Pharmacodynamic processes and disease progression are increasingly characterized with pharmacometric models. However, modelling options for discrete-type responses remain limited, although these response variables are commonly encountered clinical endpoints. Types of data defined as discrete data are generally ordinal, e.g. symptom severity, count, i.e. event frequency, and time-to-event, i.e. event occurrence. Underlying assumptions accompanying discrete data models need investigation and possibly adaptations in order to expand their use. Moreover, because these models are highly non-linear, estimation with linearization-based maximum likelihood methods may be biased. The aim of this thesis was to explore pharmacometric methods and novel models for discrete data through (i) the investigation of benefits of treating discrete data with different modelling approaches, (ii) evaluations of the performance of several estimation methods for discrete models, and (iii) the development of novel models for the handling of complex discrete data recorded during (pre-)clinical studies. A simulation study indicated that approaches such as a truncated Poisson model and a logit-transformed continuous model were adequate for treating ordinal data ranked on a 0-10 scale. Features that handled serial correlation and underdispersion were developed for the models to subsequently fit real pain scores. The performance of nine estimation methods was studied for dose-response continuous models. Other types of serially correlated count models were studied for the analysis of overdispersed data represented by the number of epilepsy seizures per day. For these types of models, the commonly used Laplace estimation method presented a bias, whereas the adaptive Gaussian quadrature method did not. Count models were also compared to repeated time-to-event models when the exact time of gastroesophageal symptom occurrence was known. Two new model structures handling repeated time-to-categorical events, i.e. events with an ordinal severity aspect, were introduced. Laplace and two expectation-maximisation estimation methods were found to be performing well for frequent repeated time-to-event models. In conclusion, this thesis presents approaches, estimation methods, and diagnostics adapted for treating discrete data. Novel models and diagnostics were developed when lacking and applied to biological observations.

INTRODUÇÃO: A combinação de sepse e insuficiência renal com necessidade de diálise é bastante comum nas Unidades de Terapia Intensiva e esta situação tem elevada mortalidade. Um desafio neste cenário é prescrever a dose correta dos antimicrobianos para o tratamento destas infecções. Em pacientes críticos e hemodinamicamente instáveis que necessitam de terapia renal substitutiva, um dos métodos mais utilizados é a diálise contínua, entretanto, recentemente, tem-se utilizado a diálise de baixa eficiência - conhecida como SLED, da abreviação do inglês \"sustained low-efficiency dialysis\". Esta modalidade de terapia renal substitutiva combina características da hemodiálise contínua com a intermitente, utilizando o equipamento da diálise intermitente, com menores fluxos sanguíneos e de dialisato, e com vantagem de menor custo. Apesar do fluxo mais baixo, por ser utilizado tempo mais prolongado, a SLED frequentemente resulta em maior clearance e especula-se que a remoção dos fármacos seria maior. Há escassez de estudos que avaliaram a farmacocinética e farmacodinâmica de antimicrobianos em pacientes submetidos à SLED.OBJETIVOS: Avaliar adequação farmacodinâmica de meropenem e vancomicina em pacientes submetidos a diálise estendida de baixa eficiência. Avaliar a depuração paramêtros farmacocinéticos durante a sessão de SLED. MÉTODOS: Foi realizado estudo prospectivo descritivo observacional com coleta de material biológico julho de 2012 a julho de 2014 HC-FMUSP. Foram incluídos pacientes submetidos à SLED em uso de vancomicina e/ou meropenem. Foram coletadas amostras de sangue seriadas (tempos: imediatamente antes do início da sessão de diálise, 0,5h, 1h, 2h, 4h após o início do tratamento e ao final da sessão). A quantificação dos antimicrobianos foi realizada através dos métodos analíticos de quantificação em Cromatografia Líquida de Alta Eficiência (CLAE). Os parâmetros farmacocinéticos foram calculados apenas durante a sessão de diálise utilizando-se o software WinNonlin. A área sob a curva foi