Poor dose optimization in cancer trials leads to poor patient outcomes in tumor suppression and drug tolerance as well as failures in the drug development process. Most phase I clinical cancer trials still use traditional dose-finding methods, which are inadequate for evaluating novel cancer therapies, such as molecularly targeted agents and immunotherapies. Traditional approaches include using rule-based designs instead of model-based designs, assuming one dose should be recommended to all patients, and assuming the higher the dose, the better. This dissertation aims to address each of the inefficiencies that exist in phase I trials to optimize patient and trial outcomes in oncology, specifically in settings where the patient population is heterogeneous, i.e., settings where eligibility criteria have been expanded or settings evaluating a therapy that targets multiple tumor types and mutations. In the first part of this work, we address the inefficiencies of rule-based designs and the barrier to implementation of model-based designs. We use published phase I trials that used the most common rule-based method, the 3+3 design, and compare the trial outcomes to those obtained with novel model-based designs. In the second part of the work, we propose a broadened eligibility dose-finding design to address the situation of unknown patient heterogeneity in phase I cancer trials where eligibility is expanded, and multiple eligibility criteria could potentially lead to different optimal doses for patient subgroups. Lastly, we address patient heterogeneity in efficacy by developing a dose-optimization design that accounts for patient-specific characteristics, toxicity, pharmacokinetic data, and efficacy to identify the target population and inform the optimal dose for each subpopulation. The findings in each work highlight the advantages of model-based designs, particularly when tailored for the therapy and patient population in question. Using published dose-finding trials, we show that novel designs would recommend different doses about 40% of the time and confirm the advantages of these designs compared with the 3 + 3 design, as suggested previously by simulation studies. When accounting for heterogeneity in toxicity, the broadened eligibility design identified when the expanded subpopulation should be recommended a lower dose due to their tolerance and identified the criteria affecting toxicity at least 60% of the time in simulation studies. The dose-optimization design, focusing on heterogeneity in efficacy, demonstrated that a model-based approach to identifying the target population can be effective. Further, in the presence of heterogeneity, patient characteristics relating to molecular tumor characteristics were identified correctly, and a different optimal dose was recommended for each identified target subpopulation. The simulation studies of all proposed designs show that accounting for heterogeneity, even when the source of heterogeneity is unknown, is beneficial. In addition, the simulation studies highlight the poor performance of a naive method that recommends one dose for all. Our findings in this dissertation reveal the large proportion of the patient population that will be incorrectly dosed if inappropriate dose-finding designs are used. While we cannot directly understand the effect of dose selection on cancer trial outcomes, it is likely that not handling characteristics of novel cancer therapies early on contributes to the high attrition rates of cancer trials and the toxicity burden encountered in later trials and post-approval studies.

The purpose of this study was to examine the pharmacokinetics of d-xylose in man as a function of age with particular emphasis on its absorption characteristics. This study required the development of a specific and sensitive method for the quantitation of xylose from plasma and urine. Following a clean-up procedure, plasma or urine samples are concentrated and undergo two sequential derivatization steps and then are quantitated by capillary column gas liquid chromatography (GLC). D-Xylose is frequently quantitated by a tedious colorimetric assay involving the use of thiourea, a proven animal carcinogen. We have evaluated a more expedient colorimetric assay employing less toxic reagents. Based upon these comparisons the "phloroglucinol" method has been recommended as a replacement for the currently used clinical method for quantitating d-xylose. The human studies revealed age related changes in some but not all d-xylose disposition parameters. Systemic, renal, and nonrenal clearances all declined with advancing age. The terminal elimination half life increased with age. Age had very little influence on the various volumes of distribution. In general, parameters relating to oral absorption showed no age-related dependence. In contrast to what is generally believed, the bioavailability of d-xylose did not decline with age. Lastly, this dissertation addresses the problem of how infusion data may best be fit. Concentration-time data may be fit by a nonlinear regression algorithm in two ways; (1) concentration-time data may be collected and fit both during infusion and after infusion is terminated, (2) concentration-time data may be collected only after the infusion is terminated and be fit as a bolus. Concentration-time data were computer simulated with random error and we found that fitting the entire curve gave the most accurate estimates of disposition parameters.

