Exosome microneedle transdermal drug release simulation method based on sustained-release kinetic model

Abstract

The application discloses an exosome microneedle transdermal drug release simulation method based on a slow-release kinetic model, and belongs to the technical field of biological medicine simulation. The soluble microneedle co-loaded with engineered exosomes and finasteride is used to obtain in-vitro drug release experimental data, and a variety of drug release kinetic models are used to fit the optimal model and parameters. A three-dimensional geometric model containing a multi-layer skin and a hair follicle structure is constructed, a diffusion-release coupling partial differential equation set is established, the time and space concentration distribution of the drug around the skin and the hair follicle is obtained through finite element solution, the output of the effective concentration maintenance time around the hair follicle and the whole body exposure amount prediction value is obtained, and multi-target prescription parameter optimization can be carried out. The application also discloses a simulation system and a computer readable storage medium. According to the simulation parameter calibrated according to the actual measurement experiment, the mass transfer difference between the small molecule drug and the nanocapsule is distinguished, the digital prediction of the microneedle preparation can be realized, the physical trial production is reduced, and simulation support is provided for the research and development of the androgenic alopecia microneedle drug preparation.

Description

Technical Field

[0001] This invention relates to the field of biomedical simulation technology, specifically to a simulation method for transdermal drug delivery via exosome microneedles based on a sustained-release kinetic model. Background Technology

[0002] Androgenetic alopecia is the most common type of hair loss in clinical practice. Its pathological root lies in the continuous attack of hair follicles by dihydrotestosterone (DHT), leading to follicle miniaturization, decreased activity of hair follicle stem cells, and ultimately, progressive hair loss. Current clinical treatments include oral finasteride, which inhibits 5α-reductase to lower DHT levels and promote hair regeneration; however, systemic administration can easily cause adverse reactions such as sexual dysfunction. Topical minoxidil suffers from low transdermal penetration and strong local irritation. Soluble microneedles, leveraging their minimally invasive transdermal advantages, can directly deliver exosomes and small-molecule drugs to the hair follicles in the scalp dermis. Mesenchymal stem cell-derived engineered exosomes can activate hair follicle stem cells, repairing damaged follicles. Co-loading exosomes and finasteride onto soluble microneedles can simultaneously achieve hair follicle repair and local protection, avoiding the systemic side effects of oral administration. This approach holds great potential in the local treatment of androgenetic alopecia. However, the composition of the microneedle matrix, the geometry of the needle, and the ratio of the two active substances to drug loading can all directly alter the release and diffusion behavior of exosomes and finasteride within the scalp. Traditional R&D models mainly rely on extensive animal experiments and in vitro transdermal tests to screen formulations. These experiments are time-consuming, costly, and it is difficult to directly observe the actual concentration of the drug in the dermis and hair follicle area over time. Furthermore, formulation optimization lacks theoretical simulation tools to support it.

[0003] Numerous academic publications have explored transdermal drug delivery via exosome microneedles. Song et al., in their study published in *Nano-Micro Letters*, constructed an exosome microneedle delivery patch and investigated the impact of microneedle structure on the transdermal depth and release rate of exosomes, demonstrating that microneedles can significantly improve the skin delivery efficiency of nanovesicles. In the area of ​​drug delivery simulation, Bhatnagar et al. conducted finite element method (FEM) simulations of transdermal drug delivery via microneedles, using kinetic models to simulate the release and diffusion process of small molecule drugs in the microneedle-skin system, providing a theoretical reference for microneedle formulation design. However, most existing academic reports simply treat exosomes as ordinary nanoparticles for simulation, ignoring the particle size sieving and diffusion-limited effect of exosome vesicles during release from the hydrogel matrix. When the simulation object is a two-component co-delivery system of "nanovesicle exosomes + small molecule finasteride", the existing simulation framework fails to distinguish the release mechanisms of the two substances with huge differences in physicochemical properties. It lacks a coupled kinetic model that can simultaneously take into account matrix erosion, vesicle-limited diffusion, tissue metabolic clearance, and capillary absorption. It also lacks quantitative simulation evaluation methods for effective concentration in the hair follicle target area and systemic exposure risk. Relying solely on existing academic simulation schemes, it is difficult to directly use them for prescription prediction and parameter optimization of co-loaded exosome-finasteride hair growth microneedles.

[0004] Domestic patents have also been published. Patent CN113694005A discloses a soluble microneedle array loaded with stem cell exosomes. This patent enables the loading of stem cell exosomes into hyaluronic acid microneedles for skin repair, and discloses the microneedle material composition, preparation process, and basic in vitro efficacy evaluation methods, which has to some extent promoted the development of exosome microneedle products. However, this patent and similar existing patents all focus on the formulation and preparation process of the microneedle product itself, relying solely on in vitro and in vivo experiments to verify performance. They have not established corresponding drug release kinetic simulation methods, making it impossible to numerically predict the spatiotemporal concentration distribution of exosomes and small molecule drugs within the microneedles in the multi-layered scalp tissue and around hair follicles before formulation preparation; nor can they quantitatively predict the systemic exposure level resulting from drug absorption via capillaries after local administration, making it impossible to complete multi-objective formulation reverse optimization in the early stages of research and development. In other words, while existing patents have solved the technical problems of physical preparation of exosome microneedles, there are still significant shortcomings in the digital simulation and prediction of formulation development. It is still necessary to repeatedly prepare multiple batches of samples to carry out trial and error experiments, resulting in high R&D costs and long iteration cycles. Summary of the Invention

[0005] To address the aforementioned technical issues, this application discloses a simulation method for transdermal drug delivery via exosome microneedles based on a sustained-release kinetic model, comprising the following steps:

[0006] A soluble microneedle array loaded with exosomes and finasteride was constructed. The soluble microneedle array includes a substrate and multiple microneedles arranged in an array on the substrate. The microneedles are composed of a soluble polymer matrix material, engineered exosomes loaded with TGF-β, and finasteride.

[0007] To obtain in vitro drug release experimental data, the soluble microneedle array was placed in a transdermal diffusion experimental device, and the release liquid was collected at different time points. The cumulative release amount of exosomes and finasteride was measured to obtain the in vitro cumulative release curve.

[0008] A sustained-release kinetic model was established, and multiple drug release kinetic models were used to fit the obtained in vitro cumulative release curves for the release behavior of exosomes and finasteride. The drug release kinetic models included at least the zero-order release model, the first-order release model, the Higuchi model, and the Korsmeyer-Peppas model. The optimal sustained-release kinetic model for exosomes and finasteride and its corresponding model parameters were determined based on the goodness of fit. The model parameters included the release rate constant and the release exponent.

[0009] A three-dimensional geometric model of the scalp skin is constructed. Based on the histological structure of the scalp skin, a multi-layer three-dimensional geometric model including the stratum corneum, active epidermis, dermis and hair follicle structure is constructed. The initial insertion position, insertion depth and initial drug concentration of the microneedle body are set as initial conditions in the three-dimensional geometric model.

[0010] A set of partial differential equations for transdermal drug release was established. Based on the determined optimal sustained-release kinetic model and combined with Fick's diffusion law, a set of diffusion-release coupled partial differential equations for exosomes and finasteride in each layer of the multilayer three-dimensional geometric model was established. The set of partial differential equations includes the diffusion coefficient, partition coefficient and degradation rate parameters of the microneedle matrix of each tissue layer.

[0011] The constructed three-dimensional geometric model was meshed, boundary conditions were set, and the finite element method was used to numerically solve the established partial differential equations to obtain the spatiotemporal concentration distribution of exosomes and finasteride in different layers of scalp skin and tissues around hair follicles.

[0012] The duration of effective concentration of exosomes and finasteride around hair follicles in the dermis, and the predicted systemic exposure of the drug through transdermal absorption into the capillary network.

[0013] Preferably, the construction of the soluble microneedle array loaded with exosomes and finasteride specifically includes:

[0014] Mesenchymal stem cells were overexpressed with TGF-β by lentiviral transfection, and the cell culture supernatant was collected and engineered exosomes loaded with TGF-β were extracted by ultracentrifugation.

[0015] Finasteride was dissolved in an organic solvent to prepare a finasteride solution, and the engineered exosomes loaded with TGF-β were resuspended in a buffer solution to prepare an exosome suspension.

[0016] Hyaluronic acid, polyvinylpyrrolidone and sodium carboxymethyl cellulose were dissolved in deionized water to prepare a polymer matrix solution.

[0017] The finasteride solution and the exosome suspension were added sequentially to the polymer matrix solution, stirred and mixed, and then injected into the PDMS microneedle mold. The mixture was centrifuged to fully fill the micropores of the mold. After scraping off the excess solution, the mixture was dried and molded, and then demolded to obtain the soluble microneedle array.

[0018] Preferably, the establishment of the sustained-release kinetic model involves fitting the obtained in vitro cumulative release curve using multiple drug release kinetic models, specifically targeting the release behavior of finasteride from exosomes. This includes:

[0019] Using the zero-level release model according to the formula The in vitro cumulative release curve was fitted, wherein... for Cumulative release percentage at any given time This is the initial release amount. The zero-order release rate constant;

[0020] Using the first-order release model according to the formula The in vitro cumulative release curve was fitted, wherein... This is the first-order release rate constant;

[0021] Using the Higuchi model according to the formula The in vitro cumulative release curve was fitted, wherein... Higuchi release rate constant;

[0022] Using the Korsmeyer-Peppas model according to the formula The in vitro cumulative release curve was fitted, wherein... The Korsmeyer-Peppas release rate constant. To release the index, when The release mechanism is dominated by Fick diffusion, when The time is a synergistic mechanism of anomalous diffusion and matrix dissolution, when At that time, matrix dissolution was the dominant process;

[0023] Calculate the coefficient of determination for each drug release kinetic model. Select the AIC value of the Akaike Information Criterion. The model with the largest and smallest AIC value is taken as the optimal sustained-release kinetic model for the corresponding drug.

[0024] Preferably, the construction of the three-dimensional geometric model of the scalp skin is based on the histological structure of the scalp skin, constructing a multi-layered three-dimensional geometric model including the stratum corneum, active epidermis, dermis, and hair follicle structure, specifically including:

[0025] The stratum corneum is set as a flat layer with a thickness of 10-20 μm, the active epidermal layer is set as a flat layer with a thickness of 50-100 μm, and the dermis is set as a flat layer with a thickness of 1000-2000 μm.

[0026] A cylindrical hair follicle structure is constructed in the dermis, the diameter of the hair follicle structure is set to 100-300μm, and the depth extends from the surface of the stratum corneum to the interior of the dermis 500-1500μm;

[0027] The microneedle body is modeled as a cone array, with the bottom diameter of the cones being 200-400 μm, the height being 500-800 μm, and the center-to-center distance between adjacent cones being 500-1000 μm;

[0028] The tip coordinates of the microneedle are set to the target area next to the hair follicle structure, so that the horizontal distance between the tip of the microneedle and the hair follicle structure is 50-200μm.

