Breezula (Clascoterone): Is Systemic Absorption Really Negligible?
Published by AmpleLab Research
When Cosmo Pharmaceuticals announced its Phase 3 results for clascoterone 5% solution (Breezula) in December 2025, the release described the drug as blocking DHT at the follicle receptor "without systemic absorption", and quoted an investigator praising "real, measurable regrowth with negligible systemic exposure." In April 2026, the 12-month release tied its safety findings to "the negligible systemic absorption previously seen in earlier clinical studies."
That framing matters, because a major rationale for a topical antiandrogen is to concentrate its action at the target tissue while minimising systemic exposure. But clascoterone is not an unstudied molecule. Its 1% cream, Winlevi, is an approved acne treatment with published pharmacokinetic data, regulator-reviewed labelling and a UK summary of product characteristics. Those documents describe measurable blood levels that rise over the first days of use, and a reversible effect on adrenal function in a minority of people at maximal use.
This article sets out what those studies measured, what they can and cannot tell us about a scalp solution, and what has not yet been published. For the efficacy side of the story, see our Breezula Phase 3 article; this piece is about exposure.
What Cosmo Has Said, and What It Has Shown
Three statements are worth separating, because they come from different places. The mechanism paragraph of the December 2025 topline release, written by the company, describes DHT being blocked at the follicle receptor "without systemic absorption". The same release quotes Dr Maria Hordinsky of the University of Minnesota on "negligible systemic exposure". And the April 2026 release says its 12-month findings "are consistent with the negligible systemic absorption previously seen in earlier clinical studies", alongside "no significant systemic hormonal side effects observed".
The release reports that clascoterone 5% "maintained a safety and tolerability profile comparable to vehicle" across 12 months. It contains no adverse-event rates, no laboratory results (adrenal, cortisol or potassium), no plasma concentrations, and no count of the men in the 12-month follow-up. It does not say which earlier studies it means by "previously seen", and says the full dataset is planned for submission to a peer-reviewed journal.
Cosmo's July 2026 half-year results say only that scientific publications from the programme are expected to follow, with no timing. Until the data appear, these are statements, not results. That is how company topline releases work, and it is not a criticism of Cosmo in particular. It does mean the claim has to be tested against the published studies, which is what the rest of this article does.
What the Acne Studies Measured
The most direct human data come from the 1% cream, because it went through formal pharmacokinetic and adrenal testing for its acne approval. The clearest published example is an open-label study of 42 people aged 12 and over with moderate-to-severe acne, 20 adults and 22 adolescents, who applied the cream twice daily for 14 days.
Clascoterone 1% cream was applied twice daily for 14 days to the face, chest and/or back. Plasma levels reached steady state by day 5 and were about 1.8 to 2.1 times higher than after the first dose. Mean peak concentration (Cmax) was 4.4 ng/mL in adults and 4.6 ng/mL in adolescents, with wide variation between individuals (coefficients of variation of 67% and 103%). The metabolite cortexolone stayed below the 0.5 ng/mL quantification limit in both groups. Three of 42 people (7%) had an abnormal adrenal test, covered below. The authors are affiliated with Cassiopea SpA, the compound's original developer, which Cosmo has since acquired, and they concluded that the cream was safe and well tolerated over 14 days.
The US prescribing information gives more detail on the adult cohort of that same study (20 people applying about 6 g of cream twice daily for two weeks, a large amount chosen to approximate a worst case). By day 14, mean Cmax was 4.5 ng/mL (standard deviation 2.9), the area under the curve over a dosing interval was 37.1 h·ng/mL (SD 22.3), and the average concentration was 3.1 ng/mL (SD 1.9). Around 84% to 89% of clascoterone in plasma is bound to protein in vitro. The UK summary of product characteristics (Glenmark, last revised January 2025) gives the same figures. The 2025 paper described below confirms that the 4.5 ng/mL figure comes from this phase 2 study, so the label and the 2019 paper are one dataset, not two.
Section 12.3 reports the pharmacokinetics of the adult cohort described above, with steady state by day 5 and the main expected metabolite, cortexolone, generally at or below 0.5 ng/mL. Section 5.2 carries a warning that HPA axis suppression "was observed and may occur during or after treatment". The label also notes that elevated potassium levels were seen in some trial participants.
