3D Multi-agent models for protein release from PLGA spherical particles with complex inner morphologies

Abstract

In order to better understand and predict the release of proteins from bioerodible micro-or nanospheres, it is important to know the influences of different initial factors on the release mechanisms. Often though it is difficult to assess what exactly is at the origin of a certain dissolution profile. We propose here a new class of fine-grained multi-agent models built to incorporate increasing complexity, permitting the exploration of the role of different parameters, especially that of the internal morphology of the spheres, in the exhibited release profile. This approach, based on Monte-Carlo (MC) and Cellular Automata (CA) techniques, has permitted the testing of various assumptions and hypotheses about several experimental systems of nanospheres encapsulating proteins. Results have confirmed that this modelling approach has increased the resolution over the complexity involved, opening promising perspectives for future developments, especially complementing in vitro experimentation.

Quantitative multi-agent models for simulating protein release from PLGA bioerodible nano- and microspheres

“…move to a neighbouring site with specified probability. 4 Validation with experimental data for quantitative measurements Table 1 presents a list of variables, the values of which need to be determined in order to perform a simulation aiming validation of a model version (several versions are available for a number of internal configurations of the spheres [23]), or prediction of the dissolution profile in a given experimental situation.…”

“…Figure 9, (b) illustrates performance of the model calibrated to simulate release of carbonic anhydrase from microspheres of size ≃1 µm, described, like previous spheres, in [9]. More details on this example are available in [23], which focuses on the importance of correctly modelling the internal mor- …”

“…A more detailed description about modelling the internal morphology of the spheres is given in a parallel paper [23]. …”

Parallelisation strategies for large scale cellular automata frameworks in pharmaceutical modelling

“…In the present high performance computing era, the resources to process such models efficiently are readily available, but algorithmic solutions in this field tend to lag behind. The key problem to solve is the expense in terms of time and data load of cross-process communication present in parallelisation of such models, a consequence of the fact that many of them incorporate agent-like behaviour within the CA framework [2].…”

A Mathematical Model for Predicting Release Rate from Multiple Emulsions and Micro/Nanospheres

“…In most cases, for modeling the release process from multiple emulsions and microemulsions, the existing analytical models and simulations of diffusion-controlled drug release from polymeric non-degrading (porous/macroporous) microspheres can be applied [5][6][7]. Diffusion inside permeable liquid membranes is modeled similarly to microspheres, but with a lower diffusion coefficient.…”