Combination chemotherapies that introduce multi-agent treatments to target cancer cells are emerging as new paradigms to overcome chemotherapy resistance and side effects involved with conventional monotherapies. In environmental toxicology, characterizing effects of mixtures of toxins rather than simply analyzing the effect of single toxins are of significant interest. In order to determine such combination effects, it is necessary to systematically investigate interactions between different concentration-dependent components of a mixture. Conventional microtiter plate format based assays are efficient and cost-effective, however are not practical as the number of combinations increases drastically. Although robotic pipetting systems can overcome the labor-intensive and time-consuming limitations, they are too costly for general users. Microfluidic live cell screening platforms can allow precise control of cell culture microenvironments by applying accurate doses of biomolecular mixtures with specific mixing ratios generated through integrated on-chip microfluidic gradient generators. This thesis first presents a live cell array platform with integrated microfluidic network-based gradient generator which enables generation and dosing of 64 unique combinations of two cancer drugs at different concentrations to an 8 by 8 cell culture chamber array. We have developed the system into a fully automated microfluidic live cell screening platform with uniform cell seeding capability and pair-wise gradient profile generation. This platform was utilized to investigate the gene expression regulation of colorectal cancer cells in response to combination cancer drug treatment. The resulting cell responses indicate that the two cancer drugs show additive effect when sequential drug treatment scheme is applied, demonstrating the utility of the microfluidic live cell assay platform. However, large reagent consumption and difficulties of repeatedly generating the exact same concentrations and mixture profiles from batch to batch and device to device due to the fact that the generated gradient profiles or mixing ratios of chemicals have to rely on stable flow at optimized flow rate throughout the entire multi-day experiment limit the widespread use of this method. Moreover, producing three or more reagent mixtures require complicated microchannel structures and operating procedures when using traditional microfluidic network-based gradient generators. Therefore, an on-demand geometric metering-based mixture generator which facilitates robust, scalable, and accurate multi-reagent mixing in a high-throughput fashion has been developed and incorporated with a live cell array as a microfluidic screening platform for conducting combination drug or toxin assays. Integrated single cell trapping array allowed single cell resolution analysis of drugs and toxin effects. Reagent mixture generation and precise application of the mixtures to arrays of cell culture chambers repeatedly over time were successfully demonstrated, showing significantly improved repeatability and accuracy than those from conventional microfluidic network-based gradient generators. The influence of this improved repeatability and accuracy in generating concentration specified mixtures on obtaining more reliable and repeatable biological data sets were studied.

Receptor Tyrosine Kinase (RTK) inhibitors, such as erlotinib/tarceva, have been introduced in the past decade as a promising therapeutic option in Head and Neck Squamous Cell Carcinoma (HNSCC), however, they lack significant efficacy as single agents. As a result, RTK inhibitors require a combination based therapeutic approach with other treatment modalities. To uncover such a combination of agents, we performed a high throughput Prestwick library screen that included 1200 compounds approved by the FDA on HNSCC cell lines and found that monensin, a coccidial antibiotic, synergistically enhanced the cytotoxicity of erlotinib. RT-PCR revealed that monensin induced the expression of Activation of Transcription Factor (ATF) 3 and its downstream target C/EBP homologous protein (CHOP) which are key regulators of apoptosis. Furthermore, RNA-Seq analysis suggests that monensin augments erlotinib cytotoxicity by disturbing lipid and sterol biosynthesis. Therefore, identifying the mechanism of action exerted by monensin may open alternative avenues of cancer treatment.

Platinum-based chemotherapies are effective agents in the treatment of a wide variety of human cancers. However, patients with recurrent disease can become resistant to platinum-based chemotherapy, leading to low overall survival rates. Activating transcription factor 3 (ATF3) is a stress-inducible gene that is a regulator of cisplatin-induced cytotoxicity. ATF3 protein expression was upregulated after cytotoxic doses of cisplatin treatment in a panel of cell lines. A chromatin immunoprecipitation assay showed that upon treatment with cisplatin, ATF3 directly bound to the CHOP gene promoter and this correlated with an increase in CHOP protein expression. In a 1200 compound library screen performed on cancer cell lines, disulfiram, a dithiocarbamate drug, was identified as an enhancer of the cytotoxic effects of cisplatin. This increased cytotoxic action was likely due to disulfiram and cisplatin’s ability to induce ATF3 independently through two separate mechanisms, namely the MAPK and integrated stress pathways. Furthermore, ATF3 protein and mRNA levels were variable amongst human ovarian and lung cancer tissues, suggesting the potential for basal expression of ATF3 to be predictive of cisplatin treatment response. Thus, understanding ATF3’s role in cisplatin-induced cytotoxicity will lead to novel therapeutic approaches that could improve this drug’s efficacy.