The most thorough pharmacokinetic dataset is a 2025 paper reporting five Phase 1 studies of the 1% cream, which includes the only longer-term exposure data we found.
The paper reports five Phase 1 studies of clascoterone 1% cream: single-dose and 14-day multiple-dose studies in healthy volunteers (24 each), a six-week steady-state study in 8 patients with acne, and skin irritation (36) and sensitisation (250) studies. In the six-week study, patients applied 6 g (60 mg) once daily to the face and parts of the chest and back. On day 42, mean Cmax was 2.71 ng/mL (coefficient of variation 73%), mean Cmin 0.35 ng/mL, average concentration 1.39 ng/mL and AUC over the dosing interval 33.45 h·ng/mL. Trough concentrations and urinary recovery showed no further increase after the first application through to day 42, and the authors report that no accumulation was observed. In the 14-day study of 4 mL (40 mg) or 8 mL (80 mg) once daily, mean plasma levels after the 10th and 14th applications ranged from 0.23 to 2.42 ng/mL, with high variability, and less than 1% of the applied dose appeared in urine as conjugated clascoterone. The authors report no clinically relevant changes in laboratory tests, vital signs or ECGs. The studies were funded by Cosmo and Intrepid Therapeutics, and the authors include employees of Cassiopea and Sun Pharma; two authors hold stock options in Cassiopea and have served as senior officers of Cosmo.
What the Peer-Reviewed Paper Says
The pharmacokinetic paper does not say "no systemic absorption". Its abstract says the Phase 1 findings "supported low systemic absorption" of clascoterone 1% cream. It is a developer-affiliated paper, and it covers the cream, not the scalp solution.
Three features stand out across these studies. First, the levels are low, single-digit nanograms per millilitre, but they were consistently detectable in the cream studies that measured them. Second, they rise over the first days of use (about 1.8 to 2.1 times between the first dose and steady state in the twice-daily 14-day study) and then plateau rather than climbing further, at least over the six weeks tested in eight patients. Third, the variation between people is large, with coefficients of variation of 67% and 103% for Cmax in the 2019 study and 73% in the six-week study, so some individuals will run well above the average.
The Scalp Solution: What the QT Study Adds
None of the above used the scalp solution. The closest published human data for the AGA formulation come from a 2021 Phase 1 study designed to test whether clascoterone affects the heart's QT interval, which required deliberately high exposure.
Thirty-two healthy volunteers were randomised 3:1 to clascoterone (24) or vehicle (8). Active participants applied 3 mL of a 7.5% solution (225 mg) to the scalp and both thighs twice daily on days 1 to 3 and once on day 4, a total of 450 mg on a full day. Mean Cmax rose from 4.81 ng/mL on day 1 to 7.65 ng/mL on day 4, and the area under the curve over 24 hours rose from 57.28 to 113.94 ng·h/mL, roughly a doubling by our calculation from the reported means. Cortexolone, a metabolite the body also makes naturally, was quantifiable in 21 participants by day 4 against 5 on day 1 (day-4 mean Cmax 1.27 ng/mL), although the paper does not separate drug-derived from endogenous cortexolone. There was no clinically meaningful QTc prolongation. The abstract states that the solution developed for androgenetic alopecia "leads to a measurable systemic concentration and accumulation of the antiandrogen". The paper reports no cortisol or adrenal measurements.
Three caveats matter. The solution was 7.5%, not the 5% used in the Phase 3 trials, and the dose was deliberately supratherapeutic. The study ran for four days, so it says nothing about 12 months of use. And it measured no adrenal response.
How does that compare with the scalp dose? The trial registry entry for SCALP 1 lists 1.5 mL of 5% solution twice daily to the vertex and temples. By simple arithmetic that is 75 mg per application and 150 mg a day. That is a little above the roughly 120 mg a day delivered by the maximal-use cream regimen (about 6 g of 1% cream, twice daily), about two and a half times the 60 mg a day (6 g of 1% cream once daily) in the six-week study, and a third of the 450 mg a day in the QT study. These are our calculations from the registered volumes and concentrations, and the registry record, last updated in February 2024, may not reflect exactly how the trials were run.
The point is not that either study predicts scalp exposure. Vehicle, application site and duration all differ, so the doses cannot be used to interpolate a plasma level. The point is that the dose used in the AGA trials sits in the range where measurable levels have already been documented, which makes the absence of published exposure data for the trial formulation the main gap.
