Artificial Neural Networks to Predict the Apparent Degree of Supersaturation in Supersaturated Lipid-Based Formulations: A Pilot Study

Abstract

In response to the increasing application of machine learning (ML) across many facets of pharmaceutical development, this pilot study investigated if ML, using artificial neural networks (ANNs), could predict the apparent degree of supersaturation (aDS) from two supersaturated LBFs (sLBFs). Accuracy was compared to partial least squares (PLS) regression models. Equilibrium solubility in Capmul MCM and Maisine CC was obtained for 21 poorly water-soluble drugs at ambient temperature and 60 °C to calculate the aDS ratio. These aDS ratios and drug descriptors were used to train the ML models. When compared, the ANNs outperformed PLS for both sLBFCapmulMC (r2 0.90 vs. 0.56) and sLBFMaisineLC (r2 0.83 vs. 0.62), displaying smaller root mean square errors (RMSEs) and residuals upon training and testing. Across all the models, the descriptors involving reactivity and electron density were most important for prediction. This pilot study showed that ML can be employed to predict the propensity for supersaturation in LBFs, but even larger datasets need to be evaluated to draw final conclusions.

A dataset of formulation compositions for self-emulsifying drug delivery systems

“…In the context of oral lipid-based formulations, ML and computational techniques have played a role in early-stage development, notably based on small molecule drug solubility screening 15 . Preliminary ML modeling has been used to predict drug supersaturation in lipid-based formulations and increases in the apparent solubility of drug upon dispersion of SEDDS 16 , 17 . In these cases, a limited number of formulation compositions (i.e., two representative examples) were explored.…”

Comparative Analysis of Chemical Descriptors by Machine Learning Reveals Atomistic Insights into Solute–Lipid Interactions

“…Over the past decade, data-driven approaches have made significant strides in predicting solubility in formulation vehicles utilizing quantitative structure property relationships (QSPR) . Conceptually, the accuracy of any machine learning model depends on the quality of the data set, the algorithm used, as well as the way in which molecular properties are being encoded.…”

“…Over the past decade, data-driven approaches have made significant strides in predicting solubility in formulation vehicles utilizing quantitative structure property relationships (QSPR). 10 Conceptually, the accuracy of any machine learning model depends on the quality of the data set, the algorithm used, as well as the way in which molecular properties are being encoded. Pioneering research in understanding and predicting solubility in triglycerides has been conducted utilizing various modeling techniques and features such as two-dimensional (2D) and three-dimensional (3D) descriptors, as well as solvation parameters.…”