JPKD系统预测紫癜性肾炎患儿他克莫司血药浓度的适用性研究

Applicability of the JPKD system for predicting tacrolimus blood concentrations in children with Henoch-Schonlein purpura nephritis

关键词:  他克莫司  紫癜性肾炎  JPKD  血药浓度  预测能力

Key words:  tacrolimus  henoch-schonlein purpura nephritis  JPKD  blood concentration  predictive ability

摘要:

目的 探讨JPKD群体药代动力学软件对紫癜性肾炎患儿他克莫司血药浓度的预测能力及影响因素分析。方法:以2016年6月-2023年6月就诊于昆明儿童医院使用他克莫司治疗并定期进行谷浓度监测的紫癜性肾炎患者为研究对象进行回顾性研究。使用JPKD预测他克莫司剂量调整后的理论血药浓度,Pearson分析结果的相关性；权重偏差(WRES)评估评估JPKD 系统的预测能力。采用单因素和多因素Logistic回归分析筛选JPKD系统预测准确性的影响因素。受试者工作特征(ROC)曲线评价影响因素对软件预测准确性的判断价值。结果:共纳入43例HSPN患儿132例次血药浓度监测。他克莫司调整剂量后血药浓度的实测值为4.31±1.86 (1.12~10.55) ng·mL-1,JPKD系统预测的血药浓度为4.32±2.26 (1.39~9.86) ng·mL-1。平均绝对权重偏差和相对预测误差分别为39.03%和7.64%。Pearson分析表明预测值与实测值具有相关性,预测的准确率为51.81%。单因素分析显示年龄、体重、身高、剂量、AST、Scr、PLT和合并使用五酯胶囊影响JPKD预测准确性。进一步Logistic回归分析显示,仅AST影响JPKD预测准确性。ROC曲线显示患儿ASL值对JPKD预测准确性AUC(95%CI)=0.751(0.644～0.857,P＜0.01),对应的截止值为26.0 U·L-1。结论:JPKD对紫癜性肾炎患儿他克莫司血药浓度具有一定的预测能力,但需结合肝功能制定个体化方案。

Abstract:

ABSTRACT: OBJECTIVE To investigate the predictive ability of JPKD population pharmacokinetic software for tacrolimus blood concentrations in children with Henoch-Schonlein purpura nephritis and to analyze the factors influencing it. METHODS: A retrospective study was conducted with patients with purpura nephritis who attended Kunming Children's Hospital for treatment with tacrolimus from June 2016 to June 2023 and were regularly monitored for trough concentrations. JPKD was used to predict theoretical tacrolimus dose-adjusted blood concentrations, and Pearson analysis was used to analyze the correlation of the results; weighted deviation (WRES) assessment was used to evaluate the predictive ability of JPKD. Univariate and multivariate logistic regression analyses were used to screen for factors influencing the predictive accuracy of the JPKD. Subject work characteristics (ROC) curves were evaluated to assess the judgmental value of the influencing factors on the predictive accuracy of the software. RESULTS: A total of 43 children with HSPN were included in 132 blood concentration monitoring sessions. The measured tacrolimus dose-adjusted blood concentration was 4.31±1.86 (1.12~10.55) ng·mL-1, and the predicted blood concentration by JPKD was 4.32±2.26 (1.39~9.86) ng·mL-1. The mean absolute weighting bias and relative prediction error were 39.03% and 7.64%, respectively.Pearson analysis showed that the predicted values were correlated with the measured values, and the accuracy of prediction was 51.81%. Univariate analysis showed that age, weight, height, dose, AST, Scr, PLT and combined use of pentothal capsules affected the predictive accuracy of JPKD. Further logistic regression analysis showed that only AST affected JPKD prediction accuracy.ROC curve showed that children's ASL values had a positive effect on JPKD prediction accuracy AUC (95% CI) = 0.751 (0.644 to 0.857, P < 0.01), corresponding to a cut-off value of 26.0 U·L-1.CONCLUSION: The JPKD has a positive effect on the predictive accuracy of tacrolimus in children with purpura nephritis. blood concentration has some predictive ability, but an individualized regimen in combination with liver function is needed.