[0029] Preferably, the establishment of the partial differential equations for transdermal drug release, based on the determined optimal sustained-release kinetic model and combined with Fick's diffusion law, specifically includes establishing a diffusion-release coupled partial differential equation system for exosomes and finasteride in each layer of the multilayer three-dimensional geometric model, including:

[0030] A diffusion equation based on matrix dissolution coupling was established to investigate the release process of finasteride in the microneedle matrix:

[0031]

[0032] in, This refers to the local concentration of finasteride. The position- and time-dependent effective diffusion coefficient tensor of finasteride. For the time-varying degradation rate function of the microneedle matrix, The source term function for the dissociation of finasteride from the matrix-bound state to the free state;

[0033] To investigate the release process of exosomes from the microneedle matrix, a confined diffusion equation based on particle size sieving effect was established:

[0034]

[0035] in, This refers to the local concentration of exosomes. For depends on the instantaneous porosity of the matrix Equivalent hydrodynamic radius of exosomes The effective diffusion coefficient, The inactivation decay constant of exosomes in the tissue environment;

[0036] The effective diffusion coefficient Calculate as follows:

[0037]

[0038] in, The diffusion coefficient of exosomes in free aqueous solution is given. The porosity tortuosity correction index. The particle size attenuation coefficient is the sieve size reduction factor. As a matrix in Average mesh size at any given time;

[0039] A stratified diffusion equation was established to investigate the diffusion process of finasteride and exosomes in different layers of scalp skin:

[0040]

[0041] Among them, superscript Indicates the skin tissue layer. subscript Indicates the type of drug. , For drugs At the organizational level The diffusion coefficient in For drugs At the organizational level The metabolic clearance rate constant in the figure.

[0042] Preferably, the step of meshing the constructed three-dimensional geometric model, setting boundary conditions, and numerically solving the established partial differential equations using the finite element method specifically includes:

[0043] An adaptively refined grid is used in the microneedle area and the area surrounding the hair follicle structure, while a coarser grid is used in the peripheral area of ​​the dermis away from the microneedles and hair follicles. The total number of grid cells is controlled between 500,000 and 2 million.

[0044] A flux-free boundary condition is set on the outer surface of the stratum corneum, as shown in the formula. ,in The boundary normal vector;

[0045] A convection clearance boundary condition is set at the junction of the dermis and the capillary network, as shown in the formula. ,in The capillary wall permeability coefficient, The initial setting for the drug concentration in the blood is zero.

[0046] At the interface between adjacent tissue layers, concentration continuity and flux conservation conditions are set, and the formula is as follows: and ,in This is the distribution coefficient of the drug between two adjacent tissue layers;

[0047] The implicit backward Euler time discretization scheme was adopted, with the time step set to adaptive variable step size. The initial time step size was 1 minute, the maximum time step size was 30 minutes, and the total simulation time was set to 72 hours.

[0048] Preferably, the effective concentration maintenance time of the exported exosomes and finasteride around the hair follicles in the dermis, and the predicted systemic exposure of the drug through transdermal absorption into the capillary network, specifically include:

[0049] Extract the finasteride concentration time series data of each grid node within a 50-200μm range around the hair follicle structure, and calculate the duration for which the finasteride concentration remains above a preset effective concentration threshold, as the effective concentration maintenance time of finasteride. ;

[0050] Time-series data of exosome concentration were extracted from each grid node within a 50-200 μm range around the hair follicle structure. The duration during which the exosome concentration remained above a preset effective concentration threshold was calculated as the effective concentration maintenance time of exosomes. ;

[0051] At the capillary boundary, the predicted systemic exposure to finasteride is calculated as follows:

[0052]

[0053] in, This refers to the total body exposure. Total simulation duration, For the curved surface of capillary boundaries, The concentration of finasteride at the dermal capillary interface;

[0054] The Compared with the reference value for systemic exposure to oral finasteride, when When the systemic safety of local microneedle administration is determined to be satisfactory when the proportion is below the preset threshold of the systemic exposure reference value of the oral finasteride.

[0055] Preferably, the method further includes a parameter optimization step for the sustained-release kinetic model:

[0056] The optimization objective is to maximize the sum of the effective concentration maintenance time of the exosomes and finasteride around the hair follicles. The optimization variables are the loading concentration of finasteride in the microneedle, the loading concentration of the exosomes, and the ratio of the polymer matrix material. The systemic exposure of finasteride does not exceed a preset safety threshold as a constraint. A multi-objective optimization model is constructed.

[0057] The multi-objective optimization model is iteratively solved using a particle swarm optimization algorithm or a genetic algorithm to obtain the optimal combination of microneedle prescription parameters that achieves synergistic synergy between hair follicle repair and protection.

[0058] The exosome microneedle transdermal drug delivery simulation system based on a sustained-release kinetic model is applicable to any of the exosome microneedle transdermal drug delivery simulation methods based on a sustained-release kinetic model, including:

[0059] The microneedle construction module constructs a soluble microneedle array loaded with exosomes and finasteride. The soluble microneedle array includes a substrate and multiple microneedles arranged in an array on the substrate. The microneedles are composed of a soluble polymer matrix material, engineered exosomes loaded with TGF-β, and finasteride.

[0060] The drug release data acquisition module places the soluble microneedle array in a transdermal diffusion experimental device, collects the release liquid at different time points, and measures the cumulative release amount of exosomes and finasteride to obtain the in vitro cumulative release curve.

[0061] The kinetic modeling module uses multiple drug release kinetic models to fit the obtained in vitro cumulative release curves for exosome and finasteride release behavior, and determines the optimal sustained-release kinetic model and its corresponding model parameters for exosome and finasteride based on the goodness of fit.

[0062] The geometric modeling module is based on the histological structure of the scalp skin, constructing a multi-layer three-dimensional geometric model including the stratum corneum, active epidermis, dermis and hair follicle structure, and setting the initial insertion position, insertion depth and initial drug concentration of the microneedle body as initial conditions.

[0063] Based on the determined optimal sustained-release kinetic model and combined with Fick's diffusion law, the equation construction module establishes a set of diffusion-release coupled partial differential equations for exosomes and finasteride in each layer of the multi-layer three-dimensional geometric model.

[0064] The numerical solution module meshes the constructed three-dimensional geometric model, sets boundary conditions, and uses the finite element method to numerically solve the established partial differential equations to obtain the spatiotemporal concentration distribution results of exosomes and finasteride in various layers of scalp skin and tissues around hair follicles.

[0065] Based on the numerical solution, the results analysis module outputs the effective concentration maintenance time of exosomes and finasteride around the hair follicles in the dermis, as well as the predicted value of the systemic exposure of the drug through transdermal absorption into the capillary network.

[0066] A computer-readable storage medium storing a computer program, wherein when executed by a processor, the computer program implements the steps of a simulation method for transdermal drug delivery using exosome microneedles based on a sustained-release kinetic model.

[0067] Compared with the prior art, the technical solution of this application has the following technical effects:

[0068] The sustained-release kinetic model constructed in this invention can complete the model parameter calibration based on in vitro drug release experimental data. By fitting and screening multiple drug release kinetic models, suitable release models for exosomes and finasteride are obtained. This method can closely match the release behavior of real samples, ensuring the accuracy of subsequent numerical simulations and preventing model parameters from deviating from the actual formulation release characteristics.

[0069] This invention constructs a three-dimensional geometric model of the scalp skin, including the stratum corneum, active epidermis, dermis, and hair follicle structure, which can reproduce the real multi-layered tissue structure of the scalp and the spatial morphology of hair follicles. By setting the geometry of the microneedle and the initial conditions for insertion, the spatial positional relationship of the microneedles after they act on the scalp can be reproduced, providing a geometrically simulated carrier that conforms to physiological reality for the drug diffusion process.

[0070] This invention establishes multiple sets of diffusion-release coupled partial differential equations to describe the finasteride matrix dissolution and release process and the restricted diffusion behavior of exosomes due to particle size sieving, while also incorporating the metabolic clearance effects of various skin tissue layers. It can distinguish the mass transfer patterns of two substances with significantly different physicochemical properties within the microneedle matrix and skin tissue, and fully reconstruct the mass transfer changes of each substance.

[0071] This invention employs the finite element method to numerically solve the equations, focusing on the microneedles and the surrounding area of ​​hair follicles through mesh refinement, and using appropriate boundary conditions and time discretization schemes for calculation. It can stably output the effective concentration maintenance time around hair follicles and the predicted values ​​of systemic exposure, obtaining objective simulation output results and achieving digital characterization of the effects of microneedle drug delivery.

[0072] This invention employs a multi-objective parameter optimization process, using the effect on the hair follicle periphery as the optimization objective and the systemic exposure level as a constraint for iterative solution. It can output microneedle prescription parameter combinations adapted to application requirements, providing a reference direction for microneedle formulation development, facilitating rapid formulation screening, and supporting the research and development of hair growth microneedle formulations.

[0073] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the preferred embodiments of this application are described in detail below with reference to the accompanying drawings.

[0074] The above and other objects, advantages and features of this application will become more apparent to those skilled in the art from the following detailed description of specific embodiments in conjunction with the accompanying drawings. Attached Figure Description

[0075] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In all drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn to scale.

[0076] Based on the description of the figures and their corresponding technical content in the document, the titles of the figures are as follows:

[0077] Figure 1 This is a flowchart of the overall simulation method for transdermal drug delivery via exosome microneedles based on a sustained-release kinetic model;

[0078] Figure 2 A three-dimensional geometric model of the scalp and skin, and a spatial coupling architecture diagram of microneedles and hair follicles;

[0079] Figure 3 This is a module architecture diagram of a finite element method for solving the partial differential equations for transdermal drug delivery.

[0080] Figure 4 This is a graph showing the in vitro cumulative release curve of finasteride and exosomes and the fitting verification of the kinetic model.

[0081] Figure 5 Simulation verification diagram of microneedle matrix degradation kinetics and spatiotemporal concentration field of drug in dermis;

[0082] Figure 6Pareto frontier and optimal prescription validation plot for multi-objective optimization of effective concentration maintenance time in hair follicle target area and systemic exposure;

[0083] Figure 7 This is a flowchart illustrating the engineered exosome acquisition process according to an embodiment of the present invention.

[0084] Figure 8 This is a schematic diagram illustrating the preparation process of a soluble microneedle array loaded with exosomes and finasteride according to an embodiment of the present invention.

[0085] Figure 9 This is a schematic diagram illustrating the drug delivery process of soluble microneedles inserted into the scalp skin according to an embodiment of the present invention;

[0086] Figure 10 A comparative schematic diagram of different types of microneedle structures and transdermal drug delivery methods;

[0087] Figure 11 This is a simulation system architecture diagram for transdermal drug delivery via exosome microneedles based on a sustained-release kinetic model. Detailed Implementation

[0088] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. In the following description, specific details such as specific configurations and components are provided merely to help fully understand the embodiments of this application. Therefore, those skilled in the art should understand that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this application. In addition, for clarity and brevity, descriptions of known functions and structures are omitted in the embodiments.

