文献速递治疗药物监测与药物基因组学特殊人群用药

机器学习模型与新建群体药代动力学模型预测儿童白消安暴露的性能比较

Comparing the Predictive Performance of Machine Learning Models with a Newly Developed Population Pharmacokinetic Model for Busulfan in Paediatric Patients.
PubMed 文献 · Clin Pharmacokinet · 2026-09-22

本研究基于丹麦67名儿童(637个观测)的回顾性丰富采样数据建立白消安群体药代动力学(popPK)模型及随机森林、XGBoost机器学习模型,并在荷兰另一中心171名儿童(1871个观测)的TDM数据上进行外部验证。一室模型拟合最佳,第2、3天清除率较第1天分别下降10.4%和13%。popPK与两种机器学习模型外部预测中位PE为-4.82%至-7.44%,中位绝对PE约15%,约80%数据预测误差在±30%内;popPK模型下0、0.5、1、4小时或0、1、2、4小时采样可充分预测浓度与暴露。

要点

  • 一室popPK模型拟合最佳,白消安清除率第2、3天较第1天分别下降10.4%和13%。
  • popPK与随机森林、XGBoost外部验证中位绝对PE约15%,约80%预测在±30%内。
  • popPK模型可采用0、0.5、1、4小时或0、1、2、4小时有限采样预测暴露。

核心结论

  • 新建popPK模型可支持儿童白消安个体化剂量调整并减少TDM采血样本。
  • 随机森林与XGBoost对白消安血浆浓度的预测性能与popPK模型(群体预测)相当。
  • 机器学习方法需更大数据集才能描述完整药代动力学特征。

研究设计与样本

  • 基于丹麦单中心67名儿童(637个观测)的回顾性丰富采样数据建立popPK与机器学习模型。
  • 在荷兰另一中心171名儿童(1871个观测)的TDM数据集上以预测误差进行外部评估。
  • 对popPK与机器学习方法均评估了有限采样方案。

主要发现

  • 一室模型为最佳拟合,第2天和第3天清除率较第1天分别下降10.4%和13%。
  • 外部数据集中popPK、随机森林、XGBoost的中位PE为-4.82%至-7.44%,中位绝对PE约15%。
  • 约80%的数据预测误差在±30%以内。

关键数值与推荐

  • popPK模型下输注后0、0.5、1、4小时或0、1、2、4小时采样可充分预测白消安血浆浓度与暴露。

临床怎么用

  • 儿童白消安TDM可考虑采用popPK模型结合有限采样方案以减少采血次数。
  • 机器学习模型可用于浓度预测,但完整药代动力学描述仍需更大数据集支持。

局限性

  • 机器学习方法需要更大数据集才能描述完整药代动力学特征。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

原文

BACKGROUND AND OBJECTIVES: Busulfan, used in conditioning regimens prior to haematopoietic stem cell transplantation, has a narrow therapeutic index; therefore, due to pharmacokinetic variability, therapeutic drug monitoring (TDM) is employed to ensure drug exposure levels associated with optimal clinical outcomes. This study aimed to facilitate dose adjustment and potentially reduce the number of samples required for TDM in clinical practice. METHODS: Population pharmacokinetic (popPK) and machine learning (ML) models (Random Forest and XGBoost) were developed using a local, retrospective, extensive-sampling dataset of 67 children (637 observations, Denmark) and externally evaluated using prediction error (PE) on a TDM dataset from another centre with 171 children (1871 observations, Netherlands). Limited sampling schedules were evaluated for both methods. RESULTS: A one-compartment model was the best data fit, showing time-varying clearance decreases of 10.4% and 13% on Days 2 and 3, respectively, compared to Day 1. The popPK, Random Forest, and XGBoost models performed well on the external dataset for concentration prediction, with median PE ranging from -4.82 to -7.44%, median absolute PE of approximately 15%, and approximately 80% of data predicted within ±30% of the PE. Sampling strategies at 0, 0.5, 1, 4 h or 0, 1, 2, 4 h post-infusion adequately predicted busulfan plasma concentration and exposure using the popPK model. CONCLUSIONS: A new popPK model was developed to support individualised dose adjustments in children using fewer blood samples. Both Random Forest and XGBoost could predict busulfan's plasma concentration with performance like that of the popPK model (population prediction). However, a larger dataset is needed for ML methods to describe a complete pharmacokinetic profile.

来源:Clin Pharmacokinet,查看原文。版权归原作者所有。

引用

  1. Yang T, Thastrup M, Masmas TN, et al. Comparing the Predictive Performance of Machine Learning Models with a Newly Developed Population Pharmacokinetic Model for Busulfan in Paediatric Patients. Clin Pharmacokinet. 2026. PMID: 42773369. doi:10.1007/s40262-026-01702-9

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。