Molecular Simulation and Statistical Learning Methods toward Predicting Drug–Polymer Amorphous Solid Dispersion Miscibility, Stability, and Formulation Design

Abstract

Amorphous solid dispersions (ASDs) have emerged as widespread formulations for drug delivery of poorly soluble active pharmaceutical ingredients (APIs). Predicting the API solubility with various carriers in the API–carrier mixture and the principal API–carrier non-bonding interactions are critical factors for rational drug development and formulation decisions. Experimental determination of these interactions, solubility, and dissolution mechanisms is time-consuming, costly, and reliant on trial and error. To that end, molecular modeling has been applied to simulate ASD properties and mechanisms. Quantum mechanical methods elucidate the strength of API–carrier non-bonding interactions, while molecular dynamics simulations model and predict ASD physical stability, solubility, and dissolution mechanisms. Statistical learning models have been recently applied to the prediction of a variety of drug formulation properties and show immense potential for continued application in the understanding and prediction of ASD solubility. Continued theoretical progress and computational applications will accelerate lead compound development before clinical trials. This article reviews in silico research for the rational formulation design of low-solubility drugs. Pertinent theoretical groundwork is presented, modeling applications and limitations are discussed, and the prospective clinical benefits of accelerated ASD formulation are envisioned.

Atomistic Descriptors for Machine Learning Models of Solubility Parameters for Small Molecules and Polymers

“…−0.65SHBa + 1.301SHdsCH − 5.580nAcid − 0.618SssCH2 − 15.805SHBd +15.53nHeavyAtom + 10.667 (11) This model does not contain any unreasonably small or large factors for descriptors, which indicates that there are no irrelevant or redundant descriptors. Figure 4 shows that MLR performed well for the training set of small molecules and the LOOCV, according to the R 2 score and other metrics.…”

“…Although new GC approaches are being developed, a general model that can cover a wide range of polymer species and polymer properties is not available [ 10 ]. Atomistic simulations employing force fields and interatomic potential functions are another tool for predicting polymer properties [ 5 , 11 ]. However, accurate SP predictions using atomistic simulations are computationally demanding, especially for polymers and compounds with complex structures [ 12 ].…”

“…In this regard, data-driven approaches based on machine learning (ML) models have become an appealing alternative to simple empirical approaches and atomistic simulations. ML models have been developed to predict the physical or chemical properties of materials with good accuracy, including solubility parameters [ 11 ]. However, the predictive power of ML models depends heavily on the availability of accurate and consistent target data covering a wide structural and compositional range, as well as unbiased descriptors, i.e., readily available observables that can be linked to the target property [ 13 ].…”

DPD Modelling of the Self- and Co-Assembly of Polymers and Polyelectrolytes in Aqueous Media: Impact on Polymer Science

“…Since the early application of Monte Carlo simulations by Ulam and his colleagues in the Manhattan Project more than 70 years ago, computer modelling and computer simulations in particular have become important tools of physical, chemical, biological and material research [85][86][87][88][89]. They enable checking of the correctness of theoretical hypotheses and predictions and provide data that are either inaccessible or barely accessible by experiments.…”

Molecular Dynamic Approach to Predict the Miscibility of Excipients for Lipid-based Formulations.

“…This may lead to hindrance to drug dissolution in the GI tract which invariably affects the bioavailability of oral dosage form [1]. This has been overcome by enhancing the API solubility with the use of drug delivery technologies such as particle amorphization, lipid-based systems, size reduction, salt formation, self-emulsi cation, complexation with cyclodextrins, cosolvent systems, micellar and surfactant systems [2].…”