Search

Background: Available in-vitro and animal studies indicate that inflammation impacts cytochromes P450 (CYP) activityviamultiple and complex transcriptional and post-transcriptional mechanisms, depending on the specific CYP isoforms and the nature of inflammation mediators. It is essential to review the current published data on the impact of inflammation on CYP activities in adults to support drug individualization based on comorbidities and diseases in clinical practice. Methods: This systematic review was conducted in PubMed through 7th January 2021 looking for articles that investigated the consequences of inflammation on CYP activities in adults. Information on the source of inflammation, victim drugs (and CYPs involved), effect of disease-drug interaction, number of subjects, and study design were extracted. Results: The search strategy identified 218 studies and case reports that met our inclusion criteria. These articles were divided into fourteen different sources of inflammation (such as infection, autoimmune diseases, cancer, therapies with immunomodulator…). The impact of inflammation on CYP activities appeared to be isoform-specific and dependent on the nature and severity of the underlying disease causing the inflammation. Some of these drug-disease interactions had a significant influence on drug pharmacokinetic parameters and on clinical management. For example, clozapine levels doubled with signs of toxicity during infections and the concentration ratio between clopidogrel's active metabolite and clopidogrel is 48-fold lower in critically ill patients. Infection and CYP3A were the most cited perpetrator of inflammation and the most studied CYP, respectively. Moreover, some data suggest that resolution of inflammation results in a return to baseline CYP activities. Conclusion: Convincing evidence shows that inflammation is a major factor to be taken into account in drug development and in clinical practice to avoid any efficacy or safety issues because inflammation modulates CYP activities and thus drug pharmacokinetics. The impact is different depending on the CYP isoform and the inflammatory disease considered. Moreover, resolution of inflammation appears to result in a normalization of CYP activity. However, some results are still equivocal and further investigations are thus needed.

Acute nociceptive pain management in children is a major public health concern. Effective and safe pain treatment is essential, but safety data cannot be simply extrapolated from adults to children due to pharmacokinetic and pharmacodynamic specificities. In addition, the frequent absence of child-specific data, the difficulty to assess drug tolerability, and the infants' inability to communicate properly and voluntarily report adverse drug reactions make children more vulnerable to safety issues. Awareness of the possible toxicity of analgesics is important but should not lead to suboptimal dosing and underuse of analgesia. A better assessment and individualization of treatment should allow effective prescribing of analgesics in more secure conditions. This article aims to review the safety of acetaminophen, nonsteroidal anti-inflammatory drugs, and opioids in children and the precautions that should be taken.

Antidepressants, mainly tricyclic and non-selective reuptake inhibitors of serotonin antidepressants, are part of the treatment of chronic pain. The management is complicated by a large interindividual variability of efficacy and tolerance. Important part of this variability is associated with nucleotide polymorphisms of genes encoding enzymes involved in the pharmacokinetics and pharmacodynamics of these molecules. Identification of these genetic variants could to predict clinical consequences and allowed individualized adjustments in medication or dosage. This article presents the current knowledge on the influence of genetics on the efficacy and adverse effects of antidepressants used in chronic pain treatment.

In Switzerland, medical cannabinoids can be prescribed under compassionate use after special authorization in justified indications such as refractory pain. Evidence of efficacy in pain is limited and the clinical benefit seems to be modest. Their drug-drug interactions (DDI) profile is poorly documented. Cytochromes P450 (CYP) 2C9 and 3A4 are involved in the metabolism of tetrahydrocannabinol and cannabidiol, which implies possible DDI with CYP450 inhibitor and inducer, such as anticonvulsivants and HIV protease inhibitors, which may be prescribed in patients with neuropathic pain.

Adverse drug events (ADE) are a major public health issue, with drug-drug interactions (DDI) being one of well-recognized causes of ADE that could be preventable by the use of DDI screening software. We compared the ability of four programs to detect clinically important DDI. We tested 62 drug pairs with and 12 drug pairs without clinically important DDI. Lexi-Interact and Epocrates were the most sensitive (95%) compared to the Compendium and Theriaque (80 and 73%, respectively). The Compendium and Theriaque also showed the lowest negative predictive value. All programs showed high specificity and positive predictive value. The qualitative assessment showed the best performances for Compendium and Lexi-Interact. The last one seems to be the best screening program, but the Compendium is in French and is freely available.

