No description available.

La warfarine est un médicament anticoagulant possédant un faible index thérapeutique et une grande variabilité intra et interindividuelle dans la réponse au traitement. Les facteurs déterminants de la réponse à la warfarine ne sont pas tous connus et la présente étude vise à tester l'hypothèse que la pratique régulière d’activité physique puisse y être associée. Nous avons évalué si l’activité physique, mesurée à l’aide de 2 questionnaires différents, était associée à la dose de warfarine et au pourcentage de temps passé à l'intérieur de l'intervalle thérapeutique ciblé (time in therapeutic range : TTR). L’étude a été menée chez les 1064 participants de la Cohorte warfarine de l’Institut de Cardiologie de Montréal (ICM) et chez 618 utilisateurs de warfarine issus de la Biobanque de l’ICM. Nous avons trouvé que, dans les deux cohortes, les patients actifs nécessitaient une dose hebdomadaire moyenne plus élevée que les patients inactifs. L’association perdurait lorsque le modèle statistique était ajusté pour différentes variables connues pour influencer la réponse à la warfarine, telles que le génotype aux gènes CYP2C9 et VKORC1, l’âge, la taille, le poids, et l’INR ciblé. L’INR ciblé est décidé par le médecin et il correspond généralement à 2,0 – 3,0 ou 2,5 – 3,5. Les patients de la Cohorte warfarine avaient aussi plus de chances d’avoir un TTR inférieur à 60%, donc d’être moins stables. La pratique régulière d’activité physique est donc un facteur déterminant de la dose thérapeutique de warfarine et la pratique d'activité physique intensive est associée à un TTR plus faible. / Warfarin is an oral anticoagulant agent with a narrow therapeutic index. Dosing of warfarin is highly variable among patients and it may also vary in time for the same patient. All factors influencing warfarin response are not known and this study aims to elucidate if regular physical activity is one important factor. We evaluated whether warfarin dosage and the time in therapeutic range (TTR) were associated with RPA. RPA was measured via 2 different questionnaires. The study was conducted by using 1,064 patients in the Quebec Warfarin Cohort (QWC) and 618 patients from the Genetic-Hospital Cohort. Both cohorts are hosted at the Montreal Heart Institute. In both cohorts, we found that active patients required higher weekly doses of warfarin than inactive patients. The association was maintained when the model was adjusted for variables known to influence warfarin response, i.e. CYP2C9 and VKORC1 genotypes, age, height, weight and targeted INR. The targeted INR is fixed by the physicist and is usually between 2.0 and 3.0 or between 2.5 or 3.5. Active patients from the QWC were more susceptible to have a TTR inferior to 60%, i.e. to be unstable. RPA influences warfarin response and intensive RPA is associated with a greater instability in treatment response.

Les effets indésirables médicamenteux (EIM) ont des répercussions considérables tant sur la santé que sur l'économie. De 1,9% à 2,3% des patients hospitalisés en sont victimes, et leur coût a récemment été estimé aux alentours de 400 millions d'euros pour la seule Allemagne. De plus, les EIM sont fréquemment la cause du retrait d'un médicament du marché, conduisant à des pertes pour l'industrie pharmaceutique se chiffrant parfois en millions d'euros.De multiples études suggèrent que des facteurs génétiques jouent un rôle non négligeable dans la réponse des patients à leur traitement. Cette réponse comprend non seulement les effets thérapeutiques attendus, mais aussi les effets secondaires potentiels. C'est un phénomène complexe, et nous nous tournons vers l'apprentissage statistique pour proposer de nouveaux outils permettant de mieux le comprendre.Nous étudions différents problèmes liés à la prédiction de la réponse d'un patient à son traitement à partir de son profil génétique. Pour ce faire, nous nous plaçons dans le cadre de l'apprentissage statistique multitâche, qui consiste à combiner les données disponibles pour plusieurs problèmes liés afin de les résoudre simultanément.Nous proposons un nouveau modèle linéaire de prédiction multitâche qui s'appuie sur des descripteurs des tâches pour sélectionner les variables pertinentes et améliorer les prédictions obtenues par les algorithmes de l'état de l'art. Enfin, nous étudions comment améliorer la stabilité des variables sélectionnées, afin d'obtenir des modèles interprétables. / Adverse drug reaction (ADR) is a serious concern that has important health and economical repercussions. Between 1.9%-2.3% of the hospitalized patients suffer from ADR, and the annual cost of ADR have been estimated to be of 400 million euros in Germany alone. Furthermore, ADRs can cause the withdrawal of a drug from the market, which can cause up to millions of dollars of losses to the pharmaceutical industry.Multiple studies suggest that genetic factors may play a role in the response of the patients to their treatment. This covers not only the response in terms of the intended main effect, but also % according toin terms of potential side effects. The complexity of predicting drug response suggests that machine learning could bring new tools and techniques for understanding ADR.In this doctoral thesis, we study different problems related to drug response prediction, based on the genetic characteristics of patients.We frame them through multitask machine learning frameworks, which combine all data available for related problems in order to solve them at the same time.We propose a novel model for multitask linear prediction that uses task descriptors to select relevant features and make predictions with better performance as state-of-the-art algorithms. Finally, we study strategies for increasing the stability of the selected features, in order to improve interpretability for biological applications.

