagues (Journal of Natural Products, 2008) found that the major cannabinoids are potent against methicillin-resistant Staphylococcus aureus in the lab, apparently through mechanisms unrelated to conventional antibiotic resistance. Eighteen years later there is still no licensed cannabinoid antibacterial and no trial in infected patients.
Status: real in vitro finding, no human evidence, no product.
7. The Summary Table
| Condition | Best evidence | Level |
|---|---|---|
| Uraemic pruritus | Cream with endocannabinoid-like lipids (PEA type), not a cannabis-derived product; Szepietowski et al., Wrocław Medical University, 2005; 21 patients, uncontrolled | Small, uncontrolled; indirect relevance to cannabis |
| Psoriasis | Mechanism only | No controlled evidence |
| Herpes simplex | In vitro inhibition (1980); worse disease in animals (1986) | No human trial. Effective antivirals exist |
| Sickle cell pain | 23-patient randomised crossover; primary outcome not significant, only mood interference significant | Small RCT, mixed |
| Leukaemia | Cell-line apoptosis via CB2 | Preclinical only |
| MRSA | Potent in vitro activity, 2008 | Preclinical only |
8. Why Grouping Them Is the Honest Choice
Each of these could be written up as its own article with an encouraging headline. Doing that would suggest six substantial bodies of research where there is really one small uncontrolled study of a non-cannabis cream, one small mixed trial, and four items with no human trial at all.
All six follow the same pattern: an early-stage observation gets quoted without saying how early it is. The observations are not made up. The way they are presented is the problem.
If you have used a topical cannabis preparation for a skin condition, the useful details are what was in the base, what else you were using, and whether the condition is one that fluctuates anyway. And if you are thinking about treating a mole or a non-healing skin lesion at home, please see a doctor first. That one is not negotiable.
This article is general educational information, not medical advice. Several conditions above have effective conventional treatments that should not be delayed.