The three-dimensional structure of the 5α-R I type protein (PDB ID: 3LLQ) was obtained from the Protein Data Bank for structural fitting and subsequent docking analyses (Linstrom, 1997; Linstrom and Mallard, 2001). Protein structure was acquired in the standard .pdb format. The preparation process involved the removal of water molecules, excision of superfluous chains, and the addition of polar hydrogens and charges. The refined protein structures were then saved in .pdb for further in silico studies. Molecular docking experiments were conducted using a personal computer equipped with an Intel Core i7-12700U CPU (2.3 GHz) and 16 GB of RAM, running Windows 11 (64-bit OS). To validate the docking procedure, the native protein ligand was initially docked to ensure methodological consistency.

Subsequent molecular docking of 5α-R was performed with four ligands: salicylic acid, bakuchiol, quercetin as a main compound of P. mume extract and shikimic acid. A comprehensive computational methodology was implemented in this investigation for the execution of molecular docking analyses. A multi-tiered protocol was employed, wherein the DiffDock (Corso et al., 2022) model was selected due to its optimal suitability for blind docking methodologies, in which no a priori knowledge regarding binding site localization isx presumed. This diffusion-based algorithmic approach was utilized for the prediction and characterization of ligand positioning within protein binding cavities. In docking process, Diffdock algorithm created 10 conformation and calculate Diffdock-confidence score for each of them. Subsequently, quantitative assessment of protein-ligand binding affinities was performed utilizing GNINA (McNutt et al., 2021) with grid, a scoring function operating on deep learning principles and calculated specifically for the top-rank (higger Diffdock-confidence score) conformation derived from DiffDock.

Concurrently, systematic analysis of protein structure surface interactions was conducted through the application of MASIF (Molecular Surface Interaction Fingerprinting) (Gainza et al., 2020).

The highest-scoring docking poses of shikimic acid were further refined through molecular dynamics (MD) simulations using GROMACS v.2023.1 (Abraham et al., 2015) with MPI support. Each protein–ligand complex was set up for MD simulation employing the CHARMM36 force field (Huang and MacKerell, 2013) for the protein and CGenFF parameters (Vanommeslaeghe et al., 2010) for the ligand. The systems were embedded in a cubic solvent box filled with TIP3P water molecules (Mark and Nilsson, 2001), ensuring a minimum padding of 1.0 nm between the protein surface and the box edges. System neutrality was achieved by introducing counterions.

Initial geometry optimization was carried out via the steepest descent algorithm, followed by equilibration in the NVT ensemble (100 ps) and then in the NPT ensemble (100 ps). A production MD trajectory was generated for 20 ns under NPT conditions at 298.15 K and 1 bar, with temperature and pressure regulated by the Nosé–Hoover thermostat (Evans and Holian, 1985) and the Parrinello–Rahman barostat (Ke et al., 2022), respectively. Coordinates were recorded every 10 ps for downstream analyses.

Binding free energies were estimated using the MM/PBSA approach (Genheden and Ryde, 2015) as implemented in gmx_MMPBSA v.1.6.4 (Valdés-Tresanco et al., 2021). The calculation was based on the final 10 ns of each production trajectory, sampling frames at 100-ps intervals. The Poisson–Boltzmann (PB) implicit solvation model was used with a grid spacing of 0.5 Å and an ionic strength of 0.1 M (Stein et al., 2019).

Skikimic acid (CAS 138-59-0) was obtained from Shaanxi Huike Botanical Development Co., Ltd. (Xian city, China). Prunus mume fruit extract standardized for quercetin (1–5 wt%) was prepared by Danjoungbio Co., Ltd. (Gangwon-do, Republic of Korea). Azelaic acid (CAS 123-99-9) was obtained from Sunfine Global Co., Ltd. (Chungcheongbuk-do, Republic of Korea). Salicylic acid (CAS 69-72-7) stabilized by methyl cyclodextrin was purchased from Shangdong Binzhou Zhiyuan Biotechnology Co., Ltd. (Shangdong, China).

