In theory, siRNAs can inhibit every known cancer-causing gene through sequence-specific RNA interference. However, almost twenty years after the discovery of RNA interference, the use of siRNAs as targeted molecular medicines remains challenging due to comprehensively poor pharmacokinetic properties of siRNA. Naked siRNA molecules are rapidly excreted through the urine and cannot inherently enter cells or access the cytosol through endosomal escape, resulting in limited bioavailability within tumor cells after systemic administration. Strategies to complex negatively-charged siRNA into cationic polyion complexes (polyplexes) have been effective for the treatment of diseases in the liver where polyplexes naturally biodistribute. But the same polyplexes have shown limited success in oncology due to rapid disassembly within the kidneys, off-target accumulation within the liver, and limited on-target accumulation within tumor tissue. Thus, a broader set of polyplex physicochemical parameters remain to be optimized in order to improve siRNA delivery to tumors after systemic administration. Here, we show that fine-tuning hydrophobic stabilizing forces of siRNA polyplexes, through altering either the polymer carrier or siRNA molecule, can simply and effectively improve siRNA bioavailability and accumulation within solid breast tumors. In both cases, increased polyplex stability through the optimization of core hydrophobicity decreased rapid renal clearance and led to appreciable increases in blood circulation, tumor accumulation, and intratumoral siRNA bioactivity. Our updated siRNA polyplex technology enabled the first selective, therapeutic silencing of mTORC2 in HER2-amplified breast tumors. Due to the prominent role of mTORC2 within the oncogenic PI3K-Akt-mTOR pathway, selective mTORC2 inhibition slowed tumor growth through the induction of cell death and cooperated with the HER2 receptor tyrosine kinase inhibitor, lapatinib, to kill HER2-amplified tumor cells and halt tumor growth. In sum, this work systematically elucidates the impact of core hydrophobicity on siRNA polyplex performance in vivo, illustrates the broad potential for therapeutically inhibiting currently âundruggableâ cancer-causing oncogenes, and highlights the specific therapeutic potential of selectively inhibiting mTORC2 as a tumor cell killing strategy in HER2-amplified breast cancers.

Osteoarthritis (OA) is a disease characterized by degradation of joints with the development of painful inflammation in the surrounding tissues. Post-traumatic osteoarthritis (PTOA) is OA that develops following a traumatic injury to the joint. Currently, there are a limited number of treatments for this disease and many of these only provide temporary, palliative relief. Here, we discuss polymer drug delivery systems that can provide targeted and sustained delivery of imaging and therapeutic agents to OA-affected sites. Polymer based microparticles were investigated as a therapeutic for PTOA. The inherent antioxidant function of poly(propylene sulfide) (PPS) microspheres (MS) was dissected for different reactive oxygen species (ROS), and therapeutic benefits of PPS-MS were explored in mechanically induced PTOA. PPS-MS scavenged hydrogen peroxide (H2O2), hypochlorite, and peroxynitrite but not superoxide in vitro in cell-free and cell based assays. Elevated ROS levels were confirmed in a mouse model of PTOA. In the PTOA model, PPS-MS reduced matrix metalloproteinase activity. These results suggest that local delivery of PPS-MS to site of PTOA reduces articular cartilage destruction. These results motivate further exploration of PPS as a stand-alone, locally-sustained antioxidant therapy and as a material for microsphere-based, sustained local drug delivery to inflamed tissues at risk of ROS damage.

Duchenne muscular dystrophy (DMD) is a progressive myopathy caused by mutations in the dystrophin gene, leading to contraction-induced damage, inflammation, and necrosis in skeletal and cardiac muscles. Reliable methods of characterizing DMD cardiomyopathy are essential for effective pharmacological therapy. Myocardial circumferential strain (εcc) measured via harmonic phase (HARP) analysis is commonly used to measure cardiomyopathy. Heart rate variability (HRV) data can be used to quantify autonomic compensation in diseased patients, potentially providing an additional method of disease characterization. In this retrospective investigation, we hypothesized that 1) our custom HARP algorithm would be correlated with equivalent results from standard clinical software, and 2) that the εcc results would be correlated with HRV parameters. Twenty-eight boys with DMD were studied (ages 8-21). Cardiac MRI data included spatial modulation of magnetization (SPAMM)-tagged images, acquired throughout the cardiac cycle. HARP analysis was used to calculate peak Lagrangian εcc. Forty-eight-hour Holter monitoring data were acquired. Parasympathetic input-associate power (PIAP) was determined by spectral analysis of R-R intervals observed during sleep to determine the proportion of total power in the high-frequency band. The εcc results from our HARP algorithm were correlated with clinically-used values (r=0.79, p<1.0x10-7). PIAP measurements from two randomly sampled periods did not differ significantly (p=0.51) and had high reliability (intraclass correlation=0.923, p<0.001; n=28). However, PIAP did not prove to be a significant measure of disease characterization. Though promising patterns exist, key study limitations must be addressed in order to conclude that HRV parameters can provide an alternate method of assessing DMD.

