Purposes: The purpose of this study was to estimate the incidence rate of epilepsy in 2009 among various patient populations of Arizona seniors enrolled in Medicaid. Additionally, this study aimed to assess the incremental and relative economic and healthcare resource burden of new-onset epilepsy compared with those without newly diagnosed epilepsy. Methods: Arizona Medicaid claims data years 2008 through 2010 were used for this analysis. Patients who were at least 65 years of age as of January 1, 2009 and were continuously enrolled in any coverage group pertaining to Arizona Medicaid for at least 12 months during 2008 and 2009 were considered eligible for the 2009 cohort; patients who were continuously enrolled during 2008 but not at all during 2009 were excluded. In addition to meeting the criteria for the 2009 cohort, incident cases of 2009 must also have epilepsy-related healthcare claims, a one-year clean period during which they cannot have any epilepsy-related healthcare claim, and that their first epilepsy-related healthcare claim in 2009 cannot have with a diagnosis code of 345.x1. The outcome variables assessed included incidence rate of being diagnosed with epilepsy, total monthly healthcare costs, total monthly inpatient costs, total monthly outpatient costs, total monthly prescription costs, incidence rate of inpatient stay, and incidence rate of physician visits. Results: The incidence rate of epilepsy in 2009 for this population was 7.9 per 1,000. Patients with epilepsy-related disease-based risk factors and of younger age had significantly greater incidence rate compared with their counterparts. In general, total monthly healthcare costs and incidence rate of inpatient stay were approximately three times greater in patients with newly diagnosed epilepsy compared with their counterpart (p < 0.001); incremental total monthly healthcare costs in patients with newly diagnosed epilepsy reached 2,077 US dollars. Conclusion: The Arizona Medicaid population appeared to have higher incidence rate compared with the US Medicare and the general US population. Additionally, Arizona Medicaid beneficiaries with newly diagnosed epilepsy had significantly greater healthcare costs and resource utilization compared with those without epilepsy.

Purpose: The purpose of this research was to examine and compare readmissions between the rural and urban areas for dual eligible patients in Arizona. The study also examined if living in an area of high socioeconomic status as opposed to living in an area of low socioeconomic status affects patient's risk of all-cause-30-day readmission. Methods: The study used data on dual eligible members who were continuously enrolled in AHCCCS and the University of Arizona Health Network between January 1st, 2011 and November 30th, 2012. The outcomes of interest in this study were risk of all-cause-30-day readmission, length of stay at index admission, and cost of readmission (i.e. amount paid by the University of Arizona Health Network). Log-binomial regression, Poison regression, and gamma regressions were used to model risk of readmission, length of stay at index admission, and cost of readmissions respectively. Results: Readmission of dual eligible patients in this sample was not related to residential location of patients. All-cause-30-day readmission did not differ for rural dual eligible patients and urban dual eligible patients. Readmission of dual eligible patients in this sample was not related to socioeconomic conditions. Dual eligible patients discharged in the year 2011 had a significantly higher risk of all-cause-30-day readmission than those discharged in the year 2012 (RR=1.05; p=0.03). Dual eligible patients discharged to skilled nursing facilities had a higher all-cause-30-day readmission risk compared to those discharged home. Length of stay at index admission was not associated with residential location. Patients admitted in the year 2011 were more likely to stay longer than those admitted in 2012 (IRR=1.13; p<.0001).The longer a dual eligible patient stayed during their readmission, the more their cost of readmission. (RR=1.33; p=0.0024). Conclusions: Readmission was neither associated with residential location nor socioeconomic condition. Cost associated with all-cause-30-day readmissions was not associated with residential location of the patients. The study also concludes that length of stay at index admission did not differ with socioeconomic conditions of the areas patients lived. It also concludes that year of discharge was associated with their risk of all-cause-30-day readmission.