Human sebocytes (Celprogen, United States) purchased from were grown in 75 cm² flasks (Corning, United States), cultivated and expanded in an incubator at 37 °C in the presence of 5% CO₂, using a culture medium with 0.03% Tween 20 (Sigma). When confluence was reached, the cells were seeded in 96-well plates (Corning, United States) to determine the non-cytotoxic concentrations of the evaluated substance.

Cell viability of dermal sebocytes was determined colorimetrically using MTT (3-[4,5-dimethylthiazol-2-yl]-2,5 diphenyl tetrazolium bromide) dye (Sigma Chemical, St. Louis, Mo., United States). For the assay the evaluated substance was prepared in culture medium with 0.03% Tween 20 (Sigma Chemical, St. Louis, Mo., United States) and added to the 96 well plate at a serial dilution in the range of 100.00–0.003 mg/mL using the dilution factor of 3.16¹⁰. Sebocyte cells were used in the amount of 10⁵ cells per well in a 96-well plate. Pre-culture was carried out for 48 h before the test. Basal control was represented as cells without treatment. The number of surviving cells was fixed with MTT at 5 mg/mL (30 µL/well), incubated for another 4 h and absorbance measurements were performed at 570 nm using Multiskan GO monochromator (Thermo Scientific, Finland). Cell viability was expressed as a percentage relative to the control. The number of biological repetitions was 6. Statistical data processing was performed to calculate the mean, standard deviation, and trend for each biological repetition. The normality of the data was controlled using the Shapiro-Wilk test. Comparisons were made using a linear mixed-effects model in case of normality; in case of nonparametric statistics, the Wilcoxon test was used.

The study was performed on skin sebocytes after culturing under standard conditions, using three maximum nontoxic concentrations from previous testing. Cells were additionally stimulated with 10 nM testosterone, simulating the inflammatory hormonal stress conditions of acne, and cultured for 72 h in an atmosphere with 5% CO₂ at 37 °C. For each experimental condition (basal, testosterone-induced stress, and active substances concentrations), three wells were plated, each containing ∼90,000 sebocytes. The cell supernatant was then obtained, purified, and the 5α-R type I amount was assessed by ELISA enzyme-linked immunosorbent assay using Human 3-oxo-5-alpha-steroid-4-dehydrogenase ELISA Kit (5α-reductase type I, ABclonal, United States, catalog code RK12279). Quantification was performed spectrophotometrically at a wavelength of 450 nm using a Multiskan GO monochromator (Thermo Fisher Scientific, Finland). After the assay, protein levels (pg/mL) were obtained by applying the measured optical density (OD) to the linear regression equation derived from the standard curve provided by the kit. Thus, data normalization relies directly on the standard curve calibration.

Salicylic acid and azelaic acid, known for the treatment of acne and related conditions (Lu et al., 2019; King et al., 2023), were used as positive comparison controls. This study included three technical replicates per condition. No additional normalization (e.g., total protein quantification) was applied.

To evaluate the data obtained, the ANOVA test was used, which also allowed measuring the variation in results, comparing data between groups. The Bonferroni post-test was then applied, which reinforced and made the result presented in the ANOVA test even more precise. A significance level of 5% was used in both assessments.

A novel phytochemical combination of shikimic acid and P. mume extract was implemented into the thin hydrocolloid cosmetic patches, as shown in Table 1.

The pH of the final patch formulation (semi-finished product) was measured to be between 5.5 and 7.5 at 25 °C. For stability, the patches were stored in dry, well-ventilated areas at temperatures ranging from +5 °C to +25 °C and at a relative air humidity not exceeding 75%, protected from direct sunlight and heat sources.

This clinical research was performed with thin polymer acne-patches soaked with composition of shikimic acid and P. mume extract to evaluate its compatibility with human skin (irritant potential) under normal conditions of use and reasonably foreseeable misuse conditions. Thin patches consisting of Table 1 ingredients were used as the base for the ingredients introduction. As a positive control, thin cosmetic patches containing 0.5 wt% encapsulated salicylic acid as the main active for acne treatment were selected.