determinada para a vancomicina. Para o meropenem, calculou-se o tempo acima da MIC. Resultados: Foram incluídos 24 pacientes tratados com vancomicina e 21 com meropenem eforam obtidas 170 amostras de plasma. As concentrações médias de vancomicina sérica e meropenem: antes da sessão de SLED foram 24,5 e 28,0 ?g / ml, respectivamente; e após SLED 14 e 6 ?g / ml, respectivamente. A depuração média foi de 41% para a vancomicina e 78% para o meropenem. Para vancomicina, 22 (96%), 19 (83%) e 16 (70%) pacientes teriam atingido o alvo (AUC0-24 > 400) considerando-se MIC 0,5; <= 1mg/l e <= 2 mg/l respectivamente. Para meropenem, 19 (95%), 18 (90%) e 11 (55%) pacientes teriam atingido a meta (70% de tempo acima da CIM) se infectados com isolados com MIC <= 1, <= 4 e <= 8 mg/l, respectivamente. Conclusões: Em pacientes críticos, meropenem evancomicina foram removidas durante o SLED. Entretando, a maioria dos pacientes alcançaria alvo PK-PD, excepto para CIMs mais altas. Sugerimos doses de manutenção de 1g a cada 12 ou 8 horaspara meropenem. Para a vancomicina, deve-se utilizar abordagem mais individualizada com monitorização sérica, uma vez que ensaios comerciais são disponíveis / Background: Antibiotic dosing is a challenge in critically ill patients undergoing renal replacement therapy. Our aim was to evaluate pharmacokinetics and pharmacodynamics of meropenem and vancomycin in patients undergoing SLED.Methods: ICU patients undergoing SLED, receiving meropenem and/or vancomycin, were prospectively evaluated. Blood samples were collected at the start of SLED and 0.5; 1; 2; 4 and 6 hours later. Antimicrobial levels were determined by HPLC. Noncompartimental pharmacokinetic analysis was performed. Area under the curve was determined for vancomycin. For meropenem, time above MIC was calculated. Results: 24 patients receiving vancomycin and 21 receiving meropenem were included; 170 plasma samples were obtained. Median serum vancomycin and meropenem concentrations: before SLED were 24.5 and 28.0 ?g/ml, respectively; and after SLED 14 and 6 ?g/ml, respectively. Mean removal was 41% for vancomycin and 78% for meropenem. For vancomycin, 22 (96%), 19(83%) and 16(70%) patients would have achieved the target (AUC0-24>400) considering MIC 0.5; <= 1mg/l and <= 2 mg/l, respectively. For meropenem, 19 (95%), 18 (90%) and 11(55%) patients would have achieved the target (70% of time above MIC) if infected with isolates with MIC <= 1, <= 4 and <= 8mg/l, respectively. Conclusions: In critically ill patients, meropenem and vancomycin were removed during SLED. Despite this, overall high PK/PD target attainment was obtained, except for higher MICs. We suggest maintenance doses of 1g tid or bid for meropenem. For vancomycin, more individualized approach using therapeutic drug monitoring should be used, as commercial assays are available

Durante a cirurgia de revascularização do miocárdio com circulação extracorpórea e hipotermia (CEC-H) ocorre alteração na efetividade do propofol e na sua farmacocinética realizada a partir das concentrações plasmáticas do propofol total no decurso do tempo. A ligação do propofol à proteína plasmática parece estar alterada em consequência de diversos fatores incluindo a hemodiluição e a heparinização que ocorre no início da circulação extracorpórea, uma vez que se reportou anteriormente que a concentração plasmática do propofol livre aumentou durante a realização da circulação extracorpórea normotérmica. Por outro lado, a infusão alvo controlada é recomendada para manter a concentração plasmática do propofol equivalente ao alvo de 2 &#181g/mL durante a intervenção cirúrgica com CEC-H. Se alterações significativas na hipnose do propofol ocorrem nesses pacientes, então o efeito aumentado desse agente hipnótico poderia estar relacionado à redução na extensão da ligação do fármaco as proteínas plasmáticas; entretanto, o assunto ainda permanece em discussão e necessita de investigações adicionais. Assim, o objetivo do estudo foi investigar as concentrações plasmáticas de propofol livre em pacientes durante a revascularização do miocárdio com e sem o procedimento de CEC-H através da abordagem PK-PD. Dezenove pacientes foram alocados e estratificados para realização de cirurgia de revascularização do miocárdio com circulação extracorpórea (CEC-H, n=10) ou sem circulação extracorpórea (NCEC, n=9). Os pacientes foram anestesiados com sufentanil e propofol alvo de 2 &#181g/mL. Realizou-se