Personalized modeling for drug concentration prediction using Support Vector Machine

Building a personalized model to describe the drug concentration inside the human body for each patient is highly important to the clinical practice and demanding to the modeling tools. Instead of using traditional explicit methods, in this paper we propose a machine learning approach to describe the relation between the drug concentration and patients' features. Machine learning has been largely applied to analyze data in various domains, but it is still new to personalized medicine, especially dose individualization. We focus mainly on the prediction of the drug concentrations as well as the analysis of different features' influence. Models are built based on Support Vector Machine and the prediction results are compared with the traditional analytical models.

Médecine génomique et Maladies infectieuses [Genomic medicine and infectious diseases].

Relentless progress in our knowledge of the nature and functional consequences of human genetic variation allows for a better understanding of the protracted battle between pathogens and their human hosts. Multiple polymorphisms have been identified that impact our response to infections or to anti-infective drugs, and some of them are already used in the clinic. However, to make personalized medicine a reality in infectious diseases, a sustained effort is needed not only in research but also in genomic education.

Cancer omics strategies for personalized treatments

The Consensus Molecular Subtypes (CMSs) classification stratifies colorectal cancer (CRC) into four well-defined molecular subgroups, providing incredible support to personalized medicine. Indeed, the huge inter-patient heterogeneity observed in CRC makes it difficult to define a therapeutic strategy from which every patient can benefit. Unfortunately, so far really few targetable biomarkers are known in the CRC setting, leading to an urgent need for new targeted therapies. Here we performed a bioinformatic meta-analysis over a cohort of 1700 CMS-stratified CRC patients, identifying a negative correlation between high levels of anaplastic lymphoma kinase (ALK) expression and relapse-free survival, exclusively in the CMS1 subtype. No correlation with ALK expression was pointed out in the other three subgroups. The association of ALK with CMS1 led to generate the hypothesis that ALK pharmacological inhibition may elicit therapeutic potential in this subgroup. Thus, we tested ALK inhibitors and an ALK-directed ADC on several CRC in vitro models, stratified according to the CMS classification as well as on CRC patient-derived organoids and mice. ALK interception strongly inhibited CMS1-cells, organoids, and tumor proliferation and was responsible for the dampening of ALK activation along with the downstream. Mechanistically, we found that CMS1 cells display several mRNA copies of both ALK and ALKAL2 ligand, suggesting a role for ALK abundance in the differential response to its inhibition. Collectively, these findings support the hypothesis that ALK may represent an attractive target for CMS1 colorectal cancer therapy.

A farmacogenética e a medicina personalizada

A farmacogenética tem por objetivo a identificação de diferenças genéticas entre indivíduos que possam influenciar a resposta à terapêutica farmacológica, melhorando a sua eficácia e segurança. Associado à farmacogenética surge a “medicina personalizada”, ou seja, em oposição à existência de um fármaco que consiga tratar todos os pacientes, o tratamento individualizado parece o caminho mais promissor, uma vez que reduz o risco de reações adversas por toxicidade (segurança), adequa a dose ao indivíduo, evitando excessos ou défices (dose) e evita a metodologia de tentativa erro na escolha do fármaco (eficácia). A farmacogenética é relevante para a resposta individual ao fármaco por duas vias distintas: a farmacocinética e a farmacodinâmica. A variabilidade genética pode afetar a forma como um fármaco pode ser absorvido, ativado, metabolizado ou excretado, podendo conduzir assim a uma variabilidade na resposta. De entre o número infindável de possíveis exemplos, nesta revisão apresentam-se exemplos relacionados com os genes do Citocromo P450, do gene NAT2 e do gene da Colinesterase. As diferenças genéticas entre os indivíduos podem ainda afetar a resposta ao fármaco pela sua farmacodinâmica, ou seja, a resposta específica do alvo ao fármaco. De entre a multiplicidade de alvos de fármacos existentes serão apresentados exemplos do gene da G6PD e do VKORC1. Apesar de alguns dados científicos indicarem benefício para o paciente, ainda está longe de a farmacogenética fazer parte da prática clínica de rotina, talvez porque os custos-benefícios ainda não foram avaliados de forma precisa.

Urothelial bladder cancer progression: lessons learned from the bench

Urothelial bladder carcinoma (UBC) is an intricate malignancy with a variable natural history and clinical behavior. Despite developments in diagnosis/prognosis refinement and treatment modalities, the recurrence rate is high, and progression from non-muscle to muscle invasive UBC commonly leads to metastasis. Moreover, patients with muscle-invasive or extra-vesical disease often fail the standard chemotherapy treatment, and overall survival rates are poor. Thus, UBC remains a challenge in the oncology field, representing an ideal candidate for research on biomarkers that could identify patients at increased risk of recurrence, progression, and chemo-refractoriness. However, progress toward personalized medicine has been hampered by the unique genetic complexity of UBC. Recent genome-wide expression and sequencing studies have brought new insights into its molecular features, pathogenesis and clinical diversity, revealing a landscape where classical pathology is intersected by the novel and heterogeneous molecular groups. Hence, it seems plausible to postulate that only an integrated signature of prognostic/predictive biomarkers inherent in different cancer hallmarks will reach clinical validation. In this review, we have summarized ours and others' research into novel putative biomarkers of progression and chemoresistance that encompass several hallmarks of cancer: tumor neovascularization, invasion and metastasis, and energy metabolism reprogramming of the tumor microenvironment.