BACKGROUND AND PURPOSE: There is high interindividual variability in the activity of drug-metabolizing enzymes catalysing the oxidation of oxycodone [cytochrome P450 (CYP) 2D6 and 3A], due to genetic polymorphisms and/or drug-drug interactions. The effects of CYP2D6 and/or CYP3A activity modulation on the pharmacokinetics of oxycodone remains poorly explored. EXPERIMENTAL APPROACH: A randomized crossover double-blind placebo-controlled study was performed with 10 healthy volunteers genotyped for CYP2D6 [six extensive (EM), two deficient (PM/IM) and two ultrarapid metabolizers (UM)]. The volunteers randomly received on five different occasions: oxycodone 0.2 mg x kg(-1) and placebo; oxycodone and quinidine (CYP2D6 inhibitor); oxycodone and ketoconazole (CYP3A inhibitor); oxycodone and quinidine+ketoconazole; placebo. Blood samples for plasma concentrations of oxycodone and metabolites (oxymorphone, noroxycodone and noroxymorphone) were collected for 24 h after dosing. Phenotyping for CYP2D6 (with dextromethorphan) and CYP3A (with midazolam) were assessed at each session. KEY RESULTS: CYP2D6 activity was correlated with oxymorphone and noroxymorphone AUCs and C(max) (-0.71 < Spearman correlation coefficient rhos < -0.92). Oxymorphone C(max) was 62% and 75% lower in PM than EM and UM. Noroxymorphone C(max) reduction was even more pronounced (90%). In UM, oxymorphone and noroxymorphone concentrations increased whereas noroxycodone exposure was halved. Blocking CYP2D6 (with quinidine) reduced oxymorphone and noroxymorphone C(max) by 40% and 80%, and increased noroxycodone AUC(infinity) by 70%. Blocking CYP3A4 (with ketoconazole) tripled oxymorphone AUC(infinity) and reduced noroxycodone and noroxymorphone AUCs by 80%. Shunting to CYP2D6 pathway was observed after CYP3A4 inhibition. CONCLUSIONS AND IMPLICATIONS: Drug-drug interactions via CYP2D6 and CYP3A affected oxycodone pharmacokinetics and its magnitude depended on CYP2D6 genotype.

A bioanalytical method was developed for the simultaneous determination of paracetamol and ketorolac enantiomers in human plasma using two-dimensional liquid chromatography-mass spectrometry. Separation was first achieved in a reversed-phase C18 column by using a gradient solvent system consisting of 0.1% aqueous formic acid and acetonitrile (ACN). The effluent between 8.9 and 9.9 min, corresponding to phenacetin and racemic ketorolac peaks, was transferred to a polysaccharide-based chiral column (ChiralPak AD-RH) by using a six-port switching valve. Ketorolac enantiomers were subsequently separated on the chiral column using an isocratic mobile phase composed of ACN/0.1% formic acid 50:50 (v/v). The total run-time was less than 18 min. This innovative strategy prolongs the lifetime of chiral columns by avoiding damages due to the sample matrix. The detection was carried out with an ion trap mass spectrometer equipped with an electrospray ionisation source. The tested ranges were 0.05-20 microg/ml for paracetamol and 0.005-2 microg/ml for each ketorolac enantiomer. This method was fully validated and showed good performances in terms of trueness (80-110%) and precision (6.7-13.2%). The mean extraction recoveries were 60%, 72% and 76% for paracetamol, R-ketorolac and S-ketorolac, respectively. Finally, this procedure was successfully applied to a pharmacokinetic study.

