RADT-14. TOWARDS IMAGE-GUIDED MODELING OF PATIENT-SPECIFIC RHENIUM-186 NANOLIPOSOME DISTRIBUTION VIA CONVECTION-ENHANCED DELIVERY FOR GLIOBLASTOMA MULTIFORME

Convection-enhanced delivery (CED) of Rhenium-186 nanoliposomes (RNL) is a promising approach to provide precise delivery of large, localized doses of radiation with the goal of extending overall survival for patients with recurrent GBM. A central component of successful CED, is achieving optimal catheter placement for delivery of the therapy. While surgical planning software exists for this purpose, current approaches are designed for small molecules and therefore are not appropriate for larger particles like RNL. To address this concern, we have developed a mathematical model to predict the distribution of RNL via CED on a patient-specific basis. The model is defined on the 3D brain domain which consists of 1) pressure and flow fields generated by accounting for catheter infusion, flow through brain, and fluid loss into capillaries, and 2) the transport of RNL governed by an advection-diffusion equation. We utilize pre-operative MRI to assign patient-specific tissue geometry and properties (e.g., diffusivity, conductivity), and calibrate the model with SPECT measurements within 24 h post the RNL delivery. This model is implemented on one patient enrolled in NCT01906385. The accuracy of model calibration and prediction is evaluated by the Dice score and concordance correlation coefficient (CCC) between modeled and measured distributions of RNL. Our model calibration achieves Dice scores of 0.80, 0.81, 0.69 and CCC of 0.92, 0.93, 0.73 for RNL distributions at the mid-delivery, end of delivery, and 24 h after the delivery, respectively. Long-term model prediction achieves Dice scores of 0.69 and 0.52 at 144 h and 196 h after the delivery, respectively, and CCC of 0.57 and 0.31. Preliminary results demonstrate a proof-of-concept for a patient-specific model to predict the spatiotemporally-resolved distribution of nanoparticles. Ongoing efforts focus on improving our model by accounting backflow and angle of catheter placement, and applying to more patients. Funding: NIH R01CA235800, CPRIT RR160005.

Correct regionalisation of a tissue primordium is essential for coordinated morphogenesis

During organ development, tubular organs often form from flat epithelial primordia. In the placodes of the forming tubes of the salivary glands in the Drosophila embryo, we previously identified spatially defined cell behaviours of cell wedging, tilting and cell intercalation that are key to the initial stages of tube formation. Here we address what the requirements are that ensure the continuous formation of a narrow symmetrical tube from an initially asymmetrical primordium whilst overall tissue geometry is constantly changing. We are using live-imaging and quantitative methods to compare wild-type placodes and mutants that either show disrupted cell behaviours or an initial symmetrical placode organisation, with both resulting in severe impairment of the invagination. We find that early transcriptional patterning of key morphogenetic transcription factors drives the selective activation of downstream morphogenetic modules, such as GPCR signalling that activates apical-medial actomyosin activity to drive cell wedging at the future asymmetrically-placed invagination point. Over time, transcription of key factors expands across the rest of the placode and cells switch their behaviour from predominantly intercalating to predominantly apically constricting as their position approaches the invagination pit. Misplacement or enlargement of the initial invagination pit leads to early problems in cell behaviours that eventually result in a defective organ shape. Our work illustrates that the dynamic patterning of the expression of transcription factors and downstream morphogenetic effectors ensures positionally fixed areas of cell behaviour with regards to the invagination point. This patterning in combination with the asymmetric geometrical set-up ensures functional organ formation.

Buffering by Transporters Can Spare Geometric Hindrance in Controlling Glutamate Escape

The surface of astrocyte processes that often surround excitatory synapses is packed with high-affinity glutamate transporters, largely preventing extrasynaptic glutamate escape. The shape and prevalence of perisynaptic astroglia vary among brain regions, in some cases providing a complete isolation of synaptic connections from the surrounding tissue. The perception has been that the geometry of perisynaptic environment is therefore essential to preventing extrasynaptic glutamate escape. To understand to what degree this notion holds, we modelled brain neuropil as a space filled with a scatter of randomly sized, overlapping spheres representing randomly shaped cellular elements and intercellular lumen. Simulating release and diffusion of glutamate molecules inside the interstitial gaps in this medium showed that high-affinity transporters would efficiently constrain extrasynaptic spread of glutamate even when diffusion passages are relatively open. We thus estimate that, in the hippocampal or cerebellar neuropil, the bulk of glutamate released by a synaptic vesicle is rapidly bound by transporters (or high-affinity target receptors) mainly in close proximity of the synaptic cleft, whether or not certain physiological or pathological events change local tissue geometry.

Hindbrain neuropore tissue geometry determines asymmetric cell-mediated closure dynamics in mouse embryos

Gap closure is a common morphogenetic process. In mammals, failure to close the embryonic hindbrain neuropore (HNP) gap causes fatal anencephaly. We observed that surface ectoderm cells surrounding the mouse HNP assemble high-tension actomyosin purse strings at their leading edge and establish the initial contacts across the embryonic midline. Fibronectin and laminin are present, and tensin 1 accumulates in focal adhesion-like puncta at this leading edge. The HNP gap closes asymmetrically, faster from its rostral than caudal end, while maintaining an elongated aspect ratio. Cell-based physical modeling identifies two closure mechanisms sufficient to account for tissue-level HNP closure dynamics: purse-string contraction and directional cell motion implemented through active crawling. Combining both closure mechanisms hastens gap closure and produces a constant rate of gap shortening. Purse-string contraction reduces, whereas crawling increases gap aspect ratio, and their combination maintains it. Closure rate asymmetry can be explained by asymmetric embryo tissue geometry, namely a narrower rostral gap apex, whereas biomechanical tension inferred from laser ablation is equivalent at the gaps’ rostral and caudal closure points. At the cellular level, the physical model predicts rearrangements of cells at the HNP rostral and caudal extremes as the gap shortens. These behaviors are reproducibly live imaged in mouse embryos. Thus, mammalian embryos coordinate cellular- and tissue-level mechanics to achieve this critical gap closure event.