The Adrenal Question
The HPA axis is the hormonal chain linking the brain's hypothalamus and pituitary to the adrenal glands, which make cortisol. A cosyntropin stimulation test checks it: the person receives a synthetic form of the pituitary hormone ACTH, and cortisol is measured 30 minutes later. A low response suggests the adrenals are being held back, as can happen with potent topical corticosteroids.
Clascoterone gets this test because it is not a pure antiandrogen. The FDA's risk review notes that it has weak glucocorticoid activity, and its metabolite cortexolone (11-deoxycortisol) is an intermediate in the body's own cortisol synthesis, which the TGA's product information describes as having weak glucocorticoid properties. The developer-affiliated 2025 paper says the same, describing cortexolone as generally inert apart from weak glucocorticoid activity.
In the 2019 study, 3 of 42 people (7%) had a post-stimulation cortisol below the study's 18 µg/dL cut-off at the end of treatment, at 14.9 to 17.7 µg/dL. All three recovered within four weeks of stopping, and none showed clinical evidence of adrenal suppression. The US label reports the same finding by age group (1 of 20 adults, 5%, and 2 of 22 adolescents, 9%) and carries it as a warning. The UK SmPC advises considering withdrawal if suppression develops, and lists large treated areas, prolonged use and occlusive dressings as factors that increase absorption.
Read calmly, that is a small, reversible signal in a short, high-dose test of the cream. The adrenal data we found all come from short tests of the cream, 14 days of use, not 12 months. For the 5% scalp solution, the QT study measured no adrenal response, and the SCALP 1 registry entry lists physical examinations, ECGs and blood and urine safety labs without mentioning adrenal testing. The one scalp-solution cortisol statement we found is from a 2019 company release, covered below.
That statement, from the Phase 2 dose-ranging study of more than 400 men in Germany, is worth reading closely. In its April 2019 results release, Cassiopea said cortisol was tested in a subgroup of patients "to verify that Clascoterone is free from systemic steroid activity", that the mean absolute changes in cortisol "were similar among groups", and added that this was "proving that Clascoterone has no systemic effect on cortisol". The arms included 2.5%, 5% and 7.5% solutions twice daily, so the Phase 3 strength and schedule were covered.
Its limits are clear from what it leaves out. It is a company statement rather than a peer-reviewed report. The release does not say how many patients were in the subgroup, when cortisol was sampled, whether a stimulation test was used, or give any values. A group mean can also look unchanged while a few individuals shift, which is the pattern the 1% cream study found (3 of 42 below cut-off). "Proving" is a strong word for an unquantified subgroup mean. It is reassuring, but it is a summary, not the data.
The Studies Side by Side
The five datasets side by side, with the formulation, dose and exposure each one covers:
What "Measurable" Does and Doesn't Mean
None of this shows harm. A drug can be detectable in blood without producing a clinical effect, and the numbers above are low and strongly protein-bound, with the weakly glucocorticoid-active metabolite barely quantifiable in the cream studies. The adrenal signal was small, reversible and unaccompanied by clinical features. If Cosmo's full dataset bears out its reported 12-month tolerability comparable to vehicle, that would be an encouraging result.
What the data do show is that "negligible" and "without systemic absorption" are stronger than the published pharmacokinetics support. The developer-affiliated papers themselves are more measured: the 2025 five-study paper concludes that its findings supported "low systemic absorption" of the cream, and elsewhere describes "limited" and "minimal" exposure. "Low" or "limited" describes the studies more accurately than "negligible" or "without". The difference matters most to the people likeliest to read the claim: men considering a topical antiandrogen precisely because they want to avoid systemic hormonal effects. It also matters for how this class is compared. Our RU58841 article covers RU58841, an older compound with the same mechanism whose bloodstream levels were measured in a human study, and our Breezula vs KX-826 comparison explains why the datasets for these drugs cannot be set side by side.
There is also an open question the pharmacokinetic studies do not answer: whether these plasma levels do anything at the androgen receptor or to hormones controlled by it. The studies above report drug concentrations, adrenal tests, ECGs and local tolerability, and we found no serum androgen measurements in the sources we reviewed. That is the gap the "no significant systemic hormonal side effects" claim will have to fill with numbers.