There has been tremendous advancement in sequencing technologies; the rate at which sequencing data can be generated has increased multifold while the cost of sequencing continues on a downward descent. Sequencing data provide novel insights into the ecological environment of microbes as well as human health and disease status but challenge investigators with a variety of computational issues. This thesis focuses on three common problems in the analysis of high-throughput data. The goals of the first project are to (1) develop a statistical framework and a complete software pipeline for metagenomics that identifies microbes to the strain level and thus facilitating a personalized drug treatment targeting the strain; and (2) estimate the relative content of microbes in a sample as accurately and as quickly as possible. The second project focuses on the analysis of the microbiome variation across multiple samples. Studying the variation of microbiomes under different conditions within an organism or environment is the key to diagnosing diseases and providing personalized treatments. The goals are to (1) identify various statistical diversity measures; (2) develop confidence regions for the relative abundance estimates; (3) perform multi-dimensional and differential expression analysis; and (4) develop a complete pipeline for multi-sample microbiome analysis. The third project is focused on batch effect analysis. When analyzing high dimensional data, non-biological experimental variation or “batch effects” confound the true associations between the conditions of interest and the outcome variable. Batch effects exist even after normalization. Hence, unless the batch effects are identified and corrected, any attempts for downstream analyses, will likely be error prone and may lead to false positive results. The goals are to (1) analyze the effect of correlation of the batch adjusted data and develop new techniques to account for correlation in two step hypothesis testing approach; (2) develop a software pipeline to identify whether batch effects are present in the data and adjust for batch effects in a suitable way. In summary, we developed software pipelines called PathoScope, PathoStat and BatchQC as part of these projects and validated our techniques using simulation and real data sets.

Dans la lutte contre le cancer, la médecine personnalisée est une stratégie nécessaire et très prometteuse. En effet, il est primordial de pouvoir tester l'innocuité et l'efficacité de médicaments anticancéreux sur des échantillons provenant du patient lui-même, du fait de la diversité des réponses entre patients. Le but est d'améliorer la performance des soins et d'éviter des traitements inutiles et même parfois nocifs pour le patient. Ainsi, l'exemple de la chimiothérapie illustre parfaitement cette stratégie. Le cout élevé des molécules thérapeutiques, la nocivité de ces molécules et les réponses variées des patients face à une même molécule implique le recours aux tests de ces molécules sur un échantillon provenant du patient lui-même. Il en résulte un intérêt croissant dans le développement de tests simples, robustes et peu couteux permettant l'évaluation de la chimio sensibilité de cellules biologiques issues d'une biopsie. Les problématiques liées à la mise en place de tels tests sont la quantité de cellules disponibles dans une biopsie, la diversité des molécules thérapeutiques à tester et également le choix d'une technique de détection permettant de suivre la cinétique d'action des molécules sur les cellules biologiques. L'une des réponses à la faible quantité de cellules est le développement de tests dans des environnements microfluidiques qui nécessitent donc l'intégration et la miniaturisation d'une technique de détection. La stratégie qui sera étudiée dans cette thèse est l'utilisation de l'impedancemetrie par le biais d'électrodes inter digitées d'Oxyde d'Etain et d'Indium (ITO) pour l'analyse quantitative de l'état de cellules de cancer du sein pour des applications de criblage de médicaments anticancéreux. Ce matériau