Les biomarqueurs sanguins peuvent être utilisés pour définir le pronostic des patients ou permettre de déterminer les altérations moléculaires des cancers, et peut-être pouvoir guider les traitements de thérapies ciblées.Les cellules tumorales circulantes sont le reflet de la cascade métastatique et de la progression tumorale. La détection et la caractérisation des CTC est un domaine clé de la recherche dans le cancer. Cependant, il n’existe pas de méthode standard pour la détection des CTC, et le premier objectif de notre étude a été de comparer deux systèmes de détection des CTC basé sur l’expression de l’antigène EpCAM (CellSearch), ou la taille des cellules (ISET). Nos résultats montrent qu’il existe une bonne corrélation pour la détection des CTC dans les cancers du sein ou de la prostate, mais pas dans les cancers bronchiques. Ces résultats suggèrent qu’il est nécessaire de développer d’autres techniques de détection des CTC pour l’énumération et la caractérisation pour permettre une médecine de précision.A ce jour il n’existe aucun marqueur validé pour prédire l’efficacité des antiangiogéniques. Les CEC et CEP sont des marqueurs prometteurs. Dans notre étude, nous avons fait l’hypothèse que les CEC et les CEP pouvaient être pronostic de la survie des patients inclus dans les études de phases précoces. Nos résultats montrent qu’un taux élevé de CEP est associé à un mauvais pronostic, et que les CEP pourraient permettre de mieux sélectionner les patients. En conclusion, les marqueurs sanguins comme les CTC, les CEC ou les CEP peuvent être utilisés comme des facteurs pronostiques ou permettre une caractérisation moléculaire, et être une partie intégrante des programmes de médecine de précision. / Non-inasive biomarkers detected in the blood could be use for risk) stratification or molecular classification in advanced cancer patients, and could be a guide for molecular targeted therapies. Circulating tumor cells reflect the metastatic cascade and the cancer progression. The detection and molecular characterization of circulating tumor cells (CTCs) are a key area of translational cancer research. However, there is no universal method to detect CTC, and the primary objective of our study was to compare CTC detection systems based on the expression of the EpCAM antigen (CellSearch assay) or on cell size (ISET assay). Our results showed concordant results in CTC detection in breast and prosatet cancer patients, but not in lung cancer patients. These results suggest that we need to develop other CTC-detection techniques CTC for enumeration and characterization in order to to contribute to guiding specific targeted.To date, no biomarker has been validated for the prediction of efficacy of antiangiogenic agents in patients with advanced cancer. CEC and CEP counts have recently emerged as a potential candidate. In our study, we hypothesised that CEC and CEP are prognostic in patients enrolled in phase I. Our results showed that High CEP levels are associated with poor prognostics and could provide a new tool for patient selection in early anticancer drug trials.In conclusion, non invasive biomarkers such as CTC or CEC, CEP detectable in the blood could be used in the clinic as prognostic factors or surrogates for traditional tumor biopsies, and be a major component of precision medicine.

No description available.

No description available.