[0089] It should be understood that the phrase "an embodiment" or "this embodiment" throughout the specification means that a specific feature, structure, or characteristic related to the embodiment is included in at least one embodiment of this application. Therefore, "an embodiment" or "this embodiment" appearing throughout the specification does not necessarily refer to the same embodiment. Furthermore, these specific features, structures, or characteristics can be combined in any suitable manner in one or more embodiments.

[0090] Furthermore, reference numerals and / or letters may be repeated in different examples within this application. Such repetition is for the purpose of simplification and clarity and does not in itself indicate a relationship between the various embodiments and / or settings discussed.

[0091] In this article, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can mean: A exists alone, B exists alone, and A and B exist simultaneously. The term " / and" in this article describes another type of relationship between related objects, indicating that two relationships can exist. For example, A / and B can mean: A exists alone, and A and B exist alone. In addition, the character " / " in this article generally indicates that the related objects before and after it are in an "or" relationship.

[0092] In this article, the term "at least one" is merely a description of the relationship between related objects, indicating that there can be three relationships. For example, "at least one of A and B" can mean: A exists alone, A and B exist simultaneously, or B exists alone.

[0093] It should also be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion.

[0094] Example 1 mainly describes a simulation method for transdermal drug delivery via exosome microneedles based on a sustained-release kinetic model, such as... Figure 1 As shown, a soluble microneedle array loaded with exosomes and finasteride was constructed. The soluble microneedle array includes a substrate and multiple microneedles arranged in an array on the substrate. The microneedles are composed of a soluble polymer matrix material, engineered exosomes loaded with TGF-β, and finasteride.

[0095] To obtain in vitro drug release experimental data, the soluble microneedle array was placed in a transdermal diffusion experimental device, and the release liquid was collected at different time points. The cumulative release amount of exosomes and finasteride was measured to obtain the in vitro cumulative release curve.

[0096] A sustained-release kinetic model was established, and multiple drug release kinetic models were used to fit the obtained in vitro cumulative release curves for the release behavior of exosomes and finasteride. The drug release kinetic models included at least the zero-order release model, the first-order release model, the Higuchi model, and the Korsmeyer-Peppas model. The optimal sustained-release kinetic model for exosomes and finasteride and its corresponding model parameters were determined based on the goodness of fit. The model parameters included the release rate constant and the release exponent.

[0097] A three-dimensional geometric model of the scalp skin is constructed. Based on the histological structure of the scalp skin, a multi-layer three-dimensional geometric model including the stratum corneum, active epidermis, dermis and hair follicle structure is constructed. The initial insertion position, insertion depth and initial drug concentration of the microneedle body are set as initial conditions in the three-dimensional geometric model.

[0098] A set of partial differential equations for transdermal drug release was established. Based on the determined optimal sustained-release kinetic model and combined with Fick's diffusion law, a set of diffusion-release coupled partial differential equations for exosomes and finasteride in each layer of the multilayer three-dimensional geometric model was established. The set of partial differential equations includes the diffusion coefficient, partition coefficient and degradation rate parameters of the microneedle matrix of each tissue layer.

[0099] The constructed three-dimensional geometric model was meshed, boundary conditions were set, and the finite element method was used to numerically solve the established partial differential equations to obtain the spatiotemporal concentration distribution of exosomes and finasteride in different layers of scalp skin and tissues around hair follicles.

[0100] Based on the numerical solution results, the effective concentration maintenance time of exosomes and finasteride around the hair follicles in the dermis, the concentration distribution cloud map, and the predicted value of the systemic exposure of the drug after transdermal absorption into the capillary network are output to evaluate the repair and protection effect of local microneedle administration on hair follicles.

[0101] Engineered exosomes loaded with TGF-β were prepared. Human umbilical cord mesenchymal stem cells (hUC-MSCs) were overexpressed with transforming growth factor-β (TGF-β) via lentiviral transfection. The third-generation self-inactivated lentiviral vector pLVX-TGFβ1-Puro was used. After packaging the lentiviral vector into HEK293T cells, the viral titer was determined using the limiting dilution method and controlled within a certain range. Within the range.

[0102] Third- to fifth-generation hUC-MSCs were seeded in 6-well plates at a density of 2 × 10^5 cells / well. When the cell confluence reached 50%-60%, lentiviral suspension was added at ratios of MOI of 10, 15, 20, 25, and 30, respectively. Polybrene is used to improve infection efficiency at 37°C and 5%. After culturing under the specified conditions for 24 hours, the medium was replaced with fresh complete medium, and culturing continued for another 48 hours. Then, [the following was added] After 72 hours of screening with puromycin, the surviving cells are engineered mesenchymal stem cell lines that stably overexpress TGF-β (denoted as TGFβ1-MSCs).

[0103] Lentiviral transfection efficiency With MOI and infection time The changes satisfy the following first-order dynamic relationship: ,in, The upper limit of saturated transfection efficiency (actual measurement is) ), The infection rate constant (measured as follows) ), This represents the effective contact time for infection. When MOI=20, At that time, theoretical transfection efficiency The actual flow cytometry results were as follows: This matches the theoretical value.

[0104] TGFβ1-MSCs as The culture medium was seeded at a density of 70%–80% in T175 culture flasks for expansion. After washing three times with PBS, the medium was replaced with serum-free exosome-specific medium (DMEM / F12 containing exosome-free fetal bovine serum substitute). The culture was then continued at 37°C with a 5% concentration of... Incubate for 48–72 hours under the specified conditions. Collect the cell culture supernatant and centrifuge at 4°C using the following gradient procedure:

[0105]

[0106] In the fourth step, after centrifugation, the supernatant was discarded, and the precipitate was resuspended in pre-cooled PBS buffer (pH 7.4, containing 0.1% BSA). The precipitate was then washed and purified by ultracentrifugation at 100,000×g for 70 minutes. Finally, the precipitate was resuspended in an appropriate amount of PBS to obtain a high-purity engineered exosome suspension loaded with TGF-β.

[0107] like Figure 7 The diagram shows the complete process flow for obtaining engineered exosomes in this embodiment, which includes four steps in sequence: sample pretreatment, exosome precipitation, exosome resuspension, and exosome purification. After pretreatment to remove cells and large debris, the cell culture supernatant is used to collect the exosome precipitate. The obtained precipitate is resuspended in buffer solution and further purified to remove impurities such as proteins and vesicles, ultimately yielding a high-purity engineered exosome suspension for microneedle drug delivery.

[0108] Exosome recovery rate Calculate as follows:

[0109]

[0110] in, The final particle concentration of the resuspension. The volume of the resuspension liquid. This represents the initial particle concentration of the supernatant. The initial supernatant volume is given. The measured recoveries in three independent experiments were 38.2%, 45.6%, and 51.3%, respectively, with a mean of [missing value]. .

[0111] The particle size distribution and concentration of exosomes were determined using nanoparticle tracking analysis (NTA, NanoSight NS300). Three independent measurements showed that the exosome particle size distribution ranged from 52 to 176 nm, with a peak particle size of [missing value]. The average particle size is Particle concentration is The total protein concentration was determined by the BCA protein quantification method. The protein / particle ratio is Western blot analysis confirmed that exosome marker proteins CD9, CD63, CD81, and TSG101 were positively expressed, while the negative marker Calnexin was negative, meeting the identification criteria of the International Society for Extracellular Vesicles (ISEV).

[0112] The content of TGF-β1 in exosomes was quantitatively detected by ELISA. Before the assay, the exosome membrane was lysed with 0.1% Triton X-100 to release the internal contents, and then the procedure was followed according to the kit instructions. The encapsulation efficiency of TGF-β1 in exosomes was measured. The calculation formula is:

[0113]

[0114] in, The total concentration of TGF-β1 measured after lysis. This represents the concentration of free TGF-β1 in the supernatant of unlysolved exosomes. This represents the original total concentration of TGF-β1 in the cell culture supernatant. (Measured) , , .

[0115] Prepare a finasteride solution. Finasteride (molecular weight 272.38 Da, purity ≥99%) is a white crystalline powder with extremely low solubility in water. About time To improve its dispersion uniformity in the polymer matrix solution, a co-solvent method was used for preparation. Specific procedures: Accurately weigh 200 mg of finasteride powder and add it to 4 mL of anhydrous ethanol. Stir magnetically at 600 rpm for 45 minutes at a constant temperature water bath of 37°C until completely dissolved. Then, slowly add 6 mL of deionized water dropwise while continuing stirring, and continue stirring for 20 minutes to obtain a finasteride solution in an ethanol-water mixture with a volume fraction of 40%. The measured concentration was [missing information]. .

[0116] Solubility of finasteride in ethanol-water mixed solvent With ethanol volume fraction The changes satisfy the extended Jouyban-Acree cosolvent model: ;in, ( ), ( ), , , These are model constants, obtained by fitting experimental data using nonlinear regression. , At that time, the model predicted the solubility as , and the measured value The deviation is 3.2%.

[0117] The prepared finasteride solution was filtered through a 0.22 μm PVDF membrane to remove impurities, and then the pH was adjusted to [value missing] using 0.1 M HCl and 0.1 M NaOH. The osmotic pressure was measured using a freezing point osmometer. It meets the compatibility requirements for subsequent mixing with exosomes.

[0118] Preparation of polymer matrix solution, the soluble polymer matrix material is sodium hyaluronate (HA, molecular weight) Polyvinylpyrrolidone (PVP, K30 grade, molecular weight) ) and sodium carboxymethyl cellulose (CMC-Na, medium viscosity grade, molecular weight It is a compound of three polymers in a specific mass ratio.

[0119] The preparation process is as follows: Weigh 3.0g of sodium hyaluronate powder and slowly sprinkle it into 80mL of deionized water. Incubate at room temperature (…). Under these conditions, mechanically stir at 300 rpm for 3 hours until completely dissolved to form a transparent, viscous HA solution; weigh 5.0 g of PVP powder and add it to the above HA solution, continuing to stir for 1.5 hours until completely dissolved; weigh 1.5 g of CMC-Na powder and slowly add it to the mixed solution, raising the water bath temperature to [temperature missing]. Stir at 400 rpm for 4 hours until completely dissolved, and finally bring the volume to 100 mL with deionized water to obtain a polymer matrix solution with a HA:PVP:CMC mass ratio of 6:10:3 and a total polymer concentration of 19.5% (w / v).

[0120] The rheological properties of the polymer matrix solution were determined using a rotational rheometer (Anton Paar MCR302, cone-plate geometry, cone angle 1°, diameter 50 mm). Characterization under certain conditions. Steady-state shear scan (shear rate) The results show that the matrix solution exhibits shear-thinning pseudoplastic behavior, and its rheological data can be accurately fitted by the Carreau-Yasuda model.

[0121]

[0122] in, Zero shear viscosity (measured) ), Infinite shear viscosity (measured) ), Characteristic relaxation time (measured) ), Yasuda index (measured) ), Power-law exponent (actual measurement) ).

[0123] Zero-shear viscosity of polymer matrix solutions The relationship between the polymer composition and the following mixing rules applies:

[0124]

[0125] in, ( (viscosity of water) , , These are the intrinsic viscosities of each polymer. The mass concentration of each polymer ( ), This is the mixed Huggins constant. Theoretical prediction. The deviation from the measured value is 3.0%.

