Evaluating the performance of machine‐learning regression models for pharmacokinetic drug–drug interactions

Abstract

Combination therapy or concomitant drug administration can be associated with pharmacokinetic drug–drug interactions, increasing the risk of adverse drug events and reduced drug efficacy. Thus far, machine‐learning models have been developed that can classify drug–drug interactions. However, to enable quantification of the pharmacokinetic effects of a drug–drug interaction, regression‐based machine learning should be explored. Therefore, this study investigated the use of regression‐based machine learning to predict changes in drug exposure caused by pharmacokinetic drug–drug interactions. Fold changes in exposure relative to substrate drug monotherapy were collected from 120 clinical drug–drug interaction studies extracted from the Washington Drug Interaction Database and SimCYP compound library files. Drug characteristics (features) were collected such as structure, physicochemical properties, in vitro pharmacokinetic properties, cytochrome P450 metabolic activity, and population characteristics. Three different regression‐based supervised machine‐learning models were then applied to the prediction task: random forest, elastic net, and support vector regressor. Model performance was evaluated using fivefold cross‐validation. Strongest performance was observed with support vector regression, with 78% of predictions within twofold of the observed exposure changes. The results show that changes in drug exposure can be predicted with reasonable accuracy using regression‐based machine‐learning models trained on data available early in drug discovery. This has potential applications in enabling earlier drug–drug interaction risk assessment for new drug candidates.

Utility of PBPK Modeling in Predicting and Characterizing Clinical Drug Interactions

“…Indeed, successful examples of the use of regression-based ML models being used to predict DDI due to inhibition or induction can be found in the literature (Gill et al, 2023). Whereas middle-out or top-down PBPK models are often used later in the development stage of clinical candidates, the appropriate use of AI and ML to more rapidly inform bottom-up PBPK models earlier in drug discovery should allow for more facile and sooner use of PBPK modeling to predict DDI, so long as the appropriate level of confidence in such an approach can be achieved.…”

Advancing Precision Medicine: A Review of Innovative In Silico Approaches for Drug Development, Clinical Pharmacology and Personalized Healthcare

“…Gill et al used regression-based ML to predict drug exposure changes due to interactions [ 195 ]. The model, with 78% accuracy within twofold observed changes, highlighted early drug-discovery features for risk assessment.…”

“…The model, with 78% accuracy within twofold observed changes, highlighted early drug-discovery features for risk assessment. Despite potential biases, it showcased ML’s power in capturing relationships, aiding decision-making in drug discovery [ 195 ]. Song et al focused on DDI prediction using similarity-based ML, achieving an AUROC exceeding 0.97 [ 196 ].…”

Machine learning in toxicological sciences: opportunities for assessing drug toxicity

“…The MACCS fingerprint models performed best (based on accuracy and area under the ROC curve) (Li et al, 2023). Another study focused more on the dominant drugmetabolizing isoform, cytochrome P450 3A4, found that regression-based ML (using a support vector regressor) was most effective for drugs from 120 studies (Gill et al, 2023).…”