présente l'avantage d'être transparent permettant ainsi des mesures d'impédance qui pourrait être couplées à des mesures optiques dans un environnement microfluidique. Dans une première partie, nous avons caractérisé et étudié des électrodes inter digitées d'or et d'ITO pour des mesures d'impédance avec des cellules cancéreuses. Cette caractérisation par spectroscopie d'impédance réalisée dans des solutions de milieu de culture en présence et absence de cellules, ont permis de démontrer que la différence de sensibilité entre ces deux matériaux provenait à la fois d'une différence de comportement résistif mais également d'une différence d'impédance interfaciale, dans les deux cas à la défaveur de l'ITO. Après ce constat, nous avons donc poursuivi l'étude afin d'évaluer les capacités de l'ITO pour des mesures de chimio sensibilité de la molécule 5-fluorouracil et également proposé une stratégie pour améliorer la sensibilité de l'ITO tout en conservant sa transparence. La stratégie développée consiste en la modification de la surface de d'électrodes d'ITO avec de l'oxyde d'iridium pour améliorer la sensibilité de l'ITO, tout en gardant sa transparence. Cette approche est intéressante pour pouvoir concevoir un dispositif transparent et facile à coupler avec un système d'observation microscopique dans un environnement microfluidique / In the fight against cancer, personalized medicine is a necessary and very promising strategy. In fact, it is essential to be able to test the safety and effectiveness of anticancer drugs on samples from the patient, due to the diversity of responses between patients. The aim is to improve the performance of health care and avoid unnecessary and sometimes harmful treatments. Thus, chemotherapy perfectly illustrates this strategy. The high cost of therapeutic molecules, the harmfulness of these molecules and the varied responses of patients involve the use of tests with chemotherapeutic molecules on samples coming from cancer patients. This results in a growing interest in the development of simple, robust and inexpensive tests for assessing the chemo sensitivity of biological cells from a biopsy. The problems related to carrying out such tests are the quantity of cells available in a biopsy, the diversity of the therapeutic molecules to be tested and also the choice of a detection technique, able to monitor the kinetics of action of the molecules on the biological cells. One solution to the small amount of cells is to carry out the tests in microfluidic environments which therefore require the integration and miniaturization of a detection technique. The strategy that will be studied in this thesis is the use of electrical impedance with interdigitated electrodes of indium tin oxide (ITO) for the quantitative analysis of the state of breast cancer cells for screening applications of anticancer drugs. This material has the advantage of being transparent allowing impedance measurements that could be coupled to optical measurements in a microfluidic environment. In the first part, we characterized and studied interdigitated electrodes of gold and ITO for impedance measurements with cancer cells. This impedance spectroscopy characterization carried out in culture medium solutions, in the presence and absence of cells, demonstrated that the difference in sensitivity between these two materials comes from a difference in resistive behavior and also from a difference in interfacial impedance, in both cases to the disadvantage of ITO. After this, we continued the study to evaluate the capabilities of ITO for chemosensitivity measurements using the molecule 5 fluorouracil and we suggested a strategy to improve the sensitivity of ITO while maintaining its transparency. The strategy developed consists of modifying the surface of ITO electrodes with iridium oxide to improve the sensitivity of the ITO, while keeping its transparency. This approach is interesting for developing a transparent device and easy to couple with a microscopic observation system in a microfluidic environment

No description available.

