Pharmacometric modelling is widely used in many aspects related to type 2 diabetes mellitus (T2DM), for instance in the anti-diabetes drug development, and in quantifying the disease progression of T2DM. The aim of this thesis were to improve the design of early phase anti-diabetes drug development studies with the focus on the power to identify mechanism of drug action (MoA), and to characterize and quantify the progression from prediabetes to overt diabetes, both the natural progression and the progression with diet and exercise interventions, using pharmacometrics modelling. The appropriateness of a study design depends on the MoAs of the anti-hyperglycaemic drug. Depending on if the focus is power to identify drug effect or accuracy and precision of drug effect, the best design will be different. Using insulin measurements on top of glucose has increase the power to identify a correct drug effect, distinguish a correct MoA from the incorrect, and to identify a secondary MoA in most cases. The accuracy and precision of drug parameter estimates, however, was not affected by insulin. A natural diabetes disease progression model was successfully added in a previously developed model to describe parameter changes of glucose and insulin regulation among impaired glucose tolerance (IGT) subjects, with the quantification of the lifestyle intervention. In this model, the assessment of multiple short-term provocations was combined to predict the long-term disease progression, and offers apart from the assessment of the onset of T2DM also the framework for how to perform similar analysis. Another previously published model was further developed to characterize the weight change in driving the changes in glucose homeostasis in subjects with IGT. This model includes the complex relationship between dropout from study and weight and glucose changes. This thesis has provided a first written guidance in designing a study for pharmacometrics analysis when characterizing drug effects, for early phase anti-diabetes drug development. The characterisation of the progression from prediabetes to overt diabetes using pharmacometrics modelling was successfully performed. Both the natural progression and the progression with diet and exercise interventions were quantified in this thesis.

生物醫藥產業對人類健康福祉影響甚鉅，同時也屬於技術、資本密集、開發期長、風險偏高的產業，在新藥開發的歷程中，臨床試驗是開發中藥品自「試驗階段」邁向「市場階段」的絕對關鍵過程，其重要性不言而喻。對藥廠而言，臨床試驗若能有效率地被執行，一方面可縮短試驗時間減少花費；另一方面則是搶得先機佔據市場，及早實現獲利。因應臨床試驗重要且繁複需求的臨床試驗服務公司（Contract Research Organization, CRO）便應蘊而生。 CRO產業在美國的發展已逾40年，在奠基於過去的競爭基礎之外，其CRO產業已朝向「便利性」與「客製化」等利基市場佈局，以滿足新的顧客價值主張，並創新商業模式，追求新的成長機會。本研究在介紹美國CRO之產業市場營運模式之外，也以個案分析方式，自國內CRO公司與藥廠/生技公司的互動、合作，探討CRO的營運模式是否符合客戶需求，並嘗試以<白地策略>書中四個核心市場要素：顧客價值主張、利潤公式、關鍵流程與關鍵資源，思索CRO公司應如何經營市場白地。希望借鏡國外CRO公司的演進，反饋予國內CRO產業的未來發展及策略調整參考。 / Bio-pharmaceutical industry on human health well-being highly influential, but also a industry of technical, capital intensive, long development periods, and high risk. Among the course of drug development, clinical trials are absolute the key to the process, and advance the development process from the pilot phase to the market phase. For pharmaceutical companies, if the clinical trial can be executed efficiently, they can shorten the test time and spending; seize the market in advance, and reap profits as soon as possible. In response to the important and complex clinical trial requirements, the Contract Research Organization (CRO) prospers and takes advantage of a favorable situation. The evolvement of the CRO industry in US is more than 40 years. In the foundation base in the past, the CRO industry has been towards convenience and customization and other niche market distribution, to meet new customer value propositions, innovative business models and pursue new growth opportunities. In this thesis, the author not only introduces the CRO’s business models in US, but also explores the CRO’s business models in Taiwan by way of case studies. Through the interaction and cooperation between domestic pharmaceutical companies and CROs, the author wants to find out if the business models of CROs are in line with customer needs. The author also wants to quest if the CROs can learn how to manage and operate a white space which the CROs hope to seize by way of the book Seizing the White Space lists four fundamental building blocks including customer value proposition, profit formula, key resources, and key processes that make a company business model works. The domestic CROs can adjust their strategy and business model for fitting customer’s value proposition.

