Ghazi, Noha
(2024)
Developing analytical data-informed statistical models for predicting the physical stability of drug-polymer solid dispersions using high-throughput 2D printed arrays.
PhD thesis, University of Nottingham.
Abstract
Introduction:
Amorphous solid dispersions (ASDs) offer a promising strategy to address the poor solubility challenges in over 40% of newly discovered active pharmaceutical ingredients (APIs). Despite the potential advantages of solid dispersions, some challenges would hinder the development of ASDs such as scaling up and physical instability that lead to phase segregation and possible further recrystallization of the API, particularly during the storage period. Traditional preparation methods of ASDs as well as analytical techniques to assess formulation stability may require samples on a milligram-to-gram scale. However, whilst a number of drugs and polymers have been studied in the literature, the number of researched APIs and their loading with different polymers is still very limited, and model development is often done for each system in isolation, making it difficult to draw any general conclusions regarding the physicochemical properties of APIs that would be highly correlated to the stability of those ASDs.
Therefore, this thesis aims to combine the use of a novel high-throughput miniaturised screening approach for printing nano-arrays of a relatively larger number of samples than previously reported using a minimal amount of materials in picoliter with statistical modelling of the outcome stability data. That model could correlate the stability of solid dispersion formulations with the physicochemical properties of the utilized APIs. Then to test the predictive power of 2D inkjet printing through 3D inkjet printing, highlighting its utility in anticipating the physical stability of scaled-up drug-polymer dispersions.
Methods:
2-D Microarray Printing for Screening of Pharmaceutical Solid Dispersions. The use of printed nano-arrays in pre-formulation and solid-form screening, employing pico-litres of drugs with various physicochemical properties is investigated here. The study involves preliminary techniques, such as manual and contact printing, to assess formulation printability, with a subsequent focus on 2D inkjet printing on a nanogram scale. The novelty of the research lies in sample quantity, addressability, and the ease of analysis through exploring a large number of drug-polymer loadings formulations used, around 930 drug/polymer loadings in triplicates (around 2793 loadings) to build an extensive library of drug-polymer combinations. This is much more than has been achieved before in other studies while taking into consideration performing this safely using a minimal (nano-gram) amount of materials. The stability data collected for the prepared ASDs at accelerated conditions for six months is to be used as input in the statistical modelling.
Multiple Linear Regression Modelling for the Stability of Solid Dispersions. Through building an extensive library of formulations developed from 23 different drugs combined within two polymeric matrices sufficient data was gathered to form a statistical-based model. Stability under accelerated conditions served as a critical input for developing multiple linear regression models predicting the stability of amorphous solid dispersions. The models incorporate variables related to the stability of solid dispersions, such as hydrogen bond acceptors, heteroatoms, and oxygen atoms within drug molecules. Rigorous validation through Leave-One-Out Cross-Validation ensures the reliability of the models by confirming the same trend between measured and predicted stability data of different APIs.
3-D Inkjet Printing of Solid Dispersions. This is done to explore the viability of 3D inkjet printing to produce formulations highlighted in the 2D screening with a relatively high drug loading using poorly soluble APIs. The results demonstrate the effectiveness of the 3D printing process in assessing the stability of drug-polymer dispersions. The agreement between 2D and 3D stability outcomes reinforces the utility of 2D printing in early-stage formulation development, allowing for the anticipation of physical stability in scaled-up dispersions.
Results:
The 2D inkjet printing approach, with its capacity for high-throughput screening, proved to be a valuable tool for assessing drug-polymer miscibility limits and predicting stability outcomes. The extensive library of formulations enabled the development of multiple linear regression models, providing insights into the critical physicochemical properties influencing the stability of amorphous solid dispersions. The 3D inkjet printing results further validated the predictive power of the 2D approach. The formulations exhibited stability consistent with predictions, showcasing the potential for using 2D printing as a reliable tool for early-stage formulation development.
Conclusions:
In conclusion, this thesis introduces a novel approach to solid-dispersion formulation screening. The use of printed nano-arrays and statistical modelling enhances efficiency, minimizes material requirements, and broadens the scope of samples evaluated. The predictive power of 2D inkjet printing is confirmed through 3D inkjet printing, highlighting its utility in anticipating the physical stability of scaled-up drug-polymer dispersions. This work not only advances pharmaceutical formulation but also sets the stage for the practical application of these statistical models in the industry.
Actions (Archive Staff Only)
 |
Edit View |
Tools
Tools
