Graduation date: 1999

The “magic bullet” theory stated by Paul Ehrlich more than one century ago foresees that drugs need to reach their target to exert their pharmacological activity. When dealing with antibiotics, this implies not only a capacity to interact with bacteria, but also to rejoin them in the infected compartment. The latter property is directly linked to pharmacokinetics. In this Thesis, we have focused our interest on cellular pharmacokinetics of antibiotics in macrophages, reasoning that (i) this model would be relevant for predicting the access of antibiotics to intracellular bacteria, and (ii) deciphering mechanisms of drug entry, distribution, and efflux in single cells would provide pieces of information applicable to the understanding of general pharmaco-kinetics. Accumulation and distribution were studied for oritavancin, a semi-synthetic glycopeptide currently in phase III of clinical development. This molecule is characterized by a highly bactericidal activity against staphylococci, including multiresistant strains, which would be of interest for the eradication of the intracellular forms of these bacteria. We show that oritavancin enters macrophages by adsorptive endocytosis and accumulates to very large amounts in lysosomes, with cellular concentrations as high as 300-fold the extracellular ones. Accordingly, it exerts a concentration-dependent bactericidal activity against Staphylococcus aureus multiplying in phagolysosomes. In parallel, however, it also causes morphological and biochemical alterations, characterized by the deposition of material of heterogeneous aspect and the accumulation of phospholipids and cholesterol. These data point to the difficulty of dissociating high cellular accumulation and cellular toxicity and plead for the interest of in vitro models in the evaluation of the intracellular activity of antibiotics, the early assessment of drug safety profile and the orientation of further in vivo studies. Active efflux was evaluated for macrolides and quinolones, two antibiotic classes accumulating in macrophages and therefore usually considered as useful in the treatment intracellular infections. Macrophages indeed express multidrug transporters in their pericellular membrane, which can extrude a large variety of drugs presenting as common feature an amphiphatic character. We show that macrolides (and azithromycin in particular) are substrates for P-glycoprotein, while quinolones (and ciprofloxacin in particular) are substrates for an MRP-like transporter. Inhibiting these pumps increases the intracellular activity of these antibiotics by enhancing their cellular accumulation. We also describe different mode of transport for these antibiotics, which essentially reflect their variable fluxes through the membrane. These data underline the major role constitutive efflux plays in the modulation of the pharmacokinetics of drugs, and, hence, of their pharmacodynamics. They encourage the setting-up of large-scale screenings aimed at evaluating drug-transporter interactions in lead optimization processes. We may therefore conclude to the usefulness of cellular models in the study of the pharmacokinetics of drugs and of its consequences for pharmacodynamics and toxicity, and suggest the implementation of such models in the early development of new drugs.

New approaches quantifying the effect of treatment are needed in oncology to improve the drug development process and to enable treatment optimization for existing therapies. This thesis focuses on the development of pharmacometric models for biomarkers, side effects and efficacy in order to identify predictors of clinical response in anti-cancer drug therapy. The variability in myelosuppression was characterized in six different cytotoxic anticancer treatments to evaluate a model-based dose individualization approach utilizing neutrophil counts as a biomarker. The estimated impact of inter-occasion variability was relatively low in relation to the inter-individual variability, indicating that myelosuppression is predictable from one treatment course to another. The approach may thereby be useful for dose optimization within an individual. To further study and to identify predictors for the severe side effect febrile neutropenia (FN), the relationship between the shape of the myelosuppression time-course and the probability of FN was characterized. Patients with a rapid decline in neutrophil counts and high drug sensitivity were identified to have a higher probability of developing FN compared with other patients who experience grade 4 neutropenia. Predictors of clinical response in patients receiving sunitinib for the treatment of gastro-intestinal stromal tumor (GIST) were identified by the development of an integrated modeling framework. Drug exposure, biomarkers, tumor dynamics, side effects and overall survival (OS) were linked in a unified structure, and univariate and multivariate exposure variables were tested for their predictive capacities. The soluble biomarker, sVEGFR-3 and tumor size at start of treatment were found to be promising predictors of overall survival, with decreased sVEGFR-3 levels and smaller baseline tumor size being predictive of longer OS. Also hypertension and neutropenia was identified as predictors of OS. The developed modeling framework may be useful to monitor clinical response, optimize dosing in sunitinib and to facilitate dose individualization.