生技產業為我國未來六大明星產業之一，除政府成立生技創投基金，民間企業也陸續加入這波生技投資行列，如永豐餘集團旗下的上智生技創投，與潤泰集團旗下的鑽石生技投資。以籌資來源而言，又分為借款融資關係(負債端)的外部資金，以及股東投資關係(權益端)的自有資金兩種，對於公司經營各有優缺點，亦應取得平衡。唯目前多數為權益端的資金投入，尤其以該產業中風險最高的新藥研發公司為例，仍普遍高達95%以上的股東權益比率。顯示其籌資來源有限，且難以吸引負債端的投資者參與。而這樣的資金來源比例，除不符合企業融資順位理論於公司成長階段的籌資策略與負債權益比率，權益端資金多以短期獲得高利潤為目的，以資金性質亦不適合占資產達95%以上之比例。 以目前負債端籌資管道，新藥研發公司多數利用台灣中小企業信用保證基金直保部或經濟部促進產業創新或研究發展貸款計畫專案申請，唯融資額度上限遠不足以支付藥物開發費用，且非一般負債端直接經由銀行評估取得融資之方式。綜觀國際業態，單一全新藥物開發至上市平均需約USD8億元(約NTD240億元)不等，而台灣公司的研發策略多數為分段發展或老藥新用(藥物重新定位)策略，但仍有高度資金需求。唯銀行、負債端投資者普遍缺乏投入該產業的意願，主要顧慮為具冗長的產品研發週期業態、高度不確定性的產品上市審查、長期臨床試驗伴隨的高額成本。此外，對於資金專注研發之新藥研發公司，亦面臨擔保品不足之問題。而實務上，負債端資金提供者如銀行，對於複雜的生技領域與新藥研發公司業態不甚了解，為降低融資意願的另一主因。 故本研究旨在建立一套適用於新藥研發公司之一般性價值評估方式，解決此雙方認知差異問題，以增加更多元的籌資管道。其中，本文參考其他國家評估方法，選擇其中針對技術型公司、發展久遠的韓國技術信用保證基金KOTEC評估模式，導入台灣微脂體、基亞生物科技、賽德醫藥科技3間新藥研發公司個案作一評估。並於最後研究結論，經由分析比較個案公司間歷年經營狀況，得出公司整體與個別質、量性指標項目量化的相對分數，以台灣微脂體分數157分最高，基亞生技次之。本研究亦參考個案評估狀況，得出該類公司較佳的一般性經營策略結論，發現公司創立早期可先以開發週期短、風險較低的老藥新用開發以代替副業產生短期營收的效用，同時累積本業開發經驗，待時機成熟再轉入全新藥物開發為一攻守兼具的經營模式，以供新藥研發公司參考。此外，本研究屬於探索性研究，僅於評估新藥研發公司分數階段，尚未轉換為公司融資評等。該部分尚待具一定案源量後，以統計模型將評估分數與還款違約率關聯性做一分析，方能計算融資評等。而建立內部評等模型、資訊系統對台灣銀行規模而言，為一額外高昂成本，亦建議可效法韓國由政府主導為可行方式之一。 / The biotechnology industry is one of the six future stars of the industries in Taiwan. The government established Biotechnology Venture Capital (BVC), and the more and more private companies joined the procession of biotech investments, such as the two famous biotech funds, Taiwan Global BioFund (TGB) and Diamond BioFund Inc.. According to sources of funding, we can divided them into two groups: one is the loan of external funds (liability side), and the other is the shareholder investment of internal funds (equity side), both of them have different advantages and disadvantages for the company, and the company should strike a balance between these advantages and disadvantages. However, the majority of the funds are invested from the equity side, especially the new drug development companies, which are the highest risk types in the industry, and most of their equity ratio is higher than 95 %. This information indicates the limited sources of funding, and the difficulty to attract liability side’s investors to participate. That proportion of funding sources doesn’t comply with the company’s financing strategy and debt to equity ratio in the growth stage of the enterprise life cycle in the pecking order theory, and equity side’s funds are not suitable for accounting for more than 95% of assets in balance sheet because most of them want to get high profits in the short-term. Currently, major new drug development companies usually apply for loans from the Direct Guarantee Dept. of the Small & Medium Enterprise Credit Guarantee Fund of Taiwan (Taiwan SMEG) or the Promote Industrial Innovation or R&D Loan Program of Industrial Development Bureau in Taiwan, but the amount of loan is insufficient to cover the costs for the new drug development, and this method is not a general way to obtain liability side’s financing from the bank’s direct evaluation. In the international situation, the progress from development to sale of a single new drug spends about US $800 million (about NT $24 billion) on average. Despite Taiwan's R&D strategies only cover the sectional development progress or the policy of the new usage of old drugs (drug repositioning), there is still a high degree of capital requirement. However, in the present, banks and other liability side’s investors still lack the will to invest in the new drug development companies. These investors concern about several major problems, including the lengthy product development cycle, high uncertainty of the product examination and approval, the high cost of long-term clinical trials in this industry. In addition, these companies are also faced with the problem of lacking collateral, because they invest much money in new drug R&D. On the other hand, liability side’s investors, such as banks, don’t understand the complex field of new drug development companies' business models, and this situation becomes another reason for