Ziel: Entwicklung eines gesundheitsökonomischen Evaluationskonzepts zum Nachweis einer Kostenreduktion unter gleichzeitiger Optimierung des medizinischen Nutzens durch pharmakogenetisches Screening bei Erstdiagnose einer Schizophrenie. Finale Zielsetzung ist die Aufnahme der pharmakogenetischen a priori-Diagnostik für die Indikation Schizophrenie in die GKV-Regelversorgung. Methodik: Basierend auf dem aktuellen Stand gendiagnostischer Forschung sowie der evidenzbasierten Schizophrenietherapie wurde eine prospektive, randomisierte und kontrollierte, dreiarmige, offene, multizentrische Pilotstudie im Paralleldesign über 3 Jahre konzeptioniert. Studienpopulation: 300 Patienten (1:1:1) im Alter von 18 bis 65 Jahren mit erstmaliger F20-Diagnose (ICD-10). Interventionen: pharmakogenetisches Screening und integrierte Versorgung; integrierte Versorgung; Standardversorgung. Die Erhebung des medizinischen Nutzens erfolgt durch Messung des klinischen Outcome bzgl. der patientenrelevanten Endpunkte Mortalität, Morbidität, Lebensqualität und Nebenwirkungen zu definierten Zeitpunkten. Perspektivisch relevante Kosten werden im "piggy back"-Verfahren ermittelt. Ergebnisse: Angesichts zurzeit bestehender Limitationen im deutschen Gesundheitssystem (z.B. unzureichendes intersektorales Schnittstellenmanagement bei der Arzneimittelversorgung und Informationsweitergabe) wird die Integrierte Versorgung als geeignete Versorgungsform für den Nutzennachweis eingestuft. Die Integrierte Versorgung stellt jedoch momentan nicht den allgemeinen Standard der psychiatrischen Patientenversorgung dar. Aus GKV-Perspektive wesentliche Kostentreiber der Schizophrenietherapie sind Rückfälle, Krankenhausaufenthalte, Arbeitslosigkeit und vorzeitige Verrentung. Eine Verringerung der Häufigkeit dieser Parameter könnte z.B. zu einer Reduktion der Erstjahres-Behandlungskosten (zurzeit ca. 30% der Gesamtkosten) führen. Die Kosten-Effektivitäts-Analyse erweist sich als Studienform mit der geringsten Anfälligkeit für Bias und Confounder. Trotz einer vergleichsweise hohen externen Validität ist das Studiensetting nicht uneingeschränkt übertragbar auf die Versorgungsrealität des deutschen Gesundheitssystems. Es existiert aktuell keine generelle Empfehlung für den Einsatz der Gendiagnostik zur Steuerung der Arzneimitteltherapie in Psychiatrie. Ebenso hat die integrierte Versorgung bisher keinen umfassenden Einzug in den psychiatrischen Behandlungsalltag gefunden, so dass die beschriebenen Limitationen einen positiven Nutzennachweis erschweren. Dennoch ist das Konzept als praktisch umsetzbar zu bewerten. Schlussfolgerung: Bei dieser Faktenlage ist das Interesse der GKV an der Veranlassung einer gesundheitsökonomischen Evaluation mit dem Ziel einer Erstattungsfähigkeit des a priori durchgeführten pharmakogenetischen Screenings bei Schizophrenie als eher gering einzustufen. Jedoch lassen das Update der S3-Praxisleitlinie mit dem Einbezug der strukturierten u. integrierten Versorgung sowie der Aktionsplan „Individualisierte Medizin“ des Bundesforschungsministeriums auf eine Fokussierung auf diese Fragestellung und veränderte Interessenlage bzgl. der Initiierung der Pilotstudie hoffen. Weitere Forschungstätigkeit sowie die praktische Erprobung neuer gendiagnostischen Verfahren sind, basierend auf versorgungsbezogenen Pilotstudien wie der hier konzeptionierten, fachübergreifend erforderlich, um die Relevanz der Methodik für den psychiatrischen Versorgungsalltag zu belegen. / Purpose: Development of a health-economic investigation method to study whether a cost reduction under concurrent optimisation of the medical use exists by using pharmacogenetic a- priori- screening with first diagnosis of a schizophrenia. Final objective is the reimbursement of pharmacogenetic diagnostics for the indication schizophrenia in the German health statutory insurance (GKV). Methods: A prospective, randomised and controlled, 3-armed, parallel, open, multicentre pilot study with a duration of 3 years was designed based on the actual status of genetic-diagnostic research as well as the evidence-based therapy of schizophrenia. Study population: 300 patients (1:1:1) aged 18 to 65 years with initial F20 diagnosis (ICD-10). Interventions: pharmacogenetic screening and integrated care; integrated care; standard care. For evaluation of the medical benefit the clinical outcome is measured at defined times with regard to the patients' relevant endpoints mortality, morbidity, quality of life and side effects. In perspective relevant costs are determined by "piggy back" procedure. Results: In view of actually existing limitations within the German health system (e.g., insufficient intersectional medication and information management) the integrated care is considered being a suitable setting to demonstrate the advantage of using pharmacogenetic screening. Nevertheless, the integrated care does not show the general standard of the psychiatric patient's care at the moment. From GKV perspective essential cost drivers of schizophrenia therapy are relapses, hospital stays, unemployment and untimely superannuation. Diminishing the rate of these parametres could lead, e.g., to a reduction of the first year medical costs (at the moment approx. 30% of the total expenses). The cost-effectiveness analysis seems to be the study form with the slightest susceptibility to bias and confounding. In spite of a relatively high external validity the study setting is not unconditionally transferable to the German health system. Currently no general recommendation exists for the application of the genetic diagnostics to manage medication therapy in psychiatry. Up to now also the integrated care has not found a comprehensive entry in psychiatric practice, so that the described limitations are complicating a positive use proof. Nevertheless, the investigational concept can be regarded as feasible. Conclusion: Based on the existing situation the GKV's interest in performing a health-economic evaluation, which is focussed on the reimbursement of pharmacogenetic a priori-diagnostics in schizophrenia, is considered to be low. However, the situation may change in view of the expected update of the S3-practise guideline with the focus on structured and integrated care as well as the action plan „individualised medicine“ of the German federal research ministry. Thus, there is hope for changing interests in a pilot study. Based on care-related pilot studies as presented here, further research activities and practical testing of recent gene diagnostic procedures are necessary to demonstrate the relevance of the methodology for psychiatric practice.

