基于有限采样法的骨肉瘤患者甲氨蝶呤体内暴露量估算模型的建立与验证

目的 采用有限采样法估算骨肉瘤患者静脉滴注大剂量甲氨蝶呤(high dose methotrexate, HD-MTX)后的体内药物暴露量,即药时曲线下面积(area under the drug concentration versus time curve, AUC_0-t)。方法 34名骨肉瘤患者共进行67个周期HD-MTX化疗,分别于滴注结束时、开始滴注10,24,48和72 h收集患者血液样本,采用化学发光免疫法(CMIA)进行MTX血药浓度测定。计算得到患者MTX临床药动学参数后,经多元线性回归,建立基于有限采样法的AUC_0-t估算模型。采用Bootstrap法和Jackknife法进行模型准确性验证,采用Bland-Altman分析法评估有限采样法与经典药动学方法之间的一致性。通过绘制ROC曲线,评估AUC_0-t对HD-MTX不良反应的预测性能。结果 综合考虑临床可操作性及准确性,选择两点(c_max、c_{24 h})模型用于MTX AUC_0-t的估算,回归方程为AUC_0-t=6.36c_max+161.30c_{24 h},r²为0.933,平均预测误差为0.47%,均方根误差为6.78%。AUC_0-t对肝功能损伤、骨髓抑制及消化道反应等药物不良反应具有良好的预测能力,P<0.05。结论 基于c_max、c_{24 h}两点估算的有限采样法可以用于骨肉瘤患者HD-MTX药物暴露量的估算,方法简便,易于临床开展,可为MTX个体化用药提供参考。

Establishment and Validation of a Limited Sampling Strategy to Estimate Methotrexate Exposure in Patient with Osteosarcoma

OBJECTIVE To establish limited sampling strategy (LSS) to estimate area under the drug concentration versus time curve (AUC_0-t) of osteosarcoma patients who had high dose methotrexate (HD-MTX) intravenous infusion. METHODS Thirty-four osteosarcoma patient had 67 cycle of HD-MTX chemotherapy. Blood samples were collected at the end of the methotrexate infusion and 10, 24, 48, 72 h after the start of drug administration. Chemiluminescence microparticle immunoassay was used to determine the MTX serum concentration. Multivariate regression analysis was conducted to establish models for the estimation of methotrexate exposure. The model was further verified by Bootstrap and Jackknife method. Consistency of the AUC_0-t results were compared between classic pharmacokinetics calculation and proposed limited sampling strategy estimation. Receiver operating characteristic (ROC) curves were drawn to evaluate the predictive capacity of AUC_0-t to MTX adverse effects. RESULTS Taking clinical operation and method accuracy into consideration, two sampling time point model (c_max, c_{24 h}) was selected, the regression equation was AUC_0-t=6.36c_max+161.30c_{24 h}, r²=0.933, MPE=0.47%, RMSE=6.78%. AUC_0-t could be a good predictor of severe adverse events during HD-MTX chemotherapy, P<0.05. CONCLUSION The established limited sampling strategy to estimate the exposure based on c_max and c_{24 h} could be used to MTX individualized therapy in clinical practice.