[0126] The prepared polymer matrix solution was allowed to stand at 4°C for 18 hours to remove bubbles, and then set aside for later use.

[0127] The finasteride solution was slowly added dropwise to the polymer matrix solution at a volume ratio of 1:8, and the mixture was stirred at 250 rpm for 20 minutes to ensure uniform dispersion of finasteride in the polymer matrix. At this point, the concentration of finasteride in the mixture was... Diluting the ethanol volume fraction to approximately 4.4% does not affect the gelation behavior of the polymer matrix.

[0128] The engineered exosome suspension loaded with TGF-β was prepared according to the final concentration. Add to the above mixed solution. To avoid rupture of the exosome membrane structure due to high-speed stirring, this step involves gentle stirring at 100 rpm for 15 minutes. After mixing, take samples for NTA analysis to confirm that the exosome particle size distribution did not change significantly (peak particle size deviation before and after mixing). This indicates that the mixing process did not cause damage to exosomes.

[0129] A PDMS (polydimethylsiloxane, Sylgard184) microneedle negative mold is provided. The negative mold is prepared by casting using an SU-8 photoresist master mold. The micropores are arranged in an inverted conical array, and the single-pore parameters are: opening diameter... ,depth cone angle Hole spacing (center to center) A single array region is There are a total of 100 micropores.

[0130] The drug-loaded mixture is The amount of product was evenly applied to the surface of the PDMS female mold, and then the mold was placed in a centrifuge. (about Centrifuge for 15 minutes under the specified conditions. During centrifugation, the mixture fills the microporous cavity under centrifugal force, reaching a filling depth of... Over time The changes satisfy the following filling dynamics model based on Hagen-Poiseuille flow:

[0131]

[0132] in, For micropore depth, The density of the mixture, The centrifugal angular velocity, The average distance from the micropore to the centrifuge shaft. The micropore opening radius is 1. This represents the effective viscosity of the mixture at the centrifugal shear rate. Due to the high shear rate during centrifugation (approximately...), this is... ), calculated according to the Carreau-Yasuda model Substituting into the calculation yields the characteristic filling time constant. ,when hour, This means the filling depth reached 98.0% of the micropore depth. The measured filling rate was observed using cross-sectional SEM. This is highly consistent with the theoretical value.

[0133] After centrifugation, use a stainless steel scraper to remove excess solution along the surface of the negative mold. Place the filled negative mold in a constant temperature and humidity drying oven. relative humidity Drying was carried out for 18 hours under specific conditions. During the drying process, the solution inside the micropores gradually dehydrated and solidified from the surface to the bottom, and the shrinkage rate of the needle body was controlled within a certain range. .

[0134] After drying, pour the base solution (an aqueous solution containing 15% w / v PVA and 8% w / v PVP) onto the surface of the negative mold. The amount used is... Under the same conditions, the substrate was dried for another 10 hours to form a flexible substrate with sufficient mechanical strength. Finally, it was carefully peeled and demolded from the PDMS negative mold to obtain a complete soluble microneedle array patch.

[0135] like Figure 8 The diagram shows the fabrication process of the soluble microneedle array in this embodiment. The fabrication process is divided into two main stages: microneedle tip preparation and backing layer preparation and microneedle demolding. First, the tip solution is dropped into the PDMS mold, and the solution is fully filled into the mold cavity by vacuuming. After drying, the needle body is formed. Then, the backing solution is dropped into the mold, and the process is repeated by vacuuming and drying to form the substrate structure. Finally, the complete soluble microneedle array is obtained by demolding.

[0136] The microneedle array was characterized using an optical microscope (Olympus BX53) and a scanning electron microscope (SEM, Hitachi SU8010). SEM results showed that the needles were fully formed, with smooth surfaces free of bubbles and defects, and the tip curvature radius was [not specified]. needle base diameter Needle height (Compared to mold depth) (Shrinkage of approximately 1.8%).

[0137] Axial compression tests were performed on a single needle using a TextureAnalyzer (Brookfield CT3) with a stainless steel flat probe, and the loading rate was measured. The force-displacement curve was recorded until the needle broke. The breaking force of a single needle was recorded. The theoretical prediction model is based on the Euler buckling correction formula:

[0138]

[0139] in, The equivalent elastic modulus of the microneedle matrix material (measured by nanoindentation) is ), This is a simplified expression for the average moment of inertia of a conical cross section. For an effective length coefficient (a cone with one end fixed and the other free). Theoretical prediction. The measured average value of 10 needles was All of them are greater than the critical force required to penetrate the skin. .

[0140] The actual drug loading of finasteride and exosomes in the microneedle array was determined by HPLC and NTA, respectively. The actual drug loading of finasteride in a single microneedle array (100 needles) was: (Theoretical feed rate) Drug loading efficiency The actual drug loading capacity of exosomes is (Theoretical feed rate) Drug loading efficiency The reason why exosomes have lower drug loading efficiency than finasteride is that some exosomes are lost due to changes in membrane osmotic pressure during centrifugation and drying.

[0141] like Figure 9 The diagram illustrates the delivery of the prepared soluble microneedle patch to the skin. After the microneedle patch is pressed onto the skin surface, the sharp microneedles penetrate the skin barrier and enter the dermal layer. The needles gradually dissolve upon contact with tissue fluid, releasing and diffusing the encapsulated exosomes and finasteride into the tissue surrounding the hair follicle. Figure 10 As shown, the structure, transdermal and drug release behavior of five types of microneedles, namely solid microneedles, coated microneedles, hollow microneedles, soluble microneedles and porous microneedles, were compared. Compared with other types of microneedles, soluble microneedles release drugs by dissolving the needle matrix itself, without the need for additional drug delivery power and without producing solid microneedle residues, making them suitable for local delivery of active biological components such as exosomes.

[0142] A transdermal diffusion experimental system was constructed, using a modified Franz vertical diffusion cell (PermeGear, USA) as the in vitro transdermal diffusion experimental device. The diffusion cell consists of a donor chamber (volume...) Effective diffusion area ) and receptor chamber (volume) The sample consists of a donor chamber and a recipient chamber, which are fixed together by stainless steel clamps to secure the excised skin sample.

[0143] Preparation of the acceptor solution: Mix PBS buffer (pH 7.4, 0.01M) and anhydrous ethanol at a volume ratio of 75:25, and add 0.05% (w / v) sodium azide as an antibacterial agent. The reason for choosing PBS containing 25% ethanol as the acceptor solution is that finasteride's solubility in pure PBS is only... The sink condition could not be maintained, but the solubility increased after adding 25% ethanol. This is much greater than the highest possible concentration of the acceptor fluid that could be reached during the experiment, ensuring that the leakage conditions are always met.

[0144] Placed at the bottom of the receptor chamber The magnetic stir bar, with its stirring speed calibrated via a tachometer, is as follows: The entire diffusion tank is immersed in a constant-temperature water bath circulation system, with the water bath temperature set at [temperature value missing]. To simulate skin surface temperature, the diffusion cell containing the receptor fluid was subjected to a 30-minute isothermal equilibration process before the experiment to ensure the temperature of the receptor fluid remained stable at a certain level. .

[0145] Fresh pig ear skin (obtained within 4 hours of slaughter) was selected as a transdermal barrier model. The tissue structure of pig ear skin is highly similar to that of human scalp skin, with comparable epidermal thickness, hair follicle density, and lipid composition. Preparation process: After rinsing the pig ear thoroughly with running tap water, the skin was peeled off along the cartilage surface with a scalpel, subcutaneous fat tissue was removed, and the skin was cut to a uniform thickness using an electric skin harvester (Zimmer, Aesculap). .

[0146] Skin samples were taken for histological characterization: fixed in 4% paraformaldehyde, embedded in paraffin, and sectioned (thickness). After staining with H&E, the thickness of each layer was measured under an optical microscope. Measured results: stratum corneum thickness... active epidermal layer thickness Hair follicle density Average hair follicle diameter .

[0147] The prepared skin sample was cut to a diameter... The round, flat pieces were wrapped in aluminum foil and then placed... Store in a refrigerator (shelf life not exceeding 30 days). Before use, thaw at room temperature for 30 minutes, rinse with PBS buffer, and then soak in receptor solution for later use.

[0148] Skin samples were fixed between the donor and recipient chambers of a Franz diffusion cell, with the dermal side facing the recipient chamber. Soluble microneedle arrays were cut to contain... The patch consists of nine needles. It is applied manually by aligning it with the center of the skin and applying pressure. During application, a digital force sensor (Mark-10, ESM303) monitors the pressure in real time and controls the total applied force. (equivalent to a single needle) Press and hold for 45 seconds.

[0149] Depth of microneedle insertion into the skin The insertion depth model, based on elastoplastic contact mechanics, is determined by a combination of factors including pressure, skin mechanics, and needle geometry.

[0150]

[0151] in, Apply pressure evenly to a single needle. The critical puncture pressure (related to the fracture toughness of the stratum corneum). The penetration sharpness index is used. The measured parameters are: , .when hour, That is, the depth of penetration is 98.5%, approximately Fluorescence imaging of frozen sections of skin after application was performed using a confocal laser scanning microscope (CLSM). The measured depth reached by the needle tip was [missing information]. It penetrates the stratum corneum and active epidermal layer to reach the interior of the dermis, with a deviation of 2.6% from the theoretical prediction.

[0152] At preset time points (0.25h, 0.5h, 1h, 2h, 4h, 6h, 8h, 12h, 18h, 24h, 36h, 48h, 60h, 72h), all samples were removed from the recipient chamber using a pipette. Immediately replenish the recipient fluid with an equal volume of preheated solution. Fresh recipient fluid. Three parallel diffusion cells were set up at each time point.

[0153] Finasteride concentration determination: HPLC-UV method (Agilent 1260 Infinity II), chromatographic conditions: ZORBAX Eclipse Plus C18 column (…). , The mobile phase was acetonitrile-water (45:55, v / v), with isocratic elution and a flow rate of [missing information]. Column temperature Detection wavelength Injection volume The linear range of the standard curve is... The regression equation is , Detection limit .

[0154] Exosome concentration determination: When preparing the exosome suspension, PKH26 red fluorescent dye (Sigma-Aldrich, excitation / emission wavelength) was first used. Labeling of exosomal lipid membranes. Labeling method: PKH26 was labeled with... The final concentration was added to the exosome suspension and incubated at 37°C for 20 minutes, followed by washing twice after ultracentrifugation at 100,000 × g for 70 minutes to remove free dye. The labeling efficiency was determined using a fluorescence spectrophotometer (Hitachi F-7100). Receptor fluid samples at each time point were first centrifuged at 10,000×g for 10 minutes to remove protein precipitates, and then the exosomes were concentrated by ultracentrifugation at 100,000×g for 60 minutes. Fluorescence intensity was measured after resuspending in PBS. The linear range of the standard curve was [value missing]. , .