coleta seriada de sangue para estudo farmacocinético e o efeito foi monitorado através do índice bispectral (BIS) para medida da profundidade da hipnose no período desde a indução da anestesia até 12 horas após o término da infusão de propofol, em intervalos de tempo pré-determinados no protocolo de estudo. As concentrações plasmáticas foram determinadas através de método bioanalítico pela técnica de cromatografia líquida de alta eficiência. A farmacocinética foi investigada a partir da aplicação do modelo aberto de dois compartimentos, PK Solutions v. 2. A análise PK-PD foi realizada no Graph Pad Prisma v.5.0 após a escolha do modelo do efeito máximo (EMAX sigmóide, slope variável). Os dados foram analisados utilizando o Prisma v. 5.0, p<0,05, significância estatística. As concentrações plasmáticas de propofol total foram comparáveis nos dois grupos (CEC-H e NCEC); entretanto o grupo CEC-H evidenciou aumento na concentração do propofol livre de 2 a 5 vezes em função da redução na ligação do fármaco às proteínas plasmáticas. A farmacocinética do propofol livre mostrou diferença significativa entre os grupos no processo de distribuição pelo prolongamento da meia vida e aumento do volume aparente, e no processo de eliminação em função do aumento na depuração plasmática e redução na meia vida biológica no grupo CEC-H. A escolha do modelo EMAX sigmóide, slope variável foi adequada uma vez que se evidenciou alta correlação entre os valores do índice bispectral e as concentrações plasmáticas do propofol livre (r2>0.90, P<0.001) para os pacientes investigados. / During coronary artery bypass grafting (CABG) surgery with cardiopulmonary bypass (CPB) profound changes occur on propofol effect and on kinetic disposition related to total drug plasma measurements in these patients. It was reported that drug plasma binding could be altered as a consequence of hemodilution and heparinization before starts CPB since free propofol plasma levels was increased by twice under normothermic procedure. In addition, the target controlled infusion (TCI) is recommended to maintain propofol plasma concentration (2 &#181g/mL) during CABG CPB-H intervention. However, whether significant changes that occur in propofol hypnosis in these patients could be related to the reduction on the extension of drug plasma binding remain unclear and under discussion until now. Then, the objective of this study was to investigate propofol free plasma levels in patients undergoing CABG with and without CPB by a pharmacokinetics-pharmacodynamics (PK-PD) approach. Nineteen patients were scheduled for on-pump coronary artery bypass grafting (CABG-CPB, n=10) or off-pump coronary artery bypass grafting (OPCABG, n=9) were anesthetized with sufentanil and propofol TCI (2 &#181g/mL). Blood samples were collected for drug plasma measurements and BIS were applied to access the depth of hypnosis from the induction of anesthesia up to 12 hours after the end of propofol infusion, at predetermined intervals. Plasma drug concentrations were measured using high-performance liquid chromatography, followed by a propofol pharmacokinetic analysis based on two compartment open model, PK Solutions v.2; PK-PD analysis was performed by applying EMAX model, sigmoid shape-variable slope and data were analyzed using Prisma v. 5.0, considering p<0.05 as significant difference between groups. The total propofol plasma concentrations were comparable in both groups during CABG; however it was shown in CPB-group significant increases in propofol free plasma concentration by twice to fivefold occur as a consequence of drug plasma protein binding reduced in these patients. Pharmacokinetics of free propofol in CPB-H group compared to OPCAB group based on two compartment open model was significantly different by the prolongation of distribution half-life, increases on plasma clearance, and biological half-life shortened. In addition, the kinetic disposition of propofol changes in a different manner considering free drug levels in the CPB-H group against OPCAB group as follows: prolongation of distribution half-life and increases on volume of distribution, remaining unchanged biological half-life in spite of plasma clearance increased. BIS values showed a strong correlation with free drug levels (r2>0.90, P<0.001) in CPB-H group and also in OPCAB group by the chosen EMAX model sigmoid shape-variable slope analyzed by GraphPad Prisma v.5.0.

