Fabrication of dissolvable microneedles or DMNs for mechanical testing.
Dissolvable microneedles or DMNs were fabricated at room temperature in a clean room using the procedure as previously described33. Briefly, appropriate volume of sterile polymer solution (2.08% HA of MW 5,000 Da, 4.16% PVA of MW 32000, 4.16% sucrose, in weight to volume of water) were poured into silicone molds and left under the moisture control atmosphere of 5% humidity until dry. VitC-gluta-DMNs were prepared similarly except that vitC and glutathione were added to the polymer solution. DMNs of two different needle lengths were fabricated, the 350 µm DMN patch (1.5 cm in diameter or 1.76 cm2 circular patch containing 89 needles positioned at the tip to tip distance of 1,120 μm; each needle is nail-shaped with 180 x 180 x 200 μm (W x L x H) square column and, 150 μm height of the square pyramid on the top) and the 550 µm DMN patch (1.5 cm in diameter or 1.76 cm2 circular patch containing 89 needles positioned at the tip to tip distance of 1,110 μm; each needle is nail-shaped with 200 x 200 x 400 μm (W x L x H) square column and, 150 μm height of the square pyramid on the top). The needles were sitting on a 500 µm thick base made of the same material as the needles. All prepared DMNs were examined under stereomicroscope (Olympus DP22, Tokyo, Japan).
Fabrication of detachable dissolvable microneedles or DDMNs for stability experiment, ex vivo, in vivo and clinical test in volunteers
Detachable DMNs (so called DDMNs) were fabricated using the micro-molding method in a similar fashion to the above DMN preparation except that the base of needles is attached to water penetrable membrane sheet. Briefly, for red dye-loaded DDMNs, appropriate volume of sterile polymer solution containing red dye (2.08% HA of MW 5,000 Da, 4.16% PVA of MW 32000, 4.16% sucrose, 2.52% ponceau 4R, 1.56% azorubine in weight to volume of water) were poured into silicone molds and left under the moisture control atmosphere of 5% humidity until dry. Then the water penetrable membrane was attached and the DDMN patch were put under moisture control atmosphere of 2.5% humidity until completely dry. VitC-gluta-DDMNs were prepared similarly except that dye solution was replaced with vitC and glutathione solution. Two different lengths of needle of DDMN patches were fabricated, the 350 µm DDMN patch (1.5 cm in diameter or 1.76 cm2 circular patch containing 89 needles positioned at the tip to tip distance of 1,120 μm; each needle is nail-shaped with 180 x 180 x 200 μm (W x L x H) square column and, 150 μm height of the square pyramid on the top) and the 550 µm DDMN patch (1.5 cm in diameter or 1.76 cm2 circular patch containing 89 needles positioned at the tip to tip distance of 1,110 μm; each needle is nail-shaped with 200 x 200 x 400 μm (W x L x H) square column and, 150 μm height of the square pyramid on the top). All prepared DDMNs were examined under stereomicroscope.
Mechanical property of DMNs
Compressive strength of DMN arrays was determined by measuring displaced distance as the tested sample was compressed at an applied axial force rate of 1 mm/min (with the maximum force of 100 N or until microneedle broke) through a universal testing machine (Shimadzu EZ-S, Shimadzu Corporation, Tokyo, Japan) operated at 32 °C and a relative humidity of 7.4 ± 2%.
Tested DMN samples included 1) the DMN patch with needle length of 350 μm containing 380 μg vitC and 95 μg glutathione (high dose-vitC-gluta-350-DMN patch which corresponds to 40% vitC and 10% glutathione in the needle matrix), 2) the DMN patch with needle length of 550 μm containing 664 μg vitC and 166 μg glutathione (high dose-vitC-gluta-550-DMN patch which corresponds to 40% vitC and 10% glutathione in the needle matrix), 3) the DMN patch with needle length of 350 μm containing 190 μg vitC and 47.5 μg glutathione (low dose-vitC-gluta-350-DMN patch which corresponds to 20% vitC and 5% glutathione in the needle matrix), and 4) the DMN patch with needle length of 550 μm containing 332 μg vitC and 83 μg glutathione (low dose-vitC-gluta-550-DMN patch which corresponds to 20% vitC and 5% glutathione in the needle matrix).
Ex vivo skin penetration and diffusion
Full thickness of fresh porcine (crossbred pig, Sus scrofa domesticus) ear skin and fresh Wistar rats (Rattus norvegicus) dorsal skin (obtained from Department of Pathology, Faculty of Veterinary Science, Chulalongkorn University, and National Laboratory Animal Center, Mahidol University, Thailand, respectively) were cleaned with water, shaved with hair clipper, and dried with tissue paper. All fresh skin pieces were used immediately.