This thesis describes a novel "Tablet-in-a-Bottle" oral suspension formulation. Ingredients with unstable physical or chemical characteristics can be placed in a core tablet, and then dry compression coated with an outer layer which provides separation from other components. The new suspension formulation comprises fast disintegrating clavulanic acid (KCA) tablets with a powder mixture containing amoxicillin. Hardness, friability, flow properties and weight uniformity of tablets for three different formulations were investigated and were all improved in a third formulation. Stability tests under different humidities were conducted. Amoxicillin and clavulanic acid in the new formulations showed the same stabilities when compared with the marketed product Augmentin��. Preliminary pharmacokinetics and bioavailability of one new formulation were evaluated by comparing in vitro release rates and in vivo urinary excretion rates. In vitro dissolution studies were carried out according to the USP XXIII paddle method. The new formulation showed faster release rates during the first hour when stirring speed was 25 rpm. However, when 75 rpm stirring speed was applied, the dissolution profiles for the new formulation and the reference marketed product were identical. A randomized two-way crossover bioequivalence study was designed to evaluate the bioavailabilities. Cmax, Tmax and AUC[subscript o--->t] of amoxicillin were within ��20% of the reference pharmacokinetic values. However, Cmax and Tmax of clavulanate were not within ��20%. Bioeqivalence between this new suspension formulation and the marketed product (Augmentin��) were evaluated using a two one-sided t-test. There is not sufficient statistical support with this test to conclude that the two products are bioequivalent. However, this is most likely due to small sample size and high intersubject variation and statistical support for bioequivalence is expected in a larger study group. / Graduation date: 1998

Several polymer systems including P(MAA-g-EG) and P(MAA-co-NVP) with crosslinking agents TEGDMA and PEGDMA1000, monomer-to-solvent ratios of 67:33, 60:40, and 50:50, and particle sizes of <75 microns, 90-150 microns, and 150-212 microns were synthesized for use with protein-transporter conjugates. All synthesized systems were characterized by SEM which demonstrated the visual size, surface features, and surface textures of the polymer microparticles. Insulin-transferrin and calcitonin-transferrin conjugates were successfully synthesized using the protein crosslinker SPDP, binding the two proteins with a disulfide bond. The multi-step conjugation reactions used to create the conjugates were analyzed by the use of UV spectroscopy and HPLC to ensure the quality of the final products. In both conjugation reactions, the final product yield was found to be over 70%. The in vitro loading and release characteristics for insulin-transferrin and calcitonin-transferrin were separately investigated. By testing loading and release using a number of different polymer systems with different synthesis parameters, it was possible to optimize the hydrogel carriers for use with each of the conjugates independently. Upon optimization, the ideal system for use with insulin-transferrin and calcitonin-transferrin was found to be P(MAA-g-EG) microparticles of <75 microns formed using a PEGDMA1000 crosslinker and a 50:50 monomer-to-solvent ratio for both conjugates through separate optimization processes. This optimized polymer carrier was found to release upwards of 50% of loaded insulin-transferrin conjugate and near 90% of loaded calcitonin-transferrin conjugate. The insulin-transferrin conjugate was further evaluated through the use of cellular and animal models. Using cellular models, the insulin-transferrin conjugate was shown to increase transport relative to insulin by a factor of 7, achieving an apparent permeability of 37 x 10⁹ cm/s. Also, in the presence of polymer microparticles, the insulin-transferrin conjugate increased transport by a factor of 14 times relative to insulin, achieve an apparent permeability of 72.8 x 10⁹ cm/s. The presence of the microparticles near the cells was found to improve conjugate transport by nearly 100%. The preliminary animal studies verified the successful synthesis of the insulin-transferrin conjugate as well as demonstrated the bioactivity of the insulin portion of the molecule by achieving a drop in blood glucose level upon subcutaneous injection. / text