<p>Despite advancements in therapies, such as surgery, irradiation (IR) and chemotherapy, outcome for patients suffering from glioblastoma (GBM) remains fatal; the median survival time is only about 15 months. Even with novel therapeutic targets, networks and signaling pathways being discovered, monotherapy with such agents targeting such pathways has been disappointing in clinical trials. Poor prognosis for GBM can be attributed to several factors, including failure of drugs to cross the blood-brain-barrier (BBB), tumor heterogeneity, invasiveness, and angiogenesis. Development of tumor resistance, particularly to temozolomide (TMZ) and IR, creates a substantial clinical challenge.</p><p> </p><p>The primary focus of the work described herein was to develop a modeling and simulation approach that could be applied to rationally develop novel combination therapies and dose regimens that mitigate resistance development. Specifically, TMZ was combined with small molecule inhibitors that are either currently in clinical trials or are approved drugs for other cancer types, and which target the disease at various resistance signaling pathways that are induced in response to TMZ monotherapy. To accomplish this objective, an integrated PKPD modeling approach was used. A PK model for each drug was first defined. PK models were subsequently linked to a PD model description of tumor growth dynamics in the presence of a single drug or combinations of drugs. A key outcome of these combined PKPD models was tumor static concentration (TSC) curves of TMZ in combination with small molecule inhibitors that identify combination drug exposures predicted to arrest tumor growth. This approach was applied to TMZ in combination with abemaciclib (a dual CDK4/6 small molecule inhibitor) based on data from a published study evaluating abemaciclib (ACB) efficacy in combination with TMZ in a U87 GBM xenograft model. TSC was also constructed for TMZ in combination with RG7388 (MDM2 inhibitor) based on the data from an in-vivo study that evaluated effects on tumor growth suppression of these small molecule inhibitors in combination with TMZ in GBM 10 patient derived xenografts.</p><p>In GBM 43 mouse xenografts, emergence of resistance to TMZ treatment was identified. Thus, a resistance integrated PKPD model was developed to predict tumor growth kinetics after treatment with TMZ in GBM 43 tumors. Population PK models in immune deficient NOD.Cg-<em>Prkdc^scid Il2rg^tm1Wjl</em>/SzJ (NSG) mice for TMZ and small molecule inhibitors (GDC0068/RG7112) were developed based on a combination of data obtained from an in-vivo study and published sources. Subsequently, PK models were linked to tumor volume data obtained from GBM 43 subcutaneous xenografts. Model parameters quantifying tumor volume dynamics were precisely estimated (coefficient of variation < 40%) compared to a base tumor growth inhibition model in GBM 43 that did not incorporate resistance development. Graphical diagnostics of the resistance incorporated PKPD tumor growth inhibition model demonstrated a superior fit compared to the base model, and accurately captured the emergence of resistance to the TMZ monotherapy treatment observed in the GBM 43 patient derived xenograft model.</p>

Indiana University-Purdue University Indianapolis (IUPUI) / Pharmacodynamics (PD) studies the relationship between drug concentration and drug effect on target sites. This field has recently gained attention as studies involving PD Drug-Drug interactions (DDI) assure discovery of multi-targeted drug agents and novel efficacious drug combinations. A PD drug combination could be synergistic, additive or antagonistic depending upon the summed effect of the drug combination at a target site. The PD literature has grown immensely and most of its knowledge is dispersed across different scientific journals, thus the manual identification of PD DDI is a challenge. In order to support an automated means to extract PD DDI, we propose Pharmacodynamics Miner (PD-Miner). PD-Miner is a text-mining tool, which is capable of identifying PD DDI from in vitro PD experiments. It is powered by two major features, i.e., collection of full text