Study was carried out in compliance with the most recent recommendations of the Helsinki Declaration (64th WMA General Assembly, Fortaleza, Brazil, October 2013) and has followed the “Guidelines for the Assessment of Skin Tolerance of Potentially Irritant Cosmetic Ingredients,” (Walker et al., 1997). After a favorable opinion from the ethical committee of CosmoProdTest (protocol code DR079860, 28 March 2024), the participants provided written informed consent to voluntarily participate in this clinical study.

A prospective open-label, non-randomized, single-center, split-body within-subject controlled study was conducted to determine the effect of thin acne patches containing a composition of shikimic acid and P. mume extract. Fifteen male and female volunteers, aged 18–28 years, mean age 19.8 years, with oily or combined skin type were involved. The study included volunteers with mild to moderate acne. Inclusion criteria were: volunteer’s agreement to carefully and diligently follow the instructions provided by the investigator, age 18–28 years, skin type: combination, oily, or prone to oiliness; clinical examination data: oily shine on the facial skin, enlarged pores, open and closed comedones, single papules, pustules, post-inflammatory spots; papulopustular form of acne of mild severity (presence on the face of closed and open comedones, an insignificant number of papules and pustules). Exclusion criteria were: allergic history, increased sensitivity to the components of the investigational products, dermatological diseases in the acute phase, problematic pigmentation; oncological diseases, blood diseases, autoimmune diseases; endocrinopathies affecting the condition of the skin; severe injuries and surgical interventions during the last 6 months; any cosmetic procedures in the observation area in the last 3 months preceding the study and during the study period (including the use of cosmetic products with phytoestrogens, AHA acids, retinoids within the last 6 months, laser or mechanical resurfacing, medium or deep chemical peels within the last 6 months, implantation of gels, threads, botulinum toxin injections, mesotherapy, biorevitalization, plasmolifting, ozone therapy within the last 2 months, etc.); any plastic surgery in the observed area in the last 12 months; systemic or topical use of medicinal products which, in the investigator’s opinion, may influence the outcome of the study during the study period and in the 6 months preceding the study; visiting a solarium; various diets; alcohol, smoking, narcotic drugs abuse. Volunteers met the inclusion criteria and had no exclusion criteria.

The study was designed as a split-body within-subject, using the investigational patch and the patch with salycilic acid (0.5 wt.%) in the same subject in parallel, on two close body areas with similar acne spots. The main objective was to confirm non-inferiority of the investigational product to the patch used as a standard for acne treatment. Subjects used the product under dermatologist control according to their individual personal hygiene routine and in accordance with the following guidelines: determine the size of the skin inflammation, clean the skin in the usual way, stick the patch on the inflamed area, evaluate the inflamed area after 6–8 h and remove the patch in case of positive results. In case the inflammation signs were still preasent observation was continued up to 48 h or until resolution.

Before the study, the sensitizing and irritating properties were assessed by a single application of the investigated product with the composition on the skin under dermatologist supervision. Additionally, functional parameters of the skin - sebummetry, comedogenicity, erythema and pigmentation indices when using the products were evaluated. Before the study, all subjects underwent a preliminary examination, including physical examination, analysis of concomitant diseases, thorough collection of dermatologic, allergologic and pharmacologic anamnesis. All studies were conducted in the same conditions of humidity and air temperature 22 °C ± 2 °C and relative humidity 40%–60%. The effects were evaluated using certified medical equipment: Cutometer^® 580 MPA equipped with a built-in sensor Sebumeter^® SM 815 (sebumetry); mexameter with a special lamp with 2 wavelengths (mexametry); Tewameter^® TM 300 (tevametry); UV lamp equipped with a magnifying glass manufactured by “IONTO-COMED GmbH” (Germany).