[0155] Based on the drug concentration and sample volume at each time point, the cumulative release amount and cumulative release percentage are calculated using the following formula:

[0156]

[0157]

[0158] in, For the first The cumulative release amount at each point in time. For the first Drug concentration in the recipient fluid at each time point The total volume of the receptor chamber. For each sample volume (full replacement). This represents the total drug load in the microneedle.

[0159] Actual release data showed that finasteride was significantly released within 0.25 hours. The cumulative release over 4 hours is 12h is 24h is 48h is Reached in 72 hours Exosome release was significantly delayed compared to finasteride: no release was detected at 0.25 h, and at 2 h... 12h is 24h is 48h is 72h is The reason for the slow release rate and low cumulative release of exosomes is that the exosome particle size ( When the effective pore size is close to or exceeds that of the matrix in the early stages of degradation, it is limited by a significant particle size sieving effect, and some exosomes are inactivated during drying and release.

[0160] The total mass transfer resistance of the drug from microneedles to the recipient fluid in the Franz diffusion cell system Resistance released from microneedle matrix Diffusion resistance of different layers of skin and receptor fluid boundary layer resistance It consists of three parts connected in series: ;

[0161] in, Effective diffusion area. Boundary layer mass transfer coefficient. With stirring speed The relationship was determined through dissolution experiments of naphthazoline standards:

[0162]

[0163] Actual measurement , .when At that time, finasteride ,correspond much smaller and (All in) The magnitude indicates that the resistance of the receptor fluid boundary layer is negligible, and the system is in drug release / skin diffusion control mode.

[0164] The in vitro cumulative release curves of the obtained finasteride and exosomes ( (each time point was repeated three times), and nonlinear weighted least squares regression fitting was performed using the following four standard drug release kinetic models:

[0165] (1) Zero-level release model:

[0166] (2) First-level release model:

[0167] (3) Higuchi model:

[0168] (4) Korsmeyer-Peppas model (only considering the early stage of release) (Data fitting)

[0169] The fitting method uses weighted least squares, with weighting coefficients... ,in For the first The standard deviations of three repeated measures at each time point. The fitting results of each model are summarized in the table below:

[0170]

[0171] The AICc values ​​show that the Korsmeyer-Peppas model fits the early release data of both drugs best, but it should be noted that the KP model only... It is valid within the specified range, but cannot describe the subsequent release behavior.

[0172] Construct a two-drug differentiation correction model, such as Figure 4 As shown, Figure 4 (a) Shows a comparison of the fitting of zero-order, first-order, and Higuchi models to the cumulative release data of finasteride and exosomes over the entire time period. The three standard models show differences in the later stages of release ( Systematic biases were observed in all models: the zero-level and first-level models overestimated the late-stage release rate, while the Higuchi model underestimated the late-stage release amount. Figure 4 (b) demonstrates the fit of the Korsmeyer-Peppas standard model across the entire time frame, showing its performance in... The predicted values ​​deviated from the experimental data, with a maximum deviation of 8.5%.

[0173] To overcome the limitations of the standard model, a physics-driven correction term is introduced into the Korsmeyer-Peppas model framework:

[0174] For finasteride, being a small-molecule hydrophobic drug, the expansion of the matrix pore size during soluble matrix swelling leads to an accelerated release effect in the later stages. A logarithmic correction term is introduced to describe the enhancement of the release rate by matrix swelling:

[0175]

[0176] in, Let KP be the release rate constant. To release the index, The swelling enhancement amplitude factor, These are the swelling kinetic rate parameters. Nonlinear fitting results: , , , , AICc= . lie in The range indicates that the release mechanism of finasteride is anomalous transport, namely the synergistic coupling of Fick diffusion and matrix dissolution.

[0177] For exosomes, due to their nanoscale vesicle structure, the release process is simultaneously affected by the particle size sieving effect limited by the matrix mesh size and the inactivation decay of the vesicles themselves in the tissue fluid environment due to changes in osmotic pressure and protease degradation. Therefore, an exponential decay correction term is introduced:

[0178]

[0179] in, Let KP be the release rate constant. To release the index, This is a parameter representing the exosome decay rate. The decay kinetics exponent. Nonlinear fitting results: , , , , AICc= . This indicates that the release mechanism of exosomes is dominated by matrix dissolution (Case-II transport), meaning that the release rate is mainly controlled by the degradation / dissolution rate of the polymer matrix rather than pure diffusion.

[0180] The asymptotic 95% confidence intervals of the parameter estimates were obtained using the Fisher information matrix. Calculation of the inverse matrix:

[0181]

[0182] =

[0183] coefficient of variation of all fitted parameters All values ​​were less than 8.5%, indicating that the parameter identification had good statistical accuracy.

[0184] like Figure 4 (c) shows the standardized residuals for the four standard models and modified models. Time series analysis and lag autocorrelation function (ACF) analysis were performed. The residual sequence of the standard KP model showed significant positive autocorrelation (lag 1 ACF = 0.62, exceeding the 95% confidence band), indicating a systematic omission in the model structure. The residuals of the modified model exhibited an approximately white noise distribution (ACF of all lag orders were within the range of 0.62). Within the range), the Durbin-Watson statistics are respectively and (Close to the ideal value of 2.0), the p-values ​​of the Ljung-BoxQ test are all greater than 0.15, so the null hypothesis of residual independence is accepted.

[0185] Construct a three-dimensional geometric model of the scalp skin, such as Figure 2 As shown, based on scalp skin histological data and the measured pig ear skin histological parameters in step two, the following three-layer geometric structure was constructed sequentially along the depth direction (defined as the positive z-axis pointing inwards):

[0186] Stratum Corneum (SC): The outermost dense barrier structure, with a thickness of... The stratum corneum is composed of differentiated corneocytes embedded in a lipid matrix, forming a "brick-and-mortar" structure. In the geometric model, the stratum corneum is modeled as a horizontal plate of uniform thickness, with the horizontal modeling area being... , of Rectangular region. The material properties of the stratum corneum are set as follows: effective porosity. tortuosity factor lipid volume fraction Moisture content .

[0187] Viable Epidermis (VE): Located beneath the stratum corneum, with a thickness of... The active epidermal layer consists of multiple layers of living cells and contains a small amount of interstitial fluid. The material property is set as: effective porosity. tortuosity factor Cell density Integral number of liquids .

[0188] Dermis: Located beneath the active epidermis, with a thickness of [missing information]. The dermis consists of type I and type III collagen fiber networks, elastic fibers, fibroblasts, and a hyaluronic acid-rich extracellular matrix (ECM), containing a rich capillary network and lymphatic vessels. The material property is set as: effective porosity. tortuosity factor Collagen fiber volume fraction Fibroblast density Capillary surface area density .

[0189] Embedded modeling of hair follicle structures involves embedding cylindrical hair follicle structures into the dermis. Each hair follicle structure extends along the z-axis from the surface of the stratum corneum (…). Extending downwards into the dermis, the total length The hair follicle structure, from the inside out, consists of the following concentric cylindrical / hemispherical structures:

[0190] Hair shaft: Located at the geometric center of the hair follicle, with a diameter of... In the model, it is set as an impermeable solid region (the drug does not enter the inside of the hair shaft).

[0191] Inner Root Sheath (IRS): Encloses the hair shaft; wall thickness , outer diameter The IRS consists of Henle and Huxley layers, with a dense structure and high effective porosity. , tortuosity ;

[0192] Outer Root Sheath (ORS): Encloses the IRS, with a wall thickness of [missing information]. , outer diameter The ORS structure is similar to the skin, with an effective porosity of , tortuosity ;

[0193] Dermal papilla (DP): Located at the base of the hair follicle, it is modeled as a hemispherical structure with a radius of... The center coordinates are located at The dermal papilla contains abundant blood vessels and dermal papilla cells (density). The hair follicle is the core area for hair follicle growth signals.

[0194] Seven hair follicle structures are arranged in a hexagonal close-packed pattern within the modeling area, with the center-to-center spacing of the hair follicles being... The corresponding density is approximately .

[0195] The microneedle body is modeled as a truncated cone geometry. Geometric parameters of the needle body include: total height. needle base radius needle tip radius The angle between the generatrix on the side of the needle body and the axis. Microneedle array Square array arrangement, center-to-center distance between adjacent needles .

[0196] The spatial positioning strategy of microneedle arrays is as follows: The projection point of a hair follicle center onto the skin surface... Using the origin as a reference, The center of the microneedle array is offset from the reference origin. The position is such that the horizontal distance between the needle tip closest to the hair follicle in the array and the outer wall of the hair follicle ORS is [missing information]. (i.e., the edge of the needle body is close to the outer wall of the hair follicle) The overlapping area (where the microneedle is close to the sidewall of the hair follicle during insertion) is the horizontal distance from the tip of the needle closest to the hair follicle. .

[0197] Microneedle insertion depth Needle tip coordinates Ensure the needle tip is positioned in the middle of the dermis. ).

[0198] Initial drug loading concentration of finasteride in the microneedle body (Corresponding to drug loading of 9 doses per tablet) Initial drug loading concentration of exosomes .

[0199] The area around the hair follicle structure is defined, starting from the outer wall of the hair follicle ORS and extending outwards. Circular target area As the target area for drug action (including ORS, hair papilla and surrounding dermal tissue).

[0200] A kinetic model of microneedle matrix degradation was developed, showing that after soluble microneedles penetrate the dermis, they swell and gradually degrade in a tissue fluid environment with high water content. The matrix degradation process follows a mixed mechanism of surface dissolution and bulk degradation. The degree of matrix degradation was defined. for The ratio of the mass of degraded matrix to the initial matrix mass at any given time:

[0201]

[0202] in, The overall degradation rate constant is Degradation kinetic index ( Indicates self-accelerated degradation. Indicates slowed degradation. (This is primary degradation).

[0203] Degradation rate constant Taking into account the coupled effects of temperature and moisture content, the Arrhenius-humidity correction relationship is satisfied:

[0204]

[0205] in, For reference conditions ( , The baseline degradation rate under ( ) This refers to the water content of the dermal tissue. A moisture content sensitive index, For degradation activation energy, Let be the ideal gas constant. ( Substituting the parameters, we can calculate... The in vitro mass loss experiment (microneedles were immersed in pH 7.4 PBS) was conducted. (Timed weighing) Calibration .

[0206] Instantaneous porosity of microneedle matrix and swelling ratio Dynamic changes during the degradation process:

[0207]

[0208]

[0209] in, The initial porosity in the dry state. The limiting porosity after complete degradation. The maximum swelling ratio, This is a swelling decay parameter (describing the swelling and shrinkage caused by the collapse of the matrix structure in the later stages).

[0210] like Figure 5 As shown in (a), Internally, the degradation degree rapidly increased from 0 to 0.42, the porosity increased from 0.08 to 0.44, and the swelling ratio first increased and then decreased. (about It reaches its peak at that time. It then slowly retracted; The degradation rate tends to slow down. The degradation rate reached 0.85, and the porosity was 0.82. The degradation rate was 0.95, and the matrix was almost completely dissolved.

