Denifanstat for Acne: What the Trial Data Actually Shows

Denifanstat is an oral fatty acid synthase inhibitor that has completed phase 3 testing for moderate to severe acne in China. As of 3 August 2026 it is approved nowhere. Its New Drug Application was accepted for review by China’s National Medical Products Administration in December 2025, and in the United States the developer has said it plans to file an Investigational New Drug application in mid-2026, which is the step that precedes American trials rather than approval.

The results being quoted online come from two different places, and the difference matters. One phase 2 trial has been published in a peer-reviewed journal, and its headline endpoint missed significance. The phase 3 numbers everyone repeats come from the sponsor’s own press release and a conference presentation, and have not appeared in a peer-reviewed journal or in the trial registry. Both sets are below, each with its source attached.

What denifanstat is

Denifanstat, also called ASC40 and TVB-2640 in the trial registry, blocks fatty acid synthase, the enzyme that builds fatty acids inside cells. In skin, the target is the sebaceous gland. Ascletis describes the drug as a first-in-class FASN inhibitor that reduces sebum production by inhibiting de novo lipogenesis in sebocytes, modulates cytokine responses, and suppresses Th17 cell-mediated inflammation. That description comes from the developer, and the clinical trials test the result rather than the mechanism.

The class matters because it is genuinely new. Existing acne drugs work through retinoid signalling, antibacterial action, hormones, or keratolysis. A fatty acid synthase inhibitor works upstream of sebum itself, which is why the drug attracted attention. Sagimet Biosciences developed the molecule and licensed rights in Greater China to Ascletis Pharmaceuticals, which ran the acne trials.

One caveat about the target is worth stating, since it complicates the story rather than supporting it. A proteome-wide Mendelian randomisation study published in the Journal of Cosmetic Dermatology quantified 4,709 circulating proteins in 35,559 Icelandic individuals and integrated the data with acne genome-wide association studies covering 34,422 acne patients and 364,991 controls. Fatty acid synthase showed a protective effect against acne (OR 0.768; 95% CI 0.676 to 0.872; p = 4.685E-05), while tissue inhibitor of metalloproteinases 4 was associated with increased risk (OR 1.169; 95% CI 1.103 to 1.241; p = 1.956E-07). A genetic association pointing the opposite way to a drug’s mechanism does not overturn trial results, and the authors themselves list FASN as a promising target. It does mean the biology is less settled than a press release implies.

Where the drug stands on 3 August 2026

Ascletis announced on 10 December 2025 that China’s National Medical Products Administration had accepted its New Drug Application for denifanstat in moderate to severe acne vulgaris. Acceptance means the regulator agreed to review the file. It is not approval, and a drug under review can still be rejected or approved with restrictions.

In the United States, Sagimet states it plans to file an Investigational New Drug application for denifanstat in moderate to severe acne in mid-2026, and anticipates advancing into a registrational phase 3 trial in the second half of 2026. An IND is permission to start testing in humans in that country. No approval application exists in the United States or the European Union.

For anyone reading this in Dubai, the practical answer is short. There is no route to obtain denifanstat for acne here, no clinic can prescribe it, and any seller offering it online is selling something unverified.

The published phase 2 trial, and why its headline missed

One denifanstat acne trial has been published in a peer-reviewed journal: a phase 2, randomised, double-blind, placebo-controlled, dose-escalation, multi-centre trial run in China and registered as NCT05104125, published in the Journal of the European Academy of Dermatology and Venereology.

Patients received denifanstat orally at 25 mg (n = 45), 50 mg (n = 44), 75 mg (n = 45) or placebo (n = 45) once daily after dinner for 12 weeks. On total lesion counts, all three doses beat placebo: median reductions of −53.2% at 25 mg, −61.3% at 50 mg and −53.1% at 75 mg, against −34.2% on placebo, all with p < 0.05.

The endpoint that regulators care most about did not separate. The proportion of participants achieving at least a 2-grade reduction in Investigator’s Global Assessment was 31.1% on 25 mg, 31.8% on 50 mg, 22.2% on 75 mg and 15.6% on placebo, and the paper states that the difference among the four groups was not significant, with p = 0.336. Note also that the highest dose performed worst on that measure, which is not how a clean dose-response looks.

Study drug-related treatment-emergent adverse events were 48.9%, 47.7%, 62.2% and 48.9% across the 25 mg, 50 mg, 75 mg and placebo groups. The most common were dry eye, dry skin, urine protein positive, skin peeling and conjunctivitis, all grade 1 to 2, with no serious drug-related events. The authors’ own conclusion is measured: 50 mg once daily was generally well tolerated and showed potential to reduce total lesion counts, and larger and longer term studies are needed to confirm efficacy and safety.

The phase 3 numbers, and where they come from

The figures that circulate in beauty and industry coverage are phase 3 figures, and they are stronger than the phase 2 ones. Their source is a press release issued by Ascletis on 17 September 2025, describing results presented in the Late Breaking News sessions of the European Academy of Dermatology and Venereology Congress 2025 in Paris.