One more systemic signal is potassium. The FDA's review observes that clascoterone is structurally similar to spironolactone, which can raise potassium, so a similar effect is biologically plausible, and the US label notes elevated potassium in some trial participants. The reported rates were close to vehicle in the US label and the TGA product information (5% against 4%, and 3.7% against 3.9%, respectively), while the FDA review gives a somewhat larger difference (6.7% against 3.9%). The TGA adds that the Phase 3 acne trials ran no clinical laboratory evaluations. The SCALP registry lists blood safety labs, so potassium data should exist for the hair programme, but none has been reported.
What Would Settle It
Four pieces of data would resolve most of this. Steady-state plasma levels for the 5% solution at the trial dose on the scalp. Numerical cortisol results from the scalp solution, ideally including an appropriately timed assessment of adrenal function after prolonged use, rather than relying on the 14-day cream study. Potassium and other laboratory results from the SCALP programme. And serum androgen levels, with rates of sexual-function adverse events set against vehicle.
Cosmo targets a US filing in the first quarter of 2027 and a European filing in the second quarter, according to its July 2026 release. Regulators' review documents, like the Winlevi risk review used throughout this article, tend to publish exactly these numbers, so the public record should improve as the filings progress. In the meantime, the sensible reading is the cautious one: clascoterone is designed to act locally, and the evidence so far is compatible with low systemic exposure, but "low" has been measured and "none" has not. For the wider efficacy and comparison picture, see our clascoterone vs finasteride claim article and our guides to finasteride vs dutasteride and topical finasteride.
Frequently Asked Questions
Does Breezula (clascoterone 5%) get into the bloodstream?
Based on the available studies, yes, to a measurable extent. In a Phase 1 study of a 7.5% scalp solution at a supratherapeutic dose, clascoterone was quantifiable in plasma and exposure roughly doubled between day 1 and day 4. The 1% acne cream also produced low but measurable levels, around 4 to 5 ng/mL at peak in the twice-daily maximal-use study and about 2.7 ng/mL in a six-week once-daily study. Whether levels with the 5% solution at the Phase 3 dose differ has not been published.
Is Breezula's systemic absorption negligible?
Cosmo's releases and a quoted investigator use that word, but the releases give no plasma data. Published studies of clascoterone report low but measurable levels that rise over the first days of dosing, so "low" is a more accurate description than "negligible" or "none". Low does not mean harmful, and the 12-month safety data are reported as comparable to vehicle but are not yet published in full.
Can clascoterone suppress adrenal function?
At maximal use of the 1% cream for 14 days, a small proportion of people had a post-stimulation cortisol at or just below the cut-off, 3 of 42 in one study, and all recovered within four weeks with no clinical signs. The US label lists HPA axis suppression as a warning. For the 5% scalp solution, the only adrenal statement we found in the public record is a 2019 company release saying mean cortisol changes in a Phase 2 subgroup were similar across groups, with no values or subgroup size given. We found no peer-reviewed adrenal data.
How does clascoterone compare with oral finasteride systemically?
They work differently. Oral finasteride lowers DHT throughout the body, while clascoterone leaves DHT production alone and blocks the androgen receptor, with systemic exposure that is low but measurable. No head-to-head trial has compared their systemic effects, so the comparison rests on separate bodies of evidence.
Is Breezula approved?
No. Breezula is an investigational drug that is not approved by any regulator and cannot be bought through any legitimate channel. Cosmo's July 2026 half-year release targets a US filing in the first quarter of 2027 and a European filing in the second quarter, with no approval date.
Is Winlevi the same as Breezula, and does its label apply?
They share the active molecule, clascoterone, but Winlevi is a 1% cream approved for acne and Breezula is a 5% scalp solution. Winlevi's label is the best regulator-reviewed source on clascoterone absorption, but it describes a different strength, vehicle and application site, so it is a reference point rather than a prediction for the scalp solution.
Selected Research
This article is provided for educational purposes and does not constitute medical advice. Breezula (clascoterone 5% solution) is an investigational drug not approved by any regulator and not available for purchase through any legitimate channel. Winlevi (clascoterone 1% cream) is approved for acne only. AmpleLab products are cosmetic formulations, unrelated to clascoterone, and are not intended to diagnose, treat, cure, or prevent any condition.
AmpleLab.