El texto completo de este trabajo no está disponible en el Repositorio Académico UPC por restricciones de la casa editorial donde ha sido publicado. / Background: Pulmonary tuberculosis is a highly prevalent disease in low-income countries; clinical prediction tools allow healthcare personnel to catalog patients with a higher risk of death in order to prioritize medical attention. Methodology: We conducted a literature search on prognostic models aimed to predict mortality in patients diagnosed with pulmonary tuberculosis. We included prospective and retrospective studies where prognostic models predicting mortality were either developed or validated in patients diagnosed with pulmonary tuberculosis. Three reviewers independently assessed the quality of the included studies using the PROBAST tool (Prediction model study Risk of Bias Assessment Tool). A narrative review of the characteristics of each model was conducted. Results: Six articles (n = 3553 patients) containing six prediction models were included in the review. Most studies (5 out of 6) were retrospective cohorts, only one study was a prospective case-control study. All the studies had a high risk of bias according to the PROBAST tool in the overall assessment. Regarding the applicability of the prediction models, three studies had a low concern of applicability, two high concern and one unclear concern. Five studies developed new prediction rules. In general, the presented models had a good discriminatory ability, with areas under the curve fluctuating between 0.65 up to 0.91. Conclusion: None of the prognostic models included in the review accurately predict mortality in patients with pulmonary tuberculosis, due to great heterogeneity in the population and a high risk of bias. / Revisión por pares

Prognostic models have clinical appeal to aid therapeutic decision making. In the UK, the Nottingham Prognostic Index (NPI) has been used, for over two decades, to inform patient management. However, it has been commented that NPI is not capable of identifying a subgroup of patients with a prognosis so good that adjuvant therapy with potential harmful side effects can be withheld safely. Tissue Microarray Analysis (TMA) now makes possible measurement of biological tissue microarray features of frozen biopsies from breast cancer tumours. These give an insight to the biology of tumour and hence could have the potential to enhance prognostic modelling. I therefore wished to investigate whether biomarkers can add value to clinical predictors to provide improved prognostic stratification in terms of Recurrence Free Survival (RFS). However, there are very many biomarkers that could be measured, they usually exhibit skewed distribution and missing values are common. The statistical issues raised are thus number of variables being tested, form of the association, imputation of missing data, and assessment of the stability and internal validity of the model. Therefore the specific aim of this study was to develop and to demonstrate performance of statistical modelling techniques that will be useful in circumstances where there is a surfeit of explanatory variables and missing data; in particular to achieve useful and parsimonious models while guarding against instability and overfitting. I also sought to identify a subgroup of patients with a prognosis so good that a decision can be made to avoid adjuvant therapy. I aimed to provide statistically robust answers to a set of clinical question and develop strategies to be used in such data sets that would be useful and acceptable to clinicians. A unique data set of 401 Estrogen Receptor positive (ER+) tamoxifen treated breast cancer patients with measurement for a large panel of biomarkers (72 in total) was available. Taking a statistical approach, I applied a multi-faceted screening process to select a limited set of potentially informative variables and to detect the appropriate form of the association, followed by multiple imputations of missing data and bootstrapping. In comparison with the NPI, the final joint model derived assigned patients into more appropriate risk groups (14% of recurred and 4% of non-recurred cases). The actuarial 7-year RFS rate for patients in the lowest risk quartile was 95% (95% C.I.: 89%, 100%). To evaluate an alternative approach, biological knowledge was incorporated into the process of model development. Model building began with the use of biological expertise to divide the variables into substantive biomarker sets on the basis of presumed role in the pathway to cancer progression. For each biomarker family, an informative and parsimonious index was generated by combining family variables, to be offered to the final model as intermediate predictor. In comparison with NPI, patients into more appropriate risk groups (21% of recurred and 11% of non-recurred patients). This model identified a low-risk group with 7-year RFS rate at 98% (95% C.I.: 96%, 100%).