Le design standard des essais randomisés de phase III suppose le recrutement d'un grand nombre de patients pour assurer un risque α de 0.025 unilatéral et une puissance d'au moins 80%. Ceci s'avérer difficile dans les maladies rares, ou encore si le traitement cible une population spécifique définie par un sous-type moléculaire rare. Nous avons évalué par simulation la performance d'une série d'essais randomisés. Au terme de chaque essai, s'il est associé à une amélioration significative, le traitement expérimental devient le contrôle de l'essai suivant. Les designs ont été évalués pour différents taux de recrutement, différentes sévérités de la maladie, et différentes distributions hypothétiques des effets d'un futur traitement. Nous avons montré, que sous des hypothèses raisonnables, une série d'essais de plus petite taille et avec un risque α relâché est associée à un plus grand bénéfice à long terme que deux essais de design standard. Nous avons enrichi cette approche avec des designs plus flexibles incluant des analyses intermédiaires d'efficacité et/ou futilité, et des designs adaptatifs à trois bras avec sélection de traitement. Nous avons montré qu'une analyse intermédiaire avec une règle d'arrêt pour futilité était associé à un gain supplémentaire et à une meilleure maitrise du risque, contrairement aux règles d'arrêt pour efficacité qui ne permettent pas d'améliorer la performance. Les séries d'essais à trois bras sont systématiquement plus performants que les séries d'essais à deux bras. Dans la troisième de la thèse, nous avons étudié les essais randomisés évaluant un algorithme de traitement plutôt que l'efficacité d'un seul traitement. Le traitement expérimental est déterminé selon la mutation. Nous avons comparé deux méthodes basées sur le modèles de Cox à effets aléatoires pour l'estimation de l'effet traitement dans chaque mutation : Maximum Integrated Partial Likellihood (MIPL) en utilisant le package coxme et Maximum H-Likelihood (MHL) en utilisant le package frailtyHL. La performance de la méthode MIPL est légèrement meilleure. En présence d'un effet traitement hétérogène, les deux méthodes sousestime l'effet dans les mutations avec un large effet, et le surestime dans les mutations avec un modeste effet. / Large sample sizes are required in randomized trials designed to meet typical one-sided α-level of 0.025 and at least 80% power. This may be unachievable in a reasonable time frame even with international collaborations. It is either because the medical condition is rare, or because the trial focuses on an uncommon subset of patients with a rare molecular subtype where the treatment tested is deemed relevant. We simulated a series of two-arm superiority trials over a long research horizon (15 years). Within the series of trials, the treatment selected after each trial becomes the control treatment of the next one. Different disease severities, accrual rates, and hypotheses of how treatments improve over time were considered. We showed that compared with two larger trials with the typical one-sided α-level of 0.025, performing a series of small trials with relaxed α-levels leads on average to larger survival benefits over a long research horizon, but also to higher risk of selecting a worse treatment at the end of the research period. We then extended this framework with more 'flexible' designs including interim analyses for futility and/or efficacy, and three-arm adaptive designs with treatment selection at interim. We showed that including an interim analysis with a futility rule is associated with an additional survival gain and a better risk control as compared to series with no interim analysis. Including an interim analysis for efficacy yields almost no additional gain. Series based on three-arm trials are associated with a systematic improvement of the survival gain and the risk control as compared to series of two-arm trials. In the third part of the thesis, we examined the issue of randomized trials evaluating a treatment algorithm instead of a single drugs' efficacy. The treatment in the experimental group depends on the mutation, unlike the control group. We evaluated two methods based on the Cox frailty model to estimate the treatment effect in each mutation: Maximum Integrated Partial Likellihood (MIPL) using package coxme and Maximum H-Likelihood (MHL) using package frailtyHL. MIPL method performs slightly better. In presence of a heterogeneous treatment effect, the two methods underestimate the treatment effect in mutations where the treatment effect is large, and overestimates the treatment effect in mutations where the treatment effect is small.

No description available.