Targeted cancer therapeutics represent the advent of a new therapeutic age, brought forth by the small molecule tyrosine kinase inhibitor (TKI) imatinib (Gleevec®). Imatinib is able to cause complete and sustained remissions in patients with chronic myelogenous leukemia (CML) driven by the Abelson (ABL) kinase, which caused a massive paradigm shift in how cancer is treated. The following research has been completed to extend the principles of imatinib therapy to the rearranged during transfection (RET) kinase. The RET kinase is involved in driving the pathology of medullary thyroid cancer (MTC), papillary thyroid carcinoma (PTC), certain non-small cell lung cancers (NSCLC), chronic myelomonocytic leukemia (CMML), tamoxifen resistant breast cancer, and Spitz melanoma. A heavily diverse population of solid and liquid carcinomas are driven by the RET oncogene, and patients presenting with these cancers could significantly benefit from a RET inhibitor. Previous drug discovery campaigns identified RET activity after therapeutic development for an unrelated kinase, as the case with vandetanib (Calpresa®) and cabozantinib (Cometriq®). Both agents fail to achieve dominant activity on RET and are more active on the vascular endothelial growth factor receptor 2 (VEGFR2), yet still achieve efficacy in RET driven tumors. This likely results from interrupting the oncogene cooperation between RET and VEGFR2; VEGFR2 provides the nutrients through angiogenesis that RET requires to promote proliferation and survival. We hypothesized that an equipotent RET/VEGFR2 dual inhibitor could maximize inhibiting the cooperation between RET and VEGFR2 in RET driven cancers. The inhibitor should be developed to maintain activity on all known RET mutations for treatment durability. In that case, the RET oncogene, despite mutating, will always be inhibited. Through research efforts, Pz-1 was identified as a sub-nanomolar, equipotent inhibitor of both RET (IC₅₀<0.001 µM) and VEGFR2 (IC₅₀<0.001 µM). Pz-1 was found active on every known, clinically relevant RET mutant tested at an IC₅₀≤0.001 µM. Through RET-driven xenograft models, Pz-1 was found active at an oral dose as low as 0.3 mg/kg/day.

Several methods have been proposed for the prediction of aqueous solubility. This study proposes the SCRATCH model for the aqueous solubility estimation of a compound directly from its structure. The algorithm utilizes predicted melting points and predicted aqueous activity coefficients for the solubility estimation, reflecting the truly predictive nature of the model. It uses two additive, constitutive molecular descriptors (enthalpy of melting and aqueous activity coefficient) and two non-additive molecular descriptors (symmetry and flexibility). The melting point prediction is trained on over 2200 compounds whereas the aqueous activity coefficient is trained on about 1640 compounds, making the model very rigorous and robust. The model is validated using a 10-fold cross- validation.A comparison with the General Solubility Equation suggests that the SCRATCH predicted aqueous solubilities have a slightly more average absolute error. This could result due to the fact that SCRATCH uses two predicted parameters whereas the GSE utilizes only one predicted property. Although the GSE is simpler to use, the drawback of requiring an experimental melting point is overcome in SCRATCH which can predict the aqueous solubility of a compound just based on its structure and no experimental values.

The United States Environmental Protection Agency was established in 1970 to control, limit, and regulate pollutant entry into the environment. The primary sources of pollutants are motor vehicle emissions, chemical plants, production factories, land fills, and natural or man-made catastrophes. Persistent organic pollutants have been known to cause such aliments as cancer, respiratory disease, and birth defects. These compounds can also cause irreversible environmental effects such as ozone depletion.The amounts of pollutants in air, water, soil, and organic matter can be correlated with the octanol solubility, vapor pressure, octanol-air partition coefficient, and air-water partition coefficient. The estimation of physical properties plays an important role in understanding the fate of organic pollutants. Although it is more desirable to measure such properties, their estimations can be of great importance in conserving resources and minimizing exposure.In this dissertation new equations for the estimation of these properties are generated. This is accomplished without the use of fitted parameters or regression analysis. The only experimental input parameters are the transition temperatures. The transition properties were estimated from molecular structure. The average absolute errors for the estimated properties are less than one log unit from the experimental values.