Introduction. Alors qu’elle constitue le seul traitement curatif du cholangiocarcinome intrahépatique (CCIH), la résection reste associée à un taux de récidive supérieur à 60% et un taux de survie réelle à 5 ans inférieur à 20%. Une estimation fiable du pronostic ainsi qu’une meilleure compréhension de la biologie tumorale est essentielle pour améliorer le pronostic.Méthodes. A l’appui des données clinico-biologiques de deux larges cohortes de patients avec CCIH réséqué (MSKCC, n=189 et AFC, n=522), trois objectifs ont été explorés. Tout d’abord, définir quel modèle pronostique publié est le plus performant. Ensuite, définir la fiabilité de l’évaluation pronostique préopératoire à partir de, respectivement, l’imagerie, des microARN (miR) circulants diagnostiques et du profil génomique tumoral. Enfin, évaluer l’impact pronostique de la survenue d’événements périopératoires tels que transfusion et morbidité.Résultats. Premièrement, les nomogrammes apportaient une meilleure estimation pronostique en comparaison à la classification AJCC 7ème édition. Deuxièmement, la taille et la multifocalité tumorale sur l’imagerie préopératoire permettaient de différencier deux groupes de patients de pronostic clairement distincts (p<0,001). L’existence d’une mutation d’un gène de remodelage de la chromatine (BAP1, ARID1A, PBRM1) tendait à être associé à une survie sans récidive plus favorable qu’en l’absence de mutation (p=0,09). Alors qu’ayant un potentiel comme marqueur diagnostique circulant, miR21 et miR221 n’étaient pas associé à la survie. Troisièmement, la transfusion peropératoire n’impactait pas la survie à long terme alors que la survenue d’une complication sévère (grade Dindo-Clavien > 2) était indépendamment associée à une survie globale plus courte (p=0,002).Conclusion. Alors que les nomogrammes postopératoires apportent une meilleure estimation pronostique, le développement de modèles pronostiques préopératoires est faisable notamment à partir de l’imagerie et de marqueurs biologiques tumoraux complémentaires. / Introduction. Complete resection stands as the only curative option for intrahepatic cholangiocarcinoma (IHCC). Still, prognosis remains poor after resection due to a recurrence rate over 60% leading to actual 5-year survival rates below 20%. Reliable prognostic estimation and better understanding of tumor biology would be of interest for improving IHCC prognosis.Methods. Using clinical and biological data from two large cohort of resected IHCC (MSKCC, n=189 and AFC, n=522), three objectives have been explored. First, assessing the performances of different published prognostic models. Second, defining the reliability of preoperative prognostic estimation using imaging, tumoral genomic profiling and circulating tumoral microRNA (miR). Third, evaluating the prognostic impact of perioperative events such as blood transfusion and morbidity.Results. First, nomograms displayed better prognostic accuracy over the AJCC 7th edition staging system. Second, tumor size and multifocality on preoperative imaging allowed patient stratification in groups statistically different regarding prognosis (p<0.001). Further, the presence of chromatine remodeling gene mutations (BAP1, ARID1A, PBRM1) tended towards longer recurrence-free survical (p=0,09). Some diagnostic circulating miR such as miR21 and miR221 were not associated with survival. Third, in contrast with intraoperative transfusion, the occurrence of severe morbidity (Dindo-Clavien grade > 2) was independently associated with shorter overall survival (p=0.002).Conclusion. Nomograms outperform conventional staging sytem. Preoperative prognostic estimation is feasible and reliable using imaging. Identifying new prognostic biomarkers would help refining preoperative prognostic estimation.

In this dissertation, we present a collection of manuscripts describing the development of prognostic models designed to assist clinical decision making. This work is motivated by limitations of commonly used techniques to produce accessible prognostic models with easily interpretable and clinically credible results. Such limitations hinder prognostic model widespread utilization in medical practice. Our methodology is based on Rough Set Theory (RST) as a mathematical tool for clinical data anal- ysis. We focus on developing rule-based prognostic models for end-of life care decision making in an effort to improve the hospice referral process. The development of the prognostic models is demonstrated using a retrospective data set of 9,103 terminally ill patients containing physiological characteristics, diagnostic information and neurological function values. We develop four RST-based prognostic models and compare them with commonly used classification techniques including logistic regression, support vector machines, random forest and decision trees in terms of characteristics related to clinical credibility such as accessibility and accuracy. RST based models show comparable accuracy with other methodologies while providing accessible models with a structure that facilitates clinical interpretation. They offer both more insight into the model process and more opportunity for the model to incorporate personal information of those making and being affected by the decision.