Pathology as the Cornerstone of Human Tissue Banking: European Consensus Expert Group Report

Aside from ethical considerations, the primary requirement for usage of human tissues in basic or translational research is the thorough characterization of tissues. The second, but equally essential, requirement is that tissues be collected, processed, annotated, and preserved in optimal conditions. These requirements put the pathologist at the center of tissue banking activities and of research aimed at discovering new biomarkers. Pathologists not only provide information identifying the specimen but also make decisions on what materials should be biobanked, on the preservation conditions, and on the timeline of events that precede preservation and storage. This central position calls for increased recognition of the role of the pathologist by the biomolecular community and places new demands on the pathologist's workload and scope of scientific activities. These questions were addressed by an Expert Group Meeting of the European Biological and Biomolecular Research Infrastructure (BBMRI). While detailed recommendations are published elsewhere (Bevilacqua et al., Virchows Archivs, 2010, in press), this article outlines the strategic and technological issues identified by the Expert Group and identifies ways forward for better integration of pathology in the current thrust for development of biomarker-based "personalized medicine.

MR MAQ: algorisme de Read Mapping utilitzant la plataforma Hadoop

L’èxit del Projecte Genoma Humà (PGH) l’any 2000 va fer de la “medicina personalitzada” una realitat més propera. Els descobriments del PGH han simplificat les tècniques de seqüenciació de tal manera que actualment qualsevol persona pot aconseguir la seva seqüència d’ADN complerta. La tecnologia de Read Mapping destaca en aquest tipus de tècniques i es caracteritza per manegar una gran quantitat de dades. Hadoop, el framework d’Apache per aplicacions intensives de dades sota el paradigma Map Reduce, resulta un aliat perfecte per aquest tipus de tecnologia i ha sigut l’opció escollida per a realitzar aquest projecte. Durant tot el treball es realitza l’estudi, l’anàlisi i les experimentacions necessàries per aconseguir un Algorisme Genètic innovador que utilitzi tot el potencial de Hadoop.

New frontiers in therapeutic resistance in cancer.

Despite advances in personalized medicine and targeted therapies, therapeutic resistance remains a persistent dilemma encountered by clinicians, scientists and patients. In this article we summarize the highlights of the third Quebec Conference on Therapeutic Resistance in Cancer. This unique meeting provided researchers and clinicians with insights into: intrinsic and acquired resistance; tumor heterogeneity; complexities of biomarker-driven trials; challenges of 'omics data analysis; and models of clinical applications of personalized medicine. Emphasized throughout the conference was the importance of collaborations - between industry and academia, and between basic researchers and clinicians - so that therapeutic resistance can be studied where it matters most, in patients.

The future key role of LC-high-resolution-MS analyses in clinical laboratories: a focus on quantification.

For the last decade, high-resolution (HR)-MS has been associated with qualitative analyses while triple quadrupole MS has been associated with routine quantitative analyses. However, a shift of this paradigm is taking place: quantitative and qualitative analyses will be increasingly performed by HR-MS, and it will become the common 'language' for most mass spectrometrists. Most analyses will be performed by full-scan acquisitions recording 'all' ions entering the HR-MS with subsequent construction of narrow-width extracted-ion chromatograms. Ions will be available for absolute quantification, profiling and data mining. In parallel to quantification, metabotyping will be the next step in clinical LC-MS analyses because it should help in personalized medicine. This article is aimed to help analytical chemists who perform targeted quantitative acquisitions with triple quadrupole MS make the transition to quantitative and qualitative analyses using HR-MS. Guidelines for the acceptance criteria of mass accuracy and for the determination of mass extraction windows in quantitative analyses are proposed.

Les tests génétiques peuvent-ils être utiles pour la prévention cardiovasculaire? [Are genetic tests useful for cardiovascular prevention?].

Recent progresses in genetics have opened new avenues to further our understanding of the pathophysiological mechanisms underlying cardiovascular disease, raising, new expectations in the field of personalized medicine. Genetic tests may have a high predictive value for rare monogenic diseases. The situation is very different for common polygenic diseases, such as myocardial infarction, type 2 diabetes or stroke. The results from recent genome-wide association studies have provided useful information for research, but have not yet been proven to be clinically useful. It is therefore currently not recommended to conducted genetic testing to guide cardiovascular prevention neither in clinical nor in public health settings.

Gene-environment interactions of selected pharmacogenes in arterial hypertension.