The clinical study included several visits to a dermatologist. Visit 0 served to obtain informed consent and visual assessment of skin condition using instrumental diagnostic methods. After that, thin patches were applied to selected areas with early inflammation. Visit 1 occurred 8 h after the start of the study and application of the patch to the inflamed area. The dermatologist removed the patches, after which the volunteer rested for 10 min to rule out false reactions to the patch. Next, the dermatologist performed clinical examination, local status assessment, and visual inspection. If the area of early inflammation did not resolve in 8 h, the dermatologist measured skin parameters and applied new thin patches. Visit 2 occurred 24 h after the first application of the study products. The dermatologist removed the patches, after which the volunteer rested for 10 min to rule out false reactions to the patch. Next, the dermatologist performed clinical examination, assessment of localized status, and visual inspection. If the area of early inflammation did not resolve in 24 h, the dermatologist measured skin parameters and applied new thin patches. Visit 3 occurred 48 h after the first application of the study products. The dermatologist removed the patches, after which the volunteer rested for 10 min to rule out false reactions to the patch. Next, the dermatologist performed clinical examination, assessment of localized status, and visual inspection. If the early inflammation area did not resolve in 48 h, the dermatologist measured the skin parameters and made a note in the individual patient’s medical record.

To assess efficacy measures a significance level of α = 0.05 was adopted for all statistical analyses. The primary hypothesis tested was the equality of means. The assumption of normality for the sampling distribution of the mean was inferred from the normality of the underlying data, which was assessed using the Shapiro-Wilk test. For within-group comparisons (baseline and post-treatment measurements), a paired Student’s t-test was employed. For comparisons between different patches groups, Welch’s t-test was used. To control the false discovery rate (FDR) arising from multiple comparisons, the p-values were adjusted using the Benjamin-Hochberg procedure. The obtained results were processed using the Statistica package (StatSoft, United States, Ver. 8.0).

Four ligands were selected to evaluate the interaction with 5α-R in silico: salicylic acid, bakuchiol, quercetin as a main phytochemical of P. mume extract and shikimic acid. Salicylic acid is one of the best known acne management components (Lu et al., 2019). Bacuchiol is a plant-derived component with anti-inflammatory and antibacterial effects, because of which it is widely used in dermatology and considered as a promising anti-acne agent (Puyana et al., 2022; Mascarenhas-Melo et al., 2024). Quercetin, found in Prunus mume fruit (Jang et al., 2018), has shown efficacy in the treatment of acne of bacterial nature due to its complex anti-inflammatory and antioxidant properties (Lim et al., 2021; Bo and Li, 2025). Shikimic acid has only recently entered the world of cosmetics and there are very few researches of testing this compound’s effects in humans. However, it has been reported to have potential for anti-acne applications due to its antibacterial and moisturizing properties (Batory and Rotsztejn, 2022).

According to the results of molecular docking, all of the studied substances–bakuchiol, quercetin, salicylic acid, and shikimic acid–were found interact with the active center of 5α-R (Figure 2). Across 10 docking simulations, the best-scoring poses consistently reproduced the experimental ligand conformation with an average RMSD of 2.06 Å (±0.44 Å).

The highest affinity to the 5α-R active center was shown by quercetin, but this compound is characterized by low stability due to its susceptibility to oxidative and light degradation and poor water solubility, which makes it limited for use in cosmetics (Osojnik Črnivec et al., 2024). Bakuchiol also has high affinity for the 5α-R active site, but its topical application is hampered by its low permeability across the epidermal barrier (Lu et al., 2025). Salicylic acid is one of the classical components for acne therapy, but it often causes dryness and irritation of the skin and therefore needs to be supplemented with moisturizers (Chularojanamontri et al., 2014).

Despite exhibiting relatively low binding affinity, shikimic acid is positioned within the steroid-binding pocket of the 5α-R active site. Optimal spatial complementarity and electrostatic interactions with critical catalytic residues are maintained in this orientation. Consequently, significant inhibitory efficacy is achieved through precise molecular recognition and stabilization within the enzymatic cleft, despite suboptimal affinity metrics. Shikimic acid is still not widely used in cosmetic products, but it has good stability (Chen et al., 2014) and moisturizing properties (Batory and Rotsztejn, 2022), so this substance was chosen for further work.

Molecular docking provides only an initial estimate of binding affinity and is inherently limited by the inaccuracies of scoring functions in reliably predicting true binding free energies (10.5772/intechopen.85991). For futher validation docking position of shikimic acid (Diffdock-confidence score equal, we used MM/PBSA analysis based on the Poisson–Boltzmann method yielded binding free energies of −6.53 ± 1.02 kcal/mol for 5-alpha-reductase. Analysis of the molecular dynamics (MD) trajectories indicated that paeoniflorin remained stably positioned within the corresponding docking binding site throughout the simulation (Figure 3).