Precision cancer medicine is an area of research that involves complex bioinformatics software for the clinical analysis pipelines that sequencing data is passed through. A research team at Karolinska Institutet (KI) is running clinical trials to test new methods of precision cancer medicine and is working on creating a new web application that will be integrated with their two existing applications to aid bioinformaticians in their work of running pipelines and producing clinical reports for the hospitals. Recent publications show that web tools are used for improving usability of bioinformatics software, but have not discussed the inclusion of pipeline management into these systems. This study aims at exploring how the new web system can impact the complexity of the daily work within the trials for users with limited computational experience. This was done by developing a high-fidelity prototype of the new application using a user-centred design approach that included agile design work, usability tests and a focus group with the bioinformaticians at KI. The test sessions and focus group discussion indicate that the new application can speed up the weekly batch-wise sample processing by decreasing the need for users to interact with the command-line. It is also suggested that an integration of different applications can lead to a more efficient workflow within the team. The implication for the development of bioinformatics tools is that there are potentially important benefits to be found in making more usable interfaces for users with different experiences and areas of expertise. / Precisionsmedicin som cancerbehandling är ett forskningsområde som involverar komplex bioinformatisk mjukvara för de kliniska analyspipelines som sekvenseringsdata skickas igenom. En forskargrupp vid Karolinska Institutet (KI) driver kliniska studier för att testa nya metoder för precisionsmedicin mot cancer och arbetar med att skapa en ny webbapplikation som ska integreras med deras två befintliga applikationer för att hjälpa bioinformatiker i deras arbete med att köra pipelines och producera kliniska rapporter för sjukhusen. Nyligen publicerad forskning visar att webbverktyg används för att förbättra användbarheten av bioinformatisk programvara, men har inte berört hantering av pipelines som en del i dessa system. Denna studie syftar till att utforska hur det nya webbsystemet kan påverka tids- och komplexitetsaspekter i det dagliga arbetet inom de kliniska studierna. Detta gjordes genom att utveckla en high-fidelity prototyp av den nya applikationen, baserat på en användarcentrerad approach innefattande agilt designarbete, användbarhetstester och en fokusgrupp med bioinformatikerna på KI. Testsessionerna och fokusgruppsdiskussionen indikerar att den nya applikationen kan påskynda arbetet med de veckovisa omgångarna med prover genom att minska behovet för användare att interagera med kommandoraden. Det föreslås också att en integration av olika applikationer kan leda till ett effektivare arbetsflöde inom forskningsgruppen. Innebörden för utvecklingen av bioinformatiska verktyg är att det potentiellt kan finnas viktiga fördelar med att göra mer användbara gränssnitt för användare med olika erfarenheter och kompetensområden.

Early detection reduces mortality in lung cancer, but it is also considered as a challenge for oncologists and for healthcare systems. In addition, screening modalities like CT-scans come with undesired effects, many suspected patients are wrongly diagnosed with lung cancer. This thesis contributes to solve the challenge of early lung cancer detection by utilizing unique data consisting of self-reported symptoms. The proposed method is a predictive machine learning algorithm based on natural language processing, which handles the data as an unstructured data set. A replication of a previous study where a prediction model based on a conventional multivariate machine learning using the same data is done and presented, for comparison. After evaluation, validation and interpretation, a set of variables were highlighted as early predictors of lung cancer. The performance of the proposed approach managed to match the performance of the conventional approach. This promising result opens for further development where such an approach can be used in clinical decision support systems. Future work could then involve other modalities, in a multimodal machine learning approach. / Tidig lungcancerdiagnostisering kan öka chanserna för överlevnad hos lungcancerpatienter, men att upptäcka lungcancer i ett tidigt stadie är en av de större utmaningarna för onkologer och sjukvården. Idag undersöks patienter med riskfaktorer baserat på rökning och ålder, dessa undersökningar sker med hjälp av bland annat medicinskt avbildningssystem, då oftast CT-bilder, vilket medför felaktiga och kostsamma diagnoser. Detta arbete föreslår en maskininlärninig algoritm baserad på Natural language processing, som genom analys och bearbetning av ostrukturerade data, av patienternas egna anamneser, kan prediktera lungcancer. Arbetet har genomfört en jämförelse med en konventionell maskininlärning algoritm baserat på en replikering av ett annat studie där samma data behandlades som strukturerad. Den föreslagna metoden har visat ett likartat resultat samt prestanda, och har identifierat riskfaktorer samt symptom för lungcancer. Detta arbete öppnar upp för en utveckling mot ett kliniskt användande i form av beslutsstödsystem, som även kan hantera elektriska hälsojournaler. Andra arbeten kan vidareutveckla metoden för att hantera andra varianter av data, så som medicinska bilder och biomarkörer, och genom det förbättra prestandan.