O transtorno de humor associado ao transtorno de estresse pós-traumático (TEPT) tem, em geral, desfechos clínicos mais graves. Embora essa associação esteja consolidada no transtorno depressivo, esse não é o caso no transtorno bipolar (TB). Os poucos estudos realizados acerca dessa comorbidade demonstraram piora na qualidade de vida e aumento no número de tentativas de suicídio associado aos pacientes com TEPT e TB. Nenhum estudo, entretanto, avaliou o impacto do TEPT em características centrais do TB, como o número de episódios de humor ou funcionamento psicossocial. Por outro lado, estudos pré-clínicos sugerem que existe um fenômeno de sensibilização cruzada entre o TB e o TEPT. Foi proposto que a redução do brain-derived neurotrophic fator (BDNF) e alterações de marcadores inflamatórios poderiam ser mecanismos associados a esta sensibilização. Com relação a última hipótese, as alterações em marcadores inflamatórios estão bem consolidadas no TB, porém esse não é o caso no TEPT. Com a finalidade de aprofundar o estudo dessas questões, os dois primeiros estudos que compõem essa tese abordaram aspectos clínicos e biológicos relacionados a esses transtornos. O primeiro é uma metanálise e metaregressão que demonstrou que o TEPT está associado a níveis aumentados de IL-6, IL-1β, TNF-α e interferon-γ na circulação periférica. Além disso, o tempo de doença foi positivamente associado aos níveis de IL-1β, enquanto a gravidade dos sintomas foi positivamente associada aos níveis de IL-6. Esses achados podem não só apresentar um novo mecanismo biológico para explicar o fenômeno de sensibilização cruzada entre TEPT e TB, mas também abre novos horizontes na busca de novas estratégias terapêuticas e diagnósticas para o TEPT. O segundo foi um estudo caso-controle que avaliou características clínicas centrais do TB e demonstrou pela primeira vez que indivíduos com TB e comorbidade com TEPT apresentam mais episódios maníacos e pior funcionamento psicossocial. Ademais, o início dos episódios maníacos e o uso de substâncias de abuso surgiram de maneira mais precoce nesses indivíduos. Além desses dois estudos, optamos por abordar a influência da comorbidade com os transtornos ansiosos (como o TEPT) em um desfecho clínico trágico em pacientes com transtorno de humor: o suicídio. Esse estudo foi realizado em uma população diferente da do estudo clínico anterior, incluindo pacientes com depressão maior ou com TB, e foi focado em apenas um desfecho. Nosso objetivo era criar uma ferramenta que pudesse gerar um escore de probabilidade para cada paciente com transtorno de humor que refletisse o risco de tentar suicídio. Utilizando técnicas de machine learning, demonstramos que o risco de suicídio em indivíduos com transtorno de humor pode ser estimado objetivamente a partir de variáveis clínicas facilmente obtidas e variáveis demográficas. Embora tenhamos encontrado uma boa acurácia (72%) e área (area under the curve) (77%), futuros estudos podem integrar dados de variáveis de marcadores biológicos (genética, neuroimagem, neurocognição, etc.), usando também técnicas de machine learning, para atingir uma acurácia ainda maior. O uso de um instrumento objetivo é o primeiro passo na busca de um tratamento mais personalizado para aqueles pacientes com alto risco de se suicidar. / The comorbidity between mood disorders and posttraumatic stress disorder (PTSD) is associated with poorer clinical outcomes. Although this association is well established in major depressive disorder, fewer studies included patients with bipolar disorder (BD). These studies report that patients with BD and PTSD show increased number of suicide attempts and worse quality of life. However, no clinical studies so far have reported the impact of comorbid PTSD on core features of BD, such as