Co-trimoxazole, a combination of a 5-to-l ratio of Sulphamethoxazole (SMZ) and Trimethoprim (TMP) , is a highly effective, broad-spectrum antibacterial agent. Since its introduction in 1968, it has been extensively used in infections of the respiratory and urinary tracts. Co-trimoxazole was developed by the systematic investigation of a series of compounds whose mechanism of action was already known. As early as 1950 synergy between sulphonamides and 2,4-diaminopyrimidines was reported. This was to be expected as both groups of drugs exert their antibacterial activity by interfering with the same biochemical pathway in bacteria. TMP was chosen from among many 2,4-diaminopyrimidines tested because of its good antibacterial activity and low toxicity. SMZ was chosen from the sulphonamides available for combination with TMP because of similarity of their biological half-lives. The widespread use of the combination coupled with the fact that monitoring of the levels of all drugs in the body is becoming increasingly important has stimulated research into rapid and efficient methods for the analysis of TMP and SMZ in biological fluids. Another consequence of the immense popularity of the combination is the appearance on the market of several generic preparations of Co-trimoxazole. It is now generally recognized that drug products from different manufacturers which are chemically equivalent may not be therapeutically equivalent. This is due to the fact that the absorption rate and/or bioavailability (extent of absorption) of a poorly soluble drug may be markedly affected by its release rate from the product and by its subsequent dissolution rate in gastrointestinal fluids. Hence bioequivalence of these various products should be established

The use of the oral route of administration for the treatment of young children with antibiotics can at times be problematic since, factors such as nausea, vomiting, taste and/or smell, in addition to the challenges associated with the administration of suspensions, may contribute to poor patient compliance. In such cases, the use of the rectal route of administration may be appropriate. Therefore, suppositories containing 250 mg azithromycin (AZI) were manufactured and assessed for potential as an antibiotic suppository dosage form. Suppositories, containing AZI dihydrate were manufactured by the fusion method, using different grades of PEG, Witepsol® and Suppocire® bases. The rate and extent of AZI release was evaluated using USP apparatus I, and samples were analyzed using a validated HPLC method. Differences in the rate and extent of AZI release were observed with the greatest amount of AZI being released from PEG formulations. The rate and extent of AZI release from formulations manufactured using fatty bases were influenced by physicochemical properties, such as melting rate and hydroxyl value, of the bases. In addition drug partitioning appeared to favor the lipid phase and had a negative impact on AZI release characteristics. Two different formulation approaches were used in an attempt to increase the rate and extent of AZI release from fatty base formulations. The use of surfactants significantly increased AZI release from formulations manufactured with fatty bases with high hydroxyl values. The use of urea or Povidone K25 in combination with AZI as a physical mixture or solid dispersion did not increase the rate and extent of AZI release from the fatty suppositories, to any significant extent. The mechanism of drug release was evaluated using several mathematical models, including the Higuchi, Korsmeyer- eppas, Zero and, First order models. In addition, in vitro dissolution profiles were characterized by the difference and similarity factors, f1 and f2 and by use of the Gohel similarity factor, Sd. AZI release kinetics were best described by the Higuchi and Korsmeyer-Peppas models and the values of the release exponent, n, revealed that drug release was a consequence of the combined effects of AZI diffusion, rate of melting of the base and partitioning of the drug which can be considered to be anomalous release.

