Mass spectrometric methods, including EI, CI, FAB and ESI LC/MS have been surveyed as tools for identification and characterization of compounds of natural origin exhibiting biological activity. Bioactive catechins isolated from green tea were analyzed by mass spectral methods: EI spectra provided both molecular weight and structural information, including epimer differentiation. FAB mass spectrometry gave both molecular weight and structural information on all compounds. ESI LC/MS provided unambiguous MW information on all compounds and some additional structural data on compounds ECG and EGCG. ESI LC/MS provided a means for separation of all compounds in a mixture and is an appropriate method for analysis of a crude extract of this plant material. Based on the result from biological testing, showing that quinic acid derivatives possess considerable anti-HIV activity, four analogs of dicaffeoylquinic acid were characterized by mass spectral methods. An attempt was made to design a mass spectral method allowing the differentiation between the analogs. FAB mass spectral analysis provided good MW information for all compounds. In addition, ions representing the elimination of water from MH+ ion of the 3,5-DCQA-OAc clearly differentiated this isomer from the 3,4-DCQA-OAc. MIKES analysis of the MH+ ions of the acetate derivative confirmed the isomer specific water loss. ESI provided unambiguous MW information on all compounds, confirming that loss of water is specific for the 3,5-DCQA-OAc. The extract of the CSF in patients with ALL was surveyed for a suitable biomarker which would indicate brain tissue damage following therapy. Phospholipid levels in CSF in three groups of patients receiving different CNS propylaxis were monitored during the course of treatment and the elevated levels were correlated to the cognitive impairment evaluation results. As a result of the CNS propylaxis, the levels of phospholipids in CSF are significantly elevated, indicating disruption of brain cell membranes. Two major classes of phospholipid were identified by FAB mass spectral analysis, PC and SM. Their elevated levels were inversely correlated with the decreased scores from cognitive testing. A close correlation was found between the PC levels and some test scores.

Enzyme inhibition studies are conducted to characterize enzymes and to examine drug-drug interactions. To characterize the inhibitory process (competitive, non-competitive and uncompetitive) and to determine the inhibitory constant (Kᵢ), data analysis techniques (e.g., Dixon, Lineweaver-Burk, etc.) are used to linearize the inherently non-linear rate of substrate metabolism vs. substrate concentration data. These techniques were developed before the general use of computers. However, many investigators still rely on these techniques in spite of the easy availability of non-linear regression fitting programs. In Chapter 2, three methods (simultaneous nonlinear regression fit (SNLR); Dixon; non-simultaneous, nonlinear fit [K(m,app)]) were compared for estimating Ki from simulated data sets generated from a competitive inhibition model equation with 10% CV added random error to the data values. Of the three methods, the SNLR method was found to be the most robust, the fastest and easiest to implement. The K(m,app) method also gave good estimates but was more time-consuming. The Dixon method failed to give accurate and precise estimates of Kᵢ. The purpose of the study in Chapter 3 was to examine the minimal experimental design needed to obtain reliable and robust estimates of Kᵢ (as well as V(max) and K(m)). Four cases were examined. In the experimental design that relied upon the least amount of data, a control data set was fit simultaneously with one of the substrate-inhibitor pairs (25-10 or 250-100 μM). A total of 4 rate values were analyzed per fit (i.e., 3 control + 1 inhibitor value). A total of 100 data sets were fit per substrate-inhibitor pair. The preceding was repeated for a random error of 20 %CV. Thus, the total number of experiments was reduced from 108 (in Chapter 2) to 12 (in Chapter 3) (Case IV). Good estimates of the enzyme kinetic parameters were obtained. In Chapter 4, the ability of the SNLR method to identify the correct mechanism of inhibition was evaluated; competitive or noncompetitive enzyme-inhibition. Two experimental designs were examined ("conventional, non-optimal" and "semi-minimal"). The semi-minimal design was successful in discriminating between the two enzyme-inhibition mechanisms even for data with 30 %CV added random error.