Motivated by the urgent need for new molecular targets and novel agents to treat pancreatic cancer, a target-based approach to drug discovery was implemented that led to the identification, validation and targeting of the Aurora family of kinases. The Aurora kinases (A, B and C) are mitotic serine/threonine kinases involved in various aspects of mitosis, including centrosome separation, bipolar spindle assembly, chromosome alignment and cytokinesis. In this dissertation, the potential use of the Aurora kinases as therapeutic targets to treat pancreatic cancer was investigated. It was found that both Aurora A and Aurora B are overexpressed in pancreatic adenocarcinomas, suggesting that some cancer cells are dependent upon their activity for continued proliferation and survival. To validate this hypothesis, antisense oligonucleotides were used in cell-based assays to evaluate the biological consequences of Aurora A and/or Aurora B inhibition. It was found that perturbations in Aurora kinase signaling result in cell cycle arrest and apoptosis. The biological fingerprints of Aurora A and Aurora B inhibition were compared and contrasted in an effort to identify the superior therapeutic target. It was concluded that an Aurora A-targeted therapy may have some beneficial consequences; however, a therapeutic approach discriminating between Aurora A and Aurora B is not straightforward. A fragment-based approach relying heavily on computer modeling was used to design and identify a nanomolar inhibitor of the Aurora kinases; however, it showed activity only at high micromolar concentrations in cell-based evaluations suggesting the compound possessed unfavorable characteristics that limited its biological activity. The preclinical development of analogues of the compound discovered by the work presented in this dissertation is ongoing. Finally, the pancreas-specific overexpression of Aurora A kinase was shown to be insufficient to induced pancreatic tumorigenesis in a mouse transgenic model.

The folates are a family of structurally similar, water-soluble, B vitamins, documented to play prominently in human health and disease. The potential impact of folate nutrition has been demonstrated by large-scale epidemiological and nutritional studies indicating that decreased folate intake increases the risk of cancer development. Human skin is particularly prone to the development of carcinomas and it is established that skin cancer risk correlates with exposure to the complete carcinogen ultraviolet radiation (UVR) in the form of sunlight. Recently a link between skin, sunlight, and folate has emerged from studies demonstrating that folate species are degraded by exposure to wavelengths of UVR contained within the solar spectrum. It is hypothesized that the unique physiology, function, and environment of skin combine to make skin tissue prone to folate deficiencies and that folate supplementation is a promising strategy for the prevention of skin cancer. However, many questions regarding folate nutrition within human skin must be answered before strategies to modulate folate nutrition may be rationally designed and safely implemented. This work presents novel means to examine skin-specific folate nutrition, including an analytical method to quantify individual folate species in human keratinocytes adaptable for the analysis of intact skin tissue and innovative cultured keratinocyte models of both acute and chronic folate deficiencies. It is demonstrated that folate deficiencies in skin tissue are possible and even likely as proliferating human keratinocytes are unable to maintain intracellular folate concentrations when nutrient conditions are limited and exposure to UVR results in biologically relevant folate degradation. Folate deficiency in human keratinocytes is observed to have potential pro-carcinogenic consequences including S-phase proliferation arrest, increased inherent DNA strand breaks, increased uracil misincorporation into DNA, and deficiencies in DNA damage repair, which are reversed when folate nutrient levels are optimized. The presented work characterizes the relationship between intracellular folate species and environmental carcinogens known to induce skin cancer and addresses challenges facing supplementation strategies for specifically improving folate nutriture in human skin. In total, this report broadens our understanding of folate nutrition in human skin and demonstrates that optimization of folate nutrition holds promise as a cancer preventive strategy.

Thesis (Ph. D.)--University of Michigan.

Thesis (Ph. D.)--University of Michigan.

Dissertation (Ph.D.)--University of Michigan.