Small interfering RNA (siRNA) can potently and specifically suppress translation of any gene, including intracellular targets traditionally considered âundruggableâ. However, emergence of translational siRNA therapies has remained slow, with the primary challenge being the formidable anatomical and physiological barriers that must be overcome to deliver siRNA to its intracellular site of action in target cell types. Polymeric nanoparticle (NP) carriers can protect the siRNA against degradation and transport it into the cell, but these polyelectrolyte systems are hampered by poor in vivo stability and toxic/immunogenic effects. Hydrophobic modification of siRNA is a promising approach to improve the pharmacokinetic properties of siRNA without the complexity and toxicity of traditional NP carriers. siRNA conjugated to the lipid palmitic acid (PA) acts synergistically with two distinct polymer NP carriers, improving carrier stability, pharmacokinetics, and cellular internalization, leading to enhanced gene silencing efficacy at reduced polymer doses. Modification of siRNA with PA is thus a powerful strategy to broaden the therapeutic index of NP-based strategies. Additionally, conjugation of an albumin-binding hydrophobe (termed L2) to siRNA broadly enhances its pharmacokinetic profile, biodistribution to tumors, and tissue penetration capacity. siRNA-L2 consistently outperformed a commercial NP carrier in tumor accumulation and biocompatibility and elicited significant and sustained in vivo tumor gene silencing, highlighting its potential as a translational and potentially transformative approach to improve systemic RNAi cancer therapies. The benefits conferred through PA and L2 conjugation establish a pivotal role for hydrophobic siRNA conjugates in the advance of siRNA to medical application.

With the rapid emergence of antibiotic-resistant bacteria and the lack of antibiotics in the development pipeline, bacteremia and sepsis are becoming an increasing health concern. Rapid diagnosis currently suffers from the need for the amplification of the bacterial signal, which is often only accomplished by overnight blood culture. Without a proper diagnosis, effective treatment may not be administered, resulting in increased mortality. Extracorporeal bacterial separation methods remove bacteria from whole blood, making them a promising avenue for both diagnosis and treatment. Specifically, micromagnetic separation removes bacteria bound to paramagnetic beads using a strong magnetic field. A small-footprint microfluidic device for micromagnetic separation was fabricated and characterized by volumetric flow testing and computational fluid dynamics. The framework for a computational analysis of micromagnetic separation was also developed, incorporating a magnetostatic finite element analysis of the device and a physiologically-based pharmacokinetic model of bacteremia. Small-footprint microfluidic devices were shown to be able to capture bacteria, but at rates and with overall capacities inadequate as an effective treatment for bacteremia. As such, these devices could be considered as means of rapid diagnosis; the development of high-throughput extracorporeal blood-cleansing devices is required for bacteremia treatment.

Integrating personalized risk predictions into clinical decision support requires well-calibrated models, yet model accuracy deteriorates as patient populations shift. Understanding the influence of modeling methods on performance drift is essential for designing updating protocols. Using national cohorts of Department of Veterans Affairs hospital admissions, we compared the temporal performance of seven regression and machine learning models for hospital-acquired acute kidney injury and 30-day mortality after admission. All modeling methods were robust in terms of discrimination and experienced deteriorating calibration. Random forest and neural network models experienced lower levels of calibration drift than regressions. The L-2 penalized logistic regression for mortality demonstrated drift similar to the random forest. Increasing overprediction by all models correlated with declining event rates. Diverging patterns of calibration drift among acute kidney injury models coincided with predictor-outcome association changes. The mortality models revealed reduced susceptibility of random forest, neural network, and L-2 penalized logistic regression models to case mix-driven calibration drift. These findings support the advancement of clinical predictive analytics and lay a foundation for systems to maintain model accuracy. As calibration drift impacted each method, all clinical prediction models should be routinely reassessed and updated as needed. Regression models have a greater need for frequent evaluation and updating than machine learning models, highlighting the importance of tailoring updating protocols to variations in the susceptibility of models to patient population shifts. While the suite of best practices remains to be developed, modeling methods will be an essential component in determining when and how models are updated.

The spatial distribution of genetic variation within proteins is shaped by evolutionary constraint and thus can provide insights into the functional importance of protein regions and the potential pathogenicity of protein alterations. To facilitate the spatial analysis of coding variation in protein structure, we develop PDBMap, an automated pipeline for mapping genetic variants into all solved and predicted protein structures. We then comprehensively evaluate the 3D spatial patterns of constraint on human germline and somatic variation in 4,568 solved protein structures. Different classes of coding variants have significantly different spatial distributions. Neutral missense variants exhibit a range of 3D constraint patterns, with a general trend of spatial dispersion driven by constraint on core residues. In contrast, germline and variants are significantly more likely to be clustered in protein structure space. Finally, we demonstrate that this difference in the spatial distributions of disease-associated and benign germline variants provides a signature for accurately classifying variants of unknown significance (VUS) that is complementary to current approaches for VUS classification.