Previous work has suggested that triclabendazole, a member of the benzimidazole group of compounds, possessed efficacy against Schistosoma mansoni. In view of recent indications in praziquantel treatment failures and loss of sensitivity, it is imperative that new anti-schistosomals are developed as contingent treatment options, while resistance alleles, if any, remain at low frequencies. While recent studies have indicated that triclabendazole monotherapy exert weak anti-schistosomal effects, the combinatorial application of triclabendazole with praziquantel has not been explored. To assess this hypothesis, triclabendazole and its metabolites were initially assessed against the many life-stages of Schistosoma mansoni in vitro. Combinatorial drug and isobologram analyses against adult Schistosoma mansoni was also performed, and subsequently applied against other parasitic models (Giardia duodenalis and Haemonchus contortus) to assess the specificity of such effects. Subsequently, the drug combinations were assessed against Schistosoma mansoni in vivo. To further assess the suitability of combinatorial drug applications, an economic model was developed to project the cost-efficacy of praziquantel-triclabendazole drug combinations in a global focus. It was concluded that triclabendazole and its metabolites possessed good efficacy against immature schistosomula, albeit weak efficacy against adult Schistosoma mansoni. Upon combination with praziquantel, however, a strong synergistic effect against adult worms were observed in vitro. Praziquantel and triclabendazole were also shown to possess unique and independent ovicidal modes of action that can be of clinical significance. More importantly, in vivo drug trials concluded that the combinations exerted additive effects against Schistosoma mansoni harbored in mice. Economic modeling and cost-effectiveness analyses further demonstrated the feasibility of this drug combination, and may represent a new line of treatment against mansonial schistosomiasis

Diss. (sammanfattning) Stockholm : Karolinska institutet, 2006. / Härtill 5 uppsatser.

Thesis (Ph.D.)--University of Western Australia, 2007.

Thesis (M.S.)--Rochester Institute of Technology, 1991. / Spine title: Neural networks and optimization problems. Typescript. Includes bibliographical references (leaves 53-54).

In this paper we describe a technique of classifier combination used in a human identification system. The system integrates all available features from multi-modal sources within a Bayesian framework. The framework allows representinga class of popular classifier combination rules and methods within a single formalism. It relies on a “per-class” measure of confidence derived from performance of each classifier on training data that is shown to improve performance on a synthetic data set. The method is especially relevant in autonomous surveillance setting where varying time scales and missing features are a common occurrence. We show an application of this technique to the real-world surveillance database of video and audio recordings of people collected over several weeks in the office setting.

In asthma therapy, the use of theophylline to prevent bronchial spasm and glucocorticoids to decrease inflammation is widely indicated. Apart from the acute asthma attack oral theophylline is treated for chronic therapy in order to minimize inflammation and to enhance the efficiency of corticosteroids and recover steroids’ anti-inflammatory actions in COPD treatment. The preferred application route for respiratory disease treatment is by inhalation, such as dry powder inhalers (DPI) being the delivery systems of first choice. As shown recently, there is an advantageous effect if the drugs are given simultaneously which is caused by a synergistic effect at the same target cell in the lung epithelia. Therefore, it seems rational to combine both substances in one particle. This type of particle has the advantage over a combination product containing both drugs in a physical mixture which occurs rather randomly deposition leading to API segregation and non-dose-uniformity. Dry powder inhalers (DPIs) is a type of therapeutic pharmaceutical formulations usually present in the solid form. Due to the nature of the solid-state, an understanding of chemical and physical properties must be established for acquiring optimum performance of the active pharmaceutical ingredients (APIs). In recent year, generation of DPIs is a destructive procedure to meet the micron size. Such processes are inefficient and difficult to control. Moreover, according to current researches on combination APIs formulation, this type of DPIs performed a greater variability in does delivery of each active, leading to poor bioavailability and limit clinical efficient. This result suggest that combination formulations require advanced quality and functionality of particles with suitable physicochemical properties. Hence, in order to production of binary and combination DPIs products, the aim of this study was to develop the spray drying and ultrasonic process for engineering of combination drug particles that will be delivered more efficiently and independently of dose variations to the lung. Microparticles were produced by spray drying or/and ultrasonic technique. The processing parameters and addition of excipients (polymers) were optimized using a full factorial design such that microparticles were produced in a narrow size range suitable for inhalation. Employing excipients resulted in high saturation environment leading to minimized sphere particles when compared to conventional solvent. Solid state characterization of microparticles using powder x-ray diffraction and differential scanning calorimetry indicated that the particles contained crystalline but no cocrystal. The combination particles comparable to or better than micronized drug when formulated as a powder blended with lactose. It was concluded that the use of HPMC enhanced crystallinity suitable for inhalation; and combination particles improved uniform distribution on the stage of NGI.

The purpose of this study was to examine the training and development of teachers in multi-grade teaching in selected Eastern Cape primary schools with a particular focus on the integration of vertical and horizontal knowledge in post-training. Multigrade teaching is the combining of learners of different grade levels in one classroom taught by one teacher. To gain better understanding of the construct of vertical and horizontal integration of knowledge within post-training environment, this qualitative case study design sampled eight teachers, four school principals and three trainers, purposively. In order to evaluate the training and development provided for teachers in multi-grade classrooms the researcher examined the training programmes, the training curriculum and transfer of training inputs. The results of the study indicate that post-training is the determinant of the extent of the effectiveness of the transfer of training skills, knowledge and attitudes gained in multi-grade teaching training. The study found a massive gap that currently exists between training in multi-grade teaching and the actual transferability of such training in multi-grade contexts. Positive transfer of training in multi-grade teaching strategies was found to be skills related to teaching strategies, lesson management, curriculum integration and social components of multi-grade teaching. Negative transfer of training included classroom organisation, lesson planning, timetabling, curriculum adaptation and assessment. Although multi-grade teacher training programme is assumed to change behaviours, attitudes, impart knowledge and improve teaching skills, the results of the study indicate that training inputs might not necessarily transform into classroom practice. In order to facilitate and ensure vertical and horizontal integration of knowledge and skills and transfer of training to classroombased teaching, recurrent training in multi-grade teaching is therefore, suggested.