Predicting busulfan exposure in patients undergoing hematopoietic stem cell transplantation using machine learning techniques

ABSTRACT

Purpose

This study aimed to establish an optimal model to predict the busulfan (BU) area under the curve at steady state (AUC_ss) by using machine learning (ML).

Patients and methods

Seventy-nine adult patients (age ≥18 years) who received BU intravenously and underwent therapeutic drug monitoring from 2013 to 2021 at Fujian Medical University Union Hospital were enrolled in this retrospective study. The whole dataset was divided into a training group and test group at the ratio of 8:2. BU AUC_ss were considered as the target variable. Nine different ML algorithms and one population pharmacokinetic (pop PK) model were developed and validated, and their predictive performance was compared.

Results

All ML models were superior to the pop PK model (R2 = 0.751, MSE = 0.722, 14 and RMSE = 0.830) in model fitting and had better predictive accuracy. The ML model of BU AUC_ss established through support vector regression (SVR) and gradient boosted regression trees (GBRT) had the best predictive ability (R² = 0.953 and 0.953, MSE = 0.323 and 0.326, and RMSE = 0.423 and 0.425).

Conclusion

All the ML models can potentially be used to estimate BU AUC_ss with the aim of facilitating rational use of BU on the individualized level, especially models built by SVR and GBRT algorithms.

KEYWORDS:

• Machine learning
• Pharmacokinetic
• Busulfan
• Area under the curve
• Prediction
• Therapeutic drug monitoring

Acknowledgments

We thank the General Direction of the Fujian Medical University Union Hospital and Fujian Medical University for continuous encouragement in our studies.

Declaration of interest

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Ethics statement

The authors state that study was carried out in compliance with the guidelines of the Helsinki Declaration with the approval of the Fujian Medical University Union Hospital Research Ethics Committee. The requirement for informed consent was waived owing to its retrospective nature.

Author contributions

X Wu and M Liu participated in the research design and method execution. D Li and J Zhao analyzed the data, drafted the manuscript, and performed the analysis with constructive discussions. B Xu, H Huang and Y Zheng contributed to data analysis and plotting. S Han and X Wu polished the language. All authors revised and approved the manuscript.

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/17512433.2023.2226866

Additional information

Funding

This paper was funded by the Joint Funds for the innovation of science and Technology, Fujian Province (NO. 2019Y9059, China).