Over the past few decades many researchers have investigated the utility of in vitro - in vivo correlations for the assessment of dosage forms. These investigations are, however, dependent on reproducible dissolution data and well conducted biostudies in order to establish meaningful and robust correlations. Despite the fact that the establishment of such correlations is perhaps idealistic, considerable interest has still been shown in this area of research. Various Controlled/Modified Release Dosage Forms (CMRD's) of theophylline, a weakly basic drug, and indomethacin, a weakly acidic drug, were assessed in order to establish in vitro - in vivo correlations. Dissolution rate studies were carried out using either the USP basket or paddle apparatus. The dissolution rate studies were conducted in a range of dissolution media of varying pH. Bioavailability studies were conducted on the dosage forms used by the Biopharmaceutics Research Institute at Rhodes University. The results of these biostudies were kindly made available for use in this research project. Type A correlations were established using a mathematical simulation process whereby expected in vivo responses are simulated and compared to actual profiles obtained for the dosage forms. In order to perform the simulations the dissolution rate profiles were stripped and using linear regression and the methods of residuals the dissolution rate order and the relevant dissolution rates were obtained. The results of the s imulations indicated that the in vivo serum concentration-time curves could be accurately predicted for the theophylline dosage forms but to a lesser extent, for the indomethacin formulations. The dissolution rate studies indicated that the paddle method is a suitable method for dissolution rate studies of theophylline CMRD's, although it appeared that the optimum pH of the dissolution medium was formulation dependent. Dissolution rate studies conducted on indomethacin formulations indicated that the USP specified basket method for extended-release indomethacin formulations was not able to distinguish between two formulations which exhibited different in vivo profiles. The conversion to the paddle method was, however, able to highlight the differences between these formulations. The use of three dimensional topographs to depict dissolution rate profiles was demonstrated for formulations of both theophylline and indomethacin. The topographs enabled the successful differentiation between bioinequivalent formulations. The dissolution rate profiles were also fitted to the Wei bull equation and the parameters obtained from this were compared to the Weibull parameters obtained from the in vivo absorption plots obtained using the Wagner-Nelson method. The results indicated that the Weibull function was suitable to describe both the in vivo and in vitro data. The following recommendations for the preformulation dissolution studies of weakly acidic and weakly basic drugs are proposed. The dissolution rate studies of weakly acid drugs, such as indomethacin, should be carried out over a range of pH utilising the paddle apparatus. Three dimensional topographs based on the dissolution data should be constructed and used as a comparative tool for different formulations. Based on these comparisons the appropriate formulation can then be selected for a pilot scale in vivo bioavailability study. The dissolution rate studies of weakly basic drugs, such as theophylline, should be carried out over a range of pH utilising the paddle apparatus. The dissolution data should then be used to simulate the expected in vivo profile and on this basis the appropriate formulation selected for a pilot scale bioavailability study. The above approach to the preformulation studies of new CMRO's would allow for the more careful selection of new dosage forms and could thus eliminate costly and unnecessary bioavailability studies performed on inferior formulations.

The dissolution profiles of eight lots of paracetamol tablets representing seven different tablet brands are determined in a USP rotating basket assembly and a stationary basket-rotating paddle apparatus. The in vitro data are expressed in terms of dissolution parameters and inter-tablet differences are assessed statistically using analysis of variance (ANOVA) and the Scheffe test. Highly significant differences are observed between a number of the tablets at the 95% confidence level. Representative tablets from the dissolution rate study and a control dose of paracetamol dissolved in water are subsequently investigated in a 4 x 4 latin square design bioavailability trial. Serum and urine samples are collected and assayed for paracetamol alone (serum) and together with its metabolites (urine) by means of high pressure liquid chromatography. The in vivo data are expressed in terms of bioavailability parameters and differences between the test doses are assessed by means of ANOVA. No significant differences are observed between the dosage forms at the 95% confidence level.

Numerous experiments in recent years have indicated differences in the bioavailability of corticosteroids from seemingly identical topical dosage forms. The human blanching assay was utilized in this study to assess the comparative blanching activities of various locally manufactured proprietary corticosteroid preparations. The first experiment was performed to assess the relative blanching activities of six semi - solid preparations containing the same concentration of betamethasone 17-valerate. The preparations used were Betnovate cream and ointment, Persivate cream and ointment and Celestoderm-V cream and ointment. This was followed, in the second experiment, by the investigation of the blanching activities of two lotions containing betamethasone 17-valerate (Betnovate and Celestoderm-V) and a lotion containing betamethasone 17,21- dipropionate (Diprosone). The third experiment involved a study of six semi-solid proprietary corticosteroid-containing formulations, viz. Dermovate (clobetasol propionate) cream and ointment, Betnovate (betamethasone 17-valerate) cream and ointment and Eumovate (clobetasone butyrate) cream and ointment. This investigation was prompted by claims in advertisements in the medical media that Dermovate is therapeutically more efficacious than Betnovate which is more efficacious than Eumovate. The penultimate experiment in this study served the purpose of finding a corticosteroid-containing preparation that falls into the moderately potent group of corticosteroid formulations, as described in the United Kingdom MIMS. This preparation was used in the final experiment which was undertaken to ascertain the potency category of Florone (diflorasone diacetate) cream and ointment.