The evaluation of cytotoxicity revealed that shikimic acid had a favorable toxicological profile on the skin sebocyte model in the studied concentration interval from 0.001% to 1.0% wt. Cell viability was maintained throughout the entire concentration range of shikimic acid and above the threshold value of 70% (EC₇₀), indicating a gentle effect on skin cells involved in the acne pathogenesis (Table 2; Figure 4). At the same time, the cytotoxicity of shikimic acid is about 15%–20% less than that of salicylic acid, often used for topical acne management.

Next, the effect of P. mume extract and shikimic acid combination on dermal sebocyte viability was tested. It was found that cell survival was maintained at a high level (85%–102% of live cells), throughout the studied range of 0.001–1.0 wt.% compared to the basal control. Throughout the concentration range from 0.001% to 0.2%, the survival of cells treated with the combination of shikimic acid and P. mume extract is only slightly lower compared to those treated with pure shikimic acid (average difference between 3% and 5%). However, starting at a concentration of 0.6%, the survival rate of cells treated with the combination drops sharply to 90% and further to 85%, which is about 15% lower than the survival rate under pure shikimic acid. This indicates that at concentrations up to 0.6% the combination gently affects the metabolic activity of cells and can be used for topical acne treatment (Table 3).

After shikimic acid exposure, the amount of 5α-R in skin sebocytes in vitro decreased by a value ranging from 67.5% to 101% depending on the acid concentration against the background of hormonal stress and returned to the initial physiological level (Table 4). This means that at a concentration of 0.3% shikimic acid completely eliminates the hormonal stress caused by excessive testosterone level (p < 0.001).

It was also checked whether the 5α-R-modulating properties of shikimic acid in the composition with P. mume fruit extract were not changed. According to the results of 5α-R quantification in skin sebocytes after the composition components were added to the culture medium, it turned out that in the combined presence of P. mume fruit extract not only does not impair the properties of shikimic acid, but shows an enhanced effect. This effect is well expressed even at the minimum concentration of 0.03% of the active ingredient studied by us (Figure 5).

The combination of P. mume extract and shikimic acid in a mass ratio of 1:1 reduced the amount of 5α-R by more than 100% (p < 0.001) at a concentration of 0.3% under testosterone-induced hormonal inflammatory stress (10 nM) compared to basal levels after 48 h of skin dermal cell culture (Table 5). Thus, the combination of shikimic acid with P. mume extract completely eliminates the effects of hormonal stress by reducing 5α-R back to basal levels. At the same time, this combination is almost 20% more effective than a similar concentration of shikimic acid alone (p < 0.001) and almost 30% more effective than similar concentrations of azelaic and salicylic acids (p < 0.01).

Also the combination with P. mume fruit extract was found to exhibit significant bioactivity, effectively affecting 5α-R in sebocytes at concentrations ranging from 0.03% to 1.0%. Shikimic acid alone has a more moderate impact on reducing the expression of 5α-R in skin sebocyte cells, and for a comparable effect, it needs to be used 10 times more than the tested combination. The action of 0.3% shikimic acid is equivalent to that of 0.03% of the combination with P. mume fruit extract (p < 0.001). Thus, P. mume fruit extract has an enhansing effect, allowing to reduce the concentration of shikimic acid by 10 times while maintaining the high biological activity of the combination. At the same time, there is no information in the literature about the fact that P. mume fruit extract itself has inhibitory activity against 5α-R. It is assumed that P. mume extract may have a synergistic effect, but more detailed study of the component synergism and its effect on various biological targets is required further.