The use of analgesics is based on the empiric administration of a given drug with clinical monitoring for efficacy and toxicity. However, individual responses to drugs are influenced by a combination of pharmacokinetic and pharmacodynamic factors that can sometimes be regulated by genetic factors. Whereas polymorphic drug-metabolizing enzymes and drug transporters may affect the pharmacokinetics of drugs, polymorphic drug targets and disease-related pathways may influence the pharmacodynamic action of drugs. After a usual dose, variations in drug toxicity and inefficacy can be observed depending on the polymorphism, the analgesic considered and the presence or absence of active metabolites. For opioids, the most studied being morphine, mutations in the ABCB1 gene, coding for P-glycoprotein (P-gp), and in the micro-opioid receptor reduce morphine potency. Cytochrome P450 (CYP) 2D6 mutations influence the analgesic effect of codeine and tramadol, and polymorphism of CYP2C9 is potentially linked to an increase in nonsteroidal anti-inflammatory drug-induced adverse events. Furthermore, drug interactions can mimic genetic deficiency and contribute to the variability in response to analgesics. This review summarizes the available data on the pharmacokinetic and pharmacodynamic consequences of known polymorphisms of drug-metabolizing enzymes, drug transporters, drug targets and other nonopioid biological systems on central and peripheral analgesics.

Recent research has revealed the presence of opioid receptors in inflamed peripheral tissues. This gives rise to the possibility of treating the pain caused by wounds with the advantage of reducing the secondary effects related to the use of opioids and at the same time rendering their use more efficacious. The theory of a peripheral analgesic action of opioids has been tested in a variety of situations, in particular in brachial plexus blocks and intra-articular injections. In addition, the analgesic effect of opioids has been tested by cutaneous application for various conditions but on only a limited number of patients. On the basis of these case reports, we cannot conclude that the peripheral use of opioids is efficacious and the topical use of opioids cannot be recommended unless done within the framework of a controlled clinical trial.

Cancer pain can be effectively controlled in most patients by classical pharmacological treatment. We retrospectively studied the characteristics and factors associated with non responsive pain. Between 1989 and 1996, 1767 patients were referred to our pain center; 831 (47%) had cancer pain and from 787 evaluable cases 118 (15%) experienced non-controlled pain whereas good pain control was achieved within a few days in 669 (85%) patients. Gender, age, cancer type, metastasis, initial pain intensity, nociceptive or neuropathic components and administration of adjuvant therapies were similar in both groups. On the other hand, diffuse pain, abdominal pain, terminal care, near death and doses of strong opioids were significantly different. Factors associated with therapeutic failure were conflicts, life and complications and breakthrough pain. In the presence of refractory cancer pain the factors predictive of therapeutic failure should be identified in order to optimize individual pain treatment.

A retrospective study over the last 5 years on drug-induced liver injury was performed in the Internal Medicine Department of the University Hospital of Geneva. According to WHO and Council for International Organizations of Medical Sciences (CIOMS) criteria for drug-induced liver adverse events, 142 cases were identified, with 255 drugs fulfilling causality criteria of certain, probable or possible. 63 patients (44%) suffered severe consequences of liver injury, including 9 deaths. Drugs reported belonged to the therapeutic classes of antibiotics (43%), cardiovascular (20%), analgesics (10%), gastrointestinal (9%), psychotropic (7%), and others (11%). Despite bias inherent to the retrospective methodology, only 10% of the side effects were spontaneously reported to the pharmaco-vigilance board. This is in accordance with the known reporting rate. It can be concluded that spontaneous reports are to be regarded only as signals.

Factors relating to working status, i.e. whether the subject is still at work or on workmen's compensation, have been studied in chronic pain patients. Sociodemographic and pain variables were not correlated with absence from work, nor was diagnosis. However, beliefs and attitudes concerning pain differed between patients still working and those on workmen's compensation. In chronic pain patients medical attention should focus not only on description of symptoms but also on the "illness behaviour" dimension.

The purpose of this study was to evaluate the dose-response of paracetamol and to assess its plasma concentration-effect relationship. According to a randomized, double-blind, and placebo controlled design healthy volunteers (n = 11) received 0.5, 1 and 2 g paracetamol (as propacetamol) intravenously (15 min). The extent of analgesic effect was measured over 8 hours by objective pain threshold (R-III) monitoring after selective transcutaneous electrical stimulation of the ipsilateral sural nerve. Maximal R-III threshold increase is at 2 hours after 0.5 g and 2.5 hours after 1 g and 2 g. The delayed peak effects, when compared to peak plasma concentrations, draw counter clockwise hysteresis. Duration of paracetamol antinociceptive effect increased concomitantly to dosage increment. Intravenous paracetamol exerts a dose-dependent central analgesia.

