Prediction of In Vivo Pharmacokinetic Parameters and Time–Exposure Curves in Rats Using Machine Learning from the Chemical Structure

Abstract

Animal pharmacokinetic (PK) data as well as human and animal in vitro systems are utilized in drug discovery to define the rate and route of drug elimination. Accurate prediction and mechanistic understanding of drug clearance and disposition in animals provide a degree of confidence for extrapolation to humans. In addition, prediction of in vivo properties can be used to improve design during drug discovery, help select compounds with better properties, and reduce the number of in vivo experiments. In this study, we generated machine learning models able to predict rat in vivo PK parameters and concentration–time PK profiles based on the molecular chemical structure and either measured or predicted in vitro parameters. The models were trained on internal in vivo rat PK data for over 3000 diverse compounds from multiple projects and therapeutic areas, and the predicted endpoints include clearance and oral bioavailability. We compared the performance of various traditional machine learning algorithms and deep learning approaches, including graph convolutional neural networks. The best models for PK parameters achieved R 2 = 0.63 [root mean squared error (RMSE) = 0.26] for clearance and R 2 = 0.55 (RMSE = 0.46) for bioavailability. The models provide a fast and cost-efficient way to guide the design of molecules with optimal PK profiles, to enable the prediction of virtual compounds at the point of design, and to drive prioritization of compounds for in vivo assays.

Computational Predictions of Nonclinical Pharmacokinetics at the Drug Design Stage

“…Figures S14, S16). Similar results were recently reported for the GP rat CLp model, calling for additional research needed into reliable quantification of uncertainty.…”

“…For example, if focusing on the drug-design stage results (i.e., in the absence of in vitro data), Obrezanova et al showed RMSLE values of 0.35, 0.31, and 0.57 for predictions of rat CLp, Vss, and oral F, respectively. 9 This result is comparable to our AutoGluon and GP average RMSLE values of 0.42, 0.39, and 0.57 for rat CLp, Vss, and oral F, respectively.…”

“…Our exploratory data set included mouse and rat PK studies performed at Roche in the last 30 years, consisting of 9,685 unique structures matched with corresponding aggregated PK properties (Supplementary Table 1). To our knowledge, this is one of the biggest nonclinical PK data sets described thus far, with rat data alone matching recent publications 9,21,23 and exceeding the number of publicly reported clinical PK results. 10,65,66 Compared to the well-known Lombardo data set of human PK results, 67 our rodent data set showed a significantly higher MW (median of 464 vs 371 Da), lipophilicity (median calc.…”

“…logP of 3.8 vs 2.0; logD of 2.61 vs 0.7), and polarity (TPSA of 100 vs 89 Å), with up to 53% of the molecules beyond the classical Lipinski drug-like space. 62 Compared to a recently published rat PK data set, 9 our compounds appear similarly lipophilic (median logD of approximately 2.6 vs 2.3). As our data set represented an exploratory drug discovery space tested in two rodent species (mouse and rat with overlap of 9%) and observed PK properties spanned a broad dynamic range (median property range of 700-fold; Supplementary Table 3), we hypothesized that this PK data set offers a possibility for a proof-of-concept study for applicability of computational PK at the drug-design stage.…”

“…Data set limitations, in terms of compound number and diversity, may also influence Multi-Task approaches, all of which showed comparable performance to Single-Task models (Supplementary Table 6), with the exception of improved overprediction prediction bias and correlation for oral F. This observation is in accordance with findings from Obrezanova et al, who report overall equivalent performance between Single-Task and Multi-Task models for most PK parameters. 9 Middling performance of neural networks (ChemProp, AttentiveFP, and DeepChem GCNN), consistently inferior to AutoGluon and GP, offers additional evidence for the data set limit hypothesis. To increase the data set size and diversity for graph neural networks, we tested a number of pretraining approaches with relevant in vitro data, 12 focusing on predictions of CLp and F. Even if observed effects were modest and statistically insignificant (Figure 4), pretrained ChemProp models of CLp offered consistent improvements of precision, bias, and correlation.…”