number of mood episodes and functional impairment. On the other hand, preclinical studies showed that there is a cross-sensitization between BD and PTSD. Accordingly, it was proposed that the brain-derived neurotrophic factor (BDNF) and changes of inflammatory markers might be the biological underpinnings of this cross-sensitization. Although the changes in inflammatory markers are well established in BD, this is not the case for PTSD. In view of the above, the first two studies of this thesis addressed clinical and biological issues related to BD and PTSD. The first is a meta-analysis and metaregression study that showed increased levels of IL-6, IL-1β, TNF-α and interferon-γ in peripheral circulation among patients with PTSD. Furthermore, illness duration was positively associated with IL-1β levels, while the severity of the symptoms was positively associated with IL-6 levels. These findings may not only provide a new biological mechanism to explain the cross-sensitization between PTSD and BD, but also opens a new avenue in the search for new therapeutic targets and diagnostic strategies for PTSD. The second was a case-control study that assessed core clinical features of BD showed increased manic episodes and functional impairment among patients with BD and comorbid PTSD. In addition, those patients were younger when they started the manic episodes and substance use. Moreover, we chose to address the influence of comorbid anxiety disorders (such as PTSD) in suicide in patients with mood disorders. This study was conducted in a different population, including patients with major depressive disorder or BD, and it was focused on only one outcome. Suicide is a tragic clinical outcome, but highly preventable. Our aim was to develop a clinical tool using machine learning techniques to estimate a probability score at individual level to stratify the risk of a patient with a mood disorder attempts suicide. Therefore, our study showed that the risk of suicide in individuals with a mood disorder can be objectively estimated from easily assessed clinical and demographic variables. Although we have found a good accuracy (72%) and area under the curve (77%), future studies may integrate data from biological markers (genetic, neuroimaging, neurocognition, etc.) also using machine learning techniques to achieve a higher accuracy. This objective instrument is the first step towards a more personalized treatment for those patients at high risk of suicide.

Acute leukemias (AL) comprise a heterogeneous group of hematologic malignancies, and individual patient responses to treatment can be difficult to predict. Monitoring of minimal residual disease (MRD) is thus very important and holds great potential for improving treatment strategies. Common MRD targets include immunoglobulin heavy chain or T-cell receptor gene rearrangements, recurrent cytogenetic abnormalities and mutations in important hematological genes. Whereas in the majority of adult acute lymphoblastic leukemia patients a suitable MRD target can be identified, in adult acute myeloid leukemia patients well-characterized targets are found in only half of cases. The identification of new specific molecular markers of leukemic blasts for MRD assessment, particularly in AML patients, is therefore highly desirable. Our aim was to develop a flexible strategy for mapping of cytogenetically identified unique clone-specific abnormalities down to the single nucleotide level and, based on the sequence, design a specific real-time PCR assay for MRD assessment in AL patients without any previously described MRD marker. Using a combination of cytogenetic (chromosome banding, chromosome microdissection), molecular cytogenetic (mFISH, mBAND) and molecular biological (next- generation sequencing, long-range...