The technological applicability of shikimic acid in cosmetic patches consisting of the components described in Table 1 was evaluated. Production of thin patches from hydrocolloid mass is a complex technological process that requiring a special method of preparation and composition introduction. It is necessary for the outer hydrogel layer to be impermeable and provide a protective film, while the inner layer has colloidal properties to bind exudate at the inflammation site. The stability and efficacy of such a construct depends on the preparation temperature and the interaction of the polymers in its composition, so technologically cosmetic patches are more complicated comparing to other topical acne treatments (Kuo et al., 2021). In this study, the technological complexity was further increased by the fact that shikimic acid is a water-soluble substance, and the thin patches have less than 5% free water to dissolve it.

When patches samples were made, it was found that shikimic acid at a concentration of 0.5 wt.% significantly impaired their stability and appearance. These changes were manifested by the precipitation of shikimic acid crystals over the entire patches surface, presumably due to its low solubility in anhydrous media. Thus, the patch samples had an uneven surface, crystalline inclusions, softening, decreased adhesion to the skin surface, color change during the stability test at 42 °C ± 2 °C, and impaired plasticity of the colloidal mass (Figure 6). The agglomerates formation was visible to the naked eye and confirmed by optical microscopy. Besides aesthetic imperfection, such changes critically affect the hydrocolloid patch properties. Patches with such a structure will unevenly release active ingredients into the skin and provide poor adhesion to the skin (checked by gluing the patch to the back of the previously cleaned hand (at the base of the thumb), the patch should instantly stick without bubbles or creases, fit tightly to the skin, not shift or slide across the surface), preventing delivery of the substance deep into the inflamed pore in acne. Shikimic acid crystals may also irritate the skin due to their moderate acidic properties. This means that despite their potential efficacy, in practice such patches are completely unsuitable for use on human skin.

The use of propylene glycol, glycerol and 1,2-hexanediol as solvents did not solve this technological problem. However, there are reports in the literature that plant extracts rich in polyphenols are able to unpredictably change the rheological properties of hydrocolloid films depending on the substances used: increase or decrease the viscosity of the solution, change thermostability and fluidity (Silva-Weiss, 2014). In particular, P. mume extract has similar properties due to its content of a large amount of flavonoids related in chemical structure to shikimic acid (Zhao et al., 2023). Thus, the P. mume extract addition potentially promoted the homogeneous dissolution of shikimic acid in the patches and ensured their stability over a long shelf life. To evaluate this hypothesis, a combination of shikimic acid and P. mume fruit extract in a mass ratio of 1:1 was prepared and its technological applicability as a stabilizing agent for thin acne patches was tested.

To preserve the patches consumptive and mechanical properties, it is necessary to make a premix of shikimic acid in the extract of P. mume fruits in a mass ratio of 1:1, with gradual addition of shikimic acid and a magnetic stirring at a speed of at least 300 rpm for at least 20 min. Under such conditions, shikimic acid starts to mutually dissolve in the P. mume fruit extract due to the high content of amphiphilic and lipophilic compounds, especially flavonoids, stabilizing shikimic acid due to related chemical structure. After preparation, the premix was introduced into the final hydrocolloid mass without heating under regular stirring until complete dissolution.

It turned out that the addition of P. mume fruit extract really increased the solubility of shikimic acid in the practically anhydrous hydrophobic hydrocolloid mass without crystal formation and without loss of elastic-viscosity-plastic properties of the patches. Patches produced by this technology had improved adhesion, stability and absence of inclusions, irregularities, air bubbles and crystals over the entire surface area (Figure 7).

After stable patches with improved adhesion and 5α-R-modulating properties and safety on human sebocytes were obtained, the safety and efficacy of these patches for treatment of mild to moderate acne were clinically investigated. Lespedeza cuneata extract and Lactobacillus ferment extract were also included in the patch formula to eliminate post-acne signs and provide an antibacterial effect. Both named components have proven safety and are used in the cosmetic industry (Lee et al., 2018; Ingredient, 2022; Dou et al., 2023).

No serious adverse events were observed in the subject group during the patch approbation throughout the clinical study period, including home application, which indicates the good tolerability of these patches. At the final stage after 48 h, a clinical study showed positive dynamics of the skin condition in the absolute majority of the volunteers: elimination or resolution of early inflammed skin signs, removal of redness, elimination of subjective discomfort, even skin tone. In addition, there was no resumption of inflammatory processes in the patch application sites, pores were not clogged and the original skin condition was not aggravated.