Many Active Pharmaceutical Ingredients (API) such as the antiviral agent famciclovir (FCV) are required for paediatric treatment but are not commercially available in age-appropriate dosage forms. It is common practice to prepare oral liquid dosage forms using commercially available tablets, capsules or powdered API and then dispersing or dissolving the crushed and/or powdered materials in a vehicle that the patient can swallow. Vehicles that are commonly used for this purpose include methylcellulose, syrup or combinations of these carriers where possible or commercially available suspending agents such as Ora-Sweet®, if available, can be used. However, several critical factors are overlooked when manufacturing extemporaneous formulations including, but not limited to, physical and chemical properties of the API, excipients, compatibility, stability and bioavailability issues. A stability-indicating High Performance Liquid Chromatography (HPLC) method for the analysis of FCV was developed and validated according to the International Conference on Harmonization (ICH) guidelines. The method is sensitive, selective, precise, accurate and linear over the concentration range 2-120 μg/ml. The stability of 25 mg/ml FCV formulations was assessed in vehicles manufactured from syrup simplex, hydroxypropyl methylcellulose (HPMC), Ora-Sweet® and an aqueous buffer (pH 6) following storage at 25 °C/60% RH and 40 °C/75% RH over six (6) to eight (8) weeks. The shelf life of the products was calculated as the longest period of storage for approximately 90% of the added FCV to be recovered. Formulations were manufactured using syrup simplex or HPMC with methylparaben and propylparaben individually or in combination and with sodium metabisulphite, ascorbic acid or citric acid as antioxidants. The resultant products were subject to quality control analysis for API content, viscosity, pH and appearance and the resultant data were subject to statistical analysis. The degradation rates were calculated for each product and a degradation profile plotted. The degradation rates of FCV in extemporaneous formulations were compared to those of FCV manufactured using a commercially available suspending agent and a buffered vehicle. FCV undergoes major degradation in the presence of sucrose, as observed for formulations in which the vehicle was syrup and Ora-Sweet®. FCV was found to be most stable when dissolved/dispersed in an HPMC vehicle incorporating sodium metabisulphite and a combination of parabens. The formulation that exhibited the maximum stability was manufactured using an aqueous solution buffered to pH 6. Due to the enhanced stability of FCV when added to a buffered vehicle a formulation in which an HPMC vehicle buffered to pH 6 with sodium metabisulphite, methylparaben and propylparaben was selected for optimisation using a Central Composite Design approach (CCD). In this way it was possible to establish a relationship between input variables such as pH, % w/v HPMC, % w/v antioxidant and % w/v preservative and the responses selected for monitoring by means of response surface modelling. A quadratic model was found to be the most appropriate to describe the relationship between input and output variables. Thirty batches of product were randomly manufactured according to the CCD and analysed to establish the stability in respect of viscosity, pH and the amount of FCV remaining following storage and the data were fitted to models using Design-Expert® software. A correlation between input variables and the responses was best described by a quadratic polynomial model. Analysis of Variance indicated that the response surface models were significant (P-value < 0.0001). The pH to which a FCV formulation was buffered was the most significant factor to effect the % drug content and the ultimate pH of the formulation, while the % w/v HPMC had the most significant effect on the viscosity of the product. The optimum composition for the manufacture of an oral liquid FCV formulation was predicted using the optimisation function of the Design-Expert® software. A low % error of prediction was established, indicating that the model is robust and that RSM is an appropriate formulation optimisation tool as it has a high prognostic ability. A liquid FCV formulation was developed, optimised and found to be suitable for its intended purpose. However further optimisation is required in respect of colourants, sweeteners and/or flavourants. The approach followed is useful in ensuring the development of quality products and can be applied in future.