reducing the financing will. Therefore, we should establish a general evaluation method applicable to new drug development companies, to solve the problem of cognitive differences between liability side’s investors and the borrowers, and expand the funding sources of these companies. This article refers to the actual evaluation method in other countries, chooses the most suitable and well developed evaluation model --- Korea Technology Finance Corporation (KOTEC)’s evaluation method for the technology-based company, and utilizes the method to evaluate three cases of the new drug development companies in Taiwan, including Taiwan Liposome Co., Medigen Biotechnology Crop., and CytoPharm, Inc.. In conclusion of the study, by analyzing and comparing the three companies’ operating situations in recent years, we can get relative quantified scores from the companies’ overall and individual qualitative, quantitative indicators, and the result is that Taiwan Liposome Co. gets the highest score, 157 points, then Medigen Biotechnology Crop. gets the middle one. This study also refers the case situations, to find a better general business strategy for such companies. We find that new drug development company in the early stage can focus on new usage development of old drugs ,which has advantages of short development cycle and lower risk, to replace the sideline that generates short-term revenue, and accumulate the experience of drug development. When the time is ripe, it can transfer to new drug development. This way is the general suggestion of both offensive and defensive business model for new drug development companies. In addition, this study is an exploratory research, which only focuses on the evaluation stage, and has not converted the result into a corporate financing credit rating. To calculate financing credit ratings, we require a lot of historical cases data to establish a statistical analysis model, and link evaluation scores with repayment default rates. The establishment of an internal rating model or information system incurs high additional costs for the size of the banks in Taiwan, so the recommended one of the possible ways is that we can follow the example led by the South Korea Government.

The purpose of this diploma thesis is to understand financing of drug development from an enterprise risk management perspective as well as to critically assess the efficiency of the ISO framework and risk management techniques used for determining whether to fund drug development or not. The diploma thesis is divided into theoretical and practical part. The first part starts with perception and assessment of uncertainty and risk in the past. It describes how risk-averse individuals attempted to deal with uncertainty and different risk. This is followed by the evolution of traditional risk management into the fast developing enterprise risk management. The text further analyses commonly used risk management standards COSO ERM and ISO 31000:2009. However, the main focus is on the critical assessment of analytical tools which are frequently used for evaluating and assessing risks, especially financial ones, during drug development. The theoretical part is finished by a drug development process, whose phases are briefly described. The practical part was written in co-operation with AstraZeneca, a top-notch pharmaceutical company. The overview of its business is preceded by an explanation of current issues in the pharmaceutical industry. Furthermore, the risk analysis is conducted with respect to the ISO framework. Subsequently, selected risk assessment techniques are applied on the simplified financial model of two different drugs, which was created based on AstraZeneca's real data. These risk assessment tools are used in different phases of drug development so it could be seen clearly how the results are changing during a project. The outcomes of this risk analysis are compared with original plans used by AstraZeneca which were used for deciding whether to fund drug development or not.