Rheumatoid Arthritis is an inflammatory joint disease that is one of the most common autoimmune diseases in the world. The treatment usually starts with a first-line treatment called Methotrexate, but it is often insufficient. One of the most common second-line treatments is Tumor Necrosis Factor inhibitors (TNFi). Although some patients respond to TNFi, it has a risk of side effects, including infections. Hence, ability to predict patient responses to TNFi becomes important to choose the correct treatment. This work presents a new approach to predict if the patients were still on TNFi, 1 year after they started, by using a generative neural network architecture called Variational Autoencoder (VAE). We combined a VAE and a classifier neural network to create a supervised learning model called Supervised VAE (SVAE), trained on two versions of a tabular dataset containing Swedish register data. The datasets consist of 7341 patient records, and our SVAE achieved an AUROC score of 0.615 on validation data. Nevertheless, compared to machine learning models previously used for the same prediction task, SVAE achieved higher scores than decision trees and elastic net but lower scores than random forest and gradient-boosted decision tree. Despite the regularization effect that VAEs provide during classification training, the scores achieved by the SVAEs tested during this thesis were lower than the acceptable discrimination level. / Reumatoid artrit är en inflammatorisk ledsjukdom och är en av de vanligaste autoimmuna sjukdomarna i världen. Medicinsk behandling börjar ofta med Metotrexat. Vid brist på respons så fortsätter behandlingen ofta med Tumor Necrosis Inhibitors (TNFi). På grund av biverkningar av TNFi, såsom ökad risk för infektioner, är det viktigt att kunna prediktera patienters respons på behandlingen. Här presenteras ett nytt sätt att prediktera om patienter fortfarande stod på TNFi ett år efter initiering. Vi kombinerade Variational Autoencoder (VAE), ett generativt neuralt nätverk, med ett klassificeringsnätverk för att skapa en övervakad inlärningsmodell kallad Supervised VAE (SVAE). Denna tränades på två versioner av svenska registerdata, vilka innehöll information om 7341 patienter i tabellform. Vår SVAE-modell uppnådde 0,615 AUROC på valideringsdata. I jämförelse med maskininlärningsmodeller som tidigare använts för samma prediktionsuppgift uppnådde SVAE högre poäng än Decision Tree och Elastic Net men lägre poäng än Random Forest och Gradient-Boosted Decision Tree. Trots regulariseringseffekten som VAE ger under träning så var poängen som de testade SVAEmodellerna uppnår lägre än den acceptabla diskrimineringsnivån.

Introdução: A apneia obstrutiva do sono (AOS) é uma doença altamente prevalente, caracterizada pela obstrução recorrente da faringe durante o sono. Apesar do quadro clínico ser marcado por ronco e sonolência diurna excessiva e do maior risco cardiometabólico associado à AOS, uma grande parcela dos pacientes diagnosticados permanece sem nenhum tratamento. Dessa forma, são necessárias estratégias com o objetivo de otimizar as opções de tratamento para os pacientes com AOS. Objetivos: Esta tese é composta pela compilação de três artigos com o objetivo geral de avaliar os fatores fisiopatológicos anatômicos da AOS que podem influenciar na variabilidade individual de resposta ao tratamento. Os objetivos específicos de cada artigo são: Artigo 1) Avaliar se o padrão de obstrução da faringe influencia no efeito da mudança de decúbito de supino para lateral na patência da via aérea superior; Artigo 2) Avaliar as diferenças na complacência das regiões da faringe e sua associação com padrões de curva inspiratória; Artigo 3) Avaliar a influência da estrutura faríngea envolvida na obstrução e a colapsabilidade da via aérea superior na eficácia do aparelho intra-oral (AIO) no tratamento da AOS. Métodos: Foram recrutados pacientes com diagnóstico prévio de AOS com idade entre 21 a 70 anos. Artigo 1: Os indivíduos foram avaliados com sonoendoscopia e registro simultâneo do fluxo aéreo durante o sono natural em decúbito supino e lateral. Artigo 2: os indivíduos foram avaliados com sonoendoscopia e registro simultâneo da pressão faríngea durante o sono natural. Artigo 3: Os indivíduos foram avaliados na primeira noite com sonoendoscopia e, em duas noites adicionais, foram submetidos à