This retrospective study considers the influence of chronological age on the perception and consequential effects of chronic non-cancer pain in 203 patients referred to an ambulatory pain clinic. One patient in four (50) was 65 years or older. Pain had existed for more than a year in 66%. It was referred to an "unbearable" by 54% of elderly adults vs 44% (NS) of the younger adult population, and as "intense" by 61% vs 44% (p < 0.05). A higher proportion of neurogenic pain (54% vs 26%, p < 0.05) could explain this significant difference. Cognitive factors could also have amplified the painful syndrome since 59% (vs 54%) of the elderly considered they had received insufficient information on the potential consequences of their painful condition. However, despite a higher occurrence of other severe illnesses in the elderly (52% vs 35%, p < 0.05), they appear to worry less about their health problems (58% vs 78%, p < 0.01). Finally, in spite of increased severity of pain in the elderly, depression scores and consequential effects are similar in the two populations. In conclusion, the intensity of chronic pain appears to be more severe in the elderly, perhaps due to its more frequent neuropathic origin. Nonetheless, its repercussions on daily life activity seem no worse than for younger adults, and could reflect better coping strategies towards chronic pain.

The aim of this retrospective study was to examine factors related to symptomatic pain treatment in 992 hospitalized patients referred to the pain clinic. These patients accounted for 1440 different pains (1.5 pains/patient), of nociceptive (59%), or neuropathic (24%) origin, a combination of both (16%) and of unidentified (1%) origin. Patients resistant to therapy did not differ from nonresistant patients (62%) in terms of age, sex, or number of pains. Although recommended pharmacological treatments were pursued, the neuropathic origin of pain was a factor in resistance to treatment. Among the neuropathic pains resistant to treatment (42%) a central nervous origin is more frequently found. Resistance to pain was more frequently encountered after strokes, primitive tumor of the central nervous system or traumatic plexopathy, whereas peripheral lesions of the nervous system were more easily controlled by symptomatic treatment. This study shows that central neuropathic pain is more frequently associated with resistance to recommended pharmacological symptomatic treatments. This observation underlines the limits of conventional medical therapy.

Les cytochromes P450 (CYP450) représentent les principales enzymes du métabolisme de phase I. Pour mesurer l'activité de ces CYP450, le phénotypage revêt une importance particulière en clinique car ces enzymes présentent une grande variabilité intra- et inter-individuelle pouvant influencer la réponse thérapeutique. Les principaux objectifs de cette thèse étaient d’identifier, développer et/ou valider des biomarqueurs urinaires ou plasmatiques pour évaluer de manière sûre et non invasive l'activité des CYP450 chez l’être humain. Plus précisément, certains biomarqueurs endogènes identifiés dans la littérature scientifique pour le CYP1A2 et le CYP3A4/5 ont fait l'objet d'études plus approfondies afin de pallier partiellement le manque de données cliniques nécessaires à leur validation. Par ailleurs, une analyse métabolomique non ciblée a été réalisée dans le but d’identifier de nouveaux marqueurs de l'activité du CYP2D6. Le présent travail représente une avancée dans la découverte et la mise en œuvre de biomarqueurs des CYP450 chez l'être humain.