Skin penetration of the low dose-vitC-gluta-350-DDMN patch, and the low dose-vitC-gluta-550-DDMN patch was investigated. Red dye, vitC and glutathione were co-loaded into the DDMNs. The amounts of dyes were 9.97 and 6.17 µg of ponceau and azorubine in each of the low dose-vitC-gluta-350-DDMN, and 17.43 and 10.79 µg of ponceau and azorubine in each of the low dose-vitC-gluta-550-DDMN patch. The experiment was initiated by pressing the DDMN patch against the stratum corneum side of the porcine skin piece with the force of 20 N. Then the backing of the DDMNs was wet with 2 drops of water and the patch was hand-pressed against the skin for another 2 min before removing the backing from the skin. The skin was immediately subjected to cross sectioning and the cut surface of sectioned skin was straightaway visualized under the stereomicroscope.
Diffusion test was performed on red dye-loaded DDMNs. The red dye-loaded DDMN patch used in the experiment was the 550 µm needle-height-DDMNs containing 470 µg dye/patch (290 µg ponceau and 180 µg azorubine). The DDMNs were applied on the skin, then the skin pieces were placed on wet filter paper (soaked with 0.01 M PBS buffer of pH 7.4). At designated times (0, 30, 60, 120 min post administration), the skin piece was surgically sectioned along the line of the embedded needle array, and the sectioned skin piece was immediately visualized under the stereomicroscope. Each control skin piece was topically applied with the red dye solution containing the same amount of dye as in each DDMN patch (470 µg dye in 20 µL water spread on 1.76 cm2 area of the skin) and was processed similarly.
Observation of skin surfaces of human volunteers
The in vivo skin surface monitoring experiment on human volunteers was carried out using the low dose-vitC-gluta-350-DDMN patch and the low dose-vitC-gluta-550-DDMN patch. DDMN application was carried out at the inner forearm skin of four volunteers. Each volunteer was administered with both DDMN samples (2 min application time) at location 2 cm apart. The skin sites were photographed by stereomicroscope at various times (0, 30, 60, 120 min) post the administration. Control skin (the nearby area with no DDMN administration) was also photographed.
The same volunteer was also administered with four conventional DMNs at location 2 cm apart. The DMN administration was carried out with the DMN patch being left on the skin (no needle detachment feature). Each DMN patch was removed at different times, 0, 30 60 and 120 min post the start of the application. Skin surface photograph was taken immediately after the removal of the DMN patch.
Optimization of DDMN formulation to stabilize the loaded vitC
Stability of vitC in solution. Three antioxidants including coenzyme Q10, vitamin E and glutathione, were experimented for their effect on vitC stability51,52,53. VitC solutions containing various amounts of each antioxidant were prepared. Final concentration of vitC was 0.002% (w/v) in water and final concentrations of antioxidants tested in the experiment were 0.002, 0.001, 0.0005, 0.00025 and 0.000125% (w/v). The tested solution was kept in normal glass test tube under normal natural indoor light (UVB (280–320 nm) of 0.05–0.06 mW/cm2 and UVA (320–400 nm) of 3.0–5.0 mW/cm2), and at designated time the solution was subjected to UV-visible absorption analysis (maximum absorption of vitamin C is 265 nm) using the CARY 100 Bio UV-visible spectrometer (Palo Alto, USA). Another set of the same sample was kept in the dark and analyzed similarly.
VitC stability in DDMNs. Each of the freshly prepared DDMNs co-loaded with glutathione at 25% of the amount of vitC (low and high dose-vitC-gluta-350-DDMN patch) was kept separately in individually sealed foil package at 25 (RT), 40 and 50°C. At specific time points, the DDMN patches were taken out and dissolved in 10 mL water under light-proof condition. The obtained solutions were immediately subjected to UV-visible absorption analysis as described above. At each time point, freshly prepared vitC standard solutions were used to construct calibration curve. Experiment was carried out in triplicate.
In vivo skin irritancy
All animal experiments were approved by Chulalongkorn University Institutional Animal Care and Use committee (CU-IACUC) (Protocol no.1973015). The study was carried out in compliance with the ARRIVE guidelines 2.0 checklist (Table S1) and all methods were carried out in accordance with the guidelines and CU-IACUC regulations. The small sample size was selected to minimize the numbers of animal used. Eighteen male Wistar rats (Rattus norvegicus) with the weight of 300-400 g) (Nomura Siam International Co., Ltd., Bangkok, Thailand) were housed in groups of two and given 2 weeks to acclimate to the housing environment (temperature:22±2°C, humidity: 50±20% and a standard 12:12 light: dark cycle). Eighteen, fourteen-week-old, male Wistar rats were randomly divided into 2 groups: groups of day 1 and day 7. Five types of DDMN samples were tested for skin irritation using in vivo skin at the back of rats. These groups were 1) low dose-vitC-gluta-350-DDMN patch, 2) high dose-vitC-gluta-350-DDMN patch), 3) high dose-vitC-gluta-550-DDMN patch, and 4 and 5) unloaded DDMNs of 350 and 550 μm needle lengths.