articles and in vitro PD ontology. The in vitro PD ontology currently has four classes and more than hundred subclasses; based on these classes and subclasses the full text corpus is annotated. The annotated full text corpus forms a database of articles, which can be queried based upon drug keywords and ontology subclasses. Since the ontology covers term and concept meanings, the system is capable of formulating semantic queries. PD-Miner extracts in vitro PD DDI based upon references to cell lines and cell phenotypes. The results are in the form of fragments of sentences in which important concepts are visually highlighted. To determine the accuracy of the system, we used a gold standard of 5 expert curated articles. PD-Miner identified DDI with a recall of 75% and a precision of 46.55%. Along with the development of PD Miner, we also report development of a semantically annotated in vitro PD corpus. This corpus includes term and sentence level annotations and serves as a gold standard for future text mining.

Pharmacokinetic (PK)- and pharmacodynamic (PD) modeling are useful tools whenassessing treatment effect. A patient’s adherence can potentially be rate-limiting, since it isthe first process in a chain of processes that determines treatment effect. Therefore agreater system taking into consideration PKPD as well as adherence models couldpotentially unlock a greater system understanding. This study focuses on investigating thefeasibility of combining models concerning adherence, PK and PD. An extensive mapping of previously made work on the topics of PKPD model developmentand adherence models concerning type 2 diabetes was conducted. Results concluded thatthere are gaps in research regarding adequate adherence-scoring methods that easily can belinked to dosing regimens. Furthermore, there is lacking research regarding feedback fromexposure-response to adherence. A simple model was implemented to provide a proposedlinkage inhowthe connection could be made between adherence and a PKPD-model.Sensitivity analysis showed that the adherence scoring used (Summary of DiabetesSelf-Care Activities measure, SDSCA) had a moderate correlation to the final response onfasting plasma glucose (Spearman ρ=−0.478∗∗∗). This result suggests that adherenceshould be considered as a relatively important factor to weave in to systems models ofpharmacology and future research should be made on further developing modelsimplementing both social factors, such as adherence, as well as pharmacologic response. Apossible way could be linking dose regimen to adherence scoring. / Farmakokinetiska (PK)- och farmakodynamiska (PD) modeller är användbara verktyg vid utvärdering av effekten av en behandlingsplan. Patientens åtlydnad tilll läkemedelsordinationen kan potentiellt vara en begränsande faktor för behandlingsprocessen. Att utveckla större system som täcker farmakologiska- samt åtlydnadsmodeller skulle potentiellt kunna vara en väg till en förhöjd förståelse angående farmakologiska system. Denna studie fokuserar på att undersöka genomförbarheten av att koppla samman modeller angående farmakokinetik, farmakodynamik samt åtlydnadsmodeller. En omfattande kartläggning av tidigare utfört arbete angående utvecklingen av PKPD-modeller och åtlydnadsmodeller som utgick fr ̊an typ 2 diabetes utfördes. Resultatet av studien visade en avsaknad av forskning gällande definieringen och kvantifieringen avhur man mäter åtlydnad för simuleringssyften. Ytterligare saknades det forskning rörande system med återkoppling från farmakologiska segment av ett system tillbaka till åtlydnadsdelarna. En enkel modell implementerades som ett förslag till hur en potentiell sammankoppling skulle kunna utföras. En känslighetsanalys utfördes och visade att poängskattningen för åtlydnad, SDSCA (Summary of Diabetes Self-Care Activities), hade en måttlig korrelation (Spearman ρ=−0.478∗∗∗) till den slutgiltiga koncentrationen av glukos i plasma. Detta resultat innebär att åtlydnad har en koppling till förbättrandet av hyperglykemi och bör därför inte exkluderas vid framtida utveckling av modeller för farmakologi. En länk skulle kunna vara kopplingen mellan ”åtlydnads-poäng” och doseringsregim.

