Objective: To explore the optimal ratio of dihydrotestosterone and hydroxyflutamide (hereinafter referred to as DH), construct a dual release system of androgen and its antagonist, and analyze the application effect of this system in the repair of full-thickness burn wounds in mice. Methods: This study was an experimental study. The HaCaT cells were divided into blank group (without drug culture), low baseline group, medium baseline group, and high baseline group according to the random number table (the same grouping method below), and the last three groups of cells were cultured by adding three different ratios of DH. Under a medium ratio, the mass of dihydrotestosterone in the three baseline groups from low to high was 1.4, 2.8, and 4.0 µg, respectively, and the mass of hydroxyflutamide was 1.2, 1.6, and 2.0 µg, respectively. On this basis, under a small ratio, the mass of dihydrotestosterone was reduced by half and the mass of hydroxyflutamide was increased by half; under a large ratio, the mass of dihydrotestosterone was increased by half and the mass of hydroxyflutamide was reduced by half. After culture of 2 days, the cell proliferation level was detected by cell counting kit 8 ( n =4). Sixteen 6-8-week-old male BALB/c mice were used to establish a full-thickness burn wound on the back and divided into blank group, small ratio group, medium ratio group, and large ratio group, with 4 mice in each group. On post injury day (PID) 7, normal saline containing different ratios of DH was locally dropped to the wounds of mice in the last three groups of mice (the total mass of DH in the three ratio groups from small to large was 127.5, 165.0, and 202.5 µg, respectively, and the mass ratios of dihydrotestosterone to hydroxyflutamide (hereinafter referred to as drug mass ratio) were 8∶9, 8∶3, and 8∶1, respectively), afterwards, the administration was repeated every 48 hours until PID 27; normal saline was dropped to the wound of mice in blank group at the aforementioned time points. The wound healing status on PID 0 (immediately), 7, 14, 21, and 28 was observed, and the wound healing rates on PID 7, 14, 21, and 28 were calculated ( n =4). On PID 28, the wound tissue was taken, which was stained with hematoxylin and eosin for observing re-epithelialization and with Masson for observing collagen fibers, and the proportion of collagen fibers was analyzed ( n =3). Twenty 6-8-week-old male BALB/c mice were used to establish a full-thickness burn wound on the back and divided into ordinary scaffold group, small proportion scaffold group, medium proportion scaffold group, and large proportion scaffold group (with 5 mice in each group). On PID 7, the wound was continuously dressed with a polycaprolactone scaffold without drug and a polycaprolactone scaffold containing DH with a drug mass ratio of 1∶3, 1∶1, or 3∶1 (i.e. the dual release system of androgen and its antagonist, with total mass of DH being about 1.7 mg) prepared by using electrospinning technology until the end of the experiment. Histopathological analyses of tissue ( n =3) at the same time points as those in the previous animal experiment were performed. On PID 7 and 14, the wound exudates were collected and the relative abundance of bacterial communities was analyzed using 16S ribosomal RNA high-throughput sequencing ( n =3). Results: After culture of 2 days, under a small ratio, the proliferation levels of HaCaT cells in low baseline group and high baseline group were significantly higher than the level in blank group ( P <0.05). As the time after injury prolonged, the wounds of all four groups of mice continued to shrink. On PID 14, the wound healing rate of mice in large ratio group was 72.5% (61.7%, 75.1%), which was close to 53.3% (49.5%, 64.4%) in blank group ( P >0.05); the wound healing rates of mice in small and medium ratio groups were 74.2% (71.0%, 84.2%) and 70.4% (65.1%, 74.4%), respectively, which were significantly higher than the rate in blank group (with both Z values being -2.31, P <0.05). On PID 21, the wound healing rate of mice in small ratio group was significantly higher than that in blank group ( Z =-2.31, P <0.05). On PID 28, the wounds of mice in the three ratio groups were completely re-epithelialized and the epidermis was thicker than that in blank group; compared with that in blank group, the collagen fiber content in the wound tissue of mice in the three ratio groups was higher and arranged more orderly, and the proportions of collagen fibers in the wound tissue of mice in small and large ratio groups were significantly increased ( P <0.05). On PID 28, the wounds of mice in ordinary scaffold group were partially epithelialized, while the wounds of mice in the three proportion scaffold groups were almost completely epithelialized. Among them, the wounds of mice in small proportion scaffold group had the thickest epidermis. The proportion of collagen fibers in the wound tissue of mice in small proportion scaffold group was significantly increased compared with that in ordinary scaffold group ( P <0.05). On PID 7, the bacterial communities with high relative abundance in the wound exudation of mice in the four groups included bacteria of Corynebacterium , Staphylococcus , and Rhodococcus . On PID 14, the bacterial communities with high relative abundance in the wound exudation of mice in the four groups included bacteria of Stenotrophomonas , Rhodococcus , and Staphylococcus , and the number of bacterial species in the wound exudation of mice in the three proportion scaffold groups was more than that in ordinary scaffold group. Conclusions: When the drug mass ratio is relatively small, DH has the effect of promoting the proliferation of HaCaT cells. The ratio of 8∶9 is the optimal mass ratio of dihydrotestosterone to hydroxyflutamide, and DH with this mass ratio can promote re-epithelialization and collagen deposition of full-thickness burn wounds in mice, and promote wound healing. The constructed dual release system of androgen and its antagonist with DH in a 1∶3 drug mass ratio contributes to the re-epithelialization and collagen deposition of the full-thickness burn wounds in mice, and can improve the diversity of wound microbiota.