# In Silico Predictions Driving the Development of 3D-Printed Drug Delivery Systems

## TL;DR

This paper shows how computer predictions can help design better 3D-printed drug delivery systems by reducing trial-and-error in excipient selection and improving printability and drug release.

## Contribution

The study introduces a framework using in silico predictions of excipient miscibility and molecular dynamics simulations to optimize 3D-printed drug formulations.

## Key findings

- Miscibility parameters predicted drug-excipient printability with strong correlation to experimental results.
- Molecular dynamics simulations accurately predicted dissolution behavior based on cohesive energy density.
- Hydrophilic carriers showed faster drug release, while hydrophobic carriers enabled sustained release.

## Abstract

Background: Three-dimensional printing (3DP) is a promising technology for advancing pharmaceutical research by enabling the production of personalized drug products. However, its progress has been hindered by the conventional trial-and-error approach to excipient selection and optimization. Methods: In this study, the blend module was employed to determine the miscibility parameters—mixing energy (Emix) and Flory–Huggins interaction parameter (χ) to find the right excipients and drug–excipient ratio and examine the incorporation of plasticizers and lipids to enhance printability. Furthermore, molecular dynamics (MD) simulations were employed to calculate the cohesive energy density (CED) for predicting the dissolution behavior of 3DP formulations. Results: Data from 51 formulations were analyzed, enabling correlation and experimental validation of the in silico predictions. The predicted miscibility values demonstrated a strong correlation with experimental printability results. Furthermore, using a miscibility parameter, it was possible to accurately forecast minor changes in the drug-to-excipient ratio, plasticizer/lipid concentration, and hot-melt extrusion (HME) temperature that affect printability. Hydrophilic carriers exhibited lower CED values corresponding to faster drug release. In contrast, more hydrophobic carriers revealed high CED values, indicating stronger drug entrapment and sustained release. Conclusions: The miscibility parameters and MD-simulated CED values provide a practical framework for early-stage, high-throughput excipient screening. Overall, in silico prediction offers a viable strategy for modeling the entire 3DP workflow, minimizing the need for trial-and-error experimentation, and thereby accelerating the clinical translation of 3DP drug delivery systems.