Background: Most drugs are excreted in breastmilk, so breastfed children may be potentially exposed when their mothers take medications. Data on the safety of drug use by breastfeeding women is sparse and similarly the data on the occurrence of adverse drug reactions during breastfeeding is lacking. Objectives: Reviewing the current literature on reported cases and studies of the adverse drug reactions involving breastfed infants with medicated mothers. Finding mechanisms to explain the impact of mother’s drugs on their infants. Method: A literature search was conducted via PubMed, Embase and Google Scholar with the keywords “adverse drug reaction”, “breast feeding”, “breastfed neonates/newborns/infants”, from 23rd September 2020 to 3rd October of 2020. Case reports and studies were selected to be reviewed and analyzed. Results: 84% of cases of ADRs were encountered in breastfed infants younger than two months. They were serious in most cases and touched mainly neurological and gastro-intestinal systems. That was associated with the most involved drugs used by breastfeeding mothers including psychotropic drugs, analgesics, and antiepileptic drugs. Conclusion: Even if most of commonly used drugs are compatible with breastfeeding, they can pass in breastmilk and cause ADRs in breastfed infants. So, there is a risk for infants to develop adverse drug reactions, because of their vulnerability especially during the two first months of life. They are more prone to ADRs since they present pharmacokinetic and pharmacodynamic immaturity. Sometimes ADRs happen while the infant or the mother has a particularity such as a genetic polymorphism (i.e. rapid metabolizer), or hypersensitivity to the drug. Reported cases and studies are rarely and hardly found in the literature, there is a lack of information concerning this subject. It may be explained because pregnant and breastfeeding women are ethically excluded from clinical trials, and also for infants’ security considerations. Perhaps, there are not many cases of ADRs in breastfed infants in practice, and cases may be underreported like all pharmacovigilance. Nevertheless, in cases at risk, a close follow-up, a raise of parents’ awareness about warning symptoms, and a monitoring of drug concentration in breastmilk can be very useful to detect early any ADRs and to avoid tragic situations. Key words: Breastfeeding, lactation, breastfed infants, neonates, newborns, adverse drug reactions

The variability of a drug's response between patients is an expected feature of most drug treatments, it can lead to therapeutic failures or unwanted adverse drug reactions. This therapeutic phenomenon is increased by pharmacokinetic variability implying that, for the same dose, the effective drug concentration in body fluids is highly variable. In addition to pharmacokinetics, the interindividual pharmacodynamic variability impacts the response of receptors in equal concentration of drug. This work is focused on polymorphism in drug metabolizing enzymes, especially N-acetyltransferase 2 (NAT2).

Neuroinflammation is a complex mechanism linked to several brain diseases and many undesired side effects associated with opioid treatments, and results from the interactions of cells within the CNS such as astrocytes and brain endothelial cells. Albeit astrocytes are the most abundant cells of the CNS there is currently limited information about their role in neuroinflammatory processes and their involvement in the onset of morphine-induced side effects. In the present thesis we combined in vitro models and mass spectrometry-based proteomics to investigate and characterise the inflammatory phenotype of human astrocytes, brain endothelial cells and their released extracellular vesicles at protein level. Moreover, we extensively studied the effects, at translational and post-translational level, of morphine and its two main metabolites M3G and M6G in human astrocytes, and reported new information which is important to deepen our knowledge about the participation of these cells in the development of opioid-induced side effects.

La variabilité pharmacocinétique est une source majeure de variabilité de réponse aux médicaments, via notamment le métabolisme des médicaments par les cytochromes P450 (CYP). Parmi ceux-ci, le CYP2D6 représente un intérêt clinique considérable, en raison du grand nombre de médicaments qu'il métabolise et d'un polymorphisme important, principalement expliqué par des variations génétiques (facteur intrinsèque) ainsi que par des interactions médicamenteuses (facteur extrinsèque). Ce travail de thèse consiste en l'étude de l'impact du polymorphisme génétique sur les interactions médicamenteuses impliquant le CYP2D6. La thèse est divisée en trois volets majeurs: étude clinique mettant en évidence un effet du polymorphisme génétique sur l'amplitude des interactions médicamenteuses, étude in vitro investiguant les mécanismes des interactions gène-médicament-médicament, étude in silico de prédiction des interactions gène-médicament-médicament. Le présent travail souligne la nécessité de prise en compte des différents facteurs intrinsèques et extrinsèques de variabilité pharmacocinétique pour l'optimisation thérapeutique dans le contexte de la médecine de précision.