Physiologically based pharmacokinetic (PBPK) modeling is a tool used in drug discovery and human health risk assessment. PBPK models are mathematical representations of the anatomy, physiology and biochemistry of an organism. PBPK models, using both compound and physiologic inputs, are used to predict a drug’s pharmacokinetics in various situations. Tissue to plasma partition coefficients (Kp), a key PBPK model input, define the steady state concentration differential between the tissue and plasma and are used to predict the volume of distribution. Experimental determination of these parameters once limited the development of PBPK models however in silico prediction methods were introduced to overcome this issue. The developed algorithms vary in input parameters and prediction accuracy and none are considered standard, warranting further research. Chapter 2 presents a newly developed Kp prediction algorithm that requires only readily available input parameters. Using a test dataset, this Kp prediction algorithm demonstrated good prediction accuracy and greater prediction accuracy than preexisting algorithms. Chapter 3 introduced a decision tree based Kp prediction method. In this novel approach, six previously published algorithms, including the one developed in Chapter 2, were utilized. The aim of the developed classifier was to identify the most accurate tissue-specific Kp prediction algorithm for a new drug. A dataset consisting of 122 drugs was used to train the classifier and identify the most accurate Kp prediction algorithm for a certain physico-chemical space. Three versions of tissue specific classifiers were developed and were dependent on the necessary inputs. The use of the classifier resulted in a better prediction accuracy as compared to the use of any single Kp prediction algorithm for all tissues; the current mode of use in PBPK model building. With built-in estimation equations for those input parameters not necessarily available, this Kp prediction tool will provide Kp prediction when only limited input parameters are available. The two presented innovative methods will improve tissue distribution prediction accuracy thus enhancing the confidence in PBPK modeling outputs.

Whole-body PBPK models were developed based on both the intestinal traditional model (TM) and segregated-flow model (SFM) to describe codeine sequential metabolism in man/rat. Model parameters were optimized with Scientist® and Simcyp® simulator to predict literature data after oral (p.o.) and intravenous (i.v.) codeine administration in man/rat. In vivo codeine PK studies on rats were performed to provide more data for simulation. The role of fm’ (fractional formation clearance of morphine from codeine) in model discrimination between the TM and SFM was investigated. A greater difference between the [AUC_M3G/AUC_Morphine]p.o. and [AUC_M3G/AUC_Morphine]i.v. ratio existed for the SFM, especially when the fm’ was low. It was found that our tailor-made PBPK models using Scientist® were superior to those from Simcyp® in describing codeine sequential metabolism. Residual sum of squares and AUC’s were calculated for each model, which demonstrated superiority of the SFM over TM in predicting codeine sequential metabolism in man/rat.

Obesity in pregnancy has been associated with an elevated risk for neural tube defects, though it is unknown if this is linked to a lower folate status in obese women. Studies have identified a reduced folate status among obese women even after controlling for folate intake. Thus, it is possible that folic acid pharmacokinetics are altered in the obese body. In this study, we compared the pharmacokinetics of folic acid in obese and non-obese women of childbearing age, following administration of a weight-adjusted dose. Area under the concentration-time curve was found to be significantly higher in the obese group, with the dose per kilogram lean body weight most strongly predicting systemic exposure. Estimation of the daily dose required to achieve protective blood concentrations did not identify a need to change supplementation recommendations for obese women. Accordingly, current guidelines appear to suggest adequate doses for obese women of childbearing age.

Whole-body PBPK models were developed based on both the intestinal traditional model (TM) and segregated-flow model (SFM) to describe codeine sequential metabolism in man/rat. Model parameters were optimized with Scientist® and Simcyp® simulator to predict literature data after oral (p.o.) and intravenous (i.v.) codeine administration in man/rat. In vivo codeine PK studies on rats were performed to provide more data for simulation. The role of fm’ (fractional formation clearance of morphine from codeine) in model discrimination between the TM and SFM was investigated. A greater difference between the [AUC_M3G/AUC_Morphine]p.o. and [AUC_M3G/AUC_Morphine]i.v. ratio existed for the SFM, especially when the fm’ was low. It was found that our tailor-made PBPK models using Scientist® were superior to those from Simcyp® in describing codeine sequential metabolism. Residual sum of squares and AUC’s were calculated for each model, which demonstrated superiority of the SFM over TM in predicting codeine sequential metabolism in man/rat.