Les relations entre la structure des composés chimiques et leurs propriétés sont complexes et à haute dimension. Dans le processus de développement de médicaments, plusieurs proprié- tés d’un composé doivent souvent être optimisées simultanément, ce qui complique encore la tâche. Ce travail explore deux représentations des composés chimiques pour les tâches de prédiction des propriétés. L’objectif de ces représentations proposées est d’améliorer l’explicabilité afin de faciliter le processus d’optimisation des propriétés des composés. Pre- mièrement, nous décomposons l’algorithme ECFP (Extended connectivity Fingerprint) et le rendons plus simple pour la compréhension humaine. Nous remplaçons une fonction de hachage sujet aux collisions par une relation univoque de sous structure à bit. Nous consta- tons que ce changement ne se traduit pas par une meilleure performance prédictive d’un perceptron multicouche par rapport à l’ECFP. Toutefois, si la capacité du prédicteur est ra- menée à celle d’un prédicteur linéaire, ses performances sont meilleures que celles de l’ECFP. Deuxièmement, nous appliquons l’apprentissage automatique à l’analyse des paires molécu- laires appariées (MMPA), un paradigme de conception du développement de médicaments. La MMPA compare des paires de composés très similaires, dont la structure diffère par une modification sur un site. Nous formons des modèles de prédiction sur des paires de com- posés afin de prédire les différences d’activité. Nous utilisons des contraintes de similarité par paires comme MMPA, mais nous utilisons également des paires échantillonnées de façon aléatoire pour entraîner les modèles. Nous constatons que les modèles sont plus performants sur des paires choisies au hasard que sur des paires avec des contraintes de similarité strictes. Cependant, les meilleurs modèles par paires ne sont pas capables de battre les performances de prédiction du modèle simple de base. Ces deux études, RCFP et comparaisons par paires, visent à aborder la prédiction des propriétés d’une manière plus compréhensible. En utili- sant l’intuition et l’expérience des chimistes médicinaux dans le cadre de la modélisation prédictive, nous espérons encourager l’explicabilité en tant que composante nécessaire des modèles cheminformatiques prédictifs. / The relationships between the structure of chemical compounds and their properties are complex and high dimensional. In the drug development process, multiple properties of a compound often need to be optimized simultaneously, further complicating the task. This work explores two representations of chemical compounds for property prediction tasks. The goal of these suggested representations is improved explainability to better understand the compound property optimization process. First, we decompose the Extended Connectivity Fingerprint (ECFP) algorithm and make it more straightforward for human understanding. We replace a collision-prone hash function with a one-to-one substructure-to-bit relationship. We find that this change which does not translate to higher predictive performance of a multi- layer perceptron compared to ECFP. However, if the capacity of the predictor is lowered to that of a linear predictor, it does perform better than ECFP. Second, we apply machine learning to Matched Molecular Pair Analysis (MMPA), a drug development design paradigm. MMPA compares pairs of highly similar compounds, differing in structure by modification at one site. We train prediction models on pairs of compounds to predict differences in activity. We use pairwise similarity constraints like MMPA, but also use randomly sampled pairs to train the models. We find that models perform better on randomly chosen pairs than on pairs with strict similarity constraints. However, the best pairwise models are not able to beat the prediction performance of the simpler baseline single model. Both of these investigations, RCFP and pairwise comparisons, aim to approach property prediction in a more explainable way. By using intuition and experience of medicinal chemists within predictive modelling, we hope to encourage explainability as a necessary component of predictive cheminformatic models.

Indiana University-Purdue University Indianapolis (IUPUI) / Signal transducer and activator of transcription 3 (STAT3) is constitutively activated in malignant tumors, and its activation is associated with high histological grade and advanced cancer stage. STAT3 has been shown to play important roles in multiple aspects of cancer aggressiveness including proliferation, survival, self-renewal, migration, invasion, angiogenesis and immune response by regulating the expression of diverse downstream target genes. Thus, inhibiting STAT3 promises to be an attractive strategy for treatment of advanced tumors with metastatic potential. We firstly identified a STAT3 inhibitor, inS3-54, by targeting the DNA-binding site of STAT3 using an in-silico screening approach; however, inS3-54 was finally found not to be appropriate for further studies because of low specificity on STAT3 and poor absorption in mice. To develop an effective and specific STAT3 inhibitor, we identified 89 analogues for the structure-activity relationship analysis. By using hematopoietic progenitor cells isolated from wild-type and STAT3 conditional knockout mice, further studies showed that three analogues (A18, A26 and A69) only inhibited STAT3-dependent colony formation of hematopoietic progenitor cells, indicating a higher selectivity for STAT3 than their parental compound, inS3-54. These compounds were found to (1) inhibit STAT3-specific DNA binding activity; (2) bind to STAT3 protein; (3) suppress proliferation of cancer cells harboring aberrant STAT3 signaling; (4) inhibit migration and invasion of cancer cells and (5) inhibit STAT3-dependent expression of downstream targets by blocking the binding of STAT3 to the promoter regions of responsive genes in cells. In addition, A18 can reduce tumor growth in a mouse xenograft model of lung cancer with little effect on body weight. Taken together, we conclude that it is feasible to inhibit STAT3 by targeting its DNA-binding domain for discovery of anticancer therapeutics.

