Preparation, Screening, Selection, and Ex Vivo Skin Permeability Evaluation of Curcumin-Loaded Microemulsions for Topical Delivery: A Two-Level Full Factorial Approach

Abstract

Background: Curcumin is a polyphenolic natural product with well-documented anti-inflammatory and antioxidant activity; however, its use as a topical agent is constrained by aqueous insolubility, poor partitioning across the stratum corneum, and Biopharmaceutics Classification System (BCS) class IV behavior. Objectives: This study aimed to prepare, screen, and characterize curcumin-loaded microemulsions; to quantify their capacity to enhance delivery across excised rat skin relative to an aqueous suspension; and to select a lead candidate among the tested compositions. Methods: Eight microemulsions were prepared using oleic acid and Transcutol P (oil phase), Tween 80/Span 20 (surfactant), and propylene glycol (co-surfactant), according to a 23 full factorial design. The independent variables were total oil content (5% vs. 50% w/w), water-phase fraction (5% vs. 10% w/w), and the surfactant-to-co-surfactant (Smix) ratio (1:1 vs. 3:1). Droplet size, polydispersity index (PDI), viscosity, in vitro release, and ex vivo permeation through rat abdominal skin (Franz cells; barrier integrity verified) were evaluated. Effects were modeled using ordinary least-squares regression and tested by analysis of variance (ANOVA) against a pure-error term pooled from triplicate measurements; reduced models, confidence intervals, and lack-of-fit tests were used to evaluate the permeation responses. Results: The validated UV-spectrophotometric assay was linear (y = 68.019x − 0.0025; R2 = 0.9997), precise (relative standard deviation [RSD] < 2.5%), and accurate (recovery 96.58 - 105.64%), with a limit of detection (LOD) of 0.22 µg/mL and a limit of quantification (LOQ) of 0.68 µg/mL. Droplet size ranged from 34.0 ± 1.0 to 158.0 ± 2.0 nm. Oil content was the dominant factor affecting size, PDI, steady-state flux (Jss), and permeability (P < 0.0001 for size), whereas the Smix ratio governed the apparent diffusion coefficient (P = 0.037). Reduced models for flux and the diffusion coefficient showed no significant lack of fit (P = 0.529 and P = 0.598, respectively). The lead formulation F5 (5% oil, 5% water, Smix 3:1) exhibited a droplet size of 34.0 ± 1.0 nm, a PDI of 0.201 ± 0.004, a 24-h cumulative release of 94.95% (Higuchi kinetics, R2 = 0.9994, n = 0.417), a steady-state flux of 6.16 ± 0.60 µg/cm2/h versus 0.63 ± 0.10 µg/cm2/h for the aqueous comparator, and the shortest lag time (0.82 ± 0.36 h), reaching approximately 133 µg/cm2 of cumulative 24-h permeation versus approximately 11 µg/cm2 for the comparator. Enhancement ratios across the series ranged from 7.2 to 10.5, expressed as ex vivo screening ratios. Conclusions: The microemulsion platform improved curcumin release and skin permeation relative to an aqueous suspension. F5 was identified as the lead candidate among the eight compositions screened, rather than as a formal optimum, because a two-level design without center points cannot detect curvature. The aqueous suspension was a non-solubilized negative control, not a positive control, and formal sink conditions were not maintained at the 24-h endpoint; both points are addressed in the Limitations. F5 represents a promising vehicle for an ex vivo and in vitro proof-of-concept topical curcumin product, warranting subsequent in vivo safety evaluation, including formal skin irritation testing, and efficacy evaluation.