The development of effective therapies for inner ear disorders, particularly sensorineural hearing loss (SNHL), requires robust experimental models that accurately evaluate drug delivery systems and therapeutic agents. The round window membrane (RWM) represents the principal anatomical entry site for transtympanic (TT) drug administration to the inner ear and therefore plays a critical role in determining therapeutic efficacy. This review provides a comprehensive overview of the anatomical and physiological characteristics of the RWM and critically examines the experimental approaches used to investigate drug transport across this barrier. In vivo, ex vivo, and in vitro models are compared with respect to their methodological characteristics, translational relevance, advantages, limitations, and most appropriate applications. In vivo models provide integrated information on pharmacokinetics, biodistribution, safety, and therapeutic outcomes, but are limited by biological variability, alternative routes of inner-ear entry, ethical considerations, and cost. Ex vivo systems preserve native tissue architecture in a controlled environment, although tissue availability, viability, and inter-donor variability remain important constraints. In vitro models enable mechanistic studies and higher-throughput screening while reducing animal use but currently lack the full cellular and anatomical complexity of the native human RWM and require further validation. Analytical, imaging, and computational methods used to quantify transport, distribution, and retention are also summarized. Together, these complementary platforms support the rational development of next-generation ototherapeutic strategies.
Nanoparticles have been extensively studied as drug carriers for tumor theranostics. However, the rapid clearance effect of the mononuclear phagocyte system (MPS) seriously hinders their efficient accumulation at the tumor site. To address this limitation, a bioinspired nanotherapeutic platform, SPDGDOX NCs, was developed to utilize a “don’t-eat-me signaling” strategy, enabling effective evasion of the MPS. The SPDGDOX NCs comprised the following components: i) a core structure of PDA nanoparticles after modification with PEI; ii) the CD47-mimicking peptide fragment “self-peptide” (SP) conjugated to the NCs to convey the “don’t-eat-me” signal; iii) the paramagnetic metal ion Gd3+ and DOX linked to NCs as agents for MR imaging and antitumor chemotherapy, respectively. Briefly, upon circulation in the bloodstream, peptide SP on SPDGDOX NCs could activate the CD47-SIRPa immune response. Subsequently, SPDGDOX NCs could evade phagocytosis, facilitating persistent circulation in the bloodstream, thus enhancing their enrichment in tumors. As confirmed by various in vitro and in vivo studies, SPDGDOX NCs exhibited robust anti-tumor efficacy and excellent T1-weighted MRI effect, significantly inhibiting tumor growth and prolonging survival time. Overall, the SPDGDOX NCs provide a convenient strategy for improving MRI-traceable anticancer efficacy, which is expected to have future clinical applications in cancer theranostics.
Atopic dermatitis (AD) is a chronic inflammatory skin disease characterized by skin barrier dysfunction and immune dysregulation, which may influence topical drug penetration and therapeutic efficacy. This study evaluated a microneedle-assisted approach to enhance the cutaneous delivery of compound betamethasone in AD. Franz diffusion cell experiments demonstrated 3.29-fold and 4.67-fold increases in the 24-h cumulative permeation of betamethasone dipropionate (BDP) and betamethasone sodium phosphate (BSP), respectively, compared with passive application. In vivo fluorescence imaging showed greater penetration depth of fluorescein sodium following microneedle treatment, while histological examination revealed transient epidermal disruption. In a 2,4-dinitrochlorobenzene (DNCB)-induced AD mouse model, the combined treatment was associated with reduced epidermal thickening, mast cell accumulation, and serum IgE and IL-13 levels. In a pilot split-body clinical study involving 28 participants with mild-to-moderate AD, the microneedle + compound betamethasone-treated side showed greater improvements in Scoring Atopic Dermatitis (SCORAD) intensity and numeric rating scale (NRS) pruritus scores than the contralateral microneedle + saline control side after 7 days. No serious adverse events were observed during the study period. These findings demonstrate enhanced cutaneous permeation of compound betamethasone following microneedle-assisted application and provide preliminary evidence of therapeutic activity and short-term tolerability in AD.
The development of polymeric nanoparticles remains largely empirical, requiring extensive experimental screening to identify suitable formulation and processing conditions for each active pharmaceutical ingredient (API). Machine learning (ML) offers opportunities to support data-driven drug-delivery development, but existing approaches are often restricted to individual compounds and lack external validation. Here, we developed NANO-AI, a two-step ML framework integrating formulation variables, process parameters, and molecular descriptors to predict nanoparticle formation and key physicochemical attributes. The framework was trained on 925 experimental formulations encompassing nine physicochemically diverse APIs. In the first step, classification models predicted successful nanoparticle formation, achieving an area under the receiver operating characteristic curve of 0.823. In the second step, regression models predicted particle size, polydispersity index, zeta potential, and encapsulation efficiency, with R² values above 0.95 for size, zeta potential, and encapsulation efficiency. Interlaboratory experiments demonstrated robustness across operators and experimental environments. Importantly, zero-shot validation using previously unseen APIs showed that the framework could generate useful predictions across external compounds spanning different degrees of physicochemical similarity to the training APIs. These findings establish NANO-AI as a generalizable data-driven framework for rational polymeric nanocarrier development, potentially reducing empirical screening and supporting formulation development across chemically diverse therapeutic compounds.
Biodegradable polymer-based microneedle (MN) patches enable self-administered, long-acting drug delivery. Controlled drug release from MNs requires controlled material distribution and minimal porosity in the MNs. MN detachment in the skin requires MNs without film formation beyond the MN base. To achieve these goals, we studied solvent removal dynamics during MN patch fabrication by a solvent-casting method to better control distribution of polymer and drug in the MN patch mold. More specifically, we designed casting solvents to control (i) solvent evaporation rate, (ii) solvent absorption rate into polydimethylsiloxane (PDMS) molds, and (iii) solubility of biodegradable polymer (poly (lactic acid)) and drug (etonogestrel, for long-acting contraception) in the solvent. We found that polymer film formation was prevented by using solvents with minimal mold absorption and moderate evaporation rates. We further demonstrated that solvent-nonsolvent mixtures titrated to the solubility limit of the polymer and beyond the solubility limit of the drug (forming a suspension of fine precipitate) also produced MNs without films. This occurred even when using solvents that absorbed into the molds. To better achieve sustained drug release, we showed that centrifuge-based casting and heat annealing reduced surface and bulk MN porosity. The resulting MNs achieved efficient detachment in porcine skin and sustained release of etonogestrel for two months in-vitro. This study establishes a framework to control solvent removal dynamics during MN patch fabrication to optimize biodegradable polymer MNs for long-acting delivery of etonogestrel and other drugs.