Indiana University-Purdue University Indianapolis (IUPUI) / Traditional treatment strategy development for diseases involves the identification of target proteins related to disease states, and the interference of these proteins with drug molecules. Computational drug discovery and virtual screening from thousands of chemical compounds have accelerated this process. The thesis presents a comprehensive framework of computational drug discovery using system biology approaches. The thesis mainly consists of two parts: disease biomarker identification and disease treatment discoveries. The first part of the thesis focuses on the research in biomarker identification for human diseases in the post-genomic era with an emphasis in system biology approaches such as using the protein interaction networks. There are two major types of biomarkers: Diagnostic Biomarker is expected to detect a given type of disease in an individual with both high sensitivity and specificity; Predictive Biomarker serves to predict drug response before treatment is started. Both are essential before we even start seeking any treatment for the patients. In this part, we first studied how the coverage of the disease genes, the protein interaction quality, and gene ranking strategies can affect the identification of disease genes. Second, we addressed the challenge of constructing a central database to collect the system level data such as protein interaction, pathway, etc. Finally, we built case studies for biomarker identification for using dabetes as a case study. The second part of the thesis mainly addresses how to find treatments after disease identification. It specifically focuses on computational drug repositioning due to its low lost, few translational issues and other benefits. First, we described how to implement literature mining approaches to build the disease-protein-drug connectivity map and demonstrated its superior performances compared to other existing applications. Second, we presented a valuable drug-protein directionality database which filled the research gap of lacking alternatives for the experimental CMAP in computational drug discovery field. We also extended the correlation based ranking algorithms by including the underlying topology among proteins. Finally, we demonstrated how to study drug repositioning beyond genomic level and from one dimension to two dimensions with clinical side effect as prediction features.

Indiana University-Purdue University Indianapolis (IUPUI) / ALDH isoenzymes are known to impact the sensitivity of certain neoplastic cells toward cyclophosphamides and its analogs. Despite its bone marrow toxicity, cyclophos-phamide is still used to treat various recalcitrant forms of cancer. When activated, cyclo-phosphamide forms aldophosphamide that can spontaneously form the toxic phospho-ramide mustard, an alkylating agent unless detoxified by ALDH isozymes to the carbox-yphosphamide metabolite. Prior work has demonstrated that the ALDH1A1 and ALDH3A1 isoenzymes can convert aldophosphamide to carboxyphosphamide. This has also been verified by over expression and siRNA knockdown studies. Selective small molecule inhibitors for these ALDH isoenzymes are not currently available. We hypothe-sized that novel and selective small molecule inhibitors of ALDH3A1 would enhance cancer cells’ sensitivity toward cyclophosphamide. If successful, this approach can widen the therapeutic treatment window for cyclophosphamides; permitting lower effective dos-ing regimens with reduced toxicity. An esterase based absorbance assay was optimized in a high throughput setting and 101, 000 compounds were screened and two new selective inhibitors for ALDH3A1, which have IC50 values of 0.2 µM (CB7) and 16 µM (CB29) were discovered. These two compounds compete for aldehyde binding, which was vali-dated both by kinetic and crystallographic studies. Structure activity relationship dataset has helped us determine the basis of potency and selectivity of these compounds towards ALDH3A1 activity. Our data is further supported by mafosfamide (an analog of cyclo-phosphamide) chemosensitivity data, performed on lung adenocarcinoma (A549) and gli-oblastoma (SF767) cell lines. Overall, I have identified two compounds, which inhibit ALDH3A1’s dehydrogenase activity selectively and increases sensitization of ALDH3A1 positive cells to aldophosphamide and its analogs. This may have the potential in improving chemotherapeutic efficacy of cyclophosphamide as well as to help us understand better the role of ALDH3A1 in cells. Future work will focus on testing these compounds on other cancer cell lines that involve ALDH3A1 expression as a mode of chemoresistance.