Abstract
Background: A key challenge in modern pharmaceutical research is developing predictive models for drug formulation behavior. Since permeability is closely linked to molecular properties, considering a broad of characteristics is essential for building reliable predictive tools. Near-infrared spectroscopy (NIR), a non-destructive, non-invasive, and chemically specific method, offers a powerful alternative to current gold-standard methods approved by regulatory agencies. Objectives: This study aims to apply a partial analytical quality by design (AQbD) approach to enhance the understanding and development of NIR and RP-HPLC methodologies. Methods: The employment of NIR with multivariate data analysis enabled the establishment of chemometric models for the classification and quantification of bifonazole (BFZ) in cream formulations. Results: An analytical target profile (ATP) was defined to guide the selection of critical method variables and support method design and development activities. Risk assessment was carried out using an Ishikawa diagram. For the RP-HPLC method, key performance parameters such as peak area, theoretical plates, tailing factor, and assay were evaluated, while NIR spectra and BFZ concentration were considered for method performance. The quantification models enabled the accurate determination of BFZ content, yielding results of 8.48 mg via NIR and 8.34 mg via RP-HPLC, with an RSD of 1.25%. Conclusions: These findings demonstrate the robustness and reliability of the models, making them suitable for routine quality control of BFZ formulations. Future research should aim to explore its use for monitoring permeation dynamics in real time and integrating it into regulatory frameworks to standardize its application in pharmaceutical quality control and formulation development.