Obesity in pregnancy has been associated with an elevated risk for neural tube defects, though it is unknown if this is linked to a lower folate status in obese women. Studies have identified a reduced folate status among obese women even after controlling for folate intake. Thus, it is possible that folic acid pharmacokinetics are altered in the obese body. In this study, we compared the pharmacokinetics of folic acid in obese and non-obese women of childbearing age, following administration of a weight-adjusted dose. Area under the concentration-time curve was found to be significantly higher in the obese group, with the dose per kilogram lean body weight most strongly predicting systemic exposure. Estimation of the daily dose required to achieve protective blood concentrations did not identify a need to change supplementation recommendations for obese women. Accordingly, current guidelines appear to suggest adequate doses for obese women of childbearing age.

El capítulo de "Resultados" ha sido retirado siguiendo indicaciones de la autora, al existir participación de empresas, convenio de confidencialidad o la posibilidad de generar patentes. / El principio activo sometido a estudio es irosustat, un novedoso inhibidor irreversible de la sulfatasa esteroidal (STS), que actúa bloqueando la síntesis de estrógenos que están implicados en el desarrollo y progresión del cáncer de mama hormonodependiente. El objetivo principal de esta investigación es evaluar las características farmacocinéticas preclínicas de irosustat. Para ello, se investigan en tres especies animales, rata, perro y mono, los procesos de absorción, distribución y eliminación que intervienen en el paso del compuesto por el organismo. El primer aspecto característico del producto, y relevante desde un punto de vista farmacocinético, es su unión reversible a la anhidrasa carbónica II (CA II) del interior de los eritrocitos. La farmacocinética de irosustat depende de la magnitud de dicha unión: el aumento de la unión a CA II prolonga su semivida en sangre, mientras que un desplazamiento de dicha unión provoca una aceleración en la eliminación. Este fenómeno ha supuesto una dificultad para el estudio farmacocinético de irosustat, especialmente en cuanto a la manipulación y conservación de las muestras biológicas. Por un lado, se han determinado simultáneamente las concentraciones de irosustat en plasma y sangre, mediante una metodología consistente en cromatografía líquida de alta eficacia acoplada a espectrometría de masas, con un límite de cuantificación de 0.025 ng/ml en ambas matrices biológicas. Por otro lado, se ha establecido el protocolo a seguir para manipular correctamente las muestras de plasma y sangre. La absorción de irosustat es relativamente rápida, puesto que los valores de Cmax en plasma se alcanzan generalmente alrededor de 1 hora post administración en rata, perro y mono. En plasma de rata y perro, la biodisponibilidad absoluta de irosustat en 1% metilcelulosa es incompleta y decrece al incrementar las dosis. Los valores hallados en rata oscilan entre 27.0% y 11.9% en el intervalo de dosis de 3 a 300 mg/kg, y en perro entre 84.1% y 44.5% a las dosis de 0.25 a 25 mg/kg. Cuando la biodisponibilidad se calcula a partir de las concentraciones de irosustat en sangre, los resultados hallados son muy superiores debido al efecto protector que ejercen los eritrocitos frente a la degradación y/o el metabolismo. La ausencia de linealidad cinética a las dosis superiores se ha atribuido a una falta de absorción de irosustat, debido a su limitada solubilidad. En referencia a la distribución, irosustat presenta una alta fijación a proteínas plasmáticas (>96%). Al administrar 14C irosustat a la rata, se observa que la radioactividad se distribuye ampliamente en los tejidos y no existe un fenómeno de fijación asociada a la melanina. Utilizando las herramientas de la farmacocinética poblacional, se ha construido un modelo no lineal de efectos mixtos en rata y perro, que se caracteriza por incorporar un proceso de distribución no lineal que considera la entrada inmediata de irosustat al interior de los eritrocitos para explicar el comportamiento farmacocinético de irosustat en plasma y sangre. El aclaramiento plasmático de irosustat es de 0.129 l/h en la rata y de 31.9 l/h en el perro, con variabilidades interindividuales de 3.8% y 33%, respectivamente. El volumen de distribución poblacional es de 0.11 l en la rata y de 4.5 l en el perro, no pudiéndose incluir variabilidad interindividual para este parámetro. Irosustat se elimina extensivamente metabolizado por el hígado y su principal vía de excreción es mayoritariamente la vía renal en las tres especies ensayadas. / The compound under study is irosustat, a novel irreversible inhibitor of steroid sulfatase (STS), which works by blocking the synthesis of estrogen that are involved in the development and progression of hormone dependent breast cancer. The main objective is to evaluate the preclinical pharmacokinetic properties of irosustat. Thus, the processes of absorption, distribution and elimination are investigated in rat, dog and monkey. The main characteristic of irosustat, that affects its pharmacokinetic behavior, is its reversible binding to carbonic anhydrase II (CA II) within erythrocytes. In order to gain insight into how binding to CA II within erythrocytes influences irosustat pharmacokinetics, we comparatively assess its concentration in plasma and blood following different dose levels. Irosustat absorption is relatively rapid, since the plasma Cmax values are usually reached about 1 hour after administration in rat, dog and monkey. In rat and dog plasma, the absolute bioavailability of irosustat in 1% methylcellulose is incomplete and decreases with increasing dose levels. The values found in the rat range between 27.0% and 11.9% at doses of 3 to 300 mg/kg, and in the dog range between 84.1% and 44.5% at doses of 0.25 to 25 mg/kg. When the bioavailability is calculated from irosustat concentrations in blood, the results found are much higher due to the protective effect of erythrocytes against degradation and/or metabolism. About distribution, irosustat exhibits high plasma protein binding (> 96%). When 14C irosustat is administered to rats, radioactivity is distributed widely in tissues. Using the tools of population pharmacokinetics, it has been built a model in rat and dog, characterized by incorporating a nonlinear distribution process that considers irosustat immediate entry into erythrocytes to explain the pharmacokinetic behavior of the compound in plasma and blood. The plasma clearance of irosustat is 0.129 L/h in rats and 31.9 L/h in dogs, with interindividual variability of 3.8% and 33%, respectively. The population volume of distribution is 0.11 L in the rat and 4.5 L in the dog, without being possible to include interindividual variability for this parameter. Irosustat is eliminated extensively metabolized by the liver and its major route of excretion is via the kidney in the three animal species tested.