Les effets indésirables médicamenteux liés aux substances psychotropes sont des évènements fréquents, qui peuvent engendrer un transfert à l'hôpital somatique. Relever la nature de ces effets indésirables médicamenteux et identifier les substances et les facteurs de risque s'avère important pour sécuriser le processus de prescription des psychotropes. Les études sur les effets indésirables médicamenteux sont nombreuses au sein de la littérature mais peu ont été publiées spécifiquement sur les patients hospitalisés dans une unité psychiatrique. Le but de ce travail consiste donc à recenser le nombre de transferts d'un hôpital psychiatrique vers un service des urgences en lien avec un effet indésirable médicamenteux, et à étudier l'imputabilité et l'évitabilité des effets indésirables liés à des substances psychotropes, au sein de ces transferts. Les résultats intermédiaires de cette analyse sont présentés ici.

Une des principales sources de la variabilité de réponse aux médicaments est la variabilité pharmacocinétique pour laquelle les interactions médicamenteuses constituent une des explications les plus fréquentes. Ce travail de thèse consiste en grand partie en la conception et la réalisation de deux études cliniques dans lesquelles les interactions entre les médicaments antiplaquettaires clopidogrel, prasugrel, ticagrelor et le ritonavir, un médicament couramment prescrit chez des patients infectés par le VIH, ont été évaluées. Pour la première fois, le technique de modélisation pharmacocinétique basée sur la physiologie, appelée PBPK, a été utilisé afin de prédire l'ampleur de ces interactions. Il s'agit d'une méthode innovante et prometteuse de prédiction du devenir du médicament dans le corps humain sur la base des données in vitro et in vivo du médicament ainsi que des caractéristiques physiopathologiques du sujet virtuel. Une adaptation de dose permettra de réduire les conséquences cliniques de ces interactions.

Drug metabolizing enzymes, such as cytochromes P450, and transporters, such as P-glycoprotein, play a crucial role in drug disposition. Their activity is subject to great interindividual variability which can cause differences in drug concentrations and result in therapeutic failure or side effects. In this thesis, a cocktail approach allowing for simultaneous assessment of the in vivo activity of these proteins has been developed. In a first step, an LC/MS-MS method for the quantification of the cocktail probes and their metabolites in DBS has been developed and validated. A first clinical study showed that modulation of cytochromes and P-glycoprotein activities can be reliably predicted using this approach. A second clinical study was performed in order to verify the interaction potential between the probe drugs within the cocktail and thus fully validate the latter. Few clinical study and clinical practice examples of the utility and application of this approach are also presented.

L'acénocoumarol est un anticoagulant oral dérivé de la coumarine dont le mode d'action repose sur l'inhibition d'une enzyme-clé du cycle de régénération de la vitamine K, la sous-unité 1 de la vitamine K époxyde réductase. Son utilisation est rendue complexe par les causes suivantes: sa marge thérapeutique étroite, son métabolisme dépendant des cytochromes P450 et sa susceptibilité aux interactions médicamenteuses. La conséquence de ces différents facteurs est l'apparition d'effets indésirables se traduisant le plus souvent par des hémorragies. L'acénocoumarol est métabolisé par les cytochromes CYP2C9, 2C19 et 1A2. Ceux-ci et plus particulièrement les CYP2C9 et 2C19 sont sujets à des polymorphismes génétiques (génotypes) influençant leur activité (phénotypes). Sur la base de celle-ci, des individus peuvent être considérés comme étant métaboliseurs rapides, normaux, intermédiaires ou lents. Le but de ce travail consiste à effectuer une étude de corrélation génotype-phénotype pour les CYP2C9 et 2C19 dans une cohorte de patients anticoagulés par acénocoumarol. Une corrélation est obtenue lorsque le phénotype prédit par le génotype correspond au phénotype mesuré.

En Europe, 19 % des adultes souffrent de douleurs chroniques, qui affectent leur vie sociale et professionnelle. Ces douleurs peuvent être d'origine nociceptive (lésions des tissus périphériques) ou neuropathique (dysfonctionnement du système nerveux central ou périphérique). L'impact psychologique engendré est important. En fonction des publications,20 à 50 % des patients douloureux chroniques développent un épisode dépressif majeur. La médication suit les directives de l'OMS qui a défini 3 paliers antalgiques en fonction de l'intensité des douleurs. Les opioïdes constituent la médication de choix pour les douleurs d'intensité modérée à sévère. Leur utilisation remonte au 19 ème siècle, et connait actuellement une augmentation notable. La polémique naissante au sujet de leur emploi dans les douleurs chroniques a attiré notre attention. En 2008, une étude aux Etats-Unis a démontré que 73,8 % des décès suite à une overdose médicamenteuse pouvaient être attribués aux opioïdes. Le but de ce travail est d'évaluer les caractéristiques des patients, de leurs douleurs et des médicaments antalgiques consommés, par le biais d'un questionnaire.