Mise en contexte : Le décès neurologique correspond à la perte permanente des fonctions cérébrales. La validité de son diagnostic repose sur des conditions préalables et l’exclusion des facteurs confondants. L’objectif du mémoire est d’investiguer le rôle potentiellement confondant des médicaments sédatifs et analgésiques (MSA) sur le diagnostic de décès neurologique (DDN). Méthodes : 1) Nous avons réalisé une étude de la portée des recommandations mondiales s’intéressant au rôle des médicaments sédatifs et analgésiques lors du diagnostic de décès neurologique. Nous avons extrait les informations pertinentes à nos questions et les références qui les soutiennent. 2) Nous avons réalisé une étude transversale (étude Pinkie) nichée dans la cohorte INDex « CT-Perfusion for neurological diagnostic evaluation: a prospective Canadian multicenter diagnostic test study ». Nous avons comparé les concentrations plasmatiques de MSA lors du DDN et investigué l’association statistique entre MSA et DDN. Résultats : 1) La revue de la portée a démontré la variabilité des recommandations dans le domaine et le peu de preuves qui les soutiennent. Néanmoins, 67% des recommandations suggéraient de doser les MSA lors du DDN. 2) Nous avons observé dans l’étude Pinkie que les concentrations plasmatiques de MSA sont faibles et qu’elles ne sont pas associées à la probabilité de déclarer un patient en décès neurologique. Conclusion : Dans le cadre d’un diagnostic de décès neurologique effectué selon les guides de pratique au Canada, les médicaments sédatifs et analgésiques ne sont pas associés à la probabilité de diagnostic de décès neurologique. / Background: Neurological death is defined by the permanent loss of cerebral function. The validity of its diagnosis relies on prerequisite conditions and the exclusion of confounding factors. The goal of this work is to investigate the potential confounding role of sedatives and analgesics drugs (SAD) on the neurological death determination (NDD). Methods: 1) We conducted a scoping review of global guidelines addressing the role of sedative and analgesic drugs in the neurological death determination. We have extracted the information relevant to our questions and the references that support them. 2) We conducted a cross-sectional study (Pinkie study) nested in the INDex cohort “CT-Perfusion for neurological diagnostic evaluation: a prospective Canadian multicenter diagnostic test study”. We compared plasma concentrations of SAD during NDD and investigated the statistical association between SAD and NDD. Results: 1) The scoping review demonstrated the variability of recommendations in the field and the limited evidence supporting them. Nevertheless, 67% of recommendations suggested dosing SAD during NDD. 2) We observed in the Pinkie study that the plasma concentrations of SAD are low and that they are not associated with the probability of declaring a patient neurologically dead. Conclusion: In the context of a neurological death determination made according to the Canadian practice guidelines, sedative and analgesic drugs are not associated with the probability of neurological death diagnosis.