Warfarin is one of the most widely used anticoagulants. Therapy is complicated by warfarin’s narrow therapeutic range and pronounced variability in individual dose requirements. Although warfarin therapy is uncommon in children, it is crucial for children with certain congenital or acquired heart diseases. Treatment in children is especially difficult due to the lack of i) a decision support tool for efficient and consistent dose adjustments, and ii) a flexible warfarin formulation for accurate and reproducible dosing. The overall aim of this thesis was to develop a PKPD-based pharmacometric model for warfarin that describes the dose-response relationship over time, and to identify important predictors that influence individual dose requirements both in adults and children. Special emphasis was placed on investigating the contribution of genetic factors to the observed variability. A clinically useful pharmacometric model for warfarin has been developed using NONMEM. The model has been successfully reformulated into a KPD-model that describes the relationship between warfarin dose and INR response, and that is applicable to both adults and children. From a clinical perspective, this is a very important change since it allows the use of information on dose and INR that is available routinely. The model incorporates both patient and clinical characteristics, such as age, weight, CYP2C9 and VKORC1 genotype, and baseline and target INR, for the prediction of an individualised starting dose. It also enables the use of information from previous doses and INR observations to further individualise the dose a posteriori using a Bayesian forecasting method. The NONMEM model has been transferred to a user-friendly, platform independent tool to aid use in clinical practice. The tool can be used for a priori and a posteriori individualisation of warfarin therapy in both adults and children. The tool should ensure consistent dose adjustment practices, and provide more efficient individualisation of warfarin dosing in all patients, irrespective of age, body weight, CYP2C9 or VKORC1 genotype, baseline or target INR. The expected outcome is improved warfarin therapy compared with empirical dosing, with patients achieving a therapeutic and stable INR faster and avoiding high INRs that increase the risk of bleeding.