O transtorno de humor associado ao transtorno de estresse pós-traumático (TEPT) tem, em geral, desfechos clínicos mais graves. Embora essa associação esteja consolidada no transtorno depressivo, esse não é o caso no transtorno bipolar (TB). Os poucos estudos realizados acerca dessa comorbidade demonstraram piora na qualidade de vida e aumento no número de tentativas de suicídio associado aos pacientes com TEPT e TB. Nenhum estudo, entretanto, avaliou o impacto do TEPT em características centrais do TB, como o número de episódios de humor ou funcionamento psicossocial. Por outro lado, estudos pré-clínicos sugerem que existe um fenômeno de sensibilização cruzada entre o TB e o TEPT. Foi proposto que a redução do brain-derived neurotrophic fator (BDNF) e alterações de marcadores inflamatórios poderiam ser mecanismos associados a esta sensibilização. Com relação a última hipótese, as alterações em marcadores inflamatórios estão bem consolidadas no TB, porém esse não é o caso no TEPT. Com a finalidade de aprofundar o estudo dessas questões, os dois primeiros estudos que compõem essa tese abordaram aspectos clínicos e biológicos relacionados a esses transtornos. O primeiro é uma metanálise e metaregressão que demonstrou que o TEPT está associado a níveis aumentados de IL-6, IL-1β, TNF-α e interferon-γ na circulação periférica. Além disso, o tempo de doença foi positivamente associado aos níveis de IL-1β, enquanto a gravidade dos sintomas foi positivamente associada aos níveis de IL-6. Esses achados podem não só apresentar um novo mecanismo biológico para explicar o fenômeno de sensibilização cruzada entre TEPT e TB, mas também abre novos horizontes na busca de novas estratégias terapêuticas e diagnósticas para o TEPT. O segundo foi um estudo caso-controle que avaliou características clínicas centrais do TB e demonstrou pela primeira vez que indivíduos com TB e comorbidade com TEPT apresentam mais episódios maníacos e pior funcionamento psicossocial. Ademais, o início dos episódios maníacos e o uso de substâncias de abuso surgiram de maneira mais precoce nesses indivíduos. Além desses dois estudos, optamos por abordar a influência da comorbidade com os transtornos ansiosos (como o TEPT) em um desfecho clínico trágico em pacientes com transtorno de humor: o suicídio. Esse estudo foi realizado em uma população diferente da do estudo clínico anterior, incluindo pacientes com depressão maior ou com TB, e foi focado em apenas um desfecho. Nosso objetivo era criar uma ferramenta que pudesse gerar um escore de probabilidade para cada paciente com transtorno de humor que refletisse o risco de tentar suicídio. Utilizando técnicas de machine learning, demonstramos que o risco de suicídio em indivíduos com transtorno de humor pode ser estimado objetivamente a partir de variáveis clínicas facilmente obtidas e variáveis demográficas. Embora tenhamos encontrado uma boa acurácia (72%) e área (area under the curve) (77%), futuros estudos podem integrar dados de variáveis de marcadores biológicos (genética, neuroimagem, neurocognição, etc.), usando também técnicas de machine learning, para atingir uma acurácia ainda maior. O uso de um instrumento objetivo é o primeiro passo na busca de um tratamento mais personalizado para aqueles pacientes com alto risco de se suicidar. / The comorbidity between mood disorders and posttraumatic stress disorder (PTSD) is associated with poorer clinical outcomes. Although this association is well established in major depressive disorder, fewer studies included patients with bipolar disorder (BD). These studies report that patients with BD and PTSD show increased number of suicide attempts and worse quality of life. However, no clinical studies so far have reported the impact of comorbid PTSD on core features of BD, such as number of mood episodes and functional impairment. On the other hand, preclinical studies showed that there is a cross-sensitization between BD and PTSD. Accordingly, it was proposed that the brain-derived neurotrophic factor (BDNF) and changes of inflammatory markers might be the biological underpinnings of this cross-sensitization. Although the changes in inflammatory markers are well established in BD, this is not the case for PTSD. In view of the above, the first two studies of this thesis addressed clinical and biological issues related to BD and PTSD. The first is a meta-analysis and metaregression study that showed increased levels of IL-6, IL-1β, TNF-α and interferon-γ in peripheral circulation among patients with PTSD. Furthermore, illness duration was positively associated with IL-1β levels, while the severity of the symptoms was positively associated with IL-6 levels. These findings may not only provide a new biological mechanism to explain the cross-sensitization between PTSD and BD, but also opens a new avenue in the search for new therapeutic targets and diagnostic strategies for PTSD. The second was a case-control study that assessed core clinical features of BD showed increased manic episodes and functional impairment among patients with BD and comorbid PTSD. In addition, those patients were younger when they started the manic episodes and substance use. Moreover, we chose to address the influence of comorbid anxiety disorders (such as PTSD) in suicide in patients with mood disorders. This study was conducted in a different population, including patients with major depressive disorder or BD, and it was focused on only one outcome. Suicide is a tragic clinical outcome, but highly preventable. Our aim was to develop a clinical tool using machine learning techniques to estimate a probability score at individual level to stratify the risk of a patient with a mood disorder attempts suicide. Therefore, our study showed that the risk of suicide in individuals with a mood disorder can be objectively estimated from easily assessed clinical and demographic variables. Although we have found a good accuracy (72%) and area under the curve (77%), future studies may integrate data from biological markers (genetic, neuroimaging, neurocognition, etc.) also using machine learning techniques to achieve a higher accuracy. This objective instrument is the first step towards a more personalized treatment for those patients at high risk of suicide.