Solicitud de embargo por publicación en revista indexada. / OBJECTIVE. To synthesize the evidence regarding prognostic models to predict mortality in patients diagnosed with pulmonary tuberculosis. METHODOLOGY. The current study followed the guidelines of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (The PRISMA Group, 2020) Statement. A literature search on prognostic models aimed to predict mortality in patients diagnosed with pulmonary tuberculosis was conducted by three revisers. We included prospective and retrospective studies where prognostic models predicting mortality were either developed or validated in patients diagnosed with pulmonary tuberculosis. Three reviewers independently assessed the quality of the included studies using the PROBAST tool. (¨Prediction model study Risk Of Bias Assessment Tool¨), which assesses both the risk of bias (RoB) and the applicability of each model. A descriptive analysis of each of the prediction models developed, their performance and the population characteristics of each article was conducted. RESULTS. Only 6 articles met the selection criteria. There was a total of 6 prognostic rules, one in each article. Most studies (5 out of 6) were retrospective cohorts, only 1 study was a prospective case-control study. When adding the population of all the studies, there were a total of 3,553 participants, with samples ranging from 103 participants to 1070 participants. All the studies had a high risk of bias according to the PROBAST tool in the overall assessment. The overall assessment showed that 3 studies had a low concern of applicability, 2 high concern and 1 unclear concern. Only 5 studies developed new prediction rules. In general, the presented models had a good discriminatory ability, with areas under the curve fluctuating between 0.65 up to 0.91. The predictive model with the highest discriminatory power was the one reported by Horita, et - al. with an AUC of 0.910 in the development cohort and 0.893 in the validation cohort. CONCLUSION. Considering that pulmonary tuberculosis is a highly prevalent disease in low-income countries, it would be very useful to have quality tools that allow healthcare personnel to be able to catalog patients with a higher risk of death so that they can receive priority medical attention. / OBJETIVO Sintetizar la evidencia acerca de modelos pronósticos que predicen mortalidad en pacientes con tuberculosis pulmonar. METODOLOGÍA. El siguiente estudio sigue las guías PRISMA del año 2020 (Preferred Reporting Items for Systematic Reviews and Meta-Analyses). Se realizó una búsqueda literaria, por tres revisores, de modelos pronósticos que se enfocaban en predecir mortalidad en pacientes diagnosticados con tuberculosis pulmonar. Se incluyeron estudios prospectivos y retrospectivos, donde los modelos pronósticos que predecían mortalidad habían sido desarrollados o validados en pacientes con tuberculosis pulmonar. De manera independiente, tres revisores evaluaron la calidad de los estudios incluidos usando la herramienta PROBAST (¨Prediction model study Risk Of Bias Assessment Tool¨), la cual evalúa el riesgo de sesgo y la aplicabilidad de cada modelo. Se realizó un análisis descriptivo de cada modelo de predicción, su performance, y las características de la población. RESULTADOS. Solo 6 artículos cumplieron los criterios de selección. Hubo un total de 6 modelos pronósticos, uno en cada artículo. La mayoría de los estudios (5 de 6) fueron cohortes retrospectivas, y solo uno fue un estudio de casos y controles prospectivo. Al sumar la población total de los estudios, hubo un total de 3,553 participantes, con muestras desde 103 hasta 1070 participantes. Todos los estudios obtuvieron un alto riesgo de sesgo, de acuerdo a la herramienta PROBAST, en la evaluación global. Además, la evaluación global mostró que 3 estudios obtuvieron una baja preocupación de aplicabilidad, 2 alta preocupación y un estudio preocupación indeterminada. Solo 5 estudios desarrollaron nuevas reglas de predicción, mientras que uno válido una ya existente. En general los modelos de predicción mostraron una buena habilidad discriminatoria, con valores de área bajo la curva que fluctuaban entre 0.65 hasta 0.91. El modelo de predicción con mayor poder discriminatorio fue el reportado por Horita, et – al con un valor de área bajo la curva de 0.910 en la cohorte de desarrollo y 0.893 en la cohorte de validación. CONCLUSIÓN. Tomando en cuenta que la tuberculosis pulmonar es una enfermedad prevalente en países de desarrollo, sería útil contar con herramientas que ayuden a los profesionales de la salud a catalogar a los pacientes con mayor riesgo de mortalidad, para que así ellos puedan recibir atención médica prioritaria. / Tesis

No description available.