Le développement de nouveaux médicaments ou traitements contre le cancer requiert des années de travail préclinique avant de se rendre aux essais cliniques et ultimement le marché. Malheureusement, la grande majorité des composés ne réussiront pas cette transition et ne démontreront pas de bénéfices en essais cliniques. Dans le but de réduire l’attrition au long du processus de développement des médicaments, la modélisation mathématique est de plus en plus utilisée dans la recherche préclinique pour investiguer et optimiser les traitements pour améliorer les probabilités de succès de thérapies potentielles. Les modèles mécanistiques visent à incorporer les mécanismes d’action d’un médicament ainsi que les interactions physiologiques et cellulaires pour approfondir notre compréhension des systèmes et des effets thérapeutiques. La dissertation suivante traite de l’implémentation the modèles mécanistique hétérogènes dans des contextes précliniques pour la recherche contre le cancer. Le second chapitre discute du cancer des cellules granulosa ovarienne and du développement d’un modèle mathématique pour investiguer le potentiel d’une thérapie combinatoire qui inclut une chimiothérapie et une immunothérapie produisant une protéine en lien avec le facteur de nécrose tumorale (TRAIL) à l’aide d’un virus oncolytique (VO). Le modèle considère les cellules tumorales à travers les stades de la mitose, l’infection de ces cellules par le VO et la pression du système immunitaire inné sur la population de cellules tumorales. Le modèle incorpore aussi des modèles pharmacocinétique/pharmacodynamie (PK/PD) pour TRAIL et le médicament chimio thérapeutique, composé activateur de procaspase-1 (PAC-1). Cela inclue un modèle PK mécanistique décrivant la liaison de TRAIL à son récepteur ainsi qu’un modèle pharmacocinétique à deux compartiments pour PAC-1 dans le but d’intégrer les deux concentrations dans une fonction d’effets combinés affectant la population de cellules cancéreuses. À travers les simulations, nous avons déterminé les doses minimales requises et le schéma posologique optimal pour PAC-1 pour minimiser la croissance tumorale. Nous avons aussi établi un scénario permettant d’éradiquer la tumeur à l’aide d’un VO possédant un taux d’infection plus grand qu’initialement testé. 4 Dans le chapitre 3, nous présentons différentes approches pour inclure la variabilité inter-individuelle dans des modèles mécanistiques et discutons de leur bénéfices et désavantages. Nous décrivons comment les modèles PK de population (PopPK) informent sur la moyenne des paramètres d’une cohorte, la variation provenant des covariables et comment cette variabilité dans les paramètres permet d’étudier différentes dynamiques à travers une population. Dans une cohorte, la variabilité peut être généré par des algorithmes en assurant que les patients virtuels générés possèdent des paramètres et des résultats réalistiques. Nous discutons aussi des cohortes in silico pouvant prédire un intervalle de résultats and de scénarios potentiels d’un traitement. Ces essais cliniques virtuels sont très utiles en pharmacologie quantitative de systèmes (QSP). Enfin, nous présentons une application d’un modèle PopPK utilisant 300 patients virtuels dans un modèle QSP pour la différentiation des cellules souches mammaires affectées par des doses d’estrogène. Nous investiguons l’effet de cette thérapie hormonale sur la différentiation cellulaire pour son application potentiel pour traiter le cancer du sein triple négatif (TNBC) puisque la prolactine a été proposée dans des modèles expérimentaux pour forcer la différentiation cellulaire des cellules cancéreuses. Notre modèle et les résultats obtenus servent de preuve de concept pour continuer la recherche des méthode pharmacologiques pour induire la différentiation des cellules souches permettant de réduire la sévérité et la plasticité des cellules cancéreuses. / Developing novel cancer drugs or therapies requires years of preclinical work before translation to clinical trials and ultimately the market. Unfortunately, an overwhelming majority of compounds will fail to make this transition and will show no benefit in trials. To reduce attrition along the drug development pipeline, mathematical modelling is increasingly used in preclinical work to investigate and optimize treatment scenarios, in the hope of improving the success rate of potential therapies. Mechanistic models aim to incorporate the mechanisms of actions of drugs and physiological/cellular interactions to provide a deeper understanding of the system and rationally investigate therapeutic effectiveness. This thesis focuses on the implementation of heterogeneous, mechanistic mathematical models in preclinical contexts in cancer drug development. The first chapter of this thesis provides an overview of mathematical oncology and the drug discovery pipeline by presenting different tumour growth models and the integration of therapeutic effect through pharmacokinetic/pharmacodynamic (PK/PD) models. The second