In silico formulation prediction of drug/cyclodextrin/polymer ternary complexes by machine learning and molecular modeling techniques

Advancing algorithmic drug product development: Recommendations for machine learning approaches in drug formulation

Artificial intelligence is a rapidly expanding area of research, with the disruptive potential to transform traditional approaches in the pharmaceutical industry, from drug discovery and development to clinical practice. Machine learning, a subfield of artificial intelligence, has fundamentally transformed in silico modelling and has the capacity to streamline clinical translation. This paper reviews data-driven modelling methodologies with a focus on drug formulation development. Despite recent advances, there is limited modelling guidance specific to drug product development and a trend towards suboptimal modelling practices, resulting in models that may not give reliable predictions in practice. There is an overwhelming focus on benchtop experimental outcomes obtained for a specific modelling aim, leaving the capabilities of data scraping or the use of combined modelling approaches yet to be fully explored. Moreover, the preference for high accuracy can lead to a reliance on black box methods over interpretable models. This further limits the widespread adoption of machine learning as black boxes yield models that cannot be easily understood for the purposes of enhancing product performance. In this review, recommendations for conducting machine learning research for drug product development to ensure trustworthiness, transparency, and reliability of the models produced are presented. Finally, possible future directions on how research in this area might develop are discussed to aim for models that provide useful and robust guidance to formulators.

Cyclodextrins and Their Derivatives as Drug Stability Modifiers

Cyclodextrins (CDs) are cyclic oligosaccharides that contain a relatively hydrophobic central cavity and a hydrophilic outer surface. They are widely used to form non-covalent inclusion complexes with many substances. Although such inclusion complexes typically exhibit higher aqueous solubility and chemical stability than pure drugs, it has been shown that CDs can promote the degradation of some drugs. This property of stabilizing certain drugs while destabilizing others can be explained by the type of CD used and the structure of the inclusion complex formed. In addition, the ability to form complexes of CDs can be improved through the addition of suitable auxiliary substances, forming multicomponent complexes. Therefore, it is important to evaluate the effect that binary and multicomponent complexes have on the chemical and physical stability of complexed drugs. The objective of this review is to summarize the studies on the stabilizing and destabilizing effects of complexes with CDs on drugs that exhibit stability problems.

A comparative study on inclusion complex formation between formononetin and β-cyclodextrin derivatives through multiscale classical and umbrella sampling simulations

Formononetin, a naturally occurring isoflavone exhibits a wide range of therapeutic applications including antioxidant, anti-tumor, antiviral, anti-diabetic and neuroprotective activities. However, the low hydro-solubility of formononetin has limited its prospective use in cosmetic, neutraceutical and pharmaceutical industries. Cyclodextrins (CDs), especially β-CD and its derivatives have emerged as promising agents to improve the water solubility of poorly hydrosoluble compounds by the formation of inclusion complexes. We employed multiscale (1000 ns) explicit solvent and umbrella sampling molecular dynamics (MD) simulations to study the interactions and thermodynamic parameters of inclusion complex formation between formononetin and five most commonly used β-CD derivatives. Classical MD simulations revealed two possible binding conformations of formononetin inside the central cavity of hydroxypropyl-β-CD (HP-β-CD), randomly methylated-β-CD (ME-β-CD), and sulfobutylether-β-CD (SBE-β-CD). The binding conformation with the benzopyrone ring of formononetin inside the central cavity of β-CD derivatives was more frequent than the phenyl group occupying the hydrophobic cavity. These interactions were supported by a variety of non-bonded contacts including hydrogen bonds, pi-lone pair, pi-sigma, and pi-alkyl interactions. Formononetin showed favorable end-state MD-driven thermodynamic binding free energies with all the selected β-CD derivatives, except succinyl-β-CD (S-β-CD). Furthermore, umbrella sampling simulations were used to investigate the interactions and thermodynamic parameters of the host-guest inclusion complexes. The SBE-β-CD/formononetin inclusion complex showed the lowest binding energy signifying the highest affinity among all the selected host-guest inclusion complexes. Our study could be used as a standard for analyzing and comparing the ability of different β-CD derivatives to enhance the hydro-solubility of poorly soluble molecules.

Integrating Machine Learning in the Coarse-Grained Molecular Simulation of Polymers

Machine learning (ML) is having an increasing impact on the physical sciences, engineering, and technology and its integration into molecular simulation frameworks holds great potential to expand their scope of applicability to complex materials and facilitate fundamental knowledge and reliable property predictions, contributing to the development of efficient materials design routes. The application of ML in materials informatics in general, and polymer informatics in particular, has led to interesting results, however great untapped potential lies in the integration of ML techniques into the multiscale molecular simulation methods for the study of macromolecular systems, specifically in the context of Coarse Grained (CG) simulations. In this Perspective, we aim at presenting the pioneering recent research efforts in this direction and discussing how these new ML-based techniques can contribute to critical aspects of the development of multiscale molecular simulation methods for bulk complex chemical systems, especially polymers. Prerequisites for the implementation of such ML-integrated methods and open challenges that need to be met toward the development of general systematic ML-based coarse graining schemes for polymers are discussed.

Anomalous Properties of Cyclodextrins and Their Complexes in Aqueous Solutions

Cyclodextrins (CDs) are cyclic oligosaccharides that emerged as industrial excipients in the early 1970s and are currently found in at least 130 marketed pharmaceutical products, in addition to numerous other consumer products. Although CDs have been the subject of close to 100,000 publications since their discovery, and although their structure and properties appear to be trivial, CDs are constantly surprising investigators by their unique physicochemical properties. In aqueous solutions, CDs are solubilizing complexing agents of poorly soluble drugs while they can also act as organic cosolvents like ethanol. CDs and their complexes self-assemble in aqueous solutions to form both nano- and microparticles. The nanoparticles have diameters that are well below the wavelength of visible light; thus, the solutions appear to be clear. However, the nanoparticles can result in erroneous conclusions and misinterpretations of experimental results. CDs can act as penetration enhancers, increasing drug permeation through lipophilic membranes, but they do so without affecting the membrane barrier. This review is an account of some of the unexpected results the authors have encountered during their studies of CDs as pharmaceutical excipients.

