Chemotherapy plays an important role in the treatment of cancer. However, these drugs also cause death of non-malignant cells, resulting in severe side-effects. In addition, drug resistance may exist. Predictive tools for dose and drug selection are therefore warranted. In this thesis predictive pharmacometric models were developed for the main dose-limiting side-effect, neutropenia, and for treatment response following chemotherapy. Neutropenia is associated with a high risk for life-threatening infections and leads frequently to reduced dose delivery and thereby suboptimal treatment of the tumor. A better characterization of the dynamics of docetaxel induced neutropenia was obtained by simultaneous analysis of neutrophils and leukocytes. The fraction of neutrophils was shown to change over the time-course, hence leukocytes and neutrophil counts are not interchangeable biomarkers. Sometimes neutrophil count is reported as categorical severity of neutropenia (Grade 0-4). A method was developed that allowed analysis of these data closer to its true continuous nature. The main regulatory hormone of neutrophils is granulocyte colony stimulating factor (G-CSF). Although recombinant G-CSF is used as supportive therapy to prevent neutropenia, little is known of how the endogenous G-CSF concentrations vary in patients following chemotherapy. A prospective study was carried out and simultaneous analysis of endogenous G-CSF and neutrophils following chemotherapy enabled a more mechanistic model to be developed that also could verify the self-regulatory properties of the physiological system. Patient characteristics were investigated using a pharmacokinetic-myelosuppression model for docetaxel in patients with normal and impaired liver function. The model was a useful tool in evaluating different dosing strategies and a reduced dosing scheme was suggested in patients with poor liver function, thereby enabling docetaxel treatment in this patient population which has previously been excluded. Treatment of acute myeloid leukemia with daunorubicin and cytarabine is associated with drug resistance and high variability in pharmacokinetics, which was partly explained for daunorubicin by peripheral leukocyte count. An integrated model of the in vitro drug sensitivity and treatment response showed that in vitro drug sensitivity was predictive for treatment outcome in this patient population and may therefore be used for choice of drug. The developed pharmacometric models in this thesis may be useful in the optimization of treatments schedules for existing and new drugs as well as to assist in drug and dose selection to improve therapy in an individual patient. The models and methods presented may also facilitate pooled analysis of data and demonstrate principles which could be useful for the pharmacometric community.

Sunitinib, an oral vascular endothelial growth factor receptor, is a first-line option for metastatic renal cell carcinoma and widely used in clinical practice. Despite the proven benefit of sunitnib in metastatic renal cell carcinoma, patients may suffer from a variety of adverse events including hypertension, fatigue, hypothyroidism, hand?foot skin reactions, rash, depigmentation, and myelosuppression. Myelosuppression is usually mild, transient and resolves during the two weeks at the end of each cycle where no drug is taken. We present a case of severe and early grade 3 neutropenia and thrombocytopenia occurring two weeks into a six-week cycle. Because of the extreme nature of the toxicity, CYP 3A4 polymorphisms were explored. The patient was found to be heterozygous for CYP 3A4*22, at least partially explaining the early-onset and severity of myelosuppression. This pharmacogenetics information resulted in a rechallenge of dose-reduced sunitinib, which was well tolerated by the patient. The current state of pharmacogenomics concerning sunitinb is also presented, and the need for greater research in this area is highlighted.

Paclitaxel (Taxol®) is now widely used against breast, ovarian and non-small-cell lung cancer. Anticancer agents generally have narrow therapeutic indices, often with myelosuppression (mainly neutropenia) as dose-limiting side effect. A further complicating factor is that paclitaxel when given as Taxol® has a nonlinear pharmacokinetic (PK) behaviour in plasma. Identifying risk groups more sensitive to chemotherapy due to either a PK or pharmacodynamic (PD) interindividual variability is of importance. The aim of the thesis was to develop predictive mechanism-based PK and PD models applicable for paclitaxel. PK and PK/PD models were developed for patient data from studies with relatively frequent sampling or sparse sampling schedules. Population analyses were performed using the software NONMEM. A pharmacokinetic model describing unbound, total plasma and blood concentrations of paclitaxel from known binding mechanisms was developed and validated. The nonlinear PK in plasma could to a large extent be explained by the micelle forming vehicle Cremophor EL (CrEL) and the unbound drug showed linear PK. Besides a binding component directly proportional to concentrations of CrEL, the model included both linear and nonlinear binding components in plasma and blood. Further, relations between the PK parameters and different demographic factors, including polymorphisms in the cytochrome P450s involved in paclitaxel metabolism, were investigated. A semi-physiological PD model for chemotherapy-induced myelosuppression was developed and applied to different anticancer drugs. The model included a self-renewal for proliferating cells, transit compartments describing the delay in observed myelosuppression and a feedback parameter reflecting the effect on the bone marrow from growth factors that can result in an overshoot in white blood cells. The system-related parameters estimated showed consistency across drugs and the difference in the drug-related parameter reflected the relative bone marrow toxicity of the drugs. Relations between demographic factors and the PD parameters were identified. The developed mechanism-based models promote a better understanding of paclitaxel PK and PD and may be used as tools in dosing individualisation and in development of dosing strategies for new administration forms and new drugs in the same area.