chapter of this thesis discusses granulosa cell tumour (GCT) of the ovary and the development of a mathematical model to investigate the potential of a combination therapy using a chemotherapy and an immunotherapy that produces tumour necrosis factor-related apoptosis-inducing ligand (TRAIL) through an oncolytic virus (OV). The model considers tumour cells throughout the phases of the cell cycle, the infection of these cancer cells by the OV, and the innate-immune pressure from the body. It also incorporates detailed PK/PD models for TRAIL and the chemotherapeutic drug, procaspase activating compound-1 (PAC-1). This includes a mechanistic receptor binding PK model for TRAIL as well as a two-compartment PK model for PAC-1 to properly integrate the concentrations of both compounds in the combination effect function applied to the cancer cell populations. Through simulations and hypothesis testing, we determined the minimal doses and ideal dosing regimens for PAC-1 that best controlled tumour growth. We also established how to successfully eradicate the tumour under the assumption of a much higher infection rate of the OV. 6 In the third chapter, we present different approaches to include inter-individual variability into mechanistic mathematical models, each with their own benefits and challenges. We describe how population PKs (PopPK) inform on cohort averages and variability due to covariates, and how to use this heterogeneity to recover the dynamics of drug treatment in patient populations. Variability in cohorts can also be generated through algorithms ensuring that virtual patients have realistic parameters and outcomes. We also touch upon in silico trials that help to predict a range of outcomes and treatment scenarios. These in silico clinical trials are highly valuable in quantitative system pharmacology (QSP) due to their predictive nature. Lastly, we present an application of PopPK using 300 generated patients in a QSP model for mammary stem cell differentiation under treatment with estrogen (estradiol). We investigate the effect of hormone therapy on mammary cell differentiation due to its potential application in triple negative breast cancer (TNBC), as prolactin has been proposed in experimental models to induce differentiation in TNBC stem cells. Our model and results serve as proof of concept for the continued investigation into pharmacological means of inducing stem cell differentiation to reduce cancer plasticity and severity.

No description available.

Indiana University-Purdue University Indianapolis (IUPUI) / Model-based drug development has become an essential element to potentially make drug development more productive by assessing the data using mathematical and statistical approaches to construct and utilize models to increase the understanding of the drug and disease. The modeling and simulation approach not only quantifies the exposure-response relationship, and the level of variability, but also identifies the potential contributors to the variability. I hypothesized that the modeling and simulation approach can: 1) leverage our understanding of pharmacokinetic-pharmacodynamic (PK-PD) relationship from pre-clinical system to human; 2) quantitatively capture the drug impact on patients; 3) evaluate clinical trial designs; and 4) identify potential contributors to drug toxicity and efficacy. The major findings for these studies included: 1) a translational PK modeling approach that predicted clozapine and norclozapine central nervous system exposures in humans relating these exposures to receptor binding kinetics at multiple receptors; 2) a population pharmacokinetic analysis of a study of sertraline in depressed elderly patients with Alzheimer’s disease that identified site specific differences in drug exposure contributing to the overall variability in sertraline exposure; 3) the utility of a longitudinal tumor dynamic model developed by the Food and Drug Administration for predicting survival in non-small cell lung cancer patients, including an exploration of the limitations of this approach; 4) a Monte Carlo clinical trial simulation approach that was used to evaluate a pre-defined oncology trial with a sparse drug concentration sampling schedule with the aim to quantify how well individual drug exposures, random variability, and the food effects of abiraterone and nilotinib were determined under these conditions; 5) a time to event analysis that facilitated the identification of candidate genes including polymorphisms associated with vincristine-induced neuropathy from several association analyses in childhood acute lymphoblastic leukemia (ALL) patients; and 6) a LASSO penalized regression model that predicted vincristine-induced neuropathy and relapse in ALL patients and provided the basis for a risk assessment of the population. Overall, results from this dissertation provide an improved understanding of treatment effect in patients with an assessment of PK/PD combined and with a risk evaluation of drug toxicity and efficacy.