Hypertension affects approximately 1 billion people worldwide. Owing to population aging, hypertension-related cardiovascular burden is expected to rise in the near future. In addition to genetic variants influencing the blood pressure response to antihypertensive drugs, several genes encoding for drug-metabolizing or -transporting enzymes have been associated with blood pressure and/or hypertension in humans (e.g., ACE, CYP1A2, CYP3A5, ABCB1 and MTHFR) regardless of drug treatment. These genes are also involved in the metabolism and transport of endogenous substances and their effects may be modified by selected environmental factors, such as diet or lifestyle. However, little is currently known on the complex interplay between environmental factors, endogenous factors, genetic variants and drugs on blood pressure control. This review will discuss the respective role of population-based primary prevention and personalized medicine for arterial hypertension, taking a pharmacogenomics' perspective focusing on selected pharmacogenes.

Conceptos generales y métodos de estudio en farmacogenética [General concepts and study methods in pharmacogenetics.]

Pharmacogenetics, the study of how individual genetic profiles influence the response to drugs, is an important topic. Results from pharmacogenetics studies in various clinical settings may lead to personalized medicine. Herein, we present the most important concepts of this discipline, as well as currently-used study methods.

Example-based support vector machine for drug concentration analysis

Machine learning has been largely applied to analyze data in various domains, but it is still new to personalized medicine, especially dose individualization. In this paper, we focus on the prediction of drug concentrations using Support Vector Machines (S VM) and the analysis of the influence of each feature to the prediction results. Our study shows that SVM-based approaches achieve similar prediction results compared with pharmacokinetic model. The two proposed example-based SVM methods demonstrate that the individual features help to increase the accuracy in the predictions of drug concentration with a reduced library of training data.

Computational analysis of the genetic and environmental contributions to disease-related human phenotypes: From predicting adverse lipid response of HIV patients receiving antiretroviral therapy to genome-wide association studies of cardiovascular and lipid related disorders

Genetic variants influence the risk to develop certain diseases or give rise to differences in drug response. Recent progresses in cost-effective, high-throughput genome-wide techniques, such as microarrays measuring Single Nucleotide Polymorphisms (SNPs), have facilitated genotyping of large clinical and population cohorts. Combining the massive genotypic data with measurements of phenotypic traits allows for the determination of genetic differences that explain, at least in part, the phenotypic variations within a population. So far, models combining the most significant variants can only explain a small fraction of the variance, indicating the limitations of current models. In particular, researchers have only begun to address the possibility of interactions between genotypes and the environment. Elucidating the contributions of such interactions is a difficult task because of the large number of genetic as well as possible environmental factors.In this thesis, I worked on several projects within this context. My first and main project was the identification of possible SNP-environment interactions, where the phenotypes were serum lipid levels of patients from the Swiss HIV Cohort Study (SHCS) treated with antiretroviral therapy. Here the genotypes consisted of a limited set of SNPs in candidate genes relevant for lipid transport and metabolism. The environmental variables were the specific combinations of drugs given to each patient over the treatment period. My work explored bioinformatic and statistical approaches to relate patients' lipid responses to these SNPs, drugs and, importantly, their interactions. The goal of this project was to improve our understanding and to explore the possibility of predicting dyslipidemia, a well-known adverse drug reaction of antiretroviral therapy. Specifically, I quantified how much of the variance in lipid profiles could be explained by the host genetic variants, the administered drugs and SNP-drug interactions and assessed the predictive power of these features on lipid responses. Using cross-validation stratified by patients, we could not validate our hypothesis that models that select a subset of SNP-drug interactions in a principled way have better predictive power than the control models using "random" subsets. Nevertheless, all models tested containing SNP and/or drug terms, exhibited significant predictive power (as compared to a random predictor) and explained a sizable proportion of variance, in the patient stratified cross-validation context. Importantly, the model containing stepwise selected SNP terms showed higher capacity to predict triglyceride levels than a model containing randomly selected SNPs. Dyslipidemia is a complex trait for which many factors remain to be discovered, thus missing from the data, and possibly explaining the limitations of our analysis. In particular, the interactions of drugs with SNPs selected from the set of candidate genes likely have small effect sizes which we were unable to detect in a sample of the present size (<800 patients).In the second part of my thesis, I performed genome-wide association studies within the Cohorte Lausannoise (CoLaus). I have been involved in several international projects to identify SNPs that are associated with various traits, such as serum calcium, body mass index, two-hour glucose levels, as well as metabolic syndrome and its components. These phenotypes are all related to major human health issues, such as cardiovascular disease. I applied statistical methods to detect new variants associated with these phenotypes, contributing to the identification of new genetic loci that may lead to new insights into the genetic basis of these traits. This kind of research will lead to a better understanding of the mechanisms underlying these pathologies, a better evaluation of disease risk, the identification of new therapeutic leads and may ultimately lead to the realization of "personalized" medicine.