Cochlear pharmacokinetics - Micro-computed tomography and learning-prediction modeling for transport parameter determination

• Published in: Hearing research 2019-09, Vol.380, p.46-59
• Main Authors: Moudgalya, Sanketh S., Wilson, Kevin, Zhu, Xiaoxia, Budzevich, Mikalai M., Walton, Joseph P., Cahill, Nathan D., Frisina, Robert D., Borkholder, David A.
• Format: Article
• Language: English
• Subjects: 3-D image registration; Animals; Cochlea; Cochlea - diagnostic imaging; Cochlea - metabolism; Computer Simulation; Contrast agent; Contrast Media - administration & dosage; Contrast Media - pharmacokinetics; Diffusion; Infusions, Parenteral; Inner ear; Iopamidol - administration & dosage; Iopamidol - pharmacokinetics; Mice, Inbred CBA; Micro-computed tomography; Models, Biological; Mouse; Optimization; Pharmacokinetics; Segmentation; Tissue Distribution; X-Ray Microtomography
• ISSN: 0378-5955; 1878-5891
• DOI: 10.1016/j.heares.2019.05.009

Inner ear disorders such as sensorineural deafness and genetic diseases may one day be treated with local drug delivery to the inner ear. Current pharmacokinetic models have been based on invasive methods to measure drug concentrations, limiting them in spatial resolution, and restricting the research to larger rodents. We developed an intracochlear pharmacokinetic model based on an imaging, learning-prediction (LP) paradigm for learning transport parameters in the murine cochlea. This was achieved using noninvasive micro-computed tomography imaging of the cochlea during in vivo infusion of a contrast agent at the basal end of scala tympani through a cochleostomy. Each scan was registered in 3-D to a cochlear atlas to segment the cochlear regions with high accuracy, enabling concentrations to be extracted along the length of each scala. These spatio-temporal concentration profiles were used to learn a concentration dependent diffusion coefficient, and transport parameters between the major scalae and to clearance. The LP model results are comparable to the current state of the art model, and can simulate concentrations for cases involving different infusion molecules and different drug delivery protocols. Forward simulation results with pulsatile delivery suggest the pharmacokinetic model can be used to optimize drug delivery protocols to reduce total drug delivered and the potential for toxic side effects. While developed in the challenging murine cochlea, the processes are scalable to larger animals and different drug infusion paradigms. [Display omitted] •Micro-Computed Tomography (μCT): cochlear imaging during in vivo infusion of ISOVUE at the ST base through a cochleostomy.•3-D Registration: Cochlear atlas geometrically transformed to the μCT images with transfer of atlas labels to the scans.•Spatio-temporal concentration extraction from the μCT images for scala tympani, scala vestibuli, scala media.•Learning-Prediction model: Learning concentration dependent diffusion coefficient.•Learning-Prediction model: Learning transport parameters between major scalae, and to clearance.•Forward simulation of cochlear pharmacokinetics using learned parameters.