Cette étude vise à étudier le risque lié à la prise des médicaments pendant la grossesse et l'issue de cette grossesse, et à évaluer le rapport bénéfice/risque d'un traitement médicamenteux reçu pendant la grossesse à partir d'un collectif de consultations issue du service de pharmacologie et toxicologie cliniques des HUG. Notre but principal est de calculer le taux de malformations, d'avortements, de fausses couches et de toutes autres complications dues à l'exposition médicamenteuses durant la grossesse. Une analyse des classes de médicaments les plus souvent représentés dans ce collectif selon la classe ATC (classification anatomique, thérapeutique et chimique) et la catégorie de risques pendant la grossesse et leur impact sur la grossesse sera également effectuée. Enfin le taux de malformation sera comparé avec celui exprimé chez une population générale.

Une des grandes classes de médicaments antiplaquettaires utilisés dans les maladies cardiovasculaires est la famille des thiénopyridines. Une revue de la littérature approfondie a tout d'abord été réalisée sur la pharmacologie et la variabilité de réponse interindividuelle de ces médicaments. La deuxième partie montre les études in vitro qui ont été effectuées pour caractériser le métabolisme de ces agents antiplaquettaires. Les interactions médicamenteuses entre les thiénopyridines et les antirétroviraux comme le ritonavir ont ensuite été étudiées in vitro puis in vivo avec la réalisation d'une étude clinique sur des volontaires sains. Les résultats de ces études ont confirmé une interaction pharmacocinétique entre le ritonavir et le prasugrel liée à l'inactivation du CYP3A par le ritonavir, provoquant une diminution de la production de métabolite actif du prasugrel. Dans une dernière partie, le potentiel inducteur/inhibiteur du CYP3A4 par les anti-VIH a été investigué à l'aide d'une nouvelle lignée cellulaire hépatique.

Clinical studies have pointed out a possible interaction between aspirin and omeprazole. In a recent study in aspirin treated patients, proton pump inhibitors (PPIs) were linked to an increased risk of adverse cardiovascular events. A recent proteomic study has described the presence of proton pumps in human platelets. This work looked into the modulation of aspirin antiplatelet effect by omeprazole and the effect of omeprazole alone on platelet activity. Incubation of omeprazole with aspirin, compared to aspirin alone, showed no significant difference on thromboxane A2 generation (p>0.05). However, it significantly (p>0.05) decreased platelet aggregation and granular secretion. Incubation of omeprazole in plasma caused a significant decrease of TxA2 generation and platelet granular secretion compared to control conditions, but aggregation remained unchanged. Omeprazole may enhance the antiplatelet effect of aspirin by another mechanism than COX-1 inhibition. The hypothesis about proton pumps role in the growth and stabilisation of the platelet plug is a given, but further studies are needed.

La présente thèse s'inscrit dans un contexte global d'amélioration de la qualité et de la sécurité de la prescription de l'acénocoumarol au sein des Hôpitaux Universitaires de Genève. Dans un premier temps, un état des lieux de la prescription de cet anticoagulant oral a été réalisé. Le deuxième grand volet de ce travail a été consacré à la réalisation d'une étude clinique observationnelle de cohorte. Cette étude avait pour but d'évaluer l'impact de certaines variations génétiques sur la réponse au traitement anticoagulant ainsi que sur les interactions médicamenteuses chez des patients hospitalisés débutant un traitement d'acénocoumarol. Finalement, une étude de transport in vitro est venue clore ce travail de recherche afin d'étudier le transport par la P-glycoprotéine des trois principaux dérivés coumariniques utilisés à travers le monde (acénocoumarol, warfarine, phenprocoumone) ainsi que du rivaroxaban, un anticoagulant oral apparu récemment sur le marché.