Hematopoiesis is the highly regenerative process of producing billions of blood cells each day, including white blood cells, red blood cells, and platelets. Given the relatively short life span of these mature cells, hematopoiesis is dependent on stem and progenitor cells to generate renewed progeny, which represents a tightly regulated process. This includes cell intrinsic and external factors, and where dysregulation can lead to anemia and cancer. As such, the hematopoietic hierarchy has been intensely studied for nearly a century and represents a gold standard model of cell fate and developmental biology, in research and clinical applications. Cellular models, such as in vitro culture and human-mouse xenografts in vivo, have been developed to explain complex phenomena pertaining to hematopoiesis and also interrogate processes which are too invasive to study in humans. Hematopoietic generation is required beyond sustaining homeostasis, and progenitors can be damaged through cytotoxic injuries such as radiation and standard chemotherapy, and also undergo leukemic transformation. There are two main treatment modalities for leukemia patients (a) receiving a stem cell transplant, and (b) drug or radiation-based therapy. In the former, shortages of donors and stem cells has remained an unmet clinical need for decades. In the latter, selective targeting of genetic mutations has become a successful standard-of-care in leukemias such as chronic myelogenous leukemia and acute promyelocytic leukemia. However, in the most common adult hematologic malignancy, chronic lymphocytic leukemia (CLL), similar targeting therapies have not been developed. Altogether, shortages of stem cells from healthy donors, chemotherapy-induced immune dysfunction, and a lack of targeted therapies, all reinforce the immediate need for innovative cellular models to address these clinical problems. To generate additional sources of human hematopoietic progenitors for laboratory study, human PSCs have been used. Unlike hematopoietic progenitor cells collected from healthy and leukemic donors, human pluripotent stem cells (PSC) can be easily propagated and expanded in vitro. PSCs can generate hematopoietic progenitor cells, but they remain poorly understood and have not been robustly applied to solve the aforementioned deficiencies related to patient treatment. Importantly, the biological regulation of both hematopoiesis and PSCs has been experimentally confirmed to significantly deviate between humans and other animals, such as mice, further reinforcing the importance of human-specific cell models of hematopoiesis. Therefore, I hypothesized that human stem cell models provide a focused approach to interrogate the regulation of hematopoiesis from the apex of the hierarchy, which can be used to understand the promotion of healthy hematopoiesis and understand malignant transformation. Collectively, the data presented within this thesis offer a deeper conceptualization of human stem cell models and the deconvolution of several complex components of hematopoietic regulation. This work has revealed novel, clinically relevant, and actionable targets to ultimately enable the promotion of healthy hematopoiesis on multiple fronts. / Thesis / Doctor of Philosophy (PhD) / This thesis presents research on novel molecular and genetic regulatory pathways of self-renewal and differentiation in models of healthy and malignant human hematopoiesis. The origin of healthy hematopoietic regulation stems from a large body of work spanning decades and encompasses many efforts by others to derive hematopoietic stem cells from human pluripotent cells. The development of a genetic model for the malignant regulation of CLL was truly serendipitous, was propelled through robust and intriguing results that begged for further exploration, and filled a clinical gap in identifying actionable targets in CLL. Lastly, these two projects, along with my supportive roles in other published works throughout my graduate studies, instructed me to develop a human-mouse transplant model to uncover the biology of regenerating healthy hematopoiesis during injury.

Paclitaxel is one of the most successful anti-cancer drugs, particularly in the treatment of breast cancer and ovarian cancer. For the investigation of the interaction between paclitaxel and MD-2 protein, and development of new antagonists for lipopolysaccharide, several C10 A-nor-paclitaxel analogs have been synthesized and their biological activities have been evaluated. In order to reduce the myelosuppression effect of the paclitaxel, several C3â ² and C4 paclitaxel analogs have been synthesized and their biological evaluation have been studied. / Master of Science

General attrition rates in drug development pipeline have been recognized as a necessity to shift gears towards new methodologies that allow earlier and correct decisions, and the optimal use of all information accrued throughout the process. The quantitative science of pharmacometrics using pharmacokinetic-pharmacodynamic models was identified as one of the strategies core to this renaissance. Coupled with Optimal Design (OD), they constitute together an attractive toolkit to usher more rapidly and successfully new agents to marketing approval. The general aim of this thesis was to investigate how the use of novel pharmacometric methodologies can improve the design and analysis of clinical trials within drug development. The implementation of a Monte-Carlo Mapped power method permitted to rapidly generate multiple hypotheses and to adequately compute the corresponding sample size within 1% of the time usually necessary in more traditional model-based power assessment. Allowing statistical inference across all data available and the integration of mechanistic interpretation of the models, the performance of this new methodology in proof-of-concept and dose-finding trials highlighted the possibility to reduce drastically the number of healthy volunteers and patients exposed to experimental drugs. This thesis furthermore addressed the benefits of OD in planning trials with bio analytical limits and toxicity constraints, through the development of novel optimality criteria that foremost pinpoint information and safety aspects. The use of these methodologies showed better estimation properties and robustness for the ensuing data analysis and reduced the number of patients exposed to severe toxicity by 7-fold.  Finally, predictive tools for maximum tolerated dose selection in Phase I oncology trials were explored for a combination therapy characterized by main dose-limiting hematological toxicity. In this example, Bayesian and model-based approaches provided the incentive to a paradigm change away from the traditional rule-based “3+3” design algorithm. Throughout this thesis several examples have shown the possibility of streamlining clinical trials with more model-based design and analysis supports. Ultimately, efficient use of the data can elevate the probability of a successful trial and increase paramount ethical conduct.