Dokoumetzidis Aristides, Macheras Panos
Queen's University of Belfast, Medical Biology Centre, UK.
J Pharmacokinet Pharmacodyn. 2009 Apr;36(2):165-78. doi: 10.1007/s10928-009-9116-x. Epub 2009 Apr 2.
We explore the use of fractional order differential equations for the analysis of datasets of various drug processes that present anomalous kinetics, i.e. kinetics that are non-exponential and are typically described by power-laws. A fractional differential equation corresponds to a differential equation with a derivative of fractional order. The fractional equivalents of the "zero-" and "first-order" processes are derived. The fractional zero-order process is a power-law while the fractional first-order process is a Mittag-Leffler function. The latter behaves as a stretched exponential for early times and as a power-law for later times. Applications of these two basic results for drug dissolution/release and drug disposition are presented. The fractional model of dissolution is fitted successfully to datasets taken from literature of in vivo dissolution curves. Also, the proposed pharmacokinetic model is fitted to a dataset which exhibits power-law terminal phase. The Mittag-Leffler function describes well the data for small and large time scales and presents an advantage over empirical power-laws which go to infinity as time approaches zero. The proposed approach is compared conceptually with fractal kinetics, an alternative approach to describe datasets with non exponential kinetics. Fractional kinetics offers an elegant description of anomalous kinetics, with a valid scientific basis, since it has already been applied in problems of diffusion in other fields, and describes well the data.
我们探讨使用分数阶微分方程来分析呈现反常动力学的各种药物过程的数据集,即非指数动力学且通常由幂律描述的动力学。分数阶微分方程对应于具有分数阶导数的微分方程。推导了“零阶”和“一阶”过程的分数阶等效形式。分数零阶过程是幂律,而分数一阶过程是米塔格 - 莱夫勒函数。后者在早期表现为拉伸指数,在后期表现为幂律。展示了这两个基本结果在药物溶解/释放和药物处置方面的应用。将溶解的分数阶模型成功拟合到从体内溶解曲线文献中获取的数据集。此外,将所提出的药代动力学模型拟合到一个呈现幂律终末相的数据集。米塔格 - 莱夫勒函数能很好地描述小时间尺度和大时间尺度的数据,并且相对于在时间趋近于零时趋于无穷大的经验幂律具有优势。从概念上将所提出的方法与分形动力学进行比较,分形动力学是描述非指数动力学数据集的另一种方法。分数阶动力学为反常动力学提供了一种优雅的描述,具有有效的科学依据,因为它已被应用于其他领域的扩散问题,并且能很好地描述数据。