Abstract
INTRODUCTION
. Despite the widespread use of solid dosage forms (SDFs) with modified release (MR), there is a paucity in the literature of comprehensive reviews covering the full development cycle — from selection of a technological platform to the application of digital optimization tools. Existing publications typically focus on individual technological platforms without addressing digital optimization.
AIM.
To systematically review and critically evaluate modern approaches to optimi zing the manufacturing technology for obtaining of modified-release solid dosage forms, and to determine the role of digital tools in enhancing the efficiency of pharmaceutical development.
DISCUSSION.
The analysis of scientific publications showed that hydrophilic matrices based on hydroxypropyl methylcellulose (HPMC) remain the leading platform for the development of prolonged-release formulations. However, controlling the release of highly soluble drugs requires combining hydrophilic and hydrophobic polymers. Osmotic systems provide a pH-independent release profile, but their use is associated with the risk of gastrointestinal irritation. Multiparticulate forms reduce the likelihood of unintentional release of dose dumping; in such systems, the quality of the polymer film coating governs the reproducibility of release kinetics. Hot-melt extrusion combined with 3D printing using fused deposition modeling (FDM) enables the creation of a desired release profile by varying the geometry of the dosage form. It has been established that the Quality by Design concept and machine lear ning methods substantially reduce the experimental workload, although challenges related to the interpretability of deep neural networks and the lack of external validation of predictive models remain unresolved.
CONCLUSIONS.
Matrix systems based on HPMC hold a leading position among prolonged-release technologies owing to their scalability and predictable kinetics; however, the optimal choice of technology for solid dosage forms with modified release is determined by the physicochemical properties of the substance and the pharmacokinetic profile. The application of design of experiments methods and machine learning in the development of MR SDFs can substantially reduce the number of laboratory experiments. A promising direction for future research lies in the development of interpretable predictive models and the adaptation of the regulatory framework to dosage forms produced by additive manufacturing (three-dimensional printing).