[0211] The drug release rate from the microneedle matrix is ​​determined by both matrix degradation-driven drug dissociation and intramatrix diffusion. (Definition of drug) In spatial location and time Release source item for:

[0212]

[0213] in, This is the initial drug loading concentration. This is a spatial indicator function for the microneedle body (value 1 within the geometric region occupied by the needle body, and 0 in other regions). This is the time derivative of the degradation rate of the Weibull-type matrix. The release rate normalization mapping function is constructed based on the optimal sustained-release kinetic model determined in step three:

[0214] For finasteride: ;

[0215] For exosomes: ;

[0216] The transport of finasteride (molecular weight 272.38 Da, logP = 3.03) in various layers of scalp skin is controlled by a triple mechanism of diffusion, interstitial fluid convection, and tissue metabolism. In the first... Layered organization ( In ), the concentration field of finasteride Satisfy the following partial differential equations: ;

[0217] Effective diffusion coefficient Calculations based on the modified Mackie-Meares pore model: ;in, The diffusion coefficient of finasteride in free aqueous solution (based on the Stokes-Einstein equation). calculate, ), This represents the ratio of molecular size to pore size. The parameters and calculation results for each layer are shown in the table below:

[0218]

[0219] Convection velocity The interstitial fluid in the dermis slowly seeps under the influence of the tissue hydraulic gradient, with a velocity on the order of magnitude of... The direction is along the depth direction of the dermis (z-direction), set The seepage rate in the stratum corneum and active epidermal layer is approximately zero.

[0220] Metabolic items Finasteride is metabolized by the CYP3A4 enzyme in skin tissue, and its metabolic kinetics are described using the Michaelis-Menten model. Among them, the dermis , Active epidermal layer , The stratum corneum has no metabolic activity. ).when At this time, metabolism approximates first-order kinetics, with a rate constant. .

[0221] Exosomes (particle size) The surface carries a negative charge and has a zeta potential. The transport of exosomes in the dermis is subject to multiple limitations, including steric hindrance from the collagen fiber network, interstitial fluid drag, cellular uptake, and enzymatic degradation. (Exosome concentration field) Satisfy the following partial differential equations: Spatiotemporal dependent diffusion coefficient tensor The study comprehensively considered the four effects of collagen fiber anisotropy, matrix viscosity hindrance, particle size sieving, and steric hindrance: ;

[0222] The baseline diffusion coefficient of exosomes in isotropic free aqueous solutions (Stokes-Einstein equation). );

[0223] This is the orientation tensor of dermal collagen fibers (a second-order symmetric positive definite tensor), whose principal values ​​are determined by the Hermans orientation factor of dermal collagen fibers. Decide: , Around the hair follicle Within this range, collagen fibers are arranged radially along the hair follicle. , , In areas far from hair follicles, It is approximately isotropic;

[0224] This refers to the viscosity-restricting factor caused by the concentration of hyaluronic acid. , The HA released from the degradation of the microneedle matrix creates localized high-concentration areas, significantly increasing the viscous resistance to exosome diffusion;

[0225] Particle size sieving factor, , The effective pore size of the local ECM in the dermis ( (This depends on collagen density). When , hour, This indicates that the effective diffusion coefficient of exosomes in the dermis is only about 3% of that of free water;

[0226] The steric hindrance Boltzmann factor, This represents the repulsive potential energy between exosomes and collagen fibers. Based on the Ogston model... ,in Where is the radius of the collagen fiber. When... hour, , .

[0227] Comprehensive calculations show that the effective diffusion coefficient of exosomes in the dermis is... It is 5 orders of magnitude lower than finasteride, which explains the experimental phenomenon that exosome release and diffusion are significantly slower than finasteride.

[0228] Cellular uptake items: The rate constant for the uptake of exosomes by fibroblasts and dermal papilla cells via clathrin-mediated endocytosis (determined by in vitro co-culture experiments). This refers to local cell density in the dermal papilla region. Ordinary dermal area .

[0229] Enzymatic degradation items: This is the rate constant (half-life) of exosome inactivation in the tissue environment due to membrane protein degradation and lipid peroxidation. ).

[0230] Initial conditions ( time):

[0231]

[0232]

[0233] in The spatial domain occupied by the microneedle.

[0234] The boundary conditions are set as follows:

[0235] outer surface of skin ( (Plane, excluding the microneedle insertion area): zero-flux Neumann boundary,

[0236]

[0237] Deep dermal capillary boundaries Convection clears Robin boundaries.

[0238]

[0239] Among them, capillary clearance rate Calculations based on the Kedem-Katchalsky membrane transport model:

[0240]

[0241] For finasteride: (Small molecules can pass through freely) , ,calculate .

[0242] For exosomes: (Nanoparticles cannot pass through the capillary wall.) , .

[0243] hair follicle target area boundary : Permeation-accumulation coupling boundary, ;in , (Finasteride has a partition coefficient >1 in ORS due to lipid enrichment). , .

[0244] Adjacent tissue layer interfaces (SC / VE interface and VE / Der interface): concentration continuity and flux conservation. ;

[0245] Interlayer allocation coefficient: (Finasteride is enriched in stratum corneum lipids). ; (Exosomes cannot penetrate the stratum corneum) .

[0246] Modeling region side boundary Periodic symmetric boundary conditions (simulating the repeatability of a single microneedle-follicle unit in an array). ;

[0247] like Figure 3 As shown, the finite element solution system includes a preprocessing module (geometry import, material property allocation and mesh generation), a physics configuration module (PDE weak form establishment and boundary condition application), an adaptive time-stepping solver module, and a postprocessing module (field variable extraction, integral calculation and visualization output).

[0248] The 3D geometric model is spatially discretized using a tetrahedral unstructured mesh. The mesh size distribution employs an adaptive control strategy based on prior estimation of the concentration gradient. Local mesh size... The calculation formula is:

[0249]

[0250] in, The maximum allowable mesh size globally. Let the order of the shape function be . Here is the regularization constant. The distance from the unit to the nearest microneedle or hair follicle. For reference distance, This is the distance decay exponent.

[0251] Following the above strategy, the mesh size varies from the microneedle surface to the outer wall of the hair follicle. Gradually transition to the periphery of the modeling area The final generated mesh contains 1.42 million tetrahedral elements and 286,000 nodes. Mesh independence verification shows that when the mesh is refined to 2.35 million elements, the maximum deviation in the average concentration of the hair follicle target area is 0.72%. Changes It was confirmed that the 1.42 million-cell mesh met the accuracy requirements.

[0252] Implicit backward Euler scheme is used to discretize time. At time step... Within this framework, after discretizing the PDE using the Galerkin weak form, a system of nonlinear algebraic equations is obtained:

[0253]

[0254] in For the quality matrix, The stiffness matrix is ​​dependent on the concentration field (and is nonlinear due to the concentration dependence of the metabolic term and diffusion coefficient). The load vector contributed by the source term and boundary conditions. For the first The node concentration vector at each time step.

[0255] The nonlinear equations are solved using the Newton-Raphson iterative method. In the... In the next Newton iteration, solve the linearized incremental equation:

[0256]

[0257] in For Jacobian matrices, Let be the residual vector. When Convergence is determined at that time.

[0258] The linear subsystem employs a preprocessed GMRES iterative solver (with restart parameters). The preconditioner employs the Algebraic Multigrid (AMG) method, which has algebraic convergence tolerance. .

[0259] Time step An adaptive strategy based on local truncation error estimation is adopted. Two sets of solutions with first-order and second-order accuracy are calculated simultaneously, and the difference between them is used as the error estimate. The step size adjustment formula is:

[0260]

[0261] in , (Euler is first order backwards). , Initial step size (After the initial thrust, the rapid release phase) As the release becomes smoother, the stride length gradually increases, reaching a maximum stride length. Total simulation duration There are approximately 580 time steps in total.

[0262] The total computation time for a single simulation on a high-performance workstation (Intel Xeon Gold 6348×2, 56 cores, 256GB DDR4 RAM, NVMe SSD) is approximately 4.2 hours, with a peak memory usage of approximately 68GB.

[0263] like Figure 5 (b) and Figure 5 As shown in (c), they respectively demonstrate , , , Four moments along the depth of the dermis Two-dimensional concentration distribution thermograms of finasteride and exosomes at cross-sections (color mapping ranges are respectively) and ).

[0264] Finasteride concentration field ( Figure 5 (b) Upon release of finasteride from the nine microneedles, an independent circular high-concentration region (peak concentration) is formed around the needles. (No concentration field superposition has occurred between the individual needles; the concentration at the edge of the hair follicle target area is approximately...) ; At this time, the concentration fields of adjacent needles begin to merge, forming a continuous elliptical high-concentration region, with an average concentration of [missing information] within the hair follicle target area. (Higher than MEC=) ); At that time, the concentration field further expanded to Range, average concentration in the hair follicle target area Furthermore, due to the high logP of finasteride, concentration "hot spots" appear in the lipid-rich region of ORS. At that time, the overall concentration level decreased, and the average concentration in the hair follicle target area decreased. It is still higher than MEC.

[0265] Exosome concentration field ( Figure 5 (c) At that time, exosomes are released only from a thin layer on the surface of the needle, and the diffusion range is extremely limited. Each needle body forms an independent, small-scale high-concentration spot (peak value). The concentration within the hair follicle target area is close to zero. As matrix degradation accelerates and pore size increases, the exosome diffusion range expands to... The needle closest to the hair follicle (horizontal distance) The released exosomes begin to reach the outer wall of the hair follicle ORS, with a concentration of approximately [missing information] at the target area edge. ; At that time, the exosome diffusion range reached The concentration fields of multiple needles are superimposed around the hair follicle, resulting in an average concentration in the target area. (Higher than MEC=) ); At that time, due to continuous uptake and inactivation degradation, the average concentration in the target area decreased to It is still slightly higher than MEC.

[0266] Finasteride in the hair follicle target area Duration of maintaining effective concentration Calculate using the following formula:

[0267]

[0268] in The target volume average concentration. This represents the lowest effective concentration of finasteride to inhibit 5α-reductase type II activity (based on in vitro enzyme activity inhibition assays). MEC (1.2 times that of the baseline prescription parameters). Simulation calculation .

[0269] Duration of effective exosome concentration Similarly, calculate:

[0270]

[0271] in This represents the minimum effective concentration of exosomes to promote dermal papilla cell proliferation (based on CCK-8 assay). Benchmark formulation. .

[0272] Exposure of finasteride into systemic circulation via absorption through dermal capillaries Calculate using the following formula:

[0273]

[0274] in This is the apparent volume of distribution of finasteride. For liver clearance rate. Simulation calculation under baseline prescription. .

[0275] Steady-state 24-hour systemic exposure to finasteride (1 mg / day) (Reported values ​​in the literature). Systemic exposure ratio of microneedle local drug delivery. The levels are far lower than those of oral administration, indicating that local microneedle administration has a significant advantage in systemic safety.

[0276] like Figure 6 As shown, the optimization objective is to maximize the sum of the weighted effective concentration maintenance times:

[0277]

[0278] in , These are weighting coefficients (emphasizing the DHT-inhibiting effect of finasteride). Optimization variables. The feasible region for each variable is: , , , , .