Re-envisioning the design of nanomedicines: harnessing automation and artificial intelligence

INTRODUCTION

Interest in nanomedicines has surged in recent years due to the critical role they have played in the COVID-19 pandemic. Nanoformulations can turn promising therapeutic cargo into viable products through improvements in drug safety and efficacy profiles. However, the developmental pathway for such formulations is non-trivial and largely reliant on trial-and-error. Beyond the costly demands on time and resources, this traditional approach may stunt innovation. The emergence of automation, artificial intelligence (AI) and machine learning (ML) tools, which are currently underutilized in pharmaceutical formulation development, offers a promising direction for an improved path in the design of nanomedicines.

AREAS COVERED

the potential of harnessing experimental automation and AI/ML to drive innovation in nanomedicine development. The discussion centers on the current challenges in drug formulation research and development, and the major advantages afforded through the application of data-driven methods.

EXPERT OPINION

The development of integrated workflows based on automated experimentation and AI/ML may accelerate nanomedicine development. A crucial step in achieving this is the generation of high-quality, accessible datasets. Future efforts to make full use of these tools can ultimately contribute to the development of more innovative nanomedicines and improved clinical translation of formulations that rely on advanced drug delivery systems.

Discrimination of β-cyclodextrin/hazelnut (Corylus avellana L.) oil/flavonoid glycoside and flavonolignan ternary complexes by Fourier-transform infrared spectroscopy coupled with principal component analysis

The goal of the study was the discrimination of β-cyclodextrin (β-CD)/hazelnut (Corylus avellana L.) oil/antioxidant ternary complexes through Fourier-transform infrared spectroscopy coupled with principal component analysis (FTIR-PCA). These innovative complexes combine the characteristics of the three components and improve the properties of the resulting material such as the onsite protection against oxidative degradation of hazelnut oil unsaturated fatty acid glycerides. Also, the apparent water solubility and bioaccessibility of the hazelnut oil components and antioxidants can be increased, as well as the controlled release of bioactive compounds (fatty acid glycerides and antioxidant flavonoids, namely hesperidin, naringin, rutin, and silymarin). The appropriate method for obtaining the ternary complexes was kneading the components at various molar ratios (1:1:1 and 3:1:1 for β-CD hydrate:hazelnut oil (average molar mass of 900 g/mol):flavonoid). The recovering yields of the ternary complexes were in the range of 51.5-85.3% and were generally higher for the 3:1:1 samples. The thermal stability was evaluated by thermogravimetry and differential scanning calorimetry. Discrimination of the ternary complexes was easily performed through the FTIR-PCA coupled method, especially based on the stretching vibrations of CO groups in flavonoids and/or CO/CC groups in the ternary complexes at 1014.6 (± 3.8) and 1023.2 (± 1.1) cm^-1 along the second PCA component (PC₂), respectively. The wavenumbers were more appropriate for discrimination than the corresponding intensities of the specific FTIR bands. On the other hand, ternary complexes were clearly distinguishable from the starting β-CD hydrate along the first component (PC₁) by all FTIR band intensities and along PC₂ by the wavenumber of the asymmetric stretching vibrations of the CH groups at 2922.9 (± 0.4) cm^-1 for ternary complexes and 2924.8 (± 1.4) cm^-1 for β-CD hydrate. The first two PCA components explain 70.38% from the variance of the FTIR data (from a total number of 26 variables). Other valuable classifications were obtained for the antioxidant flavonoids, with a high similarity for hesperidin and naringin, according to FTIR-PCA, as well as for ternary complexes depending on molar ratios. The FTIR-PCA coupled technique is a fast, nondestructive and cheap method for the evaluation of quality and similarity/characteristics of these new types of cyclodextrin-based ternary complexes having enhanced properties and stability.

The applications of Machine learning (ML) in designing dry powder for inhalation by using thin-film-freezing technology

Dry powder inhalers (DPIs) are one of the most widely used devices for treating respiratory diseases. Thin--film--freezing (TFF) is a particle engineering technology that has been demonstrated to prepare dry powder for inhalation with enhanced physicochemical properties. Aerosol performance, which is indicated by fine particle fraction (FPF) and mass median aerodynamic diameter (MMAD), is an important consideration during the product development process. However, the conventional approach for formulation development requires many trial-and-error experiments, which is both laborious and time consuming. As a state-of-the art technique, machine learning has gained more attention in pharmaceutical science and has been widely applied in different settings. In this study, we have successfully built a prediction model for aerosol performance by using both tabular data and scanning electron microscopy (SEM) images. TFF technology was used to prepare 134 dry powder formulations which were collected as a tabular dataset. After testing many machine learning models, we determined that the Random Forest (RF) model was best for FPF prediction with a mean absolute error of ± 7.251%, and artificial neural networks (ANNs) performed the best in estimating MMAD with a mean absolute error of ± 0.393 μm. In addition, a convolutional neural network was employed for SEM image classification and has demonstrated high accuracy (>83.86%) and adaptability in predicting 316 SEM images of three different drug formulations. In conclusion, the machine learning models using both tabular data and image classification were successfully established to evaluate the aerosol performance of dry powder for inhalation. These machine learning models facilitate the product development process of dry powder for inhalation manufactured by TFF technology and have the potential to significantly reduce the product development workload. The machine learning methodology can also be applied to other formulation design and development processes in the future.