A Machine Learning Workflow for Tumour Detection in Breasts Using 3D Microwave Imaging

A two-stage workflow for detecting and monitoring tumors in the human breast with an inverse scattering-based technique is presented. Stage 1 involves a phaseless bulk-parameter inference neural network that recovers the geometry and permittivity of the breast fibroglandular region. The bulk parameters are used for calibration and as prior information for Stage 2, a full phase contrast source inversion of the measurement data, to detect regions of high relative complex-valued permittivity in the breast based on an assumed known overall tissue geometry. We demonstrate the ability of the workflow to recover the geometry and bulk permittivity of the different sized fibroglandular regions, and to detect and localize tumors of various sizes and locations within the breast model. Preliminary results show promise for a synthetically trained Stage 1 network to be applied to experimental data and provide quality prior information in practical imaging situations.

Hindbrain neuropore tissue geometry determines asymmetric cell-mediated closure dynamics

Gap closure is a common morphogenetic process. In mammals, failure to close the embryonic hindbrain neuropore (HNP) gap causes fatal anencephaly. We observed that surface ectoderm cells surrounding the mouse HNP assemble high-tension actomyosin purse-strings at their leading edge and establish the initial contacts across the embryonic midline. The HNP gap closes asymmetrically, faster from its rostral than caudal extreme, while maintaining an elongated aspect ratio. Cell-based physical modelling identifies two closure mechanisms sufficient to describe tissue-level HNP closure dynamics; purse-string contraction and directional cell crawling. Combining both closure mechanisms hastens gap closure and produces a constant rate of gap shortening. Purse-string contraction reduces, whereas crawling increases gap aspect ratio, and their combination maintains it. Closure rate asymmetry can be explained by embryo tissue geometry, namely a narrower rostral gap apex. At the cellular level, our model predicts highly directional cell migration with a constant rate of cells leaving the HNP rim. These behaviours are reproducibly live-imaged in mouse embryos. Thus, mammalian embryos coordinate cellular and tissue-level mechanics to achieve this critical gap closure event.

Microtissue geometry and cell-generated forces drive patterning of liver progenitor cell differentiation in 3D

Investigating the role of mechanical signaling on stem and progenitor cell differentiation in three-dimensional (3D) microenvironments is key to fully understanding these processes. Towards this, we implemented a hydrogel microwell based method to produce arrays of multicellular microtissues in constrained geometries, which cause distinct profiles of mechanical signals in 3D. We applied this platform to a model liver development system to investigate the impact of tissue geometry and mechanical stress on liver progenitor cell bipotential differentiation into hepatocyte-like and biliary-like cells. We fabricated 3D liver progenitor cell microtissues of varied geometries, including cylinder and toroid, and used image segmentation on confocal images to track individual single cell phenotype within defined spatial coordinates. These studies demonstrated patterning of hepatocytic differentiation to the outer shell of the cylinder and toroid microtissues, except at the inner diameter surface of the toroid tissues. Biliary differentiation was distributed throughout the microtissue interior regions and was additionally increased in toroid tissues compared to cylinder tissues. We used finite element modeling to predict stress distributions in these microtissues which demonstrated that cell-cell tension correlated with hepatocytic fate, while compression correlated with decreased hepatocytic differentiation and increased biliary differentiation. Overall, this combined approach that integrates microscale fabrication, imaging and analysis, and mechanical modeling serve as a demonstration of how microtissue geometry can drive patterning of mechanical stresses that regulate cell differentiation trajectories. It also can serve as a platform for the further investigation of tissue morphogenetic signaling mechanisms in the liver as well as other stem cell differentiation contexts.

Functional Grading of a Transversely Isotropic Hyperelastic Model with Applications in Modeling Tricuspid and Mitral Valve Transition Regions

Surgical simulators and injury-prediction human models require a combination of representative tissue geometry and accurate tissue material properties to predict realistic tool–tissue interaction forces and injury mechanisms, respectively. While biological tissues have been individually characterized, the transition regions between tissues have received limited research attention, potentially resulting in inaccuracies within simulations. In this work, an approach to characterize the transition regions in transversely isotropic (TI) soft tissues using functionally graded material (FGM) modeling is presented. The effect of nonlinearities and multi-regime nature of the TI model on the functional grading process is discussed. The proposed approach has been implemented to characterize the transition regions in the leaflet (LL), chordae tendinae (CT) and the papillary muscle (PM) of porcine tricuspid valve (TV) and mitral valve (MV). The FGM model is informed using high resolution morphological measurements of the collagen fiber orientation and tissue composition in the transition regions, and deformation characteristics predicted by the FGM model are numerically validated to experimental data using X-ray diffraction imaging. The results indicate feasibility of using the FGM approach in modeling soft-tissue transitions and has implications in improving physical representation of tissue deformation throughout the body using a scalable version of the proposed approach.