(M1130-08-50) Personalized Drug Delivery via 3D-Printed Tablets: Optimizing Patient-Specific Release Kinetics Using a Central Composite Design

Purpose:  Personalized medicine requires dosage forms tailored to individual patient needs, including adjustable dosing and release profiles. The study evaluates 3D-printed tablets as a platform for customizable drug delivery, focusing on how polymer composition and print parameters modulate release kinetics to optimize therapeutic outcomes. An understanding of factors influencing release rate will facilitate the design of customized 3D-printed tablets that can be programmed to release drug at a predetermined rate. Our study shows that release rate can be tailored to account for the physico-chemical properties of the drug by modulating properties of the polymer blend and the print conditions. In this project we set out to print 3D-tablets from 3 different polymer blends, varying the infill density and infill pattern to quantitatively assess their individual contributions to release of a drug through development of a regression equation that could characterize drug release kinetics. Statistical analyses were used to solve for the infill density and infill pattern contribution to controlling release. 

Methods: Custom polymer filaments were extruded on a Three-Tec ZE 12mm twin screw extruder from blends of either HPC/PEG, PVP/HPC/PEG or EC/HPC/PEG which represented a range of physical and mechanical properties. These polymer blends together with the print conditions were used to optimize drug release rates. 3D printed tablets with infill density (20, 50, 80, 100%) and infill patterns (grid, concentric, hexagonal) were printed using the HPC/PEG polymer filaments. The Central Composite Face-Centered (CCD) design was used to efficiently test different print conditions to understand the relationship between the infill density and infill pattern and release rate. Drug release of the 3D-printed tablets was assessed by dissolution testing with spectrophotometric analysis of a model drug, methylene blue. The model was validated by selecting 3 different print conditions (30% infill/hexagonal, 40% infill/concentric, 90% infill/ grid) and comparing the actual (experimentally measured) and predicted (calculated) release rates.

Results: 3D-printed tablets demonstrated controlled delivery over 24 hours. Increasing the infill density reduced the drug release rate as the polymer restricted the flow of buffer solution into the tablets. Drug release was rapid from the hexagonal design due to the large internal surface area. Drug release from the PVP/HPC/PEG polymer was the fastest, while the EC/HPC/PEG polymer tablets remained as an intact shell. The HPC/PEG polymer gelled, potentially limiting drug release. Higuchi analyses confirmed that drug release from the tablets occurred through a diffusion mechanism and the slopes of the Higuchi analysis were assigned as a Response Variable for the Minitab analysis of the CCD. The regression equation that was derived was: R = 67.100 - 0.711a - 0.133b + 0.00294a² + 0.00303b² - 0.00247ab. Where R=release rate, a=infill density, b=infill pattern. This regression equation and corresponding response surface plots demonstrated primarily that infill density was more effective than infill pattern at modulating drug release. The Center Point Sample (50% infill density, grid pattern) was replicated 5 times and demonstrated statistical reproducibility. A good correlation was observed between the experimentally measured and calculated release rates (r²=0.8937) validating the regression equation developed.

Conclusion: CCD is an efficient statistical optimization technique to determine the contribution of different polymer compositions and print conditions in complex systems. In this study the HPC/PEG was amenable to printing under varying print conditions and infill density was more effective than infill pattern in modulating release rate. With this information, different print conditions can be selected to customize drug release rates. CCD allowed for derivation of a quantitative regression analysis that facilitated a mechanistic understanding of the impact of print conditions on drug release.

Acknowledgements: Ashland