Ignacio Hernández González
Isotopes Centre, San José de las Lajas, Mayabeque, Cuba
Objectives: A nonlinear mixed effect model is applied to solve a very sparse PK design where gammagraphic imaging was used to quantify drug uptake in specific compartments.
Methods: Data were drawn from real experiments using blood sampling and imaging procedures to build a compartmental model of tumor uptake of radiolabeled drug. Based on practical consideration, two situations of incomplete individual profiles were evaluated: overlapping and no overlapping time point distribution. In all cases five blood samples were considered per time point. Serial tumor uptake was simulated as obtained from gammagraphic images of radiolabeled drugs. All pharmacokinetic profiles were analyzed by means of nonlinear mixed effect models using the MONOLIX software [1] version 4.2 (Lixoft, France).
Results: All sampling schedules yield the same results when computed using the MONOLIX software and the SAEM algorithm. Population and individual pharmacokinetic parameters were accurately estimated with three or five determination per sampling point. According with the used methodology and software tool, it can be an expected result, but demonstrating the method performance in such situations, allow us to select a more flexible design using a very small number of animals in preclinical research. It is particularly important because imaging procedures take several minutes and a non-restricted experimental design helps in planning and executing experiments. For example, peptides with a very fast blood clearance from blood need flexible enough design to have a good concentration profile in a very short time, together with a probable long lasting tumor uptake. In radiotherapy, is crucial in the estimation of dose delivered to target and non-target tissues. The combination with imaging procedures allows us to construct a completely structured compartmental analysis. So, compartments are no longer hypothetical or invisible.
Conclusions: Imaging radiolabeled drugs can help to build compartmental models based on quantification of real tissue uptake rather than defining hypothetical compartments. Nonlinear mixed effect models can be used with a very sparse data of blood samples and tissue uptake values, contributing to the reduction of animals used in biomedical research. With the introduction of molecular imaging in drug development and research [2] we can be able to introduce the molecular based pharmacokinetic models.
References:
[1] Samson, A., Lavielle, M., and Mentré, F. (2007). The SAEM algorithm for group comparison tests in longitudinal data analysis based on non-linear mixed-effects model. Statistics in Medicine, 26: 4860–75.
[2] Rudin M, Weissleder R. (2003) Molecular imaging in drug discovery and development. Nature Review, 2: 123-131.
Reference: PAGE 23 (2014) Abstr 3311 [www.page-meeting.org/?abstract=3311]
Poster: Methodology - Other topics