OBJECTIVE: The purpose of this paper is to establish population pharmacokinetics model of intravenous infusing Busulfan HSCT patients by NONMEM, to evaluate intra-individual and inter-individual variable, and explore CL and Vd influenced by physiological and pathological factors.METHODS:1)High performance liquid chromatography method was established to determine busulfan concentration in human plasma.2)Concentrations of 17 of intensive sampling IV Bu patients was determined. All possible 2ã€3and4 concentration points combinations was evaluated by stepwise linear regression to evaluate AUC. Sampling blood according to the time point of good correlation combilations.3)35 of HSCT IV Bu patients was investigated. Physiological and pathological factors such as ageã€sexã€heightã€weightã€doseã€liver and renal functions was collected. Evaluated them by NONMEM through establishing population pharmacokinetics model and verifies the final model by Bootstrap.RESULTS: 1) Chromatographic conditions are as follow, the mobile phase:methanol :water=75:25,flow rate:1.0ml/min, wave length:280nm.The linear regression function is Y=0.2226X-0.0224,r=0.9998,with recovery of(83.37±1.94)%,and the RSD of intra-day and inter-day are less than 15%. 2) The linearity relationship of the model established by 2 and 3 drug concentration points is good. The best regression models of the intravenous infusion first dosage are AUC0-t=97.823+178.62C240+170.961C300ã€AUC0-t=-477.93+142.088C180+814.64C240-379.429C360ã€AUC0-t=-178.381+166.86C120+234.881C150-229.209C180+238.58C240;models of the ninth dosage are AUC0-t=-2.939+109.353C120+217.992C240ã€AUC0-t=-93.301+211.974C15+65.11C135+114.243C180ã€AUC0-t=-32.586-19.652C120+101.04C360+294.327C300+265.361C240. 3)The final model function is CL=θ1×exp(θ3×WT)×exp(θ4×YFJB)×exp(η1);Vd=θ2×exp(θ5×WT)×exp(θ6×YFJB)×exp(η2)。CONCLUSION:In this work, analysis method was established for detecting concentration of busulfan in human plasma. Limited sampling strategy AUC functions was evaluated. Population pharmacokinetics model was established by NONMEM and model verification was investigated by Bootstrap. The final model is precise and stable and can provide the foundation of personalized medicine. |