(T1330-11-64) Physiologically Based Pharmacokinetic Model for Miltefosine in Mice

Purpose: Visceral leishmaniasis is a devastating infectious disease caused by protozoan parasites in the genus of Leishmania and it is highly fatal if left untreated. Miltefosine, a hexadecyl phosphocholine drug, is the only FDA approved oral drug for the treatment of leishmaniasis. The unusual properties of miltefosine, such as very high protein binding and extensive tissue distribution, causes prolonged elimination half-life in humans and preclinical animal models. Leishmaniasis is characterized by hepatomegaly and splenomegaly which results in enlargement of liver and spleen. It has been previously reported that albumin concentration decreased in hepatosplenic patients¹. Leishmanial infection also caused variations in tissue distribution for example, Amphotericin B². Therefore, it may be assumed that absorption and distribution of miltefosine varies during leishmania infection.

Methods: Physiologically based pharmacokinetic (PBPK) modeling is a computer simulation approach that uses physiological and drug-related properties to study the pharmacokinetics of the drugs. In this study, a whole-body oral PBPK model was initially built for healthy mice using the MATLAB SimBiology software. The mouse physiological parameters were obtained from literature and drug related parameters like plasma protein binding, blood to plasma partitioning ratio, tissue partition coefficient and drug metabolism were either determined by in vitro studies or extracted from the literature. The organ compartments in the PBPK model were connected via the ordinary differential equations (ODE), which describe the time dependent changes in drug concentrations. Initial studies supported PBPK models using permeability limited absorption mechanism in the kidney and remainder organ compartments.

Results: The pharmacokinetic (PK) parameters estimated were absorption rate constant, elimination rate constant, clearance, and permeability limited distribution rate constants. These values are provided along with the standard errors in the Table 1. Some of the parameter standard error values are higher and utilization of experimental values in the future might reduce the standard error values. The local sensitivities of these parameters were calculated, and small intestine absorption rate constant was found to be the most sensitive parameter and stomach absorption and metabolism rate constants were found to be the least sensitive parameters to the venous drug concentrations. The predicted plasma time profile for healthy mice were within 2-fold of the observed profile, demonstrating its validity (Fig 1). Furthermore, the healthy mice PBPK model predictions were compared with observed plasma concentration time profiles of leishmania infected mice³. It was observed that only 66% of the healthy model predicted data fell within 2-fold error of the leishmania infected mice observed data (Fig 2a). On modifying some absorption and distribution related model parameters, like small intestine absorption rate constant, intestinal transit rate, kidney permeability constant and remainder permeability constant, an infected mice model was generated. The predictions from this model were able to match the observed plasma concentration time profile curves (Fig 2b).

Conclusion: The reduction in albumin content caused by splenomegaly led to an increase in miltefosine oral absorption rate from the intestine. The decrease in distribution rate of miltefosine as observed for amphotericin B, could also have been due to invasion of parasites into macrophages in leishmaniasis conditions. Therefore, these observations clearly indicate the changes in PK of miltefosine in diseased conditions.

References: 1. Leite, L. A. C.; Domingues, A. L. C.; Lopes, E. P.; Ferreira, R. d. C. d. S.; Pimenta Filho, A. d. A.; Fonseca, C. S. M. d.; Santos, B. S. d.; Lima, V. L. d. M. Relationship between splenomegaly and hematologic findings in patients with hepatosplenic schistosomiasis. Revista brasileira de hematologia e hemoterapia 2013, 35, 332-336.
2. Kagan, L.; Gershkovich, P.; Wasan, K. M.; Mager, D. E. Physiologically based pharmacokinetic model of amphotericin B disposition in rats following administration of deoxycholate formulation (Fungizone(R)): pooled analysis of published data. AAPS J 2011, 13 (2), 255-264. DOI: 10.1208/s12248-011-9267-8.
3. Voak, A. A.; Harris, A.; Coteron-Lopez, J. M.; Angulo-Barturen, I.; Ferrer-Bazaga, S.; Croft, S. L.; Seifert, K. Pharmacokinetic/pharmacodynamic relationships of liposomal amphotericin B and miltefosine in experimental visceral leishmaniasis. PLoS neglected tropical diseases 2021, 15 (3), e0009013.


Estimated PBPK model parameters for healthy mice


Comparison of plasma concentration time profile of model predicted data versus observed data from healthy mice


Infected mice plasma concentration time profiles before and after parameter modification