The trial is registered as NCT06192264: a phase 3, randomised, double-blind, placebo-controlled, multi-centre study of 480 participants across 41 sites in China, triple-masked so that participant, care provider and investigator were all blinded, running 84 days of treatment and completed on 19 May 2025. Participants were randomised 1:1 to denifanstat 50 mg once daily or placebo. According to the company’s release, at week 12 in the intention-to-treat analysis:

• IGA treatment success 33.17 on denifanstat versus 14.58 on placebo
• Total lesion count reduced by 57.38% versus 35.42%
• Inflammatory lesion count reduced by 63.45% versus 43.21%
• Non-inflammatory lesion count reduced by 51.85% versus 28.94%
• p < 0.0001 for every one of those comparisons
• Treatment-emergent adverse events in 58.6% of the denifanstat group and 56.3% of the placebo group, the most common being dry skin (6.3% versus 2.9%) and dry eye (5.9% versus 3.8%), with all drug-related events reported as grade 1 to 2 and none serious

Two qualifications belong next to those numbers, and neither is a reason to dismiss them. First, they are sponsor-reported. A conference presentation and a company press release have not been through peer review, and a reader cannot check the analysis, the handling of dropouts, or the full adverse event table. Second, as of 3 August 2026 the results section of the ClinicalTrials.gov record for NCT06192264 is empty, and a PubMed search for denifanstat and acne on the same date returned three records: the phase 2 trial report, a commentary titled “Denifanstat in acne therapy: A promising step forward”, and an expert opinion piece on FASN inhibitors in acne. The phase 3 paper is not among them.

What that means in practice: treat the phase 3 figures as the developer’s account of a completed trial, which is what they are, and wait for the published paper before treating them as settled. That is the same standard any dermatology journal applies.

The 52-week study, and what an open-label design can show

A second phase 3 study, registered as NCT06248008, enrolled 240 patients with moderate to severe acne in China. Ascletis reported topline results on 29 January 2026. Its primary endpoints were safety measures: incidence of treatment-emergent adverse events, incidence of serious adverse events, and incidence of discontinuation due to adverse events. The company reported that most events were grade 1 or 2, with no drug-related grade 3 or 4 events, no drug-related serious adverse events and no deaths.

The design point matters more than the numbers. This study was open-label with no control arm, meaning everyone knew what they were taking and there was no placebo group to compare against. That design is appropriate and normal for long-term safety, which is what it was built to measure. It cannot establish how well a drug works, because improvement over a year in an open study cannot be separated from the natural course of acne, which DermNet notes tends to improve after the age of 25.

What the evidence still does not cover

Four gaps are worth naming plainly, because none of them appear in the marketing coverage.

All controlled efficacy data runs to 12 weeks. Acne is a chronic condition managed over years, and nothing published so far shows what happens when the drug stops. Every trial to date was conducted in China, so the results have not been reproduced in other populations, including in the Gulf. There is no head-to-head trial against isotretinoin, doxycycline, or a topical retinoid, which means any claim that denifanstat is gentler or stronger than an existing treatment is a comparison nobody has run. And the peer-reviewed record consists of one phase 2 trial whose primary global-assessment endpoint did not reach significance.

None of this makes the drug a failure. A first-in-class mechanism reaching a completed phase 3 with an accepted regulatory filing is a real result. It does mean the honest description is “promising and unfinished” rather than “new standard”.

What moderate to severe acne means in numbers

The trials enrolled “moderate to severe” acne, and that phrase has a definition. DermNet classifies moderate acne as a total lesion count of 30 to 125 and severe acne as a total lesion count above 125. Acne affects males and females of all ethnicities and is prevalent in adolescents and young adults, with 85% of 16 to 18 year olds affected.

Two features move a case up the scale regardless of count: nodules and pseudocysts, and scarring. Scarring is the reason dermatologists escalate treatment early rather than waiting. Redness and pigment marks left after a spot resolves usually fade. A scar is a change in the structure of the skin, and once it forms, treating it is a separate and harder problem.

What is available in Dubai now

Acne treatment does not wait on a drug in regulatory review. DermNet’s summary of standard care runs as follows.

For mild acne: topical agents such as benzoyl peroxide, azelaic acid, and tretinoin or adapalene gel, along with some topical antibiotics such as clindamycin; newer topical agents such as clascoterone; low-dose combined oral contraceptive; antiseptic or keratolytic washes containing salicylic acid; light or laser therapy; and probiotics, which may have some effect but only after three months of use.

For moderate acne: everything above, plus a tetracycline such as doxycycline at 50 to 200 mg daily for around six months, with erythromycin or trimethoprim if doxycycline is not tolerated. Antiandrogen therapy with cyproterone acetate plus ethinylestradiol, or spironolactone, may be considered in women not responding to a low-dose combined oral contraceptive, particularly with polycystic ovaries. Isotretinoin is often used if acne is persistent or treatment-resistant, and intralesional steroid injections can help acute larger lesions.