Toxicity is a major limitation in clinical use of most anticancer drugs. Liposomes, especially targeted long-circulating liposomes, provide the possibility of delivering drugs specifically to targeted cancer tissues, thus increasing anticancer activity and minimizing toxicity. 2-4'-Amino-3'-methylphenylbenzothiazole (AMPB), a potent anticancer drug, is inappropriate for traditional oral or parenteral formulations because of its severe dose-limiting hepatotoxicity. Several PEG-coated liposomal formulations were developed by using different drug/lipid ratios. Particle size and encapsulation efficiency of each formulation were investigated; the most stable liposomal formulation was selected for animal testing. The formulation with AMPB/egg phosphatidyl choline/cholesterol/1, 2-distearoyl-sn-glycero-3-phosphoethanolamine-N-polyethylene glycol 2000, PEG2000-DSPE in a 1/5/5/1 molar ratio is the best formulation. This formulation contains 2 mg/ml AMPB with encapsulation efficiency above 95%, average particle size 120-150 nm. Daunorubicin is a well known anti cancer agent. To minimize its cardiotoxicity, targeted folate-PEG-liposomes were developed in this study. The pH-gradient loading method was used to increase the drug loading efficiency. Above 97% loading efficiency was reached by creating a 3 to 4 unit pH difference across the liposome membrane. The final folate-PEG-liposomml formulation contained 2.5 mg/ml daunorubicin HCI, with average particle size of 110∼120 nm, pH of ∼7.4, and a drug/lipid ratio 1/20 (w/w). The solvent, chloroform, commonly used for liposome preparation, is harmful to humans. Therefore, halothane, a commonly used inhalation anesthetic, was used in this study in place of chloroform to prepare liposomes. AMPB and several other proprietary anticancer agents were formulated in liposomes by using halothane and chloroform. No obvious differences in physicochemical properties were observed between halothane and chloroform mediated liposomes.

The AQUAFAC (Aqueous Functional Group Activity Coefficients) method for predicting aqueous activity coefficients proposed by Myrdal et al. (1992) is expanded to include for hydrogen bonding groups: hydroxy, carboxy, nitro and amino. Activity coefficients can be used to estimate aqueous solubility. Using aqueous solubility data, from a number of sources, for a set of subsituted aromatic compounds, group or q values are derived. Group values have been generated for a number of substituents, none have included hydrogen bonding groups (Myrdal et al., 1992,1993). Q values are related to activity coefficients through the following relationship: log gammaw = Sigmaniqi where log gammaw is the log of the activity coefficient, qi is the group value subtituent i and ni is the group frequency. Ortho effects between hydrogen bonding groups is also examined. Intramoleculat hydrogen bonding involving carboxy substituted compounds, in this research, does appear to affect aqueous solubility.

<p>After hundreds of years of use the medical properties of Marijuana have been marginalized in our society. Qualitative interview data was collected from medical marijuana patients and knowledgeable producers and activists in Michigan about their perceptions and observations on the medical use of marijuana. Patients consistently reported using marijuana to substitute or wean off prescription drugs. All patients and producers who were taking opiate narcotics claimed they reduced overall drug use, especially opiates, by using medical marijuana. Patients and caregivers also claimed medical marijuana was preferred over opiates, eased withdrawal from opiates, and in some cases was perceived as more effective at relieving pain. Other issues explored included the operation of the Michigan Medical Marijuana Act, the formation and operation of medical marijuana centers in the face of countervailing State and federal, opposition, and the varieties and effects of different strains of medical marijuana. </p><p> Keywords: Medical Marijuana, State and Federal Marijuana Laws, Michigan Marijuana, Controlled Substances, Drugs, Qualitative Interviews </p>

The case-crossover design is a new epidemiological method that evolved around binary exposures and the binomial distribution. We develop a new approach of data analysis for this design based on the actual exposure occurrence times, such as those available from computerized prescription databases. Assuming an exponential distribution for the inter-exposure onset times, we derive two new matched-paired estimators of the odds-ratio, one weighted the other unweighted. A simulation study demonstrates that both new estimators based on the exponential distribution are more efficient than the classical estimator based on the binomial distribution and that the unweighted estimator appears to be the most valid. These new estimators of the odds-ratio are also more flexible and amenable to verifying some of the assumptions behind the case-crossover design. We illustrate this approach with data on 54 asthma deaths identified from the Saskatchewan Health databases, to assess the association with the use of inhaled beta-agonists.