polissonografia com e sem AIO para determinação do índice de apneiahipopneia (IAH) e para a medida da pressão crítica de fechamento da faringe (Pcrit). Resultados: Artigo 1: Foram avaliados 24 pacientes (17 homens, idade: 53±6 anos, IAH:48±28 eventos/hora). Em pacientes com obstrução associada a língua (n=10), não houve aumento significativo do pico de fluxo inspiratório e da ventilação minuto com a mudança de decúbito de supino para lateral. A posição lateral resultou em diminuição da ocorrência do colapso de epiglote e aumento de 45% na ventilação minuto entre os pacientes com obstrução da epiglote (n=6). Artigo 2: Foram avaliados 14 pacientes (9 homens, idade: 51±5 anos, IAH: 56±32 eventos/hora). Comparada à região retroglossal, a região retropalatal foi mais estreita (19,2 [23,9] mm2 versus 55,0 [30,7]mm2; p < 0,001) e apresentou maior complacência (3,2 ± 2,1 mm2/cmH2O versus 2,1 ± 1,8 mm2/cmH2O; p < 0,001). A dependência ao esforço negativo foi positivamente associada ao estreitamento da área retropalatal (r=0,47; p=0,001). Artigo 3: Foram avaliados 25 pacientes (17 homens, idade: 49±11 anos, IAH: 51±24 eventos/hora). O AIO reduziu a Pcrit em 3,9±2,4 cmH2O e o IAH em 69%. A redução da Pcrit foi maior nos pacientes com a língua posteriorizada. A ppresença da língua posteriorizada (p=0,03) e menor colapsabilidade da faringe (Pcrit < 1 cmH2O) no momento inicial (p=0,04) foram determinantes de melhor resposta terapêutica demonstrada pela maior redução do IAH com o AIO (83% versus 48%; p < 0,001). Conclusões: O padrão de obstrução e a colapsabilidade da faringe são fatores determinantes na resposta individual às modalidades terapêuticas alternativas para tratamento da AOS como terapia posicional e AIO. Pacientes com colapso de epiglote apresentaram melhora da patência da faringe com o decúbito lateral e assim podem se beneficiar da terapia posicional para AOS. A região retropalatal apresentou menor área e maior complacência comparada à região retroglossal nos pacientes com AOS. Finalmente, pacientes com língua posteriorizada e menor colapsabilidade da faringe são bons candidatos ao uso do AIO para tratamento da AOS / Introduction: Obstructive sleep apnea (OSA) is a highly prevalent disease characterized by recurrent pharyngeal obstruction during sleep. Despite symptoms such as snore and excessive daytime sleepiness and, the higher cardiometabolic risk related to OSA, a large proportion of patients do not receive any treatment for the disease. Therefore, strategies to improve management approaches for OSA are necessary. Objectives: This thesis consists of the compilation of three articles with a general aim of investigating the anatomic factors involved in OSA pathogenesis that could affect the individual variability of treatment responses. The specific aims of each article are: Article 1: To evaluate if the pattern of pharyngeal obstruction influences the effect of changing from supine to lateral position on upper airway patency; 2) To compare the compliance of each pharyngeal level and its association with inspiratory flow patterns; 3) To evaluate the effect of pharyngeal collapsibility and the pharyngeal structure causing collapse on the efficacy of oral appliance therapy for OSA. Methods: Patients previously diagnosed with OSA ranging from 21 to 70 years old were recruited for the studies. Article 1: Patients underwent upper airway endoscopy with simultaneous recordings of respiratory airflow in both supine and lateral position during natural sleep. Article 2: Patients underwent upper airway endoscopy with simultaneous recordings of pharyngeal pressure during natural sleep. Article 3: Patients underwent upper airway endoscopy on the first night. On two additional overnight studies, polysomnography was performed with and without an oral appliance to determine apnea-hypopnea index (AHI), and to measure the critical closing pressure (Pcrit). Results: Article 1: Twenty-four patients (17 men, 53±6 years old, AHI:48±28 events/hour) were studied. Patients with tongue-related obstruction (n=10) showed no improvement in airflow, and the tongue remained posteriorly located. Epiglottic obstruction was virtually abolished with lateral positioning and ventilation increased by 45% compared to supine position. Article 2: Fourteen patients (9 men, 51±5 years old, AHI: 56±32 events/hour) were studied. Compared