Prior to DDMN administration, rats were anesthetized. The hairs on the dorsal back skin were trimmed with hair clipper (less than 1 mm length) and the shaved skin was cleaned and dry. DDMNs were administered (as described above) randomly by a veterinarian using double blind technique, regarding the location at the back of the rats. Gross and histopathological aspects were observed. Each animal was randomly administered with maximum 3 DDMN patches on its back.
For gross pathology, the gross lesions were assessed over the signs of skin reaction on day 1 and day 7 post single administration using the procedure of ISO 10993-10: Biological evaluation of medical devices-Part 10, Tests for irritation and skin sensitization (Draize test) and OECD 404 guideline for testing of chemicals (acute dermal irritation and corrosion). The irritation score concerning the signs of skin erythema and edema after DDMN application was calculated by dividing the sum of irritation scores of all animals in each group with the total number of animals in that group. All irritation scores, the cumulative irritation indexes, were categorized according to the following criteria: 0-0.4: negligible irritation, 0.5-1.9: slight irritation, 2-4.9: moderate irritation, 5-8: severe irritation, by 2 independent veterinary pathologists. The scoring system used for skin reaction is shown in Table S2 in the supplementary information or SI.
For histopathology, the full thickness of skins from all experimental groups on day 1 and day 7 post single administration were collected and fixed in 10% buffered formalin before processed by the routine hematoxylin&eosin (H&E) staining and observed under the light microscope by 2 independent veterinary pathologists using single blind technique. The skin samples were evaluated in terms of epidermal and dermal parameters that modified from Machtinger et al. (2004)54 and ISO 10993-10:2010.
Treatment of hyperpigmentation in human
Hypopigmentation efficacy of DDMN patches was evaluated in human volunteers through a randomized triple blind-controlled study, conducted at Srinakharinwirot skin center, Bangkok, Thailand from July 2021 to August 2021. The study was approved by the research ethic committee of Srinakharinwirot University (SWEC/F-095/2564) and was later registered in the Thai Clinical Trial Registry (TCTR20210903004) on 03/09/2021. All methods were performed in accordance with the relevant guidelines and regulations of the Declarations of Helsinki. Patients who have had allergic reaction to vitC or glutathione, history of keloid, concurrent skin infection on face, previous used of any medication and laser that have whitening effect (hydroquinone, arbutin, azelaic acid, kojic acid, licorice extract, Lakoocha extract, vitC, vitamin E, soybean extract, resorcinol, alpha/beta-hydroxy acid, nicotinamide) for 1 week before enrolled, were excluded. Seventeen patients were enrolled, and eleven patients completed the study. Six patients could not complete the follow up due to COVID-19 pandemics situation in Bangkok, Thailand. The patients were age over 18 years old who had more than 3 lesions of post inflammatory hyperpigmentation from acne on face. Informed written consent was obtained from each volunteer patient before the study.
The post inflammatory hyperpigmented lesions on the face of all patients were divided into three groups, lesion A, B and C, by a computer-generated random sequence: lesion A was treated by unloaded-350-DDMNs, lesion B was treated by high dose-vitC-gluta-350-DDMN patch, and lesion C was treated by high dose-vitC-gluta-550-DDMN patch.
Volunteers applied the appropriated DDMNs on the designated clean and dry studied area by themselves. DDMN patch was applied daily before bedtime for 4 weeks. During the entire study period, volunteers had normal routine of skin cleansing and makeup. Personal and medical history was obtained at baseline. Colorimeter was used to assess melanin index and luminance (L*) at week 0 (baseline), 1, 2, 3 and 4. Melanin index was represented the skin darkness with scale ranging from 0 (white) to 100 (total black). The L* was represented skin lightness with scale ranging from 0 (total black) to 100 (white). The percentage of melanin reduction and L* reduction was calculated as follows:
Percentage of melanin reduction = [(melanin index W0 – melanin index Wt) / melanin index W0)] x100
Percentage of L* reduction = [(L* W0 – L* Wt)/ L* W0)] x100
Where W0 was week 0, and Wt was week of the measurement.
High number of the percentage of melanin reduction corresponds to large decrease in melanin.
Low number of the percentage of L* reduction corresponds to large increase in skin lightening.
Statistical analysis
The data from animal study were analyzed by one-way ANOVA followed by the Tukey’s multiple comparison post hoc test using GraphPad Prism (version 5.03) software (GraphPad, USA). The data from human clinical study were analyzed by mixed linear model using STATA (version 14.0) software (Stata Corp LLC, USA). Differences were considered significant at p<0.05.