For severe acne: referral to a dermatologist. Oral antibiotics are often used at higher doses than normal, and oral isotretinoin is usually recommended in suitable patients. If there is fever, joint pain, bone pain, or ulcerated or extensive skin lesions, a blood count should be arranged and the referral is urgent.

One practical note on sequencing. DermNet’s advice on benzoyl peroxide includes being patient, because acne responds very slowly to treatment and it may take several months to notice an improvement. Judging a regimen at three weeks is the most common reason people abandon something that was working.

Isotretinoin, since every denifanstat article compares them

Isotretinoin is the drug denifanstat is measured against in press coverage, so its actual profile is worth stating rather than summarising as “harsh”.

In acne, isotretinoin reduces sebum production, shrinks the sebaceous glands, reduces follicular occlusion, inhibits bacterial growth and has anti-inflammatory properties. Daily doses for acne range from under 0.1 to over 1 mg per kg of body weight, and courses have often been restricted to 16 to 30 weeks. Some prescribers target a cumulative dose of 120 to 140 mg/kg hoping to reduce relapse, though DermNet notes the evidence for that remains controversial, and the general trend has been towards lower doses.

The costs are specific and well documented. Dry lips and cheilitis occur in 100% of patients on 1 mg/kg/day. Dry skin, fragile skin, dermatitis, dry nostrils and nosebleeds, dry or irritable eyes, facial redness, sunburn on sun exposure, temporary hair loss and brittle nails are all listed among common effects. Patients with significant liver or kidney disease, high blood fats, diabetes or depression may be advised not to take it or to take a lower dose with regular follow-up.

The absolute rule concerns pregnancy. Isotretinoin must not be taken in pregnancy, or if there is a significant risk of pregnancy, because of a very high risk of serious congenital abnormalities. Pregnancy must be excluded before and during treatment. Anyone who could biologically have a child should take precautions during treatment and for four weeks after stopping, and if sexually active, two reliable methods of contraception are advised; the low-dose progesterone mini-pill alone is not recommended. Blood donation is not allowed while taking it. No contraceptive precautions are necessary for men, since isotretinoin has not been shown to affect sperm or to cause birth defects in children fathered by men taking it.

Set against that, the interest in a drug that lowers sebum without teratogenicity is easy to understand. Whether denifanstat is that drug is exactly what the missing head-to-head trials would tell us.

Risks that apply to the treatments you can actually get

Denifanstat carries no advice for a reader in Dubai, because nobody here can take it. The treatments available do carry advice.

Doxycycline is a photosensitiser and can cause unexpected sunburn or acute painful onycholysis, so skin and nails need protecting from sun exposure, which in Dubai is not a footnote. It must be taken upright with copious water and about an hour before lying down, to reduce oesophagitis, and it is better tolerated after food. Tetracyclines must not be taken by pregnant or breastfeeding women, or by children under 12, because they discolour growing teeth and may cause enamel hypoplasia.

Topical retinoids cause retinoid dermatitis with erythema, peeling and dry skin, along with irritant contact dermatitis and sun sensitivity, and they make eczema worse through the same drying effect. Some are contraindicated in pregnancy, and DermNet states that women of childbearing age must use effective contraception due to the teratogenic effects of retinoids. Adapalene is the least irritating topical retinoid and tretinoin the most, and applying at night with a sunscreen and moisturiser combination during the day is the standard way to manage both problems.

Benzoyl peroxide causes expected mild dryness and can bleach clothing, towels and bedding, so it should dry fully before the skin touches fabric. Serious allergic reactions including anaphylaxis have been reported, though rarely. Chemical peels carry an increased risk of dyspigmentation and of hypertrophic and keloid scarring in Fitzpatrick skin types IV to VI, which describes a large share of patients in Dubai, and DermNet advises waiting three to six months before repeating a moderate depth peel.

One habit prevents most avoidable trouble: start one new treatment at a time, so that a reaction identifies its own cause.

When acne needs a doctor now

See a doctor without waiting if acne is leaving scars, or if there are nodules or cysts under the skin. The same applies when three months of over-the-counter treatment has changed nothing, or when acne begins abruptly in adulthood alongside irregular periods or excess hair growth in women, which can point to a hormonal cause. Add one more: acne that is affecting mood, sleep or willingness to leave the house. That last one is a medical reason, not a soft one.

Seek care urgently if acne comes with fever, joint pain, bone pain, or ulcerated or extensive skin lesions. Swelling of the lips, tongue or throat, difficulty breathing or swallowing, or a widespread rash with faintness after starting any medication is an emergency: in the UAE call 998 for an ambulance, and anyone feeling faint should lie down with their legs raised until help arrives.