In the hospital setting, infant pain is frequently poorly diagnosed and managed. In the present study we investigated the behavioural response to severe pain in the developing nervous system as well as the degree to which morphine can produce adequate analgesia. / We describe the behavioural response in 3 and 15 day old rat pups to intraplantar injection of formalin at concentrations capable of inducing c-Fos. These concentrations are 10-fold higher than that necessary to evoke continuous pain behaviour over a period 40--60 minutes. Dose-effect relations for morphine were examined to determine if pain continued to increase with formalin concentration, in the presence of maximal levels of pain behaviour. / Pain behaviour increased in intensity and duration with formalin concentration in infant rats, despite the presence of a behavioural ceiling, as illustrated by the increased morphine requirement for analgesia. Non-specific pain behaviour and sleep/wake states were altered by severe pain. The analgesic effects of morphine were found to be independent of its sedative effects in both age groups. Moreover, in rat pups, full analgesia cannot be obtained by subtoxic doses of morphine when high formalin concentrations are used. These data suggest that severe pain is not fully controllable in the infant due to immaturity of the substrates of opioid analgesia, and raise ethical questions about the use of high formalin concentrations in immature animals. The data also underline the need for adequate pain measures in human infants that can distinguish between pain relief, sedation and coping behaviour.

The graphical presentation of survival functions by the Kaplan-Meier method is very useful for reporting results from clinical trials or cohort studies. However, this method is based on the "intent to treat" approach that assumes that the exposure under study is fixed over time, as are the covariates that are in addition assumed to be balanced between groups. These situations are not often the case in observational studies, where the exposure may varies over time, and covariates may be both time-fixed and time-dependent. In this thesis, we compare two methods for estimation of the survival function with time-varying exposure to the time-fixed approach. We apply these techniques to a cohort study of the effectiveness of inhaled corticosteroids in the treatment of chronic obstructive pulmonary disease (COPD). / The two methods are an extension of the Kaplan-Meier estimator and the Nelson-Aalen estimator based on the Cox proportional hazards model. Both methods are based on the assumption that the effect of exposure is acute. (Abstract shortened by UMI.)

<p> There are a number of adverse drug events per year due to ingesting the wrong medication. The present study examined whether including a picture of the medicine on pharmaceutical labels increased the participants' ability to identify the medicine that they should be taking. Participants were shown fake pharmaceutical labels that included either a picture or text description of the medicine's appearance, and were asked to identify the medicine being depicted or answer questions about the information found on the label. Participants performed better on questions relating to the content on the label than on questions asking them to identify the medicine. However, if the depiction was a color picture of the medication, participants' performance better than if the depiction was a black-and-white picture or text description. Thus, a color picture may help patients identify their medicine and reduce the number of cases involving the ingestion of the wrong medication.</p>

<p> Focusing medicine to disease locations is a needed ability to treat a variety of pathologies. During chemotherapy, for example, typically less than 0.1% of the drugs are taken up by tumor cells, with the remaining 99.9% going into healthy tissue. Physicians often select the dosage by how much a patient can physically withstand rather than by how much is needed to kill all the tumor cells. The ability to actively position medicine, to physically direct and focus it to specific locations in the body, would allow better treatment of not only cancer but many other diseases. </p><p> Magnetic drug targeting (MDT) harnesses therapeutics attached to magnetizable particles, directing them to disease locations using magnetic fields. Particles injected into the vasculature will circulate throughout the body as the applied magnetic field is used to attempt confinement at target locations. The goal is to use the reservoir of particles in the general circulation and target a specific location by pulling the nanoparticles using magnetic forces. </p><p> This dissertation adds three main advancements to development of magnetic drug targeting. Chapter 2 develops a comprehensive ferrofluid transport model within any blood vessel and surrounding tissue under an applied magnetic field. Chapter 3 creates a ferrofluid mobility model to predict ferrofluid and drug concentrations within physiologically relevant tissue architectures established from human autopsy samples. Chapter 4 optimizes the applied magnetic fields within the particle mobility models to predict the best treatment scenarios for two classes of chemotherapies for treating future patients with hepatic metastatic breast cancer microtumors.</p>