to the retroglossal airway, the retropalatal airway was smaller at end-expiration (p < 0.001), and had greater absolute and relative compliances (p < 0.001). NED was positively associated with retropalatal relative area change (r=0.47; p < 0.001). Article 3: Twenty-five patients (17 men, 49±11 years old, IAH: 51±24 events/hour) were studied. Oral appliance therapy reduced Pcrit by 3.9±2.4 cmH2O and AHI by 69%. Oral appliance lowered Pcrit by 2.7±0.9 cmH2O more in those with posteriorly-located tongue compared to those without (p < 0.008). Posteriorly-located tongue (p=0.03) and lower baseline collapsibility (p=0.04) were significant determinants of a greater-than-average AHI response to therapy (83% versus 48%, p < 0.001). Conclusions: The pattern of obstruction and pharyngeal collapsibility are determinants of the individual response to alternative OSA treatment such as positional therapy and oral appliance. Patients with epiglottic obstruction showed significant improvement with lateral sleeping position and, therefore may benefit from positional therapy for OSA. The retropalatal airway had a smaller area and a greater compliance compared to the retroglossal airway in OSA patients. Finally, patients with posteriorly located tongue plus less-severe collapsibility are good candidates for oral appliance therapy

Introdução: A apneia obstrutiva do sono (AOS) é uma doença altamente prevalente, caracterizada pela obstrução recorrente da faringe durante o sono. Apesar do quadro clínico ser marcado por ronco e sonolência diurna excessiva e do maior risco cardiometabólico associado à AOS, uma grande parcela dos pacientes diagnosticados permanece sem nenhum tratamento. Dessa forma, são necessárias estratégias com o objetivo de otimizar as opções de tratamento para os pacientes com AOS. Objetivos: Esta tese é composta pela compilação de três artigos com o objetivo geral de avaliar os fatores fisiopatológicos anatômicos da AOS que podem influenciar na variabilidade individual de resposta ao tratamento. Os objetivos específicos de cada artigo são: Artigo 1) Avaliar se o padrão de obstrução da faringe influencia no efeito da mudança de decúbito de supino para lateral na patência da via aérea superior; Artigo 2) Avaliar as diferenças na complacência das regiões da faringe e sua associação com padrões de curva inspiratória; Artigo 3) Avaliar a influência da estrutura faríngea envolvida na obstrução e a colapsabilidade da via aérea superior na eficácia do aparelho intra-oral (AIO) no tratamento da AOS. Métodos: Foram recrutados pacientes com diagnóstico prévio de AOS com idade entre 21 a 70 anos. Artigo 1: Os indivíduos foram avaliados com sonoendoscopia e registro simultâneo do fluxo aéreo durante o sono natural em decúbito supino e lateral. Artigo 2: os indivíduos foram avaliados com sonoendoscopia e registro simultâneo da pressão faríngea durante o sono natural. Artigo 3: Os indivíduos foram avaliados na primeira noite com sonoendoscopia e, em duas noites adicionais, foram submetidos à polissonografia com e sem AIO para determinação do índice de apneiahipopneia (IAH) e para a medida da pressão crítica de fechamento da faringe (Pcrit). Resultados: Artigo 1: Foram avaliados 24 pacientes (17 homens, idade: 53±6 anos, IAH:48±28 eventos/hora). Em pacientes com obstrução associada a língua (n=10), não houve aumento significativo do pico de fluxo inspiratório e da ventilação minuto com a mudança de decúbito de supino para lateral. A posição lateral resultou em diminuição da ocorrência do colapso de epiglote e aumento de 45% na ventilação minuto entre os pacientes com obstrução da epiglote (n=6). Artigo 2: Foram avaliados 14 pacientes (9 homens, idade: 51±5 anos, IAH: 56±32 eventos/hora). Comparada à região retroglossal, a região retropalatal foi mais estreita (19,2 [23,9] mm2 versus 55,0 [30,7]mm2; p < 0,001) e apresentou maior complacência (3,2 ± 2,1 mm2/cmH2O versus 2,1 ± 1,8 mm2/cmH2O; p < 0,001). A dependência ao esforço negativo foi positivamente associada ao estreitamento da área retropalatal (r=0,47; p=0,001). Artigo 3: Foram avaliados 25 pacientes (17 homens, idade: 49±11 anos, IAH: 51±24 eventos/hora). O AIO reduziu a Pcrit em 3,9±2,4 cmH2O e o IAH em 69%. A redução da Pcrit foi maior nos pacientes com a língua posteriorizada. A ppresença da língua posteriorizada (p=0,03) e menor colapsabilidade