Related Experiment Video
Updated: Jan 18, 2026
Author Spotlight: Generating Neuronal Phenotypic Profiles - A Protocol to Culture and Image Human Midbrain Dopaminergic Neurons
Published on: July 7, 2023
Fraction-based Linear Extrapolation (FLEX) Method for Predicting Human Pharmacokinetic Clearance: Advanced Allometric
Yuki Umemori1, Koichi Handa2, Saki Yoshimura1
1Axcelead Tokyo West Partners, Inc. Translational Science, Discovery DMPK, Hino-Shi, Tokyo, 191-0065, Japan.
Accurate human clearance prediction is crucial for drug development. A new method combining threshold-based scaling and machine learning improves predictions for compounds with low unbound fractions, aiding early-stage drug decisions.
Area of Science:
- Pharmacokinetics and Drug Metabolism
- Computational Chemistry and Cheminformatics
- Drug Discovery and Development
Background:
- Accurate prediction of human clearance (CL) is vital in early drug development.
- Single Species Scaling (SSS) using rat pharmacokinetic (PK) data is common but less accurate for compounds with very low unbound plasma fraction (fu,plasma).
- Existing methods lack a systematic approach to address the limitations of SSS for compounds with extremely low fu,plasma.
Purpose of the Study:
- To develop and validate a novel approach for improving human CL prediction, particularly for compounds with low fu,plasma.
- To systematically validate the Single Species Scaling unbound (SSS fu Rat) method using an independent dataset.
- To integrate threshold-based allometry with machine learning for enhanced predictive accuracy.
Main Methods:
- Developed Fraction-based Linear EXtrapolation SSS (FLEX-SSS fu Rat), a method that adaptively switches between SSS fu Rat and SSS Rat based on an optimized fu threshold.
- Derived optimal thresholds and scaling coefficients using a training set of 200 compounds.
- Built a random forest (RF) machine learning model utilizing molecular descriptors and validated both models with an external dataset of 62 compounds.
Main Results:
- All five predictive models demonstrated comparable performance.
- A consensus model combining FLEX-SSS fu Rat and RF achieved the best results.
- The consensus model predicted human CL within a 2-fold error for 40.3% of compounds, with only 16.1% exceeding a 5-fold error, and a geometric mean fold error (GMFE) of 2.7.
Conclusions:
- This study provides the first systematic validation of SSS fu Rat on an independent dataset.
- The integration of threshold-based allometry and machine learning significantly enhances the accuracy of human CL prediction.
- The developed approach supports more informed decision-making for first-in-human dose selection in drug development.
More Related Videos
04:09Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma
Published on: October 10, 2018
08:59An Intestine/Liver Microphysiological System for Drug Pharmacokinetic and Toxicological Assessment
Published on: December 3, 2020
Related Concept Videos
01:17Clearance Models: Noncompartmental Models
The noncompartmental approach capitalizes on extensive sampling data, correlating the volume of distribution to systemic exposure and the administered dosage. This method enables...
01:26Pharmacokinetic Models: Comparison and Selection Criterion
Physiological models take a detailed approach by considering specific molecular processes. They can predict drug distribution, metabolism, and elimination changes, providing a comprehensive understanding of how drugs interact with the body.
00:56One-Compartment Open Model for IV Bolus Administration: Estimation of Clearance
In the one-compartment open model for intravenous (IV) bolus administration, clearance is estimated by dividing the elimination rate by the plasma drug concentration. This equation leverages the elimination rate constant and the apparent...
00:59Model-Independent Approaches for Pharmacokinetic Data: Noncompartmental Analysis
One important characteristic of noncompartmental analyses is that drug exposure increases proportionally with increasing doses. This...
01:01Dosage Regimens: Partial Pharmacokinetic Parameters
01:06Model Approaches for Pharmacokinetic Data: Distributed Parameter Models
The distributed parameter models are specifically designed to account for variations and differences in some drug classes. This model is particularly useful for assessing regional concentrations of anticancer or...