[0279] The constraint is a safety constraint on whole-body exposure:

[0280]

[0281] in (That is, the systemic exposure of microneedle drug delivery does not exceed 5% of that of oral administration).

[0282] The solution is obtained using the Multi-Objective Particle Swarm Optimization (MOPSO) algorithm. Algorithm parameters: particle swarm size. External archive capacity Maximum number of iterations Inertial weight from linearly decreasing to Learning factor Mesh adaptive mutation probability The particle position update formula is:

[0283]

[0284]

[0285] in Select from external archives sorted by crowding distance.

[0286] like Figure 6 As shown, after 300 iterations of convergence, a total of 42 non-dominated solutions were obtained on the Pareto front. The front curve exhibits a typical concave characteristic: at low... area( Increasing the drug loading can significantly improve (Slope is large); at high area( ), The growth rate tends to saturate (the slope approaches zero), indicating that there is diminishing marginal utility.

[0287] Considering both efficacy and safety, the knee point is selected as the recommended optimal prescription based on the Pareto frontier: , HA:PVP:CMC = 5.5:11:3.5 (total concentration 20%). Simulation predictions for this formulation: , , , .

[0288] To verify the reliability of the simulation predictions, a microneedle array was prepared according to the optimal prescription parameters and an in vitro transdermal experiment was conducted. The average concentration maintenance time of finasteride in the hair follicle target area (extracted through skin sections and fluorescence imaging) was measured within 72 hours. (Prediction 48.2h, deviation 5.4%), exosomes (Predicted duration 33.5 hours, deviation 8.0%), total exposure ratio (Prediction 2.8%, deviation 10.7%). The deviations between the simulation predictions and experimental verification values ​​are all within 10%, verifying the accuracy and reliability of the simulation model.

[0289] In this embodiment, the soluble microneedles corresponding to the simulation object are identified, and in vitro drug release experiments are conducted to obtain release curves. The optimal kinetic model is selected by fitting multiple drug release models. A three-dimensional geometric model of the scalp containing the stratum corneum, active epidermis, dermis and hair follicles is constructed. A diffusion-release coupled partial differential equation system is built, and numerical solutions are obtained using the finite element method. The effective concentration maintenance time around the hair follicle and the predicted index of systemic exposure are output, realizing the digital simulation characterization of the drug delivery effect of microneedles.

[0290] Example 2: This example describes an exosome microneedle transdermal drug delivery simulation system based on a sustained-release kinetic model, such as... Figure 11 As shown, it includes:

[0291] The microneedle construction module constructs a soluble microneedle array loaded with exosomes and finasteride. The soluble microneedle array includes a substrate and multiple microneedles arranged in an array on the substrate. The microneedles are composed of a soluble polymer matrix material, engineered exosomes loaded with TGF-β, and finasteride.

[0292] The drug release data acquisition module places the soluble microneedle array in a transdermal diffusion experimental device, collects the release liquid at different time points, and measures the cumulative release amount of exosomes and finasteride to obtain the in vitro cumulative release curve.

[0293] The kinetic modeling module uses multiple drug release kinetic models to fit the obtained in vitro cumulative release curves for exosome and finasteride release behavior, and determines the optimal sustained-release kinetic model and its corresponding model parameters for exosome and finasteride based on the goodness of fit.

[0294] The geometric modeling module is based on the histological structure of the scalp skin, constructing a multi-layer three-dimensional geometric model including the stratum corneum, active epidermis, dermis and hair follicle structure, and setting the initial insertion position, insertion depth and initial drug concentration of the microneedle body as initial conditions.

[0295] Based on the determined optimal sustained-release kinetic model and combined with Fick's diffusion law, the equation construction module establishes a set of diffusion-release coupled partial differential equations for exosomes and finasteride in each layer of the multi-layer three-dimensional geometric model.

[0296] The numerical solution module meshes the constructed three-dimensional geometric model, sets boundary conditions, and uses the finite element method to numerically solve the established partial differential equations to obtain the spatiotemporal concentration distribution results of exosomes and finasteride in various layers of scalp skin and tissues around hair follicles.

[0297] Based on the numerical solution, the results analysis module outputs the effective concentration maintenance time of exosomes and finasteride around the hair follicles in the dermis, as well as the predicted value of the systemic exposure of the drug through transdermal absorption into the capillary network.

[0298] The microneedle construction module corresponds to the initial simulation object definition stage in the simulation workflow. This stage involves internal software parameter configuration and does not involve the physical fabrication of actual microneedles. The unit inputs the structural and compositional information of the object to be simulated: including the substrate, the arrayed microneedles, and the material composition of the needles consisting of a soluble polymer matrix, TGF-β-loaded engineered exosomes, and finasteride. This unit imports the structural composition information of the physical sample into the simulation pipeline, corresponding to the physical processes of engineered exosome preparation and microneedle array formation in the specific implementation method, thus defining the computational object for all subsequent simulation calculations.

[0299] The drug release data acquisition module receives the raw detection data from the in vitro transdermal experiments, corresponding to the in vitro drug release experiment stage in the Franz diffusion cell in the protocol. It receives the detection results of the release solution at different sampling time points, and after processing, obtains the in vitro cumulative release curves of exosomes and finasteride. The raw experimental data comes from the actual measurement results of HPLC and nanoparticle tracking analyzer. The processed release curves are directly transmitted downstream as the data source for kinetic fitting calculations, ensuring that the simulation parameters originate from real experiments and are not purely arbitrary assumptions.

[0300] The kinetic modeling module receives the in vitro cumulative release curve from the upstream source and performs fitting calculations on four types of drug release models: zero-order release, first-order release, Higuchi, and Korsmeyer-Peppas. The fitting effect is evaluated based on the coefficient of determination and the Akaike information criterion, selecting the optimal sustained-release model for each exosome and finasteride, and outputting key model parameters such as the release rate constant and release index. Corresponding to the kinetic model fitting and parameter identification section in the specific implementation, this module completes the conversion of experimental data into simulation model parameters, providing experimentally calibrated parameters for subsequent partial differential equations.

[0301] The geometric modeling module constructs a multi-layered three-dimensional geometric computational domain based on measured parameters of scalp histology. This domain fully reproduces the stratum corneum, active epidermis, dermis, and hair follicle structure. Simultaneously, the microneedle insertion location, insertion depth, and initial drug concentration are recorded as initial conditions for the entire finite element calculation. This unit reproduces the thickness of each skin layer, the geometric dimensions of hair follicles, and the relative spatial positions between microneedles and hair follicles, constructing a three-dimensional simulation geometric carrier that closely resembles the physiological characteristics of the human scalp.

[0302] The equation construction module reads the optimal model parameters output by the kinetic modeling module and assembles a complete set of diffusion-release coupled partial differential equations in conjunction with Fick's diffusion law. The equation set distinguishes the distinct mass transfer behaviors of small molecule finasteride and nanovesicle exosomes, incorporating matrix dissolution effects, exosome particle size-restricted diffusion, and tissue metabolic consumption effects. It also includes relevant physical parameters such as the diffusion coefficient, partition coefficient, and matrix degradation rate of each skin layer, generating a complete mathematical model that can be used for solving.

[0303] After obtaining the 3D geometric model and coupled equations, the numerical solution module performs model mesh generation. Mesh refinement is applied to areas with drastic concentration changes, such as the areas around microneedles and hair follicles, while a coarse mesh is used in the outer regions to balance computational overhead. A complete set of physical boundary conditions is configured: no flux boundary on the outer skin surface, capillary convection clearance boundary at the dermal base, and interlayer concentration and flux conservation boundary. The finite element solver is then invoked to iteratively solve the nonlinear equations, ultimately outputting spatiotemporal concentration field data for the entire scalp region over time. This corresponds to the entire numerical calculation process in the example, including mesh generation, time discretization, and Newton-Raphson iterative solution.

[0304] The results analysis module reads the complete spatiotemporal concentration field output by the numerical solution module, extracts the concentration time series of each grid node within a specified range around the hair follicle, and calculates the effective concentration maintenance duration of the two substances. Then, it performs integral calculations on the capillary boundary surface to obtain the predicted results of systemic drug exposure. The unit only outputs objective quantitative numerical indicators and does not perform subjective qualitative evaluations. The output results can determine whether the drug in the hair follicle target area can reach the effective therapeutic threshold, and can also predict the amount of drug entering the bloodstream after local administration. This is used to measure the ability of local administration to avoid the systemic side effects of oral finasteride, aligning with the overall research and development objectives of the invention.

[0305] This application also provides a computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor, it implements the steps of a simulation method for transdermal drug delivery of exosomes via microneedles based on a sustained-release kinetic model.

[0306] The storage medium is the physical carrier that accompanies this simulation solution, and the computer program is stored inside the medium. When the program is run by the processor, the entire simulation workflow can be completed: configuring the simulation object, processing in vitro drug release data, fitting sustained-release kinetics, 3D geometric modeling of the scalp, constructing a diffusion-release coupling equation set, performing finite element analysis, and outputting the effective concentration maintenance time around the hair follicle and the predicted index of systemic exposure. It can also perform multi-objective prescription optimization calculations.

[0307] This embodiment employs a modular simulation system to implement the aforementioned simulation method in software, with each functional module working collaboratively to complete the entire process. The complex simulation logic is encapsulated into executable functional units, lowering the operational threshold for simulation calculations, facilitating repeated adjustments to microneedle target parameters for multiple simulation comparisons, improving the efficiency of simulation work, and enabling technicians to quickly assess microneedle formulation plans.

[0308] The above are merely preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. For those skilled in the art, the present invention can have various modifications and variations. Any changes, modifications, substitutions, integrations, and parameter changes made to these embodiments within the spirit and principles of the present invention, without departing from the principles and spirit of the present invention, through conventional substitutions or to achieve the same function, fall within the scope of protection of the present invention.

Claims

1. An exosome microneedle transdermal drug release simulation method based on a sustained-release kinetic model, characterized in that, Includes the following steps: A soluble microneedle array loaded with exosomes and finasteride was constructed. The soluble microneedle array includes a substrate and multiple microneedles arranged in an array on the substrate. The microneedles are composed of a soluble polymer matrix material, engineered exosomes loaded with TGF-β, and finasteride. To obtain in vitro drug release experimental data, the soluble microneedle array was placed in a transdermal diffusion experimental device, and the release liquid was collected at different time points. The cumulative release amount of exosomes and finasteride was measured to obtain the in vitro cumulative release curve. A sustained-release kinetic model was established, and multiple drug release kinetic models were used to fit the obtained in vitro cumulative release curves for the release behavior of exosomes and finasteride. The drug release kinetic models included at least the zero-order release model, the first-order release model, the Higuchi model, and the Korsmeyer-Peppas model. The optimal sustained-release kinetic model for exosomes and finasteride and its corresponding model parameters were determined based on the goodness of fit. The model parameters included the release rate constant and the release exponent. A three-dimensional geometric model of the scalp skin is constructed. Based on the histological structure of the scalp skin, a multi-layer three-dimensional geometric model including the stratum corneum, active epidermis, dermis and hair follicle structure is constructed. The initial insertion position, insertion depth and initial drug concentration of the microneedle body are set as initial conditions in the three-dimensional geometric model. A set of partial differential equations for transdermal drug release was established. Based on the determined optimal sustained-release kinetic model and combined with Fick's diffusion law, a set of diffusion-release coupled partial differential equations for exosomes and finasteride in each layer of the multilayer three-dimensional geometric model was established. The set of partial differential equations includes the diffusion coefficient, partition coefficient and degradation rate parameters of the microneedle matrix of each tissue layer. The constructed three-dimensional geometric model was meshed, boundary conditions were set, and the finite element method was used to numerically solve the established partial differential equations to obtain the spatiotemporal concentration distribution of exosomes and finasteride in different layers of scalp skin and tissues around hair follicles. The duration of effective concentration of exosomes and finasteride around hair follicles in the dermis, and the predicted systemic exposure of the drug through transdermal absorption into the capillary network.