Acne signs elimination and supression at the early stage was achieved in 73% of the group of volunteers using study patches (11 out of 15 people) after 24 h. In 53% of volunteers who used thin patches with a composition based on shikimic acid and P. mume extract, the complete resolution of early inflammatory elements, regression of subjective sensations and visible improvement of skin condition after 8 h was noted. This demonstrates the rapid effect and hormonal stress normalization in mild acne (Figure 8). In the remaining 27% of volunteers inflammatory elements completely resolved after 48 h. The dermatologist noted the early resolution of inflammation with high efficacy and the patches’ safe impact on the skin. Early inflammatory elements disappeared in 24 h, while itching and redness were markedly reduced after one application within 8 h.

Using a positive control–patches with 0.5 wt.% salicylic acid–the acne signs elimination and suppression at the early stage was achieved in 73% of volunteers after 8–24 h, but with a predominant proportion after 24 h of constant patch contact with the skin. In 33% of volunteers using thin patches with salicylic acid, complete resolution of early inflammatory elements, regression of subjective sensations and visible improvement of skin condition was noted after 8 h. In the remaining 27% of volunteers, the inflammatory elements resolved after 48 h. Through in vitro study of 5α-R amount evaluation in human skin sebocytes, the combination of shikimic acid and P. mume was shown to be almost 30% more effective than salicylic acid. In this regard, the hypothesis of the research was that patches with salicylic acid may take longer time to produce a clinically significant effect than patches with shikimic acid and Prunus mume extract.

According to the results obtained by dynamic study of functional indices in the group of subjects using the biophysical research methods, an improvement in skin physiological parameters was found when using thin patches with a composition of shikimic acid and P. mume extract and with salicylic acid (Table 6). The combination of shikimic acid and P. mume fruit extract in a mass ratio of 1:1 significantly (p < 0.001) reduced the skin sebum content during 48 h of patches application, which allows normalizing the inflammation zone and reducing the influence of sebocytes on the acne pathogenesis. Patches with 0.5% salicylic acid also significantly (p < 0.01) reduced the skin sebum content during 48 h of patches application. A difference in efficacy was observed between the two types of acne patches, however, this difference did not reach statistical significance (p = 0.12).

The results obtained confirm the in vitro findings in terms of time frame. The comparatively rapid (from 8 h) inflammation area reduction is associated with the antioxidant and anti-inflammatory properties of P. mume extract and shikimic acid. The subsequent skin improvement after 24 and 48 h is likely to mediated by 5αR modulation in sebocytes, which acts through epigenetic mechanisms and requires time to take effect.

It should be noted though that the study had certain limitations: the trial was planned as a pilot proof of concept study and the sample size (15 per group) was quite small (Julious, 2005). The results of this trial confirm safe test patch profile and good skin compatibility. Also the results show non-inferiority of the test therapy as compared to the control with a tendency to potentially higher efficacy. So the future research is stipulated with a bigger number of subjects and more thorough skin parameters assessment, potentially combining in vivo and in vitro assessments.

These results should be contextualized within existing acne therapies. In vitro studies included azelaic acid as a positive control, a well-established topical agent for acne. However, to achieve its therapeutic effect, azelaic acid typically requires high concentrations of 15%–20% in formulations, which are often associated with local side effects such as pain, itching, and dryness (Feng et al., 2024). In clinical study, salicylic acid was used as a positive control, which is often classified as quasi-drug—a regulatory category for products falling between cosmetics and pharmaceuticals. Other first-line topical treatments have inherent limitations. Retinoids, for instance, exert their effects by binding to nuclear retinoic acid receptors and modulating gene transcription (Huang et al., 2014). This genomic mechanism results in a cumulative effect, with significant clinical improvement often taking several weeks to manifest. Benzoyl peroxide, another acne treatment gold standart, possesses a potent antibacterial action against Cutibacterium acnes but, to the best of our knowledge, there is no substantial evidence documenting a direct inhibitory effect on I type 5α-reductase activity (Eid et al., 2023).