Patient care is complex and imperfect. Understanding and improving patient care requires clinical datasets and scientific methodology. We designed a set of methods to characterize the state of patient care and identify opportunities for improvement from electronic health record (EHR) data. The state of patient care is the distribution of patients throughout a clinical workflow. An opportunity for improvement is a means to shift patient distribution away from suboptimal states. We tested our methods within Vanderbilt University Medical Centerâs (VUMC) EHR system and the adjuvant endocrine therapy domain. Our methods divide into three aims: 1) Determine sufficiency of the data, 2) Characterize the state of care, and 3) Identify opportunities for improvement. Data sufficiency is the rise and persistence of data in an EHR system. We built metrics for data sufficiency that can be used in cohort and data selection. We find that despite inconsistent and missing data, we can leverage EHR data for studies on patient care. To characterize the state of patient care, we built a state diagram for adjuvant endocrine therapy at VUMC, and used EHR data to determine the distribution of patient across states. We measured drug choice frequencies, rates of adverse events, and recurrence rates. We also determined the extent to which EHR data can characterize complete patient care. To identify an opportunity for patient care improvement, we identified a suboptimal state (failure to follow-up) among VUMC adjuvant endocrine therapy patients and framed a classification problem using EHR data. We used supervised machine learning to predict follow-up and identify significant predictors that may inform on improvement. Patients that fail to follow-up may receive the majority of their care outside of VUMC. Follow-up could be improved by 1) referral to VUMC primary care provider or 2) documenting where patients follow-up to reduce ambiguity of care. These methods characterized the state of patient care and opportunities for improvement among an adjuvant endocrine therapy patient population using VUMCâs EHR data. We believe these methods are extensible to other EHR systems and other healthcare domains. These methods are valuable for drawing new clinical knowledge from clinical datasets.

Master of Science / Department of Electrical and Computer Engineering / Steven Warren / Smart phones and other mobile technologies can be used to collect and display physiological information from subjects in various environments – clinical or otherwise. This thesis highlights software app reference designs that allow a smart phone to receive, process, and display biomedical data. Two research projects, described below and in the thesis body, guided this development. Android Studio was chosen to develop the phone application, after exploring multiple development options (including a cross-platform development tool), because it reduced the development time and the number of required programming languages. The first project, supported by the Kansas State University Johnson Cancer Research Center (JCRC), required a mobile device software application that could determine the hemoglobin level of a blood sample based on the most prevalent color in an image acquired by a phone camera, where the image is the result of a chemical reaction between the blood sample and a reagent. To calculate the hemoglobin level, a circular region of interest is identified from within the original image using image processing, and color information from that region of interest is input to a model that provides the hemoglobin level. The algorithm to identify the region of interest is promising but needs additional development to work properly at different image resolutions. The associated model also needs additional work, as described in the text. The second project, in collaboration with Heartspring, Wichita, KS, required a mobile application to display information from a sensor bed used to gather nighttime physiological data from severely disabled autistic children. In this case, a local data server broadcasts these data over a wireless network. The phone application gathers information about the bed over this wireless network and displays these data in user-friendly manner. This approach works well when sending basic information but experiences challenges when sending images. Future work for both project applications includes error handling and user interface improvements. For the JCRC application, a better way to account for image resolution changes needs to be developed, in addition to a means to determine whether the region of interest is valid. For the Heartspring application, future work should include improving image transmissions.

Advanced neuroimaging techniques provide the possibility to non-invasively understand and monitor white matter during development and disease. While data from quantitative MRI techniques, such as multiexponential T2 (MET2) and quantitative magnetization transfer (qMT), correlate with myelin content, neither provide an absolute measure of the myelin volume fraction (MVF). Additionally, in preclinical studies, despite time-intensity and small tissue samples, histology remains the gold standard for quantitatively assessing changes in myelin content and white matter microstructural properties, such as myelin thickness and the g-ratio (ratio of axon radius to myelinated fiber radius). Therefore, the work in this dissertation first established and validated methods for MVF imaging from MET2 and qMT against quantitative electron microscopy. We show strong agreement in adult, control mice along with three mouse models of white matter disease. Next, we applied MVF imaging in mice during normal development and observe good agreement between MET2 and qMT and with expected myelin development. To further investigate specific changes in myelin microstructure, recent methods proposed measuring the g-ratio from MRI (gMRI). We revised the model and displayed with quantitative histology that gMRI provides an axon-area-weighted g-ratio. Calculating gMRI requires an accurate measure of MVF; thus, we utilize our MVF imaging techniques to measure gMRI in mouse brain and detect changes in g-ratio with disease in agreement with quantitative histology. In short, we develop and validate measures of MVF and g-ratio from MRI which have the potential to non-invasively provide more specific and thorough assessment of white matter not obtainable with currently used methods.