2. The exosome microneedle transdermal drug release simulation method based on the slow-release kinetic model according to claim 1, characterized in that, The construction of the soluble microneedle array loaded with exosomes and finasteride specifically includes: Mesenchymal stem cells were overexpressed with TGF-β by lentiviral transfection, and the cell culture supernatant was collected and engineered exosomes loaded with TGF-β were extracted by ultracentrifugation. Finasteride was dissolved in an organic solvent to prepare a finasteride solution, and the engineered exosomes loaded with TGF-β were resuspended in a buffer solution to prepare an exosome suspension. Hyaluronic acid, polyvinylpyrrolidone and sodium carboxymethyl cellulose were dissolved in deionized water to prepare a polymer matrix solution. The finasteride solution and the exosome suspension were added sequentially to the polymer matrix solution, stirred and mixed, and then injected into the PDMS microneedle mold. The mixture was centrifuged to fully fill the micropores of the mold. After scraping off the excess solution, the mixture was dried and molded, and then demolded to obtain the soluble microneedle array.

3. The exosome microneedle transdermal drug release simulation method based on the slow-release kinetic model according to claim 2, characterized in that, The establishment of a sustained-release kinetic model involves fitting the obtained in vitro cumulative release curves using multiple drug release kinetic models, specifically targeting the release behavior of finasteride from exosomes. These models include: Using the zero-level release model according to the formula The in vitro cumulative release curve was fitted, wherein... for Cumulative release percentage at any given time This is the initial release amount. The zero-order release rate constant; The in vitro cumulative release profile was fitted to the equation wherein k is the first order release rate constant; and is the first order release rate constant. The Higuchi model was used to fit the in vitro release data according to the equation The in vitro cumulative release curves were fitted, where is the Higuchi release rate constant. Using the Korsmeyer-Peppas model according to the formula The in vitro cumulative release curve was fitted, wherein... The Korsmeyer-Peppas release rate constant. To release the index, when The release mechanism is dominated by Fick diffusion, when The time is a synergistic mechanism of anomalous diffusion and matrix dissolution, when At that time, matrix dissolution was the dominant process; Calculate the coefficient of determination for each drug release kinetic model. Select the AIC value of the Akaike Information Criterion. The model with the largest and smallest AIC value is taken as the optimal sustained-release kinetic model for the corresponding drug.

4. The exosome microneedle transdermal drug delivery simulation method based on a sustained-release kinetic model according to claim 3, characterized in that, The construction of the three-dimensional geometric model of the scalp skin, based on the histological structure of the scalp skin, involves building a multi-layered three-dimensional geometric model including the stratum corneum, active epidermis, dermis, and hair follicle structure, specifically including: The stratum corneum is set as a flat layer with a thickness of 10-20 μm, the active epidermal layer is set as a flat layer with a thickness of 50-100 μm, and the dermis is set as a flat layer with a thickness of 1000-2000 μm. A cylindrical hair follicle structure is constructed in the dermis, the diameter of the hair follicle structure is set to 100-300μm, and the depth extends from the surface of the stratum corneum to the interior of the dermis 500-1500μm; The microneedle body is modeled as a cone array, with the bottom diameter of the cones being 200-400 μm, the height being 500-800 μm, and the center-to-center distance between adjacent cones being 500-1000 μm; The tip coordinates of the microneedle are set to the target area next to the hair follicle structure, so that the horizontal distance between the tip of the microneedle and the hair follicle structure is 50-200μm.

5. The exosome microneedle transdermal drug delivery simulation method based on a sustained-release kinetic model according to claim 4, characterized in that, The partial differential equations for transdermal drug release are established based on a determined optimal sustained-release kinetic model and Fick's diffusion law. Specifically, a diffusion-release coupled partial differential equation system for exosomes and finasteride is established in each layer of the multilayer three-dimensional geometric model. A diffusion equation based on matrix dissolution coupling was established to investigate the release process of finasteride in the microneedle matrix: ， in, This refers to the local concentration of finasteride. The position- and time-dependent effective diffusion coefficient tensor of finasteride. For the time-varying degradation rate function of the microneedle matrix, The source term function for the dissociation of finasteride from the matrix-bound state to the free state; To investigate the release process of exosomes from the microneedle matrix, a confined diffusion equation based on particle size sieving effect was established: ， in, This refers to the local concentration of exosomes. For depends on the instantaneous porosity of the matrix Equivalent hydrodynamic radius of exosomes The effective diffusion coefficient, The inactivation decay constant of exosomes in the tissue environment; The effective diffusion coefficient Calculate as follows: ， in, The diffusion coefficient of exosomes in free aqueous solution is given. The porosity tortuosity correction index. The particle size attenuation coefficient is the sieve size reduction factor. As a matrix in Average mesh size at any given time; A stratified diffusion equation was established to investigate the diffusion process of finasteride and exosomes in different layers of scalp skin: ， Among them, superscript Indicates the skin tissue layer. subscript Indicates the type of drug. , For drugs At the organizational level The diffusion coefficient in For drugs At the organizational level The metabolic clearance rate constant in the figure.

6. The exosome microneedle transdermal drug delivery simulation method based on a sustained-release kinetic model according to claim 5, characterized in that, The process of meshing the constructed three-dimensional geometric model, setting boundary conditions, and numerically solving the established partial differential equations using the finite element method specifically includes: An adaptively refined grid is used in the microneedle area and the area surrounding the hair follicle structure, while a coarser grid is used in the peripheral area of ​​the dermis away from the microneedles and hair follicles. The total number of grid cells is controlled between 500,000 and 2 million. A flux-free boundary condition is set on the outer surface of the stratum corneum, as shown in the formula. ,in The boundary normal vector; A convection clearance boundary condition is set at the junction of the dermis and the capillary network, as shown in the formula. ,in The capillary wall permeability coefficient, The initial setting for the drug concentration in the blood is zero. At the interface between adjacent tissue layers, concentration continuity and flux conservation conditions are set, and the formula is as follows: and ,in This is the distribution coefficient of the drug between two adjacent tissue layers; The implicit backward Euler time discretization scheme was adopted, with the time step set to adaptive variable step size. The initial time step size was 1 minute, the maximum time step size was 30 minutes, and the total simulation time was set to 72 hours.

7. The exosome microneedle transdermal drug delivery simulation method based on a sustained-release kinetic model according to claim 6, characterized in that, The duration of effective concentration maintenance of finasteride in the dermal hair follicles by the exported exosomes, and the predicted systemic exposure of the drug through transdermal absorption into the capillary network, specifically include: Extract the finasteride concentration time series data of each grid node within a 50-200μm range around the hair follicle structure, and calculate the duration for which the finasteride concentration remains above a preset effective concentration threshold, as the effective concentration maintenance time of finasteride. ; Time-series data of exosome concentration were extracted from each grid node within a 50-200 μm range around the hair follicle structure. The duration during which the exosome concentration remained above a preset effective concentration threshold was calculated as the effective concentration maintenance time of exosomes. ; At the capillary boundary, the predicted systemic exposure to finasteride is calculated as follows: ， in, This refers to the total body exposure. Total simulation duration, For the curved surface of capillary boundaries, The concentration of finasteride at the dermal capillary interface; The Compared with the reference value for systemic exposure to oral finasteride, when When the systemic safety of local microneedle administration is determined to be satisfactory when the proportion is below the preset threshold of the systemic exposure reference value of the oral finasteride.

8. The exosome microneedle transdermal drug delivery simulation method based on a sustained-release kinetic model according to claim 1, characterized in that, It also includes a parameter optimization step for the sustained-release kinetic model: The optimization objective is to maximize the sum of the effective concentration maintenance time of the exosomes and finasteride around the hair follicles. The optimization variables are the loading concentration of finasteride in the microneedle, the loading concentration of the exosomes, and the ratio of the polymer matrix material. The systemic exposure of finasteride does not exceed a preset safety threshold as a constraint. A multi-objective optimization model is constructed. The multi-objective optimization model is iteratively solved using a particle swarm optimization algorithm or a genetic algorithm to obtain the optimal combination of microneedle prescription parameters that achieves synergistic synergy between hair follicle repair and protection.

9. A simulation system for transdermal drug delivery using exosomes and microneedles based on a sustained-release kinetic model, applicable to the simulation method for transdermal drug delivery using exosomes and microneedles based on a sustained-release kinetic model as described in any one of claims 1-8, characterized in that, include: The microneedle construction module constructs a soluble microneedle array loaded with exosomes and finasteride. The soluble microneedle array includes a substrate and multiple microneedles arranged in an array on the substrate. The microneedles are composed of a soluble polymer matrix material, engineered exosomes loaded with TGF-β, and finasteride. The drug release data acquisition module places the soluble microneedle array in a transdermal diffusion experimental device, collects the release liquid at different time points, and measures the cumulative release amount of exosomes and finasteride to obtain the in vitro cumulative release curve. The kinetic modeling module uses multiple drug release kinetic models to fit the obtained in vitro cumulative release curves for exosome and finasteride release behavior, and determines the optimal sustained-release kinetic model and its corresponding model parameters for exosome and finasteride based on the goodness of fit. The geometric modeling module is based on the histological structure of the scalp skin, constructing a multi-layer three-dimensional geometric model including the stratum corneum, active epidermis, dermis and hair follicle structure, and setting the initial insertion position, insertion depth and initial drug concentration of the microneedle body as initial conditions. Based on the determined optimal sustained-release kinetic model and combined with Fick's diffusion law, the equation construction module establishes a set of diffusion-release coupled partial differential equations for exosomes and finasteride in each layer of the multi-layer three-dimensional geometric model. The numerical solution module meshes the constructed three-dimensional geometric model, sets boundary conditions, and uses the finite element method to numerically solve the established partial differential equations to obtain the spatiotemporal concentration distribution results of exosomes and finasteride in various layers of scalp skin and tissues around hair follicles. Based on the numerical solution, the results analysis module outputs the effective concentration maintenance time of exosomes and finasteride around the hair follicles in the dermis, as well as the predicted value of the systemic exposure of the drug through transdermal absorption into the capillary network.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the exosome microneedle transdermal drug delivery simulation method based on a sustained-release kinetic model as described in any one of claims 1 to 8.

Images