da faringe (Pcrit < 1 cmH2O) no momento inicial (p=0,04) foram determinantes de melhor resposta terapêutica demonstrada pela maior redução do IAH com o AIO (83% versus 48%; p < 0,001). Conclusões: O padrão de obstrução e a colapsabilidade da faringe são fatores determinantes na resposta individual às modalidades terapêuticas alternativas para tratamento da AOS como terapia posicional e AIO. Pacientes com colapso de epiglote apresentaram melhora da patência da faringe com o decúbito lateral e assim podem se beneficiar da terapia posicional para AOS. A região retropalatal apresentou menor área e maior complacência comparada à região retroglossal nos pacientes com AOS. Finalmente, pacientes com língua posteriorizada e menor colapsabilidade da faringe são bons candidatos ao uso do AIO para tratamento da AOS / Introduction: Obstructive sleep apnea (OSA) is a highly prevalent disease characterized by recurrent pharyngeal obstruction during sleep. Despite symptoms such as snore and excessive daytime sleepiness and, the higher cardiometabolic risk related to OSA, a large proportion of patients do not receive any treatment for the disease. Therefore, strategies to improve management approaches for OSA are necessary. Objectives: This thesis consists of the compilation of three articles with a general aim of investigating the anatomic factors involved in OSA pathogenesis that could affect the individual variability of treatment responses. The specific aims of each article are: Article 1: To evaluate if the pattern of pharyngeal obstruction influences the effect of changing from supine to lateral position on upper airway patency; 2) To compare the compliance of each pharyngeal level and its association with inspiratory flow patterns; 3) To evaluate the effect of pharyngeal collapsibility and the pharyngeal structure causing collapse on the efficacy of oral appliance therapy for OSA. Methods: Patients previously diagnosed with OSA ranging from 21 to 70 years old were recruited for the studies. Article 1: Patients underwent upper airway endoscopy with simultaneous recordings of respiratory airflow in both supine and lateral position during natural sleep. Article 2: Patients underwent upper airway endoscopy with simultaneous recordings of pharyngeal pressure during natural sleep. Article 3: Patients underwent upper airway endoscopy on the first night. On two additional overnight studies, polysomnography was performed with and without an oral appliance to determine apnea-hypopnea index (AHI), and to measure the critical closing pressure (Pcrit). Results: Article 1: Twenty-four patients (17 men, 53±6 years old, AHI:48±28 events/hour) were studied. Patients with tongue-related obstruction (n=10) showed no improvement in airflow, and the tongue remained posteriorly located. Epiglottic obstruction was virtually abolished with lateral positioning and ventilation increased by 45% compared to supine position. Article 2: Fourteen patients (9 men, 51±5 years old, AHI: 56±32 events/hour) were studied. Compared to the retroglossal airway, the retropalatal airway was smaller at end-expiration (p < 0.001), and had greater absolute and relative compliances (p < 0.001). NED was positively associated with retropalatal relative area change (r=0.47; p < 0.001). Article 3: Twenty-five patients (17 men, 49±11 years old, IAH: 51±24 events/hour) were studied. Oral appliance therapy reduced Pcrit by 3.9±2.4 cmH2O and AHI by 69%. Oral appliance lowered Pcrit by 2.7±0.9 cmH2O more in those with posteriorly-located tongue compared to those without (p < 0.008). Posteriorly-located tongue (p=0.03) and lower baseline collapsibility (p=0.04) were significant determinants of a greater-than-average AHI response to therapy (83% versus 48%, p < 0.001). Conclusions: The pattern of obstruction and pharyngeal collapsibility are determinants of the individual response to alternative OSA treatment such as positional therapy and oral appliance. Patients with epiglottic obstruction showed significant improvement with lateral sleeping position and, therefore may benefit from positional therapy for OSA. The retropalatal airway had a smaller area and a greater compliance compared to the retroglossal airway in OSA patients. Finally, patients with posteriorly located tongue plus less-severe collapsibility are good candidates for oral appliance therapy