Page 346 - IJB-8-4

Three-Dimensional Printing Technologies for Drug Delivery Applications
Table 14. Regulatory considerations for materials, processes and products when applying 3D printing technologies
Features
Materials The ink should go through a strict quality control evaluation to guarantee its homogeneity and traceability as
with any raw material in manufacturing processes [129] .
In case of powder, a standardized procedure is required to ensure a uniform particle size and polydispersity index [129] .
As in any manufacturing process, quality control is carried out to evaluate impurities, among other elements. It
is necessary to have a procedure to evaluate unreacted photoinitiators, free radicals or even degradation [132] .
Furthermore, it is necessary to evaluate the level of toxicity and to verify if it is acceptable or not [132] .
Processes Proper documentation of all printing parameters and procedures will reduce the risk of inaccuracies in the
manufacturing process [129] .
Laser beam energy density, scanning speed, deposition velocity, and humidity are parameters that need to be
measured to guarantee the consistency of the manufacturing process. As it is known, these parameters have a
great impact on the physical characteristics of the final product [129] .
Products Specialized quality control tests of the final product should be performed to obtain an accurate result of the
design. Some examples for quality control tests are surface laser scanning, micro-CT, and various printer
monitoring strategies [129] .
A cleaning process of the finished product should be carried out regardless the 3D printing process that was
used to remove, for example, support material, residual monomers, etc. [129] .

Furthermore, 3D printing techniques require a design (CAD) images to allow ML algorithms to
unique production environment and/or the use of some identify the difference among designs, calculating their
specific resources, such as a highly specialized laser [127] . complexity [143] . Also in 2022, a study working on ML to
Current challenges of 3D printing technologies for drug predict 3D printing performance parameters of different
delivery applications can lead to a long trial and error formulations, such as processing temperatures (extrusion
process before transforming it from a laboratory to a and printing temperatures), feedstock characteristics, and
revolutionary manufacturing process [127,133] . Large-scale printability, was published. Ong et al. mined data on hot-
manufacturing represents a big challenge as explained in melt extrusion (HME) and fused deposition modeling
previous sections; different techniques and processes have (FDM), and an extensive range of different 3DP
emerged and the evolution to mass production and further formulations to optimize product design without having
commercialization will also require an entire ecosystem it physically. Through this research, it was discovered
where academy, industry, and government participate in that the simulated drugs had accurate release profiles;
to facilitate all the essential conditions [128,131,133] . this represents a strong advantage in terms of time saving
Despite the challenges presented to date, 3D printed because each iteration would take days [127] .
drug delivery system has a promising future that will ML has also been used in decision trees for HME
change the course of current healthcare. Synergic efforts in and artificial neural networks (ANNs) to enhance the
different fields are required. They include a sustainability quality of drug products throughout the pharmaceutical
focus to produce eco-friendly and physiologically safe workflow. In addition, ANNs have correctly predicted
excipients and filaments, research to reduce waste of 3D the dissolution profiles of ibuprofen-loaded Printlets™
printing processes, and studies for the improvement of the fabricated using DLP [144,145] . Moreover, ML has
dosage accuracy until the incursion to digitalization [131] . contributed to predicting the required force for penetration
Machine learning (ML), which is an application of of 3D printed microneedle arrays (MLA) as well as the
artificial intelligence (AI) to enable pattern recognition capabilities for their insertion into the skin [146] .
from large and complex datasets, is gaining presence Considering the advances worldwide, a strong
in the 3D printing field [134-141] . This tool contributes to emphasis on collaborative work in the digital era is
product quality and productivity by in situ monitoring, expected to happen. In the future, automatization and
optimizing design and process parameters, and speeding robotics will be a reality, giving rise to more innovative
up the microstructure evolution prediction [142] . and efficient 3D printed drugs. In a more distant future,
In this context, ML has been applied in different a big transformation from 3D to 4D printed drugs is
3D printing techniques to estimate performance and foreseen, and this next generation of drugs will come
quality indicators. Recently in 2022, an integration of true thanks to shape memory materials [10,147] . This new
ML and 3D printing through a graphical user interface technology enables adaptability and dynamic response to
for printing parameter optimization was published. While the structures according to the desired effect or shape [147] .
the majority of 3D printing research considers orthogonal 4D printing will bring unique characteristics in bio-
designs, authors employed nine different computer-aided robotics for medical purposes, such as drug delivery,

338 International Journal of Bioprinting (2022)–Volume 8, Issue 4