Le clopidogrel est une prodrogue qui doit être bioactivée par les cytochromes P450 pour pouvoir inhiber l'agrégation plaquettaire. Une grande variabilité de la réponse à ce médicament est observée. L'origine de cette variabilité est multifactorielle avec une composante génétique et environnementale. Les CYP2C19, 3A4, 3A5 et 2B6 sont les plus impliqués dans la bioactivation du clopidogrel. Le ritonavir est un puissant inhibiteur des cytochromes P450 3A4/5 et 2B6. On suppose qu'une administration simultanée du clopidogrel et du ritonavir aurait un effet plus prononcé chez un métaboliseur lent du CYP2C19 que chez un métaboliseur rapide. Le but de ce travail est l'évaluation in vitro de l'interaction entre le clopidogrel et le ritonavir en utilisant des microsomes hépatiques humains normaux et déficients en CYP2C19

Les effets indésirables médicamenteux (EIMs) sont des événements qui apparaissent fréquemment chez les patients hospitalisés, et qui peuvent également être à l'origine de l'admission de ces patients aux soins intensifs. Un certain nombre de facteurs de risque sont identifiés comme étant des éléments favorisant la survenue des EIMs. Le but de ce travail consiste à étudier la fréquence des interactions médicamenteuses et d'observer si elles constituent un facteur de risque de développer un EIM. A partir des données de l'étude sur le rôle du médicament dans les motifs de transfert de l'étage aux soins intensifs, une analyse centrée sur les interactions est réalisée. Les traitements médicamenteux de chacun des patients inclus dans l'étude sont analysés en recherchant les interactions médicamenteuses au moyen du logiciel Lexi-Interact™ Online. Le nombre de patients qui présentaient une interaction médicamenteuse, le nombre d'interactions par patient et le type d'interactions ont été spécifiquement relevés. Les différences entre les groupes ont été évaluées en appliquant deux tests paramétriques ; le test de t de Student et le test du Khi carré. Pour les variables présentant une distribution asymétrique, le test non paramétrique de Mann-Whitney est réalisé. Le groupe admis pour un EIM est comparable au groupe contrôle hormis le nombre moyen de médicaments administrés et la survenue d'une insuffisance rénale aiguë. Il ressort aussi, que le nombre de patients avec au moins une interaction est plus élevé dans le groupe de patients avec EIMs sans toutefois démontrer une différence significative entre les groupes (p=0.14). Les patients admis avec un EIM présentent en moyenne 11.8 ± 8.9 interactions par patient alors que les patients admis sans EIMs sont sujets à une moyenne de 8.3 ± 7.0 interactions (p = 0.0012). Les interactions pharmacodynamiques et les interactions de catégorie de risque C, c'est-à-dire qui nécessitent un suivi du patient, se manifestent le plus souvent. Parmi les interactions pharmacocinétiques, les interactions se situent essentiellement au niveau des CYP3A4, 2C9 et 1A2. Les interactions médicamenteuses sont l'un des facteurs de risque qui participe au transfert des patients des unités de soins aux soins intensifs, en association avec la polymédication et l'insuffisance rénale aiguë. Le poids de ces interactions comme facteur de risque est encore à déterminer.

Il s'agit d'une étude rétrospective comparant des toxicomanes actifs ou inactifs hospitalisés aux Hôpitaux Universitaires de Genève entre janvier 1999 et juin 2003 et bénéficiant ou non d'une substitution par méthadone. Un total de 167 patients recevant de la méthadone ont été inclus et comparés à un groupe contrôle de 80 toxicomanes sans substitution. Différentes variables démographiques, biologiques et pharmacologiques ont été analysées. Dans le groupe méthadone, la prévalence de l'allongement du QTc de 0.5 secondes ou plus était de 16.2% versus 0 dans le groupe contrôle. La dose de méthadone, la présence d'inhibiteurs du cytochrome P-450 3A4, la valeur de potassium et la fonction hépatique ont été identifiés comme facteurs contribuant à l'allongement du QT.